Copyright by Juan Carlos Jimenez Saldarriaga 1997 Structural Styles of the Andean Foothills, Putumayo Basin, Colombia by Juan Carlos Jimenez, B.S. THESIS Presented to the Faculty of the Graduate School of The University of Texas at Austin in Partial Fulfillment of the Requirements for the Degree of MASTER OF SCIENCE IN GEOLOGICAL SCIENCES The University of Texas at Austin December, 1997 Structural Styles of the Andean Foothills, Putumayo Basin, Colombia Approved by Supervising Committee: ·Xl ' ' ., --1,,. zr­ i 'lf/1~ ./ . ~·,,1 '-~./ ­ .. . .. ~ ,.,. Paul Mann Dedication To you Patty ... my love, my friend, my life ... Mis hijos Juanes y Juli que algun dia entenderan Tavo, continua guiandonos con sabiduria para inculcar en nuestros hijos la rectitud, honestidad y valentia que siempre demostraste, algun dia estaremos contigo nuevamente ... Acknowledgements I thank ECOPETROL, the Colombian National Oil Company, for their support and confidence received for my studies at The University of Texas at Austin. Special thanks to Dr. Randy Marrett for his friendship, guidance and advice during all phases of this work, " ... this is going to take you just TEN minutes ... ". Thanks to Dr. Paul Mann for all his interest, recommendations and technical supervision. Special thanks to my friends Edgar Kairuz, Nelson Alvarez, and Vicky Velez for their valuable help with the data. Gracias a nuestras familias, abuelo Pipao, abuela duz, abuelo Kikago y abuela moma por que siempre estuvieron pendientes de todo detalle para que nuestros asuntos en Colombia funcionaran perfectamente ... y funcionaron. Special thanks to David and Pam for their love, friendship and help. Thanks Cesar, Claudita compas, Jimmy and Claudia partners. Students from the Geology Department contributed positively in different stages of the study. In particular: Jennifer Beall, Javier Moros, Richard Weiland, Marta Beltran, Felix Dfaz, Rion Camerlo, Mario Aranda and Rogerio Souza. Thanks to Dr. Noel Tyler, Dr. Robert Hanford, Roger Tyler and Andrew Scott for their friendship and help. Abstract Structural styles of the Andean foothills, Putumayo Basin, Colombia Juan Carlos Jimenez, MSGeoSci The University of Texas at Austin, 1997 Supervisor: Randall Marrett Interpretation of seismic profiles, earthquake fault-plane solutions, radar images, and geometry of structures suggests that two different structural styles are viable alternatives for the Putumayo basin in Colombia. An eastern domain, varying in width from 4 to 13 km, might be characterized by strike-slip faulting parallel to the Andes because it exhibits similar structures to those formed in restraining bend settings, an example is the Orito fold, the largest known oil field in the basin. Correlation of seismic reflections with wells into the Orito fold and foreland indicates a post-Miocene age for this structure. Previous interpretations of contractional dip-slip movement on Andes-parallel structures, as proposed by Portilla (1991) with faults involving basement, are also viable. A 15 km-width western domain is interpreted as a region of foreland­dipping rocks uplifted above their regional level by wedging of pre-Cretaceous (?) rocks beneath known Jurassic rocks. Above the Jurassic rocks thin-skinned deformation occurs inside of the Mesozoic and Cenozoic sedimentary cover, also in the form of wedging. Mesozoic and Paleozoic (?) rocks were injected into of a late Cretaceous-early Paleocene unit composed of shale. The western domain is truncated to the west by a major reverse fault that places Paleozoic rocks over Mesozoic and Cenozoic rocks. Table of Contents List of Figures ................... ....................................................................................... x List of Tables .............. .............................................. ......................................... .... xii INTRODUCTION .................. ................................ ........ ......................................... 1 Problem and Significance .... ................................. ......................................... . 1 Location of Study Area ....... ............................................................................ 3 Previous Work on Structural Style in the Northern Andean Fold-Thrust Belt. ................................... ............................................................ ..... .. .. 4 Data Sources ................................................................................... ................ 7 REGIONAL TECTONICS ............................................... ................ ..................... 11 Tectonic Setting ......... ................................................................................... 11 History of Plate Motion ................................................................................ 13 Timing of Deformation ................................... ............ .................................. 13 Active Deformation Kinematics ................................................................... 15 New Analysis of Intracontinental Earthquake Kinematics ........................... 18 Additional Evidence for Strike-Slip Faulting ............................................... 28 STRUCTURES IN THE PUTUMA YO BASIN ................................................... 33 Reflection-Seismic Observations and Structural Interpretation ................... 33 Eastern Domain ..................... .................... .......................................... 38 Western Domain ............... ................. .................................. ................ 43 Structural models .......................................................................................... 51 Dip-Slip Faulting Model ...... .... ............................................................ 53 Strike-Slip Faulting Model ......... .............................................. ........... 57 CONCLUSIONS ................................................................................................... 61 APPENDIX A. Seismic Surveys ........................................................................... 64 APPENDIX B. Well Information ...... ............................... ..................................... 67 Bibliography ............................................................................................... ........... 69 Vita ................................................ .... .. .. .. .... .. .. ............... ...................................... 74 List of Figures Figure 1: Location map of the Putumayo basin .............................................. ... 2 Figure 2: Surface geology map .......................................................................... 8 Figure 3: Information used in this study ....... ..................................................... 9 Figure 4: Tectonic features in northwestern South America ............................ 12 Figure 5: Generalized cross-sectional model for the basin .............................. 14 Figure 6: Portion of the seismic line 0730 ....................................................... 16 Figure 7: Map of earthquake fault-plane solutions .......................................... 20 Figure 8: Analysis of thrust events ................................................................... 23 Figure 9: Analysis of strike-slip events ........................................................... . 24 Figure 10: Comparison of kinematic directions .............................................. ... 26 Figure 11: Shuttle Imaging Radar ........................... .......................... ................. 29 Figure 12: Map of surf ace structural features ............................ ........................ 31 Figure 13: Stratigraphy of formations in the basin .......................... .................. 34 Figure 14: Seismic line 1889 ..................... ......................................................... 36 Figure 15: Structural contour map near the base of Cretaceous ... ..................... 37 Figure 16: Seismic line 1460 ........ ...................................................................... 39 Figure 17: Orito fault-plane map ........................................................................ 41 Figure 18: Orito fault-plane section ................................................................... 42 Figure 19: Seismic line 2210 .............................................................................. 44 Figure 20: Seismic line P-2210 ................................. ......................................... 45 Figure 21: Seismic line 2240 ............................. ................ ................................. 46 Figure 22: Seismic line 1685 .............................................................................. 48 Figure 23: Seismic line 1245 ............................................................................. .49 Figure 24: Cross sections location ..................................................................... 54 Figure 25: Cross section #1 ........ ........................................................................ 56 Figure 26: Cross section #2 ................................................................................ 59 List of Tables Table 1: Earthquake focal mechanism solutions for intracontinental events in the northern Andes ........................................................................ 19 Table 2: Analytical results of seismic events .................................................. 22 Table 3: Kinematic directions and their 95% uncertainty cones .................... 25 INTRODUCTION PROBLEM AND SIGNIFICANCE The discovery over the last decade of giant fields at the edge of the Llanos basin sparked renewed interest in hydrocarbon exploration in the foothills of the Andean Cordillera, Colombia. Hydrocarbon accumulations in the area are structurally trapped. This study focuses on defining the structural style present in the Andean foothills near the Putumayo basin, located to the south of the Llanos basin (Fig. 1). Analysis of existing surface, well-log, and seismic-reflection data, synthesis of existing seismological data, and new analysis of space-based radar images suggest that strike-slip deformation is a viable alternative to the previously interpreted contractional deformation in the Andean foothills in southern Colombia. The possibility of significant strike-slip deformation in the foothills of the Andean Cordillera in the Putumayo basin of Colombia offers a suite of new opportunities. From the exploration and development point of view, types of plays not found in a fold and thrust belt might be recognized. Structures or prospects already defined and interpreted should be reevaluated in the light of strike-slip deformation, which might decrease risk. Processing of seismic profiles using parameters appropriate for the structural setting could improve the image quality. These considerations should also be evaluated in the case of other foreland basins like the Llanos basin. ao· 70° O' Figure 1. Location map showing the study area in the Putumayo basin, Colombia. (Modified from Amaya, 1996). Previous interpretations of structural styles in the Putumayo basin area, based on surf ace geologic, 2D seismic, and well log data, focus on shortening perpendicular to the Cordillera as a result of dip-slip on faults oriented parallel to the mountain front (Portilla, 1991; BEICIP-ECOPETROL, 1988; Western Atlas, 1995; Cordoba et al., 1997). However, geometric relationships between faults and folds, interpretation from radar images of right-lateral offsets coincident with lineaments parallel to the range, well-documented regional fault systems having range-parallel strike-slip motion, and evidence of seismicity on strike-slip faults suggest shortening occurs by strike-slip as well as dip-slip on range-parallel faults. Folding coincident with a bend in one strike-slip fault indicates inhibition of lateral motion (restraining bend). LOCATION OF STUDY AREA The Putumayo basin, covering over 50,000 km2 , is located in the southwestern part of Colombia and extends into Ecuador and Peru where it is called the Oriente and Marafion basins, respectively. In Colombia the basin is bounded to the north by the Vaupes Arc, to the west by the Andean Cordillera and to the east by the Guyana Shield (Fig. 1). This study focuses on the 15 km-wide fold and thrust belt occurring parallel to the Andean Cordillera and running along the western side of the Putumayo basin (Fig. 1). This area extends from near the border with Ecuador at 0.5° to 1 ° north latitude in Colombia, comprising over 5,000 km2• PREVIOUS WORK ON STRUCTURAL STYLE IN THE NORTHERN ANDEAN FOLD-THRUST BELT Published information regarding the structural style and evolution of the fold and thrust belt of the Putumayo basin is scarce. An unpublished Master's thesis from South Carolina University by Portilla (1991) is the most extensive and integrated structural study developed in the fold and thrust belt of the Putumayo basin. In this study, the structural style is defined based mainly on interpretation of seismic reflection profiles, several of them were used also in this study. Portilla (1991) identified three main tectonic zones: the Front Range characterized by a system of west-dipping imbricate thrusts, the Inner Foothills characterized by a system of imbricate back-thrusts and associated hanging-wall monoclines, and the Outer Foothills where asymmetric anticlines associated with high-angle reverse faults constitute the main structural features. Portilla (1991) describes the structural style of the Putumayo fold and thrust belt as a "product of the interaction of a deep foreland propagating "thick-skinned" crustal deformation and its consequent shallow "thin-skinned" deformation in the Mesozoic and Cenozoic sedimentary cover". A late Eocene to Recent contraction resulted in the formation of the fold and thrust belt (Portilla, 1991 ). Several regional studies have been undertaken in the basin, the most recent concerning the petroleum systems of the Putumayo basin (Cordoba et al., 1997). In this report the structural style is characterized as thick-skinned deformation of Paleocene age associated with "deep detachments" and subsequent reactivation. Cordoba et al. (1997) cite the Orito structure as an example of this type of deformation. The Orito structure is the one of the largest structures known in the Putumayo basin and contains the largest oil field discovered in the basin. A later thin-skinned deformation developed in the westernmost part of the basin, overprinting the older thick-skinned structures. Finally, associated with the youngest contractional event, thick-skinned deformation thrusted pre-Cretaceous igneous rocks on top of Cretaceous and Tertiary sedimentary rocks in the westernmost part of the basin. This last event of thick-skinned deformation terminated with strike-slip faulting seen in deformation of Quaternary terraces (Cordoba et al., 1997). Western Atlas International (1995) completed a regional evaluation of the Upper Magdalena and Putumayo basins. The study described the structural style in the foothills of the Putumayo basin, as "a poorly-formed duplex zone to the east" and "an imbricate zone to the southwest". This imbricate zone extends westwards beneath the Andean Cordillera. The Orito fault, the most prominent structure in the area, is classified as a high-angle reverse fault which formed during the Miocene. A regional evaluation of the Putumayo basin by BEICIP-ECOPETROL ( 1988) defined the structural features in the area as predominantly steeply-dipping basement-involved reverse faults of post-Cretaceous age. Some of these structures were interpreted as inversion of normal faults with a minor component of strike­slip movement as seen in the Oriente basin of Ecuador to the south of the Putumayo basin (Balkwill et al., 1995). Reverse faulting affected the entire sedimentary column with a predominantly eastward vergence. Strike-slip movement on east-west striking faults, crosscutting and displacing the reverse faults, was also mentioned. Other published reports describing the structural style of the Putumayo basin are based mainly on interpretation of data from the Oriente and Marafion basins in Ecuador and Peru, respectively, and extrapolated to the Putumayo basin in Colombia. Balkwill et al. ( 1995) interpreted the structures of the Oriente basin in Ecuador as small-scale examples of the type of large-scale, high-relief basement-involved foreland structures present in the Sierras Pampeanas of Argentina (Jordan and Allmendinger, 1986) and in the Laramide province of the western United States (Berg, 1962). Balkwill et al. (1995) also identified small­scale intrastratal detachments in the Oriente basin and interpreted them as "local adjustment to inversion of basement faults". They highlighted the absence, at least in seismic profiles, of the large-scale thin-skinned deformation described in the Llanos basin of Colombia (e.g. Cooper et al, 1995). A study of the petroleum systems along the foldbelt associated with the Marafion-Oriente-Putumayo basins (Marksteiner and Aleman, 1996) interpreted thick-skinned Middle Eocene age deformation (Incaic phase) followed by a renewal of uplift during the Late Miocene-Pliocene (Andean Orogeny). In summary, previous work concluded that the structural style of the Putumayo basin is predominantly basement-involved contraction in the easternmost part of the fold and thrust belt, and thin-skinned in the westernmost zone. Thin-skinned deformation was described by the different authors as occurring in the form of duplexes, imbricate zones in the southwest, and a triangle zone. Almost all deformation was interpreted to be result of dip slip, although recent strike-slip faulting was mentioned (BEICIP-ECOPETROL, 1988; Cordoba et al., 1997). DATA SOURCES Surf ace geologic data were obtained from a map compiled by Cordoba et al. (1997). The map integrates information mapped by several petroleum companies during acquisition of the various seismic surveys (Fig. 2). A data set of 16 earthquake fault-plane solutions, mainly obtained from the seismological reports of Harvard University and calculated using the centroid­moment tensor method (Dziewonski et al., 1981 ), was analyzed to characterize active tectonics in the region (Chinn and Isacks, 1983; Suarez et al., 1983; Harvard Seismology, 1997). A set of 725 km of 2D seismic lines was studied (Fig. 3; Appendix A). The data consist of digital migrated seismic profiles interpreted with a vertical scale of 3.75 inches/second, taken in twelve different surveys from 1988-1994. A 3D seismic volume acquired in 1992 was also examined (Fig. 3). The 3D survey consists of a grid of lines at 25 m spacing, covering a total area of 24 km2 • Seismic interpretation focused mainly on the 2D lines due to the limited coverage of the 3D seismic. Three synthetic seismograms and one check shot log were used to correlate seismic reflectors with formation tops. Well log data from 17 wells (Appendix B) were used to identify or confirm potential zones of faulting based on repetition of stratigraphic levels and LITHOLOGY GEOLOGICAL SYMBOLS Quaternary Terraces and Alluvium '-r-.. DEPOSITIONAL CONTACT Tertiary sedimentary rocks Volcanic rocks (Tertiary) '-4.../ THRUST FAULT Orito Gp. (Mioc.-Plio.) _._ NORMAL FAULT Pepino Fm. (Eocene) /°i./ ANTICLINAL AXIAL TRACE Rumiyaco Fm . (Cret-Paleoc) ~ Materna Fm.(Jurassic) FAULT INFERRED Villela Fm.(Cretaceous) -Sedim., metam.and igneous rocks(Paleozoic) -STUDY AREA Caballes Fm .(Cretaceous) l!!iLJ Metamorphic rocks (Precambrian) Figure 2. Geologic map of study area in the Putumayo basin. (Modified from Cordoba et al., 1997). 8 Figure 3. Map showing location of data used in this study. 9 anomalies in thickness. A common log suite includes gamma ray, sonic, and resistivity curves. A radar image at approximately 1 :300,000 scale was used to interpret geomorphic features and correlate it with the geologic map to identify apparent strike-slip offsets. The radar image is from the high resolution ( 12 meters per pixel) SIR-A (Shuttle Imaging Radar) acquired in 1987 showing an oblique view (looking towards the west) in panchromatic mode. REGIONAL TECTONICS TECTONIC SETTING The Putumayo basin is a foreland basin located east of the Andean Cordillera in Colombia, in the northwestern corner of South America (Fig. 4 ). Its tectonic evolution through time, as well as that of other sub-Andean basins, has been related to the convergence of the Nazca and South American plates since Late Cretaceous times (Pardo-Casas and Molnar, 1987). The boundary between the Nazca and South American plates is a convergent margin defined by the Ecuador-Colombia trench, a Wadati-Benioff zone (a zone of shallow (< 70 km depth) and intermediate to deep (> 70 km depth) seismic hypocenters), and the Andean magmatic arc. Active movement between the Nazca and South American plates is reflected by seismic activity in the area. The subduction zone is defined by a zone of shallow earthquakes that parallels the western coastline of Ecuador and southern Colombia (Pennington, 1981; Chinn and !sacks, 1983). A series of earthquake focal mechanisms show nodal planes striking parallel to the trench and include normal and thrust events typical of a subduction zone environment (Dewey, 1972; Pennington, 1981; Chinn and Isac ks, 1983; Suarez et al., 1983; Ego et al., 1996). Another belt of seismic epicenters is concentrated in the Andean Cordillera, predominantly along its easternmost flank (Chinn and !sacks, 1983; Suarez et al., 1983). 10' O' 4• so· CARIBBEAN PLATE rate= 18± 6 mm/yr NAZ CA PLATE 3• ~ rate=60 mm/yr so· 100 km L-.....1 70° 10' O' Figure 4. Tectonic features in northwestern South America. The dark arrows show relative movement between each plate, microplate or block with respect to stable South American plate (1) Villavicencio station, (2) Bogota station (3) Malpelo Island station and (4) San Andres Island station. Measurements from 1991to1996 GPS campaigns (Kellogg and Trenkamp, 1997). HISTORY OF PLATE MOTION Convergence between the N azca and South American plates has been proposed as the mechanism to explain the uplift of the Andean Cordillera (e.g., Pardo-Casas and Molnar, 1987). Based on plate reconstruction from magnetic anomalies and oceanic fracture zones, Pardo-Casas and Molnar ( 1987) calculated a rate of convergence of about 100 mm/yr in an east-west direction during the Late Cenozoic. Changes in convergence rates, with faster periods during Late Eocene and Miocene-Pliocene times, are correlated with two pulses of relatively intense tectonic activity in the Peruvian Andes (Pardo-Casas and Molnar, 1987). GPS (Global Positioning System) constraints on active interplate motion, show convergence between the Nazca and South American plates of about 60 mm/yr in an east-west direction since 1991 (Kellogg and Trenkamp, 1997). Comparison of convergence data obtained by GPS and plate reconstruction methods reveals a reduction in convergence rate. This change in the kinematics between plates is not discernible in the plate reconstruction method because limitations of resolution. TIMING OF DEFORMATION A Triassic-Jurassic rifting event is widely recognized in the eastern basins of Venezuela, Colombia, and Ecuador (e.g., Mada and Mojica, 1981) (Fig. 5). In Colombia red beds, volcanic flows and marine sedimentary rocks represent the most important deposits formed during this rift phase (Saldana or Motema Fm.) (Portilla, 1991). During Cretaceous times a contractional tectonic event at about WEST EAST Andean (c) Cordillera ~ Putumayo basin l< x l< x " l< " ~ x X' l( ;l,lll; JC Late Eocene-Present (b) Andean Cordillera Putumayo basin xx " l(, x " " x x x " x x ~ ...­" " .,i:.. Cretaceous-Early Tertiary Paleo Andean (a) Cordillera x l< x x X'. x x x " Middle Paleozoic-Early Cretaceous Figure 5. Generalized cross-sectional model for Middle Paleozoic-Present evolution of the Putumayo basin. (a) deformation dominated by rifting, (b) deposition of marine sediments and continental sediments at top non-evi­dence of important deformation in the Putumayo basin, ( c) principal contractional event recorded in sedimentary rocks in the Putumayo basin. (Modified from Cordoba et al., 1997). 120 Ma produced deformation in the Central Cordillera in Colombia (Colletta et al., 1990). Folding, thrusting and uplifting of the Eastern Cordillera has been attributed to a strong shortening event during the Neogene (Andean Orogeny) caused by the convergence of the Panama block with South America (Mann and Burke, 1984). Major structures observed in the seismic profiles from the Putumayo basin (e.g., the Orito structure) involve stratigraphic levels younger than those dated as Oligocene-Miocene by Cordoba et al. ( 1997) (Fig. 5), therefore a good upper age constraint for this deformation is not available. A succession of parallel and harmonically-folded seismic reflectors is correlated in the foreland with stratigraphic units from Cretaceous to Middle Miocene in age (Fig. 6). These major structures were possibly formed during Miocene-Pliocene times, as proposed in the Llanos basin of Colombia to explain the strong shortening event in the Eastern Cordillera (Colletta et al., 1990) and shortening in the Peruvian Andes (Megard et al., 1984 ). Because deformation observed in the Putumayo basin is possibly younger than Middle Miocene age and plate-reconstruction resolution is poor for this time period, the relevant kinematic setting is best inferred from GPS and earthquake data (i.e. active deformation). ACTIVE DEFORMATION KINEMATICS Interpretation of intraplate earthquake focal mechanisms by Pennington (1981) concluded that right-lateral strike-slip faulting is occurring along the eastern border of the Andean Cordillera in Colombia. Colletta et al. (1990), however, interpreted this seismicity to the south of their study area as related to South L-0730 S.P 550 650 750 0.0 __,,/ Quater~y alluvium ~~~ _ ~?, ___ --~ ~.----Interval representing-~~~ ' ~ ~cks of Oligocene------.::._---_ Interval representing \ 1ocene age ----------=::-----rocks of Oligocene­ ~ ~~~~~~--------~ \ =---~--..::..___ ~ Mioceneage ----~-------~ ---=:... 1.0 ------,.. ~\ ------~ ~,Nearto LateEocene Near tOJ) Late Eocene ~· ~~ ~ __L:::::::;---'O \ i ,. --::::::-...... ~ --------------­ ~' e;q.b:::::-....-.....:::: . ~ .... ....., ilse ....,. .Q;,e~e 2 ~ear base Cretaceous .0 11 km 1 l=N=e=ar=ba=s=e=c=r=e=ta=c=eo=u=s~~:::; South North Figure 6. Portion of seismic line L-0730 shows Orito anticline. Notice deforma­tion of Oligocene-Miocene strata. Stratigraphic units correlated with wells info­reland to the east of section. the obliquity of the chain with respect to the shortening direction, and although mountain belt the particular area of their study has the same trend, the shortening direction was assumed to be perpendicular to the front of the Eastern Cordillera and dip-slip movement was assumed to dominate (Colletta et al., 1990). Conclusions differing from those presented by Pennington ( 1981 ), also based on interpretation of intracontinental earthquakes, were given by Chinn and !sacks (1983) and by Suarez et al. (1983). These authors proposed that shallow focal depths ( < 70 km) combined with mainly dip-slip focal mechanisms indicate that much of the crustal deformation in the eastern Andes is occurring in the form of basement-involved horizontal shortening perpendicular to the mountain range. Although, Chinn and !sacks (1983) recognized that some right-lateral strike-slip movement also occurs in the eastern Andes in Ecuador and southern Peru. Intraplate motion determined from GPS campaigns from 1991 to 1996 shows relative movement between the Andean Cordillera and the South American plate in a northeast direction at a rate of 8 ± 2 mm/yr (measured between Bogota and Villavicencio in Colombia), Similar results were found nearer to the Putumayo basin between the city of Pasto and Villavicencio (Kellogg and Trenkamp, 1997). Active right-lateral strike-slip faulting occurring along the eastern border of the Andean Cordillera in Colombia is consistent with both the interpretation based on earthquake focal mechanisms by Pennington ( 1981) and these GPS data. NEW ANALYSIS OF INTRACONTINENTAL EARTHQUAKE KINEMATICS Because many new data are now available on the seismicity of the northern Andes, I have undertaken a new analysis of intracontinental earthquake kinematics. Sixteen fault-plane solutions from intracontinental earthquakes (Table 1) were examined to observe the recent activity in the Andes of Ecuador and central-southeastern Colombia, 11 of which were not available to Suarez et al. (1983) or Chinn and !sacks (1983). None of the earthquakes occurred in the Putumayo basin or in the portion of the Andean Cordillera bounding it. However, several of the earthquakes centered to the south and to the north (Events 1, 2, 4, 6, 8, 14 and 15) indicate that strike-slip faulting is occurring in a broad and elongated zone ranging from the eastern area of the Andean Cordillera to the foreland of the Llanos basin, and from Ecuador to central Colombia passing along the Putumayo basin (Fig. 7). A graphical kinematic method was used to analyze the earthquake data (Marrett and Allmendinger, 1990). The graphical kinematic analysis uses the principal axis orientations of average incremental strain (P-and T-axes for shortening and extension, respectively) and obtains directional maxima of the shortening and extension axes from Bingham distribution statistics, which assume that to a first order fault kinematics are independent of fault size (Marrett and Allmendinger, 1990). The moment tensor summation method was not used because it typically weights the data such that the biggest event dominates the results, whereas the uniform weighting of the data in the Bingham distribution Event Source number Suarez et al., 1983 day month 10 5 year 63 reference Seismic moment Lat No. (dyne*cm) 0 N 8.85E+25 -2.20 Long °W 77.60 depth km 16 sense fault of slip ;trike(rhr: NR 230 fault dip 88 Slip trend 50 Slip T-axis plunge trend 4 5 T-axis plunge I P-axis trend 95 P-axis plun.e;e 4 2 Chinn&lsacks, 1983 9 2 67 32 4.77E+26 2.93 74.83 36 TR 20 62 23 6 340 24 243 15 Suarez et al. , 1983 23 2 73 13 9.47E+24 -2.16 78.33 10 TL 156 28 293 20 322 60 98 23 4 Chinn&lsacks, 1983 27 9 74 45 l.OOE+25 2.72 71.37 44 NR 225 89 45 0 0 90 5 Chinn&lsacks, 1983 6 10 76 46 1.00E+25 -0.76 78.75 33 TR 13 20 71 18 56 61 259 27 6 Harvard CMT 31 3 83 B033183A 3.51E+24 2.45 76.81 56.8 TR 26 76 27 5 342 13 251 6 7 Harvard CMT 19 5 83 B051983A 8.10E+23 0.11 76.63 23 TR 60 49 99 36 39 62 304 3 ...... \0 8 9 Harvard CMT Harvard CMT 6 6 3 3 87 87 B030687D B030687A l.17E+25 4.90E+25 0.31 0.10 77.73 77.77 15 15 NR TR 226 198 40 20 35 258 9 17 185 243 26 61 71 86 41 27 I0 Harvard CMT 6 3 87 C030687B 6.37E+26 -0.06 77.84 15 TR 195 27 277 26 269 71 100 19 11 Harvard CMT 22 9 87 C092287B 4.07E+25 -0.89 78.24 15 TR 218 42 244 22 201 52 90 15 12 Harvard CMT 22 9 87 B092287C 1.07E+25 -0.98 78.24 19.4 TR 197 42 240 31 189 63 80 9 13 Harvard CMT 11 8 90 B081190A 9.45E+23 O.Ql 78.15 15 TL 323 45 100 35 157 64 258 5 14 Harvard CMT 26 12 92 Bl22692E 7.00E+24 -1.15 77.92 15 TR 200 46 210 10 171 38 63 22 15 Harvard CMT 6 6 94 C060694J 1.84E+26 2.93 75.94 15 TR 206 76 209 10 163 17 72 3 16 Harvard CMT 25 8 96 B082596B l.83E+24 1.12 77.99 15 TR 172 48 210 35 152 61 55 4 Table I: Focal mechanism data for intracontinental earthquakes in the northern Andes and adjacent foreland between 3°N and 3°S. Explanation of symbols: N=normal, T=thrust, L=left, R=right, (rhr)=right hand rule, T-axis=extension axis, and P-axis=shortening axis. Fault planes chosen as nodal plane most parallel to major faults identified by surface geology and structural contour maps. 80° 70° CARIBBEAN PLATE NAZCA PLATE • 10° VENEZUELA COLOMBIA ~4 PERU Figure 7: Map of earthquake fault-plane solutions. Earthquakes shown are sha­llow focus ( <70 km) intracontinental events. Event numbers correspond to those in Table 1. oo analysis provides an average of the kinematics. This is appropriate because of the geologically short time period represented by the instrumental seismic record. Due to their heterogeneous kinematics, the earthquakes were separated into three groups (Table 2): events with extension axes plunging steeper than 45° (Fig. 8), events with shortening axes plunging steeper than 45° and events with both shortening and extension axes plunging shallower than 45° (Fig. 9). These three groups represent thrust (N=9), normal (N=O), and strike-slip events (N=7), respectively. Plots of shortening (P) and extension (T) axes for each group of events were constructed and Bingham distribution statistics were used to determine directional maxima of the shortening and extension axes (Figs. 8 and 9). Pseudo-fault plane solutions were also plotted to represent the average kinematics of each group of events (Figs. 8 and 9). For thrust events, the maximum shortening axis plunges shallowly to the east and the extension maximum plunges steeply to the southwest (Fig. 8). The shortening axis trend for thrust events approximately coincides with the east-west convergence direction between the Nazca and South American plates calculated from plate reconstruction (Pardo-Casas and Molnar, 1987) as well as from GPS measurements (Table 3 and Fig. 10) (Kellogg and Trenkamp, 1997). The majority of thrust earthquakes occurred at depths ranging from 15 to 44 km in Ecuador along the central part of the Andean Cordillera (Fig. 7), and not along the eastern part of the Andes as described by Suarez et al. (1983). The maximum crustal thickness calculated along the Trans-Andean Geophysical Profile in south-central Colombian Andes is about 45 km for the central part of the Andean Cordillera and Group Events N Eigenvector l Eigenvector 2Eigenvector 3Eigenvalue I Eigenvalue 2Eigenvalue 3 Nodal Plane Nodal Plane Thrust Strike-slip Normal 3,5,7,9, 10, 11, 12, 13, 16 1, 2, 4, 6, 8, 14, 15 None 9 7 214,81 358, 7 89,5 0.35 0.02 -0.40 187,40 352,51 169, 6 299,80 79, 7 0.41 0.02 -0.42 214,80 124,89 N N Table 2. Analytical results of three groups of seismic events from Table 1. N indicates number of events. Eigenvectors are maximum kinematic axes calcu­lated by Bingham distribution statistics. Eigenvectors 1 and 3 indicate maximum extension and shortening orientations respectively. Eigenvalues represent magnitude of eigenvectors. Nodal planes represent the fault plane and the auxiliary plane perpendicular to the slip direction. Figure 8: Analysis of thrust events. (a) Shows faults, slip lineations, and arrows indicate sense of slip. (b) Shaded and white dihedra re­present average kinematics, with shortening (P) and extension (T) axes of individual events shown as diamonds and open boxes, res­pectively Figure 9: Analysis of strike-slip events. (a) Shows faults, slip linea­tions, and arrows indicate sense of slip. (b) Shaded and white dihe­dra represent average kinematics, with shortening (P) and extension (T) axes of individual events shown as solid diamonds and open bo­xes, respectively. Kinematics Method Source Movement azimuth 95% Uncertainty Cones as Half Apical Angles Nazca-South America plate motion Plate reconstruction [a] 90° 12° Nazca-South America plate motion GPS [b] 86° 20 Andean Cordillera-South America plate motion GPS [b] 37° 130 Earthquake shortening axis (thrusts) Bingham analysis of fault-plane solutions [c] 88.5° 10.2° Earthquake shortening axis (strike-slip) Bingham analysis of fault-plane solutions [c] 78.6° 10.1 ° Slip direction of strike-slip earthquakes Bingham analysis of fault-plane solutions [c] 34.0° 5.2° Table 3. Kinematic directions and their 95% uncertainty cones. GPS=Global Positioning System. Sources: [a] Pardo-Casas and Molnar (1987); [b] Kellogg and Trenkamp (1997); [c] this study. Uncertainty values for plate reconstruction and GPS results direc­tly from ellipses in published figures (Pardo-Casas and Molnar, 1987; Kellogg and Trenkamp, 1997). Uncertainty values for earthquake kinematic data determined using equations for estima­tion of the principal axis of a symmetric bipolar distribution of axial data (Fisher et al., 1987). Figure 10: Comparison of kinematic directions and their 95% uncertain­ty cones on lower hemisphere equal angle net. (a) Represents Nazca­South America plate convergence direction (from Pardo-Casas and Mol­nar, 1987). (b) and ( c) Represent relative motion between Nazca-South American plates and Andean Cordillera-South American plate, respec­tively, determined from GPS (Kellogg and Trenkamp, 1997). (d) and (e) Represent earthquake-kinematics shortening (thrust and strike-slip, res­pectively). (f) Represents the slip direction of strike-slip earthquakes. Eartquake analyses were determined using linked Bingham analyses of fault-plane solutions for shallow intraplate events between 3°N and 3°S (Chinn and !sacks, 1983; Suarez et al., 1983; Harvard Seismology, 1996). approximately 25 km for the western part of the Putumayo basin (Case et al., 1973). These earthquakes show that basement-involved faulting is occurring in the central part of the mountain range. If shortening in shallow detachments is occurring, it is doing so aseismically or mostly in events of small magnitude. For strike-slip events, the maximum shortening axis plunges shallowly to the east-northeast and the extension maximum plunges shallowly to the south­southeast (Fig. 9). Shortening axis trend is parallel to the E-W convergence-vector direction between the Nazca and South American plates (Pardo-Casas and Molnar, 1987; Kellogg and Trenkamp, 1997). However, the slip directions of strike-slip events agree well with the movement direction between the Andean Cordillera and stable South America determined using GPS (Table 3 and Fig. 10) (Kellogg and Trenkamp, 1997). Strike-slip events are mainly located at depths ranging from 15 to 57 km in the eastern part of the Andean Cordillera and in the foreland of the Llanos basin. The preferred nodal planes, striking almost parallel to the mountain range, suggest that strike-slip movement is occurring parallel to the Andean Cordillera and involving basement. Basement-involved faulting is characteristic of strike-slip faults (Sylvester, 1988). In summary, comparison of late Miocene to present Nazca-South America plate kinematics derived from plate reconstruction (Pardo-Casas and Molnar, 1987), recent ( 1991-1996) kinematics derived from GPS measurements between Nazca-South America and Andean Cordillera-South America (Kellogg and Trenkamp, 1997), and historic earthquake kinematics in the Ecuador and Colombian Andes reveals the following characteristics (Fig. 10). The cones of 95% confidence for Late Miocene to present Nazca-South American plate convergence directions calculated from plate reconstruction and recent GPS campaigns, and historic earthquake shortening for thrust and strike-slip events overlap. The cone of 95% confidence for historic earthquake slip directions for strike-slip events overlaps the cone of 95% confidence for GPS derived Andean Cordillera-South American plate motion, but these directions are statistically distinct from the other kinematics studied. ADDITIONAL EVIDENCE FOR STRIKE-SLIP FAUL TING Current literature makes little mention of strike-slip faults parallel to the mountain range bounding the Putumayo basin. Cordoba et al. ( 1997) mention recent strike-slip faulting evidenced by "tilting of terraces, scarps, and offsets of streams". In the study by BEICIP-ECOPETROL (1988), strike-slip faulting is mentioned as occurring on a north-south striking fault system present in the foreland of the Putumayo basin and is assigned a pre-Tertiary age. Strike-slip faulting has also been interpreted in the form of "lateral ramps or transfer zones" associated with the latest major thrusting event (Cordoba et al., 1997). I interpreted a radar image of the Putumayo basin from the northeastern part of the study area (Fig. 11). Geomorphic features in the images were correlated with the geologic map to guide interpretation. Apparent right-lateral offsets of Mesozoic and Cenozoic rocks are aligned with long linear canyons in the mountain range (sites 1 and 2 in Fig. 11). These long linear NE-trending features are interpreted as apparent right-lateral strike-slip faults striking parallel to the mountain front. Figure 11. Shuttle Imaging Radar (SIR-C) showing apparent dextral offset in an upper Cretaceous-Lower Paleocene unit (KTr) (1), and in an older Cretaceous unit (Ksm) (2). Well-documented regional evidence for right-lateral strike-slip fault systems is found to the west of the study area in the Andean Cordillera. The southern continuation of the Garzon fault, approximately 30 km to the west of the study area (Fig. 12), bounds the southeastern side of the Upper Magdalena basin and strikes parallel to the mountain range. This fault was interpreted as a right­lateral strike-slip fault from radar images based on identification of "right­stepping releasing bend basins along the fault" (Chorowicz et al., 1996). A slip rate of 1.5 mm/yr during the last 2 Ma was calculated based on the presence of triangular faceted scarps, offsets of the piedmont slope estimated to be 3 km and offsets of small active alluvial fans ranging from 300 m to 1 km (Chorowicz et al., 1996). South of the Putumayo basin in Ecuador, the Garzon fault continues into the Chingual-La Sofia fault (La Bonita fault) (Fig. 12), another system of active right-lateral strike-slip faults. Movement of the Chingual-La Sofia fault was calculated at a rate of 7.0 ± 3.0 mm/yr between 37,000 ± 630 and 8,600 ± 60 yr BP (Ego et al., 1996). The most prominent offset produced by the Chingual-La Sofia fault is in the Rio Chingual, which is offset by 7.5 to 10.5 km (Ego et al., 1996). Towards the south, the Chingual-La Sofia fault steps to the western side of the Andean Cordillera where it is called the Pallatanga fault (Fig. 4). Movement along the right-lateral Pallatanga fault was estimated from offset glacial deposits at a rate of 4.0 ± 1.0 mm/yr during the last 10,000-12,000 yr (Winter and Lavenu, 1989). GEOLOGICAL SYMBOLS A VOLCANO ~ANTICLINAL AXIS '""'--./ THRUST FAULT STUDY AREA _..__ NORMAL FAULT COUNTRY LIMIT --FAULT INFERRED The zone between the Pallatanga fault and the Chingual-La Sofia faults is called the "Inter-Andean Depression" and is interpreted as a restraining bend (Ego et al., 1996). The restraining bend is bounded by right-lateral strike-slip faults striking approximately northeast-southwest, parallel to the trend of the Andean Cordillera. Characteristic structures present in the restraining bend include asymmetric folds with north-south trending axes and opposing vergence (Lavenu, 1994). STRUCTURES IN THE PUTUMAYO BASIN As shown in the previous chapter, regional and local evidence suggests that important strike-slip faulting is occurring parallel to the Andean Cordillera. In this chapter strike-slip faulting is considered as a possible explanation for specific structures inside the Putumayo basin, in terms of the expression of these structures in 2D seismic profiles and in map view. Structural cross sections were constructed to test two models of structures in the Putumayo basin: (1) pure dip­slip fault movement and (2) oblique-slip faulting with important vertical and horizontal components of movement. REFLECTION-SEISMIC 0BSERVA TIONS AND STRUCTURAL INTERPRETATION The Putumayo basin has a known sedimentary fill of about 4 km in thickness composed of Mesozoic and Cenozoic rocks (Fig. 13). Paleozoic and Precambrian crystalline rocks have not been penetrated by wells on the basin (Fig. 2), but high grade metamorphic rocks, mainly gneiss, crop out in fault contact over Cretaceous and Tertiary rocks (Case, et al., 1973). However, Paleozoic sedimentary rocks are suspected to exist below Mesozoic rocks in the Putumayo basin because they were drilled approximately 100 km to the south in Ecuador (Sacha-deep well), where they consist of elastic sequences topped by limestones (Kirkpatrick, 1994 ). Mesozoic rocks are described in outcrops along the western side of the study area as intercalated igneous intrusive and volcanic rocks of Middle Jurassic age and sedimentary rocks of continental origin (Mojica and Llinas, 1984), VJ .j::. representing the oldest units drilled in the Putumayo basin. The se1sm1c expression for the top of Jurassic strata is characterized by the deepest, strong and continuous seismic reflection, present in most of the seismic profiles (Fig. 14). Cretaceous rocks consist of shales, limestones and minor sandstones, found in outcrops and wells throughout the basin. A basal Cretaceous sandstone dated as Aptian-Albian (Garzon and Rueda, 1994; Morales et al., 1996 a and b) is the main hydrocarbon reservoir in the basin. Due to its economic importance a structural contour map near the base of Cretaceous was constructed (Fig. 15). The main source rock is Albian-Maastrichtian in age (Morales et al., 1996 b). Basal­Cretaceous rocks have a strong and continuous seismic-reflector pattern that is easily correlated throughout the seismic profiles. Cenozoic rocks are constituted by continental deposits with thick shaly units at the base. This shaly unit is dated as Paleocene-Late Cretaceous (Cordoba et al., 1997) and has poor seismic expression (Fig. 14). Farther up in the section a very strong and continuous seismic reflection in the Cenozoic interval was correlated to a Late Eocene conglomeratic package intercalated with shale dated by Cordoba et al. (1997). The uppermost interval, dated as Miocene, is a thick interval of discontinuous seismic reflections (Fig. 14). Based on the seismic patterns observed, two different structural styles are present in two domains of the fold and thrust belt in the western part of the Putumayo basin (Fig. 15). An eastern domain comprises structures associated with steeply-dipping faults that possibly involve basement rocks, the largest of which is the Orito structure (Fig. 14). A western domain comprises foreland­ WEST EAST ORITO ANTICLINE FORELAND L-1889 ;?;~ S.P 150 200 250 300 350 400 450 500 ~~ ~ -=..._ =. Oligocene 10 present interval -= --=--=----=._--_ ......_ ~ _ _ ---------~ ------........ ~"' L­-_ NeartoplateEoccne ---=---------------~ ~~........... ~-,___ ...£ --=--................. _ r-==_ -~ ---------=--=------=------~~,__ 4r-= ­ -----------=-=--------=------..... -..........~......... -::..._ --~==---1.0 ~ ~~~~~~==~~y~N~c~a~rt~o~p~la~tc~Eoc~c~:-c Near base Cretaceous --t<\. ~------= ~ -~~ --=--~---=-­ ~o~~ -----­ ~,... ~ 2.0 l1kml O UC Near base Cretaceous \..;.) 0\ ; ---------------~Legend 10 km Fault ~ Contour in m above sea level. Producing well • Contour interval 250 m. Non-producing or dry well ' Figure 15. Structural contour map in depth near the base of Cretaceous. Dashed line through map separates structural domains. Shaded areas are structural highs. 37 dipping strata uplifted from their regional level that become sub-horizontal in the westernmost part of the domain (Fig. 16). Eastern Domain The dominant structure in the eastern domain is the Orito structure (Fig. 15), which contains the largest fold in the Putumayo basin. Seismic expression of the Orito fold is represented by a curving of parallel and approximately continuous seismic reflectors representing the base of Cretaceous to Miocene section (Fig. 14). The Orito fold, with a dominantly NNW-SSE trending axis, is approximately 30 km long, varying in width from 15 km in the southwestern part of the structure to only 4 km in the northeastern part. The fold has a gently dipping (5°) southwestern limb and a steep (45°) northeastern limb (Fig. 15). A very steeply dipping east-southeast flank of the Orito fold is cut by the Orito fault. The north-northwest flank is limited by the San Antonio-Caribe fault (Fig. 15). Over 1,600 meters of structural relief separate the Orito fold from a relatively undeformed area toward the foreland, considered in this study as the regional level for the stratigraphic sequence. The Orito fault is imaged in seismic profiles as a 1-km wide "noisy" zone where seismic reflector continuity is lost. Evidence for the existence of the fault is displayed in Figure 6, where the hanging wall and footwall of the Orito fault are shown. This fault zone can be traced in the seismic profiles from the surface down to approximately 2.0 seconds (two-way travel time) (Fig. 14). A strong velocity pull-up of seismic reflectors in the footwall of the fault is characteristic. The Orito fault is a foreland-vergent, NNE-SSW striking, major fault with an important NORTHWEST SOUTHEAST L-1460 S.P 2~0 Pz I Romerillo-1 v NORTHWEST SOUTHEAST reverse component of movement and a maximum throw of 800 milliseconds (over 2,000 meters) (Fig. 14). The Orito fault is a nearly straight-mappable feature for a distance of more than 20 km. The strike of the Orito fault forms angles of 50° with both the trend of the Orito fold axis (Fig. 15) and the trend of the Andean Cordillera. Along strike towards north the Orito fault steps to the left into another fault, called here the Orito north fault. An approximate separation of 4 km is observed between the northern tip of the Orito fault and the southern tip of the Orito-north fault (Fig. 15). Structural contours on the Orito fault plane (Fig. 17) near the northern tip show a very steeply dipping plane that lose displacement and dies out near the top of the footwall Upper Eocene reflector. Towards the south, the fault plane above the Upper Eocene footwall cutoff continues steeply dipping and apparently flattens downward. Between lines L-1889 and L-187 5 the fault plane changes its geometry by flattening upward and becoming steeper dipping downward. South of line L-1875 the fault-plane dip becomes constant at about 45° (Fig. 17). A fault-plane section of the Orito fault was constructed by projecting the hanging wall and footwall cutoffs of seismic reflectors to a vertical plane, following the method described by Allan (1989) (Fig. 18). The purpose of this fault-plane representation was to describe the along-strike behavior of displacement on the Orito fault. The footwall cutoffs remain at the same depth (the same two way travel time values) along strike (Fig. 18). This strongly suggests, as said before, that apparent uplift in the footwall of the Orito fault as z -- Ul ~ ~-~~ ~,;::;,.-:...----­ =---=~70on--~---:-:.--­ ~ "L501J ­ ....... .::::_-~"" -------~-:....-::.~--------. -----<.11QJ)---~· ­ I --_ lso0-~~~1---__ ··~1000-:::~~~~-_ l,-07'0 ~--.~---=_:;I~ ' Surface !~::::::::::::='" -­ ~ ·~ I I <.I• ls C('J 00 '!:> - ... 1~ I~ i:l lt--.l 0 tv 1-J N 0 1­0 0 0 Figure 17. Ori to fault-plane map. Solid lines are subsurface depth-structural contours, and dashed lines indicate footwall cutoffs. Notice absence of the Orito fault in the line L-P2210 .p... N SOUTH SEISMIC LINES 0 V) 0 0 N c-00 0 00 00 °' 00 N -N N N I I I I ~ .....i .....i .....i .....i I I I I I "1 .. -cocei\e "Ne~to\'. '-'a~e P ------­ Near ton Late Eocene ---------... -----~.________ -­ ~------------------~-~ --­ f"'..pfac.eous --:=,Near base Cretaceo .... ._ ____ ----~~-~------~~~--~---~~ ~~~-~~~~~~----~~-------~~~­ Near base Cretaceous ---­ IO km NORTH 0 0 ­ N N N N N p.. I I .....i .....i I I --0.0 f/J ~ u i:.il f/J ~ --1.0 i:.il ~ --2.0 Figure 18. Orito fault-plane section. Solid lines represent hanging-wall cutoffs, dashed lines represent footwall cutoffs. Notice the structural low in the hanging wall at line L-1889 and the loss of throw in the fault towards line L-2220. shown in Figure 14 is due to velocity pull-up effect. The hanging-wall cutoffs of the upper two reflectors form a structural low in time in the center of the section (Fig. 18). This is interpreted as a change in the dip of the fault. After the fault cuts the lowest two reflectors it flattens through the upper two reflectors (Fig. 17). Another characteristic of the Orito fault is the loss of throw along strike as it disappears northward near line L-2220 (Fig. 15). A set of seismic profiles organized from south to north and oriented perpendicular to the strike of the Orito and Orito north faults (Figs. 19, 20 and 21) illustrates the tips of both faults and the zone between both tips. The fault zones, seen clearly in lines L-2210 (Orito fault) and L-2240 (Orito north fault) (Figs. 19 and 21), exhibit the typical pattern characterized by discontinuous reflectors and seismically noisy zones. However, seismic profile L P-2210 (Fig. 20), located between the lines described before, shows continuous and parallel seismic reflectors of the Orito fold between the two faults. The fault zones are absent in this line. The geometric relationship between the very steeply dipping northeastern limb of the Orito fold with the stepover between faults is similar, at least morphologically, to folds formed as the result of shortening in a restraining bend between two strike-slip faults. Western Domain A 15 km-wide domain extends westward from the eastern domain section to the mountain front. In general, an almost continuous and parallel panel of foreland-dipping seismic reflectors characterizes this broad zone. However, some local differences are observed along the western termination of the seismic t WEST EAST WEST EAST L-2240 L-2240 S.P 100 150 I I -====-:_ ....__ '-.....' ,_ ..... °' ~ L'km I ~ profiles. Three sub-domains are differentiated. The northern sub-domain is characterized by foreland-dipping seismic reflectors (Fig. 22) truncated to the west by the Sucio fault (Fig. 2 and 15). The central sub-domain is characterized by a similar foreland-dipping panel of seismic reflectors which flattens out to the west (Fig. 16). Farther to the west, the horizontal seismic reflectors are truncated by the Sucio fault. In the southern sub-domain, the foreland-dipping panel of seismic reflectors diverges in the westernmost part of the seismic profile (Fig. 23). A wedge-shaped zone separates an upper sequence of parallel foreland-dipping reflectors from a lower group of reflectors that continues sub-horizontally below the wedge (Fig. 23). The upper panel of foreland-dipping seismic reflectors in the southern sub­domain (Fig. 23) was correlated based on surface geology (Fig. 2) with units dated Late Mesozoic (?) to Cenozoic in age (Cordoba et al., 1997). The lower panel is correlated with Mesozoic rocks (mainly Cretaceous) based on observations in the Bagre west-1 well (Fig. 15) (Cordoba et al., 1997). Seismic resolution inside the wedge is poor, but outcrops of units within the wedge were identified as Cretaceous in age and mapped as dipping towards the foreland at angles close to 40° (Cordoba et al., 1997) (Fig. 2). The stacking of Mesozoic units inside the wedge and Mesozoic units beneath the wedge strongly suggests wedging occurred within Upper Mesozoic-Lower Cenozoic shale (Villeta?­Rumiyaco FMS.). Two interpretations were considered as alternatives to wedging to explain the tilting of the 15 km-wide western domain, but due to inadequate subsurface NORTHWEST SOUTHEAST L-1685 S.P l?O 3po 3?0 4po 4?0 5po "' -o.o I· i km .1 NORTHWEST SOUTHEAST L-1685 0.0 1.0 Figure 22. Seismic line L-1685 shows a panel of foreland-dipping seismic reflec­tors truncated to the west by a major thrust. Picks in the upper section are indica­ted by black triangles in the right side of seismic profile. NORTHWEST SOUTHEAST NORTHWEST SOUTHEAST L-1245 information I prefer the wedge interpretation because it is the simplest explanation of the structure. The first alternative interpretation is based on uplift of basement blocks similar to those found in the eastern domain. However, it is difficult to explain in such interpretation why the reflectors over a basement structure in the western domain behave in a complete different manner from those in the eastern domain, although the Mesozoic and Cenozoic stratigraphy is similar. Different basement rocks where faults are propagating at depth cannot be ruled out, but basement characteristics below these domains are unknown. Another alternative interpretation considered was based on duplexes. However, the only structure in front of the western domain is the Orito structure, which would be required to account for all the eastward displacement occurring below the 15 km-wide western domain. This implies a greater contraction in the eastern domain that can be justified. Based on evidence from the seismic profiles, the structure of the western domain of the folded belt is interpreted to consist of wedging at two different stratigraphic levels. The upper wedge, found only in the southern sub-domain, contains Mesozoic and Paleozoic (?) rocks and might be bounded by floor and roof thrusts that propagated along Upper Mesozoic-Lower Cenozoic shale (Fig. 23). The lower level is required to explain uplift of foreland-dipping rocks throughout the western domain and is interpreted as wedging involving pre­Cretaceous rocks. Cretaceous and younger rocks were interpreted to occur only in the thrust sheet above the roof thrust of the lower wedge. The sub-horizontal rock panels present in the structurally highest position of the seismic profiles and the absence of a major backthrust emerging at the surface along the seismic lines support the interpretation of a wedge bounded by a passive roof thrust that propagated beneath the oldest known stratigraphic unit in this part of the basin. As a consequence of the low seismic resolution in the pre-Cretaceous rocks, floor and roof thrusts for the lower zone wedge were not identified. However, as mentioned previously, sedimentary rocks of Jurassic-Triassic and Paleozoic ages are found in the Oriente basin in Ecuador. These strata, expected to occur in the Putumayo basin, might serve as potential detachment zones. STRUCTURAL MODELS The fold and thrust belt in the Putumayo basin can be separated into eastern and western domains (Fig. 15) having different styles of deformation. The structural style in the eastern domain is characterized by folding as response to slip on steeply dipping faults. Uplift and vertical offset of stratigraphic units along the Orito and Orito north faults is explained by a dip-slip component of movement. The northern end of the Orito fold is interpreted as a response to a horizontal-slip component of movement on the same faults. The structural style in the western domain, characterized by uplift and tilting of the 15 km wide belt of foreland-dipping rocks, is interpreted to be the result of injection of intercutaneous wedges along locally thin-skinned faults into the foreland stratigraphic sequence. The western domain exhibits thin-skinned triangle-zone geometry as defined by Price (1986). Wedging occurs at two different stratigraphic levels and apparently involves different stratigraphic units inside the wedges. An upper level is interpreted as wedging of Paleozoic (?) and Mesozoic rocks into the Mesozoic­Cenozoic interface. The lower level of wedging occurs beneath Cretaceous rocks, but there is no evidence regarding the age of the rocks inside the wedge. In the westernmost part of the domain, Paleozoic rocks are thrusted over the Mesozoic and Cenozoic sequence. Existence of a thin-skinned triangle zone as mechanism to explain foreland-dipping rocks uplifted from their regional level in the Putumayo Basin was also proposed by other authors (e.g. Portilla, 1991). In contrast, a previous interpretation of the Orito structure as an anticline in the hanging wall of a "splay of a deeper blind basement-rooted thrust" (Portilla, 1991) suggests that the formation of the Ori to anticline is a result of dip-slip on the Orito fault. A different but equally possible interpretation is proposed in this study. The Ori to anticline is interpreted as occurring in a restraining bend, a feature formed as result of strike-slip displacement inhibition where the right­lateral Orito fault steps to the left into the Orito-north fault. The Orito and Orito­north faults are possibly different shallow level fault segments of a single throughgoing strike-slip fault at basement level. In an attempt to disprove one or both of the structural interpretations depicted above, two balanced cross sections were constructed. A cross section is a vertical section that represents the geometry of structures as they appear today, after deformation (Marshak and Mitra, 1988). Balanced cross sections are structural profiles drawn to conserve the original cross-sectional area of the rocks that were subject to deformation. The sections must observe some structural rules (Dahlstrom, 1969) that provide a means to test and improve the interpretation. A balanced cross section must depict structural geometries that are consistent with the family of structural types (i.e. the structural style) that are known to exist in the area of study. In addition, it must be possible to restore a balanced cross section to a geologically plausible pre-deformation state. This means that the section has been reassembled such that fault displacements have been removed and folds have been straightened out (Marshak and Mitra, 1988). Previous work involving balanced cross sections is unknown in this part of the Andean Cordillera and Putumayo basin. Two vertical cross-sections were constructed and restored to test two interpretations: (1) the Orito fold as result of dip-slip on the Orito fault and (2) the Orito fold as result of right-lateral motion on the Orito fault. Cross-section orientations were chosen based on analyses of earthquake fault-plane solutions (Figs. 8 and 9 and Table 2) and surface geology (Fig. 2). Dip-Slip Faulting Model Cross-section #1 (Fig. 24) was constructed in two segments. The first segment runs across the western domain and is oriented in a northwest-southeast direction, perpendicular to the structural fabric mapped at the surface (Fig. 24). A second segment runs across the eastern domain of the fold and thrust belt. It is oriented in a west-east direction, roughly perpendicular to the strike of the Orito fault and approximately parallel to the movement plane calculated from dip-slip earthquake fault-plane solutions (Fig. 8). This cross-section assumes the folds LITHOLOGY Quaternary Terraces and Alluvium Tertiary sedimentary undif. Volcanic rocks (Tertiary) Orito Gp (Mioc.-Plio) Pepino Fm (Eocene) Rumiyaco Fm (Cret-Paleoc) Villela Fm .(Cretaceous) 111111 Caballos Fm.(Cretaceous) GEOLOGICAL SYMBOLS '--"' DEPOSITIONAL CONTACT ....._,, THRUST FAULT _.._ NORMAL FAULT /iv ANTICLINAL AXIAL TRACE --FAULT INFERRED [ijffi] Motema Fm.(Jurassic) -CROSS SECTION -Sedimentary, metamorphic and igneous rocks -• Area of the structural contour (Paleozoic) map of figure 15 [fi1U Metamorphic Precambrian rocks Figure 24. Geologic map showing locations of cross sections 1 and 2. 54 formed as a result of dip-slip on the different faults. The cross-section was constructed along the seismic profiles L-1460 and L-1820 (Fig. 24 ). The eastern segment of the cross section (Fig. 25) corresponds to the foreland and the footwall of the Orito fault is considered as the regional level of strata constituted by relatively underfomed stratigraphic units that are imaged as sub-horizontal and parallel reflectors on the seismic profiles. The dip of the Orito fault within the Mesozoic and Cenozoic rocks was determined based on the fault­plane structural contour map (Fig. 17). The fault does not become listric at depth because the hanging-wall units toward the west do not return to their regional levels and therefore the fault must involve basement. The Orito fault is the eastern limit of the Orito fold shown in the cross section. The Orito fold is a monocline limited to the west by the San Antonio-Caribe fault. In this area shortening within the Mesozoic section and deeper (?) is accommodated by the thin-skinned San Antonio-Caribe fault, whereas shortening within the Cenozoic section is accommodated by folding above the tip of the San Antonio-Caribe fault. The geometry of the Orito fold and San Antonio-Caribe fault were well constrained by seismic profiles, and show the San Antonio-Caribe fault as a backthrust formed in the tip of the passive roof thrust of the wedge uplifting the western domain. The segment of cross section #1 within the western domain (Fig. 25) was constructed based on the seismic interpretation and correlation with wells drilled through Mesozoic and Cenozoic rocks. In order to explain uplift of these foreland-dipping rocks a wedge interpretation was assumed, as justified above. Inadequate data from wells and seismic profiles prevented unique identification of depths for the wedge floor and roof thrusts. The passive roof thrust arbitrarily interpreted to lie at the shallowest depth possible, at the base of the package of Mesozoic and Cenozoic reflectors. The depth of the floor thrust was constrained by the depth of the roof thrust where the overlaying reflectors reach the level they maintain elsewhere in the Orito fault hanging wall. The floor thrust was drawn to be horizontal for simplicity. At the westernmost part of the western domain, the Mesozoic section is truncated by a major reverse fault carrying Paleozoic rocks in the hanging wall. This zone is only diffusely imaged at the end of the seismic profiles, so fragmentary information constraining this fault block was taken from the surf ace geologic map (Fig. 2). Cross section #1 meets the criteria for a balanced cross section. Consequently, the dip-slip faulting model must be considered as a viable interpretation of structure in the Putumayo basin. Strike-Slip Faulting Model The maximum displacement of most faults typically occurs near the middle of their traces and not near of their tips. Ifthe Orito fault is a dip-slip fault, then it would be an exception to this rule of thumb. Where the Orito fault loses its throw northward, the Orito fold has the highest structural relief (Fig. 15). The structural relief of the fold consequently does not correspond well with the fault throw along the fault. The highest structural relief of the Orito fold does correspond closely with the stepover from the Orito and Orito north faults, suggesting the possibility of strike-slip movement. Cross-section #2 was constructed in a vertical plane approximately parallel to the slip direction determined from strike-slip earthquake fault-plane solutions (Fig. 24). Ifthe Orito fault moves with a slip vector that trends parallel to the slip direction of the strike­slip earthquakes, then the movement plane of deformation in a restraining bend would be approximated by the plane of cross-section #2. This cross section was constructed based on the structural contour map (Fig. 15) and seismic profile intersections, and was located where strike-slip displacement along the presumed right-lateral Orito fault is inhibited in the proposed restraining bend (Fig. 26). The strike-slip model in cross-section #2 (Fig. 26) shows an anticline over 25 km wide with a backlimb dip of 5° and a forelimb dip of 45°. The Orito fold forelimb represents shortening above the tip of a deep-rooted fault that connects the shallow level Orito and Orito north faults. This deep-rooted fault is not imaged in any seismic profile and is interpreted in the section based only on the Orito fold geometry (Fig. 26). The Orito fold backlimb is proposed to be a result of isostatically driven flexural bending, as proposed in the model for the flexural cantilever (Kusznir et al., 1991). The flexural-cantilever model was proposed to explain farfield footwall uplift and hanging wall subsidence in response to isostatic forces in extensional environments. Although this model was proposed for normal faulting and no literature applying this model to contractional deformation was found, I believe the basic concepts of this model explain the Orito fold geometry as result of slip on a deep rooted fault. The distance from the fault stepover at which the backlimb strata return to their regional levels agrees approximately with the distance SOUTHWEST 1-1820 Lf2200 \2210 L-1 889 L-2220 L-P22 IO L-rs 1 I I Dep