- T HE GEOLOGICAL SOCIETY

- OF AMERICA®

Special Paper 436

Formation and Applications of the Sedimentary Record
in Arc Collision Zones

Edited by

Amy E. Draut
U.S. Geological Survey
USGS Pacific Science Center
400 Natural Bridges Drive
Santa Cruz, California 95060
USA

Peter D. Clift
School of Geosciences
University of Aberdeen
Aberdeen, AB24 3UE
UK
and
DFG-Research Centre Ocean Margins
Geowissenschaften, Universität Bremen
Klagenfurter Strasse
28359 Bremen
Germany

David W. Scholl
Emeritus Senior Scientist
U.S. Geological Survey
345 Middlefield Road
Menlo Park, California 94025
USA
and
Stanford University (Emeritus)
USA

Special Paper 436

3300 Penrose Place, P.O. Box 9140 Boulder, Colorado 80301-9140 USA

2008
Copyright © 2008, The Geological Society of America, Inc. (GSA). All rights reserved. GSA grants
permission to individual scientists to make unlimited photocopies of one or more items from this volume
for noncommercial purposes advancing science or education, including classroom use. For permission to
make photocopies of any item in this volume for other noncommercial, nonprofit purposes, contact the
Geological Society of America. Written permission is required from GSA for all other forms of capture
or reproduction of any item in the volume including, but not limited to, all types of electronic or digital
scanning or other digital or manual transformation of articles or any portion thereof, such as abstracts,
into computer-readable and/or transmittable form for personal or corporate use, either noncommercial
or commercial, for-profit or otherwise. Send permission requests to GSA Copyright Permissions,
3300 Penrose Place, P.O. Box 9140, Boulder, Colorado 80301-9140, USA.

Copyright is not claimed on any material prepared wholly by government employees within the scope of
their employment.

Published by The Geological Society of America, Inc.

3300 Penrose Place, P.O. Box 9140, Boulder, Colorado 80301-9140, USA
www.geosociety.org

Printed in U.S.A.

GSA Books Science Editor: Marion E. Bickford and Donald I. Siegel

Library of Congress Cataloging-in-Publication Data

Formation and applications of the sedimentary record in arc collision zones / edited by Amy E. Draut,
Peter D. Clift, David W. Scholl.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-8137-2436-2 (pbk.)
1. Island arcs. 2. Rocks, Carbonate—Analysis. 3. Convergent margins. 4. Plate tectonics. 5. Geology,
Stratigraphic. I. Draut, Amy E. II. Clift, P. D. (Peter D.) III. Scholl, David W.
QE511.2.F67 2008
551.1′36—dc22

2007048410

Cover: Oligocene shelf sandstones and shales of the paleo-Chinese passive margin exposed on the
northern coast of Taiwan. The rocks are buried and deformed during the ongoing collision of the Luzon
Arc with the passive margin of southern China. Towards the north of Taiwan the mountains experience
gravitational collapse and extension, culminating in subsidence and formation of the Ilan Plain basin and
Okinawa Trough. Photo by Andrew Lin.

10 9 8 7 6 5 4 3 2 1
 Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v

1. Preservation of forearc basins during island arc–continent collision: Some insights

from the Ordovician of western Ireland . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Paul D. Ryan

2. Basin formation by volcanic arc loading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Dave Waltham, Robert Hall, Helen R. Smyth, and Cynthia J. Ebinger

3. Cenozoic arc processes in Indonesia: Identification of the key influences on the

stratigraphic record in active volcanic arcs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
Robert Hall and Helen R. Smyth

4. Carbonate-platform facies in volcanic-arc settings: Characteristics and controls on

deposition and stratigraphic development . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
Steven L. Dorobek

5. Sediment waves in the Bismarck Volcanic Arc, Papua New Guinea . . . . . . . . . . . . . . . . . . . . . . . 91

Gary Hoffmann, Eli Silver, Simon Day, Eugene Morgan, Neal Driscoll, and Daniel Orange

6. The Lichi Mélange: A collision mélange formation along early arcward backthrusts
during forearc basin closure, Taiwan arc-continent collision . . . . . . . . . . . . . . . . . . . . . . . . . . . 127
Chi-Yue Huang, Chih-Wei Chien, Bochu Yao, and Chung-Pai Chang

7. Oblique subduction in an island arc collision setting: Unique sedimentation, accretion,

and deformation processes in the Boso TTT-type triple junction area, NW Pacific . . . . . . . . . . 155
Yujiro Ogawa, Yoshihiro Takami, and Sakiko Takazawa

8. The West Crocker formation of northwest Borneo: A Paleogene accretionary prism . . . . . . . . 171
Joseph J. Lambiase, Tan Yaw Tzong, Amelia G. William, Michael D. Bidgood, Patrice Brenac,
and Andrew B. Cullen

9. Temporal changes in the composition of Miocene sandstone related to collision

between the Honshu and Izu Arcs, central Japan . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 185
Koichi Okuzawa and Ken-ichiro Hisada

10. Cenozoic volcanic arc history of East Java, Indonesia: The stratigraphic record
of eruptions on an active continental margin. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199
Helen R. Smyth, Robert Hall, and Gary J. Nichols

iii
iv Contents

11. New constraints on the sedimentation and uplift history of the Andaman-Nicobar
accretionary prism, South Andaman Island . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 223
R. Allen, A. Carter, Y. Najman, P.C. Bandopadhyay, H.J. Chapman, M.J. Bickle, E. Garzanti,
G. Vezzoli, S. Andò, G.L. Foster, and C. Gerring

12. Post-collisional collapse in the wake of migrating arc-continent collision in the

Ilan Basin, Taiwan. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257
Peter D. Clift, Andrew T.S. Lin, Andrew Carter, Francis Wu, Amy E. Draut, T.-H. Lai, L.-Y. Fei,
Hans Schouten, and Louis Teng

13. The Guerrero Composite Terrane of western Mexico: Collision and subsequent
rifting in a supra-subduction zone . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279
E. Centeno-García, M. Guerrero-Suastegui, and O. Talavera-Mendoza

14. Tectonic architecture of an arc-arc collision zone, Newfoundland Appalachians . . . . . . . . . . . 309

Alexandre Zagorevski, Cees R. van Staal, Vicki McNicoll, Neil Rogers, and Pablo Valverde-Vaquero

15. The Catalina Schist: Evidence for middle Cretaceous subduction erosion of
southwestern North America . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 335
M. Grove, G.E. Bebout, C.E. Jacobson, A.P. Barth, D.L. Kimbrough, R.L. King, Haibo Zou,
O.M. Lovera, B.J. Mahoney, and G.E. Gehrels

16. Sedimentary response to arc-continent collision, Permian, southern Mongolia . . . . . . . . . . . . 363

C.L. Johnson, J.A. Amory, D. Zinniker, M.A. Lamb, S.A. Graham, M. Affolter, and G. Badarch

17. Links among mountain building, surface erosion, and growth of an accretionary
prism in a subduction zone—An example from southwest Japan . . . . . . . . . . . . . . . . . . . . . . . . 391
Gaku Kimura, Yujin Kitamura, Asuka Yamaguchi, and Hugues Raimbourg

Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
 Preface
The Sedimentary Record in Arc Collision Zones

Sediment deposited in volcanic arc settings yields the best available record of active-margin evolu-
tion spanning long periods of geologic time. The tectonic, erosional, and magmatic geochemical histories
preserved in arc sedimentary records, and particularly in arc collision zones, hold the key to understanding
multiple, inter-connected geologic processes: the formation and destruction of continental crust, changes in
plate configuration and rates of plate motion, subduction-zone deformation and associated seismogenesis,
and, as is becoming increasingly apparent, links between tectonically driven rock uplift and climate.
Arc sedimentary and volcanic records from many locations have been used to decipher tectonic evolu-
tion of modern and ancient convergent margins (e.g., Dickinson et al., 1982; Scholl et al., 1983; Tatsumi et
al., 1983; Ingersoll, 1983; Harbert et al., 1986; Dewey and Ryan, 1990; Gill et al., 1994; Fackler-Adams et
al., 1997; Trop et al., 2003; Busby et al., 2005; Clift et al., 2005; Huang et al., 2006), and geochemical studies
of arc sediment and volcaniclastic deposits have done much to illuminate possible origins of continental crust
(e.g., Pearcy et al., 1990; Miller and Christensen, 1994; Draut et al., 2002). Though much has been learned
from these records and their applications thus far, many important questions remain regarding controls on for-
mation of the sedimentary record in arc environments, where tectonically generated topography is typically
complex. A greater understanding of how and where accommodation space is created, and of depositional and
reworking processes in arc settings, is essential for determining the ability of arc sediment to preserve accu-
rately the tectonic and magmatic signals for which they are commonly studied (e.g., Marsaglia and Ingersoll,
1992; Busby and Ingersoll, 1995; Draut and Clift, 2006). Better comprehension of how arc sediment forms
and evolves will facilitate new and more advanced applications of arc stratigraphy as well as, perhaps, inhibit-
ing inferences for which the sedimentary record provides insufficient or inconclusive evidence.
This compilation of papers resulted indirectly from a Geological Society of America Penrose Con-
ference held in October 2005, co-sponsored by the British Sedimentological Research Group (Geological
Society of London) and the International Association of Sedimentologists. Approximately half of the papers
in this volume were authored by participants in that meeting, which focused on lessons in tectonics, climate,
and eustasy gained from the stratigraphic record preserved in arc collision zones.
The first set of papers in this collection focuses on formation and evolution of the sedimentary record in
arc settings and arc collision zones, concentrating on modern intra-oceanic examples. Two modeling papers
examine basin formation around intra-oceanic arcs: Ryan presents a model indicating that forearc-basin
topography can evolve as a function of eclogite formation, which causes basin subsidence (forming accom-
modation space) and may explain the problematic preservation of forearc basins during arc-continent colli-
sion and orogeny. Waltham et al. model flexural subsidence due to loading of arc and forearc lithosphere by
a volcanic arc, and find that the predicted loads are sufficient to account for much or all of basin subsidence
within 100 km of active arc volcanoes, a process suggested to be particularly common in older, mature arcs.
A related study of Indonesian arc stratigraphy by Hall and Smyth identifies key influences on development of
the sedimentary record in active volcanic arcs, including complexities of basin stratigraphy such as extension,
subduction polarity reversal, collision events, and fluctuating accommodation space associated with variable
loading of the lithosphere by the arc itself. Dorobek presents a comprehensive analysis of evolving carbonate
sedimentation processes around volcanic arcs through time. He shows that although voluminous volcanic
activity is harmful to reef growth, most of the volcaniclastic output of arc volcanoes is trapped in basins.
This allows carbonate reefs to act as sensitive recorders of vertical tectonic motions in forearc regions. Hoff-
mann et al., using new high-resolution seismic images from Papua New Guinea, examine the reworking of
volcaniclastic sediment around volcanic arcs; their study demonstrates that variations in sediment thickness
caused by the formation of current-driven bedforms warrant caution when comparing bed thicknesses across
sedimentary sections sampled in cores. Huang et al. explore mélange development in the active arc-continent
collision zone of Taiwan. Ogawa et al. review the tectonic development of the unique trench-trench-trench
triple junction between the Honshu and Izu Arcs, now located south of Tokyo. They also present a physical
analogue model showing formation of synthetic and antithetic faulting and topography within this oblique
arc collision zone and evaluate stress partitioning between the basement and overlying sediment cover.

v
vi Preface

The second half of the volume presents new applications of arc sedimentary records. Lambiase et al. use
new sedimentology, micropaleontology, and structural analysis of turbidite sequences in Borneo to interpret
them as an accretionary prism and examine in detail the syn-depositional deformation structures that occur
within such a system. Okuzawa and Hisada have reconstructed the tectonic evolution of collision between
the Izu and Honshu Arcs, Japan, using provenance of pre- and syn-collisional Miocene clastic deposits.
They focus on heavy-mineral studies of clinopyroxene, garnet, and spinel to show a switch in provenance in
the trench fill during the middle Miocene driven by initial uplift and erosion of the Honshu Arc after colli-
sion with the Izu arc massif. Smyth et al. trace the complete eruptive history of a well preserved Indonesian
Eocene–Miocene volcanic arc from initiation to termination, and present evidence for contamination of vol-
canism by Archean–Cambrian continental crust underneath the arc. These authors also note that the volume
of acidic arc volcanism has previously been underestimated because weathered acidic volcaniclastic material
was misinterpreted as terrigenous erosion products. Allen et al. use sediment provenance, 40Ar-39Ar mica
dating, and fission-track data to constrain the uplift history and sources of sediment entering the accretionary
prism of the Sunda subduction zone as recorded in the Andaman Islands. This study demonstrated that the
rising Himalaya did not contribute greatly to this accretionary wedge, but that erosion switched from being
local, arc-derived to continental Myanmar around 40 Ma. Clift et al. use a variety of geological and geophysi-
cal data sets to reconstruct exhumation and subsidence rates along the Ilan Basin, Taiwan (western limit of
the Okinawa Trough), to show that subduction polarity reversal associated with a migrating, steady-state
oblique collision between the Luzon arc and the passive margin of southern China causes rapid formation
of deep marginal basins along reactivated detachment faults that are filled by detritus from the orogen. In a
comprehensive summary paper, Centeno-Garcia et al. demonstrate the utility of arc lithofacies and isotopic
signatures to reconstruct the highly complex tectonic evolution of western Mexico, including multiple stages
of terrane collision and rifting. Zagorevski et al. present new U-Pb ages constraining timing of volcanism,
sedimentation, and the age of Ordovician arc-arc collision during the closure of the Iapetus Ocean and
formation of the Appalachian-Caledonide orogen, developing a tectonic model analogous to modern arc-
arc collisions in the southwestern Pacific. Grove et al. use new detrital zircon U-Pb age data to propose a
subduction-erosion origin for portions of the Catalina Schist in the western United States, in contrast to the
long-held theory that the metamorphic complex formed entirely during nascent subduction. The final two
papers provide new insights into possible tectonic-climate coupling in arc-collisional settings: Johnson et
al. use provenance, paleoclimatic, and stratigraphic data to construct a regional synthesis of Permian arc-
continent collision in Eurasia, encompassing the timing of tectonic events and implications for climatic
changes associated with orographic effects of regional uplift. Kimura et al. investigate intriguing feedback
mechanisms among orogenesis, surface erosion via the Asian monsoon, growth of the Nankai Trough accre-
tionary prism by incorporation of sediment eroded from the orogen, and generation of large earthquakes that
promote additional sediment erosion.
It is expected that the utility and applications of arc sedimentary records will expand considerably
with upcoming research initiatives such as those of the Integrated Ocean Drilling Program (IODP) that will
address processes in modern arc settings, including the Nankai Trough, Izu-Bonin forearc, and the collision
zone between the Yakutat terrane and North America, among others. New advances in analytical techniques
will improve age controls on tectonic evolution of active margins and, it is hoped, offer greater insight into
the links and feedback processes among collision-related uplift, precipitation, sediment flux to subduction
zones, and seismic activity with its associated societal hazards.

Amy E. Draut
Peter D. Clift
David W. Scholl
 Preface
REFERENCES CITED
Busby, C.J., and Ingersoll, R.V., eds., 1995, Tectonics of sedi-
mentary basins: Blackwell Science, 592 p.
Busby, C.J., Bassett, K., Steiner, M.B., and Riggs, N.R., 2005,
Climatic and tectonic controls on Jurassic intra-arc basins
related to northward drift of North America, in Anderson,
T.H., Nourse, J.A., McKee, J.W., and Steiner, M.B., eds.,
The Mojave-Sonora megashear hypothesis: Development,
assessment, and alternatives: Geological Society of Amer-
ica Special Paper 393, p. 359–376.
Clift, P.D., Pavlis, T., Debari, S.M., Draut, A.E., Rioux, M., and
Kelemen, P.B., 2005, Subduction erosion of the Juras-
sic Talkeetna-Bonanza arc and the Mesozoic accretion-
ary tectonics of western North America: Geology, v. 33,
p. 881–884.
Dewey, J.F., and Ryan, P.D., 1990, The Ordovician evolution of
the South Mayo trough, western Ireland: Tectonics, v. 9,
p. 887–901.
Dickinson, W.R., Ingersoll, R.V., Cowan, D.S., Helmond,
K.P., and Suczek, C.A., 1982, Provenance of Franciscan
graywackes in coastal California: Geological Society of
America Bulletin, v. 93, p. 95–107, doi: 10.1130/0016-
7606(1982)93<95:POFGIC>2.0.CO;2.
Draut, A.E., Clift, P.D., Hannigan, R.E., Layne, G., and Shi-
mizu, N., 2002, A model for continental crust genesis by
arc accretion: Rare earth element evidence from the Irish
Caledonides: Earth and Planetary Science Letters, v. 203,
p. 861–877, doi: 10.1016/S0012-821X(02)00931-7.
Draut, A.E., and Clift, P.D., 2006, Sedimentary processes in
modern and ancient oceanic arc settings: evidence from the
Jurassic Talkeetna Formation of Alaska and the Mariana
and Tonga arcs, western Pacific: Journal of Sedimentary
Research, v. 76, p. 493–514, doi: 10.2110/jsr.2006.044
Fackler-Adams, B.N., Busby, C.J., and Mattinson, J.M., 1997,
Jurassic magmatism and sedimentation in the Palen Moun-
tains, southeastern California: Implications for regional
tectonic controls on the Mesozoic magmatic arc: Geologi-
cal Society of America Bulletin, v. 109, p. 1464–1484, doi:
10.1130/0016-7606(1997)109<1464:JMASIT>2.3.CO;2.
Gill, J.B., Hiscott, R.N., and Vidal, P., 1994, Turbidite geo-
chemistry and evolution of the Izu-Bonin Arc and con-
tinents: Lithos, v. 33, p. 135–168, doi: 10.1016/0024-
4937(94)90058-2.
Harbert, W., Scholl, D.W., Vallier, T.L., Stevenson, A.J., and
Mann, D.M., 1986, Major evolutionary phases of a forearc
vii
basin of the Aleutian terrace: Relation to North Pacific tec-
tonic events and the formation of the Aleutian subduction
complex: Geology, v. 14, p. 757–761, doi: 10.1130/0091-
7613(1986)14<757:MEPOAF>2.0.CO;2.
Huang, C.-Y., Yuan, P.B., and Tsao, S.J., 2006, Temporal and
spatial records of active arc-continent collision in Taiwan:
A synthesis: Geological Society of America Bulletin,
v. 118, p. 274–288, doi: 10.1130/B25527.1.
Ingersoll, R.V., 1983, Petrofacies and provenance of Late Meso-
zoic forearc basin, northern and central California: The
American Association of Petroleum Geologists Bulletin,
v. 67, p. 1125–1142.
Marsaglia, K.M., and Ingersoll, R.V., 1992, Compositional
trends in arc-related, deep-marine sand and sandstone: A
reassessment of magmatic-arc provenance: Geological
Society of America Bulletin, v. 104, p. 1637–1649, doi:
10.1130/0016-7606(1992)104<1637:CTIARD>2.3.CO;2.
Miller, D.J., and Christensen, N.I., 1994, Seismic signature and
geochemistry of an island arc: A multidisciplinary study
of the Kohistan accreted terrane, northern Pakistan: Jour-
nal of Geophysical Research, v. 99, p. 11623–11642, doi:
10.1029/94JB00059.
Pearcy, L.G., Debari, S.M., and Sleep, N.H., 1990, Mass balance
calculations for two sections of island arc crust and impli-
cations for the formation of continents: Earth and Plane-
tary Science Letters, v. 96, p. 427–442, doi: 10.1016/0012-
821X(90)90018-S.
Scholl, D.W., Vallier, T.L., and Stevenson, A.J., 1983, Arc,
forearc, and trench sedimentation and tectonics; Amlia
corridor of the Aleutian Ridge, in Watkins, J.S., and
Drake, C.L., eds., Studies in continental margin geology:
American Association of Petroleum Geologists Memoir
34, p. 413–439.
Tatsumi, Y., Sakuyama, M., Fukuyama, H., and Kushiro, I.,
1983, Generation of arc basalt magmas and thermal struc-
ture of the mantle wedge in subduction zones: Journal of
Geophysical Research, v. 88, p. 5815–5825.
Trop, J.M., Ridgway, K.D., and Spell, T.L., 2003, Sedimentary
record of transpressional tectonics and ridge subduction
in the Tertiary Matanuska Valley–Talkeetna Mountains
forearc basin, southern Alaska, in Sisson, V.B., Roeske,
S.M., and Pavlis, T.L., eds., Geology of a transpressional
orogen during ridge-trench interaction along the North
Pacific margin: Geological Society of America Special
Paper 371, p. 89–118.
 The Geological Society of America
Special Paper 436
2008

Preservation of forearc basins during island arc–continent collision: Some insights from
the Ordovician of western Ireland

Paul D. Ryan*
Department of Earth and Ocean Sciences, National University of Ireland, Galway, Ireland

ABSTRACT

A new model is proposed for the problematic preservation of an Ordovician

forearc basin, which records a complete sedimentary record of arc-continent colli-
sion during the Grampian (Taconic) orogeny in the west of Ireland. The South Mayo
Trough represents an arc and forearc complex developed above a subduction zone
in which the slab dipped away from the Laurentian passive margin. The collision
of this arc with Laurentia caused the Middle Ordovician Grampian orogeny. How-
ever, the South Mayo Trough, in the hanging wall of this collision zone, remained a
site of marine sedimentation during the entire process. Early sediments show der-
ivation from an island-arc complex, an ophiolitic backstop, and polymetamorphic
trench sediments. These are conformably overlain by marine deposits derived from a
more evolved arc complex and an emerging juvenile orogen. This transition is dated
as being coeval with the Grampian metamorphism of the Laurentian footwall. The
problem remains as to why subsidence continued in a basin on the hanging wall. It is
proposed that the suppression of the expected topography is due to the nature of the
Laurentian continental margin. Geophysical and geological evidence suggests that
this was a volcanic margin during Neoproterozoic rifting. It is argued that the sub-
duction of this margin caused the formation of eclogites, which reduced its buoyancy.
Simple numerical models are presented which show that this is a viable mechanism
for the suppression of topography during early stages of arc-continent collision and
hence for the preservation of forearcs.

Keywords: forearc, preservation, Grampian Orogeny, arc-continent collision, eclogite.

INTRODUCTION
Island arc–continent collision is an important mechanism
for continental growth. However, proposed models for this pro-
cess are varied and complex. A forearc can collide with a passive
continental margin (Fig. 1A), initiating orogeny at that margin,
as suggested for the Middle Ordovician of South Mayo, Ireland
(Ryan and Dewey, 1991); the early Eocene of Kamchatka (Kon-
stantinovskaia, 2000, 2001); the Neoproterozoic of the Tuva-
Mongolia Massif (Kuzmichev et al., 2001) and of the Anti-Atlas
of Morocco (Thomas et al., 2002). This process can result in slab
break-off and a reversal of subduction polarity (McKenzie, 1969)
such as in South Mayo (Ryan and Dewey, 1991) or Taiwan (Teng,
1996; Clift et al., 2003) and the creation of an active continental

*paul.ryan@nuigalway.ie

Ryan, P.D., 2008, Preservation of forearc basins during island arc–continent collision: Some insights from the Ordovician of western Ireland, in Draut, A.E., Clift,
P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special Paper 436,
p. 1–9, doi: 10.1130/2008.2436(01). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All rights reserved.

1
2 Ryan
B C
D to the partial or complete subduction of
E F
margin (Fig. 1B). Alternatively, a new subduction zone can initi-
ate elsewhere in the plate to the rear of the arc (Fig. 1C), as has
been proposed for the late Archaean Wutaishan orogeny (Wang et
al., 1996). The forearc can be either thrust in its entirety over the
footwall, the preferred model for South Mayo (Ryan and Dewey,
1991; Dewey and Mange, 1999; Clift et al., 2004), or become
part of an imbricated stack driving subsidence in a foreland basin
developed upon the footwall as in Taiwan (van der Werff, 1995).
Alternatively, modeling shows that if the overriding oceanic plate
fails in the backarc or arc region, the arc and forearc can be wholly
or partly subducted (Fig. 1D) if they are not sufficiently buoyant
(Konstantinovskaia, 2000; Boutelier et al., 2003). This leads to an
early orogenic event, with little of the arc complex preserved and
perhaps ultrahigh pressure metamorphism in the continental mar-
gin. A forearc can also collide with an active margin (Fig. 1E), a
model used for the collision of the Kohistan arc with Asia during
the Late Cretaceous (Khan et al., 1997). This should produce a
climactic orogenic event following a phase of crustal growth by
addition of subduction-related magmatism. However, an alternate
model is that the Kohistan backarc collided with an active Asian
margin (Rolland et al., 2000; O’Brien, 2001; Robertson and Col-
lins, 2002).
In all such models the preservation potential of the forearc
is low. If obducted, it lies at high structural levels in the hanging
Figure 1. Sketches showing various pro-
posed mechanisms affecting the forearc
region during island arc–continent col-
lision: (A) Initial precollision configu-
ration. (B) Obduction of the arc and
forearc, followed by subduction polar-
ity reversal. (C) Obduction of the arc
and forearc, followed by initiation of a
new subduction zone elsewhere in the
system, which may have either polarity.
(D) Weakness in the arc or backarc leads
the forearc and arc without reversal of
sense of subduction with accompany-
ing ultrahigh pressure metamorphism
of the continental margin. (E) Two op-
posing magmatic arcs, or an island arc
and a magmatic arc collide. (F) Neither
forearc is obducted, and a transform
boundary is created.
wall of an orogeny and is likely to be removed rapidly by ero-
sion, especially if slab break-off occurs. Alternatively, it may be
subducted or dismembered by strike-slip faulting.
This contribution discusses the geology of western Ireland,
where an Ordovician island arc is believed to have collided with
the Laurentian passive margin (Ryan and Dewey, 1991), causing
the Grampian orogeny (475–462 Ma) prior to the final closure of
the Iapetus Ocean in Late Silurian times (420–416 Ma). Remark-
ably, a significant proportion of the forearc, including a possible
forearc basin (Dewey and Ryan, 1990), the South Mayo Trough, is
preserved in spite of the complex having been in the hanging wall
of the earliest orogenic phase. It will be argued that the geology
of the footwall to this system, the Laurentian continental margin,
was critical in the preservation of the South Mayo forearc.
ORDOVICIAN GEOLOGY OF WESTERN IRELAND
The main geological divisions of western Ireland are shown
in Figure 2. The rocks to the north of the Fair Head–Clew Bay
Line have a Grenvillian basement (Daly, 1996, 2001; Daly and
Flowerdew, 2005), overlain by Neoproterozoic sediments of the
Dalradian Supergroup, which were deposited from pre–720 Ma
to post–595 Ma (Condon and Prave, 2000). Mafic dikes cut the
basement, postdate all Grenvillian fabrics, and are isotopically
 Insights from the Ordovician of western Ireland 3

Figure 2. (A) Geology of western and northwestern Ireland. (B) Geology of South Mayo. Fm.—Formation; Gp.—Group; Cplx.—complex;
FCL—Fair Head–Clew Bay Line; SUF—Southern Upland Fault.

and geochemically matched with pre-deformational dikes in the
overlying Dalradian (Daly and Flowerdew, 2005). Both base-
ment and cover were affected by the Grampian orogeny between
475 Ma and 462 Ma (Daly and Flowerdew, 2005; Flowerdew et
al., 2005). In the extreme south the Dalradian Supergroup con-
tains mafic volcanics with crossite (Gray and Yardley, 1979).
This region is commonly interpreted as making up part of the
Laurentian passive margin to the Iapetus Ocean, which opened
in latest Neoproterozoic times, the Dalradian sediments having
been deposited during rifting and breakup of Rodinia (Strachan
and Holdsworth, 2000).
The Clew Bay Complex lies to the south of the Fair Head–
Clew Bay Line and is associated with a strong magnetic lineament
(Max and Riddihough, 1975; Max et al., 1983) and is regarded
as the westward continuation of the Highland Boundary Fault in
Scotland. The complex is highly sheared and comprises two main
components: a sedimentary mélange underlying most of Clew
Bay (Dewey and Ryan, 1990) and a dismembered ophiolite in
the south (Ryan et al., 1983). Heavy mineral (Dewey and Mange,
1999) and isotopic studies (Chew, 2003) support a mature con-
tinental provenance for the sediments. The Clew Bay Complex
has been variously interpreted as part of the Dalradian (Phillips,
1973), a lateral equivalent to the Highland Border Complex of
Scotland (Harper et al., 1989), a shear carpet formed beneath
advancing ophiolitic nappes (Dewey and Shackleton, 1984), or
an accretionary complex above a south-verging subduction zone
(Dewey and Ryan, 1990). The age of the complex is uncertain,
although a Cambrian–Ordovician age is likely on provenance
4 Ryan
grounds (Dewey and Mange, 1999), and it shares the same struc-
tural history as the Dalradian to the north with the S2 fabric in
both dated at ca. 460 Ma (Chew et al., 2003). Provenance stud-
ies also suggest a stratigraphic linkage between this complex and
the South Mayo Trough by Early to Middle Ordovician times
(Dewey and Ryan, 1990).
The South Mayo Trough (Fig. 2B) contains a thick sequence
of Lower to Middle Ordovician sediments that are preserved in a
large syncline, the Mweelrea-Partry Syncline. The northern limb
contains an Arenig-Llanvirn sequence that exceeds 9 km in thick-
ness in which early sediments show derivation from a primitive
island arc complex, an ophiolitic backstop, and polymetamorphic
trench sediments. These are conformably overlain by marine
clastic sequences derived from a more evolved arc complex and
an emerging juvenile (Grampian) orogen (Dewey and Mange,
1999). Associated volcanic complexes, which lie mainly in the
southern limb, are of primitive arc affinities near the base (Ryan
et al., 1980; Clift and Ryan, 1994; Draut et al., 2004) and show an
increasing proportion of Laurentian-derived melt upsection (Draut
et al., 2004). Dewey and Ryan (1990) interpreted the South Mayo
Trough as a forearc basin on the grounds that it lay between an arc
(volcanic complexes) and an accretionary prism (the Clew Bay
Complex) which in turn had been thrust over a passive continen-
tal margin, Laurentia, whose leading edge had been subducted. It
is remarkable that the South Mayo Trough, interpreted as lying
in the obducted hanging wall, apparently provides a continuous
record of sedimentation during Grampian deformation of the
Laurentian margin, interpreted as lying in the footwall.
The Dalradian rocks of Connemara (Fig. 1) are in fault con-
tact with the South Mayo Trough (Dewey et al., 1970; Ryan and
Archer, 1978) and were probably emplaced outboard of the oce-
anic terrane of South Mayo by Late Ordovician lateral motion
(Hutton, 1987). The southern boundary is a SE-verging thrust,
the Mannin Thrust (Leake et al., 1983), which emplaced migma-
titic metasediments and intrusives over evolved arc rocks dated
at 474.6 ± 5.5 Ma (Friedrich, 1998; Draut and Clift, 2002) that
show chemical similarities to volcanics within the South Mayo
Trough. This suggests that Connemara lay structurally beneath
an evolving arc by Llanvirn times (Draut and Clift, 2002). U/Pb
studies of syntectonic gabbros and later granitic melts date D2 of
the Grampian orogeny at 473 Ma, and post-D3 rapid uplift that
took place before 462 Ma (Friedrich et al., 1999a, b).
The Grampian (Taconic) orogeny was a Middle Ordovician
event that affected the Laurentian margin prior to final closure
of the Iapetus Ocean, resulting from the collision of an arc with
the Laurentian margin, followed by a subduction polarity reversal
(Dewey and Shackleton, 1984; Ryan and Dewey, 1991). Island
arc–continent collision began in Arenig times (Draut and Clift,
2001), with hard collision being reflected in an increase of conti-
nental component in the arc melt (Draut and Clift, 2001; Draut et
al., 2004). Provenance studies in the South Mayo Trough suggest
initiation in an oceanic forearc, with later sediments recording
the stripping of an evolved arc and a juvenile orogen (Dewey and
Mange, 1999; Clift et al., 2004). This change in provenance was
coeval with the emplacement of the D2 gabbros in Connemara at
473 Ma (Dewey and Mange, 1999). Radiometric evidence from
both North Mayo and Connemara suggests that the entire event
took <20 m.y., with initial collision taking ~10 m.y. and subduc-
tion flip a further 5 m.y. (Dewey and Mange, 1999; Soper et al.,
1999; Draut and Clift, 2001).
Evidence for the reversal of subduction polarity beneath
the Laurentian margin after the Grampian orogeny is found in
Middle to Upper Ordovician and Silurian strata of Connemara
and South Mayo. The South Connemara Complex, to the south
of the Southern Upland Fault (SUF, Fig. 2), was probably formed
in a Late Ordovician trench of opposite polarity adjacent to the
Laurentian margin (Ryan and Dewey, 2004). Silurian deposits
in north Connemara and South Mayo (Fig. 2) contain abundant
arc detritus (Menuge et al., 1995; Dewey and Mange, 1999)
and probably formed in a supra-subduction zone setting (Wil-
liams and Harper, 1991; Menuge et al., 1995). These deposits are
mostly shallow marine and of 2.5–3 km in thickness, suggesting
only modest stretching of a normally buoyant lithosphere without
topography (equivalent to a β factor of ~1.25). Critically, where
the Silurian sedimentary and volcanic sequences rest unconform-
ably upon the Ordovician rocks, the earliest slaty cleavage, if pres-
ent, cuts both sequences. Structural and stratigraphic evidence,
therefore, suggests that prior to Silurian deposition the Ordovi-
cian sequence of South Mayo was only modestly deformed, of
anchimetamorphic grade, and near sea level (Dewey and Ryan,
1990). If the arc-continent collision model is accepted for South
Mayo, a mechanism must, therefore, have operated to suppress
the topography of the forearc during and after this event.
ISOSTATIC CONSEQUENCES OF ARC-CONTINENT
COLLISION
The likely topographic response of an arc-continent colli-
sion is investigated in two ways. Firstly, a flexural isostatic model
is used to estimate the uplift of the hanging wall for a 10 m.y.
period from 485 to 475 Ma (Dewey and Mange, 1999; Draut and
Clift, 2002), when arc-continent collision was thought to have
taken place. Secondly, a simple model, assuming Airy isostasy,
is used to estimate the likely uplift during slab break-off, which
is assumed to have led to the emplacement of mafic plutons and
quartz-diorite gneisses in Connemara between 475 and 470 Ma
(Friedrich et al., 1999a, b). The aim is to investigate the condi-
tions that are required to keep the forearc basin beneath sea level
during both of these events.
The flexural response of the lithosphere during collision, but
prior to slab break-off, is estimated using a two-dimensional (2D)
kinematic flexural model. The present across-strike width of the
rocks affected by the Grampian orogeny (the Grampian and North-
Western terranes of Murphy et al., 1991) is ~100–140 km. Rocks
to the north in Donegal were metamorphosed to greenschist facies
(Pitcher and Berger, 1972), suggesting that the Grampian orogen
extended farther north. Strike swing and post-Grampian collapse
(Alsop, 1991) make it difficult to estimate the original width of
 Insights from the Ordovician of western Ireland 5

the orogen, but allowing for such factors, Dewey and Mange
(1999) suggest 200 km as a reasonable value. It is assumed that
an arc complex, including South Mayo, was emplaced over the
Laurentian margin along a thrust that nucleated upon the sub-
duction zone. A hanging wall of 210 km was moved over a flat
28 km in depth (ca. 0.8 GPa) for 100 km and then up a ramp at
an angle of 14.2° for a further 110 km (Fig. 3). A specific gravity
of 2200 kg m–3 was assumed for sediments in the flexural basin.
A value of 2840 kg m–3, the mean density of diorite, was used
for both the arc material of the hanging wall and a continental
basement containing up to 30% mafic material for the footwall.
The thrust was moved at 20 mm.yr –1 so that the deformation
front moved 200 km in 10 m.y. The actual rate of overthrusting
may have been greater (Dewey and Mange, 1999). The hang-
ing wall was moved along flow lines parallel to the ramp–flat
surface. Erosion was modeled using equation 5 below, and the
model was compensated using flexural isostasy (Watts, 2001)
after each increment of movement. Different elastic thicknesses
(Te) of 2.0 km (Fig. 3A) and 20.0 km (Fig. 3C) were used to
estimate the response of flexurally weak and flexurally strong
systems, respectively. The exact position of the South Mayo
forearc basin within such a system is unknown but is taken as
being between 100 and 200 km south of the thrust tip.
The model with low Te (Fig. 3A) predicts some 3 km of uplift
in the forearc region accompanied by ~1.5 km of erosion and a
flexural basin on the Laurentian margin with 0.5 km of sediment
(Fig. 3B). The model with high Te (Fig. 3C) predicts some 6 km
of uplift in the forearc region accompanied by 3–4 km of ero-
sion and a flexural basin on the Laurentian margin with 3.4 km of
sediment (Fig. 3D). No Ordovician flexural basin is recorded in
Ireland, nor is there any evidence for one farther north offshore
on the WIRELINES seismic reflection section (Klemperer et al.,
1991). Thus it is concluded that the overthrusting of the arc hang-
ing wall over the continental margin footwall was likely to have
taken place in a low Te regime. There still remains the problem
that if the assumptions made in this simple mechanical model

A
B

C
D

Figure 3. (A) Flexural isostatic model in which a hanging wall of 28 km is emplaced at 20 mm.yr –1 over a continental
footwall. The heavy upper line represents the erosion surface after 10 m.y. at the end of the model run. The dashed line
represents the thrust surface. The erosion constant is 5 m.y.–1, and the elastic thickness (Te) is 2 km. (B) Subsidence-time
curve for the foreland basin produced at the toe of the thrust system in (A). (C) A similar model to that in (A) but with Te
= 20 km. (D) Subsidence-time curve for the foreland basin produced at the toe of the thrust system in (C).
6 Ryan
are reasonable (see caption for Fig. 3), the South Mayo Trough
was likely to have been above sea level prior to slab break-off at
ca. 475 Ma.
A simple technique is employed to estimate the possible
topography associated with subsequent slab break-off. Fig-
ure 4A shows a typical tectonic cartoon of the process of island
arc–continent collision at the point of subduction reversal (based
upon Dewey and Mange, 1999). Airy compensation is assumed
in the light of the previous flexural analysis. Topography (h)
with respect to the summit of the oceanic ridges is given by (see
Dewey et al., 1993):
( ρ a − ρ l ) ( lz − h )
h= , (1)
ρc
where ρc, ρa, ρl are the specific gravities of the crust, astheno-
sphere, and lithosphere, respectively, and lz is the thickness of the
lithosphere. Where ρl may be estimated from:
ρl =
[(Cz − h) ρc + (lz − Cz ) ρm ],
• •
(2)
[lz − h]
B South Mayo speculatively positioned
where Cz is the thickness of the crust and ρm the specific gravity
of the mantle. The following values were assumed for the specific
gravities: ρc = 2840 kg.m–3; ρm = 3330 kg.m–3; and ρa = 3281 kg.
m–3. The value for ρa places the upper surface of the lithosphere
with Cz = 30 km and lz = 125 km at sea level.
If topography above sea level (d) is corrected for the mean
depth of the mid-ocean ridges we obtain:
d + 2.64 = h . (3)
When (2) and (3) are substituted in (1) and simplified,
we get:
d=
[Cz (ρm − ρc )+ lz (ρa − ρm )] − 2.64. (4)
ρa
Equation 4 can be used to estimate the topography of any
geodynamic cartoon if the thicknesses of crust and mantle are
measured at regular intervals across the model (Fig. 4). Although
this may differ from dynamic topography, it provides a simple
Figure 4. (A) Isostatic consequences
of an arc-continent collision and sub-
duction polarity flip model taken from
Dewey and Mange (1999). The heavy
solid line in the upper topographic grid
shows the maximum topography pro-
duced by this model, assuming Airy
compensation; the dashed line shows
the likely topography after 10 Ma ero-
sion at an erosion constant of 5 m.y.–1.
(B) Isostatic consequences of both the
model in Figure 4A and the thrust mod-
els from Figure 3A, with the geology of
onto the hanging wall, assuming lower
crustal-phase changes. The heavy lines
in the upper topographic grid show
the likely maximum topography for
the model in Figure 3A, assuming the
lower crust below 40 km is either 100%
eclogite (solid line) or 50% eclogite
(dashed line). The lighter lines show the
maximum topography of the tectonic
model in Figure 4A, assuming that the
lower crust below 40 km is either 100%
eclogite (solid line) or 50% eclogite
(dashed line).
 Insights from the Ordovician of western Ireland
check on the admissibility of any such cartoon. The height of a
feature zt originally at z0, allowing for subsequent erosion during
a given interval (t m.y.), can be estimated using:
ρc
⎛ . k .t ⎞ in some regions (Pitcher and Berger, 1972). Correlation with the
z t = z0 ⎜ 1 − e ρm ⎟,
⎜ ⎟
⎝ ⎠
where k is an erosion constant (m.y.–1). Values for k vary from
1 m.y.–1 for Midwestern basins in the United States to 5 m.y.–1 for
the Moroccan Rif and 20 m.y.–1 for the New Zealand Alps (esti-
mated from data in Goudie, 1995). The topography associated
with slab break-off and the remaining topography after 10 m.y.
are shown in Figure 4A. Approximately 5 km of material would
have been removed from the forearc region, assuming an erosion
constant k = 5 m.y.–1 (Fig. 4A).
This analysis suggests that the South Mayo Trough, or any
other similar forearc complex, would have been unlikely to sur-
vive arc-continent collision followed by slab break-off unless the
topography was depressed by some other factor. It is unlikely
that this was due to some dynamic process, as the South Mayo
region became a site of Silurian marine sedimentation follow-
ing the Grampian orogeny in basins with modest β factors (see
above), suggesting that the crust was never far above sea level in
the post-Grampian interval.
The model suggests uplift rates in the forearc region on
the order of 2 mm.yr –1 to 5 mm.yr –1 during collision and slab
break-off. Similar values of 5 mm.yr –1 are recorded in Sumba
where a forearc basin is overriding the Australian margin (van
der Werff, 1995); this is a region that has been suggested as an
analogy for the Grampian orogeny (Dewey and Mange, 1999).
However, if such rates were sustained over the duration of the
Grampian orogeny, it is unlikely that the South Mayo Trough
would have been preserved. Some feature of the Grampian col-
lision zone must, therefore, have depressed the topography of
this orogen. Three possibilities exist. Firstly, South Mayo may
not have been involved in the Grampian arc-continent collision,
but the abundance of evidence reviewed above makes this seem
highly unlikely. Secondly, slab pull may have initially reduced
the topography dynamically. However, the emplacement of mafic
plutons during the principal D2/D3 Grampian deformation in
Connemara is generally believed to have been related to delami-
nation of the downgoing Iapetan slab at the time of, not after,
peak deformation. A third possibility is that the nature of the con-
tinental basement in the footwall may have buffered rapid uplift.
ROLE OF THE LAURENTIAN BASEMENT
Deep reflection seismic studies (WIRELINE, Klemperer et
al., 1991) show that the basement north of the Fair Head–Clew
Bay Line off the shore of western and northwestern Ireland is
anomalously reflective between ~1.2 s two-way traveltime
(TWTT) and the Moho at ~10 s TWTT. Klemperer et al. (1991)
attributed this to the presence of considerable volumes of basaltic
7
material within the basement of either Tertiary or Neoproterozoic
age. Mafic dikes and sills intruded in Argyll Group rocks occur
throughout the Dalradian sequence and are particularly preva-
lent in Donegal, composing up to 20% of the surface geology
(5) Tayvallich lavas of Scotland suggest that they were emplaced at
ca. 595 Ma (Halliday et al., 1989). The Dalradian in northwestern
Ireland is associated with positive, moderate-frequency magnetic
anomalies (Armstrong et al., 1998), which are interpreted as hav-
ing been caused by these mafic bodies. In Donegal Bay, Carbon-
iferous deposits overstep the Dalradian metasediments, and the
associated broad, negative magnetic anomalies mask the pattern
associated with the Dalradian rocks. This relationship would not
have occurred if the mafic rocks had been post-Carboniferous.
Furthermore, the WIRELINE survey did not pass close to any
known Tertiary seamounts. Thus it is most likely that the base-
ment reflectivity is associated with the Neoproterozoic basalt
swarms emplaced during rifting in Argyll Group times and that
in this region the Laurentian margin was a volcanic margin.
The presence of voluminous basalt within the deep crust
provides a mechanism for depressing topography during con-
vergence (Dewey et al., 1993). If mafic rocks convert to eclo-
gite, their specific gravity increases by up to 20%, and other
continental rocks on converting to their eclogite facies equiva-
lents will show increases in specific gravity of between 5% and
10% (Dewey et al., 1993). This is likely to occur below 40 km
in a refrigerated system such as a subducting continental mar-
gin. The effect of any such conversion to eclogite facies for the
lower crust is to increase the overall specific gravity ρc' of the
crust as follows:
ρe ∗ Ez + ρc ∗ (Cz − Ez )
ρc' = , (6)
Cz
where ρe is the density of the eclogitized crust, and Ez is the
thickness of the eclogitic layer. Figure 4B shows the effect of
substituting the modified values for ρc' into equation 4 on the
assumption that all the crust in the collision zone beneath 40 km
is eclogitized and that this region has a basaltic component of
either 100% or 50%.
The analysis is performed for two possible scenarios.
The first is the model of Dewey and Mange (1999), shown in
Figure 4A, which assumes that the crust on the leading edge of
the Laurentian margin is thickened during convergence. The sec-
ond is based upon the flexural model shown in Figure 3A, which
assumes that the Laurentian crust is thick beneath the overriding
forearc but thins toward the ocean-continent boundary. The first
model predicts that the forearc region will have a mean topogra-
phy of a few hundred meters above sea level. The frontal region
of the arc would be ~4 km below sea level as a result of the nega-
tive buoyancy attributable to the thickened eclogite-facies lower
crust in this region. The presence of lherzolite within the litho-
spheric mantle would also depress topography as the depth of
8 Ryan
the mantle slab increases and the aluminous phases convert from
spinel to garnet below 90 km. However, this effect is small. If
a 10% lherzolite component containing 10% spinel is assumed,
the maximum depth in this model (Fig. 4B) would increase by
30 m on the southern side of the forearc basin. The second model
suggests that the forearc region could be ~200 m below sea level
but that much of the frontal arc region would be above sea level
if there was 100% eclogite. The second model seems to provide
a better fit to the geology of South Mayo, where sedimentation
is believed to have continued during collision and slab break-off.
However, given the uncertainties in any such analysis, it is diffi-
cult to differentiate between the two models. The important point
to note is that significant production of eclogite within the foot-
wall to the collision zone can depress the topography sufficiently
to allow sedimentation to continue in the forearc region during
arc-continent collision.
DISCUSSION AND CONCLUSIONS
It is suggested that the production of significant volumes of
eclogite at the base of the footwall during arc-continent collision
may depress the topography sufficiently to preserve the forearc
within the hanging wall. This process is likely to be important
only along volcanic margins or those that contain older basic
massifs. In this context, it is worthy of note that in the Anti-Atlas
orogen, where a forearc basin (Sarhro Group) has been preserved
during arc-continent collision, the continental basement contains
dike swarms and mafic volcanics (Ifzwane Suite) related to rift-
ing (Thomas et al., 2002).
This analysis has the limitations of any 2D static model.
It cannot, for instance, distinguish between the effects of slab
break-off along the length of a continental margin or gradual
tearing (Clift et al., 2003). However, any out-of-plane effects
of a tearing slab should assist in depressing topography. There
also remains the problem that a root of eclogites has not been
recorded in any deep geophysical survey of western Ireland.
However, the Grampian orogen underwent extensive orogenic
collapse in the Late Ordovician (Alsop, 1991; Clift et al., 2004)
during which time such a root would have been exhumed and
retrogressed to amphibolite. The buoyancy gained during the
decompression of the eclogites could have buffered topography
and limited subsidence during this phase (Dewey et al., 1993).
Such a mechanism may account for the cessation of sedimenta-
tion within the South Mayo Trough. An alternative explanation
would be to apply the model of Jull and Kelemen (2001) and
suggest that the crustal eclogitic root beneath Laurentia became
detached and sank into the mantle in late post-Grampian times.
This would have driven rapid uplift in the arc region of the model
in Figure 4A and in the accretionary complex in the model of
Figure 4B. Both regions were sites of Silurian sedimentation
(Fig. 2) onto a low-metamorphic-grade basement, making such
a process unlikely, at least beneath the South Mayo Trough.
REFERENCES CITED
Alsop, G.I., 1991, Gravitational collapse and extension along a mid crustal
detachment: The Lough Derg Slide, northwest Ireland: Geological Maga-
zine, v. 128, p. 345–354.
Armstrong, G.D., Brown, C., and Mars, J.E., 1998, The Irish continental
crust revealed by potential field methods: Technical Program Extended
Abstracts, Society of Exploration Geophysicists, 68th Annual Meet-
ing, New Orleans, Louisiana, September 1998; CD-ROM and Extended
Abstracts Volume.
Boutelier, D., Chemenda, A., and Burg, J.-P., 2003, Subduction versus accretion
of intra-oceanic volcanic arcs: Insight from thermo-mechanical analogue
experiments: Earth and Planetary Science Letters, v. 212, p. 31–45, doi:
10.1016/S0012-821X(03)00239-5.
Chew, D.M., 2003, Structural and stratigraphic relationships across the con-
tinuation of the Highland Boundary Fault in western Ireland: Geological
Magazine, v. 140, p. 73–85, doi: 10.1017/S0016756802007008.
Chew, D.M., Daly, J.S., Page, L.M., and Kennedy, M.J., 2003, Grampian oro-
genesis and the development of blueschist facies metamorphism in west-
ern Ireland: Geological Society [London] Journal, v. 160, p. 911–924.
Clift, P.D., and Ryan, P.D., 1994, Geochemical evolution of an Ordovician
island arc, South Mayo, Ireland: Geological Society [London] Journal,
v. 151, p. 329–342.
Clift, P.D., Schouten, H., and Draut, A.E., 2003, A general model of arc conti-
nent collision and subduction polarity reversal from Taiwan and the Irish
Caledonides, in Larter, R., and Leat, P., eds., Intra-Oceanic Subduction
Systems; Tectonic and Magmatic Processes: Geological Society [Lon-
don] Special Publication 219, p. 81–98.
Clift, P.D., Dewey, J.F., Draut, A.E., Chew, D.M., Mange, M., and Ryan, P.D.,
2004, Rapid tectonic exhumation, detachment faulting and orogenic
collapse in the Caledonides of western Ireland: Tectonophysics, v. 384,
p. 91–113, doi: 10.1016/j.tecto.2004.03.009.
Condon, D.J., and Prave, A.R., 2000, Two from Donegal: Neoproterozoic epi-
sodes on the northeast margin of Laurentia: Geology, v. 28, p. 951–954,
doi: 10.1130/0091-7613(2000)28<951:TFDNGE>2.0.CO;2.
Daly, J.S., 1996, Pre-Caledonian history of the Annagh Gneiss Complex, north-
western Ireland, and correlation with Laurentia-Baltica: Irish Journal of
Earth Sciences, v. 15, p. 5–8.
Daly, J.S., 2001, The Precambrian, in Holland, C.H., ed., The Geology of Ire-
land (2nd edition): Edinburgh, Scottish Academic Press, p. 7–45.
Daly, J.S., and Flowerdew, M.J., 2005, Grampian and late Grenville events
recorded by mineral geochronology near a basement cover contact in north
Mayo, Ireland: Geological Society [London] Journal, v. 162, p. 163–174,
doi: 10.1144/0016-764903-150.
Dewey, J., and Mange, M., 1999, Petrography of Ordovician and Silurian sedi-
ments in the western Ireland Caledonides: Tracers of a short-lived Ordovi-
cian continent arc collision orogeny and the evolution of the Laurentian
Appalachian Caledonian margin, in MacNiocaill, C., and Ryan, P.D., eds.,
Continental Tectonics: Geological Society [London] Special Publication
164, p. 55–107.
Dewey, J.F., and Ryan, P.D., 1990, The Ordovician evolution of the South Mayo
Trough, western Ireland: Tectonics, v. 9, p. 887–901.
Dewey, J.F., and Shackleton, R.M., 1984, A model for the evolution of the
Grampian tract in the early Caledonides and Appalachians: Nature, v. 312,
p. 115–121, doi: 10.1038/312115a0.
Dewey, J.F., McKerrow, W.S., and Moorbath, S., 1970, The relationship
between isotopic ages, uplift and sedimentation during Ordovician times
in western Ireland: Scottish Journal of Geology, v. 6, p. 133–145.
Dewey, J.F., Ryan, P.D., and Andersen, T.B., 1993, Orogenic uplift and col-
lapse, crustal thickness, fabrics and metamorphic phase changes: The
role of eclogites, in Prichard, H.M., et al., eds., Magmatic Processes and
Plate Tectonics: Geological Society [London] Special Publication 76,
p. 325–343.
Draut, A.E., and Clift, P.D., 2001, Geochemical evolution of arc magmatism dur-
ing arc-continent collision, South Mayo, Ireland: Geology, v. 29, p. 543–
546, doi: 10.1130/0091-7613(2001)029<0543:GEOAMD>2.0.CO;2.
Draut, A.E., and Clift, P.D., 2002, The origin and significance of the Delaney
Dome Formation, Connemara, Ireland: Geological Society [London]
Journal, v. 159, p. 95–103.
 Insights from the Ordovician of western Ireland
Draut, A.E., Clift, P.D., Chew, D.M., Cooper, M.J., Taylor, R.N., and Hannigan,
R.E., 2004, Laurentian crustal recycling in the Ordovician Grampian Orog-
eny: Nd isotopic evidence from South Mayo, western Ireland: Geological
Magazine, v. 141, p. 195–207, doi: 10.1017/S001675680400891X.
Flowerdew, M.J., Daly, J.S., and Whitehouse, M.J., 2005, 470 Ma granitoid
magmatism associated with the Grampian Orogeny in the Slishwood
Division, NW Ireland: Geological Society [London] Journal, v. 162,
p. 563–575, doi: 10.1144/0016-784904-067.
Friedrich, A.M., 1998, 40Ar/39Ar and U-Pb geochronological constraints on
the thermal and tectonic evolution of the Connemara Caledonides [Ph.
D. thesis]: Cambridge, Department of Earth, Atmospheric and Planetary
Sciences, Massachusetts Institute of Technology, 228 p.
Friedrich, A.M., Hodges, K.V., Bowring, S.A., and Martin, M.W., 1999a, Geo-
chronological constraints on the magmatic, metamorphic, and thermal
evolution of the Connemara Caledonides, western Ireland: Geological
Society [London] Journal, v. 156, p. 1217–1230.
Friedrich, A.M., Bowring, S.A., Martin, M.W., and Hodges, K.V., 1999b, Short-
lived continental magmatic arc at Connemara, western Irish Caledonides:
Implications for the age of the Grampian orogeny: Geology, v. 27, p. 27–
30, doi: 10.1130/0091-7613(1999)027<0027:SLCMAA>2.3.CO;2.
Goudie, A.S., 1995, The Changing Earth: Rates of Geomorphological Pro-
cesses: London, Blackwell, 302 p.
Gray, J.R., and Yardley, B.W.D., 1979, A Caledonian blueschist from the Irish
Dalradian: Nature, v. 278, p. 736–737, doi: 10.1038/278736a0.
Halliday, A.N., Graham, C.R., Aftalion, M., and Dymoke, P., 1989, The depo-
sitional age of the Dalradian Supergroup: U-Pb and Sm-Nd isotopic stud-
ies of the Tayvallich Volcanics, Scotland: Geological Society [London]
Journal, v. 146, p. 3–6.
Harper, D.A.T., Williams, D.M., and Armstrong, H.A., 1989, Stratigraphical
correlations adjacent to the Highland Boundary Fault in the west of Ire-
land: Geological Society [London] Journal, v. 146, p. 381–384.
Hutton, D.H.W., 1987, Strike slip terranes and a model for the evolution of the
British and Irish Caledonides: Geological Magazine, v. 124, p. 405–425.
Jull, M., and Kelemen, P.B., 2001, On the conditions for lower crustal convec-
tive instability: Journal of Geophysical Research, v. 106, p. 6423–6446,
doi: 10.1029/2000JB900357.
Khan, M.A., Stern, R.J., Gribble, R.F., and Windley, B.F., 1997, Geochemical
and isotopic constraints on subduction polarity, magma sources and pal-
aeogeography of the Kohistan Arc, northern Pakistan: Geological Society
[London] Journal, v. 154, p. 935–946.
Klemperer, S.L., Ryan, P.D., and Snyder, D.B., 1991, A deep seismic reflec-
tion transect across the Irish Caledonides: Geological Society [London]
Journal, v. 148, p. 149–164.
Konstantinovskaia, E.A., 2000, Geodynamics of an Early Eocene arc–continent
collision reconstructed from the Kamchatka Orogenic Belt, NE Russia:
Tectonophysics, v. 325, p. 87–105, doi: 10.1016/S0040-1951(00)00132-3.
Konstantinovskaia, E.A., 2001, Arc–continent collision and subduction rever-
sal in the Cenozoic evolution of the Northwest Pacific: An example from
Kamchatka (NE Russia): Tectonophysics, v. 333, p. 75–94, doi: 10.1016/
S0040-1951(00)00268-7.
Kuzmichev, A.B., Bibikova, E.V., and Zhuravlev, D.Z., 2001, Neoprotero-
zoic (~800 Ma) orogeny in the Tuva-Mongolia Massif (Siberia): Island
arc–continent collision at the northeast Rodinia margin: Precambrian
Research, v. 110, p. 109–126, doi: 10.1016/S0301-9268(01)00183-8.
Leake, B.E., Tanner, P.W.G., Singh, D., and Halliday, A.N., 1983, Major south-
ward thrusting of the Dalradian rocks of Connemara, western Ireland:
Nature, v. 305, p. 210–213, doi: 10.1038/305210a0.
Max, M.D., and Riddihough, R.P., 1975, Continuation of the Highland Bound-
ary Fault in Ireland: Geology, v. 3, p. 206–210, doi: 10.1130/0091-
7613(1975)3<206:COTHBF>2.0.CO;2.
Max, M.D., Ryan, P.D., and Inamdar, D.D., 1983, A magnetic deep structural
geology interpretation of Ireland: Tectonics, v. 2, p. 431–451.
McKenzie, D.P., 1969, Speculations on the consequences and causes of plate
motions: Geophysical Journal of the Royal Astronomical Society, v. 18,
p. 1–32.
Menuge, J.F., Williams, D.M., and O’Connor, P.D., 1995, Silurian turbidites
used to reconstruct a volcanic terrain and its Mesoproterozoic basement
in the Irish Caledonides: Geological Society [London] Journal, v. 152,
p. 269–278.
9
Murphy, F.C., Anderson, T.B., Daly, J.S., Gallagher, V., Graham, J.R., Harper,
D.A.T., Johnston, J.D., Kennan, P.S., Kennedy, M.J., Long, C.B., Morris,
J.H., O’Keefe, W.G., Parkes, M., Ryan, P.D., Sloan, R.J., Stillman, C.J.,
Tietzsch-Tyler, D., Todd, S.P., and Wrafter, J.P., 1991, An appraisal of
Caledonian suspect terranes in Ireland: Irish Journal of Earth Sciences,
v. 11, p. 11–41.
O’Brien, P.J., 2001, Subduction followed by collision: Alpine and Himalayan
examples: Physics of the Earth and Planetary Interiors, v. 127, p. 277–
291, doi: 10.1016/S0031-9201(01)00232-1.
Phillips, W.E.A., 1973, The pre-Silurian rocks of Clare Island, Co. Mayo, Ire-
land, and the age of metamorphism of the Dalradian in Ireland: Geologi-
cal Society [London] Journal, v. 129, p. 585–606.
Pitcher, W.S., and Berger, A.R., 1972, The Geology of Donegal: A Study of
Granite Emplacement and Unroofing: New York, John Wiley, 435 p.
Robertson, A.H.F., and Collins, A.S., 2002, Shyok Suture Zone, N Pakistan:
Late Mesozoic-Tertiary evolution of a critical suture separating the oce-
anic Ladakh Arc from the Asian continental margin: Journal of Asian
Earth Sciences, v. 20, p. 309–351, doi: 10.1016/S1367-9120(01)00041-4.
Rolland, Y., Pecher, A., and Picard, C., 2000, Middle Cretaceous back-arc for-
mation and arc evolution along the Asian margin: The Shyok Suture Zone
in northern Ladakh (NW Himalaya): Tectonophysics, v. 325, p. 145–173,
doi: 10.1016/S0040-1951(00)00135-9.
Ryan, P.D., and Archer, J.B., 1978, The Lough Nafooey fault: A Taconic
structure in western Ireland: Geological Survey of Ireland Bulletin, v. 2,
p. 255–264.
Ryan, P.D., and Dewey, J.F., 1991, A geological and tectonic cross-section of
the Caledonides of western Ireland: Geological Society [London] Journal,
v. 148, p. 173–180.
Ryan, P.D., and Dewey, J.F., 2004, The South Connemara Group reinterpreted:
A subduction-accretion complex in the Caledonides of Galway Bay, west-
ern Ireland: Journal of Geodynamics, v. 37, p. 513–529, doi: 10.1016/
j.jog.2004.02.018.
Ryan, P.D., Floyd, P.A., and Archer, J.B., 1980, The stratigraphy and petro-
chemistry of the Lough Nafooey Group (Tremadocian), western Ireland:
Geological Society [London] Journal, v. 137, p. 443–458.
Ryan, P.D., Sawal, V.K., and Rowlands, A.S., 1983, Ophiolitic mélange sepa-
rates ortho- and para-tectonic Caledonides in western Ireland: Nature,
v. 302, p. 50–52, doi: 10.1038/302050a0.
Soper, N.J., Ryan, P.D., and Dewey, J.F., 1999, Age of the Grampian Orogeny
in Scotland and Ireland: Geological Society [London] Journal, v. 148,
p. 173–180.
Strachan, R.A., and Holdsworth, R.E., 2000, Late Neoproterozoic (<750 Ma)
to Early Ordovician passive margin sedimentation along the Laurentian
margin of Iapetus, in Woodcock, N., and Strachan, R., eds., Geological
History of Britain and Ireland: London, Blackwell, p. 73–87.
Teng, L.S., 1996, Extensional collapse of the northern Taiwan mountain belt:
Geology, v. 24, p. 949–952, doi: 10.1130/0091-7613(1996)024<0949:
ECOTNT>2.3.CO;2.
Thomas, R.J., Chevallier, L.P., Gresse, P.G., Harmer, R.E., Eglington, B.M.,
Armstrong, R.A., de Beer, C.H., Martini, J.E.J., de Kock, G.S., Macey,
P.H., and Ingram, B.A., 2002, Precambrian evolution of the Sirwa Win-
dow, Anti-Atlas Orogen, Morocco: Precambrian Research, v. 118, p. 1–
57, doi: 10.1016/S0301-9268(02)00075-X.
van der Werff, W., 1995, Cenozoic evolution of the Savu Basin, Indonesia:
Fore-arc basin response to arc-continent collision: Marine and Petroleum
Geology, v. 12, p. 247–262, doi: 10.1016/0264-8172(95)98378-I.
Wang, K.Y., Li, J.L., Hao, J., Li, J., and Zhou, S., 1996, The Wutaishan oro-
genic belt within the Shanxi Province, northern China: A record of late
Archaean collision tectonics: Precambrian Research, v. 78, p. 95–103,
doi: 10.1016/0301-9268(95)00071-2.
Watts, A.B., 2001, Isostasy and Flexure of the Lithosphere: Cambridge, Eng-
land, Cambridge University Press, 458 p.
Williams, D.M., and Harper, D.A.T., 1991, End-Silurian modifications of Ordo-
vician terranes in western Ireland: Geological Society [London] Journal,
v. 148, p. 165–171.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Basin formation by volcanic arc loading

Dave Waltham
Robert Hall
Helen R. Smyth
Cynthia J. Ebinger
SE Asia Research Group, Department of Geology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK

ABSTRACT

This paper quantifies the flexural subsidence expected from loading by a volcanic
arc. The resulting mathematical model shows that the arc width should grow with
time and that the subsidence beneath the load can be estimated from the observed
arc width at the surface. Application of this model to the Halmahera Arc in Indonesia
shows an excellent fit to observations if a broken-plate model of flexure is assumed.
The model also gives an excellent fit to data from East Java, also in Indonesia, where it
is possible to forward model gravity anomalies. In particular, the depth, location, and
width of the depocenter-associated gravity low are accurately reproduced, although
the model does require a high density for the volcanic arc (2900 kg m–3). This may
indicate additional buried loads due, for example, to magmatic underplating. Our
main conclusion is that loads generated by the volcanic arc are sufficient to account
for much, if not all, of the subsidence in basins within ~100 km of active volcanoes at
subduction plate boundaries, if the plate is broken. The basins will be asymmetrical
and, close to the arc, will contain coarse volcaniclastic material, whereas deposits far-
ther away are likely to be volcaniclastic turbidites. The density contrast between arc
and underlying crust required to produce the Indonesian arc basins means that they
are unlikely to form in young intraoceanic arcs but may be common in older and more
mature arcs.

Keywords: volcanic arc, loading, flexural subsidence, Indonesian arcs.

INTRODUCTION mechanism responsible for basin formation. Dewey (1980) sug-

gested that these basins result from extension caused by rollback
This paper investigates whether the isostatic load of volcanic of the subduction hinge; i.e., the location of subduction moves
arcs could be responsible for a significant fraction of the subsid- progressively away from the overriding plate and so drags the
ence frequently observed in volcanic arc settings. Arc volcanism overriding plate with it. Another suggestion is that extensional
occurs in the overriding plate of a subduction system by melt- stresses are set up by secondary mantle currents created by sub-
ing of the mantle wedge owing to introduction of volatiles car- duction (Toksöz and Bird, 1977). It is also possible that basins
ried beneath it by the subducting plate. Basins commonly form result from loading by magmatic underplating, as has been sug-
both trenchward of the arc (the forearc) and behind the arc (the gested for loading of oceanic lithosphere by volcanic islands
backarc). However, there is no consensus concerning the precise (Watts et al., 1985). Another possibility (Bahlburg and Furlong,
Waltham, D., Hall, R., Smyth, H.R., and Ebinger, C.J., 2008, Basin formation by volcanic arc loading, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation
and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special Paper 436, p. 11–26, doi: 10.1130/2008.2436(02). For
permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All rights reserved.

11
12 Waltham et al.
1996; Smith et al., 2002) is that the arc volcanism itself produces
significant near-surface loads leading to the formation of flex-
ural basins. This last suggestion is the one pursued in this paper,
although we emphasize that basins form in arcs by a number of
mechanisms, and we are not proposing that all basins form in this
way. The model has particular relevance to basins that form very
close to the volcanic arc. The proximal parts of such basins will
be characterized by coarse volcanic debris of primary volcanic
and sedimentary origin, typically terrestrial to shallow marine,
and may include mass-flow deposits of different types.
The idea that the lithosphere flexes in response to volcanic
loads is not new, although it has most commonly been applied
to loading of oceanic lithosphere by island chains associated
with hot-spot volcanism (Watts and Cochran, 1974; Watts et al.,
1997). Bahlburg and Furlong (1996) applied a continuous-plate
flexure model to subsidence in the Ordovician foreland of north-
western Argentina. Their model used geographically extensive
volcanic loads and was able to produce >8 km of subsidence. In
a similar study using the same modeling algorithm (attributed to
Slingerland et al., 1994), Smith et al. (2002) modeled subsidence
of the Abiquiu Embayment in the Rio Grande Rift, southwestern
USA. The Rio Grande study also employed a volcanic load that
filled most of the resulting basin, but the calculations of Smith
et al. (2002) accounted for only 800 m of subsidence in that
area. Other authors (e.g., Karner and Watts, 1983; Nunn et al.,
1987) applied similar models to continental subsidence resulting
from thrust-emplaced loads on an unbroken plate. These studies
showed that gravity profiles were consistent with flexural subsid-
ence, although, in most cases, additional subsurface loads were
required to explain the observed subsidence.
In this paper we produce a semianalytical model relating the
observed surface width of the loading volcanic arc to the result-
ing subsidence. We also look at the consequences of assuming a
broken-plate rather than a continuous-plate model for the under-
lying lithosphere. Our approach principally differs from that
employed by others in that it predicts the volcanic load from
first principles, thus providing constraints on loading that are
independent of those obtained, for example, from gravity mod-
eling. Such an approach is difficult for geometrically complex
thrust-emplaced loads (Nunn et al., 1987) but, as shown here, is
relatively straightforward for a volcanic arc load.
EVIDENCE FOR ARC LOADING IN INDONESIA
In many volcanic arcs, thick sequences of sedimentary and
volcaniclastic rocks in basins are close to, and on both sides of,
the volcanic arc. Much, if not all, of the material in the basins
is derived from the arc itself, and much of it is very coarse. In
Indonesia (Fig. 1), many basins near modern volcanic arcs have
proved rich in hydrocarbons and consequently have been the tar-
get of exploration work. The North, Central, and South Sumatra
Basins, and the offshore Northwest Java and East Java Basins are
examples. However, these basins are typically >100 km from the
active volcanic arc and have been described as backarc basins
(e.g., Busby and Ingersoll, 1995), although they are not floored
by oceanic crust and lack many features of backarc basins (Smyth
et al., 2007, this volume). Nor are they retro-arc foreland basins
formed in response to thrusting and dynamic subsidence driven
by subduction (e.g., DeCelles and Giles, 1996), as pointed out by
Moss and McCarthy (1997). The Sunda Shelf basins have many
characteristics of rift basins (Cole and Crittenden, 1997; Hall and
Morley, 2004). However, the basins we are concerned with are
much closer to the arc itself. Seismic lines cross many of these
basins, but it is rare for them to approach close to the volcanic arc
because the volcanic-rich sequences are normally considered to
be relatively unprospective as far as hydrocarbon source rocks are
concerned. Also, potential reservoir rocks commonly have prob-
lems with loss of porosity and permeability owing to breakdown
of unstable grains, and close to the arcs, seismic data are difficult
to acquire and interpret. We have studied a number of Indone-
sian volcanic arcs in the field and have worked on the sequences
in basins close to the volcanic arcs. The ages and histories of
the basins suggest that their development is related closely to the
development of the volcanic arc. We have chosen two different
arcs in Indonesia for which the history of the basin and the arc is
known, and where volcanic loading is a plausible explanation of
at least some of the basin subsidence. These are South Halmahera
and East Java (Fig. 1), where the arcs are built on different types
of crust.
South Halmahera
Halmahera has a long volcanic arc history extending from
the Late Jurassic, almost entirely intraoceanic (Hall et al., 1995).
The present arc is in its final stages of activity, as subduction has
nearly eliminated the Molucca Sea, and the Sangihe and Halma-
hera Arcs (Fig. 2), on the west and east sides of the Molucca Sea,
are actively colliding. The Molucca Sea plate dips east under
Halmahera and west under the Sangihe Arc in an inverted U-shape
(McCaffrey et al., 1980). Seismicity shows ~200–300 km of lith-
osphere subducted beneath Halmahera, whereas the Benioff zone
associated with the west-dipping slab can be identified to a depth
of at least 600 km beneath the Sangihe Arc.
The modern arc (Figs. 1 and 2) resumed activity in the Qua-
ternary, after a brief decline in volcanic activity, and the axis of the
arc moved ~50 km west from its late Miocene to Pliocene posi-
tion (Hall et al., 1988b, 1995). Both these arcs are built on older
volcanic arcs active during Cretaceous and early Cenozoic time.
The Quaternary and Neogene Halmahera Arcs have a chemical
character typical of intraoceanic arcs (Morris et al., 1983; Forde,
1997). There is no evidence of continental crust beneath them
except at the southernmost end of the volcanic arc (Morris et al.,
1983), on Bacan and Obi, where movement on strands of the
Sorong fault zone brought slivers of the continental crust beneath
the arc in the last few million years (Hall et al., 1995; Ali et al.,
2001). Mapping of the Halmahera Arc shows that the basement
is ophiolitic, formed in an early Mesozoic intraoceanic arc (Hall
et al., 1988a; Ballantyne, 1992), overlain by Cretaceous, Eocene,
 Basin formation by volcanic arc loading 13

Hainan

Manila Trench
Indochina
PHILIPPINE
South SEA

Phil
China PLATE

ippi
Sea Philippines
Se man

ne
a n 10oN
law
a
da

Pat
Pa

Trench
An

ta
ni
Chumphon Con
Mergui Sulu
Son
Malay Sea
North West Sabah
Sumatra Malay Sandakan
peninsula Natuna
East Sabah
Natuna
Sarawak Brunei Celebes CAROLINE
Penyu Tarakan Sea PLATE

Sea ca
C Halmahera

c
Su ent Sarawak

Molu
ma ral Muara
tra Sunda
Borneo 0o
Su

Shelf Kutai
Kalimantan
nd

Sumatra
a

South Barito Sulawesi

Tr

Sumatra
Asem New
en

Asem Seram Guinea

ch

West
Java Banda Sea Aru
East Java
Islands
Java Lombok
Jav Bali Timor Arafura
aT 10oS
SUNDALAND ren
ch Sumba rou
gh
helf
Shelf
CORE im or T ul S
T Sah

Indian
Ocean
INDIAN-AUSTRALIAN PLATE
90oE 100oE 110oE 120oE 130oE

Figure 1. Major basins of the southern Sundaland margin and location of the Halmahera and Java Arcs.

and Oligocene arc volcanic rocks. Shallow Miocene marine car-
bonates unconformably overlie all the older rocks.
The Neogene Halmahera Arc became active at ca. 11 Ma at
its southern end on Obi (Hall et al., 1995). In SW Halmahera,
volcanic activity began a little later as magmatism propagated
north. Basins formed on each side of the arc. In South Halma-
hera, activity in the volcanic arc, and subsidence on each side of
the arc, began in the late Miocene at ca. 8 Ma. To the west, tur-
bidites and debris flows were deposited in the forearc, and to the
east (Fig. 2), a marine basin developed in Weda Bay (Hall, 1987;
Hall et al., 1988b; Nichols and Hall, 1991).
On the western side of the SW arm of Halmahera (Fig. 2)
there was subsidence and deposition of at least 1000 m of subma-
rine slope deposits in the forearc (Nichols and Hall, 1991). In the
forearc, subsidence significantly exceeded the supply of mate-
rial. In contrast, on the backarc side, sediment supply broadly
kept pace with subsidence. On land in the SW arm of Halmahera
there are between 2800 and 3800 m of sedimentary rocks in the
basin east of the arc (Nichols and Hall, 1991). They were depos-
ited close to the arc in shallow water and rest unconformably on
pre-Miocene volcanic rocks or locally on shallow-water lower to
middle Miocene limestones. The sequence fines up from fan-delta
conglomerates into sandstones, mudstones, and limestones. Shal-
low marine deposition continued into the Pliocene. The depth of
water at the time of deposition of the upper part of the sequence
is uncertain but was no more than a few hundred meters.
The Miocene–Pliocene sequence interpreted from the seis-
mic sections (Fig. 3) across Weda Bay varies in thickness. It
rests unconformably on a karstified limestone surface and is up
to 2000 m thick in local depocenters but is typically ~1000 m
thick. Hence the basin was markedly asymmetric, with the great-
est thickness of sediment adjacent to the Halmahera Arc on the
western side of the basin. The local depocenters offshore, and
the fan deltas onshore, are thickest close to interpreted volca-
nic centers. There is little evidence of faulting associated with
basin formation on seismic lines (Fig. 3), and there is no indica-
tion of rifting. The lower-middle Miocene limestones represent
an approximate sea-level datum and probably covered the whole
area; close to the arc they were removed by erosion before depo-
sition of the basin sequence. The Halmahera Basin was actively
 Cotobato

Ph
Trench Talaud PHILIPPINE

ilip
0 km 200

pin
Islands SEA 4°N
N

e
Snellius

Tre
Ridge

nch
Sangihe
Morotai
Basin East
rc
CELEBES Morotai
ihe A

SEA Siau Central Rise

Molucca
Sang

Sea

t
hrus
North Sula

rc
wesi Trenc Ridge Morotai 2°N

ra A
h

ra T

ahe
ahe
Mayu

Halm
Halm
st

Halmahera
Thru

Tifore Ternate
Sulawesi North arm B
gihe

Tidore
San

MOLUCCA Waigeo 0°
Togian Islands Gorontalo SEA C Weda
Basin Gebe
Bay
Gag
Batanta
Bacan
Bird’s
East
Salawati Head
arm lt
g Fau
Obi Soron
Misool 2°S
Banggai Islands
Sula Islands
122°E 124°E 126°E 128°E 130°E A
Sangihe Halmahera
Thrust Thrust 0
W Accretionary complex E
Arc Forearc Forearc Arc ARC

km based on
seismic lines
SANGIHE HALMAHERA
MOLUCCA SEA PLATE based on
field
B 5
observations 0 km 20 C

Figure 2. (A) Location and major tectonic features of the Molucca Sea region. Small, black-filled triangles are modern volcanoes. Bathymetric
contours are at 200, 2000, 4000, and 6000 m. Large barbed lines are subduction zones, and small barbed lines are thrusts. (B) Cross section
across the Halmahera and Sangihe Arcs on section line B. Thrusts on each side of the Molucca Sea are directed outward toward the adjacent arcs,
although the subducting Molucca Sea plate dips east beneath Halmahera and west below the Sangihe Arc. (C) Inset is the restored cross section
of the Miocene–Pliocene Weda Bay Basin of SW Halmahera on section line C, flattened to the Pliocene unconformity, showing estimated thick-
ness of the section.
 0
0 km 10

1
TWT secs

3
0

seabed

1 Pliocene unco
TWT secs

Miocene-Pliocene nformity
volcanic arc-derived basin fill

2 top lime
stone
top basem
ent
A
3

0
0 km 10

1
TWT secs

Halmahera
2

3
0
0°
seabed Weda Bay
Pliocene u
1 nconformit
y
TWT secs

Miocene-Pliocene A
volcanic arc-derived basin fill

top limestone
2
B
B 128°E C
3
Figure 3. Uninterpreted seismic lines and interpreted sections across parts of the sedimentary sequences deposited at the western edge of Weda
Bay, Halmahera. Locations of lines A and B are shown on inset map C. There is little evidence of faulting on the sections and no indication of a
rift character to the basin. The Miocene–Pliocene arc-derived sequence is deposited on top of middle Miocene shallow marine limestones and is
terminated by an intra-Pliocene unconformity. The sequence thickens toward the Miocene–Pliocene volcanic arc. TWT—two-way traveltime.
16 Waltham et al.

subsiding from the late Miocene until the middle to late Pliocene,
an interval of ~5 m.y. Based on field measurement of sections
and biostratigraphic dating, Nichols and Hall (1991) estimated
that during deposition of this interval the western side of the
basin, adjacent to the arc, subsided at least 2.8 km, representing
an average subsidence rate of at least 47 cm/1000 yr. Subsidence
rates for the eastern part of the basin, ~100 km from the arc, were
between 17 and 34 cm/1000 yr.
At ca. 3 Ma, sedimentary rocks of this basin were thrust west
over the Neogene arc, and later there was thrusting of the forearc
from the west. The Quaternary Halmahera Arc is built uncon-
formably upon all these rocks. The thrusting is a result of colli-
sion of the Sangihe and Halmahera Arcs. In Weda Bay there has
been recent subsidence, and its deepest parts are almost 2000 m
below sea level. Part of this subsidence is probably due to thrust
loading to the west, but part may result from movements along
splays of the Sorong Fault that appear to control the form of the
present-day depocenter (Nichols and Hall, 1991).
East Java
East Java (Fig. 4) is situated on the continental margin of
Sundaland. There has been subduction to the south of Java, along
the Java Trench, since the early Cenozoic (Hall, 2002). The base-
ment of most of East Java has previously been interpreted as arc
and ophiolitic material accreted to the continental margin in the
Late Cretaceous, but our work (Smyth et al., this volume) has
shown that there is old continental crust beneath the Southern
Mountains of East Java. Our work in East Java also suggests
that the present northward subduction of Indian-Australian litho-
sphere began in the middle Eocene. The oldest Cenozoic rocks
resting on older basement are terrestrial conglomerates without
volcanic material, but a short distance above these rocks volca-
nic debris appears in middle Eocene sediments and increases in
abundance upsection (Smyth, 2005).
There is a record of two volcanic arcs (Fig. 4) in East Java
(Smyth et al., this volume). An early Cenozoic arc formed in the

o o
110 E 112 E
o
114 E Approximate location of the Oligocene-Miocene
volcanoes of the Southern Mountains Arc
\ Deposits of the Southern Mountains Volcanic
o
7S
Arc
Modern volcanoes of the Sunda Arc

Limit of the strong negative gravity anomaly

defined by -150 µms-2 contour
Approximate line of section (after
o
8S Pertamina, 1996)

0 50 100

kilometers

Modern Arc
N S

East Java Sea: Kendeng Basin: Southern Mountains Arc:

Edge of the Sunda Shelf Thick sequence behind Eocene to Early Miocene
Eocene to Early Miocene arc volcanic arc

5 km

Basement

100 km

Figure 4. Location of the Eocene to early Miocene volcanic arc in the Southern Mountains of East Java, the Kendeng Basin, and the modern arc.
Inferred volcanic centers of the Southern Mountains Arc are shown with open triangles, and active volcanoes are shown with solid triangles. A
cross section of the Kendeng Basin along the line indicated is based on Pertamina (1996). The Kendeng Basin sequence is interpreted to thicken
toward the Southern Mountains volcanic arc.
 Basin formation by volcanic arc loading
Southern Mountains and was active from the middle Eocene to
the early Miocene (ca. 42–18 Ma). It erupted material ranging
in composition from andesite to rhyolite (Smyth, 2005). The
arc formed a chain of volcanic islands during the early Ceno-
zoic that were initially similar in character to volcanoes of the
present-day Izu-Bonin-Mariana Arc and the Aleutian Islands,
except that behind the arc was a broadly shallow marine shelf
and no deep oceanic backarc basin. The volcanoes were terres-
trial, but some of their products were deposited close to the arc
in a marine setting, and they formed separate small islands. The
Oligocene volcanic centers are well preserved and have a spac-
ing similar to volcanoes of the modern arc on Java (Smyth et
al., this volume). They are predominantly andesitic and have
been described as the “Old Andesites” (van Bemmelen, 1949).
However, the volcanoes erupted considerable volumes of more
siliceous material in explosive Plinian eruptions, and this mate-
rial was dispersed widely as ash. The volume of siliceous mate-
rial has been overlooked in descriptions of volcanic activity on
Java. More details of the stratigraphy are given in Smyth et al.
(this volume).
Figure 5. Bouguer gravity anomaly map of East Java and location of the modeled profile discussed in the text.
17
Immediately behind, and to the north of, the Southern
Mountains Arc is the deep Kendeng Basin (Fig. 4). The basin
is long (at least 400 km) and narrow (100–120 km) and trends
east-west, parallel to the Southern Mountains Arc. The basin
is characterized by a strong negative Bouguer gravity anomaly
(Fig. 5), which exceeds −580 μms–2, and extends from west to
east. The basin formed during the middle Eocene (Untung and
Sato, 1978). The Kendeng Basin succession is not well exposed
but contains much volcanic debris carried north from the arc. The
oldest rocks are not seen in situ but are sampled by mud vol-
canoes currently erupting through the basin sequence. They are
terrestrial and shallow marine rocks similar to those deposited
close to the arc during the Eocene (de Genevraye and Samuel,
1972). The Kendeng Basin succession records a deepening of
the basin with time, and during the Oligocene thick sequences
of volcaniclastic turbidites and pelagic mudstones were depos-
ited, suggesting that subsidence exceeded the supply of material.
Seismic lines across the northern parts of the depocenter show
that the Kendeng Basin sequence thickens toward the Southern
Mountains Arc and is ~3 km thick in the north (Fig. 4; Pertamina,
18 Waltham et al.
1996). Untung and Sato (1978) suggested that the deeper parts of
the basin contain ~6 km of section. Gravity calculations suggest
there may be as much as 10 km of sediment in its thickest parts.
The increase in thickness of the sequence toward the arc and the
importance of volcanic debris suggest that most of the Kendeng
Basin fill was derived from the arc. Because the Kendeng Basin
is so poorly exposed, and there are no seismic lines crossing the
entire basin, it is impossible to assess the role of faulting in the
basin’s development. South- and north-dipping thrusts shown on
interpreted sections across the basin (Pertamina, 1996) could be
reactivated normal faults.
To the north of the Kendeng Basin was a carbonate and clas-
tic shelf during the early Cenozoic. This was the edge of the Sun-
daland continent. During the Eocene to early Miocene there were
elongate emergent ridges, trending roughly NE-SW and oblique
to the arc, separated by depressions. These ridges contain ter-
restrial clastic sediments and coals at their base, thought to be
Eocene, overlain by shallow marine Eocene to Miocene clastic
sediments and platform carbonates. There are no reports of vol-
canic material in the shelf sequences, although reported clay lay-
ers may be fine-grained volcanogenic air-fall tuff deposited far
from the arc. At the shelf edge there is volcanic material, includ-
ing clays, zircons, and volcanic quartz (Smyth, 2005). The basin
edge at the southern edge of the shelf is partly exposed in a fold-
thrust belt where there was 10–30 km of shortening, suggested
to have occurred during the Pliocene (de Genevraye and Samuel,
1972) or late Miocene.
The Southern Mountains Arc extends to the south coast of
Java, and there has been almost no hydrocarbon exploration off-
shore directly south of the coast, so little is known about this
region. Marine geophysical studies show ~1 km of sedimentary
cover of unknown age on basement in small forearc basins (Kopp
et al., 2006).
The Southern Mountains Arc has been elevated and tilted
since the early Miocene and now dips uniformly to the south at
~30° (Smyth, 2005). On land the arc is ~40 km wide. Arc activity
ceased in the early Miocene, followed by a period of little or no
volcanic activity (Smyth et al., this volume). Volcanic arc activ-
ity resumed in the late Miocene ~50 km north of the Southern
Mountains Arc, and the modern volcanoes are built on the Ken-
deng Basin (Smyth et al., this volume). The products of the late
Miocene to Holocene arc are more basic than the older arc and
are predominantly basaltic andesites to andesites (e.g., Soeria-
Atmadja et al., 1994; van Bemmelen, 1949).
Causes of Basin Formation
In South Halmahera and East Java there is association
between volcanic arc activity and basin formation. Both areas
were emergent or close to sea level at the time of basin initiation.
Subsidence began as the volcanic arc became active. Both basins
have asymmetrical profiles, with the greatest thickness of mate-
rial close to the volcanic arc. Clastic input came from the vol-
canic arc, and both basins are dominated by volcanic debris. In
Halmahera, local depocenters contain sequences that are thick-
est close to inferred volcanic centers. Some crustal extension is
likely in both volcanic arcs and is probably required in order to
allow magma to reach the surface. However, in Halmahera there
is little evidence for extensional faulting in the basin sequences,
and no significant faulting is seen on seismic lines. In neither
basin is there evidence for a typical rift character, nor is there
evidence for significant crustal extension. In neither region are
there oceanic backarc basins. There is no evidence for thrusting
before or during sedimentation. Thrusting, unrelated to arc devel-
opment, occurred in both areas after the basins formed and filled.
In Halmahera, arc-arc collision caused thrusting, first from the
backarc side and then from the forearc side, directed toward the
volcanic arc. In East Java there was mainly northward-thrusting of
the Kendeng Basin at its northern edge some time after volcanic
arc activity ceased. All these observations suggest that volca-
nic activity contributed in some way to basin formation, possibly
through loading by the volcanic arc itself or possibly by weaken-
ing of the plate, or by a combination of both.
MATHEMATICAL MODEL
In this section we develop a simple analytical model for
describing how load-generated subsidence is controlled by an
accumulating volcanic arc. The aim is to produce the simplest
possible model capable of testing whether the observed sub-
sidence is compatible with the likely size of load. The primary
assumptions of our model are that subsidence is caused by a
volcanic arc–generated line load and that the subsidence directly
under the load is proportional to the size of the load, i.e.,
s = kV, (1)
where s is subsidence, k is a constant, and V is the load (Fig. 6).
For flexure of a uniform, unbroken beam, k is given by (Turcotte
and Schubert, 2002, equations 3.127, 3.131, and 3.135)
k = π/(2xbΔρ1g), (2)
where xb is the flexural-bulge to arc-center distance, Δρ1 is the
density contrast between the basin fill and the asthenosphere,
and g is the acceleration from gravity. For a broken plate the
equivalent formula is (Turcotte and Schubert, 2002, equations
3.127, 3.141, and 3.144)
k = 3π/(4xbΔρ1g). (3)
The resulting basin is assumed to be filled by sediments up
to a horizontal surface and also filled by the volcanic arc but with
a triangular (in cross section) subaerial load caused by the current
volcanic edifice itself. The load therefore consists of two com-
ponents, a load resulting from the density excess of the buried
volcanic arc plus a load resulting from the subaerial volcanic edi-
fice. Hence
 Basin formation by volcanic arc loading 19

Xb
W

Sediment
fill

Flexing Lithosphere

Asthenosphere

Figure 6. Schematic diagram showing concept of basin formation by volcanic arc loading. W—arc width; xb—flexural bulge to arc-center dis-
tance; s—subsidence.

edifice) is unlikely to be available, and so a theoretical model for

⎧ s
⎫ arc width as a function of subsidence is required.
V = ⎨Δρ2 ∫ wds' + ρβw2 ⎬ g , (4) Combining equations 1 and 4, differentiating with respect to
⎩ 0 ⎭ s, and using the boundary condition that w(0) = 0 produces,

where Δρ2 is the density excess of the volcanic arc over the 1 = Aw + 2Bww′, (6)
remaining basin fill, w(s) is the arc width as a function of subsid-
ence, ρ is the volcanic arc density, and β is given by where the prime indicates differentiation and

β = 0.5 tan(θ) = h/W, (5) A = kgΔρ2, (7)

where θ is the volcano slope, h is the final volcano height, and B = kgβρ, (8)
W the final arc width. Note that equation 4 implies that there
was no significant deformation of the volcanic arc during basin with k given by equation 2 or 3. The solution to ordinary differ-
subsidence. ential equation 6 is (Appendix)
Given a well-constrained cross-sectional geometry, equa-
tions 1 through 4 alone would be sufficient to test the concept w = (1/A) f(A2s/B), (9)
that basins can be formed by flexural loading resulting from a
volcanic arc. In practice the required information (e.g., detailed where f(x) is the function shown in Figure 7A. Hence, the width
geometry of the arc deposits beneath the present-day volcanic of the arc increases rapidly in the early stages of subsidence (i.e.,
 A
f
0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

0
0 1 2 3 4 5 6
x

B Arc Width (km)

0 5 10 15 20 25 30 35 40
0

2
Subsidence (km)

Continuous-plate (CP)
Co
6 Broken-plate (BP) nti
nu
ou
BP sediment density +10% s-p
Br

lat
o

e
ke
n-

BP arc density +10%

pl
at

8
e

BP basin width +10%

Arc

Ba
sin

BP slope +10%
de

wi
ns

dt
ity

10
+1
+1

0%
0%

12

Figure 7. (A) Loading function derived in Appendix. Arc width increases with subsidence along a suitably scaled version of this curve. (B) Sen-
sitivity analysis. The highest dashed line shows the result for the continuous-plate model, equation 9, and the average values from Table 1. The
solid line shows the result for the broken-plate model, equation 9, and the average values from Table 1. The other curves show the result of
increasing each parameter by 10% using the broken-plate model.
 Basin formation by volcanic arc loading
when A2s/B is small) and asymptotically approaches a maximum
width of 1/A as subsidence becomes large. Figure 7B illustrates the
effect of arc width on basin subsidence predicted from equation 9.
The “standard” simulation uses the average values of the param-
eters given in Table 1 and assumes a broken-plate model (i.e., it
uses equation 3 rather than equation 2). This simulation shows
that highly significant subsidence can be obtained from reasonable
parameter values given sensible widths for a volcanic arc.
Figure 7B also shows the sensitivity to parameter uncertain-
ties. Each of the curves is obtained by increasing a single parame-
ter 10% above its standard value. The result of using a continuous
rather than a broken plate is also shown. Figure 7B demonstrates
that the most important factor is whether or not the plate can be
considered broken. Other factors have relatively minor influ-
ences, although uncertainties in the arc density are, perhaps, the
most significant factor. However, it should be noted that a 10%
uncertainty in arc density is unlikely, whereas a 10% uncertainty
in volcanic slope or in basin width is realistic.
In the following section, observations on the geometry and
gravity profiles from two Indonesian arcs are used to test the
model developed in this section.
APPLICATION TO INDONESIAN ARCS
The modern Halmahera volcanoes have elevations up to
1730 m and diameters at sea level of their base of between 10
and 25 km, typically ~20 km for those on land. Most of the active
volcanoes with smaller diameters are offshore volcanic islands
(Makian, Tidore, and Ternate) rising from sea-floor depths
of several hundred meters, and therefore their diameter at the
base is not known. A diameter of 20 km for all the volcanoes
at their base is reasonable. The volcanoes of the Quaternary arc
have been active for no more than 2 m.y. and probably less. The
dimensions of the Neogene volcanic centers are not known pre-
cisely, partly because of the nature of the exposure in rainforest
terrain, and partly because they are overthrust by backarc and
forearc basin rocks. The distribution of their products identified
during mapping of the islands at 1:250,000 scale suggests that
they were larger than the Quaternary volcanoes, consistent with
their longer period of activity from 8 to 3 Ma. A width of the arc
of 20 km is therefore a reasonable value. Like the Neogene arc,
the exact width of the Halmahera backarc basin is uncertain. Its
eastern edge is in Weda Bay at water depths of 2 km, and detailed
seismic lines cover only the western side of the offshore basin.
TABLE 1. PARAMETERS USED IN THE FLEXURAL MODEL FOR HALMAHERA
Parameter Minimum Maximum
3
Basin fill density (kg/m ) 2400 2200
3
Volcanic arc density (kg/m ) 2700 2700
Mantle density (kg/m3) 3400 3400
Basin width (km) 60 80
Subsidence (m) 3250 3350
Slope (°) 10 6 Minimum slope gives maximum arc width
Arc width (km) 10 25
Arc width (km) 17 28
Arc width (km) 21 34
21
Although the backarc basin has been thrust westward, the very
coarse character of the Neogene fan-delta deposits suggests they
were deposited within a few kilometers of the arc. The seismic
lines show that the basin deposits thin eastward, away from the
arc, and suggest that the basin width was between 60 and 80 km.
The estimated thickness of the sequence close to the Neogene arc
is between 2800 and 3800 m (Nichols and Hall, 1991) and up to
2 km offshore farther from the arc.
In addition to this information, the mathematical model
requires densities for the basin fill, mantle, and volcanic arc.
Based upon Hamilton (1979), we assume a basin-fill density of
2300 ± 100 kg/m3, a mantle density of 3400 kg/m3, and a vol-
canic arc density of 2700 kg/m3. These densities, together with
the observed basin width of 70 ± 10 km, subsidence of 3300
± 500 m, and β = 0.07 ± 0.02 (i.e., volcano slopes of 8 ± 2°), give
a predicted arc width of 23 ± 5 km for a broken-plate model, in
good agreement with the observed volcano diameters. The con-
tinuous plate model predicts an arc width of 28 ± 7 km. Hence,
the observed subsidence and arc width are compatible with basin
formation by arc loading and suggest that a broken-plate model
fits the data better than a continuous-plate model of lithospheric
flexure. The parameters used, and the results obtained, for the
Halmahera example are summarized in Table 1.
In our second test case we do not, unfortunately, have good
constraints upon basin subsidence, but, on the other hand, we
do have a gravity profile that can be directly compared with a
simulated profile based upon our mathematical model of arc-
loaded flexure.
In East Java the volcanoes are much larger than those on
Halmahera. Several of the present-day volcanoes have elevations
>3000 m, and all except two of East Java’s modern volcanoes are
between 1600 and 3300 m high. The volcanoes typically have
diameters of 50–55 km at their base. The modern arc began activ-
ity ~10–12 m.y. ago. As in Halmahera, the size of the volcanic
centers of the Southern Mountains Arc is more difficult to assess.
They have a similar distribution and spacing to the modern vol-
canoes. Several of the centers can be mapped up to 40 km north
of the coast, and the present steep coastline is not the southern
limit of the volcanoes. Thus, an arc width of 50 km, similar to the
diameter of the modern volcanoes, is reasonable and consistent
with an interval of arc activity of ~23 m.y. The basin width can be
determined with greater certainty; it is between 100 and 120 km
from the arc margin to the shelf edge, and there is estimated to
have been 10–30 km of shortening on thrusts at the shelf edge.
Comments
Minimum density gives maximum arc width
Andesite
Standard
Gives estimated elastic thickness of 4.5–7.1 km
See text
O bs e r v e d
Predicted broken-plate model
Predicted continuous-plate model
22 Waltham et al.

Independent estimates of basin thickness are unavailable for this
area, but gravity data (see Fig. 3) can be used to test the model.
The Bouguer anomaly data along an East Java profile is
shown in Figure 8. Figure 9 shows the same data after a regional
trend was removed owing to the gravity signature of a subduct-
ing lithospheric plate deep beneath Java. The gravity modeling
was performed by an add-on to our main arc-load modeling pro-
gram, which simply summed contributions from a large number
of small, rectangular, constant-density slabs using well-known
expressions (e.g., see Telford et al., 1990). The small density
contrast between the mantle lithosphere and the asthenosphere
was ignored, and a single density was assumed for the entire
mantle. A forward model of gravity over an arc-loaded, flexing,
broken elastic lithosphere is also shown in Figure 9. Flexure and
gravity model parameters are given in Table 2. Note that β =
0.05 implies a modern volcano height of 2500 m and a slope
of 6°. Using the parameters listed in Table 2, the mathematical
model predicts a subsidence of 6.9 km beneath the arc.
The key features of Figure 9 are the gravity low owing to
the basin depocenter, but also the gravity high at the south end

A 1400 S N
1200

1000

800
Bouguer Anomaly (gu)

600

400

200

-200

-400

-600

-800
B S N
900

800

700

600
Topography (m)

500

400

300

200

100

0
50 100 150 200 250 300
Distance (km)
Figure 8. Bouguer anomaly (A) and topographic profile (B) for East Java for the line shown in Figure 5.
 Basin formation by volcanic arc loading 23

600
S N
400

200
Bouguer Anomaly (gu)

-200

-400

-600

-800

-1000

-1200
50 100 150 200 250 300 350

Distance (km)
Figure 9. Bouguer anomaly from Figure 8 after removal of regional trend. The solid line is the predicted gravity anomaly based on a filled basin
produced by subsidence caused by volcanic arc loading of a broken elastic plate.

TABLE 2. PARAMETERS USED IN THE GRAVITY AND is a simple relationship between arc width at the surface and
FLEXURAL MODELING OF THE EAST JAVA BASIN the amount of subsidence. This model is strongly supported by
Parameters Units analysis of the Halmahera Arc, whose width agrees closely with
3
Basin fill density (kg/m ) 2200 that predicted by our model from the observed subsidence. Our
3
Volcanic arc density (kg/m ) 2900
Crustal density (kg/m )3
2700 model is further supported by the strong similarity between the
3
Crustal thickness (kg/m ) 40 observed Bouguer anomaly on East Java to that predicted by a
Mantle density (kg/m3) 3400 model of basin subsidence above a broken plate. The gravity pro-
Basin width (km) 150
Arc width (km) 50 files agree closely in both the depth and width of the anomaly
Slope (°) 6 produced by the basin.
Subsidence pred icted (m) 6900 For Halmahera, the subsidence observed can be accounted
for entirely by arc loading if a broken-plate model is assumed.
of the profile between 50 and 150 km, which is due to the posi- This is consistent with weakening of the arc crust by magmatic
tive effect of the volcanic arc itself. The remaining discrepancies heating. For East Java, the modeling suggests that an additional
between modeled and observed gravity profiles occur at distinct load is required to produce the observed subsidence. One obvi-
topographic highs and would be removed by suitable topographic ous contributor could be crustal extension, and we cannot rule
corrections. The basin-fill density is at the low end of figures given out extensional faulting, as the Kendeng Basin is poorly exposed
by Hamilton (1979). The crustal thickness of 40 km is possibly and there are no seismic lines crossing the basin. However, the
too high, although Hamilton (1979) shows a speculative thick- large volumes of highly acidic volcanic products that erupted in
ness beneath the Java magmatic arc of >30 km, and the value is East Java in the Southern Mountains Arc suggest significant dif-
well within the plausible range for a mature arc built on conti- ferentiation of arc magmas before eruption, which should have
nental crust (e.g., Kay and Mpodozis, 2001). However, the arc produced large volumes of basic cumulates (S. Sparks, 2005, per-
density has been set rather high, and this may reflect the need for sonal commun.). The relatively high density used in the model
further loads to produce the observed gravity high near the arc. would be consistent with large volumes of dense cumulates at
deep crustal levels. Magmatic underplating by such cumulates
DISCUSSION would have provided an additional load as in volcanic islands
(Watts et al., 1985).
The key prediction of the mathematical model of basin for- Probably the biggest uncertainty in the modeling and inter-
mation by volcanic arc loading, as presented here, is that there pretation is the density contrast between crust and arc. Little is
24 Waltham et al.

known about the crustal structure and thickness of any Indone- volcanoes and the underlying crust, and therefore the volcanoes
sian arc. The model assumes a normal continental crust for the do not act as a load. In continental margin arcs such as Java, and
arc basement with a load similar in composition to andesites. long-lived intraoceanic arcs such as Halmahera, there is a larger
This is reasonable in East Java, where the basement is continental density contrast between the deeper crust and the eruptive prod-
crust and preserved volcanic centers are predominantly andes- ucts of the arc. In these cases the volcanoes act as a significant
ites. It is also likely, based on observations of the character of load and produce flexural basins close to the arc. The absence of
small areas of exposed basement and oil-company drilling of the deep basins in arcs in some parts of Indonesia and elsewhere in
basement, that accreted arc and ophiolitic material includes ser- the world may be due to the small density contrast between vol-
pentinites, which would lower the average crustal density. For canoes and the underlying crust in those regions.
Halmahera, Milsom et al. (1996) concluded that the crust was at
least 20 km thick and had a bulk density approaching that of con- APPENDIX
tinental rather than oceanic crust. Local gravity highs are associ-
ated with Paleogene arc volcanic rocks rather than the ophiolites, Equation 6 is
and nowhere do the ophiolites have the high gravity fields asso-
ciated with classic ophiolites. Although much of Halmahera is 1 = Aw + 2Bww′. (A1)
underlain by ophiolitic rocks, significant serpentinization of the
ultramafic parts could account for the low crustal density. The Changing the variable to
products of the modern arc (Morris et al., 1983) and the Neogene
arc (Forde, 1997; Macpherson et al., 2003) are almost entirely x = A2s/B (A2)
basaltic andesites and andesites.
and scaling using
CONCLUSIONS
f = Aw (A3)
The role of volcanoes in the development of basins in arc
regions has generally been ignored, overlooked, or not consid- then gives
ered important. This is possibly because most studies have been
concerned with larger true backarc basins and forearc basins both 1 = f + 2 f f ′, (A4)
of which are formed much farther from the arc. However, in sev-
eral Indonesian arcs there are thick sequences of sedimentary where f ′ now indicates differentiation with respect to x. Equation
and volcaniclastic rocks in deep basins very close to the arc. The A4 is a nonlinear ordinary differential equation (ode), which does
mathematical modeling shows that volcanic loading can make not have an analytical solution in terms of elementary functions.
a contribution to basin subsidence in these settings, and may However, at small values of f the second term on the right side
be the primary cause. Volcanic loading can account for basins dominates, so that 1 ~2 f f ′ with a solution of f = x0.5. At large val-
close to the arc but is not relevant to basins >~100 km from the ues of f the first term dominates, and the solution asymptotically
arc. The basins produced by arc loading are asymmetrical, and approaches f = 1. The function f can be estimated using finite
basin sequences are thickest close to the arc. These are likely to differencing with
be dominated by very coarse, terrestrial and shallow marine, pri-
mary volcanic and volcaniclastic rocks that may include a variety f ~(fi + fi+1)/2 (A5)
of mass-flow deposits. Locally and intermittently the supply of
material from the arc may exceed subsidence, and there may be and
rapid vertical and lateral variations in grain size and changes from
shallow marine to terrestrial deposits. Farther from the arc, sub- f ′ ~(fi+1 – fi)/Δx, (A6)
sidence typically exceeds supply, and the more proximal deposits
pass laterally into deeper water sedimentary deposits such as tur- which, after substitution into (A4), yields
bidites. The distribution and character of the material, particularly
close to the arc, has some similarities to rocks described by Draut 2
− Δx ⎛ Δx ⎞ Δx
and Clift (2006) from the Jurassic Talkeetna Formation of Alaska. f i+1 = + ⎜ ⎟ + fi 2 − f i + Δx , (A7)
However, in the Alaskan Jurassic arc and in the modern Mari- 4 ⎝ 4 ⎠ 2
ana and Tonga Arcs, accommodation space is not likely to have
been created by volcanic loading but had already existed. This is which, together with the boundary condition that f0 = 0, allows
because in young intraoceanic settings the volcanoes rise from the function in Figure 2 to be calculated. The resultant function
great depths above the ocean floor. In young intraoceanic arcs has the required properties that f ~x0.5 for small x and f ~1 for
there is also likely to be little density difference between the arc large x.
 Basin formation by volcanic arc loading
ACKNOWLEDGMENTS
Our work on Indonesian arcs has been supported at different
times by the University of London Central Research Fund,
NERC, and the Royal Society, but mainly by a consortium of
oil companies whose membership has changed with time. We
thank colleagues in Indonesia at the Geological Research and
Development Center Bandung (now the Geological Survey of
Indonesia), LIPI, and Institut Teknologi Bandung for help with
field work. We thank Chris Elders and Andi Salahuddin for dis-
cussion of Weda Bay seismic interpretation.
REFERENCES CITED
Ali, J.R., Hall, R., and Baker, S.J., 2001, Palaeomagnetic data from a Meso-
zoic Philippine Sea Plate ophiolite on Obi Island, Eastern Indonesia:
Journal of Asian Earth Sciences, v. 19, p. 535–546, doi: 10.1016/S1367-
9120(00)00053-5.
Bahlburg, H., and Furlong, K.P., 1996, Lithospheric modeling of the Ordovi-
cian foreland basin in the Puna of northwestern Argentina: Tectonophys-
ics, v. 259, p. 245–258, doi: 10.1016/0040-1951(95)00129-8.
Ballantyne, P.D., 1992, Petrology and geochemistry of the plutonic rocks of
the Halmahera ophiolite, eastern Indonesia, an analogue of modern oce-
anic forearcs, in Parson, L.M., et al., eds., Ophiolites and Their Modern
Oceanic Analogues: Geological Society [London] Special Publication 60,
p. 179–202.
Busby, C.J., and Ingersoll, R.V., 1995, Tectonics of Sedimentary Basins: Cam-
bridge, Massachusetts, Blackwell, 579 p.
Cole, J.M., and Crittenden, S., 1997, Early Tertiary basin formation and the
development of lacustrine and quasi-lacustrine/marine source rocks on
the Sunda Shelf of SE Asia, in Fraser, A.J., et al., eds., Petroleum Geol-
ogy of SE Asia: Geological Society [London] Special Publication 126,
p. 147–183.
DeCelles, P.G., and Giles, K.A., 1996, Foreland basin systems: Basin Research,
v. 8, p. 105–123, doi: 10.1046/j.1365-2117.1996.01491.x.
de Genevraye, P., and Samuel, L., 1972, The geology of Kendeng Zone (East
Java): Proceedings of Indonesian Petroleum Association 1st Annual Con-
vention, Jakarta, p. 17–30.
Dewey, J.F., 1980, Episodicity, sequence and style at convergent plate margins,
in Strangeway, D.W., ed., The Continental Crust and Its Mineral Deposits:
Geological Association of Canada Special Paper 20, p. 553–573.
Draut, A.E., and Clift, P.D., 2006, Sedimentary processes in modern and ancient
oceanic arc settings: Evidence from the Jurassic Talkeetna Formation of
Alaska and the Mariana and Tonga Arcs, Western Pacific: Journal of Sedi-
mentary Research, v. 76, p. 493–514, doi: 10.2110/jsr.2006.044.
Forde, E.J., 1997, The geochemistry of the Neogene Halmahera Arc, eastern
Indonesia [Ph.D. thesis]: University of London, 268 p.
Hall, R., 1987, Plate boundary evolution in the Halmahera region, Indonesia:
Tectonophysics, v. 144, p. 337–352, doi: 10.1016/0040-1951(87)90301-5.
Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and
the SW Pacific: Computer-based reconstructions, model and animations:
Journal of Asian Earth Sciences, v. 20, p. 353–434, doi: 10.1016/S1367-
9120(01)00069-4.
Hall, R., and Morley, C.K., 2004, Sundaland Basins, in Clift, P., et al., eds., Con-
tinent-Ocean Interactions within the East Asian Marginal Seas: American
Geophysical Union Geophysical Monograph 149, p. 55–85.
Hall, R., Audley-Charles, M.G., Banner, F.T., Hidayat, S., and Tobing, S.L.,
1988a, Basement rocks of the Halmahera region, Eastern Indonesia: A
Late Cretaceous–Early Tertiary arc and fore-arc: Geological Society
[London] Journal, v. 145, p. 65–84.
Hall, R., Audley-Charles, M.G., Banner, F.T., Hidayat, S., and Tobing, S.L.,
1988b, Late Paleogene–Quaternary geology of Halmahera, Eastern Indo-
nesia: Initiation of a volcanic island arc: Geological Society [London]
Journal, v. 145, p. 577–590.
25
Hall, R., Ali, J.R., Anderson, C.D., and Baker, S.J., 1995, Origin and motion
history of the Philippine Sea Plate: Tectonophysics, v. 251, p. 229–250,
doi:10.1016/0040-1951(95)00038-0.
Hamilton, W., 1979, Tectonics of the Indonesian region: U.S. Geological Sur-
vey Professional Paper 1078, 345 p.
Karner, G.D., and Watts, A.B., 1983, Gravity anomalies and flexure of the
lithosphere at mountain ranges: Journal of Geophysical Research, v. 88,
p. 10,449–10,477.
Kay, S.M., and Mpodozis, C., 2001, Central Andean ore deposits linked
to evolving shallow subduction systems and thickening crust: GSA
Today, v. 11, no. 3, p. 4–9, doi: 10.1130/1052-5173(2001)011<0004:
CAODLT>2.0.CO;2.
Kopp, H., Flueh, E.R., Petersen, C.J., Weinrebe, W., Wittwer, A., and Meramex
Scientists, 2006, The Java margin revisited: Evidence for subduction ero-
sion off Java: Earth and Planetary Science Letters, v. 242, p. 130–142,
doi: 10.1016/j.epsl.2005.11.036.
Macpherson, C.G., Forde, E.J., Hall, R., and Thirlwall, M.F., 2003, Geochemi-
cal evolution of magmatism in an arc-arc collision: The Halmahera and
Sangihe arcs, eastern Indonesia, in Larter, R.D., and Leat, P.T., eds., Intra-
oceanic Subduction Systems: Tectonic and Magmatic Processes: Geolog-
ical Society [London] Special Publication 219, p. 207–220.
McCaffrey, R., Silver, E.A., and Raitt, R.W., 1980, Crustal structure of the
Molucca Sea collision zone, Indonesia, in Hayes, D.E., ed., The Tectonic
and Geologic Evolution of Southeast Asian Seas and Islands, Part 1:
American Geophysical Union Geophysical Monograph 23, p. 161–178.
Milsom, J., Hall, R., and Padmawidjaja, T., 1996, Gravity fields in eastern
Halmahera and the Bonin Arc: Implications for ophiolite origin and
emplacement: Tectonics, v. 15, p. 84–93, doi: 10.1029/95TC02353.
Morris, J.D., Jezek, P.A., Hart, S.R., and Gill, J.B., 1983, The Halmahera
island arc, Molucca Sea collision zone, Indonesia: A geochemical survey,
in Hayes, D.E., ed., The Tectonic and Geologic Evolution of Southeast
Asian Seas and Islands, Part 2: American Geophysical Union Geophysi-
cal Monograph 27, p. 373–387.
Moss, S.J., and McCarthy, A.J., 1997, Discussion—Foreland basin systems:
Basin Research, v. 9, p. 171–176.
Nichols, G.J., and Hall, R., 1991, Basin formation and Neogene sedimentation
in a backarc setting, Halmahera, eastern Indonesia: Marine and Petroleum
Geology, v. 8, p. 50–61, doi: 10.1016/0264-8172(91)90044-2.
Nunn, J.A., Czerniak, M., and Pilger, R.H., Jr., 1987, Constraints on the struc-
ture of Brooks Range and Colville Basin, Northern Alaska, from flexure
and gravity analysis: Tectonics, v. 6, p. 603–617.
Pertamina, B.P.P.K.A., 1996, Petroleum geology of Indonesian basins, prin-
ciples, methods and application. Volume IV: East Java Basins, 107 p.
Slingerland, R., Harbaugh, J.W., and Furlong, K., 1994, Simulating Clastic
Sedimentary Basins: New York, Prentice Hall, 220 p.
Smith, G.A., Moore, J.D., and McIntosh, W.C., 2002, Assessing roles of vol-
canism and basin subsidence in causing Oligocene–Lower Miocene sedi-
mentation in the northern Rio Grande Rift, New Mexico, USA: Journal of
Sedimentary Research, v. 72, p. 836–848.
Smyth, H.R., 2005, Eocene to Miocene basin history and volcanic history in
East Java, Indonesia [Ph.D. thesis]: University of London, 476 p.
Smyth, H.R., Hall, R., and Nichols, G.J., 2008, Early Cenozoic volcanic arc
history of East Java, Indonesia: The stratigraphic record of eruptions on a
continental margin in a tropical setting, in Draut, A.E., et al., eds., Lessons
from the Stratigraphic Record in Arc Collision Zones: Geological Society
of America Special Publication 436 (this volume).
Soeria-Atmadja, R., Maury, R.C., Bellon, H., Pringgoprawiro, H., Polvé, M., and
Priadi, B., 1994, Tertiary magmatic belts in Java: Journal of Southeast Asian
Earth Sciences, v. 9, p. 13–17, doi: 10.1016/0743-9547(94)90062-0.
Telford, W.M., Geldart, L.P., and Sheriff, R.E., 1990, Applied Geophysics:
Cambridge, UK, Cambridge University Press, 770 p.
Toksöz, M.N., and Bird, P., 1977, Formation and evolution of marginal basins
and continental plateaus, in Talwani, M., and Pitman, W.C., eds., Island
Arcs, Deep Sea Trenches and Back Arc Basins: American Geophysical
Union Maurice Ewing Series 1, p. 379–393.
Turcotte, D.L., and Schubert, G., 2002, Geodynamics (2nd edition): Cambridge,
UK, Cambridge University Press, 456 p.
Untung, M., and Sato, Y., 1978, Gravity and geological studies in Java, Indo-
nesia: Geological Survey of Indonesia and Geological Survey of Japan,
Special Publication 6, 207 p.
26 Waltham et al.
van Bemmelen, R.W., 1949, The Geology of Indonesia: Nijhoff, The Hague,
Government Printing Office, 732 p.
Watts, A.B., and Cochran, J.R., 1974, Gravity anomalies and flexure of the
lithosphere along the Hawaiian-Emperor seamount chain: Royal Astro-
nomical Society Geophysical Journal, v. 38, p. 119–141.
Watts, A.B., ten Brink, U.S., Buhl, P., and Brocher, T., 1985, A multichannel
seismic study of lithospheric flexure across the Hawaiian-Emperor sea-
mount chain: Nature, v. 315, p. 105–111, doi:10.1038/315105a0.
Watts, A.B., Peirce, C., Collier, J., Dalwood, R., Canales, J.P., and Henstock,
T.J., 1997, A seismic study of lithospheric flexure in the vicinity of Tener-
ife, Canary Islands: Earth and Planetary Science Letters, v. 146, p. 431–
447, doi: 10.1016/S0012-821X(96)00249-X.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Cenozoic arc processes in Indonesia: Identification of the key influences on the

stratigraphic record in active volcanic arcs

Robert Hall
Helen R. Smyth
SE Asia Research Group, Department of Geology, Royal Holloway University of London, Egham, Surrey, TW20 0EX, UK

ABSTRACT

The Indonesian region includes several volcanic island arcs that are highly active
at the present day, and also contains a record of Cenozoic volcanic activity owing to
subduction of oceanic lithosphere at the margins of SE Asia. As a result of long-term
subduction, there is a high regional heat flow, and a weak crust and lithosphere, as
identified in other subduction zone backarcs. The stratigraphic record in the Indo-
nesian region reflects a complex tectonic history, including collisions, changing plate
boundaries, subduction polarity reversals, elimination of volcanic arcs, and exten-
sion. The arcs have not behaved as often portrayed in many arc models. They mark
subduction but were not continuously active, and it is possible to have subduction
without magmatism. Subduction hinge retreat was accompanied by significant arc
volcanism, whereas periods of hinge advance were marked by reduction or cessation
of volcanic activity. Growth of the region occurred in an episodic way, by the addition
of ophiolites and continental slivers, and as a result of arc magmatism. In Indonesia,
relatively small amounts of material were accreted from the downgoing plate during
subduction, but there is also little evidence for subduction erosion. During collision
the arc region may fail, resulting in thrusting, and the weakest point is the position of
the active volcanic arc itself. Volcanic arcs shift position suddenly, and arcs can disap-
pear during collision by overthrusting. Arcs are geologically ephemeral features and
may have very short histories in comparison with most well-known older orogenic
belts. The stratigraphic record of the basins within arc regions is complex. Because of
a weak lithosphere the character of sedimentary basins may be unusual, and basins
are commonly very deep and subside rapidly. There is a high sediment flux. The vol-
canic arc itself influences the stratigraphic record and basin development. The load
imposed by the volcanic arc causes flexure and provides accommodation space. The
volcanic arc thus can form the basin and supply most of its sediment. Tropical pro-
cesses influence the mineralogy and apparent maturity of the sediment, especially
volcanogenic material. A complex stratigraphy will result from the waxing and wan-
ing of volcanic activity.

Keywords: Cenozoic, Sunda Arc, Banda Arc, Sangihe Arc, Halmahera Arc, Indone-
sia, stratigraphic record.

Hall, R., and Smyth, H.R., 2008, Cenozoic arc processes in Indonesia: Identification of the key influences on the stratigraphic record in active volcanic arcs, in
Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America
Special Paper 436, p. 27–54, doi: 10.1130/2008.2436(03). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All
rights reserved.

27
28 Hall and Smyth

INTRODUCTION background on the region, and a summary of the history of several

The size of Indonesia, the abundance of volcanic activity,
and its long volcanic history make it an important area for under-
standing arc processes and influences of volcanic arcs on the
stratigraphic record. Hamilton (1988) commented that the arcs
of Indonesia–southern Philippines–western Melanesia are in “the
region of greatest modern variety and complexity,” but the region
remains understudied and often overlooked. The long history of
subduction beneath Indonesia has influenced arc development.
Subduction has produced a thin and warm lithosphere beneath
the upper plate, which is unlike that of older stable continents and
their margins. These features have influenced the development of
the Indonesian arcs, which differ in many ways from textbook
examples of arcs described from other parts of the world. They
were formed at the margins of a large, weak subduction backarc
region and consequently seem to have been unusually responsive
to plate boundary forces. Because of their youth it is not pos-
sible to study the deep levels of the arcs, but the arc history that
is revealed by their stratigraphy offers insights into the develop-
ment of older orogens and older volcanic arcs.
In this paper we identify some important features of Indo-
nesian volcanic arcs, particularly those that influence the strati-
graphic record of arc activity. Because the geology of Indonesia is
not familiar to many readers, we begin our discussion with some
of the most important Cenozoic arcs in Indonesia. We then dis-
cuss particular features of the record of arc activity and attempt to
identify those that may be of general interest in understanding arc
processes and those that may be useful in interpreting the record
of activity in older arcs elsewhere in the world. Our discussion
of the stratigraphic record in Indonesian arcs is based mainly on
studies carried out by the SE Asia Research Group in several dif-
ferent arcs over many years, especially in Java and Halmahera,
but also in arcs of Sumatra, Sulawesi, and Borneo. These studies
have included extensive field work and related work documented
in theses, but some of the results are unpublished and in places
we may appear to make statements unsupported by easily acces-
sible literature; we hope this will be forgiven, and we felt it was
justified in our attempt to give an account of the region in a paper
of reasonable length.
Indonesia and Large Explosive Volcanic Events
Indonesia is a large country that includes >18,000 islands
and stretches >5000 km from west to east (Fig. 1). Wallace (1869)
tried to draw attention to its size in his book The Malay Archi-
pelago by comparing the island of Borneo to the British Isles,
and subsequently Umbgrove (1938) and van Bemmelen (1949)
compared the region to areas more familiar to North American

Andaman
10°N Sea South China Sea
Philippines

5°N Malay Sunda

Peninsula Shelf Sangihe
Celebes
Sea
Borneo Bird’s
Nias Halmahera Head
0°
ts

Molucca
ai

Sea
Str
sar

Sulawesi
kas

Indian
Ocean Sumatra
Ma

5°S Java Sea

Banda
Flores Sea
Sea
Java

Timor
0 500 1,000
10°S Northwest Shelf
kilometers Australia
100°E 110°E 120°E 130°E

Figure 1. Indonesia shown on a digital elevation model (DEM), based on Shuttle Radar Topography Mission (SRTM) data merged with the
Sandwell and Smith (1997) bathymetry.
 Cenozoic arc processes in Indonesia
and European readers. Figure 2 is a modern attempt to show the
size of the region by comparing Indonesia to the conterminous
United States. According to the Smithsonian Global Volcanism
Program (Smithsonian, 2006), Indonesia includes 150 Holocene
volcanoes; 95 of these could be considered active because they
have erupted since A.D. 1500. Indonesia has a long record of
volcanic activity, recording subduction related to northward
movement of the Indian-Australian plate and westward motion
of the Pacific plates. Cenozoic subduction of Indian-Australian
oceanic lithosphere beneath most of Indonesia began after Aus-
tralia separated from Antarctica and moved rapidly northward
from ~45 m.y. ago (Hall, 2002).
With a long history of volcanic activity, and a well-known
explosive character in historical times, it is surprising that the
only exceptionally large eruptions known from Indonesia in the
Cenozoic are those of Toba. Of the Holocene volcanoes, 32 have
records of very large eruptions with a volcanic explosive index
(VEI) >4; 19 of these have erupted in the past 200 yr, includ-
ing Tambora in 1815 (VEI = 7) and Krakatau in 1883 (VEI =
6). There are only 4 volcanoes in the Smithsonian global data
set of large-volume Holocene explosive eruptions with a VEI of
7; the other 3 erupted in prehistorical times and pre-date 4000
B.C. Tambora, on the island of Sumbawa, is known for its impact
on global climate, and its 1815 eruption resulted in the North-
ern Hemisphere’s “year without a summer” in 1816, when crops
failed, causing famine and population movements (Harington,
1992; Zeilinga de Boer and Sanders, 2002). The 74 ka eruption
of Toba (Chesner and Rose, 1991; Chesner et al., 1991), on the
5000 km
Figure 2. Indonesia and Malaysia compared to the USA. The box is 60° from east to west and 25° from north to south.
29
island of Sumatra, was the largest in the last 2 m.y., and its effect
on climate must have been even more devastating. It has been
speculated that the eruption led to accelerated glaciation and a
bottleneck in human evolution (Ambrose, 2003; Rampino and
Self, 1992, 1993a, b).
For the last 36 m.y., 42 large explosive events are recorded
in a recent global compilation (Mason et al., 2004); 33 of
these large explosive events are recorded from North America,
whereas Toba alone is situated in a humid tropical climate. The
record has been interpreted to indicate that there were two pulses
in Cenozoic global volcanism: one between 36 and 25 Ma, and
the other since 13.5 Ma. However, Indonesia remains a rela-
tively remote area, which is difficult to explore because of the
absence of infrastructure, difficult terrain, and rain forest cover.
We suggest that the absence of exceptionally large eruptions
in the stratigraphic record from Indonesia has more to do with
preservation and sampling than with volcanic activity. As we
report below, there is indeed evidence of large explosive events
in Indonesia during the Cenozoic, and no doubt more of these
remain to be discovered. The lack of information on large erup-
tions in Indonesia is not dissimilar to the general lack of infor-
mation on Indonesia’s volcanic history. This partly reflects the
absence of modern detailed studies, but ideas about arcs may
be unduly influenced by results from more accessible or bet-
ter exposed regions, from ocean drilling, and from geochemi-
cal studies. However, as with large eruptions the stratigraphic
record in Indonesia has some valuable insights to reveal; this
paper identifies some of them.
30 Hall and Smyth

ARC ACTIVITY IN INDONESIA
The distribution of modern-day volcanoes in Indonesia
shows clearly that most volcanic activity is related to subduction
(Fig. 3). This is also emphasized by the distribution of seismic-
ity (Cardwell and Isacks, 1978; Cardwell et al., 1980; Engdahl
et al., 1998; England et al., 2004). Global positioning system
(GPS) observations (Bock et al., 2003; Kreemer et al., 2000;
Rangin et al., 1999), seismicity, and volcanic activity indicate
rapid plate movements and considerable tectonic complexity.
Present-day Indonesian volcanic activity can be regarded as the
products of four separate volcanic arcs (Fig. 3): the Sunda Arc,
stretching from Sumatra through Java to the east, the horseshoe-
shaped Banda Arc of eastern Indonesia, the Halmahera Arc
of the North Moluccas, and the Sangihe Arc, extending from
Sulawesi into the southern Philippines. Here we give first a brief
summary of the Cenozoic history of each of these arcs and the
character of the crust on which they are constructed to provide
the background to our commentary on Java, east Indonesia, and
the Moluccas. More detailed discussion of each of these arcs,
literature references, and animated reconstructions of the tec-
tonic development are in Hall (2002) and Garwin et al. (2005).
Java: The Sunda Arc
Indonesia is situated on the southern margin of the Eurasian
plate at the edge of Sundaland (van Bemmelen, 1949; Hamilton,
1979; Hutchison, 1989; Katili, 1975). The Sunda Arc has been
active since the Eocene (e.g., Garwin et al., 2005; Hall, 2002;
Hamilton, 1977) and is the product of subduction of the Indian-
Australian plate beneath the Sundaland margin. Sundaland
(Fig. 4) is the continental core of SE Asia, comprising Indochina,
the Thai-Malay Peninsula, Sumatra, Java, Borneo, and the shal-
low marine shelf (Sunda Shelf) between them, and formed by
amalgamation of continental blocks during the Triassic. There
is a Proterozoic basement (Liew and McCulloch, 1985; Liew
and Page, 1985) intruded by Permian–Triassic granites formed
by subduction and postcollisional thickening of the continental
crust (Hutchison, 1989, 1996). In Borneo the oldest rocks known
are Paleozoic metamorphic rocks intruded by Mesozoic granites.
From Sumatra to Borneo the continental core is surrounded by
Mesozoic ophiolitic and arc igneous rocks, and possible frag-
ments of continental material, accreted during the Cretaceous.
In Sumatra and West Java the Sunda Arc is built in part on the
continental margin of Sundaland. In East Java it is constructed
on arc and ophiolitic rocks accreted at the Mesozoic active mar-
gin and in part on a continental fragment that was added by a
Cretaceous collision.
In Sumatra a record of subduction-related magmatism
extends back into the early Mesozoic or late Paleozoic (McCourt
et al., 1996), but in Java the oldest record of subduction is Cre-
taceous. The Sundaland Cretaceous active margin is interpreted
to have run the length of Sumatra into West Java and then turned
northeast into SE Borneo (Hamilton, 1979). In Central Java and
SE Borneo there is evidence for Cretaceous high pressure–low
temperature subduction-related metamorphism (Parkinson et

5°N

Sangihe Talaud
Arc
Morotai
Mayu Halmahera
Borneo Arc
0°
Sunda
Kalimantan
Su

Shelf
m
at

Sulawesi
ra

Obi
ait
Su

r Str
nd

5°S Seram
a

Buru
assa
Ar

Java Sea
c

Mak

Banda Arc
Java Wetar
Bali Flores
Sunda
10°S
Arc
Lombok
Timor

95°E 100°E 105°E 110°E 115°E 120°E 125°E 130°E

Figure 3. Present-day volcanoes in Indonesia (Smithsonian, 2006) and the principal volcanic arcs discussed in the text: Sunda Arc, Banda Arc,
Halmahera Arc, and Sangihe Arc. Indonesia is shaded in gray, and the area of Malaysia is unfilled.
 Cenozoic arc processes in Indonesia 31

Manila Trench
EURASIAN PHILIPPINE
PLATE SEA PACIFIC
PLATE OCEAN

Philip
South
China Philippines

pine
Andaman Sea
10°N
PACIFIC

Tren
Sea

c
PLATE

h
MESOZOIC Borneo Sulu
SUNDALAND Sea
CORE Suture


Molucca


Sunda
Shelf
Celebes
Sea
Suture CAROLINE
PLATE
Molucca
Sea
Sorong
SUNDALAND Suture 0°

t
trai
Sun

S
sar
da

Se
ram
kas
Bismarck
Tre

Sea
Ma
nc

Tro
Sulawesi
h

Banda

u
Java Sea Banda

gh
Suture Sea
Suture New Guinea
ADDED BY
EARLY CENOZOIC
Arafura 10°S
g h
Java Trou Shelf
INDIAN Trench
Tim
or
OCEAN Sahul
ADDED BY Shelf
EARLY MIOCENE
Coral

0 km 1000
INDIAN-AUSTRALIAN Sea

PLATE AUSTRALIA
90°E 100°E 110°E 120°E 130°E 140°E 20°S

Figure 4. Wide plate boundary zone of Indonesia. The light shaded area is the zone of collision between the Eurasian, Indian-Australian, and
Pacific–Philippine Sea plates, which can be considered as a wide suture zone, within which there are smaller sutures shown in darker shading.
Lines show incremental growth of the Indonesian region at different stages by addition of fragments of continental, arc, and ophiolitic rocks.
Modified from Hall and Wilson (2000).

al., 1998). In Indonesia there is almost no Paleocene record, and
it appears that most of Sundaland was emergent. All published
plate tectonic reconstructions show subduction beneath Java
before 45 Ma, but there is no record of volcanic activity in Java
in the early Cenozoic (Smyth et al., this volume). We suggest
there was a cessation of subduction beneath Java in the Late
Cretaceous following the Cretaceous collision of a continental
fragment at the Sundaland margin (Smyth et al., this volume).
Our recent and current work in Java suggests that subduction
resumed in the middle Eocene, forming a volcanic arc in the
Southern Mountains that ran the length of Java. At this time,
ca. 45 Ma (Fig. 5), the rate of Australia-Antarctica separation
increased, and Australia began to move northward relatively rap-
idly. Since then there has been continuous subduction of Indian
Ocean lithosphere beneath the Sunda Arc. During the Eocene
and Oligocene the position of the volcanic arc remained broadly
fixed, and from Sumatra to Sulawesi abundant volcanic activ-
ity accompanied northward subduction of the Indian-Australian
plate. In the Java-Sulawesi sector of the Sunda Arc, volca-
nism greatly diminished during the early and middle Miocene,
although northward subduction of Indian-Australian lithosphere
continued (Fig. 5). Following Australian continental collision
in east Indonesia, the Sunda subduction hinge advanced north-
ward as a result of counterclockwise rotation of the Borneo-Java
part of the Sundaland margin, and this is interpreted to have led
to termination of magmatism, despite continued subduction,
because of the absence of fresh material replenishing the mantle
wedge (Macpherson and Hall, 1999, 2002). At the end of the
middle Miocene at ca. 10 Ma, volcanic activity resumed in the
Java-Sulawesi sector of the Sunda Arc, after the termination
of Borneo-Java rotation. Magmatism began again in Java with
increased vigor, but in a position north of the Paleogene arc,
forming the modern arc (Fig. 5). During most of the Cenozoic
since the Eocene, Java was a chain of volcanic islands, and only
in the last 5 m.y. has the island emerged as an extensive area
of land.
90°E 120°E 30°N 90°E 120°E 30°N
45 Ma 15 Ma
Shikoku PACIFIC
Basin PLATE

PACIFIC
PLATE
Luzon
Parece
Proto-South Vela
China Sea Izu
Palawan Basin

Bo
Sulu

nin
Sea

Ma

Su
ma
rian
Su

tra
as
Celebes

ma
Sea West 0° 0°

tr
Borneo

a
Borneo Philippine
c
Ar
Basin

Su
gihe

nd
San Molucca

a
Ja
va Sulawesi
Eas Sea
Ar
c West t P Java
Sulawesi hil Halmahera
ip pin
es
–
Ha INDIAN-AUSTRALIAN Banda
lm PLATE Embayment New Solomon
INDIAN-AUSTRALIAN ah
er Halmahera Guinea Sea
PLATE a
Arc

20°S 20°S

90°E 120°E 30°N

25 Ma 5 Ma
PACIFIC
PLATE
South
China
Sea
Mariana
Trough

ea th

o
aS u
h in -So
Pr
C ot
Parece Celebes
Vela Sea
c

Basin Ayu
Sea ca

0°
Molu

Trough

Celebes
Sea Bismarck
Caroline Sea
Sea Banda
Sea New
Guinea
Bird’s
Head
Sula INDIAN-AUSTRALIAN
INDIAN-AUSTRALIAN Spur
PLATE PLATE
New
Guinea 20°S
 Cenozoic arc processes in Indonesia 33

Figure 5. Reconstructions of the Indonesian region, based on Hall (2002). 45 Ma: The Pacific plate was moving broadly NW and subducting be-
neath the east Asian margin and the eastern margin of the Philippine Sea plate, forming the Izu-Bonin-Marianas Arc. Spreading centers linked the
West Philippine Sea Basin and the Celebes Sea. Australia was moving NNE on the Indian-Australian plate, and northward subduction produced
the Sunda Arc and the East Philippines–Halmahera Arc. The proto–South China Sea was subducting southward beneath the north Borneo–Luzon
margin. The parts of the Pacific and Indian-Australian plates shown in blue without anomalies have been subducted since 45 Ma. 25 Ma: The
East Philippines–Halmahera–South Caroline Arc collided with the Australian margin in New Guinea. Collision between Australia and SE Asia
began with initial contact of the Sula Spur, the continental promontory east of the Bird’s Head of New Guinea, and East Sulawesi. These events
caused major reorganization of plate boundaries. Active spreading in the South China Sea was caused by southward subduction of the proto–South
China Sea. 15 Ma: Subduction continued beneath the Sunda Arc, but arc volcanic activity ceased in Java between ca. 20 and 10 Ma as the hinge
advanced, owing to collision of the Australian margin in Sulawesi. Arc terranes in New Guinea moved westward in a wide left-lateral strike-
slip zone. Locking of splays at the western end of the fault zone induced eastward subduction of the Molucca Sea; westward subduction of the
Molucca Sea beneath the Sangihe Arc had begun at ca. 25 Ma. 5 Ma: Volcanic activity resumed in Java at ca. 10 Ma and has continued until the
present. Hinge rollback into the Banda Embayment south of the Bird’s Head produced the arcuate subducted slab beneath the Banda Arc and led
to backarc spreading in the Banda Sea. The Molucca Sea was in the process of elimination by subduction at its west and east sides beneath the
Sangihe and Halmahera Arcs.

Eastern Indonesia: The Banda Arc
East of Java and Borneo, Indonesia can be considered to
be a wide and complex suture zone (Fig. 4), and even today
it can be tectonically described only in terms of several small
plates and multiple subduction zones (Hall, 2002; Hall and Wil-
son, 2000; Hamilton, 1979). Eastern Indonesia is the product of
a long period of extension, subduction, and collision and has
a complex basement; there is some continental crust but also
much more arc and ophiolitic basement in comparison with
western Indonesia. The Banda Arc is the horseshoe-shaped arc,
which today extends east from Flores to Buru (Fig. 3), pass-
ing through Timor and Seram and includes both an outer non-
volcanic arc and an inner volcanic arc. Although there was a
volcanic arc in eastern Indonesia during the Paleogene, and
parts of this and possibly older arcs may be found in the highest
nappes of Timor and other Banda islands, we do not term them
the Banda Arc. The older volcanic arc rocks formed part of the
Paleogene Sunda Arc. The Banda volcanic arc is young and has
been active only for ~10 m.y. (Abbott and Chamalaun, 1981;
Honthaas et al., 1999), and the horseshoe-shaped Banda Arc is a
phenomenon of the same time interval. It formed by subduction
of an embayment within the northward-moving Australian plate
(Charlton, 2000; Hall, 1996, 2002; Hamilton, 1979). The arc
developed within the collision zone after the Australian margin
collided with the former active margin of the Sunda Arc.
From the Eocene the Sunda Arc can be traced east through
Sulawesi into the East Philippines and Halmahera Arcs in the
Pacific. During the Paleogene (Fig. 5) subduction was northward
beneath Sulawesi and the East Philippines until the first collision
of Australian crust, the Sula Spur, with SE Asia in the early Mio-
cene ~25 m.y. ago (Fig. 5). The collision caused the long subduc-
tion system to separate into two parts. West of Sulawesi, northward
subduction continued in the Sunda Arc, but to the east subduction
ceased, and the Australia–Philippine Sea plate boundary became
a strike-slip system. Between the two, the Australia–SE Asia col-
lision formed a mountain belt in East Sulawesi. By the late middle
Miocene, at ca. 12 Ma, convergence in East Sulawesi could no
longer be accommodated by orogenic contraction. At this time the
oldest oceanic lithosphere in the Indian Ocean, of Late Jurassic
age, arrived at the eastern end of the Java Trench. This area of
old crust north of the NW Shelf of Australia formed an embay-
ment in the Australian margin, the proto–Banda Sea. Because of
its age and thickness, the Jurassic ocean lithosphere fell away rap-
idly, causing the Banda subduction hinge to roll back rapidly to
the south and east, inducing massive extension in the overlying
plate. In western Sulawesi this first induced extensional magma-
tism, which began at ca. 11 Ma (Polvé et al. 1997). As the hinge
rolled back into the Banda Embayment it led to formation of the
Banda Arc and the opening of the North Banda Sea, as described
by Hamilton (1988). Further rollback caused the opening of the
Flores Sea and later the South Banda Sea (Fig. 5). South of the
Bird’s Head microcontinent the rollback of the subduction hinge
resulted in collision of the Banda volcanic arc at ca. 3 Ma with
the Australian margin in the region of Timor and the cessation of
volcanic activity in this segment of the arc. After collision, conver-
gence ceased south of the volcanic arc, and new plate boundaries
developed north of the arc between Flores and Wetar and to the
north of the South Banda Sea. Within the Bird’s Head microconti-
nent there has been significant shortening and probable intraconti-
nental subduction within the last 3 m.y. at the Seram Trough.
Moluccas: The Halmahera and Sangihe Arcs
The Halmahera and Sangihe Arcs both have long volcanic
histories. The modern Halmahera Arc is constructed on older
arcs, of which the oldest known is an intraoceanic arc formed
in the Pacific in the Mesozoic (Hall et al., 1988a, 1995) pre-
sumably built on older oceanic crust. The modern Sangihe Arc
is constructed on arcs formed at the Pacific margin in the early
Cenozoic (Hall, 2002); again the deepest parts are built on older
oceanic crust (Evans et al., 1983). Both of the currently active
arcs formed during the Neogene. They are unusual in that they
are the only arcs in the world that are currently colliding.
The pre-Eocene history of these arcs is not well known. At
45 Ma (Fig. 5) the Halmahera Arc was far out in the western
34 Hall and Smyth
Pacific but was situated on the southern margin of the Philip-
pine Sea plate beneath which there was northward subduction
of Indian-Australian lithosphere. Between 45 and 25 Ma the
Philippines-Halmahera Arc remained at approximately the same
latitude above a north-dipping subduction zone north of Austra-
lia. At ca. 25 Ma (Fig. 5) there was the most important Cenozoic
change in plate boundaries in the region. Arc-continent collision
between the East Philippines–Halmahera Arc and the New Guinea
margin terminated northward subduction of oceanic lithosphere
north of Australia, and a major left-lateral strike-slip boundary
developed in northern New Guinea. Arc terranes were translated
westward within this strike-slip system. The arc terranes at the
southern edge of the Philippine Sea plate moved in a clockwise
direction along the northern New Guinea margin within the left-
lateral Sorong strike-slip zone. At the western end of the left-
lateral fault system there was subduction beneath the Sangihe Arc,
and shortening and uplift in East Sulawesi. The Philippine Sea
plate moved northward as it rotated clockwise, accompanied by
complex strike-slip faulting and minor subduction at its western
edge in the Sangihe Arc and Philippines (Fig. 5). The Molucca
Sea double subduction system was initiated at ca. 15 Ma (Fig. 5).
Subduction on the west side of the Molucca Sea in the Sangihe
Arc had started soon after the 25 Ma plate reorganization, but
the oldest Neogene volcanic rocks in the Halmahera Arc have
ages of ca. 11 Ma (Baker and Malaihollo, 1996). Initiation of
east-directed Halmahera subduction probably resulted from the
locking of one of the strands of the left-lateral Sorong fault zone
at the southern edge of the Molucca Sea. In eastern Indonesia the
Figure 6. Contoured heat-flow map for SE Asia, based on the database of Pollack et al. (1990, 1993) and oil company
compilations (Kenyon and Beddoes, 1977; Rutherford and Qureshi, 1981).
Molucca Sea has been eliminated by subduction at both its east-
ern and western sides (Fig. 5). The Sangihe–North Sulawesi Arc
is now thrusting over the Halmahera Arc in the Molucca Sea,
as discussed below. The central Molucca Sea mélange wedge
and ophiolites represent the forearc basin and basement of the
Sangihe Arc, which will soon have completely overridden the
Halmahera Arc (Hall, 2000).
RESULTS OF LONG-TERM SUBDUCTION
There is a great contrast between present tectonic activity,
manifested by seismicity and volcanism at the margins, and the
apparent interior stability of Sundaland, which has led some
authors to describe the core as a shield or craton (e.g., Ben-
Avraham and Emery, 1973; Gobbett and Hutchison, 1973; Tjia,
1996). It is true that most of the areas of extreme relief, high
elevations, and actively rising mountains are in eastern Indonesia,
but the core is certainly not a craton. Most of the Sunda Shelf is
flat and close to sea level (Fig. 1), and even the large island of
Borneo has generally low elevations with the exception of the
isolated 4 km peak of Mount Kinabalu in the north. However, the
picture of Indonesia as an active volcanic margin surrounding a
stable continental region is misleading. There has been significant
deformation within Sundaland during the Cenozoic. Heat flow,
seismic tomography, and geological observations indicate that
the Sundaland continental core north of Indonesia is unusual.
Sundaland has high surface-heat-flow values (Fig. 6), typi-
cally more than 80 mW/m2 (Artemieva and Mooney, 2001; Hall
 Cenozoic arc processes in Indonesia 35

and Morley, 2004). At the margins of Sundaland, high heat flow is
related to subduction-related processes and magma rise, but in the
interior this is not a result of arc magmatism as sometimes sug-
gested (e.g., Nagao and Uyeda, 1995). High heat flow values are
recorded in the interior Sundaland basins from the Gulf of Thai-
land to west Borneo >800 km from the active volcanoes of the
Sunda Arc and similar distances from oceanic crust of the South
China Sea. The hot interior of Sundaland appears to be the conse-
quence of high upper-crustal heat flow from radiogenic granites
and their erosional products, the insulation effects of thick sedi-
ments, and a high mantle heat flow (Hall and Morley, 2004).
P and S wave seismic-tomography models (Fig. 7) show
that Sundaland is an area of low velocities in the lithosphere
and underlying mantle (e.g., Bijwaard et al., 1998; Lebedev and
Nolet, 2003; Ritsema and van Heijst, 2000; Widiyantoro and
van der Hilst, 1997) in contrast to Indian and Australian conti-
nental lithosphere to the NW and SE, which are colder, thicker,
and stronger. Low mantle velocities are commonly interpreted
in terms of elevated temperature, and this is consistent with
regional high heat flow, but they may also partly reflect elevated
mantle volatile contents, partial melting, or seismic anisotropy
(Lebedev and Nolet, 2003). The high heat flows seen region-
ally across Sundaland, and the generally low mantle velocities
observed in tomographic models, suggest mantle heat-flow val-
ues on the order of 40 mW/m2 (Hall and Morley, 2004), which
are at the high end of the range estimated globally (Artemieva
and Mooney, 1999). Thus they suggest that the region is under-
lain by a thin and weak lithosphere.
Similar lithosphere has been identified in other subduction
zone backarcs (Hyndman et al., 2005). Subduction has produced
a different mantle beneath the upper plate, associated with a lith-
osphere that is not like that of older stable continents and their
margins. The high heat flow and thin, weak lithosphere are the
consequence of long-term subduction at the Sundaland margins,
notably at the Indonesian arcs. These features have influenced arc
development in some ways not identified in many arc models.
ISLAND ARC MODELS
Many models of island arc development were formulated in
the 1960s and 1970s during the rapid period of development of the
plate tectonic theory (e.g., Dewey, 1980; Dewey and Bird, 1970;
Dickinson, 1973, 1974, 1977; Dickinson and Seely, 1979; Karig,
1971), and they are still commonly reproduced in textbooks.
Much of the subsequent work on island arcs in the past 30 yr has
focused on geochemistry and magmatic processes, as it seems
to be widely thought that our knowledge of tectonic processes
in volcanic arcs is complete. A common arc model (Fig. 8) sug-
gests that after the initiation of subduction (a problem in itself),
a volcanic arc develops, typically between 70 and 100 km from
the subduction trench (Dickinson, 1973, 1977; Karig and Shar-
man, 1975); this width increases with the age since arc initiation.
An accretionary prism formed by progressive transfer of material
from the downgoing slab to the upper plate develops near the
trench (Scholl et al., 1986). As time proceeds the accretionary
prism becomes larger and wider (Dickinson, 1973, 1977), under-
thrusting leads to elevation of the arcward part of the prism which
may form a forearc high, and the arc-trench gap progressively
widens. At the same time, debris from the volcanic arc is depos-
ited in the forearc in basins in front of the arc, and the size and
width of these basins increase with time. Continued arc magma-
tism is interpreted to lead to thickening of the crust beneath the
arc. The backarc region is usually not shown in the tectonic mod-
els; in some arcs backarc basins are formed, floored by oceanic
crust, whereas in others no oceanic backarc basins are formed.
The reasons for the formation of backarc basins, which are found
mainly in the western Pacific, are still not clear. However, these
basins are some distance from the arc, typically >100 km. Little

Figure 7. Depth slices through S20RTS S-wave tomographic model (Ritsema and van Heijst, 2000) for SE Asia. High
shear velocities are represented by blue, and low shear velocities by red, with an intensity that is proportional to the am-
plitude of the shear velocity perturbations. The range in shear velocity variation is given below each map.
36 Hall and Smyth

Volcanic arc Forearc basin

A Trench
Oceanic crust B Structural in
folding (kneading)

crystalline

Offscraped oceanic
deposits

SEDIMENT
sedimentary SUBDUCTION
C
Subduction complex
ocean crust
crystalline

Lithosphere
small
Forearc basin accretionary
body
D
Asthenosphere

crystalline
Forearc basin
SUBDUCTION
EROSION

SEDIMENT
SUBDUCTION
E

30 km
CRATONIC
MASSIF

SUBDUCTION 50 km
100 km EROSION

Figure 8. Some models for arc development. (A) The arc-trench gap increases in width as progressive accretion occurs at the trench (modified
from Dickinson, 1977). As discussed in the text, this type of model is based in part on observations from Indonesia, but new mapping suggests
a less important role for subduction accretion. (B, C, D, E) The range of tectonic processes at subduction margins, from accretion to erosion,
modified from Scholl et al. (1980).

attention has been given to the area directly behind the volca-
nic arc, and most arc-derived debris is usually shown to move
into the forearc rather than into the area behind the arc. In recent
years, in many accounts of arcs in other regions, and in modeling
of arcs, the arc model has become more sophisticated. Partly as
a result of discoveries made by the Ocean Drilling Program, fea-
tures such as subduction erosion (e.g., Kay et al., 2005; Ranero
and von Huene, 2000; von Huene and Scholl, 1991) and chang-
ing slab dips (e.g., Funiciello et al., 2003; Kay et al., 1987; Van-
nucchi et al., 2004) have been interpreted as important influences
on arc development (Fig. 8). However, there has been no ocean
drilling of the Indonesian arcs, and few offshore investigations
of the forearcs; we are still dependent on field studies and the
knowledge acquired by hydrocarbon exploration in Indonesia to
interpret arc history and tectonic processes.
The Indonesian arcs differ from these models in a number
of ways. Arc magmatism may be discontinuous despite continu-
ous subduction. In many Indonesian arcs there is little evidence
of major addition of new material by accretion at the trench, but
equally there is little evidence for subduction erosion. There are
sudden movements of the position of the volcanic arc, and there
may be evidence of significant thrusting within the arc itself,
not always with obvious cause; collisions, subduction accretion,
subduction erosion, or changing angles of subduction do not
offer solutions. In general, in the Indonesian arcs there is evi-
dence of a staccato development, possibly related to processes
 Cenozoic arc processes in Indonesia
of arc-continent collision but in some cases with no clear reason.
Observations of seismicity on the basis of well-located hypo-
centers (Engdahl et al., 1998; England et al., 2004) show that
along the modern arc there are sectors that are aseismic or signif-
icantly less seismically active in comparison with other sectors.
This may indicate differences in coupling of the downgoing and
overriding plate at the subduction zone, and variations in cou-
pling and decoupling in time as well as space could contribute
to a discontinuous tectonic history. This could also be related to
dynamics of flow in the mantle wedge (S. Lamb, 2006, personal
commun.; Oncken et al., 2006).
Large-scale dynamic processes have undoubtedly influenced
the stratigraphic record of the Indonesian arcs. However, smaller
scale effects are also observed. The growth of arc volcanoes has
in some parts of the region contributed to basin development by
flexural loading. Furthermore the products of the Indonesian vol-
canic arcs have been modified by processes related to eruption and
weathering in a tropical setting. Both these features may have rel-
evance to the interpretation of ancient arcs and their stratigraphic
record, as discussed below. We begin with the larger scale effects
and move on to more subtle features of arc stratigraphy.
Magmatism and Subduction
The association between subduction and magmatism is well
known. Early plate tectonic models (e.g., Oxburgh and Turcotte,
1970) attributed magmatism to frictional heating at the Benioff
zone, melting of the subducting slab, and compressional shorten-
ing of the overriding plate in the arc-trench gap. Melts are now
considered to originate primarily in the mantle wedge above
the subduction zone and are thought to result from the input of
volatiles that lower the mantle solidus. In Indonesia, almost all
Cenozoic magmatism seems to have been influenced by subduc-
tion, although it has not always been the direct result of melting
of the mantle wedge above the subducting slab. In some cases
a subduction signature is inherited from previous subduction
events. The geochemistry of igneous rocks (e.g., Pearce and
Peate, 1995) is often used to infer the tectonic setting in ancient
arcs, but if a subduction signature is inherited, volcanic arcs may
be mistakenly identified in ancient orogenic belts as subduction
related when they are not. Macpherson and Hall (1999, 2002)
highlighted areas in Indonesia where there has been arc-like
magmatism but subduction is not active, such as south Sulawesi
(Polvé et al., 1997) or northern Java, where volcanoes are very far
from the active arc and have an ultra-potassic character (Edwards
et al., 1994; Macpherson, 1994). In these places the character of
magmatism reflects older subduction events that have enriched
the mantle.
However, the association between magmatism and subduc-
tion is commonly not straightforward. Some arc models portray
compressional arcs with active magmatism, and contraction is
often thought to be typical of active arcs, probably because the
subduction model causes us to think of plate convergence. In
the Indonesian arcs, contraction is commonly associated with
37
cessation of arc activity. Hamilton (1988, 1995) pointed out that
subduction zones are often portrayed as static systems with the
subducting slab rolling over a stationary hinge and sliding down
a slot fixed in the mantle, but this is not a good model. He empha-
sized the necessity of viewing subduction in a dynamic way and
argued that it is most commonly driven by gravity acting on the
subducting slab, and most subduction zones are characterized by
retreat of the hinge with time, or rollback. Extension of the over-
riding plate is known to be common in arc settings (e.g., Chase,
1978; Dewey, 1980; Elsassar, 1971; Hamilton, 1988, 1995) and
may be the normal condition for a volcanic arc. For the period
since 25 Ma, tectonic models (Hall, 1996, 2002) suggest that
hinge retreat (Fig. 9) in most subduction zones in SE Asia and
the western Pacific was accompanied by significant arc volca-
nism (Macpherson and Hall, 1999, 2002) and in some zones by
marginal basin formation.
Extension in arcs may be essential at regional and local
scales to induce melting, to provide pathways for magmas, and
to vary the rate of supply of volatile components to the mantle
wedge. In particular, hinge retreat allows replenishment of the
mantle wedge by an inflow of hot, undepleted mantle (Andrews
and Sleep, 1974; Furukawa, 1993). This can also cause ablation
of the base of the overriding plate (Furukawa, 1993; Iwamori,
1997; Rowland and Davies, 1999), and the mantle that upwells
may melt by decompression. These processes can maintain
steady-state arc magmatism at a relatively constant distance from
the trench over prolonged periods.
Where subduction is characterized by a fixed hinge, or
by hinge advance, magmatism may decline or cease, although
subduction continues. A fixed hinge may result from coupling
of the subducting and overriding plate or from collision, caus-
ing the two plates to move in the same direction as the slab is
being subducted. In these cases, magmatism may cease, although
subduction continues, simply because the mantle wedge is not
being replenished, becomes depleted, and can no longer melt. For
the period since 25 Ma (Fig. 9), hinge advance in SE Asia and
the western Pacific was accompanied by reduction in magma-
tism (Macpherson and Hall, 1999, 2002). The Sunda Arc pro-
vides a good example. From Java eastward the hinge advanced
from ca. 20 to 10 Ma, and magmatic activity declined. In East
Java, Paleogene volcanic activity culminated with a short-lived
explosive phase (Smyth et al., this volume) and then declined sig-
nificantly or terminated; magmatism resumed at ca. 12–10 Ma.
There is no indication that subduction ceased or slowed during
the period of reduced magmatism, and the Indian-Australian
plate continued to move northward relative to Eurasia at a similar
rate to that during the Paleogene. The reduction in magmatism
can be understood as the result of hinge advance driven by the
collision of Australia in eastern Indonesia (Hall, 1996, 2002) and
manifested by counterclockwise rotation of the Borneo region
(Fuller et al., 1999). When rotation of Borneo ceased, the hinge
retreated again and magmatism resumed, especially in the Banda
Arc, where dramatic retreat of the subduction hinge required a
massive influx of fertile mantle beneath the overriding plate.
 25-5 Ma
Japan
25-5 Ma

u
u ky Izu-Bonin
Ry 25-5 Ma

Hinge
Advance

Phil
ippi
Hinge
nes
Mariana Retreat
25-15 Ma
Sangihe Mariana
Sunda 15-5 Ma
25-5 Ma

Halmahera
15-5 Ma Solomons
New
Sunda-Banda Hebrides
20-10 Ma 10-5 Ma

A: Andaman Sea
5-0 Ma BS: Banda Sea
Bi: Bismarck Sea
M: Mariana Trough
Wd: Woodlark Basin
Japan

u
u ky Izu
Ry Bonin

Hinge
Phil

Manila Advance
ippi

Trench M
ne T

Hinge
A
Mariana Retreat
renc
Sa
ng

h
ihe

Halmahera
North
Sulawesi
Bi New Britain
Solomons
Sunda
BS
da Wd
Ban

Figure 9. Movements of subduction hinges in the Neogene, modified from Macpherson and Hall (2002). Regional plate reorganizations occurred
at ca. 25 Ma and 5 Ma. Major coastal outlines are shown for reference. For subduction zones shown without shading, there was no significant
movement of the hinge. Bold letters indicate areas of young marginal basin formation.
 Cenozoic arc processes in Indonesia
Subduction Accretion and Subduction Erosion
A feature of many early arc models is an increase in the width
of the arc-trench gap with increasing age, commonly attributed to
accretion, and some of these models were based on observations
from Indonesia (e.g., Dickinson, 1973; Moore and Karig, 1980).
In contrast, in many arcs the role of subduction erosion in the
development of the arc has become widely recognized as impor-
tant (e.g., Scholl et al., 1980; von Huene et al., 2004). In Indo-
nesia these contrasting processes are difficult to assess, because
the age and nature of the crust between the arc and the trench in
Indonesian arcs is unexplored or poorly known. There is a great
need for detailed exploration of these parts of all arcs. However,
observations on land suggest that neither subduction accretion
nor erosion has had a significant impact on arc history.
Change in Width of Arc-Trench Gap with Age
There is good evidence of Early Cretaceous subduction and
widespread evidence from Sumatra to SE Borneo of middle to
Late Cretaceous collisions accompanied by ophiolite emplace-
ment (Miyazaki et al., 1998; Parkinson et al., 1998; Wakita, 2000).
As noted above, there is almost no Paleocene stratigraphic record
in Indonesia, and the whole region seems to have been emergent.
In many parts of Indonesia, poorly dated, typically terrestrial
rocks interpreted to be Paleocene or Eocene rest unconformably
on older basement rocks. In East Java the oldest sedimentary
rocks that rest unconformably on ophiolitic basement are middle
Eocene or older and lack volcanic debris (Smyth, 2005; Smyth
et al., this volume). There is little evidence for latest Cretaceous
to early Eocene volcanic activity in most of the Sundaland mar-
gin between Sumatra and Sulawesi. The Cenozoic stratigraphic
record, commonly volcanogenic, begins in the middle Eocene,
and in Java the current phase of subduction along the arc began
after 45 Ma, when Australia began to move rapidly northward.
Although it is commonly assumed that subduction at the Sunda
Trench in the Late Cretaceous (e.g., Metcalfe, 1996; Hall, 2002)
continued into the Cenozoic, we question this assumption. Ham-
ilton (1988) stated that the modern subduction system in Java was
inaugurated no earlier than late Oligocene time. We, too, consider
that the present subduction began only in the early Cenozoic,
although we now know of evidence for subduction beneath Java
before the late Oligocene (Smyth, 2005; Smyth et al., this vol-
ume). However, there is no evidence in Java to support the sugges-
tion of late Paleogene southward subduction or an arc collision in
or with Java in the Paleogene (Hamilton, 1988).
In the Sunda Arc between Sumatra and Java the arc-trench
gap is rather large, between 300 and 350 km (Fig. 1), and subduc-
tion began in the early Cenozoic. If the arc-trench gap increased
progressively with age, there should be evidence of movement
of either the volcanic arc or the trench during the Cenozoic. In
East Java the volcanic arc remained in the same position from the
Eocene to the early Miocene (Smyth, 2005; Smyth et al., this vol-
ume). It then ceased activity, and a new arc formed ~10 m.y. later,
50 km to the north, where it has remained since the late Miocene.
39
Near the trench there is some evidence for a young accretion-
ary complex constructed against an older backstop (Fig. 10),
interpreted to be an older accretionary complex (Kopp et al.,
2002, 2006; Schlüter et al., 2002), but this is only ~50 km wide.
Schlüter et al. (2002) suggest there was a progressive growth of
the arc-trench gap south of Sumatra since the Paleocene, but the
age control on this development is poor. In contrast, Kopp et al.
(2006) suggest there has been subduction erosion at the Sunda
margin south of Java, and if this is correct, the arc-trench gap
must have been wider in the past. The seismic lines on which
these interpretations are based do not show a progressive growth
of the accretionary zone but rather a young accretionary com-
plex juxtaposed against an older one (Fig. 10). This relationship
could equally be interpreted in terms of an abrupt change in the
subduction history, and we note that on land there is evidence for
such a change; the volcanic arc ceased activity at ca. 20 Ma and
resumed activity ~10 m.y. later, 50 km north of its previous posi-
tion. Why this happened is not clear. At present the available data
are simply not adequate to distinguish different models.
However, investigations of the Sumatra forearc in the Nias
region do provide the basis for testing the interpretation of arc
growth by continuous accretion. Nias, at the north end of the
Sumatra forearc (Fig. 1), is commonly cited as the classic exam-
ple of a forearc high elevated by accretion, following the work of
Moore and Karig (1980) on the basis of their mapping of part of
the island. Later mapping of the whole island and other islands of
the forearc does not support this model (Samuel, 1994; Samuel
et al., 1995, 1997; Samuel and Harbury, 1996). There is little evi-
dence on Nias for the progressive addition of material. The ophi-
olitic basement includes Cretaceous rocks and locally is overlain
unconformably by a thin Eocene marine cover. Most of the island
consists of a thick sequence of Oligocene to lower Miocene deep
marine clastic sedimentary rocks and a thick sequence of lower to
upper Miocene shallow marine clastic sedimentary rocks. There
are a few limestones and some tuff layers in the sequence. The
island became emergent in the Pliocene. The ophiolitic base-
ment was in place by the Eocene, and the deep marine sedimen-
tary rocks above the basement were deposited on it. However,
the sedimentary rocks are not material scraped off the downgo-
ing plate but represent the fill of a forearc basin, later inverted
(Fig. 11), in which most material was carried from the direction
of Sumatra. The Oligocene to early Miocene development of the
outer forearc was characterized by extension, followed by early
Miocene inversion. Mélanges are not subduction accretion com-
plexes but are diapiric in origin and formed by mud volcanism
that continues today. The forearc islands have a complex history
of local uplift and subsidence to which addition of material at the
trench has contributed, but the Sumatra forearc does not fit very
well to the model of continuous widening and accretion inter-
preted from the early land-based studies.
Mélanges
Mélanges were formed at numerous stages in the develop-
ment of arcs in Indonesia (Hall and Wilson, 2000; Hamilton,
40 Hall and Smyth

SW NE
0 JAVA
A SEA Outer Arc High Mentawai
2 Forearc Basin
INDIAN B
OCEAN C
4
TWT (s)

10°S Java Mentawai

Trenc Wrench
100°E h 110°E Accretionary Wedge I
6 Fault
Accretionary Wedge II
Sunda Trench
8 Continental crust

10 A Oceanic crust of Indian Ocean Plate 50 km

SW ACCRETIONARY DOMAIN NE
0 FOSSIL Forearc Strata
ACTIVE
Trench infill
FRONTAL PRISM
-10 OUTER HIGH OCEANIC-TYPE SHELF
CRUST
Depth (km)

Oceanic mantle Sub Shallow Mantle

duct
ing S
-30 lab

B 50 km
-40
S OUTER FOREARC HIGH FOREARC BASIN N
0
Basin Sediments:
1 Slope Deposits Hemipelagic / Pelagic sediment Normal
and Volcanic Ashes
Faulting
2
Trench Vertical Movement Post-sedimentary
TWT (s)

and Doming buckling

3

6
C 40 km

Figure 10. Cross sections across the Sunda forearc, based on marine geophysical investigations. (A) At the south end of Sumatra the accretionary
wedge is interpreted to be ~300 km wide and constructed against continental crust beneath the forearc basin (Schlüter et al., 2002). (B) Interpret-
ed tectonic units of the Sunda margin south of West Java (Kopp et al., 2002), where the accretionary domain is interpreted to have an active part
and an older part, juxtaposed against oceanic-type crust beneath the forearc basin. (C) Section south of Central Java (Kopp et al., 2006) shows
an uplifted forearc high, covered by thin sediments interpreted as evidence of subduction erosion. The age of the unconformity is unknown, but
the thin sediments above the unconformity suggest a young age. A similar unconformity on land is late Miocene to Pliocene, which developed
after northward thrusting within Java.

1979). Marine geophysical studies suggest that mélanges are
formed at Indonesian subduction zones (Kopp et al., 2002;
Schlüter et al., 2002), but they are also formed at later stages dur-
ing collision (Silver and Moore, 1978). In the central Molucca
Sea (Fig. 12) mélanges were formed at two stages. Mélanges
reported from Talaud (Moore et al., 1981) and present on Mayu
were not formed during the present arc-arc collision but are older
rocks forming part of the pre-Neogene basement of the Sangihe
forearc. Presumed mélanges of the present collision complex are
all submarine and constitute part of the bathymetrically shallow
and seismically incoherent volume of sediment in the central
Molucca Sea. Mélanges have also formed far from subduction
zones without collision (Samuel et al., 1997). In Timor, mélange
formation is attributed by some authors mainly to thrusting
related to collisional processes (Harris et al., 1998) and by oth-
ers mainly to diapiric processes expressed as modern eruption of
mud volcanoes (Barber et al., 1986). Active mud volcanism in
East Java (Smyth, 2005), behind both the Paleogene and mod-
ern arcs, erupts rock samples from deeper parts of the basin
directly north of the modern arc and brings to the surface blocks
 Cenozoic arc processes in Indonesia 41

Lahewa Sub-basin Mujoi Sub-basin Gomo Sub-basin Mola Basement High

MID-MIOCENE– SEDIMENT & OLIGOCENE–

MELANGE LATE PLIOCENE–
EARLY PLIOCENE EARLY MIOCENE
COMPLEX PLEISTOCENE
Coast SEDIMENT Coast

MID-MIOCENE–
EARLY PLIOCENE

BASEMENT
SW NE

0 V=H 10 Km

SW NE
Accretionary Forearc Sumatran fault zone,
Indian Ocean Trench wedge Nias magmatic arc & backarc
basin

Continental crust

0 V=H 60 Km
Moho

+
Overriding
continental
lithosphere

Figure 11. Cross sections across the Sumatra forearc, modified from Samuel and Harbury (1996). Mapping of Nias by Samuel (1994) shows
that the area between Sumatra and Nias was in extension for much of the Cenozoic and that most of the island’s sediments were derived from
Sumatra. The island has not emerged as a forearc high as the result of progressive subduction accretion. V=H—vertical and horizontal scales
are equal.

of the oldest parts of the basin sequence entrained in overpres-
sured muds. This is clearly not related to subduction because the
Benioff zone is >100 km below the site of mud volcanism. The
mélanges at the surface, and the presumed equivalents beneath
the surface, are the result of overpressures related to rapid depo-
sition of thick sedimentary sequences, hydrocarbon generation,
and young thrusting very far from the subduction zone.
Mélanges and Ophiolites
The emplacement of ophiolites is often linked to the pro-
cesses of mélange formation and accretion. Between the Sangihe
and Halmahera Arcs (Fig. 12) is a collision complex in which
there are ophiolites and mélanges. The Talaud Ridge at the center
of the Molucca Sea was interpreted on the basis of marine geo-
physics as a mélange wedge (McCaffrey et al., 1980; Silver and
Moore, 1978) including slices of the Molucca Sea lithosphere
(Fig. 13). These rocks are exposed on Talaud and on two tiny
islands along the ridge, Mayu and Tifore. The ophiolites were
interpreted by McCaffrey (1983, 1991) as part of the subducted
Molucca Sea plate, which has been emplaced by the collision
of the two arcs (Fig. 13). However, on Talaud a post–middle
Miocene sedimentary sequence rests unconformably on ophio-
lites that are middle Eocene or older (Moore et al., 1981). The
middle Miocene to Pleistocene rocks are tuffaceous sandstones,
siltstones, and shales with intercalations of limestone, marl, and
conglomerate in which the sediments are dominated by volcanic
debris. They were deposited in deep water by turbidity currents.
On Talaud, Neogene strata are in many places little deformed,
and they are moderately to strongly deformed locally and typi-
cally in narrow zones adjacent to mélange rocks or ophiolites.
On Mayu, coastal exposures of highly indurated mélange yield
Eocene ages (Baker, 1997). All of the mélange is probably older
than middle Miocene, and none appears to be related to the pres-
ent collision.
The similarity in character and structure of the ophiolitic
rocks to those of the basement complex on Halmahera (Hall
et al., 1988a) suggest that much of the mélange formed dur-
ing Eocene or older deformation events. The interpretation of
the ophiolites as part of the subducted Molucca Sea lithosphere
requires some complex arguments (McCaffrey, 1991) to account
42 Hall and Smyth
EURASIAN PLATE
Sulu Sea
km
00
18
Celebes Sea
Sulawesi
Banda Sea
1800 km
Figure 12. Molucca Sea region, showing the converging Halmahera and Sangihe Arcs, displayed in a three-dimensional (3D) diagram that
represents, in simplified form, the geometry of the converging plates in east Indonesia. Modified from Hall et al. (1995).
for their elevation instead of their disappearance by subduction.
In contrast, if the ophiolites are interpreted as part of the Eocene
or older basement (Fig. 13), as they are in Halmahera, and they
formed the basement to the forearc, the gravity model is much
easier to understand. Where the Halmahera forearc and arc have
been significantly overthrust the Sangihe forearc has been jacked
up. The wide Molucca Sea collisional complex is composed
of the accretionary wedges of both arcs. The basement of the
Sangihe forearc is exposed where it thrusts over this wedge. The
ophiolitic rocks of the central Molucca Sea are not part of the
Molucca Sea plate but are the basement of the Sangihe forearc
(Hall, 2000). The ophiolites were the basement before the mid-
dle Miocene.
Accretion or Erosion?
In many Indonesian arcs, there seems to have been relatively
little growth of the forearc region by the addition of material
from the downgoing plate. Many so-called accretionary com-
plexes are in fact dominated by material derived from the arc. At
the northwestern end of the Sunda Arc there is undoubtedly sedi-
ment at the distal end of the Bengal Fan (Curray, 1994) that is
being added to the overriding plate, although most of this is prob-
ably mud-dominated sediment (Stow et al., 1989). Farther south,
south of south Sumatra and Java, the amount of sedimentary
cover on the subducting plate is much thinner, yet the arc-trench
Philippine
Fault Philippine
Trench
PHILIPPINE SEA
Mindanao PLATE
Cotobato
Trench
Sangihe Arc
Halmahera Arc
Halmahera
MOLUCCA SEA
PLATE
SORONG FA
ULT ZONE
New Guinea
AUSTRALIAN PLATE
gap is just as wide. The role of the upper plate in providing mate-
rial is well illustrated at the eastern edge of the Philippine Sea
plate in the Nankai margin, where classic fold and thrust struc-
tures are formed in the thick “accretionary wedge” (Mikada et
al., 2005; Taira et al., 1991). The sediment in the upper part of the
prism is dominated by clastic material carried along the trough
from the Japan–Izu-Bonin Arc collision zone, and relatively
small amounts of material are contributed from hemipelagic
sediments on the subducting slab (Karig and Ingle et al., 1975;
Klein and Kobayashi et al., 1980; Taira et al., 1991). In Indonesia
the role of the structurally upper plate is well illustrated by the
eastern Makassar Strait, where there has been no subduction, yet
accretionary-style structures have developed in sand-dominated
sedimentary wedges that have built out from Sulawesi (Puspita et
al., 2005). These structures closely resemble the structures devel-
oped in subduction accretionary complexes, and major eastward
subduction during the Cenozoic has been suggested by some
authors (e.g., Charlton, 2000; Guntoro, 1999). However, most
authors have interpreted the West Sulawesi margin to be either
a passive margin with oceanic crust to the west (e.g., Bergman
et al., 1996; Cloke et al., 1999; Hall, 1996; Hamilton, 1979) or a
rifted margin with the thinned continental crust to the west (e.g.,
Calvert, 2000; Situmorang, 1982). Recent offshore seismic data
show that there has been no subduction (Fraser et al., 2003; Pus-
pita et al., 2005) and suggest that the Makassar Strait is floored
 Accretionary complex
Accreted
SANGIHE TALAUD sediments and crust SNELLIUS RIDGE
of Halmahera forearc MINDANAO
Forearc basin Backarc basin
0 v
v
v
v
v
v
v v v v v v
v v v v v v v v v v
10 v v v v v v v v
v Arcv crust v
v v Forearc crust v v v v Backarc crust
v v v v v
v v
v v v
v Arc
v
crust v
v
v
v
20 v
v
v
v
v
v
v
v
v v
v
v v
v v v v v v
Talaud
30
EURASIAN PLATE rc c
ru st
PHILIPPINE SEA PLATE A
Kilometers

40 Sangihe Arc rea Halmahera Arc

1
Fo
50 Snellius
Sangihe Ridge
60 Sangihe mantle Halmahera mantle

70 B
80
90
MOLUCCA SEA PLATE A Mayu
Morotai

A
ES
C

ER
100

AH
LA

LM
SU

HA
s

NW SANGIHE
Accretionary complex
MOROTAI
SE
Forearc basin Backarc basin
Forearc basin
0 v v v v v v
v v v v v
v v v v v v v v v
v v v
10 v v v v v Forearc crust Forearc crust v v
Arc crust v v v
v v v v v v
Backarc crust
v Arc crust v v v v v
v
v
v
v
v
v
v v v
20 v v v v v v v v v v v v v
v v v v v v v
v v
30
Kilometers

40
EURASIAN PLATE PHILIPPINE SEA PLATE
Sangihe Arc Halmahera Arc
50
60
Sangihe mantle Halmahera mantle
70 MOLUCCA SEA PLATE
80
90
B
100
100
50
Calculated MGAL
0
Observed -50
-100
Talaud-Mayu -150
0 50 100 Ridge 150 200 250

-0.59 -0.59 -1.83 km

-0.43 -0.43 10
0.40

20

0.44 C 30

WNW Sangihe Arc Halmahera Arc ESE

100 0 km 100 200 300
0
10
20
2.86
30
Kilometers

40
50 3.40 g/cm3
60
70
+0.20 +0.20
80 3.34 +0.05 +0.05
90 Mantle density D
100

Figure 13. Cross sections across the Molucca Sea (Hall, 2000) drawn at the same vertical and horizontal scales (A, B) in comparison with those
based on gravity modeling (C, D) by McCaffrey et al. (1980). In section A at the latitude of Talaud, the entire arc and forearc of the Halmahera
Arc has been overridden by the Sangihe forearc. Ophiolites interpreted as part of the Sangihe forearc basement are exposed in the Talaud Islands.
Farther south (section B), only part of the forearc has been overridden, but the Halmahera Arc in Morotai was overridden by its own backarc in
an earlier thrusting episode. In contrast, McCaffrey et al. (1980) interpret the ophiolites as part of the subducted Molucca Sea plate.
44 Hall and Smyth
by extended continental crust (Nur’aini et al., 2005; Puspita et
al., 2005). The structures are foreland-type fold and thrust belts,
and developed as westward thrusting has progressed since the
Pliocene, all the material having been derived from Sulawesi.
There certainly has been addition of material to the continen-
tal margin of Sundaland during the Cenozoic (Fig. 4). However,
the additions occurred during relatively short periods of time, at
widely spaced intervals, and were related to the collision of con-
tinental fragments at the Sundaland margin. The material added
included arc and ophiolitic rocks, which may in some cases rep-
resent backarc basins within the continental margin or the rem-
nants of material subducted during the arc-continent collision. In
Indonesia, arc growth seems to have occurred discontinuously
rather than by continuous steady-state addition of material at the
trench that led to gradual widening of the arc-trench gap.
Disappearance of Arcs
Westward subduction of the Molucca Sea beneath the
Sangihe Arc probably began in the early Miocene. Eastward sub-
duction of the Molucca Sea plate beneath Halmahera began in
the middle Miocene. The double subduction zone was initiated
at this time, forming a new plate, the Molucca Sea plate, separate
from the Philippine Sea plate. The oldest volcanic rocks dated
from the Halmahera Arc are 11 Ma in Obi at its southern end, and
they become younger to the north (Baker and Malaihollo, 1996).
The Molucca Sea was eliminated from south to north, and the
two forearcs began to collide. The result of the collision was that
the Halmahera Arc system (meaning the entire region between
the trench and backarc region) failed repeatedly, with thrusting in
different directions at different stages in the collision. First, the
backarc was thrust over the volcanic arc, and later the forearc was
thrust toward the volcanic arc. In south Halmahera the backarc
region was thrust onto the forearc, in places entirely eliminating
the Neogene arc. At the southern end of the Halmahera Arc on
Obi the arc was thrust onto the forearc (Ali and Hall, 1995; Ali et
al., 2001). After west-vergent thrusting, volcanism in the Halma-
hera Arc resumed between Bacan and north Halmahera. At the
south end of the arc on Obi, and at the north end from Morotai
northward, volcanism ceased. In the northern Molucca Sea the
Sangihe forearc was then thrust east onto the Halmahera forearc
and arc. In the region between Morotai and the Snellius Ridge,
parts of the Neogene Halmahera Arc and forearc have now dis-
appeared. Farther south, this east-vergent thrusting carried the
Halmahera forearc onto the flanks of the active Halmahera Arc,
and pre-Neogene rocks of the Halmahera forearc basement are
now exposed in islands of the Bacan group and off the coast of
northwest Halmahera.
Cross sections drawn across the present-day collision zone
from south to north can also be considered to display the sequence
of events in time, and a series of sections illustrating the earlier
stages in the collision can be inferred from the geology of the
Halmahera Arc (Fig. 14). An interesting consequence of the col-
lision is that the Halmahera Arc is being progressively overrid-
den by the Sangihe Arc, and in the northern Molucca Sea almost
the whole of the forearc and volcanic arc have disappeared. This
process is not complete, and it is likely that much, if not all, of
the Halmahera Arc will disappear beneath the Sangihe Arc. This
process is certainly not subduction erosion, as intended by most
authors (Scholl et al., 1980; von Huene and Scholl, 1991; von
Huene et al., 2004). However, neither is it subduction accretion,
as usually envisaged. The arc-trench gap of the Sangihe Arc has
increased since the two arcs came into collision, but the arc-
trench gap of the Halmahera Arc has diminished.
Arc Movements and Changing Slab Dips
Observations in Indonesian arcs raise the question of whether
there is such a thing as a compressional arc. Arcs may certainly
be compressed, but when contraction occurs volcanic arc activity
ceases. Magmatism may resume when compression ceases, but
the stratigraphic record suggests it does not persist during con-
traction. In the Banda Arc, magmatic activity terminated in the
Wetar sector of the volcanic arc at ca. 4 Ma, when arc-continent
collision began in Timor, and was never resumed during stacking
of nappes (Audley-Charles, 2004). In the Halmahera Arc, vol-
canic activity ceased as arc-arc collision began but resumed in a
new location shortly afterward. The repeated failures in different
places in the entire arc system in Halmahera suggest that inter-
mittent thrusting was followed by periods of relaxation in which
arc magmatism resumed.
Such abrupt movements in the position of the volcanic arc
are not uncommon and may follow periods when an arc is com-
pressed, but they may be due to other tectonic causes. Sudden
shifts in position of the volcanic arc could be related to shorten-
ing of the entire arc system (forearc, arc, and backarc), which
may involve thrusting of the backarc or the forearc toward the
arc, the magmatic arc itself being the weakest point. Field obser-
vations in Halmahera show that the site of thrusting was close
to the former active arc. Experimental and numerical modeling
studies also show that the active arc is the weakest point in the
whole arc system (Shemenda, 1994; Tang and Chemenda, 2000;
Tang et al., 2002). These studies show that failure occurs directly
beneath the arc. It can take place on faults dipping toward the
trench or arc, and this depends on the thickness of subducted
crust and flexural rigidity of the upper plate.
Since the middle Pliocene (Fig. 14) the Halmahera Arc sys-
tem failed more than once close to the site of the active volcanic
arc, presumably reflecting its weakness owing to mineralogy and
magmatism. Active volcanism has most recently resumed in a
restricted area at the center of the arc chain, over a distance of
~200 km, in comparison with the 700 km length of the early Plio-
cene arc. The arc-arc collision explains why the arc volcanism
may have ceased and then resumed as stresses changed, but this
does not really account for the change in position of the arc. The
Quaternary volcanoes are situated ~50 km west of the Pliocene
centers (Fig. 15). Hall et al. (1988b) speculated that the shift
may be related to a steepening dip of the subducted slab beneath
Halmahera. A highly complex geometry of subducted slabs lies
 Talaud ridge

0 Ma

Philippine Sea plate

Halmahera Arc and forearc overridden

by Sangihe forearc
Molucca Sea plate sinks deeper
Emergence of ophiolitic basement of
Sangihe forearc

0 Ma

Halmahera forearc overridden by Sangihe forearc

Molucca Sea plate sinks
Local emergence of islands in ‘collision complex’

0 Ma

Halmahera Arc fails again

Overthrusting of forearc region: detachments
deep within basement
Minor backthrusting at back of Sangihe forearc

Sangihe Arc Halmahera Arc

2 Ma

Molucca Sea plate

Volcanic activity ceases in Halmahera arc

Arc completely overridden by backarc region

Figure 14. Cross sections across the Molucca Sea drawn at same vertical and horizontal scales to illustrate the sequence of convergence of the
Halmahera and Sangihe Arcs since 2 Ma. The lowermost section is inferred from geological mapping on land. The upper three sections are drawn
at different latitudes across the Molucca Sea from south (bottom) to north (top). They can be considered to represent the sequence of events in
the last ~2 m.y. that have led to the structure at the latitude of Talaud, where convergence is most advanced. Collision has resulted in the almost
complete elimination of the Halmahera Arc and forearc at the latitude of Talaud.
A 127oE 129oE

I
TA
RO
O
M
o
2N

Modern volcanoes of the

Halmahera Arc
Pleistocene to modern
Molucca Sea Halmahera Arc

Miocene to Pliocene Arc

HALMAHERA

0o Weda Bay

0 50 100
BACAN
Km

B o
110 E
o
112 E
Modern volcanoes of the Sunda Arc

Oligo-Miocene volcanic centers of

the Southern Mountains Arc

7oS

EAST JAVA

SUND
A ARC

SOUT
MOUN HERN
TAIN
ARC S
8oS

0 50 100

Km

Figure 15. Abrupt movements of the volcanic arcs in (A) Halmahera (ca. 2 Ma) and (B) Java (ca. 10 Ma) took place, as discussed in the text.
 Cenozoic arc processes in Indonesia
beneath the Halmahera and Sangihe Arcs and the southern Philip-
pines (Fig. 12). The west-dipping Philippine slab has now arrived
at depths where it would hit the subducted Halmahera slab and
increase its dip, thus shifting the arc to the west if melting occurs
at a constant depth. Macpherson et al. (2003) showed that dif-
ferences in the geochemistry of Quaternary lavas in comparison
with Neogene lavas of the Halmahera Arc are consistent with an
increased sediment flux interpreted to indicate an increase in dip
of the subducted slab.
In East Java, activity in the Southern Mountains Arc termi-
nated in the early Miocene after a period of >20 m.y. at the same
location. About 10 m.y. later a new arc formed, 50 km north of
the older arc, that has remained at the same position since the
late Miocene (Fig. 15). Why the volcanic arc moved north is not
known. There is no evidence for a collision at the Java Trench.
There was Neogene contraction in Java, although it is not well
dated. It is possible that the slab dip remained at the same angle,
the depth to the Benioff zone remained constant, and therefore
contraction of the arc meant that when volcanism resumed the
new arc formed at the same distance from the trench. Several
possible explanations are suggested by Smyth et al. (this vol-
ume), but all are speculative.
The abrupt change in position of the volcanic arc is a feature
of several Indonesian arcs, and the causes are not known. How-
ever, we do not know of examples of gradual movement of the
arc with increasing age of the magmatic arc, as might be expected
from many arc models. This could mean that slab dip gradually
declines as the arc-trench gap increases, so that the arc remains in
the same position. Obviously this idea is difficult to test, as there
is no way of measuring slab dip in the past. The dip of slabs in
most Indonesian arcs is typically steep where the slab reaches
>~200 km in depth. The depth to the Benioff zone beneath active
arcs is also much more variable than previously thought (England
et al., 2004). Perhaps magmatic activity essentially pins the arc in
one position until it ceases, and then there are a number of factors
which could control the location of the resumption of volcanic
activity, including thickness or strength of the crust, preexisting
weaknesses in the crust, or changing dip of the subducting slab.
At present, all are conjectures and illustrate how incomplete our
knowledge of arcs remains.
Volcanic Arc Loading
Volcanoes exert a load. Since most basic to intermediate
arc volcanoes are denser than average continental crust, long-
lived volcanism should contribute to subsidence. Several of
the Indonesian arcs have deep filled basins close to, and both
in front of and behind, the arc. Few of these basins have been
explored seismically, and their deep structure is unknown. How-
ever, there are indications that arc loading may have contributed
to their formation, because the basin sequences thicken toward
the arc, and the timing of basin development is closely linked to
activity in the arc. As discussed by Nichols and Hall (1991) and
Smyth et al. (this volume), the basins are not what are normally
47
considered backarc basins. They are found behind the arc but
are much closer to it than typical backarc basins. They are not
underlain by newly formed oceanic crust, nor are they charac-
terized by obvious extension. They are filled with volcanic arc
material, mainly reworked as sediments. The basins formed in
a setting that may have been extensional (as noted above, pos-
sibly the typical condition of a volcanic arc) or neutral, using
the terminology of Dickinson (1995) and Busby and Ingersoll
(1995), but certainly not compressional. They are not retro-arc
basins, and although there may be evidence of thrusting, this
occurred after the basin formed.
East Java and Halmahera provide two examples. In Java the
deep Kendeng Basin behind the arc is filled with a thick sequence
of volcanic and sedimentary rocks mostly derived from the arc
itself. The basin formed in the middle Eocene, when the volca-
nic arc began its activity. Modeling by Smyth (2005) suggested
that at least part of the accommodation space was created by the
load of the volcanic arc. The basin in Weda Bay (Nichols and
Hall, 1991) behind the Halmahera Arc has been investigated as
part of oil exploration, and seismic lines close to the arc show no
indication of fault control on subsidence (see Waltham et al., this
volume). Again, a likely important contribution to subsidence is
flexural loading by the volcanic arc. This suggestion is explored
in more detail by Waltham et al. (this volume) in a mathematical
model. They conclude that volcanic loading can make a contri-
bution to basin subsidence in arc settings within ~200 km of the
arc, which may be the primary cause. In continental margin arcs
such as Java, and long-lived intraoceanic arcs such as Halmahera,
there is a significant density contrast between the deeper crust
and the eruptive arc products. In these cases the volcanoes form
a load that can produce or contribute to the formation of flexural
basins close to the arc. Elsewhere the absence of deep basins in
arcs may be due to the small density contrast between volcanoes
and the underlying crust.
Sediment Character in Indonesian Arcs
All the Indonesian arcs have been in an equatorial position
throughout the Cenozoic (Hall, 2002), and as a result the arc
products have formed in a tropical climate. Tropical processes
have several effects on rocks and grains within them. Deep tropi-
cal weathering of well-jointed rocks can lead to a high degree of
in situ rounding of material that eventually falls out of the out-
crop (Fig. 16). The onion-skin type of weathering of outcrops
is common in Indonesia, and rounded clasts are formed at all
scales from huge boulders to pebbles. The rounded clasts may
be incorporated in volcaniclastic and sedimentary rocks without
any transport whatsoever. Weathering of material also leads to
rapid destruction of unstable rock fragments and minerals. This
is particularly important for volcanic-derived material that con-
tains many unstable minerals and rock fragments such as mafic
minerals, feldspars, and clays. The interpretation of transport his-
tory and maturity based on grain shape and light mineral modes
can be very misleading in tropical settings. Two recent studies of
48 Hall and Smyth
Figure 16. Tropical weathering of rocks, resulting in rounding of clasts before they are even removed from the outcrop. Rounded clasts are
produced at all scales from small pebbles to boulders. Both photographs show rocks exposed by quarrying of the Pendul Diorite, Jiwo Hills,
East Java.
sedimentary rocks deposited close to active margins in the Indo-
nesian region illustrate the effects of tropical weathering (van
Hattum, 2005; Smyth, 2005). In both cases the tectonic setting
is known, but if these examples had come from much older oro-
genic belts the interpretation could well be very different.
There was subduction of the proto–South China Sea beneath
north Borneo (Hall, 2002; Hazebroek and Tan, 1993; Hutchi-
son, 1996; Tongkul, 1991) between the Eocene and early Mio-
cene. A large sedimentary fan, the Crocker Fan, formed at this
active margin. All sandstones are quartz-rich, and their composi-
tions plotted on conventional ternary diagrams suggest they are
mature recycled orogenic products, but there are some anomalies
(van Hattum et al., 2003; van Hattum, 2005). The oldest sand-
stones show the greatest compositional maturity, and although
younger sandstones plot as recycled orogenic sediments they
are less mature, which is inconsistent with their derivation by
recycling of older sedimentary rocks as suggested by Hutchison
et al. (2000) and William et al. (2003). The sandstone textures
are in marked contrast to their apparent compositional maturity
(van Hattum, 2005; van Hattum et al., 2006). They are immature,
typical first-cycle sandstones; grains are angular to subangular,
and there are few clasts that suggest recycling. The sandstones
are poorly sorted and have a muddy matrix and very low poros-
ity, and they contain abundant euhedral and subhedral zircon
grains typical of first-cycle sandstones. Subrounded and (rare)
rounded grains are less abundant. Tourmaline also occurs pre-
dominantly as unabraded grains. The shapes and lack of abrasion
of zircons and tourmalines indicate that long-distance transport is
unlikely. Zircon and tourmaline typically make up >70% of the
heavy mineral assemblages in the Crocker Fan sandstones. Their
abundance indicates erosion from acid plutonic rocks, and their
shapes suggest a nearby source area. Apatite would normally also
be abundant in material derived from acid plutonic rocks, but it
is commonly absent, and those grains present are typically pitted
or partially dissolved. This suggests that the abundance of apatite
reflects weathering rather than provenance. Apatite is stable dur-
ing burial but is susceptible to acidic weathering, which is com-
mon in humid tropical settings (Morton, 1984). Plots of chemical
indices of weathering and alteration (Fig. 17) show that very few
of the sandstones are similar in composition to the fresh source
rock (van Hattum, 2005). The composition has changed between
erosion and deposition. The tropical setting of the sandstones
needs to be considered before provenance can be interpreted.
In East Java, Eocene to lower Miocene quartz-rich sand-
stones plot on commonly used ternary diagrams as recycled oro-
genic or cratonic interior–derived sediments (Fig. 18), but this
interpretation is even more misleading than that for the Borneo
sandstones. Examination of quartz grains shows that many have
a volcanic origin, and a volcanic provenance is supported by the
abundance of fresh euhedral zircons. The sandstones have previ-
ously been interpreted as eroded from a continental Sundaland
source because they are rich in quartz and well sorted, but careful
examination reveals abundant evidence of their volcanic origin
on the basis of textures, light mineral constituents, quartz charac-
ter, clay mineralogy, and zircon character and ages (Smyth, 2005;
Smyth et al., this volume). These quartz-rich rocks are not the
product of recycling of material derived from continental source
regions but are the result of explosive volcanic events such as the
eruption of crystal-rich magma, followed by air-fall sorting and
subsequent epiclastic reworking (Cas and Wright, 1987; Walker,
1972). In tropical settings, intense weathering rapidly causes the
breakdown of labile minerals, mineral aggregates, and lithic frag-
ments. The redeposited sediment will be rich in resistant minerals
such as quartz and heavy minerals like zircon, and it has a higher
percentage of quartz per unit volume than the source material.
The high rates of weathering observed in tropical settings
can also have a significant influence on the preservation of
unconsolidated volcanic deposits. The potential for preserving
volcaniclastic deposits on steep terrestrial slopes is low unless
they are rapidly overlain by lavas or buried under younger arc
 Cenozoic arc processes in Indonesia 49

100 100
Strongly weathered
90 95 Inc
rea
sin
CIW (100)[Al2O3/(Al2O3+CaO+Na2O)]

80 gc
90 he
mi
ca
70
lm
85 atu
rity
60 80
HUMID

SiO2%
50 75

40 70

30 65
Fresh (source) rock
ARID
20 60

10 55
Strongly altered
0 50
0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25
CIA (100)[Al2O3/(Al2O3+CaO+Na2O+K2O)] Al2O3+K2O+Na2O

Figure 17. (A) Chemical index of weathering (CIW; Harnois, 1988) plotted against chemical index of alteration (CIA; Nesbitt and Young, 1984).
(B) Chemical maturity, expressed as a function of percentage of SiO2 and total percentage of Al2O3 + K2O + Na2O. After Suttner and Dutta (1986)
for Sabah sandstones from van Hattum (2005).

deposits. The breakdown and recycling of soft, nonwelded ashes
can be very rapid in terrestrial settings with intense precipita-
tion. Lahars are common hazards on many of Java’s volcanoes
such as Merapi (Lavigne et al., 2000). Preservation is more likely
if the volcanic material is deposited in a marine setting below
the storm-wave base, but even this material may be apparently
more mature than would be expected in a nontropical environ-
ment because of eruptive sorting of material in the ash cloud.
The East Java sandstones are well sorted, quartz-rich, and clearly
deposited in shallow marine settings, but they were derived from
volcanic material produced by explosive eruptions that occurred
a short period before their deposition. We know of other quartz-
rich sandstones in Indonesia that we and others have previously
interpreted as having a continental provenance, and our experi-
ence of sandstones in Borneo and East Java now causes us to
consider more carefully if some of these interpretations may have
been wrong. Quartz character and heavy mineral studies (Smyth,
2005; Smyth et al., this volume) are of particular importance for
reassessment. In older orogenic belts, similar misinterpretations
could easily be made, especially if the ancient climate is not con-
sidered or known.
CONCLUSIONS: LESSONS FOR ANCIENT ARCS
The Indonesian arcs differ in many ways from those com-
monly portrayed in many arc models. Are the Indonesian arcs typi-
cal, or are they one member of a spectrum of arcs? The absence of
well-documented major eruptions in the Cenozoic stratigraphic
record of Indonesian arcs is likely to reflect their tropical setting
and the lack of detailed studies in difficult terrain and relatively
inaccessible areas. An extremely large eruption, possibly on the
scale of Toba, terminated the Eocene to early Miocene phase of
arc activity in the Southern Mountains of East Java (Smyth, 2005)
at ca. 20 Ma, and other such events will surely be recognized in
the future. The tropical setting means that some features may be
typical only of tropical arcs—for example, the rapidly enhanced
maturity of sandstones and enrichment in quartz in volcano-
genic debris deposited in arc basins. Nonetheless, in ancient arcs
it is still necessary to consider the likely climatic setting before
interpreting provenance, both in terms of transport distances and
source regions.
Other features, such as the importance of extension in arcs,
the influence of volcanic arc loading on sedimentary basin devel-
opment, the critical role of hinge movement in magmatism, and
the weakness of the arc, have more general applicability. Certain
features of Indonesian arcs could also be of general relevance but
could simply reflect particular differences between tectonic pro-
cesses in SE Asia and those of other arc regions—for example,
the absence of a relationship between arc-trench gaps and dura-
tion of magmatic arc activity, emplacement of ophiolites, and for-
mation of mélanges. The long history of subduction in the region
has resulted in a thin, warm, and weak lithosphere, and although
the region is not typical of continental crust it does share many
features with other subduction zone backarcs (Hyndman et al.,
2005). The behavior of Indonesian arcs may be more representa-
tive of arcs in the upper plate of long-lived subduction margins
in which the lithosphere becomes unusually responsive to small
changes in plate boundary forces.
Indonesian arcs appear to have had relatively short lives
in comparison with many ancient arcs, but this is probably a
50 Hall and Smyth

A Q B
CRATON Continental Block
INTERIOR
Recycled Orogen
TRANSITIONAL Magmatic Arc
CONTINENTAL
Q - Total free quartz
F - Feldspar
L - Total lithic
RECYCLED fragments
OROGEN
T

DISSECTED ARC
LIF
UP
NT
ME

TRANSITIONAL
SE

ARC UNDISSECTED
BA

ARC
1mm
F L

C D

200 µm 200 µm

Figure 18. (A) Quartz sandstones from East Java plotted on a QFL ternary diagram, which could be interpreted as indicating a reworked conti-
nental orogenic provenance. The sandstones are actually first-cycle products of rapid reworking of volcanogenic material. Examples of volcanic
quartz grains from the Jaten Formation, Pacitan, East Java. (B) Bipyramidal quartz grain. (C) Scanning electron microscope (SEM) image of a
quartz shard. (D) SEM image of rounded melt embayments in quartz.

reflection of their young age and the greater ease with which
events of different arcs can be resolved in the relatively recent
past. Pre-Cenozoic arcs with lives of tens of millions of years
may include several phases of arc activity, and several different
arcs. The complexity of tectonic evolution in the region shows
that a great deal can happen in a 45 m.y. time span, and much
of this complexity could be missed in the study of ancient arcs.
This could be an explanation of the apparently compressional
arc; short-lived contractional events in older arcs may appear
to be contemporaneous with arc magmatism when in fact they
punctuate arc activity. The history of the Neogene Halmahera
Arc shows that an arc can be formed and destroyed in a short
time and may ultimately leave little trace in the stratigraphic
record. Even in the young arcs of Indonesia, dating of events is
not good, and more and better dating is required to understand
and interpret arc processes. We also need to know more about
the deep structure of arcs. For example, very little is known
about the crust between the arc and trench in any Indonesian
arc, and almost nothing is known of the thickness and character
of the crust beneath the volcanic arc. Most arcs in the western
Pacific are similarly poorly known. At the very least, the record
of the Indonesian arcs shows that some features of arc models
need to be questioned and that our understanding of tectonic
processes in volcanic arcs needs to be improved.
 Cenozoic arc processes in Indonesia
ACKNOWLEDGMENTS
Our knowledge of Indonesian arcs has been acquired through
many projects carried out by the SE Asia Research Group at
London University and Royal Holloway by many different
people. We thank all of them. Our work has been supported at
different times by the University of London Central Research
Fund, the Natural Environment Research Council (NERC), and
the Royal Society, but mainly by a consortium of oil compa-
nies whose membership has changed with time. We have been
fortunate with help and support from colleagues in Indonesia
at the Geological Research and Development Centre Bandung
(now the Geological Survey of Indonesia), Lemigas, Indone-
sian Institute of Sciences (LIPI), and Institut Teknologi Band-
ung. We thank Tim Charlton, Ron Harris, and Dave Scholl for
helpful comments on an earlier version of the paper, and Wim
Spakman for help in producing Figure 7.
REFERENCES CITED
Abbott, M.J., and Chamalaun, F.H., 1981, Geochronology of some Banda Arc
volcanics, in Barber, A.J., and Wiryosujono, S., eds., The Geology and
Tectonics of Eastern Indonesia: Bandung, Indonesia, Geological Research
and Development Centre Special Publication 2, p. 253–268.
Ali, J.R., and Hall, R., 1995, Evolution of the boundary between the Philippine
Sea Plate and Australia: Palaeomagnetic evidence from eastern Indonesia:
Tectonophysics, v. 251, p. 251–275, doi: 10.1016/0040-1951(95)00029-1.
Ali, J.R., Hall, R., and Baker, S.J., 2001, Palaeomagnetic data from a Meso-
zoic Philippine Sea Plate ophiolite on Obi Island, Eastern Indonesia:
Journal of Asian Earth Sciences, v. 19, p. 535–546, doi: 10.1016/S1367-
9120(00)00053-5.
Ambrose, S.H., 2003, Did the super-eruption of Toba cause a human population
bottleneck? Reply to Gathorne-Hardy and Harcourt-Smith: Journal of
Human Evolution, v. 45, p. 231–237, doi: 10.1016/j.jhevol.2003.08.001.
Andrews, D.J., and Sleep, N.H., 1974, Numerical modelling of tectonic flow
behind island arcs: Geophysical Journal of the Royal Astronomical Soci-
ety, v. 38, p. 237–251.
Artemieva, I.M., and Mooney, W.D., 1999, Mantle heat flow in stable conti-
nental regions: A global study: Eos (Transactions, American Geophysical
Union), v. 80, p. F967.
Artemieva, I.M., and Mooney, W.D., 2001, Thermal thickness and evolution
of Precambrian lithosphere: A global study: Journal of Geophysical
Research, v. 106, p. 16,387–16,414, doi: 10.1029/2000JB900439.
Audley-Charles, M.G., 2004, Ocean trench blocked and obliterated by Banda
forearc collision with Australian proximal continental slope: Tectono-
physics, v. 389, p. 65–79, doi: 10.1016/j.tecto.2004.07.048.
Baker, S., and Malaihollo, J., 1996, Dating of Neogene igneous rocks in the
Halmahera region: Arc initiation and development, in Hall, R., and
Blundell, D.J., eds., Tectonic Evolution of SE Asia: Geological Society
[London] Special Publication 106, p. 499–509.
Baker, S.J., 1997, Isotopic dating and island arc development in the Halmahera
Region, Eastern Indonesia [Ph.D. thesis]: University of London, 331 p.
Barber, A.J., Tjokrosapoetro, S., and Charlton, T.R., 1986, Mud volcanoes,
shale diapirs, wrench faults, and melanges in accretionary complexes,
Eastern Indonesia: American Association of Petroleum Geologists Bul-
letin, v. 70, p. 1729–1741.
Ben-Avraham, Z., and Emery, K.O., 1973, Structural framework of Sunda
Shelf: American Association of Petroleum Geologists Bulletin, v. 57,
p. 2323–2366.
Bergman, S.C., Coffield, D.Q., Talbot, J.P., and Garrard, R.A., 1996, Tertiary
tectonic and magmatic evolution of western Sulawesi and the Makassar
Strait, Indonesia: Evidence for a Miocene continent–continent collision,
in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE Asia: Geo-
logical Society [London] Special Publication 106, p. 391–429.
51
Bijwaard, H., Spakman, W., and Engdahl, E.R., 1998, Closing the gap between
regional and global travel time tomography: Journal of Geophysical
Research, v. 103, p. 30,055–30,078, doi: 10.1029/98JB02467.
Bock, Y., Prawirodirdjo, L., Genrich, J.F., Stevens, C.W., McCaffrey, R., Sub-
arya, C., Puntodewo, S.S.O., and Calais, E., 2003, Crustal motion in Indo-
nesia from Global Positioning System measurements: Journal of Geo-
physical Research, v. 108, doi:10.1029/2001JB000324.
Busby, C.J., and Ingersoll, R.V., eds., 1995, Tectonics of Sedimentary Basins:
Cambridge, Massachusetts, Blackwell Science, 579 p.
Calvert, S.J., 2000, The Cenozoic geology of the Lariang and Karama regions,
Western Sulawesi, Indonesia [Ph.D. thesis]: University of London, 353 p.
Cardwell, R.K., and Isacks, B.L., 1978, Geometry of the subducted lithosphere
beneath the Banda Sea in eastern Indonesia from seismicity and fault
plane solutions: Journal of Geophysical Research, v. 83, p. 2825–2838.
Cardwell, R.K., Isacks, B.L., and Karig, D.E., 1980, The spatial distribution
of earthquakes, focal mechanism solutions and subducted lithosphere in
the Philippines and northeast Indonesian islands, in Hayes, D.E., ed., The
Tectonic and Geologic Evolution of Southeast Asian Seas and Islands:
American Geophysical Union, Geophysical Monograph 23, p. 1–35.
Cas, R.A.F., and Wright, J.V., 1987, Volcanic Successions: London, Allen and
Unwin, 528 p.
Charlton, T.R., 2000, Tertiary evolution of the Eastern Indonesia collision com-
plex: Journal of Asian Earth Sciences, v. 18, p. 603–631, doi: 10.1016/
S1367-9120(99)00049-8.
Chase, C.G., 1978, Extension behind island arcs and motions relative to hot-
spots: Journal of Geophysical Research, v. 83, p. 5385–5387.
Chesner, C.A., and Rose, W.I., 1991, Stratigraphy of the Toba tuffs and the
evolution of the Toba caldera complex, Sumatra, Indonesia: Bulletin of
Volcanology, v. 53, p. 343–356, doi: 10.1007/BF00280226.
Chesner, C.A., Rose, W.I., Deino, A., Drake, R., and Westgate, J.A., 1991, Erup-
tive history of earth’s largest Quaternary caldera (Toba, Indonesia) clarified:
Geology, v. 19, p. 200–203, doi: 10.1130/0091-7613(1991)019<0200:
EHOESL>2.3.CO;2.
Cloke, I.R., Milsom, J., and Blundell, D.J.B., 1999, Implications of gravity data
from East Kalimantan and the Makassar Straits: A solution to the origin
of the Makassar Straits?: Journal of Asian Earth Sciences, v. 17, p. 61–78,
doi: 10.1016/S0743-9547(98)00056-7.
Curray, J.R., 1994, Sediment volume and mass beneath the Bay of Bengal: Earth
and Planetary Science Letters, v. 125, p. 371–383, doi: 10.1016/0012-
821X(94)90227-5.
Dewey, J.F., 1980, Episodicity, sequence and style at convergent plate boundar-
ies: Geological Association of Canada Special Paper 20, p. 553–573.
Dewey, J.F., and Bird, J.M., 1970, Mountain belts and the new global tectonics:
Journal of Geophysical Research, v. 75, p. 2625–2647.
Dickinson, W.R., 1973, Widths of modern arc–trench gaps proportional to past
duration of igneous activity in associated magmatic arcs: Journal of Geo-
physical Research, v. 78, p. 3395–3417.
Dickinson, W.R., 1974, Plate tectonics and sedimentation, in Dickinson, W.R.,
ed., Tectonics and Sedimentation: Society of Economic Paleontologists
and Mineralogists Special Publication 22, p. 1–27.
Dickinson, W.R., 1977, Tectono-stratigraphic evolution of subduction con-
trolled sedimentary assemblages, in Talwani, M., and Pitman, W.C., eds.,
Island Arcs, Deep Sea Trenches, and Back-Arc Basins: Maurice Ewing
Series 1: Washington, American Geophysical Union 1, p. 33–40.
Dickinson, W.R., 1995, Forearc basins, in Busby, C.J., and Ingersoll, R.V., eds.,
Tectonics of Sedimentary Basins: Cambridge, Massachusetts, Blackwell
Science, p. 221–262.
Dickinson, W.R., and Seely, D.R., 1979, Structure and stratigraphy of forearc
regions: American Association of Petroleum Geologists Bulletin, v. 63,
p. 2–31.
Edwards, C.M.H., Menzies, M.A., Thirlwall, M.F., Morris, J.D., Leeman, W.P.,
and Harmon, R.S., 1994, The transition to potassic alkaline volcanism in
island arcs: The Ringgit-Besar Complex, east Java, Indonesia: Journal of
Petrology, v. 35, p. 1557–1595.
Elsassar, W.M., 1971, Sea-floor spreading as thermal convection: Journal of
Geophysical Research, v. 76, p. 1101–1112.
Engdahl, E.R., van der Hilst, R., and Buland, R., 1998, Global teleseismic
earthquake relocation with improved travel times and procedures for
depth determination: Bulletin of the Seismological Society of America,
v. 88, p. 722–743.
52 Hall and Smyth
England, P., Engdahl, R., and Thatcher, W., 2004, Systematic variation in the
depths of slabs beneath arc volcanoes: Geophysical Journal International,
v. 156, p. 377–408, doi: 10.1111/j.1365-246X.2003.02132.x.
Evans, C.A., Hawkins, J.W., and Moore, G.F., 1983, Petrology and geochem-
istry of ophiolitic and associated volcanic rocks on the Talaud Islands,
Molucca Sea Collision Zone, Northeast Indonesia, in Hilde, T.W.C., and
Uyeda, S., eds., Geodynamics of the Western Pacific–Indonesian Region:
American Geophysical Union and Geological Society of America, Geo-
dynamic Ser. 11, p. 159–172.
Fraser, T.H., Jackson, B.A., Barber, P.M., Baillie, P.W., and Myers, K., 2003,
The West Sulawesi Fold Belt and other new plays within the North Makas-
sar Straits—A prospectivity review: Proceedings of Indonesian Petroleum
Association 29th Annual Convention, Jakarta, p. 429–450.
Fuller, M., Ali, J.R., Moss, S.J., Frost, G.M., Richter, B., and Mahfi, A., 1999,
Paleomagnetism of Borneo: Journal of Asian Earth Sciences, v. 17, p. 3–
24, doi: 10.1016/S0743-9547(98)00057-9.
Funiciello, F., Faccenna, C., Giardini, D., and Regenauer-Lieb, K., 2003,
Dynamics of retreating slabs: 2. Insights from three-dimensional labora-
tory experiments: Journal of Geophysical Research, v. 108, p. 2207, doi:
10.1029/2001JB000896.
Furukawa, Y., 1993, Depth of the decoupling plate interface and thermal
structure under arcs: Journal of Geophysical Research, v. 98, p. 20,005–
20,013.
Garwin, S., Hall, R., and Watanabe, Y., 2005, Tectonic setting, geology and gold
and copper mineralization in Cenozoic magmatic arcs of Southeast Asia
and the West Pacific: Economic Geology and Bulletin of the Society of
Economic Geologists, v. 100, p. 891–930.
Gobbett, D.J., and Hutchison, C.S., eds., 1973, Geology of the Malay Pen-
insula (West Malaysia and Singapore): New York, Wiley-Interscience,
438 p.
Guntoro, A., 1999, The formation of the Makassar Strait and the separation
between SE Kalimantan and SW Sulawesi: Journal of Asian Earth Sci-
ences, v. 17, p. 79–98, doi: 10.1016/S0743-9547(98)00037-3.
Hall, R., 1996, Reconstructing Cenozoic SE Asia, in Hall, R., and Blundell,
D.J., eds., Tectonic Evolution of SE Asia: Geological Society [London]
Special Publication 106, p. 153–184.
Hall, R., 2000, Neogene history of collision in the Halmahera region, Indone-
sia, in Proceedings of the Indonesian Petroleum Association 27th Annual
Convention, Jakarta, p. 487–493.
Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and
the SW Pacific: Computer-based reconstructions, model and animations:
Journal of Asian Earth Sciences, v. 20, p. 353–434, doi: 10.1016/S1367-
9120(01)00069-4.
Hall, R., and Morley, C.K., 2004, Sundaland Basins, in Clift, P., et al., eds., Con-
tinent–Ocean Interactions within the East Asian Marginal Seas: American
Geophysical Union Geophysical Monograph 149, p. 55–85.
Hall, R., and Wilson, M.E.J., 2000, Neogene sutures in eastern Indonesia:
Journal of Asian Earth Sciences, v. 18, p. 781–808, doi: 10.1016/S1367-
9120(00)00040-7.
Hall, R., Audley-Charles, M.G., Banner, F.T., Hidayat, S., and Tobing, S.L.,
1988a, Basement rocks of the Halmahera region, eastern Indonesia: A
Late Cretaceous–Early Tertiary arc and fore-arc: Geological Society
[London] Journal, v. 145, p. 65–84.
Hall, R., Audley-Charles, M.G., Banner, F.T., Hidayat, S., and Tobing, S.L.,
1988b, Late Paleogene–Quaternary geology of Halmahera, eastern Indo-
nesia: Initiation of a volcanic island arc: Geological Society [London]
Journal, v. 145, p. 577–590.
Hall, R., Ali, J.R., Anderson, C.D., and Baker, S.J., 1995, Origin and motion
history of the Philippine Sea Plate: Tectonophysics, v. 251, p. 229–250,
doi: 10.1016/0040-1951(95)00038-0.
Hamilton, W., 1977, Subduction in the Indonesian region, in Talwani, M.,
and Pitman, W.C., eds., Island Arcs, Deep Sea Trenches, and Back-Arc
Basins: Maurice Ewing Series 1: Washington, American Geophysical
Union 1, p. 15–31.
Hamilton, W., 1979, Tectonics of the Indonesian region: U.S. Geological Sur-
vey Professional Paper 1078, 345 p.
Hamilton, W.B., 1988, Plate tectonics and island arcs: Geological Soci-
ety of America Bulletin, v. 100, p. 1503–1527, doi: 10.1130/0016-
7606(1988)100<1503:PTAIA>2.3.CO;2.
Hamilton, W.B., 1995, Subduction and magmatism, in Smellie, J.L., ed., Volca-
nism Associated with Subduction at Consuming Plate Margins: Geologi-
cal Society [London] Special Publication 81, p. 3–28.
Harington, C.R., ed., 1992, The Year without a Summer: World Climate in
1816: Ottawa, Canadian Museum of Nature, 576 p.
Harnois, L., 1988, The CIW index: A new chemical index of weathering: Sedi-
mentary Geology, v. 55, p. 319–322, doi: 10.1016/0037-0738(88)90137-6.
Harris, R.A., Sawyer, R.K., and Audley-Charles, M.G., 1998, Collisional
melange development: Geologic associations of active melange-forming
processes with exhumed melange facies in the western Banda orogen,
Indonesia: Tectonics, v. 17, p. 458–479, doi: 10.1029/97TC03083.
Hazebroek, H.P., and Tan, D.N.K., 1993, Tertiary tectonic evolution of the NW
Sabah continental margin: Geological Society of Malaysia Bulletin, v. 33,
p. 195–210.
Honthaas, C., Maury, R.C., Priadi, B., Bellon, H., and Cotten, J., 1999, The
Plio–Quaternary Ambon arc, eastern Indonesia: Tectonophysics, v. 301,
p. 261–281, doi: 10.1016/S0040-1951(98)00227-3.
Hutchison, C.S., 1989, Geological Evolution of South-East Asia, Oxford
Monographs on Geology and Geophysics: Oxford, UK, Clarendon Press,
376 p.
Hutchison, C.S., 1996, The ‘Rajang Accretionary Prism’ and ‘Lupar Line’
problem of Borneo, in Hall, R., and Blundell, D.J., eds., Tectonic Evolu-
tion of SE Asia: Geological Society [London] Special Publication 106,
p. 247–261.
Hutchison, C.S., Bergman, S.C., Swauger, D.A., and Graves, J.E., 2000, A
Miocene collisional belt in north Borneo: Uplift mechanism and isostatic
adjustment quantified by thermochronology: Geological Society [Lon-
don] Journal, v. 157, p. 783–793.
Hyndman, R.D., Currie, C.A., and Mazzotti, S., 2005, Subduction zone back-
arcs, mobile belts, and orogenic heat: GSA Today, v. 15, no. 2, p. 4–9.
Iwamori, H., 1997, Heat sources and melting in subduction zones: Journal of Geo-
physical Research, v. 102, p. 14,803–14,820, doi: 10.1029/97JB01036.
Karig, D.E., 1971, Origin and development of marginal basins in the Western
Pacific: Journal of Geophysical Research, v. 76, p. 2542–2561.
Karig, D.E., and Ingle, J.C., Jr., et al., eds., 1975, Initial Reports of the Deep Sea
Drilling Project, Leg 31: Washington, U.S. Government Printing Office,
927 p., doi: 10.2973/dsdp.proc.31.1975.
Karig, D.E., and Sharman, G.F., III, 1975, Subduction and accretion in
trenches: Geological Society of America Bulletin, v. 86, p. 377–389, doi:
10.1130/0016-7606(1975)86<377:SAAIT>2.0.CO;2.
Katili, J.A., 1975, Volcanism and plate tectonics in the Indonesian island arcs:
Tectonophysics, v. 26, p. 165–188, doi: 10.1016/0040-1951(75)90088-8.
Kay, S.M., Godoy, E., and Kurtz, A., 2005, Episodic arc migration, crustal
thickening, subduction erosion, and magmatism in the south-central
Andes: Geological Society of America Bulletin, v. 117, p. 67–88, doi:
10.1130/B25431.1.
Kay, S.M., Maksaev, V., Moscoso, R., Mpodozis, C., and Nasi, C., 1987, Prob-
ing the evolving Andean lithosphere: Mid–late Tertiary magmatism in
Chile (29°–30°30′S) over the modern zone of subhorizontal subduction:
Journal of Geophysical Research, v. 92, p. 6173–6189.
Kenyon, C.S., and Beddoes, L.R., eds., 1977, Geothermal Gradient Map of
Southeast Asia: Jakarta, Indonesia, South East Asia Petroleum Explora-
tion Society and Indonesian Petroleum Association, 50 p.
Klein, G. deV., and Kobayashi, K., et al., editors, 1980, Initial Reports of the
Deep Sea Drilling Project, Leg 58: Washington, U.S. Government Print-
ing Office, 1022 p., doi: 10.2973/dsdp.proc.58.1980.
Kopp, H., Klaeschen, D., Flueh, E.R., Bialas, J., and Reichert, C., 2002, Crustal
structure of the Java margin from seismic wide-angle and multichannel
reflection data: Journal of Geophysical Research, v. 107, doi:10.1029/
2000JB000095.
Kopp, H., Flueh, E.R., Peterson, C.J., Weinrebe, W., Wittwer, A., and Meramax
Scientists, 2006, The Java margin revisited: Evidence for subduction ero-
sion off Java: Earth and Planetary Science Letters, v. 242, p. 130–142, doi:
10.1016/j.epsl.2005.11.036.
Kreemer, C., Holt, W.E., Goes, S., and Govers, R., 2000, Active deforma-
tion in eastern Indonesia and the Philippines from GPS and seismic-
ity data: Journal of Geophysical Research, v. 105, p. 663–680, doi:
10.1029/1999JB900356.
Lavigne, F., Thouret, J.C., Voight, B., Suwa, H., and Sumaryono, A., 2000,
Lahars at Merapi volcano, Central Java: An overview: Journal of Vol-
canology and Geothermal Research, v. 100, p. 423–456, doi: 10.1016/
S0377-0273(00)00150-5.
Lebedev, S., and Nolet, G., 2003, Upper mantle beneath Southeast Asia
from S velocity tomography: Journal of Geophysical Research, v. 108,
doi:10.1029/2000JB000073.
 Cenozoic arc processes in Indonesia
Liew, T.C., and McCulloch, M.T., 1985, Genesis of granitoid batholiths of Pen-
insular Malaysia and implications for models of crustal evolution: Evi-
dence from a Nd–Sr isotopic zircon study: Geochimica et Cosmochimica
Acta, v. 49, p. 587–600, doi: 10.1016/0016-7037(85)90050-X.
Liew, T.C., and Page, R.W., 1985, U–Pb zircon dating of granitoid plutons from
the West Coast of Peninsular Malaysia: Geological Society [London]
Journal, v. 142, p. 515–526.
Macpherson, C.G., 1994, New analytical approaches to the oxygen and carbon
stable isotope geochemistry of some subduction-related lavas [Ph.D. the-
sis]: University of London, 206 p.
Macpherson, C.G., and Hall, R., 1999, Tectonic controls of geochemical evolu-
tion in arc magmatism of SE Asia, in Proceedings of 4th PACRIM Con-
gress: Bali, Indonesia, Australian Institute of Mining and Metallurgy,
p. 359–368.
Macpherson, C.G., and Hall, R., 2002, Timing and tectonic controls in the
evolving orogen of SE Asia and the western Pacific and some implications
for ore generation, in Blundell, D.J., et al., eds., The Timing and Location
of Major Ore Deposits in an Evolving Orogen: Geological Society [Lon-
don] Special Publication 204, p. 49–67.
Macpherson, C.G., Forde, E.J., Hall, R., and Thirlwall, M.F., 2003, Geochemi-
cal evolution of magmatism in an arc–arc collision: The Halmahera and
Sangihe arcs, eastern Indonesia, in Larter, R.D., and Leat, P.T., eds., Intra-
Oceanic Subduction Systems: Tectonic and Magmatic Processes: Geo-
logical Society [London] Special Publication 219, p. 207–220.
Mason, B.G., Pyle, D.M., and Oppenheimer, C., 2004, The size and frequency
of the largest explosive eruptions on Earth: Bulletin of Volcanology, v. 66,
p. 735–748, doi: 10.1007/s00445-004-0355-9.
McCaffrey, R., 1983, Seismic wave propagation beneath the Molucca Sea
arc–arc collision zone, Indonesia: Tectonophysics, v. 96, p. 45–57, doi:
10.1016/0040-1951(83)90243-3.
McCaffrey, R., 1991, Earthquakes and ophiolite emplacement in the Molucca
Sea collision zone, Indonesia: Tectonics, v. 10, p. 433–454.
McCaffrey, R., Silver, E.A., and Raitt, R.W., 1980, Crustal structure of the
Molucca Sea collision zone, Indonesia, in Hayes, D.E., ed., The Tectonic
and Geologic Evolution of Southeast Asian Seas and Islands: American
Geophysical Union, Geophysical Monograph Ser. 23, p. 161–178.
McCourt, W.J., Crow, M.J., Cobbing, E.J., and Amin, T.C., 1996, Mesozoic and
Cenozoic plutonic evolution of SE Asia: Evidence from Sumatra, Indone-
sia, in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE Asia:
Geological Society [London] Special Publication 106, p. 321–335.
Metcalfe, I., 1996, Pre-Cretaceous evolution of SE Asian Terranes, in Hall, R.,
and Blundell, D.J., eds., Tectonic Evolution of SE Asia: Geological Soci-
ety [London] Special Publication 106, p. 97–122.
Mikada, H., Moore, G.F., Taira, A., Becker, K., Moore, J.C., and Klaus, A.,
eds., 2005, Proceedings of Ocean Drilling Program, Scientific Results,
190/196 [Online]: http://www-odp.tamu.edu/publications/190196SR/
190196sr.htm (November 2006).
Miyazaki, K., Sopaheluwakan, J., Zulkarnain, I., and Wakita, K., 1998, A jadeite-
quartz-glaucophane rock from Karangsambung, central Java, Indonesia:
The Island Arc, v. 7, p. 223–230, doi: 10.1046/j.1440-1738.1998.00164.x.
Moore, G.F., and Karig, D.E., 1980, Structural geology of Nias island, Indo-
nesia: Implications for subduction zone tectonics: American Journal of
Science, v. 280, p. 193–223.
Moore, G.F., Kadarisman, D.K., Evans, C.A., and Hawkins, J.W., 1981, Geol-
ogy of the Talaud Islands, Molucca Sea Collision Zone, northeast Indone-
sia: Journal of Structural Geology, v. 3, p. 467–475, doi: 10.1016/0191-
8141(81)90046-8.
Morton, A.C., 1984, Stability of detrital heavy minerals in Tertiary sandstones
from the North Sea basin: Clay Minerals, v. 19, p. 287–308, doi: 10.1180/
claymin.1984.019.3.04.
Nagao, T., and Uyeda, S., 1995, Heat flow distribution in southeast Asia with
consideration of volcanic heat: Tectonophysics, v. 251, p. 153–159, doi:
10.1016/0040-1951(95)00084-4.
Nesbitt, H.W., and Young, G.M., 1984, Prediction of some weathering trends
of plutonic and volcanic rocks based on thermodynamic and kinetic con-
siderations: Geochimica et Cosmochimica Acta, v. 48, p. 1523–1534, doi:
10.1016/0016-7037(84)90408-3.
Nichols, G.J., and Hall, R., 1991, Basin formation and Neogene sedimentation
in a backarc setting, Halmahera, eastern Indonesia: Marine and Petroleum
Geology, v. 8, p. 50–61, doi: 10.1016/0264-8172(91)90044-2.
Nur’Aini., S., Hall, R., and Elders, C.F., 2005, Basement architecture and
sedimentary fills of the North Makassar Straits basin: Proceedings of
53
Indonesian Petroleum Association 30th Annual Convention, Jakarta,
p. 483–497.
Oncken, O., Chong, G., Franz, G., Giese, P., Götze, H.-J., Ramos, V.A., Strecker,
M.R., and Wigger, P., eds., 2006, The Andes—Active Subduction Orog-
eny: Berlin, Springer, 570 p.
Oxburgh, E.R., and Turcotte, D.L., 1970, Thermal structure of island arcs:
Geological Society of America Bulletin, v. 81, p. 1665–1688, doi:
10.1130/0016-7606(1970)81[1665:TSOIA]2.0.CO;2.
Parkinson, C.D., Miyazaki, K., Wakita, K., Barber, A.J., and Carswell, D.A.,
1998, An overview and tectonic synthesis of the pre-Tertiary very-high-
pressure metamorphic and associated rocks of Java, Sulawesi and Kali-
mantan, Indonesia: The Island Arc, v. 7, p. 184–200, doi: 10.1046/j.1440-
1738.1998.00184.x.
Pearce, J.A., and Peate, D.W., 1995, Tectonic implications of the composition
of volcanic arc magmas: Annual Review of Earth and Planetary Sciences,
v. 23, p. 251–285, doi: 10.1146/annurev.ea.23.050195.001343.
Pollack, H.N., Hurter, S., and Johnson, J.R., 1990, The new global heat flow
data compilation: Eos (Transactions, American Geophysical Union),
v. 71, p. 1604.
Pollack, H.N., Hurter, S.J., and Johnson, J.R., 1993, Heat flow from the Earth’s
interior: Analysis of the global data set: Reviews of Geophysics, v. 31,
p. 267–280, doi: 10.1029/93RG01249.
Polvé, M., Maury, R.C., Bellon, H., Rangin, C., Priadi, B., Yuwono, S., Joron,
J.L., and Soeria-Atmadja, R., 1997, Magmatic evolution of Sulawesi
(Indonesia): Constraints on the Cenozoic geodynamic history of the Sun-
daland active margin: Tectonophysics, v. 272, p. 69–92, doi: 10.1016/
S0040-1951(96)00276-4.
Puspita, S.D., Hall, R., and Elders, C.F., 2005, Structural styles of the offshore
West Sulawesi fold belt, North Makassar Straits, Indonesia: Proceedings
of Indonesian Petroleum Association 30th Annual Convention, Jakarta,
p. 519–542.
Rampino, M.R., and Self, S., 1992, Volcanic winter and accelerated glacia-
tion following the Toba super-eruption: Nature, v. 359, p. 50–52, doi:
10.1038/359050a0.
Rampino, M.R., and Self, S., 1993a, Bottleneck in human evolution and the
Toba eruption: Science, v. 262, p. 1955, doi: 10.1126/science.8266085.
Rampino, M.R., and Self, S., 1993b, Climate–volcanism feedback and the Toba
eruption of ca. 74,000 years ago: Quaternary Research, v. 40, p. 269–280,
doi: 10.1006/qres.1993.1081.
Ranero, C.R., and von Huene, R., 2000, Subduction erosion along the Mid-
dle America convergent margin: Nature, v. 404, p. 748–752, doi:
10.1038/35008046.
Rangin, C., Spakman, W., Pubellier, M., and Bijwaard, H., 1999, Tomographic
and geological constraints on subduction along the eastern Sundaland
continental margin (South-East Asia): Bulletin de la Société Géologique
de France, v. 170, p. 775–788.
Ritsema, J., and van Heijst, H.-J., 2000, Seismic imaging of structural heteroge-
neity in the Earth’s mantle: Evidence for large-scale mantle flow: Science
Progress, v. 83, p. 243–259.
Rowland, A., and Davies, J.H., 1999, Buoyancy rather than rheology con-
trols the thickness of the overriding mechanical lithosphere at subduc-
tion zones: Geophysical Research Letters, v. 26, p. 3037–3040, doi:
10.1029/1999GL005347.
Rutherford, K.J., and Qureshi, M.K., eds., 1981, Geothermal gradient map
of Southeast Asia (2nd edition): Jakarta, Indonesia, South East Asia
Petroleum Exploration Society and Indonesian Petroleum Association,
51 p.
Samuel, M.A., 1994, The structural and stratigraphic evolution of islands at the
active margin of the Sumatran Forearc, Indonesia [Ph.D. thesis]: Univer-
sity of London, 345 p.
Samuel, M.A., and Harbury, N.A., 1996, The Mentawai fault zone and defor-
mation of the Sumatran forearc in the Nias area, in Hall, R., and Blundell,
D.J., eds., Tectonic Evolution of SE Asia: Geological Society [London]
Special Publication 106, p. 337–351.
Samuel, M.A., Harbury, N.A., Jones, M.E., and Matthews, S.J., 1995, Inver-
sion of an outer-arc ridge: The Sumatran Forearc, Indonesia, in Buchanan,
J.G., and Buchanan, P.G., eds., Basin Inversion: Geological Society [Lon-
don] Special Publication 88, p. 473–492.
Samuel, M.A., Harbury, N.A., Bakri, A., Banner, F.T., and Hartono, L., 1997,
A new stratigraphy for the islands of the Sumatran Forearc, Indonesia:
Journal of Asian Earth Sciences, v. 15, p. 339–380, doi: 10.1016/S0743-
9547(97)87720-3.
54 Hall and Smyth
Sandwell, D.T., and Smith, W.H.F., 1997, Marine gravity anomaly from Geosat
and ERS 1 satellite altimetry: Journal of Geophysical Research, v. 102,
p. 10,039–10,054, doi: 10.1029/96JB03223.
Schlüter, H.U., Gaedicke, C., Roeser, H.A., Schreckenberger, B., Meyer, H.,
Reichert, C., Djajadihardja, Y., and Prexl, A., 2002, Tectonic features of
the southern Sumatra–western Java forearc of Indonesia: Tectonics, v. 21,
p. 1047, doi: 10.1029/2001TC901048, doi: 10.1029/2001TC901048.
Scholl, D., von Huene, R., Vallier, T.L., and Howell, D.G., 1980, Sedimentary
masses and concepts about tectonic processes at underthrust ocean mar-
gins: Geology, v. 8, p. 564–568, doi: 10.1130/0091-7613(1980)8<564:
SMACAT>2.0.CO;2.
Scholl, D., Vallier, T.L., and Stevenson, A.J., 1986, Terrane accretion, produc-
tion and continental growth: A perspective based on the origin and tec-
tonic fate of the Aleutian–Bering Sea region: Geology, v. 14, p. 43–47,
doi: 10.1130/0091-7613(1986)14<43:TAPACG>2.0.CO;2.
Shemenda, A.I., 1994, Subduction: Insights from Physical Modelling: Norwell,
Massachusetts, Kluwer Academic Publishers, 215 p.
Silver, E.A., and Moore, J.C., 1978, The Molucca Sea collision zone, Indone-
sia: Journal of Geophysical Research, v. 83, p. 1681–1691.
Situmorang, B., 1982, The formation and evolution of the Makassar basin,
Indonesia [Ph.D. thesis]: University of London, 313 p.
Smithsonian, 2006, Global volcanism program: http://www.volcano.si.edu/
(March 2006).
Smyth, H.R., 2005, Eocene to Miocene basin history and volcanic history in
East Java, Indonesia [Ph.D. thesis]: University of London, p. 476.
Smyth, H.R., Hall, R., and Nichols, G.J., 2008, this volume, Cenozoic volcanic
arc history of East Java, Indonesia: The stratigraphic record of eruptions
on an active continental margin, in Draut, A.E., et al., eds., Lessons from
the Stratigraphic Record in Arc Collision Zones: Geological Society of
America Special Publication 436, doi: 10.1130/2008.2436(10).
Stow, D.A.V., Cochran, J.R., and ODP Leg 116 Shipboard Scientific Party,
1989, The Bengal Fan: Some preliminary results from ODP drilling: Geo-
Marine Letters, v. 9, p. 1–10, doi: 10.1007/BF02262812.
Suttner, L.J., and Dutta, P.K., 1986, Alluvial sandstone composition and paleo-
climate; I. Framework mineralogy: Journal of Sedimentary Research,
v. 56, p. 329–345.
Taira, A., Hill, I., Firth, J., et al., 1991, Proceedings of Ocean Drilling Program,
Initial Reports, v. 131: College Station, Texas, Ocean Drilling Program
[Online]: http://www-odp.tamu.edu/publications/131_IR/VOLUME/
CHAPTERS/ir131_06.pdf (December 2006).
Tang, J.-C., and Chemenda, A.I., 2000, Numerical modelling of arc–continent
collision: Application to Taiwan: Tectonophysics, v. 325, p. 23–42, doi:
10.1016/S0040-1951(00)00129-3.
Tang, J.C., Chemenda, A.I., Chery, J., Lallemand, S., and Hassani, R., 2002,
Compressional subduction regime and initial arc–continent collision:
Numerical modeling, in Byrne, T., and Liu, C.-S., eds., Geology and Geo-
physics of an Arc–Continent Collision, Taiwan: Geological Society of
America Special Paper 358, p. 177–186.
Tjia, H.D., 1996, Sea-level changes in the tectonically stable Malay–Thai pen-
insula: Quaternary International, v. 31, p. 95–101, doi: 10.1016/1040-
6182(95)00025-E.
Tongkul, F., 1991, Tectonic evolution of Sabah, Malaysia: Journal of South-
east Asian Earth Sciences, v. 6, p. 395–406, doi: 10.1016/0743-
9547(91)90084-B.
Umbgrove, J.H.F., 1938, Geological history of the East Indies: American Asso-
ciation of Petroleum Geologists Bulletin, v. 22, p. 1–70.
van Bemmelen, R.W., 1949, The Geology of Indonesia: Nijhoff, The Hague,
Government Printing Office, 732 p.
van Hattum, M.W.A., 2005, Provenance of Cenozoic sedimentary rocks of
northern Borneo [Ph.D. thesis]: University of London, 457 p.
van Hattum, M.W.A., Hall, R., and Nichols, G.J., 2003, Provenance of north-
ern Borneo sediments: Proceedings of Indonesian Petroleum Association
29th Annual Convention, Jakarta, p. 305–319.
van Hattum, M.W.A., Hall, R., Pickard, A.L., and Nichols, G.J., 2006, SE Asian
sediments not from Asia: Provenance and geochronology of North Bor-
neo sandstones: Geology, v. 34, p. 589–592, doi: 10.1130/G21939.1.
Vannucchi, P., Galeotti, S., Clift, P.D., Ranero, C.R., and von Huene, R., 2004,
Long-term subduction-erosion along the Guatemalan margin of the Mid-
dle America trench: Geology, v. 32, p. 617–620, doi: 10.1130/G20422.1.
von Huene, R., and Scholl, D., 1991, Observations at convergent margins con-
cerning sediment subduction, subduction erosion, and even the growth of
the continental crust: Reviews of Geophysics, v. 29, p. 279–316.
von Huene, R., Ranero, C.S., and Vannucchi, P., 2004, Generic model of sub-
duction erosion: Geology, v. 32, p. 913–916, doi: 10.1130/G20563.1.
Wakita, K., 2000, Cretaceous accretionary–collision complexes in central Indo-
nesia: Journal of Asian Earth Sciences, v. 18, p. 739–749, doi: 10.1016/
S1367-9120(00)00020-1.
Walker, G.P.L., 1972, Crystal concentrations in ignimbrites: Contributions to
Mineralogy and Petrology, v. 36, p. 135–146, doi: 10.1007/BF00371184.
Wallace, A.R., 1869, The Malay Archipelago: Hong Kong, Periplus, 515 p.
Waltham, D., Hall, R., Smyth, H.R., and Ebinger, C.J., 2008, this volume, Basin
formation by volcanic arc loading, in Draut, A.E., et al., eds., Lessons
from the Stratigraphic Record in Arc Collision Zones: Geological Society
of America Special Publication 436, doi: 10.1130/2008.2436(02).
Widiyantoro, S., and van der Hilst, R.V., 1997, Mantle structure beneath Indo-
nesia inferred from high-resolution tomographic imaging: Geophysical
Journal International, v. 130, p. 167–182, doi: 10.1111/j.1365-246X.1997.
tb00996.x.
William, A.G., Lambiase, J.J., Back, S., and Jamiran, M.K., 2003, Sedimentol-
ogy of the Jalan Selaiman and Bukit Melinsung outcrops, western Sabah:
Is the West Crocker Formation an analogue for Neogene turbidites off-
shore?: Geological Society of Malaysia Bulletin, v. 47, p. 63–75.
Zeilinga de Boer, J.Z., and Sanders, D.T., 2002, Volcanoes in Human History:
Princeton, New Jersey, Princeton University Press, 295 p.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Carbonate-platform facies in volcanic-arc settings: Characteristics and controls on

deposition and stratigraphic development

Steven L. Dorobek*
Department of Geology & Geophysics, Texas A&M University, College Station, Texas 77843, USA

ABSTRACT

Shallow-marine carbonate facies from volcanic-arc settings provide an important,

but commonly overlooked, record of relative sea-level change, differential subsidence-
uplift, paleoclimate trends, and other environmental changes. Carbonate strata are
thin where volcanic eruptions are frequent and voluminous, unless shallow, bathy-
metric highs persist for long periods of time and volcaniclastic sediment and erupted
materials are trapped in adjacent depocenters. Carbonate platforms and reefs can
attain significant thickness, however, if subsidence continues after volcanic activity
ceases or the volcanic front migrates. The areal extent of shallow-marine carbonate
sedimentation is likewise affected by differential tectonic subsidence, although car-
bonate platforms are most laterally extensive during transgressive to highstand condi-
tions and when arc depocenters are filled with sediment.
Tectonic controls on shallow-marine carbonate sedimentation in arc depocenters
include (1) coseismic fault displacements and associated surface deformation; (2) long-
wavelength tectonic subsidence related to dynamic mantle flow, flexure, lithospheric
thinning, and thermal subsidence; and (3) large-scale plate deformation related to
local conditions of subduction.
Depositional controls on carbonate sedimentation in arc depocenters include
(1) the frequency, volume, and style of volcanic eruptions; (2) accumulation rates for
siliciclastic-volcaniclastic sediment; (3) the frequency, volume, and dispersal paths of
erupted material; (4) (paleo)wind direction, which influences both carbonate facies
development directly and indirectly by controlling the dispersal of volcanic ash and
other pyroclastic sediment, which can bury carbonate-producing organisms; (5) the
frequency and intensity of tsunami events; and (6) volcanically or seismically trig-
gered mass-wasting events, which can erode or bury carbonate strata.
Regarding platform morphologies in arc-related settings, (1) fringing reefs or
barrier reef systems with lagoons may develop around volcanic edifices throughout
the long-term evolution of volcanic arcs; (2) local reefs and mounds may build on
intrabasinal, fault-bounded highs within underfilled forearc, intra-arc, and backarc
basins; (3) isolated platforms with variable platform margin-to-basin transitions are

*Current address: Maersk Oil & Gas, Esplanaden 50, DK 1263 Copenhagen, Denmark.

Dorobek, S.L., 2008, Carbonate-platform facies in volcanic-arc settings: Characteristics and controls on deposition and stratigraphic development, in Draut,
A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special
Paper 436, p. 55–90, doi: 10.1130/2008.2436(04). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All
rights reserved.

55
56 Dorobek
common in “underfilled” and tectonically active depocenters; and (4) broad ramps
and rimmed carbonate shelves are typically found in tectonically mature and sedi-
ment-filled depocenters.
Keywords: carbonate platforms, reefs, carbonate facies, subsidence, basins, depocen-
ters, carbonate stratigraphy.
INTRODUCTION
Shallow-marine carbonate facies may constitute an impor-
tant part of the stratigraphy that forms in volcanic-arc settings,
although they have received relatively minor attention from geo-
scientists (cf. Dickinson, 1995, 2001; Smith and Landis, 1995;
Soja, 1996; Wilson and Bosence, 1997; Nunn, 1998a; Wilson,
2002; Wilson and Lokier, 2002; Wilson and Vecsei, 2005; Bos-
ence, 2005). Shallow-marine carbonate facies in volcanic-arc
settings are important recorders of sea-level changes, differen-
tial subsidence across arc depocenters, paleoclimate, and other
environmental conditions. Thus, careful analysis of these strata
is critical for reconstructing the complicated geological histories
of volcanic-arc depocenters. Significant petroleum accumula-
tions are also stored in carbonate reservoirs in some volcanic-arc
depocenters (e.g., Indonesia backarc basin; Sharaf et al., 2005),
although these rather limited examples may be associated with
special petroleum-system conditions.
This paper presents a review of the tectonic and depositional
controls that influence shallow-marine carbonate systems associ-
ated with volcanic arcs along convergent margins. Examples of
carbonate systems from modern and ancient settings are used to
illustrate key principles, although it is important to note that rela-
tively few, thoroughly documented examples exist from which
to draw conclusions or construct predictive models. Detailed
stratigraphic information for carbonate successions that formed
in a variety of arc depocenters during different times in Earth
history is simply not available. Instead, the main objective of this
paper is to describe how tectonic deformation, other depositional
controls, and siliciclastic and volcaniclastic sedimentation can
influence shallow-marine carbonate facies development in arc
settings. More complete documentation of carbonate successions
from modern arc settings and from the rock record will undoubt-
edly lead to refinement of the basic models presented here, but I
hope this paper serves as a general guide to the analysis of arc-
related carbonate systems.
TECTONIC CONTROLS ON ARC-RELATED
CARBONATE SYSTEMS
Tectonic deformation along convergent plate margins
strongly influences shallow-marine carbonate depositional sys-
tems in volcanic-arc depocenters. Differential tectonic subsid-
ence or uplift in any tectonic setting can be attributed to (1) local,
fault-controlled deformation, which can involve fault systems
that only cut upper crustal levels to truly lithosphere-scale fault
systems; and (2) long-wavelength, basin-scale differential subsid-
ence and uplift, which typically involve lithosphere-scale defor-
mation processes (e.g., flexure, lithospheric thinning, or thermal
subsidence) or dynamic mantle flow (i.e., dynamic topography-
subsidence; cf. Gurnis, 1990a, b, 1991). Convergent margins with
active subduction, however, are unique tectonic settings where
coseismic displacements along great subduction faults may cause
long-wavelength (>100 km) surface deformations that can sig-
nificantly affect depositional systems on the upper (or overriding)
plate. This style of differential tectonic subsidence and uplift is
exclusive to active subduction zones and was dramatically illus-
trated during the December 2004 Sumatran earthquake (Fig. 1).
These various scales of tectonic deformation typically inter-
act simultaneously and in highly complex ways in volcanic-arc
settings, making the subsidence and uplift histories of arc dep-
ocenters difficult to characterize, predict, and understand. Other
tectonic characteristics of volcanic-arc settings can also affect
carbonate sedimentation, such as large-scale aspects of subduc-
tion, plate motion, and plan-view geometries of plate margins.
The following section first introduces large-scale tectonic con-
trols on arc depositional systems and progresses to consider
more local tectonic deformation and its effects on deposition in
arc depocenters.
Intra-Oceanic Island Arcs versus Continental-Margin
Arc Systems
Whether a volcanic arc forms as an intraoceanic or a
continental-margin arc system is of first-order importance for
shallow-marine carbonate systems. Intraoceanic island arcs
form on the leading edge of an overriding oceanic plate. In
contrast, continental-margin arc systems form where a slab
of oceanic lithosphere is subducted beneath an upper plate of
continental lithosphere. Arc-related basins along both types of
plate margins may be similar, although the initial topography
of the overriding plate and rheological differences between
oceanic and continental lithosphere of the upper plate impose
different conditions on basin development.
Volcanism associated with intraoceanic arc systems ini-
tially occurs in submarine environments. Successive subma-
rine eruptive events cause progressive shallowing of volcanic
edifices until they eventually build to within the photic zone,
and shallow-marine carbonate deposition can begin, which
may take place after millions of years of eruptive activity. In
 Carbonate-platform facies in volcanic-arc settings 57

14°N Extensional, Contractional, Strike-Slip, or Neutral

Arc Settings
Andaman Islands
Convergent margins can be classified according to the
12°N regional styles of deformation that affect the upper plate (cf.
Dewey, 1980). Various sectors of the upper plate can be charac-
terized by extensional, contractional, strike-slip, or very minor
(i.e., neutral) deformation. Extensional deformation can develop
10°N within (1) the thermally weakened basement of the arc massif,
(2) the backarc region, or (3) the forearc region, especially where
there is significant tectonic erosion during subduction (Clift et al.,
1998). Most deforming volcanic arcs have at least local strike-
8°N Nicobar Islands
slip fault zones, even where contractional or extensional styles
of deformation are dominant. In compressional arcs, thrust and
reverse faults typically form in the backarc region on the upper
plate; these faults are typically antithetic to the dip of the sub-
6°N ducting slab. Compressional retroarc foreland basin settings are
not considered here, although carbonate facies in extensional
TR

backarc basins are discussed in a later section.

EN

It also is important to note that some dip-trending sections

S
um
CH

4°N 24 Dec 2004 Sumatra across subduction margins show compressional deformation in
at
ra
epicenter Sumatra one segment of an arc system but other styles of deformation
Simeulue in other parts of the system. For example, the Sumatran forearc
region is generally considered to be an “accretionary” forearc
2°N Uplifted shoreface (Clift and Vannucchi, 2004), yet the arc itself is cut by the Suma-
28 March 2005
epicenter tran Fault, which is a major, through-going, strike-slip fault sys-
Subsided shoreface tem that extends along the entire length of Sumatra. Although
0 200 km Nias the Sumatran forearc region is generally thought to be accreting,
0° active zones of transtensional faulting cut at high angles across
90°E 92°E 94°E 96°E 98°E the forearc basin and partition the Sumatran forearc into multiple
depocenters. These closely juxtaposed styles of tectonic defor-
Figure 1. Distribution of coseismic uplift and subsidence across the
Andaman Islands and Western Indonesian forearc after the Decem- mation are described here simply to demonstrate that it can be
ber 2004 and March 2005 major earthquakes. Displacement along the inappropriate to make general characterizations about the over-
subduction fault between the subducting Indian Ocean plate and the all tectonic style of a particular arc system without examining
overriding Indonesian forearc propagated rapidly northward from the long-term deformation and subsidence histories across the entire
epicenters (indicated by stars) toward the Andaman Islands. Areas of system, from the trench to the backarc region, and understanding
coastal uplift are indicated by the stippled pattern. Areas of coastal
subsidence are indicated by dark shading. Dashed line represents inter- the kinematic history of the deformation. Accurate, large-scale
preted subsidence hinge line. Data obtained from SAR interferometry kinematic analyses may be impossible in strongly tectonized,
(InSAR) analyses. Modified from Japan Geographical Survey Institute ancient arc settings.
Web site.
Margin Geometries, Relative Plate Motion, and Changes
in Subduction
contrast, continental-margin arc systems may actually begin as
subaerial settings and require either a significant eustatic rise or Plate-motion vectors and the plan-view geometry of conti-
tectonic subsidence to submerge them to photic depths, when nental margins (i.e., before and during convergence and collision)
shallow-marine carbonate deposition can begin. also influence subduction or arc collision along specific margin
Although both oceanic and continental lithosphere will segments (Dewey, 1980; Thomas, 1983; Bradley, 1989). The
bend in response to the growth of surface loads like volcanic arc initiation of subduction or collision usually will be diachronous
systems, they have different flexural rigidities so their flexural along strike where the motion of one plate is highly oblique to
response will differ. Continental lithosphere also may have more another plate margin. If either plate is rotating about a vertical
inherent strength variations than oceanic lithosphere because of pole, then the rate of subduction or shortening caused by collision
its typically long and complicated geologic history. These varia- typically increases with distance from the pole of rotation. The
tions in strength and the presence of preexisting faults can influ- presence of salients or recesses along either plate margin will also
ence basin development in continental-margin arc systems. influence which segments along either margin are first to collide.
58 Dorobek
The angle of convergence or subduction also strongly influ-
ences the styles and distribution of deformation within the over-
riding plate. Where the relative motion of both plates is nearly
perpendicular to the trench axis, and trench rollback is slower
than the velocity at which the overriding plate approaches the
subduction zone, compressional styles of deformation should
dominate in the overriding plate (cf. Dewey, 1980). In contrast,
oblique subduction characteristically produces some type of
transtensional, transpressional, or pure strike-slip deformation in
the overriding plate. The style and geographic location of upper-
plate deformation (e.g., within the arc-trench gap, the volcanic
arc, or in the backarc region) and volcanic activity depend on
(1) the angle of convergence; (2) the dip of the subducted slab;
(3) whether any seamounts, active spreading-ridge segments,
fracture zones, or other rough surface features on the subducted
slab arrive at the subduction zone; (4) the degree of mechani-
cal coupling between the subducting and overriding plates; and
(5) the location and orientation of lateral strength variations or
preexisting fault zones in the overriding plate, which might be
reactivated during convergence. Other mechanical or rheological
attributes of plates or microplates involved along the convergent
margin may influence deformation in the overriding plate but
probably in less significant ways.
The timing of collision and styles of deformation associated
with convergent margins will change if plate-motion vectors vary
over time. Changes in the relative motion of plates also cause
intraplate stress trajectories to be reoriented over time, and con-
sequently the styles and patterns of deformation within plate inte-
riors should evolve, especially where preexisting basement fault
systems are reactivated by in-plane stress.
Magmatic activity in the overriding plate will also be affected
by changes in the nature of subduction (cf. Otsuki, 1989). Where
the subducted slab has a shallow dip, the location of the volcanic
front generally is farther inboard from the trench, or the margin
may lack substantial volcanic and magmatic features altogether
where “flat-slab” subduction occurs. The overriding plate may
also be deformed where the dip of the subducted plate becomes
very shallow (nearly horizontal) and there is mechanical coupling
between the subducted plate and the underside of the overrid-
ing plate (e.g., Neogene deformation of the Chilean-Argentinean
foreland; Jordan et al., 1983; Jordan and Allmendinger, 1986).
If the dip of the subducted slab steepens over time, trench roll-
back and extension may occur across much of the arc (Uyeda
and Kanamori, 1979; Otsuki, 1989). The plan-view geometry of
the overriding plate may also control curvature of the subduct-
ing plate, downdip from the subduction zone, which in turn may
influence deformation across backarc regions in the upper plate
(cf. Cahill and Isacks, 1992).
Where arc-continent or continent-continent collision has
occurred, subduction will usually become very limited or may
cease altogether. In these cases, the subduction zone may jump
farther outboard and also change dip (or polarity) so that contin-
ued plate convergence can be accommodated by continued sub-
duction of oceanic lithosphere. Where continental plates or arc
microplates meet along a convergent margin, some blocks might
be “extruded,” or pushed laterally away from the zone of conver-
gence (e.g., Tapponnier et al., 1982; Coward, 1990; Maynard et
al., 1997), although the amount of extrusion and lateral transla-
tion of blocks away from the collision zone varies. Significant
shortening may continue to occur along the convergence zone,
but lateral translation of fault-bounded blocks may be required
to accommodate additional convergence once the crust within
the collision zone cannot be tectonically thickened anymore.
Regional strike-slip-fault networks that bound the translated
blocks accommodate the extrusion. The sense of slip on these
faults may change over time as one block progressively plows
into the adjoining plate and more inboard blocks move laterally
to accommodate additional shortening in the collision zone.
Tectonic Drivers of Long-Wavelength Differential
Subsidence in Arc Settings
Volcanic-arc settings are characterized by rapidly changing
and complexly distributed subsidence patterns, which reflect the
interactions of multiple tectonic drivers that operate along sub-
duction margins. These drivers of differential long-wavelength
subsidence generally involve deformation or long-term rheo-
logical changes of the entire lithosphere, and include flexure,
thermal subsidence, magmatic underplating, and thinning in
extensional arcs (Fig. 2). Dynamic mantle flow can also influ-
ence long-wavelength subsidence patterns in the upper plate (i.e.,
dynamic topography; Gurnis, 1990a, b, 1991). Long-wavelength
subsidence patterns will influence the distribution, plan-view
dimensions, and ultimate thickness of carbonate platforms, with
the most widespread development of carbonate facies generally
occurring during waning stages of subduction, when tectonic
subsidence rates are progressively slowing and arc depocenters
are more likely to become filled with sediment.
Deformation in the forearc region, especially within the sub-
duction complex, can result in other types of long-wavelength
subsidence or uplift that are not related to lithospheric defor-
mation. The subduction complex can deform in multiple ways,
including complex internal thrusting, backthrusting, normal
faulting, and gravitational sliding. These different types of
deformation are thought to occur as the subduction complex
tends to maintain a wedge shape or “critical taper” (Davis et al.,
1983; Dahlen et al., 1984). Underplating and subduction ero-
sion also can cause long-wavelength surface deformation across
the top of the subduction complex (Fig. 3). Deformation of the
subduction complex can, in turn, cause differential subsidence
in the outer part of the forearc basin because the subduction
complex commonly forms the seaward limit of this depocenter.
This differential subsidence affects shallow-marine carbonate
deposition in the outer part of the forearc basin when the basin
is filled with sediment and the basin floor is within the photic
zone. Along convergent margins where the subduction com-
plex is absent or poorly developed (e.g., with sediment-starved
trenches or where subduction erosion dominates), depositional
 Carbonate-platform facies in volcanic-arc settings 59

DS DS
Fsc Fsc
Tva Tva Tbaoc Tbaoc

Ffbf Fbabf Ffbf

Fbabf Tra

(little or no thermal
subsidence of old
trapped oceanic crust) 0 100 km
A 0 100 km
B

DS DS
Fsc Fsc
Tva Tva

Ffbf Frfb Ffbf Tba

0 200 km
C 0 200 km D

Basin Fill Dynamic Subsidence

Continental or Arc Crust Flexural Subsidence

Oceanic Crust Thermal Subsidence

Figure 2. Causes of long-wavelength (>100 km) tectonic subsidence across the overriding plate for different types of convergent margins. Flex-
ural subsidence patterns (indicated by dashed black lines): Fsc—flexural subsidence of subducted plate caused by load of the subduction complex;
Ffbf—flexural subsidence of forearc region caused by sediment load within the forearc basin; Fbabf—flexural subsidence of backarc region caused
by sediment load within the backarc basin; Frfb—flexural subsidence of retroarc foreland basin region caused by sediment load and bordering
thrust belt. Thermal subsidence patterns (indicated by dotted lines): Tva—thermal subsidence caused by cooling magmatic arc, after arc be-
comes extinct; Tbaoc—thermal subsidence of oceanic crust created along backarc spreading center; Tra—thermal subsidence caused by cooling
of remnant magmatic arc in extensional backarc settings; Tba—thermal subsidence caused by cooling of rifted backarc continental or arc crust.
DS—long-wavelength dynamic subsidence across retroarc settings (indicated by solid black line). In-plane stress may also affect any flexural
subsidence patterns, depending on whether stress is compressive or tensile (in-plane stress effects are not shown here). Note how various tectonic
subsidence mechanisms may interfere with subsidence patterns caused by other drivers. Patterns are highly generalized. Any of these drivers
of long-wavelength subsidence can be counter-affected by shortening or tectonic inversion across the overriding plate. See text for additional
discussion. (A) Intra-oceanic volcanic arc with old oceanic crust trapped in backarc region. There may be little or no thermal subsidence of the
old, trapped oceanic crust. (B) Backarc region underlain by newly created oceanic crust formed by backarc spreading axis. (C) Compressional
backarc region (retroarc foreland basin system) underlain by continental crust. (D) Extensional backarc region underlain by rifted continental
crust. Synrift subsidence due to crustal thinning is not shown; only postrift thermal subsidence is shown for the extensional retroarc basin.

gradients from the arc to trench tend to be steep for long periods
of time, and shallow-water carbonate deposition will be limited
mostly to fringing reefs and narrow platform systems around
the volcanic arc.
Fault-Related Surface Deformation in Arc Depocenters
Displacement along individual faults in arc settings can
cause shorter wavelength surface deformation. This surface
deformation dramatically influences carbonate deposition if the
deformation occurs in shallow-water environments. Although
a thorough description of surface deformations caused by fault
displacements is beyond the scope of this paper, typical coseis-
mic surface deformations are described in Table 1. Important
aspects of these surface deformations include the following:
1. Coseismic surface deformations are geologically instan-
taneous and can amount to several meters where fault
displacements are greatest, which is typically found at
the midpoint of fault surfaces along which displacement
occurs.
2. Recurrence intervals of earthquakes that cause significant
surface deformation in arc depocenters may be several
60
A uplift of trench-slope break
caused by back-rotation of imbricate
thrust sheets; local uplift patterns
controlled by individual thrust slivers
B uplift of subduction complex
over zone of underplating;
uplift geometry controlled by
geometry of duplex system
C uplift of trench-slope break
caused by back-thrusting as
subduction complex attempts to
maintain critical taper;
local fault-related folding
backthrusting
D regional trench-slope
adjustment and local
fault-controlled subsidence
extensional
faulting
E subsidence caused by tectonic
erosion or sediment subduction;
compaction has similar effect
tectonic erosion
beneath
subduction sediment
complex subduction
Dorobek
decades to several thousand years. Even with compara-
tively long recurrence intervals, surface subsidence and
uplift can accrue to many tens or even hundreds of meters
within a few thousand or tens of thousands of years, which
will have obvious effects on carbonate facies develop-
ment. Longer earthquake recurrence intervals may allow
carbonate platforms to “recover” and fill new accommo-
dation space that was created during an earthquake.
3. Coseismic displacements on great subduction faults can
cause large areas of carbonate platforms and reefs in arc
depocenters to subside or to be instantly uplifted above
sea level (Fig. 1). Coastal and intertidal environments will
be most greatly affected. These events generally affect the
forearc region only, where mechanical coupling between
the subducting and overriding plates is strongest.
4. Coseismic fault displacements can generate tsunamis or
cause mass wasting that can obliterate or bury large areas
of shallow-marine carbonate deposition.
5. Some faults may creep aseismically, and although their
related surface deformations may slowly develop, sig-
nificant amounts of uplift or subsidence can still accrue
over time. Aseismic fault creep may be most common in
subduction complexes where faults cut weakly lithified,
highly sheared, fluid-rich materials.
Active faults in arc depocenters can control carbonate-plat-
form morphology, platform margin-to-basin profiles, and lateral
facies changes during deposition. Active faults can segment pre-
existing carbonate platforms or control the location of carbonate
Figure 3. Possible mechanisms for internal deformation and evolu-
tion of subduction complexes, with associated short- to intermediate-
wavelength surface deformation. Local surface deformations common-
ly are expressed as complex patterns of uplift, which are superimposed
on longer-wavelength patterns of tectonic subsidence. Figure modified
from Dickinson (1995). (A) Accretionary offscraping of successive
imbricate thrust panels (Seely et al., 1974). For growing subduction
complexes, back-rotation of imbricate thrusts causes progressive uplift
of crest of subduction complex (i.e., trench-slope break region). There
also may be local fault-related folding, which creates antiformal sur-
face deformations. (B) Selective subduction and underplating of thrust
duplexes (Sample and Fisher, 1986). Underplating at the base of the
subduction complex can cause uplift over the zone of underplating.
Structural styles related to underplating are often poorly imaged on
seismic profiles but probably are most likely expressed as a duplex
zone. (C) Backthrusting of subduction complex toward forearc basin
(Silver and Reed, 1988). Uplift of trench-slope break and distal part of
forearc basin may occur as the subduction complex maintains a critical
taper. Backthrusting in arcward side of subduction complex generally
causes uplift in trench-slope break region, with local surface deforma-
tion caused by fault-related folding. (D) Denudational normal faulting.
Both regional subsidence and local fault-controlled subsidence may
occur across trench slope as the subduction complex maintains a criti-
cal taper via extensional deformation. (E) Subduction erosion along
décollement beneath subduction complex (cf. von Huene et al., 1982).
Subduction erosion or sediment subduction along underside of sub-
duction complex creates subsidence across subduction complex. Com-
paction of subduction complex may have similar effects.
 Carbonate-platform facies in volcanic-arc settings
Table 1. TYPICAL COSEISMIC SURFACE DEFORMATIONS FOR
DIFFERENT FAULT TYPES
Fault type Fault length Surface uplift-
involved subsidence
Normal faults 5–50 km 0–3 m
Reverse/thrust faults 5–100 km 0–5 m
Subduction faults 100–1000+ km 0–5 m
Strike-slip faults:
Transfer fault 1–50 km 0–3 m
Major intraplate or plate 50–500+ km 0–3 m
boundary
facies tracts wherever displacement rates exceed the ability of
carbonate facies to fill newly generated accommodation space
on either side of a fault. Carbonate lithofacies deposited around
faults that were active during deposition may contain greater
amounts of gravity-flow deposits, tsunamites, or other cata-
strophic-event beds.
DEPOSITIONAL CONTROLS ON CARBONATE
SEDIMENTATION IN ARC SETTINGS
The accumulation of noncarbonate sediments and eruptive
products has important effects on carbonate deposition in arc
depocenters. Various depositional and erosional processes that
are characteristic of arc settings can also affect carbonate deposi-
tion. The relative influence of these factors on carbonate deposi-
tion may depend on the developmental stage of the arc system.
Erosional and Depositional Modifications to
Bathymetric Profiles
The bathymetry of arc depocenters can be rapidly modified
by subaerial or submarine deposition of volcanic materials or
large-scale mass-wasting events. Arc settings are also seismically
active during active subduction, and large-magnitude earthquakes
commonly generate significant ground displacements, so slope
failure occurs frequently. Although erosional and depositional
modifications to shallow-water depositional profiles are relatively
common in arc depocenters, they are (1) episodic and thus diffi-
cult to predict and correlate in stratigraphic successions, (2) vari-
able in lateral extent, and (3) involve highly variable amounts of
rock materials.
Sediment Flux to Arc Depocenters
Sediment flux from the volcanic arc or uplifted arc-massif
rocks can have many effects on carbonate production in adja-
cent depocenters. High sediment flux can completely overwhelm
carbonate-producing benthic organisms. Lesser amounts of sedi-
ment flux might lead to mixed carbonate-siliciclastic-pyroclastic
successions. The caliber of sediment supply from volcanic-arc and
arc-massif rocks can also determine the types of carbonate sedi-
ment produced. For example, late Pliocene carbonate facies in the
New Zealand forearc show major lateral changes in sediment types
61
and stratal patterns (Caron et al., 2004). Elongated, tide- and storm-
influenced balanid (barnacle) shoals grew on local antiformal highs
near the crest of the subduction complex. In contrast, arc-attached,
low-gradient shelf and ramp profiles were built along the arc side
of the forearc basin. Carbonate strata are more sheetlike along the
flank of the arc, are generally transgressive in character, and were
dominated by epifaunal bivalves. These characteristics are thought
to reflect a higher flux of sediment from the adjacent arc.
The transport mechanisms and volume of material involved
in sediment gravity flows will control the size of carbonate plat-
forms or reefs that might be buried during a single event. The
frequency of catastrophic depositional events within a particular
segment of the arc will determine whether adjacent carbonate
platforms and reefs can recover from a series of events. If rela-
tively low volumes of volcanogenic material are involved in any
catastrophic event, and the events occur with low frequency or
are widely dispersed along strike, then carbonate platforms and
reefs may recover and even utilize lava flows or volcaniclastic
facies as substrates.
Mass wasting can also occur at any time along steep vol-
canic edifices, although these catastrophic events are triggered
more often by volcanic earthquakes or by inflation of the volca-
no’s surface as magma ascends into subsurface magma chambers
prior to an eruption. Antecedent drainage networks on the flanks
of a volcano may control dispersal of low to moderate volumes
of volcanogenic material, although dispersal of large-volume,
density-stratified pyroclastic flows are generally less affected by
antecedent topography. The flow paths of only the basal, high-
density part of pyroclastic flows may be diverted by antecedent
topography, such that the resultant coarse-grained deposits are
confined to ravinement floors on the flanks of a volcano. These
high-density flows then spread laterally, where they become
unconfined in lower hillslope regions where antecedent valleys
are less incised and surface gradients flatten. In contrast, the low-
density upper parts of pyroclastic flows are commonly stripped
from the basal, high-density part of the flows. The low-density
parts of pyroclastic flows can flow over topographic obstructions
and lay down more areally extensive deposits. Low-density pyro-
clastic flows can also flow for several kilometers across seawater.
Lava flow paths will also be controlled by antecedent topography,
unless the volume of lava exceeds the ability of valleys and hill-
slope ravinements to contain the flows.
Underfilled versus Filled Arc Depocenters
Sedimentary basins are often described as “underfilled” or
“filled” (cf. Covey, 1986; Flemings and Jordan, 1990; Jordan,
1995), which describes whether depositional systems deposit sed-
iment into a depocenter or carry sediment past a former depocen-
ter. Although a thorough discussion of this concept is beyond the
scope of this paper, it is useful for understanding where and when
widespread carbonate platforms might form in arc depocenters.
Although unique conditions of subduction can determine sub-
sidence and basin-filling patterns in arc systems, arc depocenters
62 Dorobek
are generally more likely to become filled (1) during waning stages
of subduction when tectonic subsidence slows down, (2) in arc
depocenters with high sediment accumulation rates, or (3) in arc
depocenters that begin near base level at the onset of subduction.
Previously deposited siliciclastic sediment and erupted
materials in filled depocenters can provide broad substrates for
shallow-marine carbonate deposition. Areally extensive carbonate
platforms most commonly form when rising sea level floods these
substrates. In addition, tectonic surface deformations become
more important for shallow-water carbonate depositional systems
in filled depocenters. If a filled depocenter is submerged to water
depths that are within wave base and typical tidal ranges (<20 m
water depth), even subtle tectonic surface deformations will sig-
nificantly affect shallow-water carbonate depositional systems.
Oceanographic Circulation
The types of carbonate sediment that accumulate in arc
depocenters will be influenced by the location of the arc system
with respect to global climatic zones or oceanographic circula-
tion systems. Cool-water carbonate facies (James, 1997) will be
more typical of high-latitude arcs or where cold, upwelling water
reaches shallow-water parts of an arc system. Cool-water carbon-
ate deposits typically consist of the skeletal remains of heterotro-
phic and light-independent biota that are dominated by suspen-
sion feeders (e.g., bryozoans) and filter feeders (e.g., barnacles),
which are extremely sensitive to the amount of suspended sedi-
ments and dissolved nutrients in the water column (Henrich and
Freiwald, 1995). Thus, extensive cool-water carbonate produc-
tion may be difficult to establish close to high-latitude arc sys-
tems because of high sediment flux from the arc.
Some arc systems may also have unique physiography with
respect to oceanographic circulation so that anomalous water
masses either enter oceanic passageways or upwelling cold-water
masses impinge on shallow-marine settings and influence carbon-
ate facies development over large parts of the arc system. For exam-
ple, the present-day Gulf of California, Mexico, is a backarc setting
where active seafloor spreading has created an open passage at the
southern end of the gulf. Seasonal variations in upwelling cause
latitudinal variations in nutrient levels within the Gulf of Califor-
nia (Halfar et al., 2004). In the southern part of the gulf, carbonate
production is oligotrophic-mesotrophic and coral reef–dominated.
The central gulf is dominated by mesotrophic-eutrophic, red-algal
sediments, and the northernmost parts of the gulf are dominated by
eutrophic, molluscan-bryozoan sediment assemblages.
Wind Direction
Paleowind direction directly influences the distribution of
high-energy, wave-agitated carbonate-facies belts within car-
bonate platforms (Eberli and Ginsburg, 1989). Volcanic arcs can
build significant topographic relief such that if the arc lies in a
climatic belt with strong prevailing winds, wave-agitated facies
tracts may preferentially build on the windward side of the arc
system, whereas the leeward side of the arc may be dominated
by muddier, low-energy carbonate facies. Volcanic arcs can
also have significant orographic influence on precipitation, with
windward sides of the arc being wetter overall than the leeward
sides. Higher precipitation on the windward side of an arc may
mean greater chemical and physical weathering, with a greater
flux of sediment, especially finer grained sediment, to adjacent
marine depocenters. High weathering, erosion, and transport of
sediment to the windward side of an arc can greatly suppress
carbonate production.
Prevailing winds can also greatly influence the dispersal of
erupted pyroclastic material, especially volcanic ash. Leeward
depocenters may have much greater accumulations of volcanic
ash (Sigurdsson et al., 1980), which also can suppress carbon-
ate production.
Nutrient Supply
High nutrient levels in seawater suppress carbonate produc-
tion (Hallock and Schlager, 1986). A high nutrient flux from
arc systems may be related to intense chemical weathering of
arc rocks, eruption products, or, in the case of modern settings,
anthropogenic input, and can have major deleterious effects on
carbonate production. Upwelling of cold, nutrient-rich deep
water can also suppress carbonate production or influence car-
bonate-sediment types in some arc settings (Halfar et al., 2004).
CARBONATE DEPOSITION WITHIN ARC
DEPOCENTERS
Shallow-marine carbonate platforms and reefs can form
in forearc, intra-arc, and backarc settings. Various drivers and
expressions of tectonic subsidence and uplift affect each setting
and may vary as the arc system evolves. Temporal and spatial
changes in differential tectonic subsidence have the greatest
effects on carbonate sedimentation when arc depocenters are
nearly filled to sea level. Depositional processes and sediment
flux from the arc system also vary between arc depocenters,
which also can influence carbonate deposition.
The carbonate-platform classification scheme of Read
(1985) is used throughout this paper. Stratigraphic intervals that
record progressive surface deformation during deposition are
growth strata (cf. Vergés et al., 2002). Carbonate growth strata
are deposited around or on top of actively growing tectonic struc-
tures and can be used to reconstruct both the kinematic history of
individual structures and basin-scale patterns of tectonic subsid-
ence and uplift.
Carbonate Platforms and Reefs across the Forearc Region
Many forearc regions are characteristically too deep for
shallow-marine carbonate sedimentation, especially during early
stages of subduction and basin development. Thus, extensive
shallow-marine carbonate platforms are not likely to develop
 Carbonate-platform facies in volcanic-arc settings
across the forearc region until deformation or basin filling has
provided suitable substrates for carbonate deposition.
Using the classification scheme of Seely (1979) and Dick-
inson and Seely (1979) for forearc settings, and assuming that
the forearc region is at the proper (paleo)latitude, carbonate-
platform development is possible where (1) the crest of the sub-
duction complex forms a shoal water or emergent ridge (Figs. 4,
5); (2) the forearc basin is largely filled with sediment so that a
broad, shallow, shelf-like substrate extends from the volcanic arc
toward the trench-slope break (Fig. 5); (3) the arc massif in front
of the volcanic arc is shallow and provides a suitable substrate
for carbonate sedimentation; and (4) material flux (i.e., volca-
niclastic sediment, ash fall, lava flows) from the volcanic arc
does not suppress shallow-marine carbonate deposition across
the forearc region.
These conditions are similar to those required for carbonate-
platform development in other tectonically active basins. Silici-
clastic and volcanic flux must be sufficient to fill depocenters
to sufficient levels so that regional substrates are available for
carbonate sedimentation under the proper sea-level conditions
(generally high-frequency transgressive events) but not over-
whelm and terminate carbonate production. Until a forearc basin
is nearly filled with sediment and a suitable shelf-like profile is
constructed, only structurally high parts of the subduction com-
plex or frontal part of the volcanic arc will be shallow enough for
carbonate-platform development.
Carbonate Platforms and Reefs on Subduction Complexes
The first area where shallow-water carbonate facies may
develop across forearc settings is the subduction complex
(Figs. 4, 6). Most carbonate bodies associated with ancient sub-
duction complexes consist of highly recrystallized and sheared
pelagic carbonate facies that have been tectonically accreted and
incorporated into the subduction complex. Carbonate-capped
seamounts on the subducting plate or carbonate strata on conti-
nental or arc microplates may also be obducted or accreted to the
frontal part of subduction complexes (e.g., Permian Nabeyama
Formation, central Japan, Minoura, 1992; Daiichi-Kashima
Seamount, Japan Trench, Cadet et al., 1987; Shin Tani, 1989;
Trench
Slope Arc - Trench Gap
Trench
Slope
Slope Break
Trench
Basins Forearc Basin
Fill
? ? ?
? ??
Subducting Slab ? ??
of Oceanic Crust
20
Subduction
10
km Complex
10 20 30 40 50 km
63
Eratosthenes Seamount, Mediterranean Sea, Galindo-Zaldívar
et al., 2001). Tectonic uplift and growth of the subduction com-
plex, however, can also raise parts of the subduction complex
into the photic zone and possibly even above sea level, creating
substrates for shallow-marine carbonate deposition.
Reefs and associated carbonate facies can fringe emergent
parts of the subduction complex, such as in emergent islands of
the western Indonesian forearc. Variably sized isolated platforms
can also form on shallow, submerged parts of subduction com-
plexes. Carbonate-platform sequences that form on subduction
complexes are rarely more than a hundred meters thick, and apart
from fringing reef systems, the platforms commonly have ramp-
like profiles that mimic the depositional gradients across the top
of the subduction complex. Carbonate-platform sequences on
subduction complexes typically do not thicken or steepen over
time and then evolve into rimmed platform morphologies, which
reflects the relatively short life of the platforms, the rapidly
changing patterns of uplift and subsidence across the top of the
subduction complex, and the deleterious effects of volcaniclastic
material supplied by the volcanic arc.
Carbonate platforms and reefs on top of subduction com-
plexes may be rare in the rock record because (1) the arcward
side of the subduction complex may not become shallow enough
for carbonate-platform development until the subduction com-
plex has become tectonically uplifted or the forearc basin has
become nearly filled with sediment, which are conditions likely
to develop only for short periods of time in the development of
forearc regions with highly accretionary subduction complexes
(Clift and Vannuchi, 2004); or (2) thin carbonate-platform strata
that formed on the subduction complex are misinterpreted as
allochthonous units that have been tectonically incorporated
into the subduction complex. Recognizing carbonate facies that
formed on shallow to emergent parts of highly deformed subduc-
tion complexes requires careful age dating and identification of
depositional contacts (typically angular unconformities). Older
carbonate sequences that form on top of, but are subsequently
incorporated into, actively deforming subduction complexes will
likely have highly sheared contacts with other lithotectonic units
in the subduction complex.
Figure 4. Schematic cross section of
the forearc region. Stippled pattern
Volcanic represents undeformed or only mildly
Front
deformed forearc basin fill. The forearc
Eroded basin shown here is a largely filled dep-
arc rocks ocenter. Crust beneath the forearc basin
may be highly variable in composition
Pluton
but is shown in this diagram as mafic,
Arc Massif
probably oceanic crust. Some forearc re-
gions may not have the well-developed
subduction complex that is depicted in
this cross section. Modified from Dick-
Arc Magma
inson (1995).
Source
64 Dorobek

UNDERFILLED FILLED rates for subduction complexes can be derived by comparing the
elevations of well-dated, subaerially exposed Quaternary reef
terraces to estimates for eustatic sea-level curves for Quaternary
Tsb Tsb time (Fig. 7).
Tr Tr “Chimneys” of precipitated carbonate minerals have also
A. Sloped B. Sloped been documented from dredge samples and by observations
(slope basin) (slope basin) made from submersible vehicles across the surface of many mod-
ern subduction complexes (Kulm et al., 1986; Kulm and Suess,
1990; Haggerty, 1991; Hattori et al., 1995). These irregular col-
umns of precipitated carbonate are associated with fluid seeps
Tsb Tsb that discharge from permeable fault zones that cut to the top of
Tr Tr the deforming subduction complex (cf. Moore and Vrolijk, 1992).
The seeps provide nutrients for chemosynthetic fauna, including a
C. Ridged D. Terraced complex microbial community. Abiogenic and biologically medi-
(submerged ridge) (deep marine) ated carbonate precipitation occurs as the cold, methane-rich pore
fluids discharge from the episodically transmissive faults. These
Tsb Tsb carbonate chimneys are not limited to photic depths and can form
anywhere along the top of the subduction complex as long as it
is above the carbonate compensation depth. Volumetrically, how-
Tr Tr
ever, carbonate chimneys represent only a small fraction of the
E. Ridged F. Shelved total sediment accumulation within the subduction complex.
(shoal-water ridge) (shallow marine) Ancient examples. Pre-Quaternary examples of shallow-
marine carbonate facies deposited on subduction complexes are
Tsb Tsb rarely described in the literature. Ancient seep-related carbonate
chimneys are reported in Eocene strata of the Barbados subduc-
tion complex (Larue and Suess, 1985).
Tr Tr One example is provided by Upper Cretaceous(?) to middle
G. Ridged H. Benched Eocene fringing reefs and carbonate ramp sequences that formed
(emergent ridge) (terrestrial) at intermittent times during development of the Central Ameri-
can forearc, Costa Rica (Lundberg, 1982; Seyfried et al., 1991;
Figure 5. Morphologic variants of forearc basins, with basin fill stip- DiMarco et al., 1995; Kolarsky et al., 1995). The Upper Creta-
pled and vertically exaggerated for clarity. Left-side diagrams show
underfilled forearc basins (most common for island arcs); right-side ceous(?) (Campanian–Maastrichtian) Barra Honda platform is a
diagrams show filled forearc basins (most common for continental- cryptic, <100-m-thick, mud-rich platform sequence with rudist
margin arc systems). Symbols: dashed lines—sea level; Tr—trench; reefs on its windward side that formed on remnants of an under-
Tsb—trench-slope break (coincides with shelf-slope break for shelved lying Cretaceous arc-forearc complex (Seyfried et al., 1991).
forearcs). Modified from Seely (1979) and Dickinson and Seely DiMarco et al. (1995) suggest that the Barra Honda platform
(1979); from Dickinson (1995).
strata may be largely Paleocene in age. Kolarsky et al. (1995)
also interpreted the basement beneath the Paleogene forearc
region of southern Costa Rica and adjacent Nicaragua and Pan-
Quaternary examples. Quaternary carbonate reefs and ama as uplifted Cretaceous oceanic crust or seamounts. These
associated facies fringe emergent parts of modern subduction contrasting interpretations are mentioned here to illustrate how
complexes in the forearc regions of Sumatra (Simeulue, Nias, difficult it can be to interpret the history of these complex set-
Siberut, Sipora, and Enggano islands), the Solomon Islands tings. Regardless, the Barra Honda platform underwent episodes
(Tetepare, Rendova, Ranongga islands), and the Lesser Antilles of shallow submergence and carbonate accumulation, followed
(Barbados). There is little published information on the thickness by intermittent subaerial exposure and karstification, which have
and lithologies of offshore Holocene reefs and related facies in been attributed to superimposed eustatic fluctuations and tectonic
these examples. uplift caused by accretionary growth of an underlying subduction
Many modern subduction complexes are partially emergent. complex (Seyfried et al., 1991). Continued development of the
Pliocene to Quaternary reefs and other shallow-water facies may Central American arc-forearc system during latest Cretaceous to
crop out as a series of emergent coastal terraces across the crest Paleocene time was associated with further structural and bathy-
of the subduction complex. These terraces record both eustatic metric differentiation of the subduction complex (or “outer arc”
sea-level changes and progressive uplift of the subduction com- of Seyfried et al., 1991), forearc basin, and volcanic arc. Fora-
plex (Mesolella et al., 1969; Wheeler and Aharon, 1991; Mann et minifer-rich ramp facies prograded from both sides of the sub-
al., 1998; Nunn, 1998a, b, 2000; Dickinson, 2001). In fact, uplift duction complex during Late Cretaceous to late Paleocene uplift
 Carbonate-platform facies in volcanic-arc settings 65

SUBDUCTION COMPLEX DEPOZONE

platform types:
• isolated platforms & buildups
substrates: FOREARC BASIN DEPOZONE
• bathymetric crest of subduction complex platform types:
• may be smaller buildups within subduction complex; • small isolated buildups & reefs
probably more common in arcward parts on structural highs
• may extend arcward into forearc basin if distal • may be faulted or structurally
part of forearc basin is filled complicated by later
deformation events
substrates:
• intrabasinal structural highs
• transpressional to transtensional
structures may be typical

?
2 km ?
1 VE ~ 10 ?

10 20 30 40 50 km

Oceanic crust

‘Transitional’ forearc crust ARC-FLANK & FOREARC

BASIN DEPOZONE
Arc massif platform types:
• small isolated buildups & reefs
Subduction complex • arc-fringing platforms & reefs
• typically thin (<100 m thickness)
Trench fill • ramps to low-relief rimmed shelf profiles
• platforms tend to become wider as forearc basin fills
Forearc basin fill with sediment
substrates:
Carbonate platform/reef facies • faulted arc basement
• volcaniclastic substrates during later stages of filling

Figure 6. Schematic cross section of forearc region, showing generalized types and locations of carbonate platforms.
The forearc basin shown here is largely filled with sediment. Underfilled conditions would characterize the forearc basin
during early stages, with smaller platforms and buildups on structural highs or building from shallow, submerged parts of
the arc massif. Crust beneath forearc basin may be highly variable in composition, but is shown in this diagram as mafic,
probably oceanic crust. Note how carbonate platforms across the forearc region tend to become more areally extensive
over time as the forearc basin fills with sediment or as the subduction complex grows, both of which provide progres-
sively wider substrates for shallow carbonate sedimentation. Basin model modified from Dickinson (1995).

of the outer arc. Eocene mixed carbonate-siliciclastic strata (Mal
País Formation) contain foraminiferal limestone units up to sev-
eral tens of meters thick. Strata of the Mal País Formation also
overlie, with angular unconformity, older strata that draped base-
ment rocks of the Nicoya Peninsula. Nicoya basement rocks are
thought to represent an exhumed subduction complex by some
workers (cf. Lundberg, 1982).
Carbonate Platform and Reefs in Forearc Basins
Carbonate platforms and reefal buildups make up much of
the stratigraphic fill in some forearc basins. If the forearc basin is
at the proper (paleo)latitude, tropical carbonate facies can build
away from the seaward side of the volcanic arc or arc massif
(Fig. 6). This is commonly the bathymetrically shallowest part
of the arc-trench system when subduction is active and the vol-
canic arc is being constructed by volcanic activity. Carbonate-
platform facies can accumulate to several hundred meters thick
and extend for tens to several hundred kilometers along strike.
Volcaniclastic strata typically separate platform sequences
within the forearc basin fill and reflect pulsed eruptive activity
or episodic reorganization of dispersal systems that deliver vol-
caniclastic sediment to the forearc basin (Beaudry and Moore,
1985; Dickinson, 1995).
Carbonate deposition in forearc basins is influenced by pro-
cesses that operate at different time scales. At relatively short
time scales (i.e., 105–106 yr), thin fringing reefs and narrow plat-
form systems can form around volcanic edifices at virtually any
time in the history of a volcanic arc and forearc basin, as long as
66 Dorobek

wave-cut notches
A indicate short stillstands

relative sea level

30 m
time
20

10
Figure 7. Technique for determining
0 high-resolution rates and magnitudes
of tectonic uplift or relative sea-level
change in shallow-marine carbonate fa-
E = present cies. Wave-cut notches are a response to
B elevation A (= duration of uplift)
Meters falling sea level, resulting from either
Displacement (D)

app rate = D/A an actual eustatic fall or tectonic uplift.

are During short-term stillstands within a
=E-e

nt 40 longer-term trend of relative sea-level

Elevation (relative & real)

displacement (apparent)
sea fall, waves have time to erode coastal
re lev exposures and create notches. From
la e l Burbank and Anderson (2001).
tiv
e = elevation e
present
at time of
formation sea level se 0
a
le
ve
l
rea - 40
l
sea
an nt lev
m e em e e
isp
la c l
d
- 80
120 80 40 0
Age (ka)

submerged parts of volcanic features are within the photic zone
and sediment flux from the arc does not overwhelm carbonate
production. Development of laterally extensive and relatively
thick carbonate platforms, however, requires both longer periods
of time, probably on the order of several million years, long-term
tectonic subsidence, and the continued availability of shallow-
water substrates. These conditions are more likely to develop
during later stages of forearc-basin evolution, when tectonic
deformation and arc volcanism are diminishing, forearc subsid-
ence rates are slowing, and the forearc basin is largely filled with
sediment (cf. Dickinson, 1995).
Proximity to the volcanic arc means that the forearc basin
will be greatly influenced by sediment flux from the arc. Dur-
ing earlier phases of forearc-basin evolution, volcanic eruptions
and seismic events may be more frequent. Catastrophic eruptions
or mass wasting of volcanic edifices can instantly terminate car-
bonate systems in forearc basins, either by burying carbonate-
producing organisms or by incorporating previously deposited
carbonate strata into mass flows.
Platform types. The locations and morphologies of carbon-
ate platforms found in forearc basins will be strongly controlled
by the evolving topography-bathymetry across the basin, which
depends on the evolutionary stage of the basin. In immature,
underfilled forearc basins, fringing reefs and narrow, steep-sided
platforms may form only along the trench-facing side of the arc
and arc massif. During intermediate stages of forearc basin evolu-
tion, isolated mounds and reefal buildups might form on remnant
or actively forming structural highs across any part of the forearc
basin; intrabasinal structural highs are most common in forearc
basins that are segmented by transcurrent strike-slip faults. For
example, transtensional deformation of the late Miocene Hel-
lenic forearc basin on Crete created half-grabens where footwall
highs became sites for local patch-reef development within the
basin (ten Veen and Postma, 1999).
 Carbonate-platform facies in volcanic-arc settings
As the forearc basin is filled with sediment and the frontal
part of the arc massif is eroded, regional depositional gradients
generally flatten, and broad substrates develop where shallow-
marine carbonate systems can develop. During mature, filled
stages of forearc-basin development, carbonate systems com-
monly construct ramp or low- to moderate-relief shelf profiles
that prograde into the forearc basin from erosional remnants of
the volcanic arc (cf. Beaudry and Moore, 1985; Dickinson, 1995).
Carbonate sedimentation, however, may be terminated at any
time across a forearc basin during major eruptive events or during
massive hillslope failure that force carbonate-producing environ-
ments to be reestablished elsewhere within the basin. Short- to
long-term tectonic uplift of the arc and arc massif may cause car-
bonate environments to become subaerially exposed and either
terminate carbonate deposition or force the carbonate factory to
move elsewhere.
Local seep-related carbonate buildups have also been identi-
fied in some ancient forearc basins. Examples include the Juras-
sic Fossil Bluff Group, Antarctica (Kelly et al., 1995) and the
Eocene Joes River mélange, Barbados (Larue and Suess, 1985).
Quaternary examples. Many modern forearc basins from
tropical settings are bordered by extensive fringing reefs and
carbonate platforms that built away from the volcanic arc or
arc massif. For example, Quaternary fringing reefs and narrow
rimmed platforms are found along segments of the eastern Indo-
nesian forearc basin, where sediment flux from the volcanic arc
is relatively low. In contrast, modern carbonate facies are much
less extensive in forearc depocenters along the western Indone-
sian archipelago because there is greater sediment flux from the
volcanic arc. Carbonate deposition is suppressed in the forearc
basins of Sumatra and Java because the modern volcanic arc
has sufficient relief to cause significant orographic effects on
precipitation. The winter monsoon blows eastward across the
forearc basins of Sumatra and Java and brings heavy rainfall to
southwest-facing slopes of the western Indonesian volcanic arc.
Meteoric runoff actually lowers salinity in the Java Sea to ~30
ppt during the winter monsoon season. The seasonal decrease in
salinity and increase in the amount of suspended sediment deliv-
ered to the shallow shelf areas around Sumatra and Java suppress
carbonate sedimentation. Thus, modern fringing reefs are geo-
graphically limited to the northwestern and southeastern tips of
Sumatra and a few offshore islands within the proximal forearc
basin. Subsurface data suggest that volcaniclastic flux from the
arc had less influence on carbonate sedimentation during early
Miocene time, when carbonate shelves prograded away from the
Sumatran arc massif (Beaudry and Moore, 1985), which might
reflect a weaker monsoon at that time.
Modern microatolls (i.e., massive, ring-shaped coral heads
up to a few meters in diameter) provide a high-resolution record
of subtle uplift and subsidence across the Sumatran forearc basin
(Fig. 8; Sieh et al., 1999; Zachariasen et al., 1999). Modern coral
heads in shallow-water areas surrounding the outer ridge of the
Sumatran forearc basin grow upward and outward until they reach
annual lowest low tide (Zachariasen et al., 1999), at which point
67
they can only grow outward. Relative sea-level changes at any
given location across the Sumatran forearc region then shape the
internal growth structure of these microatolls. Coral heads grow-
ing on stable substrates develop nearly flat upper surfaces (Sieh
et al., 1999). When subjected to gradual submergence, the living
coral colony along the outer parts of these microatolls develops
a raised rim, whereas their older, dead interiors become depres-
sions. Microatolls subjected to gradual emergence typically
develop conical shapes because growth is limited to progressively
lower and outer parts of the expanding structure. Careful analysis
of growth bands within coral heads from the western coast of
Sumatra and the outer-arc islands of western Indonesia showed
temporal and regional patterns of submergence and emergence
across the Sumatran forearc over the last several hundred years,
with microatoll uplift during the giant earthquake of 1833 serving
as a key correlation horizon (Sieh et al., 1999; Zachariasen et al.,
1999). The microatoll record suggests rapid submergence over
several decades prior to the 1833 earthquake. Rates of pre-1833
submergence ranged from 5 to 11 mm/yr and increased trench-
ward across the Sumatran forearc (Zachariasen et al., 1999). The
microatolls record coseismic uplift during the 1833 earthquake
and can be observed at six sites along a 125-km length of the
subduction zone, with uplift on the order of 1–2 m. Similar to the
interseismic submergence patterns, coseismic uplift also appar-
ently increased toward the trench.
The great Sumatran earthquakes of December 2004 and
March 2005 produced significant, instantaneous uplift and sub-
sidence across the Sumatran forearc (Fig. 1). The differential
coseismic subsidence and uplift were expressed over a 300 km
distance from the trench axis and >12° of latitude (>2000 km);
modern coral reefs and coastal deposits were uplifted tens of
centimeters to >2 m locally. A regional, arc-parallel hinge line
apparently separates uplifted from subsided parts of the forearc,
with uplifted regions on the trenchward side of the forearc basin.
This uplift may be preserved in the rock record as a regional dis-
conformity within shallow-water strata. It would be difficult to
recognize whether these disconformable surfaces in forearc strata
were related to coseismic surface deformation or high-frequency
sea-level falls. Regardless, the recent Sumatran earthquake pro-
vides an important reminder of the rapid uplift and subsidence
that can occur in forearc settings.
Other modern forearc basins where Quaternary carbonate
facies are found include (1) the forearc basin along the southeast-
ern side of North Island, New Zealand (Kamp and Nelson, 1988),
and (2) the island of Espiritu Santo, Vanuatu (Wells, 1988; John-
son and Greene, 1988; Greene et al., 1988; Cabioch et al., 1998).
North Island, New Zealand. In the modern forearc basin of
North Island, New Zealand, thin veneers of carbonate sediment
are being deposited over only ~3% of the forearc region (Kamp
and Nelson, 1988). This limited area of modern shallow-water
carbonate sedimentation reflects the combined deleterious effects
of the relatively cool-water conditions of offshore eastern New
Zealand, the high-energy wave-swept conditions of this forearc
setting, and the flux of siliciclastic sediment from North Island.
 Figure 8. (A) Schematic cross sections
that depict the effect of relative sea-level
changes on coral growth. The outermost
solid-black growth band depicts the most
recent coral growth band. Solid gray
A (a) (b) bands are dead coral growth bands that
grew under the same sea-level conditions
SL as the living coral. Dashed growth bands
Living
are dead bands that grew under differ-
coral ent, prior sea-level conditions. (a) Hemi-
SL surface SL spherical coral head below highest level
of survival (HLS), with a new skeleton
accreting in a concentric fashion on the
time Hemispherical Coral Head time Cylindrical Coral Head
outside of the head. (b) The same coral
head under the same sea-level conditions
after it has reached HLS. No additional
(c) (d) upward growth is possible, although lat-
eral accretion continues along the sides
? of the coral head that are below HLS.
(c) “Hat” morphology of a coral head
SL SL that grew up to HLS, but then underwent
a drop in HLS. The part of the coral
exposed above HLS has died. Lateral
time “Hat” Microatoll time “Cup” Microatoll growth below the new HLS develops a
lower outer rim around a higher center.
The elevation difference between the
two flats is a measure of the amount of
B 10 emergence. (d) “Cup” morphology of
a microatoll that underwent a sea-level
centimeters

UP rise after the coral had been growing at

e

em
nc

HLS. The coral grows upward toward the

e rg
ge

enc new HLS, constrained only by its growth

e HLS
er

rate. Upward and outward growth over

bm

in the old HLS surface produces a raised

su

0 1995 outer rim, indicative of submergence.

66 67 68 69 The elevation of the new HLS is not re-
64 65 7071 727374
62 63 corded by the coral until the coral grows
75 76 up to it. From Zachariasen et al. (1999).
1961 77
78 (B) Cross section of a microatoll from
79 the Indonesian forearc. The X-rayed
80 thin slab reveals a clear record of annual
81 growth bands that expanded radially out-
82
83 ward (from left to right) at ~1 cm/yr. The
84
85 HLS of the coral during the past 35 yr is
86
87 91 92 93 96
88 89 90
recorded in the topography of the coral’s
9495 97
upper surface. The arrows track the ap-
sand substrate parent rise of sea level in the 1960s and
its subsequent apparent fall. Growth pat-
terns in corals such as these can provide
high-resolution records of true sea-level
change or the effects of incremental co-
seismic subsidence or uplift. From Sieh
et al. (1999).
 Carbonate-platform facies in volcanic-arc settings
Beyond the modern shelf, carbonate sediments (<0.5 m thick)
cover the tops of isolated, elliptical banks (10–50 km2 in area)
that rise from the slope to depths of 150–500 m (e.g., Madden
Banks). The banks may represent the submarine expression of
surface deformation on top of the subduction complex. Carbon-
ate sediments that form thin veneers on these submarine ridges
consist of coarse skeletal sand (dominantly a cool-water bivalve-
foraminifer assemblage with lesser coral, brachiopod, gastropod,
and echinoid fragments). Patchy mosaics of coarse skeletal sand
are also found on the shallow shelf updip from the slope banks.
These skeletal-sand accumulations on the modern shelf largely
consist of bryozoans, corals, barnacles, bivalves, and benthic
foraminifers and are associated with local patches (a few to tens
of square kilometers in area) of gravelly sediment on rugged, cur-
rent-swept seafloor topography.
Espiritu Santo, Vanuatu. Holocene carbonate sedimentation
around the island of Espiritu Santo, Vanuatu, occurs above actively
uplifting volcanic-arc basement (cf. Cabioch et al., 1998). The
Cenozoic tectonic history of Espiritu Santo is complex (Meffre
and Crawford, 2001). The Vanuatu (or New Hebrides) archipelago
can be divided into Western, Central, and Eastern Belts (Greene
et al., 1988), which record discrete and geographically separated
phases of volcanic activity. The Western Belt, which includes
Espiritu Santo, consists largely of upper Oligocene to middle Mio-
cene calc-alkaline lavas and associated volcaniclastic rocks that
formed the volcanic arc above a westward-dipping but currently
extinct subduction zone to the east. Subduction along the eastern
side of Vanuatu ceased during late Miocene time and switched to
the western side of the archipelago, where an eastward-dipping
subduction zone now exists. Late Miocene to Pliocene volcanic
rocks of the Eastern Belt record the volcanic arc that formed above
the east-dipping subduction zone. Late Pliocene–Holocene vol-
canic rocks of the Central Chain record another westward shift
of the volcanic arc toward the New Hebrides Trench. Remnants
of the older Oligocene–Miocene volcanic arc of the Western Belt
40 S wave-cut
notches
30
20
coral reefs
Drill holes
10
0 10,964
-10
13,959
12,061
15,780
-20
14,463
18 16
20
20,961
-30 21,303
>30,000
-40
(m) 0 20 40
69
now form the basement rocks of the present-day Vanuatu forearc
region. A subduction complex is absent along this margin owing
to its young age and the limited amount of sediment delivered to
the trench.
The central part of the Western Belt, including Espiritu
Santo, has been actively uplifting throughout Quaternary time
owing to subduction of the d’Entrecasteaux Ridge at this loca-
tion. The uplift history of Espiritu Santo is recorded by a series of
emergent and notched Holocene reef terraces that were deposited
on the uplifting Oligocene–Miocene volcanic substrate (Fig. 9;
Cabioch et al., 1998). Although we know the present-day tectonic
setting of Espiritu Santo, and the Holocene reef terraces should
be described as having formed in a forearc setting, it might be
difficult to make the same tectono-stratigraphic interpretation if
similar carbonate terraces were found in the rock record. More
likely, these reef terraces would be described as “intra-arc basin”
deposits (see discussion below about intra-arc carbonate facies),
simply because they are found stratigraphically above deformed,
arc-related rocks. Seismic profiles from offshore Espiritu Santo
also show drowned Quaternary carbonate reefs and reef caps (1–
6 km across, some reefs up to 300 m relief, reef tops at ~300 m
water depth) and “mounded” sediment bodies (375–450 m water
depth) (Fig. 10; Johnson and Greene, 1988; Greene et al., 1988),
which formed on submerged and complexly deformed parts of
the relict arc basement. The locations of some reefs and shelf
margins are probably fault controlled. These highly variable pat-
terns of Quaternary carbonate sedimentation around Espiritu
Santo reflect the complex differential subsidence and uplift that
can occur across young, incipiently formed forearc regions. In
this case, Espiritu Santo represents subaerial remnants of an old
volcanic arc that were deformed and became basement for the
forearc region of a younger volcanic arc.
It is worth noting that Quaternary carbonate facies appar-
ently are extremely rare within forearc basins along the west-
facing, convergent margins of North and South America, even
6 ka reef flat N
Figure 9. Cross section of the uplifted
Tasmaloum fringing reef, Vanuatu,
6 showing the 6, 8, 10, 12, 14, 16, 18, and
8 20 ka time lines. Only the most char-
10 acteristic dates are reported (those in
present mean sea level
italics are 14C dates converted to calen-
12 dar years). A 6 ka reef flat is now 40 m
above sea level, which shows the rapid
14
? uplift rates that have characterized this
? forearc region over the last 6000 yr.
volcanic
? From Cabioch et al. (1998).
substrate?
Late Pleistocene sand
60 120 180 240m
 Vanua
A 165°
Sa
170°
B Lava 0 50km

V OT
nt
a
Cr
uz
Is
la
14° S

TANDE TRO
n
Area
ds
HEBR
NORT TRENCH Vanikolo shown
N Santa
Basin
in B Maria
IDES
H NE

UGH
r
10 cm/y
Section
W

Vanua Lava Banks

E F.Z. shown
HOLM
Basin
Santa Maria HAZEL in C
SMFZ
Central FZ 15°
Basin AB
Espiritu
D’ENTRECASTEAUX Santo Aoba
PACIFIC PLATE
ZONE
Espiritu
ba
Ambrym
Malakula
AMFZ Santo Ao
EFZ
Epi

AUSTRALIA-INDIA PLATE

Pent
CO
RI

ecos
OL

Volcanic Associations VFZ

t
IS

Late Miocene to Holocene 16°

SOUT

(Central Chain)
TRO

Late Miocene to early Pliocene

NE TR
M
H

(Eastern Belt) Ambr

UG

al ym

W EN
Late Oligocene to middle Miocene ak
H

ul

HE CH
(Western Belt)
NE

BR
20°
W

Tectonic Elements

ID
E
HE

Uplifted ridge

S
BR

Rifted basin Epi

ID

167° E 168°
ES

Intra-arc basin
TR

Insular shelf
EN

Major fracture zone

CH

Trace of plate contact, dashed

where buried Insular slope
RE
0 100 200km ACTU
ER FR
Relative plate motion
HUNT
ZONE Deep basin

Quaternary
0
Quaternary
reefs C
1
UMiocene-
~3 km Pliocene
water 2 SECONDS (TWT)
Quaternary depth
East Santo slope deposits
basin fault
3
5 km Maewo
fault
4
Quaternary Eastern
North Aoba
167°E 168°E
basin fault
UMiocene-Pliocene 5
15°S
Maewo
Espiritu Aoba
Santo UOligocene-MidMiocene 6
 Carbonate-platform facies in volcanic-arc settings 71

Figure 10. (A) Main geologic and tectonic features of the New Hebrides Arc. SMFZ—Santa Maria fracture zone; ABFZ—Aoba fracture zone;
AMFZ—Ambrym fracture zone; EFZ—Epi fracture zone; VFZ—Vulcan fracture zone. Modified from Greene et al. (1988). (B) Main geologic
and tectonic features of the central New Hebrides Arc, Espiritu Santo region. Basement rocks of Espiritu Santo consist of late Oligocene to
middle Miocene volcanic and volcaniclastic rocks erupted from a paleo-volcanic arc (Western Belt rocks) that formed above a now-extinct,
westward-dipping subduction zone. During late Miocene time, however, the subduction zone switched to the western side of the New Hebrides
Arc (present-day New Hebrides Trench), and an east-dipping subduction zone developed beneath arc rocks of the Western Belt, which includes
Espiritu Santo. The active volcanic arc to this subduction zone consists of volcanic centers that make up the Central Chain, to the east of the
Western Belt. Thus, Western Belt rocks now constitute the basement of the modern forearc region of the Central Chain, and Espiritu Santo is
in a transitional tectonic state. Depocenters that formed during extension of the older Oligocene–Miocene arc (Western Belt rocks) might be
considered intra-arc basins. The volcanic centers of the Western Belt are no longer active, however; so the present-day depocenters should be
considered part of the forearc of the Miocene–Holocene volcanic arc (i.e., the Central Chain). Modified from Greene et al. (1988). (C) Interpreted
seismic-reflection profile across the East Santo and North Aoba basins, New Hebrides arc system. Note Quaternary pinnacle reefs–buildups along
the outer, fault-controlled margin of the East Santo shelf. Modified from Greene et al. (1988).

those margin segments that lie within tropical climatic zones.
The limited amounts of modern carbonate sedimentation along
these margins probably reflects the strong coastal upwelling sys-
tems that bring cold, nutrient-rich bottom water onto the shallow,
narrow shelf areas that border the western sides of these forearc
basins. In contrast to modern continental margins, extensive
Paleozoic carbonate platforms formed along the eastern side of
the paleo–Pacific Ocean because marginal oceanographic barri-
ers (e.g., island arcs, orogenic belts, and microplates) may have
protected the carbonate platforms from the deleterious effects of
upwelling (Whalen, 1995).
Quaternary carbonate reefs and shallow-platform facies are
also found in some relict forearc basin settings where subduction
is no longer active but where the basin’s bathymetry and structural
features were inherited from previous times of active subduction.
In some ways, these basins might be considered forearc succes-
sor basins. For example, the Indispensable Basin of the eastern
Solomon Islands has been interpreted as either a relict forearc
basin or the leading edge of the partially subducted Ontong-Java
Plateau (Vedder and Bruns, 1989). The Ontong-Java Plateau
resisted southwestward-directed subduction when it collided with
the Solomon Islands arc in middle to late Miocene time, forcing a
change in subduction polarity and initiation of a younger, north-
eastward-directed subduction zone beneath the southwestern
side of the Solomon Islands (Vedder and Bruns, 1989). Tectonic
structures and remnant bathymetry along the relict forearc side
of the Solomon Islands continue to influence modern sedimenta-
tion there, even ~15 m.y. after the change in subduction polarity.
For example, seismic profiles along the southwestern slope of the
Indispensable Basin show several drowned Pleistocene(?) reefs
at ~300–500 m water depths along a steep, faulted(?) escarpment
just southeast of Santa Isabel Island (Niem, 1989). The steep
bathymetric gradient across the forearc region and the low relief
and relatively small surface area of the relict volcanic arc sug-
gest that this relict forearc basin will probably remain underfilled
for a long time into the future, thus limiting the potential surface
area for shallow-marine carbonate sedimentation. Bathymetric
profiles around offshore parts of Viti Levu, Fiji, are structurally
controlled and record older Neogene to possibly Quaternary
deformation. Faults that cut the older Neogene forearc basin and
frontal arc massif influenced the location, dimensions, and mor-
phology of Neogene to Holocene reefs across the area (Fig. 11;
Rodd, 1993).
Ancient examples. Ancient examples of forearc-basin car-
bonate sequences include (1) Paleogene carbonate platforms on
transpressional ridges that transected the forearc basin of Costa
Rica (Krawinkel and Krawinkel, 1996); (2) Oligocene to Miocene
reefs and platform facies (Tau, Wailotua, and Qalimare Lime-
stones) in the Outer Melanesian forearc basin on the island of
Viti Levu, Fiji (Hathway, 1994, 1995; Rodd, 1993); (3) Miocene
Rethymnon Formation, eastern Crete (Pomoni-Papaioannou et
al., 2003); (4) Miocene mounded carbonate buildups to carbon-
ate shelf systems in several sub-basins of the Indonesian forearc
(Beaudry and Moore, 1985; Matson and Moore, 1992; van der
Werff, 1996); and (5) Pliocene “cool-water” skeletal limestones,
Hawke’s Bay, New Zealand (Kamp et al., 1988; Ballance, 1993;
Caron et al., 2004).
The present-day Terraba belt represents the inverted Tertiary
forearc region of Costa Rica (Krawinkel and Krawinkel, 1996).
This basin underwent three stages of basin evolution: (1) an ini-
tial transpressional phase during Late Cretaceous–Eocene time,
which created the transpressional ridges that served as substrates
for carbonate-platform facies; (2) a transtensional phase during
Oligocene to early Miocene time, when the forearc region was
affected by transtensional fault systems and was subsiding rap-
idly, causing the carbonate platforms to drown; and (3) a sec-
ond transpressional phase of deformation from middle Miocene
to Holocene time. Thus, tectonic deformation across the Costa
Rican forearc basin initially created the substrates for carbonate
sedimentation but ultimately caused the platforms to drown.
Upper Oligocene to Miocene carbonate intervals exposed on
the island of Viti Levu, Fiji, and imaged on seismic profiles from
adjacent offshore areas consist of thin reefs or platform sequences
that are interstratified with volcaniclastic strata (Figs. 11, 12;
Hathway, 1994, 1995; Rodd, 1993). The Qalimare Limestone,
which crops out on Viti Levu, is at least 300 m thick and consists
of coral-algal reefs and mounded facies that formed on the edge
of a shallow platform in front of the arc massif. Seismic profiles
across the Bligh Water Basin from the northern offshore shelf
of Viti Levu have intrabasinal highs that served as substrates for
A 177°E 178° 179° 180°
0
1000
50

ORM

00
ATF

10
P L
EF
A RE 50
0
T SE
0

EVU
50
G REA L
UA
VAN 500
500 10
00
RM

00
17°S 10 500 WATER BASIN
FO
AT

H
PL

IG 150° 160° 170° 180°E

BL SOLOMON ISLANDS
A

500
AW

PAPUA 10°S
NEW GUINEA VANUATU
FIJI

00
50
S

10
0 CORAL SEA
YA

NEW
50 20°
0 CALEDONIA

AUSTRALIA
30°
VITI LEVU
REWA BAU TASMAN SEA
50
WATERS 0 40°
BASIN 1000km NEW
18° BASIN ZEALAND

00
0 50
100 0

10
10
00

50
BA 0 100 Pliocene sediments
0
R
AV
I Pliocene volcanics
BA
SIN
ZONE
Late Oligocene-Mid
N

500
SI

0 0 Miocene sediments
100
BA

50 & volcanics
VA

Late Eocene
SU

RE

sediments & volcanics

TU
AC

Basin outline
19°
FR

R Major fault
TE
0
50
N 0 20 40 km
100
0
HU

B
 Carbonate-platform facies in volcanic-arc settings 73

Figure 11. (A) Geologic map of Fiji. Most of the platform areas are shallow-water carbonate environments, although some platform areas are
presently at subphotic depths (i.e., drowned or incipiently drowned). Also note the steep, typically fault-bounded edges of most platform areas,
as indicated by solid black lines. The islands that make up Fiji represent the eroded and faulted remnants of an older volcanic arc–forearc basin
system. Bathymetry in meters. From Rodd (1993). (B) Geologic cross section across Fiji and adjacent platform areas. This section shows various
isolated platforms and buildups in the relict Bligh Water forearc basin. Note how faults controlled platform locations, dimensions, and morphol-
ogy. From Rodd (1993).

local late Oligocene to middle Miocene reef development within
the basin.
The Miocene Rethymnon Formation (<60 m thick) in the
Apostoli Basin, eastern Crete, consists of skeletal-rhodalgal lime-
stone facies that were deposited on a warm, temperate ramp that
developed along the arcward side of the Neogene forearc basin
in Crete. Skeletal and rhodalgal grain types in the Rethymnon
Formation are indicative of nontropical carbonate facies, but the
presence of large benthic foraminifers suggests warm, temperate
conditions (Pomoni-Papaioannou et al., 2003). The Rethymnon
Formation onlaps the underlying Pandanassa Formation (allu-
vial fan and fluvial floodplain deposits) and Apostoli Formation
(shoreface sandstone, Herostegina sands, and gray-marl facies),
suggesting general shallowing of the Apostoli Basin and possible
uplift along the arcward flank of the basin during deposition of
the Rethymnon Formation.
Seismic profiles across several late Oligocene–Holocene
sub-basins of the western Indonesian forearc show carbonate
buildups and shelflike profiles of Miocene age that formed on the
arcward side of these forearc depocenters (Beaudry and Moore,
1985; Matson and Moore, 1992). Most of the isolated buildups
appear as mounded features that likely formed on structural highs
during transgressive events. The shelflike carbonate profiles may
also record deposition during generally transgressive conditions.
Both the carbonate buildups and shelflike platform sequences are
overlain by progradational siliciclastic strata supplied from the
arcward side of the basin during regressive events.
Pliocene strata in the forearc basin on North Island, New
Zealand (Fig. 13), include coarse-grained, shallow-water, skel-
etal-sand sheets, which locally contain steeply dipping clinoform
packages (Kamp et al., 1988). These Pliocene skeletal sand bod-
ies formed as outer forearc basin deposits. The axes of the sand
bodies are oriented subparallel to faults and deformation fabrics
within the subduction complex farther east, which suggests that
the sand-body locations and orientations may have been con-
trolled by complex surface deformations that developed above
parts of the subduction complex that lie beneath the outer forearc
basin. The skeletal sand bodies were probably deposited during
active broad wavelength folding of the forearc basin, as indicated
by (Fig. 13): (1) successively older Pliocene limestone strata
having progressively steeper dips and greater elevations (>300 m
above sea level) along the eastern flank of the basin; (2) succes-
sively younger Pliocene carbonate strata, shifted laterally toward
the central axis of the forearc basin, especially along the more
uplifted eastern flank of the basin; (3) older Pliocene carbonate
sand bodies concentrated along the eastern and western flanks of
the forearc basin, grading laterally into siliciclastic facies in the
basin axis, whereas younger carbonate intervals are thinner and
more sheetlike in character; and (4) unconformities within the
Pliocene section that merge, becoming composite unconformi-
ties on the flanks of the forearc basin (Kamp and Nelson, 1987,
1988; Caron et al., 2004).
General tectono-stratigraphic model for carbonate facies
in forearc regions. Although detailed studies of carbonate strata
deposited across forearc regions are relatively rare, previous stud-
ies of the tectono-geomorphologic evolution of forearc regions
provide insight into the possible stratigraphic evolution of car-
bonate systems that might form there. The evolution of carbonate
platforms and reefs across the forearc region will be influenced
by (Fig. 6):

Sea level

Figure 12. Schematic tectonic and dep-

Isolated platform
on faulted high
ositional model for southwestern Viti
Levu during early to middle Miocene
arc uplift = increased time. View approximately toward pres-
sediment flux to
ent northeast. Note how fault-controlled
s if
forearc basin
s isolated platforms of the lower to mid-
c ma dle Miocene Qalimare Limestone were
Ar deposited on the frontal side of the vol-
canic arc, which was rising isostatically
because of crustal thickening. This in
turn caused more erosion of the arc and
greater flux of epiclastic sediment to the
forearc basin. From Hathway (1994).
74 Dorobek

Age N.Z.
(Ma) Stages LOCALITIES & LITHOSTRATIGRAPHY, HAWKE’S BAY, NEW ZEALAND

1.5 Limestone
400
Sandstone
2.0 Wn 180 Sandy mudstone
Covered
65
2.5 0 Unconformity
0
200
Wm
3.0 0
1000 200
Wp 0

3.5
0
35

4.0 Wo
0
0
0
4.5
15°SE Wn 5°SE 2°NW 8°-15°NW 18°-20°NW 10°NW
Wn Wo
Wm Wp
Wo Wm Tt
1 km Wp Wo
Forearc Basin Subduction Complex

Figure 13. Pliocene stratigraphy across the forearc basin of North Island, New Zealand. Measured stratigraphic sections at
specific locales across the forearc basin are shown in the upper part of this figure; stratigraphic thickness in meters. Note
the association of limestone with siliciclastic units and the progressive displacement of successively younger limestone
units toward the basin axis. The cross section does not show all the structures distributed across this forearc basin and
implies only broad, synformal folding over time. Most Pliocene limestone was deposited along the outer, trenchward side
of the basin, although younger carbonate units shifted progressively westward over time, toward the arc. Also note the
progressive decrease in dip of successively younger units (see lower cross-sectional view). These growth stratal patterns
reflect local surface deformation above fault-propagation folds and more regional differential subsidence/uplift across
the North Island forearc basin, which is expressed as the progressive westward shift of shallow-marine carbonate sedi-
mentation across the basin. New Zealand stage names and their relative ages include: Wn—Nukumaruan (late Pliocene-
Pleistocene); Wm—Mangapanian (late Pliocene); Wp—Waipipian (middle Pliocene); Wo—Opoitian (early Pliocene);
Tt—Tertiary. No vertical exaggeration. From Kamp and Nelson (1988).

1. Whether the subduction complex is accretionary or non-
accretionary. Accretionary subduction complexes grow
laterally as trench sediments are added to the frontal part
of the deforming wedge. Thus, accretionary subduction
complexes commonly widen and shallow over time and
provide ever-increasing potential surface areas for shal-
low-water carbonate sedimentation.
2. Differential uplift or subsidence across the forearc region.
For example, if a formerly submerged subduction com-
plex is uplifted and becomes emergent, any preexisting
carbonate platforms or reefs across the crest of the sub-
duction complex will become subaerially exposed, and
carbonate depositional environments will be forced to
shift laterally off the flanks of the subduction complex.
The total area available for shallow-water carbonate
sedimentation may actually decrease, especially if uplift
creates steep bathymetric gradients across the top of the
subduction complex. If the upper surface of the subduc-
tion complex is below photic depths at the beginning,
however, uplift may bring the surface into shallow-water
depths and allow a carbonate factory to develop. If the
top of the subduction complex is already emergent or
at shallow water depths, continued uplift of the subduc-
tion complex may produce progradational to downstep-
ping stratal patterns within carbonate platforms or fring-
ing reefs that cap the subduction complex. In contrast,
 Carbonate-platform facies in volcanic-arc settings
regional subsidence of the subduction complex (owing
to subduction erosion, flexure of the upper plate, or
other tectonic causes) might cause preexisting carbon-
ate platforms and reefs either to drown or to backstep
toward any remaining shallow-water parts of the subduc-
tion complex. Similar complex chronostratigraphic and
facies relationships may develop above other types of
actively growing intrabasinal structural highs within the
forearc basin or along the proximal arcward side of the
forearc basin.
3. Transcurrent faults that can partition differential subsid-
ence or uplift across the forearc region. Transcurrent faults
that cut the forearc region are most common where oblique
convergence occurs (e.g., Western Indonesian forearc).
Transcurrent faults can partition parts of the forearc into
antiformal highs or fault-bounded horst and graben seg-
ments, which control the location of platforms or isolated
reefs (cf. Krawinkel and Krawinkel, 1996).
Given the complex subsidence and uplift histories of
forearc regions, carbonate-platform and reef strata that form
across the forearc region during active subduction will likely
be thin (<100 m thick) and may be bounded by either subaerial-
exposure surfaces or drowning surfaces with complex chro-
nostratigraphic relationships. The thickness and compositions
of interstratified noncarbonate facies will be highly variable,
depending on:
1. The evolving bathymetry and patterns of differential
subsidence or uplift across the forearc region, which
may change rapidly over relatively short time scales. It
is possible for the forearc region to transition repeatedly
between phases of subsidence and uplift, and regions of
subsidence and uplift may be closely juxtaposed in space
and time. The amount of time that lapses between epi-
sodes of uplift or subsidence is another critical factor that
determines whether episodic deformation forces carbon-
ate facies tracts to relocate to other shallow-water parts of
the forearc region.
2. The flux of volcanogenic and siliciclastic sediment into
the forearc basin. Rapidly prograding mixed carbonate-
siliciclastic shelf facies may build from the volcanic arc
into a forearc basin during later stages of basin evolu-
tion when the volcanic arc has built significant relief,
the basin is filled with sediment, and the basin is slowly
subsiding or uplifting. Progradational siliciclastic facies
commonly will be supplied by a series of alluvial and
deltaic point sources, where drainage networks reflect
local climate and the irregular to ridgelike topography
of the volcanic arc. In humid settings, the arc may act
almost like a line source of sediment, with many local
drainages supplying volcaniclastic sediment to the
forearc basin, especially along windward-facing slopes
of high-elevation archipelagos. Shallow-marine carbon-
ate sedimentation may be suppressed where there is a
large flux of volcanogenic material (either as eruptive
75
products or as epiclastic sediment) to the forearc. Silici-
clastic deposits can serve as substrates for shallow-marine
carbonate sedimentation during transgressive events if
sediment supply is relatively low.
Shallow-marine carbonate strata may provide an excellent
record of even subtle patterns of differential subsidence or uplift
across the forearc region. In many cases, dip-trending transects
through mature forearc basins show stratal patterns that indicate
that the basins behaved more or less as broad, synformal dep-
ocenters, with smaller-scale, structural irregularities superim-
posed on this broader wavelength deformation. Across the area
of the trench-slope break, deformation of the underlying subduc-
tion complex typically creates a broad-wavelength (tens of kilo-
meters) uplift that is oriented subparallel to the trench axis. As
previously described, wedge accretion and related tectonic defor-
mation may bring the top of the subduction complex into shallow
water depths and allow a carbonate factory to develop, or continue
until the subduction complex becomes emergent and the area for
shallow-water carbonate sedimentation may actually decrease.
Uplift of the subduction complex may also force progradational
to downstepping stratal patterns within carbonate-platform strata
or fringing reefs that were deposited on the crest or flanks of the
shallow subduction complex. In contrast, regional subsidence of
the forearc region may cause carbonate platforms and reefs either
to drown, backstep, or shift to any remaining shallow-water parts
of the forearc region.
Smaller-scale seafloor irregularities across the trench-slope
break may be the surface expression of fault-propagation folds or
other surface deformations across the top of independently mov-
ing, fault-bounded slivers within the subduction complex (e.g.,
Pliocene forearc basin, North Island, New Zealand; Kamp and
Nelson, 1987, 1988; Kamp et al., 1988).
Reactivation of preexisting fault zones in forearc basement
rocks or syndepositional deformation of older forearc-basin
strata can also create broad wavelength folds that influence facies
patterns and growth stratal patterns within more axial parts of
the forearc basin. Intrabasinal forearc highs may (1) become sub-
strates for shallow-marine carbonate facies if they reach shoal-
water depths, (2) influence dispersal of sediment gravity flows
shed from adjacent shallow-water platform areas if the highs are
more deeply submerged, or (3) be eroded by marine or subaerial
processes as they are uplifted to progressively shallower depths
or possibly even above sea level. Thus, the broad-wavelength
seafloor characteristics of intrabasinal fold axes would have simi-
lar effects on sedimentation as surface deformations across the
trench-slope break.
Lastly, transcurrent faults can partition the forearc region
into separate fault-bounded depocenters with intervening highs.
Transcurrent faults may strongly partition differential subsidence
or uplift across the forearc region and are especially common
where oblique convergence occurs. Vertical offset on the trans-
current faults may create steep bathymetric gradients that control
the location of platform margins or isolated reefs on the interven-
ing highs between the faults.
76 Dorobek
Carbonate Platform and Reefs within Intra-Arc Basins
Soja (1996) reviewed island-arc carbonate sequences, and
although the tectonic setting of some ancient examples cited
therein is somewhat ambiguous, most probably formed within
intra-arc basins or were at least built on volcanic edifices. Island-
arc carbonate sequences may characteristically have endemic
faunas, because many volcanic arcs are biogeographically iso-
lated from more laterally extensive, land-attached carbonate
platforms, where faunal radiation is more likely to occur (Soja,
1992, 1996). It is not clear if a tendency toward endemism has
any effect on carbonate-facies types, stratal architecture, or
platform-reef morphology.
Volcanic arcs may also serve as “refuges” for reef-building
faunas when local conditions become stressed or during global
extinction events. For example, while most reef-building organ-
isms became extinct during Late Triassic time, a few coral species
seemed to have survived into Early Jurassic time in volcanic-arc
settings that were subsequently accreted to western North Amer-
ica (cf. Stanley and McRoberts, 1993).
Carbonate reefs and shallow-marine facies commonly fringe
volcanic edifices within many intra-arc basins, especially if the
volcanic arc is at the proper (paleo) latitude. Carbonate reefs and
associated facies may be more common on the windward side of
volcanic archipelagos because reef-building organisms prefer to
construct frameworks on the windward side of islands and land-
masses, and abundant ashfall deposition on the downwind side of
the arc may inhibit carbonate sedimentation. Thick, areally exten-
sive carbonate-platform sequences are apparently rare in intra-arc
basins because (1) these basins typically have steep bathymet-
ric profiles when they are underfilled, so there is little available
substrate for shallow-marine carbonate sedimentation; (2) there
is abundant volcanic activity that inhibits carbonate sedimenta-
tion; (3) there commonly are frequent phases of rapid subsidence
or uplift, during which time carbonate platforms either might be
drowned or terminated by subaerial exposure; and (4) frequent
seismic events may trigger massive failure along the flanks of
volcanic edifices, which can either incorporate and destroy large
stretches of preexisting platforms and reefs or bury shallow-
marine carbonate environments with mass-flow deposits.
The thickest and most areally extensive carbonate platforms
and reefal buildups within intra-arc basins probably form during
late stages of basin evolution or following the last phases of vol-
canic activity when (1) volcanic edifices are likely to be subsiding
rather than being uplifted or undergoing active volcanic construc-
tion; (2) the basins are more likely to be filled with volcaniclastic
sediment, which will provide additional shallow-water substrates
for carbonate accumulation; (3) volcanic topography has been
subdued by erosion so that a wider surface area is available for
carbonate sedimentation during sea-level highstands; and (4) vol-
canic eruptions and seismic events have ceased or are much less
frequent than during earlier stages of basin development.
Carbonate platforms and reefs can form during active phases
of arc volcanism as long as volcanic activity constructs features
that are submerged and within the photic zone, and the volume
of volcanogenic material dispersed to intra-arc basins is not great
enough to suppress carbonate production (cf. Bardintzeff et al.,
1985). Polymict breccias, however, dominate the stratigraphy in
many intra-arc basins and may provide the only record of past
carbonate sedimentation. These breccias can be the dominant
lithofacies in intra-arc basin successions that are >5 km thick and
consist largely of volcanic and volcaniclastic clasts, with lesser
amounts of sandstone, mudstone, and shallow-marine carbonate
facies. Platform- and reef-derived clasts (with dimensions up to
several hundred meters) within these sedimentary breccias may
provide the only evidence for former shallow-marine carbonate
systems in many intra-arc basins (cf. Soja, 1990).
Intra-Arc Basin Classification and Tectonic Models
Intra-arc basins are depocenters within volcanic arcs and
are underlain by arc-massif crust. Smith and Landis (1995) clas-
sified intra-arc basins as volcano-bounded, fault-bounded, or
hybrid types. Volcano-bounded intra-arc basins are depocenters
found between constructional volcanic edifices. Volcano-bounded
basins associated with continental-margin subduction systems
typically have only thin sediment fills, which Smith and Landis
(1995) suggested are due to the generally high-standing charac-
ter and low subsidence rates of their continental-crust substrates.
In contrast, volcano-bounded basins from intraoceanic settings
may accumulate thick volcanogenic strata, possibly because they
subside more rapidly or begin as deep-water depocenters on oce-
anic crust. Erosion rates, however, are generally much higher in
continental-margin arc systems than in intraoceanic arc systems,
because continental-margin arc systems are largely subaerial fea-
tures, whereas volcanic highs and intra-arc basins may be largely
submerged in intraoceanic arc systems. Thus, erosion of volcanic
highs also affects the amount of sediment delivered to intra-arc
basins and the rate at which they are filled.
Fault-bounded intra-arc basins are defined by fault networks.
Fault patterns within most volcanic arcs and first-motion stud-
ies of intra-arc earthquakes indicate that extension or strike-slip
deformation causes most fault-bounded intra-arc basins to form.
Fault-bounded intra-arc basins typically have bounding faults that
trend parallel to the volcanic arc, although some basins and their
main bounding fault systems trend transverse to the volcanic arc.
Fault-bounded intra-arc basins also may subside more rapidly
than volcano-bounded depocenters (Smith and Landis, 1995).
Transverse oblique-slip fault systems that delineate some
intra-arc basins may accommodate block rotations along the arc
axis, such as in the Aleutian forearc region (Geist et al., 1988).
The curved trend of the Aleutian arc system, the oblique subduc-
tion that occurs along this convergent margin, and the strong cou-
pling between the subducting and overriding plates may explain
the block rotations. Differential subsidence of narrow crustal
blocks between transverse faults may create submarine channels
that funnel sediment gravity flows to the forearc basin or trench
slope. Arc-parallel strike-slip faults can also accommodate sig-
nificant lateral displacement of arc segments on either side of the
 Carbonate-platform facies in volcanic-arc settings
fault zone (e.g., Semangko or Sumatra Fault, Beck, 1983; Philip-
pine Fault, Karig et al., 1986; Sarewitz and Lewis, 1991).
Hybrid intra-arc basins have elements of both volcano- and
fault-bounded basins. Hybrid basins are defined by Smith and
Landis (1995) as depocenters with fault-defined margins, but
where most of the topographic or bathymetric relief is defined
by constructional volcanic features. Arc massifs and associated
intra-arc basins typically undergo complex deformation histories
and differential subsidence as a consequence of changes in sub-
duction parameters (dip angle, obliquity, and rate of subduction),
dynamic mantle flow, changes in surface loads, and magmatic
activity. These large-scale tectonic and magmatic processes and
resultant patterns of uplift and subsidence within the volcanic arc
change over time scales on the order of 5–20 m.y. (Smith and
Landis, 1995). As a result of the constantly changing tectonic
and magmatic conditions that characterize the long-term history
of most volcanic arcs, intra-arc basins may transition rapidly
between volcano-bounded, fault-bounded, and hybrid types.
All styles of deformation have been recognized within volca-
nic arcs and their related intra-arc basins. Extensional and strike-
slip structures are generally more common within the arc massif
than contractional structures, unless arc-continent collision has
occurred. The history of deformation within intra-arc settings is
poorly preserved because it is obscured and overprinted by mul-
tiple eruptive and deformation events, thick sediment accumu-
lations may deeply bury and obscure fault networks, or the arc
and its related basins are eventually incorporated into collisional
orogenic belts and original tectono-stratigraphic relationships
are destroyed. Many arc systems also record flips in subduction
polarity such that former forearc regions may become backarc or
interarc settings, or vice versa. Preexisting faults within the arc
massif also may be reactivated by a different sense of displace-
ment during subduction-polarity flips or other changes in sub-
duction characteristics. These complicated tectonic histories may
make it difficult to characterize the geometry, distribution, and
kinematics of fault networks that develop within a volcanic arc.
Extensional deformation is common where trench rollback
causes the volcanic arc to split apart. An active seafloor-spreading
ridge may form if the arc is completely stretched apart, leaving a
remnant arc on the trailing side of the spreading axis and an active
arc on the other side. In these cases, intra-arc rift basins evolve
into an extensional backarc setting. Intra-arc rifting may occur
faster than in intracontinental rift settings, with correspondingly
much faster rates of synrift subsidence (Yamaji, 1990). Some vol-
canic arcs, such as the Marianas arc system, have been repeatedly
split by intra-arc rifting and seafloor spreading (Hamilton, 1979).
Other arcs have been split apart during a flip in subduction polar-
ity; changes in subduction polarity may not occur synchronously
along all segments of the convergent margin. These factors may
create problems with basin classification and understanding of
the tectonic processes that cause complex spatial and temporal
patterns of uplift and subsidence. Examples of Cenozoic volca-
nic arcs that have been rifted apart during changes in subduction
polarity include the Solomon Islands (Johnson, 1979; Kroenke,
77
1984; Bruns et al., 1989) and Vanuatu archipelagos (Carney et al.,
1985; Greene et al., 1988).
The fault-bounded basement highs that delineate intra-arc
basins are parallel or oblique to arc segments. These structural
highs typically serve as (1) minor sediment source areas if they
are subaerially exposed, (2) substrates for shallow-marine sedi-
mentation if they are only slightly submerged, or (3) bathymetric
obstructions to sediment gravity flows if the structural highs are
more deeply submerged. Although seismic data are limited from
recently rifted arcs like the Solomon Islands and Vanuatu, the
arc-parallel or arc-crossing highs that separate intra-arc depocen-
ters within these arcs probably represent oblique transfer zones.
Oblique dip-slip displacement on the faults that bound the intra-
arc highs allows extensional strain to be transferred between nor-
mal fault systems within the arc. These intra-arc fault-bounded
highs may influence sedimentation within the arc for a long time
after they form, especially if extension stops before the arc is
completely rifted apart.
Transtensional, transpressional, and pure strike-slip styles of
deformation within the arc and intra-arc basins are more likely to
develop where there is oblique subduction. As the convergence
direction becomes more oblique or subparallel to the trend of
the subduction system, pure strike-slip fault networks should
become more common. Along some modern arc systems with
highly oblique subduction angles, regional strike-slip faults may
cut through and translate long segments of the volcanic arc (e.g.,
Sumatran or Semagang Fault along the island of Sumatra, west-
ern Indonesia; Philippine Fault, Philippine archipelago).
Regional shortening within most volcanic arcs typically
does not develop until arc-continent or continent-continent colli-
sion occurs and magmatic activity within the arc has long ceased.
The limited amount of shortening within many volcanic arcs (and
backarc regions) may indicate that mechanical coupling between
the overriding and subducting plates diminishes with increasing
distance from the trench (cf. Uyeda and Kanamori, 1979).
Platform Types
Most carbonate sequences within intra-arc basins consist of
either fringing reefs constructed on volcanic or plutonic basement
or barrier reefs with relatively narrow (<10 km wide) back-reef
and lagoonal facies. These fringing reefs and narrow platforms
can build relatively steep depositional profiles, as indicated by
mountainside exposures of ancient examples and bathymet-
ric profiles across modern intra-arc basins. Fringing reefs also
may form within collapsed calderas, although these probably are
minor, short-lived accumulations that are terminated or blasted
away by subsequent eruptions.
Carbonate platforms and reefs also form in fault-bounded
intra-arc basins where fault patterns and displacement histories
strongly influence platform dimensions, the location of platform-
margin facies tracts, and platform morphology. Small, isolated
platforms and pinnacle reefs are found on fault-bounded highs
within some intra-arc basins of Cenozoic island arcs in the South
Pacific. Local fault-bounded basement highs that border some of
78 Dorobek
these intra-arc basins serve as substrates for larger isolated plat-
forms. For example, Woodhall (1985) suggested that conjugate-
fault patterns controlled the locations of volcanic vents and Qua-
ternary reefs on the Exploring Isles of the northern Lau Ridge.
Other platform types are rare. Ramp-like to nonrimmed
platforms and scattered mounded buildups may develop on the
leeward sides of volcano-bounded intra-arc basins. Ramps may
still have relatively steep depositional gradients owing to the
steepness of the underlying, volcanically constructed anteced-
ent topography. Leeward carbonate ramps may be relatively thin
in the rock record because volcanic ash from active volcanoes
preferentially accumulates in leeward intra-arc depocenters and
suppresses or terminates carbonate sedimentation (cf. Eldredge
and Kropp, 1985).
Intra-arc platforms and reefs exhibit complex patterns of
stratigraphic development, because intra-arc basins are character-
ized by complex spatial and temporal patterns of differential sub-
sidence and uplift during their evolution. During active tectonic
deformation, platforms and reefs may rapidly expand or shrink
over short time scales (<100,000 yr), especially during icehouse
times when high-amplitude glacio-eustatic sea-level changes are
more likely to occur (cf. Read, 1995). Intra-arc fault systems
commonly are steeply dipping and develop significant surface
scarps (100 to >1000 m), and volcanic edifices also commonly
have dramatic bathymetric gradients, both of which limit the area
over which carbonate facies tracts can shift laterally during rela-
tive sea-level changes. During late stages of basin development,
when intra-arc depocenters are more likely to be filled, carbon-
ate platforms and reefs may utilize volcanogenic successions as
substrates and develop into more areally extensive systems. Dif-
ferential subsidence and uplift across arc terranes, however, can
still be significant during later stages of basin evolution and can
cause complex stratigraphic development within laterally exten-
sive carbonate platforms.
Quaternary examples. Quaternary intra-arc carbonate
platforms and fringing reefs are found in the Solomon Islands,
Fiji, and the Kadavu island group (Nunn and Omura, 1999;
Nunn, 2000); Vanuatu (Macfarlane et al., 1988); Lau Ridge
(Woodhall, 1985); Tonga Ridge (Cunningham and Anscombe,
1985); the Indonesian archipelago (Sumatra, Java, Bali, islands;
Wells, 1988); the submerged Grenadine Bank and other, locally
drowned pinnacles and isolated buildups on volcanic substrates
elsewhere in the Lesser Antilles (D’Anglejan and Mount-
joy, 1973; Bouysse, 1984; Dey and Smith, 1989); the Ryukyu
Islands of southwest Japan (Iryu et al., 1995; Nakamori et al.,
1995; Sagawa et al., 2001); and various parts of the Philippines.
In many cases, Quaternary reefs and narrow fringing carbonate
platforms have been uplifted to form a series of notched terraces
that are exposed near the modern coastlines of these islands.
These uplifted Quaternary limestone terraces attest to the rapid
uplift rates (average rates up to 1 cm/yr) that characterize many
modern volcanic-arc settings.
Many Quaternary reefs or shallow-marine carbonate facies
have developed in successor intra-arc basins (e.g., Quaternary
carbonate facies from the Grenadine Bank and some northwest-
ern islands, Lesser Antilles, D’Anglejan and Mountjoy, 1973;
Bouysse, 1984; Dey and Smith, 1989; Quaternary Ucuna Lime-
stone of the Neogene Takalau Limestone Group, Lau Ridge,
Woodhall, 1985; Quaternary carbonate units from islands of the
Western and Eastern Belts of Vanuatu, Macfarlane et al., 1988).
In a successor intra-arc basin the volcanic arc has been dormant
or extinct for a long time, yet regional topography-bathymetry is
still strongly influenced by the inherited tectonic-volcanic geo-
morphology. In some cases, carbonate facies may be deposited
within an arc segment where the volcanoes are extinct or dor-
mant, even though other parts of the volcanic archipelago may be
active. Long periods of geologic time (10 m.y. or more) may also
pass between major eruptive episodes.
This is the case in the Grenadine Islands, Lesser Antilles,
where intra-arc carbonate facies form a submerged and incipi-
ently drowned carbonate platform (the Grenadine Bank) that is
currently at 20–40 m water depths. The Grenadine Bank prob-
ably formed during a Pleistocene lowstand of sea level on eroded
remnants of the Lesser Antilles volcanic arc (D’Anglejan and
Mountjoy, 1973; Dey and Smith, 1989). Active volcanoes are
found within the same archipelago on the islands of Saint Vincent
and Grenada, which are to the north and south, respectively, of
the Grenadine Bank. Given that the Grenadine Bank is the same
general distance from the subduction trench as are the active vol-
canoes on Saint Vincent and Grenada, renewed volcanic activity
should be expected at some future time there. The lack of recent
volcanic activity within the Grenadine Islands, however, and the
prevailing easterly wind allow extensive modern carbonate sedi-
mentation to occur on the eastern part of the Grenadine Bank
(D’Anglejan and Mountjoy, 1973; Dey and Smith, 1989).
The Kallinago Depression in the northern Lesser Antilles
(Fig. 14) represents a volcano-bounded intra-arc basin that formed
when volcanism shifted from an outer (trenchward) Eocene
to early Miocene arc to an inner late Miocene to Holocene arc
farther west. The Kallinago Depression diminishes and merges
southward into a single arc near the island of Guadeloupe. Iso-
lated, early(?) Pleistocene buildups (e.g., Luymes Bank; Fig. 14)
have formed on the steep, volcanic substrates surrounding this
relatively young active arc.
The Lau Ridge is another example where Quaternary car-
bonate facies are accumulating in a remnant intra-arc setting.
The Lau Ridge represents the remains of a late Oligocene–early
Miocene volcanic arc that was left behind as a trailing, extinct arc
when the Lau Basin opened by seafloor spreading and separated
the Tonga Ridge from the Lau Ridge. Emergent to shallowly sub-
merged parts of extinct volcanic centers that compose the Lau
Ridge continue to serve as substrates for Holocene carbonate
sedimentation (mostly skeletal sand facies; Woodhall, 1985).
In the Vanuatu (New Hebrides) arc, Oligocene–Miocene
volcanic and volcaniclastic rocks of the Western Belt of the
archipelago represent products from a volcanic arc that was
rifted apart during middle Miocene (ca. 14 Ma) time. Renewed
arc volcanism jumped eastward to form the mostly late Miocene
 Carbonate-platform facies in volcanic-arc settings 79

A’ A A’
Luymes
1000 Anguilla Anguilla 0
3000 Bank

400
2000
1000 Shelf

0
20
Ka
0
llin Barbuda
A ag
o

20
0

0
St. Kitts
Kallinago

500
De
Depression 1
pr
Eocene-

TWT (sec)
es
Early Miocene Arc
si
on
Montserrat
0
100

Guadeloupe 2
20
00
Late Miocene-
Holocene Arc

lt?
f au
1000
200
St. Lucia 0
0 10 20 km
0 50 100 km 3
Contours in meters

Martinique

Figure 14. Map of northern Lesser Antilles and Kallinago Depression and interpreted seismic section. The Kallinago
Depression is a volcano-bounded intra-arc basin that formed between the inactive Eocene–early Miocene arc and younger
late Miocene–Holocene arc of the Lesser Antilles. Note how volcanic axes converge southward to a single volcanic arc
(i.e., Martinique). Interpreted seismic profile (location of section A–A′ shown in upper left of base map), showing slightly
deformed sedimentary fill within the Kallinago Depression (profile from Bouysse, 1984). Luymes Bank is a drowned
carbonate buildup whose top is at ~150–200 m water depth at present. Bathymetry in meters. Modified from Smith and
Landis (1995).

to late Pliocene Eastern Belt of the Vanuatu arc. Continued rift-
ing between the two belts ultimately led to the youngest volcanic
centers within the Central Belt, which formed volcanically con-
structed and fault-bounded highs during late Pliocene to Holocene
time (Macfarlane et al., 1988). Pliocene to Holocene carbonate
facies form fringing reef complexes around extinct volcanoes of
the Western and Eastern Belts and around active volcanoes of the
Central Belt. Subduction polarity, however, apparently changed
during Miocene time from west-dipping to east-dipping subduc-
tion, and the subduction zone switched from the eastern to the
western side of the arc. Given the complex tectonic history of the
New Hebrides arc, it is difficult to classify some of the depocen-
ters as either intra-arc or forearc basins because many are in a
protracted, transitional phase of development.
Ancient examples. Ancient examples of intra-arc carbonate
platforms, small banks, and reefs include (1) the Silurian Heceta
Formation, Alexander Terrane, southeastern Alaska (Soja, 1990);
(2) the Middle Devonian Kennett Formation, Lower Carbonifer-
ous Bragdon and Baird Formations, and Lower Permian McCloud
Limestone, Eastern Klamath Mountains, California (Demirmen
and Harbaugh, 1965; Watkins, 1985, 1993a, b; Watkins and
Flory, 1986); (3) Upper Cretaceous lenticular reef bodies in the
Clifton Member of the Mount Peace Formation, Lucea Inlier,
western Jamaica (Grippi, 1980); (4) the Oligocene(?) Paumbapa
Formation on Sumba Island and offshore, Indonesia (Fortuin et
al., 1997; Rutherford et al., 2001); (5) local mounded buildups
that are scattered throughout the Tertiary section on islands and
offshore structures of the Tonga Ridge (Cunningham and Ans-
combe, 1985; Herzer and Exon, 1985); (6) Miocene to Pliocene
reefal limestone and carbonate-platform facies (Tokalau Lime-
stone Group) on various islands of the Lau Ridge (Woodhall,
1985); (7) lower to middle Miocene skeletal limestones (Tari
Formation and Qalimare Limestone) on Viti Levu, Fiji (Hath-
way, 1994, 1995); and (8) the Miocene Lelet Limestone, which
was deposited on the eroded remnants of the New Ireland vol-
canic arc, Papua New Guinea (Stewart and Sandy, 1988). These
intra-arc carbonate facies typically are complexly interstratified
with pyroclastic sands and tuffs, volcaniclastic breccias, coarse-
grained pebbly sands, siltstone, mudstone, and lava flows, and
also are commonly cut by steeply dipping syn- and postdeposi-
tional faults.
Other possible ancient examples of intra-arc carbonate
sequences include a Lower Jurassic coral reef in the Telkwa
Range, British Columbia (Poulton, 1989; Stanley and McRoberts,
80 Dorobek
1993); thin, Lower Jurassic coral biostromes in the Chocolate For-
mation, Arequipa massif, Peru (Wells, 1953; Stanley and McRob-
erts, 1993); Middle Jurassic coral biostromes from the Wallowa
Terrane, west-central Idaho, USA (Stanley and Beauvais, 1990);
and Upper Triassic carbonate units (including coral reefs) of the
Martin Bridge Formation, Wallowa Mountains, Oregon, USA
(Stanley, 1982; Stanley and Senowbari-Daryan, 1986), and at
Lime Peak, southern Yukon Territory, Canada (Reid and Ginsburg,
1986). Although the Mesozoic examples from North America are
typically described as “island arc carbonates,” their specific paleo-
geographic settings remain somewhat ambiguous.
Unlike Quaternary intra-arc carbonate platforms and reefs,
for which only limited subsurface data are available, ancient
sequences provide better constraints on the thickness and internal
facies of intra-arc carbonate platforms and reefs. Most intra-arc
carbonate sequences are several tens of meters thick, although
some are several hundred meters thick. Shallow-marine, muddy
carbonate facies of the Miocene Lelet Limestone on New Ireland,
northeastern Papua New Guinea, may be an exception, as they
reportedly are up to 1 km thick, with offshore equivalents of the
Lelet Limestone ranging from 1 to 2 km thick (Exon et al., 1986;
Stewart and Sandy, 1988). The lateral extent of ancient intra-arc
carbonate facies is usually much more difficult to constrain than
for Quaternary examples, because ancient intra-arc basins may
be subjected to multiple phases of postdepositional deformation.
General Tectono-Stratigraphic Model for Intra-Arc
Carbonate Platforms and Reefs
The different types of intra-arc basins influence the types of
carbonate platforms and reefs that form in them. In general, the
rapid rates of tectonic deformation in intra-arc settings, the likeli-
hood of catastrophic eruptions, and the close proximity of high-
relief source areas for siliciclastic sediment (i.e., the volcanic arc)
cause carbonate sedimentation to be only intermittent and short-
lived within intra-arc basins.
Fringing reefs and narrow platforms are the likely domi-
nant type of carbonate systems found in most intra-arc settings
(Fig. 15). These rarely attain significant thickness (i.e., >100 m
thick) before a catastrophic eruption or mass-wasting event
buries the carbonate system. Mass-wasting events also might
incorporate carbonate strata, causing large-scale disruption or
transport of shallow-marine carbonate facies into deeper-water
settings. Displaced blocks of reef and shallow-water carbonate
facies within deep-water shale and volcaniclastic deposits may
provide the only evidence of the former intra-arc platforms.
There may be differences between intra-arc carbonate suc-
cessions from intraoceanic and continental-margin arc systems.
In intraoceanic arcs, shallow-marine carbonate facies will likely
not appear within intra-arc basins until the volcanic arc has built
itself into the photic zone. Volcano-bounded and hybrid intra-arc
basins may dominate over fault-bounded basins in young intra-
oceanic volcanic arcs. The basal part of the basin-filling sequence
in these basins will likely consist of deep-water volcaniclastic
and pelagic-hemipelagic deposits. After volcanic activity has
raised the arc substrate into the photic zone, thin carbonate pack-
ages may form rings around the volcanic substrates.
With continued arc development (in either intraoceanic or
continental-margin arc systems), intra-arc basins may evolve
into fault-bounded depocenters, where faults then control the
areal dimensions and locations of carbonate platforms and reefs
as well as the location of facies tracts and the overall morphol-
ogy of platform margin-to-basin transitions. Carbonate platforms
will likely have steep, fault-controlled margins in profile and
have linear segments in plan view, both of which would reflect
the influence of underlying fault zones. The trend of many lin-
ear platform-margin segments within fault-bounded intra-arc
basins might be predictable if the underlying fault zones develop
with consistent orientations or as conjugate fault sets that reflect
regional stress orientations.
As tectonic and volcanic activity wane during later stages
of basin development, and intra-arc basins progressively fill
with sediment (largely volcaniclastic), carbonate platforms may
become thicker and more areally extensive. Remnant volcanic-
and fault-controlled topography, however, may continue to influ-
ence the dimensions of platforms and the locations of platform-
margin-to-basin facies transitions until enough basin-filling sedi-
ment has filled the antecedent topography.
Carbonate Platforms and Reefs in Backarc Basins
Carbonate sedimentation also occurs in backarc settings on
a variety of tectonic structures and bathymetric highs constructed
by depositional processes. As in other arc depocenters, the areal
extent, thickness, morphology, and internal stratigraphy of car-
bonate platforms and reefs will be determined by the tectonic
history of the backarc and whether it is filled with sediment.
Backarc-Basin Types and Tectonic Models
Backarc basins form on the back side of subduction-related
volcanic arcs (Fig. 16). Extensive geophysical surveys of many
modern backarc regions over the last 20 yr or so have led to
the recognition of three main types of backarc basins (Dick-
inson, 1974; Karig, 1983; Ingersoll, 1988; Marsaglia, 1995):
(1) extensional backarc basins that form by rifting and seafloor
spreading within or behind continental-margin or intraoceanic
arcs, (2) remnant-ocean backarc basins that form by entrap-
ment of old oceanic crust and are associated with intraoceanic
subduction zones, and (3) compressional backarc basins, which
are more commonly classified as retro-arc foreland basins (see
discussion below). Marine backarc basins may be underlain
by continental to transitional crust if they form by extension
behind a continental-margin arc system. In contrast, backarc
basins will be underlain by highly subsided oceanic crust if the
basin forms along an intraoceanic subduction zone, and rem-
nant oceanic crust is trapped behind the newly formed volca-
nic arc. Backarc basins also can be underlain by young oceanic
crust if they form where rifting in the backarc region ultimately
leads to seafloor spreading.
 Carbonate-platform facies in volcanic-arc settings
VOLCANO-BOUNDED INTRA-ARC BASINS
platform types:
• isolated platforms & buildups, commonly
steep-sided
• thin fringing platforms & reefs during early
stages; complexly interstratified with
volcanics, volcaniclastics
• may be preferential development on
windward side of arc
• complex chronostratigraphic relationships
between individual carbonate sequences
• may be drowned platforms adjacent to
actively growing platforms
• platforms become larger & thicker as
depocenters fill, but eruptions can have
catastrophic effects on platforms
substrates:
• volcanic edifices & volcaniclastic fill
Volcano-bounded
intra-arc basins
Figure 15. Types of carbonate platforms and buildups within intra-arc basins. Platform thicknesses, widths, and relief are
not to scale. Only generalized platform types and buildups are shown. Isolated platforms and fringing platforms-reefs
are most common. Stratigraphic relationships and depositional contacts with structural features and various geomorpho-
logic surfaces across intra-arc basins are also highly generalized. Modified from Smith and Landis (1995) to emphasize
carbonate-platform facies.
Extensional backarc basins are commonly associated with
subduction zones where trench rollback occurs. Many exten-
sional backarc basins actually begin as fault-bounded intra-arc
basins, when a volcanic arc begins to rift apart. An extensional
backarc basin might come into existence when there is the
first evidence for a topographically expressed, but volcanically
extinct, remnant arc that is separated from an active volcanic
arc by a rift zone. There likely will be a distinct bathymetric
or topographic axis to the newly formed extensional backarc
basin at this point in its tectonic evolution. This definition for
an extensional backarc basin does not depend on the amount of
stretching across the rift zone or on whether rifting has ceased
or seafloor spreading has begun. The remnant arc also can be
both a source area for siliciclastic sediment and a substrate for
shallow-marine carbonate sedimentation, although carbonate
facies may be deposited only during early stages of the basin’s
history, before the remnant arc has undergone significant ther-
mal subsidence.
Backarc basins that are underlain by trapped fragments of
older oceanic crust have different tectonic and subsidence histo-
ries than those that form by backarc extension. Remnant-ocean
backarc basins are associated with intraoceanic subduction
zones, and their dimensions depend on where subduction is initi-
ated within the overriding plate and the nearest bathymetric high
81
FAULT-BOUNDED INTRA-ARC BASINS
platform types:
• isolated platforms & buildups
• similar controls & relationships as for
volcano-bounded basins (windward side,
complex chronostratigraphy, etc.)
• extensional deformation common; post-rift
thermal subsidence will affect
stratigraphic development
• internal growth strata possible if platforms
on actively deforming fault-bounded highs
substrates:
• volcanics, volcaniclastics, faulted arc
basement or volcanics/volcaniclastics
• platforms become larger & thicker as
depocenters fill or as deformation wanes
Fault-bounded
intra-arc basins
(commonly a continental margin) that defines the distal side of
the basin (see below).
Backarc regions typically are areas of overall crustal exten-
sion or strike-slip deformation. Crustal shortening is rare in most
backarc regions except for some continental-margin arc systems
where the subducting and overriding plates are strongly coupled
(e.g., segments of the Andean and western Indonesian backarc
regions and the Sea of Japan backarc region). In cases where sig-
nificant crustal shortening occurs in the backarc region of con-
tinental-margin arc systems, the flexural basins that develop are
more appropriately described as retro-arc foreland basins, which
are not considered further in this paper.
The styles and intensity of deformation in the backarc
region depend largely on (1) the tectono-evolutionary stage (or
maturity) of the arc system; (2) the relative motions (e.g., angle
of convergence) between the subducting and overriding plates;
(3) the age, thickness, dip, and crustal type of the subducting slab;
(4) possible changes in subduction polarity during convergence;
(5) whether trench rollback is possible; (6) whether a spreading
ridge, seamount, or other large-scale seafloor features have been
subducted; or (7) whether arc-continent or continent-continent
collision has begun.
Extensional backarc regions are dominated by normal-fault
networks that are subparallel or slightly oblique to the trend of
82 Dorobek
old oceanic crust
trapped by intraplate
subduction zone
A tal-margin volcanic arcs include the Andaman Sea, the Okinawa
0 100 km
new oceanic crust
created by
back-arc spreading
B structural features as the Sumisu rift zone (Letouzey and Kimura,
0 100 km
continental crust beneath
retro-arc foreland basin
(compressional arc system)
C arc; examples include the Andean and western Indonesian “back-
0 200 km
rifted continental crust beneath
extensional back-arc basin
D
0 200 km
Figure 16. Cross sections through different types of convergent mar-
gins, showing crustal types of backarc regions. (A) Intraoceanic volca-
nic arc, with old oceanic crust trapped in backarc region. (B) Backarc
region underlain by newly created oceanic crust created by spreading
axis. (C) Compressional backarc region (retro-arc foreland basin sys-
tem) underlain by continental crust. (D) Extensional backarc region
underlain by rifted continental crust.
the volcanic arc. As in other extensional settings, oblique-slip
transfer faults trend at high angles to basin-bounding normal
faults and accommodate displacement transfer between offset or
nonparallel normal faults. This style of extensional deformation
is characteristic of the 700-km-long Izu-Bonin intraoceanic arc
system, which is considered to be an example of an extensional
arc system where an incipient backarc basin(s) is forming (Taylor
et al., 1990, 1991; Klaus et al., 1992). The extensional sub-basins
along the Isu-Bonin arc are typically asymmetric, 25–110-km-
long and 25–40-km-wide features. Most basin-bounding normal
faults trend subparallel to the axis of the arc system. Similar to
continental rift systems, oblique transfer faults and accommo-
dation zones link oppositely dipping normal faults that bound
the rift depocenters. Submarine volcanic centers appear to be
concentrated along bends in the rift system that are interpreted
as accommodation zones.
Modern examples of backarc extension behind continen-
Trough, and the Japan Sea. Highly oblique subduction from west-
ernmost Indonesia northward to the continental margin of Myan-
mar, strong coupling between the subducting Indian-Australian
plate and overriding Eurasian plate, and subduction of an aseis-
mic ridge may explain the backarc rifting and seafloor spreading
in the Andaman Sea (Eguchi et al., 1980). The Okinawa Trough
is an incipient continental backarc basin and displays similar
1986). The Japan Sea is a fully developed continental-margin
backarc basin, where contraction or transpression has occurred
along its eastern side, and subduction may be in its beginning
stages (Kikuchi et al., 1991; Tamaki, 1995; Okamura et al., 1995;
Takano, 2002).
Retro-arc foreland basins are flexural depocenters that result
from crustal shortening on the continental side of the volcanic
arc” settings. In comparison, crustal shortening within backarc
basins (sensu stricto) along intraoceanic convergent margins is
apparently rare. The limited amount of shortening in the backarc
region of most intraoceanic volcanic arcs probably reflects weak
coupling between the subducting and overriding plates along
these types of convergent margins.
Platform Types
The lateral extent and cumulative thickness of carbonate
facies within backarc basins will vary, depending on the tectonic
origin(s) and evolutionary stage of the basin (Fig. 17). For back-
arc basins that form by entrapment of old oceanic crust behind
the arc, carbonate facies may be entirely limited to shallow sub-
merged areas on the backarc side of the arc massif, and there
may be no other shallow-water substrates for carbonate platforms
across the backarc region. Most emergent or shallow-water sub-
strates in these backarc basins will be constructed by volcanic
activity. Continued growth of carbonate platforms and reefs in
these backarc settings may be frequently interrupted by volcanic
eruptions or mass-wasting events so that reefs and other carbon-
ate facies must be reestablished after a catastrophic event destroys
or buries them (Eldredge and Kropp, 1985).
In contrast, extensional backarc basins will have bathymet-
ric profiles that reflect stretching of the arc massif or other types
of backarc lithosphere. During initial rifting of the volcanic arc,
carbonate sequences may be deposited in intra-arc basins (see
above). Where stretching continues to the point of whole litho-
sphere failure and a seafloor-spreading ridge develops, a backarc
basin is clearly identifiable, and trailing rifted margins develop
on either side of the basin. There is no consensus, however, as to
when an intra-arc basin structurally, bathymetrically, or sedimen-
tologically transitions into a backarc basin.
Apparently there are no preserved examples of thick, long-
lived carbonate platforms within backarc basins. Backarc basins
 INTRA-OCEANIC BACKARC SETTINGS COMPRESSIONAL BACKARC SETTINGS
A pinnacle reef or B smaller isolated ramp sequence ramp sequence
outer reef tract platforms, shoals, and reefs constructed on older constructed on foreland
shelf/ramp sequence
may be on antiformal highs above foreland basin strata side of basin
constructed on older
fault-controlled back-arc thrust belt (commonly post-orogenic)
volcaniclastic strata

TRAPPED OCEANIC CRUST

RETRO-ARC FORELAND BASIN
0 50 km 0 200 km

more regional ramp/shelf sequences

during post-rift stages; flexural

EXTENSIONAL BACKARC SETTINGS: C may contain syn-rift platforms

on fault-bounded highs
onlap and post-rift thermal
subsidence; substrates include
C. Backarc intracontinental rifts if depocenters are marine post-rift fill and continental basement
D. Backarc intracontinental rifts leading
to spreading and conjugate
passive margins
E. Intra-arc rifting leading to spreading
0 200 km
EXTENSIONAL BACK-ARC BASINS

may contain syn-rift platforms

on fault-bounded highs thermally subsiding may contain syn-rift platforms
if depocenters are marine, conjugate passive margin;
D but more extensive ramps/shelves
form during post-rift phase
similar controls on regional platform E on fault-bounded highs
if depocenters are marine,
thermally subsiding remnant
volcanic arc provides substrate
development as passive margins but more extensive ramps/shelves for short-lived isolated
during post-rift phase platforms and reefs

BACK-ARC SPREADING RIDGE

BACK-ARC SPREADING RIDGE
0 100 km
0 100 km

Figure 17. Different types of backarc settings and regional carbonate platforms that may form in them. In all examples shown, the subduction
zone and forearc region (not shown) are left of the volcanic arc (indicated by triangle). Horizontal scales are generalized; no vertical scale implied.
Only general patterns of platform development are shown; internal stratigraphy and details about platform morphology will vary significantly in
each backarc setting. (A) Intraoceanic volcanic arc, with old oceanic crust trapped in backarc region. Carbonate platforms and reefs can form on
the backarc side of intraoceanic arcs. In general, substrates for these backarc carbonate systems will be constructional volcanic edifices, eroded
arc-massif basement, or volcanic-volcaniclastic deposits. Local fault-controlled bathymetry may control platform dimensions and facies patterns.
(B) Retro-arc foreland-basin system. Extensive carbonate sedimentation can occur in retro-arc foreland basins if they are slightly underfilled
and the continental foreland is below sea level. Thus, carbonate sedimentation is probably more common during underfilled stages of retro-arc
foreland-basin development. Carbonate ramps are common, especially along the distal (foreland) side of the basin. Various styles and intensities
of deformation across the foreland can significantly affect carbonate sedimentation. Foreland-basin carbonate systems are discussed in more
detail in later sections. (C) Extensional backarc region underlain by rifted continental crust. If rifted continental basement is flooded early, synrift
carbonate platforms may form on fault-bounded basement highs. More extensive platforms may develop during late post-rift stages, when the
extensional backarc basin is nearly filled and thermal subsidence dominates. (D) Backarc region underlain by newly created oceanic crust created
by spreading axis. Where stretching in the backarc region progresses to the point of whole lithosphere failure, a seafloor spreading axis will de-
velop. If stretching began within an intra-arc setting, the trailing remnant arc becomes extinct and undergoes relatively rapid thermal subsidence.
The remnant arc can serve as a substrate for carbonate-platform development, although these are usually short-lived platforms. (E) If stretching
began within an intra-arc setting, conjugate passive margins would develop on both sides of the newly created backarc ocean basin. Both conju-
gate margins should behave like other passive margins. Thus, carbonate platforms and reefs should become more areally extensive over time as
the margins undergo thermal subsidence and progressively wider potential substrates become available for carbonate sedimentation.
84 Dorobek
that are underlain by trapped oceanic crust are deep-water set-
tings almost from their inception, so shallow-marine carbon-
ate facies will not be deposited until the volcanic arc shoals to
shallow-water conditions. Even after the volcanic arc has built
itself up to shallow-water depths, backarc carbonate facies gen-
erally will consist only of fringing reefs or narrow rimmed-shelf
platforms that are constructed on the steep-sided volcanic edifice.
More areally extensive carbonate platforms can be constructed
only after appropriate substrates for shallow-marine carbonate
sedimentation have been constructed, which requires some com-
bination of erosion of the volcanic edifice (by both subaerial and
submarine processes) and deposition of volcaniclastic sediment
and lava flows.
Similarly, most carbonate sequences within extensional
backarc basins consist of fringing reefs or relatively narrow
ramps and rimmed shelves that are constructed (1) on the backarc
side of the active volcanic arc and arc massif, (2) for a short time
on the remnant arc as it drifts away from the active arc, or (3) on
fault-bounded basement highs that partition extensional backarc
regions into sub-basins. Postrift thermal subsidence ultimately
submerges the drifting remnant arc to subphotic water depths so
that any productive carbonate factories on the remnant-arc crust
are soon terminated. Scattered pinnacle reefs and mounded build-
ups that are typically less than a few hundred meters thick may
be constructed on synrift topography in backarc regions. These
buildups apparently are drowned soon after they form because of
the combined effects of rapid subsidence and short-term eustatic
sea-level rise.
Compressional backarc settings are more like retro-arc fore-
land basins, so they are not discussed here. Regardless, carbonate
platforms and reefs are common in these settings, forming on
fault-bounded highs that result from inversion tectonics. Isolated
platforms and buildups are common morphologies.
Quaternary examples. Quaternary carbonate platforms and
reefs are found in the backarc basins of the eastern Indonesian
archipelago, the Marianas, and Fiji. Quaternary carbonate reefs
and platforms from modern backarc settings may overlap, either
geographically or conceptually, with intra-arc carbonate systems,
especially where fault-bounded intra-arc basins evolve into back-
arc basins during progressive stretching of a volcanic arc.
Most Quaternary carbonate systems in backarc basins consist
of (1) predominantly fringing reefs or barrier reef tracts with nar-
row back-reef “lagoons” that are constructed on the backarc side
of the volcanic arc, (2) drowned pinnacle reefs or mounded car-
bonate buildups that also typically form on fault-bounded highs
in the backarc region, or (3) relatively narrow platforms that are
constructed on faulted backarc basement highs that either extend
from the volcanic arc or serve as accommodation zones that par-
tition extensional backarc regions into separate sub-basins. The
present-day abundance of fringing reef tracts and drowned build-
ups reflects both the high-amplitude (i.e., >100 m) character of
the post-Pleistocene sea-level rise and the typically steep struc-
tural or volcanic topography that was flooded by this rise in sea
level. The obvious structural controls on platform dimensions and
locations reflect the ongoing faulting and underfilled character
of many Quaternary backarc basins, which typically are in early
stages of development (i.e., <5 m.y. old). Carbonate platforms
and reefs are rare in more tectonically mature backarc basins,
which have undergone more tectonic subsidence, and potential
substrates for carbonate sedimentation are below photic depths.
Ancient examples. Ancient examples of backarc carbonate
platforms and reefs include (1) thin intervals of skeletal limestone
in the dominantly siliciclastic lower member (150 m total thick-
ness) of the middle Cretaceous Olvidada Formation, Baja Cali-
fornia Norte, Mexico (Phillips, 1993); (2) Upper Triassic (Norian)
coral buildups and carbonate foreslope(?) “conglomerates” (70 m
total thickness), Puale Bay, Alaska Peninsula (Wang et al., 1988);
and (3) possibly various Triassic carbonate-platform sequences
from Western North America (Stanley, 1982). The interpretation
of many of these examples as backarc-basin carbonate sequences
is based largely on paleogeographic reconstructions where the
relative position of the volcanic arc and the direction of subduc-
tion are known. Future work may ultimately show, however, that
some of these examples formed in forearc or intra-arc basins.
General Tectono-Stratigraphic Model for Carbonate Facies
Deposited across Backarc Regions
General tectono-stratigraphic models for carbonate platforms
and reefs that form in backarc settings depend on whether the
backarc basin is extensional or nonextensional (Fig. 17). Given
the relative randomness of eruptive events and shifting dispersal
patterns for volcaniclastic sediment, there may not be a general-
ized and recurrent stratigraphic succession that develops in back-
arc regions. Carbonate sedimentation, however, is probably more
common, long-lived, and areally extensive during later stages of
backarc basin evolution, when subsidence rates should be slower,
volcanic activity is waning, deposition of volcaniclastic and epi-
clastic sediment has created broad substrates for carbonate plat-
form development, and arc topography could be highly eroded so
that less sediment is supplied to the backarc region.
In remnant-ocean backarc basins that form where older oce-
anic crust is trapped behind an intraoceanic volcanic arc, carbon-
ate facies will likely develop only after volcanic edifices have built
themselves up to within photic depths (Fig. 17A). Narrow shelves
or fringing reef systems may form along the backarc side of the
newly formed volcanic arc. Carbonate platforms can become
more areally extensive on the backarc side of these intraoceanic
arcs when deposition of volcaniclastic sediment and lava flows
have built broad, shallow-water substrates. Active volcanism
and tectonic deformation of the volcanic arc will prevent thick
carbonate accumulations from developing, but as volcanism and
deformation wane, thicker platform sequences might develop.
Thus, the oldest carbonate sediments in remnant-ocean backarc
basins likely record when the growing volcanic arc first shallowed
into photic depths. These initial deposits might be followed by
complexly interlayered sequences of thin carbonate facies with
thicker volcaniclastic deposits and lava flows. The thickest and
perhaps most areally extensive carbonate sequences record the
 Carbonate-platform facies in volcanic-arc settings
end of volcanic and tectonic activity in remnant-ocean backarc
basins. Rare(?) crustal shortening might occur at any evolutionary
stage in remnant-ocean backarc basins and could generate surface
deformations (probably related to fault-propagation folding) or
cause regional flexural subsidence across the backarc region that
might also influence carbonate-sedimentation patterns.
In contrast, carbonate sedimentation in extensional backarc
basins will likely be influenced by the fault systems that accom-
modate stretching (Fig. 17C–E). Initial stretching will likely
occur within the hot and weak lithosphere that underlies the vol-
canic arc, so arc-related basement rocks may provide initial sub-
strates for the earliest carbonate strata. Most carbonate platforms
in modern extensional backarc basins consist of fringing reefs or
narrow rimmed shelves that build from the steep tectonic, volca-
nic, or erosional topography on either side of the basin. There are
no reported examples of synrift platforms or reefs within more
axial, deep-water parts of modern extensional backarc basins, but
they should be recognizable in ancient backarc basins by their
association with extensional fault systems and by their internal
growth stratal patterns. As stretching proceeds, the evolving
synrift topography must influence the location, dimensions, and
internal growth stratigraphy of carbonate platforms and reefs,
just as in marine intracontinental rift settings. If extension pro-
gresses to the point at which a spreading axis develops, the rifted
basement of the tectonically separated remnant and still-active
arc terranes may begin to subside independently. The remnant
arc will likely undergo faster rates of postrift subsidence than the
active-arc side of the extensional backarc basin, because there
will not be a nearby heat source that will prevent the remnant
arc from thermally subsiding. In fact, continued volcanic activ-
ity, changes in the properties of the subducting slab, or various
subduction-related geodynamic processes may even cause uplift
along the active-arc side of the extensional backarc region.
Carbonate sedimentation, however, may be most extensive in
extensional backarc basins during their late synrift to early postrift
(i.e., “drift”) stages of development when there are numerous
potential substrates across the backarc region where fringing and
isolated platforms can develop (cf. Dorsey and Kidwell, 1999).
These isolated platforms will reflect the strong influence of rem-
nant synrift topography, where fault-bounded highs are substrates
for carbonate platforms and structural lows trap siliciclastic and
volcaniclastic sediment from the adjacent active and remnant arcs.
Later in the postrift history, when sediments have filled the rem-
nant synrift topography, broad depositional surfaces may serve as
substrates for carbonate platforms during relative sea-level rises.
Along the remnant-arc side of the extensional backarc basin, the
inactive arc will be erosionally beveled and continue to undergo
thermal subsidence. Sediment supply from the remnant arc and
subsidence rates should progressively decrease over time, which
may also enhance development of broad carbonate platforms
during later postrift phases. In contrast, carbonate sedimentation
along the active-arc side of the basin may never become areally
extensive or long-lived as long as there is continued flux of volca-
nogenic material to the backarc region.
85
Tectonically Influenced Unconformities in Arc Depocenters
Unconformity development across carbonate platforms in
arc depocenters can affect the distribution of porosity and per-
meability within these platforms. Tectonically enhanced uncon-
formities will be best developed within carbonate platforms that
form where tectonic uplift may occur, such as:
1. The crest of accretionary subduction complexes, where
the internal deformation history of the subduction com-
plex controls the character of surface deformations that
affect unconformity development. Disconformities that
might cap shallowing-upward carbonate sequences on
the crest of the subduction complex can grade laterally
into conformable contacts toward more axial parts of the
forearc basin (Caron et al., 2004).
2. Intra-arc basins, which might be uplifted as a volcanic arc
grows. Entire isolated platforms or only minor segments
of barrier- or fringing-reef systems may be capped by tec-
tonically enhanced unconformities in intra-arc settings,
depending on the scale of arc uplift and whether uplift
patterns are affected by local faulting.
Carbonate platforms in extensional and neutral backarc
basin settings are not likely to have major, tectonically enhanced
unconformities because these settings are generally undergoing
subsidence, with little uplift throughout much of their history. In
contrast, compressional backarc settings (i.e., retro-arc foreland
basins) may have tectonically enhanced unconformities in the
wedge-top, forebulge, and backbulge depocenters (cf. Dorobek,
1995; Jordan, 1995; DeCelles and Giles, 1996).
SUMMARY
Carbonate platforms and reefs provide sensitive records
of differential subsidence and uplift, environmental conditions,
and interactions with other depositional systems in arc settings.
In general, detailed stratigraphic studies are lacking for all arc-
related depocenters because original stratal relationships are usu-
ally poorly preserved, physical access in many arc settings (both
ancient and modern) is difficult, high-quality seismic profiles and
extensive well data are not widely available, and actual strati-
graphic relationships are highly complex in three dimensions.
Thus, it is difficult to construct general stratigraphic models for
any arc depocenter.
Long-wavelength tectonic-subsidence patterns in arc sys-
tems will control the spatial distribution of shallow-water car-
bonate facies over long time scales and ultimately will control
the thickness of carbonate strata. Local, fault-controlled dif-
ferential subsidence will influence the sizes, morphology, and
internal facies patterns of carbonate platforms that develop in
any arc depocenter.
Siliciclastic and volcaniclastic flux will come largely from
the volcanic arc or arc-massif rocks. Carbonate facies may be bet-
ter developed with greater distance from a volcanic arc, although
carbonate facies may even accumulate along the flanks of the arc
86
if drainage patterns divert siliciclastic diment elsewhere, volcanic
eruptions are infrequent or are of low volume, or major wind-
ward-leeward effects are caused by arc topography.
Carbonate-platform morphology can be highly variable
across arc depocenters and is largely dependent on the availability
of tectonic or depositionally constructed substrates. For tropical
carbonate systems, these substrates generally need to be at photic
depths. Cool-water carbonate facies in arc depocenters, however,
might form over a broader range of water depths, but they will
still be limited by the availability of suitable substrates. Other
factors, such as nutrient levels and wind direction, also influence
carbonate-facies development in arc depocenters, although the
relative influence of these controls is highly specific to any par-
ticular arc setting.
I hope this review paper serves as a starting point for analy-
sis of these important systems, although more detailed studies,
especially of relatively young or modern arc settings, are needed.
Unraveling the history of more highly deformed, ancient exam-
ples will require the insight provided by studies from younger arc
settings as well as careful tectonic reconstructions, structural res-
torations, accurate age dating, and detailed sedimentologic and
stratigraphic analysis of the ancient examples.
ACKNOWLEDGMENTS
Much of the research for this paper was completed while I was
on leave from Texas A&M University during the 2000–2001
academic year. ExxonMobil Upstream Research Company
(EMURC) supported the research during my leave and provided
an excellent work environment. I am indebted to colleagues at
EMURC, both past and present, but most notably Jim Markello,
who was a major proponent for the study. Comments and sug-
gestions by Dan Bosence, John Reijmer, Sam Rice, Moyra Wil-
son, and co-editor Peter Clift improved the manuscript and are
most appreciated.
REFERENCES CITED
Ballance, P.F., 1993, The New Zealand Neogene forearc basins, in Ballance,
P.F., ed., South Pacific Sedimentary Basins, Sedimentary Basins of the
World 2: Amsterdam, Elsevier Science, p. 177–193.
Bardintzeff, J.M., Brousse, R., and Gachon, A., 1985, Conditions of building
coral reefs on a volcano: Mururoa in Tuamotu and Rurutu in Australes
(French Polynesia), in Gabrie, C., and Salvat, B., eds., Proceedings of Fifth
International Coral Reef Congress: Miami, Florida, Rosenstiel School of
Marine and Atmospheric Science, University of Miami, p. 401–405.
Beaudry, D., and Moore, G.F., 1985, Seismic stratigraphy and Cenozoic evolu-
tion of West Sumatra forearc basin: American Association of Petroleum
Geologists Bulletin, v. 69, p. 742–759.
Beck, M.E., Jr., 1983, On the mechanism of tectonic transport in zones of
oblique subduction: Tectonophysics, v. 93, p. 1–11, doi: 10.1016/0040-
1951(83)90230-5.
Bosence, D.W.J., 2005, A new, genetic classification of carbonate platforms
based on their basinal and tectonic setting in the Cenozoic: Sedimentary
Geology, v. 175, p. 49–72, doi: 10.1016/j.sedgeo.2004.12.030.
Bouysse, P., 1984, The Lesser Antilles island arc: Structure and geodynamic
evolution, in Biju-Duval, B., Moore, J.C., et al., Initial Reports of the
Deep Sea Drilling Project: Washington, U.S. Government Printing Office,
v. 78, p. 83–103.
Dorobek
Bradley, D.C., 1989, Taconic plate kinematics as revealed by foredeep stratigra-
phy, Appalachian orogen: Tectonics, v. 8, p. 1037–1049.
Bruns, T.R., Vedder, J.G., and Culotta, R.C., 1989, Structure and tectonics
along the Kilinailau Trench, Bougainville–Buka Island region, Papua
New Guinea, in Vedder, J.G., and Bruns, T.R., eds., Geology and Offshore
Resources of Pacific Island Arcs—Solomon Islands and Bougainville,
Papua New Guinea Regions: Houston, Circum-Pacific Council for Energy
and Mineral Resources, Earth Science Series, v. 12, p. 93–123.
Burbank, D.W., and Anderson, R.S., 2001, Tectonic Geomorphology: Cam-
bridge, Massachusetts, Blackwell Science, 274 p.
Cabioch, G., Taylor, F.W., Récy, J., Lawrence Edwards, R., Gray, S.C., Faure,
G., Burr, G.S., and Corrège, T., 1998, Environmental and tectonic influ-
ence on growth and internal structure of a fringing reef at Tasmaloum
(SW Espiritu Santo, New Hebrides island arc, SW Pacific), in Camoin,
G.F., and Davies, P.J., eds., Reefs and Carbonate Platforms in the Pacific
and Indian Oceans: International Association of Sedimentologists Special
Publication 25, p. 261–277.
Cadet, J.P., Kobayashi, K., Aubouin, J., Bouleque, J., Deplus, C., Dubois, J.,
von Huene, R., Jolivet, L., Kanazawa, T., Kasahara, J., Koizurni, K., Lal-
lemand, S., Nakamura, Y., Pautot, G., Suyehiro, S., Tokuyama, H., and
Yamazaki, T., 1987, The Japan trench and its juncture with the Kuril
trench: Cruise results of the Kaiko project leg 3: Earth and Planetary Sci-
ence Letters, v. 83, p. 267–284, doi: 10.1016/0012-821X(87)90071-9.
Cahill, T., and Isacks, B.L., 1992, Seismicity and shape of the subducted Nazca
Plate: Journal of Geophysical Research, v. 97, p. 17,503–17,529.
Carney, J.N., Macfarlane, A., and Mallick, D.I.J., 1985, The Vanuatu island
arc—An outline of the stratigraphy, structure, and petrology, in Nairn,
A.E.M., et al., eds., The Ocean Basins and Margins—The Pacific Ocean:
New York, Plenum Press, v. 7a, p. 685–718.
Caron, V., Nelson, C.S., and Kamp, P.J.J., 2004, Contrasting carbonate deposi-
tional systems for Pliocene cool-water limestones cropping out in central
Hawke’s Bay, New Zealand: New Zealand Journal of Geology and Geo-
physics, v. 47, p. 697–717.
Clift, P., and Vannucchi, P., 2004, Controls on tectonic accretion versus ero-
sion in subduction zones: Implications for the origin and recycling of
the continental crust: Reviews of Geophysics, v. 42, p. RG2001, doi:
10.1029/2003RG000127.
Clift, P.D., MacLeod, C.J., Tappin, D.R., Wright, D., and Bloomer, S.H.,
1998, Tectonic controls on sedimentation and diagenesis in the Tonga
trench and forearc, SW Pacific: Geological Society of America Bul-
letin, v. 110, p. 483–496, doi: 10.1130/0016-7606(1998)110<0483:
TCOSAD>2.3.CO;2.
Covey, M., 1986, The evolution of foreland basins to steady state: Evidence
from the western Taiwan foreland basin, in Allen, P., and Homewood,
P., eds., Foreland Basins: International Association of Sedimentologists
Special Publication 8, p. 77–90.
Coward, M.P., 1990, The Precambrian, Caledonian and Variscan framework to
NW Europe, in Hardman, R.F.P., and Brooks, J., eds., Tectonic Events
Responsible for Britain’s Oil and Gas Reserves: Geological Society [Lon-
don] Special Publication 19, p. 3–36.
Cunningham, J.K., and Anscombe, K.J., 1985, Geology of Eua and other islands,
Kingdom of Tonga, in Scholl, D.W., and Vallier, T.L., eds., Geology and
Offshore Resources of Pacific Island Arcs—Tonga Region: Houston, Cir-
cum-Pacific Council for Energy and Mineral Resources, Earth Science
Series, v. 2, p. 221–257.
Dahlen, F.A., Suppe, J., and Davis, D., 1984, Mechanics of fold-and-thrust belts
and accretionary wedges: Cohesive Coulomb theory: Journal of Geophys-
ical Research, v. 89, p. 10,087–10,101.
D’Anglejan, B., and Mountjoy, E.W., 1973, Submerged reefs of the east-
ern Grenadines shelf margin: Geological Society of America Bul-
letin, v. 84, p. 2445–2454, doi: 10.1130/0016-7606(1973)84<2445:
SROTEG>2.0.CO;2.
Davis, D., Suppe, J., and Dahlen, F.A., 1983, Mechanics of fold-and-thrust
belts and accretionary wedges: Journal of Geophysical Research, v. 88,
p. 1153–1172.
DeCelles, P.G., and Giles, K.A., 1996, Foreland basin systems: Basin Research,
v. 8, p. 105–123, doi: 10.1046/j.1365-2117.1996.01491.x.
Demirmen, F., and Harbaugh, J.W., 1965, Petrography and origin of Perm-
ian McCloud Limestone of northern California: Journal of Sedimentary
Petrology, v. 35, p. 136–154.
 Carbonate-platform facies in volcanic-arc settings
Dewey, J.F., 1980, Episodicity, sequence, and style at convergent plate boundar-
ies, in Strangway, D.W., ed., The Continental Crust and Its Mineral Depos-
its: Geological Association of Canada Special Paper 20, p. 533–573.
Dey, S., and Smith, L., 1989, Carbonate and volcanic sediment distribution pat-
terns on the Grenadines Bank, Lesser Antilles Island Arc, Eastern Carib-
bean: Bulletin of Canadian Petroleum Geology, v. 37, p. 18–30.
Dickinson, W.R., 1974, Plate tectonics and sedimentation, in Dickinson, W.R.,
ed., Tectonics and Sedimentation: Society of Economic Paleontologists
and Mineralogists Special Publication 22, p. 1–27.
Dickinson, W.R., 1995, Forearc basins, in Busby, C.J., and Ingersoll, R.V., eds.,
Tectonics of Sedimentary Basins: Cambridge, Massachusetts, Blackwell
Science, p. 221–261.
Dickinson, W.R., 2001, Paleoshoreline record of relative Holocene sea levels on
Pacific islands: Earth-Science Reviews, v. 55, p. 191–234, doi: 10.1016/
S0012-8252(01)00063-0.
Dickinson, W.R., and Seely, D.R., 1979, Structure and stratigraphy of forearc
regions: American Association of Petroleum Geologists Bulletin, v. 63,
p. 2–31.
DiMarco, G., Baumgartner, P.O., and Channell, J.E.T., 1995, Late Cretaceous–
early Tertiary paleomagnetic data and a revised tectonostratigraphic subdi-
vision of Costa Rica and western Panama, in Mann, P., ed., Geologic and
Tectonic Development of the Caribbean Plate Boundary in Southern Cen-
tral America: Geological Society of America Special Paper 295, p. 1–27.
Dorobek, S.L., 1995, Synorogenic carbonate platforms and reefs in foreland
basins: Controls on stratigraphic evolution and platform/reef morphology,
in Dorobek, S.L., and Ross, G.M., eds., Stratigraphic Evolution of Fore-
land Basins: SEPM (Society for Sedimentary Geology) Special Publica-
tion 52, p. 127–147.
Dorsey, R.J., and Kidwell, S.M., 1999, Mixed carbonate-siliciclastic sedimen-
tation on a tectonically active margin: Example from the Pliocene of Baja
California Sur, Mexico: Geology, v. 27, p. 935–938, doi: 10.1130/0091-
7613(1999)027<0935:MCSSOA>2.3.CO;2.
Eberli, G.P., and Ginsburg, R.N., 1989, Cenozoic progradation of northwestern
Great Bahama Bank, a record of lateral platform growth and seal-level
fluctuations, in Crevello, P.D., Wilson, J.L., Sarg, J.F., and Read, J.F., eds.,
Controls on Carbonate Platform and Basin Development: SEPM Special
Publication 44, p. 339–352.
Eguchi, T., Uyeda, S., and Maki, T., 1980, Seismotectonics and tectonic history
of the Andaman Sea, in Toksöz, M.N., et al., eds., Oceanic Ridges and
Arcs: New York, Elsevier, p. 425–441.
Eldredge, L.C., and Kropp, R.K., 1985, Volcanic ashfall effects on intertidal
and shallow-water coral reef zones at Pagan, Mariana Islands: Tahiti, Pro-
ceedings of 5th International Coral Reef Congress, p. 195–200.
Exon, N.F., Stewart, W.D., Sandy, M.J., and Tiffin, D.L., 1986, Geology of the
eastern New Ireland Basin, northeastern Papua New Guinea: Bureau of
Mineral Resources, Journal of Australian Geology and Geophysics, v. 10,
p. 39–51.
Flemings, P.B., and Jordan, T.E., 1990, Stratigraphic modeling of fore-
land basins: Interpreting thrust deformation and lithosphere rheology:
Geology, v. 18, p. 430–434, doi: 10.1130/0091-7613(1990)018<0430:
SMOFBI>2.3.CO;2.
Fortuin, A.R., Van der Werff, W., and Wensink, H., 1997, Neogene basin history
and paleomagnetism of a rifted and inverted forearc region, on- and off-
shore Sumba, Eastern Indonesia: Journal of Asian Earth Sciences, v. 15,
p. 61–88.
Galindo-Zaldívar, J., Nieto, L.M., Robertson, A.H.F., and Woodside, J.M.,
2001, Recent tectonics of Eratosthenes Seamount: An example of sea-
mount deformation during incipient continental collision: Geo-Marine
Letters, v. 20, p. 233–242, doi: 10.1007/s003670000059.
Geist, E.L., Childs, J.R., and Scholl, D.W., 1988, The origin of summit basins
of the Aleutian Ridge: Implications for block rotation of an arc massif:
Tectonics, v. 7, p. 327–341.
Greene, H.G., Macfarlane, A., and Wong, F.L., 1988, Geology and offshore
resources of Vanuatu—Introduction and summary, in Greene, H.G., and
Wong, F.L., eds., Geology and Offshore Resources of Pacific Island
Arcs—Vanuatu Region: Houston, Circum-Pacific Council for Energy and
Mineral Resources, Earth Science Series, v. 8, p. 1–25.
Grippi, J., 1980, Geology of the Lucea Inlier, western Jamaica: Journal of the
Geological Society of Jamaica, v. 19, p. 1–24.
Gurnis, M., 1990a, Bounds on global dynamic topography from Phanero-
zoic flooding of continental platforms: Nature, v. 344, p. 754–756, doi:
10.1038/344754a0.
87
Gurnis, M., 1990b, Plate-mantle coupling and continental flooding: Geophysi-
cal Research Letters, v. 17, p. 623–626.
Gurnis, M., 1991, Continental flooding and mantle-lithosphere dynamics, in
Sabadini, R., et al., eds., Glacial Isostasy, Sea Level and Mantle Rheol-
ogy: Dordrecht, Netherlands, Kluwer Academic Publishers, p. 446.
Haggerty, J.A., 1991, Evidence from fluid seeps atop serpentine seamounts in
the Mariana Forearc; clues for emplacement of the seamounts and their
relationship to forearc tectonics, in Meyer, A.W., et al., eds., Evolution of
Mesozoic and Cenozoic Continental Margins: Marine Geology, v. 102,
p. 293–309, doi: 10.1016/0025-3227(91)90013-T.
Halfar, H., Godinez-Orta, L., Mutti, M., and Valdez-Holguin, B., 2004, Nutri-
ents and temperature controls on modern carbonate production: An exam-
ple from the Gulf of California, Mexico: Geology, v. 32, p. 213–216, doi:
10.1130/G20298.1.
Hallock, P., and Schlager, W., 1986, Nutrient excess and the demise of
coral reefs and carbonate platforms: Palaios, v. 1, p. 389–398, doi:
10.2307/3514476.
Hamilton, W., 1979, Tectonics of the Indonesian region: U.S. Geological Sur-
vey Professional Paper 1078, 345 p.
Hathway, B., 1994, Sedimentation and volcanism in an Oligocene-Miocene
intra-oceanic arc and fore-arc, southwestern Viti Levu, Fiji: Geological
Society [London] Journal, v. 151, p. 499–514.
Hathway, B., 1995, Deposition and diagenesis of Miocene arc-fringing plat-
form and debris-apron carbonates, southwestern Viti Levu, Fiji: Sedimen-
tary Geology, v. 94, p. 187–208, doi: 10.1016/0037-0738(94)00086-A.
Hattori, M., Kanie, Y., Oba, T., and Akimoto, K., 1995, Environmental condi-
tions of carbonates and chemosynthetic animal communities associated
with cold seepage zones along the subduction zone in Sagami Bay, central
Japan: Fossils, v. 60, p. 13–22.
Henrich, R., and Freiwald, A., 1995, Controls on modern carbonate sedimenta-
tion on warm-temperate to arctic coasts, shelves and seamounts in the
northern hemisphere: Implications for fossil counterparts: Facies, v. 32,
p. 71–108, doi: 10.1007/BF02536865.
Herzer, R.H., and Exon, N.F., 1985, Structure and basin analysis of the southern
Tonga forearc, in Schole, D.W., and Valier, T.L., eds., Geology and off-
shore resources of Pacific island arcs—Tonga Region: Houston, Circum-
Pacific Council for Energy and Mineral Resources, Earth Science Series,
v. 2, p. 55–73.
Ingersoll, R.V., 1988, Tectonics of sedimentary basins: Geological Soci-
ety of America Bulletin, v. 100, p. 1704–1719, doi: 10.1130/0016-
7606(1988)100<1704:TOSB>2.3.CO;2.
Iryu, Y., Nakamori, T., Matsuda, S., and Abe, O., 1995, Distribution of marine
organisms and its geological significance in the modern reef complex
of the Ryukyu Islands: Sedimentary Geology, v. 99, p. 243–258, doi:
10.1016/0037-0738(95)00047-C.
James, N.P., 1997, The cool-water carbonate depositional realm, in James, N.P.,
and Clarke, J.A.D., eds., Cool-water Carbonates: SEPM (Society for Sed-
imentary Geology) Special Publication 56, p. 1–20.
Johnson, D.P., and Greene, H.G., 1988, Modern depositional regimes, offshore
Vanuatu, in Greene, H.G., and Wong, F.L., eds., Geology and Offshore
Resources of Pacific Island Arcs—Vanuatu Region: Houston, Circum-
Pacific Council for Energy and Mineral Resources, Earth Science Series,
v. 8, p. 287–299.
Johnson, R.W., 1979, Geotectonics and volcanism in Papua New Guinea: A
review of the late Cainozoic: BMR Journal of Australian Geology and
Geophysics, v. 4, p. 181–207.
Jordan, T.E., 1995, Retroarc foreland and related basins, in Busby, C.J., and
Ingersoll, R.V., eds., Tectonics of Sedimentary Basins: Cambridge, Mas-
sachusetts, Blackwell Science, p. 331–362.
Jordan, T.E., and Allmendinger, R.W., 1986, The Sierra Pampeanas of Argen-
tina: A modern analogue of Rocky Mountain foreland deformation:
American Journal of Science, v. 286, p. 737–764.
Jordan, T.E., Isacks, B.L., Allmendinger, R.W., Brewer, J.A., Ramos, V.A.,
and Ando, C.J., 1983, Andean tectonics related to geometry of subducted
Nazca plate: Geological Society of America Bulletin, v. 94, p. 341–361,
doi: 10.1130/0016-7606(1983)94<341:ATRTGO>2.0.CO;2.
Kamp, P.J.J., and Nelson, C.S., 1987, Tectonic and sea level controls on non-
tropical Neogene limestones in New Zealand: Geology, v. 15, p. 610–613,
doi: 10.1130/0091-7613(1987)15<610:TASCON>2.0.CO;2.
Kamp, P.J.J., and Nelson, C.S., 1988, Nature and occurrence of modern and
Neogene active margin limestones in New Zealand: New Zealand Journal
of Geology and Geophysics, v. 31, p. 1–20.
88 Dorobek
Kamp, P.J.J., Harmsen, F.J., Nelson, C.S., and Boyle, S.F., 1988, Barnacle-
dominated limestone with giant cross-beds in a non-tropical, tide-swept,
Pliocene forearc seaway, Hawke’s Bay, New Zealand: Sedimentary Geol-
ogy, v. 60, p. 173–195, doi: 10.1016/0037-0738(88)90118-2.
Karig, D.E., 1983, Temporal relationships between back arc basin formation
and arc volcanism with special reference to the Philippine Sea, in Hayes,
D.E., ed., The Tectonic and Geologic Evolution of Southeast Asian Seas
and Islands (Pt. II): American Geophysical Union, AGU Monograph
Series, no. 27, P. 318–325.
Karig, D.E., Sarewitz, D.R., and Haeck, G.D., 1986, Role of strike-slip
faulting in the evolution of allochthonous terranes in the Philippines:
Geology, v. 14, p. 852–855, doi: 10.1130/0091-7613(1986)14<852:
ROSFIT>2.0.CO;2.
Kelly, S.R.A., Ditchfield, P.W., Doubleday, P.A., and Marshall, J.D., 1995,
An Upper Jurassic methane-seep limestone from the Fossil Bluff Group
forearc basin of Alexander Island, Antarctica: Journal of Sedimentary
Research, v. 65, p. 274–282.
Kikuchi, Y., Tono, S., and Funayama, M., 1991, Petroleum resources in the
Japanese island-arc setting: Episodes, v. 14, p. 236–241.
Klaus, A., Taylor, B., Moore, G.F., Murakami, F., and Okamura, Y., 1992, Back-
arc rifting in the Izu-Bonin island arc: Structural evolution of Hachijo
and Aoga Shima rifts: Island Arc, v. 1, p. 16–31, doi: 10.1111/j.1440-
1738.1992.tb00054.x.
Kolarsky, R.A., Mann, P., and Monechi, S., 1995, Stratigraphic development of
southwestern Panama as determined from integration of marine seismic
data and onshore geology, in Mann, P., ed., Geologic and Tectonic Devel-
opment of the Caribbean Plate Boundary in Southern Central America:
Geological Society of America Special Paper 295, p. 159–200.
Krawinkel, H.E., and Krawinkel, J.J., 1996, From strike-slip depositional set-
ting to fold and thrust belt; Tertiary to Recent structural and sedimentary
development of a forearc basin (Terraba Basin, southern Costa Rica):
American Association of Petroleum Geologists annual convention, San
Diego, California, May 19–22, 1996, Annual Meeting Abstracts, v. 5,
p. 78–79.
Kroenke, L.W., 1984, Cenozoic tectonic development of the southwest Pacific:
United Nations ESCAP, CCOP/SOPAC, Technical Bulletin 6, 122 p.
Kulm, L.D., and Suess, E., 1990, Relationship between carbonate deposits and
fluid venting; Oregon accretionary prism, in Langseth, M.G., and Moore,
J.C., eds., Special section on the Role of fluids in sediment accretion,
deformation, diagenesis, and metamorphism in subduction zones: Journal
of Geophysical Research, v. 95, p. 8899–8915.
Kulm, L.D., Suess, E., Moore, J.C., Carson, B., Lewis, B.T., Ritger, S.D.,
Kadko, D.C., Thornburg, T.M., Embley, R.W., Rugh, W.D., Massoth,
G.J., Langseth, M.G., Cochrane, G.R., and Scamman, R.L., 1986, Oregon
subduction zone; venting, fauna, and carbonates: Science, v. 231, p. 561–
566, doi: 10.1126/science.231.4738.561.
Larue, D.K., and Suess, E., 1985, Eocene subduction-driven vent community,
Joes River melange, Barbados: Eos (Transactions, American Geophysical
Union), v. 66, p. 1097.
Letouzey, J., and Kimura, M., 1986, The Okinawa Trough: Genesis of a back-
arc basin developing along a continental margin: Tectonophysics, v. 125,
p. 209–230, doi: 10.1016/0040-1951(86)90015-6.
Lundberg, N., 1982, Evolution of the slope landward of the Middle America
Trench, Nicoya Peninsula, Costa Rica, in Leggett, J.K., ed., Trench-
Forearc Geology: Sedimentation and Tectonics on Modern and Ancient
Active Plate Margins: Geological Society [London] Special Publication
10, p. 131–147.
Macfarlane, A., Carney, J.N., Crawford, A.J., and Greene, H.G., 1988, Van-
uatu—A review of onshore geology, in Greene, H.G., and Wong, F.L.,
eds., Geology and Offshore Resources of Pacific Island Arcs—Vanu-
atu Region: Houston, Circum-Pacific Council for Energy and Mineral
Resources, Earth Science Series, v. 8, p. 45–92.
Mann, P., Taylor, F.W., Lagoe, M.B., Quarles, A., and Burr, G., 1998, Acceler-
ating late Quaternary uplift of the New Georgia island group (Solomon
island arc) in response to subduction of recently active Woodlark spread-
ing center and Coleman seamount: Tectonophysics, v. 295, p. 259–306,
doi: 10.1016/S0040-1951(98)00129-2.
Marsaglia, K.M., 1995, Interarc and backarc basins, in Busby, C.J., and Inger-
soll, R.V., eds., Tectonics of Sedimentary Basins: Cambridge, Massachu-
setts, Blackwell Science, p. 299–329.
Matson, R.G., and Moore, G.F., 1992, Structural influences on Neogene subsid-
ence in the central Sumatra fore-arc basin, in Watkins, J.S., et al., eds.,
Geology and Geophysics of Continental Margins: American Association
of Petroleum Geologists Memoir 53, p. 157–181.
Maynard, J.R., Hofmann, W., Dunay, R.E., Bentham, P.N., Dean, K.P., and Wat-
son, I., 1997, The Carboniferous of western Europe: The development of a
petroleum system: Petroleum Geoscience, v. 3, p. 97–115.
Meffre, S., and Crawford, A.J., 2001, Collision tectonics in the New Hebri-
des arc (Vanuatu): Island Arc, v. 10, p. 33–50, doi: 10.1046/j.1440-
1738.2001.00292.x.
Mesolella, K.J., Matthews, R.K., Broecker, W.S., and Thurber, D.L., 1969, The
astronomical theory of climate change: Barbados data: Journal of Geol-
ogy, v. 77, p. 250–274.
Minoura, K., 1992, Dolomitization of Permian reef carbonates in a Jurassic
subduction complex, central Japan: Sedimentary Geology, v. 80, p. 41–
52, doi: 10.1016/0037-0738(92)90030-U.
Moore, J.C., and Vrolijk, P., 1992, Fluids in accretionary prisms: Reviews of
Geophysics, v. 30, p. 113–135.
Nakamori, T., Iryu, Y., and Yamada, T., 1995, Development of coral reefs of
the Ryukyu Islands (southwest Japan, East China Sea) during Pleisto-
cene sea-level change: Sedimentary Geology, v. 99, p. 215–231, doi:
10.1016/0037-0738(95)00045-A.
Niem, A.R., 1989, Coral and limestone, carbonate cements in siliciclastic rocks,
and chalk from dredge samples, Solomon Islands and Bougainville, Papua
New Guinea, in Vedder, J.G., and Bruns, T.R., eds., Geology and Offshore
Resources of Pacific Island Arcs—Solomon Islands and Bougainville,
Papua New Guinea Regions: Houston, Circum-Pacific Council for Energy
and Mineral Resources, Earth Science Series, v. 12, p. 175–202.
Nunn, P.D., 1998a, Pacific Island Landscapes: Suva, Fiji, University of the
South Pacific, Institute of Pacific Studies, 318 p.
Nunn, P.D., 1998b, Late Cenozoic emergence of the islands of the northern
Lau-Colville Ridge, southwest Pacific, in Stewart, I.S., and Vita-Finzi, C.,
eds., Coastal Tectonics: Geological Society [London] Special Publication
146, p. 269–278.
Nunn, P.D., 2000, Significance of emerged Holocene corals around Ovalau and
Motuiki islands, Fiji, southwest Pacific: Marine Geology, v. 163, p. 345–
351, doi: 10.1016/S0025-3227(99)00114-0.
Nunn, P.D., and Omura, A., 1999, Penultimate Interglacial emergent reef
around Kadavu Island, Southwest Pacific: Implications for late Quater-
nary island-arc tectonics and sea-level history: New Zealand Journal of
Geology and Geophysics, v. 42, p. 219–227.
Okamura, Y., Watanabe, M., Morijiri, R., and Satoh, M., 1995, Rifting and
basin inversion in the eastern margin of the Japan Sea: Island Arc, v. 4,
p. 166–181, doi: 10.1111/j.1440-1738.1995.tb00141.x.
Otsuki, K., 1989, Empirical relationships among the convergence rate of plates,
rollback rate of trench axis and island-arc tectonics: Laws of convergence
rate of plates: Tectonophysics, v. 159, p. 73–94, doi: 10.1016/0040-
1951(89)90171-6.
Phillips, J.R., 1993, Stratigraphy and structural setting of the mid-Cretaceous
Olvidada Formation, Baja California Norte, Mexico, in Gastil, R.G., and
Miller, R.H., eds., The Prebatholithic Stratigraphy of Peninsular Califor-
nia: Geological Society of America Special Paper 279, p. 97–106.
Pomoni-Papaioannou, F., Drinia, H., and Dermitzakis, M.D., 2003, Neogene
non-tropical carbonate sedimentation in a warm temperate biogeographic
province (Rethymnon Formation, Eastern Crete, Greece): Sedimentary
Geology, v. 154, p. 147–157, doi: 10.1016/S0037-0738(02)00127-6.
Poulton, T.P., 1989, A Lower Jurassic coral reef, Telkwa Range, British Colum-
bia: Canadian Society of Petroleum Geologists Memoir 13, p. 754–757.
Read, J.F., 1985, Carbonate platform facies models: American Association of
Petroleum Geologists Bulletin, v. 69, p. 1–21.
Read, J.F., 1995, Overview of carbonate platform sequences, cycle stratigraphy
and reservoirs in greenhouse and ice-house worlds, in Read, J.F., et al.,
eds., Milankovitch Sea Level Changes, Cycles and Reservoirs on Carbon-
ate Platforms in Greenhouse and Icehouse Worlds: SEPM (Society for
Sedimentary Geology) Short Course Notes 35, p. 1–102.
Reid, R.P., and Ginsburg, R.N., 1986, The role of framework in Upper Tri-
assic patch reefs in the Yukon (Canada): Palaios, v. 1, p. 590–600, doi:
10.2307/3514709.
Rodd, J.A., 1993, New reef targets for oil and gas exploration in Fiji, Southwest
Pacific: Geological Society of Malaysia Bulletin, v. 33, p. 313–330.
Rutherford, E., Burke, K., and Lytwyn, J., 2001, Tectonic history of Sumba
Island, Indonesia, since the Late Cretaceous and its rapid escape into the
forearc in the Miocene: Journal of Asian Earth Sciences, v. 19, p. 453–
479, doi: 10.1016/S1367-9120(00)00032-8.
 Carbonate-platform facies in volcanic-arc settings
Sagawa, N., Nakamori, T., and Iryu, Y., 2001, Pleistocene reef development
in the southwest Ryukyu Islands, Japan: Palaeogeography, Palaeocli-
matology, Palaeoecology, v. 175, p. 303–323, doi: 10.1016/S0031-
0182(01)00377-7.
Sample, J.C., and Fisher, D.M., 1986, Duplexes and underplating in an ancient
accretionary complex, Kodiak Islands, Alaska: Geology, v. 14, p. 160–
163, doi: 10.1130/0091-7613(1986)14<160:DAAUIA>2.0.CO;2.
Sarewitz, D.R., and Lewis, S.D., 1991, The Marinduque intra-arc basin, Phil-
ippines: Basin genesis and in situ ophiolite development in a strike-slip
setting: Geological Society of America Bulletin, v. 103, p. 597–614, doi:
10.1130/0016-7606(1991)103<0597:TMIABP>2.3.CO;2.
Seely, D.R., 1979, The evolution of structural highs bordering major forearc
basins: American Association of Petroleum Geologists Memoir 29,
p. 245–260.
Seely, D.R., Vail, P.R., and Walton, G.G., 1974, Trench slope model, in Burk,
C.A., and Drake, C.L., eds., The Geology of Continental Margins: New
York, Springer-Verlag, p. 249–260.
Seyfried, H., Astorga, A., Amann, H., Calvo, C., Kolb, W., Schmidt, H., and
Winsemann, J., 1991, Anatomy of an evolving island arc: Tectonic and
eustatic control in the south Central American fore-arc area, in Mac-
donald, D.I.M., ed., Sedimentation, Tectonics and Eustasy—Sea-level
Changes at Active Margins: International Association of Sedimentolo-
gists Special Publication 12, p. 151–163.
Sharaf, E., Simo, J.A., Carroll, A.R., and Shields, M., 2005, Stratigraphic
evolution of Oligocene–Miocene carbonates and siliciclastics, East Java
basin, Indonesia: American Association of Petroleum Geologists Bulletin,
v. 89, p. 799–819.
Shin Tani, 1989, Detailed topographic study of the Daiichi-Kashima Seamount:
Palaeogeography, Palaeoclimatology, Palaeoecology, v. 71, p. 31–47, doi:
10.1016/0031-0182(89)90028-X.
Sieh, K., Ward, S.N., Natawidjaja, D., and Suwargadi, B.W., 1999, Crustal defor-
mation at the Sumatran subduction zone revealed by coral rings: Geophys-
ical Research Letters, v. 26, p. 3141–3144, doi: 10.1029/1999GL005409.
Sigurdsson, H., Sparks, R.S.J., Carey, S.N., and Huang, T.C., 1980, Volcano-
genic sedimentation in the Lesser Antilles arc: Journal of Geology, v. 88,
p. 523–540.
Silver, E.A., and Reed, D.L., 1988, Backthrusting in accretionary wedges: Jour-
nal of Geophysical Research, v. 93, p. 3116–3126.
Smith, G.A., and Landis, C.A., 1995, Intra-arc basins, in Busby, C.J., and Inger-
soll, R.V., eds., Tectonics of Sedimentary Basins: Cambridge, Massachu-
setts, Blackwell Science, p. 263–298.
Soja, C.M., 1990, Island arc carbonates from the Silurian Heceta Formation of
southeastern Alaska (Alexander terrane): Journal of Sedimentary Petrol-
ogy, v. 60, p. 235–249.
Soja, C.M., 1992, Potential contributions of ancient oceanic islands to evolu-
tionary theory: Journal of Geology, v. 100, p. 125–134.
Soja, C.M., 1996, Island-arc carbonates: Characterization and recognition in
the ancient geologic record: Earth-Science Reviews, v. 41, p. 31–65, doi:
10.1016/0012-8252(96)00029-3.
Stanley, G.D., Jr., 1982, Triassic carbonate development and reefbuilding in
western North America: Geologische Rundschau, v. 71, p. 1057–1075,
doi: 10.1007/BF01821118.
Stanley, G.D., Jr., and Beauvais, L., 1990, Middle Jurassic corals from the Wal-
lowa Terrane, west-central Idaho: Journal of Paleontology, v. 64, p. 352–
362.
Stanley, G.D., Jr., and McRoberts, C.A., 1993, A coral reef in the Telkwa
Range, British Columbia: The earliest Jurassic example: Canadian Jour-
nal of Earth Sciences, v. 30, p. 819–831.
Stanley, G.D., and Senowbari-Daryan, B., 1986, Upper Triassic, Dachstein-
type, reef limestone from the Wallowa Mountains, Oregon; first
reported occurrence in the United States: Palaios, v. 1, p. 172–177, doi:
10.2307/3514511.
Stewart, W.D., and Sandy, M.J., 1988, Geology of New Ireland and Djaul
Islands, northeastern Papua New Guinea, in Marlow, M.S., et al., eds.,
Geology and Offshore Resources of Pacific Island Arcs—New Ireland and
Manus Regions, Papua New Guinea: Houston, Circum-Pacific Council
for Energy and Mineral Resources, Earth Science Series, v. 9, p. 13–30.
Takano, O., 2002, Changes in depositional systems and sequences in response
to basin evolution in a rifted and inverted basin: An example from the
Neogene Niigata-Shin’etsu basin, Northern Fossa Magna, central
Japan: Sedimentary Geology, v. 152, p. 79–87, doi: 10.1016/S0037-
0738(01)00286-X.
89
Tamaki, K., 1995, Opening tectonics of the Japan Sea, in Taylor, B., ed., Backarc
Basins: Tectonics and Magmatism: New York, Plenum Press, p. 407–420.
Tapponnier, P., Peltzer, G., LeDain, A.Y., Armijo, R., and Cobbold, P., 1982,
Propagation extrusion tectonics in Asia: New insights from simple experi-
ments with plasticene: Geology, v. 10, p. 611–616, doi: 10.1130/0091-
7613(1982)10<611:PETIAN>2.0.CO;2.
Taylor, B., Brown, G., Fryer, P., Gill, J.B., Hochstaedter, A.G., Hotta, H., Lang-
muir, C.H., Leinen, M., Nishimura, A., and Urabe, T., 1990, ALVIN–Sea
Beam studies of the Sumisu Rift, Izu-Bonin Arc: Earth and Planetary Sci-
ence Letters, v. 100, p. 127–147, doi: 10.1016/0012-821X(90)90181-V.
Taylor, B., Klaus, A., Brown, G.R., and Moore, G.F., 1991, Active tectonics of
the north and central Aegean Sea: Journal of Geophysical Research, v. 96,
p. 19,611–19,622.
ten Veen, J.H., and Postma, G., 1999, Neogene tectonics and basin fill patterns
in the Hellenic outer-arc (Crete, Greece): Basin Research, v. 11, p. 223–
241, doi: 10.1046/j.1365-2117.1999.00097.x.
Thomas, W.A., 1983, Continental margins, orogenic belts, and intracra-
tonic structures: Geology, v. 11, p. 270–272, doi: 10.1130/0091-
7613(1983)11<270:CMOBAI>2.0.CO;2.
Uyeda, S., and Kanamori, H., 1979, Back-arc opening and the mode of subduc-
tion: Journal of Geophysical Research, v. 84, p. 1049–1061.
van der Werff, W., 1996, Variation in forearc basin development along the
Sunda Arc, Indonesia: Journal of Southeast Asian Earth Sciences, v. 14,
p. 331–349, doi: 10.1016/S0743-9547(96)00068-2.
Vedder, J.G., and Bruns, T.R., 1989, Geologic setting and petroleum prospects
of basin sequences, offshore Solomon Islands and eastern Papua New
Guinea, in Vedder, J.G., and Bruns, T.R., eds., Geology and Offshore
Resources of Pacific Island Arcs: Solomon Islands and Bougainville,
Papua New Guinea Regions: Houston, Circum-Pacific Council for Energy
and Mineral Resources, Earth Science Series, v. 12, p. 287–322.
Vergés, J., Marzo, M., and Muñoz, J.A., 2002, Growth strata in foreland
settings: Sedimentary Geology, v. 146, p. 1–9, doi: 10.1016/S0037-
0738(01)00162-2.
von Huene, R., Langseth, M., Nasu, N., and Okada, H., 1982, A summary of
Cenozoic tectonic history along the IPOD Japan Trench transect: Geolog-
ical Society of America Bulletin, v. 93, p. 829–846, doi: 10.1130/0016-
7606(1982)93<829:ASOCTH>2.0.CO;2.
Wang, J., Newton, C.R., and Dunne, L., 1988, Late Triassic transition from bio-
genic to arc sedimentation on the Peninsular terrane: Puale Bay, Alaska
Peninsula: Geological Society of America Bulletin, v. 100, p. 1466–1478,
doi: 10.1130/0016-7606(1988)100<1466:LTTFBT>2.3.CO;2.
Watkins, R., 1985, Volcaniclastic and carbonate sedimentation in late Paleo-
zoic island-arc deposits, Eastern Klamath Mountains, California:
Geology, v. 13, p. 709–713, doi: 10.1130/0091-7613(1985)13<709:
VACSIL>2.0.CO;2.
Watkins, R., 1993a, Carbonate bank sedimentation in a volcaniclastic arc set-
ting: Lower Carboniferous limestones of the Eastern Klamath terrane,
California: Journal of Sedimentary Petrology, v. 63, p. 966–973.
Watkins, R., 1993b, Permian carbonate platform development in an island-
arc setting, Eastern Klamath Mountains, California: Journal of Geology,
v. 101, p. 659–666.
Watkins, R., and Flory, R.A., 1986, Island arc sedimentation in the Middle
Devonian Kennett Formation, Eastern Klamath Mountains, California:
Journal of Geology, v. 94, p. 753–761.
Wells, J.W., 1953, Mesozoic invertebrate faunas of Peru, part 3, Lower Juras-
sic corals from the Arequipa region: American Museum Novitiates 1631,
14 p.
Wells, S.M., 1988, Coral Reefs of the World, Volume 3: Central and Western
Pacific: Nairobi, Kenya/IUCN, Gland, Switzerland, and Cambridge, UK,
UNEP (United Nations Environment Programme), Regional Seas Direc-
tories and Bibliographies, 329 p.
Whalen, M.T., 1995, Barred basins: A model for eastern ocean basin car-
bonate platforms: Geology, v. 23, p. 625–628, doi: 10.1130/0091-
7613(1995)023<0625:BBAMFE>2.3.CO;2.
Wheeler, C.W., and Aharon, P., 1991, Mid-oceanic carbonate platforms as oce-
anic dipsticks: Examples from the Pacific: Coral Reefs, v. 10, p. 101–114,
doi: 10.1007/BF00571828.
Wilson, M.E.J., 2002, Cenozoic carbonates in SE Asia: Implications for equa-
torial carbonate development: Sedimentary Geology, v. 147, p. 295–428,
doi: 10.1016/S0037-0738(01)00228-7.
Wilson, M.E.J., and Bosence, D.W.J., 1997, Platform-top and ramp deposits of
the Tonasa Carbonate Platform, Sulawesi, Indonesia, in Fraser, A.J., et al.,
90
eds., Petroleum Geology of Southeast Asia: Geological Society [London]
Publication 126, p. 247–279.
Wilson, M.E.J., and Lokier, S.J., 2002, Siliciclastic and volcaniclastic influences
on equatorial carbonates; insights from the Neogene of Indonesia: Sedi-
mentology, v. 49, p. 583–601, doi: 10.1046/j.1365-3091.2002.00463.x.
Wilson, M.E.J., and Vecsei, A., 2005, The apparent paradox of abundant foramol
facies in low latitudes: Their environmental significance and effect on
platform development: Earth-Science Reviews, v. 69, p. 133–168, doi:
10.1016/j.earscirev.2004.08.003.
Woodhall, D., 1985, Geology of the Lau Ridge, in Scholl, D.W., and Vallier,
T.L., eds., Geology and Offshore Resources of Pacific Island Arcs—
Dorobek
Tonga region: Houston, Circum-Pacific Council for Energy and Mineral
Resources, Earth Science Series, v. 2, p. 351–378.
Yamaji, A., 1990, Rapid intra-arc rifting in Miocene, Northeast Japan: Tecton-
ics, v. 9, p. 365–378.
Zachariasen, J., Sieh, K., Taylor, F.W., Edwards, R.L., and Hantoro, W.S., 1999,
Submergence and uplift associated with the giant 1833 Sumatran subduc-
tion earthquake: Evidence from coral microatolls: Journal of Geophysical
Research, v. 104, p. 895–919, doi: 10.1029/1998JB900050.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Sediment waves in the Bismarck Volcanic Arc, Papua New Guinea

Gary Hoffmann
Eli Silver
Simon Day
Eugene Morgan
Earth Sciences Department, University of California, Santa Cruz, California 95064, USA

Neal Driscoll
Scripps Institution of Oceanography, La Jolla, California 92093, USA

Daniel Orange*
AOA Geophysics, Castroville, California 95012, USA

ABSTRACT

In the Bismarck Volcanic Arc in Papua New Guinea, six fields of sediment waves
were imaged with sonar. Sediment structures observed in seismic data and swath
bathymetry are not unique and can result from predominantly continuous (bottom)
currents, or episodic (turbidity) currents, or from deformation of sediment. Two of
these wave fields overlap and appear to be of turbidity-current origin and modified by
bottom currents, with one field unconformably overlying the other field. A field off the
coast of Dakataua caldera displays an arcuate morphology, and a series of enclosed
depressions within the field suggests creation by extensional deformation of rapidly
deposited sediment. Scour features in side-scan imagery suggest turbidity-current
activity, which also likely modifies the sediment waves. The wave field is isolated from
hyperpycnal currents, however, suggesting that in the absence of a shelf, coastal ero-
sion and small landslides can produce semiregular gravity-driven sediment flows that
deposit in deep (>1400 m) water. In Kimbe Bay a fourth sediment-wave field also dis-
plays arcuate morphology and enclosed depressions within the field. This wave field
is found within a bay >40 km from shore and also appears to have been formed by a
combination of extensional deformation of sediment and energetic current activity.
Two additional fields in Hixon Bay are fed by small and medium rivers (<~450 m3/s
mean annual discharge) draining volcanoes and mountainous regions. One small field
appears within a slide scar, suggesting that the initial topography of the scar provided
the conditions for early sediment-wave growth. A much larger field is best explained
by repeated hyperpycnal currents originating from the Pandi River. We cored a series
of upward-fining, graded sequences consistent with a turbidity-current origin. Ages
from these cores and measurements of relative thickness in sub-bottom imagery of the

*Present address: Black Gold Energy, Plaza Kemang Timor 22, Jakarta 12510, Indonesia

Hoffmann, G., Silver, E., Day, S., Morgan, E., Driscoll, N., and Orange, D., 2008, Sediment waves in the Bismarck Volcanic Arc, Papua New Guinea, in Draut,
A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special
Paper 436, p. 91–126, doi: 10.1130/2008.2436(05). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All
rights reserved.

91
92 Hoffmann et al.

field constrain deposition rates for the field and suggest that a large part of the Pandi
River discharge must be bypassing the shelf and depositing on the sediment-wave
field in deep water (>1200 m). These findings suggest that the sedimentary record
in arc collision zones will be dominated by mass-wasting deposits very close to vol-
canoes, and by river discharge depositing in select, extent regions far from shore.
Because sedimentation rates can vary by a factor of 2 between the two flanks of a
sediment wave, care must be taken when comparing bed thickness across an entire
sedimentary section.

Keywords: sediment waves, Papua New Guinea, Bismarck Sea, marine geology,
geomorphology.

INTRODUCTION giving us the rare opportunity to examine the distinguishing cri-

More than half of the total suspended sediment supplied by
rivers to the sea originates from small to medium mountain rivers
(<~450 m3/s mean annual discharge; Mulder et al., 2003), espe-
cially those in tectonically active regions such as the Bismarck
Volcanic Arc in Papua New Guinea (Milliman and Syvitski,
1992). At another active arc-continent collision, Taiwan, as much
as 42% of the sediment discharge to the ocean occurs by hyper-
pycnal flows (>40 g/L; Warrick and Milliman, 2003), typically
during flood events (Dadson et al., 2005). Large earthquakes can
cause an increase in the occurrence of hyperpycnal flows and the
percentage of sediment discharged at hyperpycnal concentrations
for some rivers for short intervals, owing to landslides (Warrick
and Milliman, 2003; Dadson et al., 2005). Under certain condi-
tions, these hyperpycnal turbidity currents can result in deep-sea
sediment waves (H.J. Lee et al., 2002; Wynn and Stow, 2002;
Schwehr et al., 2007).
The Bismarck Volcanic Arc differs from arc-collisional set-
tings such as Taiwan and other regions where turbidity-current
sediment waves have been identified (for example, the Var field
southeast of France; Migeon et al., 2001) by having active vol-
canism. Widely dispersed tephra may increase the frequency of
hyperpycnal discharges (Hayes et al., 2002), and ashfall from
volcanic eruptions may also directly produce turbidity currents
(Fiske et al., 1998). Bottom currents, rapid slope failure, and slow
soft-sediment deformation, for instance caused by repeated earth-
quake loading, are also processes that have been suggested as
being capable of producing or greatly modifying sediment-wave
fields (Gardner et al., 1999; Lee and Chough, 2001; H.J. Lee et
al., 2002; Wynn and Stow, 2002). O’Leary and Laine (1996), H.J.
Lee et al. (2002), and Wynn and Stow (2002) provide criteria to
distinguish among these various processes.
Because large undulations (hundreds of meters to a few kilo-
meters) are easily identifiable in sonar surveys of the seafloor, and
hyperpycnal turbidity currents are greatly affected by climate,
uplift, and seismicity (Mulder et al., 2003), turbidity-current-
generated sediment-wave fields should be ideal targets for study-
ing these factors in arc and arc-collisional environments. In the
Bismarck Sea we imaged multiple sets of large sediment waves,
teria using multiple wave fields within the same data set. We can
also test if these criteria can be consistently applied to differenti-
ate wave-forming processes, and what sediment waves formed by
these processes tell us about dispersal of sediment.
We begin with a discussion of the processes that create sedi-
ment waves and morphologically similar features. One of these
processes, turbidity currents, is surrounded by some contro-
versy, and so we proceed with an examination of the term and
its relation to hyperpycnal currents. We follow the discussion of
sediment-wave-generation processes and turbidity currents with
our specific observations and discuss them in terms of their dis-
tinguishing features and how they relate to climate and sediment
transport in the Bismarck Sea.
STUDY LOCATION
Papua New Guinea has two seasons, the summer monsoon
season and the winter trade-wind season. Although cyclonic
events are uncommon, near-continuous tropical rain produces
frequent landslides and consequent large sediment discharge to
the coastal ocean from the mountains, as well as by floods any
time of year (McAlpine and Keig, 1983; Pickup, 1984; Walsh
and Nittrouer, 2003). The sediment load for the entire island of
New Guinea is high, at ~1.7 × 109 t yr–1, roughly the estimated
load of all North American rivers combined (Milliman, 1995).
The Bismarck Volcanic Arc is located in northern Papua
New Guinea and forms the southern boundary of the Bismarck
Sea (Fig. 1). In the eastern half of the arc the Solomon Sea plate
is subducting beneath the Bismarck Sea plate at the New Britain
Trench (Johnson, 1979; Taylor, 1979). In the west, subduction
has ceased as the Finisterre and Adelbert terranes, parts of a rem-
nant Paleogene volcanic arc that lies in the Bismarck forearc, col-
lide with the Australian plate and are being uplifted to form the
Finisterre and Adelbert Mountains (Abbott et al., 1994; Pigram
and Davies, 1987).
The volcanism in the arc is related to subduction of the
Solomon Sea plate and occurs behind the remnant Paleogene
arc, which also includes the island of New Britain. The subma-
rine environment surrounding the Bismarck Volcanic Arc was
 Sediment waves in the Bismarck Volcanic Arc 93

144˚ 146˚ 148˚ 150˚ 152˚

-3˚ -3˚
ismic Zone
Bismarck Sea Se
Schouten Islands Bismarck Sea

-4˚ k R.
Manam Garove R -4˚
Sepi Ritter debris
TE
Karkar Dakataua KBE Fig 14
Tolokiwa flow GP
HB

WB
Fig 5 Fig 9
Adelbert Fig 2 L

F
-5˚ Mts. U Pandi -5˚
Ne Sakar WP KB B ts.
w Long G ai M River
Gu P K kan
ine Umboi Na
aF
old
Ra New Britain
mu
-6˚ an
dT
-M
ark Finisterre Ritter -6˚
hr ha Mts.
us m
tB Fa
elt ult ench
in Tr
100 km ew B
rita
-7˚ N -7˚
Solomon Sea
-8˚ -8˚
144˚ 146˚ 148˚ 150˚ 152˚
Figure 1. Location map showing regional setting of the wave fields of the Bismarck Volcanic Arc. The New Britain Trench accommodates
subduction of the Solomon Sea plate. New Britain and the Finisterre and Adelbert terranes are parts of a remnant Paleogene arc. The modern
subduction-related volcanic arc is offset to the north of the remnant arc. From west to east, some of the modern volcanoes include the Schouten
Islands, Manam, Karkar, Long, Tolokiwa, Umboi, Sakar, Ritter, Garove, Dakataua and the Willaumez Peninsula (WP), Pago (P), Karai (K), Gal-
losfulo (G), Bamus (B), Lolobau (L), Ulawun (U), and Rabaul (R), which lies on the northeast tip of the Gazelle Peninsula (GP) on the island of
New Britain. The Willaumez Peninsula is composed of a north-south string of volcanoes. The locations of several other volcanoes are shown as
small black circles. Major faults in the region include the Ramu-Markham Fault and the Wide Bay Fault (WBF). Two new faults were mapped
during this survey: the Kimbe Bay Escarpment (KBE) in Kimbe Bay (KB), east of the Willaumez Peninsula, and the Torkoro Escarpment (TE),
in the Hixon Bay region (HB) north of Lolobau. The outline of the debris flow from the 1888 collapse of Ritter Island is shown. The Sepik River
is the largest river on the north coast of Papua New Guinea. It drains a large part of the New Guinea Fold and Thrust Belt. The Pandi River drains
a large part of the Nakanai Mountains in New Britain, and the eastern flanks of Ulawun and Bamus volcanoes. The locations of Figures 2, 5, 9,
and 14 are shown.

the target of this study. We mapped submarine volcanic collapse that juts northward from New Britain into the Bismarck Sea,
features, including deposits of the 1888 Ritter Island collapse ending in Dakataua caldera. To the east of the Willaumez Pen-
(Johnson, 1987), for input into tsunami models (e.g., Ward, insula is Kimbe Bay, which is bounded on three sides by large
2001). The locations of a selection of volcanoes in the arc are volcanoes. It is also the location of a newly mapped fault, the
shown in Figure 1. The arc extends westward to the Schouten Kimbe Bay Escarpment.
Islands between 143° E and 145° E, where it meets the Bismarck Collapse blocks were found near several volcanoes dur-
Sea seismic zone. It extends east to ~152° E at Rabaul caldera ing the survey, including off Tolokiwa Island and off Dakataua
on the northeast tip of the Gazelle Peninsula. The study area caldera. At both these locations, sediment-wave fields were also
includes the Schouten Islands in the west but goes only so far mapped nearby. Sediment-wave fields were also mapped in
as the Hixon Bay area just west of the Gazelle Peninsula. The Kimbe Bay and the Hixon Bay region (boxed regions of Fig. 1).
Gazelle Peninsula is moving northwest relative to the rest of These sediment-wave fields form the basis of the current paper.
the island of New Britain along the left-lateral Wide Bay Fault
(Fig. 1; Madsen and Lindley, 1994). The Wide Bay Fault defines SEDIMENT-WAVE FORMATION AND DISTINCTIVE
the eastern edge of the Hixon Bay region, where we mapped a FEATURES
new fault, the Torkoro Escarpment.
The study area includes the deep-water (>500 m depth) Sediment waves, in the most common use of the term, refers
environment around the volcanoes in the arc, including on to any undulating pattern that is caused by differential sediment
either side of the Willaumez Peninsula, a string of volcanoes deposition and erosion over time, analogous to dunes or antidunes,
94 Hoffmann et al.
as opposed to undulating patterns that are caused by slope failure
such as slow deformation or slide events (H.J. Lee et al., 2002;
Wynn and Stow, 2002). Large undulations with wavelengths of
100 m to >10 km are observed in a variety of settings, including
on channel levees, axes, and mouths or in canyons (Normark et
al., 1980; Kidd et al., 1998; Nakajima and Satoh, 2001; Migeon
et al., 2001; Lewis and Pantin, 2002); on the flanks of volcanic
islands (Wynn et al., 2000); in troughs (Howe, 1996); and on pla-
teaus and continental slopes (Lee and Chough, 2001; O’Leary
and Laine, 1996). Short-wavelength (0.1–1 m) sediment waves
have been observed in settings such as terrestrial rivers (Mohrig
and Smith, 1996; Jerolmack and Mohrig, 2005a, 2005b; Cole-
man et al., 2005). Owing to imaging limitations in the deep-sea
environment, however, we confine our current discussion to large
sediment waves.
Two processes are recognized as producing sediment
waves as defined here, namely turbidity currents and bottom
currents (Wynn and Stow, 2002; Schwehr et al., 2007). Waves
of turbidity-current origin are created by numerous episodic
events preferentially depositing sediment on the upstream flank
of the waves in antidune fashion when the flow is in the Froude
number regime of ~0.5–1.9 (Hand, 1974; Bowen et al., 1984;
Wynn et al., 2000; H.J. Lee et al., 2002). The Froude number is
the ratio between current speed and the speed of gravity waves
in a density-stratified fluid, such as a sediment-laden-current
layer in seawater. Recent research suggests that initiation of
a turbidity-current sediment-wave field may require breaks in
slope or preexisting undulations, but once started, these wave
fields may grow on their own (Ercilla et al., 2002; H.J. Lee et
al., 2002; Kubo and Nakajima, 2002).
Deep-sea sediment waves caused by turbidity currents gen-
erally have wavelengths of 200 m to 7 km, and wave heights of
2–70 m, and occur on slopes ranging from 0.1° to 0.7° (Wynn and
Stow, 2002; H.J. Lee et al., 2002). In profile, the waves migrate
upstream and upslope because of preferential deposition on the
upstream flank (Wynn and Stow, 2002; H.J. Lee et al., 2002).
Internal reflections are commonly continuous between waves
(H.J. Lee et al., 2002). Wave crests tend to be oriented perpen-
dicular to the local slope, as turbidity currents travel downslope
(Wynn and Stow, 2002; H.J. Lee et al., 2002; Schwehr et al.,
2007). Wave sequences progressively thin downslope, wave
heights can decrease downslope, and some reflections pinch out
downslope in profile, since sediment concentration and depo-
sition in turbidity currents progressively decrease downslope
(Mulder et al., 1998). Thus deposition in the distal end will be
from large events that occur only rarely (H.J. Lee et al., 2002).
Wave crests tend to be linear or sinuous in plan view, with some
bifurcation (Wynn and Stow, 2002; H.J. Lee et al., 2002). Waves
formed by turbidity currents are often associated with other evi-
dence of turbidity-current activity, such as channels, levees, and
scour features. In their review, H.J. Lee et al. (2002) note that the
overall cross-sectional shape of a sediment-wave field is com-
monly concave upward.
The generally accepted concept of bottom-current sediment-
wave generation is that internal waves induced within a current
in a density-stratified medium flowing over topography will dif-
ferentially deposit and erode sediment on the seafloor (Flood,
1988). Sediment waves formed by bottom currents have simi-
lar dimensions as turbidity-current-generated waves (Wynn and
Stow, 2002). Wave crests tend to be linear or sinuous with bifurca-
tions (Wynn and Stow, 2002). Bedforms often migrate upcurrent
but can also migrate downcurrent (Wynn and Stow, 2002). They
are distinct from turbidity-current-generated sediment waves in
that they can form on very low slopes or flat seafloor, and wave
crests are often aligned obliquely to the slope because the cur-
rents are contour-parallel (Wynn and Stow, 2002). Both bottom-
and turbidity-current-generated waves tend to display asymmetry
between flanks in backscatter imagery (H.J. Lee et al., 2002).
Undulating patterns can also arise from deformation, how-
ever. These can occur from extension (O’Leary and Laine, 1996)
or compression (Lee and Chough, 2001; Hill et al., 2004). These
undulating patterns may be related to rapid slope failure, such as
slumps (H.J. Lee et al., 2002), or to slow deformation caused by
intermittent earthquake loading (Lee and Chough, 2001). These
features can be up to 10 km in wavelength, and up to 100 m in
wave height (Wynn and Stow, 2002), and they can occur on steep
slopes or very shallow slopes (<0.5°; Lee and Chough, 2001).
In profile, deformational features are commonly associated with
faulting (O’Leary and Laine, 1996; H.J. Lee et al., 2002). Plan-
form imagery of these features is rare, but wave crests may be
arcuate, without bifurcation (Wynn and Stow, 2002).
Bottom currents, turbidity currents, and deformation can all
interact in the creation of sediment-wave fields (Howe, 1996;
Faugères et al., 2002; H.J. Lee et al., 2002). In this manuscript
we will try to tease out the dominant and subordinate processes
responsible for creating these morphologic “sediment-wave”
features.
TURBIDITY FLOWS AND HYPERPYCNAL
CURRENTS
Considerable controversy exists regarding the nature of
flow and deposits of turbidity currents, as well as the terms
used in the literature that apply to various sediment transport
regimes (Kneller and Buckee, 2000; Shanmugam, 2000; Mul-
der and Alexander, 2001; Mulder et al., 2001; Shanmugam,
2002; Mulder et al., 2002). For the sake of clarity in our present
discussion, we note an overlap between the terms hyperpycnal
current and turbidity flow, and we define our particular usage
in this paper.
Following Mulder and Alexander (2001), hyperpycnal cur-
rent refers to sediment-laden river discharge that is denser than
the body of water into which it enters. The flow can travel long
distances in a semidiscrete layer, entraining fluid and possibly
additional sediment if it is exerting shear stresses on the bed
high enough to cause erosion (Skene et al., 1997; Mulder and
 Sediment waves in the Bismarck Volcanic Arc
Alexander, 2001; Mulder et al., 2003). For river discharge enter-
ing seawater, the sediment concentration needs to be at least
~40 kg m–3 for the flow to be negatively buoyant (Mulder and
Syvitski, 1995; Warrick and Milliman, 2003). Depending on the
river, hyperpycnal-discharge events can occur about once every
100 yr, or as often as once per year (Mulder et al., 2003).
Turbidity flow refers to sediment suspended primarily by the
upward component of turbulence (Lowe, 1982; Middleton, 1993;
Stow et al., 1996; Shanmugam, 2000; Mulder and Alexander,
2001). Such flows contain suspended sediment concentrations
up to 10% by volume (Shanmugam, 2000; Mulder and Alexan-
der, 2001). They can be generated by a number of mechanisms,
such as entrainment of sediment into surrounding seawater at
the upper surface of a debris flow. Other mechanisms include
hyperpycnal river discharge, and concentration processes in a
low-density layer (Lowe, 1982; Middleton and Hampton, 1976;
Parsons et al., 2001) such as a layer of tephra fallout during erup-
tions (Fiske et al., 1998). Coastal erosion during storm events
that affect easily eroded sequences could also generate turbidity
currents (Hayes et. al., 2002). Scully et al. (2002), for instance,
showed that energetic water waves increase the capacity for crit-
ically stratified gravity flows to transport sediment. Combined
with direct turbidity-current generation from ashfall and indirect
formation from reworking of tephra, the presence of volcanoes
may significantly increase the occurrence of turbidity-current
generation in this setting over other settings. We will therefore
use turbidity current to refer to a sustained turbidity flow gen-
erated by any mechanism, and hyperpycnal current to refer to
turbidity currents generated owing to hyperpycnal discharge at
river mouths.
METHODS
Multibeam bathymetric data were collected aboard the R/V
Kilo Moana on cruise KM0419, using a hull-mounted SIMRAD
EM-120 echo sounder. Backscatter data were collected using a
towed MR1 side-scan sonar system. Side-scan data were grid-
ded in a 16 m grid. Multibeam bathymetry and backscatter data
were collected along the 1000 km length of the arc in a swath
up to 100 km wide. Bathymetric data were median filtered, using
a 500 m kernel. High-resolution Compressed High Intensity
“Radar” Pulse (CHIRP) sub-bottom sonar data were collected
with an Edgetech 1–6 kHz swept frequency with a 50 ms dura-
tion. The CHIRP data are unmigrated. This system transmitted
approximately every 5–10 m along selected transects. Penetration
was up to a few tens of meters. Twelve short (<2 m) gravity and
piston cores were collected, primarily targeting volcanic-collapse
debris flows identified in side-scan imagery.
We examine four sediment-wave fields that were identified
in the Bismarck Arc in order to determine the various genera-
tion mechanisms of the sediment-wave fields. Doing so will help
us understand how material is transported through the arc sys-
tem and how important slope failure is compared to bottom and
95
turbidity currents. We begin in the central part of the arc north
of Tolokiwa, and then continue eastward to look at Dakataua,
Kimbe Bay, and Hixon Bay (Fig. 1).
TOLOKIWA WAVE FIELDS
Tolokiwa Island volcano, near the center of the Bismarck
Volcanic Arc (Fig. 1), has undergone at least one lateral collapse
in the Holocene. The sediment waves occur in a field north of the
island on the northern side of a broad channel ~8 km wide and
40 m deep that grades at ~0.2° to 0.5° from west to east into the
basin (Fig. 2).
Two sediment-wave fields overlap in this region. The first is
composed of large sediment waves with wavelengths of 600 m
to 1 km and wave heights of ~10 m to 40 m (Fig. 3). The second
field is composed of smaller sediment waves with wavelengths
of 100 m to 400 m and wave heights of a few meters (Fig. 3;
Fig. 4C, E). Waves in both sets are identifiable in side-scan
images as having a distinctly higher backscatter along the eastern
flanks (Fig. 3).
The field of large sediment waves is overlapped by the field
of small sediment waves, with some small waves appearing
between and overlying large waves (Fig. 3; Fig. 4C). The crests
of the small waves are oriented slightly oblique to their local
slope. This obliquity varies from wave to wave but is <30°. The
crests of the large waves are perpendicular to the local slope.
The high-backscatter eastern, downslope flanks of the waves
correspond to the thinner, more steeply dipping beds of the waves
(Fig. 4). The large sediment waves commonly show differential
deposition, with the upslope side receiving a factor of 2 or more
accumulated sediment than the downslope side. The crests of the
sediment waves migrate upslope through time (Fig. 4A, B). The
small waves lap onto the large waves (Fig. 4C, D), and they over-
lie parallel reflectors that begin at a depth of 5 m (Fig. 4E).
DAKATAUA WAVE FIELD
The Dakataua wave field resides on the northern flank of
Dakataua caldera. The field is 24 km by 18 km and elongated
N-S (Fig. 5). The average north-dipping slope of the field is 1.4°,
with the upslope flanks of the waves dipping south between 1°
and 1.5°, and the downslope flanks dipping north ~10°. The wave
field as a whole forms a topographic high of up to 200 m relative
to the flatter seafloor that surrounds the wave field. The seafloor
to the east of the wave field is deeper than that to the west. This
difference decreases northward, although north of the 240-m-
high ridge at the north end of the field the seafloor maintains an
eastward-dipping component of ~0.15° (Fig. 5).
Between the caldera and the wave field, the seafloor dips
smoothly at ~5°. In the side-scan data the waves generally do
not exhibit asymmetry; however, the downslope flanks nearest
the caldera display high backscatter (Fig. 6). Scour features are
apparent on the 5° slope leading into the wave field (Fig. 6).
 Fig 4

Ritter debris flow

es
Large Sed Waves av
dW
Se
all
Sm

Collapse
Blocks Fig 3

5 km

No Tolokiwa
Data Island

Figure 2. Bathymetry of Tolokiwa sediment-wave fields and surroundings. Two distinct fields overlap—one with wavelengths of 0.6–1 km,
labeled large sed waves, and one with wavelengths of 200–400 m, labeled small sed waves. Locations of Figures 3 and 4 are shown. Bathymetric
contour interval is 25 m, with annotated contours in bold every 200 m.
 Ritter
debris
flow
Fig 4

Large waves
Small waves

147˚ 36' 147˚ 39' 147˚ 42' 147˚ 45' 147˚ 48'
-5˚ 02' -5˚ 02'
0
17860
17
0

40
172

17

Ritter
1820
1800

2 km debris
0

flow
170

-5˚ 04' -5˚ 04'

Fig 4

-5˚ 06' -5˚ 06'

Large waves
Small waves

-5˚ 08' -5˚ 08'

147˚ 36' 147˚ 39' 147˚ 42' 147˚ 45' 147˚ 48'
Figure 3. Side-scan mosaic (top) and bathymetry (bottom), contoured at 2 m intervals of Tolokiwa sediment-wave fields. Location of Figure 4
shown. The waves are apparent in the side-scan imagery as features with higher acoustic backscatter (darker shades) in the western flanks than
eastern flanks. Insonification direction is north-south. Nadir lines run east-west at 5°02′ S and 5°06′ S. Black east-west lines are sea-surface
reflection artifacts. In both parts of the figure, short dashed lines show local trend of slope. In side-scan image, several wave crests are denoted
with short, thin lines to highlight the slight obliquity of the crest-normal to the slope. Dotted lines outline the sediment-wave fields. Note in the
side-scan image that some small waves appear within the large waves in the region of overlap.
 upslope migration (Left) The upslope flank is 9.2 m thick, measured to the lowest strong
reflection; the downslope flank, measured to the same reflection, is as thin
as 4.5 m. These measurements assume a sound velocity of 1500 m/s.

(Below) Small sediment waves onlapping

large sediment waves

W
A E

2.38
15 m VE = 17
2.40
500 m
2.42 D
TWT (s)

2.44

2.46
2.48
2.50
2.52

5m

125 m

B E
Note that reflections on the upslope sides Parallel reflections beneath small wavelength
of the waves thin or pinch out downslope sediment waves

(Left) Small sediment waves appearing

in between two large sediment waves,
and onlapping them.

Figure 4. CHIRP profile of eastern end of sediment-wave field north of Tolokiwa Island where the field merges with the ripple field. Location
shown in Figure 2. VE—vertical exaggeration; TWT—two-way traveltime. Enlarged boxes demonstrate particular features of note. All enlarged
boxes are at the same scale, which is 4 times larger than and at the same vertical exaggeration as the base figure.
 150 00© 150 05© 150 10© 150 15©
-4 40© -4 40©
Dakataua Lineation

20
00 10 km
Dakataua
Wave Field

20
4 45© -4 45©

00
Fig. 7

Fig. 8

20
00
4 50© -4 50©

1800
Collapse Block
Field
0

16
180

00

4 55© -4 55©
1400
00
16
northwest
Wave ridge
Field

5 00© Fig. 6
-5 00©
No
Data ua
ata
ak
L. D

Dakataua Caldera

150 00© 150 05© 150 10© 150 15©

Figure 5. Location map showing wave field in relation to Dakataua caldera and bathymetry (in meters). Contour interval is 25 m, with annotated
contours in bold every 200 m. Locations of Figures 6, 7, and 8 are shown. A lineation, shown as the dash-dot line, runs through the field approxi-
mately N-S. The lineation ends slightly west of the 240-m-high ridge at the north end of the field.
 150˚ 00' 150˚ 05' 150˚ 10'

4 km
Dakataua Lineation

-4˚ 45' -4˚ 45'

Fig. 7

Fig. 8
-4˚ 50' -4˚ 50'

collapse
blocks

High backscatter
downslope
flanks
-4˚ 55' -4˚ 55'

northwest ridge High backscatter scour streaks

150˚ 00' 150˚ 05' 150˚ 10'

Figure 6. Side-scan mosaic of Dakataua sediment wave field. Locations of Figures 7 and 8 shown. Scale is 1.6 times that of Figure 5. High
backscatter is dark. The two elongated regions of high backscatter in the field (~4°52′ S and 4°53′ S) are in the troughs between waves. Solid
lines denote the slope break between wave crests and steep downslope flanks of waves. Dashed lines denote enclosed depressions. The Dakataua
lineation is shown as the dash-dot line.
 Sediment waves in the Bismarck Volcanic Arc
The waves themselves are 1–4 km in length, with wave
heights of ~50 m, as measured from the upslope trough to the
crest. In plan view they are irregular, although generally arcu-
ate, with crests largely continuous across the width of the field
(Fig. 6). The crests are sinuous, and some are bifurcate (Figs. 6
and 7), and they also tend to be wider than the troughs between
waves (Fig. 7). Many of the troughs between waves contain
enclosed depressions (Figs. 6 and 7). Some of the depressions are
associated upslope with steep slopes that display high backscat-
ter, but many are not (Fig. 6), and the depressions are not associ-
ated with high backscatter scour.
A lineation runs N-S through the field (Fig. 5), characterized
by eastward-dipping slopes, widened depressions, and pinched-
out wave crests (Fig. 7). The CHIRP failed to penetrate more
than 1 m throughout this field (Fig. 8).
KIMBE BAY WAVE FIELD
Whereas the Tolokiwa and Dakataua waves are both iso-
lated from large inputs of fluvial sediment, Kimbe Bay is fed on
three sides by a multitude of small rivers that drain volcanoes.
The Kimbe Bay wave field lies near the center of the bay, almost
50 km from shore, adjacent to the Kimbe Bay Escarpment.
The Kimbe Bay wave field is >30 km long and 15 km wide
(Fig. 9). In plan view the waves are generally arcuate, although
irregular, with wavelengths varying from 1 to 4 km and wave
heights reaching >80 m in some areas (Fig. 9). The average slope
of the wave field is 1°. Within the field, many enclosed depres-
sions attain a depth of ~100 m and are found between wave crests
(Fig. 10). Some low-wave-height features (~10 m) are found
west of the main sediment-wave field. The main wave field lies
on a topographic high of up to 250 m relative to the sea floor
to the west (Fig. 9). To the east, a crevice ~2 km wide and up
to ~100 m deep separates the wave field from the Kimbe Bay
Escarpment. To the south of the wave field is a topographic ridge
at least 200 m high (Fig. 9). Several channels lead into the crev-
ice between the wave field and the escarpment. One channel is
associated with high backscatter as it cuts through the ridge east
of the Kimbe Bay Escarpment (Fig. 11), then drops 200 m into
a plunge pool (S.E. Lee et al., 2002), which is >100 m deep as
measured from the downslope side of the channel (Fig. 9).
The crevice increases in width northward, to nearly 6 km,
where depressions appear (Fig. 9). The depressions and the
escarpment appear as very high backscatter features in side-scan
imagery (Fig. 11). The high backscatter regions within these
depressions occur on the southern slopes, and the depressions
attain depths up to 75 m (Fig. 12). A lineation is contrasted against
the background sediment of the wave field, appearing as a higher
backscatter region nearly 10 km long and 1 km wide (Fig. 11).
The steeper, downslope sides of the sediment waves pro-
duce higher backscatter than the rest of the wave field (Fig. 11).
The slopes of the high backscatter faces of the waves approach
101
10° (Fig. 10). The high backscatter of the downslope sides of
the waves is associated in the CHIRP profiles with thinning
sequences and upslope migration (Fig. 13). But in most cases
(Fig. 13A, B, D) the upslope migration is apparent only on the
steep downslope flanks, and only in a few areas can upslope
migration be seen clearly across a whole wave, which might be
due to cutting across some waves obliquely (Fig. 13C). Addition-
ally, the character of the reflections varies from wave to wave,
especially in terms of penetration depth. Some wave crests are
imaged to a depth of ~8 m (Fig. 13B, C), whereas wave crests
both upslope and downslope of these crests are imaged only to a
depth of 2–3 m (Fig. 13A, D).
HIXON BAY
Hixon Bay allows us to test whether fluvial sedimentation is
an important process far from shore in this system. It is several
kilometers wide and the receptacle for the Pandi River (Fig. 14).
This river drains ~800 km2 of the Nakanai Mountains and the
eastern flanks of Ulawun and Bamus volcanoes (Fig. 1).
Backscatter, Morphology, and Sub-bottom Profiles of the
Hixon Bay Sediment-Wave Field
A large sediment-wave field is found in the Tokoro Trough
of Hixon Bay (Figs. 14–16). This trough attains depths >1900 m,
and two channels feed into the trough from the Pandi River
(Fig. 14). The sediment-wave field (Fig. 15) begins shortly below
the point at which channel scour diminishes between Torkoro and
Mele Reefs. A slide scar is present east of Lolobau Ridge and is
shown in more detail in Figure 17.
In side-scan images (Fig. 18), high backscatter scour fea-
tures highlight the Pandi River channels. The slide scar shows
a stippled pattern of high backscatter in the upper part, and the
edges are highlighted as high backscatter streaks in the lower
half of the slide scar. The sediment-wave field (Fig. 18) appears
in backscatter as a pattern of alternating higher and lower back-
scatter, with regions of varying wavelengths. We divide the wave
field into six groups on the basis of morphology (Fig. 16), and we
summarize their characteristics in Table 1.
The six groups of waves are also distinct from one another
in sub-bottom imagery (Table 2). We correlate a transparent
layer between stronger reflections across group 2 and into group
4 (Fig. 19). Reflection thickness variations above this layer
indicate that wave-crest deposition rates decrease monotoni-
cally downslope. The thickness varies by a factor of 1.5 across
the group 2 waves, and by nearly a factor of 2 from group 2 to
group 4 (Figs. 19–21). We also measured the relative thickness
of reflector sequences at the distal end of the wave field, tak-
ing advantage of continuous reflectors in group 4 (Fig. 22D).
From the upslope side to the downslope side of Figure 22D the
sequence thins by ~20%.
 150˚ 04' 150˚ 06' 150˚ 08'
-4˚ 46' -4˚ 46'
19
20

1 km 19
0 80
196
1940 1960
1900 1940
1880 192
0
1860
bifurcation
1880
-4˚ 48' -4˚ 48'
1900

1860 1880

Fig. 8
18
20
0
184 1900
1860
18
0 40
1860 180
-4˚ 50' 1820 -4˚ 50'
17

19
Dakataua Lineation

80

00
182
bifurcation 0
180
0
80

bifurcation
17

182
0
0
176
80

-4˚ 52' 1780 1740 -4˚ 52'

16

0
172

1700

150˚ 04' 150˚ 06' 150˚ 08'

Figure 7. Bathymetry of Dakataua wave field, contoured at a 4 m interval, with annotated contours in bold every 20 m.
Location of CHIRP line depicted in Figure 8 shown. The downslope sides are consistently steeper than the upslope sides
of the features. Note the bifurcation near the center of the CHIRP line shown in Figure 8 and elsewhere. The Dakataua
lineation is characterized by eastward-dipping slopes, widened depressions, and pinched-out wave crests.
 5m
CHIRP penetration is about 1 m.
250 m Diffractions give the illusory
appearance of deeper penetration.

S N

30 m VE = 13.9
2.40
1 km
2.44

2.48
T W T (s)

2.52

2.56
Data Gap

2.60

2.64

Downslope, the subsurface remains

largely impenetrable.

Figure 8. CHIRP profile of Dakataua wave field. Location shown in Figure 7. VE—vertical exaggeration; TWT—two-way traveltime. Detailed
enlargements are at the same vertical exaggeration and 3.5 times the scale as the base figure. Both demonstrate the extremely limited penetration
(~1 m) that characterizes this wave field.
 Fig 12

Depressions

K imb e Bay Escarp me nt

Kimbe Wave Field

Plunge Pool
Fig 10

Crevice Channels
Fig 13

10 km
No Data

Figure 9. Bathymetry of the Kimbe Bay sediment-wave field. Contour interval is 25 m, with annotated contours in bold every 200 m. Side-scan
mosaic of the same region at the same scale is shown in Figure 11. Motion along the newly discovered Kimbe Bay Escarpment is unknown. In
the region between the escarpment and the wave field are several depressions associated with high backscatter. The outlines of these depressions
correspond to regions of high backscatter. Detailed bathymetry and backscatter imagery of the pull-apart basins are shown in Figure 12. Loca-
tions of Figures 10 and 12 are shown. CHIRP line depicted in Figure 13 is shown.
 150˚ 38' 150˚ 40' 150˚ 42' 150˚ 44'

0
180
1 km
-5˚ 00' -5˚ 00'

18
1825

00
00
18
5
1775 180
0 177
00
175
0 18

Fig 13

17
-5˚ 02' 1725
-5˚ 02'

75
1700 1750
175
25

0
17

1700

17
1675

25
1675

1725 1675
1700

50
16

-5˚ 04' 00 -5˚ 04'

16 17
00
1625
16
75

16
1625
50
00
16

1675
16
00

-5˚ 06' -5˚ 06'

172
5

150˚ 38' 150˚ 40' 150˚ 42' 150˚ 44'

Figure 10. Detailed bathymetry of Kimbe Bay sediment-wave field. Contour interval is 5 m, with annotated contours in bold every 25 m. CHIRP
line depicted in Figure 13 is shown; the scale of this line is 2.9 times that of Figures 9 and 11. Note the pronounced irregular morphology of
the waves, with an overall arcuate pattern of alternating crests and troughs; superimposed on this is a chaotic assortment of peaks and valleys,
resulting in highly sinuous structures.
 150˚ 30' 150˚ 35' 150˚ 40' 150˚ 45' 150˚ 50'
-4˚ 50' -4˚ 50'
Fig 12

Lin
ea
tio
n
-4˚ 55'

K imb e Bay Escarp me nt

-4˚ 55'

Fig 10 Plunge Pool

-5˚ 00' -5˚ 00'

Crevice
Fig 13

Channel
Scour
-5˚ 05' -5˚ 05'
10 km

-5˚ 10' -5˚ 10'

150˚ 30' 150˚ 35' 150˚ 40' 150˚ 45' 150˚ 50'
Figure 11. Side-scan mosaic of the same region depicted in Figure 9, and at the same scale. Dotted outline of field same as in Figure 9. Loca-
tions of Figures 10 and 12 are shown. CHIRP line depicted in Figure 13 is shown. Low backscatter is light, and high backscatter is dark. High
backscatter pattern within the wave field is associated with troughs. High-backscatter lineation is oriented 33.7° with respect to the general trend
of the Kimbe Bay Escarpment. Solid lines denote the slope break between wave crests and steep downslope flanks of waves. Dashed lines denote
enclosed depressions.
 150˚ 38' 150˚ 40' 150˚ 42' 150˚ 44'
-4˚ 50' -4˚ 50'

-4˚ 51' -4˚ 51'

-4˚ 52' -4˚ 52'

-4˚ 53'
Lin

-4˚ 53'
ea
tio
n

-4˚ 54' -4˚ 54'

1 km

150˚ 38'
150˚ 38' 150˚ 40'
150˚ 40' 150˚ 42'
150˚ 42' 150˚ 44'
150˚ 44'
-4˚ 50' -4˚ 50'

00
20
19
2050 75 165
19 0
50
-4˚ 51' 50 -4˚ 51'
20

2025

20
202 75
5
-4˚ 52' 50 -4˚ 52'
20 20
25

-4˚ 53' 20 -4˚ 53'

Lin

00
ea

25
tio

20
n

-4˚ 54' -4˚ 54'

1 km

150˚ 38' 150˚ 40' 150˚ 42' 150˚ 44'

Figure 12. Close-up side-scan mosaic and bathymetry of enclosed depressions in the crevice, and the backscatter lineation noted in Figure 11.
Contour interval is 5 m, with annotated contours in bold every 25 m. Scale is 2.8 times that of Figures 9 and 11.
The waves display some downslope thinning (B) and
apparent upslope migration on their downslope B
flanks (A and B). The irregularity of the morphology
(see Figure 12) results in the CHIRP profiles imaging 10 m
reflections from adjacent surfaces at different 500 m
depths (A).
seafloor reflections
from different locations

upslope
A migration

sequence thins
upslope migration downslope

S N

2.20
30 m VE = 15
1 km
2.30
T W T (s)

2.40

2.50

upslope
C migration D 10 m
500 m

upslope
migration

Apparent upslope migration is present in most of the waves (C and D). In some cases this is true for the whole wave (C), but for
most of the waves, upslope migration is only apparent on the steep, downslope flanks where it becomes difficult to distinguish
between internal structure and diffractions (A, B, and D). Penetration varies widely from wave to wave, with some crests imaged
to a depth of up to 6 to 8 m (B and C), and others only imaged to a depth of 2 to 3 m (A and D), suggesting sedimentation
conditions are not consistent across the wave field.
Figure 13. CHIRP profile across a part of the Kimbe Bay wave field. Location shown in Figures 9, 10, and 11. VE—vertical exaggeration;
TWT—two-way traveltime. Detailed enlargements are at 4.2 times the scale and at the same vertical exaggeration as the base figure.
 151˚ 10' 151˚ 15' 151˚ 20' 151˚ 25' 151˚ 30' 151˚ 35'
-4˚ 35' No Data
-4˚ 35'
Lolobau
ent

1 800
00
Trough
rpm

16
No Data
Esca

Torkoro
Trough
e
oro

idg

-4˚ 40' -4˚ 40'

R
Tork

oro

Eas
ork

00
12

t To
st T

rko
14
We

00

ro R
sediment waves

idg
e
No Data

00
-4˚ 45' Mele -4˚ 45'
16
e Rf
slid Mele
12
00 Seamount

Fig. 17

-4˚ 50' Fig. 16 -4˚ 50'

Torkoro Reef
Lolobau Ridge 5 km
Channels No Data

-4˚ 55' Lolobau -4˚ 55'

Island Hixon Bay

Pandi New Britain

Ulawun Volcano River
-5˚ 00' -5˚ 00'
151˚ 10' 151˚ 15' 151˚ 20' 151˚ 25' 151˚ 30' 151˚ 35'
Figure 14. Bathymetry of the Hixon Bay region. Contour interval is 25 m, with annotated contours in bold every 200 m. Several channels feed
into the imaged region, including those from the Pandi River, which drains part of the Nakanai Mountains (see Fig. 1) and Ulawun volcano. The
Pandi River channels fade out as they enter the region between the Mele Seamount and Torkoro Reef. This is where a major sediment-wave field
begins. This field is shown in more detail in Figure 16, the location of which is shown here. A slide scar is observed on the slope east of West
Torkoro Ridge and Lolobau Ridge. This scar is shown in more detail in Figure 17, the location of which is shown here.
 151˚ 10' 151˚ 15' 151˚ 20' 151˚ 25' 151˚ 30' 151˚ 35'
-4˚ 35' -4˚ 35'
Ch
an
ne
ls
Lolobau co
ent

ur
Trough
m
carp

Torkoro
Trough
Es

e
oro

idg

-4˚ 40' -4˚ 40'

R
Tork

Large sediment waves

oro

Cha
Lower slide nne

Eas
l sco
ork

scar ur

t To
st T

rko
We

Small sed wvs

ro R
idg
Upper slide

e
scar
-4˚ 45' Mele -4˚ 45'
lower Rf
backscatter

Ch
Mele

an
Seamount

ne
stippled pattern

ls
co
u r
Fig. 17

-4˚ 50' Fig. 16 -4˚ 50'

Torkoro Reef
Lolobau Ridge Channel scour 5 km

-4˚ 55' Lolobau -4˚ 55'

Island Hixon Bay

Pandi New Britain

Ulawun Volcano River
-5˚ 00' -5˚ 00'
151˚ 10' 151˚ 15' 151˚ 20' 151˚ 25' 151˚ 30' 151˚ 35'
Figure 15. Hixon Bay area, with the location and scale the same as for Figure 14. Low backscatter is light, and high backscatter is dark. Regions
without data are white. The slope of the Torkoro Escarpment is notably high in backscatter. The slide scar shows up clearly as a pattern of high
backscatter. The backscatter pattern is distinctly different between the upper slide scar and the lower slide scar. The wave field within the box
locating Figure 16 can be seen as a repetitive pattern of higher and lower backscatter. Both large sediment waves and small sediment waves can
be seen. The large sediment waves are seen as a longer wavelength, lower contrast pattern, whereas the small sediment waves are seen as a shorter
wavelength, higher contrast pattern. Channels are associated with high-backscatter scour.
 151˚ 18' 151˚ 20' 151˚ 22' 151˚ 24' 151˚ 26'

1650
-4˚ 38' -4˚ 38'

Fig 22
18

BB08
50
17
75

-4˚ 40' -4˚ 40'

1825

BB09
17

1800
50
17

1725 1675
00

1675 1775
-4˚ 42' 50 -4˚ 42'
17

1700
1725
0
165

Group 5
5

Group 6
162

75
1700 16

-4˚ 44' Group 4 1650 -4˚ 44'

r
sca 0
de 160 21
Sli Fig
15

1625
1550 25
75

4
1525
0 1

1600
Gr

1 km
140
ou

75
15 Seamount
p3
1500

50
-4˚ 46' 15 -4˚ 46'
1500

1475
1350

145
0

Group 2
13
50

1300 25
14

50
12
Fig

00
-4˚ 48' 12 -4˚ 48'
Fig
19

0 142
115 5
20

0 140
110 0
0 1375 Group 1
100

Torkoro Reef

151˚ 18' 151˚ 20' 151˚ 22' 151˚ 24' 151˚ 26'
Figure 16. Bathymetry of the Hixon Bay main sediment-wave field. Contour interval is 5 m, with annotated contours in bold every 25 m. Sedi-
ment waves are divided into five groups, based on morphology. Locations of CHIRP profiles through sediment waves in Figures 19 through 22
are shown. Arrows highlight the channels that lead to the sediment-wave field. See text for details on the groups of waves. Core holes BB08 and
BB09 are shown.
112 Hoffmann et al.

151˚ 14' 151˚ 15' 151˚ 16' 151˚ 17'

14
125 14 25
0 00
13
75
-4˚ 44' -4˚ 44'

13
0.5 km

50
region of undulating bathymetry

25
1175
13
-4˚ 45' -4˚ 45'
1300
Figure 17. Detailed bathymetry of the slide
scar in the Hixon Bay region. Contour inter-
1275

val is 5 m, with annotated contours in bold

p
12

ar every 25 m. The slide scar and scarp are de-

50

-4˚ 46' Sc -4˚ 46' lineated with the dotted line. In the upslope
portion of the scar, an undulating pattern is
122

seen. A CHIRP line that cuts across these

5

features is shown in Figure 25.

-4˚ 47' 26 -4˚ 47'

Fig 0
120

-4˚ 48' -4˚ 48'

151˚ 14' 151˚ 15' 151˚ 16' 151˚ 17'

Gravity Cores from the Hixon Bay Sediment-Wave Field
We collected two gravity cores to examine the nature of sedi-
ment being deposited and its rate of deposition. The core locations
are shown in plan view in Figure 16 and in cross section in Fig-
ure 22. Core BB09 was taken in the trough between two waves,
just upslope of a 1-km-wavelength, 20-m-high wave. The slope
on which it is located dips <0.05° to the south. Core hole BB08 is
3.5 km downslope from core hole BB09 on the downslope flank
of a wave near the very end of the field. The wave has a length of
1 km and a height of 10 m. The north-dipping flank of the wave
on which core hole BB08 is located has a slope of 1°. Core BB09
is 164 cm long, and core BB08 is 85 cm long (Fig. 23). Depths in
the cores are not corrected for compaction during collection.
Core BB09 penetrated through a unit of gray clay containing
pumice dropstones (~130 cm; Fig. 23). Core BB08 penetrated
into, but not through, this same unit (~80 cm). In both cores a
series of silty and sandy tephras is interbedded with fine silt, silty
clay, and clay units (Fig. 23). The well-sorted fine-black-sand
tephra at 105 cm depth in core BB09 correlates physically with
the well-sorted fine-black-sand tephra at 53 cm depth in core
BB08 (Fig. 23). If this correlation holds, then the sedimentation
rate between the clay-pumice unit and this tephra unit was slightly
lower in core BB09 than in core BB08. Subsequently, however,
sedimentation would have been more rapid by nearly an order of
magnitude in core BB09 than in core BB08, because of the cor-
relation of the well-sorted fine white and black sand unit at 65 cm
depth in core BB09 and at 42 cm depth in core BB08 (Fig. 23).
In both cases, this white and black sand unit is topped with a pale
gray clay unit that is slightly thicker in core BB08. Radiocarbon
dating of planktonic foraminifers in core BB08 indicates that the
white and black sand unit was deposited between 373 and 242 yr
ago. The pale gray clay unit above it then took more than 200 yr
to emplace, at a rate of roughly 0.5 mm yr–1.
Calibrated radiocarbon dates of planktonic foraminifers from
both cores are shown plotted in Figure 24. The depths of the dates
from each core have been normalized to the top of the well-sorted
white and black sand layer we assume is from a single, discrete
event affecting both sites. Ages in both Figure 23 and Figure 24
are given in years before 2004, when the cores were collected.
Ages are taken from median probability dates, with upper and
lower ranges shown.
Sedimentation patterns are broadly similar at the sites of
cores BB08 and BB09, but in detail there are important differ-
ences, in particular the thickness of the pale clay layer above the
white and black sand layer, the relative thicknesses between the
 151˚ 18' 151˚ 20' 151˚ 22' 151˚ 24' 151˚ 26'

-4˚ 38' -4˚ 38'

Fig 22
-4˚ 40' -4˚ 40'

high backscatter
spots
-4˚ 42' -4˚ 42'

Group 5
Group 6

-4˚ 44' -4˚ 44'

r Group 4
sca
de
Sli 21
Fig
1 km
Seamount
Gr
ou
p3

-4˚ 46' -4˚ 46'

Group 2
Fig
Fig

-4˚ 48' -4˚ 48'

20
19

scour
Group 1

Torkoro Reef

151˚ 18' 151˚ 20' 151˚ 22' 151˚ 24' 151˚ 26'
Figure 18. Side-scan mosaic of the central Hixon Bay region sediment-wave field. Dark tones are high backscatter. Region and scale are the same
as for Figure 16. Locations of the CHIRP lines depicted in Figures 19 through 22 are shown. The different groups of waves in the sediment-wave
field display different acoustic characteristics, with some groups displaying strong asymmetry between upslope and downslope flanks (groups 3
and 6), and others displaying weak or no asymmetry. High backscatter spots mark the slopes of a few group 4 waves nearest the slide scar.
114 Hoffmann et al.

TABLE 1. BACKSCATTER AND MORPHOLOGICAL CHARACTERISTICS OF HIXON BAY SEDIMENT WAVES

Group number Wave length Wave height Backscatter contrast Wave crests Channel
(m) (m)
1 600–800 20–30 None Slightly sinuous Yes
2 ~1000 30–80 Faint Somewhat sinuous No
3 500, D.D. 20, D.D. High Linear or somewhat sinuous Yes
4 1000–2000 30–40 High contrast spots near slide scar Slightly sinuous No
5 ~1000 15–20 Faint Slightly sinuous No
6 100–500 10–20 Faint Linear or slightly sinuous No
Note: Backscatter contrast refers to the flanks of individual waves. Channel denotes whether the group is associated with a
channel directly upslope. D.D.—decreasing downslope.

TABLE 2. SUB-BOTTOM PROFILE CHARACTERISTICS OF HIXON BAY SEDIMENT WAVES

Group number Penetration depth Reflector character Upslope Continuous reflectors Figure reference
(m) migration
1 <5 N.D. N.D. N.D. 19
2 Up to 20 Well stratified, D.T. Yes Between some waves 19, 20
3 <5 Primarily N.D. N.D. 20
diffractions
4 Up to 20, Well stratified, D.T. Yes Between some waves, 19, 21, 22
less at distal end at distal end
5 ~10 Well stratified, D.T. Yes Between some waves 21, 22
6 <5 Primarily N.D. N.D. 21
diffractions
Note: D.T.—downslope thinning, meaning the sequence is observed to thin downslope; N.D.—no data, meaning the group was not
imaged clearly in sub-bottom profiles.

well-sorted fine-black-tephra layer and the white and black sand
layer, and the number of upward-fining units above the white
and black sand layer (Fig. 23). The heterogeneity between the
cores is not surprising because the wave field as a whole is het-
erogeneous, and sedimentation patterns vary across the field as
a whole.
Hixon Bay Region Slide
The upper part of the Hixon Bay slide contains undulating
features, and sediment dispersed from slope failures within the
slide scar may deposit on the distal end of the sediment-wave
field (Fig. 17). The slide scar is apparent but diverges from a
half-circle shape in its southern end. A clearly defined chute runs
downslope toward Torkoro Trough, although part of the northern
boundary of this chute is obscured by undulations that compose
most of the upper part of the slide scar. These undulations are
reminiscent of the sediment waves discussed above, with a wave-
length of roughly 1 km and wave heights of up to 40 m. Either
side of the slide chute appears as a channel ~5 m deeper than the
center of the chute (Fig. 17).
In side-scan imagery these channels are associated with high
backscatter streaks (Fig. 15). The waveforms appear as alternat-
ing high and low backscatter regions, with the steeper, downslope
flanks associated with high backscatter. Where the overall slope
is flattest the contrast in backscatter is much less.
In profile, one wave shows clear upslope migration (Fig. 25).
The complex morphology and slopes of up to 4° within the slide
scar prevent clear imaging of the field with a shallow-towed
CHIRP system. The downslope end of the slide scar also appears
to have been buried (Fig. 16).
DISCUSSION
Based solely on the criteria given in O’Leary and Laine
(1996), H.J. Lee et al. (2002), and Wynn and Stow (2002), all
of the imaged wave fields fail simple classification. Nonetheless,
they provide important clues to the origins of sediment-wave
fields. Sediment waves, although well-recognized features, are
not ubiquitous but form only under very specific conditions.
Once the origin of a given wave field has been established, it
yields important information on broader processes.
Tolokiwa Sediment-Wave Fields
The upslope migration and acoustic asymmetry observed in
the large sediment waves northeast of Tolokiwa suggest a current
origin. The dimensions of the waves and the average slope of
the field are consistent with either a turbidity-current origin or
a bottom-current origin. The region is seismically active, and so
the low slopes (<0.5°) are also consistent with a deformational
origin, but the fields are not immediately associated with a fault.
 sequence thins
downslope

SE NW

1.75
VE = 8.2
50 m
Group 1 1 km
1.85
Levee
TWT (S)

1.95

2.05
Group 2
2.15
Group 4

2.25

upslope migration

10 m
200 m
continuous reflections
between features

Figure 19. CHIRP line across Hixon Bay sediment waves. Location shown in Figure 18. Detailed enlargements are at the same vertical exaggera-
tion and at 5.1 times the scale of the central base figure. The black tick marks indicate a transparent layer between a series of reflections that is
correlated throughout this part of the wave field. Angled lines highlight upslope migration in the wave field. This CHIRP line transects a small
part of group 1, a small levee, group 2, and a small part of group 4. VE—vertical exaggeration; TWT—two-way traveltime. The thickness of
deposits above the transparent layer increases upslope from ~5 m in group 4 to 6 m at the bottom of group 2 to 9 m at the top of group 2. This
measurement is for the thickness at the wave crest and assumes a sound velocity of 1500 m s–1.
 erosional surface
10 m 200 m

SE NW

1.60

1.70 VE = 6.6
channel 50 m 1 km
1.80
levee
TWT (s)

1.90

2.00
Group 2
2.10 Group 3
2.20

Figure 20. CHIRP line across the Hixon Bay sediment-wave field. Location shown in Figure 18. The line begins within the channel leading to
group 3. It then crosses the levee of the channel at a highly oblique angle. The line then crosses the largest waves of group 3. Because the group
3 waves are small, diffractions dominate the CHIRP image. Penetration in the group 3 waves is also <5 m. The group 2 waves display upslope
migration, as highlighted by the angled lines. VE—vertical exaggeration; TWT—two-way traveltime.
 10 m
250 m

2.10
W E

2.20
TWT (s)

outside of
wave field Group 6
2.30

50 m VE = 13.9
Group 4 Group 5
1 km
2.40

10 m
250 m

Figure 21. CHIRP line across the Hixon Bay sediment waves. Location shown in Figure 18. This line cuts obliquely across the southern end
of group 4, group 5, and the central part of group 6, and slightly beyond. The tick marks indicate a transparent layer that we interpret to be the
same transparent layer seen in Figure 19. VE—vertical exaggeration; TWT—two-way traveltime. The group 4 waves appear to display upslope
migration, as highlighted by the angled lines. Note the decreasing penetration from group 4 to group 5 to group 6.
 B Upslope migration
C BB09

S N

2.30
50 m VE = 16
TWT (s)

2 km
2.40
Group 4
BB09 Torkoro
Group 5 BB08 Trough
2.50

continuous reflections

A D BB08
10 m 500 m

Upslope migration

Figure 22. CHIRP profile of Hixon Bay sediment wave field. Location shown in Figure 18. VE—vertical exaggeration; TWT—two-way travel-
time. Core holes BB08 and BB09 are shown. This line crosses the group 5 waves perpendicular to their crests. Each of these waves displays clear
upslope migration, as shown in the two enlargements (A and B). In the group 4 waves shown here, upslope migration can be seen (angled line
in panels C and D). Also apparent is a downslope thinning of the sequence. Reflections in panel D are continuous and grade into parallel reflec-
tions in Torkoro Trough, and thin by ~20% from the upslope-most wave in panel D to the downslope-most wave. This thinning is also seen when
comparing cores BB08 and BB09 (see Fig. 23).
Depth (cm)
age before BB09 description description
BB08 age before
2004
2004
0 faintly bedded brownish-gray
0
fine silt fining up to clay upward-fining silty-clay
to clay

faintly bedded brownish-gray thin dark tephra

fine silt fining up to clay bearing silts and clay 97
upward-fining silt to clay
faintly bedded brownish-gray
fine silt fining up to clay
25 upward-fining silt to 25
basal mixed fine sand poorly sorted silty clay
faintly bedded brownish-gray
90 fine silt fining up to clay pale gray clay 187
faintly bedded brownish-gray sorted fine white and
fine silt fining up to clay 242
black sand
silty clay 373
pale clay 437
upward-fining fine-silt to
50 homogeneous brownish-gray silty clay
50
silty clay well sorted fine black sand tephra
119 sand injection upward-fining fine-silt to 437
silty clay
pale gray clay silty clay 573
sorted fine white and black sand fine silt
tephra silt/fine sand
poorly sorted upward-fining
clay and silt brown-gray silty clay
75 75
gray clay with pumice
brown mud dropstones

laminated silt and silty-clay

black sandy tephra

gray clay
100 tephra sand injection 100
black well sorted fine
sand tephra

interbedded clays and tephras

125 125
gray clay with pumice dropstones

pale gray well sorted silt

tephra

150 pale gray homogenous clay 150

silty tephra

Figure 23. Cores BB09 and BB08 from the Hixon Bay region sediment-wave field, with calibrated radiocarbon ages of planktonic foraminifers.
Both cores penetrated a gray clay unit containing pumice dropstones. Above this unit is a series of tephra deposits and silt, silty clay, and clay
deposits. Both units contain a thin layer of well-sorted white sand with black grains (65 cm depth in core BB09, and 42 cm depth in core BB08),
topped by a unit of pale gray clay. Ages (in years before 2004) from this pale gray clay unit indicate that it was deposited slowly in comparison with
the upward-fining silt and clay units found above it in both cores. Depths are uncorrected for compaction of sediment during core collection.
 0

-10
4 mm
6 mm y

yr (
-1 left

-20
r-1 (righ

axis)
t axis)

Depth in BB09 (cm)

Depth in BB08 (cm)

-30
Top of pale clay layer (BB08)

-40 Top of well sorted white and black sand layer

-50

-60

-70

Data from BB08 Data from BB09

-80
0 100 200 300 400 500 600 700
Age (years before 2004)

Figure 24. Depth-age plot of radiocarbon dates from planktonic foraminifers from cores BB08 and BB09 in the Hixon Bay sediment-wave field.
Age is in years before 2004, when the cores were collected. Depths have been normalized to the top of the well-sorted white and black sand
layer we observe in both cores, and which we assume is from a single, discrete event affecting both sites. Depth within core BB08 is shown on
left axis. Depth within core BB09 is shown on right axis. Median ages are shown, with upper and lower ranges shown. For comparison, lines
corresponding to example sedimentation rates for each axis are drawn. The sedimentation rate for core BB09 since the late nineteenth century is
~6 mm yr–1. The average sedimentation rate for core BB08 since the event that produced the white and black sand layer is about two-thirds the
average sedimentation rate in core BB09 since the same event. Thus, 4 mm yr–1, using the scale of the left axis, is shown.
 One wave in the field (right)
shows apparent upslope-
migration

1.50
Scarp

1.60
VE = 11.9
30 m
TWT (s)

1 km
1.70

1.80

1.90

5m
200 m

Figure 25. CHIRP line through Hixon Bay slide. Detailed enlargements are at the same vertical exaggeration and at 5 times the scale of
the base figure. VE—vertical exaggeration; TWT—two-way traveltime. Upslope migration is clear only in one wave in this field. Location
of figure shown in Figure 17.
122 Hoffmann et al.

1.9

1.8

1.7
relative deposition thickness

1.6

1.5

1.4

1.3

1.2

1.1

1
0 1 2 3 4 5 6 7
distance (km)
distance (km)

Figure 26. Plot of relative deposition thickness on wave crests in the Hixon Bay sediment wave field as a function of
relative distance. Measurements are taken from seven wave crests shown in Figure 19, normalized to the thickness of the
most distal wave crest measured. These are plotted as diamonds. Three measurements are taken from wave crests in the
distal end of the waves shown in Figure 22, plotted as crosses. These measurements are also normalized to the most distal
wave crest in the set. Distance is kilometers upslope from the most distal measurement in the set. The proximal measure-
ment from Figure 22 (the cross at 5.5 km) is extrapolated to the wave crest from the site of core hole BB09 and assumes
a relative sedimentation rate at the site of this core hole of 1.55 times the sedimentation rate at the site of core hole BB08.
The plot shows that these two sets of measurements, which are separated by 10 km, are consistent with a downslope ex-
ponential decay of sedimentation rate from wave crest to wave crest. Short-dashed line shows an exponential curve with
a characteristic decay length of 7.6 km.

We interpret that some of the small waves are younger features
than the large waves, owing to the onlapping contact that the
small waves have with the large waves. Nevertheless, other small
waves interspersed with the larger waves could have formed con-
comitantly. Because of their acoustic asymmetry, and the oblique
orientation of their wave crests with respect to local gradients,
we interpret the small waves to be of bottom-current origin. The
lack of evidence for turbidity-current activity in the vicinity of
the wave fields is consistent with this interpretation.
The origin of the large waves remains inconclusive. We sug-
gest, however, that the consistent upslope migration more likely
indicates a current origin than a deformational origin. The near-
est possible source of turbidity currents to the field is debris flows
from Tolokiwa, but the orientation of the waves is inconsistent
with currents flowing from south to north. One alternative is that
the large waves were also generated by bottom currents. In this
case, the differences in morphology between the large waves
and small waves would imply a difference in bottom-current
velocity and direction, and possibly a change in density stratifi-
cation within the ambient seawater (Flood, 1988). However, the
crests of the large waves are oriented perpendicular to the local
slope, suggesting that they were formed by turbidity currents
derived from shallow regions to the west. The small waves thus
may have formed subsequent to or concomitantly with the large
waves as they were modified by bottom currents.
Dakataua Sediment-Wave Field
Features in the Dakataua wave field suggest both a deforma-
tional origin and a turbidity-current origin. The decreasing wave
height with distance from the caldera (Fig. 5), combined with the
decreasing high backscatter scour features in side-scan imagery
(Fig. 6), suggest that turbidity currents may play a role. The field
is isolated from hyperpycnal currents, however. The wave field
begins ~5 km from shore, south of a steep slope that drops more
than 1200 m in this space. We suggest that coastal erosion by
 Sediment waves in the Bismarck Volcanic Arc
energetic waves during large storms could result directly in tur-
bidity currents as sediment is concentrated in the surf zone during
peaks of storm activity and then released during lulls in wave
action (Scully et al., 2002). Energetic waves could also result in
turbidity currents by destabilizing the slope of the caldera, result-
ing in numerous small slope failures of weakly consolidated sedi-
ment that rapidly entrains seawater to become a turbidity flow
(Marr et al., 2001). Additionally, ashfall from eruptions can result
in a hyperconcentrated sediment layer on the sea surface, which
can result in turbidity currents (Fiske et al., 1998; Parsons et al.,
2001). Lastly, turbidity flows could be triggered by earthquakes.
Turbidity-current-generated sediment waves proximal to volca-
nic sources but lacking fluvial input are recognized elsewhere
(Wynn et al., 2000).
The arcuate and irregular morphology of the waves, as well
as the large enclosed depressions in the field, suggest that defor-
mation of some kind may play a role in the formation of these
sediment waves. The Dakataua lineation (Fig. 7) may be a fault
with some degree of downthrow to the east. We suggest that the
Dakataua wave field is formed by some combination of turbidity
currents and deformation. Lacking sub-bottom data, however, the
origin of the features remains speculative.
Kimbe Bay Sediment-Wave Field
The Kimbe Bay wave field is far from a potential source of
hyperpycnal currents. The highly irregular and relatively arcu-
ate plan-view morphology and the many enclosed depressions in
the field suggest a deformational origin. The waves display some
subsurface characteristics of current activity, however (Fig. 13),
in upslope migration of the wave crests. The presence of nearby
channels and plunge pools indicates that turbidity currents are
active in the vicinity of this field. The crevice to the east of the
field may be related to motion along the Kimbe Bay Escarpment
if this is an active fault, or the crevice may be the site of focused
channel flow. The depressions in the crevice may be tectonically
controlled, although without subsurface imagery we cannot con-
firm this interpretation. The lineation in the field (Fig. 11) may be
the site of a splay fault related to the Kimbe Bay Escarpment.
Like the Dakataua field, the Kimbe Bay field forms a topo-
graphic high relative to the surrounding seafloor, suggesting
higher sedimentation rates within the field. The two fields are also
similar in terms of wave height, wavelength, the presence of deep
enclosed depressions within the fields, and irregular morphology.
Both fields are closely related to signs of turbidity-current activ-
ity and possible faults. We suggest that the Kimbe Bay field is
also formed by a combination of turbidity currents and deforma-
tion, although we cannot determine which process dominates.
Hixon Bay Sediment-Wave Field
The morphological heterogeneity of this sediment-wave
field suggests that different parts of the field formed under dif-
ferent conditions. However, the groups we have defined are not
123
entirely distinct from one another, with some wave crests continu-
ous through the boundaries of two or more groups (Fig. 18). This
overlap suggests that either the morphology predates deposition
by currents or that currents interact with topography, depend-
ing on their magnitude, and that a continuum of magnitudes and
slightly different flow paths will result in indistinct boundaries
between flow conditions, giving rise to the variability in the mor-
phology of the field. Hyperpycnal currents that originate from the
Pandi River will tend to follow the channels leading toward the
wave field, as seen by the high backscatter scour in these chan-
nels (Fig. 18).
A turbidity-current origin for the Hixon Bay sediment
waves is strongly supported by sub-bottom data. The similarity
of reflections from wave to wave (Figs. 19–22), the progressive
downslope thinning of reflection sequences (Figs. 19, 22), the
continuity of reflections between waves (Figs. 19, 22), and the
ubiquitous upslope migration of waves (Figs. 19–22) all argue
against a deformational origin and in favor of a current origin.
Torkoro Trough is enclosed on three sides (Fig. 14). Cores taken
from the sediment-wave field show a series of upward-fining,
graded units consistent with turbidites (Fig. 23). These obser-
vations argue against a bottom-current origin and in favor of a
turbidity-current origin of the wave field. The heterogeneous
flow conditions across the wave field may be largely the result
of focused flow down the channels and different magnitudes of
flow events.
The high-backscatter regions within the slide scar (Fig. 25)
and the high-backscatter spots on the upslope flanks of three
sediment waves near the base of the slide scar (Fig. 18) suggest
active sediment transport through the slide scar. Some of this
sediment may reach the distal portion of the sediment-wave field.
However, the undulating topography in the upslope part of the
slide scar (Fig. 17) implies a site of active deposition. This depo-
sition, together with the large, mountainous drainage area of the
Pandi River in comparison with the catchments of rivers drain-
ing the east flank of Lolobau or the west flank of Ulawun, and
also the absence of channels or turbidity scour leading toward
the slide scar, all suggest that hyperpycnal discharge from the
Pandi River is the dominant source for the upward-fining units in
the cores. Slope failure and turbidity flows caused by rapid sedi-
mentation and triggered by earthquakes also may explain some
of the heterogeneity of the field. We conclude that the Hixon
Bay sediment-wave field is of hyperpycnal-current origin, at
least insofar as active growth is concerned. We cannot determine
whether or not the field grew from initial topography created by
some other process.
Sediment Supply to the Hixon Bay Sediment-Wave Field
The Hixon Bay sediment-wave field demonstrates that sig-
nificant quantities of sediment are capable of reaching the Kimbe
Bay wave field, >40 km from shore. The Hixon Bay sediment-
wave field covers ~70 km2 (Fig. 16). The waves display a pattern
of decreasing deposition downslope. Using a transparent layer
124 Hoffmann et al.
that can be correlated through much of the field, we measured the
relative thicknesses of deposits on seven wave crests (Fig. 19).
Farther downslope in the distal end of the field, we measured
three more by tracing continuous reflections (Fig. 22D) and by
extrapolating the difference in sedimentation rates observed in
cores BB08 and BB09 to the wave crest nearest core hole BB09
(Fig. 22C), taking advantage of continuous reflections between
this core and the wave crest.
Although flow conditions vary throughout the field, we
assume that sediment is primarily, although not exclusively,
originating as discharge from the Pandi River. The similarity of
cores BB08 and BB09 (Fig. 23) and the similarity of reflection
sequences from wave to wave in groups 2, 4, and 5 (Figs. 19–22)
are consistent with this assumption. We assume no discontinuities
in sedimentation rate between groups. That reflector thickness
decreases steadily downslope from wave to wave, even between
groups (Figs. 19, 22) is consistent with this assumption. The
waves display upslope migration, with thicker upslope sediment
deposits and thinner downslope sediment deposits. We assume
that the thickness of sediment at a wave crest is approximately
the average sediment thickness for a given wave. We normalize
the seven measurements of thickness from Figure 19 to the thick-
ness measurement of the wave crest farthest downslope in this
set. We similarly normalize the three measurements of thickness
from the distal end of the wave field (Fig. 22). Lastly, we assume
a power-law distribution of hyperpycnal-current events from the
Pandi River (Warrick and Milliman, 2003; Dadson, et al., 2005).
Thus, deposition rates are greater in the upslope part of the wave
field because this region receives sediment from many small, fre-
quent events as well as from large, infrequent events, whereas the
downslope part of the wave field receives sediment only from the
large, infrequent events (H.J. Lee et al., 2002).
A power-law distribution of events is expected to produce an
exponential decay of sedimentation rates away from the source.
We find that the two sets of normalized relative-thickness mea-
surements are consistent with a single exponential decay rate
(Fig. 26). Linear functions will also fit both sets of measure-
ments. However, linear functions are not invariant to scalar multi-
plication. Because we are normalizing each set of measurements,
a single linear function cannot fit both sets. The characteristic
decay rate of an exponential function, however, is invariant to
scalar multiplication. The characteristic decay length of this curve
is ~7.6 km. That is, from wave crest to wave crest, sedimenta-
tion rates are inferred to decrease by half 7.6 km downslope from
any given point in the field. Based on the above assumptions, we
extrapolate this decay rate to the remainder of the field and cali-
brate it with sedimentation rates inferred from radiocarbon dating
of cores BB08 and BB09 (Fig. 24).
Because we are interested in modern discharge of the Pandi
River, we use only the upper parts of the cores, estimating sedi-
ment accumulation in the distal end of the field at 4 mm yr–1
(Fig. 24). Note that this estimate does not account for compression
of the cores during collection and so is a lower bound estimate.
However, some of this is likely to be pelagic sedimentation, and
some of this may be related to currents flowing down the Hixon
slide scar (Fig. 25). Thus, we conservatively halve this estimate of
sediment flux arriving from the Pandi River. Integrating, the aver-
age deposition rate for the whole field is ~5 mm yr–1. Multiplying
by the area of the field, this yields a volumetric deposition of 3.5
× 105 m3 yr–1. Assuming a bulk density of the sediment of 2200 kg
m–3 results in a mass deposition rate of 7.7 × 105 t yr–1. If this
estimate represents a significant portion of Pandi River discharge,
then it places a testable lower bound on the sediment supply of
this system.
We cannot characterize the uncertainty in this estimation, but
it suggests that a large part of the Pandi River sediment bypasses
the Papua New Guinea shelf and is deposited in deep water.
Because we observe sedimentation rates of >4 mm yr–1 nearly
40 km from shore, we conclude that fluvial sediment entering
Kimbe Bay can easily account for the deposition we observe in
the Kimbe Bay wave field, although probably not the morphol-
ogy itself. The rivers entering Kimbe Bay are smaller than the
Pandi River, on the whole, and 40 km from shore the Hixon Bay
sediment-wave field is dying out, not just beginning, as is the
case in Kimbe Bay. Thus, although the Kimbe Bay field may
be dominantly of turbidity-current origin, it is probably not of
hyperpycnal-current origin.
We note that previous work has suggested that as much as
90% of sediment from the Sepik River may bypass the shelf as
sediment gravity flows (Walsh and Nittrouer, 2003). This infer-
ence was based on a comparison of sediment discharge from the
Sepik and measurements of depositional thickness on the shelf.
The measured thickness of deposits only accounted for 10% of
expected discharge. A similar argument was made by Dadson
et al. (2005) regarding sediment discharged at hyperpycnal con-
centrations from the Choshui River in Taiwan. The Hixon Bay
sediment waves present the opposite side of the story for the
Pandi River. Results from this field suggest that on active tec-
tonic margins with narrow shelves, hyperpycnal flows play an
important role in delivering sediment to the ocean. This may be
capable of explaining missing shelf deposits elsewhere, such as
off the Sepik River, the Choshui River, and the Eel River (Walsh
and Nittrouer, 2003; Dadson, et al., 2005; Sommerfield and Nit-
trouer, 1999).
CONCLUSIONS
We image six sediment-wave fields in the Bismarck Volca-
nic Arc. The two Tolokiwa fields appear to have formed by tur-
bidity currents and were subsequently or concomitantly modified
by bottom currents. The Kimbe Bay and Dakataua wave fields
appear to be of deformational and turbidity-current origin. In
Hixon Bay, two hyperpycnal-current-generated sediment-wave
fields are observed. One is a small, apparently young field in its
initial stages of growth. The other field is an older, more stable
feature related to discharge from the Pandi River. Measurements
from this wave field indicate that a large part of sediment dis-
charge from the Pandi River is deposited in deep water.
 Sediment waves in the Bismarck Volcanic Arc
The evidence for turbidity currents in the vicinity of the
Dakataua and Kimbe Bay wave fields, and that both fields are
topographic highs relative to the surrounding seafloor, suggest
focused deposition within the area of these wave fields. Between
these fields and the Hixon Bay wave fields, sediment waves with
turbidity-current components of formation appear to be a com-
mon feature in the Bismarck Volcanic Arc, particularly where
sedimentation rates are relatively high. In these locations, depo-
sition rates can vary greatly in the space of a few kilometers, both
along the strike of the arc because of focused sedimentation, and
across the strike because of the variation of sedimentation rates
between the upslope and downslope flanks of a sediment wave.
Thus, care must be taken when applying interpretations of a sedi-
mentary section to the remainder of the arc environment.
Between these fields and the Tolokiwa fields, we do not
observe large sediment waves. Topographic relief in this region
is lower than in eastern New Britain, and so we expect that sedi-
mentation rates are significantly lower. Thus, sediment waves
with turbidity-current contributions to formation may be an indi-
cator of relatively high sedimentation rates, although the inverse
may not be true. We are not aware of large turbidity-current sedi-
ment waves that are recognized in the sedimentary record. This is
probably due largely to an inherent difficulty in recognizing such
large, subtle features from outcrops, particularly if the sequence
is at all deformed. Sections in which high sedimentation rates
have been recognized thus may be good places for reexamining
the sedimentary record for sediment-wave fields.
ACKNOWLEDGMENTS
Funding for this research was provided by the Marine Geology
and Geophysics section of the U.S. National Science Founda-
tion, grant OCE-0327004 to E. Silver and S. Ward, and grant
OCE-0328278 to N. Driscoll. We are grateful for the advice and
support of many people in carrying out this research, including
Hugh Davies, James Robins, Bruce Appelgate and the Hawaii
Mapping Research Group, Russell Perembo and Susan John of
the University of Papua New Guinea for picking foraminifers,
Michaele Kashgarian of the radiocarbon facility at Lawrence
Livermore National Laboratory for assistance with dating, the
Oregon State University coring team, and the captain and crew
of the research vessel Kilo Moana. We thank A. Draut (Associ-
ate Editor), H. Lee, J. Sample, and K. Straub for their thorough
and insightful reviews of this manuscript. Most of the figures for
this paper were generated in part using Generic Mapping Tools.
REFERENCES CITED
Abbott, L.D., Silver, E.A., Thompson, P.R., Filewicz, M.V., Schneider, C., and
Abdoerrias, 1994, Stratigraphic constraints on the development and tim-
ing of arc-continent collision in northern Papua New Guinea: Journal of
Sedimentary Research, v. 64, p. 169–183.
Bowen, A.J., Normark, W.R., and Piper, D.J.W., 1984, Modeling of turbidity
currents on Navy submarine fan, California Continental Borderland: Sedi-
mentology, v. 31, p. 169–185, doi: 10.1111/j.1365-3091.1984.tb01957.x.
125
Coleman, S.E., Zhang, M.H., and Clunie, T.M., 2005, Sediment-wave develop-
ment in subcritical water flow: Journal of Hydraulic Engineering, v. 131,
p. 106–111, doi: 10.1061/(ASCE)0733-9429(2005)131:2(106).
Dadson, S., Hovius, N., Pegg, S., Dade, W.B., Horng, M.J., and Chen, H., 2005,
Hyperpycnal river flows from an active mountain belt: Journal of Geo-
physical Research, v. 110, p. F04016, doi: 10.1029/2004JF000244.
Ercilla, G., Belén, A., Wynn, R.B., and Baraza, J., 2002, Turbidity current sedi-
ment waves on irregular slopes: Observations from the Orinoco sediment-
wave field: Marine Geology, v. 192, p. 171–187, doi: 10.1016/S0025-
3227(02)00554-6.
Faugères, J.-C., Gonthier, E., Mulder, T., Kenyon, N., Cirac, P., Griboulard,
R., Berné, S., and Lesuavé, R., 2002, Multi-process generated sediment
waves on the Landes Plateau (Bay of Biscay, North Atlantic): Marine
Geology, v. 182, p. 279–302, doi: 10.1016/S0025-3227(01)00242-0.
Fiske, R.S., Cashman, K.V., Shibata, A., and Watanabe, K., 1998, Tephra dis-
persal from Myojinsho, Japan, during its shallow submarine eruption of
1952–1953: Bulletin of Volcanology, v. 59, p. 262–275, doi: 10.1007/
s004450050190.
Flood, R.D., 1988, A lee wave model for deep-sea mudwave activity: Deep-Sea
Research, v. 35, p. 973–983, doi: 10.1016/0198-0149(88)90071-4.
Gardner, J.V., Prior, D.B., and Field, M.E., 1999, Humboldt slide—A large
shear-dominated retrogressive slope failure: Marine Geology, v. 154,
p. 323–338, doi: 10.1016/S0025-3227(98)00121-2.
Hand, B.M., 1974, Supercritical flow in density currents: Journal of Sedimen-
tary Petrology, v. 44, p. 637–648.
Hayes, K., Montgomery, D.R., and Newhall, C.G., 2002, Fluvial sediment trans-
port and deposition following the 1991 eruption of Mount Pinatubo: Geo-
morphology, v. 45, p. 211–224, doi: 10.1016/S0169-555X(01)00155-6.
Hill, J.C., Driscoll, N., Weissel, J.K., and Goff, J.A., 2004, Large-scale elon-
gated gas blowouts along the U.S. continental margin: Journal of Geo-
physical Research, v. 109, p. B09101, doi: 10.1029/2004JB002969.
Howe, J.A., 1996, Turbidite and contourite sediment waves in the northern Rock-
all Trough: North Atlantic Ocean: Sedimentology, v. 43, p. 219–234.
Jerolmack, D., and Mohrig, D., 2005a, Interactions between bed forms: Topog-
raphy, turbulence, and transport: Journal of Geophysical Research, v. 110,
p. F02014, doi: 10.1029/2004JF000126.
Jerolmack, D.J., and Mohrig, D., 2005b, A unified model for subaqueous
bed form dynamics: Water Resources Research, v. 41, p. W12421, doi:
10.1029/2005WR004329.
Johnson, R.W., 1979, Geotectonics and volcanism in Papua New Guinea: A
review of the late Cainozoic: Bureau of Mineral Resources, Journal of
Australian Geology and Geophysics, v. 4, p. 181–207.
Johnson, R.W., 1987, Large-scale volcanic cone collapse: The 1888 slope fail-
ure of Ritter volcano, and other examples from Papua New Guinea: Bul-
letin of Volcanology, v. 49, p. 669–679, doi: 10.1007/BF01080358.
Kidd, R.B., Lucchi, R.G., Gee, M., and Woodside, J.M., 1998, Sedimentary
processes in the Stromboli Canyon and the Marsili Basin, SE Tyrrhenian
Sea: Results from sidescan sonar surveys: Geo-Marine Letters, v. 18,
p. 146–154, doi: 10.1007/s003670050062.
Kneller, B., and Buckee, C., 2000, The structure and fluid mechanics of tur-
bidity currents: A review of some recent studies and their geological
implications: Sedimentology, v. 47, p. 62–94, doi: 10.1046/j.1365-
3091.2000.047s1062.x.
Kubo, Y., and Nakajima, T., 2002, Laboratory experiments and numerical simu-
lations of sediment-wave formation by turbidity currents: Marine Geol-
ogy, v. 192, p. 105–121, doi: 10.1016/S0025-3227(02)00551-0.
Lee, H.J., Syvitski, J.P.M., Parker, G., Orange, D., Locat, J., Hutton, E.W.H.,
and Imran, J., 2002, Distinguishing sediment waves from slope failure
deposits: Field examples, including the ‘Humboldt slide,’ and model-
ing results: Marine Geology, v. 192, p. 79–104, doi: 10.1016/S0025-
3227(02)00550-9.
Lee, S.E., Talling, P.J., Ernst, G.G.J., and Hogg, A.J., 2002, Occurrence and origin
of submarine plunge pools at the base of the US continental slope: Marine
Geology, v. 185, p. 363–377, doi: 10.1016/S0025-3227(01)00298-5.
Lee, S.H., and Chough, S.K., 2001, High-resolution (2–7 kHz) acoustic and
geometric characters of submarine creep deposits in the South Korea Pla-
teau: East Sea: Sedimentology, v. 48, p. 629–644, doi: 10.1046/j.1365-
3091.2001.00383.x.
Lewis, K.B., and Pantin, H.M., 2002, Channel-axis, overbank and drift sedi-
ment waves in the southern Hikurangi Trough, New Zealand: Marine
Geology, v. 192, p. 123–151, doi: 10.1016/S0025-3227(02)00552-2.
126 Hoffmann et al.
Lowe, D.R., 1982, Sediment gravity flows. II. Depositional models with spe-
cial reference to the deposits of high-density turbidity currents: Journal of
Sedimentary Petrology, v. 52, p. 279–297.
Madsen, J.A., and Lindley, I.D., 1994, Large-scale structures on Gazelle Penin-
sula, New Britain: Implications for the evolution of the New Britain Arc:
Australian Journal of Earth Sciences, v. 41, p. 561–569.
Marr, J.G., Harff, P.A., Shanmugam, G., and Parker, G., 2001, Experiments on
subaqueous sandy gravity flows: The role of clay and water content in
flow dynamics and depositional structures: Geological Society of America
Bulletin, v. 113, p. 1377–1386, doi: 10.1130/0016-7606(2001)113<1377:
EOSSGF>2.0.CO;2.
McAlpine, J.R., and Keig, G., 1983, Climate of Papua New Guinea: Canberra,
Australian National University Press, 200 p.
Middleton, G.V., 1993, Sediment deposition from turbidity currents: Annual
Review of Earth and Planetary Sciences, v. 21, p. 89–114, doi: 10.1146/
annurev.ea.21.050193.000513.
Middleton, G.V., and Hampton, M.A., 1976, Subaqueous sediment transport
and deposition by sediment gravity flows, in Stanley, D.J., and Swift,
D.J.P., eds., Marine Sediment Transport and Environmental Management:
New York, Wiley, p. 197–218.
Migeon, S., Savoye, B., Zanella, E., Mulder, T., Faugères, J.-C., and Weber, O.,
2001, Detailed seismic-reflection and sedimentary study of turbidite sedi-
ment waves on the Var Sedimentary Ridge (SE France): Significance for
sediment transport and deposition and for the mechanisms of sediment-
wave construction: Marine and Petroleum Geology, v. 18, p. 179–208,
doi: 10.1016/S0264-8172(00)00060-X.
Milliman, J.D., 1995, Sediment discharge to the ocean from small mountainous
rivers: The New Guinea example: Geo-Marine Letters, v. 15, p. 127–133,
doi: 10.1007/BF01204453.
Milliman, J.D., and Syvitski, J.P.M., 1992, Geomorphic/tectonic control of sed-
iment discharge to the ocean: The importance of small mountain rivers:
Journal of Geology, v. 100, p. 525–544.
Mohrig, D., and Smith, J.D., 1996, Predicting the migration rates of sub-
aqueous dunes: Water Resources Research, v. 32, p. 3207–3217, doi:
10.1029/96WR01129.
Mulder, T., and Alexander, J., 2001, The physical character of subaqueous sedi-
mentary density flows and their deposits: Sedimentology, v. 48, p. 269–
299, doi: 10.1046/j.1365-3091.2001.00360.x.
Mulder, T., and Syvitski, J.P.M., 1995, Turbidity currents generated at river
mouths during exceptional discharges to the world oceans: Journal of
Geology, v. 103, p. 285–299.
Mulder, T., Syvitski, J.P.M., and Skene, K.I., 1998, Modeling of erosion and
deposition by turbidity currents generated at river mouths: Journal of
Sedimentary Research, v. 68, p. 124–137.
Mulder, T., Migeon, S., Savoye, B., and Faugères, J.-C., 2001, Inversely graded
turbidite sequences in the deep Mediterranean: A record of deposits from
flood-generated turbidity currents?: Geo-Marine Letters, v. 21, p. 86–93,
doi: 10.1007/s003670100071.
Mulder, T., Migeon, S., Savoye, B., and Faugères, J.-C., 2002, Reply to dis-
cussion by Shanmugam on Mulder et al. (2001, Geo-Marine Letters,
v. 21, p. 86–93), Inversely graded turbidite sequences in the deep Medi-
terranean: A record of deposits from flood-generated turbidity currents?:
Geo-Marine Letters, v. 22, p. 112–120, doi: 10.1007/s00367-002-0096-8.
Mulder, T., Syvitski, J.P.M., Migeon, S., Faugères, J.-C., and Savoye, B., 2003,
Marine hyperpycnal flows: Initiation, behavior, and related deposits. A
review: Marine and Petroleum Geology, v. 20, p. 861–882, doi: 10.1016/
j.marpetgeo.2003.01.003.
Nakajima, T., and Satoh, M., 2001, The formation of large mudwaves by turbid-
ity currents on the levees of the Toyama deep-sea channel, Japan Sea: Sed-
imentology, v. 48, p. 435–463, doi: 10.1046/j.1365-3091.2001.00373.x.
Normark, W.R., Hess, G.R., Stow, D.A.V., and Bowen, A.J., 1980, Sediment
waves on the Monterey Fan levee: A preliminary physical interpretation:
Marine Geology, v. 37, p. 1–18, doi: 10.1016/0025-3227(80)90009-2.
O’Leary, D.W., and Laine, E., 1996, Proposed criteria for recognizing intrastratal
deformation features in marine high resolution seismic reflection profiles:
Geo-Marine Letters, v. 16, p. 305–312, doi: 10.1007/BF01245561.
Parsons, J.D., Bush, J.W.M., and Syvitski, J.P.M., 2001, Hyperpycnal plume for-
mation from riverine outflows with small sediment concentrations: Sedi-
mentology, v. 48, p. 465–478, doi: 10.1046/j.1365-3091.2001.00384.x.
Pickup, G., 1984, Geomorphology of tropical rivers: I. Landforms, hydrology,
and sedimentation in the Fly and lower Purari, Papua New Guinea, in
Schick, A.P., ed., Channel Processes—Water, Sediment, Catchment Con-
trols: Braunschweig, Catena Supplement 5, p. 1–17.
Pigram, C.J., and Davies, H.L., 1987, Terranes and the accretion history of the
New Guinea orogen: Bureau of Mineral Resources, Journal of Australian
Geology and Geophysics, v. 10, p. 193–211.
Schwehr, K., Driscoll, N., Tauxe, L., 2007, Origin of continental margin mor-
phology; submarine-slide or downslope current-controlled bedforms, a
rock magnetic approach: Marine Geology, v. 240, p. 19–41.
Scully, M.E., Friedrichs, C.T., and Wright, L.D., 2002, Application of an ana-
lytical model of critically stratified gravity-driven sediment transport and
deposition to observations from the Eel River continental shelf, North-
ern California: Continental Shelf Research, v. 22, p. 1951–1974, doi:
10.1016/S0278-4343(02)00047-X.
Shanmugam, G., 2000, 50 years of the turbidite paradigm (1950s–1990s):
Deep-water processes and facies models—A critical perspective:
Marine and Petroleum Geology, v. 17, p. 285–342, doi: 10.1016/S0264-
8172(99)00011-2.
Shanmugam, G., 2002, Discussion on Mulder et al. (2001, Geo-Marine Letters,
v. 21, p. 86–93), Inversely graded turbidite sequences in the deep Medi-
terranean: A record of deposits from flood-generated turbidity currents?:
Geo-Marine Letters, v. 22, p. 108–111, doi: 10.1007/s00367-002-0100-3.
Skene, K.I., Mulder, T., and Syvitski, J.P.M., 1997, INFLO1: A model predicting
the behaviour of turbidity currents generated at river mouths: Computers
& Geosciences, v. 23, p. 975–991, doi: 10.1016/S0098-3004(97)00064-2.
Sommerfield, C.K., and Nittrouer, C.A., 1999, Modern accumulation rates
and a sediment budget for the Eel shelf: A flood-dominated depositional
environment: Marine Geology, v. 154, p. 227–241, doi: 10.1016/S0025-
3227(98)00115-7.
Stow, D.A.V., Reading, H.G., and Collinson, J.D., 1996, Deep seas, in Reading,
H.G., ed., Sedimentary Environments; Processes, Facies, and Stratigra-
phy: Oxford, UK, Blackwell Science, p. 395–453.
Taylor, B., 1979, Bismarck Sea: Evolution of a backarc basin: Geology, v. 7,
p. 171–174, doi: 10.1130/0091-7613(1979)7<171:BSEOAB>2.0.CO;2.
Walsh, J.P., and Nittrouer, C.A., 2003, Contrasting styles of off-shelf sediment
accumulation in New Guinea: Marine Geology, v. 196, p. 105–125, doi:
10.1016/S0025-3227(03)00069-0.
Ward, S.N., 2001, Landslide tsunami: Journal of Geophysical Research, v. 106,
p. 11,201–11,215, doi: 10.1029/2000JB900450.
Warrick, J.A., and Milliman, J.D., 2003, Hyperpycnal sediment discharge from
semiarid southern California rivers: Implications for coastal sediment
budgets: Geology, v. 31, p. 781–784, doi: 10.1130/G19671.1.
Wynn, R.B., and Stow, D.A.V., 2002, Classification and characterization
of deep-water sediment waves: Marine Geology, v. 192, p. 7–22, doi:
10.1016/S0025-3227(02)00547-9.
Wynn, R.B., Masson, D.G., Stow, D.A.V., and Weaver, P.P.E., 2000, Turbidity
current sediment waves on the submarine slopes of the western Canary
Islands: Marine Geology, v. 163, p. 185–198, doi: 10.1016/S0025-
3227(99)00101-2.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

The Lichi Mélange: A collision mélange formation along early arcward backthrusts
during forearc basin closure, Taiwan arc-continent collision

Chi-Yue Huang
Department of Earth Sciences, National Cheng Kung University, Tainan, Taiwan, and Research Center of Ocean Environment
and Technology, National Cheng Kung University, Tainan, Taiwan

Chih-Wei Chien
Department of Earth Sciences, National Cheng Kung University, Tainan, Taiwan

Bochu Yao
Guangzhou Marine Geological Survey, Ministry of Land and Resource, P.R.China, Guangzhou 510075

Chung-Pai Chang
Center for Space and Remote Sensing Research, National Central University, Chungli 320, Taiwan

ABSTRACT

Marine surveys show that the submarine Huatung Ridge extends northward to
the Lichi Mélange in the southwestern Coastal Range, suggesting that formation of
the Lichi Mélange is related to arcward thrusting of the forearc strata in the western
part of the North Luzon Trough during active arc-continent collision off southern Tai-
wan. A new seismic survey along the 21° N transect across the North Luzon Trough in
the incipient arc-continent collision zone further reveals that deformation of the Hua-
tung Ridge occurred soon after sedimentation in the western forearc basin, whereas
sedimentation was continuous in the eastern part of the remnant North Luzon Trough
until the complete closure of the forearc basin approaching SE Taiwan. This suggests
that the sequence in the Huatung Ridge can be coeval with just the lower sequence of
the remnant-forearc-basin strata. Multiple lines of new evidence, including micropa-
leontology, clay mineralogy, and fission track analyses along the Mukeng River and
its tributary key sections, are used to test this thrusting-forearc-origin hypothesis of
the Lichi Mélange.
In the SW Coastal Range the Lichi Mélange lies between the collision suture of
Longitudinal Valley to the west and the Taiyuan remnant forearc basin to the east. A
field survey indicates that the Taiyuan forearc-basin sequence and its volcanic base-
ment were thrust westward over the Lichi Mélange along the east-dipping Tuluan-
shan Fault. The Lichi Mélange shows varying degrees of fragmentation of strata,
mixing, and shearing. An apparently wide range of facies is present, from the weakly
sheared broken formation facies, with discernible relict sedimentary structures, to
the intensely sheared block-in-matrix mélange facies, with pervasively scaly foliation

Huang, C.Y., Chien, C.W., Yao, B., and Chang, C.P., 2008, The Lichi Mélange: A collision mélange formation along early arcward backthrusts during forearc basin
closure, Taiwan arc-continent collision, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Colli-
sion Zones: Geological Society of America Special Paper 436, p. 127–154, doi: 10.1130/2008.2436(06). For permission to copy, contact editing@geosociety.org.
©2008 The Geological Society of America. All rights reserved.

127
128 Huang et al.

dipping to the SE. Sedimentological study reveals that the subangular to subrounded,
fractured, matrix-supported metasandstone conglomerates in the pebbly mudstone
layers are repeatedly found in the broken formation facies of the Lichi Mélange. Their
composition and occurrence are identical to the deep-sea-fan conglomerate beds in
the Taiyuan remnant-forearc-basin strata to the east. Benthic foraminiferal faunas
are similar in the Lichi Mélange, regardless of the varying intensity of shearing and
strata disruptions, and are compatible with the benthic foraminiferal fauna in the
Taiyuan remnant-forearc-basin turbidites, supporting the interpretation that the pro-
tolith of the Lichi Mélange was originally deposited in the North Luzon Trough. Age
determination of planktic microfossils further demonstrates that the Lichi Mélange is
early Pliocene (3.5–3.7 Ma), implying that this mélange was deposited in a short time
and that deformation occurred soon after its deposition. The early Pliocene age of the
Lichi Mélange is coeval with just the lower part of the Taiyuan remnant forearc stra-
ta, and is much younger than the upper forearc sequence (3–1 Ma). Thus the Taiyuan
coherent-forearc-basin strata (3.7–1 Ma) were deposited continuously in the remnant
North Luzon Trough regardless of the deformation in its western part (the proto-
mélange). This scenario is an analogue for the modern configuration of the Huatung
Ridge–remnant North Luzon Trough off the southern Coastal Range in the active
arc-continent collision zone north of lat 21° N.
In addition to its kaolinite content (11–15%), the clay mineral composition of
the Lichi Mélange is compatible with the Taiyuan remnant forearc turbidites. In the
Coastal Range, kaolinites are found only in the volcanic rocks of the Tuluanshan For-
mation. This additional kaolinite in the Lichi Mélange could not have been derived
from the exposed accretionary prism to the North Luzon Trough by sedimentary mass
slumping, because no such volcanic rocks are now exposed in the accretionary prism
west of the Coastal Range. Instead, they could have been derived from the Tuluan-
shan Formation when it was emplaced into the Lichi Mélange by thrusting during
the last 1 Ma when the Luzon arc-forearc was accreted to form the southern Coastal
Range. Thus the kaolinites of the volcanic arc rocks were redistributed into the Lichi
Mélange by fluid flows along the ubiquitous geological fractures in the mélange, con-
sistent with the field occurrences of the large, rootless, fault-bounded volcanic rocks
of andesitic breccia, tuff, and agglomerates that were floating in the intensely sheared
block-in-matrix mélange facies of the Lichi Mélange.
Mélange is commonly considered to develop in the accretionary prism of a sub-
duction zone. However, the Lichi Mélange in the SW Coastal Range originated from
the thrust forearc strata, representing a unique forearc mélange for orogenic belts
worldwide. The young age and wide distribution—especially the continuous offshore-
onshore connection—of the Lichi Mélange provides a unique example for further
research into active modern mélange-forming processes by forearc thrusting during
progressive closure of the forearc basin in this active region of arc-continent collision.

Keywords: Lichi Mélange, coastal range in eastern Taiwan, forearc collision mélange,
initial arc-continent collision, forearc closure.

INTRODUCTION
The Coastal Range in eastern Taiwan has resulted from the
accretion of the Luzon Arc-forearc to the exhumed metamorphic
basement of eastern Taiwan (Fig. 1) during the arc-continent col-
lision in the last 2 m.y. (Chai, 1972; Biq, 1973; Huang et al., 2000,
2006). Now the location of initial arc-continent collision lies off
southern Taiwan (Fig. 1; Huang and Yin, 1990; Huang et al.,
1992; Reed et al., 1992; Lundberg et al., 1997; Malavieille et al.,
2002). Stratigraphic and geochemical studies have shown that the
Coastal Range is composed of three accreted Miocene–Pliocene
volcanic islands, three Plio-Pleistocene remnant forearc basins,
two intra-arc basins, and the Pliocene Lichi Mélange (Fig. 2).
The accreted Miocene–Pliocene volcanic islands (from north to
south: Yuehmei, Chimei, and Chengkuangao) are composed of
andesite, agglomerates, and tuff of the Tuluanshan Formation,
 119 o E 120 o E 121 o E 122 o E 123 o E

China V
V V
V
o
25 o N 25 N
Okinawa

ge
an
Trough

R
Taiwan Taiwan

n
hil ls
Strait

sh a
r al
ent

ueh
24o N

F o ot
24o N
C

Hs

Ra n .
ge
V
L.
Ry
uk
n
yu
er

ge

stal
Ran III
est

Tr e n
c h
C oa
W

23o N 23o N
ch

TT

Arc
n
Tr e

Philippine
SLT
Fig. 10

HTR

Sea
HP

II
K a lope
ing

22o N 22o N
Luzon
p
S
o

Hengchun

South
Ridge
a

N L T
l
ani

China
M

21o N I 21o N
North

Sea

o o o o o
119 E 120 E 121 E 122 E 123 E

Figure 1. Shaded relief map showing regional tectonics inland and off the shore of Taiwan and lines of three E-W seismic profiles. The South
China Sea oceanic crust is subducting eastward along the Manila Trench, developing the North Luzon Arc. The accretionary prism of the Manila
subduction system extends from the submarine Hengchun Ridge northward to the Miocene deep-sea sediments in the Central Range–Hengchun
Peninsula. The ophiolite-bearing Kenting Mélange in the Hengchun peninsula represents the subduction complex within the accretionary prism.
Subsequently, the North Luzon Arc collided obliquely with the underthrusting Eurasian continent north of 21° N. The collision resulted in de-
formation of the western part of the North Luzon Trough as the Huatung Ridge backthrusted eastward, thereby closing the North Luzon Trough
from south to north. The Luzon Arc and forearc are accreting on eastern Taiwan, as the Coastal Range and the Huatung Ridge further extend
directly northward as the Lichi Mélange in the southernmost Coastal Range. I—intraoceanic subduction zone; II—initial arc-continent collision
zone; III—advanced arc-continent collision zone; LM—Lichi Mélange; KM—Kenting Mélange; NLT— North Luzon Trough; HTR—Huatung
Ridge; TT—Taitung Trough; SLT—Southern Longitudinal Trough; HP—Hengchun Peninsula; L.V.—Longitudinal Valley.
130 Huang et al.

Chen et al., 1990; Huang et al., 1992, 1995, 2000). In the Coastal
TECTONOSTRATIGRAPHIC Range the chaotic Lichi Mélange, lying between the remnant Loho
o
MAP OF 24 and Taiyuan forearc basins to the east and the collision suture of
THE COASTAL RANGE Hualien the Longitudinal Valley to the west (Fig. 2), has long been con-
sidered the key to understanding the arc-continent tectonic pro-
Longitudinal Valley v
cesses in Taiwan (Hsu, 1956; Biq, 1971, 1973; Wang, 1976; Page
v

E
Post-collision Sequence v and Suppe, 1981). The Lichi Mélange occurs in a zone ~2 km

NG
in the suture basin v
v
wide with characteristics of block-in-matrix fabrics, preferred
Forearc Basin Sequence foliation in scaly argillaceous matrix, and extensional web and
RA v
v
Shuilien Basin Yuehmeiv Shuilien boudinage structures in the sandstone blocks (Fig. 4; Chen, 1997;
v Chang et al., 2000). The variously sized blocks (millimeters to
L

Loho Basin v
RA

kilometers) of different lithologies in the Lichi Mélange include

v
Taiyuan Basin the andesite suite (andesite, volcanic breccias, tuffs, and volca-
NT

v
v
v
Collision Complex niclastic turbidites), the ophiolitic suite (serpentine, gabbro, and
CE

Lichi Melange
pillow basalt), and the sedimentary suite (sandstone, sandstone-
v
shale interbeds, shale, and limestone; Biq, 1971; Wang, 1976;
Luzon Volcanic Arc v
Hsu, 1976; Liou et al., 1977; Page and Suppe, 1981). The andesite
lt
au

v
v v Yuehmei
ei F

Volcanic island v v suite is derived from the Luzon Volcanic Arc, whereas the origin
im

+ v
of the dismembered ophiolite blocks may represent the oceanic
Ch

v
KLS
Chimei +
t

v v volcanic island + +
aul

+
+ +
+ crust of either the South China Sea (Suppe et al., 1981; Chung
y f

v Chengkuangao Chimei v
TLS + and Sun, 1992) or the Philippine Sea plate beneath the Luzon
alle

v v
v +
volcanic island v +
al V

v v v forearc-Arc (Juan et al., 1980; Malavieille et al., 2002). On the

Intra-Arc v v v
din

v
v v
v v other hand, the sedimentary blocks include the weakly lithified
gitu

Chingpu Basin v v
v
v Pliocene turbidite units (tens of meters to a kilometer in size) with
Lon

v v
Chengkung Basin v
similar lithology, age, and sedimentary turbidite structures as the
Loho v
v
remnant-coherent-forearc-basin strata of the Coastal Range, and
the angular, well-lithified, whitish quartz-rich, feldspathic sand-
lt

SEA
fau

v
v
stones of late Miocene age (meters to kilometers in size). The lat-
an

Antung v v
sh

v v
an

v
v
ter unit is similar to the deep-sea-fan sandstones in the upper part
lu

Fuli v
v v
Tu

v
Chengkuangao of the accretionary prism in the Hengchun Peninsula (southern
NE

v v
v
v
v
v
v
extension of the Central Range; Cheng et al., 1984; Huang et al.,
Chengkung 1997) and is not observed in the coherent-forearc-basin turbidites
PPI

v
v
v
v
v (Pliocene–Pleistocene age) of the Coastal Range.
Kuanshan v
Fig. 9B
I LI

v
Taiyuan The origin of the Lichi Mélange has been variously inter-
o o
23 Fig. 8 23 preted (Fig. 5). It was first considered to be a subduction complex
Tungho
PH

v
v v
v v
that developed in the former Manila Trench during subduction of
v
v v the South China Sea oceanic crust in the Pliocene–Pleistocene
v
v
(Fig. 5A; Biq, 1971, 1973) before the arc-continent collision.
ange

v
Then a sedimentary slumping olistostrome model was proposed
an ge

N
ange

(Fig. 5B; Wang, 1976; Ernst, 1977; Page and Suppe, 1981; Lin
t al lRR
C en tr al R
CCooaasst a .V.

Lichi Fig. 9A
L

0 10 km and Chen, 1986). The Lichi Mélange was regarded as a sedi-

Taitung
121 30'
Figure 2. Tectonostratigraphic map of the Coastal Range (modified
from Huang et al., 1995). Rectangles denote locations of Figures 8
and 9A, B.
whereas the remnant forearc basins (Shuilien, Loho, and Taiyuan)
and the intra-arc basins (Pliocene Chingpu basin on Chimei vol-
canic island and Pleistocene Chengkung basin on Chengkuangao
volcanic island) are filled with turbidites (Takangkou and Chimei
Formations) derived from the accretionary prism. (Figs. 2 and 3;
mentary olistostrome deposited in the western part of the North
Lutao
Luzon Trough forearc basin by eastward mass wasting from
o
the exposed accretionary prism (the Central Range) during the
Pliocene–Pleistocene arc-continent collision, thus displaying a
facies change with the host non-sheared coherent-forearc-basin
strata to the east (Fig. 5B). These two disparate models were
proposed on the basis of on-land field surveys and mineralogical
studies of the dismembered ophiolite blocks and large sandstone
blocks in the Lichi Mélange without any constraint from regional
marine geological considerations. Intensive marine surveys off
southern Taiwan in the early 1990s began to provide detailed
bathymetric maps and regional tectonic structures. These investi-
gations do not support the tectonic subduction origin of the Lichi
 Biostratigraphy
LITHOSTRATIGRAPHY
EPOCH

TIME (Ma)
Longitudinal COASTAL RANGE
Valley

NORTH MIDDLE SOUTH

Suture Collision
Basin Melange Shuillien Chingpu Loho Taiyuan Chengkung
Forearc Intra-arc Forearc Forearc Intra-arc
Sequence Basin Sequence Sequence Basin
Pinanshan
PLEISTOCENE

0.7 Conglomerate

NN19 1.1 Chimei

N22 Chimei Formation
1.5 Formation
Takangkou
CN13 1.9
Formation
NN18 Chimei
NN17 2.3 Shuilien Takangkou
§Ê
Shuilien
Takangkou (Arc collapse)
Conglo- Conglomerate (Intra-arc)
N21 merate Takangkou
PLIOCENE

CN12 2.7
Formation Formation Formation Tungho Ls.
NN16
3.1
§¥§º
Formation
Takangkou
Tungho Ls.
Takangkou Fomation
Formation
Formation
3.5
NN15 CN11 Lichi Melange
N19/20

NN14 3.9
C (Arc collapse)
NN13 4.3
(Intra-arc)
CN10

4.7
(Chengkuangao Volcanic Island
B
NN12 5.1 Tuluanshan Formation
A
N18 ¥§•¶©Ls.
Kangkou
MIOCENE

5.5
NN11 ( Yuehmei ( Chimei
6 Volcanic
Volcanic
Island ) Island )
15 Tuluanshan Formation

Figure 3. Tectonostratigraphy of the Coastal Range.

Figure 4. Occurrence of angular, sheared,

whitish quartz-rich, feldspathic sand-
stone blocks in the intensely sheared
block-in-matrix mélange facies (δ-grade)
of the Lichi Mélange at sample location
LM-4 (location shown in Fig. 9A). The
large sandstone block with a black dot in
the center is ~8 m wide.
 A Subduction Model (Plio-Pleistocene)
Figure 5. Models proposed for the origin of the
Subduction Complex North Luzon Trough Luzon Arc
Lichi Mélange. (A) Tectonic subduction com-
Lichi Melange forearc Basin Tuluanshan Fm.
plex model (Biq, 1971, 1973; after Chang et al.,
South China Sea
Takangkou & 2001). (B) Sedimentary slumping olistostrome
Chimei Fms.
~ V Huatung Basin model (Page and Suppe, 1981). (C) Tectonic col-
~ V V
~ ~~ ~
V
V
V V
V
Arc volcanics V lision complex model (modified from Chang et
~ V
al., 2001): (1) continuous turbidite sedimentation
in the whole North Luzon Trough forearc basin
(3.5–3.7 Ma); (2) arcward (eastward) backthrust-
Eurasian Plate Philippine Sea Plate ing, resulting in formation of the proto–Lichi
Mélange, similar to the present-day Huatung
Ridge in the western part of the forearc basin;
the backthrusting also resulted in emplacement
of the deep-sea-fan sandstone from the accre-
tionary prism in the Hengchun Peninsula to the
B Slumping Olistostrome Model (Pliocene) proto–Lichi Mélange (3–2 Ma); (3) westward
accretion and thrusting of the Luzon Arc-forearc
Uplifted accretion ary prism
to form the southern Coastal Range (<1 Ma) and
(Central Range) Coastal Range formation of the modern Lichi Mélange (for de-
North Luzon Trough
tails, see text). L—lower part of remnant forearc
NW Erosion Forearc Basin SE strata; U—upper part of forearc strata; 1—ac-
Sub
mar
cretionary prism; 2—Lichi Mélange; 3—Luzon
ine s
lide Arc; 4—Pinanshan Conglomerate in suture basin
flow ~ 2000 m
Mixed Terrain of of the Longitudinal Valley. Blocks in the Lichi
coh ern t
fore arc tur bid ites Mélange: a—deep-sea sandstone in accretionary
Continent -derived Turbiditic Rocks Lichi a t io n prism of Hengchun Peninsula; b—oceanic crust
ne

Melange Form
Ta ka ng ka ou
Zo

and Ophiolitic Rocks V V

Formation V of Luzon Arc-forearc; c—volcanic rocks of Lu-
ult

V V V
V
Tuluanshan
V + +
V V V
V
V
V
V
zon Arc.
Fa

V V V
+Chimei+
V V V V V V V V
+ + + + +
V V V V V Igneous complex V V

C Tectonic Collision Model (Plio-Pleistocene)

Legend
(3) <1 Ma
Longitudinal

Coastal Range (southern part)

(Taiyuan Basin)

Volcanic Island)
(Chengkuangao
W

Forearc Basin
E
Accretionary

Longitudinal Che
n

Luzon Arc
Volc gkuang
Valley

Valley Lichi Melange Taiyuan

anic a
Basin
Islan o
Prism

d
Lichi
Centr
U Melange
L
~ ~~
al Ra ~ ~ ~~ 4
~

~
~~

~~~~~
~ ~

nge
~
~

~ ~ ~~ ~ ~ ~
~

~ c U
North Luzon Arc 2 b L 3
~~

Taiyuan
Trough Ch 1 a
(2) 3 ~ 2 Ma (Remnant North Vo engk
Proto-Lichi Melange Luzon Trough) lca ua
Hengchun nic nga
Peninsula
Southern Isla o
e)
(Huatung Ridg
Remnant Forearc Basin
Longitudinal nd
Trough

Accre Suture L
U

tiona basin

ry Pri L

sm North
Luzon Arc

(1) 3.5 ~ 3.7 Ma on Trough

Hengchun
Hengchun North Luz
Peninsula
Peninsula

Acc Forearc Basin

retio L
nary canic bas
ement

Pr
st ~ Vol
anic cru
ism Oce
 The Lichi Mélange
Mélange, because the Lichi Mélange lies to the east off the accre-
tionary wedge instead of within the accretionary prism (Huang
and Yin, 1990; Reed et al., 1992; Huang et al., 1992, 2000; Liu
et al., 1998; Malavieille et al., 2002). The marine surveys show
a direct connection between the submarine Huatung Ridge and
the Lichi Mélange in the southernmost Coastal Range (Fig. 6;
Huang et al., 1992; Huang, 1993). This led to the proposal of a
tectonic collision model, which suggests that the Lichi Mélange
originated from the thrusting and shearing of the forearc strata,
similar to the modern Huatung Ridge in the western part of the
North Luzon Trough during initial arc-continent collision in the
early Pliocene (Huang et al., 2000; Chang et al., 2000, 2001).
The purpose of this study is to integrate independent lines
of new evidence, including a new marine 240-fold multichannel
seismic profile across the North Luzon Trough as well as on-land
field observations and studies of the planktic and benthic fora-
miniferal fauna, clay mineral composition, and fission track pat-
terns of zircon grains in the Lichi Mélange to compare with the
Taiyuan remnant-forearc-basin strata to the east to constrain the
origin of the Lichi Mélange. Most study samples were collected
from the Mukeng River and its tributary south of Kuanshan in
133
the southwestern Coastal Range (Fig. 2), where the best sections
of the Lichi Mélange are exposed, showing different grades of
strata fragmentation, mixing, and shearing. The exposures here
led to various interpretations of the origin of the Lichi Mélange
in previous studies.
STUDY METHODS
Marine Surveys
A new multichannel seismic survey along a 21° N transect
across the North Luzon Trough off southern Taiwan (Figs. 1 and
7) was conducted in 1999, using the R/V Tanbao. The capabili-
ties of this vessel included 240 channels, an I/O digital streamer,
12.5 m of the interval between channels, a 50 m shot distance, a
250 m offset, 2 ms of sample, and 10 s of recording. The energy
source was a TI sleeve gun–airgun array. The total size was 3000
C.I., and the pressure was 2000 psi. The results of the survey were
then integrated with two 6-channel seismic profiles (MW9006–
26 and MW9006–31, Fig. 1; Chang et al., 2001) previously con-
ducted by the R/V Moana Wave in 1990 to show development of
Figure 6. Geological continuation from
the Lichi Mélange in the southernmost
Coastal Range to the submarine Hua-
tung Ridge (deformed forearc unit by
eastward thrusting) in the initial arc-
continent collision zone (marine image
adapted from Malavieille et al., 2002).
PC—Pinanshan Conglomerate in Lon-
gitudinal Valley.
 MW 9006-31 10km
Luzon
Southern
Two-way travel time (sec)

0.0 W Huatung Volcanic E 0.0

Longitudinal Ridge Arc
1.0 Trough 1.0
Taitung
2.0 Trough 2.0
Lutao
3.0 Volcanic 3.0
4.0 Island 4.0

5.0 5.0

MW 9006-26 Southern
0.0
W Longitudinal Huatung Luzon E
Trough Ridge Volcanic
1.0 Arc 1.0
Two-way travel time (sec)

2.0
North 2.0
Luzon
3.0 3.0
Trough Lanhsu
4.0 4.0
Volcanic
5.0 5.0
Island
6.0 6.0
7.0 7.0
8.0 8.0
GMGS 973
Hengchun Ridge
Accretionary Prism
W 8921 8961 9001 9041 9081 9121 9161 9201 9241 9281 9321 9361 9401 9441 9481 9521 9561 9601 9641 9681 9721 E

Luzon Volcanic Arc

Two-way travel time (sec)

3.0 Proto-Huatung 3.0

Ridge North Luzon Trough
4.0 6 Forearc Basin 4.0
5
4
5.0 3 5.0
2
1
6.0 6.0
Batan
Volcanic
7.0 Island 7.0

8.0 8.0

Figure 7. Three seismic profiles, from north to south (see Fig. 1 for location), showing progressive formation of the Huatung Ridge and closure of
the North Luzon Trough forearc basin. Line GMGS-973 further revealed syndeformational sedimentation in the North Luzon Trough, regardless
of deformation of the proto–Huatung Ridge in the western part of the trough filled with six sequences of strata separated by five unconformities
resulting from episodic arc-continent collisions.
 The Lichi Mélange 135

the Huatung Ridge along progressive closure of the North Luzon
Trough by arc-continent collision.
On-Land Study
Raymond’s (1984) classification of different grades of strata
fragmentation, mixing, and shearing—from coherent facies (α-
grade), weakly sheared broken formation facies (β-grade) to
strongly sheared dismembered facies (γ-grade), and intensely
sheared mélange facies (δ-grade)—was followed for mapping
the distribution and occurrence of the Lichi Mélange along the
Mukeng River sections (Fig. 8) and supplementary localities in
the southernmost Coastal Range (Fig. 9). Both planktic and ben-
thic foraminifers in 200 g of scaly argillaceous matrix (γ- to δ-
grade shearing) or the Bouma Te part of weakly sheared turbidite
layers (β-grade) were identified. Distributions of the main ben-
thic foraminifers were semiquantitatively plotted against stratig-
raphy according to their relative abundance. The results were then
compared with the benthic foraminifers in the coherent-forearc-
basin turbidites in the Taiyuan Basin (Chang, 1967). Multivari-
ate cluster analysis was used to compare the similarity of benthic
foraminifers within the Lichi Mélange. Benthic foraminifers that

Sector 1 Sector 2 Sector 3

(Fig. 12 A, B) (Fig. 13 A, B) (Fig. 14 A, B, C)
1 N

N
500 m 19
18 16
D

Fault
17 15
A B C 14
2 N
I IIa IIb
50
60
r
ve
4
85
Ri 5
04
65
Mushi Taiyuan
13
River

65 85
2 Bridge
11 Forearc
06 10
03 n g 12 Basin
40 u ke 07 09 IV
3 N 1 01 M 08 20
3 Nanshi
05 22 Bridge
21 65 24 H
Peinan

02

nge
23
I 25

Ra
85

27 26

al
II III
han
ntr
4 N
A Sector 3

nge
Ce
28
(Fig. 14 D, E)
s

Ra
C
Tu l u a n

IV
al
31 K
29 ast
30
32 33
35 D
Co

50
5 N 55 L
34 39 Taitung
36
60
37 40 Raymond (1984)
65 45

38
1 ~5 Stress study sites of Chang et al. (2001)

Sector 1 Sector 2
(Fig. 12 C, D) (Fig. 13 C,D)
- grade - grade - grade Sandstone Block Conglomerates Igneous Block

Figure 8. Detailed geological map and sampling locations of the study area along the Mukeng River and its southern tributary sections. Ray-
mond’s (1984) classification system in the right bottom corner is followed to map the distribution of various sheared units of the Lichi Mélange
(modified from Chang et al., 2000). The section is divided into three sectors according to their exposure sequences from west to east. Field oc-
currence and distributions of major benthic foraminifers are shown in Figures 12–14. K—Kuanshan; L—Lichi; H—Hualien.
 o o
121 15'E 121 17'30''E

lt
N Fuli V
BV

f au
g
V V o

fon
23 10'N

e
ng
0 1 km 1.0

Frequency
V V

Yu
40.5 Ma TF-6

Ra
12.5 Ma
V V 0.5 64.5 Ma
LM-10

al
2.5 Ma
TF-9 V V

n tr
0.0
TF-8 0 50 100 150 200 250 300
Ce 23 V V 1.0
3.0 Ma

Frequency
ey
TF-7
V
ll 70 V V
Va
0.5 15.6 Ma
V

lt
100 Ma
V V

fa u
al

V 0.0
d in

V V 0 50 100 150 200 250 300

i tu

2.72 Ma
V

an
76 1.0

Frequency
V V 5.8 Ma
ng

n sh V TF-9
Lo

l ua
0.5
V TF-7 92 Ma
Tu

g V V V
0.0
0 50 100 150 200 250 300
1.0 2.08 Ma

Frequency
V V 23 V V
TF-8
0.5
V
V V 0.0
g V V 0 50 100 150 200 250 300

L egend V V Age (Ma)

Gravel and sand TF-6
(Holocene) Takangkou Formation
Lichi Melange V V
g
With ophiolite 62 Taiyuan 1 .0 88 Ma

Frequency
Tungho Limestone o 18 Ma
(Late Pliocene) 23 N
Takangkou Formation
0 .5
LM-10
(Pliocene-Pleistocene)
V Tuluanshan Formation
(Miocene-Pliocene)
Slate and schist V V 0 .0
(Eocene-Miocene) Tungho 0 50 100 150 200 250 300
o o
121 15'E 121 17'30''E Age (Ma)
Lichi Melange

N A 1 .0 120 Ma
F requency

60 Ma
30 58 LM-1
42 33 65
0 .5 12.8 Ma
40 50
40 65
0 .0
0 100 200 300 400 500
58 1 .0
Frequency

0 1 2 Km 9.6 Ma
120 Ma LM-2
PEI NAN RIVER

0 .5
12
48 20
72 0 .0
42 LM-5 0 100 200 300 400 500
60 1 .0
F requency

50 9.76 Ma LM-3
Lichi 48 0 .5
110 Ma
LM-2 0 .0
0 50 100 150 200 250 300
50
1 .0
Frequency

22 66 Ma
LM-3 LM-4
LM-1 g g
0 .5 14.8 Ma

0 .0
LM-4 0 50 100 150 200 250 300
g Legend 1 .0
11 Ma
Frequency

Pin an sh an Con glome r st e

LM-5
L ic h i Me lan ge 130 Ma
0 .5
S an d st on e
Blocks in

L ime st on e
melange

Agglome r at ic an d e sit e
B asalt 0 .0
G ab b ro 0 50 100 150 200 250 300
S e r p e n t in e , Pe r id ot it e
Tak an gk ou For mat ion Age (Ma)
Lichi Melange

Figure 9. Fission track records of zircon grains separated from whitish quartz-rich feldspathic sandstone blocks in
the Lichi Mélange (A) and from the coherent forearc turbidite layers of the Taiyuan remnant forearc basin (B).
 The Lichi Mélange 137

occur in <5 samples, and the samples that contain <50 specimens
of benthic foraminifers in 200 g of sediments, were deleted dur-
ing computation of the cluster analysis.
Mineral compositions of the clay fraction (<2µm) of the col-
lected samples were also identified by X-ray powder diffraction
following the method described by Lin and Chen (1986). The
results were then compared with those of Lin and Chen (1986),
who had systematically documented the clay mineral composi-
tions in the Lichi Mélange and the coherent-forearc-basin strata
in the southern Coastal Range.
In addition, fission tracks of the zircon grains separated
from the Bouma Ta part of the Taiyuan remnant-forearc-basin
turbidites were counted to compare with the results obtained
from the whitish quartz-rich feldspathic sandstone blocks in
the sheared Lichi Mélange near Fuli and the type locality of the
Lichi Mélange along the Peinan River (Fig. 9). The detection and
counting methods of the fission track age analysis described by
Liu (1982) and Liu et al. (2001) were followed. The results of the
zircon fission track records from both the Lichi Mélange and the
Taiyuan remnant-forearc-basin turbidites were then compared
with the samples collected from the nonmetamorphosed deep-sea
sandstones in the accretionary prism of the Hengchun Peninsula
(Fig. 10) to constrain the origin of the whitish sandstone blocks
in the Lichi Mélange.
RESULTS
Marine Surveys off the Southern Coastal Range
The new 240-fold multichannel seismic profile (GMGS-973
in Fig. 1) across the North Luzon Trough (forearc basin) along
the 21° N transect is shown in Figure 7. The profile shows syn-
deformational sedimentation in the North Luzon Trough. The
forearc basin is filled by six sequences of strata separated by five
unconformities. All of these six sequences in the western part of
the North Luzon Trough are tilted upward and deformed gently
by eastward thrust, whereas the strata in the eastern part of the
trough remain flat, suggesting no deformation at all (Fig. 7).

O
22 10'N 1.0
50-65 Ma

Frequency
HP-3
HP-4 5 km 0.5
HP-4
Shallow - marine slope basin

LEGEND
Mt
Holocene

0.0
Recent 0 50 100 150 200 250 300
Alluvium 1.0
HP-3
HP-2 15-30 Ma
Taiping Fm. Frequency 85-100 Ma
Pleistocene

Szekou Fm.
Ls
Hengchun 0.5
Limestone HP-3
~~ unconformity ~~
Pliocene

5 Kt Ma Maanshan Fm.
0.0
~~ unconformity ~~ 0 50 100 150 200 250 300
15-25 Ma
1.0
Slope and trench-fill turbidites

55-65 Ma
Frequency

Hengchun
HP-2 Kenting Melange
in accretionary prism

Kt
Middle - Late Miocene

O
22 00'N (subduction complex)
0.5 HP-2
Mt Mutan Fm.
Ma HP-1
Mt lower fan turbidites
7
upper- middle fan
conglomerates and
sandstones 0.0
0 50 100 150 200 250
Loshui Fm.
Ls 1.0
middle fan sandstones 15-30 Ma
Frequency

Hengchun Peninsula 70-85 Ma

0.5
HP-1

O O
120 40'E 120 50'E
0.0
0 50 100 150 200 250
Age (Ma)
Figure 10. Fission track records of zircon grains separated from deep-sea-fan sandstones in the accretionary prism of Hengchun Peninsula. The
fission-track age pattern of these deep-sea sandstones is similar to that obtained from the whitish quartz-rich feldspathic sandstone blocks in the
Lichi Mélange (Fig. 9A).
138 Huang et al.
Geology of the Lichi Mélange Exposed along the
Mukeng River
The Lichi Mélange is exposed in the southwestern and south-
ernmost parts of the Coastal Range (Figs. 1 and 2). The Lichi
Mélange lies between the Longitudinal Valley to the west and the
remnant forearc basins (Loho Basin and Taiyuan Basin) and vol-
canic rocks to the east (Fig. 2). In the northern Coastal Range the
Lichi Mélange could have been eroded away by the river flowing
northward along the eastern side of the Longitudinal Valley. The
Longitudinal Valley, a 2–3-km-wide collision suture between the
Central Range (accretionary prism and exhumed metamorphic
basement) and the Coastal Range (accreted arc-forearc; Fig. 2)
accommodates a part of the shortening across southeastern Tai-
wan (Yu et al., 1997). Between Antung and Kuanshan the volca-
nic basement beneath the Taiyuan forearc strata has been thrust
westward over the Lichi Mélange along the Tuluanshan Fault
(Fig. 2). Consequently, the Tuluanshan volcanics are exposed
between the Lichi Mélange to the west and the remnant Taiyuan
forearc basin to the southeast (Fig. 2). In the Coastal Range this
is the only place where the volcanic rocks appear west of the
remnant forearc basins, and they lie between the Lichi Mélange
and the Taiyuan remnant forearc basin. Otherwise, the volca-
nic andesitic rocks are always observed to the east of remnant-
forearc-basin strata (Fig. 2; Hsu, 1956, 1976; Huang et al., 1995).
The east-dipping Tuluanshan Fault runs in a NNE-SSW direc-
tion, representing the last accretion of the Luzon Arc-forearc to
the exhumed Eurasian continent (eastern Central Range) as the
southern part of the Coastal Range in the last 1 m.y. (Chang et al.,
2000, 2001; Huang et al., 2000, 2006).
Figure 8 shows a detailed geological map of the Lichi
Mélange exposed along the Mukeng River and its southern trib-
utary in the southwestern Coastal Range (Chang et al., 2001).
The coherent turbidite sequence (Takangkou Formation) of the
Taiyuan remnant forearc basin is thrust westward over the Lichi
Mélange along the Tuluanshan Fault (Fig. 8). West of this fault
the study section shows various degrees (β-, γ-, and δ-grades of
Raymond, 1984) of strata fragmentation, mixing, and shearing.
The sequences with γ-grade and δ-grade shearing (dismembered
and mélange facies) exhibit characteristics of block-in-matrix
features (Fig. 4), with pervasively scaly foliation dipping to the
southeast (measurements of foliations at five localities are shown
in Figure 8; Chang et al., 2000) without discernible bedding (units
I, II, III, and IV in Fig. 8), whereas the weakly disturbed (β-grade
shearing or broken formation facies) sequences (units A, B, C,
and D in Fig. 8) still preserve distinct Bouma turbidite sequences
and clear bedding, which generally strike NE and dip SE. How-
ever, faults for various thickness are commonly observed in the
intervals of different grades (centimeters thick in β-grade, several
meters thick in γ-grade, and hundred meters thick in δ-grade) of
shearing units and also along the boundaries between the dif-
ferent units (Fig. 11; Chang et al., 2001). From west to east the
study profiles are divided into three sectors (Fig. 8) for convenient
comparisons and discussion. Sector 1 includes a γ-grade unit and
a β-grade unit (units I and A, respectively), whereas sector 2 is
composed of two γ-grade units (units IIa and IIb) and two β-grade
units (units B and C) in the north (Fig. 8). Units IIa and IIb are
combined together as the thick δ-γ-grade unit (unit II) in the south.
Sector 3 displays a γ-grade unit III, a thick and less disturbed (β-
grade) unit D, and a thick (~500 m along the river), highly sheared
Figure 11. Ubiquitous shearing planes
with fault gouges in the Lichi Mélange,
especially in mélange facies unit IV (δ-
grade of Raymond, 1984; Fig. 8).
 The Lichi Mélange
(γ- and δ-grades) unit IV directly west of the Tuluanshan Fault
(Fig. 8). Angular ophiolite blocks, predominantly of gabbro and
whitish sandstone blocks, are found restrictedly in units I, II, III,
and IV with dismembered facies or mélange facies. In contrast,
pebbly mudstone layers composed of subangular to subrounded
metasandstone conglomerates are found only in the less disturbed,
broken formation facies of units A, C, and D (Fig. 8).
Planktic Foraminifers and Age Determination of the
Lichi Mélange
Planktic foraminifers, including Sphaeroidinella dehiscens,
Globorotalia tumida, G. multicamerata, Neogloboquadrina
altispira, and Sphaeroidnellopsis seminulina, are commonly
found in most of the samples from the Lichi Mélange along the
Mukeng River and its tributaries. They show a persistent early
Pliocene age (upper Zone N19-20), consistent with results
obtained previously by calcareous nannoplankton (Zone NN15;
Chi et al., 1981; Chi, 1982; Barrier and Müller, 1984).
Benthic Foraminifers in Various Degrees of Sheared Units
of the Lichi Mélange
Distributions of benthic foraminifers in the various degrees of
sheared strata in the Lichi Mélange along the Mukeng River and
its tributary are shown in Figures 12–14. As a result of turbidite
deposition, the benthic foraminifers in the Lichi Mélange are the
mixed fauna of the indigenous deep-water taxa and the displaced
shallow-marine species. The deep-marine fauna include the cos-
mopolitan species of the genera Bulimina, Uvigerina, Fontbotia,
Pullenia, Hyalinea, Melonis, Gyroidinoides, and Cyclammina,
living on the continental slope to abyssal plain of the present
oceans. The shallow-marine fauna include species of the genera
Amphistegina, Ammonia, Operculina, Asterorotalia, Pseudorota-
lia, Calcarina, Elphidium, Lenticulina, and Cellanthus, living in
warm waters of the shallow shelf (<200 m). In each study sample,
individuals of these shallow- and deep-marine genera make up
60%–80% of the total benthic foraminiferal specimens (Chien,
2003). For convenient discussion, the species with similar ranges
of water depth or ecology are shown together in Figures 12–14.
Clay Mineral Composition
Clay mineral compositions of the sheared mudstone in the
broken formation facies or scaly argillaceous matrix and the
quartz-rich sandstone blocks in the mélange facies of the Lichi
Mélange along the Mukeng River and its southern tributary are
listed in the Appendix. The samples are all characterized by bear-
ing illite (average, 51.5%), chlorite (average, 19.5%), mixed-
layer clay minerals (average, 11.7%), and kaolinite (average,
11.1%). Smectite is minor (<3%). Kaolinite is always found in
the samples of the Lichi Mélange regardless of the mudstones in
the broken formation facies, the scaly argillaceous matrix, or the
angular whitish sandstone blocks in the mélange facies.
139
DISCUSSION
Significance of Marine Geology off the Southern Coastal
Range in the Active Arc-Continent Collision Region
Marine seismic investigations off southeastern Taiwan reveal
the progressive closure of the forearc basin owing to the east-
ward (arcward) thrusting of the forearc basin strata to develop
the Huatung Ridge (Fig. 7; Reed et al., 1992). Profile GMGS-
973 along the 21° N transect across the North Luzon Trough
(Fig. 7), where the Luzon arc collides incipiently against the Eur-
asian continent, further shows that (1) active syndeformational
sedimentation occurs in the forearc basin; once the sediments are
deposited in the trough, the sequence in the western part of the
forearc basin is deformed and then covered unconformably by
the overlying sequence; (2) folding and thrusting in the western
part of the forearc basin have occurred since the early history
of forearc sedimentation; (3) sedimentation is continuous in the
eastern part of the forearc basin regardless of active deforma-
tion in the west; and (4) the eastward thrusting accommodates
the space between the backstop of the accretionary prism and
the Luzon Arc. In compiling two 6-channel seismic profiles in
the north (Fig. 7; MW9006–26 and MW9006–31; Chang et al.,
2001), regional marine geology shows that the deformed Hua-
tung Ridge forearc becomes larger and larger so that finally the
remnant North Luzon Trough is closed at its northern terminus,
approaching the southern Coastal Range (Fig. 6). Meanwhile, the
deformed ridge crest is thrust upward so that sediments can no
longer be deposited on the Huatung Ridge.
Because the Huatung Ridge connects northward with the
Lichi Mélange in the southernmost Coastal Range (Fig. 6), it is
conceivable to consider that the Lichi Mélange has originated
from the deformed forearc basin strata like the modern Huatung
Ridge and to predict that the Lichi Mélange could have the fol-
lowing geological characteristics:
A. The Lichi Mélange should lie west of the remnant forearc
basin;
B. The Lichi Mélange would contact the coherent forearc
basin strata by faulting;
C. The depositional time interval of the Lichi Mélange would
be less than that of the entire forearc basin sequences, and
the depositional age of the Lichi Mélange can be equiva-
lent only with the lower sequence, but older than the upper
sequence, of the remnant-coherent-forearc-basin strata;
D. Regardless of shearing intensity and facies, the microfos-
sils within the Lichi Mélange should be similar to and
compatible with the fauna in the remnant forearc basin to
the east;
E. The depositional bathymetry of the Lichi Mélange would
be similar to the present North Luzon Trough (1500–
3000 m);
F. The relict sedimentary structures and the host sediment
composition of the Lichi Mélange would be similar to
those of the coherent-forearc-basin turbidites;
140 Huang et al.

A
SHALLOW MARINE TAXA DEEP-SEA TAXA
Sheared

B
grades
Age

Sector Ammonia spp.

Amphistegina spp. Asterorotalia spp. Calcarina calcar Fontbotia wuellerstorfi
Lenticulina spp. Hyalinea balthica Pullenia bulloides Uvigerina spp. Gyroidinoides spp. Cyclammina spp.
1 Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Elphidium spp. Cibicidoides spp.
Melonis spp.

Unit Sample
4 00 m

Number
N19 /20 ; NN15 (3.7 ~ 3.5 Ma)

04-0
300 m

04-1
A
04-2

04-3
200 m

04-4

03
I
100 m

02

01
0m

Figure 12. (A) Examples of field occurrence and sampling localities of units I and A, respectively, in part of sector 1. (B) Distribution of major
benthic foraminifers along the Mukeng River section.

G. The early eastward thrusting in the forearc basin in the example, sedimentary slumping or mud diapirism—responsible
initial arc-continent collision stage and the final west- for formation of the Lichi Mélange.
ward thrusting of the Luzon Arc-forearc during the
advanced arc-continent collision accreted to the Coastal Significance of Field Surveys and Age Determination
Range could have resulted in accretions of the accre-
tionary wedge materials west of the forearc basin and Field surveys along the Mukeng River and its tributary show
the volcanic arc rocks beneath the forearc basin to the that the Lichi Mélange lies west of the Taiyuan remnant forearc
highly sheared, block-in-matrix, dismembered facies basin and is thrust over by the coherent-forearc-basin strata along
and mélange facies of the Lichi Mélange; however, these the Tuluanshan Fault (Fig. 8). Neither the Lichi Mélange nor the
blocks would not appear in the weakly sheared broken volcanic rocks are observed in the Taiyuan remnant forearc basin
formation facies of the Lichi Mélange or the remnant- east of the Tuluanshan Fault, consistent with predictions A and B
coherent-forearc-basin strata. in the preceding subsection.
If our different lines of new data agree well with these geo- Biostratigraphic study reveals that the protolith of the Lichi
logical characteristics or predictions listed above, we would Mélange was deposited in a narrow age range within Zone N19-20
prefer that the Lichi Mélange originated along the arcward of planktic foraminifers or Zone NN15 (3.5–3.7 Ma) of calcare-
thrusts during forearc basin closure by active arc-continent col- ous nannoplankton (Figs. 12–14; Chi et al., 1981; Chi, 1982; Bar-
lision. On the other hand, if the new data do not fit well with the rier and Müller, 1984). This age is coeval with the lower remnant
predictions, we would consider the other possible origins—for forearc basin sequence (Fig. 3) but is older than the upper forearc
 The Lichi Mélange 141

29

C
SHALLOW MARINE TAXA DEEP-SEA TAXA
D
Sheared
grades

Sector
Age

Ammonia spp.
Amphistegina spp. Asterorotalia spp. Calcarina calcar Fontbotia wuellerstorfi
1 Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Elphidium spp.
Lenticulina spp. Hyalinea balthica
Cibicidoides spp.
Pullenia bulloides
Melonis spp.
Uvigerina spp.Gyroidinoides spp. Cyclammina spp.

Unit Sample
300 m

Number
N19; / 20 NN15 (3.7 ~ 3.5 Ma)

29

A
20 0 m

28
1 00 m

I
27
0m

Figure 12. (C) Occurrence and imbricate structure of the fractured, subangular to subrounded metasandstone conglomerates in two pebbly mud-
stone layers of broken formation facies in unit A. (D) Distribution of major benthic foraminifers in part of sector 1 along the southern tributary
section (detailed position shown in Fig. 8).

sequence (Zones N21–22 of planktic foraminifers, Chang, 1967; Significance of the Benthic Foraminiferal Study
Zones NN16–19 of calcareous nannoplankton, <3.5–1.15 Ma,
Horng and Shea, 1996). This indicates that (1) the turbidites in Figure 15 shows the result of cluster analysis of the benthic
the proto–Lichi Mélange were deposited in a short time (~0.2 foraminifers in the Lichi Mélange. No samples of the mélange
m.y.), (2) deformation (at ca. 3.5 Ma) of the proto–Lichi Mélange facies (δ-grade of shearing) appear in Figure 15 because of their
occurred as soon as these turbidites were deposited (3.5–3.7 Ma), strong depression or their broken or incomplete preservation
and (3) after deformation of the proto–Lichi Mélange the young from intensive compression or shearing, which prevents precise
turbidites (3.5–1 Ma) were deposited continuously in the upper taxonomic determinations from the few fossil remains. There-
part of the Taiyuan remnant forearc basin, a scenario analogue fore, the samples collected from mélange facies are omitted from
with the modern configuration of the Huatung Ridge–remnant computation.
North Luzon Trough off the southern Coastal Range in the mod- The samples are clustered into two main groups. Up to 79%
ern active arc-continent collision zone (Fig. 7). The results of bio- of the samples are clustered together as group A, whereas 21% of
stratigraphic study are consistent with prediction C in the preced- the samples (only 8 samples) are clustered as group B. Samples
ing subsection. of group B were collected primarily from the weakly sheared,
142 Huang et al.

A
A
B SHALLOW MARINE TAXA DEEP-SEA TAXA
Sheared

Sector
grades
Age

Ammonia spp.
Amphistegina spp. Asterorotalia spp. Calcarina calcar Fontbotia wuellerstorfi
Pullenia bulloides
2 Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Elphidium spp.
Lenticulina spp. Hyalinea balthica
Cibicidoides spp.
Melonis spp.
Uvigerina spp. Gyroidinoides spp. Cyclammina spp.

Unit Sample
Number
11
500 m

10-1
N19 / 20 ; NN15 (3.7 ~ 3.5 Ma)

10-2
C
400 m

10-3
10-4
300 m

09-1

09-2
20 0 m

09-3
IIb
08-1
08-2
07-1
B 07-2
1 00 m

06-2
06-1
05
IIa
0m

Figure 13. (A) Examples of field occurrence and sampling localities of unit C. (B) Distribution of major benthic foraminifers along the Mukeng
River section in the north.

broken formation facies (β-grade) turbidites of unit C (6 sam- The result implies that the benthic foraminifers in the Lichi
ples) and two individual samples from the dismembered facies Mélange are similar to each other regardless of different degrees
(γ-grade) of unit IIb (sample 6–2) in sector 2 and unit III (sample of shearing intensity, consistent with prediction D in the first sub-
38–2) of sector 3 (Fig. 8). Foraminifers in group B predominate section under “Discussion.”
in the indigenous deep-water species without significant shal- Foraminifers in the Taiyuan remnant forearc sequence in
low-marine assemblages (Fig. 13B), whereas samples of group the southern Coastal Range were studied by Chang (1967).
A contain both the indigenous deep-water and the displaced Unfortunately, the abundance of each benthic foraminifer was
shallow-water fauna (Figs. 12–14). Moreover, each group A and reported semiquantitatively (R, <6; F, 6–10; C, >11–20; A, >21–
group B contains the samples collected from the units of the 50). A comparison of the present work with the result of Chang
dismembered facies and the broken formation facies (Fig. 15). (1967) can be made by cross-checking the faunal lists and
 The Lichi Mélange 143

D SHALLOW MARINE TAXA DEEP-SEA TAXA

Sheared

Age
grades

Sector Ammonia spp.

Amphistegina spp. Asterorotalia spp. Calcarina calcar Fontbotia wuellerstorfi
Lenticulina spp. Hyalinea balthica Pullenia bulloides Uvigerina spp. Gyroidinoides spp. Cyclammina spp.
2 Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Elphidium spp. Cibicidoides spp.
Melonis spp.

Sample
570 m

Unit Number
37-1
37-2
37-3
500 m

37-4

37-5
N19 / 20; NN15 (3.7 ~ 3.5 Ma)

C 36
400 m

35
33-1
33-2
33-3
33-4
30 0 m

34
200 m

32

II
100 m

31

30
0m

Figure 13. (C) Examples of field occurrence and sampling localities of unit C. (D) Distribution of major benthic foraminifers in part of sector 2
along the southern tributary section (detailed position shown in Fig. 8).

relative abundance of the major taxa in both studies (Table 1). used. Our study uses a newer classification of benthic
The results suggest that the benthic foraminifers in the Lichi foraminifers by Loeblich and Tappan (1988), whereas
Mélange (this study) and the Taiyuan remnant forearc basin tur- Chang (1967) followed an older classification of Loe-
bidites (Chang, 1967) are similar: blich and Tappan (1964).
1. The benthic foraminifers are all characterized by high 2. Planktic foraminifers are abundant (>85% of total fora-
diversity (Taiyuan forearc basin turbidites: 40 genera, miniferal individuals), and benthic foraminifers are pre-
68 species, in Chang, 1967; Lichi Mélange: 86 genera dominantly calcareous genera (>75%) in both Taiyuan
and 167 species, this study) and low abundance of each forearc turbidites and the Lichi Mélange.
species (85% species < 6 individuals/200 g of sediments 3. Foraminifers in both Taiyuan forearc turbidites and the
in the Lichi Mélange and coherent forearc basin turbi- Lichi Mélange are composed of indigenous deep-marine
dites). The difference in species number between these fauna and shallow-marine species (Table 1). The deeper
two studies is due to the different classification system marine taxa include Bulimina striata d’Orbigny, B. cf.
 A

C SHALLOW MARINE TAXA DEEP-SEA TAXA

Sheared
grades

Sector
Age

Ammonia spp.
Amphistegina spp. Asterorotalia spp. Calcarina calcar Fontbotia wuellerstorfi
Pullenia bulloides
Lenticulina spp. Hyalinea balthica Uvigerina spp.Gyroidinoides spp. Cyclammina spp.
3 Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Elphidium spp. Cibicidoides spp.
Melonis spp.

Unit Sample
Number
4 80 m

19
40 0 m

15-2
N19 / 20 ; NN15 (3.7 ~ 3.5 Ma)

D 15-1

16-1
300 m

16-2
16-3
16-4
16-5
18
17
2 00 m

14
14-1
III 13-1
13-2
12-1
1 00 m

12-2
12-3
12-4
12-5
12-6
0m

Figure 14. (A, B) Examples of field occurrence and sampling localities of units III and D, respectively. (C) Distribution of major benthic fora-
minifers in part of sector 3 along the Mukeng River section in the north.
 26
D

E Sector
SHALLOW MARINE TAXA DEEP-SEA TAXA
Sheared
grades
Age

Ammonia spp.
3 Amphistegina spp. Asterorotalia spp. Calcarina calcar
Operculina spp. Pseudorotalia spp. Cellanthus craticulatus
Fontbotia wuellerstorfi
Lenticulina spp. Hyalinea balthica Pullenia bulloides
Melonis spp.
Uvigerina spp.Gyroidinoides spp. Cyclammina spp.
Elphidium spp. Cibicidoides spp.

Unit Sample
7 00 m

Number
600 m

IV 26
500 m

N19 /20 ; NN15 (3.7 ~ 3.5 Ma)

25
400 m

24
30 0 m

D
200 m

23
100 m
0m

22

Figure 14. (D) Examples of field occurrence and sampling locality 26 in mélange facies of unit IV. (E) Distribution of major benthic foramini-
fers in part of sector 3 along the southern tributary of the Mukeng River (detailed position shown in Fig. 8).
146 Huang et al.

Rescaled Distance Cluster Combine

0 5 10 15 20 25
+ + + + + +
8-1 Mudstone
8-2 Shale
9-1 Mudstone Group B
10-4 Mudstone
6-2 Shale
10-3 Mudstone
38-2 Mudstone
10-1 Mudstone
12-2 Mudstone
15-1 Sandstone
11 Mudstone
13-2 Sandstone
4-0 Siltstone
4-1 Sandstone Figure 15. Multivariate cluster analysis
4-2 Sandstone of the samples collected from the Lichi
37-4 Shale Mélange, based on benthic foraminiferal
37-5
Group A assemblage. The study samples are clus-
Mudstone
5 Mudstone tered into two main groups. Each group
16-1 Mudstone includes samples with various degrees of
16-2 Mudstone shearing. For details of both groups, see
15-1 Shale the text.
18 Mudstone
37-3 Shale
2 Mudstone
4-4 Mudstone
28 Mudstone
27 Mudstone
16-4 Mudstone
38-5 Mudstone
31 Mudstone
16-5 Mudstone
17 Mudstone
19 Mudstone
25 Mudstone
36 Mudstone
23 Mudstone
24 Mudstone
38-1 Mudstone
14 Mudstone

rostrata Brady, Uvigerina probocidea Schwager, U. his-
pida Schwager, U. subperegrina Cushman and Klein-
pell, Fontbotia wuellerstorfi (Schwager), Gyroidinoides
orbicularis (d’Orbigny), Pullenia bulloides (d’Orbigny),
Melonis nicobarensis (Cushman), and M. pompilioides
(Fichtel and Moll). These deep-marine benthic foramin-
ifers are cosmopolitan species living in water depths of
1500–3000 m of modern deep oceans (Van Morkhoven
et al., 1986). The shallow-marine species are Amphi-
stegina radiata (Fichtel and Moll); A. lessonii d’Orbigny;
Operculina complanata (Defrance); O. ammonoides
(Gronovius); Asterorotalia gaimardii inermis Billman,
Hottinger and Oesterle; A. subtrispinosa (Ishizaki);
A. yabei (Ishizaki); Pseudorotalia indopacifica (Thal-
mann); P. schroeteriana (Parker and Jones); Calcarina
calcar d’Orbigny; Cellanthus craticulatus (Fichtel and
Moll); Elphidium crispum (Linné); and Lenticulina
limbosa (Reuss). Today, these shallow-marine benthic
foraminifers are typical of the Kuroshio Current fauna
living in the outer shelf facing the open ocean (Wang,
1985; Huang, 1989).
4. The indigenous deep-marine benthic foraminifers in the
Lichi Mélange and the Taiyuan remnant forearc basin tur-
bidites are also the major fauna recovered in the present
hemipelagic muds of the North Luzon Trough (Huang,
1993), whereas the shallow-marine species are commonly
found ubiquitously in the turbidite layers off southern and
northeastern Taiwan, including the North Luzon Trough
forearc basin and the Okinawa Trough backarc basin.
They were displaced from the shallow shelf by turbidity
currents (Huang et al., 2005; Chiu, 2005).
The conclusions of this benthic foraminiferal study thus
indicate that the host rocks of the Lichi Mélange were part of
the North Luzon Trough forearc sequences before deformation.
The conclusions agree well with predictions D and E in the first
subsection under “Discussion.”
 The Lichi Mélange
TABLE 1: COMPARISON OF BENTHIC FORAMINIFERAL TAXA AND THEIR ABUNDANCE
IN THE LICHI MÉLANGE (THIS STUDY) AND THE TAIYUAN FOREARC TURBIDITES (CHANG, 1967)
Lichi Mélange
(this study)
Shallow-marine taxa:
Amphistegina radiata (Fichtel and Moll)
Amphistegina lessonii d’Orbigny
Operculina complanata (Defrance)
Operculina ammonoides (Gronovius)
Ammonia takanabensis (Ishizaki)
Asterorotalia gaimardii inermis Billman,
Hottinger and Oesterle
Asterorotalia subtrispinosa (Ishizaki)
Asterorotalia yabei (Ishizaki)
Pseudorotalia indopacifica (Thalmann)
P. cf. schroeteriana (Parker & Jones)
Calcarina calcar d’Orbigny
Cellanthus craticulatus (Fichtel and Moll)
Elphidium crispum (Linné)
Lenticulina limbosa (Reuss)
Lenticulina orbicularis (d’Orbigny)
Lenticulina iota (Cushman)
Lenticulina cf. cushmani (Galloway and Wissler)
Lenticulina spp.
Deep-marine taxa:
Bulimina striata d’Orbigny
Fontbotia wuellerstorfi (Schwager)
Hyalinea balthica (Gmelin)
Cibicidoides cf. crebbsi (Hedberg)
Cibicidoides subhaidingerii (Parr)
Gyroidinoides neosoldanii (Brotzen)
Gyroidinoides orbicularis (d’Orbigny)
Pullenia bulloides (d’Orbigny)
Melonis nicobarensis (Cushman)
Melonis pompilioides (Fichtel and Moll)
Uvigerina asperula Czjzek
Uvigerina hispida Schwager
Uvigerina peregrina Cushman
Uvigerina proboscidea Schwager
Uvigerina subperegrina Cushman and Kleinpell
Cyclammina japonica Asano kaiensis Fukuta and
Shinoki
Cyclammina tani Ishizaki
Cyclammina cf. trullissata (Brady)
Sediment Composition
Our field surveys recognize that three distinct end-member
facies are present in the Lichi Mélange: weakly sheared broken
formation facies (β-grade), strongly sheared dismembered facies
(γ-grade), and highly sheared mélange facies (δ-grade). Like the
coherent forearc basin turbidites east of the Tuluanshan Fault,
the weakly sheared, broken formation facies still preserves dis-
tinct turbidite sedimentary structures like the Bouma turbidite
sequence Ta–Te, and bedding without any angular sheared ophi-
olite or sedimentary blocks. The turbidite layers of the broken
formation facies are primarily of the classical sandy flysch type
(Figs. 12–14; C and D facies of Mutti and Lucchi, 1978). The
basal part of these classical turbidite layers (Bouma Ta–Tc) is
composed primarily of quartz, like the lower part of the coher-
ent forearc sequences. Slate chips, which are commonly found in
147
Taiyuan Forearc Basin
(Chang, 1967)
Shallow-marine taxa:
Amphistegina radiata (Fichtel and Moll)
Operculina ammonoides (Gronovius)
Pseudorotalia gaimardii compressiuscula (Brady)
Pseudorotalia schroeteriana (Parker & Jones)
Elphidium sp.
Lenticulina atlantica (Barker)
Lenticulina calcar (Linné)
Lenticulina iota (Cushman)
Lenticulina nikobarensis (Schwager)
Lenticulina spp.
Deep-marine taxa:
Bulimina striata d’Orbigny
Fontbotia wuellerstorfi (Schwager)
Gyroidinoides orbicularis (d’Orbigny)
Pullenia bulloides (d’Orbigny)
Melonis pompilioides (Fichtel and Moll)
Uvigerina (Siphouvigerina) ampullacea Brady
Uvigerina hispida Schwager
Uvigerina crassicostata Schwager
Uvigerina excellens Todd
Uvigerina proboscidea Schwager
Uvigerina subperegrina Cushman and Kleinpell
Cyclammina orbicularis Brady
Cyclammina pusilla Brady
Cyclammina tani Ishizaki
the upper coherent forearc basin turbidites (<3 Ma; Teng, 1982),
are never observed significantly in the Lichi Mélange. However,
lenses of pebbly mudstones (each layer 1–2 m thick; Fig. 12C)
are found in the weakly sheared broken formation facies (units
A, C, and D in Fig. 8). These subangular to subrounded, matrix-
supported conglomerates (each pebble 5–30 cm in size) are
composed primarily of metasandstone and are commonly frac-
tured (Fig. 12C). The orientation of these conglomerates shows
imbricate structure (Fig. 12C), representing typical channel fill-
ings derived from the exposed Eocene metasandstone beds in the
eastern Central Range. Similar pebbly mudstone layers with a
composite thickness of 150 m can be traced in the lower part of
the entire Taiyuan remnant forearc basin (Yao et al., 1988). Such
feeder-channel conglomerate beds, generally called the Shuil-
ien Conglomerate, occur in stratigraphic levels in the remnant-
forearc-basin sequences of the Coastal Range (Fig. 3).
148 Huang et al.
Therefore, sedimentological features, including occurrence
and composition of conglomerates in the pebbly mudstone lay-
ers and the turbidite structures preserved in the broken formation
facies, support the view of the Lichi Mélange as part of the North
Luzon Trough forearc basin sequences in agreement with predic-
tion F in the first subsection under “Discussion.”
Significance of Clay Mineral Composition
Clay mineral compositions in the Coastal Range were stud-
ied by Wang and Yang (1975), Lin and Chen (1986), Buchovecky
and Lundberg (1988), and Yao et al. (1988). Lin and Chen (1986)
documented systematically that the clay mineral compositions of
the muddy matrix and sedimentary blocks in various facies of
the Lichi Mélange were similar to each other in containing illite,
chlorite, kaolinite, and a little semectite, which are similar to the
clay mineral assemblages in the Taiyuan remnant-forearc-basin
turbidites except for kaolinite. This was not expected, because
sedimentological and micropaleontological studies all show that
the Taiyuan forearc basin turbidites and the Lichi Mélange were
derived from the same source and were all deposited in the same
North Luzon Trough. Consequently, they should contain simi-
lar clay mineral composition just as they contain similar pebble
composition and the benthic foraminiferal fauna.
Lin and Chen (1986) claimed that the Lichi Mélange
could have two sediment sources: a continental, low-grade-
metamorphosed sediment source (accretionary prism of the Cen-
tral Range) to provide abundant illite and chlorite, and a weath-
ered, low-latitude, high-rainfall archipelagic volcanic source for
derivation of kaolinite by eastward sedimentary mass slumping.
In contrast, the Taiyuan coherent-forearc turbidite sequences
could have only a continental low-grade-metamorphosed sedi-
ment source (Yao et al., 1988; Buchovecky and Lundberg, 1988).
Therefore, there is no kaolinite in the Taiyuan coherent-forearc-
basin sequences. Our present work confirms the results of Lin
and Chen (1986) that there is significant kaolinite in the Lichi
Mélange, regardless of shearing intensity, and agrees that the Lichi
Mélange and the Taiyuan remnant-forearc-basin strata were all
deposited in the North Luzon Trough. However, marine geology
off the Coastal Range shows that the North Luzon Trough forearc
basin lies west of the Luzon Volcanic Arc (Figs. 1 and 2). If the
volcanic arc has provided the weathered kaolinite to the forearc
basin, as Lin and Chen (1986) claimed, the Taiyuan forearc basin
turbidites in the eastern part of the North Luzon Trough would
have received more kaolinite than the Lichi Mélange in the west-
ern part of the North Luzon Trough, unless part of the volcanic
arc had been thrust to exhume the metamorphic Central Range in
eastern Taiwan and exposed as the source for providing kaolin-
ite to the Lichi Mélange in the western part of the North Luzon
Trough but not to the remnant Taiyuan forearc basin strata in the
east. However, no such Neogene volcanic arc rocks are found in
the eastern Central Range today. Moreover, south of lat 23° N the
Taiyuan remnant forearc basin and the Chengkuangao volcanic
island were accreted as the southern Coastal Range during the
last 1 m.y. (Huang et al., 2006). At that time the configuration
of the Central Range–Coastal Range thus should not have dif-
fered much from how it looks today. This implies that it may be
improbable to consider that the kaolinite in the Lichi Mélange
was derived from the volcanic arc in the eastern Central Range
by eastward sedimentary mass slumping.
In the Coastal Range, kaolinite is found only in the volcanic
rocks of the Tuluanshan Formation (Wang and Yang, 1975), sug-
gesting that the source of kaolinite in the Lichi Mélange could
have been related to the tectonic disturbance of the Tuluanshan
volcanic rocks, as follows:
1. Field surveys show that the volcanic rocks of the Tulu-
anshan Formation beneath the Taiyuan forearc basin are
thrust westward over the Lichi Mélange along the Tulu-
anshan Fault (Fig. 8), and this is the only region in which
the Tuluanshan volcanics are exposed between the Lichi
Mélange and the Loho-Taiyuan remnant forearc basins;
2. The volcanic sequence of the Tuluanshan Formation never
appears within the remnant forearc basins;
3. Large, isolated, rootless volcanic blocks of volcaniclas-
tics, andesitic agglomerates, and tuffs are present in the
Lichi Mélange from Loho to Lichi (Fig. 2; Hsu, 1956,
1976). These large blocks are fault bounded in the dis-
membered or mélange facies of the Lichi Mélange. Near
Antung, hot springs are present along the fault lines where
the fault-bounded tuff and andesitic agglomerates occur
in the Lichi Mélange.
We believe that such thrusting, fragmentation, and mixing of
the volcanic rocks within the mélange facies have allowed incor-
poration of the kaolinites from the Tuluanshan Formation into the
entire Lichi Mélange, but not to the overlying coherent-forearc-
basin strata, by fluid flow (Fisher, 1996) along the ubiquitously
sheared planes or fractures within the Lichi Mélange.
Therefore, characteristic occurrence of clay mineral compo-
sition in the Lichi Mélange does not conflict with the conclu-
sion that the protolith of the Lichi Mélange was deposited in the
western part of the North Luzon Trough forearc basin. Instead,
occurrence of characteristic kaolinite in the Lichi Mélange is
strongly related to the structures in the southern Coastal Range,
showing tectonic involvement of the volcanic basement beneath
the forearc basin in the formation of the Lichi Mélange during
closure of the North Luzon Trough, when the Luzon arc-forearc
was accreted to the southern Coastal Range during the last 1 m.y.
Consequently, the conclusion of this clay-mineral-composition
study is consistent with field surveys and prediction G in the first
subsection under “Discussion.”
Significance of the Whitish Quartz-Rich Feldspathic
Sandstone Blocks in the Lichi Mélange
In the Lichi Mélange, some allochthonous blocks—such as
the angular, whitish, well-lithified, quartz-rich, feldspathic sand-
stone blocks and ophiolite blocks—occur in the block-in-matrix
dismembered facies and mélange facies. These blocks have not
 The Lichi Mélange
been observed in the weakly sheared broken formation facies of
the Lichi Mélange or in the remnant-forearc-basin sequences.
The origin of these allochthonous sandstone blocks is problem-
atic. Like the kaolinite, these whitish angular sandstone blocks
have been interpreted as being derived from the exposed accre-
tionary prism by eastward sedimentary slumping toward the
western North Luzon Trough, the site of the Lichi Mélange (Page
and Suppe, 1981; Lin and Chen, 1986). However, no independent
evidence supports this slumping hypothesis.
Similar angular, sheared, quartz-rich sandstone blocks (1–
15 cm in size), embedded in the late Pliocene sheared mudstone
(ca. 3 Ma), have been dredged from the modern Huatung Ridge
but have never been collected from the remnant North Luzon
Trough off the Coastal Range (Huang et al., 1992; Huang, 1993).
The sandstone blocks and the associated mudstone in the Hua-
tung Ridge contain not only illite and chlorite but also significant
kaolinite similar to what we found in the Lichi Mélange (Huang
et al., 1992).
To understand the significance of these allochthonous sand-
stone blocks in the Lichi Mélange, we studied fission tracks of
zircon grains, separated from the whitish quartz-rich feldspathic
sandstone blocks in the Lichi Mélange, and also separated from
the coarse grains in the Taiyuan remnant-forearc-basin turbi-
dite layers.
Fission track analyses of the zircon grains, separated from
the angular, whitish sandstone blocks (>2 m) in the dismembered
facies and the mélange facies in Fuli, and the type locality of the
Lichi Mélange (Fig. 8), show a partial annealing feature with two
age peaks concentrated at the early–middle Miocene (9–18 Ma)
and the late Mesozoic (66–130 Ma; Fig. 9A). However, the zir-
con grains of the Taiyuan remnant-forearc-basin turbidites show
a distinctly different fission track pattern: one that is completely
reset in a tuff sample in the lower part of the forearc sequence
(2.08 Ma in sample TF-8; Fig. 9B) or partially reset with a young
peak of Pliocene age (2.5 Ma in sample TF-6; 3.0 Ma in sample
TF-7; 2.7 Ma in sample TF-9) and an old tail of Mesozoic age
(64–100 Ma; Fig. 9B). Although the fission track records of the
zircon grains all show a similar partial annealing feature, there
is a significant difference between them: a higher frequency of a
young peak (<7 Ma) in the Taiyuan remnant-forearc-basin turbi-
dites (Fig. 9B), but a higher frequency of an old peak (>66 Ma,
Mesozoic) in the angular whitish sandstone blocks of the Lichi
Mélange (Fig. 9A). The frequency for the Mesozoic age in the
Lichi Mélange samples is as high as, or even higher than, the
frequency for the Neogene age. On the contrary, the coherent
forearc turbidites show a distinct late Neogene age peak but a
very low frequency for a Mesozoic age.
Taiwan has undergone subduction and collision tectonics
since the Miocene. The fission track study of zircons has been
used successfully to constrain the source of sediment provenance
and geothermal history during the subduction and collision tec-
tonics of Taiwan (Liu, 1982; Liu et al., 2001; Huang et al., 2006;
Fuller et al., 2006). These studies have shown that fission tracks
of zircon grains from the Miocene deep-marine turbidites of the
149
accretionary prism (the western Central Range; Fig. 1) were par-
tially reset, and the zircons separated from the exhumed metamor-
phic basement (eastern Central Range; Fig. 1) were totally reset.
Our samples all show partial annealing characteristics, indicating
that they were all derived from the accretionary prism. But the
young peak age of the Taiyuan remnant-forearc-basin strata (2.7–
2 Ma) is much younger than that of the sheared whitish sandstone
blocks of the Lichi Mélange (9–18 Ma), suggesting that not only
the source, but also the mechanism, for emplacing these sheared
angular sandstone blocks into the Lichi Mélange may have sig-
nificantly differed from the turbidite sedimentation in the Taiyuan
remnant forearc basin. Moreover, these angular, sheared, whitish
quartz-rich sandstone blocks have not been observed in the Tai-
yuan remnant-forearc-basin strata, indicating that they may not
have been delivered to the Lichi Mélange by turbidity currents as
Taiyuan remnant-forearc-basin turbidite layers.
To identify the source of these quartz-rich, feldspathic sand-
stone blocks in the Lichi Mélange, we also analyzed four samples
collected from the deep-sea sandstones in the accretionary prism
of the Hengchun Peninsula for comparison (Fig. 10). These
samples also show a partial reset feature with two age peaks:
a young Neogene peak (15–30 Ma) and an old Mesozoic peak
(50–100 Ma; Fig. 10). The frequency of the Mesozoic peak is as
high as the Neogene peak, similar to that found in the whitish
sandstone blocks of the Lichi Mélange (Fig. 9A). These sand-
stone blocks in the Lichi Mélange were determined to be late
Miocene in age (Chi, 1982; Barrier and Müller, 1984), consistent
with the deep-sea-fan sandstones in the accretionary prism of the
Hengchun Peninsula (Chang, 1966; Huang et al., 1997) but older
than the Pliocene–Pleistocene Taiyuan remnant-forearc-basin
strata and the Pliocene protolith of the Lichi Mélange.
Three routes are possible through which the late Miocene
deep-sea-fan sandstones in the late Miocene accretionary prism
could have been transported to the Pliocene dismembered facies
and mélange facies of the Lichi Mélange: (1) by erosion from
the exposed accretionary prism (now the Central Range, before
exhumation of the underthrust Eurasian continental materials
during the last 2.5 m.y.; Huang et al., 2006) and then transported
to the North Luzon Trough by turbidity flow longitudinally from
north to south along the trough axis; (2) by erosion as (1) but
transported eastward by mass slumping into the western part
of the North Luzon Trough (Page and Suppe, 1981); or (3) by
eastward thrusting to move the non-exposed late Miocene accre-
tionary prism materials upward into the deformed forearc strata
(proto–Lichi Mélange, like the modern Huatung Ridge) in the
western part of the North Luzon Trough.
It is improbable that the first and second routes could have
caused the difference in the fission-track age patterns of zircons
between the Lichi Mélange and the Taiyuan remnant forearc
turbidites, because both processes would have provided similar
zircon grains from the same source (the exposed accretionary
prism) to the North Luzon Trough. Therefore, they should have
similar fission-track age patterns in the eastern or the western part
of the trough either by longitudinal turbidity flow or by eastward
150 Huang et al.
mass slumping. In contrast, the third route could have provided
the zircon grains from different sources by different mechanisms
independently to the Lichi Mélange and to the Taiyuan remnant-
forearc-basin turbidites. As the marine seismic profiles (Fig. 7)
demonstrate, we believe that the blocks of whitish sandstone could
have been thrust eastward from the non-exposed accretionary
prism unconformably beneath the western North Luzon Trough
forearc basin to the deformed proto–Lichi Mélange (Huatung
Ridge) in the western part of the North Luzon Trough, instead
of having been derived from the exposed accretionary prism by
turbidity flows or mass slumping downward to the entire North
Luzon Trough. By such eastward backthrusting, the late Miocene
deep-sea sandstone is found restricted to the strongly sheared,
block-in-matrix, dismembered facies and the mélange facies of
the Lichi Mélange but not to the weakly sheared relict, broken
formation facies of the Lichi Mélange or the Taiyuan remnant-
forearc-basin strata. This eastward thrusting mechanism is exactly
like that of the Huatung Ridge, where we found similar angular,
whitish sandstone blocks with kaolinite, or where the proto–Lichi
Mélange is developing (Fig. 7), and is consistent with prediction
G in the first subsection under “Discussion.”
Source of the East Taiwan Ophiolite
Dismembered ophiolite blocks are well known in the Lichi
Mélange in the Coastal Range, eastern Taiwan, and are com-
monly known as the East Taiwan Ophiolite. However, the source
and the mechanism for emplacing these oceanic-affinity blocks
within the Lichi Mélange are still the subject of debate. Along
the Mukeng River the number and size of these blocks (usually
0.1–2 m) are much less common and are smaller than those in
the Kuanshan and Lichi areas (Fig. 1). Along the Mukeng River
sections the East Taiwan Ophiolite blocks do not occur with
the whitish sandstone blocks. However, they both appear to be
restricted to the intensely sheared, dismembered facies or the
mélange facies of units I, II, III, and IV (Fig. 8). In the Lichi area
the larger East Taiwan Ophiolite blocks (several to hundreds of
meters) do occur commonly with the whitish sandstone blocks
(Liou et al., 1977; Page and Suppe, 1981). There are two pos-
sibilities for the source of the East Taiwan Ophiolite blocks: the
South China Sea oceanic crust or the oceanic basement beneath
the Luzon Arc-forearc of the Philippine Sea plate.
If the East Taiwan Ophiolite was part of the South China
Sea oceanic crust, its age would be younger than late Oligocene
because this sea opened in 32–17 Ma (Taylor and Hayes, 1980).
Microfossils recovered from a thin red shale intercalated within
this ophiolite sequence indicated an age of 15 Ma (Zone NN5 of
calcareous nannoplanktons; Huang et al., 1979), which would be
close to the last phase of the opening of the South China Sea. This
conclusion may favor an oceanic-crust source for the South China
Sea (Suppe et al., 1981). However, no East Taiwan Ophiolite–like
rock occurs in the present Central Range. These ophiolite blocks,
presumably of a South China Sea source, could not have been
derived from the exposed accretionary prism (Central Range)
by sedimentary slumping, because in the Lichi area the ophiol-
ite blocks occur with the whitish sandstone blocks, which were
improbably emplaced into the Lichi Mélange by mass slumping
(discussed previously). Therefore, like the concurrent late Mio-
cene angular sandstone blocks, the East Taiwan Ophiolite also
could have been thrust eastward from the accretionary prism of
the Hengchun Peninsula during the early stage of arc-continent
collision. However, in the accretionary prism the ophiolite blocks
are observed only in the Kenting Mélange of the Hengchun Pen-
insula (Fig. 10), and they differ from the East Taiwan Ophiolite
in composition. For example, the chromatite that is commonly
found in the Kenting Mélange of the Hengchun Peninsula (Chu
et al., 1988) has never been observed in the East Taiwan Ophio-
lite. Unlike the deep-sea-fan sandstones that occur ubiquitously
in the Hengchun Peninsula (Huang et al., 1997), the ophiolite
blocks in the Kenting Mélange are present in a highly limited
area (Fig. 10). Therefore, it would have been improbable for these
ophiolite blocks to have been thrust eastward from the accretion-
ary prism of the Hengchun Peninsula.
From a regional tectonic view, we prefer the interpretation
that the East Taiwan Ophiolite may represent the oceanic crust of
the Philippine Sea Plate (Juan et al., 1980; Malavieille et al., 2002).
These ophiolite blocks could have been westward thrust from the
oceanic crust beneath the North Luzon Trough forearc basin or
the Luzon Arc basement during the final westward accretion of
the Luzon Arc-forearc to eastern Taiwan during the last 1 m.y.
(Fig. 5C, 3). The age of the Luzon Arc is no later than middle
Miocene, consistent with the age of the East Taiwan Ophiolite.
The occurrence of ophiolitic rocks in the Banda forearc basement
has been documented from the Bobonaro Mélange of the Timor
arc-continent collision orogen by thrusting emplacement (Harris
et al., 1998). It is also possible that the East Taiwan Ophiolite
could have been developed by suprasubduction (Pearce et al.,
1984) by which the lower crustal materials, such as serpentinite
and peridotite, could be found in a forearc basin setting, similar
to the modern Mariana forearc (Fryer et al., 1985).
Global Significance of the Lichi Mélange: Comparison
with the Bobonaro Mélange in the Timor Arc-Continent
Collision Orogen
Mélange is generally considered to be developed within the
accretionary prism by scraping off the underthrusting oceanic
crust and the overlying deep-sea marine sediments along the
décollement fault in the subduction zone (Silver and Beutner,
1980; Cowan, 1985; Cloos and Shreve, 1988), and the mechanism
responsible for formation of mélange could be tectonic thrusting,
sedimentary slumping, or mud diapirism (Cowan, 1985; Barber
et al., 1986). However, the Pliocene Lichi Mélange was devel-
oped in the forearc basin rather in the accretionary prism. The
sedimentary mass-slumping process is unlikely as the major
mechanism responsible for formation of the Lichi Mélange, as
discussed previously. Today, two small mud volcanoes, each
100–200 wide, are found within the Lichi Mélange (Yang et al.,
 The Lichi Mélange
2004). However, in comparison with the wide distribution of the
Lichi Mélange (50 km long, 2 km wide), such small areas of mud
volcanoes could not have played an important role in the devel-
opment of the Lichi Mélange.
Today, mélange can be found also in the Timor arc-continent
collision region. The Bobonaro Mélange in Timor Island is char-
acterized by active development, long extension from offshore
to onshore for more than 2000 km, a young age (since 3 Ma),
and good exposures. Other characteristics include a wide variety
of rock types in block-in-matrix facies caused by polymecha-
nisms of tectonic detachment, mud diapirism, and sedimentary
slumping (Barber et al., 1986; Harris et al., 1998). In comparing
the Lichi Mélange in the Coastal Range, eastern Taiwan, with
the Bobonaro Mélange in Timor, of the Banda Arc region, it is
interesting to note similarities and differences. For example, the
Lichi Mélange and the Bobonaro Mélange are both of a young
age (ca. 3 Ma), forming mélanges today in active arc-continent
collision tectonics, traceable from onshore to offshore, involving
forearc basement tectonics, and showing variable degrees of rock
shearing, fragmentation, and mixing from weakly sheared bro-
ken formation facies to block-in-clay mélange facies. However,
the Lichi Mélange developed in a forearc basin setting rather than
in an accretionary prism setting, as did the Bobonaro Mélange.
Therefore, no continental fragment of the lower plate, like the
Australian pre- and post-rifting sequences in the Bobonaro
Mélange, was involved during formation of the Lichi Mélange.
On the other hand, the Luzon Volcanic Arc was involved during
formation of the Lichi Mélange, but the Banda Volcanic Arc was
not involved in formation of the Bobonaro Mélange. The micro-
fossil composition of the Lichi Mélange (Pliocene) is much sim-
pler than that of the Bobonaro Mélange (Jurassic to Pleistocene).
Deep-sea pelagic sediments such as radiolarian ooze are pres-
ent in the Bobonaro Mélange, but only hemipelagic muds and
turbidites in the Lichi Mélange. Mud diapirism and sedimentary
mass slumping were insignificant in the development of the Lichi
151
Mélange, but they played an important role in the development of
the Bobonaro Mélange.
CONCLUSIONS
New seismic surveys along the 21° N transect across the
North Luzon Trough show a syndeformational sedimentation
in the forearc basin. Episodic deformation has occurred in the
western part of the North Luzon Trough since the early history of
forearc sedimentation by the active collision between the Luzon
Arc and the Eurasian continent. The deformation resulted in for-
mation of the Huatung Ridge, which continues northward to the
Lichi Mélange in the southernmost Coastal Range, suggesting
a direct geological relationship between formation of the Lichi
Mélange and the submarine Huatung Ridge. Multiple lines of
new evidence, including on-land field surveys, planktic and ben-
thic foraminiferal study, clay mineral composition, and zircon
fission-track analyses, confirm that the Lichi Mélange was part
of the North Luzon Trough forearc sequences (Fig. 5C, 1), rep-
resenting a collision mélange formed along arcward (eastward)
backthrusting during forearc closure when the Luzon arc collided
incipiently with the Eurasian continent.
The formation of the Lichi Mélange involved two thrust
events: (1) early arcward backthrusting in the western part of the
North Luzon Trough to form the proto–Lichi Mélange, like the
Huatung Ridge, during progressive closure of the forearc basin
in the initial arc-continent collision stage (ca. 3 Ma; Figure 5C,
2), followed by (2) westward thrusting of the Luzon Arc-forearc
onto eastern Taiwan to form the present southern Coastal Range
during the advanced arc-continent collision stage during the last
1 m.y. (Fig. 5C, 3).
The Lichi Mélange provides a unique example in compari-
son with worldwide orogenic belts in that the mélange has devel-
oped by tectonic thrusting along closure of the forearc basin in
the region of active arc-continent collision.
152 Huang et al.

APPENDIX

APPENDIX. CLAY MINERAL COMPOSITION (IN PERCENTAGES) OF CLAY FRACTIONS ALONG THE MUKENG RIVER
AND ITS SOUTHERN TRIBUTARY SECTIONS IN THE LICHI MÉLANGE
No. Sample Lithology Illite Chlorite Kaolinite Smectite Mixed-layer
E MK01-1- Sheared mudstone 47 18 13 Trace to absent 22
E MK02- Sheared mudstone 43 20 9 Trace to absent 28
E MK03-0- Non-sheared silty mudstone 56 11 6 Trace to absent 27
E MK03-3- Slightly sheared mudstone 57 14 8 Trace to absent 21
E MK03-4- Sheared mudstone 48 19 12 Trace to absent 21
E MK04- Sheared mudstone 52 22 10 Trace to absent 16
C MK06-1- Sheared mudstone 47 14 11 Trace to absent 28
C MK10- Sheared mudstone 51 21 10 Trace to absent 18
B MK12- Non-sheared mudstone 56 18 10 Trace to absent 16
B MK13-1- Sheared mudstone 52 19 11 Trace to absent 18
D MKN01- Non-sheared mudstone 48 22 11 Trace to absent 19
A MKN02-2- Slightly sheared mudstone 47 19 12 Trace to absent 22
A MKN02-4- Slightly sheared mudstone 49 19 11 Trace to absent 21
A MKN02-6 Sheared mudstone 59 12 11 Trace to absent 18
A MKN03-1- Sheared mudstone 53 21 11 Trace to absent 15
A MKN04- Sheared mudstone 56 21 10 Trace to absent 13
A MKN05-2- Non-sheared shale 53 16 10 Trace to absent 21
A MKN05-4- Non-sheared mudstone 44 23 14 Trace to absent 19
A MKN05-6- Sheared mudstone 47 20 10 Trace to absent 23
A MKN07- Sheared mudstone 39 17 15 Trace to absent 29
A MKN08- Non-sheared mudstone 63 19 14 Trace to absent 4
F MKS01- Non-sheared shale 67 21 8 Trace to absent 4
F MKS02- Non-sheared mudstone 52 14 13 21
F MKS03- Sheared mudstone 53 15 14 18
F MKS03S- Non-sheared siltstone 46 19 14 21
F MKS04-2- Non-sheared mudstone 64 23 7 6
F MKS04-4- Sheared mudstone 63 19 12 6
F MKS05-1- Non-sheared mudstone 53 18 10 19
F MKS05-2- Non-sheared mudstone 55 18 10 17
F MKS06- Non-sheared mudstone 52 19 12 17
G MKS08- Sheared mudstone 52 16 9 23
G MKS09- Mud in conglomerate layer 52 19 9 20
G MKS10- Sheared mudstone 53 18 11 18
TF-06a Mudstone 54 24 6 16
TF-07 Mudstone 60 23 5 12
TF09 Mudstone 50 21 8 21
LM-1 mud Mudstone 51 19 14 16
YF02C Mudstone 53 18 6 23
YF03C Mudstone 54 16 15 15
YF01C Sandstone 28 50 15 7
LM-1 Sandstone 42 23 22 13
LM01 Sandstone 38 27 14 21
FK01C Sandstone 42 15 17 16

Average: 51.5 1 9.5 11.1 17.8

For comparison, average clay mineral composition of

43.7 9.8 13.7 3.7 28.9
Lin and Chen (1986):
 The Lichi Mélange
ACKNOWLEDGMENTS
This study was financially supported by grants from the National
Science Council (NSC93–2116-M-006–001; NSC94–2116-
M006–005) and the Research Center of Ocean Environment and
Technology, NCKU, to C.Y. Huang, and grant G2000046705
from the National Important Basic Research and Development
Projects to B. Yao. We appreciate the crew members of the R/V
Tanbao for their great efforts during the seismic surveys across
the North Luzon Trough off southern Taiwan. The authors also
appreciate the constructive discussions with R. Harris and D.
Reed in comparing the Lichi Mélange in eastern Taiwan with
the Bobonaro Mélange in Timor. The manuscript was improved
by comments from A. Basu and an anonymous reviewer.
REFERENCES CITED
Barber, A.J., Tjokrosapoetro, S., and Charlton, T.R., 1986, Mud volcanoes,
shale diapirs, wrench faults and mélanges in accretionary complex, east-
ern Indonesia: American Association of Petroleum Geologists Bulletin,
v. 70, p. 1729–1741.
Barrier, E., and Müller, C., 1984, New observations and discussions on the ori-
gin and age of the Lichi Mélange: Geological Society of China Memoir,
v. 6, p. 303–325.
Biq, C., 1971, Comparison of mélange tectonics in Taiwan and in some other
mountain belts: Petroleum Geology of Taiwan, no. 9, p. 79–106.
Biq, C., 1973, Kinematic pattern of Taiwan as an example of actual continent-
arc collision: Taipei, Report of the Seminar on Seismology: US-ROC
Cooperative Science Program, abstract, p. 21–26.
Buchovecky, E.J., and Lundberg, N., 1988, Clay mineralogy of mudstones from
the southern Coastal Range, eastern Taiwan: Unroofing of the orogen ver-
sus in-situ diagenesis: Acta Geologica Taiwanica, no. 26, p. 247–261.
Chai, B.H.T., 1972, Structure and tectonic evolution of Taiwan: American Jour-
nal of Science, v. 272, p. 389–422.
Chang, C.P., Angelier, J., and Huang, C.Y., 2000, Origin and evolution of a
mélange: The active plate boundary and suture zone of the Longitudinal
Valley, Taiwan: Tectonophysics, v. 325, p. 43–62, doi: 10.1016/S0040-
1951(00)00130-X.
Chang, C.P., Angelier, J., Huang, C.Y., and Liu, C.S., 2001, Structural evolu-
tion and significance of a mélange in a collision belt: The Lichi Mélange
and the Taiwan arc-continent collision: Geological Magazine, v. 138,
p. 633–651.
Chang, L.S., 1966, A biostratigraphic study of the Tertiary in the Hengchun
peninsula, Taiwan, based on smaller foraminifera (III: Southern Part):
Proceedings of the Geological Society of China, no. 9, p. 53–63.
Chang, L.S., 1967, A biostratigraphic study of the Tertiary in the Coastal
Range, eastern Taiwan, based on smaller foraminifera (I: Southern Part):
Proceedings of the Geological Society of China, no. 10, p. 64–76.
Chen, C.H., Shieh, Y.N., Lee, T., Chen, C.H., and Mertzman, S.A., 1990, Nd-
Sr-O isotopic evidence for source contamination and an unusual mantle
component under Luzon arc: Geochimica et Cosmochimica Acta, v. 54,
p. 2473–2483, doi: 10.1016/0016-7037(90)90234-C.
Chen, W.S., 1997, Mesoscopic structures developed in the Lichi Mélange dur-
ing the arc-continent collision in the Taiwan region: Journal of the Geo-
logical Society of China, v. 40, p. 415–434.
Cheng, Y.M., Huang, C.Y., Yeh, J.J., and Chen, W.S., 1984: The Loshui For-
mation: Deeper water sandstones in the Hengchun Peninsula, southern
Taiwan: Acta Geologica Taiwanica, no. 22, p. 35–52.
Chi, W.R., 1982, The calcareous nannofossils of the Lichi mélange and the
Kenting mélange and their significance in the interpretation of plate tec-
tonics of the Taiwan region: Ti-Chih (Geology), v. 4, p. 99–112.
Chi, W.R., Namson, J., and Suppe, J., 1981, Stratigraphic record of plate inter-
actions in the Coastal Range of eastern Taiwan: Geological Society of
China Memoir, v. 4, p. 155–194.
Chien, C.W., 2003, Forearc origin of the Lichi Mélange in Coastal Range,
eastern Taiwan: Micropaleontological evidence [M.S. thesis]: Institute of
Geosciences, National Taiwan University, 74 p.
153
Chiu, Y.L., 2005, Sedimentological study of Site 1202, Leg 195, in southern
Okinawa Trough off NE Taiwan [M.S. thesis]: Department of Earth Sci-
ences, National Cheng Kung University, 80 p.
Chu, H.T., Shen, P., and Jeng, R.C., 1988, The origin of chromatite from the
Kenting Mélange, southern Taiwan: Proceedings of the Geological Soci-
ety of China, v. 31, p. 33–52.
Chung, S.L., and Sun, S.S., 1992, A new genetic model for the East Taiwan
Ophiolite and its implications for Dupal domains in the Northern Hemi-
sphere: Earth and Planetary Science Letters, v. 109, p. 133–145, doi:
10.1016/0012-821X(92)90079-B.
Cloos, M., and Shreve, R.L., 1988, Subduction-channel model of prism accre-
tion, mélange formation, sediment subduction, and subduction erosion
at convergent plate margins: 1. Background and description: Pageoph,
v. 128, p. 455–500, doi: 10.1007/BF00874548.
Cowan, D., 1985, Structural styles in Mesozoic and Cenozoic mélanges in
the western Cordillera of North America: Geological Society of Amer-
ica Bulletin, v. 96, p. 451–462, doi: 10.1130/0016-7606(1985)96<451:
SSIMAC>2.0.CO;2.
Ernst, W.G., 1977, Olistostromes and included ophiolite debris from the
Coastal Range of eastern Taiwan: Geological Society of China Memoir,
v. 2, p. 97–114.
Fisher, D.M., 1996, Fabrics and veins in the forearc: A record of cyclic fluid
flow at depth of 15 km, in Bebout, G.E., et al., eds., Subduction: Top to
Bottom: American Geophysical Union Monograph 96, p. 75–89.
Fryer, P., Ambos, E.L., and Hussong, D.M., 1985, Origin and emplace-
ment of Mariana forearc seamounts: Geology, v. 13, p. 774–777, doi:
10.1130/0091-7613(1985)13<774:OAEOMF>2.0.CO;2.
Fuller, C.W., Willett, S.D., Fisher, D., and Lu, C.Y., 2006, A thermomechanical
wedge model of Taiwan constrained by fission-track thermochronometry:
Tectonophysics, v. 425, p. 1–24, doi: 10.1016/j.tecto.2006.05.018.
Harris, R.A., Sawyer, R.K., and Audley-Charles, M.G., 1998, Collision mélange
development: Geologic associations of active mélange-forming processes
with exhumed mélange facies in the western Banda orogen, Indonesia:
Tectonics, v. 17, p. 458–479, doi: 10.1029/97TC03083.
Horng, C.S., and Shea, K.S., 1996, Dating of the Plio-Pleistocene rapidly
deposited sequence based on integrated magneto-biostratigraphy: A case
study of the Madagida-Chi section, Coastal Range, eastern Taiwan: Jour-
nal of the Geological Society of China, v. 39, p. 31–58.
Hsu, T.L., 1956, Geology of the Coastal Range, eastern Taiwan: Geological
Survey of Taiwan Bulletin, v. 8, p. 39–64.
Hsu, T.L., 1976, The Lichi Mélange in the Coastal Range framework: Taiwan
Geological Survey Bulletin, v. 25, p. 87–95.
Huang, C.Y., 1989, Implication of the Post-Lushanian faunal change for the
occurrence of Kuroshio Current in the early late Miocene: Foraminiferal
evidence from the Chuhuangkeng section, northern Taiwan: Proceedings
of the Geological Society of China, v. 32, p. 21–45.
Huang, C.Y., 1993, Bathymetric ridge and trough in the active arc-continent
collision region off southeastern Taiwan: Reply and discussion: Journal
of the Geological Society of China, v. 36, p. 91–109.
Huang, C.Y., and Yin, Y.C., 1990, Bathymetric ridges and troughs in the active
arc-continent collision region off southeastern Taiwan: Proceedings of the
Geological Society of China, v. 33, p. 351–372.
Huang, C.Y., Shyu, C.T., Lin, S.B., Lee, T.Q., and Sheu, D.D., 1992, Marine
geology in the arc-continent collision zone off southeastern Taiwan:
Implications for late Neogene evolution of the Coastal Range: Marine
Geology, v. 107, p. 183–212, doi: 10.1016/0025-3227(92)90167-G.
Huang, C.Y., Yuan, P.B., Song, S.R., Lin, C.W., Wang, C., Chen, M.T., Shyu,
C.T., and Karp, B., 1995, Tectonics of short-lived intra-arc basins in the
arc-continent collision terrane of the Coastal Range, eastern Taiwan: Tec-
tonics, v. 14, p. 19–38, doi: 10.1029/94TC02452.
Huang, C.Y., Wu, W.Y., Chang, C.P., Tsao, S., Yuan, P.B., Lin, C.W., and Xia,
K.Y., 1997, Tectonic evolution of accretionary prism in the arc-conti-
nent collision terrane of Taiwan: Tectonophysics, v. 281, p. 31–51, doi:
10.1016/S0040-1951(97)00157-1.
Huang, C.Y., Yuan, P.B., Lin, C.W., and Wang, T.K., 2000, Geodynamic pro-
cesses of Taiwan arc-continent collision and comparison with analogs in
Timor, Papua New Guinea, Urals and Corsica: Tectonophysics, v. 325,
p. 1–21, doi: 10.1016/S0040-1951(00)00128-1.
Huang, C.Y., Chiu, Y.L., and Zhao, M., 2005, Core description and preliminary
sedimentology features of silt and sand depositions of ODP Site 1202D,
Leg, 195, southern Okinawa Trough: Terrestrial, Atmospheric and Oce-
anic Sciences, v. 16, p. 19–44.
154 Huang et al.
Huang, C.Y., Yuan, P.B., and Tsao, S.J., 2006, Temporal and spatial records of
active arc-continent collision in Taiwan: A synthesis: Geological Society
of America Bulletin, v. 118, p. 274–288, doi: 10.1130/B25527.1.
Huang, T.C., Chen, M.P., and Chi, W.R., 1979, Calcareous nannofossils from
the red shale of the ophiolite-mélange complex, eastern Taiwan: Geologi-
cal Society of China Memoir, v. 3, p. 131–138.
Juan, V.C., Lo, H.J., and Chen, C.C., 1980, Genetic relationships and emplace-
ment of the exotic basic rocks enclosed in the Lichi mélange, east Coastal
Range, Taiwan: Proceedings of the Geological Society of China, no. 23,
p. 56–68.
Lin, S.B., and Chen, G.T., 1986, Clay minerals from the Lichi Mélange and its
adjacent formations in the Coastal Range, eastern Taiwan: Acta Geologica
Taiwanica, no. 24, p. 319–356.
Liou, J.G., Suppe, J., and Ernst, W.G., 1977, Conglomerates and pebbly mud-
stones in the Lichi Mélange, eastern Taiwan: Geological Society of China
Memoir, v. 2, p. 115–128.
Liu, C.S., Liu, S.Y., Lallemand, S., Lundberg, N., and Reed, D.L., 1998, Digital
elevation model offshore Taiwan and its tectonic implications: Terrestrial,
Atmospheric and Oceanic Sciences, v. 9, p. 705–738.
Liu, T.K., Hsieh, S., Chen, Y.G., and Chen, W.S., 2001, Thermo-kinematic evo-
lution of the Taiwan oblique-collision mountain belt as revealed by zircon
fission track dating: Earth and Planetary Science Letters, v. 186, p. 45–56,
doi: 10.1016/S0012-821X(01)00232-1.
Liu, T.W., 1982, Tectonic implication of fission track ages from the Central
Range, Taiwan: Proceedings of the Geological Society of China, no. 25,
p. 22–37.
Loeblich, A., Jr., and Tappan, H., 1964, Treatise on Invertebrate Paleontology,
Part C, Protista 2, Sarcodina, Chiefly “Thecamoebians” and Foraminifer-
ida: Geological Society of America and University of Kansas Press, 900
p.
Loeblich, A., Jr., and Tappan, H., 1988, Foraminiferal Genera and Their Clas-
sification: New York, Van Nostrand Reinhold, 970 p.
Lundberg, N., Reed, D.L., Liu, C.S., and Lieske, J., Jr., 1997, Forearc-basin
closure and arc accretion in the submarine suture zone south of Taiwan:
Tectonophysics, v. 274, p. 5–23, doi: 10.1016/S0040-1951(96)00295-8.
Malavieille, J., Lallemand, S.E., Dominguez, S., Deschamps, A., Lu, C.Y., Liu,
C.S., and Schnurle, P., and the ACT Scientific Crew, 2002, Arc-continent
collision in Taiwan: New marine observations and tectonic evolution:
Geological Society of America Special Paper 358, p. 187–211.
Mutti, E., and Lucchi, F.R., 1978, Turbidites of the Northern Apennines: Intro-
duction to facies analysis: International Geology Review, v. 44, p. 125–
166.
Page, B.M., and Suppe, J., 1981, The Pliocene Lichi Mélange of Taiwan: Its
plate tectonic and olistostromal origin: American Journal of Science,
v. 281, p. 193–227.
Pearce, J.A., Lippard, S.J., and Robert, S., 1984, Characteristics and tectonic
significance of suprasubduction zone ophiolites, in Kokelaar, B.P., and
Howell, M.F., eds., Marginal Basin Geology: Geological Society [Lon-
don] Special Publication 16, p. 77–94.
Raymond, L.A., 1984, Classification of mélanges, in Raymond, L.A., ed.,
Melanges: Their Nature, Origin, and Significance: Geological Society of
America Special Paper 198, p. 7–20.
Reed, D.L., Lundberg, N., Liu, C.S., Luo, B.Y., 1992, Structural relations along
the margins of the offshore Taiwan accretionary wedge: Implications for
accretion and crustal kinematics: Acta Geologica Taiwanica, no. 30, p.
105–122.
Silver, E.A., and Beutner, E.C., 1980, Mélanges: Geology, v. 8, p. 32–34, doi:
10.1130/0091-7613(1980)8<32:M>2.0.CO;2.
Suppe, J., Liou, J.G., and Ernst, W.G., 1981, Paleogeographic origins of the
Miocene East Taiwan Ophiolite: American Journal of Science, v. 281,
p. 228–246.
Taylor, B., and Hayes, D.E., 1980, The tectonic evolution of the South China
Sea Basin, in Hayes, D.E., ed., The Tectonic and Geologic Evolution of
the Southeast Asian Seas and Islands (Part I): American Geophysical
Union Monograph 23, p. 89–104.
Teng, L.S., 1982, Stratigraphy and sedimentation of the Shuilien Conglomer-
ate, northern Coastal Range, eastern Taiwan: Acta Geologica Taiwanica,
no. 21, p. 201–220.
Van Morkhoven, F.P.C.M., Berggren, W.A., and Edwards, A.S., 1986, Cenozoic
cosmopolitan deep-water benthic foraminifera: Bulletin des Centres de
Recherches Exploration-Production, Elf-Aquitaine Memoir, v. 11, 406 p.
Wang, C.S., 1976, The Lichi Formation of the Coastal Range and arc-continent
collision in eastern Taiwan: Geological Survey of Taiwan Bulletin, v. 25,
p. 73–86.
Wang, P.X., Zhang, J., and Min, Q., 1985, Distribution of foraminifera in surface
sediments of the East China Sea, in Wang, P., ed., Marine Micropaleontol-
ogy of China, China Ocean Press: Berlin, Springer-Verlag, p. 34–69.
Wang, Y., and Yang, C.N., 1975, Expandable clay deposits and X-ray diffraction
study of montmorillonite from Coastal Range, Taiwan: Acta Geologica
Taiwanica, no. 18, p. 14–25.
Yang, T.F., Yeh, G.H., Fu, C.C., Wang, C.C., Lan, T.F., Lee, H.F., Chen, C.H.,
Walia, V., and Sung, Q.C., 2004, Composition and exhalation flux of
gases from mud volcanoes in Taiwan: Environmental Geology, v. 46,
p. 1003–1011, doi: 10.1007/s00254-004-1086-0.
Yao, T.M., Tien, P.L., and Wang-Lee, C., 1988, Clay mineralogical studies on
the Neogene formations, Taiyuan basin, southern Coastal Range, eastern
Taiwan: Acta Geologica Taiwanica, no. 26, p. 263–277.
Yu, S.B., Chen, H.Y., and Kuo, L.C., 1997, Velocity field of GPS stations in
the Taiwan area: Tectonophysics, v. 274, p. 41–59, doi: 10.1016/S0040-
1951(96)00297-1.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Oblique subduction in an island arc collision setting: Unique sedimentation, accretion,

and deformation processes in the Boso TTT-type triple junction area, NW Pacific

Yujiro Ogawa*
Doctoral Program in Earth Evolution Sciences, University of Tsukuba, Tsukuba 305-8572, Japan

Yoshihiro Takami
College of Natural Science, University of Tsukuba, Tsukuba 305-8571, Japan

Sakiko Takazawa*
College of Natural Science, University of Tsukuba, Tsukuba 305-8571, Japan

ABSTRACT

The NW corner of the Pacific Ocean is a place of unique Tertiary tectonism,

which provides one of the clearest examples of arc-arc collision. Voluminous Creta-
ceous rhyolitic-granitic magmatism along the continental margin continues into the
Paleogene. In contrast, Miocene island arc volcanism follows Eocene boninitic mag-
matism in the Izu-Mariana Arc, in association with the opening of backarc basins,
including those in the Philippine and Japan Seas. The triple junction between the
Eurasian, Philippine Sea, and Pacific plates arrived in the area south of Tokyo dur-
ing the Miocene, just as the Japan Sea was opening. After the beginning of Philip-
pine Sea plate subduction to the north, the Izu Island Arc began to collide obliquely
with the Honshu Arc. As a result, this unique tectonic setting in the NW Pacific has
produced a miniature Alpine-type orogenic belt (Tanzawa) in the collisional center,
whereas in the eastern part of the Izu Arc sediment has been actively accreting in that
forearc. Such settings have resulted in systematic accretionary prism formation from
the early Miocene in the Boso-Miura peninsular area to the present in the Sagami
Trough area. We modeled the tectonics by a simple sandbox experiment. Systematic
fault and fracture patterns of the oblique subduction type are predicted to occur dur-
ing arc-arc collision.

Keywords: island arc collision, triple junction, forearc sedimentation, forearc sliver,
accretionary prism, analogue modeling.

*Corresponding author, Ogawa: fyogawa45@yahoo.co.jp. Present address, Takazawa: Ministry of Economy, Trade and Industry of Japan, Kasumigaseki, Tokyo
100-8901, Japan.

Ogawa, Y., Takami, Y., and Takazawa, S., 2008, Oblique subduction in an island arc collision setting: Unique sedimentation, accretion, and deformation processes
in the Boso TTT-type triple junction area, NW Pacific, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record
in Arc Collision Zones: Geological Society of America Special Paper 436, p. 155–170, doi: 10.1130/2008.2436(07). For permission to copy, contact editing@
geosociety.org. ©2008 The Geological Society of America. All rights reserved.

155
156 Ogawa et al.

INTRODUCTION
Plate tectonic triple junctions provide special geologic and
tectonic conditions in which unique topography, sedimentation,
and deformation can form. Among the various types of triple
junctions (McKenzie and Morgan, 1969), a trench-trench-trench
(TTT) type of triple junction has been observed only off the Boso
Peninsula, central Japan, in the northwestern corner of the Pacific
Ocean. This is called the Boso triple junction (Ogawa et al.,
1989; Seno et al., 1989) (Fig. 1). At present, another similar type
of convergent triple junction is known nearby just north of Mount
Fuji, which is called the Fuji triple junction. Thus, in a narrow
area beneath the Tokyo region, two converging triple junctions
make a very complicated tectonic setting (Sato et al., 2005).
The geometry and tectonics of the Boso triple junction were
discussed by Seno et al. (1989) and Huchon and Labaume (1989).
The triple junction was originally called the off-Boso triple junc-
tion by Ogawa et al. (1989). The “central Japan triple junction”
of Lallemant et al. (1996) is not an adequate term, because there
are two triple junctions in central Japan. Seismic profiling and
related submarine researches were performed during the Japan-
France KAIKO project (Renard et al., 1987; Nakamura et al.,
1987; Pautot et al., 1987). After this project the principal results
were published by Ogawa et al. (1989), Seno et al. (1989), Soh et
al. (1988, 1990), and Lallemant et al. (1996), and many submarine
and on-land research projects were conducted. Most important is
the recognition that the northwestern part of the triple junction in
the Sagami Trough is a place of active sedimentation, deforma-
tion, and accretion along the oblique subduction boundary caused
by the collision of the Izu Arc with the Honshu forearc. However,
the geology and tectonics have not yet been fully explained with
reference to the oblique subduction setting.
In this paper we first review the submarine topography and
geology along the present Sagami Trough area on the basis of the
marine work of Kato et al. (1983), Kong et al. (1984), Ohkouchi
(1990), and Iwabuchi et al. (1990), and many other works related
to the KAIKO project as well as the terrestrial geology of Ogawa
and Horiuchi (1978), Ogawa (1982), and Hanamura and Ogawa
(1993), and others. We then present new evidence, including sim-
ple two-dimensional (2D) and three-dimensional (3D) analogue
shear-box experiments for oblique subduction (Takami, 1999;
Takazawa, 2003) in order to explain the uniqueness of the TTT
type of the triple junction area in terms of the present sedimenta-
tion and deformation in an oblique subduction arc-arc collisional
plate boundary.
TECTONIC SETTING
The northwestern Pacific Rim has a long tectonic his-
tory, dating from the Mesozoic and characterized by Jurassic-
Cretaceous Pacific plate subduction under NE Asia. The Japa-
nese main islands are the products of such long-term subduction
(Seno and Maruyama, 1984; Maruyama and Seno, 1986; Taira
et al., 1989; Ogawa et al., 1997; Isozaki, 1996). Since the Oligo-
cene and early to middle Miocene, when the Shikoku Basin, the
easternmost backarc basin of the Philippine Sea plate, opened,
the present Izu-Mariana Island Arc front has been moving from
west to east just south off the shore of Honshu (Kobayashi, 1984,

Figure 1. Index map of the Izu Arc collision zone with the Honshu Arc since the middle Miocene (left) to the present
(right). Izu forearc accretion occurs on the NW of the Boso triple junction. Plates A, B, and C are the Pacific, Philippine
Sea, and Eurasian plates, respectively. Plate C, just on the NW side of the triple junction, was converted to the North
American plate during the Quaternary, when the boundary between the North American and Eurasian plates extended to
central Honshu. After Hanamura and Ogawa (1993).
 Oblique subduction in an island arc collision setting 157

1995). After seafloor spreading in the Shikoku Basin ceased, the
Izu Island Arc and its NE edge, coincident with the triple junc-
tion, arrived close to its present position. Subsequently the Shi-
koku Basin lithosphere began to subduct to the NE under Honshu
(Seno and Maruyama, 1984). Consequently, the Izu Island Arc
is now connected to central Honshu, but it also extends as far
south as the Mariana Arc. The ridge of the Izu-Mariana Arc itself
has collided with the Honshu Arc in central Honshu in several
stages (Matsuda, 1978; Ogawa, 1985; Ogawa et al., 1985; Soh
et al., 1988; Aoike, 1999; Sato et al., 1999). The timing of the
first collision is synchronous with the opening of the Japan Sea
at 15 Ma (Niitsuma, 1988; Takahashi and Saito, 1997). After that
time the northeastern corner of the Philippine Sea plate has sup-
plied Izu-Mariana forearc materials to the Honshu forearc that
have been accreted to form the Miura-Boso accretionary prisms
(Figs. 2 and 3) (Ogawa, 1982, 1985; Ogawa and Taniguchi, 1988;
Hanamura and Ogawa, 1993; Takahashi and Saito, 1997; Yama-
moto et al., 2000, 2005).
The age of the Izu Arc extends at least to the Eocene
(ca. 45 Ma; Seno and Maruyama, 1984). According to the age
determination and spreading history of the Philippine Sea plate
(Kinoshita, 1980; Okino et al., 1994; Kobayashi et al., 1995), the
arc was once far to the south, in the Southern Hemisphere, and it
rotated clockwise during its northward movement to its present
position (Seno and Maruyama, 1984; Hall et al., 1995; Takahashi
and Saito, 1997). Because of spreading in the backarc basins fol-
lowing rifting of the volcanic-island-arc ridges, the oldest parts
of the Philippine Sea plate are now in both the westernmost and
easternmost parts of the present Philippine Sea plate (Kobayashi,
1984, 1995; Kobayashi et al., 1995).
The easternmost part of the Philippine Sea plate, the remnant
of the leading edge of the Izu forearc at the trenches, is composed
of ophiolitic rocks, ranging from Cretaceous to Miocene, but is
dominated by Eocene boninitic rocks and younger island arc vol-
canic materials (Fujioka and Sakamoto, 1999; Ishiwatari et al.,
2006). A similar stratigraphy is preserved as the remnant of the
“Mineoka plate” and exposed onshore as the Mineoka Ophiolite
in the Boso Peninsula (Fig. 4) (Ogawa and Taniguchi, 1988; Sato
et al., 1999; Sato and Ogawa, 2000; Ogawa and Takahashi, 2004).
Similarly, parts of the Philippine Sea plate are found in northern
New Guinea as the North New Guinea plate (Seno, 1984; Lus et
al., 2004), comprising Cretaceous to Eocene basaltic rocks and
their pelagic cover (Ogawa and Taniguchi, 1987, 1988; Mohiud-
din and Ogawa, 1998; Hirano et al., 2003; Ogawa and Sashida,
2005). Such oceanic plate history is now preserved in the Boso
Peninsula and has been documented by the detailed structural
and stratigraphic analysis of the Mineoka Ophiolite Belt (Hirano
et al., 2003; Takahashi et al., 2003; Ogawa and Takahashi, 2004;
Mori and Ogawa, 2005).
The Mineoka Ophiolite Belt has lain in a forearc sliver fault
zone since the middle Miocene, after which time there has been a
systematic development of accretionary prisms south of the Boso-
Miura Peninsulas (Ogawa, 1983; Ogawa and Taniguchi, 1988;
Lallemant et al., 1996). These prisms are formed of Izu forearc
and paleo–Sagami Trough sediments derived from an area east of
the Izu collision zone (Fig. 2) (Ogawa, 1982, 1985; Ogawa and
Taniguchi, 1987, 1988; Hanamura and Ogawa, 1993; Yamamoto
et al., 2000, 2005; Yamamoto and Kawakami, 2005). Thus the
Sagami Trough is now the modern oblique subduction boundary
at the northeasternmost corner of the Philippine Sea plate and has
been a place of sedimentation and deformation between the Izu
volcanic and Honshu continental arcs (Figs. 2 and 4). The Sag-
ami Trough is also a place of large hazardous earthquakes in the
historic and present ages, as typified by the Kanto earthquakes
(Shishikura, 2003; Sato et al., 2005). The relative present motion
of the Philippine Sea plate to the North American plate is toward
N35°W at a rate of 2.5 cm/yr (Fig. 4; Seno, 1993).
The Miura-Boso Peninsulas contain good outcrops of Mio-
cene to Pliocene Izu forearc, deep-sea, volcaniclastic deposits
that are now incorporated within accretionary prisms and covered

Figure 2. Schematic view of the model for the accretion of Izu forearc sediments to the Honshu forearc along the proto-
Sagami Trough. View from the south. A voluminous amount of Izu volcaniclastic sediments plus some trench axial chan-
nel sediments from the Honshu Arc were deposited both in the Izu forearc and the proto–Sagami Trough, finally to be
accreted to the Honshu forearc. Adapted from Hanamura and Ogawa (1993).
158 Ogawa et al.

Figure 3. Simplified trace map of the Tertiary formations on the Miura (M, west) and Boso (B, east) Peninsulas. Shaded areas represent lower
Miocene and older sedimentary rocks. Dashed doubled lines approximately trace the present plate subduction–related faults. Neogene sediments
to the north of the Mineoka Ophiolite Belt are composed of forearc basin sediments, whereas those to the south are mostly of accretionary prism
material. After Ogawa and Taniguchi (1988).

unconformably with shallow-water, coarser grained, volcanicla-
stic sedimentary rocks of the forearc basin or the trench slope
type (Hanamura and Ogawa, 1993; Yamamoto and Kawakami,
2005) (Figs. 2 and 3). The former Izu forearc sediments were
generally deep-water deposits, formed in several kilometers of
water depth (Kitazato, 1997), mostly hemipelagic (Soh et al.,
1989, 1991; Stow et al., 1998, 2002), partly contourite (Lee and
Ogawa, 1998), and include many intercalations of turbiditic or
fall-type volcaniclastic deposits (tuff and lapilli, conglomerate,
and sandstone) (Soh et al., 1989). All these sedimentary rocks
are composed of Izu Arc volcanic rocks or their derivatives. Such
sequences are divided into three prisms from north to south and
from older to younger, passing from the Emi Group (Hirono and
Ogawa, 1998), to the Miura Group (Ogawa et al., 1989; Hanamura
and Ogawa, 1993; Yamamoto and Kawakami, 2005; Yamamoto
et al., 2005; Yamamoto, 2006), and finally to the Chikura Group
(part of which may be forearc basin deposits) (Saito, 1992).
Such accretionary prisms are highly deformed by domi-
nant thrusts and folds, although most of the original sedimen-
tary structures are preserved (Figs. 2 and 3). The reason for the
good preservation is mostly due to the fact that the deforma-
tion occurred under semilithified conditions, which involved the
most significant flow or shear along bedding planes and the con-
centration of strain in faulted or folded regions (Ogawa, 1982;
Yamamoto et al., 2005; Yamamoto, 2006). As a result, the sedi-
ments between the faults and the deformed sections preserve
 Oblique subduction in an island arc collision setting 159

35 N

140 E

Figure 4. Tectonic map of the Sagami Trough. MP—Miura Peninsula; BP—Boso Peninsula; SB—Sagami Basin; MSTB—Middle Sagami
Trough Basin; OBC—Okinoyama Banka Chain; SOT—So-oh Trough; BC—Boso Canyon; NB—North Basin; MF—Mogi Fan at the triple junc-
tion. Triangle shows the relative plate motion at present between NAM (North American plate), PHS (Philippine Sea plate), and PAC (Pacific
plate). Relative motion of the Eurasian plate took place in the early Quaternary. After Seno and Maruyama (1984). Line A is the seismic profile
shown in Figure 5, and the other lines are those of the KAIKO project, some of which are shown in Figure 6.

their original structures. Therefore, along the coast, exposures
on the uplifted benches permit clear observation of sedimen-
tary structures, including grading, lamination, bioturbation, and
other features.
Ogawa and his students have published descriptions of
the terrestrial geology along the Miura-Boso Peninsula coasts
(Hanamura and Ogawa, 1993; Yamamoto et al., 2000, 2005;
Yamamoto and Kawakami, 2005; Yamamoto, 2006). The land
geology is summarized as follows:
1. Thick forearc basin sedimentary rocks have accumulated
since the early Miocene over the Mineoka Belt (in the
north) and in the middle Boso Peninsula, whereas far-
ther south several accretionary prisms have developed
between the Izu forearc and Honshu forearc. The lat-
ter sediments were originally deposited in the deep sea,
partly in the Izu forearc and partly along the boundary
between the Izu and Honshu Arcs, in settings similar to
the present Sagami Trough (Figs. 2–4).
2. Sedimentary rocks in the accretionary prisms are domi-
nated by volcaniclastic sedimentary rocks (tuff, lapilli,
and their derivatives) together with hemipelagic, muddy
sedimentary rocks or contourite background deposits.
3. Sedimentary rocks in the accretionary prisms are deformed
by many thrust faults but still preserve original sedimen-
tary textures and structures because of their rapid uplift
after accretion, although there has locally been much liq-
uefaction, injection, slumping, and faulting.
4. Rotations of the accretionary prisms around a vertical axis
are strong, and mostly on the NW side of the Miura-Boso
Peninsulas and reaching on average 30°–50° clockwise
rotation. Locally, however, rotation has been as much as
80°, owing to the collision between the Izu and Honshu
Arcs (Kanamatsu et al., 1996). Collision has occurred
(6–3 Ma) in the southern Boso Peninsula between the
accretionary prism and the forearc or the overlying trench
slope sedimentation (Yamamoto and Kawakami, 2005).
160 Ogawa et al.
TOPOGRAPHIC AND GEOLOGIC SUMMARY OF THE
SAGAMI TROUGH
The Sagami Trough is the present day plate boundary
between the North American plate to the NE and the Philip-
pine Sea plate to the SW. The latter is now subducting obliquely
under the former at a rate of 2.5 cm/yr. The SE end of the Sagami
Trough is coincident with the Boso triple junction on the Izu-
Ogasawara (Bonin) trench floor (Fig. 4). The topography of the
Sagami Trough is divided approximately into three segments,
from the NW in the Sagami Bay area (Sagami Basin), to the
middle Sagami Trough area (Middle Sagami Trough Basin), and
finally the Boso triple junction area in the SE (Fig. 4). Each area
has characteristic topography and internal structure (Nakamura
et al., 1984, 1987; Pautot et al., 1987). Seismic profiling has been
acquired, but direct observations by submersible are limited, and
no drilling has yet been carried out.
Sagami Bay Area
The Sagami Bay area is occupied by a wide basin, the Sag-
ami Basin, which is filled with a thick wedge of trench-fill sedi-
ments in the Sagami Trough (Kato et al., 1983; Kong et al., 1984;
Ohkouchi, 1990) (Fig. 5). Pleistocene to Holocene sediments
in the basin are not highly deformed but are thrust-bounded to
the NE. A fold and thrust belt has developed on the NE side,
where Miocene sedimentary rocks are deformed. Such Neogene
formations, ranging in age from early Miocene to Pleistocene,
are overlain by multiple unconformities. Terrestrial geology and
stratigraphy show that NW-trending folds and thrusts are more
strongly deformed in comparison with the older formations
(Kimura, 1976; Ogawa, 1982; Ohkouchi, 1990).
Systematic younging of these formations toward the SW is
recognized in the Miura and Boso Peninsulas, where accretionary
prism development toward the SW is well documented (Ogawa
and Taniguchi, 1988; Ogawa et al., 1989). The accreted sedimen-
tary rocks either were originally deposited on the flanks of the Izu
Arc or in the trench (Fig. 2) (Hanamura and Ogawa, 1993). The
only structural difference in the Sagami Bay area is the forma-
tion of a dissecting fault pattern subperpendicular to the dominant
NW lineament, which we suggest to be a systematic Riedel shear
system, as described in the section “Riedel Shear Experiments.”
The northeastern part of the Sagami Bay features the
Okinoyama Bank Belt, trending from NW to SE as a trench
slope break. A nearly parallel, slightly oblique en echelon pat-
tern of NW-SE faults and subperpendicular NE-SW faults cut the
entire trench slope area in the NE part of the basin (Fig. 6). The
Okinoyama Bank Belt has steep cliffs along its SW boundary, and
many Calyptogena (representative chemosynthetic clam) colonies
have been known at the foot of these cliffs (Kanie, 1999). These
en echelon faults are thought to be active and coincident with the
Sagami Tectonic Line (Kimura, 1976) (Fig. 5), corresponding to
the deformation front or splay fault of the Philippine Sea plate
subduction (Kato et al., 1983; Sato et al., 2005).
Figure 5. Seismic profile of the Sagami
Basin and its interpretation along line
A in Figure 4, showing a large amount
of trench-wedge movement in the Sa-
gami Trough (Sagami Basin). The Sa-
gami Tectonic Line (S.T.L.) SW of the
Okinoyama Bank Chain (O.B.C.) is an
offscraping or splay thrust. Adapted from
Kato et al. (1983). A—Holocene sedi-
ments; B—Pleistocene; C—Pliocene;
D—middle to upper Miocene; E—lower
Miocene.
 Oblique subduction in an island arc collision setting 161

Figure 6. Topographic features of the Sagami Basin and a new tectonic interpretation as the Riedel shears of each fault system (after Takami,
1999). Line A is the seismic profile shown in Figure 5. OIS—Oiso Spur; SMK—Sagami Knoll; MZK—Manazuru Knoll; MUK—Miu-
ra Knoll; MSK—Misaki Knoll; OYB—Okinoyama Bank; TKC—Tokyo Canyon; HIIs—Hatsushima Island; KMF—Kozu-Matsuda Fault;
AGC—Ashigara Canyon.

Nakamura et al. (1984) attempted to explain the en echelon
pattern with subperpendicular faults by proposing a special type
of plate boundary. These workers proposed that “eduction” is
occurring in this region, i.e., that the underlying plate (the Phil-
ippine Sea plate) does not subduct but is instead pulled away
from the overriding plate (North American plate) with an exten-
sional sense of motion. However, Kato et al. (1983) and Ohk-
ouchi (1990) interpreted seismic profiles to indicate SW-vergent
thrust faults in the area of the Okinoyama Bank Belt, forming
an accretionary prism (Fig. 5). Ohkouchi (1990) suggests that all
of Sagami Bay is characterized by a continuous anticline, based
on seismic profile interpretation, and concludes that horizontal
compression has occurred around this area.
In addition, all the mechanistic solutions for earthquakes
and tsunamis, and the resultant crustal movement from the
Taisho-Kanto earthquake in 1923 (M = 7.9), indicate that this
boundary is not undergoing eduction but subduction (Ando, 1974;
Aita, 1993; Sato et al., 2005). If so, this prompts the question of
why dissection of the topography occurs in the Sagami Bay area
and whether this is different from other areas. The 2D effects of
oblique subduction can be analyzed by a simple sandbox experi-
ment described in the section “Riedel Shear Experiments.”
Middle Sagami Trough Area
The middle Sagami Trough area is situated south off the
shore of the Boso Peninsula as a NW-trending longitudinal basin
(Middle Sagami Trough Basin in Fig. 4), demarcated by two
faults, coincident with the NE Boso Escarpment and the SW Awa
Canyon (Figs. 4 and 7). The interpretation of topography and
162 Ogawa et al.

seismic profiles obtained during the KAIKO project indicates
that the convex topography of the basin-fill sediments between
the two faults represents a modern accretionary prism within an
oblique subduction setting (Tanahashi, 1986; Ogawa et al., 1989)
(Figs. 7 and 8). This part of the present accretionary prism is col-
liding along its NW side with the Boso Peninsula, where a sharp
thrust-fault scarp has developed in the So-oh Trough (Fig. 8). The
Genroku-Kanto earthquake in 1703 (M = 8.2) occurred beneath
this scarp, just off the Boso Peninsula (Shishikura, 2003). Many
thrust faults in en echelon fashion were developed in the southern
fault zone, whereas almost straight faults with dominant strike-
slip motion were developed along the northern boundary of the
basin (Fig. 8). A forearc basin lies on the NE side of the accre-
tionary complex. The Sagami Trough canyon flows in a meander-
ing fashion at the foot of the fault scarp (Boso Escarpment) along
the northeastern boundary (Boso Canyon) (Figs. 4 and 8).
The sediments in the basin were accreted from Izu forearc
basin sediments. Some are slide deposits from the increase in the
slope angle, causing slope instability, as the Izu Arc is colliding
with the island of Honshu (Tokuyama et al., 1988). Within the
basin, oblique faults with superficial fractures are being system-
atically developed (Fig. 8). This set of fractures may be inter-
preted as R1 shears, as discussed in the section “Riedel Shear
Experiments.” The 3D effect of oblique subduction can be suc-
cessfully analyzed by a simple sandbox experiment.
Boso Triple Junction Area
The Boso triple junction area has a unique setting both in
topography and internal structure (Ogawa et al., 1989; Seno et
al., 1989; Iwabuchi et al., 1990). This area is not only tectonically
but also topographically unstable. In a large sense the triple junc-
tion area is a large, deep triangular basin (Figs. 4 and 9), divided
into three parts, one on the NW side, where a wide basin is up to
7400 m deep (North Basin in Fig. 4); another is the trench floor at
9400 m water depth, and the third is surrounded by steep slopes.
The first basin has some deep sea terraces along which mud vol-
canoes and dissecting channels have developed (Ogawa et al.,
1989). Canyons connecting to land on the northern side of the
wide basin are channeled toward this basin but show significant

30 km

(twtsec) 10 km

Figure 7. Seismic profiles of the middle Sagami Trough Basin, showing the Izu forearc sediments (right) accreted to the
left in the Sagami Trough (SGT). BOC—Boso Canyon; AWC—Awa Canyon; MYC—Miyake Canyon; twt—two-way
traveltime. The Miyake Canyon is within the Izu Arc, and the Boso and Awa Canyons are in the Sagami Trough. Adapted
from Ogawa et al. (1989). The bottom profile is from the KAIKO I Research Group (1986).
 Oblique subduction in an island arc collision setting
140°E
55
ne
Li
Line 49
Normal fault
Reverse fault
Fracture
Figure 8. Topographic division of the Middle Sagami Trough Basin area, adapted from Tanahashi (1986). Thrust faults and normal faults and
fractures exhibit en echelon patterns of a Riedel shear system.
meandering (Fig. 4). The canyons appear to be deformed and
dragged by the downwarping of the basin basement (Soh et al.,
1990), as they are not directly oriented toward the deepest part
but show clockwise rotation. We suggest that older canyons were
rotated by such downwarping (Fig. 9).
The main channel in the negative basin flows out through a
narrow gorge into the Izu-Ogasawara (Bonin) trench floor at a
depth of 9400 m. At this point it forms the deep-sea Mogi Fan
(Nakamura et al., 1987; Soh et al., 1988; Ogawa et al., 1989;
Seno et al., 1989) (Figs. 4 and 9). The fan deposits were thrust
and folded locally prior to accretion to the Izu forearc (Geologi-
cal Survey of Japan, 1982; Soh et al., 1988; Ogawa et al., 1989).
Huge deep-sea submarine slide bodies are recognized (Fig. 10).
Surface topographies have many unstable slopes, shown by large-
scale concave and convex cliffs and scarps by submarine sliding.
This was observed in the form of rough ridges and slides dur-
ing unmanned submersible dives (KR99–11 Shipboard Scientific
Party, 2000).
In total the triple junction area is a wide, 100-km-scale tri-
angular area of very unstable topography that is still in a state
of active collapse. Because the present relative motion of the
163
141°E
35°N
34°30'N
surrounding three plate (Fig. 4) indicates that the present triple
junction is unstable, the triple point is moving to the NW, chang-
ing the system of the junction. The topography (Figs. 9 and 10)
indicates such instability, resulting in the unique sedimentation,
accretion, and resultant erosion by gravitational instability.
RIEDEL SHEAR EXPERIMENTS
In order to better understand the tectonics around the Sagami
Trough, we performed two types of simple sandbox tests: one in
2D to model Riedel shears caused by horizontal displacement,
and another in 3D to simulate oblique subduction.
2D Riedel Shear-Box Test
For 2D tests, Bartlett et al. (1981), Naylor et al. (1986), and
many others have run simple shear-box tests and recognized
systematic development of Riedel shears. Unfortunately, these
experiments had no free zone on the displacement basement.
We prepared a new experiment, using a rubber sheet, which
was dislocated along a displacement basement, as shown in
164 Ogawa et al.

140˚E 141˚E 142˚E 143˚E

Boso Peninsula

35˚N 35˚N

M. S. T. B.
Sa
gam
i Tr
oug
h Boso Triple Junction

34˚N 34˚N

33˚N 33˚N
140˚E 141˚E 142˚E 143˚E

-9000 -6500 -6000 -5500 -5000 -4500 -4000 -3500 -3000 -1000

Figure 9. Bathymetric map showing the topography in and around the Boso triple junction, courtesy of JAM-
STEC (JAMSTEC R/V Mirai and other data edited by M. Nakanishi, 2006, personal commun.). Strong lin-
eaments from NW to SE from the Boso Peninsula side (B) continue to the middle part of the trench slope,
gradually becoming obscure in the triangle area near the triple junction. Notice a trench floor submarine fan in
the center, which is deformed by frontal thrust. M—Miura Peninsula; B—Boso Peninsula; I—Izu Peninsula.

Figure 11 (left), relative to the superjacent powder. The differ-
ences and characteristics of our experiment in comparison with
conventional methods are as follows: (1) A rubber sheet is used
as a cover for the dislocating board, and the center of the rub-
ber sheet, 5–15 cm wide, is mobile (freely sheared) while the
margins are settled by resin. (2) Quartz powder sediment, with a
grain size averaging 8 µm in diameter, is set on this rubber in a
layer varying from 1 to 4 cm thick. (3) The board is dislocated in
a right-lateral sense (dextrally) by a continuously moving motor
at a constant speed of 2 mm/s. The strain generated by this stress
appears as en echelon folds. The strain of the rubber sheet in the
en echelon folds is transferred to the overlying sediment. (4) The
strain in the form of fractures on the powder surface during dis-
location is recorded by camera. Because the Taisho-Kanto earth-
quake occurred on a right-lateral (dextral) oblique-slip fault in
the Sagami Basin area, our experimental box was designed to
efficiently mimic this type of horizontal deformation effect with
minor vertical motions.
Our experiment revealed several different results, described
as follows. First, R1 (synthetic Riedel shear) with right-lateral
displacement occurred, which was followed by R2 (antithetic)
shear with a left-lateral sense. Finally, P-shear (thrust shear) with
right-lateral displacement occurred (Fig. 11, right).
Theoretically R1 and R2 are associated as conjugate shear
planes, but in fact the former is more strongly developed, and in
general R2 rarely or only seldom occurs. The P-plane is not a the-
oretically resultant fracture but experimentally occurs as a result
of the accumulation of material to produce horizontally compres-
sive pressure into a thrust-component–bearing shear plane (Nay-
lor et al., 1986).
This stress-strain relation occurs instantaneously under
simple shear. The en echelon folds in the rubber sheet produce
a stress field in the overlying sediment that is different in each
domain of the folded rubber. As a result, the stress field makes a
conjugate set of shears, R1 and R2.
The P-shears occur at the left-overstep part of R1, or
around the R2 shear zones. This is explained as the result of
compressional stress around these areas that gives the shears
a thrust component. P-shear occurs slightly after R1 and R2.
This is because the compressional stress increases according
 Oblique subduction in an island arc collision setting 165

10 km (V. E. = 5)

Figure 10. Three-dimensional (3D) topography at the triple junction (left), showing a 10-km-scale large submarine slide body that has collapsed
onto the trench floor at 9000 m depth (KR99–11 Shipboard Scientific Party, 2000). The dive site of the remotely operated vehicle (ROV) Kaiko
is shown by an arrow, with drawings (right) by Y. Ogawa from video views from the ROV of the collapsed slope. Sediments and rocks of the
slope are composed of inclined strata of Pliocene age with minor blocks of middle Miocene age. V.E.—vertical exaggeration. Depths below sea
level shown in meters.

to the displacement of the basement. The vertical displacement
of R1 and R2 is also due to the compressional stress driven by
the left-lateral (sinistral) overstep of R1. Sediment moved in
the direction of both R1 and R2, after which sediment accumu-
lated within the wedge between the two. This uplifts the sedi-
ment, and small-scale pressure ridges form. As displacement
increases, R1 and R2 change their form into sigmoidal geom-
etries because R1 and R2 are connected.
The 85-km-long, 45-km-wide fault plane for the Taisho-
Kanto earthquake fault in the Sagami Basin area (Ando, 1974;
Aita, 1993) suggests that many subordinate fault planes are
responsible for strain. In fact, at the time of that earthquake, the
Sagami Tectonic Line corresponded to the main fault. In locali-
ties far from the epicenter at the southernmost tip of the Boso
Peninsula and south of the Miura Peninsula, motion was recorded
on exposed active faults. These facts indicate that plate motion
has accommodated not only one fault plane but associated fault
systems as well. Furthermore, two asperity parts have recently
been discovered for the Taisho-Kanto earthquake (Sato et al.,
2005) that reveal the diversity of the slip plane. This realization
indicates that the fault had some irregular concentrations of stress
and strain.
Thus the main fault of the Taisho-Kanto earthquake had
diverse subfaults. Independent preexisting faults or newly formed
faults both moved simultaneously. The subducted-landward, over-
riding plate is thought to have deformed by ductile processes out-
side the asperity region. This model leads us to think that the duc-
tile rubber sheet, overlain by quartz powder, effectively mimics
the system, with local folds being similar to the asperity regions.
Comparison with the Topography of Sagami Bay
In relation to plate rheology, Shimamoto (1989) showed that
in a thin plate, as in an island arc, where dislocation is dispersed
over a wide area, the plate boundary comprises a wide zone, and
the island arc is deformed internally. In our study region the Sag-
ami Basin area corresponds to the Izu Arc collision zone, and the
northern part acts as the hanging wall of the subduction zone in
which the North American plate undergoes complicated defor-
mation and fracturing (dislocation) (Lallemant et al., 1996). In
166 Ogawa et al.

Figure 11. 2D Riedel shear box-test equipment with a 10-cm-wide free zone of rubber with right-lateral displacement (left), and one of the rep-
resentative results (right), where R1, R2, and P show synthetic Riedel shear, antithetic Riedel shear, and thrust shear, respectively. After Takami
(1999). D.T.—Dislocating Table; M.R.S.—Movable Rubber Sheet; U.M.R.S.—Unmovable Rubber Sheet.

addition, some convex topography reflects asperities in strain or
stress concentrations (Sato et al., 2005). If such factors support
these facts, then Riedel shears (R1 and R2) must develop as con-
jugate shear fractures in the superficial sediment layers above the
hanging wall. We now compare examples of topography from the
Sagami Bay area with our experimental results.
Based on the experiments, the overriding plate was predicted
to deform by ductile shear, resulting in conjugate Riedel shears,
which we discuss below. The relative plate-movement direction
along the Sagami Trough indicates that the Sagami Tectonic Line
has oblique dextral dislocation (Fig. 4). If we assume that the
strike of the Taisho-Kanto earthquake fault was N59°W (Aita,
1993), and this is considered to be the Y-shear (main shear) direc-
tion, then the Sagami Tectonic Line must be comparable to R1.
Our experiments indicate that R2 should also occur, and the
direction corresponds well to the actual submarine topography.
In order to corroborate the possibility of these fault systems,
we plotted rose diagrams to graphically represent the directions
of canyons and faults on the landward slope of Sagami Bay, as
well as all the fractures in our experiments with <3 cm disloca-
tion with all rubber widths (Fig. 12). It is clear that the directions
in the two diagrams are similar. The internal angle between R1
and R2 in the experiments is ~10° smaller than the actual topog-
raphy, but this may be due to the rotation of topography, as shown
in experiments discussed below.
It was not easy to infer the horizontal displacement sense of
the topography just from bathymetric maps, but the faults around
the Okinoyama Bank Chain are straightly aligned, suggesting
 Oblique subduction in an island arc collision setting 167

a tectonic origin owing to high-angle, horizontal-displacement

faults (Fig. 6). Seismic profiles indicate that the western bound-
ary of the Miura Knoll (part of the Okinoyama Bank Chain) has
a very steep dip (Fig. 5), suggesting horizontal displacement on
a fault. As for the vertical displacement of the Sagami Tectonic
Line, the evidence of the Taisho-Kanto earthquake shows that the
Okinoyama Bank Chain was uplifted by a thrust component.
The experiments are associated with small displacement, but
the actual topography indicates uniform positive relief only on the
eastern side of the Sagami Tectonic Line. The difference between
the model and the observation may be due to our neglecting
the vertical displacement of the Taisho-Kanto earthquake. The
effect where the basement plate has a thrust-component–bearing
oblique fault that affects the overlying sediment was modeled in
3D by Naylor et al. (1986)’s sandbox experiments (Mandl, 1988).
This result shows that the basement dislocation may undergo R1
strain, with a thrust-faulting effect within an en echelon arrange-
ment. Naylor et al.’s experiment predicts uplift along the eastern
side of the Sagami Tectonic Line as a “flower structure.”
The comparison of experiment and topography suggests
that, although some of the submarine fault displacement is not
known in terms of the sense of motion, the present topography of
the Sagami Bay correlates well with the experiment in terms of
the dominant fault and fracture directions. In addition, our study
indicates that the submarine topography is mostly influenced by
a regional dextral shear field.

3D Oblique Subduction Test

As 3D experimental tests are rare, and oblique subduction

models in particular had not previously been performed, we made
a simple but effective test using a steel frame and stiff paper board
(Takazawa, 2003). Wheat powder with colored-chalk layers and
chocolate-powder mesh on the surface was prepared over the sub-
duction boundary in order to mimic the trench wedge sediments.
The inclined board (sloping 30° downward) was moved steadily
by hand at variable degrees of obliqueness. Fractures on the sur-
face and faults and folds inside were shown by cutting the model
with a knife after the movement. For each different degree of
obliqueness (0°, parallel to the trench, to 90° in 10° steps) the sur-
face fracture development and internal structures were recorded Figure 12. Rose diagrams of Riedel shear fractures of sandbox test
by photograph. Representative results are shown in Figure 13. with right-lateral displacement (above; see Fig. 11) and topographic
lineament of the Sagami Basin (below; see Fig. 6). Each dominant di-
Results of 3D Tests rection indicates synthetic Riedel shears and antithetic Riedel shears of
right-lateral displacement, respectively. After Takami (1999).

Figure 13 shows the most representative case, in which the

obliqueness angle was 30°. Surface fractures developed system-
atically. First, Riedel R1 shears formed, then locally R2 shears
were seen, but these were relatively rare. Another direction of
shear, interpreted as thrust shears (P-shears) formed at a later
stage. Internally within the wedge, thrusts and their associated
folds systematically developed (Fig. 13, left). In some cases
the internal thrusts developed during the initial stage, which is
similar to the accretionary prism model, because the thrust on
the ocean side formed at a later stage. In most cases, one or two
back thrusts developed as the deformation proceeded, as part of
a flower structure.
This simple experiment clearly indicates that in cases of
greater obliqueness (especially <40°), stronger Riedel shears
develop, but the effect of thrust and fold development is more
or less the same as in the case of smaller obliqueness (close to
168 Ogawa et al.

Figure 13. Results of 3D oblique subduction tests of right-lateral sense with 30° obliquity. Profile (left), and top surface
fractures (right). Notice components of fractures on the surface of the oblique subduction; synthetic Riedel shears and
thrust shears are distinct. Compare the experiment with the real examples from the Middle Sagami Trough Basin in Fig-
ure 8. After Takazawa (2003).

normal convergence). However, in the less oblique experiment,
with convergence at more than 70°, no strong P-shears formed.
The best fit to the Middle Sagami Trough Basin area in
terms of the surface fracture system is for 20°–30° of oblique-
ness (Fig. 13). This corresponds to the natural case in the area,
where measured obliqueness is ~30° (Figs. 4 and 8). In the So-oh
Trough, between the Middle Sagami Trough area and the Sagami
Bay area (NW corner in Fig. 8), just in front of the Boso Pen-
insula, obliqueness is almost normal, and no Riedel shears are
observed on modern bathymetric maps. Therefore, we conclude
that even a simple 3D oblique subduction model is successful in
mimicking the natural features of the seafloor (Takazawa, 2003).
CONCLUSIONS
In this study we verified the origin of the fault system and
related topography of the Sagami Trough and used experimen-
tal modeling to understand the development of Riedel shears in
oblique subduction settings within an island arc collision zone in
a TTT-type triple junction. In experiments, well-defined shears
associated with horizontal displacement appeared both in 2D and
3D tests. The orientation of faults along the Sagami Trough, both
in the northwestern part (Sagami Bay area) and in the middle
part (Middle Sagami Trough Basin area) can be explained as the
dominant two directions oblique to the main fault, which cor-
relate well with R1 and R2 shears in sandbox experiments. The
Sagami Tectonic Line can be understood as a plate-boundary
thrust fault, where an oblique fault has a horizontal displacement
with a thrust component.
These results indicate that the fault system and topography
can be attributed to Riedel shears, which generate wide regional
surface deformation driven by a dextral shear zone that is trig-
gered by highly oblique subduction. More detailed constraints
are needed to verify a direct link between the topography, geol-
ogy, and tectonics. We suggest that direct observation and sam-
pling at the seabed, supplemented by drilling and dredging,
are needed to understand the triple junction structure in three
dimensions. Only by doing this can our proposed model be rig-
orously tested.
The different appearance of shear patterns in the Sagami
Bay area and the Middle Sagami Basin area may be attributed
to the different relationships between the basement and the
overlying sediment cover. In the Sagami Basin a wide and thick
accretionary wedge above an irregular basement (modeled with
folds of the rubber sheet) reflects strain-asperity zones and is
deformed by strain partitioning. As a result, both R1 and R2
shears are formed continuously. In contrast, in the case of the
Middle Sagami Trough Basin the strain is transferred in a more
regular fashion to the sediments from the subduction, forming
a regular shear pattern. Further quantitative analysis by more
sophisticated methods is needed to more fully understand this
complicated region, but the simple experiments described here
offer a plausible solution.
ACKNOWLEDGMENTS
We are grateful to Peter Clift and Amy Draut for inviting us to
the 2006 Penrose Conference in Price, Utah. Peter Clift, Roland
von Huene, and one anonymous reviewer kindly read the early
version of the manuscript, and gave us critical suggestions for
revision. Some bathymetric charts and figures were prepared
through the courtesy of JAMSTEC and Masao Nakanishi.
 Oblique subduction in an island arc collision setting
REFERENCES CITED
Aita, I., 1993, Historical tsunamis in NW Sagami Bay and its numerical model:
Journal of Geography, Tokyo Geographical Society, v. 102, p. 427–436.
Ando, M., 1974, Seismo-tectonics of the 1923 Kanto Earthquake: Journal of
Physics of the Earth, v. 22, p. 263–277.
Aoike, K., 1999, Tectonic evolution of the Izu collision zone: Research Report
of Kanagawa Prefectural Museum of Natural History, v. 3, p. 111–151.
Bartlett, W.L., Freedman, M., and Logan, J.M., 1981, Experimental folding and
faulting of rocks under confining pressure: Tectonophysics, v. 79, p. 255–
277, doi: 10.1016/0040-1951(81)90116-5.
Fujioka, K., and Sakamoto, I., 1999, Structure of the Izu-Bonin Arc: Research
Report of Kanagawa Prefectural Museum of Natural History, v. 3, p. 5–20.
Geological Survey of Japan, 1982, Geological Map of the Northern Ogasawara
Arc: Marine Geology Map Series 18.
Hall, R., Ali, J.R., Anderson, C.D., and Baker, S.J., 1995, Origin and motion
history of the Philippine Sea plate: Tectonophysics, v. 251, p. 229–250,
doi: 10.1016/0040-1951(95)00038-0.
Hanamura, Y., and Ogawa, Y., 1993, Layer-parallel faults, duplexes, imbricate
thrust and vein structures of the Miura Group: Key to understanding the
Izu fore-arc sediment accretion to the Honshu forearc: Island Arc, v. 3,
p. 126–141.
Hirano, N., Ogawa, Y., Saito, K., Yoshida, T., Sato, H., and Taniguchi, H., 2003,
Multi-stage evolution of the Tertiary Mineoka Ophiolite, Japan: New geo-
chemical and age constraints, in Dilek, Y., and Robinson, P., eds., Ophio-
lite in Earth History: Geological Society [London] Special Publication
218, p. 279–298.
Hirono, T., and Ogawa, Y., 1998, Duplex arrays and thickening of accretionary
prisms—An example from Boso Peninsula, Japan: Geology, v. 26, p. 779–
782, doi: 10.1130/0091-7613(1998)026<0779:DAATOA>2.3.CO;2.
Huchon, P., and Labaume, P., 1989, Central Japan triple junction: A three-
dimensional compression model: Tectonophysics, v. 160, p. 117–133,
doi: 10.1016/0040-1951(89)90387-9.
Ishiwatari, A., Yanagida, Y., Li, Y.-B., Ishii, T., Haraguchi, S., Koizumi, K., Ich-
iyama, Y., and Umekawa, M., 2006, Dredge petrology of the boninite- and
adakite-bearing Hahajima Seamount of the Ogasawara (Bonin) forearc:
An ophiolite or a serpentinite seamount?: Island Arc, v. 15, p. 102–118,
doi: 10.1111/j.1440-1738.2006.00512.x.
Isozaki, Y., 1996, Anatomy and genesis of subduction-related orogen: A new
view of subdivision and evolution of the Japanese Islands: Island Arc,
v. 5, p. 289–320, doi: 10.1111/j.1440-1738.1996.tb00033.x.
Iwabuchi, Y., Asada, A., and Kato, Y., 1990, Multichannel seismic profiling of
the Boso triple junction: Marine Survey and Technology, v. 2, p. 29–38.
KAIKO I Research Group, 1986, Topography and Structure of Trenches around
Japan—Data Atlas of Franco-Japanese KAIKO Project, Phase I: Ocean
Research Institute, University of Tokyo, Institut Français de Recherches
pour l’Exploitation de la Mer (IFREMER) and Centre National de la
Recherche Scientifique (CNRS), 305 p.
Kanamatsu, T., Herroro-Bervera, E., Taira, A., Saito, S., Ashi, J., and Furumoto,
A.S., 1996, Magnetic fabric development in the Tertiary accretionary
complex in the Boso and Miura Peninsulas of central Japan: Geophysical
Research Letters, v. 23, p. 471–474, doi: 10.1029/96GL00147.
Kanie, Y., 1999, Cenozoic System and geological structure of the Miura Penin-
sula and Tokyo Bay to Boso Peninsula, southern-central Japan: Research
Report of Kanagawa Prefecture Museum of Natural History, v. 9, p. 79–
94.
Kato, S., Sato, Y., and Sakurai, M., 1983, Multichannel seismic survey in the
Nankai: Suruga and Sagami Troughs: Report of Hydrographic Research,
v. 18, p. 1–23.
Kimura, M., 1976, Marine geology in the Sagami-Nada Sea and its vicinity:
Geological Survey of Japan, Marine Geological Map Series 3, 9 p.
Kinoshita, H., 1980, Paleomagnetism of sediment cores from Deep Sea Drilling
Project, Leg 58, Philippine Sea, in Klein, G.DeV., Kobayashi, K., et al.,
Initial Reports of the Deep Sea Drilling Project, v. 58: Washington, D.C.,
Government Printing Office, p. 765–768.
Kitazato, H., 1997, Paleogeographic changes in central Honshu, Japan, during
the late Cenozoic in relation to the collision of the Izu-Ogasawara Arc
with the Honshu Arc: Island Arc, v. 6, p. 144–157, doi: 10.1111/j.1440-
1738.1997.tb00166.x.
Kobayashi, K., 1984, Subsidence of the Shikoku back-arc basin: Tectonophys-
ics, v. 102, p. 105–117, doi: 10.1016/0040-1951(84)90010-6.
169
Kobayashi, K., 1995, Role of subducted lithospheric slab in uplift and subsid-
ence of the northwestern margins: Marine Geology, v. 127, p. 119–144,
doi: 10.1016/0025-3227(95)00007-L.
Kobayashi, K., Kasuga, S., and Okino, K., 1995, Shikoku Basin and its mar-
gins, in Taylor, B., ed., Back-Arc Basins in the Western Pacific: New York,
Plenum, p. 381–405.
Kong, Y.S., Kagami, H., Tokuyama, H., Igarashi, C., and Nasu, N., 1984, Sub-
bottom geological structures of Sagami Bay: University of Tokyo, Bul-
letin of Ocean Research Institute, v. 16, p. 1–91.
KR99–11 Shipboard Scientific Party, 2000, Onboard Report of KR99–11
Cruise, JAMSTEC.
Lallemant, S.J., Le Pichon, X., Thoue, F., Henry, P., and Saito, S., 1996, Shear
partitioning near the central Japan triple junction: The 1923 great Kanto
earthquake revisited—I: Geophysical Journal International, v. 126,
p. 871–881.
Lee, I.T., and Ogawa, Y., 1998, Bottom-current deposits in the Miocene-Pliocene
Misaki Formation, Izu forearc area, Japan: Island Arc, v. 7, p. 315–329.
Lus, W.Y., McDougall, I., and Davies, H.L., 2004, Age of the metamorphic sole
of the Papuan Ultramafic Belt ophiolite, Papua New Guinea: Tectono-
physics, v. 392, p. 85–101, doi: 10.1016/j.tecto.2004.04.009.
Mandl, G., 1988, Mechanics of Tectonic Faulting: New York, Elsevier Science,
p. 141–144.
Maruyama, S., and Seno, T., 1986, Orogeny and relative plate motions: Exam-
ple of the Japanese Islands: Tectonophysics, v. 127, p. 305–329, doi:
10.1016/0040-1951(86)90067-3.
Matsuda, T., 1978, Collision of the Izu-Bonin arc with central Honshu: Ceno-
zoic tectonics of the fossa magna, Japan: Journal of Physics of the Earth,
v. 26, Supplement, p. S409–S421.
McKenzie, D.P., and Morgan, W.J., 1969, Evolution of triple junction: Nature,
v. 224, p. 125–133, doi: 10.1038/224125a0.
Mohiuddin, M.M., and Ogawa, Y., 1998, Late Paleocene to middle Miocene
pelagic sequences in the Boso Peninsula, Japan: New light on NW Pacific
tectonics: Island Arc, v. 7, p. 301–314.
Mori, R., and Ogawa, Y., 2005, Transpressional tectonics of the Mineoka Ophiolite
Belt in a trench–trench–trench-type triple junction, Boso Peninsula, Japan:
Island Arc, v. 14, p. 571–581, doi: 10.1111/j.1440-1738.2005.00485.x.
Nakamura, K., Shimazaki, N., and Yonekura, A., 1984, Subduction, bending
and eduction—Present and Quaternary tectonics of the northern border
of the Philippine Sea plate: Bulletin of Société Géologie de France, v. 26,
p. 221–243.
Nakamura, K., et al., 1987, Oblique and near collision subduction, Sagami
and Suruga Troughs—Preliminary results of the French-Japanese Kaiko
cruise, Leg 2: Earth and Planetary Science Letters, v. 83, p. 229–242, doi:
10.1016/0012-821X(87)90068-9.
Naylor, M.A., Mandl, G., and Supersteinjn, C.H.K., 1986, Fault geometries
in basement-induced wrench faulting under different initial stress states:
Journal of Structural Geology, v. 8, p. 737–752, doi: 10.1016/0191-
8141(86)90022-2.
Niitsuma, N., 1988, Neogene tectonic evolution of southwest Japan: Modern
Geology, v. 12, p. 497–532.
Ogawa, Y., 1982, Tectonics of some forearc fold belts in and around the arc-arc
crossing area in central Japan, in Legget, J.K., ed., Trench-Forearc Geol-
ogy: Geological Society [London] Special Publication 10, p. 49–61.
Ogawa, Y., 1983, Mineoka ophiolite belt in the Izu forearc area—Neogene
accretion of oceanic and island arc assemblages in the northeastern cor-
ner of the Philippine Sea plate, in Hashimoto, M., and Uyeda, S., eds.,
Accretion Tectonics in the Circum-Pacific Region: Tokyo, Terrapub,
p. 245–260.
Ogawa, Y., 1985, Variety of subduction and accretion processes in Cretaceous
to Recent plate boundaries around southwest and central Japan: Tectono-
physics, v. 112, p. 493–518, doi: 10.1016/0040-1951(85)90192-1.
Ogawa, Y., and Horiuchi, K., 1978, Two types of accretionary fold belts in cen-
tral Japan: Journal of Physics of the Earth, v. 26, Suppl., S321–336.
Ogawa, Y., and Sashida, K., 2005, Lower Cretaceous radiolarian bedded chert
from the Mineoka Belt, Boso Peninsula, Japan: Journal of the Geological
Society of Japan, v. 111, p. 624–627.
Ogawa, Y., and Takahashi, A., 2004, Seafloor spreading, obduction and triple
junction tectonics of the Mineoka ophiolite, Central Japan: Tectonophys-
ics, v. 392, p. 131–141, doi: 10.1016/j.tecto.2004.08.007.
Ogawa, Y., and Taniguchi, H., 1987, Ophiolitic melange in forearc area and
formation of the Mineoka belt: Science Report, Kyushu University: Geol-
ogy, v. 15, p. 1–23.
170 Ogawa et al.
Ogawa, Y., and Taniguchi, H., 1988, Geology and tectonics of the Miura-Boso
Peninsulas and the adjacent area: Modern Geology, v. 12, p. 147–168.
Ogawa, Y., Horiuchi, K., Taniguchi, H., and Naka, J., 1985, Collision of the
Izu arc with Honshu and the effects of oblique subduction in the Miura-
Boso Peninsulas: Tectonophysics, v. 119, p. 349–379, doi: 10.1016/0040-
1951(85)90046-0.
Ogawa, Y., Seno, T., Tokuyama, H., Akiyoshi, H., Fujioka, K., and Taniguchi,
H., 1989, Structure and development of the Sagami Trough and off-Boso
triple junction: Tectonophysics, v. 160, p. 135–150, doi: 10.1016/0040-
1951(89)90388-0.
Ogawa, Y., Tamaki, K., Isozaki, Y., and Shikazono, N., 1997, Japan, in Moores,
E.M., and Fairbridge, R.W., eds., Encyclopedia of European and Asian
Regional Geology: London, Chapman and Hall, p. 436–449.
Ohkouchi, N., 1990, Active structure and tectonics of the Sagami Bay: Journal
of Geography, Tokyo Geographical Society, v. 99, p. 458–470.
Okino, K., Shimakawa, Y., and Nagaoka, S., 1994, Evolution of the Shikoku
Basin: Journal of Geomagnetics and Geoelectronics, v. 46, p. 463–479.
Pautot, G., et al., 1987, Deep-sea submersible survey in the Suruga, Sagami
and Japan Trenches: Preliminary results of the 1985 Kaiko cruise, Leg 2:
Earth and Planetary Science Letters, v. 83, p. 300–312, doi: 10.1016/0012-
821X(87)90073-2.
Renard, V., et al., 1987, Trench triple junction off central Japan—Preliminary
results of French–Japanese 1984 Kaiko cruise, Leg 2: Earth and Planetary
Science Letters, v. 83, p. 243–256, doi: 10.1016/0012-821X(87)90069-0.
Saito, S., 1992, Stratigraphy of Cenozoic strata in the southern terminus area of
Boso Peninsula, Central Japan: Contributions of the Institute of Geology
and Paleontology, Tohoku University, v. 93, p. 1–37.
Sato, H., and Ogawa, Y., 2000, Sulfide minerals in peridotites as tectonic indi-
cators for genesis of ophiolitic rocks: Example from peridotite in the
Hayama-Mineoka Belt, central Japan, in Dilek, Y., Moores, E., Ethon, D.,
and Nichols, A., eds., Ophiolites and Oceanic Crust: Geological Society
of America Special Paper 349, p. 427–437.
Sato, H., Taniguchi, H., Takahashi, N., Mohiuddin, M.M., Hirano, N., and
Ogawa, Y., 1999, Origin of the Mineoka Ophiolite: Journal of Geography,
Tokyo Geographical Society, v. 108, p. 203–215.
Sato, H., Hirata, N., Koketsu, K., Okaya, D., Abe, S., Kobayashi, R., Matsub-
ara, M., Iwasaki, T., Ito, T., Ikawa, T., Kawanaka, T., Kasahara, K., and
Harder, S., 2005, Earthquake source fault beneath Tokyo: Science, v. 309,
p. 462–464, doi: 10.1126/science.1110489.
Seno, T., 1984, Was there a North New Guinea plate?: Geological Survey of
Japan Report, v. 263, p. 29–42.
Seno, T., and Maruyama, S., 1984, Paleogeographic reconstruction and origin
of Philippine Sea: Tectonophysics, v. 102, p. 53–84, doi: 10.1016/0040-
1951(84)90008-8.
Seno, T., Ogawa, Y., Tokuyama, H., Nishiyama, E., and Taira, A., 1989, Tec-
tonic evolution of the triple junction off central Honshu for the past 1
million years: Tectonophysics, v. 160, p. 91–116, doi: 10.1016/0040-
1951(89)90386-7.
Seno, T., Stein, S., and Gripp, A.E., 1993, A model for the motion of the Philip-
pine Sea plate consistence with NUVEL-1 and geological data: Journal of
Geophysical Research, v. 89, p. 17941–17948.
Shimamoto, T., 1989, Rheology of rocks and plate tectonics: Kagaku, v. 59,
p. 170–181.
Shishikura, M., 2003, Cycle of interpolate earthquake along the Sagami Trough,
deduced from tectonic geomorphology: Bulletin of Earthquake Research
Institute: University of Tokyo, v. 78, p. 245–254.
Soh, W., Taira, A., and Tokuyama, H., 1988, A trench fan in the Izu-Ogasawara
Trench on the Boso Trench triple junction, Japan: Marine Geology, v. 82,
p. 235–249, doi: 10.1016/0025-3227(88)90143-0.
Soh, W., Taira, A., Ogawa, Y., Taniguchi, H., Pickering, K.T., and Stow, D.A.V.,
1989, Submarine depositional processes for volcaniclastic sediments in
the Mio–Pliocene Misaki Formation, Miura Group, Central Japan, in
Taira, A., and Masuda, F., eds., Sedimentary Facies in the Active Plate
Margin: Tokyo, Terrapub, p. 619–630.
Soh, W., Tokuyama, H., Fujioka, K., Kato, S., and Taira, A., 1990, Morphology
and development of a deep-sea meandering canyon (Boso Canyon) on
an active plate margin, Sagami Trough, Japan: Marine Geology, v. 91,
p. 227–241, doi: 10.1016/0025-3227(90)90038-L.
Soh, W., Pickering, K.T., Taira, A., and Tokuyama, H., 1991, Basin evolution
in the arc–arc Izu Collision Zone, Mio–Pliocene Miura Group, Central
Japan: Geological Society [London] Journal, v. 148, p. 317–330.
Stow, D.A.V., Taira, A., Ogawa, Y., Soh, W., Taniguchi, H., and Pickering, K.T.,
1998, Volcaniclastic sediments, processes interaction and depositional
setting of the Mio-Pliocene Miura Group, SE Japan: Sedimentary Geol-
ogy, v. 115, p. 351–381, doi: 10.1016/S0037-0738(97)00100-0.
Stow, D.A.V., Ogawa, Y., Lee, In Tae, and Mitsuzawa, K., 2002, Neogene
contourites, Miura-Boso forearc basin, SE Japan, in Stow, D.A.V., et al.,
eds., Deep-Water Contourite Systems: Modern Drifts and Ancient Series,
Seismic and Sedimentary Characteristics: Geological Society [London]
Memoirs, v. 22, p. 409–419.
Taira, A., Tokuyama, H., and Soh, W., 1989, Accretion tectonics and evolution
of Japan, in Ben-Avraham, Z., ed., The Evolution of the Pacific Ocean
Margins: Oxford, UK, Oxford University Press, p. 100–132.
Takahashi, M., and Saito, K., 1997, Miocene intra-arc bending at an arc–arc
collision zone, central Japan: Island Arc, v. 6, p. 168–182, doi: 10.1111/
j.1440-1738.1997.tb00168.x.
Takahashi, A., Ogawa, Y., Ohta, Y., and Hirano, N., 2003, The nature of faulting
and deformation in the Mineoka Ophiolite, NW Pacific Rim, in Dilek, Y.,
and Robinson, P., eds., Ophiolite in Earth History: Geological Society
[London] Special Publication 218, p. 299–314.
Takami, Y., 1999, Tectonic interpretation of submarine topography in the Sag-
ami Bay based on Riedel shear experiments [B.S. thesis]: Tsukuba, Japan,
University of Tsukuba, 51 p.
Takazawa, S., 2003, The submarine topography and tectonics in the middle of
Sagami Trough based on the subduction model experiment [B.S. thesis]:
Tsukuba, Japan, University of Tsukuba, 60 p.
Tanahashi, M., 1986, Structures of sediments and tectonics of Sagami Trough:
Earth Monthly, v. 8, p. 238–245.
Tokuyama, H., et al., 1988, Multi-channel seismic profile in the south of Izu-
Oshima Island: Kazan (Volcano), 2nd Ser., v. 33, p. 67–77.
Yamamoto, Y., 2006, Systematic variation of shear-induced physical properties
and fabrics in the Miura–Boso accretionary prism: The earliest processes
during off-scraping: Earth and Planetary Science Letters, v. 244, p. 270–
284, doi: 10.1016/j.epsl.2006.01.049.
Yamamoto, Y., and Kawakami, S., 2005, Rapid tectonics of the Late Miocene
Boso accretionary prism related to the Izu–Bonin arc collision: Island
Arc, v. 14, p. 178–198, doi: 10.1111/j.1440-1738.2005.00463.x.
Yamamoto, Y., Ohta, Y., and Ogawa, Y., 2000, Implication for the two-stage
layer-parallel faults in the context of Izu forearc collision zone—Exam-
ples from the Miura accretionary prism, central Japan: Tectonophysics, v.
325, p. 133–144, doi: 10.1016/S0040-1951(00)00134-7.
Yamamoto, Y., Mukoyoshi, H., and Ogawa, Y., 2005, Large and small struc-
tural characteristics in shallow burial accretionary prism on land: Rapidly
uplifted Neogene accreted sediments in the Miura-Boso Peninsula: Tec-
tonics, v. 24, p. TC5008, doi: 10.1029/2005TC001823.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

The West Crocker formation of northwest Borneo: A Paleogene accretionary prism

Joseph J. Lambiase*
Tan Yaw Tzong*
Amelia G. William*
Department of Petroleum Geoscience, Universiti Brunei Darussalam, Tungku, Brunei BE 1410

Michael D. Bidgood
Patrice Brenac
GSS International, Aberdeenshire, UK

Andrew B. Cullen
Sabah Shell Sendirian Berhad, Miri, Malaysia

ABSTRACT

An integrated structural, stratigraphic, and sedimentological analysis of the

West Crocker formation in northwest Borneo suggests that it is best interpreted as an
accretionary prism. The structural geology provides clear evidence of at least two epi-
sodes of syndepositional folding and thrust faulting. A probable Eocene age, indicated
by foraminiferal and palynological assemblages, differs from the generally accepted
Oligocene to early Miocene age and is consistent with deposition of the West Crocker
formation during a phase of tectonism at the northwest Borneo margin. Sandstones
within the West Crocker formation were deposited by high-density turbidity currents
that constructed relatively small, progradational lobes in a slope apron environment,
and trace fossil assemblages confirm bathyal water depths of ~1000 m or more. The
composition of the sandstones, which contain abundant feldspars and lithic fragments,
suggests that their provenance was the first-cycle product of an eroded orogenic belt,
whereas immature textures indicate a short distance of transport.

Keywords: turbidites, syntectonic, accretionary prism, Borneo.

INTRODUCTION Formation (e.g., Hutchison et al., 2000; Fig. 1A). The Crocker-
Temburong Formation has long been recognized as a thick suc-
The West Crocker formation is an informal name for distin- cession of turbidites with northward-directed paleocurrents that
guishing the western and northwestern outcrops of a Paleogene was deposited in response to collisional tectonics associated with
succession in northwest Borneo that constitutes the sand-rich Late Cretaceous to early Miocene rifting in the South China Sea
Crocker Formation and its coeval shaly equivalent, the Temburong (Stauffer, 1967; Wilson, 1964; Tan and Lamy, 1990; Hutchison,
*Present address, Lambiase: Lambiase Geoscience Pte. Ltd., Singapore; e-mail: joe_lambiase@yahoo.com. Present address, Tzong: Sabah Shell Sendirian Berhad,
Miri, Malaysia. Present address, William: ExxonMobil Exploration and Production Malaysia Inc., Kuala Lumpur, Malaysia.

Lambiase, J.J., Tzong, T.Y., William, A.G., Bidgood, M.D., Brenac, P., and Cullen, A.B., 2008, The West Crocker formation of northwest Borneo: A Paleogene
accretionary prism, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological
Society of America Special Paper 436, p. 171–184, doi: 10.1130/2008.2436(08). For permission to copy, contact editing@geosociety.org. ©2008 The Geological
Society of America. All rights reserved.

171
172 Lambiase et al.
Figure 1. (A) Location of Borneo within southeast Asia and the distri-
bution of the West Crocker and Temburong formations, and the Rajang
Group (undifferentiated Crocker and Trusmadi Formations). (B) Loca-
tion of the Bukit Melinsung (BM), Jalan Salaiman (JS), and Maju (MJ)
outcrops; all are along main roads that are shown as dashed lines.
1996). Various authors have interpreted the unit as one major
turbidite fan system derived from a relatively distant southerly
source (e.g., Crevello, 2002; Tongkul, 1987). However, Hutchison
(2005) notes that it is difficult to envision a single fan system per-
sisting through the entire Paleogene to early Miocene age span of
the Crocker-Temburong Formation, and unconformities are sus-
pected if not proven. Also, the structural style and burial history
of the various parts of the formation differ significantly, suggest-
ing that the unit comprises a number of successive fan systems,
of which the West Crocker formation is the youngest and least
deeply buried (Hutchison, 2005).
The West Crocker formation lies directly below the Deep
Regional Unconformity of Levell (1987), a profound angular
unconformity that was generated during an orogenic event that
uplifted Paleogene and older rocks, and marks the transition from
that uplift to the rapid subsidence that accompanied deposition of
the thick (up to 14 km), overlying Neogene shallow-marine clas-
tic succession (Fig. 2). The Deep Regional Unconformity is gen-
erally regarded as being of late early Miocene age (e.g., Sandal,
1996), although it is diachronous, and some workers prefer a mid-
dle Miocene age (e.g., Hutchison et al., 2000). In the Luconia and
Balingian provinces of Sarawak a regional unconformity of simi-
lar age separates cycle II and cycle III (Madon, 1999; Madon and
Abolins, 1999) and generally marks the transition from extension
to regional subsidence. There has been no direct age dating of the
West Crocker formation, because the few microfossils that have
been recovered are not age-diagnostic within the Eocene to Mio-
cene (Hutchison, 2005). However, the West Crocker formation is
regarded as a sandy facies that is stratigraphically equivalent to the
shaly Temburong Formation (Fig. 2). The Temburong Formation
is the youngest part of the Crocker Formation-Temburong Forma-
tion complex and was dated as early Oligocene to early Miocene
by Wilson (1964), using foraminifera. Therefore, the West Crocker
formation is presumed to lie directly below the Deep Regional
Unconformity that marks the boundary with the overlying Neo-
gene shallow-marine clastic sedimentary rocks (Fig. 2).
It is generally agreed that the turbidites were deposited in a
foreland basin that developed in response to uplift of the Rajang
Group accretionary prism during the Sabah orogeny that began
in the late Eocene (ca. 36 Ma; e.g., Tongkul, 1990; Hutchison,
1996), although the paleogeography of northwest Borneo and
the exact timing of events during the early Tertiary are poorly
constrained. The thickness of the West Crocker succession is
uncertain partly because postdepositional thrust faulting makes
stratigraphic reconstruction between outcrops nearly impossible;
Wilson (1964) suggested that at least 6000 m of section is pres-
ent, whereas Tongkul (1987) argues for ~1000 m.
By the late Oligocene the West Crocker formation was
being thrust over an area of buried continental shelf and attenu-
ated continental crust called Dangerous Grounds (Hutchison,
2004; Hutchison et al., 2000). The SW-directed underthrusting of
Dangerous Grounds during subduction generated NW-SE com-
pression that was responsible for the complex structures of the
Eocene to middle Miocene sedimentary rocks in western Sabah,
including the West Crocker strata (Tongkul, 1990, 1991). Apa-
tite fission-track data indicate that the West Crocker sedimentary
rocks were buried to 4–8 km and were exhumed during develop-
ment of the middle Miocene regional unconformity (Hutchison
et al., 2000).
 West Crocker formation of northwest Borneo
Present Study
This study investigated the sedimentology, stratigraphic
architecture, and structural style of the West Crocker formation
turbidites in order to develop a tectono-stratigraphic model for
their deposition. Three outcrops were selected for detailed study:
Bukit Melinsung, Jalan Salaiman, and Maju (Fig. 1B). The first
two were chosen because they are much less deformed than
most. One of them has the thickest continuously exposed verti-
cal succession, whereas the other represents the most continu-
ous lateral exposure. Approximately 350 m of vertical section is
exposed continuously in the Bukit Melinsung outcrop (Fig. 1B),
although lateral continuity is limited to ~50 m. About 70% of the
succession is very fine to granular sandstone, and the remaining
30% is shale. Individual beds in the Jalan Salaiman outcrop near
Kota Kinabalu (Fig. 1B) can be traced laterally for ~500 m. They
173
Figure 2. Early Tertiary stratigraphy of
northwest Borneo, with suggested revisions
to the scheme of Hutchison et al. (2000).
consist of very coarse to coarse grained, 1–6-m-thick sandstones
interbedded with thinner shales in a 130-m-thick, continuously
exposed vertical succession.
The Maju outcrop was selected because it covers a large area
and has a wide variety of structures in >800 m of exposed sec-
tion, representing >400 m of stratigraphic succession (Fig. 1B).
The succession has been deformed by multiple folding and fault-
ing events that have generated thrust faults, normal faults, and a
wide range of fold types.
SEDIMENTOLOGY AND STRATIGRAPHY
The outcrops were logged to define grain size trends, sedi-
mentary structures, the nature of bed contacts, trace fossil dis-
tribution, and stacking patterns. Pollen and foraminifer assem-
blages were determined from the interbedded shales for the
174 Lambiase et al.

interpretation of age and depositional environment, respectively,
and the sandstones were analyzed petrographically to evaluate
gross composition.
Sedimentary Facies
The outcrops comprise three facies, which are designated
as thick sandstones, thin interbedded sandstones and shale,
and shale.
Thick Sandstones
Thick sandstones generally occur as amalgamated beds,
although a few are isolated between thin shales. Amalgam-
ated units generally consist of 4–25 sandstone beds with a total
thickness of 6–33 m (Fig. 3A). Within the amalgamated units,
individual beds generally thicken and coarsen upward. Their
basal contacts are straight and sharp, and relatively few are
scoured. Flute casts, although not common, indicate a northerly
paleoflow direction when corrected for structural dip (Fig. 3B).
Load structures are moderately common.
Within each unit the lower sandstone beds are relatively thin
(0.3–1.0 m) and generally grade upward from coarse or very
coarse sand to fine sand, laminated silt, and finally shale. Most
beds are massive at the base and pass upward into parallel lami-
nation (Fig. 3C), although some have convoluted bedding that
passes upward into parallel lamination. Sandstone beds 1–6 m
thick, rarely up to 10 m thick, occur near the top of the amalgam-
ated units, although Hutchison (2005) reports that 10-m-thick
beds are more common in a few other outcrops. The thick sand-
stones generally are coarse to very coarse sand, massive, and

Figure 3. (A) Amalgamated sandstones in the Jalan Salaiman outcrop. (B) Flute casts at the base of a bed in the Maju outcrop. (C) Typical West
Crocker formation turbidite sandstone, with a scoured base and a sharp upper contact with the overlying shale. The sandstone is massive, with
very coarse sand and granules near the base, and grades into parallel laminated, medium sand. The unit comprises Bouma divisions Ta, Tb, Td,
and Te. (D) Climbing ripples in the siltstone cap of a turbidite sandstone.
 West Crocker formation of northwest Borneo
structureless. Grain size distribution is uniform throughout but
grades into medium sand a few centimeters from the top of the
bed. Parallel lamination and, less commonly, contorted lamina-
tion occur locally. The uppermost parts of the massive sand bod-
ies locally have thin laminations of carbonaceous material and
are rippled or wavy-bedded.
Siltstones and shales that cap the sandstones are usually lam-
inated and contain abundant carbonaceous material; isolated rip-
ple cross-stratification and climbing ripples occur locally in thin
sandy interbeds (Fig. 3D). Generally, the sandstones in the amal-
gamated units have flat, straight, and sharp upper contacts; the
few that appear to have gradational upper contacts and scoured
basal contacts generally are in the lower part of the unit.
Thin Interbedded Sandstones and Shales
The interbedded sandstone and shale facies comprises
monotonous alternations of 4–8-cm-thick sandstones and shale
within a 0.5–1.5-m-thick unit. The sandstone layers generally are
fine- to medium-grained sand. However, some beds grade upward
from coarse- to medium-grained sand to parallel-laminated sand
and silt; some layers have discrete and climbing ripples or paral-
lel lamination, commonly with a high content of carbonaceous
material. Most of the sandstones have sharp basal and top con-
tacts and are capped by thin, homogeneous and laminated shale
layers, although a few have gradational contacts.
Shale
Shale units are homogeneous and range from 0.5 to 5.0 m in
thickness. The lower parts of the shale units usually contain thin
interbeds of fine to silty sand that exhibit straight, sharp contacts
and ripple cross-lamination.
Trace Fossils and Microfossils
Microfossils
Microfossil analysis yielded minimal information about
the age and depositional setting of the West Crocker forma-
tion because the shale samples contain only a few agglutinating
foraminifera (1–6 individuals per sample). All are either Haplo-
phragmoides spp. or Recurvoides spp., except for one possible
Trochammina sp., that are normally indicative of a dysaerobic,
deep-water setting and an undifferentiated Tertiary age.
Pollen recovery generally was poor, with many of the paly-
nomorphs consisting of reworked Cretaceous forms or those with
an undifferentiated Tertiary age such as Dicolpopollis spp. How-
ever, the presence of Inaperturopollenites spp., Liquidambar
spp., Tiliaepollenites spp., and several Palmae taxa suggests an
early Eocene to early Oligocene age.
Trace Fossils
A relatively diverse assemblage of trace fossils occurs in the
sandstones of the three outcrops, but the number of beds with
visible traces is low, and each locality tends to have a low diver-
sity, and some are even monospecific. The traces that have been
175
identified are Ophiomorpha, Thalassinoides, Terebellina, Plano-
lites, Paleophycus that are 0.3–0.5 cm in diameter, Asterosoma,
and Rhizocorallium. Ophiomorpha range from 1.0 to 2.5 cm in
diameter and are up to 1.3 m long. Tajul Anuar Jamaluddin (1989)
identified a similar trace fossil assemblage, plus Crossopodia, in
other West Crocker formation outcrops.
Trace fossils are rare within most of the shales and are mostly
Helminthoida and Megagrapton of the deep-water Nereites ich-
nofacies (Pemberton et al., 1992) as well as Planolites (?) that are
0.3–0.5 cm in diameter. However, the apparent low abundance
of trace fossils in the finer grained facies is clearly a product of
intense, tropical surface weathering. A recently excavated area of
the Jalan Salaiman outcrop revealed several bedding plane sur-
faces covered with abundant, high relief Megagrapton, Nereites,
and Ophimorpha rudis traces that are indicative of bathyal water
depths (Fig. 4; Pemberton et al., 1992). When the site was revis-
ited 6 weeks later, the traces were barely visible.
Petrography
The modal composition of 36 sandstones was analyzed in
thin section. They generally contain 40%–60% quartz, 15%–25%
feldspar, and 10%–15% clay matrix (Fig. 5A). The feldspar frac-
tion is mostly microcline and orthoclase, with a few occurrences
of plagioclase. Lithic rock fragments account for 5%–20% of
most samples but reach nearly 30% in a few examples. Thus, the
sandstones are mostly subarkoses with a few sublitharenites and
litharenites (Fig. 5B; Pettijohn et al., 1987). Twenty of the samples
plot within the field that defines a continental block as the source
of sediment, whereas the other 16 appear to have been derived
from a recycled orogen (Fig. 5B; Dickinson and Suczek, 1979).
The overall distribution is consistent with sandstones derived
from older sediment that was the first-cycle product of an eroded
orogenic belt, in this case almost certainly the Rajang Group.
The sandstones are texturally immature, as all samples are
poorly sorted and most grains are angular to subangular, although
subrounded grains are moderately common. The matrix is made
of very fine grains and clay minerals; much of the clay fraction
comprises authigenic clays formed during diagenesis or surface
weathering. Similarly, most of the feldspars in the framework con-
stituents have been replaced by sericite, kaolinite, muscovite, and
a few unidentifiable authigenic clay minerals. The petrographic
analysis indicates that the sandstones were originally very imma-
ture and contained a considerable amount of feldspar and deposi-
tional matrix before diagenesis and surface weathering.
Stratigraphy
Generally, the three sedimentary facies stack into coarsening-
upward parasequences that range in thickness from 25 to 65 m.
The basal shale in each parasequence is overlain by interbedded
sand and shale, followed by an amalgamated sandstone unit at the
top (Fig. 6). The sandstone beds within the amalgamated units
thicken and coarsen upward, which is a typical progradational
176
Figure 4. Trace fossils in the shale beds of the West Crocker formation
belong to the bathyal Nereites ichnofacies (Pemberton et al., 1992). They
include (A) Megagrapton, (B) Nereites, and (C) Ophiomorpha rudis.
Lambiase et al.
Figure 5. (A) Thin section of a typical poorly sorted, texturally imma-
ture West Crocker formation sandstone with some component grains
labeled as quartz (Q), polycrystalline quartz (PQ), microcline (M), and
feldspar (F). (B) Gross composition of selected West Crocker forma-
tion sandstones, based on the classification scheme of Pettijohn et al.
(1987). The provenance fields for recycled orogens and continental
blocks are from Dickinson and Suczek (1979).
stacking pattern (Ricci-Lucchi and Valmori, 1980; Normark,
1978; Macdonald, 1986; Walker, 1992).
Bukit Melinsung is the only outcrop with a vertical succes-
sion that is relatively undeformed and thick enough to discern
stratigraphic trends among parasequences. There, the lower four
parasequences thicken upward, with a corresponding increase in
the proportion of sandstone, indicating that they comprise a sin-
gle progradational parasequence set (Fig. 7). The parasequence
set is incomplete, because a relatively large fault forms its upper
boundary. The upper four parasequences are much sandier than
the lower four. However, they also generally thicken and coarsen
upward and make up part of another progradational parasequence
set (Fig. 7).
 West Crocker formation of northwest Borneo
Figure 6. Facies stacking pattern of a parasequence in the Jalan Salaiman
outcrop. Relative grain size is schematic and not to scale.
STRUCTURAL STYLE
All of the West Crocker formation outcrops show some
degree of structural deformation. Most are cut by at least sev-
eral, mostly thrust, faults and are folded to some degree but
are not intensely deformed. However, the Maju outcrop cov-
ers a large area and exposes the temporal relationships among
the various structures and thus offers a unique opportunity to
analyze the structural development of the West Crocker forma-
tion. The outcrop consists of three stratigraphic successions; two
are separated by a thrust fault across which the offset cannot be
determined because it is more than the exposed length of the
fault plane, and the other boundary is stratigraphic (Fig. 8). The
oldest of the three successions forms the hanging-wall of the
177
Figure 7. Generalized sequence stratigraphy
of the Bukit Melinsung outcrop, revised from
Mohd. Khalid Jamiran (1999). Note that the
thicker sandstones are units composed of stacked
beds that thicken and coarsen upward within the
unit, whereas individual beds fine upward. Sev-
eral faults and folds within the succession are not
shown; a relatively large thrust fault forms the
upper boundary of the lower parasequence set.
Relative grain size is schematic and not to scale.
Figure 8. Schematic structural cross-section of the Maju outcrop, showing the major stratigraphic subdivisions. Note that the hanging-wall
of the major thrust fault is more highly deformed than the footwall. Locations of the photographs in Figures 9 and 10 are as indicated.
After Tan (2005).
 West Crocker formation of northwest Borneo 179

large thrust fault, and the younger two successions are in the
footwall. The hanging-wall succession has undergone at least
three phases of deformation, of which the older two are com-
pressional and the third extensional, whereas the footwall suc-
cessions were affected only by the second compressional event
and the extensional phase.

Hanging-Wall Succession

Approximately 70 m of the hanging-wall succession is

exposed laterally in near-dip and near-strike surfaces that are a
total of 175 m long and range from 8 to 12 m high. The succes-
sion includes all three of the sedimentary facies described previ-
ously. The first phase of deformation was generated by NE-SW
compression that created two thrust faults that verge southwest
and a relatively large fold that was later refolded during a sub-
sequent NW-SE compressional event. The refolded fold is a key
structure for determining the relative timing of the deformational
events and is bounded by faults to the west and south. It is a
tight recumbent, refolded fold with two fold axes, the older of
which reflects the first, NE-SW compressional event (Fig. 9A).
The rock mass in the core of the tight fold comprises thick, dis-
rupted, overturned sandstone beds, and the remainder is brec-
cia. Structures formed during the first compressional event are
restricted to the hanging-wall succession, suggesting that the
deformation took place before deposition of the footwall suc-
cessions and that these structures were synsedimentary at the
formation time scale.
The second, NW-SE phase of compression generated the
second fold axis in the refolded fold and a series of high-angle,
NW-verging thrust faults that cut across the folded beds, plus
a number of other thrust faults and folds. Parasitic folds are
commonly associated with the larger folds. Generally, thicker,
competent sandstone beds are gently folded, whereas shale-
rich intervals have well-developed detachment folds, indicat- Figure 9. Structural features in the hanging-wall and lower footwall
ing a close relationship between lithology and fold geometry successions of the Maju outcrop. See Figure 8 for locations. (A) Part
that is pervasive in the outcrop. Angular, 5–30-cm-diameter of the large refolded fold in the hanging-wall. The apparent anticline
sandstone clasts within a sheared shale matrix form an inco- on the left side of the figure is actually an overturned syncline; the
hesive, foliated fault breccia on some of the faults. Part of the younging direction is toward the bottom of the photo. (B) Highly de-
formed succession in the hanging-wall that is interpreted as a slump;
hanging-wall succession is characterized by complex folds and the younging direction is variable and locally indeterminable. Some of
sandstone blocks within sheared shale that are interpreted as a the sandstones are outlined with a white line for easier visualization
slump (Fig. 9B). Several small-scale (1–5 m), isolated folds are of the structure. (C) Parasitic folds in a shale bed above gently folded
within the slump. sandstones that form an anticline in the lower footwall succession. The
A few minor normal faults with minimal displacement were younging direction is toward the top of the photo.
generated by the third and final phase of deformation. All the
young normal faults across the entire outcrop have small dis-
placements and are viewed as the product of postorogenic gravi- surface. All three sedimentary facies are present within the succes-
tational collapse after compression rather than the reflection of a sion that was deformed by the second, NW-SE–oriented compres-
significant, extensional tectonic event (Tan, 2005). sional event and later extensional faulting. Structures generated
by NW-SE compression dominate the outcrop and especially the
Lower Footwall Succession footwall successions. Indeed, the entire outcrop can be interpreted
as part of the southeast limb of a large, disharmonic anticline with
The lower footwall succession is 75 m thick and covers an three other folds representing relatively large-scale parasitic folds
area 8 to 10 m high and ~300 m laterally on a near-strike exposure on the anticline (Tan, 2005).
180 Lambiase et al.

The lower footwall succession is a gently folded anticline

within a thick sandstone-dominated succession. There are a num-
ber of parasitic folds that are restricted to shale-rich intervals up
to 7 m thick, and a series of thrust faults that cut the succession
(Fig. 9C). The thrust faults are more abundant in the NW limb
than in the SE limb; this may indicate that the compressional
stress was NW to SE directed, which coincides with the South
China Sea spreading direction (Tan and Lamy, 1990). This sug-
gests that at least some of the deformation in the West Crocker
formation may be the product of regional tectonics, and further
implies that the West Crocker formation may be age-equivalent
to the early Oligocene to early Miocene (32–16 Ma) South China
Sea spreading (Briais et al., 1993). Some of the thrust faults
exhibit drag folding, and others preferentially follow shale-rich
bedding planes. The extensional deformation phase is repre-
sented by several minor normal faults that cut the succession.

Upper Footwall Succession

The ~275 m of upper footwall succession strata also includes

all three sedimentary facies and is exposed in near-strike and
near-dip surfaces that total ~550 m in length and are 8 to 35 m
high. The basal unit is a thick sandstone that directly overlies
the gently folded anticline of the lower footwall succession. The
basal unit also has been folded, but the sandstone beds pinch
out against the flank of the underlying fold, creating an onlap
geometry with a dip difference of 11° across the onlap surface
(Fig. 10A). The onlap angle far exceeds depositional dip in simi-
lar turbidite systems (Kumar and Slatt, 1984), which indicates
that the underlying fold had developed, or was growing, when
the overlying sandstones were deposited, and is clear evidence
of synsedimentary deformation. Dips of up to 12° or 13° are
common on growing structures in unconsolidated sediments in
modern turbidite systems (Morley, 2007), so there is no need to
invoke burial and/or consolidation before folding of the underly-
ing strata, or a significant amount of time, between folding and
deposition of the onlapping sandstones.
The lower part of the upper footwall succession dips con-
sistently at 52°–60° and is cut by numerous thrust faults that
were generated by the NW-SE compressional event, some of
which form a complex fault zone. Near the top of the succession
is a large-scale fold with a NW-SE–striking fold axis in which
dips increase upsection to the point at which the beds eventu-
ally become overturned. The folding caused a distinct thickness
change in a shale unit that is detached from an overlying sandstone Figure 10. Structural features in the upper footwall
bed in which there is no thickness change; this demonstrates the succession of the Maju outcrop. See Figure 8 for lo-
competency of the sandstones relative to the shales (Fig. 10B). cations. (A) Onlap relationship between the upper and
There are also associated small-scale detachment folds within an lower footwall successions. A white line highlights the
overturned thin interbedded sandstone and shale unit (Fig. 10C). onlap surface. (B) Abrupt lateral change in shale thick-
ness adjacent to a fold detachment, as indicated by the
The fold is cut by several thrust faults, some of which are east white arrows. The younging direction is to the right of
verging and some of which have associated breccia. A number of the photograph. (C) Small-scale detachment fold in the
normal faults that represent the final, extensional phase of defor- overturned beds associated with a larger scale anticline.
mation cut the succession and the older structures. The arrow indicates the younging direction.
 West Crocker formation of northwest Borneo
TECTONO-STRATIGRAPHIC MODEL
Depositional Model
The sedimentary structures found in the West Crocker for-
mation indicate that the sandstones were deposited by turbid-
ity currents, as noted in previous studies (e.g., Stauffer, 1967;
Tongkul, 1990). Most beds contain partial Bouma sequences
with divisions Ta, Tb, and Td; Ta and Td; Ta and Tc; Ta and
Te; or just Ta. These Ta–Tc dominant, massive, structureless
but graded amalgamated sandstone beds with sharp and straight
upper and lower contacts are typical of high-density turbidites
(e.g., Stow et al., 1996; Lowe, 1982; Walker, 1992; Stow and
Johansson, 2000). The generally structureless and massive tur-
bidite sandstones indicate rapid deposition, probably by col-
lapse fallout from high-density turbidity currents (Stow and
Johansson, 2000) caused by an abrupt decrease in slope angle
along the transport path. Parallel lamination and climbing rip-
ples near the top of the sandstones and in the shale caps are the
consequence of rapid and continuous fallout of finer grains dur-
ing the final stages of deposition. The sharp bed contacts are the
product of sediment bypassing and sequential sedimentation of
different particle populations from turbulent high-density flows
transporting very coarse to fine grains in suspension (Lowe,
1982; Stow et al., 1996).
The very immature texture, angularity, and poor sorting of
the sandstones suggest a nearby sediment source and a short
transport distance, as does the presence of angular to subangular
feldspars that are easily destroyed by extensive transport (Caro-
zzi, 1993; Harwood, 1988). Many of the trace fossils belong to
the Cruziana ichnofacies that is indicative of a shelfal environ-
ment with water depths normally <200 m (Pemberton et al.,
1992; Frey and Pemberton, 1984). However, those in the shale,
Ophiomorpha rudis, Nereites, Helminthoida, and Megagrapton,
are deep-water traces of the Nereites ichnofacies (Pemberton et
al., 1992). In such cases, the assemblage in the shales is the most
reliable water depth indicator, confirming that the West Crocker
turbidites probably were deposited in >1000 m of water (G. Pem-
berton, 2004, personal commun.).
All the sedimentary characteristics are most consistent
with deposition on a slope apron (Macdonald, 1993; Reading
and Richards, 1994). Sands were deposited rapidly at the toe of
the slope, where they formed small-scale lobes; the relatively
thin sandstone beds (nearly all are <6 m, and most are <2 m
thick) suggest that the depositional lobes were not large but
had radii in the range of 1–5 km or less (Stow and Johansson,
2000). This results in a surface slope on the lobes of <1° even
for the thickest parasequence of 65 m, which corresponds well
with other sand-rich slope aprons, e.g., the middle Tonkawa
Sandstone of the Anadarko Basin in the U.S. Midcontinent
(Kumar and Slatt, 1984). The coarsening and thickening
upward nature of most of the parasequences suggests that the
lobes were progradational.
181
Age and Tectonic Setting
Age of the West Crocker Formation
The age of the West Crocker formation is poorly constrained
because few age-diagnostic microfossils have been recovered.
Many workers assign an Oligocene to early Miocene age to the
formation (e.g., Hutchison et al., 2000; Hutchison, 2004; Tong-
kul, 1990, 1991), although some acknowledge that it may be as
old as Eocene (e.g., Rangin et al., 1990; van Hattum et al., 2003).
Pollen data from this study suggest that the three outcrops are
Eocene deposits, although the age of the entire West Crocker for-
mation could be variable.
A preliminary palynological analysis, using identical ana-
lytical techniques, also indicates an Ecoene age for the uncon-
formably overlying Melingan Formation in Sabah (Fig. 2). It is
important to note that in Sabah the Melingan Formation consists
of steeply dipping, high-energy storm beds that are markedly
different from the gently dipping, low-energy, wave-dominated
sandstones with minor limestones in Sarawak and Brunei that
were dated as early Miocene by Wilson (1964). This suggests
that either the Melingan Formation has a broader age range than
previously recognized (Fig. 2) or that the Sabah and Sarawak-
Brunei outcrops belong to different stratigraphic units.
Assuming that the stratigraphic relationship between the
West Crocker and Melingan formations as defined by previous
workers (e.g., Liechti et al., 1960; Wilson, 1964; Tate, 1974) and
illustrated in Figure 2 is accurate, then the West Crocker forma-
tion probably is Eocene to Oligocene rather than Oligocene to
lower Miocene, and the stratigraphy should be revised as sug-
gested in Figure 2. This view indicates that most of the West
Crocker formation is not a sandy facies equivalent of the shaly
Temburong Formation, which resolves the problematic paleo-
geographic facies distribution whereby the sandy fans of the West
Crocker formation lie N-NE of the Temburong Formation shales
in a more distal position.
Tectonic Setting
Several aspects of the structural evolution of the Maju out-
crop indicate that at least part of the West Crocker formation
underwent synsedimentary deformation, which differs from the
traditional view that all the deformation is postdepositional (e.g.,
Tongkul, 1990, 1991; Hutchison, 1996, 2005). The foremost
aspect is that two different phases of deformation affected the
hanging-wall succession, whereas only one of them deformed
the footwall. The onlap relationship between the lower and upper
footwall successions (Fig. 8) not only confirms synsedimentary
deformation but suggests that compressional tectonics was ongo-
ing while most, if not all, of the units were deposited. Therefore,
the compressional deformation of the deep-water turbidites is
older than the uplift of the West Crocker formation.
It is generally accepted that early Tertiary compression
was related to subduction along the northwest Borneo margin,
coupled with seafloor spreading related to the opening of the
182 Lambiase et al.
South China Sea (e.g., Tongkul, 1990; Hutchison, 2005). In
Tongkul’s (1990) model, NW-SE compression was generated
by SE-directed underthrusting of the Dangerous Grounds dur-
ing late Oligocene (ca. 25 Ma) subduction. Both phases of com-
pressional deformation must have been completed by the time
that subduction ceased; many workers think that this occurred
in the middle Oligocene (ca. 30 Ma; e.g., Tongkul, 1990),
although some believe that it persisted to the early middle Mio-
cene (Hutchison, 1996). Structures generated by NW-SE com-
pression are abundant throughout the West Cocker formation
(Stauffer, 1967; Tongkul, 1990, 1991; Hutchison, 1996, 2005),
but the older, NE-SW compressional phase has not been docu-
mented previously.
The compressional tectonic setting, synsedimentary defor-
mation, and slope apron depositional model suggest that the West
Crocker formation is an accretionary prism. Sediments were sup-
plied episodically from uplifted Rajang Group strata in the active
orogen and transported down a relatively steep slope to the adja-
cent basin. They accumulated at the base of the slope in a series
of small fans that were deformed by at least two phases of synde-
formational, compressional folding and thrust faulting.
DISCUSSION
The probable Eocene accretionary prism interpretation for
the West Crocker formation not only provides the best reconcili-
ation of the structural data and sedimentological data collected
in the present study, but it overcomes many of the problems and
inconsistencies of previous interpretations. One of the impor-
tant sedimentological issues is the size and type of fan system
that the West Crocker formation represents. Many workers inter-
pret the West Crocker formation as a large submarine fan that
was fed from a point source to the SW, relatively far from the
West Crocker depocenter (e.g., Stauffer, 1967; Tongkul, 1987;
Hutchison, 1996). This argument is partially based on paleogeo-
graphic reconstructions that suggest that marine waters covered
present-day Sabah prior to the uplift that produced the Deep
Regional Unconformity (e.g., Hall, 2002), thereby precluding
a sediment source to the east. However, the thin sandstone beds
and small lobes of the West Crocker formation are too small
for a fan associated with a relatively large river (Reading and
Richards, 1994).
Another important issue is the provenance area for the West
Crocker formation sediment. Van Hattum et al. (2003) concluded
that the textural immaturity of the West Crocker formation reflects
a sediment source from nearby, first-cycle orogenic sandstones,
possibly the uplifted Rajang Group in Sarawak. These precursor
sandstones probably were the erosional products of granites in
the Schwaner Mountains in southern Borneo (Fig. 1A). Van Hat-
tum et al. (2003) also attribute the subordinate, ophiolite-derived
material in the West Crocker formation (Stauffer, 1967) to a sec-
ondary provenance area to the east.
The samples analyzed in the present study contain consider-
ably more feldspar (15%–25% as opposed to 0%–13%) and more
lithic fragments than van Hattum et al. (2003) report. This sup-
ports their observation that the West Crocker formation is com-
positionally less mature than the rest of the Crocker Formation
and also suggests that an easterly provenance area was relatively
more important. Similarly, William et al. (2003) conclude that
the West Crocker formation was derived from nearby, uplifted
Rajang Group sediments because of the composition and imma-
ture texture, although they prefer an easterly provenance area
based on petrographic analysis, regional sedimentary systems,
and stratigraphic relationships.
Synsedimentary deformation is more compatible with an
Eocene rather than with an Oligocene to early Miocene age for
the West Crocker formation. The younger age would require that
strong compression, and probably subduction, continued into the
early Miocene (ca. 20 Ma), whereas many workers (e.g., Tong-
kul, 1990) contend that subduction ended in the middle Oligo-
cene (ca. 30 Ma). An Eocene age also provides ample time for
the estimated 4–8 km of burial that the West Crocker under-
went before it was uplifted and exposed in the middle Miocene
(Hutchison, 2005); the younger age necessitates that both the
burial and uplift occurred in the early middle Miocene and were,
therefore, extremely rapid.
The West Crocker formation lies below a regional middle
Miocene unconformity beneath which the Oligocene to early
Miocene geology is characterized by extension in the Natuna Sea
(Darman and Sidi, 2000), Sarawak (Mat-Zin, 1997; Madon, 1999),
and the Dangerous Grounds off the shore of western Sabah (Mil-
som et al., 1997). In offshore western Sabah there is pre–middle
Miocene extension, and many late Miocene anticlines cored with
middle Miocene sedimentary rocks are interpreted as an inversion
of older structures (Bol and van Hoorn, 1980; Madon et al., 1999).
The problem posed by development of a major thrust belt syn-
chronous with regional extension has been answered previously
by having extension in the South China Sea induce subduction
beneath Borneo, thereby causing the Sabah orogeny (Hutchison,
1996; Hall, 2002). An Eocene age for the West Crocker forma-
tion eliminates this problem and calls into question models for the
region’s early Neogene tectonic evolution.
CONCLUSIONS
An integrated analysis of the structural geology and sedimen-
tology of the West Crocker formation concludes that the abun-
dant sandstone beds were deposited from high-density turbidity
currents and formed small, prograding lobes on a deep-water
slope apron. The composition and immature texture of the sands
necessitates a short transport distance, suggesting that uplifted
Rajang Group sediment was the source of the West Crocker
formation. Micropaleontological data and stratigraphic relation-
ships indicate that at least this part of the West Crocker formation
is Eocene rather than Oligocene to early Miocene. Nearly all of
the structural deformation is compressional and synsedimentary;
the West Crocker formation is a syntectonic deposit, probably an
accretionary prism.
 West Crocker formation of northwest Borneo
ACKNOWLEDGMENTS
The authors wish to thank George Pemberton for assistance with
trace fossil identifications and Nurhayati and Rhodora Tiamsing
for help with drafting. Mohd. Khalid Jamiran contributed to the
interpretation of stratigraphic architecture, and the manuscript
benefited greatly from discussions with Stefan Back and com-
ments by reviewers Charles Hutchison, Peter Clift, and an anony-
mous reviewer. Amelia William and Tan Yaw Tzong wish to thank
ExxonMobil Exploration and Production Malaysia and Universiti
Brunei Darussalam, respectively, for financial support.
REFERENCES CITED
Bol, A.J., and van Hoorn, B., 1980, Structural styles in western Sabah offshore:
Geological Society of Malaysia Bulletin, v. 12, p. 1–16.
Briais, A., Patriat, P., and Tapponnier, P., 1993, Updated interpretation of mag-
netic anomalies and seafloor spreading stages in the South China Sea:
Implications for the Tertiary tectonics of Southeast Asia: Journal of Geo-
physical Research, v. 98, p. 6299–6328.
Carozzi, A.V., 1993, Sedimentary Petrography: Englewood Cliffs, NJ, Prentice
Hall, 263 p.
Crevello, P.D., 2002, The great Crocker submarine fan: A world class foredeep
turbidite system: Jakarta, 28th Indonesia Petroleum Association Proceed-
ings, v. 1, p. 377–407.
Darman, H., and Sidi, F.H., 2000, An Outline of the Geology of Indonesia:
Jakarta, Indonesian Association of Geologists, 192 p.
Dickinson, W.R., and Suczek, C.A., 1979, Plate setting and sandstone com-
position: American Association of Petroleum Geologists Bulletin, v. 63,
p. 2164–2182.
Frey, R.W., and Pemberton, S.G., 1984, Trace fossil facies models, in Walker,
R.G., ed., Facies Models (2nd edition): Geological Association of Can-
ada, p. 189–207.
Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and
the SW Pacific: Computer-based reconstructions, model and animations:
Journal of Asian Earth Sciences, v. 20, p. 353–431, doi: 10.1016/S1367-
9120(01)00069-4.
Harwood, G., 1988, Microscopic techniques: II. Principles of sedimentary
petrography, in Tucker, M., ed., Techniques in Sedimentology: Oxford,
UK, Blackwell Science, p. 108–173.
Hutchison, C.S., 1996, Geological evolution of South East Asia: Kuala Lumpur,
Geological Society of Malaysia, 350 p.
Hutchison, C.S., 2004, Marginal basin evolution: The southern South China
Sea: Marine and Petroleum Geology, v. 21, p. 1129–1148, doi: 10.1016/
j.marpetgeo.2004.07.002.
Hutchison, C.S., 2005, Geology of Northwest Borneo: Amsterdam, Elsevier,
421 p.
Hutchison, C.S., Bergman, S.C., Swauger, D.A., and Graves, J.E., 2000, A
Miocene collisional belt in north Borneo: Uplift mechanism and isostatic
adjustment quantified by thermochronology: Geological Society [Lon-
don] Journal, v. 157, p. 783–793.
Kumar, N., and Slatt, R.M., 1984, Submarine-fan and slope facies of Tonkawa
(Missourian–Virgilian) sandstone in deep Anadarko Basin: American
Association of Petroleum Geologists Bulletin, v. 68, p. 1839–1856.
Levell, B.K., 1987, The nature and significance of regional unconformities in
the hydrocarbon-bearing Neogene sequences offshore West Sabah: Geo-
logical Society of Malaysia Bulletin, v. 21, p. 55–90.
Liechti, P., Roe, F.N., Haile, N.S., and Kirk, H.J.C., 1960, The Geology of Sar-
awak, Brunei and the Western Part of North Borneo: British Borneo Geo-
logical Survey Bulletin, v. 3, 360 p.
Lowe, D.R., 1982, Sediment gravity flows II. Depositional models with spe-
cial reference to the deposits of high-density turbidity currents: Journal of
Sedimentary Petrology, v. 52, p. 279–297.
Macdonald, D.I.M., 1986, Proximal to distal sedimentological variation in a
linear turbidite trough: Implications for the fan model: Sedimentology,
v. 33, p. 243–259, doi: 10.1111/j.1365-3091.1986.tb00534.x.
183
Macdonald, D.I.M., 1993, Controls on sedimentation at convergent plate mar-
gins: International Association of Sedimentologists Special Publication
20, p. 225–257.
Madon, M., 1999, Geological setting of Sarawak, in Petroleum Geology and
Resources of Malaysia: Kuala Lumpur, Petronas, p. 273–290.
Madon, M., and Abolins, P., 1999, Balingian Province, in Petroleum Geology
and Resources of Malaysia: Kuala Lumpur, Petronas, p. 343–367.
Madon, M., Leong, K.M., and Anuar, A., 1999, Sabah Basin, in Petroleum Geol-
ogy and Resources of Malaysia: Kuala Lumpur, Petronas, p. 501–542.
Mat-Zin, I.C., 1997, Tectonics, evolution and sedimentation of the Sarawak
Basin: Geological Society of Malaysia Bulletin, v. 41, p. 41–52.
Milsom, J., Holt, R., Ayub, D.B., and Smail, R., 1997, Gravity anomalies and
deep structural controls at the Sabah-Palawan margin, South China Sea,
in Petroleum Geology of Southeast Asia: Geological Society [London]
Special Publication 126, p. 417–427.
Mohd, Khalid Jamiran, 1999, Sedimentology and sequence stratigraphy of
Bukit Melinsung, Papar, Sabah [M.S. thesis]: Universiti Brunei Darus-
salam, 50 p.
Morley, C.K., 2007, Growth of folds in a deepwater setting: Basin Research
(in press).
Normark, W.R., 1978, Fan valleys, channels, and depositional lobes on mod-
ern submarine fans; characters for recognition of sandy turbidite envi-
ronments: American Association of Petroleum Geologists Bulletin, v. 62,
p. 912–931.
Pemberton, S.G., MacEachern, J.A., and Frey, R.W., 1992, Trace fossil facies
models: Environmental and allostratigraphic significance, in Walker, R.G.,
and James, N.P., eds., Facies Models: Response to Sea Level Change,
Geotext 1: Geological Association of Canada, p. 47–72.
Pettijohn, F.J., Potter, P.E., and Siever, R., 1987, Sand and Sandstone (2nd edi-
tion): New York, Springer-Verlag, 571 p.
Rangin, C., Bellon, H., Bernard, F., Letouzey, J., Muller, C., and Sanudin, T., 1990,
Neogene arc-continent collision in Sabah, Northern Borneo, Malaysia:
Tectonophysics, v. 183, p. 305–319, doi: 10.1016/0040-1951(90)90423-6.
Reading, H.G., and Richards, M., 1994, Turbidite systems in deep-water basin
margins classified by grain size and feeder system: American Association
of Petroleum Geologists Bulletin, v. 78, p. 792–822.
Ricci Lucchi, F., and Valmori, E., 1980, Basin-wide turbidites in a Miocene,
over-supplied deep-sea plain: A geometrical analysis: Sedimentology,
v. 27, p. 241–270, doi: 10.1111/j.1365-3091.1980.tb01177.x.
Sandal, S.T., ed., 1996, The Geology and Hydrocarbon Resources of Negara
Brunei Darussalam: Bandar Seri Begawan, Syabas, 243 p.
Stauffer, P.H., 1967, Studies in the Crocker Formation, Sabah, in Stauffer, P.H.,
ed., Geological Papers 1966: Geological Survey of the Borneo Region of
Malaysia, Bulletin 8, p. 1–13.
Stow, D.A.V., and Johansson, M., 2000, Deep-water massive sands: Nature, ori-
gin and hydrocarbon implications: Marine and Petroleum Geology, v. 17,
p. 145–174, doi: 10.1016/S0264-8172(99)00051-3.
Stow, D.A.V., Reading, H.G., and Collinson, J.D., 1996, Deep seas, in Reading,
H.G., ed., Sedimentary Environments and Facies (3rd edition): Oxford,
UK, Blackwell Science, p. 395–453.
Tajul Anuar Jamaluddin, 1989, Struktur sedimen dalam Formasi Crocker di
kawasan Tamparuli, Sabah: Geological Society of Malaysia Bulletin,
v. 24, p. 135–157.
Tan, D.N.K., and Lamy, J.M., 1990, Tectonic evolution of the NW Sabah con-
tinental margin since the Late Eocene: Geological Society of Malaysia
Bulletin, v. 27, p. 241–260.
Tan, Y.T., 2005, Structural geology of the Maju outcrop, West Crocker For-
mation, Kota Kinabalu, Sabah, Malaysia [M.S. thesis]: Universiti Brunei
Darussalam, 72 p.
Tate, R.B., 1974, Palaeo-environmental studies in Brunei: Brunei Museum
Journal, v. 3, p. 285–305.
Tongkul, F., 1987, The sedimentology and structure of the West Crocker For-
mation in the Kota Kinabalu area, Sabah, in GEOSEA IV Proceedings:
Jakarta, Indonesia Association of Geologists, p. 135–156.
Tongkul, F., 1990, Structural style and tectonics of Western and Northern Sabah:
Geological Society of Malaysia Bulletin, v. 27, p. 227–239.
Tongkul, F., 1991, Tectonic evolution of Sabah, Malaysia: Journal of South-
east Asian Earth Sciences, v. 6, p. 395–405, doi: 10.1016/0743-
9547(91)90084-B.
van Hattum, M.W.A., Hall, R., and Nichols, G.J., 2003, Provenance of northern
Borneo sediments: Jakarta, 29th Indonesia Petroleum Association Pro-
ceedings, v. 1, paper IPA03-G-016.
184 Lambiase et al.

Walker, R.G., 1992, Turbidites and submarine fans, in Walker, R.G., and James, Wilson, R.A.M., 1964, The geology and mineral resources of the Labuan and
N.P., eds., Facies Models: Response to Sea Level Change, Geotext 1: Padas Valley area, Sabah, Malaysia: Geological Survey of the Borneo
Geological Association of Canada, p. 239–263. Region of Malaysia, Memoir 17.
William, A.G., Lambiase, J.J., Back, S., and Mohd, Khalid Jamiran, 2003, Sedi-
mentology of the Jalan Salaiman and Bukit Melinsung outcrops, western
Sabah: Is the Crocker Formation an analogue for Neogene turbidites off-
shore?: Geological Society of Malaysia Bulletin, v. 47, p. 63–75. MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007

Printed in the USA

 The Geological Society of America
Special Paper 436
2008

Temporal changes in the composition of Miocene sandstone related to collision between

the Honshu and Izu Arcs, central Japan

Koichi Okuzawa*
Research Center for Deep Geological Environments, National Institute of Advanced Industrial Science and Technology,
Tsukuba 305-8567, Japan

Ken-ichiro Hisada
Graduate School of Life and Environmental Sciences, University of Tsukuba, Tsukuba 305-8572, Japan

ABSTRACT

The Izu Arc has been colliding with the Honshu Arc in central Japan since
ca. 15 Ma. In order to understand the provenance changes related to this collision,
we studied lower to middle Miocene sandstones in and around the collision zone by
analyzing their framework composition and the chemistries of detrital clinopyroxene,
garnet, and chromian spinel.
Sandstone deposited in the trench and forearc basin of the Honshu Arc prior
to collision includes grains of detrital garnet and chromian spinel, which originated
mainly from granites and low pressure-temperature (P-T) metamorphic rocks, and
forearc peridotite, respectively, parts of the Honshu Arc. The forearc and trench-fill
sandstones differ in terms of their framework composition; sedimentary lithics are
more abundant in the forearc sandstone than in the trench. The two groups of sedi-
ments were supplied from different parts of the Honshu Arc.
The lower part of the clastic sequence deposited within the Izu Arc is composed
mainly of volcaniclastic rocks and yields detrital clinopyroxenes that originated from
the Izu Arc. In contrast, the upper part is similar to the lower Miocene trench-fill
deposits in terms of its framework composition and the chemistry of detrital garnet
and chromian spinel. This reflects a change in provenance triggered by the initial con-
tact of the Izu Arc and the trench between the Eurasian and Philippine Sea plates.
The lower part of the middle Miocene trench-fill that was deposited following
initial contact is also similar to the lower Miocene trench-fill. The upper part, how-
ever, resembles lower Miocene sedimentary rocks of the forearc basin. This suggests
that the transport path was changed by collision. During the initial stages of collision
between the Honshu and Izu Arcs, the Honshu Arc was preferentially uplifted, and
therefore supplied most of the detritus to the collision zone.

Keywords: sandstone, provenance, framework composition, detrital, clinopyroxene,

garnet, chromian spinel, Honshu Arc, Izu Arc, collision.

*Present address: Institute for Geo-Resources and Environment, National Institute of Advanced Industrial Science and Technology, Tsukuba 305-8567, Japan;
koichi-okuzawa@aist.go.jp

Okuzawa, K., and Hisada, K., 2008, Temporal changes in the composition of Miocene sandstone related to collision between the Honshu and Izu Arcs, central
Japan, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of
America Special Paper 436, p. 185–198, doi: 10.1130/2008.2436(09). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of
America. All rights reserved.

185
186 Okuzawa and Hisada

INTRODUCTION
The Izu Arc, an oceanic island arc, has been colliding with the
Honshu Arc, a continental arc, in central Japan since ca. 15 Ma
(Fig. 1); this zone of collision is termed the Izu Collision Zone
(Taira et al., 1989). There are four main blocks of accreted Izu
Arc crust: the Koma, Misaka, Tanzawa, and Izu Blocks (Fig. 2).
Each block was accreted to the Honshu Arc at different times
since ca. 15 Ma (Niitsuma and Akiba, 1985; Amano, 1986, 1991;
Aoike, 1999). The accreted blocks are associated with thick clas-
tic sequences showing coarsening-upward trends (e.g., Ito, 1985,
1987) and regression-type facies (e.g., Huchon and Kitazato,
1984). These sequences are interpreted to have been deposited
in the trench between the Honshu Arc and the accreted blocks
(Amano, 1986, 1991; Koyama, 1991, 1993). The collision his-
tory of each block was examined in previous studies on the
basis of the lithology and geochemistry of the accreted blocks,
sequence stratigraphy, and the geochronology and deformation
history of trench-fill, as summarized by Amano et al. (1999) and
Aoike (1999).
Although the Izu Collision Zone is a suitable area for studies
that seek to understand the nature of sediments generated in arc-
arc collision settings, there are few petrographical studies of sedi-
ments from the collision zone. According to Soh (1986), sand-
stone and conglomerate from upper Miocene to Pliocene trench-
fill sedimentary rocks are composed mainly of sedimentary lithics
rather than volcanic lithics. In contrast, sandstone from the lower
Miocene accretionary complex adjacent to the collision zone is
conspicuously rich in quartz and feldspar (Watanabe and Iijima,
1990). It appears that a change in provenance occurred during
the early to middle Miocene (ca. 13–17 Ma) at approximately the
same time that the collision zone formed.
Several types of detrital heavy minerals have been examined
in various provenance studies to determine the source rocks of
the sediment. Chemical analysis of detrital heavy minerals has
proved to be a powerful tool in provenance studies (e.g., Mor-
ton, 1991). Chromian spinel, garnet, and clinopyroxene have
also been used to infer the nature of the sediment source rocks
(e.g., Hisada and Arai, 1993; Takeuchi, 1994; Cawood, 1983).
Chromian spinel generally originates from mafic to ultramafic

138°E 140°E
A B
150°
130°

140°
120°

38°N
50°
North 100 km

Eurasia American
Plate Plate

40°
Japan Sea
HG
r c
h uA Hd
ns
Ho 36°N
Pacific Sb
Izu

Plate 30°
Ch
Arc

Mn Sh Hy
Philippine Sea Ry
Inner zone

Plate
Boso Pen.
L
MT
140°

Mn
130°

Fig. 2
Izu a

Cretaceous to Tertiary accretionary complex

Outer zone

Jurassic accretionary complex

34°N
rc

Permian accretionary complex

High P/T type metamorphic rocks
Low P/T type metamorphic rocks
Circum-Izu massif serpentinite Forearc peridotite (Kurosegawa Belt)
Ophiolite
Continental block

Figure 1. Geologic outline of the Honshu Arc. (A) Plate tectonic setting of Japan. (B) Geologic map of basement rocks of the central Honshu
Arc. Modified from Isozaki (1996). MTL—Median Tectonic Line; Hd—Hida Belt; HG—Hida Gaien Belt; Mn—Mino Belt; Ry—Ryoke Belt;
Sb—Sambagawa Belt; Ch—Chichibu Belt (including Northern Chichibu, Kurosegawa, and Southern Chichibu Belts); Sh—Shimanto Belt (in-
cluding Setogawa, Kobotoke, Hayama, and Mineoka Belts); Hy—Hayama; Mn—Mineoka.
 Collision between the Honshu and Izu Arcs 187

138°E 139°E

N
Koma Group
Ko
bo
tok
eB
Study area elt

ISTL

Mt. Fuji
Belt

Mt. Hakone
Figure 2. Geologic map of the Izu Col-
anto

lision Zone. Compiled from Tsuchi

(1986), Koyama (1993), Sugiyama

Sa
Shim

(1995), and Martin and Amano (1999).

ga
35°N

mi
ISTL—Itoigawa-Shizuoka Tectonic Line.
The study area is indicated by the three

Tro
rectangular boxes.

ugh
Suruga Trough

20km

Ryuso Group Trench fill sediments

Kurami Group Setogawa Izu Block Izu
Oigawa Group Belt Tanzawa Block Collision
Setogawa Group Misaka Block Zone
+ Intrusive rocks Koma Block
Cretaceous-Oligocene

rocks; accordingly, the occurrence of detrital chromian spinel
can be used to constrain the timing of emplacement of mafic
or ultramafic rocks. The chemistry of spinel can also provide
clues to the tectonic setting in the source region (e.g., Arai,
1992; Barnes and Roeder, 2001). Garnet is commonly found
within metamorphic rocks and granite. Garnet has a compli-
cated chemistry, and the chemical composition of metamorphic
garnet depends upon both the parent rock and the metamor-
phic grade (Miyashiro, 1953). Clinopyroxene occurs as a stable
phase in almost all types of igneous rocks, as well as various
rocks that formed under conditions of both regional and contact
metamorphism (Deer et al., 1982). Although chemical analy-
sis of detrital heavy minerals is a useful tool for understanding
provenance, there has previously been no example applying it
to the sedimentary rocks in the Izu Collision Zone.
This paper describes the framework composition, mineral-
ogy, and mineral chemistry of sandstone from pre- to syncolli-
sional successions within the Izu Collision Zone. The chemistry
of detrital minerals was compared with the mineral chemistry of
potential source rocks proximal to the study area in order to iden-
tify provenance changes caused by arc-arc collision.
OUTLINE OF REGIONAL GEOLOGY
The geology of the central Honshu Arc is characterized by
a zonation of geological units (Fig. 1B). The central part of the
Honshu Arc is divided into the Inner and Outer Zones, which
are separated by the Median Tectonic Line (MTL, Fig. 1B). The
Inner Zone is composed of four belts that constitute a continen-
tal block (Hida Belt), an ophiolite (Hida Gaien Belt), a Jurassic
188 Okuzawa and Hisada

accretionary complex (Mino Belt), and low pressure-temperature Izu Collision

Setogawa Belt Zone
(P-T) metamorphic rocks (Ryoke Belt; Figure 1B; Isozaki, 1996). 10
Cretaceous to early Tertiary granites (e.g., Ryoke Granite) and western central &

Middle
felsic volcanic rocks (e.g., Nohi Rhyolite) are widely distributed part eastern
parts Mm
throughout the Inner Zone. The Outer Zone consists of a high P-T 15 Kurami Group

Miocene
metamorphic accretionary complex (Sambagawa Belt), a Juras- Ks
sic accretionary complex (Northern Chichibu Belt), an ophiolite Koma

Early
(Kurosegawa Belt), a Jurassic accretionary complex (Southern Group
20
Chichibu Belt), and a Cretaceous to Tertiary accretionary com-
plex (Shimanto, Setogawa, Kobotoke, Hayama, and Mineoka
Oigawa
Belts; Figure 1B; Isozaki, 1996). The Tertiary accretionary com- 25 Group
plex (Setogawa, Kobotoke, Hayama, and Mineoka Belts) includes

Late
Age (Ma)
a number of serpentinite bodies, which are called the Circum–Izu

Oligocene
Massif serpentinites (Arai, 1991); these four accrectionary com- 30
plex belts are also termed the Circum–Izu Massif Serpentine Belt
(Arai, 1991). These serpentinite bodies are accompanied by gab-

Early
bro and basaltic rocks of the mid-ocean-ridge basalt (MORB) 35
type and intraplate type (e.g., Ogawa and Taniguchi, 1988; Hirano

Late
et al., 2003), and are considered to have formed as crust gener-
ated by seafloor spreading in a backarc environment (Arai, 1991).

Eocene
40
Alternatively, Arai and Okada (1991) and Arai (1994) proposed

that the Circum–Izu Massif serpentinites were initially heteroge-
neous in their petrological characteristics. Arai (1994) proposed
that the incipient Circum–Izu Massif serpentinites were formed in
an arc environment. The serpentinites were later amalgmated with
gabbro and basaltic rocks (e.g., Ogawa and Taniguchi).
The study area covers parts of the Setogawa Belt and the
Izu Collision Zone (Fig. 2). An early Miocene to early middle
Miocene accretionary complex is distributed in the Setogawa
Belt and is composed of three stratigraphic units, the Setogawa,
Oigawa, and Ryuso Groups (Fig. 2). In the western part of the
Setogawa Belt, the Setogawa Group is covered by the lower Mio-
cene Kurami Group and younger sedimentary rocks, which were
deposited in a forearc basin (Fig. 2). The Izu Collision Zone com-
prises accreted Izu Arc crust and trench-fill material (Osozawa et
al., 1990; Hiroki and Matsumoto, 1999; Amano, 1991). The four
lower to middle Miocene groups were targeted in this study in
order to reconstruct the provenance changes associated with the
early stage of Honshu-Izu collision: the Setogawa, Oigawa, and
Kurami Groups of the Setogawa Belt and the Koma Group of the
Izu Collision Zone. The Ryuso Group in the Setogawa Belt was
excluded from this study because the group is composed mostly
of volcanic rocks (Sugiyama, 1995) and is inadequate for the
framework composition work.
The Setogawa Group comprises basalt, limestone, chert,
tuffaceous mudstone, and alternating sandstone and mudstone
(Fig. 3; Sugiyama, 1995). The tuffaceous mudstone contains
tectonic blocks of serpentinite, which is a component of the
Circum-Izu Massif serpentinites (Arai, 1991), intercalated with
conglomerate containing clasts of serpentinite. The Oigawa
Group consists mainly of alternating sandstone and mudstone
with minor amounts of conglomerate and sandstone (Sugiyama,
1995). The ages of the clastic rocks in the Setogawa and Oigawa
Groups are early Miocene (24–19 Ma) and early to early middle
E. Middle
50 Setogawa Group
clastic rocks tuffaceous mudstone
chert limestone volcanic rocks
Figure 3. Correlation diagram of the stratigraphic units targeted in this
study. Ks—Kushigatayama Subgroup; Mm—Momonoki Subgroup.
Compiled from Koyama (1993), Sugiyama (1995), and Hiroki and
Matsumoto (1999).
Miocene (ca. 22–15 Ma), respectively (Fig. 3; Sugiyama, 1995).
A total of seven sandstone samples were collected from the lower
Miocene accretionary complex: three from the Setogawa Group,
and four from the Oigawa Group.
The Kurami Group crops out in the SW part of the Setogawa
Belt (Fig. 2), where it unconformably overlies the Setogawa Group
(Fig. 3). The Kurami Group consists of basal conglomerate, sand-
stone, alternating sandstone and mudstone, tuffaceous mudstone,
and mudstone in ascending stratigraphic order (Ibaraki, 1986;
Hiroki and Matsumoto, 1999). The group was deposited in the
late early Miocene (18–15.5 Ma; Hiroki and Matsumoto, 1999)
and represents forearc basin sediments (Hiroki, 1995; Hiroki and
Matsumoto, 1999). Three sandstone samples were collected from
the lower part of the Kurami Group.
The Koma Group consists of two subgroups: the lower part
is the Kushigatayama Subgroup, and the upper is the Momo-
noki Subgroup (Fig. 3; Kosaka and Tsunoda, 1969), with the
Kushigatayama Subgroup being subdivided into lower and
upper parts (Koyama, 1991, 1993). The lower part of the Kushi-
gatayama Subgroup is composed mainly of tholeiitic intermedi-
ate to mafic volcaniclastic rocks and lava flows (Shimazu and
 Collision between the Honshu and Izu Arcs 189

Ishimaru, 1987). The upper part consists mainly of intermediate A Qm Setogawa Group
to mafic tuff intercalated with mudstone, sandstone, and con-
glomerate. The Momonoki Subgroup consists of alternating Oigawa Group
sandstone and mudstone, conglomerate, mudstone, siliceous CI Kurami Group
mudstone, and felsic tuff (Koyama, 1991, 1993). The sequence Kushigatayama Subgp.
shows a coarsening-upward trend (Aoike, 1999). Conglomer- TC QR Momonoki Subgp.
ate clasts within the Momonoki Subgroup are derived from
both the Honshu and Izu Arcs (Koyama, 1993). The Kushigata- Mixed
TR
yama Subgroup was deposited within the Izu Arc, whereas the Type A
Momonoki Subgroup represents trench-fill sediment (Amano, DA Type B
1991; Aoike, 1999). The Kushigatayama and Momonoki Sub- BU
groups were deposited at 16–15 Ma and 15–13.5 Ma, respec-
TA LR
tively, and are in faulted contact (Aoike, 1999). We collected
a total of nine sandstone samples from the Koma Group: three Type C UA

from the lower part of the upper Kushigatayama Subgroup, two F Lt

from the uppermost Kushigatayama Subgroup, and two each
from the lower and upper Momonoki Subgroup. B Ls Qm: monocrystalline quartz
F: total feldspar
METHODS Lt: total lithics
Ls: sedimentary lithics
Lvf: felsic volcanic lithics
We mostly collected medium-grained sandstone samples Lvm: intermediate to mafic
from the Setogawa, Oigawa, Kurami, and Koma Groups. Thin volcanic lithics
sections were stained for K-feldspar and point-counted for 500
points per sample, following the Gazzi-Dickinson method (Inger-
soll et al., 1984). Type B
Heavy liquid separation was performed to collect clinopy-
roxene, garnet, and chromian spinel from the selected 13 samples. Type A
A total of 100 g of each specimen was crushed for heavy liquid Type C
separation. Separated heavy minerals were then mounted and thin
sectioned. The chemistry of these grains was analyzed using elec-
tron probe microanalysis (JXA8621 Super Microprobe; JEOL, Lvf Lvm
Tokyo) at the Research Facility Center for Science and Technol- Figure 4. Framework composition data for sandstone analyzed in the
ogy, University of Tsukuba, Japan. Operating conditions were 20 present study. (A) Qm-F-Lt diagram. (B) Ls-Lvf-Lvm diagram. Com-
kV accelerating voltage, 10 nA specimen current, and ~10-μm positional fields in Qm-F-Lt diagram are after Dickinson et al. (1983).
beam diameter. As no zoning was detected microscopically, we CI—craton interior; TC—transitional continent; BU—basement uplift;
analyzed the center of each grain. The results of all the micro- QR—quartzose recycled; TR—transitional recycled; LR—lithic recy-
cled; DA—dissected arc; TA—transitional arc; UA—undissected arc.
probe analyses were published previously by Okuzawa (2004).

RESULTS

Framework Composition of Sandstone
Sandstone in the study area can be divided into types A,
B, and C on the basis of framework composition, as indicated
by the fields shown in Figure 4. Sandstone of the Setogawa
and Oigawa Groups, the uppermost part of the Kushigatayama
Subgroup (sample 02102504, Table 1), and the lower part of
the Momonoki Subgroup (samples 02102101 and 02102207,
Table 1) belongs to type A. This type of sandstone consists
mainly of monocrystalline quartz (14%–41%), felsic volcanic
lithics (11%–31%), and plagioclase (10%–25%), with minor
amounts of K-feldspar (1%–11%; Table 1). Type A sandstone
plots in the field of dissected to transitional magmatic arc (Dick-
inson et al., 1983; Fig. 4A).
Sandstone of the Kurami Group, the uppermost Kushiga-
tayama Subgroup (sample 02102506, Table 1), and the upper
Momonoki Subgroup (samples 02102208 and 02102303,
Table 1) are classified as type B. This type of sandstone is com-
posed mainly of felsic volcanic lithics (13%–36%), monocrystal-
line quartz (12%–22%), sedimentary lithics (mainly mudstone:
10%–19%), and plagioclase (7%–17%), with minor K-feldspar
(0%–5%) and polycrystalline quartz (0%–6%; Table 1). Type
B sandstone is richer in sedimentary lithics than that of type
A (Fig. 4B) and plots in the field of transitional magmatic arc
(Dickinson et al., 1983; Fig. 4A).
Sandstone from the lower part of the upper Kushigata-
yama Subgroup (samples 02102501, 02102502, and 02102503,
Table 1) belongs to type C. This sandstone consists mainly of
190 Okuzawa and Hisada

TABLE 1. FRAMEWORK COMPOSITIONS OF SANDSTONE SAMPLES

If Mtx
Sample Other Total
Group Qm Qp K-fel. Pl. Lvf Lvi Ls H.m. S.m. and and
no. lithics (%)
Aut. Cem.
Setogawa 02092607 36.6 0.4 11.2 15.4 15.6 0.0 0.2 0.0 0.8 2.8 0.4 16.7 100.0
02093005 38.0 0.0 8.6 18.6 13.8 0.0 0.0 0.0 0.4 0.8 0.0 19.8 100.0
02093008 14.8 0.0 1.4 23.8 29.8 0.0 0.4 0.0 0.4 2.0 1.4 26.0 100.0

Oigawa 02122603 22.6 0.6 2.0 15.0 26.6 0.0 0.6 0.0 0.4 1.0 0.0 31.2 100.0
02122604 35.2 0.8 3.4 10.0 30.8 0.0 0.0 0.0 0.4 0.4 0.0 19.0 100.0
02122606 41.2 0.0 7.0 15.4 10.8 0.0 0.0 0.0 2.6 1.2 0.0 21.8 100.0
02122609 36.4 0.2 4.2 16.2 20.2 0.0 0.4 0.0 0.2 1.8 0.0 20.4 100.0

Kurami 02122803 22.4 4.6 4.0 11.2 22.4 0.0 19.2 0.0 0.8 0.0 0.0 15.4 100.0
02122807 11.6 2.2 5.0 12.0 20.8 0.8 9.6 0.0 1.2 0.0 0.0 36.8 100.0
02122810 21.0 3.6 5.0 7.0 26.8 0.0 18.4 0.0 0.4 0.2 0.0 17.6 100.0

Koma 02102501 1.6 0.8 0.2 8.8 27.0 17.4 2.6 0.0 5.6 4.4 0.0 31.6 100.0
(Kushigata-yama 02102502 0.0 0.0 0.0 34.2 15.2 23.6 0.2 0.0 1.2 9.8 0.0 15.8 100.0
Subgroup) 02102503 9.6 0.6 2.8 25.2 27.8 4.0 5.8 0.0 0.0 0.0 9.2 15.0 100.0
02102504 14.2 0.6 6.8 24.6 23.6 1.4 4.8 0.0 1.2 1.0 0.0 21.8 100.0
02102506 17.6 0.4 0.4 17.2 13.2 0.6 13.0 0.0 0.6 0.2 0.0 36.8 100.0

(Momonoki 02102101 27.6 0.0 1.6 15.4 23.4 0.0 0.6 0.0 0.2 0.4 0.0 30.8 100.0
Subgroup) 02102207 36.6 0.2 5.2 12.4 23.6 0.6 3.8 0.0 0.2 0.2 0.0 17.2 100.0
02102208 13.8 0.2 1.8 12.6 21.4 5.8 13.2 0.0 0.8 1.0 0.2 29.2 100.0
02102303 19.4 5.8 4.8 9.4 35.8 0.2 10.0 0.0 0.4 0.4 0.2 13.6 100.0
Note: Qm—monocrystalline quartz; Qp—polycrystalline quartz; K-fel.—K-feldspar; Pl.—plagioclase; Lvf—felsic volcanics; Lvi—intermediate to mafic
volcanics; Ls—sedimentary lithics; H.m.—heavy minerals; S.m.—secondary minerals; If and Aut.—intrabasinal fragments and authigenic minerals; Mtx
and Cem.—matrix and cement.

felsic to mafic volcanic fragments (32%–44%) together with pla-
gioclase (9%–34%; Table 1). Type C sandstone plots in the field
of undissected and transitional magmatic arc (Dickinson et al.,
1983; Fig. 4A).
Chemistry of Detrital Clinopyroxene
Sandstone from the lower part of the upper Kushigatayama
Subgroup (type C sandstone) yields abundant detrital clinopy-
roxene grains. Almost all of these grains are colorless to light
green and are angular to subrounded. We obtained chemical anal-
yses of 30 detrital clinopyroxenes. Twenty-eight of the analyzed
grains are augite (Fig. 5A). They are Ca-rich and Fe-poor, with
low contents of TiO2 (0.1–0.4 wt%), Cr2O3 (0.0–0.1 wt%), Na2O
(0.2–0.4 wt%), and total Al (total Al as Al2O3: 0.6–1.5 wt%). The
other two grains are augite and diopside, and are characterized
by a higher Cr content (Cr2O3 >0.5 wt%) and total Al (3.1–4.2
wt%), with TiO2 and Na2O contents of 0.3 wt% and 0.1–0.2 wt%,
respectively.
Eighty-eight detrital garnets were analyzed from the Oigawa
Group, of which 70 are classified as pyrope-rich almandines (Pyr
= 5–45 mol%), 10 are spessartine-rich almandines (Sps = 9–40
mol%), six are Ca-rich almandines (Grs + And = 14–23 mol%),
and two are almandine-rich spessartines (Sps = 52 and 62 mol%;
Fig. 6A).
Forty-nine detrital garnets were obtained from the Kurami
Group, of which 24 are spessartine-rich almandines (Sps = 7–44
mol%), 20 are pyrope-rich almandines (Pyr = 8–41 mol%), three
are Ca-rich almandines (Grs + And = 14–17 mol%), one is an
almandine-rich spessartine (Sps = 45 mol%), and one is an Mg-
rich spessartine (Sps = 53 mol%; Fig. 6B).
Sixty-one detrital garnets were analyzed from the Koma
Group. Their compositional range is similar to that of garnets
from the Kushigatayama and Momonoki Subgroups (Fig. 6C).
These grains are mostly pyrope-rich almandines (Pyr = 10–45
mol%) and spessartine-rich almandines (Sps = 6–45 mol%),
with just seven grains of Ca-rich almandines (Grs + And = 6–37
mol%) and one of andradite-rich grossular (Grs = 89 mol%).

Chemistry of Detrital Garnet Chemistry of Detrital Chromian Spinel

A total of 269 chemical analyses of detrital garnets were
obtained from the Setogawa, Oigawa, Kurami, and Koma Groups,
with the exception of the lower part of the upper Kushigatayama
Subgroup. Seventy-one grains of detrital garnet were analyzed
from sandstones of the Setogawa Group: 49 grains are classified
as pyrope-rich almandine (Pyr = 7–45 mol%), 15 grains are spes-
sartine-rich almandines (Sps = 9–47 mol%), 3 grains are Ca-rich
almandines (Grs + And = 11–27 mol%), 3 grains are almandine-
rich spessartines (Sps = 43–63 mol%), and one is a Ca-rich spes-
sartine (Sps = 40 mol%; Fig. 6A).
All the sandstone samples that contain detrital garnets also
yield reddish-brown to black detrital chromian spinels. We obtained
a total of 324 chemical analyses of chromian spinel grains.
The chemical characteristics of detrital chromian spinels
from the Setogawa, Oigawa, Kurami, and Koma Groups are simi-
lar. The spinels from the four groups are characterized by a wide
range in Cr# [= Cr/(Cr+Al) atomic ratio], from 0.2 to 1.0 (Fig. 7).
These spinels show a negative trend in terms of the relationship
between Mg# [= Mg/(Mg+Fe2+) atomic ratio] and Cr# (Fig. 7).
Values of Fe3+# [= Fe3+/(Cr+Al+Fe3+) atomic ratio] are mostly
 Collision between the Honshu and Izu Arcs 191

A Diopside DIOPSIDE Hedenbergite Detrital garnets from the Setogawa and Oigawa Groups
HEDENBERGITE have similar chemical characteristics (Fig. 6A), and are
L. Kushigatayama
mostly spessartine-rich almandine and pyrope-rich alman-
Subgroup dine (Fig. 6A). Spessartine-rich almandine is found in granitic
rocks and low-grade to amphibolite facies metamorphic rocks
AUGITE (Fig. 6A; Nanayama, 1997). Low and high P-T metamorphic
PIGEONITE rocks and granites from the Honshu Arc (Fig. 1) in the area
proximal to the Setogawa Belt are candidates as source rocks
for the spessartine-rich almandine. Intrusion of granitic rocks in
CLINOENSTATITE CLINOFERROSILITE
the Izu Collision Zone postdates sedimentation of the Setogawa
Enstatite Ferrosilite and Oigawa Groups, and they cannot be source rocks for the
detrital garnet. Most of the detrital spessartine-rich almandines
plot in the same field as garnets from the granites and low P-T
B metamorphic rocks (Fig. 6A). Accordingly, we conclude that the
detrital spessartine-rich almandines were probably derived from
0.04 these rocks.
Orogenic calcalkali basalts Detrital pyrope-rich almandines from the Setogawa and
Oigawa Groups plot in the fields of amphibolite, pegmatite, low-
grade metamorphic rocks, and granulite facies metamorphic rocks
Ti

(Fig. 6A; Nanayama, 1997). Thirty-five grains of the almandine

Orogenic tholeiitic basalts
0 that contain garnets with Mg contents >10 wt% have not been
0 0.1 0.2 0.3
Al total
Figure 5. Chemical composition data for detrital clinopyroxenes
analyzed in the present study. The subdivision scheme (A) is after
Morimoto (1988). The range of volcanic rocks of the lower Kushigata-
yama Subgroup is surrounded by broken lines (Shimazu and Ishimaru,
1987). The fields of orogenic calc-alkali and tholeiitic basalts (B) are
after Leterrier et al. (1982).
<0.1, whereas TiO2 wt% is dominantly in the range 0.0–0.5,
although some grains record Fe3+# values >1.0.
PROVENANCE OF PRE-COLLISION SEDIMENTARY
ROCKS
Trench Sedimentary Rocks
Sandstone of the Setogawa and Oigawa Groups (labeled
Type A in Fig. 4) that was deposited in the trench prior to col-
lision (Aoike, 1999) contains fragments of granite, felsic vol-
canic rock, and sedimentary rocks such as chert and mudstone.
Pre–early Miocene granitic rock, felsic volcanic rock, and chert
have not been found in the Izu Arc. Possible sources for these
rock fragments include Cretaceous to early Tertiary granites
and felsic volcanic rocks from the Inner Zone of the Honshu
Arc, as well as a Jurassic to Cretaceous accretionary complex
in the Honshu Arc.
garnet analyzed from the lower Miocene Setogawa and Oigawa
Groups have relatively high Mg contents (MgO >10 wt%). The
Mg/Fe ratio of Ca-poor garnet is known to increase with increas-
ing metamorphic grade (e.g., Coleman et al., 1965). Granulites
found in the Honshu Arc or in the Izu Arc (e.g., Teraoka et al.,
1998), although pyrope-rich almandines have been reported from
Jurassic to Cretaceous sandstones of the Honshu Arc (Adachi
and Kojima, 1983; Teraoka et al., 1999). Such garnets have also
been described from the Sino-Korean Massif of mainland Asia
(e.g., Jiang, 1988), which is likely to have been the source of the
detrital pyrope-rich garnets in the lower Miocene Setogawa and
Oigawa Groups. Because detrital garnets are rare in the sand-
stones of the Setogawa and Oigawa Groups, they may have been
reworked from pre-Miocene sandstones.
The chemistry of detrital garnets from the middle Miocene
trench-fill sedimentary rocks of the Koma Group is similar to that
of detrital garnets from the lower Miocene Setogawa and Oigawa
Groups (Figs. 6A, C). However, the Sea of Japan had begun to
open in the middle Miocene (>15 Ma; e.g., Otofuji et al., 1985),
and it is therefore difficult to envisage that the Sino-Korean Mas-
sif supplied detrital garnet to the study area at this time. Pyrope-
rich garnets of the Koma Group were probably reworked from
pre–middle Miocene sandstones.
Detrital chromian spinels obtained from trench sedimentary
rocks of the Setogawa and Oigawa Groups are characterized by
a wide range in Cr#, low Fe3+#, and low TiO2 content, although
some have a higher TiO2 content (>1.0 wt%; Fig. 7A). A high
Cr#, coupled with low Ti and Fe3+ concentrations, is character-
istic of chromian spinels from forearc peridotite (Bloomer and
Fisher, 1987; Ishii et al., 1992; Parkinson and Pearce, 1998) or
boninites (Arai, 1992; Barnes and Roeder, 2001). Chromian
spinels with low TiO2 content plot mainly within the field of
forearc peridotite (Fig. 7A[i]). Consequently, forearc peridotite
192 Okuzawa and Hisada

A Lower Miocene trench sedimentary rocks B Lower Miocene forearc basin sedimentary
Setogawa Group (N=71)
rocks
Kurami Group (N=49)
Oigawa Group (N=88)
Alm GNF Pyr Alm GNF
Pyr
Sps Sps
ECF ECF
PG+Met.
PG+Met.

PG+Low Met. PG+Low Met.

Skarn Skarn
GNF ECF GNF ECF
APF Gro+And APF Gro+And
Alm Pyr Alm Pyr

C Middle Miocene collision zone sedimentary rocks Granites and metamorphic rocks
in the Ryoke Belt
Alm GNF Pyr
Sps Metamorphic rocks in the
Sambagawa Belt
ECF
Sps: spessartine
PG+Met. Alm: almandine
Pyr: pyrope
Gro: grossular
And: andradite

PG+Low Met. PG: pegmatite

Skarn Low Met.: low metamorphic rock
Koma Group (N=61) APF: amphibolite facies
Momonoki Subgroup ECF: eclogite facies
GNF
Kushigatayama GNF: granulite facies.
ECF
Subgroup
APF
Alm Pyr Gro+And
+And
Figure 6. Chemical composition data for detrital garnets. Garnets from (A) lower Miocene trench sedimentary rocks; (B) lower Miocene forearc
basin sedimentary rocks; (C) middle Miocene collision zone sedimentary rocks. Pegmatite and four metamorphic fields with solid lines are after
Nanayama (1997). The fields of Sambagawa Belt metamorphic rocks are compiled from Higashino et al. (1981, 1984), Enami (1983), Sonobe
and Takasu (2000), and Sakurai (2000). The field of granites and metamorphic rocks in the Ryoke belt is compiled from Ono (1969, 1975a, b,
1976, 1977, 1981), Asami and Hoshino (1980), Asami et al. (1982), Nureki et al. (1982), Seo et al. (1982), Shiba (1982, 1989), Wang (1985),
Takagi and Nagahama (1987), Enami (1988), and Miyazaki (1999).

is a possible source rock for most of the detrital chromian spinels
that have low TiO2 concentrations.
TiO2-rich (>1.0 wt%) detrital chromian spinels are derived
from a different source rock than that yielding the low-Ti spi-
nels. High-Ti spinels with intermediate Cr# are found in intra-
plate magmas (Arai, 1992; Barnes and Roeder, 2001). Intra-
plate basalt is therefore a suitable source rock for the high-Ti
detrital spinels.
Several bodies of mafic and ultramafic rocks occur close to
the study area: forearc peridotites of the Kurosegawa Belt, peri-
dotites and basaltic rocks of the high P-T Sambagawa Metamor-
phic Belt, and backarc basin peridotites and basaltic rocks of the
Circum–Izu Massif Serpentine Belt (Fig. 1). All of these mafic
and ultramafic rocks are exposed within the Honshu Arc. There
is no possible source rock for detrital chromian spinel known
in the Izu Arc. Basalts from the Circum–Izu Massif Serpentine
 A Lower Miocene trench sedimentary rocks
1
Cr
i) ii) Setogawa Group (N=125)
iii)
Oigawa Group (N=111)
5
Cr# [Cr/(Cr+Al)]

TiO2wt%
0 Mg# [Mg/(Mg+Fe2+)] 1 0
Cr# [Cr/(Cr+Al)] 1 Al Fe3+

B Lower Miocene forearc basin sedimentary rocks

1 Cr
i) ii) iii)
Kurami Group (N=43)
Forearc
5
peridotite
Cr# [Cr/(Cr+Al)]

Intraplate basalts of Circum-Izu

Massif Serpentine Belt
TiO2wt%

Kurosegawa
serpentinite

Circum-Izu
Massif Serpentinites

0 Mg# [Mg/(Mg+Fe2+)] 1 0
Cr# [Cr/(Cr+Al)]
1 Al Fe3+

C Middle Miocene collision zone sedimentary rocks

1 Cr
i) ii) Koma Group (N=86) iii)
Momonoki Subgroup
5
Kushigatayama Subgroup
Cr# [Cr/(Cr+Al)]

TiO2wt%

Sambagawa
greenstone and peridotite

0 1
Mg# [Mg/(Mg+Fe2+)] 1 0
Cr# [Cr/(Cr+Al)] Al Fe3+

Figure 7. Chemical composition data for detrital chromian spinels analyzed in the present study. The range of the Circum–Izu Massif serpentinites is
enclosed by thin lines (Arai et al., 1990). The range of the intraplate basalts in the Circum–Izu Massif Serpentine Belt is surrounded by a thick line
(Okuzawa and Hisada, 2004). The range of the forearc peridotite is surrounded by a dotted line (Bloomer and Fisher, 1987; Parkinson and Pearce,
1998). The range of the Kurosegawa serpentinite is circled by a broken line (Okuzawa et al., 2004). The range of the peridotite and greenstone in the
Sambagawa Belt is surrounded by a dotted thin line (Uesugi and Arai, 1999).
194 Okuzawa and Hisada
Belt have intraplate and MORB-type ocean floor signatures
(e.g., Ogawa and Taniguchi, 1988; Hirano et al., 2003) and are
interpreted as having been formed within a backarc setting or a
mid-ocean-ridge setting (Hirano et al., 2003). However, Arai and
Okada (1991) and Arai (1994) reported an occurrence of high Cr#
detrital spinels from tectonic blocks of serpentine-rich sandstone
within tectonic blocks of the Circum–Izu Massif serpentinites.
Although the sedimentary age of the serpentine-rich sandstone
is unsettled, Arai (1994) concluded that those sandstones were
deposited immediately after emplacement of the Circum–Izu
Massif serpentinites, and that these incipient serpentinites had
petrological characteristics of peridotite formed in an arc setting.
Many low-TiO2 detrital chromian spinels from the Setogawa
and Oigawa Groups do not plot within the field of spinels from
the Kurosegawa forearc peridotite, but instead within the field
of spinels from the recent Circum–Izu Massif serpentinites
(Fig. 7A[i], [ii]). There are three possible combinations of source
rocks for the low-TiO2 detrital spinels: (1) ancient Circum–Izu
Massif serpentinites that supplied all the low-TiO2 detrital spi-
nels found in the Setogawa and Oigawa Groups, (2) ancient
Circum–Izu Massif serpentinites and other forearc peridotites
and/or boninites that supplied the low-TiO2 detrital spinels, and
(3) other forearc peridotites and boninites that supplied all the
low-TiO2 detrital spinels. In the Boso Peninsula (Fig. 1) the lower
Miocene Mineoka Group and middle Miocene Sakuma Group
contain detrital spinels with low Ti content and a wide range of
Cr# (Cr# = 0.3–0.9: Okuzawa and Hisada, 2004). These authors
concluded that the Circum–Izu Massif serpentinites had supplied
detrital chromian spinels to both groups, because both groups
include conglomerate containing clasts of serpentinite, gabbro,
and basalts, which probably originated from the ophiolitic rocks
in the Circum–Izu Massif Serpentine Belt. On this basis, the first
scenario above appears to be the most plausible.
Most of the high-TiO2 chromian spinels found in the Setogawa
and Oigawa Groups plot within the field of alkaline basalt from the
Circum–Izu Massif Serpentine Belt (Fig. 7A[ii]). Peridotites and
basalts in the high P-T Sambagawa Metamorphic Belt also contain
high-TiO2 spinels (Fig. 7A[ii]; Uesugi and Arai, 1999), however,
most of the detrital spinels have a lower Fe3+# than spinels in the
peridotite and basalt of the Sambagawa Belt (Fig. 7A[iii]). The
high-TiO2 detrital chromian spinels were probably derived from
intraplate basalts of the Circum–Izu Massif Serpentine Belt.
The detrital chromian spinels with higher values of Mg#
and Cr# also plot outside the fields of the forearc peridotites
and Circum–Izu Massif serpentinites in an Mg#-Cr# diagram
(Fig. 7A[i]). Intraplate basalt from the Circum–Izu Massif Ser-
pentine Belt contains some spinels with higher values of Mg#
and Cr# than those of the forearc peridotites and Circum–Izu
Massif serpentinites (Ishida et al., 1990); however, most of the
detrital chromian spinels have lower TiO2 contents than spinels
in the intraplate basalts. According to Barnes and Roeder (2001),
chromitites within ophiolite contain chromian spinels with high
Mg# and Cr# and low Fe3+# and TiO2 contents. On this basis,
some of the detrital spinels analyzed in the present study appear
to be derived from chromitite-bearing ophiolite. It is known that
the Circum–Izu Massif serpentinites include a small amount of
chromitite (Kitahara, 1954).
In summary, the framework composition of sandstones and
the chemistry of detrital chromian spinels and garnets indicate
that prior to collision the trench sediments were supplied from
the Honshu Arc.
Forearc Basin Sedimentary Rocks
Sandstone of the upper lower Miocene Kurami Group
(labeled Type B in Fig. 4) was deposited in the forearc basin
of the Honshu Arc and has a different framework composi-
tion from that of lower Miocene trench sedimentary rocks:
The forearc basin sandstone is richer in sedimentary lithics in
comparison with the trench sandstone, although the assemblage
of detrital grains (Table 1) and the chemical characteristics of
detrital chromian spinels and garnets (Figs. 6 and 7) are simi-
lar. Marsaglia and Ingersoll (1992) concluded that very little
compositional variation is related to forearc basin geometry
or to submarine transport from the forearc into the trench. The
compositional differences recorded between lower Miocene
trench and forearc basin sedimentary rocks indicate that detri-
tus from the two groups of rocks was derived from different
provenances. The forearc basin sedimentary rocks appear to
have had their source from a different part of the Honshu Arc
in comparison with the trench-fill sediments. The detritus of the
Setogawa Group may have been axial currents along the trench
axis, because the framework composition of the Setogawa
Group sandstone resembles the trench sandstone of the lower
Miocene Mineoka Group in the Boso Peninsula (Watanabe and
Iijima, 1990; Okuzawa, 2004).
CHANGES IN PROVENANCE FOLLOWING
COLLISION
The lower part of the upper Kushigatayama Subgroup sand-
stone comprises sandstone of type C (see above; Fig. 4). This
rock contains intermediate to mafic volcanic lithics (Fig. 4B)
and abundant detrital clinopyroxene. A volcanic arc is therefore a
likely provenance for this rock type. We used the discrimination
diagram of Leterrier et al. (1982) to infer the tectonic setting of
the source rock on the basis of detrital clinopyroxene geochemis-
try (Fig. 5B). Two detrital clinopyroxenes with Cr2O3 >0.5 wt%
are excluded from analysis because they might have been derived
from different source rocks. Plots of detrital clinopyroxenes from
the lower part of the upper Kushigatayama Subgroup form a tight
cluster on the discrimination diagram, with most grains plotting
in the field of orogenic tholeiitic basalts (Fig. 5B). On this basis,
we conclude that detrital clinopyroxenes from the Kushigata-
yama Subgroup were derived from orogenic tholeiitic basalts.
Lower Miocene volcanic rocks from the Honshu Arc con-
sist of both tholeiitic and calc-alkaline rocks (e.g., Shuto et al.,
1988). Basalts from the Circum–Izu Massif Serpentine Belt
 Collision between the Honshu and Izu Arcs 195

have intraplate and MORB-type ocean floor geochemical sig- A Lower part of the upper Kushigatayama Subgroup
natures (e.g., Ogawa and Taniguchi, 1988; Hirano et al., 2003). (eve of the collision)
The chemistry of clinopyroxene in these rocks is different Type A
from that of orogenic tholeiitic basalts. Clinopyroxenes within rc Type B
ua
high P-T metamorphic rocks from the Sambagawa Belt show Ho nsh
Type C
omphacite and aegirine-augite compositions (Higashino et al.,
.
1981; Sakurai, 2000) and are therefore different from detrital i Gp
am
Kur
clinopyroxene grains in the upper Kushigatayama Subgroup. . arc
ted Gp aG
p. Izu
Volcanic rocks from the lower Kushigatayama Subgroup are cre awa w
c
A tog Oiga
island arc tholeiites and contain clinopyroxenes that are mainly Se ma
taya
higa
diopside and augite, with minor chrome diopside (Fig. 5A; Phi
lipp Kus gp.
ine S u b
Shimazu and Ishimaru, 1987). The chemical characteristics of the Sea
Pla
te
detrital clinopyroxenes from the lower part of the upper Kushiga-
tayama Subgroup correspond with those of clinopyroxene-bearing
volcanic rocks from the lower Kushigatayama Subgroup. Fujioka B Upper part of the upper Kushigatayama Subgroup
and lower Momonoki Subgroup (begining of collision)
and Saito (1992) reported chemical analyses of detrital clinopy-
roxenes from Oligocene to Pleistocene volcaniclastic sediments
and sedimentary rocks sampled during Ocean Drilling Program arc
shu
Hon
(ODP) Leg 126 at the Izu Arc. Detrital clinopyroxenes from the
p.
upper Kushigatayama Subgroup analyzed in the present study are ubg
no ki S
similar to those from Oligocene to Miocene sedimentary rocks mo
Mo
sampled from ODP Leg 126 sites. In summary, detritus from the . arc
ted a Gp p. Izu
cre wa G
lower part of the upper Kushigatayama Subgroup appears to have Ac togaw Oiga
Se ma
been derived from the Izu Arc (Fig. 8A). taya
Phi u s higa
K gp.
Sandstone from the uppermost Kushigatayama Subgroup lipp
ine Sub
Sea
and the lower Momonoki Subgroup is classified within our type Pla
te
A category (see above; Fig. 4) and is similar to sandstone within
the lower Miocene trench-fill (Fig. 8B) in terms of the chemis- C Upper Momonoki Subgroup (late stage of collision of
try of detrital heavy minerals (Figs. 6 and 7). The change from Koma Block)
type C to type A that is observed stratigraphically upward within
the Kushigatayama Subgroup is interpreted as a change in prov-
arc
enance related to the arrival of the Izu Arc at the Honshu Trench. shu
Hon
This indicates that the basin within the Izu Arc in which the p.
ubg
Kushigatayama Subgroup was deposited had moved close to the no ki S
mo
trench by 16–15 Ma (Aoike, 1999). Mo
. arc
Sandstone of the upper Momonoki Subgroup is classified as ted a Gp p Izu
cre G
Ac togaw wa
type B and contains detrital heavy minerals of similar chemistry Se Oig
a ma
& taya
to those within the lower Miocene Kurami Group, deposited in Phi u s higa
lip K gp.
the forearc basin (Figs. 6 and 7). As discussed above, sandstone pine
Sea Sub
Pla
of type B was supplied from a different source area of the Hon- te
shu Arc from that of type A (Fig. 8C). The change in provenance
recorded within the Momonoki Subgroup appears to have been Figure 8. Temporal changes to framework composition of the Koma
caused by a change in the passage of detritus to the trench owing Group sandstone at three stages: (A) Lower part of the upper Kushi-
to progressive collision between the two arcs. gatayama Subgroup. (B) Upper part of the upper Kushigatayama Sub-
Conglomerate within the upper Momonoki Subgroup contains group and the lower Momonoki Subgroup. (C) Upper Momonoki Sub-
group. See the text for explanation of types A, B, and C sandstones.
volcanic clasts derived from both the Kushigatayama Subgroup
and the Honshu Arc (Koyama, 1993); however, judging from the
dominance of sedimentary lithics over intermediate to mafic vol-
canic lithics in sandstone (Fig. 4A) and the lack of detrital clino- CONCLUSIONS
pyroxene, sediments of the Momonoki Subgroup were probably
derived mainly from the Honshu Arc. During the early stages of To understand the change in provenance resulting from col-
collision within the Izu Collision Zone, the Honshu Arc appears to lision between the Izu and Honshu Arcs, we studied the frame-
have been preferentially uplifted and thereby supplied more detri- work composition of sandstone and the chemical composition of
tus to the trench in comparison with sources in the Izu Arc. detrital clinopyroxene, garnet, and chromian spinel from lower to
196 Okuzawa and Hisada
middle Miocene coarse-grained sedimentary rocks deposited in
and around the Izu Collision Zone. Sediments that were depos-
ited prior to collision were derived from the Honshu Arc, includ-
ing Cretaceous to early Tertiary granites, low P-T metamorphic
rocks, Jurassic to Cretaceous accretionary complexes, and forearc
peridotite and/or backarc basin ophiolite. Sedimentary rocks of
the pre-collisional trench and forearc basin of the Honshu Arc
differ in terms of the framework composition of sandstone.
Sediments of the lower part of the upper Kushigatayama Sub-
group were derived from volcanic rocks of the Izu Arc, whereas
those of the uppermost Kushigatayama Subgroup and the lower
Momonoki Subgroup resemble lower Miocene trench fill. This
change in provenance was caused by the arrival of the Izu Arc at
the Honshu Trench. Sandstone of the upper part of the Momo-
noki Subgroup is similar to lower Miocene sedimentary rocks
deposited in the forearc of the Honshu Arc. This suggests that
the passage of detritus to the trench changed following collision.
In the early stages of collision between the Izu and Honshu Arcs
the Honshu Arc was preferentially uplifted and consequently pro-
vided the bulk of the detritus supplied to the trench.
ACKNOWLEDGMENTS
We appreciate the informative and helpful discussions with
Shoji Arai on the chemistry of chromian spinel and with Yujiro
Ogawa on tectonics in the Boso Peninsula. Comments on the
manuscript by Peter D. Clift, Andy Morton, Kantaro Fujioka,
and an anonymous reviewer are acknowledged. We extend our
thanks to Teruo Ohtsubo, Naoto Hirano, Jun-ichiro Kuroda, and
Shoichiro Tokumine for their help in the field work. Thanks are
due to Norimasa Nishida for his help with microprobe analysis.
REFERENCES CITED
Adachi, M., and Kojima, S., 1983, Geology of the Mt. Hikagedaira area, east
of Takayama, Gifu Prefecture, central Japan: Journal of Earth Sciences,
Nagoya University, v. 31, p. 37–67.
Amano, K., 1986, Southern Fossa Magna as multiple collision belt: Earth
Monthly (Chikyu), v. 8, p. 581–585 (in Japanese).
Amano, K., 1991, Multiple collision tectonics of the South Fossa Magna in
Central Japan: Modern Geology, v. 15, p. 315–329.
Amano, K., Martin, A.J., Tanakadate, H., Kanaguri, S., Yoda, N., and Aizu, T.,
1999, Tectonics and basin formation in the arc-arc collision zone: The
example of the South Fossa Magna in central Japan: Structural Geology
(Journal of the Tectonic Research Group of Japan), no. 43, p. 11–20 (in
Japanese with English abstract).
Aoike, K., 1999, Tectonic evolution of the Izu Collision Zone: Research Report
of the Kanagawa Prefectural Museum of Natural History, v. 9, p. 111–151
(in Japanese with English abstract).
Arai, S., 1991, The Circum–Izu Massif peridotite, central Japan, as back-arc
mantle fragments of the Izu-Bonin arc system, in Peters, Tj., et al., eds.,
Ophiolite Genesis and Evolution of the Oceanic Lithosphere: Dordrecht,
Netherlands, Kluwer Academic Publishers, p. 801–816.
Arai, S., 1992, Chemistry of chromian spinel in volcanic rocks as a potential
guide to magma chemistry: Mineralogical Magazine, v. 56, p. 173–184,
doi: 10.1180/minmag.1992.056.383.04.
Arai, S., 1994, The Circum–Izu Massif Serpentine Belt: Geoscience Reports
of Shizuoka University, v. 20, p. 175–185, in Japanese with English
abstract.
Arai, S., and Okada, H., 1991, Petrology of serpentine sandstone as a key to tec-
tonic development of serpentine belts: Tectonophysics, v. 195, p. 65–81,
doi: 10.1016/0040-1951(91)90144-H.
Arai, S., Ito, M., Nakayama, N., and Masuda, F., 1990, A suspect serpentinite
mass in the Tokyo Bay area: Petrology and provenance of serpentinite
pebbles in the upper Cenozoic system in the Boso Peninsula, central
Japan: Journal of the Geological Society of Japan, v. 96, p. 171–179 (in
Japanese with English abstract).
Asami, M., and Hoshino, M., 1980, Staurolite-bearing schists from the Hongu-
san area in the Ryoke metamorphic belt, central Japan: Journal of the
Geological Society of Japan, v. 86, p. 581–591.
Asami, M., Hoshino, M., Miyakawa, K., and Suwa, K., 1982, Metamorphic
conditions of staurolite schists of the Ryoke metamorphic belt in the
Hazu-Hongusan area, central Japan: Journal of the Geological Society of
Japan, v. 88, p. 437–450 (in Japanese with English abstract).
Barnes, J.S., and Roeder, P.L., 2001, The range of spinel compositions in ter-
restrial mafic and ultramafic rocks: Journal of Petrology, v. 42, p. 2279–
2302, doi: 10.1093/petrology/42.12.2279.
Bloomer, S.H., and Fisher, R.L., 1987, Petrology and geochemistry of igneous
rocks from the Tonga Trench–A non-accreting plate boundary: Journal of
Geology, v. 95, p. 469–495.
Cawood, P.A., 1983, Modal composition and detrital clinopyroxene geo-
chemistry of lithic sandstones from the New England Fold Belt (east
Australia): A Paleozoic forearc terrane: Geological Society of America
Bulletin, v. 94, p. 1199–1214, doi: 10.1130/0016-7606(1983)94<1199:
MCADCG>2.0.CO;2.
Coleman, R.G., Lee, D.E., Beatty, L.B., and Brannock, W.W., 1965, Eclog-
ites and eclogites: Their differences and similarities: Geological
Society of America Bulletin, v. 76, p. 483–508, doi: 10.1130/0016-
7606(1965)76[483:EAETDA]2.0.CO;2.
Deer, W.A., Howie, R.A., and Zussman, J., 1982, An Introduction to the Rock-
Forming Minerals (2nd edition): London, Longman, 696 p.
Dickinson, W.R., Bread, L.S., Brakenridge, G.R., Erjavec, J.L., Ferguson, R.C.,
Inman, K.F., Kenepp, R.A., Lindberg, F.A., and Ryberg, P.T., 1983, Prov-
enance of North American Phanerozoic sandstones in relation to tectonic
setting: Geological Society of America Bulletin, v. 94, p. 222–235, doi:
10.1130/0016-7606(1983)94<222:PONAPS>2.0.CO;2.
Enami, M., 1983, Petrology of pelitic schists in the oligoclase-biotite zone of
the Sanbagawa metamorphic terrain, Japan: Phase equilibria in the high-
est grade zone of a high-pressure intermediate type of metamorphic belt:
Journal of Metamorphic Geology, v. 1, p. 141–161, doi: 10.1111/j.1525-
1314.1983.tb00269.x.
Enami, M., 1988, Chemistry of garnet of the Ryoke metamorphic belt: Journal
of Geological Society, Nagoya, no. 50, p. 71–80 (in Japanese).
Fujioka, K., and Saito, S., 1992, Composition of heavy minerals from sands and
sandstones of the Izu-Bonin arc, Leg 126, in Taylor, B., Fujioka, K., et al.,
eds., Proceedings of the Ocean Drilling Program, Scientific Results, Vol-
ume 126: College Station, Texas, Ocean Drilling Program, p. 155–169.
Higashino, T., Sakai, C., Otsuki, M., Itaya, T., and Banno, S., 1981, Electron
microprobe analyses of rock-forming minerals from the Sanbagawa met-
amorphic rocks, Shikoku, Part I. Asemi River area: Science Report of
Kanazawa University, v. 26, p. 73–122.
Higashino, T., Sakai, C., Kurata, H., Enami, M., Hosotani, H., Enami, M., and
Banno, S., 1984, Electron microprobe analyses of rock-forming minerals
from the Sanbagawa metamorphic rocks, Shikoku, Part II. Sazare, Kotu
and Bessi Areas: Science Report of Kanazawa University, v. 29, p. 37–64.
Hirano, N., Ogawa, Y., Saito, K., Yoshida, T., Sato, H., and Taniguchi, H., 2003,
Multi-stage evolution of the Tertiary Mineoka ophiolite, Japan: New
geochemical and age constraints, in Dilek, Y., and Robinson, P.T., eds.,
Ophiolites in Earth History: Geological Society [London] Special Publi-
cation 218, p. 279–298.
Hiroki, Y., 1995, Sea-level changes in the early to early middle Miocene series,
central Honshu, Japan: Journal of the Geological Society of the Faculty of
Science, University of Tokyo, Section II, v. 22, p. 251–284.
Hiroki, Y., and Matsumoto, R., 1999, Magnetostratigraphic correlation of Mio-
cene regression-and-transgression boundaries in central Honshu, Japan:
Journal of the Geological Society of Japan, v. 105, p. 87–107.
Hisada, K., and Arai, S., 1993, Detrital chrome spinels in the Cretaceous San-
chu sandstone, central Japan: Indicator of serpentinite protrusion into a
fore-arc region: Palaeogeography, Palaeoclimatology, Palaeoecology,
v. 105, p. 95–109, doi: 10.1016/0031-0182(93)90109-V.
 Collision between the Honshu and Izu Arcs
Huchon, P., and Kitazato, H., 1984, Collision of the Izu block with central Japan
during the Quaternary and geological evolution of the Ashigara area: Tec-
tonophysics, v. 110, p. 201–210, doi: 10.1016/0040-1951(84)90261-0.
Ibaraki, M., 1986, Planktonic foraminiferal datum levels recognized in the
Neogene sequence of the Kakegawa area, and their relationship with
the lithostratigraphy: Journal of the Geological Society of Japan, v. 92,
p. 119–134 (in Japanese with English abstract).
Ingersoll, R.V., Bullard, T.F., Ford, R.L., Grimm, J.P., and Pickle, J.D., 1984,
The effect of grain size on detrital modes: A test of the Gazzi-Dickinson
point-counting method: Journal of Sedimentary Petrology, v. 54, p. 103–
116.
Ishida, T., Arai, S., and Takahashi, N., 1990, Metamorphosed picrite basalts
in the northern part of the Setogawa belt, central Japan: Journal of the
Geological Society of Japan, v. 96, p. 181–191 (in Japanese with English
abstract).
Ishii, T., Robinson, P.T., Maekawa, H., and Fiske, R., 1992, Petrological studies
of peridotites from diapiric serpentinite seamounts in the Izu-Ogasawara-
Mariana forearc, Leg 125, in Fryer, P., Pearce, J.A., et al., eds., Proceed-
ings of the Ocean Drilling Program, Scientific Results, Volume 125: Col-
lege Station, Texas, Ocean Drilling Program, p. 445–485.
Isozaki, Y., 1996, Anatomy and genesis of a subduction-related orogen: A new
view of geotectonic subdivision and evolution of the Japanese Islands:
Island Arc, v. 5, p. 289–320, doi: 10.1111/j.1440-1738.1996.tb00033.x.
Ito, M., 1985, The Ashigara Group: A regressive submarine fan–fan delta
sequence in a Quaternary collision boundary, north of Izu Peninsula,
central Honshu, Japan: Sedimentary Geology, v. 45, p. 261–292, doi:
10.1016/0037-0738(85)90005-3.
Ito, M., 1987, Middle to Late Miocene foredeep basin successions in an arc-arc
collision zone, northern Tanzawa Mountains, central Honshu, Japan: Sed-
imentary Geology, v. 54, p. 61–91, doi: 10.1016/0037-0738(87)90004-2.
Jiang, Y., 1988, Genetic mineralogy and minerogenetic characteristics of Early
Precambrian iron formation in Nei Monggol Zizhiqu: Bulletin of the
Tianjin Institute of Geology and Mineral Resources, v. 19, p. 1–56 (in
Chinese with English abstract).
Kitahara, J., 1954, On the chromite from the Hiroosa mine, Shizuoka Prefec-
ture: Journal of Mineralogy: Petrology and Economic Geology, v. 38,
p. 176–186 (in Japanese with English abstract).
Kosaka, T., and Tsunoda, F., 1969, Geology of the Koma Massif in the western
part of Yamanashi Prefecture: Journal of the Geological Society of Japan,
v. 75, p. 127–140 (in Japanese with English abstract).
Koyama, A., 1991, Collision of the Kushigatayama Block with the Honshu Arc
during the Middle Miocene: Modern Geology, p. 15, p. 331–345.
Koyama, A., 1993, Collision of the Kushigatayama block during the Middle
Miocene, northwestern part of South Fossa-Magna, central Japan: Mem-
oirs of Geological Society of Japan, no. 42, p. 245–254 (in Japanese with
English abstract).
Leterrier, J., Maury, R.C., Thonon, P., Girard, D., and Marchal, M., 1982,
Clinopyroxene composition as a method of identification of the magmatic
affinities of paleo-volcanic series: Earth and Planetary Science Letters,
v. 59, p. 139–154, doi: 10.1016/0012-821X(82)90122-4.
Marsaglia, K.M., and Ingersoll, R.V., 1992, Compositional trends in arc-related,
deep-marine sand and sandstone: A reassessment of magmatic-arc prov-
enance: Geological Society of America Bulletin, v. 104, p. 1637–1649,
doi: 10.1130/0016-7606(1992)104<1637:CTIARD>2.3.CO;2.
Martin, A.J., and Amano, K., 1999, Facies analyses of Miocene subaqueous vol-
caniclastics in the Koma Mountains, South Fossa Magna, central Japan:
Journal of the Geological Society of Japan, v. 105, p. 552–572.
Miyashiro, A., 1953, Calcium-poor garnet in relation to metamorphism: Geo-
chimica et Cosmochimica Acta, v. 4, p. 179–208, doi: 10.1016/0016-
7037(53)90014-3.
Miyazaki, K., 1999, Thermobarometry of the Tsukuba Metamorphic Rocks:
Bulletin of the Geological Society of Japan, v. 50, p. 515–525 (in Japa-
nese with English abstract).
Morimoto, N., 1988, Nomenclature of pyroxenes: Mineralogical Magazine,
v. 52, p. 535–550, doi: 10.1180/minmag.1988.052.367.15.
Morton, A.C., 1991, Geochemical studies of detrital heavy minerals and their
application to provenance research, in Morton, A.C., et al., eds., Develop-
ment in Sedimentary Provenance Studies: Geological Society [London]
Special Publication 57, p. 31–45.
Nanayama, F., 1997, An electron microprobe study of the Amazon fan, in
Flood, R.D., Piper, D.J.W., et al., eds., Proceedings of the Ocean Drilling
197
Program, Scientific Results, Volume 155: College Station, Texas, Ocean
Drilling Program, p. 147–168.
Niitsuma, N., and Akiba, T., 1985, Neogene tectonic evolution and plate sub-
duction in the Japanese island arcs, in Nasu, N., et al., eds., Formation of
Active Ocean Margins: Tokyo, Terra Scientific Publishing (TERRAPUB),
p. 75–108.
Nureki, T., Asami, M., Shibata, T., and Ohira, K., 1982, The Ryoke Belt of
the southwestern part of Shiaku-shoto area in the Seto-naikai Inland Sea:
Journal of the Geological Society of Japan, v. 88, p. 499–510 (in Japanese
with English abstract).
Ogawa, Y., and Taniguchi, H., 1988, Geology and tectonics of the Miura-Boso
Peninsulas and the adjacent area: Modern Geology, v. 12, p. 147–168.
Okuzawa, K., 2004, Reconstruction of paleogeography and unraveling of uplift
history in collisional zones using detrital heavy minerals—Setogawa-
Mineoka Belt and Himalaya-Bengal system [Ph.D. thesis]: Tsukuba,
Japan, University of Tsukuba, 324 p.
Okuzawa, K., and Hisada, K., 2004, Detrital chromian spinels from the Mio-
cene Haccho Formation of the Mineoka Group and the Sakuma Group,
Boso Peninsula, central Japan: Journal of the Geological Society of Japan,
v. 110, p. 237–243 (in Japanese with English abstract).
Okuzawa, K., Yagi, N., Hisada, K., and Arai, S., 2004, Detrital chromian spinels
from the Minokuchi and Uzuhiki Formations in the Itsukaichi area of the
Kanto Mountains, central Japan: Tsukuba, Japan, Annual Report of the
Institute of Geoscience, University of Tsukuba, no. 30, p. 45–51.
Ono, A., 1969, Zoning of the metamorphic rocks in the Takato-Sioziri area,
Nagano Prefecture: Journal of the Geological Society of Japan, v. 75,
p. 521–536 (in Japanese with English abstract).
Ono, A., 1975a, Spatial distribution, size and chemistry of garnet in a hornfels
from the Takato area, Ryoke metamorphic belt, central Japan: Journal of
the Japanese Association of Mineralogists, Petrologists, and Economic
Geologists, v. 70, p. 169–173.
Ono, A., 1975b, Compositions and sizes of garnet crystals from a hornfels, Ten-
ryukyo region, central Japan: Journal of the Japanese Association of Min-
eralogists, Petrologists, and Economic Geologists, v. 70, p. 194–199.
Ono, A., 1976, Crystal growth and zoning of garnet from the Ryoke metamor-
phic rocks of central Japan: Journal of the Japanese Association of Miner-
alogists, Petrologists, and Economic Geologists, v. 71, p. 308–325.
Ono, A., 1977, Chemical reactions at the boundary between gneiss and amphib-
olite in the Ryoke metamorphic terrain at Takato, Japan: Journal of the
Geological Society of Japan, v. 83, p. 33–40.
Ono, A., 1981, Geology of the Takato-Kashio area, Ryoke belt, central Japan:
Journal of the Geological Society of Japan, v. 87, p. 249–257 (in Japanese
with English abstract).
Osozawa, S., Sakai, T., and Naito, T., 1990, Miocene subduction of an active
mid-ocean ridge and origin of the Setogawa Ophiolite, central Japan:
Journal of Geology, v. 98, p. 763–771.
Otofuji, Y., Matsuda, T., and Nohda, S., 1985, Opening of the Japan Sea inferred
from the paleomagnetism of the Japan arc: Nature, v. 317, p. 603–604,
doi: 10.1038/317603a0.
Parkinson, I.J., and Pearce, J.A., 1998, Peridotites from the Izu-Bonin-Mariana
forearc (ODP Leg 125): Evidence for mantle melting and melt-mantle
interaction in a supra-subduction zone setting: Journal of Petrology, v. 39,
p. 1577–1618, doi: 10.1093/petrology/39.9.1577.
Sakurai, T., 2000, Chemical compositions of the constituent minerals of the
Gazo mass, a tectonic block in the Sambagawa metamorphic belt, Besshi
district, central Shikoku, Japan: Geoscience Report of Shimane Univer-
sity, v. 19, p. 167–185.
Seo, T., Suzuki, M., and Yokoyama, S., 1982, Ryoke metamorphic rocks in
the Uoshima area, Southwest Japan, in Yamamoto, H., ed., Petrological
Study of the Ryoke Belt, Report of Grant-in-Aid for Scientific Research
(A)(1981): Munakata, Japan, Fukuoka University of Education, p. 7–18
(in Japanese).
Shiba, M., 1982, Metamorphic conditions of the Tsukuba metamorphic rocks:
Journal of the Japanese Association of Mineralogists, Petrologists, and
Economic Geologists, v. 77, p. 345–455 (in Japanese with English
abstract).
Shiba, M., 1989, Grandite-pyralspite garnets from the Tsukuba and Hidaka
metamorphic rocks: Science Reports of the Hirosaki University, v. 36,
p. 175–183 (in Japanese with English abstract).
Shimazu, M., and Ishimaru, I., 1987, Neogene volcanic rocks of the eastern part
of the Koma Mountains, Yamanashi Prefecture: Journal of the Japanese
198 Okuzawa and Hisada
Association of Mineralogists, Petrologists, and Economic Geologists,
v. 82, p. 382–394 (in Japanese with English abstract).
Shuto, K., Takimoto, T., Sakai, A., Yamazaki, T., and Takahashi, T., 1988, Geo-
chemical variation with time of the Miocene volcanic rocks in northern
part of the Northeast Japan arc: Journal of the Geological Society of
Japan, v. 94, p. 155–172 (in Japanese with English abstract).
Soh, W., 1986, Reconstruction of Fujikawa trough in Mio-Pliocene age and its
geotectonic implication: Kyoto, Japan, Memoirs of the Faculty of Science,
Kyoto University, Series of Geology and Mineralogy, v. 52, p. 1–68.
Sonobe, M., and Takasu, A., 2000, Major and trace element chemistry of the
garnets within the Sambagawa pelitic schists in the Asemigawa area,
central Shikoku, Japan: Geoscience Report of Shimane University, v. 19,
p. 151–166.
Sugiyama, Y., 1995, Geology of the northern Setogawa Belt in the Akaishi
Mountains and the formation of the Setogawa accretionary complex: Bul-
letin of the Geological Society of Japan, v. 46, p. 177–214 (in Japanese
with English abstract).
Taira, A., Tokuyama, H., and Soh, W., 1989, Accretion tectonics and evolution
of Japan, in Ben-Avraham, Z., ed., Evolution of the Pacific Ocean Margin:
New York, Oxford University Press, p. 100–123.
Takagi, H., and Nagahama, H., 1987, The Ryoke Belt in the Hiki Hills, north-
eastern marginal area of the Kanto Mountains: Journal of the Geological
Society of Japan, v. 93, p. 210–215 (in Japanese with English abstract).
Takeuchi, M., 1994, Changes in garnet chemistry show a progressive denu-
dation of the source areas for Permian-Jurassic sandstones, Southern
Kitakami Terrane, Japan: Sedimentary Geology, v. 93, p. 85–105, doi:
10.1016/0037-0738(94)90030-2.
Teraoka, Y., Suzuki, M., and Kawakami, K., 1998, Provenance of Cretaceous
and Paleogene sediments in the Median Zone of Southwest Japan: Bul-
letin of the Geological Society of Japan, v. 49, p. 395–411 (in Japanese
with English abstract).
Teraoka, Y., Okumura, K., Suzuki, M., and Kawakami, K., 1999, Clastic sedi-
ments of the Shimanto Supergroup in Southwest Japan: Bulletin of the
Geological Society of Japan, v. 50, p. 559–590 (in Japanese with English
abstract).
Tsuchi, R., 1986, 1:200,000 Geologic Map of Shizuoka Prefecture: Shizuoka,
Japan, Shizuoka Prefecture.
Uesugi, J., and Arai, S., 1999, The Shiokawa peridotite mass in the Mikabu
belt, central Japan, as a cumulate from intra-plate tholeiite: Memoirs of
Geological Society of Japan, no. 52, p. 229–242 (in Japanese with English
abstract).
Wang, G., 1985, A Ca-Mn-Fe garnet found in the Ryoke metamorphic rocks at
the Wazuka area, Kyoto Prefecture: Journal of the Japanese Association of
Mineralogists, Petrologists, and Economic Geologists, v. 80, p. 459–462
(in Japanese with English abstract).
Watanabe, Y., and Iijima, A., 1990, Evolution of the Tertiary Setogawa-
Kobotoke-Mineoka turbidite fills: Tokyo, Journal of the Faculty of Sci-
ence, University of Tokyo, Section II, v. 20, p. 425–441.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Cenozoic volcanic arc history of East Java, Indonesia: The stratigraphic record of
eruptions on an active continental margin

Helen R. Smyth*
Robert Hall
Gary J. Nichols
SE Asia Research Group, Geology Department, Royal Holloway University of London, Egham TW20 0EX, UK

ABSTRACT

The stratigraphic record of volcanic arcs provides insights into their eruptive
history, the formation of associated basins, and the character of the deep crust
beneath them. Indian Ocean lithosphere was subducted continuously beneath Java
from ca. 45 Ma, resulting in formation of a volcanic arc, although volcanic activity
was not continuous for all of this period. The lower Cenozoic stratigraphic record
on land in East Java provides an excellent opportunity to examine the complete
eruptive history of a young, well-preserved volcanic arc from initiation to termi-
nation. The Southern Mountains Arc in Java was active from the middle Eocene
(ca. 45 Ma) to the early Miocene (ca. 20 Ma), and its activity included significant
acidic volcanism that was overlooked in previous studies of the area. In particu-
lar, quartz sandstones, previously considered to be terrigenous clastic sedimentary
rocks derived from continental crust, are now known to be of volcanic origin. These
deposits form part of the fill of the Kendeng Basin, a deep flexural basin that formed
in the backarc area, north of the arc. Dating of zircons in the arc rocks indicates
that the acidic character of the volcanism can be related to contamination of mag-
mas by a fragment of Archean to Cambrian continental crust that lay beneath the
arc. Activity in the Southern Mountains Arc ended in the early Miocene (ca. 20 Ma)
with a phase of intense eruptions, including the Semilir event, which distributed ash
over a wide area. Following the cessation of the early Cenozoic arc volcanism, there
followed a period of volcanic quiescence. Subsequently arc volcanism resumed in
the late Miocene (ca. 12–10 Ma) in the modern Sunda Arc, the axis of which lies
50 km north of the older arc.

Keywords: East Java, Indonesia, stratigraphic record, Cenozoic arc volcanism.

*Present address: Cambridge Arctic Shelf Programme (CASP), 181A Huntingdon Road, Cambridge CB3 0DH, UK

Smyth, H.R., Hall R., and Nichols, G.J., 2008, Cenozoic volcanic arc history of East Java, Indonesia: The stratigraphic record of eruptions on an active continental
margin, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of
America Special Paper 436, p. 199–222, doi: 10.1130/2008.2436(10). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of
America. All rights reserved.

199
200 Smyth et al.
INTRODUCTION
The island of Java lies within the Indonesian archipelago
between the landmasses of Eurasia and Australia, on the margin
of the Eurasian plate. The southeastern part of this plate is known
as Sundaland (Fig. 1), which is the Mesozoic continental core
of SE Asia (e.g., van Bemmelen, 1949; Hamilton, 1979). Java is
located on Sundaland’s southern margin.
Many studies of volcanic arcs are based on well-exposed
areas of arcs active long ago, such as the Cretaceous Allistos
Arc in Baja California (Busby et al., 2006) and the Cambrian–
Ordovician Lough Nafooey Arc of the Irish Caledonides (e.g.,
Draut and Clift, 2001). There are few studies of younger arcs
in areas where active volcanism continues today. One reason for
this is that younger arc products cover older arc sequences. Fur-
thermore, many modern arcs are in tropical areas, such as Indo-
nesia, the Philippines, and the Western Pacific, that are difficult
to work in, commonly are not well exposed, and are remote. The
modern volcanoes on Java form part of the Sunda Arc, which is
well known for volcanic activity, including the famous nineteenth
century eruptions of Krakatau and Tambora, and the Pleistocene
eruption of Toba. However, despite a long volcanic record in
Java, owing to subduction of Indian Ocean lithosphere during the
Cenozoic, little is known of its pre-Pleistocene arc history.
Since the middle Eocene (ca. 45 Ma) there has been north-
ward subduction of the Indo-Australian plate at the Java Trench
to the south of Java (Hall, 2002). Consequently, Java is essen-
tially a volcanic island, and a long history of arc volcanism is
recorded in its Cenozoic stratigraphy. Products of both active and
ancient volcanic arcs can be observed. An east-west–trending
chain of >30 modern volcanoes, forming part of the Sunda Arc,
Malaya
SOUTH
CHINA
SEA
0° Borneo
Sumatra
Ja
va
SUNDALAND
Tre
JAVA
n
ch
SEA
INDIAN Java
OCEAN
10°S
100°E 110°E
Figure 1. Current plate tectonic setting of western Indonesia. The mar-
gin of Sundaland (Hamilton, 1979) is defined by a black dashed line.
The black arrows show relative plate motion of boundaries—upper
plate movement relative to lower (McCaffrey, 1996). The study area is
enclosed by a rectangle.
creates the central spine of the island of Java (Fig. 2). An arc of
older, Eocene to Miocene volcanoes is parallel to and south of the
Sunda Arc and is known as the Southern Mountains Arc (Smyth
et al., 2005).
Exposures of basement rocks are rare in East Java, and
before this study the basement was thought to have formed dur-
ing the Cretaceous, when fragments of arc and ophiolitic mate-
rial were accreted to the southern margin of Sundaland (Wakita,
2000). There are no continental basement rocks at the surface
or reported from drilling in East Java. The basement is often
referred to as “transitional,” as to the north and west in Sunda-
land the basement is continental, and in the east near to Flores, it
is oceanic (e.g., Hamilton, 1979).
The Cenozoic stratigraphic record preserved on land in East
Java provides an excellent opportunity to examine the eruptive
history of a young, well-preserved volcanic arc. The present-
day arc, active since the middle to late Miocene (ca. 12–10 Ma;
Soeria-Atmadja et al., 1994), is immediately obvious, but the
importance of the older Southern Mountains Arc has previously
been underestimated. The relatively basic volcanic products of
this arc, the “Old Andesites” (van Bemmelen, 1949), are widely
known because they are well exposed and form prominent topo-
graphic features. However, the acidic products of the arc, which
are widespread, have been largely overlooked. As a result of the
work reported here, abundant quartz-rich sedimentary rocks in
East Java are now known to be the primary and epiclastic prod-
ucts of acidic volcanism but were previously interpreted as conti-
nental sediments (e.g., Harahap et al., 2003). The acidic nature of
the erupted material is a reflection of the character of the underly-
ing crust, which as a result of new zircon dating is now thought to
include a fragment of Archean continental crust.
The full cycle of the Paleogene Southern Mountains Arc,
from initiation to termination, can be documented in the stratigra-
phy of East Java. This paper reports the stages of arc development
and the sequence and timing of events within the arc, determined
by detailed field investigations accompanied by provenance anal-
ysis and isotopic dating. Most of the middle Miocene was marked
by a period of volcanic quiescence that followed the termination
of Paleogene arc volcanism in the early Miocene (ca. 20 Ma).
Near the end of the middle Miocene (ca. 12–10 Ma) the modern
Sunda Arc began activity in its present position, 50 km north of
the Southern Mountains Arc. Possible explanations for the shift
in position of arc volcanism are discussed later in this paper.
PRESENT STRUCTURE OF EAST JAVA
East Java can be subdivided into three parts, broadly par-
allel to the elongation of the island, representing (1) the early
Cenozoic Southern Mountains Arc, (2) a deep basin north of the
arc, and (3) a marine shelf north of the basin (Figs. 3 and 4). The
modern arc is built mainly on top of the basin north of the early
Cenozoic arc. We first summarize the principal features of these
components of Java and then go on to discuss the stratigraphy of
the Southern Mountains Arc.
 Cenozoic volcanic arc history of East Java, Indonesia 201

110o E o
112 E EAST JAVA SEA
o
114 E A
 Muria
igh
ang H
Remb
o
7 S
MADURA
Dieng Plateau
Kendeng Hills Surabaya

Karangsambung Madura Straits

Nanggulan EAST JAVA
Jiwo Hills

o
8 S
Yogyakarta Southern Mountains

0 50 100
INDIAN OCEAN
Kilometers

o
110 E
o
112 E
o
114 E B

o
7 S

Merapi

o
8 S

0 50 100
Kilometers

Figure 2. East Java location map. (A) Important locations and topographic features. The black star marks the epicenter of the 27 May 2006
earthquake, and the currently active volcano Merapi is marked by the black triangle. (B) Digital elevation model from SRTM data (Shuttle Radar
Transect Mission, NASA) of East Java, with the centers of Oligocene–Miocene volcanism marked.

Southern Mountains Arc
There are few exposures of basement rocks in Java, and in
East Java they are known only in the western part of the study
area. Based upon this limited evidence, the basement has been
interpreted to be arc and ophiolitic rocks of Cretaceous age. A
volcanic arc was built upon basement rocks from the middle
Eocene to the Miocene in southern Java (Smyth, 2005), which
is known to extend from East Java into West Java (Soeria-
Atmadja et al., 1994). The stratigraphic thickness of the arc
202 Smyth et al.

110o E 112o E 114o E

A
Java Sea


Section line
o
7 S SUNDA SHELF
Madura Island

KENDENG BASIN

Merapi
CENTRAL JAVA
PROVINCE

MODERN SUNDA ARC

8o S
SOUTHERN MOUNTAINS
ARC
Indian Ocean

0 50 100 B
EAST JAVA
Kilometers FORE-ARC BASIN

EOCENE TO MIOCENE
SOUTHERN
FORE-ARC
SUNDA SHELF KENDENG BASIN MOUNTAINS
BASIN
ARC
A B
?? ??
0 50 100
Km

Figure 3. Structure of East Java in map-and-sketch Eocene to Miocene profile, showing the three structural provinces—Southern Mountains Arc,
Kendeng Basin, and the edge of the Sunda Shelf—and the modern Sunda Arc building on top.

products is >2500 m, and within this sequence andesites are well
known (van Bemmelen, 1949; Soeria-Atmadja et al., 1994), but
acidic volcanic rocks have not been previously reported. The
zone containing the arc products is 50 km wide. The Southern
Mountains Arc is now uplifted and partially eroded. The strata
typically dip uniformly toward the south between 20° and 30°.
Kendeng Basin
The Kendeng Basin lies directly behind and to the north of
the Southern Mountains Arc (Fig. 2). The deposits of the basin are
poorly exposed. The depocenter is marked by a strong negative
Bouguer gravity anomaly of more than −580 μm–2 (Fig. 5). The
negative anomaly can be traced eastward into a negative marine
free air anomaly (Sandwell and Smith, 1997), which extends from
the Straits of Madura eastward to the north of Bali. At the west
end of the Kendeng Basin a relatively abrupt change in the charac-
ter of the anomaly is around the modern volcanoes of Merapi and
the Dieng Plateau (Fig. 5). To the west, in West Java, the anomaly
is not well defined, as it becomes positive (40 μm–2).
No basement rocks are exposed or known from drilling in this
region. The basin fill has an age range of middle Eocene to Mio-
cene, similar to the Southern Mountains Arc (de Genevraye and
Samuel, 1972). The basin is east-west oriented, at least 400 km
long, parallel to the Southern Mountains Arc, and filled with a
succession of volcaniclastic turbidites and pelagic mudstones that
are reported to be at least 6 km thick (Untung and Sato, 1978).
Gravity calculations indicate that the basin may contain as much
as 10 km of sediment (C.J. Ebinger, 2005, personal commun.).
Based on field observations, seismic sections, and regional gravity
interpretation (Smyth, 2005), the basin is estimated to have been
~100–120 km wide during the early Cenozoic. It is now partially
exposed at the surface in the Kendeng Fold-Thrust Belt, where
there is an estimated 10–30 km of shortening (de Genevraye and
Samuel, 1972; Smyth, 2005); de Genevraye and Samuel (1972)
report that deformation commenced in the Pliocene.
 Cenozoic volcanic arc history of East Java, Indonesia 203

110o E 112o E 114o E

Rembang Java Sea

Muria 0 50 100
Kilometers

7 S
o SUNDA SHELF Madura Island
Semarang
(geology not shown)
Dieng
Plateau
Surabaya
KENDENG BASIN

Merapi

Karangsambung
8o S
SOUTHERN MOUNTAINS
Yogyakarta
ARC
Indian Ocean
Jiwo Hills Pacitan
Southern Mountains Arc

Intrusive rocks Kendeng Basin

Kendeng

Modern volcanic and alluvial

Basin

Middle Miocene Carbonates Approximate southern deposits

Sunda Shelf

Middle Miocene reworked limit of the Kendeng

Basin (not exposed) Pliocene clastic and
volcaniclastic deposits
carbonate rocks


Oligo-Miocene volcanics
Modern Arc

Post-peneplane carbonate Fault

Eocene Sedimentary rocks Sunda Arc rocks
Basement rocks - Modern volcanic Fault (inferred)
Shelf edge clastic and
Cretaceous and older centers carbonate rocks

Figure 4. Simplified geological map of East Java, showing the main geological provinces and stratigraphic units.

Edge of Sunda Shelf
To the north of the Kendeng Basin (Figs. 2 and 4) the hills of
North Java and the offshore East Java Sea constitute an area inter-
preted to be the edge of the early Cenozoic Sunda Shelf (Hamil-
ton, 1979). This region has been the focus of most hydrocarbon
exploration. Pre-Cenozoic basement rocks sampled by drilling
are known to be ophiolitic and arc rocks, which include chert and
basic volcanic and metasedimentary rocks (e.g., Hamilton, 1979).
Basin development began in the Eocene. There are between 2000
and 6000 m of Eocene to Pliocene shallow marine clastic and
extensive carbonate sedimentary rocks within fault-controlled
basins (e.g. Ardhana, 1993; Ebanks and Cook, 1993).
The sedimentary sequences have been deformed since the
late Miocene by numerous open, east-west–oriented folds; north-
ward-verging, east-west–oriented thrusts interpreted to be Plio-
cene in age; and ENE-WSW normal faults (Chotin et al., 1984;
Hoffmann-Rothe et al., 2001).
Modern-Day Volcanic Arc
The active volcanoes of the Sunda Arc are built mainly on
the Kendeng Basin (Figs. 3 and 4) but locally overlap the edge
of the Southern Mountains Arc. Arc activity commenced in
its present position at ca. 10 Ma in the late Miocene (Soeria-
Atmadja et al., 1994). The volcanoes exceed 3000 m in elevation
and form the central spine of the island (Fig. 2). The average com-
position of the present-day volcanic products is basaltic andesite
(Nicholls et al., 1980), which is much more basic than the average
composition of the Southern Mountains Arc products. Most of
the active volcanoes are situated ~100 km above the subducting
slab (England et al., 2004). There are also a number of unusual
K-rich backarc volcanoes (Edwards et al., 1991), which occur to
the north of the axis of the arc, including Muria (Fig. 2). The erup-
tive and epiclastic products of the Sunda Arc cover a significant
proportion of the island, contributing to its fertile soils.
STRATIGRAPHIC RECORD OF THE EARLY
CENOZOIC SOUTHERN MOUNTAINS ARC
The sedimentary rocks in the Southern Mountains Arc in
East Java were deposited on the basement above a poorly dated
regional unconformity (Fig. 6). The unconformity separates
Upper Cretaceous basement rocks and the Cenozoic succession,
suggesting a long period, potentially up to 30 m.y., during which
uplift and erosion occurred. The stratigraphic succession above
204
 -500 to -250
7o S SUNDA SHELF
Dieng
Plateau
75
0m
500
m
250m
o
8 S Merapi
SOUTHERN MOUNTAINS ARC
0 50 100
Kilometers
o o
110 E
Figure 5. Bouguer gravity anomaly map of East Java. The map also shows topographic contours at 250 m intervals.
the unconformity in the Southern Mountains Arc and the Ken-
deng Basin has been subdivided into three “synthems” (Smyth
et al., 2005), each representing a different period of arc activity
(Fig. 6). The term synthem has been used because much of the
work undertaken and published in East Java has been hydrocar-
bon oriented, and as a result the term sequence is often used with
seismic and sequence stratigraphic implications (e.g., Catuneanu,
2006). We have therefore chosen to avoid the term sequence for
the major stratigraphic subdivisions and in this study used the
term synthem, defined as an unconformity-bounded stratigraphic
package (Rawson et al., 2001). The three synthems are:
• Synthem One: records the initiation of arc volcanism and
the early stages of arc development during the middle
Eocene (ca. 45 Ma) to early Oligocene (ca. 34–28 Ma).
• Synthem Two: records the growth and termination of arc vol-
canism in the Southern Mountains Arc during the late Oli-
gocene (ca. 28–23 Ma) to the early Miocene (ca. 20 Ma).
• Synthem Three: records widespread carbonate growth,
accompanied by the erosion and redeposition of rocks
from earlier synthems during the middle Miocene (ca. 20–
10 Ma), with no significant volcanic activity.
In the account of the stratigraphy that follows, the strati-
graphic ages have been converted to numerical ages, using the
Gradstein et al. (2004) time scale.
Synthem One: Initiation of the Southern Mountains Arc
Most parts of the Southern Mountains rocks of Synthem
One are covered by younger deposits and are exposed only at
Smyth et al.
-750 to -500
-250 to 0
0 to 250
250 to 500
500 to 750
750 to 1000
1000 to 1250
1250 to 1500
1500 to 1750
KENDENG BASIN 1750 to 2000
µms-2
o
112 E 114 E
Karangsambung, Nanggulan, and the Jiwo Hills (Fig. 2). Syn-
them One is not exposed in the Kendeng Basin but was sampled
in blocks brought to the surface by mud volcanoes.
Southern Mountains Arc
Basement exposures in East Java are rare and are found only
in the western part of the area at Karangsambung and the Jiwo
Hills (Fig. 2). The exposed basement rocks are Cretaceous in age
(Parkinson et al., 1998; Wakita and Munasri, 1994) and include
basaltic pillow lavas, radiolarian cherts, various metasedimen-
tary lithologies, quartz-mica schist, and high-grade metamor-
phic rocks including jadeite-quartz-glaucophane–bearing rocks
and eclogites. Quartz veins are commonly observed within the
basement exposures and have not been identified in the overly-
ing Cenozoic rocks. The high-grade metamorphic rocks have
been interpreted to indicate subduction zone metamorphism
(Miyazaki et al., 1998). The basement rocks are interpreted to
represent fragments of arc and ophiolitic material accreted to the
margin of Sundaland during the Late Cretaceous. Similar rocks
have previously been assumed to extend beneath the rest of East
Java, where there is no information from surface exposures or
drilling (e.g., Hamilton, 1979; Parkinson et al., 1998).
Synthem One is ~1000 m thick but is exposed only in the
western part of the Southern Mountains Arc in East Java. The
oldest sedimentary rocks rest directly on the basement above a
regional angular unconformity. They are poorly dated fluvial con-
glomerates and interbedded sandstones (Figs. 7A and 8) and are
at least 50 m thick, but their total thickness is difficult to assess
owing to limited and patchy exposures. These are the only rocks
 TECTONIC ENVIRONMENT SEDIMENT
Age (Ma) SETTING OF DEPOSITION CHARACTER
0
KEY

VOLCANIC ACITIVITY
PLIOCENE Active arc volcanism Volcaniclastic turbidites

RESURGENCE OF
UPLIFT Erosion or non deposition Crystal-rich volcaniclastics
AND
Subsidence Proximal volcanic rocks
5 EROSION
Super-eruption Siliciclastic sediments
Reefal limestones Basal conglomerates
Foraminiferal packstones Unconformity
10
CARBONATE ROCKS
SYNTHEM THREE

PLATFORMS AND
TURBIDITE FANS
EPICLASTIC

CARBONATE
REWORKING OF
EXTINCT VOLCANIC
VOLCANICS +
REWORKED

ARC AND
15 WIDESPREAD
MIOCENE

CARBONATE
DEPOSITION

Fig.11A
SEMILIR ERUPTION
20
EXPLOSIVE VOLCANISM

TERRESTRIAL-MARINE
VOLCANIC DEPOSITS

VOLCANIC ISLANDS
SYNTHEM TWO

IN TERRESTRIAL-
SHALLOW MARINE-
OLIGO-MIOCENE

EXPLOSIVE SLOPE. SOME Fig.10

VOLUMINOUS EPICLASTIC
25 REWORKING.
VOLCANISM
OLIGOCENE

DEPOSITION IN THE
ARC, INTRA-ARC,
FORE-ARC AND
BACK-ARC.

30
EUSTATIC EVENT APPARENT GAP IN SECTION
NO EVIDENCE OF
DEFORMATION

Fig.7C
TERRESTRIAL-SHELF-BASIN

35
EARLY BASIN SEQUENCE

OPEN - DEEP
TRANSGRESSION

MARINE
EOCENE
SYNTHEM ONE

40 SLUMPING
VOLCANIC
ACTIVITY SHALLOW MARINE

Fig.7B
SUBSIDENCE DELTAIC
45
Fig.7A
NO VOLCANIC TERRESTRIAL
ACTIVITY

~30 m.y. time gap

CRETACEOUS

CRETACEOUS
+ OLDER
BASEMENT

Figure 6. Simplified stratigraphic column of the Southern Mountains Arc. The column shows the three synthems men-
tioned in the text and indicates the phases of arc development recorded by the sedimentary succession.
206 Smyth et al.

A Basal part of Synthem One

Clay Silt Sst Cong
B Lower part of Synthem One
Clay Silt Sst Cong
C Upper
Clay
part of Synthem One
Silt Sst Cong
1.7 TOP

10 1.6

3
1.5

9
1.4
Thickness (m)

Lithology Logs A and B

Thickness (m)

Thickness (m)
Organic-rich muds
8 2 Quartz-rich sandstones
+ siltstones

1.2 Conglomerates
Lithology Logs C

Increasing volcanic material

7 Crystal-rich
1.1 volcaniclastic sandstone
Tuffaceous mudstones Figure 7. Stratigraphic logs illustrating
2 + siltstones features of the type sections of Syn-
1m
Sedimentary structures them One. (A) Basal conglomerates
6
and sandstones of the Lukulo Member,
Rip up clasts
0.9 Karangsambung Formation. (B) Quartz-
Parallel lamination rich sandstones and interbedded organ-
5 0.8
Cross lamination ic-rich muds of the middle Eocene Kali
Trough cross Songo Member, Nanggulan Formation.
lamination Volcanic components increase upsec-
Convolute lamination
0.7 tion. (C) Volcanogenic sandstones and
Channels tuffaceous mudstones of the Jetis Mem-
4
0.6
Irregular erosive ber, Nanggulan Formation.
bed base
Planar sharp bed base
1
0.5 Conglomerate
3

0.4

2
0.3

0.2
1

0.1

0 0 0
BASAL TERRESTRIAL SANDSTONES + TERRESTRIAL - SHALLOW MARINE MARINE VOLCANOGENIC
CONGLOMERATES SANDSTONES + ORGANIC MUDSTONES SANDSTONES + MUDSTONES
o, o
Loc. 7.5477 S, 109.6694 E Loc. 7.7323 S, 110.1972 E Loc. 7.7395 S, 110.1887 E

exposed onshore that contain no fresh volcanic material (Fig. 9). and mudstones (Fig. 8). The middle Eocene age of the lower
They are dominated by quartz grains and metamorphic and igne- part of the succession is based on palynomorphs in the coals
ous lithic clasts (Figs. 8 and 9), including vein and metamorphic and organic-rich mudstones (Lelono, 2000), and the occurrence
quartz, chert, phyllite, schist, metasedimentary rock, and basalt. of Nummulites and Discocyclina higher in the section. The
The lithologies identified as clasts are typical of those exposed in lower part is at least 200 m thick and is a noncalcareous unit of
the Cretaceous basement (described above) and are interpreted to well-bedded coals, conglomerates, quartz-rich sandstones, and
be the product of erosion and reworking of these basement rocks. organic-rich muds. The coals and conglomerates are restricted
The terrestrial conglomerates and sandstones lack palynomorphs to the lower 50 m. The conglomerates are commonly channel-
and so cannot be directly dated, but they are overlain by a succes- ized, are ~10–75 cm thick, and contain a range of lithic clasts
sion of well-dated middle Eocene strata (Lelono, 2000). similar to those identified in the basal section directly overly-
The middle Eocene to lower Oligocene of the Southern ing the basement. The coals vary laterally in thickness from 5
Mountains Arc is represented by a transgressive succession, from to >50 cm and have an average vitrinite reflectance of 0.4% Ro
base to top, of coals and conglomerates, sandstones, siltstones, (Smyth, 2005). For normal and high geothermal gradients typical
 Cenozoic volcanic arc history of East Java, Indonesia 207

A WEST EAST B
Foraminiferal packstone
30 750 Slump - schist and
foraminifer mixture
OLIGOCENE

Volcaniclastic sandstones
Conglomerate clast
Quartz-rich sandstones lithology types
Quartz
Basal Conglomerate Chert
Phyllite
35
Basalt
500
Metasediments

C D
40
EOCENE

250
Thickness (m)

45

Cretaceous and older basement lithologies

E F

Figure 8. Field observations of Synthem One. (A) Sketch stratigraphy of the middle Eocene–early Oligocene deposits. Volcanic material has
increased upsection from ca. 42 Ma. (B) Field photograph of the basal conglomerate. The pie chart shows the abundance of quartz and lithic
clasts. These conglomerates contain no acidic volcanic material. (C) Quartz-rich sandstones and coals from the lower part of Synthem One. (D)
Interbedded quartz-rich sandstones and organic-rich and tuffaceous mudstones, upsection from photo C. The sandstones contain a significant
percentage of volcanic quartz, plagioclase feldspar laths, and pyroclastic zircons. (E) Volcanogenic mudstone rich in planktonic foraminifers. (F)
Volcanogenic sandstones with abundant volcanic lithic fragments.

of arc regions this would imply very shallow burial depths of <1
or 2 km (Madon et al., 1997; Watts, 1997).
Above this transgressive succession lie thinly laminated
quartz-rich sandstones and organic-rich mudstones that are at
least 150 m thick. These sediments were deposited in a ter-
restrial setting with some marine incursions indicated by the
presence of calcareous zones rich in Nummulites and Discocy-
clina. In addition to the organic-rich mudstones there are sev-
eral gray mudstones, which are rich in smectite. The quartz-rich
sandstones contain metamorphic lithic clasts, like those of the
conglomerates, but also contain volcanic lithic clasts such as
pumice and andesite. In addition, a significant proportion of
the quartz has a volcanic origin (the identification of volcanic
quartz is discussed further below). At the base of the quartz-
rich sandstones the quartz is dominated by metamorphic grains,
but this content gradually decreases in its relative contribution
upsection, and at the top, volcanic quartz is dominant (Fig. 9B).
The heavy mineral assemblage contains zircons that yielded
spot sensitive high-resolution ion microprobe (SHRIMP) U-
Pb ages of 41.8 ± 1.6 and 42.7 ± 1.5 Ma (Smyth, 2005). These
ages are the same as the biostratigraphically determined ages
for the host rocks, indicating that volcanic activity occurred
208 Smyth et al.

Basement
A B TOP

80 20

60 40 Up-section
increase in volcanic
quartz within the
quartz-rich
sandstones of
40 60 Synthem One (Fig
7B), accompanied
by a decrease in
metamorphic
quartz.
20 80

80 60 40 20 Other
Volcanic
sources

BOTTOM
Pre-Middle Eocene basal Late Eocene-Oligocene
conglomerates and sandstones volcaniclastic deposits
(Fig. 7A) (Fig. 7C)
Thin section analysis of quartz type %
Middle Eocene quartz-rich Up-section increase in volcanic Metamorphic Sedimentary
sandstones (Fig. 7B) material Volcanic Unknown
Plutonic

C D

Figure 9. Petrographic details of Synthem One. Scale bars are 1 mm. (A) Basement, volcanic, and other sources are de-
picted in a triangular diagram, showing the upsection trend identified in East Java, with a reduction in basement-derived
material and an increase in volcanic contribution. (B) Pie charts showing the quartz types identified within quartz-rich
sandstones of East Java; example from the Nanggulan Formation. Volcanic quartz increases upsection. (C) Photomi-
crograph of the basal sandstone, dominated by metamorphic quartz. (D) Volcaniclastic turbidite dominated by laths of
plagioclase feldspar.

as the sediments were deposited. Higher in the section, in the
sandstones, a number of Nummulites-rich zones mark the onset
of fully marine conditions.
Above the nummulitic units the sandstones become
increasingly arkosic and rich in fresh laths of plagioclase feld-
spar (Fig. 9D), volcanic quartz, volcanic lithic fragments, elon-
gate volcanic zircons, and volcanic clays such as smectite and
zeolites (Fig. 9A). These sandstones are interbedded with tuff-
aceous mudstones. Water depth is uncertain but is interpreted
to be deeper than that for the nummulitic zones. There are no
diagnostic sedimentary structures, but the sandstones contain
pelagic foraminifers.
Higher in the section the presence of volcaniclastic turbi-
dites indicates an increase in water depth. Some turbidites are
 Cenozoic volcanic arc history of East Java, Indonesia
characterized by Bouma divisions A, C, and E, but more com-
monly C, D, and E. They are >150 m thick, with individual
bed thicknesses ranging from 5 to 50 cm. The turbidites have a
diverse planktonic foraminiferal assemblage (Fig. 8E), includ-
ing Helicosphaera euphratis, H. reticulate, and H. wolcoxonii
(P. Lunt, 2002, personal commun.), which provides an age of
NP18 (36.8–36.2 Ma) for the lower parts of the succession. The
turbidites extend into the lower Oligocene, based on biostratig-
raphy (P. Lunt, 2002, personal commun.). They are dominated
by volcanic debris (Fig. 9) such as laths of plagioclase feldspar,
volcanic lithic clasts, and volcanogenic clays, and there are no
metamorphic lithic clasts within the coarser beds. The heavy
mineral assemblage is dominated by volcanic zircons, which
yielded a weighted mean SHRIMP U-Pb age from 17 grains of
41 ± 1.4 Ma, similar to the biostratigraphic age of the lower part
of the section, suggesting some reworking.
The upper boundary of Synthem One is an intra-Oligo-
cene unconformity, which is interpreted to be the result of sea
level change, as the sedimentary rocks directly above and below
the gap have similar bedding orientations with no indication of
deformation. This could be a local sea level change but could
also be a global change. Unconformities of this age that record a
global intra-Oligocene sea level fall are widely known, from the
Haq et al. (1987) curve (30 Ma), the Marshall Paraconformity
(32–29 Ma) in the Canterbury Basin, New Zealand (Fulthorpe et
al., 1996), and the carbonates of Baldwin County, Alabama, USA
(Baum et al., 1994).
Within Synthem One the contribution of volcanic mate-
rial increases upsection as the proportion of basement-derived
material (metamorphic quartz, metamorphic lithic fragments,
and illite, chlorite, and serpentinite clays) decreases, and the
upper Eocene sedimentary rocks are almost entirely dominated
by volcanic debris (Fig. 9A, B). The volcanic centers supplying
this material cannot be separately mapped owing to their limited
exposure, but they are interpreted to follow the same trend and
occupy the same positions as the volcanoes of the Oligocene–
Miocene arc (see below).
Kendeng Basin
In the Kendeng Basin there are limited exposures, and Syn-
them One is not seen at the surface, but blocks of the older litholo-
gies have been brought to the surface by Pleistocene and modern
mud volcanoes. These blocks are terrestrial and are composed
of shallow marine sandstones and conglomerates (de Genevraye
and Samuel, 1972) similar in character to the middle Eocene sed-
imentary rocks in the Southern Mountains Arc.
Synthem Two: Growth and Catastrophic Termination of
Arc Volcanism
Southern Mountains Arc
The upper Oligocene to lower Miocene deposits of Synthem
Two are primary volcanic rocks and epiclastic rocks. These rocks
are exposed extensively throughout the Southern Mountains Arc
209
(Fig. 4) and within the fold-thrust belt of the Kendeng Basin. The
oldest rocks of this synthem in the Southern Mountains Arc are
reworked bioclastic tuffaceous mudstones dated biostratigraphi-
cally as Oligocene, NP24 (27.3–30 Ma; M. Fadel, 2002, personal
commun.).
Throughout the late Oligocene and early Miocene the vol-
canic activity in the Southern Mountains Arc was extensive,
explosive, and intermediate to acidic in composition. The depos-
its range from andesite to rhyolite, with an average SiO2 con-
tent of 67 wt% (Smyth, 2005), and include thick mantling tuffs
(Fig. 10A), crystal-rich tuffs, block and ash flows, pumice-lithic
breccias, andesitic breccias (Fig. 10B), silicic lava domes, and
lava flows. The Oligocene–Miocene volcanic centers can be
mapped (Smyth, 2005) by the occurrence of vent proximal facies,
and at least 13 centers are presently exposed (Fig. 2), which show
a similar spacing to the volcanoes of the present-day arc. The
thickness of the proximal volcanic deposits ranges from 250 to
>2000 m.
Thick (>700 m) successions of volcaniclastic sedimentary
rocks surround the volcanic centers. These reworked deposits are
commonly interbedded with beds of unreworked mantling tuffs
and volcanic breccias >1 m in thickness. The reworked volcani-
clastic beds vary in thickness from 5 to >100 cm and are crystal-
and volcanic lithic-rich sandstones and tuffaceous mudstones.
They commonly contain abundant fragments of charcoal, indi-
cating the presence of vegetated slopes on the terrestrial volcanic
centers. Both terrestrial and shallow marine deposits have been
identified; the latter do not contain abundant bioclastic mate-
rial but are weakly calcareous, and their upper bedding surfaces
are commonly intensely bioturbated with Cruziana-type facies
traces. There are slump folds and mass wasting deposits on the
flanks of the volcanic centers.
In the Southern Mountains Arc is a record of a major erup-
tion, the Semilir Eruption, toward the end of the period of arc
activity. Extensive deposits of this eruption are widespread to
the east of Yogyakarta (Fig. 4). The Semilir and Nglanggran
Formations (Fig. 10A, B) are the products of this event, and
they were deposited in a short period, possibly during one
eruptive phase between 21 and 19 Ma (Smyth, 2005). Based on
measured sections, the combined thickness of the two forma-
tions varies between 250 and 1100 m. They are well exposed
over an area of 800 km2, and the total volume of volcanic mate-
rial is estimated to be at least 480 km3. The Semilir Formation
(Fig. 6) is a thick accumulation of dacitic air-fall, pyroclastic
surge and flow deposits produced by an explosive eruption.
Both terrestrial and shallow marine deposits have been identi-
fied, indicating that the erupted material entered the sea from
the flanks of the volcanic center. The Nglanggran Formation
(Fig. 6) is a series of monomict andesitic volcanic breccias.
The individual beds are up to 10 m thick, have flat bases and
tops, and can be mapped for tens of kilometers. The breccias
contain some blocks >3 m across. They are interpreted as vent
proximal facies such as flow breccias or block-and-ash–flow
deposits. These deposits mark the end of volcanism in the
210 Smyth et al.

A B

Basement
C
D

80 20

60 40
Quartz Type %
Volcanic Unknown

40 60
SOUTHERN MOUNTAINS ARC
Kebobutak
Jaten
20 80

Volcanic 80 60 40 20 Other
Sources

Figure 10. Details of Synthem Two. (A) Field photograph of thick, well-bedded ashes (bar is 1m). (B) Field photograph
of monomict andesitic breccia; the breccia clasts commonly exceed 1 m in size (bar is 1m). (C) Basement, volcanic, and
other sources are depicted in a triangular diagram, showing a selection of formations from Synthem Two. These deposits
lack material of basement origin. (D) Pie chart showing the quartz types identified petrographically in a quartz-rich sand-
stone in the upper part of Synthem Two. The quartz is dominated by volcanic quartz; the image is a bipyramidal grain
from the Jaten Formation, Pacitan (bar is 1m).

Southern Mountains Arc. There is no significant break within
or between the Semilir and Nglanggran Formations, based on
zircon dating and biostratigraphic dating of the overlying for-
mations (Smyth, 2005).
A number of quartz-rich sandstones (Fig. 10D) are in the
upper part of Synthem Two in the Southern Mountains Arc
(Smyth, 2005). Volcanic zircons from the sandstones form a
single population with ages that are the same as that of the
Semilir Formation (Smyth, 2005) and are contemporane-
ous with the Semilir Eruption. The quartz within these sand-
stones is entirely of volcanic origin (Fig. 10D). Volcanic quartz
grains commonly appear clear and bright in thin section, have
 Cenozoic volcanic arc history of East Java, Indonesia
nonundulose extinction, and are monocrystalline (Leeder,
1982). These features and other characteristics, including
well-developed crystal faces, bipyramidal shape, melt embay-
ments, and melt inclusions, can be used to distinguish volcanic
quartz from other types such as metamorphic, vein, plutonic,
and sedimentary quartz. Volcanic crystal-rich sandstones can
be produced by primary eruptive mechanisms and/or second-
ary epiclastic processes. The sandstones and other quartz-rich
volcaniclastic rocks in East Java previously were interpreted
as continental siliciclastic deposits (Harahap et al., 2003), and
it is for this reason that many of the acidic products of the
Southern Mountains Arc have been overlooked (as discussed
below). Here they are interpreted to have a volcanic origin and
to be largely the product of the Semilir Eruption with some
subsequent reworking.
Kendeng Basin
Synthem Two exposures in the Kendeng Basin are very lim-
ited, occurring only in a small thrust-bound sliver. They comprise
poorly lithified Globigerina-rich tuffaceous mudstones ranging
in age from late Oligocene to early Miocene (de Genevraye
and Samuel, 1972). The mudstones are at least 85 m thick (de
Genevraye and Samuel, 1972) but have limited surface exposure.
The mudstones are volcanogenic and apparently lack continen-
tal terrigenous material. The volcanic material is fine grained,
reworked, and distal in character, very different from that identi-
fied within the arc at this time. The abundance of Globigerina
and other planktonic foraminifers is indicative of pelagic sedi-
mentation in an open marine setting at water depths of a few hun-
dred meters or more. Thick volcanogenic turbidites are reported
from wells in this area (P. Lunt, 2002, personal commun.), but no
descriptions of the stratigraphy or well log interpretations have
been published.
Edge of the Sunda Shelf
The offshore Eocene to Pliocene successions in the East Java
Sea that have been investigated during hydrocarbon exploration
are not reported to contain any volcanic material (e.g., Matthews
and Bransden, 1995). Field investigations during this study in
northeast Java have identified distal volcanic material on the
edge of the Sunda Shelf within carbonates, including thin tuff
layers, zones rich in smectite clays, and concentrations of volca-
nic quartz and zircons. This suggests that there may be more vol-
canic debris offshore than recognized up to now, and that some of
the clay layers may be fine air-fall ash deposits.
The stratigraphic record of Synthem Two within the South-
ern Mountains Arc and Kendeng Basin accounts for only vol-
canic and volcaniclastic rocks at this interval. There is no evi-
dence of input of material from basement or other sources. This
indicates that the volcanic arc was the only source of sediment
at this time. There are no significant exposures of carbonates
within Synthem Two in the Southern Mountains Arc or Kend-
eng Basin, but carbonates do occur farther to the north on the
edge of the Sunda Shelf.
211
Synthem Three: Reworking of the Southern Mountains Arc
Southern Mountains Arc
In the Southern Mountains, to the south of the Oligocene–
Miocene volcanic centers, extensive calcareous volcanogenic
turbidites (Fig. 11) occur with Bouma divisions A, C, D, and E.
The turbidites are at least 500 m thick and have been dated within
nannofossil subzones NN2–NN8 (Kadar, 1986) that correspond
to dates between 19 and 10 Ma. Beds vary in thickness from 20
to 75 cm, and upper bedding surfaces are commonly bioturbated
with traces of Cruziana facies. Flute casts indicate that the flow
direction was toward the southeast, and there are slump folds
with southeasterly vergence. Thin section examination shows
that volcanic crystals and lithic fragments are well rounded, and
there is no fresh or unreworked volcanic debris present. Dating of
zircons (see below) supports the field and petrographic evidence
that these are reworked volcaniclastic deposits, as the zircon ages
are similar to the age range of rocks in Synthem Two. Synthem
Three is therefore interpreted to be the product of reworking of
older arc rocks of Synthem Two rather than the product of con-
temporaneous volcanism.
Several tuff beds have been identified at the top of the tur-
bidite sequence in the Southern Mountains but are not the result
of volcanic activity in the Southern Mountains Arc. The tuffs are
distal air-fall deposits and yield zircons with ages between 12
and 10 Ma (P.J. Hamilton, 2003, personal commun.). These tuffs
are the product of the modern Sunda Arc (Soeria-Atmadja et al.,
1994), and the ages mark the initiation of volcanic activity some
50 km to the north of the Southern Mountains Arc.
In addition to the turbidites, there are the first widespread
carbonates in the Southern Mountains (Fig. 11D). The carbon-
ates range in age from late early Miocene to middle Miocene
(Lokier, 2000; Smyth, 2005) and formed isolated reefs and exten-
sive carbonate platforms, which can be observed overstepping
the deposits of Synthem Two. The limestones are at least 200 m
thick and are the source of the carbonate within the volcanogenic
turbidites.
Kendeng Basin
Volcanogenic sedimentary rocks identified in the Kendeng
Basin include channelized volcanic lithic conglomerates, crystal-
rich and volcanic lithic-rich sandstones, tuffaceous mudstones,
and Globigerina mudstones. A thickness of up to 400 m of rocks
is exposed (Smyth, 2005), and hydrocarbon exploration shows
there is up to 3000 m in the basin (de Genevraye and Samuel,
1972). The sandstones contain Bouma divisions B, C, and E and
are locally cut by the channelized conglomerates. These conglom-
erates can cut as much as 20 cm into the underlying beds and are
up to 1 m thick. Slump folds are common, and most verge toward
the northeast. Measurements of scours, grooves, channel struc-
tures, flutes, and slumping indicate that sediment was transported
from the south toward the north. The Globigerina mudstone is
similar in character to those of Synthem Two. A sample from
the top of the exposed section yielded a biostratigraphic age
212 Smyth et al.

A Clay Silt Sand Loc. 7.8727oS B

F M C 110.5496oE

20
Thickness (m)

Bioturbation
Asymmetrical ripples
Parallel lamination
Convolute parallel
lamination
Breccia
C
Pebbles
15 Figure 11. Stratigraphic details of Syn-
them Three. (A) Logged section typi-
cal of the volcanogenic turbidites of
Synthem Three, Sambipitu Formation.
Synthem Three

(B) Field photograph showing asym-

metrical ripples on the upper bedding
surfaces of the volcanogenic turbidites,
Sambipitu Formation. (C) Field photo-
graph of volcanogenic turbidites within
10 the Kendeng Basin, Kerek Formation.
(D) Field photograph of the reefal car-
bonates typical of Synthem Three in the
Southern Mountains Arc, Campurdarat
Formation.

D
5

Upper part of
0 Synthem Two

of foraminifer subzones N8–N9 (M. Fadel, 2003, personal com- in the East Java Sea (Bishop, 1980; van Bemmelen, 1949), and
mun.) corresponding to ages between 17.2 and 14.4 Ma. redeposited on the edge of the Sunda Shelf.

Edge of the Sunda Shelf INTERPRETATION OF THE EOCENE TO MIOCENE

On the edge of the Sunda Shelf are a number of lower to mid-
dle Miocene quartz-rich sandstones. They were previously inter-
preted to have a continental provenance and to have been derived
from reworking of basement rocks of Sundaland (e.g., Ardhana,
1993; Sharaf et al., 2005), but new studies (Smyth et al., 2007)
have shown that these sandstones contain a significant proportion
of volcanic quartz, volcanic zircons, and reworked volcanogenic
clays. These volcanic particles are interpreted to have fallen as
ash onto the Sunda Shelf, to have been subsequently reworked
and mixed with material derived from uplifted basement blocks
STRATIGRAPHY
Initiation of the Southern Mountains Arc
The first evidence of arc volcanism in East Java is identified
as middle Eocene at ca. 42 Ma with the first occurrence of tuff
layers, volcanic lithic fragments, volcanic quartz, laths of plagio-
clase feldspar, and volcanic zircons. Prior to this, the sedimentary
rocks, which are the oldest rocks exposed on East Java, lack evi-
dence of contemporaneous arc volcanism and are the only rocks
 Cenozoic volcanic arc history of East Java, Indonesia
exposed on East Java that contain no volcanic debris. This sug-
gests that there was no arc volcanism on Java prior to 42 Ma,
which also implies no subduction of the Indian-Australian plate
beneath Java at this time.
Growth and Development of the Southern Mountains Arc
Following the initiation of volcanism in the Southern Moun-
tains the contribution of volcanic debris increased rapidly, and
the contribution of basement-derived material decreased. The
volcanic centers probably formed an east-west chain of volcanic
islands separated by interarc basins, much like the present Izu-
Bonin-Mariana and Aleutian Islands Arcs. A narrow volcaniclas-
tic shelf built up around the isolated volcanic islands, where thick
sequences of volcanic and epiclastic material were deposited. To
the north and directly behind the arc, the Kendeng Basin began
to subside, and material was transported from the volcaniclastic
shelf into the basin.
No reef carbonates are preserved within Synthem Two near
the arc, suggesting that environmental conditions, such as the
influx of volcanic detritus, prevented reef growth or, alternatively,
that these carbonates were not preserved or exposed. However,
some distance to the north, on the edge of the Sunda Shelf, a
large area of carbonate platform was developing at this time;
the shelf received volcanic debris as air fall, but much less than
that deposited close to the arc, and the volcanic material did not
substantially inhibit carbonate growth. Wilson and Lokier (2002)
showed that volcanic input can influence carbonate development
close to active arcs without killing carbonate-producing organ-
isms and causing carbonate deposition to cease.
During the Oligocene to early Miocene, volcanic activity
along the Southern Mountains Arc was at its most voluminous
and explosive. The volcaniclastic shelf close to the active volcanic
centers increased in width and thickness. Material was fed into the
Kendeng Basin, and only in the deepest part of the basin was there
pelagic sedimentation without a volcanogenic component.
Termination of Volcanism in the Southern Mountains Arc
Following a long period of arc activity between 42 and
18 Ma, volcanism in the Southern Mountains Arc ceased. The
final stages of activity were marked by a phase of explosive vol-
canism, which included a major event, the Semilir Eruption. The
age range, thicknesses, area, and estimated volumes of volcanic
deposits in the vicinity of the Toba volcanic center (Chesner
and Rose, 1991) are similar to those of the Semilir center, and
the deposits of the Semilir Eruption may be distributed, like the
Youngest Toba Tuffs (Song et al., 2000), over large parts of SE
Asia. Work is in progress to assess their distribution.
Lull and Resurgence of Volcanic Activity
Following the termination of volcanism, there was wide-
spread erosion of the deposits of Synthem Two and extensive
213
reef growth. Volcanogenic turbidites built out as thick aprons
northward from the inactive arc into the Kendeng Basin. Sedi-
mentation was rapid and on an unstable northward-dipping slope,
and the deposits began to fill the accommodation space within
the basin. Material also traveled southward from the eroding arc
into the Java forearc. On fault-bounded highs or in the shelter of
the extinct volcanoes, reefs and carbonate platforms developed,
marking the first period of extensive carbonate growth within the
Southern Mountains during the middle Miocene (ca. 20–10 Ma).
Arc volcanism resumed in the late Miocene, after a lull of
~8 m.y., 50 km to the north of the extinct Southern Mountains
Arc at the position of the modern Sunda Arc.
CHARACTER OF THE CRUST BENEATH THE
SOUTHERN MOUNTAINS ARC
Little is known of the crust beneath many young arcs, espe-
cially in Indonesia, and East Java is no exception. However, study
of the stratigraphy of the Southern Mountains Arc has provided
some new and surprising insights into the character of the deep
crust beneath the arc.
As discussed above, exposures of basement rocks in East
Java are limited and are restricted to the west of the study area.
These rocks have been interpreted to be fragments of arc and
ophiolitic material accreted to the continental margin of Sunda-
land during the Late Cretaceous (Hamilton, 1979; Wakita, 2000).
It has been generally considered that these rocks are typical of the
basement beneath the whole of East Java. This is supported by oil
company drilling on land in East Java and farther to the north in
the East Java Sea, where deep wells penetrate a varied basement,
including basic and acidic volcanic rocks, slaty metasedimentary
rocks, quartzites, and cherts (e.g., Hamilton, 1979; Matthews and
Bransden, 1995). However, the abundance of acidic volcanism
in the Southern Mountains Arc and the dating of zircons indicate
that the crust beneath much of the southern part of East Java is
very different, and is much older.
Volcanism
The intermediate to acidic volcanic rocks of the Southern
Mountains Arc range in composition from andesite to rhyolite
(60–77 wt% SiO2), with an estimated average SiO2 content of
67 wt%. In addition, many of the high-level intrusive bodies
exposed within the Southern Mountains Arc are granodiorites.
These rocks are considerably more evolved than the basic to
intermediate eruptive products of the present-day volcanoes. The
chemistry of the modern arc volcanic rocks in East Java (Hand-
ley, 2006; Nicholls et al., 1980; Wheller et al., 1987) indicates
that they are the products of relatively primitive subduction melts,
which reveal no significant interaction with underlying crust. The
difference in chemistry of the products of the Southern Moun-
tains Arc and the Sunda Arc could be explained in a number of
ways: fractional crystallization of more basic magma, interaction
of more basic magma with felsic crust, and partial melting of
214 Smyth et al.
felsic crust. Evidence from zircons (discussed below) suggests
that involvement of continental basement in petrogenesis is the
most likely explanation.
Ancient Zircons
At the beginning of this study there were very few good dates
for volcanic rocks of the older arc on Java (50 analyses), and
all published ages were K-Ar dates (Ben-Avraham and Emery,
1973; Soeria-Atmadja et al., 1994). No acidic rocks had been
dated. During this study, zircons from 16 Cenozoic samples of
igneous (9), volcaniclastic (3), and sedimentary rocks (4) were
dated by the SHRIMP U-Pb method at Curtin University of Tech-
nology, Australia, using the methods described in Smyth et al.
(2005, 2007). More than 453 spot ages were measured, and 270
of these were Cretaceous and older.
Prior to SHRIMP dating, the expected age range of the zir-
con samples was Cenozoic to Cretaceous. This reflects the age of
arc activity known from Java, and a contribution from the Cre-
taceous arc and ophiolitic terranes interpreted to form the base-
ment of East Java. However, in addition to the expected ages, an
unexpected range of Archean–Cambrian grains was identified in
a large number of samples (Smyth, 2005, 2007). The samples
containing Cretaceous ages and those containing Cambrian and
older zircons occur in distinct areas of East Java.
Lithology and Zircon Age Range
The five intrusive and four extrusive igneous rocks analyzed
contain a range of Cambrian and older zircons (n = 155) but lack
Cretaceous zircons. In contrast, the three volcaniclastic rocks
analyzed yielded a significant number of Cretaceous zircons (n
= 42) but only one Cambrian or older zircon. The four quartz-
rich sandstones analyzed yielded varied zircon ages. All of these
sandstones contain Cretaceous grains (n = 33). Three sandstones
from the Southern Mountains and Kendeng Basin contain a range
of Cambrian and older grains (n = 22). One sandstone from the
edge of the Sunda Shelf contains a single Jurassic–Triassic grain
but no older grains.
Age Ranges and Spatial Distribution
The samples containing different zircon ages do have a clear
geographical pattern. The Archean grains are found only in rocks
of the Southern Mountains (Fig. 12B), and the igneous rocks of
this area contain no Cretaceous zircons (Fig. 12A). Cretaceous
zircons are found only in the north and west of East Java, and
these zircons have been identified only in sedimentary and vol-
caniclastic rocks. This is interpreted to indicate that the South-
ern Mountains are underlain by material providing Cambrian to
Archean zircons but no Cretaceous zircons. To the north and west
of the Southern Mountains the underlying rocks do not include
Cambrian to Archean material but do include Cretaceous mate-
rial. Figure 12C shows the distribution of basement rocks inter-
preted from the zircon ages.
If all the zircon ages are grouped together, five age peaks
can be identified (Fig. 13): a Cenozoic peak (5–42 Ma) recording
volcanism in the Southern Mountains and Sunda Arcs, a Creta-
ceous peak (65–135 Ma) consistent with the interpreted age of
basement, a Cambrian to Neoproterozoic peak (500–1000 Ma), a
Mesoproterozoic to Paleoproterozoic peak (1000–2500 Ma), and
an Archean peak (2500–3200 Ma). The Cambrian and older ages
were not expected, based on knowledge of the regional geology.
Source of the Ancient Zircons
Sundaland, the continental core of SE Asia, is the closest
and most obvious source of the very old zircons (Figs. 1 and 13).
Rocks that could have provided abundant old zircons include gran-
ites of SW Borneo (Hamilton, 1979), the Malay Peninsula (Liew
and Page, 1985), the Thai-Malay Tin Belt (Cobbing et al., 1986),
and Sumatra (Imtihanah, 2000; McCourt et al., 1996) (Fig. 13).
However, in these areas there are no known Archean rocks, nor
is there any indication that they are underlain by Archean crust.
Geochemical and isotopic studies suggest a basement no older
than Proterozoic in areas of Sundaland, such as the Thai-Malay
peninsula, that have been studied (e.g., Liew and Page, 1985). The
closest area with extensive granites is the Schwaner Mountains of
SW Borneo. Paleogene sedimentary rocks of north Borneo contain
debris eroded mainly from the Schwaner granites and the Malay
Tin Belt, including detrital zircons (van Hattum et al., 2006). The
zircon populations are dominated by Cretaceous zircons with age
peaks different from those of East Java; there are abundant Perm-
ian–Triassic zircons, rare in East Java; and Precambrian zircons
are mainly Paleoproterozoic with very rare Archean grains. The
differences in zircon ages rule out a Schwaner or Thai-Malay
provenance, and other granite sources in Sundaland are even more
distant and paleogeographically unlikely.
The largest and closest area of continental crust of Archean
age is Australia, which formed part of Gondwana until the Cre-
taceous. It has been suggested that small Gondwana continental
fragments collided with the east Sundaland margin in Sulawesi
during the Cretaceous (e.g., Wakita et al., 1996; van Leeuwen and
Muhardjo, 2005), although the source of the fragments was not
identified. There is evidence of Late Jurassic–Early Cretaceous
rifting of continental fragments from the northwest and west Aus-
tralian margins (Müller et al., 2000) during continental breakup
preceding the separation of India from Gondwana. One of these
continental fragments could have collided with the Sundaland
margin. Bergman et al. (1996) speculated that there could be
old continental crust beneath west Sulawesi, on the basis of lead
isotopic compositions of Neogene plutonic rocks, which were
suggested to have had an Australian origin. Basement rocks in
western Australia are dated as Proterozoic and Archean. Detrital
zircons from young sediments of the Perth Basin, derived from
the erosion of western Australian basement, have been well stud-
ied and dated (Cawood and Nemchin, 2000; Pell et al., 1997;
Sircombe and Freeman, 1999). The zircon populations in Perth
 Cenozoic volcanic arc history of East Java, Indonesia 215

A o
110 E
o
112 E Inherited population includes
Cretaceous zircons
 NO Cretaceous zircons

o
7S

8oS

0 50 100

Kilometers

B o
110 E
o
112 E
Inherited population contains

 NO Archean zircons
Archean zircons

7oS
Figure 12. Distribution of samples that
contained zircon populations of (A)
Cretaceous ages, and (B) Archean ages
(Smyth, 2005). (C) Interpreted character
of the basement terranes in East Java.
o
8S

0 50 100

Kilometers

C 110oE 112oE
Edge of Sunda shelf: Nature of the basement
 uncertain; potentially arc and ophiolitic fragments
accreted during the Cretaceous.

o
7S
CENTRAL JAVA:
Accreted
Cretaceous KENDENG BASIN: Nature of the basement
metamorphosed uncertain; potentially arc and ophiolitic
arc and ophiolitic fragments accreted during the Cretaceous
crust.

o
8S SOUTHERN MOUNTAINS ARC:
Underlain by a fragment of continental
crust of Gondwana origin

0 50 100
Kilometers

Basin sedimentary rocks are remarkably similar in age to those
of the Southern Mountains in East Java. Both areas yield zircons
with 4200–2500 Ma ages (predominantly 3200–2500 Ma) simi-
lar to the Archean Yilgarn and Pilbara Blocks, 2000–1600 Ma
and 1300–1000 Ma ages similar to the Paleoproterozoic and
Mesoproterozoic Capricorn and Albany-Fraser orogenic belts, and
800–500 Ma ages similar to the Neoproterozoic Pinjara orogenic
belt (Fig. 14). A west Australian origin is a geographically simple
and plausible explanation for the Precambrian zircons of East
Java, but how did they become incorporated in igneous rocks?
216 Smyth et al.

Malay
Peninsula
Sabah
Brunei CELEBES
SEA
Sarawak Malino Metamorphic
SUNDA Borneo Complex
SHELF
Kalimantan

it
Stra
0°

r
Schwaner

assa
INDIAN
Mountains

ins
OCEAN

Mounta s

Mak
Meratu
Sulawesi
Sumatra
Su

JAVA
ch
nd

r
West Java SEA aA
aT

Basin jn aw
ren

u East Java
rim
ch

Extent of continental Ka Sea Basin FLORES

crust of Sundaland SEA
(Hamilton, 1979) Java Bali Lombok
Granites of Sundaland
(Cobbing, 2005;
10°S van Hattum, 2005)
Java Trench
100°E 110°E 120°E

Figure 13. Extent of area interpreted to be underlain by continental crust, and locations of granitic bodies that may have
been available for erosion during the Cenozoic. The NE-SW–trending Karimunjawa Arch, which was a significant topo-
graphic barrier throughout the Cenozoic, is also denoted.

Incorporation of Zircons into East Java Igneous Rocks
Zircons are highly refractory and are well known for sur-
viving repeated episodes of melting (e.g., Hanchar and Hoskin,
2003). Ancient zircons could have been introduced into Paleo-
gene magmas of the Southern Mountains Arc either by subduc-
tion of Gondwana-derived sediments or crust, or by passage of
magmas through a previously emplaced Gondwana-origin conti-
nental fragment at depth beneath the Southern Mountains.
Subduction of Gondwana Sediment
Sediment from western Australia deposited on the Indo-Aus-
tralian plate could have been carried north to the Java Trench as
Australia moved north. However, even today a fan similar in size
to the modern Bengal Fan would be required to bring material
from Australia to the site of subduction at the Java Trench, and
there is no evidence of a sediment supply and distribution system
like this in the past. The huge Bengal Fan is fed by erosion from
the Himalayas, but there is no evidence for a comparable orogenic
belt in Australia during Cenozoic times. During the early Ceno-
zoic, Australia lay much farther south of its current position (Hall,
2002), and there is little sedimentary cover today on the Indian-
Australian plate south of Java (Masson et al., 1990; Kopp, 2002)
in what would have been more proximal parts of such a fan.
It is unlikely that material could have been introduced
into the Java Trench by axial transport from the Himalayas
via the Bengal Fan, or from the Bird’s Head microcontinent,
New Guinea. If sediment was produced by the erosion of the
Himalayas, it would require transport via the Bengal Fan to the
Java Trench and total transport distances >4000 km. Although
mud-rich material is considered to have reached northern
Sumatra in the distal parts of the Bengal Fan (Curray et al.,
2002), there were topographic barriers at the Investigator Frac-
ture Zone and the Ninety East Ridge, and even then a further
2000 km of along-trench transport. Today there is no material
from the Bengal Fan reported in the Java Trench off the shore
of East Java (e.g., Masson et al., 1990) and no significant sedi-
ment cover on the Indian plate south of Java (Kopp, 2002). It
is equally unlikely that sediment from the Himalayan region
could have traveled southeastward from Indochina across Sun-
daland, bypassing numerous basins and associated structural
highs (Hall and Morley, 2004) to enter East Java. This would
require transport distances >3500 km.
The basement rocks of the Bird’s Head microcontinent were
not available for erosion during most of the Cenozoic, and the
Bird’s Head was the site of deposition of thick marine carbon-
ates of the New Guinea limestones (Fraser et al., 1993; Pieters
et al., 1983). Even had the Bird’s Head been supplying large vol-
umes of sediment, along-trench transport distances >2000 km
would have been required for material to enter the subduction
zone beneath Java. This is equivalent to material from southern
Mexico or northern Washington State entering the Allistos Arc
 Cenozoic volcanic arc history of East Java, Indonesia 217

40
CENOZOIC
VOLCANISM
Range of U-Pb SHRIMP zircon ages from the A
Southern Mountains Arc, East Java (Smyth,
2005). N = 453

35

30
Numbers of grains

CRETACEOUS
BASEMENT
25

20

15

10

0
500 1000 1500 2000 2500 3000
Figure 14. Comparison of the U-Pb
PHANEROZOIC NEOPROT MESOPROT PALEOPROTEROZOIC ARCHEAN
SHRIMP ages of (A) East Java (Smyth,
2005), and (B) Western Australia (Bruguier
40 Range of U-Pb SHRIMP ages from placer B et al., 1999; Sircombe and Freeman, 1999).
deposits in Western Australia (Bruguier et al, 1999; Revised from Smyth et al. (2005).
Sircombe and Freeman, 1999) and some of the
probable source regions. N = 260
35

30 PINJARA ALBANY-FRASER CAPRICORN

YILGARN CRATON
OROGEN OROGEN OROGEN
500-800 Ma 1000-1300 Ma 1600-2000 Ma 2500-4200 Ma
Numbers of grains

25

20

15

10

500 1000 1500 2000 2500 3000

Age (Ma)

(Busby et al., 2006) of Baja California, or for northern Norway or
Alabama to have been sources for the Lough Nafooey Arc (Draut
and Clift, 2001) of the Irish Caledonides.
Subduction of a Gondwana Fragment
Subduction of continental crust is even less likely. Buoy-
ancy forces make subduction of continental crust difficult, and
even if these were overcome, the fragment would have been
required to supply material to the arc from the middle Eocene
to the early Miocene, a period of ~20 m.y. The fragment of crust
either remained stationary beneath the Southern Mountains or
was 1500 km in length, based on the present subduction rate of
75 mm/year (McCaffrey, 1996), or on Cenozoic plate tectonic
reconstructions (Hall, 2002).
218 Smyth et al.
Accretion of a Gondwana Fragment
The most likely explanation is that continental crust was
already present beneath the Southern Mountains Arc when sub-
duction began in the middle Eocene. We suggest that this was a
western Australia–rifted continental fragment that collided with
the margin of Sundaland during the Late Cretaceous and may
have terminated the Cretaceous phase of subduction. Subduction
was renewed in the middle Eocene, and Cenozoic melts were
contaminated by interaction with continental crust beneath the
Southern Mountains. The distribution of pre-Cenozoic zircons
indicates that north of the Southern Mountains there is no conti-
nental basement, and the deep crust is arc or ophiolitic, perhaps
representing material accreted during Cretaceous subduction or
underplated during collision. The continental fragment could
have been quite large and may be traceable into Sulawesi. Zircon
dating and geochemical evidence (van Leeuwen and Muhardjo,
2005) show that the Malino Metamorphic Complex of NW
Sulawesi (Fig. 13) contains zircons with ages up to 3500 Ma.
DISCUSSION
The stratigraphic record in East Java provides insights into
the nature of the deep crust beneath the arc, the history of Indian-
Australian plate subduction, and the contribution of the volcanic
arc to basin formation. It also poses some questions about the
continuity and location of volcanic activity, which have relevance
for other arcs.
Deep Character of the East Java Volcanic Arc
Little is known about the deep crust beneath the young arcs
of the Western Pacific and Indonesia. There have been few seis-
mic refraction studies and few xenolith studies, and the deep crust
is almost invariably covered by younger volcanic and sedimen-
tary rocks. In East Java the old zircons in the Eocene to Miocene
arc products provide evidence of the character of the deep crust.
The insight gained was completely unexpected. When this study
began, it was thought that the entire region was underlain by Cre-
taceous arc and ophiolitic fragments, as suggested by Hamilton
(1979). The distribution of ages shows that north of the South-
ern Mountains there is indeed crust of this character, as indicated
by the small exposures of basement, but beneath the Southern
Mountains themselves there is a fragment of continental crust of
Gondwana character and western Australian origin.
The presence of a continental fragment beneath the arc
accounts for the unusually acidic character of arc volcanism in
the Eocene to early Miocene arc. The acidic products of this arc
have been overlooked partly because the resistant “Old Andes-
ites” are topographically so much more obvious, although acidic
volcanic and minor intrusive rocks are well exposed in many parts
of the Southern Mountains. However, the main reason why the
acidic products of the Eocene to early Miocene arc were missed
by earlier studies is because they were erupted explosively and
dispersed as volcanic ash, the ash was reworked into sediments,
and the processes of eruption and tropical weathering combined
to remove the unstable volcanic constituents, leaving well-sorted
quartz-rich sandstones. These were previously interpreted as hav-
ing been eroded from a continental Sundaland source, but careful
examination of the sandstones reveals abundant evidence of their
volcanic origin on the basis of textures, light mineral constitu-
ents, quartz character, clay mineralogy, and zircon character and
ages (Smyth, 2005).
Age of Subduction
The character of the oldest parts of the East Java stratigraphic
successions indicates that Cenozoic volcanic activity began in the
middle Eocene (ca. 45 Ma). There is little evidence to support the
common assumption (e.g., Hall, 2002; Heine et al., 2004; Met-
calfe, 1996; Scotese et al., 1988) that subduction continued from
the Mesozoic into the Cenozoic without significant interruption.
The stratigraphy of East Java indicates that older subduction
probably terminated in the Late Cretaceous with the collision of
a continental fragment rifted from western Australia. We suggest
that the continental fragment extended from East Java to North
Sulawesi. Ophiolites were emplaced from Java to North Borneo
during this collision. The oldest sedimentary rocks above the
basement in East Java are of uncertain age but are middle Eocene
or older and lack volcanic debris. There is little evidence for lat-
est Cretaceous to early Eocene volcanic activity in most of the
Sundaland margin between Sumatra and Sulawesi. Subduction
resumed in the middle Eocene when Australia began to move
northward rapidly after 45 Ma (Hall, 2002; Müller et al., 2000;
Schellart et al., 2006), and has continued to the present day.
Formation of the Kendeng Basin
When volcanic activity resumed in the middle Eocene a basin
began to form directly north of the arc. Most of the investigations
of sedimentary basins on land in East Java and in the East Java
Sea (Fig. 1) have been carried out as part of hydrocarbon explora-
tion activity, and little attention has been given to areas close to
the modern arc. The Sunda Shelf basins are typically >100 km
from the arc and are characterized by many features typical of
fault-controlled basins; explanations for their origin have there-
fore interpreted them as types of backarc basins resulting from
subduction rollback, rift and thermal sag, or strike-slip faulting
(e.g., Hamilton, 1979; Matthews and Bransden, 1995).
The Kendeng Basin is not a typical backarc basin. Backarc
basins (Taylor and Karner, 1983) may form by trench rollback,
causing generation of oceanic crust (Dewey, 1980; Karig, 1971),
rifting of continental crust (Kobayashi, 1985), gravitational col-
lapse in the wake of arc-continent collision (Clift et al., 2003),
or by the trapping of old oceanic crust behind an arc (Scholl et
al., 1986). Retroarc basins (Dickinson, 1974) form as compres-
sive foreland basins behind a volcanic arc. Busby and Ingersoll
(1995) defined backarc basins as oceanic basins behind intraoce-
anic magmatic arcs, and continental basins behind continental
 Cenozoic volcanic arc history of East Java, Indonesia
margin arcs that lack foreland fold and thrust belts. Marsaglia
(1995) identified similar categories of backarc basins with the
addition of boundary basins resulting from extension along plate
boundaries with translational components. Busby and Ingersoll
(1995) claim that the Sunda Shelf basins of Indonesia are non-
extensional backarc basins formed in a neutral strain regime,
although the basis for this assertion is not clear, because there is
abundant evidence for rifting of many of these basins (e.g., Cole
and Crittenden, 1997; Hall and Morley, 2004).
There is no evidence for either newly formed oceanic crust
under the Kendeng Basin or for strike-slip faulting. The basin is
much closer to the arc than other backarc basins and most of the
“backarc basins” of the Sunda Shelf. The late Cenozoic deforma-
tion in the region led to ~10–30 km of contraction at the outer
edge of the basin, so throughout the early Cenozoic this basin was
very close to the arc. There was extension in the Sunda Shelf from
the Eocene, but nowhere was oceanic crust formed. The Kend-
eng Basin is an asymmetrical depression, the deepest part of the
basin lies directly behind the arc, and the Eocene to lower Mio-
cene volcanic and sedimentary sequence thins toward the edge of
the Sunda Shelf (Waltham et al., this volume). The basin fill was
derived mainly from the Southern Mountains Arc on the south
side of the basin. The basin began to form at the same time as
arc activity began, and its subsidence history is closely linked to
activity in the volcanic arc. There are no seismic lines crossing the
basin, and it is not well exposed at the surface, so we cannot assess
the role of faulting in the basin development, but it does not have
the typical synrift-postrift stratigraphy of many other Sunda Shelf
basins. There is no evidence of compressional loading having
contributed to basin formation. Thrusting in East Java occurred in
the late Neogene long after the Kendeng Basin had formed. The
close relationship between volcanic activity and basin subsidence
suggests that the two are linked, and we suggest that the load of
the volcanic arc was the major cause of basin subsidence. The
contribution of volcanic arc loading to basin formation in Indone-
sia is discussed in greater detail by Waltham et al. (this volume).
Movement of the Volcanic Arc
Activity in the Southern Mountains Arc terminated during
the early Miocene (ca. 20 Ma). After a period with little mag-
matism, a new episode of arc volcanism began in East Java at
ca. 12–10 Ma in a new location. The Miocene to Holocene arc is
parallel to the Southern Mountains Arc but ~50 km north of it. A
significant reduction in volcanic activity in the middle Miocene
is a well-known feature of the Sunda Arc from Java eastward,
although subduction was continuous during this period (Hall,
2002). Vigorous volcanic activity resumed at the end of the mid-
dle Miocene in the Sunda and Banda Arcs at ca. 10 Ma (Hall,
2002; Macpherson and Hall, 2002).
Macpherson and Hall (1999, 2002) suggest that the decline in
volcanic activity resulted from northward advance of the subduc-
tion hinge. Hinge advance prevented replenishment of the mantle
wedge by fertile mantle, and consequently subduction-induced
219
melting was inhibited as the wedge became depleted. The cause
of hinge advance was the result of the collision of Australia with
eastern Indonesia, and the counterclockwise rotation of the Bor-
neo-Java region this collision induced (Hall, 2002). Arc volcanism
resumed when hinge advance ceased and the mantle wedge was
replenished by fertile mantle (Macpherson and Hall, 2002).
Why the volcanic arc moved to its new position 50 km north
of the Southern Mountains Arc is uncertain. There is evidence of
thrusting and contraction in Java in the late Neogene, but its tim-
ing is not precisely known; this is part of our current research. It
is possible that contraction was linked to large-scale arc dynam-
ics, such as coupling of the subduction and overriding slab. There
is evidence of an old consolidated accreted material in the Java
forearc that acts as a backstop to the active accretionary prism
(Kopp et al., 2001). This suggests two phases of forearc develop-
ment, and these may be linked to the two phases of arc volcanism
on land in East Java. It is also possible that when the subduction
hinge advanced and the mantle wedge was not being replenished,
one result was a rigid, stronger overriding lithospheric plate that
later failed when hinge advance ceased and weaker, warmer man-
tle replenished the wedge. Another control could have been the
nature of the crust beneath the arc. The zircon ages indicate that
there was a change in basement type north of the Southern Moun-
tains and that the boundary between continental and ophiolitic
crust at depth may have been a preexisting structural weakness
that influenced the position at which melts could rise when arc
activity resumed.
Other possibilities for explaining arc migration include sub-
duction erosion and a change in dip of the subducting slab. Sub-
duction erosion seems unlikely, as the width of the arc-trench gap
today is >300 km, and because marine data support active accre-
tion at that time (e.g., Kopp et al., 2001). A change in subduction
angle cannot be ruled out, but it is notable that today the slab
dip has increased to a very high angle after the slab descended
to 100 km, as seen in seismicity data (England et al., 2004). A
lower, not a higher, angle would be expected if the arc had moved
north owing to the change in dip of the slab.
The jump in position of the volcanic arc is a feature of other
Indonesian arcs (e.g., Halmahera Arc; Nichols and Hall, 1991)
and reflects the changing stresses at the plate boundary over time.
We prefer an explanation that links the change in position of the
arc to plate reorganization in the region. However, it is difficult to
test different hypotheses. Comparison of arc development to the
east and west of East Java, from Bali eastward, and toward West
Java and Sumatra, might provide insights, but unfortunately even
less is known of arc history in these regions than in East Java.
This East Java study shows that our understanding of arc tec-
tonics is still incomplete and that studies of arc stratigraphy are
essential if our knowledge of arc dynamics is to be improved.
CONCLUSIONS
The case study of early Cenozoic arc volcanism in East Java
shows the importance of examining arc stratigraphy. Insights
220 Smyth et al.
have been gained into the eruptive history, formation of a deep
basin behind the arc, and the character of the deep crust.
The early Cenozoic stratigraphy of East Java provides
a record of a cycle of arc activity from initiation in the mid-
dle Eocene (ca. 42 Ma) to termination in the early Miocene
(ca. 20 Ma). The Kendeng Basin, directly behind the Southern
Mountains Arc, contains >6 km of volcaniclastic and sedimen-
tary rocks. The basin is not a typical backarc basin, and its sub-
sidence history is linked to volcanic activity within the South-
ern Mountains Arc.
The final stage of volcanic activity in the Southern Moun-
tains Arc is marked by the Semilir Eruption (ca. 20 Ma), which
distributed ash over a wide area and may be comparable to the
Pleistocene eruption of Toba in Sumatra. Following this phase of
major eruptions, there was a lull in volcanic activity during the
middle Miocene, followed by resumption in arc activity to the
north of the Southern Mountains Arc, along the axis of the mod-
ern Sunda Arc during the late Miocene. The mechanisms that
resulted in the decline in volcanism, and northward movement in
arc axis, are not yet understood and show that our understanding
of arcs is still incomplete.
The stratigraphic record of volcanic arcs can provide insights
into the character of the deep crust. The entire East Java region
was previously thought to be underlain by Cretaceous arc and
ophiolitic fragments, but Archean to Cambrian zircons within
acidic products of the Southern Mountains Arc point to the
occurrence of a continental crust of Gondwanan character and
western Australian origin beneath the old arc. This continental
fragment is thought to have collided with Sundaland during the
Cretaceous and is interpreted to have terminated the Cretaceous
phase of subduction. The extent of this fragment is not known
but may be traceable into Sulawesi. The use of inherited zircons
to determine the character of the deep crust may be applicable in
other arcs.
ACKNOWLEDGMENTS
This project was funded by the SE Asia Research Group at
Royal Holloway University, supported by an oil company con-
sortium. Zircons were analyzed using the SHRIMP facility at
Curtin University of Technology, Perth, Australia. Financial
assistance for SHRIMP dating was provided by the University
of London Central Research Fund and CSIRO, Australia. The
analyses were undertaken by Joseph Hamilton (CSIRO and
University of the West Indies) and Pete Kinny (Curtin Univer-
sity of Technology). We are grateful to LIPI (Lembaga Ilmu
Pengetahuan Indonesia) for field-work permissions. We wish to
thank Eko Budi Lelono, Peter Lunt, Theo van Leeuwen, Moyra
Wilson, Colin Macpherson, Heather Handley, Cindy Ebinger,
and Dave Waltham. Marcelle BouDagher-Fadel of University
College London provided biostratigraphic analyses. Peter Clift,
Jason Ali, and an anonymous reviewer are thanked for their
extensive reviews, which greatly improved this manuscript.
REFERENCES CITED
Ardhana, W., 1993, A depositional model for the early middle Miocene Ngray-
ong Formation and implications for exploration in the East Java basin:
Jakarta, Proceedings, Indonesian Petroleum Association Annual Conven-
tion, 22nd, p. 395–443.
Baum, J.S., Baum, G.R., Thompson, P.R., and Humphery, J.D., 1994, Stable
isotope evidence for relative and eustatic sea-level changes in Eocene
to Oligocene carbonates, Baldwin County, Alabama: Geological
Society of America Bulletin, v. 106, p. 824–839, doi: 10.1130/0016-
7606(1994)106<0824:SIEFRA>2.3.CO;2.
Ben-Avraham, Z., and Emery, K.O., 1973, Structural framework of Sunda
Shelf: American Association of Petroleum Geologists Bulletin, v. 57,
p. 2323–2366.
Bergman, S.C., Coffield, D.Q., Talbot, J.P., and Garrard, R.J., 1996, Tertiary
tectonic and magmatic evolution of Western Sulawesi and the Makassar
Strait, Indonesia: Evidence for a Miocene continent-continent collision,
in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE Asia: Geo-
logical Society [London] Special Publication 106, p. 391–430.
Bishop, W.P., 1980, Structure, stratigraphy and hydrocarbons offshore southern
Kalimantan, Indonesia: American Association of Petroleum Geologists
Bulletin, v. 64, p. 37–58.
Bruguier, O., Bosch, D., Pidgeon, R.T., Byrne, D.I., and Harris, L.B., 1999,
U-Pb chronology of the Northampton Complex, Western Australia—
Evidence for Grenvillian sedimentation, metamorphism and deformation
and geodynamic implications: Contributions to Mineralogy and Petrol-
ogy, v. 136, p. 258–272, doi: 10.1007/s004100050537.
Busby, C.J., and Ingersoll, R.V., 1995, Tectonics of Sedimentary Basins: Cam-
bridge, Massachusetts, Blackwell Science, 579 p.
Busby, C.J., Adams, B.F., Mattinson, J., and Deoreo, S., 2006, View of an intact
oceanic arc, from surficial to mesozonal levels: Cretaceous Alisitos arc,
Baja California: Journal of Volcanology and Geothermal Research, v. 149,
p. 1–46, doi: 10.1016/j.jvolgeores.2005.06.009.
Catuneanu, O., 2006, Principles of Sequence Stratigraphy: Amsterdam, Else-
vier, 375 p.
Cawood, P.A., and Nemchin, A.A., 2000, Provenance record of a rift basin: U/Pb
ages of detrital zircons from the Perth Basin, Western Australia: Sedimen-
tary Geology, v. 134, p. 209–234, doi: 10.1016/S0037-0738(00)00044-0.
Chesner, C.A., and Rose, W.I., 1991, Stratigraphy of the Toba tuffs and the
evolution of the Toba caldera complex, Sumatra, Indonesia: Bulletin of
Volcanology, v. 53, p. 343–356, doi: 10.1007/BF00280226.
Chotin, P., Giret, A., Rampnoux, J.P., Rasplus, L., and Suminta, S., 1984,
Etude de la fracturation de l’ile de Java, Indonesie: Bulletin de la Société
Géologique de France, v. 7, p. 1259–1262.
Clift, P.D., Schouten, H., and Draut, A.E., 2003, A general model of arc-
continent collision and subduction polarity reversal from Taiwan and the
Irish Caledonides, in Larter, R.D., and Leat, P.T., eds., Intra-Oceanic Sub-
duction Systems; Tectonic and Magmatic Processes: Geological Society
[London] Special Publication 219, p. 81–98.
Cobbing, E.J., 2005, Granites, in Barber, A.J., Crow, M.J., and Milsom, J.S.,
eds., Sumatra: Geology, Resources, and Tectonic Evolution: Geological
Society [London] Memoirs 31, p. 267–282.
Cobbing, E.J., Mallick, D.I.J., Pitfield, P.E.J., and Teoh, L.H., 1986, The gran-
ites of the Southeast Asian tin belt: Geological Society [London] Journal,
v. 143, p. 537–550.
Cole, J.M., and Crittenden, S., 1997, Early Tertiary basin formation and the devel-
opment of lacustrine and quasi-lacustrine/marine source rocks on the Sunda
Shelf of SE Asia, in Fraser, A.J., et al., eds., Petroleum Geology of SE Asia:
Geological Society [London] Special Publication 126, p. 147–183.
Curray, J.R., Emmel, F.J., and Moore, D.G., 2002, The Bengal Fan: Morphol-
ogy, geometry, stratigraphy, history and processes: Marine and Petroleum
Geology, v. 19, p. 1191–1223, doi: 10.1016/S0264-8172(03)00035-7.
de Genevraye, P., and Samuel, L., 1972, The geology of Kendeng Zone (East
Java): Jakarta, Proceedings, Indonesian Petroleum Association Annual
Convention, 1st, p. 17–30.
Dewey, J.F., 1980, Episodicity, sequence and style at convergent plate boundar-
ies: Geological Association of Canada Special Paper 20, p. 553–573.
Dickinson, W.R., 1974, Plate tectonics and sedimentation, in Dickinson, W.R.,
ed., Tectonics and Sedimentation: Tulsa, Society of Economic Paleontolo-
gists and Mineralogists, p. 204.
 Cenozoic volcanic arc history of East Java, Indonesia
Draut, A.E., and Clift, P., 2001, Geochemical evolution of arc magmatism dur-
ing arc-continental collision, South Mayo, Ireland: Geology, v. 29, p. 543–
546, doi: 10.1130/0091-7613(2001)029<0543:GEOAMD>2.0.CO;2.
Ebanks, W.J.J., and Cook, C.B.P., 1993, Sedimentology and reservoir proper-
ties of Eocene Ngimbang clastics sandstones in cores of the Pagerungan-5
well, Pagerungan Field, East Java Sea, Indonesia, in Atkinson, C.D., et al.,
eds., IPA Core Workshop Notes, Clastic Rocks and Reservoirs of Indone-
sia: Jakarta, Indonesian Petroleum Association, p. 9–36.
Edwards, C., Menzies, M., and Thirlwall, M., 1991, Evidence from Muriah,
Indonesia, for the interplay of supra-subduction zone and intraplate pro-
cesses in the genesis of potassic alkaline magmas: Journal of Petrology,
v. 32, p. 555–592.
England, P., Engdahl, R., and Thatcher, W., 2004, Systematic variation in the
depths of slabs beneath arc volcanoes: Geophysical Journal International,
v. 156, p. 377–408, doi: 10.1111/j.1365-246X.2003.02132.x.
Fraser, T.H., Bon, J., and Samuel, L., 1993, A new dynamic Mesozoic stratig-
raphy for the West Irian micro-continent Indonesia and its implications:
Jakarta, Proceedings, Indonesian Petroleum Association Annual Conven-
tion, 22nd, p. 707–761.
Fulthorpe, C.S., Carter, R.M., Miller, K.G., and Wilson, J., 1996, Marshall
Paraconformity: A mid-Oligocene record of inception of the Antarctic
Circumpolar and coeval glacio-eustatic lowstand?: Marine and Petroleum
Geology, v. 13, p. 61–77, doi: 10.1016/0264-8172(95)00033-X.
Gradstein, F., Ogg, J., and Smith, A., 2004, A Geological Time Scale 2004:
Cambridge, UK, Cambridge University Press, 589 p.
Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and
the SW Pacific: Computer-based reconstructions, model and animations:
Journal of Asian Earth Sciences, v. 20, p. 353–434, doi: 10.1016/S1367-
9120(01)00069-4.
Hall, R., and Morley, C.K., 2004, Sundaland basins, in Clift, P., et al., eds., Con-
tinent-Ocean Interactions within the East Asian Marginal Seas: American
Geophysical Union Geophysical Monograph 149, p. 55–85.
Hamilton, W., 1979, Tectonics of the Indonesian Region: U.S. Geological Sur-
vey Professional Paper 1078, 345 p.
Hanchar, J.M., and Hoskin, P.W.O., eds., 2003, Zircon: Reviews in Mineralogy
and Geochemistry, v. 53, 500 p.
Handley, H., 2006, Geochemical and Sr-Nd-Hf-O isotopic constraints on volca-
nic petrogenesis at the Sunda arc, Indonesia [Ph.D. thesis]: Durham, UK,
University of Durham, 268 p.
Haq, B.U., Hardenbol, J., and Vail, P.R., 1987, Chronology of fluctuating sea
levels since the Triassic: Science, v. 235, p. 1156–1167, doi: 10.1126/sci-
ence.235.4793.1156.
Harahap, B.H., Bachri, S., Baharuddin, Surwarna, N., Panggabean, H., and
Simanjuntak, T.O., 2003, Stratigraphic Lexicon of Indonesia: Bandung,
Geological Research and Development Centre, 729 p.
Heine, C.R., Müller, D., and Gaina, C., 2004, Reconstructing the lost East-
ern Tethys ocean basin: Convergence history of the SE Asian margin and
marine gateways, in Clift, P., et al., eds., Continent-Ocean Interactions
within the East Asian Marginal Seas: American Geophysical Union Geo-
physical Monograph 149, p. 37–54.
Hoffmann-Rothe, A., Ritter, O., and Hanna, W.F., 2001, Magnetotelluric and
geomagnetic modelling reveals zones of very high electrical conductivity
in the upper crust of Central Java: Physics of the Earth and Planetary Inte-
riors, v. 124, p. 131–151, doi: 10.1016/S0031-9201(01)00196-0.
Imtihanah, 2000, Isotopic dating of the Sumatran Fault System [MPhil. thesis]:
University of London, 150 p.
Kadar, D., 1986, Neogene planktonic foraminiferal biostratigraphy of the south
central Java area, Indonesia: Geological Research and Development Cen-
tre Special Publication 5, 83 p.
Karig, D.E., 1971, Origin and development of marginal basins in the Western
Pacific: Journal of Geophysical Research, v. 76, p. 2542–2561.
Kobayashi, K., 1985, Sea of Japan and Okinawa Trough, in Naira, A.E.M., et
al., eds., The Pacific Ocean, Volume 7: The Ocean Basins and Margins:
New York, Plenum Press, p. 419–458.
Kopp, H., 2002, BSR occurrence along the Sunda margin: Evidence from seis-
mic data: Earth and Planetary Science Letters, v. 197, p. 225–235, doi:
10.1016/S0012-821X(02)00484-3.
Kopp, H., Flueh, E.R., Klaeschen, D., Bialas, J., and Reichert, C., 2001, Crustal
structure of the central Sunda margin at the onset of oblique subduction:
Geophysical Journal International, v. 147, p. 449–474, doi: 10.1046/
j.0956-540x.2001.01547.x.
221
Leeder, M.R., 1982, Sedimentology: Process and Product: London, George
Allen and Unwin, 344 p.
Lelono, E.B., 2000, Palynological study of the Eocene Nanggulan Formation,
central Java, Indonesia [Ph.D. thesis]: University of London, 453 p.
Liew, T.C., and Page, R.W., 1985, U-Pb zircon dating of granitoid plutons from
the west coast of peninsular Malaysia: Geological Society [London] Jour-
nal, v. 142, p. 515–526.
Lokier, S.W., 2000, The Miocene Wonosari Formation, Java, Indonesia: Vol-
caniclastic influences on carbonate platform development [Ph.D. thesis]:
University of London, 648 p.
Macpherson, C.G., and Hall, R., 1999, Tectonic controls of geochemical evolu-
tion in arc magmatism of SE Asia, in Proceedings, PACRIM Congress,
4th, Australian Institute of Mining and Metallurgy, p. 359–368.
Macpherson, C.G., and Hall, R., 2002, Timing and tectonic controls on magma-
tism and ore generation in an evolving orogen: Evidence from Southeast
Asia and the western Pacific, in Blundell, D.J., et al., eds., The Timing
and Location of Major Ore Deposits in an Evolving Orogen: Geological
Society [London] Special Publication 204, p. 49–67.
Madon, M.B., Azlina, A., and Wong, R., 1997, Structural evolution, maturation
history and hydrocarbon potential of the Penyu basin, offshore peninsula
Malaysia, in Howes, J.V.C., and Noble, R.A., eds., Proceedings of the
International Conference on Petroleum Systems of SE Asia and Australia:
Jakarta, Indonesian Petroleum Association, p. 403–424.
Marsaglia, K.M., 1995, Interarc and backarc basins, in Busby, C.J., and Inger-
soll, R.V., eds., Tectonics of Sedimentary Basins: Cambridge, Massachu-
setts, Blackwell Science, p. 299–330.
Masson, D.G., Parson, L.M., Milsom, J., Nichols, G., Sikumbang, N., Dwi-
yanto, B., and Kallagher, H., 1990, Subduction of seamounts at the Java
Trench: A view with long-range sidescan sonar: Tectonophysics, v. 185,
p. 51–65, doi: 10.1016/0040-1951(90)90404-V.
Matthews, S.J., and Bransden, P.J.E., 1995, Late Cretaceous and Cenozoic
tectono-stratigraphic development of the East Java Sea Basin, Indonesia:
Marine & Petroleum Geology, v. 12, p. 499–510, doi: 10.1016/0264-
8172(95)91505-J.
McCaffrey, R., 1996, Slip partitioning at convergent plate boundaries of SE
Asia, in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE Asia:
Geological Society [London] Special Publication 106, p. 3–18.
McCourt, W.J., Crow, M.J., Cobbing, E.J., and Amin, T.C., 1996, Mesozoic and
Cenozoic plutonic evolution of SE Asia: Evidence from Sumatra, Indone-
sia, in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE Asia:
Geological Society [London] Special Publication 106, p. 321–335.
Metcalfe, I., 1996, Gondwanaland dispersion, Asian accretion and evolution of
eastern Tethys: Australian Journal of Earth Sciences, v. 43, p. 605–624.
Miyazaki, K., Sopaheluwakan, J., Zulkarnain, I., and Wakita, K., 1998, A jade-
ite-quartz-glaucophane rock from Karangsambung, central Java, Indone-
sia: Island Arc, v. 7, p. 223–230, doi: 10.1046/j.1440-1738.1998.00164.x.
Müller, R.D., Gaina, C., and Clarke, S., 2000, Seafloor spreading around Aus-
tralia, in Veevers, J.J., ed., Billion-Year Earth History of Australia and
Neighbours in Gondwanaland: Sydney, GEMOC Press, p. 18–28.
Nicholls, I.A., Whitford, D.J., Harris, K.L., and Taylor, B., 1980, Variation in
the geochemistry of mantle sources for tholeiitic and calc-alkaline mafic
magmas, western Sunda volcanic arc, Indonesia: Chemical Geology,
v. 30, p. 177–199, doi: 10.1016/0009-2541(80)90105-9.
Nichols, G.J., and Hall, R., 1991, Basin formation and Neogene sedimentation
in a backarc setting, Halmahera, eastern Indonesia: Marine and Petroleum
Geology, v. 8, p. 50–61, doi: 10.1016/0264-8172(91)90044-2.
Parkinson, C.D., Miyazaki, K., Wakita, K., Barber, A.J., and Carswell, D.A.,
1998, An overview and tectonic synthesis of the pre-Tertiary very-high-
pressure metamorphic and associated rocks of Java, Sulawesi and Kali-
mantan, Indonesia: Island Arc, v. 7, p. 184–200, doi: 10.1046/j.1440-
1738.1998.00184.x.
Pell, S.D., Williams, I.S., and Chivas, A.R., 1997, The use of protolith zircon-
age fingerprints in determining the protosource areas for some Australian
dune sands: Sedimentary Geology, v. 109, p. 233–260, doi: 10.1016/
S0037-0738(96)00061-9.
Pieters, P.E., Pigram, C.J., Trail, D.S., Dow, D.B., Ratman, N., and Sukamto,
R., 1983, The stratigraphy of western Irian Jaya: Bandung, Bulletin of
Geological Research and Development Centre, v. 8, p. 14–48.
Rawson, R.F., Allen, P.M., Brenchley, P.J., Cope, J.C.W., Gale, A.S., Evans,
J.A., Gibbard, P.L., Gregory, F.J., Hailwood, E.A., Hesselbo, S.P., Knox,
R.W.O.B., Marshall, J.E.A., Oates, M., Riley, N.J., Smith, A.G., Trewin,
222 Smyth et al.
N., and Zalasiewicz, J.A., 2001, Stratigraphical Procedure: Bath, Geo-
logical Society [London], 64 p.
Sandwell, D.T., and Smith, W.H.F., 1997, Marine gravity anomaly from Geosat
and ERS 1 satellite altimetry: Journal of Geophysical Research, v. 102,
p. 10,039–10,054, doi: 10.1029/96JB03223.
Schellart, W.P., Lister, G.S., and Toy, V.G., 2006, A Late Cretaceous and Ceno-
zoic reconstruction of the Southwest Pacific region: Tectonics controlled
by subduction and slab rollback processes: Earth-Science Reviews, v. 76,
p. 191–233, doi: 10.1016/j.earscirev.2006.01.002.
Scholl, D., Vallier, T.L., and Stevenson, A.J., 1986, Terrane accretion, produc-
tion and continental growth: A perspective based on the origin and tec-
tonic gate of the Aleutian–Bering Sea region: Geology, v. 14, p. 43–47,
doi: 10.1130/0091-7613(1986)14<43:TAPACG>2.0.CO;2.
Scotese, C.R., Gahagan, L.M., and Larson, R.L., 1988, Plate tectonic recon-
structions of the Cretaceous and Cenozoic ocean basins: Tectonophysics,
v. 155, p. 27–40, doi: 10.1016/0040-1951(88)90259-4.
Sharaf, E., Simo, J.A., Carrol, A.R., and Shields, M., 2005, Stratigraphic evolu-
tion of Oligocene–Miocene carbonates and siliciclastics, East Java basin,
Indonesia: American Association of Petroleum Geologists Bulletin, v. 89,
p. 799–819.
Sircombe, K.N., and Freeman, M.J., 1999, Provenance of detrital zircons
on the Western Australian coastline—Implications for the geological
history of the Perth basin and denudation of the Yilgarn craton: Geol-
ogy, v. 27, p. 879–882, doi: 10.1130/0091-7613(1999)027<0879:
PODZOT>2.3.CO;2.
Smyth, H., 2005, Eocene to Miocene basin history and volcanic activity in East
Java, Indonesia [Ph.D. thesis]: University of London, 476 p.
Smyth, H., Hall, R., Hamilton, J.P., and Kinny, P., 2005, East Java: Cenozoic
basins, volcanoes and ancient basement: Jakarta, Proceedings, Indonesian
Petroleum Association Annual Convention, 30th, p. 251–266.
Smyth, H.R., Hamilton, P.J., Hall, R., and Kinny, P.D, 2007, The deep crust
beneath island arcs: Inherited yircons reveal a gondwana continental frag-
ment beneath East Java, Indonesia: Earth and Planetary Science Letters
(EPSL), v. 258, p. 269–282.
Soeria-Atmadja, R., Maury, R.C., Bellon, H., Pringgoprawiro, H., Polve, M., and
Priadi, B., 1994, Tertiary magmatic belts in Java: Journal of Southeast Asian
Earth Sciences, v. 9, p. 13–17, doi: 10.1016/0743-9547(94)90062-0.
Song, S.R., Chen, C.H., Lee, M.Y., Yang, T.F., Iizuka, Y., and Wei, K.Y., 2000,
Newly discovered eastern dispersal of the youngest Toba Tuff: Marine
Geology, v. 167, p. 303–312, doi: 10.1016/S0025-3227(00)00034-7.
Taylor, B., and Karner, G.D., 1983, On the evolution of marginal basins:
Reviews of Geophysics and Space Physics, v. 21, p. 1727–1741.
Untung, M., and Sato, Y., 1978, Gravity and Geological Studies in Java, Indo-
nesia: Geological Survey of Indonesia and Geological Survey of Japan
Special Publication 6, 207 p.
van Bemmelen, R.W., 1949, The Geology of Indonesia: The Hague, Nijhoff,
Government Printing Office, 732 p.
van Hattum, M.W.A., 2005, Provenance of Cenozoic rocks of northern Borneo
[Ph.D. thesis]: University of London, 457 p.
van Hattum, M.W.A., Hall, R., Pickard, A.J., and Nichols, G.J., 2006, Southeast
Asian sediments not from Asia: Provenance and geochronology of north
Borneo sandstones: Geology, v. 34, p. 589–592, doi: 10.1130/G21939.1.
van Leeuwen, T.M., and Muhardjo, 2005, Stratigraphy and tectonic setting of
the Cretaceous and Paleogene volcanic-sedimentary successions in north-
west Sulawesi, Indonesia: Implications for the Cenozoic evolution of
Western and Northern Sulawesi: Journal of Asian Earth Sciences, v. 25,
p. 481–511, doi: 10.1016/j.jseaes.2004.05.004.
Wakita, K., 2000, Cretaceous accretionary-collision complexes in central Indo-
nesia: Journal of Asian Earth Sciences, v. 18, p. 739–749, doi: 10.1016/
S1367-9120(00)00020-1.
Wakita, K., and Munasri, B.W., 1994, Cretaceous radiolarians from the Luk-
Ulo Melange Complex in the Karangsambung area, Central Java, Indo-
nesia: Journal of Southeast Asian Earth Sciences, v. 9, p. 29–43, doi:
10.1016/0743-9547(94)90063-9.
Wakita, K., Sopaheluwakan, J., Miyazaki, K., and Zulkarnain, I., and Munasri,
1996, Tectonic evolution of the Bantimala Complex, South Sulawesi,
Indonesia, in Hall, R., and Blundell, D.J., eds., Tectonic Evolution of SE
Asia: Geological Society [London] Special Publication 106, p. 353–364.
Waltham, D., Hall, R., Smyth, H.R., and Ebinger, C., 2008, this volume, Basin
formation by volcanic loading, in Draut, A.E., et al., eds., Lessons from
the Stratigraphic Record in Arc Collision Zones: Geological Society of
America Special Publication 436, doi:10.1130/2008. 2436(02).
Watts, K.J., 1997, The Northern Nam Con Son Basin petroleum system, based
on exploration data from block 04D2, Vietnam, in Howes, J.V.C., and
Noble, R.A., eds., Proceedings of the International Conference on Petro-
leum Systems of SE Asia & Australia: Jakarta, Indonesian Petroleum
Association, p. 481–498.
Wheller, G.E., Varne, R., Foden, J.D., and Abbot, M.J., 1987, Geochemistry
of Quaternary volcanism in the Sunda-Banda arc, Indonesia, and three-
component genesis of island-arc basalt magmas: Journal of Volcanol-
ogy and Geothermal Research, v. 32, p. 137–160, doi: 10.1016/0377-
0273(87)90041-2.
Wilson, M.E.J., and Lokier, S.W., 2002, Siliciclastic and volcaniclastic influences
on equatorial carbonates: Insights from the Neogene of Indonesia: Sedi-
mentology, v. 49, p. 583–601, doi: 10.1046/j.1365-3091.2002.00463.x.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

New constraints on the sedimentation and uplift history of the Andaman-Nicobar

accretionary prism, South Andaman Island

R. Allen
Department of Environmental Science, Lancaster University LA1 4YQ, UK

A. Carter
Research School of Earth Sciences, Birkbeck and University College London, Gower St., London WC1E 6BT, UK

Y. Najman
Department of Environmental Science, Lancaster University LA1 4YQ, UK

P.C. Bandopadhyay
Geological Survey of India, Geodata Division, Salt Lake, Kolkata, 91, India

H.J. Chapman
M.J. Bickle
Department of Earth Sciences, Cambridge University, Downing St., Cambridge CB2 3EQ, UK

E. Garzanti
G. Vezzoli
S. Andò
Dipartimento di Scienze Geologiche e Geotecnologie, Universita Milano-Bicocca, Piazza della Scienza 4, 20126 Milano, Italy

G.L. Foster
Department of Earth Sciences, Bristol University, Queens Rd., Bristol BS8 1RJ, UK

C. Gerring
Department of Earth Sciences, The Open University, Walton Hall, Milton Keynes MK7 6AA, UK

ABSTRACT

The Andaman Islands are part of the Andaman-Nicobar Ridge, an accretionary

complex that forms part of the outer-arc ridge of the Sunda subduction zone. The
Tertiary rocks exposed on the Andaman Islands preserve a record of the tectonic
evolution of the surrounding region, including the evolution and closure of the Tethys
Ocean. Some of the Paleogene sediments on Andaman may represent an offscraped
part of the early Bengal Fan. Through field and petrographic observations, and use of
a number of isotopic tracers, new age and provenance constraints are placed on the

Allen, R., Carter, A., Najman, Y., Bandopadhyay, P.C., Chapman, H.J., Bickle, M.J., Garzanti, E., Vezzoli, G., Andò, S., Foster, G.L., and Gerring, C., 2008, New
constraints on the sedimentation and uplift history of the Andaman-Nicobar accretionary prism, South Andaman Island, in Draut, A.E., Clift, P.D., and Scholl,
D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special Paper 436, p. 223–255, doi:
10.1130/2008.2436(11). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All rights reserved.

223
224 Allen et al.

key Paleogene formations exposed on South Andaman. A paucity of biostratigraphic

data poorly define sediment depositional ages. Constraints on timing of deposition
obtained by dating detrital minerals for the Mithakhari Group indicate sedimenta-
tion after 60 Ma, possibly younger than 40 Ma. A better constraint is obtained for the
Andaman Flysch Formation, which was deposited between 30 and 20 Ma, based on
Ar-Ar ages of the youngest detrital muscovites at ca. 30 Ma and thermal history mod-
eling of apatite fission-track and U-Th/He data. The latter record sediment burial and
inversion (uplift) at ca. 20 Ma. In terms of sediment sources the Mithakhari Group
shows a predominantly arc-derived composition, with a very subordinate contribution
from the continental margin to the east of the arc. The Oligocene Andaman Flysch at
Corbyn’s Cove is dominated by recycled orogenic sources, but it also contains a sub-
ordinate arc-derived contribution. It is likely that the sources of the Andaman Flysch
included rocks from Myanmar affected by India-Asia collision. Any contribution of
material from the nascent Himalayas must have been minor. Nd isotope data discount
any major input from cratonic Greater India sources.

Keywords: Andaman, accretionary wedge, arc, subduction, thermochronology, prov-

enance, uplift.

INTRODUCTION aim of this paper is to build on previous field-based studies and

The Andaman-Nicobar Islands are part of an accretionary
complex that forms the outer arc ridge of the northern Sunda
subduction zone (Fig. 1). The Andaman Islands are in the south-
eastern part of the Bay of Bengal and make up part of a 3000–
5000 km chain that runs from the Myanmar Arakan-Yoma down
to Sumatra and Java in the south. The Indian plate is subducting
northward below the Euarasian plate and obliquely below the
Sino-Burman plate along the Burma-Andaman-Java Trench. The
structure of the Andaman Islands comprises an accretionary prism
formed by an imbricate stack of east-dipping fault slices and folds
that young to the west (Fig. 2), linked to a westward-shifting sub-
duction zone (e.g., Roy, 1992; Pal et al., 2003). The geology of an
accretionary wedge is complex, reflecting its dynamic environ-
ments and involving subduction, folding, and thrusting. Deposi-
tional ages and environments can change abruptly over relatively
short distances, and uplift leads to recycling of sediment from the
eroding wedge. Throughout subduction, new material introduced
at the bottom of the accretionary wedge is accreted, uplifted, or
subducted. Some of the accreted material may be uplifted and
brought to the seafloor. Slope basins may develop behind folds in
the accreted sedimentary rocks and trap sediment in a deep-water
environment. Simultaneously, shallow-water sediments such as
reefs can form on the prism top and be eroded and transported
down the slope. Given this inherent complexity, many accretion-
ary complex rocks are referred to as mélange; thus unraveling
the sedimentation history in a subduction-accretionary setting is
a major task.
Interpretation of the geology of the Andaman Islands is
hindered by the lack of isotopic age constraints, limited bio-
stratigraphy, and poor outcrop exposure (Bandopadhyay and
Ghosh, 1999, Bandopadhyay, 2005; Pal et al., 2003, 2005). The
apply petrologic, isotopic, and thermochronometric techniques
to better understand the provenance, sediment deposition, and
uplift history.
The origin of the Andaman Flysch has been debated for
>20 yr. It has been variously proposed that the Andaman Flysch
was derived from the Irrawaddy Delta (Karunakaran et al., 1968;
Pal et al., 2003) or, alternatively, from Bengal Fan material shed
from the nascent Himalaya sourced either directly or by emplace-
ment as an allochthon into the accretionary prism by oblique sub-
duction (Curray et al., 1979; Curray, 2005). We are particularly
interested to determine whether the Himalayan-Tibetan orogen
contributed sediment to the Andaman Islands, because this might
reveal information on the early evolution of the orogen not pre-
served elsewhere.
LITHOLOGIES
Previous Work
The current stratigraphy (Table 1) of the Andaman Islands
is based on lithological mapping and can be traced back to
the pioneering work of Oldham (1885), who first divided the
Andaman geology into an older Port Blair Series and a younger
Archipelago Series, separated by volcanic rocks and serpenti-
nites later recognized as an ophiolite. Over the past 50 yr the
stratigraphy has been modified and formation names changed,
but it was not until the 1960s that paleontological constraints
were used to place the Paleogene–Neogene lithostratigraphic
units within a temporal framework (Guha and Mohan, 1965;
Karunakaran et al., 1968). The stratigraphy now comprises four
units, which, in ascending order, are Cretaceous sedimentary
rocks and ophiolite, the Eocene Baratang-Mithakhari Group,
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism 225

90 100 drilled to the east and west of the islands (Fig. 2) that helped
Himalaya g es
n place the exposed geology within the context of the accretionary
Ra Mogok setting. Below we briefly describe the main rock units exposed
Belt

a
Shillong

rm
on the Andaman Islands and report key observations from previ-

obu
Bengal Basin Myanmar Shan
ous field and petrographic studies.

I nd
Plateau
Central
India Basin 20 Ophiolite
Bengal Fan Thailand The Andaman ophiolite contains the main components of an
ophiolite sequence that includes upper mantle–depleted harzbur-
Andam gites and dunites, lower crustal cumulate gabbros and peridotites,
and upper crustal sheeted dikes, pillow lavas, and marine pelagic
sediments (Halder, 1985; Ray et al., 1988; Roy, 1992). However,
Andaman
the sequence is tectonically disturbed, and much of the crustal
an Nicobar R

Sea
10
section is deformed and difficult to identify in the field. Pillow
Mergui lavas are abundant, but sheeted dikes have been identified only
Ma

basin in disrupted small-scale faulting and folding. Both massive and

lay

layered gabbros are recognized in the preserved ophiolite. South

idge

Pe

Andaman has the best preserved and most complete sequence of

nin

ophiolite, which extends for ~30 km from Corbyn’s Cove in the

su
la

north to Chiriyatapu in the south (Fig. 3).

Su

0
Pelagic Sedimentary Rocks
ma
tra

The topmost part of the ophiolite complex contains thin and

discontinuous lenses and streaks, and laterally continuous (at
95 E outcrop scale) bedded sequences of pelagic sedimentary rocks
consisting of jasper, chert, cherty limestone, and shales (Bando-
Myanmar padhyay and Ghosh, 1999). Outcrops commonly show evidence
Bengal Fan of significant deformation and folding (Fig. 4).

Bedded Chert
Irrawaddy delta Rhythmic alternations of centimeter-thick, milky white chert
15 N and millimeter- to centimeter-thick reddish-brown and purple
shale-mudstone beds constitute the bedded chert facies. Chert and
shale normally show uniform (0.5–4.0-cm-thick) beds that have
sharp bases and tops and planar contacts. Some 10–15-cm-thick
nch

North beds, and massive beds, of chert are present. Soft-sediment defor-
Andaman v
e

vv mation is evident in some localities. Radiolarians are preserved

obar Tr

v Andaman Sea to varying degrees in most cherts and indicate a Late Cretaceous
v
Richie’s to Paleocene depositional age, which constrains the underlying
an - Nic

arc

Archipelago ophiolite sequence to a Late Cretaceous age.

& Havelock
nic

South v Shale
a
Andam

Andaman
Volc

v At outcrop the shale facies form interbedded sequences

vv
vv
vv
Figure 1. General location map of Andaman Islands and location of
potential source regions. Lower figure corresponds to boxed area in
top figure.
the Eocene to upper Oligocene Port Blair–Andaman Flysch
Group, and the lower to upper Miocene Archipelago Group. In
the mid-1970s the Indian state Oil and Natural Gas Commission
(ONGC) carried out a detailed seismic reflection study across
the Andaman Islands, calibrated against offshore boreholes
of extremely variable thickness and lateral continuity. Basaltic
volcanic rocks occur as thin intercalations, conformable lenses,
and, at places, small crosscutting dikes. Thin beds of fine-grained
sandstone, siltstone, and cherty limestone are also present. There
is evidence of soft-sediment deformation and slump folds as well
as cutting by normal and thrust faults. Some shales are clearly
tuffaceous with plagioclase phenocrysts, vitric fragments, pum-
ice clasts, and diagenetically altered volcanic lithic fragments.
Chlorite is abundant in the matrix of the altered tuff and tuffa-
ceous shale. Occasionally sharp edged, cuspate or platy, fresh
glass shards can be found (Bandopadhyay and Ghosh, 1999).
226 Allen et al.

A Andaman - Nicobar A'

0-
two-way time (sec) West East
1- 50 km

2-

3- Middle to
Andaman - Nicobar
nch

Ridge
South Andaman
4- B B'
a Tre

0-
N
Sund

1000-

depth (m)
2000-

3000-
B
B'
25 Km
Melange/ Neogene Archipelago Group
Cretaceous

A
A' deformed
Upper

pelagic cover Andaman Flysch

Paleogene
Ophiolite Mithakhari Group

Figure 2. Cross section through Andaman Ridge, based on seismic sections from Roy (1992) and Curray (2005) that show
the underlying structure of the accretionary wedge, comprising a series of folded east-dipping thrust slices.

TABLE 1. SIMPLIFIED STRATIGRAPHY OF THE ANDAMAN ISLANDS

Approximate depositional Group Formation Lithology
age range
Cross-stratified and graded sandstones, silty
mudstones and limestones marls, and chalky
Miocene to Pliocene Archipelago Group
limestones

Andaman Flysch Group Bouma sequences, sandstone-shale and

(formerly Port Blair mudstones
Oligocene–late Eocene(?)
Group)

Pebbly and coarse to fine-grained volcaniclastic

Namunagarh Grit
sandstones and grits
Mithakhari Group
Interstratified massive and graded polymict
Early to middle Eocene(?) (formerly Baratang and
Hope Town Conglomerate conglomerates, massive cross-stratified and
Port Meadow Groups)
graded sandstones, shales, and thin coals
Lipa Black Shale Pyritiferous black shale

Pillow lava, basalt, gabbro, pyroxinite,

harzburgite, serpentinite, andesite, diorite,
Late Cretaceous to
Ophiolite Group plagiogranite, rhyolite, serpentinized
Paleocene(?)
harzburgite, pyroxinite, and pelagic sediments;
radiolarian chert and hematitic mudstones

Mithakhari Group
The Mithakhari Group consists of immature gravels and
coarse- to fine-grained sandstones, pebbly to fine-grained pyro-
clastic sandstones, and minor thin beds of mudstones and coal.
Whereas the Mithakhari Group dominates the outcrop geology of
the Andaman Islands, particularly in North and Middle Andaman,
a paucity of good exposures and poor access make it difficult to
obtain continuous sections, and exposures are limited to isolated
stone quarries, coastal areas, and road cuts. In South Andaman
the Mithakhari Group occurs as a north-south–trending outcrop
that extends for ~50 km, but the best sections, exposing the least
weathered outcrops, are found only near Hope Town, Mungleton,
Namunagarh, and Chiriyatapu (Fig. 3). Karunakaran et al. (1968)
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism 227

92.30' 92.38' 92.43' A)

Mithakhari Hope
South Group Town

Andaman
Namunagarh
Mungleton
Port Blair 11.39'
Hobdaypur

Andaman Corbyn's
12 Flysch Cove

Ophiolite Figure 3. General location map of An-

daman Islands and local geology (A, B)
for the areas on South Andaman Island
3km
sampled for this study.

B)
Andaman
Andaman
Flysch
Flysch Ophiolite

A) Mithakhari
Group

Chiriyatapu 11.29'

B) 92.42'

Hope Town Conglomerates

This unit is best seen near Hope Town on South Andaman
Island, where ~6 m of conglomerates and pebbly sandstones are
well exposed (Fig. 5), interbedded with thin beds of greenish
gray coarse- and fine-grained sandstones. The sequence shows
fining- and thinning-upward sequences with evidence of slump-
ing and soft sediment deformation. Bed contacts are generally
sharp and planar, and some evidence of fluvial channels can be
found. Conglomerates are polymict, of mainly basic-ultrabasic
sources, and with subordinate to minor amounts of andesite, sedi-
mentary limestone, and cherts plus sporadic mudstone clasts and
metamorphic quartz.

Namunagarh Grit Formation

Figure 4. Folded cherts and shales that represent the pelagic cover to
the Cretaceous ophiolite exposed on the shore at Chiriyatapu, South
Andaman Island.
first introduced the term Mithakhari Group, dividing the group
into a lower Lipa Black Shale, a middle Hope Town Conglom-
erate, and an upper Namunagarh Grit Formation (Table 1). The
Lipa Black Shale is a minor unit and not well exposed, and so
will not be considered further.
This unit is characterized by coarse- to fine-grained sand-
stones and siltstone, with minor conglomerate at the base. On
South Andaman the type section and best exposures are found in
quarry sections near Namunagarh village (Fig. 3). These display
3–5-m-thick, green, matrix-supported sandstones. The sandstones
are well bedded and laterally persistent along the quarry sections,
and consist of coarse- and fine-grained beds. The coarse-grained
beds, at the base of the section, are ≥1 m thick, with sharp non-
erosive contacts. The finer grained beds consist of 4–8-cm-thick
228
A)
1m
B)
10cm
Figure 5. Massive polymict conglomerates exposed at Hope Town
Quarry (A), South Andaman Island. These polymict conglomerates
are dominated by basic and ophiolitic clasts (B) whereas the matrix
contains abundant volcanic glass and felsitic volcanic grains, suggest-
ing a dominant volcanic arc source, although zircon fission-track data
also show evidence of continental Mesozoic sources.
beds of fine- to medium-grained volcaniclastic sandstones inter-
bedded with thin mudstones. Some large fragments of volcanic
rock fragments, including elongated pumice lapilli, are present,
which resemble floating clasts in turbidites. For a long time the
sandstones exposed at Namunagarh stone quarries had been
described as graywacke formed from weathering and erosion of
accreted ophiolite (Acharyya et al., 1989). Recently Bandopad-
hyay (2005) identified beds with abundant pyroclasts, including
vesiculated glass fragments, pumice clasts and shards, euhedral
feldspars, and angular lithic fragments diagnostic of tuff, indicat-
ing that some of the Namunagarh Grit beds were derived from
direct volcanic arc sources (Fig. 6).
The polymict conglomerates (Hope Town Conglomerate)
and grits (Namunagarh Grit Formation) are interpreted as having
Allen et al.
Figure 6. Namunagarh Quarry. Thin sections show that the sediments
are largely tuffaceous with clear evidence of fresh arc volcanic mate-
rial including devitrified glass.
been derived from the ophiolite and its pelagic–shallow marine
cover. The succession, which includes thin coals and gypsum, was
deposited in a delta-slope setting with facies associations ranging
from subaerial alluvial plain to prodelta slope (Chakraborty et al.,
1999). Its depositional age is not well defined owing to the lack
of distinct biostratigraphic evidence, but shallow benthic fora-
minifers in the Hope Town Conglomerate, including Nummulites
atacicus, constrain the age from late Ypresian to early Lutetian
(Karunakaran et al., 1968). Many of the foraminifers, however,
are broken and abraded (i.e., reworked). The relationship of the
Namunagarh grits to the Hope Town conglomerates is not clear,
although the Namunagarh grits are presumed to be younger.
Andaman Flysch
The Andaman Flysch is a siliciclastic turbidite sequence
deposited on a submarine fan. It is bounded between the
Mithakhari Group below and the Archipelago Group above.
The (misleading) term flysch is derived from the resemblance
of the turbidites to the classic Bouma turbidites described in the
Swiss Alps. Similar looking beds are seen throughout the Anda-
man Islands; hence the Andaman Flysch is described as crop-
ping out over a N-S strike length of 200 km from the southern
part of South Andaman to the northern tip of North Andaman.
The overall thickness is not well defined, with estimates varying
from 750 m (Roy, 1983) to 3000 m (Pal et al., 2003). The best
and most completely documented exposures are found on South
Andaman at Corbyn’s Cove (Fig. 3), where outcrops of steep,
westerly dipping beds are seen adjacent to the pillow basalt of
the ophiolite sequence (Fig. 7), although the nature of the con-
tact is uncertain. Individual sandstone beds can be traced along
strike for distances of several kilometers, but the total thickness is
only 250–300 m. Current directional structures in sandstone beds
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism
Figure 7. Outcrop of the Andaman Flysch, looking south across
Corbyn’s Cove (see Fig. 3) toward outcrops of pillow basalts and
ophiolite.
include flute casts, groove casts, and current bedding. The orien-
tation of flute casts at the base of overturned sandstone beds near
Corbyn’s Cove reveals southward-directed paleocurrents (Pal et
al., 2003). The relationship between the turbidites and underlying
lithostratigraphic units is unclear. Onlap of the Andaman Flysch
with the Mithakhari Group has been reported (Chakraborty and
Pal, 2001), but no supporting evidence was found in this study.
However, there is a marked change in lithology and provenance
with up to 50% quartz in the Andaman Flysch in contrast to
the relatively quartz-free Mithakhari Group (Pal et al., 2003).
Lithic fragments in the Andaman Flysch range from micaceous
metamorphic clasts diagnostic of continental sources to cherts,
basalts, and weathered volcanic glass consistent with derivation
from volcanic arc and ophiolite sources. Biostratigraphic evi-
dence is vague and spans the Oligocene to the early Miocene,
ca. 36–21 Ma (Pal et al., 2003).
Archipelago Group
The Archipelago Group represents the topmost stratigraphic
unit of the Tertiary succession. The lower units comprise basal
conglomerates and sandstones, overlain by calcareous arenites
of the Strait Formation. This is followed by chalk and limestone
with some argillaceous limestones and shale, described as the
Melville Limestone (Shell Limestone) Formation. Deposition
was mostly in a slope environment (Pal et al., 2005; Roy, 1983).
These sedimentary rocks most likely covered most of Andaman,
but recent uplift and erosion means that today only small patches
can be found on the main islands, with most exposures confined
to Havelock Island and associated smaller islands to the east of
South Andaman in Richie’s Archipelago (Fig. 1). Radiolarians,
planktonic foraminifers, and calcareous nannofossils from John
Lawrence Island (Singh et al., 2000) indicate a maximum depo-
sitional age for the calcareous chalk in the Archipelago Group
of 18.3–15.6 Ma (Burdigalian to Serravallian), but elsewhere the
Archipelago Group may span any age from Miocene to Pliocene
(Pal et al., 2005).
229
METHODOLOGICAL DETAILS
The aim of this work is to provide improved understanding
of the burial and uplift history of the accretionary wedge and
provenance of the constituent sedimentary rocks, now exposed
on the Andaman Islands, through the application of thermochro-
nometric, petrographic, and geochemical analyses of these rocks.
Multiple proxies are used to get the best insight to the source
provenance and to avoid any potential bias that might arise by
relying on a single type of mineral. For example, mica is gener-
ally not present in arc-derived volcanic rocks, whereas apatite
and zircon are. Sampling is confined to South Andaman, where
there are accessible exposures of the main type localities, stud-
ied by Bandopadhyay and Ghosh, (1999), Chakraborty and Pal
(2001), Pal et al. (2003, 2005), and Bandopadhyay (2005). Both
Middle and North Andaman are less developed and thus are less
accessible, with extensive areas of jungle that include large tribal
reserves that are restricted to aboriginal peoples.
Samples from South Andaman were collected for detrital
thermochronometric, heavy-mineral, biostratigraphic, and Sm-Nd
whole rock and single grain analyses.
Zircon and Apatite Fission-Track Analysis
Apatite and zircon fission-track (FT) analyses are used to
define provenance and low-temperature histories, and for places
where sediments have been subjected to significant (>1.5 km)
burial to determine their postdepositional burial and uplift his-
tory. Fission tracks in apatite are sensitive to relatively low tem-
peratures (typically <60–110 °C) and are ideally suited to con-
strain levels of postdepositional burial and timing of subsequent
rock uplift and exhumation (e.g., Green et al., 1995). Zircon
FT data record higher temperature cooling histories (typically
~200–310 °C) and for sedimentary rocks are more suited to
provenance studies that record either volcanic formation ages or
postmetamorphic cooling-exhumation ages (Carter, 1999). Sam-
ples for FT analysis were irradiated in the well-thermalized (Cd
ratio for Au >100) Hifar Reactor at Lucas Heights in Australia.
Apatite grain compositions were monitored using etch pits and
direct measurement on a JEOL microprobe using a defocused 15
KeV Beam to prevent F and Cl migration. Durango apatite and
rock salt were used as standards. Samples with mixed ages, indi-
cated by χ2 values <5% and large age-dispersion values (>20%),
were deconvolved into their principal age components using the
approach of Sambridge and Compston (1994) and incorporating
the method of Galbraith and Green (1990).
Apatite Helium Analysis
Apatite helium dating was used to provide additional con-
straint on rock uplift histories. This method complements FT
dating, as it is sensitive to closure at ~60 °C, the temperature at
which FT begins to lose sensitivity to changes in cooling rate.
Helium ages were based on replicate analyses of apatite grains
230 Allen et al.
handpicked to avoid mineral and fluid inclusions. Each selected
grain was first photographed, and then its dimensions were
recorded for later alpha-ejection correction. Samples were loaded
into platinum microtubes for helium outgassing and U/Th deter-
mination. Outgassing was achieved using an induction furnace
at a temperature of 950 °C. The abundance of 4He was measured
relative to a 99.9% pure 3He spike in a Pfeiffer Prisma 200 quad-
rupole mass spectrometer. The quantification of U/Th was per-
formed on an Agilent 7500 quadrupole mass spectrometer using
spiked solutions of the dissolved apatite. Repeated analysis of
the California Institute of Technology (CIT) laboratory Durango
apatite standard gives an age of 31.3 ± 1.2 Ma (2σ), based on 39
analyses. This error of the mean (6.7%), combined with the U/Th
and He analytical uncertainties, is used as a measure of the total
uncertainty in sample age.
Ar-Ar Age Dating of Detrital White Micas
The 40Ar/39Ar age of a detrital muscovite records the timing of
cooling and exhumation (or crystallization) of rocks in the source
region, aiding the discrimination of potential source regions for
clastic sequences (e.g., Sherlock and Kelley, 2002; Haines et al.,
2004). For this study, single grains were totally fused using an
infrared laser ablation microprobe (IRLAMP). Samples were
monitored using the GA1550 biotite standard with an age of 98.8
± 0.5 Ma (Renne et al., 1998). The calculated J value for the sam-
ples was 0.0138 ± 0.000069. Blanks were measured both before
and after each pair of sample analyses, and the mean of the two
blanks was used to correct the sample analyses for the measured
isotopes. Overall mean blank levels for 40Ar, 39Ar, and 36Ar were
(378, 6, and 11) × 10−12 cm3 at a standard temperature and pres-
sure. The resulting analyses were also corrected for mass spec-
trometer discrimination, 37Ar decay, and neutron induced interfer-
ences. The correction factors used were (39Ar/37Ar)Ca = 0.00065,
(36Ar/39Ar)Ca = 0.000264, and (40Ar/39Ar)K = 0.0085; these were
based on analyses of Ca and K salts. Samples were irradiated for
33 h in the McMaster University reactor (Canada).
U-Pb Dating of Detrital Zircon
Zircon U-Pb dating reflects the time of zircon growth, which
in most cases is the igneous rock’s crystallization age. The U-Pb
system is mostly unaffected by high-grade metamorphism and
is effectively stable up to ~750 °C (Cherniak and Watson, 2001;
Carter and Bristow, 2000). Zircon U-Pb ages from detrital grains
in a sedimentary rock are therefore expected to be representa-
tive of the range of crustal ages within the contributing drainage
basin. Samples for this study were analyzed at University College
London by laser ablation-inductively-coupled-plasma mass spec-
trometry (LA-ICP-MS) using a New Wave 213 aperture imaged
frequency quintupled laser ablation system (213 nm) coupled to
an Agilent 750 quadrupole ICP-MS. Real-time data were pro-
cessed using GLITTER and repeated measurements of the exter-
nal zircon standard PL (Svojtka et al., 2001; TIMS reference age
337.1 ± 0.7 Ma) to correct for instrumental mass bias. The results
have not been corrected for common lead or ranked according to
degree of discordance, as the latter involves choosing an arbitrary
value and is therefore open to analyzer bias.
Sm-Nd Isotope Analysis
Whole rock Sm and Nd isotope data in sedimentary rocks
are widely used to fingerprint sediment source. 143Nd/144Nd
ratios are generally normalized and expressed in epsilon units as
deviation from a chondritic uniform reservoir (CHUR), where
εNd = 0. A single epsilon unit is equivalent to a difference in the
143
Nd/144Nd ratio at the 4th digit. For clastic sedimentary rocks,
εNd will in part represent the weighted average of the time when
the sediment sources were extracted from the mantle. When melt
is extracted from the mantle, it has a lower Sm/Nd ratio than its
parent and therefore evolves over time to have a lower εNd than
CHUR; the residual has a higher Sm/Nd than CHUR, evolving to
a higher εNd over time.
For this study, sandstone and mudstone samples were collected
from type localities from the Andaman Flysch and the Mithakhari
Group. Whole rock samples were ignited overnight at 900 °C to
remove any organic material. Dissolution and analytical methods
follow Ahmad et al. (2000), with the exception that the samples
were spiked with a mixed 150Nd-149Sm spike and the 143Nd-144Nd
ratios were measured on the spiked fraction. εNd is calculated
relative to the present day (i.e., at t = 0) using CHUR 143Nd/144Nd
= 0.512638. Sm and Nd blanks were <10−3 of the sample, and the
laboratory Johnson Mathey Nd internal standard gave 143Nd/144Nd
= 0.511119 ± 5 (1σ = 24) over the period of the analyses. As whole
rock εNd typically represents the weighted average of sediment
sources, further insight into the origin of the Andaman Flysch was
achieved by analyzing the Nd isotopic character of single apatite
grains. This was achieved using a 193 nm homogenized ArF New
Wave/Merchantek laser ablation system linked to a ThermoFini-
gan Neptune multicollector mass spectrometer at the University of
Cambridge (UK). All ablation was carried out in a He environment
and mixed with Ar and N after the ablation cell. Laser spot sizes
were 65–90 µm. During the analytical period, standards repro-
duced to better than 0.5 epsilon units, while samples typically gave
internal precisions of 1–2 epsilon units. The full methodology of
this in situ approach is detailed in Foster and Vance (2006).
Petrography and Heavy Minerals
A total of 400 points were counted in six selected samples
according to the Gazzi-Dickinson method (Dickinson, 1985). A
classification scheme of grain types allowed for the collection of
fully quantitative information on the sampled sandstones. Trans-
parent dense minerals were counted on grain mounts according
to the “ribbon counting” method, and 200 minerals were counted
also to assess the percentage of etched and corroded grains. Dense
minerals were concentrated with sodium metatungstate (density,
2.9 g/cm3) using the 63–250 µm fraction treated with hydrogen
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism
peroxide, oxalic acid, and sodium ditionite to eliminate organic
matter, iron oxides, and carbonates, respectively.
RESULTS AND INTERPRETATION
Biostratigraphy
Attempts to identify new, more robust biostratigraphic con-
trol for the Paleogene sedimentary rocks, based on nannofossils,
failed owing to the barren nature of the mudstones. Samples of
mudstone were taken from the Archipelago Group, Andaman
Flysch Group, and the Mithakhari Group. Whereas nannofossils
are present in the calcareous Neogene sedimentary rocks (Archi-
pelago Group) (Singh et al., 2000), we can only conclude that
nannofossils either were never present in the Paleogene sedimen-
tary rocks or have since been dissolved by weathering or dis-
solution below the carbonate compensation depth. Similarly, the
sampled rocks yield few (as yet undated) foraminifers. Those
found were either broken or abraded, consistent with rework-
ing. At Chiriyatapu, clasts of limestone were present within the
coarse-grained Mithakhari sedimentary rocks (Fig. 8A). These
were found to have shallow marine reef assemblages, including
Nummulites spp., small miliolids and rare Morozovella spp., frag-
ments of rhodophyte algae, and dasycladaceans (Belzungia spp.)
of Thanetian–Ypresian age (ca. 58–49 Ma) (Fig. 8B).
Fission-Track and (U/Th)-He Thermochronometry
Samples from the Mithakhari Group contained relatively
low concentrations of heavy minerals and yielded fewer apa-
tites and zircons than those from the Andaman Flysch. For this
B) A)
C)
Figure 8. A and B show clasts of limestone from Mithakhari sedi-
ments. These contain shallow-marine reef assemblages that in thin sec-
tion (C) are seen to include Nummulites spp., small miliolids and rare
Morozovella spp., fragments of rhodophyte algae, and dasycladaceans
(Belzungia spp.) that indicate a Thanetian–Ypresian depositional age.
No in situ outcrops of these beds have been found on South Andaman.
231
reason, FT data sets from the Mithakhari Group were smaller in
comparison with those from the Andaman Flysch. Nevertheless
the results adequately provide a measure of the underlying detri-
tal FT signatures for each key lithostratigraphic unit. Data are
summarized as radial plots in Figures 9 and 10, and tabulated in
Table A1 (Appendix).
The interpretation of detrital apatite FT data requires com-
paring youngest ages with sediment depositional age. If all single
grain ages are less than depositional age, total resetting took place
(generally indicating burial heating to >100–120 °C; e.g., Green
et al., 1995, their Fig. 12), whereas if the population of measured
single grain ages ranges from younger to older than depositional
age, partial resetting has taken place (generally indicating burial
heating to <100 °C). The main issue with interpreting the Anda-
man FT data is that suitable depositional ages are missing, pre-
venting robust use of FT analysis to determine exhumation rates.
Samples from the Hope Town Conglomerate collected at the
Hope Town quarry (Fig. 5) yield a single population of apatites
with an age of 57 ± 9 Ma. The large uncertainty in age is due
to low uranium concentrations. The zircon ages comprise three
age modes, with the youngest (majority of analyzed grains) at
61 ± 2 Ma, within error of the apatite age. The other zircon ages
indicate sources with Late Cretaceous and Permian cooling sig-
natures. Given that the zircon grain ages from the Hope Town
Conglomerate are at or older than the Eocene biostratigraphic age
(maximum age owing to the reworked nature of the fossils), the
FT ages must reflect different sources. Furthermore, given that
the apatite grains have the same age as the youngest zircon ages,
the apatite data must also reflect provenance.
The youngest apatite and zircon ages at ca. 60 Ma constrain
depositional age to being at or after this time for the Hopetown
Conglomerate Formation at this location. The apatites are notice-
ably euhedral, contain variable chlorine, and are low in uranium
(Fig. 11), typical of volcanic apatites, consistent with sample
petrography that records a dominant flux of volcanic detritus
(see below). Similar FT ages and volcanic affinities are seen in
the apatite and zircon data from the Mungleton quarry (20 km
inland from the Hope Town quarry; Fig. 5), exposing a 6-m-thick
sequence of interbedded greenish-gray, fine- to medium-grained
sandstones, siltstones, and mudstones, underlain by conglom-
erates. The succession here was described as the Namunagarh
Grit Formation. In contrast, the apatite FT data from a quarry at
Namunagarh village (also mapped as Namunagarh Grit Forma-
tion) gave a single population age of 40 ± 4 Ma, based on 24
grains. The zircon content of this sample was too low for FT
analysis. This quarry, studied by Bandopadhyay (2005), com-
prises tuffaceous beds with well-preserved pumice fragments and
glass shards. The apatites are euhedral and have variable chlorine
and low uranium contents typical of volcanic sources. Detrital
assemblages in these rocks contain pyroxenes, epidote, sphene,
green-brown hornblende, and chrome spinel that are consistent
with provenance from a volcanic arc. Major diagenetic dissolu-
tion is also evident, so it is questionable whether the apatite age
reflects source or resetting. No track lengths were measured as
 Apatite Andaman Flysch Zircon
Partially reset by burial Unreset by burial
AND-1a 100 AND-1a
400
Central Age: 31±2 Ma Central Age: 61±5 Ma 300
P(X^2): 0.0% P(X^2): 0.0% 200
Relative Error: 47% 147±10
Relative Error: 29%
50 No. of grains: 43
No. of grains: 38 100
40 +2
+2
0 30 0
67±4
-2 50
-2
20
44±2
% relative error % relative error
10 25
0 0
10 20 30 40 10 20 30 40
Precision (1/sigma) Precision (1/sigma)

AND-1B 400 342±83

AND-1B 300
Central Age: 32±2 Ma 100 Central Age: 54±4 Ma 200
P(X^2): 0.7% P(X^2): 0.0%
Relative Error: 22% 150
Relative Error: 41%
No. of grains: 60 No. of grains: 32 101±14
+2 50 100
40
0 +2
30
-2 0 ?50
20 -2 41±2
% relative error % relative error
10
0 0
10 20 10 20 30 40
Precision (1/sigma) Precision (1/sigma)

AND-3 100 AND-3 616±102

600
Central Age: 32±2 Ma 80 Central Age: 56±1 Ma 450
P(X^2): 0.7% P(X^2): 73% 300
Relative Error: 39% 60 Relative Error: 0%
No. of grains: 50 No. of grains: 15 150

+2 40 +2
0 30 0
-2 20 -2 60

36±2
?30
% relative error % relative error
0 10 0
10 20 30 40 10 20 30 40
Precision (1/sigma) Precision (1/sigma)
100 AND-4
AND-4
Central Age: 68±8 Ma 600
Central Age: 31±2 Ma
P(X^2): 0.09% P(X^2): 0.0% 270±16
300
Relative Error: 20% Relative Error: 70%
50 No. of grains: 42 200
No.of grains: 47
40
+2
+2 100
0 30
0
-2 -2
20 58±4
50
% relative error % relative error
0 10 0 38±1
10 20 30 40 10 20 30 40
Precision (1/sigma) 25
Precision (1/sigma)

Figure 9. Radial plots of fission-track data from the Andaman Flysch, showing the distribution of single grain
ages. Where the sample data comprise mixed grain ages the deconvolved age modes are also shown.
 Apatite Zircon
Unreset by burial Unreset by burial
Hopetown Conglomerate

Central Age: 57±9 Ma Central Age: 92±11 Ma 600

P(X^2): 69% P(X^2): 0.0% 400
Relative Error: 5% 200 Relative Error: 56% 280±32
No. of grains: 19 No. of grains: 25 200
+2
80
60 +2 129±10
0 100
40 0
-2 -2
20 61±2
10
% relative error % relative error
0 0 30
10 20 10 20 30
Precision (1/sigma) Precision (1/sigma)

Namunagarh Grit
Mangluton Mangluton
Central Age: 69±9 Ma Central Age: 75±8 Ma 400
P(X^2): 0.0% 400 314±51 P(X^2): 0.0% 226±21
Relative Error: 65% 300 Relative Error: 42% 200
No. of grains: 21 No. of grains: 18
200
+2 151±68 100
0 +2
100 0
-2
-2
50 55±8 59±2
% relative error 50
% relative error
0 0
10 20 10 10 20 30 40 30
Precision (1/sigma) Precision (1/sigma)

Namunagarh Chiriyatapu
Central Age: 40±4 Ma 100 Central Age: 143±22 Ma
P(X^2): 96.0% 80 P(X^2): 0.0%
Relative Error: 0% Relative Error: 93% 800
600
No. of grains: 24 60 No. of grains: 38 400
+2 329±17
+2
0 40 0 200
-2
-2
100
20 91±7

% relative error 10 % relative error

0 0 40±3
10 20 30 30
10 20 30
Precision (1/sigma) Precision (1/sigma)

Figure 10. Radial plots of fission-track data from the Mithakhari Group, showing the distribution of single grain
ages. Where the sample data comprise mixed grain ages the deconvolved age modes are also shown.
234 Allen et al.
1.5
Apatite grain composition
Hopetown Conglomerate
Andaman Flysch
1
Chlorine (wt%)
0.5
0
0 20 40 60 80
Uranium (ppm)
a result of the low spontaneous track density. A sample of the
Namunagarh Grit Formation from Chiriyatapu on the southern
coast of South Andaman yielded only zircon, which produced
three FT provenance age modes at 40 ± 3 Ma, 91 ± 7 Ma, and
329 ± Ma. The youngest zircon FT age mode, at 40 ± 3 Ma, indi-
cates that the Thanetian–Ypresian limestone clasts found in these
rocks were derived from erosion of significantly older material
(by ca. 16 Ma). Thus we conclude that the age of the Mithakhari
Group is constrained to be younger than 60 Ma at two locations,
and younger than 40 Ma at a third location. Nevertheless, the
volcanic origin of the material, with little sign of reworking, sug-
gests that the rocks were deposited only shortly after the recorded
mineral ages. This is consistent with diachronous deposition in
local basins, as expected in a subduction zone setting (e.g., Draut
and Clift, 2006).
Apatite and zircon FT data from four samples collected
through the section of Andaman Flysch at Corbyn’s Cove are
closely similar in age (most are 35–40 Ma) to the youngest age
modes in the Namunagarh Grit Formation beds. The youngest
ages comprise most analyzed grains for zircon and all analyzed
grains for the apatites. In addition, two of the samples show zir-
con age modes at 58–67 Ma, similar to the age modes detected in
the Mithakhari Group. Three of the samples also show Mesozoic
and Paleozoic FT ages. The Andaman Flysch is widely consid-
ered to be Oligocene in age, although biostratigraphic evidence
Figure 11. Plot comparing uranium and
chlorine for apatite grains from the An-
daman Flysch and Hopetown Conglom-
erate, Mithakhari Group. The plot clearly
shows that the apatite came from differ-
ent sources. The low uranium Hopetown
Conglomerate apatites, which also have
variable amounts of chlorine, are typical
of volcanic apatites. The volcaniclastic
petrography and euhedral form of the
apatites support this and suggest that
they came from arc related sources.
100 120
is not robust (Pal et al., 2003). While both zircon and apatite FT
ages are older than the Oligocene, some partial resetting (burial-
related heating) may have taken place.
Thermal Modeling
To constrain postdepositional thermal history, apatites from
one of the samples were analyzed by the (U/Th)-He method
(Table A2, Appendix). Replicates yielded an FT corrected age
of 16 ± 1 Ma, crudely representing the time at which the sam-
ple cooled to <50 ± 10 °C. To define more robustly the sample
postdepositional thermal history, the FT and helium data can be
jointly modeled. Ideally this requires incorporating a sample
depositional age, but, as discussed, this is not well defined, and
so we resorted to using the youngest detrital ages as an upper
limit for the time of deposition. For the Andaman Flysch, deposi-
tion must have taken place at or after 35–30 Ma on the basis of
the youngest argon mica ages. In addition the Andaman Flysch
sedimentary rocks must have been at or near surface tempera-
tures by the middle Miocene (ca. 16 Ma) because in the Hobday-
pur area of South Andaman (Fig. 2) these rocks are seen to be
conformably overlain by sedimentary rocks of the Archipelago
Group, although in the eastern part of the island Archipelago
rocks are juxtaposed with rocks of the Mithakhari Group along a
faulted contact (Pal et al., 2005). A phase of reburial up to ~60 °C
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism 235

is also required on the basis of recent diagenetic evidence from
the Archipelago Group sedimentary rocks (Pal et al., 2005).
With these constraints the combined apatite FT and helium data
were modeled, using the data-driven modeling approach of Gal-
lagher (1995) that combines multicompositional FT annealing
and helium diffusion models (Ketcham et al., 1999; Meesters
and Dunai, 2002). The best-fit solution (Fig. 12) highlights
three important stages: (1) a requirement for deposition and
burial to peak temperatures of ~80–90 °C between ca. 30 and
25 Ma, (2) uplift to the surface between ca. 25 and 20 Ma, and
(3) reburial in the Miocene (ca. 25–5 Ma) to temperatures of
50–60 °C (broadly equivalent to depths of ~1–1.5 km, assuming
geothermal gradients of 30 °C/km).
depositional age
Depositional
age range
0
20
Temperature °C
Detrital Argon Data
Four samples were analyzed from the Andaman Flysch,
Corbyn’s Cove section, which yielded 114 grain ages. One addi-
tional sample from Mithakhari Group sedimentary rock at Chiri-
yatapu was also analyzed (33 grain ages). The results are dis-
played as probability plots in Figure 13 and tabulated in Table A3
(Appendix). Ages for the Mithakhari Group sample extend from
70 Ma to the Archean, with the bulk of ages spanning the Meso-
zoic and Paleozoic. The Andaman Flysch ages are in general
younger, ranging from 30 Ma to the Proterozoic. Most of the
ages (71 grains) are <200 Ma, with 20 grains falling between
30 and 60 Ma, and 32 grains between 60 and 80 Ma. Given that
postdepositional heating was modest, as constrained by the FT
data, which measure lower closure temperatures than argon, the
young mica ages, some of which are younger than the FT data,
Oldest point to an Oligocene depositional age: i.e., the Andaman Flysch
of Archipelago was deposited at or after 30 Ma.
Group
Modeled results Detrital Zircon U-Pb Data
are consistent with

Archipelago Group
40 (Pal et al., 2005) Preliminary zircon U-Pb data from the Andaman Flysch
60 diagenetic data
Youngest that indicate (Fig. 14; Table A4, Appendix) show a large Proterozoic popula-
80 U-Pb and maximum burial tion, with a few grains showing Archean and Cretaceous–Eocene
argon temperatures of
100 source
range of max burial ages. The youngest zircons (five grains) give an average age of
45-55°C
120 ages 48 ± 5 Ma, which overlaps both zircon FT and mica ages. These
50 40 30 20 10 0 grains are euhedral and have concentric zoning typical of mag-
Age (Ma) matic zircon, suggesting a direct contribution from an early
20 Eocene igneous source rather than reworking of older sedimen-
Obs. Age : 36 Ma
Number of tracks

Pred. Age : 35 Ma tary deposits.

Obs. Mean length : 13.24μm
10 Pred. Mean length 13.66μm
Obs. S.D. : 1.61μm Petrography and Heavy Mineral Data
Pred. S.D. : 1.61μm
Oldest track 39 Ma
The very fine to fine-grained turbidites of the Andaman
1 2 3 4 5 6 7 8 9 10 1112 1314151617181920 Flysch have an intermediate quartz content (Q 49 ± 2%) (Fig. 15)
Track Length (microns) and contain subequal amounts of feldspar (F 22 ± 6%; plagio-
clase feldspar 39 ± 9%) and lithic grains (L 29 ± 6%) as seen
He/P
Helium diffusion model 50 in Table A5A (Appendix). The latter chiefly consist of very low
40
rank to medium rank and subordinately high rank metapelite-
He Obs : 13.0 metapsammite grains (Garzanti and Vezzoli, 2003), indicating
He Pred: 14.7 Ma 30 provenance from a collisional orogen (Dickinson, 1985). Volca-
Grain R: 75.0 microns
20 nic grains are mainly felsitic and are significant (Lv 8 ± 1%), sug-
gesting subordinate contributions from a volcanic arc. Transparent
10
heavy-mineral assemblages in the Andaman Flysch (Table A5B,
0 Appendix) have very low concentrations (0.3 ± 0.2%). Ultra-
50 40 30 20 10 0
Age (Ma)
stable species (zircon, tourmaline, rutile, chromian spinel) rep-
resent 21 ± 2% of the assemblage, and ferromagnesian minerals
Figure 12. Best-fit thermal-history model for apatite fission-track and (pyroxenes, amphiboles) are invariably absent, clearly showing
(U-Th)/He data from the Andaman Flysch. Key features are a require- the strong influence of diagenetic dissolution.
ment for maximum burial temperatures between ca. 20 and 30 Ma
and rapid cooling at ca. 20 Ma. Values given are for observed data
The very coarse grained sandstones of the Namunagarh Grit
(Obs.), model predicted values (Pred.), and standard deviation (S.D.). Formation chiefly consist of plagioclase and microlitic, lath-
For the helium plot, R is the grain radius and He/P is the helium con- work, and felsitic volcanic grains. The quartz content is low (Q
centration profile. 14 ± 10%), and only a few K-feldspar and granitic lithic grains
236 Allen et al.

Andaman Flysch-1A
(90 grains)
Mithakhari Group

Relative probability
Relative probability

Chiriyatapu
(33 grains)

0 500 1000 1500 2000 2500 3000

0 400 800 1200 1600 2000 2400 2800 Age (Ma)
Age (Ma)

Andaman Flysch
Relative probability

Corbyn’s Cove
Relative probability

(114 grains)

0 200 400 600 800 1000 1200 1400 1600 0 50 100 150 200 250 300 350 400
Age (Ma) Age (Ma)

Figure 14. Probability plots showing the distribution of concordant de-

trital zircon U-Pb ages analyzed in this study.

Younger grain ages

Relative probability

(71 grains)
occur, indicating provenance from a largely undissected volcanic
arc (Marsaglia and Ingersoll, 1992). A few terrigenous (shale-
sandstone) and very low rank to medium rank metapelite and
metapsammite grains also occur, suggesting minor reworking
of sedimentary and low-grade metasedimentary sequences. Sig-
nificant chert, along with traces of metabasite and serpentinite
grains, point to minor contributions from oceanic rocks.
In the Namunagarh Grit Formation, opaque grains represent
21 ± 16% of total heavy minerals. Transparent heavy-mineral
assemblages have poor concentrations, pointing to significant
20 60 100 140 180 220
diagenetic dissolution. These mainly include pyroxenes (mostly
Age (Ma) green augite; 35 ± 24%) and epidote (34 ± 19%), with subordinate
Figure 13. Probability plots showing the distribution of detrital argon sphene (11 ± 8%), green-brown hornblende (8 ± 8%), chromian
mica ages analyzed in this study. spinel (8 ± 8%), minor apatite (2 ± 1%), garnet, rutile, and other
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism
Q Lmf
A) Andaman Flysch
Namunagarh Grit
+ Bengal Fan
Irrawaddy 125
+ Irrawaddy 180
O
O
O
Irrawaddy 80
rec
ton
yc
led
cra
oro
ge
n O
magmatic arc
O
O O
F L Lv + Lmv
titanium oxides, consistent with provenance from a volcanic arc.
It must be kept in mind, however, that the original composition of
the Namunagarh Grit Formation sandstones (abundant volcanic
glass and ferromagnesian minerals) was markedly different with
respect to orogen-derived turbidites, and therefore that dissolu-
tion reactions of ferromagnesian minerals may have progressed
at different rates and in the presence of partially buffered or even
saturated interstitial waters.
Sm-Nd Whole Rock and Single Grain Analyses
Sandstone and mudstone samples from the Mithakhari
Group yielded whole rock εNd values of –7.2 and 3.1(+), respec-
tively. The Andaman Flysch samples gave whole rock values of
εNd –11.1 and –8.2 for the sandstone sample and mudstone sam-
ple, respectively. Such values are lower than those of Himalayan
sources, which typically range between εNd –12 and –26 (Galy
and France-Lanord., 2001) and are inconsistent with Indian
Shield sources (Peucat et al., 1989). Very little data exist for
the relevant Myanmar source regions. Analyses of modern river
sediment from the Irrawaddy River give εNd values between
−10.7 (Colin et al., 1999) and −8.3 (this paper: Table 2). The
Western Indo-Burman ranges (Fig. 16), which may have been
exhuming by the Oligocene (Mitchell, 1993), have more nega-
tive values (Colin et al., 2006; Singh and France-Lanord, 2002;
this paper, Table 2, and Table A6A, Appendix). Figure 16, which
compares the whole rock data from the Andaman Flysch with
possible source areas, shows that the most probable source area
lies within Myanmar.
To understand the source area in more detail, and in particu-
lar the significance of apatite from the Andaman Flysch, which
appears to come from more distant sources in comparison with
local arc and ophiolite sources in the Mithakhari Group sedi-
mentary rocks, Nd isotope composition was measured on single
grains of apatite from the Andaman Flysch. Apatite typically has
237
Figure 15. Petrography of Tertiary sand-
stones of South Andaman. Whereas
composition of the Namunagarh Grit
B) Formation sandstones points to prov-
enance from an undissected-transitional
magmatic arc, a recycled orogenic prov-
enance is clearly indicated for the An-
+ daman Flysch (Dickinson, 1985). These
very fine to fine-grained turbidites com-
pare closely with modern sands of ho-
mologous grain size from the Irrawaddy
O
Delta. Q—quartz; F—feldspars; L—
O lithic grains; Lv + Lmv—volcanic and
low-rank metavolcanic; Ls + Lms—sed-
O imentary and low-rank metasedimenta-
Lms + Ls ry; Lmf—high-rank metamorphic.
Sm/Nd ratios of 0.2–0.5, therefore for non-age–corrected apatite
the 143Nd/144Nd ratio tends to reflect the Nd isotopic composition
of their parent whole rock, enabling a direct linkage between ther-
mochronometric data and source in terms of Nd isotopic signa-
ture. This approach is particularly effective if the source rocks are
young and/or their potential range of Sm-Nd has been quantified.
It is also important to note that this approach yields information
that is not the same as sediment whole rock values that record an
average value (weighted by Nd concentration) of source rocks.
It is also possible that the detrital apatite may be dominated by
a single apatite-rich source. The majority (~60%) of the single
grain analysis of apatite from the Andaman Flysch produced εNd
values of –5 to +5, typical of juvenile volcanic whole rock val-
ues and contrasting significantly with the bulk sediment values
of εNd –11.1 and –8.2 (Fig. 17; Table A6A, B, Appendix). The
restricted range of εNd values for these apatites, which gave par-
tially reset FT ages between 30 and 40 Ma (Table A1, Appen-
dix), imply a single source region for the apatite grains. Little
is known about the apatite Sm-Nd systematics for the source
rocks of the Andaman Flysch. However, as Figure 17 shows, the
Andaman Flysch apatite Nd isotope data contrast with apatite Nd
data from a Holocene sand dominated by Himalayan sources col-
lected from the Bengal Basin near Joypur, West Bengal. This plot
clearly shows that there is little evidence for material eroded from
Himalayan sources in this sample.
DISCUSSION
Constraints on Sedimentation and Uplift
New thermochronometric evidence and field observations
from type locations on South Andaman, combined with existing
biostratigraphy and petrography, provide an improved chronol-
ogy for deposition and uplift. The accretionary setting means
that sedimentation history will be intimately tied to subduction
 TABLE 2. TYPICAL SIGNATURES OF POTENTIAL SOURCE REGIONS
40
Source region Rock description, heavy minerals, and petrography Sm–Nd U-Pb ages of zircons Ar-39Ar ages of white Zircon fission-track
Whole rock mica ages
εNd(0)
Himalaya (Southern Flank)
Drained by Ganges and tributaries In the Oligocene, less metamorphosed “Higher Signature today: Signature today: Signature today: Signature today:
Oligocene bedrock characteristics Himalayan protolith” would likely have been Higher Himalayan As determined from As determined from No FT data available
interpolated from Eocene and Miocene exposed. Rocks were low-grade metamorphic, as determined Ganges sediments: Ganges tributaries: for Ganges
foreland basin rocks sub-garnet grade, nonmicaceous (as determined from Ganges: av. mostly >400 Ma– Neogene peak; sediment; He data
Bedrock signal today taken from modern from Miocene foreland basin rocks).1 –17.52 Precambrian. Grains subordinate grains <55 Ma, Pliocene–
river sediments; higher Himalaya <30 Ma rare. Grains 30– spanning to Pleistocene peak9
dominates detritus 400 Ma very rare5 Precambrian7
Oligocene bedrock: Oliogocene bedrock:
Eocene: av. –83 Oligocene bedrock: Oligocene bedrock: Eocene: peaks at 45
Miocene: Eocene and Miocene: Micas absent in Eocene, Ma, 119 Ma, and
–14 to –174 Tertiary grains absent; very rare in late 343 Ma;10
65–400 Ma grains very Oligocene–Miocene8 Miocene: peaks at 30
rare6 Ma, 60–75 Ma,
117 Ma, 300–370
Ma11

Burma
Region drained by Irrawaddy Cretaceous arc rocks and Triassic forearc-backarc Irrawaddy: Irrawaddy: Irrawaddy: Irrawaddy:
Data from modern Irrawaddy River sediments on metamorphic basement. Mogok –10.7, –8.313 Proterozoic, Ordovician, Cretaceous to Miocene Neogene,
sediment. Shan-Thai block lies to east, schists, gneisses, and intrusives, Shan-Thai Jurassic–Cretaceous, grains; peak 30–55 Paleogene,
forearc-backarc of Indo-Burman Proterozoic–Cretaceous sedimentary rocks on Paleogene, and Ma15 Cretaceous
Ranges (IBR) to west. schist basement. Irrawaddy River sediment plots Neogene grains14 grains15
Paleocontinental margin. in “recycled orogenic” province field of QFL plot
(Dickinson, 1985).12

Western Indo-Burman Ranges Dark-gray, very fine grained volcanic arenite of 3 samples give Precambrian: Archean and No mica present in pre- Cretaceous and
Accretionary prism; may have been Eocene (Oligocene?) age. Only limited mica and values of –4.0; 1 Proterozoic; Cretaceous Miocene sedimentary Paleocene to
exhuming and thus a sediment source heavy minerals. Many shale and siltstone lithics.16 sample (farther and Paleogene (55–65 rocks16 Eocene grains16
in late Eocene–Oligocene. Data north) gives a Ma) grains16
determined from Arakan modern river value of –7.416
sands (R. Allen et al., unpublished
data) unless otherwise stated.

Indian Shield
Dominantly Arc hean craton. Subordinate Predominantly gneisses and granites; opaques, Values from south Dominantly Archean19 No data available No data available
Proterozoic mobile belts and orthoproxene, and sillimanite found in rivers and east craton
Gondwanan sedimentary cover. draining east craton.17 only;
–30 to –3618
1
Najman and Garzanti (2000).
2
Galy and France-Lanord (2001).
3,4
Robinson et al. (2001); Najman et al. (2000).
5
Campbell et al. (2005).
6
DeCelles et al. (2004).
7
Brewer et al. (2003); Najman and Pringle (unpublished data).
8
DeCelles et al. (1998); Najman and Garzanti (2000).
9
Campbell et al. (2005).
10
Najman et al. (2005).
11
Najman et al. (2004, 2005).
12
Mitchell (1993); Pivnik et al. (1998); Bertrand et al. (1999); R. Allen et al. (unpublished data).
13
Colin et al. (1999); R. Allen et al. (unpublished data).
14
Bodet and Schärer (2000).
15
R. Allen et al. (unpublished data).
16
R. Allen et al. (unpublished data).
17
Mallik (1976).
18
Peucat et al. (1989); Saha et al. (2004).
19
Mishra and Rajamani (1999); Auge et al. (2003).
 Andaman Flysch
Namunagarh Grit
IndoBurman Ranges
Modern Irrawaddy River
-30 to -36
Indian Shield
Himalayan Sources

present-day
Brahmaputra /
Ganges TSS
HHS
LHS

-28 -24 -20 -16 -12 -8 -4 0 4 8

Epsilon Nd
Figure 16. Plot comparing Sm-Nd whole rock values from the Namunagarh Grit Formation and Andaman Flysch For-
mation with data from possible source regions (Colin et al., 2006; DeCelles et al., 2004; Peucat et al., 1989; Singh and
France-Lanord, 2002). The data do not support the Himalayan region as the main source. TSS—Tethys Sedimentary
Sequence; HSS—Higher Himalayan Sequence; LHS—Less Himalayan Sequence.

0.5134

0.5132 Holocene, Bengal delta

Andaman Flysch
0.5130 +8

0.5128 +4

0
143 Nd / 144 Nd

0.5126

Epsilon Nd
0.5124 Field of typical Himalayan -4
apatites
0.5122 -8

0.5120 -12

0.5118 -16

0.5116 -20

0.5114 -24

0.5112
0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4
147Sm / 144Nd
Figure 17. Single grain Nd measured on apatite from the Andaman Flysch Formation compared with a Holocene sand
from the Bengal Basin known to have its source in the Himalayas. The clear difference between these two samples sup-
ports the Sm-Nd whole rock data plotted in Figure 16.
240 Allen et al.

behavior, offscraping, accretion, and uplift. While in such a set-
ting uplift may be viewed as a more or less continuous process,
we recognize some distinct uplift events that have affected the
South Andaman geology.
The earliest conglomerates and grits of the Mithakhari
Group record subaerial exposure and erosion of the arc sequence
and shallow marine limestone cover, as described above, con-
straining the age of ophiolite as pre–Mithakhari Formation (early
to middle Eocene) and post–shallow marine sedimentation. The
petrography of the Mithakhari Group is indicative of an undis-
sected volcanic arc with minor reworking of sedimentary and
low-grade metasedimentary sequences and minor contributions
from oceanic igneous rocks.
Our data provide maximum and minimum age constraints
for deposition of the Andaman Flysch. Detrital Ar-Ar mica and
FT data constrain maximum depositional age to ca. 30 Ma. The
minimum depositional age is constrained by the thermal history
modeling, which indicates that burial to maximum temperatures
of ~90 ± 5 °C was reached ca. 25–20 Ma, ending with a phase
of rapid cooling at ca. 20 Ma. This implies a discrete episode
of major uplift at ca. 20 Ma consistent with deposition of the
Archipelago Group shallow marine volcanic-rich sedimentary
rock from ca. 18 Ma onward (Pal et al., 2005; Singh et al., 2000),
which was deposited on top of the Andaman Flysch. Uplift at
ca. 20 Ma is also regionally significant, coinciding with major
stratigraphic changes in the Irrawaddy Delta, Mergui Basin
(Fig. 1), in sediments accreted onto the Andaman-Nicobar Ridge
(Curray, 2005), and formation of an unconformity in the Indo-
Burman Ranges (Acharyya et al., 1989). Uplift of the Andaman
Ridge at 20 Ma is also coincident with the uplift of the Owen and
Murray Ridges in the Arabian Sea (Mountain and Prell, 1990),
suggestive of a wider plate tectonic trigger.
An outstanding question is, what was the original maximum
thickness of the Andaman Flysch? Hydrocarbon-exploration

180-
convergence (mm/yr)
Rate of Indian

150-

100-
Lee and Lawver, 1995

Guillot et al., 2003

convergence
(from N)
Angle of

50-
40-
30-
20-
10-

deposi-
uplift

Archipelago Group Figure 18. A comparison between In-

tion
dian convergence history (after Lee and
unconformity Lawver, 1995, and Guillot et al., 2003)
and the uplift and sedimentation history
deposi-
uplift

Andaman Flysch of the rocks studied from South Anda-

tion man Island.
contact unclear
Namunagarh
Mithakhari

continuous
Group

Grit
Hopetown deposition
Conglom. ?????
unconformity

Ophiolite pelagic uplift deposition

cover sediments

? ? ?
Ophiolite obduction
? ? ?
0 20 40 60 80 100
Age (Ma)
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism 241

wells drilled to the east and the west of Middle Andaman indicate TARIM QD AL
thicknesses of <1000 m (Roy, 1992), whereas outcrop estimates KL NORTH CHINA
range from ~750 to 3000 m (Pal et al., 2003). Results from ther- QT QS SPG
mal history modeling indicate burial of the sampled Andaman
Flysch beds beneath at least 2.5 km of cover (assuming geother- LHASA
mal gradients in the range of 35–30 °C). In a growing accretion- SOUTH CHINA
ary wedge, burial is as likely to be due to thrust stacking as it
is to sedimentation. Crucially ~2.5 km of burial is incompatible India WB
with the unannealed apatite data from the stratigraphically lower
Mithakhari Group, highlighting the fact that these different units
must have been deposited in different locations (sub-basins?)

SIBUMA
within the accretionary wedge. Thermal history modeling also

INDOCH
shows a phase of accelerated cooling from ca. 10 to 5 Ma. This
is probably tied to a phase of regional Miocene–Pliocene uplift
linked to spreading in the Andaman Sea driven by increases

SU
in the subduction rate and dip of the subducted slab (Khan

and Chakraborty, 2005). This last phase of uplift, the result of
squeezing between subduction in the west and extension in the
east, is likely responsible for the present topography of the Anda-
man Islands.
The relationship of these episodes of uplift and erosion to
regional events can be appreciated when these new constraints
are compared against reconstructions of Indian plate convergence
history. Figure 18 plots the India-Asia convergence rate and con-
vergence angle from the studies of Lee and Lawver (1995) and
more recently from that of Guillot et al. (2003). The graph shows
that the studied outcrops of Andaman Flysch were deposited at a
time of more northerly convergence and ended when subduction
shifted to a less oblique angle. Seismic lines taken from across
the Andaman Arc show how the angle of subduction can influ-
ence wedge development where the intensity of deformation,
which increases from north to south, changes with obliquity of
subduction (Curray, 2005). This relationship may explain the
evidence for a discrete episode of uplift, involving the Corbyn’s
Cove Andaman Flysch, at ca. 20 Ma.
Sediment Provenance
The Mithakhari Group shows clear evidence of a dominant
contribution from an arc, in keeping with its interpreted forearc
basin depositional environment and in line with previous prov-
enance work on the basis of petrography (Bandopadhyay, 2005).
The rocks plot in the magmatic arc field on the QFL petrographic
plot (Fig. 15); volcanic fragments and glass shards are common,
and mineral grains, e.g., apatite, are euhedral. A positive whole
rock εNd signal (+3) is indicative of derivation from a mafic-
juvenile source. The source was likely that of the Cretaceous–
Eocene arc that stretched from the Himalaya collision zone
through Myanmar to Sumatra (Mitchell, 1993), as reflected in
the FT ages of the euhedral apatites. A subordinate continental
source is confirmed by petrography and Precambrian–Paleozoic
Ar-Ar mineral grain ages. Such a source was probably the con-
tinental margin of the Shan-Thai Block (Sibumasu) extending
INA
QD = Quidam terrane
QS = Qamdo-Simao terrane SWB
QT = Qiantang terrane
KL = Kunlun terrane
AL = Alao Shan terrane
SPG = Song Pan Ganzi terrane
WB = West Burma
SWB = Southwest Burma
Figure 19. Map showing the principal terranes that have amalgamated
to form SE Asia (after Metcalfe, 1996).
down to peninsular Thailand, adjacent to the forearc (Metcalfe,
1996) (Fig. 19).
The data show a clear change in petrographic and isotopic
characteristics between the Mithakhari Group and the overlying
Andaman Flysch. Andaman Flysch composition represents the
first sedimentary material to plot in the recycled orogenic prov-
ince (Fig. 15). Initial results from the pilot study of Sm-Nd whole
rock values differ markedly between the Mithakhari Group and
the Andaman Flysch, particularly when the finer grained facies
are compared. The positive εNd signal of the Mithakhari Group
contrasts strongly with the more negative εNd signal (−11) of the
Andaman Flysch mudstone, suggestive of a contribution from
older continental crustal sources (Fig. 15). Detrital micas are of
radically different ages, with the age peaks for the Mithakhari
Group falling within the Precambrian–Paleozoic spectrum,
whereas those from the Andaman Flysch are typically of late
Mesozoic and Tertiary age (Fig. 13). A provenance contrast is
not apparent from the zircon FT data (Figs. 9 and 10), as the
detrital modes are similar. The apatite FT data cannot be directly
compared, as the Andaman Flysch apatites have seen some post-
depositional partial resetting. However, comparison of apatite
242 Allen et al.

chemistry (chlorine-uranium content, Fig. 11) does highlight a
major difference in the apatite source.
To some extent these differences between the Mithakhari
Group and the Andaman Flysch may reflect different mixtures
of the same sources. The continental-derived material that domi-
nates the Andaman Flysch represents only a minor contribution
in the arc-dominated Mithakhari Group. Variations may have as
much to do with depositional setting than source variation. In
this regard, sedimentary rocks of the Namunagarh Grit Forma-
tion may represent a perched forearc basin, whereas the Anda-
man Flysch represents deep-water turbidites.
It was proposed by Karunakaran et al. (1968) and Pal et
al. (2003) that these “recycled orogenic” Andaman Flysch rocks
were sourced by the Irrawaddy River of Myanmar and deposited
in a forearc basin. In contrast, Curray et al. (1979) considered
the Andaman Flysch to be trench sediments offscraped from
the Himalayan-derived Bengal Fan on the downgoing Indian
slab. Significant contribution from the northward-drifting cra-
tonic Greater India can be ruled out by the dissimilar petrog-
raphy, Sm-Nd signature, and mineral cooling ages (Fig. 16;
Table 2) as well as by the apatite Nd data (Fig. 17). No apa-
tites were measured with significantly negative εNd values (e.g.,
−30) diagnostic of Indian Shield sources. The mica argon ages
older than 50 Ma, present in both the Mithakhari Group and the
Andaman Flysch, are consistent with known sources along the
western margin of the Shan-Thai Block (Sibumasu) adjacent to
the forearc. The presence in the Andaman Flysch of a small but
recognizable population of mica ages between 30 and 40 Ma
is not consistent with these Sibumasu sources, as the youngest
granites on the Thai peninsula are ca. 50 Ma (Charusiri et al.,
1993). Possible sources for the Tertiary mica ages might instead
include a region affected by India-Asia collision at ca. 50 Ma,
possible candidates being either the nascent Himalaya to the
north, Transhimalaya, or a more northeastern source—namely,
Myanmar locations of the Burman-Thai Block—where Tertiary
metamorphism, magmatism, and grain isotopic characteristics
and ages are ascribed to the effects of the India-Asia collision
(Bertrand et al., 1999; 2001, Bodet and Schärer, 2000; Barley et
al., 2003). Isotopic characteristics of these Myanmar rocks are
incompletely documented at present. Those that are available are
taken from bedrock, and analyses of sands from the Irrawaddy
River, which drains the areas under discussion—i.e., the Central
Myanmar Basin, to its east the Mogok Belt and Shan Plateau, and
to its west the Indo-Burman Ranges (Bodet and Schärer, 2000;
Colin et al., 2006; Singh and France-Lanord, 2002; Table 2, this
study)—as verified by 30 Ma Ar-Ar ages from the Irrawaddy

Paleogene
nas
cen
Eurasia
fore tH Tsangpo River, flows
land imalay
bas a along Himalayan
ins
suture draining east to
South China Sea
rivers draining
southern flanks continental Figure 20. Cartoon illustrating the Paleo-
drainage gene paleogeography and paleodrainage
of SE Asia. Although this sketch simpli-
volc

fies the location of the subduction zone,

Sh

Greater India Early the Asian margin and Andaman crust

anic

an

Bengal would have changed throughout the Pa-

-T

Fan
arc

leogene as Greater India moved north-

ha

ward. The cartoon does serve to illustrate

i-

Paleogene the general location of continental sourc-

Su

location of es and drainages to the Andaman region.

Andaman
nd

crust
ala
nd
vo und
lca a T
S

nic re

Indian Ocean
ar nch
c
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism
region (Table 2). Discrimination between these two possible
sources, the Himalayan or Myanmar region, requires an under-
standing of the different paleogeography and paleo-drainage of
the region in the Oligocene in comparison with the present day,
as discussed below (Fig. 20).
Prior to the India-Asia collision, the northward extension of
the Sunda Arc as far as Pakistan permits correlation between the
southern margin of Asia in the Himalayan region with equiva-
lent rocks in Myanmar (Mitchell, 1993) and farther south. Sub-
sequent to continental collision at ca. 50 Ma (Rowley, 1996), the
Himalayan thrust belt started to develop in the north while the
eastern (Myanmar) region remained an active continental margin
with, for example, the Central Basin along which the Irrawaddy
now drains, the site of the continental margin between the sub-
duction zone to the west and the arc to the east (Pivnik et al.,
1998). Movement along the Sagaing dextral strike-slip fault (ini-
tiation in pre-Miocene times: Mitchell, 1993) and the opening of
the Andaman Sea in the late Miocene (Curray, 2005; Khan and
Chakraborty, 2005) were responsible for changes in the coastal
paleogeography of this region.
In the Paleogene the Yarlung-Tsangpo River (draining
the Himalayan Arc and suture zone) probably drained into the
South China Sea prior to its capture by the Irrawaddy River, and
finally the Brahmaputra River in the Neogene (Clark et al., 2004,
2005). Thus, Ganges-type rivers—those draining only the south-
ern slopes of the Himalayan mountain belt, with no arc com-
ponent—would have contributed the Himalayan signal to the
Indian Ocean (Fig. 17). The southern slopes of the Himalayas
consist of Indian crust. Characteristics of these rocks changed
with time as Neogene Himalayan metamorphism subsequently
increased the metamorphic grade and overprinted the metamor-
phic cooling ages of minerals and rocks exposed at the surface
today. In contrast, the northeastern Myanmar region was devoid
of Indian crustal rocks, and no thrust belt was present to bar the
Asian and arc sources from draining south. Prior to the open-
ing of the Andaman Sea, rivers draining the Myanmar region
presumably would have drained into the trench-forearc system,
which acted as a “sink,” preventing deposition farther west onto
the Indian plate.
The Andaman Flysch records mixed orogenic and arc prov-
enance. The young (<100 Ma) U-Pb–dated detrital zircons from
the Andaman Flysch are consistent with derivation from igneous
sources, most likely an eastern (Myanmar) provenance based on
similar aged grains found in modern Irrawaddy River sediment
(Bodet and Schärer, 2000), as grains with young U-Pb zircon
ages are extremely rare in the southern flanks of the Himalayas
(Campbell et al., 2005; DeCelles et al., 1998, 2004).
Myanmar sources also can adequately account for the
majority of the “recycled orogenic” component of the detritus
in the Andaman Flysch. Both petrography and εNd whole rock
signatures, especially of the fine-grained material, are nearly
identical to those of the modern Irrawaddy River (Figs. 15 and
16). 40Ar-39Ar and FT data from the Irrawaddy River also are
243
similar to data obtained from the Andaman Flysch, showing a
mica age peak of ca. 30–60 Ma with a population of older Creta-
ceous grains and zircon FT ages of Miocene and Paleogene (20–
65 Ma) to Cretaceous (Table 2). Nevertheless, some disparities
are evident. The lack of zircon grains with Jurassic U-Pb ages
for the Andaman Flysch is surprising, considering their preva-
lence in the Mogok Belt (Barley et al., 2003) and their presence
in modern Irrawaddy River sand (Bodet and Schärer, 2000).
Recourse to a Himalayan contribution through the Bengal Fan
is not required to explain the recycled orogenic component of
the Andaman Flysch. However, a minor contribution from the
Himalayan thrust belt cannot be ruled out: for example, grains
aged 100–1500 Ma in the Andaman Flysch are common to both
Himalayan and Myanmar rock types. In addition, the intermedi-
ate whole rock εNd values may represent a mixture between a
more negative Himalayan source and more positive Myanmar
sources, as implied by the single grain apatite analyses. The pos-
sibility of dual provenance, with Himalayan-derived Bengal Fan
material and Myanmar-derived material meeting and mixing in
the trench, is a model that would be consistent with the regional
seismic data, which show folding and uplift of Bengal Fan sedi-
ments at the base of the slope (Curray, 2005). Definitive discrim-
ination between Myanmar and Himalayan sources, if possible
at all, awaits more information on the source rock geology of
Myanmar, in particular that of the Shan Plateau and Mogok Belt,
and analysis of as-yet-undrilled Oligocene sediments preserved
in the Bengal Fan, and coeval sedimentary rocks of the Central
Myanmar Basin.
CONCLUSIONS
The Mithakhari Group was deposited in the late Paleocene–
Eocene, with a maximum age of ca. 60 Ma. This current study
shows that the sedimentary rocks are predominantly arc derived
from a proximal source in keeping with the interpreted forearc
depositional environment. A subordinate contribution from an
older continental source was most likely to have been the western
Sibumasu margin, but a Transhimalayan arc unit source cannot
be ruled out.
The Oligocene Andaman Flysch was deposited between 30
and 20 Ma, and then uplifted by 20 Ma. It shows a clear change
in petrographic and isotopic characteristics from the Mithakhari
Group and is composed mainly of “recycled orogenic” sources
with a subordinate arc provenance. It represents the earliest
record in the region of major influx from a continental area. The
recycled orogenic component is most simply explained by ero-
sion from the northeastern (Myanmar) continental region. Iso-
topic and petrographic differences between formations may be
explained to some extent by different mixtures and contributions
from the same source regions. Although dual provenance is a
favorable model for this region, detailed discrimination between
Himalayan and Myanmar sources, at present, awaits more data
from Myanmar source rocks.
 TABLE A1. FISSION-TRACK ANALYTICAL DATA FOR ANDAMAN ISLANDS–SOUTH ANDAMAN
Sample Latitude Mineral No. of ρd ρs ρι Central age Age components Mean S.D. No. of
Dispersion
Longitude grains Nd NS Ni ±1σ (Ma) track tracks
length
(Pχ2) RE% (μm)

MITHAKHARI GROUP
Hopetown Conglomerate Formation
Hopetown N11°41.407′ Zircon 25 0.528 12.16 4.32 0.0 56.2 91.8 ± 10.8 61 ± 2 (16) 129 ± 10 (5) 280 ± 32 (4)
Quarry E92°43.501′ (3450) (4942) (1756)
Apatite 19 1.106 0.064 0.209 0.03 69.1 56.5 ± 8.5 57 ± 9 (19)
(6131) (58) (191)
Namunagarh Grit Formation
Mungleton N11°34.223′ Zircon 18 0.535 10.972 4.872 0.0 42.3 75.15 ± 7.9 59 ± 2 (15) 226 ± 21(3)
Quarry E92°39.541′ (3450) (4182) (1857)
Apatite 21 1.106 0.150 0.246 0.0 64.5 68.9 55 ± 8 (19) 151 ± 68 (1) 314 ± 51 (1)
(6131) (322) (946)
Namunagarh N11°41.339′ Apatite 24 1.106 0.060 0.282 99.6 0.0 39.8 ± 4.2 40 ± 4 (24)
Quarry E92°41.106′ (6131) (108) (506)
Chiriyatapu N11°42.005′ Zircon 38 0.525 12.53 2.255 0.0 92.6 143.0 ± 22 40 ± 3 (9) 91 ± 7 (8) 329 ± 17 (22)
E93°32.301′ (3450) (7468) (1344)

ANDAMAN FLYSCH
AND-1A N11°38.430′ Apatite 38 1.106 0.264 1.653 0.02 28.8 30.85 ± 2.2 31 ± 2 (38) 13.25 ± 0.17 1.51 76
E92°45.292′ (6131) (497) (3116)
Zircon 43 0.516 8.341 4.07 0.0 44.5 62.15 ± 4.6 44 ± 2 (24) 67 ± 4 (12) 147 ± 10 (6)
(3450) (6439) (3147)
AND-1B N11°38.457′ Apatite 57 1.106 0.499 2.499 7.3 16.3 37.15 ± 1.8 37 ± 2 (57) 13.06 ± 0.20 1.75 75
E92°45.318′ (6131) (738) (3696)
Zircon 32 0.515 6.115 3.685 0.0 41.5 53.75 ± 4.4 41 ± 2 (26) 101 ± 14 (6)
(3450) (3085) (1859)
AND-3 N11°38.459′ Apatite 50 1.106 0.497 2.556 0.03 22.6 35.85 ± 1.9 36 ± 2 (50) 13.24 ± 0.22 1.61 54
E92°45.368′ (6131) (852) (4739)
Zircon 15 0.108 10.24 4.828 0.0 73.0 58.55 ± 11.3 36 ± 2 (12) 616 ± 102 (3)
(3450) (1930) (910)
AND-4 N11°39.461′ Apatite 47 11.106 0.427 2.652 0.04 20.2 30.75 ± 1.5 31 ± 2 (47) 13.42 ± 0.23 1.80 62
E92°45.417′ (6131) (923) (5733)
Zircon 42 0.505 10.63 4.506 0.0 70.1 67.95 ± 7.6 38 ± 1 (20) 58 ± 4 (10) 270 ± 16 (12)
(3450) (7575) (3210)
6 –2
Note: (i) Track densities are (x10 tr cm ); numbers of tracks counted (N) shown in brackets. (ii) Analyses by external detector method using 0.5 for the 4π/2π geometric correction factor. (iii) Ages calculated
using dosimeter glass CN-5; (apatite) ξCN5 = 338 ± 4; CN-2 (zircon) ξCN2 = 127 ± 4 calibrated by multiple analyses of IUGS apatite and zircon age standards (see Hurford, 1990). (iv) Pχ2 is probability for obtaining χ2
value for v degrees of freedom, where v = no. crystals –1. (v) Central age is a modal age, weighted for different precisions of individual crystals (see Galbraith and Green, 1990). (vi) Age modes deconvolved using
approach of Sambridge and Compston (1994) and Galbraith and Laslett (1993). RE—% relative error about the central age; S.D.—standard deviation; bold type—apatite; bold italic type—zircon ages.
 TABLE A2. RAW APATITE-HELIUM DATA
4 4 4 238 232
Sample Location He HB He-HB %S.D. in He Absolute ± U ± Th ± He age* Error Grain FT FT
(ncc) (ncc) (ncc) Q± (atoms) (ng) (%) (ng) (%) (Ma) ± 7% radius corrected corrected
age
(Ma)
AND-A N11°38.430′ E92°45.292′ 0.356 0.023 0.333 0.452 8.95E + 09 4.04E + 07 0.108 1.2 0.537 1.33 11.704 0.819 77 0.81 14.4
AND-B N11°38.457′ E92°45.318′ 0.209 0.021 0.188 0.119 5.05E + 09 6.01E + 06 0.062 2.18 0.215 1.8 13.756 0.962 66 0.76 18.1
AND-C N11°38.459′ E92°45.368′ 0.222 0.018 0.204 0.119 5.48E + 09 6.52E + 06 0.058 1.98 0.315 1.4 12.664 0.886 82 0.81 15.6
Note: All samples are taken from the Andaman Flysch Formation at Corbyn’s Cove. Helium ages are based on replicate analyses of apatite grains. The total uncertainty in sample age
is based on the reproducibility of 39 analyses at the California Institute of Technology laboratory. Durango apatite standard (this error of the mean = 6.7%) combined with the U/Th and He
analytical uncertainties. HB—hot blank; ncc—nanno cc; S.D.—standard deviation; FT—fission track.
*Uncorrected He age.
 TABLE A3. RAW ARGON DATA
Andaman Flysch (Corbyn’s Cove section)
40 39 38 37 36 40 39
A1B Ar Ar Ar Ar Ar Ar*/ Ar ± Age (Ma) ±
Ms 1 0.68068 0.213559 0.003536 0.005284 0.000779 2.109962 0.020067 51.8 0.5
Ms 2 0.203759 0.060257 0.000705 0.000626 6.48E-05 3.063559 0.006231 74.7 0.4
Ms 3 0.451179 0.185698 0.002571 0.002597 0.000354 1.865821 0.022963 45.9 0.6
Ms 4 0.329659 0.110714 0.001661 0.002407 0.000334 2.085147 0.038311 51.2 1.0
Ms 5 0.37799 0.046614 0.000787 0.000994 0.00013 7.286482 0.032957 172.9 1.1
Ms 6 0.269821 0.037161 0.000531 0.000573 5.98E-05 6.784923 0.089994 161.5 2.2
Ms 8 2.277688 0.11746 0.001906 0.002831 0.000274 18.7013 0.031643 414.2 2.0
Ms 9 2.143176 0.070945 0.00093 0.000402 2.99E-05 30.08465 0.047492 626.5 2.8
Ms 10 1.384602 0.074126 0.001043 0.001053 0.00018 17.96253 0.023198 399.5 1.9
Ms 11 0.253605 0.015673 0.000291 5.75E-05 0 16.18154 0.035317 363.6 1.8
Ms 12 0.259847 0.082319 0.00117 0.000824 7.98E-05 2.870187 0.003993 70.1 0.4
Ms 13 3.391288 0.385463 0.004783 0.003164 0.000389 8.499624 0.011701 200.1 1.0
Ms 15 0.790418 0.252174 0.003255 0.003477 0.000284 2.801527 0.02637 68.4 0.7
Ms 16 0.531821 0.070046 0.000772 0.000291 5.49E-05 7.360776 0.036057 174.5 1.2
Ms 17 0.323132 0.105295 0.001329 0.001905 0.000189 2.537016 0.031682 62.1 0.8
Ms 18 0.832588 0.098415 0.001318 0.000894 4.98E-05 8.310543 0.035372 195.9 1.2
Ms 19 0.176399 0.061786 0.00071 0.000428 7.99E-05 2.472932 0.067916 60.5 1.7
Ms 20 0.458665 0.086807 0.001068 0.00179 0.00013 4.842794 0.034779 116.7 1.0
Ms 21 0.269899 0.070387 0.000792 0.000117 0.0002 2.994993 0.047513 73.1 1.2
Ms 22 1.721992 0.356929 0.004472 0.009619 0.000637 4.296717 0.010036 103.9 0.6
Ms 23 1.216759 0.459022 0.005575 0.00187 0.00031 2.451514 0.006712 60.0 0.3
Ms 24 0.085958 0.027352 0.000404 0.000195 0 3.142686 0.011197 76.6 0.5
Ms 25 0.130443 0.038038 0.000618 0.002262 0.000284 1.219882 0.023433 30.1 0.6
Ms 26 2.371162 0.106391 0.001354 0.001404 0.000115 21.96896 0.014042 477.7 2.1
Ms 27 0.314057 0.064276 0.000828 0.000507 4.99E-05 4.656836 0.012754 112.4 0.6
Ms 28 0.336883 0.064142 0.000767 0.000156 8E-05 4.883813 0.04699 117.7 1.2
Ms 29 0.999109 0.106174 0.001329 0.00041 0.000125 9.062499 0.029797 212.6 1.2
Ms 30 0.124323 0.075656 0.000961 0.000488 7.49E-05 1.350842 0.039184 33.3 1.0
Ms 31 1.604755 0.163177 0.001993 0.000528 0.000145 9.572115 0.020485 223.8 1.1
Ms 32 0.176028 0.107915 0.001318 0.000411 8.49E-05 1.398714 0.006655 34.5 0.2
Ms 33 0.094113 0.056227 0.000787 0.000782 5.48E-05 1.385834 0.005724 34.2 0.2
40 39 38 37 36 40 39
A3 Ar Ar Ar Ar Ar Ar*/ Ar ± Age (Ma) ±
Ms 1 0.274897 0.083084 0.000981 0.00039 0.000145 2.793324 0.037215 68.2 1.0
Ms 2 0.444523 0.143237 0.00185 0.001424 0.000445 2.186142 0.029916 53.6 0.8
Ms 3 0.297818 0.112368 0.0014 0.000746 0.000115 2.348489 0.038715 57.5 1.0
Ms 4 1.30819 0.466538 0.005703 0.002087 0.000109 2.734713 0.013397 66.8 0.5
Ms 5 0.346651 0.128071 0.001671 0.000476 0.000155 2.349362 0.023267 57.6 0.6
Ms 6 3.632999 0.500069 0.006087 0.001156 0.000305 7.084947 0.012826 168.3 0.9
Ms 8 0.311219 0.100244 0.001252 0.000476 0.000185 2.55964 0.004776 62.6 0.3
Ms 9 0.281996 0.054435 0.00068 0.000511 0.000105 4.611138 0.010622 111.3 0.6
Ms 10 0.180333 0.044863 0.000593 0.000392 2.49E-05 3.855628 0.00766 93.5 0.5
Ms 11 0.518299 0.095269 0.001186 0.000528 0.000125 5.053064 0.006582 121.6 0.6
Ms 12 0.604087 0.186293 0.002397 0.000648 0.000155 2.997079 0.016331 73.1 0.5
Ms 13 0.514859 0.086648 0.001109 0.000392 2.49E-05 5.857044 0.034464 140.2 1.0
Ms 15 0.190225 0.061068 0.000797 0.000672 0.000145 2.414197 0.048452 59.1 1.2
Ms 16 0.138431 0.029423 0.000409 0.000534 0.000115 3.551304 0.003826 86.3 0.4
Ms 17 0.560451 0.085584 0.001022 6.89E-05 6.5E-05 6.324168 0.019494 151.0 0.9
Ms 18 1.927513 0.246058 0.003455 0.002792 0.000814 6.855692 0.06817 163.1 1.7
Ms 19 0.549605 0.188385 0.002402 0.00138 0.000265 2.502356 0.018307 61.2 0.5
Ms 20 0.199519 0.062808 0.000767 0.000897 0.000185 2.307352 0.05385 56.5 1.3
Ms 21 0.792813 0.159513 0.00209 0.002278 0.000214 4.573025 0.019797 110.4 0.7
Ms 22 0.531614 0.155877 0.00209 0.001657 0.000385 2.681443 0.009091 65.6 0.4
Ms 23 0.290937 0.092139 0.001272 0.000881 7.98E-05 2.901768 0.033228 70.8 0.9
Ms 24 0.481724 0.087635 0.000905 0.00038 9.99E-05 5.160096 0.036978 124.1 1.0
Ms 25 1.163152 0.352631 0.004461 0.001418 0.00038 2.980373 0.009667 72.7 0.4
Ms 26 0.272653 0.094083 0.001191 0.005552 8.85E-05 2.619921 0.032203 64.1 0.8
Ms 27 0.422343 0.128727 0.00161 0.000571 0.00011 3.028746 0.023467 73.9 0.7
Ms 28 0.24212 0.086183 0.001145 0.000242 9.94E-06 2.775297 0.03496 67.8 0.9
Ms 29 0.358268 0.11507 0.001441 0.000277 2.99E-05 3.036643 0.005754 74.1 0.4
Ms 30 0.178945 0.044807 0.000537 0.000211 3.49E-05 3.763259 0.033264 91.3 0.9
Ms 31 0.273151 0.093756 0.001232 0.000316 0.000115 2.551233 0.016583 62.4 0.5
Ms 32 0.173772 0.089489 0.00115 1.76E-05 5E-05 1.77673 0.016944 43.7 0.5
(continued)
 TABLE A3. RAW ARGON DATA (continued)
Andaman Flysch (Corbyn’s Cove section)
40 39 38 37 36 40 39
A4 Ar Ar Ar Ar Ar Ar*/ Ar ± Age (Ma) ±
Ms 1 1.757257 0.228175 0.003051 0.001952 0.000479 7.080388 0.0316 168.2 1.1
Ms 2 0.350481 0.117885 0.001625 0.000598 0.000135 2.63508 0.030076 64.4 0.8
Ms 3 0.226229 0.086674 0.001165 0.000669 0.000115 2.218651 0.038875 54.4 1.0
Ms 4 0.84056 0.203536 0.00254 0.000722 0.000305 3.687254 0.018742 89.5 0.6
Ms 5 0.387886 0.07517 0.001007 0.000476 3.49E-05 5.023023 0.044232 120.9 1.2
Ms 6 0.827359 0.256024 0.00326 0.001164 0.00039 2.781793 0.00436 68.0 0.3
Ms 8 0.124755 0.077014 0.001012 0.000247 0.00011 1.198084 0.007833 29.6 0.2
Ms 9 3.009747 0.428003 0.005417 0.000953 0.000195 6.897607 0.010471 164.0 0.8
Ms 10 1.209192 0.187151 0.00232 0.000706 0.000105 6.29557 0.006128 150.3 0.7
Ms 11 0.497906 0.161033 0.00184 0.000212 7.49E-05 2.954419 0.018552 72.1 0.6
Ms 12 3.00958 0.046496 0.000608 0.000395 7.99E-05 64.22046 0.067476 1144.9 4.3
Ms 13 0.239454 0.097718 0.001129 0.000897 0.00013 2.058057 0.005784 50.5 0.3
Ms 15 1.727941 0.097393 0.001247 3.59E-05 7.5E-05 17.51444 0.014541 390.6 1.8
Ms 16 0.259758 0.122808 0.001441 0.000395 3.49E-05 2.0312 0.024229 49.9 0.6
Ms 17 0.102435 0.029899 0.000307 0 1.5E-05 3.277664 0.007241 79.8 0.4
Ms 18 0.565702 0.067938 0.000869 0.000198 3.49E-05 8.174691 0.007531 192.8 0.9
Ms 19 0.461503 0.030601 0.000363 0 5E-06 15.03286 0.049761 340.1 1.9
Ms 20 0.084844 0.03198 0.000358 3.6E-05 9.99E-06 2.560674 0.046289 62.7 1.2
Ms 21 0.051231 0.028354 0.000327 0.000144 9.96E-06 1.70302 0.116858 41.9 2.9
Ms 22 0.539888 0.175027 0.002034 0.000648 0.000225 2.705021 0.007126 66.1 0.4
Ms 23 0.224278 0.125809 0.001584 0.000793 0.000115 1.513072 0.003637 37.3 0.2
Ms 24 2.435008 0.11556 0.001303 0.000343 8.99E-05 20.84143 0.030878 456.1 2.1
Ms 25 0.28137 0.043934 0.000537 0.000162 3E-05 6.202931 0.034616 148.2 1.1
Ms 26 1.447467 0.066026 0.00092 0.001368 0.000185 21.09624 0.0327 461.0 2.1
Ms 27 0.131921 0.043514 0.000777 0.001825 0.000255 1.303279 0.01007 32.2 0.3
Ms 28 0.381247 0.061408 0.000899 0.001902 0.000264 4.935659 0.026829 118.9 0.8
Ms 29 0.400761 0.030962 0.000603 0.001484 0.000155 11.46811 0.056235 265.1 1.7
Ms 30 0.124287 0.055912 0.000761 0.001294 0.000205 1.141272 0.059711 28.2 1.5
Ms 31 0.077119 0.024908 0.000368 0.000686 8.98E-05 2.030571 0.013455 49.9 0.4
Ms 32 0.042558 0.024805 0.000409 0.000952 4.97E-05 1.123077 0.119683 27.7 2.9
Ms 33 0.057102 0.020714 0.000225 0.000305 5.99E-05 1.901885 0.160137 46.7 3.9
Mithakhari Group, Chiriyatapu
40 39 38 37 36 40 39
Ar Ar Ar Ar Ar Ar*/ Ar ± Age (Ma) ±
Ms 1 0.8284261 0.0365259 0.0004242 0.0004699 0.0000000 22.6805064 0.0750144 491.3 2.6
Ms 2 0.7446774 0.0323472 0.0003628 0.0000000 0.0000550 22.5188268 0.0957791 488.2 2.8
Ms 3 0.7987535 0.0472551 0.0006388 0.0002743 0.0000249 16.7471471 0.0197808 375.1 1.7
Ms 4 0.3026851 0.0938337 0.0011243 0.0001764 0.0000750 2.9897197 0.0315933 72.9 0.8
Ms 5 0.6288437 0.0690488 0.0008483 0.0003334 0.0000449 8.9150267 0.0157014 209.3 1.0
Ms 6 0.2056404 0.0084045 0.0001022 0.0000000 0.0000000 24.4679238 0.0604556 524.8 2.5
Ms 8 2.0267770 0.0181880 0.0002760 0.0003637 0.0000349 110.8679577 0.1272294 1674.6 5.6
Ms 9 1.3058958 0.0589862 0.0007666 0.0002830 0.0000949 21.6634575 0.0479489 471.9 2.3
Ms 10 1.0895856 0.0467023 0.0006133 0.0004044 0.0000849 22.7933251 0.0380826 493.4 2.3
Ms 11 1.7582128 0.0745294 0.0008279 0.0005462 0.0000899 23.2345815 0.0322423 501.7 2.3
Ms 12 0.7200124 0.0307043 0.0004191 0.0003844 0.0001499 22.0072636 0.1088272 478.5 3.0
Ms 13 0.7633280 0.0327602 0.0004344 0.0003442 0.0000499 22.8502553 0.0157553 494.5 2.2
Ms 15 1.4321915 0.0604223 0.0007308 0.0002229 0.0000499 23.4587880 0.0323211 506.0 2.3
Ms 16 3.1189393 0.0694722 0.0008943 0.0007095 0.0000998 44.4702431 0.0356104 863.3 3.5
Ms 17 0.5782068 0.0518422 0.0006746 0.0002434 0.0000349 10.9540815 0.0577358 254.0 1.7
Ms 18 1.2848422 0.0599626 0.0007052 0.0001623 0.0000750 21.0580001 0.0540885 460.2 2.3
Ms 19 0.5290525 0.0217523 0.0003015 0.0002639 0.0000199 24.0509065 0.1038536 517.1 3.0
Ms 20 5.1561196 0.0246141 0.0003628 0.0002639 0.0000799 208.5188362 0.2883512 2444.9 6.9
Ms 21 0.6433217 0.0339743 0.0003577 0.0001625 0.0000500 18.5010347 0.0472177 410.2 2.1
Ms 22 1.1030049 0.0956311 0.0011039 0.0005283 0.0001499 11.0708902 0.0175933 256.5 1.3
Ms 23 0.3852341 0.0186428 0.0002044 0.0000000 0.0000000 20.6639642 0.0133670 452.6 2.0
Ms 24 0.8201835 0.0345528 0.0004804 0.0002349 0.0000499 23.3100383 0.0867000 503.2 2.7
Ms 25 0.2651809 0.0818851 0.0009965 0.0008358 0.0001048 2.8603315 0.0027919 69.8 0.3
Ms 26 0.4454907 0.0287515 0.0004242 0.0007315 0.0001248 14.2118131 0.0282976 323.1 1.6
Ms 27 0.2823107 0.0124546 0.0001073 0.0000000 0.0000000 22.6671453 0.0457513 491.0 2.3
Ms 28 7.5439887 0.1794691 0.0021310 0.0008891 0.0001048 41.8625264 0.0295576 822.6 3.3
Ms 29 0.4338841 0.0171655 0.0002402 0.0000000 0.0000000 25.2765779 0.0434406 539.8 2.5
Ms 30 0.9228281 0.0225529 0.0003373 0.0004192 0.0000299 40.5267640 0.1467725 801.4 4.0
Ms 31 2.4734857 0.0295677 0.0004191 0.0005765 0.0000398 83.2566784 0.1423053 1380.1 5.1
Ms 32 1.5381052 0.0256109 0.0003475 0.0004719 0.0000000 60.0566236 0.1661772 1089.0 4.7
Ms 33 0.3426377 0.0155174 0.0001942 0.0002623 0.0000299 21.5108920 0.0746433 469.0 2.5
 TABLE A4. ZIRCON U-Pb DATA (UNCORRECTED AGES) FOR ANDAMAN FLYSCH SAMPLE 1A
206 238 207 235 207 206 206 238 207 235 207 206
Grain no. Pb/ U ± Pb/ U ± Pb/ Pb ± Pb/ U ±1σ Pb/ U ±1σ Pb/ Pb ±1σ
ratio ratio ratio Age (Ma) Age (Ma) Age (Ma)
8 0.56820 0.01491 18.27837 0.00211 0.23368 0.01172 2900.4 61.3 3004.5 46.4 3077.4 77.9
21 0.52114 0.01876 17.80391 0.00319 0.24751 0.01869 2704.0 79.5 2979.2 72.0 3168.9 114.9
18 0.45548 0.02377 10.17671 0.00390 0.16214 0.01534 2419.6 105.3 2451.0 83.7 2478.1 151.3
28 0.30919 0.01417 5.34072 0.00228 0.12476 0.01304 1736.7 69.8 1875.4 88.2 2025.4 174.4
1 0.25134 0.01476 3.17496 0.00192 0.09164 0.01199 1445.4 76.0 1451.1 95.3 1459.9 230.3
37 2.68537 0.14191 0.23801 0.00394 0.08188 0.00394 1376.4 20.5 1324.4 39.1 1242.7 99.7
7 0.20471 0.00761 2.27006 0.00089 0.08055 0.00685 1200.5 40.7 1203.0 56.6 1210.5 158.8
13 0.19499 0.00490 2.12506 0.00101 0.07915 0.00497 1148.4 26.5 1157.0 42.0 1175.9 119.3
31 2.04475 0.18301 0.19077 0.00536 0.07779 0.00536 1125.6 29.0 1130.5 61.0 1141.5 172.8
16 0.19076 0.00770 2.00881 0.00087 0.07645 0.00752 1125.5 41.7 1118.5 63.8 1106.9 184.9
23 0.19007 0.00612 2.22350 0.00114 0.08469 0.00712 1121.8 33.2 1188.5 58.4 1308.4 154.9
2 0.18013 0.00766 2.41204 0.00081 0.09716 0.00881 1067.7 41.8 1246.2 60.4 1570.4 160.9
9 0.17942 0.00506 1.81366 0.00083 0.07343 0.00503 1063.8 27.7 1050.4 42.7 1025.9 132.6
5 0.17514 0.00495 1.78122 0.00078 0.07384 0.00488 1040.3 27.1 1038.6 40.6 1037.0 128.1
3 0.15851 0.00357 1.66696 0.00100 0.07632 0.00404 948.5 19.8 996.0 31.7 1103.5 102.3
35 1.54569 0.07373 0.15281 0.00239 0.07342 0.00239 916.7 13.4 948.8 29.4 1025.7 94.7
4 0.15069 0.00383 1.56797 0.00082 0.07553 0.00458 904.8 21.5 957.6 35.5 1082.6 117.0
29 0.15001 0.00490 1.70213 0.00107 0.08191 0.00734 901.0 27.5 1009.3 58.2 1243.4 166.2
12 0.13848 0.00336 1.34274 0.00107 0.07044 0.00431 836.1 19.1 864.4 34.2 941.1 120.5
39 1.43337 0.20932 0.12528 0.00491 0.08302 0.00491 760.9 28.2 903.0 87.3 1269.6 266.2
26 0.12184 0.00484 1.12034 0.00076 0.06648 0.00719 741.2 27.8 763.1 57.1 821.5 210.9
6 0.11741 0.00687 0.91179 0.00108 0.05641 0.00919 715.6 39.6 658.0 75.1 467.6 325.9
22 0.09140 0.00258 0.76164 0.00058 0.06035 0.00473 563.8 15.2 575.0 34.3 616.3 160.8
11 0.08908 0.00363 0.96494 0.00048 0.07869 0.00822 550.1 21.5 685.8 49.1 1164.5 194.1
32 0.70554 0.09115 0.08743 0.00262 0.05857 0.00262 540.3 15.6 542.1 54.3 551.3 264.1
25 0.08390 0.00418 0.73573 0.00076 0.06343 0.00948 519.4 24.8 559.9 62.5 722.7 288.8
34 0.54771 0.03128 0.06830 0.00115 0.05821 0.00115 425.9 6.9 443.5 20.5 537.1 122.9
10 0.03673 0.00221 0.25590 0.00024 0.05061 0.01065 232.5 13.7 231.4 41.9 223.1 425.1
40 0.16376 0.10117 0.02074 0.00132 0.05728 0.00132 132.3 8.4 154.0 88.3 501.8 982.4
19 0.01567 0.00179 0.10372 0.00025 0.04802 0.02354 100.2 11.3 100.2 45.6 99.0 874.6
27 0.01427 0.00191 0.10618 0.00022 0.05378 0.03759 91.3 12.1 102.5 67.0 361.9 1092.3
33 0.09256 0.01678 0.01397 0.00047 0.04809 0.00047 89.4 3.0 89.9 15.6 103.5 384.6
17 0.01183 0.00088 0.07776 0.00012 0.04771 0.01416 75.8 5.6 76.0 21.1 84.0 585.7
30 0.09413 0.02893 0.01041 0.00056 0.06564 0.00056 66.7 3.5 91.3 26.9 795.0 545.5
24 0.00927 0.00087 0.06214 0.00010 0.04851 0.01680 59.5 5.5 61.2 19.9 124.2 660.0
20 0.00910 0.00077 0.11849 0.00010 0.09449 0.02208 58.4 4.9 113.7 23.6 1518.0 386.4
38 0.05721 0.00743 0.00799 0.00020 0.05193 0.00020 51.3 1.3 56.5 7.1 282.3 275.3
15 0.00778 0.00052 0.05499 0.00005 0.05135 0.01135 49.9 3.3 54.4 11.2 256.6 441.6
36 0.13338 0.08419 0.00723 0.00081 0.13385 0.00081 46.5 5.2 127.1 75.4 2149.0 828.8
14 0.00705 0.00024 0.04889 0.00004 0.05037 0.00528 45.3 1.6 48.5 4.8 212.2 225.9
 TABLE A5A. PETROGRAPHIC DATA
S it e Formation Region Sample Grain size Q KF P Lv Lc Lp Lch Lm M HM Total MI
(μm)
Corbyn’s Cove section N11°38′43.0″ E92°45′29.2″ Andaman Flysch S Andaman FT1B 115 50 13 5 4 0 1 0 25 1 1 100.0 175
Corbyn’s Cove section ~100 m upsection from 1A Andaman Flysch S Andaman FT3 140 48 17 11 3 0 2 0 16 2 1 100.0 160
Corbyn’s Cove section below hotel Andaman Flysch S Andaman FT4 130 47 10 9 5 0 3 1 24 1 0 100.0 150
Namunagarh Quarry Namunagarh Grit S Andaman NAM 3D 1500 2 1 41 52 0 0 0 1 0 2 100.0 2
Hopetown Quarry N11°41′40.7″ E92°43′50.1″ Namunagarh Grit S Andaman NAM 25A 850 21 2 27 40 0 2 2 5 0 1 100.0 15
Mungleton Quarry N11°34′22.3″ E92°39′54.1″ Namunagarh Grit S Andaman NAM 26 1800 19 1 30 45 0 1 1 3 0 0 100.0 11
Irrawaddy River ~ Nyaungdoun Modern river sand Myanmar MY05 23A — 47 11 11 6 0 6 1 15 2 1 100.0 112
Note: The MI (“Metamorphic Index”; Garzanti and Vezzoli, 2003) expresses the average rank of metamorphic rock fragments in the studied samples, and varies from 0 in detritus from
sedimentary and volcanic cover rocks to 500 in detritus from high-grade basement rocks. Q—quartz; KF—potassium feldspar; P—plagioclase; Lv—volcanic; Lc—carbonate lithic fragments
(including marble); Lp—terriginous lithic fragments (shale, siltstone); Lch—chert lithic fragments; Lm—metamorphic lithic fragments; M—micas; HM—heavy minerals.
 TABLE A5B. HEAVY-MINERAL DATA

HM% vfs–fs
HM% transparent
% transparent
% opaque
% turbid
Total
Zircon
Dravite
Schorlite
Rutile
Sagenite
Sphene

S it e Formation Sample
Corbyn’s Cove section N11°38′43.0″ E92°45′29.2″ Andaman Flysch FT1B 0.9 0.5 54 9 37 100 3 4 4 5 0 3
Corbyn’s Cove section ~100 m upsection from 1A Andaman Flysch FT3 0.4 0.2 44 7 49 100 4 4 5 5 2 2
Corbyn’s Cove section below hotel Andaman Flysch FT4 0.6 0.2 35 8 57 100 4 3 6 7 1 0
Namunagarh Quarry Namunagarh Grit NAM 3D 4.0 1.5 37 38 25 100 0 0 0 0 0 1
Hopetown Quarry N11°41′40.7″ E92°43′50.1″ Namunagarh Grit NAM 25A 1.3 0.6 42 16 42 100 1 0 0 1 0 15
Mungleton Quarry N11°34′22.3″ E92°39′54.1″ Namunagarh Grit NAM 26 1.0 0.5 53 7 40 100 0 0 0 1 0 16
Irrawaddy River ~ Nyaungdoun River sand MY05 23A 5.0 4.0 79 3 17 100 0 0 0 0 0 3

Anatase
Brookite
Ti aggregates
Apatite
Xenotime
Monazite
Barite
Others
Blue-green
hornblende
Green
hornblende
Green-brown
hornblende
Brown
hornblende

Site Formation Sample

Corbyn’s Cove section N11°38′43.0″ E92°45′29.2″ Andaman Flysch FT1B 0 0 24 10 0 0 0 0 0 0 0 0
Corbyn’s Cove section ~100 m upsection from 1A Andaman Flysch FT3 0 0 30 17 0 1 0 0 0 0 0 0
Corbyn’s Cove section below hotel Andaman Flysch FT4 0 0 24 11 0 0 0 0 0 0 0 0
Namunagarh Quarry Namunagarh Grit NAM 3D 0 0 0 3 0 0 0 0 1 2 20 0
Hopetown Quarry N11°41′40.7″ E92°43′50.1″ Namunagarh Grit NAM 25A 0 0 3 1 0 0 0 0 0 0 0 0
Mungleton Quarry N11°34′22.3″ E92°39′54.1″ Namunagarh Grit NAM 26 0 0 1 3 0 0 0 0 0 0 0 0
Irrawaddy River ~ Nyaungdoun River sand MY05 23A 0 0 0 0 0 0 0 0 44 2 2 1
(continued)
 TABLE A5B. HEAVY-MINERAL DATA (continued)

Oxyhornblende
Glaucophane
Sodic amphiboles
Tremolite
Actinolite
Antofillite
Other amphiboles
Hypersthene
Olivine
Spinel
Epidote
Clinozoisite

Site Formation Sample

Corbyn’s Cove section N11°38′43.0″ E92°45′29.2″ Andaman Flysch FT1B 0 0 0 0 0 0 0 0 0 2 18 0
Corbyn’s Cove section ~100 m upsection from 1A Andaman Flysch FT3 0 0 0 0 0 0 0 0 0 3 2 0
Corbyn’s Cove section below hotel Andaman Flysch FT4 0 0 0 0 0 0 0 0 0 2 0 0
Namunagarh Quarry Namunagarh Grit NAM 3D 0 0 0 0 0 0 0 0 0 0 11 0
Hopetown Quarry N11°41′40.7″ E92°43′50.1″ Namunagarh Grit NAM 25A 0 0 0 0 0 0 0 0 0 16 43 0
Mungleton Quarry N11°34′22.3″ E92°39′54.1″ Namunagarh Grit NAM 26 0 0 0 0 0 0 0 0 0 7 44 0
Irrawaddy River ~ Nyaungdoun River sand MY05 23A 0 0 0 0 0 2 0 0 0 0 33 0

Zoisite
Other epidotes
Prehnite
Pumpellyite
Chloritoid
Lawsonite
Carpholite
Garnet
Staurolite
Andalusite
Kyanite
Sillimanite
Total transparent

Site Formation Sample

Corbyn’s Cove section N11°38′43.0″ E92°45′29.2″ Andaman Flysch FT1B 0 0 0 0 12 0 0 14 0 0 0 0 100
Corbyn’s Cove section ~100 m upsection from 1A Andaman Flysch FT3 0 0 0 0 12 0 0 13 0 0 0 0 100
Corbyn’s Cove section below hotel Andaman Flysch FT4 0 0 0 0 25 0 0 15 1 0 0 0 100
Namunagarh Quarry Namunagarh Grit NAM 3D 0 0 0 0 0 0 0 0 0 0 0 0 100
Hopetown Quarry N11°41′40.7″ E92°43′50.1″ Namunagarh Grit NAM 25A 0 2 0 0 0 0 0 0 0 0 0 0 100
Mungleton Quarry N11°34′22.3″ E92°39′54.1″ Namunagarh Grit NAM 26 0 0 0 0 0 0 0 1 0 0 0 0 100
Irrawaddy River ~ Nyaungdoun River sand MY05 23A 0 1 0 0 0 0 0 6 0 0 0 0 100
Note: HM—heavy minerals; vfs—very fine sand; fs—fine sand.
 TABLE A6A. WHOLE ROCK Sm-Nd DATA
147 144 143 144
Sample Location Formation / GPS Lithology Sm Nd Sm/Nd Sm/ Nd Nd/ Nd 1σ εNd
(ppm)
Andaman Islands
AN05-31 Namunagarh Quarry Namunagarh Grit Sandstone 4.83 25.3134 0.1909 0.1154 0.512271 16 –7.2
AN05 3E Namunagarh Quarry Namunagarh Grit Mudstone 1.99 7.18860 0.2769 0.1674 0.512799 19 3.1
AF1A Corbyn’s Cove Andaman Flysch Mudstone 11.17 50.6900 0.2204 0.1332 0.512068 10 –11.1
AF2 Corbyn’s Cove Andaman Flysch Sandstone 5.05 25.7129 0.1966 0.1188 0.512220 12 –8.2

Myanmar—Rivers draining Eocene rocks

MY05 8A Lemyu River N20°49.212′, E93°18.576′ MR sand 4.02 20.0489 0.2010 0.1215 0.512300 18 –7.4
MY05 15A Kyeintuli River N17°57.139′, E94°33.087′ MR sand 3.33 16.4129 0.2035 0.1250 0.512434 14 –4.0
MY05 17B Thanlwe River N18°59.097′, E94°15.244′ MR mud 3.48 17.6156 0.1977 0.1195 0.512427 18 –4.1
MY05 22B Thandwe River N18°27.466′, E94°23.563′ MR mud 3.60 18.0331 0.1997 0.1207 0.512424 14 –4.2
MY05 23B Irrawaddy River N18°48.391′, E95°12.218′ MR mud 3.67 17.9317 0.2052 0.124 0.512210 18 –8.3
Note: MR—modern river sample; GPS—Global Positioning System.
 TABLE A6B. SINGLE-GRAIN APATITE Sm-Nd DATA
147 144 143 144
Sample Formation / Age Lithology Sm/ Nd 2 S.D. of NIST610 2σ Nd/ Nd 2σ ppm εNd
(internal) (0)
1 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.123272 0.001140 0.000069 0.512855 134 4.24
2 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.070462 0.000408 0.000065 0.512803 127 3.21
3 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.090724 0.000581 0.000058 0.512237 114 –7.83
4 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.073481 0.000171 0.000054 0.512588 105 –0.98
5 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.082561 0.000180 0.000031 0.512485 61 –2.98
6 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.100785 0.002208 0.000062 0.512404 121 –4.57
7 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.106736 0.000552 0.000046 0.512211 90 –8.33
8 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.093898 0.000488 0.000033 0.511800 65 –16.35
9 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.108342 0.001112 0.000047 0.512545 93 –1.81
10 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.099159 0.000388 0.000033 0.512669 64 0.61
11 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.113837 0.000346 0.000051 0.512794 99 3.03
12 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.120108 0.000750 0.000044 0.512621 86 –0.33
13 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.078972 0.000114 0.000067 0.512291 131 –6.77
14 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.067596 0.000236 0.000075 0.512425 147 –4.16
15 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.114968 0.000317 0.000043 0.511822 83 –15.92
16 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.105069 0.001427 0.000067 0.512792 131 3.01
17 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.117803 0.000238 0.000055 0.512607 108 –0.60
18 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.085210 0.000165 0.000056 0.512321 109 –6.19
19 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.102799 0.000424 0.000051 0.512782 99 2.80
20 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.082935 0.000202 0.000136 0.512486 266 –2.96
21 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.136207 0.000550 0.000118 0.512192 230 –8.71
22 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.131323 0.000741 0.000062 0.512791 122 2.99
23 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.244791 0.003438 0.000076 0.512607 148 –0.61
24 Andaman Flysch AF1A Corbyn’s Cove Apatite grain 0.172124 0.000850 0.000079 0.512771 154 2.59
Note: Global Positioning System (GPS) reference for AF1A Corbyn’s Cove: N11°38.430' E92°45.292'; S.D.—standard deviation.
254 Allen et al.
ACKNOWLEDGMENTS
This paper has benefited from thoughtful reviews by Peter Clift,
Joseph Curray, and Christophe Colin. The work was funded by
NERC grants NE/B503192/1 and NER/S/A/2004/12158, with
additional support from the Royal Society short-visits scheme.
REFERENCES CITED
Acharyya, S.K., Ray, K.K., and Roy, D.K., 1989, Tectonic stratigraphy and
emplacement history of the ophiolite assemblage from Naga Hills and
Andaman island arc, India: Journal of the Geological Society of India,
v. 33, p. 4–18.
Ahmad, T., Harris, N., Bickle, M., Chapman, H., Bunbury, J., and Prince, C.,
2000, Isotopic constraints on the structural relationships between the
Lesser Himalayan Series and the High Himalayan Crystalline Series, Gar-
hwal Himalaya: Geological Society of America Bulletin, v. 112, p. 467–
477, doi: 10.1130/0016-7606(2000)112<0467:ICOTSR>2.3.CO;2.
Auge, T., Cocherie, A., Genna, A., Armstrong, R., Guerrot, C., Mukherjee,
N.M., and Patra, R.N., 2003, Age of the Baula PGE mineralization (Oris-
sam India) and its implications concerning the Singhbhum Archaean
nucleus: Precambrian Research, v. 121, p. 85–101, doi: 10.1016/S0301-
9268(02)00202-4.
Bandopadhyay, P.C., 2005, Discovery of abundant pyroclasts in Namunagarh
Grit, South Andaman; evidence for arc volcanism and active subduction
during the Palaeogene in the Andaman area: Journal of Asian Earth Sci-
ences, v. 25, p. 95–107, doi: 10.1016/j.jseaes.2004.01.007.
Bandopadhyay, P.C., and Ghosh, M., 1999, Facies, petrology and depositional
environment of the Tertiary sedimentary rocks around Port Blair, South
Andaman: Journal of the Geological Society of India, v. 52, p. 53–66.
Barley, M.E., Pickard, A.L., Zaw, K., Rak, P., and Doyle, M.G., 2003, Jurassic
to Miocene magmatism and metamorphism in the Mogok metamorphic
belt and the India-Eurasia collision in Myanmar: Tectonics, v. 22, p. 1019,
doi: 10.1029/2002TC001398.
Bertrand, G., Rangin, C., Maluski, H., Han, T.A., Thein, M., Myint, O., Maw,
W., and Lwin, S., 1999, Cenozoic metamorphism along the Shan scarp
(Myanmar): Evidences for ductile shear along the Sagaing fault or the
northward migration of the eastern Himalayan syntaxis?: Geophysical
Research Letters, v. 26, p. 915–918, doi: 10.1029/1999GL900136.
Bertrand, G., Rangin, C., Maluski, H., and Bellon, H., 2001, Diachronous cool-
ing along the Mogok Metamorphic Belt (Shan scarp, Myanmar): The trace
of the northward migration of the Indian syntaxis: Journal of Asian Earth
Sciences, v. 19, p. 649–659, doi: 10.1016/S1367-9120(00)00061-4.
Bodet, F., and Schärer, U., 2000, Evolution of the SE-Asian continent from U-
Pb and Hf isotopes in single grains of zircon and baddeleyite from large
rivers: Geochimica et Cosmochimica Acta, v. 64, p. 2067–2209, doi:
10.1016/S0016-7037(00)00352-5.
Brewer, I.D., Burbank, D.W., and Hodges, K.V., 2003, Modelling detrital cool-
ing-age populations: Insights from two Himalayan catchments: Basin
Research, v. 15, p. 305–320, doi: 10.1046/j.1365-2117.2003.00211.x.
Campbell, I.H., Reiners, P.W., Allen, C.M., Nicolescu, S., and Upadhyay, R.,
2005, He-Pb double dating of detrital zircons from the Ganges and Indus
Rivers: Implication for quantifying sediment recycling and provenance
studies: Earth and Planetary Science Letters, v. 237, p. 402–432, doi:
10.1016/j.epsl.2005.06.043.
Carter, A., 1999, Present status and future avenues of source region discrimina-
tion and characterization using fission-track analysis: Sedimentary Geol-
ogy, v. 124, p. 31–45, doi: 10.1016/S0037-0738(98)00119-5.
Carter, A., and Bristow, C.S., 2000, Detrital zircon geochronology: Enhancing
the quality of sedimentary source information through improved method-
ology and combined U-Pb and fission track techniques: Basin Research,
v. 12, p. 47–57, doi: 10.1046/j.1365-2117.2000.00112.x.
Chakraborty, P.P., and Pal, T., 2001, Anatomy of a forearc submarine fan:
Upper Eocene–Oligocene Andaman Flysch Group, Andaman Islands,
India: Gondwana Research, v. 4, p. 477–487, doi: 10.1016/S1342-
937X(05)70347-6.
Chakraborty, P.P., Pal, T., Dutta Gupta, T., and Gupta, K.S., 1999, Facies pat-
tern and depositional motif in an immature trench-slope basin, Eocene
Mithakhari Group, Middle Andaman, India: Journal of the Geological
Society of India, v. 53, p. 271–284.
Charusiri, P., Clark, A.H., Farrar, E., Archibald, D., and Charusiri, B., 1993,
Granite belts in Thailand: Evidence from the 40Ar/39Ar geochronological
and geological syntheses: Journal of Southeast Asian Earth Sciences, v. 8,
p. 127–136, doi: 10.1016/0743-9547(93)90014-G.
Cherniak, D.J., and Watson, E.B., 2001, Pb diffusion in zircon: Chemical Geol-
ogy, v. 172, p. 5–24, doi: 10.1016/S0009-2541(00)00233-3.
Clark, M.K., Schoenbohm, L.M., Royden, L., Whipple, K.X., Burchfiel, B.C.,
Zhang, X., Tang, W., Wang, E., and Chen, L., 2004, Surface uplift, tec-
tonics, and erosion of eastern Tibet from large-scale drainage patterns:
Tectonics, v. 23, p. TC1006, doi: 10.1029/2002TC001402.
Clark, M.K., House, M.A., Royden, L.H., Whipple, K.X., Burchfiel, B.C.,
Zhang, X., and Tang, W., 2005, Late Cenozoic uplift of southeastern
Tibet: Geology, v. 33, p. 525–528, doi: 10.1130/G21265.1.
Colin, C., Turpin, L., Bertaux, J., Desprairies, A., and Kissel, C., 1999, Ero-
sional history of the Himalayan and Burman ranges during the last two
glacial-interglacial cycles: Earth and Planetary Science Letters, v. 171, p.
647–660, doi: 10.1016/S0012-821X(99)00184-3.
Colin, C., Turpin, L., Bertaux, J., Frank, N., Kissel, C., and Duchamp, S., 2006,
Evolution of weathering patterns in the Indo-Burman Ranges over the
last 280 kyr: Effects of sediment provenance on 87Sr/86Sr ratios tracer:
Geochemistry, Geophysics, Geosystems, v. 7, p. 1–16, doi:10.1029/
2005GC000962.
Curray, J.R., 2005, Tectonics and history of the Andaman Sea region:
Journal of Asian Earth Sciences, v. 25, p. 187–232, doi: 10.1016/
j.jseaes.2004.09.001.
Curray, J.R., Moore, D.G., Lawver, L.A., Emmel, F.J., Raitt, R.W., Henry, M.,
and Kieckhefer, R., 1979, Tectonics of the Andaman Sea and Burma, in
Watkins, J., et al., eds., Geological and Geophysical Investigations of
Continental Margins: American Association of Petroleum Geologists
Memoir 29, p. 189–198.
DeCelles, P.G., Gehrels, G.E., Quade, J., and Ojha, T.P., 1998, Eocene to
early Miocene foreland basin development and the history of Himalayan
thrusting, western and central Nepal: Tectonics, v. 17, p. 741–765, doi:
10.1029/98TC02598.
DeCelles, P.G., Gehrels, G.E., Najman, Y., Martin, A.J., and Garzanti, E., 2004,
Detrital geochronology and geochemistry of Cretaceous–Early Miocene
strata of Nepal: Implications for timing and diachroneity of initial Hima-
layan orogenesis: Earth and Planetary Science Letters, v. 227, p. 313–330,
doi: 10.1016/j.epsl.2004.08.019.
Dickinson, W.R., 1985, Interpreting provenance relations from detrital modes
of sandstones: Provenance of arenites, in Zuffa, G.G., ed., NATO ASI:
Dordrecht, Netherlands, Reidel, v. 148, p. 333–361.
Draut, A.E., and Clift, P.D., 2006, Sedimentary processes in modern and ancient
oceanic arc settings: Evidence from the Jurassic Talkeetna Formation of
Alaska and the Mariana and Tonga Arcs, western Pacific: Journal of Sedi-
mentary Research, v. 76, p. 493–514, doi: 10.2110/jsr.2006.044.
Foster, G.L., and Vance, D., 2006, In situ Nd isotopic analysis of geological
materials by laser ablation MC-ICP-MS: Journal of Analytical Atomic
Spectrometry, v. 21, p. 288–296, doi: 10.1039/b513945g.
Galbraith, R.F., and Green, P.F., 1990, Estimating the component ages in a finite
mixture: Nuclear Tracks and Radiation Measurements, v. 17, p. 197–206,
doi: 10.1016/1359-0189(90)90035-V.
Galbraith, R.F., and Laslett, G.M., 1993, Statistical models for mixed fission
track ages: Nuclear Tracks and Radiation Measurements, v. 21, p. 459–
470, doi: 10.1016/1359-0189(93)90185-C.
Gallagher, K., 1995, Evolving temperature histories from apatite FT data: Earth
and Planetary Science Letters, v. 136, p. 421–435, doi: 10.1016/0012-
821X(95)00197-K.
Galy, A., and France-Lanord, C., 2001, Higher erosion rates in the Himalaya:
Geochemical constraints on riverine fluxes: Geology, v. 29, p. 23–26, doi:
10.1130/0091-7613(2001)029<0023:HERITH>2.0.CO;2.
Garzanti, E., and Vezzoli, G., 2003, A classification of metamorphic grade
in sands based on their composition and grade: Journal of Sedimentary
Research, v. 73, p. 830–837.
Green, P.F., Duddy, I.R., and Bray, R.J., 1995, Applications of thermal history
reconstruction in inverted basins, in Buchanan, J.G., and Buchanan, P.G.,
eds., Basin Inversion: Geological Society [London] Special Publication
88, p. 149–165.
 Sedimentation and uplift history of the Andaman-Nicobar accretionary prism
Guha, D.K., and Mohan, M., 1965, On the ostracod from the Neogene of Anda-
man Island: Journal of the Geological Survey of India, v. 9, p. 58–66.
Guillot, S., Garzanti, E., Baratoux, D., Marquer, D., Mahéo, G., and de
Sigoyer, J., 2003, Reconstructing the total shortening history of the NW
Himalaya: Geochemistry, Geophysics, Geosystems, v. 4, p. 1064, doi:
10.1029/2002GC000484.
Haines, P.W., Turner, S.P., Kelley, S.P., Wartho, J.-A., and Sherlock, S.C.,
2004, Ar/Ar dating of detrital muscovite in provenance investiga-
tions: A case study from the Adelaide Rift Complex, South Australia:
Earth and Planetary Science Letters, v. 227, p. 297–311, doi: 10.1016/
j.epsl.2004.08.020.
Halder, D., 1985, Some aspects of the Andaman Ophiolite Complex: Geologi-
cal Survey of India, v. 115, p. 1–11.
Hurford, A.J., 1990, Standardization of fission track dating calibration: Recom-
mendation by the Fission Track Working Group of the IUGS Subcommis-
sion on Geochronology: Chemical Geology, v. 80, p. 177–178.
Karunakaran, C., Ray, K.K., and Saha, S.S., 1968, Tertiary sedimentation in
Andaman-Nicobar geosyncline: Journal of the Geological Society of
India, v. 9, p. 32–39.
Ketcham, R.A., Donelick, R.A., and Carlson, W.D., 1999, Variability of apa-
tite fission-track annealing kinetics: III. Extrapolation to geological time-
scales: American Mineralogist, v. 84, p. 1235–1255.
Khan, P.K., and Chakraborty, P.P., 2005, Two-phase opening of Andaman Sea:
A new seismotectonic insight: Earth and Planetary Science Letters, v.
229, p. 259–271.
Lee, T.T., and Lawver, L.A., 1995, Cenozoic plate reconstruction of South-
east Asia: Tectonophysics, v. 251, p. 85–138, doi: 10.1016/0040-
1951(95)00023-2.
Mallik, T.K., 1976, Shelf sediments of the Ganges delta with special empha-
sis on the mineralogy of the western part, Bay of Bengal, Indian Ocean:
Marine Geology, v. 22, p. 1–32, doi: 10.1016/0025-3227(76)90007-4.
Marsaglia, K.M., and Ingersoll, R.V., 1992, Compositional trends in arc-related,
deep-marine sand and sandstone; a reassessment of magmatic-arc prov-
enance: Geological Society of America Bulletin, v. 104, p. 1637–1649,
doi: 10.1130/0016-7606(1992)104<1637:CTIARD>2.3.CO;2.
Meesters, A.G.C.A., and Dunai, T.J., 2002, Solving the production diffusion
equation for finite diffusion domains of various shapes; Part II, Appli-
cation to cases with alpha-ejection and nonhomogeneous distribution of
the source: Chemical Geology, v. 186, p. 57–73, doi: 10.1016/S0009-
2541(01)00423-5.
Metcalfe, I., 1996, Pre-Cretaceous evolution of SE Asian terranes, in Hall, R.,
and Blundell, D., eds., Tectonic Evolution of Southeast Asia: Geological
Society [London] Special Publication 106, p. 97–122.
Mishra, M., and Rajamani, V., 1999, Significance of the Archaean bimodal vol-
canics from the Ramagiri schist belt in the formation of the Eastern Dhar-
war craton: Journal of the Geological Society of India, v. 54, p. 563–583.
Mitchell, A.H.G., 1993, Cretaceous–Cenozoic tectonic events in the western
Myanmar (Burma)–Assam region: Geological Society [London] Journal,
v. 150, p. 1089–1102.
Mountain, G.S., and Prell, W.L., 1990, A multiphase plate tectonic history of
the southeast continental margin of Oman, in Robertson, A.H.F., et al.,
eds., The Geology and Tectonics of the Oman Region: Geological Society
[London] Special Publication 49, p. 725–743.
Najman, Y., and Garzanti, E., 2000, Reconstructing early Himalayan tec-
tonic evolution and paleogeography from Tertiary foreland basin sedi-
mentary rocks, northern India: Geological Society of America Bul-
letin, v. 112, p. 435–449, doi: 10.1130/0016-7606(2000)112<0435:
REHTEA>2.3.CO;2.
Najman, Y., Bickle, M., and Chapman, H., 2000, Early Himalayan exhumation:
Isotopic constraints from the Indian foreland basin: Terra Nova, v. 12, p.
28–34, doi: 10.1046/j.1365-3121.2000.00268.x.
Najman, Y., Johnson, C., White, N.M., and Oliver, G., 2004, Evolution of the
Himalayan foreland basin, NW India: Basin Research, v. 16, p. 1–24, doi:
10.1111/j.1365-2117.2004.00223.x.
255
Najman, Y., Carter, A., Oliver, G., and Garzanti, E., 2005, Provenance of early
foreland basin sediments, Nepal: Constraints to the timing and diachro-
neity of early Himalayan orogenesis: Geology, v. 33, p. 309–312, doi:
10.1130/G21161.1.
Oldham, R.D., 1885, Notes on the geology of Andaman Islands: Record of the
Geological Survey of India. v. 18, p. 135–145.
Pal, T., Chakraborty, P.P., Gupta, T.D., and Singh, C.D., 2003, Geodynamic
evolution of the outer-arc–forearc belt in the Andaman Islands, the cen-
tral part of the Burma-Java subduction complex: Geological Magazine,
v. 140, p. 289–307, doi: 10.1017/S0016756803007805.
Pal, T., Gupta, T.D., Chakraborty, P.P., and Gupta, S.C.D., 2005, Pyroclastic
deposits of Mio-Pliocene age in the Arakan Yoma-Andaman-Java subduc-
tion complex, Andaman Islands, Bay of Bengal, India: Geochemical Jour-
nal, v. 39, p. 69–82, doi: 10.2343/geochemj.39.69.
Peucat, J.J., Vidal, P., Bernard-Griffiths, J., and Condie, K.C., 1989, Sr, Nd and
Pb isotopic systematics in the Archaean low-to-high-grade transition zone
of southern India: Syn-accretion granulites: Journal of Geology, v. 97,
p. 537–550.
Pivnik, D.A., Nahm, J., Tucker, R.S., Smith, G.O., Nyein, K., Nyunt, M., and
Maung, P.H., 1998, Polyphase deformation in a fore-arc/back-arc basin,
Salin sub-basin, Myanmar (Burma): American Association of Petroleum
Geologists Bulletin, v. 82, p. 1837–1856.
Ray, K.K., Sengupta, S., and Van Den Hul, H.J., 1988, Chemical characters of
volcanic rocks from Andaman ophiolite, India: Geological Society [Lon-
don] Journal, v. 145, p. 392–400.
Renne, P.R., Swisher, C.C., Deino, A.L., Karner, D.B., Owens, T.L., and
DePaolo, D.J., 1998, Intercalibration of standards, absolute ages and
uncertainties in 40Ar/39Ar dating: Chemical Geology, v. 145, p. 117–152,
doi: 10.1016/S0009-2541(97)00159-9.
Robinson, D.M., DeCelles, P.G., Patchett, P.J., and Garzione, C.N., 2001,
The kinematic evolution of the Nepalese Himalaya interpreted from Nd
isotopes: Earth and Planetary Science Letters, v. 192, p. 507–521, doi:
10.1016/S0012-821X(01)00451-4.
Rowley, D.B., 1996, Age of initiation of collision between India and Asia: A
review of stratigraphic data: Earth and Planetary Science Letters, v. 145,
p. 1–13, doi: 10.1016/S0012-821X(96)00201-4.
Roy, S.K., 1992, Accretionary prism in Andaman forearc: Geological Survey of
India Special Publication 29, p. 273–278.
Roy, T.K., 1983, Geology and hydrocarbon prospects of Andaman–Nicobar
basin, in Bhandari, L.L., ed., Petroliferous Basins of India: Petroleum
Asia Journal, p. 37–50.
Saha, A., Basu, A.R., Garzione, C.N., Bandyopadhyay, P.K., and Chakrabarti,
A., 2004, Geochemical and petrological evidence for subduction-accretion
processes in the Archean Eastern Indian Craton: Earth and Planetary Sci-
ence Letters, v. 220, p. 91–106, doi: 10.1016/S0012-821X(04)00056-1.
Sambridge, M.S., and Compston, W., 1994, Mixture modelling of multi-compo-
nent data sets with application to ion-probe zircons: Earth and Planetary
Science Letters, v. 128, p. 373–390, doi: 10.1016/0012-821X(94)90157-0.
Sherlock, S.C., and Kelley, S.P., 2002, Excess argon in HP-LT rocks: A UV
laserprobe study of phengite and K-free minerals: Chemical Geology,
v. 182, p. 619–636, doi: 10.1016/S0009-2541(01)00345-X.
Singh, O.P., Subramanya, S.M., and Sharma, V., 2000, Early Neogene multiple
microfossil biostratigraphy, John Lawrence Island: Andaman Sea: Micro-
palaeontology, v. 46, p. 343–352.
Singh, S.K., and France-Lanord, C., 2002, Tracing the distribution of erosion
in the Brahmaputra watershed from isotopic compositions of stream
sediments: Earth and Planetary Science Letters, v. 202, p. 645–662, doi:
10.1016/S0012-821X(02)00822-1.
Svojtka, M., Ko, J., and Venera, Z., 2001, Dating granulite-facies structures and
the exhumation of lower crust in the Moldanubian Zone of the Bohemian
Massif: International Journal of Earth Sciences, v. 91, p. 373–385.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Post-collisional collapse in the wake of migrating arc-continent collision in the

Ilan Basin, Taiwan

Peter D. Clift
School of Geosciences, University of Aberdeen, Aberdeen, AB24 3UE, UK, and DFG-Research Centre Ocean Margins (RCOM),
Geowissenschaften, Universität Bremen, Klagenfurter Strasse, 28359 Bremen, Germany

Andrew T.S. Lin

Department of Earth Sciences, National Central University, 300, Jungda Road, Jungli, Taiwan

Andrew Carter
School of Earth Sciences, University and Birkbeck College London, Gower Street, London, WC1E 6BT, UK

Francis Wu
State University of New York at Binghamton, Department of Geological Sciences and Environmental Studies, P.O. Box 600,
Binghamton, New York 13902-6000, USA

Amy E. Draut
U.S. Geological Survey, Pacific Science Center, 400 Natural Bridges Drive, Santa Cruz, California 95060, USA

T.-H. Lai
L.-Y. Fei
Central Geological Survey, 2, Lane 109, Hua-Hsin Street, Chung-Ho, Taipei, Taiwan 235

Hans Schouten
Department of Geology and Geophysics, Woods Hole Oceanographic Institution, Woods Hole, Massachusetts 02543, USA

Louis Teng
Department of Geosciences, National Taiwan University, Roosevelt Road, Taipei, Taiwan

ABSTRACT

The Ilan Basin of northern Taiwan forms the western limit of the Okinawa Trough,
where the trough meets the compressional ranges of central Taiwan. Apatite fission-
track ages of 1.2 ± 0.5 Ma and 3.5 ± 0.5 Ma, measured north and south of the basin,
respectively, indicate faster exhumation rates in the Hsüehshan Range to the north
(>1.6 mm/yr) than in the Backbone Range to the south (0.7 mm/yr). Reconstructed
subsidence rates along the northern basin margin are also faster than in the south
(6–7 compared with 3–5 mm/yr). Global positioning system (GPS) and active seismo-
logical data indicate motion of the southern basin margin to the east and southeast.

Clift, P.D., Lin, A.T.S., Carter, A., Wu, F., Draut, A.E., Lai, T.-H., Fei, L.-Y., Schouten, H., and Teng, L., 2008, Post-collisional collapse in the wake of migrating
arc-continent collision in the Ilan Basin, Taiwan, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc
Collision Zones: Geological Society of America Special Paper 436, p. 257–278, doi: 10.1130/2008.2436(12). For permission to copy, contact editing@geosociety.
org. ©2008 The Geological Society of America. All rights reserved.

257
258 Clift et al.

We propose that the Ilan Basin is being formed as a result of extension of northern
Taiwan, largely controlled by a major southeast-dipping fault, modeled at ~30° dip,
and mapped as a continuation of the Lishan Fault, a major thrust structure in the
Central Ranges. Flexural rigidity of the lithosphere under the basin is low, with elastic
thickness ~3 km. A southwest-migrating collision between the Luzon Arc and south-
ern China, accompanied by subduction polarity reversal in the Ryukyu Trench, has
allowed crustal blocks that were previously held in compression between the Eurasian
and Philippine Sea plates to move trenchward as they reach the northern end of the
collision zone. Subduction polarity reversal permits rapid extension and formation of
the Ilan Basin and presumably, at least, the western Okinawa Trough, as a direct con-
sequence of arc-continent collision, not because of independent trench rollback forces.
This conceptual model suggests that migrating arc-continent collision causes the rapid
formation of deep marginal basins that are then filled by detritus from the adjacent
orogen, and that these should be common features in the geologic record.

Keywords: collision, extension, erosion, subduction, seismology.

INTRODUCTION
The Earth’s continental crust is generally considered to be
largely generated by magmatism along active plate margins, a
process that balances the long-term loss of continental crust back
into the mantle through subduction zones (e.g., von Huene and
Scholl, 1991; Rudnick and Fountain, 1995; Clift and Vannucchi,
2004). This balance is maintained because arc crust generated in
oceanic subduction settings is not entirely subducted when these
features collide with other trench systems or passive continental
margins. Instead, continental masses appear to be built up as a
result of the progressive accretion of primitive arc blocks to older,
preexisting continental blocks. Understanding the tectonism of
arc-continent collision is important, not only because this pro-
cess is fundamental to the conservation of the total volume of
continental crust, but also because it is likely the most important
method by which active continental margins are created (e.g.,
Casey and Dewey, 1984; Konstantinovskaya, 1999).
Accreted oceanic island-arc terranes have been recognized
in a number of orogenic suture zones (e.g., Kohistan in the Hima-
laya, Talkeetna in southern Alaska, South Mayo in the Irish Cale-
donides), and the process of arc-continent collision forms an inev-
itable part of the plate tectonic cycle. Arc-continent and continent-
continent collisions are unlikely to be identical because of the
different mechanical properties of arc and cratonic lithospheres.
In this paper we investigate the later stages of arc-continent colli-
sion, specifically the collapse of the collisional mountain belts to
form sedimentary basins whose fill might be used to understand
the collision process.
In order to understand the collapse of arc collisional moun-
tains, we examine the Ilan Basin of northern Taiwan. Taiwan,
located off the coast of southeastern China, comprises some of
the tallest and most rapidly uplifting and eroding mountains in
East Asia. The orogen is formed by the collision between the oce-
anic Luzon Arc and the passive margin of southern China (Suppe,
1981; Chemenda et al., 1997), which is itself built on the remains
of a Cretaceous continental arc (Davis et al., 1997). The collision
between the Luzon oceanic arc and mainland Asia is oblique, so
that the collision point must have migrated along the Asian con-
tinental margin through time. At any given time the north end
of the island will represent arc units and Chinese passive-margin
sequences that began to be involved in collision earlier than those
in the south, where collision is just starting. Thus northern Tai-
wan is the optimal place to examine the processes that occur dur-
ing and after the peak of arc-continent collision. These processes
form the subject of this paper.
Radiometric dating of the metamorphic rocks exposed in
the Central Ranges, combined with biostratigraphic dating of
orogenic sediment from the Coastal Ranges, indicates that col-
lision of this part of the Luzon Arc with mainland Asia started
ca. 6–9 Ma (e.g., Suppe, 1981; Teng, 1990; Sibuet et al., 2002;
Malavieille et al., 2002; Huang et al., 2006). A wide consensus in
the community agrees that Taiwan represents a simple collision
between the Luzon Arc and mainland Asia, an assumption that
underlies this present study. Nonetheless, we note that alternative
models have been advanced to explain the mountain building.
One theory involves collision of an exotic terrane with the Chi-
nese passive margin, followed by later collision of the Luzon Arc
(Lu and Hsü, 1992). Another model invokes compressional tecto-
nism following collision of the Luzon Arc with the Ryukyu Arc,
which would previously have extended farther southwest along
the southern margin of China (Hsu and Sibuet, 1995; Sibuet and
Hsu, 1997).
It is unclear when the collision between Luzon and Eurasia
began. Whereas some models favor collision to have initiated
only ca. 6–9 Ma, effectively just to the east of the present col-
lision zone (e.g., Suppe, 1984; Sibuet and Hsu, 1997; Huang et
al., 2000, 2006), other models suggest a more steady-state colli-
sion that may have started earlier (e.g., Suppe, 1984; Teng, 1990,
1996; Clift et al., 2003). In the steady-state model the modern
 Migrating arc-continent collision in the Ilan Basin, Taiwan 259

collision represents a snapshot of a continuous collision process
that is migrating to the southwest along the Chinese margin,
accreting at least part of the crust of the Luzon Arc to the edge of
Eurasia, while generating a new Ryukyu Arc-Trench system and
associated Okinawa Trough in its wake. Debate continues as to
whether the Okinawa Trough is an active rift system formed by
trench tectonic forces linked to the Ryukyu subduction zone and
propagating to the southwest into the northern tip of the Taiwan
orogen (e.g., Suppe, 1984; Sibuet et al., 1998; Wang et al., 1999),
or whether it represents the product of gravitational collapse of
the Taiwan orogen, following subduction polarity reversal (e.g.,
Teng, 1996; Clift et al., 2003). In the latter case, extension of
the Ilan Basin has occurred because of the reversal of subduc-
tion polarity following arc-continent collision, which removes
the stresses that support the high topography in central Taiwan.
The earliest phases of extension form the Ilan Plain Basin. The
Okinawa Trough represents the continuation of this extension
to form a deep marine basin offshore. Without the tectonic push
from the Philippine Sea plate the orogen is able to extend because
of the free edge provided by the new trench. Specifically, there is
no need for slab break-off to allow polarity reversal or as a trigger
of orogenic extension.
REGIONAL GEOLOGY
The Ilan Basin forms the onshore western limit of the Oki-
nawa Trough and lies in northeast Taiwan (Figs. 1 and 2). The
basin is supplied with sediment from the southwest via the Lan-
yang River. This river terminates in a small delta, facing into the
deeper waters of the Okinawa Trough. To the north the basin is
bounded by exposures of shale and sandstone of the Meichi, Sze-
leng, and Chiayang Formations, which are dated as Oligocene,
and typically interpreted to be part of the passive margin of China
(Suppe, 1981; Teng, 1990; Lundberg et al., 1997), as well as minor
amounts of the underlying Eocene Tachien Sandstone, which
represents a synrift deposit, predating Oligocene seafloor spread-
ing (Fig. 3). The whole sedimentary sequence was deformed and
thrust to the northwest during the Taiwan orogeny, forming the

East China Shelf

Figure 3 h
Tr oug
wa
Figure 2 Okina
rc
kyu A
Ryu
it
ra
St
an Figure 1. Regional bathymetric map of
iw
Ta the Taiwan region, showing the collision
ch between the Luzon Arc and the passive
Ryukyu Tren margin of southern China. The Okinawa
Trough is in active extension to the east
Gagua Ridge

of Taiwan, and its western end comes

onshore in the Ilan Plain. Dashed white
line marks the base of the continental
slope in the South China Sea.
Philippine Sea
h
ug

Arc

Relative plate
Tro
h

motion,7 cm/yr
nc

on
n
e

zo

Luz
Tr

Lu
ila

South China Sea

an

th
r
M

No

Luzon
260
an
hsh
üe
Hs
Well TL
Ilan
Lanyang River
Well CH
ult
Fa
n
ha
Lis
Backbone Range
modern Hsüehshan Range. Peak metamorphic grades are low
(e.g., Chen, 1984). Liu et al. (2001) used zircon fission-track data
to show that these units had not been buried beyond the partial
annealing temperature for that mineral (~200 °C; Tagami et al.,
1998) and inferred a lower greenschist metamorphic facies.
The southern edge of the basin is marked by outcrops of the
shale-rich Miocene Lushan Formation, which Liu et al. (2001)
considered to have reached prehnite-pumpellyite facies, and
thus is of a slightly lower grade than the Hsüehshan Range. The
Lushan Formation makes up much of the Backbone Range and
unconformably overlies the Tananao Schist locally (Suppe et al.,
1976), which is formed largely of Paleozoic and Mesozoic sedi-
mentary rocks, metamorphosed to greenschist grade. The old-
est sedimentary rocks overlying the Tananao Schist are Eocene
conglomerates of the Pilushan Formation, but the onlap is time
transgressive between basin and highs on the paleo-Chinese
margin. Localized slices of higher grade Tananao metamor-
phic rocks, including mafic amphibolites, are well exposed on
the coast at Tungao, south of Suao (Chen and Jahn, 1998). The
Tananao Schist is dated to have been metamorphosed during the
Clift et al.
1.2 ± 0.5 Ma
Well TC
Figure 2. Shaded topographic map of
the Ilan Plain area, showing the location
Philippine Sea of the drill sites considered in this study,
the position of apatite fission-track sam-
ples (with average age of the dominant
population), and the orientation of topo-
graphic profiles shown in Figure 5. Map
is from NASA Shuttle Radar data, made
available through www.geomapapp.org.
Suao
Tungao
3.5 ± 0.5 Ma
Mesozoic and subsequently overprinted by greenschist condi-
tions in the Pliocene (Lo and Yu, 1996; Wang et al., 1998).
Southwest of the Ilan Basin the Lushan and Tatungshan For-
mations are juxtaposed across the Lishan Fault. In central Taiwan
the Lishan Fault is manifested as a major, steep, NW-dipping
thrust structure, having an ESE-vergent geometry, interpreted by
Clark et al. (1993) as an antithetic backthrust relative to the domi-
nant northwest-vergent thrust stack in the Central Ranges (Wu,
1978). Changes in stratigraphic thickness across the fault suggest
that it is a reactivated structure that originally formed within an
extensional rift structure on the South China Sea passive margin
(Teng et al., 1991).
NEOTECTONICS OF THE LISHAN FAULT
Although the Lishan Fault is a NW-dipping thrust in central
Taiwan, it can be readily traced northward to the SW apex of
the Ilan Plain. In this region the fault is overturned and has an
extensional sense of motion at the western limit of the Ilan Basin,
meaning that the slip sense remains the same. The Lishan Fault is
 120 121 122

Pleistocene
volcanic rock Miocene

Taipei Basin
25
Pliocene-
Pleistocene Ilan
h an
e hs Plain
Paleogene
Hs

nge
Ra
Recent

ne
Alluvium

bo
ck
Longitudinal

Ba
Valley 24
(accreted Luzon
Forearc)

Lishan
Fault
Accreted
Chinese Line of section
margin
Coastal
Range 23
(accreted
Luzon Arc)

Paleozoic-
Mesozoic

22

Lishan Tananao Subduction

Fault Complex
Hsüehshan Backbone
Range Longitudinal
Miocene Valley Fault Coastal
Plio-Pleistocene Range

Figure 3. Simplified geological map of Taiwan, showing the map tectonic units that constitute the island and the location
of the Ilan Basin along the strike of the Lishan Fault, which separates the Hsüehshan and Backbone Ranges. The lower
cross section is modified from Lee et al. (2002) and shows the large-scale structure of the thrust belt in central Taiwan and
the geometry of the Lishan Fault, which subsequently has become the dominant structure of the Ilan Plain Basin.
262 Clift et al.
such a major tectonic division that it can be readily traced to pass
from being a thrust in central Taiwan to an extensional fault in
the Ilan Plain (Teng and Lee, 1996). In the western Ilan Basin the
Lishan Fault forms a normal fault of relatively high angle (>60°)
that separates the Hsüehshan Range from the Pleistocene basin
fill (Teng and Lee, 1996). The topographic break along the north-
ern edge of the basin is commonly sharp and suggestive of an
active neotectonic structure (Fig. 4A; Shyu et al., 2005). Expo-
sures of the sedimentary rocks that form the Hsüehshan Range
B relatively fresh exposure of sandstone
are relatively fresh and prominent at the northern edge of the
basin along much of its length. Figure 4B shows an example of
well-bedded, relatively unweathered sedimentary rocks, exposed
close to the trace of the fault near the southwestern end of the
basin. In this area the southern boundary of the basin is also sharp
and consistent with neotectonic faulting, as shown in map view
(Fig. 2) and in topographic profiles drawn across the strike of the
basin (Fig. 5). However, the shaded topographic map (Fig. 2) also
shows that the southern basin margin is less linear and poorly
Figure 4. Field photographs. (A) Steep
escarpment along the northeast edge
of the Ilan Plain, testifying to the fault-
bounded nature of the boundary. (B) A
and shale exhumed by the Lishan Fault
at the western end of the Ilan Plain, tes-
tifying to the recent activity of the fault
along the northern edge of the basin.
2m
 Migrating arc-continent collision in the Ilan Basin, Taiwan 263

Eastern transect
N S
800
Elevation (m)

600

400

200 Lishan Fault

0
0 10 20 30 40
1400 Figure 5. Topographic profiles across
Western transect the Ilan Plain Basin, showing the asym-
metry of the basin margins, especially
1200 in the east, where the northern, fault-
N S controlled margin rises rapidly owing to
1000 unloading along the Lishan Fault.
Elevation (m)

800
Lishan Fault
600

400

200

0
0 10 20 30 40
Distance (km)

defined in comparison with the northern boundary. Exposures
along the southern basin margin are generally not fresh but show
strong deep weathering of the basement, inconsistent with recent
tectonic exhumation. The eastern part of the plain in particular
shows a strong topographic contrast between a gently sloping
southern margin and a steep northern margin (Fig. 5).
This general contrast between the northern and southern
expressions of the Lishan Fault suggests that the northern bound-
ary is mostly fault controlled. However, the relationship is less
clear in the south and is more consistent with erosion of a tilted
hanging block rather than an uplifted footwall block. Although
Shyu et al. (2005) proposed a neotectonic fault along the east-
ernmost part of the southern boundary, it is noteworthy that this
feature has little topographic expression, suggesting that motion
on this structure is rather less than on the Lishan Fault. Sharp
increases in topographic gradient within the Backbone Range
south of the basin boundary nonetheless indicate that faulting is
active in this region.
Although the eastern and western ends of the northern margin
of the Ilan Basin are well defined, and apparently fault-controlled,
these trends do not align with one another but instead are separated
by a section ~10 km wide with less topographic relief (Fig. 2). We
infer a relay section in the Lishan Fault extension in this area,
where motion is transferred laterally. As is often the case in fault
relay zones in rifts, this area is marked by rivers that enter the basin
laterally, adding to the sediment provided by the axial Lanyang
River (e.g., Leeder and Gawthorpe, 1987).
DRILLING CONSTRAINTS ON NEOTECTONISM
Drilling by the Central Geological Survey of Taiwan has
allowed the basement contact along the faulted northern edge of
the Ilan Basin to be determined at depth as well as at outcrop.
Cores from three sites were examined in this project, wells TC,
TL, and CH, trending east to west close to the northern edge of
the basin (Fig. 2). These wells intersect the basement at depths of
17, 84, and 38 m, respectively, suggesting an average dip of the
basement-cover interface under the northern edge of only ~10°,
although this is over a short distance and may not be representa-
tive of the dip over the whole basin width. The overlying sedi-
ment comprises sequences of alluvial, coastal, and some shallow
marine sediment that has been dated as late Pleistocene to Holo-
cene by AMS 14C methods (Lai and Hsieh, 2003). The sediment
recovered at well CH includes fine-grained muds, silts, and fine
sands but is dominated by thick-bedded debris-flow gravels, with
clasts up to 10 cm across (Fig. 6). Well TL recovered the finest
West East
Well CH Well TL Well TC

M. Sand
M. Sand
M. Sand

C. Sand
C. Sand
C. Sand

F. Sand
F. Sand
F. Sand

Cong.
Cong.
Cong.

Mud
Mud
Mud

Silt
Silt
Silt
0 20 0

10 30 10

20 40 20

Basement
30 50 30
Depth (m)

40 60 40
Basement

50 70

60 80
Basement

70 90

80 100
Figure 6. Sedimentary logs showing the cored sequences found along the northern edge of the Ilan Plain, where Pleisto-
cene sediment is deposited over low-grade metamorphosed sedimentary rock, commonly below a zone of fault breccia
that marks a detachment surface.
 Migrating arc-continent collision in the Ilan Basin, Taiwan
grained sand of the three drilled sequences, though the sediment
is still characterized by coarse sands and gravels. Well TC recov-
ered a sequence of medium- to thick-bedded sands and muds.
In each location the sediment is underlain by well-lithified, low-
grade metamorphosed shale, typical of that seen in outcrop in the
Hsüehshan Range.
All drilled basement samples are distinguished by the pres-
ence of discrete brecciated layers, generally 2–5 m thick at the
basement contact but also present within the basement below that
level. These breccias are monomict and comprise only fragments
of metamorphosed shale (Fig. 7). The lack of material in the
breccias from the shallower basin fill sediment probably reflects
the fact that they are loose, unconsolidated materials and unable
to support the confining stresses needed to be incorporated into
the gouge. Texturally the breccias are generally very angular,
unsorted, and matrix supported. The breccias are relatively well
lithified in comparison with the overlying soft Pleistocene sedi-
ment. We interpret these rocks to represent fault breccias formed
by cataclasis in the shallow subsurface of the Lishan Fault. The
brittle, angular, broken character of the fault-breccia clasts is
apparent in thin section. Critically, the fault breccias postdate the
development of cleavage in the metapelites that form the footwall
(Fig. 8). As a result, motion on the fault is constrained to being
post–peak metamorphism and relatively shallow (<8 km) in the
crust. The breccias are preferentially formed close to the basement
contact, because this is the fault plane exposed to the surface by
the exhumation of the footwall block. Not all the breccia and fault
layers lie exactly along the basement-sediment contact, indicat-
ing that the Lishan Fault has been active along several fault splays
within the footwall block during uplift of the Hsüehshan Range.
The drill data confirm the presence of a major southeast-dipping
structure that controls the basin’s northern margin (Fig. 9).
A B
5 cm
265
Because the sediment cored in the Ilan Basin is all close
to sea level and preserves evidence of the environment of its
sedimentation, these materials can be used to determine rates of
tectonic subsidence across the plain. In the Ilan Basin the fast
rate of the sediment supply, largely from the Lanyang River,
appears generally to match rates of basement subsidence, keep-
ing the basin full while allowing shoreline progradation toward
the Okinawa Trough. Lai and Hsieh (2003) demonstrated that
subsidence reaches a maximum rate of ~19 mm/yr around the
mouth of the Lanyang River and decreases toward the edges
of the plain, both north and south, as well as toward the west.
However, what is important toward understanding the tectonics
of the Ilan Basin is the recognition that the northern part of the
plain subsides faster than the southern sector, reaching average
Holocene rates of 6–7 mm/yr versus 3–5 mm/yr. This disparity
is consistent with the observations presented above that indicate
a dominant asymmetric character to the basin structure. We thus
favor basin formation above a dominant south-dipping exten-
sional detachment (Fig. 9).
MODELING BASIN GEOMETRY
The large-scale structure of the basin can be considered and
modeled using a theoretical forward of an asymmetric basin, con-
trolled by a dominant detachment. We employed the flexural can-
tilever model of Kusznir et al. (1991) to forward model the defor-
mation and subsidence that would be expected to result from
the extension measured across the normal fault that bounds the
northern edge of the basin, and to test the idea that the regional
dip is 10°, as inferred from the drilling. Although the flexural
cantilever model is not the only, or universally accepted, theo-
retical model for extensional basins, it does have a number of
Figure 7. Photographs of cored pre-
Pleistocene rocks from well CH in the
western Ilan Basin. (A) Fault-brecciated
zone at 75.78 m depth. (B) Coherent
block of metapelite from 84.50 m depth.
266 Clift et al.

A B

Figure 8. Photomicrographs of thin sec-

tions cut from basement rocks in well
CH. (A) Typical fine-grained charac-
ter of the basement metapelites, with a
well-developed cleavage. (B) Fault rock
from within one of the breccia zones,
demonstrating low-temperature shatter-
ing textures.

250 µm 500 µm

Distance (km)
0 10 20 30 40
1000
N S
500
Figure 9. Schematic cross section of the
Elevation (m)

0 regional basin structure through the east-

ern Ilan Basin, showing the dominant
control of a south-dipping detachment
500 in controlling uplift of the Hsüehshan
Range and preferential subsidence of
the northern plain. Shading patterns are
1000 not intended to accurately portray de-
formation structures in the footwall and
1500
hanging-wall blocks.
Lishan Fault

2000
Lushan Formation (Miocene)
Pleistocene to Recent sediments
meta-sedimentary rocks
Paleogene low grade
Mesozoic subduction complex meta-sedimentary rocks

important features that allow the basin tectonics to be understood
at a first-order level, and it has been used to effect in understand-
ing basins in several parts of the world (Kusznir et al., 1995; Rob-
erts et al., 1993).
The model is complicated by our lack of subsurface data and
constraint on the degree of extension across the fault, although
this is also limited by the total size of the basin. For modeling
purposes we chose a default of 5 km horizontal extension and a
crustal thickness of 40 km. In this approach we made no attempt
to replicate the basin morphology but simply extended a model
continental lithosphere using the major detachment fault and
predicted what sort of basin this would form. Flexural rigidity
is expected to be low in an arc orogenic setting. Effective elas-
tic thickness (Te) is only 13 km in the Taiwan foreland (Lin and
Watts, 2002) and is expected to be rather less in the orogenic
core that we consider here. The deformation in the flexural can-
tilever model assumes brittle faulting in the upper crust, set at
10 km thickness here, and ductile deformation distributed in a
sinusoidal fashion over a wavelength of 100 km. The results of
the models are shown in Figure 10.
 Migrating arc-continent collision in the Ilan Basin, Taiwan 267

Figure 10. Proposed basin geometries and comparisons with possible model basins in order to limit the possible faulting
geometries and strength of the lithosphere under the Ilan Plain. (A) Proposed basin cross section inferred from outcrop,
coring, and gravity constraints. (B–D) Forward models, made using the flexural cantilever model of Kusznir et al. (1991)
and a 30° detachment fault, with variations in the flexural rigidity of the lithosphere. (E–G) Forward models using a 10°
detachment fault, showing the effects of varying amounts of extension and flexural rigidity. (H) Forward models using a
45° detachment fault, showing a poor match to observed basin shape. Te—elastic thickness.

An important result of our forward models is that basins
generated above a 10° detachment surface do not produce real-
istic basin geometries (Fig. 10A, E–G), mostly because they
do not predict significant footwall uplift of the size seen in the
Hsüehshan. Also, 10° detachments fail to produce footwall
uplift even when Te is increased to 10 km (Fig. 10A, G), or if
the extension across the fault is increased to 10 km. We conclude
that the 10° dip inferred from coring cannot be representative of
the basin. The best first-order fit to basin geometry was acquired
using a 30° detachment, with a 5 km extension and a Te of 3 km
(Fig. 10A, B). Although this is not a unique solution, it does give
some guidance as to the type of fault and mechanical state of
the crust under the Ilan Plain. Changing flexural rigidity to either
1 km or 5 km changed the basin width to less close fits (Fig. 10A,
C, D). The modeling is also effective at eliminating very high
angle faults as possible mechanisms, because even a 45° fault
produces too much footwall uplift, hanging-wall subsidence, and
a wide basin (Fig. 10A, H).
268
FISSION-TRACK ANALYSIS
The exhumation history of the ranges around the Ilan Basin
can be examined using low-temperature thermochronometry.
Although previous studies have used fission-track methods in the
central parts of Taiwan (e.g., Liu et al., 2000, 2001; Willett et
al., 2003), no similar studies around the Ilan Basin have been
undertaken. Liu et al. (2001) used zircon fission-track methods to
date the cooling history and included a transect north of the basin,
across the Hsüehshan Range. Fission-track studies in zircon grains
record cooling through a partial annealing zone of 200–320 °C
(Tagami et al., 1998), and so this method is sensitive to exhu-
mation driven by erosion and has been widely used in orogenic
exhumation studies. However, Liu et al. (2001) recorded zircon
fission tracks north of the Ilan Basin that are mostly older than
the Taiwan orogeny. Central ages are as low as 22 Ma close to the
topographic front of the Hsüehshan Range but become older far-
ther north, consistent with more exhumation at the margin of the
basin (Liu et al., 2001). That the zircon fission tracks are this old
indicates that they have been partially but not fully reset by late
Miocene–Holocene collision-related burial. A lower temperature
thermochronometer is thus required to reconstruct the cooling of
rock units in shallower parts of the crust.
In this study we employed the fission-track method applied
to apatite, which records cooling through ~125–60 °C over time
scales of 1–10 m.y. (Green et al., 1989). Apatite fission-track
analysis was performed at University College, London, UK.
Polished grain mounts were etched with 5N HNO3 at 20 °C for
20 s to reveal the spontaneous fission tracks. Subsequently, the
uranium content of each crystal was determined by irradiation,
which induced fission of 235U. The induced tracks were registered
in external mica detectors. The samples for this study were irradi-
ated in the thermal facility of the Hifar Reactor at Lucas Heights,
Australia. The neutron flux was monitored by including Corning
glass dosimeter CN-5, with a known uranium content of 11 ppm,
at either end of the sample stack. After irradiation, sample and
dosimeter mica detectors were etched in 48% HF at 20 °C for
45 min. Only crystals with sections parallel to the c-axis were
counted, as these crystals have the lowest bulk etch rate. To avoid
biasing results through preferred selection of apatite crystals, the
samples were systematically scanned, and each crystal encoun-
tered with the correct orientation was analyzed, irrespective of
track density. The results of the fission-track analysis are pre-
sented in Table 1.
TABLE 1. FISSION-TRACK ANALYTICAL DATA
Sample no./ Mineral No. Dosimeter Spontaneous
ρd ρs
field no. of Nd Ns
crystals
TW114-1 Apatite 30 1.387 3845 0.185
TW116-3 Apatite 15 1.387 3845 0.009
6 –2
Note: (i) Track densities are (×10 tr cm ) numbers of tracks counted (N) shown in parentheses; (ii) analyses by external detector method using 0.5 for the 4π/2π geometric
correction factor; (iii) ages calculated using dosimeter glass CN-5; (apatite) ξCN5 = 338 ± 4; CN-2; (zircon) ξCN2 = 127 ± 5 calibrated by multiple analyses of IUGS apatite and zircon
age standards (see Hurford, 1990); (iv) Pχ is probability for obtaining χ value for v degrees of freedom, where v = no. crystals – 1; (v) central age is a modal age, weighted for
2 2
different precisions of individual crystals (see Galbraith, 1990).
Clift et al.
Of 12 samples analyzed, only 2 yielded sufficient apatite
to produce meaningful results. Fortunately these samples came
from opposite sides of the basin (Fig. 2) and allow the exhuma-
tion histories of the margins to be compared. Central ages of 1.2
± 0.5 Ma and 3.5 ± 0.5 Ma were recorded at the coast on the
north and south basin margins, respectively (Fig. 11). The south-
ern sample contains a minor number of older grains (older than
150 Ma), reflecting an earlier phase of cooling. However, for this
study we focus on most of the population, which records cooling
linked only to Pliocene–Holocene tectonism. The single, young
grain population seen on the northern margin indicates total reset-
ting and recent, rapid cooling. Exactly why the southern margin
sample contains a few grains that are not reset at 3.5 Ma is not
clear. However, the vast majority of grains in that sample show a
well-defined 3.5 Ma cooling trend (Fig. 11). The spread of ages
would be greater if only partial annealing had occurred. We con-
clude that the sample underwent rapid cooling through the AFTA
annealing temperature zone ca. 3.5 Ma. Our result indicates more
recent cooling of the north margin in comparison with the south,
as might be expected for an asymmetric basin controlled by a
south-dipping detachment. What is surprising is that this result is
the opposite of Liu et al.’s (2001) zircon fission-track result that
has central ages of only 1.5 ± 0.3 Ma in the south close to Suao
and 22.2 ± 2.3 Ma along the northern margin, which would imply
the opposite sense of motion. Such motion is not consistent with
the structural asymmetry presented above. Although the apatite
and zircon results are consistent in the north, it is impossible to
have younger fission-track ages for zircon than for apatite, call-
ing into question whether the zircon or apatite data from Suao are
representative of the region. Assuming a geothermal gradient of
30 °C/km (Barr and Dahlen, 1988; Willett et al., 2003), the apa-
tite data imply average exhumation rates of 1.6 ± 0.3 km/m.y. in
the north and 0.7 ± 0.1 km/m.y. in the south. Because we demon-
strate the importance of extensional motion on the Lishan Fault
in generating uplift of the Hsüehshan Range, much of the exhu-
mation on the north margin may be linked to tectonic unroofing
by detachment faulting rather than by erosion.
SEISMIC EVIDENCE
Observation of recent earthquakes is important to under-
standing the neotectonic evolution of the Ilan Basin, as they
provide a measure of where faulting is active and a snapshot of
the current stress regime. The seismicity used here is a subset of
Induced Age dispersion Central age Age components
ρi
2
Ni Pχ RE% (Ma) mode 1 mode 2 mode 3
±1 (number of grains)
141 3.639 2952 0.0 260.4 14.4 ± 7.0 3.5 ± 0.5 22 ± 12 294 ± 78
(26) (2) (2)
13 1.666 2116 19.8 97.0 1.2 ± 0.5 Single pop
 Migrating arc-continent collision in the Ilan Basin, Taiwan 269

400 However, a coherent array of deep earthquakes is not seen under

TW114-1 300
Central age: 14.4 ± 7 Ma 200 the western Okinawa Trough in the 0–50 km depth range consid-
P(χ2): 0.0% 150 ered here. Our deeper profile (Fig. 12) indicates that the Philip-
Relative error: 256%
100 pine oceanic slab is deeper in this location and that therefore the
Number of crystals: 30 array of seismic events seen in Figure 14A–C reflects a separate
lithospheric structure. The NE-trending seismic zone in the cen-
tral part of the Ilan Plain is populated by shallow normal-faulting
events with NW-SE directed T-axes (extension). In the southeast
basin, shallow seismicity appears to be arranged into two shallow
+2 10 E-W–trending zones dominated by sinistral strike-slip events and
0 N-S–oriented normal faulting, but with T-axes in the same general
-2
direction as in the basin center. The two areas of intense seismic-
3.5 ± 0.5 Ma ity are not that different in that both show a shallow layer of seis-
% Relative error
63 10 0 micity between a narrow range of 8–13 km. Analysis of the map
0 and sections in Figures 13 and 14 reveals the subtle differences
10 20 30 40 in seismicity associated with different belts. In the first place, the
Precision (1/sigma)
shallow events are concentrated in a fairly narrow depth range.
TW116-3 This is curious; one possible explanation is that the seismic layer
Central age: 1.2 ± 0.5 Ma 4 is limited on its upper surface by the presence of relatively incom-
P(χ2): 19.8% petent and therefore nonbrittle sediment. In contrast, the lower
Relative error: 97%
limit of seismic activity may be bounded by hot middle crust at
Number of crystals: 15
~10 km depth, where ductile deformation precludes seismogen-
+2 3 esis. Sections B and E (Fig. 14) show the depth distribution of the
2 E-W–trending strike-slip events around 24°34′N (southern end
of the profiles). The NE-SW–trending belt does show well in the
0 1 cross section, with only a few events >30 km in Figure 14B, C.
No evidence exists for a shallow-dipping active slip plane bound-
ing the basin to the north on the basis of seismicity. Whatever fault
-2 motion we can confirm is either diffused or concentrated in a nar-
0 row depth range, probably between the bottom of the sedimentary
% Relative error
53 8 layer and the top of the basement, judging from the rheological
0 properties of these materials.
10 20 30 40 50
Precision (1/sigma)
The events >50 km deep under the basin axis are interpreted
as thrust earthquakes, consistent with their association with the
Figure 11. Radial plots of apatite fission-track analyses from Samples Ryukyu subduction zone, which is now starting under northern
TW114-1 and TW116-3, respectively, from the southern and northern Taiwan by lateral motion of the Philippine Sea plate into a tear
margins of the Ilan Basin. Locations shown in Figure 2.
in the passive margin lithosphere of southern China (Lallemand
et al., 2001). However, analysis of earthquakes shallower than
20 km from the nearby Okinawa Trough shows a dominant
extensional character, with extension perpendicular to the strike
events initially reported by the Central Weather Bureau of Tai- of the trough, i.e., NW-SE. These shallow earthquakes form an
wan. The phase data for these events are used for relocation by offshore continuation of the shallow events seen under the Ilan
means of the double-difference method of Waldhauser and Ells- Basin, which are similarly interpreted as being extensional. No
worth (2000). The relocation procedure minimizes the errors in strong evidence exists from earthquakes to support the presence
hypocentral determination because of lateral velocity variations. of a currently active major south-dipping detachment (e.g., the
The resulting events tend to cluster around known structures, as Lishan Fault). However, if the major extensional fault is shal-
shown by Waldhauser (2001). low dipping, then it is too shallow to store elastic stresses and to
The Benioff zone in this area is below 50 km and strikes nearly be seismogenic under the northern part of the Ilan Plain. Deeper
E-W and dips at ~30° to the north (Fig. 12; Wu et al., 1997). In the faulting under the southern edge of the basin is consistent with
Ilan area the deeper seismicity (~50 km) is concentrated along the these asymmetric structural models. Strike-slip faulting in the
southwestern margin of the basin, as well as more generally across southeast basin may be interpreted as part of the strain accom-
the northern half of the basin. The seismicity in the upper 50 km modation related to the bend in the tectonic fabric from a N-S ori-
is shown in Figure 13. Cross sections through the basin show that entation in central Taiwan to an E-W orientation in the Ryukyu
the deeper earthquakes form a NW-dipping array (Fig. 14A–C). Arc. The strain observed indicates motion of the southern margin
270 Clift et al.
S N
Elevation
3
(km)
0
0
Depth (km)
50
100
0 50 100 150
Distance (km)
of the Ilan Basin to the ESE. The sense of motion is the opposite
of that expected from the bending of the structural fabric, but it
is consistent with the lateral motion of orogenic crust away from
the Taiwan mountains toward the Ryukyu Trench, as might be
predicted for orogenic collapse driven by a change in stresses
triggered by generation of that trench.
GEODYNAMIC EVOLUTION
Additional constraints on the nature of current tectonic strain
are provided by global positioning system (GPS) monitoring of
the region. Here we consider the motions reconstructed by Chang
et al. (2003), as shown in Figure 15. Relative to stable southern
China the ranges north of the Ilan Basin are moving slowly to
the northwest (~10–15 mm/yr), effectively a continuation of the
west-directed thrusting that characterizes most of Taiwan, albeit
slightly rotated clockwise in this particular area. In contrast, GPS
locations in the Ilan Basin itself are moving almost due east,
while the basement around Suao on the southern boundary is
being displaced toward the southeast at 15–20 mm/yr. The net
result indicates motion of the Hsüehshan and Backbone Ranges
away from one another, consistent with the seismic evidence
for extension across the basin. The eastward motion of the Ilan
Basin shows the same sense of motion as the strike-slip fault
data derived from the fault plane solutions of the seismic data.
It should be noted that the major strike-slip structures (Fig. 11)
are south of the GPS stations on the Ilan Plain. The GPS data
thus indicate a broader zone of crustal extrusion than that based
121˚ 122˚
25˚ 25˚
Figure 12. Cross section off the east
coast of Taiwan, showing recent seis-
micity down to 150 km depth. The loca-
tion of the section is shown as a black
line on the map (right). Seismicity di-
vides broadly into the shallow seismic-
ity seen in detail in Figures 13 and 14,
and the well-defined north-dipping Be-
24˚ 24˚ nioff zone of the Ryukyu Arc.
121˚ 122˚
on the seismology alone, implying extrusion of this crustal block
toward the Okinawa Trough. We do recognize the importance of
the strike-slip zone resolved by the seismic data and note that
GPS motions are more south-oriented, directly toward the trench,
south of that lineament.
The GPS motion data can be understood in the context
of the tectonic setting that is characterized by E-W compres-
sion in central Taiwan, contrasting with N-S subduction to the
east along the Ryukyu Trench. In practice the Ryukyu Trench
provides a free edge, allowing the southward motion of crustal
blocks in the Ryukyu forearc, compared with the compression
by the Luzon-China collision (Fig. 16). If Taiwan is considered
an arc-continent collision orogen that is progressively migrat-
ing along the Chinese margin toward the southwest, then the
passive-margin units that are compressed and uplifted by col-
lision may then collapse and extend, once the restraining but-
tress of the Philippine Sea plate has been removed and sub-
duction with the opposite polarity is initiated. The load of the
Taiwan orogen causes flexure of the underlying Chinese conti-
nental crust (Yu and Chou, 2001; Lin and Watts, 2002; Lin et
al., 2003). Where that load slides southeastward away from the
margin’s edge in northern Taiwan, the continental crust quickly
rebounds and regains much of its normal thickness (Rau and
Wu, 1995). Thus SE-directed motion of the basement south of
the Ilan Basin reflects collapse of the flank of the Taiwan orogen
and motion of the Backbone Range toward the Ryukyu Trench.
Extension in the Okinawa Trough, driven by southward motion
of the Ryukyu forearc toward the trench also provides space for
 121˚30’ 121˚36’ 121˚42’ 121˚48’ 121˚54’
25˚00’

D E F

?
?
24˚54’ ?
?

? ?
?
?
?
?
?
? ?
24˚48’

C ? ?
A B
?
? ?

?
24˚42’ ? ? ? ?
? ? ?
?? ?
?
? ?
?
?
?
?
?
24˚36’ ?

? ?
? ?
?

km ?
?

0 5 10
24˚30’

0 50 100 150
Hypocentral Depth (km)
Figure 13. Map showing the location of earthquake hypocenters around the Ilan Basin, with the dominant extensional
character of recent events at depths <20 km, whereas deeper events are thrust motions linked to the Ryukyu subduction
zone and the Taiwan collision. Parallel lines A, B, C and D, E, F denote cross sections shown in Figure 14.
272 Clift et al.

SE NW SE NW SE NW
Depth (km) 4 A B C
0
-4
0
Altitude (km)

25

50
0 25 50 0 25 50 0 25 50
Distance (km) Distance (km) Distance (km)
S N S N S N
4
Depth (km)

D E F
0
-4
0
Altitude (km)

25

50
0 25 50 0 25 50 0 25 50
Distance (km) Distance (km) Distance (km)

Figure 14. Cross sections down to 50 km depth across the basin show concentrated seismicity ~13–8 km depth. Those in
the 13–50 km range are evidently associated with collision of the Philippine Sea and Eurasian plates, and the subduction
events are at depths >50 km in this region. Section locations are shown in Figure 13.

the eastward motion of material in the Ilan Basin, driven by the
gravitational potential of the thickened crust in central Taiwan.
This motion is partially accommodated by strike-slip tectonism
as seen in the seismic data, as well as by extension, especially
that focused on the major fault bounding the Ilan Plain to the
north, inferred to be the Lishan Fault, reversed in direction in
comparison with its thrust sense in the Central Ranges.
DISCUSSION
The fact that the Lishan Fault can be traced along strike from
a major thrust fault into an extensional detachment is consistent
with the extension in the Ilan Basin being the product of post-
orogenic collapse, triggered by subduction polarity reversal, as
argued by Teng (1996), and not the result of extension propa-
gating from the Okinawa Trough, driven by trench forces. The
extensional fault that controls the Ilan Basin does not represent
a westward propagating rift of oceanic origin cutting across the
older orogenic fabric, but rather a reversal of motion on thrusts
as compressional stresses are released toward the northern end of
the orogen. The major bounding fault can be clearly traced as an
extension of the Lishan Fault, which is a major thrust structure
in the Central Ranges but is overturned and moving as an exten-
sional structure where it reaches the Ilan Plain. The initiation of
 Figure 15. Map showing the motions of crustal
blocks around the Ilan Basin relative to stable
Eurasia, recorded by Chang et al. (2003) using
Ilan Basin GPS methods over the period 1990–1995.

Velocity 50 mm/yr

South of Taiwan
Chinese passive Accretionary North Luzon Luzon
Active oceanic subduction margin Complex Trough Arc
NW SE
sealevel

oceanic crust Arc lower crust

continental crust and mantle
Lishan Fault
thrust
Central Taiwan
Compression and
metamorphism sealevel
Figure 16. Schematic depiction of the origin of
the Ilan Basin as a result of gravitational col-
lapse of the Taiwan Central Ranges during a
SW-migrating collision of the Luzon Arc and
mainland China.

Northeastern Taiwan Lishan Fault

Exhumation and Taipei Basin detachment
formation of Ilan Basin
Ilan Basin
sealevel
274 Clift et al.
the Ryukyu Trench removes the compressive force of the Philip-
pine Sea plate from the orogen and allows the edifice to extend
as the trench provides a free edge toward which material can be
displaced. In turn this suggests that at least the southwestern end
of the Okinawa Trough has also formed as a result of migrat-
ing arc-continent collision and subsequent orogenic collapse and
is not generated by subduction slab rollback (cf. Suppe, 1984).
This is an important revision of the generally held belief that the
Okinawa Trough is formed by trench forces, principally rollback
of the trench, much as suggested for the Mariana Trough or the
Lau Basin (e.g., Hawkins, 1974). However, in the rollback model
the presence of the westernmost propagating end of the rift at
the northern tip of Taiwan would be coincidental, whereas rift-
ing related to collision and subduction polarity reversal would
predict this position. In practice, little evidence exists for active
rift propagation. The Okinawa Trough is not V-shaped as seen in
other western Pacific basins, where a propagating rift culminates
in seafloor spreading. The Okinawa Trough appears to be open-
ing to the west, yet the older eastern parts of the basin are no
wider than that closest to Taiwan. In the Okinawa Trough, rifting
is followed by a cessation of extension, whereas in contrast roll-
back and arc rifting are normally followed by further extension in
the form of seafloor spreading.
Evidence that the westward migration of the Okinawa
Trough is matched by arc migration was provided by Shinjo
(1999), who noted that middle Miocene volcanic rocks from the
southern Ryukyu Arc are not subduction related but instead are
similar to intraplate volcanism seen in China. Such an observa-
tion implies that the Ryukyu Arc is new and that the subduction-
related Ryukyu volcanic front has propagated into the area since
that time. This hypothesis is also supported by the geochemistry
of recent volcanic rocks from the southernmost Okinawa Trough
(Chung et al., 2000). In a collapse model the Okinawa Trough
becomes increasingly younger to the south, the opposite of the
rollback model in which the Okinawa Trough spreading centers
are propagating into the basin away from Taiwan.
It is noteworthy that active magmatism and faulting that cuts
right to the seafloor in the Okinawa Trough is restricted to that
part of the basin closest to Taiwan (Sibuet et al., 1998). In contrast,
middle to late Miocene (6–9 Ma) extension ages are recorded in
the northern Okinawa Trough (Letouzey and Kimura, 1985). The
northern Okinawa Trough thus is either the remnant of an earlier
arc-continent collision, as favored by Clift et al. (2003), or has a
separate origin from that part of the basin closer to Taiwan. The
basement of the southern Okinawa Trough is inferred to be the
extended remnants of the Taiwan orogen, equivalent to the Back-
bone and Hsüehshan Ranges. The recognition of a continuous
Taiwan-Sinzi folded zone under the SE edge of the East China
Sea (Hsiao et al., 1999) would suggest a continuous migration of
the orogen from Sinzi at ca. 12 Ma to present-day Taiwan. Vol-
canism is the manifestation of the new arc volcanic front to the
Ryukyu subduction zone, which overlies the nonvolcanic forearc
ridge only in the central and northern Ryukyu Arc, where active
extension has ceased.
Here we propose that the Ilan Basin can best be under-
stood in the context of a migrating arc collision, and especially
as the culmination of gravitational collapse of the resultant oro-
gen, having taken place over relatively short periods of geologic
time (Fig. 17). Collision of the modern Taiwan section of the
Luzon Arc with China is thought to have begun ca. 6–9 Ma (e.g.,
Dorsey, 1988; Teng, 1990; Sibuet et al., 2002; Malavieille et al.,
2002; Huang et al., 2006) and is already finished and in a state
of rapid exhumation around the Ilan Basin. Byrne and Crespi
(1997) reported extension throughout the Backbone and Central
Ranges. The stronger extension seen around the Ilan Basin can
be understood as an extension of this, made possible as the new
Ryukyu Trench form, removing the compressive stresses of col-
lision and allowing the flank of the Taiwan orogen to move later-
ally into that space.
Figure 18 shows the relationship between tectonically driven
rock uplift rates and exhumation rates in Taiwan, assuming that
“hard collision” in northern Taiwan initiated ca. 5 Ma, whereas in
the south hard collision between the Luzon Arc and China is just
starting at the modern coast. Initial collision between the Luzon
forearc and the Chinese margin begins farther south, with the devel-
opment of an accretionary prism, which progressively overthrusts
the North Luzon Trough (forearc basin; Lundberg et al., 1997).
Regional trends in rock uplift rates can be determined from the cur-
rent elevations and the age of the collision, together with estimates
for the amount of exhumation derived from the metamorphic-
grade and fission-track data (e.g., Dadson et al., 2003). Although
in some areas modern rates of uplift have been determined by dat-
ing terraces (e.g., Lin, 1969; Peng et al., 1977; Vita-Finzi and Lin,
1998), these terraces are necessarily limited to the coastal regions,
mostly in the Coastal Ranges of eastern Taiwan.
Exhumation rates driven by erosion reach a peak in the
south of the island, because rates of rock uplift are highest dur-
ing the most intense period of collisional compression between
Luzon and China; these are partly balanced by erosion driven
largely by precipitation but also by tectonic extension (Crespi et
al., 1996; Teng et al., 2000). Exhumation and vertical uplift rates
decrease abruptly toward the northern end of the Central Ranges,
especially around the Ilan Basin, although active motion along a
detachment reversing the Lishan Fault causes increased exhuma-
tion in the Hsüehshan Range. The rates of burial and exhumation
are some of the highest known on Earth. Exhumation rates at
Nanga Parbat in the Pakistan Himalaya reach 5 mm/yr (Zeitler et
al., 1993), slightly less than the rates seen in the Central Ranges
(Dadson et al., 2003). Although exhumation rates are lower in
the Hsüehshan Range, reaching >1.6 mm/yr adjacent to the
Lishan Fault, these levels are still relatively high, far exceeding
rates of 0.33 mm/yr in the Cascade Range of the northwestern
United States (Reiners et al., 2003), somewhat more than the
~1 mm/yr found in the Alaska Range (Fitzgerald et al., 1993),
and comparable to the 2.5–0.8 mm/yr measured by Tippett and
Kamp (1993) in the southern Alps of New Zealand. Clearly arc-
continent collision forms some of the most dynamic geology on
the planet.
 Migrating arc-continent collision in the Ilan Basin, Taiwan
East China Sea
Si nzi
n-
wa
Tai
orogenic
collapse
collisional
orogen
collision
migration
early
collision
-5 m.y.
oceanic
arc
South China Sea
Luzon
CONCLUSIONS
A variety of geological and geophysical data demonstrates
active extension along a NNW-SSE axis across the Ilan Basin of
northern Taiwan. The structure of the basin appears to be largely
controlled by a SE-dipping detachment fault, likely dipping at
~30°, that causes uplift of the Hsüehshan Range to the north of the
basin and preferential fast subsidence of the northern Ilan Basin.
Apatite fission-track ages confirm geomorphic evidence for faster
exhumation of the northern margin of the basin, reaching rates of
at least 1.6 ± 0.3 km/m.y. The main basin-bounding extensional
fault is mapped as a continuation of the Lishan Fault, a major thrust
structure from central Taiwan. The southern Backbone Range is
shown by seismic and GPS data to be moving southeast toward
the newly formed western Ryukyu Trench. Formation of a free
edge in the trench and release of the E-W compressive stresses
allow the Taiwan orogen to collapse in the Ilan Basin–Okinawa
Trough region. Because the Taiwan orogen is migrating to the
southwest along the passive margin of southern China, we sug-
gest that the extension and associated basin formation must also
move in this direction. Although the Ilan Basin was formed by the
275
t
ldbel
Fo
h
Tr oug
wa
O kina
Arc Figure 17. Shaded bathymetric map of the
kyu
Ryu Taiwan region, showing the collisional oro-
gen, the opposing subduction polarities,
and the Okinawa Trough opening in the
wake of orogenic collapse. The numbered
lines adjacent to the plate boundary show
the inferred time of peak arc collision be-
+4 m.y.
tween Luzon and China. Map is labeled to
show the different stages of arc-continent
collision along strike. Wide white dashed
line shows location of the Taiwan-Sinzi
Fold Belt, interpreted as remnants of the
former collisional orogen. Black dashed
7 cm/yr line shows the location of the modern arc
volcanic front, focused by extension in the
Okinawa Trough close to Taiwan.
Philippine Sea
start of collapse onshore, the process reached its culmination off-
shore in the Okinawa Trough, indicating that at least some of this
basin has formed as a result of collision and not by trench rollback
forces as previously believed (cf. Suppe, 1984). We propose that
mountain building and rifting of the Ilan Plain and the Okinawa
Trough are all results of a single common process, arc-continent
collision, rather than being the unique interaction of a collisional
orogen and a propagating backarc rift. Like the Alboran Sea in the
western Mediterranean the Ilan-Okinawa Trough shows how far
postorogenic collapse may drive extension and basin formation
(e.g., Platt and Vissers, 1989), although in this case without the
need to invoke delamination of the mantle lithosphere.
Because arc-continent collision is a common process in the
plate-tectonic cycle we anticipate that collapse basins similar to
the Ilan Basin should be a common feature and should be recog-
nizable in ancient collision zones. Although the Ordovician South
Mayo Trough of the Irish Caledonides has been interpreted to be
such a basin (e.g., Clift et al., 2003), the scarcity of other exam-
ples likely reflects a lack of understanding of existing sequences
rather than the absence of gravitational collapse in ancient arc
collisional events.
276 Clift et al.

Date of collision (Ma)

5 4 3 2 1 0

Line
of se
ction

3000 3000
Okinawa Tr.

Ilan Basin
Altitude (m)

Altitude (m)
2000 2000

1000 1000 Figure 18. Summary of the trends in

rock uplift and exhumation rate along
the length of the island of Taiwan, de-
0 0
picted also as a time progression of
hard collision during initial arc-conti-
-1000 -1000 nent collision passing into extension,
20 12 exhumation, and collapse, culminating

Exhumation Rate (mm/yr.)

cum. exhumation in formation of the Okinawa Trough.
10 Exhumation rates are from Dadson et
Exhumation (km)

15 al. (2003). Gray line shows long-wave-

8 length topography.
Cumulative

10 6

4
5 exhumation rate
2

0
25
Rate of rock uplift (mm/yr.)

cum. uplift
Cumulative rock uplift (km)

20
10

15
5
10
0
5
rate of uplift
-5
0

-5 -10
400 300 200 100 0
Distance (km)

ACKNOWLEDGMENTS fieldwork from an NSC (Taiwan) grant of 93–2116-M-008–001.

F.T.W. acknowledges support from the U.S. National Science
P.C. thanks the University of Aberdeen, College of Physical Sci- Foundation’s Continental Dynamics Program. We thank Don
ences; the Alexander von Humboldt Foundation; and the Joint Fisher, Neil Lundberg, Dave Topping, and Patrick Barnard for
Oceanographic Institutions for contributions to the cost of this their helpful reviews, as well as editor Dave Scholl for his help
research. A.T.L. is grateful for financial support for conducting in improving this manuscript.
 Migrating arc-continent collision in the Ilan Basin, Taiwan
REFERENCES CITED
Barr, T.D., and Dahlen, F.A., 1988, Thermodynamic efficiency of brittle fric-
tional mountain building: Science, v. 242, p. 749–752, doi: 10.1126/sci-
ence.242.4879.749.
Byrne, T., and Crespi, J., 1997, Kinematics of the Taiwan arc-continent colli-
sion and implications for orogenic process: Tapei, Taiwan, International
Conference and Sino-American Symposium on Tectonics of East Asia,
p. 38.
Casey, J.F., and Dewey, J.F., 1984, Initiation of subduction zones along trans-
form and accreting plate boundaries, triple-junction evolution, and forearc
spreading centres—Implications for ophiolitic geology and obduction, in
Gass, I.G., et al., Ophiolites and Oceanic Lithosphere: Geological Society
[London] Special Publication 13, p. 269–290.
Chang, C.-P., Chang, T.-Y., Angelier, A., Kao, H., Lee, J.-C., and Yu, S.-B.,
2003, Strain and stress field in Taiwan oblique convergent system: Con-
straints from GPS observation and tectonic data: Earth and Planetary Sci-
ence Letters, v. 214, p. 115–127, doi: 10.1016/S0012-821X(03)00360-1.
Chemenda, A.I., Yang, R.K., Hsieh, C.H., and Groholsky, A.L., 1997, Evolu-
tionary model for the Taiwan collision based on physical modeling: Tecto-
nophysics, v. 274, p. 253–274, doi: 10.1016/S0040-1951(97)00025-5.
Chen, C.H., 1984, Determination of lower greenschist facies boundary by K-
mica-chlorite crystallinity in the Central Range, Taiwan: Proceedings of
the Geological Society of China, v. 27, p. 41–53.
Chen, J., and Jahn, B.M., 1998, Crustal evolution of southeastern China: Nd and
Sr isotopic evidence: Tectonophysics, v. 284, p. 101–133, doi: 10.1016/
S0040-1951(97)00186-8.
Chung, S.L., Wang, S.L., Shinjo, R., Lee, C.S., and Chen, C.H., 2000, Initia-
tion of arc magmatism in an embryonic continental rifting zone of the
southernmost part of Okinawa Trough: Terra Nova, v. 12, p. 225–230,
doi: 10.1046/j.1365-3121.2000.00298.x.
Clark, M.B., Fisher, D., and Lu, C.-Y., 1993, The Hsueshan Range of Taiwan: A
crustal scale pop-up structure: Tectonics, v. 12, p. 205–217.
Clift, P.D., and Vannucchi, P., 2004, Controls on tectonic accretion versus
erosion in subduction zones: Implications for the origin and recycling
of the continental crust: Reviews of Geophysics, v. 42, p. RG2001, doi:
10.1029/2003RG000127.
Clift, P.D., Schouten, H., and Draut, A.E., 2003, A general model of arc-conti-
nent collision and subduction polarity reversal from Taiwan and the Irish
Caledonides, in Larter, R.D., and Leat, P.T., eds., Intra-Oceanic Subduc-
tion Systems; Tectonic and Magmatic Processes: Geological Society
[London] Special Publication 219, p. 81–98.
Crespi, J., Chan, Y.C., and Swaim, M., 1996, Synorogenic extension and
exhumation of the Taiwan hinterland: Geology, v. 24, p. 247–250, doi:
10.1130/0091-7613(1996)024<0247:SEAEOT>2.3.CO;2.
Dadson, S., Hovius, N., Chen, H., Dade, W.B., Hsieh, M.L., Willett, S., Hu,
J.C., Horng, M.J., Chen, M.C., Stark, C.P., Lague, D., and Lin, J.C., 2003,
Links between erosion, runoff variability and seismicity in the Taiwan
orogen: Nature, v. 426, p. 648–651, doi: 10.1038/nature02150.
Davis, D.W., Sewell, R.J., and Campbell, S.D.G., 1997, U-Pb dating of Meso-
zoic igneous rocks from Hong Kong: Geological Society [London] Jour-
nal, v. 154, p. 1067–1076.
Dorsey, B.J., 1988, Provenance evolution and un-roofing history of modern
arc-continent collision: Evidence from petrography of Plio-Pleistocene
sandstones, eastern Taiwan: Journal of Sedimentary Petrology, v. 58,
p. 208–218.
Fitzgerald, P.G., Stump, E., and Redfield, T.F., 1993, Late Cenozoic uplift of
Denali and its relation to relative plate motion and fault morphology: Sci-
ence, v. 259, p. 497–499, doi: 10.1126/science.259.5094.497.
Galbraith, R.F., 1990, The radial plot: Graphical assessment of spread in ages:
Nuclear Tracks, v. 17, p. 207–214.
Green, P.F., Duddy, I.R., Laslett, G.M., Hegarty, K.A., Gleadow, A.J., and Lov-
ering, J., 1989, Thermal annealing of fission-tracks in apatite; 4, Quantita-
tive modeling techniques and extension to geological timescales: Chemi-
cal Geology, v. 79, p. 155–182.
Hawkins, J.W., 1974, Geology of the Lau Basin, a marginal sea behind the
Tonga arc, in Burk, C.A., and Drake, C.L., eds., The Geology of Conti-
nental Margins: New York, Springer-Verlag, p. 505–520.
Hsiao, L.-Y., Lin, K.-A., Huang, S.T., and Teng, L.S., 1999, Structural charac-
teristics of the southern Taiwan-Sinzi folded zone: Petroleum Geology of
Taiwan, v. 32, p. 133–153.
277
Hsu, S.-K., and Sibuet, J.-C., 1995, Is Taiwan the result of arc-continent or arc–
arc collision?: Earth and Planetary Science Letters, v. 136, p. 315–324,
doi: 10.1016/0012-821X(95)00190-N.
Huang, C.Y., Yuan, P.B., Lin, C.W., and Wang, T.K., 2000, Geodynamic pro-
cesses of Taiwan arc-continent collision and comparison with analogs in
Timor, Papua New Guinea, Urals and Corsica: Tectonophysics, v. 325,
p. 1–21, doi: 10.1016/S0040-1951(00)00128-1.
Huang, C.Y., Yuan, P.B., and Tsao, S.H., 2006, Temporal and spatial records of
active arc-continent collision in Taiwan: A synthesis: Geological Society
of America Bulletin, v. 118, p. 274–288, doi: 10.1130/B25527.1.
Hurford, A.J., 1990, Standardization of fission-track dating calibration: Recom-
mendation by the Fission-track Working Group of the IUGS Subcommis-
sion on Geochronology: Chemical Geology, v. 80, p. 177–178.
Konstantinovskaya, E.A., 1999, Geodynamics of island arc–continent collision
in the western Pacific margin: Geotectonics, v. 33, p. 15–34.
Kusznir, N.J., Marsden, G., and Egan, S.S., 1991, A flexural cantilever simple
shear/pure shear model of continental extension, in Roberts, A.M., et al.,
eds., The Geometry of Normal Faults: Geological Society [London] Spe-
cial Publication 56, p. 41–61.
Kusznir, N.J., Roberts, A.M., and Morley, C.K., 1995, Forward and reverse
modeling of rift basin formation, in Lambiase, J.J., ed., Hydrocarbon
Habitat in Rift Basins: Geological Society [London] Special Publication
80, p. 33–56.
Lai, T.H., and Hsieh, M.L., 2003, Late-Quaternary vertical rock-movement
rates of the coastal plains of Taiwan: Abstracts, Annual Meeting of Geo-
logical Society of China, p. 119.
Lallemand, S., Front, Y., Bijwaard, H., and Kao, H., 2001, New insights on
3-D plates interaction near Taiwan from tomography and tectonic
implications: Tectonophysics, v. 335, p. 229–253, doi: 10.1016/S0040-
1951(01)00071-3.
Lee, J.C., Chu, H.-T., Angelier, J., Chan, Y.-C., Hu, J.-C., Lu, C.-Y., and Rau,
R.-J., 2002, Geometry and structure of northern surface ruptures of the
1999 Mw = 7.6 Chi-Chi, Taiwan earthquake: Influence from inherited
fold belt structures: Journal of Structural Geology, v. 24, p. 173–192, doi:
10.1016/S0191-8141(01)00056-6.
Leeder, M.R., and Gawthorpe, R.L., 1987, Sedimentary models for extensional
tilt-block/half-graben basins, in Coward, M.P., et al., eds., Continental
Extensional Tectonics: Geological Society [London] Special Publication
28, p. 139–152.
Letouzey, J., and Kimura, M., 1985, Okinawa Trough genesis; structure and
evolution of a backarc basin developed in a continent: Marine and Petro-
leum Geology, v. 2, p. 111–130, doi: 10.1016/0264-8172(85)90002-9.
Lin, A.T., and Watts, A.B., 2002, Origin of the West Taiwan basin by orogenic
loading and flexure of a rifted continental margin: Journal of Geophysical
Research, v. 107, p. 2185, doi: 10.1029/2001JB000669.
Lin, A.T., Watts, A.B., and Hesselbo, S.P., 2003, Cenozoic stratigraphy and sub-
sidence history of the South China Seamargin in the Taiwan region: Basin
Research, v. 15, p. 453–478, doi: 10.1046/j.1365-2117.2003.00215.x.
Lin, C.C., 1969, Holocene geology of Taiwan: Acta Geologica Taiwanica, v.
13, p. 83–126.
Liu, T.K., Chen, Y.G., Chen, W.S., and Jiang, S.H., 2000, Rates of cooling and
denudation of the Early Penglai Orogeny, Taiwan, as assessed by fission-
track constraints: Tectonophysics, v. 320, p. 69–82, doi: 10.1016/S0040-
1951(00)00028-7.
Liu, T.K., Hsieh, S., Chen, Y.G., and Chen, W.S., 2001, Thermo-kinematic evo-
lution of the Taiwan oblique-collision mountain belt as revealed by zircon
fission track dating: Earth and Planetary Science Letters, v. 186, p. 45–56,
doi: 10.1016/S0012-821X(01)00232-1.
Lo, C.H., and Yu, T.F., 1996, 40Ar/39Ar dating of high-pressure rocks in the
Tananao basement complex, Taiwan: Journal of the Geological Society of
China, v. 39, p. 13–30.
Lu, C.-Y., and Hsü, K.-J., 1992, Tectonic evolution of the Taiwan mountain
belt: Petroleum Geology of Taiwan, v. 27, p. 21–46.
Lundberg, N., Reed, D.L., Liu, C.-S., and Lieske, J., 1997, Forearc-basin clo-
sure and arc accretion in the submarine suture zone south of Taiwan: Tec-
tonophysics, v. 274, p. 5–23, doi: 10.1016/S0040-1951(96)00295-8.
Malavieille, J., Lallemand, S.E., Dominguez, S., Deschamps, A., Lu, C.-Y., Liu,
C.-S., and Schnurle, P., and the ACT scientific crew, 2002, Arc-continent
collision in Taiwan: New marine observations and tectonic evolution, in
Byrne, T.B., and Liu, C.-S., eds., Geology and Geophysics of an Arc-Con-
tinent Collision, Taiwan: Geological Society of America Special Paper
358, p. 187–211.
278 Clift et al.
Peng, T.-H., Li, Y.-H., and Wu, F.T., 1977, Tectonic uplift rates of the Taiwan
Island since the early Holocene, in Ho, C.S., A Special Volume Dedicated
to Biq Chingchang on the Occasion of His Retirement: Taipei, Taiwan,
Geological Society of China Memoir 2, p. 57–69.
Platt, J.P., and Vissers, R.L.M., 1989, Extensional collapse of thickened conti-
nental lithosphere: A working hypothesis for the Alboran sea and Gibraltar
arc: Geology, v. 17, p. 540–543, doi: 10.1130/0091-7613(1989)017<0540:
ECOTCL>2.3.CO;2.
Rau, R.-J., and Wu, F.T., 1995, Tomographic imaging of lithospheric structures
under Taiwan: Earth and Planetary Science Letters, v. 133, p. 517–532,
doi: 10.1016/0012-821X(95)00076-O.
Reiners, P.W., Ehlers, T.A., Mitchell, S.G., and Montgomery, D.R., 2003, Cou-
pled spatial variations in precipitation and long-term erosion rates across
the Washington Cascades: Nature, v. 426, p. 645–647, doi: 10.1038/
nature02111.
Roberts, A.M., Yielding, G., Kusznir, N.J., Walker, I., and Dorn-Lopez, D.,
1993, Mesozoic extension in the North Sea: Constraints from flexural
backstripping, forward modeling and fault populations, in Parker, J.R.,
ed., Petroleum Geology of NW Europe: Geological Society [London],
Proceedings of the 4th Conference, p. 1123–1136.
Rudnick, R.L., and Fountain, D.M., 1995, Nature and composition of the con-
tinental crust; a lower crustal perspective: Reviews of Geophysics, v. 33,
p. 267–309, doi: 10.1029/95RG01302.
Shinjo, R., 1999, Geochemistry of high Mg andesites and the tectonic evolution
of the Okinawa Trough–Ryukyu Arc system: Chemical Geology, v. 157,
p. 69–88, doi: 10.1016/S0009-2541(98)00199-5.
Shyu, J.B.H., Sieh, K., Chen, Y.G., and Liu, C.S., 2005, Neotectonic architec-
ture of Taiwan and its implications for future large earthquakes: Journal of
Geophysical Research, v. 110, p. B08402, doi: 10.1029/2004JB003251.
Sibuet, J.-C., and Hsu, S.K., 1997, Geodynamics of the Taiwan arc–arc
collision: Tectonophysics, v. 274, p. 221–251, doi: 10.1016/S0040-
1951(96)00305-8.
Sibuet, J.-C., Deffontaines, B., Hsu, S.K., Thareau, N., Le Formal, J.P., and Liu,
C.S., 1998, Okinawa Trough backarc basin; early tectonic and magmatic
evolution: Journal of Geophysical Research, v. 103, p. 30,245–30,267,
doi: 10.1029/98JB01823.
Sibuet, J.-C., Hsu, S.K., Le Pichon, X., le Formal, J.P., Reed, D., Moore, G., and
Liu, C., 2002, East Asia plate tectonics since 15 Ma: Constraints from the
Taiwan region: Tectonophysics, v. 344, p. 103–134, doi: 10.1016/S0040-
1951(01)00202-5.
Suppe, J., 1981, Mechanics of mountain building and metamorphism in Tai-
wan, in Ho, C.S., and Ernst, W.G., Papers Presented at the ROC-USA
Seminar on Plate Tectonics and Metamorphic Geology: Taipei, Taiwan,
Geological Society of China Memoir 4, p. 67–89.
Suppe, J., 1984, Kinematics of arc-continent collision, flipping of subduction,
and backarc spreading near Taiwan, in Tsan, S.F., et al., eds., A Special
Volume Dedicated to Chun-Sun Ho on the Occasion of His Retirement:
Geological Society of China Memoir 6, p. 21–33.
Suppe, J., Wang, Y., Liou, J.G., and Ernst, W.G., 1976, Observation of some
contacts between basement and Cenozoic cover in the Central Range, Tai-
wan: Proceedings of the Geological Society of China, v. 19, p. 59–70.
Tagami, T., Galbraith, R.F., Yamada, G.M., and Laslett, G.M., 1998, Revised
annealing kinetics of fission-tracks in zircon and geological implications,
in Van den Haute, P., and De Corte, F., eds., Advances in Fission-track
Geochronology: Amsterdam, Kluwer Academic Press, p. 99–112.
Teng, L.S., 1990, Geotectonic evolution of late Cenozoic arc-continent col-
lision in Taiwan: Tectonophysics, v. 183, p. 57–76, doi: 10.1016/0040-
1951(90)90188-E.
Teng, L.S., 1996, Extensional collapse of the northern Taiwan mountain belt:
Geology, v. 24, p. 949–952, doi: 10.1130/0091-7613(1996)024<0949:
ECOTNT>2.3.CO;2.
Teng, L.S., and Lee, C.-T., 1996, Geodynamic appraisal of seismogenic faults
in northeast Taiwan: Proceedings of the Geological Society of China,
v. 39, p. 125–142.
Teng, L.S., Wang, Y., Tang, C.-H., Huang, C.-Y., Huang, T.-C., Yu, M.-S., and
Ke, A., 1991, Tectonic aspects of the Paleogene depositional basin of
northern Taiwan: Proceedings of the Geological Society of China, v. 34,
p. 313–336.
Teng, L.S., Lee, C.-T., Tsai, Y.-B., and Hsiao, L.-Y., 2000, Slab breakoff as a
mechanism for flipping of subduction polarity in Taiwan: Geology, v. 28,
p. 155–158, doi: 10.1130/0091-7613(2000)28<155:SBAAMF>2.0.CO;2.
Tippett, J.M., and Kamp, P.J.J., 1993, Fission-track analysis of the Late Ceno-
zoic vertical kinematics of continental Pacific crust, South Island, New
Zealand: Journal of Geophysical Research, v. 98, p. 16,119–16,148.
Vita-Finzi, C., and Lin, J.C., 1998, Serial reverse and strike slip on imbricate
faults: The Coastal Range of east Taiwan: Geology, v. 26, p. 279–282,
doi: 10.1130/0091-7613(1998)026<0279:SRASSO>2.3.CO;2.
von Huene, R., and Scholl, D.W., 1991, Observations at convergent margins
concerning sediment subduction and the growth of continental crust:
Reviews of Geophysics, v. 29, p. 279–316.
Waldhauser, F., 2001, HypoDD: A computer program to compute double-dif-
ference hypocenter locations: U.S. Geological Survey Open-File Report
01-113 (Menlo Park, California).
Waldhauser, F., and Ellsworth, W.L., 2000, A double-difference earthquake
location algorithm: Method and application to the northern Hayward
fault: Bulletin of the Seismological Society of America, v. 90, p. 1353–
1368, doi: 10.1785/0120000006.
Wang, K.L., Chung, S.L., Chen, C.-H., Shinjo, R., Yang, T.F., and Chen, C.H.,
1999, Post-collisional magmatism around northern Taiwan and its relation
with opening of the Okinawa Trough: Tectonophysics, v. 308, p. 363–376,
doi: 10.1016/S0040-1951(99)00111-0.
Wang, P.L., Lin, L.H., and Lo, C.H., 1998, 40Ar/39Ar dating of mylonitiza-
tion in the Tananao schist, eastern Taiwan: Geological Society of China
Journal, v. 41, p. 159–183.
Willett, S.D., Fisher, D., Fuller, C., Yeh, E.-C., and Lu, C.-Y., 2003, Erosion
rates and orogenic-wedge kinematics in Taiwan inferred from fission-
track thermochronometry: Geology, v. 31, p. 945–948, doi: 10.1130/
G19702.1.
Wu, F.T., 1978, Recent tectonics in Taiwan: Journal of Physics of the Earth,
v. 26, p. S265–S299.
Wu, F.T., Rau, R.J., and Salzberg, D.H., 1997, Taiwan tectonics: Thin-skinned or
lithospheric collision: Tectonophysics, v. 274, p. 191–220, doi: 10.1016/
S0040-1951(96)00304-6.
Yu, H.-S., and Chou, Y.-W., 2001, Characteristics and development of the
flexural forebulge and basal unconformity of western Taiwan fore-
land basin: Tectonophysics, v. 333, p. 277–291, doi: 10.1016/S0040-
1951(00)00279-1.
Zeitler, P.K., Chamberlain, C.P., and Smith, H.A., 1993, Synchronous anatexis,
metamorphism and rapid denudation at Nanga Parbat (Pakistan Himalaya):
Geology, v. 21, p. 347–350, doi: 10.1130/0091-7613(1993)021<0347:
SAMARD>2.3.CO;2.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

The Guerrero Composite Terrane of western Mexico: Collision and subsequent rifting in
a supra-subduction zone

E. Centeno-García*
Instituto de Geología, Universidad Nacional Autónoma de México, Ciudad Universitaria, México D.F. 04510, México

M. Guerrero-Suastegui
O. Talavera-Mendoza
Unidad Académica de Ciencias de la Tierra, Universidad Autónoma de Guerrero, AP 197, Taxco el Viejo, Guerrero, México

ABSTRACT

The Guerrero Composite Terrane of western Mexico is the second largest ter-
rane in North America. Mostly characterized by submarine volcanism and formed
by five terranes, the Guerrero records vast and complex subduction-related processes
influenced by major translation and rifting. It is composed of the Teloloapan, Gua-
najuato, Arcelia, Tahue, and Zihuatanejo Terranes. The Teloloapan Terrane is made
up of Lower Cretaceous island-arc (IA) andesitic to basaltic submarine lava flows,
interbedded with limestone and shallow-marine volcaniclastic rocks. The Guanajuato
and Arcelia Terranes are characterized by Lower Cretaceous supra-subduction
ophiolite successions formed by deep-marine volcanic and sedimentary rocks with
mid-oceanic-ridge basalt (MORB), oceanic-island basalt (OIB), and island-arc basalt
(IAB) signatures. These two terranes are placed between the continent and the more
evolved arc assemblages of the Zihuatanejo Terrane. The Tahue Terrane is composed
of Paleozoic accreted arc and eugeoclinal sedimentary rocks, Triassic rift-related
metaigneous rocks, and overlain unconformably by pillow basalts, limestone, and
volcaniclastic rocks. The Zihuatanejo Terrane was formed by Triassic ocean-flank to
ocean-floor assemblages accreted in Early Jurassic time (subduction complexes). The
subduction complexes are overlain by Middle Jurassic–evolved volcanic arc rocks,
which are in turn unconformably overlain by Early and Late Cretaceous subaerial
and marine arc-related volcano-sedimentary assemblages.
Mesozoic stratigraphy at the paleocontinental margin of Mexico (Oaxaquia and
Mixteca Terranes) is formed by Triassic submarine fan turbidites accreted during
Early Jurassic time; Middle Jurassic–evolved volcanic arc rocks are unconformably
covered by a Late Jurassic to Cretaceous calcareous platform.
Six stages in the tectonic evolution are proposed on the basis of the stratigraphic
and deformational events recorded in western Mexico: (1) A passive or rifting margin
developed along the western margin of continental Mexico throughout the Triassic. A

*centeno@servidor.unam.mx

Centeno-García, E., Guerrero-Suastegui, M., and Talavera-Mendoza, O., 2008, The Guerrero Composite Terrane of western Mexico: Collision and subsequent rift-
ing in a supra-subduction zone, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones:
Geological Society of America Special Paper 436, p. 279–308, doi: 10.1130/2008.2436(13). For permission to copy, contact editing@geosociety.org. ©2008 The
Geological Society of America. All rights reserved.

279
280 Centeno-García et al.

thick siliciclastic turbiditic succession of the Potosi Submarine Fan was accumulated
on the paleo-continental shelf-slope and extended to the west in a marginal oceanic
basin. (2) Subduction began in the Early Jurassic, and the turbidites of the Potosi
Fan with slivers of the oceanic crust were accreted, forming a wide subduction prism.
(3) Exhumation of the accretionary prism and development of a Middle Jurassic con-
tinental arc onto the paleo-continental margin (Oaxaquia and Mixteca Terrane) took
place, and also in the Zihuatanejo Terrane. (4) Intra-arc strike-slip faulting and rift-
ing of the Middle Jurassic continental arc took place along with migration of the
subduction toward the west and development of a calcareous platform in Oaxaquia
and the Mixteca Terrane (continental Mexico). (5) Drifting of the previously accreted
Tahue and Zihuatanejo Terranes formed a series of marginal arc-backarc systems,
or one continuously drifting arc with intra-arc and backarc basins during Early to
middle Cretaceous time. (6) Deformation of the arc assemblages, and development of
Santonian to Maastrichtian foreland and other basins, date the final amalgamation of
the Guerrero Composite Terrane with the continental margin.

Keywords: Guerrero Terrane, Mexico, tectonics, Triassic subduction complex, Creta-

ceous arc volcanism.

INTRODUCTION suggested that the Guerrero Composite Terrane might represent

The present configuration of continental Mexico was built
after accretion of basement remnants and oceanic terranes. Dur-
ing most of their Mesozoic history, Proterozoic to Paleozoic
accreted terranes formed a relatively narrow neck of land adja-
cent to the North American craton. This was bordered on its
eastern side by rifting and on its western side by active subduc-
tion. Thus Mexico is probably one of the most suitable regions
in North America for studying the interaction between these two
differing tectonic scenarios. We suggest in this paper, based on
evidence recorded in the stratigraphy of the Guerrero Composite
Terrane and surrounding terranes, that the almost continuously
subducting Pacific margin of Mexico was directly influenced by
extensional tectonics associated with the breakup of Pangea and
the formation of the Gulf of Mexico.
The Guerrero Composite Terrane (Campa and Coney, 1983)
constitutes approximately one-third of Mexico. As originally
described, it is the largest of all the Mexican terranes and probably
the second largest of the North America Cordillera after Wrangel-
lia (Campa and Coney, 1983; Centeno-García et al., 1993a). The
Guerrero Composite Terrane is characterized mostly by subma-
rine and locally subaerial volcanic and sedimentary successions
that range in age from Jurassic (Tithonian) to middle–Late Creta-
ceous (Cenomanian), and scarce exposures of older rocks. A wide
variety of models has been proposed for the origin of the Guerrero
Composite Terrane. Like other terranes of the North America Cor-
dillera, it was first interpreted as an exotic terrane formed by a far-
traveled Cretaceous oceanic arc. Some authors have suggested that
it was an oceanic arc terrane that was accreted to nuclear Mexico
in Late Cretaceous time via a westward-dipping subduction zone
that closed a major ocean basin (Lapierre et. al., 1992; Tardy et
al., 1994; Dickinson and Lawton, 2001, etc.). Other authors have
one or more complex systems of two or three peripheral arcs that
developed relatively close to the continent (Campa and Ramírez,
1979; Ramírez-Espinosa et al., 1991; Mendoza and Suastegui,
2000; Centeno-García et al., 2003; Centeno-García, 2005). Some
models even proposed that the arc was autochthonous and was
built upon Proterozoic continental crust of nuclear Mexico (de
Cserna, 1978; Elías-Herrera and Sánchez-Zavala, 1990). In other
words, there is a model for each likely possibility, but each lacks
strong supporting evidence.
New findings on the stratigraphy, discussed in this paper,
suggest a more complex evolution, implying a series of accre-
tions to the continent followed by rifting, and later by collision.
In this paper we attempt to present our insights into the evolu-
tion of western Mexico gained from examining the stratigraphy
and structure, and the geochemical and geochronological data, of
such a vast area. However, we discuss in this paper only strati-
graphic units and localities that are keys for reconstructing the
tectonic evolution. This paper synthesizes the work done by
many authors. Although there is the need for more geochrono-
logical and detailed field work, we consider that the preliminary
tectonic model presented in this paper is consistent with the evi-
dence collected to date.
OVERVIEW OF THE GUERRERO COMPOSITE AND
NEIGHBORING TERRANES
The stratigraphy of western Mexico is synthesized in this
paper under the framework of tectono-stratigraphic terranes,
which are regions that share the same geological history and
are bounded by major faults. As mentioned before, by the early
Mesozoic, the Paleozoic and Proterozoic terranes were already
accreted to the southern part of the North American craton.
 Guerrero Composite Terrane of western Mexico 281

Those that already formed part of the continental margin dur-
ing the Mesozoic were Oaxaquia and the Mixteca, Parral, and
Cortes Terranes (Fig. 1). Terranes accreted or displaced during
the Mesozoic were those of the Guerrero Composite, the Central,
as well as terranes of the western Baja California Peninsula. The
latter will not be reviewed in this paper. A brief summary of the
stratigraphy is described as follows; more detailed descriptions of
key areas and events are discussed later.
OAXAQUIA
At the end of the Paleozoic, Proterozoic basement terranes
of Gondwanan affinity were already accreted to the southern part
of the North American craton. The largest of these is the Oaxa-
quia block (Fig. 1), a crustal fragment, subcontinent in size, of
Grenville affinity (Ortega-Gutiérrez et al., 1995). This crustal
block forms the backbone of eastern Mexico and is referred to
herein as continental Mexico for the Mesozoic. Oaxaquia has
a Precambrian (1157–900 Ma) crystalline basement (gneisses
and anorthosites; Patchett and Ruíz, 1987; Ortega-Gutiérrez et
al., 1995; Ramírez-Ramírez, 1992; Lawlor et al., 1999; Solari
et al., 2003; Keppie et al., 2003). It is covered by Paleozoic
sedimentary rocks (Fig. 2) that are capped by Permian volca-
nic and volcaniclastic rocks (McKee et al., 1999; Stewart et al.,
1999; Rosales-Lagarde et al., 2005). Triassic (Carnian–Norian)
sedimentary rocks (La Ballena Formation) are exposed at the
western margin of Oaxaquia (Labarthe et al., 1982; Silva-
Romo, 1993; Tristán-Gonzalez and Torres-Hernández, 1994;
Centeno-García and Silva-Romo, 1997; Barboza-Gudiño et al.,
1998, 1999, 2004; Bartolini et al., 2002). These rocks are made
up of a thick succession of turbidites (Fig. 2) deposited in a
submarine fan environment named the Potosi Fan (Centeno-
García, 2005).
Triassic rocks of the Potosi Fan were deformed prior to deposi-
tion of Jurassic volcanic-volcaniclastic rocks (Centeno-García and
Silva-Romo, 1997). They are interpreted as a Jurassic continental
arc and rest unconformably on the Triassic Potosi Fan. Jurassic
arc strata are made up of subaerial andesitic-rhyolitic lava flows,
interbedded with volcaniclastic rocks (Silva-Romo, 1993). The
arc sequence changes transitionally upsection to shallow-marine

115° 110° 105° 100° 95° 90°

USA Cenozoic volcanism

C
a
Te b o M ÉXICO Oaxaquia
Mo
r r rc j av North Cortes Terrane
an a e-
30° e So America 30°
no Parral Terrane
ra La
Ba
bia Central Terrane
Me

Cortes Fa
O ult Mixteca Terrane
ga

Terrane Sa a x
sh

n M aq
uia Guerrero Composite Terrane
ea

arc
r

os
Fa
ult Tahue Terrane
Arcelia Terrane N
Mojave-Sonora
Megashear Zihuatanejo 25°
25°
G

Early Mesozoic Co

Teloloapan
ue

Central
Fol

Fig. 6
Terrane Guanajuato
rr

d
and
er
o

O
Thr
C

ax
ust

.4
om

g aq
Fi
Bel
po

ui
nt

GULF OF
t Fr
in
PA

si

a
en

on
te

ta

MÉXICO
l
C

t
M
Te

ar
IF

gi
rr

20°
IC

20°
an
e
O
C

Mixteca
E

Fig. 8
A
N

A
IC

AL
ÉX

EM
M

T
UA
G

115° 110° 105° 100° 95° 90°

Figure 1. Map showing main tectono-stratigraphic terranes, major faults mentioned in the text, and locations of Figures 4, 6, and 8.
282 Centeno-García et al.

GUERRERO COMPOSITE TERRANE

Guanajuato
Teloloapan

Oaxaquia
Caborca
Mixteca

Central
Arcelia

Cortes
Parral
Tahue
Zihuatanejo
Maastrichtian
Coast Huetamo Northern basic-andesitic
Campanian submarine volcanism
Santonian andesitic
Coniacian submarine volcanism
Turonian
❖ rhyolitic-andesitic
submarine volcanism
Cenomanian
? rhyolitic-andesitic
Albian ❖ continental volcanism
Aptian
volcaniclastic rocks
terrestrial and shallow
Barremian
marine sedimentary
? ? ●
●
rocks
Hauterivian siliciclastic and
Valanginian ●
volcaniclastic rocks
Berriasian limestone, shale
● and evaporites
Tithonian ●
●
siliciclastic turbidites
Kimmeridgian
● ●
?
Oxfordian ● accretionary prism
Callovian ●
Bathonian metamorphic arc
Bajocian volcanic and
Aalenian ● sedimentary rocks
Toarcian gneiss and schist
Pliensbachian
●
Sinemurian

Hettangian
Rhaetian

Norian

Carnian
Ladinian
Early Triassic

Paleozoic
Proterozoic

Figure 2. Simplified stratigraphic columns for Oaxaquia and terranes mentioned in the text. They show the age range (in Ma) of sedimentation
and magmatism for western and central Mexico. Geochronological data are represented as follows: Black circles are U/Pb ages, and diamonds
are Ar/Ar and K/Ar ages.

volcaniclastic rocks, limestone, and some evaporites (Fig. 2;
Silva-Romo, 1993; Tristán-González and Torres-Hernández,
1994; Barboza-Gudiño et al., 2004). Calcareous sedimentation
in Oaxaquia ranges in age from late Oxfordian–Kimmeridgian to
Turonian and is interpreted as the southern extension of the North
American seaway. A major change upsection from calcareous to
clastic sedimentation occurred at the uppermost part of the Cre-
taceous, forming a thick succession of sandstone, shale, and con-
glomerate (Caracol Formation; Silva-Romo, 1993). Oaxaquia is
overthrust by the Guerrero Composite Terrane (Fig. 1).
 Guerrero Composite Terrane of western Mexico
MIXTECA TERRANE
The basement of the eastern Mixteca Terrane is made up of
pre-Mississippian polydeformed metamorphic rocks of the Acat-
lán Complex (Ortega-Gutiérrez, 1981; Ruíz et al., 1988; Yañez et
al., 1991). This complex is considered to be the result of complex
interactions between Gondwana and Laurentia previous and dur-
ing the assembling of Pangea (Ortega-Gutiérrez et al., 1999). It
is unconformably overlain by Permian sedimentary rocks, which
are in turn overlain unconformably by Middle Jurassic volcanic
and sedimentary rocks (Fig. 2; García-Díaz et al., 2004). At the
western part of the terrane, near the limit with the Guerrero Com-
posite Terrane, partly metamorphosed volcanic and volcaniclastic
rocks are exposed (Taxco Schist and Chapolapa Formation; de
Cserna and Fries, 1981; Talavera-Mendoza, 1993; Campa and Iri-
ondo, 2004). The Taxco Schist is made up of andesitic to rhyolitic
lavas and volcaniclastic rocks of Early Cretaceous age (Talavera-
Mendoza, 1993; Campa and Iriondo, 2004). The Taxco Schist
is unconformably overlain by a thick limestone succession of
Albian to Cenomanian age and by Turonian–Maastrichtian clastic
rocks (Mexcala Formation; Campa and Ramírez, 1979; Talavera-
Mendoza et al., 1995). Contacts between the Mixteca Terrane and
Oaxaquia, as well as between the Mixteca and Guerrero Com-
posite Terranes, are partially exposed. The Mixteca Terrane is on
strike-slip fault contact with Oaxaquia, and rocks of the Guerrero
Composite Terrane are thrust over the Mixteca Terrane.
PARRAL TERRANE
The Parral Terrane (Figs. 1 and 2) was first defined by
Pacheco et al. (1984) and Coney and Campa (1987) and was rede-
fined by Centeno-García (2005). The basement of the Parral Ter-
rane is formed by Devonian to Carboniferous metamorphic rocks
(Pescadito Schist; Eguiluz and Campa, 1982; Araujo and Arenas,
1986; Zaldivar and Garduño, 1984). These Paleozoic metamor-
phic rocks are unconformably overlain by red beds and volca-
nic successions (Nazas Formation; Pantoja-Alor, 1963), which
change transitionally to Tithonian limestone (Araujo and Arenas,
1986; Contreras-Montero et al., 1988). Cretaceous calcareous
and clastic sedimentation of the Parral Terrane is laterally con-
tinuous with the calcareous-clastic deposits that cover Oaxaquia
and the Central Terrane. Relationships among the Parral, Cen-
tral, and Cortes Terranes and the Parral Terrane and Oaxaquia are
unknown, because the contacts are covered by Cretaceous lime-
stone or by Cenozoic volcanic successions. Therefore, the exact
locations of their boundaries are unknown but are inferred by the
difference in styles of deformation of the Cretaceous rocks.
CABORCA AND CORTES TERRANES
The Caborca Terrane has a Proterozoic basement older than
1.7 Ga (Anderson and Silver, 1981), covered by a thick Paleozoic
sedimentary succession. It has been interpreted to be a displaced
block of continental North America, originally located along the
283
Paleozoic “miogeosyncline” of Nevada. It was transferred toward
the south by Middle to Late Jurassic time via the Mohave-Sonora
megashear (Anderson and Silver, 1979, 2005; Stewart et al.,
1990). The Cortes Terrane is interpreted as an autochthonous ter-
rane to North America, which probably evolved at the margin of
the Caborca Terrane (Stewart et al., 1990). It is made up of a thick
succession of Paleozoic deep-marine turbidites that were thrust
over platform limestone of the Caborca Terrane (Figs. 1 and 2).
The Cortes Terrane is interpreted as continental-slope depos-
its, and it is considered the southern extension of the Paleozoic
Cordilleran “eugeoclinal” deposits from Nevada and California
(Poole and Madrid, 1988; Coney and Campa, 1987; Stewart et al.,
1990). The previously deformed Paleozoic deep-marine rocks of
the Cortes Terrane are overlain by Triassic (Carnian–Norian) ter-
restrial and marine sedimentary rocks (Stewart et al., 1990; Stew-
art and Roldán-Quintana, 1991). The Triassic rocks are overlain
by Cretaceous red beds and volcanic rocks (Stewart and Roldán-
Quintana, 1991). Contact relationships between the Cortes and
Guerrero Composite Terranes have not been well constrained,
but the contact is inferred to be a Late Cretaceous thrust fault.
CENTRAL TERRANE
The nature of the basement of the Central Terrane is unknown,
but it is assumed to be different from the Proterozoic basement of
Oaxaquia because its oldest exposed rocks near its contact are
a subduction-related accretionary complex (Taray Formation;
Anderson et al., 1990; Diaz-Salgado et al., 2003; Anderson et al.,
2005; Centeno-García, 2005). The subduction zone on which the
Taray Formation was deformed was probably constructed along
the Oaxaquia continental margin between Late Permian and Early
Jurassic time (Diaz-Salgado et al., 2003; Anderson et al., 2005).
The complex is unconformably overlain by Oxfordian subaerial
rhyolitic to andesitic volcanic rocks and red beds (Jones et al.,
1995). These rocks change transitionally to shallow-marine lime-
stone that ranges in age from Late Jurassic to Late Cretaceous
(Córdoba-Méndez, 1964). The location of the northern and east-
ern contact between the Central Terrane and Oaxaquia is inferred
on the basis of the location of the last exposures of Paleozoic–
early Mesozoic rocks, and a contrast in deformation styles of Cre-
taceous rocks in both (Fig. 1). The contact between the Central
and Guerrero Composite Terranes has not been studied in detail
but is inferred on the basis of the distribution of the northernmost
exposures of Cretaceous marine volcanic rocks that belong to the
Guerrero Composite Terrane. Structural trends on both sides of
the contact suggest that the Central Terrane is overthrust by the
Guerrero Composite Terrane to the south (Fig. 1). The thrusting is
inferred to have occurred about Late Cretaceous time.
GUERRERO COMPOSITE TERRANE
Areas with large volumes of Lower Cretaceous volcanic
and volcaniclastic rocks, located toward the west of Oaxa-
quia and the Mixteca Terrane, were originally grouped as the
284 Centeno-García et al.
Guerrero Terrane by Campa and Coney (1983) and were, 10
years later, divided into the Tahue, Nahuatl, and Tepehuano
Terranes by Sedlock et al. (1993). Subsequent regional map-
ping has shown that the divisions proposed by Campa and
Coney (1983) are closer to the field locations of faults delim-
iting the terranes than those of Sedlock et al. (1993). There-
fore, more recent reviews of the terrane distribution of Mexico
(e.g., Centeno-García, 2005) have been based on Campa and
Coney (1983). The Guerrero is a composite terrane, formed by
at least five terranes: Tahue, Zihuatanejo, Guanajuato, Arcelia,
and Teloloapan (Figs. 1 and 2; Talavera-Mendoza et al., 1995;
Mendoza and Suastegui, 2000; Centeno-García et al., 2003;
Centeno-García, 2005). Their stratigraphy is briefly described
from NNW to ESE (Fig. 1):
Tahue Terrane
The Tahue Terrane contains the oldest rocks found so far
within the Guerrero Composite Terrane (Fig. 2; Centeno-García,
2005). These rocks comprise Ordovician marine rhyolitic-
andesitic lavas and clastic and calcareous rocks, all deformed
and metamorphosed to low-greenschist facies (El Fuerte Com-
plex; Mullan, 1978; Roldán-Quintana et al., 1993; Poole and
Perry, 1998). These rocks may have originated as an oceanic
arc that apparently was accreted previous to the deposition of
Pennsylvanian–Permian deep-marine sedimentary rocks (San
José de Gracia Formation; Carrillo-Martínez, 1971; Gastil et
al., 1991; Arredondo-Guerrero and Centeno-García, 2003;
Centeno-García, 2005). These deep-marine turbidites are strongly
deformed but do not show the metamorphism of the El Fuerte
Complex; thus an unconformable contact relationship between
these two units is inferred. Paleozoic rocks of the Tahue Terrane
are unconformably overlain by Cretaceous marine arc volcanic
rocks and are interpreted as part of the Guerrero Arc (Ortega-
Gutiérrez et al., 1979; Henry and Fredrikson, 1987; Roldán-
Quintana et al., 1993; Freydier et al., 1995). These rocks are also
cut by mafic and ultramafic intrusions that are part of the same
Cretaceous arc magmatism (Henry and Fredrikson, 1987; Gas-
til et al., 1999; Arredondo-Guerrero and Centeno-García, 2003).
Therefore, the Paleozoic units form the basement upon which
the arc was built. The Tahue Terrane also contains metamorphic
rocks of Triassic age (Keppie et al., 2006). The contact relation-
ship between the Cortes and Tahue Terranes has not been studied
in detail, but it is inferred to be a thrust (Fig. 1; Roldán-Quintana
et al., 1993). The contact between the Tahue and Zihuatanejo Ter-
ranes is not exposed.
Zihuatanejo Terrane
The Zihuatanejo Terrane is the largest of all terranes that
form the Guerrero Composite Terrane (Fig. 1). It extends north
of the Mexican Volcanic Belt and along the Pacific Coast of Mex-
ico (Centeno-García et al., 1993a, 1993b; Talavera-Mendoza et
al., 1995; Mendoza and Suastegui, 2000). Its basement is made
up of large volumes of Triassic (Norian) quartz-rich turbidites
(sandstone and shale) that are tectonically imbricated (Campa
et al., 1982; Centeno-García et al., 1993a, 1993b). The turbi-
dites form a matrix within which are blocks and slabs of pillow
basalts, diabase, banded gabbros, chert, and limestone (Fig. 2).
These rocks have received different names at different outcrops:
Zacatecas Formation, Arteaga Complex, and Las Ollas Complex
(Burckhardt and Scalia, 1906; Ranson et al., 1982; Cuevas-Pérez,
1983; Monod and Calvet, 1991; Centeno-García and Silva-
Romo, 1997; Talavera-Mendoza, 2000; Centeno-García et al.,
2003). The deformation of these rocks varies from gently folded
strata to highly sheared block-in-matrix textures, and their meta-
morphism ranges from none to high-greenschist–amphibolite
facies (Centeno-García et al., 2003). Blueschist facies have been
reported only in one locality (Las Ollas Complex; Talavera-
Mendoza, 2000). These lithologies are interpreted to constitute
an Upper Triassic(?)–Lower Jurassic subduction-related accre-
tionary complex.
Scattered exposures of rocks of Middle to Late Jurassic–
evolved arc volcanism lie along the Pacific Coast of the Zihua-
tanejo Terrane. These rocks are made up of submarine rhyo-
litic lavas and volcaniclastic rocks, and granitoids that were
emplaced in rocks of the accretionary complex (Bissig et al.,
2003; Centeno-García et al., 2003). The Middle to Upper Juras-
sic arc rocks were in turn deformed and exhumed previous to the
deposition of uppermost Jurassic–Cretaceous arc-related strata
(Centeno-García et al., 2003).
The Cretaceous arc succession ranges from Berriasian to
Cenomanian in age, and it includes andesitic, basaltic, and some
rhyolitic volcanic and volcaniclastic rocks, interbedded with
limestone, evaporites, and some red beds (Grajales and López,
1984). The arc succession contains abundant fossils such as rud-
ists, gastropods, microfossils, fossil logs, and vertebrates.
This arc succession was deformed prior to the intrusion of
large granitoids of latest Cretaceous to Paleogene age (Schaaf
et al., 2000). Also, uppermost Cretaceous (Santonian to Maas-
trichtian) red beds and volcanic rocks rest unconformably on all
previous units (Altamira Areyán, 2002; Benammi et al., 2005).
The contact between the Zihuatanejo Terrane and Oaxaquia is
exposed at its northern limit, where Cretaceous arc rocks of the
Zihuatanejo Terrane are thrust over shallow-marine limestone of
Oaxaquia. Its contact with the Arcelia and Guanajuato Terranes
is inferred to be an east-verging thrust, but it is covered by upper-
most Cretaceous and Cenozoic red beds and volcanic rocks.
Guanajuato Terrane
The Guanajuato Terrane has been interpreted as a com-
plete crustal section through a primitive island arc that appears
to lack an older basement (Ortiz-Hernandez et al., 1991; Ortiz-
Hernandez, 1992). It has also been interpreted as the remains of
an oceanic basin that lay between the Guerrero arc and the conti-
nental margin (Freydier et al., 2000). This terrane was formed by
a series of tectonic slivers that placed lower crust rocks (gabbro,
 Guerrero Composite Terrane of western Mexico
tonalite, serpentinite, wehrlite, and dike swarms) on pillow
basalts, rhyolitic tuffs, volcanic turbidites, chert, and black detri-
tal limestone (Quintero-Legorreta, 1992; Ortiz-Hernandez et al.,
1992; Lapierre et al., 1992; Monod et al., 1990; Martínez-Reyes,
1992; Ortiz-Hernandez et al., 2003). These rocks were poorly
dated as Tithonian–Hauterivian in age (Ortiz-Hernandez et al.,
2003; Hall and Mortensen, 2003). Previously deformed volca-
nic turbidites are unconformably overlain by Aptian–Albian
limestone (Ortiz-Hernandez et al., 2003). This suggests that
sedimentation and at least one phase of deformation occurred
previous to the Aptian–Albian (Ortiz-Hernandez et al., 2003).
At present the Guanajuato Terrane is thrust over the calcareous
platform of Oaxaquia (Ortiz-Hernández et al., 2002). Contact
relationships between the Guanajuato and Zihuatanejo Terranes
have not been constrained.
Arcelia Terrane
The Arcelia Terrane is made up of basaltic pillow lavas and
ultramafic bodies, black shale and chert, and volcanic turbidites,
all intensively deformed and partly metamorphosed (Ramírez-
Espinosa et al., 1991; Talavera-Mendoza et al., 1995). It is char-
acterized by Early Cretaceous deep-marine primitive arc or arc-
related oceanic facies and shows the least evolved magmatism
of all the arc successions of the Guerrero Composite Terrane
(Talavera-Mendoza et al., 1995; Mendoza and Suastegui, 2000).
The Arcelia Terrane appears to lack an older basement (Talavera-
Mendoza et al., 1995; Mendoza and Suastegui, 2000). Rocks of
the Arcelia Terrane apparently were thrust over the assemblages
of the Teloloapan Terrane, and were in turn overthrust by rocks
of the Zihuatanejo Terrane. However, these contacts are inferred
because they are covered by younger red beds.
Teloloapan Terrane
The Teloloapan Terrane consists of two distinct regions: the
eastern region is characterized by shallow-marine volcanic and
sedimentary deposits (Fig. 2), and the western region by deeper
volcanic and sedimentary facies (Guerrero-Suastegui et al., 1991;
Ramírez-Espinoza et al., 1991; Talavera-Mendoza et al., 1995;
Mendoza and Suastegui, 2000; Guerrero-Suastegui, 2004). Both
are marine arc assemblages, which vary in composition from
basalt-andesite to scarce dacite-rhyolite (Talavera-Mendoza
et al., 1995). This unit contains microfossils (radiolarians and
coccoliths), gastropods, and bivalves that range in age from
Hauterivian to Aptian; these rocks change transitionally upsec-
tion to Aptian–Albian island-arc carbonates (Guerrero-Suastegui
et al., 1991; Ramírez-Espinoza et al., 1991; Talavera-Mendoza et
al., 1995). The Teloloapan Terrane (Fig. 1) is exposed in the east-
ernmost parts of the Guerrero Composite Terrane. It is charac-
terized structurally by a complex thrust-fault system that verges
eastward. Its Lower Cretaceous rocks are severely deformed and
metamorphosed in low-grade greenschist facies. The Teloloapan
Terrane overrides either Lower to Middle Cretaceous platform
285
carbonates or Upper Cretaceous clastic sediments that belong
to the Mixteca Terrane (Fig. 1; Campa and Ramirez, 1979). The
nature of its basement remains unknown. Metamorphic rocks
that are exposed near the northwestern boundary of the Teloloa-
pan Terrane with the Arcelia Terrane have been interpreted as a
possible basement for the former (Elías-Herrera and Sánchez-
Zavala, 1990; Sanchez-Zavala, 1993). The rocks in this area are
of uncertain age and origin.
TECTONIC MODEL
The most abundant rocks of the Guerrero Composite Terrane
are marine, and rarely subaerial, arc volcanic and sedimentary
successions that range in age from latest Jurassic (Tithonian) to
middle Late Cretaceous (Cenomanian). The composition of the
few scattered exposures of older units suggests a complex earlier
tectonic evolution. These older rocks were not taken into con-
sideration for the tectonic models proposed by previous authors
(de Cserna, 1978; Campa and Ramirez, 1979; Elías-Herrera and
Sánchez-Zavala, 1990; Tardy et al., 1994; Lapierre, et. al., 1992;
Dickinson and Lawton, 2001, etc.). Based on the available infor-
mation, we identified six main tectonic stages in the evolution of
the Guerrero Composite Terrane. These stages are represented in
Figure 3 and are briefly described in this section. Detailed discus-
sion of the data that support the reconstruction of each stage is
presented in the following section.
Stage I: Collision of a Paleozoic Oceanic Arc?—Basement
of the Tahue Terrane
The basement of the Tahue Terrane (Fig. 3) is composed of
the early Paleozoic accreted volcanic-sedimentary rocks of the
El Fuerte Metamorphic Complex. There are not enough data
available to constrain the origin of this complex. Preliminary
interpretations considered these rocks as remnants of Gondwa-
nan crust accreted during the formation of Pangea (Poole et al.,
2005). In this model, metamorphic rocks of El Fuerte could be
the western continuation of basement rocks of the Parral Terrane
(Figs. 1 and 2). An alternative interpretation is that the El Fuerte
Complex may be a displaced fragment of the early Paleozoic arc
(Antler Arc) that collided with the western continental margin of
North America during late Paleozoic time (Burchfiel et al., 1992;
Sánchez-Zavala et al., 1999; Dickinson, 2004; Centeno-García,
2005). Carboniferous deep-marine turbidites (San José de Gracia
Formation) that apparently cover the lower Paleozoic arc rocks
unconformably may be correlative with deep-marine sedimen-
tary rocks exposed in the eastern peninsular ranges of Baja Cali-
fornia and the southwestern Cordillera of North America (Gastil
et al., 1991; Centeno-García, 2005).
In either of the two scenarios, deformed Paleozoic rocks of
the Tahue Terrane are the basement upon which Cretaceous vol-
canism was built, indicating an earlier history of accretion of the
Guerrero Composite Terrane than was previously interpreted by
other authors.
W Tahue
(Guerrero Composite Terrane) Cortes
E
Passive margin sediments Stage I
Arc collision and
Caborca development of a passive
margin
Paleozoic
Vizcaíno? Zihuatanejo (Guerrero Composite) Oaxaquia
and Central terranes Continent margin
Siliciclastic turbidites share the same provenance Stage II
n
Potosí Submarine Fan
Vizcaino?
Marginal oceanic basin
Oaxaquia and Mixteca
with active rift volcanism
Arteaga Basin

Triassic
Carnian-Norian
Vizcaíno? Guerrero Central Oaxaquia
Potosi Fan Stage III
Taray Complex (deformed) Accretion via subduction
Arteaga Complex wide accretionary prism
?
? Oaxaquia and Mixteca Early Jurassic
pre 180-163 Ma

Guerrero Central Oaxaquia

transtension? Stage IV
Tumbiscatío Continental arc and
158 Ma
Granitoid contemporaneous strike-slip
163-158 Ma and extension (roll-back of
+ the subducting plate?)
? +Oaxaquia and Mixteca
Middle to
Late Jurassic
Sub
duc
tion

Guerrero Composite Terrane Central Oaxaquia (north)

and Mixteca
Zihuatanejo/Tahue Arcelia
Guanajuato Teloloapan Stage V
transtension? Formation of a multiple arcs
Calcareous K platform system or a single arc with
intra-arc/back arc rifting
Cretaceous
? ? ? Oaxaquia and Mixteca Berriasian-Cenomanian
Sub
duc
tion

Guerrero Composite Terrane Central / Oaxaquia Stage VI a

Zihuatanejo/Tahue Arcelia Teloloapan and Mixteca (south) Deformation and locally
Guanajuato
metamorphism, cut by
granitoids 105 Ma compression 105 Ma granitoids in the
Zihuatanejo terrane
Oaxaquia and Mixteca Cretaceous
Cenomanian?-pre Santonian

Red Beds Stage VI b

Marine turbidites Active continental arc in
the west and syntectonic
marine turbidites (foreland)
in the east and deformation
Cretaceous
Santonian?-Maastrichtian

Figure 3. Tectonic models for the evolution of western Mexico, showing the alternating stages of subduction-collision and rifting.
 Guerrero Composite Terrane of western Mexico
Stage II: Late Triassic Passive Margin—Deposition of the
Potosi Fan
The paleo-continental edge of Mexico lay approximately
at the western boundary of the Oaxaquia and Mixteca Terranes
in the early Mesozoic (Fig. 1; Centeno-García, 2005). Thus the
Central and Guerrero Composite Terranes (Fig. 1) were accreted
or displaced to their present position during the Mesozoic. Sedi-
mentation along the western continental margin of Oaxaquia was
dominated by large volumes of siliciclastic turbidites (quartz-rich
sandstone and shale) that were deposited in the distal continental
shelf or at the continental slope at least during Carnian–Norian
time (Fig. 3). Accretionary complexes that form the basement of
the Central Terrane and parts of the Guerrero Composite Terrane
(Zihuatanejo Terrane) are formed largely (up to 60% of the total
area of exposures) by similar quartz-rich sandstone and shale
turbidites that made up the matrix within which blocks of vari-
able composition are embedded. These turbidites in the accreted
terranes contain fossils of the same age as those from turbidites
deposited at the continental slope of Oaxaquia.
Detrital zircon ages obtained from turbidites from all the
localities of the Carnian–Norian turbidites, from Oaxaquia to
basal accretionary complexes of the Central and Zihuatanejo Ter-
ranes, show the same populations, which suggest that the fan tur-
bidites spread into a marginal oceanic basin that was later accreted
to the continental margin. These siliciclastic rocks are grouped as
the Potosi Fan (Centeno-García, 2005) and are important because
they can be traced from Oaxaquia to the present Pacific Coast
of Mexico, and they tie together the Central Terrane, the west-
ernmost part of the Guerrero Composite Terrane (Zihuatanejo
Terrane), and the continental margin of southern North America
(Oaxaquia) during Late Triassic time. Thus, the Potosi Subma-
rine Fan may have been a large sedimentary feature, probably
close to the dimensions of the present Bengal Fan.
There is no evidence of Triassic magmatism in continental
Mexico, and detrital zircon geochronology of the fan turbidites
show that the youngest age populations are much older than dep-
ositional ages in all the studied localities of the Potosi Fan (Fig. 3;
Centeno-García et al., 2005; Centeno-García, 2005). Therefore,
the Potosi Fan probably was deposited across a passive margin,
or at least a margin that had no active subduction along the length
of the fan at the time of deposition.
Stage III: Accretion of the Potosi Fan to the Continental
Margin via Subduction—Basement of the Central and
Zihuatanejo Terranes
All the Triassic units of central and western Mexico are
strongly deformed and partially metamorphosed, indicating that a
major compressional event occurred during latest Triassic–Early
Jurassic time. This event is characterized by tight folding, shear-
ing, and axial cleavage in the continent-slope deposits of the
Potosi Fan in Oaxaquia (La Ballena Formation), and block-
in-matrix texture in the Taray Formation (Central Terrane),
287
in the Zacatecas Formation, and in the Arteaga and Las Ollas
Complexes (Zihuatanejo Terrane). These last three units formed
in the distal ocean-floor zone of the Potosi Fan. The presence
of mélanges (Arteaga Complex and Taray Formation) as well as
blueschist in the Las Ollas Complex (Zihuatanejo Terrane) indi-
cates that deformation occurred in a subduction zone. During this
deformational event the turbidites of the Potosi Submarine Fan,
with slivers of the oceanic crust and its sedimentary cover, were
accreted to the continent. This accretionary prism apparently
was very wide, as suggested by the large areas that are floored
by it. Whether the subducting slab was dipping toward the west
(under an oceanic arc) or the east (under continental Mexico)
has not been constrained. There are two isolated reports of dated
Early Jurassic volcanic rocks in Oaxaquia (Barboza-Gudiño et
al., 2004; Fastovsky et al., 2005), but whether they are part of a
continental arc or not is not known. Evidence of contemporane-
ous oceanic-arc magmatism is exposed in the Vizcaíno Penin-
sula of Baja California (Kimbrough and Moore, 2003), where
Triassic–Jurassic volcanic rocks have geochemical signatures of
primitive arc affinity. It is possible that the rocks in the Vizcaíno
Peninsula represent a displaced fragment of an oceanic arc that
accreted to the Arteaga and Las Ollas Complexes of the western
Guerrero Composite Terrane, which in turn accreted to the Taray,
Zacatecas, and La Ballena Formations, but this model needs to be
supported by more evidence.
Stage IV: Late Jurassic Continental Arc—Overlapping
Assemblage for Guerrero Composite Terrane, Central
Terrane, Oaxaquia, and Mixteca Terrane
Subaerial volcanic and sedimentary rocks, as well as shallow
porphyritic intrusives, dikes, and sills, overlie or cut previously
deformed Triassic sedimentary rocks in Oaxaquia and rocks of
the accretionary prism in the Central Terrane. These rocks range
in age from 174 to 158 Ma (Jones et al., 1995; Barboza-Gudiño
et al., 2004). A common attribute of all the outcrops of these
rocks is that they are mostly rhyolitic in composition, with minor
dacitic-andesitic lava flows and tuffs, and show evolved-arc geo-
chemical signatures (Centeno-García and Silva-Romo, 1997;
Centeno-García, 2002; Centeno-García and Díaz-Salgado, 2002).
Coeval volcanic rocks have been reported in the Mixteca Terrane
as well, suggesting that arc volcanism was widespread in conti-
nental Mexico at that time (García-Díaz et al., 2004). Rocks of
similar age range and similar evolved-arc geochemical signatures
are exposed in the western Zihuatanejo Terrane of the Guerrero
Composite Terrane (Bissig et al., 2003; Centeno-García et al.,
2003). This suggests that the Guerrero Composite Terrane may
have been incorporated into the continental margin by that time.
Summarizing the data described above: (1) Triassic base-
ment rocks of the Zihuatanejo Terrane (Guerrero Composite Ter-
rane) share a provenance linkage with rocks of the same age in
Oaxaquia and the Central Terrane; (2) all Triassic rocks, from
those deposited on the paleo-continent’s margin of Mexico to
those within the accreted terranes, were deformed previous to the
288 Centeno-García et al.
development of a Late Jurassic continental arc; and (3) evolved
Upper Jurassic continental-arc volcanism was widespread among
continental Mexico and accreted terranes (Central and Zihua-
tanejo Terranes). On the basis of these facts, we propose in this
paper that the first accretion of the Guerrero Terrane occurred
during latest Triassic–Early Jurassic time instead of near the
end of the Cretaceous, as previously proposed by other authors.
Therefore, the Late Jurassic magmatic event represents an over-
lapping assemblage that stitches all the terranes of central and
western Mexico for that period.
Stage V: Late Jurassic–Early Cretaceous Intra-Arc Strike
Slip(?)–Rifting of the Continental Arc—Drifting of the
Guerrero Composite Terrane
It has been proposed that major lateral displacements
occurred during the activity of the Jurassic continental arc of
stage IV (Anderson and Silver, 2005). Therefore, the arc was
originally in a more northerly position, and it was displaced, via
the Mojave-Sonora Megashear, to its present position in central
Mexico prior to, or at, the early stage of development of the cal-
careous platform (Anderson and Silver, 2005).
Whether this major strike-slip system existed or not has been
widely discussed (see GSA Special Paper 393). We consider that
extensive geological evidence of major tectonism during and
after arc volcanism exists (see following discussion). The cessa-
tion of magmatism in the Central Terrane and Oaxaquia suggests
a change in the location of the subduction zone. Then, a major
regional calcareous platform developed over the arc and other
older rocks. This major transgression initiated the deposition
of limestone on Oaxaquia, and on the Mixteca and Central Ter-
ranes. Calcareous sedimentation in central and eastern Mexico
was characterized by high subsidence rates (Goldhammer, 1999).
Arc magmatism continued only in a small area in the western
Mixteca Terrane and became widespread in the Guerrero Com-
posite Terrane. Although there is some overlap in age ranges of
arc volcanism among the terranes that form the Guerrero Com-
posite Terrane, there is a general trend from older ages in eastern
Oaxaquia and the Central Terrane to younger ages in the western
Guerrero Composite Terrane (Fig. 3). This suggests a possible
W-SW migration of the subduction zone. We propose that dur-
ing and after the continental arc activity (Late Jurassic–Early
Cretaceous time), large amounts of extension and lateral trans-
lations may have occurred (see inferred faults in Fig. 1). This
extensional-transtensional(?) event split the continental arc, ini-
tiating the drifting of parts of previously accreted oceanic rocks
(basements of the Tahue and Zihuatanejo Terranes) and the gen-
eration of new oceanic crust (Guanajuato and Arcelia Terranes).
With the data available, it seems that volcanic activity at the
northern Zihuatanejo Terrane and at the Guanajuato and Teloloa-
pan Terranes was restricted to latest Jurassic–Early Cretaceous
time (Fig. 3). In contrast, in the Arcelia, Tahue, and southern
Zihuatanejo Terranes, arc volcanism apparently continued up
to Albian–Cenomanian time (Fig. 3). Geochemical and isotopic
compositions of most of the Upper Jurassic–Cretaceous igneous
rocks of the different arc assemblages of the Guerrero Composite
Terrane suggest primitive sources, with little or no influence on
an evolved continental crust (e.g., Ortiz-Hernandez et al., 1991;
Lapierre et al., 1992; Centeno-García et al., 1993a; Freydier et al.,
1995; Gastil et al., 1999; Talavera-Mendoza et al., 1995; Men-
doza and Suastegui, 2000, among others). Basalts with ocean-
island (OI) and mid-oceanic-ridge basalt (MORB) signatures of
the Arcelia and Guanajuato terranes (Lapierre et al., 1992; Ortiz-
Hernandez et al., 2003; Mendoza and Suastegui, 2000) suggest
the influence of a mantle source for the magmatism.
Regional differences in the strata suggest abrupt lateral
changes in the depositional environments from shallow marine
to deep marine. Also, lateral differences in thickness of the suc-
cessions suggest that they may have been deposited in alternate
subsiding basins and basement highs where the deposits draped
thinly or were absent. These major geological differences suggest
that intra-arc rifting was considerable and was probably associ-
ated with a complex paleogeography of marginal arc and backarc
systems in western Mexico. Whether or not the different terranes
of the Guerrero Composite Terrane were formed in a single arc
has not been constrained. Some authors proposed that the Guer-
rero Terrane formed from a complex system of two or three arcs
(Ramírez-Espinosa et al., 1991; Mendoza and Suastegui, 2000).
However, no Cretaceous subduction-related accretionary prisms
have been identified within any of the terranes of the Guerrero
Composite Terrane.
Stage VI: Final Accretion of the Guerrero Composite
Terrane, and Development of a New Continental Arc
A major Late Cretaceous–early Paleogene orogenic phase is
recorded throughout Mexico, coeval to the Sevier and Laramide
orogenies in western North America. This event is associated with
the Mexican Fold and Thrust Belt of the Sierra Madre Oriental.
Apparently, final amalgamation of the Guerrero Composite Ter-
rane occurred during this orogenic event, and volcanic and sedi-
mentary rocks of the Teloloapan, Guanajuato, Zihuatanejo, and
Tahue Terranes were thrust over the calcareous platform rocks
of Oaxaquia and the Central, Cortes, and Mixteca Terranes. The
amount of tectonic transport apparently is significant, as xeno-
liths of Precambrian continental crust were found in Cenozoic
volcanic rocks that erupted onto accreted rocks of the Guanajuato
Terrane (Urrutia-Fucugauchi and Uribe-Cifuentes, 1999). Sig-
nificant tectonic transport is also suggested by the amount of
shortening that produced tight folding and major thrusting within
the northern Zihuatanejo Terrane and the Arcelia and Teloloapan
Terranes (Salinas-Prieto et al., 2000). In contrast, deformation of
Cretaceous rocks in the southern parts of the Zihuatanejo Terrane
formed wide regional anticlines, and some overturned folds and
minor thrust faults locally. The structures generally trend NW-
SE, although locally some structures trend N-S and E-W.
Santonian terrestrial sedimentation covers unconformably
the previously deformed arc assemblages of the Zihuatanejo
 Guerrero Composite Terrane of western Mexico
Terrane (Benammi et al., 2005). Synorogenic sedimentary basins
(Caracol Formation in Oaxaquia, and Mexcala Formation in the
Mixteca Terrane) containing clasts derived from the Guerrero
Composite Terrane suggest that these terranes were deformed
and exhumed by that time. In addition, synorogenic sedimenta-
tion overlaps the Arcelia and Teloloapan terranes (Miahuatepec
Formation), which suggests that these two terranes were also
amalgamated during the same orogenic event (Mendoza and Sua-
stegui, 2000; Guerrero-Suastegui, 2004). All these synorogenic
basins range in age from Turonian to Maastrichtian. In addition,
Paleocene granitoids along the coast cut the previously folded
units of the Zihuatanejo Terrane and suggest a Late Cretaceous–
early Paleogene deformation.
Therefore, final amalgamation of the Guerrero Composite
Terrane occurred between Santonian and Turonian–Maastrich-
tian time.
DISCUSSION
This section summarizes the stratigraphic, structural, and
geochemical data that support the proposed stages for the tec-
tonic evolution of western Mexico.
Stage I: Origin of the Basement of the Tahue Terrane
Exposures of pre-Cretaceous rocks in northwest Mexico
are scattered; thus contact relationships among them can only
be indirectly inferred (Figs. 1 and 4). Approximate distribution
of the contacts among the terranes of western Mexico (Caborca,
Cortes, and Tahue; Figs. 1 and 4) was outlined on the basis of
the geographic distribution of pre-Cretaceous outcrops and lat-
eral changes in the isotopic signatures of Cretaceous–Paleogene
granitoids (Valencia-Moreno et al., 2001). Thus the nature of the
contacts and the amount of displacement among different base-
ments are unknown. In this section the main stratigraphic units
that define the terranes are described following a NW to SE tran-
sect throughout the Paleozoic rocks of the Caborca, Cortes, and
Tahue Terranes (Guerrero Composite Terrane).
At the southern margin of the Caborca Terrane a thick shel-
fal limestone succession is exposed that contains Carboniferous–
Permian fusulinids and other shallow-marine fossil fauna (Stew-
art et al., 1990). These rocks are overridden by a north-verging
major thrust fault that places deeper marine sedimentary rocks
of the Cortes Terrane on the shelfal rocks of the Caborca Ter-
rane (Fig. 1; Coney and Campa, 1987; Poole and Madrid, 1988;
Stewart et al., 1990).
Basal metamorphic rocks are not exposed in the Cortes Ter-
rane, but its basement has been interpreted as thinned Proterozoic
rocks, perhaps the same as in the Caborca Terrane, or else Pro-
terozoic metamorphic rocks different from those of the Caborca
Terrane (McDowell et al., 1999; Valencia-Moreno et al., 1999;
Valencia-Moreno et al., 2001). The deep-marine sedimentary
rocks of the Cortes Terrane are sandstone and shale turbidites,
graptolitic shale, chert, and layered barite that range in age from
289
Ordovician to Devonian–Early Mississippian and were deformed
during the Mississippian (Poole and Madrid, 1988; Stewart et al.,
1990). These rocks are in turn overlain by Upper Carboniferous
and Permian turbidites (Fig. 5; Poole and Madrid, 1988; Stew-
art et al., 1990; Poole et al., 2005). They all were deposited in
a deep-marine environment and are interpreted to be part of the
Paleozoic continental slope-rise deposits of western North Amer-
ica (Poole and Madrid, 1988; Stewart et al., 1990). All these units
of the Cortes Terrane were deformed and thrust over the Caborca
Terrane by Late Permian to Early Triassic time, and they are
unconformably covered by Upper Triassic terrestrial and marine
sedimentary rocks (Stewart et al., 1990). Therefore, the Caborca
and Cortes Terranes were assembled by early Mesozoic time.
The nature of the contact between the Cortes and Tahue Ter-
ranes (the latter belonging to the Guerrero Composite Terrane)
has not been mapped in detail. It is inferred to be a thrust that
verges toward the north, and it is probably north of El Fuerte
town in Sinaloa State (Fig. 5), based on the northernmost expo-
sures of Cretaceous marine volcanic rocks of the Guerrero Ter-
rane (Servais et al., 1982; Henry and Fredrikson, 1987; Roldán-
Quintana et al., 1993; Freydier et al., 1995).
The oldest Paleozoic rocks of the Tahue Terrane (Guer-
rero Composite Terrane) are exposed in the area of El Fuerte
(El Fuerte Complex; Figs. 4 and 5). The El Fuerte Complex is
formed by marine rhyolitic to andesitic lava flows and volca-
niclastic rocks, interbedded with quartz-rich sandstone, shale,
and thin-bedded limestone (Mullan, 1978; Roldán-Quintana et
al., 1993; Poole and Perry, 1998). All these various components
are deformed and metamorphosed to greenschist facies (Mul-
lan, 1978; Roldán-Quintana et al., 1993). Sedimentary rocks of
the El Fuerte Complex contain Ordovician conodonts (Poole
and Perry, 1998). Preliminary geochemical analyses indicate a
calc-alkaline island-arc affinity for the volcanic rocks of the El
Fuerte Complex, and are similar to those from coeval Paleo-
zoic arc rocks in the Klamath Mountains of Northern California
(Lapierre et al., 1987). However, more detailed geochemical
and geochronological work needs to be done to constrain their
origin and relationships.
Upper Paleozoic deep-marine sedimentary rocks are ex-
posed south of El Fuerte, in San José de Gracia town, Mazat-
lán City, and other scattered localities in Sinaloa State (Figs. 4
and 5). These rocks belong to the San José de Gracia Formation
(Carrillo-Martínez, 1971; Gastil et al., 1991; Arredondo-Guerrero
and Centeno-García, 2003) and are made up of quartz-rich sand-
stone and shale turbidites, thin-bedded calcareous debris flows,
black shale, and chert. The turbidites contain olistoliths of lime-
stone with chert nodules, which in turn contain Middle Pennsyl-
vanian to Early Permian fossils at the San José de Gracia locality
(Carrillo-Martínez, 1971; Gastil et al., 1991). The San José de
Gracia Formation has been interpreted as deposits in a deep-
marine environment (Gastil et al., 1991). The contact between
the El Fuerte Complex and the San José de Gracia Formation
is not exposed. However, major differences in deformation and
metamorphism (turbidites of the San José de Gracia Formation
 Hermosillo

e
a Terran
Caborc a n e
e rr
Cortes T

Caborca Terrane
Limestone (Lower Cretaceous)
Ciudad Shallow marine limestone
Obregón S O N O R A and shale (Paleozoic)
Cortes Terrane
Navojoa ne
erra e Terrestrial and marine
27°
s T rran sedimentary rocks (Triassic)
rte Te
Co ue
T a h Deep Marine turbidites, chert
and barite (Paleozoic)

Sonobari
El Fuerte

C
H
San Jose de Gracia
N

I
26°

H
U
A
Sinaloa de Leyva-Porohui

H
U
A
D
U
R
25° A

Overlapping units
N

CULIACÁN
Recent sediments
G
O

Recent basalts
Miocene-Pliocene
Volcanic and sedimentary
(conglomerate-sandstone)
Granitoids (Paleocene-Oligocene)
Upper Cretaceous-Oligocene
Tarahumara Formation and Lower and Upper 24°
Volcanic Supergroup

Tahue Terrane (Guerrero Composite Terrane)

Granitoids (Upper Jurassic-Lower Cretaceous)
Cretaceous Guerrero Arc
MAZATLÁN
Marine volcanic and sedimentary rocks
San Francisco Gneiss (Triassic) 23°
San Jose de Gracia Fm. (Carboniferous)
El Fuerte Complex (Ordovician)
km
0 25 50 100

109° 108° 107° 106°

Figure 4. Geologic map of Sinaloa and the southern Sonora states, showing the geology of Caborca, Cortes, and Tahue Terranes (after
Carrillo-Martínez, 1971; Mullan, 1978; Gastil et al., 1978; Henry and Fredrikson, 1987; Stewart and Roldán-Quintana, 1991; Ortega et
al., 1992; and our own work).
 WEST EAST
western Oaxaquia
Maastrichtian Caracol Fm.
Campanian shale and sandstone
foreland-basin fill
Santonian GUERRERO COMPOSITE TERRANE
Coniacian
Turonian Tahue Terrane Zihuatanejo Terrane Guanajuato Terrane Central Terrane
Cenomanian
La Borda, Limestone
Albian El Saucito and Titonian to Aptian
ChilitosFms. Tuna Mansa
Aptian Lower Cretaceous diorite
Aptian? Cerro Pelon Transitional Limestone
volcanic turbidites tonalite
Barremian volcanic turbidites Santa Ana from continental to Tithonian to Aptian
limestone, tuff and calcareous debris flows, dike swarm marine shale,
chert, pillowed lavas tuff and chert Esperanza Fm. sandstone, evaporites,
Hauterivian and dikes deformed Transitional
pillowed lavas and dikes limestone, chert, limestone
Valanginian and in part deformed and in part volcaniclastic turbidites Kimmeridgian-Tithonian from continental to
Berriasian metamorphosed metamorphosed felsic lavas 146 Ma marine shale,
Tithonian felsic and La Luz sandstone, evaporites,
basic lavas basaltic flows La Perlita limestone
Kimmeridgian and tuff limestone
150-147 Ma Albian-Aptian felsic lavas and Oxfordian-Tithonian
Oxfordian
Callovian
massive limestone, volcaniclastics felsic lavas and
Bathonian
black limestone continental volcaniclastic rocks
Bajocian unconformity shales and sanstone Oxfordian continental
Aalenian
unconformity unknown age
Arperos Fm.
Toarcian Hauterivian-Tithonian? Felsic lavas and
Pliensbachian deep marine volcaniclastic volcaniclastic rocks
unconformity turbidites,chert, limestone 189-172 Ma
Sinemurian

Hettangian
unconformity unconformity
Rhaetian Francisco gneiss
Upper Triassic Taray Fm. La Ballena Fm.
Norian Zacatecas Fm. Paleozoic-Triassic? Carnian-Norian
migmatized gneisses Norian-Carnian
and amphibolites matrix is quartz-rich turbidites deformed quartz-rich
matrix is quartz-rich turbidites blocks of pillow lavas, chert, turbidites
Carnian blocks of pillow lavas serpentinite and limestone continent-slope facies
accretionary complex accretionary complex
Ladinian El Fuerte Complex ?
Early Triassic Ordovician
metamorphosed
volcanic and quartz-rich linked by provenance (Potosí Fan)
Paleozoic turbidites, limestone,
chert, basalts and rhyolites
Figure 5. Simplified stratigraphic columns for Oaxaquia and terranes north of the Transmexican Volcanic Belt. The columns are in an east-west order. They include the Tahue Terrane,
which is part of the Guerrero Composite Terrane. Vertical scale shows the age range (in Ma).
292 Centeno-García et al.
are strongly deformed but not metamorphosed) indicate that the
contact is probably an angular unconformity.
Both units of the Tahue Terrane (El Fuerte Complex and
San José de Gracia Formation) are important because they can
constrain the paleogeography of the northern Guerrero Terrane.
Preliminary single-grain, detrital zircon geochronology from
the quartz-rich sandstone from the turbidites of the San José de
Gracia Formation shows populations that have a North American
affinity (Centeno-García et al., unpublished data) and are similar
to those from Paleozoic rocks in Baja California, in the Cortes
Terrane, and in Nevada (Gehrels et al., 2002).
The stratigraphy, geochemistry, and provenance of the Paleo-
zoic rocks suggest that the Tahue Terrane (Guerrero Composite
Terrane) was linked to the tectonic evolution of the western conti-
nental margin of North America, probably up to Permian–Triassic
time. After that, there were major differences in the composition
of the Mesozoic sedimentary cover of the Caborca-Cortes Ter-
ranes with respect to that of the Tahue Terrane. Therefore, it is
likely that a fragment of previously accreted island-arc and con-
tinent-margin assemblages drifted from the continental margin
sometime in the early Mesozoic.
Contact relationships between the Paleozoic sedimentary
rocks of the San José de Gracia Formation (Tahue Terrane) and
the Triassic subduction-related complex of the Zihuatanejo Ter-
rane are unknown because the contact is covered by younger
rocks. However, the Tahue and Zihuatanejo Terranes share simi-
lar Cretaceous volcanic and sedimentary cover.
Stages II and III: Triassic Potosi Fan and Its Accretion to
the Continental Margin
There are few exposures of Triassic rocks in Mexico, and
they are limited to the Caborca and Cortes Terranes, western
Oaxaquia, the Central and Zihuatanejo Terranes, and a small out-
crop in the Vizcaíno Peninsula in Baja California. Triassic rocks
have not been found in the Mixteca Terrane or in other terranes
of Mexico. In this section we briefly describe Triassic rocks of
the Cortes and Tahue Terranes (Barranca Group and Francisco
Gneiss) and focus on the marine Triassic rocks of Oaxaquia (La
Ballena Formation), the Central Terrane (Taray Formation), and
the Guerrero Composite Terrane (Zacatecas Formation, and the
Arteaga and Las Ollas Complexes).
Barranca Group and Francisco Gneiss
Triassic (Carnian–Norian) sedimentary rocks of the Cortes
Terrane are made up of fluvial sandstone and shale that contain
abundant coal layers (Barranca Group; Stewart and Roldán-
Quintana, 1991). These sediments were deposited unconform-
ably on previously deformed Paleozoic deep-marine rocks. The
Triassic fluvial deposits change transitionally up the column to
shallow-marine siliciclastic deposits. These rocks have no evi-
dence of contemporaneous volcanism. In contrast, Triassic rocks
of the Tahue Terrane (Guerrero Composite Terrane) are made up
of metamorphosed igneous rocks of the Francisco Gneiss near
Sonobari (Figs. 4 and 5; Mullan, 1978; Keppie et al., 2006).
The Francisco Gneiss is made up of migmatized gneisses and
amphibolites that have within-plate and continental tholeiite geo-
chemical signatures (Keppie et al., 2006). This suggests that the
Tahue and Cortes terranes may have been geographically sepa-
rated by that time.
La Ballena Formation
Triassic rocks of Oaxaquia crop out on its western margin,
near its boundary with the Guerrero Composite Terrane (Figs. 1
and 6). They are grouped as the La Ballena Formation (Silva-
Romo, 1993; Silva-Romo et al., 2000), and their largest expo-
sures are in the Peñón Blanco, Charcas, and Real de Catorce
areas (Fig. 6; Silva-Romo, 1993; Tristán-González and Torres-
Hernández, 1994; Barboza-Gudiño et al., 2004). The La Ballena
Formation is made up of quartz-rich sandstone and shale, and
scarce conglomerates deposited as small channel-fill lenses. The
sedimentary structures of these Triassic rocks indicate deposi-
tion mostly by turbidity currents, although some debris flows
and large slumps are present. This sequence contains abundant
trace fossils and ammonites and bivalves of Late Triassic (Car-
nian) age at the Peñón Blanco and Charcas areas (Cantu-Chapa,
1969; Silva-Romo et al., 2000; Bartolini et al., 2002). Sedimen-
tary structures and fossil fauna suggest that the deposition of this
unit occurred in a submarine fan that developed on an external
platform or continental slope setting. These rocks form part of
the Potosi Submarine Fan (Centeno-García, 2005). The origi-
nal thickness is unknown, but up to 4640 m was penetrated by
exploration drilling without reaching the base of the succession
(López-Infanzón, 1986).
Taray Formation
Similar marine siliciclastic rocks crop out at the Pico de
Teyra region in the Central Terrane (Figs. 5 and 6). They belong
to the Taray Formation, made up of highly deformed quartz-rich
turbidites (sandstone and shale) interbedded with some black
chert and scarce detrital limestone that contains fragments of cri-
noids, gastropods, corals, bivalves, and bryozoans (Diaz-Salgado
et al., 2003). The Taray siliciclastic turbidites form a matrix
within which blocks of black and green chert, pillow basalts, ser-
pentinite, and crystallized limestone can be found (Figs. 5 and 6;
Diaz-Salgado et al., 2003). The age of this unit remains undeter-
mined; however, there are reports of fusulinids from one of the
limestone blocks (Anderson et al., 1990). The youngest detrital
zircons collected from the sedimentary matrix are Late Permian
in age (Diaz-Salgado et al., 2003). There is also a report of molds
of bivalves of possible Carnian age (Barboza-Gudiño et al., 1999;
Bartolini et al., 2002). Thus deposition of the sedimentary matrix
should have occurred between the Late Permian to the Late Trias-
sic. The Taray Formation has a block-in-matrix structural style,
formed by centimeter-size blocks to blocks of hundreds of meters
in size, all in a highly sheared sedimentary matrix. This charac-
teristic is typical of a subduction accretionary complex (Ander-
son et al., 1990, 2005; Diaz-Salgado et al., 2003).
 Guerrero Composite Terrane of western Mexico 293

102° 101° 100°

Cenozoic cover
Caopas Oaxaquia and Central Terrane
U/Pb 158 Ma Concepción
Ju-K Calcareous Platform
del Oro
Pico Ju felsic and intermediate lava
de flows, epiclastic rocks and redbeds
Teyra
Tr Quartz-rich shale and sandstone
turbidites (Potosí Fan)
24° Huiznopala Tr(?) Accretionary complex
Granjeno blocks of pillow basalt, ultramafic,
chert, and marble in a sedimentary
quartz-rich matrix
Real de Matehuala
Central Catorce
Huizachal
Proterozoic and Paleozoic
U/Pb190 Ma metamorphic and sedimentary
terrane rocks

Guanajuato Terrane
Ju-K pillow basalts, deep marine
23° Charcas volcanic turbidites, chert, and mafic
and ultramafic plutons

Oaxaquia Zihuatanejo Terrane

Zacatecas Ju-K pillow basalts, volcanic
Peñón turbidites, detrital limestone
U/Pb 147-150 Ma Blanco
Tr Accretionary complex
blocks of pillow basalt, in a
San Luis Potosí sedimentary quartz-rich matrix
22°
Gu ter
an ran

Zihuatanejo 100 km
aju e

terrane
at
o

U/Pb146 Ma
Guanajuato

Figure 6. Geologic map of central Mexico, showing main stratigraphic units of Oaxaquia and the Central, Zihuatanejo, and Guanajuato Terranes
(modified from Ortega et al., 1992). Tr—Triassic; Ju—Jurassic; K—Cretaceous.

Zacatecas Formation
The oldest rocks of the Zihuatanejo Terrane in its northern-
most exposure are Triassic in age as well (Fig. 6). They make
up the Zacatecas Formation, which crops out in a small tectonic
window at the western margin of Zacatecas City (Fig. 6; Burck-
hardt and Scalia, 1906; Ranson et al., 1982; Cuevas-Pérez, 1983;
Monod and Calvet, 1991). This formation is made up of quartz-
rich turbidites (sandstone and shale) that contain blocks of pil-
low basalts that have MORB geochemical signatures (Fig. 7;
Centeno-García and Silva-Romo, 1997). The Zacatecas Forma-
tion contains fossil ammonites and bivalves of Late Triassic (Car-
nian) age (Burckhardt and Scalia, 1906; Bartolini et al., 2002).
Its contact with the La Borda Formation of Late Jurassic(?)–Cre-
taceous age is inferred to have been originally an unconformity,
but it was sheared and detached during Late Cretaceous thrust-
ing and folding (Fig. 3). Rocks of the Zacatecas Formation show
structures associated with two distinct deformational events, one
of them prior to the deformation that is recorded in the Creta-
ceous rocks as well. The small size of the outcrop prohibits con-
straints on the tectonic origin of the Zacatecas Formation, but its
lava flows and siliciclastic turbidites are similar to those from the
Arteaga Accretionary Complex, which is exposed in the southern
part of the Zihuatanejo Terrane.
Arteaga Complex
More exposures of Triassic(?) rocks are found in the south-
ern part of the Zihuatanejo Terrane (Fig. 7). Their largest outcrops
are located in the Arteaga, Placeres del Oro, and Tiquicheo areas
(Arteaga Complex) and near Zihuatanejo City (Las Ollas Complex)
(Fig. 8; Centeno-García et al., 1993a, 1993b; Talavera-Mendoza
et al., 1995; Mendoza and Suastegui, 2000). The Arteaga Com-
plex is made up of quartz-rich turbidites (sandstone and shale),
 WEST EAST
Mixteca Terrane
GUERRERO COMPOSITE TERRANE (western Continent-margin)
Maastrichtian
Mexcala Fm.
Campanian Zihuatanejo Terrane Arcelia Terrane TeloloapanTerrane shale and sandstone
Santonian Coastal region Huetamo region foreland-basin fill
Coniacian subaerial to marine
Turonian
shale, sandstone,
Cenomanian evaporites, limestone, Comburindio and
Albian
basaltic-andesitic Mal Paso Fms.
lava flows shallow marine
Aptian Albian-Cenomanian Cuautla and
limestone,volcaniclastic Limestone Morelos Fms.
subaerial few marine rocks, few andesitic-basaltic Aptian-Albian Limestone
Barremian volcaniclastic rocks, lava flows Villa de Ayala Fm. Aptian-Albian
felsic lavas Aptian Albian MORB basaltic pillow unconformity
Hauterivian Aptian lavas, IA lavas and Taxco Schist,
Angao and San felsic and
Valanginian volcaniclastic rocks Zicapa and
Lucas Fms. dikes, chert, deep andesitic
Berriasian and limestone volcanic turbidites, chert marine volcaniclastic Chapolapa
Tithonian age not well felsic and basic lavas turbidites lavas and felsic lavas and
Kimmeridgian constrained volcaniclastics volcaniclastics
unconformity shallowing upward Early Cretaceous shallow marine in the
Oxfordian Neocomian continental affinity
Callovian
Cuale east, deep marine in 127-133 Ma
Bathonian felsic lavas and the west
Bajocian Tumbscatío granitoids Las Lluvias
Aalenian
145-137 Ma
163-155 Ma Ignimbrite
Toarcian 168-179 Ma
unconformity
Pliensbachian

Sinemurian

Hettangian
Rhaetian
Arteaga Complex
Norian Norian-Carnian
matrix is quartz-rich turbidites Acatlan Complex
Carnian blocks of pillow lavas, chert, mid-Paleozoic
serpentinite and limestone poly deformed sedimentary
Ladinian accretionary complex and volcanic rocks
Early Triassic continent-continent collision
Paleozoic

Figure 7. Simplified stratigraphic columns for the terranes described in the text that are south of the Transmexican Volcanic Belt, except for the Guanajuato Terrane, and include the
Mixteca Terrane and the Teloloapan, Arcelia, and Zihuatanejo Terranes (Guerrero Composite Terrane). Vertical scale shows the age range (in Ma). MORB—mid-oceanic-ridge basalt;
IA—island arc.
 Zihuatanejo Terrane Teloloapan Terrane Mixteca Terrane Overlapping assemblages
Huetamo area, Cretaceous Lower Cretaceous arc assemblages Aptian-Albian Calcareous Miahuatepec Formation
arc assemblage, mostly marine shallow marine andesitic to basaltic Platform Upper Cretaceous clastic rocks
volcaniclastic rocks, limestone and lava flows and volcaniclastic rocks, foreland basin-fill (overlaps
some pryroclastic and lava flows massive and reefal limestone (east) Lower Cretaceous continental Arcelia and Teloloapan terranes)
Coastal Cretaceous arc assemblage deep marine lava flows, calcareous arc assemblages, marine and Mexcala Formation
shallow marine and terrestrial rhyolite, debris flows and volcanic turbidites (west) terrestrial rhyolite to andesite lava Turonian-Maastrichtian clastic rocks
andesite and some basalt, limestone, flows and epiclastic rocks, foreland basin-fill (overlaps
volcaniclastic and basement-derived Arcelia Terrane quartz-rich clastic rocks Mixteco and Teloloapan terranes)
clastic rocks Cretaceous IAB pillow lavas, Paleozoic Acatlán Complex CutzamalaFormation
Middle to Upper Jurassic fine-grained volcanic turbidites Santonian-Maastrichtian red beds
plutons and chert (deep marine) (overlaps Zihuatanejo and Arcelia
Las Ollas Subduction Complex Xolapa Terrane terranes)
Cretaceous MORB pillow lavas,
(age unknown) fine-grained volcanic turbidites Jurassic to Cretaceous Upper Cretaceous to Paleogene
and chert (deep marine) migmatites, gneisses, and plutons granitoids
Arteaga Subduction Complex
(Triassic) Cenozoic volcanic and
sedimentary rocks
104° 103° 102° 101°

Zapotitlán Nevado de Colima

Observed terrane boundary
(thrust fault)
Tecalitlán
Minatitlán
Zitácuaro inferred terrane boundary
Jilotlán (thrust fault)
Colima Pijuamo inferred terrane boundary
Tepalcatepec (strike-slip fault)
Manzanillo Nueva
Italia Rivers Roads
Tejupilco 19°
Tecoman
Z I H U ATA N E J O T E R R A N E State limit
Coalcoman
Aguililla 100°
TERRANE

U/Pb 163+3 Ma
Aquila Infiernillo Dam Huetamo
Taxco
Tumbiscatío
e

Balsas
Ar/Ar 152-158 Ma River Teloloapan
N Arteaga Iguala
Arcelia
RRANE
marine

deep marin

iver
PA as R
CIF Bals

IC
shallow

OC MIXTECA TERRANE
18°
ARCELIA

EA
N Playa Azul Zicapa
Lázaro Cárdenas
TELOLOAPAN TE

10 0 20 40
Zihuatanejo
km Chilpancingo
Chapolapa
XOLAPA TERRANE

Figure 8. Geologic map of southwestern Mexico, showing the simplified geology of the Mixteca, Teloloapan, Arcelia, and Zihuatanejo Terranes (after Campa and Ramirez, 1979;
Ortega et al., 1992; Talavera-Mendoza et al., 1995; Corona-Chávez and Israde-Alcántara, 1999; Mendoza and Suastegui, 2000; Centeno-García et al., 2003). IAB—island-arc basalt;
MORB—mid-oceanic-ridge basalt.
296 Centeno-García et al.
black and green chert, and mafic tuff that form a matrix that con-
tains blocks and slabs of pillow basalts, diabase, banded gabbros,
chert, and limestone, all deformed in a block-in-matrix structural
style (Centeno-García et al., 2003). Chert layers contain radio-
larians of Triassic (Ladinian–Carnian) age (Campa et al., 1982).
Pillow basalts and gabbros have oceanic geochemical signatures
(MORB; Centeno-García et al., 1993a; Centeno-García et al.,
2003). Sedimentary structures preserved in some exposures of
unmetamorphosed turbidites, along with the affinity of the few
fossils found in the sedimentary rocks of the matrix, suggest that
the sequence was deposited in a deep-ocean environment. Appar-
ently the quartz-rich turbidites were contemporaneous with oce-
anic magmatic activity, as they are interbedded with volcaniclas-
tic rocks (Centeno-García et al., 2003). The block-in-matrix style
of deformation of the Arteaga Complex, as well as its lithological
associations, indicate that it was formed in a subduction accre-
tionary prism. Metamorphism ranges from none to amphibolite
facies; blueschist facies has not been found in the area.
Las Ollas Complex
The Las Ollas Complex forms part of the Zihuatanejo
Terrane and is exposed near Zihuatanejo City (Figs. 7 and 8;
Talavera-Mendoza, 2000). This complex is a tectonic mélange
formed by highly sheared blocks of metabasalt, banded and mas-
sive gabbro, metadolerite, ultramafic rocks, and shale and quartz-
rich sandstone (Talavera-Mendoza, 2000). These blocks are
enveloped in a highly sheared clastic (quartz-rich sandstone) or
serpentinitic matrix (Talavera-Mendoza, 2000). Blueschist facies
were reported by Talavera-Mendoza (1993, 2000). Geochemical
compositions of the basalts are typical of MORB and primitive
oceanic-arc magmas (Talavera-Mendoza, 2000). 40Ar/39Ar and
K/Ar ages obtained from amphibole from several metagabbro
blocks range from 223 Ma to 96 Ma (Permian to early Ceno-
manian) (Delgado, 1982; A. Iriondo, 2003, personal commun.).
This has been interpreted to be the subduction complex of the
Cretaceous arc (Vidal-Serratos, 1991; Talavera-Mendoza, 1993);
however, its contact relationships with Cretaceous arc-related
rocks, and similarities with the Arteaga Complex, suggest an ear-
lier origin.
Quartz-rich turbidites from the La Ballena Formation of
Oaxaquia (continental Mexico), the matrix of the Taray Forma-
tion of the Central Terrane, and the Arteaga and Las Ollas Com-
plexes and the Zacatecas Formation of the Zihuatanejo Terrane
(Guerrero Composite Terrane) have similar and distinctive com-
positions and detrital-zircon age populations (Centeno-García
et al., 2005; Talavera-Mendoza et al., 2007). Therefore, Triassic
sedimentation of the central and western terranes of Mexico is
linked by provenance. The youngest zircon age populations from
all the samples (latest Permian) are much older than the depo-
sitional ages of the turbidites (Carnian–Norian), which means
that there was no active volcanism at that time. In other words,
there is no evidence of Late Triassic continental arc volcanism in
Mexico. Zircon age populations of the Potosi Fan are different
from those of Triassic quartz-rich sandstone from the Caborca
and Cortes Terranes (González-León et al., 2005) but are simi-
lar to those from Triassic fluvial sedimentary rocks of Arizona
(Anderson, 2006). This suggests that at the end of the Triassic
the terranes of central and western Mexico may have been to the
north of their present locations.
Based on this evidence, we propose that the margin of the
western paleo-continent of Mexico was passive or rifting at the
end of the Triassic. This passive margin received abundant clastic
sedimentation, forming the large Potosi Fan. Sediments of this
fan were deposited on oceanic crust (Arteaga Basin in Fig. 3).
When subduction started, slivers of the ocean floor were tectoni-
cally mixed with the already existing passive-margin quartz-rich
turbidites that were forming the Taray and Zacatecas Formations
as well as the Arteaga and Las Ollas Complexes. Whether the
ocean basin that was covered by sediments of the Potosi Fan was
an active marginal oceanic basin, a marginal backarc basin, or an
open ocean–continent flank is still uncertain. The only potential
evidence of association of the Potosi Fan sediments with tholei-
itic oceanic volcanism is the volcaniclastic rocks interbedded
with the siliciclastic turbidites in the Arteaga Complex, as the
volcaniclastic rocks have geochemical signatures between primi-
tive island arc and MORB (Centeno-García et al., 2003).
At least two phases of deformation are found in all the Trias-
sic rocks of Oaxaquia and the Central and Zihuatanejo Terranes.
The first event comprised strong shearing and tight folding, and
the block-in-matrix structures. A second event was recorded
only in the Arteaga Complex. This event deformed the Jurassic
granitoids as well, and it is characterized by a mylonitic fabric.
The third event was common to all the Triassic units and is also
recorded in the Jurassic and Cretaceous cover sediments, and it
is characterized by axial cleavage, open to tight folding, reverse
faulting, and thrusting.
The time of accretion of the Central Terrane with Oaxaquia
is assumed to have been prior to the Middle Jurassic, because the
La Ballena Formation of Oaxaquia and the Taray Formation of
the Central Terrane were deformed and locally metamorphosed
prior to deposition of Upper Jurassic terrestrial volcanic and clas-
tic formations (Tristán-González and Torres-Hernández, 1994;
Jones et al., 1995; Silva-Romo et al., 2000). The Zihuatanejo
Terrane (Guerrero Composite Terrane) was also accreted at that
time, because the Arteaga Complex is cut by granitoids of Middle
Jurassic age as well (Centeno-García et al., 2003).
The subduction zone that formed the Taray and Zacatecas
Formations, and the Arteaga and Las Ollas Complexes, was
probably constructed along the continental margin of Oaxaquia
in Early Jurassic time. Whether the subducting slab was dipping
to the east or to the west has not been determined.
Stage IV: Jurassic Continental Arc of Western Mexico
Erosion and exhumation of the accreted continental slope
sediments and the accretionary complexes occurred prior to the
initiation of Middle to Late Jurassic magmatism. This is indicated
by the major angular unconformity that separates the Jurassic arc
 Guerrero Composite Terrane of western Mexico
succession from the deformed Triassic rocks of Oaxaquia and
the Central and Zihuatanejo Terranes. Jurassic arc magmatism
has also been identified in the Mixteca Terrane. The Jurassic arc
rocks have different names at different locations; they are hereby
described by their occurrence in different terranes:
Nazas, Huizachal, and La Joya Formations in Oaxaquia
The La Ballena Formation (Oaxaquia) is unconformably
overlain by the volcanic rocks and red beds of the Nazas Forma-
tion in the Peñón Blanco, Charcas, and Real de Catorce areas
(Figs. 5 and 6; Silva-Romo, 1993; Tristán-González and Torres-
Hernández, 1994; Barboza-Gudiño et al., 2004). The Nazas For-
mation is made up of dacitic and minor rhyolitic and andesitic
lava flows and pyroclastic flows, dikes, and porphyritic shallow
intrusives. The volcanic rocks are interbedded with conglomer-
ate, sandstone, and scarce paleosols. Conglomerate is formed
mostly by volcanic clasts and a few clasts of sandstone and shale
derived from the underlying La Ballena Formation. The volca-
niclastic conglomerate and sandstone form lens-shaped bedding
with low-angle cross-bedding, interbedded with some debris
flows, suggesting that they were deposited in a terrestrial (fluvial
and alluvial fan) environment.
Although their age has not been well constrained at all
the exposures, there is a report of U/Pb ages as old as 189 Ma
at a subaerial volcanic-sedimentary succession in Huizachal
(Huizachal Formation; Fastovsky et al., 2005), which might not
belong to the same volcanic arc event (Figs. 5 and 6). Rocks of
the Nazas Formation at Real de Catorce yielded U/Pb ages of
172 ± 5 Ma (Barboza-Gudiño et al., 2004). The Nazas Forma-
tion changes transitionally upward to shallow-marine volcanicla-
stic rocks, some evaporites, and thin-bedded limestone, which in
turn become a thick limestone succession in the Peñón Blanco
and Charcas areas (Fig. 6). The basal part of this limestone suc-
cession contains late Oxfordian–Kimmeridgian fossil faunas
(Centeno-García and Silva-Romo, 1997). In contrast, there is
an internal angular unconformity in the Real de Catorce local-
ity (Fig. 6), which separates in two units, the terrestrial volcanic
and sedimentary successions (Nazas and La Joya Formations;
Barboza-Gudiño et al., 2004). The upper La Joya Formation
changes transitionally upward to shallow-marine volcanic sand-
stone and shale interbedded with thin limestone strata. The oldest
fossils reported from the base of the limestone succession in Real
de Catorce are Oxfordian in age (Barboza-Gudiño et al., 2004).
Caopas, Rodeo, and Nazas Formations of the Central Terrane
The volcanic cover of the Taray Formation (Central Terrane;
Figs. 5 and 6) belongs to the Caopas, Rodeo, and Nazas Forma-
tions (Córdoba-Méndez, 1964; López-Infanzón, 1986; Jones et
al., 1995). The Rodeo and Nazas Formations are lateral equiva-
lents of the same rocks but named differently in separate out-
crops (Díaz-Salgado, 2004). Both units are made up of rhyolitic
to andesitic lava flows and dikes, and pyroclastic deposits that are
interbedded with fluvial sedimentary rocks, mostly sandstone and
conglomerate (Anderson et al., 1990, 1991; Jones et al., 1995;
297
Díaz-Salgado, 2004). The Caopas Formation was formed by
shallow porphyritic intrusives. Felsic volcanic rocks of the Rodeo
Formation yielded a K-Ar age of 183 Ma (López-Infanzón, 1986),
and the Caopas Formation a U/Pb age of 158 Ma (Jones et al.,
1995). Terrestrial volcaniclastic rocks of the Rodeo Formation
are interpreted to have been deformed previous to the deposition
of late Oxfordian limestone (Anderson et al., 1991; Bartolini et
al., 2002). However, in another locality nearby the Nazas Forma-
tion changes transitionally upward to shallow-marine calcareous
rocks that range in age from Late Jurassic to Late Cretaceous
(Córdoba-Méndez, 1964; Díaz-Salgado, 2004).
All these volcanic-sedimentary units are interpreted in this
work as the first overlapping succession that stitches the Central
Terrane with Oaxaquia. The Caopas and Rodeo Formations, as
well as the Nazas Formation, are interpreted as continental intra-
arc assemblages (Jones et al., 1995).
Las Lluvias Ignimbrite of the Mixteca Terrane
Jurassic arc volcanism was also recorded in the Mixteca
Terrane in which ignimbrites, interbedded with fluvial and
shallow-marine siliciclastic deposits, yielded U/Pb ages of 168.2
± 1.2 Ma, 177.3 ± 1.5 Ma, and 179.1 ± 1.5 Ma (Campa and Iri-
ondo, 2003).
Cuale Assemblage and Tumbiscatio Granitoids of the
Zihuatanejo Terrane
Evidence of coeval Jurassic magmatism has been found
in two localities in the Zihuatanejo Terrane (Guerrero Com-
posite Terrane). One of the exposures is NE of Puerto Vallarta
City, in the Cuale mining district, and the other locality is in the
Tumbiscatio region, both along the Pacific Coast (Figs. 7 and
8). Rocks at Cuale contain volcanogenic massive sulfide (VMS)
deposits and are composed of submarine rhyolitic lavas and tuffs,
volcanic sandstone with evolved-arc geochemical affinity (Bis-
sig et al., 2003), and shale that yielded U/Pb ages of 162.4 and
155.9 Ma (Bissig et al., 2003). These rocks are strongly deformed
and partially metamorphosed, and their contact with Cretaceous
unmetamorphosed marine volcanic and sedimentary successions
has not been determined.
Two Jurassic granitoids crop out in the Tumbiscatio region.
They were emplaced in previously deformed sedimentary rocks
of the Arteaga Complex, and vary in composition from grano-
diorite to granite to quartz monzonite. Their geochemical com-
positions are typical of calc-alkaline subduction-related gran-
ites, which are more evolved than granitoids of Cretaceous and
Cenozoic ages from the same area. Both granitoids show intense
shearing and internal deformation. Grajales and López (1984)
obtained one K/Ar date of Late Jurassic age (158 Ma). U/Pb
isotopic analysis yielded a 163 Ma age, and Ar/Ar ages are 158
and 152.4 Ma (Centeno-García et al., 2003). The igneous rocks
of the Cuale and Tumbiscatio regions have strong similarities
in geochemical composition and age with volcanic rocks of the
Central Terrane (Caopas, Rodeo, and Nazas); thus we suggest
that they probably originated in the same volcanic arc. Therefore,
298 Centeno-García et al.
the Arteaga Complex was probably accreted to the continental
margin, either near or along the strike from central Mexico.
The Jurassic volcanic event did not produce a thick strati-
graphic column and apparently did not have large volumes of
volcanic products. The column changes transitionally upward to
shallow-marine calcareous rocks. Therefore, the lithologic asso-
ciations and vertical facies changes of this volcanic-sedimentary
event are similar to those of a continental rift. However, the scarce
geochemical analyses from its volcanic rocks suggest an arc set-
ting (Jones et al., 1995). These rocks have been interpreted as the
southern continuation of the Jurassic continental arc that devel-
oped along the southwestern margin of North America (Jones et
al., 1995).
It has been proposed that major strike-slip faults were prob-
ably active during the arc activity (Mojave-Sonora Megashear;
Jones et al., 1995). This could explain the fact that the Potosi Fan
is south of its possible continental fluvial correlative in Arizona,
as well as the southward displacement of the Tahue Terrane.
Whether or not the Jurassic volcanic event was coeval with
a major transform fault has not been well documented. The
evidence in favor of an important synsedimentary deformation
involving major extension is as follows: (1) Minor synsedimen-
tary normal faults and local angular unconformities are present
within the Jurassic arc volcanic and sedimentary successions,
and pre-Cretaceous mylonitic shearing is recorded in the Juras-
sic granitoids of the Tumbiscatio region (Zihuatanejo Ter-
rane). (2) Arc magmatism suddenly ceased in the Central Ter-
rane and Oaxaquia, followed by a rapid transgression recorded
in a few meters of transitional sedimentation. (3) Subsidence
rates apparently were significant during the early stages of the
Oxfordian–Kimmeridgian marine sedimentation, because the
calcareous rocks show evidence of deeper sedimentation at
higher stratigraphic levels as well as overall rapid sedimenta-
tion. (4) Although fault planes have been obliterated by younger
deformational events, they have been inferred by the rapid lat-
eral changes in thickness and facies of the calcareous succession
through the interval from the end of the Jurassic to the Early
Cretaceous. (5) In addition, major regional lineaments have
been identified in central and eastern Oaxaquia, including the
San Marcos and La Babia Faults (Fig. 1) (Goldhammer, 1999;
Chávez-Cabello et al., 2005).
Stage V: Rifting of the Guerrero Terranes and Formation
of a Complex Arc System
In this section we list the main stratigraphic features of the
volcanic-sedimentary successions of the Guerrero Composite
Terrane and the Mixteca Terrane. Arc volcanism was absent in
Oaxaquia and the Central Terrane through the end of the Juras-
sic and the Cretaceous. During this period, oceanic crust was
emplaced toward the east of Oaxaquia in the Gulf of Mexico,
and continuous subsidence prevailed throughout the Early Creta-
ceous, resulting in a thick calcareous platform that covered all the
Central Terrane and Oaxaquia.
Although much detailed work needs to be done in order to
reconstruct the paleogeography of western Mexico during the Cre-
taceous, the available evidence indicates three important features:
1. Magmatism prograded generally east to west through
time, from the oldest ages in the Oaxaquia and Mixteca
Terranes to the youngest ages in the coastal Zihuatanejo
Terrane. There is some overlap of age ranges for the vol-
canism among the different terranes, e.g., volcanism of
the Mixteca Terrane overlaps in age with part of the vol-
canism of the Teloloapan Terrane (Guerrero Composite
Terrane). However, on a large scale, Albian–Cenomanian
volcanism is absent in the Mixteca and the Teloloapan
Terranes, and it is widespread in the coastal region of the
Zihuatanejo and Arcelia Terranes.
2. Magma chemistry changed through time toward a more
primitive melt. The Middle Jurassic volcanic and intru-
sive rocks in all the terranes show mostly felsic-evolved
continental-arc geochemical signatures, including the
Mixteca Terrane and Oaxaquia. In contrast the Cretaceous
volcanic rocks of the Guerrero Composite Terrane range
from tholeiitic basalts to andesites, with few rhyolites.
They show more primitive island-arc (IA) geochemical
signatures overall, and some even have MORB to oceanic-
island basalt (OIB) signatures. The Mixteca Terrane is the
exception to this trend; its magmatism remained evolved,
with continental arc signatures, into the Cretaceous.
3. Within different assemblages of the Guerrero Composite
Terrane there are major differences in the stratigraphy,
sediment composition, and depositional environments.
And the Guerrero Composite Terrane overall is different
from the volcanic-sedimentary rocks of the Mixteca Ter-
rane to the east. In their present distribution, areas with
shallow-marine and terrestrial volcanic-sedimentary suc-
cessions alternate with areas with deep-marine volcanic-
sedimentary successions, and suggest a complex paleo-
geography for that time.
These three features are hereby interpreted as evidence of
intra-arc rifting-translation. We propose, as a hypothesis to be
tested, that the subduction zone might have migrated to the west.
This would have produced thinning of the crust, which in turn
would have originated more primitive IA geochemical signatures
of the magmas and promoted the development of deep basins.
Whether the amount of extension was large enough to develop
oceanic basins and several parallel subduction zones has not
been determined.
The stratigraphy, depositional environments, age, and geo-
chemical affinities of the main units are summarized by terrane.
First, those of southern Mexico are described, following a section
from east to west. Next, Cretaceous rocks of the northern terranes
are described from east to west as well.
Mixteca Terrane
Three localities with Early Cretaceous volcanism have been
identified in the western Mixteca Terrane near the contact with
 Guerrero Composite Terrane of western Mexico
the Guerrero Composite Terrane: the Taxco Schist and the Chapo-
lapa and Zicapa Formations (Fries, 1960; de Cserna and Fries,
1981; Talavera-Mendoza, 1993; Campa and Iriondo, 2003; Fitz
et al., 2002). The Taxco Schist is made up of submarine andesitic
to rhyolitic lava flows and tuffs interbedded with epiclastic rocks
and quartz-rich sandstone and shale (de Cserna and Fries, 1981;
Talavera-Mendoza, 1993). Its volcanic rocks have a continental-
arc geochemical affinity, more evolved than contemporaneous
magmatism from the Guerrero Composite Terrane (Talavera-
Mendoza; 1993; Centeno-García et al., 1993a). The Zicapa For-
mation is made up of dacitic to rhyolitic lava flows interbedded
with fluvial deposits (Fitz et al., 2002). The Chapolapa Formation
is composed mostly of marine lava flows and epiclastic rocks.
The abundance of quartzites within the volcanic-sedimentary
successions of the Taxco Schist and Zicapa Formation suggests
that a crystalline basement was exposed during the arc activity.
U/Pb dating of lavas from the Taxco Schist by sensitive
high-resolution ion microprobe (SHRIMP) methods yielded
130–131 Ma ages (Campa and Iriondo, 2004), and from the
volcanic-volcaniclastic rocks of the Zicapa Formation, 127 Ma
(Fitz et al., 2002). Lava flows from the Chapolapa Formation
have 129–133 Ma SHRIMP U/Pb ages. The Taxco Schist shows
one phase of deformation and metamorphism prior to the deposi-
tion of Aptian–Albian carbonates. Thus, Early Cretaceous vol-
canic rocks of the Mixteca Terrane are unconformably covered
by a carbonate platform that ranges in age from Early to middle
Cretaceous (Fries, 1960).
The limestone succession in the western Mixteca Terrane
changes upward to a thick clastic succession (Mexcala Forma-
tion) of Turonian to Maastrichtian age (Guerrero-Suastegui,
2004). The Mexcala Formation is made up of alternating sand-
stone, shale, and conglomerate, deposited in deltaic and subma-
rine-fan environments (Figs. 7 and 8). It is a synorogenic deposit
(foreland basin-fill) associated with regional thrusting and fold-
ing of both the Guerrero Composite and Mixteca Terranes at the
end of the Cretaceous. Therefore, the Mexcala Formation is the
first overlapping assemblage that stitches the Guerrero Composite
Terrane and the Mixteca Terrane, and marks the final amalgama-
tion of the Guerrero Composite Terrane to continental Mexico.
Teloloapan Terrane
The Teloloapan Terrane (Figs. 1 and 8) is exposed in the east-
ernmost parts of the Guerrero Composite Terrane. This terrane is
characterized structurally by a complex east-vergent thrust-fault
system. Its rocks are severely deformed and metamorphosed to
low-grade greenschist facies. The Teloloapan Terrane overrides
either Cretaceous platform carbonates or Upper Cretaceous silic-
iclastic rocks that belong to the Mixteca Terrane (Figs. 7 and 8;
Talavera-Mendoza et al., 1995).
The nature of the basement of the Teloloapan Terrane
remains unknown. Metamorphic rocks of the Tejupilco area
(Fig. 8) were interpreted as a possible basement for the Telo-
loapan Terrane by Elías-Herrera and Sánchez-Zavala (1990),
and Sanchez-Zavala (1993). These authors suggested that the
299
Tejupilco volcanic-sedimentary sequence might represent an arc
assemblage older than the rest of the Guerrero Terrane magma-
tism. They based this conclusion on U-Pb dates from associ-
ated sulfide deposits. The ages they obtained vary broadly from
Carnian (227 Ma) to Oxfordian (156 Ma). However, the same
volcanic-sedimentary rocks were considered a part of the Cre-
taceous arc assemblage by other authors (Campa and Ramirez,
1979; Talavera-Mendoza et al., 1995).
The arc assemblage of the Teloloapan Terrane consists of
two distinct regions with different volcanic and sedimentary
rocks. The eastern region is characterized by shallow-marine
deposits, and the western region is composed of deeper facies
(Guerrero-Suastegui et al., 1991; Ramírez-Espinoza et al., 1991;
Talavera-Mendoza et al., 1995; Mendoza and Suastegui, 2000;
Guerrero-Suastegui, 2004).The stratigraphy of the eastern region,
from base to top, is made up of basaltic to andesitic pillow and
massive lava flows, volcanic breccias, and pyroclastic flow depos-
its (Villa de Ayala Formation; Talavera-Mendoza et al., 1995).
These deposits are interbedded with epiclastic sandstone and
conglomerate. Primary structures in the volcaniclastic rocks sug-
gest a marine depositional environment (Guerrero-Suastegui et
al., 1991; Guerrero-Suastegui, 2004). Storm deposits, coral frag-
ments, and other fossils suggest shallow and warm waters. This
unit contains fossil gastropods and bivalves that range in age from
Hauterivian to Aptian (Guerrero-Suastegui et al., 1991; Ramírez-
Espinoza et al., 1991; Talavera-Mendoza et al., 1995).
Geochemical analyses of volcanic rocks of the Villa de Ayala
Formation of the Teloloapan Terrane indicate that the magma-
tism is calc-alkaline and similar to that of active intraoceanic arcs
(Talavera-Mendoza, 1993; Talavera-Mendoza et al., 1995; Lapi-
erre et al., 1992; Mendoza and Suastegui, 2000; Centeno-García
et al., 1993a). The base of the Villa de Ayala Formation is not
exposed. The maximum thickness is considered to be ~3000 m
(Guerrero-Suastegui, 2004). The volcanic succession of this for-
mation changes transitionally upward to thick, massive reefal
limestone of the Teloloapan Formation. At the base the Teloloa-
pan Formation is composed of intertidal limestone interbedded
with volcaniclastic rocks containing rudists and nerineas of late
Aptian–early Albian age (Guerrero-Suastegui et al., 1991, 1993;
Guerrero-Suastegui, 2004). Thus magmatism ceased prior to the
late Aptian (Guerrero-Suastegui et al., 1991; Mendoza and Sua-
stegui, 2000; Guerrero-Suastegui, 2004). The Teloloapan Forma-
tion grades upward into the Pachivia Formation of Turonian age,
which is made up of shale and fine-grained sandstone and shale.
The Pachivia Formation is the western equivalent of the Mexcala
Formation of the Mixteca Terrane and indicates that the Telo-
loapan and Mixteca Terranes were already in close proximity
(Guerrero-Suastegui et al., 1991; Talavera-Mendoza et al., 1995;
Guerrero-Suastegui, 2004).
The stratigraphy of the western part of the Teloloapan Ter-
rane comprises submarine basaltic, andesitic, and felsic lava
flows and volcaniclastic rocks (Villa de Ayala Formation) depos-
ited in deeper water conditions than the sediments of the eastern
Teloloapan Terrane. It is in transitional contact upsection with
300 Centeno-García et al.
the Acapetlahuaya Formation, composed of thin-bedded volca-
nic shale and sandstone at the base, at some localities interbed-
ded with dark, thinly laminated limestone. It changes transition-
ally upward to shale, with little or no volcanic material at the
top (Campa and Ramirez, 1979; Guerrero-Suastegui, 2004). This
unit has been highly tectonized, making it difficult to calculate
its original thickness and contact relationships. Apparently, the
Acapetlahuaya Formation changes laterally toward the west and
overlies transitionally the volcaniclastic deposits of the Villa de
Ayala Formation. Its upper contact with the Amatepec Forma-
tion is highly tectonized. The Acapetlahuaya Formation contains
ammonoids, and radiolarians that are late Aptian in age (Campa
et al., 1974; Guerrero-Suastegui et al., 1993; Talavera-Mendoza
et al., 1995; Guerrero-Suastegui, 2004).
The Amatepec Formation is made up of thin-bedded black
detrital limestone and is devoid of volcanic material. It is inter-
preted as deep-basin–slope deposits. This formation is tightly
folded and overlies either the Villa de Ayala or the Acapetlahuaya
Formation. It is late Albian to early Cenomanian in age, based on
calcispherulids, planktonic foraminifers, and radiolarians (Campa
and Ramirez, 1979; Guerrero-Suastegui et al., 1991, 1993;
Talavera-Mendoza, 1993; Talavera-Mendoza et al., 1995). The
deep-marine limestone is overlain by turbiditic sandstone-shale
successions of the Miahuatepec Formation (Talavera-Mendoza et
al., 1995). Fossils have not been found, but it is at least post–early
Cenomanian because of its stratigraphic position. The Miahuate-
pec Formation was deposited, during the amalgamation of the
Zihuatanejo, Arcelia, and Teloloapan Terranes, in a thrust-related
basin (Guerrero-Suastegui et al., 1991; Ramírez-Espinoza et al.,
1991; Talavera-Mendoza et al., 1995; Mendoza and Suastegui,
2000; Guerrero-Suastegui, 2004).
The ages of magmatism of the Teloloapan Terrane have
been poorly constrained by the limited fossils found in the vol-
caniclastic levels. A few U/Pb isotopic ages from felsic lavas at
the base of the succession range in age from 137.4 to 145.9 Ma
(Tithonian–Hauterivian; Mortensen et al., 2003). Thus, magma-
tism of the Teloloapan Terrane is in part contemporaneous with
that of the Mixteca, Guanajuato, and Zihuatanejo Terranes.
There are three distinctive differences in the Cretaceous stra-
tigraphy between the Mixteca and Teloloapan Terranes: (1) Vol-
canism of the Mixteca Terrane is more evolved, and its isotopic
signatures show influence of old continental crust in the magma
generation. In contrast, volcanism of the Teloloapan Terrane is
more primitive and has no traces of contamination by old con-
tinental crust (Centeno-García et al., 1993a; Talavera-Mendoza
et al., 1995; Mendoza and Suastegui, 2000). (2) Metamorphic
and quartz clasts are abundant (up to 70%) in the sandstones that
are interbedded with volcanic rocks in the Mixteca Terrane but
are absent throughout the stratigraphic column of the Teloloapan
Terrane. (3) Magmatism ceased in the Mixteca Terrane before
the Aptian, and part of the volcanic-sedimentary succession was
deformed and metamorphosed (Taxco Schist). In contrast, vol-
canism continued in the Teloloapan Terrane until Aptian–Albian
time, and no internal deformation has been identified.
The arc volcanism of the Mixteca and Teloloapan Terranes
has been interpreted as part of a single arc-backarc system in
which volcanism of the Mixteca Terrane would be the backarc
basin (Cabral-Cano et al., 2000; Monod et al., 1994). An alter-
native interpretation is that these two terranes belong to differ-
ent arcs, separated by a double-dipping subduction of an oceanic
basin (Guerrero-Suastegui, 2004).
Arcelia Terrane
Thrust over the Teloloapan Terrane is the Arcelia Terrane
(Guerrero Composite Terrane), which shows deeper marine
facies and less evolved magmatism than the rest of the arc suc-
cessions of the Guerrero Composite Terrane (Talavera-Mendoza
et al., 1995; Mendoza and Suastegui, 2000). This terrane is made
up of basaltic pillow lavas and ultramafic bodies, black shale and
chert, and fine-grained volcanic turbidites (Fig. 7), all intensively
deformed and partly metamorphosed (Ramírez-Espinoza et al.,
1991; Talavera-Mendoza et al., 1995). The chert layers contain
radiolarians reported as Albian–Cenomanian in age (Dávila and
Guerrero, 1990; Ramírez-Espinoza et al., 1991). Ar/Ar and K/Ar
ages (93.4–105 Ma; Delgado et al., 1990; Ortiz and Lapierre,
1991; Elías-Herrera, 1993) are compatible with biochronology,
but detrital zircon ages from volcanic turbidites are older (mean
age, 130 Ma; Talavera-Mendoza et al., 2007). Geochemical sig-
natures of the Arcelia magmas are similar to those in recent prim-
itive IAs and oceanic basins (MORB) (Talavera-Mendoza, 1993;
Talavera-Mendoza et al., 1995; Mendoza and Suastegui, 2000).
There are no exposures of older rocks in the Arcelia Terrane, and
no clasts of older metamorphic or sedimentary rocks have been
found in its sedimentary strata. Mendoza and Suastegui (2000)
suggest that this terrane is entirely oceanic, that it may have origi-
nated as an independent oceanic arc and backarc basin, and that
it represents partly developed oceanic crust. An alternative inter-
pretation is that the Arcelia Terrane could also be a backarc basin
of the Zihuatanejo Terrane (Centeno-García et al., 2003a).
Southern Part of the Zihuatanejo Terrane
Uppermost Jurassic–Cretaceous volcanic-sedimentary assem-
blages of the Zihuatanejo Terrane can be grouped in three main
regions: northern Zihuatanejo Terrane (Zacatecas area), Huetamo
area, and coastal Zihuatanejo-Colima region (Figs. 7 and 8). The
uppermost Jurassic to Cretaceous strata of the southern Zihuatanejo
Terrane are not as strongly deformed as those in other terranes, and
original contact relationships and complete stratigraphic columns
are well preserved. The strata are characterized by numerous lat-
eral facies changes and internal erosional and angular unconformi-
ties. The geographic distribution of the facies is highly irregular,
and it has not yet been determined in detail. Therefore, compiling,
correlating, and synthesizing the stratigraphy of the area is difficult
because it varies considerably from one locality to another. The
stratigraphy of the northern Zihuatanejo Terrane (Zacatecas area)
is described later.
The stratigraphic column of the southern Zihuatanejo Ter-
rane in the Huetamo area is made up of Triassic basement rocks
 Guerrero Composite Terrane of western Mexico
of the Arteaga Complex, overlain by uppermost Jurassic to Cre-
taceous volcanic and sedimentary cover. These rocks are thrust
over the Arcelia Terrane (Figs. 7 and 8). Arc-related rocks of the
Huetamo region (Figs. 7 and 8) overall have been formed by a
thick succession of alternating shale, sandstone, and conglom-
erate, with scattered basaltic pillow lavas, submarine ignimbrite
flows, and other intermediate pyroclastic and epiclastic flows in
the lower parts of the succession (Angao and San Lucas For-
mations; Pantoja, 1959; Guerrero-Suastegui, 1997). These arc-
related rocks lie unconformably on the Arteaga Complex (Figs. 7
and 8). Fossils of Late Jurassic age have been reported from the
Angao Formation (Pantoja, 1959), although the major exposures
of volcaniclastic rhythmic sedimentary rocks are Berriasian to
upper Aptian (Guerrero-Suastegui, 1997). Their depositional
environment changes upsection from deep to shallow marine.
The volcaniclastic rocks of the San Lucas Formation change
toward the top to thick limestone zones with fossil ammonites,
orbitolinids, and rudists of late Aptian–early Albian age (El
Cajon and Mal Paso Formations; Guerrero-Suastegui, 1997;
Pantoja-Alor and Caballero, 2003). This sequence alternates with
or changes laterally into marine and terrestrial volcanic sandstone
and conglomerate (Comburindio Formation; Guerrero-Suastegui,
1997; Pantoja-Alor and Caballero, 2003). The conglomerate is
covered by massive, thick packets of limestone (Huetamo For-
mation) that contain fossils of late Albian–Cenomanian age. This
unit is found only in the central parts of the Huetamo region (Pan-
toja, 1990).
The arc succession of the Zihuatanejo Terrane in the Huetamo
area was deformed prior to the deposition of a thick, subaerial
red-bed succession that is interbedded with volcanic rocks (Cut-
zamala Formation of Campa and Ramirez, 1979) and is related
to a continental arc of Santonian–Maastrichtian age (Altamira-
Areyán, 2002; Benammi et al., 2005).
The oldest Cretaceous rocks of the Zihuatanejo Terrane in
the Zihuatanejo-Colima region of coastal Mexico that have been
penetrated by drilling are Berriasian–Hauterivian in age (Alberca
Formation; Cuevas, 1981). The lower member of the Alberca
Formation is made up of interbedded black shale, sandstone, and
limestone, and some tuff. The upper member is composed mostly
of andesitic-basaltic lava flows interbedded with limestone and
shale. The Alberca Formation changes transitionally upward to
andesitic and basaltic lava flows, with some rhyolitic flows, inter-
bedded with pyroclastic (intermediate tuffs and ignimbrites) and
epiclastic deposits. It contains limestone packets interbedded with
subaerial conglomerate and sandstone, red siltstone, and some
evaporites, and continues into limestone with scarce basaltic pil-
low lavas at the top (Tecalitlán, Tepalcatepec, and Madrid Forma-
tions). The age range of these units, based on their fossil content,
is Barremian to Cenomanian (Grajales and López, 1984).
Along the west coast between the cities of Colima and
Zihuatanejo are exposures of an important succession of red
beds, alternating with lesser amounts of limestone in compari-
son with other areas of the Guerrero Composite Terrane. The
assemblage is made up of rhyolitic lavas (lava flows, breccias,
301
and ignimbrites) and minor andesitic and dacitic lavas (Tecalitlán
Formation, Titzupa-La Unión assemblage, Playitas Formation,
etc.; Ferrusquía et al., 1978; Grajales and López, 1984; Pantoja
and Estrada, 1986; Centeno-García et al., 2003). These units are
interbedded with epiclastic deposits such as tuff, volcanic shale,
and sandstone, and some conglomerate. The assemblage also
contains thin beds of limestone containing orbitolinids, gastro-
pods, and some pelecypods of late Albian–Cenomanian age (Fer-
rusquía et al., 1978; Grajales and López, 1984). Raindrop marks,
desiccation polygons, and dinosaur footprints can be found in this
succession (Ferrusquía et al., 1978). The lower parts of the Cre-
taceous succession are missing in the Arteaga region, where non-
marine and shallow-marine volcanic and volcaniclastic rocks of
Aptian–Albian age rest unconformably on the Arteaga Complex.
Overall, Cretaceous volcanic rocks of the southern Zihua-
tanejo Terrane show geochemical and isotopic signatures that
suggest a transitional composition between oceanic island arcs
and active continental margins (Centeno-García, 1994; Freydier
et al., 1997; Mendoza and Suastegui, 2000). The high potassium
content, abundance of felsic lavas, and trace element abundances
of these volcanic rocks are similar to those observed in IAs where
the crust is thick (>~20 km), allowing magmatic differentiation
(Centeno-García, 1994).
Rocks of the southern Zihuatanejo Terrane are distinctive
from the rest of the terranes because they were deposited in
shallow-marine and fluvial environments, contain fossil verte-
brates, and show calc-alkaline volcanism more evolved than that
of the Teloloapan and Arcelia Terranes. Sedimentary rocks inter-
bedded with the volcanic flows contain clasts of their basement
rocks, made up of sandstone, quartz, and mylonitic granite. Thus
its stratigraphy is similar to that of arcs constructed on intermedi-
ate crust with a previous history of accretions. The presence of
fossil vertebrates suggests proximity to the continent.
Northern Guerrero Terrane
Following a section from east to west in the northern part of
the Guerrero Terrane, the main stratigraphic characteristic is an
absence of rocks similar to those of the Mixteca or Teloloapan Ter-
rane. Instead, deep-marine volcanic-sedimentary successions of
the Guanajuato Terrane were thrust directly over limestone of the
calcareous platform of Oaxaquia. Contact relationships between
the Guanajuato Terrane and the northern Zihuatanejo Terrane are
unconstrained because the contact is covered by younger units. It
is inferred that the Guanajuato Terrane is overthrust by the Zihua-
tanejo Terrane on the basis of regional vergence of the structures.
Contact relationships between the Tahue and Zihuatanejo Ter-
ranes are unknown because the contact is covered by overlapping
Cenozoic assemblages.
Guanajuato Terrane
The succession at the Guanajuato Terrane has been described
as a complete stratigraphic column of an accreted volcanic arc,
as its assemblages vary from the roots of the arc (gabbros and
302 Centeno-García et al.
diabases, and dike swarms) to pillow basalts, interbedded with
thin-bedded siltstone, shale, chert, and fine-grained volcanic
sandstone (Figs. 6 and 7; Ortiz-Hernandez et al., 1991; Ortiz-
Hernandez, 1992). However, all the different stratigraphic levels
are in the form of tectonic slivers (Fig. 7), with the deepest mafic
levels (gabbro, tonalite, serpentinite, wehrlite, dike swarms)
thrust over the upper stratigraphic levels (pillow basalt and vol-
canic turbidites).
The uppermost thrust sheet is made up of ultramafic-mafic
rocks of the Cerro Pelón tonalite and the Tuna Mansa diorite.
These ultramafic rocks are thrust over a succession incorporat-
ing a diabasic feeder dike swarm, basaltic pillow lavas (La Luz
basalts), rhyolitic tuffs (Cubilete Tuff), and a deep-marine vol-
caniclastic succession made up of sandstone and shale turbi-
dites, chert, and black detrital limestone (Esperanza Formation;
Quintero-Legorreta, 1992; Ortiz-Hernandez et al., 1992; Ortiz-
Hernandez et al., 2003). Basalts of this assemblage show geo-
chemical signatures similar to present primitive volcanic island
arcs (Ortiz-Hernandez, 1992).
The third and lowermost structural level (Fig. 7) is com-
posed of a thick turbidite succession of volcanic graywackes,
quartzites, micritic limestone, radiolarian chert, black shale,
and rare conglomerate resting on basaltic pillow lavas (Arperos
Formation; Ortiz-Hernandez et al., 1992; Lapierre et al., 1992;
Quintero-Legorreta, 1992; Monod et al., 1990; Martínez-Reyes,
1992; Ortiz-Hernandez et al., 2003). Pillow basalts at the base
of the Arperos Formation are more alkaline than the La Luz
basalts and show OI geochemical signatures (Ortiz-Hernandez
et al., 2003). The Arperos Formation is unconformably overlain
by the Aptian–Albian La Perlita Limestone (Ortiz-Hernandez et
al., 2003).
It is difficult to reconstruct the role of the Guanajuato Ter-
rane in the tectonic evolution of western Mexico because of the
lack of enough geochronological data. The only U/Pb zircon
age reported from the area comes from the El Gordo volcano-
genic massive sulfide ore deposit (Hall and Mortensen, 2003),
which is considered part of the lowermost succession by Hall
and Mortensen, (2003), but it is at the stratigraphic level of the
second thrust sheet (Cubilete tuff?) in the stratigraphy proposed
by Ortiz-Hernandez et al. (1992). The age of a rhyolite from El
Gordo volcanogenic massive sulfide ore deposit reported by Hall
and Mortensen (2003) yielded a 146.1 Ma U/Pb age. There are
also reports of badly preserved radiolarians from the Arperos For-
mation that are not in good enough condition to be age indicators
(possibly Valanginian–Turonian in age), but a report of nanno-
fossils suggests a Tithonian–Hauterivian age (Ortiz-Hernandez
et al., 2003). Other ages reported from the Guanajuato area are
from K/Ar analyses and seem to have been reset by later thermal
events (Ortiz-Hernandez et al., 1992, 2003). The sedimentary
rocks of the La Luz and Arperos Formations seem to be distal
volcanic turbidite deposits, but the abundance of limestone asso-
ciated with the pillow lavas suggests that deposition occurred
above the carbonate compensation depth (Ortiz-Hernandez et al.,
2003). Aptian–Albian limestone of the La Perlita Formation rests
unconformably on the Arperos Formation and suggests that sedi-
mentation and at least one phase of deformation occurred prior
to the Aptian–Albian (Ortiz-Hernandez et al., 2003). Whether or
not this deformation is related to the accretion of the Guanajuato
Terrane to the continental margin has not been determined. At
present the Guanajuato Terrane is thrust over the calcareous
platform of Oaxaquia in the San Miguel de Allende area (Ortiz-
Hernández et al., 2002).
Rocks of the Guanajuato Terrane have been correlated with
the Arcelia Terrane, and both were interpreted as having formed
part of an oceanic arc independent of the Zihuatanejo and other
arc terranes (Ortiz-Hernandez et al., 1992). Also, these rocks
are considered relicts of an oceanic basin consumed by subduc-
tion related to the arc of the Zihuatanejo Terrane (Lapierre et al.,
1992; Tardy et al., 1994). An alternative preliminary interpreta-
tion, based on provenance and stratigraphy, is that the Guanajuato
Terrane may have been the backarc basin of the Zihuatanejo Ter-
rane (Centeno-García et al., 2003).
Zihuatanejo Terrane
The Upper Jurassic–Cretaceous stratigraphy of the Zacate-
cas area in the northern Zihuatanejo Terrane is very different than
the stratigraphy of the neighboring Central Terrane and Oaxa-
quia (Figs. 5–7). Whereas the strata in the northern Zihuatanejo
Terrane are mostly composed of volcanic and volcaniclastic
rocks, northern Oaxaquia and the Central Terrane were covered
by a thick, shallow-marine calcareous platform during the Late
Jurassic–Cretaceous (Centeno-García and Silva-Romo, 1997).
This suggests that the Zihuatanejo Terrane was probably under-
going dislocation from the continental margin during that time.
The arc stratigraphy of the Zacatecas area is formed by the
La Borda, Chilitos, and El Saucito Formations (de Cserna, 1976;
Yta et al., 1990; Olvera-Carranza et al., 2001; Olvera-Carranza,
2002). These three formations are made up of pillow basalts and
volcanic breccias, interbedded with thin-bedded siltstone, shale,
chert, and volcanic sandstone and conglomerate, with scarce fel-
sic tuff beds and detrital limestone (Centeno-García and Silva-
Romo, 1997; Olvera-Carranza, 2002). The chert layers contain
radiolarian fossils of Neocomian(?) to Aptian–Albian(?) age (Yta
et al., 1990; Olvera-Carranza, 2002). However, older U/Pb ages
have been reported (150–148 Ma) from the base of the succes-
sion (Danielson, 2000; Mortensen et al., 2003). Lapierre et al.
(1992) and Freydier et al. (1995) characterized this magmatism
as primitive IA and OI basalts. Sedimentary structures and fos-
sil content suggest that the La Borda, El Saucito, and Chilitos
Formations were deposited as distal turbidites and grain flows in
a volcaniclastic submarine apron (Centeno-García et al., 2003).
These Jurassic–Cretaceous arc successions contain important
volcanogenic, massive sulfide ore deposits (Yta et al., 1990; Dan-
ielson, 2000; Mortensen et al., 2003).
Tahue Terane
Cretaceous successions of the Tahue Terrane are exposed
mostly in the Sinaloa de Leyva–Porohui region (Fig. 4). They
 Guerrero Composite Terrane of western Mexico
were formed by submarine pillow lavas, volcaniclastic rocks,
shale, and limestone. They contain Albian ammonites (Ortega-
Gutiérrez et al., 1979; Freydier et al., 1995; Gastil et al., 1999),
but Ar/Ar ages from the lavas are younger (86 Ma; Gastil et al.,
1999), suggesting resetting. The basaltic lavas show MORB and
OIB geochemical affinities, but the volcaniclastic rocks are more
evolved and show IA geochemical signatures (Freydier et al.,
1995; Gastil et al., 1999). Although this volcanic succession has
been interpreted as the northern continuation of the Arperos For-
mation of the Guanajuato Terrane, and part of a major oceanic
basin that originally lay between the Guerrero arc and the con-
tinent (Tardy et al., 1994; Lapierre, et. al., 1992; Dickinson and
Lawton, 2001), the stratigraphy does not support such a tectonic
scenario because (1) the Cretaceous volcanic rocks rest uncon-
formably on a Paleozoic basement, (2) the stratigraphy and facies
associations are not indicative of deep-pelagic sedimentation and
oceanic-ridge volcanism, and (3) the Guanajuato successions
apparently are older than the arc assemblages of the Tahue and
other parts of the Guerrero Composite Terrane.
SUMMARY
• The stratigraphy of the Guerrero Composite Terrane of
western Mexico is characterized by a series of terranes
whose basements were formed by Paleozoic to Triassic
fragments of oceanic arcs, continental slope sediments,
and ocean floor assemblages that were accreted to the con-
tinent and consecutively rifted and translated.
• Metamorphosed Ordovician volcanic and marine sedimen-
tary rocks and a thick succession of deep-marine turbidites
of the NW Guerrero Composite Terrane (Tahue Terrane)
make up the record of a middle Paleozoic collision and
development of a Carboniferous to Permian passive mar-
gin. These rocks might be equivalent to the early Paleozoic
Antler Arc and eugeoclinal sedimentation in the SW Cor-
dillera of North America.
• The continental margin during the early Mesozoic was
located in the middle of Mexico, approximately along the
boundary between Oaxaquia and the Central–Guerrero
Composite Terranes. This continental margin was active
during the Permian–Carboniferous, when a continental arc
developed in Oaxaquia.
• Permian–Carboniferous arc-related magmatism ceased,
and a passive or rifted margin developed along the west-
ern continental margin of Mexico, extending throughout
the Triassic. This development is suggested by the thick
submarine siliciclastic turbidite succession that accumu-
lated on the western paleo-continental shelf–slope region
(Potosi Submarine Fan). The siliciclastic fan turbidites are
mostly continent-derived, quartz-rich sandstone, siltstone,
and shale, containing fossils of Carnian–Norian age.
• The Potosi Fan is interpreted as passive-margin deposits,
as there is no evidence of contemporaneous magmatism
either in the stratigraphy or in the provenance.
303
• The siliciclastic rocks of the Potosi Fan extended to the
west in a marginal oceanic basin (Arteaga Basin) that at
present forms the basement of the Zihuatanejo Terrane of
the Guerrero Composite Terrane.
• The first compressional event that deformed the Triassic
rocks originated tight folding, shearing, and axial cleavage
in the La Ballena Formation, and block-in-matrix texture
in the Taray and Zacatecas Formations and the Arteaga
Complex. This deformation was related to subduction
along the early Mesozoic continental margin. It may have
started sometime between the Late Triassic and Early
Jurassic, accreting the turbidites of the Potosi Submarine
Fan, with slivers of the oceanic crust, to the continent.
• Whether the subducting slab was dipping toward the west
or the east is not well constrained, but the accretionary
prism apparently was very wide. Evidence of contempo-
raneous oceanic arc magmatism is found in the Vizcaíno
Peninsula, where a volcanic sequence of primitive arc affin-
ity is exposed. It is possible that the rocks in the Vizcaino
Peninsula represent a displaced fragment of an oceanic arc
that accreted to the Arteaga Complex of the Guerrero Com-
posite Terrane, but this model needs more evidence.
• Arc-related volcanic and sedimentary rocks unconform-
ably overlie the deformed Triassic rocks of Oaxaquia and
the Central and Guerrero Composite Terranes. They are
characterized by continental rhyolitic to andesitic lava
flows, interbedded with fluvial and alluvial deposits. The
succession shows minor angular unconformities, probably
related to tilting. These rocks have been interpreted as the
southern continuation of the Jurassic continental arc that
developed along the southwestern margin of the United
States. Magmatism was active from ca. 163 to 155 Ma
(Callovian–Oxfordian), although older volcanic rocks
have been reported for eastern Mexico (189 Ma). The
Jurassic arc shows more evolved geochemical signatures
than the subsequent volcanic events.
• During and after the continental arc activity (Late
Jurassic–Early Cretaceous), large amounts of extension
and lateral translations probably occurred, as suggested by
the changes in the stratigraphy. It has been proposed that
major strike-slip faults were probably active during the
arc activity (Mojave-Sonora Megashear). Arc magmatism
ceased in central Mexico, and considerable subsidence
and extension is evidenced by the fast deepening of the
calcareous platform that developed over the arc rocks.
• Major stratigraphic, geochemical, and isotopic differences
are evident in the different Cretaceous stratigraphic assem-
blages among the Guerrero terranes. They are, from east to
west: Andesitic-basaltic submarine lava flows and tuff (IA
geochemical signatures), interbedded with limestone and
shallow-marine volcaniclastics (Teloloapan Terrane) that
were thrust over contemporaneous but more evolved arc
successions and the calcareous platform of southern con-
tinental Mexico (Mixteca Terrane). Ophiolite successions,
304 Centeno-García et al.
with deep-marine volcanic and sedimentary rocks with
MORB, OIB, and IA signatures (Guanajuato and Arcelia
Terranes), are placed between the continent and the more
evolved arc in the north (Zihuatanejo Terrane) and between
the two shallow-marine arcs (Teloloapan and Zihuatanejo
Terranes) in the south.
• These major geological differences suggest that intra-arc
rifting was considerable and originated a series of marginal
arc-backarc systems in western Mexico, with complex
paleogeography. Two possible scenarios can be proposed for
the Cretaceous paleogeography of western Mexico: (1) that
there was one single rifting arc, with westward migration
of the magmatism and development of deep-marine intra-
arc and backarc basins (Guanajuato and Arcelia Terranes);
and (2) that rifting during the end of the Jurassic was large
enough to allow the formation of multiple marginal island
arcs, separated by oceanic backarc basins.
• The proposed timing of the final amalgamation of the
Guerrero terranes to the margin of older terranes that form
the eastern part of Mexico is Turonian to Maastrichtian,
as suggested by the age span of foreland basins associated
with the deformation of the arc. Overlapping the previ-
ously deformed Arcelia and Zihuatanejo Terranes, a new
arc developed along the coast by Santonian time.
ACKNOWLEDGMENTS
This paper is a contribution to PAPIIT projects IN109605–3
and IN116599, funded by the Universidad Nacional Autónoma
de México (UNAM), and to projects UC-MEXUS Exotic ver-
sus Fringing Arc Models: Implications for the Growth of Con-
tinents, and SEP/2003 C02 42642. Special thanks are due J.
Stock, C. Busby, C. Vita-Finzi, and A.E. Draut for their reviews
and comments, which greatly improved the paper.
REFERENCES CITED
Altamira Areyán, A., 2002, Las litofacies y sus implicaciones de la cuenca sedi-
mentaria Cutzamala-Tiquicheo, Estado de Guerrero y Michoacán, México
[M.S. thesis]: México, Universidad Nacional Autónoma de México, Insti-
tuto de Geología, 79 p.
Anderson, C.E., 2006, U-Pb ages of detrital zircons in the Holbrook Member
of the Moenkopi Formation near Winslow, Arizona: Geological Society of
America Abstracts with Programs, v. 38, no. 5, p. 10.
Anderson, T.H., and Silver, L.T., 1979, The role of the Mojave-Sonora
megashear in the tectonic evolution of northern Sonora, in Anderson, T.H.,
and Roldán-Quintana, J., eds., Geology of Northern Sonora: San Diego,
California, Geological Society of America Annual Meeting, Guidebook,
Field Trip 27, p. 59–68.
Anderson, T.H., and Silver, L.T., 1981, An overview of Precambrian rocks in
Sonora, México: Universidad Nacional Autónoma de México, Instituto de
Geología: Revista Mexicana de Ciencias Geológicas, v. 5, p. 131–139.
Anderson, T.H., and Silver, L.T., 2005, The Mojave-Sonora megashear—Field
and analytical studies leading to the conception and evolution of the
hypothesis, in Anderson, T.H., et al., eds., The Mojave-Sonora Megashear
Hypothesis: Development, Assessment, and Alternatives: Geological
Society of America Special Paper 393, p. 1–50.
Anderson, T.H., McKee, J.W., and Jones, N.W., 1990, Jurassic(?) melange in
north-central Mexico: Geological Society of America Abstracts with Pro-
grams, v. 22, no. 3, p. 3.
Anderson, T.H., McKee, J.W., and Jones, N.W., 1991, A northwest trending,
Jurassic fold nappe, northernmost Zacatecas, Mexico: Tectonics, v. 10,
p. 383–401.
Anderson, T.H., Jones, N.W., and McKee, J.W., 2005, The Taray Formation:
Jurassic(?) mélange in northern Mexico—Tectonic implications, in
Anderson, T.H., et al., eds., The Mojave-Sonora Megashear Hypothesis:
Development, Assessment, and Alternatives: Geological Society of Amer-
ica Special Paper 393, p. 427–455.
Araujo, M.J., and Arenas, P.R., 1986, Estudio tectonico-sedimentario, en el mar
mexicano Estados de Chihuahua y Durango: Boletín Sociedad Geológica
Mexicana, v. 47–2, p. 43–71.
Arredondo-Guerrero, P., and Centeno-García, E., 2003, Geology of the Mazat-
lán region, southern Sinaloa state, Mexico: Geological Society of America
Abstracts with Programs, v. 35, no. 4, p. 71.
Barboza-Gudiño, J.R., Tristan-Gonzalez, M., and Torres-Hernandez, J.R.,
1998, The Late Triassic–Early Jurassic active continental margin of west-
ern North America in northeastern Mexico: Geofisica Internacional, v. 37,
p. 283–292.
Barboza-Gudiño, J.R., Tristan-Gonzalez, M., and Torres-Hernandez, J.R.,
1999, Tectonic setting of pre-Oxfordian units from central and northeast-
ern Mexico: A review, in Bartolini, C., et al., eds., Mesozoic Sedimentary
and Tectonic History of North-Central Mexico: Geological Society of
America Special Paper 340, p. 197–210.
Barboza-Gudiño, J.R., Hoppe, M., Gómez-Anguiano, M., and Martínez-
Macías, P.R., 2004, Aportaciones para la interpretación estratigráfica y
estructural de la porción noroccidental de la Sierra de Catorce, San Luis
Potosí, México: Universidad Nacional Autónoma de México, Instituto de
Geología: Revista Mexicana de Ciencias Geológicas, v. 21, p. 299–319.
Bartolini, C., Cantú-Chapa, A., Lang, H., and Barboza-Gudiño, R., 2002, The
Triassic Zacatecas Formation in Central Mexico, paleotectonic, paleo-
geographic and paleobiogeographic implications, in Bartolini C., et al.,
eds., The Western Gulf of Mexico Basin: Tectonics, Sedimentary Basins
and Petroleum Systems: American Association of Petroleum Geologists
Memoir 75, p. 295–315.
Benammi, M., Centeno-García, E., Martinez-Hernandez, E., Morales-Gamez,
M., Urrutia-Fucugauchi, J., and Tolson, G., 2005, Presencia de dinosau-
rios en la Barranca Los Bonetes en el sur de México (Región de Tiquicheo,
Estado de Michoacán) y sus implicaciones cronoestratigráficas: Universi-
dad Nacional Autónoma de México, Instituto de Geología: Revista Mexi-
cana de Ciencias Geológicas, v. 22, p. 429–435.
Bissig, T., Mortensen, J.K., and Hall, B., 2003, The volcano-sedimentary set-
ting of the Kuroko type Vhms District of Cuale, Jalisco, Mexico: Geologi-
cal Society of America Abstracts with Programs, v. 35, no. 4, p. 61.
Burchfiel, B.C., Cowan, D.S., and Davis, G.A., 1992, Tectonic overview of the
Cordilleran orogen in the western United States, in Burchfiel, B.C. et al.,
eds., The Cordilleran Orogen, Conterminous United States: Boulder, Col-
orado, Geological Society of America, The Geology of North America, v.
G-3, p. 407–479.
Burckhardt, C., and Scalia, S., 1906, Géologie des environs de Zacatecas:
Mexico, International Geological Congress, 10th, Excursion Guidebook
16, 26 p.
Cabral-Cano, E., Lang, H.R., and Harrison, C.G.A., 2000, Stratigraphic assess-
ment of the Arcelia–Teloloapan area, southern Mexico: Implications for
southern Mexico’s post-Neocomian tectonic evolution: Journal of South
American Earth Sciences, v. 13, p. 443–457.
Campa, M.F., and Coney, P.J., 1983, Tectono-stratigraphic terranes and mineral
resource distributions in Mexico: Canadian Journal of Earth Sciences,
v. 20, p. 1040–1051.
Campa, M.F., and Iriondo, A., 2003, Early Cretaceous protolith ages for metavol-
canic rocks from Taxco and Taxco Viejo in southern Mexico: Geological
Society of America Abstracts with Programs, v. 35, no. 4, p. 71.
Campa, M.F., and Iriondo, A., 2004, Significado de dataciones Cretácicas de
los arcos volcánicos de Taxco, Taxco Viejo y Chapolapa, en la evolución
de la plataforma Guerrero-Morelos: Unión Geofísica Mexicana: Reunión
Nacional de Ciencias de la Tierra, GEOS, v. 24, p. 173.
Campa, M.F., and Ramirez, J., 1979, La Evolución Geológica y la Metalogéne-
sis de Guerrero: Universidad Autónoma de Guerrero, Ser. Técnico-Cientí-
fica, v. 1, 84 p.
Campa, M.F., Campos, M., Flores, R., and Oviedo, R., 1974, La sequencia
Mesozoíca volcanico-sedimentaria metamorfizada de Ixtapan de la Sal,
Mex.-Teloloapan, Gro.: Soc. Geol. Mex. Bull., v. 35, p. 7–28.
 Guerrero Composite Terrane of western Mexico
Campa, M.F., Ramirez, J., and Bloome, C., 1982, La secuencia volcanico-sedi-
mentaria metamorfizada del Triasico (Ladiniano-Carnico) de la region de
Tumbiscatio, Michoacan: Sociedad Geológica Mexicana, VI Convención
Geológica Nacional, 6 Resúmenes, 48 p.
Cantu-Chapa, A., 1969, Una nueva localidad Triásico Superior en México:
Revista Instituto Mexicano del Petróleo, v. 1, p. 71–72.
Carrillo-Martínez, 1971, Geología de la Hoja San José de Gracia, Sinaloa
[senior thesis]: Mexico City, Universidad Nacional Autónoma de México,
Facultad de Ingeniería, 154 p.
Centeno-García, E., 1994, Tectonic evolution of the Guerrero Terrane, western
Mexico [Ph.D. thesis]: Tucson, University of Arizona, 220 p.
Centeno-García, E., 2002, Overview of Mesozoic subduction-related magmatic
events of Mexico and their relationship with the rest of the cordillera: Geo-
logical Society of America Abstracts with Programs v. 34, no. 5, p. A-97.
Centeno-García, E., 2005, Review of Upper Paleozoic and Lower Mesozoic
stratigraphy and depositional environments of central and west Mexico:
Constraints on terrane analysis and paleogeography, in Anderson, T.H.,
et al., eds., The Mojave-Sonora Megashear Hypothesis: Development,
Assessment, and Alternatives: Geological Society of America Special
Paper 393, p. 233–258.
Centeno-García, E., and Díaz-Salgado, C., 2002, Estratigrafía y Geoquímica
de las rocas volcánicas de la Formación Huizachal en la Región de Aram-
berri, Estado de Nuevo León, III Reunión Nacional de Ciencias de la
Tierra, Puerto Vallarta: Actas INAGEQ, v. 8, p. 244.
Centeno-Garcia, E., and Silva-Romo, G., 1997, Petrogenesis and tectonic
evolution of central México during Triassic-Jurrasic time: Universidad
Nacional Autónoma de México, Instituto de Geología: Revista Mexicana
de Ciencias Geológicas, v. 14, p. 244–260.
Centeno-García, E., Ruíz, J., Coney, P., Patchett, J.P., and Ortega, G.F., 1993a,
Guerrero Terrane of Mexico: Its role in the Southern Cordillera from new
geochemical data: Geology, v. 21, p. 419–422.
Centeno-García, E., García-Díaz, J.L., Guerrero-Suastegui, M., Ramírez-Espi-
noza, J., Salinas-Prieto, J.C., and Talavera-Mendoza, O., 1993b, Geology
of the southern part of the Guerrero Terrane, Ciudad Altamirano-Teloloa-
pan area, in Ortega, G.F., et al., eds., Terrane Geology of Southern México:
Universidad Nacional Autónoma de México, Instituto de Geología, First
Circum-Pacific and Circum-Atlantic Terrane Conference, Gto. Mexico:
Guidebook of Field Trip B, p. 22–33.
Centeno-García, E., Olvera-Carranza, K., Corona-Esquivel, R., Camprubí, A.,
Tritlla, J., and Sanchez-Martinez, S., 2003a, Depositional environment
and paleogeographic distribution of the Jurassic-cretaceous arc in the
western and northern Guerrero Terrane, Mexico: GSA 99th Cordilleran
Section Annual Meeting Abstracts with Programs, v. 35, n. 4, p. 76.
Centeno-García, E., Corona-Chavez, P., Talavera-Mendoza, O., and Iriondo,
A., 2003b, Geology and tectonic evolution of the Western Guerrero ter-
rane—A transect from Puerto Vallarta to Zihuatanejo, México, in Geo-
logic Transects across Cordilleran México: Guidebook for Field Trips of
the 99th GSA Cordilleran Section Meeting, UNAM Instituto de Geologia
Publicación Especial no. 1, p. 201–228.
Centeno-García, E., Gehrels, G., Diaz-Salgado, C., and Talavera-Mendoza, O.,
2005, Zircon provenance of Triassic (Paleozoic?) turbidites from central
and western Mexico: Implications for the early evolution of the Guerrero
Arc: Geological Society of America Abstracts with Programs, v. 37, no.
4, p. 12.
Chávez-Cabello, G., Aranda-Gómez, J.J., Molina-Garza, R.S., Cossío-Tor-
res, T., Irving, R., Arvizu-Gutiérrez I.R., and González-Naranjo, G.A.,
2005, La falla San Marcos: Una estructura jurásica de basamento mul-
tirreactivada del noreste de México, in Alaniz-Álvarez, S.A., and Nieto-
Samaniego, A.F., eds., Boletín de la Sociedad Geológica Mexicana, Volu-
men Conmemorativo Del Centenario, Grandes Fronteras Tectónicas De
México, v. 57, p. 27–52.
Coney, P.J., and Campa, M.F., 1987, Lithotectonic terrane map of México (west
of the 91st meridian): U.S. Geological Survey Miscellaneous Field Stud-
ies Map and Report MF-1874-D, scale 1:10,000,000, 1 sheet.
Contreras-Montero, B., Martínez-Cortes, A., and Gómez-Luna, M.A., 1988,
Bioestratigrafía y sedimentología del Jurásico Superior en San Pedro del
Gallo, Durango, México: Revista del Instituto Mexicano del Petróleo, v.
20, p. 5–29.
Córdoba-Méndez, D.A., 1964, Geology of Apizolaya Quadrangle (east half),
Northern Zacatecas, México [M.A. thesis]: University of Texas at Austin,
111 p.
305
Corona-Chávez, P., and Israde-Alcántara, I., 1999, Carta Geológica del Estado
de Michoacán: Map, scale 1:250,000.
Cuevas, S.F., 1981, Prospecto Tpalcatepec: México, D.F., Petróleos Mexicanos,
IGPR-164, Internal report (unpublished).
Cuevas-Pérez, E., 1983, The geological evolution of the Mesozoic in the State
of Zacatecas Mexico: Zentralblatt für Geologie und Palaeontologie, Teil
I: Allgemeine, Angewandte: Regionale und Historische Geologie, v. 1,
p. 190–201.
Danielson, T.J., 2000, Age, paleotectonic setting, and common Pb isotope sig-
nature of the volcanogenic massive sulfide deposit, southeastern Zacate-
cas State, central Mexico [M.S. thesis]: Vancouver, University of British
Columbia, 120 p.
Dávila, V.M., and Guerrero, M., 1990, Una edad basada en radiolarios para la
secuencia volcánica-sedimentaria de Arcelia, Estado de Guerrero: Socie-
dad Geológica Mexicana, 10th Convención Geológica Nacional, Libro de
Resúmenes, p. 83.
De Cserna, Z., 1976, Geology of the Fresnillo area, Zacatecas, México: Geo-
logical Society of America Bulletin, v. 87, p. 1191–1199.
De Cserna, Z., 1978, Notas sobre la geología de la región comprendida entre
Iguala, Ciudad Altamirano y Temascaltepec, Estados de Guerrero y
México: Sociedad Geológica Mexicana, Libro Guía de la Excursión
geológica a Tierra Caliente, p. 1–25.
De Cserna, Z., and Fries, C., Jr., 1981, Hoja Taxco 14Q-h(7), and Resumen de
la Geología de la Hoja Taxco, Estados de Guerrero, México y Morelos:
Universidad Nacional Autónoma de México, Insituto de Geología, Carta
Geologica de Mexico: Map Ser., scale 1:100,000, and 47 p. text.
Delgado, A.L., López, M.M., York, D., and Hall, C.M., 1990, Geology and geo-
chronology of ultramafic localities in the Cuicateco and Tierra Caliente
Complexes, southern Mexico: Geological Society of America Abstracts
with Programs v. 22, no. 7, p. 326.
Delgado, L.A., 1982, Descripción preliminar de la geología y mecánica
de emplazamiento del complejo ultrabásico de Loma Baya, Guer-
rero, México: Universidad Nacional Autónoma de México, Instituto de
Geofísica: Geofísica Internacional, v. 25, p. 537–558.
Díaz Salgado, C., 2004, Caracterización tectónica y procedencia de la for-
mación Taray, región de Pico de Teyra, estado de Zacatecas [M.S. thesis]:
Universidad Nacional Autónoma de México, Posgrado en Ciencias de la
Tierra, Instituto de Geología, 95 p.
Diaz-Salgado, C., Centeno-García, E., and Gehrels, G., 2003, Stratigraphy,
depositional environments, and tectonic significance of the Taray Forma-
tion, northern Zacatecas state, Mexico: Geological Society of America
Abstracts with Programs, v. 35, no. 4, p. 71.
Dickinson, W., 2004, Evolution of the North American Cordillera: Annual
Review of Earth and Planetary Sciences, v. 32, p. 13–45, doi: 10.1146/
annurev.earth.32.101802.120257.
Dickinson, W.R., and Lawton, T.F., 2001, Carboniferous to Cretaceous assem-
bly and fragmentation of Mexico: Geological Society of America Bul-
letin, v. 113, p. 1142–1160.
Eguiluz, A.S., and Campa, M.F., 1982, Problemas tectónicos del sector de San
Pedro El Gallo, en los Estados de Chihuahua y Durango: Boletín Asoci-
ación Mexicana de Geólogos Petroleros, v. 34, p. 5–42.
Elías-Herrera, M., 1993, Geology of the Valle de Bravo and Zacazonapan areas,
south-central Mexico, in Ortega-Gutiérrez, F., et al., eds., Proceedings of
the First Circum-Pacific and Circum-Atlantic Terrane Conference: Gua-
najuato, Mexico, p. 12–21.
Elías-Herrera, M., and Sánchez-Zavala, J.L., 1992, Tectonic implications of
a mylonitic granite in the lower structural levels of the Tierra Caliente
Complex (Guerrero Terrane), Southern México: Universidad Nacional
Autónoma de México, Instituto de Geología: Revista Mexicana de Cien-
cias Geológicas, v. 9, p. 113–125.
Fastovsky, D.E., Hermes, O.D., Strater, N.H., Bowring, S.A., Clark, J.M., Mon-
tellano, M., and Hernandez R., 2005, Pre–Late Jurassic, fossil-bearing
volcanic and sedimentary red beds of Huizachal Canyon, Tamaulipas,
Mexico, in Anderson, T.H., et al., eds., The Mojave-Sonora Megashear
Hypothesis: Development, Assessment, and Alternatives: Geological
Society of America Special Paper 393, p. 259–282.
Ferrusquía, I., Applegate, S.P., and Espinosa, L., 1978, Rocas volcanosedi-
mentarias mesozoicas y huellas de dinosaurios en la región surocciden-
tal pacífica de México: Universidad Nacional Autónoma de México,
Instituto de Geología: Revista Mexicana de Ciencias Geológicas, v. 2,
p. 150–162.
306 Centeno-García et al.
Fitz, D.E., Campa, M.F., and López, M.M., 2002, Fechamiento de lavas ande-
síticas de la Formación Zicapa, en el límite oriental de la Plataforma
Guerrero Morelos: Actas INAGEQ, v. 8, p. 178.
Freydier, C., Lapierre, H., Tardy, M., Coulon, C., Martinez-Reyes, J., and
Orsini, J.B., 1995, Les formations magmatiques de Porohui (Sinaloa);
temoins de l’evolution géodynamique mesozoique et tertiaire des Cordil-
leres mexicaines: Comptes Rendus de l’Académie des Sciences, Ser. 2:
Sciences de la Terre et des Planètes, v. 321, p. 529–536.
Freydier, C., Lapierre, H., Briqueu, L., Tardy, M., Coulon, Ch., and Martínez,
J., 1997, Volcaniclastic sequences with continental affinities within the
Late Jurassic–Early Cretaceous intraoceanic Arc terrane (western Mex-
ico): Journal of Geology, v. 105, p. 483–502.
Freydier, C., Lapierre, H., Ruiz, J., Tardy, M., Martinez-R.J., and Coulon, C.,
2000, The Early Cretaceous Arperos basin: An oceanic domain dividing
the Guerrero arc from nuclear Mexico evidenced by the geochemistry of
the lavas and sediments: Journal of South American Earth Sciences, v. 13,
p. 325–336.
Fries, C., Jr., 1960, Geología del Estado de Morelos y de partes adyacentes de
México y Guerrero, región central meridional de México: Universidad
Nacional Autónoma de México, Instituto de Geología, Boletín, no. 60,
236 p.
García-Díaz, J.L., Tardy, M., Campa Uranga, M.F., and Lapierre H., 2004,
Geología de la Sierra Madre del Sur en la región de Chilpancingo y Oli-
nalá, Guerrero, una contribución al conocimiento de la evolución geo-
dinámica del margen Pacífico mexicano a partir del Jurásico: Unión
Geofísica Mexicana, Reunión Nacional de Ciencias de la Tierra, GEOS,
v. 24, p. 173.
Gastil, G., Miller, R., Anderson, P., Crocker, J., Campbell, M., Buch, P.,
Lothringer, C., Leier-Engelhardt, P., DeLattre, M., Hoobs, J., and Roldán-
Quintana, J., 1991, The relation between the Paleozoic strata on opposite
sides of the Gulf of California, in Pérez-Segura, E., and Jacques-Ayala,
C., eds., Studies of Sonoran Geology: Geological Society of America
Special Paper 254, p. 1–7.
Gastil, G., Rector, R., Hazelton, G., Al-Riyami, R., Hanes, J., Farrar, E., Boeh-
nel, H., Ortega-Rivera, A., and Guzman, J.G., 1999, Late Cretaceous pil-
low basalt, siliceous tuff and calc-turbidite near Porohui, northern Sinaloa,
Mexico, in Bartolini, C., et al., eds., Mesozoic sedimentary and tectonic
history of north-central Mexico: Geological Society of America Special
Paper 340, p. 145–150.
Gastil, R.G., Morgan, G.J., Krummenacher, D., 1978, Mesozoic history of
peninsular California and related areas east of the Gulf of California, in
Howell, D.G., McDougall, K.A., eds., Mesozoic Paleogeography of the
western United States: Los Angeles, California, Pacific Section, Society
of the Economic Paleontologists and Mineralogists, p. 107–115.
Gehrels, G.E., Stewart, J.H., and Ketner, K.B., 2002, Cordilleran-margin
quartzites in Baja California—Implications for tectonic transport: Earth
and Planetary Science Letters, v. 199, p. 202–210.
Goldhammer, R.K., 1999, Mesozoic sequence stratigraphy and paleogeographic
evolution of northeast Mexico, in Bartolini, C., et al., eds., Mesozoic Sedi-
mentary and Tectonic History of North-Central Mexico: Geological Soci-
ety of America Special Paper 340, p. 197–210.
González-León, C., Stanley, G.D., Gehrels, G., and Centeno-Garcia, E., 2005,
New data on the lithostratigraphy, zircon and isotope provenance, and
paleogeographic setting of the El Antimonio Group, Sonora, Mexico, in
Anderson, T.H., et al., eds., The Mojave-Sonora Megashear Hypothesis:
Development, Assessment, and Alternatives: Geological Society of Amer-
ica Special Paper 393, p. 259–282.
Grajales, M., and López, M., 1984, Estudio petrogenético de las rocas ígneas y
metamorficas en el Prospecto Tomatlan-Guerrero-Jalisco: Instituto Mexi-
cano del Petróleo, Subdirección de Tecnología y Exploración, Proyecto
C-1160 (unpublished).
Guerrero-Suastegui, M., 1997, Depositional history and sedimentary petrology
of the Huetamo sequence, Southwestern Mexico [M.S. thesis]: University
of Texas at El Paso, 95 p.
Guerrero-Suastegui, M., 2004, Depositional and tectonic history of the Guer-
rero Terrane, Sierra Madre de Sur; with emphasis on sedimentary suc-
cessions of the Teloloapan area, southwestern Mexico [Ph.D. thesis]: St.
John’s, Newfoundland, Memorial University, 600 p.
Guerrero-Suastegui, M., Ramírez-Espinosa, J., Talavera-Mendoza, O., and
Campa-Uranga, M.F., 1991, El desarrollo carbonatado del Cretácico Infe-
rior asociado al arco de Teloloapan, Noroccidente del Estado de Guerrero:
Convención sobre la evolución Geológica Mexicana, 1er Congreso Mexi-
cano de Mineralogía, Pachuca, Memoir, p. 67–70.
Guerrero-Suastegui, M., Ramírez Espinosa, J., Gómez Luna, M.E., González
Casildo, V., and Martínez Cortes, A., 1993, Depósitos de tormenta y fauna
fósil asociada del Albiano Superior (Formación Teloloapan) Noroeste del
Estado de Guerrero: Sociedad Mexicana de Paleontología, IV Congreso
Nacional de Paleontología, Memorias, p. 93–97.
Hall, B.V., and Mortensen, J.K., 2003, Bimodal-siliciclastic massive sulfide
deposits of the Leon-Guanajuato District, Central Mexico: Geological
Society of America Abstracts with Programs, v. 35, no. 4, p. 61.
Henry, C.D., and Fredrikson, G., 1987, Geology of part of southern Sinaloa,
Mexico, adjacent to the Gulf of California: Geological Society of America
Map and Chart Ser., v. MCH063, p. 1–14.
Jones, N.W., McKee, J.W., Anderson, T.H., and Silver, L.T., 1995, Jurassic vol-
canic rocks in northeastern Mexico: A possible remnant of a Cordilleran
magmatic arc, in Jaques-Ayala, C., et al., eds., Studies on the Mesozoic of
Sonora and Adjacent Areas: Geological Society of America Special Paper
301, p. 179–190.
Keppie, J.D., Dostal, J., Cameron, K.L., Solari, L.A., Ortega-Gutiérrez, F., and
Lopez, R., 2003, Geochronology and geochemistry of Grenvillian igne-
ous suites in the northern Oaxacan Complex, southern Mexico: Tectonic
implications: Precambrian Research, v. 120, p. 365–389.
Keppie, J.D., Dostal, J., Miller, B.V., Ortega-Rivera, A., Roldán-Quintana, J.,
and Lee, J.W.K., 2006, Geochronology and geochemistry of the Fran-
cisco Gneiss: Triassic continental rift tholeiites on the Mexican margin of
Pangea metamorphosed and exhumed in a Tertiary core complex: Interna-
tional Geology Review, v. 48, p. 1–16.
Kimbrough, D.L., and Moore, T.E., 2003, Ophiolite and volcanic arc assem-
blages on the Vizcaino Peninsula and Cedros Island, Baja California Sur,
Mexico: Mesozoic forearc lithosphere of the Cordilleran magmatic arc,
in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA, A Volume in Honor of R. Gordon Gastil:
Geological Society of America Special Paper 374, p. 43–71.
Labarthe, G., Tristán, M., and Aguillón, R.A., 1982, Estudio geológico-minero
del área de Peñón Blanco, estados de San Luis Potosí y Zacatecas: Insti-
tuto de Geología y Metalurgia, Universidad Autónoma de San Luis Potosí,
Folleto Técnico no. 76, 80 p.
Lapierre, H., Brouxel, M., Albarède, F., Coulon, C., Lecuyer, C., Martin, P.,
Mascle, G., and Rouer, O., 1987, Paleozoic and Lower Mesozoic mag-
mas from the eastern Klamath Mountains (North California) and the
geodynamic evolution of northwestern America: Tectonophysics, v. 140,
p. 155–177.
Lapierre, H., Ortiz, L.E., Abouchami, W., Monod, O., Coulon, C., and Zim-
mermann, J.L., 1992, A crustal section of an intra-oceanic island arc: The
Late Jurassic–Early Cretaceous Guanajuato magmatic sequence, central
Mexico: Earth and Planetary Science Letters, v. 108, p. 61–77.
Lawlor, P.J., Ortega-Gutiérrez, F., Cameron, K.L., Ochoa-Camarillo, H., Lopez,
R., and Sampson, D.E., 1999, U–Pb geochronology, geochemistry, and
provenance of the Grenvillian Huiznopala Gneiss of Eastern Mexico: Pre-
cambrian Research, v. 94, p. 73–99.
López-Infanzón, M., 1986, Petrologia y radiometria de rocas igneas y metamor-
ficas de Mexico: Boletin de la Asociacion Mexicana de Geologos Petrole-
ros, v. 38, p. 59–98.
Martínez-Reyes, J., 1992, Mapa geológico de la sierra de Guanajuato, escala
1:100,000, con Resumen de la geología de la sierra de Guanajuato [map]:
Universidad Nacional Autónoma de México, Instituto de Geología, Carta
Geológica de México Ser., scale 1:100,000.
McDowell, F.W., Housh, T.B., and Wark, D.A., 1999, Nature of the crust beneath
west-central Chihuahua, Mexico, based upon Sr, Nd, and Pb isotopic com-
positions at the Tomochic volcanic center: Geological Society of America
Bulletin, v. 111, p. 823–830, doi: 10.1130/0016–7606(1999)111<0823:
NOTCBW>2.3.CO;2.
McKee, J.W., Jones, N.W., and Anderson, T.H., 1999, Late Paleozoic and
early Mesozoic history of the Las Delicias terrane, Coahuila, Mexico, in
Bartolini, C., et al., eds., Mesozoic Sedimentary and Tectonic History of
North-Central Mexico: Geological Society of America Special Paper 340,
p. 161–189.
Mendoza, O.T., and Suastegui, M.G., 2000, Geochemistry and isotopic compo-
sition of the Guerrero Terrane (western México): Implications for the tec-
tonomagmatic evolution of southwestern North America during the Late
Mesozoic: Journal of South American Earth Sciences, v. 13, p. 297–324.
 Guerrero Composite Terrane of western Mexico
Monod, O., and Calvet, P., 1991, Structural and stratigraphic reinterpretation
of the Triassic units near Zacatecas, Zac., Central Mexico: Evidence of a
Laramide nappe pile: Zentralblatt für Geologie und Palaeontologie, Teil
I: Allgemeine, Angewandte: Regionale und Historische Geologie, v. 1,
p. 1533–1544.
Monod, O., Lapierre, H., Chiodi, M., Martínez-Reyes, J., Calvet, Ph., Ortiz-
Hernández, E., and Zimmermann, J.L., 1990, Réconstitution d’un arc
insulaire intra-océanique au Mexique central: La séquence volcanopluto-
nique de Guanajuato (Crétacé inférieur): Comptes Rendus de l’Académie
des Sciences de Paris, Sér. 2, v. 310, p. 45–51.
Monod, O., Faure, M., and Salinas, J.C., 1994, Intra-arc opening and closure
of a marginal sea: The case of the Guerrero Terrane (SW Mexico): Island
Arc, v. 3, p. 25–34, doi: 10.1111/j.1440-1738.1994.tb00002.x.
Mortensen, J.K., Hall, B.V., Bissig, T., Friedman, R.M., Danielson, T., Oliver,
J., Rhys, D.A., and Ross, K.V., 2003, U-Pb zircon age and Pb isotopic
constraints on the age and origin of volcanogenic massive sulfide deposits
in the Guerrero Terrane of central Mexico: Geological Society of America
Abstracts with Programs, v. 35, no. 4, p. 61.
Mullan, H.S., 1978, Evolution of part of the Nevadan Orogen in northwestern
Mexico: Geological Society of America Bulletin, v. 89, p. 1175–1188,
doi: 10.1130/0016-7606(1978)89<1175:EOPOTN>2.0.CO;2.
Olvera-Carranza, K., 2002, Estudio estratigráfico-estructural del sector central
de la Sierra de Zacatecas, México [B.E. thesis]: Universidad Autónoma de
Nuevo León, Facultad de Ciencias de la Tierra, 70 p.
Olvera-Carranza, K., Centeno, E., and Camprubí, A., 2001, Deformation
and distribution of massive sulphide deposits in Zacatecas, Mexico, in
Pietrzynski, A., et al., eds., Mineral Deposits at the Beginning of the 21st
Century: Lisse, Swets and Zeitlinger, p. 313–316.
Ortega, G.F., Mitre, S.L.M., Roldán, Q.J., Aranda, G.J., Morán, Z.D., Alaníz,
A.S., and Nieto, S.A., 1992, Carta Geológica de la República Mexicana:
Universidad Nacional Autónoma de México, Instituto de Geología: Map,
scale 1:2,000,000, 5th edition.
Ortega-Gutiérrez, F., 1981, Metamorphic belts of southern Mexico and their
tectonic significance: Geofísica Internacional, v. 20, p. 177–202.
Ortega-Gutiérrez, F., Elías-Herrera, M., Reyes-Salas, M., Ortega-Gutiérrez, F.,
Prieto-Vélez, R., Zúñiga, Y., and Flores, S., 1979, Una secuencia volcano-
plutónica-sedimentaria cretácica en el norte de Sinaloa; ¿un complejo
ofiolítico?: Universidad Nacional Autónoma de México, Instituto de
Geología: Revista Mexicana de Ciencias Geológicas, v. 3, p. 1–8.
Ortega-Gutiérrez, F., Ruíz, J., and Centeno-García, E., 1995, Oaxaquia—A
Proterozoic microcontinent accreted to North America during the
Late Paleozoic: Geology, v. 23, p. 1127–1130, doi: 10.1130/0091-
7613(1995)023<1127:OAPMAT>2.3.CO;2.
Ortega-Gutiérrez, F., Elias-Herrera, M., Reyes-Salas, M., Macias-Romo, C.,
and Lopez, R., 1999, Late Ordovician–Early Silurian continental colli-
sional orogeny in southern Mexico and its bearing on Gondwana–Lauren-
tia connections: Geology, v. 27, p. 719–722.
Ortiz, E., and Lapierre, H., 1991, Las secuencias toleíticas de Guanajuato y
Arcelia, México centro-meridional: Remanentes de un arco insular intra-
oceánico del Jurásico superior–Cretácico inferior: Zentralblatt für Geolo-
gie und Palaeontologie, Teil I: Allgemeine, Angewandte: Regionale und
Historische Geologie, v. 6, p. 1503–1517.
Ortiz-Hernandez, E.L., 1992, L’arc intra-océanique allocthone Jurassique supéri-
eur–Crétacé inférieur du domaine Cordillerain Mexicain (Guerrero Ter-
rane): Pétrographie, géochimie et mineralisations associées des segments de
Guanajuato et de Palmar Chico–Arcelia consequences paleogéographiques
[thesis]: Université Joseph Fourier-Grenoble, France, 109 p.
Ortiz-Hernandez, E.L., Yta, M., Talavera, O., Lapierre, H., Monod, O., and
Tardy, M., 1991, Origine intra-océanique des formations volcano-pluto-
niques d’arc du Jurassique supérieur–Crétacé inférieur du Mexique cen-
tro-méridional: Comptes Rendus de l’Académie des Sciences (Paris) (Ser.
2), v. 305, p. 1093–1098.
Ortiz-Hernandez, E.L., Chiodi, M., Lapierre, H., Monod, O., and Calvet, P.,
1992, El arco intraoceánico alóctono (Cretácico Inferior) de Guanajuato-
características petrográficas, geoquímicas, estructurales e isotópicas del
complejo filonianao y de las lavas basálticas asociadas, implicaciones
geodinámicas: Universidad Nacional Autónoma de México, Instituto de
Geología: Revista Mexicana de Ciencias Geológicas, v. 9, p. 126–145.
Ortiz-Hernández, E.L., Flores-Castro, K., and Acevedo-Sandoval, O.A., 2002,
Petrographic and geochemical characteristics of upper Aptian calc-alka-
line volcanism in San Miguel de Allende, Guanajuato state, Mexico: Uni-
307
versidad Nacional Autónoma de México, Instituto de Geología: Revista
Mexicana de Ciencias Geológicas, v. 19, p. 81–90.
Ortiz-Hernandez, E.L., Acevedo-Sandoval, O.A., and Flores-Castro, K., 2003,
Early Cretaceous intraplate seamounts from Guanajuato, central México,
geochemical and mineralogical data: Universidad Nacional Autónoma de
México, Instituto de Geología: Revista Mexicana de Ciencias Geológicas,
v. 20, p. 27–40.
Pacheco, G.C., Castro, M.R., and Gómez, M.A., 1984, Confluencia de terrenos
estratotectónicos en Santa María del Oro, Durango, México: Revista del
Instituto Mexicano del Petróleo, v. 16, p. 7–20.
Pantoja, A.J., 1959, Estudio Geológico de Reconocimiento de la región de
Huetamo, Estado de Michoacán: Consejo de Recursos Naturales no
renovables: Boletin (Instituto de Estudios de Poblacion y Desarrollo
[Dominican Republic]), v. 50, p. 1–33.
Pantoja, A.J., 1990, Redefinición de las unidades estratigráficas de la secuencia
Mesozoica de la región de Huetamo-Altamirano, Estados de Michoacán y
Guerrero: Convención Geológica Nacional, Resúmenes, p. 66.
Pantoja, A.J., and Estrada, B.S., 1986, Estratigrafía de los alrededores de la
mina de fierro de El Encino, Jalisco: Sociedad Geológica Mexicana: Bole-
tin (Instituto de Estudios de Poblacion y Desarrollo [Dominican Repub-
lic]), v. 47, p. 1–15.
Pantoja-Alor, J., 1963, Hoja Biseca 13R-k(3), con resumen de la geología de
la Hoja San Pedro del Gallo, Estado de Durango: Universidad Nacional
Autónoma de México, Instituto de Geología: Carta Geológica de México
Ser., scale 1:100,000.
Pantoja-Alor, J. and Gómez-Caballero, J.A., 2003, Main geologic and biostrati-
graphic features of the Cretaceous of southwestern Mexico (Guerrero
terrane), guidebook for field trips of the 99th GSA Cordilleran Section
Meeting: UNAM Instituto de Geologia Publicación Especial, no. 1, p.
229–260.
Patchett, P.J., and Ruíz, J., 1987, Nd isotopic ages of crust formation and meta-
morphism in the Precambrian of eastern and southern México: Contri-
butions to Mineralogy and Petrology, v. 96, p. 523–528, doi: 10.1007/
BF01166697.
Poole, F.G., and Madrid, R.J., 1988, Allochthonous Paleozoic eugeoclinal rocks
of the Barita de Sonora mine area, central Sonora, México, in Rodríguez-
Torres, ed., El Paleozoico de la región central del Estado de Sonora:
Libreto Guía de la Excursión Geológica para el Segundo Simposio sobre
la Geología y Minería del Estado de Sonora, Excursiones de Campo, Uni-
versidad Nacional Autónoma de México, Hermosillo, Sonora, p. 32–41.
Poole, F.G., and Perry, W.J., Jr., 1998, Laurentia-Gondwana continental margins
in Northern Mexico, and their late Paleozoic collision, IGCP Project 376
annual meeting: Instituto de Geología, Universidad Nacional Autónoma
de México, México City, México, Laurentia-Gondwanan connections
before Pangea, Program and Abstracts, p. 27.
Poole, F.G., Perry, W.J., Madrid, R.J., and Amaya-Martínez, R., 2005, Tectonic
synthesis of the Ouachita-Marathon-Sonora orogenic margin of southern
Laurentia: Stratigraphic and structural implications for timing of defor-
mational events and plate-tectonic model, in Anderson, T.H., et al., eds.,
The Mojave-Sonora Megashear Hypothesis: Development, Assessment,
and Alternatives: Geological Society of America Special Paper 393, p.
543–596.
Quintero-Legorreta, O., 1992, Geología de la región de Comanja, Estados
de Guanajuato y Jalisco: Universidad Nacional Autónoma de México,
Instituto de Geología: Revista Mexicana de Ciencias Geológicas, v. 10,
p. 6–25.
Ramírez-Espinosa, J., Campa, M.F., Talavera, O., and Guerrero, M., 1991, Car-
acterización de los arcos insulares de la Sierra Madre del Sur y sus impli-
caciones tectónicas: Convención sobre la evolución Geológica Mexicana,
1er Congreso Mexicano de Mineralogía, Pachuca, Memoir, p. 163–166.
Ramírez-Ramírez, C., 1992, Pre-Mesozoic geology of Huizachal-Peregrina
Anticlinorium, Ciudad Victoria, Tamaulipas, and adjacent parts of eastern
Mexico [Ph.D. thesis]: University of Texas at Austin, 450 p.
Ranson, W.A., Fernandez, L.A., Simmons, W.B., Jr., and de la Vega, E.S., 1982,
Petrology of the metamorphic rocks of Zacatecas, México: Sociedad
Geológica Mexicana, v. 43, p. 37–59.
Roldán-Quintana, J., Gonzalez-Leon, C.M., and Amaya-Martinez, R., 1993,
Geologic constraints on the northern limit of the Guerrero Terrane in
northwestern Mexico, in Ortega Gutiérrez, F., et al., eds., First Circum-
Pacific and Circum-Atlantic Terrane Conference: Guanajuato, México,
Proceedings, p. 124–127.
308 Centeno-García et al.
Rosales-Lagarde, L., Centeno-García, E., Dostal, J., Sour-Tovar, F., Ochoa-
Camarillo, H., and Quiroz-Barroso, S., 2005, The Tuzancoa Formation of
Hidalgo State: Evidence of a Carboniferous–Permian submarine arc built
on continental crust in eastern Mexico: International Geology Review,
v. 47, p. 901–919.
Ruíz, J., Patchett, P.J., and Ortega-Gutiérrez, F., 1988, Proterozoic and Pha-
nerozoic basement terranes of México from Nd isotopic studies: Geologi-
cal Society of America Bulletin, v. 100, p. 274–281, doi: 10.1130/0016–
7606(1988)100<0274:PAPBTO>2.3.CO;2.
Salinas-Prieto, J.C., Monod, O., and Faure, M., 2000, Ductile deformations of
opposite vergence in the eastern part of the Guerrero Terrane (SW Mex-
ico): Journal of South American Earth Sciences, v. 13, p. 389–402.
Sanchez-Zavala, J.L., 1993, Secuencia volcanosedimentaria Jurásico Superior-
Cretácico Arcelia Otzoloapan (Terreno Guerrero), area Valle de Bravo-
Zacazonapan, Estado de México: Petrografía, Geoquímica, Metamor-
fismo e Interpretación Tectónica [M.S. thesis]: Universidad Nacional
Autónoma de México, Facultad de Ciencias, Mexico, 91 p.
Sánchez-Zavala, J.L., Centeno-García, E., and Ortega-Gutiérrez, F., 1999,
Review of Paleozoic stratigraphy of Mexico and its role in the Gondwana-
Laurentia connections, in Ramos, V.A., and Keppie, J.D., eds., Laurentia-
Gondwana Connections before Pangea: Geological Society of America
Special Paper 336, p. 211–226.
Schaaf, P., Böhnel, H., and Pérez-Venzor, P.A., 2000, Pre-Miocene palaeogeog-
raphy of the Los Cabos Block, Baja California Sur: Geochronological and
palaeomagnetic constraints: Tectonophysics, v. 318, p. 53–69.
Sedlock, R.L., Ortega-Gutiérrez, F., and Speed, R.C., 1993, Tectonostrati-
graphic Terranes and Tectonic Evolution of Mexico: Geological Society
of America Special Paper 278, 153 p.
Servais, M., Rojo, Y.R., and Colorado, L.D., 1982, Estudio de las rocas basicas
y ultrabasicas de Sinaloa y Guanajuato; postulacion de un paleogolfo de
Baja California y de una digitacion tethysiana en Mexico central: Geomi-
met, v. 3, p. 53–71.
Silva-Romo, G., 1993, Estudio de la Estratigrafía y Estructuras Tectónicas de
la Sierra de Salinas, Estados de San Luis Potosí y Zacatecas [M.S. the-
sis]: Universidad Nacional Autónoma de México, Facultad de Ciencias,
México, 111 p.
Silva-Romo, G., Arellano Gil, J., Mendoza Rosales, C., and Nieto Obregon,
J., 2000, A submarine fan in the Mesa Central, Mexico: Journal of South
American Earth Sciences, v. 13, p. 429–442.
Solari, L.A., Keppie, J.D., Ortega-Gutiérrez, F., Cameron, K.L., Lopez, R., and
Hames, W.E., 2003, ~990 and ~1,100 Grenvillian tectonothermal events
in the northern Oaxacan Complex, southern Mexico: Roots of an orogen:
Tectonophysics, v. 365, p. 257–282.
Stewart, J.H., and Roldán-Quintana, J., 1991, Upper Triassic Barranca Group:
Nonmarine and shallow-marine rift-basin deposits of northwestern Mex-
ico, in Pérez-Segura, E., and Jacques-Ayala, C., eds., Studies of Sonoran
Geology: Geological Society of America Special Paper 254, p. 19–36.
Stewart, J.H., Poole, F.G., Ketner, K.B., Madrid, R.J., Roldán-Quintana, J.,
and Amaya-Martínez, R., 1990, Tectonics and stratigraphy of the Paleo-
zoic and Triassic southern margin of North America, Sonora, México, in
Gehrels, G.E., and Spencer, J.E., eds., Geologic Excursions through the
Sonoran Desert Region, Arizona and Sonora: Arizona Geological Survey
Special Paper 7, p. 183–202.
Stewart, J.H., Blodgett, R.B., Boucot, A.J., Carter, J.L., and Lopez, R., 1999,
Exotic Paleozoic strata of Gondwanan provenance near Ciudad Victoria,
Tamaulipas, Mexico, in Ramos, V.A., and Keppie, J.D., eds., Laurentia-
Gondwana Connections before Pangea: Geological Society of America
Special Paper 336, p. 227–252.
Talavera-Mendoza, O., 1993, Les formations orogéniques mésozoïques du
Guerrero (Mexique méridional): Contribution à la connaissance de
l’évolution géodynamique des cordillères mexicaines [Ph.D. thesis]: Uni-
versité Joseph Fourier-Grenoble I, France, 462 p.
Talavera-Mendoza, M.O., 2000, Mélanges in southern México: Geochemistry
and metamorphism of Las Ollas complex (Guerrero Terrane): Canadian
Journal of Earth Sciences, v. 37, p. 1309–1320.
Talavera-Mendoza, O., Ramirez-Espinosa, J., and Guerrero-Suástegui, M.,
1995, Petrology and geochemistry of the Teloloapan subterrane, a Lower
Cretaceous evolved intra-oceanic island-arc: Geofísica Internacional,
v. 34, p. 3–22.
Talavera-Mendoza, O., Ruiz, J., Gehrels, G., Valencia, V., and Centeno-García,
E., 2007, Detrital zircon U/Pb geochronology of southern Guerrero and
western Mixteca arc successions (southern Mexico): new insights for the
tectonic evolution of southwestern North America during Late Mesozoic:
Geolgical Society of America Bulletin, v. 119, no. 9/10, p. 1052 1065,
doi: 10.1130/B26016.1.
Tardy, M., Lapierre, H., Freydier, C., Coulon, C., Gill, J.B., Mercier de Lepinay,
B., Beck, C., Martinez, J., Talavera, M., Ortiz, E., Stein, G., Bourdier,
J.L., and Yta, M., 1994, The Guerrero suspect terrane (western Mexico)
and coeval arc terranes (the Greater Antilles and the Western Cordillera of
Colombia): A late Mesozoic intra-oceanic arc accreted to cratonal Amer-
ica during the Cretaceous: Tectonophysics, v. 234, p. 49–73.
Tristán-González, M., and Torres-Hernández, J.R., 1994, Geología del área
de Charcas, Estado de San Luis Potosí, 1994: Universidad Nacional
Autónoma de México, Instituto de Geología: Revista Mexicana de Cien-
cias Geológicas, v. 11, p. 117–138.
Urrutia-Fucugauchi, J., and Uribe-Cifuentes, R.M., 1999, Lower crustal xeno-
liths from the Valle de Santiago Maar Field, Michoacan-Guanajuato
Volcanic Field, Central Mexico: International Geology Review, v. 41,
p. 1067–1081.
Valencia-Moreno, M., Ruíz, J., and Roldán-Quintana, J., 1999, Geochemistry
of Laramide granitic rocks across the southern margin of the Paleozoic
North American continent, Central Sonora, Mexico: International Geol-
ogy Review, v. 41, p. 845–857.
Valencia-Moreno, M., Ruíz, J., Barton, M.D., Patchett, P.J., Zürcher, L., Hod-
kinson, D., and Roldán-Quintana, J., 2001, A chemical and isotopic study
of the Laramide granitic belt of northwestern Mexico: Identification of the
southern edge of the North American Precambrian basement: Geological
Society of America Bulletin, v. 113, p. 1409–1422, doi: 10.1130/0016–
7606(2001)113<1409:ACAISO>2.0.CO;2.
Vidal-Serratos, R., 1991, Estratigrafía y tectónica de la región de Zihuatanejo,
Estado de Guerrero, Sierra Madre del Sur: Convención sobre la evolución
Geológica Mexicana, 1er Congreso Mexicano de Mineralogía, Pachuca,
Memoir, p. 231–233.
Yañez, P., Ruiz, J., Patchett, P.J., Ortega-Gutiérrez, F., and Gehrels, G., 1991,
Isotopic studies of the Acatlan Complex, southern Mexico: Implications
for Paleozoic North American tectonics: Geological Society of America
Bulletin, v. 103, p. 817–828.
Yta, M., Lapierre, H., Monod, O., and Wever, P., 1990, Magmatic and struc-
tural characteristics of the Lower Cretaceous arc-volcano-sedimentary
sequence of Saucito-Zacatecas-Fresnillo (central Mexico), geodynamic
implications: Munich, Geowisenschaftliches Lateinamerika, Kolloquium,
Memoir, p. 21.11–23.11.
Zaldivar, R.J., and Garduño, M.V.H., 1984, Estudio estratigráfico y estructural
de las rocas del Paleozoico Superior de Santa Maria del Oro, Durango,
y sus implicaciones tectónicas [abstract]: Reunión Anual, Sociedad
Geológica de México, p. 37–38.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Tectonic architecture of an arc-arc collision zone, Newfoundland Appalachians

Alexandre Zagorevski*
Cees R. van Staal*
Vicki McNicoll
Neil Rogers
Geological Survey of Canada, 601 Booth Street, Ottawa, Ontario, K1A 0E8, Canada

Pablo Valverde-Vaquero
Instituto Geológico y Minero de España (IGME), La Calera 1, Tres Cantos (Madrid), 28760, Spain

ABSTRACT

The Appalachian-Caledonian orogen records a complex history of the closure of

the Cambrian−Ordovician Iapetus Ocean. The Dunnage Zone of Newfoundland pre-
serves evidence of an Ordovician arc-arc collision between the Red Indian Lake Arc,
which forms part of the peri-Laurentian Annieopsquotch accretionary tract (ca. 480–
460 Ma), and the peri-Gondwanan Victoria Arc (ca. 473–453 Ma). Despite the similar-
ity in age, the coeval arc systems can be differentiated on the basis of the contrasts
that are apparent across the suture zone, the Red Indian Line. These contrasts include
structural and tectonic history, stratigraphy, basement characteristics, radiogenic lead
in mineral deposits, and fauna. The arc-arc collision is considered in terms of modern
analogues (Molucca and Solomon Seas) in the southwest Pacific, and the timing is
constrained by stratigraphic relations in the two arc systems. The Victoria Arc occu-
pied a lower-plate setting during the collision and underwent subsidence during the
collision, similar to the Australian active margin and Halmahera arcs in the south-
west Pacific. The timing of the subsidence is constrained by three new ages of volcanic
rocks in the Victoria Arc (457 ± 2; 456.8 ± 3.1; 457 ± 3.6 Ma) that immediately predate
or are coeval with deposition of the Caradoc black shale. In contrast the Red Indian
Lake Arc contains a sub-Silurian unconformity and a distinct lack of Caradoc black
shale, suggesting uplift during the collision. The emergent peri-Laurentian terranes
provided detritus into the newly created basin above the Victoria Arc. The evidence of
this basin is preserved in the Badger Group, which stratigraphically overlies the peri-
Gondwanan Victoria Arc but incorporated peri-Laurentian detritus. Thus the Badger
Group forms a successor basin(s) over the Red Indian Line. Following the collision,
subduction stepped back into an outboard basin, the Exploits-Tetagouche backarc,
closing the Iapetus Ocean along the Dog Bay Line in the Silurian. Correlative tracts in

*Email, Zagorevski: azagorev@nrcan.gc.ca; Present address, van Staal: Geological Survey of Canada, 625 Robson Street, Vancouver, British Columbia V6B
5J3, Canada.

Zagorevski, A., van Staal, C.R., McNicoll, V., Rogers, N., and Valverde-Vaquero, P., 2008, Tectonic architecture of an arc-arc collision zone, Newfoundland Appa-
lachians, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of
America Special Paper 436, p. 309–333, doi: 10.1130/2008.2436(14). For permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of
America. All rights reserved.

309
310 Zagorevski et al.
the Northern Appalachians and British Caledonides support the Ordovician arc-arc
collision; however, the evidence is less obvious than in Newfoundland.
Keywords: arc-arc collision, Dunnage Zone, U/Pb geochronology, Sandbian and Katian
black shale, Red Indian Line.
INTRODUCTION
The Cambrian–Ordovician closure of the main tract of the
Iapetus Ocean generated a diverse set of arc-backarc terranes and
microcontinents that were accreted sequentially to the Lauren-
tian margin, leading to significant growth of the eastern margin
(present coordinates) of North America and the formation of
the Appalachian-Caledonian orogen. The reconstruction of the
paleogeography of the Iapetus Ocean requires identification of
the various terranes and their tectonic setting and has important
implications for a wide variety of topics ranging from the distri-
bution of mineral deposits to the tectonic development of conver-
gent margins and continental growth.
Newfoundland lies in a critical position in the Appalachians
because it forms the geological link between the Northern Appa-
lachian and the Caledonide segments of this once-continuous
orogen. As such the development of ideas in Newfoundland has
greatly influenced the evolution of thought on the tectonic devel-
opment of the Appalachian-Caledonian orogen as a whole. In
addition to occupying a key position in the orogen, Newfound-
land is also perhaps the best constrained portion of the Appala-
chians, featuring excellent coastal exposures, detailed mapping,
high-density geochemical, isotopic, geochronological, and geo-
physical data that include two crustal-scale seismic-reflection
transects in central Newfoundland (van der Velden et al., 2004).
The central mobile belt of Newfoundland comprises a
complex tectonic collage of arc-backarc terranes that formed
within the Iapetan realm outboard of Laurentia and Gondwana.
The juxtaposition of these terranes was piecemeal, involving
several arc-continent and arc-arc collisions, with paleogeo-
graphic complexity similar to the modern arcs in the southwest
Pacific (e.g., van Staal et al., 1998). Similar to the southwest
Pacific (Hall, 2002), the accretionary processes in the Iapetus
operated on very short time scales, with the accretion of many
arc-backarc terranes that occurred within 5–10 m.y. following
their formation (Lissenberg et al., 2005a; Zagorevski et al.,
2006, 2007a). Consequently, the resolution of these processes
requires highly detailed geochronology, which is available in
central Newfoundland.
This paper reviews the evidence for an Ordovician arc-arc
collision recorded in the central mobile belt of the Newfound-
land Appalachians. Three new U-Pb zircon geochronologi-
cal ages constrain the timing of volcanism and sedimentation
in the Exploits Subzone and the age of the arc-arc collision.
Select structural data for central Newfoundland is presented
and integrated into a tectonic model, which also incorporates
the seismic-reflection data of van der Velden et al. (2004). This
tectonic model is considered in the context of the tectonics and
sedimentation in recent active arc-arc collisions in the south-
west Pacific, specifically the Huon-Finisterre (Solomon Sea)
and Halmahera-Sangihe (Molucca Sea) arc-continent and arc-
arc collisions (e.g., Abers and McCaffrey, 1994; Pubellier et al.,
1999; Whitmore et al., 1999).
Modern Arc-Arc Collisions
Prior to reviewing the evidence for the arc-arc collision in
central Newfoundland we briefly introduce two modern ana-
logues and review the key evidence that would enable the rec-
ognition of arc-arc collisions in ancient orogens. We have cho-
sen the Huon-Finisterre arc-continent collision in the Solomon
Sea (Abbott et al., 1994; Abers and McCaffrey, 1994) and the
Halmahera-Sangihe arc-arc collision in the Molucca Sea (Lalle-
mand et al., 1998; Pubellier et al., 1999) because they are active
and preserve different stages of collision along the strike of the
orogen. In addition, we believe that they most closely resemble
the relationships in Newfoundland.
Solomon Sea
The Solomon Sea plate is currently subducting under the Bis-
marck Arc to the north (New Britain Trench) and under the active
Australian continental margin to the south (Trobriand Trough:
Fig. 1; e.g., Abers and McCaffrey, 1994). In the western Solo-
mon Sea the Australian-Bismarck-Solomon-plate triple junction
forms the site of a modern arc-arc collision where the Bismarck
Arc is colliding with the active Australian margin. The collision is
diachronous from the northwest to the southeast and has occurred
northeast of the triple junction, forming the Huon and Finisterre
Ranges on the Huon Peninsula. The deformation is accommo-
dated by southwest-directed thrusting of the Bismarck Arc and
its accretionary prism over the Australian active margin along the
Ramu-Markham Fault, a continuation of the New Britain Trench
on land. An important out-of-sequence thrust, the Gain Thrust,
cuts across the Ramu-Markham Fault and emplaces the volcanic
rocks of the Bismarck Arc over its accretionary prism (Fig. 2).
Continuation of movement on the Gain thrust may in time struc-
turally mask the accretionary prism, leading to the juxtaposition
of the Bismarck Arc and active Australian margin.
As a result of the collision and loading by the Bismarck Arc,
parts of the Australian continental shelf underwent rapid subsid-
ence (2 km over the last 348 m.y.: Galewsky et al., 1996) indi-
cated by the presence of the drowned carbonate platform above
 144º 146ºº 148ºº 150ºº 152º 154º 156º 120 º 125 º 130 º
2º

2º
Bismarck Sea

Philippine
Sea
South Bismark Plate
4º

4º

Ph

5º
ilip
5º
Bismarck Arc
SA

p
ine
Hu
on
Celebes Sea SP

Tre
-
Ra Fin New Britain

n
i
Ma nge ster
6º

North Sulawes
nch

ch
6º
i
r s re Tre Trench
Fa kham Mark ri tain
ult ham
C wB HA
d anyoncp Ne Solomon Sea

0º
0º
Trob
Pa

r iand Molucca
pu

MS Trou Sea
an
8º

gh

8º
Pe
ni

Sul
ns
ul

North
a

awe
Woodlark Basin Banda
Sea

si
Australian
10º

Banda

10º
Plate

5º
5º
Sea
144º 146º 148º 150º 152º 154º 156º 120 º 125 º 130 º
Subduction zone/
Volcano Fault
reverse fault

Figure 1. Simplified tectonic map of Solomon Sea and Molucca Sea. Digital base modified from Bird (2003); boundaries modified from Abers
and McCaffrey (1994), Pubellier et al. (1999), Milsom (2001), and Bird (2003). Abbreviations: dcp—drowned carbonate platform; HA—Halma-
hera Arc; SA—Sangihe Arc; SP—Snellius Plateau; MS—Morobe Shelf.

A Molucca Sea collision zone B Solomon Sea collision zone

Ramu-Markham Gain Bismarck
Celebes Sangihe Pujada Snellius Fault Thrust Arc
Sea Arc Ridge Plateau

7ºN Australian

?
Active Margin
Figure 2. Synthetic sections of the Mo-
148º10′ E lucca Sea and Solomon Sea collision
zones, demonstrating key tectonic rela-
Molucca Slab
tionships. (A) Based on Lallemand et
Sea Plate
al. (1998) and Pubellier et al. (1999).
(B) Modified best gravity fit model of
Markham
Cotabato Sangihe Sangihe Philippine Trobriand New Britain Abbott et al. (1994). Projection of the
Canyon
Trench Arc forearc Trench Trough Trench slab at depth at latitude 148°N is based
turbidites
on Abers and McCaffrey (1994). Large
s
black arrows indicate uplift of the Huon-
Snelliu u Finisterre terrane and subsidence of the
a
5ºN P la t e ? Solomon ? Morobe Shelf west of the cross section.
Sea Plate
Molucca
Sea Plate
148º55′ E

sedimentary overriding subducting accretionary oceanic upper recent

basin arc arc wedge crust mantle volcano
312 Zagorevski et al.
the Morobe Shelf (Fig. 1; Galewsky and Silver, 1997). The degree
of subsidence of the Australian margin decreases to the east of
the Australia-Bismarck-Solomon-plate triple junction, where the
collision has not yet occurred (Galewsky and Silver, 1997). The
emergent orogen to the west of the triple junction supplies abun-
dant sediment into the Solomon Sea. The detritus is dominantly
transported as turbidity currents parallel to the trench along the
Markham Canyon and is deposited on the subducting Molucca
Sea plate and Australian margin (Galewsky and Silver, 1997).
Molucca Sea
The Molucca Sea is the site of the collision between the fac-
ing Sangihe and Halmahera Arcs, which formed as a result of the
double subduction of the Molucca Sea plate to the west and east,
respectively (Fig. 1; e.g., Hall, 2002). The collision initiated in
the northern Molucca Sea during the Pliocene and migrated to
the south, where the collision is still in its infancy (Pubellier et
al., 1999; Hall, 2000). The Molucca Sea plate has been entirely
overridden by the collisional complex derived primarily from the
Sangihe forearc. In the southern Molucca Sea, where the collision
is the least advanced, the Halmahera forearc is in part overridden
by the Sangihe forearc. To the north, the Halmahera forearc is
completely overridden by the Sangihe forearc, and the Snellius
Plateau, the continuation of the Halmahera Arc to the north, is
subducting at present (Lallemand et al., 1998; Pubellier et al.,
1999; Hall, 2000). As a result of subduction, the Snellius Plateau
underwent rapid subsidence, as evidenced by the presence of a
drowned carbonate shelf, whereas parts of the overriding colli-
sional complex have been uplifted above sea level (Talaud Island
and Pujada Ridge: Figs. 1, 2).
The Molucca Sea plate is entirely overridden by the colli-
sional complex but is still actively subducting in the south Molucca
Sea. To the north, evidence from deep earthquakes suggests that
the Molucca Sea plate has delaminated from the overriding plate
beneath Mindanao (Lallemand et al., 1998). Mindanao is char-
acterized by widespread Pliocene and younger syn- to postcol-
lisional basaltic to dacitic magmatism. The compositions of the
magmas range from calc-alkaline to shoshonitic. The diversity of
the magmas and the geochemistry has been related to the delami-
nation of the Molucca Sea plate (Sajona et al., 2000).
Recognition of Ancient Arc-Arc Collisions
The juxtaposition of coeval arc terranes, such as those
described for the Molucca and Solomon Seas, presents a dif-
ficulty with respect to their identification and interpretation in
ancient orogens. In the case of both collision zones described
above, the facing arc terranes may be differentiated on the basis
of the (1) tectonic history and basement characteristics, (2) geo-
chemistry, and (3) stratigraphic record. (1) Prior to the collision
the Halmahera Arc-backarc underwent a period of shortening that
resulted in thrusting of the backarc rocks onto the arc and a gap
in volcanism, whereas the Sangihe Arc was not shortened during
this time. The Australian margin in the Solomon Sea area was not
active until recently, whereas the basement to the Bismarck Arc
should have undergone several tectono-thermal events relating
to protracted arc volcanism, based on the reconstructions of Hall
(2002). (2) The geochemistry of the volcanic rocks of the Halma-
hera and Sangihe Arcs shows distinct characteristics (Macpher-
son et al., 2003), which may be utilized in their separation.
(3) The pre- and syncollisional stratigraphy may prove extremely
useful in differentiation of the arc sequences. Both the Australian
margin and the Halmahera Arc–Snellius Plateau were drowned
prior to and during the collision, while the overriding arcs and
forearc basins were undergoing uplift and erosion. Emergence
of the overriding arc and collisional orogen supplies syntectonic
sediment into the basin created above the subducting arc. The
appearance of sediment with provenance consistent with deriva-
tion from the overriding arc thus would provide information on
the timing of the emergence of the orogen. Finally, the delamina-
tion of the subducting oceanic plate may lead to extensive syn- to
postcollisional magmatism, marking the end of the collision, such
as in Mindanao (Sajona et al., 2000).
The recognition of ancient arc-arc collision zones similar to
those in the southwest Pacific thus would require an approach
that integrates stratigraphy, geochemistry, geochronology, struc-
tural geology, and geophysics, combined with a good understand-
ing of basement-cover relationships and affinity of the various
terranes. We believe that the quality, quantity, and distribution
of data make the Newfoundland Appalachians unique among
ancient orogens in that they permit detailed reconstruction of the
tectonic architecture of an Ordovician arc-arc collision.
GEOLOGICAL SETTING
The complexity of the closure of the Iapetus Ocean is
reflected in the zonal division of the northern Appalachians,
where five zones have been defined: Humber, Dunnage, Gander,
Avalon, and Meguma (Fig. 3; Williams, 1995b). The Humber
Zone represents the Cambrian–Ordovician Laurentian margin.
The Dunnage Zone (the central mobile belt) lies outboard of the
Humber Zone and contains allochthonous ensialic and ensimatic
arc-backarc complexes that formed within the Iapetus Ocean. Far-
ther outboard, the Gander, Avalon, and Meguma Zones represent
microcontinents derived from Gondwana (see van Staal, 2005,
for a review). The margin of Laurentia experienced three oro-
genic episodes from Ordovician to Early Devonian time, namely
the Taconic, Salinic, and Acadian. These episodes occurred as a
result of the successive arrival of the Dashwoods, Gander, and
Avalon microcontinents (van Staal, 2005).
The Dunnage Zone is subdivided into the peri-Laurentian
Notre Dame and the peri-Gondwanan Exploits Subzones (Fig. 3;
Williams et al., 1988). Other subsequent subdivisions (Williams,
1995a) reflect minor regional differences internal to these two
subzones that are beyond the scope of this paper. The Notre Dame
Subzone comprises predominantly Early to Middle Ordovician
Notre Dame Arc plutons that intruded a ribbon microcontinent,
referred to as the Dashwoods microcontinent (Waldron and van
Staal, 2001), as well as accreted peri-Laurentian terranes. The
 Tectonic architecture of an arc-arc collision zone 313

59° 57° 55° 53°

Laurentian
51° N 51°
Margin
Humber Zone

Dunnage Zone Dunnage Zone

50° 50°
Notre Dame Subzone
Gulf of
e
on

St. Lawrence Notre Dame

rZ

Bay Gondwanan
be

Margin
m

e e
Hu

on m

L
bz D a

RI Dunnage Zone:
Su tre

49° 49°
o

Exploits Subzone
N

e
Exploits Zon
Subzone Gander Zone
er
nd

Z Figure 3. Tectono-stratigraphic subdivi-

LS
Ga

V sions of Newfoundland (after Williams,

ne

Avalon Zone
Zo

1995a) and northern Appalachians

n
o

(from Hibbard et al., 2004). 1—Popelo-

al

Meguma Zone
Av

gan inlier; 2—Bathurst Supergroup;

3—Chain Lakes Massif and Jim Pond
Formation; 4—Shelburne Falls Arc;
50 0 50 100
47° 5—Bronson Hill Arc; RIL—Red Indian
Kilometres
Line; VLSZ—Victoria Lake shear zone;
BVBL—Baie Verte – Brompton Line.
59° 57° 55° 53°
45 50
°W
°W

°W

°N °N
70

60
65

BVBL
RIL
1
3
2

50
CF BVBL °N
4 5 RIL

BBF
CB
40 F
°N DHF
°W
°W

40 45
°W

°N °N
60
65
70

Exploits Subzone contains peri-Gondwanan ensialic and ensi-
matic arc-backarc complexes. The two subzones are differenti-
ated on the basis of faunal, paleomagnetic, isotopic, and struc-
tural contrasts (Williams et al., 1988).
Notre Dame–Exploits Relationships in Newfoundland
The identification of the marked geological contrasts
between the Notre Dame and Exploits Subzones has led to rec-
ognition of the Red Indian Line (Fig. 3; Williams et al., 1988),
which is now accepted as the main suture zone in the north-
ern Appalachians, along which several thousand kilometers of
Iapetus oceanic lithosphere has been consumed (e.g., van Staal,
2005). Following the formation of the Red Indian Line the con-
vergence continued by means of subduction in Iapetan marginal
basins and seaways underlain by oceanic crust (e.g., Exploits-
Tetagouche backarc basin: van Staal et al., 1998). In Notre
Dame Bay the Red Indian Line is marked by late brittle faults
and locally by Ordovician mélange (McConnell et al., 2002;
Williams et al., 1988), although the exact nature and location of
this boundary is debated (e.g., Arnott et al., 1985; Wasowski et
al., 1986; van der Voo et al., 1991). In central Newfoundland the
surface trace of the Red Indian Line is marked by a mélange-
phyllonite belt (Fig. 3; Rogers and van Staal, 2002; Zagorevski
et al., 2006, 2007b), which coincides with the surface projec-
tion of a major crustal-scale reflector, interpreted to be a fault,
in seismic-reflection profiles (van der Velden et al., 2004). The
surface trace of the Red Indian Line is cut out in central New-
foundland by the Devonian Victoria Lake shear zone (Fig. 3;
Valverde-Vaquero and van Staal, 2002), which is a continuation
314 Zagorevski et al.

of the Noel-Paul’s Line (Williams et al., 1988). At depth the Red
Indian Line is truncated by a seismic reflector interpreted to be a
Devonian wedging-related fault (van der Velden et al., 2004).
Notre Dame Subzone
The Notre Dame Subzone in Newfoundland comprises
several composite terranes formed outboard of the Lauren-
tian margin. The western margin and central part of the Notre
Dame Subzone is dominated by the Dashwoods microconti-
nent (Waldron and van Staal, 2001) and its Notre Dame Arc
suprastructure. The eastern margin of the Notre Dame Sub-
zone contains a tectonic collage of continental and intraoce-
anic supra-subduction zone complexes collectively referred to
as the Annieopsquotch accretionary tract (Figs. 4–6; van Staal
et al., 1998; Lissenberg et al., 2005b; Zagorevski et al., 2006).
The Annieopsquotch accretionary tract formed above a west-
dipping subduction zone and was facing the main Iapetan basin.
As such, it contains the most outboard arc-backarc complexes
of the Notre Dame Subzone and provides constraints on the
development of the Notre Dame Subzone and its juxtaposition
with the Exploits Subzone.
Subduction in the Annieopsquotch accretionary tract initi-
ated at ca. 480 Ma, leading to the formation of the Annieops-
quotch ophiolite belt (Dunning and Krogh, 1985; Dunning,
1987; Lissenberg et al., 2005b). This may have occurred in a
reentrant in the Dashwoods margin (Waldron and van Staal,
2001; Zagorevski et al., 2006). Subsequently, the subduction
developed under the Dashwoods margin, forming a continen-
tal arc and backarc by ca. 473 Ma (Buchans Group and Lloyds
River ophiolite complex: Dunning et al., 1987; Swinden et al.,
1997; Zagorevski et al., 2006). This arc rifted, forming an oce-
anic basin (Pickett, 1987) and a younger arc, represented by the
Red Indian Lake Group in central Newfoundland (Zagorevski
et al., 2006).
Red Indian Lake Group
The Red Indian Lake Group comprises mafic, andesitic, and
felsic volcanic rocks, epiclastic sedimentary rocks, red shale, and
ferruginous chert. Tholeiitic mafic volcanic rocks are predominant
at the stratigraphic base, while calc-alkaline mafic, andesitic, and
felsic volcanic rocks become more common at the stratigraphic
top, suggesting a maturing arc sequence (Zagorevski et al., 2006).
Negative εNd values of felsic volcanic rocks (−6.4 to −7.7) and

N 57° 55° L
RI
12
11 13
10
49°
B
U 49°
R
G

5
6
8
4 9
7
3
1

48°
48°
2

57° 55°

Annieopsquotch Badger Victoria-Exploits forearc, Penobscot

accretionary tract Group arc and backarc volcanic Arc
and sedimentary rocks

Figure 4. Distribution of Annieopsquotch accretionary tract, Victoria Arc, Penobscot basement, and Badger Group in
Newfoundland (after Colman-Sadd et al., 1990). 1—Dashwoods Subzone; 2—Windsor Point Group; 3—Annieops-
quotch ophiolite belt; 4—Lloyds River ophiolite complex; 5—Buchans Group; 6—Red Indian Lake Group; 7—Pats
Pond Group and Wigwam Brook Group; 8—Sutherlands Pond group; 9—Tally Pond Group and Noel Paul’s Brook
group; 10—Exploits Group; 11—Wild Bight Group; 12—Summerford Group; 13—Dunnage mélange; GRUB—Gan-
der River Ultrabasic Belt; RIL—Red Indian Line.
 Red Indian Exploits Subzone
Notre Dame Subzone Line
Llandovery Wenlock
Silurian

Springdale Group and equivalents Botwood Group

430
Silurian
440

Badger Group
*
Llanvirn Caradoc Ash.
450

Ordovician
Upper

Widsor Point
Group (2)

WBG
460

NPBg

VICTORIA ARC
Figure 5. Relationships between selected
Ordovician Ordovician

WwBG
tectono-stratigraphic units in New-
Middle
Age (Ma)

Sutherlands
Pond group (8)
foundland. Subdivision of the Ordovi-
Summerford Group
(accreted seamount; 12)
470

Red Indian
Lake Group (6)
Arenig

cian Period into epochs follows McKer-

row and van Staal (2000). Numbers in
Dashwoods
Subzone(1)

Hgs parentheses refer to the locations shown

Buchans
Group (5)
Lloyds River
Ophiolite Complex (4)
Lower
480

in Figure 4. Hgs—Harpoon gabbro

Tremadoc

suite; NPBg—Noel Paul’s Brook group;

Dunnage
melange (13)

Penobscot orogeny
Annieopsquotch
ophiolite belt (3)

WwBG—Wigwam Brook Group. See

text for discussion.
490

Cambrian

*
Upper

Pats Pond
Group (7)

PENOBSCOT ARC
Wild Bight
Group (11)
Exploits
Group
500

(10)
Annieopsquotch accretionary
Cambrian

tract
Middle

Tally
Pond
510

Group
(9)
Victoria Lake Supergroup
Ophiolite Peri-Gondwanan Penobscot
Black shale
volcanic rocks arc
Peri-Laurentian Sedimentary
Limestone Plutonic rocks
volcanic rocks rocks
U/Pb zircon age Hiatus

Map Buchans
Unclassified Red Indian Location
contact Line
Lithoprobe Red Indian Lake
Thrust fault seismic 48o45’
reflection 56o30’
profiles
Notre Dame Subzone
Annieopsquotch
accretionary tract o
57 15' Figure 6. Simplified geology of cen-
tral Newfoundland (after Zagorevski et
Exploits Subzone S1
al., 2006). Stereonets are lower hemi-
Victoria Lake Supergroup n=162
sphere, equal area, contoured using
Victoria arc 48 30’
o
the Kamb method, with the number of
o
57 00' points indicated. Stereonets of S1 are
Penobscot arc for the indicated portions of the mapped
ak
e area, south and southwest of Red Indian
sL S1 Lake. The S2 stereonet is for the entire
10 km d
L loy n=289 mapped area south and southwest of
57o45' Red Indian Lake.
to ria Lake
Vic
o
S2
48 15'
n=190
57o30’ S1
n=284
316 Zagorevski et al.
abundant Mesoproterozoic zircon inheritance suggest that this arc
formed on continental crust. The range of U/Pb zircon ages of the
volcanic rocks of the Red Indian Lake Group (462 +2/–9, 463
± 3, 464.8 ± 3.5, 465 ± 2 Ma: Zagorevski et al., 2006) indicates
that arc volcanism is predominantly Llanvirn (i.e., 459–465 Ma:
McKerrow and van Staal, 2000). The Red Indian Lake Group is
the most outboard and the youngest sequence identified within
the Annieopsquotch accretionary tract. Correlatives of the Red
Indian Lake Group likely occur throughout the Annieopsquotch
accretionary tract and may include the undated Crescent Lake
Formation (Fig. 3; Bostock, 1988) and parts of the Cottrell’s Cove
Group (Arenig: in O’Brien, 2003; i.e., 465–480 Ma: McKerrow
and van Staal, 2000) in the Notre Dame Bay region. The Red
Indian Lake Group and its correlatives are henceforth referred to
as the Red Indian Lake Arc, which comprises the youngest arc
sequence in the Annieopsquotch accretionary tract.
Notre Dame Subzone Cover
The Notre Dame Subzone is unconformably overlain by the
volcaniclastic and sedimentary rocks of the spatially restricted
Windsor Point Group in southwestern Newfoundland (Fig. 4;
453 +5/–4 Ma: Dube et al., 1996); however, the most wide-
spread unconformity is sub-Silurian. The Notre Dame Subzone
is regionally overlain unconformably by the rocks of the Silurian
Springdale Group or its equivalents (Chandler et al., 1987; Dun-
ning et al., 1990). In central Newfoundland these rocks comprise
felsic to andesitic volcanic rocks, breccia, polymictic conglom-
erate, and red sandstone (e.g., van Staal et al., 2005a, b). The
provenance of the polymictic conglomerate suggests derivation
predominantly from the Notre Dame Subzone. Some clasts bear
a predepositional foliation, indicating deformation and exhuma-
tion of the Ordovician tectonites by Silurian time.
Exploits Subzone
To the east of the Red Indian Line the Exploits Subzone in
Newfoundland can be subdivided into two distinct arc-related
sequences (Fig. 4; Jenner and Swinden, 1993; O’Brien et al.,
1997; MacLachlan and Dunning, 1998a, b; van Staal et al.,
1998; Rogers et al., 2006; Zagorevski et al., 2007a). The older
sequence, the Penobscot arc and backarc, contains ca. 513–
485 Ma volcaniclastic and sedimentary rocks that formed above
an east-dipping subduction zone along the margin of Ganderia
(Rogers et al., 2006; Zagorevski et al., 2007a), a postulated peri-
Gondwanan microcontinent (van Staal et al., 1996). A gap in
magmatism in the Exploits Subzone (485–478 Ma) is correlated
with the collision of the Penobscot arc with Ganderia at ca. 480–
485 Ma and obduction of the Penobscot backarc ophiolites (Jen-
ner and Swinden, 1993) onto the Newfoundland Gander Zone
(Fig. 3; e.g., Colman-Sadd et al., 1992; Tucker et al., 1994; van
Staal et al., 1998). The Penobscot arc is stratigraphically over-
lain by ca. 473–453 Ma volcano-sedimentary rocks (Fig. 4; e.g.,
MacLachlan and Dunning, 1998b; O’Brien et al., 1997; Evans
and Kean, 2002; Zagorevski et al., 2007a), which are referred to
as the Victoria Arc in central Newfoundland. The Victoria Arc is
a correlative of the Popelogan arc of New Brunswick, Canada
(Fig. 3: Popelogan inlier; van Staal et al., 1998).
Victoria Arc
The tectonic setting of the Victoria Arc, and its correla-
tive Popelogan arc in New Brunswick, are well constrained in
the northern Appalachians (e.g., van Staal et al., 1998). The
earliest Victoria Arc volcanism in Newfoundland occurred at
ca. 473 Ma (Fig. 4; Wild Bight Group: MacLachlan and Dun-
ning, 1998b). Coeval eruption of arc- and backarc-related vol-
canic rocks (Exploits Group: O’Brien et al., 1997; Wild Bight
Group: MacLachlan and Dunning, 1998b) indicates an exten-
sional phase of arc magmatism, which led to the rifting of the
Victoria and Popelogan arcs from Ganderia and the opening of
the Exploits-Tetagouche backarc basin in Newfoundland and
New Brunswick (Fig. 3; O’Brien et al., 1997; MacLachlan and
Dunning, 1998b; van Staal et al., 1998; Valverde-Vaquero et al.,
2006; Bathurst Supergroup: van Staal et al., 2003). Abundant fos-
sil and some age-dating evidence indicate that the Victoria Arc
was active until at least middle Llanvirn time (e.g., O’Brien et
al., 1997). For example, Victoria Arc–related basalts of the Sops
Head Complex in Notre Dame Bay form peperitic contacts with
Llanvirn limestone (McConnell et al., 2002).
In central Newfoundland the Victoria Arc is represented by
the Sutherlands Pond group (ca. 457–462 Ma: Dunning et al.,
1987; Rogers et al., 2005; see following), Noel Paul’s Brook group
(ca. 457–465 Ma: van Staal et al., 2005c), and Wigwam Brook
Group (453 ± 4 Ma: Zagorevski et al., 2007a). The youngest dated
Victoria Arc–related volcanic rocks in the lower Wigwam Brook
Group (van Staal et al., 2005c; Zagorevski et al., 2007a) comprise
a coarsening-up sequence of felsic tuff and breccia that are inter-
calated with turbiditic sandstone, conglomerate, black shale, and
basalt with arclike chemistry. The upper Wigwam Brook Group
comprises turbiditic sandstone and black shale, suggesting cessa-
tion of arc magmatism (Zagorevski et al., 2007a).
Exploits Backarc
Similarly to the Victoria Arc, the Exploits-Tetagouche back-
arc was magmatically active until the late Llanvirn (e.g., Law-
rence Head Formation: O’Brien et al., 1997). The coeval sedi-
mentation in the Exploits backarc was continuous from the early
Arenig until the Caradoc (Caradoc: 449–459 Ma: McKerrow and
van Staal, 2000) and mostly consisted of deep-marine sediments
interrupted by occasional pulses in magmatism, such as the erup-
tion of backarc-basin basalts of the Laurence Head Formation
(O’Brien et al., 1997). The end of the Llanvirn was marked by
cessation of magmatism, deposition of chert and regionally dis-
continuous limestone, followed by the black shale of the Law-
rence Harbour Formation (Williams, 1995a).
In central Newfoundland, backarc volcanism is preserved in
the Red Cross Group, which comprises felsic volcanic rocks (466
± 3 Ma: Valverde-Vaquero et al., 2006), basalt, turbiditic sandstone,
black shale, and calcareous shale locally intruded by gabbro (457
 Tectonic architecture of an arc-arc collision zone
± 3 Ma: Valverde-Vaquero et al., 2006). Similar to the rest of the
Exploits Subzone, the upper parts of the Red Cross Group com-
monly contain abundant black shale and calcareous shale.
Exploits Subzone Cover
One of the key characteristics of the Exploits Subzone is
the presence of the regionally transgressive Caradoc black shale
cover, accompanied by the general cessation of volcanism (Fig. 5;
van der Pluijm et al., 1987; Williams, 1995a). In the Notre Dame
Bay area (Fig. 3), this time was marked by deposition of the black
shale of the Lawrence Harbour and Shoal Arm Formations (e.g.,
Williams et al., 1995; O’Brien et al., 1997). In central Newfound-
land the Wigwam Brook Group (Zagorevski et al., 2007a) and
Stanley Waters formation (see following discussion; van Staal et
al., 2005c) mirror the relationships in Notre Dame Bay and indi-
cate cessation of volcanism accompanied-followed by deposition
of the Caradoc black shale.
The cessation of arc magmatism and deposition of black shale
are commonly assumed to have occurred at the Llanvirn–Caradoc
boundary (e.g., O’Brien, 2003). However, in the Wigwam Brook
Group this transition occurred in the middle Caradoc (453 ± 4 Ma:
Zagorevski et al., 2007a). Volcanism of similar age may also have
been the source of felsic volcanic boudins, which are common in
the mélange belts of central Newfoundland (Rogers and van Staal,
2002; Valverde-Vaquero et al., 2006). The cessation of volcanism
and the subsidence of the Victoria Arc thus appear to have been
somewhat diachronous along strike, and may have started earlier
in Notre Dame Bay. Alternatively this may be an artifact of sparse
dating in Notre Dame Bay or of local basin architecture such that
parts of the basin were effectively isolated from the volcaniclastic
input and received only background sedimentation.
Regionally the Ashgill to Wenlock (424–449 Ma: McKer-
row and van Staal, 2000) Badger Group (Williams et al., 1993)
stratigraphically overlies the Caradoc black shale and comprises
an upward-coarsening sedimentary sequence of deep-marine tur-
bidites to shallow-marine conglomerate. Structural and sedimen-
tological investigations of the Badger Group indicate deposition
in a syntectonic setting (e.g., Kusky et al., 1987; Williams et al.,
1995). The detrital provenance of the sedimentary rocks suggests
derivation from the Notre Dame Subzone (see Williams et al.,
1995), confirmed by the U/Pb detrital zircon data (McNicoll et
al., 2001). The syntectonic setting and provenance have led to the
interpretation of the Badger Group as a successor basin or basins
over the Red Indian Line in the forearc region (arc-trench gap)
of the Late Ordovician to Early Silurian subduction complex
formed during the closure of the Tetagouche-Exploits backarc
basin (Kusky et al., 1987; Pickering, 1987; van Staal et al., 1998;
Valverde-Vaquero et al., 2006).
STRUCTURAL GEOLOGY
Central Newfoundland displays a relatively simple macro-
scopic structure with internally folded structural panels bounded
by curviplanar shear zones (Fig. 6). In the Annieopsquotch
317
accretionary tract, for example, the thickness of the individual
tectonic panels varies between 0 and 5 km, indicating partial to
total excision of tectono-stratigraphic units along strike. Detailed
investigations of the structural geology indicate that the macro-
scopic simplicity is deceptive and that the rocks have undergone
multiple phases of deformation. Seven deformation episodes
(D1 through D7) have been recognized in central Newfoundland,
based on overprinting relationships as well as available age and
stratigraphic constraints (Zagorevski et al., 2007b). The regional
geometry of the rocks is primarily the result of D1, D2, and D4 epi-
sodes, although only D1 structures formed during the arc-arc col-
lision. Hence, only these are discussed below. D3, 5–7 so far have
been shown to be present only on a mesoscopic scale (Zagorevski
et al., 2007b) and therefore are not imperative for this chapter.
The oldest phase of deformation identified in central New-
foundland (D1) was heterogeneous and led to strain localization
into northwest-dipping brittle to ductile shear zones ten to several
hundred meters in thickness. These deformation zones are char-
acterized by mylonite, phyllonite, and/or black shale mélange that
mark the boundaries between the various tectono-stratigraphic
units (Figs. 6, 7). In proximity to the Red Indian Line in cen-
tral Newfoundland, D1 shear zones accommodated SSE-directed
motion (Lissenberg, 2005; Zagorevski et al., 2007b).
D1 shear zones are overprinted by upright moderate- to
shallow-plunging asymmetric F2 folds and axial planar S2 folia-
tion. S2 is commonly a composite foliation formed by transposi-
tion of S1 by tight to isoclinal F2 folds (Figs. 6, 7). The composite
S1–2 is the regionally dominant foliation. The enveloping surface
of the F2 folds dips northwest, suggesting that the D1 shear zones
were originally shallow northwest-dipping structures prior to the
D2 structures. In addition, the emplacement of “old over young”
rocks suggests that the D1 shear zones represent thrust faults
(Fig. 8; Zagorevski et al., 2007b). This interpretation is supported
by areas of weak D2 strain, such as Buchans (Fig. 6), where a
southeast-directed antiformal thrust stack is developed (Calon
and Green, 1987).
D2 strain is commonly concentrated along D1 shear zones
where the D1 tectonites provided a strong planar anisotropy. This
resulted in transposition of S1 and reactivation of D2-steepened
D1 shear zones as southeast-directed reverse faults. Similar to D1
shear zones, D2 reverse faults also accommodated SSE-directed
translation. The differentiation of D1 and D2 structures in areas
of high D2 strain thus can be tenuous in the absence of overprint-
ing relationships.
D4 structures include S4 foliation, F4 folds, and D4 shear
zones that overprint D1 and D2 structures. Although the D4 struc-
tures timewise are not related to the arc-arc collision discussed
in this paper, they are locally paramount to the kinematic inter-
pretation of the geometry of the deformed rocks. D4 structures
were associated with formation of a northwest-directed thrust
and fold belt in central Newfoundland. An excellent example of
a D4 structure is the Devonian Victoria Lake shear zone (Fig. 6;
Valverde-Vaquero et al., 2006). This shear zone emplaces the
metamorphic tectonites of the Gander Zone over the Exploits
318 Zagorevski et al.

A B C D

S1 S1

S5
F1
S2
S4 (composite) S2

Figure 7. Representative relationships between D1 and D2 structures. (A) Horizontal outcrop of polydeformed black shale mélange with a basaltic
knocker marking the boundary between the Sutherlands Pond group and the Penobscot arc. (B) Horizontal outcrop of highly deformed Ordovi-
cian limestone along the same boundary, with sinistral boudinage of a vein. (C) Vertical photomicrograph of F2-folded D1 metabasic tectonite
along the boundary between Lloyds River ophiolite complex and Annieopsquotch ophiolite belt (see Fig. 5). (D) Vertical outcrop demonstrating
the relationships between F1 folded quartz veins and upright F2 folds in felsic tuff of the Red Indian Lake Group.

A B Notre AAT RIL Victoria

RIL
N Dame Arc Lake Sg.

10
Ganderia
Laurentia 20
B
RIL 30

Moho 40

C Notre
50 km

C Dame Arc RIL

10
Ganderia
20
Laurentia Laurentia 30
40
50 km

Figure 8. Interpretation of D1-D2 relationships in central Newfoundland. (A) Interpretation of the three-dimensional geometry of the Red Indian
Line, based on the asymmetry of F2 folds, which indicate a northwest-dipping enveloping surface. Inset shows a close-up of the central area with
folded shear zones (light gray) and folded unconformity between the Pats Pond and Wigwam Brook groups (dark gray). Location of seismic-
reflection profiles is indicated by panels B and C. Interpretation of the D1-D2 relationship in panel A is consistent with the seismic-reflection
profiles shown in panels B and C (simplified from van der Velden et al., 2004). AAT—Annieopsquotch accretionary tract; RIL—Red Indian
Line; Sg—supergroup.

Subzone and in part buries the D1 and D2 shear zones, including
the Red Indian Line.
Ages of D1 and D2 Deformation Zones
The ages of the D1 zone have been obtained from the syn-
tectonic intrusions in the westernmost units of the Annieops-
quotch accretionary tract. These syntectonic intrusions span
from Arenig to early Caradoc (ca. 470 Ma Ar/Ar hornblende;
468 ± 2 Ma U/Pb zircon; ca. 464–458 Ma U/Pb zircon: Lissen-
berg et al., 2005a). The oldest ages of D1 deformation predate
the youngest peri-Laurentian arc volcanism in the Annieops-
quotch accretionary tract (Fig. 5; i.e., Red Indian Lake Group:
Zagorevski et al., 2006). Therefore, Arenig to early Llanvirn
deformation is strictly peri-Laurentian and is not related to the
accretion of any peri-Gondwanan terranes. This deformation is
probably a result of back thrusting that initiated following the
Taconic collision of the Dashwoods microcontinent with the
 Tectonic architecture of an arc-arc collision zone 319

Laurentian margin at ca. 470 Ma (Waldron and van Staal, 2001;
van Staal et al., 2006).
The earliest constraints on the start of deformation that
affected both the Notre Dame and Exploits Subzones are pro-
vided by the stratigraphic relationships. The dismemberment of
the Caradoc black shale unit into mélange and broken formation
along the Red Indian Line, without incorporation of Ashgill-age
rocks, suggests that the D1 deformation that affected both sub-
zones was under way in the late Caradoc (also observed in the
Sops Head–Boones Point Complex: McConnell et al., 2002).
Hence, although the D1 deformation was protracted, lasting from
at least 470 Ma until at least 450 Ma, only the youngest part of
the D1 deformation is related to the accretion of peri-Gondwa-
nan terranes.
The upper limit of D1 deformation and the span of D2 defor-
mation are constrained by the Silurian plutonic rocks (435–427
Ma: Dunning et al., 1990; Whalen et al., 2006; Zagorevski et
al., 2007b). Some of these plutons stitch major D1 shear zones
and contain rare enclaves of D1 tectonites. Despite displaying
cross-cutting relationships, the Silurian plutonic rocks are also
deformed along the shear zones at amphibolite facies within
error of the age of intrusion and as such are interpreted as syn-
D2 plutons (Zagorevski et al., 2007b). Hence, D2 is constrained
by these plutons to have started by at least 435 Ma and contin-
ued to at least 426 Ma (Zagorevski et al., 2007b). The age of the
D2 deformation is also constrained by the unconformable Silu-
rian cover of the Notre Dame Subzone. The age of the red beds
(ca. 429 Ma: Dunning et al., 1990), their intimate association
with D2 reverse faults, and the presence of clasts containing pre-
depositional S1 foliation indicate that they were deposited during
the D2 deformational episode, following the exhumation of the
D1 tectonites.
U/Pb GEOCHRONOLOGY
U-Pb thermal ionization mass spectrometry (TIMS) and
sensitive high-resolution ion microprobe (SHRIMP II) analyses
were conducted at the Geological Survey of Canada (GSC) in
order to constrain the timing of the arc volcanism and black shale
deposition in the Exploits Subzone. U-Pb TIMS analytical meth-
ods are outlined in Parrish et al. (1987), with treatment of analyti-
cal errors after Roddick et al. (1987). SHRIMP II analyses were
conducted using analytical procedures described by Stern (1997),
with standards and U-Pb calibration methods following Stern and
Amelin (2003). U-Pb TIMS and SHRIMP analyses are presented
in Tables 1 and 2, respectively, and are plotted in concordia dia-
grams with errors at the 2σ level (Fig. 9). Further details on the
U-Pb analytical techniques are presented in the Appendix. Geo-
chronology sample locations are indicated in Figure 10.
Penny Brook Formation
A sample (RAX05–900, Z8777; UTM zone 21, 594708 E,
5474803 N; NAD 83: Fig. 10) from a thin seam of felsic tuff,

Penny Brook 0.082 505

520 Stanley Waters formation
Formation Fine grained felsic tuff
0.082 Felsic Tuff 495 835
500 Ma
0.078 480 485 Z3
0.078
Pb/238U

460 475
0.074
440 465
206

0.070 Z2 Z1
Concordia age = 0.074 Z1
420 455
457.5 ± 2.7 Ma LI = 457 ± 3.6 Ma
0.066 SHRIMP II data (MSWD=0.1)
400
TIMS data 100 µ m
0.062
0.4 0.5 0.6 0.7 0.8 0.54 0.57 0.60 0.63 Figure 9. U/Pb SHRIMP II and TIMS
Sutherlands 0.082 505 concordia diagrams (2σ, decay con-
520 Sutherlands Pond
Pond group group stants included). MSWD—mean square
0.082 Rhyolite 500 Felsic volcanic of weighted deviates.
Z4A
485
Pb/238U

0.078 480 0.078

475
460
0.074
206

465
440
0.070 0.074
Concordia age = 455 Z5A
420
456.8 ± 3.1 Ma 457 ± 2 Ma
0.066 SHRIMP II data TIMS data
400
0.062 0.070
0.4 0.5 0.6 0.7 0.8 0.54 0.57 0.60 0.63 0.66 0.69
207 235 207 235
Pb/ U Pb/ U
320
Badger Bay
Victoria Lake
5477000 Supergroup
Badger
Group
Shoal
l’s
au p
Arm
el P rou
Fm. No ok g
Bro
5475000
Wild Bight 457±3 Ma
Group 5355000
593000 596000 480000
Figure 10. Simplified geology of the geochronology sample locations (after McConnell et al., 2002; Rogers et al., 2005; van Staal et al., 2005c).
For explanation of the fills, see Figure 5. From Dunning et al. (1987).
contained within a section of sandstone, was collected from the
Penny Brook Formation at the top of the Wild Bight Group ~30 m
below the stratigraphic contact with the Shoal Arm Formation
(B. O’Brien, 2005, personal commun.). The sample contained
a small number of euhedral prismatic zircons, many of which
contain abundant inclusions. Backscatter SEM images reveal
faint oscillatory zoning in many of the grains. As the sample did
not yield much zircon, and the presence of xenocrystic zircon in
this thin tuff unit was a possibility, the sample was analyzed by
SHRIMP. A concordia age, utilizing all of the SHRIMP analy-
ses, is calculated to be 457.5 ± 2.7 Ma (mean square of weighted
deviates [MSWD] of concordance and equivalence = 0.82; n =
18) (Fig. 9; Table 1). This date is interpreted to be the crystalliza-
tion age of the felsic tuff.
Noel Paul’s Brook Group
Crystal tuff from the Stanley Waters formation of the
Noel Paul’s Brook group (van Staal et al., 2005c) was sampled
along the eastern shore of the Stanley Waters arm of Victoria
Lake (VL01–9104; Z7334; UTM zone 21, 418768 E, 5356382
N; Fig. 10). The Stanley Waters formation forms part of a nar-
row but regionally extensive belt of coarse volcanogenic gray-
wacke, fine-grained crystal tuff, and minor intercalations of
green volcanogenic siltstone and shale (van Staal et al., 2005c).
This sequence is directly adjacent to the ca. 565 Ma Valentine
Lake monzonite (Evans and Kean, 2002). Although the contact
is not exposed, it is presumed to be a reverse fault. The volca-
nogenic siltstone grades into Caradoc black shale, which caps
the sequence. Therefore, the tuff is interpreted to represent the
upper part of a volcanogenic sequence at the base of the Caradoc
black shale, which forms a regional cover for the Victoria Lake
Supergroup. This sample yielded abundant stubby to elongated
prismatic zircon grains with melt inclusions through their cores
(Fig. 9). The three zircon fractions analyzed define a mixing line
(MSWD = 0.114) with an upper intercept of 835 ± 170 Ma and
a lower intercept of 456 ± 12 Ma (Table 1; Fig. 9). This line is
Zagorevski et al.
Sutherlands Pond gr.
Mafic volcanic Red Indian
5358000 Felsic volcanic Lake
Stanley
Waters Red Indian
Wigwam
Lake Group Brook Group
457±4 Ma ck te 462+4/-2 Ma1
ba i
p le Su 5399000
ip ive
Cr rus 457±1 Ma
t 457±3 Ma
In
Victoria Lake
Supergroup
5396000
Victoria Lake
483000 518000 524000
anchored by fraction Z1 with a concordant age of 457 ± 3.6 Ma.
This concordant age provides the best estimate for the extrusion
age of the tuff.
Sutherlands Pond Group
The first Sutherlands Pond group sample (RAX01–904,
z7157; UTM zone 21, 517905 E, 5398510 N: Fig. 10) was col-
lected from a quartz-feldspar phyric rhyolite unit that contains
small black quartz phenocrysts. This rhyolite occurs as small
beds within a rhyolitic breccia sequence that typifies the Tims
Creek formation (Rogers et al., 2005). The sample contains abun-
dant euhedral, prismatic zircons. Three-multigrain zircon TIMS
analyses from this sample are highly discordant (15%–57%)
owing to the presence of inherited zircon (Table 2, not plotted).
Backscatter and cathodoluminescence scanning electron micro-
scope (SEM) images of the zircons reveal the presence of inher-
ited cores rimmed by thick, oscillatory-zoned magmatic rims. A
concordia age, calculated from the SHRIMP analyses of mag-
matic grains and rims (n = 21), is 456.8 ± 3.1 Ma (MSWD of
concordance and equivalence = 1.6) (Table 1; Fig. 9). Inherited
cores and entirely xenocrystic grains analyzed from this rhyo-
lite range in age from ca. 970 to 1300 Ma (analyses in Table 2
marked by an asterisk, not plotted). The date of 456.8 ± 3.1 Ma is
interpreted as the crystallization age of the rhyolite.
A second sample from the Sutherlands Pond group
(MRB01–06, z7098; UTM zone 21, 521960 E, 5398445 N:
Fig. 10) was collected from an aphanitic, altered, brecciated
felsic volcanic rock. A very small number of fair quality, small
euhedral zircons with abundant inclusions and fractures were
retrieved from this sample. There was only enough material to
analyze two multigrain fractions. One TIMS analysis is 34% dis-
cordant and contains a large inherited component (Z4A; Table 2;
Fig. 9). The other analysis (Z5A) is concordant, with an age of
456.9 ± 1.3 Ma. Although there is only one concordant analysis,
the age of 457 ± 2 Ma is consistent with the crystallization age
of the RAX01–904 sample from the Sutherlands Pond group.
 TABLE 1. U/Pb SHRIMP ANALYTICAL DATA
Ages (Ma)
204 204 208 207 206 207 206 207
U Th Th Pb* Pb Pb ± 204Pb Pb ± 208Pb Pb ±207Pb Pb ± 206Pb Corr Pb ± 207Pb Pb ± 206Pb Pb ± 207Pb
206 206 206 206 235 235 238 238 206 206 238 238 206 206
Spot name (ppm) (ppm) U (ppm) (ppb) Pb Pb f(206)204 Pb Pb U U U U Coeff Pb Pb U U Pb Pb
RAX05-900 (Z8777): Felsic tuff, Penny Brook Formation
8777-7.1 116 33 0.297 8 4 0.000539 0.000380 0.0093 0.0830 0.0166 0.5217 0.0643 0.0730 0.0011 0.244 0.0518 0.0062 454 7 278 278
8777-10.1 316 86 0.282 23 1 0.000058 0.000128 0.0010 0.0910 0.0053 0.5737 0.0283 0.0738 0.0009 0.362 0.0564 0.0026 459 5 467 106
8777-11.1 287 101 0.363 21 0 0.000010 0.000010 0.0002 0.1197 0.0028 0.5618 0.0111 0.0730 0.0009 0.692 0.0558 0.0008 454 5 444 32
8777-13.1 462 223 0.498 35 7 0.000229 0.000063 0.0040 0.1553 0.0048 0.5493 0.0141 0.0732 0.0009 0.561 0.0544 0.0012 456 5 388 49
8777-14.1 290 144 0.512 22 3 0.000162 0.000142 0.0028 0.1613 0.0063 0.5744 0.0264 0.0732 0.0010 0.422 0.0569 0.0024 455 6 489 95
8777-15.1 214 94 0.454 16 5 0.000353 0.000137 0.0061 0.1280 0.0064 0.5520 0.0268 0.0723 0.0010 0.408 0.0554 0.0025 450 6 426 103
8777-16.1 277 131 0.488 21 3 0.000179 0.000163 0.0031 0.1546 0.0097 0.5521 0.0304 0.0728 0.0011 0.385 0.0550 0.0028 453 6 412 119
8777-17.1 191 72 0.390 14 3 0.000213 0.000181 0.0037 0.1235 0.0094 0.5641 0.0322 0.0734 0.0010 0.346 0.0557 0.0030 457 6 442 125
8777-18.1 359 115 0.331 27 3 0.000148 0.000092 0.0026 0.1035 0.0046 0.5682 0.0202 0.0742 0.0009 0.465 0.0556 0.0018 461 6 435 72
8777-19.1 369 214 0.599 30 3 0.000134 0.000073 0.0023 0.1807 0.0080 0.5711 0.0227 0.0755 0.0012 0.511 0.0548 0.0019 469 7 406 79
8777-20.1 441 237 0.555 34 6 0.000223 0.000076 0.0039 0.1702 0.0058 0.5481 0.0162 0.0735 0.0009 0.523 0.0541 0.0014 457 5 375 58
8777-27.1 282 114 0.418 21 4 0.000226 0.000193 0.0039 0.1370 0.0098 0.5627 0.0327 0.0721 0.0009 0.336 0.0566 0.0031 449 5 476 127
8777-24.1 266 102 0.398 19 3 0.000164 0.000138 0.0028 0.1173 0.0072 0.5629 0.0260 0.0725 0.0010 0.409 0.0563 0.0024 451 6 464 97
8777-29.1 372 169 0.470 29 1 0.000041 0.000069 0.0007 0.1518 0.0040 0.5749 0.0166 0.0748 0.0009 0.535 0.0557 0.0014 465 6 441 56
8777-28.1 293 170 0.598 23 1 0.000055 0.000099 0.0010 0.1866 0.0052 0.5672 0.0200 0.0738 0.0009 0.460 0.0557 0.0018 459 5 441 72
8777-45.1 166 70 0.435 13 4 0.000368 0.000217 0.0064 0.1333 0.0094 0.5549 0.0380 0.0741 0.0010 0.323 0.0543 0.0036 461 6 385 154
8777-44.1 200 73 0.378 15 3 0.000272 0.000138 0.0047 0.1129 0.0063 0.5684 0.0263 0.0745 0.0010 0.408 0.0554 0.0024 463 6 427 98
8777-41.1 108 30 0.289 7 7 0.001031 0.000365 0.0179 0.0703 0.0143 0.4621 0.0601 0.0715 0.0013 0.258 0.0469 0.0059 445 8 43 278
8777-39.1 771 453 0.606 63 2 0.000037 0.000054 0.0007 0.1904 0.0031 0.5951 0.0153 0.0759 0.0010 0.597 0.0569 0.0012 472 6 486 47
RAX01-904 (Z7157): Rhyolite, Sutherlands Pond group
7157-46.1 47 35 0.773 4 4 0.001226 0.000525 0.0213 0.2544 0.0211 0.5945 0.0848 0.0711 0.0015 0.267 0.0607 0.0084 443 9 627 330
7157-87.1 432 277 0.663 34 3 0.000114 0.000036 0.0020 0.2109 0.0033 0.5508 0.0109 0.0712 0.0009 0.725 0.0561 0.0008 443 5 456 31
7157-43.1 559 369 0.682 44 2 0.000059 0.000049 0.0010 0.2135 0.0030 0.5584 0.0140 0.0714 0.0009 0.619 0.0568 0.0011 444 6 482 45
7157-39.1 495 228 0.475 37 1 0.000049 0.000033 0.0009 0.1524 0.0020 0.5660 0.0099 0.0718 0.0008 0.754 0.0572 0.0007 447 5 498 26
7157-84.1 152 104 0.710 12 4 0.000420 0.000106 0.0073 0.2295 0.0053 0.5736 0.0204 0.0719 0.0009 0.478 0.0579 0.0018 448 6 524 71
7157-55.1 385 251 0.673 30 2 0.000094 0.000046 0.0016 0.2140 0.0028 0.5592 0.0113 0.0722 0.0008 0.663 0.0561 0.0009 450 5 458 34
7157-78.1 123 66 0.559 9 5 0.000630 0.000139 0.0109 0.1678 0.0062 0.5335 0.0258 0.0723 0.0011 0.435 0.0535 0.0024 450 7 351 102
7157-86.1 126 35 0.285 9 2 0.000296 0.000151 0.0051 0.0992 0.0062 0.5832 0.0283 0.0723 0.0012 0.458 0.0585 0.0025 450 7 548 98
7157-99.1 364 232 0.658 29 5 0.000227 0.000067 0.0039 0.2040 0.0035 0.5530 0.0152 0.0727 0.0010 0.609 0.0552 0.0012 452 6 420 50
7157-83.1 274 177 0.665 22 2 0.000117 0.000054 0.0020 0.2081 0.0033 0.5723 0.0129 0.0728 0.0009 0.627 0.0571 0.0010 453 5 494 39
7157-92.1 269 171 0.656 21 3 0.000163 0.000059 0.0028 0.2079 0.0049 0.5714 0.0165 0.0729 0.0010 0.586 0.0569 0.0013 454 6 486 53
(continued)
 TABLE 1. U/Pb SHRIMP ANALYTICAL DATA (CONTINUED)
Ages (Ma)
204 204 208 207 206 207 206 207
U Th Th Pb* Pb Pb ± 204Pb Pb ± 208Pb Pb ±207Pb Pb ± 206Pb Corr Pb ± 207Pb Pb ± 206Pb Pb ± 207Pb
206 206 204 206 206 235 235 238 238 206 206 238 238 206 206
Spot name (ppm) (ppm) U (ppm) (ppb) Pb Pb f(206) Pb Pb U U U U Coeff Pb Pb U U Pb Pb
RAX01-904 (Z7157): Rhyolite, Sutherlands Pond group (continued)
7157-77.1 281 152 0.557 22 5 0.000274 0.000081 0.0048 0.1739 0.0037 0.5568 0.0215 0.0729 0.0008 0.408 0.0554 0.0020 454 5 427 81
7157-69.1 208 68 0.339 15 5 0.000386 0.000102 0.0067 0.1034 0.0043 0.5683 0.0195 0.0737 0.0009 0.461 0.0559 0.0017 458 5 450 70
7157-20.1 312 166 0.550 25 1 0.000044 0.000093 0.0008 0.1796 0.0055 0.5716 0.0176 0.0739 0.0008 0.466 0.0561 0.0015 460 5 457 62
7157-45.3 425 135 0.329 31 5 0.000199 0.000048 0.0035 0.1039 0.0024 0.5726 0.0131 0.0740 0.0009 0.631 0.0561 0.0010 460 5 457 40
7157-91.1 69 3 0.040 5 3 0.000650 0.000306 0.0113 0.0112 0.0115 0.5582 0.0520 0.0740 0.0012 0.291 0.0547 0.0049 460 7 400 215
7157-37.1 287 118 0.426 22 4 0.000228 0.000086 0.0040 0.1332 0.0037 0.5638 0.0171 0.0741 0.0009 0.509 0.0552 0.0015 461 5 420 60
7157-101.1 229 95 0.431 17 4 0.000248 0.000083 0.0043 0.1351 0.0037 0.5666 0.0177 0.0743 0.0009 0.480 0.0553 0.0015 462 5 423 63
7157-14.1 416 227 0.563 33 2 0.000081 0.000047 0.0014 0.1776 0.0025 0.5741 0.0111 0.0745 0.0008 0.648 0.0559 0.0008 463 5 449 33
7157-4.1 498 346 0.717 41 0 0.000010 0.000010 0.0002 0.2272 0.0019 0.5791 0.0079 0.0752 0.0008 0.844 0.0558 0.0004 468 5 446 17
7157-6.1 112 63 0.586 9 2 0.000230 0.000138 0.0040 0.1869 0.0063 0.5954 0.0274 0.0761 0.0010 0.391 0.0567 0.0024 473 6 481 97
7157-7.1 142 100 0.723 12 1 0.000110 0.000100 0.0019 0.2293 0.0050 0.6135 0.0208 0.0767 0.0009 0.464 0.0580 0.0018 476 5 531 68
7157-5.1* 73 51 0.718 13 1 0.000090 0.000095 0.0016 0.2278 0.0047 1.5942 0.0493 0.1614 0.0021 0.527 0.0717 0.0019 964 12 976 55
7157-100.1* 69 92 1.375 15 7 0.000717 0.000162 0.0124 0.4230 0.0079 1.5824 0.0687 0.1624 0.0027 0.498 0.0707 0.0027 970 15 947 80
7157-79.1* 67 49 0.757 13 1 0.000123 0.000095 0.0021 0.2431 0.0052 1.7060 0.0463 0.1680 0.0021 0.555 0.0737 0.0017 1001 11 1032 47
7157-88.1* 50 27 0.544 9 5 0.000678 0.000211 0.0118 0.1539 0.0090 1.6470 0.0895 0.1686 0.0030 0.442 0.0708 0.0035 1004 17 953 104
7157-18.1* 124 44 0.366 22 3 0.000162 0.000049 0.0028 0.1089 0.0032 1.7502 0.0324 0.1729 0.0020 0.701 0.0734 0.0010 1028 11 1026 27
7157-36.1* 95 28 0.302 17 5 0.000380 0.000086 0.0066 0.0891 0.0039 1.7502 0.0451 0.1748 0.0023 0.601 0.0726 0.0015 1039 12 1003 43
7157-101.2* 117 38 0.339 21 2 0.000128 0.000098 0.0022 0.1050 0.0040 1.8462 0.0515 0.1791 0.0026 0.618 0.0748 0.0017 1062 14 1062 45
7157-37.2* 75 13 0.182 13 6 0.000474 0.000118 0.0082 0.0551 0.0047 1.8560 0.0606 0.1850 0.0027 0.554 0.0728 0.0020 1094 15 1007 57
7157-87.2* 340 165 0.501 70 1 0.000021 0.000030 0.0004 0.1527 0.0016 2.0648 0.0307 0.1932 0.0023 0.849 0.0775 0.0006 1139 12 1134 16
7157-80.1* 129 66 0.526 28 3 0.000121 0.000066 0.0021 0.1605 0.0032 2.2415 0.0476 0.2008 0.0027 0.722 0.0810 0.0012 1180 15 1220 29
7157-45.1* 81 25 0.320 19 2 0.000148 0.000069 0.0026 0.0975 0.0044 2.8578 0.0708 0.2226 0.0038 0.762 0.0931 0.0015 1296 20 1490 31
Note: See Stern (1997). Uncertainties reported at 1σ (absolute) and are calculated by numerical propagation of all known sources of error; f(206)204 refers to mole fraction of total 206Pb that is due to common Pb, calculated
using the 204Pb method; common Pb composition used is the surface blank.
 TABLE 2. U-Pb TIMS ANALYTICAL DATA
Isotopic ratios†† Ages (Ma)##
Fract.* Description† 206 208 207 206 207 206 207 207
Wt. U Pb§ Pb# Pb** Pb Pb ± 1SE Pb ± 1SE Corr.§§ Pb ± 1SE Pb Pb Pb Disc
204 206 235 238 206 238 235 206
(ug) (ppm) (ppm) Pb (pg) Pb U Abs U Abs Coeff. Pb Abs U ± 2SE U ± 2SE Pb ± 2SE (%)
MRB01-06 (Z7098): Brecciated felsic volcanic rock, Sutherlands Pond group
Z4A (42) Co,Clr,Eu,St,aIn,fFr,M1° 15 251 22 1983 10 0.19 0.70183 0.00134 0.07973 0.00008 0.769 0.06385 0.00008 494.5 1.0 539.9 1.6 736.5 5.5 34.1
Z5A (48) Co,Clr,Eu,St,aIn,fFr,M3° 18 239 19 2503 8 0.21 0.56821 0.00102 0.07344 0.00011 0.696 0.05611 0.00007 456.9 1.3 456.9 1.3 456.8 5.8 0.0

RAX01-904 (Z7157): Rhyolite, Sutherlands Pond group

Z1 (10) Co,Clr,Eu,Pr,rIn,Dia 26 128 15 7290 3 0.18 1.10375 0.00233 0.11263 0.00025 0.859 0.07107 0.00008 688.0 2.9 755.2 2.3 959.5 4.7 29.8
Z2 (7) Co,Clr,Eu,Pr,rIn,Dia 30 129 12 2313 9 0.14 0.74132 0.00191 0.08806 0.00019 0.793 0.06106 0.00010 544.1 2.2 563.2 2.2 641.2 6.7 15.8
Z3A (4) Co,Clr,Eu,St,rIn,Dia 29 206 33 9106 6 0.22 2.14224 0.00323 0.13993 0.00019 0.953 0.11103 0.00005 844.3 2.1 1162.5 2.1 1816.4 1.7 57.0

VL01-9104 (Z7334): Felsic Tuff, Noel Paul’s Brook group

Z1 (18) Co,Clr,Pr,Eu,aIn,NM5° 6 219 17 1262 5 0.12 0.58395 0.00231 0.07490 0.00011 0.500 0.05654 0.00020 465.6 1.3 467.0 3.0 473.8 15.3 1.8
Z2 (19) Co,Clr,Pr,Eu,aIn,NM5° 10 579 45 2367 11 0.15 0.58595 0.01076 0.07446 0.00124 0.572 0.05708 0.00093 463.0 14.9 468.3 13.8 494.4 70.1 6.6
Z3 (20) Co,Clr,Pr,Eu,aIn,NM5° 20 327 26 3228 10 0.13 0.63209 0.00456 0.07926 0.00056 0.967 0.05784 0.00011 491.7 6.6 497.4 5.7 523.7 8.1 6.3
*All fractions are zircon and have been abraded following the method of Krogh (1982). Number in parentheses refers to the number of grains in the analysis.
†
Zircon descriptions: Co—colorless; Clr—clear; fFr—few fractures; aIn—abundant inclusions; rIn—rare inclusions; Eu—euhedral; Pr—prismatic; St—stubby prism; Dia—diamagnetic; M1°—magnetic @ 1.8A, 1°SS; M3°—
magnetic @ 1.8A, 3°SS; NM5°—nonmagnetic @ 1.8A, 5°SS.
§
Radiogenic Pb.
#
Measured ratio, corrected for spike and fractionation.
**Total common Pb in analysis corrected for fractionation and spike.
††
Corrected for blank Pb and U and common Pb; errors quoted are 1σ absolute; procedural blank values for this study ranged from 0.1 to 0.3 pg for U and 2 to 10 pg for Pb; Pb blank isotopic composition is based on the
analysis of procedural blanks; corrections for common Pb were made using Stacey and Kramers (1975) compositions.
§§
Correlation coefficient.
##
Corrected for blank and common Pb; errors quoted are 2σ in Ma.
324 Zagorevski et al.
Hence we interpret the 457 ± 2 Ma as the crystallization age of
the Sutherlands Pond group felsic volcanic rock.
DISCUSSION
Identification of four new ages of Caradoc peri-Gondwanan
volcanism in the Victoria Arc, in addition to a previously obtained
age (453 ± 3 Ma: Zagorevski et al., 2007a), demonstrates that Cara-
doc volcanism was much more widespread than previously postu-
lated (e.g., O’Brien et al., 1997; MacLachlan et al., 2001), allowing
more precise timing constraints to be placed on the deposition of the
cover and the cessation of magmatism in the Victoria Arc. The age
obtained for the Penny Brook Formation extends the age range of
volcanism in Notre Dame Bay until 457.5 ± 2.7 Ma and constrains
the maximum age of sedimentation in the overlying manganiferrous
chert, shale, and black shale of the Caradoc Shoal Arm Formation,
which forms part of the Victoria Arc cover sequence. The volcanic
rocks from the Sutherlands Pond and Noel Paul’s Brook groups are
the same age (457 ± 2; 456.8 ± 3.1; 457 ± 3.6 Ma). Because they
predate deposition of most of the black shale, these ages constrain
the maximum age of deposition of the cover of the Victoria Arc
independently from sparse fossil evidence.
Contrasting Ordovician Arc Systems: Red Indian Lake
and Victoria Arcs
The ages obtained in this study and by Zagorevski (2006)
indicate that the Victoria Arc was magmatically active from
ca. 473 Ma until 453 Ma. The Victoria Arc magmatism was
coeval with the Annieopsquotch accretionary tract, where mag-
matism initiated at ca. 480 Ma (Annieopsquotch ophiolite belt:
Dunning and Krogh, 1985; Lissenberg et al., 2005b) and con-
tinued until ca. 460 Ma (Red Indian Lake Arc: Zagorevski et al.,
2006). The subsequent juxtaposition of these coeval arc systems
along the RIL can make their separation very difficult in the
absence of a multidisciplinary data set.
Williams et al. (1988) noted the broad contrasts in stratig-
raphy, structure, fauna, plutonic rocks, radiogenic lead in min-
eral deposits, magnetic anomalies, and gravity anomalies as the
basis for separation of the Dunnage Zone into the Notre Dame
and Exploits Subzones. Detailed mapping in central Newfound-
land has revealed additional criteria that can be utilized for the
separation of the arc systems, namely the zircon provenance and
the Sm/Nd isotopic characteristics of volcanic rocks. In the fol-
lowing sections we discuss the differences in the basement char-
acteristics, tectonic history, and cover sequences between the
Red Indian Lake and Victoria Arcs. These will then be utilized
to reconstruct the Ordovician arc-arc collision in Newfoundland
based, in part, on modern analogues in the southwest Pacific.
Basement Characteristics
The Red Indian Lake and Victoria Arcs can be effectively dif-
ferentiated on the basis of the zircon inheritance and the Sm/Nd
isotopic characteristics of felsic volcanic rocks. The lowest εNd val-
ues recorded in the volcanic rocks of the Exploits Subzone occur
in the youngest rocks (Wigwam Brook Group: εNd −4), whereas
most of the rocks have generally maintained positive to slightly
negative εNd values (εNd +8 to −1) with TDM ages generally in the
1.30–0.6 Ga range (Fig. 11; Zagorevski et al., 2006). In contrast to
the Exploits Subzone, volcanic rocks of the Notre Dame Subzone
display consistently lower εNd values, with εNd in felsic volcanic
rocks commonly in the −3 to −10 range (Whalen et al., 1997; Lis-
senberg et al., 2005a; Zagorevski et al., 2006), indicating a greater
contribution of continental crust and/or more mature continental
crust (Fig. 11).
Zircon inheritance supports the Sm/Nd isotope data and
indicates that the volcanic rocks of the Exploits Subzone have
been deposited on continental basement that underwent tectono-
magmatic events at ca. 560 Ma and ca. 900–1200 Ma (Zagor-
evski et al., 2007a). This is supported by the zircon inheritance in
the Sutherlands Pond group rhyolite (ca. 970–1200 Ma: Table 2).
This basement is similar to the Proterozoic Crippleback Igneous
Suite and Sandy Lake Group (ca. 560 Ma, TDM ~1300: Kerr et al.,
1995; Rogers et al., 2006), which stratigraphically underlie the
oldest parts of the Penobscot arc in Newfoundland (Rogers et al.,
2006). Similar relationships occur in the broadly correlative New
River Belt in New Brunswick (Johnson and McLeod, 1996). This
basement likely represents a fragment of Ganderia (van Staal et
al., 2004), a Gondwana-derived microcontinent, which is charac-
terized by zircon provenance in the 0.54–0.55, 0.6–0.8, 1.0–1.55,
and 2.5–2.7 Ga age ranges (van Staal et al., 1996), εNd values
greater than −4, and TDM ages of late Precambrian igneous rocks
spanning 0.9 to 1.35 Ga (Kerr et al., 1995; Rogers et al., 2006).
Notre Dame/ Exploits Subzone Gander
Dashwoods sz. Zone
8
3
Age (Ma)
εNd
480 460 460 500 540
-2
-7 Legend
Mafic
Felsic
Gneiss
-12
Figure 11. Sm/Nd isotopic evolution of the Notre Dame and Exploits
Subzones. Compiled from Jenner and Swinden (1993), Kerr et al.
(1995), Swinden et al. (1997), Whalen et al. (1997), MacLachlan and
Dunning (1998a, b), Rogers (2004), Lissenberg et al. (2005b), Rogers
et al. (2006), and Zagorevski et al. (2006, 2007a).
 Tectonic architecture of an arc-arc collision zone
The ensialic plutonic and volcanic rocks of the Annieops-
quotch accretionary tract and the Red Indian Lake Arc have abun-
dant inheritance in the 935–1845 Ma age range (Dec et al., 1997;
Zagorevski et al., 2006), consistent with the presence of Laurentian
basement at depth (e.g., Cawood et al., 2001; Cawood and Nem-
chin, 2001; Cawood et al., 1995). The presence of inheritance in
the 1.0–1.55 Ga age ranges is common to both subzones and thus
is not a reliable differentiation tool. However, zircon inheritance
in the 1.7–1.9 Ga age range is generally associated with the Lau-
rentian margin (Cawood and Nemchin, 2001). In addition, zircon
inheritance in the 520–565 Ma age range appears to be restricted
to the peri-Gondwanan terranes in the Dunnage Zone, owing
to the extensive magmatism of this age in the Ganderia-derived
basement (Dunning and O’Brien, 1989; Evans et al., 1990; Rog-
ers et al., 2006). Although ca. 550 Ma magmatism (e.g., Cawood
et al., 2001, and references therein) related to the opening of Iape-
tus is present in Laurentia (Cawood et al., 2001), it is generally
restricted to the Humber Zone.
Sm/Nd isotopic signatures and the presence of locally abun-
dant zircon inheritance in felsic volcanic rocks indicate the con-
tribution of continental basement to both the Red Indian Lake and
Victoria Arcs. The contrasts in the Sm/Nd isotopic characteristics
and the presence of specific age ranges of zircon indicate that the
nature of the arc basement is fundamentally different and reflects
the peri-Laurentian and peri-Gondwanan derivation of the Red
Indian Lake and Victoria Arcs, respectively (Fig. 11).
Tectonic History
The Annieopsquotch accretionary tract and the Victoria Arc
underwent distinctly different tectonic histories in the Early to
Middle Ordovician. The Annieopsquotch accretionary tract
underwent deformation throughout much of its history, starting
at ca. 470 Ma and lasting until the collision with the Victoria Arc
(Fig. 5; Lissenberg et al., 2005a). Some of this deformation may
be related to the simultaneous collision between the Dashwoods
microcontinent and the Laurentian margin (Waldron and van
Staal, 2001). In contrast, the initiation of magmatism in the Vic-
toria Arc at ca. 473 Ma followed a short orogenic episode (Fig. 5;
Penobscot orogeny: Colman-Sadd et al., 1992) in the Exploits
Subzone and Gander Zone. From the onset of magmatism, the
continental Victoria Arc was extensional (e.g., MacLachlan and
Dunning, 1998b; O’Brien et al., 1997; Rogers et al., 2003), open-
ing a wide Tetagouche-Exploits oceanic backarc basin in New-
foundland and New Brunswick (van Staal et al., 1998).
The earliest deformation that is common to both the Red
Indian Lake and Victoria Arcs started during Caradoc time. This
is constrained by the involvement of the Caradoc black shale and
volcanic rocks in the mélange zones that mark the D1 thrusts,
including the Red Indian Line (e.g., McConnell et al., 2002;
Rogers and van Staal, 2002). The D1 mélange zones most likely
represent the syncollisional terrane boundaries formed during
the accretion of the Victoria Arc to the Laurentian margin. Shear
sense indicators along the D1 shear zones suggest sinistral oblique
325
SSE-directed emplacement of the Red Indian Lake Arc over the
Victoria Arc. This observation is confirmed by seismic-reflection
studies of Thurlow et al. (1992) and van der Velden et al. (2004),
which clearly show the Red Indian Line to emplace the Annie-
opsquotch accretionary tract and the Red Indian Lake Arc over
the Exploits Subzone and the Victoria Arc (Fig. 8). Interpreta-
tion of the migrated seismic reflection data by van der Velden et
al. (2004) suggests that the Victoria Arc and its basement were
subducted to a depth of at least 18 km and underthrust the Annie-
opsquotch accretionary tract by at least 60 km (Fig. 8).
Cover Sequences of the Notre Dame and Exploits Subzones
The contrast between the cover sequences of the Notre Dame
and Exploits Subzones is highly distinctive in the Newfoundland
Appalachians and can serve as a reliable tool to differentiate the
two arc systems. Adhering to the original observations of Williams
et al. (1988) the Notre Dame Subzone, and therefore the Annie-
opsquotch accretionary tract, contain a sub-Silurian unconformity
that is overlain by Silurian continental red sandstones and volcanic
rocks. In contrast, the rocks of the Exploits Subzone, and thus the
Victoria Arc, are overlain by Caradoc black shale and display a gen-
erally continuous sedimentation through the Ordovician-Silurian
boundary (i.e., Badger Group: Williams et al., 1995). Although
several models for the depositional setting of the Badger Group
have been proposed (Wasowski et al., 1986), the potential tectonic
causes for the contrasts between the subzones have not yet been
addressed. We now briefly discuss several key observations that
allow a comparison with modern arc-arc collisions.
Parts of the Victoria Arc were covered by late Llanvirn lime-
stone (Figs. 5, 7; e.g., Williams, 1995b; O’Brien et al., 1997;
McConnell et al., 2002), suggesting relatively shallow water
prior to the collision. The deposition of limestone was locally
accompanied by formation of mélange, suggesting initiation of
deformation in the Caradoc (e.g., Sops Head Complex: McCon-
nell et al., 2002). This was followed by widespread deposition
of Caradoc black shale (Fig. 5; e.g., Williams, 1995b). In central
Newfoundland, Caradoc black shale of the Laurence Harbour
Formation unconformably overlies the volcanic rocks of the Tally
Pond Group (ca. 513 Ma: Rogers et al., 2005, 2006), suggesting
that the Penobscot basement of the Victoria Arc was locally emer-
gent prior to the Caradoc. The deposition of black shale above the
Victoria Arc and its basement, followed by deposition of deep-
marine turbidites of the Badger Group (Williams et al., 1993),
support rapid submergence of the Victoria Arc (e.g., Williams,
1995b). The zircon provenance of the turbidites in the lower Bad-
ger Group suggests derivation from the Laurentian margin or the
Notre Dame Subzone (McNicoll et al., 2001), indicating that the
collision of the Victoria Arc and the Annieopsquotch accretion-
ary tract had already started by Early Ashgill time.
In modern arc-arc collisions, the downgoing plate (i.e., Snel-
lius Plateau and Morobe Shelf) undergoes rapid subsidence,
as indicated by drowned carbonate platforms (e.g., Abers and
McCaffrey, 1994; Galewsky and Silver, 1997; Pubellier et al.,
326 Zagorevski et al.

1999). At the same time the overriding plate is rapidly uplifted and
forms an emergent orogen (e.g., Talaud Island, Huon-Finisterre
Ranges: Fig. 1). The detritus derived from the emergent orogen is
transported parallel to the trench and deposited as thick turbidite
sequences such as the Markam Canyon turbidites (Galewsky and
Silver, 1997; Whitmore et al., 1999). This detritus would have its
source in the overriding plate.
The concurrence of the subsidence of the Victoria Arc and
the influx of the peri-Laurentian detritus suggest that this was
caused by the collision with the Red Indian Lake Arc (Figs. 12,
13). Consistent with this interpretation, structural and seismic-
reflection studies indicate that the Victoria Arc occupies a lower
plate setting with respect to the Red Indian Lake Arc (Fig. 8; e.g.,
van der Velden et al., 2004). Hence it was partially subducted
under the Red Indian Lake Arc, resulting in loading of the Vic-
toria Arc crust, rapid subsidence, and widespread deposition of
black shale above the arc and backarc. This is consistent with the
absence of the Upper Wild Bight Group equivalents in central
Newfoundland, which were probably overridden by the Annie-
opsquotch accretionary tract during the arc-arc collision (Fig. 8)

A 465–460 Ma Red Indian Iapetus Dunnage Victoria

Exploits-
Lake group Tetagouche
(AAT) Ocean melange arc
backarc

Accreted
terranes
Iapetus
Ocean Plate

B 460–455 Ma Uplift of peri-Laurentian Loading of Exploits-

terranes RIL Victoria arc Tetagouche
backarc
SHC

forearc
Molucca Figure 12. Tectonic model for the
Iapetus
subduction? Sea Plate
Ocean Plate Caradoc arc-arc collision in central
Newfoundland (based on van Staal et
al., 1998). AAT—Annieopsquotch ac-
cretionary tract; DBL—Dog Bay Line;
RIL—Red Indian Line; SHC—Sops
C 455–450 Ma Emergent peri-Laurentian Caradoc black Head Complex; WPG—Windsor Point
terranes shale and Group.
RIL volcanic rocks Future
WPG DBL

initiation of
subduction
Slab
detachment

Iapetus
Ocean Plate

D 450–430 Ma Sub-Silurian unconformity RIL Badger Future

basin DBL
 Tectonic architecture of an arc-arc collision zone
c. 450 Ma
Laurentia
Baltica
RILA
VPA
AB
ETB a VPA—Victoria-Popelogan Arc.
deri
Gan
lonia
Ava
Gondwana
similar to the Snellius Plateau (Fig. 2). The loading and subsid-
ence of the Victoria Arc would have occurred contemporaneously
with the Caradoc arc-arc collision, as indicated by the formation
of the Caradoc black shale mélange (Fig. 12; e.g., Sops Head
Complex: McConnell et al., 2002).
Any subduction of the buoyant Victoria Arc crust and its
continental basement would have most likely resulted in the
emergence of the Annieopsquotch accretionary tract and the
Notre Dame Subzone above sea level (e.g., Huon-Finisterre
Range: see previous discussion). Although no unconformities of
the right age have been yet described within the Annieopsquotch
accretionary tract in central Newfoundland, an upper Caradoc
unconformity has been defined in southwestern Newfoundland
below the Windsor Point Group (453 +5/–4 Ma: Dube et al.,
1996), which unconformably overlies the Notre Dame Subzone
plutons (ca. 469–488 Ma Cape Ray Igneous Complex: Dube et
al., 1996). We interpret this unconformity and the lack of late
Caradoc rocks in the Annieopsquotch accretionary tract to imply
that the Annieopsquotch accretionary tract had indeed breached
sea level, leading to nondeposition and/or erosion through-
out the Caradoc (Fig. 12). Hence, the influx of turbidites with
327
Figure 13. Caradoc paleogeography of
arc-arc collision in the Appalachian-
Caledonian orogen, slightly modified
from van Staal et al. (1998). Original
figure courtesy of Conall MacNiocaill.
ETBAB—Exploits-Tetagouche backarc
basin; RILA—Red Indian Lake Arc;
peri-Laurentian provenance and their deposition on the subsided
Victoria Arc (McNicoll et al., 2001) signaled the emergence of the
Notre Dame Subzone and Laurentia during the arc-arc collision.
Syncollisional Magmatism
The presence of abundant Caradoc volcanism in the Exploits
Subzone indicates that volcanism and deformation were coeval.
Caradoc felsic volcanic rocks in the Victoria Arc achieve distinctly
low εNd values (εNd −4 in the Wigwam Brook Group: Zagorevski
et al., 2006; εNd −4 in the Sutherlands Pond group: Rogers, 2004)
that are atypical for the Exploits Subzone but common in the
Notre Dame Subzone (Fig. 11). The presence of Caradoc black
shale and moderately low εNd values distinguishes the Caradoc
volcanic rocks of the Exploits Subzone from the peri-Laurentian
felsic volcanic rocks of the Red Indian Lake Group (εNd −6.4 to
−7.7; Fig. 11; Zagorevski et al., 2006); however, the distinctly low
εNd values in the Caradoc volcanic rocks are rather atypical for the
Exploits Subzone, which generally contained positive to slightly
negative εNd values throughout its history (Fig. 11). This may have
resulted from tapping of a larger proportion or a different part of
328 Zagorevski et al.
the Ganderian basement during magma generation, leading to a
rapid change in εNd values in the Caradoc.
An alternate solution is more appealing in terms of an arc-
arc collision model, although it would be difficult to prove. The
low εNd in the Sutherlands Pond group rhyolite may be a result of
subduction of peri-Laurentian sediment below the Victoria Arc.
Comparison with the geochemical evolution of the Halmahera-
Sangihe arc collision zone (Macpherson et al., 2003) indicates
that this is a feasible, albeit not a unique, solution. In both the
Halmahera and Sangihe Arcs, the geochemistry of volcanic rocks
that erupted prior to and during the collision indicates increased
sediment flux through the subduction zone. The increased sedi-
ment flux could be a result of the collision of the facing accretion-
ary prisms, as is probably the case in the Molucca Sea, or deposi-
tion of emergent orogen-derived sediment along active trenches
(e.g., Markham Canyon turbidites in the Trobriand Trough). A
similar explanation can be utilized for the Wigwam Brook Group
felsic tuff; however, since the tuff may be epiclastic in nature,
the low εNd values could also have been achieved by mixing with
sediment of low εNd values. Hence, the most likely source would
still be the emergent peri-Laurentian terranes.
Although the discussion of the chemical characteristics of
the volcanic rocks is beyond the scope of this paper the (pre-)
Caradocian volcanic rocks have a range of compositions from
arclike to nonarc, which suggests eruption in an extensional
suprasubduction zone setting (e.g., Penny Brook Formation:
McConnell and O’Brien, 2000; Sutherlands Pond group: Evans
and Kean, 2002; Rogers, 2004; Rogers et al., 2005; Diversion
Lake Group: Swinden et al., 1989; Evans and Kean, 2002; Rog-
ers et al., 2005; Sops Head Complex: McConnell et al., 2002). An
alternate setting for the syncollisional extensionlike magmatism
may be above a delaminated Iapetus slab (Fig. 12), similar to the
Pliocene and younger volcanic rocks in Mindanao (Sajona et al.,
2000). This is an appealing model because it would also explain
the change of isotopic characteristics of the younger volcanic
rocks in the Exploits Subzone, as both peri-Gondwanan and peri-
Laurentian sources can be involved.
Along-Strike Variations
Peri-Laurentian magmatism in the Annieopsquotch accre-
tionary tract (ca. 460–480 Ma: Dunning and Krogh, 1985;
Zagorevski et al., 2006) was in part coeval with the Victoria Arc
(ca. 453–473 Ma: e.g., MacLachlan and Dunning, 1998b; Zago-
revski et al., 2007a), requiring subduction on both sides of the
Iapetus Ocean, which culminated in a Molucca Sea–type arc-arc
collision (Fig. 13; van Staal et al., 1998). The key evidence that
can be used to infer the presence of distinct and separated arc
sequences in Newfoundland include structural and tectonic his-
tory, stratigraphy, basement characteristics, radiogenic lead in
mineral deposits, fauna, zircon provenance, and Sm/Nd isotopic
characteristics (see previous discussion; Williams et al., 1988).
Extrapolation of these interpretations is difficult outside of New-
foundland, where the equivalents of the Dunnage Zone tend to
be poorly preserved owing to metamorphism and deformation as
well as the presence of extensive cover sequences.
In addition to these obstacles, the age of deformation, tec-
tonic relationships, nature of the basement, and preservation of
tectonic elements can vary significantly along the strike of the
collision zone. First, both the Molucca and Solomon Sea arc-arc
collisions are diachronous along strike of the orogen; hence the
duration of arc volcanism, the age of the cover sequences, and the
timing of deformation will also be diachronous (Fig. 1). Second,
the polarity of thrusting can change along strike and may result
in stratigraphic differences in the cover sequences. In the cen-
tral Molucca collision zone, for example, the Halmahera Arc is
subducted under the Sangihe Arc. At the same time, back thrusts
emplace the Halmahera Arc over the Sangihe Arc in the north-
ern Molucca Sea (Fig. 2). Hence the polarity of thrusting changes
from north to south. Third, long-range correlation of the Appa-
lachian terranes may require adjustments to the local basement
characteristics, as the nature of the basement is unlikely to remain
exactly the same along the whole strike of the orogen. Finally, the
absence of tectonic elements can result from either strike-slip exci-
sion (e.g., Elders, 1987) or the local peculiarity of the colliding arc
systems. Analogue experiments suggest that the behavior of the
overriding arc during the collision will depend on the strength of
the arc lithosphere and the retroarc region; i.e., the arc and forearc
may be deformed but preserved, the forearc may be subducted,
the forearc and part of the arc may be subducted, or the whole arc
may be subducted (Boutelier et al., 2003). The lack of preserva-
tion of the forearc basin to the Red Indian Lake Arc, for example,
indicates that the forearc from the central Newfoundland section
was removed by either strike-slip translation or subduction.
Northern Appalachians
Despite the potential problems with preservation and long-
range correlations, equivalents of the Annieopsquotch accretion-
ary tract and Victoria Arc are recognized elsewhere in the north-
ern Appalachians. The remnants of the Dashwoods microconti-
nent and its Notre Dame Arc suprastructure in New England are
preserved in the Shelburne Falls Arc (ca. 470–485 Ma: Karabi-
nos et al., 1998) and the Chain Lakes Massif (>473 Ma: Gerbi et
al., 2006) both of which have clear peri-Laurentian affinities. The
Annieopsquotch accretionary tract is largely obscured in Atlantic
Canada by cover sequences; however, correlatives of the Annie-
opsquotch accretionary tract include the Boil Mountain Complex
and the Jim Pond Formation (ca. 477–485 Ma: Coish and Rog-
ers, 1987; Gerbi et al., 2006), which are adjacent to the Chain
Lakes Massif (Fig. 3) as well as to the Caucomgomoc inlier (van
Staal et al., 1998).
The equivalents of the Victoria Arc (Fig. 3: Popelogan Inlier:
Wilson, 2003; Munsungun inlier: Ayuso and Schulz, 2003; Ayuso
et al., 2003) and backarc (Fig. 3: Bathurst Supergroup: van Staal
et al., 2003) are well preserved in New Brunswick and Maine;
however, the equivalents of the Victoria Arc in western Maine,
New Hampshire, and Vermont are less clear. The Middle to Late
Ordovician Bronson Hill Arc (Moench and Aleinikoff, 2003),
 Tectonic architecture of an arc-arc collision zone
which lies to the east of the peri-Laurentian Shelburne Falls Arc,
is a probable correlative of the Munsungun inlier, which is an
equivalent of the Popelogan Arc in Maine (Ayuso and Schulz,
2003; Ayuso et al., 2003). The Bronson Hill Arc contains two
distinct sequences: the Ammonoosuc (457–470 Ma) and Qui-
mby (435–456 Ma) sequences, which are thought to have formed
above subduction zones of opposite polarity (Moench and Aleini-
koff, 2003). Although the Ammonoosuc sequence has been tec-
tonically correlated with the peri-Laurentian Shelburne Falls Arc
(e.g., Moench and Aleinikoff, 2003), the contact relationships are
obscured by the Connecticut Valley Trough and younger deposits,
and the age span of the Ammonoosuc sequence overlaps with the
age of the Victoria Arc, although relationships with the Penobscot
Arc basement have not been identified. In addition, the Ammo-
noosuc sequence shares several characteristics with the peri-
Gondwanan Victoria Arc (van Staal et al., 1998; Hibbard et al.,
2004), including an important period of extension as exemplified
by the Chickwolnepy sheeted intrusions (ca. 467 Ma; Fitz, 2002;
Moench and Aleinikoff, 2003). The cover sequences are also
remarkably similar to the Victoria Arc. The type Ammonoosuc
volcanic rocks are conformably overlain by the Caradoc sulfidic
black shale of the Partridge Formation, followed by an uncon-
formity and deposition of marine graywacke of the Ashgill Qui-
mby Formation (Moench and Aleinikoff, 2003). The presence of
a Caradoc-Ashgill unconformity between the Ammonoosuc and
Quimby sequences (Moench and Aleinikoff, 2003) differentiates
the Bronson Hill Arc from the Victoria Arc; however, this can be
easily reconciled by considering the along-strike variations in an
arc-arc collision zone.
The Quimby sequence, overlying the Ammonoosuc sequence,
comprises marine shales, graywackes, and conglomerates
intruded by Upper Ordovician to Lower Silurian plutons (Moench
and Aleinikoff, 2003). The Quimby sequence corresponds to the
Bronson Hill Arc in southern New Hampshire and central Mas-
sachusetts (454–442 Ma: Tucker and Robinson, 1990; Karabinos
et al., 1998). This phase of the Bronson Hill Arc development has
been interpreted to have occurred following an arc collision of the
Shelburne Falls Arc and the Ammonoosuc sequence of the Bron-
son Hill Arc with the Laurentian margin (Karabinos et al., 1998;
Moench and Aleinikoff, 2003). The collision was followed by
establishment of a west-dipping subduction zone along the com-
posite Laurentian margin following a polarity flip (Karabinos et
al., 1998; Moench and Aleinikoff, 2003).
The tectonic evolution of New England broadly mirrors the
Ordovician arc-arc collision in Newfoundland. The accretion of
the Ammonoosuc sequence of the Bronson Hill Arc to the com-
posite Laurentian margin, comprising the peri-Laurentian Shel-
burne Falls Arc and Jim Pond Formation, occurred at ca. 457 Ma,
followed by a magmatic gap lasting several million years, deep-
marine sedimentation, and probable subduction reversal (Moench
and Aleinikoff, 2003). This closely agrees with the accretion of
the Victoria Arc to the Red Indian Lake Arc, where the Victoria
arc was loaded, leading to subsidence and deep-marine sedimen-
tation (Fig. 12B, C). Subsequent convergence was accommodated
329
by a change of polarity and a step-back of the subduction into the
Tetagouche-Exploits backarc (Fig. 12C, D) and formation of the
Quimby sequence of the Bronson Hill Arc.
British Caledonides
A direct correlation between Newfoundland and the British
Caledonides is difficult, because the well-preserved sedimen-
tary sequences of the late Llanvirn to the Late Silurian Southern
Uplands accretionary complex (e.g., Legget, 1987) are largely
absent in the northern Appalachians, a feature that may reflect
strike-slip excision or subduction. Van Staal et al. (1998) proposed
a correlation of the Annieopsquotch accretionary tract with the
South Connemara Group in Ireland (e.g., Ryan and Dewey, 2004)
and the correlative Northern Belt of the Southern Uplands terrane
in Scotland. This correlation was in part supported by Armstrong
and Owen (2001), who proposed a new terrane, Novantia, to be
a correlative of the Annieopsquotch accretionary tract. Novantia
has been subducted underneath the Southern Uplands; however,
its presence was deduced on the basis of geophysical modeling.
Armstrong and Owen (2001) also recognized correlatives of the
Victoria Arc in the Grangegeeth Terrane of eastern Ireland and
proposed that these have also been in part subducted under the
Southern Uplands. The juxtaposition of the Novantia and Grang-
egeeth terranes in the Ordovician was subsequently supported by
the presence of distinctly peri-Gondwanan (ca. 560 Ma) zircons
in the Caradoc Portpatrick Formation of the Southern Uplands
(Phillips et al., 2003).
Comparison of the Newfoundland arc-arc collision with
the Southern Uplands illustrates the importance of along-strike
variation in a collisional setting. The Southern Uplands accre-
tionary complex contains both the peri-Laurentian and the peri-
Gondwanan provenance, indicating emergence of both Lau-
rentia and the Grangegeeth Terrane during the collision. The
Badger Group in Newfoundland, on the other hand, contains a
distinctly peri-Laurentian provenance (McNicoll et al., 2001),
supporting the emergence of the peri-Laurentian terranes and
subsidence of the Victoria Arc during the collision.
CONCLUSIONS
Comparison with the modern analogues indicates that the rec-
ognition of ancient arc-arc collision zones requires an integrated
and multidisciplinary approach. In Newfoundland, two coeval but
distinct and separated arc systems are preserved in the Dunnage
Zone: the peri-Laurentian Red Indian Lake Arc (ca. 473–460 Ma)
and the peri-Gondwanan Victoria Arc (ca. 473–453 Ma). The two
arc systems collided above a doubly dipping subduction zone
during early to middle Caradoc time along the Red Indian Line
(Figs. 12, 13), the main Iapetus suture zone, marking the end of
the Taconic orogeny and the start of the Salinic orogeny. Follow-
ing the collision the subduction stepped back into the Exploits-
Tetagouche backarc basin, leading to the closure of that tract
along the Silurian Dog Bay Line (Williams et al., 1993; van Staal
et al., 1998).
330 Zagorevski et al.
The separation of the Annieopsquotch accretionary tract and
the Victoria Arc is supported by contrasts in structural and tec-
tonic history, stratigraphy, basement characteristics, radiogenic
lead in mineral deposits, and fauna that are apparent across the
Red Indian Line (Williams et al., 1988). A brief review of the
northern Appalachians and the British Caledonides reveals great
similarity in the correlative tracts that also preserve evidence of
an Ordovician arc-arc collision, although the interpretation of the
polarity of subduction zones (e.g., Armstrong and Owen, 2001;
Moench and Aleinikoff, 2003) differs from the models based
on Newfoundland (e.g., van Staal et al., 1998). Future research
should aim to reconcile these differences. The timing of the arc-
arc collision appears to have been restricted to the Caradoc in all
of the correlative tracts. This similarity may reflect either coeval
collision along the entire length of the tract or the postcollisional
dispersal of the accreted terranes along the Laurentian margin
(e.g., Elders, 1987).
APPENDIX: U-Pb GEOCHRONOLOGY ANALYTICAL
TECHNIQUES
SHRIMP II analyses were conducted at the Geological
Survey of Canada (GSC) using analytical procedures described
by Stern (1997), with standards and U-Pb calibration methods
following Stern and Amelin (2003). Zircons from the samples
were cast in 2.5 cm diameter epoxy mounts (GSC mount 373 for
sample RAX05–900 and GSC mount 254 for sample RAX01–
904 along with fragments of the GSC laboratory standard zircon
(z6266, with 206Pb/238U age = 559 Ma). The midsections of the
zircons were exposed using 9, 6, and 1 µm diamond compound,
and the internal features of the zircons were characterized with
backscatter electrons (BSE) and cathodoluminescence (CL) uti-
lizing a Cambridge Instruments scanning electron microscope
(SEM). Mount surfaces were evaporatively coated with 10 nm
of high-purity Au. Analyses were conducted using a 16O– primary
beam, projected onto the zircons at 10 kV. The sputtered area
used for analysis was ca. 25 µm in diameter with a beam current
of ~5 nA and 13 nA for RAX05–900 and RAX01–904, respec-
tively. The count rates of 10 isotopes of Zr+, U+, Th+, and Pb+
in zircon were sequentially measured over 6 scans with a single
electron multiplier and a pulse counting system with deadtime of
22 ns. Off-line data processing was accomplished using custom-
ized in-house software. The 1σ external errors of 206Pb/238U ratios
reported in Table 1 incorporate a ±1.0% error in calibrating the
standard zircon (see Stern and Amelin, 2003). No fractionation
correction was applied to the Pb isotope data; common Pb cor-
rection utilized the measured 204Pb/206Pb and compositions mod-
eled after Cumming and Richards (1975). The 206Pb/238U ages for
the analyses have been corrected for common Pb using both the
204 and 207 methods (Stern, 1997), but there is generally no sig-
nificant difference in the results.
U-Pb TIMS analytical methods utilized in this study are out-
lined in Parrish et al. (1987). Heavy mineral concentrates were
prepared by standard crushing, grinding, WilfleyTM table, and
heavy liquid techniques. Mineral separates were sorted by mag-
netic susceptibility using a FrantzTM isodynamic separator. Mul-
tigrain zircon fractions analyzed were very strongly air abraded
following the method of Krogh (1982). Treatment of analyti-
cal errors follows Roddick et al. (1987) with errors for the ages
reported at the 2σ level (Table 2). U-Pb concordia diagrams are
presented in Figure 9. The data are plotted in concordia diagrams
with errors at the 2σ level (Fig. 9), using Isoplot v. 2.49 (Ludwig,
2001) to generate the plots.
A concordia age (Ludwig, 1998) is calculated for some of
the samples presented in this paper. A concordia age incorporates
errors in the decay constants and includes both an evaluation of
concordance and an evaluation of equivalence of the data (how
well the data fit the assumption that they are repeated measure-
ments of the same point). The calculated concordia ages and errors
quoted in the text are at 2σ, with decay constant errors included.
ACKNOWLEDGMENTS
This paper is a contribution to the Targeted Geoscience Initia-
tive 3–Appalachians (2005–2010) and Red Indian Line Tar-
geted Geoscience Initiative projects (2000–2003). B. O’Brien
(Geological Survey of Newfoundland and Labrador) is thanked
for the insightful field trip through Notre Dame Bay and cen-
tral Newfoundland that helped to inspire this manuscript. The
manuscript has benefited from reviews from S. Castonguay,
D. Chew, A. Draut, and P. Karabinos. This is Geological Survey
of Canada Contribution 20060169.
REFERENCES CITED
Abbott, L.D., Silver, E.A., and Galewsky, J., 1994, Structural evolution of a
modern arc-continent collision in Papua New Guinea: Tectonics, v. 13,
p. 1007–1034, doi: 10.1029/94TC01623.
Abers, G.A., and McCaffrey, R., 1994, Active arc-continent collision; earth-
quakes, gravity anomalies, and fault kinematics in the Huon-Finisterre
collision zone, Papua New Guinea: Tectonics, v. 13, p. 227–245, doi:
10.1029/93TC02940.
Armstrong, H.A., and Owen, A.W., 2001, Terrane evolution of the paratectonic
Caledonides of northern Britain: Geological Society [London] Journal,
v. 158, p. 475–486.
Arnott, R.J., McKerrow, W.S., and Cocks, L.R.M., 1985, The tectonics and dep-
ositional history of the Ordovician and Silurian rocks of Notre Dame Bay,
Newfoundland: Canadian Journal of Earth Sciences, v. 22, p. 607–618.
Ayuso, R.A., and Schulz, K.J., 2003, Nd-Pb-Sr isotope geochemistry and
origin of the Ordovician Bald Mountain and Mount Chase massive sul-
fide deposits, northern Maine: Economic Geology Monographs, v. 11,
p. 611–630.
Ayuso, R.A., Wooden, J.L., Foley, N.K., Slack, J.F., Sinha, A.K., and Persing,
H., 2003, Pb isotope geochemistry and U-Pb zircon (SHRIMP-RG) ages
of the Bald Mountain and Mount Chase massive sulfide deposits, north-
ern Maine; mantle and crustal contributions in the Ordovician: Economic
Geology Monographs, v. 11, p. 589–609.
Bird, P., 2003, An updated digital model of plate boundaries: Geochemistry,
Geophysics, Geosystems—G3, v. 4, p. 52.
Bostock, H.H., 1988, Geology and petrochemistry of the Ordovician volcano-
plutonic Robert’s Arm Group, Notre Dame Bay, Newfoundland: Geologi-
cal Survey of Canada Bulletin 369, 84 p.
Boutelier, D., Chemenda, A., and Burg, J.P., 2003, Subduction versus accretion
of intra-oceanic volcanic arcs; insight from thermo-mechanical analogue
experiments: Earth and Planetary Science Letters, v. 212, p. 31–45, doi:
10.1016/S0012-821X(03)00239-5.
 Tectonic architecture of an arc-arc collision zone
Calon, T.J., and Green, F.K., 1987, Preliminary results of a detailed structural
analysis of the Buchans Mine area: Geological Survey of Canada Paper,
v. 86–24, p. 273–288.
Cawood, P.A., van Gool, J.A.M., and Dunning, G.R., 1995, Collisional tecton-
ics along the Laurentian margin of the Newfoundland Appalachians: Geo-
logical Association of Canada Special Paper, v. 41, p. 283–301.
Cawood, P.A., and Nemchin, A.A., 2001, Paleogeographic development of the
East Laurentian margin; constraints from U-Pb dating of detrital zircons
in the Newfoundland Appalachians: Geological Society of America Bul-
letin, v. 113, p. 1234–1246, doi: 10.1130/0016-7606(2001)113<1234:
PDOTEL>2.0.CO;2.
Cawood, P.A., McCausland, P.J.A., and Dunning, G.R., 2001, Opening Iape-
tus; constraints from the Laurentian margin in Newfoundland: Geologi-
cal Society of America Bulletin, v. 113, p. 443–453, doi: 10.1130/0016-
7606(2001)113<0443:OICFTL>2.0.CO;2.
Chandler, F.W., Sullivan, R.W., and Currie, K.L., 1987, The age of the Spring-
dale Group, western Newfoundland, and correlative rocks; evidence for a
Llandovery overlap assemblage in the Canadian Appalachians: Transac-
tions of the Royal Society of Edinburgh: Earth Sciences, v. 78, p. 41–49.
Coish, R.A., and Rogers, N.W., 1987, Geochemistry of the Boil Mountain ophi-
olitic complex, Northwest Maine, and tectonic implications: Contributions
to Mineralogy and Petrology, v. 97, p. 51–65, doi: 10.1007/BF00375214.
Colman-Sadd, S., Hayes, J., Knight, I., and Pereira, C.P.G., 1990, The geology
of the Island of Newfoundland; map 90–01: Report of Activities—New-
foundland Geological Survey Branch, v. 1990, p. 24.
Colman-Sadd, S.P., Dunning, G.R., and Dec, T., 1992, Dunnage-Gander rela-
tionships and Ordovician orogeny in central Newfoundland; a sediment
provenance and U/Pb age study: American Journal of Science, v. 292,
p. 317–355.
Cumming, G.L., and Richards, J.R., 1975, Ore lead in a continuously chang-
ing Earth: Earth and Planetary Science Letters, v. 28, p. 155–171, doi:
10.1016/0012-821X(75)90223-X.
Dec, T., Swinden, H.S., and Dunning, R.G., 1997, Lithostratigraphy and geo-
chemistry of the Cottrells Cove Group, Buchans-Roberts Arm volcanic
belt; new constraints for the paleotectonic setting of the Notre Dame Sub-
zone, Newfoundland Appalachians: Canadian Journal of Earth Sciences,
v. 34, p. 86–103.
Dube, B., Dunning, G.R., Lauziere, K., and Roddick, J.C., 1996, New
insights into the Appalachian Orogen from geology and geochronology
along the Cape Ray fault zone, Southwest Newfoundland: Geological
Society of America Bulletin, v. 108, p. 101–116, doi: 10.1130/0016-
7606(1996)108<0101:NIITAO>2.3.CO;2.
Dunning, G.R., 1987, Geology of the Annieopsquotch Complex, Southwest
Newfoundland: Canadian Journal of Earth Sciences, v. 24, p. 1162–1174.
Dunning, G.R., and Krogh, T.E., 1985, Geochronology of ophiolites of the
Newfoundland Appalachians: Canadian Journal of Earth Sciences, v. 22,
p. 1659–1670.
Dunning, G.R., and O’Brien, S.J., 1989, Late Proterozoic–early Paleozoic crust
in the Hermitage Flexure, Newfoundland Appalachians; U/Pb ages and
tectonic significance: Geology, v. 17, p. 548–551, doi: 10.1130/0091-
7613(1989)017<0548:LPEPCI>2.3.CO;2.
Dunning, G.R., Kean, B.F., Thurlow, J.G., and Swinden, H.S., 1987, Geochro-
nology of the Buchans, Roberts Arm, and Victoria Lake groups and Mans-
field Cove Complex, Newfoundland: Canadian Journal of Earth Sciences,
v. 24, p. 1175–1184.
Dunning, G.R., O’Brien, S.J., Colman-Sadd, S.P., Blackwood, R.F., Dickson,
W.L., O’Neill, P.P., and Krogh, T.E., 1990, Silurian orogeny in the New-
foundland Appalachians: Journal of Geology, v. 98, p. 895–913.
Elders, C.F., 1987, The provenance of granite boulders in conglomerates of the
northern and central belts of the Southern Uplands of Scotland: Geologi-
cal Society [London] Journal, v. 144, p. 853–863.
Evans, D.T.W., and Kean, B.F., 2002, The Victoria Lake Supergroup, central
Newfoundland—Its definition, setting and volcanogenic massive sulphide
mineralization: Newfoundland Department of Mines and Energy, Geo-
logical Survey, Open File NFLD/2790, 68 p.
Evans, D.T.W., Kean, B.F., and Dunning, G.R., 1990, Geological studies, Vic-
toria Lake Group, central Newfoundland: Government of Newfoundland
and Labrador, Department of Mines and Energy, Geological Survey,
Report 90–1, p. 131–144.
Fitz, T.J., 2002, Magmatism in an extensional setting within the Bronson Hill
belt: The Chickwolnepy Intrusions of northern New Hampshire: Physics
and Chemistry of the Earth, v. 27, p. 97–108.
331
Galewsky, J., and Silver, E.A., 1997, Tectonic controls on facies transitions in an
oblique collision; the western Solomon Sea, Papua New Guinea: Geologi-
cal Society of America Bulletin, v. 109, p. 1266–1278, doi: 10.1130/0016-
7606(1997)109<1266:TCOFTI>2.3.CO;2.
Galewsky, J., Silver, E.A., Gallup, C.D., Edwards, R.L., and Potts, D.C., 1996,
Foredeep tectonics and carbonate platform dynamics in the Huon Gulf,
Papua New Guinea: Geology, v. 24, p. 819–822, doi: 10.1130/0091-
7613(1996)024<0819:FTACPD>2.3.CO;2.
Gerbi, C.C., Johnson, S.E., and Aleinikoff, J.N., 2006, Origin and orogenic role
of the Chain Lakes massif, Maine and Quebec: Canadian Journal of Earth
Sciences, v. 43, p. 339–366, doi: 10.1139/E05-112.
Hall, R., 2000, Neogene history of collision in the Halmahera region, Indone-
sia: Proceedings of the Indonesian Petroleum Association, 27th Annual
Convention, p. 487–493.
Hall, R., 2002, Cenozoic geological and plate tectonic evolution of SE Asia and
the SW Pacific: Computer-based reconstructions, model and animations:
Journal of Asian Earth Sciences, v. 20, p. 353–431, doi: 10.1016/S1367-
9120(01)00069-4.
Hibbard, J., van Staal, C., Rankin, D., and Williams, H., 2004, Lithotectonic
map of the Appalachian Orogen, Canada–United States of America: Geo-
logical Survey of Canada, Map 2096A, scale 1:500,000.
Jenner, G.A., and Swinden, H.S., 1993, The Pipestone Pond Complex, central
Newfoundland; complex magmatism in an eastern Dunnage Zone ophiol-
ite: Canadian Journal of Earth Sciences, v. 30, p. 434–448.
Johnson, S.C., and McLeod, M.J., 1996, The New River Belt; a unique segment
along the western margin of the Avalon composite terrane, southern New
Brunswick, Canada: Geological Society of America Special Paper 304,
p. 149–164.
Karabinos, P., Samson, S.D., Hepburn, J.C., and Stoll, H.M., 1998, Taconian
Orogeny in the New England Appalachians; collision between Lau-
rentia and the Shelburne Falls Arc: Geology, v. 26, p. 215–218, doi:
10.1130/0091-7613(1998)026<0215:TOITNE>2.3.CO;2.
Kerr, A., Jenner, G.A., and Fryer, B.J., 1995, Sm-Nd isotopic geochemistry of
Precambrian to Paleozoic granitoid suites and the deep-crustal structure
of the southeast margin of the Newfoundland Appalachians: Canadian
Journal of Earth Sciences, v. 32, p. 224–245.
Krogh, T.E., 1982, Improved accuracy of U-Pb ages by creation of more con-
cordant systems using an air abrasion technique: Geochimica et Cosmo-
chimica Acta, v. 46, p. 637–649, doi: 10.1016/0016-7037(82)90165-X.
Kusky, T.M., Kidd, W.S.F., and Bradley, D.C., 1987, Displacement history of
the Northern Arm Fault, and its bearing on the post-Taconic evolution of
north-central Newfoundland: Journal of Geodynamics, v. 7, p. 105–133,
doi: 10.1016/0264-3707(87)90067-6.
Lallemand, S.E., Popoff, M., Cadet, J.-P., Bader, A.-G., Pubellier, M., Rangin,
C., and Deffontaines, B., 1998, Genetic relations between the central
and southern Philippine Trench and the Sangihe Trench: Journal of Geo-
physical Research, B, Solid Earth and Planets, v. 103, p. 933–950, doi:
10.1029/97JB02620.
Leggett, J.K., 1987, The Southern Uplands as an accretionary prism; the impor-
tance of analogues in reconstructing palaeogeography: Geological Soci-
ety [London] Journal, v. 144, p. 737–752.
Lissenberg, C.J., 2005, The origin and tectonic evolution of the Annieopsquotch
ophiolite belt, Newfoundland Appalachians: Ontario, Canada, University
of Ottawa, 149 p.
Lissenberg, C.J., Zagorevski, A., McNicoll, V.J., van Staal, C.R., and Whalen, J.B.,
2005a, Assembly of the Annieopsquotch accretionary tract, Newfoundland
Appalachians; age and geodynamic constraints from syn-kinematic intru-
sions: Journal of Geology, v. 113, p. 553–570, doi: 10.1086/431909.
Lissenberg, C.J., van Staal, C.R., Bedard, J.H., and Zagorevski, A., 2005b, Geo-
chemical constraints on the origin of the Annieopsquotch ophiolite belt,
Newfoundland Appalachians: Geological Society of America Bulletin, v.
117, p. 1413–1426, doi: 10.1130/B25731.1.
Ludwig, K.R., 1998, On the treatment of concordant uranium-lead ages: Geo-
chimica et Cosmochimica Acta, v. 62, p. 665–676, doi: 10.1016/S0016-
7037(98)00059-3.
Ludwig, K.R., 2001, User’s manual for Isoplot/Ex rev. 2.49: A Geochronologi-
cal Toolkit for Microsoft Excel: Berkeley, California, Berkeley Geochro-
nological Center, Special Publication 1a, 55 p.
MacLachlan, K., and Dunning, G., 1998a, U-Pb ages and tectonomagmatic
relationships of early Ordovician low-Ti tholeiites, boninites and related
plutonic rocks in central Newfoundland, Canada: Contributions to Miner-
alogy and Petrology, v. 133, p. 235–258, doi: 10.1007/s004100050450.
332 Zagorevski et al.
MacLachlan, K., and Dunning, G.R., 1998b, U-Pb ages and tectonomagmatic
relationships of Middle Ordovician volcanic rocks of the Wild Bight
Group, Newfoundland Appalachians: Canadian Journal of Earth Sciences,
v. 35, p. 998–1017, doi: 10.1139/cjes-35-9-998.
MacLachlan, K., O’Brien, B.H., and Dunning, G.R., 2001, Redefinition of the
Wild Bight Group, Newfoundland; implications for models of island-arc
evolution in the Exploits Subzone: Canadian Journal of Earth Sciences,
v. 38, p. 889–907, doi: 10.1139/cjes-38-6-889.
Macpherson, C.G., Forde, E.J., Hall, R., and Thirlwall, M.F., 2003, Geochemi-
cal evolution of magmatism in an arc-arc collision; the Halmahera and
Sangihe Arcs, eastern Indonesia: Geological Society [London] Special
Publication 219, p. 207–220.
McConnell, B., and O’Brien, B., 2000, Stratigraphic and petrogenetic relation-
ships between arc and non-arc basalts in western and southern parts of
the Ordovician Wild Bight Group, Dunnage Zone, Newfoundland: St.
John’s, NF, Canada, Government of Newfoundland and Labrador Current
Research Report, v. 2000–1, p. 217–230.
McConnell, B.J., O’Brien, B.H., and Nowlan, G.S., 2002, Late Middle Ordovi-
cian olistostrome formation and magmatism along the Red Indian Line,
the Laurentian Arc–Gondwanan Arc boundary, at Sops Head, Newfound-
land: Canadian Journal of Earth Sciences, v. 39, p. 1625–1633, doi:
10.1139/e02-084.
McKerrow, W.S., and van Staal, C.R., 2000, The Palaeozoic time scale reviewed:
Geological Society [London] Special Publication 179, p. 5–8.
McNicoll, V., van Staal, C.R., and Waldron, J.W.F., 2001, Accretionary his-
tory of the northern Appalachians: SHRIMP study of Ordovician–Silurian
syntectonic sediments in the Canadian Appalachians: St. John’s, New-
foundland: Geological Association of Canada/Mineralogical Association
of Canada Annual Meeting Abstracts, v. 26, p. 100.
Milsom, J., 2001, Subduction in eastern Indonesia; how many slabs?: Tectono-
physics, v. 338, p. 167–178, doi: 10.1016/S0040-1951(01)00137-8.
Moench, R., and Aleinikoff, J.N., 2003, Stratigraphy, geochronology, and
accretionary terrane settings of two Bronson Hill arc sequences, northern
New England: Physics and Chemistry of the Earth, v. 28, p. 113–160.
O’Brien, B., Swinden, H.S., Dunning, G.R., Williams, S.H., and O’Brien, F.H.C.,
1997, A peri-Gondwanan arc–back arc complex in Iapetus; Early–Mid
Ordovician evolution of the Exploits Group, Newfoundland: American
Journal of Science, v. 297, p. 220–272.
O’Brien, B.H., 2003, Geology of the central Notre Dame Bay region (Parts of
NTS areas 2E/3,6,11) northeastern Newfoundland: Government of New-
foundland and Labrador, Department of Mines and Energy, Geological
Survey, Report 03–03, 147 p.
Parrish, R.R., Roddick, J.C., Loveridge, W.D., Sullivan, R.W., and Anonymous,
1987, Uranium-lead analytical techniques at the Geochronology Labora-
tory, Geological Survey of Canada: Geological Survey of Canada Paper,
v. 87–2, p. 3–7.
Phillips, E.R., Evans, J.A., Stone, P., Horstwood, M.S.A., Floyd, J.D., Smith,
R.A., Akhurst, M.C., and Barron, H.F., 2003, Detrital Avalonian zircons in
the Laurentian Southern Uplands terrane, Scotland: Geology, v. 31, p. 625–
628, doi: 10.1130/0091-7613(2003)031<0625:DAZITL>2.0.CO;2.
Pickering, K.T., 1987, Wet-sediment deformation in the Upper Ordovician
Point Leamington Formation; an active thrust-imbricate system during
sedimentation, Notre Dame Bay, north-central Newfoundland, in Jones,
M.F., and Preston, R.M.F., eds., Deformation of Sediments and Sedi-
mentary Rocks: Geological Society [London] Special Publication 29,
p. 213–239.
Pickett, J.W., 1987, Geology and geochemistry of the Skidder Basalt: Geologi-
cal Survey of Canada Paper, v. 86–24, p. 195–218.
Pubellier, M., Bader, A.G., Rangin, C., Deffontaines, B., and Quebral, R., 1999,
Upper plate deformation induced by subduction of a volcanic arc; the Snel-
lius Plateau (Molucca Sea, Indonesia and Mindanao, Philippines): Tecto-
nophysics, v. 304, p. 345–368, doi: 10.1016/S0040-1951(98)00300-X.
Roddick, J.C., Loveridge, W.D., and Parrish, R.R., 1987, Precise U/Pb dating of
zircon at the sub-nanogram Pb level: Chemical Geology: Isotope Geosci-
ence Section, v. 66, p. 111–121, doi: 10.1016/0168-9622(87)90034-0.
Rogers, N., 2004, Red Indian Line geochemical database: Geological Survey of
Canada Open File 4605.
Rogers, N., and van Staal, C.R., 2002, Toward a Victoria Lake Supergroup; a
provisional stratigraphic revision of the Red Indian to Victoria lakes area,
central Newfoundland: Newfoundland Geological Survey Branch, Cur-
rent Research, Report 02–1, p. 185–195.
Rogers, N., van Staal, C.R., and Theriault, R., 2003, Volcanology and tec-
tonic setting of the northern Bathurst mining camp; Part I, Extension
and rifting of the Popelogan Arc: Economic Geology Monographs, v. 11,
p. 157–179.
Rogers, N., van Staal, C.R., Pollock, J., and Zagorevski, A., 2005, Geology,
Lake Ambrose and part of Buchans, Newfoundland (NTS 12-A/10 and
part of 12-A/15): Map, scale 1:50,000.
Rogers, N., van Staal, C.R., McNicoll, V., Pollock, J., and Zagorevski, A.W.,
2006, Neoproterozoic and Cambrian arc magmatism along the eastern
margin of the Victoria Lake Supergroup: A remnant of Ganderian base-
ment in central Newfoundland? in McCausland, P.J.A., Murphy, J.B., and
MacNiocaill, C., eds., Endings and Beginnings: Paleogeography of the
Neoproterozoic-Cambrian Transition: Amsterdam, Elsevier, Precambrian
Research, v. 147, p. 320–341.
Ryan, P.D., and Dewey, J.F., 2004, The South Connemara Group reinterpreted:
A subduction-accretion complex in the Caledonides of Galway Bay, west-
ern Ireland: Journal of Geodynamics, v. 37, p. 513–529, doi: 10.1016/
j.jog.2004.02.018.
Sajona, F.G., Maury, R.C., Pubellier, M., Leterrier, J., Bellon, H., and Cotten,
J., 2000, Magmatic source enrichment by slab-derived melts in a young
post-collision setting, central Mindanao (Philippines): Lithos, v. 54,
p. 173–206, doi: 10.1016/S0024-4937(00)00019-0.
Stacey, J.S., and Kramers, J.D., 1975, Approximation of terrestrial lead iso-
tope evolution by a two-stage model: Earth and Planetary Science Letters,
v. 26, p. 207–221, doi: 10.1016/0012-821X(75)90088-6.
Stern, R.A., 1997, The GSC Sensitive High Resolution Ion Microprobe
(SHRIMP); analytical techniques of zircon U-Th-Pb age determina-
tions and performance evaluation: Geological Survey of Canada, Current
Research, Report 1997-F, 31 p.
Stern, R.A., and Amelin, Y., 2003, Assessment of errors in SIMS zircon U-
Pb geochronology using a natural zircon standard and NIST SRM 610
glass: Chemical Geology, v. 197, p. 111–146, doi: 10.1016/S0009-
2541(02)00320-0.
Swinden, H.S., Jenner, G.A., Kean, B.F., and Evans, D.T.W., 1989, Volcanic
rock geochemistry as a guide for massive sulphide exploration in central
Newfoundland: Report of Activities—Mineral Development Division (St
John’s, 1985), v. 89–1, p. 201–219.
Swinden, H.S., Jenner, G.A., and Szybinski, Z.A., 1997, Magmatic and tectonic
evolution of the Cambrian–Ordovician Laurentian margin of Iapetus; geo-
chemical and isotopic constraints from the Notre Dame Subzone, New-
foundland: Geological Society of America Memoir 191, p. 337–365.
Thurlow, J.G., Spencer, C.P., Boerner, D.E., Reed, L.E., and Wright, J.A., 1992,
Geological interpretation of a high resolution reflection seismic survey at
the Buchans Mine, Newfoundland: Canadian Journal of Earth Sciences,
v. 29, p. 2022–2037.
Tucker, R.D., and Robinson, P., 1990, Age and setting of the Bronson Hill mag-
matic arc; a re-evaluation based on U-Pb zircon ages in southern New
England: Geological Society of America Bulletin, v. 102, p. 1404–1419,
doi: 10.1130/0016-7606(1990)102<1404:AASOTB>2.3.CO;2.
Tucker, R.D., O’Brien, S.J., and O’Brien, B.H., 1994, Age and implications of
Early Ordovician (Arenig) plutonism in the type area of the Bay du Nord
Group, Dunnage Zone, southern Newfoundland Appalachians: Canadian
Journal of Earth Sciences, v. 31, p. 351–357.
Valverde-Vaquero, P., and van Staal, C.R., 2002, Geology and magnetic anoma-
lies of the Exploits-Meelpaeg boundary zone in the Victoria Lake area
(central Newfoundland); regional implications: Newfoundland Geologi-
cal Survey Branch, Current Research, Report 02–1, p. 197–209.
Valverde-Vaquero, P., van Staal, C.R., McNicoll, V., and Dunning, G.R., 2006,
Mid–Late Ordovician magmatism and metamorphism along the Gander
margin in central Newfoundland: Geological Society [London] Journal,
v. 163, p. 347–362, doi: 10.1144/0016-764904-130.
van der Pluijm, B.A., Karlstrom, K.E., and Williams, P.F., 1987, Fossil evidence
for fault-derived stratigraphic repetition in the northeastern Newfoundland
Appalachians: Canadian Journal of Earth Sciences, v. 24, p. 2337–2350.
van der Velden, A.J., van Staal, C.R., and Cook, F.A., 2004, Crustal structure, fossil
subduction, and the tectonic evolution of the Newfoundland Appalachians;
evidence from a reprocessed seismic reflection survey: Geological Society
of America Bulletin, v. 116, p. 1485–1498, doi: 10.1130/B25518.1.
van der Voo, R., Johnson, R.J.E., van der Pluijm, B.A., and Knutson, L.C.,
1991, Paleogeography of some vestiges of Iapetus; paleomagnetism of the
Ordovician Robert’s Arm, Summerford and Chanceport groups, central
 Tectonic architecture of an arc-arc collision zone
Newfoundland: Geological Society of America Bulletin, v. 103, p. 1564–
1575, doi: 10.1130/0016-7606(1991)103<1564:POSVOI>2.3.CO;2.
van Staal, C.R., 2005, North America; Northern Appalachians, in Selley, R.C.,
et al., eds., Encyclopedia of Geology: Oxford, Elsevier, p. 81–92.
van Staal, C.R., Sullivan, R.W., and Whalen, J.B., 1996, Provenance of tec-
tonic history of the Gander Zone in the Caledonian/Appalachian Orogen;
implications for the origin and assembly of Avalon: Geological Society of
America Special Paper 304, p. 347–367.
van Staal, C.R., Dewey, J.F., MacNiocaill, C., and McKerrow, W.S., 1998, The
Cambrian–Silurian tectonic evolution of the Northern Appalachians and
British Caledonides; history of a complex, west and southwest Pacific-
type segment of Iapetus, in Blundell, D.J., and Scott, A.C., eds., Lyell:
The Past Is the Key to the Present: Geological Society [London] Special
Publication 143, p. 199–242.
van Staal, C.R., Wilson, R.A., Rogers, N., Fyffe, L.R., Langton, J.P., McCutch-
eon, S.R., McNicoll, V., and Ravenhurst, C.E., 2003, Geology and tec-
tonic history of the Bathurst Supergroup, Bathurst mining camp, and its
relationships to coeval rocks in southwestern New Brunswick and adja-
cent mine; a synthesis: Economic Geology Monographs, v. 11, p. 37–60.
van Staal, C.R., McNicoll, V., Valverde-Vaquero, P., Barr, S.M., Fyffe, L.R.,
and Reusch, D.N., 2004, Ganderia, Avalonia, and the Salinic and Acadian
orogenies: Geological Society of America Abstracts with Programs, v. 36,
no. 2, p. 128–129.
van Staal, C.R., Lissenberg, C.J., Pehrsson, S., Zagorevski, A., Valverde-
Vaquero, P., Herd, R.K., McNicoll, V., and Whalen, J.B., 2005a, Geology,
Puddle Pond, Newfoundland (NTS 12-A/05): Map, scale 1:50,000.
van Staal, C.R., Valverde-Vaquero, P., Zagorevski, A., Boutsma, S., Pehrsson,
S., van Noorden, M., and McNicoll, V., 2005b, Geology, King George IV
Lake, Newfoundland (NTS 12-A/04): Map, scale 1:50,000.
van Staal, C.R., Valverde-Vaquero, P., Zagorevski, A., Rogers, N., Lissenberg,
C.J., and McNicoll, V., 2005c, Geology, Victoria Lake, Newfoundland
(NTS 12-A/06): Map, scale 1:50,000.
van Staal, C.R., Whalen, J.B., McNicoll, V.J., Pehrsson, S., Lissenberg, C.J.,
Zagorevski, A., van Breemen, O., and Jenner, G.A., 2007, The Notre
Dame arc and the Taconic orogeny in Newfoundland, in Hatcher, R.D.,
Jr., Carlson, M.P., McBride, J.H., and Martínez-Catalán, J.R., eds., 4-D
Framework of Continental Crust: Geological Society of America Memoir
200, p. 511–552, doi: 10.1130/2007.1200(26).
Waldron, J.W.F., and van Staal, C.R., 2001, Taconian Orogeny and the accre-
tion of the Dashwoods Block; a peri-Laurentian microcontinent in
the Iapetus Ocean: Geology, v. 29, p. 811–814, doi: 10.1130/0091-
7613(2001)029<0811:TOATAO>2.0.CO;2.
Wasowski, J.J., Jacobi, R.D., Elliott, C.G., Williams, P.F., Arnott, R., McKer-
row, W.S., and Cocks, L.R.M., 1986, The tectonics and depositional his-
tory of the Ordovician and Silurian rocks of Notre Dame Bay, Newfound-
land; discussions and reply: Canadian Journal of Earth Sciences, v. 23,
p. 583–590.
Whalen, J.B., Jenner, G.A., Longstaffe, F.J., Gariepy, C., and Fryer, B.J., 1997,
Implications of granitoid geochemical and isotopic (Nd, O, Pb) data from
333
the Cambrian–Ordovician Notre Dame Arc for the evolution of the central
mobile belt, Newfoundland Appalachians: Geological Society of America
Memoir 191, p. 367–395.
Whalen, J.B., McNicoll, V.J., van Staal, C.R., Lissenberg, C.J., Longstaffe, F.J.,
Jenner, G.A., and van Breemen, O., 2006, Spatial, temporal and geochem-
ical characteristics of Silurian collision-zone magmatism, Newfoundland
Appalachians: An example of a rapidly evolving magmatic system related
to slab break-off: Lithos, v. 89, p. 377–404.
Whitmore, G.P., Crook, K.A.W., and Johnson, D.P., 1999, Sedimentation in a
complex convergent margin; the Papua New Guinea collision zone of the
western Solomon Sea: Marine Geology, v. 157, p. 19–45, doi: 10.1016/
S0025-3227(98)00132-7.
Williams, H., 1995a, Chapter 2: Temporal and spatial divisions, in Williams,
H., ed., Geology of the Appalachian-Caledonian Orogen in Canada and
Greenland: Geological Survey of Canada, p. 23–42.
Williams, H., 1995b, Chapter 3: Dunnage Zone—Newfoundland, in Williams,
H., ed., Geology of the Appalachian-Caledonian Orogen in Canada and
Greenland: Geological Survey of Canada, p. 142–166.
Williams, H., Colman-Sadd, S.P., Swinden, H.S., and Anonymous, 1988, Tec-
tonic-stratigraphic subdivisions of central Newfoundland: Geological
Survey of Canada Paper, v. 88–1B, p. 91–98.
Williams, H., Currie, K.L., and Piasecki, M.A.J., 1993, The Dog Bay Line; a
major Silurian tectonic boundary in Northeast Newfoundland: Canadian
Journal of Earth Sciences, v. 30, p. 2481–2494.
Williams, H., Lafrance, B., Dean, P.L., Williams, P.F., Pickering, K.T., and van
der Pluijm, B.A., 1995, Chapter 4: Badger belt, in Williams, H., ed., Geol-
ogy of the Appalachian-Caledonian Orogen in Canada and Greenland:
Geological Survey of Canada, p. 403–413.
Wilson, R.A., 2003, Geochemistry and petrogenesis of Ordovician arc-related
mafic volcanic rocks in the Popelogan Inlier, northern New Brunswick:
Canadian Journal of Earth Sciences, v. 40, p. 1171–1189, doi: 10.1139/
e03-034.
Zagorevski, A., Rogers, N., McNicoll, V., Lissenberg, C.J., van Staal, C.R., and
Valverde-Vaquero, P., 2006, Lower to Middle Ordovician evolution of
peri-Laurentian arc and back-arc complexes in the Iapetus: Constraints
from the Annieopsquotch Accretionary Tract, central Newfoundland:
Geological Society of America Bulletin, v. 118, p. 324–342, doi: 10.1130/
B25775.1.
Zagorevski, A., McNicoll, V., van Staal, C., and Rogers, N., 2007a, Upper
Cambrian to Upper Ordovician peri-Gondwanan island arc activity in the
Victoria Lake Supergroup, central Newfoundland: Tectonic development
of the northern Ganderian margin: American Journal of Science, v. 307,
p. 339–370.
Zagorevski, A., van Staal, C.R., and McNicoll, V.J., 2007b, Distinct Taconic,
Salinic and Acadian deformation along the Iapetus suture zone, New-
foundland Appalachians: Canadian Journal of Earth Sciences, v. 44.
doi:10.1139/E07-037.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

The Catalina Schist: Evidence for middle Cretaceous subduction erosion of southwestern
North America

M. Grove*
Department of Earth & Space Sciences, University of California, Los Angeles, 3806 Geology, Los Angeles, California 90095, USA

G.E. Bebout
Department of Earth & Environmental Sciences, Lehigh University, 31 Williams Drive, Bethlehem, Pennsylvania 18015, USA

C.E. Jacobson
Department of Geological and Atmospheric Sciences, 253 Science I, Iowa State University, Ames, Iowa 50011-3212, USA

A.P. Barth
Department of Earth Sciences, Indiana–Purdue University, 723 West Michigan Street, Indianapolis, Indiana 46202, USA

D.L. Kimbrough
Department of Geological Sciences, San Diego State University, 5500 Campanile Drive, San Diego, California 92182-1020, USA

R.L. King
School of Earth and Environmental Sciences, Washington State University, Pullman, Washington 99164-2812, USA

Haibo Zou
O.M. Lovera
Department of Earth & Space Sciences, University of California, Los Angeles, 3806 Geology, Los Angeles, California 90095, USA

B.J. Mahoney
Department of Geology, University of Wisconsin–Eau Claire, Phillips 157, Eau Claire, Wisconsin 54702-4004, USA

G.E. Gehrels
Department of Geosciences, University of Arizona, Tucson, Gould-Simpson Building 529, Tucson, Arizona 85721, USA

ABSTRACT

The Catalina Schist underlies the inner southern California borderland of south-
western North America. On Santa Catalina Island, amphibolite facies rocks that
recrystallized and partially melted at ca. 115 Ma and at 40 km depth occur atop an
inverted metamorphic stack that juxtaposes progressively lower grade, high-pressure/
temperature (PT) rocks across low-angle faults. This inverted metamorphic sequence

*marty@oro.ess.ucla.edu

Grove, M., Bebout, G.E., Jacobson, C.E., Barth, A.P., Kimbrough, D.L., King, R.L., Zou, H., Lovera, O.M., Mahoney, B.J., and Gehrels, G.E., 2008, The Catalina
Schist: Evidence for middle Cretaceous subduction erosion of southwestern North America, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and
Applications of the Sedimentary Record in Arc Collision Zones: Geological Society of America Special Paper 436, p. 335–361, doi: 10.1130/2008.2436(15). For
permission to copy, contact editing@geosociety.org. ©2008 The Geological Society of America. All rights reserved.

335
336 Grove et al.

has been regarded as having formed within a newly initiated subduction zone. How-
ever, subduction initiation at ca. 115 Ma has been difficult to reconcile with regional
geologic relationships, because the Catalina Schist formed well after emplacement of
the adjacent Peninsular Ranges batholith had begun in earnest. New detrital zircon
U-Pb age results indicate that the Catalina Schist accreted over a ~20 m.y. interval.
The amphibolite unit metasediments formed from latest Neocomian to early Aptian
(122–115 Ma) craton-enriched detritus derived mainly from the pre-Cretaceous wall
rocks and Early Cretaceous volcanic cover of the Peninsular Ranges batholith. In
contrast, lawsonite-blueschist and lower grade rocks derived from Cenomanian sedi-
ments dominated by this batholith’s plutonic and volcanic detritus were accreted
between 97 and 95 Ma. Seismic data and geologic relationships indicate that the
Catalina Schist structurally underlies the western margin of the northern Penin-
sular Ranges batholith. We propose that construction of the Catalina Schist com-
plex involved underthrusting of the Early Cretaceous forearc rocks to a subcrustal
position beneath the western Peninsular Ranges batholith. The heat for amphibolite
facies metamorphism and anatexis observed within the Catalina Schist was supplied
by the western part of the batholith while subduction was continuous along the mar-
gin. Progressive subduction erosion ultimately juxtaposed the high-grade Catalina
Schist with lower grade blueschists accreted above the subduction zone by 95 Ma.
This coincided with an eastern relocation of arc magmatism and emplacement of the
ca. 95 Ma La Posta tonalite-trondjhemite-granodiorite suite of the eastern Peninsular
Ranges batholith. Final assembly of the Catalina Schist marked the initial stage of the
Late Cretaceous–early Tertiary craton-ward shift of arc magmatism and deformation
of southwestern North America that culminated in the Laramide orogeny.

Keywords: Catalina Schist, U-Pb, zircon, subduction erosion.

INTRODUCTION intervening rocks (Page, 1981; Hall, 1991) or by lateral transport

Subduction erosion is a fundamental convergent margin pro-
cess that tectonically shortens the forearc region by displacing
rocks from the overriding plate to subcrustal positions and/or into
the asthenosphere (Scholl et al., 1980; von Huene and Scholl,
1991, 1993). Roughly half of all present-day convergent margins
exhibit geologic and/or geophysical evidence for subduction-
related removal of rocks from the forearc regions of the overrid-
ing plate to positions beneath the arc and/or into the mantle (e.g.,
Clift and Vannucchi, 2004). The globally average rate at which
continental debris is transported toward the mantle is significant
(~2.5 km3/year; Scholl and von Huene, 2007). Unfortunately, its
consumptive nature means that direct evidence for subduction
erosion in ancient convergent margins is generally ambiguous or
lacking. Indirect evidence for subduction erosion such as “miss-
ing” or “telescoped” forearc crust is difficult to distinguish from
alternative tectonic processes such as margin-parallel strike-slip
faulting (e.g., Karig, 1980). A classic illustration of the difficulty
in deciding between subduction erosion and margin-parallel
strike-slip dispersal of forearc rocks is provided by the Nacimiento
fault in west-central California (Page, 1970). There, plutonic and
metamorphic rocks of the Salinian block are juxtaposed against
a high-pressure, low-temperature mélange of the Sur-Obispo
terrane. This anomalous juxtaposition of lithotectonic belts has
been explained both by underthrusting (subduction erosion) of
of hundreds (Hill and Dibblee, 1953; Suppe, 1970; Dickinson,
1983; Dickinson et al., 2005) to thousands of kilometers (Page,
1982; McWilliams and Howell, 1982; Vedder et al., 1983; Debi-
che et al., 1987).
Like the Sur-Obispo terrane, seismic data and geologic
relationships indicate that the high-pressure/temperature (PT)
Catalina Schist of southern California is also in direct contact
with its coeval magmatic arc. This relationship was revealed by
middle Miocene extension that exhumed the Catalina Schist from
beneath the northwestern margin of the Peninsular Ranges batho-
lith across an east-dipping Miocene detachment fault (Crouch
and Suppe, 1993) (Fig. 1). The dramatic extension of the border-
land occurred during a middle Miocene microplate capture event
(e.g., Nicholson et al., 1994) triggered by Pacific–North Ameri-
can shearing along the margin (e.g., Atwater, 1970). During this
event, Late Cretaceous forearc strata and underlying Jurassic and
earliest Cretaceous basement rocks were ripped from the western
margin of the Peninsular Ranges batholith, rotated clockwise,
and translated northwestward into the outer California border-
land along a detachment system underlain by the Catalina Schist
(Wright, 1991; Crouch and Suppe, 1993; Bohannon and Geist,
1998; Ingersoll and Rumelhart, 1999).
Although the Catalina Schist contains abundant blue-
schist facies rocks typical of subduction zone environments, it
also has higher temperature lithologies that define an inverted
Figure 1. Geologic setting of the Catalina Schist. Location map in the upper right shows the distribution of high-
pressure/temperature (PT) subduction complexes and Cretaceous–Tertiary batholiths in southwestern North America.
The offshore distribution of subduction complexes is estimated primarily from Crouch and Suppe (1993), Bohannon
and Geist (1998), Sedlock (1988a, b), Bonini and Baldwin (1998), and Fletcher et al. (2007). The box outlines the loca-
tion of the southern California continental borderland. The simplified map of the borderland shown in the lower left is
based upon Bohannon and Geist (1998), whereas cross section Y–Y′ is after Crouch and Suppe (1993). Symbols P and
P* denote formerly contiguous rocks that dextrally sheared and rotated clockwise during middle Miocene rifting. The
box in the lower left shows the location of Catalina Island. The geologic map and cross section X–X′ of Catalina Island
in the upper left are after Platt (1976). The distribution of lawsonite-albite and actinolite-albite rocks is from Altheim et
al. (1997). Note that epidote-blueschist and epidote-amphibolite rocks are distinguished from Platt’s (1976) greenschist
unit. A coherent kilometer-scale mass of epidote-amphibolite facies metagabbro that directly underlies the amphibolite
unit was included by Platt (1976) within his amphibolite unit. K–T—Cretaceous–Tertiary.
338 Grove et al.
metamorphic sequence that includes amphibolite facies rocks
at the highest structural levels (see inset in Fig. 1). These partly
melted rocks underwent peak grade conditions at ca. 115 Ma
(Suppe and Armstrong, 1972; Mattinson, 1986; Sorensen and
Barton, 1987; Grove and Bebout, 1995; Anczkiewicz et al., 2004).
A longstanding explanation for the unusually high-temperature,
inverted metamorphism of the Catalina Schist is that it reflects
progressive accretion of early subducted rocks beneath unrefri-
gerated mantle lithosphere during nascent subduction (e.g., Platt,
1975). Although Platt’s (1975) nascent subduction hypothesis
was based upon the Catalina Schist, he extended it to explain the
origin of meter- to kilometer-scale high-grade blocks within the
Franciscan Complex and other accretionary complexes along the
western margin of North America (e.g., Platt, 1975; Cloos, 1985;
Wakabayashi, 1990, 1992, 1999; Anczkiewicz et al., 2004; Wak-
abayashi and Dumitru, 2007).
The nascent subduction hypothesis makes sense for the Fran-
ciscan Complex because the blocks are coeval with a major pulse of
Middle Jurassic magmatic activity along the western North Amer-
ican continental margin (Evernden and Kistler, 1970; Saleeby and
Sharp, 1980; Stern et al., 1981; Chen and Moore, 1982; Wright
and Fahan, 1988; Barton et al., 1988; Staude and Barton, 2001;
Irwin, 2002). The Middle Jurassic orogenesis included formation
of the Coast Range ophiolite (Hopson et al., 1981; Shervais et
al., 2005), which is considered by some (e.g., Stern and Bloomer,
1992) to be a product of the initiation of Middle Jurassic subduc-
tion (see also Dickinson et al., 1996). The equivalence in age of
the Coast Range ophiolite and the high-grade blocks of the Fran-
ciscan Complex corresponds well with a common relationship
exhibited by many of the world’s ophiolites: namely, age equiva-
lence of the ophiolite and its metamorphic sole (Jamieson, 1986;
Hacker, 1990a, b; Wakabayashi and Dilek, 2000).
Whereas establishment of a new subduction regime dur-
ing the Middle Jurassic may generally explain the occurrence of
high-grade rocks of this age within the Franciscan Complex of
central and northern California, an equivalent event at ca. 115 Ma
off the southern California margin makes little sense with respect
to the convergent margin evolution of southwest North America.
Emplacement of the 750-km-long Peninsular Ranges batho-
lith had begun by Middle Jurassic time (Shaw et al., 2003) and
was well under way prior to the proposed subduction-initiation
event at 115 Ma (Silver and Chappell, 1988; Kistler et al., 2003).
Furthermore, no middle Cretaceous ophiolite exists within the
southern California region. Only fragments of older forearc base-
ment are preserved. The latter is best represented by Triassic and
Middle Jurassic ophiolitic rocks that border the southern part of
the Peninsular Ranges batholith (Kimbrough and Moore, 2003).
Just as in the case of the Franciscan Complex, the Middle Juras-
sic ophiolite of west-central Baja California is associated with
high-PT blocks of equivalent age on the Vizcaino Peninsula and
Cedros Island (Fig. 2; Baldwin and Harrison, 1989, 1992).
This paper presents new detrital zircon U-Pb results from
metagraywackes of the major metamorphic units of the Cata-
lina Schist that clarify its accretion history. We confirm a genetic
relationship to forearc and batholith rocks of the adjacent Penin-
sular Ranges batholith. These new results indicate that the anom-
alously high-T, high-PT amphibolite facies rocks of the Catalina
Schist had been deposited, accreted, and metamorphosed at peak
grade conditions ~15–20 m.y. before the lawsonite-blueschist and
lower-temperature, high-PT rocks of the complex were accreted.
We conclude that amphibolite facies, epidote-amphibolite, and
possibly epidote blueschist units of the Catalina Schist formed
from forearc strata and basement rocks that were underthrust
and sheared together with lithospheric mantle beneath the west-
ern margin of the Peninsular Ranges batholith in a subduc-
tion erosion process. Our model is consistent with recent deep
seismic-reflection imaging of subparallel megathrusts within
modern subduction zones (e.g., Calvert, 2004). Continued sub-
duction erosion ultimately juxtaposed the high-grade rocks of the
Catalina Schist with lawsonite-blueschists and lower grade rocks
that had formed within a subduction zone setting.
BACKGROUND
Catalina Schist High-Pressure/Temperature Complex
High-PT rocks distributed along the western margin of
southern and Baja California appear to represent the southern
continuation of the better exposed Franciscan Complex of cen-
tral and northern California (Fig. 1; Woodford, 1924; Suppe and
Armstrong, 1972; Kilmer, 1979; Blake et al., 1984). Within the
southern California region, on-land exposures of a high-PT com-
plex referred to as the Catalina Schist occur on Santa Catalina
Island (Bailey, 1941; Platt, 1976) and the Palos Verdes Penin-
sula (Woodring et al., 1946; Dibblee, 1999). More widespread
submarine exposures of the Catalina Schist occur throughout the
inner southern California borderland, and equivalent basement is
detected in boreholes within the western and southwestern Los
Angeles Basin (Schoellhamer and Woodford, 1951; Yerkes et al.,
1965; Yeats, 1968, 1973; Sorensen, 1985, 1988a, 1988b; Wright,
1991; Crouch and Suppe, 1993; Bohannon and Geist, 1998; ten
Brink et al., 2000).
The Catalina Schist on Santa Catalina Island has been
described by Woodford (1924), Bailey (1941), Platt (1975, 1976),
Sorensen and Barton (1987), Sorensen (1986, 1988a, b), Sorensen
and Grossman (1989), Bebout and Barton (1989, 1993, 2002),
and Grove and Bebout (1995) and is only briefly summarized
here. It consists of metasedimentary, metavolcanic, and ultramafic
protoliths that were metamorphosed and sheared together under
amphibolite facies to lawsonite-blueschist facies and lower grade
conditions during the middle Cretaceous (Suppe and Armstrong,
1972; Platt, 1976; Mattinson, 1986; Sorensen and Barton, 1987;
Sorensen, 1988a, 1988b; Grove and Bebout, 1995). Individual
tectonic slices (Fig. 1) contain rocks of broadly equivalent meta-
morphic grade. These are juxtaposed across low-angle faults in an
apparently inverted metamorphic sequence (Platt, 1976; Fig. 1).
The structurally highest unit is an amphibolite facies shear zone
composed primarily of intercalated and metasomatically altered
Figure 2. Geologic map of the Peninsular Ranges batholith, modified after Gastil et al. (1975) and Kimbrough et al.
(2001). Generalized stratigraphic relationships of the forearc region in the northern Vizcaino Peninsula modified after
Kimbrough et al. (2001), Kimbrough and Moore (2003), and references cited within these papers. Simplified stratigraphy
of the northern Santa Ana Mountains is based upon references provided within Lovera et al. (1999). Western limit of the
forearc is the restored pre–middle Miocene position based upon Bohannon and Parsons’s (1995) reconstruction and our
interpretation of aeromagnetic data presented by Langenheim and Jachens (2003). PRB—Peninsular Ranges batholith;
TTG—tonalite, trondjhemite, granodiorite.
340 Grove et al.
mafic and former harzburgite/dunite protoliths (Sorensen and
Barton, 1987; Sorensen, 1988b; Bebout and Barton, 1993, 2002).
The proportion of sediment, particularly immature graywacke,
increases structurally downward. At the structurally lowest levels
the lawsonite-blueschist and lower grade units are sediment dom-
inated. Each of the major metamorphic units is characterized by
meter- to kilometer-scale, compositionally heterogeneous shear
zones that appear to have facilitated metasomatic fluid infiltration
at near peak-grade conditions (Bebout and Barton, 1989, 1993,
2002; King et al., 2006, 2007).
Primary geologic mapping of the Catalina Schist on Santa
Catalina Island was performed by Bailey (1941) and Platt (1976).
Although we rely heavily upon Platt’s (1976) mapping of the
central part of the island, we recognize different tectonometa-
morphic units than he did. Platt’s (1976) Catalina greenschist
unit has been subdivided into epidote-amphibolite and epidote-
blueschist units (Grove and Bebout, 1995). Results presented
in Grove and Bebout (1995) and in this paper indicate that the
epidote-amphibolite and epidote-blueschist rocks had impor-
tant differences in accretion and metamorphic history. In addi-
tion, we have delineated lawsonite-albite and actinolite-albite
facies units on the west end of the island. These lower grade
rocks appear to structurally underlie a lawsonite-blueschist unit
(Altheim et al., 1997).
Because the Catalina Schist terrane was significantly
extended as it was exhumed during the early middle Miocene
formation of the inner continental borderland (Wright, 1991;
Crouch and Suppe, 1993; Nicholson et al., 1994; Bohannon and
Geist, 1998; Ingersoll and Rumelhart, 1999), original contacts
bounding the major units within the schist are likely to have been
strongly modified or excised altogether by middle Tertiary struc-
tures. In addition, the extent to which the Santa Catalina Island
exposures of Catalina Schist represent the much larger area of
high-PT rocks that underlie the inner continental borderland is
difficult to assess. Study of widely distributed deposits of the San
Onofre Breccia indicates that the Santa Catalina Island exposures
are probably broadly representative of the overall terrane (Stu-
art, 1979). The San Onofre Breccia is an early middle Miocene
deposit that accumulated as the Catalina Schist was exhumed
(Wright, 1991; Crouch and Suppe, 1993). Clast counts performed
by Stuart (1979) appear to be broadly consistent with an unroof-
ing sequence in which amphibolite and epidote-amphibolite
grade Catalina Schist and low-PT mafic basement from the hang-
ing wall were much more abundant at lower stratigraphic levels
within the San Onofre Breccia. Our own observations confirm
this relationship for the thick deposits of this breccia that accumu-
lated along the western margin of the Peninsular Ranges batho-
lith (Fig. 1). In contrast, the stratigraphically younger San Onofre
Breccia, including deposits laid down atop extended borderland
crust underlain by the Catalina Schist, tends to be dominated by
lawsonite-blueschist and lower-grade detritus. Based upon these
relationships, we are reasonably confident that our sampling of
the Catalina Schist on Santa Catalina Island is broadly represen-
tative of the Catalina Schist terrane as a whole.
Forearc Basement and Strata
The relatively complete forearc sequence of the Vizcaino
Peninsula at the southern end of the Peninsular Ranges batholith
provides the basis to interpret the forearc of the highly disrupted
southern California borderland (Figs. 1, 2). On the Vizcaino Pen-
insula and on Cedros Island, both Late Triassic (221 Ma) and Mid-
dle Jurassic (173 Ma) ophiolite sequences (Moore, 1985; Kim-
brough, 1985; Kimbrough and Moore, 2003) are the basement for
Upper Jurassic, Lower Cretaceous, Upper Cretaceous, and lower
Cenozoic arc volcanic and forearc sedimentary rocks (Boles,
1978; Kilmer, 1979; Kienast and Rangin, 1982; Boles and Lan-
dis, 1984; Patterson, 1984; Smith and Busby, 1993; Busby et al.,
1998; Kimbrough et al., 2001; Critelli et al., 2002; Busby, 2004).
High-PT Cretaceous rocks of the Western Baja terrane constitute
an accretionary complex that is juxtaposed beneath the Triassic
and Jurassic ophiolitic rocks well outboard of the batholithic mar-
gin (Suppe and Armstrong, 1972; Moore, 1986; Sedlock, 1988a,
b; Baldwin and Harrison, 1989; Baldwin and Harrison, 1992).
Only fragmentary evidence exists for the early Mesozoic
forearc basement within the southern California area. Within
the outer California continental borderland, offshore drilling
and seismic exploration detect mafic basement beneath forearc
strata (Bohannon and Geist, 1998; ten Brink et al., 2000). Sur-
face exposures of lower Mesozoic basement rocks that occur on
Santa Cruz Island (the Willows Complex of Weaver and Nolf,
1969; Hill, 1976; Mattinson and Hill, 1976) are correlated with
the Coast Range ophiolite (Jones et al., 1976). The associated
Santa Cruz Island Schist and the correlative Santa Monica slate
of the western Transverse Ranges exhibit distinctly arclike com-
positions and could be equivalent to accreted Jurassic terranes of
the Sierran Foothills (Sorensen, 1985, 1988a). “Saussurite” gab-
bro similar to altered zones of the Willows Complex gabbros is
in fault contact with the Catalina Schist on Santa Catalina Island
(Platt, 1976) and within the subsurface of the southwestern Los
Angeles Basin (Schoellhamer and Woodford, 1951; Yeats, 1973;
Sorensen, 1985, 1988b). Saussurite gabbro clasts are a major
component within the San Onofre Breccia (Stuart, 1979) and
confirm that low-PT mafic basement was exhumed along with
the Catalina Schist during middle Miocene borderland rifting.
Cenomanian and younger forearc strata onlap the west-
ern margin of the northern Peninsular Ranges batholith (Fig. 2;
Woodring and Popenoe, 1942; Yerkes et al., 1965; Flynn, 1970;
Nordstrom, 1970; Peterson and Nordstrom, 1970; Kennedy and
Moore, 1971; Sundberg and Cooper, 1978; Schoellhamer et al.,
1981; Nilsen and Abbott, 1981; Bottjer et al., 1982; Bottjer and
Link, 1984; Fry et al., 1985; Girty, 1987; Bannon et al., 1989).
Thick sections of Upper Cretaceous strata also occur throughout
the outer borderland (Howell and Vedder, 1981; Vedder, 1987;
Bohannon and Geist, 1998). The existence of Lower Cretaceous
forearc strata is far less certain, however. Sedimentary rocks of this
age have not been described anywhere along the western margin
of the northwestern Peninsular Ranges, the Santa Monica Moun-
tains, or on the offshore islands. Their existence in the subsurface
 The Catalina Schist
is mostly inferred in all areas that have been drilled (Howell and
Vedder, 1981; Vedder, 1987; Bohannon and Geist, 1998; ten Brink
et al., 2000). Our model for the formation of the Catalina Schist
explicitly accounts for the scarcity of these Lower Cretaceous
forearc rocks in the southern California region (see Discussion).
Peninsular Ranges Batholith
The Peninsular Ranges batholith of southern and Baja Cali-
fornia constitutes a classic Cordilleran continental margin batho-
lith (Fig. 2; Larsen, 1948; Jahns, 1954; Gastil et al., 1975). The
better studied northern segment of this batholith consists of lon-
gitudinal western and eastern zones based on age, petrology, pre-
batholithic wall rock, geophysical parameters, and depth and style
of emplacement (Gastil et al., 1981; Baird and Miesch, 1984; Tay-
lor, 1986; Gromet and Silver, 1987; Silver and Chappell, 1988;
Ague and Brimhall, 1988; Todd et al., 1988; Hill and Silver, 1988;
Gastil, 1993; Johnson et al., 1999; Lovera et al., 1999; Todd et al.,
2003; Kistler et al., 2003; Langenheim and Jachens, 2003). The
oldest recognized intrusive rocks are gneissic S-type plutons that
occupy the medial zone of the Peninsular Ranges batholith and
yield Middle Jurassic emplacement ages (Todd and Shaw, 1985;
Thomson and Girty, 1994; Shaw et al., 2003; Kistler et al., 2003).
Cretaceous plutons as old as 140 Ma also occur within this batho-
lith (Silver and Chappell, 1988; Alsleben et al., 2005; D.L. Kim-
brough, personal observ.). These are cut by a regionally extensive
dike swarm emplaced at ca. 130–120 Ma (Böhnel et al., 2002;
D.L. Kimbrough, personal observ.). The well-developed west-
ern zone of the Peninsular Ranges batholith is composed mainly
of 125–100 Ma gabbro to monzogranite plutons with primitive
island arc geochemical affinities. The eastern zone of the batho-
lith is defined by a belt of large-volume, 95 ± 3 Ma tonalite,
trondjhemite, and low-K granodiorite (TTG) plutons (Gastil et
al., 1975; Silver and Chappell, 1988; Walawender et al., 1990)
that constitute the La Posta TTG suite (Tulloch and Kimbrough,
2003; Kimbrough and Grove, 2006). These Na- and Al-rich plu-
tons have a deep, garnet-present, melt-source signature, as seen
in high Sr, Ba, Sr/Y, and La/Yb. Field and thermobarometric data
indicate emplacement at ~2–6 kbar pressures (Rothstein and
Manning, 2003) followed by rapid Cenomanian–Turonian uplift
and denudation at rates of ~1–2 mm/yr (Krummenacher et al.,
1975; Lovera et al., 1999; Johnson et al., 1999; Kimbrough et al.,
2001; Ortega Rivera, 2003; Grove et al., 2003b). A delayed late
Campanian–Maastrichtian phase of uplift (e.g., Krummenacher
et al., 1975) appears to be related to Laramide shallow subduction
and removal of the deep crustal and lithospheric roots of the La
Posta belt (Lovera et al., 1999; Grove et al., 2003a, b).
SAMPLING AND METHODS
Detrital Zircon U-Pb Age Measurements
A total of 645 U-Pb zircon ages were measured from 33 sam-
ples of amphibolite through sub-blueschist facies metagraywacke
341
of the Catalina Schist collected from Santa Catalina Island. Many
of the samples examined were previously studied by Grove and
Bebout (1995). We focused upon metasedimentary rocks from
each of the major metamorphic units in order to obtain an upper
bound upon the depositional age of their sedimentary protolith
and to determine sediment provenance. In our U-Pb analysis of
zircons from the Catalina Schist, we employed secondary ion-
ization mass spectrometry (SIMS) methods using the UCLA
Cameca ims 1270 ion microprobe. The extensive metamorphic
recrystallization that affected zircons from the amphibolite and
epidote-amphibolite units was significantly ameliorated by the
high spatial resolution of the ion microprobe in conjunction with
cathodoluminescence (CL) imagery (e.g., only ~1 nanogram of
sputtered zircon required to yield a U-Pb age). The techniques are
described in Grove et al. (2003b) with additional details included
in the GSA Data Repository.1
We analyzed conventionally sectioned and polished zircons
in epoxy mounts. The zircons were hand selected from heavy
mineral concentrates produced from standard crushing, density,
and magnetic methods. Further details are included in the Data
Repository. Because the number of zircons measured from indi-
vidual samples was typically small owing to the low yields real-
ized during mineral separation, we pooled data from multiple
samples to obtain statistically meaningful results for each of the
major tectonic units of the Catalina Schist.
As noted above, metamorphic zircon growth was a signifi-
cant issue for some of the grains we examined, particularly for
the metasediments from the amphibolite unit. Anczkiewicz et al.
(2004) also described evidence for metamorphic zircon growth
in reconnaissance U-Pb zircon measurements they reported from
the Catalina Schist amphibolite unit. Although we selected our
analysis sites on the basis of morphologic and optical criteria
in CL imagery, a significant number of the spot analyses from
zircons of the amphibolite and epidote-amphibolite units over-
lapped regions affected by metamorphic recrystallization. Meta-
morphic overgrowths on igneous zircons generally have Th/U
<0.1; Kröner et al., 1994; Rubatto, 2002; Williams and Claesson,
1987). We have found a cutoff of Th/U = 0.1 to be empirically
well supported by detrital zircon results from the Cenomanian and
younger, pluton-derived strata that overlie the Peninsular Ranges
batholith. In our sampling of these units (Mahoney et al., 2005),
1506 out of 1527, or 98.6% of the zircon U-Pb analyses, yielded
Th/U values equal to or greater than 0.1 (Fig. 3A). The average
Th/U value measured was 0.48 ± 0.41, with 3.3% of the analyses
falling within the Th/U = 0.1–0.2 range. Thus a cutoff of Th/U
= 0.1 is useful for identifying detrital igneous zircons adversely
affected by metamorphic recrystallization. Accordingly, we have
excluded all analyses with Th/U <0.1 from the summary plots or
calculations presented below (Fig. 3B).
Complete data tables of U-Pb age measurements, litho-
logic descriptions, and sample locations are available from the
1
GSA Data Repository item 2008105 is available online at www.geosociety.
org/pubs/ft2008.htm, or on request from editing@geosociety.org or Documents
Secretary, GSA, P.O. Box 9140, Boulder, CO 80301, USA.
342 Grove et al.
Figure 3. (A) Measured zircon U-Pb age versus Th/U of Cretaceous
forearc sedimentary rocks of the Peninsular Ranges batholith. Data,
from Mahoney et al. (2005), demonstrate that virtually all detrital igne-
ous zircons derived from the Peninsular Ranges have Th/U >0.1 (see
also Williams and Claesson, 1987). (B) Equivalent plot for zircon re-
sults from the Catalina Schist amphibolite unit. Based upon cathodo-
luminescence imaging and other criteria, we regard a Th/U = 0.1 as a
meaningful cutoff to distinguish analyses in which the sputter pit over-
lapped metamorphic zircon growth (open symbols). The filled sym-
bols represent detrital grains that we consider to be largely unaffected
by metamorphic zircon growth (see also Kröner et al., 1994; Rubatto,
2002). Zircon results with Th/U <0.1 have been excluded from all sub-
sequent plots and calculations.
GSA Data Repository. In this paper, quoted uncertainties are
±1σ errors unless otherwise specified. The U-Pb ages are gener-
ally 206Pb/238U values for <1 Ga zircons and 207Pb/206Pb ages for
older grains. Analyses with age uncertainties >10% have been
discarded. The U-Pb ages of zircons that exhibited resolvable
206
Pb/238U versus 207Pb/235U discordance are 207Pb/206Pb values.
Rb-Sr Measurements
To clarify the significance of previous 40Ar/39Ar phengite
step-heating results (Grove and Bebout, 1995) that had indicated a
possible Middle Jurassic age for garnet-bearing blueschist blocks
within the lawsonite-blueschist mélange, we carried out Rb-Sr
measurements with one of the blocks (330–4B) that had been
analyzed in this previous study. All column chemistry and mass
spectrometer measurements were carried out at UCLA. There,
Rb and Sr were separated in cation exchange columns containing
AG50W-X8 resin, using 2.5N HCl. The Sr isotopic compositions
were measured with a VG54–30 multicollector thermal ionization
mass spectrometer. 87Sr/86Sr ratios were normalized to 86Sr/88Sr
= 0.1194. In this analysis session, the measured 87Sr/86Sr value
for our Sr standard (NBS 987) value was 87Sr/86Sr = 0.710239
± 16 (2σ, n = 13). The Rb and Sr concentrations were measured
by isotope dilution using VG54–30. Model Rb-Sr isochron ages
were calculated using ISOPLOT/Ex Version 3 (Ludwig, 2003).
RESULTS
Amphibolite Facies
Five biotite + muscovite + garnet ± kyanite-bearing meta-
graywackes from the amphibolite unit yielded abundant subhe-
dral to well-rounded zircon. Alternatively, both metachert samples
examined contained only metamorphic zircon. Although exter-
nal overgrowths <1–5 µm or more were common, zircons also
exhibited internal areas of patchy diffuse recrystallization in CL
imagery. Of the 169 analyses, 48 (28%) overlapped metamorphic
zircon based upon Th/U (Fig. 3B). About half of the affected anal-
yses yielded obviously mixed ages. U-Pb ages calculated for the
remaining analyses with Th/U <0.1 fell between 107 and 126 Ma,
with a peak at 116 ± 6 Ma. This result is in good agreement with
independent estimates for the timing of amphibolite unit recrys-
tallization (ca. 115 Ma; Suppe and Armstrong, 1972; Mattinson,
1986; Grove and Bebout, 1995; Anczkiewicz et al., 2004).
The data for zircons with Th/U >0.1 clearly indicate that the
sedimentary protolith of the amphibolite facies metagraywackes
contained a large proportion of craton-derived detritus (Fig. 4A).
Roughly 50% of the analyses were from Middle Proterozoic
zircon (Fig. 4G). Unfortunately, many of these grains exhibited
variable Pb loss and high degrees of discordance. This condition
complicates characterization of the Middle Proterozoic age dis-
tribution. Although 207Pb/206Pb ages more accurately approximate
the crystallization age than the highly discordant U-Pb ages, they
Figure 4. Relative probability plots of detrital zircon U-Pb age distributions from major units of the Catalina Schist
on Catalina Island. (A) Amphibolite. (B) Epidote-amphibolite. (C) Epidote-blueschist. (D) Lawsonite-blueschist.
(E) Actinolite-albite. (F) Lawsonite-albite. Note that we use a split horizontal axis at 300 Ma and that relative probability
plots between 300 and 3000 Ma have a 2× scaling factor to improve resolution of the overall age distribution. (G) True
scale cumulative probability spectra for all units.
344 Grove et al.
are insufficiently precise to reveal strong clustering of ages. Nev-
ertheless, the subdued maxima that occur at 1.15, 1.40, and 1.65–
1.80 Ga are compatible with a southwestern North American
provenance. Age peaks also occur at 126, 145, and 162 Ma. The
five youngest zircons measured from the amphibolite unit meta-
graywackes yielded a weighted mean age of 122 ± 3 Ma. These
analyses yielded a mean Th/U value of 0.30 ± 0.20 and were from
oscillatory zoned regions in Cl− imagery that likely reflect igne-
ous crystallization. Accordingly, we regard 122 ± 3 Ma as a geo-
logically meaningful upper bound upon the depositional age of
the sedimentary protolith.
Epidote-Amphibolite Facies
Five muscovite ± biotite ± garnet epidote-amphibolite facies
metagraywackes yielded zircon that was generally more subhe-
dral to euhedral grains than the well-rounded grains common
in the amphibolite unit metagraywackes. CL imaging and Th/U
indicate that metamorphic recrystallization of detrital grains was
less common than in the amphibolite unit. Only 13 of 135 or
10% of analyses of the zircons from epidote-amphibolite facies
metagraywackes had Th/U <0.1. The U-Pb age distribution indi-
cates a diminished cratonal provenance relative to amphibolite
facies metagraywackes; only ~20% of the zircons yielded Middle
Proterozoic ages (Fig. 4G). Although fewer analyses are avail-
able, the overall age distribution of Middle Proterozoic zircons
in epidote-amphibolite metagraywackes is broadly similar to that
of the amphibolite facies rocks (Fig. 4B). The relative lack of
Proterozoic grains is accompanied by a proportionate increase of
<200 Ma grains. Distinct peaks are present at 115 and 126 Ma,
with a strong peak at 150 Ma (Fig. 4B). The five youngest detrital
zircons from the epidote-amphibolite metagraywackes yielded
an age of 113 ± 3 Ma, with a mean Th/U value of 0.42 ± 0.05.
Epidote-Blueschist Facies
Thirty analyzable zircons were recovered from a single
metagraywacke sample intercalated with Na-amphibole + clino-
zoisite + albite-bearing mafic rocks (glaucophanic greenschists
of Platt, 1975; Sorensen, 1986). Six similar-appearing samples
from equivalent field settings failed to yield zircon. This poor
zircon yield probably reflects abundant volcanic detritus in the
protolith. Nearly all of the zircons recovered consisted of clear
subhedral to euhedral grains. There was no evidence for meta-
morphic zircon growth either petrographically or in terms of
measured Th/U. Two distinct maxima occur at 103 and 142 Ma
(Fig. 4C). Although results from the epidote-blueschist facies
unit are transitional between those obtained from the amphibo-
lite and epidote-amphibolite facies units and lower grade parts of
the Catalina Schist (Fig. 4G), the epidote-blueschist facies unit
appears more closely allied with the lower grade units, based
upon the age of the youngest zircons detected in each population.
The five youngest zircons measured yielded an average U-Pb age
of 101 ± 3 Ma with a mean Th/U of 0.63 ± 0.23.
Lawsonite-Blueschist Facies and Lower Grade Rocks
Data bearing upon the sedimentary protoliths of the
lawsonite-blueschist and lawsonite-albite, and albite-actinolite
facies units are discussed together because of the highly similar
results produced (Fig. 4D–G). Overall, we obtained 164 analyses
from 11 lawsonite-blueschist facies rocks, 59 analyses from 3
albite-actinolite facies rocks, and 82 analyses from 7 lawsonite-
albite facies metagraywackes. No metamorphic zircon growth
was detected; Th/U values were all >0.1. The Ca-rich metagray-
wackes prevalent in low-grade parts of the Catalina Schist over-
whelmingly contain clear subhedral to euhedral grains. Most of
these grains yield U-Pb ages of 95–130 Ma (Fig. 4D–F). Late
Jurassic–Early Cretaceous zircon was greatly subordinate, and
only trace quantities of Proterozoic zircon were present. Because
the youngest zircon U-Pb ages measured from the three lowest
grade units all agreed within error (96 ± 3 Ma, 97 ± 3 Ma, and 98
± 3 Ma for lawsonite-blueschist, actinolite-albite, and lawsonite-
albite, respectively), they define a 97 ± 3 Ma upper bound for the
depositional age of the sedimentary protolith for all of the low-
grade Catalina Schist units.
Garnet-Bearing Blueschist Block in Lawsonite-Blueschist
Mélange
Garnet-bearing blueschist blocks occur within mélange
zones in the lawsonite-blueschist unit. Phengite 40Ar/39Ar results
for two such blocks indicate that they formed prior to peak-grade
recrystallization in the amphibolite unit (Grove and Bebout,
1995). To confirm this relationship we undertook Rb-Sr mea-
surements of phengite, Na-amphibole, and whole rock for one of
the samples (330–4B; Fig. 5A). Results are shown in Figure 5B.
Phengite, Na-amphibole, and whole-rock results do not define
a statistically meaningful isochron. However, the model Rb-Sr
age defined by phengite + Na-amphibole, 135 Ma, is identical
to the total gas age calculated from 40Ar/39Ar step-heating results
obtained from 330 to 4B phengite.
DISCUSSION
Evaluation of the Nascent Subduction Model for the
Catalina Schist
The origin of the amphibolite facies rocks of the Catalina
Schist is central to assessing the tectonic significance of this ter-
rane. Amphibolite facies metamorphism and anatexis are atypi-
cal features of subduction complexes and are thought to reflect
unusual processes along a convergent margin. High-temperature
conditions accompany the earliest developmental stages of sub-
duction before the hanging wall is refrigerated (Platt, 1975;
Cloos, 1985; Peacock, 1987). Although ridge subduction and/or
slow underflow of very young oceanic crust (e.g., Peacock, 1987,
1992; Hacker, 1990a, 1990b; Peacock et al., 1994) can also pro-
duce high-temperature metamorphism in subduction zones, there
 The Catalina Schist
is no evidence that either process affected the middle Cretaceous
southwestern margin of North America.
The detrital zircon results (Fig. 6) require substantial modi-
fication of Platt’s (1975) hypothesis that amphibolite through
blueschist facies rocks formed in an inverted metamorphic aure-
ole in response to transient heating during nascent subduction at
ca. 115 Ma. Sequential accretion of the major tectonic units of the
345
Figure 5. (A) Garnet-bearing blueschist
block in mélange matrix of the lawson-
ite blueschist unit. (B) Rb-Sr results for
phengite, Na-amphibole, and whole
rock. A statistically meaningful isochron
is not defined. Phengite and Na-amphi-
bole define a model isochron (135 Ma)
that is identical to the total gas 40Ar/39Ar
age yielded by this sample.
Catalina Schist over a ~20 m.y. interval is required. As indicated in
Figure 6A, the protolith of the amphibolite unit metagraywackes
was deposited between 122 Ma (youngest detrital zircon ages)
and 115 Ma (the time of peak-grade recrystallization). In contrast,
the sedimentary protolith of the lawsonite-blueschist and lower
grade units was deposited after 97 Ma or at least 18 m.y. after
peak-grade recrystallization of the amphibolite unit (Fig. 6C, D).
346 Grove et al.

Figure 6. Estimated temperature-time histories for major units of the Catalina Schist. (A) Amphibolite. (B) Epidote-
amphibolite. (C) Epidote-blueschist. (D) Lawsonite-blueschist and lower grade units of the Catalina Schist. Yellow circles
represent measured detrital zircon U-Pb ages. Maximum bounds upon the depositional age of the sedimentary protolith
represent average and standard deviation of U-Pb ages of the five youngest zircons with Th/U >0.1 (15 youngest zircons
for composite of the three lowest grade units). Data sources for other thermochronometry include Suppe and Armstrong
(1972), Mattinson (1986), Grove and Bebout (1995), and Anczkiewicz et al. (2004). Note that we have varied the bulk
closure temperatures assigned to micas from 400 °C for coarse-grained muscovites within the high-grade units to 350 °C
for finer grained, less retentive white mica within lawsonite-albite rocks.

Available data for the Catalina Schist amphibolite unit
indicate that it cooled slowly (ca. 25 °C/m.y.) from peak-grade
conditions (650–700 °C) at 115 Ma to K-Ar muscovite closure
temperatures (350–400 °C) at 105–100 Ma (Fig. 6A; Grove
and Bebout, 1995). The Catalina epidote-amphibolite unit also
cooled slowly (Fig. 6B): a 7–12 m.y. interval separated peak-
grade recrystallization of epidote-amphibolite mafic gneisses and
cooling through phengite K-Ar closure. Slow cooling exhibited
by the amphibolite unit is compatible with very slow subduction
or subduction of very young oceanic lithosphere. However, envi-
ronments such as these are far too warm to permit formation of
lawsonite-blueschist and lower temperature metamorphic rocks.
Within the Catalina Schist, only the lawsonite-blueschist and
lower grade rocks record metamorphic conditions that require
formation within a low-T, high-PT subduction zone. The amphib-
olite and epidote-amphibolite units were retrograded under
greenschist facies conditions, as was the epidote-blueschist unit
(Platt, 1976; Sorensen, 1986; Bebout and Barton, 1989; Grove
and Bebout, 1995). Given their present association with lawson-
ite-blueschist and lower grade rocks, we find it remarkable that
the amphibolite and epidote-amphibolite units exhibit such little
evidence for equilibration at blueschist facies conditions (Platt,
1976; Sorensen, 1986; Bebout and Barton, 1989; Grove and
Bebout, 1995). Blueschist facies overprinting of higher grade
 The Catalina Schist 347

rocks is the rule for subduction complexes of the North American

Cordillera (Ernst, 1988). For example, tectonic blocks of garnet
amphibolite in the Franciscan and Shuksan subduction com-
plexes are heavily overprinted by blueschist facies assemblages
(e.g., Brown et al., 1982; Cloos, 1985; Wakabayashi, 1990). In
contrast, evidence for blueschist facies overprinting within the
Catalina Schist amphibolite unit is obscure and limited to sparse
veining by pumpellyite and trace amounts of lawsonite in white
mica + zoisite + albite assemblages that replace oligoclase in
amphibolite facies rocks (Grove and Bebout, 1995). We interpret
the available thermochronology and greenschist facies overprint-
ing of the high-grade units of the Catalina Schist to indicate that
they occupied a subcrustal position characterized by 400–500 °C
temperatures prior to 100–95 Ma. The lack of blueschist facies
overprinting within the amphibolite and epidote-amphibolite
units indicates to us that these rocks were sufficiently distant
from an active subduction zone to be largely unaffected by the
low geothermal gradient (<5–10 °C/km) recorded by the Fran-
ciscan and other Cordilleran subduction complexes during the
middle Cretaceous.

Provenance Ties Linking the Catalina Schist to the

Peninsular Ranges Batholith

King et al. (2007) measured whole-rock Pb isotopes from

mélange matrix sampled from the amphibolite, lawsonite-
blueschist, and lawsonite-albite units of the Catalina Schist.
Although details of the U-Th-Pb systematics indicate the pos-
sible elemental redistributions during the devolatization history
of the schist (see King et al., 2007), it is remarkable how well the
measured Pb isotopic compositions of the sediment-dominated
lawsonite-blueschist and lawsonite-albite mélange matrix agree
with those measured from granitoids of the adjacent Peninsular
Ranges batholith (Fig. 7A, B; Kistler et al., 2003). The whole-
rock Pb isotopic compositions from the mafic- and ultramafic-
dominated mélange of the Catalina Schist amphibolite unit also
overlap strongly with the Peninsular Ranges batholith (Fig. 7C)
but tend to more radiogenic values than the lower grade units
(Fig. 7A, B). The amphibolite unit mélange results are more dif-
ficult to interpret than those from the lower grade units, because
the sources of Pb are far less uncertain owing to the scarcity of
sedimentary material within the mélange (see Bebout and Bar-
ton, 1993, 2002). Nevertheless, these results are consistent with
a cratonal provenance similar to that exhibited by southeastern

Figure 7. Whole-rock Pb isotopic data of King et al. (2007) for:

(A) Lawsonite-albite. (B) Lawsonite-blueschist. (C) Amphibolite
unit mélange matrix of the Catalina Schist. Also shown are whole-
rock Pb isotopic data for northern Peninsular Ranges batholith plutons
(red circles; Kistler et al., 2003) and equivalent data for southeastern
Arizona (orange circles; Wooden and Miller, 1990). The yellow field
represents the generalized distribution of whole-rock Pb yielded by the
Mojave–Transverse Ranges province (based upon Barth et al., 1995,
and personal observ. from J. Wooden, D. Coleman, and A. Barth).
348 Grove et al.
Arizona (Wooden and Miller, 1990) and can be differentiated
from the Mojave–Transverse Ranges province (Fig. 7C; Barth et
al., 1995; J. Wooden, D. Coleman, and A. Barth, personal observ.).
Because the early Mesozoic wall rocks of the Peninsular Ranges
batholith are dominated by craton-derived Proterozoic zircon
that indicates a similar source (see below), we speculate that the
Proterozoic zircon contributed from the cratonally derived sedi-
ment eroded from the batholith’s wall rocks recrystallized dur-
ing mélange formation, liberating the highly radiogenic Pb that
was dispersed throughout the amphibolite unit mélange as it was
recrystallized and metasomatically altered (Bebout and Barton,
1993, 2002; King et al., 2007).
In the same manner as the Pb isotope data discussed above,
the detrital zircon results from the Catalina Schist also indicate
a close genetic relationship between the Catalina Schist and the
Peninsular Ranges batholith: (1) early Mesozoic, Paleozoic, and
Proterozoic detrital zircon from the two terranes are similar; and
(2) middle Cretaceous detrital zircon age distributions from the
lawsonite-blueschist and lower grade Catalina Schist are virtu-
ally identical to the distribution of crystallization ages from the
northern Peninsular Ranges batholith and the detrital age distri-
bution from the adjacent Upper Cretaceous forearc rocks.
Lower Mesozoic flysch wall rocks of the Peninsular Ranges
batholith crop out within the deeply denuded axial zone of the
northern batholith (Fig. 2) and represent a major potential source
of early Mesozoic, Paleozoic, and Proterozoic zircon. A com-
posite detrital zircon distribution based upon results obtained by
Morgan et al. (2005) from three different localities in southern
California (Bedford Canyon Formation, French Valley area, and
Julian Schist; localities 1–3 in Fig. 2) and northern Baja Califor-
nia (Vallecitos area; locality 4 in Fig. 2) is shown in Figure 8.
As indicated, the age distribution defined by >200 Ma detrital
zircon from the Catalina Schist (Fig. 8A) is consistent with a
source region similar to the Peninsular Ranges batholith wall
rocks (Fig. 8B). The expected age distribution for detritus that
originated from the southwest North American craton is repre-
sented by late Miocene to Holocene sands from the Colorado
River system (Fig. 8C). Well-defined peaks at 1.0–1.2, 1.45,
and 1.65–1.75 Ga are distinctive of the basement assemblage
(Gehrels and Stewart, 1998). Early Paleozoic and latest Neopro-
terozoic (400–650 Ma) zircon was contributed by supracrustal
cover rocks that had their source in the Appalachian and Oua-
chita orogenic belts (Dickinson and Gehrels, 2003). Finally,
Permian–Triassic zircon is derived from the earliest Phanero-
zoic magmatic arcs established along the southwestern North
American margin (Barth and Wooden, 2006; González-León et
al., 2006).
The lawsonite-blueschist and lower grade rocks of the Cata-
lina Schist that contain Upper Cretaceous forearc strata of the
Peninsular Ranges are very similar (Fig. 8E), as is the distribu-
tion of U-Pb zircon crystallization ages from the adjacent batho-
lith (Fig. 8F; Silver and Chappell, 1988; Walawender et al., 1990;
Kistler et al., 2003; D.L. Kimbrough, personal observ.). The fact
that the Catalina Schist results are skewed to slightly older ages is
readily understood in terms of the Cenomanian age of the proto-
lith. Cenomanian-age forearc strata from the Peninsular Ranges
batholith are also skewed to older ages because they were depos-
ited prior to massive exhumation of the ca. 95 Ma La Posta plu-
tonic suite within the eastern batholith (Mahoney et al., 2005).
Most of the results from the Upper Cretaceous forearc rocks are
Campanian or Maastrichtian and hence are enriched in detritus
from the eastern plutonic zone.
In order to further evaluate the strength of the provenance
tie between the Catalina Schist and the Peninsular Ranges, we
have carried out ternary mixing calculations. We have identified
three distinctive components: (1) lower Mesozoic wall rocks, rep-
resented by the Bedford Canyon, French Valley, Julian Schist,
and Vallecitos flysch localities (Fig. 9A; Morgan et al., 2005);
(2) Early Cretaceous volcanics, represented by results from
Lower Cretaceous sandstones and volcanics within the volcanic
arc (Fig. 9B; Alsleben et al., 2005; D.L. Kimbrough, personal
observ.); and (3) Upper Cretaceous forearc, represented by
Cenomanian–Maastrichtian strata distributed along the western
margin of the Peninsular Ranges batholith (Fig. 9C; Mahoney
et al., 2005). Cumulative probability plots for these three com-
ponents are shown in Figure 9D, along with equivalent curves
representing the major units of the Catalina Schist.
We have linearly mixed these three components to obtain
the best fit to the detrital zircon age distributions from the Cata-
lina Schist. Results for the amphibolite unit, epidote-amphibolite
unit, epidote-blueschist unit, and lawsonite-blueschist and lower-
grade units are shown in Figure 9E through H, respectively. The
relative proportions of the three end members required to pro-
duce these fits are shown in Figure 9I. As indicated in Figure 9I,
the detrital zircon provenance signature of the earliest accreted
material within the amphibolite unit of the Catalina Schist is best
approximated by a 68:32 mixture of sediment derived predomi-
nantly from lower Mesozoic wall rocks and the Early Cretaceous
volcanic arc, respectively. Such a high proportion of cratonally
derived material within the amphibolite unit metasediments
makes sense, given the abundance of Middle Proterozoic zircon,
the aluminous and quartz-rich nature of the protolith, and the fact
that pegmatites, metasediments, and mélange sampled from the
amphibolite unit tend to yield a relatively radiogenic Sr, Nd, and
Pb isotopic signature (i.e., similar to evolved arc crust; Bebout
and Barton, 1993, 2002; King et al., 2006, 2007).
The provenance signature of sediment shed from the Penin-
sular Ranges batholith shifted dramatically between early Aptian
to early Cenomanian time (Mahoney et al., 2005). A parallel
shift is recorded by the Catalina Schist (Fig. 4). In the case of
the epidote-amphibolite unit, the relative proportions of lower
Mesozoic wall rock to Early Cretaceous volcanic arc sediment
diminishes to 44:56 (Fig. 9I). Compositionally, there is still a
resolvable cratonal contribution in the epidote-amphibolite meta-
graywackes. A further shift is exhibited by the epidote-blueschist
metagraywackes. Their age distribution is well modeled by a
20:58:22 mixture of lower Mesozoic wall rock, Lower Creta-
ceous volcanic arc, and Upper Cretaceous forearc sediment,
 The Catalina Schist 349

Figure 8. Relative probability plots (200–2200 Ma) of detrital zircon U-Pb ages. (A) All results from the Catalina Schist.
(B) Representative early Mesozoic flysch wall rocks of the northern Peninsular Ranges batholith (Morgan et al., 2005);
data originate from localities 1–4 in Figure 2. (C) Representation of the southwestern North American detrital zircon
provenance signature as represented by late Miocene–Holocene sediments of the Colorado River system (D. Kimbrough
and M. Grove, personal observ.). Yellow bands denote the primary Middle Proterozoic crystallization age maxima con-
tributed by cratonal basement and supracrustal rocks of southwestern North America. (D) All detrital zircon U-Pb age
results from the lawsonite-blueschist and lower grade units of the Catalina Schist. (E) Representative Late Cretaceous
(K) detrital zircon data from the forearc of the northern Peninsular Ranges batholith (PRB) (Mahoney et al., 2005).
(F) Pluton U-Pb zircon crystallization ages from the northern Peninsular Ranges batholith (Silver and Chappell, 1988;
Walawender et al., 1990; Kistler et al., 2003; D. Kimbrough, personal observ.).

respectively (Fig. 9I). The provenance shift was completed by
the time the protolith of the lawsonite-blueschist and lower grade
metagraywackes was deposited. The detrital zircon age distribu-
tion of the latter is well described by a 49:51 mixture of Lower
Cretaceous volcanic arc and Upper Cretaceous forearc sediment,
respectively (Fig. 9I).
Implications of Pre-115 Ma High-PT Tectonic Blocks
within Low-Grade Catalina Schist
Grove and Bebout (1995) reported 40Ar/39Ar results for two
garnet-bearing, blueschist blocks within the lawsonite-blueschist
unit of the Catalina Schist that indicated that the blocks predated
115 Ma peak-grade metamorphism within the amphibolite unit.
Because the pronounced age gradients yielded by both samples
reached 150–160 Ma at the highest temperatures of gas release,
Grove and Bebout (1995) speculated that the garnet-bearing
blueschist blocks formed during the Middle Jurassic and had the
same tectonic significance as similar high-grade blocks present
within the Franciscan Complex along the margin to the north
(Wakabayashi, and Dumitru, 2007, and references cited therein)
and within the western Baja terrane along the margin to the
south (Baldwin and Harrison, 1989, 1992). The concordance of
the 135 Ma Rb-Sr and K-Ar model ages from sample 330–4B
(Fig. 5) indicate that 40Ar/39Ar results from it cannot be explained
by excess 40Ar contamination. Petrographic observations of
Figure 9. Relative probability plots (0–2000 Ma) of detrital zircon U-Pb age results used as end members in ternary
mixing calculations: (A) Early Mesozoic flysch wall rocks of the northern Peninsular Ranges batholith (PRB) from
Morgan et al. (2005). (B) Early Cretaceous volcanic sandstones (Alsleben et al., 2005) and volcanic rocks (D. Kim-
brough, personal observ.) of the northern Peninsular Ranges batholith. (C) Late Cretaceous forearc strata of the northern
Peninsular Ranges batholith (Mahoney et al., 2005). (D) Cumulative probability plots of the three end members. Best-fit
ternary-mixing results: (E) Amphibolite unit. (F) Epidote-amphibolite unit. (G) Epidote-blueschist unit. (H) Lawsonite-
blueschist (and lower grade) units of the Catalina Schist. (I) Ternary diagram showing best-fit solutions. Note the general
trend away from Peninsular Ranges wall rock and volcanic zircon signature exhibited by the oldest accreted units toward
a pluton-dominated zircon provenance that characterizes the youngest rocks within the Catalina Schist. Mz—Mesozoic;
K—Cretaceous.
 The Catalina Schist
sample 330–4B indicate that the K-Ar and Rb-Sr model ages are
most sensibly interpreted as mixed ages that resulted from partial
replacement of 0.1–1 mm diameter Middle Jurassic phengite by
much finer grained (<50 µm) middle Cretaceous phengite. This
evidence for prior (Middle Jurassic?) subduction metamorphism
along the margin is in good agreement with the geologic history
of the Peninsular Ranges batholith. Conversely, pre–115 Ma sub-
duction along the segment of the margin underplated by the Cata-
lina Schist makes it difficult to explain how mantle lithosphere
composing the overriding hanging wall could have retained suf-
ficient heat to produce amphibolite facies metamorphism within
a newly formed subduction zone at 115 Ma.
Subduction Erosion Model for the Formation of the
Catalina Schist
The results and geologic relationships described above
lead us to conclude that the amphibolite, epidote-amphibolite,
and possibly the epidote-blueschist units of the Catalina Schist
are unlikely to have formed within the same subduction zone
in which the lawsonite-blueschist and lower grade rocks of the
Catalina Schist were also accreted and metamorphosed. Whereas
studies of plate kinematics involving the Farallon, Kula, and
Pacific plates are not definitive, they indicate that continuous sub-
duction of comparatively old oceanic crust likely prevailed along
the southwestern North American margin throughout the middle
Cretaceous (Engebretsen et al., 1986; Stock and Molnar, 1988).
Accordingly, we favor a model in which the Catalina amphibolite
and epidote-amphibolite units represent Early Cretaceous forearc
strata and basement (Fig. 10A) that were underthrust and meta-
morphosed beneath a forearc thrust that was well separated from
the deeper subduction megathrust (Fig. 10B). A modern analogue
for paired megathrusts depicted in Figure 10B is imaged by deep
seismic-reflection data from the northern Cascadia subduction
zone (Calvert, 2004). Forearc thrusting on this scale is capable
of displacing forearc rocks to subcrustal positions and is conse-
quently an important manifestation of subduction erosion.
In the Catalina Schist, high-PT amphibolite facies meta-
morphism and partial melting may have occurred when a part
of the Early Cretaceous forearc (Fig. 10A) was underthrust to a
position beneath the western margin of the northern Peninsular
Ranges batholith between 122 and 115 Ma (Fig. 10B). This event
coincided broadly with previously documented intra-arc thrust-
ing within the adjacent Peninsular Ranges batholith (Gastil et al.,
1981; Todd et al., 1988; Silver and Chappell, 1988; Thomson and
Girty, 1994; Busby et al., 1998; Johnson et al., 1999: Schmidt
et al., 2002; Wetmore et al., 2002; Schmidt and Paterson, 2002;
Wetmore et al., 2003; Busby, 2004). Shortening of the forearc
could have been triggered by lateral expansion related to batho-
lith emplacement coupled with collapse of thin, hot backarc
crust that had separated the Peninsular Ranges batholith from the
southwestern craton margin in the Early Cretaceous.
Underthrusting of the Early Cretaceous forearc rocks to a
stalled position beneath the western margin of the magmatic arc
351
(Fig. 10B) explains the heat source required to produce amphibo-
lite facies metamorphism and anatexis within the Catalina amphib-
olite unit. It also explains why the amphibolite unit underwent
protracted greenschist facies metamorphic conditions after peak
metamorphism occurred at 115 Ma. Greenschist facies, rather
than blueschist facies overprinting, is expected if the amphibo-
lite unit was metamorphosed close to the magmatic arc and far
removed from the subduction zone.
Based upon detrital zircon results, the epidote-amphibolite
unit was accreted after 113 ± 3 Ma or several million years
after peak grade recrystallization in the overlying amphibolite
unit (Fig. 6B). Accretion of the epidote-amphibolite unit cre-
ated an imbricate thrust stack beneath the western margin of the
Peninsular Ranges batholith (Fig. 10C). Peak grade recrystalliza-
tion of the epidote-amphibolite unit occurred at 110–107 Ma on
the basis of 40Ar/39Ar hornblende results from the mafic gneiss
directly underlying the amphibolite unit (Fig. 6B). Just as in the
case of the amphibolite unit, greenschist facies metamorphic
conditions likely persisted within the epidote-amphibolite unit
for up to 10 m.y. on the basis of 40Ar/30Ar phengite results from
epidote-amphibolite unit metagraywackes that indicate that cool-
ing below 400–350 °C was delayed until ca. 97 Ma (Fig. 6B).
Only fragmentary constraints are available for the meta-
morphic evolution of the epidote-blueschist unit. Previous work
had indicated that the rocks of the epidote-amphibolite unit and
epidote-blueschist unit were contiguous (the Catalina greenschist
unit of Platt, 1976) and shared a common greenschist facies over-
print (Platt, 1976; Sorensen, 1986). Our detrital zircon results
from a single sample of epidote-blueschist metasediment inter-
calated with clinozoisite-albite–bearing blueschists cast doubt
upon the likelihood that the epidote-amphibolite and epidote-
blueschist units were closely related prior to ca. 101 Ma. Results
from this single sample indicate that the epidote-blueschist rocks
were accreted after 101 ± 3 Ma or roughly 6 m.y. after peak-grade
recrystallization of the epidote-amphibolite unit had occurred. The
epidote-blueschist assemblages could reflect subcrustal accretion
beneath the collapsing forearc at an intermediate position between
the subduction zone and the arc. Greenschist facies overprinting
of the epidote-blueschist assemblages resulted as the rocks were
underthrust beneath the epidote-amphibolite and higher grade
rocks of the Catalina Schist between 101 and 97 Ma.
The lawsonite-blueschist and lower grade rocks of the Cata-
lina Schist were accreted after 97 ± 3 Ma during a major Ceno-
manian pulse of synbatholith erosion (Fig. 10D; Kimbrough et
al., 2001). Evidence that accretion took place within the sub-
duction zone is provided by the lawsonite-blueschist and lower
grade mineralogy and the fact that phengite 40Ar/39Ar ages and
the youngest detrital zircon ages coincide, indicating that accre-
tion, metamorphism, and cooling took place rapidly (Fig. 6D).
Merging of the higher grade units of the Catalina Schist with the
subduction complex likely occurred by ca. 95 Ma and triggered
tectonic denudation via low-angle extensional faulting that atten-
uated the crust and established the presently observed contact
relationships within the Catalina Schist (e.g., Platt, 1986).
(A)

(B)

(C)

(D)

(E)
 The Catalina Schist 353

Figure 10. Subduction erosion model for the formation of the Catalina Schist. (A) Early Cretaceous geometry of the convergent margin of the
northern Peninsular Ranges batholith. The backarc extension and sedimentation shown are inferred from geologic relationships within east-
central Peninsular Ranges batholith and formerly adjacent mainland Mexico. Detrital zircon results from Early Cretaceous volcanic sediments
(Alsleben et al., 2005) require close proximity to southwestern North America. (B) Initial underthrusting of Early Cretaceous forearc beneath
northwestern Peninsular Ranges and amphibolite facies metamorphism and anatexis at 115 Ma. Amphibolite unit stalls beneath western batholith
at a position well separated from active subduction zone. Evidence for intra-arc shortening at this time is detailed within the text. (C) Under-
thrusting of epidote-amphibolite unit during progressive subduction erosion of the Early Cretaceous forearc beneath northwestern Peninsular
Ranges batholith. Shortening in the backarc region preconditions the crust for deep melting (>40 km) to generate La Posta magmatism at 95 Ma.
(D) Accretion of the lawsonite-blueschist and lower grade units of the Catalina Schist at 95 Ma occurs concomitantly with a major pulse of de-
nudation and erosion within the Peninsular Ranges batholith (e.g., Kimbrough et al., 2001). Deep, garnet-involved melting to generate La Posta
magmatism is facilitated by overthickened arc-ophiolite basement in the former backarc, focused asthenospheric counterflow caused by initiation
of flat subduction and eastern relocation of magmatic arc, and massive devolatilization of deep underplated Catalina Schist within the subduc-
tion channel. (E) Laramide flat subduction tectonically erodes deep crustal root beneath eastern Peninsular Ranges batholith between 80 and
65 Ma, causing renewed deep exhumation of the eastern part of this batholith (Grove et al., 2003b). Tr—Triassic; J—Jurassic; K—Cretaceous;
Pz—Paleozoic; TTG—tonalite, trondjhemite, granodiorite.

Devolatilization and Fluid Flow within the Catalina Schist
Recent studies (Bebout et al., 1999, 2007; Bebout, 2007)
have discussed varying extents of devolatilization in the tectono-
metamorphic units of the Catalina Schist within the context of
varying prograde P-T paths that took place in a subduction zone
setting (also see discussion in Grove and Bebout, 1995). Whereas
the subduction erosion model (Fig. 10) does not affect the general
conclusions of this previous work as they pertain to devolatiliza-
tion histories, the sources and pathways for infiltrating fluids that
metasomatized the higher grade rocks of the Catalina Schist do
require further consideration within the context of the subduction
erosional model presented here. Bebout and Barton (1989) and
Bebout (1991a, b) proposed that, in the amphibolite unit, the O
isotope compositions in mélange and other more permeable zones
reflect infiltration by externally derived aqueous fluids with δ O
18
the forearc. Pervasive hydration of mantle lithosphere underly-
ing the forearc region is consistent with observations, in mod-
ern forearcs, of extensive hydration in the hanging wall leading
to distinctive seismic-velocity signatures (Bostock et al., 2002;
Brocher et al., 2003) and forearc serpentinite diapirs (Fryer et
al., 1995). As continued subduction erosion brought the loci of
forearc thrusting nearer to the active subduction thrust system
(see Fig. 10C, D, E), progressively lower T rocks, such as those
in the epidote-amphibolite and epidote-blueschist rocks, would
similarly have been infiltrated by these fluids emanating from
the still lower T (lawsonite-blueschist) domains produced in the
active subduction zone thrust.
Relationship to Emplacement of the La Posta Tonalite-
Trondjhemite-Granodiorite Suite

inherited from prior equilibration with similar metasedimentary
rocks but at lower temperatures (~300–600 °C). The source of the
fluids that ascended into the underplated (but still hot) amphibo-
lite unit was primarily related to 400–550 °C chlorite breakdown
reactions in rocks equivalent to those within the lower grade units
of the Catalina Schist (Bebout, 1991a, b; also see discussion in
Grove and Bebout, 1995).
In the subduction erosion model (Fig. 10), fluids with appro-
priate lower T metasedimentary signatures derived from the
subducting slab are required to traverse a significant thickness
(tens of kilometers) of mantle lithosphere in order to infiltrate
the amphibolite unit at near peak-grade metamorphic conditions.
We suggest that this could have occurred without substantial
modification of the δ O of the fluids. Devolatilization of sub-
18
ducting oceanic crust and trench fill is considered capable of
hydrating mantle lithosphere over broad regions (Peacock, 1993;
Humphreys et al., 2003). Infiltration of aqueous fluids derived
from the subduction zone likely began during, or prior to, the
Middle Jurassic. This previous history was likely sufficient to
thoroughly hydrate and isotopically re-equilibrate the interven-
ing lithospheric mantle with lower temperature, slab-derived flu-
ids prior to the onset of Early Cretaceous subduction erosion of
At ca. 100 Ma the locus of Peninsular Ranges batholith mag-
matism shifted abruptly eastward in conjunction with intrusion
of the voluminous La Posta tonalite-trondjhemite-granodiorite
plutonic suite (Fig. 2; Gastil et al., 1975; Silver and Chappell,
1988; Walawender et al., 1990; Tulloch and Kimbrough, 2003;
Kimbrough and Grove, 2006). The La Posta suite emplacement
involved a sustained (ca. 98–92 Ma) magmatic flux of >100 km3/
m.y./km strike length over >1200 km (Kimbrough and Grove,
2006). The high Sr, Ba, Sr/Y, Na2O, Al2O3, and highly fraction-
ated rare-earth-element patterns exhibited by these plutons indi-
cate deep, garnet-involved melting of a fundamentally mafic
source region (Gromet and Silver, 1987). Oxygen and Rb-Sr
isotopic measurements reported by Taylor (1986), Gromet and
Silver (1987), Silver and Chappell (1988), Hill and Silver (1988),
and Kistler et al. (2003) have revealed that the northern La Posta
belt features elevated δ18O (9–12‰ whole-rock equivalent) at
intermediate initial 87Sr/86Sr values (typically 0.704–0.708). The
supracrustal contribution implied by these combined isotopic
attributes cannot be accounted for by high-level assimilation of
highly radiogenic, early Mesozoic, cratonally derived flysch host
rocks (Shaw et al., 2003). These characteristics are more readily
explained by partial melting and/or devolatilization of isotopically
354 Grove et al.
primitive low-grade metasedimentary and metavolcanic rocks
and altered oceanic crust within the deep source region (Taylor,
1986; Gromet and Silver, 1987).
Emplacement of the La Posta belt plutons involved an east-
ern relocation of the locus of magmatism within the Peninsular
Ranges over a strike length of 1200 km (Fig. 10D; Kimbrough
and Grove, 2006). This event was the initial stage of an eastern
sweep of magmatism into northern Mexico that occurred dur-
ing the Late Cretaceous–early Tertiary (e.g., Silver and Chappell,
1988; McDowell et al., 2001; Staude and Barton, 2001; Henry et
al., 2003). Subduction erosion (Fig. 10D) could have delivered
the Catalina Schist into close proximity to the La Posta source
region between 100 and 95 Ma. The coincidence of crustal thick-
ening of the former backarc extensional basin (compare Fig. 10A
with 10D) focused asthenospheric corner flow (Fig. 10D), and
massive devolatilization of the Catalina Schist (Fig. 10D) may
have set up the optimal conditions required to trigger the La Posta
TTG flare-up (Kimbrough and Grove, 2006). Recent δ18O mea-
surements performed by Kimbrough and Grove (2006) indicate
that the La Posta plutons have high δ18O only in the northern
segment of the batholith, where it might have been underplated
by the Catalina Schist (Fig. 2). This is a good indication that
Catalina Schist subduction erosion was sufficiently important to
modify the source region characteristics of the La Posta belt mag-
matism. An important implication is that substantial fractions
of subducted forearc supracrustal material can be recycled via
subduction erosion processes and incorporated into arc batholith
magmas over a short (<10 Ma) time interval.
Relationship to Laramide Underthrusting
The accretion of the lowest grade units of the Catalina Schist
beneath the western Peninsular Ranges and the eastern relocation
of the La Posta plutonic belt to the eastern Peninsular Ranges at
95 Ma can be viewed as important precursors to the Laramide
craton-ward shift of arc magmatism and contractional deforma-
tion (Coney and Reynolds, 1977; Dickinson and Snyder, 1978).
During the Laramide episode of Late Cretaceous–early Tertiary
deformation, large tracts of southern California and southwest-
ern Arizona were underplated by the high-PT Pelona Schist and
related schists, with no intervening lithospheric mantle preserved
(Fig. 10E; Ehlig, 1968, 1981; Crowell, 1968, 1981; Yeats, 1968;
Haxel and Dillon, 1978; Haxel et al., 2002; Burchfiel and Davis,
1981; Jacobson, 1983, 1990; Jacobson et al., 1988, 2002, 2007;
Dillon et al., 1990; Malin et al., 1995; Wood and Saleeby, 1997;
Saleeby, 2003; Grove et al., 2003a). Late Cretaceous eclogitic
xenoliths have been recovered from kimberlite pipes as far east as
northeastern Arizona (Usui et al., 2003).
The earliest recognized accretion of the distinctive Pelona
and related schists occurred at 91 ± 1 Ma along the southwest-
ernmost tip of the Sierra Nevada batholith (Rand Schist within
the San Emigdio Mountains; Saleeby, 2003; Grove et al., 2003a).
Lawsonite-blueschist and the lower grade Catalina Schist rep-
resent nearly equivalent material underplated at a slightly older
time (i.e., between 97 and 95 Ma) beneath the northern Peninsular
Ranges (Fig. 6B). Thermochronology and detrital zircon results
obtained from the NW-SE–trending belt of schist exposures have
revealed that schist underplating associated with the Laramide
event was widespread by 80–70 Ma beneath the middle Creta-
ceous arc (Jacobson, 1990; Jacobson et al., 2000; Barth et al.,
2003; Grove et al., 2003a). By 70–60 Ma, cratonal rocks were
being underplated by schist at positions well east of the medial
Cretaceous arc (Grove et al., 2003a; Usui et al., 2003; Jacobson
et al., 2007).
Laramide shallow subduction processes tectonically removed
the deep lithospheric mantle roots of the La Posta belt within
the northern Peninsular Ranges batholith between 80 and 65 Ma
(Fig. 10E). Receiver function seismic studies indicate that the
deep crust and lithospheric mantle roots no longer exist beneath
or within the northeastern Peninsular Ranges batholith (Ichinose
et al., 1996; Lewis et al., 2001). Whereas this batholith exhib-
ited predominantly syn- to late-batholithic cooling in its southern
extent (>85 Ma; e.g., Ortega Rivera, 2003), the northeastern seg-
ment of the batholith was further characterized by a delayed and
very significant pulse of rapid cooling between 80 and 65 Ma
(Krummenacher et al., 1975; George and Dokka, 1994; Lovera
et al., 1999; Grove et al., 2003b). Lovera et al. (1999) and Grove
et al. (2003b) attributed this cooling to denudation related to the
removal of lithospheric mantle and the underplating of schist dur-
ing Laramide shallow subduction. Holk et al. (2006) reported that
Laramide-age deformation of the northern Peninsular Ranges
was associated with infiltration of high δD and δ18O fluids that
are most readily explained by devolatilization of subducted oce-
anic crust or underplated volcanogenic sediments. The impact of
the Laramide orogeny on the northern Peninsular Ranges appears
to have contrasted significantly with its effect upon the south-
ern Sierra Nevada batholith, where an inflection in the Laramide
subduction zone (Pickett and Saleeby, 1993; Malin et al., 1995;
Saleeby, 2003) apparently allowed the deep crust and upper man-
tle to be preserved until much more recently (Ducea and Saleeby,
1996, 1998; Zandt et al., 2004).
CONCLUSIONS
1. Major tectonometamorphic units of the Catalina Schist
were successively accreted over a 15–20 m.y. interval,
beginning with the amphibolite unit at ca. 120–115 Ma
and concluding with the lawsonite-blueschist and lower
grade lithologies by 97–95 Ma.
2. The amphibolite unit resided at high temperatures for
a protracted period of time (10–15 m.y.), whereas the
lawsonite-blueschist and lower grade units that most
likely formed near or within the subduction zone were
deposited, accreted, metamorphosed, and cooled over
a time interval too brief to be resolved by the methods
employed in this study (<3 m.y.).
3. The provenance of the Catalina Schist metasedimentary
rocks shifted as a function of time of accretion (and now,
 The Catalina Schist
metamorphic grade). Metagraywackes from the earliest
accreted amphibolite unit were derived from an early
Aptian sediment that apparently originated from ero-
sion of Late Triassic–Jurassic flysch wall rocks and Early
Cretaceous volcanics of the Peninsular Ranges batholith.
Successively younger accreted materials became enriched
with Early Cretaceous plutonic zircon from the Peninsular
Ranges. The last accreted lawsonite-blueschist and lower
grade rocks were derived from Turonian sediment with a
detrital zircon provenance virtually identical to similarly
aged sediment within the Peninsular Ranges forearc.
4. The Catalina Schist likely does not represent a synchro-
nously formed, inverted metamorphic aureole arising
from nascent subduction. The highest grade portions of
the complex appear to have formed by a subduction ero-
sion process in which parts of the forearc were under-
thrust beneath the western margin of the Peninsular
Ranges batholith at ca. 122–115 Ma. Progressive subduc-
tion erosion of the forearc by continued underthrusting in
the forearc region ultimately juxtaposed the higher grade
units of the Catalina Schist with the subduction complex
by 97–95 Ma. Only the lawsonite-blueschist and lower
grade rocks are considered to have originated along the
dominant subduction zone thrust within a high-PT ther-
mal regime characteristic of such settings.
5. The accretion of the Catalina Schist marked the initial
stage of shallowing subduction and inboard migration of
magmatism and sediment underplating that culminated
in the Late Cretaceous–early Tertiary Laramide orogeny.
The sedimentary protolith of the youngest Catalina schist
units are nearly identical in age and provenance to the
oldest representative of the eugeoclinal schists emplaced
beneath the southwesternmost Sierra Nevada batholith at
ca. 92 Ma.
ACKNOWLEDGMENTS
We gratefully acknowledge discussions with Pat Abbott,
Suzanne Baldwin, Andy Barth, Mark Barton, Ned Brown,
Cathy Busby, Mark Cloos, Mihai Ducea, Ray Ingersoll, Peter
Lonsdale, John Platt, Jason Saleeby, Dave Scholl, Sorena
Sorensen, John Wakabayashi, and Jim Wright that helped shape
ideas presented in this paper. We particularly thank Bob Bohan-
non, Mark Cloos, John Platt, Sorena Sorensen, and John Wak-
abayashi, whose reviews of previous versions of the manuscript
led to significant improvements. Work carried out by Brian
Altheim and Cathy Christoffel, as parts of their M.S. research
at Lehigh University, provided important information on low-
grade metamorphism within the Catalina Schist that helped
guide our sampling. Helge Alsleben, Brian Mahoney, and J.R.
Morgan provided permission to use detrital zircon U-Pb age
data from the Peninsular Ranges batholith to compare with our
results from the Catalina Schist. Joe Wooden and Drew Cole-
man shared unpublished whole-rock Pb isotopic data from the
355
Mojave–Transverse Ranges province. Field support on Catalina
Island was provided by the Catalina Island Conservancy, with
special thanks to Frank Starkey for providing logistical support
and assistance in the field. Michelle Hopkins prepared samples
and performed cathodoluminescence imaging of zircon grains.
J.R. Morgan assisted with ion microprobe analysis. The ion
microprobe facility at UCLA and the Laserchron Center at the
University of Arizona are partly supported by grants from the
Instrumentation and Facilities Program, Division of Earth Sci-
ences, U.S. National Science Foundation.
REFERENCES CITED
Ague, J., and Brimhall, G.H., 1988, Magmatic arc asymmetry and distribu-
tion of anomalous plutonic belts in the batholiths of California: Effects
of assimilation, crustal thickness, and depth of crystallization: Geologi-
cal Society of America Bulletin, v. 100, p. 912–927, doi: 10.1130/0016-
7606(1988)100<0912:MAAADO>2.3.CO;2.
Alsleben, H., Wetmore, P., Gehrels, G., and Paterson, S., 2005, Using detrital
zircon geochronology to track island arc collision: Geological Society of
America Abstracts with Programs, v. 37, no., 7, p. 551–552.
Altheim, B.K., Christoffel, C.A., Bebout, G.E., and Grove, M., 1997, Detec-
tion of subtle variations in low-T, high-P/T metamorphism in the Cata-
lina Schist; tectonic and kinetic implications: Program with Abstracts,
Geological Association of Canada, Mineralogical Association of Canada,
Canadian Geophysical Union, Joint Annual Meeting, v. 22, p. 2.
Anczkiewicz, R., Platt, J.P., Thirlwall, M.F., and Wakabayashi, J., 2004, Fran-
ciscan subduction off to a slow start: Evidence from high-precision Lu-Hf
garnet ages on high-grade blocks: Earth and Planetary Science Letters,
v. 225, p. 147–161, doi: 10.1016/j.epsl.2004.06.003.
Atwater, T., 1970, Implications of plate tectonics for the Cenozoic tectonic
evolution of western North America: Geological Society of America
Bulletin, v. 81, p. 3513–3536, doi: 10.1130/0016-7606(1970)81[3513:
IOPTFT]2.0.CO;2.
Bailey, E.H., 1941, Mineralogy, petrology, and geology of Catalina Island, Cal-
ifornia [Ph.D. thesis]: Stanford, California, Stanford University, 193 p.
Baird, A.K., and Miesch, A.T., 1984, Batholithic Rocks of Southern Califor-
nia—A Model for the Petrochemical Nature of Their Source Materials:
U.S. Geological Survey Professional Paper 1284, 24 p.
Baldwin, S.L., and Harrison, T.M., 1989, Geochronology of blueschists from
west-central Baja California and the timing of uplift in subduction com-
plexes: Journal of Geology, v. 97, p. 149–163.
Baldwin, S.L., and Harrison, T.M., 1992, The P-T-T history of blocks in
serpeninite matrix melange, west-central Baja California: Geologi-
cal Society of America Bulletin, v. 104, p. 18–31, doi: 10.1130/0016-
7606(1992)104<0018:TPTTHO>2.3.CO;2.
Bannon, J.L., Bottjer, D.J., Lund, S.P., and Saul, L.R., 1989, Campanian/Maas-
trichtian stage boundary in southern California: Resolution and implica-
tions for large-scale depositional patterns: Geology, v. 17, p. 80–83, doi:
10.1130/0091-7613(1989)017<0080:CMSBIS>2.3.CO;2.
Barth, A.P., and Wooden, J.L., 2006, Timing of magmatism following initial
convergence at a passive margin, southwestern U.S. Cordillera, and ages
of lower crustal magma sources: Journal of Geology, v. 114, p. 231–245,
doi: 10.1086/499573.
Barth, A.P., Wooden, J.L., Tosdal, R.M., Morrison, J., Dawson, D.L., and Hernly,
B.M., 1995, Origin of gneisses in the aureole of the San Gabriel anorthosite
complex and implications for the Proterozoic crustal evolution of south-
ern California: Tectonics, v. 14, p. 736–752, doi: 10.1029/94TC02901.
Barth, A.P., Wooden, J.L., Grove, M., Jacobson, C.E., and Pedrick, J.N., 2003,
U-Pb zircon geochronology of rocks in the Salinas Valley region of Cali-
fornia: A reevaluation of the crustal structure and origin of the Salinian
block: Geology, v. 31, no. 6, p. 517–520.
Barton, M.D., Battles, D.A., Bebout, G.E., Capo, R.C., Christensen, J.N., Davis,
S.R., Hanson, R.B., Michelsen, C.J., and Trim, H.E., 1988, Mesozoic
contact metamorphism in the western United States, in Ernst, W.G., ed.,
Metamorphism and Crustal Evolution of the Western United States (Rubey
Volume 7): Englewood Cliffs, New Jersey, Prentice Hall, p. 110–178.
356 Grove et al.
Bebout, G.E., 1991a, Field-based evidence for devolatilization in subduction
zones: Implications for arc magmatism: Science, v. 251, p. 413–416, doi:
10.1126/science.251.4992.413.
Bebout, G.E., 1991b, Geometry and mechanisms of fluid flow at 15 to 45 kilo-
meter depths in an early Cretaceous accretionary complex: Geophysical
Research Letters, v. 18, p. 923–926.
Bebout, G.E., 2007, Trace element and isotopic fluxes/subducted slab, in Trea-
tise on Geochemistry, v. 3: New York, Elsevier, p. 1–50, doi: 10.1016/
B978-008043751-4/00231-5.
Bebout, G.E., and Barton, M.D., 1989, Fluid flow and metasomatism in a sub-
duction zone hydrothermal system: Catalina Schist terrane, California:
Geology, v. 17, p. 976–980, doi: 10.1130/0091-7613(1989)017<0976:
FFAMIA>2.3.CO;2.
Bebout, G.E., and Barton, M.D., 1993, Metasomatism during subduction: Prod-
ucts and possible paths in the Catalina Schist, California: Chemical Geol-
ogy, v. 108, p. 61–92, doi: 10.1016/0009-2541(93)90318-D.
Bebout, G.E., and Barton, M.D., 2002, Tectonic and metasomatic mixing in a
high-T, subduction zone mélange—Insights into the geochemical evolu-
tion of the slab-mantle interface: Chemical Geology, v. 187, p. 79–106,
doi: 10.1016/S0009-2541(02)00019-0.
Bebout, G.E., Ryan, J.G., Leeman, W.P., and Bebout, A.E., 1999, Fractionation
of trace elements during subduction-zone metamorphism: Impact of con-
vergent margin thermal evolution: Earth and Planetary Science Letters,
v. 171, p. 63–81, doi: 10.1016/S0012-821X(99)00135-1.
Bebout, G.E., Bebout, A.E., and Graham, C.M., 2007, Cycling of B, Li, and
LILE (K, Cs, Rb, Ba, Sr) into subduction zones: SIMS evidence from
micas in high P/T metasedimentary rocks: Chemical Geology, vol. 239, p.
284–304, doi: 10.1016/j.chemgeo.2006.10.016.
Blake, M.C., Jr., Jayko, A.S., Moore, T.E., Chavez, V., Saleeby, J.B., and Seel,
K., 1984, Tectonostratigraphic terranes of Magdalena Island, Baja Cali-
fornia Sur, in Frizzel, V.A., Jr., ed., Geology of the Baja California Pen-
insula: Society of Economic Paleontologists and Mineralogists, Pacific
Section, v. 39, p. 183–191.
Bohannon, R.G., and Geist, E.L., 1998, Upper crustal structure and Neogene
tectonic development of the California continental borderland: Geologi-
cal Society of America Bulletin, v. 110, p. 779–800, doi: 10.1130/0016-
7606(1998)110<0779:UCSANT>2.3.CO;2.
Bohannon, R.G., and Parsons, T., 1995, Tectonic implications of post-30 Ma
Pacific and North American relative plate motions: Geological Society of
America Bulletin, v. 107, no. 8, p. 937–959.
Böhnel, H., Delgado-Argote, L.A., and Kimbrough, D.L., 2002, Discordant
paleomagnetic data for middle Cretaceous intrusive rocks from northern
Baja California: Latitude displacement, tilt, or vertical axis rotation?: Tec-
tonics, v. 21, doi: 10.1029/2001TC001298.
Boles, J.R., 1978, Basin analysis of the Eugenia Formation (Late Jurassic),
Punta Eugenia area, Baja, California, in Howell, D.G., and McDougall,
K.A., eds., Mesozoic Paleogeography of the Western United States,
Pacific Coast Paleogeography Symposium 2: Sacramento, California,
Society of Economic Paleontologists and Mineralogists, p. 493–498.
Boles, J.R., and Landis, C.A., 1984, Jurassic sedimentary mélange and asso-
ciated facies, Baja California, Mexico: Geological Society of America
Bulletin, v. 95, p. 513–521, doi: 10.1130/0016-7606(1984)95<513:
JSMAAF>2.0.CO;2.
Bonini, J.A., and Baldwin, S.L., 1998, Mesozoic metamorphic and middle to
late Tertiary magmatic events on Magdalena and Santa Margarita islands,
Baja California Sur, Mexico: Implications for the tectonic evolution of the
Baja California continental borderland: Geological Society of America
Bulletin, v. 110, p. 1094–1104, doi: 10.1130/0016-7606(1998)110<1094:
MMAMTL>2.3.CO;2.
Bostock, M.G., Hyndman, R.D., Rondenay, S., and Peacock, S.M., 2002, An
inverted continental Moho and the serpentinization of the forearc mantle:
Nature, v. 417, p. 536–538, doi: 10.1038/417536a.
Bottjer, D.J., and Link, M.H., 1984, A synthesis of Late Cretaceous southern
California and northern Baja California paleogeography, in Crouch, J.K.,
and Bachman, S.B., eds., Tectonics and Sedimentation along the Califor-
nia Margin: Fullerton, California, Society of Economic Paleontologists
and Mineralogists, Pacific Section, p. 171–188.
Bottjer, D.J., Colburn, I.P., and Cooper, J.D., 1982, Late Cretaceous Deposi-
tional Environments and Paleogeography, Santa Ana Mountains, South-
ern California: Los Angeles, Society of Economic Paleontologists and
Mineralogists, Pacific Section, 121 p.
Brocher, T.M., Parsons, T., Tréhu, A.M., Snelson, C.M., and Fisher, M.A., 2003,
Seismic evidence for widespread serpentinized forearc upper mantle along
the Cascadia margin: Geology, v. 31, p. 267–270, doi: 10.1130/0091-
7613(2003)031<0267:SEFWSF>2.0.CO;2.
Brown, E.H., Wilson, D.L., Armstrong, R.L., and Harakal, J.E., 1982, Petrologic,
structural, and age relations of serpentinite, amphibolite, and blueschist in
the Shuksan Suite of the Iron Mountain–Gee Point area, North Cascades,
Washington: Geological Society of America Bulletin, v. 93, p. 1087–1098,
doi: 10.1130/0016-7606(1982)93<1087:PSAARO>2.0.CO;2.
Burchfiel, B.C., and Davis, G.A., 1981, Mojave Desert and environs, in Ernst,
W.G., ed., The Geotectonic Development of California (Rubey Volume
1): Englewood Cliffs, New Jersey, Prentice Hall, p. 217–252.
Busby, C.J., 2004, Continental growth at convergent margins facing large ocean
basins: A case study from Mesozoic Baja California, Mexico: Tectono-
physics, v. 392, p. 241–277, doi: 10.1016/j.tecto.2004.04.017.
Busby, C.J., Smith, D.P., Morris, W.R., and Adams, B., 1998, Evolutionary
model for convergent margins facing large ocean basins: Mesozoic Baja
California (Mexico): Geology, v. 26, p. 227–230, doi: 10.1130/0091-
7613(1998)026<0227:EMFCMF>2.3.CO;2.
Calvert, A.J., 2004, Seismic reflection imaging of two megathrust shear zones
in the northern Cascadia subduction zone: Nature, v. 428, p. 163–167,
doi: 10.1038/nature02372.
Chen, J.H., and Moore, J.G., 1982, Uranium-lead isotopic ages from the Sierra
Nevada batholith: Journal of Geophysical Research, v. 87, p. 4761–4784.
Clift, P., and Vannucchi, P., 2004, Controls on tectonic accretion versus erosion in
subduction zones: Implications for the origin and recycling of the continen-
tal crust: Reviews of Geophysics, v. 42, doi: 10.1029/2003RG000127.
Cloos, M., 1985, Thermal evolution of convergent plate margins—Thermal
modeling and reevaluation of isotopic Ar-ages for blueschists in the Fran-
ciscan complex of California: Tectonics, v. 4, p. 421–433.
Coney, P.J., and Reynolds, S.J., 1977, Cordilleran Benioff zones: Nature, v. 270,
p. 403–406, doi: 10.1038/270403a0.
Critelli, S., Marsaglia, K.M., and Busby, C.J., 2002, Tectonic history of a Juras-
sic backarc-basin sequence (the Gran Cañon Formation, Cedros Island,
Mexico), based on compositional modes of tuffaceous deposits: Geologi-
cal Society of America Bulletin, v. 114, p. 515–527, doi: 10.1130/0016-
7606(2002)114<0515:THOAJB>2.0.CO;2.
Crouch, J.K., and Suppe, J., 1993, Late Cenozoic tectonic evolution of the
Los Angeles Basin and inner California borderland; a model for core
complex–like crustal extension: Geological Society of America Bul-
letin, v. 105, p. 1415–1434, doi: 10.1130/0016-7606(1993)105<1415:
LCTEOT>2.3.CO;2.
Crowell, J.C., 1968, Movement histories of faults in the Transverse Ranges and
speculations on the tectonic history of California, in Dickinson, W.R., and
Grantz, A., eds., Proceedings of Conference on Geologic Problems of San
Andreas Fault System: Stanford University Publications in the Geological
Sciences, v. 11, p. 323–341.
Crowell, J.C., 1981, An outline of the tectonic history of southeastern California,
in Ernst, W.G., ed., The Geotectonic Development of California (Rubey
Volume 1): Englewood Cliffs, New Jersey, Prentice Hall, p. 583–600.
Debiche, M.G., Cox, A., and Engebretson, D., 1987, The Motion of Alloch-
thonous Terranes across the North Pacific Basin: Geological Society of
America Special Paper 207, 49 p.
Dibblee, T.W., Jr., 1999, Geologic map of the Palos Verdes Peninsula and
vicinity (Redondo Beach, Torrance, and San Pedro Quadrangles) Los
Angeles County, California: Dibblee Geological Foundation Map DF-70
(Ehrenspeck, H.E., Ehlig, P.L., and Bartlett, W.L., eds.), scale 1:24,000,
colored, 4 cross-sections.
Dickinson, W.R., 1983, Cretaceous sinistral strike slip along Nacimiento fault
in coastal California: American Association of Petroleum Geologists Bul-
letin, v. 67, p. 624–645.
Dickinson, W.R., and Gehrels, G.E., U-Pb ages of detrital zircons from Permian
and Jurassic eolian sandstones from the Colorado Plateau, USA: paleo-
geographic implications: Sedimentary Geology, v. 163, p. 29–66.
Dickinson, W.R., and Snyder, W.S., 1978, Plate tectonics of the Laramide orog-
eny, in Matthews, V., III, ed., Laramide Folding Associated with Base-
ment Block Faulting in the Western United States: Geological Society of
America Memoir 151, p. 355–366.
Dickinson, W.R., Hopson, C.A., and Saleeby, J.B., 1996, Alternate origins of
the Coast Range ophiolite (California): Introduction and implications:
GSA Today, v. 6, no. 2, p. 1–10.
 The Catalina Schist
Dickinson, W.R., Ducea, M., Rosenberg, L.I., Greene, H.G., Graham, S.A.,
Clark, J.C., Weber, G.E., Kidder, S., Ernst, W.G., and Brabb, E.E., 2005,
Net Dextral Slip, Neogene San Gregorio–Hosgri Fault Zone, Coastal
California: Geologic Evidence and Tectonic Implications: Geological
Society of America Special Paper 391, 43 p.
Dillon, J.T., Haxel, G.B., and Tosdal, R.M., 1990, Structural evidence for north-
eastward movement on the Chocolate Mountains thrust, southeasternmost
California: Journal of Geophysical Research, v. 95, p. 19,953–19,971.
Ducea, M.N., and Saleeby, J.B., 1996, Buoyancy sources for a large, unrooted
mountain range, the Sierra Nevada, California: Evidence from xenolith
thermobarometry: Journal of Geophysical Research, v. 101, p. 8229–
8244, doi: 10.1029/95JB03452.
Ducea, M.N., and Saleeby, J.B., 1998, The age and origin of a thick mafic-
ultramafic keel from beneath the Sierra Nevada batholith: Contribu-
tions to Mineralogy and Petrology, v. 133, p. 169–185, doi: 10.1007/
s004100050445.
Ehlig, P.L., 1968, Causes of distribution of Pelona, Rand, and Orocopia Schists
along the San Andreas and Garlock faults, in Dickinson, W.R., and
Grantz, A., eds., Proceedings of Conference on Geologic Problems of San
Andreas Fault System: Stanford University Publications in the Geological
Sciences, v. 11, p. 294–306.
Ehlig, P.L., 1981, Origin and tectonic history of the basement terrane of the San
Gabriel Mountains, central Transverse Ranges, in Ernst, W.G., ed., The
Geotectonic Development of California (Rubey Volume 1): Englewood
Cliffs, New Jersey, Prentice Hall, p. 253–283.
Engebretsen, D.C., Cox, A., and Gordon, R.G., 1986, Relative motions between
oceanic and continental plates in the Pacific Basin: Geological Society of
America Special Paper 206, 59 p.
Ernst, W.G., 1988, Tectonic history of subduction zones inferred from retrograde
blueschist P-T paths: Geology, v. 16, p. 1081–1085, doi: 10.1130/0091-
7613(1988)016<1081:THOSZI>2.3.CO;2.
Evernden, J.F., and Kistler, R.W., 1970, Chronology of Emplacement of Meso-
zoic Batholithic Complexes in California and Western Nevada: U.S. Geo-
logical Survey Professional Paper 623, 42 p.
Fletcher, J.M., Grove, M., Kimbrough, D.L., Lovera, O.M., and Gehrels, G.E.,
2007, Ridge-trench interactions and the Neogene tectonic evolution of the
Magdalena Shelf: Insights from detrital zircon U-Pb ages from the Mag-
dalena fan and adjacent areas: Geological Society of America Bulletin, v.
119, no. 11-12, p. 1313–1336.
Flynn, C.J., 1970, Post-batholithic geology of La Gloria–Presa Rodriquez
area, Baja California, Mexico: Geological Society of America Bul-
letin, v. 81, p. 1789–1806, doi: 10.1130/0016-7606(1970)81[1789:
PGOTLG]2.0.CO;2.
Fry, J.G., Bottjer, D.J., and Lund, S.P., 1985, Magnetostratigraphy of displaced
Upper Cretaceous strata in southern California: Geology, v. 13, p. 648–
651, doi: 10.1130/0091-7613(1985)13<648:MODUCS>2.0.CO;2.
Fryer, P., Mottl, M., Johnson, L., Haggerty, J., Phipps, S., and Maekawa, H.,
1995, Serpentine bodies in the forearcs of Western Pacific convergent
margins: Origin and associated fluids, in Taylor, B., and Natland, J., eds.,
Active Margins and Marginal Basins of the Western Pacific: American
Geophysical Union Geophysical Monograph 88, p. 259–279.
Gastil, R.G., 1993, Prebatholithic history of Peninsular California, in Gastil,
R.G., and Miller, R.H., eds., The Prebatholithic Stratigraphy of Peninsular
California: Geological Society of America Special Paper 279, p. 145–156.
Gastil, R.G., Phillips, R.P., and Allison, E.C., 1975, Reconnaissance Geology
of the State of Baja California: Geological Society of America Memoir
140, 170 p.
Gastil, R.G., Morgan, G.J., and Krummenacher, D., 1981, The tectonic history
of Peninsular and adjacent Mexico, in Ernst, W.G., ed., The Geotectonic
Development of California (Rubey Volume 1): Englewood Cliffs, New
Jersey, Prentice Hall, p. 284–306.
Gehrels, G.E., and Stewart, J.H., 1998, Detrital zircon geochronology of
Cambrian to Triassic miogeoclinal and eugeoclinal strata of Sonora,
Mexico: Journal of Geophysical Research, v. 103, p. 2471–2487, doi:
10.1029/97JB03251.
George, P.G. and Dokka, R.K., 1994, Major Late Cretaceous cooling events in the
eastern Peninsular Ranges, California and their implications for Cordille-
ran tectonics: Geological Society of America Bulletin, v. 106, p. 903–914.
Girty, G.H., 1987, Sandstone provenance, Point Loma formation, San Diego,
California: Evidence for uplift of the Peninsular Ranges during the
Laramide orogeny: Journal of Sedimentary Petrology, v. 57, p. 839–844.
357
González-León, C.M., Stanley, G.D., Jr., Gehrels, G.E., and Centeno-García,
E., 2006, New data on the lithostratigraphy, detrital zircon and Nd iso-
tope provenance, and paleogeographic setting of the El Antimonio Group,
Sonora, Mexico, in Anderson, T.H., et al., eds., The Mojave-Sonora
Megashear Hypothesis: Development, Assessment, and Alternatives:
Geological Society of America Special Paper 393, p. 259–282.
Gromet, L.P., and Silver, L.T., 1987, REE variations across the Peninsular
Ranges batholith: Implications for batholithic petrogenesis and crustal
growth in magmatic arcs: Journal of Petrology, v. 28, p. 75–125.
Grove, M., and Bebout, G.E., 1995, Cretaceous tectonic evolution of coastal
southern California: Insights from the Catalina Schist: Tectonics, v. 14,
p. 1290–1308, doi: 10.1029/95TC01931.
Grove, M., Jacobson, C.E., Barth, A.P., and Vucic, A., 2003a, Temporal and
spatial trends of Late Cretaceous–Early Tertiary underplating of Pelona
and related schists beneath southern California and southwestern Arizona,
in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p. 381–406.
Grove, M., Lovera, O.M., and Harrison, T.M., 2003b, Late Cretaceous cooling
of the east-central Peninsular Ranges Batholith (33 degrees N): Relation-
ship to La Posta Pluton emplacement, Laramide shallow subduction, and
forearc, in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern
Mexico and the Southwestern USA: Geological Society of America Spe-
cial Paper 374, p. 355–379.
Hacker, B.R., 1990a, Simulation of the metamorphic and deformational history
of the metamorphic sole of the Oman Ophiolite: Journal of Geophysical
Research, v. 95, p. 4895–4907.
Hacker, B.R., 1990b, The role of deformation in the formation of metamor-
phic gradients: Ridge subduction beneath the Oman Ophiolite: Tectonics,
v. 10, p. 455–473.
Hall, C.A., 1991, Geology of the Point Sur–Lopez Point Region, Coast Ranges,
California: A Part of the Southern California Allochthon: Geological
Society of America Special Paper 266, 40 p.
Haxel, G.B., and Dillon, J.T., 1978, The Pelona-Orocopia Schist and Vin-
cent–Chocolate Mountain thrust system, southern California, in How-
ell, D.G., and McDougall, K.A., eds., Mesozoic Paleogeography of the
Western United States: Society of Economic Paleontologists and Min-
eralogists, Pacific Section, Pacific Coast Paleogeography Symposium 2,
p. 453–469.
Haxel, G.B., Jacobson, C.E., Richard, S.M., Tosdal, R.M., and Grubensky, M.J.,
2002, The Orocopia Schist in southwest Arizona: Early Tertiary oceanic
rocks trapped or transported far inland, in Barth, A.P., ed., Crustal Evolu-
tion of the Southwestern United States: Geological Society of America
Special Paper 365, p. 99–128.
Henry, C.D., McDowell, F.W., and Silver, L.T., 2003, Geology and geochronol-
ogy of granitic batholith complex, Sinaloa, México: Implications for Cor-
dilleran magmatism and tectonics, in Johnson, S.E., et al., eds., Tectonic
Evolution of Northwestern Mexico and the Southwestern USA: Geologi-
cal Society of America Special Paper 374, p. 237–273.
Hill, D.J., 1976, Geology of the Jurassic basement rocks, Santa Cruz Island,
California, and correlation with other Mesozoic basement terranes in
California, in Howell, D.G., ed., Aspects of the Geologic History of the
California Continental Borderland: American Association of Petroleum
Geologists, Pacific Section, Miscellaneous Publications, v. 24, p. 16–56.
Hill, M.L., and Dibblee, T.W., 1953, San Andreas, Garlock, and Big Pine faults,
California—A study of the character, history, and tectonic significance
of their displacements: Geological Society of America Bulletin, v. 64,
p. 443–458, doi: 10.1130/0016-7606(1953)64[443:SAGABP]2.0.CO;2.
Hill, R.I., and Silver, L.T., 1988, San Jacinto intrusive complex: Constraints on
crustal magma chamber processes from strontium isotope heterogeneity:
Journal of Geophysical Research, v. 93, p. 10,373–10,388.
Holk, G.J., Taylor, H.P., and Gromet, P.L., 2006, Stable isotope evidence for
large-scale infiltration of metamorphic fluids generated during shallow
subduction into the eastern Peninsular Ranges mylonite zone (EPRMZ),
Southern California: International Geology Review, v. 48, p. 209–222.
Hopson, C.A., Mattinson, J.M., and Pessagno, E.A., Jr., 1981, Coast Range
Ophiolite, western California, in Ernst, W.G., ed., The Geotectonic Devel-
opment of California (Rubey Volume 1): Englewood Cliffs, New Jersey,
Prentice Hall, p. 418–510.
Howell, D.G., and Vedder, J.G., 1981, Structural implications of stratigraphic
discontinuities across the southern California continental borderland, in
358 Grove et al.
Ernst, W.G., ed., The Geotectonic Development of California (Rubey Vol-
ume 1): Englewood Cliffs, New Jersey, Prentice Hall, p. 535–559.
Humphreys, E., Hessler, E., Dueker, K., Farmer, G.L., Erslev, E., and Atwater,
T., 2003, How Laramide-age hydration of North American lithosphere
by the Farallon slab controlled subsequent activity in the Western United
States: International Geology Review, v. 45, p. 575–595.
Ichinose, G.A., Day, S., Magistrale, H., Prush, T., Vernon, F., and Edelman, A.,
1996, Crustal thickness variations beneath the Peninsular Ranges, south-
ern California: Geophysical Research Letters, v. 23, p. 3095–3098, doi:
10.1029/96GL03020.
Ingersoll, R.V., and Rumelhart, P.E., 1999, Three-stage evolution of the Los
Angeles Basin, Southern California: Geology, v. 27, p. 593–596, doi:
10.1130/0091-7613(1999)027<0593:TSEOTL>2.3.CO;2.
Irwin, W.P., 2002, Correlation of the Klamath Mountains and Sierra Nevada:
U.S. Geological Survey Open-File Report 02–490.
Jacobson, C.E., 1983, Structural geology of the Pelona Schist and Vincent
thrust, San Gabriel Mountains, California: Geological Society of Amer-
ica Bulletin, v. 94, p. 753–767, doi: 10.1130/0016-7606(1983)94<753:
SGOTPS>2.0.CO;2.
Jacobson, C.E., 1990, The 40Ar/39Ar geochronology of the Pelona Schist and
related rocks, southern California: Journal of Geophysical Research,
v. 95, p. 509–528.
Jacobson, C.E., Dawson, M.R., and Postlethwaite, C.E., 1988, Structure, meta-
morphism, and tectonic significance of the Pelona, Orocopia, and Rand
Schists, southern California, in Ernst, W.G., ed., Metamorphism and
Crustal Evolution of the Western United States (Rubey Volume 7): Engle-
wood Cliffs, New Jersey, Prentice Hall, p. 976–997.
Jacobson, C.E., Barth, A.P., and Grove, M., 2000, Late Cretaceous protolith
age and provenance of the Pelona and Orocopia Schists, southern Califor-
nia: implications for evolution of the Cordilleran margin: Geology, v. 28,
p. 219–222.
Jacobson, C.E., Grove, M., Stamp, M.M., Vucic, A., Oyarzabal, F.R., Haxel,
G.B., Tosdal, R.M., and Sherrod, D.R., 2002, Exhumation history of the
Orocopia Schist and related rocks in the Gavilan Hills area of southeast-
ernmost California, in Barth, A.P., ed., Crustal Evolution of the South-
western United States: Geological Society of America Special Paper 365,
p. 129–154.
Jacobson, C.E., Grove, M., Vucic, A., Pedrick, J., and Ebert, K.A., 2007,
Exhumation of the Orocopia Schist and associated rocks of southeastern
California: Relative roles of erosion, synsubduction tectonic denudation,
and middle Cenozoic extension, in Cloos, M., Carlson, W.D., Gilbert,
M.C., Lious, J.G., and Sorensen, S., eds., Convergent Margin Terranes
and Associated Regions: A Tribute to W.G. Ernst: Geological Society of
America Special Paper 419, p. 1–37.
Jahns, R.H., 1954, Geology of the Peninsular Ranges Province, southern Cali-
fornia and Baja California, in Jahns, R.H., ed., Geology of Southern Cali-
fornia: California Division of Mines Bulletin, v. 170, p. 29–52.
Jamieson, R.A., 1986, PT paths from high temperature shear zones beneath
ophiolites: Journal of Metamorphic Geology, v. 4, p. 3–22, doi: 10.1111/
j.1525-1314.1986.tb00335.x.
Johnson, S.E., Tate, M.C., and Fanning, C.M., 1999, New geologic mapping
and SHRIMP U-Pb zircon data in the Peninsular Ranges batholith, Baja
California, Mexico: Evidence for a suture?: Geology, v. 27, p. 743–746,
doi: 10.1130/0091-7613(1999)027<0743:NGMASU>2.3.CO;2.
Jones, D.L., Blake, M.C., Jr., and Rangin, C., 1976, The four Jurassic belts of
northern California and their significance to the geology of the south-
ern California Borderland, in Howell, D.G., ed., Aspects of the Geologic
History of the California Continental Borderland: American Association
of Petroleum Geologists, Pacific Section, Miscellaneous Publication 24,
p. 343–362.
Karig, D.E., 1980, Material transport within accretionary prisms and the
‘knocker’ problem: Journal of Geology, v. 88, p. 27–39.
Kennedy, M.P., and Moore, G.W., 1971, Stratigraphic relations of Upper Creta-
ceous and Eocene formations, San Diego coastal area, California: Ameri-
can Association of Petroleum Geologists Bulletin, v. 55, p. 709–722.
Kienast, J.R., and Rangin, C., 1982, Mesozoic blueschists and melanges
of Cedros Island (Baja California, Mexico): A consequence of nappe
emplacement or subduction?: Earth and Planetary Science Letters, v. 59,
p. 119–138, doi: 10.1016/0012-821X(82)90121-2.
Kilmer, F.J., 1979, A geological sketch of Cedros Island, Baja California, Mex-
ico, in Gastil, R.G., and Abbott, P.L., eds., Baja California Geology: San
Diego State University (field trip guidebook for the 1979 Geological Soci-
ety of America Annual Meeting, San Diego, California), p. 1–28.
Kimbrough, D.L., 1985, Tectonostratigraphic terranes of the Vizcaino Pen-
insula and Cedros and San Benitos islands, Baja California, Mexico, in
Howell, D.G., ed., Tectonostratigraphic Terranes of the Circum-Pacific
Region: Houston, Texas, Circum-Pacific Council for Energy and Mineral
Resources, Earth Science Series, v. 1, p. 285–298.
Kimbrough, D.L., and Grove, M., 2006, The eastern Peninsular Ranges batho-
lith magmatic flare-up: Insight from zircon U-Pb ages and oxygen iso-
tope ratios: Backbone of the Americas, Patagonia to Alaska, Geological
Society of America Specialty Meetings Abstracts with Programs, no. 2,
p. 107.
Kimbrough, D.L., and Moore, T.E., 2003, Ophiolite and volcanic arc assem-
blages on the Vizcaino Peninsula and Cedros Island, Baja California Sur,
México: Mesozoic forearc lithosphere of the Cordilleran magmatic arc,
in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p. 43–71.
Kimbrough, D.L., Smith, D.P., Mahoney, B.J., Moore, T.E., Grove, M., Gastil,
R.G., Ortega-Rivera, A., and Fanning, M.C., 2001, Forearc-basin sedimen-
tary response to rapid Late Cretaceous batholith emplacement in the Pen-
insular Ranges of southern and Baja California: Geology, v. 29, p. 491–
494, doi: 10.1130/0091-7613(2001)029<0491:FBSRTR>2.0.CO;2.
King, R.L., Bebout, G.E., Moriguti, T., and Nakamura, E., 2006, Elemental
mixing systematics and Sr-Nd isotope geochemistry of mélange for-
mation: Obstacles to identification of fluid sources to arc volcanics:
Earth and Planetary Science Letters, v. 246, p. 288–304, doi: 10.1016/
j.epsl.2006.03.053.
King, R.L., Bebout, G.E., Grove, M., Moriguti, T., Nakamura, E., 2007, Boron
and lead isotope signatures of subduction-zone mélange formation:
Hybridization and fractionation along the slab-mantle interface beneath
volcanic arcs: Chemical Geology, doi: 1016/j.chemgeo.2007.01.009.
Kistler, R.W., Wooden, J.L., and Morton, D.M., 2003, Isotopes and Ages in the
Northern Peninsular Ranges Batholith, Southern California: U.S. Geo-
logical Survey Open-File Report 03–489, 45 p.
Kröner, A., Jaeckel, P., and Williams, I.S., 1994, Pb-loss patterns in zircons
from a high-grade metamorphic terrain as revealed by different dating
methods: U–Pb and Pb–Pb ages for igneous and metamorphic zircons
from northern Sri Lanka: Precambrian Research, v. 66, p. 151–181, doi:
10.1016/0301-9268(94)90049-3.
Krummenacher, D., Gastil, R.G., Bushee, J., and Dupont, J., 1975, K-Ar appar-
ent ages, Peninsular Ranges batholith, southern California and Baja Cali-
fornia: Geological Society of America Bulletin, v. 86, p. 760–768, doi:
10.1130/0016-7606(1975)86<760:KAAPRB>2.0.CO;2.
Langenheim, V.E., and Jachens, R.C., 2003, Crustal structure of the Penin-
sular Ranges batholith from magnetic data: Implications for Gulf of
California rifting: Geophysical Research Letters, vol. 30, no. 11, doi:
10.1029/2003GL017159.
Larsen, E.S., 1948, Batholith and Associated Rocks of Corona, Elsinore, and
San Luis Rey Quadrangles, Southern California: Geological Society of
America Memoir 29, 182 p.
Lewis, J.L., Day, S.M., Magistrale, H., Castro, R.R., Astiz, L., Rebollar, C.,
Eakins, J., Vernon, F.L., and Brune, J.N., 2001, Crustal thickness of the
Peninsular Ranges and Gulf Extensional Province in the Californias: Jour-
nal of Geophysical Research, p. 13,599–13,612, doi: 2001JB000178.
Lovera, O.M., Grove, M., Kimbrough, D.L., and Abbott, P.L., 1999, Magni-
tude and time scales in the exhumation of arc crust: An approach based
upon analysis of detrital closure age distributions: Journal of Geophysical
Research, v. 104, p. 29,419–29,438, doi: 10.1029/1999JB900082.
Ludwig, K.R., 2003, ISOPLOT/EX Version 3: Berkeley, California, Berkeley
Geochronology Center Special Publication 4.
Mahoney, J.B., Kimbrough, D.L., Grove, M., Jacobson, C.E., and Mustard,
P.S., 2005, Variations in Late Cretaceous batholith exhumation patterns
along the western North American margin: Geological Society of Amer-
ica Abstracts with Programs, v. 37, no. 7, p. 482.
Malin, P.E., Goodman, E.D., Henyey, T.L., Li, Y.G., Okaya, D.A., and Saleeby,
J.B., 1995, Significance of seismic reflections beneath a tilted exposure of
deep continental crust, Tehachapi Mountains, California: Journal of Geo-
physical Research, v. 100, p. 2069–2087, doi: 10.1029/94JB02127.
Mattinson, J.M., 1986, Geochronology of high-pressure–low-temperature
Franciscan metabasites: A new approach using the U-Pb system, in Evans,
 The Catalina Schist
B.W., and Brown, E.H., eds., Blueschists and Eclogites: Geological Soci-
ety of America Memoir 164, p. 95–106.
Mattinson, J.M., and Hill, D.J., 1976, Age of plutonic basement rocks, Santa
Cruz Island, in Howell, D.G., ed., Aspects of the Geologic History of the
California Continental Borderland: American Association of Petroleum
Geologists, Pacific Section, Miscellaneous Publications, v. 24, p. 53–58.
McDowell, F.W., Roldán-Quintana, J., and Connelly, J.N., 2001, Duration
of Late Cretaceous–early Tertiary magmatism in east-central Sonora,
Mexico: Geological Society of America Bulletin, v. 113, p. 521–531, doi:
10.1130/0016-7606(2001)113<0521:DOLCET>2.0.CO;2.
McWilliams, M.O., and Howell, D.G., 1982, Exotic terranes of western Cali-
fornia: Nature, v. 297, p. 215–217, doi: 10.1038/297215a0.
Moore, T.E., 1985, Stratigraphic and tectonic significance of the Mesozoic
tectonostratigraphic terranes of the Vizcaíno peninsula, Baja California
Sur, Mexico, in Howell, D.G., ed., Tectonostratigraphic Terranes of the
Circum-Pacific Region: Houston, Circum-Pacific Council for Energy and
Mineral Resources, p. 315–329.
Moore, T.E., 1986, Petrology and tectonic implications of the blueschist-bear-
ing Puerto Nuevo mélange complex, Vizcaino Peninsula, Baja Califor-
nia Sur, Mexico, in Evans, B.W., and Brown, E.H., eds., Blueschists and
Eclogites: Geological Society of America Memoir 164, p. 43–58.
Morgan, J.R., Kimbrough, D.L., and Grove, M., 2005, Detrital U/Pb zircon
ages from the Peninsular Ranges Mesozoic flysch belt of southern and
Baja California: Peninsula Geological Society, 7th International Meeting,
Abstracts with Programs, p. 20.
Nicholson, C., Sorlien, C.C., Atwater, T., Crowell, J.C., and Luyendyk, B.P.,
1994, Microplate capture, rotation of the western Transverse Ranges,
and initiation of the San Andreas transform as a low-angle fault system:
Geology, v. 22, p. 491–495, doi: 10.1130/0091-7613(1994)022<0491:
MCROTW>2.3.CO;2.
Nilsen, T.H., and Abbott, P.L., 1981, Paleogeography and sedimentology of
Upper Cretaceous turbidites, San Diego, California: American Associa-
tion of Petroleum Geologists Bulletin, v. 65, p. 1256–1284.
Nordstrom, C.E., 1970, Lusardi Formation—A post-batholithic Cretaceous con-
glomerate north of San Diego, California: Geological Society of Amer-
ica Bulletin, v. 81, p. 601–605, doi: 10.1130/0016-7606(1970)81[601:
LFAPCC]2.0.CO;2.
Ortega-Rivera, A., 2003, Geochronological constraints on the tectonic history
of the Peninsular Ranges Batholith of Alta and Baja California: Tectonic
implications for western México, in Johnson, S.E., et al., eds., Tectonic
Evolution of Northwestern Mexico and the Southwestern USA: Geologi-
cal Society of America Special Paper 374, p. 297–335.
Page, B.M., 1970, Sur-Nacimiento fault zone of California: Continental margin
tectonics: Geological Society of America Bulletin, v. 81, p. 667–690, doi:
10.1130/0016-7606(1970)81[667:SFZOCC]2.0.CO;2.
Page, B.M., 1981, The southern Coast Ranges, in Ernst, W.G., ed., The Geo-
tectonic Development of California (Rubey Volume 1): Englewood Cliffs,
New Jersey, Prentice Hall, p. 329–417.
Page, B.M., 1982, Migration of Salinian composite block, California, and dis-
appearance of fragments: American Journal of Science, v. 282, p. 1694–
1734.
Patterson, D.L., 1984, Los Chapunes and Valle sandstones: Cretaceous petrofa-
cies of the Vizcaino basin, Baja California, Mexico, in Frizzell, V.A., Jr.,
ed., Geology of the Baja California Peninsula: Society of Economic Pale-
ontologists and Mineralogists, Pacific Section, v. 39, p. 161–172.
Peacock, S.M., 1987, Creation and preservation of subduction-related metamor-
phic gradients: Journal of Geophysical Research, v. 92, p. 12,763–12,781.
Peacock, S.M., 1992, Blueschist-facies metamorphism, shear heating, and P-T-t
paths in subduction shear zones: Journal of Geophysical Research, v. 97,
p. 17,693–17,707.
Peacock, S.M., 1993, Large-scale hydration of the lithosphere above sub-
ducting slabs: Chemical Geology, v. 108, p. 49–59, doi: 10.1016/0009-
2541(93)90317-C.
Peacock, S.M., Rushmer, T., and Thompson, A.B., 1994, Partial melting of
subducting oceanic crust: Earth and Planetary Science Letters, v. 121,
p. 227–244, doi: 10.1016/0012-821X(94)90042-6.
Peterson, G.L., and Nordstrom, C.E., 1970, Sub–La Jolla unconformity in
vicinity of San Diego, California: American Association of Petroleum
Geologists Bulletin, v. 54, p. 265–274.
Pickett, D.A., and Saleeby, J.B., 1993, Thermobarometric constraints on the
depth of the exposure and conditions of plutonism and metamorphism at
359
deep levels of the Sierra Nevada batholith, Tehachapi Mountains, Califor-
nia: Journal of Geophysical Research, v. 98, p. 609–629.
Platt, J.P., 1975, Metamorphic and deformational processes in the Francis-
can Complex, California: Some insights from the Catalina Schist ter-
rane: Geological Society of America Bulletin, v. 86, p. 1337–1347, doi:
10.1130/0016-7606(1975)86<1337:MADPIT>2.0.CO;2.
Platt, J.P., 1976, The petrology, structure, and geologic history of the Catalina
Schist terrain, southern California: University of California Publications
in Geological Sciences, v. 112, 111 p.
Platt, J.P., 1986, Dynamics of orogenic wedges and the uplift of high-pressure
metamorphic rocks: Geological Society of America Bulletin, v. 97, p. 1037–
1053, doi: 10.1130/0016-7606(1986)97<1037:DOOWAT>2.0.CO;2.
Rothstein, D.A., and Manning, C.E., 2003, Geothermal gradients in continental
magmatic arcs; constraints from the eastern Peninsular Ranges Batholith,
Baja California, Mexico, in Johnson, S.E., et al., eds., Tectonic Evolution
of Northwestern Mexico and the Southwestern USA: Geological Society
of America Special Paper 374, p. 337–354.
Rubatto, D., 2002, Zircon trace element geochemistry, partitioning with garnet
and the link between U-Pb ages and metamorphism: Chemical Geology,
v. 184, p. 123–138, doi: 10.1016/S0009-2541(01)00355-2.
Saleeby, J.B., 2003, Segmentation of the Laramide Slab—Evidence from the
southern Sierra Nevada region: Geological Society of America Bul-
letin, v. 115, p. 655–668, doi: 10.1130/0016-7606(2003)115<0655:
SOTLSF>2.0.CO;2.
Saleeby, J.B., and Sharp, W.D., 1980, Chronology of the structural and pet-
rologic development of the southwest Sierra Nevada foothills, Califor-
nia: Geological Society of America Bulletin, v. 91, p. 317–320, doi:
10.1130/0016-7606(1980)91<317:COTSAP>2.0.CO;2.
Schmidt, K.L., and Paterson, S.R., 2002, A doubly vergent, fan structure in the
Peninsular Ranges batholith: Transpression or complex flow along a crustal-
scale discontinuity?: Tectonics, v. 21, doi: 10.1029/2001TC001353.
Schmidt, K.L., Wetmore, P.H., Johnson, S.E., and Paterson, S.R., 2002, Along-
strike variation in the origin of an ocean-continent basement join in the
Mesozoic Peninsular Ranges batholith: To what degree does an island arc
collision drive orogeny?, in Barth, A., ed., Crustal Evolution of the South-
western United States: Geological Society of America Special Paper 365,
p. 49–72.
Schoellhamer, J.E., and Woodford, A.O., 1951, The floor of the Los Angeles
basin, Los Angeles, Orange, and San Bernardino Counties, California:
U.S. Geological Survey Oil and Gas Investigations Map OM-117.
Schoellhamer, J.E., Vedder, J.G., Yerkes, R.F., and Kinney, D.M., 1981, Geol-
ogy of the Northern Santa Ana Mountains, California: U.S. Geological
Survey Professional Paper 420-D, 109 p.
Scholl, D.W., von Huene, R., Vallier, T.L., and Howell, D.G., 1980, Sedimentary
masses and concepts about tectonic processes at underthrust ocean mar-
gins: Geology, v. 8, p. 564–568, doi: 10.1130/0091-7613(1980)8<564:
SMACAT>2.0.CO;2.
Scholl, D.W., and von Huene, R., 2007, Crustal recycling at modern subduction
zones applied to the past—Issues of growth and preservation of continen-
tal basement, mantle geochemistry, and supercontinent reconstruction, in
Hatcher, R.D., Jr., Carlson, M.P., McBride, J.H., and Martínez Catalán,
J.R., eds., The 4D Framework of Continental Crust: Geological Society
of America Memoir 200, p. 9–32.
Sedlock, R.L., 1988a, Metamorphic petrology of a high-pressure, low-tem-
perature subduction complex in west-central Baja California, Mexico:
Journal of Metamorphic Geology, v. 6, p. 205–233, doi: 10.1111/j.1525-
1314.1988.tb00416.x.
Sedlock, R.L., 1988b, Tectonic setting of blueschist and island-arc terranes of
west central Baja California, Mexico: Geology, v. 16, p. 623–626, doi:
10.1130/0091-7613(1988)016<0623:TSOBAI>2.3.CO;2.
Shaw, S.E., Todd, V.R., and Grove, M., 2003, Jurassic peraluminous gneissic
granites in the axial zone of the Peninsular Ranges, southern California,
in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p.157–183.
Shervais, J.W., Murchey, B.L., Kimbrough, D.L., Renne, P.R., and Hanan, B.,
2005, Radioisotopic and biostratigraphic age relations in the Coast Range
Ophiolite, northern California: Implications for the tectonic evolution of
the Western Cordillera: Geological Society of America Bulletin, v. 117,
p. 633–653, doi: 10.1130/B25443.1.
360 Grove et al.
Silver, L.T., and Chappell, B.W., 1988, The Peninsular Ranges batholith: An
insight into the evolution of the Cordilleran batholiths of southwestern
North America: Royal Society of Edinburgh Transactions: Earth Sci-
ences, v. 79, p. 105–121.
Smith, D.P., and Busby, C.J., 1993, Mid-Cretaceous crust extension recorded
in deep marine half-graben fill, Cedros Island, Mexico: Geological
Society of America Bulletin, v. 105, p. 547–562, doi: 10.1130/0016-
7606(1993)105<0547:MCCERI>2.3.CO;2.
Sorensen, S.S., 1985, Petrologic evidence for Jurassic island arc-like basement
rocks in the southwestern Tranverse ranges and California continental
borderland: Geological Society of America Bulletin, v. 96, p. 997–1006,
doi: 10.1130/0016-7606(1985)96<997:PEFJIB>2.0.CO;2.
Sorensen, S.S., 1986, Petrologic and geochemical comparison of the blueschist
and greenschist units of the Catalina Schist terrane, southern California,
in Evans, B.W., and Brown, E.H., eds., Blueschists and Eclogites: Geo-
logical Society of America Memoir 164, p. 59–75.
Sorensen, S.S., 1988a, Tectonometamorphic significance of the basement rocks
of the Los Angeles Basin and the California Continental Borderland, in
Ernst, W.G., ed., Metamorphism and Crustal Evolution of the Western
United States (Rubey Volume 7), p. 998–1022.
Sorensen, S.S., 1988b, Petrology of amphibolite-facies mafic and ultramafic
rocks from the Catalina Schist, southern California: Metasomatism and
migmatization in a subduction zone metamorphic setting: Journal of
Metamorphic Geology, v. 6, p. 405–435, doi: 10.1111/j.1525-1314.1988.
tb00431.x.
Sorensen, S.S., and Barton, M.D., 1987, Metasomatism and partial melting in a
subduction complex: Catalina Schist, southern California: Geology, v. 15,
p. 115–118, doi: 10.1130/0091-7613(1987)15<115:MAPMIA>2.0.CO;2.
Sorensen, S.S., and Grossman, J.N., 1989, Enrichment of trace elements in gar-
net amphibolites from a paleo-subduction zone: Catalina Schist, southern
California: Geochimica et Cosmochimica Acta, v. 53, p. 3155–3177, doi:
10.1016/0016-7037(89)90096-3.
Staude, J.M., and Barton, M.D., 2001, Jurassic to Holocene tectonics, mag-
matism, and metallogeny of northwestern Mexico: Geological Soci-
ety of America Bulletin, v. 113, p. 1357–1374, doi: 10.1130/0016-
7606(2001)113<1357:JTHTMA>2.0.CO;2.
Stern, R.J., and Bloomer, S.H., 1992, Subduction zone infancy: Examples from
the Eocene Izu-Bonin-Mariana and Jurassic California arcs: Geological
Society of America Bulletin, v. 104, p. 1621–1636, doi: 10.1130/0016-
7606(1992)104<1621:SZIEFT>2.3.CO;2.
Stern, T., Bateman, P.C., Morgan, B.A., Newell, M.F., and Peck, D.L., 1981, Iso-
topic U-Pb Ages of Zircon from the Granitoids of the Central Sierra Nevada,
California: U.S. Geological Survey Professional Paper 1185, 17 p.
Stock, J., and Molnar, P., 1988, Uncertainties and implications of the Late Cre-
taceous and Tertiary position of North America relative to the Farallon,
Kula, and Pacific plates: Tectonics, v. 7, p. 1339–1384.
Stuart, C.J., 1979, Middle Miocene paleogeography of coastal southern Cali-
fornia and the California borderland: Evidence from schist-bearing sedi-
mentary rocks, in Armentrout, J.M., et al., eds., Cenozoic Paleogeography
of the Western United States: Society of Economic Paleontologists and
Mineralogists, Pacific Section, Field Guide 3, p. 29–44.
Sundberg, F.A., and Cooper, J.D., 1978, Late Cretaceous depositional envi-
ronments, northern Santa Ana Mountains, Southern California, in How-
ell, D.G., and McDougall, K.A., eds., Mesozoic Paleogeography of the
Western United States: Society of Economic Paleontologists and Min-
eralogists, Pacific Section, Pacific Coast Paleogeography Symposium 2,
p. 535–546.
Suppe, J., 1970, Offset of late Mesozoic basement terrains by the San Andreas
fault system: Geological Society of America Bulletin, v. 81, p. 3253–
3258, doi: 10.1130/0016-7606(1970)81[3253:OOLMBT]2.0.CO;2.
Suppe, J., and Armstrong, R.L., 1972, Potassium-argon dating of Franciscan
metamorphic rocks: American Journal of Science, v. 272, p. 217–233.
Taylor, H.P., 1986, Igneous rocks: Isotopic case studies of Circum-Pacific mag-
matism, in Valley, J.W., et al., eds., Stable Isotopes in High-Temperature
Geological Processes: Mineralogical Society of America, p. 273–317.
ten Brink, U.S., Zhang, J., Brocher, T.M., Okaya, D.A., Klitgord, K.D., and Fuis,
G.S., 2000, Geophysical evidence for the evolution of the California inner
continental borderland as a metamorphic core complex: Journal of Geo-
physical Research, v. 105, p. 5835–5857, doi: 10.1029/1999JB900318.
Thomson, C.N., and Girty, G.H., 1994, Early Cretaceous intra-arc ductile
strain in Triassic–Jurassic and Cretaceous continental margin arc rocks,
Peninsular Ranges, California: Tectonics, v. 13, p. 1108–1119, doi:
10.1029/94TC01649.
Todd, V.R., and Shaw, S.E., 1985, S-type granitoids and an I-S line in the Pen-
insular Ranges batholith, southern California: Geology, v. 13, p. 231–233,
doi: 10.1130/0091-7613(1985)13<231:SGAAIL>2.0.CO;2.
Todd, V.R., Erskine, B.G., and Morton, D.M., 1988, Metamorphic and tectonic
evolution of the northern Peninsular Ranges Batholith, southern Cali-
fornia, in Ernst, W.G., ed., Metamorphism and Crustal Evolution of the
Western United States (Rubey Volume 7): Englewood Cliffs, New Jersey,
Prentice Hall, p. 894–937.
Todd, V.R., Shaw, S.E., and Hammarstrom, J.M., 2003, Cretaceous plutons of
the Peninsular Ranges Batholith, San Diego and westernmost Imperial
Counties, California: Intrusion across a Late Jurassic continental margin,
in Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p. 185–235.
Tulloch, A.J., and Kimbrough, D.L., 2003, Paired plutonic belts in convergent
margins and the development of high Sr/Y magmatism: Peninsular Ranges
Batholith of Baja California and Median Batholith of New Zealand, in
Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p. 275–295.
Usui, T., Nakamura, E., Kobayashi, K., Maruyama, S., and Helmstaedt, H.,
2003, Fate of the subducted Farallon plate inferred from eclogite xenoliths
in the Colorado Plateau: Geology, v. 31, p. 589–592, doi: 10.1130/0091-
7613(2003)031<0589:FOTSFP>2.0.CO;2.
Vedder, J.G., 1987, Regional geology and petroleum potential of the southern
California Borderland, in Scholl, D.W., et al., eds., Geology and Resource
Potential of the Continental Margin of Western North America and Adja-
cent Ocean Basins: Beaufort Sea to Baja California: Houston, Circum-
Pacific Council for Energy and Mineral Resources, p. 403–447.
Vedder, J.G., Howell, D.G., and McLean, H., 1983, Stratigraphy, sedimenta-
tion, and tectonic accretion of exotic terranes, southern Coast Ranges,
California, in Watkins, J.S., and Drake, C.L., eds., Studies in Continental
Margin Geology: American Association of Petroleum Geologists Memoir
34, p. 471–496.
von Huene, R., and Scholl, D.W., 1991, Observations at convergent margins
concerning sediment subduction, subduction erosion, and the growth of
continental crust: Reviews of Geophysics, v. 29, p. 279–316.
von Huene, R., and Scholl, D.W., 1993, The return of sialic material to the mantle
indicated by terrigeneous material subducted at convergent margins: Tec-
tonophysics, v. 219, p. 163–175, doi: 10.1016/0040-1951(93)90294-T.
Wakabayashi, J., 1990, Counterclockwise P-T-t paths from amphibolites, Fran-
ciscan Complex, California: Relics from the early stages of subduction
zone metamorphism: Journal of Geology, v. 98, p. 657–680.
Wakabayashi, J., 1992, Nappes, tectonics of oblique plate convergence, and met-
amorphic evolution related to 140 million years of continuous subduction,
Franciscan Complex, California: Journal of Geology, v. 100, p. 19–40.
Wakabayashi, J., 1999, Subduction and the rock record: Concepts developed
in the Franciscan Complex, California, in Sloan, D., et al., eds., Clas-
sic Cordilleran Concepts: A View from California: Geological Society of
America Special Paper 338, p. 123–133.
Wakabayashi, J., and Dilek, K., 2000, Spatial and temporal relations between
ophiolites and their subophiolitic soles: A test of models of forearc ophio-
lite genesis, in Dilek, Y., et al., eds., Ophiolites and Oceanic Crust: New
Insights from Field Studies and Ocean Drilling: Geological Society of
America Special Paper 349, p. 53–64.
Wakabayashi, J., and Dumitru, T.A., 2007, 40Ar/39Ar ages from coherent high-
pressure metamorphic rocks of the Franciscan Complex, California:
Revisiting the timing of metamorphism of the world’s type subduction
complex: International Geology Review, v. 49, no. 10, p. 873–906.
Walawender, M.J., Gastil, R.G., Clinkenbeard, J.P., McCormick, W.V., East-
man, B.G., Wernicke, R.S., Wardlaw, M.S., Gunn, S.H., and Smith, B.M.,
1990, Origin and evolution of the zoned La Posta–type plutons, eastern
Peninsular Ranges batholith, southern and Baja California, in Anderson,
J.L., ed., The Nature and Origin of Cordilleran Magmatism: Geological
Society of America Memoir 174, p. 1–18.
Weaver, D.W., and Nolf, B., 1969, Geology of Santa Cruz Island (map), in
Weaver, D.W., ed., Geology of the Northern Channel Islands, Southern
California Borderland: American Association of Petroleum Geologists,
Pacific Section, Special Publication, scale 1:24,000.
 The Catalina Schist
Wetmore, P.H., Schmidt, K.L., Paterson, S.R., and Herzig, C., 2002, Tectonic
implications for the along-strike variation of the Peninsular Ranges
batholith, southern and Baja California: Geology, v. 30, p. 247–250, doi:
10.1130/0091-7613(2002)030<0247:TIFTAS>2.0.CO;2.
Wetmore, P.H., Herzig, C., Schultz, P.W., Alsleben, H., Paterson, S.R., and
Schmidt, K.L., 2003, Cretaceous tectonic evolution of the western zone
of the Peninsular Ranges batholith of southern and Baja California, in
Johnson, S.E., et al., eds., Tectonic Evolution of Northwestern Mexico
and the Southwestern USA: Geological Society of America Special Paper
374, p. 93–116.
Williams, I.S., and Claesson, S., 1987, Isotopic evidence for the Precambrian
provenance and Caledonian metamorphism of high grade paragneisses
from the Seve Nappes, Scandinavian Caledonides: Contributions to Min-
eralogy and Petrology, v. 97, p. 205–217, doi: 10.1007/BF00371240.
Wood, D.J., and Saleeby, J.B., 1997, Late Cretaceous–Paleocene extensional
collapse and disaggregation of the southernmost Sierra Nevada batholith:
International Geology Review, v. 39, p. 973–1009.
Wooden, J.L., and Miller, D.M., 1990, Chronologic and isotopic framework for
Early Proterozoic evolution of the eastern Mojave Desert region, SE Cali-
fornia: Journal of Geophysical Research, v. 95, p. 20133–20146.
Woodford, A.O., 1924, The Catalina Metamorphic Facies of the Franciscan
series: University of California Publications in Geological Sciences,
v. 15, p. 49–68.
Woodring, W.P., and Popenoe, W.P., 1942, Upper Cretaceous formations and
faunas of southern California: American Association of Petroleum Geolo-
gists Bulletin, v. 26, p. 166–176.
Woodring, W.P., Bramlette, M.N., and Kew, W.S.W., 1946, Geology and Pale-
ontology of the Palos Verdes Hills, California: U.S. Geological Survey
Professional Paper 207, 145 p.
361
Wright, J.E., and Fahan, M.R., 1988, An expanded view of Jurassic orogenesis
in the western United States Cordillera: Middle Jurassic (pre-Nevadan)
regional metamorphism and thrust faulting within an active arc environ-
ment, Klamath Mountains, California: Geological Society of America
Bulletin, v. 100, p. 859–876, doi: 10.1130/0016-7606(1988)100<0859:
AEVOJO>2.3.CO;2.
Wright, T.L., 1991, Structural geology and tectonic evolution of the Los Ange-
les Basin, California, in Biddle, K.T., ed., Active Margin Basins: Ameri-
can Association of Petroleum Geologists Memoir 52, p. 35–134.
Yeats, R.S., 1968, Southern California structure, seafloor spreading, and history
of the Pacific Basin: Geological Society of America Bulletin, v. 79, p. 1693–
1702, doi: 10.1130/0016-7606(1968)79[1693:SCSSSA]2.0.CO;2.
Yeats, R.S., 1973, Newport-Inglewood fault zone, Los Angeles basin, Cali-
fornia: American Association of Petroleum Geologists Bulletin, v. 57,
p. 117–135.
Yerkes, R.F., McCulloh, T.H., Schoellhamer, J.E., and Vedder, J.G., 1965, Geol-
ogy of the Los Angeles Basin, California—An Introduction: U.S. Geo-
logical Survey Professional Paper 420-A, 57 p.
Zandt, G., Gilbert, H., Owens, T.J., Ducea, M., Saleeby, J., and Jones, C.H.,
2004, Active foundering of a continental arc root beneath the southern
Sierra Nevada in California: Nature, v. 431, p. 41–46, doi: 10.1038/
nature02847.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Sedimentary response to arc-continent collision, Permian, southern Mongolia

C.L. Johnson*
Geology and Geophysics, WBB 609, University of Utah, 135 South 1460 East, Salt Lake City, Utah 84112, USA

J.A. Amory
Salym Petroleum Development, Griboedova Street #2, 625000 Tyumen, Russia

D. Zinniker
Geology and Geophysics, Yale University, P.O. Box 2080109, New Haven, Connecticut 06520, USA

M.A. Lamb
Geology Department, OWS 153, University of St. Thomas, 2115 Summit Avenue, St. Paul, Minnesota 55105, USA

S.A. Graham
Geological and Environmental Sciences Department, Stanford University, 450 Serra Mall (Building 320) #118, Stanford,
California 94305, USA

M. Affolter
Geology and Geophysics, WBB 609, University of Utah, 135 South 1460 East, Salt Lake City, Utah 84112, USA

G. Badarch†
Institute of Geology and Mineral Resources, Mongolian Academy of Sciences, Ulaanbaatar 210351, Mongolia

ABSTRACT

The Eurasian Tien Shan–Yin Shan suture is a ~3000-km-long boundary between

Paleozoic arc and accretionary complexes (the Altaids) and Precambrian microcon-
tinental blocks (Tarim and North China block). Stratigraphic data are presented
from localities in southern Mongolia spanning >800 km along the northern margin
of the suture. Facies descriptions, climatic indicators, sandstone provenance, and
paleocurrent data help reconstruct Permian basin evolution during and following
arc-continent collision, and results are integrated with previously published data to
create a preliminary regional synthesis. Upper Permian strata of southern Mongolia
comprise fluvial successions in the southwest and marine turbidite deposits in the
southeast. Floral assemblages show mixing of Siberian craton and North China block
communities, indicating their close proximity to Mongolia by Permian time. There
was a rapid transition from humid environments in the Late Permian to more arid

*cjohnson@earth.utah.edu
†
Deceased.

Johnson, C.L., Amory, J.A., Zinniker, D., Lamb, M.A., Graham, S.A., Affolter, M., and Badarch, G., 2008, Sedimentary response to arc-continent collision,
Permian, southern Mongolia, in Draut, A.E., Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record in Arc Collision Zones:
Geological Society of America Special Paper 436, p. 363–390, doi: 10.1130/2008.2436(16). For permission to copy, contact editing@geosociety.org. ©2008 The
Geological Society of America. All rights reserved.

363
364 Johnson et al.

conditions in the Early Triassic, corresponding to the global Permian–Triassic bound-

ary event, but which may also reflect more local driving mechanisms such as rain
shadow effects. Permian sandstones from Mongolia have undissected to dissected arc
provenance, with little contribution from continental or recycled orogen sources. The
timing of the nonmarine-marine facies transition and cessation of arc magmatism
broadly supports earlier collision along the western part of the suture zone than along
the eastern part (e.g., Late Carboniferous–Late Permian). However, when regional
geologic constraints are integrated, a more complex model involving differential rota-
tion of Tarim and the North China block is preferred. Late Paleozoic rocks of southern
Mongolia have been subsequently dismembered along Mesozoic–Cenozoic strike-slip
faults, and thus also represent the long-term record of intracontinental deformation
within accreted, heterogeneous crust.

Keywords: collisional tectonics, arc accretion, palynology, sedimentary basins,

Mongolia.

INTRODUCTION the timing, mechanisms, and sedimentary record of Mesozoic and

Continental growth via accretion is well illustrated by the
Cenozoic India-Eurasia collision (Harrison et al., 1992; Rowley,
1996). This phenomenon broadly represents the entire Phanero-
zoic history of the region, as demonstrated by a series of generally
southward-younging suture zones extending from the Siberian
craton through Mongolia and China (Zhang et al., 1984; Fig. 1).
Such rapid continental growth not only highlights an exceptional
protracted record of terrane accretion but also illustrates sub-
sequent intracontinental deformation along inherited zones of
crustal weakness, even far inboard of the active margins (Tappon-
nier et al., 1982; Burchfiel et al., 1989). Multiple studies highlight
younger collisional events in this region (Graham et al., 1993;
Graham, 1996; Yin and Nie, 1996). However, the suture zone
that formed the main nucleation point for Mesozoic accretion
remains poorly understood; specifically, the boundary between
Paleozoic arc terranes that lie south of the Siberian craton and
tectonic blocks in China that are floored mainly by Precambrian
continental crust (e.g., North China block and Tarim–Central Tien
Shan blocks; Watson et al., 1987). This accretionary complex (the
Altaid tectonic collage or “Altaids” after Şengör and Natal’in,
1996, Fig. 1; cf. the Central Asian Orogenic Belt of Jahn, 1999) is
well exposed in the study area of southern Mongolia. As such, this
region constitutes a significant portion of one of the largest zones

Figure 1. Map of major tectonic units,

boundaries, and fault zones in east-
central Asia. ATF—Altyn Tagh fault;
BS—Bogda Shan; CTS—Central Tien
Shan; EGFZ—East Gobi fault zone;
JH—Junggar-Hegen suture; NTS—
North Tien Shan; QFS—Qilian fault sys-
tem; SIB—Siberian craton; THB—Tur-
pan-Hami basin; TLF—Tan Lu fault;
UB—Ulaanbaatar (capital of Mongolia).
 Sedimentary response to arc-continent collision
of arc accretion on earth (Şengör et al., 1993), so there is much to
learn about arc-arc collision as well as terminal arc-continent col-
lision, the subject of this paper. Few integrated studies of Permian
rocks in southern Mongolia are available, so this study is a first
step toward documenting that record.
The boundary between the Altaids and Precambrian micro-
continental blocks of north China has been variously named the
Suolon, Solonker, Junggar-Hegen, South Mongolia–Hinggan,
and Tien Shan–Yin Shan sutures (Şengör et al., 1993; Wang and
Mo, 1995; Badarch et al., 2002; Cope et al., 2005; Shi, 2006).
The term Tien Shan–Yin Shan suture is favored because it implies
a regional continuation of the collision zone from northwest to
northeast China (Fig. 1; Yin and Nie, 1996; Xiao et al., 2004),
a correlation that is supported by this study. The geology of this
region is poorly defined on many levels beyond these problem-
atic naming conventions. The latitude and extent of the suture
zone relative to the China-Mongolia political border is debated
(Xiao et al., 2003). Assuming that collision involved a combined
Tarim–North China block moving north relative to the Altaids
(this assumption is discussed more fully below; Enkin et al.,
1992; Ziegler et al., 1996), the Tien Shan–Yin Shan suture spans
>3000 km from northwest to northeast China, making it one of
the longest and oldest tectonic boundaries within the amalgam-
ated crustal material south of the Siberian craton.
Formation of this suture during closure of the Paleo-Asian
ocean (Dobretsov et al., 1994) is generally thought to have been
complete by Late Permian to Early Triassic time, but alternate
interpretations range from Devonian to Cretaceous (reviewed in
Xiao et al., 2003). Timing relationships across the region have
also been interpreted as either broadly synchronous or highly dia-
chronous collision from west to east (Amory, 1996; Yin and Nie,
1996). Although many studies provide important constraints for
parts of the suture zone (particularly in northwest China; Cole-
man, 1989; Windley et al., 1990; Laurent-Charvet et al., 2003),
in general the paleogeography of the entire closing ocean basin
is poorly understood in terms of polarity of subduction, and the
number and nature of magmatic arcs that may have been active
(Lamb and Badarch, 2001; Zhou et al., 2001; Xiao et al., 2004).
Without a better understanding of the location, timing, and
mechanisms associated with this event, it is difficult to evaluate
implications such as whether postorogenic collapse could have
been active in the Permian (Tang, 1990) or later into the Trias-
sic (Davis et al., 2004), whether later intracontinental deforma-
tion has been concentrated along major crustal boundaries, and
whether regional climate change in the Permian of China can be
attributed to a rain shadow effect created by collisional orogens
(Cope et al., 2005).
This study reviews known, inferred, and newly constrained
relationships relevant to the tectonic and stratigraphic history on
either side of the nominal collision zone in order to extract a new
model regarding its formation. Reconstruction of the Tien Shan–
Yin Shan suture is aided by decades of research in northern China.
We first review these studies, with an emphasis on available basin
analysis constraints, to provide a regional tectonic context and to
365
highlight outstanding questions related to regional tectonic evo-
lution. Complementary data are then presented from seven main
localities spanning >800 km from southwestern to southeastern
Mongolia (Table 1; Fig. 2). New data and observations are derived
primarily from Upper Permian accretionary and postaccretionary
sedimentary complexes. Considered in the regional context, these
data help constrain the timing, mechanisms, and implications of
this early phase of continental growth in Asia. Similarly, the arc
setting of southern Mongolia in the middle to late Paleozoic has
been compared to the modern southwest Pacific Ocean (Lamb
and Badarch, 2001), and additional field constraints permit eval-
uating this analogy and various models that describe the strati-
graphic record of interarc and arc-continent collisions (Ingersoll
et al., 1995; Huang et al., 2006).
LATE PALEOZOIC SEDIMENTATION AND
TECTONICS OF NORTH CHINA
The southern margin of the Altaids is inferred to follow
either a strongly arcuate shape roughly parallel to the modern
political border, or alternatively a more continuous east-west–
trending boundary across northern China (Fig. 1, Junggar-
Hegen versus Tien Shan–Yin Shan sutures; Zhang et al., 1984;
Yin and Nie, 1996). Tarim and the North China block represent
continental margins with Precambrian crust, forming distinctive
southern boundaries to the accretionary complex. In contrast,
the Altaid tectonic collage primarily comprises late Paleozoic
subduction-accretion complexes with probably only small and
isolated microcontinent blocks (Şengör et al., 1993). Confusion
over the location of the suture that marks final closure of the
Paleo-Asian ocean is exacerbated by this rather diffuse northern
margin, which consists of several magmatic arc and ophiolite
zones, and thus potentially multiple terrane boundaries.
Northwest China Tectonostratigraphy
Tarim and the North China block bound the Altaids south
of Mongolia (Fig. 1). These two microcontinents are inferred
by many to have amalgamated prior to the Ordovician (Yin and
Nie, 1996; Heubeck, 2001). Similar middle to upper Paleozoic
TABLE 1. LATITUDE AND LONGITUDE COORDINATES FOR
KEY STUDY AREAS*
Shin Jinst N 44° 30', E 99° 18'
Noyon Uul N 43° 12', E 101° 54'
N 43° 09', E 102° 04'
Tavan Tolgoy N 43° 33', E 105° 31'
Naftgar Uul N 43° 14', E 105° 12'
Tsaagan Tolgoy N 42° 53', E 105° 27'
Nomgon N 42° 57', E 108° 20'
N 42° 46', E 108° 29'
Hovsgol N 43° 35', E 109° 31'
N 43° 37', E 109° 47'
*See Figure 2.
366 Johnson et al.

Shin Jinst Faults associated with the Permian granite Ordovician-Silurian sequences,
East Gobi fault zone undifferentiated
Mesozoic-Cenozoic rocks, Permian sedimentary
undifferentiated and volcanic rocks Rocks mapped as Precambrian,
Devonian-Carboniferous arc including carbonate klippen and
44º N Triassic sedimentary rocks sequences, undifferentiated possible Mesozoic tectonites

in
b as
o bi Hovsgol
Tavan Tolgoy tG
s
Noyon Uul Ea
Naftgar Uul
Nomgon
Tsaagan
China Tolgoy
border-Mongolia

42º N Onch
Hayrhan Bulgan Uul N
99º E

108º E
0 50 100 km
102º E

105º E
Southwest localities South-central localities Southeast localities

Figure 2. Geologic map of southern Mongolia, showing main study areas. This “timeslice” map is generalized and modified from Tomurtogoo
(1999) and Yanshin (1989) and highlights Triassic and older rocks most relevant to this study. See Table 1 for latitude and longitude coordinate
locations of the study areas.

stratigraphy on either side of the Qilian Shan supports this early
assembly (Yang et al., 2001; Ritts et al., 2004), and no clear
Phanerozoic suture has been found between Tarim and the North
China block. On the contrary, paleomagnetic data indicate sepa-
rate apparent polar wander paths throughout the late Paleozoic,
and such reconstructions imply that the North Tien Shan and Yin
Shan segments of the suture zone formed at about the same time
(Carboniferous–Permian) but at very different latitudes (McEl-
hinny et al., 1981; Li et al., 1988; Enkin et al., 1992; Zhao et al.,
1996). Floral distributions generally contradict this interpreta-
tion, instead indicating that Tarim and the North China block
were at similar latitudes in the Late Permian (Ziegler et al.,
1996). Paleozoic paleomagnetic poles from Tarim may be over-
printed, not fully reconciled with Mesozoic–Cenozoic shorten-
ing and vertical axis rotations, and/or compared with problem-
atic Eurasian reference poles (Gilder et al., 1996; Roger et al.,
2003). With respect to late Paleozoic collision with the Altaids,
it seems that Tarim and the North China block can be treated as
constituting a northward-migrating Andean-style margin, but it
is important to note that the paleogeography is poorly known
and controversial.
Amalgamation of northwest China occurred in two main
phases, beginning with collision of Tarim and the Central Tien
Shan by Early Carboniferous time (Fig. 1; Carroll et al., 1990;
Windley et al., 1990; Allen et al., 1993; Zhou et al., 2001; Wang
et al., 2005). The second phase of amalgamation in northwest
China is inferred to represent the southern boundary of the Altaids
in northwest China and thus is of primary interest for this study.
Alternate interpretations place the boundary to the north of the
Junggar basin (e.g., Junggar-Hegen suture, Fig. 1; Zhang et al.,
1984; Watson et al., 1987). Although crustal thickness throughout
northwest China ranges up to 50 km, this is mainly attributed to
post-Paleozoic sedimentary basin deposits and Cenozoic crustal
thickening owing to India-Asia collision (Wang et al., 2003). No
Precambrian rocks have been found surrounding the Junggar
basin, in contrast to widespread Paleozoic ophiolite complexes
and mafic igneous rocks that are present (Ren et al., 1987; Cole-
man, 1989; Wang et al., 2003). Thus the Junggar basin does not
seem to be underpinned by Precambrian continental crust on the
basis of multiple lines of evidence, and the area is better grouped
into Altaid arc affinities (Şengör et al., 1993).
Collision between the combined Tarim–Central Tien Shan
and the North Tien Shan–Bogda Shan Arcs occurred between
Late Carboniferous and Early Permian time (ca. 300–280 Ma),
based on radiometric dating of arc-related igneous rocks (Fig. 1;
Yin and Nie, 1996; Dumitru et al., 2001; Zhou et al., 2001). Sedi-
mentary basin analyses provide further constraints on formation
of the North Tien Shan. The youngest marine rocks directly north
of Tarim (in the Junggar basin) are Upper Carboniferous–Lower
Permian turbidites and shallow-marine sandstone, mudstone, and
carbonates (Carroll et al., 1995). Overlying Permian deposits are
nonmarine, with at least 3000 m of upward-coarsening succes-
sions ranging from lacustrine mudstone to coarse alluvial-fluvial
conglomerate (Carroll et al., 1990).
Apparent climatic shifts accompany and postdate collision
in the Junggar and Turpan basins (connected in the Permian–
Triassic, Hendrix et al., 1992; Carroll et al., 1995). There is a
Late Permian transition from organic-rich, wood- and coal-bear-
ing fluvial and lacustrine rocks associated with relatively humid
climates, to braided fluvial red beds with common desiccation
 Sedimentary response to arc-continent collision
features (Carroll, 1991; Wartes et al., 2002). Arid conditions
persisted from the latest Permian to the Early Triassic, whereas
coal-bearing, wet-climate rocks were once again widespread
in the Middle to Late Triassic (Hendrix et al., 1992; Greene et
al., 2001). Along with paleocurrent and provenance data, these
lithologic and climatic changes have been attributed to syndepo-
sitional deformation during uplift of the North Tien Shan in the
Early Permian and continued deformation and partitioning of the
flexural Junggar-Turpan-Hami basin from the Late Permian into
the Mesozoic (Hendrix, 2000; Greene et al., 2001; Vincent and
Allen, 2001). Carroll et al. (1992) also linked evidence for saline
versus fresh-water depositional conditions in Permian lacustrine
oil shale to an orographic effect, driven by increased precipitation
runoff adjacent to the North Tien Shan.
The Junggar basin likely originated in an arc setting (back-,
intra- or interarc), and by the late Paleozoic it was a trapped ocean
basin, evolving into a flexural foreland with ongoing collisional
deformation in the Mesozoic (Allen et al., 1991; Carroll et al.,
1990; Vincent and Allen, 2001). Sediment transport in the Perm-
ian Junggar trapped ocean basin is inferred to have been toward
the west-northwest, as indicated by regional paleocurrent data
(Carroll et al., 1995). This may reflect closure of a wedge-shaped
ocean that opened toward the west, perhaps accommodated by
clockwise rotation of Tarim and the Central Tien Shan during
this time. This interpretation is consistent with published remnant
ocean-basin models, where sediment is dispersed outboard of the
migrating suture zone and mainly parallel to colliding margins
(Ingersoll et al., 1995). However, it may contradict the broader
interpretation of the entire Tien Shan–Yin Shan suture having
closed progressively from west to east (Nie et al., 1990; Muel-
ler et al., 1991; Amory et al., 1994). Other geologic evidence
for Permian block rotation in northwest China includes dextral
strike-slip shear zones in the North Tien Shan with synkinematic
biotite 40Ar/39Ar ages of 290–245 Ma (Laurent-Charvet et al.,
2003). There is also paleomagnetic evidence for continued move-
ment of Tarim with respect to Siberia since the Late Permian (Li
et al., 1988; Gilder et al., 1996), so it seems permissible that rota-
tion could have begun during late Paleozoic collision, coincident
with dextral shear in the North Tien Shan.
Northeast China Tectonostratigraphy
The broader interpretation of west-to-east closure across the
entire Tien Shan–Yin Shan suture zone is based mainly on evi-
dence for younger collision along the Yin Shan belt (northeast
China) ~2000 km east of the North Tien Shan (Fig. 1; Amory et al.,
1994). Cope et al. (2005) interpreted active arc magmatism from
ca. 400–275 Ma via detrital zircon ages from Permian sandstones.
This is further constrained by a geochemical shift to inferred post-
collisional granites younger than ca. 257 Ma (Chen et al., 2000).
Dissected continental arcs are found south of the Mongolian bor-
der in the Yin Shan highlands, indicating south-directed subduc-
tion in the region through Late Permian time. Alternatively, Xiao
et al. (2003) define both a northern and southern boundary to the
367
Solonker (e.g., Yin Shan) suture in northeast China, with dual-
polarity subduction beneath two continental margins (the North
China block to the south and a Devonian continental margin to
the north), followed by Himalayan-style, final-stage continental
collision of these margins at the end of the Permian. Although
few paleomagnetic data are available for Carboniferous–Permian
rocks of the North China block (cf. Huang et al., 2001), Mesozoic
paleogeographic reconstructions suggest that Mongolian arc ter-
ranes had accreted to the North China block by the Late Permian,
and as a combined block they rotated counterclockwise relative
to the Siberian craton from the Triassic to the Cretaceous during
closure of the Mongol-Okhotsk ocean (Fig. 1; Enkin et al., 1992;
Heubeck, 2001). As noted for paleomagnetic data from northwest
China, it seems possible that this rotation began during or prior to
Permian collision.
Along the North China block boundary in the Yin Shan
Range (Fig. 1), >1000 m of Upper Carboniferous–Upper(?)
Permian sedimentary rocks were deposited in braided fluvial
to meandering fluvial and floodplain environments (Mueller et
al., 1991; Cope et al., 2005). The youngest marine rocks in this
region are Ordovician shallow-marine carbonates that uncon-
formably underlie the Carboniferous and a younger nonmarine
sequence. Significant changes occur within the upper Paleozoic
stratigraphy in northeast China. Coeval with a shift from braided
fluvial to meandering fluvial deposition, paleocurrents reversed
from north-directed dispersal in Carboniferous–Lower Permian
strata to south-directed dispersal in Upper Permian–Lower Tri-
assic(?) rocks (Cope et al., 2005). Sandstone petrography indi-
cates a recycled orogen to dissected arc provenance, representing
unroofing of a thin sedimentary cover into continental arc plutons
during the Late Permian (Cope et al., 2005).
The late Paleozoic succession in northeast China is also
marked by a climatic shift from humid-environment, coal-bearing
deposits in the Carboniferous–Early Permian to widespread arid-
climate, red bed deposition by Late Permian time (Mueller et al.,
1991). Cope et al. (2005) suggest that the climatic shift represents
a rain shadow effect created during uplift of the continental mar-
gin owing to collision, and/or northward movement of northern
China to arid subtropical latitudes during this time. Uplift and sig-
nificant crustal thickening along this margin during the Permian
may have been further supported by Late Triassic–Early Jurassic
(ca. 220–190 Ma) collapse of the orogen on the basis of radiomet-
ric ages of rocks involved in metamorphic core complex forma-
tion in northeast China (Davis et al., 2004).
Summary
Multiple lines of evidence indicate that collision between the
southern Altaids and microcontinent blocks of north China (Tarim
and the North China block) occurred during Carboniferous to
Permian time. More specifically, suturing appears to have occurred
earlier in the Tien Shan than in the Yin Shan Ranges, suggesting
diachronous west-to-east closure. However, available radiometric
data indicate that arc activity was at least partly overlapping in
368 Johnson et al.
northeast and northwest China, and the transition from nonmarine
to marine deposition is not well preserved across the length of the
suture zone. Paleomagnetic data generally support the possibility
of differential rotation of Tarim and the North China block during
the late Paleozoic, and stratigraphic studies suggest that the Jung-
gar trapped ocean basin may have opened toward the west rather
than having been contiguous to the east. Thus the prospect of a
more complex collisional event, with Tarim and the North China
block acting somewhat independently and perhaps separated by
an indenter (Lamb and Badarch, 2001), is intriguing and perhaps
supported by our new data from Mongolia. Another important but
poorly constrained aspect of this collision is what, if any, climate
response occurred during and after arc-continent collision, and
whether this might have been driven by orographic relief along
the suture zone. Rain shadow and other topographic influences
driven by collisional and postcollisional tectonics have been cited
to explain observed stratigraphic changes in Permian basins of
northwest and northeast China, but this is only one of multiple
possible driving factors.
Outstanding questions that we now seek to address include
whether the Permian of Mongolia supports or refutes the tim-
ing and location of this suture zone, as suggested by data from
the continental margin to the south. The potential climatic effects
of this event are also investigated. Additional data from the
southern Altaids help to evaluate how well various modern ana-
logues, including the southwest Pacific Ocean, may apply to this
setting. This ancient record addresses existing models of basin
evolution, evolving sedimentary systems and source areas, and
issues of preservation in comparison with modern analogues.
Finally, interpreting subsequent Mesozoic–Cenozoic deforma-
tion requires understanding the pre- and syn-assembly record of
southern Mongolia provided by this study.
LATE PALEOZOIC SEDIMENTATION AND
TECTONICS OF SOUTHERN MONGOLIA
Overview
The broad swath of land from southwestern to southeastern
Mongolia encompasses some 1000 km along the northern edge of
the Tien Shan–Yin Shan suture zone. Arc and accretionary com-
plexes that define the Altaids are well exposed along basement
uplifts in southern Mongolia (Fig. 2). Nevertheless, the number
and nature of these intraoceanic arc systems is debated; the geo-
logic complexity of the region is often represented by dozens of
individual tectonic “terranes” and intricate models for their colli-
sion and evolution (Zonenshain et al., 1990; Şengör et al., 1993;
Şengör and Natal’in, 1996; Badarch et al., 2002).
Based on geochemical data and sandstone provenance data
from >20 sites in southern Mongolia, Lamb and Badarch (2001)
proposed a simpler interpretation of two main evolving arc sys-
tems, Devonian and Carboniferous, followed by Permian closure
of the Paleo-Asian ocean. Devonian units are generally assigned
to an island arc setting above a north-dipping subduction zone,
likely close to a continental margin or highly dissected sediment
source to the north (Badarch et al., 2002). The Carboniferous arc
shifted slightly south, closer to the present-day China-Mongolia
border, and is at least partly built on Devonian strata. Carbon-
iferous sandstones suggest a possible granitic source, although
this may represent uplift and dissection of the Devonian arcs
rather than proximity to a microcontinent (Lamb, 1998). Both
systems have analogous components across the border and likely
extended west into the Bogda Shan–North Tien Shan arc sys-
tem (Fig. 1; Lamb and Badarch, 2001), albeit with some changes
toward more oceanic arc affinities to the southwest. Similarly, the
system is tentatively linked through southeast Mongolia and into
northeast China (Xiao et al., 2003). These arcs were likely more
complicated regionally, but the simpler interpretation reflects the
resolution afforded by available data (Lamb and Badarch, 2001).
This interpretation is also more representative of continuous arc
systems, such as the modern Philippine Arc or other parts of the
southwestern Pacific Ocean, where arc type, geometry, and sub-
duction polarity are variable along strike but nevertheless are part
of the same geodynamic setting.
Regional geologic maps of southern Mongolia show mul-
tiple Precambrian units, implying the existence of a microconti-
nent along this part of the Tien Shan–Yin Shan suture, e.g., the
South Gobi microcontinent (Fig. 2; Şengör et al., 1993; Xiao et
al., 2003). This interpretation is supported by a single U-Pb zir-
con age (916 ± 156 Ma) from gneiss on the Chinese side of the
Yagan-Onch Hayrhan metamorphic core complex (Fig. 2; Wang
et al., 2001); however, protolith ages have not been tested rigor-
ously across the region. High-grade gneiss-schist-mylonite com-
plexes in southeastern Mongolia are mainly Mesozoic tectonites
(Webb et al., 1999; Webb and Johnson, 2006), and in many areas
inferred protolith lithologies are more consistent with Paleozoic
arc successions than with continental crust (Johnson et al., 2001;
Webb and Johnson, 2006). In addition to metamorphic rocks,
Proterozoic marble-quartzite units are mapped throughout the
region. These are typically in flat, thrust-fault contact over Perm-
ian and younger rocks and are inferred to be klippen of alloch-
thonous lower Mesozoic thrust sheets (Webb et al., 1999). Thus
the existence of the South Gobi microcontinent is uncertain, and
in contrast to the Tarim and North China blocks the Precambrian
crust likely forms a very minor component of the southern Altaids
in Mongolia, if at all.
Although these rocks have not been described extensively
in the English-language literature, Permian sequences in south-
ern Mongolia are inferred to represent final closure of the Paleo-
Asian ocean between Mongolian arc terranes and continental
block(s) to the south (Amory, 1996). The timing of this terminal
collision is perhaps best constrained by Permian floral assem-
blages that indicate mixing between polar-temperate, Siberian
(Angaran) communities and tropical, north China (Cathaysian)
communities (Durante, 1971, 1983; Shen et al., 2006). Similarly,
interpreted syn- or postcollisional granites in northeast China
have ca. 250–230 Ma U-Pb and Rb-Sr ages, supporting Late
Permian collision (Chen et al., 2000).
 Sedimentary response to arc-continent collision 369

The ages of Permian rocks in Mongolia are generally only
poorly known. Volcanic lithologies are uncommon except for the
Early Permian (Fig. 3), and even for this interval have not been
extensively dated by modern radiometric techniques. Invertebrate
biostratigraphy for marine sections in southeastern Mongolia
relies mainly on brachiopod zonations, some with identifica-
tion of bryozoans, crinoids, and foraminifers (Manankov, 1988,
2004). Nonmarine strata are constrained by plant fossils, inverte-
brates, and spore-palynology where available. These forms tend
to be low-resolution age constraints (>1 m.y.), and the correlation
from Russian “horizons” to the most recent global absolute time
scales (Gradstein et al., 2004) is just currently under way (Shen et
al., 2006; Shi, 2006; Manankov et al., 2006). Thus it is difficult to
provide stage or even epoch assignments for many of the studied
intervals. Improving age resolution and correlations to a global
time scale are a focus of our ongoing work in Mongolia. For the
purposes of this study we retain the use of Upper-Late Permian
(Cisuralian, 299–270.6 Ma) versus Lower-Early Permian (Guada-
lupian–Lopingian, 270.6–253 Ma) terminology (Fig. 3; cf. Grad-
stein et al., 2004), as this is still widely used in regional literature
and provides the best means for suggesting regional correlations.
Generalized stratigraphic correlations (Fig. 3) show broad
subdivisions for the Permian of southern Mongolia. Lower Perm-
ian red bed and rhyolite-andesite volcanic sequences are observed
across the region, typically in fault contact with Carboniferous–
Devonian arc sequences. Overlying middle Permian strata include
shallow-marine carbonate and clastic successions, which con-
formably underlie turbidite deposits of the Lugyn Gol Formation
(Fig. 3). A regional unconformity divides these flysch deposits
from the nonmarine overlap succession, which includes coal-bear-
ing fluvial strata that graded into braided fluvial successions in the
Early Triassic (Fig. 3).

Standard Southwest South-central Southeast

Chronostratigraphy Mongolia Mongolia Mongolia
Triassic
250 and younger
Tavan Tolgoy and ?
Tsaagan Tolgoy Nonmarine Upper
~1600 m meandering Formations, ~100-300 m Permian units not
fluvial deposits, Noyon thick. Coal bearing fluvial Uncon- exposed.
Lopingian

255 Uul (Figs. 4-5). formity

deposits (Figs. 6-10).
?
Late Permian

Uncon-
260 formity
Lugyn Gol Formation,
Lugyn Gol Formation,
Guadalupian

? >500 m thick. Distal

>100 m thick (partially turbidite deposits (Figs.
265 preserved in 11, 12).
south-central Mongolia).
Distal turbidite deposits.
Uncon-
formity
270
Age (Ma)

Tavan Har and Har

? Agui Uul and Bulgan Uul Erdene Formations,
275 Formations, >3000 m >300 m thick. Shallow
thick. Shallow marine marine clastics, minor
Early Permian

carbonate and clastic carbonate and volcanic

deposits. units (Figs. 11, 12).
Ulaan Nur Formation,
280 ~300 m thick.
Rhyolitic-andesitic
volcanic rocks and
nonmarine clastic
Cisuralian

285 deposits (Figs. 4-5).

? Uncon-
contact formity
not
? exposed
290
Argalant Formation,
>100 m thick, red beds
and volcanic rocks. Argalant Formation,
>100 m thick, red beds
295 Carboniferous ? Unconformable and fault and volcanic rocks.
? sedimentary and volcanic contact with
C2 rocks thrust over Permian Carboniferous and older.
(Noyon Uul, Fig. 5E) or ? Unconformable and
P1 fault contact with
Carboniferous angular unconformity with
Carboniferous and older. Devonian.
and older

Figure 3. General chronostratigraphic correlation of Permian sedimentary and volcanic rocks from southwest to southeast Mongolia (cf. locali-
ties shown in Fig. 2). Lithologies are indicated by patterns as shown in Figure 4 (key). Formation names, approximate thicknesses, lithologies,
and contact relationships are compiled from our own work and published interpretations (Berkey and Morris, 1927; Grabau, 1931; Durante,
1971; Mossakovsky and Tomurtogoo, 1976; Zaitsev, 1974; Nikolov et al., 1981; Manankov et al., 2006). Dashed lines between sections represent
inferred lithostratigraphic correlations, which in some cases are interpreted to be time-transgressive. Time scale is from Gradstein et al. (2004),
although we have retained the use of Early versus Late Permian epochs, as discussed in the text. Age assignments are only approximate, given
the low resolution of available age constraints from Mongolia.
370 Johnson et al.
To fill in such broad correlations with more detailed field-
based constraints, we now present new data from a regional
southwest to southeast transect, highlighting localities where
Permian strata are best preserved. The sections were visited
from 1992 to 2005, and so this is a long-term compendium of
our studies. We present these observations at a reconnaissance
scale in order to emphasize regional relations; much of this inter-
pretation is derived from more detailed descriptions in Amory
(1996) and Lamb et al. (1999). This work focuses on the iden-
tification of sedimentary facies to highlight basin setting and
paleogeography, to constrain marine to nonmarine transitions,
and to identify shifting climatic and tectonic controls on sedi-
mentation. Stratigraphic relations were identified on the basis of
measuring and describing sections and, where possible, collect-
ing paleocurrent data. Climatic indicators are mainly inferred
from study of nonmarine sedimentary facies, particularly the
distinction between highly organic and coal-bearing strata that
formed in wet to seasonally humid environments versus red
beds, with common exposure and desiccation features indicative
of aridity (Parrish et al., 1982; Wright, 1992; Dubiel and Smoot,
1994). Spores and other palynomorphs from the Tsaagan Tol-
goy locality (Fig. 2) support these interpretations. Sedimentary
provenance, mainly derived from sandstone petrofacies, is also
discussed as an additional constraint on basin tectonic setting
and evolving source areas.
Southwest Localities
Permian strata in southwest Mongolia (Fig. 2) lie uncon-
formably on Lower Carboniferous and older marine sedimentary
and volcanic units. Although Devonian–Carboniferous succes-
sions have distinct arc-related geochemistry, Permian strata are
nonmarine and are interbedded with basalts that have intraplate
geochemical signatures (Lamb and Badarch, 2001). These rela-
tions support a postcollisional setting by Permian time.
At Shin Jinst (Fig. 4), Zonenshain et al. (1974) reported
Late(?) Permian plant fossils in a section of ~100 m of reddish
argillite, sandstone, and conglomerate (Fig. 5A). Although beds
are cleaved and deformed in places, sedimentary structures are
relatively well preserved. Organic material is abundant through-
out, including in-place petrified wood stumps. Sandstone beds
are typically lenticular over several meters, have sharp and
erosional bases, and are trough to low-angle cross-bedded and
current-rippled (Fig. 5B–D). These features are interpreted
as fluvial channels, and given the abundance of red and gray
mudstone, representing floodplain deposits, we suggest mainly
meandering fluvial environments. The presence of soft sedimen-
tary deformation features, climbing ripples, and syndepositional
clastic dikes may also be consistent with a delta-plain environ-
ment (Battacharya and Walker, 1992), even though distal marine
or lacustrine successions are not preserved. There is an over-
all upward-coarsening trend, possibly representing a shift from
meandering to braided fluvial deposition from Late Permian to
Early Triassic time.
Approximately 250 km southeast of Shin Jinst at Noyon
Uul (Fig. 2), ~1600 m of Upper Permian strata include inter-
bedded sandstone and mudstone intervals that are also inter-
preted to represent meandering fluvial environments (Fig. 4;
full section shown in Hendrix et al., 1996). Sandstone beds are
lenticular over ~10 m (Fig. 5F) with minor erosional scour-
ing at their bases and are upward fining, with common trough
and planar cross-bedding. Reworked terrestrial organic matter
is present throughout. The section is dominated by floodplain
deposits of variegated silt and mudstone; carbonate nodules and
minor paleosols (argillosols) are also present (Fig. 5G, H). Age
control in this part of the section is based on well-preserved
Late Permian flora and fauna (Zaitsev et al., 1973; Durante,
1976; other references in Hendrix et al., 1996). However, the
uppermost portion of the meandering fluvial succession has
been assigned an Early Triassic age, based upon tetrapod fossils
(Gubin and Sinitza, 1993). This generally fine-grained sequence
is sharply overlain by thick Middle(?) Triassic conglomerates
(Zaitsev et al., 1973; Hendrix et al., 2001). The coarse-grained
Triassic–Jurassic units at Noyon Uul were deposited in an intra-
plate setting, likely within a flexural foreland basin (Hendrix
et al., 1996). They clearly postdate the main collisional event
along the Tien Shan–Yin Shan suture but likely record ongoing
intraplate contractile deformation.
The Late Permian floral assemblage at Noyon Uul is also
notable because it contains a mixture of Cathaysian (north China
affinity) and Angaran (Siberian affinity) flora (Durante, 1971;
Amory, 1996). This is one of the early phases of population
mixing prior to complete homogenization in the Triassic, and is
detected at localities in south-central Mongolia as well. In particu-
lar, an unusual cycad genus (Guramsania) has been found at both
Noyon Uul and Tsaagan Tolgoy (Figs. 2, 6), providing further
evidence for a connection between these localities (Vakrameyev
et al., 1986). Such floral mixing in the transition zone along the
China-Mongolia border provides compelling evidence of close
approach and/or collision with the North China block (Li, 2006;
Shen et al., 2006).
Although Upper Permian fluvial strata measured within the
Noyon Uul syncline are generally consistent with at least peri-
odically wet-humid conditions (Hendrix et al., 1996), Permian
strata unconformably underlying this section are quite different.
Anatoleva (1974) describes ~315 m of distinctively red extrusive
volcanic units interbedded with sandstone and mudstone exposed
southeast of the Noyon Uul syncline (Figs. 2, 4). Our observa-
tions of this section are generally consistent with Anatoleva’s
(1974) descriptions. The lowermost unit is in thrust-fault contact
with middle–Upper Carboniferous arc sequences (Fig. 5E). It
consists of ~55 m of poorly sorted cobble-boulder conglomerate
interbedded with andesitic-rhyolitic crystalline tuffs ranging from
0.5 to 2.5 m thick. This sequence fines upward into a distinctive
red and reddish-brown succession (~140 m total) of interbedded
siltstone, sandstone, and conglomerate interbedded with tuff,
tuff breccias, and rhyolitic ash-fall deposits and rare trachyte-
basalt flows. The uppermost ~120 m of section grades to mainly
 Key for all stratigraphic sections
siderite horizon current ripples Upper Permian,
root marks low angle Shin Jinst
cross beds (m)
mud chips Upper Permian (?) - lowermost 100
low angle /
carbonate wavy bedding Triassic, Noyon Uul syncline
nodules planar laminae (Western Transect,
soft sediment Hendrix et al., 1996) trough
deformation trough cross beds cross beds
cover n=58
coal (m)
mudstone-siltstone
fine-medium grained sandstone
volcanic and volcaniclastic
sandstone
gravel-pebble conglomerate 40
limestone or dolomite
cobble-boulder conglomerate
75

Lower Permian,
SE of Noyon Uul syncline
(m) angular unconformity

300 30

250 50

200 20

150

25

100 10

50

0m 0m 0m
r t. t. l. r t. t. l. r t. t. l.
ve lts ss cg ve lts ss cg ve lts ss cg
co d/si d. ble co d/si d. ble co d/si d. ble
u e b u e b u e b
m m pe m m pe m m pe

Figure 4. Measured sections from southwest localities (see Fig. 2 for locations). Key applies to all measured sections
(Figs. 4, 6, 9, 11). The left-hand section (Lower Permian, southeast of Noyon Uul syncline) is a schematic representation
based on Anatoleva (1974) and our own field observations. The Noyon Uul syncline detailed section was measured in
1992 by M. Hendrix and E. Sobel (Western Transect of Hendrix et al., 1996; ~1500 m from base), and is redrafted with
permission. This section was measured close to the Permian–Triassic boundary, and its exact age is uncertain; however,
the depositional style is representative of the underlying, continuous Permian section. Paleocurrent measurements are
from the same approximate stratigraphic interval, measured ~10 km along strike (Sain Sar Bulag transect; Hendrix et al.,
2001). The Shin Jinst section was measured as part of this study, as described in the text.
A B

~ 30 meters

C D

E F

~10 m

G H

Figure 5. Photographs from Permian strata at the southwest localities (F–H from M. Hendrix). (A) Overview of Shin
Jinst section. (B, C) Channel cut-and-fill geometries at Shin Jinst. (D) Soft sedimentary deformation within a channel
sandstone body, Shin Jinst. (E) Lower Permian rocks southeast of Noyon Uul syncline, showing red beds and volcani-
clastic rocks in the foreground. High ridges in the background are Carboniferous sedimentary and volcanic rocks thrust
over Permian (teeth on overriding plate). (F) Overview of detailed Upper Permian section at Noyon Uul. (G) Carbonate
nodules within interdistributary mudstone-siltstone deposits, Upper Permian, Noyon Uul. Stratigraphic up is to the upper
left. (H) Variegated red-green argillosol and resistant carbonate zone, Upper Permian, Noyon Uul (hammer for scale).
 U. Permian-L.Triassic
Tsaagan Tolgoy, detailed section
U. Permian-L.Triassic (m)
Tsaagan Tolgoy, long section 70
last petrified wood fragment
(m) (at about 80 meters)
400
X

60
last siderite horizon

major shift in pollen assemblage

300 50
X Lower Triassic
(measured along strike)
detailed section at right

first nodular carbonate

X Upper Permian
color change to dark gray
to greenish-light gray mudstones

40

200
last coal bed
last Cordaites leaf impression
30

20

100

X
10

X
0m 0m
l al
r oa t. t. l. er coltst. sst. cgl.
o ve c ilts . ss cg v i
c d/s d ble
e b co d/s d. ble
u u e b
m m pe m m pe

Figure 6. Long (at left) and detailed (right) measured sections across the Permian–Triassic boundary at south-central local-
ity Tsaagan Tolgoy. Xs mark palynological samples used in Figure 7. See Figure 4 for key to lithologies and symbols.
374 Johnson et al.
cross-bedded sandstone and siltstone, which are variegated with
yellow and greenish gray more commonly than in the underly-
ing sections (Fig. 4). In general the section is coarse grained and
poorly sorted, with a trend to better sorting and rounding starting
at ~100 m from the base. Organic matter is rare, probably owing
to volcanic contributions and poor preservation under oxidizing
conditions. These observations are consistent with an alluvial fan
setting, close to an active volcanic source area, possibly grading
to braided fluvial deposition through time.
A Late Permian age for this volcanic and clastic succession
southeast of the Noyon Uul syncline has been suggested (Ana-
toleva, 1974); however, this lower section appears to have an
angular relationship with the Upper Permian strata described by
Hendrix et al. (1996) and is also quite distinctive lithologically
(with its red color, abundant volcanic rocks, and lack of coal mea-
sures). The age of the lower section is bracketed between Late
Carboniferous, which is in fault contact with this section, and
Late Permian; thus we favor an Early Permian age for these clas-
tic and volcanic units.
South-Central Localities
One of the better known nonmarine Permian successions in
south-central Mongolia is found at Tsaagan Tolgoy (also known
as Ih Uvgon, Zinniker and Badarch, 1997; Figs. 2, 6). Age control
here is excellent, based on our own and previous paleontological
analyses (Durante, 1976; Vakrameyev et al., 1986; Zinniker and
Badarch, 1997; Uranbileg, 2003; Aristov, 2005), which document
Upper Permian through Lower Triassic strata (Fig. 6). Character-
istic Permian pollen genera found at Tsaagan Tolgoy (identified
by D. Zinniker, Fig. 7) include Cordaitina, Florintes, Potonie-
isporites, and Neoraistrickia. Triassic and transition zone species
include Lueckisporites virkiae, Lundbladispora sp., and Taeni-
sporites sp. The Permian–Triassic palynological shift is similar
to that documented along the southern margin of the Junggar
basin (Ouyang and Norris, 1999).
The Upper Permian section at Tsaagan Tolgoy (Fig. 6) is
~300 m thick and consists primarily of dark, organic-rich mud-
stone, coaly zones, and thin sandstone beds (Figs. 7, 8A). Pet-
rified wood is common, including paleohorizontal and growth-
position logs encased in tabular sandstone beds (Fig. 8B). The
section is generally fine grained, with multiple concentrated
zones of centimeter-scale siderite nodules, indicating swampy,
reducing conditions (Landuydt, 1990). Sandstone units range
from thin beds to 1–4-m-scale individual and amalgamated units
and commonly exhibit convolute bedding and slumping and
upward-fining trends (Fig. 8C). Trough cross-beds and asym-
metric current ripples are also found within the discontinuous
sandstone beds. Overall, this section is consistent with poorly
drained interdistributary or backswamp deposition, with sandier
units representing mainly crevasse splay deposition or meander-
ing of minor channels across the floodplain.
Above the Permian–Triassic boundary (~280 m above the
base of the section; Figs. 7, 8D), the section at Tsaagan Tolgoy
coarsens upward abruptly, with the first appearance of pebble-
cobble conglomerate at ~400 m. Conglomeratic and sandy inter-
vals contain large-scale trough cross-beds and sharp to erosive
bases (Fig. 8E). Above 280 m, mudstone units grade abruptly
from organic-rich, dark-gray units to lighter colored tan-green
and organic-poor beds. Carbonate nodules also appear in the
upper section, and siderite is comparatively rare, indicating a
shift to relatively well-drained soils. Overall, this vertical trend
is interpreted as a change to higher energy fluvial environments,
possibly including braided-stream deposition, with more oxi-
dized and less organic-rich floodplain deposition.
A distinctive shift in floral assemblages also occurs at Tsaa-
gan Tolgoy (Fig. 7). Late Permian flora are dominated by Cor-
daitales (pre-gymnosperms), ferns, and bottryococcus algae,
whereas the Triassic assemblage includes taeniate-striate bisac-
cate pollen typically associated with evaporites and more arid
conditions. Cavate microspores (Lundbladispora sp.) and mega-
spores characteristic of Early Triassic heterosporous lycopods
(e.g., the desert xerophyte Pleuromeia) were also found in the
upper part of the section. An observed 4–6‰ decrease in the δ13C
ratio of organic carbon is coincident with changes in lithology
and pollen assemblage (Fig. 7A). Similar shifts across the Perm-
ian–Triassic boundary in marine and nonmarine sections around
the world are believed to reflect a large, globally synchronous
contribution of isotopically light carbon into the atmosphere
(Baud et al., 1989; Krull and Retallack, 2000).
Along with the lithologic changes noted previously, the
rapid floral and isotopic shift is thought to represent the demise
of Late Permian coal-bearing swamp facies and replacement with
primarily herbaceous flora tolerant of seasonal droughts. The
change may reflect the global record of warming across the Perm-
ian–Triassic boundary (Faucett et al., 1994; Retallack et al., 1996;
Berner, 2002). However, a similar aridification trend observed
within Permian rocks of northeast China has also been attributed
to a possible rain shadow effect, reflecting ongoing collision and
uplift along the Tien Shan–Yin Shan suture (Cope et al., 2005).
Supporting evidence for this climate shift is found at the Tavan
Tolgoy and Naftgar Uul localities, described below. Although
neither of these sections records the actual Permian–Triassic tran-
sition, as at Tsaagan Tolgoy, individually they provide supporting
evidence of distinctive and shifting climate regimes.
Additional Upper Permian coal-bearing fluvial deposits are
preserved within coal quarries at Tavan Tolgoy (Durante, 1971;
Fig. 9). Coal seams are individually up to 14–33 m thick and
interbedded with coarser units, including rare pebble conglom-
erate and medium- to fine-grained sandstone. Sandstones are
trough cross-bedded with sharp, erosional bases, and are lenticu-
lar, although amalgamated units are laterally continuous over tens
of meters (Fig. 10). Sandstone beds exhibit normal grading, and
fine upward to rippled siltstone and mudstone. Paleocurrent mea-
surements at Tavan Tolgoy are primarily south-southwest directed
(Fig. 9). Overall, outcrops at Tavan Tolgoy are broadly analogous
to the Upper Permian at Tsaagan Tolgoy and are consistent with a
meandering fluvial system in a relatively humid environment.
 A Relative abundance of
palynomorph groups: present common abundant

smooth or small sculptured

cavate spores (7E)
characteristic

megaspore wall fragments

Triassic taxa

spore tetrad

taeniate bisaccate (7D)

striate bisaccate
transitional

unsculptured acavate
spores
normal bisaccate

monosaccate
characteristic
Permian taxa

sculptured acavate
spores (7B and 7C)
large sculptured cavate
spores
bottryococcus

cycadopites
-23
d 13 C (‰)

-25

-27
Permian Triassic
-29
0 40 80 120 160 200 240 280 320 360 400
distance in meters from base of section

B C D E

10 µ
µmm 10 µm 10 µm 10 µm

Figure 7. (A) Spore and pollen abundance and δ13C isotope data from Tsaagan Tolgoy, plotted against distance above the base of the long sec-
tion shown in Figure 6. (B–E) Photomicrographs of select spores including Neoraistrickia (B), Lycopodiumsporites (C), Leuckisporites (D), and
Lundbladispora (E), keyed to the palynomorph groups shown above. Carbon isotopic data were collected on acid-resistant kerogen isolated for
palynological analysis, conducted at Petrobras stable isotope laboratory in 1997 following industry-standard methods.
 A B

10 cm

Figure 8. Photographs from Tsaagan Tolgoy. (A) Coaly intervals at the base of the section. (B) Petrified log in growth
position. (C) Convolute bedding in overbank deposits. (D) Transition from dark, organic-rich mudstone to tan, organic-poor
mudstone at the Permian–Triassic boundary (~290 m). (E) Trough cross-beds in a lenticular, channelized sandstone.
Figure 9. Measured sections from south-central localities Naftgar Uul and Tavan Tolgoy (see Fig. 2 for locations). Paleo-
current data from Naftgar Uul include data from possible Triassic strata overlying this section. See Figure 4 for key to
symbols and lithologies.
378 Johnson et al.
A The lower part of the section is more consistent with sedimentary
B Southeast Localities
Figure 10. Photographs of the Tavan Tolgoy coal mine section shown
in Figure 9. (A) Lenticular sandstone bodies and interbedded coals.
Person (circled) for scale. (B) Detailed photo of finely laminated silt-
coal beds in interdistributary deposits.
As at Noyon Uul, Upper Permian strata at Tavan Tolgoy and
Tsaagan Tolgoy contain uniquely mixed Angaran-Cathaysian
assemblages (Vakrameyev et al., 1986) and thus indicate that
south Mongolia and north China were in close proximity at that
time. Widespread terrestrial deposition in this area likely occurred
in syn- and postcollisional flexural basins (Hendrix et al., 2001).
Another similarity to the Noyon Uul section is that Upper Perm-
ian strata near Tavan Tolgoy overlie extrusive intermediate-felsic
volcanic rocks and interbedded reddish conglomerate, sandstone,
and siltstone that are thought to be Lower Permian (Ulaan Nur
Formation, Fig. 3; Amory, 1996). Thus there is apparently an
older climatic shift from generally arid to more humid climates
in the middle Permian, followed by a return to aridification at or
near the Permian–Triassic boundary.
In comparison with Tsaagan Tolgoy, Triassic strata are bet-
ter exposed nearby at Naftgar Uul (Figs. 2, 9). Although shown
as Permian on regional maps (Tomurtogoo, 1999), one of us (D.
Zinniker) found Late Triassic to Middle Jurassic pollen assem-
blages (Classopollis) in lacustrine facies near the top of the section.
facies found in the Triassic across the region, and thus it is likely
that these strata are younger than mapped. The section at Naft-
gar Uul includes more than 300 m of interbedded sandstone, silt-
stone, and rare granular to pebble conglomerate. Sandstone beds
contain planar laminations, low-angle cross-stratification, and
trough cross-beds with >1 m of relief. These beds typically have
sharp, slightly erosive bases, and amalgamated units are laterally
continuous over >50 m. Paleoflow directions measured in trough
cross-beds are variable but are generally west-directed (Fig. 9).
Coal is absent in the section, and only rare, poorly preserved and
oxidized plant fossils are present. Lower Triassic strata here and
at Tsaagan Tolgoy are consistent with sand-rich meandering rivers
to braided-stream environments, with oxidizing environments and
likely periodic arid-drought conditions.
The final phase of marine deposition in southern Mongo-
lia is preserved in outcrops of the Permian Lugyn Gol Forma-
tion, exposed at Nomgon and Hovsgol (Figs. 2, 11). Devonian
turbidite deposits are also reported in the area, which has led to
some confusion in mapping, where these widespread outcrops
are alternately assigned either to the Early–Middle Devonian or
the Late Permian (cf. Yanshin, 1989, and Tomurtogoo, 1999).
However, the Late Permian age of these deposits is well estab-
lished in multiple areas (Pavlova et al., 1991) and is confirmed
by our own findings in 1995 and 2005 of Late Permian crinoids
and bryozoans (J. Undariya, 2005, personal commun.). It is pos-
sible that Devonian crinoids and corals found in the section are
reworked and transported. The Lugyn Gol Formation crops out
along northeast-trending ridges at Nomgon (Fig. 2). Beds are
highly deformed and cleaved with large-scale isoclinal folding
and overturned units. Thus true thickness is difficult to deter-
mine, but we estimate a minimum thickness of ~800 m, based on
individually continuous, non-repeated sections (Fig. 11).
Permian flysch units mainly lie unconformably or in fault
contact with Carboniferous and older metasedimentary, volcanic,
and intrusive rocks, including ophiolites, carbonate klippen, and
metamorphic rocks mapped as Precambrian (Fig. 3; Tomurtogoo,
1999). The lowest part of the section includes a possible shallow-
marine succession that is marked by 10–50-m-scale upward-
coarsening and -thickening sandstone packages (Figs. 11, 12A).
Dolomitic units include crinoid, brachiopod, and bivalve coquina
and grainstone. Low-angle cross-bedding includes trough and
possible hummocky cross-stratification, and ripple marks are
also present (Figs. 12B, C). Beds are generally tabular rather
than strongly lenticular, but they do pinch out over hundreds of
meters along strike. Sandstone beds have sharp bases but gener-
ally lack channel-like scour features or bar forms, and no sub-
aerial exposure surfaces were identified. This lowest section is
interpreted as a shallow-marine succession, possibly along a flu-
vial shoreline with upward-shallowing prodeltaic parasequence
packages (Fig. 11; Posamentier and Allen, 1999). Similar units
Figure 11. Measured sections through the Upper Permian Lugyn Gol Formation at Nomgon, southeast Mon-
golia (see Fig. 2 for location). Sections are noncontinuous but are shown in stratigraphic order from oldest to
youngest (left to right). Paleocurrent data are all from uppermost stratigraphic section. See Figure 4 for key to
lithologies and symbols.
380 Johnson et al.

A B

C D

~25 m

E F

Figure 12. Photographs of the Lugyn Gol Formation exposed at Nomgon. (A) Upward-coarsening and thickening pack-
ages (lowermost stratigraphic section, Fig. 11). (B) Trough cross-beds (lowermost section). (C) Ripple marks at top of
bed, middle section. (D) Overview of turbidite sequence. (E) Single upward-fining and -thinning unit (overturned beds,
from uppermost section, Fig. 11). (F) Flute casts at the base of sandy turbidite bed.

are found to the east at Hovsgol (Fig. 2), where crinoid and bra-
chiopod grainstones are present with sporadic plant fossils, bed-
ding-plane parallel and perpendicular bioturbation, and ripples
and convolute bedding, which may also support shallow-marine
and/or prodeltaic deposition.
There is a rapid transition from the shallow-marine packages
into regular turbidite sequences (Fig. 11, uppermost section) at
Nomgon. The classic turbidite flysch architecture of the Lugyn
Gol Formation includes hundreds of meters of tabular, continuous
sandstone and shale beds (Fig. 12D). Sandstone units are centime-
ter to decimeter scale in thickness and are typically upward fining
(Fig. 12E). Basal contacts are sharp to slightly erosive and com-
monly preserve tool marks and flute casts, which generally show
south-southeast transport directions (Figs. 11, 12F). Structures
include normal grading, sporadic load structures, planar laminae,
and some current ripples at the tops of beds, corresponding to Ta,
 Sedimentary response to arc-continent collision 381

Tb, and Tc/Te petrofacies (respectively) of Mutti and Ricci Lucchi Qm

(1978). Local zones of intense cleavage preclude identification of
small-scale sedimentary structures, but the rhythmic sand-shale
interbedding and tabular, laterally continuous units (Fig. 11) are
consistent with turbidite deposition, likely within Mutti and Ricci
Lucchi’s (1978) mixed sand-mud facies (facies C).

rec
These turbidite deposits are tentatively assigned to an outer

yc
submarine fan setting that is intermediate between proximal and

led
k
loc
distal fan deposition (Walker, 1992). Coarse-grained facies, high-

oro
mixed

lb
density turbidite flows, slumping and channel-levee systems were

ge
nta
not observed, which argues against a more proximal inner fan

n
ne
setting. Similarly, the system is quite sandy, and bioturbation is
Dissected

nti
rare, arguing against a distal basin-floor fan environment. How- arc

ever, overall packaging is variable between more shaly and more
sandy intervals, and this likely indicates fluctuations in relative
sea level through time or shifting feeder channels within the fan
(Amory, 1996). The distal submarine environments observed at
Nomgon and Hovsgol are consistent with depositional settings
observed in analogous remnant ocean-basin fan systems, where
basin geometry favors trough-like axial deposition, and turbidite
pulses are funneled outboard over long distances (Graham et al.,
1975). We suggest that the underlying shallow-marine successions
formed along the margin of an extinct Carboniferous–Devonian
arc, immediately prior to final collision stage marked by rapid
subsidence and deposition of overlying turbidite successions.
Sandstone Provenance Data
Sandstone samples were collected for petrographic analysis
and point counting (Fig. 13; Table 2) to complement facies anal-
yses. Sandstone provenance studies permit evaluation of the tec-
tonic setting in which sedimentary sequences are deposited and
can demonstrate evolution of source areas through time (Dickin-
son and Suczek, 1979; Dickinson et al., 1983). In other arc colli-
sional settings, sandstone compositional trends have been attrib-
uted to erosion of magmatic arcs and unroofing of collisional
orogens (Dorsey, 1988; Graham et al., 1993) and thus record tim-
ing and evolution of suturing and postcollisional uplift.
Sandstone provenance studies are best conducted on medium-
grained sandstone samples that are relatively unaltered. Given the
level of deformation and burial in many Permian outcrops of the
region, not all localities are represented. Nevertheless, 82 Perm-
ian samples from 6 localities across southwestern to southeastern
Mongolia were point-counted using a modified Gazzi-Dickinson
method (Ingersoll et al., 1984), which compensates for the poten-
tial adverse effect of grain-size variations. Principal framework
grains were counted by standard categories (Dickinson, 1970) and
then normalized as detrital modes on standard ternary diagrams
for comparison with global modal distribution fields (Table 2;
Dickinson et al., 1983; Marsaglia and Ingersoll, 1992). In some
cases, sandstones were extensively altered, such that distinguish-
ing polycrystalline quartz and chert from altered and devitrified
zeolites or volcanic lithic fragments was impossible. Thus the
QmFLt ternary diagram (Fig. 13; Qm = monocrystalline quartz,
co
Transitional
arc
F Lt
Undissected arc
Shin Jinst (n=11) Southwest localities
Noyon Uul (n=21) (Upper Permian)
South-central
Naftgar Uul (n=18)
locality (Triassic?)
Nomgon (n=21)
Southeast localities
Hovsgol (n=10) (Upper Permian)
Effect of dynamic metamorphism
(Nomgon and Shin Jinst samples)
Figure 13. Ternary diagram of sandstone modal compositions, using
monocrystalline quartz (Qm), potassium plus plagioclase feldspar (F),
and total lithic fragments (Lt). Mean values are shown with a field
of one standard mean deviation (Table 2). Provenance fields are from
Dickinson et al. (1983), following methods described in the text.
F = plagioclase + potassium feldspar, Lt = sedimentary, meta-
morphic, and volcanic lithic fragments + carbonate and chert) is
the most reliable comparison between all localities. Recent stud-
ies of modern arc-derived sands also use the QtFL ternary plot
(Marsaglia and Ingersoll, 1992), which combines monocrystal-
line, polycrystalline, and chert into the Qt (total quartz) field.
However, Qp (polycrystalline quartz) is a minor component in
most modern arc–derived sands (Marsaglia, 1992), so use of
QmFLt modal compositions provides a reasonable comparison
for the southern Mongolia Permian sample suite.
Upper Permian sandstones from southern Mongolia range
from undissected to transitional to dissected arc provenance fields
(Fig. 13), indicating recycling or inheritance of late Paleozoic arc
sources (Graham et al., 1993). None of the samples has a recycled
orogen or continental block signature. There does not appear to be
a consistent spatial trend in samples from across southern Mongo-
lia: i.e., dissected versus undissected arc provenance fields are not
 TABLE 2. POINT COUNT DATA FROM PERMIAN SANDSTONES OF SOUTHERN MONGOLIA
Naftgar Uul
Shin Jinst Qm F Lt (continued) Qm F Lt
94-SJ-181 17.66 29.58 52.76 93-NG-14a 18.37 38.86 42.77
94-SJ-182 16.36 43.18 40.45 93-NG-15 10.83 35.04 54.13
94-SJ-230 41.11 42.78 16.11 93-NG-15a 14.46 28.19 57.35
94-SJ-231 26.98 43.25 29.76 93-NG-24 20.60 36.54 42.86
94-SJ-240 24.24 34.34 41.41 93-NG-25 17.49 28.72 53.79
94-SJ-241 25.87 11.89 62.24 93-NG-26 16.84 31.02 52.14
94-SJ-243 30.70 28.73 40.57 93-NG-27 25.27 22.87 51.86
94-SJ-245 32.75 21.62 45.63 93-NG-27a 17.36 38.14 44.50
94-SJ-246 34.67 21.07 44.27 NG Mean 16.52 33.52 49.96
94-SJ-247 31.99 18.01 50.00 NG St Dev 4.24 5.08 5.71
94-SJ-251 23.48 30.19 46.33
SJ Mean 27.80 29.51 42.69 Nomgon (Lugyn Gol Fm.)
SJ St Dev 7.03 10.24 11.47 93-LG-201 30.80 22.90 46.20
93-LG-202 38.70 22.10 39.10
Noyon Uul (TU = Tost Uul, near Noyon Uul) 93-LG-204 26.10 25.00 48.90
95-NU-2 0.00 4.55 95.45 93-LG-206 39.10 19.00 41.90
95-NU-3 0.76 15.78 83.46 93-LG-207 19.30 28.90 51.90
95-NU-4 3.57 13.74 82.69 93-LG-208 24.60 23.50 51.80
95-NU-8 5.79 3.22 91.00 93-LG-209 23.40 27.10 49.50
92-NU-8 1.43 9.07 89.50 93-LG-210 26.40 23.40 50.20
92-NU-10 1.20 19.60 79.20 93-LG-211 25.60 25.90 48.50
92-NU-13 6.20 19.90 73.90 93-LG-301 26.30 29.30 44.40
92-NU-14 6.80 22.20 71.00 93-LG-302 31.00 27.00 42.00
92-NU-15 10.59 9.85 79.56 93-LG-303 32.90 25.20 41.90
92-NU-39 5.52 4.94 89.53 93-LG-304 30.90 18.00 51.10
92-NU-40 6.10 18.30 73.30 93-LG-305 20.60 29.40 50.00
92-NU-41 8.25 18.45 73.30 93-LG-306 27.40 31.80 40.80
97-TU-1 6.40 18.50 75.10 93-LG-307 30.90 21.10 48.00
97-TU-4 8.80 12.50 78.70 93-LG-308 26.60 29.60 43.80
97-TU-7 5.40 16.40 78.30 93-LG-309 27.50 30.50 41.90
97-TU-8 5.90 10.80 83.30 93-LG-401 32.00 24.70 43.30
97-TU-13 4.80 18.30 76.80 93-LG-402 43.20 24.80 32.00
97-TU-14 3.70 15.50 80.80 93-LG-403 26.60 15.70 57.60
97-TU-15 5.90 11.80 82.20 LG Mean 29.04 25.00 45.94
97-TU-16 10.20 18.50 71.30 LG St Dev 5.78 4.17 5.54
97-TU-19 2.00 8.90 89.10
NU Mean 5.21 13.85 80.83 Hovsgol (Lugyn Gol Fm.)
NU St Dev 2.89 5.41 6.79 XO-1A 4.14 25.30
XO-1B 5.08 28.43 66.50
Naftgar Uul XO-1C 1.52 33.59 64.90
93-NG-1 15.20 41.52 43.27 XO-1D 6.57 28.71 64.72
93-NG-2 15.01 29.01 55.98 93-XO-3A 1.63 12.23 86.14
93-NG-3 18.25 31.22 50.53 93-XO-3B 1.70 14.16 84.14
93-NG-5 18.69 26.11 55.19 93-XO-3C 0.59 24.56 74.85
93-NG-6 12.00 39.50 48.50 93-XO-3E 5.23 18.90 75.87
93-NG-7 5.43 33.33 61.23 93-XO-3F 5.85 22.15 72.00
93-NG-9 15.79 34.59 49.62 93-XO-3G 1.56 17.19 81.25
93-NG-12 20.59 32.94 46.47 XO Mean 3.39 22.52 74.09
93-NG-13 20.17 40.34 39.49 XO St Dev 2.09 6.52 7.40
93-NG-14 14.91 35.50 49.59
Note: QmFLt normalized to 100%. Qm—monocrystalline quartz; F—total feldspar, Lt—total lithic fragments (see text for
explanation of methods); St Dev—standard deviation.
 Sedimentary response to arc-continent collision 383

linked to the southwestern, south-central, or southeastern study
areas, respectively. The dissected arc provenance indicated by
sandstone samples from Shin Jinst and Nomgon may in fact be an
artifact of alteration level. Dynamic, greenschist-grade metamor-
phism causes notable recrystallization and breakdown of labile
grains, including chert, calcite, and fine-grained lithic volcanic
fragments, into a slaty fabric matrix. This causes a relative com-
positional increase in the quartzofeldspathic fraction, which might
indicate a more transitional-arc origin for the original composi-
tion of these sandstones (gray arrow in Fig. 13). This may explain
the difference between Hovsgol (undissected arc) and Lugyn Gol
sand compositions, as the Lugyn Gol turbidite sand samples have
undergone a comparatively high degree of alteration.
By comparison, Naftgar Uul sandstone provenance samples
lack any metamorphic overprint. These samples are more com-
positionally mature, but they are also interpreted as possibly
younger (Triassic) deposits than the other localities plotted here.
They are distinctly richer in monocrystalline quartz. At Naftgar
Uul there is no statistical change in sandstone composition over
a 2000-m-thick sampling interval (Amory, 1996). In contrast,
sandstone composition at Noyon Uul changes drastically near the
Permian–Triassic boundary from lithic-rich to more quartzofeld-
spathic (Hendrix et al., 1996; note that only the Permian sample
compositions from Noyon Uul are presented here).
DISCUSSION
Tectonic Setting and Stratigraphic Evolution
Permian sedimentary basins of southern Mongolia formed
along the northern margin of an evolving collision zone between
amalgamated middle to late Paleozoic oceanic arc systems and
microcontinental blocks of north China. Upper Permian sedi-
mentary successions in southwestern Mongolia (e.g., Shin Jinst
and Noyon Uul localities) comprise entirely nonmarine deposits,
including meandering to braided fluvial environments. Similar
successions are found in south-central Mongolia (e.g., Naft-
gar Uul, Tavan Tolgoy, and Tsaagan Tolgoy), albeit with more
extensive coal measures. These deposits likely formed in sepa-
rate intramontane, flexural basins formed in response to collision
and continued crustal thickening along the western end of the
Tien Shan–Yin Shan suture (cf., Hendrix et al., 2001). In con-
trast, Upper Permian strata in southeast Mongolia (e.g., Nomgon
and Hovsgol) include distal turbidite sequences that record the
final phase of marine deposition in a closing, trapped ocean basin
immediately prior to final collision by Early Triassic time.
These regional facies relationships (summarized in Fig. 14)
broadly support diachronous collision along the Tien Shan–Yin
Shan suture zone. This interpretation is based mainly on the pro-
gression of youngest marine rocks from the middle Carboniferous
in southwestern Mongolia to the Late Permian in the southeast,
as well as evidence for earlier cessation of arc-related magmatism
in northwest China (Zhou et al., 2001; Cope et al., 2005). How-
ever, several caveats accompany this general interpretation. The
marine-to-nonmarine facies transition is not smoothly continu-
ous from west to east. The only classic flysch deposits preserved
in southeastern Mongolia are part of the Lugyn Gol Formation
(Nomgon and Hovsgol localities, Fig. 14). Carboniferous or
Lower Permian flysch deposits are not present in the study areas;
indeed, distal marine deposits predating the Upper Permian are
sparse throughout southern Mongolia (Lamb and Badarch, 2001).
Furthermore, Upper Carboniferous strata across the border in the

Mongol-Okhotsk ocean:
T1 marine sedimentation
in NE Mongolia

SW localities
(NU, SJ) SC localities T1 ? SE localities
(NA, TT, TsT) (NO, HO)
T1 P2
T1 Figure 14. Schematic summary of re-
Junggar: P2 P1
W-facing P1 P2 C2
gional tectonic and stratigraphic re-
P1 trapped C2 P1 lationships discussed in this study.
ocean basin? C2 EGFZ—East Gobi fault zone. South-
FZ E-facing P2 Solonker
T1 EG trapped ocean basin west (SW) localities: NU—Noyon Uul;
P2 Bulgan Uul:
offset P2 SJ—Shin Jinst. South-central (SC) lo-
P1 North Tie Yinshan
n Shan turbidites? calities: NA—Naftgar Uul; TT—Tavan
C2
Tolgoy; TsT—Tsaagan Tolgoy. South-
lt
Tarim fau Nanshan T1 east (SE) localities: NO—Nomgon;
agh
y nT North China P2 HO—Hovsgol. Age divisions are only
Alt block P1 approximate, based on available resolu-
C2 ? tion as discussed in the text.

T1 Lower Triassic nonmarine, arid climate facies 0 250 500 km

P2 Upper Permian nonmarine, humid climate facies
P1 Lower Permian marine strata (ocean basin) and/or oceanic arc sequences
C2 Upper Carboniferous
sediment dispersal direction
Inferred Late Paleozoic rotation direction.
384 Johnson et al.

Junggar basin may indicate west-directed closure and clockwise
rotation of Tarim, which suggests that perhaps Tarim and the North
China block were not colliding as a single block (Fig. 14). Thus the
sedimentary record appears to support Late Carboniferous–Early
Permian collision between the Altaids and Tarim, versus Late
Permian amalgamation between the Altaids and the North China
block; however, this does not seem to have occurred in a contigu-
ous east-west basin with simple progressive suturing proceeding
toward the east (Fig. 15).
Provenance data underscore that the progressive closure
model should not be oversimplified. A transition from undis-
sected-dissected arc to recycled orogen provenance is observed
as an unroofing trend in many arc-collisional settings (Dorsey,
1988; Clift et al., 2003). If closure of the Paleo-Asian ocean was
accommodated by progressive counterclockwise rotation and
northward movement of Tarim and North China blocks com-
bined, we might expect to see a west to east trend from dissected
to undissected arc provenance in our study areas indicating ear-
lier arc dissection in the west. No such trend was observed, nor is
there evidence for systematic cessation of arc activity from west
to east in Mongolia. Sand compositions represent recycling of
Carboniferous and Devonian arc systems (Graham et al., 1993;
Lamb and Badarch, 2001) in contrast to Permian sandstones in
northeast China, which range from dissected arc to recycled oro-
gen provenance (Fig. 13; Mueller et al., 1991; Cope et al., 2005).
This indicates a much stronger continental (quartzofeldspathic)
contribution to Permian basins along the southern margin of the
Tien Shan–Yin Shan suture in China and argues against exposure
of significant Precambrian crust in Mongolia (i.e., the hypoth-
esized South Gobi microcontinent). Finally, the lack of direct
evidence for Permian arc activity in southern Mongolia supports
south-dipping subduction beneath the North China block, at least
during this terminal collision phase.
One way of reconciling the simple regional interpretation
(of progressive west-to-east suturing) with the complexities
indicated by field data is to depart from some aspects of tra-
ditional remnant ocean-basin models. According to Ingersoll et
al. (1995, p. 363), “A remnant ocean basin is a shrinking ocean

closing
turbidite
fan system
ocean basin ESE
Altaids, southern Mongolia
SE C
SC
Yin Shan active
North magmatic arcs
SW B China
North (Nanshan) block
A Tien Shan
Tarim

subducting
WNW Altaid lithosphere

microcontinent crust

A Junggar basin and North Tien Shan, northwest China

- Late Carboniferous to Early Permian collision and closure of Junggar trapped ocean basin
from east to west during clockwise (?) rotation of Tarim.
B North Tien Shan--Yinshan suture zone
- Collisional zone formed during late Paleozoic closure of the Paleo-Asian ocean. The orogen
grows through bidirectional suturing, first westerly with closure of the Junggar ocean basin
(Late Carboniferous to Early Permian), then easterly with closure of the Solonker ocean (Early
to Late Permian). The orogenic belt comprises mainly uplifted Carboniferous-Devonian arc
sequences, which are the main source for sediment delivered by turbidite fan deposits in the
closing ocean basins to the west and east.

C Solonker ocean basin

- Early to Late Permian collision between the North China block and the Altaids of southeast
Mongolia. Suturing progresses eastward over time due to counterclockwise rotation of the
North China block relative to Mongolia. Sediment is mainly derived from the growing orogenic
belt to the west (including uplifted Carboniferous-Devonian arc sequences), and distributed in
the form of turbidite fan deposits within the trapped ocean basin.

Figure 15. Paleogeographic block diagram for the late Paleozoic of the China-Mongolia border zone, summarizing ob-
served tectonic and stratigraphic relationships discussed in this study. Cartoon is highly schematic, and no specific scale
is implied by the features shown. SW, SC, and SE refer to approximate locations of the southwest, south-central, and
southeast study areas in Mongolia, discussed in the text (cf. Fig. 14).
 Sedimentary response to arc-continent collision
basin, which is flanked by at least one convergent margin and
whose floor is typically covered by turbidites derived predomi-
nantly from associated suture zone(s).” Although this definition
broadly applies to the Permian setting in the China-Mongolia
border region, many aspects of the model are derived mainly
from end-member cases in which significant volumes of sediment
are deposited and preserved as turbidite fans—i.e., the Cenozoic
Himalayan-Bengal system and the late Paleozoic Appalachian-
Ouachita orogen (Graham et al., 1975). In these case studies,
sequential suturing of an irregular continental margin results in
transition from fine-grained marine flysch to coarse-grained ter-
restrial molasse in the direction of progressive collision (Dewey
and Burke, 1974). This predictive aspect of the remnant ocean
basin model may be broadly supported by the fact that turbi-
dite facies in southeastern Mongolia are mainly distal, outer fan
deposits, favoring long-distance funneling of submarine flows
outboard of a growing orogen. However, the flysch-molasse
transition is not well preserved across the entire suture zone, and
this part of the Altaids may be better represented by ongoing
intra-arc and microcontinent collisions with small trapped ocean
basins, as are found in parts of the modern southwest Pacific
Ocean (cf. Fig. 15).
Collision and Climate
Regional sedimentary facies relationships demonstrate chang-
ing climate regimes that accompany collision. In Mongolia, two
possible transitions from relatively arid climate to humid climate
conditions occur (Fig. 14). The older shift is poorly constrained but
seems to have occurred in the middle Permian at Noyon Uul and
Tavan Tolgoy, where red shale and sandstone beds generally rep-
resenting well-drained soils and oxidizing conditions are distinct
from overlying Upper Permian coal-bearing units. The younger
climatic shift is better constrained in southwest and south-central
Mongolia, where coal-bearing Upper Permian strata are overlain
by Lower Triassic rocks that lack extensive coal deposits. This
is illustrated by a dramatic record of floral transition to drought-
tolerant organisms at Tsaagan Tolgoy. As noted previously, this
sudden shift suggests a correlation with the global warming event
at the Permian–Triassic boundary (Retallack et al., 1996; Berner,
2002). However, orographic relief generated by local tectonic
uplift may also have influenced such climatic changes.
A similar aridification trend, though inferred to be some-
what older, is reported from northeast China, where Carbonifer-
ous and Lower Permian humid-climate facies were replaced by
arid-climate indicators that started in the Late Permian (Mueller
et al., 1991). The relationship is also present in northwest China,
based on a Late Permian transition from humid to arid environ-
ments (Carroll, 1991; Hendrix et al., 1992; Wartes et al., 2002;
Fig. 14). In the Yin Shan belt, the drying event is attributed either
to northward latitudinal migration of north China into subtropical
realms or to a rain shadow effect following development of high
topography during collision with the Altaids (Carroll et al., 1992;
Cope et al., 2005).
385
Data from Mongolia help to address these hypothesized cli-
matic forcing factors. Regarding the impact of northward plate
movement, southern Mongolia already had mixed Cathaysian-
Angaran floral assemblages by Late Permian time, underscoring
the influence of flora with Siberian and North China craton affini-
ties. However, paleomagnetic data still indicate a wide latitudinal
gap between Siberia and the North China block (and by inference
the amalgamated Mongolian arcs) extending into the Cretaceous
(Enkin et al., 1992). The sedimentary record in northeast Mongo-
lia does not support such a late collision with the Siberian craton,
although Lower Triassic marine rocks are present in northeast
Mongolia (Zonenshain et al., 1990; Badarch et al., 2002), indicat-
ing that final amalgamation with Siberia likely extended into the
early Mesozoic along this remnant ocean (the Mongol-Okhotsk
ocean, Fig. 14) to the north of the Tien Shan–Yin Shan suture. As
such, continued northward movement of both the North China
block and southern Mongolian arcs is still possible, at least
into the Triassic. However, the Permian–Triassic climatic shift
in southern Mongolia apparently postdates the middle Permian
event in China, whereas we would expect an earlier response to
the north if caused by such plate movement.
Although it is currently unclear how much crustal thicken-
ing and orographic relief accompanied collision in the southern
Altaids, Davis et al. (2004) have documented metamorphic core
complex formation in northeast China at ca. 220–190 Ma. This
may indicate collapse of the Permian–Early Triassic collisional
orogen and would support the presence of high topography
prior to collapse. Cope et al. (2005) suggested creation of a rain
shadow, resulting in red bed deposition, starting in the middle to
Late Permian. Interestingly, this coincides with Upper Permian
coal-bearing deposits in south-central Mongolia (Fig. 14), which
could also indicate trapping of moisture on the north side of an
inferred mountain range along the Tien Shan–Yin Shan suture.
The rain shadow interpretation seems permissible within the
range of available age constraints, although we note that turbidite
deposition continued in the Late Permian in southeast Mongolia
(directly north of the Yin Shan belt), suggesting slightly later col-
lision. In any case, the Early Triassic appears to have been a time
of arid climate across the region, which likely reflected the global
warming event at the Permian–Triassic boundary, perhaps modi-
fied by local topography.
IMPLICATIONS
If, as is argued here, the Tien Shan–Yin Shan suture links the
Tarim-Altaids and the North China block–Altaids collision zones,
the entire suture is >3000 km long and thus rivals some of the
longest accretionary zones known (e.g., Indian-Eurasia collision,
Ouachita-Appalachian orogen). The diffuse northern boundary
represented by the Altaids creates a particular challenge in delin-
eating the final suture zone: perhaps the best marker is the south-
ern boundary against known Precambrian-floored blocks, with
the understanding that multiple arc systems mark the collision
zone to the north. This record seems to be well represented by
386 Johnson et al.
modern accretionary processes in the southwestern Pacific ocean,
including the “triple-junction” region of convergence between
the Pacific-Philippine, Australia, and Southeast Asia plates (Hin-
schberger et al., 2005), with several magmatic arcs and micro-
plates occupying the Java, Molucca, and Banda Seas. Although
the geometry of this collision zone may not directly apply to the
Altaids, it is notable that the oceanic collision zone is marked
by multiple arc and backarc basins between major plates that are
moving independently, which may be analogous to the Siberian
craton, the Tarim–North Tien Shan, and the North China block
in the late Paleozoic. Multiple trapped ocean basins are present
along the Tien Shan–Yin Shan suture, but evidence is lacking for
continuous zipperlike regional closure (Fig. 15). This broad com-
parison is supported by other modern Pacific arc-collision ana-
logues where climatic effects of orogeny are documented (e.g., a
present-day rain shadow effect in southern Papua New Guinea) as
well as the evolution to syn- or postcollisional collapse and cre-
ation of metamorphic core complexes, as in Taiwan (Clift et al.,
2003; Huang et al., 2006). Perhaps the main strength of Permian
studies near the China-Mongolia border is that this region pro-
vides a good ancient analogue for such modern systems and pre-
dicts what the southwest Pacific might look like following termi-
nal collision and closure over the next tens of millions of years.
Strike-slip deformation is also an important aspect of the
long-term results of these accretionary complexes. The role of
syncollisional strike-slip faults in accommodating rotational clo-
sure and oblique collisions is noted in modern analogs (Moore and
Silver, 1983; Hinschberger et al., 2005; Huang et al., 2006). Late
Paleozoic strike-slip deformation has been documented in parts
of the Altaids (Şengör and Natal’in, 1996; Laurent-Charvet et al.,
2003; Zhou et al., 2001) but has not yet been well documented
in the Permian of Mongolia. Reconstruction of syncollisional
paleogeography is complicated by later strike-slip deformation
that has only recently been recognized in southern Mongolia. At
least two phases of left-lateral movement are documented along
the East Gobi fault zone (Figs. 1, 2). The first phase occurred in
the latest Triassic (ca. 207–209 Ma) associated with a northeast-
trending mylonite zone that is continuous for >250 km (Lamb et
al., 1999). This was followed by Late Cretaceous–Cenozoic reac-
tivation via brittle sinistral faulting (Johnson, 2004). The amount
of offset associated with these sinistral movements is poorly con-
strained but likely represents ~200–400 km of total slip (Webb
and Johnson, 2006). One important implication of this postcol-
lisional deformation is that attempts to reconstruct the southern
margin of the Altaids by conducting mainly west-to-east corre-
lations may incorrectly assume that these units were originally
formed at the same latitude. For example, turbidite deposits in the
Nomgon area may have equivalents along the southernmost bor-
der of Mongolia at Bulgan Uul (Figs. 2, 14; Lamb et al., 1999),
and thus the record of continued marine deposition may extend
farther westward than currently realized.
It is common for convergent margins to be reactivated as
transform or strike-slip settings from shifting convergence direc-
tions along a plate margin (e.g., Cenozoic of California; Graham,
1978) or from intracontinental deformation within an accretion-
ary complex (Tapponnier et al., 1982). The latter seems to have
been the case in Mongolia, where these two phases of sinistral
motion have been tentatively linked to post-Paleozoic collisions
along the southward-growing margin of Asia, including the
Cenozoic India-Asia collision >2000 km from the East Gobi fault
zone. Cenozoic strike-slip is also demonstrated in southwestern
Mongolia (Cunningham et al., 2003). Rather than following a
single terrane boundary, this intracontinental fault system seems
to exploit multiple zones of weakness within heterogeneous crust
of the Altaids. Thus consideration of syn-and postcollisional
strike-slip deformation is critical to reconstruction of these col-
lisional zones, and, conversely, better constraints on the location
and nature of arc-collision zones help to predict modes and loca-
tions of subsequent deformation.
CONCLUSIONS
Stratigraphic data presented here are a first step toward
unraveling the history of arc-arc and ultimately arc-continent col-
lision that is dramatically represented in the southern Altaids, par-
ticularly from north of the China-Mongolia border. Future studies
that attempt to further constrain the timing and modes of conti-
nental growth by accretion, both in this part of Asia and in analo-
gous settings worldwide, should take into account several key
constraints afforded by this study. Regional relations that extend
some 3000 km from northwest China, through south Mongolia,
to northeast China, argue for broad correlation of the Tien Shan–
Yin Shan sutures. The main suture zone is best drawn along the
northern margin of known continental blocks; tectonic boundar-
ies that extend into the Altaids represent only extensions of this
broad collision zone into a heterogeneous accretionary margin.
Generally south-dipping subduction during the Permian is
supported by the lack of active Permian arcs in southern Mongo-
lia, and by complementary studies of active Permian arc systems
in north China. In very general terms, west-to-east diachronous
closure may be argued mainly on the basis of the age of the young-
est marine strata from the Tien Shan to the Yin Shan (from Upper
Carboniferous to Upper Permian). However, paleogeographic
reconstructions based on sedimentary facies and paleocurrent
data, sandstone petrofacies, and dating of magmatic activity indi-
cate that the collisional event was not a simple progressive sutur-
ing process. There is no record of progressive arc dissection in
the Permian from southwest to southeast Mongolia, nor is there
a continuously eastward-younging flysch to molasse transition.
Sandstone provenance studies argue for very little input from
continental crust or recycled orogen sources into Permian basins
of Mongolia, in contrast to analogous deposits in north China.
Final arc-continent accretion may have included both west- and
east-facing trapped oceans and differential rotation of the North
China block and Tarim, possibly driven by uneven colliding mar-
gins with indenters that are currently poorly constrained. Thus
Permian strata of southern Mongolia are best considered an end-
member representation of published models for remnant ocean
 Sedimentary response to arc-continent collision
basin closure. This record also provides a predictive glimpse into
future results of arc-arc and arc-continent collisions in the present
southwest Pacific Ocean.
The possible implications of this collisional event are sig-
nificant. Regional climatic shifts are linked to evolution of the
accretionary belt, including the possibility of a rain shadow effect
south of the collisional orogen, which may have subsequently
collapsed during the Triassic. Similar impacts are reported in
modern arc-continent collisions such as Taiwan and Papua New
Guinea. Finally, strike-slip deformation is an important syncol-
lision mechanism to consider as a way of accommodating rota-
tional closure. The recognition of subsequent strike-slip defor-
mation is also critical to reconstruction of the suture zone, which
in southern Mongolia is now partly offset and dismembered
along the East Gobi fault zone. It is likely that multiple phases of
post-Permian movement along this fault system were driven by
continued collisions along the growing margin of Asia, and thus
the long-term record of intracontinental deformation is also tied
to this region as an inherited zone of crustal weakness.
ACKNOWLEDGMENTS
This work was supported by U.S. National Science Founda-
tion grants to Graham (EAR-9614555 and EAR-9708207) and
Johnson (EAR-0537318, Tectonics and OISE, L.E. Webb, co-
PI), and funding from the American Chemical Society (Petro-
leum Research Fund 40193-G8 to Johnson). Funds from Stan-
ford University also supported this work, including the Gradu-
ate Fellowship program, the McGee Fund, and the Stanford
Undergraduate Research Opportunity Grant. The American
Association of Petroleum Geologists and the Geological Soci-
ety of America supported field seasons in Mongolia through
several student grants to the authors. We thank Ch. Minjin and
colleagues at the Mongolian University of Science and Tech-
nology for their support, and M.S. Hendrix and L.E. Webb for
many years of collaboration. D. Zinniker thanks S. Fowell and
G. Norris for assistance in paleopalynology. Countless field
assistants and colleagues assisted in field work, including A.
Chimitsuren, J. Crider, M. Heumann, A. Keller, T. Hickson, N.
Manchuk, G. Sersmaa, E. Sobel, and J. Undariya. We appreci-
ate thoughtful and thorough reviews by A. Carroll, R. Dorsey,
M. Faure, and P. Clift, which improved the final manuscript.
This study is dedicated to our colleague, advisor, and dear
friend Gombosuren Badarch, who died during preparation of
this manuscript. This work would not have been possible with-
out Badarch’s guidance, expertise, and patience. We miss him
deeply, but his love for Gobi field work and the geology of Asia
continues to inspire us.
REFERENCES CITED
Allen, M.B., Windley, B.F., Zhang, C., Zhao, Z., and Wang, R.-G., 1991, Basin
evolution within and adjacent to the Tien Shan Range, NW China: Geo-
logical Society [London] Journal, v. 148, p. 369–378.
387
Allen, M.B., Windley, B.F., and Zhang, C., 1993, Palaeozoic collisional tecton-
ics and magmatism of the Chinese Tien Shan, central Asia: Tectonophys-
ics, v. 220, p. 89–115, doi: 10.1016/0040-1951(93)90225-9.
Amory, J., 1996, Permian sedimentation and tectonics of southern Mongolia
[M.S. thesis]: Stanford, California, Stanford University, 183 p.
Amory, J., Hendrix, M., Lamb, M., Keller, A., Badarch, G., and Tomurtogoo,
O., 1994, Permian sedimentation and tectonics of southern Mongolia;
implications for a time-transgressive collision with North China: Geologi-
cal Society of America Abstracts with Programs, v. 26, no. 7, p. 242.
Anatoleva, A.I., 1974, The structure and composition of the Permian red vol-
canogenic-sedimentary complex in southern Mongolia: Geologiya i Geo-
fizika, v. 15, p. 32–43.
Aristov, D.S., 2005, New Grylloblattids (Insecta: Gryllobattida) from the Trias-
sic of Eastern Europe, Eastern Kazakstan, and Mongolia: Paleontological
Journal, v. 39, p. 173–177.
Badarch, G., Cunningham, W.D., and Windley, B.F., 2002, A new terrane sub-
division for Mongolia; implications for the Phanerozoic crustal growth
of Central Asia: Journal of Asian Earth Sciences, v. 21, p. 87–110, doi:
10.1016/S1367-9120(02)00017-2.
Battacharya, J.P., and Walker, R.G., 1992, Deltas, in Walker, R.G., and James,
N.P., eds., Facies Models: St. Johns, Newfoundland, Geological Associa-
tion of Canada, p. 157–178.
Baud, A., Magaritz, M., and Holser, W.T., 1989, Permian–Triassic of the Tethys;
carbon isotope studies: Geologische Rundschau, v. 78, p. 649–677, doi:
10.1007/BF01776196.
Berkey, C.P., and Morris, F.K., 1927, Geology of Mongolia, in Natural History
of Central Asia, Volume 2: New York, American Museum of Natural His-
tory, 475 p.
Berner, R.A., 2002, Examination of hypothesis for the Permo–Triassic bound-
ary extinction by carbon cycle modeling: Proceedings of the National
Academy of Sciences of the United States of America, v. 99, p. 4172–
4177, doi: 10.1073/pnas.032095199.
Burchfiel, B.C., Deng, Q., Molnar, P., Royden, L., Wang, Y., Zhang, P.,
and Zhang, W., 1989, Intracrustal detachment within zones of conti-
nental deformation: Geology, v. 17, p. 748–752, doi: 10.1130/0091-
7613(1989)017<0448:IDWZOC>2.3.CO;2.
Carroll, A.R., 1991, Late Paleozoic tectonics, sedimentation, and petroleum
potential of the Junggar and Tarim basins, northwest China [Ph.D. thesis]:
Stanford, California, Stanford University, 405 p.
Carroll, A.R., Liang, Y., Graham, S.A., Xiao, X., Hendrix, M.S., Chu, J., and
McKnight, C., 1990, Junggar basin, northwest China: Trapped Late
Paleozoic ocean: Tectonophysics, v. 181, p. 1–14, doi: 10.1016/0040-
1951(90)90004-R.
Carroll, A.R., Brassell, S.C., and Graham, S.A., 1992, Upper Permian lacus-
trine oil shales, southern Junggar basin, northwest China: American Asso-
ciation of Petroleum Geologists Bulletin, v. 76, p. 1874–1902.
Carroll, A.R., Graham, S.A., Hendrix, M.S., Ying, D., and Zhou, D., 1995, Late
Paleozoic tectonic amalgamation of northwestern China: Sedimentary
record of the northern Tarim, northwestern Turpan, and southern Junggar
basins: Geological Society of America Bulletin, v. 107, p. 571–594, doi:
10.1130/0016-7606(1995)107<0571:LPTAON>2.3.CO;2.
Chen, B., Jahn, B.-M., Wilde, S., and Xu, B., 2000, Two contrasting Paleo-
zoic magmatic belts in northern Inner Mongolia, China: Petrogenesis and
tectonic implications: Tectonophysics, v. 328, p. 157–182, doi: 10.1016/
S0040-1951(00)00182-7.
Clift, P.D., Schouten, H., and Draut, A.E., 2003, A general model of arc-conti-
nent collision and subduction polarity reversal from Taiwan and the Irish
Caledonides, in Larter, R.D. and Leat, P.T., eds., Intra-oceanic Subduction
Systems: Tectonic and Magmatic Processes: Geological Society [London]
Special Publication 219, p. 81–98.
Coleman, R.G., 1989, Continental growth of northwest China: Tectonics, v. 8,
p. 621–635.
Cope, T., Ritts, B.G., Darby, B.J., Fildani, A., and Graham, S.A., 2005, Late
Paleozoic sedimentation on the northern margin of the North China
Block; implications for regional tectonics and climate change: Interna-
tional Geology Review, v. 47, p. 270–296.
Cunningham, D., Dijkstra, A.H., Howard, J., Quarles, A., and Badarch, G.,
2003, Active intraplate strike-slip faulting and transpressional uplift in the
Mongolian Altai in Storti, F., Holdsworth, R.E., and Salvini, F., Intraplate
Strike-slip Deformation Belts: Geological Society [London] Special Pub-
lication 210, p. 65–87.
388 Johnson et al.
Davis, G.A., Xu, B., Zheng, Y., and Zhang, W., 2004, Indosinian extension in
the Solonker suture zone: The Sonid Zuoqi metamorphic core complex,
Inner Mongolia, China: Beijing, China University of Geosciences, Earth
Science Frontiers, v. 11, p. 135–143.
Dewey, J.F., and Burke, K., 1974, Hot spots and continental break-up: Implica-
tions for collisional orogeny: Geology, v. 2, p. 57–60, doi: 10.1130/0091-
7613(1974)2<57:HSACBI>2.0.CO;2.
Dickinson, W.R., 1970, Interpreting detrital modes of graywacke and arkose:
Journal of Sedimentary Petrology, v. 40, p. 695–707.
Dickinson, W.R., and Suczek, C.A., 1979, Plate tectonics and sandstone com-
positions: American Association of Petroleum Geologists Bulletin, v. 63,
p. 2164–2182.
Dickinson, W.R., Beard, L.S., Brakenridge, G.R., Erjavec, J.L., Ferguson, R.C.,
Inman, K.F., Knepp, R.A., Lindberg, F.A., and Ryberg, P.T., 1983, Prov-
enance of North American Phanerozoic sandstones in relation to tectonic
setting: Geological Society of America Bulletin, v. 94, p. 222–235, doi:
10.1130/0016-7606(1983)94<222:PONAPS>2.0.CO;2.
Dobretsov, N.L., Coleman, R.G., and Berzin, N.A., eds., 1994, Geodynamic
Evolution of the Paleoasian Ocean: Russian Geology and Geophysics,
v. 35, 233 p.
Dorsey, R.J., 1988, Provenance evolution and unroofing history of a modern
arc-continent collision; evidence from petrography of Plio–Pleistocene
sandstones, eastern Taiwan: Journal of Sedimentary Petrology, v. 58,
p. 208–218.
Dubiel, R.F., and Smoot, J.P., 1994, Criteria for interpreting paleoclimate from
red beds: A tool for Pangean reconstructions, in Embry, A.F., Beauchamp,
B., and Glass, D.J., Pangea: Global Environments and Resources: Cana-
dian Society of Petroleum Geology Memoir 17, p. 295–310.
Dumitru, T., Zhou, D., Chang, R., Graham, S., Hendrix, M., Sobel, E., and Car-
roll, A., 2001, Uplift, exhumation, and deformation in the Chinese Tian
Shan, in Hendrix, M.S., and Davis, G.A., Paleozoic and Mesozoic Tec-
tonic Evolution of Central and Eastern Asia: From Continental Assembly
to Intracontinental Deformation: Geological Society of America Memoir
194, p. 71–99.
Durante, M.V., 1971, Late Permian flora of Mongolia and the southern bound-
ary of the Angaran floristic region of that time: Paleontological Journal,
v. 4, p. 511–522.
Durante, M.V., 1976, The Carboniferous and Permian Stratigraphy of Mongolia
on the Basis of Paleobotanical Data: Joint Soviet-Mongolian Scientific
Research Geological Expedition, v. 19, 280 p.
Durante, M.V., 1983, Existence of an Upper Permian mixed Cathaysio-Angar-
ian flora in Nanshan (North China): Geobios, v. 16, p. 241–242, doi:
10.1016/S0016-6995(83)80022-9.
Enkin, R.J., Yang, Z., Chen, Y., and Courtillot, V.E., 1992, Paleomagnetic con-
straints on the geodynamic history of the major blocks of China from the
Permian to the present: Journal of Geophysical Research, B, Solid Earth
and Planets, v. 97, p. 13,953–13,989.
Faucett, P.J., Barron, E.J., Robison, V.D., and Katz, B., 1994, The climatic evo-
lution of India and Australia from the Late Permian to mid-Jurassic: A
comparison of climate model results within the geologic record, in Klein,
G.D., ed., Pangea: Paleoclimate, Tectonics, and Sedimentation during
Accretion, Zenith and Breakup of a Supercontinent: Geological Society
of America Special Paper 288, p. 139–157.
Gilder, S., Zhao, X., Coe, R., Meng, Z., Courtillot, V., and Besse, J., 1996,
Paleomagnetism and tectonics of the southern Tarim basin, northwestern
China: Journal of Geophysical Research, v. 101, p. 22,015–22,031, doi:
10.1029/96JB01647.
Grabau, A.W., 1931, The Permian of Mongolia, in Natural History of Central
Asia, Volume 4: New York, American Museum of Natural History, 665 p.
Gradstein, F.M., Ogg, J.G., et al., 2004, A Geologic Time Scale 2004: Cam-
bridge, UK, Cambridge University Press, 589 p.
Graham, S.A., 1978, Role of Salinian Block in evolution of San Andreas fault
system, California: American Association of Petroleum Geologists Bul-
letin, v. 62, p. 2214–2231.
Graham, S.A., 1996, Controls on intracontinental deformation in Central Asia:
Geological Society of America Abstracts with Programs, v. 28, p. 112.
Graham, S.A., Dickinson, W.R., and Ingersoll, R.V., 1975, Himalayan-Bengal
model for flysch dispersal in Appalachian-Ouachita system: Geologi-
cal Society of America Bulletin, v. 86, p. 273–286, doi: 10.1130/0016-
7606(1975)86<273:HMFFDI>2.0.CO;2.
Graham, S.A., Hendrix, M.S., Wang, L.B., and Carroll, A.R., 1993, Collisional
successor basins of western China; impact of tectonic inheritance on sand
composition: Geological Society of America Bulletin, v. 105, p. 323–344,
doi: 10.1130/0016-7606(1993)105<0323:CSBOWC>2.3.CO;2.
Greene, T.J., Carroll, A., Hendrix, M., Graham, S., Wartes, M., and Abbink, O.,
2001, Sedimentary record of Mesozoic deformation and inception of the
Turpan-Hami Basin, Northwest China, in Hendrix, M.S., and Davis, G.A.,
eds., Paleozoic and Mesozoic Tectonic Evolution of Central and Eastern
Asia: From Continental Assembly to Intracontinental Deformation: Geo-
logical Society of America Memoir 194, p. 317–340.
Gubin, Y.M., and Sinitza, S.M., 1993, Triassic terrestrial tetrapods of Mon-
golia and the geological structure of the Sain-Sar-Bulak locality: Bul-
letin of the New Mexico Museum of Natural History and Science, v. 3,
p. 169–170.
Harrison, T.M., Copeland, P., Kidd, W.S.F., and Yin, A., 1992, Raising Tibet:
Science, v. 255, p. 1663–1670, doi: 10.1126/science.255.5052.1663.
Hendrix, M., 2000, Evolution of Mesozoic sandstone compositions, southern
Junggar, northern Tarim, and western Turpan basins, Northwest China;
a detrital record of the ancestral Tian Shan: Journal of Sedimentary
Research, v. 70, p. 520–532.
Hendrix, M., Graham, S., Carroll, A., Sobel, E., McKnight, C., Schulein, B., and
Wang, Z., 1992, Sedimentary record and climatic implications of recur-
rent deformation in the Tian Shan; evidence from Mesozoic strata of the
north Tarim, south Junggar, and Turpan basins, Northwest China: Geo-
logical Society of America Bulletin, v. 104, p. 53–79, doi: 10.1130/0016-
7606(1992)104<0053:SRACIO>2.3.CO;2.
Hendrix, M.S., Graham, S.A., Amory, J.Y., and Badarch, G., 1996, Noyon Uul
Syncline, southern Mongolia; lower Mesozoic sedimentary record of the
tectonic amalgamation of Central Asia: Geological Society of America
Bulletin, v. 108, p. 1256–1274, doi: 10.1130/0016-7606(1996)108<1256:
NUSSML>2.3.CO;2.
Hendrix, M., Beck, M., Badarch, G., and Graham, S., 2001, Triassic synoro-
genic sedimentation in southern Mongolia: Early effects of intracontinen-
tal deformation, in Hendrix, M.S., and Davis, G.A., eds., Paleozoic and
Mesozoic Tectonic Evolution of Central and Eastern Asia: From Conti-
nental Assembly to Intracontinental Deformation: Geological Society of
America Memoir 194, p. 199–214.
Heubeck, C., 2001, Assembly of central Asia during the middle and late
Paleozoic, in Hendrix, M.S., and Davis, G.A., eds., Paleozoic and Meso-
zoic Tectonic Evolution of Central and Eastern Asia: From Continental
Assembly to Intracontinental Deformation: Geological Society of Amer-
ica Memoir 194, p. 1–22.
Hinschberger, F., Malod, J.A., Rehault, J.P., Villeneuve, M., Royer, J.Y.,
and Burhanuddin, S., 2005, Late Cenozoic geodynamic evolution of
eastern Indonesia: Tectonophysics, v. 404, p. 91–118, doi: 10.1016/
j.tecto.2005.05.005.
Huang, B., Otofuiji, Y., Zhu, R., Shi, R., and Wang, Y., 2001, Paleomagnetism
of Carboniferous sediments in the Hexi Corridor; its origin and tectonic
implications: Earth and Planetary Science Letters, v. 194, p. 135–149,
doi: 10.1016/S0012-821X(01)00557-X.
Huang, C.-Y., Yuan, P.B., and Tsao, S.-J., 2006, Temporal and spatial records of
active arc-continent collision in Taiwan: A synthesis: Geological Society
of America Bulletin, v. 118, p. 274–288, doi: 10.1130/B25527.1.
Ingersoll, R.V., Fullard, T.F., Ford, R.L., Grimm, J.P., Pickle, J.D., and Sares,
S.W., 1984, The effect of grain size on detrital modes; a test of the Gazzi-
Dickinson point-counting method: Journal of Sedimentary Petrology,
v. 54, p. 103–116.
Ingersoll, R.V., Graham, S.A., and Dickinson, W.R., 1995, Remnant ocean
basins, in Ingersoll, R.V., and Busby, C.J., eds., Tectonics of Sedimentary
Basins: Malden, Massachusetts, Blackwell Science, p. 363–392.
Jahn, B., 1999, Introduction to IGCP-420; Continental growth in the Phanero-
zoic; evidence from the Central Asian orogenic belt (CAOB): Mémoires
de Géosciences Rennes, Special Issue no. 2, p. 1–5.
Johnson, C.L., 2004, Polyphase evolution of the East Gobi basin: Sedimen-
tary and structural records of Mesozoic–Cenozoic intraplate deforma-
tion in Mongolia: Basin Research, v. 16, p. 79–99, doi: 10.1111/j.1365-
2117.2004.00221.x.
Johnson, C.L., Webb, L.E., Graham, S.A., Hendrix, M.A., and Badarch, G.,
2001, Sedimentary and structural records of late Mesozoic high-strain
extension and strain partitioning, East Gobi basin, southern Mongolia, in
Hendrix, M.S., and Davis, G.A., eds., Paleozoic and Mesozoic Tectonic
Evolution of Central and Eastern Asia: From Continental Assembly to
Intracontinental Deformation: Geological Society of America Memoir
194, p. 413–434.
 Sedimentary response to arc-continent collision
Krull, E.S., and Retallack, G.J., 2000, δ13C depth profiles from paleosols across
the Permian–Triassic boundary: Evidence for methane release: Geologi-
cal Society of America Bulletin, v. 112, p. 1459–1472, doi: 10.1130/0016-
7606(2000)112<1459:CDPFPA>2.0.CO;2.
Lamb, M.A., 1998, Paleozoic sedimentation, volcanism, and tectonics of south-
ern Mongolia [Ph.D. thesis]: Stanford, California, Stanford University,
207 p.
Lamb, M.A., and Badarch, G., 2001, Paleozoic sedimentary basins and vol-
canic arc systems of southern Mongolia: New geochemical and petro-
graphic constraints, in Hendrix, M.S., and Davis, G.A., eds., Paleozoic
and Mesozoic Tectonic Evolution of Central and Eastern Asia: From Con-
tinental Assembly to Intracontinental Deformation: Geological Society of
America Memoir 194, p. 117–147.
Lamb, M.A., Hanson, A.D., Graham, S.A., Badarch, G., and Webb, L.E., 1999,
Left-lateral sense offset of upper Proterozoic to Paleozoic features across
the Gobi Onon, Tost, and Zuunbayan faults in southern Mongolia and
implications for other Central Asian faults: Earth and Planetary Science
Letters, v. 173, p. 183–194, doi: 10.1016/S0012-821X(99)00227-7.
Landuydt, C.J., 1990, Micromorphology of iron minerals from bog ordes of
the Beliian Camine area, in Douglas, L.A., ed., Soil Micromorphology: A
Basic and Applied Science: Amsterdam, Elsevier, p. 289–294.
Laurent-Charvet, S., Charvet, J., Monie, P., and Shu, L., 2003, Late Paleo-
zoic strike-slip shear zones in eastern Central Asia (NW China); new
structural and geochronological data: Tectonics, v. 22, no. 2, doi:
10.1029/2001TC901047.
Li, J.-Y., 2006, Permian geodynamic setting of Northeast China and adjacent
regions: Closure of the Paleo-Asian Ocean and subduction of the Paleo-
Pacific: Journal of Asian Earth Sciences, v. 26, p. 207–224, doi: 10.1016/
j.jseaes.2005.09.001.
Li, Y., McWilliams, M., Cox, A., Sharps, R., Li, Y., Gao, Z., Zhang, Z., and
Zhai, Y., 1988, Late Permian paleomagnetic pole from dikes of the
Tarim craton, China: Geology, v. 16, p. 275–278, doi: 10.1130/0091-
7613(1988)016<0275:LPPPFD>2.3.CO;2.
Manankov, I.N., 1988, Late Permian productida (Brachiopoda) from southeast-
ern Mongolia: Paleontological Journal, v. 32, p. 486–492.
Manankov, I.N., 2004, New species of Early Permian brachiopods and biostra-
tigraphy of the boreal basin of Mongolia: Paleontological Journal, v. 38,
p. 366–372.
Manankov, I.N., Shi, G.R., and Shen, S.-J., 2006, An overview of Permian
marine stratigraphy and biostratigraphy of Mongolia: Journal of Asian
Earth Sciences, v. 26, p. 294–303, doi: 10.1016/j.jseaes.2005.11.008.
Marsaglia, K.M., 1992, Petrography and provenance of volcaniclastic sands
recovered from the Izu-Bonin arc and the Mariana arc, Leg 126: Proceed-
ings of the Ocean Drilling Program, Scientific Results, v. 126, p. 139–154.
Marsaglia, K.M., and Ingersoll, R.V., 1992, Compositional trends in arc-related,
deep-marine sand and sandstone; a reassessment of magmatic-arc prov-
enance: Geological Society of America Bulletin, v. 104, p. 1637–1649,
doi: 10.1130/0016-7606(1992)104<1637:CTIARD>2.3.CO;2.
McElhinny, M.W., Embleton, B.J.J., Ma, X.A., and Zhang, Z.K., 1981, Frag-
mentation of Asia in the Permian: Nature, v. 293, p. 212–215, doi:
10.1038/293212a0.
Moore, G.F., and Silver, E.A., 1983, Collisional processes in the northern
Molucca Sea, in Hayes, D.E., ed., The Geology and Tectonics of East and
Southeast Asian Seas: Part 2: Geophysical Monograph, v. 27, p. 360–372.
Mossakovsky, A.A., and Tomurtogoo, O., 1976, Upper Paleozoic of Mongo-
lia: Joint Soviet Mongolian Scientific Research Geological Expedition,
Transactions, v. 15, 126 p.
Mueller, F., Rogers, J.J.W., Jin, Y., Wang, H., Li, W., Chronic, J., and Mueller,
J., 1991, Late Carboniferous to Permian sedimentation in Inner Mongolia,
China, and tectonic relationships between North China and Siberia: Jour-
nal of Geology, v. 99, p. 251–263.
Mutti, E., and Ricci Lucchi, F., 1978, Turbidites of the Northern Apennines;
Introduction to Facies Analysis: Falls Church, Virginia, American Geo-
logical Institute, 166 p.
Nie, S.-Y., Rowley, D.B., and Ziegler, A.M., 1990, Constraints on the locations
of Asian microcontinents in Palaeo-Tethys during the Late Paleozoic, in
McKerrow, W.S., and Scotese, C.R., Palaeozoic Palaeogeography and
Biogeography: Geological Society [London] Memoir, 12, p. 397–409.
Nikolov, Z., Ehehbum, C., Shishkov, G., and Sallabasheva, V., 1981, Novye
dannye o geologiy i petrologiy kamennougol’nogo mestorozhdeniya
“Tavan Tolgoy” y yuzhnoj Gobi Mongol’skoy narodnoy respubliki—New
data on the geology and petrology of the “Tavan Tolgoy” coal deposit
389
of southern Gobi in Mongolia: Spisanie na Bulgarskogo Geologichesko
Druzhestvo, v. 42, p. 107–117.
Ouyang, S., and Norris, G., 1999, Earliest Triassic (Induan) spores and pol-
len from the Junggar Basin, Xinjiang, northwestern China: Review of
Palaeobotany and Palynology, v. 106, p. 1–56, doi: 10.1016/S0034-
6667(98)00078-5.
Parrish, J.T., Ziegler, A.M., and Scotese, C.R., 1982, Rainfall patterns and the
distribution of coals and evaporites in the Mesozoic and Cenozoic: Pal-
aeogeography, Palaeoclimatology, Palaeoecology, v. 40, p. 67–101.
Pavlova, E.E., Manankov, I.N., Morozova, I.P., Solov’yeva, M.N., Suyetenko,
O.D., and Bogoslovskaya, M.F., 1991, Permian invertebrates of southern
Mongolia: Joint Soviet Mongolian Scientific Research Geological Expe-
dition, Transactions, v. 40, 173 p.
Posamentier, H.W., and Allen, G.P., 1999, Siliciclastic Sequence Stratigraphy–
Concepts and Applications: SEPM (Society for Sedimentary Geology),
Concepts in Sedimentology and Paleontology no. 7, 210 p.
Ren, J., Jiang, C., Zhang, Z., Qin, D., and Huang, T.K., 1987, Geotectonic Evo-
lution of China: Beijing, Science Press, 217 p.
Retallack, G.J., Veevers, J.J., and Morante, R., 1996, Global coal gap between
Permian–Triassic extinctions and middle Triassic recovery of peat form-
ing plants: Geological Society of America Bulletin, v. 108, p. 195–207,
doi: 10.1130/0016-7606(1996)108<0195:GCGBPT>2.3.CO;2.
Ritts, B.D., Yue, Y.J., and Graham, S.A., 2004, Oligocene–Miocene tectonics
and sedimentation along the Altyn Tagh Fault, northern Tibetan Plateau:
Analysis of the Xorkol, Subei, and Aksay basins: Journal of Geology,
v. 112, p. 207–229, doi: 10.1086/381658.
Roger, F., Arnaud, N., Gilder, S., Tapponnier, P., Jolivet, M., Brunel, M., Mala-
vieille, J., Xu, Z., and Yang, J., 2003, Geochronological and geochemical
constraints on Mesozoic suturing in east central Tibet: Tectonics, v. 22,
p. 1037, doi: 10.1029/2002TC001466.
Rowley, D., 1996, Age of initiation and collision between India and Asia: A
review of stratigraphic data: Earth and Planetary Science Letters, v. 145,
p. 1–13, doi: 10.1016/S0012-821X(96)00201-4.
Şengör, A.M.C., and Natal’in, B.A., 1996, Paleotectonics of Asia: Fragments of
a synthesis, in Yin, A., and Harrison, M., eds., The Tectonic Evolution of
Asia: Cambridge, UK, Cambridge University Press, p. 486–640.
Şengör, A.M.C., Natal’in, B.A., and Burtman, V.S., 1993, Evolution of the
Altaid tectonic collage and Palaeozoic crustal growth in Eurasia: Nature,
v. 364, p. 299–307, doi: 10.1038/364299a0.
Shen, S.-Z., Zhang, H., Shang, Q.H., and Li, W.-Z., 2006, Permian stratigraphy
and correlation of Northeast China: A review: Journal of Asian Earth Sci-
ences, v. 26, p. 304–326, doi: 10.1016/j.jseaes.2005.07.007.
Shi, G.R., 2006, The marine Permian of East and Northeast Asia: An over-
view of biostratigraphy, palaeobiogeography and palaeogeographi-
cal: Journal of Asian Earth Sciences, v. 26, p. 175–206, doi: 10.1016/
j.jseaes.2005.11.004.
Tang, K., 1990, Tectonic development of Paleozoic foldbelts at the north mar-
gin of the Sino-Korean craton: Tectonics, v. 9, p. 249–260.
Tapponnier, R., Peltzer, G., Le Dain, A.Y., Armijo, R., and Cobbold, P., 1982,
Propagating extrusion tectonics in Asia; new insights from simple experi-
ments with plasticine: Geology, v. 10, p. 611–616, doi: 10.1130/0091-
7613(1982)10<611:PETIAN>2.0.CO;2.
Tomurtogoo, O., 1999, Geological Map of Mongolia, 1:1,000,000 scale:
Ankara, Turkey, General Directorate of Mineral Research and Explora-
tion, 1 sheet.
Uranbileg, L., 2003, The new plants of Upper Permian coal deposits in southern
Mongolia: Mongolian Geoscientist, v. 23, p. 47–50.
Vakrameyev, V.A., Lebedev, Y.L., and Sodov, Z., 1986, A cycad(?) Guramsania
gen. nov. from the Upper Permian of South Mongolia: Paleontological
Journal, v. 20, p. 95–101.
Vincent, S.J., and Allen, M.B., 2001, Sedimentary record of Mesozoic intrac-
ontinental deformation in the eastern Junggar Basin, northwestern China;
response to orogeny at the Asian margin, in Hendrix, M.S., and Davis,
G.A., eds., Paleozoic and Mesozoic Tectonic Evolution of Central and
Eastern Asia: From Continental Assembly to Intracontinental Deforma-
tion: Geological Society of America Memoir 194, p. 341–360.
Walker, R.G., 1992, Turbidites and submarine fans, in Walker, R.G., and James,
N.P., eds., Facies Models: St. Johns, Newfoundland, Geological Associa-
tion of Canada, p. 239–265.
Wang, B., Faure, M., Cluzel, D., Shu, L., and Charvet, J., 2005, Tectonic evo-
lution of the northern part of the Yili Block (Western Chinese Tianshan),
in Sklyarov, E.V., ed., Structural and Tectonic Correlation across the
390 Johnson et al.
Central Asia Orogenic Collage: North-eastern Segment: Irkutsk, IGCP-
480, Guidebook and Abstract Volume of the Siberian Workshop IGCP-
480, p. 274–278.
Wang, H., and Mo, X., 1995, An outline of the tectonic evolution of China:
Episodes, v. 18, p. 6–16.
Wang, T., Zheng, Y., Gehrels, G., and Zhigou, M., 2001, Geochronological evi-
dence for existence of a South Mongolian microcontinent: Chinese Sci-
ence Bulletin, v. 46, p. 2005–2008.
Wang, Y., Mooney, W., Xuecheng, Y., and Coleman, R., 2003, The crustal
structure from the Altai Mountains to the Altyn Tagh fault, northwest
China: Journal of Geophysical Research, Solid Earth, v. 108, no. B6, doi:
10.1029/2001JB000552.
Wartes, M.A., Carroll, A.R., and Greene, T.J., 2002, Permian sedimentary
record of the Turnpan-Hami basin and adjacent regions, northwest
China: Constraints on postamalgamation tectonic evolution: Geologi-
cal Society of America Bulletin, v. 114, p. 131–152, doi: 10.1130/0016-
7606(2002)114<0131:PSROTT>2.0.CO;2.
Watson, M.P., Hayward, A.B., Parkinson, D.N., and Zhang, Z.M., 1987, Plate
tectonic history, basin development and petroleum source rock deposition
onshore China: Marine and Petroleum Geology, v. 4, p. 205–225, doi:
10.1016/0264-8172(87)90045-6.
Webb, L.E., and Johnson, C.L., 2006, Tertiary strike-slip faulting in southeast-
ern Mongolia and implications for Asian tectonics: Earth and Planetary
Science Letters, v. 241, p. 323–335, doi: 10.1016/j.epsl.2005.10.033.
Webb, L.E., Graham, S.A., Johnson, C.L., Badarch, G., and Hendrix, M.S.,
1999, Occurrence, age, and implications of the Yagan-Onch Hayrhan met-
amorphic core complex, southern Mongolia: Geology, v. 27, p. 143–146,
doi: 10.1130/0091-7613(1999)027<0143:OAAIOT>2.3.CO;2.
Windley, B.F., Allen, M.B., Zhang, C., Zhao, Z.Y., and Wang, G.R., 1990,
Paleozoic accretion and Cenozoic redeformation of the Chinese Tien
Shan Range, central Asia: Geology, v. 18, p. 128–131, doi: 10.1130/0091-
7613(1990)018<0128:PAACRO>2.3.CO;2.
Wright, V.P., 1992, Paleopedology: Stratigraphic relationships and empirical
methodology, in Martini, I.P., and Chesworth, W., eds., Weathering, Soils,
and Paleosols: Amsterdam, Elsevier, p. 475–499.
Xiao, W., Windley, B.F., Hao, J., and Zhai, M., 2003, Accretion leading to
collision and the Permian Solonker suture, Inner Mongolia, China: Ter-
mination of the central Asian orogenic belt: Tectonics, v. 22, no. 6, doi:
10.1029/2002TC001484.
Xiao, W., Windley, B.F., Badarch, G., Sun, S.H., Li, J., Qin, K., and Wang, Z.,
2004, Palaeozoic accretionary and convergent tectonics of the southern
Altaids: Implications for the growth of Central Asia: Geological Society
[London] Journal, v. 161, p. 339–342.
Yang, J., Xu, Z., Zhang, J., Chu, C., Zhang, R., and Liou, J.G., 2001, Tectonic
significance of early Paleozoic high-pressure rocks in Altun-Qaidam-Qil-
ian Mountains, northwest China, in Hendrix, M.S., and Davis, G.A., eds.,
Paleozoic and Mesozoic Tectonic Evolution of Central and Eastern Asia:
From Continental Assembly to Intracontinental Deformation: Geological
Society of America Memoir 194, p. 151–170.
Yanshin, A.L., 1989, 1:1,500,000 Map of Geologic Formations of the Mongo-
lian People’s Republic: Moscow, Academia Nauka USSR, 1 sheet.
Yin, A., and Nie, S., 1996, A Phanerozoic palinspastic reconstruction of China
and its neighboring regions, in Yin, A., and Harrison, M., eds., The Tec-
tonic Evolution of Asia: Cambridge, UK, Cambridge University Press,
p. 442–485.
Zaitsev, N.S., 1974, Stratigraphy and Tectonics of the Mongolian People’s
Republic: Joint Soviet Mongolian Scientific Research Geological Expedi-
tion, Transactions, v. 1, 148 p.
Zaitsev, N.S., Mossakovsky, A.A., and Shishkin, M.A., 1973, Type section of
Upper Paleozoic and Triassic with the first remains of Labyrinthodonts,
Southern Mongolia: Izvestiia Akademii Nauk USSR, Seriia Khimiches-
kaia, v. 7, p. 133–144.
Zhang, Z.M., Liou, J.G., and Coleman, R.G., 1984, An outline of the plate tec-
tonics of China: Geological Society of America Bulletin, v. 95, p. 295–
312, doi: 10.1130/0016-7606(1984)95<295:AOOTPT>2.0.CO;2.
Zhao, X., Coe, R.S., Gilder, S., and Frost, G.M., 1996, Paleomagnetic con-
straints on the palaeogeography of China: Implications for Gondawana-
land: Australian Journal of Earth Sciences, v. 43, p. 643–672.
Zhou, D., Graham, S.A., Chang, E.Z., Wang, B., and Hacker, B., 2001, Paleo-
zoic tectonic amalgamation of the Chinese Tian Shan: Evidence from a
transect along the Dushanzi-Kuqa Highway, in Hendrix, M.S., and Davis,
G.A., eds., Paleozoic and Mesozoic Tectonic Evolution of Central and
Eastern Asia: From Continental Assembly to Intracontinental Deforma-
tion: Geological Society of America Memoir 194, p. 23–45.
Ziegler, A.M., Rees, P.M., Rowley, D.B., Bekker, A., Li, Q., and Hulver, M.L.,
1996, Mesozoic assembly of Asia: Constraints from fossil floras, tecton-
ics, and paleomagnetism, in Yin, A., and Harrison, M., eds., The Tec-
tonic Evolution of Asia: Cambridge, UK, Cambridge University Press,
p. 371–400.
Zinniker, D.A., and Badarch, G., 1997, Reconnaissance palynology and evi-
dence of climate change across the Permian/Triassic boundary, Ih Uvgon,
southern Mongolia: Geological Society of America Abstracts with Pro-
grams, v. 29, no. 6, p. 97.
Zonenshain, L.P., Durante, M.V., Markova, N.G., Filippova, I.V., and Check-
hovich, M.V., 1974, The main features of the geological structure and
development of contiguous parts of the Mongolian and Gobi Altai Moun-
tains, in Zaitsev, N.S., ed., Stratigraphy and Tectonics of the Mongolian
People’s Republic: Joint Soviet Mongolian Scientific Research Geologi-
cal Expedition, Transactions, v. 1, p. 114–131.
Zonenshain, L.P., Kuzmin, M.I., and Natapov, L.M., 1990, Geology of the
USSR; a Plate-Tectonic Synthesis: American Geophysical Union Geody-
namics Series, v. 21, 242 p.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA
 The Geological Society of America
Special Paper 436
2008

Links among mountain building, surface erosion, and growth of an accretionary prism in
a subduction zone—An example from southwest Japan

Gaku Kimura*
Department of Earth and Planetary Science, University of Tokyo, Japan, and Institute for Frontier Research on Earth Evolution,
Japan Agency for Marine-Earth Science and Technology, Yokosuka, Japan

Yujin Kitamura
Department of Earth and Planetary Science, University of Tokyo, Japan, now at IFM-GEOMAR, Kiel, Germany

Asuka Yamaguchi
Hugues Raimbourg
Department of Earth and Planetary Science, University of Tokyo, Japan

ABSTRACT

The relationships between mountain building, surface erosion, sediment supply to

the trench, and growth of the accretionary prism are examined in southwest Japan and
the Nankai Trough. Mountain building caused by the subduction of the Philippine Sea
plate in the Nankai Trough and collision in central Japan has resulted in a rock uplift
rate of ~4 mm/yr. Surface denudation rates in the mountain regions are on the order of
3–4 mm/yr, resulting from the heavy rainfall of the Asian monsoon. This fact suggests
that mountain building is almost in an equilibrium stage in which surface erosion and
rock uplift balance each other, resulting in a constant altitude of ~2000 m.
Several drainage systems on land and in offshore submarine canyons enable
the transport of eroded sediments directly into the Nankai Trough. Most of the ter-
rigenous sediments supplied to the Nankai Trough are accreted in the subduction
zone of the Philippine Sea plate. The accretion rate of the sediments in the eastern
Nankai Trough is ~1.68 × 107 m3/yr, which is consistent with the denudation rate of
the Akaishi Mountains, contributing to the supply of 1.72 × 107 m3/yr of sediment in
central Japan.
The growth of the accretionary prism is an important controlling factor for the
onset of large earthquakes in the Nankai Trough, because the hanging wall of the
rupture area of the seismogenic zone is composed entirely of the accretionary prism.
Repeated large earthquakes with a recurrence time of ~100–200 yr, which are well
recorded in the Nankai Trough, in turn promote surface erosion through consecutive
landslides and tsunamis.

*gaku@eps.s.u-tokyo.ac.jp

Kimura, G., Kitamura, Y., Yamaguchi, A., and Raimbourg, H., 2008, Links among mountain building, surface erosion, and growth of an accretionary prism in a
subduction zone—An example from southwest Japan, in Draut, A.E, Clift, P.D., and Scholl, D.W., eds., Formation and Applications of the Sedimentary Record
in Arc Collision Zones: Geological Society of America Special Paper 436, p. 391–403, doi: 10.1130/2008.2436(17). For permission to copy, contact editing@
geosociety.org. ©2008 The Geological Society of America. All rights reserved.

391
392 Kimura et al.

Southwest Japan, with its extensive record of both erosional processes and seis-
mic events, shows the intimate long-term relationships between tectonically driven
mountain building, surface erosion under the Asian monsoon climate, growth of the
accretionary prism in the trench, and the generation of large earthquakes.

Keywords: Nakai Trough, Asian monsoon, denudation, accretionary prism, large

earthquake.

INTRODUCTION TECTONIC SETTING OF SOUTHWEST JAPAN AND

The growth of accretionary prisms is controlled mainly by
the sediment supply into the trench, which depends on surface
erosion on land and on the drainage system down to the trench.
The development of the drainage system from land to trench and
active surface erosion are necessary for the development of an
accretionary prism. Surface erosion on land is especially intense
in mountain regions under climatic conditions that include heavy
rainfall, such as tropical to subtropical rainy mountains (e.g., in
southeastern Asia) or high-latitude glacier-covered mountains
(e.g., Alaska, Aleutian Trench, or southernmost regions of South
America). Mountain building in these regions results from plate
convergence associated with continent-continent, continent-arc,
or arc-arc collisions, or intracontinental shortening.
Southwest Japan is a particularly appropriate context for
analysis of the complex interplay among active mountain build-
ing on land, active surface erosion from the Asian monsoon
climate, and the resulting development of a huge accretionary
prism in the Nankai Trough (Taira et al., 1980, 1988). Taira et al.
(1988) and Taira and Niitsuma (1986) emphasized that effective
drainage systems from the central mountain region to the Nankai
Trough in Japan have contributed to the buildup of the Nankai
accretionary prism. The mountain building in central Japan is the
tectonic result of the collision between the Izu-Bonin Arc and the
main Honshu Arc, and the collision between the southwest Japan
and northeast Japan Arcs (Sugimura, 1972; Huzita, 1980).
The Asian monsoon brings heavy rain to Japan every year
during the spring rainy season, called Tsuyu, which means con-
tinuous rain. In addition, typhoons attack the Japanese islands,
usually at the end of summer. This climatic condition enhances
the surface erosion of the mountains, and eroded debris and sedi-
ments drain into the rivers down to the ocean.
Frequent earthquakes also might promote surface erosion.
Earthquakes in the Nankai Trough have repeatedly taken place
during historic time (Ando, 1975). Most of the earthquakes in the
Nankai Trough are larger than Mw = 8. Ruff and Kanamori (1980)
suggested that the occurrence of large earthquakes in the subduc-
tion zone might be related to the development of an accretionary
prism in the trench.
Thus, mountain building, heavy surface erosion, develop-
ment of the accretionary prism, and large earthquakes in the
Nankai Trough might ultimately be linked to one another. In this
paper, we examine the linkage in southwest Japan and discuss the
relationships among these geologic processes.
MOUNTAIN BUILDING IN CENTRAL JAPAN
The tectonic setting of central to southwest Japan has resulted
from the convergence of four plates or blocks, and it is commonly
decomposed into three distinct layers (Fig. 1): the top layer is
a conjunction of the Amurian plate of southwest Japan and the
northeast Japan block, which have collided with each other in
central Japan; the second layer is the Philippine Sea plate; and
the lowest layer is the Pacific plate, which is subducting beneath
the above two layers.
The upper plate of the Japanese islands is composed mainly
of volcanic arcs and ancient accretionary complexes, whose rigid
parts consist only of upper brittle crust without rigid mantle litho-
sphere. As the island arcs are strongly deformed, they are sepa-
rated into several microblocks by large faults or tectonic disconti-
nuities. In such cases the distinction between plate boundaries and
intraplate faults is subtle, and the definition of plates differs from
one author to another, from North America (Nakamura, 1983),
to Okhotsk (Savostin and Karasik, 1981; Kimura and Tamaki,
1986), to the northeast Japan plate or block (Seno, 1985), to the
northeast Japan Arc, to Eurasia (Nakamura, 1983), to the Amu-
rian (Savostin and Karasik, 1981; Kimura and Tamaki, 1986)
plate, to the southwest Japan Arc (Fig. 1). Whether the north-
east and southwest Japan Arcs are plates or microblocks, they
are separated by a large fault in central Japan (Fig. 1; Nakamura,
1983), and ongoing convergence between the two is responsible
for mountain uplift of the Japan Alpine Mountains in central
Japan. The convergence and collision of these two arcs began in
Pliocene time (Tsunakawa and Takeuchi, 1986) and accelerated
at ca. 2 Ma (Kaizuka, 1975; Huzita, 1980; Nakamura and Uyeda,
1980; Tsunakawa and Takeuchi, 1986).
The southern part of central Japan is also the collision zone
between the Izu-Ogasawara Arc on the Philippine Sea plate and
the northeast Japan Arc (Sugimura, 1972; Fig. 1). Since the onset
of collision at ca. 15 Ma the arc crust has continuously flaked and
accreted to the upper plate of central Japan (Takahashi, 2006).
The Izu Peninsula is a colliding part of the Philippine Sea plate,
and the intracrustal breakage of the peninsula has already started
because of the collision. The Pacific side of the southwest Japan
Arc (Fig. 1) is composed mainly of Jurassic, Cretaceous, and
Tertiary accretionary complexes, which are exposed parallel to
the Nankai Trough. The complexes in central Japan are bent in a
northward-convex shape. This bending is the result of the contin-
uous collision and accretion of the Izu-Ogasawara Arc (Matsuda
 130û
130º 135û
135º 140û
140º 145û
145º
45û
45º Okhotsk 45º
Plate

40û
40º Amurian 40º
Plate NE Japan

ch
ren
10 cm/y
Arc

an T
Jap
Collision
35û
35º SW Japan 35º
Arc Saga
m
Trou i
h
gh Pacific Plate
r o ug
iT
n ka
Na
4-6 cm/y
30û
30º 30º
Figure 2,3,5 and 6
Izu-Og
ch

asawar
en

Philippine Sea
Tr
yu

Plate
uk

25û
25º 25º
a Arc
Ry

km
0 500
130º 135º 140º 145º
Figure 1. Map showing tectonic setting of the Japanese islands. Note that central Japan lies at the collision zone between
the Amurian plate of southwest Japan, the northeast Japan block, and the Izu-Ogasawara Arc on the Philippine Sea plate.
394 Kimura et al.

and Uyeda, 1971). The main part of the Philippine Sea plate is
now subducting beneath southwest Japan along the Nankai and
Sagami Troughs at a rate of 4–6 cm/yr (Seno, 1989).
The Pacific plate is subducting beneath the northeast Japan
Arc at ~10 cm/yr along the Japan Trench. A trench-trench-trench
type of triple junction lies to the southeast of central Japan at the
southeastern end of the Sagami Trough.
UPLIFT RATE OF THE MOUNTAINS IN CENTRAL
AND SOUTHWEST JAPAN
The present crustal movement of the Japanese islands has
been precisely recorded with modern geodetic observations
within eight distinct periods from 1883 to 1999 (Kunimi et al.,
2001). In addition to these observations, the global positioning
system (GPS), put in place after the 1995 Hanshin-Awaji earth-
quake, has enabled highly detailed measurements over the past
10 yr (Sagiya, 2004). The century-scale record shows that uplift
in central Japan is concentrated in mountain regions (Figs. 2, 3),
especially in the southern part—the Akaishi Mountains—which
lie to the west of the Itoigawa-Shizuoka Tectonic Line and the
Izu Peninsula.
The mountains of the Kii Peninsula, and the Shikoku and
Kyushu island regions in southwest Japan, also have undergone
some uplift, especially in their southern parts (Figs. 2, 3). The
most uplifted regions are parallel to the Nankai Trough but are
not continuous. They actually alternate with subsiding regions
characterized by flat plains, channels, and bays, resulting in
the following pattern from east to west: the Akaishi Mountains
(uplift), Ise Bay (subsidence), Kii Mountain in the Kii Peninsula

Itoigaw
Figure 4

a- S
iz

h
uo
T.L.

s ka T
Akaishi

.L.
in
Akaishi
Mounta
d Sea
I nlan
Seto
Is
e
Ba
aka
Os y O
ains

h
Bay ount
roug

Ogasawara Islands
Kii M
s
tain
ga T
M oun
oku no
e
Lin ma
Shik
Suru

ic Sh S Ku in
n Ch ion s
cto an o-m Ba
Te M
dia
n Tosa ne is
l ak
Me
nnel

Bay i
hu
us
o Cha

Ky A
gh
Bung

i Trou
nka
Na

Philippine Sea Plate

Figure 2. Map showing topographic and offshore features in central and southwest Japan. Rectangle shows the area of
Figure 4. Circles with letters show the following capes: O—Omaezaki; S—Shiono; M—Muroto; A—Ashizuri.
 Mountain building, surface erosion, and growth of an accretionary prism 395

130
130º 131º 132º 133º 134º 135º 136º 137º 138º 139º 140º
37º
A km
0 100 200 300
36º

35º

34º

B
36º
33º

32º
34º

31º
m 32º
-0.4 0 0.4 m km
-0.05 0.00 0.05 0 100 200
30º
130º 132º 134º 136º 138º 140º

Figure 3. Maps showing vertical movement of the Japanese islands. (A) 1883–1999, modified from Kunimi et al. (2001).
Note that the Akaishi Mountains in central Japan and the Pacific side of the Kii Peninsula, as well as the Shikoku and
Kyushu islands, are characterized by an uplift rate of ~40 cm/100 yr. The Seto Inland Sea region, in contrast, is subsid-
ing. (B) Vertical movement of southwest Japan, represented from GPS networks from 1996 to 2004 (Sagiya, 2004). Dots
represent 358 GPS stations.

(uplift), Kii channel (subsidence), eastern Shikoku Mountains
(uplift), Tosa Bay (stable), and the western Shikoku Mountains
in Kyushu (uplift). This geometry is characteristic of 10-km-scale
crustal folds in the along-arc orientation and is explained as a
superimposed compression that resulted from two different con-
vergence directions owing to the subduction of the Philippine Sea
plate along the Nankai Trough and the collision in central Japan
(Huzita, 1980).
The Seto Inland Sea to the north of the island of Shikoku
(Fig. 2) is a subsiding region. This aspect was explained by
the same tectonic origin of crustal-scale folding, superposition
caused by the subduction of the Philippine Sea plate, and the col-
lision in central Japan similar to that in the outer arc of southwest
Japan (Kaizuka, 1975; Huzita, 1980).
The recent compilation of geodetic observations by GPS
(Sagiya, 2004) shows almost the same vertical crustal move-
ment as was recorded earlier by long-term geodetic observations
(Fig. 3B) except for the southern tip of the outer arc. Mountain
regions in the outer arc are generally uplifted, but their south-
ern parts in Capes Omaezaki in Tokai, Shionomisaki in Kii, and
Muroto in Shikoku, from east to west, are undergoing subsidence
(Fig. 2B). This subsidence has been explained as an interseismic
phenomenon between the large earthquakes in 1944 in Tonankai
and in 1946 in Nankai, and a near-future one predicted for the
twenty-first century (Sagiya, 2004). The uplift rate in the moun-
tain region has reached a maximum of 4 mm/yr (Fig. 2) in the
Akaishi Mountains and other nearby areas.
SURFACE DENUDATION RATE OF THE MOUNTAINS
IN CENTRAL JAPAN
The relationship between sediment production and tectonic
uplift in the Japanese islands, and how the climate contributes
to sediment production, were thoroughly studied by Yoshikawa
396 Kimura et al.
(1974) and Ohmori (1978). Their contribution is a milestone in
this research field. We briefly introduce their results in this sec-
tion and discuss them in light of data acquired from geographic
information system (GIS) analyses in central Japan.
The denudation rate of the mountain region was estimated
from the amount of accumulated sediments in the water reser-
voirs, which were constructed by complete damming of the pri-
mary river system (the mainstream system important to the nation
as defined by the Ministry of Land, Infrastructure and Transport
of the Japanese Government) that drains the mountains. As the
volume of small clayey particles and solution chemicals from
erosion is likely to be smaller by several orders of magnitude
than that of solid accumulation in the reservoirs, Ohmori (1978)
ignored such components for a first-order approximation.
The period studied by Ohmori (1978) is also important.
Many small dams were constructed in upstream regions in the
tributaries of the primary rivers since the 1970s to prevent land-
slides and debris flows in the mountains. This means that sedi-
ment accumulation in the large reservoirs can no longer be eas-
ily related to the denudation rate of the mountains. Thus, the
data set collected in the 1960s is the only available quantitative
measurement of the natural erosion rate in the mountains in the
Japanese islands.
Ohmori (1978) clarified that an empirical relationship
between the mean altitude of mountains and sediment produc-
tion is expressed by
SDR = W × H a ,
where SDR = the mean annual sediment delivery rate (m3/km2/
yr), H = the mean altitude of mountains (m), W = 0.4642−3 (/yr),
and a = 2.1894. Thus SDR is almost a quadratic function of the
mean altitude. The denudation rate (DR) is obtained from
DR = SDR ×1.75 /2.68 ×10−6 ,
where 1.75 and 2.68 are the mean bulk densities of sediments
and bedrocks, respectively. Then, the relationship between the
denudation rate and the mean altitude is represented by
DR = 0.3031×10−9 × H 2.1894 .
As reviewed in the previous section, most of the mountains
in Japan are undergoing tectonic uplift. The uplift started in the
Pliocene and accelerated at ca. 2 Ma as described. Geologically
long-term uplift is a result of tectonic uplift (U) minus surface
erosion.
A geological long-term surface uplift rate dH/dt might be
expressed by
dH /dt = U − DR = U − 0.3031×10−9 × H 2.1894 .
Integrating the function from a mean altitude h1 to another mean
altitude h2,
h2
1
t= ∫ U − 0.3031 × 1 0
h1
−9
× H 2.1894
dH
.
This is a law of diminishing returns, which implies that the
mountains are equilibrated at some altitude at some time t. In
the case of 4 mm/yr of U, ~1 m.y. is enough to reach the equilib-
rium stage at which surface erosion and uplift balance each other,
resulting in a constant altitude of ~2000 m.
Figure 4B shows that most of the Akaishi Mountains are
steeper than 30°, which is the critical slope angle consistent with
internal friction of crustal rocks. Slopes steeper than 30° are not
stable and collapse easily by gravitational sliding.
By multiplying the denudation rate (DR) obtained from the
local stream system (Ohmori, 1978) by the surface of the land
whose altitude is higher than 1000 m, a total solid erosion rate is
obtained. We reanalyzed the surface areas of the Akaishi Moun-
tains higher than 1000 m, which amounted to 4.631 × 106 m2,
obtained from GIS analysis. Applying the DR from the river sys-
tems in the Akaishi Mountains in central Japan, an erosional solid
volume of ~1.72 × 107 m3/yr is estimated (Table 1).
Ohmori (1978) pointed out that the lithology of the moun-
tains and location-controlled climatic differences (such as snowy
mountains in the north or rainy mountains in the south within the
Japanese islands) are not strong factors in denudation rates. This
is important, because climatic conditions during the Last Glacial
Maximum were different from those of the present. Sea level was
much lower, and the land area was much greater, than in the pres-
ent, but low-altitude mountains might have not contributed much
to surface erosion, as suggested by the SDR-H function described
above. The present Japanese islands widely cover the climatic
zones of the Asian monsoon from south to north, but Ohmori
(1978) points out the absence of serious differences in the denu-
dation rate in northern and southern Japan.
DRAINAGE SYSTEMS TO THE TRENCH FLOOR OF
THE NANKAI TROUGH
The sediments from the uplifting mountains have not been
directly transported into the trench in many places but are instead
trapped in the forearc, e.g., on continental shelves, in forearc
basins, and in piggyback slope basins on the accretionary prism.
In the eastern part of the Nankai forearc off the Tokai district, the
main submarine canyons—Suruga Trough and Tenryu Canyon—
directly connect the on-land river systems to the trench floor of
the Nankai Trough (Fig. 5). The Sionomisaki Submarine Canyon
from the Kii Peninsula is the main drainage canyon in the middle
part of the Nankai Trough. Other, rather small canyons connect
the Shikoku and Kyushu islands with the Nankai Trough (Fig. 5).
These canyons are topographic manifestations of faults that
cut the accretionary prism (Tokuyama et al., 1999). Detailed topog-
raphy of the mouth of these canyons shows that submarine fans
(Tokuyama et al., 1999) composed of turbidites (Ike et al., 2005)
pass through the canyon channels directly from the land (Fig. 6).
 137º A 138º 139º 137º B 138º 139º

36º 36º

1.887
2.247

v.
shi Mts

u ri
ry
Kiso riv.

Akai

Ten
3.598 Mt. Fuji
3.267 Fu
ji r
iv.

Abe riv.
35º 35º

Oi riv.
Izu Pen.
Suruga Bay

Kiku riv.
slope
km degree
Pacific Ocean 0 50 0 10 20 30 40 50

137º 138º 139º

Figure 4. (A) Area of mountains higher than 1000 m (gray area). Areas shown in five distinct colors show the drainage systems that supply sediments to the main rivers. Numbers in
solid circles represent denudation rates of each drainage system calculated by Ohmori (1978). (B) Areas with slopes steeper than 20°. Solid line on land shows a contour of 1000 m in
the Akaishi Mountains. Note that areas with slopes steeper than 20°–25° are mostly higher than 1000 m.
398 Kimura et al.

TABLE 1. DENUDATION RATE* AND SURFACE AREA HIGHER

THAN 1000 m ALTITUDE IN THE AKAISHI MOUNTAINS IN
The eastern parts of the Nankai Trough are filled by turbi-
CENTRAL JAPAN dites, which are apparent on seismic profiles (Park et al., 2002a;
River system Denudation rate Surface area Erosion of solids Ike et al., 2005) most of which originate from the canyons men-
–3 6 2 6 3
(10 m/yr) (10 m ) (×10 m /yr) tioned above (see Fig. 7). The trench floor of the Nankai Trough
Fuji 3.598 1,678 6.04 gently deepens from east to west owing to subduction of the
Tenryu 2.247 2,191 4.92
topographic highs in the eastern part (Tokuyama et al., 1999).
Oi 3.267 636 2.08
Abe 3.435 55 0.19
Therefore, the turbidites flow down to the west.
Kiso 1.887 2,125 4.01 The total volume of solid sediments draining directly into the
trench per year, passing through these channels, is ~1.44–1.91 ×
Total erosion of solids 17.24
Note: Denudation rate and each river system are shown in Figure 4.
107 m3/yr, as estimated from the solid volume of denudation in
*(Ohmori, 1978) central Japan, given in the previous section.

244
9711

888
590 3457 253
1257
2903
Japan Sea 5026

Trough
20 183
136 124
69
32

Suruga
Sagami
322 476 370

on
35 Trough
1601 1326 Cany
u

1748 no
Tenry

576
4600 ma
Ku in
s
180
Shi
o ba
130 Can no-mi
yon saki
4992
3867
u gh Pacific Ocean
Tro
1530
652
i
310
n ka
Na

Philippine Sea Plate

Figure 5. Map showing mean sediment delivery rate (SDR) to the reservoirs in central to southwest Japan. Black and gray
solid circles indicate the river systems draining into the Pacific Ocean and Japan Sea, respectively. Numbers in circles
represent the volume (m3/km2/yr), modified from Ohmori (1978).
 Mountain building, surface erosion, and growth of an accretionary prism 399

130º 131º 132º 133º 134º 135º 136º 137º 138º 139º 140º
37º

36º

ts.
Akaishi M
Japan Sea

Trough
Honshu
35º

Suruga
Sagami

on
Trough

Cany
Kii Pen.
34º

u
Shikoku

Tenry
no
ma
Ku in
Shi s
o ba

KR9
Can no-m
33º yon isak
i

806
Kyushu 14
gh

-02
1-2 Pacific Ocean
d u
Tro
k ai
32º an
KR

N
98
10
-01

31º

Philippine Sea Plate

30º

m
km
-10000 0 10000 0 100 200 300

Figure 6. Map showing submarine channel systems draining down to the Nankai Trough. Black lines show survey lines
of the seismic profile shown in Figure 7.

ACCRETION RATE OF THE SOLID VOLUME IN THE taper angle varies between 2° and 11°, depending on hydrological
NANKAI ACCRETIONARY PRISM conditions within the wedge and basal décollement (Saffer and
Bekins, 2002; Wang and Hu, 2006; Kimura et al., 2007) (Fig. 7).
The mode of accretion changes at the trench-slope break The frontal accretion rate Vt per unit length along the trench
from an in-sequence and out-of-sequence thrusting mode to an is estimated from
underplating-dominant mode (Kimura et al., 2007). Terrigenous
∫
zd
sediments supplied from the mountainous region on land domi- Vt = v c (1− φ 0e−b0 z )dz
nantly return back to the land owing to frontal accretion in the 0

Nankai Trough (e.g., Moore et al., 2001; Kimura et al., 2007). In

the case of the Nankai Trough the critical taper condition (Davis (Westbrook, 1994; Shibata et al., 2006), where v c is the conver-
et al., 1983) of the accretionary prism is operated trenchward gent rate, 4 cm/yr; φ0 = 0.65, the initial porosity of the trench
from the trench-slope break (Saffer and Bekins, 2002; Wang filling sediment; b0 = 1.54, the constant of the porosity-depth
and Hu, 2006; Kimura et al., 2007), and the accretionary-wedge relationship for the accretionary prism (Bray and Karig, 1985);
400 Kimura et al.

1605 tsunamigenic rupture?

Ashizuri KR9810-01 5.4 7.9 4.6

0 0
Depth (km)

2 Deformation 2
front
4 ST 4
6 OO 6
8
Accretionary prism 8
10 Step down 10
Philippine Sea Plate
12 12
80 70 60 50 40 30 20 10 0
1944 co-seismic rupture
6.3 16.5 7.8
2 Deformation front
Kii 2
Depth (km)

KR9806-02 4 4
)
lt (OOST
Splay fau
6 6
Accretionary prism
8 8
10 Step down 10
Philippine Sea Plate
× 12 12
60 50 40 30 20 10 0
1946 co-seismic rupture
0 0
3.3 8.5 2.1
Depth (km)

2 2
Deformation front
4 OOST 4
Accretionary prism 6
6
8 8
Step down Philippine Sea Plate
10 10
60 50 40 30 20 10 0
Muroto 141-2D
Distance from deformation front (km)
Figure 7. Simplified profiles of the Nankai accretionary prism (Kimura et al., 2007). The age of the prism at a distance of 30 km from the defor-
mation front is ca. 2 Ma (Moore et al., 2001). Out-of-sequence thrusts (OOSTs) are well developed in each accretionary prism.

and zd = 1200 m is the averaged thickness of trench-filling sedi- DISCUSSION

ments involved in frontal accretion in the eastern half of the Nan-
kai Trough (Moore et al., 2001; Park et al., 2002b).
We obtain Vt = 47.95 m2/yr. This value is the growth rate
of the area of the profile of the accretionary prism normal to the
trench axis. The distance of the accretionary prism along the
trench is ~500 km from the mouth of the Suruga Trough to the
deformation front off Shikoku. The results of ocean drilling off
Shikoku show that the trench-filling sediments are dominated
by turbidites directly derived from the land, as mentioned above
(Taira and Niitsuma, 1986; Taira et al., 1992). However, large
amounts of terrigenous sediments are also supplied to the trench
through Shionomisaki Canyon and other canyons to the west,
diluting the sediment transported from the northeast. Because it
is difficult to quantify the contributions from various sources, we
consider only turbidites accreted in the trench to the east of the
mouth of Shionomisaki submarine canyon, along ~300 km of the
trench (Fig. 5). Using this value, the total volumetric accretion
rate is ~1.68 × 107 m3/yr.
Such a rapid rate of accretion suggests that the accretion-
ary prism outboard of the trench-slope break at ~30 km from the
deformation front is composed dominantly of Quaternary turbi-
dites. This is verified by direct drilling into the accretionary prism
(Moore et al., 2001) and by samples using submersible dives
(J. Ashi, 2006, personal commun.).
We quantitatively compiled data on the present mountain
building, surface erosion, and growth of the accretionary prism
in southwest Japan and the Nankai Trough. Our treatment is
just a first-order approximation and needs further development,
but these relationships are important toward understanding the
relationship and feedback among these processes, which have
been studied separately until now. We discuss in the following
section how present relationships can be extrapolated to long-
term evolution.
Relationship between Long-Term and Short-Term
Linkages
The denudation rate in the mountains reviewed by Ohmori
(1978) is averaged over ~10 yr, and the uplift rate in terms of
geodetic observation is averaged over 10–100 yr. It is question-
able whether these rates are applicable to geologic time scales of
millions of years.
Was the denudation rate accelerated or decelerated during
the most recent ice age? It is well known that high mountains in
Japan were capped by glaciers during the last ice age (Minato
et al., 1965). Such climatic conditions appear to have enhanced
surface erosion and thus have produced many more sediments
 Mountain building, surface erosion, and growth of an accretionary prism
than in the present day, as a wide consensus concludes that aver-
age denudation rates have been higher than normal since the start
of northern hemispheric glaciation ca. 2.7 Ma. During the last ice
age the average annual temperature in the Japanese islands was
~7–8 °C lower than that at present. Ohmori (1978) discussed the
difference in denudation rate between the mountains in northern
Japan from the example of Hokkaido, the northernmost island
of Japan, and those in southwest Japan. Their locations are sepa-
rated in latitude by >10°, and the difference in average annual
temperature is ~10 °C at present. Ohmori’s conclusion is that the
denudation rates are almost the same in northern, central, and
southern Japan, as reviewed. Climatic conditions in central and
southwest Japan are different from those in northern Japan. A
heavy rainy season is common in central and southwest Japan
but not in northern Japan. In contrast, northern Japan has a heavy
snowy winter, whereas southwest Japan does not have such a
severe winter. Such climatic factors may compensate each other,
resulting in similar denudation rates. Such present analysis of
the denudation rate in the Japanese islands suggests that a much
cooler climate during the last ice age might not have resulted in
higher denudation rates than at present. The East Asian summer
monsoon rains were much weaker in glacial times, and because
rainfall is a major control on erosion rates (Reiners et al., 2003;
Dadson et al., 2003) this means that denudation rates during the
last glacial maximum likely were much lower than in the present
interglacial. Thus, the wide consensus is that average denudation
rates higher than normal during an ice age should be reexamined
on a global scale.
Another factor controlling denudation rate is sea level,
which was much lower during the last ice age (Fairbanks, 1989).
Ohmori (1978) pointed out that the altitude of the mountains is
the primary factor controlling the erosion rate and the lowest part
of land does not seriously contribute to sediment production but
just affects transportation and accumulation.
All these observations by Ohmori (1978) suggest that the
denudation rate during the last ice age was not very different from
the present interglacial. The intensification of the Asian monsoon
controlled by global-scale atmospheric currents can play back to
the beginning of the Quaternary Period. In addition, the present
tectonic framework of the Japanese islands also dates back to the
beginning of the Quaternary (Huzita, 1980; Honza and Fujioka,
2004). Thus the linkage among mountain building, denudation
rates, and sediment supply into the trench might be applicable at
least to the Quaternary Period. This hypothesis is to be tested by
studying the sedimentation history in the Nankai Trough and the
northern Shikoku Basin, where drilling in the framework of the
Integrated Ocean Drilling Program will soon operate.
Onset of Large Earthquakes, the Seismogenic Zone, and
Growth of the Accretionary Prism
The growth of the accretionary prism is important for the
onset of the rupture of large earthquakes in subduction zones (Ruff
and Kanamori, 1980). Sediments covering the subducting oceanic
401
plate smooth the surface and make a widely coupled asperity in
the seismogenic zone. The rupture areas of the 1944 Tonankai
and 1944 Nankai earthquakes lie beneath and within the accre-
tionary prism (e.g., Park et al., 2002b), which developed from
accretion of sediment transported into the Nankai Trough since
the late Miocene Epoch. What controls the size of the rupture
area of large earthquakes in subduction zones is a complicated
question. One of the hypotheses involves sediment subduction:
Subducted sediments not only smooth the surface as suggested
by Ruff and Kanamori (1980), but water is released from sedi-
ments, especially those that are clayey, dynamically lubricating
the plate-boundary fault for rupture propagation (e.g., Kimura et
al., 2007). Most of the terrigenous turbidites are accreted at the
toe of the accretionary prism, but old pelagic to hemipelagic sedi-
ments underthrust, and their lithification works for the onset of the
seismogenic zone (Moore and Saffer, 2001; Kimura et al., 2007).
The dynamic weakening resulting from lubrication of the fault
might be the most likely mechanism to propagate the rupture area,
and the size of the fluid-rich area might control the magnitude of
earthquakes. In turn, the shaking of the land and tsunamis caused
by large earthquakes enhance landslides in the mountainous areas
and promote surface erosion.
Accretion versus Subduction Erosion
One of the unsolved but important questions in the research of
subduction zones concerns the balance between accretion and sub-
duction erosion (e.g., von Huene and Scholl, 1991; Clift and Van-
nucchi, 2004). The review in this paper suggests that the amount of
underthrust sediments, even in the frontal part of the prism, might
be one or two orders of magnitude smaller than accreted sedi-
ments. Only the distal parts of turbidites transported far from the
trench and deposited as hemipelagic sediments will come back to
the trench and will underthrust, together with biogenic and chemi-
cally precipitated pelagic sediments overlying oceanic basement
of the subducting plate. Subducting sediments are much smaller
in volume because underplating takes place beneath the frontally
accreted sediments. Thus the amount of sediment subduction at
the toe of the deformation front in the eastern Nankai Trough
might be 1–10 × 105 m3/yr along a unit kilometer along the trench,
because the difference between the sediment supply rate of 1.72 ×
107 m3/yr and the accretion rate of 1.68 × 107 m3/yr is the amount
that should subduct along 300 km of the trench axis. This value is
one or two orders of magnitude smaller than the value estimated
for the erosional margins (von Huene and Scholl, 1991; Clift and
Vannucchi, 2004), although the value is estimated without taking
into account the amount of basal erosion or underplating.
Long-Term Feedback System among Tectonic,
Sedimentary, and Climatic Processes
A generic relationship among tectonic, sedimentary, and cli-
matic processes is shown in Figure 8. An important link between
the tectonic and sedimentary processes is the enhancement of
402 Kimura et al.
Tectonic process Sedimentary process
Collision
Mountain building Surface erosion
Sedimentary transport
Accretionary prism Trench filling
Large earthquake Shaking & Tsunami
>Mw=8.0
Smooth the plate interface?
Fluid induced rupture propagation/lubrication?
Change the climate/ocean current system
surface erosion owing to mountain uplift and the resulting growth
of the accretionary prism owing to sediment transport into the
trench. The growth of the accretionary prism is a key for large-
earthquake generation in subduction zones, as discussed above,
and then strong shaking and tsunamis further enhance surface
erosion in the mountains on land. Thus, tectonic events and sedi-
ment production are interrelated at long-term geological scales.
ACKNOWLEDGMENTS
This study was supported by the Plate Dynamics Program of the
Japan Agency for Marine-Earth Science & Technology, and the
21st Century Center of Excellence Program of the University of
Tokyo. Discussions with G.F. Moore, Y. Kaneda, T. Shibata, S.
Okamoto, and T. Ike were quite useful. Reviews by A. Draut, C.
Wobus, P. Clift, and G.F. Moore improved the early draft. We
appreciate all these persons.
REFERENCES CITED
Ando, M., 1975, Source mechanisms and tectonic significance of historical
earthquakes along Nankai Trough, Japan: Tectonophysics, v. 27, p. 119–
140, doi: 10.1016/0040-1951(75)90102-X.
Bray, C.J., and Karig, D.E., 1985, Porosity of sediments in accretionary prisms
and some implications for dewatering processes: Journal of Geophysical
Research, v. 90, p. 768–778.
Clift, P., and Vannucchi, P., 2004, Controls on tectonic accretion versus erosion
in subduction zones: Implications for the origin and recycling of the con-
tinental crust: Reviews in Geophysics, v. 42, no. RG2001.
Dadson, S.J., Hovius, N., Chen, H.G., Dade, W.B., Hsieh, M.L., Willett, S.D.,
Hu, J.C., Horng, M.J., Chen, M.C., Stark, C.P., Lague, D., and Lin, J.C.,
2003, Links between erosion, runoff variability and seismicity in the Tai-
wan orogen: Nature, v. 426, p. 648–651, doi: 10.1038/nature02150.
Davis, J.D., Suppe, F.A., and Dahlen, Q., 1983, Mechanics of fold-and-thrust
belts and accretionary wedges: Journal of Geophysical Research, v. 88,
p. 1153–1172.
Fairbanks, R.G., 1989, A 17,000-year glacio-eustatic sea level record: Influence
of glacial melting rates on Younger Dryas events and deep ocean circula-
tion: Nature, v. 342, p. 637–642, doi: 10.1038/342637a0.
Honza, E., and Fujioka, K., 2004, Formation of arcs and backarc basins inferred
from the tectonic evolution of Southeast Asia since the Late Cretaceous:
Tectonophysics, v. 384, p. 23–53, doi: 10.1016/j.tecto.2004.02.006.
Climatic process
Glacier
Heavy rainfall
Figure 8. Diagram showing linkage
among tectonic, sedimentary, and cli-
matic systems.
Huzita, K., 1980, Role of the Median Tectonic Line in the Quaternary tecton-
ics of the Japanese Islands: Memoirs of the Geological Society of Japan,
v. 18, p. 129–153.
Ike, T., Park, J., Kaneda, Y., and Moore, G., 2005, Along-strike variations in
the structure of the Nankai Trough accretionary prism: Correlation with
variations in subducting basement topography and overlying sediments:
Eos (Transactions, American Geophysical Union), v. 86, Fall Meeting
Supplement, p. T13B–0461.
Kaizuka, S., 1975, Tectonic model for the morphology of arc trench systems,
especially for the echelon ridges and mid-arc fault: Japanese Journal of
Geology and Geography, v. 45, p. 9–28.
Kimura, G., and Tamaki, K., 1986, Collision, rotation, and back-arc spreading
in the region of the Okhotsk and Japan Sea: Tectonics, v. 50, p. 389–401.
Kimura, G., Kitamura, Y., Hashimoto, Y., Yamaguchi, A., Shibata, T., Ujiie, K.,
and Okamoto, S., 2007, Primary switch of plate boundary fault around
the up-dip limit of seismogenic subduction zone and its relationship
with change in wedge taper: Earth and Planetary Science Letters, v. 255,
p. 471–484.
Kunimi, T., Takano, Y., Suzuki, M., Saitou, T., Narita, T., and Okamura, S.,
2001, Vertical crustal movements in Japan estimated from the leveling
observations data for the past 100 years: Journal of the Geographical Sur-
vey Institute, v. 96, p. 23–37.
Matsuda, T., and Uyeda, S., 1971, On the pacific-type orogeny and its model-
extension of paired belts concept and possible origin of marginal seas:
Tectonophysics, v. 11, p. 5–27, doi: 10.1016/0040-1951(71)90076-X.
Minato, M., Gorai, M., and Hunahashi, M., 1965, The Geological Development
of the Japanese Islands: Tokyo, Tsukiji-Shokan, 442 p.
Moore, G.F., Taira, A., Klaus, A., Becker, L., Boeckel, B., Cragg, B.A., Dean,
A., Fergusson, C.L., Henry, P., Hirano, S., Hisamitsu, T., Hunze, S.,
Kastner, M., Maltman, A.J., Morgan, J.K., Murakami, Y., Saffer, D.M.,
Sanchez-Gomez, M., Screaton, E.J., Smith, D.C., Spivack, A.J., Steurer,
J., Tobin, H.J., Ujiie, K., Underwood, M.B., and Wilson, M., 2001, New
insights into deformation and fluid flow processes in the Nankai Trough
accretionary prism: Results of Ocean Drilling Program Leg 190: Geo-
chemistry, Geophysics, Geosystems, v. 2., no. 2001GC000166.
Moore, J.C., and Saffer, D., 2001, Updip limit of the seismogenic zone
beneath the accretionary prism of southwest Japan: An effect of diage-
netic to low-grade metamorphic processes and increasing effective stress:
Geology, v. 29, p. 183–186, doi: 10.1130/0091-7613(2001)029<0183:
ULOTSZ>2.0.CO;2.
Nakamura, K., 1983, Possible nascent trench along the eastern Japan Sea as
the convergent boundary between Eurasian and North American plates:
University of Tokyo, Bulletin of the Earthquake Research Institution, v.
58, p. 711–722 (in Japanese).
Nakamura, K., and Uyeda, S., 1980, Stress gradient in arc–back arc regions and
plate subduction: Journal of Geophysical Research, v. 85, p. 6419–6428.
Ohmori, H., 1978, Relief structure of the Japanese mountains and their stages
in geomorphic development: University of Tokyo, Bulletin of the Depart-
ment of Geography, v. 10, p. 31–85.
 Mountain building, surface erosion, and growth of an accretionary prism
Park, J.O., Tsuru, T., Kodaira, S., Cummins, P.R., and Kaneda, Y., 2002a,
Splay fault branching along the Nankai subduction zone: Science, v. 297,
p. 1157–1160, doi: 10.1126/science.1074111.
Park, J.O., Tsuru, T., Takahashi, N., Hori, T., Kodaira, S., Nakanishi, A., Miura,
S., and Kaneda, Y., 2002b, A deep strong reflector in the Nankai accretion-
ary wedge from multichannel seismic data: Implications for underplating
and interseismic shear stress release: Journal of Geophysical Research–
Solid Earth, v. 107, no. B4, doi:10.1029/2001JB00026.
Reiners, P.W., Ehlers, T.A., Mitchell, S.G., and Montgomery, D.R., 2003, Cou-
pled spatial variations in precipitation and long-term erosion rates across
the Washington Cascades: Nature, v. 426, p. 645–647, doi: 10.1038/
nature02111.
Ruff, L., and Kanamori, H., 1980, Seismicity and the subduction process:
Physics of the Earth and Planetary Interiors, v. 23, p. 240–252, doi:
10.1016/0031-9201(80)90117-X.
Saffer, D.M., and Bekins, B.A., 2002, Hydrologic controls on the morphology
and mechanics of accretionary wedges: Geology, v. 30, p. 271–274, doi:
10.1130/0091-7613(2002)030<0271:HCOTMA>2.0.CO;2.
Sagiya, T., 2004, A decade of GEONET: 1994–2003—The continuous GPS
observation in Japan and its impact on earthquake studies (Review): Earth
Planets Space, v. 56, p. xxix–xli.
Savostin, L.A., and Karasik, A.M., 1981, Recent plate-tectonics of the Arctic
Basin and of Northeastern Asia: Tectonophysics, v. 74, p. 111–145, doi:
10.1016/0040-1951(81)90131-1.
Seno, T., 1985, Is northern Honshu a microplate?: Tectonophysics, v. 115,
p. 177–196, doi: 10.1016/0040-1951(85)90138-6.
Seno, T., 1989, Philippine Sea plate kinematics: Modern Geology, v. 14,
p. 87–97.
Shibata, T., Orihashi, Y., Kimura, G., and Hashimoto, Y., 2006, Intermittent
underplating and sediment subduction—Implication of U-Pb dating of
zircons from the Shimanto accretionary complex, SW Japan: EOS Trans.
AGU, v. 87, Fall Meet. Suppl., Abstract T33C-0524.
Sugimura, A., 1972, Plate boundaries in the vicinity of Japan: Kagaku, v. 42,
p. 192–202 (in Japanese).
Taira, A., and Niitsuma, N., 1986, Turbidite sedimentation in the Nankai Trough
as interpreted from magnetic fabric, grain size, and detrital modal analy-
ses: Washington (U.S. Government Printing Office), Initial Reports of the
Deep Sea Drilling Project, v. 87, p. 611–632.
403
Taira, A., Tashiro, M., Okamura, M., and Katto, J., 1980, The geology of the
Shimanto Belt in Kochi Prefecture, Shikoku, Japan, in Taira, A., and
Tashiro, M., eds., Geology and Paleontology of the Shimanto Belt: Kochi,
Japan, Rinyakosaikai Press, p. 319–389.
Taira, A., Katto, J., Tashiro, M., Okamura, M., and Kodama, K., 1988, The
Shimanto Belt in Shikoku, Japan—Evolution of Cretaceous to Miocene
accretionary prism: Modern Geology, v. 12, p. 5–46.
Taira, A., Hill, I., Firth, J., Berner, U., Bruckmann, W., Byrne, T., Chabern-
aud, T., Fisher, A., Foucher, J.-P., Gamo, T., Gieskes, J., Hyndman, R.,
Karig, D., Kastner, M., Kato, Y., Lallemand, S., Lu, R., Maltaman, A.,
Moore, G., Moran, K., Olaffson, G., Owens, W., Pickering, K., Siena,
F., Taylor, E., Underwood, M., Wilkinson, C., Yamano, M., and Zhang,
J., 1992, Sediment deformation and hydrogeology of the Nankai Trough
accretionary prism: Synthesis of shipboard results of ODP Leg 131: Earth
and Planetary Science Letters, v. 109, p. 431–450, doi: 10.1016/0012-
821X(92)90104-4.
Takahashi, M., 2006, Tectonic boundary between Northeast and Southwest
Japan Arcs during Japan Sea opening: Journal of Geological Society of
Japan, v. 112, p. 14–32.
Tokuyama, H., Ashi, J., Soh, W., Kuramoto, S., and Ikeda, Y., 1999, Active Sub-
marine Faults off Tokai: Tokyo, University of Tokyo Press, 151 p.
Tsunakawa, H., and Takeuchi, A., 1986, Tertiary paleo-stress field and volcanic
activity of the Japanese islands: Kagaku, v. 50, p. 624–631.
von Huene, R., and Scholl, D.W., 1991, Observations at convergent margins
concerning sediment subduction, subduction erosion, and the growth of
continental-crust: Reviews of Geophysics, v. 29, p. 279–316.
Wang, K., and Hu, Y., 2006, Accretionary prisms in subduction earthquake
cycles: The theory of dynamic Coulomb wedge: Journal of Geophysical
Research, vol. 111, no. B6, doi:10.1029/2005JB004094.
Westbrook, G.K., 1994, Growth of accretionary wedges off Vancouver Island
and Oregon, in Westbrook, G.K., Carson, B., and Musgrave R.J., et al.,
Proceedings of the Ocean Drilling Program, Initial Reports, Volume 146:
College Station, Texas (Ocean Drilling Program), p. 381–388.
Yoshikawa, T., 1974, Denudation and tectonic movement in contemporary
Japan: Bulletin of the Department of Geography: University of Tokyo,
v. 6, p. 1–14.
MANUSCRIPT ACCEPTED BY THE SOCIETY 24 APRIL 2007
Printed in the USA