Onset of fault reactivation in the Eastern Cordillera of Colombia and 1 proximal Llanos basin; response to Caribbean South American 2 convergence in early Palaeogene time 3 4 German Bayona (*), Corporacin Geolgica ARES, Bogot, Colombia, Calle 5

44A N. 53-96 (gbayona@cgares.org) 6 Agustin Cardona, Universidad Nacional de Colombia and Corporacin 7

Geolgica ARES 8 Carlos Jaramillo, Smithsonian Tropical Research Institute and Corporacin 9

Geolgica ARES 10 Andres Mora, Instituto Colombiano del Petrleo, 11 Camilo Montes, Universidad de los Andes and Corporacin Geolgica ARES, 12 Victor Caballero, Instituto Colombiano del Petrleo, 13 Hernando Mahecha, Corporacin Geolgica ARES, 14 Felipe Lamus, Corporacin Geolgica ARES, 15 Omar Montenegro, Corporacin Geolgica ARES, 16 Giovanny Jimenez, Corporacin Geolgica ARES, 17 Andres Mesa, Hocol S.A., Bogot, Colombia 18 Victor Valencia, School of Earth and Environmental Sciences, Washington 19 State University, Pullman, WA 99164-2812 85721, USA 20 (*) Corresponding author: German Bayona (gbayona@cgares.org) 21 Number of words: 7322 22 Number of references: 78 23 Number of Tables: 2 24 Number of Figures: 9 25 Number of Appendix: 4 26 27

Abstract 28 The inversion of Mesozoic extensional structures in the northern Andes 29

controlled the location of syn-orogenic successions and dispersal of detritus 30 since latest Maastrichtian time. Our results are supported by detailed 31

geological mapping, integrated provenance (petrography, heavy minerals, 32 geochronology) analysis and chronostratigraphic correlation (palynological 33

and geochronology data) of 13 areas with Palaeogene strata across the 34 central segment of Eastern Cordillera. Thickness trends, spatial and temporal 35

variations of sedimentation rates and provenance signatures indicate that 36 mechanisms driving location of uplifts and tectonic subsidence vary among 37 syn-orogenic depocenters. In the late Maastrichtian-mid Palaeocene time, 38 crustal tilting of the Central Cordillera favored reverse reactivation of the 39

western border of the former extensional Cretaceous basin; block-tilting of the 40 hanging-wall separated two depocenters: a western depocenter (in the 41

Magdalena Valley) and an eastern depocenter (presently along the axial zone 42 of the Eastern Cordillera, Llanos foothills and Llanos basin) of the reactivated 43 fault. In Late Palaeocene to Early Eocene time, as eastern subduction of the 44

Caribbean plate and intraplate magmatic advanced eastward, reactivation of 45 older structures migrated eastward up to the Llanos basin and disrupted the 46 eastern depocenter. These three depocenters were separated by two low-47 amplitude uplifts that exposed dominantly Cretaceous sedimentary cover. 48

Additionally, syn-orogenic detrital sediments supplied from the eastward tilted 49 Central Cordillera reached areas of the axial domain of the Eastern Cordillera, 50

whereas unstable metamorphic and sedimentary fragments of the 51 easternmost depocenter were supplied by basement-cored uplifts with 52

Cretaceous and Paleozoic sedimentary cover reported in the southern Llanos 53 basin. This tectonic configuration of low-amplitude uplifts separating intraplate 54 syn-orogenic depocenters and intraplate magmatic activity in Palaeocene time 55

was primary controlled by subduction of the Caribbean plate. 56 57

58 Keywords: Inversion tectonics, onset of deformation, Palaeogene, Eastern 59 Cordillera, 60 61 Reactivation of pre-existing structures is a major response to changes in 62 tectonic regime; therefore, determination of fault reactivation timing is 63 fundamental for identifying shift in tectonic processes. Positive inversion 64 structures involve the reactivation of a former normal fault (Hayward & 65 Graham 1989; Coward 1994; Sibson 1995; Bayona & Lawton, 2003). In the 66 northern Andes, along its eastern segment, major uplifts have occurred along 67 inversion structures (e. g. Cooper et al. 1995; Colleta et al. 1997; Corts et al. 68 2006; Mora et al. 2008a). Several factors may control whether an older 69 structure may be reactivated: (a) fault and basin geometries (Butler 1989; 70 Buchanan & McClay 1991, 1992; Bayona & Thomas, 2003), (b) potential of 71 fault reactivation (e. g. pore pressure, friction, cohesion, fault gouge material; 72 Sibson 1985, 1995; Etheridge, 1986), and (c) re-organization of stresses 73 along the collided margin (Kluth & Coney 1981; Ziegler et al. 1998; Marshak 74 et al. 2000). 75 76 Inversion structures fragments former basin geometry and drives the evolution 77 of compartmented syn-orogenic basins. As fault reactivation develops, 78 tectonic fragmentation of the drainage system, depositional patterns and 79 climate conditions begin to differ between localized depocenters (Sobel et al. 80 2003; Sobel & Strecker 2003; Bayona & Thomas, 2003; Hilley & Strecker 81 2005; Strecker et al. 2007, 2009, 2011). Therefore, depositional 82 environments, stratigraphic thicknesses and provenance signatures begin to 83 vary among depocenters (Bayona & Lawton, 2003; Hain et al. 2011). 84 Presently, the Magdalena Valley basin and the Llanos foreland basin of 85 Colombia are two basins with drainage, climate and depositional systems 86 totally disrupted by the rapid vertical uplift of the Eastern Cordillera of 87 Colombia since late Miocene-Pliocene time (Fig. 1a; see Mora et al., 2008b 88 for review of the youngest events of Eastern Cordillera uplift); however, upper 89 Cretaceous rocks of the Magdalena Valley, Eastern Cordillera and Llanos 90 basin accumulated within the same basin and represent a single depositional 91 profile of a shallow platform in a passive-margin setting (Cooper et al., 1995; 92 Villamil, 1999; Cediel et al. 2003) 93

94 On the other hand, modeling of undisrupted basins, such as foreland basins 95 (DeCelles & Giles 1996), has allowed to predict patterns of deformation (i.e. 96 the advance of orogenic front and flexural subsidence), sediment routing 97 (provenance signature) and accumulation patterns (stratal geometry and 98 threedimensional lithofacies distribution) (Beaumont 1981; Jordan 1981; 99 Heller et al. 1988; Flemings & Jordan 1989; Crampton & Allen 1995; DeCelles 100 & Giles 1996; Burbank et al. 1996; Sinclair 1997; DeCelles & Horton 2003). In 101 orogens with multi-phase deformation, such as the northern Andes (Bayona et 102 al. 2008), comprehensive basin stratigraphy and provenance analyses 103 become the key to document the timing and sense of fault movement along 104 reactivated structures. 105 106 The timing of when a basin become positive feature is relevant for 107 determination of cessation for burial processes of source rocks and formation 108 of early structures in a petroleum system. Definition of the onset of positive 109 inversion in the northern Andes has been controversial, and the timing range 110 from MaastrichtianPalaeocene (Sarmiento-Rojas 2001; Corts 2004; Montes 111 et al. 2005; Bayona et al. 2008; Moretti et al., 2010; Parra et al., 2012), 112 middle-late Eocene (Toro et al. 2004; Gmez et al. 2005a; Parra et al. 2009; 113 Horton et al. 2010; Saylor et al. 2011; 2012), late Eocene (Mora et al. 2010), 114 and late Oligocene-Miocene (Villamil 1999; Cooper et al. 1995). In order to 115 address this problem of the timing of fault reactivation in a multi-phase 116 orogeny, like the Eastern Cordillera, we integrated the analysis of depositional 117 environments, thickness variations, and provenance signatures of 118 Maastrichtian-lower Eocene strata that are presently compartmentalized by 119 several structures within the double-verging Eastern Cordillera of Colombia 120 (Fig. 1). Our study specifically examines the Maastrichtian-early Palaeogene 121 time interval to document possible intraplate deformation events coeval with 122 the oblique convergence of the Caribbean plate or collision of oceanic arcs 123 with the South American plate (see review in Bayona et al. 2012). The 124 approach used in this study shows that abrupt changes in provenance 125 signatures and sedimentation rates can be used to reconstruct the early 126 phases of deformation in multi-phase orogens when low-temperature 127 thermochronometry commonly lacks resolution in defining low-amplitude 128 deformation. 129 130 131 Regional setting 132 Three major mountain belts are the result of the complex interaction of the 133 Nazca, Caribbean and South America plates since the Late Cretaceous: 134 Western Cordillera (WC), Central Cordillera (CC) and the double-verging 135 Eastern Cordillera (EC) (Fig. 1a & b). The Eastern Cordillera is interpreted as 136 a wide Cretaceous extensional basin formed during at least two stretching 137 events (Sarmiento-Rojas et al. 2006), and tectonically inverted during the 138 Cenozoic (Colleta et al. 1990; Dengo & Covey 1993; Cooper et al. 1995; Mora 139 et al. 2006). The geometry of the double-verging Eastern Cordillera changes 140 northward from a narrow belt in the southern domain that broadens in its 141 central segment where the Cordillera is bounded by east- and west-verging 142

thrust belts (Fig. 1b). Further to the north, the Eastern Cordillera bifurcates 143 into the Santander Massif and the Mrida Andes (Fig. 1a). 144 The areas selected for this study are located in the three structural domains of 145 the central segment of the Eastern Cordillera (Fig. 1c). Its western domain 146 includes a west-verging thrust and fold belt that places Cretaceous and 147 Palaeogene strata in contact with Cenozoic strata and buried structures of the 148 southern Middle Magdalena Valley basin. In the frontal splay of the west-149 verging system is the north-south trending Guaduas Syncline (Corts et al. 150 2006). The axial domain consists of tight anticlines and broad synclines 151 related to double-verging faults (McLaughlin & Arce 1975; Fajardo-Pea 152 1998; Toro et al. 2004; Bayona et al. 2008). This domain broadens northward 153 of Bogota City, and the east-verging Soapaga-Pesca-Machet Fault System 154 divides the axial domain into western axial and eastern axial domains (Fig. 155 1c). The eastern domain includes east-verging faults placing Cretaceous and 156 Cenozoic strata over Cenozoic strata of the Llanos foreland basin. Basement 157 rocks of the Quetame Massif are exposed in the southern segment of the 158 eastern domain (Fig. 1c). The northeast-trending Medina Syncline is part of 159 the frontal splay of the east-verging fault system (Parra et al. 2009). Reverse 160 structures in these three domains of the Eastern Cordillera have been 161 interpreted as an inversion of former normal faults (Cooper et al. 1995; 162 Branquet et al. 2002; Restrepo-Pace et al. 2004; Corts et al. 2006; Mora et 163 al. 2006, 2008a; Kammer & Sanchez 2006). 164 Interpretations of basin geometry for early Palaeogene vary from a single and 165 continuous foreland basin (Cooper et al. 1995; Villamil 1999; Gmez et al. 166 2005a; Horton et al. 2010; Saylor et al. 2011, 2012), a continuous negative 167 flexural basin with apparent absence of bounding thrusts (Pindell et al. 2005) 168 to a foreland basin disrupted by uplifts (Bayona et al., 2008). The location of 169 intraplate uplifts has been proposed in the western border of the basin 170 (Bayona et al. 2003; Restrepo-Pace et al. 2004; Corts et al. 2006; Moretti et 171 al. 2010), along the axial zone of the basin (Fabre 1981, 1987; Sarmiento-172 Rojas 2001; Pardo 2004), or at both borders of the basin (Fajardo-Pea 1998; 173 Villamil 1999). Geodynamic modeling of Palaeogene strata encompassing 174 the axial zone of Eastern Cordillera to the Llanos basin indicate eastward 175 migration of tectonic loads within the restored Eastern Cordillera to explain the 176 Maastrichtian-Palaeogene geometry of the syn-orogenic basin (Bayona et al. 177 2008). 178 Early Palaeogene (45-65 Ma) magmatic activity has been reported along the 179 northern Andean continental margin (Central Cordillera-Santa Marta Massif in 180 Fig. 1a & b; (Aspden et al. 1987; Cardona et al. 2011), and volcanic detrital 181 zircons have been recognized in Paleocene-lower Eocene sandstones as far 182 as 400 km from the magmatic arc (Saylor et al. 2011, 2012; Bayona et al. 183 2012). Marginal and intraplate magmatism during early Palaeogene time is 184 related to convergence and shallow subduction of the Caribbean crust 185 beneath the South America plate, and the abrupt termination of volcanism in 186 middle Eocene time to a change to a transpressive non-magmatic margin 187 (Bayona et al. 2012). 188 189

Maastrichtian-lower Eocene stratigraphy and depositional environments 190 In this section, we summarize lithological associations and depositional 191 environment interpretations proposed for Maastrichtian-lower Eocene units in 192 the three structural domains of the central segment of the Eastern Cordillera, 193 as these units record the regional shift from marine to continental depositional 194 environments as response to the accretion of oceanic terranes west of the 195 Romeral fault system (Fig. 1) (e.g. Etayo-Serna et al. 1983; McCourt et al. 196 1984). 197 Maastrichtian-lower Eocene stratigraphic units of the western domain have 198 been defined in the western flank of the Guaduas Syncline (Fig. 1c), and 199 include from base to top; the Umir, Cimarrona, Seca and Hoyon Formations 200 (Fig. 2). The lower two units change laterally and vertically from calcareous 201 dark-gray mudstones with abundant fossil content (bivalves, foraminifera), to 202 cross-bedded conglomerates and fine-grained sandstones (Gmez & Pedraza 203 1994); conglomerate beds of the Cimarrona Formation changes eastward to 204 limestone beds in the eastern flank (Gmez et al. 2003). The overlying Seca 205 Formation includes dark-colored mudstones with thin coal interbeds at the 206 base changing upsection to varicolored massive mudstones with interbeds of 207 fine to medium grained sandstones toward the top (De Porta 1966). The 208 Hoyon Formation comprises interbeds from a meter to decameter scale of 209 mudstones, sandstones and conglomerates, and it is divided into three 210 informal units. The lower Hoyon unit is a conglomerate interval with upward-211 coarsening successions, has transitional contacts with sandstones and 212 mudstones of the underlying Seca Formation and with mudstones of the 213 overlying middle Hoyon unit; the latter units comprises dominantly massive 214 red mudstones. In contrast, the upper Hoyon unit is a conglomerate interval 215 resting unconformably upon the middle Hoyon unit. Both in lower and upper 216 Hoyon units, horizontal bedding dominates over cross bedding as internal 217 structures, and an E-NE direction of transport is documented by palaeocurrent 218 measurements (Gmez et al. 2003). Gmez et al. (2003) interpreted the 219 Maastrichtian-lower Eocene succession as a regression from a shallow-220 marine environment with fan deltas (Umir & Cimarrona Formations), changing 221 to coastal and alluvial plains (Seca Formation), and ending with fluvial and 222 alluvial fan deposits (Hoyon Formation). 223 In the axial domain, Maastrichtian-lower Palaeocene strata include the 224 Guaduas Formation, overlain by Palaeocene-lower Eocene Cacho and 225 Bogot Formations to the west of the Pesca-Machet Fault, or by the Lower 226 and Upper Socha Formations to the east of the Pesca-Machet Fault (Fig. 2). 227 The Guaduas Formation rests conformably over fine to medium-grained 228 sandstones of the Guadalupe Formation with cross-beds and abundant 229 ichnofossils (Perez & Salazar, 1978; Fabre, 1981). The lower half of the 230 Guaduas Formation includes fine-grained strata with abundant coal seams, 231 whereas the upper half includes laminated mudstones, massive light-colored 232 mudstones and sandstones interbeds (Sarmiento 1992). The Cacho/Lower 233 Socha Formations are cross-bedded, fine to coarse-grained upward-fining 234 quartzarenite sandstones, locally conglomeratic; these units rest 235 disconformably upon light-colored mudstones of the Guaduas Formation 236 (Pardo 2004; Bayona et al. 2010). In the western axial domain, the Bogota 237

Formation overlies conformably the Cacho Formation and consists dominantly 238 of massive multi-colored sandy mudstone with isolated upward-fining and 239 coarsening massive to cross-bedded sandstones (Bayona et al. 2010). In the 240 eastern axial domain, the Upper Socha Formation overlies conformably the 241 Lower Socha Formation and it changes upsection from bioturbated dark-gray 242 organic-rich claystone and mudstone with thin coal seams to massive light-243 colored sandy mudstone with isolated upward-fining and coarsening 244 sandstones with cross beds, wavy lamination and ripple cross-lamination 245 (Pardo 2004). Coarse-grained to conglomeratic sandstones of the 246 Regadera/Picacho formations disconformably overlie the Bogota/Upper 247 Socha formations. Sarmiento (1992) interpreted the regression from shallow-248 marine to alluvial plain conditions for the Guaduas Formation, whereas 249 Bayona et al. (2010), Pardo (2004) and Saylor et al. (2011) interpreted 250 overlying units as deposition in fluvial systems varying from braided, 251 meandering to anastomosing system. 252 In the eastern domain, as well as in the proximal Llanos basin, the Guaduas 253 Formation is very thin or absent, and the Barco and Cuervos Formations rest 254 unconformably upon upper Cretaceous units (Cooper et al. 1995; Bayona et 255 al. 2008) (Fig. 2). The succession described below corresponds to the 256 western flank of the Medina Syncline. Maastrichtian strata consist of 257 coarsening-upward succession from medium-grained to conglomeratic 258 sandstones with cross-beds and reworked bivalves, oysters and corals of the 259 Guadalupe Formation (Guerrero & Sarmiento 1996). The Barco Formation 260 comprises fine-to medium-grained quartzose sandstones and is overlain by 261 interbeds of upward-fining sandstone and mudstone with bidirectional cross-262 bedded and bioturbated heterolithic laminated sandstone of the lower 263 Cuervos Formation. The middle and upper Cuervos Formation resembles the 264 lithological association of the Upper Socha Formation. Guerrero & Sarmiento 265 (1996) interpreted Maastrichtian strata as shallow-marine deposits, whereas 266 Cazier et al. (1995), Reyes (1996) and Bayona et al. (2008) interpreted the 267 Barco-Cuervos succession as coastal plains with tidal influence. The Mirador 268 Formation is a coarse-grained sandstone unit of lower-to-middle Eocene age 269 (Bayona et al. 2008; Parra et al. 2009) that disconformably overlies the 270 Cuervos Formation. 271 272 Selection of study areas and methods 273 In order to determine spatial and temporal changes in basin geometry for a 274 given time interval, it is crucial to consider the chronostratigraphic correlation 275 of lithological units described above at different structural domains. The 276 survey of Maastrichtian-lower Eocene strata included three areas in the 277 Guaduas Syncline (San Juan de Ro Seco East and West, and Guaduero), six 278 areas in the western axial zone (Usme, Guatavita, Machet, Checua, Tunja 279 and Paipa), five areas in the eastern axial zone (Umbita, San Antonio, 280 Rondn, Pesca and Paz de Ro) and two areas in the eastern domain 281 (Cusiana and Medina) (Fig. 1b) 282

The thickness of Maastrichtian-lower Eocene lithostratigraphic units described 283 above was measured in synclinal structures and within areas with minimum 284 structural complexity. Lithostratigraphic units were mapped in both flanks of 285 each syncline structure, and the lateral variation of thickness was calculated 286 using local structural cross sections. Samples for palynological and 287 provenance analyses (petrography, heavy mineral and detrital U/Pb zircon 288 geochronology) were collected for each measured stratigraphic section. 289 Palynology and detrital volcanic zircons supply the age control for 290 Maastrichtian-lower Eocene units. Palynology methods are described in 291 Appendix 1, and the results of 381 palynological samples are summarized in 292 Table 1. Calculation of maximum age of deposition based on U-Pb ages in 293 volcanic zircons follows the results presented in Bayona et al. (2012). 294 Sandstone petrography (123 samples), heavy minerals (46 samples) and U-295 Pb detrital zircon geochronology (25 samples) analyses were carried out in 296 Palaeocene-lower Eocene sandstones in order to compare composition, 297 heavy mineral assemblage and detrital zircons age population among 298 structural domains. Descriptions of analytical methods for each analysis are 299 in Appendix 1. Heavy minerals were grouped into ultrastable (zircon, rutile, 300 turmaline), stable (apatite, garnet), unstable (amphibole, epidote, pyroxene, 301 chlorite, talc and others) and muscovite. Zircon crystallization age sources 302 were divided into three major provinces; 65-360 Ma from the Central 303 Cordillera; 360-1300 Ma from basement rocks of the Eastern Cordillera and 304 Santander massif; zircons supplied from basement rocks in the Guyana 305 craton have Neoproterozoic ages (600-650 Ma) and populations older than 306 1300 Ma (see Horton et al. 2010a, b) for a detailed discussion of zircon 307 crystallization ages in the northern Andes). The Jurassic to Santonian 308 sedimentary cover of the Eastern Cordillera includes a mixture of the last two 309 provinces. 310 311 Thickness variation and age control of Maastrichtian- lower Eocene 312 strata 313 This section presents the most significant trends of stratigraphic thickness 314 variations for studied stratigraphic units within structural domains. Figure 2a 315 illustrates the measured stratigraphic thickness for each unit in each area and 316 the position of samples with volcanic detrital zircons. Palynological results 317 (summarized in Table 1) and calculation of maximum depositional age were 318 used to construct the chronostratigraphic correlation presented in Figure 2b. 319 In the western domain, the stratigraphic thickness of the Seca Formation 320 increases eastward and northward (Table 1, Fig. 2a). In the area of 321 Guaduero, the thickness ranges from 490 m to more than 700 m, whereas in 322 the San Juan de Rio Seco area the thickness ranges from 400 to 470 m. The 323 lower and middle intervals of the Hoyon Formation are completely recorded in 324 the eastern flank of the San Juan de Rio Seco area, and are partly eroded by 325 the unconformity in the western flank of the San Juan de Rio Seco area and 326 completely eroded in the Guaduero area (Fig. 2a). For the Seca Formation, a 327 Maastrichtian palynological age at the base and detrital volcanic zircon ages 328

at the top (61.9 2 Ma) constrain a Maastrichtian-lower Palaeocene age 329 (Table 1). The lower Hoyon unit yielded only two zircons with Paleocene age, 330 being 62 Ma the youngest zircon, whereas the maximum age of deposition for 331 the middle Hoyon unit is 56.3 1.6 Ma. The upper Hoyon yielded middle 332 Eocene to lower Oligocene palynological ages, and single zircon age of 55 333 Ma. 334 In the western axial domain, the Guaduas Formation shows its maximum 335 thickness in the central area of Checua (1098 m), and decreases to less than 336 600 m in northward, southward and eastward directions (Table 1, Fig. 2a). 337 Lower and middle intervals of the Guaduas Formation range in age from 338 uppermost Campanian to Maastrichtian, and only the uppermost strata 339 yielded a lower Palaeocene age (Fig. 2b). The Cacho Formation yielded four 340 detrital zircon ages in the range of 61 to 64 Ma in the Checua area, yielding 341 an early Palaeocene age of deposition (Fig. 2b). Thickness variations are 342 similar to the Guaduas Formation, with a maximum thickness of 245 m in the 343 Checua area. The stratigraphic thickness of the Bogota Formation is very 344 variable with a maximum of 1415 m in Usme, and a minimum of 169 m in 345 Tunja; regionally, its thickness decreases northward. Detrital volcanic zircon 346 ages and palynology indicate middle Palaeocene to lower Eocene age for this 347 unit. The middle Eocene Regadera Formation overlies disconformably strata 348 of the Bogota Formation (Bayona et al. 2010) 349 In the eastern axial domain, the thickness of the Guaduas Formation varies 350 from 290 to 609 m (Table 1, Fig. 2a) and the upper interval of the Guaduas 351 Formation is Maastrichtian in age (Fig. 2b). Only in Umbita, an eastward 352 decrease of thickness is documented from 525 to 350 m. The stratigraphic 353 thickness of the Lower Socha Formation decreases northward from 257 m in 354 Umbita to a range between 90-138 m in Paz de Ro (Pardo 2004; Saylor et al. 355 2011). Similarly, the thickness of the Upper Socha decreases from 450-460 356 m in Umbita and San Antonio to 255-280 m in Pesca and Paz de Ro. Saylor 357 et al. (2011) documents an abrupt change from 227 to 344 m of Upper Socha 358 Formation over a distance of less than 3 km. The Lower Socha Formation 359 unit is poorly dated as lower to middle Palaeocene in the Paz de Ro area 360 (Pardo 2004)(Fig. 1b), whereas the Upper Socha Formation has palynological 361 and mineral-age constraints indicating late Palaeocene-early Eocene age 362 (Table 1, Fig. 2b). The overlying Picacho Formation has been assigned an 363 early to middle Eocene age (Pardo 2004). 364 The Guaduas Formation in the Eastern domain is less than 50 m thick and 365 yielded only a Maastrichtian age (Parra et al. 2009); coeval strata in the 366 Cusiana area are not reported (Fig. 2a). This unit overlies upper Campanian 367 to lower Maastrichtian strata of the Guadalupe Formation (Guerrero & 368 Sarmiento 1996). The sandy Barco Formation is early to middle Palaeocene 369 in age, whereas the Cuervos Formation has palynological and mineral-age 370 data that indicate a late Palaeocene-early Eocene age (Bayona et al. 2008; 371 Parra et al. 2009) (Table 1, Fig. 2). The thickness of both the Barco and 372 Cuervos units decreases eastward, as well as the thickness of the lower to 373 middle Eocene Mirador Formation (Parra et al. 2009). 374 375

Chronostratigraphic correlation 376 Because stratigraphic units are diachronous (Fig. 2b), we established the 377 following boundaries at 65, 59, 55 and 49 Ma for chronostratigraphic 378 correlation of strata. The following criteria were used to determine the 379 position of those times for each stratigraphic section (Fig. 3) and calculation of 380 sedimentation rates presented in Table 2: 381

(1) Palynological data and maximum ages of deposition (Table 1) were 382 used as the most reliable information for age constraints. 383

(2) When the stratigraphic unit does not have an age, we used the age 384 determined for the same unit in an area within the same structural 385 domain. 386

(3) Sandy units (i.e. amalgamated channels in Cacho Formation) are 387 considered to have lower rates of accumulation than muddy units (i.e. 388 flood plains in the Bogota Formation). 389

Campanian-lower Maastrichtian conglomerate-bearing strata are at the 390 western and eastern domains, while fine-grained strata of the Guaduas 391 Formation accumulated over sandstones of the Guadalupe Formation in the 392 axial zone (Fig. 2b). Onset of fine-grained deposition of the Guaduas and 393 Seca formations is older in the western domain than in the axial domain, in 394 this later domain, onset of deposition is nearly coeval (Fig. 2b). The lacuna 395 (non deposition + erosion) overlying Seca and Guaduas strata increases 396 eastward among domains, and northward in the four domains. Onset of fine-397 grained deposition of the Bogot/Cuervos strata is coeval in the eastern and 398 axial domains, whereas in the western domain onset of fine-grained 399 deposition of the middle Hoyn Formation is younger than in the axial domain. 400 The Palaeocene-Eocene boundary is well constrained by palynology and 401 detrital zircon ages in the four domains (Fig. 2b), showing a correlation, from 402 west to east, of middle Hoyn/Bogota/Upper Socha/Cuervos Formations (Fig. 403 2b). Lower Eocene strata in the eastern domain and eastern axial domain 404 include sandstones of the Mirador/Picacho Formations, whereas in the 405 western axial domain corresponds to fine-grained strata of the upper Bogota 406 Formation. In the western domain, the middle Hoyon Formation in the south 407 and strata mapped as upper Seca Formation in the north have an inferred 408 mid-Palaeocene-lower Eocene age (Fig. 2b). 409 410 Provenance 411 The description of provenance signatures (sandstone petrography, heavy 412 minerals and U-Pb detrital zircon geochronology) follows the 413 chronostratigraphic intervals defined above, to understand dispersal of 414 terrigenous detritus by interval of time (see Figures 4 to 6 and raw data in 415 Appendixes 2 to 4, respectively). Glauconite was identified either as 416 authigenic (e.g. clean faces filling pores Fig. 4e) or as reworked fragment 417 (clay coating and rounded fragments Fig. 4f and 4g). 418

Upper Campanian sandstones (>70 Ma). Quartzarenite sandstones show 419 authigenic glauconite, as identified at the top of the Guadalupe Formation in 420 the Umbita (Fig. 4e) and Checua areas (e.g. Sarmiento 1992). Heavy mineral 421 assemblage of Guadalupe Formation in the Eastern domain is dominantly 422 ultrastable (Fig. 5a), showing 11.7% of titanite (Fig.5b). 423 Detrital zircons for Campanian strata from the axial and eastern domain 424 reveals a dominant Proterozoic population, with several age peaks between 425 ca. 0.96 Ga to 1.8 Ga (see also Cretaceous rock data in Horton et al. 2010b 426 and Saylor et al. 2011), whereas one sample of the western domain reveals 427 additionally a younger population 0.9 Ga 446 (Saylor et al. 2011). 447 Lower-middle Palaeocene sandstones (59-65 Ma). Provenance signature 448 defined by sandstone petrography of the eastern and western domains varies 449 in the content of the amount of quartz and lithic fragments (craton interior and 450 transitional recycled orogen, respectively), whereas provenance signature of 451 sandstones of the two axial domains overlap in the field of recycled quartzose 452 orogen (Fig.4b). Total lithic fragments content increases westward (Figs. 4b, 453 4h, 4i and 4j). Metamorphic and plutonic rock fragments increase with 454 respect to Maastrichtian sandstones; these fragments, volcanic lithic 455 fragments and feldspars are more abundant in the western domain and 456 eastern axial domain (Fig.4b). Reworked glauconite fragments were identified 457 in the Checua and Usme areas of the western axial domain. Unstable and 458 stable heavy minerals assemblages are more abundant in the western 459 domain and in Usme area, whereas in the other areas of the axial zone and 460 the eastern domain, ultrastable minerals dominate (Fig. 5a). Heavy minerals 461 concentrate in laminae in the Umbita area, and in Pesca and Paipa muscovite 462 are very common. Garnets fragments were identified in the Umbita area 463 (eastern axial domain, Fig. 5d), while pyroxene fragments were identified in 464

San Juan de Ro Seco (western domain, Fig. 5e) and Umbita (eastern axial 465 domain, Fig. 5f). 466 Cretaceous detrital zircons together with more limited Triassic (248-230 Ma), 467 Early Palaeozoic and Proterozoic detrital zircons dominate in the western 468 domain and the Usme area (Figs 6 and 7), as well as in the Paz de Ro area 469 (eastern axial domain, Fig. 6; Saylor et al. 2011). Although Phanerozoic zircon 470 populations are similar along the western axial zone, there is a northward 471 increase of Proterozoic (0.8-2 Ga) age populations (Fig. 7). In contrast, one 472 sample from the eastern domain shows older zircon age peaks of 600, 900 473 and 1500 Ma (Fig. 6). 474 Upper Palaeocene sandstones (55-59 Ma). Provenance signature of the 475 four studied domains plot near the boundary of quartzose and transitional 476 recycled orogens, being more lithic and feldspar-rich in the western domain 477 (Fig. 4c). The content of total lithic fragments increases slightly with respect to 478 lower-middle Palaeocene sandstones. Metamorphic-plutonic rock fragments, 479 as well as feldspar fragments, are more abundant in western, eastern and 480 western axial domains, whereas sedimentary rock fragments dominate in the 481 eastern axial domain (Fig. 4c). Reworked glauconite was identified in the 482 Usme (western axial domain) and Umbita (eastern axial domain) areas. 483 Unstable heavy minerals were found in the Usme (epidote), Rondn (chlorite) 484 and Medina (hornblende, Fig. 5g) areas, and in the former two areas the 485 unstable population is an important constituent (Fig. 5a). In the other areas of 486 the western, axial and eastern domains the ultrastable assemblage dominates 487 (Fig. 5a). Micas continue as important constituents in northern areas (Paipa-488 Pesca) of the axial zone. Garnet was identified in Guaduero (western domain) 489 and Medina (eastern domain) areas. 490 The detrital zircon age populations from the western domain are dominantly 491 Mesozoic (90-250 Ma; Fig. 6), whereas samples from the western axial and 492 eastern axial domains show an increase in Phanerozoic detrital zircon ages 493 (60-300 Ma) and decrease in Proterozoic (>1000 Ma) zircon ages when 494 compared with the lower-middle Palaeocene samples (Fig. 6). In contrast, the 495 eastern domain shows a bimodal distribution with minor population of 50-65 496 Ma zircons and a prominent Proterozoic population (> 1.5 Ga) (Fig. 6). 497 Lower Eocene sandstones (49-55 Ma). Provenance signature of the Mirador 498 (eastern domain) and Picacho (eastern axial domain) sandstones plot in the 499 upper boundary of quartzose recycled orogen, whereas the other sandstones 500 of the western axial and eastern axial domains (Bogota/Upper Socha 501 Formations) plot near the boundary of quartzose and transitional recycled 502 orogens (Fig. 4d). The former two units are above the unconformity, while the 503 latter two units are below the unconformity. The content of metamorphic-504 plutonic and sedimentary rock fragments is proportional in these sandstones, 505 with an increase in the content of the latter fragments in the Picacho 506 sandstones (Fig. 4d). The highest content of feldspar fragments is in the 507 western axial domain. Reworked glauconite was identified in the Usme 508 (western axial domain) and Umbita (eastern axial domain) areas. Stable and 509 unstable heavy mineral assemblages are common both in the Usme (garnet 510 in Fig. 5h; epidote and hornblende) and Umbita areas (garnet, titanite, epidote 511

and hornblende), with a dominance of micas in the Rondn area (Figs. 5a & 512 i). The ultrastable assemblage becomes dominant in quartzose sandstones of 513 the Picacho and Mirador Formations, above the unconformity (Fig. 5a); 514 however, in the Medina area (eastern domain) apatite, hornblende, pyroxene 515 and titanite are present between 2-5%. 516 Detrital zircon age populations from the western axial domain include mostly < 517 290 Ma ages with significant 58-70 Ma and 220-290 Ma age peaks; whereas 518 in the eastern axial domain, similar Phanerozoic detrital zircon age population 519 alternate with more abundant Proterozoic (>1000 Ma) zircon ages with peaks 520 as old as ca. 1.8 Ga (Fig. 6). Samples from the Upper Socha Formation 521 reported in Saylor et al. (2011) include zircons in the range of 53-57 Ma. In 522 contrast, detrital zircon population reported for one sample of the Mirador 523 Formation in the eastern domain (Horton et al. 2010a & b) is characterized by 524 zircons older than 1.4 Ga. 525 526 Spatial and temporal changes in sedimentation rates 527 Division of stratigraphic units (thickness and lithologies) into 528 chronostratigraphic intervals (age) let us to calculate sedimentation rates (see 529 Table 2 for summary). In this section, we document the spatial trend of 530 sedimentation rates for each interval of time in order to locate depocenters 531 and infer its three dimensional geometry in map and profile view (Figs. 8 & 9). 532 We use a sedimentation rate of 80 m/m.y as an arbitrary reference to enclose 533 areas with relative high sedimentation rates as depocenters. 534 Maastrichtian depocenters (65-70 Ma). Two depocenters with localized 535 areas of maximum sedimentation rates were formed at this time. In the 536 western domain, a localized depocenter formed around San de Juan de Ro 537 Seco area (95 m/m.y). Sedimentation rates decrease abruptly father north in 538 Guaduero area (3.4 m/m.y), and we infer that sedimentation rates of the 539 depocenter increased eastward (i.e., become thicker eastward; Fig. 9b). 540 Maastrichtian strata thin or are eroded by an unconformity farther to the north 541 (Ro Minero area; Restrepo-Pace et al. 2004), and are not recorded by an 542 unconformity farther south (Fusa Syncline; Bayona et al. 2003b). In contrast, 543 the depocenter in the axial domain has an elongated geometry with a NNW 544 strike (Fig. 8a). A maximum sedimentation rate of 131 m/m.y is documented 545 in Checua area; other values decrease in all directions as far as the eastern 546 domain where Maastrichtian strata are not preserved (Fig. 8a). The profile 547 geometry of the axial depocenter is like asymmetrical sag that is thicker to the 548 west. 549 Lower to middle Palaeocene depocenters (59-65 Ma). Maximum 550 sedimentation rates in the western domain are located in the East San Juan 551 de Ro Seco 122 m/m.y and decrease northward to Guaduero area to 58-76 552 m/m.y, suggesting a north-striking elongated depocenter geometry (Fig. 8b) 553 that is thicker eastward (Fig. 9c). In contrast, the former Maastrichtian 554 depocenter in the axial domain broke into two depocenters. In the western 555 axial domain, the Checua area continues as the area with a maximum 556 sedimentation rate of 83 m/m.y; sedimentation rates decreases slightly 557

southward to 55-57 m/m.y and more abruptly toward the eastern axial domain 558 where sedimentation rates are between 30-36 m/m.y. In the eastern domain, 559 a new depocenter began to form as sedimentation rates increased to 38-45 560 m/m.y. The profile geometry of the axial depocenter continues as 561 asymmetrical sag that is thicker westward, whereas the geometry of the 562 eastern domain depocenter has a wedge-like geometry that is thicker 563 westward (Fig. 9c). 564 Upper Palaeocene depocenters (55-59 Ma). During late Palaeocene time, 565 the depocenter in the western domain had similar pattern as described before, 566 with maximum sedimentation rates in the East San Juan de Rio Seco area 567 (152 m/m.y) and a profile geometry that is thicker eastward (Fig. 9c). The 568 other two depocenters described for the axial and eastern domains, 569 respectively, have a significant increase of sedimentation rates in comparison 570 with rates reported in lower-middle Palaeocene time (Fig. 8c), but profile 571 geometry retains similar. In the axial domain, sedimentation rates increases 572 to 92-143 m/m.y in the Checua-Guatavita areas, whereas farther south 573 reached 131 m/my in Usme area. In other areas farther north of the axial 574 domain, sedimentation rates range between 47 to 81 m/m.y. In the 575 depocenter of the Eastern domain, sedimentation rates reached 119 m/m.y 576 for the Medina area. 577 Lower Eocene depocenters (49-55 Ma). Only two depocenters formed at 578 this time. The depocenter in the western domain maintain its location in the 579 San Juan de Rio Seco area (up to 138 m/m.y) with a wedge-like geometry 580 thickening eastward. As in previous intervals of time, sedimentation rates 581 decrease abruptly northward to less than 12 m/m.y. The other depocenter 582 formed in the southern western axial domain, in the Usme (115 m/m.y) and 583 Guatavita (118 m/m.y) areas. Farther north and east, sedimentation rates 584 decreased to values ranging 15-45 m/m.y in the axial domain, and 0-16 m/m.y 585 in the Eastern domain. 586 587 Discussion 588 Integration of provenance, spatial and temporal variation of sedimentation 589 rates, and location of depocenters allow arising a new model of tectono-590 stratigraphic evolution of Maastrichtian-lower Eocene syn-orogenic basins. 591 We compare this model with previous tectono-stratigraphic models proposed 592 (e.g., a foreland basin setting), and then we link the evolution of these syn-593 orogenic basins with the interaction between the Caribbean and South 594 America plates. 595 Provenance signatures of Maastrichtian to middle Paleocene sandstones (Fig. 596 4a & b) show two different sources supplying sediments to the western 597 domain and the axial domain. The following criteria indicate that an uplifted 598 area with a Cretaceous sedimentary cover separated these two domains: 599 (1) reworked glauconite (Fig. 4f and 4g) and Cretaceous pollen in different 600 areas of the axial domain (Fig. 8a & b) 601

(2) increase of muddy matrix in the fabric of Palaeocene sandstones (Fig. 4f 602 and 4g) 603 (3) dominance of stable and ultrastable heavy minerals in the axial domain 604 with a localized increase of unstable minerals in the Usme area in upper 605 Palaeocene sandstones, whereas in the western domain stable and unstable 606 minerals are more common in lower-middle Palaeocene sandstones and 607 ultrastable minerals become dominant in upper Palaeocene sandstones (Fig. 608 5a) 609 (4) Northward enrichment of 900-1800 Ma detrital zircon age population in 610 sandstones of the axial zone (Fig. 7). 611 612 The four criteria presented above also indicate that the uplifted area must be 613 located nearby to the place of deposition, as pollen and glauconite are very 614 unstable fragments to survive tropical weathering and more than 40 kms of 615 fluvial transport, as documented in the fluvial sands of the Llanos basin of 616 Colombia (Amorocho et al. 2011). The reworking of compositionally mature 617 Cretaceous sandstones concentrates ultrastable heavy minerals, the muddy 618 fraction may be a product of erosion of Palaeosols (Potter et al. 2005), and 619 900-1800 Ma detrital zircon ages are very common population in Cretaceous 620 rocks (Horton et al. 2010 a & b). 621 Separation of the western and western axial domains by the uplift of 622 Cretaceous sedimentary cover is also suggested by the eastward increase of 623 thickness in the western domain, and the high sedimentation rates reported in 624 San Juan de Rio Seco area. Tectonic loads eastward of Guaduero and San 625 Juan de Ro Seco sections favored higher tectonic subsidence of the western 626 domain (Fig. 9a & b). Those tectonic loads correspond to the inversion of 627 former normal faults of the western margin of the former extensional basin 628 (Fig. 8a & 9b), as proposed by Sarmiento-Rojas (2001) and Bayona et al. 629 (2012) using geodynamic models, Cortes et al. (2006) using surface and 630 subsurface mapping relations of Palaeogene units, and Moretti et al. (2010) 631 using difference of vitrinite reflectance data of the same unit across the 632 Bituima Fault (Fig. 1). Farther to the north, Parra et al. (2012) document with 633 thermochronological data the onset of exhumation of the western flank of the 634 Eastern Cordillera in Palaeocene time. 635 Uplift of the western flank of the Eastern Cordillera was not continuous 636 southward, since continental deposits to the south of the San de Rio Seco 637 area in the western domain were connected with a fluvial system in the Usme 638 area, as suggested by similar detrital zircon population between these two 639 areas (Figs 6 and 8a & b). Northward and eastward dispersion of detritus 640 supplied from the Central Cordillera in the axial domain is further supported by 641 palaeocurrent indicators reported for Palaeocene sandstones in the axial 642 domain (see Bayona et al. 2008 for review and Saylor et al. 2011 for 643 palaeocurrent data in Paz de Ro area). 644 The presence of (1) metamorphic rock fragments; (2) garnet, apatite and 645 epidote fragments, and (3) dominance of Cretaceous and Permo-Triassic 646 zircon age populations over Jurassic age populations support the erosion of 647 Cretaceous intrusive (Antioqueo Batholith) and Permo-Triassic metamorphic 648 rocks presently exposed to the west of the Central Cordillera, adjacent to the 649

Romeral paleosuture. In contrast, the low content of Jurassic zircon age 650 population, feldspar and volcanic lithic fragments indicate that the eastern 651 flank of the Central Cordillera was scarcely exposed at this time. 652 Crustal tilting of the Central Cordillera-Santa Marta massif (Fig. 9a & b), an 653 uplift mechanism of the continental margin related to subduction of the 654 Caribbean plate (Cardona et al., 2011; Bayona et al., 2011, 2012, Ayala et al., 655 2012), explain both clockwise tilting of the Central Cordillera and reactivation 656 of intraplate structures. Reactivation of the western fault system of the former 657 extensional basin acted as the eastern boundary of the tilted rigid crustal 658 block (Fig. 9). Eastward thickening of the depocenter in the western domain 659 (Figs 8 & 9) is also explained by clockwise tilting of the Central Cordillera. 660 Similarly, clockwise tilting of the inverted hanging-wall block (Figs 9b & c) 661 explains the elongated depocenter documented in Maastrichtian to late 662 Palaeocene time (Fig. 8 a to c) in the western axial domain; sedimentation 663 rates in the western axial depocenter are equivalent to the rate determined in 664 the East San Juan de Rio Seco area in the western domain. The 665 simultaneous clockwise tilting activity of uplifted blocks may explain the similar 666 sedimentation rates in two areas that are 130 kms apart (Fig. 8); however, 667 local structures controlled patterns of sedimentation in the western domain 668 (e.g., normal faulting in Fig. 8c controlled localized deposition of fine-grained 669 strata of the middle Hoyon Formation). 670 The distance between the Central Cordillera and the eastern axial and 671 eastern domains is greater than 250 kms (Fig. 8); chemically unstable 672 fragments are not expected to survive hundreds of kilometers of fluvial 673 transport in tropical settings (Johnsson et al. 1991; Amorocho et al. 2011). 674 Reworked Cretaceous pollen, unstable heavy mineral fragments (Fig. 5; 675 garnet, titanite, hornblende, epidote, chlorite) and a high content of lithic 676 fragments are common features in upper Palaeocene-lower Eocene 677 sandstones of the eastern axial and eastern domains (Figs 4c & d). 678 Therefore, unstable terrigenous detritus in the eastern axial and eastern 679 domains was not entirely derived from the Central Cordillera and may include 680 the reactivation of eastern domains. In addition, characteristic detrital zircon 681 population derived from the Central Cordillera (70-360 Ma) is not recorded in 682 the eastern domain (Fig. 6). 683 In the late Palaeocene, mud-dominated deposition and increase in the content 684 of metamorphic lithic fragments in sandstones in the eastern domain began 685 earlier than in the axial domain (Figs 2b and 4c). Unstable titanite and 686 hornblende fragments in Campanian and upper Palaeocene strata (Fig. 5a) 687 also indicate the presence of nearby basement-cored uplifts. Detrital zircon 688 age population of upper Paleocene strata in the eastern domain (Fig. 6) is 689 similar to the > 1.2 Ga zircon population reported from a buried basement-690 cored high in the proximal Llanos basin (Ibaez et al. 2009). These buried 691 basement-cored highs, presently under the Cenozoic foreland strata, include 692 Cretaceous and/or Palaeozoic sedimentary units overlying basement rocks 693 (Bayona et al. 2007). Therefore, erosion of the sedimentary cover and 694 basement-cored uplifts supplied quartzose detritus, metamorphic lithic 695 fragments, unstable heavy minerals and > 1.2 Ga detrital zircons. 696

Reactivation of the eastern normal fault system of the former extensional 697 basin and structures in the southern Llanos basin occurred during the 698 Palaeocene. This deformation is evident in late Palaeocene time by the 699 immature composition of sandstones (Fig. 4c), and higher sedimentation rates 700 in the Eastern domain than in the eastern axial domain (Fig. 8c & 9c). 701 Location of the depositional axis in the eastern domain during the 702 Maastrichtian (Figs. 8a and 9b), and reactivation of the eastern normal fault 703 system preclude the eastward transport of detritus supplied from Central 704 Cordillera to the eastern domain (Fig. 6). Tectonic loading between the 705 western and eastern domains contribute to flexural tectonic subsidence and 706 foreland pattern of deposition between the eastern axial domain and the 707 proximal Llanos basin in Maastrichtian to late Palaeocene time (Fig. 9b & c) 708 (Bayona et al. 2008). The Soapaga-Pesca-Macheta Fault System was not 709 active up to early Eocene time, since there are not abrupt changes in 710 sedimentation rates and provenance signatures between the western axial 711 and eastern axial domains. Why this fault system did not reactivate, in 712 contrast to the faults bordering the extensional basin, needs further 713 investigation. 714 Fault reactivation along the eastern margin is coeval with the record of 715 volcanic zircons in upper Palaeocene sandstones (Bayona et al. 2012). 716 These two processes may be explained by eastward shallow subduction of 717 the Caribbean plate (Figs 9c and d) that produced intraplate magmatism and 718 fault reactivation as far as the Llanos basin. Intraplate magmatism affecting 719 Eastern Cordillera and adjacent basins to the east has been reported for 720 Cretaceous (Vasquez et al. 2010) and Mio-Pliocene time (Taboada et al. 721 2000; Vasquez et al. 2009). 722 723 Conclusions 724 Spatial and temporal variation of sedimentation rates and provenance 725 signatures were used to decipher early phases of deformation in a multi-726 phase orogen, like the Eastern Cordillera of Colombia. Maastrichtian to lower 727 Eocene strata in the western, axial and eastern domains of the central 728 segment of Eastern Cordillera support a model of eastward migration of 729 intraplate fault reactivation (positive inversion structures). Diachronous 730 intraplate fault reactivation, development of syn-orogenic basins and intraplate 731 magmatism are associated with convergence and subduction of the buoyant 732 Caribbean plate beneath the South American margin (Fig. 9). Positive 733 reactivation of normal faults bounding the western margin of the former 734 extensional basin is associated to clockwise tilting of the Central Cordillera. 735 Compressive stresses transmitted to intraplate settings reactivated faults 736 bordering the eastern margin of the extensional basin, and reactivated faults 737 as far as the proximal Llanos basin. Fault reactivation was coeval with 738 marginal and intraplate magmatism. 739 Fault reactivation in Palaeocene time produced three independent syn-740 orogenic basins separated by low-amplitude uplifts with exposure of 741 Cretaceous sedimentary cover that supplied sediments to an axial 742

depocenter. Rocks exposed along the trace of the Romeral paleosuture of the 743 Central Cordillera supplied sediments to the western and axial domains, as 744 recorded by detrital zircon population ages younger than 360 Ma, 745 metamorphic rock fragments and garnet-epidote-hornblende heavy mineral 746 assemblage. The source area that supplied sediments to the eastern 747 depocenter was presently-buried basement-cored highs in the southern 748 Llanos basin, as indicated by the dominance of >1.2 Ga zircon population 749 (Fig. 6), immature composition of upper Palaeocene sandstones, and record 750 of unstable titanite and hornblende fragments. Reworked glauconite and 751 cretaceous pollen fragments, presence of ferrogineous matrix in quartzarenite 752 sandstones implying erosion of paleosol profiles, and northward enrichment of 753 900-1800 Ma detrital zircon age populations in the axial domain support the 754 interpretation of unroofing of nearby Cretaceous sedimentary cover 755 756 Acknowledgements 757 This research is part of a multi-disciplinary project funded by Ecopetrol S.A-758 ICP (project Cronologa de la deformacin de cuencas subandinas), Hocol 759 S.A, Maurel & Prom, and Corporacin Geolgica ARES. Thanks to Johan 760 Ortiz, Giovanny Nova, Paola Montao, Oscar Molsalve, Jairo Roncancio and 761 Juan Caicedo all of whom participated in the field work for this project; to 762 Diana Ochoa, Millerlandy Romero, Andres Pardo, Silane daSilva, Manuel 763 Paez and Milton Rueda for their palynological analysis; Camilo Bustamente, 764 Carolina Ojeda, Andrea Torres and Victor Valencia for their provenance 765 analysis. We also want to thank Nelson Sanchez, Jorge Rubiano and Nestor 766 Moreno (ICP-Ecopetrol) and Mauricio Parra for beneficial discussions. Our 767 gratitude also goes to Mauricio Ibaez (U. of Arizona) who shared with us the 768 geochronological data on basement uplifts in the proximal Llanos Basin as 769 well as Andrs Reyes (Ecopetrol), Mario de Freitas and Andrs Fajardo 770 (Hocol) who gave us the permission for the publication of the data presented 771 in this paper. The manuscript was improved by editorial comments of two 772 anonymous reviewers and editor Michal Nemcok. 773 774 References: 775 AMOROCHO, R., BAYONA, G. & REYES-HARKER, A. (2011) Controls on the 776

Composition of Fluvial Sands in the Proximal Area of a Tropical 777 Foreland Basin (Colombia). Geologa Colombiana, 36, 163-177. 778

ASPDEN, J.A., MCCOURT, W. & BROOK, M. (1987) Geochemical Control of 779 Subduction-Related Magmatism: The Mesozoic and Cenozoic Plutonic 780 History of Western Colombia. Journal of the Geological Society of 781 London, 144, 893-905. 782

AYALA, C., BAYONA, G., CARDONA, A., OJEDA, C., MONTENEGRO, O., MONTES, C., 783 VALENCIA, V. & JARAMILLO, C. (2012) The Paleogene Synorogenic 784 Succession in the Northwestern Maracaibo Block: Tracking Intraplate 785 Uplifts and Changes in Sediment Delivery Systems. Journal of South 786 America Earth Sciences, in press. 787

BAYONA, G., CORTS, M., JARAMILLO, C. & LLINS, R.D. (2003). The 788 Fusagasug Succession: A Record of the Complex Latest Cretaceous-789 Pre-Miocene Deformation between the Magdalena Valley and Sabana 790 De Bogot Areas. VIII Simposio Bolivariano de Exploracin Petrolera 791 en las Cuencas Subandinas, Cartagena, Colombia. 792

BAYONA, G. & LAWTON, T. (2003) Fault-Proximal Stratigraphic Record of 793 Episodic Extension and Oblique Inversion, Bisbee Basin, Southwestern 794 New Mexico, USA. Basin Research, 15, 251-270. 795

BAYONA, G. & THOMAS, W.A. (2003) Distinguishing Fault Reactivation from 796 Flexural Deformation in the Distal Stratigraphy of the Peripheral 797 Blountian Foreland Basin. Basin Research, 15, 503-526. 798

BAYONA, G., JARAMILLO, C., RUEDA, M., REYES-HARKER, A. & TORRES, V. (2007) 799 Paleocene-Middle Miocene Flexural-Margin Migration of the 800 Nonmarine Llanos Foreland Basin of Colombia. CT&F Ciencia, 801 Tecnologa y Futuro, 3 (3), 141-160. 802

BAYONA, G., CORTES, M., JARAMILLO, C., OJEDA, G., ARISTIZABAL, J. & REYES-803 HARKER, A. (2008) An Integrated Analysis of an Orogen-Sedimentary 804 Basin Pair: Latest Cretaceous-Cenozoic Evolution of the Linked 805 Eastern Cordillera Orogen and the Llanos Foreland Basin of Colombia. 806 Geological Society of America Bulletin, 120, 1171-1197. 807

BAYONA, G., MONTENEGRO, O., CARDONA, A., JARAMILLO, C., LAMUS, F., MORN, S., 808 QUIROZ, L., RUIZ, M., VALENCIA, V., PARRA, M. & STOCKLI, D. (2010) 809 Estratigrafa, Procedencia, Subsidencia Y Exhumacin De Las 810 Unidades Palegenas En El Sinclinal De Usme, Sur De La Zona Axial 811 De La Cordillera Oriental. Geologia Colombiana, 35, 5-35. 812

BAYONA, G., MONTES, C., CARDONA, A., JARAMILLO, C., OJEDA, G. & VALENCIA, V. 813 (2011) Intraplate Subsidence and Basin Filling Adjacent to an Oceanic 814 Arc-Continental Collision; a Case from the Southern Caribbean-South 815 America Plate Margin. Basin Research, 23, 403-422. 816

BAYONA, G., CARDONA, A., JARAMILLO, C., MORA, A., MONTES, C., VALENCIA, V., 817 AYALA CALVO, R.C., MONTENEGRO, O. & IBAEZ-MEJIA, M. (2012) Early 818 Paleogene Magmatism in the Northern Andes: Insights on the Effects 819 of Oceanic Plateau-Continent Convergence. Earth and Planetary 820 Science Letters, 331-332, 97-111. 821

BEAUMONT, C. (1981) Foreland Basins. Royal Astronomical Society 822 Geophysical Journal, 65, 291-329. 823

BRANQUET, Y., CHEILLETZ, A., COBBOLD, P., BABY, P., LAUMONIER, B. & GIULIANI, 824 G. (2002) Andean Deformation and Rift Inversion, Eastern Edge of the 825 Cordillera Oriental (Guateque - Medina Area), Colombia. Journal of 826 South American Earth Sciences, 15, 391-407. 827

BUCHANAN, P. & MCCLAY, K. (1992) Experiments on Basin Inversion above 828 Reactivated Domino Faults. Marine and Petroleum Geology, 9, 486-829 500. 830

BUCHANAN, P.G. & MCCLAY, K.R. (1991) Sandbox Experiments of Inverted 831 Listric and Planar Fault Systems. Tectonophysics, 188, 97-115. 832

BURBANK, D., BECK, R. & MULDER, T. (1996) The Himalayan Foreland Basin In: 833 The Tectonic Evolution of Asia (Ed. by A. Yin & M. Harrison), 148-188. 834 Cambridge University Press, Cambridge. 835

BUTLER, R.W.H. (1989) The Influence of Preexisting Basin Structure on Thrust 836 Systems Evolution in the Western Alps. In: Inversion Tectonics (Ed. by 837

M. A. Cooper & G. D. Williams), Special Publication 44, 105122. 838 Geological Society, London. 839

CARDONA, A., VALENCIA, V., BAYONA, G., DUQUE, J., DUCEA, M., GERHELS, G., 840 JARAMILLO, C., MONTES, C., OJEDA, G. & RUIZ, J. (2011) Early Subduction 841 Orogeny in the Northern Andes: Turonian to Eocene Magmatic and 842 Provenance Record in the Santa Marta Massif and Rancheria Basin, 843 Northern Colombia. Terranova, 23, 26-34. 844

CAZIER, E.C., HAYWARD, A.B., ESPINOSA, G., VELANDIA, J., MUGNIOT, J.F. & 845 LEEL, W.G. (1995) Petroleum Geology of the Cusiana Field, Llanos 846 Basin Foothills, Colombia. America Association of Petroleum 847 Geologists Bulletin, 79, 1444-1463. 848

CEDIEL, F., SHAW, R. & CCERES, C. (2003) Tectonic Assembly of the Northern 849 Andean Block. In: The Circum-Gulf of Mexico and the Caribbean: 850 Hydrocarbon Habitats, Basin Formation and Plate Tectonics (Ed. by C. 851 Bartolini, R. Buffler & J. Blickwede), 815-848. American Association of 852 Petroleum Geologists Memoir 79. 853

COLLETA, B., HBRARD, F., LETOUZEY, J., WERNER, P. & RUDKIEWIKZ, J.L. 854 (1990) Tectonic Style and Crustal Structure of the Eastern Cordillera 855 (Colombia), from a Balanced Cross Section. In: Petroleum and 856 Tectonics in Mobile Belts (Ed. by J. Letouzey), 81-100. Editions 857 Technip, Paris 11990, Paris. 858

COLLETA, B., ROURE, F., DETONI, B., LOUREIRO, D., PASSALAQUA, H. & GOU, Y. 859 (1997) Tectonic Inheritance, Crustal Architecture and Contrasting 860 Structural Styles in the Venezuela Andes . Tectonics, 16 777794. 861

COOPER, M.A., ADDISON, F.T., ALVAREZ, R., CORAL, M., GRAHAM, R.H., 862 HAYWARD, A.B., HOWE, S., MARTINEZ, J., NAAR, J., PEAS, R., PULHAM, 863 A.J. & TABORDA, A. (1995) Basin Development and Tectonic History of 864 the Llanos Basin, Eastern Cordillera, and Middle Magdalena Valley, 865 Colombia. American Association of Petroleum Geologists Bulletin, 79, 866 1421-1443. 867

CORTES, M. (2004) Evolution Structurale Du Front Centre-Occidental De La 868 Cordillre Orientale De Colombie. Ph.D Thesis, Universit Pierre et 869 Marie Curie, Paris. 870

CORTS, M., ARISTIZABAL, J.J., BAYONA, G., OJEDA, G., REYES, A. & GAMBA, N. 871 (2006). Structure and Kinematics of the Eastern Foothills of the 872 Eastern Cordillera of Colombia from Balanced Cross-Sections and 873 Forward Modelling. IX Simposio Bolivariano Exploracion Petrolera en 874 las Cuencas Subandinas, Cartagena, Colombia. 875

COWARD, M. (1994) Inversion Tectonics. In: Continental Deformation (Ed. by 876 P. Hancock), 289304. Pergamon Press, Oxford 877

CRAMPTON, S.L. & ALLEN, P.A. (1995) Recognition of Forebulge 878 Unconformities Associated with Early Stage Foreland Basin 879 Development: Example from the North Alpine Foreland Basin. America 880 Association of Petroleum Geologists Bulletin, 79, 1495-1514. 881

DECELLES, P. & GILES, K. (1996) Foreland Basin Systems. Basin Research, 8, 882 105-123. 883

DECELLES, P.G. & HORTON, B.K. (2003) Early to Middletertiary Foreland Basin 884 Development and the History of Andean Crustal Shortening in Bolivia. 885 Geological Society of American Bulletin, 115, 58-77. 886

DENGO, C.A. & COVEY, M.C. (1993) Structure of the Eastern Cordillera of 887 Colombia: Implications for Trap Styles and Regional Tectonics. 888 American Association of Petroleum Geologists Bulletin, 77, 1315-1337. 889

DEPORTA, J. (1966) Geologa Del Extremo Sur Del Valle Medio Del 890 Magdalena. Boletn Geolgico Universidad Industrial de Santander, 22-891 23, 1-347. 892

ETAYO-SERNA, F., BARRERO, D., LOZANO, H. & OTHERS 15 AUTHORS (1983) 893 Mapa De Terrenos Geolgicos De Colombia. Ingeominas, Bogot. 894

ETHERIDGE, M.A. (1986) On the Reactivation of Extensional Fault Systems 895 Philosophical Transactions of the Royal Society of London, A317, 179896 194. 897

FABRE, A. (1981) Estratigrafa De La Sierra Nevada Del Cocuy, Boyac Y 898 Arauca, Cordillera Oriental (Colombia). Geologa Norandina, 4, 3-12. 899

FABRE, A. (1987) Tectonique Et Gneration D'hydrocarbures: Un Modle De 900 L'evolution De La Cordillre Orientale De Colombie Et Du Bassin De 901 Llanos Pendant Le Crtac Et Le Tertiaire. Archives des Sciences 902 Genve, 40, 145-190. 903

FAJARDO-PEA, G. (1998) Structural Analysis and Basin Inversion Evolutionary 904 Model of the Arcabuco, Tunja and Sogamoso Regions, Eastern 905 Cordillera, Colombia. MSc. Thesis, University of Colorado, Boulder. 906

FLEMINGS, P.B. & JORDAN, T.E. (1989) A Synthetic Stratigraphic Model of 907 Foreland Basin Development. J. Geophys. Res. - Solid Earth, 94, 908 3851-3866. 909

GMEZ, E. & PEDRAZA, P. (1994) El Maastrichtiano De La Regin De Honda - 910 Guaduas, Lmite Norte Del Valle Superior Del Magdalena: Registro 911 Sedimentario De Un Delta Dominado Por Ros Trenzados. In: Estudios 912 Geolgicos Del Valle Superior Del Magdalena (Ed. by F. Etayo-Serna), 913 20. Universidad Nacional de Colombia, Bogot. 914

GMEZ, E., JORDAN, T.E., ALLMENDINGER, R.W., HEGARTY, K., KELLEY, S. & 915 HEIZLER, M. (2003) Controls on Architecture of the Late Cretaceous to 916 Cenozoic Southern Middle Magdalena Valley Basin. Geological Society 917 of America Bulletin, 115, 131-147. 918

GMEZ, E., JORDAN, T., ALLMENDINGER, R.W. & CARDOZO, N. (2005a) 919 Development of the Colombian Foreland-Basin System as a 920 Consequence of Diachronous Exhumation of the Northern Andes. 921 Geological Society of America Bulletin, 117, 1272-1292; doi: 922 1210.1130/B25456.25451. 923

GMEZ, E., JORDAN, T., ALLMENDINGER, R.W., HEGARTY, K. & KELLEY, S. 924 (2005b) Syntectonic Cenozoic Sedimentation in the Northern Middle 925 Magdalena Valley Basin of Colombia and Implications for Exhumation 926 of the Northern Andes. Geological Society of America Bulletin, 117, 927 547569; doi: 510.1130/B25454.25451. 928

GUERRERO, J. & SARMIENTO, G. (1996) Estratigrafa Fsica, Palinolgica, 929 Sedimentolgica Y Secuencial Del Cretcico Superior Y Paleoceno Del 930 Piedemonte Llanero, Implicaciones En Exploracin Petrolera. Geologia 931 Colombiana, 20, 3-66. 932

HAIN, M.P., STRECKER, M.R., BOOKHAGEN, B., ALONSO, R.N., PINGEL, H. & 933 SCHMITT, A.K. (2011) Neogene to Quaternary Broken Foreland 934 Formation and Sedimentation Dynamics in the Andes of Nw Argentina 935 (25S). Tectonics 30, 27 p. 10.1029/2010TC002703. 936

HAYWARD, A.B. & GRAHAM, R.H. (1989) Some Geometrical Characteristics of 937 Inversion. In: Inversion Tectonics (Ed. by M. A. Cooper & G. D. 938 Williams), Special Publication 44, 1739. Geological Society, London 939

HELLER, P.L., ANGEVINE, C.L., WINSLOW, N.S. & PAOLA, C. (1988) Two-Phase 940 Stratigraphic Model of Foreland-Basin Sequences. Geology, 16, 501-941 504. 942

HILLEY, G.E. & STRECKER, M.R. (2005) Processes of Oscillatory Basin Filling 943 and Excavation in a Tectonically Active Orogen: Quebrada Del Toro 944 Basin, Nw Argentina. Geological Society of American Bulletin, 117, 945 887901. 946

HORTON, B.K., PARRA, M., SAYLOR, J.E., NIE, J., MORA, A., TORRES, V., 947 STOCKLI, D.F. & STRECKER, M.R. (2010a) Resolving Uplift of the 948 Northern Andes Using Detrital Zircon Age Signatures. GSA Today, 20, 949 4-9. 950

HORTON, B.K., SAYLOR, J.E., NIE, J., MORA, A., PARRA, M., REYES-HARKER, A. & 951 STOCKLI, D.F. (2010b) Linking Sedimentation in the Northern Andes to 952 Basement Configuration, Mesozoic Extension, and Cenozoic 953 Shortening: Evidence from Detrital Zircon U-Pb Ages, Eastern 954 Cordillera, Colombia. GSA Bulletin, 122, 14231442. 955

IBAEZ-MEJIA, M., RUIZ, J., GEHRELS, G., MORA, A., CARDONA, A. & VALENCIA, V. 956 (2009) Detrital Zircon of Low-Grade Metasediments from the Basement 957 of the North Andean Llanos Foreland Basin-Colombia: Provenance 958 Signatures of Early Iapetan Passive Margin Deposits in Northwestern 959 Amazonia. Geological Society of America Abstracts with Programs. 960 Portland, Oregon. 41, 662. 961

JARAMILLO, C., RUEDA, M. & TORRES, V. (2011) A Palynological Zonation for the 962 Cenozoic of the Llanos and Llanos Foothills of Colombia. Palynology, 963 35, 46-84. 964

JOHNSSON, M.J., STALLARD, R.F. & LUNDBERG, N. (1991) Controls on the 965 Composition of Fluvial Sands from a Tropical Weathering Environment: 966 Sands of the Orinoco River Drainage Basin, Venezuela and Colombia. 967 Geological Society of America Bulletin, 103, 1622-1647. 968

JORDAN, T.E. (1981) Thrust Loads and Foreland Basin Evolution, Cretaceous, 969 Western United States. American Association of Petroleum Geologists 970 Bulletin, 65, 2506-2520. 971

KAMMER, A. & SANCHEZ, J. (2006) Early Jurassic Rift Structures Associated 972 with the Soapaga and Boyac Faults of the Eastern Cordillera, 973 Colombia: Sedimentological Inferences and Regional Implications. 974 Journal of South American Earth Sciences, 21, 412-422. 975

KLUTH, C.F. & CONEY, P.J. (1981) Plate Tectonics of the Ancestral Rocky 976 Mountains. Geology, 9, 1015. 977

MARSHAK, S., KARLSTROM, K.E. & TIMMONS, J.M. (2000) Inversion of 978 Proterozoic Extensional Faults , an Explanation for the Pattern of 979 Laramide and Ancestral Rockies Intracratonic Deformation, United 980 States. Geology, 28, 735738. 981

MCCOURT, W.J., FEININGER, T. & BROOK, M. (1984) New Geological and 982 Geochronological Data from the Colombian Andes: Continental Growth 983 by Multiple Accretion. Journal of the Geological Society of London, 984 141, 831-845. 985

MCLAUGHLIN, D. & ARCE, M. (1975) Mapa Geologico Del Cuadrangulo K-11, 986 Zipaquira, INGEOMINAS. Bogota. 987

MONTES, C., HATCHER JR., D. & RESTREPO-PACE, P. (2005) Tectonic 988 Reconstruction of the Northern Andean Blocks: Oblique Convergence 989 and Rotations Derived from the Kinematics of the Piedras-Girardot 990 Area, Colombia. Tectonophysics, 399, 221-250. 991

MORA, A., PARRA, M., STRECKER, M.R., KAMMER, A., DIMATE, C. & RODRIGUEZ, 992 F. (2006) Cenozoic Contractional Reactivation of Mesozoic Extensional 993 Structures in the Eastern Cordillera of Colombia. Tectonics, 25. 994

MORA, A., GAONA, T., KLEY, J., MONTOYA, D., PARRA, M., QUIROZ, L., REYES, G. 995 & STRECKER, M.R. (2008a) The Role of Inherited Extensional Fault 996 Segmentation and Linkage in Contractional Orogenesis: A 997 Reconstruction of Lower Cretaceous Inverted Rift Basins in the Eastern 998 Cordillera of Colombia. Basin Research, 21, 111137. 999

MORA, A., PARRA, M., STRECKER, M.R., SOBEL, E.R., HOOGHIEMSTRA, H., TORRES, V. 1000 & VALLEJO-JARAMILLO, J. (2008b) Climatic Forcing of Asymmetric 1001 Orogenic Evolution in the Eastern Cordillera of Colombia. Geological 1002 Society of America Bulletin, 120, 930-949. 1003

MORA, A., HORTON, B., MESA, A., RUBIANO, J., KETCHMAN, R., PARRA, M., BLANCO, 1004 V., GARCA, D. & STOCKLI, D. (2010) Migration of Cenozoic Deformation 1005 in the Eastern Cordillera of Colombia Interpreted from Fission Track 1006 Results and Structural Relationships: Implications for Petroleum 1007 Systems. AAPG Bulletin, 94, 1543-1580. 1008

MORETTI, I., RODRIGUEZ, G., MAYORGA, M. & MONDRAGON, J. (2010) Integrated 1009 Exploration Workflow in the South Middle Magdalena Valley 1010 (Colombia). Journal of South America Earth Sciences, 29, 187-197. 1011

PARDO, A. (2004) Paleocene-Eocene Palynology and Palynofacies from 1012 Northeastern Colombia and Western Venezuela. Ph. D. Thesis, 1013 Universite de Liege. 1014

PARRA, M., MORA, A., JARAMILLO, C., STRECKER, M.R., SOBEL, E.R., QUIROZ, 1015 L.I., RUEDA, M. & TORRES, V. (2009) Orogenic Wedge Advance in the 1016 Northern Andes: Evidence from the Oligocene-Miocene Sedimentary 1017 Record of the Medina Basin, Eastern Cordillera, Colombia. Geological 1018 Society of America Bulletin, 121, 780-800. 1019

PARRA, M., MORA, A., LOPEZ, C., ROJAS, L.E. & HORTON, B.K. (2012) Detecting 1020 Earliest Shortening and Deformation Advance in Thrust-Belt 1021 Hinterlands: Example from the Colombian Andes. Geology, 40, 175-1022 178. 1023

PEREZ, G. & SALAZAR, A. (1978) Estratigrafia Y Facies Del Grupo Guadalupe. 1024 Geologia Colombiana, 10, 7-85. 1025

PINDELL, J.L., KENNAN, L., MARESCH, W.V., STANEK, K.-P., DRAPER, G. & HIGGS, 1026 R. (2005) Plate Kinematics and Crustal Dynamics of Circum-Caribbean 1027 Arc-Continent Interactions: Tectonic Controls on Basin Development in 1028 Proto-Caribbean Margins. In: Caribbean-South American Plate 1029 Interactions, Venezuela (Ed. by H. G. Ave Lallemant & V. B. Sisson), 1030 394, 7-52. Geological Society of America Special Paper 394. 1031

POTTER, P.E., MAYNARD, J.B. & DEPETRIS, J.P. (2005) Mud and Mudstones, 1032 Introduction and Overview. Springer. 1033

RESTREPO-PACE, P., COLMENARES, F., HIGUERA, C. & MAYORGA, M. (2004) A 1034 Fold-and-Thrust Belt Along the Western Flank of the Eastern Cordillera 1035

of Colombia - Style, Kinematics, and Timing Constraints Derived from 1036 Seismic Data and Detailed Surface Mapping. In: Thrust Tectonics and 1037 Hydrocarbon Systems (Ed. by K. McClay), 598-613. American 1038 Association of Petroleum Geologists Memoir 82. 1039

REYES, A. (1996) Sedimentology and Stratigraphy and Its Control on Porosity 1040 and Permeability Barco Formation, Cusiana Field, Colombia. Master's 1041 Thesis, University of Reading, Reading. 1042

SARMIENTO, G. (1992) Estratigrafa Y Medio De Depsito De La Formacin 1043 Guaduas. Boletn Geolgico Ingeominas., 32, 3-44. 1044

SARMIENTO-ROJAS, L.F. (2001) Mesozoic Rifting and Cenozoic Basin Inversion 1045 History of the Eastern Cordillera, Colombian Andes; Inferences from 1046 Tectonic Models. ECOPETROL-Netherlands Research School of 1047 Sedimentary Geology, Bogot. 1048

SARMIENTO-ROJAS, L.F., VAN WESS, J.D. & CLOETINGH, S. (2006) Mesozoic 1049 Transtensional Basin History of the Eastern Cordillera, Colombian 1050 Andes: Inferences from Tectonic Models. Journal of South American 1051 Earth Sciences, 21, 383-411. 1052

SAYLOR, J.E., HORTON, B.K., NIE, J., CORREDOR, J. & MORA, A. (2011) 1053 Evaluating Foreland Basin Partitioning in the Northern Andes Using 1054 Cenozoic Fill of the Floresta Basin, Eastern Cordillera, Colombia. Basin 1055 Research, 23, 377402. 1056

SAYLOR, J.E., STOCKLI, D., HORTON, B., NIE, J. & MORA, A. (2012) Discriminating 1057 Rapid Exhumation from Syndepositional Volcanism Using Detrital 1058 Zircon Double Dating: Implications for the Tectonic History of the 1059 Eastern Cordillera, Colombia. Geological Society of America Bulletin, 1060 124, 762-779. 1061

SIBSON, R.H. (1985) A Note on Fault Reactivation. Journal of Structural 1062 Geology, 7, 751754. 1063

SIBSON, R.H. (1995) Selective Fault Reactivation During Basin Inversion: 1064 Potential for Fluid Redistribution through Fault Valve Action. In: Basin 1065 Inversion (Ed. by J. G. Buchanan & P. G. Buchanan), Special 1066 Publication 88, 319. Geological Society, London. 1067

SINCLAIR, H.D. (1997) Tectonostratigraphic Model for Underfilled Peripheral 1068 Foreland Basins: An Alpine Perspective. Geological Society of America 1069 Bulletin, 109, 324-346. 1070

SOBEL, E.R., HILLEY, G.E. & STRECKER, M.R. (2003) Formation of Internally 1071 Drained Contractional Basins by AridityLimited Bedrock Incision. J. 1072 Geophys. Res. ^ Solid. 1073

SOBEL, E.R., AND M. R. STRECKER (2003) Uplift, Exhumation and Precipitation: 1074 Tectonic and Climatic Control of Late Cenozoic Landscape Evolution in 1075 the Northern Sierras Pampeanas, Argentina. Basin Research, 15, 431-1076 451. 1077

STRECKER, M.R., ALONSO, R.N., BOOKHAGEN, B., CARRAPA, B., HILLEY, G.E., 1078 SOBEL, E.R. & TRAUTH, M.H. (2007) Tectonics and Climate of the 1079 Southern Central Andes. Annu. Rev. Earth Planet. Sci., 35, 747787. 1080

STRECKER, M.R., ALONSO, R., BOOKHAGEN, B., CARRAPA, B., COUTAND, I., HAIN, 1081 M.P., HILLEY, G.E., MORTIMER, E., SCHOENBOHM, L. & SOBEL, E.R. 1082 (2009) Does the Topographic Distribution of the Central Andean Puna 1083 Plateau Result from Climatic or Geodynamic Processes? Geology, 37, 1084 643646. 1085

STRECKER, M.R., HILLEY, G.E., BOOKHAGEN, B. & SOBEL, E.R. (2011) 1086 Structural, Geomorphic and Depositional Characteristics of Contiguous 1087 and Broken Foreland Basins: Examples from the Eastern Flanks of the 1088 Central Andes in Bolivia and Nw Argentina, in Recent Advances. In: 1089 Tectonics of Sedimentary Basins (Ed. by C. Busby & A. Azor). 1090 Blackwell, Boston. 1091

TABOADA, A., RIVERA, L.A., FUENZALIDA, A., CISTERNAS, A., PHILIP, H., 1092 BIJWAARD, H., OLAYA, J. & RIVERA, C. (2000) Geodynamics of the 1093 Northern Andes: Subductions and Intracontinetal Deformation 1094 (Colombia). Tectonics, 19, 787-813. 1095

TORO, J., ROURE, F., BORDAS-LE FLONCH, N., LE CORNEC-LANCE, S. & SASSI, W. 1096 (2004) Thermal and Kinematic Evolution of the Eastern Cordillera Fold 1097 and Thrust Belt, Colombia. In: Deformation, Fluid Flow, and Reservoir 1098 Appraisal in Foreland Fold and Thrust Belts: Aapg Hedberg Series, No. 1099 1 (Ed. by R. Swennen, F. Roure & J. W. Granath), 1, 79 115. 1100

VASQUEZ, M., ALTENBERGER, U. & ROMER, R. (2009) Neogene Magmatism and 1101 Its Possible Causal Relationship with Hydrocarbon Generation in Sw 1102 Colombia. International Journal of Earth Sciences (Geol Rundsch), 98, 1103 1053-1062. 1104

VASQUEZ, M., ALTENBERGER, U., ROMER, R., SUDO, M. & MORENO-TRUJILLO, J. 1105 (2010) Magmatic Evolution of the Andean Eastern Cordillera of 1106 Colombia During the Cretaceous: Influence of Previous Tectonic 1107 Processes. Journal of South America Earth Sciences, 29, 171-186. 1108

VILLAMIL, T. (1999) Campanian-Miocene Tectonostratigraphy, Depocenter 1109 Evolution and Basin Development of Colombia and Western 1110 Venezuela. Palaeogeography, Palaeoclimatology, Palaeoecology, 153, 1111 239-275. 1112

ZIEGLER, P.A., VANWEES, J.D. & CLOETINGH, S. (1998) Mechanical Controls on 1113 Collision Related Compressional Intraplate Deformation. 1114 Tectonophysics, 300, 103129. 1115

1116 1117 Figure captions 1118 Figure 1. A) Location of the three Cordilleras in the northern Andes and 1119 Romeral fault system (RFS). B) Regional cross section from the northern 1120 Andes (modified from Bayona et al. 2012). C) Geologic map showing major 1121 structures of Eastern Cordillera and the location of selected studied areas. 1122 For reference, the studied areas were grouped into four structural domains: 1123 Western, Western Axial, Eastern Axial, and Eastern domains. 1124 Figure 2. A) Lithostratigraphic correlation of the Maastrichtian-Lower Eocene 1125 strata in the four structural domains showing lateral variation of thickness and 1126 lithology of stratigraphic units (names of lithological units are shown to the left 1127 of each structural domain). Datum= lower-middle Eocene unconformity. B) 1128 Chronostratigraphic correlation of upper Campanian - middle Eocene strata 1129 based on palynological and geochronological data. See Figure 1 for location 1130 of sections and Table 1 for a summary of thickness and chronostratigraphic 1131 data per area. 1132

Figure 3. Identification of chronostratigraphic markers in the stratigraphic 1133 column of the Umbita area, and the location of boundaries of 1134 chronostratigraphic intervals. Note the change in rate of accumulation 1135 between sandy and fine-grained intervals. See Table 2 for identification of 1136 chronostratigraphic intervals for other areas. 1137 Figure 4. A to D: Ternary provenance and lithic population diagrams for each 1138 chronostratigraphic interval showing the calculated mean value and shaded 1139 polygons represent 1 sigma uncertainty envelopes (see Appendix 2 for raw 1140 data). For lower Eocene sandstones (Fig. 4D), samples of the quartzose 1141 Picacho and Mirador Formations are plotted separately because those units 1142 are above an unconformity. Note the upsection increase of lithic content in all 1143 domains, and the enrichment of metamorphic lithic fragments in late 1144 Paleocene time for the western and eastern domains. E: Authigenic 1145 glauconite. F and G: Reworked glauconite and ferruginous matrix reported in 1146 Maastrichtian strata of the axial domain. Areas with reworked glauconite are 1147 indicated for each interval of time. H to J: Eastward increase of sandstone 1148 maturity of lower-middle Palaeocene sandstones, indicating at least three 1149 provenance fields (see 4B) from transitional recycled orogen to the west, to 1150 quartzose recycled in the axial zone, and to craton interior to the east. 1151 Figure 5. A. Lateral distribution of heavy minerals in Maastrichtian to lower 1152 Eocene sandstones. B to H. Photomicrographs of titanite, garnet, pyroxene 1153 and hornblende as example of unstable and stable minerals reported at 1154 different domains and at different age intervals; the presence and good 1155 preservation of these minerals indicate provenance from metamorphic and 1156 igneous rocks and accumulation in a first-cycle sandstone. I. Muscovite 1157 fragments are very common in northern areas of the axial zone; this mineral is 1158 an indicator of accumulation in low-energy settings. 1159 Figure 6. Detrital zircon age population distributed by age interval and 1160 structural domain. Note the dominance of young age population (70-300 Ma) 1161 in the western domain since Campanian, while this population is absent in the 1162 eastern domain; only Paleocene ages of volcanic zircons are reported in 1163 eastern domain. In western axial zone, the dominance of 70-300 Ma 1164 increases since lower-middle Palaeocene sandstones, whereas in the eastern 1165 axial zone this population increases in upper Palaeocene sandstones. 1166 Paleozoic and Proterozoic age populations dominate in the eastern axial and 1167 eastern domains in all the intervals, and in the western axial domain in upper 1168 Cretaceous rocks, and lower-middle Palaeocene rocks of the northern areas 1169 (see Fig. 7). Archean ages (>2500 Ma) are absent in all the study areas. 1170 Figure 7 Detrital zircon age populations of lower-middle Palaeocene 1171 sandstones from the western axial zone showing the northward increase of 1172 900-1800 Ma ages. The increment of this age population is related to the 1173 reworking of Cretaceous rocks, as the exposure of these units increases 1174 northward (see Figure 1). 1175 Figure 8. A to D. Palaeogeographic maps for the four chronostratigraphic 1176 intervals discussed in the text showing the location of source areas, active 1177 structures, geometry of depocenters, sedimentation rates, depositional 1178

environments and dispersion of terrigenous detritus. E. Palinspastic map for 1179 Palaeogene time of Sarmiento-Rojas (2001) used as base map for location of 1180 studied areas shown in A to D. A to D. These diagrams show the northward 1181 withdrawn of marginal deposits and how depositional systems of the western 1182 domain were disconnected since Maastrichtian time. B and C. During the 1183 Palaeocene, a single depositional system broke into two depositional systems 1184 due to eastward migration of intraplate reactivation. The depositional system 1185 in the axial zone shows the presence of low-energy settings (i.e., continental 1186 lakes); farther north, this depositional system is closed by uplift of the 1187 Santander massif at this time (Ayala et al. 2012). In contrast, the depocenter 1188 formed in the eastern domain show the northward dispersion of terrigenous 1189 material. Depocenters with sedimentation rates > 80 m/m.y. in the axial 1190 domain formed eastward to reactivated faults. (D) In Early Eocene time, 1191 magmatic activity continues along the Central Cordillera and advances 1192 onward to intraplate settings (Fig. 9c; Bayona et al. 2012); localized 1193 depocenters formed in the southern western axial domain and the western 1194 domain, and low tectonic subsidence regime dominated in eastern axial and 1195 eastern domains. 1196 Figure 9. A to C. Crustal-scale cross sections for 75, 65 and 55 Ma (see Fig. 1197 8A and C for location) showing relationship between Caribbean subduction, 1198 marginal and intraplate magmatism (from Bayona et al. 2012), crustal tilting, 1199 fault reactivation, and geometry of intraplate syn-orogenic basins. Tectonic 1200 subsidence mechanisms vary eastward from crustal tilting of the Central 1201 Cordillera to reactivation of intraplate structures. Both mechanisms are 1202 related to subduction of the Caribbean plate. In early Eocene time, 1203 reactivation in intraplate structures in both margins of the Eastern Cordillera 1204 formed a sag-like basin bordered by exposed Cretaceous rocks; deposition in 1205 the Llanos basin corresponds to filling of a foreland basin (Bayona et al., 1206 2008). D and E. Regional tectonic setting showing the shift in the style of 1207 collision of the Caribbean plate with the NW continental margin, and its effects 1208 on continental arc magmatism and reactivation of intraplate structures 1209 (modified from Bayona et al., 2012) 1210 1211 Table Captions 1212 Table 1. Measured stratigraphic thickness, palynological results and 1213 maximum age of deposition for Maastrichtian-Eocene strata (data from 1214 Bayona et al., 2012; error for the youngest age of a single crystal is not 1215 shown). These areas corresponds to location of major syncline structures, as 1216 shown if Fig. 1. 1217 Table 2. Estimated stratigraphic thickness for chronostratigraphic intervals in 1218 each study area (see Fig. 3 for explanation of methods) and calculation of 1219 sedimentation rates (m/m.y) shown in Figure 8. 1220 1221 1222

1223

Med

ina

Syn

clin

e

Los

Cob

arde

s A

ntic

line

Arca

buco

Antic

line

Gua

vio

Antic

line

La S

alin

a F

ault

Pea

deO

roFa

ult

Bit

uim

aTh

rust

Cam

bao

Thru

st

Paja

rito

Fau

lt

Yopa

l Fa

ult

Chmeza Fault

Boyac

Faul

t

Soa p

aga

Faul

t

Pe

sca

Faul

t

LaSa

lina

Faul

tBucaram

anga - Santa Marta Fault

Chucarima Fault

Guai

cara

mo

Faul

t

Ma

chet

Fa

ult

SAN JUANRIO SECO

USME

GUADUERO

GUATAVITA

CHECUA

MACHETA

TUNJAPAIPA

PESCA

RONDONSAN ANTONIO

UMBITA

CUSIANA

MEDINA

RIO

MIN

ERO

FUSA

PAZ DEL RIO

Serv

it-

Leng

up

Fault

MID

DLE

MA

GD

ALE

NA

VA

LLEY

EASTERN CORDILLERA

LLANOS BASIN

Bogota

QUETAMEMASSIF

WD

WAD

EAD

720'0"W730'0"W740'0"W

0 20 40Km.

WD

EDWAD

CC = Central CordilleraEC = Eastern Cordillera

WC = Western CordilleraSMM = Santa Marta Massif

SM = Santander Massif

BR = Baudo Range PR = Perija Range

MA = Merida AndesMV = Magdalena Valley

RFS = Romeral Fault System

= Western Axial Domain = Eastern Axial Domain= Western Domain = Eastern Domain

Structural Domains

EAD

ED

ED

EADWADWD

Fig.1. Bayona et al.

A.

Fig. 1b

C.

B.80W 70W

5S

5N

15N

WC

SMM

CC

EC

MA

PR

BRMV

RFS

SM

WOCE ANIC DOMAIN CONT INE NTAL DOMAIN

WE S TE R NCORDILLE R A

(WC )

CENTR

ALCO

R-

DILLE

RA(C

C)

E AS TE R N CORDILLE R A (E C ) LLANOS FOR E LAND BAS INBAUDOR ANGE (BR )

S OUTH AME R ICAN PLATE

PROTO-CAR IB B E AN

E

O

40

80

120

Modified from Taboada et al. (2000) and R estrepo-Pace et al. (2004)

Asthenosphere

Oceanic C rust

NAZCA PLATE

Km Rom

erla

Fault

System

(RFS

)

WD WAD EAD EDMagd

alen

aVa

lley

P LATE

GUY

ANA

CRAT

ON

60'

0"N

50'

0"N

40'

0"N

Neogene

Quaternary

Palaeogene

Lower Cretaceous

Upper Cretaceous

Upper Paleozoic

sub-Devonian

Precambrian BasementJurassic

Paleozoic Intrusive

Triassic Intrusive

LEGEND

Well data

Section from other studies

Studied areas

Syn

clin

e

Gua

duas

SSanJuan

RoSeco

EFlank

SanJuan

deRo

Seco

Upper

Hoyn

MiddleHoyn

LowerHoyn

Seca

Seca

Regadera

Fluvialplains

Bogot

Fluvialplains

Cacho

Fluvialplains

Guaduas

CoastalPlainstoFluvialPlains

Mirador

Fluvialplains

Cuervos

Barco

Picacho

UpperSocha

LowerSocha

Guaduas

Concentracin

Guaduero

Usme

S N

Machet

Checua

Tunja

S N

Pesca

Pazdel

RoUmbita

S N

Medina

Cusiana

W

SanJuan

RoSeco

WFlank

UpperHoyn

Alluvialfans

MiddleHoyn

LowerHoyn

Seca

E

Pesca-Machetafault system

Easternfault andfoldbelt

Western

faultand

foldbelt

Fluvialplains

FluvialplainstoAlluvialfans

CoastalplainstoFluvialplains

Fluvialplains

CoastalplainstoFluvialplains

CoastalPlainstoFluvialPlains

Fluvialplains

Fluvialplains

Fluvialplains

Fluvialplains

Fluvialplains

Fluvialplains

Datum

WESTERN DOMAIN AXIAL DOMAIN EASTERN DOMAINwestern axial eastern axial

A. Lithostratigraphic correlation

Guaduas

Upper

Cretaceous

LEGENDIncomplete section

U/Pb SampleU/Pb Sample with 45-65 Ma ZirconsSample codeCalculated maximum age of deposition(Only if n 3); square for youngest age ifn

PicachoFm

.

245

LowerSochaFm

.Guaduas

Fm.

350

257

GRAVEL

UNITS

THICKN

ESS(m)

METER

S(m)

LITHOLOGY

400

0

375

250

125

500

625

750

875

1250

1125

1000

Palynology

Petrography

Heavy

Minerals

EPOCH

/AG

EGuadalupe

Fm.

Lower

Eocene

Middle-Upper

Palaeocene

Maastrichtian

Lower-Middle

Palaeocene

YoungestAge

53Ma

?

?

Unknown

?

66

38

71

64

52

350/5=70m/my

257/6=43m/my

283/4=70m/my

117/2=5