Chapter 5

Spatial-Temporal Migration of Granitoid

Magmatism and the Phanerozoic Tectono-
Magmatic Evolution of the Colombian
Andes

Hildebrando Leal-Mejía, Robert P. Shaw,
and Joan Carles Melgarejo i Draper

5.1  Introduction

Granitoid magmatic rocks form important constituents of the Phanerozoic record of

the Colombian Andes. From the viewpoint of modern-day geological exposure,
granitoids manifest as volumetrically abundant plutonic, hypabyssal and volcanic
rocks, occurring as major batholiths, stocks, dike swarms and extensive volcanic
sequences. During both proto-Andean and Andean times, these granitoid expres-
sions form integral components of the temporal-tectonic evolution of Colombia and
of Northwestern South America as a whole.
Granitoid magmatism sensu lato and its relationships to plate tectonics and
global tectonic setting have become increasingly understood (e.g. Wilson 1989;
Barbarin 1999; Hamilton 1994; Stern 2002). In this context, an integral analysis of
the distribution, age, lithogeochemical and isotopic composition of granitoid rocks
in the Colombian Andes can in due process shed light upon the nature, timing and
framework of Colombian tectonic development.
Granitoid magmatism in the Colombian Andes includes a lithogeochemically
and texturally diverse suite of holocrystalline phaneritic plutonic, hypabyssal

Electronic supplementary material: The online version of this chapter (https://doi.org/10.1007/978-

3-319-76132-9_5) contains supplementary material, which is available to authorized users.
H. Leal-Mejía (*)
Mineral Deposit Research Unit (MDRU), The University of British Columbia (UBC),
Vancouver, BC, Canada
Departament de Mineralogia, Petrologia i Geologia Aplicada, Facultat de Ciències de la
Terra, Universitat de Barcelona, Barcelona, Catalonia, Spain
e-mail: [email protected]
R. P. Shaw · J. C. Melgarejo i Draper
Departament de Mineralogia, Petrologia i Geologia Aplicada, Facultat de Ciències de la
Terra, Universitat de Barcelona, Barcelona, Catalonia, Spain

© Springer Nature Switzerland AG 2019 253

F. Cediel, R. P. Shaw (eds.), Geology and Tectonics of Northwestern South America,
Frontiers in Earth Sciences, https://doi.org/10.1007/978-3-319-76132-9_5
254 H. Leal-Mejía et al.

p­ orphyritic and volcanic/volcano-sedimentary rocks and occasionally their meta-

morphosed equivalents, which, with respect to age, span the entire Phanerozoic.
Although the cartographic limits of most plutons and volcanic units have been
established and many have been described or studied on an individual basis, there
are few works which assess the regional distribution vs. temporal development of
Colombian Phanerozoic-aged granitoid magmatism as a whole. This is certainly
due to the fact that there are limited studies in Colombia which combine modern-­
day (post-1995) lithogeochemical, geochronological, petrographic and isotopic
studies to multiple intrusives of various ages on a regional level, in an attempt to
specifically trace the tectono-magmatic evolution of the entire region over the
breadth of the entire Phanerozoic.
The present synthesis of granitoid magmatism throughout the Colombian Andes
is based primarily upon geological analysis combined with zircon U-Pb age dates,
major-minor-trace element lithogeochemical analyses and isotope geochemistry
studies for granitoid intrusive and volcanic rocks, generated by various authors
since ca. 1995. The majority of the new data has been published since ca. 2006.
The composite data set provides a more precise framework upon which to recon-
struct the temporal-spatial development and migration of Phanerozoic granitoid
magmatism in Colombia. When this information is integrated with updated kine-
matic models for the tectonic evolution of the Northern Andean region, refined con-
clusions can be drawn about the distribution, nature, timing, migration and controls
upon Phanerozoic granitoid magmatism in the Colombian Andes.
The cornerstone of this presentation is derived from our own published
(e.g. Cediel et al. 2003; Leal-Mejía et al. 2011) and unpublished investigations (e.g.
Cediel et al. 1994; Cediel and Cáceres 2000; Leal-Mejía 2011), coupled with impor-
tant newer studies from numerous third-party authors (cited below), the combined
data for which permit a deeper understanding of Colombian and Northern Andean
tectono-magmatic development during the period spanning the early Paleozoic
through to the conformation of the active volcanic arcs of the present-day Northern
Andes.
The widespread distribution of granitoid rocks in the geologic record of the
Colombian Andes was recognized in numerous historic works (e.g. Trumpy 1949;
Singewald 1950; Gansser 1955, 1973; Nelson 1957; Campbell 1974), and the
approximate age of many of the major batholiths and extensive volcanic sequences
was broadly inferred based upon field relationships. With the advent of more precise
lithogeochemical analyses and radiometric age dating and the initiation of inte-
grated regional geological studies (e.g. Goldsmith et al. 1971; Feininger et al. 1972;
Tschanz et al. 1974; Feininger and Botero 1982; Sillitoe et al. 1982; Alvarez 1983;
Aspden et al. 1987), a more detailed picture of Phanerozoic granitoid magmatism in
the Colombian Andes emerged.
Early lithogeochemical studies, however, did not include important batholiths in
the northern and eastern portion of the Colombian cordilleran system (e.g. the Sierra
Nevada de Santa Marta, the Santander Massif and Mocoa), in the Serranía de San
Lucas, Segovia or Chocó Arc. Additionally, in the existing database, most trace ele-
ments, including the rare-earth elements, were often not analysed, and results were
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 255

generally not placed into a tectonic framework. With respect to radiometric age
dating, the early database from the 1970s and 1980s consisted almost exclusively of
K-Ar and Rb-Sr (isochron) ages, which had large margins of error and were subse-
quently proven in many cases, in Colombia and elsewhere, to be imprecise, erratic
and of poor repeatability.
In a first integrated attempt to place Meso-Cenozoic granitoids in Colombia into
a modern-day tectonic framework using available distribution, lithogeochemical
and radiometric age data, Aspden et al. (1987) presented a well-conceived synthesis
of subduction-related magmatism in Colombia. Based upon radiometric age data
compiled from numerous sources (see Maya 1992), they identified five magmatic
episodes (Triassic, Jurassic, Cretaceous, Paleogene and Neogene) and outlined a
regional tectonic framework for the evolution of subduction-related magmatism in
the Northern Andes. Additionally, they identified some of the key factors and con-
trols influencing granitoid arc development during the Meso-Cenozoic.
Studies presenting high-quality U-Pb (zircon) dating techniques for magmatic
rocks permit a more accurate assessment of the crystallization and inheritance
age(s) of granitoids. Such studies in the Colombian Andes, at times combined with
incipient Sr, Nd and Pb isotope data and more complete lithogeochemical analyses,
began to trickle in ca. 1995. Initial studies addressed specific plutons or sub-regions
and included a limited number of samples per intrusive body. Examples of such
works include Dörr et al. (1995) for the Paramo Rico and Santa Barbara batholiths
of the Santander Massif; Ordoñez et al. (2001) for the Sonsón Batholith; Altenberger
and Concha (2005) for the northern Ibagué Batholith (K-Ar ages only); Vinasco
(2004) and Vinasco et al. (2006) for the Permo-Triassic granitoids found throughout
the northern Central Cordillera and elsewhere; Ibañez-Mejía et  al. (2007) and
Restrepo-Moreno et al. (2007) for the Antioquia Batholith; Ordoñez-Carmona et al.
(2007a, b) for the Antioquia, Segovia and Sabanalarga batholiths; Weber et  al.
(2015) for the Santa Fé Batholith; Mejía et al. (2008), Duque (2009), and Cardona
et al. (2011) for the late Cretaceous-Paleocene intrusives of the Sierra Nevada de
Santa Marta (i.e. Santa Marta Batholith); Correa et al. (2006) for the Altavista and
San Diego Stocks (satellites to the Antioquia Batholith); Villagómez (2010) and
Villagómez et al. (2011) for the Ibagué, Antioquia and Buga batholiths; Bustamante
et al. (2010) and Zapata et al. (2016) for plutons of the southern Ibagué, Mocoa and
Garzón suites; Mantilla et al. (2012) and Bissig et al. (2014) for Phanerozoic intru-
sive rocks of the Santander Massif; Bayona et  al. (2012) and Bustamante et  al.
(2017) for Paleocene plutons of the Central Cordillera; and Montes et  al. 2012,
2015) for the Mandé and Acandí batholiths and associated plutons. Some of the
earlier works represented incomplete or in-process studies and were presented in
conference-related abstracts containing limited background information which are
difficult to access and which do not permit the full geological evaluation of the
numerical data or of the derived conclusions. Others are published in well-circulated
international journals and are readily accessible for detailed review.
Leal-Mejía (2011) presented an integrated investigation of Phanerozoic gran-
itoid magmatism related to gold metallogeny in the Colombian Andes. This tempo-
ral, lithogeochemical and tectono-magmatic study was based upon a review and
256 H. Leal-Mejía et al.

compilation of the historic to recent data cited above, in addition to the presentation
of 107 new high-precision U-Pb (zircon) dates for intrusive and volcanic rocks, sup-
ported by new K-Ar, Ar-Ar and Re-Os ages, as well as Sr, Nd and Pb isotope data
and 282 research-quality whole-rock major-minor-trace-REE lithogeochemical
analyses. The study included new data from many previously un- or understudied
Phanerozoic granitoids in the Colombian Andes such as the Pueblo Bello, Norosí-
San Martín de Loba, Segovia, Ibagué (north and south), Mariquita and Antioquia,
Buga, Sonsón-­Nariño, Mandé, Piedrancha-La Llanada and Farallones batholiths. It
identified various Permo-Triassic granitoids which had previously been mapped as
Precambrian, Jurassic or Paleocene in age. Additionally, many smaller holocrystal-
line stocks such as El Carmen, Mocoa, Irra, Jejenes, Frontino and Támesis, and
numerous clusters of Neogene hypabyssal porphyry stocks observed along the mar-
gins of the Central and Western Cordilleras and in the Santander Massif, were ana-
lysed. The present chapter draws heavily on the data and conclusions presented by
Leal-Mejía (2011).
Important contributions to the U-Pb age date, isotopic and lithogeochemical
database for Northern Andean Paleozoic through Jurassic granitoid intrusions have
recently been supplied by researchers from the University of Geneva. Such works
include Villagómez (2010) and Villagómez et al. (2011), applicable to portions of
the Western and Central Cordilleras and Sierra Nevada de Santa Marta; Van der
Lelij (2013), Van der Lelij et al. (2016) and Spikings et al. (2015), applicable to the
Santander Massif and the Sierra de Mérida (Venezuela); and Cochrane (2013),
Cochrane et  al. (2014a, b) and Spikings et  al. (2015), applicable to the Garzón
Massif and various Permo-Triassic and Jurassic granitoids outcropping along the
margins of Colombia’s Central Cordillera. Pertinent data from these works have
been reviewed, and conclusions from these authors have been integrated into the
ensuing text and graphics of this chapter.

5.2  P
 hanerozoic Tectonic Framework of the Colombian
Andes

The Colombian Andes is contained within the North Andes (Bird 2003) or Northern
Andean Block (Cediel et al. 2003; Cediel 2011; Fig. 5.1). From a geographic stand-
point, the Northern Andean Block includes the northernmost Peruvian Andes (north
of the Huancabamba Deflection), in addition to the cordilleran systems of Ecuador,
Colombia and Venezuela and the eastern Chocó Arc segment of the Panamá double
arc. The evolution of the region and its modern-day geologic, tectonic and physio-
graphic expression is the result of complex interactions between oceanic and conti-
nental tectonic plates, beginning in the mid-Proterozoic (e.g. Cediel et  al. 1994;
Ramos 1999; Restrepo-Pace and Cediel 2010; Cediel 2011). Since the Meso-­
Cenozoic, no less than four plates, including the South American continental block
(western Guiana Shield) and the Farallon, Caribbean and Nazca-Cocos oceanic
plates, have been involved.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 257

Fig. 5.1  Location of Colombia and the Northern Andean Block (Cediel et al. 2003) in relation to
microplates of northwestern South America and surrounding region as defined by Bird (2003).
Present-day microplate relative movement vectors and velocities in mm/a after Bird (2003)

Internally, the Northern Andean Block itself comprises a microplate hosting

numerous allochthonous and parautochthonous tectonic slivers of mixed oceanic,
peri-cratonic and continental affinity. The allochthonous and parautochthonous
crustal fragments have accumulated tectonically along the northwest margin of
South America during successive accretionary events and are presently broadly
bound by the Guiana Shield segment of the South American Plate to the east, the
Pacific (Nazca) Plate to the west and the Caribbean Plate to the north (Fig. 5.1).
Numerous tectonic models for the development of the Colombian Andes and
adjacent Pacific and Caribbean regions have been proposed over to past 40 years
(e.g. Etayo-Serna et al. 1983; Burke et al. 1984; Kellogg et al. 1985; Restrepo and
Toussaint 1988; Pindell et al. 1988; Cediel et al. 1994; Taboada et al. 2000). It is
now recognized that the present-day Northern Andean configuration is the result of
a history involving collision and accretion of allochthonous terranes and the devel-
opment of subduction-related arcs along the northwestern South American margin
(Restrepo and Toussaint 1988). Although emphasizing locally important features
regarding Colombian tectonic development, early tectonic models suffered from a
variety of geographic and/or temporal limitations, and in some cases the conclusions
are now known to be invalid.
258 H. Leal-Mejía et al.

More recent regional tectonic models recognize the fact that an accurate account
of Colombian tectonic evolution cannot be obtained through the imposition of
geopolitical limits upon model construction. The understanding of Colombian tec-
tonics involves an understanding of the integrated geological evolution of the entire
Northern Andean region, from northern Perú through Venezuela and Panamá.
Tectonic models dealing with pre- and early Phanerozoic time remain elusive and
necessarily generalized due to the highly fragmented geological record remaining
from this extended time period. More recent works which address the Proterozoic
and early Phanerozoic tectonic record of the Colombian Andes include Restrepo-­
Pace (1992, 1995), Cediel et al. (1994), Ramos (1999), Cediel and Cáceres (2000),
Cordani et al. (2005), Keppie (2008), Ibañez-Mejía et al. (2011), Restrepo-Pace and
Cediel (2010), Cediel (2011), and Van der Lelij et al. (2016). Similarly, the mid-late
Paleozoic record remains controversial, and relatively limited work has been com-
pleted for this time period, especially in the metamorphic rocks of the Colombian
Central Cordillera, beyond important contributions by Restrepo-Pace (1992),
Vinasco (2004), Vinasco et al. (2006), and Cochrane et al. (2014a). With respect to
the Meso-Cenozoic, various recent works recognize the critical importance of the
evolution and demise of the Farallon Plate and the birth, evolution and emplacement
of the Caribbean and Nazca-Cocos plates, with respect to the inboard tectono-­
magmatic development of the Colombian Andes, especially during the Northern
Andean Orogeny (e.g. Cediel et al. 1994; Cediel and Cáceres 2000; Maresch et al.
2000; Taboada et al. 2000; Pindell and Kennan 2001; Cediel et al. 2003; Kerr et al.
2003; Kennan and Pindell 2009; Leal-Mejía et al. 2011; Montes et al. 2012; Spikings
et al. 2015).
In a Northern Andean analysis spanning the Proterozoic to the present, Cediel
et al. (2003) describe more than 30 litho-tectonic and morpho-structural units (ter-
ranes, terrane assemblages, physiographic and morpho-structural domains, etc.),
contained within four major tectonic realms (Fig. 5.2). Each realm records distinct
and in some cases unique internal deformation styles as a response to progressive
westward accretionary continental growth along the northwestern Guiana Shield
(Amazon Craton) margin. Tectonic realms and terrane assemblages are delimited
by important regional-scale sutures and fault systems. Within this chapter, we will
adhere primarily to the tectonic nomenclature presented by and updated from
Cediel et al. (2003; Fig. 5.2), much of which is derived from historic detailed anal-
yses of Colombian tectonics, as presented by Etayo-Serna et al. (1983) and Cediel
et al. (1994).
The kinematic evolution of the Caribbean Plate and resulting large-scale Northern
Andean-Caribbean Plate interactions have been depicted in Meso-Cenozoic tec-
tonic reconstructions presented by various authors including Cediel and Cáceres
(2000), Pindell and Kennan (2001), Kerr et al. (2003), Cediel et al. (2003), Kennan
and Pindell (2009), Wright and Wyld (2011), Montes et al. (2012), and Nerlich et al.
(2014), amongst others.
Cediel and Cáceres (2000), Cediel et  al. (2003), and Cediel (2011) observed
that the geotectonic evolution of Colombia can be separated into pre-Northern
Andean Orogeny events (i.e. events prior to approximately the Aptian) and
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 259

Fig. 5.2  Distribution of Phanerozoic granitoids in relation to the major litho-tectonic and morpho-­
structural elements of the Colombian Andes. (Granitoid shapes modified after Cediel and Cáceres
2000; Gómez et al. 2007; Gómez et al. 2015a. Litho-tectonic base map adapted from Cediel et al.
2003). Age ranges based upon U-Pb (zircon) age dates compiled herein

Northern Andean Orogeny-related events beginning in the Aptian onwards. In this

context, these authors present tectono-stratigraphic reconstructions for the mid-late
Proterozoic (Orinoquiense event), the early to middle Paleozoic (Quetame/
Caparonesis event), the mid- and late Paleozoic to early and mid-Mesozoic (Bolívar
Aulacogen) and the Aptian to recent (Northern Andean Orogeny). Similar recon-
260 H. Leal-Mejía et al.

structions which focus upon the age, distribution, migration and nature of granitoid
magmatism within the tectonic configuration/evolution of the region will be presented
within the present work.

5.2.1  Tectonic Elements of the Colombian Andes

The Northern Andes represents a complex composite orogen with an extended and
intricate tectonic history involving continental collision and rifting, prolonged
taphrogeny, transpressive accertionary orogenesis, tectonic inversion and tectonic
detachment and drift, which span the mid-Proterozoic to Recent (Cediel and Cáceres
2000; Cediel et al. 2003; Cediel 2011). Exposures of granitoid intrusive ± volcanic
rocks form the vestiges of rift-related magmatism and of subduction-related mag-
matic arcs and arc segments, generated along the Northern Andean margin since
pre-Andean and proto-Andean as well as throughout Andean times, beginning in at
least the early Paleozoic. Arc segments of varying ages are often separated by major
arc-parallel or arc-transverse fault systems which, based upon geotectonic analysis,
have been identified as sutures or paleo-transform faults. In this section we describe
the important tectonic elements which form the basement complexes to granitoid
plutons ± volcanic sequences emplaced throughout the Phanerozoic record of the
Colombian Andes. A schematic representation of the tectonic elements forming
basement to Phanerozoic granitoids within the Colombian Andes is shown in
Fig. 5.2. The sequential development of Phanerozoic granitoid magmatism within
the context of these tectonic elements is proposed in a synthesis containing descrip-
tive text, tectonic reconstructions and time-space diagrams, revealed at the end of
this presentation.
Guiana Shield Realm (GSR)  The eastern foreland of the present-day Colombian
Andes is underlain by cratonic rocks of the western Guiana Shield, for which
recorded radiometric age dates ranging from ca. 2.5 to 1.5 Ga demonstrate a gen-
eral east-to-west younging trend (e.g. Kroonenberg 1982; Priem et al. 1982; Priem
et al. 1989; Cordani et al. 2005; Ibañez-Mejía et al. 2011). The westernmost mar-
gin of the Shield is marked by a ca. 1.2–0.95 Ga belt of granulite grade metamor-
phic rocks (Fig. 5.2) (Kroonenberg 1982; Priem et al. 1989; Restrepo-Pace 1995;
Cordani et al. 2005; Cardona et al. 2010a; Restrepo-Pace and Cediel 2010; Ibañez-
Mejía et al. 2011), recorded in outcrops in the Sierra Nevada de Santa Marta and
Santander and Garzón Massifs and considered to represent Grenvillian-age conti-
nent-continent interaction along the Bucaramanga–Santa Marta–Garzón fault and
suture system (Cediel et  al. 2003; Cediel 2011) during the final assembly of
Rhodinia (Cordani et al. 2005; Ibañez-Mejía et al. 2011). The resulting tectono-
thermal metamorphic event has been referred to in Colombia as the Orinoquiense
Orogen (Kroonenberg 1982; Restrepo-Pace 1995; Cediel and Cáceres 2000;
Restrepo-Pace and Cediel 2010; Cediel 2011) or Putumayo Orogen (Ibañez-Mejía
et al. 2011).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 261

Maracaibo Sub-plate Realm (MSP)  The MSP is a composite tectonic realm also
underlain by the Guiana Shield, but much of its uplift history is linked to the Meso-­
Cenozoic tectonic evolution of the region. Its northern limit, in contact with the
Caribbean Plate, is defined by the dextral Oca-El Pilar fault system and the Santa
Marta thrust front (Fig.  5.2), whilst its west margin, in contact with the Central
Tectonic Realm, is defined by the reactivated Bucaramanga-Santa-Marta fault.
Topographic relief is provided by the Santander and Quetame Massifs, the Sierra de
Mérida, the Serrania de Perijá and the Sierra Nevada de Santa Marta, the uplift his-
tory of which is linked to detachment and NW-vergent tectonic float during the
Meso-Cenozoic (Cediel and Cáceres 2000; Cediel et al. 2003). The MSP contains
numerous litho-tectonic and morpho-structural components, including exhumed
Proterozoic and early Paleozoic basement massifs (Santander, Quetame, Floresta).
Late Triassic-Jurassic ensialic extensional volcano-sedimentary basins are exposed
along the Santander Massif, Sierra Nevada de Santa Marta and Serranía de Perijá.
Uplift of the Santander Massif and Sierra Nevada de Santa Marta has unroofed
important holocrystalline granitoid batholiths of early Paleozoic and latest Triassic-­
Jurassic age.

Central Tectonic Realm (CTR)  The CTR (originally termed Central Continental
Sub-plate (CCSP) by Cediel et al. 2003) is a composite, temporally and composi-
tionally heterogeneous realm which occupies a wedge located between the Guiana
Shield Realm, the Maracaibo Sub-plate Realm, and the Western Tectonic Realm
(Fig. 5.2). It forms the basement complex which underlies the entire central portion
of the Colombian Andes. The CTR is comprised of a variety of litho-tectonic and
morpho-structural entities. Its composite metamorphic basement consists of the
Proterozoic Chicamocha Terrane and the Paleozoic to early Mesozoic Cajamarca-­
Valdivia Terrane (CA-VA). Superimposed upon these core components are Jurassic
magmatic arc segments including the San Lucas, Ibagué and Segovia blocks, the
late Cretaceous Antioquian Batholith and additional Paleocene plutons to the south,
and the Pleistocene to Recent Northern Andean volcanic arc, all of which dominate
Colombia’s physiographic Central Cordillera. The Lower, Middle and Upper
Magdalena basins and Colombia’s geologic Eastern Cordillera (EC) were also devel-
oped upon CTR metamorphic basement. The Chicamocha and Cajamarca-­Valdivia
constituents of the CTR are allochthonous to parautochthonous with respect to the
Guiana Shield autochthon, having been sutured to the shield in pre-­Andean times.
The Mesozoic to Recent components of the CTR are considered to be autochthonous
with respect to Chicamocha-CA-CV metamorphic basement.
The oldest constituent of the CTR is the Precambrian Chicamocha Terrane, inter-
preted as an embedded fragment of composite parautochthonous to allochthonous
continental crust containing relict fragments of Oaxaquia basement (Keppie and
Ortega-Gutierrez 2010) and Oaxaquian-Colombian fringing volcano-magmatic arcs
(Ibañez-Mejía et  al. 2011; Cediel 2011), amalgamated with tectonic rafts of the
westernmost Guiana Shield and welded to the autochtonous Guiana Shield margin
during the ca. 1.0–0.95 Ga event (Cediel 2011). It is represented by Mesoproterozoic
262 H. Leal-Mejía et al.

inliers exposed along portions of Colombia’s Central Cordillera (Cediel and Cáceres
2000; Cordani et al. 2005; Cediel 2011; Leal-Mejía 2011), including in the Serranía
de San Lucas where Cuadros et al. (2014) documented early Mesoproterozoic base-
ment containing a bimodal assemblage of within-plate granitoids and oceanic island
and possibly underplate metamafic rocks of juvenile, mantle-derived character.
They published zircon U-Pb crystallization ages ranging from ca. 1.54 to 1.50 Ga
for high-grade metagranitoids which exhibit ~1 Ga (Grenvillian) overprinting.
Chicamocha is bound to the west by the composite Cajamarca-Valdivia Terrane
(Cediel and Cáceres 2000; Cediel et al. 2003; Cediel 2011) which broadly coincides
with the Central Andean Terrane as described by Restrepo-Pace (1992). Cajamarca-­
Valdivia contains amphibolitic, graphitic and semi-pelitic schists and marbles,
metamorphosed to greenschist through epidote amphibolite grade and generally
assigned a Neoproterozoic to early Paleozoic age (Feininger et al. 1972; Restrepo-­
Pace 1992; Cediel and Cáceres 2000; Ordoñez-Carmona et al. 2006; Cediel 2011;
Spikings et al. 2015), in agreement with Ediacaran to Cambrian C-isotope stratigra-
phy ages for contained carbonates, as published by Silva et al. (2005). Based upon
geochemical and geological characterization studies presented by Restrepo-Pace
(1992), Cajamarca-Valdivia represents a pericratonic island arc and continental
margin accretionary prism assemblage, accreted along the western Chicamocha
Terrane. Cajamarca-Valdivia forms the basement to Carboniferous and Permian
through mid-late Triassic gneissic granitoids, meta-amphibolites and peraluminous
granites associated with the assembly and break-up of Pangaea (Vinasco et al. 2006;
Cardona et al. 2010b; Cochrane 2013; Cochrane et al. 2014a; Spikings et al. 2015).
González (2001) and Cediel (2011) suggest the terrane also contains tectonic floats
of Mesoproterozoic metamorphic continental basement rocks, including the Puqui
and El Retiro-Rio Negro gneisses, although Ordoñez-Carmona et al. (2006) note
that these units produce broadly Permo-Triassic radiometric age dates and hence
are better considered members of the Permo-Triassic suite of Vinasco et al. (2006).
The Palestina fault system denotes the suture between the Chicamocha and
Cajamarca-­Valdivia Terrane assemblages. The trace of the modern-day Palestina
fault system is the result of various reactivations during the Meso-Cenozoic Northern
Andean orogeny (Feininger 1970; Cediel and Cáceres 2000). Associated structures
include the Chapeton and Pericos faults.
It is important to note that present-day geological mapping does not permit the
precise distribution of Proterozoic, early Paleozoic and Permo-Triassic constituents
of CA-VA (Cediel and Cáceres 2000; Gómez et al. 2015a), and zircon-based U-Pb
age dating is only beginning to reveal the complexity of the CA-VA assemblage. For
example, various units which were formerly thought to be of Proterozoic or early
Paleozoic age are now known to belong to the Permo-Triassic assemblage (Restrepo-­
Pace 1992; Ordoñez-Carmona et al. 2006; Vinasco et al. 2006; Leal-Mejía 2011;
Spikings et al. 2015).
The timing of CA-VA assemblage and accretion along the western Chicamocha
margin and the origins of the Palestina fault and suture system have been examined
by various authors (Feininger 1970; Cediel et al. 1994; Restrepo-Pace 1995; Cediel
and Cáceres 2000; Cediel et al. 2003; Restrepo-Pace and Cediel 2010; Cediel 2011).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 263

Regional tectono-sedimentary analysis (Cediel et al. 1994), arc-related magmatic

patterns and available metamorphic age dates (Restrepo-Pace 1995; Van der Lelij
et al. 2016) suggest early Paleozoic accretion of CA-VA during the Quetame Orogeny
(Cediel and Cáceres 2000; Cediel 2011). Peak Barrovian conditions, including
emplacement of syn-kinematic granitoids, were attained at ca. 477–472  Ma
(Restrepo-Pace 1995; Van der Lelij et al. 2016).
Following amalgamation, important Meso-Cenozoic constituents were superim-
posed upon of the composite metamorphic basement of the Central Continental
Sub-plate. From the late Triassic-Jurassic, these include the San Lucas, Ibagué and
Segovia blocks. These composite lithotectons represent temporally/geographically
constrained, ensialic extensional basin-continental margin magmatic arc couplets.
Volcano-sedimentary basin development and subsequent emplacement of the calc-­
alkaline San Lucas and Ibagué batholiths appear to have been localized by exten-
sional reactivation of the Palestina suture. A general east-to-west younging trend of
major Jurassic arc-related batholiths, from the Santander Plutonic Group through
the San Lucas batholiths and into the Segovia Batholith, has been interpreted by
various authors (Leal-Mejía 2011; Cochrane 2013; Cochrane et al. 2014b; Spikings
et al. 2015) to reflect regional extension due to slab rollback. Cediel and Cáceres
(2000) and Cediel et al. (2003) denominated the taphrogenic framework associated
with the break-up of Pangaea and the opening of the Proto-Caribbean basin during
the Mesozoic and early Cenozoic, the Bolivar Aulacogen (see below).
The late Mesozoic history of the CTR is dominated by continued extension and
opening of the Valle Alto rift (Cediel et al. 1994; Cediel and Cáceres 2000; Cediel
et al. 2003). This period involves deep crustal and continental margin rifting, the
emplacement of alkalic-tholeiitic dike suites and mafic magmatism (Fabre and
Delaloye 1983; Vásquez et al. 2010), the invasion of the Cretaceous seaway and the
deposition of deep to shallow marine sequences over extensive areas of the CTR,
the Maracaibo Sub-plate and the continental platform of the Guiana Shield (Cediel
et al. 1994; Cediel et al. 2003; Cediel 2011; Cediel 2018; Sarmiento 2018). The lat-
est Jurassic through Cretaceous record has been exhumed and is exposed through-
out Colombia’s Eastern Cordillera and within erosional relicts such as the San Pablo
and Segovia Formations preserved within the physiographic Central Cordillera. The
early to mid-Cretaceous, from ca. 145 to 95 Ma, is characterized by an apparent
hiatus in significant granitoid magmatism throughout continental Colombia, as sug-
gested by the absence of calc-alkaline arc segments or important volumes of granit-
oid rocks of any kind dating from this time period.
The late Mesozoic and Cenozoic to Recent components of the Central Tectonic
Realm are dominated by subduction-related arc granitoids, associated with assem-
bly and accretion of the Western Tectonic Realm (detailed below) and the confor-
mation of the present-day Northern Andes volcanic arc (Stern 2004). Granitoid
assemblages, including the Cretaceous Antioquian Batholith (Feininger and Botero
1982; González 2001; Leal-Mejía 2011) and Paleocene to early Eocene intrusions
observed to the south, reflect the subduction-driven and accretionary regime
dominant during the late Cretaceous. Oligo-Miocene-Pliocene and Pleistocene to
Recent Andean-type magmatism form north–northeast trending belts and clusters of
264 H. Leal-Mejía et al.

intrusions, partially eroded volcanic edifices and active stratovolcanic cones stretching
along the western margin of the Central Tectonic Realm.
Western Tectonic Realm (WTR)  The approach, assembly and accretion of the
allochthonous Western Tectonic Realm (Fig. 5.2) provided the driving mechanism
for granitoid magmatism during the late Meso-Cenozoic Northern Andean orogeny
(Cediel and Cáceres 2000; Leal-Mejía 2011). Within the WTR three composite ter-
rane assemblages are recognized, including the Pacific (PAT) and Caribbean (CAT)
and the Chocó Arc (CHO). The Romeral and Dagua terranes of the PAT assemblage,
and the San Jacinto and Sinú terranes of the CAT assemblage, roughly correspond
to litho-tectonic units recognized by Etayo-Serna et al. (1983). The combined PAT
and CHO assemblages approximate the Provincia Litosférica Oceánica Cretácica
del Occidente de Colombia (PLOCO) of Nivia et al. (1996) and form the geographic
Western Cordillera of Colombia. All of these tectonic assemblages contain frag-
ments of oceanic crust, oceanic plateaus, aseismic ridges and/or ophiolite with asso-
ciated marine sedimentary rocks. All developed within/upon oceanic basement and,
based upon faunal assemblages (e.g. Etayo-Serna and Rodríguez 1985), paleomag-
netic data (e.g. Estrada 1995) and recent paleogeographic reconstructions (e.g.
Cediel et al. 1994; Cediel et al. 2003; Kennan and Pindell 2009; Montes et al. 2012),
and all, with the exception of components of the Romeral assemblage, are alloch-
thonous with respect to continental South America. The composite terrane assem-
blages of the Western Tectonic Realm are characterized as follows:

Pacific Terrane Assemblage (PAT)  The PAT consists of the Romeral assemblage
and Dagua and Gorgona terranes. Romeral may be interpreted as a regional-scale
tectonic melange (Cediel and Cáceres 2000), developed within an early Cretaceous,
rift-related transtensional basin along the Colombian Pacific margin during the
early Cretaceous (Nivia et al. 2006; Kennan and Pindell 2009). The Romeral assem-
blage includes intensely deformed and fragmented blocks (tectonic floats?) of
amphibolite and carbonaceous schist, high-pressure metamorphic rocks (eclogite,
blueschist), layered mafic and ultramafic complexes, marine and peri-cratonic arc-­
related volcanic rocks, ophiolite and meta-sediments, dating from the Paleozoic,
Jurassic and early Cretaceous. The suite was assembled along the Pacific margin in
tectonic contact with the CTR to the east, along the Romeral fault system (Ego et al.
1995; Cediel et al. 2003). The Romeral assemblage underlies much of the Cauca-­
Patía intermontane valley (Fig. 5.2), including the northern inter-Andean depression
to the north and south of the city of Pasto. Cediel et al. (2003) note that the alloch-
thonous vs. in situ nature of the Romeral mélange remains unclear, but current
information suggests the presence of both components of a peri-cratonic nature
deposited in a continental margin basin (Nivia et  al. 2006; Kennan and Pindell
2009) and allochthonous components formed within an intra-oceanic setting.
To the west of the Romeral melange, the Dagua terrane is comprised of an assem-
blage of oceanic mafic and ultramafic rocks (Diabasico Group) which forms the
basement for important thicknesses of flyschoid silici-clastic sedimentary rocks
including chert, siltstone, marlstone and greywacke (Dagua Group).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 265

Lithogeochemical studies (e.g. Kerr et  al. 1997; Sinton et  al. 1998) indicate the
mafic and ultramafic volcanic and intrusive rocks are of oceanic tholeiitic N- and
E-MORB affinity, interpreted to represent accreted fragments of oceanic crust,
ophiolite, aseismic ridges and/or oceanic plateaus belonging to the Farallon Plate
and Cretaceous Caribbean-Colombian Oceanic Plateau (CCOP) or Caribbean Large
Igneous Province (CLIP), as described by Kerr et al. (1997) and Sinton et al. (1998),
respectively. A mantle plume-hotspot-oceanic flood basalt origin for the CCOP/
CLIP assemblage, developed within/upon the Farallon Plate, has been proposed by
these authors. Data provided by Nerlich et al. (2014, and references cited therein)
indicates that the section of the Farallon Plate which forms basement to CCOP/
CLIP plateau rocks varies from ca. 144 Ma in the east, younging westwards to ca.
75 Ma, presently located in the westernmost Caribbean. A summary of radiometric
(mostly Ar/Ar) age dates for accreted CCOP/CLIP rocks in northern South America
and the Caribbean suggests that plateau-related mafic-ultramafic magmatism super-
imposed upon the Farallon Plate may be considered in three stages; a volumetrically
restricted phase initiated at ca. 100  Ma, followed by the widespread eruption of
oceanic plateau rocks dating from ca. 92 to 87 Ma (Kerr et al. 1997; Sinton et al.
1998; Kerr et al. 2003; Hastie and Kerr 2010; Nerlich et al. 2014). Lesser but still
significant, plateau-related basaltic magmatism was subsequently recorded between
ca. 77 and 72 Ma (Kerr et al. 1997; Sinton et al. 1998), although these dates are
considered to represent the wanning stages of CCOP-/CLIP-related magmatism.
Within the Dagua terrane, mid-Cretaceous Ar-Ar ages for oceanic plateau rocks
(Kerr et  al. 1997; Sinton et  al. 1998) are in broad agreement with mid- to late
Cretaceous biostratigraphic ages for contained oceanic sedimentary rocks contained
within the Dagua Group (Etayo-Serna and Rodríguez 1985). Radiometric age dat-
ing of the Bolivar ultramafic complex along the eastern margin of the Dagua terrane
returned U-Pb (zircon) ages ranging from ca. 97 to 95 Ma (Villagómez et al. 2011).
Additional subduction-related granitoids (our Western Group granitoids; see Sect.
5.3.4.1) appear to form part of the Greater Arc of the Caribbean assemblage (e.g.
Pindell and Kennan 2001; Hastie and Kerr 2010; Wright and Wyld 2011; Weber
et al. 2015) and were emplaced into the Farallon Plate and Dagua terrane prior to
accretion along the Colombian Pacific margin. Approach/collision of the Farallon
Plate-CCOP/CLIP assemblage, and accretion of the Dagua terrane, began in the late
Cretaceous (see Sect. 5.4.3.2), along the Cauca fault and suture system.
Further west, the Gorgona Terrane is located mostly offshore, on the southwestern
margin of the Colombian Pacific. Gorgona also represents an accreted oceanic plateau
of mantle plume affinity, containing massive basaltic and spinifex-textured komatiitic
lava flows, pillow lavas and a peridotite-gabbro complex (McGeary and Ben-Avraham
1989). Radiometric ages provided by Sinton et al. (1998) range from ca. 87 to 83 Ma;
however, paleomagnetic and lithogeochemical data, and paleogeographic reconstruc-
tions presented by Estrada (1995), Kerr and Tarney (2005) and Kennan and Pindell
(2009), suggest Gorgona has no clear correlation with the CCOP/CLIP. Gorgona is
limited to the east by the Buenaventura fault and to the west by the modern-day
Colombian trench. Accretion of Gorgona along the western margin of the Dagua
terrane took place during the Eocene (Cediel et al. 2003; Kerr and Tarney 2005).
266 H. Leal-Mejía et al.

Caribbean Terrane Assemblage (CAT)  Two principle terranes are contained within
this assemblage, the San Jacinto and Sinú (Fig. 5.2). San Jacinto includes a MORB-­
type tholeiitic basement considered a fragment of the CCOP/CLIP assemblage,
containing upper Cretaceous deep marine carbonaceous cherts, mudstones and
marlstones locally intercalated with coarser-grained siliciclastic and felsic pyroclas-
tic material and intruded by minor holocrystalline diorite to quartz diorite plutons of
poorly constrained age (e.g. El Alacrán, San Juan de Asís). San Jacinto was accreted
to the northern CTR along the San Jacinto fault during the Eocene (Cediel and
Cáceres 2000). The Sinú terrane is comprised of similar basement to San Jacinto,
overlain by turbidite sequences of Oligocene age. It was juxtaposed along the San
Jacinto margin in the Miocene.

The Chocó (Eastern Panamá) Arc (CHO)  The Chocó Arc assemblage in Colombia
(Duque-Caro 1990; Schmidt-Thomé et al. 1992; Cediel et al. 2010), together with
Campanian to Eocene mafic oceanic and intermediate arc-related plutonic rocks of
the Darién (San Blas) Range and Azuero Peninsula in Panamá (Wegner et al. 2011;
Montes et al. 2012), represents the eastern segments of the Panamá double arc. In
Colombia, the basement of the composite Chocó Arc (Fig. 5.2) is comprised of two
distinct litho-tectonic assemblages: the Cañas Gordas terrane and the El Paso
Terrane which includes the Baudó Range (Cediel et al. 2010; Redwood 2018).
Cañas Gordas consists of mixed volcanic rocks of the Barroso Fm. overlain by
fine-grained sedimentary rocks of the Penderisco Fm. The Barroso Fm. is domi-
nated by tholeiitic to calc-alkaline, massive, porphyritic and amygdaloidal basalt,
with andesitic flows, tuffs and agglomerates (Rodriguez and Arango 2013).
Sedimentary interbeds within the Barroso Fm. mapped near the town of Buriticá
contain Barremian through middle Albian fossil assemblages (González 2001 and
references cited therein). Weber et  al. (2015) consider gabbros belonging to the
Barroso Fm. to belong to the CCOP/CLIP plateau assemblage, although biostati-
graphic data suggests that Barroso is pre-CCOP and may better represent accreted
slivers of the older, Farallon Plate oceanic basement upon which the CCOP rests
(Nerlich et  al. 2014 and references cited therein). The Penderisco Fm. includes
thinly bedded, mudstone, siltstone, marlstone, greywacke and chert. Two members,
including Urrao and Nutibara, contain marine fossil assemblages dating from the
Aptian-Albian to the upper Cretaceous (González 2001 and references cited
therein), again demonstrating the diachronous nature of the Farallon-CCOP/CLIP
assemblage. The eastern margin of the Cañas Gordas terrane was intruded by the
Buriticá tonalite and the Santa Fé Batholith at ca. 100 Ma and 90 Ma, respectively
(Weber et al. 2015). The terrane assemblage was accreted to the continental margin
during the late Cretaceous to Paleocene (see Sect. 5.4.3.2).
The El Paso-Baudó components of the Chocó Arc are comprised of late
Cretaceous to Paleogene sections of tholeiitic basalt of N- and E-MORB affinity
(Goossens et al. 1977; Kerr et al. 1997), overlain by minor pyroclastic rocks, chert
and turbidite. El Paso-Baudó represents a late Cretaceous silver of the CCOP/CLIP
assemblage, considered to have formed along the trailing edge of the Caribbean
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 267

Plate. The Mandé-Acandí arc with associated plutonic, hypabyssal porphyritic

stocks and pyroclastic volcanic sequences (La Equis-Santa Cecilia Fms; Sillitoe
et al. 1982; Schmidt-Thomé et al. 1992; Leal-Mejía 2011; Montes et al. 2012) was
emplaced within El Paso oceanic basement between ca. 60 and 42 Ma (Leal-Mejía
2011; Montes et al. 2012, 2015). Ural-Alaskan-type zoned ultramafic complexes at
Alto Condoto and Mumbú were intruded into El Paso Terrane basement in the early
Miocene (Tistl 1994; Tistl et al. 1994; Cediel et al. 2010). Development of the San
Juan and Atrato basins began in the Paleogene. Final collision of the El Paso-Baudó-­
Mande assemblage along the western Cañas Gordas margin and uplift of the Baudo
Range is recorded in the Miocene (Cediel et al. 2010; Montes et al. 2012, 2015).
Faults related to the assembly and accretion of the Chocó Arc, including the
Garrapatas-Dabeiba and San Juan-Sebastian systems (Cediel and Cáceres 2000;
Cediel et  al. 2003, 2010) and Uramita system (Duque-Caro 1991; Montes et  al.
2012), reactivate, deform and/or truncate earlier structures associated with the
Romeral, Cauca and San Jacinto fault systems.

5.2.2  Structural Framework

The original emplacement of many of the granitoid arc segments in Colombia dem-
onstrates strong structural control, especially with respect to precursor basement
architecture during Proterozoic through Paleozoic and early Mesozoic times. Past
authors have noted the coaxial nature of both pre-Triassic and late Triassic – early
Jurassic structure and the distribution of the major late Triassic – through Pliocene
arc segments (Aspden et al. 1987; Cediel et al. 2003; Leal-Mejía 2011). Cediel et al.
(1994), and Cediel and Cáceres (2000) demonstrate the reactivation of pre-Triassic
structures, including the Bucaramanga–Santa Marta–Suaza (Garzón) and Palestina
fault and suture systems, in the development of late Triassic volcano-sedimentary
grabens (e.g. the Perijá, Maracaibo and Payande rifts) and the subsequent emplace-
ment of major holocrystalline granitoid batholiths.
Many of the earlier phases of late Triassic to Pliocene granitoid magmatism,
especially those from the Jurassic and Cretaceous, have suffered some degree of
post-emplacement structural modification during the late Mesozoic-Cenozoic
Northern Andean Orogeny (Cediel et al. 2003). A simplified structural framework
for the modern-day Colombian Andes is shown in Fig. 5.2. The present-day struc-
tural architecture of the region is dominated by a complex array of large-scale
strike-slip fault systems with complex transpressive movement vectors, for which, in
some cases, origins and reactivations spanning the Proterozoic to the present can be
demonstrated (Cediel and Cáceres 2000; Cáceres et  al. 2003; Cediel et  al. 2003;
Cediel 2011). The structures of greatest importance with respect to the post-­
emplacement history of late Triassic-Pliocene granitoids are the Bucaramanga–Santa
Marta–Garzón, Palestina and Romeral fault and suture systems. These fault/suture
systems presently record tens of kilometres or more of lateral offset brought on by
the sequential accretion of oceanic terranes along the Pacific margin of NW South
268 H. Leal-Mejía et al.

America and initiation of tectonic float in the Maracaibo Sub-plate, during stages of
the Northern Andean Orogeny (Cediel et  al. 2003; Kennan and Pindell 2009).
Consequent modification/deformation of late Triassic-Pliocene granitoids ranges
from localized to regional shearing and/or thrusting  – focussed along intrusive
margins  – to tectonic segmentation and km-scale sinistral or dextral translation.
Beyond these structural modifications, however, and unlike many of the Colombian
granitoids of Paleozoic age, the latest Triassic through Plio-Pleistocene granitoids,
including their coeval volcano-sedimentary sequences (where preserved), are not
regionally metamorphosed nor have they been subjected to regional-scale ductile
or penetrative deformation, prograde metamorphism or recrystallization during
post-­emplacement tectonic events.

5.3  D
 istribution, Age, Nature and Temporal-Spatial
Evolution of Colombian Phanerozoic Granitoid
Magmatism

5.3.1  Introduction

Figure 5.3 depicts the distribution of all major occurrences of Phanerozoic granit-
oid plutonic, hypabyssal and volcanic rocks throughout the Colombian Andes,
based upon available regional cartographic data as presented by Cediel and Cáceres
(2000), Gómez et al. (2007) and Gómez et al. (2015a). The distribution, nature and
temporal-spatial evolution of Colombian granitoids will herein be discussed under
four age-based headings: Paleozoic-mid-Triassic, latest Triassic-Jurassic,
Cretaceous-Eocene and latest Oligocene-Mio-Pliocene. Each of these headings
contains detailed information with respect to the age, classification, lithogeochem-
istry, isotope geochemistry and tectono-magmatic evolution of the respective gran-
itoids. In the interest of space and focus, we have opted to forgo detailed
petrographic-minerographic descriptions of the Colombian granitoid suites, and
the reader will be referred to specific bibliographic references pertaining the vari-
ous ages of Colombian granitoids in order to assess this valuable information. The
final section of this chapter presents time-space analyses and tectono-magmatic
reconstructions which place the granitoids into the integrated, evolving geotec-
tonic framework of the Colombian Andes.

5.3.1.1  U-Pb (Zircon) Age Database: Phanerozoic Magmatic Episodes

Appendix A1 reveals bibliographic reference and geographic location data of granitoids

dated by the U-Pb (zircon) method and compiled for use in this study. The quality
and provenance of all information were reviewed, and importantly, one of the
prerequisites for inclusion herein was that accurate sample location data (normally
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 269

Fig. 5.3  Distribution of the principal Phanerozoic granitoid plutonic, hypabyssal and volcanic
rocks in the Colombian Andes in relation to physiographic relief as derived from SRTM 90 m digi-
tal elevation model (DEM). (Granitoid shapes adapted from Cediel and Cáceres 2000; Gómez
et al. 2007; Gómez et al. 2015a). Age ranges based upon U-Pb (zircon) data compiled herein
270 H. Leal-Mejía et al.

Fig. 5.4  Six principle periods of Phanerozoic granitoid magmatism in the Colombian Andes as
derived from U-Pb (zircon) age dates. (Data is compiled primarily from Leal-Mejía (2011; and
works cited therein), Gómez et al. 2015b and additional references cited within the present text)

UTM or latitude-longitude coordinates tied to a specified registered datum) were

available. Approximately 287 individual data points are represented.
Figure 5.4 plots the distribution of U-Pb (zircon) crystallization age determinations
for Phanerozoic granitoids from throughout the Colombian Andes. The histogram
highlights six broad episodes of Phanerozoic granitoid magmatism in the Colombian
Andes, including early Paleozoic (ca. 485–439 Ma), Carboniferous (ca. 333–310 Ma),
Permo-Triassic (ca. 288–223  Ma), latest Triassic-Jurassic (ca. 210–146  Ma),
late Cretaceous to Eocene (ca. 100–42 Ma) and latest Oligocene to Mio-­Pliocene
(ca. 23–1.2  Ma). A continuum of this last episode, from the Pliocene to the
Pleistocene and Recent, encompassing the modern-day Colombian (Northern
Andean) volcanic arc, is evident, but not discussed in detail herein (see Marín-­Cerón
et al. 2018).
Within most of the episodes depicted in Fig. 5.4, the resolution of the available
U-Pb (zircon) data permits definition of additional sub-episodes of magmatic activity.
This resolution, when combined with available lithogeochemical and isotopic data,
permits a detailed analysis of the evolution and migration of Phanerozoic granitoid
arc segments over time. Individual sub-episodes or pulses can be interpreted to
coincide with, or denote, magmatic responses to tectonic developments throughout
the region. These sub-episodes and their tectonic significance will be discussed in
detail in the Sect. 5.3 of this chapter.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 271

5.3.1.2  Lithogeochemical Database and Data Filtering

Appendix A2 also reveals bibliographic references and sample location data for
some 561 Phanerozoic granitoids, for which whole-rock lithogeochemical analyses
are presently available. As with the U-Pb (zircon) geochronology samples, accurate
sample location data was prerequisite for inclusion within the lithogeochemical
sample database utilized herein. Most lithogeochemical samples were collected
from surface outcrops, although many samples from the Leal-Mejía (2011) study
were taken from underground exposures and diamond drill core. Approximately 85
percent of the entire lithogeochemical sample set database includes a full suite of
major, minor, trace and rare-earth elements.
The Colombian Cordilleras evolved at geographically low latitudes and have in
many areas been subjected to high rainfall sub-tropical to tropical conditions since
at least the late Cretaceous. In this context, samples collected from surface expo-
sures have potentially been subjected to surface weathering and oxidation.
Additionally, in the case of samples collected from underground exposures or dia-
mond drill core, the potential exists for syn- or post-crystallization deuteric and/or
hydrothermal alteration. In either case, the effects of the above related processes,
being visual or cryptic, can significantly alter sample lithogeochemistry.
Based upon the above, the entire 545 sample suites utilized herein were sub-
jected to element mobility analysis, permitting the identification of granitoids which
have undergone subsolidus alteration processes such as those which can markedly
affect critical alkali and aluminium indices and potentially generate misleading con-
clusions regarding the petrogenetic trends of the sample set (Davies and Whitehead
2010). The net result of subsolidus alteration is generally increased alkalinity, alu-
minity and/or silica content. In our particular case, the following general consider-
ations are applicable: (1) high loss on ignition values (LOI > 2.0) may be related to
high volatile content (reflecting possible clay, carbonate or sulphide alteration), (2)
high SiO2 content may be related to hydrothermal silicification, and (3) specific
alteration and element ratio diagrams such as the alkali-alumina molar ratio plot
(Davies and Whitehead 2006), the K-Ca-Na alteration evaluation plot (Warren
et al. 2007) and Pearce Element Ratio (PER) and General Element Ratio (GER)
diagrams (Stanley and Madeisky 1994, 1996) aid in the detection of altered samples.
The result of data filtering was the identification of 212 samples which clearly
exhibit the effects of hydrothermal alteration and/or weathering. Although the
remaining 349 sample data set may be considered limited with respect to the exten-
sive volume of magma it represents, the data are of high quality and permit the
lithogeochemical and petrogenetic characterization of the great majority of the
granitoids discussed herein.
For presentation purposes, we illustrate the major, trace and REE lithogeo-
chemical data for samples from each major magmatic episode, using the following
­diagrams: the AFM diagram (Irvine and Baragar 1971), the K2O vs. silica plot
(Peccerillo and Taylor 1976), the aluminium saturation index diagram (Barton and
Young 2002), the modified alkali-lime index (MALI) (Na2O + K2O-CaO) vs. silica, the
FeOtot/(FeOtot + MgO) vs. silica diagrams (Frost et al. 2001; Frost and Frost 2008),
272 H. Leal-Mejía et al.

the R1-R2 diagram (De la Roche et al. 1980), the C1 chondrite normalized REE
plot (McDonough and Sun 1995), the primordial mantle normalized trace element
spider diagram (Wood et al. 1979), and the granitoid tectonic discrimination dia-
gram (Ta vs. Yb) (Pearce et al. 1984).

5.3.2  P
 aleozoic to Mid-Triassic Granitoid Magmatism:
Distribution, Age and Nature

Granitoid rocks from this extended time period comprise a texturally, compositionally
and petrogenetically diverse suite, which has been subjected to a prolonged and
intense tectonic history, both during and post-dating their emplacement and cooling.
Given this observation, the geological context of what are commonly referred to as
gneissic granitoids, meta-granitoids or foliated granitoids within the Colombian geo-
logical literature is complex, and the nature, distribution and genesis of these rocks in
the Colombian Andes are relatively poorly understood. These observations may be
attributed to various causes. Firstly, although (presumed) Paleozoic through mid-Tri-
assic granitoids are of relatively widespread distribution, especially within the
Santander Massif and the physiographic Central Cordillera (Aspden et al. 1987; Ward
et al. 1973; Cediel and Caceres 2000; Vinasco et al. 2006; Gómez et al. 2015a), intru-
sions are limited to relatively small stocks and elongate or irregular-shaped bodies,
with complex outcrop patterns, commonly intercalated with other granitoids of older
or younger age. Exposure is often inhibited by thick vegetation cover, deep surficial
oxidation and soil development or hydrothermal alteration, and the cartographic limits
of many of the intrusions have yet to be clearly established. Historically, this situation
was exasperated by the fact that few reliable radiometric age dates were available for
the gneissic granitoids, and until the more recent application of U-Pb (zircon) dating
techniques, many occurrences of these rocks were presumed to be of Precambrian,
early Paleozoic or Mesozoic age, based primarily upon field relationships and consid-
erations regarding texture and/or metamorphic grade.
Based upon information provided by more recent geological, age-dating, lithogeo-
chemical and isotopic studies (e.g. Restrepo-Pace 1995; Vinasco et al. 2006; Ibañez-
mejía et  al. 2008; Cardona et  al. 2010b; Horton et  al. 2010; Montes et  al. 2010;
Leal-Mejía 2011; Leal-Mejía et al. 2011; Restrepo et al. 2011; Villagómez et al. 2011;
Mantilla et al. 2012; Van der Lelij 2013; Cochrane 2013; Cochrane et al. 2014a; Van
der Lelij et al. 2016), three broad populations of granitoids will be highlighted within
this section: (1) early Paleozoic, (2) Carboniferous and (3) Permo-Triassic.

5.3.2.1  Distribution of Early Paleozoic to Mid-LateTriassic Granitoids

Figure 5.5 highlights the distribution of early Paleozoic through mid-Triassic gran-
itoids throughout the Colombian Andes, based upon available regional geologic
mapping and compilation. For reference, the principle physiographic provinces of
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 273

the region are also shown. Early Paleozoic rocks described as granitoids, metamor-
phosed and foliated granitoids and granitic gneisses (orthogneisses) are mostly con-
tained within the Santander, Floresta and Quetame massifs of the eastern Colombian
Andes (Horton et al. 2010; Leal-Mejía 2011; Leal-Mejía et al. 2011; Mantilla et al.
2012; Van der Lelij 2013; Van der Lelij et al. 2016). In addition, punctual occur-
rences of early Paleozoic granitoids have been reported on the northern and western
flanks of the Central Cordillera, along the Otú Fault near El Bagre (Leal-Mejía
2011, Leal-Mejía et al. 2011) and along the Cauca River valley (La Miel Orthogneiss;
Vinasco et al. 2006; Villagómez et al. 2011; Martens et al. 2014) (Fig. 5.5).
Carboniferous granitoids have been reported at only one locality; the El Carmen-El
Cordero Stock, near El Bagre (Leal-Mejía 2011) (Fig. 5.5). These intrusive rocks are
of two main types: (1) early, fine-grained melanocratic, phaneritic holocrystalline to
weakly porphyritic gabbro-diorites and (2) volumetrically dominant coarse-grained,
phaneritic and holocrystalline leucocratic tonalities containing quartz, plagioclase and
minor K-feldspar (microcline), with biotite, abundant zircon and ilmenite as acces-
sory minerals. No additional granitoids of similar age have been reported in the
Colombian Andes. The El Carmen-El Cordero pluton(s) were historically undifferen-
tiated from Jurassic intrusives of the Segovia Batholith (González 2001) and refer-
ences contained therein and remain so in all but the most recent geological compilation
(e.g. Cediel and Cáceres 2000; Gómez et al. 2015a). Notwithstanding, the geological
limits of the El Carmen-El Cordero plutons have yet to be established, and the con-
tacts shown in Figs. 5.3 and 5.5 represent interpretations based upon very preliminary
field reconnaissance and the examination of DEM images.
Permian to Triassic granitoids are widely distributed in the Colombian Andes
from the border with Ecuador in the south to the Sierra Nevada de Santa Marta and
the Guajira peninsula on the Colombian Caribbean coast. The majority of these
bodies however are exposed within the northern Central Cordillera (Fig. 5.5). The
Permo-Triassic granitoid suite is exposed as relatively small bodies outcropping on
the eastern and western flanks of the Central Cordillera. Many of these bodies have
been documented under local names, but, due to their small size, do not resolve well
within regional-scale geologic maps. From S to N, confirmed granitoid gneisses of
Permo-Triassic age include the La Plata orthogranite, La Linea intrusive gneiss,
Manizales gneiss, Chinchina gneiss, the southern Sonsón Batholith (i.e. the Nariño
Batholith), the Quebrada Pácora stock, the Pantanillo intrusive gneiss, the
Cambumbía stock, the Rio Verde intrusive gneiss, the Alto de Minas intrusive
gneiss, the Abejorral intrusive gneiss, the El Buey stock, the La Honda stock, the
Amagá stock, the Pueblito diorite, the Palmitas granitic gneiss, the Horizontes
tonalite gneiss, the Montegrande granitic gneiss, the Naranjales granitic gneiss, the
Samaná granitic gneiss, the Santa Isabel gneiss, the Puquí meta-tonalite, the Nechí
Gneiss, the Los Muchachitos gneiss and the Uray Gneiss (Vinasco et  al. 2006;
Ibañez-mejía et  al. 2008; Cardona et  al. 2010b; Montes et  al. 2010; Leal-Mejía
2011; Leal-Mejía et al. 2011; Restrepo et al. 2011; Villagómez et al. 2011; Cochrane
et al. 2014a). Some of these bodies are located in Fig. 5.5. The geological limits and
age of numerous additional, small, unnamed, undated bodies of granitic orthogneiss
have yet to be clearly defined.
274 H. Leal-Mejía et al.

Fig. 5.5  Distribution of early Paleozoic through mid-late Triassic granitoids in the Colombian
Andes. Principle modern-day physiographic provinces of the region are shown for reference.
(Granitoid shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al. 2015a)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 275

The gneissic texture observed in many early Paleozoic to mid-Triassic granitoids,

especially those from the Permo-Triassic, has led to confusion with respect to their
overall abundance and distribution. In various instances, such rocks are mapped as
representatives of Precambrian basement, based upon the erroneous assumption that
the “gneissic textured” granitoids are necessarily older than their surrounding host
rocks. This assumption has been recently disproven using high-precision U-Pb
(zircon) dating techniques, which illustrate that in various instances, granitoids pre-
viously recorded as “Precambrian” in age in fact belong to the Permo-Triassic suite,
hosted within a Paleozoic or Permo-Triassic-aged basement (Vinasco et  al. 2006;
Ibañez-Mejía et al. 2008; Leal-Mejía 2011; Restrepo et al. 2011; Villagómez et al.
2011, Cochrane et al. 2014a). It is suspected that the abundance of Permo-Triassic
granitoids will increase in future studies, at the expense of the “Precambrian” suite.
It would appear that modern U-Pb dating techniques will form the best means for the
differentiation of the Precambrian and early Paleozoic vs. Permo-Triassic suites.

5.3.2.2  A
 ge Constraints on Paleozoic to Mid-Triassic Granitoid
Magmatism

The U-Pb (zircon) age distribution of Phanerozoic granitoids of pre-Jurassic age in

the Colombian Andes is presented in histogram format in Fig. 5.6. In terms of age,
pre-Jurassic magmatism in the region has been previously well recognized, in two

Fig. 5.6  Three principle periods of pre-Jurassic granitoid magmatism in the Colombian Andes, as
derived from the distribution of U-Pb (zircon) age dates. Although each period, including respective
sub-periods, is well-represented by multiple age dates, the overall distribution of these granitoids is
sparse and erratic when compared to granitoids of post latest Triassic age
276 H. Leal-Mejía et al.

specific age ranges, including the early Paleozoic (i.e. early to middle Ordovician)
and the Permian to mid-late Triassic (e.g. Goldsmith et al. 1971; Boinet et al. 1985;
Restrepo-Pace 1995; Ordoñez 2001; Cediel et  al. 2003; Vinasco et  al. 2006;
Ordoñez-Carmona et  al. 2006; Ibañez-Mejía et  al. 2008; Cardona et  al. 2010b;
Horton et al. 2010; Montes et al. 2010; Weber et al. 2010). Leal-Mejía (2011) docu-
mented previously unrecognized granitoid magmatism of Carboniferous age, in the
El Carmen-El Cordero Stock near El Bagre. Additional pre-Jurassic U-Pb zircon
ages have more recently been published, for early Paleozoic foliated granitoids in
the Angosturas district and other localities in the Santander Massif (Mantilla et al.
2012; Van der Lelij et al. 2016), for the La Miel orthogneiss to the west (Villagómez
et al. 2011; Martens et al. 2014) and for Permo-Triassic granitoids and amphibolites
of the Central Cordillera (Restrepo et al. 2011; Villagómez et al. 2011; Cochrane
et al. 2014a).
The resulting composite U-Pb (zircon) age date database permits definition of
three distinct episodes of granitoid magmatism within the Colombian Andes: (1) ca.
485–439  Ma (early Paleozoic; early to mid-Ordovician), (2) ca. 333–310  Ma
(Carboniferous) and (3) ca. 289–223 Ma (Permian to mid-late Triassic). Sub-­episodes
of granitoid magmatism are implicit within the age distribution recorded by each of
the major episodes and have been interpreted by the various authors to represent
granitoid magmatism within the evolving tectonic framework of the region during
the early Phanerozoic, as will be reviewed in Sect. 5.4.

Early Paleozoic Granitoids

Crystallization U-Pb zircon ages for early Paleozoic granitoids within Colombia’s
eastern cordilleran system span the range between ca. 485 and 439 Ma (Restrepo-­
Pace 1995; Horton et al. 2010; Leal-Mejía 2011; Mantilla et al. 2012; Martens et al.
2014; Van der Lelij et  al. 2016). Early Paleozoic magmatism is recorded in the
Santander, Floresta and Quetame massifs, in unfoliated and foliated arc-related
granitoids spanning a range between ca. 485 and 482  Ma (Horton et  al. 2010;
Mantilla et al. 2012; Van der Lelij et al. 2016). Syn-kinematic and peak metamor-
phic granitoid magmatism, coeval with medium-pressure Barrovian-type metamor-
phism (Van der Lelij et al. 2016), is recorded by foliated granitoids spanning a range
between 479.8 and 472.5  Ma, in the Santander and Floresta massifs (Restrepo-
Pace 1995; Horton et  al. 2010; Leal-Mejía 2011; Mantilla et  al. 2012; Van der
Lelij et  al. 2016). Post-metamorphic magmatism in the Santander Massif is
recorded by granitoids emplaced during post-orogenic extension and/or resumed
arc-related magmatism, returning U-Pb (zircon) ages between ca. 462.5 and
439.2 Ma (Leal-Mejía 2011; Van der Lelij et al. 2016).
To the west, additional localized occurrences of early Paleozoic granitoids/
granitic gneisses are exposed at two localities within the Central Cordillera. These
include (1) the ca. 479–443 Ma (Villagómez et al. 2011; Martens et al. 2014) La
Miel leuco-orthogneiss, composed primarily of k-feldspar, plagioclase, quartz,
muscovite and minor biotite, with a clear relict igneous texture, and (2) a
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 277

473.4  ±  6.9  Ma (Leal-Mejía 2011) unnamed granodiorite intrusion outcropping

along the Otú Fault near El Bagre (Fig. 5.5). The zircon separate for the sample was
extracted from saprolitized bedrock, and the cartographic limits and geological con-
text of this granodiorite body have yet to be completely defined.
A notable feature of the early Paleozoic zircon populations in Colombia is the com-
plex internal structure and growth zonation of individual zircon crystals (e.g. Mantilla
et al. 2012; Van de Lelij et al. 2016). In this context, beyond the interpreted crystalliza-
tion ages presented above, many of the early Paleozoic zircon populations present
multiple inheritance ages, dating from the early-mid-Proterozoic (Leal-Mejía 2011;
Mantilla et al. 2012) and ranging into the late Proterozoic, and suggest a prolonged
history of magmatism, metamorphism and crustal recycling during Proterozoic and
early Paleozoic times (e.g. Van der Lelij et al. 2016; see below).

Carboniferous Granitoids

The occurrence of the Carboniferous El Carmen-El Cordero granitoids of the north-

ern Central Cordillera was initially recorded by Leal-Mejía (2011). No additional
occurrences have appeared in recent literature, so, based upon available geochrono-
logical data, this magmatism is presently restricted to the El Carmen-El Cordero
occurrences. The El Carmen-El Cordero suite consists of early holocrystalline to
weakly porphyritic melanodiorite, which returned a U-Pb (zircon) age of
333.1 ± 4.7 Ma, whilst four samples of Na-rich, quartz, plagioclase ± biotite and
K-feldspar leucotonalite, comprising the main El Carmen Stock and associated
dikes, returned U-Pb ages ranging from ca. 326 ± 5.6 Ma to 310.6 ± 5.6 Ma (Leal-­
Mejía 2011). The precise paragenetic relationship between the melanodiorite and
the various phases of leucotonalite has yet to be deciphered as the contact between
these units is not exposed. The ca. 333–310  Ma  U-Pb (zircon) ages recorded by
Leal-Mejía (2011) were considered to represent magmatic crystallization ages.
Interestingly, unlike the early Paleozoic and Permo-Triassic (see below) granitoids
from throughout the Colombian Andes, the Carboniferous granitoids demonstrate
no indication of inheritance ages within their U-Pb (zircon) age date profiles.

Permo-Triassic Granitoids

Based upon published U-Pb (zircon) age dates, Permo-Triassic granitoids in the
Colombian Andes, including granitoid gneisses and amphibolites, span the range
from ca. 290 to 222 Ma (Fig. 5.6). Various authors (e.g. Vinasco 2004; Leal-Mejía
2011; Cochrane 2013) record complex zoning and inheritance patterns for zircons
returning Permian through early-mid-Triassic ages. Recognition of the importance
of the Permo-Triassic suite was initially revealed in the works of Vinasco (2004)
and Vinasco et al. (2006), based upon dating of the La Honda, El Buey, Abejorral
and other meta-granitoid intrusions along the Central Cordillera. Subsequent publi-
cations expanded the database for Permo-Triassic granitoids in the Central Cordillera
278 H. Leal-Mejía et al.

to include the Samaná granitic orthogneiss (244.9 ± 4.7 Ma, Ibañez-Mejía et al. 2008),

the Santa Isabel gneiss (267.8 ± 3.6 Ma, Restrepo et al. 2011), the Nechí Gneiss
(ca. 282–277 Ma, Leal-Mejía 2011; Restrepo et al. 2011), the Las Palmas migmatite
(222.0 ± 5.0 Ma, Restrepo et al. 2011), and other localized granitoid, granitic gneiss
and amphibolites bodies which have returned ages ranging from ca. 278 to 236 Ma
(Leal-Mejía 2011; Villagómez et al. 2011; Cochrane et al. 2014a). Cardona et al.
(2010b) reported granitoids ranging from ca. 264 to 288 Ma in the Sierra Nevada de
Santa Marta, whilst Montes et al. (2010) reported ca. 240 Ma ages from granitoid
samples collected from subsurface drill core from the Lower Magdalena Basin.
Finally, based upon new U-Pb (zircon) age dates and petrochemical and petro-
graphic data, Leal-Mejía (2011) recognized that the Sonsón Batholith (González
2001) consists of at least two composite plutonic bodies, the southern segment of
which returns Permo-Triassic U-Pb (zircon) age dates. Based upon early K-Ar dat-
ing, González (2001) originally assigned the Sonsón Batholith to the Jurassic.
Radiometric age dating by Leal-Mejía (2011), however, returned U-Pb (zircon) ages
of 245.4 ± 4.8 Ma and 237.2 ± 4.1 Ma, for samples collected to the south and to the
west of the town of Nariño. In this context Leal-Mejía (2011) referred to the granit-
oids returning early Triassic ages as the Nariño Batholith, however the geologic
limits of the Triassic pluton(s) have yet to be formally mapped. The ca. 237  Ma
pluton outcropping around the town of Nariño is homogeneous and holocrystalline
to weakly foliated. Regional transects across the eastern margin of the pluton to the
south of the town of Nariño reveal it is in contact with hornfelsed paragneiss con-
taining early Permian-aged zircons (Leal-Mejía 2011), similar to rocks intercalated
in the early Permian Río Verde gneiss complex to the NE of Sonsón (Vinasco et al.
2006). The SW flank of the Nariño Batholith, near San Félix, is unconformably
overlain by Aptian-Albian siliciclastic rocks of the Abejorral Formation.
In general, the age of the Nariño Batholith as described herein coincides well
with the Permo-Triassic suite documented elsewhere in the Central Cordillera by
Vinasco et  al. (2006). Based upon presently available mapping, field reconnais-
sance, regional geological trends and U-Pb age (in addition to available lithogeo-
chemical data), we suggest that the paragneiss along the eastern margin of the
Nariño Batholith represents the southern continuation of the Río Verde granitic
gneiss, whilst the western sector of the Nariño Batholith appears to represent the
southern continuation of Permo-Triassic orthogneiss, granitic gneiss and “post-­
tectonic granite”, along the western margin of the Central Cordillera, as illustrated
by Vinasco et al. (2006).

5.3.2.3  L
 ithogeochemical and Isotopic Characteristics of Early Paleozoic
to Mid-Triassic Granitoids

Lithogeochemistry

In terms of whole-rock lithogeochemistry, the Paleozoic to mid-Triassic granit-

oids exhibit significant differences in composition between the early Paleozoic,
Carboniferous and Permo-Triassic age groupings highlighted above (Fig.  5.6).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 279

Available data for each of the three age groups will be summarized separately
briefly herein.
Major, trace and rare-earth element lithogeochemical data for early Paleozoic
granitoids, as drawn from the data sets of Leal-Mejía (2011) and Van de Lelij et al.
(2016), are presented in Figs. 5.7 and 5.8. Whole-rock analyses reveal variable SiO2
contents ranging from 57.6 to 74.7 wt%, with compositions ranging from gabbro-­
diorite to tonalite through granite. The sample set defines a calc-alkaline trend, plot-
ting in the medium-K to high-K calc-alkaline fields. Most of the early Paleozoic
granitoids are weakly peraluminous, with the most altered samples exhibiting
extremely high A/NKC values (>2.0), likely due to post-crystallization hydrother-
mal alteration. Notwithstanding, two samples (10VDL23 and 10VDL47, Van der
Lelij et al. 2016) plot in the metaluminous field. With respect to the classification
scheme of Frost et al. (2001), most of early Paleozoic samples plot in the calcic and
calc-alkalic fields and are magnesian (oxidized) in composition. Trace element and
rare-earth element spider diagrams indicate fractionated arc-related magmatism
(volcanic-arc granites (VAG)) with variable negative Eu anomalies.
With respect to the Carboniferous granitoids, petrographic analysis of both the
melanodiorite and leucotonalite suites (Leal-Mejía 2011) suggests minor effects
brought on by hydrothermal alteration and/or very low-grade regional metamor-
phism. Hornblende within the melanodiorite has been replaced by pumpellyite,
prehnite, chlorite and epidote, whilst the cores of plagioclase have been altered to
sericite. Accessory biotite in the leucotonalite has been partially altered to an
assemblage containing chlorite, epidote, magnetite and titanite, whilst the cores of
plagioclase crystals are strongly sericitized. Alteration plots for the El Carmen
suite suggest some degree of major element mobility associated with these effects.
The ca. 333–310 Ma granitoid suite plots in two separate clusters on the lithogeo-
chemistry plots (Fig. 5.7). SiO2 contents are lower in melano-gabbro/diorites (48.4–
48.8  wt%) with respect to leucotonalites (68.8–72.5  wt%). Both groups exhibit
notably low K2O contents (0.04–1.22 wt%) and plot in the compositional ranges of
gabbro-norite and tonalite-granodiorite, respectively. The Na (vs. K)-rich, trondhje-
mitic nature of the leucotonalites becomes particularly evident when samples are plot-
ted on the feldspar triangle of O’Connor (1965; see Leal-Mejía 2011). The
melano-gabbro/melanodiorite members of the suite plot clearly tholeiitic, whilst the
leucotonalite series presents more evolved calc-alkaline compositions on the AFM
diagram. A composite calc-alkaline trend, however, can only be inferred by the data,
as there is a clear compositional gap in the differentiation series (Fig. 5.7). This may
be a reflection of the limited number of analyses available for the suite (n = 6) or alter-
natively may be a result of the apparent bimodal nature of the suite. Melanodiorite
samples plot in the metaluminous field, whereas leucotonalites are weakly peralumi-
nous perhaps due to alteration effects. The entire suite plots in the magnesian and
calcic fields of Frost et al. (2001). Trace element diagrams depict large-ion lithophile
element enrichment (e.g. Ba, K) and depletion of high-field strength elements
(e.g. Nb, Ta, Ti). Chondrite-normalized REE plots reveal flat patterns around 10x
chondrite concentrations for the melanodiorites (∑REE = 24.7–34.7 ppm) vs. some-
what more enriched and fractionated patterns for the leucotonalite samples
(∑REE = 49–94.1 ppm). Neither rock type produces significant Eu anomalies.
280 H. Leal-Mejía et al.

Fig. 5.7  Major element lithogeochemical plots for pre-Jurassic (i.e. early Paleozoic, Carboniferous
and Permo-Triassic) granitoids in the Colombian Andes. (a) AFM plot, curve after Irvine and
Baragar (1971); (b) K2O vs. SiO2 plot, boundary fields in grey as summarized by Rickwood
(1989); (c) alumina saturation plot after Barton and Young (2002); (d and e) MALI and Fe-index
vs. SiO2 plots, respectively, after Frost et al. (2001); (f) R1-R2 classification plot after De La Roche
et al. (1980). Th tholeiite, C-A calc-alkaline, Sh shoshonite, Gb No gabbro-norite, Gb Di gabbro-­
diorite Di Diorite Mz Di monzodiorite Mz Monzonite To Tonalite Gd Granodiorite Gr Granite Alk
Gr Alkali Granite
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 281

Fig. 5.8  Trace element and REE lithogeochemical plots for pre-Jurassic (i.e. early Paleozoic,
Carboniferous and Permo-Triassic) granitoids in the Colombian Andes. (a and b) Trace Element
and REE normalized spider-diagram plots; (c) granite discrimination Ta vs. Yb diagram after
Pearce et al. (1984). VAG volcanic-arc granites, syn-COLG syn-collisional granites, WPG within-­
plate granites, ORG ocean ridge granites

Lithogeochemical data for Permian to mid-Triassic granitoids, displayed in

Figs. 5.7 and 5.8, was sourced from various authors including Saenz (2003), Vinasco
et al. (2006), Cardona et al. (2010b), Leal-Mejía (2011), Cochrane et al. (2014a),
Rodríguez et  al. (2014) and Van der Lelij et  al. (2016). The data set represents
sample collection within diverse geographic and geologic environments throughout
the Colombian Andes and includes some samples from Ecuador and Venezuela.
Lithogeochemical interpretation of the Permo-Triassic granitoids is in some respect
difficult and complex, given the metamorphic conditions, post-emplacement tec-
tonic history, post-crystallization alteration and, in many cases, the poorly defined
geological context of this suite.
The composite data set appears bimodal with respect to SiO2 contents. The more
felsic members yield 57.6–73.5 wt% SiO2, with anomalous, higher SiO2 contents
(73.7–83.0 wt%) considered to be associated with low temperature alteration (silici-
fication). Coeval Permo-Triassic amphibolites reveal lower SiO2 contents (46.5–
52.5 wt%). The felsic series shows a well-defined medium-K to high-K calc-alkaline
282 H. Leal-Mejía et al.

trend on the AFM and K2O vs. SiO2 diagrams, whilst the amphibolites plot apart in
the respective tholeiitic field. The R1-R2 discriminational plot (De la Roche et al.
1980) depicts a clear diorite-tonalite-granodiorite-granite compositional trend for
the felsic suite and highlights the general increase in alkalinity and silica content
brought on by increasing degrees of post-crystallization alteration. On the same plot
the (apparently altered) amphibolites plot in the gabbro-diorite to gabbro-norite
field. A remarkable feature of most of the Permo-Triassic granitoids is their gener-
ally peraluminous character (e.g. Vinasco et al. 2006; Leal-Mejía 2011; Cochrane
et al. 2014a), for both altered and unaltered samples. Again, the amphibolites plot
apart in the metaluminous field. With respect to the Frost et al. (2001) classification,
the felsic subset straddles the magnesian-ferroan granitoid boundary line, although
the majority of the unaltered samples plot clearly on the magnesian side, whilst
some altered and undifferentiated samples plot ferroan. The amphibolites plot
clearly magnesian. The MALI plot (Frost et al. 2001) indicates that the felsic suite
is calc-alkalic and alkali-calcic in composition, whilst the amphibolites plot apart in
the calcic field.
The general bimodal tendency of the Permo-Triassic felsic granitoid-­amphibolite
suite is sustained within the trace element diagrams contained within Fig. 5.8.
The felsic granitoid suite reveals variable trace element patterns. Although some of
the samples suggest arc-related signatures (large-ion lithophile element enrichment,
depletion of high-field strength elements) when normalized to primordial mantle
(Fig. 5.8), Cochrane (2013) notes that when normalized to upper continental crust
compositions, his suite of ca. 275–225 Ma granitoids from Colombian and Ecuador
is indistinguishable from continental crust. REE plots reveal moderate overall REE
enrichment and moderately sloping, fractionated trends for the felsic subset. Slightly
negative or no Eu anomalies are observed. The amphibolites record essentially flat
REE patterns with approximately 10x chondrite concentrations.

Sr-Nd-Pb Isotope Geochemistry

Available Sr, Nd and Pb isotope geochemical data for early Paleozoic, Carboniferous
and Permian to mid-Triassic granitoids and amphibolites from the Colombian
Andes are shown plotted in Fig. 5.9. For comparative purposes Sr, Nd, and Pb isotope
data for early Paleozoic granitoids from the Venezuelan (Merida) Andes (Van der
Lelij et al. 2016) and for Permian-mid-Triassic granitoids (Van der Lelij et al. 2016)
and amphibolites (Chiaradia et al. 2004; Cochrane et al. 2014a), from Venezuela
and Ecuador, respectively, are also plotted.
The early Paleozoic granitoids of the Santander Massif in the eastern Colombian
Andes show notably negative εNd values (εNd(t)  =  −6.0 to −1.3) and high initial
87
Sr/86Sr ratios (87Sr/86Sr(i) = 0.70148–0.71292) (Van der Lelij et al. 2016), suggest-
ing important mixing, assimilation and/or interaction with the upper continental
crust. Lead isotope data for the early Paleozoic Santander granitoids show relatively
high (radiogenic) values (206Pb/204Pb  =  19.01–20.17, 207Pb/204Pb  =  15.68–17.79,
206
Pb/204Pb = 38.88–40.67, Van der Lelij et al. 2016) and plot over the upper crust
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 283

Fig. 5.9  Sr-Nd and Pb isotope plots for pre-Jurassic granitoids in the Colombian Andes. Additional
data for igneous and metamorphic suites from the surrounding region are included for reference

lead isotope evolution curve of the plumbotectonics model of Zartman and Doe
(1981) (Fig.  5.9). The Santander Massif Pb isotopic compositions compare well
with the Pb isotope composition of early Paleozoic granitoids from the Merida
Andes in Venezuela.
With respect to the composite Permian to mid-Triassic suite, no Sr-Nd data are
available for granitoids within the ca. 290–260  Ma age range. Considering the ca.
250–216  Ma ages, however, a subset of granitic gneisses and coeval amphibolites
(including the ca. 240 Ma Santa Elena amphibolite of Cochrane et al. (2014a) and the
ca. 216 Ma Aburrá ophiolite of Correa (2007)), from the Colombia’s Central Cordillera,
is represented. Granitoids from this subset reveal similar, evolved (upper crustal) Sr
and Nd isotope compositions (87Sr/86Sr(i) = 0.70150–0.73106, εNd(t) = −8.91 to −0.76,
Leal-Mejía 2011), when compared with the early Paleozoic suite (Fig. 5.9). These
isotopic compositions contrast markedly with the signatures for the Permo-Triassic
amphibolites of Colombia and Ecuador (87Sr/86Sr(i) = 0.70243–0.70535, εNd(t) = +3.37
to +10.18; Cochrane et al. 2014a), reflecting the bimodal nature of the Permo-Triassic
suite. A crustal provenance for the Central Cordilleran granitoids, without significant
contribution from enriched mantle sources was suggested by Vinasco et al. (2006) and
Cochrane (2013), whilst a primarily mantle-­derived source for the amphibolites was
proposed by Cochrane et al. (2014a; see below).
Pb isotope data for ca. 250–216 Ma Central Cordillera granitoids presented by
Leal-Mejía (2011) also plot over the upper crust lead isotope evolution curve of the
284 H. Leal-Mejía et al.

plumbotectonics model of Zartman and Doe (1981), (206Pb/204Pb  =  18.57–18.89,

207
Pb/204Pb = 15.64–15.69, 206Pb/204Pb = 38.6–39.2), although the data exhibit less
radiogenic compositions than those observed for the early Paleozoic granitoid suite
(Fig. 5.9). The Colombian Permo-Triassic granitoids compare well with Pb isotope
compositions for (meta-)granitoids and granitoid gneisses of similar age, hosted
within the Loja Terrane, Ecuador (Chiaradia et al. 2004; Cochrane et al. 2014a), and
also with compositions for Permo-Triassic granitoids from the Merida Andes,
Venezuela (Van der Lelij et al. 2016) (Fig. 5.9). Ecuador’s Loja Terrane (Aspden
et al. 1992) is considered the geological equivalent and southern extension of the
Cajamarca-Valdivia Terrane, which forms the basement to Colombia’s Central
Cordillera (Cediel et al. 2003). With respect to the Permo-Triassic amphibolites, the
Pb isotope composition of the ca. 240 Ma Santa Elena amphibolite (Cochrane et al.
2014a) is notably less radiogenic than the bulk of the Permo-Triassic meta-granitoid
suite (Fig. 5.9), suggesting a mantle-derived component in the source region.
In contrast to both the early Paleozoic and Permo-Triassic granitoid suites, the ca.
333–310 Ma El Carmen-El Cordero granitoids exhibit 87Sr/86Sr(i) and εNd values plot-
ting up the mantle array (Fig. 5.9). The observed low initial 87Sr/86Sr ratios and positive
εNd(t) values (87Sr/86Sr(i) = 0.70441–0.70516, εNd(t) = +0.58 to +3.79) led Leal-Mejía
(2011) to suggest a primitive, mantle-derived source for the Carmen-El Cordero suite,
without the presence of a significant crustal component. The same suite exhibits some-
what less radiogenic Pb isotope values than early Paleozoic and Permo-Triassic granit-
oids (206Pb/204Pb = 18.45–18.92, 207Pb/204Pb = 15.64–15.67, 206Pb/204Pb = 38.37–38.79,
Leal-Mejía 2011), supporting this conclusion. The El Carmen-El Cordero suite intrudes
Cajamarca-Valdivia Terrane basement. Lead isotope values for El Carmen-El Cordero
can be interpreted to plot on a mixing curve between more mantelic Pb isotope values,
as represented by the Santa Elena and Rio Piedras (Ecuador) amphibolites (Fig. 5.9),
and crustally derived Pb sources as reflected in the Pb isotope composition of the early
Paleozoic and Permo-Triassic meta-granitoid suites.

Additional Isotopic Studies for Early Paleozoic and Permo-Triassic Granitoids

In recent years, an important set of Lu-Hf isotope data has become available for
some of the early Phanerozoic granitoid suites of the Colombian Andes. Lu-Hf
isotope data, when combined with other isotope analyses (e.g. Rb-Sr, Sm-Nd), have
been used to shed additional light upon the potential source regions for granitoid
magmas subject to diverse and prolonged geological histories (e.g. Stevenson and
Patchett 1990; Deckart et al. 2010; Kurhila et al. 2010). Mantilla et al. (2012) pro-
vided Lu-Hf data for early Ordovician granitoids from the Vetas-California district
of the Santander Massif. Van der Lelij (2013) and Van der Lelij et al. (2016) com-
bined Lu-Hf data with additional Sr, Nd and Pb isotope analyses for early Paleozoic
granitoids from throughout the Santander Massif and Mérida Andes (Venezuela),
whilst Cochrane (2013), Cochrane et al. (2014a) and Spikings et al. (2015) supplied
Hf isotope data for various Permo-Triassic granitoids and amphibolites in
Colombia’s Central Cordillera and Ecuador.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 285

Mantilla et al. (2012) interpret radiogenic, initial epsilon Hf (εHfi) values of >0
for a ca. 477 Ma, calc-alkaline meta-diorite collected near Vetas-California, to be
indicative of a depleted mantle source. In this context they interpret the Vetas-­California
granitoid to represent mantle-derived magmas formed within a supra-­subduction zone
setting.
Notwithstanding, Van der Lelij (2013) and Van de Lelij et al. (2016) evaluated
early Paleozoic granitoid magmatism in the Santander Massif and Mérida Andes
(Venezuela) based upon more extensive Lu-Hf (zircon) data, combined with whole-­
rock Rb-Sr and Sm-Nd isotope analyses. These authors indicate that Lu-Hf model
ages of >1.3 Ga are restricted to syn-orogenic (arc- and collision-related) granitoids
which formed during Barrovian metamorphism and crustal thickening between ca.
499 and 472 Ma. They note that these same granitoids yield high initial 87Sr/86Sr
ratios, suggesting a melt derived from evolved, Rb-rich middle to upper crust. A
possible crustal end member source for this crust includes Precambrian basement
units which are exposed in the Garzón Massif and adjacent regions and sedimentary
rocks that host detritus derived from these units (Van der Lelij 2013).
Furthermore, Van der Lelij (2013) and Van de Lelij et  al. (2016) indicate that
subsequent early Paleozoic granitoids, which crystallized between ca. 472 and
452 Ma, yield younger Lu-Hf model ages, with low initial 87Sr/86Sr ratios, suggest-
ing that they were derived at least in part from more juvenile, Rb-poor sources. They
conclude that the overall isotopic composition of post-472 Ma granitoids suggests
melt derived from recycling of variable, lower to upper crustal end members with
unquantified contributions from enriched mantle sources (Van der Lelij 2013).
With respect to the bimodal suite of Permian through Triassic meta-granitoids
and amphibolites, Cochrane (2013) and Cochrane et al. (2014a) provided Hf isotope
data for zircon separates from 14 granitoids and 4 amphibolites collected in Ecuador
and in Colombia’s Central Cordillera. Based upon composite lithogeochemical
data, Cochrane (2013) considers the ca. 275–225 Ma meta-granitoids to be S-type
and to have been derived from an upper crustal source. He notes that coeval zircons
in most of the granitoids yield extremely large intra-sample εHfi variations (e.g.
+3.2 to −11). He considers these variations to be too large to be representative of
magmatic zircons that crystallized from a single, well-mixed source. He suggests
the εHfi variations for the meta-granitoid zircons could be accounted for by source
mixing, although he acknowledges that disequilibration reactions which fractionate
Hf within zircon could also be responsible for some of the variation. In terms of
source mixing, Cochrane (2013) indicates that xenocrystic zircon cores within the
meta-granitoid zircon population return ages ranging from ca. 275 Ma to 1.2 Ga. He
considers these ages to be representative of the range of meta-sedimentary proto-
liths involved in crustal anatexis during petrogenesis of the ca. 275–225  Ma
meta-granitoids.
Ca. 240–223 Ma amphibolites studied by Cochrane (2013) were found to yield
εHfi values that negatively correlate with their 206Pb/238U zircon ages; that is, the
older, ca. 240–232 Ma, amphibolites produced overall less positive εHfi values. He
notes that the ca. 240–232  Ma amphibolites contain complex, oscillatory zoned
286 H. Leal-Mejía et al.

zircons which produce both positive and negative εHfi values. Cochrane (2013)
interprets the data to reflect crustal contamination during older amphibolite emplace-
ment. Conversely, he notes that younger (ca. 225–223  Ma) amphibolites contain
only unzoned zircons which exhibit no intra-sample zircon εHfi variation and return
the most juvenile (i.e. positive) εHfi values (+13 to +15), which approach the
depleted mantle array. He further observes that the least radiogenic volumes of zir-
cons extracted from the amphibolites overlap with the Hf isotopic signatures of the
meta-granitoids (“crustal anatectites”). He concludes that crustal contamination
during emplacement was an important process in the petrogenesis of the older (ca.
240–232 Ma) amphibolites but became progressively less important over time, as
reflected in the isotopic composition of the younger amphibolites.

5.3.3  L
 atest Triassic-Jurassic Granitoid Magmatism:
Distribution, Age and Nature

Late Triassic-Jurassic granitoids represent the most extensive period of magmatic

activity recorded within the present-day geological exposure of the Colombian
Andes. The belt is comprised of a SSW- to NNE-oriented array of volcano-plutonic
arc segments extending from the Ecuador border to the Sierra Nevada de Santa
Marta on the Caribbean coast (Aspden et al. 1987; Cediel and Cáceres 2000; Gómez
et al. 2007; Gómez et al. 2015a). It forms the northern extension of a more extensive
system of late Triassic-Jurassic volcano-plutonic arc segments which continue into
southernmost Ecuador and northern Perú (Litherland et al. 1994; Cediel et al. 2003;
Cochrane 2013; Cochrane et al. 2014b; Spikings et al. 2015).

5.3.3.1  Distribution of Late Triassic to Jurassic Granitoids

The distribution of late Triassic-Jurassic granitoid batholiths, stocks and associated

volcanic sequences in the Colombian Andes is shown in Fig. 5.10, whilst Table 5.1
summarizes the nomenclature, ages and morpho-tectonic position of the major
batholiths and coeval volcanic/volcano-clastic sequences.
Within the Colombian Andes, volcano-plutonic rocks of late Triassic-Jurassic
age cropout in the Garzón and Santander Massifs, the Sierra Nevada de Santa Marta
and within the Central Cordillera and the Serranía de San Lucas. Major batholiths
and associated stocks within the Jurassic belt include, from south and east to north
and west, the Mocoa Batholith in the Garzón Massif (Alfonso 2000); the Santa
Bárbara-Rionegro-Mogotes batholiths and the Pescadero, La Corcova and Páramo
Rico plutons in the Santander Massif (the “Santander Plutonic Group” of Ward
et  al. 1973, also described by Royero and Clavijo 2001); the Ibagué Batholith
(Nelson 1957; Núñez 1998; Altenberger and Concha 2005); the Norosí (Guamocó)-
San Martín batholiths of the Serranía de San Lucas (the “San Lucas granitoids”
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 287

Fig. 5.10  Distribution of latest Triassic through Jurassic granitoids in the Colombian Andes.
Principle modern-day physiographic provinces of the region are shown for reference. (Granitoid
shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al. 2015a)
288 H. Leal-Mejía et al.

Table 5.1  Summary of latest Triassic and Jurassic granitoid plutonism and coeval volcanism in
the Colombian Andes (basement domains refer to litho-tectonic and morpho-structural units
defined by Cediel et al. (2003) and reviewed in text (see Fig. 5.2))
Intrusive Coeval volcanism
age
range Age range
Major (U-Pb (U-Pb Related
batholith/arc zircon, zircon, porphyritic Physiologic Basement
segment Ma) Unit Ma) intrusions region domain
Santander Ca. Jordán No None Santander MSP
Plutonic 210–196 Fm. available known massif (northwesternmost
group: age dates Guiana shield)
Santa
Bárbara -
Rionegro -
Mogotes
Batholiths
and
The
Pescadero,
La Corcova
and
Páramo Rico
plutons
Southern Ca. Southern No None Central CTR
Ibagué 189–182 Ibagué available known Cordillera
(CA-VA
Batholith Volcanics age dates -Chicamocha
(Saldaña Terrane
Fm.a) Contact)
Norosí and Ca. Noreán Ca. Santa Cruz Serranía de CTR
San Martín 189–182 Fm. 201– (ca. san Lucas (Chicamocha
batholiths 174 Ma 178 Ma) Terrane
Sierra Ca. 180 Guatapurí Ca. None Sierra MSP
Nevada de Fm. 183 Ma known Nevada de (N westernmost
Santa Marta Santa Marta Guiana shield)
batholiths
(pueblo
Bello-­
Patillal and
Aracataca-­
central)
Mocoa-­ Ca. Mocoa Ca. Mocoa Garzón Western Guiana
Garzón trend 179–173 trend 185 Ma (ca. massif shield (Amazon
batholiths Volcanics 170 Ma) Craton?)
(Saldaña
Fm.a)
(continued)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 289

Table 5.1 (continued)
Intrusive Coeval volcanism
age
range Age range
Major (U-Pb (U-Pb Related
batholith/arc zircon, zircon, porphyritic Physiologic Basement
segment Ma) Unit Ma) intrusions region domain
Northern Ca. Northern Ca. Infierno-­ Central CTR
Ibagué 169–152 Ibagué 158 Ma Chilí, Cordillera (CA-VA
Batholith Volcanics Rovira, -Chicamocha
(Saldaña Chaparral Terrane
Fm.a) (ca. Contact)
149–
146 Ma)
Segovia Ca. None No None Northern (CTR)
Batholith 167–158 Identified associated known Central Ca-VA
volcanic Cordillera
rocks
MSP Maracaibo Sub-Plate, CTR Central Continental Realm, CA-VA Cajamarca-Valdivia Terrane
The Saldaña Fm. as presently understood appears to be a regionally extensive but diachronous unit
a

which requires further subdivision

described by Clavijo et al. 2008); the Segovia Batholith (Feininger et al. 1972); and
the Aracataca, Central, Pueblo Bello and Patillal batholiths of the Sierra Nevada de
Santa Marta (Tschanz et al. 1974).
In addition to the major suites of plutonic rocks, important deposits of associated
volcanic and volcano-sedimentary strata of late Triassic-Jurassic age are preserved.
These deposits include those related to the Mocoa and Ibagué batholiths (e.g.
Saldaña Fm.), those bordering the Santander Massif (i.e. Jordán Fm.), those associ-
ated with the Norosí Batholith of the Serranía de San Lucas (i.e. Noreán Fm.) and
those observed along the south-eastern flank of the Sierra Nevada de Santa Marta
(i.e. Guatapurí Fm.). These sequences are considered to be generally penecontem-
poraneous in age with their neighbouring batholiths.

5.3.3.2  A
 ge Constrains on Late Triassic to Jurassic Granitoid
Magmatism

Geologic field relationships and historic radiometric whole-rock or mineral separate

age dates (i.e. Maya 1992; Gómez et al. 2015b) have, as a whole, provided temporal
constraint upon the emplacement of late Triassic-Jurassic granitoids in the
Colombian Andes. Regardless, lack of analytical resolution, large margins of error
and overlap in the historic data have led to the assignment of the entire suite to a
broad interval spanning the late Triassic to early Cretaceous, as reflected in the
undifferentiated intrusives displayed upon regional geologic maps (e.g. Cediel and
290 H. Leal-Mejía et al.

Cáceres 2000; Gómez et  al. 2007; Gómez et  al. 2015a). The present work has
assessed Colombian late Triassic-Jurassic granitoid magmatism from a regional
perspective. Sixty-eight high-precision zircon U-Pb age dates have been compiled
from Dörr et al. (1995), Bustamante et al. (2010), Leal-Mejía (2011), Villagómez
et al. (2011), Cochrane (2013), Mantilla et al. (2013), Van der Lelij (2013), Bissig
et al. (2014), Cochrane et al. (2014b), Van der Lelij et al. (2016) and Zapata et al.
(2016), providing data for the Norosí and San Martín batholiths, the southern and
northern segments of the Ibagué Batholith, the Segovia Batholith, the Pueblo Bello-­
Patillal Batholith and holocrystalline and porphyritic intrusive in the Garzón Massif
and Mocoa Batholith. In addition, 12 zircon U-Pb ages for late Triassic-Jurassic
volcano-sedimentary sequences, including the Noreán (eastern flank of Norosí
Batholith), Guatapurí (southeastern flank of Pueblo Bello Batholith) and Saldaña
(southern Ibagué Batholith) Fms., were compiled.
The temporal distribution of zircon U-Pb ages for late Triassic-Jurassic granit-
oids is displayed in Fig. 5.11. This histogram permits the definition of four mag-
matic sub-episodes spanning the ca. 210 and 146 Ma time period, represented by
granitoid batholith emplacement in six spatially separate arc segments (Fig. 5.10).
The oldest, ca. 210–196 Ma sub-episode, is confined to the batholiths and stocks of
the Santander Plutonic Group (Dörr et al. 1995; Mantilla et al. 2013; Bissig et al.
2014; Van der Lelij 2013; Van der Lelij et al. 2016). A second ca. 189–180 Ma
sub-­episode is recorded by the Norosí and San Martín batholiths, the Pueblo Bello-­

Fig. 5.11  U-Pb (zircon) age date populations for latest Triassic through Jurassic granitoids in the
Colombian Andes. Note the clustering of age dates for individual arc segments and how the age
populations support the overall east-to-west migration of the granitoid arc axis over time. See text
for further discussion (also see Figs. 5.10, 5.31, and 5.32)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 291

Patillal Batholith and the southern Ibagué Batholith (Leal-Mejía 2011). Zircon
U-Pb analyses for the Guatapurí and southern Ibagué volcanic formations return
ages penecontemporaneous with the age of spatially related holocrystalline plutons.
Zircon separates from volcanic rocks of the Noreán Fm. return a wider range of
ages, spanning ca. 202–172  Ma. The base of the Noreán Fm., to the east of the
Norosí Batholith, is comprised of andesite flows and felsic pyroclastic rocks with
associated diorite dikes and felsic plugs. This bimodal assembly dates from ca. 201
to 193 Ma.
A third sub-episode, emplaced at ca. 180–172  Ma, is revealed in the Mocoa
Batholith and intrusive and volcanics exposed along the margins of the Garzón
Massif (Bustamante et al. 2010; Leal-Mejía 2011; Cochrane et al. 2014b; Zapata
et al. 2016). Rhyolite tuff of spatially related volcanic rocks returned a U-Pb (zir-
con) age of 181.5 ± 1.6 Ma (Cochrane et al. 2014b). Previous work by Sillitoe et al.
(1982) provided K-Ar (magmatic biotite) ages of 210 ± 4 Ma and 198 ± 4 Ma for
the Mocoa Batholith. U-Pb (zircon) data do not support the Sillitoe et al. (1982)
K-Ar ages, although we note that the U-Pb samples were collected significantly
(>10 km) to the west of the Sillitoe et al. (1982) locations. Regardless, no evidence
for ~210 to 198 Ma magmatism along the Mocoa-Garzón trend is provided by the
U-Pb (zircon) data, and based upon our multi-sample database, we conclude that the
majority of the Mocoa-Garzón granitoids crystallized between ca. 180 and 172 Ma.
A fourth ca. 169–152 Ma sub-episode is recorded in granitoids of the northern
Ibagué (Leal-Mejía 2011; Villagómez et al. 2011; Cochrane 2013; Cochrane et al.
2014b) and Segovia batholiths (Leal-Mejía 2011). With respect to the Ibagué
Batholith, zircon U-Pb ages indicate that it is a large composite intrusive comprised
of at least two temporally spatially defined magmatic pulses at ca. 189–182 Ma and
ca. 165–152 Ma. Unfortunately, available data does not permit the precise definition
of the contact between the southern and northern sectors. The contact shown in
Fig. 5.10 is an approximation based upon field and DEM observations and historic
K-Ar age data. With respect to the Segovia Batholith (the Western Batholith of
Bogotá and Aluja 1981 or Segovia Batholith of Ballasteros 1983), the 188.9 ± 2 Ma
age presented by Cochrane (2013) pertains to a sample which is actually located
within the southern Norosí Batholith (compare our Fig. 5.10 with Cochrane 2013,
Fig. 5.1 on p. 88), well to the east of the mapped limits of the Segovia Batholith. The
Cochrane (2013) sample is herein included in the ca. 189–180 Ma San Lucas suite
and accords well with previous age dates for the Norosí Batholith.
In addition to the four major episodes of holocrystalline plutonism outlined
above, three localized events comprised of hypabyssal, porphyritic-textured dikes,
sills and stocks are observed (Fig. 5.10). These include (1) a cluster of porphyritic
dikes and sills at Santa Cruz on the NW margin of the Serranía de San Lucas, a
sample from which returned a zircon U-Pb age of 178.1  ±  5.6  Ma (Leal-Mejía
2011); (2) porphyritic stocks at Mocoa (Sillitoe et  al. 1984) a sample of which
returned a zircon U-Pb age of 170.2 ± 2.7 Ma with an inheritance ages ranging from
ca. 184 Ma to ca. 1200 Ma (Leal-Mejía 2011); and (3) numerous porphyritic stocks
emplaced along the eastern margin of the northern Ibagué Batholith. Hypabyssal
porphyry from Infierno-Chilí area returned a zircon U-Pb age of 149.3 ± 2.8 Ma
292 H. Leal-Mejía et al.

(Leal-Mejía 2011). Cochrane (2013) revealed a 146.8 ± 1.5 Ma zircon U-Pb age for
quartz porphyry near Lérida. Numerous similar, undated porphyritic stocks outcrop
along the eastern margin of the northern Ibagué Batholith, extending from Rovira
south to beyond Chaparral (Fig. 5.10) and are herein assigned to the same tem-
poral suite.

Discussion and Synthesis of Spatial Distribution of Late Triassic-Jurassic

Granitoids

In the synthesis of Meso-Cenozoic granitoid magmatism presented by Aspden et al.

(1987), “eastern” and “western” granitoid belts of late Triassic- Jurassic age were
recognized. These authors suggested that the eastern belt (including Santander and
Mocoa) may be older than the western (Ibagué-Sonsón1-Segovia2-Sierra Nevada de
Santa Marta) belt.
(Footnotes: 1The Sonsón Batholith, then thought to be of Jurassic age, has now
been shown to be a composite intrusion of Permo-Triassic and Paleocene age (see
Leal-Mejía et al. 2011). 2 Aspden et al. (1987) grouped the Norosí and San Martín
batholiths of the Serranía de San Lucas with the Segovia Batholith of Bogotá and
Aluja (1981). They provided no data for Norosí and San Martín and hence did not
recognize that they represent temporally distinct batholiths).
Spikings et al. (2015), based upon new and compiled U-Pb (zircon) age dates,
presented an analysis of late Triassic-Jurassic granitoid magmatism at the scale of
the entire Northern Andes. In Colombia they reiterate the westward migration of
granitoid magmatism from the Santander Plutonic Group to the Norosí Batholith
between ca. 196 and 189 Ma; however, they do not differentiate the ca. 168–158 Ma
Segovia Batholith as defined by Bogotá and Aluja (1981), Ballesteros (1983), Leal-­
Mejía (2011) and Leal-Mejía et al. (2011) and detailed herein. To the south, Spikings
et  al. (2015) group the northern and southern Ibagué and Mocoa-Garzón trend
batholiths as the undifferentiated Ibagué Batholith and reveal a composite U-Pb age
ranging from ca. 189 to 146 Ma. Their data demonstrate that the zircon U-Pb ages
tend to cluster within restricted ranges within distinct sectors of the batholith and
that the southern sector of the Ibagué Batholith returns significantly older ages
when compared to the Mocoa-Garzón trend and northern Ibagué (see their Fig. 5.10
and our Figs. 5.10 and 5.11).
When our zircon U-Pb database is combined with that presented by Spikings
et  al. (2015), a more detailed temporal-spatial analysis of late Triassic-Jurassic
granitoid magmatism is permitted. Four distinct age ranges are observed within at
least six separate arc segments, and the temporal-spatial migration of late Triassic-­
Jurassic granitoid magmatism based upon U-Pb (zircon) ages may be visualized in
the colour coding of Figs. 5.10 and 5.11. An E-W transect across the northern sector
of the Colombian Cordilleras highlights an east-to-west younging trend beginning
with the ca. 210–196  Ma batholiths of the Santander Massif, passing westwards
through the ca. 189–180 Ma batholiths of the Serranía de San Lucas and into ca.
167–158 Ma Segovia Batholith. This tendency is accentuated if the estimated 100
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 293

kilometres of post-Jurassic sinistral displacement along the Bucaramanga–Santa

Marta fault system (Campbell 1968; Etayo-Serna and Rodríguez 1985; Cediel et al.
2003) are restored, placing the SW sector of the Pueblo Bello-Patillal Batholith and
associated Guatapurí volcanics in very close proximity and immediately along trend
with the Serranía de San Lucas intrusive-volcanic suite (Fig. 5.10). Notably, this
restoration will not affect the position of the El Jordán Fm., which rests along the
NW margin of the Santander Massif, but on the west side of the Bucaramanga–
Santa Marta Fault. This suggests that the Jordán Fm. along with the Guatapurí and
Noreán Fms. forms remnants of a formerly unified volcanic province.
To the south, in the northern and southern Ibagué batholiths, the Mocoa Batholith
and the intrusions exposed along the Garzón Massif, the east-to-west younging trend
is not clearly defined. The ca. 188–180 Ma (southern Ibagué) and ca. 180–172 Ma
(Mocoa-Garzón) episodes migrate along a NNE-oriented axis, and an apparent
southward and eastward migration of the magmatic arc axis is recorded. Current
regional structural interpretations depict the ca. 180–172 Ma Mocoa-­Garzón intru-
sions as tectonic slices caught up in dextral oblique basement reactivation structures
responsible for Miocene uplift in the Garzón Massif (Fig. 5.2; Cediel and Cáceres
2000; Cediel et al. 2003). If restoration of an (albeit) undefined component of post-
emplacement dextral translation along the Garzón Massif structures is taken into
account, the apparent eastward migration of magmatism is reduced (although not
completely eliminated), and slices of the Mocoa Batholith become coaxial with the
ca. 189–180 Ma southern Ibagué Batholith.
Following emplacement of the Mocoa-Garzón intrusions, the ca. 166–152 Ma
northern segment of the Ibagué Batholith was intruded along trend to the NNE. Thus,
granitoid migration in the southern segment of the late Triassic-Jurassic arc during
the ca. 189–152 Ma period was primarily along a NNE-oriented axis.
With respect to the location and timing of hypabyssal, porphyritic dikes, sills and
stocks, it is observed that the three temporal suites were emplaced within and/or
along the contacts with a respective major batholith of penecontemporaneous age
(Fig. 5.10). Thus, the ca. 178 Ma Santa Cruz dikes and sills are located along the
NW margin of the ca. 189–180 Ma Norosí Batholith, the ca. 170 Ma Mocoa stocks
are located within the ca. 180–172 Ma Mocoa Batholith, and the ca. 152–146 Ma
Rovira-Lerida stocks are emplaced along the eastern margin of the ca. 168–155 Ma
northern Ibagué Batholith. In all cases, porphyritic magmatism was initiated within
ca. 2–5  m.y. following the waning of holocrystalline plutonism. The porphyritic
granitoids thus appear to represent closure-phase magmatism emplaced during the
late evolution of the respective holocrystalline arc segment, prior to wholesale arc
migration or cessation of active magmatism.
In summary, zircon U-Pb ages for late Triassic-Jurassic granitoid magmatism in
the Colombian Andes permit the temporal definition of four major magmatic epi-
sodes including granitoid batholith emplacement within six spatially separate arc
segments, in addition to three spatially temporally separate events involving late-­
stage, volumetrically minor hypabyssal porphyries. Each major batholith (or group
of batholiths in the case of the Santander Plutonic Group) is considered to represent
a temporally and spatially separate arc segment developed within the overall context
294 H. Leal-Mejía et al.

of late Triassic-Jurassic subduction-related granitoid magmatism affecting much of

Northern Andean margin during this time period (Aspden et al. 1987; Litherland
et  al. 1994; Cediel et  al. 2003; Leal-Mejía et  al. 2011; Cochrane et  al. 2014b;
Spikings et al. 2015). In Colombia, magmatism migrates over time, both along the
length of the magmatic arc axis and in a transverse sense, related to the interpreted
movement vector of the subducting Pacific Plate. Tectonic setting and evolution
during the late Triassic-Jurassic will be discussed in Sect. 5.4.2, following the
presentation of additional lithogeochemical and isotopic information.

5.3.3.3  L
 ithogeochemical and Isotopic Characteristics
of Late Triassic-­Jurassic Granitoids

Lithogeochemistry

Figures 5.12 and 5.13 present whole-rock lithogeochemical analyses for 136 samples
of late Triassic-Jurassic granitoids, including holocrystalline and hypabyssal intru-
sive and volcanic rocks, as compiled from Dörr et  al. (1995), Bustamante et  al.
(2010), Leal-Mejía (2011), Bissig et al. (2014), Cochrane et al. (2014b) and Van der
Lelij et al. (2016). Of the samples represented herein, some 36% (49 samples) are
considered altered, based upon the criteria discussed in Sect. 5.3.1.2. Altered sam-
ples are identified by the unfilled symbols used in Figs.  5.12 and 5.13 lithogeo-
chemical plots. No lithogeochemical data are available for the ca. 180 Ma batholiths
of the Sierra Nevada de Santa Marta or their coeval volcano-sedimentary sequences
(i.e. the Guatapurí Fm.).
The Colombian late Triassic to Jurassic batholiths are low-K to high-K calc-­
alkaline (Irvine and Baragar 1971; Peccerillo and Taylor 1976) in composition. All
main phase batholiths are metaluminous, with the exception of the Santander
Plutonic Group where localized peraluminous members are recorded (e.g. Bissig
et al. 2014). Specific lithogeochemical features of the individual granitoid suites are
reviewed below.

Santander Plutonic Group

The ca. 210–196 Ma Santander Plutonic Group produces the most differentiated
trend of bulk compositions, ranging from gabbro-diorite through to granite and
leucogranite (“alaskite”). Relative to the other Colombian late Triassic-Jurassic
intrusive suites, the Santander Plutonic Group reveals (1) a higher degree of alka-
linity, (2) a tendency towards weakly to strongly peraluminous compositions
(including the leucogranites of the Vetas-California area, Bissig et al. 2014) and (3)
enrichment in trace elements and REE (ΣREE = 47.42–503.96 ppm). Overall, REE
patterns reveal moderate to steep decreasing slopes ((La/Yb)N  =  3.47–34.22).
The HREE define a relatively flat pattern for the intermediate members ((Gd/
Yb)N  =  0.63–2.54) and a slightly decreasing pattern for the leucogranites ((Gd/
Yb)N = 0.51–3.35).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 295

Fig. 5.12  Major element lithogeochemical plots for latest Triassic through Jurassic granitoids in
the Colombian Andes. (a) AFM plot, curve after Irvine and Baragar (1971); (b) K2O vs. SiO2 plot,
boundary fields in grey as summarized by Rickwood (1989); (c) alumina saturation plot after
Barton and Young (2002); (d and e) MALI and Fe-index vs. SiO2 plots, respectively, after Frost
et al. (2001); (f) R1-R2 classification plot after De La Roche et al. (1980). Th tholeiite, C-A calc-­
alkaline, Sh Shoshonite, Gb No gabbro-norite, Gb Di gabbro-diorite, Di diorite, Mz Di monzodio-
rite, Mz monzonite, To tonalite, Qtz Mz quartz monzonite, Gd granodiorite, Gr granite, Alk Gr
alkali granite
296 H. Leal-Mejía et al.

Fig. 5.13  Trace element and REE lithogeochemical plots for latest Triassic through Jurassic gran-
itoids in the Colombian Andes. (a and b) Trace element and REE normalized spider diagram plots;
(c) granite discrimination Ta vs. Yb diagram after Pearce et al. (1984). VAG volcanic-arc granites,
syn-COLG syn-collisional granites, WPG within-plate granites, ORG ocean ridge granites

Norosí and San Martín Batholiths

The ca. 189–182 Ma Norosí and San Martín batholiths (San Lucas granitoids) follow
a similar although less alkalic trend to that observed in the Santander Plutonic Group.
Compositional variations are restricted to diorite through granodiorite, and no peralu-
minous tendency is observed. REE concentrations are moderate (ΣREE = 117.24–
146.08 ppm) and, as with the intermediate suite of Santander, reveal a moderately
pronounced negative Eu anomaly (Eu/Eu* = 0.58–0.82). The LREE however are dis-
tinctly less enriched producing more moderately decreasing slopes ((La/Sm)N = 2.73–
3.81). Relatively flat patterns for HREE are also observed ((Gd/Yb)N = 0.83–1.31).
Volcanosedimentary rocks of the Noreán Fm. in the Serrania de San Lucas exhibit
similar REE patterns to the Norosi and San Martin granitoids.

Southern Ibagué Batholith

The ca. 189–180 Ma granitoids of the southern Ibagué Batholith are compositionally
more variable when compared to Norosí and San Martín, ranging from metaluminous
calc-alkaline gabbro and diorite through high-K (alkali-calcic) granodiorite, quartz
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 297

monzonite and locally granite. The population appears to be bimodal, but this may be
a reflection of the relatively small sample set. Portions of the southern Ibagué Batholith
are pyroxene-dominant with lessor amounts of biotite. The REE are less enriched than
both Santander and Norosí-San Martín (ΣREE = 88.07–209.80 ppm). The decreasing
LREE slopes are somewhat steeper ((La/Sm)N = 2.89–5.83) than those observed in
Norosí-San Martín ((La/Sm)N  =  2.73–3.81), whilst Eu anomalies are only weakly
negative to slightly positive (Eu/Eu* = 0.73–1.09). The volcanic rocks of the Saldaña
Formation have slightly higher REE contents (ΣREE = 118.30–240.33 ppm) and
similar weak negative Eu anomalies (Eu/Eu* = 0.77–0.95).

Mocoa-Garzón Trend Batholiths

The ca. 180–172 Ma granitoids of the Mocoa-Garzón trend show a metaluminous,
high-K calc-alkaline (alkali-calcic) character associated with hornblende-biotite-­
bearing granodiorite to monzogranite compositions. REE contents in the phaneritic
granitoids are enriched (e.g. Altamira granite; ΣREE = 223.40 ppm), when com-
pared to the Norosí, San Martín and southern Ibagué batholiths, although not to the
degree as seen in the Santander Plutonic Group. The Mocoa porphyries
(ΣREE = 100.18–104.76 ppm) are distinctly less enriched in REE than the penecon-
temporaneous phaneritic granitoids (ΣREE = 113.95–175.06 ppm). REE patterns
include relatively steep decreasing slopes ((La/Yb)N = 7.97–16.77). Eu anomalies
are moderately negative for the phaneritic granitoids (Eu/Eu*  =  0.67–0.83) and
slightly negative to weakly positive for porphyries (Eu/Eu* = 0.95–1.10).

Northern Ibagué and Segovia Batholiths

The ca. 168–155 Ma granitoids of the northern Ibagué and Segovia batholiths pres-
ent metaluminous, low-K calc-alkaline (calc-alkalic to calcic after Frost et al. 2001)
compositions dominated by hornblende with lesser biotite-bearing diorite to quartz
diorite. REE patterns are flatter ((La/Yb)N = 3.58–16.3) and values are overall less
enriched (ΣREE = 57.63–141.49 ppm) than those observed for Norosí-San Martín
and southern Ibagué. Northern Ibagué shows slightly negative to moderately posi-
tive Eu anomalies (Eu/Eu* = 0.77–1.47), similar to southern Ibagué, whereas very
weak negative Eu anomalies are observed in samples from the Segovia Batholith
(Eu/Eu* = 0.72–1.12).

Northern Ibagué Hypabyssal Porphyry Suite

Lithogeochemical data are limited for the ca. 152–145 Ma porphyritic granitoids of
the northern Ibagué Batholith, and the data set contains analyses for various undated
hypabyssal porphyry intrusions of inferred latest Jurassic age. The suite includes
metaluminous, low-K calc-alkaline (calc-alkalic to calcic) gabbro-diorite to grano-
diorite with less enriched REE contents (ΣREE = 68.07–124.50 ppm) and similar
slopes ((La/Yb)N = 2.97–9.67) when compared to the northern Ibague and Segovia
batholiths. Eu anomalies for the porphyries are very weakly negative to slightly
positive (Eu/Eu* = 0.91–1.08).
298 H. Leal-Mejía et al.

Whole-Rock Lithogeochemistry Summary and Discussion for Late Triassic-­Jurassic

Granitoids
The composite lithogeochemical data set used in this study demonstrates the
Colombian late Triassic to Jurassic batholiths are of low-K to high-K calc-alka-
line (Irvine and Baragar 1971; Peccerillo and Taylor 1976) or magnesian, calcic
to alkali-calcic (Frost et al. 2001) in composition. All main phase batholiths are
metaluminous, with the exception of the Santander Plutonic Group where a clear
trend towards strongly peraluminous compositions is recorded in the leucogranite
suite of Bissig et al. (2014). The granitoids are dominated by biotite and/or horn-
blende diorite through granodiorite but include localized ranging gabbrodioritic,
monzonitic and granitic phases. Consistent LILE enrichment compared to HFSE
is observed, as are negative anomalies of refractory elements such as Ta, Nb and
Ti. LREE enrichment compared to HREE is recorded in all suites. Eu anomalies
range from markedly negative to slightly positive, being generally consistent
within individual batholiths. The foregoing characteristics are consistent with
classification of the Colombian granitoids as Cordilleran-type granitoids (Frost
et al. 2001), volcanic arc granitoids (Pearce et al. 1984), or K-spar and amphibole-
rich calc-alkaline granitoids (Barbarin 1999), generated in transitional to subduc-
tion-type settings. This conclusion is consistent with data and conclusions
presented by previous workers including Alvarez (1983), Aspden et  al. (1987),
Dorr et al. (1995), Bustamante et al. (2010), Leal-Mejía et al. (2011), Bissig et al.
(2014) and Spikings et al. (2015).
When the composite lithogeochemical data set for the late Triassic to Jurassic
Colombian granitoids is considered within the spatial-temporal and geological
framework for the individual batholiths presented in Fig. 5.10, the lithogeochemical
plots (Figs.  5.12 and 5.13) demonstrate clear east-to-west trends towards more
primitive (less alkaline, more magnesian, less enriched in both trace and REEs, less
fractionated REEs, weaker to no Eu anomaly) whole-rock compositions. In north-
ern Colombia this is recorded in the enriched, calc-alkalic to alkali-calcic (high-K
calc-alkaline) and peraluminous compositions of the Santander Plutonic Group,
westwards through the intermediate compositions of the San Lucas granitoids, and
into the less enriched, calcic to calc-alkalic, metaluminous compositions of the
Segovia Batholith. As shown in Fig. 5.10, this tendency coincides with the east-to-­
west younging of the major batholiths and with changes in the nature and composi-
tion of the intruded basement. In southern Colombia a similar lithogeochemical
trend from the alkali-calcic compositions of the Mocoa-Garzón granitoids to the
calcic to calc-alkalic compositions of the southern Ibagué Batholith to the west is
observed, although the age relationships are reversed, with the Mocoa-Garzón gran-
itoids being younger than the southern Ibagué Batholith. This suggests that the
major, minor, trace and rare-earth element lithogeochemistry of the individual
batholiths is less a function of age as it is of the nature of the intruded basement and,
likely, the specific tectonic framework and conditions at the time of emplacement
(Barbarin 1999; Frost et al. 2001). These themes will be discussed further following
presentation of isotopic data for the Colombian granitoids.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 299

When the lithogeochemistry of the hypabyssal porphyry suites (i.e. the ca.
178 Ma Santa Cruz dikes and sills, ca. 170 Ma Mocoa porphyries and the ca. 152–
146  Ma northern Ibagué porphyries) is compared with the respective, spatially
related, slightly older, holocrystalline batholith, in general, the porphyries tend to
(1) be less enriched in K (i.e. less alkaline), (2) be less enriched in trace elements
and REE and (3) have a less pronounce to neutral or even positive Eu anomaly.
These trends are best observed in the unaltered porphyries of the northern Ibagué
Batholith and are present but potentially modified by post-crystallization alteration
and mineralization at Santa Cruz and Mocoa. Notwithstanding, the data suggest that
the hypabyssal porphyry suites consistently reveal more mantelic compositions
when compared to the spatially related, slightly older, holocrystalline batholith.

Sr-Nd-Pb Isotope Geochemistry

Sr-Nd Isotope Geochemistry Results

Sr-Nd isotope data for Colombian late Triassic-Jurassic granitoids, including for the
Santander Plutonic Group (Restrepo-Pace, 1995; Bissig et al. 2014; Van der Lelij
et al. 2016), the Mocoa Batholith porphyries (Leal-Mejía 2011), the Norosí and San
Martín batholiths, Santa Cruz porphyries and the Noreán volcanics (Leal-Mejía
2011; Cochrane et al. 2014b), the southern Ibagué Batholith and the southern Ibagué
volcanics (Leal-Mejía 2011), the central and northern Ibagué Batholith and hyp-
abyssal porphyries (Cochrane et al. 2014b) and the Segovia Batholith (Leal-­Mejía
2011), are presented in Fig.  5.14 and tabulated in Appendix 3. For comparative
purposes, selected data sets for the Precambrian basement rocks of the Santander
Massif (Cordani et  al. 2005; Ordoñez-Carmona et  al. 2006; Bissig et  al. 2014),
Garzón Massif (Restrepo-Pace et al. 1997; Cordani et al. 2005) and Chicamocha
Terrane (Cuadros et al. 2014) are also presented.
Late Triassic–early Jurassic intrusives of the Santander Plutonic Group are char-
acterized by a wide range of highly radiogenic 87Sr/86Sr(i) ratios (0.70533 to 0.73660)
with negative εNd(t) values (−19.34 to −3.46). Data plot within a similarly disparate
field defined by samples of Precambrian and Paleozoic continental basement rocks
of the Santander Massif (Cordani et al. 2005; Ordoñez-Carmona et al. 2006; Bissig
et al. 2014).
No Sr-Nd data are available for the main phase Mocoa-Garzón trend batholiths.
Results for the ca. 170  Ma Mocoa porphyries reveal moderate to high 87Sr/86Sr(i)
ratios (~0.70600) and negative εNd(t) values (−5.60 to −3.32). εNd(t) values for the
Mocoa porphyries plot within the range presented by Restrepo-Pace et al. (1997)
and Cordani et al. (2005); however, full comparison of the data sets is hampered by
the lack of 87Sr/86Sr analyses for Garzón Massif basement.
Unlike the Santander Plutonic Group, data for the Norosí and San Martín
batholiths plot within a more discrete array, characterized by moderately high
87
Sr/86Sr(i) ratios (0.70674–0.70826) with negative εNd(t) values (−6.65 to +0.09).
Coeval volcanic rocks of the Noreán Fm. return somewhat more mantelic εNd(t)
300 H. Leal-Mejía et al.

Fig. 5.14  Sr-Nd and Pb isotope plots for latest Triassic through Jurassic granitoids in the
Colombian Andes. Additional data for igneous and metamorphic suites from the surrounding
region are included for reference

signatures than the Norosí and San Martín batholiths. Conversely, the Santa Cruz
porphyry dikes, on the western margin of the Serranía de San Lucas, record a high
87
Sr/86Sr(i) ratio (0.70851) and a slightly more negative εNd(t) value (−6.9).
The southern Ibagué Batholith and associated volcanic rocks yield mixed initial
87
Sr/86Sr ratios around the bulk Earth composition plotting within or near the mantle
array (87Sr/86Sr(i) = 0.70489 to 0.70609; εNd(t) = −0.96 to +4.83). No Sr isotope data
is available for the northern Ibagué Batholith, although Nd isotope data presented
by Cochrane et al. (2014b) record positive εNd(t) values (+0.32 to +3.86) similar to
the εNd(t) values for the southern Ibagué Batholith.
Finally, samples from the Segovia Batholith exhibit the lowest 87Sr/86Sr(i) ratios
(0.70385 to 0.70434) and positive εNd(t) values (+0.86 to +6.52), generally falling
along the mantle array.

Sr-Nd Isotope Geochemistry Summary and Discussion

Whole-rock Sr-Nd isotope data for the Colombian late Triassic-Jurassic granitoids
plot in clusters, on a per-intrusive suite basis, with relatively little overlap between
the sample sets for individual batholiths (Fig. 5.14). The overall 87Sr/86Sr(i) isotope
composition of the Colombian granitoids ranges from the highly evolved values of
the ca. 210–196 Ma Santander Plutonic Group (87Sr/86Sr(i) = 0.70533 to 0.73660) to
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 301

the juvenile values of the ca. 168–155 Ma Segovia Batholith (87Sr/86Sr(i) = 0.70385

to 0.70434). Commensurate with these data, εNd(t) values for the Santander Plutonic
Group are negative (εNd(t) = −19.34 to −3.46), whilst the Segovia Batholith returns
εNd(t) values up to +6.52. The Norosí-San Martín and southern and northern Ibagué
batholiths return intermediate 87Sr/86Sr(i) and εNd(t) values which plot in semi-­
discrete arrays between the above-mentioned data sets (Fig. 5.14).
As with the lithogeochemical data presented above, 87Sr/86Sr(i) and εNd(t) data
may be considered within a spatial-temporal and geological framework (Fig. 5.10).
In the northern sector of the Colombian Cordilleras, 87Sr/86Sr(i) and εNd(t) data reveal
an east-to-west trend of increasingly more juvenile 87Sr/86Sr(i) and εNd(t) values,
extending from the highly evolved, upper crustal-influenced compositions of the ca.
210–196 Ma Santander Plutonic Group to the mixed values of the ca. 189–180 Ma
Norosí and San Martín batholiths which cluster at the base and along the lower sec-
tion of the mantle array, into to the primative 87Sr/86Sr(i) and εNd(t) values for the
Segovia Batholith (Fig.  5.14). A similar east-to-west tendency is observed in the
south where the Mocoa porphyry, hosted within the Garzón Massif, returns mixed
crustal values at the base of the mantle array, whilst the southern Ibagué Batholith
to the west returns higher (positive) εNd(t) values which plot farther up the mantle
array (Fig. 5.14).
Placed into a geological context, the above east-to-west trend is supported by
changes in the nature of the intruded basement complex as shown in Fig. 5.10. In the
north, Sr and Nd isotope data for the Santander Plutonic Group plot within the broad
data field outlined for samples of the Bucaramanga gneiss with a distinct tendency
towards upper crustal values. This suggests partial derivation and/or contamination
of the Santander Plutonic Group granitoids from Precambrian and/or early Paleozoic
basement rocks widely exposed in the Santander Massif (Goldsmith et al. 1971). To
the west, Sr and Nd isotope data for the Norosí and San Martín batholiths of the San
Lucas region are more tightly clustered mostly near the base of and extending up to
the lower section of the mantle array. Sr-Nd isotope characterization of the base-
ment rocks in the Serranía de San Lucas is restricted to analyses for the metamafic
constituents which return 87Sr/86Sr(i) and εNd(t) values consistent with a depleted
mantle or lower crustal source, whilst 87Sr/86Sr(i) values for the felsic basement com-
ponents of the region are poorly constrained (Cuadros et  al. 2014; Fig.  5.14).
87
Sr/86Sr(i) and εNd(t) data for the San Lucas Jurassic granitoids suggests evolution
along the mantle array with a moderate degree of crustal input, perhaps derived
from the less refractory felsic components of the basement for which 87Sr/86Sr(i)
ratios have yet to be well defined (Cuadros et al. 2014). The San Lucas granitoids
have apparently assimilated significantly less upper crustal material than the
Santander Plutonic Group granitoids. Farther west, 87Sr/86Sr(i) and εNd(t) data for the
Segovia Batholith reveal juvenile values with little indication of assimilation of, or
contamination by, enriched continental crust. Host rocks for the Segovia Batholith
include, to the west, early Paleozoic metasedimentary rocks of the Cajamarca-­
Valdivia island arc assemblage (Restrepo-Pace 1992; Cediel and Cáceres 2000;
Cediel 2011), with quartzo-feldspathic gneisses of the San Lucas complex (González
1999) to the east. In either case contacts with the Segovia Batholith are faulted and
302 H. Leal-Mejía et al.

not well exposed. No Sr-Nd isotope data are available for either of these units.
Notwithstanding, the juvenile 87Sr/86Sr(i) and εNd(t) signatures of the Segovia
Batholith suggest little interaction with continental basement perhaps due to (1)
rapid batholith emplacement in a highly extensional environment and/or (2) the
absence of underlying continental basement in this region (in this context the
San Lucas gneiss may be interpreted as a tectonic float of continental basement
contained between the Otú and Palestina fault zones, as opposed to indicating the
presence of continuous continental basement beneath the Segovia region).
To the south, a similar east-to-west pattern of diminishing crustal input is sug-
gested between the Mocoa porphyry and the Ibagué Batholith. Mocoa is underlain
by Precambrian continental basement of the Garzón Massif (Cediel and Cáceres
2000; Gómez et al. 2015a); however, actual hosts for the porphyritic intrusions ana-
lysed in this study include Jurassic holocrystalline intrusive and coeval volcanic
rocks of the Mocoa-Garzón trend, for which no Sr-Nd isotope analyses are avail-
able. Our Mocoa porphyry samples plot at the base of the mantle array, within the
negative εNd(t) range documented for the Garzón Massif (Fig. 5.10); however, little,
if any, evolution of the Mocoa porphyry with respect to 87Sr/86Sr(i) is suggested.
Based upon available data, it is not possible to ascertain the influence of Garzón
Massif Precambrian basement vs. Jurassic Mocoa-Garzón trend granitoids, in the
Sr-Nd isotope composition of the Mocoa porphyry. To the west, the ca. 188–180 Ma
southern Ibagué Batholith and coeval volcanic rocks, and the ca. 166–152 Ma north-
ern Ibagué Batholith, return more mantelic signatures including mostly positive
εNd(t) values. 87Sr/86Sr(i) ratios for the southern Ibagué granitoids cluster about
0.70500 placing the composite data set along the middle mantle array. Despite their
apparent age difference, the data suggest that similar Sr-Nd isotope systematics can
be inferred for the southern and northern Ibagué batholiths. Both batholiths share a
similar tectonic position along the Palestina fault and suture separating Precambrian
Chicamocha basement from the peri-cratonic domain represented by the Cajamarca-­
Valdivia Terrane (Fig. 5.10). No Sr-Nd isotope data are available for basement rocks
along the Ibagué trend, although the Ibagué and San Lucas batholiths share a similar
structural position along their eastern margin (Fig. 5.10), and values similar to those
recorded for Chicamocha basement in the San Lucas region (Cuadros et al. 2014)
can be inferred. As such, when compared with the Santander, Garzón and San
Lucas granitoids, we interpret the mostly mantelic 87Sr/86Sr(i) and εNd(t) signatures
for the southern and northern Ibagué batholiths to reflect very limited, if any,
crustal assimilation or contamination. Rapid ascent of mantle-derived magmas,
facilitated by extensional reactivation of the preexisting Palestina suture (see Sect.
5.4.2), could result in the mantelic 87Sr/86Sr(i) and εNd(t) signatures recorded along
the Ibagué trend.
Based upon the composite Sr and Nd isotope data, factors controlling the Sr-Nd
isotope composition of the late Triassic-Jurassic granitoids included (1) the Sr-Nd
isotope composition of the magmatic source region, which, in all cases with the
possible exception of the Santander Plutonic Group, was dominated by the
(depleted?) mantle, and (2) the nature and composition of the basement complex
into which the granitoids were emplaced (Fig. 5.10). Undoubtedly, the tectonic and
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 303

structural framework at the time of emplacement was also important. Further

discussion of Sr-Nd results for the Colombian late Triassic-Jurassic granitoids will
be pursued below, following presentation of lead isotope and additional data.

Pb Isotope Geochemistry Results

Figure 5.14b and Appendix 3 present available whole-rock Pb isotope data for the
Colombian late Triassic-Jurassic granitoid suite, including for the Santander
Plutonic Group (Bissig et al. 2014; Van del Lelij et al. 2016) and for the Mocoa
porphyries, the Norosí-San Martín batholiths, Noreán volcanics and Santa Cruz
porphyries and the southern Ibagué Batholith and volcanics (Leal-Mejía 2011).
No Pb isotope analyses for the basement complexes hosting the late Triassic-­
Jurassic granitoids were produced during the present study. For comparative pur-
poses, Fig. 5.14 outlines available Pb isotopic data fields for (1) the Garzón Massif
(Ruiz et  al. 1999), applicable to the Mocoa porphyry intrusions; (2) Paleozoic
metasedimentary basement of the Loja Terrane, Ecuador (Chiaradia et  al. 2004),
which may serve as a proxy for the unknown values of the Cajamarca-Valdivia
assemblage in Colombia, given that regional correlation between the early Paleozoic
metasedimentary sequences of Ecuador and Colombia has been proposed by vari-
ous authors (Restrepo-Pace 1992; Cediel et  al. 2003; Kennan and Pindell 2009;
Spikings et al. 2015); (3) the lead isotope composition of the Piedras amphibolite,
Ecuador, considered by Chiaradia et al. (2004) to represent a Triassic MORB-type
mantle source reservoir which, based upon late Triassic-Jurassic tectonic models for
the Northern Andes (e.g. Spikings et al. 2015; Van de Lelij et al. 2016), may provide
a reasonable estimate for Triassic MORB-type mantle in Colombia; and (4) lead
isotope data for Jurassic arc-related granitoids of Ecuador (Chiaradia et al. 2004),
which may be compared directly with their penecontemporaneous Colombian
counterparts.
With respect to the Colombian granitoids, the most radiogenic values are
observed within the Santander Plutonic Group (206Pb/204Pb  =  19.12–19.44,
207
Pb/204Pb  =  15.70–15.71, 208Pb/204Pb  =  39.20–39.54; Bissig et  al. 2014; Van del
Lelij et al. 2016), which straddle the upper crust lead evolution curve of Zartman
and Doe (1981).
No Pb isotopic data are available for the main-stage batholiths of the Mocoa-­
Garzón trend. Samples of the Mocoa porphyries reveal radiogenic values, plotting
just below the Orogene lead evolution curve (206Pb/204Pb  =  18.14–18.26;
207
Pb/204Pb = 15.57–15.59; 208Pb/204Pb = 38.21–38.29), slightly less radiogenic than
the hosting Grenvillian metamorphic basement rocks of the Garzón Massif (Ruiz
et al. 1999; Fig. 5.14).
Samples from the southern Ibagué Batholith cluster above the Orogene curve
(206Pb/204Pb = 18.72–18.85; 207Pb/204Pb = 15.62–15.63; 208Pb/204Pb = 38.61–38.94)
and return significantly less radiogenic values than those of the Santander Plutonic
Group. The penecontemporaneous southern Ibagué volcanics return less radiogenic
values than the coeval batholith (206Pb/204Pb  =  17.96–18.19; 207Pb/204Pb  =  15.55–
15.56; 208Pb/204Pb = 38.26–38.43). The Norosí and San Martín batholiths of the San
304 H. Leal-Mejía et al.

Lucas region are in turn somewhat less radiogenic than the coeval southern Ibagué
Batholith, clustering to the left and just above the Orogene curve (206Pb/204Pb = 18.35–
18.61; 207Pb/204Pb = 15.60–15.63; 208Pb/204Pb = 37.90–38.45). A similar relationship
is observed between samples from the Norosí and San Martín batholiths and the
coeval Noreán volcanics, with the Noreán volcanics revealing a less radiogenic Pb
isotope composition than that observed for the coeval batholiths. Data for the
Noreán volcanics plot just below the Orogene curve (206Pb/204Pb  =  17.90–17.98;
207
Pb/204Pb = 15.560–15.64; 208Pb/204Pb = 37.48–37.63).
Finally, samples from the Segovia Batholith present a radiogenic composition
clustering between the Orogene and the upper crust curves (206Pb/204Pb  =  18.92–
18.95; 207Pb/204Pb = 15.64–15.67; 208Pb/204Pb = 38.79–38.94) and are located within
the Pb isotope compositional field for the early Paleozoic metasedimentary rocks of
the Loja Terrane (Fig. 5.14), which serve as proxy for the host Cajamarca-Valdivia
basement. Data fall just above the Orogene curve and are somewhat more radio-
genic than those for the southern Ibagué Batholith.

Late Triassic-Jurassic Pb Isotope Summary and Discussion

The overall Pb isotopic composition of the Colombian late Triassic-Jurassic granitoids
is moderately to highly evolved with all 207Pb/204Pb values exceeding 15.55 and
206
Pb/204Pb values extending as high as 19.44. Similar to the Sr-Nd isotope data
presented earlier, whole-rock Pb isotope data for the Colombian late Triassic-­
Jurassic granitoids plot in discrete arrays, on a per-batholith or granitoid suite basis,
with little overlap between the sample sets for individual batholiths (Fig. 5.14b).
With the exception of the Santander Plutonic Group, the composite data form a
linear array of clusters falling along, and at a slightly steeper slope to, the Orogene
lead evolution curve of Zartman and Doe (1981). The granitoids of the Santander
Plutonic Group form an isolated, highly radiogenic array disposed along and essen-
tially parallel to the upper crust lead evolution curve.
Chiaradia et al. (2004) interpret the elongate, sloped array produced on the ura-
nogenic diagram by the Ecuadorian Jurassic intrusions (Fig. 5.14b) to represent the
mixing of magmas derived from a relatively homogenous MORB-type mantle
whose lead isotope composition is approximated by the Triassic Piedras amphibo-
lite (Fig. 5.14b), with crustal Pb derived from the basement units which host the
Jurassic intrusions, in accord with the composite pre-Jurassic continental-oceanic
basement recorded beneath the Ecuadorian Andes (e.g. Litherland et al. 1994).
The variable character of the resulting intrusive lead isotope signatures is consid-
ered to primarily reflect variations in the composition of the basement host rocks
(Chiaradia et al. 2004).
The Colombian late Triassic-Jurassic granitoid lead data plot in discrete clusters
on a per-batholith basis. With the exception of the highly evolved lead isotopic com-
positions of the Santander Plutonic Group, the Colombia data plot coincident with
the lead isotope field for Ecuadorian Jurassic arc-related granitoids as presented
by Chiaradia et al. (2004). As noted, little lead isotope data is available for the
host basement complexes in Colombia, although, as in Ecuador, composite,
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 305

pre-­Mesozoic basement architecture has also been documented in the Colombian

Andes (Restrepo-Pace 1992; Cediel and Cáceres 2000; Cediel et al. 2003; Ordoñez-­
Carmona et al. 2006; Spikings et al. 2015). We propose, as per arguments presented
in Ecuador by Chiaradia et  al. (2004), that the observed variations in Pb isotope
composition for the Ibagué, Norosí, San Martín and Segovia batholiths may be
derived through the mixing of lead from a time-evolved MORB-type mantle source
(approximated by the Piedras amphibolite), including lead derived from the Orogene,
with lead inherited from basement complexes represented by the Chicamocha
Terrane and the Cajamarca-Valdivia Terrane. With respect to samples from the
Santander Plutonic Group, individual lead analyses plot in a linear form, essentially
parallel to the upper crust lead evolution curve of Zartman and Doe (1981), suggest-
ing that the lead isotope composition of the Santander granitoids was essentially
derived from the combined Proterozoic and early Paleozoic continental basement of
the Santander Massif (westernmost Guiana Shield and Grenvillian granulite belt),
with little or no contribution of MORB-type mantle or Orogene lead. As recorded by
the Sr-Nd isotope data, the Pb isotope composition of late Triassic-Jurassic granit-
oids in the Colombian Andes reflects the east-to-west changes in the composition of
the intruded basement units as shown in Fig. 5.10.

Synthesis and Conclusions of Lithogeochemistry and Sr, Nd and Pb Isotope

Geochemistry for Late Triassic-Jurassic Granitoids

Lithogeochemical data and Sr-Nd and Pb isotope systematics combined with U-Pb
(zircon) age dating for the Colombian late Triassic-Jurassic batholiths reveal clear
temporal-spatial trends and permit consistent qualitative conclusions with respect to
magmatic sources and evolution and the degree of contamination through crustal
anatexis or assimilation with host basement units. Lithogeochemical data indicate
that all of the late Triassic-Jurassic batholiths are of the Cordilleran (Frost et  al.
2001), volcanic arc (Pearce et  al. 1984) or calc-alkaline (Barbarin 1999) types,
typical of transitional (Barbarin 1999; in the case of the Santander Plutonic Group)
and subduction-related tectonic settings. In northern Colombia, U-Pb (zircon) age
dates demonstrate westward migration of the axis of magmatism from the ca. 209–
196 Ma Santander Plutonic Group into the ca. 189–182 Ma main-phase batholiths
of the Serranía de San Lucas and subsequently into the ca. 168–155 Ma Segovia
Batholith. Lithogeochemical and Sr-Nd isotope data document diminishing crustal
contamination and increasingly more juvenile melt compositions progressing from
east to west. Data support an upper mantle source region and the variable mixing of
mantle and crustal contributions for all batholiths, with the exception of the
Santander Plutonic Group for which significant degrees of melt contamination
through crustal anatexis and/or assimilation can be inferred. This is supported by
the findings of Van der Lelij (2013), who, based upon Lu-Hf and Sr isotope data,
concluded that Paleozoic and Mesozoic granitoids emplaced in the Santander
Massif and Merida Andes between ca. 472 and 196  Ma were primarily derived
through the recycling of Precambrian basement including lower to upper crustal
306 H. Leal-Mejía et al.

sources with limited, if any, juvenile input from the depleted mantle. A similar
conclusion can be drawn from the Pb isotope data for the Santander Plutonic Group
which suggest in situ derivation and evolution of Pb with little contribution from
Orogene or MORB-type mantle sources.
In the south, U-Pb (zircon) age dates for the southern Ibagué, Mocoa-Garzón
trend and northern Ibagué batholiths suggest south and minor eastward migration of
magmatism from the southern Ibagué to Mocoa-Garzón batholiths and subsequently
along trend to the NNE into the northern Ibagué Batholith. Lithogeochemical and
Sr-Nd and Pb isotope data reveal similar, predominantly upper mantle-derived com-
positions for the southern and northern Ibagué batholiths, despite their differences
in age. The data suggest granitoid generation from a similar magmatic source region
and emplacement under like tectonic conditions, in either case facilitated by the
suture contact between the Chicamocha and Cajamarca-Valdivia units, which lim-
ited interaction between granitoid magmas and either basement domain. Sr-Nd and
Pb isotope data are lacking for the Jurassic granitoids of the Mocoa-Garzón trend,
but geological and lithogeochemical data infer greater degrees of magma interac-
tion with the hosting Grenvillian metamorphic rocks of the western Guiana Shield
as widely exposed in the Garzón Massif (Kroonenberg 1982; Ibañez-Mejía et al.
2011; Gómez et al. 2015a), although apparently not to the same degree as observed
in the Santander Plutonic Group.
In conclusion, individual late Triassic to Jurassic granitoid batholiths of the
Colombian Andes represent temporally and spatially separate arc segments, intruded
into geologically distinct basement complexes. U-Pb (zircon) age, lithogeochemical
and Sr-Nd and Pb isotope data suggest that granitoid chemical and isotopic charac-
teristics and evolution are essentially independent of age and were primarily deter-
mined by processes within the magmatic source region for the granitoid melts and
by the composition and/or degree of interaction with the hosting basement complex.
Data for the individual batholiths reflect the spatial migration of late Triassic to
Jurassic magmatism, combined with the unique geological conditions encountered
by each granitoid arc segment at the time of emplacement. An overview and inter-
pretation of the structural framework and tectonic evolution at the time of emplace-
ment of the Colombian late Triassic to Jurassic granitoids are presented in the
magmato-tectonic synthesis contained in Sect. 5.4.2.

5.3.4  C
 retaceous to Eocene Granitoid Magmatism:
Distribution, Age and Nature

Volumetrically significant granitoids of Cretaceous to Eocene age comprise much

of the northwesternmost segment of the Colombian Andes, within the northern
Central Cordillera and within the Chocó Arc segment of the Western Cordillera.
Based upon geological exposure throughout the Colombian Andes, over 80 percent
of Colombian Cretaceous to Eocene granitoid magmatism is concentrated within
two composite intrusions and their satellite plutons, including the Antioquian and
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 307

Sonsón batholiths. Of these, the Antioquian Batholith (Feininger and Botero 1982)
and its satellites are by far the largest, occupying an exposed area exceeding some
7800 square kilometres, more than the combined area of all the remaining Colombian
Cretaceous to Eocene granitoids. The remaining granitoids, although volumetri-
cally less significant, provide important information regarding the tectonic history
of the region during the Cretaceous-Eocene.

5.3.4.1  Distribution

The distribution of major Colombian Cretaceous to Eocene granitoids including

their associated volcanic sequences, where present, is shown in Fig.  5.15. Based
upon geographic distribution and tectonic history, two broad groups of Cretaceous-­
Eocene granitoids can be recognized in Colombia, (1) an Eastern Group of autoch-
thonous continental affinity, intruding the Cajamarca-Valdivia metamorphic
basement complex, which was in situ within the Northern Andean tectonic mosaic
at the time of pluton emplacement (i.e. prior to the early-mid-Cretaceous), and (2) a
Western Group, including allochthonous granitoids of peri-cratonic or intra-oceanic
affinity, hosted within accreted oceanic volcanic and sedimentary rocks of the
Farallon Plate and CCOP/CLIP assemblage, presently underlying the cordilleran
regions and coastal plains along the Colombian Pacific to the west of the Cauca and
Garrapatas-Dabeiba fault and suture systems (Fig. 5.15).
The Eastern Group includes the late Cretaceous to Paleocene Antioquian
Batholith and its satellite plutons (Ovejas Batholith and Altavista, La Unión and La
Culebra stocks), the Paleocene Sonsón Batholith and other smaller Paleocene to
Eocene intrusives such as the El Bosque Batholith and the Mariquita, Manizales, El
Hatillo and Santa Bárbara stocks. The Santa Marta Batholith and Latal, Toribio and
Buritáca plutons, located on the leading apex of the Sierra Nevada de Santa Marta
(Tschanz et al. 1974; Mejía et al. 2008; Duque 2009; Cardona et al. 2011), are also
included within this group. Notable features of many of the Eastern Group plutons,
when compared to their Jurassic counterparts, include their generally sub-equant
shapes (length-to-width ratios mostly less than 2:1) and the lack (or lack of preser-
vation) of a coeval volcanic pile.
Within the Western Group, the Santa Fé, Sabanalarga and Buga batholiths and
the Mistrató and other minor plutons (e.g. Jejenes Stock) are hosted within
Cretaceous oceanic rocks of the Dagua and Cañas Gordas terrane assemblages. The
Western Group granitoids may be considered to form components of the CCOP/
CLIP assemblage, as discussed by Kerr et al. (1997) and Sinton et al. (1998). There
is little published geological or geochemical information regarding some of these
intrusions in Colombia, and in some cases precise radiometric age dates and litho-
geochemical information have only recently been obtained (e.g. Buga, Villagómez
et al. 2011; Santa Fé, Weber et al. 2015; Jejenes, Leal-Mejía 2011). An initial under-
standing of the origin and nature of these plutons and their relationship with their
host rocks is herein presented.
308 H. Leal-Mejía et al.

Fig. 5.15  Distribution of mid-Cretaceous to Eocene granitoids in the Colombian Andes. Principle
modern-day physiographic provinces of the region are shown for reference. (Granitoid shapes
modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al. 2015a)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 309

Farther to the west, the Paleocene-Eocene Mandé-Acandí batholiths, including

the coeval La Equis-Santa Cecilia Formation volcanic and pyroclastic rocks
(Fig. 5.15), are the most significant expression of granitoid magmatism within the
Western Group. These granitoids were generated in an intra-oceanic setting upon
late Cretaceous oceanic crust which forms the basement of the western segment of
the Chocó Arc (Montes et al. 2012, 2015).

5.3.4.2  Age Constraints on Cretaceous-Eocene Granitoid Magmatism

Recent U-Pb (zircon) age determinations for Cretaceous to Eocene granitoids in the
Colombian Andes, including intrusions from both the Eastern and Western groups,
have been conducted by various authors. Results are included for works dedicated
to the Antioquian Batholith and its surroundings (Correa et al. 2006; Ibañez-Mejía
et  al. 2007; Ordoñez-Carmona et  al. 2007a; Restrepo-Moreno et  al. 2007; Leal-­
Mejía 2011; Villagómez et al. 2011); the Sonsón Batholith (Ordoñez et al. 2001;
Leal-Mejía 2011); the Mariquita Stock (Leal-Mejía 2011); the Manizales, El Hatillo
and El Bosque plutons (Bayona et  al. 2012; Bustamante et  al. 2017); the Santa
Marta Batholith (Mejía et al. 2008; Duque 2009; Cardona et al. 2011); the Buga
Batholith (Villagómez et al. 2011); the Jejenes and Irra stocks (Leal-Mejía 2011);
and the Mandé-Acandí batholiths (Leal-Mejía 2011; Wegner et al. 2011; Montes
et al. 2012).
In total, the above data set represents over one hundred eighty-five high-­precision
U-Pb (zircon) magmatic crystallization ages which can be used to model Cretaceous
to Paleogene magmatism in Colombia. The composite data are displayed in histo-
gram format in Fig.  5.16. In many cases, the new data represent the first well-­
constrained age dates when compared to the historic largely K-Ar-based database of
Maya (1992). In other cases, the data permit a much better definition of the multiple
magmatic pulses which comprise large and complex intrusions, such as the
Antioquian Batholith.
Within the Eastern Group of granitoids, the oldest pluton is the volumetrically
minor Mariquita Stock, which produced a U-Pb (zircon) age of ca. 93.5 Ma (Leal-­
Mejía 2011). Large-scale, volumetrically significant and continuous plutonism
begins in the mid-Cretaceous with the Antioquian Batholith, including its satellite
plutons, between ca. 96 and 72 Ma. This event extends into lesser Paleocene and
Eocene magmatism at ca. 62–54 Ma, recorded in the Antioquian and Sonsón batho-
liths and Manizales, El Bosque, El Hatillo, Santa Barbara intrusions and other minor
plutons to the south.
The Antioquian Batholith is a composite poly-phase pluton emplaced in at least
four pulses, spanning the late Cretaceous to Paleocene (Fig. 5.16). The earliest ca.
96–92  Ma phase is associated with more mafic to intermediate magmatism as
­recognized in the Altavista and San Diego stocks (Correa et al. 2006) and the mafic-­
intermediate xenoliths commonly embedded within the younger felsic, main-phase
members of the batholith. Volumetrically, two distinct phaneritic-equigranular
310 H. Leal-Mejía et al.

Fig. 5.16  U-Pb (zircon) age date populations for mid-Cretaceous through Eocene granitoids in the
Colombian Andes. Note penecontemporaneous ages for the Western Group, allochthonous oceanic
suite vs. the Eastern Group, autochthonous continental suite, representing the coeval emplacement
of granitoids in distinct geotectonic environments

tonalitic to granodioritic pulses, from ca. 89 to 82 Ma and from ca. 81 to 72 Ma,
account for the majority (>90%?) of the main mass of the Antioquian Batholith and
satellite stocks. The Culebra Stock near Segovia returned an age of ca. 87.5 Ma.
Granodiorite porphyry dikes extending to the NE into the Segovia area returned an
age of ca. 86 Ma. The Ovejas Batholith returned ages ranging from ca. 76 to 72 Ma
(Restrepo-Moreno et al. 2007), whilst the La Unión Stock to the south returned ca.
73.5 Ma with inheritance from ca. 82.8 Ma (Leal-Mejía 2011).
Minor Paleocene granitoid magmatism is also recorded in isolated areas within
the Antioquian Batholith domain. The Caracolí Stock on the east-centre margin of
the batholith returned an age of ca. 60  Ma, with inheritance from ca. 79  Ma.
Medium-grained equigranular tonalite from near Providencia in the Nus River val-
ley returned various dates ranging from ca. 60 to 58 Ma, whilst a medium- to coarse-­
grained quartz biotite granite porphyry stock containing distinctive euhedral
bipyramidal quartz crystals, located west of Santo Domingo, revealed an age of ca.
60 Ma (Leal-Mejía 2011).
Volumetrically more significant Paleocene magmatism within the Eastern Group is
documented in the Sonsón Batholith. This granitoid pluton was formerly considered
to be of Jurassic age (Cediel and Cáceres 2000; González 2001; Gómez et al. 2007);
however, U-Pb (zircon) age dating reveals it is a composite body, comprised of gran-
itoid rocks of Permo-Triassic age in the south (Leal-Mejía 2011; Fig. 5.15) and of
Paleocene age (ca. 61–57 Ma) in the north (Ordoñez et al. 2001; Leal-Mejía 2011).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 311

The Sonsón Batholith is presently interpreted to include the northern sector extending
around and to the east of the town of Sonsón (Leal-Mejía 2011; Fig. 5.15). The precise
contact between these two ages of intrusive has yet to be cartographically defined.
Additional, recent Paleocene U-Pb (zircon) dates have also been reported for the
Eastern Group Manizales Stock (ca. 59  Ma, Bayona et  al. 2012), indicating that
autochthonous granitoid magmatism continued to the south of Sonsón during this
time period. A general southward and eastward younging trend for magmatism can
be inferred to continue into the Eocene with the emplacement of the El Hatillo
Stock at ca. 55 Ma (Bayona et al. 2012; Bustamante et al. 2017) and the presence of
additional granitoid plutons, including the El Bosque Batholith which also provides
a U-Pb (zircon) age of ca. 55 Ma (Bustamante et al. 2017)
Finally, within the Eastern Group plutons of northernmost Colombia, Paleocene to
Eocene granitoid magmatism spanning the age range from ca. 64 to 47 Ma (Mejía
et al. 2008; Duque 2009; Cardona et al. 2011) is recorded along the apex of the Sierra
Nevada de Santa Marta (Tschanz et  al. 1974). Detailed study of the Santa Marta
Batholith and satellite plutons including the Latal, Toribio and Buritaca stocks, by
Duque (2009), revealed emplacement of the suite in two pulses between ca. 58 and
50 Ma. The principal components of the SW Santa Marta Batholith proper and Latal
pluton, including distinctive coarse- and fine-grained phases, were intruded between
ca. 58 and 55  Ma, followed by emplacement of the NE sector of the Santa Marta
Batholith and the Buritaca and Toribio stocks, by ca. 50 Ma. Based upon the compos-
ite U-Pb (zircon) data, Duque (2009) interprets a general NE migration of magmatism
within the main-phase Santa Marta Batholith, terminating in the Buritaca pluton.
Additional, early, volumetrically minor, ca. 64–62 Ma, two-­mica trondhjemitic leuco-
granites, identified by the author (e.g. Playa Salguero), were considered unrelated to
main phase batholith emplacement (see Duque-Trujillo et al. 2018).
With respect to the Cretaceous to Eocene Western Group (CCOP/CLIP) granit-
oids located to the west of the Cauca and Garrapatas-Dabeiba fault and suture sys-
tem (Fig. 5.15), the oldest of these plutons, hosted within Cañas Gordas oceanic
basement, include the Buriticá tonalite and associated Santa Fé Batholith, which
have returned U-Pb (zircon) dates of ca. 100 Ma and 90 Ma, respectively (Weber
et al. 2015). The Sabanalarga Batholith, located in fault contact immediately to the
east (Nívia and Gómez 2005; Gómez et al. 2007), has not been dated using the U-Pb
technique but appears to represent a tectonically duplicated segment of the Santa Fé
Batholith. Further detailed mapping and age dating are required to better define the
relationships between these intrusions and the host basement complex.
Along the trend to the south of Santa Fé and Sabanalarga, the Mistrató Batholith
(Fig.  5.15) also appears within strongly tectonized Cañas Gordas volcano-­
sedimentary rocks, in intrusive/structural contact with the Barroso Fm. An ca.
85 Ma U-Pb (zircon) age was presented for the Mistrató Batholith by the Agencia
Nacional de Hidrocarburos and Universidad de Caldas (2011). A similar age was
recorded, farther south, in the southern sector of the Western Cordillera, where the
previously undated Jejenes Stock returned a U-Pb age of ca. 84 Ma (Leal-Mejía
2011). In this case, intrusive relationships with the CCOP-related Dagua terrane
are observed.
312 H. Leal-Mejía et al.

To the NNE of the Jejénes Stock, the Buga Batholith (Fig. 5.12) has returned a
U-Pb (zircon) age of ca. 92 Ma (Villagómez et al. 2011). Buga appears to have
been emplaced with pre-CCOP basement rocks of the Dagua terrane (Anaime Fm;
Nívia 1992). Both the western and eastern margins of the batholith have been
tectonically modified.
The youngest and by far largest intrusion of the Western Group allochonous
granitoids is the Mandé-Acandí Batholith (Fig. 5.15), hosted within the El Paso-­
Baudo assemblage of northwesternmost Colombia (Cediel et al. 2010). Field obser-
vations and regional magnetic data (Cediel et al. 2010) indicate the Mandé Batholith
is a composite body comprised of holocrystalline phaneritic and porphyritic phases
ranging from diorite to granodiorite and granite. It is flanked to the east and west by
the penecontemporaneous Santa Cecilia-La Equis volcanic sequence, of Paleogene
age (Cediel et al. 2010). A thermal aureole is recorded within the volcanic sequence
indicating the Mandé Batholith intrudes the volcanic pile. Leal-Mejía (2011) pro-
vided U-Pb (zircon) dates of ca. 46–44 Ma for quartz diorite porphyry which cuts
phaneritic granodiorite within the north central sector of the batholith at Pantanos.
An ca. 62 Ma (Paleocene) inheritance age, interpreted to have been donated by the
volcanic pile or main batholith, was observed for these samples. Within the northern
extension of the Mandé magmatic arc, including the Acandí Batholith in Panama’s
San Blas Range, Paleocene-Eocene U-Pb (zircon) magmatic crystallization ages are
also observed. In Colombia, Montes et al. (2012) and Montes et al. (2015) record a
maximum age of ca. 50 Ma for the Acandí Batholith.

Discussion of Spatial Distribution of Cretaceous-Eocene Granitoids

Based upon geographic distribution and geological setting, two major groups of
Colombian Cretaceous to Eocene granitoids have been identified, including 1)
Eastern Group granitoids and 2) Western Group granitoids. The Eastern Group rep-
resents autochthonous, continental granitoid magmatism of Cretaceous to Eocene
age, largely dominated by two major magmatic pulses at ca. 89–82  Ma and ca.
79–72 Ma, generating the main mass of the Antioquian Batholith, its satellite plu-
tons and the Irra Stock (ca. 70 Ma). Magmatism is rather abruptly shut down after
ca. 72 Ma but reinitiates at ca. 62–58 Ma, within and to the south of the Antioquian
Batholith, with the emplacement of various smaller plutons, the largest of which is
the Sonsón Batholith. The available U-Pb age data demonstrate the general south-
ward and eastward migration of autochthonous magmatic centres of the Eastern
Group during post main-phase Antioquian Batholith time, from the Paleocene to the
early Eocene. The Paleocene-Eocene granitoid centres can be traced from the 61 to
58 Ma Sonsón and Manizales intrusives in the north to the El Hatillo, El Bosque and
Santa Bárbara plutons to the south and east, all of which produce Paleocene-Eocene
U-Pb (Bayona et al. 2012; Bustamante et al. 2017) and/or K-Ar (Maya 1992) radio-
metric age dates.
The Western Group (CCOP/CLIP-related) granitoids may also be considered in
two spatially and temporally separate groups, including an early group (Sabanalarga/
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 313

Santa Fé, Buriticá, Jejénes, Buga, etc.) dating from ca. 100 to 82 Ma, hosted within
the Dagua-Cañas Gordas terranes, and the significantly younger ca. 50–42 Ma gran-
itoids of the Mandé-Acandí arc, hosted within the El Paso-Baudó terrane.
Emplacement of the early group is essentially penecontemporaneous with the
development of the early phases of the continental Antioquian Batholith. The
Western Group granitoids, however, are consistently hosted within oceanic terrane
assemblages which have been deemed to be allochthonous (e.g. Cediel et al. 2003;
Kerr et al. 2003; Kennan and Pindell 2009) and are considered to represent granitoid
magmatism in an intra-oceanic environment, related to the generation and migration
of the CCOP/CLIP assemblage, prior to accretion along the Colombian margin.
Further temporal and spatial differentiation of the Cretaceous to Eocene granitoids
of the Eastern and Western groups, within the context of the tectonic evolution of
the region, will be discussed in detail following the presentation of lithogeochemi-
cal and isotopic data in the following section.

5.3.4.3  L
 ithogeochemical and Isotopic Characteristics of Cretaceous
to Eocene Granitoids

Historically, little whole-rock lithogeochemical or isotopic data has been available

for the Colombian Cretaceous to Eocene granitoid suite. Older works or compila-
tions such as Alvarez (1983), Feininger and Botero (1982) or González (2001) con-
tain some basic major element oxide data but include only limited or no minor, trace
and rare-earth element data and virtually no Pb, Sr or Nd isotopic data, thus limiting
the interpretation of petrogenetic and tectonic constraints for these rocks.
Recently, with the use of ICP-based analytical techniques, studies applying com-
bined whole-rock major-minor-trace-rare-earth element studies, and additional iso-
topic analyses, to the Cretaceous-Eocene granitoid suite, have become available.
Important contributions which analyse multiple plutons at a regional scale include
Villagómez et al. (2011) and Leal-Mejía (2011). Additional studies involving spe-
cific intrusions include Ordoñez et al. (2001) for the Sonsón Batholith; Correa et al.
(2006), Ibañez-Mejía et al. (2007), Ordoñez-Carmona et al. (2007a) and Restrepo-­
Moreno et al. (2007) for the Antioquian Batholith; Wegner et al. (2011) and Montes
et  al. (2012) for the Mandé-Acandí batholiths; and Bayona et  al. (2012) and
Bustamante et  al. (2017) for the Manizales, El Hatillo and El Bosque plutons.
Representative lithogeochemical data for the Cretaceous-Eocene Eastern Group and
Western Group magmatic suite is shown in Figs. 5.17, 5.18, 5.19, and 5.20.

Lithogeochemistry

Eastern Group Granitoids

The Antioquian Batholith: The ca. 96–58 Ma Antioquian Batholith granitoid suite,
including satellite bodies (e.g. the Ovejas Batholith and the Altavista, La Unión and
La Culebra stocks) and the coeval Segovia dikes, is represented by 57 samples, 14
314 H. Leal-Mejía et al.

of which are altered. All of the magmatic phases show similar broad-scale lithogeo-
chemical features such as a metaluminous nature, within a highly differentiated
calc-alkaline compositional trend, which varies over time, from gabbro to granite.
With respect to the classification scheme of Frost et  al. (2001), the Antioquian
Batholith suite demonstrates a weakly ferroan trending to magnesian composition
with decreasing age, whilst most samples demonstrate a distinctly calcic tendency.
Trace element spider diagram patterns for the Antioquian Batholith granitoids show
magmatic arc-related signatures, with enrichment of HFSE with respect to LILE
and conspicuous negative Ta-Nb anomalies (Fig.  5.18). The REE patterns show
highly variable REE contents (ΣREE = 21.82–335.69) and also variable negative to
positive Eu anomalies (Eu/Eu* = 0.50–2.85) (Fig. 5.18).
The ca. 62–58 Ma Providencia granitoid suite, which may be considered post
main-phase batholith in age, is characterized by lower-K biotite-bearing granodio-
rite to granite with compositionally distinct (high-Na) plagioclase and “adakite-­
like” geochemical features (Richards and Kerrich 2007; e.g. high SiO2 (≥56 wt%),
Al2O3 (≥15 wt%) and Na2O (≥3.5 wt%) contents, low K2O (≤3 wt%) contents and
Sr enrichment (≥400  ppm), accompanied by depletion of Y (≤18  ppm) and Yb
(≤1.9 ppm)). Providencia suite REE trends are slightly depleted with respect to the
main phases of the batholiths (ΣREE = 47–160.87). They describe gently decreas-
ing slopes and no significant Eu anomaly.
The Irra Stock: Major, minor and trace element data for the ca. 70  Ma Irra
Stock (Figs.  5.17 and 5.18) reveal characteristics which distinguish it from the
main phases of the Antioquian Batholith. It is a metaluminous syenite of the sho-
shonite series (alkali), and it is enriched in both trace and rare-earth elements;
however the mantle-normalized plot displays positive Ba and Sr anomalies and
negative Nb, Ta, P and Ti anomalies similar to arc-related rocks. REE contents are
relatively enriched (ΣREE  =  99.16–216.5  ppm). REE plots reveal moderately
fractionated chondrite-­normalized patterns (La/Yb)N = 18.28–24.00). No signifi-
cant Eu anomaly is observed. The Irra Stock is located within the Romeral tectonic
zone and is considered to form part of an in situ phase of minor alkaline magma-
tism, similar to the Sucre intrusive suite located in a similar tectonic position within
Romeral to the north (Vinasco 2018). Both lithogeochemical and age data for the
Irra Stock are contrary to data observed for the low-K Western Group granitoids
(see below).
The Sonsón Batholith: Samples from different phases of the Sonsón Batholith
including phaneritic quartz-diorites, leucogranites and diorite porphyry dikes are
represented by five relatively unaltered samples and one leucogranite sample with
evidences of alteration (Fig.  5.17). The samples are metaluminous in nature (A/
CNK <1.1) of medium- to high-K calc-alkaline affinity, although the leucogranites
plot marginally peraluminous due to the partial metasomatic replacement of biotite
by muscovite. An arc-magmatism signature for the Sonsón Batholith samples is
revealed by the trace element spider diagram (Fig.  5.18), where higher U (12.1–
16.1  ppm) and Th (21.3–24.2  ppm) contents and positive incompatible element
anomalies (e.g., K, Th, U and Ta), accompanied by strong negative compatible
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 315

Fig. 5.17  Major element lithogeochemical plots for mid-Cretaceous through Paleocene, Eastern
Group (continental) granitoids in the Colombian Andes. (a) AFM plot, curve after Irvine and
Baragar (1971); (b) K2O vs. SiO2 plot, boundary fields in grey as summarized by Rickwood
(1989); (c) alumina saturation plot after Barton and Young (2002); (d and e) MALI and Fe-index
vs. SiO2 plots, respectively, after Frost et al. (2001); (f) R1-R2 classification plot after De La Roche
et al. (1980). Th tholeiite, C-A calc-alkaline, Sh shoshonite, Gb No gabbro-norite, Ol-Gb (olivine-)
gabbro, Gb Di gabbro-diorite, Di diorite, Mz Di monzodiorite, To tonalite, Gd granodiorite, Gr
granite Alk Gr alkali granite, Sy syenite
316 H. Leal-Mejía et al.

Fig. 5.18  Trace element and REE lithogeochemical plots for mid-Cretaceous through Paleocene,
Eastern Group (continental) granitoids in the Colombian Andes. (a and b) Trace Element and REE
normalized spider diagram plots; (c) granite discrimination Ta vs. Yb diagram after Pearce et al.
(1984). VAG volcanic-arc granites, syn-COLG syn-collisional granites, WPG within-plate granites,
ORG ocean ridge granites

element anomalies (e.g., Ba, Sr, Zr and Ti) for the leucogranites confirm their
more evolved character with respect to the quartz-diorite rocks. The REE patterns
indicate higher REE contents in the allanite-bearing quartz-diorite rocks
((ΣREE  =  127.94–142.79  ppm) with respect to the leucogranite rocks
(ΣREE = 68.56–89.48 ppm). All of the Sonsón samples show moderately fraction-
ated patterns with gentle decreasing slopes. The quartz-diorite reveals a slightly
steeper overall slope ((La/Yb)N = 10.84–11.51), with moderate negative Eu anom-
aly (Eu/Eu* = 0.54–0.72). By comparison, the leucogranite reveals stronger nega-
tive Eu anomalies (Eu/Eu*  =  0.22–0.34) and relative depletion of the heavy
rare-earth elements (HREE; La-Sm). Relatively flat light rare-earth elements
(LREE) patterns are observed for all samples from the Sonsón Batholith suite ((Gd/
Yb)N  =  0.89–1.57; Leal-Mejía 2011). Although of similar age, none of the rock
types of the Sonsón Batholith show the Na-rich “adakite-like” geochemical signa-
ture observed in the Providencia suite of the Antioquian Batholith.
The Mariquita Stock: The Mariquita Stock is a metaluminous, medium-K calc-­
alkaline granodiorite. It plots magnesian-calcic in the classification scheme of Frost
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 317

et al. (2001). Trace element and REE volcanic reveal arc signatures with moderate
REE contents (ΣREE  =  80.82  ppm), decreasing slopes ((La/Yb)N  =  3.28) and a
moderate negative Eu anomaly (Eu/Eu* = 0.55; Fig. 5.18).
Other Paleogene Eastern Group Granitoids: Other samples of Eastern Group
granitoids (Fig. 5.17) include twenty-one samples of the main phase of the Santa
Marta Batholith (Mejía et al. 2008; Duque 2009), one sample of the Manizales
Stock (Leal-Mejía 2011), three samples of the El Bosque Batholith (Leal-Mejía
2011; Bustamante et  al. 2017) and twelve samples of the El Hatillo Stock
(Bustamante et  al. 2017). Samples of the Manizales Stock and the El Bosque
Batholith appear altered, whilst those of the Santa Marta Batholith and the El
Hatillo Stock do not. In general, samples from these plutons exhibit metalumi-
nous to weakly peraluminous, medium- to marginally high-K calc-alkaline char-
acter, with SiO2 content ranging from 57 to 71 wt%. A general tendency towards
an increasingly magnesian calc-alkalic character, and increasing aluminity and
silica content, with decreasing age is observed. A notable exception to this trend
is the Eocene granitoids of the Sierra Nevada de Santa Marta, which were
emplaced under differing tectonic conditions when compared with the granitoids
of similar age located in the Central Cordillera. With respect to REE contents,
ΣREE for samples from the Manizales Stock and El Bosque Batholith are
107.67 ppm and 112.36–169.53 ppm, respectively, whereas the El Hatillo Stock
shows lower values (ΣREE = 85.18–129.54 ppm). The Manizales Stock exhibits
a fractionated REE pattern ((La/ Yb)N = 9.6) with very subtle negative Eu anomaly
(Eu/Eu* = 0.89) and a relatively flat HREE trend ((Gd/Yb)N = 1.27). The El Bosque
Batholith shows a similar decreasing slope for light rare-earth elements (La-Sm) as
in the Manizales Stock, and no significant Eu anomaly (Eu/Eu* = 0.73–0.98).
It records strong depletion of the HREE (Gd-Lu) where a concave upward (spoon-
shaped) pattern for the HREE is observed, which may be related to hornblende
fractionation in the magma source.

Western Group Granitoids

Lithogeochemical plots for the Western Group of Cretaceous to Eocene granitoids
are presented in Figs. 5.19 and 5.20. Some 72 samples in total are represented.
In contrast to the Eastern Group, 55% (40) of the samples from the Western Group
are notably affected by some degree of post-emplacement alteration. This alteration
is a reflection of the observably tectonized nature of most of the intrusions as viewed
in outcrop, where abundant faulting and jointing, coupled with high rainfall condi-
tions, have led to chemical weathering and deep oxidation. Thus, in many respects, the
quality of the data set is not on par with that from the Eastern Group. Notwithstanding,
the observations regarding the general lithogeochemical characteristics of the Western
Group granitoids revealed herein are considered valid.
All Western Group granitoids demonstrate a low-K character (K2O content
<1%). As a reflection of this observation, Western Group granitoids are mixed
magnesian-­ferroan in nature, with a distinctly calcic tendency. Leal-Mejía (2011)
observed the low-K behaviour for dioritic phases of the Buga Batholith and Jejenes
318 H. Leal-Mejía et al.

Fig. 5.19  Major element lithogeochemical plots for mid-Cretaceous through Eocene, Western
Group (oceanic) granitoids in the Colombian Andes. (a) AFM plot, curve after Irvine and Baragar
(1971); (b) K2O vs. SiO2 plot, boundary fields in grey as summarized by Rickwood (1989); (c) alu-
mina saturation plot after Barton and Young (2002); (d and e) MALI and Fe-index vs. SiO2 plots,
respectively, after Frost et al. (2001); (f) R1-R2 classification plot after De La Roche et al. (1980). Th
tholeiite, C-A calc-alkaline, Sh shoshonite, Gb No gabbro-norite, Ol-Gb (olivine-)gabbro, Gb Di
gabbro-diorite, Di diorite, Mz Di monzodiorite, Mz monzonite, To tonalite, Gd granodiorite
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 319

Fig. 5.20  Trace element and REE lithogeochemical plots for mid-Cretaceous through Eocene
Western Group (oceanic) granitoids in the Colombian Andes. (a and b) Trace element and REE
normalized spider diagram plots; (c) granite discrimination Ta vs. Yb diagram after Pearce et al.
(1984). VAG volcanic-arc granites, syn-COLG syn-collisional granites, WPG within-plate granites,
ORG ocean ridge granites

Stock, as well as for the Mandé Batholith, and noted that samples from the Jejenes
Stock and Mistrató, Buga and Mandé batholiths samples plot in the tholeiite field of
the Peccerillo and Taylor (1976) diagram. Exceptions to this general trend include
some dioritic to granodioritic samples of the Buga and Mandé batholiths which
returned K2O contents >2% (Leal- Mejía 2011).
Buga, Santa Fé and Mistrató batholiths and Jejenes Stock: The late Cretaceous
Buga, Santa Fé (including Sabanalarga) and Mistrató batholiths and Buriticá and
Jejenes Stock are represented by 65 lithogeochemical samples. Of these samples,
over 54% (35 samples) present evidence of hydrothermal alteration. Notwithstanding,
the lithogeochemical analyses establish the metaluminous character and variable
low-K tholeiitic to high-K calc-alkaline affinity of the suite (Fig. 5.19). The Jejénes
Stock sample plots in the peraluminous field, likely as a result of petrographically
observable hydrothermal alteration (saussuritization) in the analysed sample (Leal-­
Mejía 2011). Samples of the Buga, Santa Fé (Sabanalarga) and Mistrató batholiths
show bimodal compositions with more melanocratic samples of gabbroic to dioritic
320 H. Leal-Mejía et al.

composition and more leucocratic samples of granodioritic or tonalitic composition.

The trace element spider diagram patterns of unaltered samples exhibit enrichment
of HFSE with respect to LILE and negative Ta-Nb and Ti anomalies, typical of
arc-­magmatism (Fig. 5.20). The Santa Fé (Sabanalarga) Batholith and the Buriticá
Stock return REE contents ranging from ΣREE  =  3.66–130.14  ppm, with
chondrite-­normalized patterns exhibiting moderate to steep decreasing slopes
((La/Yb)N = 0.77–12.88) and slightly negative to slightly positive Eu anomalies
(Eu/Eu* = 0.74–1.40). The Buga Batholith shows variable behaviour as implied
by observed compositional variations. Data presented by Leal-Mejía (2011) and
Villagómez et  al. (2011) indicate the more mafic phases contain variable REE
contents (ΣREE = 15.42–41 ppm) with flat chondrite-normalized patterns approx-
imating ten times chondrite value ((La/Yb)N = 0.8–2.5). Other phases show higher
REE contents (ΣREE  =  229–241  ppm) and exhibit more fractionated patterns
((La/Yb)N = 21–33), with no significant Eu anomalies (Eu/Eu* = 0.95–0.98).
A third phase, with intermediate REE contents (ΣREE = 104.48 ppm), moderate
fractionation ((La/Yb)N  =  2.5) and a negative Eu anomaly (Eu/Eu*  =  0.52) are
observed. Mistrató Batholith samples show lower REE contents (ΣREE  =  2.6–
42.5 ppm) with a variable flat to decreasing slope patterns ((La/Yb)N = 0.71–10) and
subtle to strongly positive Eu anomaly (Eu/Eu*  =  1.02–2.61) when compared to
Buga Batholith samples. Felsic phases of the Mistrató Batholith reveal a moderate
decreasing slope fractionated chondrite-normalized pattern ((La/Yb)N = 10) and no
Eu anomaly (Eu/Eu* = 0.99), whilst diorites of the same intrusive show relatively
flat REE patterns ((La/Yb)N = 0.7–1.4).
The Jejénes Pluton is represented by a single sample, a saussuritized low-K
tonalite which reveals relatively flat arc-related trace element patterns similar to
the Buga and Mistrató batholiths. It contains relatively low concentrations of rare-
earth elements (ΣREE = 34.8 ppm) which reveal a slightly fractionated pattern
((La/Yb)N = 4.12) and a slight, positive Eu anomaly (Eu/ Eu* = 1.42).
The Mandé Batholith: Holocrystalline and porphyritic samples of the Paleocene-­
Eocene Mandé Batholith are represented by seven samples. Five reveal evidence of
alteration, whilst two are relatively unaltered. The samples are metaluminous with
exception of two of the altered samples which plot in the peraluminous field, likely
due to alteration. As with the Cretaceous granitoids of the Western Group, the
Mandé suite includes low-K tholeiitic and medium- to high-K calc-alkaline mem-
bers. The low-K samples record enhanced degrees of alteration. Notwithstanding,
most varieties of potassic alteration involve K-enrichment (Warren et al. 2007), and
hence the low-K samples are herein considered to reflect an originally low-K proto-
lith. This is backed by apparent REE depletion (ΣREE = 23.79–88.86 ppm) in the
low-K samples vs. the higher overall REE contents (ΣREE = 109.86–125.93 ppm)
of the medium- to high-K samples, suggesting the low-K samples are distinctly
more primitive. REE chondrite-normalized patterns for all samples show moderate
decreasing slopes ((La/Yb)N  =  2.27–5.81) with moderate to subtle negative Eu
anomalies (Eu/Eu* = 0.63–1.07).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 321

Whole-Rock Lithogeochemistry Summary and Discussion for the Cretaceous-­

Eocene Granitoids
Based upon the composite lithogeochemical data set for Cretaceous to Eocene
granitoids used in this study, it is possible to differentiate and characterize the Eastern
and Western granitoid groups derived from the spatial analysis outlined above.
Further consideration of these groups, in conjunction with U-Pb (zircon) age data,
permits the identification of subgroups within each group, as follows: the Eastern
Group includes (1) the ca. 96–72 Ma autochthonous, continental granitoids of the
Antioquian Batholith and (2) the ca. 62–49 Ma granitoids of the Providencia suite,
Sonsón Batholith and Central Cordilleran plutons to the south of the Antioquian
Batholith and the Eocene intrusives of the Sierra Nevada de Santa Marta. The
Providencia suite granitoids, although temporally belonging to the Paleocene suite,
are spatially contained within the confines of the Antioquian Batholith and display
some unique lithogeochemical characteristics which have been highlighted above.
Within the allochthonous, oceanic Western Group (CCOP-/CLIP-related) granitoids,
two groups are distinguished, (1) ca. 100–82 Ma, Santa Fé through Jejénes trend,
hosted along the tectonized margin of the Dagua and Cañas Gordas terranes, and (2)
the ca. 50–42 Ma granitoids and hypabyssal porphyries of the Mandé-­Acandí arc,
hosted within El Paso-Baudó oceanic basement.
The Antioquian Batholith granitoids demonstrate a differentiation trend from
gabbroic through quartz-dioritic compositions, with generally increasing alkalinity
and silica contents and increasing levels of REE fractionation vs. time. The later
phases reveal characteristics of within-plate and syn-collisional granitoids, suggest-
ing increased crustal input or crustal interaction/contamination (Pearce et al. 1984).
Interestingly, the Paleocene Providencia suite marks a clear return to more mantelic
less-fractionated compositions. The Irra Stock is a relatively enriched, alkali series
shoshonite emplaced within the Romeral tectonic zone at ca. 70 Ma. It is of similar
age and composition as alkali granitoids emplaced within the Romeral zone, to the
north at Sucre, as recently revealed by Vinasco (2018).
The post-Antioquian granitoids to the south demonstrate similar compositional
trends to those of the Antioquian suite, with a general tendency towards increased
alkalinity and silica contents and greater degrees of REE fractionation. Pronounced
Eu anomalies are particularly recorded for the leucocratic phases of the Sonsón
Batholith. The Sierra Nevada de Santa Marta granitoids demonstrate metaluminous,
magnesian-calcic lithochemistry granitoids with relatively typical arc-related trace
and REE profiles.
Although interpretation is obscured to some degree by notably greater levels of
post-crystallization alteration and tectonism, when compared to the Eastern Group
granitoids, the whole-rock lithogeochemistry of ca. 100–82  Ma Western Group
granitoids is clearly reflective of their spatial distribution within Farallon and CCOP/
CLIP-related, oceanic domain basement rocks. Despite alteration patterns, all of the
ca. 100–82  Ma Western Group granitoids plot within the volcanic arc granitoids
field of Pearce et al. (1984). When compared to the Eastern Group, the Santa Fé
through Jejénes trend plutons are decidedly more calcic in composition, with lesser
322 H. Leal-Mejía et al.

degrees of alkalinity and silica, HFSE and REE enrichment, typical of more primi-
tive granitoids developed within intra-oceanic vs. continental arcs. Hints of Fe
enrichment are observed within the Western Group suites on both the AFM and
FeO(total)/(FeO(total)  +  MgO) vs. silica diagrams. Notwithstanding, it is noteworthy
that the larger batholiths, including Santa Fé and Buga, were apparently sufficiently
stable and long-lived to develop more alkali- and silica-rich fractionates, relatively
enriched in HFSE.
With respect to the ca. 62–42 Ma Western Group Mandé-Acandí Batholith, it is
observed that the representing sample population is small, considering that Mandé-­
Acandí represents a far greater volume of granitoid magma, exceeding that of all of
the ca. 100–82  Ma Western Group granitoids combined. Notwithstanding, and
despite additional differences in age with respect to the Santa Fé through Jejénes
trend, the early, phaneritic holocrystalline phases of the Mandé-Acandí Batholith
record similar major, minor and trace and rare-earth element (REE) trends, r­ eflective
of development and emplacement within similar, albeit younger, allochthonous,
intra-oceanic basement. Notably, some of the younger porphyritic granitoids of the
Pantanos-Pegadorcito area, and some of the phaneritic granitoid samples, return a
much more evolved composition, richer in alkalies and silica, with enhanced trace
and REE contents, and the incipient development of negative Eu anomalies. These
observations are considered to be reflections of the longer-lived and mature devel-
opment of the Mandé-Acandí arc, complete with generation of a coeval volcanic
pile (Santa Cecilia-La Equis Fm.).
Based upon the forgoing data presentation and conclusions, variations in the age
vs. lithogeochemical expression of the entire suite of Cretaceous through Eocene
(i.e. ca. 100–ca. 40 Ma) age Eastern and Western Group granitoids of the Colombian
Andes, as observed with the Jurassic-aged correlatives, are primarily a function of
the nature and composition of the basement into which the granitoid magmas have
been emplaced. Even more so, the specific tectonic environment and conditions at
the time of granitoid emplacement, varying from continental and autochthonous to
CCOP-/CLIP-related oceanic and allochthonous, are in evidence. In the case of the
Cretaceous to Eocene granitoids, differences in granitoid composition are very
much a reflection of the dynamic and changing tectonic conditions taking place
primarily within the Pacific realm during the formulative phases of the Northern
Andean Orogeny. We now present additional geochemical and isotopic data for the
Cretaceous-Eocene granitoids of the Colombian Andes, in order to more precisely
discuss the tectono-magmatic evolution of the region in the coming sections of this
presentation.

Isotope Geochemistry

Sr-Nd Isotope Geochemistry: Results, Summary and Discussion

A relatively complete set of Sr-Nd isotope data, representative of both the Eastern
Group and Western Group of Colombian Cretaceous to Eocene granitoids, is pre-
sented in Fig.  5.21. Data for both the Eastern and Western groups and their
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 323

Fig. 5.21  Sr-Nd and Pb isotope plots for mid-Cretaceous through Eocene granitoids in the
Colombian Andes. Additional data for igneous and metamorphic suites from the surrounding
region are included for reference. Note the primitive isotopic composition of the Western Group
oceanic granitoids vs. the continental granitoids of the Eastern Group. Notwithstanding, the
strongly mantellic signature of the initial post-collisional Providencia suite in the eastern domain
is noteworthy, compared to the evident crustal contamination observed for the remainder of the
Eastern Group post-collisional suite. See text for discussion

respective subgroups, as described above, are clearly resolved, with essentially no

observed Sr-Nd compositional overlap. For reference purposes, we have included
two additional data sets within the context of Fig.  5.21: (1) data for early(?)
Cretaceous, tectonically disrupted, continental margin MORB contained within the
Romeral melange (Quebradagrande/Arquía units of Nívia et al. 1996) and (2) data
for autochthonous early to mid-Cretaceous, rift-related mafic intrusions located
within the Eastern Cordillera (Fabre and Delaloye 1983; Vázquez et al. 2010). In
general terms, it is noted that all of the data sets represented in Fig. 5.21, including
for both mafic and granitoid rock types, plot along the mantle array. Alteration of
original Sr-Nd isotopic composition is suggested, particularly in terms of increasing
87
Sr/86Sr(i) ratios, within the Cretaceous MORB and Eastern Cordilleran mafic suites.
The Cretaceous MORBs are highly tectonized and commonly demonstrate some
degree of carbonitization, as do the Eastern Cordilleran intrusions, which have been
exposed to post-emplacement basin dewatering events during late Cretaceous and
Cenozoic basin inversion (Vázquez et al. 2010; Shaw et al. 2018).
With respect to the Eastern Group Cretaceous to Eocene granitoids shown in
Fig.  5.21, Sr-Nd isotope values for the ca. 96–72  Ma Antioquian Batholith suite
324 H. Leal-Mejía et al.

indicate a mantle-derived source (Leal-Mejía 2011). The Antioquian Batholith

demonstrates a vertical array with similar initial 87Sr/86Sr(i) ratios (0.70398–0.70455)
and increasing εNd(t) values, from the ca. 87–82  Ma facies to ca. 72  Ma facies
(εNd(t)  =  +1.74 to +4.77). The younger Paleocene-Eocene plutons of the Eastern
Group similarly plot along the mantle array but show a clear shift to lower εNd(t)
values and higher 87Sr/86Sr(i) ratios, suggesting more evolved compositions with
greater degrees of crustal interaction, as previously suggested by the whole-rock
lithogeochemical data. The ca. 62–58 Ma Providencia suite, hosted within the con-
fines of the Antioquian Batholith, provides an exception to this tendency, displaying
clearly more juvenile Sr-Nd compositions. A crustal component becomes more evi-
dent in the Cretaceous Mariquita Stock, which, based upon regional geologic map-
ping, can be inferred to have interacted with disrupted continental basement of
Mesoproterozoic(?) age, along the inferred contact between the Chicamocha and
Cajamarca-Valdivia basement complexes (Fig. 5.15).
With respect to the allochthonous Western Group CCOP/CLIP granitoids, all of
the analyses displayed in Fig. 5.21 plot within the mantle source region, reflecting
their ubiquitous genesis and emplacement within the oceanic regime. Samples from
the Mandé Batholith yielded clearly mantelic Sr and Nd isotope values, with the late
porphyritic phases displaying particularly enriched εNd(t) values (to +15.26). Similar
enrichment trends are noted within the more fractionated granitoids of the Buga and
Jejénes suites.

Pb Isotope Geochemistry: Results and Summary and Discussion

Available lead isotope geochemical data for the Eastern and Western Group
Cretaceous to Eocene granitoids of the Colombian Andes are also presented in
Fig.  5.21. As with the Sr-Nd diagram, Pb data for early(?) Cretaceous Romeral
melange MORB (Quebradagrande/Arquía units; Nívia et al. 1996), and for Eastern
Cordillera rift-related mafic intrusions (Fabre and Delaloye 1983; Vázquez et  al.
2010), have been included in the diagram. In addition, the data envelope for
Paleozoic metasedimentary basement of the Loja Terrane, Ecuador (Chiaradia et al.
2004), as a proxy for the unknown values of the Cajamarca-Valdivia assemblage
(Restrepo-Pace 1992; Cediel et al. 2003) which hosts the majority of the Eastern
Group granitoids and formed the Colombian continental margin during early-mid-­
Cretaceous times, is shown.
Lead isotope geochemistry both in the ca. 96–72 Ma and the ca. 62–49 Ma sub-
groups of Eastern Group granitoids clusters in a relatively narrow range of values
(206Pb/204Pb  =  18.74–19.21, 207Pb/204Pb  =  15.58–15.68 and 208Pb/204Pb  =  38.48–
39.05), extending in a moderately steep array, between the Orogene and the upper
crust lead evolution curves of Zartman and Doe (1981). A general tendency towards
more radiogenic values, suggesting greater degrees of crustal interaction, between
the earlier Antioquian Batholith suite and the Paleocene-Eocene suites of the Central
Cordillera to the south, can be observed. Regardless, as with the lithogeochemical
and Sr-Nd data, a clear return to less interacted, more mantelic compositions is
reflected in the data from the ca. 62–58 Ma Providencia granitoid suite, contained
within the confines of the Antioquian Batholith.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 325

To the west, within the continental margin to oceanic domain, Cretaceous MORB
of the Romeral mélange is observed to be notably radiogenic (Fig. 5.21b), in keep-
ing with oceanic rocks formed along a marginal basin, receiving sediments and
rifted fragments of the relatively radiogenic continental basement represented by
the Cajamarca-Valdivia Terrane. The influence of continental margin sedimentation
may be extrapolated to explain the only slightly less radiogenic Pb isotope composi-
tion of the ca. 100–82  Ma subgroup of the Western Group CCOP/CLIP-related
granitoids (Fig.  5.21b) (the lead isotope results for which range from
206
Pb/204Pb  =  19.08–19.44, 207Pb/204Pb  =  15.67–15.70 and 208Pb/204Pb  =  38.77–
38.91). This observation supports the formation of the ca. 100–82  Ma CCOP-/
CLIP-related intra-oceanic arc in relatively close proximity to the Colombian
­continental margin, along a subduction trench which was receiving continentally
derived sediment, prior to accretion. Farther west, available lead isotope analyses
for samples from the Mandé Batholith (i.e. the Pantanos porphyry suite), however,
depict a shift to less radiogenic compositions (206Pb/204Pb  =  18.92–18.96,
207
Pb/204Pb = 15.61–15.64 and 208Pb/204Pb = 38.56–38.60). This feature can be inter-
preted to reflect development of the ca. 62–42 Ma Mandé(-Acandí) arc in a more
distal intra-oceanic environment, isolated from the influence of significant volumes
of radiogenic, continentally derived sediments.

Synthesis and Conclusions of Lithochemistry and Sr, Nd and Pb Isotope

Geochemistry for Cretaceous-Eocene Granitoids
Spatial, temporal and lithogeochemical considerations, supported by whole-rock
trace element, REE and Sr-Nd and Pb isotopic compositional data, for mid-­Cretaceous
through Eocene (ca. 100 through ca. 42 Ma) granitoids of the Colombian Andes, per-
mit the identification and consistent characterization of two broad groups of granit-
oids: the Eastern and Western groups. All of the granitoids within both groups may be
classified as volcanic arc-related granitoids; however, each group formed within a
distinct litho-tectonic setting: the Eastern Group is comprised of autochthonous
granitoids emplaced within a continental setting, whilst the Western Group is com-
prised of allochthonous, accreted arc granitoids, formed within an intra-oceanic set-
ting. Each of these principal groups may be further subdivided into two individual
subgroups, primarily based upon age constraints provided by the U-Pb (zircon) age
database, although the lithogeochemical and isotopic evolution of each subgroup is
reflective of the evolving petrogenesis and tectonic setting of the region.
The Cretaceous phases of the Eastern Group (Antioquian Batholith suite) are
subduction-related continental arc (Cordilleran) granitoids with relatively limited
degrees of crustal interaction, as recorded in their Sr-Nd and Pb isotope signatures.
Following a ca. 10 Ma hiatus in continental granitoid magmatism, the Paleocene-­
Eocene Eastern subgroup granitoids to the south of the Antioquian Batholith show
similar, although more evolved, lithogeochemical compositions and depict isoto-
pic trends towards increasing crustal interaction (e.g. Bustamante et al. 2017). A
marked exception to this trend remains the ca. 62–58 Ma Providencia suite, which
displays a consistent return to more mantelic lithogeochemical, trace element,
REE and isotopic compositions. This is an important observation from a metallo-
326 H. Leal-Mejía et al.

genic viewpoint, given that the largest Au resource presently outlined within the
entire Cretaceous-Eocene granitoid suite of the Colombian Andes is parageneti-
cally related to the Providencia granitoid suite (Leal-Mejía et al. 2010; Leal-Mejía
2011; Shaw et al. 2018).
In contrast to the Eastern Group granitoids, the Western Group plutons display
more primitive lithogeochemical characteristics reflective of their generation within
an intra-oceanic setting. Sr-Nd and Pb isotope data, however, suggest, at least locally,
a significant crustal component, interpreted to represent the relative near proximity
of the Colombian segment of the CCOP/CLIP intra-oceanic arc system and accreted
Western Group granitoids, to the continental margin during the ca. 100–82 Ma time
interval. The ca. 62–42 Ma Mandé(-Acandí) subgroup of the Western granitoids is
more reflective of primitive intra-oceanic lithogeochemical compositions, albeit with
a more highly evolved component, as recorded by the composition of younger
(ca. 46–42  Ma) porphyritic stocks observed at Pantanos and elsewhere along the
Mandé-Acandí trend. From a Sr-Nd isotope standpoint, available samples of the
Mandé suite are consistently mantle-derived, and Pb-isotope compositions are sig-
nificantly less radiogeneic than granitoids of the ca. 100–82 Ma subgroup. From a
tectonic standpoint, the Mandé-Acandí arc was generated upon CCOP crust repre-
senting the trailing edge of the CCOP plateau, following the accretion of most of the
Western Tectonic Realm (Cañas Gordas and Dagua terranes) in the late Cretaceous
(Cediel et al. 1994; Cediel et al. 2003, 2010; Spikings et al. 2015). In this context, the
Colombian Pacific margin would have been comprised of accreted oceanic materials
during genesis of the Paleocene-Eocene Mandé-Acandí granitoids, and, unlike the
ca. 100–82  Ma granitoids, the Mandé arc system would have been shielded from
exposure to continentally derived sediments of the Cajamarca-Valdivia Terrane,
accounting for its less radiogenic Pb isotope signatures.

5.3.5  L
 atest Oligocene to Pliocene Granitoid Magmatism:
Distribution, Age and Nature
5.3.5.1  Introduction

Latest Oligocene to Pliocene granitoid magmatism is widely but irregularly distributed

throughout the Colombian Andes. Unlike the magmatic periods described previ-
ously, large masses of holocrystalline granitoids of Oligo-Miocene age are not
exposed in outcrop within the Colombian Andes, certainly due to factors related to
uplift and erosional level. Notwithstanding, small batholiths and stocks of holocrys-
talline rocks and abundant hypabyssal porphyry clusters and volcanic deposits are
widespread, especially within the western Andean ranges and intermontage valleys,
and testify to the active development of granitoid magmatisn during the tectonic
development of the Colombian Andes during the latest Oligocene and Miocene.
As with previous magmatic periods, latest Oligocene to Pliocene magmatism man-
ifests as a complex distribution of spatially constrained arc segments of differing ages.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 327

Unlike the Cretaceous to Eocene intrusions, however, all latest Oligocene to

Pliocene magmatic rocks are considered to have been emplaced broadly in situ, into
continental margin metamorphic and accreted oceanic basement rocks juxtaposed
prior to and during the various phases of the Northern Andean Orogeny (see
Sect. 5.4.4). Latest Oligocene to Pliocene magmatism in Colombia merges, both
temporally and spatially, into the modern-day Pleistocene to Recent volcanic arc of
the Northern Andes. The distribution, nature and constraints upon Pleistocene to
Recent magmatism are well documented in various geological compilations (e.g.
Cediel and Cáceres 2000; Cediel et al. 2003; Stern 2004; Gómez et al. 2015a; Marín-
Cerón et al. 2018) and will not be discussed in detail herein.
Based primarily upon textural characteristics, three broad categories of latest
Oligocene to Pliocene granitoid rocks can be distinguished. These include (1) holo-
crystalline, phaneritic granitoids forming stocks and small-scale batholiths, (2)
high-level hypabyssal porphyritic rocks occurring as clusters of small-scale stocks
and dikes and (3) volcanic rocks, including flows and pyroclastic sequences.
Spatially and genetically related volcanic sequences are generally absent in the
vicinity of the deeper-level phaneritic intrusive suites, whilst the hypabyssal por-
phyry stocks may at least locally intrude a penecontemporaneous volcanic edifice.
Erosional factors may account for the lack of a volcanic component in the deeper-­
level plutons, although in some instances it is possible that a volcanic component
was never developed.

5.3.5.2  Distribution

The great majority of latest Oligocene to Miocene granitoids in Colombia are con-
centrated within the western ranges and along the intermontane valleys of Colombian
Andes, with only minor, isolated occurrences of high-level granitoids of late
Miocene and Pliocene age having been documented in the Eastern Cordillera and
Santander Massif.
The distribution of latest Oligocene to Miocene holocrystalline plutons is pri-
marily observed in the physiographic Western Cordilleras of Colombia, where
numerous small batholiths and stocks are observed (Fig.  5.22). In the southwest
these include the Piedrancha and Cuembí batholiths and associated minor stocks
observed at El Vergel, La Llanada and Cumbitara, all of which intrude CCOP-/
CLIP-related oceanic volcanic and sedimentary rocks of the Dagua and Diabásico
Groups (Arango and Ponce 1982). Regional mapping suggests that this trend of
plutons extends northwards into the region to the west of Cali (Fig. 5.22), where
little modern radiometric age or lithogeochemical data are available. Farther to the
north, within the confines of the Chocó Arc, the holocrystalline Farallones Batholith,
the Urrao pluton and the El Cerro Igneous Complex are observed to intrude Cañas
Gordas Group basement to the west of the Middle Cauca River valley (Rodríguez
and Zapata 2012; Zapata and Rodríguez 2013). The northern extension of this trend
can only be inferred based upon the appearance of small plutons within the regional
mapping database (González 2001; Gómez et al. 2015a).
328 H. Leal-Mejía et al.

Fig. 5.22  Distribution of latest Oligocene through Mio-Pliocene granitoids in the Colombian
Andes. Principal modern-day physiographic provinces of the region are shown for reference.
(Granitoid shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al.
2015a)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 329

With respect to the hypabyssal porphyritic granitoid stocks and dikes and associ-
ated volcanic rocks, these clusters are particularly well exposed along the eastern
margins of the Western Cordilleras and the western margin of the Central Cordillera,
within/along the physiographic depressions of the Patía and Upper Cauca drainage
basins (Fig. 5.22). Additional, Miocene and Pliocene hypabyssal porphyry clusters
are observed within the Central Cordillera, the Santander Massif and the Eastern
Cordillera (including the Quetame Massif).
In the south, along the Patía and Upper Cauca drainage, a ca. 200  km long
SW-NE trending series of Miocene hypabyssal porphyry clusters extends from
Arboledas (Berruecos) to the area of Cerro Bolívar and Almaguer-La Vega (Betulia
Igneous Complex) and northwards through Altamira, Dominical, Piedra Sentada
and La Sierra. It is possible that this belt continues farther north beneath Plio-­
Pliestocene to Recent volcanic cover around Popayán. Another cluster of hypabys-
sal stocks appears in the Upper Cauca basin to the north of Popayán, at Santander
de Quilichao-Buenos Aires-Suárez (París and Marín 1979).
Farther north, along both the eastern and western margins of the Middle Cauca
River valley to the north of Pereira, numerous clusters of hypabyssal granitoid por-
phyritic stocks and dikes are observed over a ca. 100 km long N-S trending mag-
matic belt (González 1990, 1993, 2001). Along this belt (Fig.  5.22), porphyry
intrusives are commonly observed, either as isolated plugs or volumetrically more
significant clusters of stocks and dikes. From south to north, some of the more
important clusters of stocks outcrop near Marsella, at Manizales-Villa María, from
Anserma to Quinchía and from Río Sucio to Supía, La Felisa, Marmato and
Valparaíso-Caramanta at Támesis, around Jericó (the Quebradona cluster) and from
Venecia-Fredonia north to Titiribí. Spatially coincident with the Middle Cauca
River valley trend are the thick volcanic, pyroclastic and volcaniclastic sequences of
the late Miocene Combia Fm., which also outcrop in the Middle Cauca River region,
primarily to the west of the river between Anserma in the south and Jericó-Tarzo in
the north. The Combia Fm. has also been mapped around Venecia-Fredonia and
Titiribí. Combia is considered to be, in part at least, the extrusive expression of the
late Miocene porphyry centres.
To the north of Titiribí, hypabyssal porphyry centres become scarce and isolated
and are of generally only inferred to be late Miocene age. This observation may be
in part a facet of the difficulty in recognition of these generally small intrusives in
regional mapping, under heavy vegetation cover and deep tropical soil profiles.
Isolated porphyritic granitoid stocks are observed to the west of Anzá and around
Buriticá, where late Miocene 40Ar-39Ar ages have been published (Lesage et  al.
2013). This northern section of the Middle Cauca porphyry belt may extend through
Peque and as far north as Puerto Libertador, where isolated, small, holocrystalline
and porphyritic granitoid stocks and dikes are observed at El Alacran and Montiel.
The age of these last units, however, has yet to be established.
Within the Central Cordillera, a significant cluster of Miocene porphyritic
granitoids is observed at Cajamarca-Salento, including the Colosa porphyry (Lodder
et al. 2010) and other surrounding hypabyssal intrusives extending as far east as
the Toche river. This cluster may extend to the north where it would be covered by
330 H. Leal-Mejía et al.

volcanic and pyroclastic rocks of the modern-day Northern Andean volcanic arc.
Farther to the north, within the Central Cordillera, an additional cluster of hypabys-
sal granitoids and associated pyroclastic rocks appears in the Manzanares-Samaná-­
Nariño (Antioquia) region. This cluster was referred to as the Río Dulce suite
(Fig. 5.22) by Leal-Mejía (2011). Isolated high-level volcanic occurrences extend
as far east as Norcasia.
Within Colombia’s eastern cordilleran system, magmatic rocks of Miocene to
Pliocene age are very scarce and only punctually developed. A localized cluster
of hypabyssal granitoids is located in the Vetas-California area of the Santander
Massif (Fig. 5.22; Mantilla et al. 2009; Leal-Mejía 2011; Bissig et al. 2014). To
the south, minor, isolated, high-level porphyritic granitoids and associated volca-
nic flows and pyroclastic rocks are observed in the Eastern Cordillera at Paipa
and Iza (Garzón 2003; Pardo et al. 2005a, b; Vesga and Jaramillo 2009), whilst
similar occurrences are observed to the south and east of Bogotá at Quetame
(Ujueta et al. 1990).
The distribution and nature of Pleistocene to Recent magmatism in Colombia
is well documented and readily observed on regional geological compilations
such as those presented by Cediel and Cáceres (2000) and Gómez et al. (2015a).
This volcanic chain, including its extension to the south into Ecuador, is com-
prised of about 75 active volcanoes which are recognized within the Andean geo-
logical literature as the Northern Volcanic Zone (e.g. Stern 2004). In Colombia,
this magmatic arc is primarily manifested along the Central Cordillera and the
Patía-Upper Cauca River physiographic depression, where extensive volcanic and
pyroclastic deposits are related to active volcanic edifices. Volcanism is domi-
nated by lavas and pyroclastic rocks of bas-andesitic, andesitic and dacitic and
occasionally basaltic composition.
Stern (2004) distinguished three separate segments comprising the active
Colombian volcanic arc, including (1) the northern segment (Cerro Bravo, Santa
Isabel, Nevado del Ruíz, Nevado del Tolima, Cerro Machín volcanoes), (2) the cen-
tral segment (Nevado del Huila, Puracé, Sotará volcanoes) and (3) the southern
segment (Cumbal, Azufral, Galeras, Doña Juana volcanoes). The northern and cen-
tral segments are located in the Central Cordillera, whereas the southern segment is
located in the Patía-Upper Cauca River depression and along the eastern margin of
the southern Western Cordillera.
Cediel et al. (2003) recognized that the regional-scale structural architecture of
the Northern Andes plays a fundamental role in the distribution and localization of
volcanic edifices in Colombia. They documented the coincidence of volcanic cone
and segment (“sub-chain”) locations, with the trace of the various paleo-suture sys-
tems active in the tectonic assembly of the Northern Andean region since the mid-­
Proterozoic. In Colombia five sub-chains were defined, including, from west to east,
the Cauca sub-chain (Chiles-Cumbal-Azufral-Olaya volcanoes), the inter-Andean
sub-chain (Galeras-Morazurco volcanoes), the Romeral-Peltetec sub-chain (La
Victoria-Chimbo-Bordoncillo-Doña Juana-Sotará-Puracé volcanoes), the Palestina
sub-chain (La Horqueta-Paletará-Huila-Tolima-Ruíz-Herveo volcanoes) and the
Suaza sub-chain (Guamués-Acevedo volcanoes).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 331

5.3.5.3  Age Constraints on Miocene to Pliocene Granitoid Magmatism

Published U-Pb (zircon) data or information regarding the precise age of Miocene-­
Pliocene magmatism in the Colombia Andes is surprisingly limited (e.g. Maya 1992;
Frantz et al. 2003; Tassinari et al. 2008; Mantilla et al. 2009; Lodder et al. 2010;
Henrichs 2013; Lesage et al. 2013; Bissig et al. 2014). Based primarily upon older
K-Ar (whole-rock or mineral separate) analyses, previous work and published geo-
logical maps in general refer to interpreted Miocene- to Pliocene-aged magmatic
rocks as being “Neogene” in age, thus historically precluding any form of detailed
analysis of the spatial appearance and migration of Miocene to Pliocene magmatic
rocks with time.
More recently, Leal-Mejía (2011) presented a study containing 24 new U-Pb
(zircon) age determinations for Miocene-Pliocene holocrystalline granitoid intru-
sions and hypabyssal porphyritic stocks, backed by various new K-Ar mineral sepa-
rate and whole-rock dates. This work permits a more precise time-space analysis
and understanding of the evolution of Miocene and Pliocene granitoid magmatism
throughout the Colombian Andes. A histogram depicting the age distribution of
Miocene to Pliocene magmatic rocks based upon recent U-Pb (zircon) age determi-
nations is presented in Fig. 5.23.
With respect to the holocrystalline, phaneritic plutons, the U-Pb (zircon) data of
Leal-Mejía (2011) indicates the oldest, ca. 23  Ma magmatic rocks are located in
southwest Colombia, extending from the Piedrancha Batholith north to El Vergel, La
Llanada and Cumbitara (Cuembí) plutons. These plutons record a latest Oligocene to
early Miocene magmatic event spanning the ca. 24–21 Ma interval. Analysis of the
Piedrancha Batholith near Piedrancha yielded an Oligo-Miocene magmatic crystalli-
zation age of ca. 23.4 Ma. To the north the El Vergel Stock returned an early Miocene
magmatic crystallization age of ca. 21.9 Ma. The Cumbitara Stock returned a crystal-
lization age of ca. 23.1 Ma. No inheritance ages were observed in any of the samples.
As noted above, based upon available mapping, sporadic plutons, which could be of
similar age, are mapped extending northwards along the trend of the Western Cordillera
into the region to the west of Cali (Fig. 5.22).
Farther to the north, for the holocrystalline, phaneritic intrusives including the
Farallones Batholith and the Urrao pluton, no new U-Pb (zircon) dates are available.
These units have historically been dated by the K-Ar (hornblende) method and have
returned ages including 11 ± 2 Ma (Calle et al. 1980; Zapata and Rodríguez 2013) and
11 to 12 Ma (Botero 1975), respectively. To the north of Urrao, along an approximate
N-S axis, Leal-Mejía (2011) recorded a K-Ar (biotite) date of ca. 11.8 Ma for the El
Cerro Igneous Complex. Regional mapping suggests this trend of plutons may extend
further northwards into the region to the north of Dabeiba (González 2001).
With respect to the porphyritic granitoid suites, U-Pb (zircon) analyses to date
have revealed ages ranging from Miocene to Plio-Pleistocene. Consideration of the
composite data set reveals a complex time-space distribution of these rock types.
Beginning in southwestern Colombia, along the Patía and Upper Cauca drain-
age, the trend of porphyritic stocks and dikes, extending from Arboledas (Berruecos)
in the south to Almaguer-La Vega (Betulia Igneous Complex), and Piedra Sentada
332 H. Leal-Mejía et al.

and La Sierra returned ages spanning the ca. 17–9 Ma interval (Leal-Mejía 2011).
The oldest magmatic crystallization age was obtained from the Dominical porphyry,
which returned an age of ca. 17 Ma. The Cerro Gordo porphyry returned a mag-
matic crystallization age of ca. 14 Ma. Holocrystalline hornblende biotite tonalite
from La Dorada and tonalite porphyry from Altamira returned similar ages of ca.
11.8 Ma and 11.6 Ma, respectively, whilst a hornblende diorite porphyry from La
Dorada returned a magmatic age of ca. 9.2 Ma. To the north of Popayán, the northern
Cauca Department hypabyssal porphyry cluster at Santander de Quilichao-­Buenos
Aires-Suárez was also dated by Leal-Mejía (2011). Diorite porphyry from near
Suárez returned a magmatic crystallization age of ca. 17.7  Ma, which compares
well with the age of similar porphyries from the Dominical area to the south.
The hypabyssal porphyry suites of the Middle Cauca return distinctly younger
ages than those of the Patía-Upper Cauca trend. Hypabyssal granitoids along the
Middle Cauca extend from Marsella in the south to Titiribí in the north. Available
U-Pb (zircon) data span the range between ca. 9 and 4 Ma (Fig. 5.23). Quartz diorite
porphyry from Quinchía (Dos Quebradas) returned a U-Pb (zircon) age of ca.
8.0 Ma (Leal-Mejía 2011). Diorite porphyry from the Marmato area returned a ca.
6.5 Ma age (Frantz et al. 2003). Granodiorite of the Támesis Stock, sampled near
Támesis, returned a U-Pb (zircon) age of ca. 7.2 Ma (Leal-Mejía 2011), in marked
contrast to historic K-Ar age data which suggested a Cretaceous age (Maya 1992).
At Yarumalito, Henrichs (2013) provided U-Pb (zircon) crystallization ages of
7.0 ± 0.15 Ma and 6.95 ± 0.16 Ma for samples of andesite and diorite, respectively.
North of Yarumalito, Leal-Mejía (2011) provided various additional U-Pb (zircon)

Fig. 5.23  U-Pb (zircon) and selected K-Ar and 40Ar-39Ar age date populations for latest Oligocene
through Mio-Pliocene granitoids in the Colombian Andes
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 333

age dates: diorite porphyry at La Aurora within the Jericó (Quebradona) porphyry
cluster, 8 km to the SE of the Jericó townsite, yielded a U-Pb (zircon) age of ca.
8 Ma. The La Mina diorite porphyry, located about 5 km to the south of Venecia,
produced an age of ca. 7.6 Ma, and hornblende granodiorite porphyry from the El
Medio creek, located a few hundred metres to the SW from the Titiribí townsite,
yielded a ca. 7.6 Ma U-Pb (zircon) age.
To the north of Titiribí, hypabyssal porphyritic granitoid magmatism along the
Middle Cauca becomes less apparent and has only been documented in isolated,
mostly undated occurrences. Leal-Mejía (2011) provided an 11.8 ± 1.1 Ma K-Ar
(magmatic hornblende) age for hornblende diorite porphyry at Buriticá. 40Ar-39Ar
step heating analysis of similar porphyry by Lesage et al. (2013) produced a horn-
blende cooling age of 7.41 ± 0.4 Ma, suggesting various phases of porphyritic dio-
rites may be present. Elsewhere, along the general northern extension of the Middle
Cauca trend, porphyritic granitoids of unconfirmed late Miocene age are observed
to the west of Anzá, north of Dabeiba, near Peque and to the SW of Puerto
Libertador, at El Alacrán and Montiel (Teheran), where small, isolated quartz dio-
rite stocks and porphyritic dikes intrude volcano-sedimentary basement rocks of
the San Jacinto terrane. Most of these occurrences are too small and isolated to be
resolved at the scale of available regional geologic maps (e.g. Cediel and Cáceres
2000; Gómez et al. 2015a). For the purposes of our analysis, these occurrences are
considered to represent the northern extension of granitoid magmatism along the
Middle Cauca trend, and we tentatively assign these rocks a late Miocene age of
emplacement.
The volcanic and pyroclastic sequences of the Combia Fm. are exposed through-
out the Middle Cauca River valley region (Fig. 5.22), and many of the Middle Cauca
trend hypabyssal porphyry intrusives are hosted within the Combia Formation.
Detailed geological mapping suggests the Combia Fm. records a transition from
Oligo-Miocene siliciclastic sedimentation along the paleo-Cauca and Sinifaná
basins (i.e. Amagá Fm.) to active granitoid magmatism during the middle to late
Miocene. Historic K-Ar dating places the Combia suite at ca. 9.1  Ma (Restrepo
et al. 1981 in Toro et al. 1999), although it is uncertain what level of the volcanic
pile this would represent. Based upon stratigraphic and cross-cutting relationships
in relation to the ca. 7 Ma hypabyssal stocks near Yarumalito, Henrichs (2013) con-
cluded that porphyry emplacement was closely related to the final stages of Combia
Fm. volcanism. Leal-Mejía (2011) dated an andesite of the upper Combia Formation
outcropping near Támesis at ca. 6.1 Ma (K-Ar, whole-rock). This is in broad agree-
ment with the ca. 6  Ma age of “Combia volcanism” suggested by Ramírez et  al.
(2006) and with K/Ar and 40Ar-39Ar mineral and whole-rock age dates produced by
various studies and compiled by Rodríguez and Zapata (2014). In reality, the
Combia Fm. is an extensive volcano-sedimentary unit locally exceeding 1000
meters in stratigraphic thickness. It is cut by many of the hypabyssal granitoids
listed above indicating that volcanism initiated before ca. 8  Ma and may have
continued to as late as ca. 4 Ma should volcanism have accompanied the full range
of hypabyssal porphyritic granitoid magmatism as indicated by available U-Pb
(zircon) age dates. Notwithstanding, lithogeochemical data presented below suggest
334 H. Leal-Mejía et al.

that the early, mafic (bas-andesitic) phases of the Combia Fm. may be related to the
ca. 12–10 Ma Farallones Batholith suite.
To the south and west of the Middle Cauca region, within the Central Cordillera,
the Cajamarca-Salento hypabyssal porphyry cluster (Fig. 5.22), including various
granitoid porphyry bodies outcropping near Cajamarca, Tierradentro, Montecristo
and Salento, was dated by Leal-Mejía (2011). In general, results span the ca. 8.3–
6.3 Ma range. Magmatic crystallization ages for the hypabyssal intrusive suite at La
Colosa (see Lodder et al. 2010) yield ages spanning the ca. 8.3–7.3 Ma interval.
Three early diorite porphyries returned U-Pb (zircon) ages between ca. 8.3 and
7.9 Ma, whilst paragenetically later granodiorite porphyries yielded slightly younger
ages of ca. 7.6 Ma and 7.5 Ma. A latest phase of quartz porphyry returned a 7.3 Ma
magmatic crystallization age. Inheritance ages obtained from zircon crystals from
the La Colosa porphyries span a wide range between ca. 1060 and 13 Ma. Elsewhere,
hypabyssal granitoid porphyry from La Morena and Tierradentro returned ages of
ca. 8.4 Ma and 8.1 Ma, respectively. A quartz diorite porphyry from the Montecristo
area returned a ca. 7.6 Ma U-Pb (zircon) age, and granodiorite porphyry collected
near Salento returned ca. 6.3 Ma.
Continuing along the Central Cordillera, to the NNE of the Cajamarca-Salento
hypabyssal porphyry cluster, beyond active volcanic cover provided by the Tolima-­
Santa Isabel-Ruíz volcanic complex, the Plio-Pleistocene Río Dulce porphyry clus-
ter (Fig.  5.22) was also revealed by U-Pb (zircon) age dating completed by
Leal-Mejía (2011). Eight U-Pb analyses were provided at Río Dulce, for the suite of
hypabyssal granitoids and associated pyroclastic volcanic rocks. Magmatism span-
ning the 2.4 Ma to 0.4 Ma interval was recorded, in at least three distinct magmatic
pulses, including at 2.4 to 2.3 Ma, 1.2 to 1.0 Ma and 0.4 Ma. The oldest magmatic
ages at Río Dulce are revealed in two granodiorite porphyry samples collected to the
SSE from the Nariño (Antioquia) townsite. Both returned the same age of ca.
2.4 Ma. A porphyry fragment from a nearby intrusive breccia returned an age of ca.
2.3 Ma. Diorite porphyry which seems to cut the breccia also returned a ca. 2.3 Ma
age. A second magmatic pulse was recognized in the Espíritu Santo-Santa Bárbara
porphyry, located about 5 km to the SE of the Nariño townsite. Quartz diorite por-
phyry from Espíritu Santo hill returned an age of ca.1.0 Ma. In addition, two sam-
ples collected along the Espíritu Santo Creek, about 3  km to the east, returned
similar ages of ca. 1.2 Ma and 1.0 Ma. Finally, in the northern Río Dulce area, dio-
rite porphyry collected at La Cabaña hill, about 12  km to the east of the Nariño
townsite, returned a Pleistocene magmatic age of ca. 0.4 Ma.
Overall, ages for the hypabyssal intrusive rocks of the Río Dulce area porphyry suite
obtained by Leal-Mejía (2011) are compared well with Pliocene to Pleistocene K-Ar
ages (Maya, 1992) and geological relationships established for the nearby Tolima-Ruíz
volcanic complex, suggesting that the apparently extinct Río Dulce cluster formed, in
its time, the northernmost extension of the modern-day Northern Andes volcanic arc.
Further north and along the trend of the active Northern Andes volcanic zone, within
the Central Cordillera (including the Serranía de San Lucas), no additional manifes-
tations of Miocene or younger granitoid magmatism have been documented or
established using modern radiometric age dating techniques (Fig. 5.22).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 335

Within Colombia’s Eastern Cordilleran system, including within the Sierra

Nevada de Santa Marta, Serranía de Perijá, Santander Massif, Eastern Cordillera
and Garzón Massif, Miocene and younger granitoid magmatism is far scarcer than
within the Central and Western Cordilleran ranges. In this context, documented
granitoids or their volcanic equivalents, of confirmed Miocene or younger age, con-
stitute less than a fraction of one percent of the geological record of the entire
­eastern Colombian Andes. Notwithstanding, localized occurrences of granitoid
rocks have in recent years been documented, and their age and occurrence contrib-
ute to an understanding of the magmato-tectonic evolution of the region as a whole.
Within the Santander Massif, a NNE-oriented generally linear trend of late
Miocene hypabyssal granitoids along the Vetas-California mining district has
recently been documented in detail by various authors including Mantilla et  al.
(2009), Leal-Mejía (2011) and Bissig et al. (2014). Mantilla et al. (2009) presented
U-Pb (zircon) data for two samples of granitoid porphyry from near the Vetas and
California townsites. Results yielded magmatic crystallization ages of ca. 9.0 Ma
and 8.4  Ma, respectively. Leal-Mejía (2011) dated hypabyssal granodiorite por-
phyry cropping at the San Celestino Mine near California, yielding a ca.
10.2 Ma U-Pb (zircon) age. Both the above cited studies noted various inheritance
ages reflected in zircon grains, including at ca. 30 Ma, 50 Ma, 180 Ma and 200 Ma.
Mantilla et al. (2009) related the older ages to inherited zircons from metamorphic
and magmatic rocks in the area, including the granitoid rocks of the Santander
Plutonic Group, and explained the younger inheritance ages as possible age mixing
in zircon grains with more complex internal structure or inherited zircon grains
from younger unidentified magmatic pulses in the area.
In recent studies more directly pertaining to gold mineralization along the Vetas-­
California trend, Rodríguez (2014) and Rodríguez et al. (2017) produced 40Ar/39Ar
studies returning Pliocene to Pleistocene plateau ages ranging from ca. 3.9 to
1.2 Ma, for alteration minerals (primarily alunite) contained within and overprinting
the Miocene porphyritic granitoids and Proterozoic through Jurassic basement
rocks of the Santander Massif. Although U-Pb (zircon) age data for porphyritic
granitoids from the sector has yet to produce such young magmatic crystallization
ages, both of the foregoing studies conclude that the Plio-Pleistocene 40Ar/39Ar ages
for alteration alunite reflect the presence of coeval granitoid magmatism, located at
shallow depths in upper crustal levels beneath the district, which has yet to be
revealed by uplift and erosion.
Some 175 km to the south of Vetas-California, within the domain of Colombia’s
Eastern Cordillera (sensu stricto), punctual occurrences of high-level porphyritic
granitoids and associated flows and pyroclastic rocks are known at Paipa-Iza.
K-Ar and 40Ar/39Ar radiometric dates at Paipa span the range between ca. 2.4 and
1.8 Ma; however, field evidence suggests volcanic activity continued into more
recent times (Cepeda and Pardo-Villaveces 2004; Pardo et al. 2005a). An addi-
tional 180 km further to the south, similar granitoids to those observed at Paipa,
although of more limited aerial extent, are located near the town of Quetame. A
K-Ar (whole-rock) age of ca. 5.6 Ma was recorded for the Quetame granitoids by
Ujueta et al. (1990).
336 H. Leal-Mejía et al.

Discussion and Synthesis of Spatial Distribution of Oligocene to Pliocene

Granitoids

Unlike some of the Cretaceous through Eocene magmatic arc segments of the
Colombian Andes, tectono-magmatic analyses (e.g. Apsden et al. 1987; Cediel et al.
2003; Leal-Mejía, 2011) indicate that all of the latest Oligocene through Pleistocene
granitoid arc segments and isolated granitoid occurrences, summarized above, are
autochthonous within the Miocene tectonic configuration and evolution of the
region. In this context, the spatial vs. temporal evolution of continental granitoids
during the Neogene is linked to pre- and syn-Neogene crustal architecture and intra-­
continental stress field evolution, coupled with the nature of oceanic vs. continental
plate interactions along the Pacific and Caribbean margins. As our magmato-­tectonic
analysis presented in Sect. 5.4.4.2 will discuss, we consider all of the major latest
Oligocene through Pliocene granitoid arc segments discussed above to be related to
the westward subduction of Pacific (Nazca) Plate crust, beginning in the late
Oligocene.
The spatial vs. temporal relationships of latest Oligocene through Plio-
Pleistocene granitoids are revealed by regional geological mapping (e.g. Cediel
and Cáceres 2000; Gómez et al. 2015a), in conjunction with relatively recent U-Pb
(zircon) and other age date studies cited above. The data reveal that the oldest gran-
itoids from this time period include holocrystalline quartz-diorite and granodiorite
of the ca. 24–21 Ma Piedrancha-Cuembi trend. This arc segment was apparently
relatively short-lived and extinct, prior to the eastward migration of arc axial mag-
matism into the hypabyssal granitoid porphyry-dominated, ca. 17–9  Ma Upper
Cauca-Buenos Aires-Suarez arc segment. Pleistocene to modern-day arc magma-
tism currently overprints the southern portion of the Upper Cauca porphyry belt, as
manifest in active volcanoes such as Galeras and Doña Juana, and continues to
migrate eastwards as manifest in volcanic fields around San Roque and the Nevado
del Huila.
Following closure of the Piedrancha-Cuembí arc segment and in the later phases
of Upper Cauca magmatism, holocrystalline granitoid magmatism abruptly reap-
pears in the northern Chocó Arc sector, of the physiographic Western Cordillera,
intruding the Cañas Gordas Terrane basement complex. Farallones-El Cerro is a
spatially and temporally distinct arc, unrelated to Piedrancha-Cuembí-Upper Cauca.
It formed to the north of paleo-transform faults within the Pacific-Nazca Plate
(Fig. 5.22) and represents the results of the differential interaction of a segmented
subducting Nazca Plate along the Colombian Pacific margin (Sect. 5.4.4.2). The ca.
12–10 Ma Farallones-El Cerro segment was also short-lived, as arc axial magma-
tism migrated eastwards into Romeral melange basement, with the appearance of
widespread porphyritic granitoids and associated volcanic rocks of the upper
Combia Fm. along the belt-like Middle Cauca arc segment, between ca. 9 and 4 Ma.
Simultaneous magmatism beginning at ca. 9 Ma was recorded to the ESE of the
Middle Cauca, within the Cajamarca-Salento porphyry cluster, hosted by Cajamarca-­
Valdivia metamorphic basement rocks of the Central Cordillera. A similar cluster of
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 337

porphyritic granitoids of Plio-Pleistocene age appears at Río Dulce, to the NNE of

Cajamarca-Salento. Both Cajamarca-Salento and Río Dulce are broadly coaxial
with the northernmost segment of the active Northern Andes volcanic arc (Stern
2004), and no additional manifestations of active Andean volcanism are manifest
north of Río Dulce (ca. 5° 30’N). Additional discussion of the development and
migration of Mio-Pliocene arc segments with respect to the evolution of the Nazca
Plate along the Colombian Pacific margin will be presented in Sect. 5.4.4.2.
Miocene through Plio-Pleistocene occurrences of granitoid magmatic rocks
within Colombia’s eastern cordilleran system are widely spaced and volumetri-
cally minor, and current data does not permit the temporal or spatial correlation of
the granitoids within the geological context of the region or with respect to the
volumetrically much more abundant Neogene granitoids which are widespread
throughout the physiographic Central and Western Cordilleras. Additional discus-
sion of the Vetas-California, Paipa and Quetame granitoids will also be supplied in
the magmato-­tectonic analysis of the region presented in Sect. 5.4.4.2.

5.3.5.4  L
 ithogeochemical and Isotopic Characteristics of Oligocene
to Pliocene Granitoids

Latest Oligocene to Pliocene granitoids in the Colombian Andes include a texturally

diverse suite of medium to coarse-grained, holocrystalline, phaneritic equigranular
and porphyritic rock types. Detailed lithogeochemical information for the suite has
only recently become available. Historic data, including basic major element analy-
ses from limited localities, was presented by Alvarez (1983). More recently, various
authors have provided more complete analysis of the Miocene and Pliocene suite,
including samples from most of the important arc segments and porphyry clusters
throughout the Colombian Andes. Data from Gil-Rodríguez (2010), Leal-Mejía
(2011), Rodríguez and Zapata (2012), Lesage et al. (2013), Zapata and Rodríguez
(2013), Bissig et al. (2014), Gil-Rodríguez (2014) and Cruz et al. (2014) are included
in the present data set.
In terms of whole-rock lithogeochemistry, 150 samples of latest Oligocene to
Plio-Pleistocene holocrystalline and hypabyssal granitoids are presented herein
(Figs. 5.24, 5.25, 5.26, and 5.27). Some 43.3% (65 samples) of the sample set are
considered altered, considering element mobility criteria described in Sect. 5.3.1.2.
This is consistent with the observation that many of the altered samples were col-
lected in areas subject to hydrothermal alteration related to porphyry-style and/or
epithermal mineralization. Altered samples are represented by unfilled symbols in
the lithogeochemical plots. The sample set as a whole is considered to be represen-
tative of the major Miocene to Pleistocene arc segments of the Colombian Andes.
The Colombian latest Oligocene to Plio-Pleistocene granitoids are medium- to
high-K calc-alkaline in composition. Most of the main holocrystalline granitoids
and hypabyssal stocks and dikes are metaluminous, with exception of samples
of Miocene porphyritic granitoids of the Santander Massif, which exhibit clear
338 H. Leal-Mejía et al.

Fig. 5.24  Major element lithogeochemical plots for mid-Cretaceous through Eocene, Western Group
(oceanic) granitoids in the Colombian Andes. (a) AFM plot, curve after Irvine and Baragar (1971);
(b) K2O vs. SiO2 plot, boundary fields in grey as summarized by Rickwood (1989); (c) alumina satu-
ration plot after Barton and Young (2002); (d and e) MALI and Fe-index vs. SiO2 plots, respectively,
after Frost et al. (2001); (f) R1-R2 classification plot after De La Roche et al. (1980). Th tholeiite, C-A
calc-alkaline, Sh shoshonite, Gb No gabbro-norite, Ol-Gb (olivine-)gabbro, Gb Di gabbro-diorite, Di
diorite, Mz Di monzodiorite, Mz monzonite, To tonalite, Gd granodiorite
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 339

Fig. 5.25  Trace element and REE lithogeochemical plots for latest Oligocene and Miocene holo-
crystalline granitoids in the Colombian Andes. (a and b) Trace element and REE normalized spider
diagram plots; (c) granite discrimination Ta vs. Yb diagram after Pearce et  al. (1984). VAG
volcanic-­arc granites, syn-COLG syn-collisional granites, WPG within-plate granites, ORG ocean
ridge granites

peraluminous affinity, and some altered samples of both, holocrystalline and por-
phyritic granitoids, which have been shifted into the peraluminous field due to sub-
solidus hydrothermal alteration. With respect to the classification scheme of Frost
et  al. (2001), many of the Neogene granitoids, both holocrystalline plutonic and
hypabyssal porphyritic, record a tendency towards ferroan compositions (Figs. 5.24
and 5.26). We feel this is a reflection of the tendency of many of the analysed sam-
ples to contain observable quantities (accessory to +2 modal percent) of pyrite,
which would affect the Fe number, but does not signify Fe enrichment in the tech-
nical (i.e. tholeiitic) sense of the term (Irvine and Baragar 1971). This interpretation
is supported by the “calc-alkaline” trend for the same samples, revealed by the
AFM diagrams in Figs.  5.24a and 5.26a. Notwithstanding alteration of the Fe
number, the presence of pyrite does not affect the generally calcic to calc-alkalic,
metaluminous designation of these samples. A brief review of specific lithogeo-
chemical features of individual Oligocene to Pleistocene granitoid arc segments
and clusters is now presented.
340 H. Leal-Mejía et al.

Fig. 5.26  Major element lithogeochemical plots for latest Oligocene and Mio-Pliocene porphy-
ritic granitoids in the Colombian Andes. Major element lithogeochemical plots for mid-Cretaceous
through Eocene, Western Group (oceanic) granitoids in the Colombian Andes. (a) AFM plot, curve
after Irvine and Baragar (1971); (b) K2O vs. SiO2 plot, boundary fields in grey as summarized by
Rickwood (1989); (c) alumina saturation plot after Barton and Young (2002); (d and e) MALI and
Fe-index vs. SiO2 plots, respectively, after Frost et al. (2001); (f) R1-R2 classification plot after De
La Roche et al. (1980). Th tholeiite, C-A calc-alkaline, Sh shoshonite, Gb No gabbro-norite, Ol-Gb
(olivine-)gabbro, Gb Di gabbro-diorite, Di diorite, Mz Di monzodiorite, Mz monzonite, To tonalite,
Gd granodiorite
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 341

Fig. 5.27  Trace element and REE lithogeochemical plots for latest Oligocene and Miocene por-
phyritic granitoids in the Colombian Andes. (a and b) Trace element and REE normalized spider
diagram plots; (c) granite discrimination Ta vs. Yb diagram after Pearce et  al. (1984). VAG
volcanic-­arc granites, syn-COLG syn-collisional granites, WPG within-plate granites, ORG ocean
ridge granites

Lithogeochemistry

Latest Oligocene-Early Miocene Holocrystalline (Phaneritic) Granitoids

The Piedrancha-La Llanada-Cuembi trend batholiths: The Piedrancha-La Llanada
batholiths are represented by four unaltered samples. These samples are metalumi-
nous, medium-K calc-alkaline rocks of diorite to tonalite composition (Fig. 5.24).
Trace element spider diagram patterns for the samples of the Piedrancha-La Llanada
batholiths show arc-magmatism patterns with enrichment of HFSE with respect to
LILE and notable negative Ta-Nb and Ti anomalies. The REE patterns show mod-
erate REE contents (ΣREE  =  29.5–40.5  ppm) and no significant Eu anomalies
(Eu/Eu* = 0.95–1.06).
The Farallones-Páramo de Frontino-El Cerro trend batholiths: The Farallones
Batholith is represented by thirteen samples, five of them with evidence of alteration.
Samples of the Farallones Batholith are mostly high-K calc-alkaline metaluminous
with some samples straddling the limit between medium- and high-K calc-alkaline
342 H. Leal-Mejía et al.

fields. Notably, some unaltered samples exhibit a more alkaline (higher K content)
affinity, plotting in the Shoshonite field. Consequently, most of the unaltered sam-
ples show a tonalitic composition, and more alkaline rocks exhibit a monzo-gabbro
to monzonite composition. Arc-related patterns are revealed by trace element spider
diagrams. Chondrite-normalized REE diagrams show low to moderate REE con-
tents (ΣREE = 39.8–112.3 ppm) and no significant negative or positive Eu anoma-
lies (Eu/Eu* = 0.82–1.11).
Additional samples of Miocene holocrystalline granitoids hosted within Cañas
Gordas basement of the northern Western Cordillera, compiled primarily from
Rodríguez and Zapata (2012) and Zapata and Rodríguez (2013), include the El
Cerro Stock (nine samples, four altered), Páramo de Frontino (five samples, two
altered), Nudillales (three samples, two altered), Carauta (three samples, one
altered), La Horqueta (two samples, one altered), Morrogacho (one unaltered sam-
ple), Valle de Perdidas (one altered sample) and Río San Juan (one altered sample)
stocks and one altered sample of a minor mafic intrusion (Fig. 5.24). All of these
samples have major and minor oxides concentrations but no trace element/REE
geochemical data, with exception of one of the altered samples of the El Cerro
Stock, presented by Leal-Mejía (2011), which has complete multi-elemental analy-
sis. The samples of these granitoids are metaluminous, and, notably, most of them
exhibit a more alkaline character (shoshonite field potash values; alkali-calcic to
alkali after Frost et al. 2001), with respect to the calcic to calc-alkaline Piedrancha-La
Llanada and Farallones plutons. This more alkaline character is also reflected in
their highly variable composition, plotting from alkali gabbro through quartz mon-
zonite and syeno-diorite, in contrast to the more dioritic-tonalitic compositions
observed at Piedrancha-La Llanada and Farallones. The trace element pattern for
the sample of the El Cerro Stock reflects some element mobility associated with
alteration, with significant depletion in Th and U, whereas the chondrite-normalized
REE pattern shows relatively low REE contents (ΣREE = 34.55 ppm) and a very
subtle positive Eu anomaly (Eu/Eu* = 1.08) (Fig. 5.25).

Miocene to Plio-Pleistocene Hypabyssal Porphyry Granitoid Suites

The Patía-Upper Cauca hypabyssal porphyry suite: Middle to late Miocene Patía-­
Upper Cauca hypabyssal porphyry suite is well-represented by thirty-three samples
of porphyritic to weakly porphyritic granitoids. Nine of these samples exhibit evi-
dence of element mobility associated with alteration. The samples are metalumi-
nous in character with a medium-K calc-alkaline affinity and compositions ranging
from diorite and granodiorite. Arc-related geochemical signatures are revealed by
patterns in the trace element spider diagram, and chondrite-normalized REE dia-
gram patterns indicate moderate to high REE contents (ΣREE = 27.27–130.98 ppm).
Slight Eu anomalies varying between slightly negative and slightly positive (Eu/
Eu* = 0.87–1.18) are recorded. HREE patterns are relatively flat ((Gd/Yb)N = 0.89–
4.48), with concentrations between two and twenty times chondrite (Fig. 5.27).
The Middle Cauca hypabyssal porphyry suite: Late Miocene hypabyssal por-
phyry granitoids of the Middle Cauca are well-represented by 38 samples, 19 of
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 343

them with evidence of element mobility related to alteration. The unaltered samples
define a highly differentiated compositional trend ranging from monzo-gabbro to
granodiorite and tonalite. They are metaluminous in character with generally
medium- to high-K calc-alkaline affinity. The suite mostly plots within the calc-­
alkalic field of Frost et al. (2001). A clear effect of hydrothermal alteration, pushing
the suite into the alkali-calcic and alkalic fields (shoshonite), is observed (Fig. 5.26).
Trace element spider diagrams of unaltered samples confirm the magmatic arc-­
related geochemical signature with evident negative Ta-Nb and Ti anomalies and
significant depletion in Th and U.  Chondrite-normalized REE diagram patterns
show moderate to relatively high REE contents (ΣREE = 36.48–141.05 ppm). Eu
anomalies vary between slightly negative to moderately positive (Eu/Eu* = 0.87–
1.21). HREE concentrations are variable, between two and twenty times the con-
centration of chondrites (Fig. 5.27).
The Cajamarca-Salento hypabyssal porphyry suite: Late Miocene hypabyssal
porphyry granitoids of the Cajamarca-Salento region are represented by twenty
samples, ten of which show evidence of alteration. The unaltered samples are meta-
luminous in character with medium- to high-K calc-alkaline affinity, whereas the
altered samples are shifted towards the peraluminous field, probably due to the
effects of alteration. Composition ranges between diorite and granodiorite-tonalite.
Trace element spider diagram patterns of unaltered samples confirm the magmatic
arc-related geochemical signature with negative Ta-Nb and Ti anomalies. Chondrite-­
normalized REE diagram patterns show moderate to relatively high REE contents
(ΣREE = 68.33–127.71 ppm) with slightly negative to slightly positive Eu anoma-
lies (Eu/Eu*  =  0.80–1.12). HREE concentrations are variable, between four and
fifteen times the concentration in chondrites (Fig. 5.27).
The Río Dulce hypabyssal porphyry suite: Pliocene to Pleistocene hypabyssal
porphyry granitoids of the Río Dulce region are represented by four samples,
although three of them show evidences of alteration. The samples plot in the meta-
luminous field with exception of the most altered sample which plots in the peralu-
minous field and are of medium-K calc-alkaline affinity with tonalite compositions.
Arc-magmatism geochemical signatures are also revealed by trace element spider
diagram patterns. Chondrite-normalized REE diagram patterns show moderate to
relatively high REE contents (ΣREE = 84.61–127.66 ppm) with steep decreasing
slopes ((La/Yb)N = 8.33–14.97) with no significant Eu anomalies (Eu/Eu* = 0.80–
1.12). HREE concentrations are around ten times the concentration in chondrites
(Fig. 5.27).
The Santander Massif hypabyssal porphyry suite: Middle to late Miocene hyp-
abyssal porphyry granitoids of the Santander Massif, especially those clustered
around Vetas-California, are well-represented by twelve samples, six of them
revealing element mobility due to alteration. The unaltered samples show highly
evolved geochemical features, including a peraluminous character and high-K calc-­
alkaline affinity. Potassium enrichment is recorded in the altered samples with K2O
values >4.0 wt%. The Vetas-California suite ranges compositionally from granodio-
rite to tonalite, with some samples plotting close to the limit with the quartz monzo-
nite and granite fields. Trace element spider diagram patterns of unaltered samples
344 H. Leal-Mejía et al.

confirm the magmatic arc-related geochemical signature, with evidently negative

Ta-Nb and Ti anomalies. Chondrite-normalized REE diagram patterns show rela-
tively high REE contents (ΣREE = 104.87–185.10 ppm) and flat to slightly negative
Eu anomalies (Eu/Eu*  =  0.77–0.97). Significant enrichment in LREE at Vetas-­
California, with respect to other Miocene hypabyssal porphyry suites (up to 110
times chondrite), is observed, accompanied by a relative depletion of MREE and
HREE (about ten times chondrite; Fig. 5.27).
Paipa-Iza and Quetame Granitoids: Available major element lithogeochemistry
for Pliocene-Pleistocene volcanic rocks in the Eastern Cordillera includes one sam-
ple of a rhyodacite porphyry at the Quetame region (Ujueta et al. 1990) and fifteen
samples of Paipa volcano products described as alkali (k-feldspar) rhyodacites and
trachytes and calc-alkaline rhyolites (Cepeda and Pardo-Villaveces 2004; Pardo
et al. 2005b). Despite evident element mobility associated to hydrothermal/volcanic
alteration and/or weathering of samples from Paipa (only four out of fifteen samples
seem to be relatively fresh), some general trends can be observed and compared the
late Miocene-Pleistocene magmatism observed to the north in the Santander Massif
region (Leal-Mejía 2011; Bissig et al. 2014; Cruz et al. 2014) (Fig. 5.26). The less
altered/weathered Paipa volcanic rocks exhibit higher silica contents (SiO2 = 68.2–
71.6  wt%) when compared to the Santander Massif porphyries (SiO2  =  63.4–
66.3 wt%), whereas the Quetame volcanic rock returned 66.4 wt% SiO2. Alumina
content in samples from Paipa and the Santander Massif porphyries is similar
(Al2O3  =  16.2–18.1  wt% and Al2O3  =  17.2–17.7  wt%, respectively), whilst the
Quetame volcanics sample shows significantly lower values (Al2O3 = 14.8 wt%).
Fe2O3, MgO, CaO and TiO2 values for the Santander Massif porphyries are signifi-
cantly higher (Fe2O3 = 2.5–4.9 wt%, MgO = 0.7–0.12 wt%, CaO = 2.4–4.0 wt% and
TiO2 = 0.3–0.5 wt%) than in the Paipa volcanics (Fe2O3 = 0.8–2.0 wt%, MgO = 0.02–
0.5 wt%, CaO = 0.2–0.8 wt% and TiO2 = 0.1–0.3 wt%). Fe2O3 and CaO values for
the Quetame volcanics sample (Fe2O3 = 2.9 wt% and CaO = 3.2 wt%) are compa-
rable to those of the Santander Massif porphyries, whereas TiO2 value (0.27 wt%)
is more in the range of the Paipa samples. The MgO value for the Quetame volca-
nics sample (2.0 wt%) is significantly higher than MgO values observed for both
Santander Massif porphyries and Paipa volcanics. Na2O values are slightly higher in
the Paipa volcanics when compared to the Santander Massif porphyries (Na2O = 5.9–
6.8 wt% and Na2O = 4.0–5.1 wt%, respectively), whereas K2O values are compa-
rable for both sample sets (K2O  =  3.4–3.7  wt% and K2O  =  3.0–3.8  wt%,
respectively).
All of the Paipa-Quetame samples show a well-defined calc-alkaline trend in the
AFM diagram (Fig. 5.26a), with samples from Paipa volcanics being slightly more
evolved/fractionated with respect to the Santander Massif porphyries. In addition,
all of the less altered samples plot in the high-K calc-alkaline field (Fig. 5.26b). The
presence of abundant K-feldspar phenocrysts in both Quetame and Paipa volcanic
rocks (Ujueta et al. 1990; Pardo et al. 2005b), as well as the Santander Massif por-
phyries (Mantilla et al. 2009; Cruz et al. 2014), suggests a more alkaline affinity for
these rocks; however, major element lithogeochemistry indicates that these rocks
are mostly metaluminous to weakly peraluminous in nature (Fig. 5.26c), and do not
plot in the peralkaline field. At Paipa, post-crystallization hydrothermal alteration
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 345

includes abundant secondary silica and pyritic sulfidation. In either case, these fac-
tors significantly reduce the confidence level of any interpretations and conclusions
drawn with respect to these granitoid suites, especially those derived based soley
upon major element lithogeochemistry.
Classification diagrams for feldspathic igneous rocks proposed by Frost et  al.
(2001) and Frost and Frost (2008) (Fig. 5.26d, e) clearly differentiate the magnesian,
oxidized, calcic to calc-alkalic Santander Massif porphyries and alkali-calcic Quetame
sample, from the ferroan (reduced) alkalic samples of the Paipa suite. Moreover, cal-
culation of the alkalinity index (AI) and the feldspathoid silica-­saturation index (FSSI)
proposed by Frost and Frost (2008) returned positive values for both indexes
(AI = 0.5–6.7, FSSI = 13.5–47.9), which confirm a silica-saturated metaluminous/
peraluminous character for these quartz-bearing rocks rather than a peralkaline char-
acter. The R1-R2 classification plot for plutonic rocks (Fig. 5.26f) also demonstrates
the alkalic affinity for rocks from the Paipa volcanics (alkali granite/quartz syenite)
with respect to more calc-alkaline rocks of the Santander Massif porphyries (grano-
diorite/tonalite) and the Quetame volcanics (quartz monzonite).

Whole-Rock Lithochemistry Summary and Discussion for the Oligocene-­

Pliocene Granitoids
Whole-rock, trace and REE data for the majority of the latest Oligocene through
Plio-Pliestocene holocrystalline and hypabyssal porphyritic granitoids of the west-
ern Colombian Andes, regardless of age, plot in consistent, narrow ranges, with
variations in the lithogeochemical composition of most samples attributable to the
effects of late or post-crystallization alteration of the alkali contents. The western
Colombian intrusive suites are metaluminous, calcic to calc-alkaline plutons with
typical arc-related trace element patterns and flat, unfractionated REE patterns lack-
ing well-developed Eu anomalies, considered typical of relatively undifferentiated,
primitive, subduction-related granitoids emplaced within oceanic crust, as repre-
sented by the Romeral mélange and Cañas Gordas and Dagua terranes, which form
the basement complexes to the western Colombian Andes. The trend towards alkali
enrichment observed in some of the small, isolated plutons associated with the ca.
12–10  Ma Farallones-El Cerro trend may be explained by late magmatic
K-metasomatism, as described petrographically by Escobar and Tejada (1992),
without invoking additional magmatic source regions or differentiation processes.
Similar arc-related major, trace and REE compositions are recorded by the
Cajamarca-Salento and Río Dulce suites, despite the observation that they are
hosted continent-ward, within Cajamarca-Valdivia Terrane basement. Cajamarca-­
Valdivia, however, is not of typically “continental” composition (Restrepo-Pace
1992; Cediel et  al. 2003; Cediel 2011) and in this context is considered to have
preserved the more primitive bulk compositions reflected in the analyses of the
Cajamarca-Salento and Río Dulce porphyry suites.
Conversely, granitoid porphyries and volcanic rocks at Vetas-California, Paipa-­
Iza and Quetame were emplaced within composite mid-Proterozoic-early Paleozoic
continental metamorphic basement within the Santander Massif and Eastern
Cordillera. All of these suites represent isolated, low-volume outliers of granitoid
346 H. Leal-Mejía et al.

rocks emplaced significantly to the east and north of the principle Miocene granit-
oid suites of the Western and Central Cordilleras and of the Pleistocene to recent
Northern Andean volcanic arc. Based upon the lithogeochemical data provided, the
Vetas-California and Quetame granitoids conform to a magnesian, calcic to alkali-­
calcic suite, whilst the Paipa-Iza granitoids are of ferroan, alkali affinity (Frost et al.
2001), and in this context, as a whole, the Santander-Eastern Cordilleran granitoids
record a bimodal distribution (Fig. 5.26). Notwithstanding, the data are limited and
geographically disperse, and do not yet permit interpretation of the potential petro-
genetic relationships between these outlier suites. Within the context of the compo-
sitional trends of the entire Oligo-Miocene to Pliocene granitoid suite presented
herein, however, the Vetas-California, Paipa-Iza and Quetame granitoids provide
the most consistently differentiated/evolved lithogeochemistry, especially in terms
of alkalinity, aluminium indices and trace and REE patterns. We interpret these
observations to reflect greater degrees of crustal interaction and assimilation/con-
tamination from the thick continental basement of the Santander Massif and
Chicamocha Terrane (Figs. 5.2 and 5.22) vs. the more primitive basement composi-
tions provided by the Cajamarca-Valdivia Terrane and the oceanic terranes of the
Western Tectonic Realm, which host the Oligo-Miocene-Pliocene granitoids of the
Central and Western Cordilleras.

Isotope Geochemistry

Sr-Nd Isotope Geochemistry, Results, Summary and Discussion

Available Sr-Nd isotope data for the entire suite of holocrystalline phaneritic and
hypabyssal porphyritic granitoid rocks of latest Oligocene to Pliocene age, with the
exception of the Paipa-Iza and Quetame occurrences, are presented in Fig. 5.28. All
of the arc segments and porphyry clusters documented above, albeit most with a
limited number of samples, are represented within the data set. In addition, for com-
parative purposes, we include Sr-Nd isotope data published for mantle and crustal
xenoliths contained within Plio-Pleistocene garnetiferous pyroclastic rocks occur-
ring at Mercaderes (Weber et  al. 2002; Rodríguez-Vargas et  al. 2005), spatially
coincident with late Miocene hypabyssal granitoid porphyry within our Upper
Cauca-Patía arc segment, which we interpret to establish a representative range of
values for the Sr-Nd composition of the mantle/crust beneath the SW Colombian
margin during the Neogene.
As displayed in Fig. 5.28, with the exception of the Vetas-California porphyry
cluster, essentially the entire latest Oligocene to Plio-Pleistocene suite of granitoids
considered within this study plot within a narrow vertical field, originating within
the mantle array and tending towards increasing εNd(t) values. 87Sr/86Sr(i) values
ubiquitously plot within a narrow range (87Sr/86Sr(i) = 0.70395 to 0.70506), with very
little scatter, with only two samples of tonalite-granodiorite from the Piedrancha-­
Cuembí suite plotting outside of this range (Fig.  5.28). The Piedrancha-Cuembí
suite is in fact the most scattered of the Miocene granitoid data sets with respect to
Sr-Nd isotope systematics, an observation we attribute to the widespread carbo-
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 347

Fig. 5.28  Sr-Nd and Pb isotope plots for latest Oligocene through Mio-Pliocene holocrystalline
and porphyritic granitoids in the Colombian Andes. Data for mantle and deep crustal xenoliths
from the Cauca-Patía area, and volcanic rocks of the Combia Fm. from the Middle Cauca area, are
plotted for reference. See text for discussion

nitization the suite has suffered, a process which could affect the 87Sr/86Sr(i) ratios of
these rocks. Notwithstanding, the Sr-Nd data sets for the Miocene-Plio-Pleistocene
granitoids of the Cajamarca-Valdivia Terrane and Western Tectonic Realm plot
essentially co-spatial with the isotopic signatures provided by the Mercaderes
mantle xenoliths.
The Sr-Nd isotope composition of the Vetas-California granitoid porphyries,
revealed in Fig.  5.28, is commensurate with the lithogeochemical compositions,
trends and conclusions outlined in Sect. 5.3.5.4. The Vetas-California suite depicts
a Sr-Nd compositional trend of decreasing εNd(t) values with increasing 87Sr/86Sr(i),
originating within the central mantle array and evolving towards crustally influ-
enced values. Additional discussion of the isotopic evolution of the latest Oligocene
through Plio-Pleistocene granitoid suite will be provided following presentation of
Pb isotope data below.

Pb Isotope Geochemistry, Results and Summary and Discussion

Available Pb isotope geochemical results for latest Oligocene to Plio-Pleistocene
granitoids of the Colombian Andes are presented in Fig. 5.28. For comparative
purposes Pb isotope data published for crustal xenoliths contained within the
348 H. Leal-Mejía et al.

Mercaderes garnetiferous pyroclastic rocks (Weber et  al. 2002) are also shown.
Figure  5.28 demonstrates that the Pb isotope composition of the entire latest
Oligocene through Plio-Pleistocene granitoid data set plots within a clustered range
with very little scatter (206Pb/204Pb  =  18.79–19.39, 207Pb/204Pb  =  15.62–15.76 and
208
Pb/204Pb = 38.68–39.21), especially evident on the 207Pb/206Pb vs. 206Pb/204Pb plot.
The latest Oligocene-­Plio-­Pleistocene data is notably well grouped, forming a tight,
steep array between the Orogene and upper crust lead evolution curves of the
Plumbotectonics model of Zartman and Doe (1981), in marked contrast to typically
more shallow arrays provided by the data sets for the latest Triassic-Jurassic granit-
oids (Fig. 5.14) and mid-­Cretaceous-­Eocene granitoids (Fig. 5.21). With the excep-
tion of three samples of granitoid porphyries from the Middle Cauca region, data of
the latest Oligocene-­Plio-­Pleistocene granitoids plots co-spatial with the range
established by the Mercaderes crustal xenoliths, and a model involving the mixing
of relatively homogenous, less radiogenic, mantle-derived magmas (as supported by
the Sr-Nd data) with a more radiogenic Pb source range, as established in the
Mercaderes crustal xenoliths, is invoked to explain the observed latest Oligocene-
Plio-­Pleistocene range of Pb compositions. The three samples of late Miocene por-
phyritic granitoids occur in close proximity within the central Middle Cauca belt
(Jericó, Venecia and Titiribí clusters). In terms of Sr-Nd isotopic composition, these
samples all plot well within the mantle array and within the range of the majority of
the Miocene-Pliocene granitoid porphyries from other regions (Fig. 5.28). In view
of this, we interpret this small population to represent the mixing of mantle-derived
Pb compositions similar to those of the granitoids from other regions, with a more
radiogenic, crustal-sourced Pb of a somewhat distinct composition to that defined
by the Mercaderes crustal xenoliths. This is in keeping with the observation that the
Romeral tectonic zone, which forms basement to the entire suite of Miocene gran-
itoid porphyries, along both the Upper Cauca-Patía and southern Middle Cauca
belts, is a heterogeneous lithotecton comprised of a mix (mélange) of rock types of
differing age and continental, peri-cratonic and oceanic provenance. In this respect,
the relatively homogenous appearance of Pb isotope compositions for the latest
Oligocene-Plio-Pleistocene suite may well be a function of the relatively few locali-
ties for which Pb isotope analyses are available, especially given the restricted dis-
tribution of crustal xenoliths such as those documented at Mercaderes.

Synthesis and Conclusions of Lithogeochemistry and Sr, Nd and Pb Isotope

Geochemistry for Oligocene-Pliocene Granitoids
Whole-rock lithogeochemistry, including major, minor, trace element and REE
data, combined with analyses documenting the Sr-Nd and Pb isotope composition
of latest Oligocene to Miocene and Plio-Pleistocene holocrystalline phaneritic and
hypabyssal porphyritic rocks from numerous localities within the northeastern, cen-
tral and western Colombian Andes permits the consistent characterization of the
entire suite as subduction-related, Cordilleran (Frost et  al. 2001), volcanic arc
(Pearce et al. 1984) or calc-alkaline (Barbarain 1999) granitoids, formed within a
continental arc setting. The principal host domains for the Neogene granitoid suite
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 349

are the oceanic basement terranes of the Western Tectonic Realm (Romeral, Cañas
Gordas, Dagua); however, important occurrences are also observed within the
Cajamarca-Valdivia Terrane, within Colombia’s physiographic Central Cordillera,
essentially coaxial with the modern-day Northern Andes volcanic arc (Stern 2004).
Isolated Miocene to Plio-Pleistocene granitoid outliers are also observed farther
east, at Vetas-California in the Santander Massif and at Paipa-Iza and Quetame, in
the Eastern Cordillera. The tectonic assembly of the region was essentially com-
plete at the time of emplacement of each arc segment or granitoid cluster, and all of
the latest Oligocene through Plio-Pleistocene granitoid suites may be considered
autochthonous with respect to the tectonic evolution of the Colombian Andes.
Review of the whole-rock lithogeochemical and isotope data for the latest
Oligocene to Plio-Pleistocene suite demonstrates remarkably consistent composi-
tional trends, despite the varied nature of the basement complexes into which the
granitoids were emplaced. Rare-earth element and isotopic trends for the majority
of the suite suggest limited degrees of magmatic fractionation and isotopic exchange
at crustal levels, consistent with the rapid emplacement of subduction-related,
mantle-­derived melts, facilitated by the preexisting structural architecture, which
includes various paleo-sutures, as exemplified by the Palestina, Romeral and Cauca
fault systems. The most evolved granitoids within the latest Oligocene-Pliocene
suite include those of the Vetas-California area, which have evidently undergone
somewhat greater degrees of fractionation, assimilation and/or isotopic exchange
with the thick continental basement exposed within the Santander Massif. Further
discussion of the nature, distribution and tectonic evolution of Neogene granitoid
magmatism in the Colombian Andes is presented in Sect. 5.4.4.2.

5.4  Phanerozoic Tectono-Magmatic Evolution

of the Colombian Andes

Aspden et al. (1987) presented a temporal-spatial analysis of granitoid magmatism

in the Colombian Andes based upon published K-Ar and Rb-Sr radiometric age
dates. They defined five episodes of subduction-related granitoid magmatism,
including the Triassic, Jurassic, Cretaceous, Paleogene and Neogene, and offered a
schematic interpretation of the tectonic framework for each episode within the gen-
eralized tectonic configuration of the entire Northern Andean region. These same
authors identified various factors which influenced the nature, distribution and
geometry of Meso-Cenozoic subduction-related granitoid arcs in Colombia. These
factors included oblique plate convergence, low-angle subduction including changes
in the angle of the subducting oceanic plate and the role of aseismic features and the
accretion of allochthonous components contained within the oceanic domain along
the Pacific margin, in the development of, and hiatuses in, the subduction process.
More recently, kinematic models for the tectonic and structural evolution of the
Northern Andes (e.g. Cediel et al. 2003; Kennan and Pindell 2009) and Caribbean
Plate (e.g. Pindell and Kennan 2001; Nerlich et al. 2014), constructed at a similar
350 H. Leal-Mejía et al.

scale to the work of Aspden et  al. (1987), have independently confirmed and
expanded upon many of the assertations presented by these early authors.
The increased resolution and widespread distribution of the present-day  U-Pb
age, lithogeochemical and isotopic database, when combined with updated con-
cepts for the geological evolution of the Colombian Andes, permit a reassessment
and more detailed reconstruction of the tectono-magmatic evolution of the region
than that afforded in the Aspden et al. (1987) analysis. In the following section, we
present sequential reconstructions detailing the Phanerozoic tectono-magmatic evo-
lution of granitoids in the Colombian Andes, based upon the major magmatic epi-
sodes defined by the U-Pb (zircon) age date, lithogeochemical and isotopic database,
as described in detail in the foregoing sections. Annotated schematic illustrations
and time-space analyses for the early Paleozoic through middle-late Triassic, latest
Triassic through Jurassic, early to middle Cretaceous, middle Cretaceous through
Eocene and latest Oligocene through Miocene-Pliocene are provided in Figs. 5.29,
5.30, 5.31, 5.32, 5.33, 5.34, 5.35, and 5.36. Descriptive text pertaining to each time
period highlights the temporal and spatial evolution of granitoid magmatism within
the litho-tectonic and morpho-structural development of the region, prior to, leading
up to and during the Meso-Cenozoic Northern Andean orogeny.
In terms of nomenclature pertaining to the various phases of tectonic develop-
ment of the Colombian Andes, we have adhered to terminology used in the work of
Cediel et al. (1994), Cediel and Cáceres (2000), Cediel et al. (2003), Cediel (2011)
and Cediel (2018). This work provides a coherent and sufficiently detailed frame-
work, at an appropriate temporal and spatial scale for the Colombian Andes, span-
ning the Proterozoic to Mio-Pliocene, within which to integrate the periods of
granitoid magmatism defined herein. Table  5.2 provides a summary of tectonic
events recorded within the Colombian Andes, as described in detail in the works of
the previously cited authors.

5.4.1  P
 re-northern Andean Orogeny Granitoids: Early
Paleozoic Through Mid-Late Triassic

Our study has identified three episodes of granitoid magmatism recorded within
the Colombian Andes, which were generated and emplaced within the context of
pre-­Northern Andean Orogeny tectono-magmatic development. These episodes
include the early Paleozoic (ca. 485–439 Ma), Carboniferous (ca. 333–310 Ma) and
Permo-­Triassic (ca. 288–223 Ma). With respect to all three episodes, the granitoid-
magmatic record is relatively sparse and punctually developed, especially when
compared to wide-spread and volumetrically exponential magmatism developed
during the Meso-Cenozoic. Indeed, we remind the reader that the full extent of all
three early Phanerozoic magmatic events has yet to be fully defined, based upon
presently available radiometric age dates vs. the resolution of existing field-based
geological mapping, which doesn’t yet recognize some of the important early
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 351

Fig. 5.29  Major litho-tectonic elements and interpreted tectonic setting of NW Colombia and
surrounding area during the late Triassic. The spatial relationship between early Paleozoic,
Carboniferous and Permo-Triassic granitoids exposed in the Colombian Andes is shown.
(Granitoid shapes modified after Cediel and Cáceres 2000; Gómez et  al. 2007; Gómez et  al.
2015a. Litho-­tectonic terrane and fault nomenclature modified after Cediel et al. 2003. See text
for additional details)
 352

Fig. 5.30  Time-space analysis of early Paleozoic through mid-late Triassic granitoids in the Colombian Andes and surrounding region, in relation to tectonic
framework, major litho-tectonic elements and orogenic events. The age and nature of individual granitoid intrusive suites of the time period are indicated. The
profile contains elements projected onto a ca. NW–SE line of section through west-central Colombia. (Litho-tectonic terrane and fault nomenclature modified
H. Leal-Mejía et al.

after Cediel et al. (2003). See text for additional details)

5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 353

Fig. 5.31  Major litho-tectonic elements and interpreted tectonic setting of NW Colombia and
surrounding area during the latest Triassic through Jurassic, highlighting the spatial-temporal
relationship between the major Jurassic arc segments exposed in the Colombian Andes. (Granitoid
shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al. 2015a. Litho-­
tectonic terrane and fault nomenclature modified after Cediel et al. 2003. See text for additional
details)
 354

Fig. 5.32  Time-space analysis of latest Triassic through Jurassic granitoids in the Colombian Andes and surrounding region, in relation to tectonic framework,
major litho-tectonic elements and orogenic events. The age and nature of granitoid intrusive suites of the same time period are indicated. The profile contains
H. Leal-Mejía et al.

elements projected onto a ca. NW–SE line of section through west-central Colombia. (Litho-tectonic terrane and fault nomenclature modified after Cediel et al.
2003. See text for additional details)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 355

Fig. 5.33  Major litho-tectonic elements and interpreted tectonic setting of NW Colombia and sur-
rounding area during the early to mid-Cretaceous. Note the absence of significant volumes of
granitoid rocks within the exposed geological record of the Colombian Andes for this time period.
(Lithological unit shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al.
2015a. Litho-tectonic terrane and fault nomenclature modified after Cediel et al. 2003. See text for
additional details)
356 H. Leal-Mejía et al.

Fig. 5.34  Major litho-tectonic elements and interpreted tectonic setting of NW Colombia and sur-
rounding area during the mid-Cretaceous to Eocene. A schematic depiction of the temporal-spatial
relationship between Eastern Group (continental) granitoids and Western Group (oceanic) granit-
oids is presented. (Granitoid shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007;
Gómez et  al. 2015a. Litho-tectonic terrane and fault nomenclature modified after Cediel et  al.
2003. See text for additional details)
 5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic…

Fig. 5.35  Time-space analysis of early Cretaceous through Eocene granitoids in the Colombian Andes and surrounding region, in relation to tectonic frame-
work, major litho-tectonic elements and orogenic events. The age and nature of granitoid intrusive suites of the same time period are indicated. The profile
contains elements projected onto a ca. NW–SE line of section through west-central Colombia. (Litho-tectonic terrane and fault nomenclature modified after
Cediel et al. 2003. See text for additional details)
357
358 H. Leal-Mejía et al.

Fig. 5.36  Major litho-tectonic elements and interpreted tectonic setting of NW Colombia and sur-
rounding area during the latest Oligocene through Mio-Plio-Pleistocene. The near modern-day
tectonic assembly of the region by the Pliocene is observed. The active Galeras-Puracé-Huila-Ruíz
volcanoes mark the trend of the modern-day calc-alkaline arc axis in the Colombian Andes.
(Granitoid shapes modified after Cediel and Cáceres 2000; Gómez et al. 2007; Gómez et al. 2015a.
Litho-tectonic terrane and fault nomenclature modified after Cediel et al. 2003. See text for addi-
tional details)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 359

Table 5.2  Summary of Colombian tectono-magmatic episodes and regional tectonic comparisons.
Litho-tectonic and morpho-structural units as defined by Cediel et  al. (2003) and indicated in
Fig. 5.2
Colombian Distribution of
tectonic phase Colombian
Time period (with age of Regional temporal granitoids
(magmatic associated comparatives (basement Tectonic regime
episode) granitoids) (Orogenies) domain) (Colombia)
Pre-Cambrian Orinoco Grenville Orogeny, Granulite Belt Collisional,
Orogeny North America in MSP (SNSM, Compressional,
(ca. Santander Accretionary
1.2–0.9 Ga) massif) and
Garzón massif
Early Quetame Caparonesis (Venezuela), MSP (Santander Collisional,
Paleozoic Orogeny Ocloy (Ecuador-Perú), massif, SNSM), Compressional,
(ca. Famantinian (Argentina), Floresta and Accretionary
485–473 Ma) Taconian-Acadian Quetame Extension after
(N. America-N. Europe) massifs, CTR ca.465 Ma
(CA-VA,
Central
Cordillera)
Carboniferous Bolívar – CTR (CA-VA, Extensional,
Aulacogen, Otú rift, rifting (failed)
(early phase) Northern
(ca. Central
333–310 Ma) Cordillera)
Permian-early Permo-­ Gondwanide Orogeny, CTR (mostly Compressional,
Triassic Triassic Alleghanian-Appalachian CA-VA, Central transpressional?
tec.-thermal Orogeny, N. America Cordillera),
event (ca. SNSM
290–250 Ma)
Mid-late Bolívar – CTR (CA-VA, Extensional,
Triassic Aulacogen Central rifting
(intermediate Cordillera),
phase) MSP (Santander
(ca. massif, SNSM),
250–216 Ma) Garzón massif
(continued)
360 H. Leal-Mejía et al.

Table 5.2 (continued)
Colombian Distribution of
tectonic phase Colombian
Time period (with age of Regional temporal granitoids
(magmatic associated comparatives (basement Tectonic regime
episode) granitoids) (Orogenies) domain) (Colombia)
Latest Bolívar – CTR (San Lucas Extensional, slab
Triassic-­ Aulacogen range, Central rollback
Jurassic (late phase) Cordillera and
(ca. CA-VA
210–146 Ma) (Segovia
Batholith), MSP
(Santander
massif, SNSM),
Garzón massif
Early Bolívar – No significant Extensional,
cretaceous Aulacogen Granitoids rifting
(culminant (Valle Alto Rift)
phase)
Mid-­ Early Andean Orogeny CTR (CA-VA, Transpressional,
cretaceous-­ Northern (Peruvian and Incaic eastern group collisional,
Eocene Andean continental
phases), Peltetec melange accretionary
Orogeny (Ecuador), Laramide andgranitoids),
(ca. Sevier Orogenies, NorthSNSM, WTR
100–42 Ma) America (western group
CCOP/CLIP
gtoids)
Earliest Late Northern Late Andean Orogeny, WTR, RM, Oblique to
Oligocene-­ Andean Perú (Quecha phase), late CTR (CA-VA), orthogonal
Mio-­Pliocene Orogeny northern Andean MSP (Santander compression,
(ca. Orogeny, Ecuador massif), EC collisional,
24–0.4 Ma) Accretionary,
Nazca plate
subduction,
back-arc
extensión?
MSP Maracaibo Sub-plate, SNSM Sierra Nevada de Santa Marta, CTR Central Continental Realm,
CA-VA Cajamarca-Valdivia Terrane, WTR Western Tectonic Realm, CCOP/CLIP Caribbean-­
Colombian Oceanic Plateau/Caribbean Large Igneous Province, RM Romeral Melange, EC
Eastern Cordillera

Phanerozoic granitoid suites which constitute the region as a whole (e.g. early
Paleozoic and Carboniferous granitoids of the Cajamarca-Valdivia Terrane). In
addition, we emphasize that most of the early Phanerozoic (meta-)granitoids are
deeply eroded, deformed and metamorphosed and have been subject to a complex
series of tectono-magmatic events following their emplacement and spanning the
Meso-Cenozoic. Within this framework, and notwithstanding, the early Paleozoic,
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 361

Carboniferous and Permo-Triassic granitoids constitute the (albeit limited) principle

magmatic record for almost 300 million years, that is, over half of the Phanerozoic
tectonic and geologic record of the Colombian Andes.
In consideration of the above, we suggest that the interpretation of detailed tec-
tonic frameworks for the early Paleozoic, Carboniferous and Permo-Triassic granit-
oids in Colombia (and the Northern Andean region in general) is a complex affair.
This is exemplified by the observation that published tectono-magmatic models for
early Phanerozoic granitoid magmatism in the Colombian Andes, as derived from
recent integrated lithogeochemical and isotopic studies (see detailed reviews and
summaries presented by Cochrane (2013), Cochrane et al. (2014a), Van der Lelij
(2013), Van der Lelij et al. (2016) and Spikings et al. (2015), are presented within
highly schematic global-scale paleo-geographic reconstructions, which are contro-
versial (Van der Lelij et al. op. cit.); Cochrane et al. 2014a; Spikings et al. op. cit.)
and often difficult to reconcile at scales applicable to the litho-tectonic and morpho-­
structural units comprising the Colombian Andes.
In the following discussion of Colombian early Phanerozoic tectono-magmatic
development, we have opted to forgo large-scale and schematic paleo-geographic
and tectonic reconstructions, which may be found in well-versed and readily acces-
sible sources (e.g. Weber et  al. 2007; Cochrane et  al. 2014a; Van der Lelij et  al.
2015; Spikings et  al. 2015). Figure  5.29 depicts the actual distribution of early
Paleozoic, Carboniferous and Permo-Triassic granitoids, as modified from the exist-
ing regional geological map base (Cediel and Cáceres 2000; Gómez et al. 2015a).
The granitoids are depicted within the context of their host basement complexes,
and the figure is annotated with information pertaining to the petrogenesis, tectonic
environment and tectonic evolution of the region, as derived from the information
sources utilized in diagram construction. No attempt at paleo-geographic recon-
struction with respect to the distribution of granitoids from the various age group-
ings has been initiated. The early Phanerozoic tectono-magmatic evolution of the
region is additionally summarized in time-space format, presented in Fig. 5.30.

5.4.1.1  T
 ectonic Framework for Early Paleozoic Granitoids:
The Quetame Orogeny and Early Bolívar Aulacogen

During the latest Proterozoic to early Paleozoic, the composite basement of the
paleo-Andean continental region in Colombia was comprised of the western margin
of the Guiana Shield (Amazon Craton), the >ca. 1.2 Ga Chicamocha Terrane paleo-­
continental allochthon and an intervening belt of mid-Proterozoic (ca. 1.2–0.9 Ga)
granulite-grade metamorphic rocks, petrogenetically dominated by recycled early-­
mid-­Proterozoic continental crust (Fig. 5.29; see Sect. 5.2.1). This assemblage com-
prised the subsiding basement to thick deposits of autochthonous marine and
epicontinental sediments of Vendian and Cambrian(?), late Ordovician, Silurian,
Devonian and Carboniferous to Permian age (Cediel et al. 1994; Silva et al. 2005).
In Colombia, early Paleozoic supracrustal sequences underwent Cordilleran defor-
mation and regional, Barrovian-type, sub-greenschist to amphibolite-grade
362 H. Leal-Mejía et al.

metamorphism (e.g. Goldsmith et al. 1971; Ward et al. 1973; Restrepo-Pace 1995),
during what has been referred to in Colombia as the Quetame Orogeny (Cediel and
Cáceres 2000; Cediel et al. 2003). Within the Northern Andean region, this event
may be compared with the Caparonensis Orogeny in Venezuela, the Ocloy Orogeny
in Ecuador and Perú as well as the Famantinian Orogeny of northern Argentina and
the Taconian-Acadian and Caledonian orogenies, of North America and Northern
Europe, respectively. Within Colombia’s Eastern Cordilleran system, early Paleozoic
granitoids associated with this orogenic framework are located within the Santander,
Floresta and Quetame massifs (Figs. 5.29 and 5.30).
The composite geologic, radiometric age date, lithogeochemical and isotopic
database for the early Paleozoic (e.g. Cediel et al. 1994; Cediel and Cáceres 2000;
Horton et al. 2010; Leal-Mejía 2011; Mantilla et al. 2012; Van de Lelij 2013) per-
mits an initial understanding of granitoid magmatism within the context of the
Quetame orogenic cycle. Based upon recent, detailed lithogeochemical and isotopic
analysis of granitoids from the eastern Colombian and Mérida (Venezuela) Andes,
Van der Lelij (2013) and Van der Lelij et al. (2016) identified three phases of gran-
itoid magmatism within the context of early Paleozoic tectono-magmatic develop-
ment, which they integrate within the interpreted geodynamic evolution of the
autochthonous pre-Andean margin. These include (1) early, ca. 499–473 Ma syn-­
kinematic and peak metamorphic granitoids, which they interpret to have been gen-
erated/emplaced during a period of compression, crustal thickening, metamorphism
and orogenesis; (2) ca. 472–452 Ma granitoids, emplaced during post-orogenic col-
lapse, extension and basin formation; and (3) ca. 452–415 Ma granitoids emplaced
during resumed compression, basin closure and crustal thickening. Although these
authors interpret the continual subduction of Iapetus oceanic crust beneath the NW
Gondwana margin during the entire ca. 499–415 Ma period (see Fig. 5.15 of Van der
Lelij et al. 2015), their detailed Hf, Sr, Nd and Pb isotope data led them to conclude
that all of the ca. 499–415 Ma granitoids are primarily composed of recycled crustal
melts, with increasing but minor contributions of enriched and depleted mantle
material during the ca. 472–452  Ma period, facilitated by active extension and
crustal thinning, respectively. In this context, the early Paleozoic granitoids appar-
ently do not represent subduction-derived melts per se, and Van der Lelij et  al.
(2016) invoke a process of lithospheric mantle upwelling and heat advection at the
base of the crust, in the generation and partitioning of primarily crustal-derived
melts.
The data of Van der Lelij (2013) and Van der Lelij et al. (2016) did not include,
however, the emerging population of early Paleozoic granitoids located significantly
to the west of the Santander-Floresta-Quetame massifs, hosted within Cajamarca-­
Valdivia Terrane metamorphic basement which underlies much of Colombia’s
Central Cordillera. Cajamarca-Valdivia is stratigraphically comprised of poly-­
deformed Vendian and early Paleozoic marine meta-sedimentary and volcanic
rocks, including the Cajamarca, Valdivia and Montebello Groups (Restrepo-Pace
1992; Cediel and Cáceres 2000; González 2001; Silva et al. 2005). Geochemical
and geological characterization studies presented by Restrepo-Pace (1992) and
paleogeographic reconstructions presented by Cediel et al. (1994) and Cediel (2011)
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 363

suggest Cajamarca-Valdivia (referred to as the Central Andean Terrane by Restrepo-­

Pace op. cit.) represents a peri-cratonic island arc and continental margin accretion-
ary prism assemblage, developed along the western Colombian continental margin
beginning in the Vendian-Cambrian and accreted along the Palestina fault and suture
system during the late Ordovician-Silurian. In this context, early Paleozoic granit-
oids contained with the Cajamarca-Valdivia assemblage, including the ca. 473 Ma
quartz diorite outcropping along the Otú Fault (Leal-Mejía 2011) and the ca. 479–
445 Ma La Miel orthogneiss (Villagómez et al. 2011; Martens et al. 2014), represent
granitoids emplaced within the peri-cratonic realm, as members of the Cajamarca-­
Valdivia arc complex. A more complete petrogenetic and tectonic characterization
of these granitoids, unfortunately, is lacking, due to the absence of lithogeochemical
analyses.
In view of the tectonic framework for early Paleozoic granitoid magmatism sum-
marized herein, we suggest the phases of syn- and post-orogenic granitoid magma-
tism documented in the eastern Colombian Andes (Goldsmith 1971; Restrepo-Pace
1995; Van der Lelij 2013; Van der Lelij et al. 2015), and granitoids contained within
the Central Cordillera (Villagómez et  al. 2011; Leal-Mejía 2011; Martens et  al.
2014) were developed within the context of the Ordovician-Silurian Quetame
Orogeny as described by Cediel and Cáceres (2000) and Cediel et al. (2003). This
orogeny appears to have been driven by the approach and accretion of the Cajamarca-­
Valdivia island arc assembly and closure of the Iapetus Ocean.
The youngest early Paleozoic granitoids from the Santander Massif and
Cajamarca-Valdivia Terrane date from ca. 439 to 445 Ma, respectively. Existing
U-Pb (zircon) age date data indicate a paucity of granitoid occurrences through-
out the Colombian Andes, spanning the period from ca. 439 to 333 Ma (Figs. 5.4,
5.5, and 5.6), indicating a general hiatus in granitoid magmatism over a ca.
100 m.y. span.
The term Bolivar Aulacogen was originally proposed by Cediel and Cáceres
(2000) and Cediel et  al. (2003) to describe the prolonged period of continental
taphrogenesis surrounding northwestern South America, beginning in the mid-late
Paleozoic and continuing through to the early Cretaceous (Cediel et al. 1994; Cediel
and Cáceres 2000; Cediel et al. 2003). In eastern Colombia and western Venezuela,
this extensional regime initiated with the development of an intercontinental rift and
deposition of marine strata in the Pennsylvanian through Permian (Sierra de Mérida,
Eastern Cordillera). The extensional regime changed briefly to transpressive in the
late Permian, as recorded by tight folds associated with strike-slip faulting observed
in the Sierra de Mérida (Marechal 1983). Rifting resumed during the Triassic (e.g.
Payandé rift, Cediel and Cáceres 2000), continued into the Jurassic (e.g. Morrocoyal
rift, Geyer 1973; Siquisique rift, Bartok et al. 1985; Perijá rift, Cediel and Cáceres
2000) and culminated in the early Cretaceous with the opening of the Valle Alto rift
(Cediel and Cáceres 2000), prior to the onset of the transpressive regime character-
istic of the proto-Northern Andean Orogeny.
Within the context of the Bolívar Aulacogen, granitoid development may be
considered within three stages. As observed above, the initial phase of the Bolívar
Aulacogen was essentially amagmatic, coinciding with a hiatus in granitoid magmatism
364 H. Leal-Mejía et al.

in Colombia extending from ca. 439 to 333 Ma. The intermediate phase is demon-
strated by an emerging record of magmatism beginning in the mid-­Carboniferous
and extending into the Permian and mid-late Triassic, dominated by granitoid ana-
tectites and bimodal granitoid-gabbo (amphibolite) assemblages. The late phase is
characterized by subduction-related volcano-plutonic arc systematics, developed
within a highly extensional regime during the Jurassic (Sect. 5.4.2.1). We will now
outline the development of granitoid magmatism during the intermediate phase of
the Bolívar Aulacogen, from the mid-Carboniferous to the Permo-Triassic.

5.4.1.2  Tectonic Framework for Carboniferous Granitoids

Regional-scale tectonic reconstructions for the Carboniferous of the Northern

Andes depict a generally passive margin. Interpreted north- to west-directed sub-
duction was localized along the conjugate Laurentian margin prior to the final amal-
gamation of Pangaea in the Permian (e.g. Keppie 2008; Ramos 2009; Van der Lelij
et al. 2016). In Colombia, detailed basin and facies analysis suggests the ca. 333–
310 Ma period was a time of flysch-type sedimentation and of general magmatic
quiescence (Cediel et al. 1994; Cediel et al. 1998; Cáceres et al. 2003). Indeed, Van
der Lelij (2013) notes that there is little evidence (on a regional level) to support the
existence of an active margin outboard of northwestern Gondwana between ca.
415 Ma and 290 Ma.
Notwithstanding, Leal-Mejía (2011) documented Carboniferous, ca. 330–
310 Ma granitoid magmatism in the El Carmen-El Cordero Stocks, hosted within
Cajamarca-Valdivia Terrane basement along the Otú Fault in Colombia’s northern
Central Cordillera (Figs. 5.3, 5.5, and 5.6). Available petrographic, lithogeochemi-
cal data (Leal-Mejía 2011; see Sect. 5.3.2.3, Figs. 5.7 and 5.8) denote an apparently
bimodal gabbro-melanodiorite-leucotonalite assemblage at El Carmen-El Cordero.
The suite is of metaluminous-weakly peraluminous, magnesian-calcic composition
(Frost et al. 2001), and all samples return strongly mantelic Sr-Nd isotope signa-
tures (Fig. 5.9). Based upon the low-K, hydrous nature of the El Carmen-El Cordero
granitoids, the suite does not appear to represent an A-type (Loiselle and Wones
1979; see Frost et  al. 2001) assemblage. A general calc-alkaline trend may be
implied on the AFM diagram, although this is inconclusive, given gaps in intermedi-
ate compositions within the differentiation series. This may simply reflect the lim-
ited sample population upon which the present classification in based (n  =  7).
Overall, however, utilizing the classification scheme of Barbarin (1999), the El
Carmen-El Cordero suite conforms well to mantle-derived, “tholeiitic” granitoids,
of the RTG (Ridge Tholeiitic Granitoids) type. RTG suites characteristically include
gabbro through tonalite, trondhjemite and plagiogranite assemblages which are
Na-rich and mantelic with low 87Sr/86Sr ratios. Such suites are interpreted to be
associated with tholeiitic, gabbro-dominant assemblages generated along oceanic
spreading ridges. The more felsic members of the series are derived in small vol-
umes through extreme crystal fractionation of basaltic melts and occur as dikes and
plutons hosted within ophiolite complexes/oceanic crust (Barbarain 1999 and refer-
ences cited therein).
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 365

In terms of age, lithochemistry and Sr-Nd isotope composition (Leal-Mejía

2011), the El Carmen-El Cordero suite presently stands unique, not only in the
Colombian Andes but for the entire Northern Andean region. The age of these intru-
sives significantly pre-dates the age of the well-documented Permo-Triassic arc-­
related (e.g. Cardona et al. 2010b) and bimodal meta-granitoids, granitoid gneisses
and amphibolites (Vinasco et  al. 2006; Cardona et  al. 2010b; Cochrane 2013;
Spikings et  al. 2015; see below). As mentioned, the geological context of the El
Carmen-El Cordero suite is not fully understood. The granitoids are hosted within
the confines of the Cajamarca-Valdivia Terrane and, based upon age constraints,
were emplace at least 70 m.y. after accretion of the Cajamarca-Valdivia assemblage
to continental Colombia. The granitoids are localized along the Otú Fault, a major
N-S striking feature, which in the past has been interpreted as a potential plate
boundary (e.g. Restrepo and Toussaint 1988; González 2001). Lithogeochemical
data supplied by Leal-Mejía (2011) and summarized herein suggest the El Carmen-El
Cordero granitoids represent a RTG suite (Barbarain 1999) complete with low-­
volume leucotonalite and trondhjemite differentiates, petrogenetically associated
with oceanic spreading and ophiolite formation. We suggest that the El Carmen-El
Cordero suite reflects the progressively extensional environment prevalent during
the intermediate stages of the Bolívar Aulacogen. The Otú Fault could represent the
longitudinal axis of a rift basin which opened to the point of at least locally produc-
ing oceanic lithosphere. The U-Pb (Zircon) ages produced to date by the El
Carmen-El Cordero assemblage suggest the Otú rift was active over a >  23  m.y.
period. No additional geological (sedimentological) record of the basin is known to
exist within the region, although it could be contained within the poly-deformed
metamorphic sequences of the Central Cordillera which have yet to be accurately
dated. Alternatively, it may have been mostly removed by erosion during Meso-­
Cenozoic tectonic events. In either case, rifting and basin formation along the Otú
Fault were short-lived and appear to have been aborted by the early Permian(?). An
ensuing period of granitoid quiescence is observed, between ca. 310 and 289 Ma
prior to the appearance of a new population of granitoid gneisses, granitoids and
amphibolites with petrographically, lithogeochemically and isotopically distinct
characteristics, during the Permo-Triassic.

5.4.1.3  Tectonic Framework for Permian to Mid-Late Triassic Granitoids

Granitoids returning Permian U-Pb (zircon) dates appear in the Colombian Andes at
ca. 289 Ma, and granitoid magmatism sensu lato continued throughout the Permian
and into the mid-late Triassic. In recent years, numerous workers have produced
disparate, localized radiometric age date, lithogeochemical and isotopic data per-
taining to the Permo-Triassic granitoid suite (e.g. Ordoñez and Pimentel 2002;
Saenz 2003; Cardona et al. 2010b; Leal-Mejía 2011; Villagómez et al. 2011; Van
der Lelij 2013; Rodríguez et al. 2014), whilst detailed, integrated studies focussed
specifically upon these rocks have been undertaken by Vinasco (2004) and Cochrane
(2013). Upon integration of these studies, a composite understanding of the tectonic
framework of Permo-Triassic granitoid magmatism can be derived.
366 H. Leal-Mejía et al.

Global tectonic reconstructions proposed by various authors (Keppie and Ramos

1999; Keppie 2004; Cocks and Torsvik 2006; Weber et  al. 2007; Cardona et  al.
2010b; Van der Lelij 2013) suggest that during the final assembly of Pangaea in the
Permian, northwestern South America was positioned along the WNW-margin of
Gondwana, at a complex juncture between Gondwana, Laurensia and numerous
loosely assembled pericratonic terranes, accumulated during closure of the Rheic
Ocean (e.g. the Middle American and Mexican terranes) but tangential to the prin-
ciple Gondwana-Laurentia suture (Ouachita-Marathon front; Keppie 2008;
Cochrane et al. 2014a; Van der Lelij 2013). Following Pangaea assembly most car-
toons depict the development of a west-facing subduction zone, suggesting the east-
ward subduction of proto-Pacific oceanic crust along much of the western Pangaean
margin (e.g. Cocks and Torsvik 2006; Keppie 2008; Cardona et al. 2010b; Cochrane
et al. 2014a; Van der Lelij et al. 2016).
In this context, based upon whole-rock lithogeochemical major, trace and rare-­
earth element analyses, Cardona et al. (2010b) and Villagómez et al. (2011) inter-
pret Permian meta-granitoids and granitoid gneisses outcropping in the Sierra
Nevada de Santa Marta (ca. 288–264 Ma) and the Central Cordillera (ca. 272 Ma),
respectively, to represent vestiges of early Permian, subduction-driven continental
margin magmatic arcs. We note, however, that the textural, mineralogical, structural
and lithogeochemical characteristics of these occurrences share many of the fea-
tures typical of the entire Permian through mid-late Triassic granitoid suite (Figs. 5.7
and 5.8), including the complexly zoned nature of contained zircons which produce
multiple inheritance ages. We suggest that in the absence of more in-depth isotopic
studies (Sr, Nd, Pb, Hf), it is premature to assign these meta-granitoids to a specific
tectonic environment based upon lithogeochemical analyses alone, especially given
the complexities which have historically been encountered in the interpretation of
other Colombian granitoid suites (e.g. early Paleozoic meta-granitoids, early
Jurassic Santander Plutonic Group).
Vinasco (2004) and Vinasco et  al. (2006) produced a detailed and integrated
petrographic, U-Pb (zircon), 40Ar-39Ar and Sr-Nd isotope and composite lithogeo-
chemical study of Permo-Triassic granitoid gneisses and less deformed granitoids
from several locations in Colombia’s Central Cordillera, and it was these authors
who were first to recognized the regional distribution and significance of this meta-­
granitoid suite. Vinasco et al. (2006) observed that inherited zircons from syntec-
tonic peraluminous granitic gneisses returned ca. 280 Ma metamorphic ages, whilst
ca. 250 Ma ages were returned from neoformed zircons. The less deformed crustal
granitoids returned ages of ca. 230 Ma. They demonstrated that, although individual
samples plot medium- to high-K calc-alkaline in composition, the suite is consis-
tently peraluminous (S-type) and that Sr-Nd isotope data suggest high degrees of
interaction, assimilation or derivation of magma from upper crustal sources. They
note that isotopic data for the ca. 230 Ma suite reveals increasing contributions of
juvenile mantle. Vinasco et al. (2006) suggest that the meta-granitoid suite is the
product of regional Permo-Triassic tectono-thermal orogenesis associated with the
assembly and break-up of the Pangaea supercontinent. A genetic model presented
by Vinasco et  al. (2006) suggests the Permian to mid-late Triassic suite records
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 367

collision-­related metamorphism at ca. 280 Ma, followed by crustal thickening and

the emplacement of syn-kinematic (gneissic) peraluminous granitoids at ca. 250 Ma.
Orogenic collapse led to the emplacement of late tectonic granitoid intrusions at ca.
230 Ma., marking the onset of Pangaea break-up in the Northern Andean region.
In addition to documenting the age and nature of Permo-Triassic granitoids in
central Colombia, Vinasco (2004) and Vinasco et al. (2006) also observed the spatial
and temporal relationship between peraluminous granitoids and amphibolite, on
both the eastern flank (e.g. Padua amphibolite) and western flank (e.g. El Retiro
amphibolite and Aburrá ophiolite) of the Central Cordillera. These authors sug-
gested the amphibolites represent mantle-derived mafic melts which played a role in
crustal anatexis and the overall petrogenesis of the ca. 230 Ma peraluminous gran-
itoids, during mid-late Triassic regional extension.
In a more recent study pertaining to the late Paleozoic-Cenozoic tectonic evolu-
tion of the Northern Andean region, Cochrane (2013) and Cochrane et al. (2014a)
provided additional lithogeochemical and isotopic analyses of the Permo-Triassic
meta-granitoids, including a detailed analysis of the spatially related amphibolites
from the Andes of both Colombia and Ecuador. The findings of these authors concur
with those suggested by Vinasco et al. (2006). Cochrane (2013) notes that important
lithogeochemical and isotopic features of ca. 275–240 Ma meta-granitoids include
whole-rock (La/Yb)N ratios of ca. 11–16, generally magmatic zircon Th/U ratios of
0.26–1.27 and zircon εHfi values between +2 and −12, which he interprets as con-
sistent with anatectites generated via relatively low degrees of crustal melting,
including a minimal juvenile component. Cochrane (op. cit.) concludes that
Permian-early Triassic granitoid magmatism in NW South America likely occurred
as a consequence of the collision and final amalgamation of western Pangaea,
although he notes that the composite lithogeochemical and isotopic data do not
unambiguously constrain the specific tectonic environment within which the
Permian-earliest Triassic anatectites formed.
Beginning at ca. 240 Ma, Cochrane (2013) and Cochrane et al. (2014a) docu-
ment the emplacement of anatectic granitoids accompanied by the appearance of
tholeiitic sills and dikes (amphibolites). These authors observed that most of the
post ca. 240 Ma crustal anatectites yield large intra-sample εHfi variations and much
lower (La/Yb)N and Th/U ratios than the Permian-early Triassic meta-granitoids,
potentially reflecting source mixing with coeval juvenile mafic magmatism. With
respect to the amphibolites emplaced between ca. 240 and 216 Ma, analyses pre-
sented by Correa (2007) and Cochrane et al. (2014a) reveal tholeiitic N-MORB to
Back Arc Basin Basalt (BABB) compositions. Cochrane et  al. (2014a) note that
early (ca. 240–232  Ma) mantle-derived tholeiites with εHfi values from +7.4 to
+11.2 yield some zircon εHfi values which suggest that older amphibolites
­assimilated continental crust, whilst the ca. 232–216 Ma amphibolites reveal dimin-
ished crustal contamination and incrementally juvenile isotopic compositions.
Cochrane et al. (2014a) present a model for the ca. 240–216 Ma anatectic granitoids
and juvenile amphibolites involving the thinning of continental lithosphere during
Pangaea break-up. They suggest the rift stage of continental disassembly involved
basaltic underplating which led to emplacement of mafic melts and, in turn, anatec-
368 H. Leal-Mejía et al.

tic melting of the continental crust. Based upon the collective data, they conclude
that rifting led to sea-floor spreading after ca. 223 Ma with ocean crust formation
occurring by ca. 216 Ma (Correa 2007; Cochrane et al. 2014a).
The foregoing tectonic models for Permo-Triassic granitoids in Colombia are
based primarily upon lithogeochemical, isotopic and petrogenetic arguments. They
provide important temporal and spatial constraints with respect to existing models
which demonstrate the taphrogenic character of the intermediate phases of the
Bolívar Aulacogen during the Permo-Triassic, as derived primarily from surface
geological and borehole mapping, geophysical studies and sedimentary facies and
basin analysis (e.g. Cediel et al. 1994; Cediel et al. 1998; Cediel and Cáceres 2000).
The magmatic vs. sedimentary-based models are particularly sympathetic begin-
ning with the onset of mid-late Triassic continental rifting.
With respect to the Permian tectonic assembly of Pangaea, however, some degree
of controversy surrounds the nature and extent of the effects of continental collision
and tectono-thermal metamorphism in Colombia, and in various locations, it is dif-
ficult to reconcile early Permian subduction(?) and continental collision within the
taphrogenic context of the Bolivar Aulacogen. For example, based upon the RTG
assemblage observed at El Carmen-El Cordero, and discussed above, an extensional
regime is observed into the late Carboniferous. Within the Quetame Massif, along
the Sumapaz Range, a near-complete upper Paleozoic to early Mesozoic strati-
graphic section is preserved (Cediel and Cáceres 2000), containing carbonate and
evaporate sequences of Carboniferous through early to middle Permian age, which
show no obvious tectono-metamorphic effects. The El Carmen-El Cordero granit-
oid suite also provides a case in point for this geological quandary. As documented
above, the El Carmen-El Cordero suite is of mid-Carboniferous age (ca. 333–
310  Ma), predating the Permian-early Triassic tectono-thermal event by over
30 million years. Detailed petrographic study of the El Carmen-El Cordero suite by
Leal-Mejía (2011), however, failed to reveal significant post-crystallization penetra-
tive deformation or metamorphic mineral assemblages beyond the pumpellyite-­
prehnite-­chlorite-epidote grade, an assemblage which could just as easily have
resulted from the low-temperature hydrothermal alteration which affects the suite
(Leal-Mejía 2011; Shaw et al. 2018).
Notwithstanding, cartoons depicting the relative position of continental Colombia
within Gondwana and with respect to Laurentia, the Middle American-Mexican ter-
ranes and the Ouachita-Marathon front during the late Carboniferous-early Permian
are highly speculative, and the majority of the recent global-scale reconstructions
suggest the region was peripheral to the principle Gondwana-Laurentia suture (e.g.
Keppie 2008; Weber et al. 2007; Cadona et al. 2010b; Van der Lelij 2013; Cochrane
et  al. 2014a). A Permo-Triassic suture per se has yet to be clearly documented
within the context of Colombia-based paleo-tectonic reconstructions (e.g. Cediel
et al. 1994), and the proximity of continental Colombia to the Ouachita-Marathon
front remains largely undetermined. The highly complex nature of the western
Pangaea juncture is evident, and the potential role of the Middle American-Mexican
terranes in stress field buffering along the collision zone has yet to be evaluated.
It is intuitive that the presence of numerous small peri-cratonic crustal fragments
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 369

will have a first-order effect upon the development of a well-defined or easily iden-
tifiable suture trace, especially if the region was located in a tangential position with
respect to the principle collision front. From a structural standpoint, tight folds asso-
ciated with late Permian strike-slip faulting in the Sierra de Mérida (Marechal 1983)
may provide a record of Pangaean assembly from within the continental autoch-
thon. Further investigations regarding the Permo-Triassic tectono-thermal event on
a Colombian vs. regional scale are clearly warranted.
In conclusion, granitoid magmatism within the Colombian Andes during the
Permian through mid-late Triassic is represented by widespread but generally small-­
volume occurrences of granitoid gneisses and anatectites, observed primarily within
the Cajamarca-Valdivia Terrane underlying much of the Central Cordillera but also
within the Sierra Nevada de Santa Marta and, to a lesser degree, in the Santander
Massif. Based upon the data sets compiled herein, these granitoids provide a mag-
matic record reflecting the tectonic history of western Pangaea during the Permian
and Triassic (e.g. Vinasco et al. 2006; Cochrane et al. 2014a). The granitoids are
clearly characterized by their ubiquitous peraluminous (S-type) nature, contrasting
markedly with the voluminous metaluminous granitoids dominating the Colombian
Andes during the Meso-Cenozoic. Most authors concur that the Permian to early
Triassic meta-granitoids and granitoid gneisses provide a record of crustal thicken-
ing and anatexis coincident with Pangaea amalgamation, whilst the bimodal mid-­
late Triassic peraluminous granite-amphibolite suite reflects continental rifting,
culminating in ocean crust formation during Pangaea disassembly. Finally, we note
that the role of subduction, as depicted in numerous large-scale paleo-tectonic
reconstructions of the western Pangaean region (e.g. Weber et al. 2007; Cardona
et al. 2010b; Cochrane et al. 2014a and references cited therein), and the contribu-
tion of subduction-derived melts (e.g. Cardona et al. 2010b; Villagómez et al. 2011),
in the petrogenesis of the Colombian Permo-Triassic granitoids, have yet to be
clearly demonstrated.
With the onset of oceanic rifting and advanced continental break-up along the
Colombian proto-Pacific margin, the intermediate phase of the Bolívar Aulacogen,
characterized by low-volume peraluminous granitoids, gave way to a regime per-
missive to the emplacement of voluminous, subduction-related metaluminous gran-
itoids. The development of large-scale Jurassic batholiths accompanied by abundant
volcanic rocks, emplaced within an extensional regime, during the late phase of
development of the Bolívar Aulacogen will now be discussed.

5.4.2  L
 ate Bolívar Aulacogen: Tectonic Framework for Latest
Triassic-Jurassic Granitoids

Schematic models for the late Triassic-Jurassic structural and tectonic evolution of
Colombia and the Northern Andes have been presented by numerous authors over
the last five decades, including Bürgl (1967), Irving (1975), Sillitoe et al. (1982),
Burke et al. (1984), Etayo-Serna et al. (1983), Aspden et al. (1987), Restrepo and
370 H. Leal-Mejía et al.

Toussaint (1988), Pindell et  al. (1988), Cediel et  al. (1994), Cediel and Cáceres
(2000), Pindell and Kennan (2001), Cediel et al. (2003), Kennan and Pindell (2009),
Cochrane et  al. (2014b) and Spikings et  al. (2015). Many of these models were
imprecise, incomplete and/or overly selective with respect to the composite data-
base of geological and cartographic information available for the region or, alterna-
tively, were drawn at scales encompassing all of NW South America and the
Caribbean, which did not permit the exposition of detailed and specific geological,
stratigraphic, radiometric age date, lithogeochemical and isotopic information.
In order to update and better constrain these models, at a scale specifically repre-
sentative of the Colombian Andes, we have integrated the late Triassic-Jurassic
radiometric age, isotopic and lithogeochemical information presented above into
the detailed paleo-facies, structural and tectonic framework provided by Cediel
et al. (1994). The resulting composite late Triassic-Jurassic tectono-magmatic con-
figuration is presented in Fig. 5.31. A summarized time-space analysis for the mag-
matic evolution of the region during the late Triassic-Jurassic is illustrated in
Fig. 5.32.
The transition from middle to late Triassic rifting and continental break-up to the
formation of late Triassic-Jurassic subduction-related magmatic arcs, marking the
late phase of the Bolívar Aulacogen, is first recorded in Colombia in the ca. 210–
196 Ma granitoids of the Santander Plutonic Group. Spikings et al. (2015) interpret
the formation of a proto-subduction zone along the NW Colombian (Gondwana)
margin beginning around this time. Late Triassic-Jurassic rift-related sedimentation
in the Maracaibo and Perijá Rifts (Cediel et  al. 1994; Cediel and Cáceres 2000;
Cáceres et  al. 2003; Cediel et  al. 2003) indicates active rifting accompanied by
emplacement of the Santander granitoid suite. Based upon the NNW orientation of
the long axis of the Santander suite, initial subduction (if present) was broadly
NE-directed. Although the granitoids are interpreted to have been emplaced in a
continental arc setting, lithogeochemical and isotopic data indicate melts were pri-
marily derived from, or mixed with, crustal sources, with a limited mantelic compo-
nent (Van de Lelij 2013; Bissig et al. 2014). A crustal source is in keeping with the
lithogeochemical and isotopic composition of Proterozoic and early Paleozoic met-
amorphic basement rocks of the Santander Massif which host the Santander Plutonic
Group. Magma generation may be more specifically related to extension-related
mantle upwelling and thermal-induced partial melting of lower crustal basement
underlying the Santander Massif than to the subduction and partial fusion of oce-
anic lithosphere per se, as would be implied in typical models for arc-related,
­calc-­alkaline granitoids. In either case, extension was insufficient to allow the
wholesale entry of mantle-derived melts into the upper crust (Van der Lelij 2013).
Following ca. 196 Ma, WNW migration of the calc-alkaline magmatic arc axis is
observed (Fig. 5.31). The ca. 189–180 Ma granitoids of the southern Ibagué, Norosí,
San Martín and Pueblo Bello-Patillal Batholiths and the ca. 180–172 Ma Mocoa-­
Garzón intrusions represent extensive subduction-related magmatism with a clear
metaluminous character, increasing mantelic component and diminishing degree of
interaction with sialic continental basement (Alvarez 1983; Dörr et al. 1995; Leal-­
Mejía et al. 2011; Cochrane 2013). Arc axis migration was accompanied by ~30
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 371

degrees of clockwise rotation of the long axis of the arc into a NNE orientation,
suggesting a shift to broadly SE-oriented subduction. A marked increase in magma
volume is represented by the ca. 189–172  Ma granitoids, all of which include a
significant explosive volcanic component (e.g. the Saldaña, Noreán, Jordán and
Guatapurí Fms.). The ca. 189–172  Ma arc segments were thus emplaced under
highly extensional conditions, in some instances coaxial to precursor Permo-­
Triassic rift-related sedimentary grabens (e.g. Payandé Rift and Ibagué Batholith;
Cediel et al. 1994; Cediel and Cáceres 2000). The slight eastward migration of the
Mocoa-Garzón intrusions with respect to the southern Ibagué Batholith suggests the
onset of a locally compressional regime in southern Colombia at the end of this
magmatic cycle, possibly related to declining rates of extension and/or shallowing
of the oceanic slab subduction angle.
Continued WNW migration of the magmatic arc axis is observed with the
emplacement of the ca. 168–155  Ma Segovia and northern Ibagué Batholiths
(Fig.  5.31). Further shifts to more juvenile, mantle-derived compositions are
observed for these batholiths, with lesser REE enrichment and Sr-Nd isotope ratios
trending into the depleted mantle array. Notwithstanding, the absence of associated
volcanic piles or evidence of coeval volcanism suggests these granitoids were
emplaced within an increasingly neutral to compressive tectonic regime. The ero-
sion of significant Jurassic volcanic stratigraphy during Cenozoic Northern Andean
orogenic events cannot however be ruled out.
To the south, in the southern Ibagué Batholith, no granitoids dating from the ca.
168 to 155 Ma episode are observed, and based upon the available data, no addi-
tional Jurassic granitoid magmatism is recorded for this area, signifying the shut-
down of subduction by ca. 172 Ma. We interpret the development of NW–SE-striking
transform fault or slab tear in the Pacific Plate (Fig. 5.31), to the north where sub-
duction continued between ca. 168 and 155  Ma, whilst to the south a complete
shutdown of the Jurassic arc in Colombia is observed.
Following the final episode of holocrystalline intrusions at ca. 152  Ma, volu-
metrically minor hypabyssal porphyry stocks were emplaced along the eastern
(back arc) margin of the northern Ibagué Batholith between ca. 152 and 145 Ma
(Fig. 5.31). They record, if anything, a net eastward migration of magmatism, sug-
gesting the (temporary) cessation of regional extension and a trend towards a more
neutral to compressive tectonic conditions during closure of late Triassic-Jurassic
arc-related granitoid magmatism.
The late Triassic-Jurassic granitoids of the Colombian Andes were generated
within a highly complex tectonic regime involving the early rifting and break-up of
western Pangaea and the separation of the Middle American terranes, followed by
the continuous broadly east-directed subduction of Pacific oceanic crust beneath
NW South America. The net result of this tectonic evolution was the temporal
development of four major granitoid episodes, with associated volcanism and hyp-
abyssal porphyry emplacement, manifest in at least six spatially distinct arc seg-
ments, emplaced within a highly extensional tectonic regime. Hamilton (1994)
notes that continental margin magmatic arcs are extensional by nature, as recorded
in the development of back-arc basins and arc-axial grabens. He cites slab-pull
372 H. Leal-Mejía et al.

(rollback) due to the sinking of dense fore-arc oceanic lithosphere into the mantle
as the major factor in the development of extension across a magmatic arc. Leal-
Mejía (2011), Cochrane et al. (2014b) and Spikings et al. (2015) considered Pacific
oceanic slab rollback an important cause of WNW granitoid arc migration in north-
ern Colombia between ca. 210 and 152 Ma. In addition to the extensional effects
caused by slab rollback, however, a net SE-directed movement vector for the South
American Plate, nearly opposite that of slab rollback, has been proposed for most of
the Jurassic and early Cretaceous (e.g. Kennan and Pindell 2009). Thus, we inter-
pret the tectonic framework for reactivation of pre-Mesozoic basement structures,
the development of middle to late Triassic continental rifts and the emplacement
of the late Triassic-Jurassic subduction-related granitoids in Colombia to be a
reflection of the extreme extensional conditions brought on by the combination of
Pacific slab rollback and SE-directed migration of the South American Plate
throughout the Jurassic.

5.4.2.1  Culmination of the Bolivar Aulacogen: Valle Alto Rift

Paleo-tectonic reconstructions suggest that interactions between the Pacific Plate

and NW South America became highly oblique or strike-slip during the Jurassic-­
Cretaceous transition (Cediel et al. 1994; Cediel et al. 2003; Keppie 2004; Kennan
and Pindell 2009; Cochrane 2013; Spiking et al. 2015), leading to shutdown of the
subduction-driven granitoid magmatism which dominated the Jurassic. After ca.
145 Ma, the geological record confirms a rift-dominated tectonic regime, with the
formation of juvenile oceanic crust along the Colombian Pacific margin (Nivia et al.
2006; Cochrane 2013; Spikings et  al. 2015) and opening of the early–middle
Cretaceous Valle Alto-Eastern Cordillera Basin Rift (Cediel et al. 1994; Cediel and
Cáceres 2000; Cediel et al. 2003). This event was marked by deep continental rift-
ing and subsidence, the invasion of the Cretaceous seaway and the deposition of
marine and epicontinental sequences over extensive areas of the Central Tectonic
Realm (including the Cajamarca-Valdivia Terrane), the Maracaibo Sub-plate and
the continental platform of the Guiana Shield (e.g. see Sarmiento 2018) (Fig. 5.33).
The axis of the Valle Alto rift is marked by Colombia’s Eastern Cordilleran basin,
which contains up to 6 km of Cretaceous marine deposits characterized by a trans-
gressive sequence of basal, restricted marine mudstones, carbonates and evaporates
overlain by progressively deeper water, reduced (carbonaceous) shales and mud-
stones, deposited in at least four diachronous subbasins (Sarmiento 2001). Small
volumes of compositionally heterogeneous rift-related alkaline and tholeiitic mafic
intrusions mark periods of maximum extension, subsidence and subbasin develop-
ment (Fabre and Delaloye 1983; Vásquez et al. 2010). The mafic intrusions range in
age from ca. 136 to 74  Ma (Fabre and Delaloye 1983; Vásquez et  al. 2010).
Lithogeochemical and isotopic data published by Vásquez et al. (2010) demonstrate
the mantle-derived character and variable degrees of LREE enrichment and contri-
bution of old crustal material to the parent melts. The oldest intrusions (Pacho, ca.
136 Ma) plot in the field of “continental basalts”, reflecting the continental character
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 373

of the early rifted crust beneath the Eastern Cordillera, whilst the younger intrusions
reveal lithogeochemical and isotopic data which is progressively more ocean-like
(Vásquez et al. 2010). Additional rift-related Cretaceous marine volcano-sedimen-
tary deposits (e.g. San Pablo, Segovia, Valle Alto and Soledad Fms., Fig.  5.33)
(Gonzalez 2001) are found as localized erosional remnants within Colombia’s
Central Cordillera.
Along the Colombian Pacific margin, the period spanning the latest Jurassic
through ca. 124 Ma was under left lateral transtension (Cediel et al. 1994; Kennan
and Pindel 2009; Fig. 5.33) and formed an active depocenter for Berriasian through
Aptian and Albian sedimentary rocks of continental margin and oceanic affinity and
mixed assemblages of tholeiitic and calc-alkaline basalt and andesite, with associ-
ated mafic and ultramafic intrusive rocks (e.g. Quebradagrande Complex, Nívia
et al. 1996). This marginal basin also contained disjointed slivers of early Paleozoic
and Permo-Triassic metamorphic rocks (e.g. Bugalagrande complex; McCourt and
Feininger 1984; Arquía Complex; Nívia et al. 1996) typical of the rifted Northern
Andean continental margin during the early Cretaceous (Litherland et  al. 1994;
Cediel et al. 2003). Plate reorganization beginning in the Aptian (Cediel et al. 1994;
Maresch et al. 2000; Pindell and Kennan 2001) led to deep burial, metamorphism
and tectonic reworking of the marginal basin assemblages along the Colombian
margin (e.g. Orrego et al. 1980; McCourt and Feininger 1984; Maresch et al. 2000;
Bustamante 2008; Maresch et al. 2009), accompanied by large-scale dextral-oblique
transpressive shearing along the Romeral fault system (Ego et al. 1995). The com-
plex tectonic architecture of the Romeral mélange (Cediel and Cáceres 2000; Cediel
et al. 2003) was established at this time.
In the early Cretaceous, the Colombian Pacific was thus dominated by a rifted
transtensional-transform margin and by plate movement vectors, which, from a
tectono-magmatic standpoint, were not conducive to the formation of subduction-­
related granitoids (e.g. Aspden et al. 1987; Cediel et al. 1994; Pindell and Kennan
2001). This observation is principally supported by the absence of subduction-­
related, calc-alkaline granitoids in continental Colombia during the period from ca.
145 to 96 Ma (Fig. 5.4), suggesting little, if any, subduction took place beneath the
Colombian continental margin during this time.
Prolonged regional extension related to the Bolivar Aulacogen and the culminant
Valle Alto rift, and the ensuing ca. 50  Ma hiatus in granitoid magmatism in the
Colombian Andes, is terminated in the mid- to late Cretaceous, when plate
­reconfiguration in the Pacific regime led to dextral oblique convergence along the
Colombian margin (Figs. 5.33 and 5.34). This shift signalled the onset of the late
Mesozoic-Cenozoic Northern Andean Orogeny (Cediel and Cáceres 2000; Cediel
et al. 2003), comprised of a series of punctuated tectono-magmatic events, including
the generation of subduction-related, calc-alkaline, continental margin and peri-
cratonic volcano-magmatic arcs and the sequential approach, collision and accre-
tion of the Western Tectonic Realm allochthonous terrane assemblages of Pacific
provenance along the Colombian Pacific and Caribbean margins.
The tectonic evolution of Colombia and the Northern Andes during this time was
intimately linked to the opening of the Proto-Caribbean basin and to the genesis and
374 H. Leal-Mejía et al.

emplacement of the Caribbean Plate (e.g. Cediel et al. 1994; Kerr et al. 1997; Pindell
and Kennan 2001; Cediel et al. 2003; Kerr et al. 2003; Kennan and Pindell. 2009).
As highlighted in the following section, significant volumes of subduction-related
granitoids reappear in the Colombian Andes at ca. 96 Ma, with emplacement of the
precursor phases of the Antioquian Batholith (Leal-Mejía 2011).

5.4.3  E
 arly Northern Andean Orogeny: Tectonic Framework
for Cretaceous-Eocene Granitoids

Within the historical context, the Northern Andean Orogeny in Colombia has been
described by various authors (e.g. Bürgl 1967; Campbell 1974; Irving 1975).
General disagreement was observed, however, with respect to the timing and spatial
distribution of events, especially concerning the timing of deformation and granit-
oid magmatism. Based upon integrated time-space analysis and considering the
nature and geological history of the pre-Andean tectonic framework, Cediel et al.
(2003) redefined the Northern Andean Orogeny to include orogenic events occur-
ring since the transition from the generally extensional-transtensional regime of the
Bolivar Aulacogen to the transpressive (accretionary) regime beginning in the mid-­
Cretaceous (Aptian-Albian) and continuing up to the present. In historic works, the
driving mechanisms behind deformation and magmatism were poorly understood.
In recent times, however, numerous works demonstrate the sequential tectonic evo-
lution of the Colombian Andes and the integral relationship between the nature,
composition, migration and emplacement of the Caribbean Plate and the tectonic
development of the Northern Andean Block as a whole (e.g. Cediel et al. 1994; Kerr
et al. 1997; Sinton et al. 1998; Pindell and Kennan 2001; Cediel et al. 2003; Kerr
et al. 2003; Kennan and Pindell 2009; Nerlich et al. 2014; Cediel 2011).
With respect to the development of granitoid magmatism during the period span-
ning the mid-Cretaceous through Eocene, as observed in outcrop, recorded upon
regional scale geological maps (Cediel and Cáceres 2000; Gómez et al. 2007; Gómez
et al. 2015a) and verified by the available radiometric age dating studies highlighted
above, two groups of granitoids within the Colombian Andes, including the Eastern
and Western Groups, may be defined. Each of these groups has been subdivided into
subgroups, based primarily upon the age vs. spatial distribution of the granitoid
intrusions (Figs. 5.15 and 5.16). Thus, within the Eastern group, the ca. 96–72 Ma
Antioquian Batholith and satellite plutons and the ca. 62–50  Ma intrusions to the
south of the Antioquian Batholith suite, and in the Sierra Nevada de Santa Marta,
may be considered. Within the Western Group, the ca. 100–84 Ma and ca. 50–42 Ma
subgroups are highlighted. We emphasize that, based upon geological setting and
geotectonic considerations, supported by lithogeochemical and isotopic arguments,
the Eastern and Western groups reflect fundamental differences in petrogenesis and
mode of emplacement: the Eastern Group is autochthonous intrusions generated in
situ within the continental regime, whilst the Western Group is allochthonous in
nature, generated within the intra-oceanic regime, prior to accretion to the Colombian
continental margin.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 375

5.4.3.1  T
 ectonic Setting for the Ca. 96–72 Ma Antioquian Batholith Arc
Segment

Figure 5.34 presents a schematic representation of the composite tectonic setting of

the mid-Cretaceous-Eocene, taking into account the distribution of granitoid rocks
from this time period, as presently recognized within the geologic mosaic of the
Colombian Andes. Figure 5.35 contains a detailed time-space analysis for the ca.
100–40 Ma granitoids within the context of the established tectonic elements of the
region.
Various tectonic reconstructions for the region surrounding NW South America
(e.g. Pindell and Kennan 2001; Cediel et al. 2003; Kennan and Pindell 2009; Wright
and Wyld 2011; Spikings et al. 2015; Weber et al. 2015), in conjunction with U-Pb
(zircon) age dating of Colombian granitoids presented herein, suggest initiation of
E- to NE-directed, dextral-oblique subduction beneath the western Colombian mar-
gin, beginning at ca. 100 Ma, resulting in the appearance of metaluminous, calcic to
calc-alkalic continental arc granitoids beginning at ca. 96 Ma.
With respect to the Antioquian Batholith and its suite of satellite plutons, mag-
matism was generated along a west-facing arc segment, within the Colombian con-
tinental block, represented by the Cajamarca-Valdivia Terrane. Three important
magmatic pulses have been identified, including early calcic gabbros and diorites
emplaced at ca. 96–92 Ma, followed by main phase batholith emplacement includ-
ing two distinct tonalitic to granodioritic suites, in two pulses, from ca. 89 to 82 Ma
and from ca. in 81 to 72 Ma, accounting for greater than 90% of batholith volume.
The ca. 89–72 Ma period would coincide with the eastward subduction of Proto-­
Caribbean and ± marginal basin crust beneath northwestern South America. The
generally dextral, transpressive regime of emplacement for the Antioquian Batholith
suite has been highlighted by numerous authors (Aspden et al. 1987; Cediel et al.
1994; Pindell and Kennan 2001; Cediel et al. 2003), accounting for the relatively
limited extent of the ca. 96–72 Ma continental arc segment, especially when com-
pared with the orthogonal subduction regimes dominating the Cretaceous granitoid
arcs observed in the Central Andes of Perú and Chile. At ca. 72  Ma, granitoid
­magmatism within the Antioquian Batholith ceases abruptly, and a hiatus of ca.
10 m.y. is recorded prior to the reinitiation of granitoid magmatism within the con-
tinental domain.

5.4.3.2  T
 ectonic Setting for the Ca. 100–84 Ma Western Group Arc
Segment

The ca. 100–84 Ma granitoids of the Western Group, including the Buriticá, Santa
Fé (Sananalarga), Mistrató, Buga and Jejénes and associated intrusive suites, form
a curvilinear arc segment which extends for over 600 km, aligned along the NNW-­
oriented tectonized front of the Western Tectonic Realm, in sutured contact with the
continental margin facies represented by the Romeral melange, immediately to the
east (Fig. 5.34). Geological, lithogeochemical and isotopic considerations indicate
the ca. 100–84 Ma Western Group granitoids represent the vestiges of a primitive
376 H. Leal-Mejía et al.

calcic to calc-alkaline arc system, generated within the intra-oceanic domain and
emplaced within the host Dagua and Cañas Gordas terrane assemblages of the
Western Tectonic Realm, prior to their accretion to the continental margin.
Many recent tectonic reconstructions focus upon the oceanic domain along the
NW margin of South America during the mid-Cretaceous through late Cretaceous
(e.g. Kennan and Pindell 2009; Wright and Wyld 2011; Nerlich et al. 2014; Spikings
et al. 2015; Weber et al. 2015). These reconstructions illustrate the appearance of
intra-oceanic arcs associated with east-facing subduction of Proto-Caribbean oce-
anic crust beneath the approaching Caribbean-Colombian Oceanic Plateau (CCOP/
CLIP; Kerr et  al. 1997, 2003; Sinton et  al. 1998). This system of primitive arcs,
emplaced within the Farallon Plate and within overlying oceanic plateau rocks, has
been variably referred to as the “Great Arc of the Caribbean” (Burke et al. 1984;
Kennan and Pindell 2009; Hastie and Kerr 2010), the “Ecuador-Colombia Leeward
Arc” (Wright and Wyld 2011), the “Greater Antillean Arc” (Nerlich et al. 2014) and
the “Rio Cala Arc” (Spikings et al. 2015). In Colombia, the ca. 100–84 Ma metalu-
minous granitoids contained within the Dagua and Cañas Gordas terranes (Fig. 5.34),
including the Buriticá, Santa Fé (Sananalarga?), Mistrató, Buga and Jejénes intru-
sives, are interpreted herein to represent accreted constituents of the Greater Arc.
Various studies address the timing and kinematics of Farallon Plate-CCOP/CLIP
assemblage collision and accretion to the Colombian margin during the late
Mesozoic, during what we herein refer to as the early Northern Andean Orogeny.
Detailed paleo-facies and stratigraphic reconstruction, and basin analysis, at the
scale of the entire Colombian Andes (Cediel et al. 1994; Cediel and Cáceres 2000;
Cáceres et al. 2003), depict the continental margin tectonic response to the approach
and sequential collision of the Cañas Gordas, Dagua and Gorgona terranes, begin-
ning in the Campanian and extending progressively continent-ward, as recorded in
uplift-related unconformities recorded in the physiographic Central and Eastern
Cordilleras and Santander Massif, extending into the Eocene and Oligocene. This
stratigraphic data is supported by the detailed study of seismic sections depicting
the subsurface structure and tectonic evolution of Meso-Cenozoic sedimentary
basins in Colombia (Cediel et al. 1998; Sarmiento 2018) and by thermochronologi-
cal data suggesting rapid exhumation in the Central Cordillera between ca. 75 and
55 Ma (Spikings et al. 2015). Reconstruction of the evolution and trajectory of the
NE-migrating CCOP/CLIP assemblage, from the Pacific realm into the inter-Amer-
ican gap, suggest (final?) docking of Farallon-CCOP/CLIP components along the
NW margin of South America at ca. 54.5 Ma (Nerlich et al. 2014), closely followed
by accretion of the Gorgona Terrane beginning in the mid-Eocene (Cediel et  al.
2003; Kerr and Tarney, 2005). Thus, the early Northern Andean Orogeny is a dia-
chronous, regional event which, in Colombia, evolved both spatially and temporally
over a span exceeding 20 m.y.
With respect to the evolution of the ca. 100 through 72 Ma subduction-related
granitoids within the region, we interpret the demise of the ca. 100–84 Ma Western
Group arc segment to be related to the near-complete, west-directed consumption of
Proto-Caribbean crust located between the Farallon-CCOP/CLIP assembly and the
Colombian margin, during CCOP/CLIP migration into the peri-cratonic realm, by
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 377

ca. 84 Ma. Continued NW-directed convergence between the Farallon-CCOP/CLIP

assembly and the continental margin was accommodated by dextral oblique trans-
form faulting (Fig. 5.34) and, more locally, by the east-directed subduction of rem-
nants of marginal basin, Proto-Caribbean, possibly leading edge slivers of
Farallon Plate oceanic crust beneath the continental margin, resulting in the
main-phase emplacement of the Antioquian Batholith suite. The Antioquian arc
segment however was generally short-lived and was rapidly extinguished at ca.
72 Ma, due to impingement of the Farallon-CCOP/CLIP assemblage upon the con-
tinental margin.
We interpret the ensuing hiatus in continental magmatism to be related to the
invasion of the Antioquian segment trench by the Farallon-CCOP/CLIP assem-
blage. This magmatic hiatus is considered to reflect various factors/events associ-
ated with post-accretionary tectonic reorganization, prior to the reinitiation of the
granitoid magmatism recorded within the post-Antioquian, ca. 62–50 Ma Eastern
granitoids subgroup. Initially, “chocking-off” of the subduction zone was due to
invasion by buoyant CCOP/CLIP fragments such as those represented by the Dagua
terrane. Continued plate convergence was dominated by dextral-oblique transpres-
sion and offshore transform faulting (Aspden et al. 1987; Pindell and Kennan 2001;
Cediel et al. 2003; Wright and Wyld 2011), within an overall regime which was not
conducive to continued subduction nor to immediate slab breakoff and subduction
reinitiation. Coupling stress beginning in the Maastrichtian was partitioned into
various structural components of the Andean mosaic. The development of the Cauca
fault and suture system (Ego et al. 1995; Cediel et al. 2003), which separates the
accreted Western Tectonic Realm assemblages from the continental margin, took
place at this time. Additional tectonic tightening and reactivation along preexisting
structures, including the Palestina (Feininger et al. 1972) and Romeral fault systems
(Ego et al. 1995; Cediel and Cáceres 2000; Cediel et al. 2003; Vinasco 2018), facili-
tated collision-related uplift of litho-tectonic units throughout the Central Tectonic
Realm.

5.4.3.3  T
 ectonic Setting for the Ca. 62–50 Ma Eastern Group Post-­
collisional Arc Segment

Following a ca. 10 Ma hiatus, continental arc granitoids reappear within the autoch-
thonous, continental domain, as recorded in our ca. 62–52 Ma Eastern Group intru-
sions (i.e. Providencia, Sonsón, Manizales, El Hatillo, El Bosque and Santa Marta
plutons), contained within the physiographic Central Cordillera and Sierra Nevada
de Santa Marta (Fig. 5.34). These intrusions represent the reinitiation of granitoid
magmatism, albeit at a much reduced rate/volume, following the initial invasion/
collision of Farallon-CCOP/CLIP components along the Colombian margin and the
extinction of the Antioquian Batholith-related magmatism. In this context we have
referred to the ca. 62–52 Ma Eastern Group intrusions as “post-collisional” granit-
oids in Figs. 5.34 and 5.35. U-Pb (zircon) age dates for these plutons illustrate
the southward migration of the post-collisional arc axis, from the Providencia
378 H. Leal-Mejía et al.

suite into the ancestral Central Cordillera to the south of the Antioquian Batholith.
The lithogeochemical and isotopic tendencies of these intrusions are clearly distin-
guished on Figs. 5.17, 5.18, and 5.21, especially with respect to the increased degree
of isotopic exchange through direct anatexis or contamination from crustal sources,
as suggested by the available Sr-Nd data. Recent Hf isotope data supplied by
Bustamante et al. (2017) additionally supports this observation. Initial εHf values
presented by these authors for the El Hatillo Stock (−0.7 to +5.6) and the El Bosque
Batholith (−4.5 to +1.3) suggest moderate to high degrees of crustal inheritance and
recycling. Indeed, the El Bosque Batholith contains inherited Permian-aged zircons
(Bustamante et  al. 2017), supplying direct evidence of the recycling of Central
Cordilleran basement (Cajamarca-Valdivia Terrane).
Leal-Mejía (2011) and Bustamante et  al. (2017) draw attention to the strong
“adakite-like” trend produced by the ca. 62 Ma Providencia suite. Bustamante et al.
(op. cit.) contrast this trend with the lower Sr/Y vs. Y ratios produced by the Sonsón
and El Bosque Batholiths and El Hatillo Stock. These authors suggest a petroge-
netic model involving magmatic differentiation at the base of a thick lower crust,
related to convergence/subduction of the CCOP lithosphere, with apparently
increasing degrees of crustal contamination as magmatism migrated southwards.
Notwithstanding, Leal-Mejía (2011) observed that potentially analogous litho-
geochemical and isotopic trends may be derived through a model involving delami-
nation of subducted oceanic lithosphere and asthenospheric upwelling, following
terrane collision. This author provides as example the work of Parada et al. (1999),
who explain the lithogeochemical and isotopic evolution of the late Jurassic-­
Cretaceous Chilean Coastal Batholith of the Central Andes as the result of collision
of an oceanic ridge with the continental margin. These authors interpreted pre-­
collisional, metaluminous, calc-alkaline magmas to be products of east-directed
subduction-related arc magmatism. Following oceanic ridge collision and the cessa-
tion of subduction-related magmatism, Parada et al. (1999) invoke a model of litho-
spheric delamination leading to the upwelling of asthenospheric mantle and
extensional deformation in the overlying continental crust, followed by the emplace-
ment of post-collisional granitoids with “adakite-like” signatures. They note that
εNd(t) values within the Coastal Batholith show a vertically increasing trend, coinci-
dent with the transition to “adakite-like” compositions. A very similar vertical
increasing εNd(t) array is observed within the Antioquian Batholith suite (Fig. 5.21),
prior to the emplacement of the “adakite-like” compositions reflected in the
Providencia and to a lesser degree El Hatillo suites (Leal-Mejía 2011).
In this context, we suggest that the reappearance of granitoid magmatism as rep-
resented by the ca. 62–52 Ma Eastern Group arc segment does not necessarily rep-
resent the resumption of subduction along the Colombian Pacific margin and could
alternatively be explained using a model of post-collisional lithospheric delamina-
tion, asthenospheric upwelling and thermally induced anatexis to generate post-­
collisional granitoid magmatism in the Central Cordillera. Such a scenario conforms
well with the punctuated nature of observed magmatism, as recorded within the
U-Pb age database vs. the proposed tectonic development of the region (Fig. 5.34),
in addition to explaining the observed lithogeochemical and isotopic trends, including
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 379

the dominantly negative εHfi values for the El Bosque Batholith. Indeed, Vinasco
(2018) interprets recent U-Pb (zircon) age and lithochemical data for the alkaline
Sucre (Antioquia) intrusions, to suggest the initiation of delamination of the
Caribbean assemblage as early as 70 Ma. The Sucre intrusions produce very similar
age and lithochemical data to that of the Irra Stock as presented herein (Figs. 5.16
and 5.17) and suggest these plutons represent the early delamination suite, emplaced
along the Romeral suture boundary.
We note that the differentiation of granitoids resulting from thermal heat trans-
fer between the mantle and lower crust during asthenospheric upwelling, from
more typical subduction-related granitoids, using basic lithogeochemical and iso-
topic analyses, is not necessarily straightforward, especially in the presence of tec-
tonically thickened continental crust, where enhanced degrees of crustal anatexis,
assimilation or contamination, may be intuitively suspected. A nearby example of
this situation has already been revealed in the evolving petrogenetic interpretation
of the early Jurassic granitoids of the Santander Plutonic Group, within the
Santander Massif, where historic interpretations (e.g. Goldsmith et  al. 1971;
Aspden et al. 1987; Dorr et al. 1995) of relatively “typical” subduction-related pet-
rogenesis have been supplanted by a model involving partial fusion of lower crustal
source rocks by asthenospheric upwelling and heat transfer, as revealed by
advanced lithogeochemical and isotopic studies, including Lu-Hf isotope analyses
(e.g. Van der Lelij 2013).
In either case, Paleocene-Eocene granitoid magmatism along the Central
Cordillera was short-lived and was abruptly extinguished again at ca. 52 Ma. As
with the shutdown of the Antioquian Batholith, extinction of the Paleocene-Eocene
arc appears to be associated with collision of another oceanic ridge, in this case
represented by the Gorgona Terrane, which was accreted to the Colombian Pacific
margin in the Eocene (Cediel et  al. 2003; Kerr and Tarney 2005). Following
emplacement of the ca. 62–52  Ma, post-collisional, Eastern Group granitoids, a
resumed, ca. 30 m.y. hiatus in subduction-related granitoid magmatism is observed,
as recorded by the absence of significant volumes of observable granitoids through-
out central continental Colombia, dating from a period extending from the early
Eocene (ca. 52  Ma) to the latest Oligocene (Figs.  5.3, 5.4, 5.16, and 5.23; Leal-­
Mejía 2011).
Although temporally related to the ca. 62–52 Ma Eastern post-collisional granit-
oid subgroup, the emplacement kinematics of the Santa Marta Batholith granitoids
are clearly distinct from those of the Central Cordillera. The detailed studies of
Mejía et al. (2008), Duque (2009) and Cardona et al. (2011) provide insight into the
tectono-magmatic evolution of Paleogene granitoids along the NW apex of the
Sierra Nevada de Santa Marta. Duque (2009) and Duque-Trujillo et al. (2018) iden-
tified various intrusive phases within and surrounding the Santa Marta Batholith,
ranging in age from ca. 64 to 50  Ma. These authors conclude that an early (ca.
64–62  Ma) suite of volumetrically minor, peraluminous leucogranites (e.g. Playa
Salguero pluton) were probably derived via anatexis of local amphibolite basement
and are petrogenetically unrelated to the Santa Marta Batholith suite per se. They
emphasize the localized, punctuated and short-lived nature of the Santa Marta
380 H. Leal-Mejía et al.

arc segment and the absence of post-ca. 50 Ma granitoid magmatism in the region,
following closure of the Santa Marta arc system, and conclude that Santa Marta
suite magmatism was not associated with a long-lived or well-established subduc-
tion zone. A model involving the forced underthrusting of thickened, buoyant oce-
anic lithosphere beneath the apex of the Santa Marta Massif was proposed. Duque
(2009) and Duque-Trujillo et al. (2018) relate the emplacement of the ca. 64–62 Ma
peraluminous leucogranites hosted within the Gaira Group accretionary complex
(Cediel and Cáceres 2000) to the partial fusion of amphibolitic basement due to the
initial interaction of the Farallon-CCOP/CLIP assemblage with the northern
Colombian margin, followed by main-phase emplacement and NE migration of
Santa Marta suite-related magmatism between ca. 58 and 50 Ma. This model is in
keeping with previous interpretations of the kinematics and temporal development
of the Gaira Group accretionary prism and Santa Marta batholith, as presented by
Cediel and Cáceres (2000) and Cediel et al. (2003; see Cediel 2018), in which detach-
ment and NW migration of the Maracaibo Sub-plate beginning in the Paleocene
(Fig.  5.34) resulted in the localized forced underthrusting of CCOP/CLIP crust,
metamorphism within the Gaira Group and punctual granitoid magmatism, as
recorded within the Playa Salguero pluton, Santa Marta Batholith and associated
plutons.

5.4.3.4  T
 ectonic Setting for the Ca. 62–40 Ma Mandé-Acandí Western
Group Arc Segment

The Paleocene-Eocene Mandé-Acandí arc assemblage (Fig.  5.15), including the

metaluminous, low-K calc-alkaline (calcic) Mandé and Acandí batholiths, volcanic
and pyroclastic rocks of the La Equis-Santa Cecilia Fms. and hypabyssal porphyry
centres at Pantanos-Pegadorcito, Murindó and Acandí and elsewhere, represents the
most significant expression of granitoid magmatism within the Western Group of
granitoids. It is the only assemblage within the ca. 100–40 Ma suite for which a
coeval volcanic member is preserved. Geological, lithogeochemical and isotopic
data for the holocrystalline Mandé-Acandí Batholith and associated hypabyssal
porphyritic rocks is consistent with an origin within an intra-oceanic arc, emplaced
within CCOP/CLIP crust as represented by the El Paso Terrane-Baudó Complex
(Cediel el at. 2009; Montes et al. 2012; Cediel 2018). Figure 5.34 depicts the pre-
docking, intra-oceanic configuration Mandé-Acandí arc and host terranes. These
same litho-tectonic units are depicted within the detailed time-space analysis pre-
sented in Fig. 5.35.
Based upon schematic reconstructions depicting the tectonic evolution of NW
South America and offshore Pacific and Caribbean domains during the Paleogene,
the Mandé-Acandí arc is contained within the trailing edge of the CCOP/CLIP
plateau, associated with a SW-facing (NE-verging), intra-oceanic subduction zone,
contained within the Farallon Plate (e.g. Aspden et al. 1987; Pindell and Kennan
2001; Kennan and Pindell 2009; Wright and Wyld 2011; Montes et  al. 2012;
Nerlich et al. 2014; Weber et al. 2015). Additional intra-oceanic granitoids associ-
ated with this subduction zone, located in Central America, date from the
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 381

Cretaceous (Buchs et  al. 2010) and include the Middle American arc series, as
depicted, for example, by Pindel and Kennan (2001) and Wright and Wyld (2011).
Magmatism related to the northern (Panamanian) segment of the Mandé-Acandí
arc may have initiated as early as 62 to 59 Ma (Wegner et al. 2011; Montes et al.
2012) however published U-Pb (zircon) crystallization ages for granitoids from the
Acandí Batholith in Colombia range from ca. 50 Ma (Montes et al. 2012, 2015). To
the south, holocrystalline and porphyritic granitoids from the Pantanos-Pegadorcito
area return U-Pb dates of ca. 45  Ma, whilst granitoids collected on the southern
margin of the Mandé Batholith returned ages of ca. 43  Ma. Thus, U-Pb (zircon)
crystallization ages suggest Mandé-Acandí is a multiphase arc, emplaced over a
period of ca. 20 m.y., with an overall younging trend from north to south (Fig. 5.15).
We note that the flare-up of the Mande-Acandí arc segment is penecontemporane-
ous with the onset of strong dextral-oblique transpression, tectonic tightening and
uplift observed within the Colombian continental block, brought on by collision of
litho-tectonic components of the Farallon-CCOP/CLIP assembly (Cañas Gordas,
Dagua terranes) along the Colombian margin beginning at ca. 75 Ma. The ensuing
“tectonic lock-up” along the NW margin of South America during the late
Cretaceous-Paleocene may have played a role in the decoupling of the Farallon
Plate from the trailing edge of the CCOP/CLIP plateau and the development of the
Paleocene-Eocene segment of the Middle American Trench, along which east-­
directed subduction of Farallon crust beneath the trailing edge of the CCOP/CLIP
plateau resulted in emplacement of the Mandé-Acandí arc.
Nerlich et  al. (2014) reconstruct the genesis, evolution and migration of the
Caribbean plate/basin into the inter-American gap, based upon the Pacific hotspot
reference frame (Wessel and Kroenke 2008) and the Global Moving Hotspot
Reference Frame (Doubrovine et al. 2012). Nerlich et al. (2014) conclude that the
Caribbean Plate docks with the South America by ca. 54.5 Ma, roughly coincident
with the switch from divergence to convergence between North and South America
(Müller et  al. 1999). Docking at 54.5  Ma is in good agreement with schematic
­tectonic models depicting the Caribbean Plate reaching its near-final resting place
during the Eocene (e.g. Pindell and Kennan 2001; Kennan and Pindel 2009).
Following docking of the Caribbean Plate, granitoid magmatism does not reap-
pear in the Colombian Andes until the Oligo-Miocene. The re-establishment of sub-
duction along the Colombian Trench and the continued convergence between the
South American Plate and the trailing edge of the CCOP/CLIP plateau are aspects
of the late Northern Andean Orogeny, described forthwith in Sect. 5.4.4.

5.4.4  T
 he Late Northern Andean Orogeny: Tectonic
Framework and Evolution for Latest Oligocene
to Plio-­Pleistocene Granitoids

As documented above, following emplacement of the ca. 62–50  Ma, post-­

collisional, Eastern Group granitoids, the observable early Eocene to latest
Oligocene geological record of the continental Colombian Andes is marked by the
382 H. Leal-Mejía et al.

absence of significant volumes of granitoids (e.g. Cediel and Cáceres 2000; Gómez
et  al. 2015a), suggesting a hiatus in subduction-related granitoid magmatism or
continental arc development during this period. It is feasible that, should such mag-
matism have existed, it could have been erased by subsequent uplift and erosion
during the later phases of the Northern Andean Orogeny. Such a contention, how-
ever, is not supported by the limited available, albeit localized, detrital zircon stud-
ies from the Colombian Andes (e.g. Nie et al. 2012; Saylor et al. 2012), which,
conversely, reveal the absence of 50 to 20 Ma detrital zircon populations. Based
upon the foregoing, we conclude that granitoid magmatism associated with sub-
duction along the Colombian Pacific margin effectively terminated with the
emplacement of the ca. 62–52  Ma post-collisional arc and accretion of the
Gorgona Terrane. Subsequent tectonic development during the ensuing ca.
30  m.y. magmatic hiatus is characterized by continued dextral compression,
transform faulting and plate reorganization along the Pacific margin and struc-
tural tightening throughout continental Colombia (Cediel et  al. 1994; Cediel
et al. 2003; Kennan and Pindell 2009; Cediel 2018).
Beginning at ca. 24 Ma, granitoid magmatism reappears in the south–western-
most Colombian Andes (Fig. 5.36), hosted within CCOP-/CLIP-related rocks of the
Dagua terrane (accreted in the late Cretaceous-Paleocene) but well to the south of
the location of the proposed trailing edge of the Caribbean Plate in the late Oligocene
(e.g. Cediel and Cáceres 2000; Pindell and Kennan 2001; Kennan and Pindell 2009;
Hastie and Kerr 2010; Montes et al. 2012; Nerlich et al. 2014). This new phase of
granitoid magmatism signals the reactivation of arc development within continental
and western Colombia, which dominates the Neogene tectono-magmatic develop-
ment of the region during the late Northern Andean Orogeny, following the early
Eocene docking of the Caribbean Plate (Nerlich et al. 2014)
Detailed time-space analysis based upon U-Pb (zircon) crystallization ages for
latest Oligocene through Miocene and Plio-Pleistocene granitoids and associated
volcanic rocks throughout the Colombian Andes (Figs. 5.22, 5.36, and 5.37) dem-
onstrates that extensive, composite “Neogene” arc magmatism recorded on regional
geologic maps of the physiographic Central and Western Cordilleras, along the
Cauca and Patia valleys and elsewhere (e.g. Cediel and Cáceres 2000; Gómez et al.
2015a), in fact consists of a complex distribution of magmatic arc segments, the
location of which is observed to migrate in time and space, in both an overall south-­
to-­north and west-to-east pattern (Cediel et al. 2003; Leal-Mejía 2011). The genesis
and spatial evolution of these arc segments may in turn be attributed to (1) com-
plexities in the late Oligocene-Miocene collision between continental South
America and the trailing edge of the Caribbean Plate, resulting in accretion of the El
Paso-Baudó Terrane, and (2) the penecontemporaneous evolution of the eastern
Farallon (Nazca-Cocos) Plate along the Colombian Pacific margin. In view of these
factors, we present observations pertaining to the convergence and collision of the
South American Plate with the Caribbean plateau and to the Miocene evolution of
the easternmost Farallon Plate, as they pertain to the magmatic evolution of conti-
nental Colombia, prior to presenting conclusions pertaining to the temporal-spatial
evolution of granitoid arc segments in the onshore realm during the latest Oligocene,
Miocene and Plio-Pleistocene.
 5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic…

Fig. 5.37  Time-space analysis of latest Oligocene through Plio-Pleistocene granitoids in the Colombian Andes and surrounding region, in relation to tectonic
framework, major litho-tectonic elements and orogenic events. The age and nature of granitoid intrusive suites of the same time period are indicated. The profile
contains elements projected onto a ca. NW–SE line of section through west-central Colombia. (Litho-tectonic terrane and fault nomenclature modified after
Cediel et al. 2003. See text for additional details)
383
384 H. Leal-Mejía et al.

5.4.4.1  C
 onvergence and Collision Between South America
and the Trailing Edge of the Caribbean Plateau

Tectonic reconstructions demonstrate that most of the CCOP/CLIP components of

the Western Tectonic Realm, including the Dagua, Cañas Gordas, Gorgona and San
Jacinto terranes, were loosely in place within the near-shore realm by the late
Oligocene to early Miocene (e.g. Cediel et  al. 1994; Pindell and Kennan 2001;
Cediel et al. 2003; Kennan and Pindell 2009; Cediel 2018) (Figs. 5.34 and 5.36). To
the west, the El Paso-Baudó segment of the Chocó Arc (including the Mandé-
Acandí Batholith) was located within the peri-cratonic realm hosted upon the trail-
ing edge of the Caribbean Plate.
Detailed analysis of the structural and kinematic evolution of the eastern Panamá
Arc (i.e. Chocó Arc) for the late Eocene through Miocene, presented by Farris et al.
(2011) and Montes et al. (2012), depicts the WNW-ESE orientation of the El Paso-­
Mandé-­Acandí-Baudó assemblage along the trailing edge of the CCOP/CLIP pla-
teau, essentially colinear to the trend of the Middle American arc, in consort with
NE-directed subduction of Farallon crust, as universally depicted in regional tec-
tonic reconstructions spanning the mid-Cretaceous and Paleogene (e.g. Aspden
et al. 1987; Pindell and Kennan 2001; Cediel et al. 2003; Kerr et al. 2003; Kennan
and Pindell, 2009; Farris et al. 2011; Wright and Wyld 2011; Montes et al. 2012;
Nerlich et al. 2014; Weber et al. 2015).
Convergence between the El Paso-Baudó Terrane and the NW Colombian mar-
gin took place in the late Oligocene (Duque-Caro 1990; Pindell and Kennan 2001;
Cediel et al. 2003; Cediel et al. 2010). Based upon structural, lithogeochemical and
thermochronological data, Farris et al. (2011) present a model depicting the colli-
sion between South America and the southern Panama (i.e. Chocó) Arc. Their model
proposes the N and W convergence of the South American block upon the Chocó
Arc (as opposed to the continued N and E migration of the Caribbean plateau), prior
to collision beginning at 23–25 Ma. This proposal is in agreement with early struc-
tural data and conclusions regarding the vergence and evolution of the Panamá
thrust and fold belts (Silver et al. 1990), plate motion data for South America (Silver
et  al. 1998; Müller et  al. 1999) and early Eocene docking constraints for the
Caribbean Plate (Nerlich et al. 2014).
Additional lithostratigraphic, radiometric, paleomagnetic, structural and thermo-
chronological data presented by Montes et al. (2012) demonstrate initial NW-vergent
thrusting and clockwise rotation due to W-E convergence of the southwestern mar-
gin of the Chocó (El Paso-Baudó) assembly with South America during the
Oligocene. Following collision at 25–23 Ma (Farris et al. 2011), continued clock-
wise rotation and E-W convergence led to closure of the Central American seaway
by ca. 15 Ma (Montes et al. 2012). Accretion and obduction of the El Paso Terrane,
including severance of the Mandé-Acandí Arc from its Caribbean roots, took place
between 15 and 12  Ma, within the broad context of the development of the
Panamanian orocline (Silver et al. 1990; Farris et al. 2011), resulting in develop-
ment of the San Juan-Sebastian (Uramita) suture system and apparent NW vergence
of the Panamá thrust and fold belt (Figs. 5.34 and 5.36), during the mid- through late
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 385

Miocene (Duque-Caro 1990; Cediel et al. 2010; Montes et al. 2012). Uplift of the
western El Paso Terrane (Baudó Range) and development of the Atrato Basin took
place between ca. 8 and 4 Ma (Cediel et al. 2010) and appear to be a feature associ-
ated with the evolution of the Farallon and Nazca-Cocos plates and the re-­
establishment of subduction along the Colombian Pacific margin (Sect. 5.4.4.2).
It is important to note that convergence between the El Paso segment of the
Chocó Arc and the NW South American margin during the Oligocene to Miocene
left no apparent record of granitoid arc development within continental Colombia
(Central Tectonic Realm, Maracaibo Sub-plate), as may be inferred by the absence
of mapped Oligo-Miocene granitoids (e.g. Cediel and Cáceres 2000; Cediel et al.
2003; Leal-Mejía 2011; Gómez et al. 2015a), or significant Oligo-Miocene detrital
zircon populations (Nie et al. 2012; Saylor et al. 2012). We interpret this observation
to reflect the near in situ emplacement of the Mandé-Acandí Arc, riding passively
within/upon the trailing edge (El Paso Terrane segment) of the Caribbean Plate.
This assemblage developed as an intact member of the CCOP/CLIP plateau as a
whole and was little transported following docking of the Caribbean Plate at ca.
54.5 Ma (Nerlich et al. 2014). We conclude that Oligocene-Miocene convergence
between the El Paso assemblage and NW South America did not involve develop-
ment of a significant intra-plate subduction zone between the Caribbean plateau and
the Colombian continental margin. This conclusion is in keeping with the argu-
ments involving the buoyancy of CCOP/CLIP lithosphere (Molnar and Atwater
1978) and with the interpretation of various authors limiting interaction of the
Caribbean-NW Colombian margin to a model involving the amagmatic, limited,
SE-directed forced underthrusting of thick CCOP/CLIP lithosphere beneath the
South American margin during this time period (Van der Hilst 1990; Van der Hilst
and Mann 1994; Cediel et al. 2003; Kerr et al. 2003; Farris et al. 2011).

5.4.4.2  E
 volution of the Farallon-Nazca-Cocos Plate System and Neogene
Reinitiation of Subduction in the Colombian Andes

Coeval with Oligo-Miocene tectonic development in NW Colombia, east-directed

subduction of the Farallon Plate beneath the Colombian Pacific margin was re-­
established by ca. 24 Ma (Leal-Mejía 2011). Throughout the Miocene and into the
Plio-Pleistocene, pulses of metaluminous, calc-alkaline magmatism in temporally
and geographically distinct volcano-magmatic arc segments are revealed by U-Pb
(zircon) crystallization ages for subduction-related granitoids (Figs.  5.36 and
5.37). This granitoid magmatism, which includes localized coeval volcanism, was
emplaced within metamorphic rocks of the Central Tectonic Realm and Maracaibo
Sub-plate and within more recently accreted oceanic rocks comprising the Western
Tectonic Realm. These basement complexes are all considered to have formed
part of the Colombian accretionary mosaic at the time of emplacement of their
contained latest Oligocene through Plio-Pleistocence granitoids, and hence the
entire Neogene granitoid suite is considered autochthonous with respect to the
continental margin.
386 H. Leal-Mejía et al.

At the end of the Oligocene, the triple junction between the Farallon, South
American and Caribbean plates was located in the near-shore Colombian Pacific,
along the south–westernmost margin of the Panamá-Choco Arc (e.g. Pindell and
Kennan 2001; Cediel et  al. 2003; Lonsdale 2005). Accretion-related transform
faults of the Garrapatas and San Juan-Sebastian sutures (including the southern
Uramita Fault of Duque-Caro (1990) and Montes et al. (2012)), marking the south-
ern margin of the Choco Arc terranes (i.e. the trailing edge of the Caribbean Plate),
were already established as broadly NE-SW corridors of crustal-scale weakness
(e.g. Barrero 1977; Duque-Caro 1991; Cediel et al. 2003) (Figs. 5.34 and 5.36). At
ca. 23 Ma, the Farallon Plate splits to form the Nazca and Cocos plates along a ca.
E-W rift that also extended into the Colombian Pacific, in the vicinity of the junc-
tion between the Middle American and South American subduction zones (Pindell
and Kennan 2001; Lonsdale 2005). Continued plate reorganization in the Pacific
realm (Lonsdale 2005) and W-directed motion of the South American Plate (Silver
et al. 1998; Farris et al. 2011) resulted in near-orthogonal convergence between the
Farallon Plate and the Colombian margin. Mid-Miocene rifting within the Nazca
Plate is marked by the formation of the E-W-oriented Sandra Rift off the Colombian
Pacific margin, which presently separates the Coiba microplate to the north from
Malpelo Ridge and associated crust to the south (Lonsdale 2005) (Fig. 5.36). Ocean
crust associated with seafloor spreading along the Sandra Rift dates from between
ca. 14 and 9 Ma (Lonsdale 2005) and comprises the oceanic slab juxtaposed along
the present-day northern Colombian Trench between ca. 5°N and 8°N. To the south,
similar crust of somewhat older (ca. 14–18 Ma) age is preserved (Lonsdale 2005).
Within the Colombian onshore realm, the analysis of earthquake hypocentral
solutions, gravity and magnetic data, tomographic imaging, petrogenetic data and
the distribution of modern-day volcanic activity has led numerous authors in recent
years to present models for Miocene to recent subduction beneath continental
Colombia (Santô 1969; Dewey 1972; Pennington 1981; Van der Hilst and Mann
1994; Taboada et  al. 2000; Sarmiento 2001; Zarifi et  al. 2007; Vargas and Mann
2013; Bissig et al. 2014; Chiarabba et al. 2015). Many of these studies are attempts
to reconcile modern-day earthquake activity observed in the Santander Massif
(Bucaramanga seismic nest), with the distribution of modern-day volcanic activity,
and localized late Miocene to Pliocene magmatism observed in the Vetas-California
area of the Santander Massif (Mantilla et al. 2009; Bissig et al. 2014) and at Paipa-­
Iza in the northernmost Eastern Cordillera (Floresta Massif) (Pardo 2005a, b). All
of the foregoing authors agree that available data suggests eastward to south-east-
ward subduction of a composite oceanic slab comprised of the segmented,
Miocene-age, Nazca Plate. Some authors suggest the Nazca Plate is undergoing
down-slab interaction beneath continental Colombia with CCOP/CLIP oceanic
crust of Cretaceous age (e.g. Pennington 1981; Taboada et  al. 2000; Zarifi et  al.
2007; Vargas and Mann 2013).
Seismic tomography and additional geophysical data presented by Pennington
(1981), Taboada et  al. (2000), Sarmiento (2001), Zarifi et  al. (2007), Vargas and
Mann (2013) and Chiarabba et al. (2015) have been interpreted to reflect an E-W
discontinuity or tear in the oceanic slab presently subducting beneath western
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 387

Colombia (e.g. the Caldas tear of Vargas and Mann 2013) (Fig. 5.36). This discon-
tinuity is inferred to be located between ca. 4.8°N and 5.2 °N, broadly coincident
with the southern end of the Serranía de Baudó (Taboada et al. 2000) and the ENE
striking segment of the San Juan Sebastian-Uramita suture system (Cediel et  al.
2010; Montes et al. 2012). Vargas and Mann (2013) and Chiarabba et al. (2015) note
that the discontinuity also coincides with the interpreted western projection of fea-
tures located within the subducting Nazca plate, including the Sandra Rift and the
Coiba Transform fault, respectively.
Santô (1969), Dewey (1972), Pennington (1981), Van der Hilst and Mann (1994),
Taboada et al. (2000), Sarmiento (2001), Cediel et al. (2003), Zarifi et al. (2007),
Vargas and Mann (2013), Bissig et al. (2014) and Chiarabba et al. (2015) interpret
the geometry and nature of the oceanic slab segments presently subducting beneath
continental Colombia, on either side of the E-W discontinuity. All authors agree that
south of ca. 5°N, the Nazca Plate is undergoing moderately steep subduction, at an
angle of between ca. 30° and 40°, steepening to >50° beneath the Eastern Cordillera.
This southern segment of “normally” dipping Nazca crust was referred to as the
Cauca segment by Pennington (1981). Active volcanism associated with Cauca seg-
ment subduction manifests in the Colombian portion of the Northern Andean volca-
nic zone, which (coincidentally) terminates at about 5°N (Figs. 5.3 and 5.36).
North of 5°N, variable interpretations of the nature and geometry of subducted
oceanic crust have been presented. Beneath the eastern Colombian Andes, seismic
data and tomographic imaging are suggestive of a dipping slab, which has long been
interpreted to be associated with abundant earthquake activity surrounding the
Bucaramanga seismic nest (Santô 1969; Dewey 1972; Pennington 1981). Pennington
(1981) referred to this shallowly dipping slab as the Bucaramanga segment. Some
authors interpret this segment to represent CCOP/CLIP lithosphere, which in turn is
interpreted to represent the down-slab prolongation of late Cretaceous CCOP/CLIP
crust exposed within the Chocó Arc (El Paso-Baudó Terrane) (Pennington 1981;
Taboada et al. 2000; Sarmiento 2001; Zarifi et al. 2007; Vargas and Mann, 2013;
Bissig et al. 2014). However, the sparse and discontinuous nature of seismic activity
recorded along the interpreted up-dip segment of the Bucaramanga slab (see pro-
files presented by Pennington 1981; Taboada et al. 2000; Vargas and Mann 2013;
Chiarabba et al. 2015) led authors to suggest that the Bucaramanga segment is no
longer connected to surface plates (Santô 1969; Dewey 1972; Sarmiento 2001; also
see Plate 2C of Taboada et al. 2000; and Fig. 5.5 of Vargas and Mann 2013).
Other authors (e.g. Pennington 1981; Zarifi et al. 2007; Vargas and Mann 2013)
suggest “apparent” up-dip continuity between the Bucaramanga segment and
inferred, shallowly-dipping Cretaceous CCOP/CLIP lithosphere beneath the Central
Cordillera, which in turn would be connected to CCOP lithosphere exposed in the
Chocó Arc. These authors interpret the abrupt northward termination of the North
Andes volcanic arc at ca. 5°N to reflect amagmatic, flat-slab subduction of CCOP
lithosphere.
Notwithstanding, 600  km to the west of the Santander Massif, along the
Colombian Trench, various authors suggest the Miocene, Coiba microplate segment
of the Nazca Plate is (also) undergoing eastward subduction, between ca. 5°N and
388 H. Leal-Mejía et al.

ca. 8°N (Aspden et al. 1987; Van der Hilst and Mann 1994; Taboada et al. 2000;
Cediel et al. 2003; Lonsdale 2005; Vargas and Mann 2013; Chiarabba et al. 2015).
Van der Hilst and Mann (1994) and Chiarabba et al. (2015) argue that the Coiba
segment contains buoyant features, such as thickened volcanic ridges, and is sub-
ducting at a lower angle when compared with the Cauca segment to the south. These
authors attribute the lack of modern volcanic arc development N of 5°N, to amag-
matic, flat-slab subduction of the Coiba microplate. Indeed, the most recent model-
ling of seismic data presented by Chiarabba et  al. (2015) suggests down-slab
continuity of the Miocene Coiba microplate, extending from the Colombia trench
into the region beneath the northeastern Colombian Andes. Thus, Chiarabba et al.
(2015) present a simplified model involving massive, down-slab devolatilization of
the thickened, Miocene, Coiba microplate, which can equally be invoked to explain
the lack of arc-related volcanism in the up-slab section to the N of 5°N, seismic
activity in the Bucaramanga nest and localized granitoid magmatism in the Santander
Massif. In this context, the model of Chiarabba et al. (2015) is more in line with
earlier modelling and arguments presented by Van der Hilst and Mann (1994) which
suggest the Nazca, and not Caribbean Plate, is subducting beneath NW Colombia
and that the presence of continuous or fragmented CCOP crust is not required to
explain the observed seismic or magmatic phenomena. These and other authors
(e.g. Cediel et al. 2003; Farris et al. 2011) suggest the Caribbean Plate is undergo-
ing, rather, S- to SE-directed forced underthrusting along the western Colombian
Caribbean margin (Fig. 5.36).
Interestingly, beyond observations regarding the absence vs. presence of modern-­
day arc-related volcanism, N and S of ca. 5°N, respectively, few proponents of the
various subducting slab models have taken into account the evolution and spatial
migration of subduction-related granitoid arc segments manifest within the western
Colombian Andes during the latest Oligocene through Miocene. This period coin-
cides with the birth and growth-related architectural evolution of the Nazca Plate
(Lonsdale 2005) and with the reinitiation of subduction-related granitoid magma-
tism throughout western Colombia, leading to the conformation of the Colombian
segment of the North Andes volcanic arc (Aspden et al. 1987; Cediel et al. 2003;
Leal-Mejía 2011). Aspden et al. (1987) highlighted the presence of Neogene gran-
itoids along the Western Cordillera and Cauca-Patia intermontane valley, which
they considered to be associated with late Oligocene to present subduction along the
entire Colombian Pacific margin. Taboada et al. (2000) related late Miocene mag-
matism observed along the Western Cordillera and Romeral mélange, between 5°N
and 7°N, to the development of a wedge of hot asthenosphere which favoured melt-
ing and granitoid magmatism between the subducting Nazca Plate and accreted
sections of the CCOP/CLIP. They suggest that the presence of CCOP/CLIP litho-
sphere beneath the Central Cordillera to the east acts as a shield which prevents the
penetration of rising melts and as such explains the absence of active volcanism to
the N of 5°N.  Cediel et  al. (2003) provided an explanation for the punctuated
emplacement and spatial-temporal evolution of Miocene holocrystalline and
porphyry-­related arc segments in western Colombia, based upon the Miocene tec-
tonic assembly of the region, involving the differential subduction of Nazca Plate
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 389

crust on either side of the paleo-Garrapatas transform fault. Chiarabba et al. (2015)
suggest that the Coiba segment of the Nazca Plate initially underwent normal, mod-
erate- to high-angle subduction, which could explain the development of subduction-­
related granitoids between 5°N and 7°N.  According to their model, the entry of
buoyant material into the Colombia trench at ca. 10 Ma led to enhanced tearing of
the Nazca slab along pre-established planes of weakness (e.g. the Coiba transform
fault), the onset of low-angle subduction N of ca. 5°N and the consequent cessation
of eastward-progressing magmatism, explaining the absence of the modern-day vol-
canic activity associated with the subducting Coiba segment.
As outlined in detail in Sect. 5.3.5 and within Figs. 5.22, 5.23, and 5.36, at least
six granitoid arc segments/clusters of latest Oligocene through Miocene and Plio-­
Pleistocene age, and of significant length, continuity and outcropping area, are
recorded within Colombia’s physiographic Western and Central Cordilleras and
along the Cauca-Patia intermontane valley. We consider the modern-day Colombian
portion of the Northern Andean volcanic zone (which itself is segmented; Cediel
et al. 2003; Stern 2004; Marín-Cerón et al. 2018), to represent a temporally separate
arc segment, although it is cospatial with, and locally superimposed upon, Mio-
Pliocene segments. In addition to the above, within Colombia’s eastern cordilleran
system, isolated granitoid occurrences of Miocene and Pliocene age are recorded at
Vetas-California in the Santander Massif and Paipa-Iza and Quetame in the Eastern
Cordillera. With respect to the generally N-S- to NNE-oriented axis of the Miocene
arc segments and the NNE trend of the modern-day Northern Andean volcanic zone,
the Vetas-Paipa-Quetame occurrences are located well to the east (on average over
150 km east) of the magmatic arc axis. We do not consider the Vetas-­Paipa-­Quetame
granitoids to constitute a definable arc segment. They are low volume, localized
occurrences which, based upon clear differences in lithochemistry and widely
spaced, non-coaxial distribution, are considered outliers with respect to the mag-
matic trends of central and western Colombia.
In the south, the ca. 23–21 Ma Piedrancha-Cuembí holocrystalline suite and the
ca. 18–9 Ma Upper Cauca-Patía porphyry suite are associated with subduction of
the southern, Cauca segment of the Nazca plate. Continued eastward migration of
the granitoid volcanic arc axis is recorded in the southern portion of the active
Colombian volcanic arc (e.g. San Roque, Huila volcanoes). To the north, the ca.
12–10  Ma Farallones-El Cerro holocrystalline suite and the ca. 9–5  Ma Middle
Cauca porphyry suite are associated with subduction of the Coiba segment.
Granitoid magmatism related to the continued subduction of the Coiba segment
records eastward migration of arc-axial magmatism, observed in the late Miocene
Cajamarca-Salento hypabyssal porphyry cluster and the Plio-Pleistocene Río Dulce,
both located within the Cajamarca-Valdivia basement rocks of the Central Cordillera.
Both of these granitoid clusters are essentially coaxial with the northern portion of
the active Colombian volcanic arc. The active Machín volcano is located on the
eastern margin of the Cajamarca-Salento porphyry cluster, whilst Plio-Pleistocene
to recent volcanic cover from the Nevado del Tolima limits exposure of this same
porphyry cluster immediately to the north. Along trend to the NNE at Río Dulce,
Plio-Pleistocene ages for hypabyssal granitoid intrusive and associated volcanic
390 H. Leal-Mejía et al.

rocks compare well with similar ages for volcanic materials from the Ruíz, Santa
Isabel and Tolima stratovolcanic complexes (Maya 1992). In this context, Rio Dulce
(ca. 5.7°N) may be interpreted to represent the northernmost expression of volca-
nism associated with the modern-day Colombian volcanic arc (Fig. 5.36). We inter-
pret the eastward migration of both the Cauca and Coiba-related arc axial magmatism
to reflect progressive shallowing of the subduction angle of the respective segments
of the Nazca oceanic crust, associated with the consumption of progressively
younger and thermally buoyant Nazca Plate lithosphere (Lonsdale 2005) probably
augmented by the entrance of buoyant aseismic features such as the Carnegie and
Sandra Ridge into the (Ecuador-)Colombia trench, effectively inhibiting the sub-
duction process (e.g. Chiarabba et al. 2015).
Composite lithogeochemical and isotopic data presented herein permit interpre-
tation of all of granitoid suites emplaced along the western Colombian convergent
margin during the Mio-Pliocene, including those located to the N of ca. 5°N (i.e. the
ca. 12 Ma Farallones-El Cerro trend and the ca. 9–5 Ma Middle Cauca trend), as
mantle-derived, metaluminous, calc-alkaline granitoids typical of subduction-­
related suites. It may be observed that, aside from differences in age, the granitoid
suites comprising the various Colombian arc segments of western Colombia are
very similar in major, minor, trace element, and isotopic composition. All of the
suites demonstrate typical gabbro through granodiorite trends with strongly man-
telic compositions and, in no instance, are enhanced levels of crustal contamination
(e.g. peraluminous tendencies, anomalously high Sr isotope compositions) implicit
in the petrogenetic trends demonstrated by the data set.
Based upon the above arguments, we interpret Neogene granitoid magmatism
throughout western Colombia (i.e. the Western and Central Cordilleras and Cauca-­
Patía intermontane valley) to be the result of the subduction of composite Nazca
crust beneath the composite Colombian margin since the late Oligocene. Differences
in the rate and style of east-dipping subduction on either side of the Cauca-Coiba
slab tear, beginning in the latest Oligocene, are reflected in the complex spatial and
temporal distribution of Colombian onshore volcano-plutonic arc magmatism
throughout the early Miocene and Plio-Pleistocene to recent (e.g. Cediel et  al.
2003). We conclude that all of the western Colombian granitoid arc segments/clus-
ters were emplaced following passage and docking of the trailing edge of the
Caribbean Plate and do not represent the subduction of Farallon-CCOP/CLIP
assemblage lithosphere per se.
Mio-Pliocene granitoids of Colombia’s Eastern Cordilleran system, including
those of the Vetas-California area in the Santander Massif and at Paipa-Iza and
Quetame, within the Eastern Cordillera sensu stricto, form volumetrically small and
isolated occurrences located over 150 km east of the subduction-related magmatic
axis defined by the active Colombian volcanic arc. Available lithogeochemical data
for this group of granitoids is incomplete and does not permit a full analysis of the
petrogenesis of these occurrences nor a complete comparison amongst themselves.
Miocene granitoid magmatism in the Santander Massif ranges from ca. 14 to
9 Ma (Mantilla et al. 2009; Leal-Mejía 2011; Mantilla et al. 2013; Bissig et al. 2014;
Cruz et  al. 2014) although recent studies suggest that unexposed magmatism of
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 391

Pliocene age is likely present at shallow depth below the trend (Rodríguez, 2014;
Rodríguez et al. 2017). The lithogeochemical and isotopic database for the Vetas-­
California granitoids is fairly complete, and the clear evolution of the suite to more
evolved (siliceous, alkaline, peraluminous) compositions when compared to the
Neogene porphyritic granitoids of western Colombia is evident (Figs.  5.26 and
5.27). Bissig et al. (2014) indicate that the hydrous, oxidized Vetas-California por-
phyries evolved from mantle-derived melts which have assimilated moderate
amounts of crustal material, potentially including Guiana Shield, granulite belt and/
or Paleozoic supracrustal rocks, typical of the basement assembly of the Santander
Massif. Bissig et al. (2014) provide radiogenic Sr, Nd and Pb isotope data which
suggest the Miocene granitoids contain juvenile material, unlike the Paleozoic and
Jurassic granitoids of the area which, based upon radiogenic Lu-Hf isotope analy-
ses, appear to primarily represent recycled ca. 1 Ga continental crust (Van der Lelij
2013; Cochrane 2013). Lu-Hf isotope analyses for the Miocene Vetas-California
porphyry suite have yet to be performed.
Within the Eastern Cordillera, some 175 and 360  km the south of Vetas-­
California, respectively, the granitoids of Paipa-Iza and Quetame reveal additional
isolated, low-volume occurrences of Mio-Pliocene granitoids situated significantly
to the east of the active Colombian volcanic arc axis. Major element lithogeochemi-
cal data from Paipa (Pardo et  al. 2005b) indicate ferroan, alkalic, peraluminous
compositions, dissimilar to the Vetas-California suite, atypical of Cordilleran
­granitoids and perhaps more akin to A-type granitoids, characteristic of melts gen-
erated in extensional environments (Frost et al. 2001). Lithogeochemical data for
Quetame is restricted to a single major-element analysis of ca. 5.6 Ma felsic por-
phyry (Ujueta et al. 1990). The analysis reveals attributes of both the Vetas-California
and Paipa suites; however, it is difficult to draw any firm conclusions based upon a
single major element lithogeochemical analysis. Notwithstanding, Pardo et  al.
(2005b) and Ujueta et al. (1990) conclude that the lithogeochemical data for both
Paipa (Iza) and Quetame, respectively, is markedly distinct from the typically calc-
alkaline (calcic to calc-alkalic after Frost et al. 2001) compositions revealed along
the active Colombian volcanic arc to the west.
From a petrogenetic standpoint, direct comparison of the Paipa-Iza-Quetame
lithogeochemical data with that of Vetas-California is difficult due to the lack of
trace, REE and radiogenic isotope data at Paipa-Iza-Quetame. From a major-­
element standpoint, however, the ferroan, alkalic nature of the Paipa-Iza granitoids
contrasts markedly with the magnesian-calc alkalic suite from Vetas-California
(Fig. 5.26), and if the Eastern Colombian granitoids of Mio-Pliocene age, although
relatively widely space in occurrence, are considered as a whole, the suite may be
considered to provide a bimodal distribution in terms of observed major element
lithochemistry.
Aside from lithogeochemical comparisons, the Vetas-California-Paipa-Iza-
Quetame suites share an important relationship with respect to distribution and
structural controls. The occurrences are aligned along a ca. NNE-axis, whose trace
is approximately parallel with respect to, and located east of (i.e. in the back-arc),
the ca. NNE-oriented axis of the active Colombian volcanic arc (Figs.  5.22 and
392 H. Leal-Mejía et al.

5.36). In addition, the Santander-Eastern Cordilleran granitoids are all located along
the trace of the Bucaramanga–Santa Marta–Garzón fault and suture system (Cediel
and Cáceres 2000; Cediel et  al. 2003) (Figs.  5.2 and 5.36), a long-lived, active,
crustal-scale feature with significant vertical continuity, which could have facili-
tated the emplacement of mantle-derived melts into the upper crust.
The regional tectonic setting and relationship of the isolated granitoid occur-
rences of Colombia’s eastern cordilleran system, to Mio-Pliocene granitoid magma-
tism and active Andean-style volcanism related to Nazca Plate subduction in the
western and central Colombian Andes, have yet to be fully established. Taken as a
whole, the present geographic position of these occurrences locates them in a back-­
arc position and along a sub-parallel NNE trend to the magmatic axis of the active
Colombian volcanic arc. The bimodal lithogeochemical composition of the
Santander-Eastern Cordillera occurrences suggests the suite as a whole may be rift-­
related. Based upon the foregoing, we suggests the Vetas-California-Paipa-Iza-
Quetame granitoids could represent indications of crustal extension, focussed along
the active Bucaramanga–Santa Marta–Garzón fault system, and rift-related magma-
tism within the back-arc of the Northern Andean volcanic zone in Colombia.
Figures 5.36 and 5.37 demonstrate complexities of the nature, geometry and tim-
ing of latest Oligocene-Miocene to Plio-Pleistocene magmatic arc development and
granitoid magmatism in Colombia. It can be observed that extensive, composite
“Neogene” granitoid magmatism in Colombia is in fact composed of a series of
more spatially temporally limited arc segments, including a bimodal suite of outlier
occurrences located in the back-arc region. Granitoid magmatism demarcating the
composite arc is observed to migrate in time and space, in both a south-to-north and
west-to-east sense. The emplacement, localization and lithochemistry of the numer-
ous arc and outlier segments were influenced by the nature and composition of vari-
ous basement complexes, facilitated by the location and reactivation of paleo-fault
and suture systems throughout the Colombian Andes.

5.5  Summary and Concluding Statement

Plutonic and hypabyssal porphyritic granitoids and locally their volcanic equiva-
lents constitute important components of the geological record of the Colombian
Andes, not only from a volumetric standpoint but additionally as a reflection of the
complex, diverse and dynamic tectonic evolution of the region. Based upon the
composite U-Pb (zircon) age date database ca. 1995–2017, the analysis presented in
this chapter has identified six principle episodes of Phanerozoic granitoid magma-
tism including early Paleozoic (ca. 485–439 Ma), Carboniferous (ca. 333–310 Ma),
Permo-Triassic (ca. 289–225 Ma), latest Triassic-Jurassic (ca. 210–146 Ma), late
Cretaceous to Eocene (ca. 100–42 Ma) and latest Oligocene to Mio-Pliocene (ca.
23–1.2  Ma). A continuum of this last episode into the Plio-Pleistocene through
Recent manifests in the modern-day Colombian (Northern Andean) volcanic arc.
The spatial distribution and analytical resolution of the U-Pb (zircon) database
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 393

permit the identification of subpopulations within the major granitoid episodes and,
in turn, a detailed analysis of the spatial and temporal migration of granitoid mag-
matism during the entire Colombian Phanerozoic.
Our analysis has integrated the major granitoid episodes into the pre-Northern
Andean Orogeny, proto-Northern Andean Orogeny and Northern Andean Orogeny
phases of Colombian tectonic evolution, and, supported by lithogeochemical and
radiometric isotope data for many of the granitoid suites, we have used the granitoid
populations as indicators of the tectonic framework in which the granitoids were
generated and emplaced. Three pre-Northern Andean Orogeny granitoid popula-
tions are identified. Early Paleozoic granitoids of the Santander, Floresta and
Quetame massifs include (1) ca. 499–473 Ma syn-kinematic and peak metamorphic
granitoids, which are interpreted to have been generated/emplaced during a period
of compression, crustal thickening, Barrovian-style metamorphism and orogenesis;
(2) ca. 472–452 Ma granitoids, emplaced during post-orogenic collapse, extension
and basin formation; and (3) ca. 452–415 Ma granitoids emplaced during resumed
compression, basin closure and crustal thickening. The role of subduction per se in
the petrogenesis of the early Paleozoic granitoids has yet to be clearly established,
given that the entire ca. 485–439 Ma suite apparently represents primarily recycled,
crustal-derived melts with limited juvenile contribution. Processes as diverse as
crustal thickening, Barrovian-type metamorphism, extension, crustal thinning,
lithospheric mantle upwelling and heat advection at the base of the crust have been
invoked in the generation of these granitoids.
Early Paleozoic granitoid magmatism in eastern Colombia may have been
brought on by approach and accretion Cajamarca-Valdivia island arc complex dur-
ing the Quetame Orogeny. The Cajamarca-Valdivia Terrane, underlying much of
Colombia’s Central Cordillera, also contains an emerging population of similarly
aged early Paleozoic granitoids, which are only now beginning to be recognized.
Based upon current interpretations, these granitoids are considered allochthonous or
peri-cratonic with respect to their Santander-Floresta-Quetame Massif counterparts
and the continental Colombian tectonic mosaic as recorded during the early
Paleozoic.
Following the emplacement of the last of the Colombian early Paleozoic granit-
oids at ca. 439 Ma, the region entered an amagmatic phase extending through to ca.
333 Ma. This granitoid hiatus denotes the onset of the Bolívar Aulacogen, a pro-
longed period of continental taphrogenesis characterized by extensional tectonics,
the development of intra-continental and continental margin rifts and deposition of
epicontinental and marine sedimentary strata in the Carboniferous through Permian.
An initial record of rift-related magmatism beginning in the mid-Carboniferous
is recorded in the El Carmen-El Cordero gabbro-leucotonalite-trondjhemite suite
hosted within Cajamarca-Valdivia Terrane basement along the Otú Fault within the
Central Cordillera. The Carmen-El Cordero granitoids represent a Ridge Tholeiitic
Granitoid assemblage (Barbarain 1999) petrogenetically associated with oceanic
spreading and ophiolite formation. We suggest that the El Carmen-El Cordero suite
reflects the progressively extensional environment prevalent during of the interme-
diate stages of the Bolívar Aulacogen. In the first instance, however, activity along
394 H. Leal-Mejía et al.

the Otú rift was short-lived, and rifting was apparently aborted during a tectono-­
thermal event which affected much of the Northern Andean region, beginning in the
early Permian.
The early Permian tectono-thermal event, including the emplacement of ca. 289–
240  Ma syn-orogenic (±subduction-related?) peraluminous granitoid gneisses, is
associated with collision and crustal thickening during the amalgamation of western
Pangaea. As with the early Paleozoic granitoids, the Permian gneissic granitoids
appear to represent primarily recycled melts derived from S-type upper crustal
sources. Following ca. 240 Ma, the resumption of rifting is registered by a wide-
spread but low-volume bimodal suite of metaluminous tholeiitic amphibolites and
peraluminous anatectic granitoids, observed to intrude the Cajamarca-Valdivia
Terrane throughout much of the Central Cordillera but also recorded in the Santander
Massif and Upper Magdalena Basin. Both amphibolites and granitoids record an
increasingly juvenile composition over time. The emplacement of this rift-related
suite culminates in seafloor spreading after ca. 223 Ma and ocean crust formation by
ca. 216 Ma, as represented by the Aburrá (Santa Elena) ophiolite. As with the early
Paleozoic granitoid suite, the role of subduction and the contribution of subduction-­
derived magmatism in the petrogenesis of the Permian and mid-late Triassic peralu-
minous granitoids is uncertain. Processes including crustal thickening and anatexis
during continental amalgamation, and regional extension, crustal thinning and
basaltic underplating during continental break-up, have been suggested as root
causes for the generation of these granitoids.
In Colombia, we suggest that the understanding of Permo-Triassic granitoid
magmatism remains in many respects at a preliminary stage. The Permo-Triassic
granitoid suite is under-represented within the Colombian geological map base, as
many of these gneissic granitoids have been historically assigned to the early
Paleozoic or Precambrian or in the case of the southern Sonsón Batholith, to the
Jurassic. The further use of resilient U-Pb (zircon) dating techniques and the identi-
fication of new or mis-assigned Permo-Triassic granitoids will oblige a return to
field-based mapping in order to define the physical limits of the Permo-Triassic
intrusive suite. The accurate representation and interpretation of this important suite
on published geologic maps will in turn permit better understanding of the tectonic
development of the region as a whole during this time period.
Following incipient continental break-up and the formation of oceanic crust
along the Colombian proto-Pacific margin beginning in the mid-Triassic, regional
extension continued. The onset of the late Bolívar Aulacogen at this time is accom-
panied by a brief hiatus in granitoid magmatism, extending from ca. 225 to 210 Ma.
Resumption of granitoid magmatism in the latest Triassic is characterized by the
appearance of a complex spatial and temporal array of voluminous, continental arc
granitoids including coeval volcanic rocks, of mostly metaluminous composition,
which are interpreted to represent subduction-derived melts. These latest Triassic-­
Jurassic granitoids and volcanic rocks are quite unlike the low-volume peralumi-
nous granitoids characteristic of previous extensional phases. They characterize a
highly extensional but subduction-related regime dominant during the late Bolívar
Aulacogen.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 395

Notwithstanding, in northern Colombia, radiogenic isotope analyses (Lu-Hf,

Sr-Nd, Pb-Pb) suggest that even some metaluminous latest Triassic-early Jurassic
granitoids of the Santander Plutonic Group remain primarily comprised of recycled
mid-Proterozoic continental material and were generated via processes involving
regional extension, asthenospheric upwelling and thermal anatexis of the lower
crust. In this context the Santander granitoids may represent a highly contaminated
transitional suite, and not represent subduction-related melts per se. Continued
extension leads to westward migration of arc axial magmatism during the mid-­
Jurassic (Sierra Nevada de Santa Marta and San Lucas batholiths) and into the late
Jurassic (Segovia Batholith). Related granitoids demonstrate increasingly juvenile
compositions and diminishing isotopic contributions from the hosting basement
rocks and are considered to represent subduction-related melts emplaced within an
overall extensional regime associated with a westward-retreating trench and slab
rollback within the proto-Farallon Plate. The Segovia Batholith may in fact repre-
sent an eroded peri-cratonic island arc developed upon rifted Cajamarca-Valdivia
Terrane basement. In southern Colombia, mid- and late Jurassic magmatism is also
clearly temporally and spatially distinguished in the southern Ibagué, Mocoa-­
Garzón and northern Ibagué batholiths. The westward migration of arc-axial mag-
matism is less well defined, however, and granitoid plutonism migrates primarily
along the NNE axial trend of the granitoid arc segments. Significant volumes of
Jurassic volcano-sedimentary rocks (e.g. Noreán, Guatapurí, Saldaña Fms.) and
hypabyssal porphyritic granitoids are preserved within the Jurassic arc segments,
especially those of middle Jurassic age, an observation we interpret to reflect the
extensional environment of arc formation. All known occurrences of hypabyssal
porphyritic rocks associated with Jurassic holocrystalline batholiths (e.g. Santa
Cruz, Mocoa, Rovira) were emplaced within 2 to 5  m.y. of the shutdown of the
associated main phase holocrystalline batholiths.
Continued extension during culmination of the Bolívar Aulacogen led to the
development of a rifted continental margin floored by Proto-Caribbean oceanic
crust, the opening of the culminant intercontinental Valle Alto rift and the invasion
of the Cretaceous seaway over much of the region. An ensuing 50  m.y hiatus in
subduction-related magmatism from between ca. 145 and 95 Ma is indicated, based
upon the absence of significant granitoid occurrences of this age range throughout
continental Colombia.
Plate reorganization in the Pacific during the early Cretaceous led to the onset of
the early Northern Andean Orogeny, marked initially by dextral transpression and
the formation of blueschist assemblages along the Colombian Pacific margin begin-
ning prior to ca. 120 Ma and followed by the appearance of a complex assemblage
of subduction-related granitoids generated within both the autochthonous continen-
tal and allochthonous oceanic realms. The Eastern Group granitoids, including pri-
marily the ca. 96–72 Ma Antioquian Batholith and its satellite plutons, represent
continental arc magmas derived via the eastward, dextral-oblique subduction of
Proto-Caribbean ± leading-edge Farallon Plate crust beneath the Colombian conti-
nental margin. Antioquian Batholith magmatism was extinguished at ca. 72 Ma due
to the collision and accretion of CCOP-/CLIP-related terranes of the Western
396 H. Leal-Mejía et al.

Tectonic Realm (Cañas Gordas, Dagua, San Jacinto). A low-volume, short-lived

post-collisional arc was reignited within the Central Cordillera, to the south of the
Antioquian Batholith, between ca. 62 and 50 Ma (Sonsón, Manizales, El Hatillo, El
Bosque, Córdoba plutons), possibly marking (1) the temporal reinitiation of sub-
duction along the Pacific margin or alternatively (2) asthenospheric upwelling and
thermally induced partial melting of lower crustal materials due to delamination of
recently subducted oceanic lithosphere. In either case, granitoid magmatism within
the continental domain was extinguished during final approach and accretion of the
Gorgona Terrane along the Pacific margin in the early Eocene.
Contemporaneous with development of the Paleocene-Eocene post-collisional
arc of the Central Cordillera, granitoid magmatism was also developed in the ca.
57–50 Ma Santa Marta Batholith and associated plutons located along the apex of
the Sierra Nevada de Santa Marta. This localized and short-lived arc segment was
generated kinematically independently from plate interactions along the Colombian
Pacific margin and is interpreted to be related to the low-angle subduction or forced
underthrusting of oceanic crust and tectonic stacking along the NW margin of the
Sierra Nevada de Santa Marta, due to the NW-directed migration of the continental
Maracaibo tectonic float.
Located to the west of the continental granitoids of ca. 96–50 Ma age and hosted
within accreted oceanic terranes of the Western Tectonic Realm, two spatially
­temporally separate groups of subduction-related granitoids are also encompassed
within early Northern Andean Orogeny development. The first, dating from ca. 100
to 84  Ma was generated during westward subduction of Proto-Caribbean crust
beneath the northward and eastward migrating Farallon-CCOP/CLIP assemblage.
These primitive, allochthonous, intra-oceanic arc granitoids (Buriticá, Santa Fé,
Sabanalarga, Mistrató, Buga, Jejénes), correlative to the Greater (or leading-edge)
Arc of the Caribbean series, were detached from their Farallon/CCOP roots in the
late Cretaceous-Paleocene during dextral-oblique collision of the Farallon-CCOP/
CLIP assembly and accretion of the Western Tectonic Realm terranes. We interpret
the late Cretaceous-Paleocene tectonic lock-up between the Farallon-CCOP/CLIP
assemblage and the northwestern South American margin to have led to the shut-
down of subduction and granitoid arc magmatism in both continental Colombia and
along the Colombian segment of the Greater (leading-edge) Arc.
We suggest that the late Cretaceous-Paleocene-Eocene tectonic pile-up of buoy-
ant, leading-edge CCOP-CLIP fragments along the NW South American margin
was also a persuading factor in the detachment and initiation of east-directed sub-
duction of the Farallon Plate beneath the trailing edge of the CCOP/CLIP assem-
blage, beginning in the Paleocene. In Colombia, this magmatism is represented by
the ca. 50–42 Ma, intra-oceanic Mandé and Acandí batholiths including associated
hypabyssal porphyritic stocks, all of which comprise the younger representatives of
the allochthonous, Western Group granitoids associated with early Northern Andean
orogenic development. Mandé-Acandí correlates with the Middle American Arc
series of Central America. The arc emplaced into the trailing edge of the CCOP,
which, in Colombia, is represented by the El Paso Terrane including the Baudó
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 397

Complex. This composite arc and oceanic basement assemblage, however, were not
accreted to the Colombia margin until the Miocene.
Many investigations of the origins and spatial vs. temporal migration of the
Farallon-CCOP/CLIP assemblage demonstrate Pacific provenance and N and E
migration into the inter-American gap, during the mid-late Cretaceous and early
Paleogene. These same investigations suggest the Caribbean Plate docked with (i.e.
was fixed with respect to) the South American Plate in the early Eocene (by ca.
54.5 Ma). We contend that the magmatic record of interactions between the South
American, Proto-Caribbean and Farallon plates and the Caribbean-Colombian
Oceanic Plateau is duly indicated by the pre- and syn- and post-collisional granitoid
arc segments within the continental domain (Eastern Group granitoids) and the
leading- and trailing-edge, intra-oceanic (Western Group) granitoids, presently
accreted along the Colombian Pacific margin. In this context, we suggest the re-­
evaluation of tectonic models which require the amagmatic, low-angle or flat-slab
subduction/consumption of large volumes of oceanic lithosphere (Pacific, Farallon,
CCOP) beneath continental Colombia, during time intervals in which the existence
of an accompanying magmatic arc within the continental cannot be demonstrated
(e.g. early-mid-Cretaceous, ca. 145–96 Ma).
Following early Northern Andean Orogeny terrane assembly and docking of the
Caribbean Plate, an additional hiatus in subduction-related granitoid magmatism in
continental Colombia, spanning the period from ca. 50 to 23 Ma, is recorded. This
hiatus is marked by the absence of outcropping granitoids, including the lack of
detrital zircon populations dating from this time interval. Autochthonous granitoid
arc-related magmatism resumed along the Colombian Pacific margin at ca. 23 Ma.
The following events characterize the tectonic and magmatic development of the
region leading up to and during the late Northern Andean Orogeny:
(1) The N and W migration of the South American Plate, relative to the stationary
Caribbean Plate, beginning as early as the Eocene. Plate interaction along the
Colombo-Caribbean margin was limited to tectonic tightening, stacking, buckling
and uplift of the San Jacinto and Sinú terranes and the forced-underthrusting of
Caribbean lithosphere. The absence of granitoid arc magmatism throughout conti-
nental Colombia, coincident with Cenozoic Colombo-Caribbean Plate interaction,
is again stressed. N and W migration of the South American Plate continued into the
mid-Miocene resulting in the culmination of the late Northern Andean Orogeny,
including closure of the Middle American Seaway, collision/accretion of the El
Paso Terrane and uplift of the Baudo Complex along the northwesternmost
Colombian margin between ca. 8 and 4 Ma.
(2) Restructuring/rifting of the Farallon Plate within the eastern Pacific domain,
resulting in development of the Nazca-Cocos plate system. Continued rifting within
the Nazca segment between ca. 20 and 9  Ma gave rise to the Cauca and Coiba
microplates, separated by the ca. E-W striking Sandra Ridge. Granitoids associated
with Nazca Plate subduction along the Colombian Pacific margin first appear within
the ca. 23–21 Ma Piedrancha-Cuembí arc segment, located in south–westernmost
Colombia, well to the south of the trailing edge of the CCOP. The progressively
398 H. Leal-Mejía et al.

shallowing angle of subduction of the southern (Cauca) segment of the Nazca Plate
led to eastward migration of the granitoid arc axis into the Cauca-Patía region
between ca. 18 and 9 Ma. Continued eastward and northward migration of the mag-
matic arc axis during the Mio-Plio-Pleistocene led to conformation of the modern-­
day Northern Andes volcanic arc in southern and central Colombia.
To the north, tectonic tightening associated with South American-CCOP plate
interaction hindered initiation of subduction related to the Coiba segment of the
Nazca Plate, with the first manifestation of subduction-related granitoids appearing
in the Farallones-Páramo Frontino-El Cerro arc segment at ca. 12–10  Ma. Again,
progressive shallowing of the subduction angle, probably due to trench clogging by
buoyant aseismic material (e.g. Sandra Ridge), led to eastward migration of arc axial
magmatism into the Middle Cauca valley and the Central Cordillera (Cajamarca-
Salento porphyry cluster) between ca. 9 and 4 Ma, followed by emplacement of the
Plio-Pleistocene Río Dulce cluster to the north and coaxial conformation of the
northernmost segment of the active Colombian volcanic arc (Ruíz-Santa Isabel-
Tolima volcanic complex).
Based upon the foregoing, all of the latest Oligocene through Plio-Pleistocene
granitoid arc/volcanic segments in the Colombian Andes are demonstrably associated
with the segmented subduction of the Nazca Plate beneath the Pacific margin, begin-
ning in the latest Oligocene, and all of the documented Oligo-Miocene arc segments
are considered autochthonous with respect to continental Colombia. In addition to
these subduction-related granitoids, minor, isolated occurrences of ­Mio-­Pliocene hyp-
abyssal and volcanic rocks (Vetas-California, Paipa-Iza, Quetame) are observed
within the Santander Massif and Eastern Cordillera, to the east of the active Colombian
volcanic arc. On the basis of major element lithochemistry, these back-arc occur-
rences comprise a bimodal suite. They form a coaxial trend with respect to the overall
NNE orientation of the Miocene through modern-day granitoid arc axis. We interpret
the Vetas-California-Paipa-Iza-Quetame occurrences to represent incipient rift-related
magmatism whose emplacement was facilitated by back-arc extension focussed along
the deep crustal conduits provided by the Bucaramanga–Santa Marta–Garzón fault
and suture system.
The age, nature and spatial vs. temporal distribution of granitoids, as presently
exposed within the Colombian geologic mosaic, provide valuable clues to the deci-
phering of the Phanerozoic tectono-magmatic history of the Colombian Andes.
Although important advances have been made in the last decade, especially with
respect to the generation of high-resolution age date, lithogeochemical and isotopic
data, much work remains to be done, in continued sampling within the context of
high-quality field-based mapping. Data verification, integration and synthesis into
the ample and evolving geological, geophysical and tectono-sedimentalogical
database which exists for the region will be an essential component of this process.
We consider the analysis presented within this chapter as preliminary and, beyond
the advance in understanding we feel it represents, would hope it will inspire con-
tinued investigation of the less studied, polemic and unresolved details regarding
Phanerozoic tectono-magmatic evolution in the Colombian Andes.
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 399

References

Agencia Nacional de Hidrocarburos (ANH), Universidad de Caldas (2011) Estudio integrado

de los núcleos y registros obtenidos de los pozos someros (slim holes) perforados por la
ANH. Agencia Nacional de Hidrocarburos, Unpublished Report, Manizales, 304 p
Alfonso R (2000) Catálogo de unidades ígneas de Colombia: Batolito de Mocoa. INGEOMINAS,
Bogotá
Alvarez JA (1983) Geología de la Cordillera Central y el occidente colombiano y petroquímica de
los intrusivos granitoides meso-cenozoicos. Boletín Geológico INGEOMINAS 26(2):1–175
Altenberger U, Concha AE (2005) Late Lower to early Middle Jurassic arc magmatism in the
northern Ibagué-Batholith/Colombia. Geología Colombiana 30:87–97
Arango JL, Ponce A (1982) Mapa Geológico del Departamento de Nariño. Memoria Explicativa.
INGEOMINAS, 40 p
Aspden JA, McCourt WJ, Brook M (1987) Geometrical control of subduction-related magmatism:
the Mesozoic and Cenozoic plutonic history of western Colombia. J Geol Soc 144(6):893–905
Aspden JA, Fortey N, Litherland M, Viteri F, Harrison SM (1992) Regional S-type granites in the
Ecuadorian Andes: possible remnants of the breakup of western Gondwana. J S Am Earth Sci
6(3):123–132
Ballesteros CI (1983) Mapa Geológico generalizado del Departamento de Bolívar. INGEOMINAS,
Bogotá
Barbarin B (1999) A review of the relationships between granitoid types, their origins and their
geodynamic environments. Lithos 46(3):605–626
Barrero D (1977) Geology of the Central Western Cordil- lera, west of Buga and Roldanillo,
Colombia Ph.D. Thesis, Colorado School of Mines
Bartok PE, Renz O, Westermann GEG (1985) The Siquisique Ophiolites, Northern Lara State,
Venezuela: a discussion on their Middle Jurassic ammonites and tectonic implications. Geol
Soc Am Bull 96:1050–1055
Barton MD, Young S (2002) Non-pegmatitic deposits of beryllium: mineralogy, geology, phase
equilibria and origin. Rev Mineral Geochem 50(1):591–691
Bayona G, Cardona A, Jaramillo C, Mora A, Montes C, Valencia V, Ayala C, Montenegro O,
Ibañez-Mejía M (2012) Early Paleogene magmatism in the northern Andes: insights on the
effects of Oceanic Plateau–continent convergence. Earth Planet Sci Lett 331-332:97–111
Bird P (2003) An updated digital model of plate boundaries. Geochem Geophys Geosyst
4(3):1027–1059
Bissig T, Mantilla LC, Hart CJR (2014) Petrochemistry of igneous rocks of the California-Vetas
mining district, Santander, Colombia: implications for northern Andean tectonics and porphyry
Cu (–Mo, Au) metallogeny. Lithos 200-201:355–367
Bogotá J, Aluja J (1981) Geología de la Serranía de San Lucas. Geología Norandina 4:49–55
Boinet T, Bourgois J, Bellon H, Toussaint JF (1985) Age et répartition du magmatisme
Prémèsozoïque des Andes de Colombie. Comptes-rendus des séances de l'Académie des
sciences Paris 300(10):445–450
Botero G (1975) Edades radiométricas de algunos plutones colombianos. Minería 27:8336–8342
Buchs DM, Arculus RJ, Baumgartner PO, Baumgartner-Mora C, Ulianov A (2010) Late Cretaceous
arc development on the SW margin of the Caribbean plate: insights from the Golfito, Costa
Rica, and Azuero, Panama, complexes. Geochem Geophys Geosyst 11(7):1–35
Bürgl H (1967) The orogenesis of the Andean system of Colombia. Tectonophysics 4:429–443
Burke K, Cooper C, Dewey JF, Mann P, Pindell JL (1984) Caribbean tectonics and relative plate
motions. In: Bonini WE, Hargraves RB, Shagam R (eds) The Caribbean–South American Plate
boundary and regional tectonics. Geological Society of America Memoir 162:31–63
Bustamante A (2008) Geotermobarometria, geoquímica, geocronologia e evolução tectônica
das rochas da fácies xisto azul nas áreas de Jambaló (Cauca) e Barragán (Valle del Cauca),
Colômbia. Ph.D. thesis, Universidade de São Paulo
400 H. Leal-Mejía et al.

Bustamante C, Cardona A, Bayona G, Mora A, Valencia V, Gehrels G, Vervoort J (2010) U-Pb

LA-ICP-MS geochronology and regional correlation of middle Jurassic intrusive rocks from
the Garzon Massif, Upper Magdalena Valley and Central Cordillera, Southern Colombia.
Boletín de Geología 32(2):93–109
Bustamante C, Cardona A, Archanjo CJ, Bayona G, Lara M, Valencia V (2017) Geochemistry
and isotopic signatures of Paleogene plutonic and detrital rocks of the Northern Andes of
Colombia: a record of post-collisional arc magmatism. Lithos 277:199–209
Cáceres C, Cediel F, Etayo F (2003) Guía introductoria de la distribución de facies sedimentarias
de Colombia, Mapas de distribución de facies sedimentarias y armazón tectónico de Colombia
a través del Proterozoico y del Fanerozoico. INGEOMINAS, Bogotá
Calle B, Toussaint JF, Restrepo JJ, Linares E (1980) Edades K/Ar de dos plutones de la parte sep-
tentrional de la Cordillera Occidental de Colombia. Geología Norandina 2:17–20
Campbell CJ (1968) The Santa Marta wrench fault of Colombia and its regional setting. Paper pre-
sented at the 4th Caribbean Geological Conference, Port of Spain, 28 March – 12 Abril 1965
Campbell CJ (1974) Colombian Andes. In: Spencer AM (ed) Mesozoic-Cenozoic Orogenic Belts.
Geological Society of London Special Publications (4):705–724
Cardona A, Chew D, Valencia VA, Bayona G, Miškovic A, Ibañez-Mejía M (2010a) Grenvillian
remnants in the Northern Andes: Rodinian and Phanerozoic paleogeographic perspectives. J S
Am Earth Sci 29(1):92–104
Cardona A, Valencia V, Garzón A, Montes C, Ojeda G, Ruiz J, Weber M (2010b) Permian to
Triassic I to S-type magmatic switch in the northeast Sierra Nevada de Santa Marta and adja-
cent regions, Colombian Caribbean: tectonic setting and implications within Pangea paleoge-
ography. J S Am Earth Sci 29(4):772–783
Cardona A, Valencia VA, Bayona G, Duque J, Ducea M, Gehrels G, Jaramillo C, Montes C, Ojeda
G, Ruíz J (2011) Early-subduction-related orogeny in the northern Andes: Turonian to Eocene
magmatic and provenance record in the Santa Marta Massif and Rancheria Basin, northern
Colombia. Terra Nova 23:26–34
Cediel F (2011) Major Tecto-sedimentary events and basin development in the Phanerozoic of
Colombia: In: Cediel F (ed) Petroleum Geology of Colombia. Regional Geology of Colombia,
vol 1. Agencia Nacional de Hidrocarburos (ANH) – EAFIT, pp 13–108
Cediel F (2018) Phanerozoic orogens of Northwestern South America: cordilleran-type orogens,
taphrogenic tectonics and orogenic float. In: Cediel F, Shaw RP (eds) Geology and tecton-
ics of Northwestern South America: the Pacific-Caribbean-Andean junction. Springer, Cham,
pp 3–89
Cediel F, Cáceres C (2000) Geological map of Colombia. Geotec, Ltd., Bogotá
Cediel F, Etayo F, Cáceres C (1994) Facies distribution and tectonic setting through the Phanerozoic
of Colombia. INGEOMINAS (ed) Geotec Ltd., Bogota
Cediel F, Shaw RP, Cáceres C (2003) Tectonic assembly of the Northern Andean Block. In:
Bartolin, C, Buffler RT, Blickwede J (eds) The circum-Gulf of Mexico and the Caribbean:
hydrocarbon habitats, basin formation, and plate tectonics. AAPG Memoir 79:815–848
Cediel F, Barrero D, Cáceres C (1998) Seismic Atlas of Colombia: Seismic expression of struc-
tural styles in the basins of Colombia. Robertson Research International, UK (ed) Geotec Ltd.,
Bogotá, vol 1 to 6
Cediel F, Restrepo I, Marín-Cerón MI, Duque-Caro H, Cuartas C, Mora C, Montenegro G, García
E, Tovar D, Muñoz G (2010) Geology and hydrocarbon potential, Atrato and San Juan basins,
Chocó (Panamá) arc, Colombia, Tumaco Basin (Pacific realm). Colombia, Agencia Nacional
de Hidrocarburos (ANH)-EAFIT, Medellín
Cepeda H, Pardo-Villaveces N (2004) Vulcanismo de Paipa. Technical Report. INGEOMINAS,
Bogotá
Chiarabba C, De Gori P, Faccenna C, Speranza F, Seccia D, Dionicio V, Prieto GA (2015)
Subduction system and flat slab beneath the Eastern Cordillera of Colombia. Geochem
Geophys Geosyst 17(1):16–27
Chiaradia M, Fontboté L, Paladines A (2004) Metal sources in mineral deposits and crustal rocks
of Ecuador (1 N–4 S): a lead isotope synthesis. Econ Geol 99(6):1085–1106
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 401

Clavijo J, Mantilla L, Pinto J, Berna L, Perez A (2008) Evolución geológica de la Serranía de San
Lucas, norte del valle medio del Magdalena y noroeste de la Cordillera Oriental. Boletín de
Geología 3045–62
Cochrane R (2013) U-Pb thermochronology, geochronology and geochemistry of NW South
America: rift to drift transition, active margin dynamics and implications for the volume bal-
ance of continents. PhD thesis, Université de Genève
Cochrane R, Spikings R, Gerdes A, Ulianov A, Mora A, Villagómez D, Putlitz B, Chiaradia M
(2014a) Permo-Triassic anatexis, continental rifting and the disassembly of western Pangaea.
Lithos 190–191:383–402
Cochrane R, Spikings R, Gerdes A, Winkler W, Ulianov A, Mora A, Chiaradia M (2014b)
Distinguishing between in-situ and accretionary growth of continents along active margins.
Lithos 202–203:382–394
Cocks LRM, Torsvik TH (2006) European geography in a global context from the Vendian to the
end of the Palaeozoic. In: Gee DG, Stephenson RA (eds) European lithosphere dynamics. Geol
Soc Lond Mem 32:83–95
Cordani UG, Cardona A, Jiménez DM, Liu D, Nutman AP (2005) Geochronology of Proterozoic
basement inliers in the Colombian Andes: tectonic history of remnants of a fragmented
Grenville belt. Geological Society of London Special Publications 246(1):329–346
Correa AM, Pimentel M, Restrepo JJ, Nilson A, Ordoñez O, Martens U, Laux JE, Junges S (2006)
U-Pb zircon ages and Nd-Sr isotopes of the Altavista stock and the San Diego gabbro: new
insights on Cretaceous arc magmatism in the Colombian Andes. Abstract presented at the V
South American Symposium on Isotope Geology (SSAGI), Punta del Este, 24–27 April 2006
Correa AM (2007) Petrogênese e evolução do ofiolito de Aburrá, cordilheira central dos Andes
colombianos. PhD Thesis, Universidade de Brasilia
Cruz N, Carrillo JA, Mantilla LC (2014) Consideraciones petrogenéticas y geocronología de las
rocas ígneas porfiríticas aflorantes en la Quebrada Ventanas (Municipio Arboledas, Norte de
Santander, Colombia): implicaciones metalogenéticas. Boletín de Geología 36(1):103–118
Cuadros FA, Botelho NF, Ordóñez-Carmona O, Matteini M (2014) Mesoproterozoic crust in
the San Lucas range (Colombia): an insight into the crustal evolution of the northern Andes.
Precambrian Res 245:186–206
Davies JF, Whitehead RE (2006) Alkali-alumina and MgO-alumina molar ratios of altered and
unaltered rhyolites. Explor Min Geol 15(1–2):75–88
Davies JF, Whitehead RE (2010) Alkali/alumina molar ratio trends in altered Granitoid rocks host-
ing porphyry and related deposits. Explor Min Geol 19(1–2):13–22
De La Roche H, Leterrier JT, Grandclaude P, Marchal M (1980) A classification of volcanic and
plutonic rocks using R1R2-diagram and major-element analyses—its relationships with cur-
rent nomenclature. Chem Geol 29(1–4):183–210
Deckart K, Godoy E, Bertens A, Saeed A (2010) Barren Miocene granitoids in the Central Andean
metallogenic belt, Chile: geochemistry and Nd-Hf and U-Pb isotope systematics. Andean Geol
37(1):1–31
Dewey JW (1972) Seismicity and tectonics of western Venezuela. Bull Seismol Soc Am
62(6):1711–1751
Dörr W, Grösser JR, Rodríguez GI, Kramm U (1995) Zircon U-Pb age of the Páramo Rico tonalite-­
granodiorite, Santander Massif (Cordillera Oriental, Colombia) and its geotectonic signifi-
cance. J S Am Earth Sci 8(2):187–194
Doubrovine PV, Steinberger B, Torsvik TH (2012) Absolute plate motions in a reference frame
defined by moving hot spots in the Pacific, Atlantic, and Indian oceans. J  Geophys Res
117(B09101):1–30
Duque JF (2009) Geocronología (U/Pb y 40Ar/39Ar) y geoquímica de los intrusivos paleógenos de
la Sierra Nevada de Santa Marta y sus relaciones con la tectónica del Caribe y el arco mag-
mático circun-Caribeño. MSc thesis, Universidad Nacional Autónoma de México
Duque-Caro H (1990) The Chocó block in the northwestern corner of South America: structural,
tectonostratigraphic, and paleogeographic implications. J S Am Earth Sci 3(1):71–84
Duque-Caro H (1991) Contributions to the geology of the Pacific and the Caribbean coastal areas
of Northwestern Colombia and South America. PhD Thesis, Princeton University
402 H. Leal-Mejía et al.

Duque-Trujillo J, Sánchez J, Orozco-Esquivel T, Cárdenas A (2018) Cenozoic magmatism of the

maracaibo block and its tectonic significance. In: Cediel F, Shaw RP (eds) Geology and tecton-
ics of Northwestern South America: the Pacific-Caribbean-Andean junction. Springer, Cham,
pp 551–594
Ego F, Sébrier M, Yepes H (1995) Is the Cauca-Patía and Romeral fault system left or right-lateral?
Geophys Res Lett 22(1):33–36
Escobar LA, Tejada N (1992) Prospección de Platino en Piroxenitas y de Oro en Skarn en la Mina
Don Diego, El Cerro, Frontino (Antioquia). B.Sc. Thesis, Universidad Nacional de Colombia,
Facultad de Ciencias, Seccional de Medellín
Estrada JJ (1995) Paleomagnetism and accretion events in the Northern Andes: PhD Thesis, State
University of New York
Etayo-Serna F, Barrero D, Lozano H, Espinosa A, González H, Orrego A, Zambrano F, Duque
H, Vargas R, Núñez A, Álvarez J, Ropaín C, Ballesteros I, Cardozo E, Forero H, Galvis N,
Ramírez C, Sarmiento L, Albers JP, Case JE, Singer DA, Bowen RW, Berger BR, Cox DP,
Hodges CA (1983) Mapa de terrenos geológicos de Colombia. Publicaciones Geológicas
Especiales, vol 14. INGEOMINAS, Bogotá
Etayo-Serna F, Rodríguez GI (1985) Edad de la Formación Los Santos. In: Etayo-Serna F, Laverde-­
Montaño F, Pava A (eds) Proyecto Cretácico: Contribuciones. Publicaciones Geológicas
Especiales, vol 16. INGEOMINAS, Bogotá
Fabre A, Delaloye M (1983) Intrusiones básicas cretácicas en las sedimentitas de la parte central
de la Cordillera Oriental. Geología Norandina 6:19–28
Farris DW, Jaramillo C, Bayona G, Restrepo-Moreno SA, Montes C, Cardona A, Mora A,
Speakman RJ, Glascock MD, Valencia V (2011) Fracturing of the Panamanian Isthmus during
initial collision with South America. Geology 39(11):1007–1010
Feininger T (1970) The Palestina fault, Colombia. Bull Geol Soc Am 81:1201–1216
Feininger T, Botero G (1982) The Antioquian Batholith, Colombia. Publicación Geológica
Especial INGEOMINAS 12:1–50
Feininger T, Barrero D, Castro N (1972) Geología de parte de los departamentos de Antioquia y
Caldas (sub-zona II-B). Boletín Geológico INGEOMINAS 20(2):1–173
Frantz JC, Ordoñez OC, Franco E, Groves DI, McNaughton NJ (2003) Marmato porphyry
intrusion, ages and mineralization. In: Abstracts of the IX Colombian Geological Congress,
Medellín, 30 July – 1 August 2003
Frost BR, Frost CD (2008) A geochemical classification for feldspathic igneous rocks. J Petrol
49(11):1955–1969
Frost BR, Barnes CG, Collins WJ, Arculus RJ, Ellis DJ, Frost CD (2001) A geochemical classifica-
tion for granitic rocks. J Petrol 42(11):2033–2048
Gansser A (1955) Ein Beitrag zur Geologie und Petrographie der Sierra Nevada de Santa Marta
(Kolumbien, Suedamarika). Schweiz. Mineral Petrogr Mitt 35(2):209–279
Gansser A (1973) Facts and theories on the Andes. J Geol Soc Lond 129:93–131
Garzón T (2003) Geoquímica y potencial minero asociado a cuerpos volcánicos de la región de
Paipa, Departamento de Boyacá, Colombia. M.Sc. thesis, Universidad Nacional de Colombia
Geyer OF (1973) Das Praekretazische Mesozoikum von Kolumbien. Geologisches Jarbuch 5:1–56
Gil-Rodríguez J  (2010) Igneous Petrology of the Colosa Gold-rich Porphyry System (Tolima,
Colombia). M.Sc. thesis, The University of Arizona
Gil-Rodríguez J  (2014) Petrology of the Betulia Igneous Complex, Cauca, Colombia. J  S Am
Earth Sci 56:339–356
Goldsmith R, Marvin RF, Mehnert HH (1971) Radiometric ages in the Santander Massif, Eastern
Cordillera, Colombian Andes. US Geol Surv Prof Pap 750D:D44–D49
Gómez J, Nivia A, Montes NE, Jiménez DM, Tejada ML, Sepúlveda MJ, Osorio JA, Gaona T,
Diederix H, Uribe H, Mora M (2007) Mapa Geológico de Colombia. INGEOMINAS, Bogotá
Gómez J, Montes NE, Nivia A, Diederix H (2015a) Mapa Geológico de Colombia 2015. Servicio
Geológico Colombiano, Bogotá
Gómez J, Montes NE, Alcárcel FA, Ceballos JA (2015b) Catálogo de dataciones radiométricas de
Colombia en ArcGIS y Google Earth. In: Gómez J, Almanza MF (eds) Compilando la geología
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 403

de Colombia - Una visión a 2015. Servicio Geológico Colombiano, Publicaciones Geológicas

Especiales 33:63–419
González H (1990) Mapa geológico generalizado del Departamento de Risaralda. INGEOMINAS,
Escala 1:200,000, Versión digital 2010
González H (1993) Mapa geológico generalizado del Departamento de Caldas - geología y recur-
sos minerales. Memoria explicativa, INGEOMINAS, 62p
González H (1999) Geología del Departamento de Antioquia. Escale 1:400.000. INGEOMINAS,
Bogotá
González H (2001) Mapa geológico del Departamento de Antioquia: Geología, recursos minerales
y amenazas potenciales. Memoria Explicativa. INGEOMINAS, Bogotá
Goossens PJ, Rose WI, Flores D (1977) Geochemistry of tholeiites of the Basic Igneous Complex
of northwestern South America. Geol Soc Am Bull 88(12):1711–1720
Hamilton WB (1994) Subduction systems and magmatism. In: Smellie JL (ed) Volcanism asso-
ciated with extension at consuming plate margins: Geological Society of London Special
Publication 81:3–28
Hastie AR, Kerr AC (2010) Mantle plume or slab window?: physical and geochemical constraints
on the origin of the Caribbean oceanic plateau. Earth Sci Rev 98(3):283–293
Henrichs I (2013) Caracterização e idade das intrusivas do sistema pórfiro Yarumalito, Magmatismo
Combia, Colombia. M.Sc. Thesis, Universidade Federal do Rio Grande do Sul
Horton BK, Taylor JE, Nie J, Mora A, Parra M, Reyes-Harker A, Stockli DF (2010) Linking sedi-
mentation in the northern Andes to basement configuration, Mesozoic extension, and Cenozoic
shortening: evidence from detrital zircon U-Pb ages, Eastern Cordillera, Colombia. Geol Soc
Am Bull 122(9–10):1423–1442
Ibañez-Mejia M, Tassinari CCG, Jaramillo-Mejia J (2007) U-Pb ages of the “Antioquian
Batholith” – Geochronological constraints of late Cretaceous magmatism in the central Andes
of Colombia. Abstract presented at the XI Congreso Colombiano de Geología, Bucaramanga,
14–17
Ibañez-Mejía M, Jaramillo-Mejía JM, Valencia VA (2008) U–Th/Pb zircon geochronology by multi-
collector LA-ICP-MS of the Samaná Gneiss: a Middle Triassic syn-tectonic body in the Central
Andes of Colombia, related to the latter stages of Pangea assembly. In: VI South American
Symposium on Isotope Geology, San Carlos de Bariloche–Argentina, 13–17 April 2008
Ibañez-Mejía M, Ruiz J, Valencia VA, Cardona A, Gehrels GE, Mora AR (2011) The Putumayo
Orogen of Amazonia and its implications for Rodinia reconstructions: new U–Pb geochro-
nological insights into the Proterozoic tectonic evolution of northwestern South America.
Precambrian Res 191(1):58–77
Irvine TN, Baragar WRA (1971) A guide to the chemical classification of the common volcanic
rocks. Can J Earth Sci 8(5):523–548
Irving EM (1975) Structural evolution of the northernmost Andes, Colombia. U.S.  Geological
Survey, Professional Paper 846
Kellogg JN, Ogujiofor IJ, Kansakar DR (1985) Cenozoic tectonics of the Panama and North Andes
blocks. In: Memoirs of the 6th Latin American Congress on Geology, vol 1. Bogotá, p. 40
Kennan L, Pindell JL (2009) Dextral shear, terrane accretion and basin formation in the Northern
Andes: best explained by interaction with a Pacific-derived Caribbean plate? Geological
Society of London Special Publications 328(1):487–531
Keppie JD (2004) Terranes of Mexico revisited: a 1.3 billion year odyssey. Int Geol Rev
46(9):765–794
Keppie JD (2008) Terranes of Mexico revisited: a 1.3 billion year Odyssey. In: Keppie JD, Murphy
JB, Ortega-Gutierrez F, Ernst WG (eds) Middle American Terranes, potential correlatives, and
Orogenic processes. CRC Press/Taylor & Francis Group, Boca Raton, 7–36
Keppie JD, Ortega-Gutierrez F (2010) 1.3–0.9 Ga Oaxaquia (Mexico): remnant of an arc/backarc
on the northern margin of Amazonia. J S Am Earth Sci 29(1):21–27
Keppie JD, Ramos VA (1999) Odyssey ofterranes in the lapetus and Rheic oceans during the
Paleozoic. In: Ramos VA, Keppie JD (eds) Laurentia-Gondwana connections before Pangea.
The Geological Society of America, Special Paper 336:267–276
404 H. Leal-Mejía et al.

Kerr AC, Tarney J (2005) Tectonic evolution of the Caribbean and northwestern South America:
the case for accretion of two Late Cretaceous oceanic plateaus. Geology 33(4):269–272
Kerr AC, Tarney J, Marriner GF, Nivia A, Saunders AD (1997) The Caribbean–Colombian
Cretaceous Igneous Province: The internal anatomy of an oceanic plateau. In: Mahoney JJ,
Coffin MF (eds) Large Igneous Provinces: Continental, oceanic, and planetary flood volca-
nism. American Geophysical Union, Geophysical monograph 100:123–144
Kerr AC, White RV, Thompson PME, Tarney J, Saunders AD (2003) No oceanic plateau—No
Caribbean plate? The seminal role of an oceanic plateau in Caribbean plate evolution. In:
Bartolin, C, Buffler RT, Blickwede J (eds) The circum-Gulf of Mexico and the Carib-bean:
hydrocarbon habitats, basin formation, and plate tectonics. AAPG Memoir 79:126–168
Kroonenberg SB (1982) A Grenvillian granulite belt in the Colombian Andes and its relation to the
Guiana shield. Geol Mijnb 61(4):325–333
Kurhila M, Andersen T, Rämö OT (2010) Diverse sources of crustal granitic magma: Lu–Hf
isotope data on zircon in three Paleoproterozoic leucogranites of southern Finland. Lithos
115(1):263–271
Leal-Mejía H (2011) Phanerozoic Gold Metallogeny in the Colombian Andes: A tectono-­magmatic
approach. Ph.D. thesis, Universitat de Barcelona
Leal-Mejía H, Shaw RP, Melgarejo JC (2011) Phanerozoic granitoid magmatism in Colombia and
the tectono-magmatic evolution of the Colombian Andes. In: Cediel F (ed) Petroleum Geology
of Colombia. Regional Geology of Colombia, vol 1. Agencia Nacional de Hidrocarburos
(ANH) – EAFIT, p 109–188
Lesage G, Richards JP, Muehlenbachs K, Spell TL (2013) Geochronology, geochemistry, and fluid
characterization of the Late Miocene Buriticá Gold Deposit, Antioquia Department, Colombia.
Econ Geol 108:1067–1097
Litherland M, Aspden JA, Jemielita RA (1994) The metamorphic belts of Ecuador. Overseas Mem
Br Geol Surv 11:1–147
Lodder C, Padilla R, Shaw RP, Garzón T, Palacio E, Jahoda R (2010) Discovery history of the
La Colosa Gold Porphyry deposit, Cajamarca, Colombia. Soc Econ Geol Spec Pub 15:19–28
Loiselle MC, Wones DR (1979) Characteristics and origin of anorogenic granites. Geol Soc Am
Abstr Programs 11:468
Lonsdale P (2005) Creation of the Cocos and Nazca plates by fission of the Farallon plate.
Tectonophysics 404(3):237–264
Mantilla LC, Valencia VA, Barra F, Pinto J, Colegial J (2009) U-Pb geochronology of porphyry
rocks in the Vetas – California gold mining area (Santander, Colombia). Boletín de Geología
(UIS) 31(1):31–43
Mantilla LC, Bissig T, Cottle JM, Hart CRJ (2012) Remains of early Ordovician mantle-derived
magmatism in the Santander Massif (Colombian Eastern Cordillera). J S Am Earth Sci 38:1–12
Mantilla LC, Bissig T, Valencia V, Hart CJR (2013) The magmatic history of the Vetas-California
mining district, Santander Massif, Eastern Cordillera, Colombia. J S Am Earth Sci 45:235–249
Maresch WV, Stöckhert B, Baumann A, Kaiser C, Kluge R, Krückhans-Lueder G, Brix MR,
Thomson M (2000) Crustal history and plate tectonic development in the southern Caribbean.
Sonderheft Zeitschrift fuer Angewandte. Andean Geol 1:283–289
Maresch WV, Kluge R, Baumann A, Pindell JL, Krückhans-Lueder G, Stanek K (2009) The occur-
rence and timing of high-pressure metamorphism on Margarita Island, Venezuela: a constraint
on Caribbean-South America interaction. Geol Soc Lond Spec Publ 328(1):705–741
Marechal P (1983) Les Temoins de Chaine Hercynienne dans le Noyau Ancien des Andes de
Merida (Venezuela): Structure et evolution tectometamorphique. Ph.D. Thesis, Universite de
Bretagne Occidentale
Marín-Cerón MI, Leal-Mejía H, Bernet M, Mesa-García J (2018) Late cenozoic to modern-day
volcanism in the Northern Andes; A geochronological, petrographical and geochemical review.
In: Cediel F, Shaw RP (eds) Geology and tectonics of Northwestern South America: the Pacific-
Caribbean-Andean junction. Springer, Cham, pp 603–641
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 405

Martens U, Restrepo JJ, Ordóñez-Carmona O, Correa-Martínez AM (2014) The Tahamí and

Anacona terranes of the Colombian Andes: missing links between the South American and
Mexican Gondwana margins. J Geol 122(5):507–530
Maya M (1992) Catalogo de dataciones isotópicas en Colombia. Boletín Geológico INGEOMINAS
32(1–3):127–188
McCourt WJ, Feininger T (1984) High pressure metamorphic rocks of the Central Cordillera of
Colombia. British Geological Survey Reprint Series 84:28–35
McDonough WF, Sun SS (1995) The composition of the earth. Chem Geol 120(3–4):223–253
McGeary S, Ben-Avraham Z (1989) The accretion of Gorgona Island, Colombia: Multichannel
seismic evidence. In: Howel DG (ed) Tectonostratigraphic terranes of the Circum-Pacific
Region. Circum-Pacific Council for Energy and Mineral Resources, Earth Sciences Series
1:543–554
Mejía P, Santa M, Ordoñez O, Pimentel M (2008) Consideraciones petrográficas, geoquímicas y
geocronologicas de la parte occidental del Batolito de Santa Marta. Revista Dyna 15:223–236
Molnar P, Atwater T (1978) Interarc spreading and Cordilleran tectonics as alternates related to the
age of subducted oceanic lithosphere. Earth Planet Sci Lett 41(3):330–340
Montes C, Guzmán G, Bayona G, Cardona A, Valencia V, Jaramillo C (2010) Clockwise rotation of
the Santa Marta Massif and simultaneous Paleogene to Neogene deformation of the Plato-San
Jorge and Cesar-Ranchería basins. Journal of South American Earth Sciences 29(4):832–848
Montes C, Cardona A, McFadden R, Morón SE, Silva CA, Restrepo-Moreno S, Ramírez DA,
Hoyos N, Wilson J, Farris D, Bayona GA, Jaramillo CA, Valencia V, Bryan J, Flores JA (2012)
Evidence for middle Eocene and younger land emergence in central Panama: implications for
Isthmus closure. Geol Soc Am Bull 124(5–6):780–799
Montes C, Cardona A, Jaramillo C, Pardo A, Silva JC, Valencia V, Ayala C, Pérez-Angel LC,
Rodríguez-Parra LA, Ramirez V, Niño H (2015) Middle Miocene closure of the American
seaway. Science 348(6231):226–229
Müller RD, Royer JY, Cande SC, Roest WR, Maschenkov S (1999) New constraints on the Late
Cretaceous/tertiary plate tectonic evolution of the Caribbean. In: Mann P (ed) Caribbean
basins. Sedimentary basins of the world, vol 4. Elsevier, Amsterdam, pp 33–59
Nelson HW (1957) Contribution to the geology of the Central and Western Cordillera of Colombia
in the sector between Ibagué and Cali. Leidsche Geologische Mededelingen 22:1–75
Nerlich R, Clark SR, Bunge HP (2014) Reconstructing the link between the Galapagos hotspot and
the Caribbean plateau. GeoResJ 1-2:1–7
Nie J, Horton BK, Saylor JE, Mora A, Mange M, Garzione CN, Basu A, Moreno CJ, Caballero V,
Parra M (2012) Integrated provenance analysis of a convergent retroarc foreland system: U–
Pb ages, heavy minerals, Nd isotopes, and sandstone compositions of the Middle Magdalena
Valley basin, northern Andes, Colombia. Earth Sci Rev 110(1):111–126
Nívia A, Gómez J (2005) El Gabro Santa Fé de Antioquia y la Cuarzodiorita Sabanalarga, una
propuesta de nomenclatura litoestratigráfica para dos cuerpos plutónicos diferentes agrupa-
dos previamente como Batolito de Sabanalarga en el Departamento de Antioquia, Colombia.
Memoirs of the X Colombian Geological Congress, Bogotá
Nívia A (1992) Mapa geológico generalizado del Departamento del Valle del Cauca. Escala
1:300.000. INGEOMINAS - British Geological Survey (BGS), Bogotá
Nivia A, Marriner G, Kerr A (1996) El Complejo Quebrada Grande: una posible cuenca marginal
intracratónica del Cretáceo Inferior en la Cordillera Central de los Andes Colombianos. In:
Abstracts of the VII Colombian Geological Congress, vol III, 108–123
Nivia A, Marriner GF, Kerr AC, Tarney J (2006) The Quebradagrande complex: a lower cretaceous
ensialic marginal basin in the Central Cordillera of the Colombian Andes. J S Am Earth Sci
21:423–436
Núñez A (1998) Catalogo de unidades litoestratigraficas de Colombia  – Batolito de Ibagué:
INGEOMINAS, Bogotá
O’connor JT (1965) A classification for quartz-rich igneous rocks based on feldspar ratios. US
Geological Survey Professional Paper B 525:79–84
406 H. Leal-Mejía et al.

Ordoñez O (2001) Caracterização Isotópica Rb-Sr e Sm-Nd dos Principais Eventos Magmáticos
nos Andes Colombianos. Ph.D. thesis, Universidade de Brasilia
Ordoñez O, Pimentel MM (2002) Rb–Sr and Sm–Nd isotopic study of the Puquı́ complex,
Colombian Andes. J S Am Earth Sci 15(2):173–182
Ordoñez O, Pimentel MM, Armstrong RA, Gioia SMCL, Junges S (2001) U-Pb SHRIMP and
Rb-Sr ages of the Sonsón Batholith. In: III South American Symposium on Isotope Geology,
Pucon - Chile, 21–24 October 2001
Ordoñez-Carmona O, Restrepo JJ, Pimentel MM (2006) Geochronological and isotopical review
of pre-Devonian crustal basement of the Colombian Andes. J S Am Earth Sci 21(4):372–382
Ordoñez-Carmona O, Pimentel M, Laux JH (2007a) Edades U-Pb del Batolito Antioqueño. Boletín
de Ciencias de la Tierra 22:129–130
Ordoñez-Carmona O, Pimentel MM, Frantz JC, Chemale F (2007b) Edades U-Pb convencionales
de algunas intrusiones colombianas. Abstract presented at the XI Congreso Colombiano de
Geología, Bucaramanga, 14–17 Aug 2007
Orrego A, Cepeda H, Rodríguez G (1980) Esquistos glaucofánicos en el área de Jambaló, Cauca
(Colombia). Geología Norandina 1:5–10
Parada MA, Nyström JO, Levi B (1999) Multiple sources of the Coastal Batholith of central
Chile (31-34°S) – geochemical and Sr-Nd isotopic evidence and tectonic implications. Lithos
46:505–521
Pardo N, Cepeda H, Jaramillo JM (2005a) The Paipa volcano, Eastern Cordillera of Colombia,
South America – volcanic Stratigraphy. Earth Sci Res J 9(1):3–18
Pardo N, Cepeda H, Jaramillo JM (2005b) The Paipa volcano, Eastern Cordillera of Colombia,
South America (Part II)  – petrography and major elements petrology. Earth Sci Res
J 9(2):148–164
París G, Marín P (1979) Mapa Geológico Generalizado del Departamento del Cauca. Memoria
Explicativa, INGEOMINAS, 38 p
Pearce JA, Harris NB, Tindle AG (1984) Trace element discrimination diagrams for the tectonic
interpretation of granitic rocks. J Petrol 25(4):956–983
Peccerillo A, Taylor SR (1976) Geochemistry of Eocene calc-alkaline volcanic rocks from the
Kastamonu area, Northern Turkey. Contrib Mineral Petrol 58(1):63–81
Pennington WD (1981) Subduction of the eastern Panama Basin and seismotectonics of north-
western South America. J Geophys Res Solid Earth 86(B11):10753–10770
Pindell JL, Cande SC, Pitman WC, Rowley DB, Dewey JF, Labrecque J, Haxby W (1988) Plate
kinematic framework for models of Caribbean evolution. Tectonophysics 155:121–138
Pindell J, Kennan L (2001) Processes & Events in the Terrane assembly of Trinidad and E.
Venezuela. In: GCSSEPM Foundation 21st annual research conference transactions, Petroleum
Systems of Deep-Water Basins, 159–192
Priem HNA, Andriessen PAM, Boelrijk NAIM, de Boorder H, Hebeda EH, Verdurmen EA, Huguett
A, Verdurmen EAT, Verschure RH (1982) Geochronology of the Precambrian in the Amazonas
region of southeastern Colombia (western Guiana shield). Geol Mijnb 61(3):229–242
Priem HNA, Kroonemberg SB, Boelrijk NAI, Hebeda EH (1989) Rb-Sr and K-Ar evidence of
1.6Ga basement underlying the 1.2Ga Garzón-Santa Marta granulitic belt in the Colombian
Andes. Precambrian Res 42:315–324
Ramírez DA, López A, Sierra GM, Toro G (2006) Edad y proveniencia de las rocas volcanico
sedimentarias de la Formación Combia en el suroccidente antioqueño – Colombia. Boletín de
Ciencias de la Tierra 19:9–26
Ramos VA (1999) Plate tectonic setting of the Andean Cordillera. Episodes 22(3):183–190
Ramos VA (2009) Anatomy and global context of the Andes: main geologic features and the
Andean orogenic cycle. In: Kay SM, Ramos VA, and Dickinson WR (eds) backbone of the
Americas: shallow Subduction, plateau uplift, and ridge and Terrane collision. Geol Soc Am
Mem 204:31–65
Redwood SD (2018) The Geology of the Panama-Chocó Arc Springer volumen – Confirm defini-
tive reference
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 407

Restrepo JJ, Toussaint JF (1988) Terrains and continental accretion in the Northern Andes.
Episodes 11:189–193
Restrepo JJ, Toussaint JF, González H (1981) Edades Mio-Pliocenas del magmatismo asociado a
la Formación Combia, Departamento de antioquia y Caldas, Colombia. Geología Norandina
3:21–26
Restrepo JJ, Ordóñez-Carmona O, Armstrong R, Pimentel MM (2011) Triassic metamorphism in
the northern part of the Tahamí Terrane of the central cordillera of Colombia. J S Am Earth Sci
32(4):497–507
Restrepo-Moreno SA, Foster DA, Kamenov GD (2007) Formation age and magma sources for the
Antioqueño Batholith derived from LA-ICP-MS Uranium-Lead dating and Hafnium-isotope
analysis of zircon grains. Geol Soc Am Abstr Programs 39(6):493
Restrepo-Pace PA (1992) Petrotectonic characterization of the Central Andean Terrane, Colombia.
J S Am Earth Sci 5(1):97–116
Restrepo-Pace PA (1995) Late Precambrian to Early Mesozoic tectonic evolution of the Colombian
Andes based on new geological, geochemical and isotopic data. PhD thesis, The University of
Arizona
Restrepo-Pace PA, Ruíz J, Gehrels G, Cosca M (1997) Geochronology and Nd isotopic data
of Grenville-age rocks in the Colombian Andes  – new constraints for late Proterozoic–
early Paleozoic paleocontinental reconstructions of the Americas. Earth Planet Sci Lett
150:427–441
Restrepo-Pace PA, Cediel F (2010) Northern South America basement tectonics and implications
for paleocontinental reconstructions of the Americas. J S Am Earth Sci 29:764–771
Richards JP, Kerrich R (2007) Adakite-like rocks – their diverse origins and questionable role in
metallogenesis. Econ Geol 102(4):537–576
Rickwood PC (1989) Boundary lines within petrologic diagrams which use oxides of major and
minor elements. Lithos 22(4):247–263
Rodríguez AL (2014) Geology, Alteration, Mineralization and Hydrothermal Evolution of the
La Bodega  – La Mascota deposits, California-Vetas Mining District, Eastern Cordillera of
Colombia, Northern Andes. M.Sc. thesis, The University of British Columbia
Rodriguez G, Arango MI (2013) Formación Barroso: arco volcanico toleitico y diabasas de
San José de Urama: un prisma acrecionario T-MORB en el segmento norte de la Cordillera
Occidental de Colombia. Boletín de Ciencias de la Tierra 33:17–38
Rodríguez G, Zapata G (2012) Características del plutonismo Mioceno superior en el segmento
norte de la Cordillera Occidental e implicaciones tectónicas en el modelo geológico del noroc-
cidente colombiano. Boletín de Ciencias de La Tierra 31:5–22
Rodríguez G, Zapata G (2014) Descripción de una nueva unidad de lavas denominada Andesitas
basálticas de El Morito-correlación regional con eventos magmáticos de arco. Boletín de
Geología 36(1):85–102
Rodríguez AL, Bissig T, Hart CJ, Mantilla LC (2017) Late Pliocene high-Sulfidation epithermal
gold mineralization at the La Bodega and La Mascota Deposits, Northeastern Cordillera of
Colombia. Econ Geol 112(2):347–374
Rodríguez G, Arango MI, Zapata G, Bermúdez JG (2014) Petrografía y geoquímica del Neis de
Nechí. Boletín de Geología 36(1):71–84
Rodríguez-Vargas A, Koester E, Mallmann G, Conceição RV, Kawashita K, Weber MBI (2005)
Mantle diversity beneath the Colombian Andes, northern volcanic zone: constraints from Sr
and Nd isotopes. Lithos 82(3):471–484
Royero JM, Clavijo J  (2001) Mapa geologico generalizado del Departamento de Santander.
Memoria Explicativa: INGEOMINAS, Bogotá
Ruiz J, Tosdal RM, Restrepo PA, Murillo-Muñetón G (1999) Pb isotope evidence for Colombia-­
southern Mexico connections in the Proterozoic. In: Ramos VA, Keppie JD (eds.) Laurentia-­
Gondwana Connections before Pangea. Geological Society of America Special Paper
336:183–197
Saenz EA (2003) Fission track thermochronology and denudational response to tectonics in the
north of The Colombian Central Cordillera. MSc Thesis, Shimane University
408 H. Leal-Mejía et al.

Santô T (1969) Characteristics of seismicity in South America. Bull Earthquake Res Inst
47:635–672
Sarmiento LF (2001) Mesozoic rifting and Cenozoic basin inversion history of the Eastern
Cordillera, Colombian Andes. Inferences from tectonic models. Ph.D. Thesis, Vrije Universiteit
Sarmiento LF (2018) Cretaceous stratigraphy and paleo-facies maps of Northwestern South
America. In: Cediel F, Shaw RP (eds) Geology and tectonics of Northwestern South America:
the Pacific-Caribbean-Andean junction. Springer, Cham, pp 673–739
Saylor JE, Stockli DF, Horton BK, Nie J, Mora A (2012) Discriminating rapid exhumation from
syndepositional volcanism using detrital zircon double dating: implications for the tectonic his-
tory of the Eastern Cordillera, Colombia. Geol Soc Am Bull 124(5–6):762–779
Schmidt-Thomé M, Feldhaus L, Salazar G, Muñoz R (1992) Explicación del mapa geológico,
escala 1:250 000, del flanco oeste de la Cordillera Occidental entre los ríos Andágueda y
Murindó, Departamentos Antioquia y Chocó, República de Colombia. Geologisches Jahrbuch
Reihe B, Band B78
Shaw RP, Leal-Mejía H, Melgarejo JC (2018) Phanerozoic metallogeny in the Colombian Andes:
a tectono-magmatic analysis in space and time. In: Cediel F, Shaw RP (eds) Geology and
tectonics of Northwestern South America: the Pacific-Caribbean-Andean junction. Springer,
Cham, pp 411–535
Sillitoe RH, Jaramillo L, Damon PE, Shafiqullah M, Escovar R (1982) Setting, characteristics and
age of the Andean porphyry copper belt in Colombia. Econ Geol 77:1837–1850
Sillitoe RH, Jaramillo L, Castro H (1984) Geologic exploration of a molybdenum-rich porphyry
copper deposit at Mocoa, Colombia. Economic Geology 79(1):106–123
Silva JC, Arenas JE, Sial AN, Ferreira VP, Jiménez D (2005) Finding the Neoproterozoic-Cambrian
transition in carbonate successions from the Silgará Formation, Northeastern Colombia:
an assessment from C-isotope stratigraphy. In: Memorias del X Congreso Colombiano de
Geología, Bogota
Silver EA, Reed DL, Tagudin JE, Heil DJ (1990) Implications of the north and south Panama thrust
belts for the origin of the Panama orocline. Tectonics 9:261–281
Silver PG, Russo RM, Lithgow-Bertelloni C (1998) Coupling of south American and African plate
motion and plate deformation. Science 279:60–63
Singewald QD (1950) Mineral resources of Colombia (other than petroleum). US Geol Surv Bull
964–B:56–204
Sinton CW, Duncan RA, Storey M, Lewis J, Estrada JJ (1998) An oceanic flood basalt province
within the Caribbean plate. Earth Planet Sci Lett 155(3):221–235
Spikings R, Cochrane R, Villagómez D, Van der Lelij R, Vallejo C, Winkler W, Beate B (2015) The
geological history of northwestern South America – from Pangaea to the early collision of the
Caribbean large Igneous Province (290–75 Ma). Gondwana Res 27:95–139
Stanley CR, Madeisky HE (1994) Lithogeochemical Exploration for Hydrothermal Ore Deposits
using Pearce Element Ratio Analysis. In: Lentz DR (ed) Alteration and Alteration Processes
associated with Ore-forming Systems. Geological Association of Canada, Short Course Notes
11:193–211
Stanley CR, Madeisky HE (1996) Lithogeochemical exploration for metasomatic zones associated
with hydrothermal mineral deposits using Pearce Element Ratio Analysis. Short Course Notes
on Pearce Element Ratio Analysis, Mineral Deposit Research Unit (MDRU), the University of
British Columbia, Canada
Stern RJ (2002) Subduction zones. Reviews of geophysics 40(4):3-1-3-38
Stern CR (2004) Active Andean volcanism: its geologic and tectonic setting. Revista geológica de
Chile 31(2):161–206
Stevenson RK, Patchett PJ (1990) Implications for the evolution of continental crust from Hf
isotope systematics of Archean detrital zircons. Geochim Cosmochim Acta 54(6):1683–1697
Taboada A, Rivera LA, Fuenzalida A, Cisternas A, Philip H, Bijwaard H, Olaya J, Rivera C
(2000) Geodynamics of the northern Andes  - Subductions and intracontinental deformation
(Colombia). Tectonics 19(5):787–813
5  Spatial-Temporal Migration of Granitoid Magmatism and the Phanerozoic… 409

Tassinari CCG, Diaz F, Buenaventura J  (2008) Age and source of gold mineralization in the
Marmato mining district, NW Colombia – a Miocene-Pliocene epizonal gold deposit. Ore Geol
Rev 33:505–518
Tistl M (1994) Geochemistry of platinum-group elements of the zoned ultramafic Alto Condoto
Complex, Northwest Colombia. Econ Geol 89(1):158–167
Tistl M, Burgath KP, Höhndorf A, Kreuzer H, Muñoz R, Salinas R (1994) Origin and emplace-
ment of tertiary ultramafic complexes in northwest Colombia: evidence from geochemistry and
K-Ar, Sm-Nd and Rb-Sr isotopes. Earth Planet Sci Lett 126(1–3):41–59
Toro G, Restrepo JJ, Poupeau G, Saenz E, Azdimousa A (1999) Datación por trazas de fision de
circones rosados asociados a la secuencia volcano-sedimentaria de Irra (Caldas). Boletín de
Ciencias de la Tierra 13:28–34
Tschanz CM, Marvin RF, Cruz J, Mehnert H, Cebulla G (1974) Geologic evolution of the Sierra
Nevada de Santa Marta area, Colombia. Geol Soc Am Bull 85:273–284
Trumpy D (1949) Geology of Colombia. Shell Unpublished Report No. 23323
Ujueta G, Macia C, Romero F (1990) Cuerpo Riodacítico del Terciario Superior en la Región de
Quetame, Cundinamarca. Geología Colombiana 17:143–150
Van der Hilst RD (1990) Tomography with P, PP and pP delay-time data and the three-dimensional
mantle structure below the Caribbean region. Ph.D. Thesis, Instituut voor Aardwetenschappen
der Rijksuniversiteit Utrecht
Van der Hilst R, Mann P (1994) Tectonic implica- tions of tomography images of subducted litho-
sphere beneath northwestern South America. Geology 22:451–454
Van der Lelij R (2013) Reconstructing north-western Gondwana with implications for the evolu-
tion of the Iapetus and Rheic Oceans: a geochronological, thermochronological and geochemi-
cal study. PhD thesis, Université de Genève
Van der Lelij R, Spikings RA, Ulianov A, Chiaradia M, Mora A (2016) Palaeozoic to early Jurassic
history of the northwestern corner of Gondwana, and implications for the evolution of the
Iapetus, Rheic and Pacific Oceans. Gondwana Res 31:271–294
Vásquez M, Altenberger U, Romer RL, Sudo M, Moreno-Murillo JM (2010) Magmatic evolution
of the Andean Eastern Cordillera of Colombia during the cretaceous: influence of previous
tectonic processes. J S Am Earth Sci 29(2):171–186
Vargas CA, Mann P (2013) Tearing and breaking off of subducted slabs as the result of colli-
sion of the Panama arc-indenter with Northwestern South America. Bull Seismol Soc Am
103(3):2025–2046
Vesga AM, Jaramillo JM (2009) Geoquímica del domo volcánico en el Municipio de Iza,
Departamento de Boyacá  – Interpretación geodinámica y comparación con el vulcanismo
Neógeno de la Cordillera Oriental. Boletín de Geología (UIS) 31(2):97–108
Villagómez D (2010) Thermochronology, geochronology and geochemistry of the Western and
Central cordilleras and Sierra Nevada de Santa Marta, Colombia: The tectonic evolution of NW
South America. PhD thesis, Université de Genève
Villagómez D, Spikings R, Magna T, Kammer A, Winkler W, Beltrán A (2011) Geochronology,
geochemistry and tectonic evolution of the Western and Central cordilleras of Colombia.
Lithos 125:875–896
Vinasco CJ (2004) Evolução crustal e história tectônica dos granitóides permo-triássicos dos
Andes do Norte. PhD thesis, Universidade de Sao Paulo
Vinasco C (2018) The romeral shear zone. In: Cediel F, Shaw RP (eds) Geology and tectonics
of Northwestern South America: the Pacific-Caribbean-Andean junction. Springer, Cham,
pp 833–870
Vinasco CJ, Cordani UG, González H, Weber M, Pelaez C (2006) Geochronological, isotopic,
and geochemical data from Permo-Triassic granitic gneisses and granitoids of the Colombian
Central Andes. J S Am Earth Sci 21:355–371
Ward DE, Goldsmith R, Cruz J, Jaramillo C, Restrepo H (1973) Geología de los cuadrangu-
los H-12 Bucaramanga y H-13 Pamplona, Departamento de Santander. Boletín Geológico
INGEOMINAS 21(1–3):1–132
410 H. Leal-Mejía et al.

Warren I, Simmons SF, Mauk JL (2007) Whole-rock geochemical techniques for evaluating hydro-
thermal alteration, mass changes, and compositional gradients associated with epithermal
Au-Ag mineralization. Econ Geol 102:923–948
Weber B, Iriondo A, Premo W, Hecht L, Schaaf P (2007) New insights into the history and origin
of the southern Maya block, SE México: U–Pb–SHRIMP zircon geochronology from meta-
morphic rocks of the Chiapas massif. Int J Earth Sci 96(2):253–269
Weber MB, Tarney J, Kempton PD, Kent RW (2002) Crustal make-up of the northern Andes –
evidence based on deep crustal xenolith suites, Mercaderes, SW Colombia. Tectonophysics
345(1):49–82
Weber M, Cardona A, Valencia V, García-Casco A, Tobón M (2010) U/Pb detrital zircon provenance
from late cretaceous metamorphic units of the Guajira peninsula, Colombia: tectonic implica-
tions on the collision between the Caribbean arc and the South American margin. J S Am Earth
Sci 29(4):805–816
Weber M, Gómez-Tapias J, Cardona A, Duarte E, Pardo-Trujillo A, Valencia VA (2015)
Geochemistry of the Santa Fé Batholith and Buriticá Tonalite in NW Colombia and evidence of
subduction initiation beneath the Colombian Caribbean plateau. J S Am Earth Sci 62:257–274
Wegner W, Wörner G, Harmon RS, Jicha BR (2011) Magmatic history and evolution of the Central
American land bridge in Panama since cretaceous times. Geol Soc Am Bull 123(3–4):703–724
Wessel P, Kroenke LW (2008) Pacific absolute plate motion since 145 Ma: an assessment of the
fixed hot spot hypothesis. J Geophys Res 113(B06101):1–21
Wilson M (1989) Igneous petrogenesis a global tectonics approach. Chapman & Hall, London
Wood DA, Tarney J, Varet J, Saunders AD, Bougault H, Joron JL, Treuil M, Cann JR (1979)
Geochemistry of basalts drilled in the North Atlantic by IPOD leg 49: implications for mantle
heterogeneity. Earth Planet Sci Lett 42(1):77–97
Wright JE, Wyld SJ (2011) Late cretaceous subduction initiation on the eastern margin of the
Caribbean-Colombian oceanic plateau: one great arc of the Caribbean (?). Geosphere 7:468–493
Zapata G, Rodríguez G (2013) Petrografía, Geoquímica y edad de la Granodiorita de Farallones y
las rocas volcánicas asociadas. Boletín de Geología 35(1):81–96
Zapata S, Cardona A, Jaramillo C, Valencia V, Vervoort J (2016) U-Pb LA-ICP-MS geochronology
and geochemistry of Jurassic volcanic and plutonic rocks from the Putumayo region (Southern
Colombia):tectonic setting and regional correlations. Boletín de Geología 38(2):21–38
Zarifi Z, Havskov J, Hanyga A (2007) An insight into the Bucaramanga nest. Tectonophysics
443(1):93–105
Zartman RE, Doe SM (1981) Plumbotectonics – the model. Tectonophysics 75:135–162