Journal of the Geological Society

Magmatic arc fault systems, deformation partitioning and emplacement of granitic

complexes in the Coastal Cordillera, north Chilean Andes (25°30'S to 27°00'S)
John Grocott and Graeme K. Taylor

Journal of the Geological Society 2002; v. 159; p. 425-443

doi:110.1144/0016-764901-124

Email alerting click here to receive free email alerts when new articles cite this article
service
Permission click here to seek permission to re-use all or part of this article
request
Subscribe click here to subscribe to Journal of the Geological Society or the Lyell Collection

Notes

Downloaded by on 2 June 2007

© 2002 Geological Society of London

Journal of the Geological Society, London, Vol. 159, 2002, pp. 425–442. Printed in Great Britain.

Magmatic arc fault systems, deformation partitioning and emplacement of granitic

complexes in the Coastal Cordillera, north Chilean Andes (258309S to 278009S)
1 2
J O H N G RO C OT T & G R A E M E K . TAY L O R
1
School of Earth Sciences and Geography, Centre for Earth and Environmental Science Research, Kingston University,
Kingston-upon-Thames KT1 2EE, UK (e-mail: j.grocott@kingston.ac.uk)
2
Department of Geological Sciences, University of Plymouth, Plymouth PL4 8AA, UK

Abstract: Three fault systems were responsible for Permian to Late Cretaceous deformation of the overriding
plate of the Andean convergent margin in the Coastal Cordillera of northern Chile (258309 to 278009S).
Displacements were linked to crustal growth expressed by the emplacement of a sequence of magmatic arcs.
The Tigrillo Fault System, active from Triassic to Early Cretaceous time, was characterized by arc-normal
extension with increasing left-oblique extension (transtension) from Early Jurassic to Early Cretaceous time.
Stretching of the crust created space for Triassic, Early Jurassic and Early Cretaceous arc basins where
epiclastic, volcaniclastic and volcanic sequences accumulated in continental to shallow marine environments.
Tabular plutonic complexes were emplaced by roof uplift–floor subsidence that allowed a vertical transfer of
material in the crust without significant horizontal extension. The Atacama Fault System was initiated at
c.132 Ma as a (mainly) left strike-slip fault during left-oblique extension of the margin. Elongate, tabular
plutonic complexes were emplaced within the Atacama Fault System between c.132 and c.106 Ma, again by
roof uplift–floor subsidence mechanisms. Ductile–brittle transitions in synplutonic mylonitic rocks of the
Atacama Fault System provided the setting for Kiruna-type Fe-apatite, and Fe oxide (with Cu and/or Au) ores.
The Chivato Fault System was active as an extensional fault system at the eastern side of the Coastal
Cordillera during displacement on the Tigrillo Fault System and later, between c.125 and c.93 Ma, as a
partitioned left-oblique extensional fault system. In post-Early Cretaceous time the Chivato Fault System was
inverted by left-oblique contraction (transpression) when NW-trending transfer faults, some probably
reactivated lateral ramps in the Tigrillo Fault System, accommodated clockwise vertical-axis rotations of 35–
458. Contraction inverted the Atacama Fault System and Tigrillo Fault System and was responsible for west-
vergent, thin-skinned, fold–thrust deformation in stratified rocks throughout the margin.

Keywords: Andes, Coastal Cordillera, island arcs, magmatism, faulting.

The Coastal Cordillera in the south–central Andes of northern
Chile provides an exceptional record of overriding plate defor-
mation at a convergent margin that has evolved from a retreating
subduction boundary, in Permian to Early Cretaceous time, to an
advancing subduction boundary in Late Cretaceous time. Grani-
tic magmatism at convergent margins is intrinsic to the growth of
continents and is an integral part of Andean geology. However,
the relationship of magma ascent and emplacement to deforma-
tion and, indeed, the very shape of the intrusions making up
Andean arc batholiths has, until recently, been obscure. Here, we
explore this fundamental relationship and show that this segment
of the Andes is a prototype for the role of granitic magmatism in
the deformation of the upper plate of continental magmatic arcs.
We focus on: (1) strains associated with emplacement of plutonic
complexes; (2) the kinematics of magmatic arc fault systems; (3)
structures responsible for vertical-axis rotations and partitioning
of deformation into arc-normal contraction and arc-parallel
strike-slip during oblique convergence.
In broad terms it is well known that overriding plate deforma-
tion in subduction systems depends on relative rates of subduc-
tion and convergence (Dewey 1980; Royden 1993; Waschbusch
& Beaumont 1996). When convergence rate and subduction rate
differ, the trench migrates with respect to a fixed point in the
overriding plate interior (Fig. 1). If the distance between the
trench and this point decreases the boundary is an advancing
subduction boundary and when it increases the boundary is a
retreating subduction boundary. Advancing subduction bound-
aries are characterized by contraction and retreating subduction
boundaries by extension in the arc and back arc (Russo & Silver
1996). These principles, however, provide only a general guide to
structural style in the overriding plate at convergent margins and
the aim here is to document in detail the response of the
overriding plate to deformation at the South American subduc-
tion boundary throughout Mesozoic time.
Geology of the Coastal Cordillera (258309S–278009S)
Devonian to Carboniferous metasedimentary rocks form the
basement in this part of the Coastal Cordillera (Figs 2 and 3)
and were originally deposited in a forearc basin (Bahlburg &
Breitkreuz 1993). Subsequently they were deformed in a late
Carboniferous, SW-vergent thrust belt with gently inclined folds
and intense cleavage–bedding anisotropy (Bell 1984). The
sequence has characteristics of a mélange SW of Chañaral, where
it may be part of a Palaeozoic accretionary prism (Bell 1987).
Permian, Triassic, Jurassic and Early Cretaceous magmatic arc
plutonic complexes were emplaced into this metasedimentary
basement (Berg et al. 1983; Brown 1991; Scheuber & Reutter
1992). They have been the focus of several geochronology
programmes which have shown that intrusive rocks and synplu-
tonic ductile shear zones, in general, young eastward (Damm &
Pichowiak 1981; Berg & Breitkreuz 1983; Naranjo & Puig 1984;
Berg & Baumann 1985; Brook et al. 1986; Dallmeyer et al.
1996; Fig. 4). The stratigraphy of corresponding volcanic arc and

425
426 J. G RO C OT T & G . K . TAY L O R
(a) (b)
Upper plate
shortening (Vc-Vr>0)
continent
advances
slab trenchward slab
roll-back (Vc) roll-back
(Vr) (Vr)
back-arc rocks of Triassic to Early Cretaceous age has been
revised recently and simplified (Godoy et al. 1997; Godoy &
Lara 1998; Lara & Godoy 1998; Marschik & Fontboté 1996;
Arévalo 1999; Fig. 4). Volcanic rocks east of the Atacama Fault
System, formerly regarded as Cerro Florida Formation (Naranjo
1978) and correlated with the Bandurrias Formation of Early
Cretaceous age near Copiapó (Segerstrom & Ruiz 1962), have
been reassigned to the La Negra Formation. The rocks of the
Bandurrias Formation are now regarded as time-equivalent to the
uppermost La Negra Formation and the lower part of the
Chañarcillo Group. They have been reassigned to the Punta del
Cobre Formation (Fig. 4).
Three major arc-parallel fault systems transect this part of the
Coastal Cordillera: the Tigrillo, Atacama and Chivato fault
systems (Fig. 2). Faults in each system are linked by NW- and
some NE-trending transfer faults. All the fault systems show
evidence of longevity and a history of reactivation. Resolving
their polyphase history is critical to understanding deformation
partitioning in the Andean margin and other similar convergent
boundaries.
Tigrillo Fault System
The Tigrillo fault separates Palaeozoic basement from stratified
rocks to the east (Fig. 2) and was recognized in part by Naranjo
& Puig (1984). It is an east-down, normal-slip fault with Triassic
to Middle Jurassic sedimentary, volcaniclastic and pyroclastic
rocks exposed in the hanging wall, and a minimum vertical
separation of 1 km. Displacement on the Tigrillo fault decreases
to the south and a set of left-stepping, normal-slip and dextral
strike-slip faults (Fig. 2) may link it to east-down, synplutonic,
ductile and ductile–brittle, dip-slip shear zones at the margins of
Las Animas and Las Tazas plutonic complexes (Fig. 3). Ages of
c.160 Ma for the Las Animas plutonic complex and c.132 Ma for
Las Tazas (Dallmeyer et al. 1996; Fig. 4) are consistent with
increments of normal slip on the Tigrillo fault. The Tigrillo fault
itself, and the overstep system linking it to displacements at the
western side of the plutonic complexes, make up the Tigrillo
Fault System.
Atacama Fault System
The Atacama Fault System is defined by an arcuate system of
faults with steeply dipping branches marked by ductile shear
zones reworked by left-slip brittle faults (Naranjo 1987; Brown
et al. 1993; Fig. 2). Ductile deformation was associated with
Upper plate
extension (Vc-Vr<0)
Fig. 1. (a) Velocity of the upper plate
toward the trench (Vc) exceeds roll-back
velocity of the lower plate (Vr) and the
no
upper plate is characterized by contractional
trenchward
motion of deformation. (b) Lower-plate roll-back (Vr)
is greater than the velocity of the continent
continental
interior towards the trench and the upper plate is
(Vc=0) characterized by extension. (After Royden
1993; Russo & Silver 1996).
emplacement of Early Cretaceous plutonic complexes between
c.132 and c.106 Ma (Grocott et al. 1994; Dallmeyer et al. 1996).
The relationship between ductile and left-slip brittle deformation
is exemplified on the eastern branch of the Atacama Fault
System, where left-slip brittle faults rework a 1 km wide,
greenschist to lower amphibolite facies, ductile shear zone at the
margin of the Sierra Dieciocho plutonic complex (Fig. 3). The
belt of mylonitic rocks has an 40 Ar/39 Ar hornblende isotope
correlation age of c.125 Ma (Dallmeyer et al. 1996), identical,
within error, to the U–Pb zircon age for the early intrusive
phases of the complex (Berg & Breitkreuz 1983; Fig. 4). High-
temperature ductile shear zones, ductile–brittle shear zones and
brittle faults all with left-slip kinematics imply a progressive,
down-temperature sequence from ductile to brittle deformation
during cooling of the plutonic complex. Late-stage magmatic, Cu
(Au) and magnetite–apatite mineralization was emplaced in the
fault system when the rocks passed through the ductile–brittle
transition. Manto Verde Mine, a regionally important Cu (Au)
ore deposit (Vila et al. 1997), is located between the left-stepping
central and eastern branches of the Atacama Fault System at a
strike-slip relay ramp breached by the Manto Verde fault
(Bonson 1998; Fig. 3).
The total amount of left strike-slip displacement on brittle
faults of the Atacama Fault System has not been estimated
reliably because cut-off points are difficult to identify (St. Amand
& Allen 1960; Arabasz 1971; Brown et al. 1993). Nevertheless,
the discontinuous and overstepping nature of the Atacama Fault
System fault branches between 258309S and 278009S implies that
displacement is unlikely to exceed a few kilometres.
Chivato Fault System
The Chivato Fault System (Godoy et al. 1997) was active during
emplacement of the Remolino plutonic complex as a ductile
shear zone with dip-slip and left strike-slip displacements (Fig.
2). The age of this complex is not well constrained, although a
K–Ar biotite age of 93 ! 3 Ma (Lara & Godoy 1998; Fig. 4)
implies that it is younger than the Sierra Dieciocho complex.
The Chivato Fault System is similar to the Tigrillo Fault System
and the Atacama Fault System in that the high-temperature
ductile shear zones along it are related to the emplacement of arc
plutons. It further resembles the Atacama Fault System in the
emplacement of magnetite–apatite mineralization during the
ductile–brittle transition in these shear zones (Bonson 1998).
Unlike the Atacama Fault System, the Chivato Fault System was
significantly inverted in Late Cretaceous time or later, over-
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
printed by brittle reverse-slip displacements and forms part of a
major belt of left-oblique contraction (transpression) at the
Coastal Cordillera–Precordillera boundary (Godoy et al. 1997;
Arévalo 1999; Fig. 2). This transpressional belt was called the
Central Valley Fault System by Taylor et al. (1998). Some of the
steeply dipping reverse faults in the Chivato Fault System may
be inverted normal faults of Jurassic to Early Cretaceous age
(Godoy et al. 1997) belonging to the Tigrillo Fault System.
Stratified rocks exposed west of the Atacama Fault System
also display west- to NW-vergent folds (Fig. 3) and occasional
thrusts consistent with arc-normal to NW–SE shortening. The
style of the structures and the amount of along-strike shortening
vary markedly across left-slip, NW-trending transfer faults. Fault
inversion has given rise to buckle folds in the hanging wall of
many faults of the Tigrillo Fault System but has not led to
reverse-slip separations on any major fault west of the Atacama
Fault System (Fig. 3). NE-trending folds in units of the
Chañarcillo Group show that contractional deformation is of
post-Albian age, and a maximum age for this deformation of
c.81 to c.77 Ma was established by Sylvester & Palacios (1992)
and Arèvalo (1999). New 40 Ar/39 Ar geochronology from just east
of the Chivato Fault System near Inca de Oro, and from much
farther south along the contractional belt near Vallenar (288309S),
implies that the shortening occurred during emplacement of a
suite of Late Cretaceous granitic intrusions in the Precordillera
between c.69 and c.65 Ma ago (Gipson 2002).
NW- and NE-trending faults
NW- and NE-trending fault zones are prominent in the Coastal
Cordillera but the NW-trending set is by far the most penetrative
(Fig. 2). Most NW faults have left strike-slip displacements that
post-date emplacement of Lower Cretaceous plutonic complexes
and displacements on the Tigrillo Fault System and Atacama
Fault System. These faults were linked kinematically to left-
oblique contraction on the Chivato Fault System (Arévalo 1999).
Included in this set are two, NW- to NNW-trending fault zones
that substantially offset the Atacama Fault System between 268S
and Taltal (Fig. 2). Brown et al. (1993) estimated a maximum
left strike-slip displacement of 70 km for these faults. This
estimate was based on restoration of cut-off points of segments
of the Atacama Fault System with both ductile and brittle
characteristics against the NW- to NNW-trending faults. We have
now recognized segments of the Atacama Fault System between
the cut-off points originally defined, so that displacement cannot
exceed 20 km (Fig. 2). These major NW- to NNW-trending faults
belong to a system that transects the Coastal Cordillera between
258 and 298S. They are linked to displacements on the Central
Valley Fault System at the Coastal Cordillera–Precordillera
boundary to the east (Taylor et al. 1998). The Chivato Fault
System is located within the northern part of the Central Valley
Fault System whereas farther south that fault system is repre-
sented by the Chañarcillo thrust belt (Arèvalo 1999; Fig. 2).
Although left-slip on most NW-trending faults in the Coastal
Cordillera is post-Early Cretaceous in age, and clearly post-dates
displacement on both the Tigrillo Fault System and the Atacama
Fault System, many NW faults host late Early Cretaceous
mineralization, consistent with earlier increments of slip and
local dilation (Colley et al. 1988). Where such faults curve into
the Atacama Fault System, as at Manto Verde (Fig. 3), kinematic
linkage with the Atacama Fault System during Early Cretaceous
time is implied (Bonson 1998). We now present further evidence
that NW-trending faults were active during Permian to Early
427
Cretaceous time, specifically from an analysis of the emplace-
ment mechanism of magmatic arc plutonic complexes.
Permian and Triassic plutonic complexes
Cifuncho
Permian and Triassic plutonic complexes comprise monzonite,
granite, granodiorite or tonalite with minor gabbro and diorite
(Brown 1991). The Cifuncho complex seems to be the oldest of
these, but its age is only poorly constrained by a discordant, U–
Pb zircon lower intercept age of 292 ! 14 Ma (Damm &
Pichowiak 1981). The external contacts are either straight, and
marked by NW-trending faults, or irregular. The latter are
characterized by sheet-like apophyses that interfinger with host
rocks (Fig. 5a). The sheets are up to 100 m thick with tabular
form, steeply dipping edges and gently dipping upper and lower
contacts parallel to the bedding–cleavage fabric anisotropy in the
basement (Fig. 3, locality a1 ). No macroscopic magmatic state or
crystal–plastic fabrics in the granitic rocks, or emplacement-
related fabrics in the host rock, were observed. Andalusite and
muscovite are present in a narrow contact aureole. In contrast,
the contact immediately to the north of this locality is straight
and trends NW with a left-separation (apparent displacement) of
2.5 km (Fig. 3, locality a2 ). It dips steeply NE, is parallel to
bedding in metasedimentary rocks and there is a narrow thermal
aureole. The dip in the basement flattens out to define an open
synform with an axial surface trace parallel to the contact. This
fold deforms the cleavage as well as bedding, implying that it
post-dates the main cleavage-forming regional deformation.
Brittle deformation at the contact is marked by steeply dipping
minor faults in a 20 m wide fracture zone. Two sets of
slickenlines are present on fracture surfaces, one down-dip and
one pitching 108NW. Nevertheless, brittle deformation is not
intense and the contact is essentially intrusive, albeit reworked
by minor dip-slip and strike-slip deformation. A lineament,
visible on Landsat TM imagery, extends away from the intrusion
in the same direction as the contact. There are no separations
across this lineament, either of vertical dykes in the intrusion to
the NW or of the Tigrillo fault and gently dipping strata in its
hanging wall to the SE. This rules out significant increments of
both strike-slip and dip-slip post-emplacement displacements on
the lineament.
Chañaral–Cachina
This elongate complex is exposed south of Cifuncho in the
footwall of the Tigrillo Fault System (Fig. 3). Geochronological
data are sparse but include a U–Pb zircon age of 230 ! 8 Ma
from the southern part of the complex (Berg & Baumann 1985).
The straight, fault-like, NW-trending northern contact is intrusive
but reworked in part by brittle faults (Naranjo & Puig 1984). A
lineament, visible on Landsat TM images, extends to the SE,
where it is a minor fault that displaces the east-dipping Tigrillo
Fault System, a strip of steeply dipping Triassic volcaniclastic
rocks of the Cifuncho Formation and the lower contact of an
Upper Jurassic sill in gently dipping Lower Jurassic strata, each
with left- or right-separations of less than 200 m (Fig. 3, locality
b1 ). Farther south, a narrow strip of basement separates the
eastern contact of the intrusion from a branch of the Tigrillo
Fault System (Fig. 3, locality b2 ). The contact is intrusive, sharp
and dips c.508E, parallel to bedding and to a 50 m thick granitic
sheet in the basement (Fig. 5b). Grain size in the intrusion
remains coarse right up to the contact with metasedimentary host
Fig. 2. Overview map of the Coastal Cordillera in northern Chile (258309S–278309S) showing the major plutonic complexes, volcanic, volcaniclastic and
sedimentary stratified units and the major fault systems: the Tigrillo Fault System (TFS), Atacama Fault System (AFS) and the Chivato Fault System
(CFS). Maps compiled from the authors’ own fieldwork, aerial photograph and Landsat image interpretation, and the maps of the Servicio Nacional de
Geologı́a y Minerı́a (Mercado 1978, 1980; Naranjo 1978; Naranjo & Puig 1984; Godoy & Lara 1998; Lara & Godoy 1998).
 Jln
base of T projected from 5 km N B AFS C
2
Tac UJm
Tac Tac Jln
Pzm 1
Pzm
Jln Jpa UJm Jln
Tac UJm LKg
Tg Jpa Jln 0
Tg Tg Pzm
Pzm Jpa
Jln -1
Tg
Pzm
Pzm
2 km synplutonic NW-trending
extensional fault fault at W oblique left-slip
with possible margin of fault
AFS internal units, AFS
synplutonic pluton (west) (east)
Las Tazas
component AFS
E (central) F
MVF
2
Las Animas Sierra

M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
Las Animas Dieciocho
1 2 3+4 5 1
Flamenco LKg
UJg Jln
Pzm 0
LJg LKg LKg
internal structure unknown UJg
LKg -1
Probable Pzm Probable
Probable Pzm
Probable Pzm
intense fabrics high synplutonic west vergent D1 folds Pzm
crenulated by deformation (schematic - not mapped) high synplutonic deformation
roof-uplift strain overprints D1 overprints D1 in host rocks Las Tazas weak vertical
in host rocks in host rocks intrusive magmatic state
no magmatic state into synplutonic fabric in pluton
fabric in pluton mylonite

Fig. 3. Structural map (see foldout) and cross-sections of the Coastal Cordillera in the Chañaral district of northern Chile (258309S–268459S). The map includes representative orientation data shown on
stereographic projections (Lambert Equal area projections) and palaeomagnetically determined vertical-axis rotations (from Randall et al. 1996). Map and cross-sections were compiled from the authors’
own fieldwork, aerial photograph and Landsat image interpretation, and the maps of the Servicio Nacional de Geologı́a y Minerı́a (Mercado 1978, 1980; Naranjo 1978; Naranjo & Puig 1984; Godoy & Lara
1998; Lara & Godoy 1998).

429
430 J. G RO C OT T & G . K . TAY L O R

50
YPRESIAN

THANET 60 (60.6 ± 1.8)

3
Cabeza deVacas
DANIAN (63 ± 2) 3

post 77Ma
(sinistral)
Hornitos

pression
70

Trans-
Fm.
MAASTRICHTIAN

1
CAMPANIAN "Ocoitas" Intrusion (77 ± 3) ? u/c
80

Cerrillos Fm.
SANTONIAN
CONIACIAN
TURONIAN 90

Sierra de Fraga
Remolino (93.3 ± 3) 2

(sinistral)
CENOMANIAN

tension

Extension in
Trans-
100
ALBIAN 4 onlap
La Borracha (106.3 ± 0.7)
110

Aeropuerto Fm.

(Sinistral)
Bandurrias Fm.
Sierra Dieciocho

Chañarcillo Gp
APTIAN 120 5

tension
(125.7 ± 0.6)

Trans-
4
(126.8 ± 0.5)
Las Tazas (126.8 ± 1.3)
7
BARREMIAN 9
(130.2 ± 0.6)5 (129.2 ± 0.5)
130 (130.3 ± 1.3) 7
HAUTERIVIAN (132.2 ± 0,7) 5
4
Monardes-Moradito
VALANGINIAN (137.6 ± 0.5)
140 (138.9 ± 0.4) 9
(142.1 ± 0.9)

Cobre Fm.
BERRIASIAN

Punta del

Arc-normal
extension
TITHONIAN 150
4
KIMMERIDGIAN Animas (153 ± 0.8) (153.4 ± 0.6)99
OXFORDIAN (155.1 ± 0.7)
7
CALLOVIAN 160 Animas (159.7 ± 1.6)
BATHONIAN

La Negra Fm.
BAJOCIAN 170

AALENIAN Bajocian
Fauna
180 200 m
TOARCIAN above
4 base
Flamenco (east)
190

Arc-normal
(188.7 ± 0.7)

extension?
PLIENSBACHIAN Flamenco (190.1 ± 1.9)
7
Azucar Fm.

hiatus
SINEMURIAN Flamenco (west) 4
Pan de

200 (198.2 ± 0.6)

HETTANGIAN
RHAETIAN
210
Aqua Chica-
Chifuncho

NORIAN 8
Vetado (217 ± 12)
Fm.

220

CARNIAN 8
230 Chañaral (230 ± 8)

LADINIAN
240
6
Chifuncho (292 ± 14)

Stage Time Plutonic Stratigraphical Tectonism

(Ma) Complexes Units

1
K-Ar total rock age (Arévalo 1994)
2,3 2 3
K-Ar biotite ages ( Lara & Godoy 1998; Arévalo 1994)
4,5 40 39 4 5
Ar/ Ar mineral ages ( Dallmeyer et al. 1996 - plutons; Dallmeyer et al. 1996 - synplutonic
mylonites)
6,7,8,9 6 7 8
U-Pb zircon ages ( Damm & Pichowiak 1981; Berg & Breitkreutz, 1983; Berg & Baumann 1985)
9 40 9
Ar/ 39 Ar total rock ages of basaltic andesites in dykes ( Dallmeyer et al. 1996)

Fig. 4. Geochronological and tectonostratigraphic chart for the Coastal Cordillera in northern Chile (258309S–278309S).
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S 431

Fig. 5. (a) Contact of Cifuncho pluton and Palaeozoic basement viewed from the east showing sheet-like structure at the pluton margin (locality a1 , Fig.
3). (b) Contact of Cifuncho pluton and Palaeozoic basement looking south near locality b2 (Fig. 3); the contact dips east concordant with bedding in
Palaeozoic metasedimentary rocks. (c) Coarse granite at southern contact of Vetado pluton at locality c, Fig. 3. (d) Roof of Barquito intrusion viewed from
locality d, Fig. 3. (e) Mylonitic migmatite in ductile shear zone at contact of Flamenco pluton with Palaeozoic basement (locality e, Fig. 3). (f) C9 fabrics
in mylonitic shear zone at margin of Flamenco pluton (locality f, Fig. 3).

rocks. These have a narrow metamorphic aureole in which
andalusite is developed. The dip in the host rock decreases away
from the contact to define an open, steeply inclined syncline in
bedding and cleavage with an axial surface that trends parallel to
the pluton margin.
Nearer Chañaral, the same contact is sliced into NE-trending
segments by NW-trending faults marked by zones of crush
breccia and silicification. The trend of contact between fault
pairs is consistently clockwise of the general trend of the contact
by 20–408, so that vertical-axis rotations, consistent with
432 J. G RO C OT T & G . K . TAY L O R

palaeomagnetic results, could have accompanied left-slip faulting

(Randall et al. 1996). The faults overprint the left-overstep
system in the Tigrillo Fault System (Fig. 2) and, in contrast to
the NW-trending fault at the northern contact of the complex,
show consistent left-separations. Slickenlines confirm these are
oblique left-slip faults (Fig. 3) some of which, traced inboard,
displace plutons as young as at c.93 Ma (Fig. 2).

Vetado
The Vetado plutonic complex lies inboard of the Chañaral–
Cachina complex. It has a U–Pb zircon age of 217 ! 12 Ma
(Berg & Baumann 1985), similar to the stratigraphic age of
Cifuncho and Aqua Chica Formations preserved in structural
basins along the Tigrillo Fault System (Fig. 4). To the west the
contact is subvertical and trends NNE. The host rocks are
intensely ductilely deformed pelitic schists with thin quartzites.
The granite remains coarse grained up to a sharp contact with
metasedimentary rocks (Fig. 5c) that are annealed and have spots
of andalusite and grains of contact-metamorphic muscovite.
Granite veins cut pelitic schists at a high angle but the veins are
not folded, showing that the main schistosity predates emplace-
ment. Close to the contact, andalusite spots are flattened in axial
surfaces of east-vergent minor folds that strike parallel to the
contact. This fabric is interpreted to be due to minor emplace-
ment-related strain. Folding of bedding and the main schistosity
defines an open, NNE-trending synform parallel to the contact
with the intrusion.

Form and emplacement mechanism of Permian and
Triassic plutonic complexes
Two of the features of the Permian and Triassic plutonic
complexes we have described are diagnostic of uplift of the rocks
forming the pluton roofs to create space for pluton emplacement.
First, many north–south-trending contacts of these complexes
terminate against zones of fracturing expressed on Landsat TM
images as NW-trending lineaments. These fault-like structures
show no mappable fault separations away from the pluton (Fig.
6a), even where they transect geological boundaries with very
different orientations (e.g. Fig. 3, NW and SE of locality a2 ).
This characteristic effectively rules out the possibility of post-
emplacement strike-slip or dip-slip (Fig. 6b and c) or of
synplutonic strike-slip (Fig. 6d), leaving the interpretation that
the fracture zone has accommodated synplutonic vertical displa-
cements, restricted to the contact, during emplacement. A second
group of structures diagnostic of roof uplift are the open,
synformal folds in bedding and cleavage that lie parallel to all
the contacts examined so far, with the exception of the southern
boundary of the Vetado complex. These are interpreted as roof-
uplift folds probably, although not necessarily, associated with
roof uplift faults (Fig. 7c and d). Implicit in this emplacement
model is that the Permian and Triassic plutonic complexes are
tabular granites with length and width dimensions that are much
greater than their thickness (McCaffrey & Petford 1997; Cruden
1998).
Lower Jurassic plutonic complexes
Barquito, Flamenco and Pajonales
Permian and Triassic intrusions do not crop out south of
Chañaral and instead plutonic complexes of an Early Jurassic arc
were emplaced into metasedimentary basement rocks now ex-
Fig. 6. Criteria allowing identification of synplutonic faults: (a)
synplutonic roof uplift produces a synform in the host rocks and/or a
pluton-side-up dip-slip fault at the pluton margin; (b) a post-
emplacement strike-slip fault offsets the pluton and lithological contacts
away from the pluton by equal amounts; (c) post-emplacement dip-slip
faults displace the pluton and lithological contacts away from the pluton
by amounts depending on the dip of the contacts, and the amount and
direction of fault slip; (d) synplutonic strike-slip faults have transform-
like characteristics with the apparent offset of the pluton opposed to the
real shear sense on the fault.
posed along the coast (Fig. 2). The intrusive rocks are tonalites
to granodiorites with some diorite, and the host rocks comprise
pelitic to semi-pelitic, strongly deformed to mylonitic schists of
the Chañaral mélange and Las Tórtolas Formation (Bell 1987).
Our fieldwork focused on the Flamenco plutonic complex, which
has a U–Pb zircon age of 190.1 ! 1.9 Ma (Berg & Breitkreuz
1983) and 40 Ar/39 Ar hornblende isotope correlation ages of
188.7 ! 0.7 Ma and 198.2 ! 0.6 Ma from the western and
eastern parts of the intrusion, respectively (Dallmeyer et al.
1996).
At Barquito, schists form a gently dipping roof (Fig. 5d) but at
Flamenco, and the north side of Pajonales pluton, the roof is not
exposed and mylonitic foliation dips steeply on all sides (Fig. 3).
The bases of the plutons are not exposed, giving rise to doubt
about their shape. Grocott & Wilson (1997) noted that the
wavelength–thickness relationships of large-scale open folds in
the mylonitic schists were inconsistent with their multilayer
characteristics and that the wavelength of these folds may reflect
the thickness of underlying tabular intrusions. Those workers
also thought that the map pattern of the pluton was due to
interference between east–west-trending and north- to NE-trend-
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
ing folds, although this idea did not fully account for irregular,
steeply dipping margins. Previously, Grocott et al. (1994) had
argued that synplutonic, top-to-the-east displacements in the
mylonitic envelope were evidence for emplacement at dilational
fault jogs during regional extension. New data from the Flamen-
co pluton allow re-evaluation of pluton shape and emplacement
mechanism and we show that a top-east, ductile shear zone
formed early during emplacement and was progressively de-
formed and recrystallized as space for the pluton was created by
roof uplift.
There are four phases of deformation in the host rocks (Fig.
8a and b). The earliest (D1 ) is associated with the SW- to W-
vergent pre-Permian folds and cleavage (S1 ), and SW- to W-
transporting shear zones that characterize basement structure
(Bell 1984, 1987). The second and third phases (D2, D3 ) are
syn-emplacement, and the last phase (D4 ) is late syn-emplace-
ment. In the outer part of the aureole, fresh andalusite post-
dates S1 and represents an early contact hornfelsing of the
host rocks. Within 500 m of the contact this andalusite has
recrystallized during intense reworking of D1 fabrics to define
a new stretching fabric (L2 ) and a new planar fabric (S2 ; Fig.
5e). The vergence direction defined by the intersection of the
433
Fig. 7. Models for granitic pluton
emplacement by roof uplift. Folds with
characteristic style and position, relative to
the plutonic complex, form during
reactivation of steeply dipping, pre-existing
fractures in extension ((a) and (d)) or
contraction ((b) and (c)). The amount of
horizontal shortening or extension is
determined by the dip of the fault and the
displacement. (e) Emplacement of granitic
pluton at a dilational jog in a low-angle
detachment fault; a greater amount of heave
(horizontal extension) is implied in
comparison with the roof-uplift models
S1 and S2 cleavages is to the east, and kinematic indicators in
zones of high D2 strain (mainly asymmetric boudins) indicate
top-east displacements during D2. Aggregates of stretched
andalusite are partly retrogressed to chlorite but the intense D2
fabric is cut by late, undeformed granite sheets, showing
unequivocally that D2 was linked to the main episode of
emplacement. Close to the contact the host rocks are migma-
titic, and coarse high-temperature D2 fabrics have been
annealed. Evidence of the third phase of deformation is
present close to the western margin of the pluton at Obispito
(Fig. 3). Here, high-grade host rocks have an intense, gently
north-plunging linear fabric, comprising an L-tectonite fabric
and a fold hinge-line lineation (L3 ), caused by crenulation of
the S2 planar fabric during D3 (Fig. 8b). At the eastern
margin, in contrast, S2 is not crenulated and there is no
obvious evidence of D3 fabrics. Late syn-emplacement defor-
mation (D4 ) is marked by conjugate sets of small C9 shear
zones (S4 ), with thin granite veins along the shear plane (Fig.
5f). S4 shear zones are widespread throughout the zone of
intense D2 strain surrounding the pluton, and the conjugate
sets accommodated bulk horizontal flattening and vertical
stretching (Fig. 8b).
434 J. G RO C OT T & G . K . TAY L O R
Fig. 8. (a) Structure in the mylonitic envelope of western part of the Flamenco plutonic complex. (b) Model for the emplacement mechanism. Planar (S2 )
and linear (L2 ) fabric orientations are plotted stereographically in Fig. 3, vi.
Emplacement mechanism for the Flamenco plutonic
complex
We envisage the emplacement of the Flamenco pluton as
essentially a two-stage process (Fig. 8b). First, a thin tabular
granite was emplaced during regional extension, softening the
host rocks at the contact and effectively introducing a ductile
detachment horizon in the upper crust. Steeply dipping exten-
sional brittle faults at higher crustal levels detached into this
zone, in which D2 planar and linear fabrics with top-east
kinematics formed. The second stage involved inflation of the
pluton with space created mainly by roof uplift. Variation in the
plunge of L2 (Fig. 3) is consistent with folding of an originally
NW-trending, D2 extension lineation and reworking of D1 –D2
planar fabrics during roof uplift (D3 ). At the western margin,
near Obispito, crenulation of S2 during D3 is consistent with
superimposition of a west-down, normal dip-slip shear strain on
an east-vergent intersection between S1 and S2 . At the eastern
margin, crenulation of S2 did not occur during roof uplift
because east-down normal dip-slip was superimposed on an east-
vergent intersection between S1 and S2 , and thus the planar
fabric was reinforced rather than crenulated during D3 (Fig. 8b).
Variation in the plunge of L2 (Fig. 3, vi) cannot be due to
stretching in roof-uplift shear zones alone because then it would
plunge consistently down-dip on the foliation, and give a small
circle distribution of poles. Conjugate C9 shear zones (D4 ) are
consistent with late flattening across the earlier fabrics and
vertical extension. This may reflect a ballooning component
during emplacement, caused when the rate of magma supply
exceeded the rate at which space was created by roof uplift,
resulting in ductile flattening of the host rocks in the high-
temperature envelope (Fig. 8b).
Stratified Triassic and Jurassic rocks
Cifuncho, Aqua Chica and Pan de Azucar Formations
The Cifuncho and Aqua Chica Formations of Late Triassic age
are exposed discontinuously along strike in the hanging wall of
the Tigrillo fault (Fig. 3). They comprise coarse clastic continen-
tal strata, with intermediate volcaniclastic rocks towards the base
(Mercado 1980). The presence of lithic clasts derived from the
immediate basement, coupled with very poor sorting, implies a
proximal deposition (Suárez & Bell 1992, 1994). The Cifuncho
Formation is overlain by the Pan de Azucar Formation, which
comprises mainly shallow marine limestone with thin tuff bands
(Davidson et al. 1976). There is no angular unconformity, but the
base of the Pan de Azucar Formation is diachronous (Naranjo &
Puig 1984). The formations were deposited in pull-apart basins
in a NW-trending fault system that extended into central
Argentina (Dalziel et al. 1987; Suárez & Bell 1992; Simpson et
al. 2001) and basaltic, andesitic and silicic volcanic rocks within
the sequences have been linked to subduction.
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
La Negra Formation
The La Negra Formation crops out in the Coastal Cordillera
between Arica and Chañaral. It is several kilometres thick and
consists of andesite and basaltic andesite lavas, volcaniclastic
breccias and high-level sills, with intercalations of marine and
continental sediments (Mpodozis & Ramos 1990). South of
Taltal, it overlies the Pan de Azucar Formation with erosional
unconformity and crops out in the hanging wall of the Tigrillo
fault. Marine sedimentary horizons range in age from Pliensba-
chian at the base near Taltal (Naranjo & Puig 1984) to Bajocian,
200 m above the base, west of Sierra Minillas (Davidson et al.
1976). Flamenco and Pajonales complexes have 40 Ar/39 Ar iso-
tope correlation ages that straddle the age of the unconformity
and imply that they are contemporary with the onset of eruption
of the La Negra volcanic arc (Fig. 4). A K–Ar age of 159 !
3 Ma for one of the Upper Jurassic sills cutting the lower part of
the sequence gives a minimum age for the formation (Naranjo &
Puig 1984).
Permian, Triassic and Jurassic deformation: summary
Structures in host rocks of Permian and Triassic plutonic
complexes point to emplacement by roof uplift permitted by
folding with brittle displacements at steeply dipping pluton
margins. This implies that mainly vertical displacements accom-
panied emplacement, and gives few clues about the arc-normal
component of deformation. More data on the dips of pluton
margins and/or the detailed shape of the folds are required before
the criteria implicit in Fig. 7 can be applied to determine whether
horizontal shortening or contraction accompanied pluton empla-
cement. Nevertheless, Late Triassic sequences were deposited in
NW-trending rifts and are consistent with overriding plate
extension and a retreating subduction boundary at this time. We
therefore predict that margin-parallel faults and/or folds that
permitted roof uplift during emplacement of the Permian and
Triassic plutonic complexes will prove to be extensional struc-
tures, many of which were subsequently reworked by later fault
displacements, particularly on the Tigrillo Fault System. Marine
transgression represented by the Pan de Azucar Formation may
well be due to thermal subsidence following Triassic rifting
(Suárez & Bell 1994).
The La Negra Formation represents an Early Jurassic volcanic
arc (Scheuber & Reutter 1992; Lucassen & Franz 1994).
Although the geochemistry is, in some respects, consistent with a
back-arc basin setting (Rogers & Hawkesworth 1989), it is the
sedimentary sequences of the Lautaro Formation farther east that
represent the corresponding back-arc (Mpodozis & Ramos
1990). Marine beds in the sequence show that subsidence kept
pace with deposition, consistent with overriding plate extension
(Scheuber & Gonzalez 1999). Space for the Lower Jurassic
Flamenco plutonic complex was created mainly by roof uplift,
which folded an early synplutonic shear zone. As the complex is
the same age as the lower part of the La Negra Formation, it
seems possible that the synplutonic shear zone at Flamenco was
part of a regional pattern of ductile to brittle extensional faults
that included contemporary basin-margin growth faults, probably
precursors to the Tigrillo Fault System, although none of these
have yet been identified. We conclude that in Triassic to Late
Jurassic time the overriding plate was characterized by exten-
sional deformation at a retreating subduction boundary and that
arc plutons were emplaced by a roof uplift mechanism during
extension.
435
Upper Jurassic and Lower Cretaceous plutonic
complexes
Las Animas and Las Tazas
The Las Animas plutonic complex is dioritic and has an age of
c.160 Ma (Dallmeyer et al. 1996). It is mostly coarse grained but
becomes finer to the north, where it is often difficult to
distinguish from volcanic rocks of La Negra Formation. Sheets
of diorite emplaced into feldspar-phyric, amygdaloidal andesite
show that the complex was intrusive into the formation, the lower
part of which also contains Upper Jurassic tonalite to granodior-
ite sills of identical radiometric age to the Las Animas complex
and that probably belong to the same suite (Naranjo & Puig
1984).
To the south, the Las Animas complex consists mainly of
diopside–hornblende–biotite diorite emplaced into greenschist
facies metasedimentary rocks of the arc basement. The diorite is
unfoliated, but in the pelitic host rocks there is a 75 m wide,
high-temperature contact aureole with coarse-grained migmatites
and a moderate planar fabric intensity. Foliation is vertical,
trends NNE and is parallel to the contact (Fig. 3). It contains a
steeply plunging stretching fabric and down-to-the-east, S–C
fabrics. A second phase of syn-emplacement deformation is
marked by conjugate sets of C9-type minor shear zones with
granite veins. Bulk strain caused by these zones is horizontal
flattening with vertical stretch. The ductile shear zone does not
continue far to the north, where the host rock lithology is more
siliceous and there is less synplutonic migmatite. Here, the
contact is discordant to west-dipping, pre-emplacement quartz
mylonite and steeply dipping C9-type minor shear zones are the
only expression of synplutonic ductile deformation.
The eastern margin of the Las Animas complex was probably
displaced below the erosion level during emplacement of Las
Tazas complex; an elongate monzonite and granodiorite plutonic
complex emplaced along the western branch of the Atacama
Fault System (Fig. 3). The Las Tazas complex has an age of
c.132 Ma determined by 40 Ar/39 Ar on hornblende (Dallmeyer et
al. 1996) and U–Pb on zircon (Berg & Breitkreuz 1983). From
the identical ages given by each method, we infer that cooling
through the blocking temperatures was rapid, consistent with
emplacement at high crustal levels (Dallmeyer et al. 1996). The
complex is bounded by steeply dipping, high-temperature mylo-
nitic shear zones (Fig. 9a) with the same 40 Ar/39 Ar cooling ages
as the intrusion (Wilson et al. 2000). The shear zone at the
western margin, in dioritic rocks of the Las Animas complex, is
characterized by early syn-emplacement, east-down, dip-slip
displacements with late syn-emplacement horizontal flattening
and vertical stretching on conjugate C9 surfaces (Wilson &
Grocott 1999). High-temperature fabrics in the shear zone,
anisotropy of magnetic susceptibility (AMS) and solid-state
fabrics in the western part of the complex are all characterized
by steep stretching lineations and east-down, non-coaxial flow.
This unequivocally demonstrates synkinematic emplacement of
the Las Tazas plutonic complex.
The northern part of the complex comprises five lithological
units defining an east-younging, sheeted structure (Wilson 1998).
From west to east, magmatic state lineation and the long axes of
the flattening-type AMS ellipsoids rotate from steep to shallow
plunge. Steeply dipping synplutonic mylonitic rocks at the east-
ern margin are characterized by shallow lineation and left-slip
S–C fabrics that overprint, very occasionally preserved, right-
slip kinematic indicators (Wilson 1996). This evidence that
strike-slip at the eastern margin and dip-slip at the western
margin of the complex were essentially contemporaneous implies
436 J. G RO C OT T & G . K . TAY L O R

Fig. 9. (a) Structure of Las Tazas plutonic complex. (b) Model for the emplacement of Las Tazas pluton. Planar and linear fabrics from the marginal
shear zones are plotted stereographically in Fig. 3. Internal structure of the plutonic complex after Wilson (1998).

that by c.132 Ma a change in arc kinematics from arc-normal complex appears fault-like but is intrusive (Fig. 9a). A fracture
extension to left-oblique extension (transtension) had already zone extends away from the contact to the NW and SE, but there
taken place. are no measurable displacements across it so that it has the
The straight, NW-trending southern contact of the Las Tazas characteristics of a synplutonic fault (Fig. 6a). The western
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
margin of the northern pluton is cut and displaced by several
NW-trending faults. These faults have consistent left-separations
along their length, extend beyond the pluton margin and have
horizontal to gently NW-plunging slickenlines. They clearly post-
date emplacement and fault-bounded sectors of the contact have
rotated clockwise relative to sectors without NW-trending faults
(Fig. 9a). This implies that vertical-axis block rotations were
associated with left-slip on these faults.
Shape and emplacement mechanism
We have argued that the Permian and Triassic plutons are gently
dipping sheets and that they have steep sides because synplutonic
faults have been exploited to allow roof uplift during emplace-
ment. The steep contacts and elongate outcrops of the Upper
Jurassic and Lower Cretaceous complexes do not, therefore, rule
out gently dipping, tabular form per se. However, in contrast to
all the older plutonic complexes, deformation in the host rocks
of the Las Animas and the Las Tazas complexes is characterized
by pluton-side-down displacements.
Two types of synplutonic fault are consistent with pluton-side-
down displacements: (1) a steeply dipping, floor depression fault
(Fig. 7a and b); (2) a releasing fault-bend in a low-angle
extensional fault system (Fig. 7e). Grocott et al. (1994) favoured
the second possibility but did not consider the alternative. Their
model required 7–10 km of horizontal extension (heave) (for Las
Animas) and c.5 km (for Las Tazas) to make space for the
plutons, together with linkage between steep faults bounding the
jog and putative low-angle extensional faults to balance the
section (Fig. 7e). Of course, Late Jurassic to Early Cretaceous
normal faults are present in the Coastal Cordillera, exemplified
by the Tigrillo fault, but these are mainly high-angle faults.
There is no evidence for the large stratigraphic omissions on
low-angle faults required to accommodate horizontal extension
of these magnitudes in the Coastal Cordillera at the present
erosion level, although this does not rule out the existence of a
detachment horizon at depth. Neither is it realistic to argue that
space for the plutons was made by extension (heave) of up to
10 km on the steeply dipping faults observed, because the
amount of vertical displacement (throw) required would necessi-
tate very large and abrupt changes in structural level across the
arc, and hence in metamorphic grade, for which there is no
evidence. We therefore interpret the synplutonic shear zones at
the western margin of the complexes as high-angle ductile
extensional faults that accommodated pluton emplacement by
floor depression and roof uplift, rather than by very large
amounts of horizontal extension (heave) (Fig. 9b). Core com-
plex-type extensional detachments reported from the Sierra de
Fraga in the back-arc domain are of later, post-Aptian, age
(Mpodozis & Allmendinger 1993; Fig. 4) but could still be
linked kinematically to high-angle normal faulting during empla-
cement of the late Early Cretaceous plutonic complexes between
the Atacama Fault System and the Chivato Fault System (Fig. 2).
Significance of magmatic state and AMS fabrics
Steeply dipping magmatic state and AMS fabrics in the Las
Tazas pluton have been used to support the view that the
northern complex, at least, has a dyke-like form (Wilson 1998).
These fabrics certainly do not exclude a roof-uplift–floor depres-
sion mechanism. Rather, steeply dipping magmatic state and
crystal plastic fabrics are to be expected, at least close to steeply
dipping marginal shear zones, and are predicted throughout the
intrusion by the cantilever model for pluton emplacement
437
(Cruden 1998), because the column of crust below the pluton,
and the crystallizing magma itself, will experience a weak
vertical shear during emplacement (Fig. 9b).
Although the northern pluton was emplaced during extension
to left-oblique extension, AMS and magmatic state fabrics are
steeply dipping flattening fabrics (Wilson 1998). An explanation
for flattening fabrics in the pluton, consistent with the tabular
shape now recognized, is that strain caused by left strike-slip was
progressively superimposed on fabrics formed by dip-slip defor-
mation, from west to east across the pluton. The resultant fabric
ellipse would be of a flattening type and late progressive left-slip
deformation could rotate the linear element towards the horizon-
tal. Where they have not been overprinted, originally east-down
kinematic indicators may then give a right-slip sense viewed on
sections cut parallel to the strike-slip movement direction,
consistent with the very localized evidence we have of this on
the eastern margin.
Late Jurassic and Early Cretaceous deformation:
summary
Upper Jurassic and Lower Cretaceous plutonic complexes have
the same age as marine sedimentary and volcaniclastic sequences
deposited in basins inboard of the magmatic arc and interpreted
to be back-arc basins (Fig. 4). The existence of a marine back-
arc basin is prima facie evidence of a component of arc-normal
extension at this time. Evidence from dyke dilation directions
points to an increasing component of left-oblique extension
(transtension) in the magmatic arc during this period (Taylor &
Randall 2000). The plutonic complexes contain basaltic andesite
dykes with dilation directions of 2498 and 2328, for Upper
Jurassic dykes in the Vetado and Las Animas complexes,
respectively, and 2058 for Lower Cretaceous dykes in the Las
Tazas complex. This change in the dilation direction implies
increasing obliquity of extension from Late Jurassic into Early
Cretaceous time and a transition from an extensional to a left-
oblique extensional (transtensional) regime.
Synplutonic deformation associated with emplacement of the
Lower Cretaceous plutonic complexes implies that this transten-
sional regime was partitioned into strike-slip and dip-slip compo-
nents. Evidence of deformation partitioning is particularly clear
for the Las Tazas plutonic complex, where the synplutonic shear
zone at the western margin is a pluton-side-down, dip-slip shear
zone and that on the eastern side is characterized by left strike-
slip deformation. Other Lower Cretaceous plutons, Sierra Diecio-
cho (c.125 Ma), La Borracha (c.106 Ma) and Remolino
(c.93 Ma), all have evidence of partitioning of strike-slip and
dip-slip components of the deformation, and show that a regime
of partitioned transtension characterized the margin throughout
Early Cretaceous time. The significance of the Atacama Fault
System and the Chivato Fault System in arc kinematics is simply
that they were synplutonic fault systems that accumulated
(mainly) the strike-slip components of this partitioned transten-
sion.
Late-stage deformation
The Permian and Triassic plutonic complexes and stratified
rocks, and faults of the Tigrillo Fault System, are folded by
large-wavelength, low-amplitude, east–west-trending folds (Fig.
3). Similarly, the sinuous outcrop of the western contact of the
Las Animas plutonic complex is, in part, caused by this type of
folding. These large, open folds are a prominent structural
element between 258 and 278S (Figs. 2 and 3) and possibly
438 J. G RO C OT T & G . K . TAY L O R

farther north (Ferraris & DiBiase 1978). They plunge gently west
and are either bends or buckles formed in response to a late
deformation that post-dates all magmatic arc development and
post-Early Cretaceous inversion of the Tigrillo Fault System and
Chivato Fault System. Evidence for late fault displacements on
the Atacama Fault System is provided by almost ubiquitous
down-dip slickenlines on fault surfaces that may be related to
post-Early Cretaceous inversion, but may also be related to
Miocene to present-day extensional deformation of the upper
crust in the Coastal Cordillera.

Vertical-axis rotation in the Coastal Cordillera

Clockwise vertical-axis rotations reaching 35–458 have been
recorded by investigations of the remanent magnetism in the
Coastal Cordillera between 258S and 278S (Fig. 3). They are
clearly an important element in overriding plate deformation, but
their origin remains controversial. What is clear is that differ-
ential shortening across the Bolivian orocline cannot account for
more than c.108 of the rotations recorded at 268S (Randall et al.
1996) and that some form of block rotation related to fault
displacements is a likely basis for a solution (Forsythe &
Chisholm 1994). Taylor et al. (1998) have argued that the
rotations took place in a post-Early Cretaceous, crustal-scale
strike-slip duplex with linked margin-parallel and NW-trending
faults defining blocks that rotated clockwise during left-oblique
contraction of the Coastal Cordillera (Fig. 10a). This model has
itself proved controversial, particularly concerning the circum-
stances in which clockwise rotation of fault blocks is possible in
a left-slip regime (Beck 1998). Here, we present new evidence
for the link between vertical-axis rotation and strike-slip faulting,
and show that such rotations are an integral part of a change of
upper-plate deformation regime at the South American subduc-
tion boundary from left-oblique extension, in Permian to Early
Cretaceous time, to left-oblique contraction in Late Cretaceous
time.
Graphical models are often used to interpret palaeomagneti-
cally determined vertical-axis rotations (Garfunkel 1989; Nur et
al. 1989; Fig. 11). Characteristic of such models is that blocks Fig. 10. Major NW-trending faults in the Coastal Cordillera (258309S–
are assumed not to deform internally during rotation, and that 298009S) (after Taylor et al. 1998). Orientation of NW-trending fault sets,
faults and blocks rotate by equal amounts. Noting that strike-slip grouped according to their age, Chañaral district (258309S–268459S; area
fault arrays are often bounded by major strike-slip faults, of Fig. 3).
Garfunkel (1989) analysed block rotation in terms of area
(volume) change and length change in the boundary (Fig. 11).
He showed that, provided the blocks are free to change length in faults and to prevent voids developing, extra rotation of the faults
the dimension parallel to the boundary zone, and assuming plane and/or shortening of the rocks in the intervening fault blocks is
strain (no area change), rotation may be either in the same, or in geometrically required at the block corners. Lengthening of fault
the opposite sense, to shearing on the boundary zone, depending blocks parallel to the boundary faults during rotation may be
on the shear sense of the fault array for a given sense of shear on balanced by folding (thickening) perpendicular to the zone. In
the boundary zone. Sense of shear on the fault array will depend the Coastal Cordillera between 258S and 278S, the east–west-
on whether shortening or extension occurs across the boundary trending, late, open folds may be an expression of thickening of
zone. In effect, the fault arrays act like a single slip system in a the fault blocks during rotation.
crystal structure and rotation takes place with a sense dependent
on the longitudinal strain across the boundary zone and not the
The link between fault displacements and vertical-axis
shear sense parallel to the boundary zone (Passchier & Trouw
rotation
1996, p. 89).
Well-known space problems implicit in this approach are In the Coastal Cordillera between 258S and 278S the remanent
particularly obvious at the corners of the blocks in the boundary magnetizations affected by vertical-axis rotations are believed to
zone and are also evident in the lengthening of the deformed be both primary and secondary, and therefore similar amounts of
domain implied by the model (McKenzie & Jackson 1983; Fig. rotation have affected rocks and magnetizations of different ages
11). In nature, the undeformable block assumption must break (Randall et al. 1996; Fig. 3). Rotations are recorded by: (1)
down, at least at the corners of rotating blocks, and is often Lower Jurassic volcanic rocks of the La Negra Formation; (2)
expressed by curvature of the array faults into the boundary Upper Jurassic basaltic–andesite dykes emplaced into the Veta-
faults (Fig. 10a). To accommodate displacement on the array do, Flamenco and Las Animas plutonic complexes; (3) Lower
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S 439

shear gradient (K) calculated rotation (R) observed rotation (R)

-0.27 9.13 12
-0.24 8.88 10
-0.21 6.13 9
α -0.35 12.73 11
-0.27 9.77 9.5

δ w
α Las Tazas shear gradient (K) vs rotation (delta)
s

calculated rotation
15
10

(delta)
5
0
-0.4 -0.3 -0.2 -0.1 0
(a) shear gradient (K)

Las Tazas shear gradient (K) vs rotation (delta)

measured rotation
15

(delta)
10
5
w
α
0
α+δ δ
s -0.4 -0.3 -0.2 -0.1 0
(b) shear gradient (K)
K = s/w α = initial angle Fig. 12. Fault slip, block rotation and fault orientation related to NW-
sin δ δ = rotation trending faults cutting the margin of the Las Tazas plutonic complex.
K= . sin(δ-α)
sin α Typical vertical-axis rotations (a) predicted by graphical analysis and (b)
s = displacement measured in situ are each of the order of 10–158, significantly less than
K = cot(α-δ) - cot δ
w = block width palaeomagnetically determined vertical-axis rotation (Fig. 3). Parameters
K and ! are defined in Fig. 11.
cot δ = cot(α-δ) - K K = shear gradient

Fig. 11. Graphical analysis of the relationship between the parameters of
displacement, block rotation and fault slip during sinistral transpression
(modified from Garfunkel 1989).
Cretaceous basaltic–andesite dykes in the Las Tazas and Remo-
lino plutonic complexes, which record a similar amount of
vertical-axis rotation in Lower Cretaceous dykes in plutonic
complexes on both sides of the Atacama Fault System. The
youngest dykes studied palaeomagnetically are Lower Cretaceous
basaltic andesites that cut mylonitic rocks with a 40 Ar/39 Ar
isotope correlation age on hornblende of c.125 Ma east of the
Atacama Fault System, but an upper age bracket for the rotations
has yet to be defined.
At the western margin of the Las Tazas, Sierra Dieciocho and
Remolino plutonic complexes, intensely foliated mylonitic rocks
cut by NW-trending faults systematically trend 10–158 clockwise
of the average trend of the shear zones themselves, and of the
Coastal Cordillera in general (Figs. 2, 3 and 9a). This observation
implies that clockwise vertical-axis rotations of at least this
magnitude have accompanied left-slip on faults in this area and is
supported by graphical analysis showing that displacement mag-
nitudes, block width and amounts of rotation (calculated and
measured) are consistent with each other (Fig. 12). The rema-
nence of late synplutonic dykes in the complex records a vertical-
axis rotation of 35.68 ! 13.28 from a site in the Las Tazas
plutonic complex just east of the faulted segment of the pluton
margin (Randall et al. 1996; Fig. 3). This is larger than would be
expected from the 10–158 rotation of the foliation noted above,
albeit at the low end of the range results from the area as a whole.
The general trend of the main stratigraphic and structural
elements of the Coastal Cordillera is parallel to the coast. At a
smaller scale, however, lithological contacts that trend NE in
blocks bounded by left-slip NW-trending faults are ubiquitous
between 258S and 278S (Figs. 3 and 10a). The problem of
interpretation of these NE trends as a consequence of vertical-
axis rotation is that it is often impossible to know the pre-
faulting trend of the contact, and hence estimate the amount of
rotation during slip on NW faults. Nevertheless, NW fault-
bounded blocks with internal lithological boundaries that strike
clockwise of the overall trend of margin are consistent with the
hypothesis that vertical-axis rotation and left-slip on NW faults
are linked (Forsythe & Chisholm 1994).
To investigate the discrepancy between structural and palaeo-
magnetic data at Las Tazas, and to provide a further test of the
hypothesis that rotations are linked to left-slip on NW-trending
faults, we plotted orientation of NW-trending faults according to
age (Fig. 10). The average trend of all NW faults is 3208, but those
440 J. G RO C OT T & G . K . TAY L O R
for which we can demonstrate unequivocally a history of displace-
ment that goes back at least as far as Late Jurassic time make a
smaller angle to the margin than younger faults. Late Jurassic to
Early Cretaceous transfer faults in the Tigrillo Fault System and
Early Cretaceous synplutonic and synmineralization faults trend
on average at 3428 (Fig. 10b, ii). Most post-Early Cretaceous
faults have a trend close to the average of 3208, and the youngest
faults in the margin, including the still active Taltal fault, trend at
3078. The range of these average trends is 358, close to typical
values of vertical-axis rotation determined palaeomagnetically.
Interpretation
Our interpretation is that early increments of post-Early Cretac-
eous displacement and up to 228 rotation (342–3208) were
accommodated on pre-existing faults of the Tigrillo Fault System,
now orientated NNW and first active during Late Jurassic to
Early Cretaceous extensional deformation as right-slip transfer
faults (Fig. 2). When these faults rotated into an unfavourable
orientation for further slip, the main set of NW faults formed
and, with the fault blocks they defined, subsequently rotated c.138
(320–3078) to their present orientation with progressive arc-
normal shortening. This implies that NW-trending faults and fault
blocks are redefined incrementally and that the amount of
vertical-axis rotation recorded palaeomagnetically may be up to
228 more than the rotation experienced by an adjacent fault
belonging to the 3208 set as, for example, at Las Tazas. NW
faults that make a high angle to the margin do so because they
are relatively young and have not rotated significantly. The
orientation of the youngest faults may simply be controlled by
the failure angle in the overriding plate, although their persis-
tence through time suggests that deep-seated, NW-trending
structures repropagate periodically through the upper crust to re-
establish a set of faults at a relatively high angle to the margin.
Relation of clockwise rotations to left-slip margin-parallel
shear sense
In the models of Garfunkel (1989), block rotations are linked to
displacements on boundary fault systems. In the Coastal Cordil-
lera, the youngest component of strike-slip diplacements on NW
faults post-date displacement on the Atacama Fault System. Some
kinematic linkage between NW faults and the Atacama Fault
System has been demonstrated at Manto Verde (Bonson 1998)
but there is no evidence of major linkage between NW faults and
the Atacama Fault System, consistent with similar magnitudes of
rotation recorded on either side of the fault system. The Atacama
Fault System cannot, therefore, be regarded as a boundary fault
to the NW fault arrays in the sense of Garfunkel (1989).
Traced east from the Atacama Fault System, NW faults meet
the Chivato Fault System–Chañarcillo thrust belt at the Coastal
Cordillera–Pre-Cordillera boundary (Taylor et al. 1998; Arévalo
1999; Fig. 2). This system is characterized by left-oblique
contraction partitioned into arc-normal thrusting and arc-parallel
left strike-slip (Arévalo 1999). Significantly, NW faults acted as
transfer faults in the thrust belt, just as they do in the stratified
sequences of the Coastal Cordillera. The combination of NW-
trending left-slip faults, clockwise vertical-axis rotation and
margin-parallel left-slip faults of the Chivato Fault System is
fully consistent with models of Garfunkel (1989), where sense of
block rotation is opposite to the sense of shear on the boundary
faults. This system has accommodated left-oblique contraction of
the margin during post-Early Cretaceous time, although an upper
age bracket for this deformation has yet to be defined.
Conclusions
Deformation of the overriding plate at the Andean subduction
boundary was characterized by extension on the Tigrillo Fault
System in Triassic to Early Cretaceous time with increasing left-
oblique extension (transtension) from Late Jurassic into Early
Cretaceous time. Permian to Lower Cretaceous plutonic com-
plexes were emplaced by roof uplift–floor depression mechan-
isms that allowed a vertical transfer of material in the crust but
this process, of itself, did not accommodate significant amounts
of horizontal extension. The Atacama Fault System was initiated
during emplacement of the Las Tazas plutonic complex at
c.132 Ma as a left strike-slip fault system. Emplacement of
plutonic complexes along the Atacama Fault System by roof
uplift–floor depression involving dip-slip displacements on ex-
tensional faults is consistent with operation of the Atacama Fault
System as part of a partitioned, left-oblique transtensional system
between 132 and 106 Ma. Farther east, at the Coastal Cordillera–
Precordillera boundary, the Chivato Fault System was active as a
partitioned left-oblique extensional fault system during emplace-
ment of Early Cretaceous plutonic complexes between c.125 and
c.93 Ma. The Chivato Fault System may also have had an earlier
history of extension that would correlate with displacements on
the Tigrillo Fault System. This style of upper-plate deformation
is consistent with the existence of a retreating subduction at the
Andean margin from Triassic time to the end of Early Cretaceous
time.
Subsequent deformation in the Coastal Cordillera was parti-
tioned into arc-parallel left strike-slip and arc-normal shortening
components. Vertical-axis clockwise rotations of 35–458 were
produced by a combination of left-slip on NW-trending faults
and left-slip on north–south-trending margin-parallel faults and
accommodated by domino-style rotation of both NW faults and
fault blocks. Whereas some NW-trending faults were initiated
during this contractional deformation, others were reactivated
lateral ramps in the Tigrillo Fault System. Arc-normal shortening
was accommodated by: (1) domino-style fault and fault block
rotation with concomitant arc-parallel extension; (2) transpres-
sion on north–south-trending fault zones, particularly the Chiva-
to Fault System at the Coastal Cordillera–Precordillera
boundary; (3) thin-skinned folding and thrusting in Triassic to
Lower Cretaceous stratified rocks. Partitioned oblique contraction
of the upper plate is consistent with operation of the Andean
convergent margin as an advancing subduction boundary in post-
Early Cretaceous time.
We would like to thank colleagues at the Geological Survey of Chile
(SERNAGEOMIN) for their contributions, often through vigorous debate,
to the ideas expressed in this paper, especially C. Arévalo, P. Cornejo, E.
Godoy, S. Matthews and A. Tomlinson. We also thank C. Bonson, M.
Brown, M. Gardeweg, M. Gipson, D. Leech, C. Mpodozis, D. Randall, P.
Treloar and J. Wilson for discussion. SERNAGEOMIN provided us with
first-class logistical support without which our fieldwork would have been
much more difficult. In particular, we are indebted to R. Morales for all
his help over many field seasons. S. Bignold (K.U.) compiled the maps
and drafted the other figures, and A. Belsham (K.U. Media Services)
processed the photographic images. Both are thanked for their help. The
research was funded by NERC research grant GR3/99/111 awarded to
G.K.T. and J.G., and a Royal Society Study Visit bursary awarded to J.G.
References
Arabasz, W.J., Jr. 1971. Geological and geophysical studies of the Atacama fault
zone in northern Chile. PhD thesis, California Institute of Technology,
Pasadena.
Arévalo, C. 1994. Carta GeologÍca de Chile, Hoja Los Loros, escala 1:100,000.
 M AG M AT I C A R C FAU LT S Y S T E M S , C H I L E A N A N D E S
Servicio Nacional de Geologı́a y Minerı́a, Santiago de Chile.
Arévalo, C. 1999. The Coastal Cordillera/Precordillera boundary in the Tierra
Amarilla area (278209–278409S; 708059–708209W) northern Chile, and the
structural setting of the Candelaria Cu–Au ore deposit. PhD thesis,
University of Kingston, Kingston-upon-Thames.
Bahlburg, H. & Breitkreuz, C. 1993. Differential response of a Devonian–
Carboniferous platform–deeper basin system to sea-level change and
tectonics, N. Chilean Andes. Basin Research, 5, 21–40.
Beck, M.E. Jr 1998. On the mechanism of crustal block rotations in the central
Andes. Tectonophysics, 299, 75–92.
Bell, C.M. 1984. Deformation produced by subduction of a Palaeozoic turbidite
sequence in northern Chile. Journal of the Geological Society, London, 141,
339–347.
Bell, C.M. 1987. The origin of the Upper Palaeozoic Chañaral melange of
northern Chile. Journal of the Geological Society, London, 144, 599–610.
Berg, K. & Baumann, A. 1985. Plutonic and metasedimentary rocks from the
Coastal Range of northern Chile: Rb–Sr and U–Pb isotopic systematics.
Earth and Planetary Science Letters, 75, 101–115.
Berg, K. & Breitkreuz, C. 1983. Mesozoische Plutone in der nordchilenischen
Küstenkordillere: petrogenese, geochronologie, Geochemie und Geodynamik
mantelbetonter Magmatite. Geotektonische Forschungen, 66, 1–107.
Berg, K., Breitkreuz, C., Damm, K.-W., Pichowiak, S. & Zeil, W. 1983. The
North-Chilean Coast Range—an example for the development of an active
continental margin. Geologische Rundshau, 72, 715–731.
Bonson, C. 1998. Fracturing, fluid processes and mineralisation in the Cretaceous
continental magmatic arc of northern Chile (258159–278159S). PhD thesis,
University of Kingston, Kingston-upon-Thames.
Brook, M., Pankhurst, R.J., Shepherd, T.J., Spiro, B., Snelling, N.J. &
Swainbank, I.G. 1986. Andean Geochronology and Metallogenesis. British
Geological Survey Open-File Report WI/IG/86/1.
Brown, M. 1991. Comparative geochemical interpretation of Permian–Triassic
plutonic complexes of the Coastal Range and Altiplano (258309–268309S),
northern Chile. Geological Society of America, Special Paper, 265, 157–177.
Brown, M., Diaz, F. & Grocott, J. 1993. Displacement history of the Atacama
Fault System 258009–278009S, northern Chile. Geology Society of America
Bulletin, 105, 1165–1174.
Colley, H., Treloar, P.J. & Diaz, F. 1988. Gold and silver mineralization in the
El Salvador region, northern Chile. In: Keays, R., Ramsey, R. & Groves, D.
(eds) The Geology of Gold Deposits: the Perspective in 1988. Economic
Geology Monograph, 6, 208–217.
Cruden, A.R. 1998. On the emplacement of tabular granites. Journal of the
Geological Society, London, 155, 853–862.
Dallmeyer, R.D., Brown, M., Grocott, J., Taylor, G.K. & Treloar, P.J.T.
1996. Mesozoic magmatic and tectonic events within the Andean plate
boundary zone, 268–278309S, north Chile: constraints from 40 Ar/39 Ar mineral
ages. Journal of Geology, 104, 19–40.
Dalziel, I.W.D., Storey, B.C., Garrett, S.W., Grunow, A.M., Herrod, L.D.B.
& Pankhurst, R.J. 1987. Extensional tectonics and fragmentation of
Gondwanaland. In: Dewey, J.F., Coward, M.P. & Hancock, P. (eds)
Continental Extensional Tectonics. Geological Society, London, Special
Publications, 28, 433–441.
Damm, K.W. & Pichowiak, S. 1981. Geodynamik und Magmengenese in der
Küstenkordillere Nordchiles zwischen Taltal und Chañaral. Geotekonische
Forschungen, 61, 1–166.
Davidson, J., Godoy, E. & Covacevich, V. 1976. El Bajociano de Sierra Minillas
(708309O; 268009S) y Sierra Fraga (698509O; 278009S), Provincia de Atacama,
Chile: edad y marco geotectónico de la Formación La Negra en esta latitud. I
Congreso Geológico Chileno, Universidad de Chile, Actas, 1, 255–272.
Dewey, J.F. 1980. Episodicity, sequence and style at convergent plate boundaries.
In: Strangeway, D.W. (ed.) The Continental Crust and its Mineral
Deposits. Geological Association of Canada, Special Publication, 20,
553–576.
Ferraris, F.B. & DiBiase, F.F. 1978. Carta Geologı́ca de Chile, Hojas
Antofagasta, No. 30, escala 1:250,000. Servicio Nacional de Geologı́a y
Minerı́a, Santiago.
Forsythe, R. & Chisholm, L. 1994. Palaeomagnetic and structural constraints on
rotations in the north Chilean coast ranges. Journal of South American Earth
Sciences, 7, 279–294.
Garfunkel, Z. 1989. Regional deformation by block translation and rotation. In:
Kissel, C. & Laj, C. (eds) Paleomagnetic Rotations and Continental
Deformation. NATO ASI Series, 254, 181–208.
Gipson, M. 2002. Palaeomagnetism and geochronology of the Andean forearc in
the Vallenar district (298S), Region III, northern Chile. PhD thesis, University
of Plymouth.
Godoy, E. & Lara, L. 1998. Carta Geologı́ca de Chile, Hojas Chañaral y Diego
del Almagro, No. 5-6, escala 1:100,000. Servicio Nacional de GeologÍa y
MinerÍa, Santiago de Chile.
Godoy, E., Lara, L. & Ugalde, H. 1997. La falla Chivato: borde oriental del
441
plutonismo asociado al sistema de falla Atacama. VII Congreso Geológico
Chileno, Universidad Católica del Norte, Antofagasta, Actas, 3, 1604–1607.
Grocott, J., Brown, M., Dallmeyer, R.D., Taylor, G.K. & Treloar, P.J.
1994. Mechanisms of continental growth in extensional arcs: an example
from the Andean plate-boundary zone. Geology, 22, 391–394.
Grocott, J. & Wilson, J. 1997. Ascent and emplacement of granitic plutonic
complexes in subduction-related extensional environments. In: Holness, M.
(ed.) Deformation-enhanced Fluid Transport in the Earth’s Crust and Mantle.
Mineralogical Society Series, 7, 173–195.
Lara, L. & Godoy, E. 1998. Carta GeologÍca de Chile, Hoja Quebrada Salitrosa,
No. 4, escala 1:100,000. Servicio Nacional de GeologÍa y MinerÍa, Santiago
de Chile.
Lucassen, F. & Franz, G. 1994. Arc-related Jurassic igneous and metamorphic
rocks in the Coastal Cordillera of northern Chile, Region Antofagasta.
Journal of Metamorphic Geology, 14, 249–265.
Marschik, R. & Fontboté, L. 1996. Copper (–iron) mineralization and super-
imposition of alteration events in the Punta del Cobre belt, northern Chile. In:
Camus, F., Sillitoe, R.H. & Peterson, R. (eds) Andean Copper Deposits:
New Discoveries, Mineralization, Styles, and Metallogeny. Society of
Economic Geologists Special Publication, 5, 171–190.
McCaffrey, K.J.W. & Petford, N. 1997. Are granites scale invariant? Journal of
the Geological Society, London, 154, 1–4.
McKenzie, D. & Jackson, J. 1983. The relationship between strain rates, crustal
thickening, palaeomagnetism, finite strain and fault movements within a
deforming zone. Earth and Planetary Science Letters, 65, 182–202.
Mercado, M. 1978. Carta Geológica de Chile, Cordillera de la Costa entre
Chañaral y Caldera, No. 27, escala 1:100,000. Instituto de Investigaciónes
Geológica de Chile, Santiago.
Mercado, M. 1980. Carta Geológica de Chile, Area Pan de Azúcar, No. 37, escala
1:100,000. Instituto de Investigaciónes Geológica de Chile, Santiago.
Mpodozis, C. & Allmendinger, R.W. 1993. Extensional tectonics: Cretaceous
Andes, northern Chile (278S). Geological Society of America Bulletin, 105,
1462–1177.
Mpodozis, C. & Ramos, V.A. 1990. The Andes of Chile and Argentina. In:
Eriksen, G.E., Cañas-Pinochet, M.T. & Reinemund, J.A. (eds) Geology of
the Andes and its relationship to hydrocarbon and mineral resources.
Circum-Pacific Council for Energy and Mineral Resources, Earth Science
Series, 11, 59–90.
Naranjo, J.A. 1978. Carta Geológica de Chile, Hoja Geologı́a de la zona interior
de la Cordillera de la Costa entre los 268009 y 268209S, No. 34, escala
1:100,000. Instituto de Investigaciónes Geológica de Chile, Santiago.
Naranjo, J.A. & Puig, A. 1984. Carta Geologı́ca de Chile, Hojas Taltal y
Chañaral, No. 62-63, escala 1:250,000. Servicio Nacional de GeologÍa y
Minerı́a, Santiago.
Naranjo, J.A. 1987. Interpretación de la actividad Cenozoica superior a lo largo
de la zona de falla Atacama, norte de Chile. Revista Geológica de Chile, 31,
43–55.
Nur, A., Ron, H. & Scotti, O. 1989. Mechanics of distributed fault and block
rotation. In: Kissel, C. & Laj, C. (eds) Paleomagnetic Rotations and
Continental Deformation. NATO ASI Series, 254, 209–228.
Passchier, C.W. & Trouw, R.A.J. 1996. Microtectonics. Springer, Berlin.
Randall, D.E., Taylor, G.K. & Grocott, J. 1996. Major crustal rotations in the
Andean margin: paleomagnetic results from the Coastal Cordillera of north-
ern Chile. Journal of Geophysical Research, 101, 15783–15798.
Rogers, G. & Hawkesworth, C.J. 1989. A geochemical traverse across the north
Chilean Andes: evidence for crust generation from the mantle wedge. Earth
and Planetary Science Letters, 91, 271–285.
Royden, L.H. 1993. The tectonic expression of slab-pull at continental convergent
boundaries. Tectonics, 12, 303–325.
Russo, R.M. & Silver, P.G. 1996. Cordillera formation, mantle dynamics and the
Wilson Cycle. Geology, 24, 511–514.
Segerstrom, K. & Ruiz, C. 1962. Carta Geológica de Chile, Hoja Copiapó, No.
6, escala 1:100,000. Instituto de Investigaciónes Geológicas de Chile,
Santiago.
St. Amand, P. & Allen, C.R. 1960. Strike-slip faulting in northern Chile.
Geological Society of America Bulletin, 71, 1965.
Scheuber, E. & González, G. 1999. Tectonics of the Jurassic–Early Cretaceous
magmatic arc of the north Chilean Coastal Cordillera (228–268S): a story of
crustal deformation along a convergent plate boundary. Tectonics, 18,
895–910.
Scheuber, E. & Reutter, K.J. 1992. Magmatic arc tectonics in the Central Andes
between 218 and 258S. Tectonophysics, 205, 127–140.
Simpson, C., Whitmeyer, S., De Paor, D.G., Gromet, R., Miro, R., Krol, M.A.
& Short, H. 2001. Sequential ductile through brittle reactivation of major
fault zones along the accretionary margin of Gondwana in central Argentina.
In: Holdsworth, R.E., Strachan, R.A., Magloughlin, J. & Knipe, R.J.
(eds) The Nature and Tectonic Significance of Fault Zone Weakening.
Geological Society, London, Special Publications, 186, 233–255.
442 J. G RO C OT T & G . K . TAY L O R

Suárez, M. & Bell, C.M. 1992. Triassic rift-related sedimentary basins in
northern Chile (248–298S). Journal of South American Earth Sciences, 6,
109–121.
Suárez, M. & Bell, C.M. 1994. Braided rivers, lakes and sabkahs of the Upper
Triassic Cifuncho Formation, Atacama Region, Chile. Journal of South
American Earth Sciences, 7, 25–33.
Sylvester, H. & Palacios, C. 1992. Transpressional structures in the
Andes between the Atacama Fault Zone and the West Fissure System at
278S, III Region, Chile. Zentralblatt für Geologie und Paläontologie, 6,
1645–1658.
Taylor, G.K.T. & Randall, D.E. 2000. Structural analysis of dyke emplacement
directions as an aid to palaeomagnetic studies: an example from northern
Chile. Geophysical Journal International, 141, 253–258.
Taylor, G.K., Grocott, J., Pope, A. & Randall, D.E. 1998. Mesozoic fault
systems, deformation and fault block rotation in the Andean forearc: a crustal
scale strike-slip duplex in the Coastal Cordillera of northern Chile.
Tectonophysics, 299, 93–109.
Vila, T., Lindsay, N. & Zamora, R. 1997. Geology of the Manto Verde copper
deposit, northern Chile: a specularite-rich, hydrothermal–tectonic breccia
related to the Atacama Fault Zone. In: Camus, F., Sillitoe, R.H. &
Petersen, R. (eds) Andean Copper Deposits: New Discoveries, Mineraliza-
tion Styles and Metallogeny. Society of Economic Geologists Special
Publication, 5, 157–170.
Waschbusch, P. & Beaumont, C. 1996. Effect of a retreating subduction zone on
deformation in simple regions of plate convergence. Journal of Geophysical
Research—Solid Earth, 101(B12), 28133–28148.
Wilson, J. 1996. Emplacement of the Las Tazas plutonic complex, Coastal
Cordillera, northern Chile. PhD thesis, University of Kingston, Kingston-
upon-Thames.
Wilson, J. 1998. Magnetic susceptibility patterns in a Cordilleran granitoid: the
Las Tazas complex, northern Chile. Journal of Geophysical Research, 103,
5257–5267.
Wilson, J. & Grocott, J. 1999. The emplacement of the Las Tazas complex,
northern Chile: the relationship between local and regional strain. Journal of
Structural Geology, 21, 1513–1523.
Wilson, J., Dallmeyer, R.D. & Grocott, J. 2000. New 40 Ar/39 Ar dates from the
Las Tazas complex, northern Chile: tectonic significance. Journal of South
American Earth Science, 13, 115–122.

Received 12 September 2001; revised typescript accepted 31 January 2002.

Scientific edity by Rob Strachan