See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/299587754

Characterization of Triassic Rifting in Peru and implications for the early

disassembly of western Pangaea

Article  in  Gondwana Research · April 2016

DOI: 10.1016/j.gr.2016.02.008

CITATIONS READS

59 2,561

8 authors, including:

Richard Spikings Mariel Reitsma

University of Geneva HRH Geology
147 PUBLICATIONS   3,980 CITATIONS    12 PUBLICATIONS   308 CITATIONS   

SEE PROFILE SEE PROFILE

Flora Boekhout Aleksandar Miskovic

University of Münster Terra Balcanica Resources
18 PUBLICATIONS   242 CITATIONS    28 PUBLICATIONS   854 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

The tectonic history of the Antarctic Peninsula during the Mesozoic: from the paleo-Pacific margin of Gondwana to the West Antarctica active margin View project

Large ignous provinces View project

All content following this page was uploaded by Aleksandar Miskovic on 31 January 2018.

The user has requested enhancement of the downloaded file.

 Gondwana Research 35 (2016) 124–143

Contents lists available at ScienceDirect

Gondwana Research

journal homepage: www.elsevier.com/locate/gr

Characterisation of Triassic rifting in Peru and implications for the early

disassembly of western Pangaea
R. Spikings a,⁎, M.J. Reitsma b, F. Boekhout c, A. Mišković d, A. Ulianov e, M. Chiaradia a, A. Gerdes f, U. Schaltegger a
a
Department of Earth Sciences and the Environment, University of Geneva, Rue des Maraichers 13, Geneva 1205, Switzerland
b
HRH Geology, 19 Silverburn Place, Aberdeen AB23 8GE, United Kingdom
c
Institut für Geologie und Paläontologie, Corrensstrasse 24, D-48149 Münster, Germany
d
Deptartment of Earth, Ocean and Atmospheric Sciences, The University of British Columbia, Vancouver V6T 1Z4, Canada
e
Institute of Mineralogy and Geochemistry, University of Lausanne, Switzerland
f
Institute of Geosciences, Mineralogy, J. W. Goethe University, Frankfurt 60438, Germany

a r t i c l e i n f o a b s t r a c t

Article history: Triassic igneous and sedimentary rocks exposed within the basement of the Andes were deposited in a series of
Received 8 December 2015 rifts, and may record the early disassembly of western Pangaea. These rocks are sporadically exposed along
Received in revised form 10 February 2016 almost the entire length of western South America, although reliable geochronological and isotopic data are
Accepted 17 February 2016
sparse. We combine new geochronological (zircon U–Pb), isotopic (Hf, Nd) and geochemical data with strati-
Available online 1 April 2016
graphic observations to constrain the age and tectonic setting of the Mitu Rift of southern Peru. The Peruvian
Handling Editor: F. Pirajno Mitu Rift is compared with other Triassic rifts in Colombia and Ecuador (Palanda Rift; 240–225 Ma), Bolivia
(Mitu Rift; Triassic), Bolivia, Chile and Argentina (e.g. Cuyo Basin; 246–230 Ma), and conclusions are reached re-
Keywords: garding the relationship between Triassic extension along the western margin of Pangaea, and the eventual for-
Pangaea mation of the Proto-Caribbean and Central Atlantic oceans. The Mitu Rift (Peru) was active during ~245–240 to
South America ~220 Ma and was synchronous with rifting along the Pacific margin of Colombia and Ecuador, along the Chilean
Triassic rifts margin and western Argentina, and probably rifting within Bolivia. Rifting north of the Huancabamba Deflection
Geochronology was accompanied by subduction and led to seafloor spreading, whereas rifting along the Peruvian and Chilean
Geochemistry
margins mainly occurred in the absence of subduction and terminated prior to the formation of extensive tran-
Mitu Group
sitional crust. Extension within Peru and Chile probably occurred via a combination of transtension, steepening
and detachment of an arrested slab. We propose that plate tectonic forces initiated the early break-up of Pangaea
by attenuating its margins and enhancing mantle upwelling. Prolonged extension may have propagated along
pre-existing weak zones that extended into the continental interior, captured melts derived from the upwelled
mantle forming a LIP (e.g. Central Atlantic Magmatic Province), became hot and weak and eventually lead to
the formation of a juvenile ocean (e.g. Central Atlantic).
© 2016 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.

1. Introduction
The amalgamation of Pangaea in the Carboniferous–Permian
coincided with a drastic reduction in plate velocities and an almost
complete pause in continental drift during the Late Permian–early
Jurassic (Vilas and Valencio, 1978). This relatively stationary period
was characterised by heat accumulation, the production of large
volumes of magmas (e.g. Kay et al., 1989) and geographically extensive
extension along western South America (e.g. Charrier et al., 2007;
Spikings et al., 2015). Increased plate velocities in the latest Triassic–
earliest Jurassic were coeval with the disassembly of Pangaea, although
the relationship between Triassic extension and the final disassembly of
Pangaea is unclear. We present a detailed stratigraphic, geochronological
⁎ Corresponding author.
E-mail address: richard.spikings@unige.ch (R. Spikings).
and geochemical study of the Triassic ‘Mitu’ rift of Peru, and compare this
phase with Middle–Late Triassic rifts in Colombia, Ecuador, Bolivia,
Argentina and Chile. This compilation provides a means to assess the
overall timing of extension along western Pangaea prior to its disassem-
bly, examine the nature of the forces that drove extension, and discuss
their influence on the break-up of Pangaea.
Extension along the northern sections of western South America
resulted in the formation of oceanic lithosphere that separated South
America from crust that currently forms part of Central America and
Mexico (Cochrane et al., 2014a; Spikings et al., 2015). However, exten-
sion along the Peruvian and Chilean margins terminated before oceanic
lithosphere could form. The aborted Mitu Rift of southern Peru experi-
enced the highest amount of extension along the Peruvian margin,
forming depocenters for syn-rift volcanic and sedimentary units
(Mitu Group; Fig. 1) that are still partly preserved. The Mitu Group
was traditionally assigned to the Permo-Triassic (Dalmayrac et al.,

http://dx.doi.org/10.1016/j.gr.2016.02.008
1342-937X/© 2016 International Association for Gondwana Research. Published by Elsevier B.V. All rights reserved.
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143 125

Volcanic and Sedimentary

Ecuador rocks
(oceanic
Quito
Foreland Late Triassic - Jurassic
plateau and A Allinccapac Volcanic Complex
arc; accreted Chocolate Fm. (216 - 135 Ma)
at 75 Ma) Volcanic arc
EC Peru: Pucará Group
Ecuador: Santiago Fm.
Limestones
Triassic
Mitu Group (Peru), red siliclastic
rocks, alkaline lavas
Plutonic rocks
5˚S Peru Jurassic
Foreland Ecuador: Abitagua, Zamora
Batholiths (185-145 Ma)
Peru: Ilo Batholith (173-152 Ma)
Triassic
Ecuador: Sabanilla Migmatite
Tres Lagunas Granite

Su
Permian

b-
EC

An
Peru: Permian granitoids

de
Carboniferous

an
Peru: Calc-alkaline
granitoids

re
gi
10˚S 500 km

on
H. Huanca
Foreland

Lima
EC

Nazca Pallpa
Plate Cordillera de
Carabaya
A
Huanicapa
Lake
15˚S Mamuera Titicaca
Ar
eq Altiplano
ui
pa
e

200-176 Ma
dg

Ilo Batholith
Ri

173-152 Ma
a
zc
Na

Chile
80˚W 75˚W 70˚W

Fig. 1. Simplified geological map of the Eastern Cordillera and coastal region of the Arequipa Terrane of Peru (Leon et al., 2000), and the southern Eastern Cordillera of Ecuador (Litherland
et al., 1994), showing the distribution of Carboniferous to Jurassic rocks. Sampled sections through the Mitu Group and the cordilleras Andahuaylas and Carabaya are labeled. U–Pb (zircon)
age ranges for the Jurassic batholiths are taken from Boekhout et al. (2012; Peru) and Spikings et al. (2015; Ecuador), and ages for the Chocolate Fm. are from Boekhout et al. (2013). Red
boxes labeled A, B are the map regions shown in Fig. 2a, and b, respectively. EC: Eastern Cordillera.

1980; Carlotto, 1998; Kontak et al., 1990; Sempere et al., 2002), al-
though a paucity of fossils has hindered accurate estimates of its age
and duration, which were largely based on the ages of the bracketing
formations. The upper levels of the underlying Copacabana Group
have been constrained by palynology and foraminifera to the Artinskian
(Doubinger and Marocco, 1981; Dalmayrac et al., 1980; the timescale of
Gradstein et al. (2012) is used throughout this manuscript). An angular
unconformity related to the Tardi-Hercynian Orogeny (Laubacher,
1978; Rosas et al., 2007) separates the Copacabana and Mitu groups
and renders the age estimate for the basal Mitu Group imprecise. The
Pucará Group, regarded by Rosas et al. (2007) to represent a thermal
sag phase that occurred subsequent to extension and deposition of the
Mitu Group, is attributed to the Late Triassic–Early Jurassic on the
basis of ammonite fossils and U‐Pb zircon ages from ash beds (Jaillard
et al., 1990; Schaltegger et al., 2008; Wotzlaw et al., 2014).
This study presents stratigraphic, geochronological (U–Pb zircon),
geochemical and isotopic (Hf and Nd) data from the sedimentary and
volcanic rocks of the syn-rift Mitu Group, and coeval granitoid intru-
sions of the southern Eastern Cordillera of Peru. These data are used to
develop a temporal and general tectonic framework for the rift se-
quence, and to propose driving forces for its formation and termination.
2. Geological framework and previous work
Most of Peru was covered by an epeiric sea in the earliest Permian,
giving rise to deposition of carbonates of the Copacabana Group
(Fig. 2; e.g. Dalmayrac et al., 1980), which contain abundant shallow
marine fossils, and intercalations of sandstone and shale. Siliciclastic
input becomes dominant in the final stages (Laubacher, 1978;
Doubinger and Marocco, 1981; Reitsma, 2012) of the upper levels of
126 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

a
12oS
0 50 100 km

217.8 ± 3.5
13oS
Ayacucho B. Machancha
Co
rdi 304.3±0.1 Sacred Valley C
ller or 236.3±0.9 (120)
243.4 ± 6.7 ad C di
eA lle
nd Huanicapa P ra
238.4 ± 4.9 ah Cusco Pallpa de
ua
yla Ca
s ra
Abancay
ba
235.1±1.8(122) ya
226.5±4.2(163)
226.1±2.2(91) 216.1±3.1 207±3.4
Macusani
14oS
223.7±0.1(71)
Sicuani
This study; U-Pb zircon (sample code MRxx) Mamuera 227.7±5.6
Reitsma (2012); U-Pb zircon 235.6±6.1
226±10
Miskovic et al. (2009); U-Pb zircon 227.4±5.4
Sample for geochemistry
74ºW 72oW 70oW
Sedimentary and Plutonic rocks
volcanic rocks
b Upper Jurassic - Cretaceous Cenozoic
o
10 30’S Huambutio, Socosani
and Huancané Fms. Jurassic
238±3.6 Siliclastic rocks
Lower Jurassic
Cerro de Pasco Triassic
246.8±4.6
Allinccapac Volcanic
Complex
Pucará Group Palaeozoic
227.4±3.8
Limestones
Triassic
Hacienda Mitu Group
Huanca Red siliciclastic rocks,
alkaline lavas
11oS
Carboniferous - Permian
Copacabana and Ambo Groups
Limestones, sandstone, shale,
0 25 km ignimbrites

Devonian (and older)

76oW 75°30’W Phyllites, schists

Fig. 2. Geological maps of the sampled regions of the Eastern Cordillera of Peru, showing the location of the Mitu Group, coeval Triassic igneous rocks, and their relationship to rocks that
unconformably overlie the Triassic units. The Jurassic Allinccapac Complex is also shown, along with the locations of rocks sampled for geochemical and geochronological analyses.
Previous and new zircon U–Pb ages are labeled. C: Calca, P: Pisac. a) Cordilleras Andahuaylas and Carabaya, b) central Eastern Cordillera. Maps modified after Leon et al. (2000).

the Copacabana Group, indicating a partial retreat of the inland sea.
The top of the Copacabana Group is assigned to the Artinskian
(lower Permian; 284–275 Ma), based on palynology (Doubinger and
Marocco, 1981) and foraminifera (Dalmayrac et al., 1980).
The poorly constrained Late Permian–Early Triassic Tardihercynian
deformation phase affected the rocks of the Eastern Cordillera of central
and south-eastern Peru (Fig. 1), giving rise to folding and an angular un-
conformity between Permo-Carboniferous rocks and the overlying
Mitu Group (Laubacher, 1978; Rosas et al., 2007). However, in many
locations this phase is only recognized as a sharp erosional surface due
to regional uplift that affected the Copacabana Group (Dalmayrac
et al., 1980).
The Mitu Group is mainly exposed on the Altiplano and in the Eastern
Cordillera of Peru (Fig. 1), while drill cores reveal the presence of the Mitu
Group in the subsurface of the Amazon foreland basin (Mathalone and
Montoya, 1995). The locally N 2000 m thick Mitu Group has been
extensively studied in southern Peru (Laubacher, 1978; Carlotto, 1998;
Marocco, 1978; Cenki et al., 2000; Panca, 2010), where it is composed
of red, siliclastic rocks that are intercalated with alkaline lava flows and
were deposited in subaerial basins. Clastic deposits are arkosic and
range from mudstone to coarse sandstones and conglomerates. Clasts
in conglomerates mainly originate from reworking of Mitu Group lavas,
and a minor component was derived from the underlying units
(Marocco, 1978; Mégard, 1978; Dalmayrac et al., 1980). Depositional en-
vironments are interpreted to be alluvial fans and fluvial systems
(Sempere et al., 2002; Cenki et al., 2000; Panca, 2010).
Volcanic rocks of the Mitu Group are mainly basaltic to andesitic
in composition, although rhyolites and (welded) ash beds are also
recognized. The mafic lavas usually have trachytic textures with up to
cm-sized plagioclase phenocrysts, with pyroxene and occasional olivine
in the matrix. Whole rock geochemistry indicates that the lavas are alka-
line with an intraplate signature (Kontak et al., 1990; Cenki et al., 2000).
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143 127

Previous age determinations of the Mitu Group are limited to K–Ar
dates of 280 ± 3 Ma and 245 ± 3 Ma for two basalts in northern
Bolivia (Kontak et al., 1990), and a U–Pb zircon age of 226 ± 10 Ma
for a rhyolitic tuff in the Sicuani area (Fig. 2; Mišković et al., 2009).
Therefore, the Permo-Triassic age that has been traditionally assigned
to the Mitu Group was largely based on the ages of the bracketing
formations, and radiometric systems that are easily disturbed.
Triassic plutons are exposed in a belt that trends parallel to the expo-
sures of the Mitu Group, and are most voluminous in the Cordillera de
Carabaya (Figs. 1 and 2) in southern Peru. The Triassic plutons step
~ 130 km to the south-west from the Cordillera de Carabaya to the
Cordillera de Andahuaylas (Fig. 2), where they diminish in volume
and disappear completely within Peru north of 10°S (Fig. 1; Mišković
et al., 2009; Reitsma, 2012). This waning of Triassic plutonism to the
north corresponds with a paucity of lava flows in the Mitu Group in
central and northern Peru (Mégard, 1978).
Helwig (1972) suggested that volcanic rocks of the Mitu Group
formed part of a main arc axis along the continental margin. However,
the distance between the Mitu Group and the present day trench (300
to 550 km), in combination with the alkaline character of the lavas led
Noble et al. (1978) and Dalmayrac et al. (1980) to reject the arc hypoth-
esis. Noble et al. (1978) proposed that the Mitu Group was deposited in
a back-arc basin, and that the arc was removed by younger tectonic pro-
cesses. Alternatively, Vivier et al. (1976) and Dalmayrac et al. (1980)
proposed a continental rift setting in the absence of an arc, and this
has remained the most widely accepted hypothesis (e.g. Carlotto,
1998; Kontak et al., 1990; Sempere et al., 2002) because it accounts
for the geochemistry of the lavas, subaerial sedimentation within
(half)-grabens and the absence of a voluminous Permo-Triassic arc.
More recently, Reitsma (2012) revived the back-arc hypothesis to ex-
plain the emplacement of Triassic granitoids in the Peruvian Eastern
Cordillera, based on geometric, geochemical and tectonic constraints.
Reitsma (2012) proposed that the Triassic arc was obliterated by high
rates of subduction erosion along the Peruvian margin during the Late
Mesozoic–Cenozoic (Clift et al., 2003; Stern, 2011).
The basins that hosted the Mitu Group in northern and central Peru
experienced a marine ingression in the Norian (Late Triassic; 227–
208 Ma). Rosas et al. (2007) suggested that marine conditions and sub-
sidence were driven by thermal sagging after extensional stresses
ceased. Platform carbonates, black shales and evaporites of the Pucará
Group (Fig. 1) accumulated in an epeiric sea. Detailed ammonite stratig-
raphy and high precision U–Pb zircon dating (Schaltegger et al., 2008;
Guex et al., 2012) has shown that the Triassic–Jurassic boundary inter-
sects the middle member of the Pucará Group. Marine conditions
persisted in northern and central Peru until the Toarcian (~ 180 Ma;
Jaillard et al., 1990; Rosas et al., 2007). However, the time span of
the Pucará Group corresponds to a hiatus in the sedimentary record in
southern Peru, where the Mitu Group is in erosional contact with Mid-
dle Jurassic to Cretaceous units (Fig. 2). Sempere et al. (2002) suggest
that the fluvio-eolian Quilcapunco Formation, which conformably over-
lies the Mitu Group northwest of Lake Titicaca (Fig. 1), is the only time
equivalent sequence of the Pucará Group in southernmost Peru.

NW H. Huanca Huanicapa Pallpa Mamuera SE

1100
Huancané
Pucará

MR229 MR172
Fm.
Socosani 1000 MR171
Fm. max.
600 1400 223±8 Ma
100
900

1200
500 80
800

MR237
40 700
1000
400
MR86 MR236
20 max. 600
235±2 Ma
800
MR234
MR41 239±2 Ma 500
300 0m

Limestone 600 MR233

Siltstone
400
Sandstone Lava MR232
max.
200 Breccia / 255±3 Ma
Conglomerate 300
400
Detrital rock: U-Pb zircon
200 MR100
Volcanic rock: U-Pb zircon and geochemistry
100
Volcanic rock: Whole rock geochemistry 200 MR115
MR99
Unconformity 100 234±1 Ma
Exposure gap
GR6-0611 in section
max. 0m
0m 246±2 Ma 0
C’bana C’bana Ambo
Group

Fig. 3. Stratigraphic sections through the Mitu Group along the Sacred Valley (Pallpa section), close to Sicuani (Mamuera section), close to Abancay (Huanicapa section) and close to Cerro
de Pasco (Hacienda Huanca section). Sections are not drawn to the same scale. The contact between the Copacabana and Mitu groups is not exposed at the base of the Huanicapa section,
and thus the local base probably does not represent the base of the Mitu Group. Ages are zircon U–Pb ages of lavas and tuffs, or the youngest age obtained from detrital zircons in
sedimentary rocks (see text).
128 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

Jurassic arc rocks are restricted to southern Peru, and include

4.92

−4.20
0.51
−0.97
5.83
5.83
6.22
5.03
5.77
−1.95
ɛNdi
metaluminous gabbros to granodiorites of the Coastal Batholith in the
Arequipa region (200 ± 1.1 Ma to 175.8 ± 1.2 Ma; Demouy et al.,
2012), and the Ilo Batholith (173.3 ± 1.3 to 151.5 ± 0.8 Ma, Boekhout

−0.71 ± 0.56 (2.8)

et al., 2012; Fig. 1). The intrusive and extrusive peralkaline Allincapac

−3.80 ± 1.0 (40)

complex was emplaced in the Cordillera de Carabaya at the end of the
Lower Jurassic. K–Ar dates of 175 ± 4 Ma (Fig. 2; Kontak et al., 1990)
and 184 Ma (no reported uncertainty; Stewart et al., 1974) are compa-

ɛHfi
rable with more recent U–Pb zircon dates of 195 ± 11 Ma and 184 ±
4 Ma (Mišković et al., 2009), which were obtained from nepheline sye-
nites. These peralkaline rocks are thought to have been formed by de-

−2.2 ± 0.4

−0.3 ± 0.8
compression melting of the shallowest asthenospheric mantle in an

ɛHfi high
extensional back-arc setting (Mišković et al., 2009; Reitsma, 2012).

3. Methods

−4.9 ± 0.2

−1.3 ± 0.5
3.1. Sampled stratigraphic sections and Triassic intrusions

ɛHfi low
Five stratigraphic sections of the Mitu Group were studied, which are
the Mamuera section near Sicuani, the Pallpa and Huancalle sections in
the Sacred Valley, the Huanicapa section near Abancay in southern Peru

LA-ICPMS
LA-ICPMS
LA-ICPMS
LA-ICPMS
LA-ICPMS

LA-ICPMS
LA-ICPMS
ID-TIMS
(Fig. 2) and the Hacienda Huanca section in the vicinity of Cerro de

Method
Pasco in central Peru (Fig. 1). Geochronological analyses of volcanic ash,
lavas and detrital grains in reworked clastic sequences were performed
in all of the sections using samples that were taken as close to their strat-

n/Ntota

95/105
96/102
93/109
79/104
igraphic base as possible (Fig. 3). The top of the Mitu Group was sampled

93/98
69/89
20/23
6/9
in the Mamuera section. The Huancané Formation, which overlies the
Mitu Group in the Sacred Valley, was sampled in the Pallpa section near
Baños Machancha (Fig. 2). Lava flows were sampled for whole rock geo-

234.3 ± 0.3 (2.5)

238.7 ± 1.8 (2.2)

chemistry throughout the sections, and in the surrounding regions.
Age± 2σ (Ma)

Geochronological data have been obtained using U–Pb zircon

176.7 ± 4.5
234.8 ± 1.4
223.4 ± 7.5
255.4 ± 3.2
246.2 ± 2.4
165.3 ± 3.3
(CA-ID-TIMS and LA-ICPMS) analyses, whole rock major oxide,
trace and REE data have been obtained using XRF and LA-ICPMS
analyses, and Nd (whole rock) and Hf (zircon) isotopic data have
been acquired using TIMS (Nd) and LA-MC-ICPMS (Hf). Methodo-
Summary of U–Pb ages and Hf and Nd isotopic data of volcanic rocks, and detrital zircons from sedimentary rocks.

logical details are presented in the appendix and the data are

White sandstone
White sandstone
Basaltic andesite
Basaltic andesite

Andesitic dacite
Trachy andesite

Rhyodacite tuff
Trachyandesite
Alkaline basalt
Alkaline basalt

Alkaline basalt

Red sandstone
Red sandstone
Red sandstone
Red sandstone
summarised in Tables 1a and 1b.
Trachy dacite
Lithology

Trachyte

4. Results: Mitu Group and bounding stratigraphic sequences

4.1. Geochronology and stratigraphy

71° 46° 48″

72° 50° 24″

71° 55′ 12″

71° 45′ 36″
70° 30′ 36″

71° 04′ 48″

71° 48′ 00″
72° 49′ 12″
70° 50′ 24″
71° 52′ 48″
71° 42′ 00″
71° 55′ 12″
71° 54′ 36″
71° 15′ 00″
Long. (W)

71° 4° 48″

76° 4° 12″
71° 4′ 48″

The stratigraphy of the Mitu Group in distinct regions is summarised

in composite sections, which have also been used to present the results
of U–Pb age determinations of sedimentary and volcanic rocks (Fig. 3).
Weighted mean and youngest detrital LA-ICP-MS and ID-TIMS U–Pb
analyses are summarised in Tables 1a and 1b, and data from individual
10° 49′ 12″
13° 16′ 12″
13° 27′ 36″
13° 52′ 12″

14° 21′ 36″

13° 27′ 36″

13° 33′ 00″

14° 21′ 00″
13° 28′ 12″
13° 35′ 24″
13° 25′ 12″
13° 27′ 36″
13° 32′ 24″
13° 15′ 36″
13° 15′ 00″
14° 12′ 36″
14° 21′ 36″

detrital and volcanic zircons are listed in Supplementary Table 2.

Lat. (S)

4.1.1. Sacred Valley

The Sacred Valley is a 50 km long valley located 20 km north of the
city of Cusco, and is regarded as the type locality for the Mitu Group
Huancané Fm.
Huancané Fm.
Stratigraphy

Allinccapac

(Fig. 2). Red siltstones, sandstones, conglomerates, alkaline basalts

Mitu Gp.

Mitu Gp.
Mitu Gp.

Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.
Mitu Gp.

and andesites crop out along the flanks of the valley, and the rocks are
folded in a south-east plunging anticline, with a fold axis that is parallel
to the length of the basin. The Copacabana Group crops out in the core
Sedimentary rocks (detrital zircons)

of the anticline, in the southern part of the Sacred Valley. An alkali gran-
Sacred Valley
Sacred Valley
Sacred Valley
Sacred Valley
Sacred Valley
Sacred Valley

ite dated at 304.3 ± 0.1 Ma (Fig. 2; zircon U–Pb; Reitsma, 2012) is in

Hda Huanca
Huanipaca
Mamuera
Macusani

contact with the Mitu Group in the north-west of the valley, although
Abancay
Location

Sicuani
Sicuani
Sicuani

Pallpa

Pallpa

the nature of the contact is unclear. The Mitu Group is overlain by the
Calca

upper Jurassic Huambutio or the lower Cretaceous Huancané Fms. via

well-preserved erosional surfaces.
Volcanic rocks

GR6-0611
MR100*
MR234*
Sample

4.1.1.1. Pallpa section (UTM 19L0196605/8510300). The 1500 m long

MR229
MR171
MR232
MR66*
MR23*
MR07*
MR18*
MR33*
MR62*
MR63*
MR73*
MR99*
Table 1a

MR61
MR86

section near the village of Pallpa has been described in detail by Panca
(2010; Figs. 2 and 3). Here, the Mitu Group overlies carbonates,
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
Table 1b
Summary of U–Pb ages and Hf, and Nd isotopic data from Triassic intrusive rocks.
Sample Location Stratigraphy Lat. (S) Long. (W)
Intrusive rocks
MR71* Coasa Triassic intrusion 13° 57′ 36″ 70° 18′ 36″
MR91* Abancay Triassic intrusion 13° 45′ 00″ 72° 55′ 12″
MR120* Marcapata Triassic intrusion 13° 33′ 36″ 70° 54′ 24″
MR122* Ocongate Triassic intrusion 13° 30′ 36″ 71° 08′ 24″
MR163* Coasa Triassic intrusion 13° 52′ 48″ 70° 10′ 48″
a: number of analyses used in weighted mean calculation over total number of analyses, discordant analyses, xenocrystic cores and analyses with slight inheritance or Pb loss are not used
in weighted mean calculation of the age.
ID-TIMS: Isotope Dilution Thermal Ionisation Mass Spectrometry.
LA-ICPMS: Laser Ablation-Inductively Coupled Mass Spectrometry.
⁎ Also analysed for geochemistry.
Numbers in parentheses are MSWD's.
Hf isotopic measurements are obtained from dated zircons; Nd isotopic measurements are on whole rock.
sandstones and ignimbrites of the Copacabana Group (Reitsma, 2012).
The earliest Mitu Group sedimentation occurred across a floodplain,
which evolved into fluvial channels at a time that was synchronous
with an increase in alkaline volcanism. High rates of sedimentation and
an increase in volcanic activity favoured the development of alluvial
fans. The sequence is overlain by eolian sandstones of the Cretaceous
Huancané Fm.
Pink, medium-grained sandstone MR232 was deposited in a fluvial
setting and was sampled ~500 m from the base of the section (Fig. 3).
104 zircons have been dated by LA-ICP-MS for a provenance study,
and 79 grains yielded concordant ages. The youngest grain cluster has
a 238U–206Pb age of 255.4 ± 3.2 Ma (Tables 1a and 1b), and most zircons
in the sample are between 450 and 550 Ma. The probability distribution
curve shows peaks at ~254, ~483 and ~529 Ma (Fig. 4a).
A green, rhyolitic crystal tuff (MR234) containing biotite and zircon
was sampled ~200 m higher up the section (Fig. 3), and LA-ICP-MS anal-
yses of 23 zircons yielded 21 concordant ages. The offset of the youngest
concordant grain with respect to the other grains was interpreted as
minor Pb-loss and therefore it was rejected from the data set, yielding
a weighted mean age of 238.7 ± 1.8 Ma (MSWD of 2.2; Fig. 5a).
A medium-grained, yellow sandstone (MR229) was taken from the
Huancané Formation (Fig. 3). All grains, including the zircons are ex-
tremely well rounded and 95 out of 105 zircons yielded concordant
ages. A single grain yields an age of 176.7 ± 4.5 Ma, and all other zircons
are Middle Triassic and older (Fig. 6a; Supplementary Table 2).
4.1.1.2. Calca (UTM 19L0183245/8531499). The eolian Huancané Forma-
tion is exposed near Baños de Machancha in the vicinity of the village of
Calca (Fig. 2), where it unconformably overlies the Mitu Group and is
overthrusted by the Tarma–Copacabana Group (Reitsma, 2012).
Eighty-nine zircons extracted from medium-grained, yellow sandstone
MR61 were dated, yielding 69 concordant ages. Numerous ages define a
rather continuous range from 165 Ma to 2605 Ma, with most grains
occurring in the 450–650 Ma interval, while older grains become pro-
gressively scarcer (Fig. 6b). The youngest, concordant zircon indicates
a maximum depositional age of 165.3 ± 3.3 Ma.
4.1.2. Sicuani (UTM 19L0275280/8411039)
The Mamuera section (Fig. 3) is a kilometre long section located
20 km south-east of the city of Sicuani (Fig. 2). The Mitu Group directly
overlies the Carboniferous Ambo Group, indicating that the Copacabana
Group is missing although it has been observed elsewhere in the Sicuani
area. The Ambo Group consists of cross-bedded coarse sandstones with
pebble-rich channel fills. The Mitu Group is represented by an alterna-
tion of lava dominated parts and red sandstone intervals (Fig. 3). The
volcanic rocks are mainly basaltic andesite with plagioclase pheno-
crysts, and in parts the lavas are very rich in mm sized vesicles that
are filled with amygdales (often calcite). The sandstone layers have a
clear volcanoclastic source and vary from fine to coarse, including
129
Lithology Age± 2σ (Ma) n/Ntota Method ɛHfi ɛNdi
Syenogranite 223.7 ± 0.1 (0.5) 3/4 ID-TIMS −0.50 ± 1.10(8.4) −1.80
Granodiorite 226.1 ± 2.2 (6.5) 25/30 LA-ICPMS −0.63 ± 0.36(1.1) −2.50
Gabbro 236.3 ± 0.9 (2.4) 25/25 LA-ICPMS 3.47 ± 0.56(2.1) 1.00
Granodiorite 235.1 ± 1.8 (4.2) 21/23 LA-ICPMS −1.06 ± 0.60(2.1) −1.03
Granodiorite 226.5 ± 4.2 (3.2) 4/25 LA-ICPMS 1.31 ± 0.27(0.9)
cross-bedding and pebble beds. Pebbles are mainly of volcanic origin
but vein quartz is also observed.
One of the oldest lavas in the section is trachy-andesite MR99,
and nine of its zircons have been dated by ID-TIMS. Two zircons yielded
Permian ages and are regarded as xenocrysts (Supplementary Table 2),
which may be derived from the previously documented Permian conti-
nental arc (e.g. Mišković et al., 2009). Despite the chemical annealing-
leaching treatment of the zircons, one grain still displays Pb-loss
although the high blank renders this measurement imprecise (Supple-
mentary Table 2). The weighted mean of the five remaining grains
yields a 206Pb–238U age of 234.3 ± 0.3 Ma (MSWD of 2.5; Fig. 5b).
Further towards the top of the section the detrital zircons of
medium-grained red sandstone MR171 were dated by LA-ICP-MS. The
detrital population is dominated by Triassic grains, with the probability
curve peaking at ~ 235 Ma, while the youngest zircon cluster yields a
U–Pb age of 223.4 ± 7.5 Ma (Fig. 4b).
4.1.3. Abancay (UTM 18L0733712/8500945)
The Huanicapa section is situated north of the city of Abancay, along
the dirt road towards the village of Huanicapa (Fig. 2). A ~100-meter
thick sequence through the Mitu Group is exposed (Fig. 3) while further
along the road black mudstones and fine sandstones of the Copacabana
Group crop out, which contain fossil remnants of leaves and stems of
Permian Lycopsids (Ianuzzi et al., 2000). The contact between the
Copacabana and Mitu groups is not exposed. The bottom of the Mitu
Group is dominated by sedimentary rocks ranging from red mudstones
to bedded sandstones, and some limestone or calcareous beds. A few chan-
nel fills are observed, as well as cavities left by gypsum rose crystals, and
fine layers are often disturbed by bioturbation. The sediments were depos-
ited across a floodplain and the limestone beds are remnants of oxbow
lakes. The top part of the section is dominated by andesitic basalts, and
the lavas are very rich in vesicles that are filled with feldspar amygdales.
A depositional hiatus separates the Mitu Group and the overlying lime-
stones and shales of the lower Jurassic Socosani Formation.
Red sandstone MR86 contains cm-scale coarse white sandstone
channels which were sampled for detrital zircon analyses. Ninety six
zircons yielded concordant ages, and two thirds of the zircon population
is Triassic with a probability curve peak at ~240 Ma, while the youngest
zircon cluster yields an age of 234.8 ± 1.4 Ma (Fig. 4c). The older dates
are interpreted as reworking of Permian and older magmatic belts.
Basalt MR41 was sampled at the bottom of the section and its geo-
chemical composition has been determined, although other lavas in
the section were not suitable for geochemical analyses due to the high
percentage of amygdales.
4.1.4. Cerro de Pasco (UTM 18L0381299/88215979)
The Hacienda Huanca section is located near Cerro de Pasco in
central Peru (Fig. 2), and exposes 620 m of the Mitu Group. Here, the
Mitu Group is underlain by lava flows of the Ambo Group and carbon-
ates and green sandstones of the Copacabana Group, and is overlain
130 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

A B
Sandstone, Mitu Group, Sacred Valley, Pallpa section
25 Sandstone, Mitu Group, Sicuani, Mamuera section
50
483 529 Ma MR232, n = 79 235 Ma MR171, n = 93
Ma
20 Youngest zircon: 40 Youngest zircon:
Relative probability

254
Ma 250.7 ± 4.9 Ma 217.3 ± 4.1 Ma
Youngest cluster (n = 9): Youngest cluster (n = 3):
Number

15 30
255.4 ± 3.2 Ma 223.4 ± 7.5 Ma

10 20

5 10

0 0
0 400 800 1200 1600 2000 2400 2800 100 500 900 1300 1700 2100
Age (Ma) Age (Ma)
C D
Sandstone, Mitu Group, Abancay, Huanicapa section Sandstone, Mitu Group, Cerro de Pasco, Huanca section
80
240 Ma 271 Ma
25
MR86, n = 96 GR6-0611, n = 93
Youngest zircon: Youngest zircon:
Relative probability

60 20
224.3 ± 7.4 Ma 241.5 ± 6.0 Ma
Youngest cluster (n = 18): Youngest cluster (n = 8):
Number

234.8 ± 1.4 Ma 15 246.2 ± 2.4Ma

40 248
Ma
10

20
5

0 0
0 250 500 750 1000 1250 1500 1750 2000 100 300 500 700 900 1100 1300
Age (Ma) Age (Ma)

Fig. 4. Histograms (blue) and probability density plots (red) of concordant LA-ICP-MS ages acquired from detrital zircons of the Mitu Group. Youngest cluster is defined as the weighted
mean age of zircons that overlap with the youngest zircon. N = number of grains plotted.

by limestones of the Pucará Group. The Mitu Group mainly consists of
fine to medium-grained red sandstones and conglomerates with a red
sandy matrix and sub-rounded volcanic and sandstone clasts (Fig. 3).
The conglomerates are grouped into fining upward sequences, each of
which is 3 to 4 m thick. Only one volcanic interval was observed,
which is located near the top of the section. The Hacienda Huanca
section is discussed in detail in Rodriguez et al. (2011).
Detrital zircons were extracted from medium-grained, red sandstone
GR6-0611 (Fig. 3), which has fine planar bedding and is located near
the base of the section. Ninety three zircons yielded ages that are
Mesoproterozoic–early Neoproterozoic and latest Carboniferous–middle
Triassic, with only 2 grains in between these age groups (Fig. 4d). The
youngest zircons cluster at 246.2 ± 2.4 Ma, and define the maximum
stratigraphic age of this sandstone.
4.2. Geochemistry of lavas of the Mitu Group
4.2.1. Major oxides (whole rock)
Geochemical data have been obtained from lavas from dispersed ex-
posures of the Mitu Group (Fig. 7). The lava flows show a high degree of
weathering, sericitization of plagioclase, and post-depositional silicifica-
tion, yielding loss on ignition (LOI) values averaging 4.5 wt.%. Whole
rock compositions have been normalized to an anhydrous composition
in the graphical representations, and the raw data is listed in
Supplementary Table 3.
Triassic lavas of the Mitu Group have SiO2 and MgO values varying
between 44–80 wt.% and 0.4–13 wt.%. Alkalis were probably partially
remobilized during weathering, so we use the subalkaline–alkaline divi-
sion of de la Roche et al. (1980; Fig. 7a), which utilises major elements.
Most lavas of the Mitu Group plot in the alkaline field except for those
with SiO2 N 60 wt.%. SiO2 content of Mitu Group lavas is inversely corre-
lated with MgO, CaO, Fe2O3 and TiO2. Tectonic discrimination using Ti
and V (Fig. 7b; Shervais, 1982) reveals the alkaline and intra-plate na-
ture of the lavas from the Sacred Valley and at Sicuani. Lavas from the
Sacred Valley and near Sicuani follow the same major oxide trends,
although those from the Abancay area yield distinctly higher Al2O3
(Fig. 7c), and lower TiO2 concentrations (Supplementary Table 3).
4.2.2. Trace and rare earth elements (whole rock)
Due to the weathered nature of the lavas, we prefer to assign
lithological names based on immobile elements. Triassic lavas of the Mitu
Group along the Sacred Valley are mainly alkaline basalts in Zr/TiO2 v
Nb/Y space (Fig. 7d; Winchester and Floyd, 1977), and the more felsic sam-
ples plot as trachy-andesite/dacite (MR33, 60 and 62) and rhyodacite
(MR234). Lavas from Sicuani and Abancay with SiO2 b 65 wt.% are sub-
alkaline, and are basaltic to andesitic. Lava MR99 and welded tuff MR115
are trachy-andesites (Sicuani, Mamuera section; Fig. 7d).
Basalts of the Sacred Valley yield a positive concave up correlation
between Ni and MgO (Fig. 7e), suggesting that Ni was efficiently re-
moved during early fractionation. Lavas from the Sicuani area follow

Fig. 5. Concordia plots of zircon U–Pb data obtained from sedimentary rocks of the Mitu Group, and Triassic intrusions within the cordilleras Andahuaylas and Carabaya. Concordant
analyses are shown in red, and have been used to calculate the weighted mean age. N = total number of obtained ages.
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143 131

A B
206Pb/238U Mitu Group, Sacred Valley, Pallpa Section
Mitu Group, Sicuani, Mamuera Section
MR234 0.0374
Rhyolitic tuff MR99
0.040 Trachy-andesite 236
250
0.0372
235

0.038 240 0.0370

234

0.0368 233
230
0.036
206Pb/238U 232 206Pb/238U age:
age: 0.0366
238.7 ± 1.8 Ma 234.34 ± 0.30 Ma
220 231 MSWD = 2.5, n = 6
MSWD = 2.2, n = 20
0.0364
0.034
0.235 0.245 0.255 0.265 0.275 0.285
0.20 0.22 0.24 0.26 0.28 0.30 0.32 0.34
C D
0.0395 0.041
MR120 MR122
Gabbro Granodiorite
250
0.0385 0.039

240

0.0375 0.037
230

0.0365 206Pb/238U age: 0.035 206Pb/238Uage:

230
236.3 ± 0.9 Ma 235.1± 1.8 Ma
MSWD = 2.4, n = 25 210 MSWD = 4.2, n = 21
0.0355 0.033
0.22 0.24 0.26 0.28 0.30 0.22 0.24 0.26 0.28 0.30 0.32
E F
MR71 206Pb/238U
age:
0.03540
MR163
224.2 Syenogranite 0.039 Granodiorite 226.5 ± 4.2 Ma
MSWD = 3.2, n = 4
0.03536 0.037
230
0.03532 223.8 0.035
220
0.033
0.03528 0.23 0.24 0.25 0.26
223.4
207Pb/235U
206Pb/238U age:
0.03524 223.73±0.09 Ma
MSWD =0.46, n =3
0.03520
0.244 0.246 0.248 0.250 0.252

G
0.041 MR91 260
Granodiorite
0.039
240
0.037

0.035 220

0.033
200
0.031
206Pb/238U
age:
0.029 226.1±2.2 Ma
MSWD =6.5, n =25
0.027
0.20 0.22 0.24 0.26 0.28 0.30
207Pb/235U
132 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

the same trend, although basaltic andesite MR172 (Mamuera section, A Huancané Fm., Sacred Valley, Pallpa section
Sicuani) has a higher Ni concentration compared to samples with the 20
565 Ma MR229
same MgO content. All lavas from the Sicuani section yield a fairly
constant Sr/Sc ratio, regardless of their MgO content (Fig. 7f), while
Yellow quartz arenite
16

Relative probability
the Sr/Sc ratio of lavas of the Sacred Valley increased during early frac-
tionation, and subsequently decreased from b 5 wt.% MgO. Basalts and
Youngest zircon:
176.7 ± 4.5 Ma

Number
andesites at Sicuani yield an inverse relationship between Y and Rb/Y
12 n = 95
(Fig. 7g), whereas alkaline basalts and trachy-andesite MR33 from the
Sacred Valley show a positive correlation between Y and Rb/Y. Rhyolitic
tuff MR234 from the Sacred Valley has a high Rb/Y, and trachy-andesite 8
MR60 and dacite MR62 have high Y concentrations.
Triassic alkaline basalts from the Sacred Valley yield low LILE/HFSE
(Fig. 8a) and no negative Nb or Ta anomalies, and thus are not character- 4
istic of subduction related magmas. More felsic samples from the Sacred
Valley (trachy-andesite MR60, trachydacite MR62, rhyolitic tuff MR234)
yield slightly negative Nb–Ta anomalies, and only trachyandesites, 0
trachydacite and rhyolite (MR33, 60, 62 and 234) have strong negative 0 400 800 1200 1600 2000 2400 2800 3200
Sr and Ti anomalies (Fig. 8a). The trace element patterns of lavas of the
Sicuani area overlap with the least enriched lavas from the Sacred B Huancané Fm., Sacred Valley, Calca
Valley, except that they show more pronounced negative Nb–Ta and 478 Ma
12 MR61
La–Ce anomalies than the latter (Fig. 8c). LILE and REE abundances are Yellow quartz arenite
generally higher in Triassic lavas of the Mitu Group than in coeval, rift-

Relative probability
10
related amphibolitised basalts in Ecuador and Colombia which resemble Youngest zircon:
E-MORB rocks. 165.3 ± 3.3 Ma
REE abundances and La/Yb ratios for Triassic lavas of the Mitu Group Number 8
n = 69
with similar SiO2 content decrease from the Sacred Valley to Sicuani and
Abancay (Figs. 7h and 8). Most basalts within the Sacred Valley lack Eu 6
anomalies and yield continually lower concentrations from light
to heavy REE (Fig. 8b). Rhyolitic tuff MR234 has the same slope but 4
with a strongly negative Eu anomaly, and trachy-andesite MR60
and trachy-dacite MR62 have flat HREE abundances. Andesitic dacite 2
MR73 and basaltic andesites MR100 and MR172 from Sicuani have flat-
ter REE slopes than the more alkaline lavas of the Sacred Valley (Fig. 7h),
0
and differ completely from the more fractionated rhyolite (MR99) and 0 400 800 1200 1600 2000 2400 2800
welded rhyolitic tuff (MR115) within the same section, which have Age (Ma)
complicated REE patterns (Fig. 8d). These latter two have strong
negative Eu anomalies and steep, negative HREE gradients. Fig. 6. Histograms (blue) and probability density plots (red) of concordant LA-ICP-MS ages
acquired from detrital zircons of the Huancané Fm. that is exposed along the Sacred Valley.
4.2.3. Nd (whole rock) and Hf (zircon) isotopic ratios Youngest cluster is defined as the weighted mean age of zircons that overlap with the
youngest zircon. N = number of grains plotted.
The 143Nd/144Ndi ratios of nine Triassic lavas of the Mitu Group range
from 0.512122 to 0.512660 (Tables 1a and 1b and Supplementary
Table 4). Nd-isotopic ratios correlate positively with incompatible ele- 5. Results: Triassic intrusions
ments (e.g. La), suggesting that the lavas remained closed with respect
to Sm and Nd, over the life span of the samples. The high Sm (N 8 ppm) 5.1. Geochronology
and Nd (N 38 ppm) concentrations of the lavas from the Sacred Valley
are characteristic of alkaline rocks, while the lower concentrations in Five Triassic felsic intrusions from the cordilleras Andahuaylas and
the Abancay and Sicuani sections trend towards values found in calc- Carabaya (Fig. 2) were dated using CA-ID-TIMS and LA-ICPMS and
alkaline rocks (Herrmann, 1970). yield weighted mean U–Pb crystallisation ages from concordant zircons
Lavas of the Sicuani area are clearly distinguishable from those of the ranging between 236.3 ± 0.9 and 223.7 ± 0.1 Ma (Fig. 10a; Tables 1a
Sacred Valley by their lower εNdi values (−4.2–0.5 and 4.8–6.2, respec- and 1b and Supplementary Table 6). Some of these dates yield
tively; Fig. 9a; Tables 1a and 1b). The εNdi value of the basaltic andesite MSWD values N2.5, perhaps revealing an anticrystic and xenocrystic
(MR23) from Abancay lies within the range of the lavas from Sicuani. component that could be investigated with additional CA-ID-TIMS anal-
εNdi versus SiO2 (Fig. 9a) shows a negative correlation for lavas of the yses. Nevertheless, these dates are considered to be useful estimates of
Sicuani area, indicating crustal contamination, while εNdi values of the time of crystallisation of the intrusions, and overlap with the
lavas of the Sacred Valley remain invariable over a compositional crystallisation ages of lavas and the maximum stratigraphic dates of
range from alkaline basalt (MR63) to trachydacite (MR62). The range sedimentary rocks within the Mitu Group. These ages overlap with a
of εNdi from the Sacred Valley overlaps with coeval gabbroic sills and peak in ages obtained from peraluminous intrusions in Colombia,
dykes in Ecuador and Colombia (εNdi 5 to 10; Spikings et al., 2015). Ecuador and northern Chile (Fig. 10b).
Hf isotopic compositions of zircon in a trachyandesite (MR99) from
Sicuani and a rhyodacitic tuff (MR234) from the Pallpa section (Sacred 5.2. Geochemistry of Triassic granitoid intrusions
Valley) yield mixed populations with ƐHft values ranging between of
−4.9 ± 0.2 to −2.2 ± 0.4 and −1.3 ± 0.5 to −0.3 ± 0.8, respectively 5.2.1. Major oxides and trace elements
(Fig. 9b; Tables 1a and 1b and Supplementary Table 5). These are signif- A majority of Triassic granitoids in the cordilleras Andahuaylas and
icantly less radiogenic than the most juvenile compositions from coeval Carabaya are classified as granodiorites to alkali granites in the cationic
gabbroic dykes in Ecuador and Colombia (ƐHft of 5 to 15; Spikings et al., scheme of De La Roche et al., 1980 (Fig. 11a), although a single
2015). gabbronorite was also sampled. These rocks mainly lie within the calc-
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143 133

a b
2000 V (ppm)
Ti/V = 10 20 Continental 50
400

6Ca+2Mg+Al
flood basalt
Peralkaline Alkaline Subalkaline 350
300
Alkali basalt

basaltalkaline
250 MORB 100
& BAB
1000 200

Calc-
150
100
50

0 0
-2000 -1000 0 1000 2000 3000 4000 0 5 10 15 20 25
4Si-11(Na+K)-2(Fe+Ti) Ti(ppm)/1000
c d
Jurassic (184-195 Ma)
22 back-arc Phonolite
Mitu Group (Triassic) 1 Com/Pant

20 Sacred Valley
Al2O3 (wt%)

Sicuani Rhyolite

Zr/TiO2
Abancay
.1 Trachyte
18 Rhyodacite/Dacite
TrachyAnd
Andesite
16 Bsn/Nph
.01 Andesite/Basalt

SubAlkaline Basalt Alkaline Basalt

14
.001
40 50 60 70 80 .01 .1 1 10
Nb/Y
e SiO2 (wt%)
200 150
f
MR41
150
100
Sr/Sc

MR172
Ni (ppm)

100

50
50

0
0 5 00 5 10 15
MgO (wt%) 10 15
MgO (wt%)
g h
12
rhyolites MR99, 115, 234
andesitic dacite MR73
10
10
La/Sm

As
sim
Rb/Y

ila 8
tio
n
Frac
tionatio
n 6
5
MR100
4
MR62 MR99
MR60 MR172

0 2 MR73
10 30 50 70 0 10 20 30 40
Y (ppm) La/Yb

Fig. 7. Geochemical data acquired from lavas of the Mitu Group. Multi-cation discrimination plot (a) from De La Roche et al. (1980), tectonic discrimination plot (b) from Shervais (1982),
lithological discrimination (d) from Winchester and Floyd (1977). Arrow in (f) indicates increasing Sr/Sc during pyroxene fractionation, and decreasing due to subsequent plagioclase
fractionation. The (pink) field defined by lavas of the Jurassic back-arc (Allinccapac Complex) is shown as a comparison (data from Reitsma (2012)).

alkaline differentiation trend on the modified alkali-lime index of Cordillera of Peru (Mišković et al., 2009), and Triassic intrusions in
Peacock (1931; Fig. 1b; Supplementary Table 3), and overlap with a the Eastern and Central cordilleras of Ecuador and Colombia
broader field encapsulating all Triassic intrusions in the Eastern (Spikings et al., 2015), respectively. The Triassic intrusions dated here
134 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

Triassic
Rock/Primitive Mantle Rock/Chondrite
a) Mitu Group,
10000
Sacred Valley
Sacred Valley, basalt
Sacred Valley, SiO2>54 wt.%
1000 100 MR62
MR60

100

10 10
b) Mitu Group, MR234
MR33, 60, Sacred Valley
1
62, 234
Sacred Valley, basalt
Sacred Valley, SiO2 >54 wt.%
0.1 Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr Sm Eu Ti Dy Y Yb Lu 1
La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

10000 c) Mitu Group d) Mitu Group

Sicuani MR100 Sicuani
Abancay MR172 Abancay
1000 100

100

10 10

1 MR115
MR99
0.1 1
Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr Sm Eu Ti Dy Y Yb Lu La Ce Pr Nd Pm Sm Eu Gd Tb Dy Ho Er Tm Yb Lu

Triassic amphibolitised basalt, Ecuador and Colombia (Spikings et al., 2015)

Fig. 8. Trace and rare earth multi-element plots of lavas of the Mitu Group, normalized to Primitive Mantle (Trace Elements; Sun and McDonough, 1989) and Chondrite (REE; Boynton, 1984).

have mildly peraluminous Aluminium Saturation Indices (Fig. 11c),
similar to coeval intrusions dated throughout the Peruvian Eastern
Cordillera by Mišković et al. (2009). However, these are far less
peraluminous than Triassic intrusions from Ecuador and Colombia
(Spikings et al., 2015).
Primitive mantle normalised trace element abundances (Fig. 11d)
reveal negative Nb and Ta anomalies, suggesting that a subduction com-
ponent was incorporated into these rocks. Trace element concentrations
normalized to the composition of average upper continental crust
(Fig. 11e; Taylor and McLennan, 1995) plot close to unity, suggesting
the subduction component was derived from melting of continental
crust. The trace element compositions are indistinguishable from coeval
S-type monzogranites in Ecuador and Colombia (Fig. 11d), and
show negative Ba, Eu, Sr and Ti anomalies, suggesting that plagioclase
and Fe–Ti oxides have fractionated. REE abundances in the Triassic
granitoids normalized to N-MORB reveal light-REE enrichment
with (La/Yb)n ranging between 63.3 and 5.7, and no correlation with
206
Pb–238U crystallisation age (Fig. 11f).

a Mitu Group (Triassic)

b
Sacred Valley
MR163 MR120
10 Sicuani MR91 MR122
6.0
Abancay
Triassic Granitoid 4.0
5 MR63
MR62
2.0
εNd(i) εHf(t)
0 0.0
MR23
-2.0 MR71
-5
-4.0

-10 -6.0
40 50 60 70 80 220 225 230 235 240
SiO2 (wt%) Zircon 238U/206Pb age±2σ (Ma)

Fig. 9. a) Nd (whole rock) and b) Hf (zircon) isotopic compositions of Triassic lavas of the Mitu Group and Triassic intrusions within the cordilleras Andahuaylas and Carabaya. Hf data
points represent single, coeval zircons, and the pink bars represent 2σ uncertainties (LA-ICPMS analyses only).
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143 135

a b
Pe r u v i a n E a s t e r n C o r d i l l e ra Peru: Eastern Cordillera
Cordilleras Andahuaylas
Frequency Colombia: Central Cordillera
and Carabaya
238U/206Pb (zircon) age±2σ (Ma)
290 This work 8
Ecuador: Eastern Cordillera
Miskovic et al. (2009)
Reitsma (2012) N. Chile (Andes and
270 Coastal Cordillera)
6

250
4

Mitu Gp.
Triassic

230

2
210

190 0
11 12 13 14 200 210 220 230 240 250 260
Latitude (°S) 238U/206Pb (zircon) age±2σ (Ma)

Fig. 10. a) Comparison between U–Pb zircon crystallisation ages of Triassic intrusions throughout the Eastern Cordillera of Peru (which includes the cordilleras Andahuaylas and
Carabaya); data compiled from this study, and Mišković et al. (2009). b) Histogram of concordant zircon U–Pb ages obtained from intrusions exposed in the cordilleras Andahuaylas
and Carabaya of Peru (this study), Eastern and Central Cordilleras of Ecuador and Colombia, respectively (Spikings et al., 2015), and the Andes and coastal cordillera of northern Chile
(Maksaev et al., 2014).

5.2.2. Nd (whole rock) and Hf (zircon) isotopic ratios
Nd (whole rock) isotopic compositions of 4 granitoids yields ƐNdi
values that range between −2.50 and 1.00 (Tables 1a and 1b), which
are similar to those obtained from lavas in the region of Sicuani
(Fig. 9a; Supplementary Table 4). Hf (zircon) isotopic measurements
of zircons from 5 granitoids yield ƐHft ranging between 3.47 ± 0.56
and − 1.06 ± 0.60 (Fig. 9b; Tables 1a and 1b and Supplementary
Table 5), which overlaps with the Hf isotopic composition of Triassic
lavas of the Mitu Group. Syenogranite MR71 yields a large dispersion
in single zircon ƐHft with an MSWD of 8.1, which may be due to partial
melting of heterogeneous continental crust, corroborating the geo-
chemical data. There is no correlation between ƐHft and crystallisation
age.
6. Interpretation
6.1. Age of the Mitu Group
6.1.1. Age of the basal Mitu Group of southern and central Peru
The Mitu Group is underlain by the Copacabana Group (Fig. 3),
whose upper levels are assigned to the Artinskian (lower Permian;
Doubinger and Marocco, 1981; Dalmayrac et al., 1980). Therefore,
sedimentation within the Mitu Group was traditionally assumed to
have started in the middle or upper Permian (e.g. Sempere et al.,
2002). However, our work suggests that deposition of the Mitu Group
did not start before the Middle Triassic.
Each sedimentary section was sampled as close to its base as possi-
ble, yielding similar depositional ages over a distance of 670 km along
orogenic strike. The age for the onset of sedimentation in the Mamuera
section is constrained by the ID-TIMS age of 234.3 ± 0.3 Ma for the first
volcanic interval (trachy-andesite MR99; Fig. 3). A similar age (238.7 ±
1.8 Ma) was obtained from tuff MR234, situated 750 m above the base of
the Pallpa section (Fig. 3). Sandstone MR232 located below the ash bed
in the Pallpa section gave a maximum detrital age of 255.4 ± 3.2 Ma
(Fig. 4a). The sandstone and the ash bed are stratigraphically less than
200 m apart (Fig. 3), and thus the older age of the youngest zircons in
the sandstone are a consequence of either i) no zircons were present
during deposition due to the absence of felsic volcanism related to the
Mitu Group prior to 239 Ma, or ii) the Mitu Group was not significantly
reworking itself in this location. We speculate that the approximately
500 m thick section of siltstones of the Mitu Group located beneath
sandstone MR232 have depositional ages that extend into the lower,
Middle Triassic. This is based on depositional constrains obtained from
the other sections. It is possible that these sedimentary rocks extend
to older dates, although there is currently no evidence to support
Permian or even Lower Triassic rocks in the Mitu Group.
The sandstone taken near the base of the Huanca section in the
Eastern Cordillera (Cerro de Pasco; GR6-0611) has a Middle Triassic
maximum depositional age (246.2 ± 2.4 Ma; Fig. 4d), which is slight-
ly older than the volcanic ages found in the Mamuera and Pallpa sec-
tions. The contact between the Copacabana and Mitu Groups is not
exposed in the Huanicapa section, and it is unclear how close sand-
stone MR86 is to the base of the Mitu Group. The maximum age of
the sandstone indicates that it is stratigraphically younger than the
samples discussed above. Interestingly, a majority of the zircons
are Triassic in age, with the probability curve peaking at ~ 240 Ma
(Fig. 4c), indicating that magmatism within the Mitu Group probably
started in the early Middle Triassic (Anisian) in the Abancay region,
and that the Mitu Group significantly reworked itself in this region
with relatively little input from outside the basin. Sandstone MR171,
which was taken towards the top of the Mamuera section, yields a com-
parable pattern where the probability curve peaks at ~235 Ma (Fig. 4b),
suggesting that the volcanic event found at the base of the section was
reworked.
Deposition of the Mitu Group started in the beginning of the
Middle Triassic (~ 245–240 Ma, Anisian), and basin initiation was ap-
proximately synchronous along orogenic strike in central and southern
Peru.
6.1.2. Age of the upper Mitu Group
The Mitu Group is overlain by carbonates of the Pucará Group in
northern and central Peru (Fig. 1). Marine invasion of the ‘Mitu’ basins
started in the Norian (Rosas et al., 2007; 227–208 Ma), constraining
the end of sedimentation within the Mitu Group in northern and central
Peru. The Mitu Group is overlain by upper Jurassic or Cretaceous sedi-
mentary rocks in most places within southern Peru (Figs. 2 and 3), via
an erosional hiatus. The stratigraphically youngest sample of the Mitu
Group that was dated in the Mamuera section (MR171) yielded a max-
imum stratigraphic age of 223.4 ± 7.5 Ma, corroborating the Norian
start of the Pucará Group in northern and central Peru.
Two eolian sandstones from the Huancané Formation in the Sacred
Valley (Fig. 3) do not tightly constrain the end of sedimentation within
the Mitu Group due to the large hiatus between the two formations.
136 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

a
2000 b
Triassic
This study Gabbronorite
1800

6Ca+2Mg+Al (millications)
Colombia/Ecuador Ol. 10
1600 Jurassic Gabbro

Na2O + K2O - CaO (wt%)

Allinccapac (back-arc)
1400 8
Gabbro
E. Cord. Peru
1200 Syenodiorite Diorite 6 Late Triassic
1000 Alkalic
Tonalite
800 4
Qtz
Neph. Granodiorite
monzonite cic
600 Syenite
Monz cal
ogran 2 li-
Syenite it
Syenograenit ka e
400 Qtz e Al lin
ka
syenite 0 al S-type
200 Alkali granite lc-
Ca Calcic
Lachlan FB
0 -2
0 500 1000 1500 2000 2500 3000 55 60 65 70 75 80
4Si - 11(Na+K)-2(Fe+Ti) (millications) SiO2 (wt%)

c d
4.0 Rock/Primitive Mantle
Peraluminous
1000 Colombia/Ecuador (Triassic)
3.5
Metalluminous

This study (Triassic)

3.0
Al/(Na+K)

2.5 100

2.0
1.5 E.Cord Peru 10
Late Triassic
1.0
0.5 Alkali
0.5 1.0 1.5 2.0 2.5 1 granite MR186
Al/(Ca+Na+K)
Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr SmEu Ti Dy Y Yb Lu

e f
Rock/Upper Continental Crust
10 (La/Yb)n
70
60
50
1
40
30
20
0.1
Alkali 10
granite MR186
0
220 225 230 235 240
Cs Rb Ba Th U Nb Ta K La Ce Pb Pr Sr P Nd Zr SmEu Ti Dy Y Yb Lu
Zircon 238U/206Pb age±2σ (Ma)

Fig. 11. Geochemistry of Triassic plutons of the cordilleras Andahuaylas and Carabaya (Peru), and comparisons with Jurassic back-arc lavas of the Allinccapac Complex (data from Reitsma
(2012)). Multi-cation discrimination plot (a) from De La Roche et al. (1980), alkaline–silica discrimination plot (b) from Peacock (1931), Aluminium Saturation indices (c) calculated using
Maniar and Piccoli (1989), and multi-element plots normalized to Primitive Mantle (Sun and McDonough, 1989) and Upper Continental Crust (Taylor and McLennan, 1995).

Both sandstones yield a tentative Middle Jurassic maximum stratigraph- which run parallel to exposures of the Mitu Group (Fig. 2). Plutons
ic age based on single youngest zircon ages (Fig. 6; MR61: 165.3 ± between Abancay and Ayacucho span between 243.4 ± 6.7 Ma and
3.3 Ma; MR229: 176.7 ± 4.5 Ma). 217.8 ± 3.5 Ma, and the majority of Triassic plutons of the Cordillera
de Carabaya were emplaced between 236.3 ± 0.9 Ma and 216.1 ±
6.1.3. Newly defined age of the Mitu Group 3.1 Ma, with one monzodiorite intruding at 207.0 ± 3.4 Ma. Emplace-
The zircon U–Pb age data suggests that sedimentation within the ment of the plutons and deposition of the Mitu Group were coeval,
Mitu Group started in the Middle Triassic (~240 Ma) and lasted until at and thus formed within the same tectonic environment.
least the Late Triassic (~ 220 Ma). Previous suggestions that the Mitu
Group was deposited during the Permo-Triassic are inaccurate, and the 6.2. Geochemical characterisation
revised stratigraphic age of the Mitu Group is ~245–240 Ma to ~220 Ma.
This new age range for the Mitu Group overlaps with recent 6.2.1. Triassic lavas of the Mitu Group
(Mišković et al., 2009) and new (this study) U–Pb zircon ages for plu- The alkaline to sub-alkaline nature of the basalts and andesites of the
tons within the Cordillera de Andahuaylas and Carabaya (Fig. 10a), Mitu Group (Figs. 7 a, b) supports a continental, intra-plate origin. Some
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
alkaline basalts from the Sacred Valley and basalt MR41 from Abancay
are relatively unfractionated (e.g. Fig. 7c), although early fractionation
of a ferromagnesian phase (probably olivine) efficiently removed com-
patible elements from the liquid (Fig. 7e). Alkaline basalts from the
Sacred Valley further evolved by pyroxene fractionation, driving an ini-
tial increase in the concentration of Al2O3 and Sr/Sc (Fig. 7c and f). The
decrease in Al concentration and Sr/Sc at ~ 5 wt.% MgO and 52 wt.%
SiO2 shows that plagioclase fractionation dominates in the more
evolved samples. The absence of negative Eu, Sr and Ba anomalies in
the alkaline basalts of the Sacred Valley (Fig. 8a, c) suggests the liquids
fractionated at depths greater than the plagioclase stability field, and
subsequently rapidly ascended to the surface. However, some basaltic
magma stalled, and was perhaps held in a crustal reservoir prior to
eruption, where it evolved to trachytic compositions via plagioclase
fractionation (e.g. MR33, 60 and 62).
Crustal assimilation significantly increases the Rb/Y ratio, and thus
trachyandesite MR99, rhyolites (115, 234) and andesitic dacite MR73
from Sicuani and the Sacred Valley attained their high SiO2 concentra-
tions from crustal assimilation (Fig. 7g). This corroborates the interme-
diate εHfi weighted mean values of zircons of − 0.7 ± 0.6 (MR234;
Tables 1a and 1b) and −3.8 ± 1.0 (MR99), and the large range of εHfi
in individual zircons and the presence of xenocrystic zircons in
trachyandesite MR99 (Supplementary Table 6). On the contrary, alka-
line basalts from the Sacred Valley only have slightly increasing Rb/Y
and trend towards evolution by fractional crystallization than via
assimilation. The high Y concentrations of trachyandesite MR60 and
trachydacite MR62 (Sacred Valley) were probably almost entirely
obtained by fractional crystallization.
Magma evolution in the region of Sicuani was dominated by crustal
assimilation, whereas fractional crystallization was significant in the re-
gion of the Sacred Valley. The range in SiO2 (44 to 62 wt.%) versus the
relatively invariable εNdi values (Fig. 9a) strongly suggests that the
lavas of the Sacred Valley evolved by varying degrees of partial melting
and/or fractional crystallization of a single source. On the contrary, the
inverse correlation between SiO2 and εNdi, combined less radiogenic
Nd isotopic compositions found in the lavas from Sicuani suggests that
these magmas were most likely enriched in incompatible elements by
varying degrees of crustal assimilation. If we assume that the lavas at
Sicuani originated from the same mantle reservoir as those in the Sacred
Valley, then assimilation of a calc-alkaline crust that was less radiogenic
would account for their lower alkalinity. Since basaltic andesite MR23
from the Abancay region plots near the SiO2 versus εNdi trend of the
lavas from the Sicuani area, we assume that it formed by mixing of
similar end-members.
6.2.2. Triassic intrusions
The dominant mechanism for generating the Triassic granodiorites
to alkali granites of the cordilleras Andahuaylas and Carabaya is
considered to be anatexis of the continental crust because their trace
element compositions plot close to unity when compared with upper
continental crust, and they are mildly peraluminous (Fig. 11). ƐHft
values for zircons span between − 1.06 and 3.47 (Fig. 9b), suggesting
that these crustal melts mixed with a juvenile component to varying de-
grees, which also erupted to form coeval alkaline to sub-alkaline lavas of
the Mitu Group. A comparison of ƐHft (zircon) with crystallisation age
reveals no trend, which suggests that neither the hot, isotopically
juvenile magmas nor crustal melts progressively dominated through
the Triassic. Similarly, a lack of any clear variation between (La/Yb)n
and time suggests that there was no systematic increase or decrease
in partial melting of the crust through the Middle Triassic. These obser-
vations contrast with coeval anatectites in Ecuador and Colombia,
where the ƐHft (zircon; mafic intrusions) is progressively more radio-
genic, and the (La/Yb)n ratios of the anatectites reduces during 240–
223 Ma (Spikings et al., 2015). These were interpreted to reflect and
increased flux of juvenile asthenospheric melts and heat through the
137
continental crust as it progressively attenuated, increasing the amount
of partial melting of the crust.
6.3. Geodynamic setting of the Mitu Rift
Sedimentological, geochemical and geochronological evidence strong-
ly suggests that the Mitu Group and coeval granitoids of the Peruvian
Eastern Cordillera were emplaced within a Middle Triassic continental
rift. First, subaerial terrigeneous sedimentary rocks of the Mitu Group
were deposited in fault controlled half grabens (Rosas et al., 2007;
Rodriguez and Cueva, 2010) throughout the Peruvian Eastern Cordillera.
The U–Pb ages of detrital zircons (Fig. 4b, c) indicates that sedimentary
input to these half grabens became progressively more restricted with
time, and the Mitu Group significantly reworked itself. If extension was
accompanied by thermal doming of the crust, this generates an efficient
topographic barrier for detrital input from beyond the attenuated zone.
Second, Triassic magmatism was dominated by alkaline lavas with
within-plate geochemical affinities, and an abundance of peraluminous
granitoid intrusions dominated by monzogranites, which are characteris-
tic of continental rift settings (e.g. Spikings et al., 2015). This interpreta-
tion is consistent with Vivier et al. (1976); Mégard (1978); Dalmayrac
et al. (1980), Kontak et al. (1985), Sempere et al. (2002) and Mišković
et al. (2009) who assign the Mitu Group to a continental rift setting.
The basins which host the siliclastic sedimentary rocks of the Mitu
Group started opening at ~ 245–240 Ma, which is considered to mark
the beginning of an extensional phase along the Peruvian margin. This
differs from previous interpetations (e.g. Sempere et al., 2002) who mis-
interpret K/Ar dates to suggest the Mitu Group was deposited in the
Permian. Extension post-dates Late Permian–Early Triassic deformation,
which has been assigned to the compressive Tardihercynian Orogeny
(e.g. Laubacher, 1978), and is represented in central Peru by a sharp ero-
sional surface that truncates the upper Copacabana Group. Marine inva-
sion of the Mitu basins started in the Norian (227–208 Ma; Rosas et al.,
2007) in northern and central Peru, perhaps due to thermal sagging of
cooling, extended crust. Similarly, the upper Mitu Group in southern
Peru is also truncated, and is in erosional contact with Middle Jurassic
sandstones. The “Mitu” Rift is considered to be a failed rift because
there is no evidence for extensive juvenile transitional crust (e.g.
Spikings et al., 2015), Triassic ophiolites, or a post-Triassic suture against
the Arequipa Terrane (Fig. 1), which may represent the western wall of
the rift axis. Consequently, the timing of extension is taken to be repre-
sented by the duration of the Mitu Group, which spans ~ 245–240 Ma
to 220 Ma, significantly revising earlier estimates which suggested rifting
was continuous from the Permian to the Triassic (e.g. Kontak et al., 1990).
Mafic magmatism during this period was characterised by alkaline
magmas derived from a presumably upwelling asthenospheric source,
which assimilated continental crust to varying degrees forming both as-
similation (Sicuani) and fractionation (Sacred Valley) dominated lavas.
The mafic lavas mainly occur south of 12°S within the Eastern Cordillera.
Felsic magmatism is dominated by peraluminous monzogranitic
intrusions that crop out along the Eastern Cordillera (including the cor-
dilleras Andahuaylas and Carabaya) between ~ 10.5 and 14°S (Fig. 2).
These intrusions generally become younger along-strike towards the
southeast (Fig. 10a), where exposure of the Mitu Group is the most
extensive within the Eastern Cordillera. This suggests that magmatic in-
trusions played an important role in controlling the pattern of extension
and rifting by thermally weakening the lithosphere in southern Peru.
Noble et al. (1978) and Reitsma (2012) suggested that extension
occurred in a back-arc setting, despite the current lack of a preserved
Triassic arc in Peru. These authors hypothesise that Triassic subduction
formed an arc, which has subsequently been obliterated by Cenozoic
tectonic erosion (e.g. Clift et al., 2003; Stern, 2011; Clift and Hartley,
2007). They cite evidence for Triassic arc magmatism in central Chile,
south of 33°S (see next section; Glodny et al., 2005), although central
Chile is currently separated from southern Peru by N 2000 km. The
Moro Mejillones Tonalite (208 ± 2 Ma; Fig. 12; Casquet et al., 2014) in
138 R. Spikings et al. / Gondwana Research 35 (2016) 124–143

northern Chile is the only Triassic arc remnant recorded proximal to
Peru, which post-dates rifting during ~ 245–240 Ma and ~ 220 Ma.
Given the absence of evidence for a Triassic arc in Peru, we propose
that the Mitu Group was deposited in a continental rift in the absence
of subduction. Franzese and Spalletti (2001) suggest that Triassic exten-
sion north of 40°S was driven by slab detachment, opening of an as-
thenospheric window, rock uplift, thermal weakening and gravitational
collapse.
Evidence suggests that Triassic and Jurassic extension in Peru do not
represent a single, continuous process, but rather that that they were
driven by unrelated forces. Norian carbonates (Pucará Fm.) and Middle
Jurassic sandstones deposited during Jurassic extension unconformably
overlie the Mitu Group, and Jurassic extensional axes in Peru are located
inboard of the Triassic rift (e.g. Jaillard et al., 1990) in northern and cen-
tral Peru, and trenchward in southern Peru (Sempere et al., 2002;
Boekhout et al., 2013). Furthermore, the latest Triassic and Jurassic are
characterised by subduction and arc magmatism in southern Peru
(Chocolate Fm. and the Ilo Batholith; Fig. 1), whereas the Triassic margin
of Peru and northern Chile was characterised by an inactive or detached
slab (e.g. Mišković et al., 2009). The termination of Triassic rifting coin-
cides with the abrupt onset of arc magmatism and subduction beneath
the Arequipa Terrane at ~216 Ma, forming the Chocolate Arc (Boekhout
et al., 2012, 2013). The initiation of subduction aborted the Mitu Rift,
presumably due to a change in plate convergence velocity.

80° 70° 60°

YUCATAN
10° Northern Andes
Continental disassembly and formation of oceanic lithosphere
Caracas
A A’
Payendé Rift ~216 Ma Central American NW Gondwana
YA

240-225 Ma Basement blocks

(Maya?)
MA

Continental break-up
and ophiolite formation
A
0°
Au
to c
hth o nous S

A’
AMAZONIA
u

Southern Peru
ns
as

Failed “Mitu” Rift and initiation of arc magmatism/compression(?)

10°
Ba

B B’
se

B’
en ~216 Ma
m

Lima Arequipa NW Gondwana

t

Ar Chocolate
B eq Entre Rios Arc
uip Branch compression
a Brasiliano
Mitu Rift
245 - 240 to 220 Ma mobile belt
Antofalla

20° (aborted)
Moro
Mejillones Tupiza
Branch
Inactive/detached slab Present day coastline
Ingaguás Subduction zone (late Cretaceous - Tertiary
Complex Fault allochthonous terranes
Ischigualasto in blue)
30° Triassic rift axis
Villa Unión Basin
236 - 231 Ma
CUENCA
CUYANA

Cuyo Basin R. PLATA

246 - 230 Ma ? Triassic red
Buenos
Aires siliciclastic rocks
Triassic anatectites
Los Menucos Fm. (mainly monzogranites)
40°
222±2 Ma Triassic alkaline lavas
Triassic conjugate margin
(currently detached)
Pata

? El Tranquilo Basin
(La Golondrina Rift) Undifferentiated Palaeozoic
onia g

Middle - Upper Triassic and younger basement

Pen

Sunsas (1.3 - 1.0 Ga)

ins

AFRICA
aged basement terranes
ula

50°
Palaeoproterozoic
and Archaean cratons

Fig. 12. Simplified illustration of the distribution of Triassic extensional basins and rifts, and associated sedimentary and igneous rocks along western South America during the Middle–
Late Triassic. Age ranges shown in red are derived from concordant zircon U–Pb (zircon) analyses of detrital zircons, and volcanic rocks (see main text). No concordant U–Pb ages have
been obtained from extensional basins in Bolivia. Approximate Triassic conjugate margins within Pangaea are shown in grey. Most of the Pacific margin of South America was extending
during ~240–220 Ma, and the loci of extension tracked relict terrane boundaries (e.g. Arequipa–Amazonia in Peru, and Sunsas-aged terranes in Argentina and Chile). Middle Triassic–Late
rifts are stratigraphically bound by erosional hiatuses and unconformities, and represent a distinct extensional episode that post-dates the extrusion of extensive rhyolitic provinces (e.g.
Choiyoi Complex), and pre-dates the onset of Andean subduction at ~208 Ma. The sections depict successful rifting north of the Huancabamba Deflection (A-A’), and the aborted Mitu rift
(B-B’). Eastern Pacific plate kinematics are drawn sinistral to South America, reflecting most structural measurements for Triassic western South America. Rock distributions taken from
Litherland et al. (1994); Leon et al. (2000); Sempere et al. (2002) and Morel et al. (2003).
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
6.4. Comparison with other regions of western South America
6.4.1. North of the Huancabamba Deflection (Colombia and Ecuador)
Pindell and Dewey (1982); Jaillard et al. (1990); Litherland et al.
(1994); Cochrane et al. (2014a) and Spikings et al. (2015; and
references therein) provide a detailed description of Triassic rifting
recorded in rocks exposed in the Cordillera Central of Colombia, and
the Cordillera Real of Ecuador. We refer to this rift zone as the Payendé
Rift (Fig. 12), which was also used by Senff (1995) to describe the
tectonic setting of Triassic, red siliclastic rocks of the Payendé Fm. in
northern Colombia. Continental rifting during 240–225 Ma (concordant
zircon U–Pb dates of bimodal magmatic rocks) is characterised by
bimodal magmatism, tholeiitic mafic lavas and metamorphosed,
peraluminous S-type crustal anatectites dominated by monzogranites.
Continental separation was complete by 225 Ma, forming tholeiitic
basalts with no evidence for contamination with continental crust,
and ophiolite sequences that yield crystallisation dates of ~ 216 Ma
(Martínez, 2007). Close temporal, faunal and geochemical similarities
suggest the conjugate margin is represented by parts of the basement
of Mexico and Guatemala (Fig. 12; e.g. the Oaxaquia Terrane and the
Acatlán Block; Spikings et al., 2015). Rifting is considered to have been
accompanied by sinistral shearing (Litherland et al., 1994) within a
back-arc setting (Cochrane et al., 2014a; Spikings et al., 2015) relative
to Triassic arc rocks that are currently preserved in Mexico (Keppie
et al., 2006; Helbig et al., 2012). The magmatic and sedimentary rocks
associated with Triassic (240–225 Ma) rifting north of the
Huancabamba Deflection (e.g. Sabanilla Migmatite, Tres Lagunas
Granite and the Cajamarca Complex) are mainly in faulted contact
with the younger late Triassic–Jurassic arc, which commenced at
~ 209 Ma in the Santander Massif (Van der Lelij et al., 2015; Spikings
et al., 2015). Jurassic extension had multiple effects, including the
emplacement of the Central Atlantic Magmatic Province during 201–
198 Ma (e.g. Jourdan et al., 2009; Blackburn et al., 2013), and the
separation of North America, Yucatan and South America at 195 Ma
(e.g. Beutel, 2009; Pindell and Kennan, 2009), forming the Proto-
Caribbean Seaway and the Central Atlantic Ocean. The relationship
between the driving forces of Triassic extension in northern South
America, and Jurassic extension which lead to the fragmentation of west-
ern Pangaea is unclear because the rift-related Triassic units are persis-
tently faulted against the Jurassic rocks, although the northern South
American margin remained passive until ~209 Ma (Spikings et al., 2015).
6.4.2. Eastern Cordillera and Subandean fold belts of Bolivia
Triassic rifting in Bolivia (Fig. 12) is defined by the distribution of
oxidised siliclastic rocks (Tiquina and Ipaguazú Fms.) in palaeograbens
(Sempere et al., 2002), abundant alkaline and tholeiitic intrusions and
lavas south of 17°S (Soler and Sempere, 1993) and few peralkaline in-
trusions (225–220 Ma; U–Pb zircon; Farrar et al., 1990) in the Eastern
Cordillera and surrounding sub-andean belts (Sempere et al., 2002).
Sempere et al. (2002) consider these to be geographically and temporal-
ly continuous with the Mitu Rift in Peru, and they define several distinct
rift axes, which is typical of continental rift environments (e.g. Red Sea–
Suez–Aqaba). However, Bertrand et al. (2014) report plagioclase
40
Ar/39Ar dates of 199–198 Ma (disturbed age spectra) from tholeiitic
basaltic sills along the Entre Rios rift branch, which they relate to the
Central Atlantic Magmatic Province. These basalts rest conformably on
the Ipaguazú Fm, and thus they do not temporally correlate with the
middle Triassic Mitu Rift of Peru. Summarising, the timing of extension
in Bolivia is poorly defined due to a paucity of accurate radiometric
dates. Sempere et al. (2002) suggest that the onset of rifting was
diachronous and becomes younger from central Peru towards Bolivia.
However, their interpretation is based on Permian K–Ar dates for the
Mitu Group in Peru, which we have shown to be erroneous. Similar to
the Mitu rift in the southern Peruvian Eastern Cordillera, rifting in
Bolivia is sealed by Jurassic fluvio-eolian sandstones (Sempere, 1995).
139
6.4.3. Chile and Argentina
A series of Middle–Late Triassic rift systems and sinistral pull-apart
basins with a NNW-SSE trend in northern and central Chile and
Argentina (Charrier, 1973) propagated along the hanging wall of previ-
ous sutures that separate Palaeozoic terranes (e.g. Ramos and Kay,
1991; Ramos, 1994). These sedimentary and volcanic successions are
mainly separated from older rocks via unconformities and disconfor-
mities (e.g. Zerfass et al., 2004), and occasionally rest upon thick
Palaeozoic continental deposits (e.g. Mancuso and Caselli, 2012). U–Pb
dates of tuffs from the Cuyo Basin (Fig. 12) span between 246 and
230 Ma (Potrerillos Fm., Spalletti et al., 2008; Rincón Blanco Group,
Barredo et al., 2012). A tuff from the rift-related Ischigualasto Fm.
(Ischigualasto-Villa Unión Basin; Fig. 12) yields a sanidine 40Ar/39Ar
date of 231.4 ± 0.3 Ma (Martinez et al., 2011 recalibration of the
227.8 ± 0.3 Ma age of Rogers et al., 1993). These are stratigraphically
consistent with ID-TIMS zircon U–Pb dates of 236–234 Ma obtained
from tuffs within the Chañares Fm., which also forms part of the
Ischigualasto-Villa Unión Basin (Marsicano et al., 2016). A tuff bed
within the Los Menucos depocenter (Fig. 12) gave a Rb–Sr whole rock
isochron date of 222 ± 2 Ma (Rapela et al., 1996). These overlap with
K–Ar dates of 238–221 Ma for monzogranites of younger plutons within
the Ingaguás Complex (Fig. 12; Mpodozis and Kay, 1992) of the Chilean
Frontal Cordillera, although potential argon loss renders these K–Ar
dates unreliable relative to the time of crystalisation. Maksaev et al.
(2014) compile and present new zircon U–Pb dates from volcanic
rocks in northern Chile, which reveals a peak in activity between 240
and 210 Ma. The Middle–Late Triassic volcanic rocks are associated
with coeval post-orogenic, peraluminous intrusions and bimodal
magmatism (e.g. Martin et al., 1999) in the high Chilean Andes and
Coastal Cordillera that yield zircon U–Pb dates that peak between 247
and 216 Ma in northern Chile (Maksaev et al., 2014; Fig. 10b). The
Late Triassic Tranquillo Fm. (Late Triassic; Fig. 12; Morel et al., 2003)
within the Permian–Triassic Golondrina Rift shows that Triassic exten-
sion can be traced to southernmost South America.
Slab steepening and detachment are thought to have occurred dur-
ing the Late Permian–Early Triassic, forming the Choiyoi Magmatic
Province (e.g. Ramos and Folguera, 2009; main activity 257–243 Ma;
e.g. Franzese and Spalletti, 2001; Llambias et al., 2003), which is widely
distributed in Chile and Argentina between 21°S and 44°S and pre-dates
the Middle Triassic basins. The Middle–Late Triassic rifts and associated
magmatic units are collectively considered to have formed subsequent
to slab-detachment, giving rise to intense crustal melting under exten-
sional tectonic conditions during a period of arrested subduction (e.g.
Kay et al., 1989; Parada et al., 2007; Spalletti et al., 2008). Subduction
recommenced during ~ 208–195 Ma (Charrier et al., 2007; Casquet
et al., 2014), coinciding with the disassembly of western Pangaea (e.g.
Spikings et al., 2015). However, other authors (e.g. Glodny et al., 2005;
Vasquez et al., 2011) suggest that subduction accompanied extension
in central Chile (34°S–37°S) throughout the Triassic, based on the pres-
ence of an accretionary wedge. Zerfass et al. (2004) also refer to some
Triassic basins throughout southern South America as back-arc basins,
although they present no evidence for subduction.
6.4.4. Summary
Several features are common to a majority of Middle–Late Triassic
rifts in South America. First, extension along the Mitu Rift (~ 245–
240 Ma to ~220 Ma) and the Payende Rift (240–225 Ma; culminating
with the formation of oceanic lithosphere at ~ 216 Ma) was synchro-
nous. Rifting may have been coeval in Bolivia, although the timing of ex-
tension is not accurately constrained by radiometric dates. Middle and
Late Triassic rifting in Chile, and western Argentina also coincided
with a peak in magmatism during 247–210 Ma (Maksaev et al., 2014),
suggesting they were driven by the same tectonic regime. Second, a ma-
jority of the Middle–Late Triassic rift sequences unconformably truncate
older rocks, and are themselves unconformably overlain by Late Triassic
and Jurassic sequences, suggesting they represent discrete events.
140 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
Finally, a majority of the Middle–Late Triassic rifts opened with sinistral
displacement (e.g. Litherland et al., 1994; Zerfass et al., 2004). These
similarities suggest that extension was driven by a single, common,
large scale process that affected almost the entire Pacific margin of
South America within Pangaea.
Middle-Late Triassic extension along the Payendé rift resulted in
the complete separation of Central American and Mexican terranes
(e.g. Maya Block) from South America, and the formation of oceanic lith-
osphere, whereas coeval rifts within Peru, Boliva, Chile and Argentina
were aborted, and did not lead to the fragmentation of South America
(Fig. 12). Late Triassic back-arc plutons in the Acatlán Complex
(Mexico; Helbig et al., 2012) were utilised by Cochrane et al. (2014a)
and Spikings et al. (2015) to propose that subduction drove rifting
along northwestern South America, which was also proposed by
Pindell and Tabbutt (1995). In contrast, no clear evidence exists for sub-
duction along the Peruvian and a majority of the Chilean margin be-
tween 240 and 210 Ma, and slab-pull forces do not seem to account
for Middle–Late Triassic extension in Peru, Bolivia and Chile. A potential
exception is a Triassic accretionary wedge in central Chile, although this
is currently only found between 34 and 37°S. Fig. 12 depicts a potential
plate kinematic scenario that could account for the common features
and differences found within the Middle–Late Triassic rifts within
South America. Plate-margin wide sinistral-directed stress, and a pauci-
ty of subduction south of the present-day Huancabamba Deflection
combined with coeval subduction to the north supports a model
where oceanic lithosphere of the eastern Pacific Ocean displaced with
a southward component along the entire margin, perhaps as a single co-
herent plate. The rate of subduction varied along-strike of the trench as
the ocean-continent angle changed, implying that the current broad-
scale topology of western South America pre-dates the Triassic. This hy-
pothesis implies that transtension may have been the dominant cause
of Middle–Late Triassic extension along the highly oblique Peruvian
and Chilean margins (e.g. Franzese and Spalletti, 2001; Zerfass et al.,
2004). Other authors have hypothesized that extension along the
Chilean margin was driven by slab detachment and lithospheric delam-
ination (e.g. Mpodozis and Ramos, 1989; Kay et al., 1989; Franzese and
Spalletti, 2001), or by slab-steepening above an active subduction zone
(Vasquez et al., 2011). Transtension may have been accompanied by
some, or all of these affects.
The rift basins and post-orogenic magmatism terminated during
~216–209 Ma in Colombia, Ecuador, Peru and northern Chile, coinciding
with the onset of subduction in those regions (Boekhout et al., 2013;
Casquet et al., 2014; Spikings et al., 2015) and the initiation of the so-
called Andean Cycle. The onset of subduction may have been associated
with compression in northern Chile (Mejillonia suspect terrane;
Casquet et al., 2014), presumably due to a change to less oblique plate
convergence directions (e.g. Pindell and Kennan, 2009). We speculate
that southern Peru may have also been compressed at ~216 Ma, giving
rise to an erosional contact between the Mitu Group and Jurassic
sandstones. This late Triassic compressional phase was rapidly followed
by extension and marine transgression, which prevailed until 140–
120 Ma along extensive tracts of the Andean Orogen e.g. (Jaillard
et al., 1990; Spikings et al., 2015). Arc magmatism may not have
commenced in central and southern Chile until the Middle Jurassic
(e.g. Kay et al., 1989; Parada et al., 2007), while other authors suggest
arc magmatism was continuous throughout the Triassic and Jurassic
within central Chile (Vasquez et al., 2011).
7. Conclusions
1. The Mitu Group consists of subaerial, oxidised fluvial and alluvial
sediments that were intercalated with alkaline basaltic lavas and
rhyodacitic tuffs, and were coeval with the emplacement of mildly
peraluminous, S-type intrusions. A combination of deposition of the
Mitu Group in half-grabens, topographic isolation from external
drainage and the alkaline to sub-alkaline character of the volcanic
rocks confirms that these rocks were deposited within a continental
rift, which formed at the same time along the Peruvian margin. The
alkaline magmas were derived from a common, juvenile source
and evolved by either fractional crystallisation at various depths
(Sacred Valley), or were dominated by crustal assimilation (Sicuani),
which perhaps reflects spatial and temporal variations in the amount
and rate of extension. These mafic magmas drove anatexis of the
upper continental crust, forming peraluminous intrusions that strike
parallel to the Mitu Group.
2. Concordant zircon U–Pb ages of lavas and tuffs, combined with ages
obtained from detrital zircons in sedimentary horizons constrain the
age of the Mitu Group from ~245–240 to ~220 Ma (Anisian–Norian),
which significantly revises earlier estimates that deposition com-
menced in the Permian (e.g. Sempere et al., 2002), and constrains
the time of Middle–Late Triassic extension in Peru. The Mitu Rift un-
conformably overlies a compressive phase of the Tardihercynian
Orogeny (Gondwanides), and its upper boundary is truncated by
an erosional unconformity, implying that it represents a distinct ex-
tensional episode that extended for a majority of the western South
American margin.
3. Extension along the Mitu Rift of Peru was synchronous with rifting
along the Pacific margin of Colombia and Ecuador, extension along
the Chilean margin and western Argentina north of ~40°S, and pos-
sibly continental rifting within Bolivia. Rifting north of the
Huancabamba Deflection was accompanied by subduction, occurred
within a backarc and ultimately led to seafloor spreading and the
separation of continental crust that subsequently formed parts of
the Mexican basement (Spikings et al., 2015). In contrast, rifting
along the Peruvian and Chilean margins mainly occurred in the ab-
sence of subduction and terminated prior to the formation of exten-
sive transitional crust, and these regions remain as aulacogens that
track relict Palaeozoic and Proterozoic terrane boundaries. The com-
bination of subduction north of the Huancabamba Deflection and a
paucity of subduction to the south supports sinistral displacement
of the eastern Pacific lithosphere relative to South America during
~ 245–209 Ma, implying that rifts in Peru, Chile and Argentina
opened by transtension above an arrested slab. The slab may have
detached, giving rise to intense crustal melting and contributing to
tensile stress (e.g. Spalletti et al., 2008).
4. Termination of rifting north and south of the Huancabamba Deflec-
tion at ~ 209 Ma corresponds with the onset of Pacific subduction,
implying a change to less oblique plate kinematics that initiated sub-
duction along almost the entire, newly established Andean margin.
This kinematic change may have been accompanied by a brief
phase of compression in southern Peru and northern Chile, prior to
the well-established prolonged period of Jurassic extension.
5. The relationship between Triassic plate margin extension and the
final fragmentation of Pangaea is unclear. However, the sequence
of events: plate margin rifting during ~245–220 Ma in the absence
of subduction, initiation of subduction and short-lived compression
at ~216–209 Ma, extrusion of a LIP at 201–198 Ma and the opening
of the proto-Caribbean Seaway and Central Atlantic at ~ 195 Ma
opens the possibility that plate tectonic forces initiated the early
break-up of Pangaea by attenuating its margins and enhancing
mantle upwelling. Prolonged extension may have propagated
along pre-existing weak zones that extended into the continental in-
terior (e.g. the South America–Yucatan–North America–Central
African sutures). These zones captured melts derived from the
upwelled mantle (e.g. Buiter and Torsvik, 2014) forming a LIP (e.g.
Central Atlantic Magmatic Province), became hot and weak and
eventually lead to the formation of a juvenile ocean. This assumes
that compression which accompanied the onset of subduction at
~216 Ma was a minor, second order effect.
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.gr.2016.02.008.
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
Acknowledgments
We thank Fernando Panca for his assistance in the field in the Sacred
Valley, INGEMMET for the logisitical support in Peru and members of the
geochemistry and geochronology groups of the univerisites of Geneva
and Lausanne for their analytical support. The manuscript was improved
by the careful reviews of Victor Ramos and an anonymous reviewer.
This project was funded by Swiss SNF projects 200020_116572 and
200020_124332.
Appendix A. Appendix
A.1. Methodology
A.1.1. Whole rock geochemistry
A.1.1.1. Major, trace and rare earth element analysis. Thirty four igneous
rocks were processed for geochemical analysis. Weathered zones were
removed prior to crushing and grinding using a jaw crusher, hydraulic
press and agate mill. Whole rock powders were transformed into
lithium tetraborate glass discs on which major oxides and trace ele-
ments were determined using an x-ray fluorescence Philips PW 2400
spectrometer. Certified reference materials BHVO (basalt) and SY-2
(syenite) were used for quality control.
Analyses of rare earth and additional trace elements (e.g. Th, U, Ta, Cs,
Hf) were performed with Laser Ablation ICP-MS with an Elan 6100 DRC
quadrupole mass spectrometer interfaced to a GeoLas 200M 193 nm
excimer ablation system. Spot analyses were done on glass discs previ-
ously used for major oxide determinations, using ablation parameters of
10 Hz, a 120 μm pit size and ~10 J/cm2 on sample density and helium as
a carrier gas. The acquisition times for the background and the ablation in-
terval were ~70 and 35 s, respectively. Dwell times per isotope ranged
from 10 to 20 ms and peak-hopping mode was employed. The ThO+/
Th+ and Ba2+/Ba+ ratios were optimized to 4.27 × 10−3 and 0.021, re-
spectively. The NIST SRM 610 synthetic glass standard was analysed for
external standardisation (Pearce et al., 1997). Raw data were reduced
offline using the LAMTRACE software (Jackson et al., 2004). CaO or Sr con-
centrations measured by XRF were used for internal standardization.
Three analyses per sample were acquired and the results averaged to ob-
tain the final concentrations of trace and rare earth elements. All mea-
surements were carried out at the Institute of Mineralogy and
Geochemistry, University of Lausanne, Switzerland.
A.1.1.2. Nd whole rock isotopic analyses. Approximately 150 mg of pow-
dered rock (b 70 μm) was dissolved overnight in 3 ml 2.2 M CH3COOH
to remove carbonate alteration. Subsequently, the samples were cleaned
by centrifuging and rinsing with ultrapure H2O. Samples were dissolved
in closed Teflon vials for 7 days on a hot plate at 140 °C with 4 ml of
conc. HF. The solution was dried on a hot plate and re-dissolved in 3 ml
of 15 M HNO3(aq) in closed Teflon vials at 140 °C, and dried down again.
Nd separation was carried out using cascade columns with TRU-spec
and Ln-spec resins following a modified method after Pin et al. (1994).
Nd isotope ratios were measured on a Thermo Scientific TRITON mass
spectrometer on Faraday cups in static mode. Sr was loaded on single
Re filaments with a Ta oxide solution and measured at a pyrometer-
controlled temperature of 1480 °C in static mode using a ‘virtual amplifier’
to cancel out biases in gain calibration among amplifiers. 87Sr/86Sr values
were internally corrected for fractionation using a 88Sr/86Sr value of
8.375209. Raw values were further corrected for external fractionation
by a value of 0.03‰, determined by repeated measurements of the
SRM987 standard (87Sr/86Sr = 0.710250). External reproducibility of
the 87Sr/86Sr ratio for the SRM987 standard is 7 ppm. Nd was loaded
onto double Re filaments with 1 M HNO3(aq) and measured in static
mode using virtual amplifier switching. 143Nd/144Nd values were inter-
nally corrected for fractionation using a 146Nd/144Nd value of 0.7219 and
the 144Sm interference on 144Nd was monitored on the mass 147 Sm
141
and corrected using a 144Sm/147Sm value of 0.206700. External re-
producibility of the JNdi-1 standard (Tanaka et al., 2000) was
b5 ppm. 87Sr/86Sr and 143Nd/144Nd whole rock compositions were
corrected for time-integrated decay of 87Rb and 147Sm using Rb, Sr,
Sm and Nd concentrations determined by LA-ICP-MS on the glass
whole rock discs and ages obtained in this study.
A.1.2. Zircon U–Pb geochronology
A.1.2.1. Laser ablation ICP-MS dating. Zircons were handpicked with a
preference for large, euhedral grains for volcanic rocks and completely
randomly for detrital samples. Zircons were mounted in epoxy and
polished with diamond paste to expose an internal surface. A total of
741 zircons extracted from 1 volcanic and 7 sedimentary rocks were
imaged by cathodoluminescence (CL). Isotopic measurements were
acquired with an Element XR sector field single-collector ICP-MS
interfaced to a NewWave UP-193 nm excimer ablation system at the
University of Lausanne, Switzerland. Operating conditions were similar
to those described in Ulianov et al. (2012). Twenty-three spots were
analysed for the volcanic rocks and 47 to 109 spots per detrital sample.
Discordant ages, inherited zircons and zircons showing signs of Pb-
loss were discarded from the calculation of weighted mean ages. A
weighted mean age was calculated using the 206Pb/238U ratio of
analytically concordant zircons.
A.1.2.2. CA-ID-TIMS. Nine zircons from trachy-andesite MR99 (Mitu
Group) were dated by CA-ID-TIMS. The methodology described in
Schoene et al. (2010) was followed for zircon annealing, leaching, disso-
lution and chemical separation of U and Pb, isotope analysis and data
treatment. Zircons were spiked with the Earthtime 205Pb–233U–235U
tracer solution. Isotopic analyses were performed at the University of
Geneva on a TRITON mass spectrometer equipped with a MasCom
electron multiplier in ion counting mode.
Due to the high precision of CA-ID-TIMS analyses, only the youngest
206
Pb/238U date or the weighted mean of the youngest data cluster was
used (Schaltegger et al., 2009; Schoene et al., 2010).
A.1.3. Lu–Hf isotope analyses
A.1.3.1. LA-MC-ICP-MS. Lu–Hf isotopic analyses on concordant, dated zir-
cons were performed with a Thermo-Scientific Neptune multi-collector
ICP-MS at JWG University, Frankfurt with a New Wave Research UP-213
laser and teardrop-shaped, low-volume ablation cell (see Gerdes and
Zeh, 2006, 2009) with helium as a carrier gas. The Lu–Hf laser spot
was drilled close to or partially overlapping the U–Pb laser spot. The
laser beam parameters were 40–50 μm spot size, and 5 Hz firing repeti-
tion rate. To correct for isobaric interferences of Lu and Yb on mass 176
the isotopes 172Yb, 173Yb and 175Lu were simultaneously monitored. The
176
Yb and 176Lu signals were calculated using a 176Yb/173Yb of 0.796218
(Chu et al., 2002) and 176Lu/175Lu of 0.02658 (JWG in-house value). The
instrumental mass bias for Hf isotopes was corrected using an exponen-
tial law and a 179Hf/177Hf value of 0.7325. In the case of Yb isotopes, the
mass bias was corrected using the Hf mass bias of the individual integra-
tion step multiplied by a daily βHf/βYb offset factor (Sláma et al., 2008;
Gerdes and Zeh, 2009). All zircon LA-MC-ICP-MS analyses were adjusted
relative to the JMC 475 176Hf/177Hf ratio of 0.282160 and the reported un-
certainties (2σ) were propagated by quadratic addition of the external
reproducibility of GJ-1 (2σ, n = 18/30) and Temora (2σ, n = 12) or
Plesovice (2σ, n = 20) and the within-run precision of each analysis (2
SE). The external reproducibility (2σ, n N 50) over more than 6 months
of reference zircon 91500, GJ-1, and Plešovice (176Hf/177Hf =
0.282298 ± 0.000026, 0.282003 ± 0.000018, and 0.282482 ±
0.000015, respectively) at JWG is about 0.005–0.009% (b 1σ).
A.1.3.2. Hf isotope analyses on solutions of ID-TIMS samples. Hf isotopic
compositions were measured on the same volume of zircon dated by
142 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
CA-ID-TIMS techniques. This involves retaining the waste eluted solu-
tion obtained from ion exchange chemistry. The dried residue was dis-
solved in 500 to 700 μl 2% HNO3(aq) and introduced into the plasma
with an Aridus II desolvating nebulizer, and analysed for Hf isotope
composition using the same multi-collector ICP-MS techniques as de-
scribed above for laser ablation analyses. Data acquisition of the solution
analyses lasted for 5 min per sample (55 ratios) and was performed
in automatic mode, bracketing groups of 5 unknowns by JMC-475
standard measurements. Six solution aliquots of standard zircon GJ-1
were analysed as an external control (0.282016 ± 0.000002, 2σ).
References
Barredo, S., Chemale, F., Marsicano, C., Ávila, J.N., Ottone, E.G., Ramos, V.A., 2012. Tectono-
sequence stratigraphy and U–Pb zircon ages of the Rincón Blanco Depocenter, north-
ern Cuyo Rift, Argentina. Gondwana Research 21, 624–636.
Bertrand, H., Fornari, M., Marzoli, A., García-Duarte, R., Sempere, T., 2014. The Central
Magmatic Province extends into Bolivia. Lithos 188, 33–43.
Beutel, E.K., 2009. Magmatic rifting of Pangaea linked to onset of South America plate
motion. Tectonophysics 468, 149–157.
Blackburn, T.J., Olsen, P.E., Bowring, S.A., McLeann, N.M., Kent, D.V., Puffer, J., McHone, G.,
Rasbury, E.T., Et-Touhami, M., 2013. Zircon U–Pb geochronology Lins the End-Triassic
extinction with the Central Atlantic Magmatic Province. Science 24, 941–945.
Boekhout, F., Spikings, R., Sempere, T., Chiaradia, M., Ulianov, A., Schaltegger, U., 2012. Me-
sozoic arc magmatism along the south Peruvian margin during Gondwana breakup
and dispersal. Lithos 146, 48–64.
Boekhout, F., Sempere, T., Spikings, R., Schaltegger, U., 2013. Late Paleozoic to Jurassic
chronostratigraphy of coastal southern Peru: temporal evolution of sedimentation
along an active margin. Journal of South American Earth Sciences 47, 179–200.
Boynton, W.V., 1984. Geochemistry of the rare earth elements: meteorite studies. In:
Henderson, P. (Ed.), Rare Earth Element Geochemistry. Elsevier, pp. 63–114.
Buiter, S.J.H., Torsvik, T.H., 2014. A review of Wilson Cycle plate margins: A role for mantle
plumes in continental break-up along sutures? Gondwana Research 26, 627–653.
Carlotto, V., 1998. Evolution Andine et Raccourcissement au niveau de Cusco (13-16S),
Perou. Enregistrement sédimentaire, chronologie, contrôles paléogepgraphiques,
évolution cinématique PhD thesis Université Joseph Fourier, Grenoble.
Casquet, C., Herve, F., Pankhurst, R.J., Baldo, E., Calderon, M., Fanning, C.M., Rapela, C.W.,
Dahlquist, J., 2014. The Mejillonia suspect Terrane (Northern Chile): Late Triassic
fast burial and metamorphism of sediments in a magmatic arc environment extend-
ing into the Early Jurassic. Gondwana Research 25, 1272–1286.
Cenki, B., Jaillard, E., Carlotto, V., 2000. Estudio petrografico-geoquimico del volcanismo
pre-Huancané de la region de Cusco-Sicuani (Sur del Peru): Interpretacion
geodinamica. Boletin de la Sociedad Geologica del Peru 89, 45–56.
Charrier, R., 1973. Interruptions of spreading and the compressive tectonic phases of the
meridonal andes. Earth and Planetary Letters 20, 242–249.
Charrier, R., Pinto, L., Rodriguez, M.P., 2007. Tectonostratgraphic evolution of the Andean
Orogen in Chile. In: Moreno, T., Gibbons, W. (Eds.), The Geology of Chile. The Geolog-
ical Society, Kondon, pp. 21–114.
Chu, N.C., Chavagnac, V., Nesbitt, R.W., Boella, R.M., Milton, J.A., German, C.R., Bayon, G.,
Burton, K., 2002. Hf isotope ratio analyses using multi-collector inductively coupled
plasma mass spectrometry: an evaluation of isobaric interference corrections. Journal
of Analytical Atomic Spectrometry 17, 1567–1574.
Clift, P.D., Hartley, A.J., 2007. Slow rates of subduction erosion and coastal underplating
along the Andean margin of Chile and Peru. Geology 35, 503–506.
Clift, P., Pecher, I., Kukowski, N., Hampel, A., 2003. Tectonic erosion of the Peruvian
forearc, Lima Basin, by subduction and Nazca Ridge collision. Tectonics 22, 1026.
Cochrane, R., Spikings, R., Gerdes, A., Ulianov, A., Mora, A., Villagómez, D., Putlitz, B.,
Chiaradia, M., 2014a. Permo-Triassic anatexis, continental rifting and the disassembly
of western Pangaea. Lithos 190-191, 383–402.
Dalmayrac, B., Laubacher, G., Marocco, R., 1980. Caractères généraux de l’evolution
géologique des Andes Péruviennes. Géologie des Andes Péruviennes 122. PRSTOM,
Paris (501 pp.).
De La Roche, H., Letterier, J., Grandclaude, P., Marchal, M., 1980. A classification of volcanic
and plutonic rocks using R1R2-diagram and major-element analyses — its relation-
ships with current nomenclature. Chemical Geology 29, 183–210.
Demouy, S., Paquette, J.-L., Blanquat, M. de S., Benoit, M., Belousova, E.A., O’Reilly, S.Y.,
García, F., Tejada, L.C., Gallegos, R., Sempere, T., 2012. Spatial and temporal evolution
of Liassic to Paleocene arc activity in southern Peru unravelled by zircon U–Pb and Hf
in-situ data on plutonic rocks. Lithos 155, 183–200.
Doubinger, J., Marocco, R., 1981. Contenu Palynologique du Groupe Copacabana (Permien
Inférieur et Moyan) sur la bordure Sud de la Cordillère de Vilcabamba, région de
Cuzco (Pérou). International Journal of Earth Sciences 70, 1086–1099.
Farrar, E., Clark, A.H., Heinrich, S.M., 1990. The age of the Zongo pluton and the
tectonothermal evolution of the Zongo–San Gaban Zone in the Cordillera Real, Boliv-
ia. I International Symposium on Andean Geodynamics, Grenoble, pp. 171–174.
Franzese, J.R., Spalletti, L.A., 2001. Late Triassic–early Jurassic continental extension in
southwestern Gondwana: a tectonic segmentation and pre-break-up rifting. Journal
of South American Earth Sciences 14, 257–270.
Gerdes, A., Zeh, A., 2006. Combined U–Pb and Hf isotope LA-(MC-)ICP-MS analyses of de-
trital zircons: comparison with SHRIMP and new constraints for the provenance and
age of an Armorican metasediment in central Germany. Earth and Planetary Science
Letters 249, 47–61.
Gerdes, A., Zeh, A., 2009. Zircon formation versus zircon alteration — new insights from
combined U–Pb and Lu–Hf in-situ LA-ICP-MS analyses, and consequences for the in-
terpretation of Archean zircon from the Central Zone of the Limpopo Belt. Chemical
Geology 261, 230–243.
Glodny, J., Lohrmann, J., Echtler, H., Gräfe, K., Seifert, W., Collao, S., Figueroa, O., 2005. In-
ternal dynamics of a paleoaccretionary wedge: insights from combined isotope
tectonochronology and sandbox modelling of the South-Central Chilean forearc.
Earth and Planetary Science Letters 231, 23–39.
Gradstein, F.M., et al., 2012. The Geologic Time Scale 2012. Elsevier, Boston, USA, p. 1129.
Guex, J., Schoene, B., Bartolini, A., Spangenberg, J., Schaltegger, U., O’Dogherty, L., Taylor,
D., Bucher, H., Atudorei, V., 2012. Geochronological constraints on post-extinction re-
covery of the ammonoids and carbon cycle perturbations during Early Jurassic.
Palaeogeography, Palaeoclimatology, Palaeoecology 346, 1–11.
Helbig, M., Keppie, J.D., Murphy, J.B., Solari, L.A., 2012. U–Pb geochronological constraints
on the Triassic–Jurassic Ayu Complex, southern Mexico: derivation from the western
margin of Pangea-A. Gondwana Research 22, 910–927.
Helwig, J., 1972. Stratigraphy, sedimentation, paleogeography and paleoclimates of
Carboniferous (‘Gondwana’) and Permian of Bolivia. American Association of Petro-
leum Geology 56, 1008–1033.
Herrmann, A.G., 1970. Yttrium and Lanthanides. In: Wedepohl, K.H. (Ed.), Handbook of
Geochemistry II-5. Springer-Verlag, Berlin, pp. 39–71.
Ianuzzi, R., Diaz-Martinez, E., Carlotto, V., 2000. Preencia de licofitas fosiles en el Permico
del borde sur de la Cordillera Vilcabamba, deflexion de Abancay, Peru. X Congreso
Peruano de Geologia, Lima, p. 201.
Jackson, S.E., Pearson, N.J., Griffin, W.L., Belousova, E.A., 2004. The application of laser
ablation-inductively coupled plasma-mass spectrometry to in situ U–Pn zircon
geochronology. Chemical Geology 211, 47–69.
Jaillard, E., Soler, P., Carlier, G., Mourier, T., 1990. Geodynamic evolution of the northern
and central Andes during early to middle Mesozoic times: a Tethyan model. Journal
of the Geological Society 147, 1009–1022.
Jourdan, F., Marzoli, A., Bertrand, H., Cirilli, S., Tanner, L.H., Kontak, D.J., McHone, G., Renne,
P.R., Bellieni, G., 2009. 40Ar/39Ar ages of CAMP in North America: implications for the
Triassic–Jurassic boundary and the 40K decay constant bias. Lithos 110, 167–180.
Kay, S.M., Ramos, V.A., Mpodozis, C., Sruoga, P., 1989. Late Paleozoic to Jurassic silicic
magmatism at the Gondwana margin: analogy to the Middle Proterozoic in North
America? Geology 17, 324–328.
Keppie, J.D., Dostal, J., Miller, B.V., Ortega-Rivera, A., Roldan-Quintana, J., Lee, J.W.K., 2006.
Geochronology and geochemistry of the Francisco Gneiss: Triassic continental rift
tholeiites on the Mexican Margin of Pangea metamorphosed and exhumed in a
tertiary core complex. International Geology Review 48, 1–16.
Kontak, D.J., Clark, A.H., Farrar, E., Archibald, D.A., Baadsgaard, H., 1990. Late Paleozoic–early
Mesozoic magmatism in the Cordillera de Carabaya, Puno, southeastern Peru: Geochro-
nology and Geochemistry. Journal of South American Earth Sciences 3, 213–230.
Kontak, D.J., Clark, A.H., Farrar, E., Strong, D.F., 1985. The rift associated Permo-Triassic
magmatism of the Eastern Cordillera: a precursor to the Andean orogeny. In: Pitcher,
W.S., Atherton, M.P., Cobbing, J., Beckinsale, R.D. (Eds.), Magmatism at a plate edge:
The Peruvian Andes. Blackie, Glasgow, and Halsted Press, New York, pp. 36–44.
Laubacher, G., 1978. Géologie des Andes Peruvienne. Géologie de la Cordillère Orientale et
de l’Altiplano au nord et nord-ouest du lac Titicaca, Pérou. ORSTOM, Paris (217 pp.).
Leon, W., Palacios, O., Vargas, L., Sanchez, A., 2000. Memoria explicativa del Mapa
Geologico del Peru (1999). Lima, Instituto Geológico Minero y Metalúrgico, Carta
Geológica Nacaional, Boletín 136, escala 1:1,000,000.
Litherland, M., Aspden, J.A., Jemielita, R.A., 1994. The metamorphic belts of Ecuador.
Overseas Memoir of the British Geological Survey 11. British Geological Survey,
Nottingham, England, p. 147.
Llambias, E.J., Quenardelle, S., Montenegro, T., 2003. The Choiyoi Group from central
Argentina: a subalkaline transitional to alkaline association in the craton adjacent
to the active margin of the Gondwana continent. Journal of South American Earth Sci-
ences 16, 243–257.
Maksaev, V., Munizaga, F., Tassinari, C., 2014. Timing of magmatism of the paleo-Pacific bor-
der of Gondwana: U–Pb geochronology of Late Paleozoic to Early Mesozoic igneous rocks
of the north Chilean Andes between 20° and 31°S. Andean Geology 41, 447–506.
Mancuso, A.C., Caselli, A.T., 2012. Paleolimnology evolution in rift basins: the
Ischigualasto-Villa Unión Basin (Central-Western Argentina) during the Triassic. Sed-
imentary Geology 275, 38–54.
Maniar, P.D., Piccoli, P.M., 1989. Tectonic discrimination of granitoids. Geological Society
of America Bulletin 101, 635–643.
Marocco, R., 1978. Géologie des Andes Péruviennes. ORSTOM, Paris (195 pp.).
Marsicano, C., Irmis, R.B., Mancuso, A.C., Mundil, R., Chemale, F., 2016. The precise tempo-
ral calibration of dinosaur origins. Proceedings of the National Academy of Sciences
113, 509–513.
Martin, M.W., Clavero, J.C., Mpodozis, C.M., 1999. Late Paleozoic to Early Jurassic tectonic
development of the high Andean Principal Cordillera, El Indio Region, Chile (29–
30°S). Journal of South American Earth Sciences 12, 33–49.
Martínez, A.M.C., 2007. Petrogenesis and evolution of Aburra Ophiolite, Colombian Andes,
Central Range Ph.D. Thesis University of Brasilia, p. 178.
Martinez, R.N., Sereno, P.C., Alcober, O.A., Colombi, C.E., 2011. A basal dinosaur from the
dawn of the dinosaur era in southwestern Pangaea. Science 331, 206–210.
Mathalone, J.M.P., Montoya, M., 1995. Petroleum geology of the Sub-Andean basins of
Peru. In: Tankard, A., Suarez, R., Welsink, H.J. (Eds.), Petroleum basins of South
America 62. American Association of Petroleum Geology Memoir, pp. 423–444.
Mégard, F., 1978. Etude géologique des Andes du Pérou central. ORSTOM, Paris (310 pp.).
Mišković, A., Spikings, R.A., Chew, D.M., Košler, J., Ulianov, A., Schaltegger, U., 2009.
Tectonomagmatic evolution of Western Amazonia: geochemical characterization
and zircon U–Pb geochronologic constraints from the Peruvian Eastern Cordilleran
granitoids. Geological Society of America Bulletin 121, 1298–1324.
 R. Spikings et al. / Gondwana Research 35 (2016) 124–143
Morel, E.M., Artabe, A.E., Spalletti, L.A., 2003. Triassic floras of Argentina: biostratigraphy,
floristic events and comparison with other areas of Gondwana and Laurasia.
Alcheringa, An Australasian Journal of Palaeontology 27, 231–243.
Mpodozis, C., Kay, S., 1992. Late Paleozoic to Triassic evolution of the Gondwana margin:
evidence from Chilean Frontal Cordilleran batholiths (28°S to 31°S). Geological Soci-
ety of America Bulletin 104, 999–1014.
Mpodozis, C., Ramos, V., 1989. The Andes of Chile and Argentina. In: Ericksen, G.E., Canas
Pinochet, M.T., Reinemund, J.A. (Eds.), Geology of the Andes and its Relation to
Hydrocarbon and Mineral Resources. Circum-Pacific Council for Energy and Mineral
Resources, Earth Science Series 11, pp. 59–90.
Noble, D.C., Silberman, M.L., Mégard, F., Bowman, H.R., 1978. Comendite (peralkaline rhy-
olites) in the Mitu Group, central Peru: evidence of Permian–Triassic extension in the
Central Andes. United States Geological Survey. Journal of Research 6, 453–457.
Panca, F., 2010. The late Permian–Lower Triassic Mitu Group, SE of Cusco (Peru). Field Re-
lations, Volcanosedimentary Facies and Geochemistry. Technische Universität
Bergakademie, Freiberg MSc thesis.
Parada, M., et al., 2007. Andean Magmatism. In: Moreno, T., Gibbons, W. (Eds.), The
Geology of Chile. The Geological Society, Kondon, pp. 115–146.
Peacock, M.A., 1931. Classification of igneous rock series. The Journal of Geology 39, 54–67.
Pearce, N.J.G., Perkins, W.T., Westgate, J.A., Gorton, M.P., Jackson, S.E., Neal, C.R., Chenery, S.P.,
1997. A compilation of new and published major and trace element data for NIST SRM
610 and NIST SRM 612 glass reference materials. Geostandards Newsletter 21, 115–144.
Pin, C., Briot, D., Bassin, C., Poitrasson, F., 1994. Concomitant separation of strontium and
samarium–neodynium for isotopic analyses in silicate samples, based on specific
extraction chromatography. Analytica Chimica Acta 298, 209–217.
Pindell, J., Dewey, J.F., 1982. Permo-Triassic reconstruction of western Pangaea and the
evolution of the Gulf of Mexico and south Caribbean region. Tectonics 1, 179–211.
Pindell, J.L., Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and
northern South America in the mantle reference frame: an update. Geological Society
of London, Special Publication 328, 1–55.
Pindell, J.L., Tabbutt, K.D., 1995. Mesozoic–Cenozoic Andean Paleogeography and Regional
Controls on Hydrocarbon Systems. In: Tankard, A.J., Suarez, R., Welsink, H.J. (Eds.), Pe-
troleum Basins of South America. American Association of Petroleum Geologists
Memoir 62, pp. 101–128.
Ramos, V.A., 1994. Terranes of Southern Gondwanaland and Their Control in the Andean
Structure (30°–33°S Latitude). In: Reutter, K.-J., Scheuber, E., Wigger, P.J. (Eds.),
Tectonics of the Southern Central Andes. Springer-Verlag, Heidelberg, pp. 249–261.
Ramos, V.A., Folguera, A., 2009. Andean flat-slab subduction through time. In: Murphy,
J.B., Keppie, J.D., Hynes, A.J. (Eds.), Ancient Orogens and Modern Analogues. Geologi-
cal Society of London Special Publications 327, pp. 31–54.
Ramos, V.A., Kay, S.M., 1991. Triassic rifting and associated basalts in the Cuyo basin,
central Argentina. Geological Society of America Special Papers 265, 79–92.
Rapela, C.W., Pankhurst, R.J., Llambías, E.J., Labudía, C.H., Artabe, A., 1996. Gondwana
Magmatism of Patagonia: Inner Cordilleran Calc-alkaline Batholiths and Bimodal
Volcanic Province. International Symposium Andean Geodynamics, Saint Malo,
France, pp. 791–994.
Reitsma, M.J., 2012. Reconstructing the Late Paleozoic–Early Mesozoic Plutonic and
Sedimentary Record of South-East Peru: Orphaned Back-arcs Along the Western
Margin of Gondwana PhD thesis Terre & Environment 111. University of Geneva,
Switzerland, p. 226.
Rodriguez, R., Cueva, E., 2010. Fallas Permo-Triásicas en la región de Chachapoyas,
contoles para exploración de yacimientos de minerales e hidrocarburos. XV Congreso
Peruano de Geología, Cusco, peru, pp. 805–808.
Rodriguez, R., Cueva, E., Carlotto, V., 2011. Geología del cuadrángulo de Cerro de Pasco
(22-k). Carta Géologica Nacional 144. INGEMMET (164 pp.).
Rogers, R.R., Swisher III, C.C., Sereno, P.C., Monetta, A.M., Forster, C.A., Martinez, R.N., 1993.
The Ischigualasto Tertrapod Assemblage (Late Triassic, Argentina) and 40Ar/39Ar
dating of dinosaur origins. Science 260, 794–797.
Rosas, S., Fontboté, L., Tankard, A., 2007. Tectonic evolution and paleogeography of the Me-
sozoic Pucará Basin, central Peru. Journal of South American Earth Sciences 24, 1–24.
Schaltegger, U., Guex, J., Bartolini, A., Schoene, B., Ovtcharova, M., 2008. Precise U–Pb con-
straints for end-Triassic mass extinction, its correlation to volcanism and Hettangian
post-extension recovery. Earth and Planetary Science Letters 267, 266–275.
Schaltegger, U., Brack, P., Ovtcharova, M., Peytcheva, I., Schoene, B., Stracke, A., Marocchi,
M., Bargossi, G., 2009. Zircon and titanite recording 1.5 million years of magma accre-
tion, crystallisation and initial cooling in a composite pulton (southern Adamello
Batholith, northern Italy). Earth and Planetary Science Letters 286, 208–218.
Schoene, B., Guex, J., Bartolini, A., Schaltegger, U., Blackburn, T.J., 2010. Correlating the
end-Triassic mass extinction and flood basalt volcanism at the 100 Ka level. Geology
38, 387–390.
View publication stats
143
Sempere, T., 1995. Phanerozoic Evolution of Bolivia and Adjacent Regions. In: Tankard, A.,
Suarez, R., Welsink, H.J. (Eds.), Petroleum Basins of South America 62. American
Association of Petroleum Geology Memoir, pp. 423–444.
Sempere, T., Carlier, G., Soler, P., Fornari, M., Carlotto, V., Jacay, J., Arispe, O., Néraudeau, D.,
Cárdenas, J., Rosas, S., Jiménez, N., 2002. Late Permian–Middle Jurassic lithospheric
thinning in Peru and Bolivia, and its bearing on Andean-age tectonics. Tectonophysics
345, 153–181.
Senff, M., 1995. Sedimentologie, Fauna und Fazies des präkretazischen Mesozoikums im
Oberen Magdalenatal von Zentralkolumbien unter besonderer Beruecksichtigung
der obertriassichen Payande Formation PhD thesis Geolische Institut, Univeristät
Guessen, Germany (114 p.).
Shervais, J.W., 1982. Ti-V plots and the petrogenesis of modern and ophiolitic lavas. Earth
and Planetary Science Letters 59, 101–118.
Sláma, J., Košler, J., et al., 2008. Plešovice zircon — a new natural reference material for
U–Pb and Hf isotopic microanalysis. Chemical Geology 249, 1–35.
Soler, P., Sempere, T., 1993. Stratigraphie, géochimie et signification paléotectonique des
roches volcaniques basiques mésozoiques des Andes boliviennes. Comptes Rendus
de l’Académie des Sciences de Paris 316, 777–784.
Spalletti, L.A., Fanning, C.M., Rapela, C.W., 2008. Dating the Triassic continental rift in the
southern Andes: the Potrerillos Formation, Cuyo Basin, Argentina. Geologica Acta 6,
267–283.
Spikings, R., Cochrane, R., Villagomez, D., Van der Lelij, R., Vallejo, C., Winkler, W., Beate, B.,
2015. The geological history of northwestern South America: from Pangaea to the
early collision of the Caribbean Large Igneous Province (290–75 Ma). Gondwana Re-
search 27, 95–139.
Stern, C.R., 2011. Subduction erosion: rates, mechanismsand its role in arc magmatism and
the evolution of the continental crust and mantle. Gondwana Research 20, 284–308.
Stewart, J.W., Evernden, J.F., Snelling, N.J., 1974. Age determinations from Andean Peru: a
reconnaissance survey. Geological Society of America Bulletin 85, 1107–1116.
Sun, S.S., McDonough, W.F., 1989. Chemical and isotopic systematics of oceanic basalts:
implications for mantle composition and processes. In: Saunders, A.D., Norry, M.J.
(Eds.), Magmatism in Ocean Basins. Geologogical Society of London Special Publica-
tion 42, pp. 313–345.
Tanaka, T., Togashi, S., et al., 2000. Jndi-1: a neodynium isotopic reference in consistency
with La Jolla neodynium. Chemical Geology 168, 279–281.
Taylor, S.R., McLennan, S.M., 1995. The geochemical evolution of the continental crust. Re-
views of Geophysics 33, 241–265.
Ulianov, A., Muntener, O., Schaltegger, U., Bussy, F., 2012. The data treatment dependant
variability of U–Pb zircon ages obtained using mono-collector, sector field, laser
ablation ICP-MS. Journal of Analytical Atomic Spectrometry 27, 663–676.
Van der Lelij, R., Spikings, R., Ulianov, A., Chiariadia, M., Mora, A., 2015. Palaeozoic to Early
Jurassic history of the northwestern corner of Gondwana, and implications for the
evolution of the Iapetus, Rheic and Pacific Oceans. Gondwana Research http://dx.
doi.org/10.1016/j.gr.2015.01.011.
Vasquez, P., Glodny, J., Franz, G., Frei, D., Romer, R.L., 2011. Early Mesozoic plutonism of
the Cordillera de la Costa (34-37S), Chile. Constraints on the onset of the Andean
orogeny. Journal of Geology 119, 159–184.
Vilas, J.F., Valencio, D.A., 1978. Palaeomagnetism and K—Ar age of the Upper Ordovician
Alcaparrosa Formation, Argentina. Geophysical Journal International 55, 143–154.
Vivier, F., Audebaud, E., Vatin-Perignon, N., 1976. Le magmatisme tardi-hercynien et
andin le long d’une transversale sud-Péruvienne: bilan géochimique des éléments in-
compatible. 4ème Réunion des Science de la Terre, Paris, p. 386.
Winchester, J.A., Floyd, P.A., 1977. Geochemical discrimination of different magma series and
their differentiation products using immobile elements. Chemical Geology 20, 325–343.
Zerfass, H., Chemale, F., Schultz, C.L., Lavina, E., 2004. Tectonics and sedimentation in
Southern South America during Triassic. Sedimentary Geology 166, 265–292.
Further reading
Cochrane, R., Spikings, R., Gerdes, A., Winkler, W., Ulianov, A., Mora, A., Chiradia, M.,
2014b. Distinguishing between in-situ and accretionary growth of continents along
active margins. Lithos 202, 382–394.