1 Introduction

The Maracaibo continental block (MCB), northwestern South America , comprises a low lying area, which includes the Maracaibo basin , on its edges prominent mountain belts (the Sierra Nevada de Santa Marta (SNSM), the Santander Massif (SM), the Mérida or Venezuelan Andes (VA), the Trujillo Mountains (TM) part of the VA, and the inner part the Serranía de Perijá (SP)), and the Cesar-Ranchería and Maracaibo basins (Fig. 13.1). The MCB can be considered as the northwesternmost fragment of the Guiana Shield, overlain by extensive Phanerozoic supracrustal sequences. During the late Cretaceous, the MCB began to migrate northwestward along the Santa Marta-Bucaramanga and Oca fault systems (Fig. 13.1; Pindell 1993). Transpressional forces generated during this process resulted in the development of different tectonic blocks such as the Mérida Andes, the Caparo block, and the Serranía de Trujillo or Trujillo block within the VA (inset in Fig. 13.2, Bermúdez et al. 2010; Figs. 13.2 and 13.5), SM, SP, and SNSM. The MCB is distinguished from other continental blocks to the northwest of the Guiana Shield by a unique and regionally constrained style of deformation, which originated during the Mesozoic-Cenozoic period as a result of interactions between the Pacific (Nazca), Caribbean, and continental South American plates (Fig. 13.1).

Fig. 13.1
figure 1figure 1

Major tectonic features of northern South America. (Modified after Colmenares and Zoback 2003)

Full size image
Fig. 13.2
figure 2figure 2

(a) Simplified geological map of the MCB and surrounding belts. (Modified from Gómez et al. 2015; Hackley et al. 2005). (b) Seismicity between 1911 and 2013 using data available from the digital library of the Geophysics Laboratory of Universidad de Los Andes, Mérida (http://lgula.ciens.ula.ve/), data provided by the Venezuelan Foundation for Seismological Investigation (FUNVISIS; http://www.funvisis.gob.ve), the National Seismological Network of Colombia (http://seisan.sgc.gov.co/RSNC/), and the United States Geological Survey (USGS; https://earthquake.usgs.gov/earthquakes/feed/v1.0/csv.php). Faults and DEM were obtained from http://www.geomapapp.org; Ryan et al. (2009) and Bermúdez et al. (2013)

Full size image

Detailed Global Positioning System (GPS ) measurements and a recent compilation of displacement rate data (e.g., offsets of glacial moraines and pyroclastic flows) suggest that a large part of the Northern Andes, including the MCB, is currently escaping to the northeast relative to “stable” South America at rates close to 6 ± 2 mm/a (Egbue and Kellogg 2010).

The purpose of this chapter is to provide a general overview of the temporal and spatial morphotectonic evolution of the different mountain belts belonging to the MCB within the Caribbean and Northern Andes tectonic framework.

2 Structural Settings of the Maracaibo Mountain Belts

The Maracaibo mountain belts are crossed by different system faults: Santa Marta-Bucaramanga fault in the west, the Valera fault in the east, the Oca fault to the north, and Boconó fault to the south (Fig. 13.1). Figure 13.2 shows a simplified geological map of the SNSM, SP, SM, and VA (Gómez et al. 2015; Hackley et al. 2005).

2.1 Sierra Nevada de Santa Marta

The triangular SNSM massif is an isolated mountain range bounded by the Oca fault, Santa Marta-Bucaramanga fault, and the Cesar lineament (Fig. 13.3). It forms the world’s highest coastal mountain range with ~5750 m elevation across ~40 km of coastline. During the Tertiary, dextral, and sinistral activity, movement of 65 and 110 km, respectively, occurred along the Oca and Santa Marta-Bucaramanga faults. Subsequently, erosional exhumation of several thousand meters led to development of the current geomorphological shape (Tschanz et al. 1974; Idárraga-García and Romero 2010).

Fig. 13.3
figure 3figure 3

Available radiometric age data for SNSM and SP (Dataset compiled from Hoorn et al. 2010; Herman et al. 2013; Gómez et al. 2015; Piraquive 2017 and references therein). Each panel shows the distribution of data derived by a different geo-thermochronological method or on a different mineral. Ages shown do not take into account any consideration of their uncertainties. In this figure SF corresponds to the Sevilla fault

Full size image

The SNSM is divided into three geological provinces comprising magmatic and metamorphic rocks separated by the Sevilla and Cesar lineaments (see Colmenares et al.’s chapter in this book; Tschanz et al. 1974; Bustamante et al. 2009):

  1. 1.

    The Sierra Nevada province, an old granulite terrane (1.3 Ga) overlain by unmetamorphosed Paleozoic and Permian(?)-Triassic rocks (Fig. 13.3).

  2. 2.

    A metamorphic terrane consisting of Precambrian amphibolite-grade metamorphic rocks overlain by Silurian phyllites and unmetamorphosed Paleozoic and Mesozoic rocks, which are typical of the Colombian Western Cordillera (Doolan 1970; Tschanz et al. 1974; Restrepo-Pace et al. 1997; Ordoñez et al. 2002; Cordani et al. 2005; Cardona et al. 2006, 2010).

  3. 3.

    The youngest province consists of three northeast-trending regional metamorphic belts, which are intruded by Tertiary plutons. These belts comprise Permian-Triassic gneiss, Jurassic schist, and Cretaceous-Paleocene greenschist and were formed in the same subduction zone setting as the plutonic belts in province (2) above.

Cardona et al. (2006) proposed Triassic-Jurassic heating and a cooling pulse in the SNSM based on the oldest individual step-heating ages from discordant hornblende and biotite 40Ar/39Ar age spectra, for a paragneiss in the Paleozoic Sevilla Complex (minimum zircon U-Pb age of 529 ± 10 Ma; Cardona et al. 2006; Fig. 13.3a). Because the age spectra suggest the existence of thermal perturbations, the 40Ar/39Ar ages are considered to be inconclusive (Villagómez 2010). Cardona et al. (2010) reported zircon and apatite (U-Th)/He ages from crystalline rocks exposed along a NW-SE traverse within the northwestern region of the Santa Marta Province, which range from 20–27 Ma and 5–24 Ma, respectively. These authors assume a high geothermal gradient of 50 °C/km and propose moderate exhumation rates of 0.16 km/Myr during the late Oligocene and 0.33 km/Myr since the middle–late Miocene.

Villagómez (2010) reported zircon and apatite fission-track (AFT) ages across the SM and SNSM. From these, 14 samples were collected along a single traverse oriented approximately perpendicular to the northwestern corner of the SNSM (Fig. 13.3). The oldest AFT ages within the SNSM were obtained from the southern and eastern Sierra Nevada Province. Six Jurassic granitoids and a single rhyolite sample proximal to the Cesar lineament (CL, Fig. 13.3) yield apatite fission-track ages between ~60 Ma and ~40 Ma. These rocks form part of the Andean-wide, Jurassic continental arc that extends along the western South American margin. Despite the fact that these samples reside in three or more distinct faulted blocks (Fig. 13.3), almost all AFT ages are indistinguishable over an elevation range of 400–2700 m. Rocks located within 10 km of the deformation zone of the Santa Marta-Bucaramanga fault yield indistinguishable AFT ages between 23 and 30 Ma (Villagómez 2010; Villagómez and Spikings 2013), which are clearly younger than less deformed regions of the same province to the east. Mean AFT lengths range between 12.21 ± 1.70 μm and 13.96 ± 0.17 μm (Table 1 in Villagómez 2010). Ages are interpreted as dating times of deformation and exhumation associated with movement along the Santa Marta-Bucaramanga fault.

The youngest AFT ages within the SNSM, ranging between 16 Ma and 28 Ma, are found to the north of the Sevilla fault (see Colmenares’s chapter in this book) in the Santa Marta and Sevilla provinces (SF, Figs. 13.3a and 13.4). Sample SN6 of Villagómez (2010) yielding an apatite fission-track age of 41.0 ± 9.6 Ma is considered to relate to the onset of exhumation.

Fig. 13.4
figure 4figure 4

Available radiometric age data for the SM and Bucaramanga fault (Dataset compiled from Hoorn et al. 2010; Herman et al. 2013; Mora et al. 2015; Gómez et al. 2015; Amaya 2016; Amaya et al. 2017 and references therein). Each panel shows the distribution of data derived by a different geo-thermochronological method or on a different mineral. Ages shown do not take into account any consideration of their uncertainties

Full size image

2.2 The Serranía de Perijá

Although it might appear that the SP is an extension of the SM in Colombia, it is convenient to describe it separately. The transition between both mountain ranges occurs approximately at latitude 9.5°N (Shagam 1975; Fig. 13.3). The eastern foothills of the SP are defined by two erosional unconformities (Bucher 1952): (1) a marine sequence of mottled claystones and siltstones of Oligocene-Miocene age, which overlaps westward with Eocene and Cretaceous marine sedimentary units and (2) sandstones of the (late Pliocene-Pleistocene?) Milagro Formation, cropping out along the eastern flank of the SP, between the village of Riecito (at about 9.5°N) to the southwest and Maracaibo city (at about 10.6 °N) to the northeast, where this unit is in direct contact with pre-Cretaceous igneous rocks. AFT data in this area indicate cooling by the late Oligocene–early Miocene (22–27 Ma) for the Jurassic Palmar granite and Paleozoic metasedimentary rocks in the Palmar High and Pliocene (ca. 3 Ma) cooling in the Totumo-Inciarte arch that affected the Jurassic La Quinta Formation and Paleozoic Lajas granite (Shagam et al. 1984). (Plate 1 in Miller 1962; Shagam et al. 1984) (Fig. 13.3).

Fault systems and folds of the SP trend N35°E. The main structures of the SP, the Perijá and Tigre faults, play a significant mechanical role, dividing the chain into several tectonic blocks. Other faults present in the area, such as the Cesar fault on the northwestern side of the SP and the Cuiba fault on the southeastern side, form the boundary faults of the SP (Fig. 13.3; Cediel’s chapter in this book). The Cuiba fault appears to be a high-angle reverse fault (Fig. 14 in Shagam 1975). The east-trending transverse Oca fault system defines the northern termination of the SP range and the SNSM block. In the northern part of this range, south-trending faults are connected with others in the Maracaibo Basin (Fig. 13.1). Crossing through the southern and south-central part of the SP, a system of transverse faults is recognized in the Cesar-Ranchería Basin (Arena Blanca trend in Miller 1962).

The Perijá-El Tigre fault system crosses diagonally through the SP and marks a syncline and structural depression that segments the range. Most of the northern SP lies to the west of the fault, including the Serranía de Valledupar and structural blocks between Cuiba and Cesar faults (Fig. 14 in Shagam 1975; Fig. 13.3). The entire southern segment of the range lies to the east of this last area forming the Rio Negro anticlinal feature and the structural feature known as the Totumo-Inciarte arch. The Perijá fault terminates or is sharply offset where it meets the Arena Blanca displacement closer to the town of El Tucuy (Plate 1 in Miller 1962).

Development of the Cretaceous Perijá trough and early Eocene movement on the margin of this trough are related to Oca fault activity. Several raised beaches formed across the Oca fault zone near Sinamaica show evidence of Pleistocene or recent movement on the fault.

On the east side of the SP, different structural features can be observed. For example, an en echelon pattern of anticlinal prongs controls a portion of the eastern front of the range, which reappears with poor development and reversed direction in the north. Other features include largely concealed anticlinal belts in the structurally high region west of Maracaibo, as well as the abrupt termination of this mountain range.

SP relief is more moderate in comparison with SNSM, SM, and VA, and the range was apparently not uplifted to the same degree as the VA (Shagam 1975). Perijá surface uplift was counterbalanced by subsidence in the Cesar Valley and the northwestern Maracaibo Basin. Miller (1962) proposed at least four distinct Tertiary surface uplift episodes in the SP: (1) Eocene growth of the Totumo arch; (2) Oligocene activity such as reactivation of the Oca fault; (3) post-Oligocene tectonism producing strong folding and faulting, which formed the current architecture of the chain; and (4) structural activity apparently extended into the late Pliocene but since that time has been somewhat subdued.

2.3 Santander Massif

The SM, located in the Eastern Cordillera, represents the western boundary of the MCB (Figs. 13.2 and 13.4; see Zuluaga et al. chapter in this book). The geological history of the SM is complex. Its basement is composed of igneous and metamorphic rocks of the Precambrian Bucaramanga gneiss and the pre-Devonian Silgará Formation, which are unconformably overlain by non-metamorphic Carboniferous and Permian clastic to calcareous sedimentary rocks of the Floresta Formation. The Silgará Formation was intruded by Mesozoic quartz monzonite, diorite, and other Cretaceous or younger porphyritic rocks. The Jurassic Jordan and Girón Formations are separated by unconformities from the Cretaceous Tambor, Rosa Blanca, Paja, Tablazo, Simití, La Luna, and Umir Formations. Igneous rocks with porphyritic-aphanitic and porphyritic-phaneritic textures are related to the Miocene and most recent magmatic events across this massif (Mantilla et al. 2011). The SM is a structurally uplifted block, bordered to the west by the Santa Marta-Bucaramanga fault and to the east by the Pamplona fault system (see Zuluaga et al. in this book). This latter fault system defines the western boundary of the MCB.

Ward et al. (1974) suggested that foliation and fold orientations in the Bucaramanga Gneiss, Silgará Formation, and orthogneissic rocks of the SM are approximately similar. Regionally, these orientations are N-S, and tend to parallel the axis of the orogenic belt (Eastern Cordillera), showing local variations and discontinuities due to the presence of intrusive bodies of the Santander Plutonic Complex and a sedimentary cover. Restrepo-Pace (1995) proposed a shear-couple model for the SM fault system. The SM fault system requires a left-lateral sense of shear along faults such as the Bucaramanga fault and the Pamplona thrust (which can be considered a part of the Pamplona fault system). Displacements along Santa Marta-Bucaramanga faults generated a structural dissection of the SM, leading to the formation of rhomboidal blocks with Precambrian to Paleozoic rocks intruded by Mesozoic granites (van der Lelij et al. 2016). Along the western flank, an exhumed zircon fission-track partial annealing zone is exposed, and since the late Miocene based on AFT, exhumation rates in the SM are in the order of 0.3–0.4 km/Myr (Amaya 2016; Amaya et al. 2017).

2.4 Venezuelan or Mérida Andes

The VA form a complex orogen which extends over a length of ~500 km in SW-NE direction (Figs. 13.1 and 13.5). The SW-NE oriented Boconó strike-slip fault runs along this mountain belt and divides it almost symmetrically into two main tectonic blocks, the Sierra la Culata (to the north) and Sierra Nevada (to the south). The VA attains a maximum elevation of 4978 m at Pico Bolívar in the Sierra Nevada block. The orogen is bounded by northern and southern thrust faults (Fig. 13.5). The VA is formed due to convergence between the Caribbean, Nazca, and South American plates and the MCB (Colletta et al. 1997). Significant surface uplift of the central part of the VA had a major impact on the deviation of the Orinoco, Amazonas, and Magdalena paleo-drainage systems, which in turn had an important influence on the biodiversity of northern South America (Hoorn et al. 1995, 2010).

Fig. 13.5
figure 5figure 5

Available radiometric age data for the VA. Each panel shows the distribution of data derived by a different geo-thermochronological method or on a different mineral. Ages shown do not take into account any consideration of their uncertainties. Note panel (G) shows main tectonic blocks (All compiled from Kohn et al. 1984; Bermúdez et al. 2010, 2011, 2014; van der Lelij et al. 2016). In this last panel, the abbreviations CATB, CB, EB, ECB, SLCB, SNB, and TB correspond to Cerro Azul Thrust, Caparo, Escalante, El Carmen, Sierra la Culata, Sierra Nevada, and Trujillo tectonic blocks, respectively

Full size image

Bermúdez et al. (2010) demonstrated an important control of inherited faults and structures on exhumation patterns across the Mérida Andes and divided the orogen into at least seven individual active tectonic blocks (Fig. 13.5). Further, using fission-track dating and 3-D thermokinematic Pecube modeling (Braun 2003; Braun et al. 2012), Bermúdez et al. (2011) showed an asynchronous exhumation history of the two tectonic blocks located in the central part of this chain. The Sierra Nevada block located to the south of the Boconó fault was exhumed since 14–12 Ma at rates of 1.5 km/Myr until about 6–4 Ma. Later these rates decreased to 0.5 km/Myr. In contrast, the El Carmen intrusive body, a sliver of the Sierra La Culata block to the north of the Boconó fault was exhumed from around 6 Ma to the present at rates of 1.5 km/Myr. This indicates that movement along the Boconó strike-slip fault includes an important oblique component, which allows for accommodation of deformation and exhumation. The exhumation rate however does not explain the high elevation of this chain. The surface uplift may be explained in part by compression of tectonic plates and in part by isostatic compensation. The Caparo and Trujillo blocks, located at the northeastern and southwestern terminations of this belt, exhibit lower mean elevations and slightly lower mean slopes than the central blocks (Sierra Nevada and Sierra La Culata blocks). Based on fission-track data, these two tectonic blocks cooled slowly from the Oligocene to late Miocene (Bermúdez et al. 2011). From a seismic point of view, the Caparo and Trujillo blocks are very different (Fig. 13.2b). Earthquakes to the south of the Caparo block are concentrated on the vertex formed by the Bramón, Central-Sur Andino, and Caparo fault systems. In this area earthquakes tend to be deep (>60 km) with magnitudes between 4.2 and 5.7. To the north of the Caparo block, seismicity is restricted to an active branch of the Boconó fault system (San Simón fault), and earthquakes with magnitudes between 3.5 and 4.5 are shallower (20–40 km). This seismic area coincides with the Escalante block, an active tectonic block with AFT ages and track length distributions similar to the Sierra La Culata block (2–6 Ma). Seismic activity in the Trujillo block is concentrated across the boundary of the blocks defined by the Valera, Boconó, and Burbusay-Carache faults (Bermúdez et al. 2010, 2013). Deeper earthquakes for this area (40–60 km) with magnitudes between 3.5 and 5.0 are focused on the Valera fault. Further, AFT ages >30 Ma are restricted to the Trujillo block (Figs. 13.6 and 13.7 in Bermúdez et al. 2010). The northern part of the Trujillo block shows an older AFT age of 145 ± 7 Ma (sample SVA-80-44; Bermúdez et al. 2010), which was obtained for the Valera Granite. This pattern reflects a pre-orogenic cooling history. Younger AFT ages (3.5–5.8 Ma) are observed in the southwest of the Valera block. The Trujillo block was not affected by the rapid exhumation observed in the central Mérida Andes. Seismicity across the Cerro Azul block, located in the southwestern foothills of the VA, is not uniform. AFT data from Proterozoic rocks located in this block suggest that the exhumation history is similar to that of the Sierra Nevada block (Figs. 13.2 and 13.5).

Fig. 13.6
figure 6figure 6

Age-elevation relationships (AER) obtained using York’s method for different thermochronometers from the SNSM, SM, and VA. For SNSM: (a) AHe-AER, (b) AFT-AER, (c) ZHe-AER. For SM: (d) AFT-AER (gray line and hexagons correspond to AER for Bucaramanga-Picacho profile, Mora et al. 2015; Amaya et al. 2017; black line and white hexagons correspond to a profile to the east of SMB fault, (e) ZFT-AER, (f) Biotite 40Ar/39Ar AER. For VA: (g) AFT-AER for different tectonic blocks, (h) AFT-AER ages only for Sierra Nevada and Sierra La Culata blocks. Probability density function of apatite fission-track ages in (i) SNSM, (j) SM, (k) VA, and (l) SP mountain belts. (m) Map of interpolated AFT ages for the circum-MCB mountain belts

Full size image
Fig. 13.7
figure 7figure 7

(a) Summary of the cooling/denudation history of circum-MCB mountain belts. (b) Current tectonic configuration of accreted terranes and plates. In this figure DAP is Dagua-Piñon terrane, SJ is the San Jacinto terrane, GOR is the Gorgona terrane, CG is the Cañas Gordas terrane, BAU is the Baudó terrane, GU-FA is the Guajira-Falcón terrane, and CAM is the Caribbean Mountain terrane (see Cediel et al. 2003; and Cediel et al. chapter in this book)

Full size image

In summary, the VA experienced rapid exhumation starting in the middle Miocene, which was controlled by tectonic activity along the Boconó fault. Using a multidisciplinary approach, in which different types of data were used to reconstruct the timing of mountain building and the surface uplift history, it has been determined that the VA had reached present-day elevations by late Miocene to Pliocene time (Kohn et al. 1984; Bermúdez et al. 2010). The evolution of the VA is contemporaneous with the evolution of the Eastern Cordillera in Colombia (Bermúdez et al. 2017), for which climatic (Mora et al. 2008, 2010) and tectonic (Parra et al. 2010) controls have been documented, particularly over the past 35 Ma.

3 Exhumation and Age-Elevation Relationships

For discriminating between differences in exhumation across the different mountain belts, we compiled a thermochronological dataset comprising 34 muscovite, 81 biotite, 55 hornblende, and 9 K-feldspar 40Ar/39Ar ages; 103 zircon and 46 apatite (U-Th)/He ages; and 2 sphene, 95 zircon, and 309 apatite fission-track ages (including ages obtained by both the LA-ICP-MS and external detector methods). The dataset used is based on previous compilations by the Colombian Geological Service (Gómez et al. 2015), as well as Herman et al. (2013), Hoorn et al. (2010), Mora et al. (2015), Amaya (2016), Piraquive (2017), and Amaya et al. (2017). The data used however are not spread evenly across the different mountain belts . For the SNSM the data are concentrated in the NW foothills. Hornblende, biotite, and K-feldspar 40Ar/39Ar ages are distributed across the western part of the Oca fault (Fig. 13.4). Zircon fission-track are also concentrated in the NW foothills and closer to the Oca fault; these vary from 36.1 to 56.8 Ma (Piraquive 2017). Apatite fission-track, zircon, and apatite (U-Th)/He data are available for the westernmost border of the SNSM (Villagómez 2010; Villagómez et al. 2011). These apatite ages are based on samples located approximately perpendicular to the chain, along a unique elevation profile (24–2340 m) for this region. Some apatite fission-track ages have been reported from between the Tierra Nueva fault and the Cesar lineament (Figs. 13.2 and 13.3), but their distribution is not sufficient for estimating age-elevation relationships. No thermochronological data have been reported from the central part of the SNSM, where maximum elevations are close to 5800 m.

For the SP (Fig. 13.3), the only available data are zircon fission-track ages generated by Shagam et al. (1984) with ages ranging between 65.2 and 156 Ma. However, the sampling pattern does not permit an evaluation of any possible age-elevation relationships.

Muscovite, hornblende, and biotite 40Ar/39Ar ages are available for the SM. However, the data are not distributed in a systematic fashion across the different tectonic blocks. Further, the sampling strategy did not include collecting rocks from across a vertical transect orthogonal to the main structures, with exception of some 40Ar/39Ar biotite ages. For this last study, it is possible to estimate an exhumation rate for a tectonic block close to the Bucaramanga fault (Fig. 13.4). No zircon and apatite (U-Th)/He age data are available for the SM. So far such age data are only available for the Eastern Cordillera located between El Socorro and Cepitá, which is an area located in the western part of the Bucaramanga fault (Caballero et al. 2013) bounded by the Ocamonte and del Suárez faults, which themselves define a tectonic block (Diederix et al. 2009). Zircon and apatite fission-track ages are available from both sides of the Bucaramanga fault (Fig. 13.4). For the western flank of the SM, apatite and zircon fission-track and (U-Th)/He ages have been reported from along two different vertical profiles. These are a Bucaramanga-Picacho profile (Amaya 2016; Mora et al. 2015; van der Lelij et al. 2016; Amaya et al. 2017) and for an area delimited by the towns of Santa Bárbara, Zapatoga, El Socorro, and Cepitá (see Mora et al. 2015; Amaya et al. 2017).

For the VA (Fig. 13.5), one K-feldspar and two hornblende 40Ar/39Ar ages have been reported (Kohn et al. 1984). Van der Lelij et al. (2016) obtained 40Ar/39Ar hornblende, muscovite, biotite, and K-feldspar age data from igneous and metamorphic rocks. Twenty-one zircon and two sphene fission-track ages were reported by Kohn et al. (1984). The areal distribution of 40Ar/39Ar, fission-track, and zircon (U-Th)/He ages reported by Bermúdez et al. (2014) do not permit an estimation of any age-elevation relationships in the Caparo and Trujillo blocks. This is in contrast to apatite fission-track ages reported by Kohn et al. (1984) and Bermúdez et al. (2010, 2011) from two vertical profiles, one each from the Sierra La Culata and Sierra Nevada blocks.

4 Discussion

4.1 Exhumation Rates

In order to estimate age-elevation relationships of the different mountain belts, we selected those areas where there was a greater density of ages and datasets were not cut by major fault systems. In Fig. 13.6, we summarize significant areas from the different mountain belts. For the SNSM, samples collected across a vertical profile by Villagómez et al. (2011) were used. For modeling the data, we distinguish three different exhumation rates as follows: 2.7 ± 0.4 km/Myr from 24 to 20 Ma, increasing to 3.5 ± 0.3 km/Myr between 20 and 16 Ma, and finally decreasing to 0.3 ± 0.1 km/Myr between 15 and 5 Ma. Villagómez et al. (2011) sampled along a vertical profile in the foothills of the SNSM massif, where steep slopes and high rainfall would be predicted to significantly increase erosive power. However, exhumation rates are relatively low. Long-term exhumation rates across the SM were calculated using 40Ar/39Ar biotite ages. These provide a rate of 0.03 ± 0.01 km/Myr, between 180 and 200 Ma, which seem to remain almost invariant between 80 and 50 Ma. More recently, between 10 Ma and 5 Ma, exhumation rates increased to 0.5 ± 0.2 km/Myr, which are very similar to those calculated for the SNSM.

For the SM, the cooling histories of tectonic blocks are similar to the VA (Mora et al. 2009; Parra et al. 2009), but the intense faulting pattern of the SM render rigorous exhumation rate estimates difficult. Two different profiles at the east of Bucaramanga fault (Fig. 13.6) provide different exhumation rates 0.5 ± 0.1 km/Ma and 0.2 ± 0.1 km/Ma, respectively, which support the notion of differential exhumation between independent tectonic blocks. Across the VA the pattern of exhumation is different than for the SM and the SNSM. The central part of the VA indicates exhumation rates ranging between 1.3 km/Myr and 1.8 km/Myr for the Sierra La Culata and Sierra Nevada blocks, respectively. However, because 3-D thermokinematic modeling has shown asymmetric behavior and changes in exhumation rates (Bermúdez et al. 2011), these rates should be considered as first-order approximations. Further, other effects such as climate, relief change, and fluid flow in areas closer to the faults need to be taken into account.

Finally, it is very important to take into account the effect of eroding topography on steady-state isotherms (Stüwe et al. 1994). In this chapter we were careful to consider such effects; further, exhumation rates across the MCB are not always high; thus the isotherms may not be unduly perturbed.

4.2 Periods of Cooling

The cooling history of the SNSM has been poorly constrained, mainly because several 40Ar/39Ar do not show plateau ages or because they exhibit disturbed spectra (Restrepo-Pace 1995; Restrepo-Pace et al. 1997; Cordani et al. 2005). These are interpreted as a consequence of reheating during the intrusion of several plutons located at shallow depths at ~200 Ma (Dörr et al. 1995; van der Lelij et al. 2016). Some of the plateau ages range between 184 ± 3 and 213 ± 3 Ma. This age range is similar to that observed in the Bucaramanga gneiss of the SM.

For the SNSM, Villagómez (2010) proposed three cooling pulses at (1) 65–58 Ma and 50–40 Ma (Jurassic granitoids), in the central Sierra Nevada Province; (2) 40–25 Ma (Precambrian gneisses) proximal to the Santa Marta-Bucaramanga Fault with higher exhumation rates at 26–29 Ma (Jurassic granitoid) close to the Sevilla lineament, in the western Sierra Nevada Province; and (3) 25–30 Ma (Paleogene quartz diorite and aplite) and 16–25 Ma (Paleogene aplite and quartz diorites and Upper Cretaceous schist) north of the Sevilla lineament. Further, this author also proposed that cooling events could be directly associated with exhumation because (i) they broadly correlate with steep gradients in the age-elevation relationships, (ii) they are not coeval with and do not immediately postdate proximal magmatic activity, and (iii) they are contemporaneous with sedimentation in the neighboring basins (Cesar and Lower Magdalena basins).

For the SM, 40Ar/39Ar thermochronometry on muscovite, hornblende, and biotite indicates two separate thermal events: (1) 150–200 Ma related to rifting processes and (2) 60–120 Ma. Zircon fission-track data collected from Jurassic and older basement rocks exposed in the SM define two age groups at 80–120 Ma and 50–70 Ma (Shagam et al. 1984), which support an exhumed partial annealing zone (Amaya et al. 2017). Apatite fission-track data from Jurassic and older basement rocks distinguish between two different cooling pulses, the first between 15 and 25 Ma and the second between 3.5 and 10 Ma, which suggest differences in timing and rates of cooling within discrete fault blocks. Shagam et al. (1984) also proposed an earliest Oligocene (~34 Ma) exhumation phase based on an interpretation that sedimentary rocks in the Eastern Piedmont were derived from erosion of the SM.

For the southern VA, 40Ar/39Ar biotite ages ranging between 185 and 480 Ma are older than those reported from SNSM and SM (van der Lelij et al. 2016). Zircon fission-track ages fall between 54 and 120 Ma, with older ages located closer to the main faults. For the apatite fission -track ages, it is possible to discriminate between four different age groups at (1) 1.8–3 Ma, (2) 4–9 Ma, (3) 15–30 Ma, and (4) 57–145 Ma (Miller 1962; Shagam 1975; Shagam et al. 1984; Kohn et al. 1984; Villagómez et al. 2011; van der Lelij et al. 2016).

4.3 Main Plate Tectonic Events

During the middle Jurassic ( ̴170 Ma), Pangea breakup resulted in the formation of Gondwana and Laurasia, leaving a series of rifts and discontinuities in the crust. The reactivation of preexisting tectonic discontinuities of different age and origin, mostly developed along the former passive margin of the South American plate, has played an important role in the evolution of the circum-MCB mountain belts. Figure 13.7 summarizes the main interactions between the relevant plates and accretion of different tectonic blocks. Preexisting tectonic discontinuities correspond to rift structures generated in the Jurassic during separation between the North American and South American plates (Fig. 13.7a). From Aptian to Albian time, the Romeral terrane was pushed by the Farallon plate toward South America (Cediel et al. 2003), and at this time the first appearance of a system of arcs (Mérida Arch) occurs, which correspond with the first tectonic inversion of Jurassic rifts. For this period thermochronologic ages closer to ~150 Ma correspond with this inversion.

From late Cretaceous to Paleocene (70–65 Ma), exhumation of the SM commenced together with slow exhumation of the proto-Eastern Cordillera (Amaya et al. 2017). This is possibly related to oblique subduction and collision of the Dagua-Piñón (DA) and San Jacinto terranes (SJ; Cediel et al. 2003), which are remnants of the Caribbean plate. From the Paleocene to Eocene (65–56 Ma), fast exhumation occurred in the Colombian Central Cordillera as well as metamorphic deformation along the SNSM, and the first reactivation of the Oca fault is also recorded during this period.

From Eocene to early Oligocene (56–28 Ma), the following events are discerned: oblique subduction and accretion of the Gorgona terrane (GOR; Cediel et al. 2003), Eocene magmatism along the MCB, magmatism along the Oca Fault, and emplacement of the Guajira-Falcón (GU-FA) and Caribbean Mountain (CAM) terranes. These events are responsible for rapid exhumation throughout SNSM and SM, but at this time exhumation was moderate in the SP and slow in the VA (Fig. 13.7a).

From late Oligocene to early Miocene (28–20 Ma), thermochronological data reflect collision of the Sinú terrane (SN; Fig. 9e) and frontal obduction of the Cañas Gordas (CG) terrane, a remnant of the Farallon Plate. Collision of the Baudó (BAU) terrane and subduction of the Nazca plate occur during this period, resulting in the onset of collision between the Panamá arc and South America.

From middle Miocene to late Miocene (14–8 Ma), collision between the Panamá arc and South American plates terminates, the main lineation of the Boconó fault is reactivated, and deviation of the Orinoco River occurs (Hoorn et al. 1995). The main exhumation of the VA is also achieved during this period. Further, rapid exhumation on both sides of the Bucaramanga fault is reported from different areas.

From Pliocene to Pleistocene (5–1 Ma), local exhumation in several areas (SM, SNSM, SP, and VA) is triggered as a consequence of the northwest migration of the MCB .

5 Conclusions

Despite the caveats outlined above, some exhumation episodes have been significant enough to be recorded throughout the different mountain belts circumventing the MCB as follows:

  1. 1.

    Reactivation of normal faults during Mesozoic rifting between ~150 and 120 Ma across the SP, the Trujillo block in the VA, the Oca fault in the SNSM, and the northern SM.

  2. 2.

    Tectono-thermal pulses between 65 and 40 Ma are recorded in the SNSM, SP, and VA. However, with the available thermochronology dataset, these events are less evident within the SM. These pulses were triggered by oblique subduction and collision of the Dagua-Piñón (DA) and San Jacinto (SJ) terranes.

  3. 3.

    Relatively fast cooling from 35 to 16 Ma is recorded in the SNSM, SM, SP, and VA. This cooling was generated by accretion of Gorgona terrane and onset of subduction of the Caribbean plate below South America. During this period, tectonic inversion of the Guajira-Falcón terrane also took place.

Continuous cooling pulses between 15 and 2 Ma are commonly recorded across the different mountain belts, including some faults that were locally reactivated during last 4–2 Ma. This continuous pulse can be related to onset of collision between the Panamá arc and South America.