Lithos 282–283 (2017) 388–402

Contents lists available at ScienceDirect

Lithos

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

New fission-track age constraints on the exhumation of the central

Santander Massif: Implications for the tectonic evolution of the Northern
Andes, Colombia
Sergio Amaya a,⁎, Carlos Augusto Zuluaga a, Matthias Bernet b
a
Departamento de Geociencias, Universidad Nacional de Colombia, Sede Bogotá, Colombia
b
Institut des Sciences de la Terre, Université Grenoble Alpes, CNRS, Grenoble, France

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

Article history: The Late Cretaceous to late Neogene exhumation history of the central Santander Massif in the Northern Andes of
Received 24 June 2016 Colombia is controlled by the geodynamic interactions between the Caribbean, South American and Nazca plates,
Accepted 24 March 2017 as well as the Neogene collision and accretion of the Panama arc. Slab-breakoff of the Caribbean plate, with the tip
Available online 29 March 2017
of the slab tear presently being located beneath Bucaramanga, and the east-west oriented Caldas tear are the
main structures relating seismic activity and Late Miocene to Pleistocene magmatic/hydrothermal activity and
Keywords:
Exhumation
associated gold mineralization in the central Santander Massif.
Fission-track analysis Here we present new apatite (AFT) and zircon fission-track (ZFT) data from 18 samples collected along two pro-
Thermal modelling files in the California–Vetas block (including the Rio Charta), to the south of the Rio Charta fault, and from Buca-
Santander Massif ramanga to Picacho on the western flank of the central Santander Massif. The fission-track data are used for time–
Rio Charta fault temperature history modelling and for estimating long-term average exhumation rates. The California–Vetas
Bucaramanga fault block in the central Santander Massif to the north of the Rio Charta fault cooled rapidly at a rate of about 24
°C/Myr between 10 and 5 Ma. Fast cooling was not related to post-magmatic cooling or hydrothermal activity,
but rather to exhumation, with rates based on apatite fission-track cooling ages on the order of 0.3–
0.4 km/Myr. However, long-term average exhumation rates since the Late Cretaceous, based on zircon fission-
track data, were only on the order of 0.1–0.2 km/Myr. Our data indicate that next to the Rio Charta fault also
the Surata fault contributed to the exhumation of the California–Vetas block.
The western flank of the central Santander Massif, shows a more complete thermal history along the Bucaraman-
ga–Picacho profile, with the exposure of an exhumed zircon fission-track partial annealing zone. Thermal history
modelling of zircon fission-track data of this profile shows that after burial and heating from about 150 Ma on
cooling at rates of 7–10 °C/Myr started at about 25 Ma. For the lower part of the profile, the early Miocene ZFT
ages indicate exhumation at rates of 0.3–0.5 km/Myr along the Bucaramanga fault, but were only about
0.1 km/Myr on the high plateau of the central Santander Massif.
© 2017 Elsevier B.V. All rights reserved.

1. Introduction
The Santander Massif in the Northern Andes of Colombia, is the
northern extension of the Eastern Cordillera. The current configuration
of the Santander Massif is the result of oblique convergence between
the Caribbean and South American plates, accretion of the Panama arc,
and rotation and translation of the continental Maracaibo block
(Fig. 1; Aleman and Ramos, 2000). These plate tectonic interactions re-
sulted in rapid exhumation of individual tectonic blocks at different
rates in the Santander Massif along the Bucaramanga fault, and the
Merida Andes along the Boconó fault during the Paleogene and Neogene
⁎ Corresponding author.
E-mail addresses: samayaf@unal.edu.co, samaya@sgc.gov.co (S. Amaya).
(Fig. 1; Bermúdez et al., 2010, 2011; Montes et al., 2005; Shagam et al.,
1984; Taboada et al., 2000).
The structure of the Santander Massif is controlled by the Bucara-
manga strike-slip fault to the west, and the Labateca–Chucurima thrust
faults to the east of the massif (Fig. 2). Several tectonic models have
been proposed for the evolution of the Santander Massif based on
field observations and thermochronological and geochronological data
(see for example Shagam et al., 1984; van der Lelij, 2013; van der Lelij
et al., 2016a, 2016b; Villagómez et al., 2011). Shagam et al. (1984) pre-
sented a wide range of apatite, zircon and sphene fission-track cooling
ages from samples collected across the massif and related the range of
cooling ages to local fault activity. Villagómez et al. (2011) used zircon
and apatite fission-track ages of 28.1 ± 3.2 Ma and 16.7 ± 3.2 Ma re-
spectively from a Jurassic granite to propose rapid cooling of the

http://dx.doi.org/10.1016/j.lithos.2017.03.019
0024-4937/© 2017 Elsevier B.V. All rights reserved.
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 389

OF

SP

B
10° N

CR
BF
MTB

Venezuela

VMMB
SL MA

WC LB

PIV
CC
EC

F
5° N G
Colombia

75° W 70° W

Fig. 1. Overview map of the Northern Andes. WC: Western Cordillera, CC: Central Cordillera, EC: Eastern Cordillera; MTB: Maracaibo Triangular Block, SNSM: Sierra Nevada de Santa Marta,
MA: Mérida Andes, SM: Santander Massif, SL: Serrania de San Lucas, SP: Sierra de Perija, LB: Llanos Basin, BA: Barinas-Apure Basin, MC: Maracaibo Basin, VMMB: Valley Middle Magdalena
Basin, CRB: Cesar-Rancheria Basin, SMBF: Santa Marta–Bucaramanga fault, OF: Oca fault, BF: Bocono fault, GF: Guaicaramo fault, PIV: Paipa-Iza Volcano.
Digital relief map (Amante and Eakins, 2008) showing the tectonic provinces of northern South America modified from Feo-Codecido et al. (1984) and Ostos et al. (2005).

Santander Massif from temperatures N 200 °C during the middle Mio-
cene. However, because of the lack of track lengths measurements in
apatite, no thermal model was presented. Van der Lelij (2013) sug-
gested that the Santander Massif started exhuming at ~ 40 Ma during
a period of accelerated convergence between the Nazca/Farallon plate
and the western margin of South America. Van der Lelij et al. (2016a)
described diachronous exhumation in distinct fault blocks of the San-
tander Massif since at least ~17 Ma at rates of 0.5–1.0 km/Myr, conclud-
ing that exhumation in the Santander Massif are strongly controlled by
secondary faults. From this previous work it is unclear whether the col-
lision of the Panamá arc with the western margin of South America at
~25 Ma (e.g. Farris et al., 2011) had significant effects on the exhuma-
tion of the Santander Massif, because exhumation supposedly started
at ~40 Ma and accelerated since ~17 Ma.
Whereas consensus seems to exist about the timing of the Cenozoic
tectonic evolution of the Santander Massif, the first-order characteris-
tics of its evolution are still under debate: Is there more than one tecton-
ic block formed within the Santander Massif during its Late Cretaceous/
early Paleocene to the present-day evolution, and which faults con-
trolled exhumation of the tectonic blocks within the Santander Massif?
The previous thermochronological studies in the Santander Massif had a
more regional sampling approach (Shagam et al., 1984; Van der Lelij,
2013; Van der Lelij et al., 2016a; Villagómez et al., 2011), whereas our
study has a rather local focus with more closely spaced sampling. The
purpose of this study is to provide new constraints on the thermal his-
tory and the timing and rate of exhumation of the central Santander
Massif between Bucaramanga and Picacho on the western flank of the
massif, and of the California–Vetas district in its central part (Fig. 2).
The California–Vetas district, which is known for its gold mineralization
(e.g. Mantilla et al., 2011, 2013; Rodriguez, 2014), is located on what we
refer to in this study as the California–Vetas block, delineated by the Rio
Charta fault to the South and South-East, and by the Surata fault system
to the West and Northwest (Fig. 2). The western flank of the central San-
tander Massif is controlled by the Bucaramanga fault.
Using fission-track data from six apatite and fifteen zircon sepa-
rates, we model the thermal history of the central Santander Massif
along two E-W oriented profiles with samples collected from sedi-
mentary, igneous and metamorphic rocks. The first profile has a
length of about 26 km, starting near Matanza at 1425 m elevation,
going up onto the high plateau of the central Santander Massif at
3961 m elevation north of Berlín, and back to about 3400 m elevation
to the west of La Laguna (Fig. 2). For this profile thermal histories
were determined for individual samples with the HeFTy program of
Ketcham (2005). The second profile has a length of about 14 km,
starting to the east of Bucaramanga at 1293 m elevation and going
up to the Picacho toll station at 3400 m elevation (Fig. 2). All ZFT
390 S. Amaya et al. / Lithos 282–283 (2017) 388–402

73°W

7°N 7°N

73°W
Fig. 2. Simplified geological map of the Santander Massif, Colombia. Locations of all samples analysed in this study are shown as squares. Apatite fission-track ages are marked in italic text,
whereas bold text indicates a zircon fission-track age. LC = La Corcova quartz-monzonite, SB = Santa Barbara quartz monzonite. The Surata fault system to the west and northwest, and
the Rio Charta fault to the south and southeast delineate the California–Vetas block.
Modified after Zuluaga et al. (in revision).

samples of this profile were also modelled with HeFTy, but with a
stacked multi-sample approach for determining a common thermal
history. The thermal history modelling results are interpreted in
the Northern Andes geodynamic context.
2. Geological setting
The Santander Massif in the Eastern Cordillera of Colombia forms
with the Sierra Nevada de Santa Marta, the Perija Mountains in
Colombia, and the Merida Andes in Venezuela, the main mountainous
belts of the triangular Maracaibo block (Fig. 1). This block is delineated
to the north by the E-W trending dextral Oca strike-slip fault, along
which Cretaceous and older rocks of the continental crust of the Guajira
Peninsula are moving towards the east (Fig. 1; Cardona et al., 2009). The
western margin of the Maracaibo block is defined by the NNW-SSE
trending sinistral Santa Marta–Bucaramanga fault system, extending
over a length of about 600 km from Santa Marta in the north to south
of Bucaramanga, where it ends defining the western margin of the San-
tander Massif (Fig. 1). The south-eastern fold-and-thrust belt of the
Merida Andes delimits the Maracaibo block to the south-east against
the South American continental plate (e.g. Colletta et al., 1997; Mann
et al., 2006; Fig. 1).
The Santander Massif crystalline basement consists of metamorphic
rocks of Precambrian to Paleozoic age, cut by Triassic and Jurassic intru-
sive bodies (Goldsmith et al., 1971; van der Lelij et al., 2016b; Ward
et al., 1974). The metamorphic basement rocks can be divided into
three units, which from bottom to top are the Bucaramanga gneiss,
the Silgará schists and the Orthogneiss units (Ward et al., 1974).
The basement of the Santander Massif is unconformably overlain by
low-grade meta-sedimentary rocks of the “Guaca–La Virgen” Formation
(formerly Floresta Formation), calcareous and clastic rocks of the
Carboniferous–Permian Diamante Formation, Tiburon and Bocas
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 391

Santander Massif simplified stratigraphy

Jurassic Cretaceous
METAMORPHIC/IGNEOUS ROCKS SEDIMENTARY ROCKS
Top: quartz sandstone, light colored withl ayers
TAMBOR Fm. Bottom: reddish brown silstone and conglomeratic
Mesozoic
RHYOLITHE
sandstone
145 Ma
Quartz monzonite, biotite Conglomeratic sandstone and conglomerate,
Jc GIRON Fm. gray - reddish - brown silstone
meduim grained, gray pink
Granodiorite, porphyritic
Jgd biotite, gray JORDAN Fm. Siltstone and fine grained sandstone,
reddish - brown and calcareous in certain parts
201 Ma
La Cocrova quartz monzonite, biotite
JTRcl
Triassic

fine grained, gray

JTRclp La Cocrova quartz monzonite, porphyritic facies
Quartz monzonite and granite, biotite
JTRcg light gray or pink and tonalite, hornblende
biotite, gray
252 Ma
BOCAS Fm. Sandstone and shale, gray to brownish-gray
Permian

TIBURRON Fm. Conglomerate with limestone pebbles

299 Ma
Carboniferous

DIAMANTE Fm. Gray limestone, gray to reddish-brown shale

Paleozoic

359 Ma
FLORESTA Fm./GUACA - LA VIRGEN Fm.
Devonian

Yellowish-gray silstone and phyllite

fine-grained sandstone

419 Ma
Silurian

ORTHOGNEISS
444 Ma
Quartz monzonite and granodiorite gneiss
Ordovician

PD SILGARA Fm.
PDs: Phyllite, schist and quarzites, less metavolcanic rocks
low to medium grade metamorphism
PDs + Jgd: PDs with numerous small granodiorite intrusions
485 Ma
Cambrian

541 Ma
Precambrian

PCab BUCARAMANGA GNEISS

PCab: Schist, gneisses and migmatites, high-grade metamorphism
PCabs + Jgd: PCa with numerous small granodiorite intrusions
PCabm: PCab with abundant small orthogneiss masses

Fig. 3. Schematic stratigraphic column of the Santander Massif Precambrian to lower Cretaceous units.
Modified after the Colombian Geological Survey 1977; van der Lelij et al. (2016b).

Formations (Fig. 3). Clastic rocks of the Tiburon Formation contain
sedimentary rock fragments that were derived from erosion of the
Diamante Formation due to epirogenic surface uplift during the Late Pa-
leozoic (Ward et al., 1974). Mesozoic units of Early and Middle Jurassic
plutonic and sedimentary rocks include the plutons that Ward et al.
(1974) called the Santander Plutonic Group. Outcrops of Jurassic intru-
sive bodies in the south and center of the Santander Massif include the
Pescadero granite to the south near Cepita, or the Santa Barbara quartz
monzonite and the Corcova Mogotes quartz monzonite in the central
Santander Massif (Fig. 2). Jurassic intrusions in the northern Santander
Massif include biotite-quartz-monzonite, granodiorite and biotite
tonalites. Lower and Middle Jurassic sedimentary rocks make up the
continental and volcanoclastic Jordan Formation, and Upper Jurassic
sedimentary rocks are represented by the Giron Formation (Fig. 3).
392
Cretaceous sedimentary rocks, such as the Tambor Formation crop out
to the east and west of the Santander Massif and are preserved locally
as erosional remnants on the massif high plateau. Cenozoic sedimentary
rocks can be found in the south-eastern and north-eastern parts of the
Santander Massif. Quaternary deposits are widely distributed to the
west of the Santander Massif, filling the alluvial valleys of the major
drainages of this region.
In the central part of the Santander Massif, between the California–
Vetas block and Bucaramanga (Fig. 2), the massif shows a particular
fault pattern with dominantly NE-SW trending faults such as the dextral
Surata fault (Fig. 2). In the eastern part of the study area, the sinistral Rio
Charta fault system forms a northwest to east oriented arc (Fig. 2),
showing small apparent vertical offset, but appreciable horizontal dis-
placement. Northeast of Bucaramanga, in the Matanza and Surata
areas, the Surata fault is within the Silgara schist.
The Bucaramanga fault separates the Bucaramanga gneiss of the
Santander Massif from Mesozoic sedimentary rocks to the west of the
Bucaramanga area. The western flank of the Santander Massif, east of
the NNW-SSE trending Bucaramanga fault, is characterized by a much
deeper level of exhumation than the eastern flank towards the Maracai-
bo basin, where a large scale strike-slip fault system is absent and Creta-
ceous and Tertiary cover units are better preserved.
3. Sampling locations
Samples were collected along two E-W profiles in the California–
Vetas block and between Bucaramanga and the Picacho toll station.
The samples from the California–Vetas block profile were collected
over an ~ 2000 m elevation range from the Surata fault near Matanza
up onto the plateau of the Santander Massif near Berlín (Figs. 2, 4). In
addition, a modern river sediment sample was collected from the Rio
Charta near the village of Charta, and two samples were collected to
the south of the Rio Charta fault on the high plateau (Fig. 2). The profile
sampled between Bucaramanga and the Picacho toll station on the
western flank of the Santander Massif plateau covers an elevation differ-
ence of ~2200 m (Figs. 2 and 4). The samples of both profiles were col-
lected from crystalline rocks, mainly the Bucaramanga gneiss, Silgara
schist, and Orthogneiss units, and plutonic rocks such as tonalite, the
Paramo Rico granodiorite, the La Corcova quartz monzonite, and sedi-
mentary rocks of the Jurassic Giron Formation and the Cretaceous
Tambor Formation (Figs. 2 and 3). The coordinates and lithologic infor-
mation of all samples are given in Table 1.
4500
4000
California -Vetas block plateau
3500
3000
lt
Elevation (m)
Surata fau
2500
2000
1500
1000
500
0
0 25 50
Fig. 4. Age-elevation relationship California–Vetas block profile. Squares denote apatite fission-track ages and circles indicate zircon fission-track ages.
S. Amaya et al. / Lithos 282–283 (2017) 388–402
4. Methods
4.1. Fission-track analysis
All samples collected in the field were crushed, pulverized and
sieved (200–80 μm fraction). The heavy mineral fraction was concen-
trated using a Wilfley table and apatite and zircon grains were separated
using standard magnetic and heavy liquid separation techniques. All
mineral separation was done at the Servicio Geologico Colombiano in
Bogotá. Apatite aliquots were mounted in epoxy resin, polished and
etched for 20 s at 21 °C in 5.5 M HNO3. Zircon aliquots were mounted
in Teflon® sheets, polished and etched in a NaOH-KOH melt in a labora-
tory oven at 228 °C for 5–30 h. Grain mounts were covered with musco-
vite external detectors for irradiation. Samples were prepared in the
fission-track laboratory of the Servicio Geologico Colombiano in Bogotá
and at the ISTerre thermochronology laboratory in Grenoble, and irradi-
ated at the FRM II research rector in Garching, Germany. A nominal
neutron flux of 8.0 n/cm2 was used for apatite sample irradiation.
IRMM540R dosimeter glasses and Durango and Fish Canyon Tuff age
standards were included in the irradiation packages for determining
fluence and zeta values. Zircon samples, Fish Canyon Tuff age standards
and CN1 dosimeter glasses were irradiated with a lower neutron flux of
0.5 n/cm2. After irradiation, muscovite external detectors were etched
for 18 min at 20 °C in 48% HF to reveal induced tracks. Samples were
analysed dry at 1250× with an Olympus BX51 optical microscope and
the FTStage 4.04 system in the fission-track laboratory of the Servicio
Geologico Colombiano in Bogotá or the ISTerre thermochronology
laboratory.
For AFT and ZFT samples central ages are presented. The central age,
as defined by Galbraith and Laslett (1993), represents an estimate of the
mean age of a discordant, over-dispersed grain age population. Over-
dispersion may be due to the intrinsic variation of radioactive decay,
given that spontaneous track formation follows a Poisson distribution,
equally to the formation of induced tracks. On the other hand, over-
dispersion may also be the indication for a complex thermal history
and partial annealing of fission tracks in apatite and zircon. The implica-
tion is that discordance in FT dating is more commonly an indication of a
mixed distribution rather than a poorly measured distribution. There
are a number of possible reasons for this (see Galbraith and Laslett
(1993) or Galbraith (2005) for details), but in our study we believe
that over-dispersion is a result of partial annealing, differences in
annealing properties, as caused by variations in chemical composition
for apatite (e.g. Donelick et al., 2005, and references therein) or
south of Rio Charta fault
outlier
Apatite
Zircon
75 100 125 150 175 200 225 250
Fission -track age (Ma)
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 393

Table 1
Sample locations and lithologies (coordinates are in the reference system WGS84).

Sample number Latitude Longitude Elevation Lithology Stratigraphic unit

(°N) (°W) (m)

Bucaramanga–Picacho profile
07SACZ-15 7.1256917 −73.9586320 1239 Quartz feldspar gneiss Bucaramanga gneiss
07SACZ-17 7.1092533 −73.0615681 1779 Quartz feldspar gneiss Bucaramanga gneiss
13SACZ-30 7.1064379 −72.9847066 3077 Mica schist Silgará schist
13SACZ-31 7.1045039 −72.9698230 3217 Orthogneiss Orthogneiss
13SACZ-32 7.1055091 −72.9667344 3400 Quartz sandstone Giron Formation?

California–Vetas profile
12SACZ-8 7.2415575 −72.8231577 3356 Sillimanite gneiss Bucaramanga gneiss
12SACZ-8B 7.2406415 −72.8220829 3360 Sillimanite gneiss Bucaramanga gneiss
12SACZ-10B 7.2373735 −72.8176779 3367 Sillimanite gneiss Bucaramanga gneiss
12SACZ-16 7.2373735 −72.7967417 3482 Quartzite Silgará schist
12-SACZ-31 7.2531214 −72.8908157 3961 Sillimanite gneiss Bucaramanga gneiss
13SACZ-10 7.2373735 −72.8171945 3340 Quartzite Orthogneiss
13SACZ-18 7.3022927 −72,9,582,222 2526 Quartz sandstone Tambor Formation
13SACZ-19 7.2785000 −72.9671600 2001 Sand modern river sediment
13SACZ-21 7.2806982 −73.0195450 1425 Mica schist Silgara schist
13SACZ-24 7.3134700 −73.0296500 1575 Quartzite Silgara schist
13SACZ-25 7.3091786 −73.0320640 1584 Diorite Jurassic intrusion

South of Rio Charta fault

12SACZ-13 7.1925390 −72.865457 3336 Sillimanite gneiss Bucaramanga GNEISS
13SACZ-11 7.1915500 −72.865060 3323 sillimanite gneiss Bucaramanga gneiss

variations in radiation damage for zircon (Brandon, 1992; Garver et al.,
2005). The central age is clearly not appropriate for finite mixed distri-
butions, but given that we deal here with samples collected from crys-
talline rocks, except for sample 13SACZ-19 from the Rio Charta near
Charta, the central age provides a better estimate for the average age
of a sample than the pooled age (Galbraith and Laslett, 1993). Further-
more, for samples with uniform, and not over-dispersed grain age spec-
tra, the central age is equal to the pooled age.
4.2. Time–temperature history modelling
We used the HeFTy program (v. 1.9.1.72) of Ketcham (2005) to
model the thermal history of individual samples from the California–
Vetas profile, for which we have apatite and zircon fission-track data.
Because of low track counts and low uranium concentrations no suffi-
cient track lengths could be measured in our apatite samples. However,
given that the mean track lengths published by van der Lelij et al.
(2016a) of apatite samples throughout the Santander Massif narrowly
range between 14.47 ± 0.1 and 15.37 ± 0.1 μm, independent of sample
location and elevation (636 m–3731 m), we use a mean track length
value from the van der Lelij et al. (2016a) data of 14.92 μm to constrain
our HeFTy models. Furthermore, the models are based on apatite
fission-track age and Dpar values, and zircon fission-track ages, and
100,000 tried paths. The constraint boxes for inverse modelling were
placed following the forward model to match observed fission-track
ages as closely as possible. Nonetheless, for comparison we ran HeFTy
models with and without constraint boxes for the California–Vetas pro-
file samples.
For the Bucaramanga–Picacho profile we also used HeFTy to deter-
mine t–T histories from ZFT data of five samples collected over an eleva-
tion difference of about 2200 m. We used the approximate depositional
age of the Giron Formation (sample 13SACZ-32) in order to constrain
the part of Late Jurassic to present-day thermal history that the sampled
metamorphic and sedimentary rocks of this profile had in common. We
therefore assigned a surface temperature of about 20 °C to sample
13SACZ-32 at the time of deposition of the Giron Formation at about
150 ± 5 Ma at the start of the model and assuming a thermal gradient
of 30 °C/km, we assigned increasing starting temperatures for the
other 4 samples of metamorphic rock depending on their elevation
today and therefore paleo-depth distance below the top sample
13SACZ-32 at 150 ± 5 Ma. We modelled the thermal histories based
both on the ZFT data of all samples and also the AFT data of sample
13SACZ32, using the ZFT cooling ages of samples 07SACZ-15 and 17 to
constrain the onset of cooling at about 25 Ma after roughly 125 Myr of
burial heating.
4.3. Exhumation rate estimates
The 1-D thermal advection model age2edot of Brandon (see Ehlers
et al., 2005) is used in this study for determining first-order exhumation
rate estimates from our AFT and ZFT data. Model input parameters in-
clude an assumed pre-exhumation steady state thermal gradient of 30
°C/km, and a mean surface temperature of 10 °C of the Santander Massif
during exhumation and surface uplift. Further input parameters include
an activation energy for 50% annealing of 147.2 (kj/mol) for apatites of
average composition and annealing parameter Beta of 2,050,000 (s−1)
(Ketcham et al., 1999). For radiation damaged zircons an activation en-
ergy of 208 (kj/mol) and a normalized frequency factor of 1 × 108 (s−1)
were used, whereas for zero damaged zircons an activation energy of
321 (kj/mol) and a normalized frequency factor of 5.566 × 1013 (s−1)
(Rahn et al., 2004; Willett and Brandon, 2013) were used. For the pur-
pose of this study such a simplistic approach is sufficient for determin-
ing long-term average exhumation rates in the Santander Massif.
5. Results
5.1. California–Vetas block profile
Four igneous and metamorphic rocks from the California–Vetas
block yield central AFT ages ranging between 29.7 ± 5.25 Ma and 7.9
± 1.1 Ma (Table 2; Figs. 2 and 4). Except for sample 12SACZ8b, obtained
from the Orthogneiss unit, all AFT ages have P(χ2) values higher than
7.9% (Table 2), suggesting that the AFT ages of each sample define single
age populations. Because of the relatively young cooling ages and the
low uranium concentration no sufficient track length measurements
of statistical value could be obtained. Dpar measurements show rela-
tively small values of b1.75 μm. Five samples from igneous and meta-
morphic rocks from the California–Vetas block yield central ZFT ages
ranging between 111.8 ± 15.5 Ma on the high plateau of the central
Santander Massif, and 37.3 ± 2.6 Ma near the Surata fault on the west-
ern flank of the massif (Table 3; Figs. 2 and 4). Samples 13SACZ-18 and
394 S. Amaya et al. / Lithos 282–283 (2017) 388–402

Table 2
Apatite fission-track data of the central Santander Massif.

Sample n ρs Ns ρi Ni ρd P(χ2) Dispersion Agea (Ma) ±2σ U (ppm) ±2σ MDpar (μm) n
(10−5 cm−2) (10−5 cm−2) (10−6 cm−2) (%) (%) Dpar

12SACZ-8B 19 0.25 29 4.71 546 1.210 1.9 43.5 9.8 4.5 6 1.0 1.21 ± 0.1 3
12SACZ-31 20 1.34 77 27.1 1557 1.269 50.6 0.9 8.6 2.2 32 1.9 1.31 ± 0.21 38
13SACZ-10 16 1.05 60 23.2 1324 1.268 39.6 12.5 7.9 2.3 28 1.7 1.86 ± 0.39 13
13SACZ-19 50 1.01 229 14.4 3266 1.270 1.9 26.1 12.6 2.3 17 1.0 - -
13SACZ-21 20 0.94 43 5.52 252 1.266 7.9 34.5 32.8 12.5 6 0.8 1.60 ± 0.24 94
13SACZ-32 30 1.14 197 15.4 2649 1.140 0.4 29.8 12.3 2.6 20 1.0 1.39 ± 0.11 29
a
Note: samples were counted dry with a BX-51 Olympus microscope at 1250× magnification. Central ages were calculated with the Binomfit program of M. Brandon (see Ehlers et al.,
2005), using a zeta value of 274.44 ± 11.56.

13SACZ-25 have a P(χ2) value N 5%, as all other samples have more com-
plex grain-age populations.
The AFT ages determined for detrital apatite grains of sample
13SACZ-19 range from 2.8 to 67 Ma (Fig. 5), with main peak ages at
11.5 ± 2.1 Ma and 31.8 ± 28.4 Ma. However, the two age peaks are
not well separated. The central age of this sample is 12.6 ± 2.3 Ma.
5.2. Bucaramanga–Picacho profile
Five samples of sedimentary, igneous and metamorphic rocks col-
lected along the Bucaramanga–Picacho profile yield central ZFT ages
ranging from 123.2 ± 10.30 Ma to 19.2 ± 2.0 Ma (Table 3; Figs. 2 &
6). Samples 07SACZ-15, 07SACZ-17 and 13SACZ-32 have P(χ2) values
of N5%, whereas the other two samples contain over-dispersed age pop-
ulations. Only sample 13SACZ-32 collected along the Bucaramanga–
Picacho profile contained sufficient apatites for analysis to yield a cen-
tral age of 12.3 ± 2.0 Ma, but with a P(χ2) value of b5% (Table 2).
5.3. Thermal history modelling
The three samples 12SACZ-8, 13SACZ-10, and 13SACZ-21 from the
California–Vetas block profile were modelled using the HeFTy software
of Ketcham (2005). Each sample was modelled twice, once using con-
straint boxes based on the forward model, and once without any con-
straint boxes (Fig. 7). For each model 100,000 paths were tried and
the number of acceptable and good paths are given in Table 4. Fig. 7
shows that there is no significant difference between the models with
and without constraint boxes, as the models are controlled by the data
and not the constraint boxes. For the two samples 12SACZ-8 and
13SACZ-10 the thermal histories are very similar, with best-fit solutions
indicating a phase of rapid cooling on the order of 20–30 °C/Myr be-
tween 10 to 5 Ma. Possibly accelerated cooling occurred already in the
late Oligocene–early Miocene, but the cooling histories between 240
°C and 120 °C are not well resolved (Fig. 7).
Sample 12SACZ-21 from the western flank of the Santander Massif,
shows a somewhat different cooling history. The phase of rapid late
Miocene cooling is not detected in this sample, but rather rapid cooling
at around 30 to 25 Ma (Fig. 7C).
The thermal history models of the samples collected along the Buca-
ramanga–Picacho profile can be stacked to emphasize the common
thermal history of the western flank of the central Santander Massif
since deposition of the Giron Formation in direct contact with Bucara-
manga gneiss near the Picacho toll station (Fig. 8). The individual ther-
mal history models of each sample can be found in the supplementary
data. Here we use the weighted mean solutions and only show the con-
straint points good and acceptable fits to make the thermal histories
more visible. The model results are shown in Table 5. The common ther-
mal history that emerges is slow burial heating because of deposition of
the Giron Formation and the subsequent Cretaceous and Cenozoic sed-
imentary units, which are partly preserved in the eastern Santander
Massif today (Fig. 2). The ZFT ages of 25–20 Ma of the two lower most
samples (07SACZ15 and 17) suggest that these rocks were heated to
temperatures exceeding the ZFT partial annealing zone, whereas the
other samples support heating with the partial annealing zone to pro-
vide partially reset central ages younger than the depositional age of
the Giron Formation. It seems from these models that cooling of the
western flank of the Santander Massif since about 25 Ma occurred at
rates of about 7–10 °C/Myr until the present day. The thermal history

Table 3
Zircon fission-track data of the central Santander Massif.

Sample n ρs (10−5 cm−2) Ns ρi (10−5 cm−2) Ni ρd (10−5 cm−2) P(χ2) Dispersion Age (Ma)a ±2σ U (ppm) ±2σ
(%) (%)

Bucaramanga–Picacho profile
07SACZ-15 11 30.7 223 0.8 224 2.09 63.9 0.5 19.2 2.0 587 81
07SACZ-17 29 28.9 745 3.2 600 2.10 50.7 9.7 24.1 1.6 442 39
13SACZ-30 20 85.7 1878 3.7 300 2.62 3.5 22.1 112.5 10.2 262 31
13SACZ-31 10 95.2 774 3.4 109 2.62 3.9 27.3 122.9 33.9 256 50
13SACZ-32 21 71.7 1458 10.3 209 2.62 70.3 0.4 123.5 10.4 196 28

California–Vetas profile
12SACZ-8 20 117.0 2152 33.0 610 3.36 0.2 21.4 67.4 5.1 391 34
12SACZ-10B 13 110.0 1760 36.6 587 3.47 0.0 58.0 66.4 11.5 420 37
12SACZ-16 20 87.5 3113 23.9 852 3.38 0.0 23.0 71.6 5.3 282 21
13SACZ-10 20 75.5 3102 13.5 554 3.39 1.9 17.4 111.8 15.5 158 14
13SACZ-18 20 81.3 2469 17.9 542 3.41 55.2 1.3 91.4 5.4 208 19
13SACZ-21 7 110.0 534 24.5 119 3.42 20.7 18.1 89.4 23.5 285 53
13SACZ-24 19 22.3 689 19.7 607 3.42 0.0 34.2 23.4 4.8 229 20
13SACZ-25 20 33.3 774 18.1 420 3.43 56.6 7.5 37.5 5.4 210 21

South of Rio Charta fault

12SACZ-13 19 63.2 2162 7.96 272 3.43 0.0 41.4 154.7 37.6 92 11
13SACZ-11 8 10.6 937 14.6 129 3.40 0.0 55.0 136.6 60.6 171 30
a
Note: fission-track age is given as central age. Samples were counted dry with a BX-51 Olympus microscope at 1250× magnification. Central age was calculated with the Binomfit program
of M. Brandon (see Ehlers et al., 2005), using zeta values of 118.48 ± 3.85 (SA) and 136.65 ± 5.17 (MB).
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 395

exhumation that did not yet allow the complete removal of not fully
reset rocks.
56Ma

40 6. Discussion
30
6.1. Thermochronological constraints on the exhumation of the Santander
20 Massif
2

The three uppermost samples from the Bucaramanga–Picacho pro-

0
-2
2Ma
103 45 29%
0 1 2 3
Fig. 5. Radial plot of the Rio Charta (13SACZ-19) detrital apatite fission-track data. The
central age is 12.6 ± 2.3 Ma, but the range of observed individual apatite fission-track
ages in this river sediment sample is between 2.8 and 66.5 Ma.
of the Santander Massif prior to the deposition of the Giron Formation
cannot be determined with our thermochronological data.
5.4. Long-term average exhumation rates
For the California–Vetas block profile long-term average exhuma-
tion rates are on the order of 0.3–0.4 km/Myr based on late Miocene
AFT ages, and 0.1–0.2 km/Myr based on the Cretaceous ZFT ages
(Fig. 9A). For the lower part of the Bucaramanga–Picacho profile the
early Miocene ZFT age indicate exhumation at rates of 0.3–0.5 km/Myr
for radiation damaged or zero-damage zircons (Fig. 9B). The partially
reset zircon ages of the three upper samples of the profile hint slow
10
file, collected between 3400 and 3077 m elevation, yield central ZFT
ages ranging from 123.2 ± 10.3 to 107.0 ± 17.1 Ma. Samples 13SACZ-
30 and 13SACZ-31 were collected from crystalline rocks whereas sam-
ple 13SACZ-32 was collected from sedimentary rocks of the Jurassic
Giron Formation, which unconformably overlies the crystalline base-
ment (Ward and Goldsmith, 1973). The ZFT data of all three samples re-
flect partially reset ages because of burial heating, rather than being
indicative of a specific tectonic or thermal event. In fact, the ZFT data
of this study from the Bucaramanga–Picacho profile, together with sam-
ple 10VDL05 of van der Lelij et al. (2016a), plotted against elevation
show that an exhumed ZFT partial annealing zone (PAZ) is exposed
4 on the western flank of the central Santander Massif (Fig. 6). This ex-
humed PAZ has a thickness of about 2.0 km, which corresponds to a
pre-exhumation thermal gradient of about 30 °C/km, depending on
hold time (Reiners and Brandon, 2006). The samples collected at 1239
and 1779 m elevation near Bucaramanga indicate that cooling caused
by exhumation of this PAZ occurred between 25 and 20 Ma (Fig. 8).
This is also consistent with the 23 ± 4.8 Ma ZFT age of sample
13SACZ-24 collected at 1586 m elevation along the California–Vetas
profile to the west of the Surata fault. In contrast, sample 13SACZ-21,
which was collected at 1425 m elevation from a granodiorite in direct
contact with the Early Cretaceous Tambor Formation has a ZFT cooling
age of 89.4 ± 23.5 Ma (Fig. 11). The sampled granodiorite and the over-
lying Tambor Formation rocks must have been displaced along the
Surata fault zone during early Miocene exhumation because the AFT
age of the same granodiorite sample corresponds within its 2 sigma
error to the about 25 Ma rapid cooling observed in the ZFT age elevation
profile (Fig. 6). Because the apparent ZFT cooling age of sample 13SACZ-
21 is much younger than both the crystallization age of the granodiorite
and the depositional age of the overlying Tambor Formation, we think
that sedimentary burial heating of the Santander Massif crystalline
rocks from the Jurassic–Early Cretaceous until the late Oligocene was
sufficient for partial resetting of the ZFT system. The phase of fast

Bucaramanga - Picacho ZFT age-elevation profile

4000

Picacho
3500

3000
Elevation (m)

Exhumed ZFT partial annealing zone

2500 ?
fast exhumation starting at aorund 25 Ma ?
2000

1500

1000
Bucaramanga
ZFT data this tsudy
500
10VDL05, van der Lelij et al. (2016)

0
0 20 40 60 80 100 120 140 160
Zircon fission-track age (Ma)

Fig. 6. Zircon fission-track age-elevation plot of the Bucaramanga–Picacho profile of data from this study and sample 10VDL05 of van der Lelij et al., 2016a). The zircon fission-track ages
follow the typical trend of an exhumed partial annealing zone.
396 S. Amaya et al. / Lithos 282–283 (2017) 388–402

A) 0 B) 0
20 12SACZ-8/8b, Bucaramanga gneiss 20 12SACZ-8/8b, Bucaramanga gneiss
3356 m elevation 3356 m elevation
40 with constraint boxes 40 without constraint boxes
60 based on forward model 60
80 80
100 100
acceptable paths
Temperature (°C)

Temperature (°C)
120 120
good paths del
140 140 mo
ard
160 forw
160
del
mo
180 ard 180
forw
200 200
220 220
best-fit model
240 240 best-fit model
260 260
280 280
300 300
100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0
Time (Ma) Time (Ma)

C) 0 D) 0
20 13SACZ-10, Orthogneiss 2E
20 13SACZ-10, Orthogneiss
3340 m elevation 3340 m elevation
40 with constraint boxes 40 without constraint boxes
60 based on forward model 60
2E

80 80

100 Temperature (°C) 100

Temperature (°C)

120 120

140 140

160
2E
160

180 180 best-fit model

200 best-fit model 200
odel
ard m
2E

220 odel 220 forw

ard m
forw
240 240

260 260

280 280

300 300
145 135 125 115 105 95 85 75 65 55 45 35 25 15 5 0 145 135 125 115 105 95 85 75 65 55 45 35 25 15 5 0
Time (Ma) Time (Ma)

E)0 F)0
20 13SACZ-21, Silgará schist 20 13SACZ-21, Silgará schist
1425 m elevation 1425 m elevation
40 with constraint boxes 40 without constraint boxes
60 based on forward model 60

80 80
100 100
Temperature (°C)

Temperature (°C)

120 120

140 140
le
od
m

160 160
it
t-f
s

180 180
be

el l
d de
200 mo 200 mo
w ard w ard
220 for 220 for

240 best-fit model 240

260 260
280 280
300 300
145 135 125 115 105 95 85 75 65 55 45 35 25 15 5 0 145 135 125 115 105 95 85 75 65 55 45 35 25 15 5 0
Time (Ma) Time (Ma)

Fig. 7. Time–temperature history models using HeFTy of Ketcham (2005) for samples 12SACZ-8/8B, 13SACZ-10 and 13SACZ-21, with and without constraints. Shown are acceptable paths
and good paths solution envelops, the forward model and the best fit model.

cooling of the western flank of the central Santander Massif at around needed for better constraining the activity of individual faults from the
25 Ma was most likely caused by fast erosional exhumation in relation Late Cretaceous to the present-day, exhumation of the different tectonic
to rapid surface uplift as a result of an oblique motion on the Bucara- blocks of the Santander Massif was controlled by the activity of primary
manga fault with a strong vertical component. The higher level of exhu- (Bucaramanga) and secondary (Surata, Rio Charta) faults.
mation on the western flank of the Santander Massif is also reflected in The geodynamic cause for accelerated exhumation starting at
its asymmetric topography with a steep western flank defined by the around 25 Ma is most likely related to the start of the collision of the
Bucaramanga fault, and a much smoother eastern flank without a Panama arc with north-western South America around that time (e.g.
major bounding strike-slip fault. So, even if more detailed data are Farris et al., 2011), and based on our thermochronological data, we
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 397

Table 4
HeFTy model results of samples with AFT and ZFT age pairs.

Sample AFT age AFT age error AFT age model GOF TL model ZFT age ZFT age error ZFT age model GOF Model Acceptable Good
(Ma) (Myr) (Ma) (μm) (Ma) (Myr) (Ma) runs fits fits

12SACZ-8 8.77 +4.20, −2.84 8.75 0.99 14.54 ± 1.15 69.8 ±4.1 69.8 1.00 100,000 19,392 7809
13SACZ-10 8.10 +2.66, −2.00 8.12 0.99 14.63 ± 1.62 112.0 ±7.0 112.0 1.00 100,000 23,011 12,986
13SACZ-21 29.6 +12.0, −8.6 29.6 1.00 14.55 ± 1.08 90.3 ±9.7 90.3 1.00 100,000 51,206 34,484
13SACZ-32 11.6 + 2.10, −1.80 11.6 1.00 14.48 ± 1.41 123.0 ±10.0 123 1.00 100,000 26,364 29,178

Unconstrained models
12SACZ-8 8.77 +4.20, −2.84 8.89 0.94 13.83 ± 1.89 69.8 ±4.1 69.3 0.91 100,000 3760 1399
13SACZ-10 8.10 +2.66, −2.00 8.03 0.95 13.76 ± 2.05 112.0 ±7.0 112.0 0.97 100,000 3359 1366
13SACZ-21 29.6 +12.0, −8.60 29.3 0.96 14.22 ± 1.59 90.3 ±9.7 90.4 0.99 100,000 14,184 9406

0
20
40
60
80
Temperature (C)

100
120
140
160
180
200
220
240
260
280
300
320

160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0

Time (Ma)

Fig. 8. Stacked HeFTy models of five samples of the Bucaramanga–Picacho profile. The top-most sample is sample 13SACZ-32 from the Jurassic Giron Formation which constrains the start
of the common thermal history. The weighted mean solution paths and forward models are shown for all samples. Onset of cooling at around 25 Ma is constrained by the central ages of
sample 07SACZ15 and 17. The model shows that cooling at rates of 7–10 °C/Myr occurred between 25 Ma and today.

think that the California–Vetas block was exhumed when changes in the
tectonic regime caused reactivation of pre-existing faults (Kammer,
1993; Osorio et al., 2008). Fault reactivation is a common phenomenon,
as major surface uplift occurred along inversion structures in the North-
ern Andes (e.g. Bermúdez et al., 2010, 2011; Colletta et al., 1997; Cooper
and Addison, 1995; Cortes et al., 2006; Mora et al., 2009).
Thermal modelling of the AFT data of samples 12SACZ-8/8b and
13SACZ-10 from the California–Vetas block profile indicate that acceler-
ated cooling on the high plateau of the central Santander Massif, started
most likely after 11 Ma (Fig. 7A–D). The question arises, whether the
fission-track ages from the high plateau record exhumational cooling
caused by regional tectonic activity and erosion, or do they record
post-magmatic cooling? AFT ages are compatible with a) zircon U-Pb
ages of porphyritic igneous rock of 13.9 ± 0.2 Ma (Mantilla et al.,
2014) and calc-alkaline intrusive rocks of 10.9 ± 0.2 to 8.4 ± 0.2 Ma
(Bissig et al., 2014; Mantilla et al., 2013); b) a ZFT age of 11.3 ±
2.8 Ma from the Bucaramanga gneiss host rock with gold bearing min-
eralization in the Vetas mining district (Urueña, 2014), and c) with a
sericite 40Ar/39Ar age of 10.9 ± 1.5 Ma for the gold mineralization
(Rojas, 2014), also in the California–Vetas mining district. All these age
data are thought to reflect very local high heat flow during the intrusion
of magmas and the upper crustal circulation of hydrothermal fluids in
the gold mining district of the central Santander Massif. For this reason
post-magmatic cooling in the central Santander Massif needs to be con-
sidered. However, the samples of the California–Vetas block profile
were collected far from the area with hydrothermal activity and gold
mineralization, and close to the Rio Charta fault (Fig. 2). Therefore, we
think that post-magmatic/hydrothermal is less likely to have affected
the apatites and that the AFT data record cooling driven by exhumation
during a time of important shortening and exhumation in the Eastern
Cordillera, possibly related to the final accretion of the Panama arc on
the north-western margin of South America (Farris et al., 2011;
Gómez et al., 2005; Parra et al., 2009; Taboada et al., 2000). Therefore,
our AFT data suggest that late Miocene exhumation of the California–
Vetas block occurred at rates of about 0.3–0.4 km/Myr (Fig. 10A). Part
of this exhumation was probably accommodated by the Surata normal
fault. The ~ 37 Ma ZFT cooling age of granodiorite sample 13SACZ-25
to the west of the Surata fault (Fig. 2) could suggest that exhumation
of the Santander Massif along the Surata fault and as that of the Bucara-
manga fault zone may have been active since ~ 40 Ma, similar to the

Table 5
HeFTy models of ZFT data from the Bucaramanga–Picacho profile.

Sample ZFT age (Ma) ZFT age error (Myr) ZFT age model (Ma) GOF Model runs Acceptable fits Good fits

13SACZ-32 123 ±10 123 0.99 10,000 4334 5627

13SACZ-31 120 ±14 120 1.00 10,000 7030 2946
13SACZ-30 111 ±8.0 111 1.00 10,000 4682 3252
07SACZ-17 24.2 ±1.6 24.2 1.00 10,000 684 374
07SACZ-15 19.4 ±2.1 19.4 1.00 10,000 1058 523
398 S. Amaya et al. / Lithos 282–283 (2017) 388–402

2.0
1.9 Fluorapatite: default FT annealing parameters:
1.8 Activation energy for 50% annealing (kcal/mole): 44.6
1.7 Annealing parameter Beta (s^-1) : 9.83E+11
1.6 Thermal Parameters:
1.5 Ts(°C)= 10 Tgrad(°C/km)= 30
Zlw(km)= 30 Tlw(°C)= 910

Exhumation rate (km/Myr)

1.4
1.3
1.2 Radiation-damaged zircon: default FT annealing parameters:
1.1 Activation energy for 50% annealing (kcal/mole): 49.77
1.0 Annealing parameter B (Myr) : 3.16E-22
0.9 Thermal Parameters:
0.8 Ts(°C)= 10 Tgrad(°C/km)= 30
0.7 Zlw(km)= 30 Tlw(°C)= 910
0.6
0.5
0.4
ZFT
0.3
AFT
0.2
0.1
0.0
0 10 20 30 40 50 60 70 80 90 100
Fission-track age (Ma)

Fig. 9. A) Relationship between fission-track age and exhumation rate determined for a 30 °C/km steady-state thermal gradient and an average surface temperature of 10 °C for the apatite
and zircon fission-track systems. The exhumation rates are estimated from the curves of the appropriate thermochronological system. B) Zircon fission-track age and exhumation rate
relationship for radiation damaged and zero damaged zircons.
Data obtained from the age2edot program of M. Brandon (see Ehlers et al., 2005).

Califronia Vetas block ZFT data

4000
West East

3500

3000
Elevation (m)

2500

2000

1500

1000

500

0
-73.05 -73 -72.95 -72.9 -72.85 -72.8 -72.75
Longitude (°W)

Fig. 10. Plot showing longitude of sample locations plotted against elevation for the California–Vetas block zircon fission-track ages. The approximate position of the Rio Charta and Surata
fault lines are indicated as stippled lines to indicate age off-set across fault lines at similar elevations.

interpretation of van der Lelij et al. (2016a). However, as discussed
above about the exhumation of the ZFT PAZ, we would expect to find
~ 40 Ma ZFT cooling ages in rocks that were near the base of the ZFT
PAZ before onset of fast cooling at around 25 Ma (Figs. 6 and 8).
The Rio Charta fault limits the California–Vetas block to the south. Al-
ready Shagam et al. (1984) demonstrated an about 4–6 Myr AFT age off-
set across the Rio Charta fault, with younger ages to the north and older
ages to the south of the fault (see Fig. 7b of Shagam et al., 1984). Our
own AFT data to the north of the Rio Charta fault are consistent with
the Shagam et al. (1984) data. Furthermore, the age off-set is also visible
in the ZFT data (Fig. 10). To the north of the Rio Charta fault most of the
central ZFT ages are ~50 Myr younger than the ZFT ages to the south of

Fig. 11. (a)–(d): Geodynamic model of the evolution of the Santander Massif. Crustal-scale cross-sections from 70 Ma to 1 Ma, showing the relationship between exhumation of the
Santander Massif and regional tectonics. (a) In Late Cretaceous and early Palaeocene times, the collision of the Caribbean Plate with the north-western South American continental margin
produced reactivation of intraplate structures that contributed to the exhumation and surface uplift of the Santander Massif. (b), (c) and (d) Regional tectonic setting, showing the collision
of the Panama Arc with the north-western South American continental and its effects on magmatism, hydrothermal activity and exhumation of the Santander Massif.
In panel a, the position of the Caribbean plate, the Central American Arc, the Panama Arch and direction of movement of the plates is taken and modified by Boschman et al. (2014); the
Caribbean dynamics subducting below South America is taken from Bayona et al. (2013). In panels b, c and d, the position of the Caribbean plate, the Central American Arc, the Panama Arch
and direction of movement of the plates is taken and modified by Boschman et al. (2014); the Caribbean dynamics by subducting in the Pleistocene below South America is taken from
Vargas and Mann (2013).
 S. Amaya et al. / Lithos 282–283 (2017) 388–402 399

D)

C)

B)

A)
400 S. Amaya et al. / Lithos 282–283 (2017) 388–402
the fault (samples 12SACZ-13 and 13SACZ-11). From the exhumed ZFT
partial annealing zone on the western flank of the central Santander Mas-
sif we know that the level of exhumation difference is only about 1–2 km
between zircons with 110 Ma and 60 Ma cooling ages (Fig. 6). The consis-
tent age off-set of both AFT and ZFT ages give the Rio Charta fault a south
dipping normal fault component. This interpretation is supported by ex-
tensional features such as triangular facets, fracture orientation and line-
ation analysis (Velandia, pers. communication).
The ZFT data of the California–Vetas block can also be used to deter-
mine first-order long-term average exhumation rate estimates which
are only about 0.1–0.2 km/Myr since the Late Cretaceous (Fig. 9B). As
these rates are slower than those derived from the AFT data, exhuma-
tion rates must have increased at some time after the Late Cretaceous,
probably first during the late Oligocene to early Miocene, as shown by
the ZFT data near the city of Bucaramanga, and further during the late
Miocene to Pliocene.
The Rio Charta drains the western flank of the California–Vetas block
to the west of the Rio Charta fault, and towards the Surata fault. The
modern river sediment sample, collected near the village of Charta
(Fig. 2), displays a range of AFT ages from 2.8 to 67 Ma, which is typical
for a detrital sample (Fig. 5). However, the central age of this sample is
12.6 ± 2.3 Ma, which corresponds largely to AFT ages found to the south
of the Rio Charta fault on the high plateau (Fig. 2), indicating the back-
ward erosion of the Rio Charta drainage.
6.2. Tectonic model
Using our and previously published thermochronological data a new
model for the thermo-tectonic evolution of the central Santander Massif
is proposed (Fig. 11). Previously Shagam et al. (1984) had identified the
California–Vetas block, but they had neither recognized the importance
of the Surata fault as the north-western and western limit of the Califor-
nia–Vetas block, nor the role it played during early Miocene to late Mio-
cene exhumation of the central Santander Massif. Van der Lelij et al.
(2016a) concluded that exhumation patterns in the Santander Massif
are strongly affected by secondary faults, and that small structural
blocks were exhumed diachronously since at least ~ 18 Ma. However,
these authors did not clearly identify the blocks and what their structur-
al limits are in the central Santander Massif, nor did they recognize that
in this part of the massif there is evidence of exhumation during the
early Miocene, as we did in this study with samples 07SACZ-15 and
07SACZ-17 along the Bucaramanga–Picacho profile (Fig. 6). Therefore,
in contrast to van der Lelij et al. (2016a, 2016b), it is demonstrated in
this study the importance that the Panama arc collision with north-
western South America during the late Oligocene-early Miocene (e.g.
Bayona et al., 2013; Farris et al., 2011; Pindell and Kennan, 2009;
Spikings et al., 2010; van der Lelij et al., 2010; Villagómez and
Spikings, 2013) had on the thermal history of the central Santander
Massif. In Fig. 11 we outline the tectonic evolution of the Santander
Massif from the Maastrichtian to the Pleistocene, and define how exhu-
mation of the Santander Massif was driven to some part by the interac-
tion of the Caribbean and South American plates and the Panama arc.
During the Maastrichtian to early Palaeocene (Fig. 11a) the Caribbean
plate collided with the north-western continental margin of South
America, producing reactivation of intraplate structures that permitted
surface uplift of the Santander Massif and erosion, supplying sediments
to the western and eastern basins. Sediment supply from the Santander
Massif basement rocks is well documented in both domains by the pres-
ence of metamorphic lithic fragments, as well as stable and unstable
heavy mineral associations (Bayona et al., 2009, 2013; Villamil, 1999).
Surface uplift of the Santander Massif was not continuous until the
late Oligocene–early Miocene when the Panama arc collided with
north-western South America and as a consequence, surface uplift and
exhumation of the western flank of the Santander Massif accelerated. Fi-
nally, during the middle Miocene to Pleistocene shallow eastward-dip-
ping subduction of the Caribbean Plate (Fig. 11c, d) produced local
intraplate magmatism and hydrothermal activity in the Santander Mas-
sif and the Paipa–Iza volcanic complex in the Eastern Cordillera during
late Miocene–Pleistocene times (Bernet et al., 2016; Jaramillo et al.,
2005; Pardo et al., 2005). It is possible that fast exhumation and surface
uplift of about 2000 m of the Santander Massif during the late Neogene
was supported by fast erosion and unloading, coupled with an isostatic
response of the massif. Strong erosion of the Santander Massif at this
time has already been suggested by Osorio et al. (2008) based on geo-
morphological evidence, and there is no evidence that the young AFT
ages in the central area of the Santander Massif are the result of partial
annealing caused by Pliocene-Pleistocene magmatic or hydrothermal
activity. Nonetheless, we think that the late Miocene phase of exhuma-
tion of the Santander Massif as well as the volcanic activity of the Paipa-
Iza volcanic complex and the hydrothermal activity in the California–
Vetas area with its associated gold mineralization, are linked to the un-
derlying regional geodynamic processes in the Northern Andes. Slab-
breakoff of the subducting Caribbean plate is most likely responsible
for the magmatic/hydrothermal and seismic activity in the Eastern Cor-
dillera and the Santander Massif (Vargas and Mann, 2013). Today, the
tip of the slab tear may be located beneath Bucaramanga, explaining
the high seismicity of this area (Vargas and Mann, 2013). The flat-slab
subduction of the Caribbean plate was impacted by the collision of the
Panama arc with north-western South America between 12 and 9 Ma
and caused a change in subduction direction and angle of the Caribbean
plate, leading to slab-breakoff propagation to the north (Vargas and
Mann, 2013). The Caldas tear which separates the Caribbean from the
not yet subducting Nazca plate extends for approximately 240 km
from the extinct Sandra oceanic ridge and the Nazca subduction zone
in the west to beneath the Eastern Cordillera and the Paipa volcanic
zone in the east. The Caldas tear may penetrate the upper crust as a
fault zone and provide a conduit for the upward rise of magmas and hy-
drothermal fluids produced by slab dehydration and melting (Vargas
and Mann, 2013), which caused the magmatic and hydrothermal activ-
ity in the California–Vetas block and the Paipa–Iza volcanic complex.
Therefore, the tectonic conditions caused by the accretion of the
Panama arc indenter to the north-western margin of the South
American plate (Gómez et al., 2005; Parra et al., 2009; Taboada et al.,
2000) were responsible for the magmatic and hydrothermal activity in
the California–Vetas Block between about 14 and 8 Ma (Bissig et al.,
2012, 2014; Mantilla et al., 2011, 2013; Raley, 2012; Rodriguez, 2014;
Rojas, 2014; Urueña, 2014). Furthermore, the local thermal perturba-
tion was synchronous with exhumation and possibly surface uplift of
the Santander Massif since the late Miocene, whereas the epithermal
mineralization in the Santander Massif between ~4 and 1.6 Ma (Bissig
et al., 2014) was coeval with the volcanic activity in the Paipa-Iza volca-
nic complex between 5.8 to about 1.8 Ma and contemporaneous with
surface uplift of the Eastern Cordillera during that time (Bernet et al.,
2016; Pardo et al., 2005; Parra et al., 2009).
7. Conclusion
AFT and ZFT data from topographic profiles of the California–Vetas
block and along the western flank of the Santander Massif between Bu-
caramanga and Picacho document the Miocene to present-day cooling
and exhumation history of individual tectonic blocks of the Santander
Massif. Thermal modelling of ZFT data from the California–Vetas block
and Bucaramanga–Picacho reveals cooling since about 25 Ma, most like-
ly caused by exhumation. This exhumation phase is seen in connection
with the initial collision of the Panamá arc with north-western South
America. Long-term average exhumation rates since the Late Cretaceous
were only about 0.1–0.2 km/Myr for the California–Vetas block and Bu-
caramanga–Picacho profile. However, the AFT data of the California–
Vetas block show that exhumation accelerated during the late Miocene
with rates of 0.3–0.4 km/Myr. The accelerated late Miocene exhumation
of the California–Vetas block is here seen in relation to the final accre-
tion of the Panama arc to the western margin of South America and
 S. Amaya et al. / Lithos 282–283 (2017) 388–402
slab-breakoff of the Caribbean plate slab beneath the Eastern Cordillera
and the Santander Massif. Exhumation of the California–Vetas block is
strongly controlled by the Surata and Rio Charta faults. The late Miocene
phase of rapid cooling and exhumation is coeval with localized magmat-
ic and hydrothermal activity in the California–Vetas block gold mining
district. Exhumation is driven by erosion and fault activity, whereas
the magmatic and hydrothermal activity is related to slab-breakoff of
the Caribbean plate and to the advance of the Caldas tear towards the
north of Bucaramanga.
RadialPlotter program of Vermeesch (2009).
Acknowledgements
We thank the Colombian Geological Service Thermochronology lab-
oratory for allowing sample preparation; we especially acknowledge
help and support from Mary Luz Peña, Yolanda Cañon, Lorena Rayo
and Cindy Urueña. This research was supported by COLCIENCIAS grants
110156933549 and UNAL grant 28170 awarded to Carlos Zuluaga, and
by a LabEx International grant and a BQR Sud grant awarded to M.
Bernet at ISTerre, Université Grenoble Alpes. The raw fission-track data
used in this study is available in the supplementary data file or from the
corresponding author: samayaf@unal.edu.co. The manuscript was im-
proved through constructive and detailed reviews by R. van der Lelij
and an anonymous reviewer.
Appendix A. Supplementary data
Supplementary data to this article can be found online at http://dx.
doi.org/10.1016/j.lithos.2017.03.019.
References
Aleman, A., Ramos, V., 2000. Northern Andes. In: Cordani, U.G., et al. (Eds.), Tectonic Evo-
lution of South America. 31st International Geological Congress. FINEP, Rio de Janeiro,
Brazil, pp. 453–480.
Amante, C., Eakins, B.W., 2008. ETOPO1 1 arc-minute global relief model: National
Geophysycal Data Center. NESDIS, NOAA. US Department of Commerce, Boulder, CO.
Bayona, G., Ayala, R.C., Ojeda, C., Cardona, A., Valencia, V., Padrón, C.E., Yoris, F., Salamanca,
J., Garcia, A., 2009. Stratigraphy and provenance of Campanian-Paleogene units in
Cesar sub-basin: contributions to tectonic evolution. Geologia Colombiana 34
(Diciembre, ISSN 0072–0992).
Bayona, G., Cardona, A., Jaramillo, C., Mora, A., Montes, C., Caballero, V., Mahecha, H.,
Lamus, F., Montenegro, O., Jimenez, G., Mesa, A., Valencia, V., 2013. Onset of fault re-
activation in the Eastern Cordillera of Colombia and proximal Llanos Basin; response
to Caribbean–South American convergence in early Palaeogene time. Geological Soci-
ety, London, Special Publications 377. http://dx.doi.org/10.1144/SP377.5.
Bermúdez, M.A., Kohn, B.P., van der Beek, P.A., Bernet, M., O'Sullivan, P.B., Shagam, R.,
2010. Spatial and temporal patterns of exhumation across the Venezuelan Andes: im-
plications for Cenozoic Caribbean geodynamics. Tectonics 29, TC5009. http://dx.doi.
org/10.1029/2009TC002635.
Bermúdez, M., Van Der Beek, P., Bernet, M., 2011. Asynchronus Miocene-Pliocene exhu-
mation of the central Venezuelan Andes. Geology 39:139–142. http://dx.doi.org/10.
1130/G31582.1.
Bernet, M., Amaya, S., Urueña, C., Peña, M., 2016. New thermo- and geochronological con-
straints on the Pliocene-Pleistocene eruption history of the Paipa-Iza volcanic com-
plex, Eastern Cordillera, Colombia. Journal of Volcanology and Geothermal Research
http://dx.doi.org/10.1016/j.jvolgeores.2016.08.013.
Bissig, T., Mantilla, F,.L.C., Rodriguez, M.,.A., Raley, C.A., Hart, C.J.R., 2012. The California-
Vetas District, eastern Cordillera, Santander, Colombia: Late Miocene porphyry and
epithermal mineralization hosted in Proterozoic gneisses and Late Triassic-Early Ju-
rassic intrusions. XVI Congreso Peruano de Geología & SEG2010 Conference, Lima,
Peru, Poster 11.
Bissig, T., Mantilla Figueroa, L.C., Hart, C.J., 2014. Petrochemistry of igneous rocks of the
California-Vetas mining district, Santander, Colombia: implications for northern An-
dean tectonics and porphyry Cu (–Mo, Au) metallogeny. Lithos 200, 355–367.
Boschman, L., vam Hinsbergen, D.J.J., Torsvik, T.H., Spakman, W., Pindell, J.L., 2014. Kine-
matic reconstruction of the Caribbean region since the Early Jurassic. Earth Science
Reviews 138, 102–136.
Brandon, M.T., 1992. Decomposition of fission-track grain-age distributions. American
Journal of Science 292, 535–564.
Cardona, A., Valencia, V., Bayona, G., Jaramillo, C., Ojeda, G., Ruiz, J., 2009. U/Pb LA-MC-ICP-
MS zircon geochronology and geochemistry from a postcollisional biotite granite of
the Baja Guajira Basin, Colombia: implications for Late Cretaceous and Neogene
Caribbean-South American tectonics. Journal of Geology 117:685–692. http://dx.
doi.org/10.1086/605776.
401
Colletta, B., Roure, F., De Toni, B., Loureiro, D., Passalacqua, H., Gou, Y., 1997. Tectonic in-
heritance, crustal architecture, and contrasting structural styles in the Venezuela
Andes. Tectonics 16:777–794. http://dx.doi.org/10.1029/97TC01659.
Cooper, M.A., Addison, F.T., 1995. Basin development and tectonic history of the Llanos
Basin, Eastern Cordillera, and Middle Magdalena Valley, Colombia. The American As-
sociation of Petroleum Geologists Bulletin 79, 1421–1443.
Cortes, M., Colleta, B., Angelier, J., 2006. Structure and tectonics of the central segment of
the Eastern Cordillera of Colombia. Journal of South American Earth Sciences 21,
437–465.
Donelick, R.A., O'Sullivan, P.B., Ketchman, R.A., 2005. Apatite fission track analysis. In:
Reiners, P.W., Ehlers, T.A. (Eds.), Low-temperature Thermochronology. Techniques,
Interpretations and Applications. Reviews in Mineralogy and Geochemistry, Mineral-
ogical Society of America 58, pp. 49–94.
Ehlers, T.A., Chaudhri, T., Kumar, S., Fuller, C., Willett, S.D., Ketcham, R., Brandon, M.T., 2005.
Computational tools for low-temperature thermochronometer interpretation. In:
Reiners, P.W., Ehlers, T.A. (Eds.), Low-temperature Thermochronology. Techniques, In-
terpretations and Applications. Reviews in Mineralogy and Geochemistry, Mineralogical
Society of America 58:pp. 589–622. http://dx.doi.org/10.2138/rmg.2005.58.22.
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 Panamian Isth-
mus during initial collision with South America. Geology 1007–1010.
Feo-Codecido, G., Smith Jr, F.D., Aboud, N., Di Giacomo, E., 1984. Basement and Paleozoic
rocks of the Venezuelan Llanos basins: Geological Society of America. Memoir 162,
173–187.
Galbraith, R.F., 2005. Statistics for Fission Tracks Analysis. Chapman and Hall, CRC, Boca
Raton.
Galbraith, R.F., Laslett, G.M., 1993. Statistical models for mixed fission track ages. Nuclear
Tracks and Radiation Measurements 21, 459–470.
Garver, J.I., Reiners, P.W., Walker, L.J., Ramage, J.M., Perry, S.E., 2005. Implications for
timing of Andean uplift from thermal resetting of radiation — damaged zircon in
the Cordillera Huayhuash, northern Peru. Journal of Geology 113, 117–138.
Goldsmith, R., Marvin, R.F., Mehnert, H.H., 1971. Radiometric ages in the Santander Mas-
sif, Eastern Cordillera, Colombian Andes. United States Geological Survey Professional
Paper 750 (D), 44–49.
Gómez, E., Jordan, T.E., Allmendinger, R.W., Hegarty, K., Kelley, S., 2005. Syntectonic Ceno-
zoic sedimentation in the northern middle Magdalena Valley Basin of Colombia and
implications for exhumation of the Northern Andes. Geological Society of America
Bulletin 117 (5–6), 547–569.
Jaramillo, J.M., Linero, P.R., Garver, J.I., 2005. Neogene`s volcanism in the Cordillera Orien-
tal of the Andes. Colombia. Earth Sci. Res. J. 9, 19–29.
Kammer, A., 1993. Steeply dipping basement faults and associated structures of the San-
tander Massif, Eastern Cordillera, Colombian Andes. Geologia Colombiana 18, 47–64.
Ketcham, R.A., 2005. Forward and inverse modeling of low-temperature
thermochronometry data. In: Reiners, P.W., Ehlers, T.A. (Eds.), Low-
temperature Thermochronology. Techniques, Interpretations and Applica-
tions. Reviews in Mineralogy and Geochemistry, Mineralogical Society of
America 58, pp. 275–314.
Ketcham, R.A., Donelick, R.A., Carlson, W.D., 1999. Variability of apatite fission-track an-
nealing kinetics: III. Extrapolation to geological timescales. American Mineralogist
84, 1235–1255.
Mann, P., Escalona, A., Castillo, M.V., 2006. Regional geologic and tectonic setting of the
Maracaibo supergiant basin, western Venezuela. American Association of Petroleum
Geologists Bulletin 90:445–477. http://dx.doi.org/10.1306/10110505031.
Mantilla, L.C., Mendoza, F.H., Bissig, T., Craig, H., 2011. Nuevas evidencias sobre el
magmatismo Miocenico en el distrito minero de Vetas-California (Macizo de Santan-
der, Cordillera Oriental, Colombia). Boletín de Geología 33, 43–58.
Mantilla, L.C., Bissig, T., Valencia, V., Hart, C.J.R., 2013. The magmatic history of the Vetas–
California mining district, Santander Massif, Eastern Cordillera, Colombia. Journal of
South American Earth Sciences 45, 235–249.
Mantilla, L.C., Cruz, N., Carrillo, J.A., 2014. Consideraciones petrogenéticas y geocronología
de las rocas ígneas porfiríticas aflorantes en la quebrada Ventanas (Municipio
Arboledas, Norte de Santander, Colombia): implicaciones metalogeneticas. Boletin
de Geología 36, 1 (enero–junio de 2014).
Montes, C., Hatcher Jr., R.D., Restrepo, P.A., 2005. Tectonic reconstruction of the northern
Andean blocks: oblique convergence and rotations derived from the kinematics of
the Piedras–Girardot area, Colombia. Tectonophysics 399, 221–250.
Mora, A., Gaona, T., Kley, J., Montoya, D., Parra, M., Quiroz, L.I., Reyes, G., Strecker, M., 2009.
The role of inherited extensional fault segmentation and linkage in contractional oro-
genesis: a reconstruction of Lower Cretaceous inverted rift basins in the Eastern Cor-
dillera of Colombia. Basin Research 21, 111–137.
Ostos, M., Yoris, F., Lallemant, H.G.A., 2005. Overview of the southeast Caribbean–South
American plate boundary zone. Geological Society of America Special Papers 394,
53–89.
Osorio, J.A., Hernandez, C., Torres, E., Botero, P., 2008. Modelo Geodinámico del
Macizo de Santander. Informe interno Subdirección de Geología Básica.
INGEOMINAS (113 pp.).
Pardo, N., Cepeda, H., Jaramillo, J.J., 2005. The Paipa Volcano, Eastern Cordillera of
Colombia,South America: volcanic stratigraphy. Earth Sciences Research Journal 9, 3–18.
Parra, M., Mora, A., Sobel, E.R., Strecker, M.R., González, R., 2009. Episodic orogenic front mi-
gration in the northern Andes: constraints from low-temperature thermochronology in
the Eastern Cordillera, Colombia. Tectonics 28, TC:4004. http://dx.doi.org/10.1029/
2008TC002423.
Pindell, J.L., Kennan, L., 2009. Tectonic evolution of the Gulf of Mexico, Caribbean and
northern South America in the mantle reference frame: an update. In: James, K.H.,
Lorente, M.A., Pindell, J.L. (Eds.), The Origin and Evolution of the Caribbean Plate. Geo-
logical Society London, Special Publications vol. 328, pp. 1–55.
402 S. Amaya et al. / Lithos 282–283 (2017) 388–402
Rahn, M.K., Brandon, M.T., Batt, G.E., Garver, J.I., 2004. A zero-damage model for fission–
track annealing in zircon. American Mineralogist 89, 473–484.
Raley, Ch., 2012. Mineralogical Characterization of Sulfide Mineralization, Alteration and
Microthermometry of Related Fluid Inclusions of the La Plata Prospect, Colombia.
(Thesis submitted in partial fulfillment of the requirements for the degree of bachelor
of science (honours) in the Faculty of Science (Geological Sciences)). The University
of British Colombia, (Vancouver). Canada.
Reiners, P.W., Brandon, M.T., 2006. Using thermochronology to understand orogenic ero-
sion. Annual Review of Earth and Planetary Sciences 34, 419–466.
Rodriguez, M.A.L., 2014. Geology, Alteration, Mineralization and Hydrothermal Evolution
of the La Bodega-La Mascota Deposits, California-Vetas Mining District, Eastern Cor-
dillera of Colombia, Northern Andes. (Unpubl. MSc thesis). The University of British
Columbia (325 pp.).
Rojas, S., 2014. Metalogenia de las mineralizaciones auríferas en la zona de Vetas, Santan-
der. (MSc thesis). Universidad Nacional de Colombia, Bogota (260 pp.).
Shagam, R., Khon, B.P., Banks, P.O., Dasch, L.E., Vargas, R., Rodriguez, G.I., Pintente, N.,
1984. Tectonic implications of Cretaceous–Pliocene fission-track ages from rocks of
the circum-Maracaibo Basin region of western Venezuela and eastern Colombia. Geo-
logical Society of America Memoirs 162, 385–412.
Spikings, R., Crowhurst, P.V., Winkler, W., Villagomez, D., 2010. Syn- and post-
accretionary cooling history of the Ecuadorian Andes constrained by their in-situ
and detrital thermochronometric record. Journal of South American Earth Sciences
30 (3-4):121–133. http://dx.doi.org/10.1016/j.jsames.2010.04.002.
Taboada, A., Rivera, L.A., Fuenzalida, A., Cisternas, A., Philip, H., Bijwaard, H., Olaya, J., Rivera,
C., 2000. Geodynamics of the northern Andes: subductions and intracontinental defor-
mation (Colombia). Tectonics 19:787–813. http://dx.doi.org/10.1029/2000TC900004.
Urueña, C.L., 2014. Metamorfismo, Exhumación y Termocronología del Neis de Bucara-
manga. (Macizo de Santander, Colombia). (MSc tesis). Universidad Nacional de
Colombia (209 pp.).
van der Lelij, R., Spikings, R.A., Kerr, A.C., Kounov, A., Cosca, M., Vhew, D., Villagomez, D.,
2010. Thermochronology and tectonics of the. Leeward Antilles: Evolution of the
southern Caribbean Plate boundary zone: Tectonics 29 (6), 1–30.
van der Lelij, R., 2013. Reconstructing North-Western Gondwana with Implications for the
Evolution of the Iapetus and Rheic Oceans: A Geochronological, Thermochronological
and Geochemical Study. (Ph.D. Thesis). University of Geneva, Swiss.
van der Lelij, R., Spikings, R., Mora, A., 2016a. Thermochronology and tectonics of the
Merida Andes and the Santander Massif, NW South America. Lithos 3803. http://dx.
doi.org/10.1016/j.lithos.2016.01.006 (PII: S0024–4937(16)00032–3, Reference).
van der Lelij, R., Spikings, R., Ulianov, A., Chiaradia, M., Mora, A., 2016b. Palaeozoic to
Early Jurassic history of the northwestern corner of Gondwana, and implications
for the evolution of the Iapetus, Rheic and Pacific Oceans. Gondwana Research
31, 271–294.
Vargas, C.A., Mann, P., 2013. Tearing and breaking off of subducted slabs as the result of
collision of the Panama Arc-Indenter with Northwestern South America. Bulletin of
the Seismological Society of America 103, 2025–2046.
Vermeesch, P., 2009. RadialPlotter: a Java application for fission-track. Luminescence and
other radial plots. Radiation Measurements 44, 409–410.
Villagómez, D., Spikings, R., 2013. Thermochronology and tectonics of the Central and
Western Cordilleras of Colombia: Early Cretaceous-Tertiary evolution of the Northern
Andes. Lithos 160 (1):228–249. http://dx.doi.org/10.1016/j.lithos.2012.12.008.
Villagómez, D., Spikings, R., Mora, A., Guzmán, G., Ojeda, G., Cortés, E., Van der Lelij, R.,
2011. Vertical tectonics at a continental crust-oceanic plateau plate boundary zone:
fission track thermochronology of the Sierra Nevada de Santa Marta, Colombia. Tec-
tonics 30. http://dx.doi.org/10.1029/2010TC002835.
Villamil, T., 1999. Campanian-Miocene tectonostratigraphy, depocenter evolution and
basin development of Colombia and Western Venezuela. Palaeogeography,
Palaeoclimatology, Palaeoecology 153, 239–275.
Ward, D.E., Goldsmith, R., 1973. Geología de los Cuadrángulos H12 Bucaramanga y H13
Pamplona, Departamento de Santander. República de Colombia, Ministerio de
Minas y Petróleos, Instituto Nacional de Investigaciones Geológico - Mineras.
Ward, D.E., Goldsmith, R., Cruz, J., Restrepo, H.A., 1974. Geology of Quadrangles H-12, H-
13, and Parts of I-12 and I-13, (Zone III), in Northeastern Santander Department,
Colombia: U.S. Geological Survey. Open file report 74–258 (466 pp.).
Willett, S.D., Brandon, M.T., 2013. Some analytical methods for converting
thermochronometric age to erosion rate. Geochemistry, Geophysics, Geosystems
14. http://dx.doi.org/10.1029/2012GC004279 (ISSN: 1525–2027).