Thesis Reference

Home , Cauca River, Palestina Fault, Toro Fault

Thesis

Lithostratigraphical and tectono-sedimentary study of the Plio-Pleistocene infill of the Interandean North Cauca Valley Basin (Colombia)

NEUWERTH, Ralph

Abstract

This investigation focus on the study of the Plio-Plesitocene deposits in a zone covering parts of the Quindío, Risaralda and Valle del Cauca departments in Central Colombia. The results can be summarized as follows : In the initial phase, a vast field campaign, a detailed sedimentological study and facies analyses have led to the differentiation of various lithological units within the Zarzal Formation, allowing the stratigraphical redefinition of the Plio-Pleistocene sediments deposited in the northern part of the Cauca Basin, on both sides of the Serranía de Santa Barbara. This sedimentological study has demonstrated the existence of soft-sediment deformations interpreted as seismites. They demonstrate an intense synsedimentary seismic activity during the Plio-Pleistocene. The last part of this research presents a tectonic study at a larger scale than the studied North Cauca Basin. A dextral strike-slip zone has been evidenced. This deformation zone seems to be responsible for the closing of the North Cauca Basin studied here.

Reference

NEUWERTH, Ralph. Lithostratigraphical and tectono-sedimentary study of the Plio-Pleistocene infill of the Interandean North Cauca Valley Basin (Colombia). Thèse de doctorat : Univ. Genève, 2009, no. Sc. 4141

URN : urn:nbn:ch:unige-189486 DOI : 10.13097/archive-ouverte/unige:18948

Available at: http://archive-ouverte.unige.ch/unige:18948

Disclaimer: layout of this document may differ from the published version.

1 / 1 UNIVERSITE DE GENEVE FACULTE DES SCIENCES Département de Géologie et Paléontologie Professeur G. E. Gorin

UNIVERSIDAD DEL QUINDÍO FACULTAD DE INGENERÍA CEIFI Profesor A. Espinosa

LITHOSTRATIGRAPHICAL AND TECTONO-SEDIMENTARY STUDY

OF THE PLIO-PLEISTOCENE INFILL OF THE INTERANDEAN

NORTH CAUCA VALLEY BASIN (COLOMBIA) ______

THESE

Présentée à la Faculté des Sciences de l’Université de Genève pour obtenir le grade de

Docteur ès sciences, mention Sciences de la Terre

par

Ralph Neuwerth

Monthey (VS)

Thèse No 4141

GENEVE Atelier de reprographie ReproMail 2012 Neuwerth, R.: Lithostratigraphical and tectono-sedimentary study of the Plio-Pleistocene infill of the interandean North Cauca valley basin (Colombia). Terre & Environnement, vol. 106, viii + 157 pp. (2012)

ISBN 978-2-940472-06-2 Section des sciences de la Terre et de l'environnement, Université de Genève, 13 rue des Maraîchers, CH-1205 Genève, Suisse Téléphone ++41-22-379.66.28 - Fax ++41-22-379.32.11 http://www.unige.ch/sciences/terre/ i

Abstract

This investigation is the result of a collaboration agreement between the Universities of Geneva (Switzerland) and Quindío (Colombia), initiated by Profs. Georges Gorin (Geneva) and Armando Espinosa (Quindío). It is part of a larger research project funded by the Swiss National Science Foundation and studying the Plio-Plesitocene deposits in a zone covering parts of the Quindío, Risaralda and Valle del Cauca departments in Central Colombia. Within the framework of the latter project, four Ph.D. and six M.Sc. theses have been carried out and six papers have been so far published in international journals. The thesis presented here attempts to integrate in an optimal way the results already obtained by other contributors to the project.

Because the North Cauca Basin studied here was significantly lacking reliable data, a multidisciplinary approach has been applied throughout the investigation. The results have improved the geological knowledge of the area and can be summarized as follows :

In the initial phase, a vast field campaign, a detailed sedimentological study and facies analyses have led to the differentiation of various lithological units within the Zarzal Formation. The latter are interstratified with each other, as well as with the volcaniclastic deposits of the Quindío-Risaralda Fan. These new data, as well as radiometric datings, have permitted the stratigraphical redefinition of the Plio-Pleistocene sediments deposited in the northern part of the Cauca Basin, on both sides of the Serranía de Santa Barbara (paper submitted to Geologica Acta).

This sedimentological study of various outcrops has demonstrated the existence of spectacular soft- sediment deformations interpreted as seismites. They demonstrate an intense synsedimentary seismic activity during the Plio-Pleistocene. These results have been published in 2006 in Sedimentary Geology.

The last part of this research presents a tectonic study at a larger scale than the studied North Cauca Basin. Because a sedimentary depositional model for the lithostratigraphical units was established, it has been necessary to take a more regional view at tectonic lineaments in order to understand the geometry of the basin in a wider context. A dextral strike-slip zone has been evidenced, comprised between the EENE-WWSW trending, dextral, transtensional Cucuana and Istmina faults. This deformation zone seems to have led to the clockwise rotation of the Romeral Fault System and the formation of NW-SE trending normal faults. The latter are probably responsible for the closing of the North Cauca Basin studied here, but also for that of the Amagá and Upper Magdalena basins. ii

Resumen

La presente investigación es el resultado del convenio existente entre las Universidades de Ginebra (Suiza) y del Quindío (Colombia), iniciado por los profesores Georges Gorin (Suiza) y Armando Espinosa (Colombia). Dicho proyecto a sido sostenido por el Fondo Nacional Suizo de la Investigación, el cual fue consagrado al estudio sedimentológico y estructural de los depósitos de edad Pliopleistocena que cubren parte de los departamentos del Quindío, Risaralda y la parte norte del departamento del Valle del Cauca. La realización de este proyecto permitió producción de cuatro tesis de doctorado, seis tesis de master y la publicación de seis artículos en revistas internacionales. El trabajo aquí presentado intenta integrar de la mejor forma los estudios realizados previamente en el marco de dicho proyecto.

La cuenca septentrional del Valle del rio Cauca carecía de estudios detallados hasta la elaboración del presente estudio. Esta problemática fue abordada de una manera multidisciplinaria y los resultados obtenidos contribuyen en el mejoramiento de los conocimientos geológicos del área de la siguiente manera.

Gracias al trabajo de campo; al estudio sedimentológico detallado y al análisis de facies las diferentes unidades litoestratigráficas existentes al interior de la Formación Zarzal pudieron ser diferenciadas. De la misma manera que las relaciones existentes con el Abanico Volcaniclástico del Quindío- Risaralda. Estos resultados junto con las dataciones radiométricas efectuadas nos permiten proponer una redefinición estratigráfica de los sedimentos Plio-Pleistocenos depositados en la parte norte de la cuenca del valle del rio Cauca; de lado y lado de la Serranía de Santa Bárbara (articulo remitido a Geológica Acta).

Esta descripción sedimentológica a la escala del afloramiento permitió igualmente de identificar impresionantes estructuras de deformación de sedimentos blandos; que fueron interpretadas como sismitas. Estas últimas, son una prueba de la actividad sísmica de la zona durante el Plio-Pleistoceno. Estos resultados fueron publicados en el año 2006 en la revista “Sedimentary Geology”.

La parte final de esta tesis propone un estudio tectónico a gran escala. Dado que elmodelode depósito de las diferentes unidades litoestratigráficas había sido previamente elaborado (en el seno de este proyecto); un análisis de las principales estructuras existentes en el área era necesario con el fin de precisar la geometría de dicha zona. La existencia de una zona de cizallamiento dextral fue evidenciada entre las fallas de componente dextral y normal de Cucuana e Itsmina orientadas EENE-WWSW. Esta zona de deformación parece ser la responsable de la rotación en sentido horario del sistema de fallas de Romeral y de la creación de las fallas normales de orientación NW – SE. Estas últimas no solo permitieron el cierre de la cuenca del Valle del río Cauca en la zona de estudio, sino que probablemente influyeron en el cierre de las cuencas de Amagá y del Valle superior del Magdalena. iii

Ré s u m é

Cette recherche est le fruit d’un accord de collaboration entre les Universités de Genève (Suisse) et du Quindío (Colombie), initié par les Professeurs Georges Gorin (Genève) et Armando Espinosa (Quindío). Elle s’insert dans un vaste projet de recherche, soutenu par le Fond National Suisse de Recherche Scientifique, portant sur l’étude des dépôts plio-pléistocènes dans une zone couvrant une partie des départements du Quindío, Risaralda et Valle del Cauca en Colombie centrale. Ce projet a permis la réalisation de quatre thèses de doctorat, six thèses de master et six articles dans des revues scientifiques internationales. Le présent travail a essayé d’intégrer de manière optimale les études préalablement réalisées dans le cadre de ce grand projet.

Cette étude a pour but de comprendre la formation ainsi que l’évolution du bassin du Cauca Nord, comprenant deux bassins sédimentaires intra-montagneux (North Cauca Valley and Quindío- Risaralda Fan) durant le dernier million d’années. Ces derniers, situés entre les cordillères centrale et occidentale de la Colombie centrale, sont séparés par une chaine montagneuse d’âge Oligo-Miocène, nommée la Serranía de Santa Barbara. Cette zone est géologiquement intéressante car tectoniquement très active. En effet, ce projet a vu le jour suite au séisme d’Armenia survenu le 20 janvier 1999 d’une magnitude de 6.2, faisant plus d’un millier de victimes. Tectoniquement, cette région est sous l’influence de trois plaques convergentes, à savoir Nazca, Caraïbes et Amérique du Sud. Les dépôts étudiés dans le cadre de ce projet représentent la plus ancienne unité stratigraphique n’ayant pas subi de déformation majeure. Leur étude détaillée permet ainsi d’améliorer la connaissance de l’évolution des Andes colombiennes entre le Néogène tardif et le Pléistocène.

Le bassin du Cauca Nord étudié ici manquant fondamentalement de données pertinentes, une approche multidisciplinaire a été nécessaire et constitue la trame de ce manuscrit. Les résultats obtenus semblent objectivement améliorer les connaissances géologiques et peuvent être résumés de la manière suivante :

Dans un premier temps, une large campagne de terrain et une étude sédimentologique détaillée de la Formation Zarzal ont permis de décrire : (1) de nombreux faciès qui n’apparaissent pas dans la littérature. Cette formation a donc été redéfinie sur la base des environnements de dépôts rencontrés : (a) fluvio-lacustre, (b) mass-flows, (c) Gilbert-type delta et (d) fan alluvial ; (2) une relation stratigraphique avec les dépôts volcaniclastiques du fan du Quindío-Risaralda, qui ont été étudiés et décrits notamment par Fernando Guarin dans ses travaux de diplôme et de doctorat liés à ce projet.

En conséquence, un nouveau groupe lithostratigraphique, nommé le Santa Barbara Group a été proposé. Celui-ci inclus la Formation Zarzal et la nouvelle Formation Quindío-Risaralda. La Formation Zarzal a été redéfinie en trois membres (Membres Obando, Ansermanuevo and Holguín) selon leurs environnements de dépôt et l’origine de leurs éléments lithologiques.

Les dépôts affleurant sur la zone d’étude ont été datés palynologiquement et géochronologiquement (méthode 40Ar/39Ar). Des pollens d’Alnus, apparus en Colombie il y a 0.8 Ma, ont été identifiés dans iv plusieurs échantillons d’argile sur l’ensemble de la zone d’étude. Les datations absolues réalisées sur des biotites prélevées dans dépôts de cendre révèlent des âges compris entre 1,06 ± 0,17 and 2,79 ± 0,5 Ma. Les sédiments du Groupe de Santa Barbara affleurant dans la zone d’étude se sont donc déposés durant le Plio-Pléistocène.

Cette description sédimentologique à l’échelle de l’affleurement a permis également de rencontrer des structures de déformation de sédiments meubles (sables de faible à moyenne granulométrie, limons et argiles) très spectaculaires qui ont fait l’objet d’une publication dans la revue Sedimentary Geology (2006).

Les différentes structures de déformation rencontrées ont été décrites morphologiquement puis classifiées dans 4 groupes comprenant 14 catégories, à savoir (1) structures de charge (simple load cast, pendulous load cast, flame structure, attached pseudonodules, detached pseudonodules) ; (2) structures d’échappement d’eau (water escape cusp, dish and pillar, pocket and pillar) ; (3) intrusions de sédiment meuble (clastic dykes and sills) ; (4) autres structures (disturbed laminites, convolute laminations, slumping, synsedimentary faults). Les mécanismes et forces de déformation sont essentiellement liés aux instabilités gravitationnelles, déshydratation, liquéfaction et déformations cassantes.

Plusieurs mécanismes sont susceptibles de déclencher de telles déformations. Les plus connus sont la surcharge sédimentaire, les courants de tempêtes et la sismicité. Aucune évidence à l’affleurement ne permettant de lier ces structures à une surcharge sédimentaire ni à des tempêtes (swalely et hummocky cross-stratification), celles sont ont donc été interprétées comme sismites. Cette interprétation est supportée par les arguments suivants : (1) l’activité sismique géologique et récente est clairement démontrée dans la zone d’étude qui est sous l’influence du système de faille Cauca-Romeral ; (2) l’intercalation d’intervalles déformés dans des couches non-déformées reflète des événements catastrophiques suivis de période de relative stabilité ; (3) les sédiments meubles de faible à moyenne granulométrie de la Formation Zarzal sont particulièrement sujets à la liquéfaction durant des épisodes sismiques ; (4) les structures rencontrées sont similaires, tant en en taille que dans leur forme, à celles décrites comme sismites dans la littérature ; (5) la large répartition géographique de déformations sont compatibles avec des sismites.

Par conséquence, l’existence de ces sismites dans les sédiments de la Formation Zarzal confirme une activité tectonique dans la zone d’étude durant le Plio-Pléistocène. Des tremblements de terre d’une magnitude supérieure à 5 sur l’échelle ouverte de Richter peuvent être admis, se basant sur la proximité de failles actives ainsi que sur le type de structures de déformation rencontrées.

La stratigraphie et l’âge des dépôts remplissant la dépression du Cauca Nord étant définis, l’activité sismique durant leur dépôt étant démontrée, l’ultime étape de cette étude a été de comprendre la formation d’un tel bassin intra-montagneux. Plusieurs modèle ont été présentés dans la littérature : graben ; bassin de pull-apart ; soulèvement de la Serranía de Santa Barbara,… v

La dernière partie de cette thèse propose ainsi une étude tectonique à plus grande échelle, une vision des grandes structures étant nécessaire pour comprendre la géométrie d’une manière globale. Pour ce faire, une large campagne de mesure de miroirs de faille sur le terrain, une interprétation minutieuse de modèles numériques et une analyse morphométrique ont été réalisées.

La cartographie des dépôts holocène et des failles principales a permis de mettre en évidence la géométrie à large échelle autour de la zone étudiée. Celle-ci montre que les bassins du Cauca Nord, du Quindío-Risaralda, d’Amagá et du Magdalena sont délimités par des failles et donc d’origine tectonique. Ce résultat ainsi que l’analyse des linéaments (1’300) ont permis de mettre en évidence une zone de cisaillement située entre les failles de Cucuana et Istmina, ou le système de failles de Cauca-Romeral a subi une rotation horaire comprise entre 9° et 16°. Cette déflexion crée des failles transtensionnelles orientée NW-SE, telle que le système de faille Otún. Le mouvement normal de ces dernières génère ainsi une surrection de la partie nord de la zone d’étude, qui produit un effet de barrage, créant en conséquence un espace d’accommodation pour le dépôt des sédiments de la Formation Zarzal. Cette interprétation semble confirmée par les résultats d’étude morphométrique sur un MNA de 30m qui démontrent une importante activité tectonique dans la zone d’étude, spécialement dans la région de La Virginia.

De plus, l’analyse structurale des dépôts de la Formation Zarzal démontre l’activité simultanée des failles d’Otún et Ibagué et du système de faille Cauca-Romeral durant le Plio-Pléistocène. Ceci est confirmé par les structures de déformation rencontrée dans ces dépôts, attestant ainsi de l’importante activité sismique durant cette période.

L’ensemble des résultats de notre équipe de recherche, associés aux données décrites dans la littérature, ont permis de proposer un modèle d’évolution cinématique simplifié du Cauca Nord, illustré dans le chapitre 7.

La période allant de l’Oligocène Tardif au Miocène précoce voit la plaque Farallon se séparer en deux nouvelles plaques : Nazca et Cocos. Cet événement coïncide avec l’initiation du système de faille Ibagué et des failles orientées ENE. La conséquence tectonique est l’augmentation du taux de surrection de la Cordillère Centrale, de développement de la vallée du Cauca et le dépôt de la Formation Cartago.

Le plissement de cette dernière commença entre l’Oligocène tardif et le Miocène précoce et la Formation La Paila s’y déposa de manière discordante. Cette période correspond à la collision du bloc Chocó-Panamá contre la plaque sud-américaine.

La Formation Zarzal s’est ensuite déposé en discordance sur la Formation La Paila durant le Plio- Pléistocène, suite à la rotation du système de faille de Cauca-Romeral. vi

De nos jours, la partie nord de la zone d’étude, ainsi que la Serranía de Santa Barbara, sont toujours en surrection, entrainant la phase actuelle de remplissage sédimentaire du bassin du Cauca Nord.

Cette thèse de doctorat ne prétend pas donner toutes les réponses aux questions posées sur la géologie de la zone d’étude. En effet, l’étude sédimentologique des dépôts plio-pléistocènes n’a été réalisée que sur les affleurements en surface et ne permet donc pas de déterminer avec exactitude la géométrie et les épaisseurs des unités lithostratigraphiques. Des données sismiques sont nécessaires pour affiner cette stratigraphie et pour proposer un modèle de facies.

De plus, le modèle de déformation proposé est dérivé d’une analyse d’un modèle numérique d’altitude. Il serait ainsi indiqué d’acquérir des données structurales sur le terrain, spécialement concernant les failles Otún, Cucuana et Istmina, ainsi qu’au pied de la Cordillère Occidentale, où les problèmes d’insécurité nous ont empêchés de nous y rendre.

Finalement, de meilleures datations des Formations Cartago et La Paila contribueraient à l’amélioration du modèle d’évolution cinématique proposé. vii

Acknowledgments

La présentation des résultats de ce travail de recherche me donne l’occasion d’exprimer ma profonde gratitude à l’égard de tous ceux qui ont, de près ou de loin, contribué à son élaboration.

Je tiens tout d’abord à exprimer mes plus vifs remerciements au Pr. Georges Gorin, mon directeur de thèse, sans qui ce projet n’aurait jamais vu le jour et que je considère non seulement comme un collègue, mais comme un ami Je tiens à relever son talent dans son domaine et à le remercier pour ses précieux conseils prodigués tout au long de cette étude. Ses fantastiques qualités humaines et sa grande disponibilité m’ont fourni une motivation supplémentaire à la réalisation de mon diplôme.

Mes remerciements s’adressent également au Pr Armando Espinosa de l’Université du Quindío (Colombie), co-directeur de thèse, et également instigateur du projet. Il m’a été d’un grand secours lors de mes nombreux déplacements en Colombie, tant par ses connaissances de la géologie locale que sur son aide logistique indispensable.

Je remercie également le Dr Mario Sartori, notre maître structuraliste à tous, pour tous ses conseils avisés, son œil sur le terrain, mais également pour sa motivation, ses anecdotes, son humour et sa constante bonne humeur.

Je remercie par anticipation le Pr. Olivier Parize pour sa participation en qualité de jury de ce travail, tout en me réjouissant de ses commentaires, sûrement pertinents, qui contribueront à l’amélioration de ce manuscrit.

Un grand merci s’adresse tout particulièrement à Fiore Suter, avec qui j’ai non seulement pu partager des mois de terrain, mon bureau dans lequel de grandes discussions ont largement contribué au présent travail, mais également un appartement et parmi les plus beaux moments de ma vie.

Un pensée chaleureuse se tourne également vers Fernando Guarin, Lina Ospina et Olivier Pahud, avec qui une réelle amitié c’est nouée, et avec qui j’ai réalisé une partie de mes missions de terrain.

Je tiens à exprimer ma plus grande reconnaissance à Rayo et Hubert, Lalo et Don Daniel pour leur aide précieuse, voire indispensable.

Plus que mes remerciement mais mon amour s’adressent à Katia, qui a été et reste mon soutien le plus précieux au monde, bacissimi !

Mille mercis également à tous mes amis, mis hermanos colombianos, particulièrement à Solecita, JuanK, LaCif, Alejo y Aleja, Lucho, Desisy y Mario, las Hermanas Negritas, Ingé Hugo.

Tous mes remerciements aussi à mes compagnons de volée, particulièrement Anouk, Jo, Julien et Fiore, avec qui le contact est toujours resté et le restera pour de nombreuse années encore je l’espère. viii

Un grand merci également non seulement au bureau 308 qui a vu se succéder, entre autre, David, Fiore, Chadia et Lina, mais également tous mes collègues du 3ème !

Je remercie aussi tous les personnes qui ont collaboré scientifiquement et techniquement : Carlos Guzmán, Richard Spikings, Rossana Martini, François, Pierrot et Peter, Olivier, Luc et Roelant, etc….

Pour finir, je tiens à remercier mes parents et mon frère Frank pour leurs encouragements et leur amour, tous mes amis, et tout spécialement Fiore Suter, mi hermanito et colocataire. Contents

Abstract i

Resumen ii

Résumé iii

Acknowledgments vii

CHAPTER I - In t r o d u c t io n 1 1.1. Description of project 2

1.2. Aims of study 2

1.3. Organisation of manuscript 4

CHAPTER II - Me t h o d s 7 2.1. Field study 8

2.2. Fault inversion method 8

2.3. Morphometry 8

References 10

CHAPTER III - Ge o l o gi c a l b a c k gr o u n d 11 3.1. Location of study area 12

3.2. Geodynamic evolution of Colombia from Jurassic to recent times 13

3.3. Stratigraphy of the studied area 17 3.4. Structural geology of the studied area 23

3.5. The Cauca Depression 28

References 35

CHAPTER IV - Se d i m e n t a r y in f il l o f An d e a n in t r a mo u n t a n e b a s in s : c a s e his t o r y o f Plio-Pl e is t o c e n e d e po s it s in t h e No r t h Ca u c a Va l l e y (Co l o mbia) 51 4.1. Introduction 52

4.2. Geological setting 54 4.3. Tectonics 56

4.4. Methods 58

4.5. Results and interpretation: New lithostratigraphical subdivision of Late Tertiary-Recent sediments 58

4.6. Discussion 74

4.7. Conclusions 76

References 77

CHAPTER V - So f t -s e d i m e n t d e f o r ma t io n in a t e c t o n ic a l l y a c t iv e a r e a : Th e Plio- Pl e is t o c e n e Za r za l Fo r ma t io n in t h e Ca u c a Va l l e y (We s t e r n Co l o mbia) 85 5.1. Introduction 86

5.2. Geology of Zarzal Formation 87

5.3. Overview of soft-sediment deformations and classifications 91

5.4. Soft-sediment deformations in the Zarzal Formation 93

5.5. Discussion 101

5.6. Conclusions 108 References 110

CHAPTER VI - Te c t o n ic s 117 6.1. Introduction 118

6.2. Large scale geometry 119

6.3. Morphometry 126

6.4. Structural analysis 130

6.5. Discussion and conclusions 145

References 147

CHAPTER VII - Co n c l u s io n s 151 7.1. Sedimentary infill of Andean intramountane basins: case history of Plio-Pleistocene deposits in the North Cauca Valley (Colombia) (chapter 4) 152

7.2. Soft-sediment deformations in a tectonically active area: The Plio-Pleistocene Zarzal Formation in the Cauca Valley (Western Colombia) (chapter 5) 152

7.3. Tectonics (chapter 6) 153

7.4. Cinematic reconstruction 153 7.5. Perspectives 155

References 156

Liste of figures and tables

CHAPTER I Fig. 1.1: Geodynamics of NW South America: velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008). Location of figure 1.2. 3 Fig. 1.2: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing departmental boundaries and location of the areas studied in the general project(in red: zone one studied by Guarin, 2008; zone II studied by Suter, 2008; zone III studied in this work). A: Ansermanuevo; Ar: Armenia; C: Cartago, V: La Virginia; P: Pereira; Z: Zarzal. CC: Central Cordillera; SSB: Serranía of Santa Barbara; WC: Western Cordillera. 4

CHAPTER III Fig. 3.1: (A) Geodynamics of NW South America: velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008). Location of study area. Abbreviations: B: Bogotá; C: Cali; CC: Central Cordillera; EC: Eastern Cordillera; WC: Western Cordillera; EFFS: Eastern Frontal Fault System; IBF: Ibagué Fault; GF: Garrapatas Fault; RFS: Romeral Fault System. (B) Digital elevation model (DEM, (USGS, 2005)) of western central Colombia showing the course of the Cauca River. The study area is located upstream of La Virginia town, at the northern termination of the Cauca Valley Basin (after (Suter et al., 2008). 12

Fig. 3.2: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing Plio- Pleistocene deposits and faults in the studied area. Abbreviations: A: Ansermanuevo; Ar: Armenia; C: Cartago, V: La Virginia; P: Pereira; Z: Zarzal. CC: Central Cordillera; SSB: Serranía of Santa Barbara; WC: Western Cordillera. AB: Aguas Bonitas Fault; Asmn: Ansermanuevo Fault; C-A: Cauca Almaguer Fault, C-P: Cauca Patía Fault; Csta: Consota Fault; Mont: Montenegro Fault; Nav: Navarco Fault; Otún: Otún Fault; Plst: Palestina Fault; Qnueva: Quebradanueva Fault; Rob: El Roble Fault; RV: Río Verde Fault; Sal: Salento Fault; San J: San Jeronimo Fault; Sev: Sevilla Fault; SR: Santa Rosa Fault; Tor: Toro Fault. 13

Fig. 3.3: Simplified cross-section across the Cauca depression and the Quindío-Risaralda Basin, see figure 3.2 for location. After (Suter et al., in review). 14 Fig. 3.4: Middle Cretaceous to Miocene plate reconstruction of northwestern South America showing two models: (A) arrival and oblique docking of the Pacific Terranes, followed by accretion of San Jacinto, Sinú , Guajira-Falcon, and Caribbean Mountain terranes and finally collision of the Chocó Arc. Grey shaded areas in all time slices represent paleotopographic swells, elevated and/or emergent areas. Red crosses represent magmatism (after (Cediel et al., 2003)). Abbreviations: BAU: Baudó terrane; CAM: Caribbean Montaine terrane; CG: Cañas-Gordas terrane; DA: Daguá-Piñon terrane; EC: Eastern Cordillera; GA: Garzón massif; GOR: Gorgona terrane; GU- FA: Guarija-Falcon terrane; MSP: Macacaibo sub-plate; PA: Panamá terrane: RO: Romeral terrane; SJ: San Jacinto terrane; SM: Sinú terrane. (B) backarc extension followed by accretion (after (Moreno and Pardo, 2002)). 15

Fig. 3.6: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing the 45 mass flow units of the Quindío-Risaralda Fan (modified after (Guarin, 2009) and the three Cartago fans (A, B and C). Abbreviations: A: Armania; C: Cartago; P: Pereira; V: La Virginia; CC: Central Cordillera; SSB: Serranía de Santa Barbara. 22

Fig. 3.7: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing faults in the studied area (modified after (Ingeominas, 1999; Nivia, 2001; Suter et al., 2008)). 24 Table 3.1: Compilation of published literature about fault characteristics in the study area and its surroundings. These faults are located in Fig. 3.7 (modified after (Suter et al., 2008)). 34

CHAPTER IV Fig. 4.1: (A) Geodynamics of NW South America and location of study area. Velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008b). Abbreviations: B: Bogotá; C: Cali; CC: Central Cordillera; EC: Eastern Cordillera; WC: Western Cordillera; EFFS: Eastern Frontal Fault System; IBF: Ibagué Fault; GF: Garrapatas Fault; RFS: Romeral Fault System. (B) Digital elevation model (DEM, (USGS, 2005)) of western central Colombia showing the course of the Cauca River. The study area is located upstream of La Virginia town, at the northern termination of the Cauca Valley Basin (after (Suter et al., 2008b). 54

Fig. 4.2: Compilation of the five 1:100’000 geological maps of INGEOMINAS (Caballero and Zapata, 1983; Parra, 1983; McCourt et al., 1984b; Nivia et al., 1995; Estrada and Viana, 1998) which cover the study area (after (Suter et al., 2008b)). 55

Fig. 4.5: Obando Member. A: stacked, trough cross-bedded bodies of black sands interpreted as braided-river channel infill; B: planar stratifications in fine silts to clays interpreted as floodplain deposits; C: diatomaceous deposits interbedded with very fine sandy beds. The white arrow indicates ash fall deposits, dated with 40Ar/39Ar method at 1.32 ± 0.07 Ma (see figure 4.7). 60 Fig. 4.6: Simplified tentative cross-sections from the Western Cordillera (WC) to the Quindío- Risaralda basin. This interpretation illustrates the interfingering of the Plio-Pleistocene units defined in this study (see Figure 4.3 for location and legend). Abbreviations: AM: Ansermanuevo Member; HM: Holguín Member; OM: Obando Member; QRFm: Quindío-Risaralda Formation; SSB: Serranía of Santa Barbara. 61

Fig. 4.7: (A) 40Ar/39Ar plateau age from the ash layer of section 18-20 in figure 4.8. 62

Fig. 4.8: Field sections in the western foothills of the Serranía of Santa Barbara (SSB) (see figure 4.4 for location and legend). In each section, the first occurrence of detritic sediments and mass flows sourced from the SSB is correlated with a black line in order to highlight the change in lithological composition. This line does not correspond to a time line. See figure 4.7 for Ar/Ar datings. 63 Fig. 4.9: Palaeocurrent directions measured in trough cross-bedded black sands, sigmoid stratifications (rose diagramme 8) and imbricated pebbles (rose diagramme 4) of Zarzal Formation. Symbols and abbreviations: n = number of palaeocurrent measurements per site; Ø = mean vector; white arrows depict the mean vector on the DEM. A: Ansermanuevo; C: Cartago; CC: Central Cordillera; CF: Cartago Fan; O: Obando; QRF: Quindío-Risaralda Fan; SSB: Serranía of Santa Barbara; Vict: La Victoria; Vir: La Virginia; WC: Western Cordillera; Z: Zarzal. The base map is the same DEM as that of figure 4.4 where sections are located. 64

Fig. 4.10: Field sections in the eastern foothills of the Western Cordillera (WC) (see figure 4.4 for location of sections). The first occurrence of alluvial conglomerates rich in black cherts, white quartz and volcanic clasts at each location is correlated with a black line in order to highlight the change in lithological composition. This line does not correspond to a time line. Legend is the same as that in figure 4.8. 65 Fig. 4.11: Ansermanuevo Member: clast-supported conglomerate rich in black chert, white quartz and volcanic elements interfingering with typical black sand deposits of the Obando Member (section 17, Fig. 4.10). Hammer for scale. 66

Fig. 4.12: Holguín Member: bottomset deposits: (A) pinching out towards the south and showing an erosive contact with the overlying foreset unit; (B) containing plant fragments (section 23, Fig. 4.8) 67

Fig. 4.13: Holguín Member: erosive contact between foreset and topset units (section 23, Fig. 4.8). 68

Fig. 4.14: Holguín Member: foreset facies (section 23, Fig. 4.8): (A) dip section of steeply dipping, sandy foresets separated by siltstone intercalations resulting from deposition of suspended load; (B) sandy foresets containing clayey soft pebbles; (C) strike-parallel section of foresets, which consist of parallel flat-lying sandstone beds. 69

Fig. 4.15: Holguín Member: topset unit consisting of stacked, trough cross-bedded alluvial deposits (well to moderately sorted sands and gravel) and mass-flow deposits. 70

Fig. 4.16: Field sections in the N-Cartago Fan (see figure 4 for location). The first occurrence of volcanic mass flows at each location is correlated with a black line in order to highlight the change in sedimentary regime. This line does not correspond to a time line, because mass flows were first deposited in the eastern part of the Quindío-Risaralda Basin before moving westwards (Espinosa, 2000; Guarin et al., 2006). Numbers associated with palaeocurrents refer to those in figure 4.9. The base map is the same DEM as that of figure 4.4 where sections are located. Legend is the same as that in figure 4.9. 72 Fig. 4.17: Quindío-Risaralda Formation. A: stratigraphical contact between (1) clast-supported conglomerates showing horizontal stratification and (2) coarse-grained sands with inverse grading; B: photograph of (2) interpreted as hyperconcentrated flow deposits; C: matrix-supported, poorly sorted conglomerates (1) interpreted as debris-flow deposits with dm-size, angular, basaltic clasts (section 6, Fig. 4.16). 72

Fig. 4.18: Example of field section in the Quindío-Risaralda Formation (after Guarin (2008)). 73 Table 4.1: Published subdivisions for the Zarzal Formation. 75

Table 4.2: Proposed new lithostratigraphical scheme of the Cauca depression and Quindío-Risaralda Basin. 76 CHAPTER V Fig. 5.1: Megatectonic framework and location of study area. 86

Fig. 5.2: Regional distribution of Plio-Pleistocene sediments in the Valle del Cauca, Risaralda and Quindío Departments. Location of Figs. 5.3 and 5.4. 87

Fig. 5.3: Geological cross-section across the Valle del Cauca and Quindío Departments. See Fig. 5.2 for location. 88

Fig. 5.4: Detailed geological and location map of study area. See Fig. 5.2 for location. 89

Fig. 5.5: E–W and N–S trending correlations of field sections in the Zarzal Formation. See Fig. 5.4 for location of field sections. 90

Table 5.1: Comparison between some classifications of soft-sediment deformation (SSD) structures showing nomenclatures and classification criteria. 91 Fig. 5.6: Types of soft-sediment deformation structures observed in the Zarzal Formation. 92

Fig. 5.7: Load structures. (A) Simple load cast (a) (section 8, see Figs. 5.4 and 5.5 for location) associated with convolute laminations (b) and water escape structure (c). (B) Pendulous load cast (section 8, see Figs. 5.4 and 5.5 for location). This structure is associated with a subvertical synsedimentary fault. Part of the deformed fine sand–clay is liquefied, probably as a result of slumping. (C) Pendulous load cast (section 8, see Figs. 5.4 and 5.5 for location) showing internal deformations (a) associated with gravity loading. 94

Fig. 5.8: Load structures. (A) Pendulous load cast (drop structure, Alfaro et al., 1997), near section 8 (see Figs. 5.4 and 5.5 for location). It forms a pocket of medium-grained sands overlying deformed, finely laminated sands. The upper part of the latter is affected by flame structures (a). (B) Attached pseudonodule made of fine-medium-grained sands, which sank into coarse-grained sands (section 8, see Figs. 5.4 and 5.5 for location). (C) Attached pseudonodule (a), water escape structure (b), simple load cast (c), convolute lamination (d) (section 8, see Figs. 5.4 and 5.5 for location). The attached pseudonodule (a) displays slightly deformed laminations. Note the clay layer within (a) which seems to have been dislocated by the water escape. 95

Fig. 5.9: Load and water-escape structures. (A) Attached (a) and detached (b) pseudonodule (section 8, see Figs. 5.4 and 5.5 for location). Note that the clayey silt has a greater density than the liquefied sand. (B) Water escape cusps (a) (section 8, see Figs. 5.4 and 5.5 for location) formed by mediumgrained sands intruding fine-grained sands. Observe the laccolith shape of the fine-grained- sand intrusion into medium-grained-sands and silts (b). This structure is capped by hardly deformed silts, which have been penetrated by a sandy sill (c). (C) Dish and pillar structure (section 8, see Figs. 5.4 and 5.5 for location). The undisturbed laminations of the overlying medium-coarse-grained sand lense proves that the underlying deformation occurred prior to its deposition and is not related to loading. 96

Fig. 5.11: Soft-sediment intrusion and other deformation structures. (A) Medium-grained sand, subvertical disconnected dyke (near section 8, see Figs. 5.4 and 5.5 for location) showing downward bending of intruded, fine-grained sediments (a) in the lower part. This feature indicates the lateral movement of the injection. (B) Disturbed laminites (near La Victoria, see Fig. 5.4 for location). (C) Convolute laminations (section 6, see Figs. 5.4 and 5.5 for location). 99 Fig. 5.12: Other soft-sediment deformation structures. (A) Slump with subhorizontal axial plane (section 8, see Figs. 5.4 and 5.5 for location). (B) Synsedimentary faulting showing both listric (a) and reverse (b) faults (section 8, see Figs. 5.4 and 5.5 for location). This structure can be assimilated to a flower structure indicative of strike-slip movements. Small-size load cast (c) and flame(d) structures are also observed. 100

Fig. 5.13: Post-depositional extensional tectonics in the Zarzal Formation (section 7, see Figs. 5.4 and 5.5 for location). 105

Fig. 5.14: Post-depositional strike-slip tectonics in the Zarzal Formation (section 8, see Figs. 5.4 and 5.5 for location). 106

Fig. 5.15: Injection dykes (section 8, see Figs. 5.4 and 5.5 for location) are a proof of extension (Rodríguez-Pascua et al., 2000). 107

CHAPTER VI Fig. 6.1: Idealized model of the Quindío-Risaralda Fan and Serranía de Santa Barbara. Black and green arrows indicate respectively horizontal and vertical direction of rotation. (After (Guarin, 2008). Abbreviations: A: Armenia, M: Montenegro, F.MTG: Montenegro fault, F.MTA: Matecaña fault, F.ARM: Armenia fault, F. PTS = Potrerillos fault, C: compresion, E: extension. 118

Fig. 6.2: (A) Summary of the data obtained in (Suter et al., 2008) and (B) comparison with the theoretical fault pattern developed under right-lateral shear system (after (Tchalenko, 1970)). (After (Suter et al., 2008)). 119

Fig. 6.3: (A) Map of relative uplift rates of the SSB fold-and-thrust range and its surrounding, Plio- Pleistocene to Recent deposits; (B) Cross-section of the SSB passing between Obando and La Victoria (see 6.3A for location) (After (Suter et al., in review)). 120

Fig. 6.4: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing Holocene depocenter areas (yellow), bounded by main faults. Abbreviations: BSFS: Bituima-La Salina Fault System; CF: Cucuana Fault; CPF: Cali-Patía Fault; DF: Doima Fault; IBF: Ibagué Fan; IF: Ibagué Fault; IST: Istmina Fault; GF: Garrapatas Fault; LSFS: La Salina Fault System; OFS: Otún Fault System; PAF: Palestina Fault; PF: Potrerillos Fault; PG: Piedras Girardot Basin; QNF: Quebradanueva Fault; QR: Quindío-Risaralda Basin; RFS: Romeral Fault System; SDL: Santo Domingo lineament; SRF: Santa Rosa Fault. Green lines represent the new faults mapped in this study (see figure 6.11 for their inferred kinematics). 121

Fig. 6.5: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing the “en- échelon” Otún Fault mapped in this study. Letters a, b and c refer to fault segments forming the Otún Fault (see figure 6.11 for kinematics). 123

Fig. 6.6: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing lineaments in the Central Cordillera and its surroundings, between 4°N and 7.4°N. 124 Fig. 6.7: Quantity-dependent rose-diagram illustrating the orientation of lineaments interpreted in figure 6.6. 126 Fig. 6.8: Summary of figure 6.6, where the principal lineaments affecting this area are grouped into families according to their strike. Abbreviations: AF: Arma Fault; GF: Garrapatas Fault; IF: Ibagué Fault; OF: Otún Fault; PF: Palestina Fault; SF: Salento Fault; SRF: Santa Rosa Fault. 127

Fig. 6.9: Simplified map drawn from figure 6.4 showing main faults and Holocene depocenters (yellow). The hatched area represents the shear zone. (See figure 6.4 for abbreviations). 128 Fig. 6.10: Quantity-dependent rose-diagram illustrating the orientation of the RFS lineaments of figure 6.8, situated (A) north of, (B) within and (C) south of the shear zone (see figure 6.9). 129 Fig. 6.11: (A) Simple shear associated with strike-slip faulting produces preferred orientation of faults, as well as different fault movements (after (Wilcox et al., 1973; Sylvester and Smith, 1976)). (B) Summary of the data obtained in this study. Grey areas represent extensional zones. Abbreviation: Ψ: shear angle. 129

Table 1: Morphometric data calculated of the studied area. 130

Fig. 6.12: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing the drainage basin asymmetry of the North Cauca Depression (yellow) and the Risaralda Basin (red). Abbreviation: Ψ: shear angle. 131

Fig. 6.13: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing the mountain- front sinuosity (Smf) of the eastern border of the Risaralda Valley (black lines), the western border of the Risaralda Valley (yellow lines), the western border of the Cauca Valley (red lines) and its eastern border formed by the SSB foothills (green lines). 132

Fig. 6.14: 30-meter resolution DEM based on radar photographs (USGS, 2005) covering the study area. White circles indicate the location of faults where striae have been observed, and white squares the location of faults in the Plio-Pleistocene deposits. Abbreviations: A: Ansermanuevo; Bel: Belalcazar; C: Cartago; Ob: Obando; R: Roldanillo; T: Toro; U: La Union; Vict: La Victoria; Vir: La Virginia; Z: Zarzal. 134

Fig. 6.15: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing (A) the distribution of the calculated palaeostress tensors and (B) the distribution of maximum horizontal σ (in red) in area A. 134

Fig. 6.16: Stereoplots representing the dip, dip azimuth, and kinematics of fault planes at each site numbered in figures 6.15 and 6.20, as well as the orientation of their calculated paleostress tensors (Wulff stereograms, lower hemisphere). 135

Fig. 6.17: Fault plane: s: slip fibers; f: crystallization fibers. The white arrow shows the motion of the missing compartment with respect to that in the picture. 136

Fig. 6.18: Histogram representing the number of tensors from area A (see figure 6.15) belonging to stress regimes versus their respective ellipsoid form parameter Ф=(σ2−σ3)/(σ1−σ3). 136 Table 2: Parameters of the 19 calculated stress tensors, with the name of the site, the orientations of σ1, σ2 and σ3, the corresponding ellipsoid form parameter (Ф) and the stress regime the tensor belongs to. The number of faults used for the calculation appears in the “n” column; Var (°) indicates the average misfit angle after the final calculation. A quality criterion between 1 (good) and 3 (bad) is given for the result. 137

Fig. 6.19: Fault plane: s: slip fiber. The white arrow shows the undeterminable motion of the missing compartment with respect to that in the picture. 138

Fig. 6.21: Histogram showing the number of tensors from area B (see figure 6.15) belonging to stress regimes versus their respective ellipsoid form parameter Ф=(σ2−σ3)/(σ1−σ3). 140

Fig. 6.22: Histogram showing the number of tensors from the whole studied area (see figure 6.15) belonging to stress regimes versus their respective ellipsoid form parameter Ф=(σ2−σ3)/(σ1−σ3). 141

Fig. 6.23: Quantity-dependent rose-diagram illustrating the orientation of the 513 faults and joints measured in the Zarzal Formation. 142

Fig. 6.24: 30-meter resolution DEM (USGS, 2005) with location of the sites where conjugate normal faults planes were measured in the Plio-Pleistocene Zarzal Formation and Quindío-Risaralda volcaniclastic Fan. The black dots in the plots (Wulff stereonets, lower hemisphere) are projections of the fault plane poles. The dip of their mean vector indicating the direction of elongation (local σ3) is represented by the black asterisks. The dip and dip azimuth of mean vectors is shown besides each plot. The values given for the sites where only one single fault plane could be measured (numbers 4, 6, 10, 14, 17 and 18; black arrows on map) correspond to the dip and dip azimuth of the fault planes. The directions of local σ3 are symbolized on the DEM by arrows. Abbreviations: A: Ansermanuevo; C: Cartago; CF: Cartago Fan; O: Obando; QRF: Quindío-Risaralda Fan; SSB: Serranía de Santa Barbara; T: Toro; Vict: La Victoria; Vir: La Virginia; WC; Western Cordillera; Z: Zarzal. 143

Fig. 6.25: Examples of synsedimentary extensional features observed in the Zarzal Formation. They are (A) covered by unfaulted beds (a) or (B) characterized by the variable thickness of the faulted blocks (b). 144

Fig. 6.26: Schematic block diagram showing main faults and Holocene depocenters (yellow) of the studied area. Abbreviations: AF: Armenia fault; CAF: Cartago Fan; CF: Cajamarca fault; PF: Potrerillos fault; QNF: Quebradanueva fault; SF: Sevilla fault. 146

CHAPTER VII

CHAPTER I

In t r o d u c t i o n

Ralph Neuwerth 2 Chapter 1

1.1. Description of project

After many years of research and numerous publications on the historical seismicity of Colombia, Prof. Armando Espinosa of the Quindío University published a revision of the seismic catalogues (Espinosa, 2004). Aware of the potential risk of a dramatic earthquake in the Armenia region, he had proposed repeatedly the need for a microzonification of the Quindío Department. Unfortunately, his proposal had not gone very far when the terrible Armenia earthquake struck the coffee region of Colombia on 20 January 1999 with a magnitude of 6.2. It killed 1230 people and destroyed more of 5600 homes. The economic impact of this earthquake led to a direct economic loss of approximately US$ 1.8 billion (Cardona, 1999). Following this tragic event and discussions between Prof. Armando Espinosa and Prof. Georges Gorin of the University of Geneva, a scientific collaboration agreement was established between the Quindío and Geneva universities. Six M. Sc. theses (Guarin, 2002; Suter, 2003; Duque, 2005; Ospina, 2007; García Londoño, 2008; Pahud, 2009) and four Ph.D theses (Guarin, 2008; Suter, 2008; Neuwerth, this work; Duque, ongoing work) have been or are being carried out within the framework of this collaboration.

This research has been supported by the Swiss National Science Foundation (grants nos. 21-67080.01 and 200020-107866) within the framework of a general project entitled: “Tectonics, neotectonics and sedimentation in active fault zone: examples of Plio-Pleistocene deposits in the Northern Andes of Central Colombia”. This investigation aimed at improving the geological knowledge about Plio- Pleistocene sediments in an area covering the Colombian Departments of Quindío, Risaralda and Valle del Cauca (Fig. 1.1), at the northern termination of the interandean Cauca Depression. This zone lies at the collision front of the Chocó-Panamá Block, where the major N-S trending Romeral Fault System (active since Cretaceous times) changes its kinematics from right-lateral in the south to left-lateral in the north (Fig. 1.1).

The studied area is located in Central Western Colombia (Figs. 1.1 and 1.2). It comprises two sedimentary basins separated by a belt of folded, pre-Pliocene continental clastic sediments (the Serranía of Santa Barbara or SSB): the North Cauca Depression to the west and the Quindío-Risaralda Basin to the east (Fig. 1.2). These two basins are filled by subhorizontal Pleistocene sediments. In the North Cauca Basin, this infill consists principally of fluvio-lacustrine deposits (the so-called Zarzal Formation), which interfinger with alluvial fans derived from the Western Cordillera in the west and from the Serranía of Santa Barbara in the east. Moreover, towards the northeast, they interfinger with the volcaniclastic mass-flow deposits which form a vast fan that infills the Quindío-Risaralda Basin and is sourced from the Cerro Bravo-Machin volcanic complex in the Central Cordillera. This area has been studied by the three PhD students involved in the project, i.e., Fernando Guarin (Guarin, 2008; area I in Fig. 1.2), Fiore Suter (Suter, 2008; area II in Fig. 1.2) and Ralph Neuwerth (area III in Fig. 1.2).

1.2. Aims of study

The North Cauca Valley Basin is a 200km long alluvial plain extending from Cali up to La Virginia (Fig. 1.1) at an altitude of 900m. It is subdivided in its northern part in the two sub-basins described Introduction 3

Fig. 1.1: Geodynamics of NW South America: velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008). Location of figure 1.2. above. These sub-basins are separated by the Serranía of Santa Barbara (SSB), an active pop-up structure forming a topographic barrier. The Oligo-Miocene rocks forming the SSB are folded and are unconformably overlain on both sides by Pliocene to Recent rocks which are subhorizontal and show extensional dislocations (Cardona and Ortiz, 1994; Pardo et al., 1994; Suter et al., 2008a; Suter et al., 2008b).

The North Cauca Valley Basin is under the influence of the N to NNE trending Romeral faults, which, in this area, presents a compressive kinematics (Paris et al., 2000; Suter et al., 2008b). 4 Chapter 1

The aims of this study are twofold: 1) to attempt to understand how such an intramountane basin has been formed; 2) to refine the existing poor stratigraphical and sedimentological knowledge of the Pliocene to Recent sediments.

1.3. Organisation of manuscript

Because this thesis deals with a multidisciplinary approach, it has been subdivided into chapters

Fig. 1.2: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing departmental boundaries and location of the areas studied in the general project(in red: zone one studied by Guarin, 2008; zone II studied by Suter, 2008; zone III studied in this work). A: Ansermanuevo; Ar: Armenia; C: Cartago, V: La Virginia; P: Pereira; Z: Zarzal. CC: Central Cordillera; SSB: Serranía of Santa Barbara; WC: Western Cordillera. Introduction 5 corresponding to the different investigated geological fields.

The second chapter presents a short overview of field work carried out to collect sedimentological and structural data, of fault inversion analysis and morphometry.

The third chapter aims at presenting the high tectonic complexity of the studied area, which is situated in an accommodation zone between three tectonic plates and at the front of the Chocó-Panamá Block indentation. The stratigraphical units cropping out in the studied area are described and related with the geodynamical evolution of northwestern South America. Finally, interrogations and hypothesis about the formation of the intramountane basin studied are exposed.

The fourth chapter focuses on sedimentology and lithostratigraphy. Twenty-two field sections have been described. They comprise data on palaeocurrents, sedimentary structures, bedding, palynological and geochronological analysis. They have permitted the identification of numerous gaps existing at present day and the proposition of a revisited lithostratigraphy of the Plio-Quaternary deposits in the North Cauca Valley. This chapter has been submitted as a manuscript to Geologica Acta.

The fifth chapter has been published in 2006 in Sedimentary Geology (Neuwerth et al., 2006). It describes soft-sediment deformations encountered at large scale and demonstrates the high level of paleosismicity that have affected the deposits in the studied area.

The sixth chapter gets into the tectonic, neotectonic and geomorphological aspects and tries to explain the northern closure of the basin. 6 Chapter 1

References

Cardona, J. F. and Ortiz, M., 1994. Aspectos estratigráficos de las unidades del intervalo Plioceno Holoceno entre Pereira y Cartago. Propuesta de definición para la Formación Pereira. Manizales, Colombia, Universidad de Caldas: 155.

Cardona, O. D., 1999. The earthquake of Armenia, Colombia, January 25, 1999, Special Report, Geohazards International, pp. 9.

Duque, A. L., 2005. Geology of the urban zone of Armenia and its application to land management (Colombia). M. Sc. thesis, Université de Genève, Geneva, Switzerland, 153 p.

Espinosa, A., 2004. Historia sísmica de Colombia (Academia Colombiana de Ciencias Exactas, Físicas y Naturales, Universidad del Quindío), CD-ROM.

García Londoño, L. F., 2008. Etude néotectonique de la Faille Armenia entre les villes de Circasia et Filandia (Quindío - Colombie). Travail de diplôme inédit.

Guarin, F., 2002. Etude du fan fluvio-volcanique du Quindío (Colombie). Geneva, Univ. of Geneva, Switzerland: 92.

Guarin, F., 2008. Etude sédimentologique du cône volcanoclastique du Quindío-Risaralda (Colombie Centrale) et sa relation avec la morphotectonique. Terre et Environnement (Université de Genève), 29, 146 p.

Ospina, L. M., 2007. Morphotectonique des dépôts quaternaires dans la région de Calarcá, Quindío (Colombie centrale). M. Sc. thesis, Université de Genève, Geneva, Switzerland, 96p.

Pahud, O., 2009. Etude sédimentologique et morphotectonique dans le Quaternaire du Valle del Cauca (Colombie). M. Sc. thesis, Université de Genève, Geneva, Switzerland.

Pardo, T. A., Moreno, S. M. and De J. Gómez, C., 1994. Evidencias de actividad neotectonica en la carretera Cartago-Ansermanuevo (Valle del Cauca, Colombia). III Conferencia colombiana de geología ambiental. Armenia, Quindío (Colombia): 181-191.

Suter, F., 2003. Géologie de la région de Playa Azul, partie occidentale distale du fan fluviovolcanique du Quindío (Serranía de Santa Barbara, Quindío et Valle del Cauca, (Colombie), Université de Genève, Suisse: 133.

Suter, F., 2008. Tectono-Sedimentary Study of the Interandean North Cauca Valley Basin, Central Western Colombia. Terre et Environnement (Université de Genève), 78, 145 p.

Suter, F., Neuwerth, R., Gorin, G. E. and Guzman, C., 2008a. Depositional model of (Plio-)Pleistocene sediments in a tectonically active zone of Central Colombia. Geologica Acta 6(2), 1-19.

Suter, F., Sartori, M., Neuwerth, R. and Gorin, G. E., 2008b. Structural imprints at the front of the Chocó-Panamá indenter: field data from the North Cauca Valley Basin, Central Colombia. Tectonophysics 460, 134-157. CHAPTER II

Me t h o d s

Ralph Neuwerth 8 Chapter 2

2.1. Field study

Extensive field work has been the most important component of this research. It corresponds to a period of some eight months over four years. It has been principally dedicated to sedimentological and structural data acquisition. The former encompasses the sedimentary logging of twenty-two field sections, the sampling of sands and silts for mineralogical study, of clays for palynological analysis and ash layers for geochronological dating. Because of the dense tropical vegetation cover, the best outcrops have been encountered near roads and urban constructions.

Structural field work has consisted of fault measurements, i.e., strike and dip of fault planes in soft sediments and strike and dip of fault planes and slickenside striae in consolidated rocks. Fracture strikes were also measured. Because of vegetation cover and strong rock weathering, this field work has been essentially carried out in river beds. Furthermore, because of the high level of insecurity in the Western Cordillera, only the foothills of the latter have been investigated.

Field work was preceded by the extensive interpretation of aerial photographs in order to locate outcrops and surface fault traces and to carry out geomorphological analysis.

2.2. Fault inversion method

This section focuses only on data analysis and quality criteria. A detailed outline of the method, its merits and limitations can be found in Angelier and Mechler (1977) and Angelier (1994). Faults data have been collected to determine palaeostress axis directions using the direct inversion method of Angelier (1990) implemented in the TectonicsFP software (Sperner et al., 1993; Ortner et al., 2002).

The stability and quality of each tensor has been estimated from the following criteria:

(1) coherency test by comparison with the right-dihedra method of Angelier and Mechler (1977);

(2) number of faults used in the inversion: a dataset consisting of less than eight faults would give a low quality result;

(3) average misfit angle: if too many faults have to be removed, the result is considered as bad,

(4) the stability of the result with respect to particular faults: when some faults have a strong influence on the result when added or removed from the dataset, the quality criterion would be bad.

Tensors were classified from 1 (excellent) to 3 (low quality).

2.3. Morphometry

Two geomorphic parameters have been used in order to estimate or quantify the neotectonic activity: the Drainage Basin Analysis and the Mountain-Front Sinuosity. Theoritical considerations can be found in Keller and Pinter (2002). Only equations are presented here in order to understand the Methods 9 values presented in chapter 5:

The Asymmetry Factor (AF) is defined as

AF = 100(Ar=At)

where Ar is the area of the basin to the right (facing downstream) of the trunk stream, and At is the total area of the drainage basin.

The mountain-front sinuosity (Smf) is defined as

Smf = Lmf/Ls

where Lmf is the length of the mountain front along the foot of the mountain, and Ls is the straight-line length of the mountain front. 10 Chapter 2

References

Angelier, J., 1990. Inversion of field data in fault tectonics to obtain the regional stress - A new rapid direct inversion method by analytical means. Geophys. J. Int. 103, 363-376.

Angelier, J., 1994. Palaeostress analysis of small-scale brittle structures. Chapter 4 in: ‘Continental Deformation’, edited by P. Hancock, Pergamon Press, 421 p. (p. 53-100). In.

Angelier, J. and Mechler, P., 1977. Sur une méthode graphique de recherche des contraintes principales également utilisable en tectonique et en séismologie: la méthode des dièdres droits. Bull. Soc. Géol. France 7(6).

Keller, A. and Pinter, N., 2002. Active Tectonics, Earthquakes, Uplift, and landscape. Prentice Hall, New Jersey, United States. pp. 362

Ortner, H., Reiter, F. and Acs, P., 2002. Easy handling of tectonic data; the programs TectonicVB for Mac and TectonicsFP for Windows. Shareware and freeware in the geosciences; II, A special issue in honour of John Butler. Pergamon, New York-Oxford-Toronto, International.

Sperner, B., Ott, R. and Ratschbacher, L., 1993. Fault-striae analysis: a turbo pascal program package for graphical presentation and reduced stress-tensor calculation. Computers Geosciences 19, 1361- 1388. CHAPTER III

Ge o l o gi c a l b a c k gr o u n d

Ralph Neuwerth 12 Chapter 3

3.1. Location of study area

The study area covers parts of three Colombian departments: Quindío (city of Armenia), Risaralda (city of Pereira) and Valle del Cauca (cities of Cartago and Zarzal). It is located between the Central and Western Cordilleras at a latitude of 4.4 – 4.8°N and a longitude of 75.8 – 76.1°W and (Figs. 3.1 and 3.2).

The North Cauca Valley Basin is subdivided in two sub-basins by a fold and thrust belt (called the Serranía de Santa Barbara or SSB): the Cauca Depression to the west and the Quindío-Risaralda Basin to the east (Fig. 3.2). These two basins are infilled by subhorizontal Pleistocene sediments. The latter consist principally of fluvio-lacustrine deposits in the Cauca Depression (the so-called Zarzal Formation), which interfinger with the volcaniclastic mass-flow deposits forming the Quindío- Risaralda Fan. The latter is sourced from the Cerro Bravo-Machin volcanic complex in the Central Cordillera (Fig. 3.2).

The intramountane North Cauca Valley Basin studied here is located immediately west of the Romeral Fault System (RFS), a palaeosuture where Palaeozoic and Cretaceous continental rocks in the east

Fig. 3.1: (A) Geodynamics of NW South America: velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008). Location of study area. Abbreviations: B: Bogotá; C: Cali; CC: Central Cordillera; EC: Eastern Cordillera; WC: Western Cordillera; EFFS: Eastern Frontal Fault System; IBF: Ibagué Fault; GF: Garrapatas Fault; RFS: Romeral Fault System. (B) Digital elevation model (DEM, (USGS, 2005)) of western central Colombia showing the course of the Cauca River. The study area is located upstream of La Virginia town, at the northern termination of the Cauca Valley Basin (after (Suter et al., 2008). Geological background 13 are adjacent to Cretaceous accreted oceanic terranes in the west (Fig 3.1 and 3.3).

3.2. Geodynamic evolution of Colombia from Jurassic to recent times

The Western Colombia geological history started with an extensive phase in the Jurassic, followed by a long period of tectonic activity associated with the movement of three distinct tectonic plates: the South American, the Pacific and the Caribbean Plates. The studied area having been affected mainly by Cenozoic tectonics, the emphasis in this description has been put on the Andean Orogeny.

Triassic and Jurassic units exposed in the Colombian Andes encompass a complex association of calcareous, siliciclastic, volcaniclastic and plutonic rocks (Bayona et al., 2006). Two tectonic settings have been proposed for these Mesozoic rocks:

(1) a backarc extension occurring behind a subduction-related magmatic arc (McCourt et al., 1984a; Pindell and Erikson, 1993; Pindell and Tabbutt, 1995; Toussaint, 1995b; a; Meschede and Frisch, 1998; Pindell and Kennan, 2001; Moreno-Sanchez and Pardo-Trujillo, 2003)

(2) an intracontinental rifting regime associated with the break-up of Pangea (Pindell and Dewey John, 1982; Ross and Scotese, 1988; Cediel et al., 2003).

Three alternative hypotheses have been proposed for the processes that might have been active during the Cretaceous (Sarmiento-Rojas et al., 2006): 14 Chapter 3

1. Backarc extension (McCourt et al., 1984a; Fabre, 1987; Toussaint and Restrepo, 1989; Cooper et al., 1995; Meschede and Frisch, 1998),

2. Passive margin (Pindell and Erikson, 1993; Pindell and Tabbutt, 1995),

3. Intracontinental rifting related to the opening of the Caribbean (Cediel et al., 2003).

These hypotheses will be better developed in the next section, where the terranes present in the studied area are described.

The Andean Orogeny started with the convergence of the Pacific Plate in the western part of Colombia. It is marked by four main events:

(1) Accretion and/or obduction along the western margin of Colombia during the Cretaceous.

(2) Relative divergence to convergence between the Americas during the Paleogene to Early Eocene.

(3) Break-up of the Farallon Plate into the Nazca and Coco plates during the Oligo-Miocene.

(4) Collision of the Chocó-Panamá-Block (CPB) into the NW corner of South America during the Miocene.

(1) The emplacement of the terranes situated west of the RFS is still a matter of debate (Fig. 3.4) .For some authors they were accreted (McCourt et al., 1984a; Restrepo and Toussaint, 1988; Restrepo- Pace, 1992; Taboada et al., 1998; Ramos and Aleman, 2000; Taboada et al., 2000; Moreno-Sanchez and Pardo-Trujillo, 2003; Chicangana, 2005b; Nivia et al., 2006), whereas others think they were obducted (Bourgois et al., 1987; Kellogg et al., 1995; Kerr et al., 1997; Kerr et al., 1998; Cediel et al., 2003).

The timing of accretion/obduction is also still debated. Most authors have recognized two major accretionary/obducting episodes in the Cretaceous on the western side of the Central Cordillera

Fig. 3.3: Simplified cross-section across the Cauca depression and the Quindío-Risaralda Basin, see figure 3.2 for location. After (Suter et al., in review). Geological background 15

Fig. 3.4: Middle Cretaceous to Miocene plate reconstruction of northwestern South America showing two models: (A) arrival and oblique docking of the Pacific Terranes, followed by accretion of San Jacinto, Sinú , Guajira-Falcon, and Caribbean Mountain terranes and finally collision of the Chocó Arc. Grey shaded areas in all time slices represent paleotopographic swells, elevated and/or emergent areas. Red crosses represent magmatism (after (Cediel et al., 2003)). Abbreviations: BAU: Baudó terrane; CAM: Caribbean Montaine terrane; CG: Cañas-Gordas terrane; DA: Daguá-Piñon terrane; EC: Eastern Cordillera; GA: Garzón massif; GOR: Gorgona terrane; GU-FA: Guarija- Falcon terrane; MSP: Macacaibo sub-plate; PA: Panamá terrane: RO: Romeral terrane; SJ: San Jacinto terrane; SM: Sinú terrane. (B) backarc extension followed by accretion (after (Moreno and Pardo, 2002)). 16 Chapter 3 and in the Western Cordillera. The first one in the Aptian-Albian corresponds to the docking of the Romeral terrane (western Central Cordillera). The second one in the Late Cretaceous-Early Paleocene corresponds to the Dagua terrane (Western Cordillera) (McCourt and Feininger, 1984; Restrepo and Toussaint, 1988; Kellogg et al., 1995; Kerr et al., 1997; Ramos and Aleman, 2000; Cediel et al., 2003) (Fig. 3.4A). Other authors (Pindell and Barrett Stephen, 1990; Pindell and Kennan, 2001) have preferred a passive margin model west of Colombia with a shift in plate polarity and accretion during the Maastrichien (Fig. 3.4B).

During latest Cretaceous times, all plate tectonic interpretations propose a convergent margin model for western Colombia: the Caribbean Plate moved northeastwards with respect to South America, whereas the Farallon Plate was subducting in a SW-NE direction underneath southern Colombia (Sarmiento-Rojas et al., 2006). Consequently, the subduction was oblique with respect to the strike of the continental margin and the Romeral Fault System was marked by transpressive right-lateral faulting (Ego et al., 1995; Paris et al., 2000; Taboada et al., 2000; Cediel et al., 2003; Chicangana, 2005b). In response to the Western Cordillera accretion, deformation in the Central and Eastern Cordilleras began during Maastrichtian times (Etayo Serna, 1994; Cooper et al., 1995; Restrepo-Pace Pedro, 1999; Villamil, 1999; Cortes and Angelier, 2005; Ramon and Rosero, 2006).

(2) The Paleogene and Early Eocene period was marked by a major stratigraphical unconformity in Colombia (Restrepo-Pace, 1999; Villamil, 1999; Gomez et al., 2003). It has been interpreted as the result of uplift in the Central Cordillera (Villamil, 1999), of accretion of the San Jacinto Terrane (Taboada et al., 2000) or the Dagua-Piñon Terrane (Cediel et al., 2003), or of an increase in convergence rate (Cooper et al., 1995). More recently, Cortes et al. (2005b) have identified from paleostress analysis a change in stress regime from E-W to WSW-ENE to NW-SE during the Late Paleocene. They have interpreted this change to be (1) the result of the transition from relative divergence to convergence between the Americas and (2) responsible for the regional unconformity.

(3) During the Late Oligocene to Early Miocene, the Farallon Plate was split into the Cocos Plate in the north and the Nazca Plate in the south (Hey, 1977b; Pennington, 1981; Taboada et al., 2000; Lonsdale, 2005). Causes for this break-up are still controversial. It may result from a major plate reorganization (Lonsdale and Klitgord, 1978; Gutscher et al., 1999; Chicangana, 2005b) or from the Galapagos triple junction activation (Hey, 1977a). The Cocos Plate moved towards the NNE and the Nazca Plate towards the E. This created both a change from oblique to orthogonal and an increase in convergence between the Nazca Plate and the continental margin. This event coincided with the initiation of the Ibague Fault system and other ENE trending right-lateral strike-slip fault systems within the Northern Andes (Acosta et al., 2007). It had various tectonic consequences in Colombia: transpressive deformations along continental faults trending N-NE in Colombia (Taboada et al., 2000), accretion of the Baudo Range (Ramos and Aleman, 2000), reactivation of strike-slip faults (Ferrari and Tibaldi, 1992; Litherland and Aspden, 1992), start of an important cycle of magmatic activity (Barberi et al., 1988; Lavenu et al., 1992) as well as an increase in the Cordillera uplift rate and, consequently, the Cauca Valley development (McCourt et al., 1984a; Ramos and Aleman, 2000). Geological background 17

(4) The last main tectonic event was the collision during the Middle Miocene of the Chocó-Panamá Block (CPB) into NW South America in an E to ESE direction (Pennington, 1981; Restrepo and Toussaint, 1988; Duque-Caro, 1990; Mann and Corrigan, 1990; Van der Hilst and Mann, 1994; Kellogg et al., 1995; Taboada et al., 2000; Trenkamp et al., 2002). It was contemporary with the onset of the major Andean tectonic phase which began 10.5 my ago and continued during Pliocene- Quaternary times (Cooper et al., 1995; Kellogg et al., 1995; Taboada et al., 2000; Cediel et al., 2003; Cortes et al., 2005b). The direction of maximum horizontal stress in the northern Andes changed from NW-SE to WNW-ESE (Cortes et al., 2005b). This collision is also thought to be responsible for the rise of the Eastern Cordillera, and uplift in the Central and Western Cordilleras of Colombia (Van der Hammen, 1958; Kroonenberg et al., 1990; Gregory-Wodzicki, 2000). This collision led to the emergence of the Panamá Isthmus, which separated the Pacific and Caribbean oceans and permitted faunal and floral exchanges between the Americas (Keigwin, 1978; Marshall et al., 1979; Marshall et al., 1982; Duque-Caro, 1990; Coates et al., 1992).

Finally, Suter et al. (2008b) have proposed that the continued movement of the Chocó–Panamá Indenter may be responsible for the observed 060- oriented right-lateral distributed shear strain, and may have closed the northern part of the Cauca Valley, thereby forming the Cauca Valley Basin.

At the present day, the Colombian Andes result from the interaction of three major converging tectonic plates (Fig. 3.1.A). With respect to the South American Plate, the Caribbean Plate moves E-SE at a velocity between 10 and 22mm/yr, whereas the Nazca Plate moves eastwards at a velocity of 50 to 78mm/yr (Pennington, 1981; Freymueller et al., 1993; Ego et al., 1996; Gutscher et al., 1999; Taboada et al., 2000; Trenkamp et al., 2002; White et al., 2003). Taboada et al. (2000) define the southern limit of the subducting Carribean slab at 5.2°N, in an area where the intermediate seismicity distribution beneath the Eastern Cordillera suggests a right-lateral E–W trending transform shear zone (TSZ), whereas Pindell et al. (2005) locate the present position of the Caribbean Plate southern edge at 4°N. In the convergence zone between these three major plates, three distinct blocks are moving and being deformed in order to accommodate the resulting stress; i.e., the Chocó-Panamá, North Andes, and Maracaibo blocks (Fig. 3.1A).

3.3. Stratigraphy of the studied area

The Plio-Pleistocene sediments studied here lie unconformably over five different structural complexes: the Palaeozoic Cajamarca and Arquía Complexes, and the pre-Cenozoic Quebradagrande, Amaime, and Western Cordillera Complexes. Moreover, at the center of the basin, they overlie a thrust and fold belt made of Cenozoic continental deposits (Serranía de Santa Barbara or SSB; see figure 3.5 for localization of stratigraphical units and faults.

3.3.1. Palaeozoic

3.3.1.1. Cajamarca Complex

The Cajamaca Complex was first described by Nelson (1957). It consists of pelitic schists, 18 Chapter 3

Fig. 3.5: Compilation of the five 1:100'000 geological maps of INGEOMINAS (Caballero and Zapata, 1983; Parra, 1983; McCourt et al., 1984; Nivia et al., 1995; Estrada and Viana, 1998) which cover the study area (after (Suter et al., 2008). quartzites, marbles, amphibolites, migmatites, granulites, amphibolites, gneisses, schists, and deformed granitoids (Gonzalez, 2001; Nivia, 2001; Núñez, 2001), which reach greenschist through lower amphibolite metamorphic grade (Cediel et al., 2003). Geochemical analyses indicate these rocks to be of intraoceanic-arc and continental-margin affinity (Restrepo-Pace, 1992). They form a parautochthonous accretionary prism of Ordovician-Silurian age (Cediel et al., 2003). The Cajamarca Complex is tectonically bounded to the west by the San Jeronimo Fault, where it rests against the Quebradagrande Complex.

3.3.1.2. Arquia Complex

The Arquia Complex is bounded to the west by the Cauca-Almaguer Fault and to the east by the Silvia- Pijao fault, where it rests respectively against the Western Cordillera Complex and the Quebradagrande Geological background 19

Complex. It has been subdivided into three lithological units in the Cauca Valley geological map (Nivia, 2001): basal schists of Bugalagrande, amphibolites of Rosario and metagabbros of Bolo Azul, the former containing amphibolitic, graphitic and micaceous schists, and quartzites. The age of the metamorphics in the Arquia Complex is poorly constrained. They may have been formed either during the Neoproterozoic or the Lower Cretaceous (Nivia et al., 2006). This complex represents a multistage zone of deformation composed of Palaeozoic, Mesozoic, and Cenozoic tectonic blocks affected by Late Jurassic–Cretaceous subduction, magmatism, and/or shear processes (Moreno and Pardo, 2003).

3.3.2. Mesozoic

3.3.2.1. Quebradagrande Complex

The Quebradagrande Complex is parallel to the western margin of the Central Cordillera and was first described by Botero (1963). It is tectonically bounded to the west and east by respectively the Silvia-Pijao and San Jeronimo faults (Maya and Gonzalez, 1996b), where it rests against the Aquia and Cajamarca Palaeozoic Complexes respectively. It is composed of an assemblage of metavolcanic and metasedimentary rocks. The protoliths of the metavolcanic rocks were basaltic to andesitic lavas and pyroclastics affected by the metamorphism of zeolite, prhenite–pumpellyite, and greenschist facies (Nivia et al., 2006). The metasedimentary rocks display a wide variation in grain size, from breccias and conglomerates to coarse sandstones with clasts of cobbles and pebbles of both volcanic rocks and chert (Gómez et al., 1995). The presence of these rocks suggests underwater volcaniclastic sedimentation produced by mass movements. Both paleontological and radiometric methods have been used to date the Quebradagrande Complex. Fossils range in age from Valanginian to Albian (140–97 Ma) (González, 1980; Gómez et al., 1995) and Toussaint and Restrepo (1978) report a K/ Ar (whole-rock) age of 105 ± 10Ma in a basalt. The depositional environment is still controversial. Some authors (Pindell et al., 1988; Pindell and Barrett Stephen, 1990; Moreno and Pardo, 2003) interpret it as an accretionary pile composed of remains of a volcanic arc and some portions of the Proto-Caribbean Plate, which were accreted and obducted on the western flank of Colombia during the Late Cretaceous (Fig. 3.4B). Others (Núñez, 2001; Nivia et al., 2006) consider it as having formed in a magmatic environment of a supra-subduction zone and during the opening of an ensialic marginal basin.

3.3.2.2. Amaime Complex

The Aimaime Complex, named Amaime Formation in the Valle del Cauca department (McCourt and Aspden, 1984; Aspden and McCourt, 1986), is limited to the east by the Cauca-Almaguer Fault (Maya and Gonzalez, 1996b). It is partially overlain by Cenozoic sedimentary rocks (Cartago and La Paila Formations) in the studied area. It corresponds to a suite of basic volcanic rocks which are generally massive tholeiitic basalts with horizons of pillow lavas, locally associated with komatiitic basalts (McCourt, 1984; McCourt et al., 1984b; Nivia, 2001). The core of the Amaime Complex consists of an ophiolitic suite (e.g., Ginebra, Venus, and Los Azules ophiolitic complexes and Anserma gabbro) intruded by igneous granitoids (e.g., Buga batholith, Támesis stock, and Sabanalarga batholith; 20 Chapter 3

McCourt et al., 1984b; Moreno and Pardo, 2003). Radiometric data obtained from plutonic rocks (e.g., Buga batholith, Támesis pluton, Irra stock) suggest a Jurassic (?)–Early Cretaceous age for the older oceanic basement (Armas, 1984; McCourt, 1984; Moreno and Pardo, 2003). Pindell and Erikson (1993), Pindell and Kennan (2001) and Moreno and Pardo (2002; 2003) propose that the Amaime-Chaucha complex contains rocks that were formed during two main tectonic events: an Albian- Maastrichtian diagonal accretion of a volcanic arc, followed by a Late Cretaceous–Tertiary progressive accretion of a portion of the Caribbean Plateau (Fig. 3.4B).

3.3.2.3. Western Cordillera Complex

The Western Cordillera Complex (“obducted and/or accreted oceanic rocks and associated intrusive” in figure 3.5) includes the Barroso and Penderisco Formations (Alvarez and González, 1978), the Cisneros and Espinal Formations, and the Diabasic Group (Barrero, 1979), the Volcánica Formation (Aspden et al., 1985; Nivia, 2001), and many informal units that make correlations difficult (e.g., Bourgois, Calle et al., 1982). It is dominated by basaltic rocks of tholeiitic MORB (mid-oceanic- ridge basalt)-type affinity, and important thicknesses of flysch-type siliciclastic sediments, including chert, siltstone, and greywacke (Cediel et al., 2003). Paleontological and radiometric data indicate an Albian(?)-Maastrichtian age, probably in the Cenomanian-Maastrichtian interval (Barrero, 1979; Aspden and McCourt, 1986; Nivia, 1996; Moreno and Pardo, 2003). The Western Cordillera Complex corresponds to a portion of the Caribbean Plate that began to be accreted to the western flank of Colombia during the Late Cretaceous (Moreno and Pardo, 2003) and to accreted fragments of oceanic crust, aseismic ridges, and/or oceanic plateaus (Cediel et al., 2003).

3.3.3. Cenozoic

3.3.3.1. Cartago Formation

The Cartago Fm consists in mudstones, sandstones, pebbly sandstones and conglomerates. It is interpreted to be continental, for the most part, with only minor marine or lacustrine influences (Keith et al., 1988). After Keith et al. (1988) and Ríos and Aránzazu (1989), it can be divided into 3 members: (1) the basal La Ribera Member showing medium to coarse, quarz-rich, conglomeratic sandstones; (2) the intermediate Piedras de Moler Member constituted by grey to greenish, fine to medium sandstones intercalated with clays; and (3) the upper Miravalles Member consisting in conglomerates and medium to coarse, conglomeratic sandstones. Geological background 21

The Cartago Fm is interpreted as a fine-grained meander belt system which evolves into braided streams and possibly coarse-grained meandering channels with time. This depositional setting is associated with a humid alluvial fan system (Keith et al., 1988). Palaeocurrents measured in the through cross-bedded sandstones of the intermediate “Piedras de Moler” Member show a SSW direction of flow (Ríos and Aránzazu, 1989).

The Cartago Fm has been correlated with the Vijes and the Guachinte/Ferreira Fms (Keith et al., 1988) and with Ferreira and Amaga Fms (McCourt, 1984).

3.3.3.2. La Paila Formation

The La Paila Fm (Van der Hammen, 1958; Alvárez, 1983; McCourt, 1984; Keith et al., 1988; Ríos and Aránzazu, 1989; Gonzalez and Nuñez, 1991; Nivia et al., 1992) is also referred to as Buga Fm (Schwinn, 1969). Its age is considered as Early Miocene by McCourt (1984) based on structural and sedimentological relationships with porphyric intrusions, as Middle Miocene by Schwinn (1969) and Nivia et al. (1992), as Middle to Late Miocene based on Ar40/Ar39 dating (12,7 Ma; Suter, unpublished), or as Miocene by Van der Hammen (1958) based on palynology.

The La Paila Fm is divided into two members: (1) the basal member comprising approximately 200 m. of dacitic tuffs, testifying to the onset of volcanic activity in the Cerro Bravo-Machin volcanic system to the east; and (2) the upper member corresponding to a sequence of sandstones, pebbly sandstones, and conglomerates. Pebbles in conglomerates indicate a source area to the NE and reflect the broad geology of the Central Cordillera: diabase, tonalite, diorite, metamorphics, quarz and some black chert. A characteristic feature is the presence of silicified wood fragments (Van der Hammen, 1958; Keith et al., 1988).

The depositional setting is interpreted as a humid alluvial fan originating in the slopes of the Central Cordillera: the clasts include igneous, metamorphic and sedimentary rocks that are present in the Central and not in the Western Cordillera. The relative abundance of conglomerates within La Paila Fm outcrops decreases westward (Keith et al., 1988; Ríos and Aránzazu, 1989).

The La Paila Fm corresponds to the La Pobreza Fm named by McCourt (1984) and have been correlated with the Combia Group (Van der Hammen, 1958) or Combia Formation (McCourt, 1984) and with the Honda Fm in the Magdalena River Valley (Van der Hammen, 1958; McCourt, 1984).

3.3.4. Plio-Pleistocene Zarzal Formation and Quindío-Risaralda Fans

The first description of subhorizontal to slightly dipping deposits of the Zarzal Formation cropping out in isolated hills of the Cauca Valley was made by Boussingault (1903), who considered these sediments as lake deposits. The name of Zarzal Formation was given by Keiser (1955, in Nelson 1962) and Van der Hammen (1955, in Van der Hammen 1958), who observed clayey and sandy sediments unconformably overlying the La Paila Fm. They were considered as Pliocene (Van der Hammen, 1958; De Porta, 1974; McCourt, 1984). 22 Chapter 3

McCourt (1984), McCourt et al. (1984b in Keith et al. (1988)) and Nivia et al. (1995) subdivided the Zarzal Formation in three units. Cardona and Ortiz (1994) interpreted different depositional environments, i.e., braided-stream, floodplain and lacustrine. They recognized the interfingering between Zarzal sediments and the volcaniclastic mass flows in the Pereira-Cartago-La Virginia area (Fig. 3.2). Geological background 23

The eastern foothills and northern end of the SSB are unconformably overlain by a succession of volcaniclastic fans. Numerous studies have been carried out in the latter deposits but they never considered the entire volcaniclastic fan complex. The latter has been successively named Armenia Formation (McCourt, 1984), Glacis del Quindío (Gonzalez and Nuñez, 1991), Pereira Formation (Cardona and Ortiz, 1994), Quindío Fan (Espinosa, 2000) and finally Quindío-Risaralda Fan (Guarin et al., 2006). This complex succession of volcaniclastic deposits has a semi-conical shape. It covers an area of over 1200 km2 (Cardona and Ortiz, 1994; Guarin, 2008). These mass flow deposits are poorly dated with an age comprised between 1.3 million and 50’000 years (Guarin, 2004). Forty- five units can be recognized according to their mineralogical, sedimentological, stratigraphical and geomorphological characteristics (Fig. 3.6) (Guarin, 2008). The most representative processes consist of debris avalanches, debris flows, hyperconcentrated flows, fluvial and ash fall deposits. The source of the sediments originates from the Cerro Bravo-Machín volcanic system (Alfaro and Aguirre, 2003). The latter is also referred to as Ruiz-Tolima volcanic system (Guarin et al., 2006) and located in the Central Cordillera (Fig. 3.1). Some mass-flow deposits are interfingering with the fluvio-lacustrine Zarzal Formation east of the SSB (Suter et al., 2008a) and near Cartago (Fig. 3.2) (Cardona and Ortiz, 1994; Neuwerth et al., 2006; Suter et al., 2008a).

Recently, a sedimentological model has been proposed to explain the relationship between the Zarzal Formation and the Quindío-Risaralda Fan deposits, showing westward progression of the volcaniclastic fans and the subsequent sedimentological damming of the Cauca Depression and Quindío-Risaralda Basin, allowing the accumulation of lacustrine, diatomitic deposits (Suter et al., 2008a).

3.4. Structural geology of the studied area

The fault pattern in the study area and its surroundings is presented in figure 3.7 and table 3.1. Four faults strikes orientation are distinguished: N to NNE, NE, ENE to EENE and NW to WWNW and presented below.

3.4.1. N to NNE striking faults

The Romeral Fault System (RFS) is one of the most active and most continuous fault systems in Colombia. It outlines the suture between oceanic rocks to the west and continental basement to the east (Cline et al., 1981; McCourt and Aspden, 1983; McCourt, 1984; Restrepo and Toussaint, 1988; Restrepo-Pace, 1992; Paris and Romero, 1994; MacDonald et al., 1996; Nivia, 1996; Taboada et al., 2000; Cortes et al., 2005a; Cortes et al., 2006). This mega-shear originated during the Mesozoic, but has been intermittently reactivated since that time (Tassinari et al., 2008). During the Cretaceous, it had a right-lateral component in southwestern Colombia (Paris and Romero, 1994). When it was reactivated during the Miocene (Vinasco, 2001, in Tassinari et al. (2008)), it dipped east and exhibited left-lateral and reverse components in northwestern Colombia (Paris and Romero, 1994; Ego et al., 1995; Taboada et al., 2000). At present time , its kinematics changes between 4° and 5°N of latitude north from right-lateral in the south to left-lateral in the north (Ego et al., 1996; Taboada et al., 2000). It forms a 20- to 40-km-wide deformed belt that separates the Western from the Central Cordilleras. 24 Chapter 3

Fig. 3.7: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing faults in the studied area (modified after (Ingeominas, 1999; Nivia, 2001; Suter et al., 2008)).

It consists of subparallel N-S to NNE-SSW striking faults including the San Jerónimo, Silvia-Pijao, and Cauca-Almaguer faults (Orrego and Paris, 1999). The same classification as that of Suter et al. (2008b) is used here, i.e. the RFS is divided into three sub-systems: The Cali-Patía Fault System in the eastern foothills of the Western Cordillera, the Quebradanueva Fault System in the SSB and its foothills, and the Cauca-Almaguer Fault System in the western foothills of the Central Cordillera.

3.4.1.1. Cauca-Almaguer Fault System

The Cauca-Almaguer fault system comprises principally the San Jerónimo, Silvia-Pijao and Cauca- Geological background 25

Almaguer faults described below, and the Navarco, Armenia and Montenegro faults.

San Jerónimo fault

The San Jerónimo fault separates the Palaeozoic metamorphic Cajamarca Complex in the Central Cordillera to the east from the Quebradagrande Complex to the west (Maya and Gonzalez, 1996; Guzmán et al., 1998) (Fig. 3.5).

Silvia-Pijao fault

This fault strikes NNE and separates the Quebradagrande Complex to the east from the Arquía Complex to the west (Gonzalez and Nuñez, 1991; Maya and Gonzalez, 1996a) (Fig. 3.5).

Cauca-Almaguer fault

The Cauca-Almaguer Fault (formerly Romeral Fault sensu Case et al. (1971) and McCourt (1984)) can be traced throughout Colombia and Ecuador (Meissner et al., 1976; Duque-Caro, 1979; Feininger and Bristow, 1980; McCourt, 1984; Aspden et al., 1992). It separates the Arquia Complex to the east from the Western Cordillera Complex to the west (Maya and Gonzalez, 1996a) (Fig. 3.5). It is overlain by the Quindío-Risaralda Fan in the studied area.

3.4.1.2. Quebradanueva Fault System

The Quebradanueva Fault System is dominated by the Quebradanueva thrust and Potrerillos back- thrust will be better described below. Together, they are responsible for the uplift of the SSB as a pop-up-like structure (Suter et al., 2008b; Suter et al., in review) (Fig. 3.3). The Quebradanueva Fault System is also comprises the Sevilla fault to the west.

Potrerillos fault

It is considered as a N-S back-thrust of the Quebradanueva Fault (Guzmán et al., 1998). This fault forms in the SSB the Potrerillos thrust which outlines the contact between the La Paila (and/or La Pobreza) and Cartago Formations (Suter et al., in review).

Quebradanueva fault

It marks the contact between the Cinta de Piedra and La Paila Formations (Caballero and Zapata, 1983) and deforms the Zarzal and La Paila fomations and the Pereira-Armenia Fan from E to W (James, 1986). After Caballero and Zapata (1983), it is composed of various ramifications (Río Barbas, Quebrada Caucho - Río Cauca and Quebrada La Arenosa - La Suiza faults), forming a 10 km wide fault zone. The neotectonic activity of the Quebradanueva thrust has been underlined, at least in the southwestern foothills of the SSB (Suter et al., in review). 26 Chapter 3

3.4.1.3. Cali-Patía Fault System

It is composed by the Ansermanuevo, Dagua-Calima, Toro, Apia, Cali-Patía and La Argelia faults. The latter three are described below:

Cali-Patía Fault

The Cali-Patía (also named Cauca-Patía or Río Cauca) Fault controls partially the Cali and Patía rivers and limits the Western Cordillera to the east and the Interandean Cauca and Patía Valley to the west (Mosquera and Orrego, 1990). This fault is related to the Western Cordillera accretion during Cretaceous and Early Cenozoic times (Salinas et al., 2007). It has been interpreted as part of the Dolores Megashear System (Cline et al., 1981).

Apia Fault

It is offset by NW-trending structures (Estrada et al., 2001).

La Argelia Fault

It is thought to be associated with the emplacement of the Chocó-Panama Block (Taboada et al., 1998) and is probably connected with the Garrapatas Fault (Paris et al., 2000).

3.4.2. NE striking faults

They are geomorphologically expressed principally by the Palestina Fault, by also by the Rio Roble and Agua Bonita faults.

Palestina Fault

The Palestina Fault extends from the Department of Antioquia in the north down to the Nevado del Ruiz (volcanic) zone in the south.

The Paletina Fault is assumed to be the result of the oblique collision of oceanic crust during the Late Cretaceous (Feiniger, 1970) and has migrated from north to south since the Eocene (Acosta et al., 2007). Furthermore, this fault contributed to the migration of Paleogene, Neogene and Quaternary magmatism and is closely related to the reactivation of NW-trending faults (Acosta et al., 2007) which seem to have displaced it (Thouret et al., 1995). The older Ruiz-Tolima volcanic massif (starting at about 0.8 Ma. and ending at about 0.2 Ma. ago) is aligned on the Palestina fault and recent explosive activity has migrated towards the intersection of the Palestina strike-slip fault with the N50°W normal faults (Thouret et al., 1995; Thouret et al., 2007).

3.4.3. ENE to EENE striking faults

Rio Verde Fault

It seems to crosscut the Silvia-Pijao and Romeral Fault Systems (INGEOMINAS, 1999; Botero et Geological background 27 al., 2004).

Ibagué Fault

The Ibagué Fault is a dextral strike-slip fault which runs in an ENE direction (Paris, 1997; Paris et al., 2000; Taboada et al., 2000; Montes and Sandoval, 2001; Audemard, 2002; Acosta et al., 2004; Bohórquez et al., 2005; Montes et al., 2005; Diederix et al., 2006) with a reverse component (Marquínez, 2001; Bohórquez et al., 2005). The dextral motion of the fault could be induced by the counterclockwise rotation of the Andean block (Acosta et al., 2004). The initiation of the Ibague fault and other ENE trending right-lateral strike-slip faults coincides with the Farallon Plate break- up some 27 Ma ago, which increased the convergence rate between the Nazca and South American plates (Chicangana, 2005a; Acosta et al., 2007). The Ibagué fault crosscuts the Central Cordillera with a minimum dextral displacement of 30 km (Montes et al., 2005). Finally, at a latitude of 4.5°N, the younger, E–W trending, right-lateral, active, major Garrapatas-Ibagué Fault zone crosscuts the Western and Central Cordilleras from the Pacific coastline up to the Magdalena Valley. This set of EENE trending, right-lateral, right-stepping, “en-échelon” active faults clearly segments and affects the Romeral Fault System, which changes strike at a latitude of 4.5°N (Suter et al., 2008b).

Santa Rosa Fault

It corresponds to the applied main shear or main fault in a theoretical fault pattern applied under right-lateral shear (Suter et al., 2008b).

Garrapatas Fault

The right-lateral (Taboada et al., 1998; Paris et al., 2000; Audemard, 2002), probably reverse and westerly-dipping (Paris et al., 2000) Garrapatas Fault crosses the Western Cordillera and outlines the south-eastern termination of the CPB (Duque-Caro, 1979; Paris and Romero, 1994; Guzmán et al., 1998; Taboada et al., 2000). It is curved and its strike becomes NNE on the eastern flank of the Western Cordillera. It is proposed as the limit between the Dagua y Cañasgordas terranes (Etayo- Serna et al., 1986; Nivia, 1996) or as a palaeotransform fault in the Farallon Plate which behaved as a strike-slip zone during the late Mesozoic–Cenozoic (Barrero, 1977). Recent onshore mapping has permitted the interpretation of the Garrapatas fault as the southern lateral ramp that facilitated the obduction of the Cañas Gordas terrane (Cediel et al., 2003). Finally, The Garrapatas, Ibagué and Río Verde faults form an E–W fractured zone at a latitude of 4.5°N, crosscutting the Western and Central Cordilleras as well as the Cauca and Magdalena Valleys. This set of EENE trending, right-lateral, right-stepping, “en-échelon” active faults clearly segments and affects the Romeral Fault System, which changes strike at a latitude of 4.5°N (Suter et al., 2008b).

3.4.4. NW to WWNW striking faultsArma Fault

It is a NW-trending fault which cuts basement rocks in the Central Cordillera and Upper Magdalena Valley (Acosta et al., 2007) and is responsible for the structural control of the Cauca River south of Medellín (López et al., 2006). This fault may be a tensional fault that developed during the Late 28 Chapter 3

Miocene when the Combia volcanics were deposited in the structural low existing in that part of the Cauca canyon (Page, 1986).

Las Cañas fault

This NW-trending left-lateral fault (Nivia et al., 1995; López, 2006) offsets the Cauca River at the latitude of Zarzal and deflects the Tertiary sequence of the Serranía de Santa Barbara (López, 2006).

The Belalcazar, Otún, Consota and Salento faults belong also to NW to WWNW striking faults but are poorly studied (Table 3.1).

3.5. The Cauca Depression

The Cauca Depression is part of the Cauca-Patía Valley, also named Graben Interandino, Cauca Patía (Acosta, 1978) or the Foso Cauca-Patía (Stutzer, 1934), which belongs to the Interandean Depression or Graben, extending for some 1,000 km from Guayaquil in Ecuador up to northern Colombia (Hörmann and Pichler, 1982).

Few models have been proposed to explain the Cauca Depression formation (Fig. 3.2). It has been interpreted as a graben-like basin (Acosta, 1978; McCourt et al., 1984a; Droux and Delaloye, 1996; MacDonald et al., 1996) or a left-lateral transtensional pull-apart basin (Kellogg et al., 1983; Alfonso et al., 1994). Recent studies have demonstrated an active, compressional, E-W trending tectonic regime in the Cauca Valley some 50 km north of Cali, which generated the thrusting of Tertiary over Quaternary sediments (López et al., 2005; López and Moreno, 2005). Radial and pure compression with horizontal σ1 and a WNW–ESE dominant direction has been reported by Suter et al. (2008b) in the SSB, east of the Cauca Valley. It has also been proposed that the uplift of the SSB is responsible for the tectonic closure of the Cauca Depression and, thereby, for its recent infill from Cali up to the Riseralda River Valley north of La Virginia (Suter et al., in review). Geological background 29

N to NNE striking faults (Cauca-Almaguer Fault System) (see section 4.1.1)

Fault Author (s) Kinematics Strike Dip Azimuth

SAN JERÓNIMO James (1986) Inverse N 15 E Vertical to 80° to the E Left-lateral? González and Núñez (1991) Inverse Map High angle to the E Right-lateral Ingeominas (1999a) Inverse N 5 W 75° to the E Left-lateral París (1997) Right-lateral Map Montes and Sandoval (2001a) Inverse NE 75° to the E Left-lateral Vergara et al. (2001) Left-lateral N 15 E High angle to the E Normal Gallego and Ospina (2003) Right-lateral Normal Bohórquez et al. (2005) Inverse NNE 65° to the E

NAVARCO James (1986) Eastern side higher N 35 E Vertical París et al. (2000) Left-lateral N 18.5 E Vertical

SILVIA-PIJAO París (1997) Inverse High angle to the E Right-lateral McCourt et al. (1984) MAP Inverse N 20 E High angle Right-lateral McCourt et al. (1985) Inverse High angle Right-lateral González and Núñez (1991) Right-lateral High angle to the E Inverse Guzman et al. (1998) N to NNE Ingeominas (1999a) Inverse N to NE E Right-lateral Montes and Sandoval (2001a) Inverse N to NE E Right-lateral Vergara et al. (2001) Left-lateral N 15 E SE Normal Botero et al. (2004) Vertical? N 15 to 25 E Minor right-lateral Ospina (2007) Left-lateral N to NNE High angle to the W Normal 30 Chapter 3

Bohórquez et al. (2005) Inverse NNE 60° to the E

ARMENIA James (1986) N 35 E Vertical? París (1997) Left-lateral N 20 E NW Guzman et al. (1998) NE to N W Ingeominas (1999) Right-lateral N 10 E W Inverse París et al. (2000a) Left-lateral N 23 E High angle to the W Normal Montes and Sandoval (2001a) Left-lateral N 10 E W to Vertical Vergara et al. (2001) Left-lateral NE High angle to the SW Normal Lalinde (2004) Inverse N 50 E Vertical Bohórquez et al. (2005) Inverse NNE 60° to the NW

CAUCA-ALMAGUER Guzman et al. (1998) Inverse ? NNE E Ingeominas (1999a) Inverse N 25 E Vertical Montes and Sandoval (2001a) Inverse N 25 E Vertical Bohórquez et al. (2005) Inverse N 50° to the E Left-lateral Tassinari et al. (2008) Left-lateral E

MONTENEGRO París (1997) Left-lateral N 20 E NW Inverse Guzman et al. (1998) Left-lateral N W Inverse Ingeominas (1999a) Inverse N 10 E E Right-lateral París et al. (2000) Left-lateral N 25 E High angle to the W Normal Montes and Sandoval (2001a) Right-lateral N 10 E E Inverse Vergara et al. (2001) Left-lateral N 20 E High angle to the SE Normal

N to NNE striking faults (Quebradanueva Fault System) (see section 4.1.2)

Fault Author (s) Kinematics Strike Dip Azimuth

SEVILLA González and Núñez (1991) Approximate Geological background 31

Cardona and Ortiz (1994) N15 E Lopez-Ramos (2003) Inverse

POTRERILLOS Guzman et al. (1998) Inverse W Lalinde (2004): ALCALA 1: W side down N 10 E Vertical ALCALA 2: Normal NS 74° to the E

QUEBRADANUEVA Caballero and Zapata (1983) Inverse High angle to the E James (1986) Inverse N 15 to 25 E 60° to 70° to the E Left-lateral E Guzman et al. (1998) Inverse NNNE E Left-lateral Pardo et al. (1994) Right-lateral N 15-20 E Bohórquez et al. (2005) Inverse NNE 60° to the E

N to NNE striking faults (Cali-Patía Fault System) (see section 4.1.3)

Fault Author (s) Kinematics Strike Dip Azimuth

CALI-PATÍA Orrego and Paris (1999) Reverse E Mantilla and Arias (2001) W Lopez-Ramos (2003) W Paris et al. (1989) Right-lateral Taboada et al. (1998) Right-lateral Montes et al. (2001) Right-lateral Rovida and Tibaldi (2005) Right-lateral

ANSERMANUEVO James (1986) Left-lateral N 20 E 75° to the W Pardo et al. (1994) Right-lateral N 15-20 E Nivia et al. (1995) Left-lateral Map

Guzman et al. (1998) Left-lateral NNE to N E Minor normal París et al. (2000) Left-lateral N 6.6 E E Bohórquez et al. (2005) Inverse NNE 60° to the W

DAGUA-CALIMA París et al. (2000) Normal N18.8°E 70° to the E-SE Taboada et al. (1998) Normal López (2006) Left-lateral

TORO Caballero y Zapata (1983) Inverse N 20 E W 32 Chapter 3

París et al. (2000) Left-lateral N6.6°E E? Taboada et al. (1998) Inverse E Ingeominas (1999a) Inverse 267 60° to the W

APIA Ingeominas (1999a) Inverse 135° 60°to the W Left-lateral Taboada et al. (1998) Inverse E Estrada et al. (2001) Left-lateral N

LA ARGELIA París et al. (2000) Right-lateral N14.5°E Inverse W Taboada et al. (1998) Right-lateral NE Inverse Ingeominas (1999a) Right-lateral 120° 60° to the W Inverse

NE striking faults (see section 4.2)

Fault Author (s) Kinematics Strike Dip Azimuth

PALESTINA Feiniger (1970) Right-lateral N to NNE Page (1986) Likely left-lateral during Quaternary James (1986) Left-lateral since N 50 E Vertical late Tertiary París (1997) Left-lateral inverse N 10 E Right lateral to the north (in Antioquia) Guzman et al. (1998) Inverse NE 75° to the NW Left-lateral París et al. (2000) Inverse N 17.8 E ± 11° Moderate to high angle Left-lateral to the W Bohórquez et al. (2005) Inverse NE 75° to the NW Left-lateral Ospina (2007) Left- Lateral NNE

RIO ROBLE Guzman et al. (1998) Inverse NE 55° to the NW

AGUA BONITA Guzman et al. (1998) Inverse NE 55° to the NW Geological background 33

ENE to EENE striking faults (see section 4.3)

Fault Author (s) Kinematics Strike Dip Azimuth

RÍO VERDE Cardona and Ortiz(1994) N 83 E Botero et al. (2004) Likely right-lateral N 85 E

IBAGUÉ París (1997) Right-lateral ENE Wrench fault, S-side up Taboada et al. (2000) Right-lateral ENE París et al. (2000) Right-lateral N 67.9 E ± 11° Vertical, S-side up Slightly oblique Montes and Sandoval (2001a) N 80 E Vertical, S-side up Marquinez (2001) Right-lateral Map Steep at the surface, Inverse decreasing at depth Audemard (2002) Right-lateral ENE Bohórquez et al. (2005) Inverse NE 70° to the NW Right-lateral Montes et al. (2005a) Right-lateral ENE Montes et al. (2005b) Right-lateral N 70 E (Suter et a., 2008b) Right-lateral EENE En-échelon

SANTA ROSA James (1986) Left-lateral (?) N 70 E 75° to the SE Normal (?) Cardona and Ortiz (1994) N 65 to 70 E 75° to the E Guzman et al. (1998) Normal observed, NE NW Inverse right-lateral expected 34 Chapter 3

Bohórquez et al. (2005) Left-lateral N 70 E 75° to the SE Normal Ingeominas (1999a) Inverse 240° 70° to the NE Right-lateral

GARRAPATAS Taboada et al. (2000) Right-lateral ENE París et al. (2000) Inverse, probably N 60.8 E ± 14° 50° to the NW right-lateral Audemard (2002) Right-lateral ENE Ingeominas (1999a) Right-lateral 35°

NW to WWNW striking faults (see section 4.4)

Fault Author (s) Kinematics Strike Dip Azimuth

ARMA Guzman et al. (1998) Normal NW 70° to the NE Right-lateral Bohórquez et al. (2005) Left-lateral Acosta et al. (2007) Left-lateral NW

BELALCAZAR Guzman et al. (1998) Normal NW 70° to the NE Right-lateral

OTÚN James (1986) Lineament N 40 W Vertical (?) Cardona and Ortiz (1994) Lineament N 56 W Guzman et al. (1998) Lineament

CONSOTA James (1986) Lineament N 40 W Vertical (?) Guzman et al. (1998) Normal NW 65° to the SW Right-lateral Ingeominas (1998) NW Bohórquez et al. (2005) Normal NW 65° to the SW Right-lateral

SALENTO González and Núñez (1991) Right-lateral N 81.5 W Vertical (Map) Guzman et al. (1998) Normal EW 80° to the N Right-lateral Ingeominas (1999) Normal N 42 W 80° to the N Right-lateral

Table 3.1: Compilation of published literature about fault characteristics in the study area and its surroundings. These faults are located in Fig. 3.7 (modified after (Suter et al., 2008)). Geological background 35

References

Acosta, C. E., 1978. El graben interandino Colombo-Ecuatoriano (Fosa Tectónica del Cauca Patía y el corredor Andino-Ecuatoriano). Boletín de geología, Universidad Industrial de Santander 12(63- 199).

Acosta, J., Lonergana, L. and Coward, M. P., 2004. Oblique transpression in the western thrust front of the Colombian Eastern Cordillera. Journal of South American Earth Sciences 17, 181–194.

Acosta, J., Velandia, F., Lonergan, L. and Mora, H., 2007. Strike-slip deformation within the Colombian Andes. Deformation of the Continental Crust: The Legacy of Mike Coward. A. C. Ries, Butler, R. W. H. & Graham, R. H. (eds). In Geological Society, London, Special Publications. 272, 303-319.

Alfaro, C. and Aguirre, A. S., 2003. Geoquímica de fuentes minerales y thermales del complejo volcánico Cerro Bravo-Machín, Colombia. IX Congreso Colombiano de Geología, Medellin.

Alfonso, C. A., Sacks, P. E., Secor, D. T., Rine, J. and Pérez, V., 1994. A Tertiary fold and thrust belt in the Valle del Cauca Basin, Colombian Andes. Journal of South American Earth Sciences 7, 387- 402.

Alvarez, E. and González, H., 1978. Geología y geoquímica del cuadrángulo I-7 Urrao, escala 1:100,000, Ingeominas, Medellín.

Alvárez, J. A., 1983. Geología de la Cordillera Central y Occidente Colombiano y petroquímica de los Intrusivos granitoides mesocenozoicos. Boletín Geológico INGEOMINAS Bogotá 26, 175.

Armas, M. d., 1984. Mapa geológico preliminar, Plancha 261-Tuluá, escala 1: 100,000, Ingeominas, Cali.

Aspden, J. A., Fortey, N., Litherland, M., Viteri, F. and Harrison, S. M., 1992. Regional S-type granites in the Ecuadorian Andes - possible remnants of the breakup of western Gondwana. Journal of South American Earth Sciences 6, 77-96.

Aspden, J. A. and McCourt, W. J., 1986. Mesozoic oceanic terrane in the Central Andes of Colombia. Geology (Boulder) 14(5), 415-418. 36 Chapter 3

Aspden, J. A., Nivia, A. and Millward, D., 1985. Mapa geológico de la Plancha 279- Dagua, escala 1: 100,000, Ingeominas, Bogota.

Audemard, F. A., 2002. Ruptura de los grandes sismos históricos venezolanos de los siglos XIX y XX revelados por la sismicidad instrumental contemporánea. XI Congreso Venezolano de Geofísica, edited by S.V.d.I. Geofísicos, pp. 8, Caracas.

Barberi, F., Coltelli, M., Ferrara, G., Innocenti, F., Navarro, J. M. and Santacroce, R., 1988. Plio- Quaternary volcanism in Ecuador. Geological Magazine 125(1), 1-14.

Barrero, D., 1977. Geology of the Central Western Cordillera, west of Buga and Roldanillo, Colombia: Ph. D. Dissertation, Colorado School of Mines, Golden, 148 p.

Barrero, D., 1979. Geology of the Central Western Cordillera, West of Buga and Roldanillo, Colombia. Publicaciones Geológicas Especiales del Ingeominas 4, 1-75.

Bayona, G., Rapalini Augusto, E. and Costanzo-Alvarez, V., 2006. Paleomagnetism in Mesozoic rocks of the Northern Andes and its implications in Mesozoic tectonics of northwestern South America. Earth Planets and Space 58(10), 1255-1272.

Bohórquez, O. P., Monsalve, M. L., Velandia, F., Gil, C. F. and Mora, H., 2005. Marco tectónico de la cadena volcánica más septentrional de la Cordillera Central de Colombia. Boletin de Geología, Universidad Industrial de Santander (UIS) 27, 55-80.

Botero, A., 1963. Contribución al conocimiento de la geología de la zona central de Antíoquia. Anales Facultad de Minas (Medellín) 57, 101.

Botero, P., García, L., Hernández, C., Sepúlveda, L. and Torres, E., 2004. Depósitos Cuaternarios Del Frente Montañoso Occidental de La Cordillera Central entre los Municipios De Calarcá Y Pijao (Quindío-Colombia): Una Evidencia de Actividad Tectónica Reciente. IV JORNADAS DE SISMOLOGÍA HISTÓRICA. V SIMPOSIO VENEZOLANO DE HISTORIA DE LAS GEOCIENCIAS. Trujillo, Venezuela, Boletín de historia de las geociencias en Venezuela Boletín de Historia de las Geociencias en Venezuela. 94: 1-7.

Bourgois, J., Toussaint, J.-F., Gonzalez, H., Azema, J., Calle, B., Desmet, A., Murcia Luis, A., Acevedo Alvaro, P., Parra, E. and Tournon, J., 1987. Geological history of the Cretaceous ophiolitic complexes of northwestern South America (Colombian Andes). Tectonophysics 143(4), 307-327. Geological background 37

Boussingault, J. B., 1903. Memorias. Biblioteca Nacional de Colombia. Bogotá.

Caballero, A. and Zapata, G., 1983. Mapa geológico de Colombia, Plancha N°224 - Pereira. INGEOMINAS, Medellín.

Case, J., Duran, L., López, A. and Moore, W., 1971. Tectonic Investigations in Western Colombia and Eastern Panamá. Geol. Soc. Am. Bull. 82(10), 2684-2711.

Cediel, F., Shaw, R. P. and Caceres, C., 2003. Tectonic assembly of the northern Andean Block. The circum-Gulf of Mexico and the Caribbean; hydrocarbon habitats, basin formation, and plate tectonics. C. Bartolini, T. Buffler Richard and F. Blickwede Jon. In American Association of Petroleum Geologists. Tulsa, OK, United States.

Chicangana, G., 2005a. The Romeral Fault System: a shear and deformed extinct subduction zone between oceanic and continental lithospheres in northwestern South America. Earth Sciences Research Journal 9(1), 51 -66.

Chicangana, G., 2005b. The Romeral Fault System: a shear and deformed extinct subduction zone between oceanic and continental lithospheres in Northwestern South America. Earth Sci. Res. J. 9(1), 51-66.

Cline, K. M., Page, W. D., Gilliam, M. L., Cluff, L. S., Arias, L. A., Benalcázar, L. G. and López, J. H., 1981. Quaternary activity on the Romeral and Cauca Faults, northwest Colombia. Revista CIAF 6(115-116).

Coates, A. G., Jackson, J. B. C., Collins, L. S., CroninT.M., Dowsett, H. J., Bybell, L. M., Jung, P. and Obando, J. A., 1992. Closure of the Isthmus of Panama; the near-shore marine record of Costa Rica and western Panama. Geological Society of America Bulletin 104(7), 814-828.

Cooper, M. A., Addison, F. T., Alvarez, R., Coral, M., Graham, R. H., Hayward, A. B., Howe, S., Martinez, J., Naar, J., Penas, R., Pulham, A. J. and Taborda, A., 1995. Basin development and tectonic history of the Llanos Basin, Eastern Cordillera, and middle Magdalena Valley, Colombia. AAPG Bulletin 79(10), 1421-1443. 38 Chapter 3

Cortes, M. and Angelier, J., 2005. Current states of stress in the northern Andes as indicated by focal mechanisms of earthquakes. Tectonophysics 403(1-4), 29-58.

Cortes, M., Angelier, J. and Colletta, B., 2005a. Paleostress evolution of the Northern Andes (Eastern Cordillera of Colombia); implications on plate kinematics of the South Caribbean region. Tectonics 24 1-27.

Cortes, M., Angelier, J. and Colletta, B., 2005b. Paleostress evolution of the Northern Andes (Eastern Cordillera of Colombia); implications on plate kinematics of the South Caribbean region. Tectonics 24(1).

Cortes, M., Colletta, B. and Angelier, J., 2006. Structure and tectonics of the central segment of the Eastern Cordillera of Colombia. Journal of South American Earth Sciences 21(4), 437-465.

De Porta, J., 1974. Lexique stratigraphique international. C. N. d. l. r. scientifique. Paris. In. ,5 4b.

Diederix, H., Audemard, F., Osorio, J., Montes, N., Velandia, F. and Romero, J. A., 2006. Modelado morfotectónico de la falla transcurrente de Ibagué, Colombia. Revista de la Asociación Geológica Argentina 61(4), 492-503.

Droux, A. and Delaloye, M., 1996. Petrography and geochemistry of Plio-Quaternary calc-alkaline volcanoes of Southwestern Colombia. Journal of South American Earth Sciences 9, 27-41.

Duque-Caro, H., 1979. Major structural elements and evolution of northwestern Colombia. In Watkins, J.E. (Ed.) Geological and geophysical investigations of continental margins. American Association of Petroleum Geologist Memoir 29, 329-351.

Duque-Caro, H., 1990. The Choco Block in the northwestern corner of South America; structural, tectonostratigraphical, and paleogeographic implications. Journal of South American Earth Sciences 3(1), 71-84.

Ego, F., Sebrier, M., Lavenu, A., Yepes, H. and Egues, A., 1996. Quaternary state of stress in the Northern Andes and the restraining bend model for the Ecuadorian Andes. Geodynamics of the Andes. F. Dewey John and H. Lamb Simon. Amsterdam, Netherlands. In Elsevier. 259; 1-3, 101- 116.

Ego, F., Sebrier, M. and Yepes, H., 1995. Is the Cauca-Patia and Romeral fault system left- or right- Geological background 39 lateral? Geophysical Research Letters 22(1), 33-36.

Espinosa, A., 2000. Mapa preliminar de fallas activas del Quindío. Primera fase: zona central. Armenia, Universidad del Quindío, Facultad de Ingenieria.

Etayo-Serna, F., Barrero, D., Lozano, H. Q., Espinosa, A., Gonzalez, H., Orrego, A., Ballesteros, I. T., Forero, H. O., Ramirez, C. Q., Zambrano-Ortiz, F., Duque-Caro, H., Vargas, R. H., Nuñez, A., Alvarez, J., Ropain, U. C., Cardozo, E. P., Galvis, N., Sarmiento, L. R., Albers, J. P., Case, J. E., Singer, D. A., Bowen, R. W., Berger, B. R., Cox, D. P. and Hodges, C. A., 1986. Mapa de terrenos geológicos de Colombia: Publicaciones especiales, Ingeominas, v. 14: Bogotá, 235 p.

Etayo Serna, F., 1994. Epilogo; a modo de historia geologica del Cretacico en el valle superior del Magdalena. Translated Title: Epilogue; the Cretaceous geological history in the upper Magdalena Valley. Pages: XX.

Fabre, A., 1987. Tectonique et géneration d’hydrocarbures: Un modèle de l’évolution de la Cordillère Orientale de Colombie et du Bassin des Llanos pendant le Crétacé et le Tertiaire. Arch. Sci. Genève 40 (Fasc. 2), 145-190.

Feiniger, T., 1970. The Palestina Fault, Colombia. Geological Society of America Bulletin 81, 1201- 1216.

Feininger, T. and Bristow, C. R., 1980. Cretaceous and Paleogene geologic history of Coastal Ecuador. Geologische Rundschau 69, 849-874.

Ferrari, L. and Tibaldi, A., 1992. Recent and active tectonics of the north-eastern Ecuadorian Andes. Journal of Geodynamics 15, 1-2.

Freymueller, J. T., Kellogg, J. N. and Vega, V., 1993. Plate motions in the North Andean region. Journal of Geophysical Research, B, Solid Earth and Planets 98(12), 21,853-21,863.

Gómez, A., Moreno, M. and Pardo, A., 1995. Edad y origen de Complejo metasedimentario de Aranzazu-Manizales en los alrededores de Manizales (Departamento de Caldas, Colombia). Geología Colombiana 19, 83-93.

Gomez, E., Jordan Teresa, E., Allmendinger Richard, W., Hegarty, K., Kelley, S. and Heizler, M., 2003. Controls on architecture of the Late Cretaceous to Cenozoic southern Middle Magdalena 40 Chapter 3

Valley Basin, Colombia. Geological Society of America Bulletin 115(2), 131-147.

Gonzalez, H., 2001. Geología de las planchas 206 Manizales y 225 Nevado del Ruíz, INGEOMINAS, Colombia.

González, H., 1980. Geología de las Planchas 167 (Sonsón) y 187 (Salamina). Ingeominas, Bol. Geol., 23(1):1-174. Bogotá.

Gonzalez, H. and Nuñez, A., 1991. Mapa Geológico generalizado del departamento del Quindío. Bogotá, Ingeominas.

Gregory-Wodzicki, K. M., 2000. Uplift history of the Central and Northern Andes; a review. Geological Society of America Bulletin 112(7), 1091-1105.

Guarin, F., 2009. Etude sédimentologique du cône volcanoclastique du Quindío-Risaralda (Colombie Centrale) et sa relation avec la morphotectonique. Terre et environnement, Section des Sciences de la Terre, Université de Genève. Ph.D. thesis.

Guarin, F., in press. Etude sédimentologique du cône volcanoclastique du Quindío-Risaralda (Colombie Centrale) et sa relation avec la morphotectonique. Terre et environnement, Section des Sciences de la Terre, Université de Genève. Ph.D. thesis.

Guarin, F., Gorin, G. and Espinosa, A., 2006. A Pleistocene stacked succession of volcaniclastic mass flows in central Colombia; the Quindio-Risaralda Fan. Debris avalanche and debris flows in volcanic terrains; origins, behaviors, and mitigation. In ALLARD et al. 16, 109-124.

Guarin, F., Gorin, G., Espinosa, A, 2004. A Plio-Pleistocene succession of volcanic massflows in Central Colombia: the Quindío-Risaralda fan. IAVCEI General Assembly, Pucón, Chile.

Gutscher, M. A., Malavieille, J., Lallemand, S. and Collot, J. Y., 1999. Tectonic segmentation of the North Andean margin; impact of the Carnegie Ridge collision. Earth and Planetary Science Letters 168(3-4), 255-270.

Guzmán, J., Franco, G., Ochoa, M., Paris, G. and Taboada, A., 1998. Proyecto para la mitigación del riesgo sísmico de Pereira Dosquebradas y Santa Rosa de Cabal: Evaluación neotectónica. Informe Final, Corporación Autónoma Regional de Risaralda, Pereira, Colombia. Geological background 41

Hey, R., 1977a. Tectonic evolution of the Cocos-Nazca spreading center. Geological Society of America Bulletin 88(10), 1404-1420.

Hey, R., 1977b. Tectonic evolution of the Cocos-Nazca spreading center. Geological Society of America Bulletin 88, 1404-1420.

Hörmann, P. K. and Pichler, H., 1982. Geochemistry, petrology and ongin of the Cenozoic volcanic rocks of the northern Andes in Ecuador. Journal of Volcanology and Geothermal Research 12, 259- 282.

Hubach, E. and Alvarado, B., 1934. Geología de los Departamentos del Valle y del Cauca, en Especial del Carbón, pp. 239, Servicio Geológico Nacional, Bogotá.

INGEOMINAS, 1999. Informe Técnico Científico del Terremoto del Quindío.

James, M. E., 1986. Estudio sismotectónico en el area del Viejo Caldas. Informe nr. 2008. Medellín, Colombia, INGEOMINAS: 113.

Keigwin, L. D., Jr., 1978. Pliocene closing of the Isthmus of Panama, based on biostratigraphical evidence from nearby Pacific Ocean and Caribbean Sea cores. Geology Boulder 6(10), 630-634.

Keith, J. F., Rine, J. M. and Sacks, P. E., 1988. Frontier Basins of Colombia, Valle del Cauca, University of South Carolina, Earth Sciences and Resources Institute, Field report: 267 p.

Kellogg, J., Godley, V. M., Ropain, C. and Bermudez, A., 1983. Gravity anomalies and tectonic evolution of northwestern South America. Caribbean Geological Conference 10th. Cartagena, Colombia: 18-31.

Kellogg, J. N., Vega, V., Stallings, T. C. and Aiken, C. L. V., 1995. Tectonic development of Panama, Costa Rica, and the Colombian Andes; constraints from Global Positioning System geodetic studies and gravity. Geologic and tectonic development of the Caribbean Plate boundary in southern Central America. P. Mann. Boulder, CO, United States. In Geological Society of America (GSA). 295, 75- 90.

Kerr, A. C., Marriner, G. F., Tarney, J., Nivia, A., Saunders, A. D., Thirlwall, M. F. and Sinton, C. W., 1997. Cretaceous basaltic terranes in western Colombia; elemental, chronological and Sr-Nd isotopic constraints on petrogenesis. Journal of Petrology 38(6), 677-702. 42 Chapter 3

Kerr , A. C., Tarney, J., Nivia, A., Marriner, G. F. and Saunders, A. D., 1998. The internal structure of oceanic plateaus; inferences from obducted Cretaceous terranes in western Colombia and the Caribbean. Tectonophysics 292, 3-4.

Kroonenberg, S., Bakker, J. G. M. and Van del Wiel, A. M., 1990. Late Cenozoic uplift and paleogeography of the Colombian Andes: Constraints on the development of high-Andean biota. Geolologie en Mijnbouw 69, 279-290.

Lavenu, A., Noblet, C., Bonhomme, M. G., Egueez, A., Dugas, F. and Vivier, G., 1992. New K-Ar age dates of Neogene and Quaternary volcanic rocks from the Ecuadorian Andes; implications for the relationship between sedimentation, volcanism and tectonics. Journal of South American Earth Sciences 5, 3-4.

Litherland, M. and Aspden, J. A., 1992. Terrane-boundary reactivation; control on the evolution of the Northern Andes. Journal of South American Earth Sciences 5(1), 71-76.

Lonsdale, P., 2005. Creation of the Cocos and Nazca plates by fission of the Farallon plate. Tectonophysics 404(237-264).

Lonsdale, P. and Klitgord, K. D., 1978. Structure and tectonic history of the eastern Panama Basin. Geological Society of America Bulletin 89(7), 981-999.

López, A. C., 2006. Analisís de deformación tectónica en el piedemonte de las Cordilleras Central y Occidental , Valle del Cauca, Colombia - contribuciones paleosísmicas. Medellín, Colombia, Universidad EAFIT: 113.

López, A. C., Sierra, G. M. and Ramírez, D. A., 2006. Vulcanismo neógeno en el suroccidente antioqueño y sus implicaciones tectónicas. Boletín de Ciencias de la Tierra, Medellín (ISSN 0120- 3630) 19, 27-42.

López, M. C., Audemard, F. A. and Velásquez, A., 2005. Evidencias geomorfológicas y estratigráficas de compresión Holocena en el Valle del Cauca (Colombia). X Congreso colombiano de geología. Bogotá. Colombia.

López, M. C. and Moreno, M., 2005. Tectónica y sedimentación en el piedemonte occidental de la Cordillera Central de Colombia. Un ejemplo en la cantera El Vínculo. X Congreso colombiano de geología. Bogotá, Colombia. Geological background 43

MacDonald, W. D., Estrada, J. J., Sierra, G. M. and Gonzalez, H., 1996. Late Cenozoic tectonics and paleomagnetism of North Cauca Basin intrusions, Colombian Andes; dual rotation modes. Tectonophysics 261(4), 277-289.

Mann, P. and Corrigan, J., 1990. Model for late Neogene deformation in Panama. Geology Boulder 18(6), 558-562.

Marquínez, G., 2001. Modelamiento del Abanico de Ibagué con métodos geofísicos, implicaciones hidrogeológicas y estructurales, Departamento de Tolima, Colombia. VIII Congreso Colombiano de Geología, Manizales, Colombia.

Marshall, L. G., Butler, R. F., Drake, R. E., Curtis, G. H. and Tedford, R. H., 1979. Calibration of the Great American Interchange. Science 204(4390), 272-279.

Marshall, L. G., Webb, S. D., Sepkoski, J. J. J. and Raup, D. M., 1982. Mammalian evolution and the Great American Interchange. Science 215, 1351-1357.

Maya, M. and Gonzalez, H., 1996a. Unidades litodémicas en la Cordillera Central de Colombia. Boletín Geológico INGEOMINAS 35, 43-57.

Maya, M. and Gonzalez, H., 1996b. Unidades litodémicas en la Cordillera Central de Colombia. Boletín Geológico, INGEOMINAS 32(2-3), 43-57.

McCourt, W. J., 1984. The geology of the Central Cordillera in the departments of Valle del Cauca, Quindío and NW Tolima. Cali, INGEOMINAS, Brit. Geol. Surv.: 37.

McCourt, W. J. and Aspden, J. A., 1983. Modelo tectónico de placas para la evolución fanerozóica de Colombia central y del sur. INGEOMINAS, Cali, Colombia, p. 36.

McCourt, W. J. and Aspden, J. A., 1984. A plate tectonic model for the phanerozoic evolution of Central and Southern Colombia: 10th Caribbean Geological Conference Transactions, p. 38-47.

McCourt, W. J., Aspden, J. A. and Brook, M., 1984a. New geological and geochronological data from the Colombian Andes; continental growth by multiple accretion. Journal of the Geological Society of London 141(5), 831-845. 44 Chapter 3

McCourt, W. J. and Feininger, T., 1984. High-pressure metamorphic rocks in the Central Cordillera of Colombia. BGS Report 16(1), 28-35.

McCourt, W. J., Mosquera, D., Nivia, A. and Nuñez, A., 1984b. Mapa Geológico Preliminar, Plancha 243 - Armenia. INGEOMINAS. Bogotá.

Meissner, R. O., Flueh, G., Stibane, F. and Berg, E., 1976. Dynamics of the active plate boundary in Southwest Colombia according to recent geophysical measurements. Tectonophysics 35, 115-136.

Meschede, M. and Frisch, W., 1998. A plate-tectonic model for the Mesozoic and early Cenozoic history of the Caribbean Plate. Tectonophysics 296, 3-4.

Montes, C., Montes, C., Hatcher, J., Robert, D. and Restrepo-Pace, P. A., 2005. Tectonic reconstruction of the northern Andean blocks: Oblique convergence and rotations derived from the kinematics of the Piedras–Girardot area, Colombia. Tectonophysics 399, 221- 250.

Montes, N. and Sandoval, A., 2001. Base de datos de fallas activas de Colombia.

Moreno-Sanchez, M. and Pardo-Trujillo, A., 2003. Stratigraphicalal sedimentological constraints on western Colombia; implications on the evolution of the Caribbean Plate. AAPG Memoir 79, 891- 924.

Moreno, M. and Pardo, A., 2002. Historia geológica del Occidende colombiano. Geo-Eco-Trop 26, 91-113.

Moreno, M. and Pardo, A., 2003. Stratigraphicalal sedimentological constraints on western Colombia; implications on the evolution of the Caribbean Plate. AAPG Memoir 79, 891-924.

Mosquera, D. and Orrego, A., 1990. Simposios de geología regional (Cordillera Central y Occidental). Informe inédito - INGEOMINAS.

Nelson, H. W., 1957. Contribution to the geology of the Central and Western Cordillera of Colombia in the sector between Ibague and Cali. Leidse Geologische Medelelingen 22, 1-76.

Nelson, H. W., 1962. Contribución al conocimiento de la Cordillera Occidental. Sección entre Cali- Buenaventura. Boletín Geologíco INGEOMINAS, Cali, 10. Geological background 45

Neuwerth, R., Suter, F., Guzman, C. A. and Gorin, G. E., 2006. Soft-sediment deformation in a tectonically active area: The Plio-Pleistocene Zarzal Formation in the Cauca Valley (Western Colombia). Sedimentary Geology 186(1-2), 67-88.

Nivia, A., 1996. The Bolivar mafic-ultramafic complex, SW Colombia; the base of an obducted oceanic plateau. Andean geodynamics. P. Soler and J. Dewey. Oxford, United Kingdom. In Pergamon. 9; 1-2, 59-68.

Nivia, A., 2001. Mapa geológico del departamento del Valle del Cauca, memoria explicativa, INGEOMINAS, Colombia.

Nivia, A., Galvis, N. and Maya, M., 1992. Mapa geológico de Colombia, plancha 242 - Zarzal, Cali.

Nivia, A., Galvis, N. and Maya, M., 1995. Mapa geológico de Colombia, plancha 242 - Zarzal. Cali.

Nivia, A., Marriner, G. F., Kerr, A. C. and Tarney, J., 2006. The Quebradagrande Complex: A Lower Cretaceous ensialic marginal basin in the Central Cordillera of the Colombian Andes. Journal of South American Earth Sciences. Tectonic evolution of the Colombian Andes 21(4), 423-436.

Núñez, A., 2001. Mapa geológico del departamento del Tolima, memoria explicativa, INGEOMINAS, Colombia.

Orrego, A. and Paris, G., 1999. Cuadrángulo N - 6, Popayán: Geología, Geoquímica y Ocurrencias Minerales (Versión Resumida del original, 1996). Publicaciones Geológicas Especiales de Ingeominas No. 22, p.1- 54.

Paris, G., 1997. Fallas potencialmente sismogénicas que pueden afectar las obras del complejo vial y cruces a desnivel de la 2a con Avenida Bolivar y de la Cejita en la Avenida República del Líbano. Alcaldía de Armenia, Armenia, Colombia.

Paris, G., Machette, M. N., Dart, R. L. and Haller, K. M., 2000. Map and database of Quaternary faults and folds in Colombia and its offshore regions. U. S. Geological Survey. Reston, VA, United States. Pages: 61, 1 sheet. 2000. 46 Chapter 3

Paris, G. and Romero, L. J. A., 1994. Fallas activas en Colombia. Boletin Geologico - Ingeominas 34(2-3), 3-25.

Pennington, W. D., 1981. Subduction of the eastern Panama Basin and seismotectonics of northwestern South America. JGR. Journal of Geophysical Research. B 86(11), 10753-10770.

Pindell, J. and Dewey John, F., 1982. Permo-Triassic reconstruction of western Pangea and the evolution of the Gulf of Mexico/Caribbean region. Tectonics 1(2), 179-211.

Pindell, J. and Erikson, J., 1993. The Mesozoic margin of northern South America. In: Salfity, J. (Ed.), Cretaceous tectonics of the Andes. Vieweg Germany, pp. 1–60.

Pindell, J. and Kennan, L., 2001. Kinematic evolution of the Gulf of Mexico and Caribbean. Program and Abstracts Society of Economic Paleontologists Gulf Coast Section Research Conference 21, 193-220.

Pindell, J., Kennan, L., Maresch, W. V., Stanek, K., Draper, G. and Higgs, R., 2005. Plate-kinematics and crustal dynamics of circum-Caribbean arc-continent interactions: Tectonic controls on basin development in Proto-Caribbean margins. Geological Society of America, Special Paper 394, 7-52.

Pindell, J. and Tabbutt, K. D., 1995. Mesozoic-Cenozoic Andean paleogeography and regional controls on hydrocarbon systems. AAPG Memoir 62, 101-128.

Pindell, J. L. and Barrett Stephen, F., 1990. Geologic evolution of the Caribbean region: a plate tectonic perspective, en Dengo, G., y Case, J.E., eds., The Caribbean Region. The Geology of North America: Boulder, Colorado, The Geological Society of America, p. 405-432.

Pindell, J. L., Cande, S. C., Pitman, W. C., Rowley, D. B., Dewey, J. F., LaBrecque, J. and Haxby, W., 1988. A platekinematic framework for models of Caribbean evolution. Tectonophysics 155, 121- 138.

Ramon, J. C. and Rosero, A., 2006. Multiphase structural evolution of the western margin of the Girardot subbasin, Upper Magdalena Valley, Colombia. Journal of South American Earth Sciences 21, 493–509.

Ramos, V. A. and Aleman, A., 2000. Tectonic evolution of the Northern Andes. Tectonic evolution of South America. G. Cordani Umberto, J. Milani Edison and A. Thomaz Filho. In In-Folo Producao Geological background 47

Editorial, Grafica e Programacao Visual. Rio de Janeiro, Brazil.2000.

Restrepo-Pace, P. A., 1992. Petrotectonic characterization of the Central Andean Terrane, Colombia. Journal of South American Earth Sciences 5(1), 97-116.

Restrepo-Pace, P. A., 1999. Field assessment of true stratigraphical thickness, stretching, shortening structural style and kinematics of the Eastern Cordillera, Colombia. Pages 58.

Restrepo-Pace Pedro, A., 1999. Field assessment of true stratigraphical thickness, stretching, shortening structural style and kinematics of the Eastern Cordillera, Colombia. Pages 58.

Restrepo, J. J. and Toussaint, J. F., 1988. Terranes and continental accretion in the Colombian Andes. Episodes 11(3), 189-193.

Ríos, P. A. and Aránzazu, J. M., 1989. Analysis litofacial del intervalo Oligoceno-Mioceno en el sector noreste de la subcuenca del Valle del Cauca, Departamento del Valle, Colombia, Universidad de Caldas. Tesis de grado: 257pp.

Ross, M. I. and Scotese, C. R., 1988. A hierarchical tectonic model of the Gulf of Mexico and Caribbean region. Tectonophysics 155, 1-4.

Salinas, R. E., González, H., González, L. M., Quintero, W. and Ruiz, S., 2007. Proyecto Modelo geológico y zonas potenciales para exploración de recursos minerales en la zona Cauca Romeral, sector Viterbo (Risaralda)- Pasto (Nariño). Colombia. - Informe de compilacion, analisis e interpretacion de informacion existente. Bogotá, Instituto Colombiano de Geología y Minería, Ingeominas: 119.

Sarmiento-Rojas, L. F., Van Wess, J. D. and Cloetingh, S., 2006. Mesozoic transtensional basin history of the Eastern Cordillera, Colombian Andes: Inferences from tectonic models. Journal of South American Earth Sciences 21, 383–411.

Schwinn, W. L., 1969. Guidebook to the geology of the Cali area, Valle del Cauca, Colombia. In Tenth Annual Field Conference, Colombian Society of Petroleum Geologists and Geophysicists: pp. 277-297.

Stutzer, O., 1934. Contribución a la geología del Foso Cauca-Patía: Compilación de los Estudios Geológicos Oficiales en Colombia, v. 2, p. 69-140. 48 Chapter 3

Suter, F., Neuwerth, R. and Gorin, G. E., in review. Tectonic closure of the Interandean Cauca Depression (Central Colombia): geomorphology and neotectonics of the Serranía de Santa Barbara fold-and-thrust range and surrounding Quaternary deposits. Geomorphology.

Suter, F., Neuwerth, R., Gorin, G. E. and Guzman, C., 2008a. Depositional model of (Plio-)Pleistocene sediments in a tectonically active zone of Central Colombia. Geologica Acta 6(2), 1-19.

Taboada, A., Dimate, C. and Fuenzalida, A., 1998. Sismotectonica de Colombia; deformacion continental activa y subduccion. Fisica de la Tierra 10, 111-147.

Taboada, A., Rivera, L. A., Fuenzalida, A., Cisternas, A., Philip, H., Bijwaard, H., Olaya, J. and Rivera, C., 2000. Geodynamics of the Northern Andes; subductions and intracontinental deformation (Colombia). Tectonics 19(5), 787-813.

Tassinari, C., Pinzon, F. and Buena, V. J., 2008. Age and sources of gold mineralization in the Marmato mining district, NW Colombia: A Miocene-Pliocene epizonal gold deposit. Ore Geology Reviews 33(3-4), 505-518.

Thouret, J. C., Cantagrel, J.-M., Robin, C., Murcia, A., Salinas, R. and Cepeda, H., 1995. Quaternary eruptive history and hazard-zone model at Nevado del Tolima and Cerro Machin Volcanoes, Colombia. Journal of Volcanology and Geothermal Research 66, 397-426.

Thouret, J. C., Ramírez, J. C., Gilbert-Malengreau, B., Vargas, C. A., Naranjo, J. L., Vandemeulebrouck, J., Valla, F. and Funk, M., 2007. Volcano–glacier interactions on composite cones and lahar generation: Nevado del Ruiz, Colombia, case study. Annals of Glaciology 45 115-127.

Toussaint, J. F., 1995a. Evolución geológica de Colombia 2. Triásico Jurásico. Contribucíon al IGCP 322 ‘‘Correlation of Jurassic events in South America’’ International Geological Correlation programme Unesco IUGS. Univ. Nacional de Colombia. Medellín, 94p.

Toussaint, J. F., 1995b. Hipótesis sobre el marco geodinámico de Colombia durante el Mesozóico temprano, Contribution to IGCP 322 Jurassic events in South America. Geol. Colombiana, Bogotá 20, 150-155. Geological background 49

Toussaint, J. F. and Restrepo, J. J., 1978. Edad K-Ar de dos rocas básicas del flanco noroccidental de la Cordillera Central. Publicaciones Especiales de Geología, 15: Facultad de Ciencias, Medellín.

Toussaint, J. F. and Restrepo, J. J., 1989. Acresiones sucesivas en Colombia; un nuevo modelo de evolución geológica. V Congr. Colomb. Geol., Bucaramanga I, 127-146.

Trenkamp, R., Kellogg, J. N., Freymueller, J. T. and Mora, H. P., 2002. Wide plate margin deformation, southern Central America and northwestern South America, CASA GPS observations. Journal of South American Earth Sciences 15(2), 157-171.

Van der Hammen, T., 1958. Estratigrafía de Terciario y Maestrichtiano continentales y tectogenesis de los Andes colombianos. Boletin Geológico VI, 67-128.

Van der Hilst, R. and Mann, P., 1994. Tectonic implications of tomographic images of subducted lithosphere beneath northwestern South America. Geology (Boulder) 22(5), 451-454.

Villamil, T., 1999. Campanian-Miocene tectonostratigraphy, depocenter evolution and basin development of Colombia and western Venezuela. Palaeogeography Palaeoclimatology Palaeoecology 153, 1-4.

Vinasco, C. J., 2001. Autilização da metodologia 40Ar-39Ar para o estudo de reativações tectônicas em zonas de cisalhamento. Paradigma: O Falhamento de Romeral nos Andes Centrais da Colômbia. Unpublished M.Sc. thesis, Instituto de Geociências, Universidade de São Paulo, 85 pp.

White, S. M., Trenkamp, R. and Kellogg, J. N., 2003. Recent crustal deformation and the earthquake cycle along the Ecuador-Colombia subduction zone. Earth and Planetary Science Letters 216(3), 231-242.

CHAPTER IV

Se d i m e n t a r y i n f i l l o f An d e a n i n t r a m o u n t a n e b a s i n s : c a s e h i s t o r y o f Pl i o -Pl e i s t o c e n e d e p o s i t s i n t h e No r t h Ca u c a Va l l e y (Co l o m b i a )

Ralph Neuwerth, Fiore Suter, Georges Gorin, Carlos Guzman

Unpublished 52 Chapter 4

4.1. Introduction

The northern termination of the Andes has to be considered separately from the rest of the Andean chain because of its complexity. It results from the convergence between three plates, i.e. the Nazca, Caribbean, and South American plates (Pennington, 1981; Restrepo and Toussaint, 1988; Freymueller et al., 1993; Kellogg et al., 1995; Taboada et al., 2000; Trenkamp et al., 2002; Cediel et al., 2003; Cortes and Angelier, 2005). In the convergence zone where most of the stress is accommodated, three intensely-deformed blocks can be distinguished: the Choco-Panamá Block (CPB), the North Andes Block and the Maracaibo Block (Fig. 4.1), the former being part of the Caribbean Plate and the latter two of the South American Plate.

The Colombian Andes are composed of three SSW-NNE trending cordilleras. They are crosscut by the Romeral fault system (Fig. 4.1) which runs between the Western and Central Cordilleras from Guayaquil up to the Caribbean Sea (Paris and Romero, 1994; Paris et al., 2000). This fault system outlines the suture between accreted terranes and continental basement (McCourt et al., 1984a; McCourt and Feininger, 1984; Aspden and McCourt, 1986; Aspden et al., 1987; Cediel et al., 2003; Nivia et al., 2006). Following the CPB collision and its subsequent indentation during the Mio-Pliocene, this originally right-lateral transcurrent fault system became left-lateral north of 5°N (Trenkamp et al., 2002).

The intramountane North Cauca Valley Basin studied in this paper is located immediately west of the Romeral fault system in the zone where the latter changes its wrench kinematics (Fig. 4.1). The studied area is located in the northern part of the Cenozoic, acretion-related, Cauca-Patía Valley Basin, which streches over 12’800 km2 (Barrero et al., 2007). From a structural point of view, the North Cauca Valley Basin is particular because an active pop- up structure forms a topographic barrier in its central part. The latter is called the Serranía de Santa Barbara (SSB; (Suter et al., in review)) and subdivides the North Cauca Valley Basin in two sub- basins: the Cauca Depression to the west and the Quindío-Risaralda Basin to the east (Fig. 4.2). The Oligo-Miocene rocks forming the SSB are folded. They are unconformably overlain on both sides of the SSB by Pliocene to Recent rocks which are subhorizontal and show both extensional dislocations (Cardona and Ortiz, 1994; Pardo et al., 1994; Suter et al., 2008a; Suter et al., 2008b) and soft-sediment deformations (Neuwerth et al., 2006). These Pliocene to recent rocks are the object of this paper.

The existence of intramountane basins showing undeformed to slightly deformed younger, Late Tertiary-recent formations unconformably overlying older, folded, Tertiary formations is not restricted to the studied area. They also occur in other parts of the Colombian Andes, such as the Amagá, Patía and the Lower and Upper Magdalena basins (Barrero et al., 2007). Therefore, the detailed analysis of such little-deformed, younger sediments infilling intramountane basins is susceptible to provide important new data for the understanding of the Late Neogene to Pleistocene evolution of the Colombian Andes. This is particularly the case for the North Cauca Valley Basin where extensive field research over the past eight years has provided important new data on these previously poorly studied Late Tertiary sediments. Lithostratigraphy 53 54 Chapter 4

Fig. 4.1: (A) Geodynamics of NW South America and location of study area. Velocities and direction of motion for the different plates and blocks with respect to South America (after (Suter et al., 2008b). Abbreviations: B: Bogotá; C: Cali; CC: Central Cordillera; EC: Eastern Cordillera; WC: Western Cordillera; EFFS: Eastern Frontal Fault System; IBF: Ibagué Fault; GF: Garrapatas Fault; RFS: Romeral Fault System. (B) Digital elevation model (DEM, (USGS, 2005)) of western central Colombia showing the course of the Cauca River. The study area is located upstream of La Virginia town, at the northern termination of the Cauca Valley Basin (after (Suter et al., 2008b).

Prior to the data presented in this paper, studies in this area have demonstrated the complex structure of the Quindío-Risaralda volcaniclastic fan (Guarin et al., 2006), the high seismic activity in the North Cauca Valley Basin during the Plio-Pleistocene to recent time (Neuwerth et al., 2006). A depositional model for the Plio-Pleistocene Zarzal Formation and the Quindío-Risaralda Fan (Suter et al., 2008a) and regional structural interpretations (Suter et al., 2008b) have been proposed. However, further field studies have (a) shown that many important facies in the Plio-Pleistocene deposits hadnot been reported and (b) pointed out the stratigraphical relationships between the different lithological units.

Consequently, the aims of this paper are: 1) to describe the Plio-Pleistocene lithological units cropping out in the North Cauca Valley Basin and illustrate the stratigraphical relationships existing between these units; 2) to propose a new lithostratigraphical subdivision of this highly seismic and so far poorly studied basin leading to a better understanding of the emplacement of these units in relation with the development of this intramountane basin. This type of study may serve as a case history that can be applied to other intramountane basins in the Colombian Andes.

4.2. Geological setting

4.2.1. Ol ig o -Mi o c e n e , f o l d e d d e p o s i t s The SSB is a small fold-and-thrust belt made of folded Oligo-Miocene rocks upon which were deposited the sedimentary units studied in this research, i.e., the Plio-Pleistocene Zarzal Formation and volcaniclastic fans (Fig 4.2). The SSB is constituted by the continental deposits of the Cartago and La Paila Formations.

The Cartago Fm (Schwinn, 1969; Ríos and Aránzazu, 1989) is also referred to as the Cinta de Piedra Fm (Hubach and Alvarado, 1934; McCourt et al., 1984b; Keith et al., 1988) or Cinta de Piedra Member of the Cauca Superior Fm (Van der Hammen, 1958). It has an age ranging from Lower Oligocene, according to palynological datings in its intermediate member (Ríos and Aránzazu, 1989), to Middle Miocene (Schwinn, 1969). McCourt (1984) considers it as Oligocene based on stratigraphical relationships. Folding in the Cartago Fm started in Late Oligocene-Early Miocene times (Keith et al., 1988; Ríos and Aránzazu, 1989) and the syn-kinematic La Paila Fm was subsequently uncomformably deposited on this incipient fold belt. The La Paila Fm (Van der Hammen, 1958; Alvárez, 1983; McCourt, 1984; Keith et al., 1988; Ríos and Aránzazu, 1989; Gonzalez and Nuñez, 1991; Nivia et al., 1992) is also referred to as Buga Fm (Schwinn, 1969). Its age is considered as Lower Miocene by McCourt (1984) based on structural and sedimentological relationships with porphyric intrusions, as Middle Miocene by Schwinn (1969) and Nivia et al. (1992), as Middle to Upper Miocene based on Ar40/Ar39 dating (12,7 Ma; Suter, unpublished), or as Miocene by Van der Lithostratigraphy 55

Hammen (1958) based on palynology. Its lithology testifies to the onset of volcanic activity in the Cerro Bravo-Machin volcanic system to the east (Fig. 4.1).

4.2.2. Pl i o -Pl e i s t o c e n e , s u b h o riz o n t a l d e p o s i t s

The first description of subhorizontal to slightly dipping deposits of the Zarzal Formation cropping out in isolated hills of the Cauca Valley was made by Boussingault (1903), who considered these sediments as lake deposits. The name of Zarzal Formation was given by Keiser (1955, in Nelson, 1962) and Van der Hammen (1955, in Van der Hammen, (1958)), who observed clayey and sandy sediments unconformably overlying the La Paila Fm. A probable Pliocene age was attributed to them (Van der Hammen, 1958; De Porta, 1974; McCourt, 1984).

McCourt (1984), McCourt et al. (1984b), Keith et al. (1988) and Nivia et al. (1995) subdivided the Zarzal Formation in three units (Table 1). Cardona and Ortiz (1994) interpreted different depositional 56 Chapter 4 environments in the Zarzal Formation, i.e., braided-stream, floodplain and lacustrine. They recognized the interfingering between Zarzal sediments and the volcaniclastic mass flows in the Pereira-Cartago- La Virginia area (Fig. 4.2 and 4.3).

The eastern foothills and northern end of the SSB are unconformably overlain by a succession of volcaniclastic fans (Fig. 4.2 and 4.3). Numerous studies have been carried out on these deposits but they never considered the entire volcaniclastic fan complex. The latter has been successively named Armenia Formation (McCourt, 1984), Glacis del Quindío (Gonzalez and Nuñez, 1991), Pereira Formation (Cardona and Ortiz, 1994), Quindío Fan (Espinosa, 2000) and finally Quindío-Risaralda Fan (Guarin et al., 2006). This complex succession of volcaniclastic deposits has a semi-conical shape. It covers an area of over 1200 km2 (Cardona and Ortiz, 1994; Guarin, 2008). These mass flow deposits are poorly dated with an age comprised between 1.3 million and 50’000 years (Guarin, 2004). Forty-five units can be recognized according to their mineralogical, sedimentological, stratigraphical and geomorphological characteristics (Guarin, 2008). The most representative processes consist of debris avalanches, debris-flows, hyperconcentrated flows, fluvial and ash fall deposits. The source of the sediments originates from the Cerro Bravo-Machín volcanic system (Alfaro and Aguirre, 2003). The latter is also referred to as Ruiz-Tolima volcanic system (Guarin et al., 2006) and located in the Central Cordillera (Fig. 4.1). Some mass-flow deposits are interfingering with the fluvio-lacustrine Zarzal Formation east of the SSB (Suter et al., 2008a) and near Cartago (Fig. 4.3) (Cardona and Ortiz, 1994; Neuwerth et al., 2006; Suter et al., 2008a).

Recently, a sedimentological model has been proposed to explain the relationship between the Zarzal Formation and the Quindío-Risaralda Fan deposits, showing the westward progression of the volcaniclastic fans and subsequent sedimentological damming of the Cauca Depression and Quindío- Risaralda Basin, allowing the accumulation of lacustrine, diatomitic deposits (Suter et al., 2008a).

4.3. Tectonics

The Plio-Pleistocene deposits studied here are marked by an absence of important deformations (Keith et al., 1988; Nivia et al., 1992). According to Ríos and Aránzazu (1989), their horizontal to

Fig. 4.3: Simplified tentative cross-section (after Suter et al. (2008b)) across the Cauca depression and the Quindío-Risaralda Basin (see figure 4.2 for location). The two zones highlighted by a frame (Cauca depression and La Vieja River) are detailed below in figure 4.6. Lithostratigraphy 57 subhorizontal position testifies to a relative tectonic stability from Pliocene to Recent times. However, McCourt (1984) and Van der Hammen (1958) reported minor dislocations probably related to recent movements along buried faults in the pre-Tertiary basement. Cardona and Ortiz (1994) and Pardo et al. (1994) also observed recent faulting. These authors mention that anticlines and synclines with amplitudes of 200 m deform the Zarzal Fm beds to the north of Obando (Fig. 4.4). Recent research has shown that these sediments are affected by superficial normal faulting associated with seismites (Neuwerth et al., 2006; Suter et al., 2008a), which is interpreted as the signature of subhorizontal landslides during earthquakes (Suter et al., 2008b).

The North Cauca Valley Basin is affected by the N-NNE trending Romeral faults. In this area, their kinematics is compressive (Paris et al., 2000; Suter et al., 2008b). However, other younger strike-slip faults crosscut this active suture inherited from the Cretaceous. The WSW–ENE trending, right- lateral, en-échelon Garrapatas-Ibagué shear zone, which bounds the study area to the south (Fig. 4.1), is interpreted as very active (Suter et al., 2008b) and the maximum shortening across Colombia takes place north of this line (Audemard, 2003). The ENE striking, right-lateral Santa Rosa Fault bounds the basin to the north and may also be of considerable importance (Fig. 4.1).

Based on inversion calculations of striated fault plane data sets, Suter et al. (2008b) obtained 29 palaeostress orientations in the folded beds of the SSB with the σ1 stress axes suggesting a WNW– ESE trend, in agreement with the neotectonic data of Paris et al. (2000) and the seismologic data of Gallego et al. (2005). This stress pattern exists at least since Upper Miocene – Lower Pliocene times (Cortes and Angelier, 2005) and is related to the eastward-southeastward movement and subsequent collision of the CPB against NW South America. This major geodynamic event is responsible for the last and strongest phase of uplift in the Colombian Andes, which affected the three cordilleras and generated a horizontal shortening exceeding 150 km (Audemard, 2002; Trenkamp et al., 2002). Various hypotheses have been proposed to explain the occurrence of the Cauca Valley intramountane basin at an altitude of 900m. Some authors have interpreted it as a graben (Acosta, 1978; Padilla, 1982; McCourt, 1984; Bermúdez et al., 1985; Droux and Delaloye, 1996; MacDonald et al., 1996) or a pull-apart basin (Kellogg et al., 1995 in Alfonso et al. (1994)), whereas other studies have demonstrated the compressive nature of the faults and folds within Cretaceous to Miocene structures (Keith et al., 1988; Ríos and Aránzazu, 1989; Alfonso et al., 1994). Neotectonic compressive activity in some faults bounding this floodplain has also been reported (López et al., 2005; López and Moreno, 2005). The Cauca Valley Basin is a 200km long alluvial plain extending from Cali up to La Virginia (Fig. 4.1) at an altitude of 900m. Recent research in the North Cauca Valley shows that most probably the existence of this basin results from a closing at its northern termination. Two hypotheses can be invoked to explain this northern closing, one based on tectonics, the other on sedimentology: 1) north of 4.5°N, the Cauca River Valley might have been closed as a result of the ongoing indentation of the CPB (Suter et al., 2008b; Suter et al., in review); 2) the rapid westward progression of the volcaniclastic Quindío-Risaralda-Cartago Fans up to the eastern foothills of the Western Cordillera might have dammed the valley towards the north (Suter et al., 2008a). 58 Chapter 4

4.4. Methods

Because it outcrops reasonably well and presents a moderate structural complexity, the Zarzal Fm has been first studied using aerial photographs, in order to draw a preliminary surface map and locate outcrops. Subsequently, new sedimentological data have been acquired through the sedimentary logging of twenty-two field sections (Fig. 4.4). The study of rock thin section and palynological slides of selected samples have complemented field data. Palynological preparations were carried out using a standard preparation method for sedimentary rocks (e.g., (Steffen and Gorin, 1993).

Laser fusion 40Ar/39Ar dating of biotites from ash layers interbedded within diatomites has been carried out in the mass-spectrometry laboratory at the University of Geneva to constrain the age of deposition of lacustrine deposits within the basin.

4.5. Results and interpretation: New lithostratigraphical subdivision of Late Tertiary- Recent sediments

4.5.1. Introduction

The existing subdivisions in the Zarzal Formation proposed by McCourt et al. (1984b, in Keith et al. (1988)) and Nivia et al. (1995) are quite different and show respectively coarsening-up and fining- upward sequences (Table 1). Moreover, they only take into account vertical variations, whereas the research presented here has revealed that this formation presents important lateral variations, which have to be taken into account in order to subdivide it. Furthermore, the interfingering of volcaniclastic mass-flow deposits from the Quindío-Risaralda Fan with the Zarzal Formation deposits (Cardona and Ortiz, 1994; Neuwerth et al., 2006; Suter et al., 2008a) highlights the stratigraphical relationship between these units. Consequently, the nomenclature of the different lithological units present in this area has to be redefined. Because seismic, well and fossil data are missing, the only stratigraphical type which can be applied in such a basin is the lithostratigraphical unit.

The revised lithostratigraphical definitions used in this paper are in agreement with the International Stratigraphical Chart (ICS). The following subdivision is proposed below: a new Santa Barbara Group including two heteropical units, the Zarzal Formation and the Quindío-Risaralda Fan (named here the Quindío-Risaralda Formation) (Table 2).

4.5.2. The Santa Barbara Group (new)

The Santa Barbara Group encompasses all the Plio-Pleistocene sediments deposited in the North Cauca Valley and Quindío-Risaralda Basins, between the Western and Central Cordilleras. Its name is derived from the Serranía de Santa Barbara, which geographically separates the zones where these deposits are encountered (Fig. 4.2).

Lithological and geomorphological criteria were used to subdivide the Santa Barbara Group into the Zarzal and Quindío-Risaralda Formations. Lithostratigraphy 59

Fig. 4.4: 30-meter resolution DEM (USGS, 2005) with location of studied field sections. Three zones are highlighted: 1: the Cartago Fan (CF) (sections 1 to 8, Fig. 4.16); 2: the eastern foothills of the Western Cordillera (WC) (sections 9 to 17, Fig. 4.10); 3: the western foothills of the Serranía of Santa Barbara (SSB) (sections 18 to 22, Fig. 4.8). Abbreviations: CC: Central Cordillera; QRF: Quindío-Risaralda Fan. 60 Chapter 4

4.5.3. The Zarzal Formation (FmZ) (redefined)

Derivation of name. Named for the village of Zarzal located south of the studied area (Fig. 4.4).

Definition. In the field, the most easily recognizable and most frequently encountered lithofacies in the Zarzal Formation are diatomites, mud, tuffaceous black sands and gravels. These deposits have been interpreted as lacustrine, floodplain and braided-stream deposits (McCourt, 1984; McCourt et al., 1984b; Keith et al., 1988; Nivia, 2001). Indeed, this definition covers the most frequently encountered sediments. However, recent field observations show that many important facies are lacking in this description. Consequently, three lithostratigraphical members have been distinguished, based on geographical, mineralogical and sedimentological criteria. Frequent intervals of soft-sediment deformations (interpreted as seismites) have been encountered in these sediments (Neuwerth et al., 2006; see below).

4.5.3.1. The Obando Member (OM) (new)

Derivation of name. Named for the village of Obando located some 25 km north of Zarzal (Fig. 4.4). The Obando Member (OM) crops out in the central part of the northern Cauca Valley basin (Fig. 4.2) and in the eastern foothills of the SSB (Suter et al., 2008a).

Type section. The type section crops out near Obando (section 18, Fig. 8 in Suter et al. (2008a). This outcrop was well described by the latter authors but not formally named. This member reaches a maximum thickness of some 33 m (section 5, Fig. 6 in Suter et al. (2008a).

Fig. 4.6: Simplified tentative cross-sections from the Western Cordillera (WC) to the Quindío-Risaralda basin. This interpretation illustrates the interfingering of the Plio-Pleistocene units defined in this study (see Figure 4.3 for location and legend). Abbreviations: AM: Ansermanuevo Member; HM: Holguín Member; OM: Obando Member; QRFm: Quindío- Risaralda Formation; SSB: Serranía of Santa Barbara.

Lithofacies (I) Fine to very coarse-grained, black to grey sands. Most frequently, these beds present stacked, trough cross-bedded bodies, often containing intraclasts of diatomite or mud soft pebbles (Fig. 4.5A). These beds are less than 50 cm thick and display erosive bases. Clasts are of volcanic composition.

(II) Very fine sands, silts and mud (Fig. 4.5B). These beds present frequent fine laminations, are continuous laterally over many tens of metres and show only minor changes in thickness. Silts often contain plant fragments, which consist mainly of well preserved fossil leaves (dicotyledonous and monocotyledonous angiosperms). Coal streaks and plant roots are also encountered. Mud has been sampled for palynological studies. Sandy and silty clasts are of volcanic composition.

(III) Diatomaceous mud and diatomite deposits. The latter are massive and sometimes finely laminated (Fig. 4.5C). Fine, non-erosive and structureless volcanic ash layers are interbedded with diatomite beds. These layers contain well oriented, angular and sorted grains of biotite, hornblende, pyroxene, feldspath, quartz and volcanic glass.

Age and relationship with other lithostratigraphic units. The lower part of the member unconformably overlies the Lower Tertiary (sections 1 and 3, Fig. 4.6 and section 15, Fig. 8 in Suter et al. (2008a). This member is interfingering with the Quindío-Risaralda Formation north of Cartago and on the eastern flank of the SSB (Fig. 4.3). It is also interfingering with the Ansermanuevo and Holguín Members, respectively on the western and eastern sides of the northern Cauca Valley Basin (Fig. 4.6).

The deposits of the Obando Member have been dated using both geochronological (Fig. 4.7) and biostratigraphical methods. Biotites in five ash layers interbedded with diatomites (sections 18, 19 and 20, Figs. 4.5C, 4.7A-E and 4.8) have been analyzed with the 40Ar/39Ar method in three sedimentary sections (sections 18-20, Fig. 4.8). Ages range between 1,06 ± 0,17 and 2,79 ± 0,5 Ma (Fig. 4.7). Note that the sample FmZ377 (section 18, Fig. 4.8 and Fig. 4.7D) presents a bad weighted plateau and age was calculated through total fusion.

Palynological analyses carried out in mud samples from different sections showed the presence of Alnus pollen (Figs. 4.10 and 16 and Figs. 6, 8 and 10 in Suter et al. (2008a). The first appearance of this tree in the high plain of Bogotá (Colombia) dates back to 0.8 Ma ago, following its southward propagation through the Isthmus of Panama from North America (Hooghiemstra and Sarmiento, 1991; Andriessen et al., 1993; Hooghiemstra and Cleef, 1995; Van der Hammen and Hooghiemstra, 1997; Torres et al., 2005). 62 Chapter 4

Fig. 4.7: (A) 40Ar/39Ar plateau age from the ash layer of section 18-20 in figure 4.8. Lithostratigraphy 63

Fig. 4.9: Palaeocurrent directions measured in trough cross-bedded black sands, sigmoid stratifications (rose diagramme 8) and imbricated pebbles (rose diagramme 4) of Zarzal Formation. Symbols and abbreviations: n = number of palaeocurrent measurements per site; Ø = mean vector; white arrows depict the mean vector on the DEM. A: Ansermanuevo; C: Cartago; CC: Central Cordillera; CF: Cartago Fan; O: Obando; QRF: Quindío-Risaralda Fan; SSB: Serranía of Santa Barbara; Vict: La Victoria; Vir: La Virginia; WC: Western Cordillera; Z: Zarzal. The base map is the same DEM as that of figure 4.4 where sections are located.

Consequently, the age of the Obando Member deposits can be constrained to the Plio-Pleistocene.

Environment of deposition. The three different lithofacies described above are interpreted as: (I) The black to grey sands correspond to the St lithofacies proposed by Miall (2006), interpreted as sinuous-crested and linguoid (3-D) dunes. Trough cross-bedding is a characteristic feature of meandering river deposits, representing the preserved foreset beds of dunes migrating downstream.

(II) The very fine sands, silts and mud correspond to the Fl lithofacies proposed by Miall (2006), interpreted as overbank, abandoned channel, or waning flood deposits. Lithostratigraphy 65 Field sections in the eastern foothills of the Cordillera Western (WC) (see figure 4.4 for location of sections). The first occurrence of alluvial Fig. 4.10: conglomerates rich in black cherts, white quartz and volcanic clasts at each location 4.8. is to a time line. Legend is the same as that in figure lithological composition. This line does not correspond correlated with a black line in order to highlight the change in 66 Chapter 4

(III) The diatomaceous mud and diatomites are interpreted as lacustrine deposits.

The St and Fl lithofacies are made of volcanic clasts which are derived from the Central Cordillera like the Quindío-Risaralda Formation (see below).

Palaeocurrents measured in the central part of the basin and surroundings of Ansermanuevo do not show any general trend (Fig. 4.9, rose diagram 1 and 2).

4.5.3.2. The Ansermanuevo Member (AM) (new)

Derivation of name. Named for the village of Ansermanuevo located 8 km northwest of Cartago (Fig. 4.4). The Ansermanuevo Member is restricted to the northwestern part of the study area.

Type section. The type section crops out between Ansermanuevo and Cartago (section 17, Fig. 4.10) and presents a minimum thickness of eight meters.

Lithofacies (I) Beds of clast-supported gravels (Fig 4.11). They show erosive bases and crude horizontal stratifications and contain angular soft pebbles (Fig. 4.11). Most deposits present imbricated clasts with an abundant sandy matrix. Beds are few decimeters up to 3 m thick and well oxidized, and present important lateral variations. The gravely to sandy matrix and the clasts have the same composition, i.e., a dominance of black cherts and white quartz (85%), volcanic material (basalts) and accessorily sedimentary clasts.

(II) Very fine- to medium-grained sands and silts with horizontal laminations that consist of millimeter- to centimeter-thick layers.

Ansermanuevo Member: clast-supported conglomerate rich in black chert, white quartz and volcanic elements interfingering with typical black sand deposits of the Obando Member (section 17, Fig. 4.10). Hammer for scale. Lithostratigraphy 67

Age and relationship with other lithostratigraphic units. The Ansermanuevo Member is interfingering with the Obando Member fluvio-lacustrine deposits and the Quindío-Risaralda Formation (sections 12, 14 and 15, Fig. 4.10). It is present only in the western part of the basin (Figs. 4.4 and 4.10), and more specifically in the upper part of the outcrops. Although no datings are available, the Ansermanuevo Member is assumed to be of Plio-Pleistocene age because of its interfingering with the dated Obando Member (see above).

Environment of deposition. The clast composition indicates that conglomerates originate from the Western Cordillera. This is confirmed by imbricated flat pebbles showing palaeocurrent directions towards the E-NE (section 16, Fig. 4.10 and rose digram 4, Fig. 4.9). The lithofacies corresponds to the Gh and Sh lithofacies proposed by Miall (2006) and respectively interpreted as longitudinal bedform and plane-bed flow (critical flow). Because of the abundance of coarse gravels alternating with sandy and silty layers and of important lateral variations, this facies is interpreted as alluvial-fan deposits.

4.5.3.3. The Holguín Member (HM) (new)

Derivation of name. The name is derived from Holguín, a village located north of Zarzal. The Holguín Member (HM) is geographically located between Zarzal and Obando. These deposits are present northwards up to Obando, some twenty-five kilometers north of Zarzal (Suter et al., 2008a).

Type section. The type section crops out around Zarzal (section 22, Fig. 4.8) and is 20 m thick.

Lithofacies. Deposits are characterized by red sands, as well as matrix- and clast- supported conglomerates. A well exposed outcrop is present south of Zarzal (section 23, Fig 4.8), in the western foothills of the SSB. It consists of a stratigraphical succession showing, from base to top, fine- grained (basal unit I), sandy sigmoidal (intermediate unit II) and horizontal coarse- grained deposits (upper unit III).

The basal unit (I) shows very fine sandy to clayey beds (Fig. 4.12A). They are composed of detrital material derived from the SSB (white quartz, black cherts and volcanic elements). Plant fragments are present (Fig. 4.12B). These horizontal beds are pinching out towards the south.

Fig. 4.13: Holguín Member: erosive contact between foreset and topset units (section 23, Fig. 4.8).

In longitudinal cross-section, this unit II is characterized by steeply dipping, coarse- to fine-grained, sigmoidal, sandy beds, each one up to 1.5 m thick and 12 m long (Fig. 4.13). These beds are separated by thin silty interbeds (Fig. 4.14A). The bed composition is similar to that of the basal unit. They contain clayey soft pebbles (Fig. 4.14B) and plant fragments. Because of the sigmoid steep angle (from 13 to 32° dip), this unit shows numerous slope instability features, such as slumps and slump scars. Furthermore, it is affected by numerous synsedimentary normal faults and soft-sediment deformations (Neuwerth et al., 2006; Suter et al., 2008a).

In transverse cross-sections, this unit II appears as parallel flat-lying sandy beds (Fig. 4.14C) that grade upwards into fine sandstones with small-scale trough cross-beddings.

The upper unit III consists of sandy and gravely deposits with an irregular and erosive base (Fig. 4.15). This unit has a minimum thickness of 6.4 m and the same lithological composition as the underlying units. The sandy and gravely lower part consists of stacked channelized bodies of well to moderately sorted clast-supported conglomerates and sands, with abundant trough cross-bedded stratifications up to 3 m thick (Fig. 4.15). Angular soft pebbles are present at the base of the upper gravel interval. The upper part of the unit is composed of disorganized, matrix-supported conglomerates, with angular and irregular-shaped chert clasts up to 10 cm in size.

Age and relationship with other lithostratigraphic units. The Holguín Member is interfingering with the dated Obando Member on the western flank of the SSB (Fig. 4.6) and is therefore considered to be of Plio-Pleistocene age.

Environment of deposition. The well exposed outcrop south of Zarzal is interpreted as a fan delta. The three units described above are considered as bottomsets (lacustrine) (I), foresets (subaqueous) (II) and topsets (subaerial) (III). Lithostratigraphy 69

The horizontal beds (I) are pinching out towards the south (i.e. in the propagation direction of the delta), confirming that they correspond to bottomsets and represent delta front deposits (Fig. 4.12A). The well- developed erosive contact with the overlying foreset succession (II), together with the proximal position of these bottomsets with respect to the foresets (Fig. 4.12A) imply a retrogradation in the typical deltaic succession (Massari and Colella, 1988). This demonstrates a regression of the lake level and/or an uplift.

The relatively thin (up to 5 meters thick) foreset unit (II) indicates a deposition in a shallow lake. Silty interbeds result from decantation of the suspended load (Fig. 4.14A) (Flores R., 1990). Palaeocurrent Fig. 4.14: Holguín Member: foreset facies (section 23, Fig. 4.8): analyses indicate a southward (mean (A) dip section of steeply dipping, sandy foresets separated by siltstone intercalations resulting from deposition of suspended vector = 192°) deltaic progradation load; (B) sandy foresets containing clayey soft pebbles; (C) (Fig. 4.9). This is surprising because strike-parallel section of foresets, which consist of parallel flat- one would expect an overall westward lying sandstone beds. current direction because of the N-S trending SSB. This could be explained by a lack of outcrops, or by the fact that sediments might have been confined between topographical highs aligned with the SSB.

Transverse foresets (II) consist of parallel flat-lying sandy beds, which grade upwards into fine- grained sandstones (Fig. 4.14C) with small-scale trough cross-beddings. This seems to indicate that the deltaic progradation was controlled by bedload deposition associated to turbidity currents along the delta face (Flores R., 1990).

The topset unit (III) consists of both alluvial (sands and conglomerates) and mass flow deposits with an irregular and erosive base (Fig. 4.15). Angular soft pebbles present at the base of the upper gravel interval testify to the existence of an alluvial plain between the mountain front and the lake. The upper part of the topset unit is composed of disorganized, matrix-supported conglomerates, with angular and irregular-shaped chert clasts reaching 10 cm in size. These characteristics permit their interpretation as mass-flow deposits. Consequently, the topset unit can be considered as having been deposited mainly by alluvial fans with ephemeral gravity flow episodes, referred to as type-A feeder system of a Gilbert-type fan delta (Postma, 1990). 70 Chapter 4

4.5.4. The Quindío-Risaralda Formation (QRFm) (new)

Derivation of name. The name is derived from the two departments where it crops out.

Fig. 4.15: Holguín Member: topset unit consisting of stacked, trough cross-bedded alluvial deposits (well to moderately sorted sands and gravel) and mass-flow deposits.

Type section. This formation is made of numerous units spreading over the Quindío-Risaralda Basin. Each of these units has a limited spatial distribution (Guarin et al., 2006; Guarin, 2008) and it is not possible to choose a representative type section. An example of field section is shown in figure 4.18. Consequently, all lithofacies described below are frequently not occurring in the same section.

Lithofacies

(IA) Clast-supported gravels with crude horizontal stratification (Figs. 4.16 and 4.17A). (IB) Tuffaceous black sands showing well-preserved trough cross-bedding and soft pebbles.

(II) Fine- to coarse-grained sands. These deposits are generally poorly stratified and without sedimentary structures, but sometimes millimetric to centimetric stacked beds with horizontal stratification and inverse grading are encountered (Fig. 4.16 and 4.17B).

(III) Matrix-supported gravels (Figs. 4.16, 4.17C and 4.18). These deposits are unstratified, poorly sorted with angular millimetric to metric clasts. These gravels present inverse grading at the base Lithostratigraphy 71 72 Chapter 4

(IV) Clast-supported, blocky deposits (Fig. 4.18). Beds are massive and poorly sorted and formed mainly by lithic material, showing sometimes matrix between clasts. They are centimeters to hundreds of meters thick, presenting sometimes jigsaw cracks indicative of differential stress (Guarin, 2008).

(V) Ash layers

Age and relationship with other lithostratigraphic units. The lower part of this formation unconformably overlies the Cretaceous Quebradagrande and Caramarca Formations in the eastern part (Figs. 4.2 and 4.18). Moreover, it is interfingering with the Obando and Ansermanuevo Members of the Zarzal Formation (Figs. 4.6 and 4.10) and presents a maximum thickness of 350 m (Guarin, 2008). Geomorphologically, it can be divided into three distinct composite fans, i.e. the Quindío-

Fig. 4.17: Quindío-Risaralda Formation. A: stratigraphical contact between (1) clast-supported conglomerates showing horizontal stratification and (2) coarse-grained sands with inverse grading; B: photograph of (2) interpreted as hyperconcentrated flow deposits; C: matrix-supported, poorly sorted conglomerates (1) interpreted as debris-flow deposits with dm-size, angular, basaltic clasts (section 6, Fig. 4.16). Lithostratigraphy 73

Risaralda, North and East Cartago fans (Suter et al., 2008a; Guarin, 2008).

A fission track analysis on zircons from the matrix of the most proximal, oldest debris-flow deposits (Guarin, 2008) in the overall fan gave an age of 1.25 ± 0.13 Ma. Other datings were carried out by

Fig. 4.18: Example of field section in the Quindío-Risaralda Formation (after Guarin (2008)).

Guarin (2008) using the 14C method on five wood samples in one of the stratigraphicalally youngest debris-flow deposits. They all gave ages older than 44.000 years. This means that the age of the QRFm units ranges from 1.3Ma to ca. 50’000 years (Guarin, 2008).

Environment of deposition. The depositional processes for each facies are interpreted as follows:

(IA) Gh lithofacies proposed by Miall (2006), interpreted as longitudinal bedform (Fig. 4.17A). (IB) St lithofacies proposed by Miall (2006) interpreted as sinuous-crested and linguoid (3-D) dunes. 74 Chapter 4

(II) Transition phase between debris-flow and normal fluvial flows, interpreted as hyperconcentrated flow (Fig. 4.17B) by Smith and Lowe (1991).

(III) Gmm lithofacies proposed by Miall (2006), interpreted as plastic debris-flow (Fig. 4.17C). (IV) Gravitational collapse of part of a volcano, interpreted as debris avalanche by Smith and Lowe (1991)

(VI) Volcanic ashes were produced by volcanic eruptions in the Cerro Bravo-Machin volcanic system (Fig. 4.1).

The lithofacies III and IV are predominant. The interbedding with gravel bars (GB) (IA) and sandy bedform (SB) (IB) is typical of the sediment-Gravity-Flow Deposits (Element SG), defined by Miall (2006).

These deposits form a large alluvial fan derived from the Cerro Bravo-Machin volcanic system (Fig. 4.3), where forty-eight sedimentary units have been geomorphologically, sedimentarily and mineralogically distinguished (Guarin, 2008).

4.6. Discussion

This sedimentological study of Plio-Pleistocene deposits in the interandean North Cauca Basin has permitted the establishment of a detailed lithostratigraphy from surface outcrops. Unfortunately, it is not possible to precisely determine the geometry and thickness of the defined lithostratigraphical units in the subsurface, because seismic data are not available.

These lithostratigraphical units infilling the North Cauca Basin testify to the uplift of the Western Cordillera and the Serranía de Santa Barbara, as well as to the volcanic activity in the Central Cordillera during Plio-Pleistocene times. Indeed, vast amounts of coarse deposits are encountered on the margins of the North Cauca Depression, i.e. on the eastern flank of the Western Cordillera and the western flank of the Serranía de Santa Barbara, corresponding respectively to the Ansermanuevo and Holguín Members. These coarse sediments are clearly related to the uplift of these mountain belts. Moreover, there is ample sedimentary evidence of a strong synsedimentary seismic activity (Neuwerth et al., 2006). Furthermore, the large volumes of debris-flows forming the Quindío- Risaralda Fan, as well as the ash-fall deposits in the Obando Member attest to the eruptive activity in the Central Cordillera during the deposition of these formations.

Therefore, this detailed sedimentary and lithostratigraphical investigation of unfolded to slightly folded sediments which have infilled this intramountane basin during Plio-Pleistocene times has brought information that may usefully complement structural investigations in the surrounding reliefs (e.g., Suter et al., 2008b). Such a case study may encourage similar studies of undeformed or only slightly deformed uppermost Tertiary sediments unconformably overlying older, Tertiary, folded rocks in other intramountane depressions of the Colombian Andes. This is particularly the case in the Magdalena (Mesa and Corpa Formations), Amagá (Irra - Tres Puertas Formation), and Patía (Popayán Formation) basins (see figure 4.1B for location), where the uppermost Tertiary infill is poorly studied.

The Irra - Tres Puertas continental Formation of Late Tertiary age crops out north of Chinchina Lithostratigraphy 75

Table 4.1: Published subdivisions for the Zarzal Formation.

(Fig. 4.1), where it unconformably overlies the Miocene Combia Formation in the Amagá basin. It comprises three members, which have been lithologically and stratigraphically distinguished, but not formally named. The following members have been distinguished from base to top: a conglomerate rich in green clasts, another one rich in white and black cherts and a volcano-sedimentary member (Estrada et al., 2001). This formation is thought to be the infill of a transpressional basin (Estrada et al., 2001) and of Late Miocene to Early Pliocene (Dueñas and Castro, 1981) or Pliocene age (Van Houten, 1976). In the Magdalena Basin (Fig. 4.1), the Plio-Pleistocene formations that might be correlated with the Santa Barbara Group are the Corpa and Mesa Formations. The Corpa Formation has so far been poorly studied and its continental part consists of alluvial deposits and fluvial terraces. It unconformably overlies the Miocene Pajuil Formation (Reyes-Santos et al., 2000; Gonzalez, 2001; Flinch, 2003). The Mesa Formation has been stratigraphically subdivided in three members, i.e., from base to top, the Las Palmas, Bernal and Lumbí Members (De Porta, 1965). The latters have been distinguished through their lithological composition. The Las Palmas Member consists mainly of gravels and sandy gravels made of basaltic and andesitic clasts. The Bernal Member conformably overlies the latter and is constituted principally of volcanic rocks and sandy beds of volcanic composition. The Lumbí Member comprises gravels, sands and white mud. This formation is 430 m thick (De Porta, 1965) and unconformably overlies the Honda Formation (Butler, 1942; De Porta, 1965). The Las Palmas and Lumbí Members have been radiometrically dated at respectively 4,3 ± 0,4 Ma y 3,5 ± 0,4 Ma (Thouret, 1989).

The Popayán Formation crops out in the Patía and Cauca Basins south of Cali (Fig. 4.1). It is constituted mainly of volcanic rocks with many intercalations of fluvio-lacustrine deposits (Hubach and Alvarado, 1934; Grosse, 1935; Hubach, 1957; Salinas et al., 2007). It is subhorizontal and unconformably overlies the Esmita and Galeón Formations (Pérez and E., 1980; Mantilla and Arias, 2001). The Popayán Fm presents a variable thickness, reaching a thickness of 600 m (Hubach, 1957) and is of Pleistocene age (Hubach, 1957; Orrego and Paris, 1999). Similarly to the North Cauca Basin studied here, the Amagá and Patía Basins are considered as 76 Chapter 4

Table 4.2: Proposed new lithostratigraphical scheme of the Cauca depression and Quindío-Risaralda Basin. collision-related basins (Barrero et al., 2007). In the latter two basins, detailed stratigraphical studies about their latest Tertiary infill are significantly lacking. Those might be very useful in orderto understand the tectonic control on their sedimentary infill.

4.7. Conclusions

Field investigations in the North Cauca interandean basin have led to the redefinition of the uppermost Tertiary Zarzal Formation on the basis of the different sedimentary environments encountered: (a) fluvio-lacustrine, (b) mass-flows, (c) Gilbert-type delta and (d) alluvial fans. Facies are mineralogically and sedimentologically distinct and heterotopic.

These sedimentary environments have been correlated not only with each other, but also with the Quindío-Risaralda volcaniclastic fan deposits derived from the Central Cordillera to the east. Therefore, a new lithostratigraphical group called the Santa Barbara Group is proposed here. It includes the redefined Zarzal and the new Quindío-Risaralda Formations. Three Members have been distinguished within the Zarzal Fm (the Obando, Ansermanuevo and Holguín Members), according to their depositional environment and the origin of their lithological constituents.

According to palynological and geochronological 40Ar/39Ar datings, the outcropping deposits of the Santa Barbara Group have a Plio-Pleistocene age.

This detailed sedimentological and lithostratigraphical study of mainly undisturbed, uppermost Tertiary, sedimentary sequences unconformably overlying folded, older Tertiary rocks has shown the close relationship existing between the Plio-Pleistocene infill of this intramountane basin and the tectonic activity affecting the bounding reliefs. It can be used as a case history for similar investigations of poorly studied sediments in other interandean basins of Colombia, such as the Amaga, Magdalena or Cauca-Patia basins. Such investigations are valuable complements to structural studies on the surrounding reliefs. Lithostratigraphy 77

Re f e r e n c e s

Alfaro, C. and Aguirre, A. S., 2003. Geoquímica de fuentes minerales y thermales del complejo volcánico Cerro Bravo-Machín, Colombia. IX Congreso Colombiano de Geología, Medellin.

Alvárez, J. A., 1983. Geología de la Cordillera Central y Occidente Colombiano y petroquímica de los Intrusivos granitoides mesocenozoicos. Boletín Geológico INGEOMINAS Bogotá 26, 175.

Andriessen, P. A. M., Helmens, K. F., Hooghiemstra, H., Riezebos, P. A. and van, d. H. T., 1993. Absolute chronology of the Pliocene-Quaternary sediment sequence of the Bogota area, Colombia. Quaternary Science Reviews 12(7), 483-501.

Aspden, J. A. and McCourt, W. J., 1986. Mesozoic oceanic terrane in the Central Andes of Colombia. Geology (Boulder) 14(5), 415-418.

Aspden, J. A., McCourt, W. J. and Brook, M., 1987. Geometrical control of subduction-related magmatism; the Mesozoic and Cenozoic plutonic history of western Colombia. Journal of the Geological Society of London 144(6), 893-905.

Audemard, F. A., 2003. Geomorphic and geologic evidence of ongoing uplift and deformation in the Mérida Andes, Venezuela. Quaternary International 101-102, 43-65.

Barrero, D., Pardo, A. s., Vargas, C. A. and Martínez, J. F., 2007. Colombian Sedimentary Basins: Nomenclature, Boundaries and Petroleum Geology, a New Proposal, ANH and B&M Exploration Ltda: 92pp. 78 Chapter 4

Bermúdez, A., Garzón, M., Evans, R. and Aucott, J., 1985. Estudio Gravimétrico del Valle del río Cauca, Dpto. del Valle. Ingeominas, Bogotá., 25pp.

Boussingault, J. B., 1903. Memorias. Biblioteca Nacional de Colombia. Bogotá.

Butler, J. W., 1942. Geology of the Honda Distric. Colombia. American Association of Petroleum Geology Bulletin, 26 (5): 793-837. Tulsa.

Caballero, A. and Zapata, G., 1983. Mapa geológico de Colombia, Plancha N°224 - Pereira. INGEOMINAS, Medellín.

Cortes, M. and Angelier, J., 2005. Current states of stress in the northern Andes as indicated by focal mechanisms of earthquakes. Tectonophysics 403(1-4), 29-58.

De Porta, J., 1965. La Estratigrafía del Cretácico superior y Terciario en el extremo Sur del Valle Medio del Magdalena. Boletín Geológico, Universidad Industrial de Santander, (19): 5-30. Bucaramanga.

De Porta, J., 1974. Lexique stratigraphique international. C. N. d. l. r. scientifique.Paris. In. 5, 4b.

Droux, A. and Delaloye, M., 1996. Petrography and geochemistry of Plio-Quaternary calc-alkaline volcanoes of Southwestern Colombia. Journal of South American Earth Sciences 9, 27-41.

Dueñas, H. and Castro, G., 1981. Asociación palinológica de la Formación Mesa en la región de Falan, Tolima, Colombia. Geol. Norandina 3(27-36).

Espinosa, A., 2000. Mapa preliminar de fallas activas del Quindío. Primera fase: zona central. Armenia, Universidad del Quindío, Facultad de Ingenieria. Lithostratigraphy 79

Estrada, J. J. and Viana, R., 1998. Mapa geológico de Colombia, Plancha 205 - Chinchiná. INGEOMINAS, Medellín.

Estrada, J. J., VIANA, R. and GONZÁLEZ, H., 2001. Geología de la plancha 205 Chinchiná, memoria explicativa. Medellín, Ingeominas.

Flinch, J. F., 2003. Structural evolution of the Sinu-Lower Magdalena area (Northern Colombia), in C. Bartolini, R. T. Buffler, and J. Blickwede, eds., The Circum-Gulf of Mexico and the Caribbean: hydrocarbon habitats, basin formation, and plate tectonics: AAPG Memoir 79, p. 776–796.

Flores R., M., 1990. Transverse and longitudinal gilbert-type deltas, Tertiary Coalmont Formation, North Park Basin, Colorado, USA.; Coarse-grained deltas. Special Publication of the International Association of Sedimentologists 10, 223-233.

Freymueller, J. T., Kellogg, J. N. and Vega, V., 1993. Plate motions in the North Andean region. Journal of Geophysical Research, B, Solid Earth and Planets 98(12), 21,853-21,863.

Gallego, A., Ospina, L. M. and Osorio, J., 2005. Sismo del Quindío del 25 de enero del 1999, evaluación morfotectónica y sismológica. Boletin de Geología, Universidad Industrial de Santander (UIS) 27, 133-150.

Gonzalez, H., 2001. Mapa Geológico del departamento de Antioquia, Memorai Explicativa, INGEOMINAS, Colombia.

Gonzalez, H. and Nuñez, A., 1991. Mapa Geológico generalizado del departamento del Quindío. Bogotá, Ingeominas.

Grosse, E., 1935. Acerca de la geología del sur de Colombia. II Compilación de estudios geológicos de Colombia, p. 139 - 231.

Guarin, F., 2008. Etude sédimentologique du cône volcanoclastique du Quindío-Risaralda (Colombie Centrale) et sa relation avec la morphotectonique. Terre et environnement, Section des Sciences de la Terre, Université de Genève. Ph.D. thesis.

Guarin, F., Gorin, G., Espinosa, A, 2004. A Plio-Pleistocene succession of volcanic massflows in Central Colombia: the Quindío-Risaralda fan. IAVCEI General Assembly, Pucón, Chile.

Hooghiemstra, H. and Cleef, A. M., 1995. Pleistocene climatic change and environmental and genetic dynamics in the North Andean montane forest and páramo. Biodiversity and conservation of neotropical montane forests. S. e. a. Churchill. New York. In Botanical Garden, 35-49.

Hooghiemstra, H. and Sarmiento, G., 1991. Long continental pollen record from a tropical intermontane basin; late Pliocene and Pleistocene history from a 540-meter core. Episodes 14(2), 107-115.

Hubach, E., 1957. Contribución a la Unidades Estratigráficas de Colombia. Instituto Geológico Nacional. Informe No. 1212. P. 1 - 166.

Hubach, E. and Alvarado, B., 1934. Geología de los Departamentos del Valle y del Cauca, en Especial del Carbón, pp. 239, Servicio Geológico Nacional, Bogotá.

Keith, J. F., Rine, J. M. and Sacks, P. E., 1988. Frontier Basins of Colombia, Valle del Cauca, University of South Carolina, Earth Sciences and Resources Institute, Field report: 267 p.

Mantilla, A. M. and Arias, J. A., 2001. Síntesis regional y caracterización estructural de la subcuenca del Patía, ECOPETROL, Bogotá: pp. 74.

Massari, F. and Colella, A., 1988. Evolution and types of fan-delta systems in some major tectonic settings. Fan deltas; sedimentology and tectonic settings. W. Nemec and R. J. Steel. Glasgow, United Kingdom. In Blackie and Son, 103-122.

McCourt, W. J., 1984. The geology of the Central Cordillera in the departments of Valle del Cauca, Quindío and NW Tolima. Cali, INGEOMINAS, Brit. Geol. Surv.: 37.

McCourt, W. J. and Feininger, T., 1984. High-pressure metamorphic rocks in the Central Cordillera of Colombia. BGS Report 16(1), 28-35.

McCourt, W. J., Mosquera, D., Nivia, A. and Nuñez, A., 1984b. Mapa Geológico Preliminar, Plancha 243 - Armenia. INGEOMINAS. Bogotá.

Miall, A. D., 2006. The Geology of Fluvial Deposits Sedimentary Facies, Basin Analysis, and Petroleum Geology. 1st ed. 1996. Corr. 4rd printing, 2006, XVI, 582 p. 504 illus., Hardcover ISBN: 978-3-540-59186-3-59186-9.

Nelson, H. W., 1962. Contribución al conocimiento de la Cordillera Occidental. Sección entre Cali- Buenaventura. Boletín Geologíco INGEOMINAS, Cali, 10.

Nivia, A., 2001. Mapa geológico del departamento del Valle del Cauca, memoria explicativa, INGEOMINAS, Colombia.

Nivia, A., Galvis, N. and Maya, M., 1992. Mapa geológico de Colombia, plancha 242 - Zarzal, Cali. 82 Chapter 4

Nivia, A., Galvis, N. and Maya, M., 1995. Mapa geológico de Colombia, plancha 242 - Zarzal. Cali.

Nivia, A., Marriner, G. F., Kerr, A. C. and Tarney, J., 2006. The Quebradagrande Complex: A Lower Cretaceous ensialic marginal basin in the Central Cordillera of the Colombian Andes. Journal of South American Earth Sciences Tectonic evolution of the Colombian Andes 21(4), 423-436.

Padilla, L. E., 1982. Un modelo estructural para el Valle del Cauca y sus implicaciones hydrogeológicas y mineras. Boletín de Geología, Universidad Industrial de Santander (UIS), Bucaramanga 15, 67- 79.

Paris, G. and Romero, L. J. A., 1994. Fallas activas en Colombia. Boletin Geologico - Ingeominas 34(2-3), 3-25.

Parra, E., 1983. Mapa geológico de Colombia, Plancha N°223 - El Cairo (Valle del Cauca), INGEOMINAS, Cali, Colombia.

Pennington, W. D., 1981. Subduction of the eastern Panama Basin and seismotectonics of northwestern South America. JGR. Journal of Geophysical Research. B 86(11), 10753-10770.

Pérez, G. and E., N. L., 1980. Estudio Geológico de la Subcuenca Patía. 48 p. Informe Geológico No. 706 de Ecopetrol, ISN 14099.

Postma, G., 1990. Depositional architecture and facies of river and fan deltas; a synthesis. Coarse- grained deltas. A. Colella and B. Prior David. Oxford, International. In Blackwell. 10, 13-27. Lithostratigraphy 83

Restrepo, J. J. and Toussaint, J. F., 1988. Terranes and continental accretion in the Colombian Andes. Episodes 11(3), 189-193.

Reyes-Santos, J. P., Monsalve, M. M. and Gonzalez, J. S., 2000. Regiones Tectono-Sedimentarias del Valle Inferior del Magdalena, Colombia: Memoria VII Simposio Bolivariano, Exploracion Petrolera en las Cuencas Subandinas, Sociedad Venezolana de Geologos, Caracas, Venezuela, September 2000, p. 310-333.

Schwinn, W. L., 1969. Guidebook to the geology of the Cali area, Valle del Cauca, Colombia. In Tenth Annual Field Conference, Colombian Society of Petroleum Geologists and Geophysicists: pp. 277-297.

Smith G., A. and Lowe, D. R., 1991. Lahars: Volcano-hydrologic events and deposition in the debris flow-hyperconcentrated flow continuum (“SEPM, Spec. Publi.”, 45). 59-70.

Steffen, D. and Gorin, G. E., 1993. Palynofacies of the Upper Tithonian-Berriasian deep-sea carbonates in the Vocontian Trough (SE France). Bull. Centre rech. Explor. Prod. Elf-Aquitaine 17, 235-247.

Suter, F., Neuwerth, R., Gorin, G. E. and Guzman, C., 2008a. Depositional model of (Plio-)Pleistocene sediments in a tectonically active zone of Central Colombia. Geologica Acta 6(2), 1-19.

Thouret, J. C., 1989. Geomorfología y cronoestratigrafía del Macizo Volcánico Ruiz - Tolima (Cordillera Central Colombiana). En: T. Van Der Hammen, S. Diaz-Piedrahita, V.J. Alvarez (Editors) Estudios de Ecosistemas Tropoandinos, volumen 3, La Cordillera Central Colombiana - Transecto Parque Los Nevados (segunda parte): 257-277. J. Cramer. Berlín.

Torres, V., Vandenberghe, J. and Hooghiemstra, H., 2005. An environmental reconstruction of the sediment infill of the Bogota Basin (Colombia) during the last 3 million years from abiotic and biotic proxies. Palaeogeography, Palaeoclimatology, Palaeoecology 226(1-2), 127-148.

USGS. (2005). “http://edc.usgs.gov/products/elevation.html.”

Van der Hammen, T., 1958. Estratigrafía de Terciario y Maestrichtiano continentales y tectogenesis de los Andes colombianos. Boletin Geológico VI, 67-128.

Van der Hammen, T. and Hooghiemstra, H., 1997. Chronostratigraphy and correlation of the Pliocene and Quaternary of Colombia. The Plio-Pleistocene boundary. T. C. Partridge. Oxford, United Kingdom. In Pergamon. 40, 81-91.

Van Houten, F., 1976. Late Cenozoic vulcaniclastic deposits, Andean foredeep, Colombia. Geol. Soc. Am. Bull. 87, 481 - 495. CHAPTER V

So f t -s e d i m e n t d e f o r m a t i o n i n a t e c t o n i c a l l y a c t iv e a r e a : Th e Pl i o - Pl e i s t o c e n e Za rz a l Fo r m a t i o n i n t h e Ca u c a Va l l e y (We s t e r n Co l o m b i a )

Ralph Neuwerth, Fiore Suter, Carlos Guzman, Georges Gorin

Sedimentary Geology (2006) 86 Chapter 5

5.1. Introduction

The Colombian Andes comprise three SSW–NNE trending, sub-parallel cordilleras (Fig. 5.1). The complex tectonic framework of this area results from the interaction between three tectonic plates (Nazca, Caribbean and South American) and the Panama–Costa Rica microplate (Pennington, 1981; Taboada et al., 2000). This interaction induced a compressive tectonic regime in the north-western part of Colombia, particularly in the studied area located between the Western and Central Cordilleras, north of the city of Cali (Fig. 5.1).

Fig. 5.1: Megatectonic framework and location of study area.

The studied area covers parts of three Colombian departments (Fig. 5.2): Quindío (city of Armenia), Risaralda (city of Pereira) and Valle del Cauca (cities of Cartago and Zarzal). This zone comprises two intramontane depressions where Plio-Pleistocene sediments were deposited: the Cauca Valley to the west displays numerous outcrops of the Pleistocene Zarzal Formation between the towns of Zarzal Soft-sediment deformations 87 and Cartago. It is separated by folded Tertiary outcrops from the Quindío–Pereira depression to the east where the Plio-Pleistocene fluvio-volcanic fans of Quindío, Pereira and Cartago accumulated. These two sedimentary units interfinger in the western part of the Cartago, Pereira and Quindío Fans. This tectonically active zone is dissected by several major SSW–NNE trending fault lineaments such as the Romeral Fault System (Fig. 5.3). In particular, the rupture at shallow depth of the Armenia fault caused the dramatic earthquake of Armenia in January 1999 (INGEOMINAS (Instituto nacional de investigaciones geológico-mineras), 1999). The Ruiz-Tolima volcanic system in the Central Cordillera is associated with these faults and formed the source of the fluvio-volcanic fans deposited to the west (Fig. 5.2).

Fig. 5.2: Regional distribution of Plio-Pleistocene sediments in the Valle del Cauca, Risaralda and Quindío Departments. Location of Figs. 5.3 and 5.4.

The Zarzal Formation encountered in the Cauca Valley was deposited in a fluvio-lacustrine environment. It contains numerous soft-sediment deformation structures, particularly in the Cartago area (Fig. 5.4). The aims of this paper are to describe the various types of deformations encountered and discuss their potential triggering mechanisms.

5.2. Geology of Zarzal Formation

The Zarzal Formation has been so far poorly studied. Boussingault (1903) was the first to describe siliceous deposits intercalated with sand and sandy clay beds in the Cartago area, where they form low-relief hills. He considered these sediments as the infill of a lake. The name Zarzal Formation was attributed in 1955 to these deposits of diatomite, clay and volcanic sand (Keiser, Nelson and Van der Hammen, unpublished, in Van der Hammen, 1958; see also De Porta, 1974).

Cardona and Ortiz (1994) were the first to analyze in detail the depositional environment of these 88 Chapter 5 sediments. They interpreted three types of facies: braided-stream channel deposits, floodplain sediments and lake deposits. They also noted the geomorphological expression of this unit: low- relief hills dissected by a well-marked drainage pattern and numerous surface fractures. They recognized traces of block tectonics younger than the Pliocene Andine orogeny and also described the interdigitation of the Zarzal Formation with the fluvio-volcanic fans of Cartago and Pereira to the east.

Fig. 5.3: Geological cross-section across the Valle del Cauca and Quindío Departments. See Fig. 5.2 for location.

So far, no precise age dating has been achieved for the Zarzal Formation. According to Van der Hammen (1958), it assumes a probable Pliocene age for the Zarzal Formation without any scientific evidence. The Zarzal Formation disconformably overlies the Miocene La Paila Formation on the western flank of the Serranía de Santa Barbara (Fig. 5.2; McCourt, 1984; Nivia et al., 1995). In the Cauca Valley, the Zarzal Formation is disconformably overlain by gravels of alluvial fans fed by the surrounding reliefs, by grey palaeosols rich in volcanic materials and by recent alluvial sediments (Nivia et al., 1995). In fact, prior to the present research, the only certainty that exists is that the Zarzal Formation postdates the Pliocene Andine compression, i.e. it has a late Pliocene to Pleistocene age.

The dramatic earthquake of Armenia in January 1999 prompted the detailed geological study of the Plio-Pleistocene deposits in the Quindío Department, especially of the fluvio-volcanic fans and their relation with the Zarzal Formation in Quindío and the Cauca Valley (Figs. 5.2 and 5.3; Guarin, 2002; Guarin et al., 2004; Gorin et al., 2003; Suter, 2003; Suter et al., 2003). Detailed field studies confirm the observations of Cardona and Ortiz (1994). The Zarzal Formation consists of autochthonous lacustrine Soft-sediment deformations 89 sediments (diatomites) with a variable degree of interfingering with volcanic, fluvio-volcanic and fluviatile influxes derived from surrounding sources. In the Cauca Valley, where numerous sections can be observed (Figs. 5.4 and 5.5), the degree of interfingering increases towards the east and ongoing studies show that the Cartago Fan (Fig. 5.2) probably supplied most of the allochthonous material. At the western edge of the Cartago Fan, volcanic mass flows are intercalated with Zarzal fluvio-lacustrine sediments and can be observed westwards up to the Ansermanuevo area (Figs. 5.4 and 5.5). Moreover, thin intercalations of volcanic ash are encountered in the diatomites. In the La Vieja and Roble valleys, east of the Serrania of Santa Barbara (Fig. 5.2), lacustrine deposits of the Zarzal Formation can also be observed underlying fluvio- volcanic sediments of the Quindío Fan (Suter, 2003). Preliminary palynological data from clays of the Zarzal Formation show a significant presence of Alnus pollen. Because the first record of this tree in Colombia dates back to 1 my (Hooghiemstra and Cleef, 1995), a large part of the Zarzal Formation is probably of Pleistocene age. These data are awaiting confirmation from ongoing radiometric dating in volcanic ashes.

Field work and aerial photographs have confirmed the evidence of recent tectonic activity, particularly the diverging drainage

Fig. 5.4: Detailed geological and location map of study area. See Fig. 5.2 for location. 90 Chapter 5

Fig. 5.5: E–W and N–S trending correlations of field sections in the Zarzal Formation. See Fig. 5.4 for location of field sections. Soft-sediment deformations 91 pattern of Zarzal outcrops in the Cauca Valley (Fig. 5.4). Most of the soft-sediment deformation structures described below are within the sections shown in Fig. 5.5.

5.3. Overview of soft-sediment deformations and classifications

Although different authors have proposed classifications, no universally accepted scheme exists for various reasons. First of all, the merely descriptive classifications do not take into account the processes which formed the structures. On the other hand, the genetic classifications are based on assumptions about the inferred processes and parameters that have acted during the deformation. In many cases, the relationships are not clear and their application to field work is difficult. Although some classifications have been proposed in order to link morphology and genetic processes, they do not answer all questions. Another difficulty is the large variety of descriptive terms. Although this paper does not aim at elaborating a new classification, a review of some proposed classifications is necessary, in order to clarify the terminology that will be used (Table 5.1). It refers mainly to Lowe (1975), Brenchley and Newall (1977), Mills (1983) and Owen (1987, 2003).

Lowe (1975) proposed a classification for water escapes structures. It is quite comprehensive but confusion may exist for some structures. Nevertheless, Lowe’s classification has been used in recent papers (Rossetti, 1999). Brenchley and Newall (1977) proposed a classification for contorted bedding, which has been updated by Mills (1983). Although the diagnostic features are relatively easy to establish, the direction of movement is difficult to infer. Moreover, it does not include soft-sediment structures such as dykes and water escape structures. More recently, Owen (1987, 2003) proposed two attractive classifications encompassing respectively all soft-sediment deformation structures and only load structures. That of 1987 intends to establish a practical scheme taking into account genetic

p () g Lowe (1975) Brenchley and Newall (1977) Owen (1987) Owen (2003) Type of Classification of Classification of contorted Expanded classification Classification of classification consolidation-related SSD bedding of SSD and working load structures classification for SSD Classification Morphology, solid–liquid Morphology and dissection Deformation mechanisms Morphology criteria kinematics and origin of of movement and driving force stresses causing flowage Proposed Convolute lamination Convolute lamination Convolute lamination Convolute lamination nomenclature Load structures Load casts Load casts Simple load casts Asymmetric load casts Pedulous load Cast Pseudonodules Attached pseudon Detached pseudon Ball and pillow Ball and pillow Ball and pillow Heave structures Slumps Slumps Dykes Clastic dykes Sills Sand volcanoes Dish structures Water-escape cusps Cusp Hydroplastic mixing layers Liquefaction layers and pockets Fluidization channels and layers Sand rolls Recumbent folds Wrinkle marks Multilayer folds Contorted heavy mineral laminae Deformed cross-bedding Recumbently folded Preglacial involutions cross-bedding

Table 5.1: Comparison between some classifications of soft-sediment deformation (SSD) structures showing nomenclatures and classification criteria. 92 Chapter 5 aspects. It can be considered as an extended version of the classifications proposed by Elliott (1965) and Allen (1982). Owen (1987) distinguishes three types of deformation mechanisms and five classes of driving forces. Simultaneously, based upon the same criteria, he suggests a classification that can be applied principally to deformation in sands. This genetic classification is very broad and covers all of the commonly encountered soft-sediment deformation structures. The second classification (Owen, 2003) is exclusively designed for load structures. It differentiates between load casts (including flame structures) and pseudonodules, which can be further subdivided. The different kinds of load structures are linked to a deformation series.

This short overview of classifications illustrates that they are either incomplete, difficult to apply in the field or include overlapping definitions. Consequently, most authors have grouped structures into morphological categories (e.g., Rossetti, 1999; Rossetti and Goes, 2000; Rodríguez-Pascua et al., 2000) or into a mixture of morphological and genetic categories (e.g., Alfaro et al., 1997).

Fig. 5.6: Types of soft-sediment deformation structures observed in the Zarzal Formation.

In the Zarzal Formation, most of the soft-sediment deformation structures reported by Lowe (1975), Mills (1983) and Owen (1987, 2003) can be observed. However, some complicated structures could not be related precisely to a classification. Consequently, the structures will first be described morphologically and grouped into four groups, further subdivided into 14 categories (Fig. 5.6). Subsequently, the processes potentially associated with their genesis will be discussed. Soft-sediment deformations 93

5.4. Soft-sediment deformations in the Zarzal Formation

Soft-sediment deformation structures in the Zarzal Formation are encountered mainly in the region between Cartago and Ansermanuevo (Fig. 5.4). Although the most frequently deformed lithologies are fine to medium-grained tuffaceous sands and clays, deformations are present in other lithologies, including diatomites and tuffaceous gravels. The following types of structures have been observed (Fig. 5.6).

5.4.1. Load structures

5.4.1.1 Load casts

The classification used here follows essentially that proposed by Owen (2003). Load casts are the most frequent structure encountered in the studied area. Although the term bcastQ is inadequate, because the structure cannot really be associated with a cast, the term is kept in order to avoid confusion in the nomenclature. Both categories proposed by Owen (2003) have been encountered.

5.4.1.1.1. Simple load casts. Their size varies from centimeters to meter. They occur in different lithologies, but mostly in tuffaceous sands and gravels overlying silty clays and diatomites (Fig. 5.7A). They show a concave profile and slightly penetrate into the underlying bed. Laminations are usually gently deformed, although in some cases this structure appears associated with more pronounced deformation, such as convolute lamination and water escape structures (Fig. 5.7A). It is similar to the sagging load castQ of Alfaro et al. (1997).

5.4.1.1.2. Pendulous load casts. Their size fluctuates from a few up to 50 cm (Fig. 5.7B). They occur in various lithologies but, similarly to the simple load casts, they are more frequently encountered in tuffaceous sands and gravels overlying clays and diatomites. Occasionally, such deformations may be observed in fine-grained sands overlying medium to coarse-grained sands. Their shape is very variable: generally they become narrower upwards (Fig. 5.7B), but this is not always the case (Figs. 5.7C and 5.8A). They show features similar to the drop structures of Anketell et al. (1970). Their base is generally sub-planar and resembles convex downward lobe, the thickness of which shows frequent lateral variations. The internal laminations are generally slightly deformed, but can also locally be strongly deformed and associated with convolute lamination and water escape structures.

5.4.1.2. Flame structures

Their size varies from centimeters to decimeters (Fig. 5.8A). In most cases, this structure is poorly developed and only rarely does it correspond to so-called mud diapirs (Owen, 2003). This structure affects only clays and diatomites. They appear always in association with load structures (Fig. 5.6).

5.4.1.3. Pseudonodules

This structure is less frequently encountered than load casts. Of the three types described by Owen (2003), the most frequently observed is the attached pseudonodule. Detached pseudonodules have 94 Chapter 5

5.4.1.3.1. Attached pseudonodules. Their size varies between 10 and 30 cm. They occur in medium and coarse-grained tuffaceous sands overlying fine and medium-grained tuffaceous sands and, sometimes, clays and diatomites. Locally, medium-grained sands overlying coarse-grained sands can be observed (Fig. 5.8B). Their shape is variable but generally corresponds to a concave profile, sometimes slightly deformed (Fig. 5.8B and C). The internal laminations are not totally concentrical but show a synclinal form, sometimes slightly deformed. They are often associated with load casts, Soft-sediment deformations 95 convolute laminations and soft sediment intrusions.

5.4.1.3.2. Detached pseudonodules. Their dimension varies between 5 and 20 cm. They are associated with fine to medium-grained sands floating in clays or fine-grained sands and display a concave-upward shape. Laminations are diffuse and slightly deformed, similar to those described by

Fig. 5.8: Load structures. (A) Pendulous load cast (drop structure, Alfaro et al., 1997), near section 8 (see Figs. 5.4 and 5.5 for location). It forms a pocket of medium-grained sands overlying deformed, finely laminated sands. The upper part of the latter is affected by flame structures (a). (B) Attached pseudonodule made of fine- medium-grained sands, which sank into coarse-grained sands (section 8, see Figs. 5.4 and 5.5 for location). (C) Attached pseudonodule (a), water escape structure (b), simple load cast (c), convolute lamination (d) (section 8, see Figs. 5.4 and 5.5 for location). The attached pseudonodule (a) displays slightly deformed laminations. Note the clay layer within (a) which seems to have been dislocated by the water escape. 96 Chapter 5

Rodríguez-Pascua et al. (2000) and thereby differing from classical pseudonodules. In some cases, this structure can be interpreted as lighter sediments sinking into denser sediment and associated with attached pseudonodules (Fig. 5.9A). This structure has been observed in association with load casts, soft sediment intrusions and convolute laminations.

Fig. 5.9: Load and water-escape structures. (A) Attached (a) and detached (b) pseudonodule (section 8, see Figs. 5.4 and 5.5 for location). Note that the clayey silt has a greater density than the liquefied sand. (B) Water escape cusps (a) (section 8, see Figs. 5.4 and 5.5 for location) formed by mediumgrained sands intruding fine-grained sands. Observe the laccolith shape of the fine-grained-sand intrusion into medium-grained-sands and silts (b). This structure is capped by hardly deformed silts, which have been penetrated by a sandy sill (c). (C) Dish and pillar structure (section 8, see Figs. 5.4 and 5.5 for location). The undisturbed laminations of the overlying medium-coarse-grained sand lense proves that the underlying deformation occurred prior to its deposition and is not related to loading. Soft-sediment deformations 97

5.4.2. Water escape structures

5.4.2.1. Water escape cusps

This structure is similar to that described by Rossetti (1999) and appears as bodies of sands that penetrate into overlying sand beds. They are morphologically similar to flame structures which occur only in mud, but differ in that cusps represent masses of underlying deformed sands that intrude into the overlying sand beds (Fig. 5.6). They display a curved shape without evidence of internal deformation (Fig. 5.9B). Their size generally varies between 20 and 30 cm.

5.4.2.2. Dish-and-pillar structures

They show the typical appearance described by many authors (e.g., Lowe and LoPiccolo, 1974): fine to medium-grained sands with shapes similar to concave-upward saucers, separated by subvertical columns of medium to coarse-grained sands (Fig. 5.9C). They are up to 15 cm wide. They do not show evidence of internal deformation, although in some cases the pillars may be slightly deformed.

5.4.2.3. Pocket-and-pillar structures

They display features similar to those described by Postma (1983). Pockets have a flat base and are concave-upward. Their height and diameter measure between 10 and 15 cm. They are filled with coarsegrained sands and fine gravels (Fig. 5.10A). Pillars are columnar, generally straight and vertical with a height between 5 and 15 cm. Their fill consists of undeformed to gently deformed coarse-grained sands. Dish-and-pillar and pocket-and-pillar structures have different origins but a similar morphology. They are abundant and are associated principally with convolute lamination and soft-sediment intrusions.

5.4.3. Soft-sediment intrusions

These structures defined by Lowe (1975) show variable morphology, composition and size. Because of this, it is hard to classify them. Together with load structures, they are the most common deformation encountered in the study area. In general, one can differentiate clastic sills from dykes.

5.4.3.1. Clastic sills

In most cases, sills are arranged like beds and can locally be confused with them. Normally they do not show internal structures, except for slightly deformed laminations. They consist predominantly of fine to medium-grained sands and occasionally coarse-grained sands. Their thickness fluctuates from 2 to 50 cm, most of them being some 10 cm thick. Their length varies from a few centimeters to 5 m. Some sills are ptygmatic, some bifurcate and others are broken up into smaller sills (Fig. 5.10B). Some are interconnected and others deformed by load structures.

5.4.3.2. Clastic dykes

Their composition is variable, consisting of predominantly coarse-grained and lesser medium-grained 98 Chapter 5

Fig. 5.10: Water escape and soft-sediment intrusion structures. (A) Pocket and pillar structure (section 8, see Figs. 5.4 and 5.5 for location). (B) Ptigmatic and bifurcated clastic dyke (section 6, see Figs. 5.4 and 5.5 for location). (C) Medium-grained sand, rooted vertical dyke intruding silty clays (near section 8, see Figs. 5.4 and 5.5 for location). This intrusion seems to be partially controlled by fractures (a). sands. They are internally massive. Dykes crosscut different types of lithologies, predominantly sands but also clays and diatomites. Their thickness varies between 5 and 30 cm. Their length may reach up to 1 m, but generally fluctuates between 10 and 50 cm. Their shape is variable, they may be contorted or fractured. One particular type of dyke (Figs. 5.10C and 5.11A) shows characteristics similar to those described by Rodríguez-Pascua et al. (2000) for dykes originating from the liquefaction of a basal sand bed. In this case, a set of dykes approximately 2 m from each other shows a clear Soft-sediment deformations 99

Fig. 5.11: Soft-sediment intrusion and other deformation structures. (A) Medium-grained sand, subvertical disconnected dyke (near section 8, see Figs. 5.4 and 5.5 for location) showing downward bending of intruded, fine-grained sediments (a) in the lower part. This feature indicates the lateral movement of the injection. (B) Disturbed laminites (near La Victoria, see Fig. 5.4 for location). (C) Convolute laminations (section 6, see Figs. 5.4 and 5.5 for location). interconnection with the basal sand feeding the intrusion. Dykes are composed of medium-grained sands. These vertical structures have a constant width of some 10 cm and their length reaches up to 1 m. Fractures seem to have locally guided the intrusion (Fig. 5.10C). They crosscut laminites, diatomites and thin beds of fine-grained sands. In the same outcrop, one can observe a set of lateral dykes not connected to a basal sand layer but laterally related to a connected dyke. This type of disconnected dyke crosscuts silts that may be contorted (Fig. 5.11C). The dimension and arrangement 100 Chapter 5 of disconnected dykes are similar to those of the connected dykes.

Because of their abundance, soft-sediment intrusions are related to almost all types of soft-sediment deformations, but most frequently with load structures, convolute laminations and dish-and-pillar structures.

5.4.4. Other structures

5.4.4.1. Disturbed laminites

This structure is associated with varve-like laminites, which consist of alternations of diatomites, clays and very fine-grained sands. The latter show a slight deformation but keep their thickness and continuity. These deformations are between 2 and 10 cm thick (Fig. 5.11B) and do not appear to be associated with other structures, except for some small clastic dykes.

5.4.4.2. Convolute laminations

This structure is not abundant. Sediments affected correspond in general to fine to medium-grained tuffaceous sands. The deformation is generally some 10 cm thick, but may reach up to 20 cm. The

Fig. 5.12: Other soft-sediment deformation structures. (A) Slump with subhorizontal axial plane (section 8, see Figs. 5.4 and 5.5 for location). (B) Synsedimentary faulting showing both listric (a) and reverse (b) faults (section 8, see Figs. 5.4 and 5.5 for location). This structure can be assimilated to a flower structure indicative of strike-slip movements. Small-size load cast (c) and flame (d) structures are also observed. Soft-sediment deformations 101 stratification is highly contorted and shows shapes similar to deformed synclines and anticlines (Fig. 5.11C). This structure is associated mainly with load structures, soft-sediment intrusions and slumps.

5.4.4.3. Slumps

This type of deformation is observed only sporadically, but can be locally quite significant. The lithologies involved are fine and medium-grained tuffaceous sands. The thickness of the deformation varies between 10 cm and 1 m (Fig. 5.12A). They may show features similar to an intraformational fold. The axial plane of the folds is generally subvertical but may also be subhorizontal (recumbent). They are associated with load structures, convolute laminations and synsedimentary faults.

5.4.4.4. Synsedimentary faults

This type of brittle structure affects intervals some 20 cm to 1 m thick. Several types of faults exist in the area studied, namely high-angle planar normal faults (Fig. 5.7B), listric normal faults and reverse faults (Fig. 5.12B). Offsets vary between 2 and 20 cm. Lithologies involved vary from one site to the other, but generally consist of fine to medium-grained tuffaceous sands and locally clays. Synsedimentary faults are associated mainly with load structures.

5.5. Discussion

5.5.1. Deformation mechanisms and driving forces

Soft-sediment deformation is the disruption of unlithified sedimentary strata (Mills, 1983). This disruption occurs in response to a deformation mechanism, a driving force and a triggering mechanism (Allen, 1982, 1986; Owen, 1987). The deformation mechanisms and driving forces have been reviewed by various authors (e.g., Allen, 1977; Lowe, 1975; Mills, 1983; Owen, 1987, 2003; Maltman, 1994; Maltman and Bolton, 2003). The deformation occurs before significant compaction of the sediments has taken place (Mills, 1983). If a driving force such as a reverse density gradient, slope failure, slumping or shear stress is acting, the sediment strength can be significantly reduced as a consequence of a process known as liquidization (Allen, 1977, 1982). As stated earlier (Anketell et al., 1970; Mills, 1983; Owen, 1987; Rossetti, 1999), various driving forces can act simultaneously during deformation, so that in many cases there is no unique cause of deformation. When sediment is liquidized, it can be deformed in response to relatively weak stresses, which under normal conditions would not affect it. Four types of liquidization can be differentiated: thixotropy, sensitivity, liquefaction and fluidization (Owen, 1987). A sequence of soft-sediment deformation structures originates from these processes.

In many cases, deformation mechanisms are initiated by an external agent, i.e. a triggering mechanism. Some of these mechanisms have been identified: artesian groundwater movement, earthquakes, storm currents and gravity flows (e.g., Lowe, 1975; Sims, 1975; Owen, 1987, 1996).

Deformation mechanisms and driving force are interpreted below for each structure. Discussion of 102 Chapter 5 the triggering mechanisms follows.

5.5.1.1. Load structures

These structures are formed in response to gravitational instability (Moretti et al., 1999). The origin of the simple load casts in the Zarzal sediments is mostly related to a reverse density gradient (Anketell et al., 1970). The driving force is associated with the difference in densities between sands/gravels and the underlying sands/silty clays and diatomites. When the liquidization occurs in both layers, a Rayleigh–Taylor instability forms (Selker, 1993). The gravitational readjustment leads simultaneously to a descent of the denser sediment and an ascent of the lighter sediment. The resulting deformation depends upon the contrast of dynamic viscosities (Anketell et al., 1970; Alfaro et al., 1997). In the case of the Zarzal Formation, the differences were minimal or the lower layer had a higher dynamic viscosity favouring the formation of simple load casts.

In some cases uneven loading probably acted as a driving force, where a marked difference in densities did not exist, as for example between fine-grained sands and silty clays. The force is associated with lateral variations in the distribution of sediment load when the substrate is liquidized and loses its capacity of support (Owen, 2003). In some cases, the presence in the lower layer of convolute laminations formed by the liquidization process seems to support this interpretation.

The pendulous load casts have a similar origin but are associated with a more advanced stage of deformation (Anketell et al., 1970; Allen, 1982; Alfaro et al., 1997, 1999; Owen, 2003). In the study area, this structure is present mainly where the difference in density is marked, for example between tuffaceous sands and clays or diatomites (Fig. 5.7B). Therefore, it is probable that a simple load cast evolves into a more deformed structure, the pendulous load cast (Owen, 2003) or into a drop structure (Alfaro et al., 1997, 1999). In some cases, the structure seems unrelated to a difference in density, for example when fine-grained sands overlie coarse-grained sands. In this case, it is likely that the deformation is linked to uneven loading, a similar mechanism to that postulated by Rodríguez-Pascua et al. (2000) for the origin of some pseudonodules. The liquefaction of the underlying layer produces a decrease in the bulk density and shear strength, thereby favouring the genesis of the structure.

Flame structures have been attributed to a high difference of dynamic viscosity between the interacting layers: the lower one has a dynamic viscosity notably lower than the upper one (Anketell et al., 1970) and consequently diapiric intrusions of fine-grained sediments take place and form flame structures (Mills, 1983). However, Owen (2003) suggests that the influence of relative viscosity on load structure morphology should be reconsidered and that the amplification rate plays an important role in the genesis of antiforms (flame structures) and synforms (load casts).

Attached pseudonodules are formed by various processes (Owen, 2003), as can be deduced from experimental data (Anketell et al., 1970; Owen, 1996; Moretti et al., 1999). In the Zarzal Formation, the most probable origin is uneven loading, because there is no marked difference between the lithology of pseudonodules and that of the associated matrix and, moreover, these deformations may be associated with convolute lamination (Fig. 5.8C). Where this structure corresponds to a normal Soft-sediment deformations 103 density gradient, the genesis is similar to that postulated by Rodríguez-Pascua et al. (2000): the liquefaction of the underlying layer generates a decrease in the bulk density and shear strength which allows the development of the pseudonodules (Fig. 5.8B).

Detached pseudonodules in the Zarzal Formation are related to the sinking of load casts in water- saturated fine-grained sediments as established by Kuenen (1958). The coexistence of load casts and detached pseudonodules supports this interpretation. However, in some cases, when the detached pseudonodule is composed of lighter material (Fig. 5.9A) than the surrounding sediment, its origin probably relates to uneven loading (Rodríguez-Pascua et al., 2000).

Some authors (Anketell et al., 1970; Owen, 2003) have proposed the existence of a deformation series for load structures. In the study area, one can observe a transition from simple load casts to pendulous load casts, attached pseudonodoles and detached pseudonodules, but without ball and pillow structures. This means that the deformation reached a relatively advanced stage.

5.5.1.2. Water escape structures

5.5.1.2.1. Water escape cusps. This structure has been interpreted by Owen (1996) as associated with local fluidization of the lower sand layer and a water escape process. The deformation produced by the water escape is superimposed on the deformation associated with a gravitationally unstable density gradient.

5.5.1.2.2. Dish-and-pillar and pocket-and-pillar structures. These structures have been interpreted by Lowe and LoPiccolo (1974) as originating during the compaction and dewatering of unconsolidated sediments. Pillars are formed during compaction associated with an explosive escape of water along vertical or subvertical columnar flow paths. Dishes are associated with dewatering and involve a complex interaction between escaping water, sediments and sedimentary structures. According to Lowe and LoPiccolo (1974), the concave-upward morphology of the dishes would indicate that the deformation was relatively important. The model proposed by Cheel and Rust (1986), whereby these structures make part of a series including the ball-and-pillow structures, does not seem appropriate in the studied sediments where no relation exists between ball-and-pillow and dish-and-pillar structures. According to some authors (e.g., Lowe and LoPiccolo, 1974; Lowe, 1975; Allen, 1982), the triggering mechanism of these structures is related to overloading and sediment gravity flows. However, other authors (e.g. Plaziat and Ahmamou, 1998; Moretti et al., 1999) associate them with seismicity. In the study area, the latter activity seems to be the principal triggering factor (see below).

According to Postma (1983), the pocket-and-pillar structures are fluid escape structures resulting from the local liquefaction and fluidization associated with high rates of sediment deposition and the presence of less permeable and texturally immature sediments. In the Zarzal Formation, the latter condition exists, whereas rapid deposition rates have not been established.

5.5.1.3. Soft-sediment intrusions

These structures have been generally interpreted as originating from the injection of liquidized 104 Chapter 5 sands into the surrounding, mostly overlying strata (e.g., Walton and O’Sullivan, 1950; Potter and Pettijohn, 1977; Lowe, 1975). Most of the intrusions in the Zarzal Formation are interpreted as liquefied intrusions (Lowe, 1975), because they are structureless and associated with deformation of the adjacent strata. Fluidized intrusions are also present in a lesser proportion and correspond to sands injected along fractures and bedding planes (Fig. 5.10C). The presence of ptygmatic intrusions (Fig. 5.10B) indicates that the injection is associated with a higher degree of compaction in the surrounding sediments (Kuenen, 1968).

The presence of dykes vertically connected to a sand layer reflects an upward movement. The lateral dykes are associated with a lateral sand flow which produces the upward/downward bending of the intruded strata (Fig. 5.11A). In both cases, the deformation is linked to the liquefaction of the basal sand (Rodríguez-Pascua et al., 2000). Some authors (e.g., Martel and Gibling, 1993; Parize et al., 1987; Parize and Fries, 2003) relate clastic dykes to sediment gravity flows and storm waves. However, in the Zarzal Formation, there is no sedimentological evidence supporting the latter mechanisms (see below).

5.5.1.4. Other structures

5.5.1.4.1. Disturbed laminites. A similar structure to this has been described by Rodríguez-Pascua et al. (2000) and interpreted as the product of ductile deformation, although they did not observe changes in thickness, but only a bending of the laminites. Probably the sediments kept some residual strength favouring a weak, ductile behaviour (Maltman and Bolton, 2003).

5.5.1.4.2. Convolute laminations. Although no unique interpretation exists with respect to the generating mechanism, this structure has generally been considered as an extremely complex form of load structure (Dzulynski and Smith, 1963; Lowe, 1975; Mills, 1983; Rossetti, 1999). Interpretation of the driving force varies among authors: some relate it to current drag or bed shear and others to slumping (Mills, 1983; Owen, 1996; Plaziat and Ahmamou, 1998). In the Zarzal Formation, these structures are not associated with clays and the probability of thixotropic behaviour is low. Consequently, the mechanism is probably linked to fluidization contrasts which create gravitational instabilities (Brenchley and Newall, 1977; Owen, 1996; Rossetti, 1999). In a few cases, the association of convolute lamination with soft-sediment intrusions suggests that a slope in the substrate may have been involved and that the structure is related to a hydroplastic deformation (Plaziat and Ahmamou, 1998).

5.5.1.4.3. Slumps. These structures are associated with the downslope movement of underconsolidated sediments under the influence of gravity. The failure occurs when the sediments are steepened beyond the stable angle of repose (Mills, 1983). Considering the minimal mixing of sediments and the general conservation of the bedding, the slumps can be classified as coherent (sensu Dzulynski, 1963) and resemble the bcontortedQ slumps mentioned by McKee et al. (1962).

5.5.1.4.4. Synsedimentary faults. This brittle deformation corresponds to a cohesive behaviour (Owen, 1987; Vanneste et al., 1999), whereby the increase in pore water pressure is not enough to Soft-sediment deformations 105 liquefy the sediments. The presence of faults and their association with undeformed strata correspond to a brittle deformation when sediments are either unconsolidated or partly consolidated (Rossetti and Goes, 2000). The coexistence of structures associated with ductile and brittle deformations can be related to differential compaction, which determines the pore pressure within the sediments: the more compacted and less saturated sediments have a brittle behaviour, whereas the less compacted and more saturated sediments show a ductile behaviour (Mohindra and Bagati, 1996; Rossetti, 1999; Rossetti and Goes, 2000).

5.5.2. Triggering mechanisms

Several mechanisms can trigger synsedimentary deformation. The best known are sediment loading (e.g. Anketell et al., 1970; Lowe and LoPiccolo, 1974), storm currents (e.g., Dalrymple, 1979; Molina et al., 1998; Alfaro et al., 2002) and seismicity (e.g., Seilacher, 1969; Lowe, 1975; Sims, 1975; Martel and Gibling, 1993; Mohindra and Bagati, 1996; Calvo et al., 1998; Lignier et al., 1998; Rossetti, 1999; Alfaro et al., 1999; Vanneste et al., 1999; Jones and Omoto, 2000; Rodríguez-Pascua et al., 2000; Bowman et al., 2004).

Fig. 5.13: Post-depositional extensional tectonics in the Zarzal Formation (section 7, see Figs. 5.4 and 5.5 for location).

Sediment loading seems to be of minor importance in the Zarzal Formation. Such processes could be associated with the rapid deposition of tuffaceous sands and gravels originating form the reworking 106 Chapter 5 of pyroclastic material and gravity flows. The weight of these sediments on water-saturated sediments could be a triggering mechanism (Owen, 1996; Jones and Omoto, 2000). However, no direct relationship between the loading of tuffaceous sediments and soft-sediment deformations has been observed in the field.

Fig. 5.14: Post-depositional strike-slip tectonics in the Zarzal Formation (section 8, see Figs. 5.4 and 5.5 for location).

Storm currents can be a triggering mechanism for soft-sediment deformations. The Zarzal Formation shows no evidence of any sedimentary structure associated with high energy (for example swaley and hummocky cross-stratification , see Molina et al., 1998). On the other hand, the vast lateral extent of the soft sediment deformation observed suggests a more regional triggering mechanism than one related to the local action of storms. Moreover, the liquefaction of sediments requires a storm wave height in excess of 6 m (Alfaro et al., 2002), something impossible in a lake like that in which the Zarzal Formation was deposited.

Seismicity is the most probable triggering mechanism of the soft sediment deformation structures encountered in this study. Seismicity can cause the fluidization of granular solids (Allen, 1982, 1986; Lowe, 1975). Although all the structures characterizing seismites (Seilacher, 1969) have not yet been encountered, a sufficient number of criteria (Bowman et al., 2004 and references therein) confirming the role of seismicity as triggering mechanism are established.

The studied area has a well-established present-day seismicity (CRQ (Corporación autonoma Regional del Quindío) and Univ. del Quindío, 1999) related to the Cauca-Romeral Fault System. The latter represents a Cretaceous paleosuture and has been active since then (McCourt, 1984). In recent times, it has been related with the earthquakes that affected the cities of Manizales, Pereira Soft-sediment deformations 107 and Armenia. Field evidence confirms that tectonic faulting (Figs. 5.13 and 5.14) and tilting (see drainage pattern, Fig. 5.4) have affected the Zarzal Formation. Moreover, the lithology of the Zarzal Formation corresponds to sediments quite to deformation when exposed to seismic waves.

The structures encountered in the Zarzal Formation are similar, both in size and shape, to those described in the field as seismites (e.g., Sims, 1975; Calvo et al., 1998; Rossetti, 1999; Vanneste et al., 1999; Rossetti and Goes, 2000; Lignier et al., 1998; Jones and Omoto, 2000; Greb and Dever, 2002; Bowman et al., 2004) or those generated experimentally (e.g., Kuenen, 1958; Owen, 1996; Moretti et al., 1999). The deformed intervals are intercalated within undeformed strata, thereby reflecting catastrophic events followed by periods of relative stability. Moreover, the deformed strata can be observed throughout a fairly large area (Figs. 5.4 and 5.5) although, so far, the lack of precise dating does not permit the correlation of the events in the different studied sections. Moreover, the classification of Wheeler (2002) provides further criteria pointing to the presence of seismites: among the six criteria (or tests) identified by Wheeler (2002) for evaluating the occurrence of seismites, Zarzal sediments fulfill four tests, the other two being inconclusive.

The abundance of soft-sediment deformation structures indicates the proximity of an active fault zone. However, further detailed field investigations are needed to precisely determine which were the active faults within the system. The soft-sediment deformation in the Zarzal Formation can be

Fig. 5.15: Injection dykes (section 8, see Figs. 5.4 and 5.5 for location) are a proof of extension (Rodríguez- Pascua et al., 2000). 108 Chapter 5 interpreted as seismites related to the activity of faults associated with the Cauca-Romeral Faults System. Field evidence demonstrates synsedimentary tectonic activity with both strike-slip (Fig. 5.12) and extensional faults (Fig. 5.7B). Similar tectonic activity continued after the deposition of the unit, as demonstrated by the occurrence of post-depositional strike-slip and extensional movements (Figs. 5.13 and 5.14). This area of Colombia is in a transpressional regime (Taboada et al., 2000), which generates strikeslip faults and associated local normal faults (pull-apart basins).

The relation between soft-sediment deformations and the magnitude of earthquakes has received much attention (e.g., Seed and Idriss, 1971; Sims, 1975; Allen, 1986; Scott and Price, 1988; Marco and Agnon, 1995; Obermaier et al., 2002). Although some authors (e.g., Seed and Idriss, 1971) consider that magnitudes from 2 to 3 are enough to trigger liquefaction, most scientists estimate that the magnitude of an earthquake should be higher than 4.5 (e.g., Marco and Agnon, 1995) to be registered in the sedimentary record. Scott and Price (1988) suggest that a magnitude of less than 5 does not cause a significant sediment liquefaction beyond 4 km from the epicenter and that a magnitude of 7 does not significantly affect sediments beyond 20 km. Considering these values and the distance to the faults that might have been active during the sedimentation of the Zarzal Formation in the Cauca Valley (e.g., some 10 km to the Quebradanueva Fault, Fig. 5.4), it can be tentatively postulated that the magnitude of the earthquakes having generated the seismites was between 5 and 7. The record of recent instrumental seismicity confirms the magnitude of earthquakes in this area (CRQ (Corporación autonoma Regional del Quindío) and Univ. del Quindío, 1999). This range of postulated magnitudes corresponds with the scale established by Rodríguez-Pascua et al. (2000) for a series of soft sediment deformation structures: they postulate that the existence of sand dykes and pseudonodules indicates magnitudes between 5 and 7 (Fig. 5.15).

5.6. Conclusions

The Plio-Pleistocene Zarzal Formation in the Cauca Valley was deposited in an intramontane depression between the Western and Central Cordilleras of Central Colombia, a zone affected by the movements of the Cauca-Romeral Fault system. Fine-grained lacustrine sediments (sands, silts, clays and diatomites) alternate with fluvial-dominated coarser sands and gravels and fluvio-volcanic mass flows derived from the Central Cordillera to the east.

Sands, silts and clays in the Zarzal Formation preserve intervals with abundant soft-sediment deformation interbedded with undeformed strata. Deformation structures can be classified into four groups: load structures (load casts, flame structures and pseudonodules), water escape structures (water escape cusps, dish and pillar, pocket and pillar), soft-sediment intrusions (clastic sills and dykes) and other structures (disturbed laminites, convolute laminations, slumps and synsedimentary faults).

Deformation mechanisms and driving forces of these deformations correlate with those known in the literature: load structures are the result of gravitational instabilities related to density differences or uneven loading; water escape structures are associated with dewatering and soft-sediment intrusions with the injection of liquidized sands into surrounding strata; disturbed laminites are the result of Soft-sediment deformations 109 ductile deformation, convolute laminations are linked to gravitational instabilities associated with inverse density gradients and slumps to gravitational downslope movements. Synsedimentary faulting is a brittle deformation that takes place when the pore pressure is not sufficient to liquefy the sediments.

Field evidence and regional geological criteria point to seismicity as the most probable triggering mechanisms for the deformations, which are consequently interpreted as seismites. There is no field evidence of the influence of sediment loading phenomena or of deformation linked to storm activity. The studied area has a history of geological and recent seismicity related to the Cauca-Romeral Fault System. The sandy lithology of the Zarzal sediments is prone to liquefaction when exposed to earthquakes and structures encountered are similar in size and shape to those encountered in seismites. Moreover, the disturbed intervals have a large areal extent and their intercalation within undeformed strata reflects the catastrophic nature of these events.

In summary, the Plio-Pleistocene Zarzal Formation represents a new case history of soft-sediment deformations related to earthquakes. 110 Chapter 5

References

Alfaro, P., Moretti, M., Soria, J.M., 1997. Soft-sediment deformation structures induced by earthquakes (seismites) in Pliocene lacustrine deposits (Guadix-Baza Basin, Central Betic Cordillera). Eclogae Geol. Helv. 90, 531–540.

Alfaro, P., Estévez, A., Moretti, M., Soria, J.M., 1999. Structures sédimentaires de déformation interprétées comme seismites dans le Quaternaire du bassin du Bas Segura (Cordillère bétique orientale). C. R. Acad. Sci. Parism, Sci. Terre et des Planetes 328, 17– 22.

Alfaro, P., Delgado, J., Estévez, A., Molina, J.M., Moretti, M., Soria, J.M., 2002. Liquefaction and fluidization structures in Messinian storm deposits (Bajo Segura Basin, Betic Cordillera, southern Spain). Int. J. Earth Sci. (Geol. Rudnsch.) 91, 505– 513.

Allen, J.R.L., 1977. The possible mechanisms of convolute laminations in graded sand beds. J. Geol. Soc. 134, 19– 31.

Allen, J.R.L., 1982. Sedimentary structures: their character and physical basis. Developments in Sedimentology vol. 30. Elsevier, Amsterdam. 663 pp.

Allen, J.R.L., 1986. Earthquake magnitude-frequency, epicentral distance, and soft-sediment deformation in sedimentary basins. Sediment. Geol. 46, 67– 75.

Anketell, J.M., Cegla, J., Dzulynski, S., 1970. On the deformational structures in systems with reversed density gradients. Ann. Soc. Geol. Pol. 40, 3 – 30.

Boussingault, J.B., 1903. Memorias vol. 2. Biblioteca Nacional de Colombia, Bogotá.

Bowman, D., Korjenkov, A., Porat, N., 2004. Late-Pleistocene seismites from Lake Issyk-Kul, The Tien Shan range, Kyrghyztan. Sediment. Geol. 163, 211 –228.

Brenchley, P.J., Newall, G., 1977. The significance of contorted bedding in the Upper Ordovician sediments of the Oslo region, Norway. J. Sediment. Petrol. 47, 819– 833.

Calvo, J.P., Rodríguez-Pascua, M.A., Martín-Velásquez, S., Jiménez, S., De Vicente, G., 1998. Microdeformation of lacustrine laminate sequences from Late Miocene formations of SE Spain: an interpretation of loop bedding. Sedimentology 45, 279– 292. Soft-sediment deformations 111

Cardona, J.F., Ortiz, M., 1994. Aspectos estratigráficos de las unidades del intervalo Plioceno Holoceno entre Pereira y Cartago. Propuesta de definición para la Formación Pereira. Grado Thesis, Universidad de Caldas, Manizales, Colombia (unpublished).

Cheel, R.J., Rust, B.R., 1986. A sequence of soft-sediment deformation (dewatering) structures in Late Quaternary subaqueous outwash near Ottawa, Canada. Sediment. Geol. 47, 77– 93.

CRQ (Corporación autonoma Regional del Quindío) and Univ. Del Quindío, 1999. Estudio de amenaza sísmica en los 26 municipios del Eje Cafetero afectados por el sismo del 25 de enero de 1999. Armenia. FOREC. 151 pp.

Dalrymple, R.W., 1979. Wave-induced liquefaction: a modern example from the Bay of Fundy. Sedimentology 26, 835– 844.

De Porta, J., 1974. Lexique stratigraphique international. Vol. 5 Amérique latine, Fasc. 4b. Centre National de la recherche scientifique, Paris.

Dzulynski, S., 1963. Directional structures in flysch. Stud. Geol. Pol. 12, 1 –136.

Dzulynski, S., Smith, A.J., 1963. Convolute lamination, its origin, preservation, and directional significance. J. Sediment. Petrol. 33, 616–627.

Elliott, R.E., 1965. A classification of subaqueous sedimentary structures based on rheological and kinematical parameters. Sedimentology 5, 193–209.

Gorin, G.E., Guarin, F., Espinosa, A., 2003. Mass-flows fluvio-volcaniques: exemple du fan du Quindío près d’Armenia (Colombie centrale). 9ème Congrès Français de Sédimentologie (Bordeaux)- Livre des résumés, Publ. vol. 38. ASF, Paris, pp. 238– 239.

Greb, S.F., Dever, G.R., 2002. Critical evaluation of possible seismites: examples from the Carboniferous of the Appalachian basin. In: Ettensohn, F.R., Rast, N., Brett, C.E.Ancient Seismites. Boulder. Geol. Soc. Amer. Spec. Paper, vol. 359, pp. 109– 125.

Guarin, F., 2002. Etude du fan fluvio-volcanique du Quindío (Colombie). M.Sc. thesis, Univ. of Geneva (Switzerland). 92 pp. (unpublished).

Guarin, F., Gorin, G., Espinosa, A., 2004. A Plio-Pleistocene Succession of Volcanic Massflows in 112 Chapter 5

Central Colombia: the Quindío-Risaralda Fan. IAVCEI General Assembly, Pucón, Chile.

Hooghiemstra, H., Cleef, A.M., 1995. Pleistocene climatic change and environmental and genetic dynamics in the North Andean montane forest and páramo. In: Churchill, S., et al., (Eds.), Biodiversity and Conservation of Neotropical Montane Forests, New York Botanical, pp. 35–49.

INGEOMINAS (Instituto nacional de investigaciones geológicomineras), 1999. Terremoto del Quindío (Enero 25 de 1999), informe técnico—científico. INGEOMINAS, Santa-Fé de Bogotá.

Jones, A.P., Omoto, K., 2000. Towards establishing criteria for identifying trigger mechanisms for soft-sediment deformation: a casestudy of Late Pleistocene lacustrine sands and clays, Onikobe and Nakayamadaira Basins, northeastern Japan. Sedimentology 47, 1211– 1226.

Kuenen, P.H., 1958. Experiments in geology. Geol. Mag. 23, 1– 28.

Kuenen, P.H., 1968. Origin of ptygmatic features. Tectonophysics 6, 143–158.

Lignier, V., Beck, C., Chapron, E., 1998. Caracterisation geometrique et texturale de pertubations synsedimentaires attribuees a des seismes, dans une formation quaternaire glaciolacustre des Alpes (T les Argiles du Trieves r). C.R. Acad. Sci. Paris, Sci. Terre et des Planètes 327, 645–652.

Lowe, D.R., 1975. Water escape structures in coarse grained sediments. Sedimentology 22, 157– 204.

Lowe, D.R., LoPiccolo, R.D., 1974. The characteristics and origins of dish and pillar structures. J. Sediment. Petrol. 44, 484– 501.

Maltman, A., 1994. Introduction and overview. In: Matlman (Ed.), The Geological Deformation of Sediments. Chapman and Hall, London, pp. 1 –35.

Maltman, A., Bolton, A., 2003. How sediments become mobilized. In: Van Rensbergen, P., Maltman, R.R., Morley, C.K. (Eds.), Subsurface Sediment Mobilization, Geol. Soc. Spec. Publ., London, vol. 216, pp. 9 – 20.

Marco, S., Agnon, A., 1995. Prehistoric earthquake deformations near Massada, Dead Sea graben. Geology 23, 695–698. Soft-sediment deformations 113

Martel, A.T., Gibling, M.R., 1993. Clastic dykes of the Devono-Carboniferous Horton Bluff Formation, Nova Scotia: storm-related structures in shallow lakes. Sediment. Geol. 87, 103–119.

McCourt, W.J., 1984. The geology of the Central Cordillera in the departments of Valle del Cauca, Quindío and NW Tolima. Report no. 8. INGEOMINAS - Brit. Geol. Surv., Cali. 37 pp.

McKee, E.D., Reynolds, M.A., Baker, C.H., 1962. Experiments on intraformational recumbents folds in cross-bedded sands. U. S. Geol. Surv. Prof. Pap. 450-D, 151– 155.

Mills, P.C., 1983. Genesis and diagnostic value of soft-sediment deformation structures—a review. Sediment. Geol. 35, 83– 104.

Mohindra, R., Bagati, T.N., 1996. Seismically induced soft-sediment deformation structures (seismites) around Sumdo in the lower Spiti valley (tethys Himalya). Sediment. Geol. 101, 69– 83.

Molina, J.M., Alfaro, P., Moretti, M., Soria, J.M., 1998. Soft-sediment deformation structures induced by cyclic stress of storm waves in tempestites (Miocene, Guadalquivir basin, Spain). Terra Nova 10, 145–150.

Moretti, M., Alfaro, P., Caselles, O., Canas, J.A., 1999. Modelling seismites with a digital shaking table. Tectonophysics 304, 369–383.

Nivia, A., Galvis, N., Maya, M., 1995. Mapa geológico de Colombia, plancha 242-Zarzal escala 1:100.000, Memoria explicativa. INGEOMINAS, Cali.

Obermaier, S.F., Pond, E.C., Olson, S.M., Green, R.A., 2002. Paleoliquefaction studies in continental settings. In: Ettensohn, F.R., Rast, N., Brett, C.E. (Eds.), Ancient Seismites, Boulder. Geol. Soc. Amer. Spec. Paper, vol. 359, pp. 13– 27.

Owen, G., 1987. Deformation processes in unconsolidated sands. In: Jones, M.E., Preston, R.M.F. (Eds.), Deformation of Sediments and Sedimentary Rocks, Geol. Soc. Spec. Publ., vol. 29, pp. 11 – 24.

Owen, G., 1996. Experimental soft-sediment deformation structures formed by the liquefaction of unconsolidated sands and some ancient examples. Sedimentology 43, 279– 293.

Owen, G., 2003. Load structures: gravity-driven sediment mobilization in the shallow subsurface. 114 Chapter 5

In: Van Rensebergen, P., Hillis, R.R., Maltman, A.J., Morley, C.K. (Eds.), Subsurface Sediment Mobilization, Geol. Soc. Spec. Publ. vol. 216. Geological Society of London, London, pp. 21– 34.

Parize, O., Fries, G., 2003. The Vocontian clastic dykes and sills: a geometric model. In: Van Rensebergen, P., Hillis, R.R., Maltman, A.J., Morley, C.K. (Eds.), Subsurface Sediment Mobilization, Geol. Soc. Spec. Publ. vol. 216. Geological Society of London, London, pp. 51–71.

Parize, O., Beaudoin, B., Fries, G., Pinault, M., Pinoteau, B., 1987. Les filons greseux sedimentaires. Notes et Memoires de TOTAL Compagnie francaise du Petrole, vol. 21, pp. 211 –233.

Plaziat, J.C., Ahmamou, M., 1998. Les differents mecanismes a l’origine de la diversite des seismites, leur identification dans le Pliocene du Saiss de Fes et de Meknes (Maroc) et leur signification tectonique. Geodin. Acta 11/4, 183– 203.

Pennington, W.D., 1981. Subduction of the eastern Panama basin and seismotectonics of Northwestern South America. J. Geophys. Res. 86, 10753–10770.

Potter, P.E., Pettijohn, F.J., 1977. Paleocurrents and Basin Analysis, 2nd. ed. Springer-Verlag, New York. 296 pp.

Postma, G., 1983. Water escape structures in the context of a depositional model of a mass flow dominated conglomeratic fan-delta (Abrioja Formation, Pliocene, Almeria Basin, Se Spain). Sedimentology 30, 91– 103.

Rodríguez-Pascua, Calvo, J.P., De Vicente, G., Gómez-Gras, D., 2000. Soft-sediment deformation structures interpreted as seismites in lacustrine sediments of the Prebetic Zone, SE Spain, and their potential use as indicators of earthquake magnitudes during the Late Miocene. Sediment. Geol. 135, 117– 135.

Rossetti, D.F., 1999. Soft-sediment deformation structures in late Albian to Cenomanian deposits, Sao Luis Basin, northern Brazil: evidence for palaeoseismicity. Sedimentology 46, 1065–1081.

Rossetti, D.F., Goes, A.M., 2000. Deciphering the sedimentological imprint of paleoseismic events: an example from the Aptian Codó Formation, northern Brazil. Sediment. Geol. 135, 137– 156.

Scott, B., Price, S., 1988. Earthquake-induced structures in young sediments. Tectonophysics 147, 165–170. Soft-sediment deformations 115

Seed, H.B., Idriss, I.M., 1971. Simplified procedure for evaluating soil liquefaction potential. J. Soil Mech. Found. Div: Am. Soc. Civ. Eng. 97, 1249– 1273.

Seilacher, A., 1969. Fault-graded beds interpreted as seismites. Sedimentology 13, 155– 159.

Selker, J.S., 1993. Expressions for the formation of load casts in soft sediments. J. Sediment. Petrol. 63, 1149– 1151.

Sims, J.D., 1975. Determining earthquake recurrence intervals from deformational structures in young lacustrine sediments. Tectonophysics 29, 141– 152.

Suter, F., 2003. Géologie de la reégion de Playa Azul, partie occidentale distale du fan fluvio- volcanique du Quindío (Serranía de Santa Barbara, Quindío et Valle del Cauca, Colombie). M. Sc. thesis, University of Geneva (Switzerland). 133 pp.

Suter, F., Espinosa, A., Gorin, G., 2003. Geología de la parte distal occidental del abanico fluvio- volcanico del Quindío. Abstract B001, IX Congreso Colombiano de Geología, Medellin, pp. 111 – 112.

Taboada, A., Rivera, L.A., Fuenzalida, A., Cisternas, A., Philip, H., Bijwaard, H., Olaya, J., Rivera, C., 2000. Geodynamics of the northern Andes: subductions and intracontinental deformation (Colombia). Tectonics 19/5, 787– 813.

Van der Hammen, T., 1958. Estratigrafía del Terciario y Maestrichtiano continentales y tectogénesis de los Andes colombianos. Bol. Geol. Serv. Geol. Natl. 6/1-3, 67– 128.

Vanneste, K., Meghraoui, M., Camelbeek, T., 1999. Late Quaternary earthquake-related soft- sediment deformation along the Belgian portion of the Feldbiss Fault, Lower Rhine Graben system. Tectonophysics 309, 57– 79.

Walton, M.S., O’Sullivan, R.B., 1950. The intrusive mechanics of clastic dikes. Am. J. Sci. 248, 1 – 21.

Wheeler, R.L., 2002. Distinguishing seismic from nonseismic soft-sediment structures: criteria from seismic-hazard analysis. In: Ettensohn, F.R., Rast, N., Brett, C.E. (Eds.), Ancient Seismites, Boulder, Geological Society of America Special Paper, vol. 359, pp. 1– 11.

CHAPTER VI

Te c t o n i c s

Ralph Neuwerth 118 Chapter 6

6.1. Introduction

A bibliographical review on structural geology, geodynamics and stratigraphy has already been presented in chapter 3. Therefore, this introductive part will focus on recent tectonic investigations carried out by (Guarin, 2008; Suter, 2008), which provide adequate information and hypotheses for this chapter.

(Guarin, 2008) has described the general shape of the Quindío-Risaralda Fan Basin and Serranía de Santa Barbara as rhomboidal (Fig. 6.1), with a preferential orientation in NE-SW and NW-SE directions. In the study zone, these directions correspond to the Palestina and Otún-Consota Fault Systems. The interaction of these fault systems is interpreted by this author as being probably responsible for a clockwise rotational movement (with respect to the horizontal plane). This movement would be induced by a single wrenching deformation associated with asymmetrical, extensional processes. Furthermore, the same author has subdivided the Quindío-Risaralda Fan in four tectonic zones: two distensional zones in the NW and SE, and two compressional zones to the NE and SW.

Fig. 6.1: Idealized model of the Quindío-Risaralda Fan and Serranía de Santa Barbara. Black and green arrows indicate respectively horizontal and vertical direction of rotation. (After (Guarin, 2008). Abbreviations: A: Armenia, M: Montenegro, F.MTG: Montenegro fault, F.MTA: Matecaña fault, F.ARM: Armenia fault, F. PTS = Potrerillos fault, C: compresion, E: extension.

(Suter et al., 2008b) has proposed a strain partitioning between the accretion-inherited Romeral Fault System (RFS) and “non-Romeral” shear fault systems. The latter are composed by (1) the ENE striking, right-lateral “Santa Rosa” fault group which indicates the shear direction and corresponds to Tectonics 119 the main fault orientation; (2) the “Ibagué” fault group forming a 19° angle with respect to the main fault and interpreted as synthetic riedels (R); and (3) the “Otun” fault group with an angle of 66° corresponding to the antithetic riedels (R’) (Fig. 6.2). North of the E–W trending lineament formed by the Ibagué and Garrapatas faults (4.5°N) the RFS changes its strike and is segmented. This E–W trending lineament probably corresponds to the southern limit of the area affected by the distributed shear strain. Furthermore, these authors have proposed that the closure of the Cauca River Valley north of La Virginia may be a consequence of the shortening generated at the Chocó Panamá Block indentation front.

Moreover, (Suter et al., in review) have interpreted the SSB as an active fold-and-thrust belt (Fig. 6.3). They have proposed that the uplift of this belt may be responsible for the tectonic closure of the Cauca Depression and consequently for its recent infill from Cali up to the Risaralda River Valley north of La Virginia.

These previous studies have provided a good, up-to-date, geological knowledge of the North Cauca Valley. To follow up on these results, a more regional view is needed in order to understand the geometry of the basin in a wider framework. Consequently, this chapter will present a large-scale geometrical analysis encompassing the mapping of Holocene deposits and main faults and lineaments in order to propose a deformation model. This model will be further validated by a geomorphological analysis and field data interpretations.

6.2. Large scale geometry

6.2.1. Holocene depocentrers

Holocene basins have been interpreted from a 90-m DEM (USGS, 2005). Criteria for this identification 120 Chapter 6 Fig. 6.3: (A) Map of relative uplift rates of the SSB fold-and-thrust range and its surrounding, Plio-Pleistocene to Recent deposits; (B) Cross-section of for location) (After (Suter et al., in review)). (see 6.3A the between Obando and La Victoria SSB passing Tectonics 121 have been the color contrast variation, erosion state and planar zone recognition. Six areas of Holocene depocenters have been identified: the Cauca Valley, Quindío-Risaralda, Amagá, Upper and Middle Magdalena Basins and the informally-named Ibagué Fan (Fig. 6.4).

The Quindío-Risaralda (Guarin, 2008) and Ibagué volcaniclastic fans present a subconical shape. They both originate from the Central Cordillera. Other basins are rhomboidal and elongated in a NNE direction, except for the Amagá Basin. Limits of the latter are not easily identifiable on the DEM image used and have been drawn on the base of geological maps (Gómez et al., 2007).

6.2.2. Main basin-bounding faults

The common geometrical features of these Holocene basins are their linearly-shaped boundaries and NNE trending preferential orientation. Furthermore, the southern and northern limits of respectively the Ibagué and Quindío-Risaralda volcaniclasic Fans are aligned on a NW-SE trending lineament, which crosscuts the Central Cordillera. This suggests that the basins described above are fault- bounded. Consequently, the main faults described in the literature have been plotted on the DEM (in red). This shows that the main faults orientated NNE, EENE and NW run along basins margins (Fig. 6.4).

However, because the northern limit of the Cauca Valley and Quindío-Risaralda Basins (i.e. the area of study) are not formally fault-constrained, special attention has been paid in this area to lineament observations on the DEM. Rotation of the DEM point of view has permitted the interpretation of an “en-échelon” fault system which crosscuts the Central and Western Cordilleras and bounds the Quindío-Risaralda and Ibagué Fans (Figs. 6.4 and 6.5). Because this fault system does not appear on any geological map, it has been formally named and mapped herein as Otún Fault System (OFS), with respect to the Otún lineament described in chapter 3.

Furthermore, the Romeral Fault System changes its strike towards the north from NNE to N-S up to NNNW, just south of the Amagá Basin (Fig. 6.6). This change coincides with a well-expressed lineament sub-parallel to the Ibagué fault, which offsetts and also rotates the western foothills of the Western Cordillera. This so-far unmapped structure has been named Istmina fault (ISF), with respect to the Istmina Deformation Zone described by (Duque-Caro, 1990; Taboada et al., 2000).

6.2.3. Lineament analysis

A topographical analysis has been carried out through the construction of a Digital Elevation Model (DEM) with Global Mapper 9 and Surfer 8 softwares, using radar photos with a 90-meter resolution

Fig. 6.5: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing the “en-échelon” Otún Fault mapped in this study. Letters a, b and c refer to fault segments forming the Otún Fault (see figure 6.11 for kinematics). 124 Chapter 6

(USGS, 2005). Because the sun angle affects the recognition of lineaments, six DEM have been constructed with a NW, NE, SW, SE, E and W light angle orientation. Some 1’300 lineaments have been identified (Fig. 6.6). Lineament recognition criteria such as geomorphological trends, anomalous orientation of drainage segments, triangular facets and distinct contrast differences have been used in this analysis.

For the sake of coherence with previous work carried out within the framework of this regional research project (Suter et al., 2008b), a similar fault nomenclature has been used, although the scale of the studied area is quite different: the N-S and NNE trending faults and lineaments have been grouped in the Romeral Fault System, the E-W to ENE trending ones in the Santa Rosa and Ibagué system, the ESE trending ones in the Salento and Ocaso system, and the NW trending ones in the Otún system.

A rose diagram has been calculated to examine lineament trends. Figure 6.7 shows the azimuthal distribution of the data. Principal lineament clusters in the area are trending N 130°-150° and N 90°-100°. The second group of dominant lineaments encompasses N 100°-110°, N 000°-010°, N 030°-040° and N 080°-090° strikes. Finally, minor clusters strike N 340°-000°, N 010°-020° and N 060°-080°.

By keeping only major lineaments and coloring them into families according to their strike, the following observations can be made (Fig. 6.8): (1) the ENE to EENE striking “Ibagué type” lineaments and (2) the NW “Otún type” (Suter et al., 2008b) are mainly represented north of 4° to 4.5° of latitude north, and much less south of this zone which corresponds to the Garrapatas and Ibagué faults; (3) the latter zone also corresponds to a major change in the strike of the Romeral Fault System, which trends NNE south of the zone and N to NNWN north of the zone; (4) NW striking lineaments seem to offset other structures.

6.2.4. Discussion

The mapping of Holocene deposits and main faults has highlighted the large-scale structural geometry around the studied area. This geometry has been simplified in figure 6.9. It shows that the Cauca Valley and Quindío-Risaralda Basins are clearly fault-bounded and therefore are tectonic basins, as well as the Amagá and Upper Magdalena Basins. Together with the lineament maps derived from the DEM (Figs. 6.6 and 6.8), this interpretation highlights a shear zone comprised between the Cucuana and Istmina faults, where the Romeral Fault System has been rotated. “Romeral-type” lineaments in figure 6.8 (black lines) situated north of (Fig. 6.10A), within (Fig. 6.10B) and south (Fig. 6.10C) of the shear zone (Fig 6.9) have been measured and plotted in rose diagrams where mean vectors have been calculated. Calculated angles indicate a clockwise rotation of the sheared block (and RFS) which varies between 9° and 16°.

A deformation ellipsoid formed by simple shear strain has been constructed (Fig. 6.11) ((Wilcox et

Fig. 6.6: 90-meter resolution DEM based on radar photographs (USGS, 2005) showing lineaments in the Central Cordillera and its surroundings, between 4°N and 7.4°N. Tectonics 125 126 Chapter 6

Fig. 6.7: Quantity-dependent rose- diagram illustrating the orientation of lineaments interpreted in figure 6.6. al., 1973; Sylvester and Smith, 1976)). The main faults bounding the Holocene basins have been added to the latter in order to estimate roughly the direction of their movements. It comes out that the RFS presents an inverse movement, the Ibagué fault is a dextral transtensive accident and the Otún Fault System is normal with a left-lateral component. These directions of movement have been also reported in figures 6.4, 6.8 and 6.9.

6.3. Morphometry

Morphometry is defined as the quantitative measurement of landscape shape (Keller and Pinter, 2002). As described in chapter 2, the SSB seems to be still in a phase of uplift (Suter et al., in review). No structural data about recent tectonic activity in the Western Cordillera have been found in the literature. In order to identify and quantify the level of neotectonic activity in the SSB, WC and North Cauca Valley, two parameters, or geomorphic indices, have been used: the drainage basin asymmetry and the mountain-front sinuosity.

The estimate of morphometric parameters (geomorphic indices) requires a high-resolution digital elevation model in order to obtain accurate results. The 30-meter DEM (USGS, 2005) available for the calculation of these parameters is not high-resolution, but accurate enough to provide a good estimation of tectonic activity.

6.3.1. Drainage Basin Asymmetry

The Drainage Basin Asymmetry can identify tectonic tilting at the scale of basins or large areas. It is sensitive to tilting perpendicular to the direction of the stream flow (Keller and Pinter, 2002). In the studied area, the main drainage is characterized by the N-NE and S-SW flows of respectively the Cauca and Risaralda meandering rivers. The latter is an affluent of the Cauca River, which it joins up at La Virginia. From the latter location, the Cauca River flows northwards and incises the cordilleras up to the northern lowlands of Colombia (Fig. 6.12). Because the course of the Cauca River is located Tectonics 127 Fig. 6.8: Summary of figure 6.6, where the principal lineaments affecting this area are grouped into families according to their strike. into families according grouped are the principal lineaments affecting this area 6.6, where Fig. 6.8: Summary of figure Arma Fault; GF: Garrapatas IF: Ibagué OF: Otún PF: Palestina SF: Salento SRF: Santa Rosa Fault. AF: Abbreviations: 128 Chapter 6 on the western part and that of the Risaralda River on the eastern part of their respective valleys, Fig. 6.9: Simplified map drawn from figure 6.4 showing main faults and Holocene depocenters (yellow). The hatched area represents the shear zone. (See figure 6.4 figure (See zone. shear the represents area hatched The (yellow). depocenters Holocene and faults main showing 6.4 figure from drawn map Simplified 6.9: Fig. for abbreviations). Tectonics 129

Fig. 6.10: Quantity-dependent rose-diagram illustrating the orientation of the RFS lineaments of figure 6.8, situated (A) north of, (B) within and (C) south of the shear zone (see figure 6.9). these two zones have been considered separately in order to calculate the asymmetry factor (AF) (see section 2.3).

The AF values are 71.5 for the Cauca Valley and 21.45 for the Risaralda Valley (in yellow and red respectively in figure 6.12 and Table 6.1). Considering that stream networks formed in stable settings should have AF values of about 50 (Keller and Pinter, 2002), the latter values suggest tilting towards the W in the Cauca Valley and towards the E in the Risaralda Valley. This abrupt change in tilting direction occurs in the la Virginia zone and also coincides with (1) the change in flow direction of the Cauca River from N-NE to E and (2) the beginning of its eroding regime. This confirms the existence of an important active tectonic lineament in the La Virginia area.

6.3.2. Mountain-Front Sinuosity

The Mountain-Front Sinuosity (Smf) is an index that reflects the balance between the tendency of erosion to produce an irregular or sinuous mountain front and active tectonics which tends to produce a relatively straight mountain front, because of an active range-bounding fault (Keller and

Fig. 6.11: (A) Simple shear associated with strike-slip faulting produces preferred orientation of faults, as well as different fault movements (after (Wilcox et al., 1973; Sylvester and Smith, 1976)). (B) Summary of the data obtained in this study. Grey areas represent extensional zones. Abbreviation: Ψ: shear angle. 130 Chapter 6

Pinter, 2002; see section 2.3). The four straight-line segments chosen in figure 6.13 correspond to the mountain fronts bordering the two valleys discussed above. Smf values comprised between 1 and 1.6 represent a mountain front associated with active, range bounding faults and values between 1.4 and 3 correspond to a mountain front with a lower, but still significant activity. Values greater than 3 indicate an inactive mountain front (Bull and McFadden, 1977; Keller and Pinter, 2002).

Results presented in Table 6.1 show that: (1) the eastern border of the Risaralda Valley (Fig 6.13, black lines) presents the highest rate of tectonic activity in the studied area, i.e. the mountain front is bounded by an active fault; (2) the western border of the Risaralda Valley (Fig. 6.13, yellow lines) is inactive to moderately active; (3) the western border of the Cauca Valley formed by the Western

A. Drainage Basin Asymmetry

Cauca Risaralda Ar At AF Ar At AF 697788 975993 71.495 14886 69446 21.435

B. Montain-Front Sinuosity

Risaralda East Risaralda West LmF Ls Smf LmF Ls Smf 1047.61 680.72 1.539 1362.77 650.05 2.096

Cauca East Cauca West LmF Ls Smf LmF Ls Smf 4910.41 2629.75 1.867 6079.12 3046 1.996

Table 1: Morphometric data calculated of the studied area.

Abbreviations: AF: drainage basin asymmetry; Ar: area of the basin to the right of the trunk stream that is facing downstream; At: total area of the drainage basin; LmF: length along the edge of the mountain-piedmont junction; Ls: straight-line length of the mountain front; SmF: mountain-front sinuosity.

Cordillera (Fig. 6.13, red lines) and its eastern border formed by the SSB foothills (Fig. 6.13, green lines) present a moderate tectonic activity.

6.4. Structural analysis

The structural results are plotted in a 30- and 90-meter DEM (USGS, 2005) (Fig. 14). They are presented below in three sub-areas marked by different types and trends of structures, i.e., the area A (section 6.4.1) and B (section 6.4.2) which bound the Cauca Depression, and the Plio-Pleistocene Tectonics 131 Fig. 6.12: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing the drainage basin asymmetry of the North Cauca Depression (yellow) and the Ψ: shear angle. Risaralda Basin (red).Abbreviation: 132 Chapter 6 Fig. 6.13: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing the mountain-front sinuosity (Smf) of the eastern border foothills of SSB the by formed the border eastern its and lines) Risaralda (red Valley Valley Cauca the of border western the lines), (yellow Valley Risaralda the of border western the lines), (black lines). (green Tectonics 133 134 Chapter 6

6.4.1. Faults in Cretaceous rocks

Fig. 6.17: Fault plane: s: slip fibers; f: crystallization fibers. The white arrow shows the motion of the missing compartment with respect to that in the picture.

6.4.1.1. Area A

The western boundary of the Cauca basin corresponds to the eastern foothills of the Western Cordillera. It is relatively rectilinear and strikes SSW-NNE. Structural data have been collected in basic volcanic rocks of the Upper Cretaceous Volcanic Formation (Kv) (Nivia et al., 1995), formerly

Fig. 6.18: Histogram representing the number of tensors from area A (see figure 6.15) belonging to stress regimes versus their respective ellipsoid form parameter Ф=(σ2−σ3)/(σ1−σ3). Tectonics 137 3 3 2 3 3 2 2 2 2 3 2 2 2 3 1 1 1 2 2-3 Quality 9 0.5 0.4 3.5 6.7 4.5 2.8 9.28 11.2 4.88 1.28 6.55 11.1 6.46 3.44 9.53 4.61 9.52 6.92 Var (°) Var 7 5 6 7 9 5 9 9 8 5 n 17 10 10 18 15 30 13 19 13 Transtension Transtension Transtension Transtension Transtension Stress regime Stress Transpression Transpression Pure extension Pure Pure extension Pure extension Pure Pure extension Pure extension Pure Radial extension Radial extension Radial extension Radial extension Pure compression Pure Radial compression Radial compression Ф 0.308 0.1543 0.2285 0.9488 0.8894 0.8161 0.9859 0.9863 0.0593 0.0339 0.4026 0.3755 0.4898 0.9373 0.0164 0.4966 0.7503 0.4379 0.9042 σ3 95 / 71 341 / 57 075 / 62 260 / 37 012 / 14 181 / 50 314 / 18 008 / 11 227 / 07 321 / 03 344 / 18 294 / 34 286 / 35 185 / 28 310 / 08 165 / 54 104 / 07 019 / 14 100 / 07 σ2 29 / 25 285 / 00 086 / 09 290 / 24 026 / 38 269 / 43 298 / 20 074 / 56 278 / 00 137 / 01 254 / 01 046 / 29 031 / 20 296 / 33 219 / 11 003 / 34 014 / 00 111 / 07 004 / 39 σ1 015 / 19 182 / 31 194 / 14 144 / 31 117 / 44 041 / 33 215 / 27 187 / 79 039 / 83 057 / 65 161 / 72 166 / 42 145 / 48 065 / 44 077 / 77 267 / 09 284 / 83 228 / 75 198 / 50 Kv/NB Kv/NB Kv/NB Kv/NB Kv/NB Kv/NB Kv/NB Kv/NB Kv/NB Kv/WB Kv/WB Kv/WB Kv/WB Kv/WB Kv/WB Kv/WB Kv/WB Tpv/NB Tpv/NB Fm/location Site name Site RN75 RN58-60-1-1 RN55-56-1 RN71-1-1 RN40-1 RN26 RN15-16-1 RN74 RN64-1-1 RN72-1-2 RN63-1 RN24-25-1 RN67 RN62-sans-0 RN61-2 RN65-1-1-1 RN69-1-1-1 RN68-1-1-1-2 RN3-10-1-1-1- 1-1-1 1 2 6 3 4 5 7 8 9 18 19 17 14 15 16 13 12 11 10 No Table 2: Parameters of the 19 tensors, calculated Table with stress the name of the site, the orientations of σ1, σ2 and σ3, ellipsoid the form corresponding parameter (Ф) the tensor belongs (°) to. indicates The the regime number average of misfitand faults anglethe used after stress for the calculation appears in the “n” column; Var result. A quality criterion between 1 (good) and 3 (bad) is given for the the final calculation. 138 Chapter 6

Fig. 6.19: Fault plane: s: slip fiber. The white arrow shows the undeterminable motion of the missing compartment with respect to that in the picture. named Diabasic Group (Kvo) (Caballero and Zapata, 1983).

Faults

Fault measurement sites are aligned along a SSW-NNE trending, some 40km long zone which corresponds to the trace of the inferred Cali-Patía fault mapped by (Nivia, 2001) from Roldanillo up to Ansermanuevo (Fig. 6.15).

The preferential fault orientation trends NNE at sites 3, 4, 5, 6, 7 and 8, EENE at site 1, and ENE and NW at site 2, with subvertical dips (Fig. 6.16).

Slikensides in the measured faults are well-preserved and coated by elongated slip fibers of chlorite (Fig. 6.17). The latter permit quite easily the determination of the movement direction. Most of the faults measured in the field present a normal and/or dextral direction of movement with high-dip angles (Fig. 6.16A and Table 6.2).

Paleostress inversions

Paleostress inversions are shown in figures 6.15A and 6.16. Except for the faults at site 5 which show a steep σ3 indicative of compressional strain (in this case pure compression, see figure 6.18), all the data present a steep σ1 indicative of extensional strain (Table 6.2 and Fig. 6.18). The latter strain is divided in three stress regimes with respect to the respective ellipsoid form parameter Ф [= (σ2−σ3)/ (σ1−σ3)]: radial extension (sites 1 and 4), pure extension (sites 2, 3 and 8) and transtension (sites 6 and 7) (method developped by (Champagnac et al., 2003)).

Maximal horizontal stresses for each site are shown as red lines in figure 6.15B. Maximum horizontal sigmas do not present significant dominant orientation. Tectonics 139

6.4.1.2. Area B

The area B corresponds to the area studied by Pahud (Pahud, 2009). It comprises a mountain range belonging to the Western Cordillera, which is delimited to the west by the Risaralda River and to

Fig. 6.20: 30-meter resolution DEM based on radar photographs (USGS, 2005) showing (A) distribution of the calculated palaeostress tensors and (B) distribution of maximum horizontal σ (in red) in area B. 140 Chapter 6 the east by the Cauca River. Structural data have been measured in basic volcanic rocks of the Upper Cretaceous Volcanic Formation (Kv) (Nivia et al., 1995), formerly named Diabasic Group (Kvo) (Caballero and Zapata, 1983), and in the Miocene porphyritic andesites of La Virginia (Tpv) (Caballero and Zapata, 2003).

Faults

Fig. 6.21: Histogram showing the number of tensors from area B (see figure 6.15) belonging to stress regimes versus their respective ellipsoid form parameter Ф=(σ2−σ3)/(σ1−σ3).

On the contrary to fault tectoglyphy described above in the western boundary, those encountered here are very badly preserved (Fig 6.19). The direction of fault movement has been difficult to determine and quality criteria are relatively bad. Consequently, results have to be considered with some caution. Most of the measured faults have a NE-trending strike and dip steeply towards the SE. Some other faults strike in a NW, WWNW and N-S direction with high-dip angles (Fig 6.16B).

Paleostress inversions

Paleostress inversions are presented in figures 6.16B and 6.20. They show a great diversity of strain, steep σ1, 2 and 3 being represented in a small area (Table 6.2 and Fig. 6.21). Tensors have also been drawn in different DEM, in order to have a better visualization (Fig. 6.20).

Sites 9 to 12 present a steep σ1 and this extensional strain is divided in two stress regimes: radial (sites 9 and 12) and pure extension (sites 10 and 11) (Fig. 6.20A). As fault orientation and dips vary a lot Tectonics 141 between sites, the maximal horizontal σ are inhomogeneous, showing no preferential orientation.

Sites 13 to 15 present a steep σ2 and the ellipsoid form parameter Ф indicates a transtensive stress regime for all the sites and the maximal horizontal σ are not homogeneous (Fig 6.20B and Table 2).

Sites 16 to 19 show a steep σ3 and this compressional strain is divided in two stress regimes: radial compression (sites 16 and 18) and transpression (sites 17 and 19) (Table 6.2). Maximal horizontal σ shows N-S to NNE-SSW orientations (Fig. 6.20C).

6.4.1.3. Interpretation

Normal faulting is surprising at the eastern boundary of the Western Cordillera because one would expect reverse faulting in an area dominated by compressive kinematics, with σ1 oriented WNW (Paris et al., 2000; Suter, 2008). The extensive pattern of these faults may be explained by the westward tilting of the Cauca basin due to sedimentary overload and active uplift of the SSB (Suter et al., 2008b). Furthermore, because site 5 presents a steep σ3 and faults are aligned along the Cali- Patía fault, normal reactivation of a segment of the latter seems plausible.

This hypothesis seems to be supported by the drainage asymmetry of the basin (see section 6.3.1), the strong volcanic activity during Plio-Pleistocene times and the strong erosion induced by the uplift of the Central and Western Cordilleras and the SSB (Thouret et al., 1990; Guarin et al., 2006; Neuwerth et al., 2006; Guarin, 2008; Suter et al., in review). Furthermore, the well-preserved state of 142 Chapter 6

Fig. 6.23: Quantity-dependent rose-diagram illustrating the orientation of the 513 faults and joints measured in the Zarzal Formation. the slickensides may indicate recent activity of this fault segment.

Datasets presented above do not show special trends. On the contrary, fault inversions show all stress regimes, thereby certainly reflecting the polyphased tectonic history of the Western Cordillera Complex (see chapter 3). Furthermore, the high rate of alteration affecting the rocks and slickensides could be explained by an active uplift of the area and the subsequent high erosion rate.

The interpretations presented above should be looked at with caution, for the following reasons: (A) out of the 74 sites analyzed in the field, only 19 have been kept because of the general bad outcrop preservation and the difficulty in determining the movement direction; (B) because the majority of the studied sites present a unimodal distribution of fault orientations (Fig. 6.16), the σ distribution is very risky and paleostress inversion results may be wrong; (C) the regional signal in the area A may be biassed by sediment loading. The variability of stress regimes (Fig. 6.22) in the studied area seems to confirm these hypotheses.

6.4.2. Zarzal Formation

Faults

Normal faulting

In the soft sediments of the Zarzal Formation, only extensive faulting has been encountered. Most displacements are small, mostly tens of centimeters and sometimes reaching a few meters. All faults Tectonics 143

Fig. 6.24: 30-meter resolution DEM (USGS, 2005) with location of the sites where conjugate normal faults planes were measured in the Plio-Pleistocene Zarzal Formation and Quindío-Risaralda volcaniclastic Fan. The black dots in the plots (Wulff stereonets, lower hemisphere) are projections of the fault plane poles. The dip of their mean vector indicating the direction of elongation (local σ3) is represented by the black asterisks. The dip and dip azimuth of mean vectors is shown besides each plot. The values given for the sites where only one single fault plane could be measured (numbers 4, 6, 10, 14, 17 and 18; black arrows on map) correspond to the dip and dip azimuth of the fault planes. The directions of local σ3 are symbolized on the DEM by arrows. Abbreviations: A: Ansermanuevo; C: Cartago; CF: Cartago Fan; O: Obando; QRF: Quindío- Risaralda Fan; SSB: Serranía de Santa Barbara; T: Toro; Vict: La Victoria; Vir: La Virginia; WC; Western Cordillera; Z: Zarzal. 144 Chapter 6 have a high angle and reach average dips of 68°. Measured faults present three preferential fault orientations, trending N to NNE, E to ENE and NW (Fig. 6.23A).

Projection of the fault plane poles has been carried out for each outcrop, as well as the mean vectors. The stereoplots, including those measured by (Pardo et al., 1994; Suter et al., 2008b) are summarized on a 30-meter DEM (USGS, 2005) (Fig. 6.24). Plots of the fracture plane poles show that faults belong to conjugate systems. In such a case, the mean vector projection is a good estimation of the local σ3 (Suter et al., 2008b). Although extension orientations do not present a preferential orientation, the mean value vector calculated from σ3 values shows a NW-SE direction (Fig 6.15).

Synsedimentary faulting

Only two synsedimentary normal faults have been encountered (Figs. 6.24, plots 1 and 18 and 6.25). They are covered by unfaulted beds (Fig. 6.25A) or characterized by the variable thickness of the faulted blocks (Fig. 6.25B) with displacements reaching five meters. (See chapter 4 for lithological and sedimentary descriptions).

Joints

Joints have been measured in each section studied in chapter IV. Orientations have been plotted in a rose diagram and show E-W to NW and NE dominant trends (Fig. 6.23B).

Interpretation

The high amount of liquefaction (see chapter 5), the large distribution of fault orientation and the Pleistocene seismic activity, as well as the regional compressive tectonic regime in the area (Guzmán et al., 1998; Gallego et al., 2005; López et al., 2005; López and Moreno, 2005; Suter et al., 2008a; Suter et al., in review) have led to relate this extensional superficial stress with lateral spreading (Audemard and De Santis, 1991; Gonzalez et al., 2004; Rastogi, 2004; Audemard et al., 2005; Suter, 2008).

However, because the Zarzal Formation deposits are strongly eroded and crop out only along the border of the basin, the statistical database is of low quality and this interpretation must be considered as tentative. In a similar way, joint orientations being widely distributed, interpretations are quite difficult. However, there is a good correlation with the main lineament sets observed in the 90-meter DEM presented above (Otún, Ibagué and RFS strike), pointing out the orientation relationship between faults and joints encountered in the Plio-Pleistocene soft sediments and large scale fault geometry. This observation illustrates the simultaneous activity of the Otún, Ibagué and RFS structures during Plio-Pleistocene time.

6.5. Discussion and conclusions

The analysis of basins and of fault and lineament geometry at a larger scale than the studied area demonstrates the existence of a shear zone constrained between the Istmina and Cucuana faults, leading to a clockwise block rotation of a maximum of 16°. This simple shear zone deflected the Romeral Fault System and created transtensional faults orientated NW-SE such as the Otún Fault Sytem. The normal movement of the latter created an uplift north of the studied area, which dammed the Cauca River Valley and thereby generated accommodation space for the deposition of the Zarzal Formation south of the fault (Fig. 6.26). It also changed the flow regime of the Cauca River north of La Virginia. Futhermore, the dextral Santa Rosa fault, an EENE structure parallel to the Ibagué and Cucuana faults seems to control the southern limit of the Cartago Fan (Fig. 6.26).

The NW-SE and EENE-WWSW fault geometry may also be responsible for the closure of the Amagá and Upper Magdalena Basins.

Furthermore, a Pliocene age is proposed for the reactivation of NW-trending faults, based on stratigraphical and geodynamical arguments. The age of the Zarzal Formation deposits has been dated geochronologically as Plio-Pleistocene (see chapter 4). On the other hand, a kinematic reconstruction of the study area proposed by (Suter et al., 2008b) indicates that the Cauca Valley Basin may have been formed by the northern closure of the valley following the indentation of the Chocó Panamá Block (CPB) and the eastward shift of the Western Cordillera. The indentation of the CPB is contemporary with the onset of the mayor Andean tectonic phase which began some 10.5 my ago and continued throughout Pliocene-Quaternary times (Cooper et al., 1995; Kellogg et al., 1995; Taboada et al., 2000; Cediel et al., 2003; Cortes et al., 2005). Because this collision occurred 10 my ago, the reactivation of NW-trending faults in the studied area during Plio-Pleistocene times seems 146 Chapter 6

Moreover, the structural analysis of the Zarzal Formation deposits demonstrates the simultaneous activity of the Otún, Ibagué and RFS structures during Plio-Pleistocene times, thereby confirming the strain partitioning proposed by (Suter et al., 2008b). This is also confirmed by the soft-sediment deformations encountered in these deposits, attesting to the important seismic activity during this period.

Finally, the Otún and Istmina Fault Systems have been formally named and mapped. They are thought to be of major importance for Plio-Pleistocene and Holocene sediment deposition. Both present an oblique geometry with a normal component, the Otún System being sinistral and the Istmina System dextral. Tectonics 147

References

Audemard, A. F. and De Santis, F., 1991. Survey of liquefaction structures induced by recent moderate earthquakes. Bulletin of the International Association of EngineeringGe ology (Boulder) 44, 5-16.

Audemard, F. A., Gomez, J. C., Tavera, H. J. and Orihuela, G. N., 2005. Soil liquefaction during the Arequipa Mw 8.4, June 23, 2001 earthquake, southern coastal Peru. Engineering Geology 78, 237- 255.

Caballero, A. and Zapata, G., 1983. Mapa geológico de Colombia, Plancha N°224 - Pereira. INGEOMINAS, Medellín.

Caballero, A. and Zapata, G., 2003. Mapa geológico de Colombia, Plancha N°224 - Pereira. INGEOMINAS, Medellín.

Champagnac, J. D., Sue, C., Delacou, B. and Burkhard, M., 2003. Brittle orogen-parallel extension in the internal zones of the Swiss Alps (South Valais). Eclogae geol. Helv. 96, 325-338.

Cortes, M., Angelier, J. and Colletta, B., 2005. Paleostress evolution of the Northern Andes (Eastern Cordillera of Colombia); implications on plate kinematics of the South Caribbean region. Tectonics 24(1).

Duque-Caro, H., 1990. The Choco Block in the northwestern corner of South America; structural, tectonostratigraphic, and paleogeographic implications. Journal of South American Earth Sciences 3(1), 71-84.

Gallego, A., Ospina, L. M. and Osorio, J., 2005. Sismo del Quindío del 25 de enero del 1999, 148 Chapter 6 evaluación morfotectónica y sismológica. Boletin de Geología, Universidad Industrial de Santander (UIS) 27, 133-150.

Gómez, J. G., Nivia, A. G., Montes, N. E., Jiménez, D. E., Sepúlveda, M. J., Gaona, T., Osorio, J. A., Diederix, H., Mora, M. and Velásquez, M. E., 2007. Atlas Geológico de Colombia, Plancha 5-08, Ingeominas.

Gonzalez, J., Schmitz, M., Audemard, F. A., Contreras, R., Mocquet, A., Delgado, J. and DeSantis, F., 2004. Site effects of the 1997 Cariaco, Venezuela earthquake. Engineering Geology (Boulder) 72, 143-177.

Neuwerth, R., Suter, F., Guzman, C. A. and Gorin, G. E., 2006. Soft-sediment deformation in a Tectonics 149 tectonically active area: The Plio-Pleistocene Zarzal Formation in the Cauca Valley (Western Colombia). Sedimentary Geology 186(1-2), 67-88.

Nivia, A., 2001. Mapa geológico del departamento del Valle del Cauca, memoria explicativa

INGEOMINAS, Colombia.

Nivia, A., Galvis, N. and Maya, M., 1995. Mapa geológico de Colombia, plancha 242 - Zarzal. Cali.

Pahud, O., 2009. Etude sédimentologique et morphotectonique dans le Quaternaire du Valle del Cauca (Colombie). M. Sc. thesis, Université de Genève, Geneva, Switzerland.

Rastogi, B. K., 2004. Damage due to the M (sub w) 7.7 Kutch, India earthquake of 2001. Tectonophysics 390(1-4), 85-103.

Suter, F., 2008. Tectono-Sedimentary Study of the Interandean North Cauca Valley Basin, Central Western Colombia. Terre et Environnement (Université de Genève), 78, 145 p.

Suter, F., Neuwerth, R., Gorin, G. E. and Guzman, C., 2008a. Depositional model of (Plio-)Pleistocene sediments in a tectonically active zone of Central Colombia. Geologica Acta 6(2), 1-19.

Sylvester, A. G. and Smith, R. R., 1976. Tectonic transpression and basement-controlled deformation in the San Andreas fault zone, Salton Trough, California. Am. Assoc. Petrol. Geol. Bull. 60 2081- 2102.

Thouret, J. C., Van Der Hammen, T., Salomons, B. and Juvigné, E., 1990. Late quaternary geology, stratigraphy and palaeoecology of the last glaciation in the colombian central Cordillera (Ruiz-Tolima massif). Symposium International “Géodynamique Andine”, Grenoble (FRA), pp. 399-402.

USGS. (2005). “http://edc.usgs.gov/products/elevation.html.”

Wilcox, R. E., Harding, T. P. and Seely, D. R., 1973. Basic Wrench Tectonics AAPG Bulletin 57(1), 74-96. CHAPTER VII

Co n c l u s i o n s

Ralph Neuwerth 152 Chapter 7

This study is the final contribution of a major investigation project that also includes the works of Ospina (2007), Guarin (2008), Suter (2008) and Pahud (2009), which have considerably enhanced the geological knowledge of the North Cauca Valley Basin and its surroundings. The major conclusions of the research presented here can be summarized as follows.

7.1. Sedimentary infill of Andean intramountane basins: case history of Plio-Pleistocene deposits in the North Cauca Valley (Colombia) (chapter 4)

Field investigations in the North Cauca interandean Basin have shown that many important sedimentary facies in the late Tertiary infill have not been reported. They have illustrated the relationships between the different stratigraphic units. This has led to the redefinition of the uppermost Tertiary Zarzal Formation on the basis of the different sedimentary environments encountered: (a) fluvio-lacustrine, (b) mass-flows, (c) Gilbert-type delta and (d) alluvial fans.

Consequently, a new lithostratigraphical group called the Santa Barbara Group has been proposed. It includes the redefined Zarzal and the new Quindío-Risaralda formations. Three members have been distinguished within the Zarzal Fm (the Obando, Ansermanuevo and Holguín Members), based on their depositional environment and the origin of their lithological constituents. According to palynological and geochronological 40Ar/39Ar datings, the outcrops of the Santa Barbara Group have a Plio-Pleistocene age.

These results illustrate the importance of undertaking such detailed sedimentological investigations in other intramountane Andean basins where similar types of deposits have been described, as for example the Amagá, Patía, Upper and Middle Magdalena Basins.

7.2. Soft-sediment deformations in a tectonically active area: The Plio-Pleistocene Zarzal Formation in the Cauca Valley (Western Colombia) (chapter 5)

Sands, silts and clays in the Plio-Pleistocene Zarzal Formation present intervals with numerous soft-sediment deformations interbedded with undeformed strata. Deformation structures have been classified into four groups: load structures (load casts, flame structures and pseudonodules), water escape structures (water escape cusps, dish and pillar, pocket and pillar), soft-sediment intrusions (clastic sills and dykes) and other structures (disturbed laminites, convolute laminations, slumps and synsedimentary faults).

Field evidence and regional geological criteria indicate that seismicity is the most probable triggering mechanisms for the deformations. The latter have been consequently interpreted as seismites. Because this disturbed intervals have a large areal extent and are intercalated within undeformed strata, this Conclusions 153 confirms the synsedimentary character of these events. It also testifies to a significant seismic activity contemporaneous to the Zarzal Formation deposition.

7.3. Tectonics (chapter 6)

The analysis of basins and of the geometry of faults and lineaments at a larger scale than the studied area demonstrates the existence of a shear zone constrained between the Istmina and Cucuana faults, leading to a clockwise block rotation of a maximum of 16°. This simple shear deflected the Romeral Fault System and created also transtensional faults oriented NW-SE such as the Otún Fault Sytem. The normal movement of the latter created an uplift north of the studied area which dammed the Cauca River Valley and thereby generated accommodation space for the deposition of the Zarzal Formation south of this fault. It also changed the flow regime of the Cauca River north of La Virginia. The same mechanism seems responsible for the closure of the Amagá and Upper Magdalena Basins.

Moreover, the structural analysis of the Zarzal Formation deposits demonstrates the simultaneous activity of the Otún, Ibagué and RFS structures during Plio-Pleistocene time. This is also confirmed by the soft-sediment deformations encountered in these deposits (chapter 5), attesting to the important seismic activity during this period.

7.4. Cinematic reconstruction

The data presented above, together with results of Guarin (2008) and Suter (2008) have permitted to reconstruct a simple cinematic evolution model of the North Cauca Valley (fig. 7.1).

During the Late Oligocene to Early Miocene, the Farallon Plate was split into the Cocos Plate in the north and the Nazca Plate in the south (Hey, 1977; Pennington, 1981; Taboada et al., 2000; Lonsdale, 2005). The Cocos Plate moved towards the NNE and the Nazca Plate towards the E. This created both a change from oblique to orthogonal and an increase in convergence between the Nazca Plate and the continental margin (Fig. 7.1A). This event coincided with the initiation of the Ibague Fault system and other ENE trending right-lateral strike-slip fault systems within the Northern Andes (Acosta et al., 2007). It had various tectonic consequences in Colombia (see chapter 3), one of which being an increase in the Cordillera uplift rate and, consequently, the Cauca Valley development and the Cartago Formation deposition (McCourt et al., 1984; Ramos and Aleman, 2000).

Folding in the Cartago Fm started in Late Oligocene-Early Miocene times (Keith et al., 1988; Ríos and Aránzazu, 1989) and the syn-kinematic La Paila Fm was subsequently uncomformably deposited on this incipient fold belt. It corresponds to the collision during the Middle Miocene of the Chocó- Panamá Block (CPB) into NW South America in an E to ESE direction (Pennington, 1981; Restrepo and Toussaint, 1988; Duque-Caro, 1990; Mann and Corrigan, 1990; Van der Hilst and Mann, 1994; Kellogg et al., 1995; Taboada et al., 2000; Trenkamp et al., 2002).

The folding and thrusting of the SSB was probably enhanced since Mio-Pliocene times and continued 154 Chapter 7

Fig. 7.1: Oligocene to present-day, schematic, simplified reconstruction of faults in the studied area. (A) Representation of the Romeral Fault System before initiation of the Ibague Fault System and other ENE trending right-lateral strike-slip fault systems. (B) Clockwise block rotation induced by EENE dextral strike-slip faulting. The latter induced initialization of NNE elongation, NW transtensional faulting and Plio-Pleistocene deposition (hatched area). (C) Present day geometry showing active, faulting, NNE elongation, shear zone and local subsidence and Holocene deposits (yellow). Conclusions 155 throughout the deposition of the La Paila Formation. The angular unconformity between the La Paila Fm and the overlying Zarzal and Quindío-Risaralda volcaniclastic deposits may record this phase of enhanced shortening, induced by the Chocó-Panamá Block collision into northwestern South America (Suter, 2008).

According to palynological and Ar40/Ar39 datings (chapter 4), deposition of the Santa Barbara Group occurred during Plio-Pleistocene times. Accommodation space elongated in a NNE direction may have been induced by SW-NW transtensional faults created by the onset of clockwise block rotation (Fig. 7.1B). This hypothesis has been also discussed by Guarin (2008) . Morever, structural analysis in Plio-Plesitocene deposits shows that NNE- (Romeral Fault System), EENE- (Ibagué Fault System) and NW-trending (Otún Fault System) faulting were simultaneously active at this time.

At present day, the northern part of the studied area (chaper 6), as well as the Serranía of Santa Barbara (Suter, 2008), are still in uplift, leading to the actual infill phase of the Cauca Valley Basin and its subsequent subsidence (Fig. 7.1C).

7.5. Perspectives

Because this sedimentological study of Plio-Pleistocene deposits in the interandean North Cauca Basin has been carried out in surface outcrops, it has not been possible to precisely determine the geometry and thickness of the defined lithostratigraphical units. Seismic data are needed to refined lithostratigraphy and propose a facies model. Moreover, the same detailed research should be applied to other interandean basins of Colombia, such as the Amaga, Magdalena or Cauca-Patia basins.

More Ar40/Ar39 datings are needed to refine the stratigraphy of the North Cauca Basin and to have a better statistical approach.

Further structural data are needed in the eastern foothills of the Western Cordillera, where field data have not been recovered, because of the high level of insecurity.

Because the deformation model described in chapter 6 has been derived from DEM analysis, it is advisable to acquire structural field data, especially around the Otún, Cucuana and Istmina Faults.

Finally, better datings of the Cartago and La Paila Formations would contribute to refine cinematic evolution models. 156 Chapter 7

References

Hey, R., 1977. Tectonic evolution of the Cocos-Nazca spreading center. Geological Society of America Bulletin 88, 1404-1420.

Keith, J. F., Rine, J. M. and Sacks, P. E., 1988. Frontier Basins of Colombia, Valle del Cauca, University of South Carolina, Earth Sciences and Resources Institute, Field report: 267 p.

Lonsdale, P., 2005. Creation of the Cocos and Nazca plates by fission of the Farallon plate. Tectonophysics 404(237-264).

Mann, P. and Corrigan, J., 1990. Model for late Neogene deformation in Panama. Geology Boulder 18(6), 558-562.

McCourt, W. J., Aspden, J. A. and Brook, M., 1984. New geological and geochronological data from the Colombian Andes; continental growth by multiple accretion. Journal of the Geological Society of London 141(5), 831-845. Conclusions 157

Ospina, L. M., 2007. Morphotectonique des dépôts quaternaires dans la région de Calarcá, Quindío (Colombie centrale). M. Sc. thesis, Université de Genève, Geneva, Switzerland, 96p.

Pahud, O., 2009. Etude sédimentologique et morphotectonique dans le Quaternaire du Valle del Cauca (Colombie). M. Sc. thesis, Université de Genève, Geneva, Switzerland.

Pennington, W. D., 1981. Subduction of the eastern Panama Basin and seismotectonics of northwestern South America. JGR. Journal of Geophysical Research. B 86(11), 10753-10770.

Ramos, V. A. and Aleman, A., 2000. Tectonic evolution of the Andes

Northern Andes. Tectonic evolution of South America. G. Cordani Umberto, J. Milani Edison and A. Thomaz Filho. In In-Folo Producao Editorial, Grafica e Programacao Visual. Rio de Janeiro, Brazil. 2000.

Restrepo, J. J. and Toussaint, J. F., 1988. Terranes and continental accretion in the Colombian Andes. Episodes 11(3), 189-193.

Suter, F., 2008. Tectono-Sedimentary Study of the Interandean North Cauca Valley Basin, Central Western Colombia. Terre et Environnement (Université de Genève), 78, 145 p.

Van der Hilst, R. and Mann, P., 1994. Tectonic implications of tomographic images of subducted lithosphere beneath northwestern South America. Geology (Boulder) 22(5), 451-454.