Lithos 160161 (2013) 228249

Contents lists available at SciVerse ScienceDirect

Lithos
journal homepage: www.elsevier.com/locate/lithos

Thermochronology and tectonics of the Central and Western Cordilleras of Colombia:

Early CretaceousTertiary evolution of the Northern Andes
Diego Villagmez 1, Richard Spikings
Section of Earth and Environmental Sciences, University of Geneva, 13 Rue des Marachers, 1205 Geneva, Switzerland

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

Article history: New thermochronological data record a complex cooling history in the Central and Western Cordilleras of
Received 22 September 2012 Colombia that is a function of Early Cretaceous to late Miocene tectonic events. Alkali-feldspar 40Ar/39Ar cooling
Accepted 15 December 2012 ages of ~138130 Ma immediately post-date the cessation of Jurassic arc-magmatism and a major unconformity
Available online 24 December 2012
within the retro-foreland region of the northern Andes. We interpret these ages as cooling driven by exhumation
in response to either compression driven by subduction of a seamount, or extension and oceanward migration of
Keywords:
Thermochronology
the slab during the earliest Cretaceous, giving rise to the Lower Cretaceous Quebradagrande arc sequence. Biotite
Colombia and alkali-feldspar 40Ar/39Ar data from the palaeocontinental margin reveal the presence of a younger cooling
Northern Andes event at 117107 Ma, which was contemporaneous with hornblende 40Ar/39Ar cooling ages obtained from
Plate tectonics mediumhigh PT metamorphic relicts of a Late JurassicEarly Cretaceous subduction channel. This cooling
Geodynamics event is attributed to exhumation driven by the collision and accretion of a fringing arc against the continental
margin, and obduction of the subduction channel onto the forearc. Inverse modelling of zircon and apatite ssion
track and (UTh)/He data from throughout the Central and Western Cordilleras reveals three periods of rapid
cooling since the Late Cretaceous. The earliest phase is recorded by Jurassic and Cretaceous granitoids that cooled
rapidly during 7565 Ma. We attribute cooling to exhumation of the continental margin during ~7570 Ma
(~1.6 km/My), which was forced by the collision and accretion of the Caribbean Large Igneous Province in the
Campanian. The Central Cordillera exhumed at moderate rates of ~0.3 km/My during ~4530 Ma, which are
also observed over widely dispersed regions along the Andean chain, and were probably caused by an increase
in continentocean plate convergence rates. Exhumation rates drastically increased in the middlelate Miocene,
with the greatest amount occurring in southern Colombia as a consequence of the collision and subduction of the
buoyant Carnegie Ridge at 15 Ma.
2013 Elsevier B.V. All rights reserved.

1. Introduction The Central Cordillera of Colombia consists of Triassic and older

The Cretaceous and Tertiary subduction history of the northern
Andes is unique within the Andean chain because it was interrupted
by the collision and accretion of arc and oceanic plateau rocks, which
are not recognised south of 5S. This study aims to quantify i) the timing
of accretion of arc and plateau rocks, and ii) the thermal and
exhumational response of the buttressing continental and indenting
oceanic rocks to collision, accretion and post-accretion subduction in
Colombia. The conclusions of the study constrain the evolution of
the post-rift Western Tethys Wilson Cycle and the early history of
the Caribbean Plate, and provide an example of how active continental
margins respond to changing plate kinematics and collision, subduction
and accretion of heterogeneous oceanic crust.
Corresponding author. Tel.: +41 223793176.
E-mail address: richard.spikings@unige.ch (R. Spikings).
1
Current address: Tectonic Analysis Inc. 14 rue Ferrier, 1202 Geneva, Switzerland.
metamorphic rocks that were partially melted during Triassic rifting
of Central American basement units away from northwestern South
America (Villagmez et al., 2011a). Jurassic subduction beneath the
Colombian margin formed a continental arc, and an Early Cretaceous
fringing arc erupted through either continental (Nivia et al., 2006) or
oceanic basement, with accompanying sedimentation within a margin-
al basin (e.g. Pindell and Kennan, 2009), culminating in compression
and basin closure. Subsequent subduction was interrupted by the colli-
sion of the Late Cretaceous, Caribbean Large Igneous Province (oceanic
plateau and its overlying arc) with the northern Andes during the Late
CretaceousEarly Tertiary (Kerr et al., 1997; Spikings et al., 2010;
Vallejo et al., 2006). The oceancontinent suture zone within the
northern Andes is well exposed in Colombia along the western ank
of the Central Cordillera. The ChocPanam block collided with
the Colombian margin during either 2523 Ma (Farris et al., 2011), at
~13 Ma (Duque-Caro, 1990) or during the Late MiocenePliocene
(Mann and Corrigan, 1990). Finally, the aseismic Carnegie Ridge currently
subducts beneath southern Colombia and northern Ecuador, although the

0024-4937/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.lithos.2012.12.008
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 229

proposed timing of collision varies between 15 Ma (Spikings et al.,
2001) and 1 Ma (Lonsdale and Klitgord, 1978).
We combine hornblende, alkali feldspar and biotite 40Ar/ 39Ar, zircon
and apatite ssion track (FT) and zircon and apatite (UTh)/He data to
generate semi-continuous Cretaceous and Tertiary thermal history
paths through ~550 C40 C for crystalline and sedimentary rocks
exposed along the Central Cordillera and accreted rocks exposed in
the Western Cordillera. These paths have been used to construct exhu-
mation histories since ~140 Ma, which have been combined with i) the
sedimentological histories of surrounding basins, ii) UPb (zircon) ages
of plutonic and metasedimentary rocks (Villagmez et al., 2011a), and
iii) geochemical analyses of intrusive and volcanic rocks (Kerr et al.,
1997; Villagmez et al., 2011a) to precisely determine the timing of
collision events, quantify the exhumational response and constrain
the post-accretionary history of the Colombian margin. Improving our
understanding of these processes is useful because they resulted in
net crustal growth of the South American Plate and can be used to
improve reconstructions of the southern margin of the Caribbean
Plate, and hence the entire Caribbean region.
2. Geological framework of the Central and Western Cordilleras
of Colombia
Previous work in Colombia divided the region into a continental
and an oceanic province, which are juxtaposed across three major

Fig. 1. Digital elevation model of northwestern South America and surrounding tectonic plates showing the main cordilleras, faults and the subducting Carnegie Ridge (background model
from Gmez et al., 2007). Cretaceous sutures are shown as thick black and yellow lines, and the three sample regions (a, b and c) are highlighted (Fig. 2). Major rock sequences of the
Central Cordillera (Colombia) and Eastern Cordillera (Ecuador) are shown. AzBF: Amazon Border Fault, CAF: CaucaAlmaguer Fault, CC: Central Cordillera, CPV: CaucaPata Valley,
EC: Eastern Cordillera, ECE: Eastern Cordillera Ecuador, GF: Garrapatas Fault, IF: Ibagu Fault; LB: Llanos Basin, MMV: Middle Magdalena Valley Basin, OB: Oriente Basin, OPF: OtPericos
Fault, PE: Peltetec Unit, PF: Palestina Fault, RC: Raspas Complex, SJF: San-Jeronimo Fault, SMF: Santa MartaBucaramanga Fault, SNSM: Sierra Nevada de Santa Marta, SZ: Sub-Andean Zone
(Ecuador), UMV: Upper Magdalena Valley Basin, WC: Western Cordillera.
230 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

fault branches, which are the San Jernimo, SilviaPijao and Cauca
Almaguer Faults (Figs. 1 and 2).
Continental crust of the Central Cordillera is exposed to the east of
the San Jernimo Fault (Fig. 1), and consists of Paleozoic gneisses that
are in unconformable contact with metamorphic rocks of the Triassic
Cajamarca Complex (Villagmez et al., 2011a). Widely dispersed
Permo-Triassic granitoids formed during Permian arc magmatism
along the western Pangea, and anatexis during its Triassic fragmenta-
tion (Cardona et al., 2010; Villagmez et al., 2011a). These lithologies
are intruded by Jurassic calc-alkaline (continental arc) granitoids of
the Ibagu Batholith (UPb zircon ages of 165170 Ma, Villagmez
et al., 2011a), and Late Cretaceous intrusive rocks of the Antioquia
Batholith (UPb zircon ages of 9584 Ma, Villagmez et al., 2011a).
The San Jernimo Fault separates Jurassic and older crystalline
rocks of the Central Cordillera from an intensively deformed belt of
volcanic (UPb zircon 114.3 3.8 Ma; Villagmez et al., 2011a) and
marine sedimentary rocks of the Quebradagrande Complex (Figs. 1
and 2), which are locally unconformably overlain by quartz-rich sedi-
mentary rocks of the Lower Cretaceous Abejorral Fm. to the east. Nivia
et al. (2006) propose that the Quebradagrande Complex formed within
a continental marginal basin during the AptianAlbian, although the
absence of continental crust to the west of the San Jernimo Fault, com-
bined with a lack of continental-derived detritus to the west of the basin
suggest that it was partly isolated from continental input (Restrepo
et al., 2009) and may have formed in an island arc that fringed the
continental margin (Pindell and Kennan, 2009; Villagmez et al.,
2011a). Alternatively, outboard continental crust may have been re-
moved by faulting.
The Quebradagrande Complex is in fault contact with tectonic slices
of garnet-bearing amphibolites, eclogites and lawsonite-glaucophane
schist of the Arqua Complex that is exposed along the western ank
of the Central Cordillera (Figs. 1 and 2). These mediumhigh PT meta-
morphic rocks crop-out between the Silvia-Pijao and the Cauca
Almaguer faults and have not yielded reliable geochronological data.
Previous radiometric ages (40Ar/ 39Ar phengite ages of 12060 Ma,
Bustamante et al., 2011) must be interpreted with caution due to the
potential for excess 40Ar in phengite extracted from high-pressure
rocks (e.g. DeJong et al., 2001).
Accreted Late Cretaceous mac and ultramac rocks occur to the
west of the CaucaAlmaguer Fault along the margins of the Cauca
Pata Valley, the Western Cordillera and the ChocPanam Terrane
(Fig. 1). Geochemical analyses suggest that lenses of mac and ultra-
mac volcanic rocks along the eastern margin of the CaucaPata Valley
(e.g. Amaime Fm.; Fig. 2B) formed in an oceanic hot-spot setting (Kerr
et al., 1997; Villagmez et al., 2011a). Radiometric analyses of the
Amaime Fm. are restricted to K/Ar ages of 10478 Ma (De Souza et al.,
1984) and a single whole rock, total fusion 40Ar/39Ar age of 76.3
1.7 Ma (Sinton et al., 1998), which do not accurately constrain its em-
placement age. The arc-related Buga Batholith intrudes these sequences
and its UPb zircon age range of 9290 Ma (Villagmez et al., 2011a)

Fig. 2. Geological maps of the study regions (see Fig. 1) within the Central and Western Cordilleras and the CaucaPata Valley of Colombia (after Gmez et al., 2007), showing sample
locations and the thermochronological ages acquired in this study. A) Northern Colombia, B) Central Colombia, and C) Southern Colombia. All ages are in Ma with an uncertainty
of 2, and sample codes are shown in blue (DV#). AzBF: Amazonian Border Fault, CPF: CaliPata Fault, HP: Hatillo Pluton, IF: Ibagu Fault, MB: Mande Batholith, OPF: Ot
Pericos Fault, PF: Palestina Fault, PP: Piedrancha Pluton, SPF: SilviaPijao Fault.
Table 1
Summary of thermochronological data from the Western and Central Cordilleras of Colombia.
40
Sample Unit Lithology Stratigraphic Elevation Latitude Longitude Phase Ar/39Ar Inverse MSWD1 Zircon FT Apatite FT Apatite MTL Zircon Apatite
age (m) N dms W dms age 2 Isochron age (Ma) age (Ma) (m) 2 (U-Th)/He (U-Th)/He
(Ma) Age 2 2 2 (c-axis age (Ma) age (Ma)
(Ma) corrected) 2 2

Central cordillera
The Anioquia and Sonsn batholiths, and their local basement (north of 5N)
DV53 Antioquia Batholith Diorite 9584 Ma 1958 61839.7 753016.3 72.9 18.8 13.43 0.32 13.4 1.4
DV54 Antioquia Batholith Diorite 9584 Ma 1958 61941.1 752849.6 Hbl 70.4 6.3 (W) 64.1 5.4 4.33 74.4 10.6 15.05 0.12
DV56 Antioquia Batholith Granite 9584 Ma 2192 60319.8 751242.7 65.5 6.0 40.2 2.6

D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

DV58 Antioquia Batholith Granite 9584 Ma 1143 60106.3 750810.8 Bt 72.3 0.3 (P) 73.2 0.8 2.22 58.5 8.0 32.0 1.9
DV58 Antioquia Batholith Granite 9584 Ma 1143 60106.3 750810.8 Or 62.6 1.1 (P) 56.5 8.6 1.26 45.5 3.1
DV63 Antioquia Batholith Qz-rich aplite 9584 Ma 777 55842.0 745731.5 74.8 7.4 64.1 9.6 20.6 1.4
DV64 Antioquia Batholith Granodiorite 9584 Ma 782 55856.5 745719.3 Pl 62.6 0.7 (W) 58.9 9.6 8.79 59.8 10.2
DV65 Cajamarca Complex Gneiss Triassic 602 55916.1 745534.5 Bt 80.8 0.3 (P) 81.9 0.6 0.18 43.5 4.8 13.69 0.29
DV67 Cajamarca Complex Mica schist Triassic 610 55614.7 745139.0 113.4 22.8
DV70 Antioquia Batholith Granodiorite 9584 Ma 2317 65814.8 752533.1 Bt 68.9 0.6 (P) 68.6 1.5 9.63 58.1 5.2 54.8 5.8 14.54 0.13 39.1 2.1
DV80 Cajamarca Complex White mica schist Triassic 431 55517.5 745008.7 41.0 6.6
DV148 Antioquia Batholith Granodiorite 9584 Ma 1473 62516.3 752305.7 55.4 5.2 53.0 4.8
DV153 Antioquia Batholith Granodiorite 9584 Ma 1360 63224.8 750719.0 66.6 8.0 69.7 8.6 14.42 0.16
DV154 Cajamarca Complex Schist Triassic 2276 54642.8 751832.0 53.1 5.6 56.6 7.4
DV155 Sonsn Batholith Granite 6555 Ma 2199 54618.2 751733.1 71.4 9.8 44.3 6.2
DV156 Sonsn Batholith Granite 6555 Ma 2486 54514.3 751800.5 47.9 5.2 48.2 4.8 14.95 0.17 30.7 1.8
DV161 Manizales Pluton Granite 5955 Ma 2735 50134.7 752332.8 43.9 4.2
DV163 Hatillo Pluton Granite ~53 Ma 1100 51043.6 745841.6 55.3 5.4 39.6 7.0
Jurassic granitoids and their local basement (southern and central Colombia)
DV01 Ibagu Batholith Granite 170165 1195 42960.1 750822.7 Or 116.4 0.9 (P) 112.2 6.3 7.78
DV02 Cajamarca Complex Gneiss ~220 Ma 685 44641.8 745754.2 Ap 63.6 9.4 32.1 7.2
DV03 Cajamarca Complex Gneiss Triassic 685 44641.8 745754.2 Ap 37.1 5.8 14.95 0.20
DV04 Ibagu Batholith Gabbro 170165 933 44700.2 745831.4 Ap 59.2 20.2 10.6 0.8
DV05 Ibagu Batholith Granodiorite 170165 1064 42427.7 751605.3 Ap 85.3 18.2 59.8 16.8 13.42 0.31 7.5 0.5
DV06 Ibagu Batholith Granite 170165 1277 42408.9 751740.3 Or 109.7 1.3 (W) 91.6 12.9 >100 81.3 10.6 35.8 4.6 13.81 0.43 47.7 2.7
DV07 Ibagu Batholith Granite 170165 1363 42425.4 751804.5 Or 109.8 2.9 (W) 86.4 14.2 >100 88.3 12.4 64.9 10.2 13.82 0.19 6.2 0.4
DV08 Cajamarca Complex Phyllite Triassic 1479 42449.6 751846.9 106.6 16.0
DV09 Ibagu Batholith Granite 170165 1354 42429.7 751811.8 Or 114.1 0.9 (W) 108.2 4.1 13.21 74.6 13.4 30.6 5.8 14.07 0.23 43.0 3.0
DV14 Cajamarca Complex Quartzite Triassic 2185 42618.5 752837.5 47.0 9.0
DV17 Cajamarca Complex Greenschist Triassic 2943 42744.1 753215.4 39.3 12.0
DV18 Triassic intrusion Gneiss ~236 Ma 3292 42819.0 753318.1 54.2 6.0 19.6 4.2
DV19 Cajamarca Complex Quartzite ~240 Ma 3292 42819.0 753318.1 38.7 18.0
DV81 Ibagu Batholith Granite 170165 919 42016.5 751147.6 Bt 130.2 2.8 (P) 136.8 10.2 2.02

(continued on next page)

231
 232
Table 1 (continued)
40
Sample Unit Lithology Stratigraphic Elevation Latitude Longitude Phase Ar/39Ar Inverse MSWD1 Zircon FT Apatite FT Apatite MTL Zircon Apatite
age (m) N dms W dms age 2 Isochron age (Ma) age (Ma) (m) 2 (U-Th)/He (U-Th)/He
(Ma) Age 2 2 2 (c-axis age (Ma) age (Ma)
(Ma) corrected) 2 2

DV81 Ibagu Batholith Granite 170165 919 42016.5 751147.6 Or 136.4 0.5 (P) 136.0 16.2 5.33 103.6 13.0
DV82 Permian intrusives Granite ~272 Ma 1051 41715.5 751359.2 Or 137.3 0.9 (W) 137.9 7.1 44.20 77.6 10.8 68.3 8.2 13.31 0.28 11.1 0.7
DV84 Permian intrusives Granite PermoTriassic 816 41354.1 751301.8 Or 134.3 0.7 (W) 133.0 10.4 28.37
DV85 Cajamarca Complex Schist Triassic 1403 42752.6 751623.1 30.5 5.8
DV86 Cajamarca Complex Schist Triassic 1420 42818.0 751631.7 35.0 8.2 9.3 0.8
DV129 Undiff. intrusion Gneiss Jurassic 2350 11002.5 765132.7 8.21 1.2 9.3 0.5 4.5 0.3
DV133 Ibagu Batholith Granodiorite ~176 Ma 1397 10458.2 764345.7 3.5 0.3
DV134 Ibagu Batholith Granodiorite ~176 Ma 1638 10432.5 764350.4 109.2 10.8 15.9 4.4
DV136 Ibagu Batholith Granite ~176 Ma 2260 10421.7 764453.5 106.8 13.6 13.7 2.2
DV139 Ibagu Batholith Dacitic porphyry ~176 Ma 1874 10501.8 764758.2 112 16.6 18.7 7.4

D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

The Quebradagrande complex
DV160 Quebradagrande C. Diorite Early Cret. 1426 50309.2 753419.0 69.4 17.2
DV176 Quebradagrande C. Diorite Early Cret. 1381 52716.0 752828.2 64.5 14.6
DV177 Quebradagrande C. Sandstone Early Cret. 2127 50706.5 752746.5 78.6 19.8
DV179 Quebradagrande C. Sandstone Early Cret. 1067 55007.1 753520.0 51.8 7.4
The Arqua complex
DV28 Arqua Complex Amphibolite E. Cretaceous 1321 42247.1 754309.0 Hbl 64.4 3.6 (P) 94.8 32.8 0.21
DV89A Arqua Complex Amphibolite E. Cretaceous 1374 41624.7' 754722.2 Hbl 117.2 9.4 (P) 109.3 20.0 1.50
DV89B Arqua Complex Amphibolite E. Cretaceous 1374 41624.7 754722.2 Hbl 112.0 3.7 (P) 114.1 16.5 0.74
The Crdoba Pluton
DV26 Crdoba Pluton Granodiorite ~80 Ma 1389 42430.9 754124.2 Hbl 67.9 1.6 (P) 65.9 5.1 1.51 45.6 7.6 13.9 2.4 6.1 0.4

Western Cordillera and the CaucaPatia Valley (Calima Terrane)

DV30 Buga Batholith Tonalite 9290 Ma 1664 35410.6 761050.4 44.8 8.4
DV42 Volcanic Fm Hornblende-gabbro 10189 Ma 1102 33705.0 763915.1 Hbl 77.6 5.7 (P) 80.4 18.1 0.59
DV91 Buga Batholith Diorite 9290 Ma 1117 35531.0 761442.4 Pl 79.2 2.6 (TF) n/a n/a 41.3 4.2
DV94 Bolvar U-mac Cmpx Pegmatite 9994 Ma 1032 42025.7 761144.0 Hbl 69.3 4.8 (TF) 67.6 15.3 1.94 67.9 7.4 61.5 9.4
DV95 Bolvar U-mac Cmpx Pegmatite 9994 Ma 1198 42002.1 761152.0 62.5 5.2 62.8 6.8
DV121 Piedrancha pluton Granodiorite ~23 Ma 1840 10822.5 775143.3 35.8 5.8 12.2 0.9

ChocPanam Terrane
DV165 Mande Batholith Diorite 4342 Ma 799 54604.7 761456.3' 41.6 5.4 37.8 5.2 14.46 0.29
DV166 Mande Batholith Diorite 4342 Ma 794 54609.4 761451.2 32.8 1.9 3.9 0.3
DV167 Mande Batholith Granodiorite 4342 Ma 782 54615.1 761451.1 46.4 5.8 38.1 5.4 13.47 0.15

P: plateau, >3 contiguous heating steps that span >50% 39Ar released (the age is the weighted mean age of the plateau).
39
W: weighted mean age over >3 contiguous heating steps that yield distinguishable ages that differ by less than 5%, and span > 50% Ar released.
TF: Total fusion age.
n/a: Not Applicable.
1
Mean square of weighted deviates of the inverse isochron linear regression.
Mineral abbreviations: Hbl: hornblende, Bt: biotite, Pl: plagioclase, Or: orthoclase.
P(2) values for all zircon and apatite ssion track ages are >5%.
(UTh)/He ages are weighted mean ages of 2 or more aliquots when the aliquot ages are indistinguishable.
Additional data for the 40Ar/39Ar, ssion track and (UTh)/He methods is provided in Supplementary Tables A1, A2 and A3.
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
constrains the minimum age for the Amaime Fm. Mac volcanic rocks
(Volcanic Fm.) and ultramac cumulates (Bolvar Ultramac Complex)
crop out in the Western Cordillera (Fig. 2) and are oceanic hot-spot
derived material (Kerr et al., 2004). Gabbros and basalts of the Volcanic
Fm. yield radiometric ages of 99.7 1.3 Ma (zircon UPb; Villagmez
et al., 2011a) and 91.7 2.7 Ma (whole rock 40Ar/ 39Ar; Kerr et al.,
1997), and the Bolvar Ultramac Complex (Fig. 2B) yields zircon
UPb ages of 97.1 2.0 Ma and 95.5 1.1 Ma (Villagmez et al.,
2011a). The igneous basement of the Western Cordillera and the
CaucaPata Valley probably forms part of the same Late Cretaceous
Caribbean Plateau, and are grouped here as the Calima Terrane (e.g.
Kerr et al., 1997, 2004; Restrepo and Toussaint, 1988; Fig. 1). This inter-
pretation is consistent with that proposed for equivalent rock sequences
within the Western Cordillera and at forearc region of Ecuador, where
they are grouped into the PinPallatanga Terrane (Luzieux et al.,
2006; Vallejo et al., 2009), which is considered to represent a relict sliver
of the Caribbean Large Igneous Province.
Cretaceous, plume-derived (Kerr et al., 1997) mac igneous rocks
crop-out along the Pacic coast, west of the Garrapatas Fault (Fig. 1),
and yield whole rock and plagioclase 40Ar/39Ar plateau ages of 78
73 Ma (Kerr et al., 1997). Toussaint and Restrepo (1994) proposed
that these form the basement of the ChocPanam Terrane (Fig. 1).
Wide disagreement exists regarding the timing of collision and accretion
of the ChocPanam Terrane with the Colombian margin, with opin-
ions varying between 25 and 23 Ma (Farris et al., 2011) and the latest
Mioceneearly Pliocene (Mann and Corrigan, 1990).
3. Previous thermochronological work
A paucity of reliable geochronological and thermochronological data
from the Central and Western Cordilleras of Colombia has hampered
previous attempts (Aspden et al., 1987; Villagmez et al., 2011a) to
temporally constrain the principal tectonic events that assembled
the rocks onto the South American plate. Previously published
thermochronological data are:
i) Apatite (UTh)/He data from the northern Central Cordillera pro-
vide evidence for exhumation-related cooling episodes during
4540 Ma and 2520 Ma (Restrepo-Moreno et al., 2009).
Fig. 3. Compilation of thermochronological ages (2) from the Central and Western Cordilleras, and the CaucaPata Valley in Colombia. Apatite FT and (UTh)/He ages young
towards the south within the Central Cordillera.
233
ii) K/Ar and Rb/Sr ages of metamorphic and granitic rocks of the Central
Cordillera range between 343 and 57 Ma (see Aspden et al., 1987).
Most of these ages have been interpreted to record thermal events
during the early Mesozoic and during ~6555 Ma (McCourt et al.,
1984), although the degree of partial resetting is undetermined.
A signicant quantity of thermochronological data has been gen-
erated in the Eastern Cordillera of Colombia (e.g. Mora et al., 2010;
Parra et al., 2012). Those data showed that the Eastern Cordillera
was subject to intense shortening and exhumation during 60
50 Ma, 4030 Ma, ~ 20 Ma and 63 Ma. Apatite FT analyses in the
southern Eastern Cordillera (Garzn Massif; Fig. 1) yielded ages of
b10 Ma, whereas zircon FT ages of ~ 120 Ma (Andriessen, 1996)
from the same region led the authors to propose that the massif
cooled rapidly during the Early Cretaceous.
4. Analytical techniques
We obtained thermochronological data using the 40Ar/ 39Ar
(hornblende, biotite, orthoclase), FT (zircon and apatite) and (UTh)/
He method (zircon and apatite). A summary of the data is presented
in Table 1 and the ages are shown in Fig. 3 (all uncertainties quoted
are 2). Complete data sets are provided in Supplementary Tables
A1 (40Ar/39Ar), A2 (FT) and A3 ((UTh)/He). Detailed analytical infor-
mation is provided in the Supplementary le analytical techniques.
Fish Canyon Tuff sanidine was used as a uence monitor for 40Ar/
39
Ar analyses, with an age of 28.02 0.16 Ma (1 internal uncertainty;
Renne et al., 1998), and ages were calculated using the 40K decay
constant of Steiger and Jger (1977).
The majority of the FT data were obtained using the external
detector method (Gleadow, 1981), although the uranium content of
four apatite samples was determined using LA-ICP-MS (e.g. Hasebe
et al., 2004) by A to Z Inc. 252Cf irradiation was used to enhance the
number of conned track-in-tracks available for length measurement
(Donelick and Miller, 1991). Maximum etch-pit diameters (Dpar)
were measured and used to assess the inter-grain variation of annealing
kinetics in apatites in some of the samples. Individual age populations
(Table A2) have been resolved using the software BINOMFIT
(Brandon, 1996) when the P(2) is b5% (Green, 1981).
234 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

A) DV01-Or: Granite, Jurassic

Ibague Batholith

36Ar / 40Ar (10-3)

300 Weighted Plateau Age: 112.26.3 Ma
Total Fusion: 123.20.3 Ma 3 40Ar/36Ar intercept: 371108

Age (Ma)
260
MSWD: 7.78
220 2
180 116.40.9 Ma
1
140
100 0
0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

B) DV06-Or: Granite, Jurassic

Ibague Batholith
140 Age: 91.612.9 Ma

36Ar / 40Ar (10-3)

40Ar/36Ar intercept: 12441200
3
MSWD: >100
Age (Ma)

109.71.3 Ma
120
2

100 1
Disturbed spectra
Total Fusion: 109.4 0.2 Ma 0
80
0 20 40 60 80 100 0.00 0.04 0.08 0.12 0.16

C) DV07-Or: Granite, Jurassic

Ibague Batholith
240
Disturbed spectra Age: 86.414.2 Ma
36Ar / 40Ar (10-3)

Total Fusion: 125.410.26 Ma 3 40Ar/36Ar intercept: 725343

200
Age (Ma)

MSWD: >100
160 109.82.9 Ma 2

120 1

80 0
0 20 40 60 80 100 0.00 0.04 0.08 0.12 0.16

D) DV09-Or: Granite, Jurassic

Ibague Batholith
36Ar / 40Ar (10-3)

Disturbed spectra Age: 108.24.1 Ma

Total Fusion: 115.1 0.3 Ma 3 40Ar/36Ar intercept: 747349
120 MSWD: 13.21
Age (Ma)

2
130
114.10.9 Ma 1
100

0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

Cumulative 39Ar Released (%) 39Ar / 40Ar

Fig. 4. 40Ar/39Ar age spectra and isotope correlation diagrams for orthoclase extracted from autochthonous rocks located north of the Ibagu Fault, Central Cordillera. Or: orthoclase.
Grey squares in the isotope correlation plots represent heating steps that are included in the age calculation, and dene the grey regression line. White squares were derived from
heating steps that dene disturbed parts of the age spectrum, and do not form part of the age calculation. Plateaus are dened according to Dalrymple and Lanphere (1974). Bt:
biotite, Hbl: hornblende, Or: orthoclase, Plag: plagioclase. All uncertainties are 2.

5. Thermochronological data from the Central Cordillera
5.1. Jurassic granitoids and their local basement (central and southern
Colombia)
5.1.1. 40Ar/ 39Ar dates (closure temperatures >~ 300 C)
Jurassic granitoids of the Ibagu Batholith (zircon UPb age of
~176 Ma, Cochrane et al., 2011) located north of the Ibagu Fault
in central Colombia yield Early Cretaceous 40Ar/ 39Ar ages (Fig. 2B).
Orthoclase extracted from four felsic intrusive rocks of the Ibagu Bath-
olith (DV01, DV06, DV07 and DV09; Fig. 4AD) yielded disturbed 40Ar/
39
Ar age spectra with weighted mean ages of ~115 Ma (DV01 and
DV09) and ~110 Ma (DV06 and DV07) over the attest regions
of their age spectra, where contiguous step ages differ by b5%, over
>60% of the total 39Ar released. These ages are younger than 40Ar/
39
Ar ages of samples located south of the Ibagu Fault (see below). Dis-
cordance is a consequence of excess 40Ar in the rst 20% of the total
39
Ar released, and 39Ar recoil, perhaps due to the presence of perthitic
lamellae. The age spectrum yielded by orthoclase DV09 (Fig. 4D) may
represent a classic volume diffusion (40Ar loss) prole from a weighted
mean age of 114.1 0.9 Ma towards a younger age of ~100105 Ma.
Permian and Jurassic granites located south of the Ibagu Fault yield
Early Cretaceous 40Ar/39Ar ages (Fig. 2B), which are older than those
obtained from granitoids located north of the fault. Biotite and ortho-
clase extracted from granite (DV81) of the Ibagu Batholith yielded
plateau ages of 130.2 2.8 Ma and 136.4 0.5 Ma, respectively
(Fig. 5A, B). Orthoclase from Permian granites DV82 (zircon UPb age
271.9 3.7 Ma; Villagmez et al., 2011a) and DV84 yielded weighted
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 235

A) DV81-Bt: Granite, Jurassic

Ibague Batholith 4

36Ar / 40Ar (10-3)

Inverse Isochron: 136.810.2 Ma
40Ar/36Ar intercept: 280.722.7
130.22.8 Ma 3
140 MSWD: 2.02

Age (Ma)
2

120 1
Weighted Plateau
Total Fusion: 127.71.6 Ma 0
100 0.00 0.02 0.04 0.06 0.08 0.10 0.12
0 20 40 60 80 100

B) DV81-Or: Granite,
Ibague Batholith
150 4
Weighted Plateau

36Ar / 40Ar (10-3)

Age: 136.016.2 Ma
Total Fusion: 134.70.3 Ma 40Ar/36Ar intercept: 299.616.6
Age (Ma)

140 3
MSWD: 5.33
130 2
136.40.5 Ma
120 1

110 0
0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12

C) DV82-Or: Granite, Permo-

Triassic Batholith
150 4
Disturbed spectra Age: 137.97.1 Ma
36Ar / 40Ar (10-3)
Total Fusion: 137.10.3 Ma 40Ar/36Ar intercept: 291.649.8
Age (Ma)

3
140 MSWD: 44.20
2
130 137.30.9 Ma
1

120 0
0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12

D) DV84-Or: Granite, Permo-

triassic Batholith
150 4
Disturbed spectra Age: 133.010.4 Ma
36Ar / 40Ar (10-3)

Total Fusion: 133.60.3 Ma 40Ar/36Ar intercept: 319.023.4

3
Age (Ma)

140 134.30.7 Ma MSWD: 28.37

2
130
1

120 0
0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12
Cumulative 39Ar Released (%) 39Ar / 40Ar

40
Fig. 5. Ar/39Ar age spectra and isotope correlation diagrams for biotite and orthoclase extracted from autochthonous rocks located south of the Ibagu Fault, Central Cordillera.

mean ages of 137.30.9 Ma and 134.30.7 Ma, respectively (Fig. 5C, D),
that do not strictly meet the plateau criteria of Dalrymple and Lanphere
(1974). However, we consider those ages to be meaningful because
contiguous age steps differ by less than 0.8% over >70% 39Ar released,
and the weighted mean ages are indistinguishable from the isotope
correlation ages (e.g. Spikings et al., 2002).
5.1.2. Fission track and (UTh)/He data (closure temperatures
b~ 300 C)
Twelve zircon FT ages from granitoids of the Jurassic Ibagu Batho-
lith and its metasedimentary basement (Cajamarca Complex) in central
Colombia range between 107 and 54 Ma, and most samples yielded
Late Cretaceous ages (Fig. 2B). With the exception of metasedimentary
rocks (Cajamarca Complex) DV14, DV17 and DV19, all zircon FT ages
yield P(2) values >5%. Zircon (UTh)/He ages of 43.0 3.0 (DV09)
and 47.7 2.7 Ma (DV06) were acquired from the Ibagu Batholith,
which lie within the same range as those obtained from northern
Colombia (Fig. 3). Apatite FT ages from metasedimentary rocks of the
Cajamarca Complex and granitoids of the Ibagu Batholith in central
Colombia range between 68.3 8.2 Ma and 19.6 4.2 Ma, with no
geographic trends (Fig. 2B). The large age variation is probably due to
displacement of the Palestina fault, the high angle reverse OtPericos
Fault and the dextral Ibagu Fault and their splays. Indistinguishable
apatite FT ages that range between 30.5 5.8 Ma and 37.1 5.8 Ma
(DV86, 85, 9, 6, 3, 2) have been obtained from samples located within
the Ibagu and OtPericos fault zones (Fig. 2B). The youngest apatite
FT age of 19.6 4.2 (DV18) Ma was obtained from a gneiss (zircon
UPb age of 237.5 5.5 Ma; Villagmez et al., 2011a) of the Cajamarca
Complex, located close to the Palestina Fault. Older apatite FT ages that
range between 59.8 16.8 Ma and 68.3 8.2 Ma (DV82, 7, 5, 4) were
obtained from rocks with a lower density of brittle fractures and faults
(Fig. 2B). Five of these samples yielded partially annealed mean track
236 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
lengths (13.7611.30 m; Table 1). Apatite (UTh)/He ages (Table 1)
obtained from various Jurassic and older (Villagmez et al., 2011a)
intrusive bodies yielded ages that span between 6.2 0.4 Ma and
11.1 0.7 Ma (DV82, 7, 5, 4), and a metasedimentary rock of the
Cajamarca Complex yielded an age of 9.3 0.8 Ma (DV86). These ages
are signicantly younger than apatite FT ages from the same region,
and are younger than apatite (UTh)/He ages obtained from the northern
Central Cordillera (Fig. 3).
Three granitoids of the Jurassic Ibagu Batholith, located in a single
fault block in southern Colombia (DV139, 134 and 136; Fig. 2C) were
collected over an elevation range of 1630 to 2260 m, and yield zircon
FT ages that range between 106.8 13.6 Ma and 112 16.6 Ma,
which overlap with the range of zircon FT ages obtained for the same
batholith in central Colombia. Orthogneiss DV129, located proximal to
the Ecuadorian border, yielded a zircon (UTh)/He age of 9.3
0.5 Ma, which is the youngest in the Central Cordillera. The same
three granitoids yielded indistinguishable apatite FT ages that range
between 13.7 2.2 Ma and 18.7 7.4 Ma (DV139, 136, 134). An
orthogneiss of the Ibagu Batholith located on a separate fault block
(Fig. 2C), yielded a younger apatite FT age of 8.2 1.2 Ma (DV129).
No useful FT track length data could be obtained from this region. The
youngest apatite (UTh)/He ages in this study have been obtained
from the southern Central Cordillera, close to the Ecuadorian border
(Figs. 2C and 3). An orthogneiss and granodiorite of the Ibagu Batholith
located at elevations between 1400 m and 2350 m yielded apatite
(UTh)/He ages of 4.5 0.3 Ma (DV129) and 3.5 0.3 Ma (DV133)
respectively.
5.2. The Antioquia and Sonsn Batholiths, and their local basement
(north of 5N)
5.2.1. 40Ar/ 39Ar dates (closure temperatures >~ 300 C)
A Triassic gneiss (DV65, Fig. 2A) located proximal to the Late
Cretaceous Antioquia Batholith in the northern Central Cordillera
yielded a biotite 40Ar/ 39Ar plateau age of 80.8 0.3 Ma (Fig. 6A).
An undeformed diorite of the Antioquia Batholith (9487 Ma;
Villagmez et al., 2011a) yielded a hornblende, weighted mean
40
Ar/ 39Ar age of 70.4 6.3 Ma (DV54; Fig. 6B) from the attest part
of a slightly disturbed region of the age spectrum. We consider the
weighted mean age to be geologically meaningful because contiguous
step-ages differ by less than 5% over ~ 90% of 39Ar released. Biotites
from two granitoids of the Antioquia Batholith yielded plateau 40Ar/
39
Ar ages of 72.30.3 Ma (DV58) and 68.90.6 Ma (DV70; Fig. 6C, D).
Feldspars extracted from granite located at the southernmost expo-
sure of the batholith yielded disturbed age spectra, albeit with indis-
tinguishable weighted mean ages of 62.6 1.1 Ma (orthoclase;
DV58) and 62.6 0.7 Ma (plagioclase; DV64; Fig. 6E, F). We consider
these ages to be interpretable because the contiguous step ages
differ by less than 3%, over 42% and 60% of the total 39Ar released,
respectively.
5.2.2. Fission track and (UTh)/He data (closure temperatures
b~ 300 C)
The Late Cretaceous Antioquia and Paleocene Sonsn (6555 Ma;
UPb zircon; Ordez-Carmona et al., 2001) batholiths, and several
basement inliers form the broad, high (mean elevation 2500 m)
plain of the Altiplano Antioqueo (Fig. 2A). Zircon FT ages from this
region show no spatial trend and range between 71.4 9.8 Ma and
47.9 5.2 Ma (Sonsn Batholith; DV156 and 155), and 66.6
8.0 Ma55.4 5.2 (Antioquia Batholith; DV153, 148, 70, 56). These
ages are either indistinguishable, or slightly older than the apatite FT
ages yielded by the same samples. Zircon (UTh)/He ages of 45.5
3.1 Ma (DV58) and 40.2 2.6 Ma (DV56) have been obtained from
granites located at 1140 and 2190 m, respectively. The zircon (UTh)/
He age of granite DV58 is younger than its apatite FT age, which may
reect enhanced He diffusion in two of the zircon aliquots, as a
consequence of radiation damage (Table A3; U >850 ppm). Apatite FT
ages from the granitoids range between 69.7 8.6 Ma and 44.3
6.2 Ma (DV156, 155, 153, 148, 70, 58), with partially annealed mean
lengths ranging between 13.740.28 m and 13.00 0.27 m. A single
mica schist of the Triassic Cajamarca Complex (DV154) yields an apatite
FT age of 56.6 7.4 Ma. Finally, apatite (UTh)/He ages from the same
rocks are 39.1 2.1 Ma (DV70), 32.0 1.9 Ma (DV58) and 30.7
1.8 Ma (DV156), which are the oldest obtained from the cordilleras of
Colombia.
Diorites located along the western slopes of the Altiplano
Antioqueo yield apatite FT ages of 74.4 10.4 Ma (DV54) and 72.9
18.8 Ma (DV53) from the Antioquia Batholith (9487 Ma; Villagmez
et al., 2011a; Fig. 2A), with mean apatite FT lengths of 13.99 0.18
and 11.70 0.44 m, respectively. These are amongst the oldest apatite
FT ages obtained in the Colombian Andes (Fig. 3). Diorite DV53 yielded
an apatite (UTh)/He age of 13.4 1.4 Ma, which is considerably youn-
ger than its apatite FT age. Granite from the Manizales pluton (K/Ar
biotite age 5955 Ma; Fig. 2A; Brook, 1984) yielded a zircon FT age of
43.9 4.2 Ma (DV161).
Zircon FT ages yielded by the Antioquia Batholith and the Hatillo
pluton along the eastern slope of the Altiplano Antioqueo are 74.8
7.4 Ma (DV63) and 55.3 5.4 Ma (DV163), respectively, which are
consistently older than their apatite FT ages. A Triassic schist of the
Cajamarca Complex (DV67), located in the same region, yielded the
oldest (albeit imprecise) apatite FT age of 113.4 22.8 Ma from the
cordilleras (Fig. 2A). Additional apatite FT ages from the eastern slope
range between 64.1 9.6 Ma and 41.0 6.6 Ma (DV80, 65, 64, 63),
and gneiss DV65 Cajamarca Complex) yielded a mean apatite FT length
of 13.69 0.29 m. The Tertiary Hatillo pluton (53.0 1.8, biotite K/Ar;
Vesga and Barrero, 1978; Fig. 2A) yielded an imprecise apatite FT age of
39.6 7.0 Ma (DV163), and aplite DV63 (Antioquia Batholith) yielded
an apatite (UTh)/He age of 20.9 1.2 Ma, which is signicantly youn-
ger than its apatite FT age.
5.3. The Quebradagrande Complex (west of the San Jeronimo Fault)
Apatite FT analyses of diorites and sandstones of the Early
Cretaceous Quebradagrande Complex (UPb zircon 114.3 3.8 Ma;
Villagmez et al., 2011a), located along the western ank of the
Central Cordillera (Fig. 2A), yielded ages that range between 78.6
19.8 and 51.8 7.4 Ma (DV160, 176, 177, 179). These are amongst
the oldest apatite FT ages obtained in the Colombian Andes (Fig. 3)
and are similar to apatite FT ages obtained from the western
Antioquia Batholith. Unfortunately, no statistically useful track length
information could be obtained.
5.4. The Arqua Complex (west of the SilviaPijao Fault)
Garnet amphibolites of the highmedium PT metamorphic rocks
of the Arqua Complex, exposed on the western border of the Central
Cordillera (Fig. 2A, B) yielded similar 40Ar/ 39Ar ages to those found in
the Jurassic Ibagu Batholith. Hornblende extracted from amphibo-
lites (DV89A and DV89B) sampled from the same outcrop yielded
imprecise and indistinguishable plateau ages of 117.2 9.4 Ma and
112.0 3.7 Ma (Fig. 7A, B). Hornblende DV89A yielded a humped
age spectrum that may be diagnostic of 39Ar recoil, and hence we con-
sider the plateau age of DV89B (112.0 3.7 Ma) to be the most accu-
rate hornblende age for this sequence, in this region. Hornblende
from a garnet amphibolite (DV28) located closer to the SilviaPijao
Fault (Fig. 2B) than the other amphibolites yielded a younger plateau
40
Ar/ 39Ar age of 64.4 3.6 Ma (Fig. 7C). This age is similar to 40Ar/
39
Ar ages obtained from continental crust exposed in the northern
Central Cordillera, and we consider it to be a reset age, which is not
indicative of the timing of crystallisation.
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 237

5.5. The Crdoba Pluton
The Crdoba pluton intrudes into the Quebradagrande Complex,
and yields a zircon UPb age of 79.7 2.5 Ma (Villagmez et al.,
2011a; Fig. 2B). Diorite (DV26) of the pluton yielded a hornblende
plateau 40Ar/ 39Ar age of 67.9 1.6 Ma (Fig. 7D), with an indistin-
guishable inverse isochron age of 65.9 5.1 Ma.
Zircon and apatite and FT ages of 45.6 7.6 Ma and 13.9 2.4 Ma,
respectively were obtained from granite DV26, and the same sample
yielded a younger apatite (UTh)/He age of 6.1 0.4 Ma.
6. Thermochronological data from Late Cretaceous accreted
mac rocks
isochron age of 80.4 18.1 Ma (Fig. 8A). Hornblende from a pegma-
tite of the Late Cretaceous Bolvar Ultramac Complex (DV94), locat-
ed along the eastern ank of the Western Cordillera, yielded a
disturbed age spectrum with a total fusion age of 69.3 4.8 Ma
(Fig. 8B), which is indistinguishable from its inverse isochron age of
67.6 15.3 Ma. The progressive reduction in age from low to high
temperature heating steps suggests that the discordance may be a
consequence of excess 40Ar that is not resolved by the inverse
isochron.
Plagioclase from a diorite (DV91) of the Buga Batholith (zircon UPb
ages span 9290 Ma; Villagmez et al., 2011a), located within the
CaucaPata Valley (Fig. 2B), yields a highly disturbed age spectra
with an uninterpretable total fusion age of 79.2 2.6 Ma (Fig. 8C).

6.1. The Western Cordillera and the CaucaPata Valley (Calima Terrane)
6.1.1. 40Ar/ 39Ar dates (closure temperatures >~ 300 C)
The mac basement of the Western Cordillera and the CaucaPata
Valley is characterised by mac crystalline rocks and minor granitoid
intrusions that are ~ 100 Ma or younger, and are allochthonous to
South America (Villagmez et al., 2011a).
With the exception of gabbro DV42, mac crystalline rocks (Figs. 1
and 2b) yield disturbed 40Ar/ 39Ar age spectra (Fig. 8). A gabbro of
the Volcanic Fm. (DV42; Fig. 2B; zircon UPb age 99.7 1.3 Ma;
Villagmez et al., 2011a) yielded an imprecise hornblende plateau
40
Ar/ 39Ar age of 77.6 5.7 Ma, with an indistinguishable inverse
6.1.2. Fission track and (UTh)/He data (closure temperatures
b~ 300 C)
Two pegmatite dykes (DV94, DV95) of the Bolvar Ultramac Com-
plex, located along the eastern ank of the Western Cordillera
(Fig. 2B) yielded zircon FT ages of 67.9 7.4 and 62.5 5.2 Ma, and
apatite FT ages of 61.5 9.4 Ma and 62.8 6.8 Ma. These samples
yielded partially annealed mean apatite FT lengths of 11.28 0.74 m
and 12.02 0.67 m and large Dpar values of 3.4 m (DV94) and
2.5 m (DV95).
Two granodiorites of the Buga Batholith (UPb zircon ages span
9092 Ma; Villagmez et al., 2011a), located west of the Cauca
Almaguer Fault within the CaucaPata Valley in central Colombia

A) DV65-Bt: Gneiss, Triassic

Cajamarca Fm.
90 4
Weighted Plateau Inverse Isochron: 81.90.6 Ma
36Ar / 40Ar (10-3)

Total Fusion: 80.2 0.2 Ma 40Ar/36Ar intercept: 182.952.8

3
Age (Ma)

80 MSWD: 0.18

80.80.3 Ma 2
70
1

60
0 10 20 30 40 50 60 70 80 90 100
0.00 0.04 0.08 0.12 0.16

B) DV54-Hbl: Diorite, Cretaceous

Antioquia Batholith
400
Disturbed spectra Age: 64.15.4 Ma
350 4
36Ar / 40Ar (10-3)

Total Fusion: 71.01.9 Ma 40Ar/36Ar intercept: 348.638.0

300
Age (Ma)

3 MSWD: 4.33
250
200
2
150
100 70.46.3 Ma
1
50
0
0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12

C) DV58-Bt: Granite, Cretaceous

Antioquia Batholith
80
Weighted Plateau Inverse Isochron: 73.20.8 Ma
4
36Ar / 40Ar (10-3)

Total Fusion: 71.40.2 Ma 40Ar/36Ar intercept: 185.098.2

MSWD: 2.22
Age (Ma)

3
70
2
72.30.3 Ma
1
60

0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12

40
Fig. 6. Ar/39Ar age spectra and isotope correlation diagrams for mineral phases extracted from autochthonous rocks of the northern (~6N) Central Cordillera.
238 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

D) DV70-Bt: Granodiorite, Cretaceous

Antioquia Batholith
80 Age: 68.61.5 Ma

36Ar / 40Ar (10-3)

3 40Ar/36Ar intercept: 329.3235.3
MSWD: 9.63

Age (Ma)
68.90.6 Ma 2
60
1
Weighted Plateau
Total Fusion: 69.10.2 Ma
40
0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12

E) DV58-Or: Granite, Cretaceous

Antioquia Batholith
Weighted Plateau Inverse Isochron: 56.58.6 Ma
100

36Ar / 40Ar (10-3)

40Ar/36Ar intercept: 439.2223.0
Total Fusion: 63.40.5 Ma 3
MSWD: 1.26
Age (Ma)

80 62.61.1 Ma
2
60
1
40

0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

F) DV64-Plag: Granodiorite, Cretaceous

Antioquia Batholith
Disturbed spectra Age: 58.99.6 Ma
36Ar / 40Ar (10-3)

80 3 40Ar/36Ar intercept: 332.3100.3

Total Fusion: 63.20.3 Ma
MSWD: 8.79
Age (Ma)

2
70 62.60.7 Ma
1
60

0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

Cumulative 39Ar Released (%) 39Ar / 40Ar

Fig. 6 (continued).

(Fig. 2B) yielded indistinguishable zircon and apatite FT ages of
41.3 4.2 Ma (diorite, DV91) and 44.8 8.4 Ma (tonalite DV30),
respectively.
Further south, a single apatite FT age of 25.8 4.8 Ma (DV121)
was obtained from granodiorite of the Miocene Piedrancha pluton
(K/Ar biotite, 23 3 Ma; Alvarez and Linares, 1979), in the southern
Western Cordillera (Fig. 2C). The same sample yielded an apatite
(UTh)/He age of 12.2 0.9 Ma, which is older than apatite (UTh)/
He ages obtained from the Central Cordillera at similar latitudes.
6.2. The ChocPanam Terrane
Two granodiorites (DV167, DV165) of the Eocene Mande Batholith
(Fig. 2A; zircon UPb ages of 4342 Ma; Cardona et al., 2010), which
form part of the ChocPanam Terrane (Fig. 1), yielded indistin-
guishable zircon FT ages of 46.4 5.8 Ma and 41.6 5.4 Ma, and
apatite FT ages of 38.1 5.4 Ma and 37.8 5.2 Ma, respectively.
Granodiorite DV166 (Mande Batholith) yielded a zircon (UTh)/He
age of 32.8 1.9 Ma, and an apatite (UTh)/He age of 3.9 0.3 Ma,
which is the youngest in northern Colombia.
7. Recovery of thermal histories
Additional thermochronological data (e.g. zircon (UTh)/He date and
grain size) yielded by those samples that constrains the t-T models
was used, when available.
7.1. 550250 C
Thermal history information has been extracted from the 40Ar/39Ar
data by estimating the closure temperatures (Tc) of each mineral phase
using experimentally derived diffusion parameters (e.g. McDougall and
Harrison, 1999). Bulk Tc values for hornblende and biotite range between
~545511 C and 360325 C, respectively (10010 C/My). We have
utilised a broad closure temperature of 29060 C for the plateau and
attest regions of age spectra obtained from orthoclase (Lovera et al.,
1991), assuming that the gas is mainly derived from the largest lattice
domains (e.g. Moni et al., 1994). We assume a bulk closure temperature
of 300225 C for plagioclase (Cassata et al., 2009). Those samples which
yielded both 40Ar/39Ar feldspar and zircon FT ages (Tc 290 C210 C;
Brandon et al., 1998) gave consistently older 40Ar/39Ar ages, conrming
our feldspar 40Ar/39Ar closure temperatures. Raman spectroscopy was
routinely performed to distinguish between orthoclase and plagioclase.
The time of each coordinate is determined after evaluation of the age
spectrum.

Semi-continuous thermal histories have been determined for sam- 7.2. 25040 C
ples that yielded data which can be reliably inverted to generate t-T
models. This restricts modelling to samples which yielded apatite FT The FT and (UTh)/He methods have been applied to zircon and ap-
dates and a statistically useful quantity of track length measurements. atite for the purpose of generating integrated thermal histories between
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 239

A) DV89A-Hbl : Garnet Amphibolite,

Arquia Fm
300 4 Inverse Isochron: 109.320.0 Ma
Weighted Plateau 40Ar/36Ar intercept: 335100

36Ar / 40Ar (10-3)

Total Fusion: 100.1 7.0 Ma
240 MSWD: 1.50
3

Age (Ma)
180 117.29.4 Ma
2
120

60 1

0 0
0 20 40 60 80 100
0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

B) DV89B-Hbl : Garnet Amphibolite,

Arquia Fm
4
Weighted Plateau Inverse Isochron: 114.116.5 Ma

36Ar / 40Ar (10-3)

40Ar/36Ar intercept: 2876.70
200 MSWD: 0.60
3 MSWD: 0.74
Total Fusion: 97.53.7 Ma
Age (Ma)

2
100
112.03.7 Ma 1

0 0
0 20 40 60 80 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12 0.14

C) DV28-Hbl: Amphibolite,
Arquia Fm.
100
Weighted Plateau 4 Inverse Isochron: 94.832.8 Ma
Total Fusion: 51.63.3 Ma 40Ar/36Ar intercept: 205.0102
36Ar / 40Ar (10-3)

80
MSWD: 0.21
Age (Ma)

3
60

64.43.6 Ma 2
40

20 1

0
0 10 20 30 40 50 60 70 80 90 100 0.00 0.04 0.08 0.12 0.16 0.20 0.24

D) DV26-Hbl: Granodiorite, Cretaceous

Cordoba batholith
100 4
Weighted Plateau Inverse Isochron: 65.95.1 Ma
36Ar / 40Ar (10-3)

Total Fusion: 70.61.3 Ma 40Ar/36Ar intercept: 319.560.2

3
Age (Ma)

80 MSWD: 1.51

2
60
67.91.6 Ma 1

40
0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10 0.12
Cumulative 39Ar Released (%) 39Ar / 40Ar

40
Fig. 7. Ar/39Ar age spectra and isotope correlation diagrams for hornblende extracted from the Arqua Complex and the Crdoba Batholith, Central Cordillera.

~25040 C (e.g. Farley and Stockli, 2002; Gleadow et al., 1986). Poten-
tial thermal history solutions have been derived by simultaneously
modelling He diffusion (Farley and Stockli, 2002; Flowers et al., 2009;
Reiners et al., 2004) and FT annealing in apatite (c-axis projected lengths;
Ketcham et al., 2007) using the HeFTy software (v 1.6.7; Ketcham, et al.,
2007). The modelling procedure predicts FT and (UTh)/He parameters
for various thermal history paths and compares them with the observed
data. A controlled random search procedure identies those thermal
histories that most closely match the analytical data.
Wide-ranging values for the temperature bounds of the zircon par-
tial annealing zone have been published. Recently, Bernet and Garver
(2005) proposed that a useful working approximation of the zircon
partial annealing zone is 290210 C, and the mid-range value of
250 C has been plotted against the zircon FT age in this study to pro-
duce a timetemperature constraint.
The diffusivity (apatite, 1.50.6 cm2/s; zircon, 0.46+0.870.3 cm2/s),
activation energy of diffusion (apatite, 330.5 kcal/mol; zircon 40
0.9 kcal/mol) of He, grain sizes, corrections for alpha-loss (Farley and
Stockli, 2002) and corrections for radiation damage accumulation were
used to model He diffusion in zircon (Reiners et al., 2004) and apatite
(Flowers et al., 2009). The apatite He partial retention zone is taken to
be ~9040 C (e.g. Farley and Stockli, 2002), whereas the zircon He par-
tial retention zone is considered to span between ~220140 C (Reiners
et al., 2004).
240 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

A) DV42-Hbl: Gabbro, Cretaceous

Palmar Stock
120 Weighted Plateau 4
Total Fusion: 72.46.0 Ma Inverse Isochron: 80.418.1 Ma

36Ar / 40Ar (10-3)

100 40Ar/36Ar intercept: 292.717.8
3

Age (Ma)
MSWD: 0.59
80
60 2
40
77.65.7 Ma 1
20
0
0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10

B) DV94-Hbl: Pegmatite, Cretaceous

Bolivar Ultramafic Complex
4
240 Disturbed spectra Inverse Isochron: 67.615.3 Ma

36Ar / 40Ar (10-3)

Total Fusion: 69.34.8 Ma 40Ar/36Ar intercept: 340.543.7
3
Age (Ma)

180 MSWD: 1.94

120 2

60 1

0
0 10 20 30 40 50 60 70 80 90 100 0.00 0.02 0.04 0.06 0.08 0.10
39Ar / 40Ar
C) DV91-Plag: Granodiorite,
Cretaceous Buga Batholith
120 Total Fusion age: 79.22.6 Ma
Age (Ma)

100

80

60

40
0 10 20 30 40 50 60 70 80 90 100
Cumulative 39Ar Released (%)

Fig. 8. 40Ar/39Ar age spectra and isotope correlation diagrams for hornblende and plagioclase extracted from allochthonous rocks of the Western Cordillera and CaucaPata Valley
of Colombia.

8. Thermal histories of autochthonous rocks of the Central
Cordillera of Colombia
8.1. Late CretaceousTertiary granitoids and their basement (northern
Colombia)
Good-t thermal histories for the Antioquia Batholith reveal an
initial period of high cooling rates at varied times between 76 Ma
and 50 Ma (Fig. 9AD). The earliest period of cooling was detected
in diorite DV54 (located close to the SilviaPijao Fault), which cooled
at rates ~ 30 C/My during ~ 7665 Ma. Diorite DV53 also cooled
rapidly during 7550 Ma, although no good-ts were obtained. Both
samples remained at approximately constant temperatures during
5015 Ma, and diorite DV53 subsequently cooled to surface tempera-
tures during 150 Ma from within the He partial retention zone. Further
east, granodiorite DV153 yielded a good-t solution with rapid cooling
between 75 and 70 Ma from ~280 C to 80 C at rates of 40 C/My.
Subsequently, moderate cooling occurred during 6550 Ma from 80 C
to 60 C at rates of 2 C/My, and no reliable information exists below
60 C. The northernmost exposures of the batholith (DV70) yielded a
good-t solution after integrating four different thermochronometers
(Fig. 9D), which reveal rapid cooling during ~7050 Ma from ~320 C
to ~100 C, at 10 C/My.
Gneiss DV65 of the Triassic Cajamarca Complex sampled along the
eastern ank of the Central Cordillera yields a best-t envelope with a
slow, approximately constant cooling rate during ~ 700 Ma (Fig. 9E).
Its biotite 40Ar/ 39Ar age (80.8 0.3 Ma) is distinguishably older than
70 Ma and hence extrapolation to temperatures >120 C is tentative,
although it supports elevated cooling rates during the Campanian.
No good-t solutions were obtained from the apatite FT and (UTh)/
He data acquired from the Paleocene Sonsn Batholith (Fig. 9F)
although granite DV156 yielded an acceptable-t showing rapid cooling
from 240 C to 100 C during 5545 Ma at rates of ~13 C/My, when
the t-T paths are extrapolated to the ZFT age. The batholith crystallised
during 6555 Ma (Ordez-Carmona et al., 2001), and cooling from
55 Ma was probably sub-solidus thermal relaxation of the pluton.
8.2. PermianJurassic granitoids and their basement (central and
southern Colombia)
8.2.1. North of the Ibagu Fault
Granitoids (DV06 and DV09) of the Ibagu Batholith and gneiss
(DV03) of the Triassic Cajamarca Complex located north of the Ibagu
Fault experienced distinctly different cooling histories to samples located
south of the fault (Fig. 10B, C, E). Orthoclase cooled through 29060 C
during ~115110 Ma, and the ages of low-temperature 40Ar/39Ar
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 241

ANTIOQUIA BATHOLITH
0
D) DV70, granodiorite
20
40
A)DV54, diorite 60
0
Unconstrained 80
20
100

Temperature (C)
Temperature (C)

40 Mean length
120
60 observed: 15.050.64 m
best-fit: 15.150.72 m 140
80 Apatite (U-Th)/He
160 Measured: 39.12.1 Ma Mean length
100
0.6 N: 118 180 Best-fit: 39.0 Ma observed: 14.540.94 m
120 GOF: 0.84 GOF: 0.80 best-fit: 14.661.03 m
AFT 200
140 AFT
Measured: 74.410.4 Ma 220 N: 191
Best-fit: 74.9 Ma 0.3 Measured: 54.85.8 Ma 0.4 GOF: 0.68
240 Best-fit: 53.8 Ma
500 GOF: 0.92
260 GOF: 0.74
520
Ar (hbl): 70.46.3 Ma 280 ZFT
540 0.0 Measured: 58.15.2 Ma
0.2
80 70 60 50 40 30 20 10 0 4 8 12 16 20 300
Time (Ma) Length (m) 320 Ar (bt): 68.90.6 Ma
340 0.0
80 70 60 50 40 30 20 10 0 4 8 12 16 20

0
B) DV53, diorite Time (Ma) Length (m)

20
Temperature (C)

40
CAJAMARCA FORMATION
60
80 E)DV65, gneiss
0
100 Unconstrained
120 40
140 Mean length
observed: 13.431.13 m
160 80
Temperature (C)

Apatite (U-Th)/He best-fit: 13.161.29 m

180 Measured: 13.41.4 Ma
Best-fit: 18.0 Ma N: 50 120
GOF: 0.11
0.30 GOF: 0.11 Mean length
160 observed: 13.671.22 m
AFT best-fit: 14.121.07 m
AFT Measured: 43.54.8 Ma
Measured: 72.918.8 Ma 0.15 200 Best-fit: 44.3 Ma N: 69
Best-fit: 56.9 Ma ? GOF: 0.98 0.30 GOF: 0.83
GOF: 0..09 240
0.00 Ar (bt): 80.80.3 Ma
90 80 70 60 50 40 30 20 10 0 481216 20 280 0.15
Time (Ma) Length (m)
320
0.00
C) DV153, granodiorite 80 70 60 50 40 30 20 10 0 481216 20
0
Unconstrained Time (Ma) Length (m)
20
40
SONSON BATHOLITH
60
80 F) DV156, granite
Temperature (C)

100 0
120 Mean length 20
140 observed: 14.410.76 m 40
AFT best-fit: 14.600.73 m
160 60
Temperature (C)

Measured: 69.78.6 Ma
180
Best-fit: 68.1 Ma 0.6 N: 86 80
200 GOF: 0.94 100
GOF: 0.71
220 120 Mean length
Apatite (U-Th)/He
ZFT Measured: 30.71.8 Ma observed: 14.950.71 m
240 0.3 140 best-fit: 14.880.91 m
Measured: 66.68.0 Ma Best-fit: 39.2 Ma
260 160 GOF: 0.14
N: 66
280 180 0.50 GOF: 0.27
AFT
300 0.0 200 Measured: 48.24.8 Ma
80 70 60 50 40 30 20 10 0 481216 20 220 Best-fit: 44.7 Ma
Time (Ma) Length (m) GOF: 0.15
240 0.25
260 ZFT
Measured: 47.95.2 Ma
280
300 0.00
600 50 40 30 20 10 481216 20
Time (Ma) Length (m)

Fig. 9. Timetemperature solutions for autochthonous rocks of the northern (~6N) Central Cordillera of Colombia obtained by inverse modelling of apatite FT age and length data,
and weighted mean (UTh)/He dates and grain size data (calculated using the weighted mean of the diffusion lengths), using the kinetic relationship of Reiners et al. (2004) for
diffusion of He in zircon, Flowers et al. (2009) for the diffusion of He in apatite and Ketcham et al. (2007) for FT annealing in apatite. A controlled random search procedure was
used to search for best-t data. Dark grey regions are envelopes for good ts and light grey are acceptable ts. The thick black line shows the statistically best-tting solution.
Measured and predicted data for the best t model are shown. Solutions were considered to be good ts when track length histograms and model ages passed the Kuiper's statistic
test with values of >0.5, and acceptable with values of >0.05. The models are extrapolated to temperatures of partial retention of argon when i) the 40Ar/39Ar ages of ferromag-
nesian phases overlap with the timing of cooling obtained from inverting the FT and (UTh)/He data, or ii) there are interpretable alkali feldspar 40Ar/39Ar ages. Dashed lines high-
light paths that have been manually interpolated using the 40Ar/39Ar data. GOF: Goodness-of-t.
242 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

Mean length
PERMIAN BATHOLITH IBAGUE BATHOLITH observed: 13.421.15 m
Mean length
observed: 13.311.16 m best-fit: 13.051.53 m
best-fit: 13.351.18 m
D. DV05, granodiorite N: 57
A. DV82, granite 0 0.30
GOF: 0.12
0 N: 70 20
20 0.30 GOF: 0.68 40
40 60 0.15

Temperature (C)
60 80
0.15
Temperature (C)

80 100
100 120 0.00
120 140 4 8 12 16 20
0.00 Apatite (U-Th)/He Length (m)
140 4 8 12 16 20 160 Measured: 7.50.5 Ma
Apatite (U-Th)/He Length (m) Best-fit: 12.8 Ma
160 Measured: 11.10.7 Ma 180 GOF: 0.16
180 Best-fit: 10.5 Ma 200
GOF: 0.91 AFT
200 220 Measured: 59.816.8 Ma
220 AFT 240 Best-fit: 46.0 Ma
Measured: 68.38.2 Ma GOF: 0.12
240 260
Best-fit: 68.3 Ma
260 ? GOF: 0.99 280 ZFT
Measured: 85.318.2 Ma
280 300
ZFT 100 80 60 40 20 0
300 Measured: 77.610.8 Ma Time (Ma)
320
Ar (ortho): 137.30.9 Ma
340 E. DV06, granite
140 120 100 80 60 40 20 0 0
AFT
Time (Ma) 20 Measured: 35.84.6 Ma
40 Best-fit: 35.4 Ma Unconstrained
GOF: 0.83
CAJAMARCA FORMATION Mean length 60
observed: 14.950.77 m
80 Zircon (U-Th)/He
B. DV03, gneiss
Temperature (C)
best-fit: 15.170.68 m Measured: 47.72.7 Ma
Temperature (C)

0 100 Best-fit: 44.3 Ma

N: 60
20 120 GOF: 0.84 Mean length
0.50 GOF: 0.94
40 AFT 140 ZFT observed: 13.81 1.41 m
Measured: 37.15.8 Ma
60 Best-fit: 36.9 Ma 160 Measured: 81.310.6 Ma best-fit: 14.06 1.09 m
N: 43
80 GOF: 0.93 0.25 180 Ar (ortho): 109.71.3 Ma 0.30 GOF: 0.58
100 200
120 220
140 0.00 240 0.15
140 120 100 80 60 40 20 0 4 8 12 16 20
Time (Ma) Length (m)
260 ?
280
300 0.00
320 4 8 12 16 20
C. DV09, granite Length (m)
340
0 AFT
20 Measured: 30.65.8 Ma 140 120 100 80 60 40 20 0
Best-fit: 29.5 Ma Time (Ma)
40 GOF: 0.73 Unconstrained
60 Mean length
Zircon (U-Th)/He
80 Measured: 43.03.0 Ma F. DV07, granite observed: 13.820.91 m
0 best-fit: 13.521.00 m
100 Best-fit: 40.4 Ma
Temperature (C)

Mean length 20
GOF: 0.81 0.50 N: 91
120 observed: 14.070.93 m
best-fit: 14.190.93 m 40 GOF: 0.05
140 ZFT
Measured: 74.66.7 Ma 60
160 0.4 N: 63 80
Temperature (C)

180 Ar (ortho): 114.10.9 Ma GOF: 0.61 0.25

100
200
120
220 0.2 140 0.00
240 Apatite (U-Th)/He
160 4 8 12 16 20
260 ? Measured: 6.20.4 Ma
180 Best-fit: 10.7 Ma Length (m)
280 0.0 GOF: 0.07
200
300 4 8 12 16 20
Length (m) 220 AFT
320 Measured: 64.910.2 Ma
240
340 Best-fit: 57.2 Ma
260 GOF: 0.14
140 120 100 80 60 40 20 0 280 ?
Time (Ma) ZFT
300 Measured: 88.312.4 Ma
320
Ar (ortho): 109.82.9 Ma
340
100 80 60 40 20 0
Time (Ma)

Fig. 10. Timetemperature solutions for autochthonous rocks of the Central Cordillera of Colombia at ~4N. See the caption of Fig. 9 for details.

heating steps of orthoclase DV09 (Fig. 4D) suggest that it continued to not been found in the good-t solutions of the remaining samples
cool until ~100 Ma, at a reduced rate. Acceptable-t envelopes for grano- north of the Ibagu Fault.
diorites DV07 and DV05 reveal elevated cooling rates during a poorly Late Eoceneearly Oligocene (4530 Ma) rapid cooling of 9 C/My
constrained period within the Late Cretaceous (Fig. 10D, F), which has is shown by all good-t models derived from north of the Ibagu Fault,
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
and is most precisely dened for granodiorite DV06 during 4540 Ma
(Fig. 10E), at a high rate of ~24 C/My. Several samples located in this
region, and in northern Colombia along the eastern ank of the Central
Cordillera, yield EoceneOligocene apatite FT ages and cooling may
have been a response to displacement of the Palestina, Ibagu and
OtPericos faults (Fig. 2B).
Almost all samples taken north of the Ibagu Fault reveal a period of
elevated cooling at rates of 4 C/My during 150 Ma, which is
more precisely dened by samples that yielded apatite (UTh)/He
data (e.g. Fig. 10A). In contrast, a majority of samples taken from the
Altiplano Antioqueo region of the northern Central Cordillera
(Fig. 2A) did not experience rapid cooling during the Miocene (Fig. 9).
8.2.2. South of the Ibagu Fault
The good-t envelope of solutions for a Permian granite (DV82)
located south of the Ibagu Fault shows rapid cooling from > 250 C
to ~ 60 C during 8070 Ma at high rates of 20 C/My (Fig. 10A). Iso-
thermal conditions prevailed until ~ 1510 Ma, when the sample
cooled at rates of 3 C/My until the present day. Granite DV82
yielded a broad at region on its orthoclase 40Ar/ 39Ar age spectrum
(Fig. 5C), suggesting that it cooled rapidly through 290 60 C during
~ 140135 Ma. Similarly, plateau and broad, at regions on the ortho-
clase 40Ar/ 39Ar age spectra of granodiorites (DV81 and DV84; Fig. 5)
of the Ibagu Batholith, located south of the Ibagu Fault shows that
they also cooled rapidly through 290 60 C during ~ 140135 Ma.
8.2.3. Southernmost Colombia
Thermal history solutions have not been obtained from rocks
located in the southern Central Cordillera because they did not yield ap-
atite FT length data. However, three granitoids of the Jurassic Ibagu
Batholith yielded apatite FT ages which i) are the youngest from the
Central Cordillera (Figs. 2C and 3), and ii) yield a weighted mean age
of ~14 Ma, which is comparable in age with apatite (UTh)/He ages
obtained from central Colombia, and the timing of the onset of cooling
of some samples in central Colombia. The apatite FT ages, when com-
bined with two Pliocene apatite (UTh)/He ages, are diagnostic of
high cooling rates in the southern Central Cordillera since the middle
Miocene. Conversely, the southern Central Cordillera yields the oldest
zircon FT ages, suggesting that it underwent a lower magnitude of
exhumation than more northern regions during ~10015 Ma.
9. Thermal histories of allochthonous rocks of the Western
Cordillera of Colombia
Good-t solutions for the Bolvar Ultramac Complex (Fig. 11A, B),
located along the eastern margin of the Western Cordillera (Fig. 2B)
reveal rapid cooling from > 250 C to 10080 C during 7565 Ma
at rates of ~ 15 C/My, followed by isothermal conditions until
~ 30 Ma. Subsequent, slower cooling (2 C/My) towards the surface
commenced at ~ 30 Ma. The timing of the onset of Late Cretaceous
cooling of this part of the Western Cordillera precisely overlaps
with that found in the central and northern regions of the Central
Cordillera.
Granitoids of the Eocene Mande Batholith cooled rapidly during
5040 Ma (Fig. 11C, D), giving rise to K/Ar hornblende ages of 56
44 Ma (Aspden et al., 1987). We consider the Eocene cooling event
to be a consequence of thermal relaxation following intrusion into
the upper crust. The sampled granitoids of the batholith yield dis-
tinctly different post-Eocene thermal histories (Fig. 11), implying
that it is likely that they reside in separate fault blocks.
10. Interpretation: Crustal exhumation and tectonic evolution
The thermal history solutions reveal periods of rapid cooling of rocks
in the Central and Western Cordilleras since ~140 Ma (Fig. 12). With
the exception of the Mande and Sonsn Batholiths, a comparison of
243
the best-t solutions with crystallisation ages suggests that cooling
was not related to sub-solidus thermal relaxation. Rather, close tempo-
ral correlations between sedimentation in the foreland Magdalena
Valley Basin (Fig. 1; E.g. Moreno et al., 2011; Nie et al., 2010; Saylor
et al., 2012) with high cooling rates in the cordilleras suggests that
cooling was a consequence of exhumation.
A present-day average geothermal gradient of 2025 C/km has
been obtained from wells in the Llanos Basin of Colombia (Bachu
et al., 1995; Fig. 1). Rock advection occurs at higher rates than heat
conduction during high exhumation rates, causing isothermal spacing
to reduce and an increase in geothermal gradients of up to 60%
(Mancktelow and Grasemann, 1997). Therefore, we assumed a con-
stant MesozoicCenozoic geothermal gradient of 30 C/km, with at
isotherms, similar to that used in previous thermochronological anal-
yses of the northern Andes (e.g. Mora et al., 2010; Spikings et al.,
2000, 2010), and a mean surface temperature of 20 C.
10.1. 145130 Ma: BerriasianHauterivian
Exhumation of the Central Cordillera during 138130 Ma (Figs. 5
and 10a; plateau alkali feldspar ages of 130138 Ma) slightly post-
dates the termination of the Jurassic continental arc (180145 Ma;
Villagmez et al., 2011a; E.g. Ibague Batholith; Fig. 1), and coincides
with a region-wide angular unconformity in the Upper Magdalena
Valley Basin (Fig. 1) in Colombia (Jaimes and de Freitas, 2006), the
Oriente Basin in Ecuador (Balkwill et al., 1995) and in northern
Peru (Jaillard et al., 1990), which spans ~140130 Ma (Fig. 13A, B).
Litherland and Aspden (1992) proposed that a major period of rock
uplift occurred in the Eastern Cordillera of Ecuador (Fig. 1) during
140120 Ma. The cessation of arc magmatism and subsequent gap in
sedimentation were previously accounted for by i) the accretion of
allochthonous terranes (e.g. Litherland and Aspden, 1992), and
ii) changes of plate motion kinematics (Jaillard et al., 1990). Exhuma-
tion may have been a consequence of compression, driven either by
the advance of the upper plate towards the oceanic plate, or the colli-
sion of an oceanic seamount (Fig. 13B), although clear evidence for an
oceanic terrane in Colombia with an age older than ~140 Ma has not
been reported. Alternatively, exhumation and interruption of the Juras-
sic arc may have occurred during extension. Mantle-derived mac rocks
have been identied in the Eastern Cordillera of Colombia, which yield
hornblende 40Ar/39Ar ages of 136120 Ma and were emplaced in a rift
setting (Vsquez et al., 2010). Early Cretaceous crustal thinning and
rifting in eastern Colombia have also been proposed in the tectonic
models of Sarmiento-Rojas et al. (2006) and Pindell and Kennan (2009).
We propose that the termination of arc magmatism was a conse-
quence of oceanward migration of the subduction zone, which accounts
for the outboard position of the younger Quebradagrande Complex
(114.3 3.8 Ma; UPb zircon; Villagmez et al., 2011a; Fig. 1), which
fringed the continental margin. Arc migration may have been driven
by a change in plate kinematics, which drove trench roll-back, exten-
sion and exhumation of the continental margin.
10.2. 130100 Ma: BarremianAlbian
Elevated exhumation rates during ~117107 Ma (Fig. 10C, E and F;
alkali feldspar 40Ar/39Ar ages of 117107 Ma) were contemporaneous
with the deposition of the upper Aptian to middle Albian, uvial
estuarine Caballos Fm. (Fig. 13C) that is exposed in the Upper Magdalena
Valley Basin (Fig. 1; Sarmiento and Rangel, 2004). Dense mineral assem-
blages and detrital zircon FT ages suggest that this sequence, along with
equivalent rocks in Ecuador (Holln Fm.) were partly derived from a
proto-Cordillera that was exhuming in the Early Cretaceous (Kennan
and Pindell, 2009; Ruiz et al., 2007). Vergara and Prssl (1994) describe
the stratigraphically underlying Barremianearly Aptian Yav Fm. and
suggest that it was derived from a cordillera located to the west. A contin-
ual supply of clastic detritus during 130100 Ma suggests that protracted
244 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

BOLIVAR ULTRAMAFIC COMPLEX MANDE BATHOLITH

0
A) DV94, pegmatite C) DV165, diorite
0
20 20
40 Unconstrained 40 Unconstrained
60 60

Temperature (C)
Temperature (C)

80 80
100 100
120 Mean length 120 Mean length
observed: 13.321.17 m 140 observed: 14.451.09 m
140
best-fit: 14.181.37 m best-fit: 14.730.97 m
160 160
180 AFT 0.4 N: 31 180 AFT 0.50 N: 52
Measured: 61.59.4 Ma GOF: 0.58 Measured: 37.85.2 Ma GOF: 0.94
200 Best-fit: 59.9 Ma 200 Best-fit: 37.1 Ma
220 GOF: 0.87 220 GOF: 0.82
240 ZFT 0.2 240 ZFT 0.25
260 Measured: 67.97.4 Ma 260 Measured: 41.65.4 Ma
280 280
300 0.0 300 0.00
90 80 70 60 50 40 30 20 10 0 4 8 12 16 20 60 50 40 30 20 10 0 4 8 12 16 20
Time (Ma) Length (m) Time (Ma) Length (m)

0
B) DV95, pegmatite 0
D) DV167, granodiorite
20 20
40 Unconstrained 40 Unconstrained
60 60
Temperature (C)
Temperature (C)

80 80
100 100
120 Mean length 120 Mean length
140 observed: 13.801.28 m 140 observed: 13.470.82 m
AFT best-fit: 14.251.30 m best-fit: 13.680.89 m
160 160 AFT
Measured: 62.86.8 Ma N: 42 Measured: 38.15.4 Ma N: 116
180 Best-fit: 63.3 Ma 0.30 180 0.4 GOF: 0.98
GOF: 0.98 Best-fit: 38.2 Ma
200 GOF: 0.88 200 GOF: 0.97
220 ZFT 220
ZFT
240 Measured: 62.55.2 Ma 0.15 240 Measured: 46.45.8 Ma 0.2
260 260
280 280
300 0.00 300 0.0
80 70 60 50 40 30 20 10 0 4 8 12 16 20 60 50 40 30 20 10 0 0 4 8 12 16 20
Time (Ma) Length (m) Time (Ma) Length (m)

Fig. 11. Timetemperature solutions for allochthonous rocks of the Western Cordillera of Colombia. See the caption of Fig. 9 for details.

exhumation continued throughout the Early Cretaceous, although the
thermochronological data suggests that it occurred at its highest rates
during ~117107 Ma.
To the west, the Aptianmiddle Albian Abejorral Fm. was deposited
within a Hautevarianearly Albian (Nivia et al., 2006) basin that may
have been formed by either crustal exure or back-arc extension
(Fig. 13C), and was derived from continental crust to the east
(Restrepo et al., 2009), possibly as a consequence of erosion of a
proto-Central Cordillera.
High exhumation rates during 117107 Ma are interpreted to have
been driven by closure of the Quebradagrande back-arc basin and colli-
sion of the Quebradagrande Arc (Villagmez et al., 2011a) with the
continental margin along the San Jernimo Fault (Figs. 1 and 13D).
The same collision event may have been responsible for exhuming
highmedium PT metamorphic rocks of the Arqua Complex (Figs. 1
and 2), forcing them to cool through ~500 C (hornblende 40Ar/39Ar
age of 112.0 3.7 Ma) and emplace along the SilviaPijao Fault.
The Arqua Complex may be the along-strike equivalent of high PT
metamorphic rocks of the Raspas Complex in Ecuador (Fig. 1), which
yield peak metamorphic ages of ~130 Ma (Lu/Hf garnet; John et al.,
2010). Similar ages (140130 Ma; plagioclase 40Ar/39Ar; Spikings
et al., 2010, unpublished data) have been obtained from gabbros of
the Peltetec Unit in Ecuador (Fig. 1), which may be the along-strike,
non-metamorphosed protolith of the Arqua Complex. Geochemical
analyses of gabbros of the Peltetec Unit (Spikings et al., 2010,
unpublished data), high-P rocks of the Raspas Complex (John et al.,
2010) and highmedium PT metamorphic rocks of the Arqua Com-
plex (Villagmez et al., 2011a) yield evidence for both MORB and OIB
rocks.
On a more regional scale, the retrogression ages of highmedium
PT metamorphic rocks exposed along the northern Andes overlap
with metamorphic zircon ages (116106 Ma; Maresch et al., 2009)
obtained from anatectites within HP-LT metagabbros at La Rinconada,
Margarita Island, southern Caribbean. Those rocks are interpreted as
back-arc basin crust that was subducted and retrogressed during
closure of the back-arc basin during 116100 Ma. The coincidence in
the timing of metamorphism recorded in the Arqua Complex, and on
Margarita Island suggests the rocks may have formed in the same tec-
tonic setting. However, Maresch et al. (2009) suggest that the HP-LT
rocks at Margarita formed on a west dipping slab of continental and
back-arc crust, which is inconsistent with the spatial relationship in
Colombia and Ecuador, where the relict slab represented by the
RaspasPeltetecArqua complex is consistently located to the west of
the arc rocks.
Amphibolite DV28 is currently mapped as part of the Arqua Com-
plex, and yields a plateau 40Ar/ 39Ar age of 64.4 3.6 (Fig. 7C). We inter-
pret this age as either i) retrogression of a sliver of the Arqua Complex
during the collision of the CLIP in the CampanianLower Tertiary, or
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249 245

Increased Batholiths; Fig. 2A, B), suggesting that the Caribbean Large Igneous
Onset of convergence
accretion of SOAM and Subduction of the Province collided with South America after 80 Ma.
of CLIP Farallon plates Carnegie Ridge
10.4. CampanianMaastrichtian

The best-t thermal history solutions for several faulted blocks of

0 North of 5N the Central Cordillera reveal a distinct period of rapid cooling during
40 DV54 7565 Ma (Fig. 12). This was synchronous with i) deposition of the
DV153
arenaceous El Cobre Fm. during the Campanian (Fig. 13G), and the
80
Maastrichtian Monserrate, La Tabla and Cimarrona Fms (Villamil,
Temperature (C)

120
DV70 1999) within a retro-foreland position (Magdalena Valley; Fig. 1),
160 ii) deposition of the CampanianMaastrichtian Nogales Fm. along
200 the western border of the growing Central Cordillera within the
240 forearc, which hosts a large proportion of metamorphic grains that
were derived from the Central Cordillera (Moreno and Pardo, 2003).
280
Clearly, rapid cooling of the buttressing rocks during 7565 Ma was
320 a consequence of rapid exhumation.
The onset of exhumation in the Campanian was driven by the colli-
500
sion of the Caribbean Large Igneous Province with the South American
540 Plate along the CaucaAlmaguer Fault (Fig. 13G). The collision event
South of 5N drove rock and surface uplift resulting in erosional exhumation
0
Western Cordillera that persisted until ~65 Ma. Previous intensive thermochronological
40 DV95
studies of the Late Cretaceous buttressing margin of Ecuador (Eastern
Temperature (C)

Central Cordillera
80 Cordillera; Fig. 1) revealed dramatically increased cooling rates at
DV94
120 7370 Ma (Spikings et al., 2010), which in combination with palaeo-
160 magnetic and sedimentological data is considered to record the timing
of collision of the Caribbean Large Igneous Province with the Ecuadorian
200
DV06 margin (Luzieux et al., 2006; Vallejo et al., 2006, 2009). Average exhu-
South of
240 Ibague Fault mation rates in the Central Cordillera of Colombia were ~1.6 km/My
DV09
280 during 7565 Ma, which are similar to an average rate of >1 km/My
320
North of (Spikings et al., 2001, 2010) obtained from the Eastern Cordillera of
Ibague Fault
Ecuador during the same time period.
140 120 100 80 60 40 20 0
Time (Ma)
10.5. Paleocenemiddle Eocene
Fig. 12. Summary of good-t thermal history solutions for a representative selection of
samples of the Central Cordillera (Late Cretaceous continental margin; light grey) and Exhumation of the buttressing and indenting margins mainly
the Western Cordillera (Late Cretaceous indentor; dark grey) that are shown in Figs. 9 ceased after 7065 Ma, although it may have continued at a signi-
11. These samples highlight the main periods of exhumation of the Central and Western cantly slower rate during 6545 Ma within specic faulted blocks
Cordillera. Not all samples experienced distinct periods of cooling and exhumation (e.g. (e.g. granodiorite DV70 and gneiss DV65; Fig. 9). Subduction of the
gneiss DV65; Fig. 9), presumably due to the orientation of faulted blocks relative to the
local stress eld. Vertical bands highlight the timing of rapid cooling and exhumation in
Faralln Plate beneath the South American Plate occurred subsequent
Colombia, and labels are sample numbers. CLIP: Caribbean Large Igneous Province. to the collision, accretion and fragmentation of the Caribbean Large
Igneous Province, forming a volcanic arc which includes the Sonsn
Batholith at 6555 Ma (Fig. 13H; Ordez-Carmona et al., 2001).

10.6. Middle Eocenelate Oligocene

ii) retrogression of a younger protolith, and DV28 does not form part of
the Arqua Complex.
10.3. 10075 Ma: Late Albian to middle Campanian
Subduction subsequent to the collision and accretion of the
Quebradagrande Arc and highmedium PT metamorphic rocks formed
a continental arc in Colombia (Fig. 13E; Villagmez et al., 2011a), which
included i) the large Antioquia Batholith (zircon UPb 9585 Ma;
Villagmez et al., 2011a), and ii) scattered plutons including the
Crdoba pluton (zircon UPb 79.7 2.5 Ma; Villagmez et al., 2011a;
Fig. 2B).
Plateau rocks of the Caribbean Large Igneous Province erupted off-
shore during 10090 Ma (Fig. 13E; Kennan, 2009; Pindell and Sinton
et al., 1998; Vallejo et al., 2009; Villagmez et al., 2011a), and
approached Colombia from the west. Oceanic plateau rocks were
intruded by an east-facing island arc (e.g. Vallejo et al., 2009), resulting
in the emplacement of the intra-oceanic Buga Batholith at 9290 Ma
(Villagmez et al., 2011a). Pre-accretionary magmatic arcs (Fig. 13F)
erupted through both the Caribbean Plateau (Buga Batholith and the
Volcanic Fm; Fig. 2B) and South America (Antioquia and Crdoba
A Palaeozoic metasedimentary rock and two Jurassic granites yield
high cooling and exhumation rates (9 C/My; ~ 0.3 km/My) during
4530 Ma with an initial higher exhumation rate during 4540 Ma
of ~ 0.8 Km/My (Fig. 10) within the Central Cordillera, proximal to
the Palestina, Ibagu and OtPericos Faults. Faulted blocks of the
Altiplano Antioqueo exhumed at a slow but continuous rate of
~0.14 km/My during 4530 Ma (Fig. 9A, C, D).
Middle Eocenelate Oligocene exhumation has been previously
identied in discrete faulted blocks of the Altiplano Antioqueo, based
on apatite (UTh)/He data (Restrepo-Moreno et al., 2009), which
yielded cooling rates of 2 to 6 C/My during 4540 Ma and ~2025 Ma.
Exhumation of the Central Cordillera during 4530 Ma was contem-
poraneous with the deposition of alluvial sedimentary rocks (Hoyn
Fm. and Gualanday Group; Jaimes and De Freitas, 2006; Villamil,
1999) within the Middle and Upper Magdalena Valley, foreland basins
(Fig. 1).
Thermochronological analyses within the Sub-Andean Zone and the
cordilleras of Ecuador (Fig. 1) reveal rapid cooling and exhumation
during 4330 Ma, with the highest rates occurring during 4335 Ma
(Spikings et al., 2010), which was synchronous with the deposition of
246 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249

A) 180-145 Ma W Ibagu Batholith E

Continental arc volcanism

Exhumation
Palaeozoic and
Triassic
basement

B) 145-130 Ma Collision, compression, outboard migration of

subduction zone
Roll-back and extension in the continent, Proto-Central Major
OR Cordillera Unconformity
migration of arc axis unidentified
Major Unconformity
Roll-back seamount?

C) 130-115 Ma Intra-oceanic arc forms close to the continent

Quebradagrande Major Quebradagrande arc Abejorral Fm Caballos Fm
arc Unconformity OR
Arqua Complex

su
bd
uc
tio
ns
top
s
SPF
D) 115-105 Ma Arqua Complex SJF OPF
Closure of the Quebradagrande back-arc basin
Accretion of the Quebradagrande arc
Obduction and accretion of high and medium P rocks
Rock uplift and exhumation of the continent

Antioquia Batholith
E) 95-90 Ma Arqua Complex
Onset of continental arc magmatism
Formation of the CLIP

Crdoba Batholith
F) 90-80 Ma Santa Marta
CLIP converges towards South America Buga Batholith Schists

G) 75-70 Ma Central Cordillera

Nogales Fm. El Cobre Fm.
Oblique collision of the CLIP with South America +
High rates of rock uplift and exhumation CAF
in the Central Cordillera

Sonsn Batholith
H) ~60 Ma
Active continental margin continues
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
conglomerates in the Oriente Basin. Elsewhere within South America,
FT analyses of the Eastern Cordilleras of Per and Bolivia (Barnes
et al., 2006) also record elevated cooling rates between 43 and 30 Ma,
implying that tectonic reactivation affected >2000 km of the South
American Pacic margin. An increase in convergence rates between
the Faralln and South American plates during 4237 Ma (Pardo-
Casas and Molnar, 1987) may be responsible for increased compres-
sion of the western South American Plate, driving rock uplift and
exhumation.
10.7. OligocenePliocene
No evidence of a signicant Oligocene exhumation event has been
detected in the Western and Central Cordilleras of Colombia (Fig. 12).
Spikings et al. (2010) proposed that fragmentation of the Faralln
Plate and the subsequent change in the subducting ocean plate
vectors at 25 Ma led to cooling and rapid exhumation in the Eastern
Cordillera of Ecuador. The absence of Oligocene cooling and exhuma-
tion in the Western and Central Cordilleras of Colombia may be a con-
sequence of stress partitioning, which preferentially reactivated the
Amazonian Border Fault System and the Santa MartaBucaramanga
Fault (Fig. 1), deforming and exhuming the Eastern Cordillera of
Colombia (Mora et al., 2010; Parra et al., 2012; Saylor et al., 2012) and
uplifting the Sierra Nevada de Santa Marta (Villagmez et al., 2011b),
isolating the Central and Western Cordilleras.
Apatite FT and (UTh)/He data from the Colombian Central Cordillera
and the Ecuadorian Eastern Cordillera (Spikings et al., 2001) reveal
distinct along strike trends in cooling and exhumation since 15 Ma
(Fig. 12). Faulted blocks of the Central Cordillera in northern Colombia
(~6N) exhumed from a depth of 1 km since ~15 Ma (Fig. 9B). Howev-
er, exhumation from a depth of 1.5 km since 15 Ma is observed across
the Central Cordillera at mid-latitudes (~4N; Fig. 10). Apatite FT ages of
~15 Ma were acquired from samples that span an elevation difference of
630 m within southernmost (~1N) Colombia, revealing a period of
middle Miocene rapid cooling and exhumation. These samples yield
the youngest apatite FT ages (~188 Ma) and the youngest apatite (U
Th)/He ages (~124 Ma), from the Central Cordillera of Colombia
(Fig. 3), suggesting that these samples exhumed from a depth of 3
1.5 km since ~15 Ma.
Spikings et al. (2001) identied a northward younging, along-strike
trend of exhumation within the Eastern Cordillera of Ecuador, where
the amount of exhumation since 15 Ma increases from 1.3 km at
~4S to 3.5 km at 0.5N. Those authors proposed that the
Carnegie Ridge collided with northwestern South America at 15 Ma,
which drove the coastal forearc towards the northnortheast forcing
transpression in northern Ecuador. Their explanation is consistent
with the timing and trend of exhumation depth in the Central Cordillera
of Colombia during 150 Ma, and is consistent with the current location
of the Carnegie Ridge beneath the EcuadorColombia border (Fig. 1).
Duque-Caro (1990) proposes that the ChocPanam terrane collided
with north-western Colombia at ~13 Ma (although Early and Late
Miocene dates have also been proposed for the timing of collision). If
that interpretation is accurate, then the collision event may have also
increased compressive stress and exhumed the crust, although it does
not easily account for the along-strike exhumation trend.
Several best-t thermal history solutions reveal rapid cooling
commencing at 105 Ma within scattered regions of the Central
Cordillera (Figs. 911), which we interpret to be a consequence of
exhumation of specic faulted massifs. Late Miocene exhumation
has been identied within the northern Western Cordillera
(Fig. 11D; Mande Batholith; DV167), although it has not been
Fig. 13. Tectonic reconstruction of Western Colombia during 18060 Ma. Thick, black arrows are indicative of exhumation determined from the thermochronological data in combination
with the stratigraphic record of the foreland basins. CAF: CaucaAlmaguer Fault, OPF: Oto-Pericos Fault, SJF: San-Jeronimo Fault, SPF: SilviaPijao Fault.
247
constrained by apatite (UTh)/He data and is tentative. Compression-
al events at ~ 9 Ma have been identied within Venezuela, Ecuador,
Peru, Bolivia and northern Argentina (see review in Spikings et al.,
2010), which suggests that they may have been driven by an increase
in the velocity of westward drift of South America, possibly driven
due to increased half-spreading rates in the equatorial Atlantic at
~ 98 Ma (Brozena, 1986).
10.8. Reactivation of the Ibagu Fault
Contrasting exhumation histories experienced by rocks located
north and south of the Ibagu Fault (Figs. 2B and 12) that dissects the
Central Cordillera suggest it was active with a vertical component
during the Early Cretaceous. Jurassic granitoids located south of the
fault exhumed at elevated rates during 138130 Ma, which post-dates
the termination of Jurassic arc magmatism by ~10 My. However, the
earliest exhumation event recorded in the same arc, north of the Ibagu
Fault occurred at 117107 Ma, and the earlier Cretaceous event is not
recorded within the temperature sensitivity realm of the data. Both re-
gions may have exhumed during 138130 Ma, although only the north-
ern wall of the Ibagu Fault was exhumed during the closure of the
back-arc basin to the Quebradagrande Arc. Using the thermal history
solutions (Fig. 10), we estimate a vertical displacement of 2.5 km
along the Ibagu Fault during 117107 Ma. Corts et al. (2006) suggest
that the Ibagu Fault is a major crustal structure that underwent
~30 km of dextral displacement during the Cenozoic. The fault sepa-
rates the Magdalena Basin into the upper and middle sub-basins
(Villamil, 1999), which have signicant differences in their structural
architecture (Corts et al., 2006). Our thermochronological data suggest
that this structure has been active since at least the late Aptian.
11. Conclusions
1. The earliest cooling event recorded in the buttressing continental
margin of Colombia occurred at 138130 Ma, which post-dates con-
tinental arc magmatism by 510 Ma. A contemporaneous major
unconformity is observed along the entire retro-foreland of the
northern Andes (Colombia and Ecuador), providing evidence for a
regional tectonic event. We ascribe this event to exhumation driven
by oceanward migration of the Jurassic subduction zone, which
formed the Early Cretaceous Quebradagrande Complex in an out-
board position. Oceanward migration of the subduction zone was
probably a result of roll-back of the trench due to changing plate
kinematics.
2. The Quebradagrande Complex accreted onto the continental mar-
gin during ~ 117107 Ma forming the San Jernimo Fault, and its
subduction channel (e.g. the Arqua Complex) accreted onto the
Quebradagrande Complex along the SilviaPijao Fault. The tectonic
mechanism for closure of the back-arc basin is unknown, although
given the likely proximity of the arc to continental crust (Fig. 13),
we suggest that it may have occurred by thrusting in the absence
of subduction zones between the arc and South American crust.
These collisions drove exhumation in distinct regions of the
Central Cordillera, and reactivated major tectonic structures such
as the Ibagu Fault. Retrogression of the M-HP/LT rocks was syn-
chronous with exhumation observed in the buttressing continen-
tal margin.
3. The Late Cretaceous Caribbean Large Igneous Province (Calima
Terrane) accreted to the Colombian margin during the Campanian
(~7570 Ma), forming the CaucaAlmaguer Fault. The collision
drove exhumation rates of ~ 1.6 km/My during 7565 Ma in
248 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
northern Colombia as an immediate response to collision and
accretion. CampanianMaastrichtian exhumation corroborates
the sudden increase in sedimentation rate in the forearc (Nogales
Fm.) and the retro-foreland (El Cobre Fm.).
4. Cooling and exhumation rates in central latitudes of the Central
Cordillera suddenly increased during ~ 4530 Ma (9 C/My;
~ 0.3 km/My), with the highest rates occurring during 4540 Ma
(~ 0.8 km/My). Low exhumation rates are observed at that time
in northern Colombia (~ 0.2 km/My). Exhumation during this
period is sporadically observed along the entire Andean chain
and was contemporaneous with the deposition of conglomerates
within the Amazon Foreland Basin, which were derived from the
west. The driving forces were provided by an increase in the
convergence rate between South America and the Faralln plate
(e.g. Spikings et al., 2010).
5. No evidence for elevated cooling and exhumation during the
Oligocene has been observed in the Western and Central Cordilleras
of Colombia, which is contrary to the exhumation history of Ecuador
(Spikings et al., 2010) and in the Sierra Nevada de Santa Marta at
the boundary between the South American and Caribbean Plates
(Villagmez et al., 2011b). We propose that activation/reactivation
of favourably oriented faults such as the Amazonian Border Fault
System and the Santa MartaBucaramanga Fault partitioned com-
pression away from the rocks of the Western and Central Cordilleras.
6. Elevated exhumation rates that exhumed rocks from depths of
3 km since ~15 Ma are observed in the southern Central Cordillera
of Colombia. Similar middle MioceneRecent exhumation rates have
been observed in northern Ecuador (Spikings et al., 2001), and the
spatial relationship between the exhumed region and the subducted
extent of the Carnegie Ridge suggests that collision and subduction
of the ridge since ~15 Ma drove rock uplift and exhumation.
Supplementary data to this article can be found online at http://
dx.doi.org/10.1016/j.lithos.2012.12.008.
Acknowledgements
The authors acknowledge the assistance provided by Prof. Andreas
Kammer, Dr. Andrs Mora and Luis Quiroz during eld work in
Colombia. Useful suggestions were provided by Dr. James Pindell
and David Chew, who examined the PhD thesis of Diego Villagomez.
Funds for the project were awarded to RS by the Swiss NSF (nos.
200021100079 and 200021103438).
References
Alvarez, J.A., Linares, E., 1979. Edades K-Ar del Batolito de Piedrancha y Stock de
Arboledes, Departmento de Narino. Publicacion Especial Geologia, 22. Universidad
Nacional de Medellin, Colombia (15 pp.).
Andriessen, P.A.M., 1996. Proterozoic to Recent thermotectonic history of the Garzon
Massif (Eastern Cordillera of the Colombian Andes) and the evolution of the
Magdalena Valley basins. Proceedings, International Workshop on Fission-Track
Dating, Gent, p. 1.
Aspden, J.A., McCourt, W.J., Brook, M., 1987. Geometrical control of subduction-related
magmatism: the Mesozoic and Cenozoic plutonic history of Western Colombia.
Journal of the Geological Society 144, 893905.
Bachu, S., Ramon, J.C., Villegas, M.E., Underschultz, J.R., 1995. Geothermal regime and
thermal history of the Llanos Basin, Colombia. American Association of Petroleum
Geologists Bulletin 79, 116129.
Balkwill, H.R., Paredes, F.I., Rodriguez, G., Almeida, J.P., 1995. Northern part of Oriente
Basin, Ecuador: reection seismic expression of structures. In: Tankard, A.J., Suarez, R.,
Welsink, H.J. (Eds.), Petroleum Basins of South America: American Association of
Petroleum Geologists, 62, pp. 559571.
Barnes, J.B., Ehlers, T.A., McQuarrie, N., O'Sullivan, P.B., Pelletier, J.D., 2006. Eocene to recent
variations in erosion across the central Andean fold-thrust belt, northern Bolivia:
implications for plateau evolution. Earth and Planetary Science Letters 248, 118133.
Bernet, M., Garver, J.I., 2005. Fission-track analysis of detrital zircon. Reviews in Miner-
alogy and Geochemistry 58, 205237.
Brandon, M.T., 1996. Probability density plot for ssion-track grain-age samples. Radi-
ation Measurements 26, 663679.
Brandon, M.T., Roden-Tice, M.K., Garver, J.I., 1998. Late Cenozoic exhumation of the
Cascadia accretionary wedge in the Olympic Mountains, northwest Washington
State. Geological Society of America Bulletin 110, 9851009.
Brook, M., 1984. New radiometric age data from SW Colombia. Ingeominas-Mision
Britanica, British Geological Survey, Colombia report (10 pp.).
Brozena, J.M., 1986. Temporal and spatial variability of seaoor spreading processes in
the Northern South Atlantic. Journal of Geophysical Research 91, 497510.
Bustamante, A., Juliani, C., Hall, C.M., Essene, 2011. 40Ar/39Ar ages from blueschists of
the Jambalo region, Central Cordillera of Colombia: implications on the styles of
accretion in the Northern Andes. Geological Acta 9, 351362.
Cardona, A., Valencia, V., Garzn, A., Montes, C., Ojeda, G., Ruiz, J., Weber, M., 2010.
Permian to Triassic I to S-type magmatic switch in the northeast Sierra Nevada
de Santa Marta and adjacent regions, Colombian Caribbean: tectonic setting and
implications within Pangea paleogeography. Journal of South American Earth
Sciences 29, 772783.
Cassata, W.S., Renne, P.R., Shuster, D.L., 2009. Argon diffusion in plagioclase and impli-
cations for thermochronometry: a case study from the Bushveld Complex, South
Africa. Geochemica et Cosmochimica Acta 73, 66006612.
Cochrane, R.S., Spikings, R.A., Winkler, W., Ulianov, A., Chiaradia, M., 2011. Tectonic
evolution of the Ecuadorian margin during the TriassicEarly Cretaceous: a geo-
chronological and geochemical study. 14th Congreso Latinoamericano de Geologa,
Medellin, p. 2.
Corts, M., Colletta, B., Angelier, J., 2006. Structure and tectonics of the central segment
of the Eastern Cordillera of Colombia. Journal of South American Earth Sciences 21,
437465.
Dalrymple, G.B., Lanphere, Marvin A., 1974. 40Ar/39Ar age spectra of some undisturbed
terrestrial samples. Geochimica et Cosmochimica Acta 38, 715738.
De Souza, H.A.F., Espinosa, A., Delaloye, M., 1984. KAr ages of basic rocks in the Patia
Valley, southwest Colombia. Tectonophysics 107, 135145.
DeJong, K., Feraud, G., Ruffet, G., Amouric, M., Wijbrans, J.R., 2001. Excess argon incor-
poration in phengite of the Mulhacen Complex: submicroscopic illitization and
uid ingress during late Miocene extension in the Betic Zone, south-eastern
Spain. Chemical Geology 178, 159195.
Donelick, R.A., Miller, D.S., 1991. Enhanced tint ssion track densities in low spontane-
ous track density apatites using 252Cf-derived ssion fragment tracks: a model
and experimental observations. Nuclear Tracks and Radiation Measurements 18,
301307.
Duque-Caro, H., 1990. The Choco Block in the northwestern corner of South America:
structural, tectonostratigraphic, and paleogeographic implications. Journal of
South American Earth Sciences 3, 7184.
Farley, K.A., Stockli, D.F., 2002. (UTh)/He dating of phosphates: apatite, monazite, and
xenotime. In: Kohn, M., Rakovan, J., Hughes, J.M. (Eds.), Phosphates. Reviews in
Mineralogy and Geochemistry, 15, pp. 559578.
Farris, D.W., Jaramillo, C., Bayona, G., Restrepo-Moreno, S.A., Montes, C., Cardona, A.,
Mora, A., Speakman, R.J., Glascock, M.D., Valencia, V., 2011. Fracturing of the Panamanian
Isthmus during initial collision with South America. Geology 39, 10071010.
Flowers, R.M., Ketcham, R.A., Shuster, D.L., Farley, K.A., 2009. Apatite (UTh)/He
thermochronometry using a radiation damage accumulation and annealing
model. Geochimica et Cosmochimica Acta 73, 23472365.
Gleadow, A.J.W., 1981. Fission-track dating methods: what are the real alternatives?
Nuclear Tracks 5, 314.
Gleadow, A.J.W., Duddy, I.R., Green, P.F., Lovering, J.F., 1986. Conned ssion track
lengths in apatite: a diagnostic tool for thermal history analysis. Contributions to
Mineralogy and Petrology 94, 405415.
Gmez, J., Nivia, A., Montes, N.E., Jimenez, D.M., Tejada, M.L., Sepulveda, J., Osorio, J.A.,
Gaona, T., Diederix, H., Uribe, H., Mora, M., 2007. Geological map of Colombia.
Ingeominas, 2nd Edition, Bogota, 1:1'000.000.
Green, P.F., 1981. A new look at statistics in ssion-track dating. Nuclear Tracks 5,
7786.
Hasebe, N., Barbarand, J., Jarvis, K., Carter, A., Hurford, A.J., 2004. Apatite ssion-track
chronometry using laser ablation ICP-MS. Chemical Geology 207, 135145.
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, 10091022.
Jaimes, E., de Freitas, M., 2006. An AlbianCenomanian unconformity in the northern
Andes. Evidence and tectonic signicance. Journal of South American Earth Sciences
21, 466492.
John, T., Scherer, E., Schenk, V., Herms, P., Halama, R., Garbe-Schnberg, D., 2010.
Subducted seamounts in an eclogite-facies ophiolite sequence: the Andean Raspas
Complex, SW Ecuador. Contributions to Mineralogy and Petrology 159, 265284.
Kennan, L., Pindell, J.L., 2009. Dextral shear, terrane accretion and basin formation in
the Northern Andes: best explained by interaction with a Pacic-derived Caribbean
Plate? Geological Society, London Special Publications 328, 487531.
Kerr, A.C., Marriner, G.F., Tarney, J., Nivia, A., Saunders, A.D., Thirlwall, M.F., Sinton,
C.W., 1997. Cretaceous basaltic terranes in Western Colombia: elemental, chrono-
logical and SrNd isotopic constraints on petrogenesis. Journal of Petrology 38,
677702.
Kerr, A.C., Tarney, J., Kempton, P.D., Pringle, M., Nivia, A., 2004. Mac pegmatites
intruding oceanic plateau gabbros and ultramac cumulates from Bolivar, Colombia:
evidence for a wet mantle plume? Journal of Petrology 45, 18771906.
Ketcham, R.A., Carter, A., Donelick, R.A., Barbarand, J., Hurford, A.J., 2007. Improved
modeling of ssion-track annealing in apatite. American Mineralogist 92, 799810.
Litherland, M., Aspden, J.A., 1992. Terrane-boundary reactivation: a control on the evo-
lution of the Northern Andes. Journal of South American Earth Sciences 5, 7176.
Lonsdale, P., Klitgord, K.D., 1978. Structure and tectonic history of the eastern Panama
Basin. Geological Society of America Bulletin 89, 981999.
 D. Villagmez, R. Spikings / Lithos 160161 (2013) 228249
Lovera, O.M., Richter, F.M., Harrison, T.M., 1991. Diffusion domains determined by 39Ar
released during step heating. Journal of Geophysical Research 96, 20572069.
Luzieux, L.D.A., Heller, F., Spikings, R., Vallejo, C.F., Winkler, W., 2006. Origin and Cretaceous
tectonic history of the coastal Ecuadorian forearc between 1N and 3S: paleomagnetic,
radiometric and fossil evidence. Earth and Planetary Science Letters 249, 400414.
Mancktelow, N.S., Grasemann, B., 1997. Time-dependent effects of heat advection and
topography on cooling histories during erosion. Tectonophysics 270, 167195.
Mann, P., Corrigan, J., 1990. Model for late Neogene deformation in Panama. Geology
18, 558562.
Maresch, W.V., Kluge, R., Baumann, A., Pindell, J.L., Krchans-Leuder, G., Stanek, K.,
2009. The occurrence and timing of high-pressure metamorphism on Margarita
Island, Venezuela: a constraint on CaribbeanSouth American interaction. Geological
Society of London Special Publications 328, 705741.
McCourt, W.J., Aspden, J.A., Brook, M., 1984. New geological and geochronological data
from the Colombian Andes: continental growth by multiple accretion. Journal of
the Geological Society 141, 831845.
McDougall, I., Harrison, T., 1999. Geochronology and Thermochronology by the 40Ar/
39Ar Method, second ed. Oxford University Press, New York.
Moni, P., Torres-Roldn, R.L., Garca-Casco, A., 1994. Cooling and exhumation of
the Western Betic Cordilleras, 40Ar/39Ar thermochronological constraints on a
collapsed terrane. Tectonophysics 238, 353379.
Mora, A., Horton, B.K., Mesa, A., Rubiano, J., Ketcham, R.A., Parra, M., Blanco, V., Garcia,
D., Stockli, D.F., 2010. Migration of Cenozoic deformation in the Eastern Cordillera
of Colombia interpreted from ssion track results and structural relationships: im-
plications for petroleum systems. American Association of Petroleum Geologists
94, 15431580.
Moreno, M., Pardo, A., 2003. Stratigraphical and sedimentological constraints on Western
Colombia: implications on the evolution of the Caribbean plate. In: Bartolini, C.,
Bufer, R.T., Blickwede, J. (Eds.), The Circum-Gulf of Mexico and the Caribbean:
Hydrocarbon habitats, basin formation, and plate tectonics: American Association
of Petroleum Geologists Memoir, 79, pp. 891924.
Moreno, C.J., Horton, B.K., Caballero, V., Mora, A., Parra, M., Sierra, J., 2011. Depositional
and provenance record of the Paleogene transition from foreland to hinterland
basin evolution during Andean orogenesis, northern Middle Magdalena Valley
Basin, Colombia. Journal of South American Earth Sciences 32, 246263.
Nie, J., Horton, B.K., Mora, A., Saylor, J.E., Housh, T.B., Rubiano, J., Naranjo, J., 2010.
Tracking exhumation of Andean ranges bounding the Middle Magdalena Valley
Basin, Colombia. Geology 38, 451454.
Nivia, A., Marriner, G.F., Kerr, A.C., Tarney, J., 2006. The Quebradagrande Complex: a
Lower Cretaceous ensialic marginal basin in the Central Cordillera of the Colombian
Andes. Journal of South American Earth Sciences 21, 423436.
Ordez-Carmona, O., Martins, P.M., Angel, C.P., 2001. Consideraciones Geocronlogicas e
Isotpicas preliminares del Magmatismo CretceoPaleoceno en el norte de la
Cordillera Central. Proceedings of VIII Congresso Colombiano Geologica Memoir
(5 pp.).
Pardo-Casas, F., Molnar, P., 1987. Relative motion of the Nazca (Farallon) and South
American plates since Late Cretaceous time. Tectonics 6, 233248.
Parra, M., Mora, A., Lopez, C., Rojas, L.E., Horton, B.K., 2012. Detecting earliest shortening
and deformation advance in thrust belt hinterlands: example from the Colombian
Andes. Tectonics 40, 175178.
Pindell, J., Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and
northern South America in the mantle reference frame: an update. In: James, K.,
Lorente, M.A., Pindell, J. (Eds.), The geology and evolution of the region between
North and South America: Geological Society of London Special Publication, 328,
pp. 155.
Reiners, P.W., Spell, T.L., Nicolescu, S., Zanetti, K.A., 2004. Zircon (UTh)/He
thermochronometry: He diffusion and comparisons with 40Ar/39Ar dating. Geochimica
et Cosmochimica Acta 68, 18571887.
Renne, P.R., Swisher, C.C., Deino, A.L., Karner, D.B., Owens, T.L., DePaolo, D.J., 1998.
Intercalibration of standards, absolute ages and uncertainties in 40Ar/39Ar dating.
Chemical Geology 145, 117152.
Restrepo, J.J., Toussaint, J.F., 1988. Terranes and continental accretion in the Colombian
Andes. Episodes 7, 189193.
Restrepo, J.J., Ordez-Carmona, O., Moreno-Snchez, M., 2009. A comment on The
Quebradagrande Complex: a Lower Cretaceous ensialic marginal basin in the Central
Cordillera of the Colombian Andes by Nivia et al. Journal of South American Earth
Sciences 28, 204205.
249
Restrepo-Moreno, S.A., Foster, D.A., Stockli, D.F., Parra, L.N., 2009. Long-term erosion
and exhumation of the Altiplano Antioqueo, Northern Andes (Colombia) from
apatite (UTh)/He thermochronology. Earth and Planetary Science Letters 278,
112.
Ruiz, G., Seward, D., Winkler, W., Maria, A.M., David, T.W., 2007. Evolution of the
Amazon Basin in Ecuador with special reference to hinterland tectonics: data
from zircon ssion-track and heavy mineral analysis. Developments in Sedimen-
tology 58, 907934.
Sarmiento, L.F., Rangel, A., 2004. Petroleum systems of the Upper Magdalena Valley,
Colombia. Marine and Petroleum Geology 21, 373391.
Sarmiento-Rojas, L.F., Van Wess, J.D., Cloetingh, S., 2006. Mesozoic transtensional basin
history of the Eastern Cordillera, Colombian Andes: inferences from tectonic
models. Journal of South American Earth Sciences 21, 383411.
Saylor, J.E., Horton, B.K., Stockli, D.F., More, A., Corredor, J., 2012. Structural and
thermochronological evidence for Paleogene basement involved shortening in
the axial Eastern Cordillera. Colombia. Journal of South American Earth Sciences.
39, 202215.
Sinton, C.W., Duncan, R.A., Storey, M., Lewis, J., Estrada, J.J., 1998. An oceanic ood basalt
province within the Caribbean plate. Earth and Planetary Science Letters 155 (34),
221235.
Spikings, R.A., Seward, D., Winkler, W., Ruiz, G.M., 2000. Low-temperature thermochronology
of the northern Cordillera Real, Ecuador: tectonic insights from zircon and apatite ssion
track analysis. Tectonics 19, 649668.
Spikings, R.A., Winkler, W., Seward, D., Handler, R., 2001. Along-strike variations in the
thermal and tectonic response of the continental Ecuadorian Andes to the collision
with heterogeneous oceanic crust. Earth and Planetary Science Letters 186, 5773.
Spikings, R.A., Foster, D.A., Kohn, B.P., Lister, G.S., 2002. Post-orogenic (b1500 Ma)
thermal history of the PalaeoMesoproterozoic Mt. Isa province, NE Australia.
Tectonophysics 349, 327365.
Spikings, R.A., Crowhurst, P.V., Winkler, W., Villagomez, D., 2010. Syn- and post accre-
tionary cooling history of the Ecuadorian Andes constrained by their in-situ and
detrital thermochronometric record. Journal of South American Earth Sciences
30, 121133.
Steiger, R.H., Jger, E., 1977. Subcommission on geochronology: convention on the use
of decay constants in geo- and cosmochronology. Earth and Planetary Science
Letters 36, 359362.
Toussaint, J.F., Restrepo, J.J., 1994. The Colombian Andes during Cretaceous times. In:
Salty, J.A. (Ed.), Cretaceous Tectonics of the Andes. Verlag Braunschweig, Wiesbaden,
Germany, pp. 61100.
Vallejo, C., Spikings, R.A., Luzieux, L., Winkler, W., Chew, D., Page, L., 2006. The early
interaction between the Caribbean Plateau and the NW South American Plate.
Terra Nova 18, 64269.
Vallejo, C., Winkler, W., Spikings, R.A., Luzieux, L., Heller, F., Bussy, F.O., 2009. Mode and
timing of terrane accretion in the forearc of the Andes in Ecuador. Geological Society
of America Memoirs 204, 197216.
Vsquez, M., Altenberger, U., Romer, R.L., Sudo, M., Moreno-Murillo, J.M., 2010. Mag-
matic evolution of the Andean Eastern Cordillera of Colombia during the Creta-
ceous: inuence of previous tectonic processes. Journal of South American Earth
Sciences 29, 171186.
Vergara, L., Prssl, K.F., 1994. Dating the Yav Formation (Aptian, Upper Magdalena
Valley, Colombia), palynological results. In: Etayo, F. (Ed.), Estudios Geolgicos
del Valle Superior del Magdalena. Departamento de Geociencias, Universidad
Nacional de Colombia, Bogot, p. 14.
Vesga, C., Barrero, D., 1978. Edades K/Ar en rocas gneas y metamrcas de la Cordillera
Central de Colombia y su implicacin geolgica. 2nd Congreso Colombiano de
Geologa, Bogot, Colombia, pp. 2131.
Villagmez, D., Spikings, R., Magna, T., Kammer, A., Winkler, W., Beltrn, A., 2011a.
Geochronology, geochemistry and tectonic evolution of the Western and Central
cordilleras of Colombia. Lithos 125, 875896.
Villagmez, D., Spikings, R., Mora, A., Guzmn, G., Ojeda, G., Corts, E., Van der Lelij, R.,
2011b. Vertical tectonics at a continental crust-oceanic plateau plate boundary
zone: ssion track thermochronology of the Sierra Nevada de Santa Marta, Colombia.
Tectonics 30 (TC4004, 18 pp.).
Villamil, T., 1999. CampanianMiocene tectonostratigraphy, depocenter evolution and
basin development of Colombia and western Venezuela. Palaeogeography,
Palaeoclimatology, Palaeoecology 153, 239275.