cenozoic tectonic evolution of the iberian peninsula

VRIJE UNIVERSITEIT

CENOZOIC TECTONIC EVOLUTION OF THE IBERIAN PENINSULA: EFFECTS AND CAUSES OF CHANGING STRESS FIELDS

ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Vrije Universiteit Amsterdam, op gezag van de rector magnificus prof.dr. T. Sminia, in het openbaar te verdedigen ten overstaan van de promotiecommissie van de faculteit der Aard- en Levenswetenschappen op donderdag 11 april 2002 om 15.45 uur in de aula van de universiteit, De Boelelaan 1105

door

Bernd Andeweg geboren te Heemskerk

promotor:

prof.dr. S.A.P.L. Cloetingh

“Only after the last fish has been caught and the last tree has been cut, man will discover he cannot eat money” Indian expression

The research reported in this thesis was carried out at the Tectonics / Structural Geology Department Faculty of Earth Sciences Vrije Universiteit De Boelelaan 1085 1081 HV Amsterdam The Netherlands y Departamento de Geodinámica Facultad de Ciencias Geológicas Universidad Complutense de Madrid 28070 Madrid Spain Financial support was provided by the Netherlands Foundation of Scientific Research (NWO) through the project “Interplay of tectonic and surface processes: vertical and horizontal motions in the Iberian Peninsula. Cenozoic tectonic evolution of the Iberian Peninsula: causes and effects of changing stress fields” (750.295.02A) Netherlands Research School of Sedimentary Geology (NSG) publication no. 20020101

ISBN 90-9015593-7 © B. Andeweg, 2002 The complete text and full color figures of this thesis will be available on the internet address: http://www.geo.vu.nl/~andb/iberia

Reading committee: Dr. Kees Biermann (Vrije Universiteit, Amsterdam) Dr. Fred Beekman (Vrije Universiteit, Amsterdam) Prof. Seth Stein (Northwestern University, Evanston, USA) Prof. Ramón Vegas (Universidad Complutense, Madrid, Spain) Dr. Gerardo de Vicente (Universidad Complutense, Madrid, Spain) Prof. Peter Ziegler (University of Basel, Switzerland)

CONTENTS Acknowledgements ................................................................................................................ I Summary.............................................................................................................................. III Samenvatting (summary in Dutch)......................................................................................... V Chapter 1 - Introduction to the tectonic setting of the Iberian Peninsula General geological evolution .............................................................................................. 1 Chapter 2 - Concepts and methodology of stress analysis 2.1 Stress in the Earth’s crust ............................................................................................ 5 2.2 Paleostress method ..................................................................................................... 7 2.2.1 Fault inversion method .......................................................................................... 7 2.2.2 Striations and solution pits on pebbles................................................................. 11 2.2.3 Regional context and (paleo)stress trajectories in Iberia...................................... 13 2.2.4 Conclusion........................................................................................................... 20 2.3 Finite element modelling of intraplate stress............................................................... 20 2.4 Stresses induced by lateral density variations ............................................................ 23 2.4.1 Concept and calculation ...................................................................................... 24 2.4.2 Testing the model-sensitivity for various input parameters................................... 28 Chapter 3 - New structural data from field studies in Northwestern and Central Iberia 3.1 New structural and kinematic indicator data for NW Iberia ......................................... 39 Bierzo Basin ................................................................................................................. 41 Monforte and Sarria Basins .......................................................................................... 51 N. Duero Basin ............................................................................................................. 52 Cantabrian/Asturian coast ............................................................................................ 53 Implications of the new data for the evolution of the stress field in NW Iberia ............... 57 3.2 Tectonic activity of the Spanish Central System during the Paleogene evidenced by structural and sedimentary geology, and apatite fission track analysis .................... 58 Introduction to the study area ....................................................................................... 58 Observations in the study area ..................................................................................... 68 Discussion and new model for the tectonic evolution of the SCS.................................. 79 Chapter 4 - Paleogene geological evolution of the Iberian Peninsula and western Mediterranean Introduction ...................................................................................................................... 83 65 Ma, KT-boundary ........................................................................................................ 87 54 Ma, L. Paleocene - E. Eocene (Ypresian) ................................................................... 88 42 Ma, M. Eocene (L. Lutetian -E. Bartonian) .................................................................. 91 36 Ma, L. Eocene (Priabonian)......................................................................................... 94 30 Ma, M. Oligocene (Rupelian - Chattian) ...................................................................... 96 27 Ma, L. Oligocene (Chattian) ...................................................................................... 114 24 Ma, L. Oligocene - E. Miocene (L. Chattian - E. Aquitanian)...................................... 116

Chapter 5 - Neogene geological evolution of the Iberian Peninsula and western Mediterranean 21 Ma, E. Miocene (L. Aquitanian - E. Burdigalian) ........................................................ 119 18 Ma, E. Miocene (L. Burdigalian) ................................................................................ 121 15 Ma, M. Miocene (L. Langhian - E. Serravallian)......................................................... 123 12 Ma, M. Miocene (L. Serravallian)............................................................................... 124 9 Ma, L. Miocene (Tortonian) ......................................................................................... 126 6 Ma, L. Miocene - E. Pliocene (Messinian - Zanclean).................................................. 128 3 Ma, L. Pliocene ........................................................................................................... 130 0 Ma, Holocene.............................................................................................................. 131 Chapter 6 - Finite element modelling of Cenozoic stress fields in the Iberian Peninsula 6.1 General model description ....................................................................................... 135 6.2 Present-day stress field:........................................................................................... 136 Observations, model geometry and boundary conditions............................................ 136 Model results .............................................................................................................. 141 Incorporating stresses induced by lateral density variations ....................................... 142 6.3 Middle Miocene (~12Ma) stress field........................................................................ 142 Observations, model geometry and boundary conditions............................................ 142 Model results .............................................................................................................. 143 Incorporating stresses induced by lateral density variations ....................................... 146 6.4 L. Oligocene (~24Ma) stress field............................................................................. 146 Observations, model geometry and boundary conditions............................................ 146 Model results .............................................................................................................. 147 Incorporating stresses induced by lateral density variations ....................................... 149 6.5 L. Eocene (~36Ma) stress field................................................................................. 150 Observations, model geometry and boundary conditions............................................ 150 Model results .............................................................................................................. 150 Incorporating stresses induced by lateral density variations ....................................... 152 6.6 L. Paleocene - E. Eocene (~54Ma) stress field......................................................... 153 Observations, model geometry and boundary conditions............................................ 153 Model results .............................................................................................................. 153 Incorporating stresses induced by lateral density variations ....................................... 155 6.7 General results and conclusions .............................................................................. 155 Chapter 7 - Synthesis ........................................................................................................ 157 Appendix A ........................................................................................................................ 160 Ridge Push as a line force.......................................................................................... 160 Ridge Push as intraplate forces .................................................................................. 162 References ........................................................................................................................ 163

ACKNOWLEDGEMENTS Completing this thesis would have been hard, if not impossible, without the support of many. I would like to thank all of them for their different way of support in –slowly, but surely- getting my thesis to the printer. I would like to mention several persons more specific, fully aware of inevitably forgetting someone... My promotor, Sierd Cloetingh, receives all credits for creating the right boundary conditions to successfully finish an intriguing PhD project on neo-tectonics. The department of Tectonics and the international environment through co-operation always were extremely inspiring. Sierd, thank you for your continuous belief in me finishing this project! Thanks are due to the members of the reading committee that spent their precious time in improving my draft. Dr. Kees Biermann (Vrije Universiteit), Dr. Gerardo de Vicente (Universidad Complutense Madrid), Prof. Seth Stein (University of Stanford), Dr. Fred Beekman (Vrije Universiteit) and Prof. Ramon Vegas (Universidad Complutense Madrid) are kindly thanked for their effort. A special word in this respect to Prof. Peter Ziegler (University of Basel): your detailed remarks were highly appreciated! Dr. David Coblentz is acknowledged for kindly sharing his code for potential energy calculations. Dr. Gerardo de Vicente and Dr. Alfonso Muñoz Martín are thanked for our joint work on the neo-tectonics of central Iberia. Without them, this project would have been only half as exciting and efficient! With regard to daily guidance I would like to mention a few people that were always helpful in discussions or any science related problem: Fred Beekman (thesis number 25 in which he is addressed in this way?!), Marlies ter Voorde (always there to make sense of the things I put on paper!), Kees Biermann, Anco Lankreijer. Prof. Peter Ziegler is thanked for a few very fruitful ‘dentist visits’. Fred Beekman and Matthias Gölke are thanked for our joint struggle in modelling faults in 3D in ANSYS. Gábor Bada is thanked for working together on science, and having a lot of fun in the mean time (the turning-the-chair record is still in my hands!). Köszönöm, Judit as well!. Both Gábor and my former roommate Ernst Willingshofer are thanked for our joint effort in trying to understand ‘extensional collapse’. Ernst is especially acknowledged, and not only for his patience in answering the phone for me over and over again. I deeply admire how he kept the good spirit alive even in less fortunate periods over the last years. Ernst, Antoinette, Carmen and Robin: all the best in the years to come! The members of the ‘Iberia-project’: Gert-Jan Weltje, Marlies ter Voorde, Wout Nijman (thanks for your enthusiasm and endurance in the field!) and Karen de Bruijne. With Karen (Catharina la Bruja), the contact was intense and very nice (‘vino blanco muy, muy seco’) during our joint field campaigns. It worked out pretty well together! I was privileged to work on a project that included numerous stays and fieldworks in Spain. Gracias a la gente de los pueblos en los que trabajamos Cogolludo, Ponferrada, San Vicente de la Barquera …¡Volveré algún día! ¡De Madrid al cielo! Gracias al grupo tectónico de la Universidad Complutense de Madrid. Gerardo, Alfonso (Poncho, ¡Aupa Atleti!), Miguel, Geles, Jorge (“Ghello Dutch people!”), Silvia, Pilar (gracias por la hospitalidad), Raul, Carmen & Carmelita, Meaza. Sin vosotros nunca hubieran sido tan interesantes mis visitas a Madrid. Agredezco a Jorge, Miguel, Gerardo, Jose Pedro Calvo y Jose Manuel Casado el haberme dejado disfrutar de su extenso conocimiento de la geología de España central. Además quiero dar las gracias a Miguel por el muy agradable trabajo de campo en Asturias (¿Beberemos otra sidra?). I would like to thank Harm Rondeel, Kees Biermann and Anne Fortuin for offering me the opportunity to assist the first year fieldwork in Spain (4 times!). All of the students: thank you for the nice days in the field. All of the assistants are thanked for the fine nights and the short holiday

I

breaks to the Embalse de Camarillas, the playa San Juan or Vilajoyosa. A special word of thank to Frank & Karen (Fuente Alamo, 1996), Thijs (‘sepia frita?’) (Hellin, 1997), Fred (‘beef special’) (Jumilla, 1998), Bart (‘Mercatorplein scoort’!), Ron and Margo (1999). And, por supuesto, Pepe and family from Meson El Alamo! The entire ancient and new E1 corridor (now ‘expanded’ to C2), with a special word to Marlies, Reini, Luis (Lucho), Harry, Tore (‘What’s that in English? Well, you know what I mean’), Taco, Matthias (witbiertje), JD, Gerco (Vissen Is Plezant), Jorge (¡Gracias! Pude practicar mi ‘español’), Stephanie, Jeroen, Sevgi, several generations of NSG-students, Kaptein Dick (& Nora) and the many foreign guests. Anco (‘biertje?’) is thanked for his support in many ways. There are several ‘groups’ of people that regularly brought some distraction in life: VUCK (Vrije Universiteit Colleague Kartclub) for some extra speed, de Ouwe Manne (indoor soccer, late night) for some fresh injuries, de Tour-directie for the biggest mental exercises and nervous afternoons in July, een ‘potje’ (binnen of in het Vondelpark). Andor (‘Hot Lips’), Erwin, Guido, Remke, Frank, Elmer, Ron (‘potje biljarten?’), Yvette, Jurgen. The Geoscoop editors, especially Michel (stay angry!), Marleen and Kees. Ceciel, thank you for your advice and tips on (not!) working with computers! GeoVUsie, thanks for all of the nice events. Donderdorst, Elfkroegentocht, VIP4life, whatever! Laurens en Aline (de Baarsjes express), Anouk (‘zeg, eh, ben jij op de fiets?’), and everyone else! Studenten, medewerkers, Onderwijsvoorlichting, Jaap, Herman, Mascha, Saskia, iedereen, die van mijn nieuwe werk een superbaan maakt. Life outside the faculty was –like it should be- more than worthwhile! Thanks to the frequent (or less frequent) diners, evenings (nights, or days) playing games, exciting trips, just drinking, or celebrating whatever: Anne & Sander (‘portje kir?’), Ilja & Daan, Mirja & Marcel, Malika & Joost, Bart & Christine, Elien & Jef (oliebollen!). Erin and Daphne, thanks for holding on inviting me over for diner! Anouk (Rein & Toke), Erik & Li, Matthijs! AV Sagitta (Olympiaplein), Phanos nowadays (in the Olympic Stadium), for the physical exercise on the runway track and Rayan for his help on getting me back on track (and behind the computer!!). Ali & Gerard, Hanneke & Serge, Marjolijn & Hugo: dank voor de gezellige weekendjes in Zeeland, etentjes in Amersfoort, Gouda en picknicks in de parken van Amsterdam-West. Mijn ouders, Bernice & Gijsbert en Celedonio, voor wie ik ben geworden! Bernice: bedankt voor je vakkundige hulp bij het maken van het boekje! En Annemieke, gelukkig kunnen we over mijn huidige werk wel praten…. Wat zouden we ook weer allemaal doen ‘na mijn promotie’? Zou het nóg leuker kunnen worden samen?

II

SUMMARY This thesis contains the results of the four years of research I carried out at the department of Structural geology/Tectonics of the Faculty of Earth Sciences at the Vrije Universiteit Amsterdam. The research forms part of the multi-disciplinary NWO program “Interplay of tectonic and surface processes: vertical and horizontal motions in the Iberian Peninsula”. The four components within this program are focusing on different aspects of this theme: (a) vertical motions using fission track analysis, (b) landscape evolution, (c) coupled erosion-sedimentation modelling in 3D and (d) tectonic modelling combined with structural geological fieldwork. The Iberian Peninsula (Portugal and Spain) has been chosen for this study because of the vast amount of available data, perfect outcrop conditions and interesting tectonic setting during the Cenozoic. The component that is described in this thesis, entitled “Cenozoic tectonic evolution of the Iberian Peninsula: causes and effects of changing stress fields” forms the tectonic/structural geological framework for the other components. As is evident from the title of this thesis, the several topics that will be addressed are: the Cenozoic tectonic evolution of the Iberian Peninsula, the effects of changing stress fields (deformation as detected by field observations) and the causes of these changes in stress field. Many regional studies have been published about the Cenozoic geology of the Iberian Peninsula. A detailed outline of the geological evolution of the entire Peninsula, however, was not yet available. Therefore, this thesis presents a compilation of the available data for the reconstruction of the tectonic evolution of the Iberian Peninsula and the western Mediterranean. Because the present-day setting of the Iberian Peninsula was essentially structured by reactivation of inherited crustal features during Tertiary times, this reconstruction has been made for the period of 65 Ma until presentday. This compilation shows low temporal and/or spatial resolution of data concerning the tectonics and the (paleo)stress field for some regions in the Iberian Peninsula. Several of these voids in knowledge have been addressed by carrying out combined structural and sedimentological fieldwork to provide new constraints on the stress field evolution. The results of these regional studies have been incorporated in the reconstruction of the entire Peninsula and will be presented in this thesis. In this way, my research provided the other components of the NWO program with information on the temporal and spatial development of the stress field and activity of larger scale tectonic features. Based on the paleo-geological reconstruction, finite element models of the Iberian Peninsula have been constructed for several time slices during the Tertiary in order to predict the stress field of that time and test different hypotheses on activity along the plate boundaries. Both the geometry and the boundary conditions have been derived directly from the paleo-geological reconstruction. The present-day situation provides the best constraints on plate (boundary) geometry, relative motion of the different plates (Africa, Eurasia) and tectonic activity deduced from seismicity. But even for this case, some of the parameters used in the modelling are a matter of debate, such as for example, the exact location of the plate boundary between Africa and the Iberian Peninsula. However, if for the present-day stress field an acceptable regional result can be obtained, then this would provide more confidence in the results obtained for periods for which the constraints are less well documented.

III

The first order numerical modelling of the present-day stress field in the Iberian Peninsula and surrounding regions uses a finite-element model of a 100km thick lithosphere with forces applied to the borders. Since it has been shown that the effect of lateral density variations (e.g. continental crust next to air in the case of a mountain range) upon the stress field is significant, this effect has been incorporated subsequently. This approach for the reconstruction of Tertiary stress fields requires a first-order estimate of paleotopography. In this way, the other components of this NWOprogram dealing with vertical motions offered feedback to the tectonic modelling. Chapter 1 is a brief introduction to the general geological evolution of the Iberian Peninsula and the western Mediterranean region. Chapter 2 offers an introduction to theoretical backgrounds of basic concepts used in the research such as (origin of) stresses in the Earth’s crust, deformation, stress inversion techniques, finite element modelling, and the concept of stress induced by lateral density variations. Extra attention will be paid to the latter concept, and the chapter contains results of sensitivity tests I performed on several basic assumptions. Chapter 3 presents structural geological data that has been gathered to fill some gaps in knowledge about the tectonic evolution of Iberia in time and space. New data is presented on the tectonic activity of the Spanish Central System during the Paleogene and for the NW of the Peninsula (Bierzo Basin and Asturian coast). In Chapter 4, a compilation is presented of the available data on the tectonic and geologic evolution of the Iberian Peninsula during the Paleogene (65-23Ma). Included are the stress field, active deformation structures, (first-order estimates on) paleotopography and sedimentary environments. This combination of data is presented in maps depicting different time-intervals, accompanied by a reference text. Chapter 5 is similar in configuration to Chapter 4, but dealing with the Neogene to present-day reconstruction (23-0Ma). Chapter 6 presents the results of numerical models on the evolution of the stress field in the Iberian Peninsula during the Cenozoic. For time slices for which stress field and paleotopography could be derived with enough confidence, model geometry and boundary conditions have been derived from the reconstructions in Chapter 4 and 5. The concept of stresses induced by lateral density variations is included in the modelling. The final chapter, Chapter 7, forms a synthesis and discusses the constraints this study provides for the tectonic evolution of the Iberian Peninsula, the implications of the model results, and possible future directions for research on this topic.

IV

SAMENVATTING (SUMMARY IN DUTCH) Mijn onderzoek heeft zich gericht op de oorzaken en gevolgen van spanningen in de aardkorst. In grote lijnen is bekend dat de oorzaken van deze spanningen gezocht moeten worden in de platentektoniek. De korst van de aarde bestaat uit verschillende ‘scherven’ (platen) die bewegen ten opzichte van elkaar. Op bepaalde plekken is die beweging uit elkaar (mid oceanische ruggen) en onvermijdelijk (ervan uitgaande dat de Aarde niet groter wordt) op andere plekken naar elkaar toe (subductie zones of gebergten). Als gevolg van krachten die samenhangen met botsing, langs elkaar bewegen of uit elkaar drijven van de platen ontstaan er spanningen in de korst. Afhankelijk van welke processen actief zijn langs de plaatranden van een bepaalde plaat, kan het spanningsveld binnen de plaat ingewikkelde tussenvormen van compressie (druk) en tensie (rek) vertonen. Zwaktezones die in de korst voorkomen, kunnen door deze spanningen in beweging komen. Dat gaat niet geleidelijk, maar met schokken (aardbevingen). Om een beter begrip te krijgen van de relatie tussen spanningen en breukbeweging is het dus erg belangrijk te weten wat voor spanningsveld in de bestudeerde plaat aanwezig is. Het gebied waarin het onderzoek is uitgevoerd is het Iberische Schiereiland (Spanje en Portugal). Dit gebied is onder andere geselecteerd omdat hier al veel waarnemingen van het spanningsveld beschikbaar waren. Daarbij komt dat bekend was dat er een opeenvolging van verschillende fases van vervorming heeft plaatsgevonden in de ‘jonge’ geologische ontwikkeling van het Iberisch Schiereiland. Vanaf ongeveer de Krijt –Tertiar grens (65 miljoen jaar geleden) botste noord ‘Iberië’ met ‘Europa’ (compressie, waardoor de Pyreneëen zich vormden). Kort daarna begon aan de oostkant van Iberië de opening van de Middellandse Zee (rek). Vervolgens begon Afrika naar het noorden te bewegen en zorgde voor botsing aan de zuidkant van Iberië (Sierra Nevada) met Noord Afrika (Atlas gebergten). Tijdens de hele geschiedenis opende in het westen de Atlantische Oceaan zich. In het eerste hoofdstuk wordt een algemeen beeld gegegeven van deze ontwikkeling. Omdat het gebied in geologisch korte tijd beïnvloed is door deze wisselende bronnen van spanning langs de plaatranden, is dit gebied zeer geschikt om de effecten hiervan op het spanningsveld te bestuderen. In het eerste hoofdstuk zal aandacht geschonken worden aan de grote lijnen van de geologische ontwikkeling van Iberië. In Hoofdstuk 2 worden de verschillende methoden en technieken die gebruikt zijn bij deze studie kort toegelicht. Daarbij wordt ingegaan op de eindige elementen methode, de paleo-spannings methode en het effect dat verschillen in korstdikte heeft op het spanningsveld in de aardkorst. De Aarde streeft naar evenwicht: bij een verdikte korst zal rek, in de afwezigheid van andere krachten op de plaat, zorgen voor verdunning. En andersom. Bij dit concept zal langer worden stil gestaan en op een aantal fundamentele punten van de theorie worden ingegaan die tot nog toe niet onderzocht waren. Uit een uitvoerige literatuur studie bleek dat van een aantal gebieden of bepaalde periodes in tijd, de gegevens over het toendertijd heersende spanningsveld summier waren of ontbraken. Om deze aan te vullen en een volledigere reconstructie mogelijk te maken, zijn extra gegevens verzameld. In Hoofdstuk 3 wordt uitgebreid ingegaan op de nieuwe gegevens en de gebieden waar die zijn verzameld. Daarbij ligt de nadruk op het

V

NE deel van het Spaanse Centraal Systeem en het noordwesten van Spanje (de Bierzoregio en de kust van Asturië). De literatuurstudie, aangevuld met de nieuwe gegevens, leverde een nauwkeurige studie op van de geologische ontwikkeling van het Iberische Schiereiland en de omliggende westelijke Mediterrane regio. Daartoe heb ik per tijdstap van 3 miljoen jaar kaarten gemaakt waarin alle beschikbare data uit literatuur en eigen studie zijn samengevoegd. Gegevens zijn gecombineerd uit verschillende vakgebieden van de aardwetenschappen: strukturele geologie voor de actieve breuken en plooien en de paleo-spanningsdata, sedimentaire geologie om te bepalen in welke gebieden erosie of sedimentatie plaatsvond en in wat voor omgeving (ondiepe zee, strand, rivier op het land, diep zee), isotopen geologie om aan de hand van splijtsporen te bepalen welke gebieden actief omhoog kwamen. Al deze gegevens zijn gebruikt om per tijdseenheid te kunnen reconstrueren hoe de spanningen in de aardkorst in Iberië in de geologische tijd gevariëerd hebben. Bij deze reconstructies hoorde ook het per tijdseenheid bepalen welke structuren de plaatranden van Iberië vormden en welke activiteit daarlangs waarschijnlijk is geweest. In hoofdstuk 4 en 5 worden deze reconstructies gepresenteerd in 15 kaarten en met tekst uitvoerig toegelicht. Hoofdstuk 4 gaat daarbij in op de ontwikkeling tijdens het Paleogeen (65-23 miljoen jaar geleden), terwijl hoofdstuk 5 zich richt op het Neogeen en het Kwartair, de afgelopen 23 miljoen jaar. Dit geheel is een zeer uitgebreide bron van referenties voor verdere studies in het Iberische Schiereiland en omgeving. Vervolgens zijn computer modellen gebouwd op basis van de reconstructies. Met deze modellen zijn de resulterende spanningsvelden berekend als gevolg van de aangenomen randvoorwaarden. Hoofdstuk 6 geeft een overzicht van de resultaten van de modelleringen, die vergeleken worden met het gereconstrueerde spanningsveld. Hieruit bleek dat het goed mogelijk is de orientatie van het spanningsveld te modelleren, maar dat de resulterende spanningsvelden te uniform zijn als ze vergeleken werden met waarnemingen (overal evenveel compressie, bijvoorbeeld). Gebruik van deze resultaten zou leiden tot verkeerde conclusies over activiteit van breuken. Op deze manier kan ook nieuwe inzichten in de plaattektoniek van een bepaalde periode verschaffen. Waarnemingen van een spanningsveld in de vroegere geschiedenis van een plaat kunnen inzicht geven in de activiteit van de plaatranden of de gereconstrueerde geometrie. Nieuw in dit onderzoek is het bestuderen van het effect dat variatie in korstdikte heeft op het spanningsveld. Als dit effect van (korst)topografie wordt meegenomen, komt het gemodelleerde spanningsveld een stuk beter overeen met de waarnemingen. Dit concept zou dan ook moeten worden toegepast in studies van spanningsvelden in de aardkorst. In het zevende en laatste hoofdstuk worden conclusies getrokken uit de vergelijking van waarnemingen en resultaten van de modellering. De conclusies zullen vervolgens in een ruimer perspectief behandeld worden.

VI

VII

Horse Shoe Abyssal Plain

n rri Go ank B

ge

Tajus Abyssal Plain

12W

36N

idge

isboa R

n

L

Arrabida do

8W

Algarve

Sa

rT we Lo

eri

L Torre -

Iberia Abyssal Plain

Galicia Interior

Miranda- Basq ue Urbese Pamplona

Landes Palteau

Bierzo

len

Gulf of Cadiz

Rhar

b

Strait of Gibraltar

El Jebha

Beti

Guadix

4W

za

an

Pyrene

nian

Northern

Aquita

a eim Al Hoc

an or b Al

ge d Ri

A

ea

teaus High Pla

0

liff Che

nS lbora dge

Yussuf Ri

Sr. Carrascoy

Murcia Cartagena/ Lower Segura

Alhama de Murcia

Vera Palomares

Nijar Sr. Alhamilla Carboneras

as Sorb

Ba

Fortuna

Lorca

Sr. Segura

cs

Sr. Nevada

Granada

Almuñecar rn ste n e W ora Alb

f

Ri

Gibraltar Arc

Gu

ad

u alq

ivi

r

na ore aM

r Sier

iana

taj o

Guad

ain Chr. Denca

38N

40N

42N

Sarria

la

Range

iel

tan

Ve n

Inner Basin Llanera

Cantabrian

Oviedo

Danois Bank

Teruel

ea Bal

res

oir an gne N ini ult Monta t. Ch Hera S Conflent Agly

untains Tell Mo

Mitidja

Hercynian structures

Paleozoic igneous rocks

Paleozoic metasediments

Mesozoic

Basement

Volcanics

Basins

in Bas n Tertiary eria g l Quaternary A

Valencia

Rivers Basin Sr. Alhamilla Elevated region Palomares Major fault

Tet Pyre nees Jaca Rossello Sr. Obarense Ainsa Tec Monforte Cerdanya Sr. Mts. de Oca Rioja C Monts De ameros La Bureba ec Trem Gu Tui Leon p a ra Sr. De la Demanda e Emporda r Mnts. g Vallfogona Ager Xinzo Olot Ca Se Srs. Eb ta Penedes Verin Duero M la a rg ro inales yu r e dtre rd Alma la Montnegre Da nas CR Bo zan lel Bo des C ro oCa ern Pra c h m t a r o ra o a S f r N Mo oy Garra sie m Vilarica nte , oz Valles e s Ambles L Vilarica , yst Zaorejas Jil na El Camp go Alba i S o r oc g od Ponsul a rra r Tajo d R Gata Rubielos Ta da l rramaorde e tes a Mora de Ciu r. D r bre B Maestrat t a S n gh lum Co Ce Guaudthern rou T edos Sr. SoMadrid a Gr l De . Sr re cia Loranca st Mijares na .E ta Menorca Ribesaldes/ len a Sr un V Alcora m Castelo a r Toledo Mnts. .T Branco al Sr rc Cabriel Mallorca Ce Valencia Ibiza ajo

Galicia

Villalba

As Pontes

Bay of Biscay

Ib

se

S ue C

Pla

Galicia Bank

Meirama

ia

us ita n

sp

-A

He

sif as M c

nc ia

n

es

r

ia er

Un

al

ko

44N

Ortegal Spur

Sr. Altomira

VIII rs

e iv

Ne

Figure 1.1 Overview of the Iberian Peninsula and western Mediterranean with the names of basins, major faults and regions that are used in the description of the geological evolution of the area (see www.geo.vu.nl/~andb/iberia for full color version).

Cenozoic tectonic evolution of the Iberian Peninsula

Chapter 1

CHAPTER 1 - INTRODUCTION TO THE TECTONIC SETTING OF THE IBERIAN PENINSULA Differential motions between tectonic plates create zones of intense deformation along their boundaries. In most cases large parts of the deformation is localized along these borders but forces related to the plate interactions also transmit stresses to the interior of the plates [Cloetingh, 1992]. If these intraplate stresses become sufficiently high to overcome the strength of the material, of which the plate is composed, then intraplate deformation related directly to plate interaction will occur (e.g. Ziegler et al. [1995]). Thus observations of intraplate deformation provide indirect information about the magnitude and type of activity along the plate boundaries. For example, interaction between the African/Arabian and Eurasian plates has generated a broad collision zone, the Himalayan-Alpine chain, running from SW Europe to SE Asia. For the Iberian Peninsula, located at the western end of this zone of convergence, the gradual opening of the North Atlantic is the most important factor in the complex pattern of differential motion between Eurasia, Africa, and Iberia over the past 120Ma. A condensed outline of the geological evolution of the Peninsula is provided in this chapter. The Cenozoic geological evolution will be elaborated in great detail in Chapter 4 and 5. The reader interested in more details on the pre-Cenozoic evolution is referred to references cited.

General geological evolution The time frame this thesis on focuses is the Cenozoic (65Ma ago to present-day). Sedimentary basins that developed during this period (Figure 1.1) cover large parts of the Iberian Peninsula. The basement underlying these basins in the western part of Iberia consists mainly of Hercynian metasediments and igneous rocks; whereas in the eastern part the basement is composed of Mesozoic series. Pre-Tertiary structures (faults, suture zones) played a major role in the Cenozoic deformation of the Peninsula.

Pre-Mesozoic Iberia was still attached to Armorica (N. France) prior to the Late Mesozoic opening of the Bay of Biscay [Garcia Mondejar, 1988]. At the time of the Hercynian orogeny (E. Carboniferous – L. Permian) Iberia was part of the Variscan arc running through Belgium, Northern France, Southern England [Ziegler, 1989]. The large-scale structural domains such as suture zones, that formed during this orogeny, can still be traced in the basement of western Iberia and play an important role in later deformation phases [Stapel, 1999]. The origin of the arc-shape of the Hercynian structures is still a matter of debate; see Dias & Ribeiro [1995] for a discussion of several models. Doblas et al. [1994] present a detailed description of the tectonic evolution from the Variscan to Alpine tectonic stages.

Mesozoic Progressive opening of the Atlantic Ocean between the Americas and at first Africa, later Iberia and finally Europe caused large differential motions between these continents (see Figures 1.2a and b). Active extension resulted in a stage of major rifting during the Mesozoic as documented by extension on all of the margins of Iberia at one time or another:

1

Chapter 1


A

E. Cret. ~110 Ma

B

L. Cret. ~85 Ma

C

E. Tert. ~50 Ma

D

M. Tert. ~35 Ma

Continent Continental and shallow marine clastic sediments Shallow marine carbonates and deeper marine sediments Marine basins floored by oceanic crust Plateau basalts

Active sea-floor spreading axis with magnetic anomalies As above but abandoned Faults/fault zones Present-day coast line 0

1000km

Figure 1.2 Progressive opening of the North Atlantic caused several periods of differential motion between the Eurasian, African, and Iberian plates. A = Early Cretaceous (~110 Ma), B = Late Cretaceous (~85 Ma), C = Early Tertiary (~50 Ma) and D = Middle Tertiary (~35 Ma). See text for explanation, after Ziegler (1988).

2


Chapter 1

(a) The western Iberian margin. Subsidence history prior to and after break up of the western Iberian margin have been described by amongst others Stapel et al. [1996a] and Rasmussen et al. [1998]. Two major stages of rifting have been recognized: the first during Late Triassic and the second in Early Jurassic. This last stage of rifting is followed by an important regional hiatus, probably related to opening of the Central Atlantic Ocean coinciding with the beginning of oceanic spreading at the Iberian Abyssal Plain, dated at 126Ma [Whitmarsh & Miles, 1995]. A third important phase of extension started in the Late Jurassic and a final phase of extension, poorly dated in the shallow parts of the margin, occured during the latest Late Jurassic to earliest Early Cretaceous. The final separation of Galica Bank and Flamish Cap has been dated at around 118 Ma. Subsequently, the area experienced regional (thermal) subsidence [Stapel, 1999]. (b) The eastern Iberian margin. The Alpine Tethys opened during the Mesozoic and, therefore, the eastern part of Iberia was locus of important rifting as well: the Iberian Basin was formed [Salas & Casas, 1993]. The westernmost traces of Mesozoic sediments related to this rift can be found as far in the continent as the Central System, where patches are preserved in ‘pop-downs’ or along major structural contacts [De Vicente et al., 1996c]. This area constituted the western extreme of the basin: thickness of the Mesozoic series increases rapidly eastward from the eastern Central System towards the Iberian Chain/Sierra de la Demanda. Subsidence analysis of the basin [Van Wees et al., 1998] as well as thermal modelling [Fernàndez et al., 1995] shows several rifting stages in eastern Iberia, the major ones being Permian-Early Triassic and Late Jurassic-Early Cretaceous of age. (c) The southern Iberian margin. From Triassic until early Liassic (E. Jurassic) the southern Iberian margin was the locus of a large shallow-water carbonate and clastic shelf development. Rifting along this margin started approximately in the Toarcian (E. Jurassic, 190Ma), resulting in a break-up of the carbonate shelf and a deepening of a part of the basin. Depressions bounded by listric faults were filled with synrift sediments. Active rifting changed to post-rift during the Early Malm (L. Jurassic, 160Ma) and related thermal subsidence lasted during the rest of the Malm and Cretaceous [Vera, 1988]. This change in rifting activity can be correlated to the development of sinistral transtensional motion along the transform zone between Africa and Iberia [Bakker et al., 1989; Biermann, 1995], along which local oceanic pull-apart basins were formed [Vera 2001]. (d) The northern Iberian margin. Here extension resulted in opening of the Bay of Biscay towards the end of the Early Cretaceous, which continued until 85 Ma [Ziegler, 1988]. Apart from widespread rifting along all of the margins of Iberia, another effect of the onset of active seafloor spreading in the Azores part of the North Atlantic (~126 Ma) and the Bay of Biscay (~115 Ma) (see Figure 1.2a) was an anti-clockwise rotation of Iberia with respect to Eurasia [Savostin et al., 1986]. This induced left lateral motion between Iberia and Eurasia coincided with collision and Late Cretaceous subduction of the Ligurian Basin onto the eastern side of Iberia [De Jong, 1990], developing the stack of the Betic nappe units [Biermann, 1995]. Towards the end of the Mesozoic at about 85 Ma (see Figure 1.2b), the opening of the Atlantic propagates between Greenland and Ireland (first along the subsequently abandoned Rockall Trough), leaving the Bay of Biscay as a failed rift [Ziegler, 1988; Srivastava et al., 1990]. The new dynamic setting led to a clockwise rotation of Eurasia with respect to Iberia causing approximately northsouth convergence. This resulted in inversion of the northern margin of Iberia, even developing into northward subduction/underthrusting of Iberia (starting in the Campanian [Puigdefàbregas & Souquet, 1986]), creating the Pyrenees.

3

Chapter 1


Cenozoic In contrast to the Mesozoic, the Tertiary and Quaternary in the Iberian Peninsula are periods dominated by compressional deformation. Deformation related to the closure of the Bay of Biscay-Pyrenean zone progressed to the west through time causing inversion of Mesozoic extensional basins [Garcia Mondejar, 1988]. Development of the Cantabrian Cordillera was related to a short-lived southward subduction of the previously formed oceanic crust in the Biscay region during latest Cretaceous to Early Eocene [Boillot & Malod, 1988] and ongoing convergence. The termination of this subduction coincided with the separation of the rotation poles of Africa and Iberia (with respect to Eurasia) as proposed by Savostin et al. [1986] at around 54 Ma (see Figure 1.2c). The start of limited differential motion between Africa and Iberia can be related to the first occurrence of large basalt flows on the western side of Greenland (pointing at the onset of seafloor spreading in this area) and is reflected in the pattern of ocean floor ages. Stresses related to the collision along the northern margin of Iberia and Eurasia were transmitted to the interior of the Iberian plate and resulted in major inversion of the Iberian Basin (see Figure 1.2d), forming the Iberian Chain [Alvaro et al., 1979], Sierra de la Demanda and Sierra de Gredos (western Spanish Central System). Final amalgamation of Iberia to Eurasia at around 30 Ma coincided with a major change in active plate boundary: the left lateral Azores-Gibraltar zone south of Iberia [Srivastava et al., 1990] is activated. Africa continued moving eastward with respect to Eurasia (including Iberia), causing an active left lateral motion plate boundary to the south of Iberia and contributing to the opening of the Valencia Trough and Balearic Basin to the east [Sabat et al., 1995]. Extension in this region started as early as Oligocene on shore in southern France (related to the Rhine-Bresse Graben system) and shifted progressively southwestward, beginning by E. Miocene (23-20 Ma) in the Alboran domain [Sanz de Galdeano, 1996]. The driving process behind the development of this extensional basin is still a matter of debate, see Ziegler et al. [2001]. Opening of the Provencal and Valencia basins has been related to the subduction of African plate beneath the Iberian-European plate [Roca, 2001]. However, the system is located at the prolongation of the Cenozoic rift system of western and central Europe, that cannot be related to back-arc extension [Ziegler, 1994; Ziegler et al. 2001]. A change in direction of convergence from NNW to NW between Africa and Eurasia in Tortonian [Mazzoli & Helman, 1994] leads to the major development of the Betics. Inversion tectonics is observed in the interior of the Iberian Plate in the Spanish Central System [De Vicente et al., 1996c] and in L. Tortonian-Messinian in the Alboran Basin [Lonergan & White, 1997]. Seismic activity in Pliocene and Pleistocene in central Iberia has been considerable [Giner Robles et al., 1996; Rodríguez Pascua, 1997]. A high level of internal deformation is demonstrated by crustal scale folding [Cloetingh et al., 2001], large-scale Pliocene uplift of several hundreds of meters in coastal areas [Janssen et al., 1993], present-day seismicity [Buforn et al., 1988] and the development of new crustal shear zones in the Alboran basin [Andeweg & Cloetingh, 2001]. All of these effects accompany the ongoing convergence between Africa and Iberia [Argus et al., 1989]. A very detailed description of the Cenozoic tectonic evolution of the Iberian Peninsula and the western Mediterranean will be presented in Chapter 4 and 5.

4


Chapter 2

CHAPTER 2 – CONCEPTS AND METHODOLOGY OF STRESS ANALYSIS This Chapter focuses on the methodological and conceptual ideas that have been applied in this research. After a short introduction on stress in the Earth’s crust, special attention will be paid to: 1) The paleostress (fault inversion) method, which has been used to determine the evolution of the stress field in the Iberian Peninsula through the Tertiary; 2) Finite element modelling, which has been applied to model the reconstructed stress fields; and, 3) The concept of stresses induced by lateral variations of density in the crust of the Earth. This concept has been incorporated in the numerical modelling of the stress field. After a short outline of the theory behind the concept, results will be presented in order to elucidate the sensitivity of the analysis with regard to the input data and applied calculations.

2.1 Stress in the Earth’s crust Planet Earth has been cooling since its formation [Sclater et al., 1980]. Driven by temperature differences related to this cooling, convection cells of mantle material drag along the lithosphere plates that move relative to one another. Formation of new crustal material and divergent motions along spreading mid-ocean ridges are compensated by material returning into the deeper levels of the Earth along convergent subduction zones. Oceanic crust is dense and, therefore, can be subducted whereas more buoyant continental upper crust in general is not recycled to mantle levels. Differential motions between plates define the actual state of stress in the lithosphere. Stress leads to deformation, the type or intensity is depending on the strength of material of which the crust is composed (the combination of rheology and thickness of stronger and weaker layers, e.g. Ranalli & Murphy [1987]). Observations of stress are limited to the outer part of the Earth and are always indirect, because only strain (deformation) is observed. Using concepts linking deformation (strain) through rheology (material parameters, depending on physical conditions) to stress, the stress field can be inferred from these observations. Principal sources of information about stress in the crust are: 1) kinematic indicator data e.g. fault slip The so-called ‘fault inversion method’ [Angelier, 1994] uses markers on fault planes to infer the direction of their movement. It is the only method to obtain quantified information on the orientation of paleostress regimes and will be discussed in section 2.2. 2) borehole break out data and stress relief measurements Both techniques are closely linked to drilling techniques. Borehole breakout data are instabilities (breakouts) in the walls of drill holes. Logging devices scan the originally circular drill holes for failures of the borehole wall and general shape of the hole. From the orientation of these observations, the orientation of the maximum horizontal compression (Shmax) can be determined. For a more detailed description on this relation between maximum horizontal stress directions, borehole instabilities and tectonic regimes see e.g. Schindler et al. [1998], Appendix A of Gölke [1996] and references therein. Stress relief measurements determine the strain relaxation (‘recovery of initial 5

Chapter 2


Seismic station 1

Seismic station 2

W ? Plane A

e av cw i ism Se

Earthquake P-area

T-area

N

A

+

P-area

T-area

B

P-area

A

B T-area

S

? Plane B

B C

+

Figure 2.1.1 Schematic representation of an earthquake and the way pressure and tension axes related to the seismic E event can be determined and displayed. Panel A: Schematic E-W crosssection of an area where a hypothetical earthquake occurs due to movement along either plane A or plane B. Note difference in first arrival of the seismic waves at the seismic stations. Panel B: 3D representation of the planes A and B from panel A and their A relation to the lower hemisphere of a sphere. = Panel C: 2D representation of the lower hemisphere of panel B. This method is the stereographic method, the standard presentation for focal D mechanism solutions.

form’) when a rock sample is separated from the volume of the surrounding rock (if a core is taken out a drill hole). This relaxation of deformation can consequently be used to deduce the magnitude and direction of the principal axes of stress. Both these methods only resolve information about the present-day stress field. 3) earth quake focal mechanism solutions Focal mechanism solutions use the first arrival of seismic waves produced by earthquakes (see Figure 2.1.1) to determine the orientation of the principal axes of stress. This first arrival (P-wave) has either compressional or dilatational polarity. Even without knowing which of the perpendicular planes (A or B in Figure 2.1.1) has moved, the quadrant in which the maximum compression should be located to produce movement of the fault can be calculated. The P (pressure) axis will be located in this quadrant (white in the Figure), whereas the T (tension) axis is in the other (gray in the Figure). The standard graphical presentation of these data is, just as many structural geological data, a projection of the lower hemisphere of a sphere. To clarify this method, Figure 2.1.1b shows the example in 3D and 2.1.1c using the stereographic projection. When several focal mechanisms in a region can be determined, the individual solutions can be combined to obtain a better constraint on the P and T directions (see Figure 2.1.1d). This is the so-called right dihedra method [Angelier & Mechler, 1977]. An advantage of focal mechanism solutions is the ability to obtain information on the stress field at deeper levels of the lithosphere. With all of these methods to infer the stress in the Earth’s crust, only the orientation and relative magnitude of the principal axes of stress (σ1, σ2 and σ3) can be determined. The ratios between the principal axes derived from their relative magnitudes are used to describe the stress fields in terms of tectonic regimes (see Figure 2.1.2). For all these methods, information can only be obtained at a local scale. Directions of stress in local settings may be related to local features such as zones of weakness, second order sources of stress, or other features that cause deviations. The World Stress Map compilation of stress indicator data ([Zoback, 1992] and http://www-wsm.physik.unikarlsruhe.de) shows that although local states of stress might vary considerably, the regional stress field in tectonic plates displays a relatively homogenous pattern on the larger scale [Coblentz & Richardson, 1995].

6


Thrust fault regime

SV=s2

SV=s3 Sh=s2

Strike-slip fault regime

SH=s1

s1>s2 >s3

Sh=s3

Chapter 2

Normal fault regime

SV=s1 SH=s1

s1>s2 >s3

SH=s2

Figure 2.1.2 The three general tectonic regimes described by the orientation of the principal axes of stress. After Bada [1999].

Sh=s3

s1>s2 >s3

In the World Stress Map program, solutions from fault slip inversion data are given a low quality rank. Focal mechanism solutions and borehole break out data are supposed to be more reliable [Zoback, 1992]. The fault inversion method, however, is the only way to obtain direct and quantified information about the stress field throughout geological history. Furthermore, comparison between stress trajectories obtained independently from focal mechanism solutions with the trajectories resulting from kinematic indicator data by Mercier et al. [1973] and SIGMA [1998] show very similar patterns. This suggests that careful analysis of paleostress regimes can provide valuable information on the orientation of the principal axes of stress for the geological past [Bergerat, 1994].

2.2 Paleostress method Several features observed in the field can be used to infer, each in a specific way, the principal axes of (paleo)stress. These include fault-slip marks, joints and fractures (see e.g. Alsaker et al. [1996] and references therein), tension-structures, pitted pebbles and mechanical twinning of calcite crystals (see González-Casado & García-Cuevas [1999] and references therein). In this thesis only fault-slip marks and pitted pebbles have been used and therefore only these methods will be described in the next sections.

2.2.1 Fault inversion method Over the last decades the fault inversion method has been applied to many regions around the world to determine the local reduced paleostress tensor for the region of interest (e.g. Angelier et al. [1986]; Bergerat [1987]; Bada [1999]; Simón-Gómez [1989]). A detailed outline of the method, its merits and limitations can be found in Angelier [1994]. I will only present the basic principles on the fault inversion (often referred to as ‘paleostress’) method.

Basics The method is based on a rather simple concept. Consider a cube (Figure 2.2.1.1) that is subjected to a stress field. For any imaginary plane that would be cut through the cube, it is possible to determine whether and in what direction movement would occur (providing the shear stress on the plane overcomes the frictional resistance). This is the case for any orientation of compression, extension or a combination of both exerted on either side of the cube.

7

Chapter 2


Figure 2.2.1.1 For any imaginary plane (two examples are dashed) through a cube on which stress is exerted, it is possible to determine whether and how movement would occur along this imaginary plane. The fault inversion techniques are based on the inversion of this concept. If on several, differently oriented, fault-planes movement is indicated by markers (see Figure 2.2.1.2), the orientation of the principal stresses can be determined.

Motion along fault planes often results in irregularities on the fault surface or secondary, minor faults developing due to the principal fault motion. In outcrops, different types of marks on fault planes (see Figure 2.2.1.2) allow determination of direction of motion along the observed fault plane. Both the dip and dip direction of the fault plane and the angle between the strike of the fault plane and the direction of movement in the plane can be measured. With enough data, the concept described above can be inverted (the reason why the method is called fault slip inversion) to obtain the orientation of the principal stress axes that produced the A B C observed movement along the fault. To get a confident solution [Delvaux, 1993], information should be obtained for a certain minimum of faults with different orientations. This minimum number of observations depends on the typical setting. D E F Mathemetically, four faults would be sufficient, but depending on the type of fault-activity (reactivation, new formed faults, strike-slip or dip-slip faulting) this can vary from 6 to 15 faults. For a set of observed faults it can be determined for G H I which combination of orientation of Figure 2.2.1.2 compression and/or extension an optimum Different types of markers on fault planes that is reached between explaining a maximum enable determination of sense of movement along number of fault-motions and obtaining a the planes: a) accretionary mineral steps, b) tectonic tool minimum error between theoretical direction marks, c) Riedel planes, d) stylolithes, of slip (α in tables) and observation. The e) stepped surfaces, f) tension gashes, absolute value of these stresses cannot be g) conjugate shear fractures, h) slickensides, i) other criteria. After Angelier computed, but the ratio between the stresses in the three perpendicular [1994]. directions provides information about the type of stress field (see Figure 2.1.2) that would be able to produce the observed deformation. This ratio is expressed as R, which is defined by R= (σ2-σ3/σ1-σ3).

8


Chapter 2

Temporal constraints on stress fields From a combination of structural and stratigraphic information, upper and lower time limits can sometimes be placed upon stress fields and changes in stress fields through time can be observed. Figure 2.2.1.3 shows this in a very schematic, idealized and theoretical situation. Such ideal conditions are, however, quite rarely available. Where such conditions are not fulfilled, dating of superimposed paleostress fields becomes a major issue, see Hancock [1994].

A

B

C

N

Figure 2.2.1.3 Theoretical example to explain how it is possible to determine changes in the stress field through geological time. In the example an E-W compression phase (panel A) is resulting in thrusting along a fault plane perpendicular to the maximum compression direction, throwing up the hanging wall of the fault. A period of erosion (panel B) and a subsequent stage of deposition (panel C) result in leveling of the fault scarp and sealing of the fault, respectively. After this, an extension oriented N-S leads to the development of normal faulting (panel D). In this way, the lower block will contain deformation related to both stress regimes, while the upper layer will testify that the compression took place before and the extension after its deposition. If the stratigraphic ages of both sequences are known, upper and lower time limits can be placed upon the stress fields.

D

Limitations Several basic assumptions are imposing limitations on the concept: (1) The mathematical definition that the three stress-axes are supposed to be orthogonal. In most methods, one of the principal axes of stress is supposed to be oriented vertically; the other two are in the horizontal plane (the so-called “Andersonian state of stress” [Anderson, 1951]). In reality this does not have to be the case. (2) The cube is made of isotropic material. Applying the paleostress method, deformation is linked to stress through the material properties of the deformed medium. In practice, geological materials most often are not homogeneous at all and/or contain pre-existing weakness zones, such as bedding planes or inherited faults. Therefore, it is still a matter of debate whether the observed strain (slip along fault planes and pressure solution marks) can be related directly to regional stress regimes (e.g. Gapais et al. [2000]). A further complication is the scale dependence. A minor volume of rock might be homogeneous and isotropic, but in a regional study inhomogeneities and anisotropies will occur in the rocks considered. Furthermore, large-scale structural inhomogeneities (anisotropies) affect the local directions of the stress field (e.g. Rebaï et al. [1992]). It is 9

Chapter 2


well known that close to fault zones (whatever their size), the principal stress axes experience rotation (e.g. Casas-Sainz & Maestro-González [1996]). The same holds, although to a lesser extent, for any weak zone and rheological contrast (which shows in e.g. cleavage angle difference between shale and sandstone). For a larger region however, rather consistent and reproducable patterns of stress orientations are observed, so deviations from the regional stress field by large-scale weakness zones can be detected [Maestro González & Casas Sainz, 1995; Zoback et al., 1982]. (3) It is assumed that observed movements along the faults occurred during a single deformation event. To explain the problem involved when several deformation events have occurred, see again the example presented before (Figure 2.2.1.3). If in a subsequent stage the top of the entire sequence is being eroded, the lower layer will contain 2 stress stages, but it cannot be determined directly that these were originally two separate events. In the example, this might result in a misinterpretation of the deformation to have been formed under a single strike-slip field. In many geological cases it is not straightforward from outcrop observation whether the rocks have been deformed by a single event or in multiple stages under possibly different orientation of the main stress field. Or more general: different reduced stress tensor of which the effects cannot be distinguished on the scale of the observation. In order to discriminate different fault sets and to minimize the subsequent error in the calculation of the stress tensor, the approach shown in the flowchart (Figure 2.2.1.4) is used. During observation in the field, sets of faults belonging to different systems are separated as far as possible into different groups. Plotting faults and striations in stereographic projection and in rose diagrams can help to constrain the fault sets better and to assign additional faults to the subsets. Another tool to group different types of faults (normal, strike-slip, inverse) under a single direction of maximum horizontal Re-evaluate fault subset separation

RIGHT DIHEDRA METHOD Approximate determination tensor TENSOR

FEX (CRATOS) Determination theoretical slip and stress regime individual faults

yes

Assigning faults to subsets of different tectonic phases

Retrodeform or rotate

rej

ec

Geometric analysis of structural elements

STEREOGRAPHIC/ ROSE DIAGRAMS

ted

fau

lts

AFAF (CRATOS) Small adjustment possible errors in measurement

No

Does fit observed and theoretical movement improve? yes

Final definition subsets

no Indications for rotations?

Does theoretical no slip and observation coincide?

Accept adjustment

TENSOR Tensor calculation Shear optimisation

compatible with Mohr-Coulomb criterium MOHR DIAGRAM yes

TENSOR Measuring kinematic indicators and structural elements in the field

Interpretation and evaluation

COMPASS

START 10

END

Figure 2.2.1.4 Flow chart of the applied methodology in obtaining tensor solution for observed sets of faults.


Chapter 2

compression is the K/dy diagram by De Vicente [1988]. This method can provide theoretical movement for faults with observed orientation of movement but unknown sense. (4) After the fault movement, the observed rock mass did not experience block rotation. In case of moderate deformation, block rotations can result in a tilt (along an axis in the horizontal plane) or spin (along a rather vertical axis) of the original orientation of the kinematic indicator data. The latter can be constrained by paleomagnetic data. Whenever indications from the field suggest late stage tilting of the observed outcrop, the observed kinematic indicator data should be rotated back to the orientation under which they have been formed originally. In some cases a post-fracturing rotation along a horizontal axis can be derived from the resulting stress field, if a consistent solution is obtained for which the principal stress axes are tilted with respect to the horizontal. (5) The rock mass moved rigidly along the discrete fault planes. Numerical experiments by Dupin et al. [1993] show that if numerous data are used and fault spacing is large, the assumption of rigid discontinuous rock mass behavior does not affect the resulting stress calculations to a great extent. Numerical experiments have been applied to prove that some of the previsouly mentioned assumptions of the fault inversion method are not valid in some cases. Cashman & Ellis [1994] showed that multiple striations on a single fault surface can result from the interaction of faults during a single deformation event in a zone with a complex fault pattern and need not reflect major changes in the stress field, as they are often interpreted. Another experiment examining the discrepancy between directions of maximum shear stress and predicted slip on fault planes in fault systems with different orientations came to similar conclusions [Pollard et al., 1993]. These numerical studies show that the relation between fault slip data and stress orientations should be considered with some caution. Application of the results of the numerical studies would require a very detailed knowledge of weakness zones present in the area of study. To determine this for a present-day setting is difficult, let alone for the geological past.

2.2.2 Striations and solution pits on pebbles Pitted surfaces on pebbles in conglomerates have long been observed and have been attributed to pressure solution (e.g. Trurnit [1968]). Sanz de Galdeano & Estévez [1981] demonstrated that, together with scratches on the pebble surface (resulting from relative movement along the matrix-clast contact), these pressure solution features can be used as kinematic indicator data to determine the orientation of regional compression. See Figure 2.2.2.1 for an example of a pebble with kinematic indicators. The process of forming kinematic indicators on pebble surfaces in a matrix is favored by (amongst others) (a) the presence of interstitial water, (b) an important difference in size between pebble and matrix grains and (c) a high contrast in hardness between matrix and pebbles components. E.g., limestone pebbles in a matrix containing quartz grains are very likely to develop striations and pressure solution features. When compared with ‘traditional’ fault slip inversion, the orientation of the maximum compression obtained by the kinematic indicators on pebbles provides very similar results ([Estévez & Sanz de Galdeano, 1983] for the Granada and Guadix-Baza basins). To produce these features relatively limited compression is required [Sanz de Galdeano & Estévez, 1981]. This makes the kinematic indicators on pebbles ideal to determine the orientation of regional compression in neotectonic studies for areas where other kinematic indicator data are less well developed or absent.

11

Chapter 2


Several methods have been proposed to determine the orientation of the principal stress axes from the deformation of pebbles in a conglomerate: (1) Sanz de Galdeano & Estévez [1981] used the conical distribution of the striations, converging to 2 opposite poles where dissolution is maximal. The orientation of the line connecting both poles is measured and is supposed to coincide with the maximum compression direction. Estimates for the other principal axes are not provided by this method. (2) Schrader [1988] describes a measurement technique putting the pebble into a system of spatial polar co-ordinates in its in situ orientation. The features on the pebble are measured in spherical angles with respect to the origin: the shape of the pebble is neglected. For near spherical pebbles deviations are small. The maximum direction of compression is measured directly on the pebble; the orientations of the intermediate and minimum compression axes are derived from the trace of the striations on the surface of the pebble. (3) Taboada [1993] proposes that the striations and polished Figure 2.2.2.1 Panel A: carbonate pebble from outcrop near Ponferrada surfaces on a single pebble can (NW Iberia) showing pressure holes and scratches by quartz be treated as a set of faults grains, indicating direction of direction of maximum developed by the local stress compression (σ1). Panel B shows interpretation. tensor around the spherical body or the displacement tensor between the pebble and the matrix. This means that measuring the orientation of the observed striations on the pebble surface would provide an ‘ideal’ set of measurements for the fault slip inversion method. This allows for the determination of the orientation of all three principal stress axes. The method is a theoretical and graphical method to obtain the trajectories of the striations on the surface of a pebble, projected on a sphere. Rodríguez Pascua & De Vicente [1998] found very similar results for the methods of Schrader [1988] and Taboada [1993] for an extended data set of pitted and scratched pebbles from the Zaorejas basin. Since the method by Schrader [1988] is more practical in use, this method is preferable, as long as pebbles are as spherical as possible.

12


Chapter 2

2.2.3 Regional context and (paleo)stress trajectories in Iberia Results of paleostress studies should be correlated carefully with regional data on mesoto macro scale structural elements in order to evaluate the regional meaning of the obtained results and in order to date its timing. An example of how important it is to interpret local results in a regional context is the Madrid Basin (MB). The intraplate Cenozoic Madrid Basin forms part of the Tajo Basin, located in the central part of Spain (see Figure 2.2.3.1). Its northern border is formed by the Spanish Central System (SCS), an intraplate mountain range with present elevation up to 2600m, standing up to 2000m above the basin. The development of this intraplate mountain range is due to multiple reactivations of late Hercynian basement faults. Most important has been the flexural response to southward trusting of the Spanish Central System. In chronological order the development of the basin can be described as follows. During the Paleogene WSW-ENE trending boundary faults of the SCS favored sinistral strikeslip movement under the NNE-SSW compression of the Pyrenees [Vegas et al., 1990]. In the north-eastern part of the SCS this led to a transpressive setting causing flowertype structures, which can be documented in the thin Mesozoic cover and early Paleogene sediments (see Chapter 3). Under this compressional stress regime, the Iberian Chain (IC) thrusted onto the NE basement of the MB. Deformation continued later under an approximately E-W compression which resulted from the superposition in place and time of the Pyrenean and Betic compression, forming the Sierra de Altomira [Muñoz Martín et al., 1998]. Upon waning of the Pyrenean compression, the situation reversed: the IC became a major dextral strike-slip zone (lateral offset estimated up to 35km, [Bergamín et al., 1996]) whereas the border faults of the SCS were oriented in favor of thrust-faulting under the Betic NNW-SSE compressional stress field. The flowertype inheritance of the SCS (see Chapter 3) favored the popping up of the mountain belt during M. Miocene. The Southern Boundary Fault (SBF) was dipping steeply, thus creating an active type of basin setting. The uplift, loading by southward thrusting of the SCS, and sedimentation created a deep foreland basin containing 2000-3500m of sediments of predominantly Late Paleogene-Late Miocene age [Querol, 1983]. The active southward thrusting of the SCS seems to have ended at around Pliocene, ongoing tectonic activity seems to be restricted to second order strike-slip faults that cut the ancient thrust front. Paleostress studies by De Vicente et al. [1996b] show an abundance of deformational structures in M. Miocene to Quaternary sediments in the central part of the MB on different scales. Both numerous paleostress tensors deduced from fault-analyses in M. Miocene-Quaternary sediments within the MB and focal mechanism solutions indicate the presence of two contemporaneous stress fields active from M. Miocene to present: NNW-SSE compression and sub-parallel NNW-SSE extension [De Vicente et al., 1996b]. Presently the major thrust front of the SBF is bisected by smaller strike-slip faults trending NNE (sinistral) or NW (dextral). Part of the seismic activity along the southeastern border of the SCS seems restricted to these strike-slip faults (see Figure 2.2.3.1). Distribution of seismicity in the MB is restricted to a very well defined zone, trending approximately parallel to the SCS, some 50-70 km to the south of the major thrust front, coinciding with the location of the flexural bulge of the SCS-MB Miocene foreland system. In this region, the major valleys show obvious straight traces, bounded by normal faults (see Figure 2.2.3.1) creating hanging valleys on either side.

13

Chapter 2


Figure 2.2.3.1 The neotectonic setting of Central Iberia shows a set of intraplate mountain ranges (Spanish Central System, Iberian Chain) separating extensive basins (Duero Basin, Madrid Basin). Upper panel: rejuvenation of the flexural bulge related to the M.Miocene southward thrusting of the SCS [Van Wees et al., 1996] is suggested by the distribution of seismicity (red dots) and normal faulting. Present-day activity of the Southern Border Fault of the SCS is restricted to strike-slip faults bisecting the major thrust front. Lower panel: cross section through the SCS and MB along profile line in upper panel, showing projected earthquake locations (with error bars) and available focal mechanism solutions. The arrows indicate superficial local stress fields.

14


Chapter 2

Deformation of Quaternary sediments is restricted to the same zone as the seismically active area. Large parts of the focal mechanism-solutions are compatible with an overall NNW-SSE compression related to ongoing convergence between Africa and Iberia, whereas some indicate extension exactly parallel to this overall compressional direction [De Vicente et al., 1996b]. These findings are compatible with flexural upbending of the lithosphere causing extensional bending stresses at shallow levels, whereas at depth compression will prevail below a neutral surface. Although more data on focal mechanisms are required to support this hypothesis, projection of the known focal mechanisms onto a cross-section seems to indicate a vertical separation between the extensional and compressional earthquakes (see lower panel in Figure 2.2.3.1). The neutral surface in this case, is due to scarceness of data, lateral projection and errors in focal depth determination not well defined. However, the lower panel of Figure 2.2.3.1 shows that the shallowness of several compressional events point to a neutral surface around 5-7km. This seems to be in the same order as the EET (Effective Elastic Thickness) value of 7km obtained by flexural basin modelling [Van Wees et al., 1996]. This value, however, might be an underestimate, since the spatial distribution of presentday seismicity and Quaternary deformation indicates a present amplification of the initial foreland-bulge of the MB-SCS due to intraplate compression. Therefore, the present-day curvature of the plate should not be considered as the result of flexural loading by the SCS only. The extension parallel to the regional compression in the MB can be explained as a superficial expression of the overall NNW-SSE compressional regime active in central Iberia from about Middle Miocene to present-day and should therefore not be regarded as the expression of a distinct extensional phase. Rather it shoud be attributed to the effect of flexural bending stresses resulting from regional compression.As described before, differential movement between Eurasia, Africa and Iberia, due to the progressive opening of the Atlantic, is reflected in Iberia by changing Tertiary stress fields. Numerous Tertiary intracontinental basins have been formed and subsequently deformed as large Late Hercynian basement faults were reactivated multiple times under these stress field changes. In this way, the infill of these basins and uplift of their borders have (indirectly) recorded changes in plate boundary activity of the Iberian plate and provide indirect information on these changes. Figure 2.2.3.2 shows a rough correlation of major unconformities and tectonic events for the larger basins in northern and central Iberia. The Tertiary basins in Iberia form good sites for paleostress studies, as their infill is episodic and young. The latter implies that the sediments witnessed only the last few deformational stages, which makes it easier to distinguish them. Moreover, in many of these basins, Tertiary sediments are deformed. Along the borders of the basins this deformation can be very pronounced. Away from the active borders, deformation becomes moderate, forming abundant deformational structures without large-scale rotations, satisfying some of the basic assumptions of the paleostress method. The internal structure of the sedimentary infill, imaged by reflection seismics (e.g. Querol [1983]; Pulgar et al. [1997]; Sabat et al. [1995]), contains a large body of information on the tectonic history of the basins and the areas bordering them. On basin wide scale, local stress fields are documented very well, both in time and place. For Central Spain (e.g. De Vicente et al. [1996b]; Muñoz Martín [1997]), the Betic region (e.g. De Ruig [1990]; Stapel et al. [1996b]; Huibregtse et al. [1998]; Jonk and Biermann [2001]), the Iberian Chain (e.g. Simón-Gómez [1989]), Ebro Basin and southeastern foreland of the Pyrenees (e.g. Guimerà [1984]), a considerable amount of data has been gathered and published.

15

16

25

20

15

10

5

Ma

Zanclean

Chattian

23.2

Aquitanian

Burdigalian

Langhian 16.8

Serravallian

11.8

Tortonian

Messinian

5.4

Piazencian

1.8

Neogene marine

1.6

Neogene mammal

Conti -nental

Gutierrez et al. 1986

Villena et al. 1996

Minor horizontal but important vertical movements

S4

4

S5

5

S6

6

S7

7

S8

8

S9

Almazan

Thrustfront of Cameros inactive

W. Ebro

Formation of the fundamental erosion surface in the Iberian Chain

C. Ebro

Onlap on the basin margins

S4

4

S5

5

S6

7 S7 6

S8

8

S9

9

Duero

Loranca

Mediavilla et al. 1996 Calvo et al. 1996 De Vicente et al. 1996

Madrid

S. of SCS Sierra Altomira

Sudden coarsening Age of unconform. defined Unconformity

Castellana/ Iberica

"Altomira"

Guadarrama 1/ Neocastellana/ Betics activation SCS

Guadarrama 2

Change from pure compression to strike slip and tension

Rodanica/ Iberomanchega 1

Iberomanchega 2

Tectonics

Chapter 2 Cenozoic tectonic evolution of the Iberian Peninsula

Figure 2.2.3.2 Tentative correlation between stratigraphy in the large Cenozoic basins of Central Iberia, tectonic events (rightmost column) and stress fields. Several events can be recognized in major part of the basins (e.g. the breaks in sedimentation in the Middle and Late Miocene), but regional events can be observed as well (e.g. the absence of the Aquitanian “Altomira” event in the Ebro Basin). Based on a Figure by Calvo et al. [1993]. Stress field results added from Gutierrez et al. [1986]; Villena et al. [1996]; Mediavilla et al. [1996]; De Vicente et al. [1996]. Abbreviations in Figure: all. = alluvial, lacus. = lacustrine, evap. = evaporitic.


Chapter 2

A major problem in reconstructing stress fields in Iberia at larger scale through time is the age of observed deformations. Dating of a specific sedimentary interval containing the kinematic indicator data is in many of the Tertiary basins very difficult because of the lack of good biostratigraphic markers in their predominantly continental sedimentary fill. Rodent teeth dating (see Daams et al. [1996] and references therein) combined with magnetostratigraphy [Krijgsman et al., 1994] have provided the only direct dating method in continental deposits in limited areas (the Loranca Basin, the Teruel-Catalayud Basin and the Vallès-Penedès Basin [Garcés et al., 1996]) and only for the Middle to Late Miocene record [Krijgsman et al., 1996]. These studies should be expanded to other time intervals and basins in the Iberian Peninsula. In the larger Duero, Tajo and Ebro basins, the continental sediments along the basin margins grade towards the basin center into lacustrine sediments that can be dated. Lateral correlation to the continental sediments provides a widely applied method of indirect dating. In the numerous smaller, isolated intramontaine basins (e.g. Bierzo, Guadiana, Zaorejas, Avila and Lozoya) where datable lacustrine sediments are scarce or absent, the absolute dating of the Tertiary (?) sediments is poor. As a best approximation these sediments are correlated based on their nature (fluvial; distal or proximal) and lithology (red conglomerates, silts, sands and clays) with better ageconstrained series of larger basins (e.g. Martín Serrano et al. [1996]). Fortunately, the littoral basins (e.g. Emporda, Maestrazo, Vallès-Penedès, Lusitania) and the basins within the Subbetic and Betic Cordillera (e.g. Vera, Sorbas, Nijar) are characterized by several marine ingressions and, therefore, dating is much more precise in these regions. Yet, even if sediments and stress field changes can be accurately dated (e.g. Huibregtse et al. [1998]; Stapel et al. [1996b]), correlation of deduced stress tensors over larger distances is not very straightforward. A compilation of the paleostress data along a NESE oriented ‘profile’ (see Figure 2.2.3.3) in the Iberian Peninsula shows that the stress field was far from homogeneous and underwent several considerable changes during Tertiary to recent times. This is yet another reason why correlation of ‘tectonic phases’ between regions based on local stress-fields is difficult. Several authors have, however, succeeded in reconstructing regional (paleo)stress fields for Iberia and surrounding areas over limited time intervals. Examples are (1) Eocene-Oligocene of the Ebro basin and Catalan Coastal Ranges (e.g. Guimerà [1984]), (2) Late Miocene to present-day in the Iberian Chain [Simón-Gómez, 1989], (3) Middle Miocene to present-day in the Madrid and Loranca basins [De Vicente et al., 1996b], (4) Middle Miocene to presentday for the Betic-Rif Cordilleras [Galindo-Zaldívar et al., 1993], (5) present-day on- and offshore Portugal [Ribeiro et al., 1996], (6) Eocene to Miocene of the Alpine foreland [Bergerat, 1987], (7) present-day Iberia [Andeweg et al., 1999a], based on limited information (focal mechanism solutions, borehole breakout data and Quaternary fault slip data) by e.g. De Vicente et al. [1996b] and (8) present-day Iberia [SIGMA, 1998]. Updated with, amongst others, information that will be presented in Chapter 3 of this thesis, a more up to date compilation of the available data for the recent and present-day stress field in the Spanish part of the Iberian Peninsula has been completed. Encompassing an enourmous number of over 105.000 fault slip and 156 focal mechanism solution data (including results of work presented in this thesis), within this Spanish national compilation two independent maps have been constructed using either of both sources of information. The resulting trajectories of Shmax do not differ much from the early version by Andeweg et al. [1999a]. By constructing two independent maps based on either source of information (Focal mechanism and fault slip data), the resulting differences of stress trajectories could be compared. At the scale of the entire Iberian Peninsula differences between the stress patterns turned out to be of minor 17

155 UNIAXIAL COMP.

STRIKESLIP

Jadraque (SE SCS) This thesis and De Bruijne et al. (2001)

N-NNE UNIAXIAL EXT.

N-NNE

1

110 STRIKE SLIP

STRIKESLIP

NE

1

SSE-S E-ESE

RADIAL EXT.

Ma

65.0

33.7

54.8

28.5

23.8

20.5

14.8 16.4

11.2

7.1

5.3

Eocene Paleocene

Aquitanian

Burdigalian

Langhian

Serravallian

Messinian

Pliocene

QUATERNARY

1.8

AGE

(NOT TO SCALE)

0

025

RADIAL EXT.

025

Sierras Exteriores Oscenses (N. Ebro) Sancho Marcen (1990)

NNE

NEOGENE

Central Ebro Basin Gutierrez(1996)

36N

NON ACTIVE TECTONICS ?

Tortonian

Alcaine (S. Ebro) Casas Sainz and Simon Gomez (1986) 4E

UNIAXIAL RADIAL COMP.

=

145

1

RADIAL EXT.

STRIKESLIP

Mijares Basin (Teruel) Paricio Cardano & Simon Gomez (1986) 4W

STRIKESLIP

NW

Almazan Basin Maestro Gonzalez & Casas Sainz (1995) 8W

STRIKESLIP

N-S comp. + E-W ext

3Ma

Teruel and Ebro Basins Simon Gomez (1989) Bardena Negra (C. Ebro) Gracia Pietro & Simon Gomez (1986)

Miocene

Oligocene

PALEOGENE

RADIAL EXT.

Siguenza (junction SCS and IC) Alvaro (1975) 12W

STRIKESLIP

150 STRIKESLIP

Alto Tajo (IC) Rodriguez Pascua et al. (1994) 38N

060

S-border of SCS Giner Robles & De Vicente (1994) 40N

STRIKESLIP

NO ACTIVE TECTONICS OR STRIKESLIP

STRIKESLIP

UNIAXIAL COMP.

Sierra Altomira Munoz Martin (1998)

RADIAL EXT.

150

150 STRIKESLIP

155 STRIKESLIP

Toledo Mountains Martin & De Vicente (1995) UNIAXIAL EXT.

NON ACTIVE TECTONICS ?

SW border of IC Munoz Martin (1993)

UNIAXIAL EXT.

UNIAXIAL COMP. 100 =

STRIKESLIP

NO ACTIVE TECTONICS ?

42N

STRIKESLIP

Southern Altomira Manera Bassa (1981) STRIKE SLIP

UNIAXIAL COMP. 055

NE Valencia province Martinez Gallego (1987) RADIAL EXT.

STRIKESLIP

STRIKE- UNIAXIAL SLIP COMP.

075

030

Alicante fold-belt De Ruig (1992)

44N

STRIKESLIP

Betic foreland (Ciudad Real) Vegas & Rincon (1995)

RADIAL EXT.

065

150 STRIKESLIP


SSW

Chapter 2

Figure 2.2.3.3 Tentative correlation between several paleostress studies in the Eastern Iberian Peninsula, arranged in a profile-like way from north (Southern Pyrenees) to south (Betic foreland). Note that correlation between stress fields is restricted to regions. In other words: different states of stress coexist in time for different regions in the Iberian Peninsula. This illustrates the need to put the paleostress data in regional and temporal context to understand the development of the regional stress fields. Data from Sancho Marcen [1990]; Gracia Prieto & Simon Gomez [1986]; Casas Sainz & Simón Gómez [1986]; Paricio Cardona & Simón Gómez [1986]; Martín Velázquez & De Vicente [1995]; Manera Bassa [1981]; Vegas & Rincón [1996]; Giner Robles et al. [1994]; Rodriguez Pascua et al. [1994]; Muñoz Martin [1993]; Muñoz Martin [1997]; De Ruig [1991]; Alvaro [1975]; Gutierrez [1986]; and Andeweg [this thesis, Chapter 3]. 1) = permutation of principal axes of stress likely. 18


Chapter 2

importance [SIGMA, 1998]. The minor differences between both results for the recent and present-day stress field, suggest once more that the use of fault slip data for the reconstruction of paleostress fields in the geological history can provide stress trajectory maps with reasonable results at larger scale. In this way, the results from a large database of fault slip data can provide indirect constraints on plate boundary processes that were active through geological history. In general, however, it is clear that throughout the Tertiary the 1. Pyrenean or ‘Iberian’ phase Paleogene- Early Neogene paleostress field in the Iberian N-S to NE-SW compression Peninsula must be related to the following active first order far field 4. Ridge Push from stress sources (see Figure 2.2.3.4): the Atlantic (1) Collision of Eurasia and Iberia, (M. -L. Cretaceous 3. Valencian phase till present day) resulting in the development of the E.Miocene - Pliocene Pyrenees along the north of the WNW-ESE extension Iberian mainland (Paleogene - E. Miocene), (2) Collision between Eurasia (Iberia being incorporated) 2. Betic phase and Africa through the Alboran M. Miocene-recent NNW-SSE compression microplate, creating the Betic Cordillera in the southern part of Iberia Figure 2.2.3.4 (M.Miocene – present-day), (3) Although temporal and spatial differences occur between Extension to the east of Iberia local and regional states of stress, the development of resulting in opening of the Valencia the stress field in the Iberian Peninsula is related to four Trough (Oligocene – M. Miocene, major sources of stress, overlapping in time and place (see text). The resulting changes in the intraplate stress resuming in Pliocene times), (4) have caused polyphase fault reactivation under different Progressive opening of the Atlantic to stress fields, basin formation and deformation, vertical the west of the Peninsula (Late motions, erosion and sedimentation. Mesozoic – present-day). From these different sources stresses were transmitted to the interior of the relatively small Iberian Peninsula. Some of the plate boundary processes were active only during a (relatively) short time span and changed in sign, magnitude and direction rapidly. This resulted in temporal and spatial superposition of the stresses leading to local stress fields with Shmax orientations changing within several hundreds of kilometers. The Sierra de Altomira presents a nice example of this principle of superposition of far field stresses. This N-S trending fold-and-thrust belt in Central Spain (see Figure 2.2.3.1) is a somewhat enigmatic feature: while the overall tectonic setting in the Iberian Peninsula was N-S oriented compression, the Sierra Altomira extruded towards the west from Late Oligocene – Early Miocene [Muñoz Martín, 1997]. The active period of this structure is well constrained to the latest Oligocene-earliest Miocene: Upper Oligocene sediments deformed in the structure, Lower Miocene sediments that onlap the Altomira structures seal the structure [Gomez et al., 1996]. During the onset of deformation the Pyrenean collision was still active, resulting in NNE-directed compression in at least NE Iberia [Guimerà, 1996]. For the latest stage of deformation (Middle Miocene) the approximately NNW Betic compression is inferred to have been the dominant stress field [De Vicente et al., 1996b]. The extrusion to the west of the Sierra de Altomira occurred under N100 directed compression, documented as the "Altomira" stress field [Muñoz Martín et al., 1998]. The local geological setting facilitated this extrusion, faulting and folding. A basement fault was active as a normal fault during the Mesozoic, resulting in a 19

Chapter 2


western culmination of a decollement level, the plastic facies of the Upper Triassic (Keuper) [Querol, 1983]. In the subsequent compressional stage both features (the basement step and culmination of the decoupling layer) generated stress concentration at the tip of the basement fault ([Van Wees et al., 1996]; [Muñoz Martín, 1997]) and decoupling was restricted to the downthrown block only (where Keuper facies was present). Finite-element studies have reproduced the N100E compression required for the observed deformation in the Sierra de Altomira region by applying a remnant of the Pyrenean compression from the north and the onset of Betic compression from the south [Muñoz Martín et al., 1998]. This explanation of the development of the Altomira structure points to a gradual change in the active plate boundaries through time, during which superposition of the Pyrenean, Atlantic and Betic stress fields induced a locally anomalous ("Altomira") stress field in central Iberia. The thickened crust in the Pyrenean region might have played an important role in this deformation as well. Just as in the present day the influence of the Pyrenees on the local stress field of the NE peninsula can still be observed.

2.2.4 Conclusion In spite of their low ‘confidence’ ranking by the WSM-classification [Zoback, 1992], and the shortcomings mentioned of the fault slip method, many regional field studies have shown fairly consistent patterns of stress orientations (for Iberia e.g. De Vicente et al. [1996b]; Guimerà [1996]; Simón-Gómez [1989]). For the geological past, the fault slip inversion is the most adequate way to obtain information about the paleostress patterns, provided there is an adequate sedimentary record preserved. If abundant observations and results (to eliminate statistical errors in the measuring procedure) from a large region (to eliminate local effects) are being evaluated with care and compared with welldocumented active larger scale structures, the method provides valuable information [Bergerat, 1994]. A compilation of stress tensor results from kinematic indicators can be used to determine the temporal and spatial evolution of the stress field of a larger area and these changes of local stress fields and patterns of intraplate deformation provide information on plate boundary processes.

2.3 Finite element modelling of intraplate stress Over the last decades numerical modelling of many kinds of tectonic processes has been applied on a wide range of geological problems (see for a review Cloetingh et al. [1998]). This range covers amongst others stress (e.g. Bada [1999]; Gölke [1996]), strain (e.g. Janssen [1996]; Spadini [1996]), extension (e.g. Van der Beek [1995]; Ter Voorde [1996]) compression, basin evolution (e.g. Lankreijer [1998]; Van Wees [1994]), sedimentation (e.g. Den Bezemer [1998]) and P-T-t modelling (e.g. Van Wees et al. [1992]; Willingshofer [2000]). Depending on the purpose of the study, different numerical concepts and techniques have been applied. The most adequate method to calculate stress and strain in a given material is the finite element method (FEM). FEM is a matrix algebraic method able to compute the behavior of complex geometries composed of multiple materials that are subjected to varying boundary conditions. The method was developed early last century primarily to calculate stress and strain in civil engineering constructions, but did not become widely used since it involves inversion of matrices, which is time consuming even for small matrices. With the advance of computer power, matrix equations could be solved on small time scales, leading to an enormous development of applications for FEM. The mathematical description and the fundamental equations that constitute the matrix inversions have been described

20


Chapter 2

concisely by amongst others Zienkiewicz & Taylor [1988a] and Zienkiewicz & Taylor [1988b]. Beekman [1994] discussed how the method could be used for geological deformation processes (bending, faulting and friction). Interested readers are referred to these studies and references therein for theoretical background. The first finite element models to calculate patterns of stress in the Earth’s crust were those by Richardson et al. [1979] in the late seventies. The models applied elastic deformation, encompassed the entire Earth and did not allow for much detail. To a first order, the modelled patterns of stress could be related to the major driving forces of the plates, in the modelling applied as boundary conditions at the edges of the plates. Vilotte et al. [1982] have been using a viscoplastic medium to apply the same concept to plate wide scales (e.g. the Indian-Asian collision). Wortel [1980], Cloetingh et al. [1982], and Cloetingh & Wortel [1986] were among the first to include more physical aspects of the forces applied to the model by considering the ridge push forces as an integrated pressure gradient due to the cooling of the lithosphere (see appendix A). These boundary forces were applied to a model with a uniform plate composed of homogeneous elastic material for the Cocos-Nazca plate and the Indo-Australian-plate. Since then, models have become more complicated, in terms of including increasingly more free parameters and concepts, like variations in material properties [Grünthal & Stromeyer, 1992], inclusion of weakness zones [Mantovani et al., 2000], and different representations of applied boundary conditions and rheology. Hu et al. [1996] for example included elasto-plastic deformation in a geometrical complex model of Taiwan. Potential energy differences induced by lateral density variations have been incorporated in the modelling by amongst others Coblentz et al. [1994]. In this way, buoyancy forces related to lateral density variations and crustal topography were introduced in the modelling, using the age-depth relationship of cooling oceanic lithosphere [Parsons & Sclater, 1977]. Finally the step was made from 2D to 3D models: Gölke et al. [1996] presented a three dimensional model for the Norwegian margin and Ragg et al. [1999] did so for Sicily including domains with different material parameters. Whereas these previous models were able to reproduce the directions of maximum horizontal compression on a large scale, the resulting local state of stress in many cases was not consistent or just not compared with observations. In the present research, the purpose is to simulate the stress fields that led to the significant intraplate deformation in the Iberian Peninsula since the Tertiary and to link deformation with vertical motions detected by fission track analysis ([De Bruijne & Andriessen, 2000; De Bruijne, 2001]). To be able to assess proximity to failure along faults with a given orientation, a better fit between the observed and modelled states of stress is required. It should be noted that the purpose of the modelling presented in this thesis is not to mimic the real world, but to study physical processes behind the observations in the geological past. Numerous parameters that have been incorporated in previous studies (e.g. crustal composition, crustal thickness, basal drag, detailed crustal configuration etc.) cannot be constrained for the present-day, let alone for the geological past. Therefore, a model is applied with only a limited amount of free parameters, which facilitates direct understanding of the effect that different plate boundary conditions have on the resulting stress field. Furthermore, the larger the set of free parameters included in a model, the harder it becomes to draw conclusions from the model results, due to the problem of non-unique solutions. This being said, numerical modelling can provide valuable insight in geological processes and enables quantitative testing of hypotheses posed on these problems. 21

Chapter 2


To be able to compare the results for the present-day stress with the paleostress field, the models in this thesis are kept as internally consistent as possible. A few parameters that have been incorporated in previous studies can hardly be constrained for the present-day, let alone for Tertiary times. Amongst these are: (1) Basal drag along the base of the plate as it moves in an absolute reference frame over the mantle. This basal drag generally is calculated by summing the forces acting on a plate, which leads to the torque of this sum acting on the plate [Richardson, 1992]; [Meijer, 1995]. The sum of the torque and the basal drag shear forces should be equal to the absolute plate motion. For the present-day situation absolute plate motion is inferred from global models such as the NUVEL-1 model [DeMets et al., 1990]. Absolute plate motion for Western Europe is extremely small (5mm/year), so it is hard to distinguish between the contributions of collisional forces and ridge push forces [Richardson, 1992]. It is, however, arguable whether basal drag calculated for the entire Eurasian plate is representative for the Iberian part of it. During the period that Iberia acted as an independent plate, basal drag might even be disregarded anyhow because small plates can be considered to be mechanically decoupled from the mantle [Melosh, 1977]. Moreover, for the reconstructions of the Iberian plate, absolute motions through time [Gordon & Jurdy, 1986] are not well documented and the estimates depend on the reconstruction model used and the relative motion vectors derived from the different rotation poles. Even for the present-day situation, differences between models observing the direction and amount of absolute plate velocity result in different values for the basal drag, see e.g. Meijer [1995]. In future studies, when more data is available on absolute motions through time, inclusion of the basal drag would help to constrain the magnitudes of the different plate boundary processes even further. (2) The contribution of circulation of mantle lithosphere [Wuming et al., 1992]. First order estimates from studies of large scale lithospheric stress induced by global mantle circulation [Steinberger, 1999] show that non-hydrostatic lithospheric stress might vary between –140 and 140 MPa. For the Iberian Peninsula values range between the low values of +40 and 0 MPa [Steinberger, 1999]. As in the case for basal drag, information of lithospheric thickness variations through time is very limited and it would be very speculative to incorporate this. Therefore it has not been included in the modelling. (3) Zones of weakness, such as faults. A fundamental problem is to what extent faults should be incorporated in the modelling. Should these be only the seismogenetic ones, which show seismic activity under the present stress field? Or all observed and inferred basement faults, which will constitute weakness zones as well but might not be activated under the present-day stress field? The definition of seismotectonically active faults is not straightforward, although attempts are made to generate databases on them (e.g. FAUST project, http://faust.ismes.it/). Even if it is possible to incorporate major faults, should not all of the small active faults be incorporated as well? The bulk deformation from very limited displacements along smaller faults is reported to be just as important as the strain accommodated by the larger structures (e.g. Sheridan [1998]). An additional major problem arises when considering the setting for geological times: how to proceed when going back in geological time and willing to determine the amount of activity of inferred faults? Until better methodologies are developed to overcome these problems, not incorporating faults seems to be the best solution for the large-scale modelling presented in this thesis.

22


Chapter 2

2.4 Stresses induced by lateral density variations Forces acting on tectonic plates can be subdivided into forces that act on the plate boundaries (e.g. collision forces) and forces acting within plates (e.g. buoyancy forces). Gravitational body forces (i.e. buoyancy forces) play a major role in intraplate deformation e.g. Richardson [1992]. These forces are the result of lateral variations in density. Any difference in density might arise from changes in crustal thickness or changes in the density of crustal material. Major lateral variations in density (see Figure 2.4.1) occur for example at (a) continental margins, (b) spreading mid-oceanic ridges, and (c) highly elevated regions. (a) At (passive) continental margins, a significant decrease of topography/bathymetry and crustal thickness towards the oceanic basin are a common result of rifting [Reemst, 1995]. This results in continental crust next to water along the surface (large density difference) and continental crust next to mantle material at deeper levels (small density difference). Ocean Ridge

w b z

Ocean Basin

Sea Water

Oceanic Crus

t

a Mantle Lithosphere

Elevated Continent

Continental Margin

w b

w c

m

m

Sea level

Continental Crust

m Equipotential

Astenosphere

Values used to determine TRS

Tl = temperature at base lithosphere Ts = temperature at surface hw = waterdepth above MOR hb = oceanic crust thickness at MOR w = seawater density b = oceanic crust density a = astenosphere density at Tl av = coefficient of thermal expansion ziso = depth of equipotential surface

c

o

1280 C o 0 C 2.5 km 7.0 km 3 1030 kg/m 3 2960 kg/m 3 3238 kg/m -5 o 3 * 10 / K 125 km

Figure 2.4.1 Figure showing four different lithospheric columns in local isostasy that have a potential energy difference with respect to each other. The difference in potential energy between a local lithospheric column and some column defining the reference tectonic state (TRS) is calculated by integration of the vertical stress in a lithospheric column, conform Coblentz et al. [1994]. The difference in potential energy between two columns divided by their distance defines the forces due to lateral density variations. The included table shows values that have been used in this study to determine the Tectonic Reference State.

(b) The spreading of an ocean basin involves uplift of a mid-oceanic ridge at which asthenosphere derived melts well up to the sea floor and upon cooling are accreted as oceanic crust to the diverging plate margins. The young crust is cooling, becomes denser and tends to sink creating abyssal plains, deep oceanic basins [Parsons & Sclater, 1977]. The age-depth relationship of oceanic crust is explained briefly in section 2.3.2 and more detailed in Appendix A. The so-called ridge push force is the (passive) effect of this cooling oceanic lithosphere rather than an active push by up-welling astenospheric material (see e.g. Bott [1991, 1993] and Ziegler et al., [2001]). The boundary between asthenosphere and lithosphere is defined as temperature dependent, moving away from the ridge, the new-formed material is cooling and, therefore, the lithosphere is thickening. Apart from the density difference between young oceanic

23

Chapter 2


lithosphere and water, the change in lithospheric thickness has a large effect on the density distribution. (c) In the case of highly elevated regions, kinetic energy that has been used to construct a pile of thrust sheets/crustal thickening is being stored as potential energy in (crustal) topography. Highly elevated regions are most often supported by a crustal root, meaning that at depth light (crustal) material replaces dense (mantle material), while along the surface air is next to dense (crustal) material in a mountain range. The presence or absence of a lithospheric slab under a mountain range will have important effects on potential deviatoric stresses [Bott, 1991].

2.4.1 Concept and calculation Consider a lithospheric column in isostatic equilibrium (see Figure 2.4.1.1, column 1). The vertical stress in the considered column is given by: Vertical stress = g[ρm (ziso - zm ) + ρc (zm – h)] where g is the gravitational acceleration, ρm and ρc are the density of mantle and crust material, ziso is the equipotential level, h is topography, and zm is the crust-mantle boundary. Column1

Thickened crust Column2

Vertical stress = g[ml (ziso-zm)+ czm-h)]

ml

1

h

c

3 4

Column2

h 1 0 2 z1

zm

z2

ziso

depth (z)

2

vertical stress (zz)

3

Area= PE difference between both columns 4

Column1

Area= PE of column1

Figure 2.4.1.1 To illustrate the concept of potential energy differences: column 2 has a higher amount of vertical stress than column 1 (gray area below line difference with dotted area) which results in a higher Potential Energy per unit area for column 2. Divided by the distance between both columns, a force pointing down slope would be resulting in tension in the thickened crust and compression in the ‘normal’ setting.

This vertical stress is through gravity directly related to the amount of material above zm and the density of this material. The potential energy (per unit area) of the specified column is defined as the integral of the vertical stress in the column. Another lithospheric column that is in isostatic equilibrium as well, but with another density distribution (column 2), might therefore have a different amount of potential energy. Thus, in Figure 2.4.1.1b, the curves denote the vertical stress of two columns and the area below these curves represents the potential energy. Differences in potential energy (shaded area in Figure 2.4.1.1b) are directly related to differences in density distribution between the columns. Divided by the distance between the considered columns, the potential energy becomes a force (per unit area). Obviously, stresses induced by lateral density variations will be largest in areas with large changes in crustal configuration whereas these forces will be oriented perpendicular to the gradient of change.

24


Chapter 2

To evaluate differences in potential energy between areas, a Tectonic Reference State (TRS) has to be defined. This TRS is a column that is supposed to be in isostatic and stress balance with mid-ocean ridges and will not be deformed in the absence of external forces. In other words: in the absence of plate tectonic forces, this is the setting the crust would finally equilibrate to. Different types of TRS have been applied in former studies. [Jones et al., 1996] adopted an astenospheric column with density ρa with its top 2.4km below sea level and no water or lithosphere above it. Based on the mean global potential energy, a column of 30km continental crust (density contrast with mantle of 488 kg/m3) at sea level equivalent to cooling oceanic lithosphere at 4.3km below sea level would represent another useful TRS [Coblentz et al., 1994]. Differences between the alternative reference states would be of the order of 1012N/m [Jones et al., 1998]. TABLE TRS shows the values used to define the TRS in this thesis (equivalent to Coblentz et al. [1994]). Another point of reference required for the calculations is an equipotential level at depth. Below this equipotential level, no compensation by density variations is assumed and therefore potential energy differences are absent. After Coblentz et al. [1994], who stated that the contribution to potential energy variations from lateral variation of density beneath a depth of 125km is negligible, I adopted an equipotential surface at this depth. This is equivalent to assuming that the thickness of the lithosphere is 125km and the asthenosphere is a homogeneous medium. The potential energy difference of an arbitrary column and the TRS can now be calculated. When the potential energy of the specified column is lower than the TRS, this column should be in deviatoric compression whereas a column with a higher potential energy should be in tension. In the absence of far field forces this would result in extensional forces dominating in thickened crustal regions versus compressional in low lands. Several authors have successfully applied the concept of stresses induced by lateral density variations to explain regional variations in large-scale (plate-wide) present-day stress fields. Examples are the Indo-Australian plate [Coblentz et al., 1995], the S. American plate [Richardson & Coblentz, 1994]; [Meijer, 1995], the western part of the Eurasian plate [Gölke & Coblentz, 1996], the Philippine plate [Pacanovsky et al., 1999] and the African plate [Coblentz & Sandiford, 1994]. The forces related to potential energy variation can regionally modify the stress field in great detail and change the settings under which a mechanical equilibrium is reached. Crustal thickness variations are important here and can be both the effect of a plate wide stress field, and cause a regional stress field. Two examples are (a) continental margins and (b) active crustal thickening. (a) Opening of oceans may cause reorganizations of spreading axes. However, in most cases present-day margins will be oriented approximately perpendicular to the ridge push forces because passive margins are the result of crustal separation and ensuing sea-floor spreading. This is for example the case at the Norwegian margin, which is oriented perpendicular to the regional stress field. In this region, the introduction of forces related to lateral density variation (extension parallel to the regional compressive stress field) results in near anisotropic stress circumstances [Gölke et al., 1996]. (b) In a zone of active thickening of continental, the bulk of thickening will occur perpendicular to the Shmax direction, leading to a major gradient in lateral density variations parallel to the regional Shmax direction. Kinetic energy is stored as high levels of potential energy in the thickened region. At a certain moment the extensional forces induced by the increasing difference in potential energy between mountain range and foreland can locally become larger than the compressional forces leading to the deformation (see Figure 2.4.1.2). So, coaxial stress tensors can be expected during tectonic history. Theoretically, this imposes a maximum altitude for thickened continental 25

Chapter 2 Regional and local forces


crust, based on (1) the magnitude of far field forces and (2) the material the crust is composed of. To what extent extensional features develop or even Resulting stress regimes lead to the ‘extensional collapse’ of a mountain range (see e.g. [Zhou & Sandiford, 1992]), depends largely on the Regional and local forces balance between far-field stresses and the stresses induced by the lateral density variations and therefore reveals information about the magnitude and orientation of the far-field stresses and Resulting stress regimes thus, indirectly about activity of plate boundary processes. Examples of limited extension on top of large actively contracting mountain ranges have been Figure 2.4.1.2 observed in (amongst others) the Andes Figure demonstrating how the local stress patterns [Meijer, 1995] and the Himalayan belt are combined results of far field stresses and [Colchen, 1999] and have been observed induced dynamic stresses. A mechanical equilibrium for the tectonic evolution of the Alpine is reached for both of the examples. In the lower panel are the induced dynamic tensile stresses larger range as well (Gosau basins than the regional compression, resulting in local [Willingshofer, 2000]). Complete extension in the mountain range. extensional collapse due to high levels of potential energy stored in zones of thickened crust can occur when the far-field stresses responsible for the build up of the orogenically thickened zone diminish by a decrease in convergence rate, rotation of the main compression direction or slab break off [Bott, 1993]. This concept has been proposed as ‘extensional collapse’ for amongst others, the Betic and Alboran region [Vissers et al., 1995]. Jones et al. [1998] linked potential energy driven extension of the Laramide orogeny (North America) to paleo-elevations of the region and calculated that elevations required for extensional collapse to occur would have had to reach over 3500m. Bada [1999] was the first to incorporate estimates of reconstructed crustal thickening in the geological past in numerical modelling of paleostress fields of the Pannonian Basin (see also Bada et al. [2001]). By comparison of modelling results with careful reconstruction of the paleostress field, these authors were able to estimate minimal crustal thickness in the Eastern Alps that would be required for extensional collapse of this region. The main uncertainties in the potential energy calculations are related to: (1) Density distribution. Especially in continental crust, which is far from homogeneous, this is a major uncertainty. If crust in a region consists of very dense material, whereas average density of continental density is used in the calculations, this will lead to an underestimation of the stresses induced by the structure. (2) Geometry of crustal thickness (Moho topography). Indirect methods for observing the transition from crust to mantle (seismic refraction and reflection) have only covered a limited extent of the world yet. In general, across mountain ranges quite a number of very well controlled geotransects have imaged the Moho very well. However, its continuation in any direction is not well resolved. (3) The influence of lithosphere thickness variations and the choice of an equipotential surface at depth. Although the density difference between asthenosphere and lithosphere is relatively small, the large volumes considered might have significant influence. In case the lower crust is decoupling the upper crust from the mantle, the forces that are related to this lithosphere topography would only have a very limited 26


Chapter 2

effect on the crustal state of stress. First order estimates from studies of large scale lithospheric stress induced by global mantle circulation [Steinberger et al., 1999] show that non-hydrostatic lithospheric stress might vary between –140 and 140MPa. For the Iberian Peninsula values only range from +40 to 0MPa. Figure 2.4.1.3 Predictions from numerical modelling of the stress field of western Europe after Gölke and Coblentz [1996]; model results for a reference thickness of 100km using a uniform elastic rheology. The panels display the predicted stress patterns for in Iberia and surrounding areas for different models. See reference and text for more detailed description.

The effect of stresses induced by lateral density variations is expected to be significant for the Iberian Peninsula as suggested by (a) large scale modelling of the European plate by Gölke et al. [1996], (b) crustal configuration in the region, and (c) seismic activity in NW Iberia, far from any active plate boundary. To motovate (a), note that finite element models on the stress field for the shelf of mid-Norway [Gölke et al., 1996] show second order tensile stresses that are large enough to create a local stress state with a small stress anisotropy. This means that the horizontal principal stresses are about equal in magnitude, which enables permutation of the maximum and intermediate principal axes. It should be noted that along the Norwegian margin, the decrease in thickness of the continental crust and the far field maximum horizontal compression are parallel. Therefore, the induced extensional forces counteract the regional compression. A large step in topography and crustal thicknesses along the western and northern coast of Iberia, similar to the mid-Norwegian shelf, causes potential energy driven tensional stresses in the same order of magnitude (about 15MPa, [Gölke & Coblentz, 1996]) as for the Norwegian margin. Results of Gölke & Coblentz [1996] show that if stresses generated by distributed ridge push and buoyancy forces are incorporated, the orientation of Shmax does not change significantly for continental Iberia with respect to the model without buoyancy forces (see Figure 2.4.1.3). The local state of stress in Iberia (ratio between the principal stress-axes), however, is altered very much as well by both reduction of the Shmax magnitude (see Figure 2.4.1.3, panel I versus panel IV) as well as by increasing importance of the other principal stresses (uniaxial compression

27

Chapter 2


versus strike-slip regime). For the oceanic part, it is clear that incorporation of these forces (compare Figure 2.4.1.3 panel II with panel III and IV) leads to a dramatic change in stress field orientation and magnitude. (b) Since convergence between Africa and Eurasia started in the Campanian (see Chapter 4 for detailed outline), Iberia was progressively subjected to compressional stresses between both plates. Additional compression from the west, related to ridge push forces of the Mid Atlantic Ridge, hindered lateral escape to the west and even added to high levels of intraplate stress. Stress was released by internal deformation: the Iberian Peninsula shows significant internal deformation, has a thickened crust (average at about 30-35 km), relatively elevated average topography (400 m, [Stapel, 1999]), and shows evidence for folding at lithospheric scale [Cloetingh et al., 2001]. Along the northern and western margins of Iberia steep gradients in crustal thickness changes occur. (c) Most seismicity in Iberia (see Figure 2.4.1.4) can be related to active plate boundaries (Mid Ocean Ridge, Azores, African-Eurasian boundary) or weak zones related to former active plate boundaries (Pyrenees/Betics). However, from November 1996 until May 1997 an area of approximately 20km2 near Lugo, located in Galicia (NW Iberia) was struck by over 250 seismic events ranging in magnitudes from 3.0 to 5.1 (Mb). Whereas these magnitudes are relatively low, the seismic activity in this region is remarkable when taking into account the large distance to any of the active plate boundaries. Focal mechanism solutions of this pronounced seismic activity indicate active normal faulting along roughly N-S trending faults ([SIGMA, 1998] and online databases Instituto Geográfico Nacional, or Harvard CMT Catalog, http://www.geo.ign.es http://www.seismology.harvard.edu/CMTsearch.html). The orientation of the regional present-day compressional stress is at large angles to the western and northern margins. An additional source of sub-parallel extensional stresses related to lateral density variations along both margins could modify the local state of stress in such a way that normal motion of the N-S oriented faults can be explained. In this respect it is remarkable that the concentration of seismicity occurred at the cross point of (weak) seismic alignments along the northern and western coast. To test this hypothesis on the occurrence of the seismicity in northwestern Iberia and to enable a better estimate of the local state of stress for regions in Iberia, the concept of lateral density variations has been incorporated in the modelling. Combined with local information on the crustal configuration (pre-existing faults, weakness zones), the proximity to failure of the crust can be evaluated, enabling a more accurate evaluation of seismic hazard.

2.4.2 Testing the model-sensitivity for various input parameters This section presents the effects of a more data-oriented approach by (a) a comparison between the effect of using filtered observed bathymetry of the ocean floor off shore Iberia or using the theoretical age-dependence curve, and (b) the effect of sampling independent data on crustal thickness from Bouguer inversion instead of using the concept of local isostasy to calculate the crustal thickness as a function of surface topography. Especially the data used as input and the methods to define crustal configuration of the concept could have important effect on the magnitudes of the induced stresses [Andeweg et al., 1999b]. E.g., in previous work by Coblentz et al. [1994] and Meijer [1995], the theoretical age-dependent depth of the oceanic crust has been used, in combination with a crustal thickness derived from sampled topography combined with the concept of isostasy.

28

10

28W

5

15

24W

20

30

20W

40

16W

50 100 600

12W

8W

2.0 3.0 4.0 5.0 6.0

Magnitude

4W

Figure 2.4.1.4 Map showing distribution, magnitude and focal depth of earthquakes from 1980-1996. Based on data by SIGMA [1998] and online databases Instituto Geográfico Nacional, (http://www.geo.ign.es) or Harvard CMT Catalog http://www.seismology.harvard.edu/CMTsearch.html). Major part of the seismic activity is related to active and former plate boundaries (Mid Ocean Ridge, Northern Africa and Pyrenees, Betics respectively).

32W

36N

40N

44N

0

Focal depth (km)

0E

4E

Cenozoic tectonic evolution of the Iberian Peninsula Chapter 2

29

Chapter 2


(a) Bathymetry The above-cited references use the theoretical age-dependent curve to determine the bathymetry of the ocean floor. After sampling the age of oceanic crust from a global ocean floor age database [Mueller et al., 1995], the depth of oceanic seabed (d) has been calculated according to the theoretical age-depth d(t) curves by Parsons & Sclater [1977] d(t) = 2500 + 350 √t for oceanic lithosphere younger than 70Ma, and d(t) = 6400 – 3200 (-t/62.8) for older ocean floor. Subsequently, this theoretical bathymetry is used in the potential energy calculations. The limit between old and young oceanic lithosphere has been chosen as the 70 Ma isochron, coinciding with the change from one formula to the other. 0

-1000

-2000

-3000

-4000

-5000

D 32W/59N

16WE/59N D’

C 38W/54N

-6000

20.5W/53.5N C’

0 -1000 -2000 -3000 -4000 -5000 -6000

11W/47.5N B’

B 32W/48N

A 30.5W/41N

10W/39N A’

60N

D’

40

20

20

D

-7000

40

C

C’

55N

50N

20

40

80

20

B’ 60

40

B

45N

A

A’

40N

80

60

100

35N

50W

30

40W

30W

20W

10W

0

180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 0

Figure 2.4.2.1 Four bathymetric profiles from the Mid Atlantic Ridge to the east show a large difference between observed depth (black dots) of oceanic crust (upgoing curves to the right due to arrival at the continental margin) and the theoretical curves (gray dots) as calculated using the age-depth relations by Parson & Sclater [1977]. Only for profile B-B’ the fit is acceptable. In the other profiles a difference of over 1000 m is normal. Lower panel shows location of profiles and age of the oceanic crust in the Northern Atlantic Ocean. Source of bathymetry: Müller et al. [1995].

Age (Ma) of oceanic crust (every 20Ma labeled in figure)


Chapter 2

Applying the theoretical curve, the mid-ocean ridge (where age of the crust is 0 Ma) would be at 2500m below water surface. In the case of the eastern North Atlantic however, the ridge deviates significantly from this theoretical depth. At the Azores triple point, the Mid Atlantic Ridge reaches up to 1400m under the water surface, further north the ridge shallows progressively to reach the surface at Iceland. It has been known that this curve almost always predicts toodeep depths for old ages as well [Stein & Stein, 1992] Furthermore, parts of the oceanic basin have been deformed during the Tertiary, for example Kings Trough. Figure 2.4.2.1 shows four profiles running from the Mid Atlantic Ridge to the coast of Western Europe. Note the large difference between the observed and theoretical curves for all of the profiles, in extreme for the northernmost profile (more than 2000m difference). These differences have an important influence on the magnitudes of the ridge push forces.

(b) Crustal thickness In order to calculate crustal thickness according to the approach of Coblentz et al. [1994] the topography of continents has been sampled from topographic databases (e.g. ETOPO5, [NOAA, 1988]) and has been filtered with a low-pass filter taking a boundary wavelength of 100km, simulating regional isostasy. This approach will eliminate local effects and topographic features related to small-scale stress sources (e.g. flexure). Then, using the concept of isostasy and reference densities of 2700kg/m3 for crust and 3300kg/m3 for mantle, two slightly different equations (see appendix of Coblentz et al. [1994]) are used to deduce crustal thickness for continental margin (thinned continental) and elevated continental settings, respectively. Shore-line is the separator between thinned continent and elevated continent. An alternative approach to obtain information on crustal thickness would be sampling independent data, as for example gravity or deep seismic data. The best method would be to use direct observations from seismic determinations of Moho depth. Although over the last decades the number of seismic data points for Iberia has been increasing (e.g. Pulgar et al. [1997]; Jabaloy et al. [1995]), data coverage is still limited and distributed inhomogeneously. To construct a grid of Moho depths from these values would likely introduce large errors. Another method is using gravity data, which are available for the entire Iberian Peninsula. The distribution of dense or light material in the crust is reflected in its gravity signal. The free air anomaly is the deviation from the geoid, while the Bouguer anomaly reflects changes in density at depth since it is obtained by correcting the Free-air gravity anomaly for topography and bathymetry. Assuming a reference density for the continental crust and a reference depth, Bouguer data can be inverted to obtain an approximation of crustal thickness, for Iberia performed by Stapel [1999]. The method is internally consistent with the isostatic calculations because reference density and depth (2700kg/m3 and 30km, respectively) are equal to those applied to determine the TRS (see Figure 2.4.1). In case of plate-wide models with elements covering hundreds of kilometers, the use of a reference density and local isostasy seems valid. A study of isostatic residuals by Stapel [1999] shows that in major part of Iberia the isostatic residuals are not zero, which indicates that topographic loads and buoyancy forces are not in balance. As a result of this, crustal thickness is over- or underestimated (excess or deficient topography for the crustal thickness and the used reference density). One way to explain the complicated isostatic residuals are dynamic forces acting on the lithosphere as for example plate boundary forces or forces generated by convection in the mantle. One should keep in mind that for the determination of the isostatic residual a reference density is used, so variations in density could explian some of the misfit. 31

Chapter 2


Especially for regions that are far from local isostasy (high isostatic residuals) crustal thicknesses differ significantly, which will have large effects on the calculated stresses induced by lateral density variations.

Testing scenarios Figure 2.4.2.2 shows a flow chart of the modelling procedure, from construction of a model through the calculation of the forces induced by lateral density variations. The ANSYS © finite element package was used to create the model and calculate the resulting stress field. Potential energy per node was calculated using ‘Potenca’ (Andeweg, not published), a modification of the potential energy calculation program by Coblentz et al. [1994]. Several routines (A2P and P2A, Andeweg, not published) have been developed to communicate between ANSYS and Potenca. START

ANSYS

Digitizing model geometry

Including Potential Energy

Create model mesh

Export and transform element/node data

Assign material properties

Potenca

A2P Sampling at nodes

Calculation crustal thickness

Potential energy per node

Transform format forces data

P2A

Potential energy difference with surrounding nodes Forces induced by lateral density variations

Sum forces at node

Additional boundary forces or constraints

Calculate stress field Re-evaluate boundary conditions

Compare results with observations

No

Reasonable fit?

Yes!

FINISH

Figure 2.4.2.2 Flow chart of the modelling procedure. Stars denote procedures that will be varied in the different scenarios. See text for discussion.

To demonstrate the effect of the different approaches on the induced forces, the input data resulting from these will be used in the potential energy calculations is presented for an area of 27 x 7 degrees (see Figure 2.4.2.3), elongated perpendicular to the margin of NW Iberia. The area is chosen to include young and old oceanic crust, continental margin, and elevated continent. For all scenarios, the area has been meshed in the same way with elements with a typical side length of about 50km. In order to obtain realistic magnitudes of forces and thus to be able to evaluate the resulting differences with respect to observations, the scenarios have been tested on a model for the European plate (further details are described in Chapter 6). A realistic collision force (see Chapter 6) has been applied along the southern boundary (African-European collision) to be able to directly compare the effect of the induced forces on effective stress levels, on the state of stress and the orientation of the principal stress axes in the Iberian Peninsula.

Scenario Ia (Coblentz) Rationale: standard method used by previous studies based on Coblentz et al. [1994]. Description: bathymetry calculated as theoretical ocean floor depth and crustal thickness calculated from filtered topography (wavelength of 100 km) using local isostasy. Results: Using the theoretical ocean floor depth to calculate the induced forces yields a rather smooth pattern of ridge push forces towards the continent. This is not a surprise, since the theoretical curve deepens gradually. The steep continental margin induces important forces pointing in the opposite direction. Integrated, these forces are of lesser importance than the ridge push forces, but their local effect on stress magnitude and 32


Chapter 2

state of stress is significant (note strong decrease of maximum horizontal compression along the northern and western margins of Iberia, see Figure 2.4.2.4). The resulting pattern of local stress states fits very well to the observations. Effective stress magnitudes (the difference between the principal stress axes) in Iberia are modelled to be of the order of ~25MPa.

Scenario Ib (Ridge close to sea-level) Rationale: in reality the Mid Atlantic Ridge is closer to sea level than the theoretical value of 2500m. Description: bathymetry calculated as theoretical ocean floor depth but starting from a ridge at 1425m instead of 2500m, for the rest of the setting equal to scenario I. Results: the curvature of the mid-oceanic ridge is similar to that in scenario Ia, but the ridge is closer to sea level. This results in a larger difference with the defined Tectonic Reference State than in scenario1a. Therefore, the associated ridge push forces are larger, which decreases the effect of the stresses induced by the margin or on-shore crustal topography. To obtain a reasonable resemblance with the directions of the observed stress field, the collision forces along the southern boundary have to be increased by over 100% resulting in relatively high stress levels in the Iberian Peninsula of over 80MPa.

Scenario II (Real bathymetry) Rationale: the ocean floor offshore Iberia shows large-scale irregularities that differ significantly from the theoretical age-depth curve. Description: bathymetry and topography sampled from ETOPO5 database and filtered with a low-pass filter applying a boundary wavelength of 100 km, crustal thickness calculated using local isostasy and the filtered values for bathymetry/topography. Results: the complex physiography of the ocean floor (e.g. Kings Trough and Gorringe Bank) distorts the smooth pattern of forces acting towards the continent. Large forces are related to these features and due to their intermediate position between ridge and continent, the forces counteract each other to large extent. Subsequently, less integrated ridge push force is transmitted to the continent (see Figure 2.4.2.5) than for the previous scenario. Although the directions of the observed and modelled trajectories yield the best fit using this approach, only limited levels of resulting effective stress (on average ~15MPa) are being reproduced for the Iberian Peninsula. These relatively low magnitudes for the effective stress would most probably not be able to account for the observed deformation.

Scenario III (Inverted Moho) Rationale: calculating crustal thickness from topography using local isostasy implies a very important but uncertain assumption (see above). Using independent data on crustal thickness would be a more correct approach. Description: topography and bathymetry sampled from ETOPO5 and filtered with lowpass filter applying a boundary wavelength of 100 km, Moho-depth sampled from Bouguer anomaly inversion (during this testing only available for region 16W/1E/35S/45N). Results: higher levels of compression enter the Iberian mainland (effective stresses ~50MPa), stresses concentrate along the a-typical margins of the Iberian Peninsula (up to more than 100MPa), see Figure 2.4.2.6. The latter is the result of the presence of local bathymetrical highs (amongst others Gorringe Bank, Galicia Bank, Vigo Seamount) that show important isostatic residuals [Stapel, 1999], in other words are not supported by thickened crust. Regions with a large isostatic residual have a strong influence on the

33

34

Oceanfloor age (in MA) of the Atlantic offshore Iberia, used to calculate 1a/b

Filtered sea floor topography

Theoretical sea floor topography

m

II

I

Sampled moho topography

Note difference

Calculated moho topography

Calculated moho topography

Figure 2.4.2.3 Figure demonstrating the differences in input data for the different scenarios and their effects on the results of the potential energy calculations. (see www.geo.vu.nl/~andb/iberia for color version)

III

Bouguer inversion Moho dataset

Local Airy isostasy

2d

Local Airy isostasy

Crustal thickness determined by:

DATA OUTSIDE RED BOX INTERPOLATED FOR PLOT ONLY

m


Shmax trajectories (observations)

Maximum horizontal compression (Shmax)

Intermediate horizontal compression

Minimum horizontal compression (Shmin)

Figure 2.4.2.4 Model result for scenario I, including a modified theoretical bathymetry and an “isostatic” topography. Left panel: predicted patterns for principal axes of stress for Iberia and surrounding region. Right panel: predicted levels of effective stress in western Europe. (see www.geo.vu.nl/~andb/iberia for color version)

100MPa

80MPa 60MPa

50MPa 40MPa

30MPa 20MPa

10MPa


35

36

0

200 km




Minimum horizontal compression (Shmin) Figure 2.4.2.5 Model result for scenario II, including true observed bathymetry and an “isostatic” topography. Left panel: predicted patterns for principal axes of stress for Iberia and surrounding region. Right panel: predicted levels of effective stress in western Europe. (see www.geo.vu.nl/~andb/iberia for color version)

100MPa

80MPa 60MPa

50MPa 40MPa

30MPa 20MPa

10MPa


0

200 km




Minimum horizontal compression (Shmin) Figure 2.4.2.6 Model result for scenario III, including true observed bathymetry and “Bouguer inverted” topography. Left panel: predicted patterns for principal axes of stress for Iberia and surrounding region. Right panel: predicted levels of effective stress in western Europe. (see www.geo.vu.nl/~andb/iberia for color version)

100MPa

80MPa 60MPa

50MPa 40MPa

30MPa 20MPa

10MPa


37

Chapter 2


induced forces. In case of Gorringe Bank, the presence of dynamic forces uplifting the shallow oceanic crust are inferred [Souriau, 1984], which can be easily related to the present-day active plate boundary. This inversion of Bouguer data to obtain Moho depth shows huge misfits with the restricted amount of seismic observations of the Moho in the Alpine regions Iberia and the North Atlantic [Stapel, 1999]. For the strongly deformed areas a possible explanation is given: the method of Bouguer inversion can only produce smooth surfaces, while large scale offsets of the Moho have been observed (e.g. line ESCIBETICAS-1 [Jabaloy et al., 1995] and ESCI-N2 [Pulgar et al., 1997]). An additional reason for the misfit is the large deviations from assuming a reference depth of 30km and an average density for continental crust in Alpine deformed regions. The same pattern of fit and misfit is observed between the calculated Moho depth using filtered present-day topography and the concept of local isostasy and the Bouguer inverted Moho database. Fit is rather good in the central parts of Iberia, but misfits occur in the deformed regions and offshore.

2.4.3 General model results and conclusions: In general the orientation of Shmax shows a rather good fit for all three approaches. The pattern of the state of stress and the levels of effective stress, however, show large differences, due to the varying magnitudes of the stress component induced by the lateral variations in gravitational potential energy for the various scenarios. When compared with observations of the present-day stress field in the Iberian Peninsula, the results obtained applying scenario I are best with regard to the local state of stress, whereas, with scenario II the best fit is obtained for the stress trajectories. The effective stress levels predicted by the various scenarios differ quite significantly in magnitude as well. Estimates of intraplate stress levels in France range between 70 ± 15MPa close to the Pyrenean front to a nearly constant 40 ± 15MPa [Rocher et al., 2000]. Compared with these levels of intraplate stress, scenario III predicts the best fit. The Mid-Atlantic Ridge and ocean floor offshore Iberian differ significantly from the theoretical curve based on cooling oceanic lithosphere. The western and northern Iberian margins have experienced moderate to intense Alpine deformation due to the opening and closure of the Kings Trough/Bay of Biscay/Pyrenean plate boundary. Large seamounts associated with large gravity anomalies line up along the western and northern margin. The crustal configuration differs significantly depending on the approach applied to determine crustal thickness. Principally, if independent direct observations of crustal thickness are available for the region (e.g. deep seismic profiling) this should be incorporated. However, as long as the knowledge of local density structure is not accurate and reference densities will be used in the calculations, the use of such a database of crustal thickness does not improve the results. While eliminating one error (calculation of crustal thickness based on isostasy), it introduces another due to the lack of information on the crustal density distribution. The same is valid for using inverted gravity data, for which a reference density has to be chosen as well. In plate wide models with large elements, the application of a reference density for crust is more justified than for the smaller element sizes used in this study. As shown in this Chapter, the orientations of the principal axes of stress are within close range for each of the scenarios. As long as the data on crustal thickness and density distribution are not accurate enough to apply the theoretically more correct scenarios II and III, the use of the internally more consistent scenario of Coblentz et al. [1994] as used by the references cited is justified. Due to the incompleteness of the input data set, even more so for the geological past, the first scenario has been used in Chapter 6.

38


Chapter 3

CHAPTER 3 - NEW STRUCTURAL DATA FROM FIELD STUDIES IN NORTHWESTERN AND CENTRAL IBERIA Based on the results of an extended literature study (presented in Chapter 4 and 5), several areas and time intervals were identified as lacking good data on the (paleo)stress field. However, in many of these areas or time intervals, additional constraints cannot be obtained, due to absence of sediments of the time interval considered. Within the framework of the SIGMA project [SIGMA, 1998], which focused on determination of the recent stress field, the northwestern part of Iberia was investigated for kinematic indicators and structural data. The NWO “Iberia” project the focus was on the NE-part of the Madrid Basin, along the southeastern border of the Spanish Central System. In the Paleogene deposits of this area growth strata were recognized during an initial survey and a combined study of sedimentology, structural geology and geothermochronology (fission tracks) was carried out in order to unravel the Paleogene tectonic history of the area and to investigate potential constraints this could yield on the tectonic evolution of central Iberia. This chapter is subdivided in two sections. The first section deals with new (paleo)stress data from northwestern Iberia and discusses the results in a more general framework. The second section is devoted to the Paleogene evolution of the northeastern Spanish Central System.

3.1 New structural and kinematic indicator data for NW Iberia The elevated basement ranges of the Cantabrian Cordillera and the NW of Iberia traditionally have been studied for the Hercynian deformation of the region (e.g. Lefort [1979]; Matte [1986] and Pérez Estaún et al. [1988] and references therein). Before opening of the Bay of Biscay, Iberia was part of the Ibero-Armorican arc, the arcuate Hercynian deformation front in Iberia, S. England, Armorica (N. France), the Ardennes (Belgium), and Newfoundland. The Hercynian structure can still be observed very well and has a strong influence on present-day topography. The Cantabrian Cordillera, however, shows marked differences in elevation and physiography with respect to the rest of western Iberia and other belts of Hercynian age (e.g. the Ardennes). Younger tectonic processes must have been active in the region in order to explain the presentday geomorphologic expression of the high (2500m) and the steep mountains towering above the flat Duero basin and at small distance from the Gulf of Biscay. Alpine deformation in the region resulted from southward subduction of the European plate under the Iberian plate during the Paleocene-Eocene [Boillot et al., 1979; Alvarez Marrón et al., 1995; Gallart et al., 1995]. Undoubtedly, stresses related to this subduction were transmitted into the nearby Iberian continent and have led to deformation. Recently, more Alpine structures were being recognized within the Cordillera e.g. [Marín, et al., 1995], offshore (Middle Miocene folding, unpublished Shell data), and along its southern border [Alonso et al., 1996]. Direct observations of Alpine deformation, however, remain limited due to scarceness of Mesozoic and Cenozoic sediments that could yield age-constraints on observed tectonic activity. Tertiary sediments do occur in the Duero Basin and several small, scattered and isolated intramontaine Tertiary basins (see Figure 3.1.1). These sediments are only dated in the Duero Basin and in the lignite39

Chapter 3


bearing As Pontes Basin. In the latter case, both Late Oligocene to Early Miocene and Middle to Late Miocene ages have been suggested, based on fossil dating and palynology. Recently, magnetostratigraphic dating [Huerta et al., 1996] has constrained the tectonic activity in the As Pontes basin to have lasted from 28.7Ma until ~21Ma. Dating of terrigenous sediments in the other small intramontaine basins was carried out by correlation to similar sediments in basins in which age constraints have been obtained (the Duero and As Pontes). This relative dating should be considered with great caution only [Martín Serrano et al., 1996], as should conclusions on timing of deformation observed in these basins based on this type of dating. Traditionally, all the basins of NW Iberia were interpreted as remnants of one larger basin ([Delmaire Bray, 1977; Sluiter and Pannekoek, 1964]) that was dominated by normal faulting and subsequently partly eroded. Within the lignitebearing As Pontes Basin, however, compression structures were discovered and the basin evolution was linked to compressive strikeslip deformation ([Garcia Aguilar, 1987; Bacelar et al., 1988; Cabrera et al., 1996; Huerta et al., 1996]). Evidence for compression has been Figure 3.1.1 put forward for several of the other Overview of the Northwestern Iberian Peninsula, the major Alpine faults and the local Cenozoic basins. basins as well (e.g. Santanach Prat [1994]). With regard to the tectonic regimes that have been proposed to form and subsequently deform the basins, marked differences occur between the numerous small basins. The basins in the northwestern extreme of the region are situated along two ~NW-SE trending strike-slip faults. They are part of a regional set of faults with the same trend. This set of faults deformed the accretionary wedge that formed during the Early Tertiary subduction in a late stage of the Pyrenean compression [Cabrera et al., 1996]. Basins (Meirama, As Pontes) along these faults have been interpreted as pull-apart basins [Ferrus Piñol, 1994] related to non-linearity in the fault traces. The Vilariça basin in NEPortugal [Cabral, 1989] is related to strike-slip motion as well, but situated along the NNE-trending Ponsul fault. The Monforte de Lemos, Maceda and Quiroga basins are most likely related to strike-slip faulting along NNE trending faults as well [Del Olmo Sanz, 1985] but observations of the border contacts are rare. The Tui basin, however, has been interpreted to have an extensional origin [Santanach Prat, 1994]. On the other hand, the Oviedo Basin is interpreted as a thrust-related basin resulting from strike-slip along the NW-SE trending Ventaniella-fault [Alonso et al., 1996] in a clear N-S compressional setting. The same tectonic regime (NS compression) is proposed for the northern edge of the Duero Basin where Tertiary sediments are inclined to vertical [Alonso et al., 1996] by southward thrusted Paleozoic basement of the Cantabrian Cordillera. The origin of the Bierzo Basin has been related to a combination of N-S compression and strike-slip tectonics based on its geographical position in between the northern Duero (NS compression) and the Vilariça (strike-slip) basin [Santanach Prat, 1994]. 40


Chapter 3

Because of the limited amount of data and the inferred differences in origin and tectonic style of the basins, it is hard to present a general tectonic model explaining the evolution of all of them. Santanach Prat [1994] argued that the structure of Tertiary basins in NW Spain is in agreement with NS-shortening. This author relates development of these basins with the Pyrenean orogeny, solely. In this region, this Pyrenean activity only lasted till the late Oligocene-Early Miocene. Present tectonic activity is suggested by increased seismic activity in northwestern Iberia, with earthquakes of magnitude (Mb) up to 5.4 [SIGMA, 1998]. This shows that compression related to the Pyrenean collision was not the last tectonic event that deformed the region. The present-day stress field inferred from focal mechanism solutions indicates approximately NNW-SSE compression (N140o -N160o), with a relatively large perpendicular extensional component [SIGMA, 1998]. As yet, paleostress data in the region were absent. This type of data was required to discriminate between the effects of the Pyrenean collision and younger deformation effects. Therefore, structural geological data were gathered in cooperation with the SIGMA project [SIGMA, 1998] and will be presented in the next sections. Based on the limited occurrence of Tertiary sediments and previous observations within the SIGMA framework, two specific areas were selected to complete the database of kinematic indicator data. The first is the Bierzo basin and its sub-basins, and the second is the Asturian/Cantabrian coast. The general geology and structural outline and data for both will be described in the next section. Finally, the observations will be discussed in the regional context of northwest Iberia.

Bierzo Basin The Bierzo basin (see Figure 3.1.2) is the largest and most impressive intramontaine basin of NW Iberia (50 by 15 to 30 km). Topography surrounding the basin is still controlled by the trends of Hercynian structures and related lithologic boundaries. The basin forms a pronounced contrast with this trend, intersecting the trend of the older structures, as is visible in the shaded topography in Figure 3.1.5a. Mountains are as high as 2200m at small distances from the basin borders are and elevated over 1700m above the base of the flat basin floor (average elevation at 420m). The basin can be subdivided into two bigger sub-basins (Ponferrada & Bembibre/Noceda) and several smaller basins (Vega de Espinareda, Ribón). Towards the southwest, the narrow WSW trending O’Barco Basin has been studied as well because it basically is the southwestern extension of the Bierzo Basin. The sedimentary infill of the Bierzo Basin consists of a complex detritic assembly of conglomerates, sand, silt, clay, and carbonate strata up to 700 meters in thickness. No fossils have been found within the sediments; so indirect dating is generally done by correlation to sediments in the nearby Duero Basin. Sluiter and Pannekoek [1964] considered all the different facies to be local variations of one single sedimentary sequence. Later studies, IGME [1982a], IGME [1982b] and Herail [1982], distinguish two complex units separated by an unconformity that was associated with an important pulse of fracturing, resulting in the individualisation of the Bierzo Basin (s.l.) and the Duero Basin. The lower unit (T1 in this study, or the Toral Formation of Herail [1982]) is attributed to the Miocene (Vindoboniense), although Delmaire Bray [1977] suggests Paleogene ages for part of the sequence. Generally the unit is composed of fine detritic sediment. At the base a coarse gravel base covers a thin layer of red clay. This coarse gravel fines upward into lime and clay with occasional carbonate material. The unit is interpreted as an alluvial fan changing from proximal setting to flood plain, deposited over gentle

41

Chapter 3


paleotopography. Based on distribution patterns of proximal and more distal facies in regional context, a sediment source located to the SW of the basin has been inferred [IGME, 1982b]. The upper sequence (T2, the Medulas and Santalla Formations of Herail [1982]) consists of red detritic sediments, generally coarser than the lower unit, and has been attributed to the Late Miocene to Pliocene. The basement beneath the basin sediments is composed of multi-phase deformed Precambrian to Paleozoic (up to Lower Devonian) metasediments and igneous bodies with a discordant, folded and fractured uppermost Carboniferous (Stephanian) basin. The Hercynian structures still form an arc running N-S near the Asturian coast, turning ~SSW in the area around the Bierzo Basin. The sutures and sedimentary provinces that can be observed in the uplifted basement in the Cantabrian range can be traced into the western Duero Basin and are supposedly present under the Bierzo and the Duero Basin. The Carboniferous strata are folded in N-S direction, approximately. Because Mesozoic and Paleogene sediments are absent in the region, it is not possible to constrain this deformation more accurately than post-Carboniferous. More to the NW (in the Magdalena Basin) Cretaceous sediments have been folded together with the Carboniferous strata [IGME, 1982a]. This suggests an Alpine age for the deformation in the area around the Bierzo Basin. The dominant structural trends in the entire basin are NNE-SSW and EW, and in the northeast and southwest an additional ENE-WSW trend is present (see Figure 3.1.2). Due to scarce outcrops, the contact between basement and Tertiary sediments can often only be observed indirectly in the field (change in color, vegetation, slope and/or lithology of clasts). Therefore, the origin of this contact remains unclear along most borders. Based on these limited observations [Sluiter and Pannekoek, 1964] present the basin borders as normal faults in their sections. Nevertheless, the authors tend to an interpretation of inverse movements along some basin border faults in their text. From this tectonic setting for the Bierzo Basin a model was derived explaining the various basins in NW Iberia to be remnants of a disected former larger basin. More recent, a limited number of direct observations of the contact indicate that both the northern and the southern borders actually are thrusts, placing Paleozoic basement over the Neogene sediments. Geophysical methods have revealed an initial offset of faults on both sides of the Noceda sub-basin of more than 700m [IGME, 1982a]. A few outcrops reveal a ~N360o/35o dipping thrust along the northern border of this sub-basin, placing Stephanian metasediments over Tertiary sediments. Based on some of these outcrops and the trace of the border contact on geological maps, the tectonic origin of the basin borders has been deduced by Santanach Prat [1994]. A set of E-W trending thrusts is interpreted as forming several branches that join up towards the east to form the frontal thrust of the Cantabrian Cordillera. The (de)formation of the Bierzo Basin has, according to Santanach Prat [1994], taken place under an intermediate state of stress between strike-slip (governing the development of most of the basins in Galicia and northern Portugal), and pure compression (responsible for the development of the Cantabrian Cordillera and the Duero Basin). In this way, this author relates the development of the Bierzo Basin mainly to the Pyrenean collision. Multiple reactivations of the border faults, even after deposition of the Tertiary sediments has been inferred, based on the distribution of the sediments and recent activity (faults are active even during Quaternary [IGME, 1982a]), has altered the shape of the basin anew. In order to constrain the amount and type of deformation that affected the basin after its initial development, structural data were gathered along the basin edges. 42

8W

Dip/dip-direction stratification Subhorizontal Estimated Fault surface

Miocene

Pliocene - Quaternary

Oval

A Proba

30

15

0

75

O BARCO Rubiana

Ambas Aguas

24

36

70

36N

40N

44N

Shmax from SIGMA

Shmax derived from pebbles

Tensor solution

51

BIERZO

15

Las Medulas

40

0

12

1

80

2

06

34

15

20

15

10

05

18

60

10

10km

Salas

46 60

15

za

Silvan

Ponferrada

Vega de Espinareda

NOCEDA

190

BEMBIBRE

El Barco de Valdeorras

Barrios

Barcena Boe o Ri

15

25

Berlanga del Bierzo

Campo

80

Ozuela

50

25

10

Fabero

2

Ponferrada

23

Toral de Merayo1 Toral de Merayo2

RIBON

VEGA DE ESPINAREDA

1

05

Sil Rio

Figure 3.1.2 Structural map of the Bierzo Basin including the subbasins of O’Barco, Noceda with tensor results from the different paleostress sites and observations of bedding dip and dip direction.

191

158

126

Bembibre

Noceda

159

127

Northwestern Duero Basin


43

Rio Sil

Chapter 3


New structural and kinematic indicator data from the Bierzo Basin During a field campaign in May 1996, partly together with Dr. Jose Manuel GonzalezCasado (Universidad Autonoma de Madrid), kinematic indicator data have been gathered in the basin. Deformation of the basin sediments is observed mainly close to the borders of the basins. Along the southern border the basin fill consists mainly of conglomerates of the upper sequence (T2). Due to the nature of the sediments, kinematic indicators are hard to find but larger scale structures have been observed more often. Along the northern edge of the basin, outcrops of the lower unit (T1) occur, offering good paleostress sites in the carbonate levels intercalated in red clay. As suggested by the complex outline of the basin, a first examination of the observed kinematic indicator data does not provide a clear pattern of deformation. The observed fault planes in sediments along the borders of the basin do not group in clear families in many cases. A few outcrops will be discussed in more detail, and the general results of the paleostress observations are presented in Table 3.1.1 and discussed at the end of this section. Fabero (Figure 3.1.3a) Just southwest of Fabero a fresh outcrop was studied at a site for a future commercial zone. In a south-facing outcrop, 23 kinematic indicator data were observed in red clay and silt with intercalations of coarser beds containing pebbles and calcreted levels. These sediments can be attributed to the upper part of the lower unit (T1) and dated at around early Miocene. Therefore, the structural data observed in the outcrop demonstrate middle Miocene or younger tectonics. Two tensors have been deduced from fault slip data: ~NS compression and ~NNW-SSE extension. An interchange between σ1 and σ3 is suggested by the similar orientations of fault planes and the principal stress axes for both tensor solutions. An evolution in time cannot be obtained for the stress field. In both cases the tensor quality is relatively low, and a lot of faults remain unexplained. Observations of overprinted fault slip data do not fit in any of the two tensors, which clearly indicate reactivation of older structures. Therefore, the area may have been affected by at least three deformation events after the middle Miocene. Rubiana (Figure 3.1.3b) Just south of the northern border fault of the O’Barco Basin, a mixture of fault rock, cemented colluvial and Miocene sediments is observed. Bedding is hard to recognize, but seems to be oriented 150/75. Striations in this material show the same trend as in the quarry above the border fault: only 2 out of the 20 kinematic indicators show strikeslip movement. The resulting tensor is of very good quality, showing pure compression in N160o direction, perpendicular to the border fault. Within the basin, a set of low angle NNE-dipping thrusts places Paleozoic rocks southward over the Tertiary basin fill. These thrusts have a perfect theoretical orientation and dip to have been formed during the observed N160o compression. The high-angle border faults might have controlled early sedimentation in the basin and were reactivated at a later stage with an inverse component. Oval (Figure 3.1.3c) Along the northern border of the O’Barco Basin, the border fault can be observed as a 70o-80o northward-dipping plane with Paleozoic in the hanging wall on top of Tertiary sediments in the footwall. NW of Oval, a quarry in Paleozoic quartzite is located just uphill of this high angle northern border thrust. At the scale of the outcrop normal faulting dominates. Kinematic indicators show mainly normal or reverse movement and only 3 out of the 34 measured faults show strike-slip movement. The latter contrasts with the 44


Chapter 3

Figure 3.1.3 Observations and paleostress results for Fabero, Rubiana and Oval. See Table 3.1.1 for explanation of abbreviations. See Figure 3.1.2 for location of the stations.

45

Chapter 3


strike-slip deformation suggested by Santanach Prat [1994]. The resulting tensors from this outcrop are: N346o directed pure compression and N101o directed pure extension. The compression is clearly related to southward thrusting, whereas the extensional event is a little enigmatic. Ribón The Romans mined the alluvial deposits of the Bierzo and NW Duero Basin for gold. They washed down huge parts of the red sediments by special techniques and gathered the gold flakes in the sediments in wool soaked in grease placed in the bedding of the man-made sedimentary fans. The most well known mine, Las Medulas, is located near the southwestern limit of the Bierzo Basin. North of Ribón, another of these ancient gold mines of the Romans created a cliff in the red sediments with in front of it a large fan, which can be dated rather confidently at being around 2000 year old. A very flat terrace of Rio Burbía consisting of well-rounded coarse pebbles cuts this fan. Presently the river is eroding its own terrace under which in the riverbed layers of red sediments are visible, dipping approximately 20o to the south. This setting shows that the area has experienced significant vertical motions in the last 2000 years leading to erosion of the anthropogenic fan, to deposition of the terrace and finally to a renewed erosional stage. The contact between the Tertiary red beds and basement is not normal since the near horizontal beds of the cliff end to the north abruptly against Paleozoic basement. The vertical offset along the steep contact can be estimated to be at least of the order of 100m, which is the height of the cliffs from the valley floor. The contact itself cannot be observed and therefore the nature of the offset cannot be determined. Parallel to the mapped contact between basement and red sediments on the eastern side of the valley, a fault in the Paleozoic basement on the opposite bank (west) of Rio Burbía has been observed. A few reverse-sinistral striations have been found, from which unfortunately no good tensor could be obtained. Toral de Merayo1 (Figure 3.1.4a) Along the southern border of the Bierzo sub-basin, near Toral de Merayo, a sand and clay-pit offers excellent outcrop conditions. An approximately 1m deep and 6m wide paleochannel is observed that has been draining towards the W, parallel to the presentday southern basin border. Interestingly, directly to the east, from where the channel has drained from a highly elevated basement block occurs presently. This proofs that relative vertical motions occurred after deposition of the Miocene sequence. The paleochannel is cut by several steep faults that cut an older normal fault. Small-scale inverse structures were observed and the steep faults reflect strike-slip motion. The tensor results show two deformation phases: an early pure extensive (Shmin ~190o) and a later pure strike-slip deformation, with Shmax oriented due N. Just some 100 meters to the west (Toral de Merayo2) in an abandoned pit, nearly pure strike-slip is observed with a N336o oriented Shmax, consistent with the phase observed in Toral de Merayo1. Ponferrada (Figure 3.1.4b) An abandoned gravel/sand quarry is located southwest of Ponferrada along the road to Molinaseca at kilometer 1.5. Nearly vertical Paleozoic quartzite (S0:230/80) is covered by a very coarse (U. Miocene?) conglomerate along a red clay tapered contact dipping gently to the north (350/10). This red clay based conglomerate is described as the base of the Miocene sediments [IGME, 1982a]. In the conglomerate some larger scale structures are present and many of the pebbles, up to the highest levels under the carbonate cemented Late Quaternary cap, are broken and show pressure solution marks. The compression direction inferred from these pebbles is ~N130oE, a direction 46


Chapter 3

that is observed in the fault inversion results at the same locality as well. A wellconstrained tensor is obtained that indicates compressive strike-slip for the majority of the faults. Salas (Figure 3.1.4c) Crossing the locality Salas de los Barrios uphill, on the left side of the road an outcrop shows a steep northwestward (340/60) dipping sequence of conglomerate, sandstone and silt. This outcrop is very close to the southern border of the basin: a few tens of meters uphill Paleozoic basement outcrops. The contact cannot be observed directly. Reverse faulting along ~N320o and ~N150o planes is obvious from the outcrop, showing zones of sheared pebble beds. Several of the conglomerate pebbles show kinematic indicators, and many pebbles of Paleozoic basement are broken indicating rather intense deformation. Both fault slip data and kinematic indicators at pebbles show a ~N310o directed maximum horizontal compression. Some hundred meters down the road normal faulting in nearly horizontal bedding can be observed. Spatial distribution of deformation types and dip direction trends in the Tertiary sediments suggest sinistral transpressive movement along this fault.

General results The results obtained by pressure solution and striated surfaces on pebbles yield a very consistent pattern of stress orientations. In all of the sites where the direction of Shmax has been determined from both pebbles and fault slip data, the solution of the first was reflected in the latter as well. The quality of the tensors is reasonable to good in most of the stations (see Table 3.1.1) even though for many of the outcrops, only approximately 70% of the kinematic indicators fit the calculated tensors. Moreover, most of the faults that can be explained by the calculated tensor are considered to be reactivated, not newly formed. Earlier phases of deformation that cannot be deduced using the paleostress method might be responsible for faults that cannot be explained by the obtained tensors. Strike-slip is the dominant state of stress, and the direction of Shmax is on average NNW. It is hard to reconstruct a temporal evolution of the stress field, because only for two outcrops it was possible to determine multiple tensors. Only in Toral de Merayo1 the stress fields have been dated relatively: the younger stress field is extensive, with extension direction near N-S. Based on this observation, a similar temporal evolution could be validated for the site Fabero. Obviously, more data are required to be able to conclude whether this extension is a local phenomenon, a later stage in the evolution of the basin or whether this represents a simultaneous coaxial N-S compression with N-S extension.

Geomorphology and spatial distribution of Tertiary sediments Combining the results obtained from the paleostress measurements with other independent data about displacements of the border faults might enable us to determine the tectonic evolution of the Bierzo Basin and surrounding sub-basins. One such data source is the spatial distribution of the Tertiary sediments. A compilation of the minimum and maximum altitude of occurrences of the two Tertiary sequences is presented in Figure 3.1.5b. Both along the basin borders, between the sub-basins and within the basins several irregularities can be observed. Because both sequences consist of fine detritic alluvial fan units, apart from a limited coarse base of the first sequence, the assumption is made that the sequences have been deposited along gentle slopes.

47

Chapter 3


Figure 3.1.4 Observations and paleostress results for Toral de Merayo1, Ponferrada and Salas. See Table 3.1.1 for explanation of abbreviations paleostress results. See Figure 3.1.2 for location of the stations. 48


Chapter 3

Paleo-topography can be inferred where the lower topographic limits differs on both sides of such an irregularity while the upper limits are coincident. Where the base and top level are displaced by the same order of magnitude, tectonic activity is assumed to have caused the anomaly. Interestingly, for the first interval a source of sediments has been inferred to be located far to the southwest of the basin [Corrochano and Carballeira, 1983]. The base of sequence T1, however, slopes in a direction towards the SW. Unfortunately; it is impossible to determine whether the base of the T1-sequence has the same age throughout the entire basin. This leaves two possible explanations: (a) the basin filled progressively with the development of the T1-sequence, onlapping the basement in a northeastern direction through time (in this case its base is not time equivalent), or (b) late stage tilting ‘reversed’ the direction of the depositional slopes (so the base is time equivalent). One of the main observations is, that if the sediments of T1 can be correlated to sediments in the Duero Basin, then either the Bierzo Basin has subsided or the western Duero Basin has been uplifted significantly since deposition of these sediments. Between the eastern Bierzo and western Duero Basin the elevation difference of the unit amounts up to 300m. This hypothesis is supported by the sequence of middle to upper Miocene sediments in the western sector of the Duero Basin, representing several alluvial depositional systems. The lower sequence is composed of typical distal alluvial inundation plains without any coarse sediment with sourcing from the west (via the Bierzo Basin), but the middle and upper sequences contain a high amount of clasts [Corrochano and Carballeira, 1983]. These upper alluvial systems indicate a reactivation of the western margin of the Duero Basin, just as the coarser deposits of T2 in the Bierzo Basin indicate reactivation of reliefs. The inferred reactivation during the MiddleLate Miocene was most likely associated with normal movement along steep NNE trending faults. This reactivation of the basin margins and tilting of blocks not only divided the Bierzo Basin from the Duero Basin, but ‘disintegrated’ the Bierzo Basin into several sub-basins as well. In spite of the neotectonic overprinting that has complicated the setting, three eastward tilted blocks can still be recognized in Figure 3.1.2 (panel B and C): (1) the western Duero, (2) the Bembibre/Noceda block and (3) the Ponferrada block. The latter two are separated by an elevated basement structure running in NNE direction from Ponferrada. Both the T1 and T2 sequences, which were deposited subsequently during Late Miocene and Pliocene, have been affected by younger tectonic activity. Especially along the northern border compression has activated approximately E-W trending low angle thrusts, which lifted up the massifs around the basin and separated the Noceda subbasin (NE) from the Bembibre Basin. Additional information on recent vertical motions and block tilting can be obtained from an analysis of the drainage pattern of the area. Small rivers (Burbía, Valcarce, Cua, not indicated on map) running through the Paleozoic basement have cut deep and steep valleys, suggesting uplift during a late stage. Along the Noceda River, a remarkable high number of terraces are present given the small size of the river and the location of the terraces at the headwaters. Moreover, the terraces are all located on the right bank of the river, which indicates that neotectonic movements are tilting the Noceda sub-basin, lifting its western part up [IGME, 1982a]. Rio Boeza (provenance from the Bembibre

49

0

400-650

280-460

320-420

10km

500

400-480340-380 340 420-480

620

460-680

Miocene

Fold

T2

540

42

0-6

A

Pliocene - Quaternary

Fault

80

820-9 800 20 >780

720

380-440 380

0

70

340-400

00 10 066

500-700

540-680 540 560 500-640

440

640

500-660

>760

600-900

760 760 720 800-870 8 820 880 -7 0 >820 6 740 620-800 660 6 720 0 >760 800 92 780 740-790 800 760 0740 800 >740 20 82 600-760 8 660-700 0 72 900 640 760-880 0 500-74

30

-9

0 83

90

0-

76

70 078 0

T1

660

0-

64

680

0

>640

0

640-820

10 0-

+

90

880-960

+

640-920

760-960

920-1060

+

1080-1135

740-1090

1040

1080-1220

1020-1260

1080-1220

1080-1120

980-1140

1140-1260

940-1080

1180-1260

B

+ + + + + + + + + + + + + C

+

700-1100

700-920

680-960

>1060

680-860

20

1080-1120

Figure 3.1.5 Height-distribution of the two sequences in the Tertiary sediments in the Bierzo Basin (s.l.). Panel A: shaded topography of the area. Panel B: lower and upper altitudes of level T1 (in black boxes) and level T2 (normal). Large steps in altitude along the basin borders show tectonic movement after deposition of the levels. Differences in offset between T1 and T2 can be observed, which indicates different tectonic situations for the period 900 31]

Sediments

n

Ri o Ja ram a

ia er

L Ba usit sin an ia n

Figure 3.2.1 The northeastern Madrid Basin at the junction of the Iberian Chain (to the east) and the SCS (to the north) in Central Iberia. Compilation from and tentative correlation between several sheets of 1:50.000 geological maps (IGME) are shown in the inset. Numbers for stratigraphic levels are from the Jadraque text and sheet [IGME, 1990b]. Paleogene sediments (darker gray) are found in two sectors near Venturada and, between Beleña de Sorbe and Huermeces.


59

Chapter 3


[1979] for the classical evolutionary model on this area). During the Mesozoic rift events that affected large parts of SW Europe [Ziegler, 1988], central Iberia formed the western border of the Iberian Basin [Salas and Casas, 1993]. This is still reflected in the different types of substratum beneath the Tertiary sediments. In the western and central part of Iberia, the substratum consists of Proterozoic and Paleozoic metasediments and intrusives with a condensed Mesozoic sedimentary sequence. To the east, in the Iberian Chain it comprises eastward-thickening Mesozoic sequences, including important decollement levels. Thus, the area investigated is located adjacent to two of the most important Cenozoic intraplate deformation belts of the Central Iberian Peninsula. As will be shown, tectonic activity of the bordering basement structures has been recorded in the basin fill of the northeastern corner of the Madrid Basin. At this end of the basin, a key area around Pinilla de Jadraque has been selected (see Figure 3.2.1) where interrelations between sedimentation and tectonic evolution could be studied in detail. The validity of the inferences from this key area for the regional geological evolution is supported by data collected elsewhere in the study area. The existing geological map of the key area around Pinilla de Jadraque [IGME, 1981a] shows a detailed analysis of the metamorphic basement structures, but a gross oversimplification of structures in the Tertiary clastic sediments. Therefore, the first objective of this study was detailed mapping of the Pinilla area. Within the Mesozoic to upper Miocene sedimentary cover evidence for Tertiary tectonic activity was derived from mesoscale folding, including growth strata, striations on fault planes, and striations and pitted surfaces of pebbles have been used for this purpose.

Proposed tectonic models for the evolution of the SCS The Alpine tectonic evolution of the study area (see Figure 3.2.1) was characterized by rigid deformation of the Hercynian basement along reactivated late Hercynian weakness zones [Doblas et al., 1994]. The overlying Mesozoic-Tertiary sedimentary cover was folded and faulted associated with block movements of the basement, due to the absence of a major decollement level. Some levels of Late Cretaceous-Paleogene evaporites are present and produce local disharmonic deformation. Central Iberia was subjected to stresses transmitted from the plate boundaries into the interior of the plate, which resulted in multiple reactivations of Late Hercynian basement faults [Andeweg et al., 1999a]. Several models have been proposed to explain the Tertiary geological evolution of the SCS. 1. As early as the first half of the last century Schmieder [1953] (but published in German in 1915) and Schwenzer [1943] (translated from a German paper of 1936) proposed a horst-and-graben structure on the basis of the geomorphologic character of the system with high-elevated plateaus, intramontaine basins (Lozoya and Amblés) and intraplate basins (Duero and Tajo). This model implies extension as the governing process in forming the system. Dominance of compression in the development of the system has been suggested for the first time in the early eighties and has become accepted widely during the last decade. Some recent geomorphologic models however, are also based upon the ‘horst-and-graben’ model [de Pedraza Gilsanz, 1989], but are unaware of the implications of this term for the tectonic setting of the structure. 2. Crustal up arching [Vegas and Banda, 1982] This model is the first that tried to explain the development of the SCS in terms of compression. The authors proposed rigid areas in the interior of the system deformed into great arches bounded by reverse faults. From a gravity profile they interpret the SCS as an unrooted structure. However, such a rootless crustal arch cannot be in local isostatic equilibrium and would require a very high level of dynamic intraplate stress to maintain the relief. 60


Chapter 3

3. Crustal pop-up [Ribeiro et al., 1990] This model suggests the existence of an important mid-crustal decoupling, running from the Betics in the south of the Iberian Peninsula all the way to the Northern Border Fault of the SCS, carrying the MB as a piggy back basin northwards. However, some major problems with this interpretation are: (a) The development of the SCS is uniquely linked to the Betic collision (starting around Middle Miocene). (b) The existence of such a mid-crustal decollement level is still a matter of debate. Some indications exist for decoupling, e.g. the Altomira phase [Muñoz Martín et al., 1998]. However, the decollement took place in the supra-crustal sediments at the level of the Keuper marl. This sedimentary decollement level pinches out west of the Sierra Altomira and therefore is not present under the Madrid Basin sediments. (c) The fault geometry of the SCS does not support this model. The Southern Border Fault (SBF), in this model the backthrust, is much steeper than the ‘frontal’ Northern Boundary Fault (NBF) and more deformation took place along the SBF (offset of about 5000m versus an estimated 1000m). This geometrical distribution of the deformation would rather suggest the inverse of the proposed model: the SCS as a frontal expression of a mid crustal detachment running from the northern Cantabrian Range to the SBF, carrying the Duero Basin as a piggy-back basin. Pulgar et al. [1997] interpreted mid-crustal delamination from deep seismic lines across the Cantabrian Range and northern border of the Duero Basin. A ‘Cantabrian’ wedge is interpreted to indent into and to split the Iberian crust apart. Such a mid-crustal detachment would provide a hypothesis for the complete absence of deformation in the Duero Basin apart from its margins. 4. Block rotation [Vegas et al., 1990] This model is based on the interpretation of wide-angle seismic profiling [Suriñach & Vegas, 1988], from which (moderate) crustal thickening was interpreted to be the result of Betic-originated intraplate compression. This strain is supposed to have localized in the SCS due to a previous dextral shear zone. Strike-slip was inferred along N030o trending faults in the Sierra de Gredos (western SCS) prior to the middle Miocene N150o compression that inverted the same fault planes. Therefore, a model is proposed of Late Cretaceous simple shear and clockwise rotation of crustal blocks, followed by a MiddleLate Miocene pure shear within the lower crust, causing reactivation of rotations, uplift and high-angle faults at boundaries. Block rotations due to strike-slip movement along the southern and northern border fault and along parallel faults in the SCS are interpreted as the mechanism governing its deformation. However, the model requires a large set of near vertical faults, trending more or less parallel, enabling the block rotations. This holds for the intramontaine late Tertiary(?) Avila Basin, but major parts of the SCS do lack evidence for such faults. Large-scale block rotations are not recognized either. 5. Pop-up structure [De Vicente et al., 1996c] De Vicente et al. [1996b] and De Vicente et al. [1996c] have described in detail the latestage tectonic evolution of central Spain and the tectono-sedimentary evolution of the Madrid Basin and the SCS from the M. Miocene intraplate deformation (related to the collision of Africa and Iberia/Eurasia). These authors relate the development of the SCS and its northern and southern foreland basins solely to NNW-SSE compression due to the collision of Africa and Iberia/Eurasia. Based on a seismic velocity model, surface geology, restoration of the Mesozoic cover over the SCS and shallow seismic reflection data in the Madrid Basin, De Vicente et al. [1996c] proposed a pop-up structural model for the SCS. The model consists of a system of pop-ups and ‘pop-downs’ with a steep SBF, thrusting the SCS over the Madrid Basin foreland. Flatter faults along the northern 61

Chapter 3


border of the SCS are oriented parallel to preexisting weakness zones in the metamorphic basement. Therefore, loading of the northern foreland is restricted, creating a less pronounced basin (the Duero Basin) due to a large horizontal but smaller vertical displacement than along the SBF. A minor problem with this model is that it explains the entire activity of the SCS as a consequence of the Betic compression from the Middle Miocene until today and disregards any effect of inheritance of prior preMiocene deformation. As a model for the late stage development of the SCS it seems to be very convenient. Nevertheless, the continuation at depth of the major faults and the existence of an upper crustal detachment level remain questionable. Deep seismic reflection lines are required to solve these problems.

Constraints from seismic data In the Madrid Basin, Amoco, Shell and Teneco carried out seismic reflection profiling in the late seventies and early eighties in a search for hydrocarbons. The seismic lines [Querol, 1983] offer limited resolution with depth, so information about the crustal configuration is restricted or absent. The lines, however, show in detail the sedimentary sequences that filled the basin and their interrelation. Tectonic activity of the basin border can be inferred from the internal architecture in the sedimentary sequence. Two profiles (Line M-7922 and Line M-7916) that run from the northern basin border (SCS) in SSE direction through part of the basin are presented in Figure 3.2.3 (see middle panel Figure 3.2.2 for location) and discussed. The NNW of the line M-7922 (Figure 3.2.3a) is located in the study area near Pinilla and the sedimentary sequence is tied by the projected wells of Baides and Santa Barbara. It reveals important constraints on episodes of fault activity. Combined with surface information (see section), growth strata dipping steeply southward can be inferred at the extreme NNW of the line. The southern flank of this growth syncline is characterized by north-dipping Paleogene reflectors. These reflectors run parallel, showing that tilting of this block occurred after the deposition of this sequence. The Mesozoic sequence shows important thickness reduction towards the Baides well, while in the Madrid Basin the Mesozoic sequence is thick and even distributed (0.5s TWT). This shows that the zone around Baides has been active as a paleohigh during Mesozoic sedimentation. The assumption of normal fault movement during the Mesozoic would fit best in the setting of the Iberian Basin. The superficial SBF interpreted by Querol [1983] cannot have acted as a normal fault, because of the low dip towards the north. Most likely, the interpreted thrust is the near-surface expression of motion along a deeper crustal fault, which might have been a normal fault during earlier stages. This thrust fault is the M. Miocene SBF of the SCS in this area, offsetting the basement about 2.5 km. In the Santa Barbara well an intra-Paleogene unconformity is observed that can be traced northwest-wards and indicates gentle uplift of the inner Madrid Basin during this period. This uplift could be related to the active thrust just SE of the well, but the unconformity is displaced by the fault itself. An alternative explanation for uplift might be the formation of a gentle bulge related to foreland basin development under the load of the SCS. Line M-7916 (Figure 3.2.3b) runs from the SCS basement just north of Madrid into the Madrid Basin. The SBF of the SCS is observed in the extreme NNW of the profile. One of the most interesting features in the profile is a similar intra-Paleogene unconformity as observed in Line M-7922. Upper Paleogene sediments onlap truncated Lower Paleogene sediments, clearly indicating an intra-Paleogene tectonic activity. This tectonic activity seems to reflect a first stage of loading by the SCS and related foreland basin development. The unconformity is tied to ~Oligocene in the nearby El Pradillo well. At this location up to 1450m of late Oligocene to early Miocene deposits have been 62

0

500

1000m

[16-18]

[27-29]

N

0

[5-7]

[8-13]

500

Barbareja

Line M-7916

Paleozoic basement

A

509

1000m

Torrelaguna

Buitrago de Lozoya

484

A’

[14]

Brihuega

Pinilla3

Marchamalo

511

Ledanca

.. Hiendelaencina Siguenza 486 485

460

Line M-7922

B B’

~ de Valdepenas la Sierra Jadraque 510

Pinilla de Jadraque (key area)

[14-15]

487

461

[15a]

[30]

?

[15b]

[16]

[27-29]

?

[5-7]

[8-13]

[14]

[15a]

[27-29]

E

A’

B’

800m

900

S

Jurassic & Triassic

L. - U. Cretaceous limestone

U. Cret./Paleocene clay and evaporite

Paleocene marl/lst.

Paleocene -Eocene marl/limestone and silt

[14-15]

[15b]

[19-24] [17-18]

Eocene conglomerates

Eoc.-Olig. Conglo/silts & clays

Olig.(?) clay/silt/cong.

U. Miocene conglomerates

Pliocene conglomerates

Pinilla1

[16]

[17-18]

[19-24]

[27-29]

[30]

[16-18]

[27-29]

Figure 3.2.2 Upper panel: approx. E-W profile through the northeastern Madrid Basin (see middle panel for location) to indicate folding of the Mesozoic-E. Tertiary cover o related to ~N050 -090 compression. Middle panel: location of profiles and seismic lines (see Figure 3.2.3). Lower panel: a N-S cross-section through the key area along Cañamares valley. Sites for which paleostress determination has been carried out are indicated.

B

800m

900

[1-4]

undifferentiated cover

Location of profiles

A

1000 800m

[30]

W


63

2.0

1.0

0.0

Thinning Mesozoic sequence

2.0

1.0

0.0

SCS

NW

(projected)

Baides-I

?

Line M-7916

SE

(projected)

SE

Santa Barbara-I

Onlapping Upper Paleogene over truncated Lower Paleogene (bulge?)

Onlapping U. Paleogene

Figure 3.2.3 Interpretation of two seismic sections through the northern Madrid Basin, perpendicular to the basin border (see middle panel Figure 3.2.2 for location). Seismic sections after Querol [1983].

Pre-Mesozoic basement

Mesozoic sequence base Utrillas base Triassic

undifferentiated reflectors L. Paleogene sequence intra-paleogene discordancy

M. Miocene and younger or poor resolution L. + U. Oligocene sequence

Line M-7922

?

SCS

Time in seconds

64

Time in seconds

NW



Chapter 3

drilled, but to the east (Santa Barbara), this level is reduced to around 800m. The unit below this sequence has not been divided into smaller segments than Maastrichtianmiddle Oligocene. Paleogene sediments drilled in the well are lacustrine-evaporitic deposits and represent distal foreland basin sediments and do not have a terrigenous nature, nor is the reflectivity pattern of the sediments interpreted as terrigenous [Querol, 1983]. The middle-upper Miocene sediments are coarse sediments, indicating a more proximal position to the thrust belt.

General outline of the sedimentary sequence in central Iberia A synthesis of the information from 6 sheets of 1:50.000 geological maps (Hiendelaencina [IGME, 1981a]; Sigüenza [IGME, 1981b]; Buitrago de Lozoya [ITGME, 1991a]; Valdepeñas de la Sierra [ITGME, 1990d]; Jadraque [ITGME, 1990b] and Torrelaguna [ITGME, 1995]), has provided an overview map of the entire southern border of the SCS (see Figure 3.2.1). Outcrops of Paleogene sediments are scarce and limited to two zones (1) Venturada and (2) Beleña de Sorbe – Huermeces. Correlation of these limited outcrops revealing Paleogene successions all along the northern border of the Madrid Basin has been problematic due to lateral facies changes and different subdivision of sedimentary units in the different sheets of the geological maps. For example, almost 80% of the map Hiendelaencina [IGME, 1981a] comprises metamorphic rocks of Precambrian-Permian age. Subdivision and detailed mapping of Tertiary and younger sediments has not been carried out as accurately as for example in the sheet located directly south of it (Jadraque [ITGME, 1990b]), which is entirely covered by Tertiary sediments and from more recent date. A tentative correlation of the Tertiary sediments along strike the contact of the SCS and MB is proposed in this section. In the study area, a large scale subdivision of the sedimentary cover can be made by defining a Mesozoic-Tertiary part consisting of Triassic, Cretaceous and Paleogene sediments and a Tertiary upper part made up of M. to U. Miocene. Numbers in brackets refer to the subdivision used in the Jadraque geological map [ITGME, 1990b] and the geologic maps presented in this section. Mesozoic-E.Tertiary During this period, the area formed the western margin of the NW–SE trending Mesozoic Iberian Basin. Therefore, the Mesozoic series overlying the metamorphic basement of the Madrid Basin is far from complete, of a marginal type, and relatively thin (several hundreds of meters). Towards the east, the thicknesses of Mesozoic strata (Muschelkalk, Keuper, and Buntsandstein) increase rapidly up to several kilometers in the Iberian Chain. For the same reason, the Keuper in the study area is formed dominantly by calcareous and siliciclastic sandstone, whereas in main part of eastern Iberia it contains significant levels of evaporites that can act as decollement level. The latest Cretaceous-Paleogene succession presents the gradual change from the marine L. Cretaceous series to the terrigenous Paleocene and younger series. Three units can be differentiated based on lithology: (a) levels of evaporites of up to 90m alternating with mud [14-15] characterize the first 800-1000m, of Maastrichtian-middle Eocene age. This is followed by (b) an approximately 200-300m thick marly series with terrigenous intercalations [16] (Eocene-Lower Oligocene) passing gradually to (c) the third unit, which reaches a thickness of 450m, consists of terrigenous sediments with carbonate levels of Late Oligocene age [17-18]. The sandstones of the middle (EoceneLower Oligocene) unit contain the first fragments of metamorphic rock of the SCS hinterland. The development of units [17-18] shows in the Jadraque area both vertical and lateral diminishing lacustrine environments by progressive building out of a fluvial 65

Chapter 3


system, most likely related to tectonic activity. Towards the east, the lacustrine carbonates become more dominant. E.Tertiary-Pliocene The Madrid Basin is filled with up to 3500 meters of Cenozoic sediments [Querol, 1983]. During large time spans of the development of the basin, lacustrine conditions prevailed in its center. Along the rims, clastic sediments shed into the basin from the active borders progressively filled the basin. The interrelations between these two facies enable an understanding of the rate of erosion in the SCS and therefore, an estimate of its tectonic activity. The Neogene series filled the Madrid Basin in endoric circumstances, forming a system of alluvial fans entering terrigenous sediments in the basin, gradually passing through complex transition facies to the lacustrine evaporites deposited in the center of the basin. All of this series is topped by the detritic-calcareous Paramo facies. Along the borders of the basin, angular unconformities within the series can be observed that towards the center of the basin change into paraconformities, clearly showing tectonic activity along the basin margins. All of the Mesozoic-E.Tertiary series have subsequently been folded and partly eroded and are covered by an angular discordant terrigenous unit that is the base of the Miocene series in the area. The oldest Miocene sediments form an undated unit of quartzitic conglomerates and calcareous breccias grading upward into red mud and intercalated marl near Arbancon [20]. In the map of Jadraque [ITGME, 1990b] this unit is attributed to lower Aragonian (~Burdigalian, L.Miocene, ~19-18Ma). To the south, near e.g. Cogolludo or Espinosa de Henares, a unit of alternating conglomerate and mud, which shows evidence for mass transport [19] discordantly covers the ‘Paleogene’ series. In the Jadraque map [ITGME, 1990b] this level is attributed to the ‘most likely lower-middle’ Aragonian (L. – M. Miocene, ~18-15Ma) based on cartographic correlation to other units in the center of the Madrid Basin that have been dated accurately. However, in both the map for Hiendelaencia [IGME, 1981a] and Sigüenza [IGME, 1981b], the top of this same unit is correlated to de Loranca del Campo outcrop, in the Madrid Basin to the south of the region, dated at upper Agenian (Lowermost Miocene, ~Aquitanian, 24-22Ma). An early Miocene age is considered in our study. The Middle Miocene is represented in the basin by series of near horizontal marl, mud, sand [21-22] and carbonate levels [23]. It shows a discordant relationship with the older units. Reptile and micro mammal findings date the top of this series at the AragonianVallesian boundary (M.-U. Miocene boundary, ~12Ma, Serravallian-Tortonian). Along the Mesozoic rim at the northern basin margin, a discordant conglomerate series forms growth synclines near Arbancon and San Andres del Congosto [24]. The unit grades upward into finer materials to the top (limestone and marl). The southern flanks of these structures are up to 2-3 times as thick as their steep northern segments. The limits of this unit have been dated as U. Vallesian - M. Turolian as base and top respectively. The progressive discordance shows tectonic activity of the border for this time-span. In the basin, the Middle-Upper Miocene sequence is topped by an approximately 5060m carbonate and marl sequence, the ‘Blanca’ formation [25-26], which forms large plains with numerous karst features at an average altitude of 1020m. Their age is uppermost Aragonian-lower Vallesian, based on several micro mammal datings and correlation. At the basin borders, erosion products of the SCS have been deposited in several units of conglomerates [27-29] during the upper Turolian –Alfambrian (late Miocene, Messinian-early Pliocene). These series of uppermost Miocene to lowermost Pliocene 66

3 40’

o

3 40’

o

Paleozoic basement

Paleogene Mesozoic

0

N

Marchamalo

o

40 55’

2000m

511

Madrid Basin o

3 15’

o ’ 3 15

Large sc

ale undu

o

3 00’

Figure 3.2.4 Structural map of the northern border of the Madrid Basin with the Spanish Central System to indicate discontinuous folds in enechelon patterns in the sedimentary cover.

3 25’

o

o

Ledanca

st5e’ m y S al 3 2

tr Cen

Brihuega

~ de Valdepenas la Sierra Jadraque 510 509

Quaternary Neogene

Torrelaguna

Buitrago de Lozoya

487

Fault Fold axis with dip direction Dip direction

lations

dri Ma

in as B d

e n

484

461

Ib ria

.. Hiendelaencina Siguenza 486 485

460

Structural elements


e

R g an

67

Chapter 3


alluvial fans discordantly overly all the underlying basement structures. Alonso Zarza et al. [1993] has described the fan systems and the paleogeographical control on their development. The last sediments of the Neogene succession in the northeastern Madrid Basin consist of 8-10m thick quartzite pebble conglomerates of late Pliocene age [30]. This unit forms extensive platforms at high elevations of about 1080 meter, and covers older upper Miocene units, e.g. [27], along an erosive unconformity that can be recognized on map scale. Younger (Quaternary) sediments are restricted mainly to terraces along the rivers that have cut through the upper Pliocene levels, into the underlying basement.

Observations in the study area Structural data In Figure 3.2.4 the trends of the most important structures in the Mesozoic to presentday sedimentary sequence of the NE Madrid Basin is shown. E-W, ENE-WSW and SSENNW are dominant directions for the large-scale folds and faults. Figure 3.2.2 shows a cross section perpendicular to the border structure (panel A) and a profile parallel to it. The latter shows broad undulations in the Mesozoic- L. Paleogene cover. Folds and faults in the thin Mesozoic-E. Tertiary cover tends to be highly discontinuous and the fold axes are not straight but show in general an E-W direction in the central parts. The southern border of the SCS is oriented at ~N070o, at a small angle (20o) with the above mentioned fold trend. This structural outline suggests that the folds in the cover formed en-echelon above a blind basement fault, which was active as a sinistral strike-slip fault during some stage in the development of the region. In two-dimensional sections, enechelon patterns of faults and folds related to strike-slip tectonics cannot be revealed. Therefore 3 parallel sections have been constructed, based on new field observations, to be able to present a three-dimensional block diagram of the key area in this study, the Pinilla area (Figure 3.2.5). This view shows the flower tulip structure, en-echelon faults and folds in the area, as well as the growth strata near Pinilla.

Sedimentary data A field study was carried out in order to describe the sedimentary sequence in the Pinilla area in detail, providing control on the correlation of the observed stratigraphy with the units in the map of Jadraque [ITGME, 1990b]. The sequence has been logged in great detail to unravel the internal structures, sedimentary features and interrelations between units that have been recognized by macro-analysis, by careful mapping at 1:18.000 and by aerial photograph study. The result of this is shows in Figure 3.2.6. The Mesozoic series can be correlated to the regional sequences, just as the lower Paleogene sequence on top of it. This is related to the nature of the latter sediments: the monotonous alternation of thin mud, evaporite and carbonate layers is interpreted as playa-like sediments, which implies broad plains without pronounced tectonic activity. The base of the sequence is dominated by mud; the middle part is a characterized by thin pale mudstone layers and the mud is coarsening upwards to siltstone. Deposits of Eocene to Oligocene age show growth strata and have been subdivided in 3 units (see for more details on the sediments De Bruijne [2001]). Unit 1 (140m thick) consists predominantly of massive sheet conglomerates with blocks up to meter scale; towards the top finer grained lenses and intercalation of sandstone occur. It is considered to have been deposited in a high energetic, fast accumulating alluvial fan system in a broad accommodation space. The base of unit 2 shows a highly lateral varying set of small erosive, channeled bodies of pebbly sandstone. They grade upward

68

0

1000m

Pinilla

Embalse de Palmaces

N

Figure 3.2.5 Three parallel NNW-SSE profiles through the Pinilla key area showing lateral connection of structures and important strike-slip component in deformation (discontinuous folding and faulting). Inset shows profile locations in simplified version of figure 3.2.6. (see www.geo.vu.nl/~andb/iberia for full color version)

N

100 m

Paleozoic basement

Triassic

Jurassic

Eocene conglomerates Paleocene marl and limestone

L.Cretaceous

Olig. clay/silt/cong.

M. Cretaceous

Miocene conglomerate Eoc.-Olig. conglomerate, silt & clay

U. Cretaceous lst.

Pliocene conglomerate

U. Cret./Paleoc. clay


69

70

17

76

Jurassic

Eoc.-Olig. conglomerate, silt & clay

20

L.Cretaceous

Olig. clay/silt/cong.

10

23

21

19

50

17

80

15

87 68

23

60

38 54

62

60

32

Paleozoic basement

Triassic

M. Cretaceous

Miocene conglomerate

Eocene conglomerates Paleocene marl and limestone

U. Cretaceous lst.

Pliocene conglomerate

U. Cret./Paleoc. clay

62 85 05

76

230/50 10

17 23

52

10

60

Fault

Fold-axis

40

285/19

Unconformity

11

80

Traced bedding plane

14 170

08

71

15

24

71

78

81

40

52

52

57

10

16

18

10

120

20

12

24

05

50 32

14

18

79

85

65

20

50

65

71

66

72

26 19

22

Pinilla de Jadraque

Paleo-current direction

07

11

26

15

14

14

14

02

13

25

140 10

16 12

06

04

32

15

81

82

66

56

85

60

16

21 05

09

85

60

0

Embalse de Palmaces

500

1000m

N


~amares Rio Can

Figure 3.2.6 Detailed geological map of the key area around the village of Pinilla de Jadraque. Significant deformation of Tertiary sediments is observed and reveals tectonic activity of the area in the early stages of the Tertiary before the major M. Miocene event that shaped much of the present-day physiography Spanish Central System. (see www.geo.vu.nl/~andb/iberia for full color version)


Chapter 3

into silty marls, clays and thin sandstone with abundant paleosols. This clearly indicates a sudden change of depositional setting between units 1 and 2: the base of unit 2 can be considered to be flood deposition of an alluvial fan, but most of the unit is attributed to more distal fan to playa-lacustrine facies. Unit 3 displays an erosive contact with unit 2 and consists of ribbon-formed conglomerate bodies alternating with marly and pebbly sandstone. In the base debris/mudflow deposits and paleosols are common. The succession fines upward into red sandy silt alternated with clean pink sandstone. Unit 3 is interpreted as deposits of a relatively small and immature fan, in comparison with the fan of Unit 1. The entire sequence is topped by erosive and only slightly folded M. Miocene conglomerates. The thicknesses of units 2 and 3 vary significantly due to synsedimentary folding (tapering). Together they form a progradational small fan system in an area with small-scale topographic differences. Unit 1 can be attributed to larger scale topography, possibly related to important fault activity in the source area (the junction between SCS and IC). Fan architecture and stacking patterns, as well as synsedimentary deformation structures indicate that at least the northeastern part of the SCS has been tectonically active from the Early Paleogene onward. Near Beleña de Sorbe and Cogolludo, the first appearance of erosion products of the SCS occurs in Upper Eocene - Lower Oligocene sands containing fragments of schist (up to 5% of the grains) [ITGME, 1990b]. Upper Eocene to Oligocene conglomerates even contain over 40% of clasts > 2cm of quartzite and schist, which show clearly provenance of the eastern SCS. While Paleocene to M. Eocene sand contains up to 15% fragments of dolomitic rocks [ITGME, 1990d] (erosion of Mesozoic cover), U. Eocene to Oligocene sand show deeper erosion into the SCS: 30% dolomite and limestone fragments and 5% fragments of schist. Increased tectonic activity of the SCS is shown by the increasing size of the erosion products: the Oligocene sediments consist of a coarsening up sequence of up to 660m mud, sand and conglomerate. The fraction of pebbles larger than 2cm show an average of ~55% Mesozoic provenance (15-35% limestone, 25-40% dolomites) and ~45% basement (35-45% quartzite, 0-7% slate and schist). The higher levels however contain higher contents of slate and schist and therefore show ongoing unroofing of the SCS. Clearly these sediments are synorogenic.

Fission track data The fission track (FT) technique has been used to contribute independent data on vertical motions and uplift/denudation of the study area. Fission tracks are deficits formed in a crystal lattice by spontaneous fission of radioactive elements of uranium. Originally they typically have a specific length of about tens of micrometers, depending on the material they were formed in. At high temperatures, these tracks close (anneal) fast after formation, while at low temperatures annealing can be disregarded. The thermal level for which this relation changes, is different for a variety of minerals. In the case of the mineral apatite, the tracks disappear immediately (“anneal”) at temperatures higher than 120o Celsius. Between 120o and 60o annealing is taking place at a considerable rate, and this zone is called the Partial Annealing Zone (PAZ, see gray box in Figure 3.2.7b). When apatite is cooling below 60 degrees, tracks are not altered. This implies that when a sample moves quickly through the PAZ, the tracks are relatively long and show a low scatter in length. Slower cooling through the PAZ causes reduced track lengths and a broad track length distribution. The temperatures defining the PAZ for apatite are low when compared to other minerals and therefore allow a reconstruction of the last part of the cooling trajectory. Using information on the increase of temperature with depth in the Earth’s crust makes it possible to infer a thermal history for the sample and ‘translate’ cooling into uplift towards the surface. An age can be inferred for the 71

Chapter 3


Figure 3.2.7 Example of the results of fission track analysis. Left panel: track-length distribution in sample (bars) and calculated from modelled thermal history (line). Right panel: modelled thermal history and age for the sample (after de Bruijne [2001]). See text for brief explanation.

moment the sample entered the partial annealing zone on its way to the surface (FTage). In order to be able to reconstruct uplift and denudation from the inferred cooling history, information is required on the geothermal gradient and topography through time. In many cases this information is not available, causing one of the main uncertainties in inferring vertical motions from fission track data. Another source of concern is the variations in chemical composition of the mineral apatite and thermal activity. This rather brief and condensed outline of the fission track method is sufficient to understand the data and inferred conclusions presented in this section. More concise reviews of the method and the involved analytical and numerical techniques can be found in amongst others Gallagher et al. [1998] and Andriessen [1995]. For this purpose, samples were taken from both the Paleogene fan deposits as the Precambrian basement of the SCS. Samples of the Paleogene (~45-40 Ma old) fan deposits yield Miocene FT-ages (~20Ma). This means that they experienced temperatures >120oC after deposition. Assuming a paleogeothermal gradient of 30oC/km, the thermal history of the samples has been modelled, which indicates that the sediments have been buried to a depth of 34km. A huge amount of sediments must have accumulated over the area between Eocene and earliest Miocene. Moreover, the same amount of overburden must haven been eroded in combination with uplift to the present elevation of 800m from ~20Ma until present. Independent results from a sample of basement (Precambrian gneiss) just north of the Paleogene fans yielded slightly older ages (~29Ma). Length distribution shows a nice peak and therefore uplift must have been rapid. An increase in uplift rate for the same period as uplift of the Paleogene sediments is inferred from the modelled thermal history. Samples of basement material more to the northwest in the SCS show very old ages for the FT-data of up to 180Ma [De Bruijne, 2001] and the thermal history for these units shows very slow cooling through the PAZ. Between these sample points, intermediate ages have been observed, which have been interpreted as resulting from block movements according to the ‘pop-up’ interpretation of De Vicente et al. [1996c]. These fission track results have been combined in a model for the evolution in time and place of uplift of several of the separate basement blocks in the SCS [De Bruijne & Andriessen, 2000]. The model suggests onset of limited uplift of the northern blocks during L.Eocene times, with an important increase up to 20Ma, and related shedding of coarse detritic material into the Madrid Basin. In the late stage evolution (middle-upper 72


Chapter 3

Miocene) the southern border fault that was activated during the initial uplift became inactive and the border stepped basinward to the present-day basin edge. This sequence of events would explain the absence of coarse Paleogene deposits along the edge of the basin in this area: uplift of the block they were deposited on in earlier stages has removed them by erosion. Uplift along the southern border of the SCS occurs mainly along the northward dipping faults; the back thrusts play a minor role in this model. The less steep northern border fault of the SCS causes only limited uplift of the northern SCS, as shown by the high ages for samples from this part of the mountain belt. A much more detailed description of fission track results for the entire SCS and their implications on the evolution of this intraplate mountain chain are addressed in a PhD thesis by De Bruijne [2001]. The fission track data and thermal histories presented in this section are taken from this thesis.

Kinematic indicator data Some publications on paleostress data are avialable for the area under consideration. Sanchez Serrano [1991] measured kinematic indicators in the region at 5 sites in both the Mesozoic basement sequence and the Miocene infill of the Madrid Basin. A predominant N140o compression was observed, but in the Mesozoic sequence an additional N060o compressional event was inferred from the fault slip data. Paleostress determination in the Madrid Basin shows a more or less consistent N140o-155o direction of Shmax from Middle Miocene to present-day [De Vicente et al., 1996a]. Kinematic indicators measured in Upper Cretaceous rocks yield similar Shmax directions of N320o/N150o (sites 1,2,3,12&18 in Table 1 of [De Vicente et al., 1996a]) and have been attributed to the Middle to Upper Miocene stress field as well. Muñoz Martín [1997] documented the stress field at the border of the Iberian Chain (IC) and the Loranca Basin (see Figure 2.2.3.1). This author showed that the stress field changed from a predominantly N055o (‘Iberian’, Oligocene) through a N100o stage (‘Altomira’, Upper Oligocene-Middle Miocene) to the Middle Miocene to present N155o ‘Guadarrama’ field. The Altomira stage has been explained [Muñoz Martín, 1997] to be the result of superposition in time and place of the Iberian and Guadarrama far field stresses caused by the Pyrenean and the Betic collision, respectively. Further constraints on the stress field in the region come from the nearby Almazan and Zaorejas basins. The Almazan Basin is located between the two branches of the Iberian Chain; Maestro González and Casas Sainz [1995] performed an extensive study of kinematic indicators in the basin. 17 out of 55 measurement sites are located in Tertiary sediments, but the resultant tensors are only dated relatively with respect to each other. Bond [1996] offers a general tectonic evolution scheme for the Almazan basin based on detailed seismic data interpretation. Combining this evolution scheme with the tensors presented by Maestro González and Casas Sainz [1995] enables establishment of the following deformation sequence: (1) Mid Eocene NE-SW compression, (2) Late Eocene/Oligocene N-S compression, (3) Mid-Late Oligocene N-S to NNE-SSW and finally (4) Early to Mid Miocene NW-SE compression. The Zaorejas Basin is located in the western Iberian Chain and formed by an E-W striking synclinal structure related to a fault-propagation fold [Rodríguez Pascua et al., 1994]. The sedimentary infill of the basin can be divided into two sequences that are separated by an unconformity. The lower sedimentary sequence (of presumably Eocene-Oligocene age) shows growth strata and therefore is syntectonic. Maximum horizontal compression in this sequence shows a N150o-160o trend. The upper sequence, overfilling the basin, is of Oligocene- Early Miocene age and shows N-S to N20o maximum horizontal compression due to late stage folding of the area. These results suggest the existence of a stress field before the Late Oligocene 73

Chapter 3


with the same Shmax as the Middle Miocene to present-day stress field. Both branches of the Iberian Chain and the southern edge of the Ebro Basin provide additional data by studies of kinematic indicator data ([Alvaro, 1975]; [Simón Gómez and Paricio Cardona, 1988] and [Perez Lorente, 1990]). Because the stress field in the region under consideration from M. Miocene to recent times was already well documented [De Vicente et al., 1996a], observation of kinematic indicator data for the E. Tertiary to M. Miocene has been a primary goal in order to unravel the tectonic evolution during this time span. Therefore kinematic indicator data (fault slip, joints, striated and pitted pebbles) were gathered in Upper Cretaceous - E. Miocene sediments along the southeastern border of the SCS. To examine the effect of the deformation of the IC on the stress field in the study area, observations were expanded sideways along the SBF. In Figure 3.2.10 the distribution of 10 sites where kinematic indicator data were gathered is shown. Extra attention was paid to the area around Pinilla de Jadraque because meso-structures in this area enabled relative and even absolute dating of several of the deformation stages. The paleostress method (see Chapter 2) was applied to determine the stress directions related to different stages of deformation. Using the method proposed by Schrader [1988], (see Chapter 2 for theoretical background) σ1-directions were determined from pitted and striated pebbles at 6 outcrops, four of which are situated within the Pinilla-area. Development of this type of kinematic indicators was facilitated by the nature of the Paleogene and Miocene sediments in the study region, consisting of quartz-matrix supported conglomerates containing a large amount of limestone clasts. Table 3.2.1 shows all the results for the different sites where kinematic indicator data have been observed; several of the sites will be discussed below in further detail. A first overview of the data suggests the existence of reactivated faults: it clearly shows a multiple deformation history and it is hard to detect clear conjugate sets of faults. Figure 3.2.2a shows a section along the Cañamares River along the village of Pinilla de Jadraque to elucidate the position and tectonic setting of outcrops Pinilla1 and Pinilla3 that will be discussed subsequently. Pinilla1 (Figure 3.2.8a) In steep southward dipping conglomerates (S0: 170/79), a total of 20 pebbles and 26 fault surfaces with kinematic indicators have been observed. The maximum compression direction derived from the pebbles indicates a mean of around 180/15. The population however can be divided in two subsets: One set (14 orientations) with dips less than 30o and a smaller set with dips up to 70o. This seems to indicate a late stage tilt of the small set, possibly related to folding. The general fracture pattern trending N355o fits well in this nearly N-S compression. Interrelations of the faults and striations on them can help to deduce the relative age of the observed stress fields in the region. Although the tensor solution is good, reactivation of the observed faults is more than likely: many of the observed faults reveal oblique slip. On a few planes dipping ~045/30 a second set of overprinted, faint dip-slip striae are observed. It is neither possible to determine movement sense nor to obtain confidence whether these are younger or older than the oblique slip components. If, based on regional knowledge these striations are supposed to show up-dip (inverse) motion, than together with more confidently determined sense of movement at a single fault plane (oriented 135/48) it is possible to obtain a confident tensor solution.

74


Chapter 3

Figure 3.2.8 Observations and paleostress results for Pinilla1-3. See Table 3.2.1 for explanation of abbreviations. See Figure 3.2.4 for location of the stations. 75

Chapter 3


Pinilla3 (Figure 3.2.8c) This key outcrop, with respect to relative dating of structures is positioned directly north of the steeply dipping U. Eocene- L. Oligocene conglomerates of the Pinilla1 outcrop. A nice fold has developed in the M. Eocene marl-limestone succession. In the core of this frontal anticline, several low angle, top to the SSW, thrusts are observed related to the first compressive deformation of the study area during the Tertiary. Closer examination reveals rather oblique striae in one case overprinting faint dip-slip striae, which suggests late stage reactivation. Aleas (Figure 3.2.9a) An outcrop located a few kilometer north of Aleas, in the core of a gentle near E-W trending syncline. The σ1 direction as inferred from pitted pebbles (N321o) is reflected in a tensor for few (6) of the fault slip data (355/10). Although the quality of this tensor is mediocre, it has been included in the analysis based on the results of the pebbles and the occurrence of a similar regional stress field. A second, better tensor is obtained in this section for a pure compressive stress regime with a small perpendicular extensive component with Shmax ~N050o. Cerezo de Mohernando (Figure 3.2.9b) In a very recent road cut just west of Cerezo de Mohernando up to 27 fault planes and 11 pebbles with kinematic indicators have been measured. The resulting σ1 direction from the pebbles, N354o, was more or less reflected in the resulting N334o Shmax of a strike-slip deformation regime that explains 14 out of the 27 faults. In this area, the resulting stress fields have a larger extensional component to them than the results in the Pinilla area. Most likely this is related to minor halokinematic movements in the triangle Aleas-Cerezo-Espinosa.

Figure 3.2.9 Observations and paleostress results for Aleas and Cerezo de Mohernando. See Table 3.2.1 for explanation of abbreviations paleostress results. See Figure 3.2.4 for location of the stations. 76


Chapter 3

Barbareja

Paleostress information obtained from:

Pebbles

This study

Pinilla3 Alcorlo

Pinilla5

P3

n Ra

Literature

n ia er Ib

Fault slip

ge

Aleas Tamajon Retiendas Valdesotos

’ 3 15 o

m

yste

al S entr

C

o 5’ 32

Pinilla1 Pinilla4 Huermeces

Madrid Basin o 3 25’

Stratigraphy

Espinosa de Henares o 3 15’

Quaternary

Pinilla2

undifferentiated cover

o 3 00’

Rio Henares

Paleozoic basement Beleña de Sorbe

Structural elements

Cerezo de Mohernando Figure 3.2.10 (above) Results of the paleostress analysis combined with result from previous studies.

Fault

Table 3.2.1 (below) Table showing all results of the observations at the paleostress stations in the study region. Code equal to the SIGMA-classification [SIGMA, 1998]. NF= total number of faults, nf= number of faults explained by tensor, orientations of the principal axes of stress and the stress ratio R. α= angle of deviation between the calculated striation and the observed one on a plane. TQR= Tensor Quality Rank, following Delvaux [1994]: A, B, or C for good, reliable, and unreliable respectively. Station

code

age sediments

Aleas

486/01

U.Eocene- U. Oligocene 1st 2nd pebbles U.Cretaceous U.Oligocene

Barbareja 460/01 Cerezo de 486/02 Mohernando

NF

pebbles

Espinosa Huermeces del Cerro Pinilla1

486/03 461/01

L.Oligocene M.Eocene- U.Oligocene

460/02

Pinilla2

460/03

Pinilla3 Pinilla4 Pinilla5

460/04 460/05 460/06

M.Eocene- U.Oligocene 1st & 3rd? 2nd pebbles M.Eocene- U.Oligocene pebbles pebbles M.Eocene- U.Oligocene M.Eocene- U.Oligocene pebbles M.Eocene- U.Oligocene pebbles

21 21 3 3 27 27 11 6 38 38 26 10 20 23 23 14 5 4

nf

s1

s2

s3

R

a

TQR

13 6 3 3 14 10 11 6 15 14 15 5 19 12 11 10 4 4

227/11 355/10 148/05 312/01 334/10 036/81 354/26 234/01 170/18 048/11 171/06 033/14 339/06 154/34 128/36 052/05 351/39 178/15

322/26 262/17

116/62 114/70

0.26 0.42

8.79 8.6

0.92 0.20

B C

Pure compressive Pure compressive

stress field

042/07 145/79 252/08

210/83 244/02 161/06

0.5 0.53 0.96

0.53 9.66 8.91

5.66 0.75 0.42

A B C

Pure compression Pure strike slip Extensive strike slip

347/86 261/04 332/78 079/11 303/00

143/04 003/73 140/02 286/75 211/76

0.42 0.57 0.46 0.5 0.6

5.55 9.17 7.45 7.63 5.76

1.08 0.65 0.69 1.13 0.43

B B B B C

Pure strike slip Pure compression Pure strike slip Pure compression Pure compression

145/31

314/59

0.6

12.7

0.56

B

Pure compression

In general two different states of tectonic stress have been detected in the study area. A compressive regime with σ1 oriented ~N140o-180o and a strike-slip regime with a N030o060o orientation for σ1. Generally, the σ1-direction derived from striated pebbles shows a very consistent near horizontal 140o/320o trend. For many of the outcrops this direction is sub-parallel to the bedding of the conglomerates. However, in Pinilla1 this direction is at high angles with the sub-vertical S0. The steeply inclined beds of the growth strata show late stage shortening by nearly conjugate sets of thrusts that fit well the ~N320o oriented

77

78

y

Loranca

ain

ria Teruel

h nC Ibe

Middle Miocene - Pliocene

Basement

Mesozoic basement

Tertiary basins

0

L.Oligocene - Early Miocene

8W

36N

40N

44N

M. Oligocene

n Ce

ys lS tra

Duero

Ce

y

lS

ra nt

tem

m ste

Tajo

Madrid

Tajo

Madrid

Loranca

Zaorejas

Alma zan

Loranca

Zaorejas

Alma zan

Teruel

Teruel

Ibe

Tajo

Madrid

Zaorejas

zan

Alma

Loranca

Teruel

L. Eocene- E. Oligocene

Duero

2W

ria

ain

ria h nC

2W 4W 2W 4W Figure 3.2.11 Overview of development of the stress field through geological time in the region, based on the compilation of paleostress data in this thesis.

Ce

lS

ra nt

m ste

Tajo

Madrid

Zaorejas

M. Eocene

4W

ain

h nC

40N

41N

y

Ibe

Duero

Ce

lS

ra nt

m ste

zan

Alma

2W

Ibe

40N

41N

Duero

4W

40N

41N

40N

41N


ain

ria h nC


Chapter 3

compression derived from the pitted pebbles. This setting forms a lower limit for the age of the development of these striated pebbles and shows clearly that folding of the Lower Paleogene series was prior to the pebble deformation. In previous studies in the area, none of the Miocene outcrops yielded another stress field than a ~N140o-155o directed Shmax. Therefore, the striated pebbles formed during the Middle Miocene to present. In the Cogolludo area, the orientation of the stress field as inferred from the pitted and striated pebbles shows changes in σ1-direction from at around 320/05 to about 140/05 pointing to recent tilting or gentle folding, most likely related to limited halokinematic motions in the area. However, other directions are present in Mesozoic to upper Oligocene levels. The strikeslip field with σ1 oriented ~N040o-050o is found in up to upper Oligocene sediments (Espinosa), so this age can be considered as the oldest age of this stress field. A cross section perpendicular to the structures formed under ~N155o compression reveals large wavelength, low amplitude folding of the Paleogene cover (see upper panel of Figure 3.2.2). This folding that can be related directly to this stage of ~N040o-065o oriented compression affected sediments up to Upper Oligocene as well, whereas Lower-Middle Miocene sediments display an angular difference with the folded strata (N. of Jadraque). Apart from some indications in outcrops Pinilla3 and Pinilla1, no hard proof can be found for an older approximately N-S compressional stress field. On the other hand, observation of the same sort of stress field in Cretaceous cover and Miocene deposits in the Madrid Basin does not a priori exclude a different age for both. A compilation of the new data on the stress field with data from literature (Figure 3.2.11) shows a clear regional evolution, which is very similar to the observations in the study area. Going back in time: (a) the present-day N-S to N140o directed Shmax trend, is observed in the entire region. In the east N-S prevails, while in the western part the NW-SE directed Shmax is dominant. This fanning of the present-day stress field is observed as well by the SIGMA results [SIGMA, 1998]. (b) The early to middle Miocene stress field was oriented near E-W in the Sierra Altomira and rotated towards NE-SW in the eastern sector. (c) During the late Eocene to late Oligocene, the dominant Shmax direction was in general between N015o and N060o. (d) Some dispersed results indicate a pre-L. Eocene near NS orientation for Shmax.

Discussion and new model for the tectonic evolution of the SCS The Paleogene sediments drilled in the Madrid Basin and their internal structure indicate that during the Paleogene an initial stage of limited loading and foreland basin development in combination with limited foreland deformation (bulge development and erosion) occurred. The distal type of Paleogene sediments suggests that the loading thrust belt was located NW-wards of the present-day location of the SBF. The thrust near Pinilla, Beleña de Sorbe, and north of Venturada might very well have been the active southern border fault during this stage. Similar timing of deformation is observed in the nearby Almazan basin. In the Mid Eocene, erosion of the Cretaceous cover of the SCS occurred, is documented in the study region, while in the Almazan region active basin subsidence started. During the Late Eocene- early Oligocene N-S oriented compression and related basin formation as observed in the Almazan basin (northwest of the SCS), which coincided with the first appearance of SCS basement erosion products in the sediments in the study area, indicate regional uplift. The Eocene fan deposits near Pinilla have been buried by up to 2-3 km of younger sediments, as revealed by fission track data. In the El Pradillo well up to 1450m of late

79

Chapter 3


Oligocene to early Miocene deposits have been drilled, but more to the east (at Santa Barbara), this level is reduced to around 800m. The unit below this sequence has not been divided into smaller segments than Maastrichtian- middle Oligocene, so part of this (1400-800m thick) sequence can be taken into account for burying the Eocene sediments. The subsequent middle Miocene – present-day uplift of the SCS was more prominent towards the west. This variation can be deduced from the fact that the proximal Paleogene alluvial fan type sediments that crop out extensively near Pinilla have been largely eroded near Venturada, which is the westernmost outcrop of Paleogene sediments along the northern Madrid Basin border. Further west, only metamorphic and plutonic basement outcrops in the same tectonic block. The same pattern is shown by the fission track results. Uplift of 2-3km for the Paleogene fans increases to up to 5km near the Sierra de Gredos. This variation can be explained by M. Miocene active strikeslip faulting accommodating a lot of deformation along and even within (west of Beleña de Sorbe) the eastern part of the region, while the same amount of shortening is accommodated in the central and western SCS by mainly thrusting. From both the model inferred from the fission track observations and the seismic reflection profiles through the basin edge, we concluded that thrusting in the M. Miocene occurred more basinward than the border fault that was active during the EoceneOligocene. The latter was not very active during the late stage (M. Miocene) compression as proven by fission track data. A possible reason for this is that the fault has been locked by the Oligocene strike-slip deformation. The proposed scenario has been tested in a quantative way by a 2D numerical model that simulates lithospheric compression and calculates the thermal consequences of this compression [Ter Voorde et al., 2001]. With the SCS geometry and the proposed scenario as input, this numerical model can reproduce the thermal histories of the fission track samples. Some of the models proposed by others for the development of the SCS can be discarded based on the data presented and synthesized here. The demonstrated tectonic activity of the SCS during the Paleogene-Oligocene took place before the main collision of Iberia with the Betic/Alboran microplate (starting M. Miocene). Therefore, models that explain the evolution of the SCS as being the result of the M. Miocene event only should be disregarded. This includes models for the SCS incorporating a crustal detachment running from the Betics to the Northern Border Fault of the SCS, transporting Iberian upper crust and the Madrid Basin as a piggyback basin northwards [Ribeiro et al., 1990]. The model proposed by De Vicente et al. [1996c] explaining the SCS as a set of pop-ups being active during M. Miocene only, seems to be incomplete with respect to the initial phase of deformation. The data presented show that some of the faults that were active during the Eocene-Oligocene were not reactivated in a later stage. Moreover, the previous deformation in the entire evolution of the SCS was of major importance in the more recent tectonic evolution. Because strike-slip deformation in an early stage of deformation has been observed, the block rotation model by Vegas et al. [1990] is appealing. However, block rotations in the northeastern SCS have been very minor, if they occurred at all.

80

Serravallian- Tortonian (~12 Ma)

M. Oligocene (~30 Ma)

MB

IC

10

20

30

40km

Flexural basin and bulge develop

Major uplift, varying for different blocks ~3500-5000 for the southern part ~1000m for the northern part

Major thrusting along southern border of SCS along other fault more basinward. Reactivation of complex strike-slip zone more difficult and limited

Popping-up of the SCS, thrusting over both northern and southern foreland

active alluvial fan

subsiding area

uplifting area

Isolated fans Overthrustiung of the IC, shedding of conglomerates.

Strike slip tectonics of basement faults SCS Flowertype structures(en-echelon folds) in the thin mesozoic cover. Local topography

0

Appr. scale

Compression direction

First deformation of Mesozoic/Lower Paleogene cover of SCS

IC major dextral strike slip

Large scale uplift of north-eastern SCS (undifferentiated block)

Large sheets of conglomerates shedding of SCS and IC filling the MB

Figure 3.2.12 Schematic illustration of the three main stages in the new model for the Cenozoic tectonic evolution of the area located at the junction of the Spanish Central System and the Iberian Chain. Left series shows a block diagram of this evolution, while the right shows a map view perspective.

Distal fan sedimentation

Active proximal fans

Oligocene-E.Miocene

Eocene-E.Oligocene

Mesozoic-E. Tertiary

Basement

N

S SC

Middle- Late Eocene (~45 Ma) N


81

Chapter 3


By combining parts of the previously proposed models with the new data, new and in time (Paleogene- Miocene) extended model for the evolution of the SCS (see Figure 3.2.12) is proposed: (1) Middle –Late Eocene (~45Ma) General but limited uplift occurs of at least the NE SCS, which leads to erosion of this hinterland and sedimentation of large conglomerate (proximal) and silt/clay (distal) sheets in the MB. Deformation of the Mesozoic/Lower Paleogene cover under NNW-SSE compression begins, which is related to the start of convergence between Eurasia and Iberia (the very onset of Pyrenean collision). (2) M. Oligocene (~30Ma) NNE-SSW to NE-SW compression occurs, caused by collision of Eurasia and Iberia along the NE margin of the Iberian plate, inverting the Mesozoic Iberian Basin into the Iberian Range. The Iberian Range is thrust onto the eastern edge of the Madrid Basin and this deformation causes left-lateral transpressional strike-slip motion along a ~N070o oriented crustal Late Hercynian fault. The thin Mesozoic-lower Paleogene cover on top of the basement is folded enechelon. As a result of transpressional flower structures, local relief develops. Progressively, structural highs develop while sedimentation continues, resulting in growth structures and isolation of local basins. These local basins eventually become filled with isolated conglomerate bodies. The strike-slip deformation related to this phase diminishes away from the active Iberian Range front. (3) Middle to late Miocene (~12-9Ma) Major differential uplift (1000-5000m) of the SCS with respect to the surrounding basins occurs. Inherited basement faults are reactivated to form a series of pop-ups and ‘pop-downs’ under NNW-SSE compression. The tectonic activity along the southern border steps basinward by activation of a new thrust. At least the NE part of the Spanish Central System has been tectonically active during the middle Paleogene. A first stage of deformation in the area took place under a stress field similar to that active during M. Miocene to present-day, so it is hard to discriminate features formed during either of the events. Deformation during this first stage of development of the mountain range and related basins has been limited, compared to the late stage evolution for the region. However, both Eocene and Oligocene compressional deformation events have contributed significantly to the distribution and type of deformation during the M. Miocene to present-day in the southeastern SCS, resulting in the present-day configuration of the Madrid Basin and Spanish Central System. The combination of data from several fields in Earth Sciences has unraveled the tectonic and geological history of the study area. This proves how important integrated studies are in present-day geology. Using either of the components as a stand-alone study, it would never have been possible to end up with a model that explains the geological structure of the region through time.

82


Chapter 4

CHAPTER 4 - PALEOGENE GEOLOGICAL EVOLUTION OF THE IBERIAN PENINSULA AND WESTERN MEDITERRANEAN Introduction The European Alpine System is a classic zone of continent-continent convergence and collision. Numerous studies over the last decades addressed the tectonic and paleogeographical evolution of the closure of the Tethys Ocean and the evolution of the Alpine orogeny (amongst others [Savostin et al., 1986]; [Biju-Dival et al., 1977]; [Dewey et al., 1989]; [Yilmaz et al., 1996]; [Ziegler, 1988] and [Stampfli et al., 2000]). Southern and Central Europe is characterized by large-scale inversion structures, arcuate fold-and-thrust belts (Western Alps, Carpathians, Betic-Rif), rotating crustal blocks, and limited subduction and extension within an overall convergent setting. Many authors have studied in great detail the puzzling contemporaneous development of all these features related to the evolution of the Western and Central Mediterranean. The wealth of observations has provoked the proposition of a wide range of ideas, models, concepts and hypotheses. The general outline of plate tectonic evolution in the Mediterranean region has been the topic of several studies over the last decades (amongst others [Yilmaz et al. 1996]; [Ziegler, 1988]; [Dercourt et al., 1986]; [Dewey et al., 1989]). In spite of the large amount of publications on the development of the region through time, all of the publications turned out to not have the temporal or spatial accuracy required for my purpose. Yilmaz et al. [1996] does not include maps for the period between 33.5 to 10.5Ma, during which major plate reorganizations occurred in the region. Some are at a very large scale [Stampfli et al., 2000] or concentrate on the central, oceanic part of the system [Dewey et al., 1989] and do not provide information on the Iberian Peninsula. Others pay attention only to a part (post-25Ma) of its Tertiary evolution [Gueguen et al., 1998], or on paleoenvironments [Dercourt et al., 1993]. Therefore, a compilation, which is presented in this Chapter and Chapter 5, was made from data from literature and additional own unpublished data. The compilation will be presented in maps showing (a) the paleo-position of present-day coastlines, (b) active tectonic structures (faults/folds), (c) sedimentary facies provinces (eroded continent/continental deposition/shallow marine/deep marine/eroded continent) (d) Paleostress results from kinematic indicator data and stress trajectories (direction of maximum horizontal compression) and for a few time spans (e) estimated paleotopography and vertical motions. The many studies mentioned before have been very useful and convenient as starting points for this compilation. In contrast to the Iberian Peninsula, my personal knowledge of e.g. Northern Africa and Southern France is limited mainly to literature. Thanks to several excellent overviews of the tectonosedimentary evolution of regions outside the Iberian Peninsula (e.g. Sissingh [2001] for SE France or Wildi [1983] for Northern Africa), the study area could be extended. This incorporation of the areas surrounding the Iberian Peninsula is relevant to understand the large-scale tectonic processes that caused the intraplate deformation of the Iberian Peninsula and to put its evolution in a broader tectonic context. The reconstructions presented should be regarded as a state of the art reference database, which will guide the interested reader to more detailed descriptions of regions and features rather than an authorative and final model for the evolution of the region.

83

Figure 4.1 Overview of the Iberian Peninsula with the names of basins, major faults and regions that are used in the description of the geological evolution of the area. (see www.geo.vu.nl/~andb/iberia for full color version) Horse Shoe Abyssal Plain

12W

36N

e ng rri Go ank B

an

Arrabida

T

8W

Algarve

do

Sa

r we

Lo

Bierzo

la

Range

iel

Miranda- Basq ue Urbese Pamplona

len

Gulf of Cadiz

a

Rharb

Strait of Gibraltar

Almuñecar

El Jebha

4W

as

Pyrenea n

ian

Norther n

Aquita n

dg

e n ra bo Al eima Al Hoc

Ri

A

0

liff Che

aus h Plate Hig

ea nS lbora dge

Yussuf Ri

Sr. Carrascoy

Murcia Cartagena/ Lower Segura

Alhama de Murcia

Vera Palomares Nijar Sr. Alhamilla Carboneras

Sorb

za Ba

Fortuna

Lorca

Sr. Segura

cs Beti

Guadix

Sr. Nevada

Granada

rn ste We oran Alb

f

Ri

Gibraltar Arc

Gu

lq da

u

r ivi

na ore ra M

Sier

iana

taj o

Guad

Ib

Tajus Abyssal Plain

Ridge Lisboa Torre -

Iberia Abyssal Plain

tan

Landes Palteau

ain Chr. Denca

38N

40N

Galicia Interior

e

se

42N

Sarria

Cantabrian

Ve n

Inner Basin Llanera

Danois Bank

Teruel

s eare Bal

oir an gne N ini ult Monta t. Ch Hera S Conflent Agly

n i Bas

untains Tell Mo

Mitidja

Alg

n eria

Hercynian structures

Paleozoic igneous rocks

Paleozoic metasediments

Mesozoic

Basement

Volcanics

Basins

Tertiary Quaternary

Rivers Basin Sr. Alhamilla Elevated region Palomares Major fault Valencia

Tet Pyre nees Jaca Rossello Sr. Obarense Ainsa Tec Monforte Cerdanya Sr. Mts. de Oca Rioja C Monts De ameros La Bureba ec Trem Gu Tui Leon p ara Sr. De la Demanda e Emporda r Mnts. g Vallfogona Ager Xinzo Olot Ca Se Srs. Eb ta Penedes Verin Duero Margina la r o y le r s ud e e r d t r Alma -D la Montnegre nas CR Bo zan lel ar Bo des C oc Ca ern Pra mo a rth a So rra f y No ra Mo o Gar sie m Vilarica nte , oz Valles e s Ambles L a Vilarica , yst Zaorejas Jil o El Camp a S on rig lb o g A d ca o Ponsul rra d R Gata a er Tajo Rubielos Ta da e tes al ram ord de Mora Ciu r. D br e Maestrat h S ntr adarern B lum e ug Co C Gu uth os r. o Tro la S S Madrid Sr. De Gred e a r i c Loranca st Mijares na .E ta Menorca Ribesaldes/ len Sr un Va Alcora m Castelo a r Toledo Mnts. .T Branco al Sr rc Cabriel Mallorca Ce Valencia Ibiza ajo

Galicia

Villalba

Oviedo

Bay of Biscay

S ue C

Pla

Galicia Bank

Meirama

As Pontes

Ortegal Spur

ni

ita Lu s

sp

-A

He

if ss Ma c ri

nc ia

n

es

r

ia er

Un

al

ko

44N

Sr. Altomira

84 rs

e iv

Ne



Chapter 4

FRANCE

Rhine

d

Ba

si n

Paris

Fo

re

la n

Bresse Al pi ne

JURA

er

n

Limagne

rth No

SWITZERLAND

Roanne

PS

AL

E PR

Rhone

Forez

Penninic Thrust

BRIAN

C, ONN A

IS

Valence

MASSIF CENTRAL

WESTERN ALPS

VERCORS

ITALY

Cevennes Fault

Valreas

Vi st re nq ue aul t

Her

ne

Nimes Camarque

o

sell

Ros

Tec

nc ra

PROVENCE Cenozoic basins

an m

n

Marseille

ra

be

Fa

ra

Aix Montpellier

Basin Fault Region

nd

GULF OF LION

G

ra

lG nt Ce

t Te

ra

nean North Pyre

Aix

Valensole

Volcanic centers

on rb

St. Chinian thrust

N

c do ue g n La

Na

MO

G NTA

Apt

Les Matelles

Aquitanian Lodeve IR

West A Forela lpine nd Ba sin

Manosque

e

Londres

O EN

Nimes Fault

Ales

Du

Montpellier thrust

GULF OF LION

45N

LIGURIAN/PROVENCAL

PYRENEES

SPAIN

Emporda

MEDITERRANEAN SEA

35N 0

10E

Figure 4.2 Overview of the Southern France with the names of basins, major faults and regions that are used in the description of the geological evolution of the area. Compiled from: Sissingh [2000], Sérrane [1995]; Vially and Trémolières [1996]; and Roure and Coletta [1996]. 85

Anti Atlas

SHAF

Tizi n'Test fault

10W

Souss

al

n

Rif

ica Afr

d

in

a

5W

Ouarzazate

SHAF

tlas High A

Sardinia

ri Ibe

lgerian North A Lesser Kabylia Greater Kabylia Mitidja Constantinois Cheliff ins Tell Mounta nd basin n forela a ic Subrif fr A Northern Tunesian Atlas

lan fore

arg

nM

llia Te

L lia by a e K ain re er Ch lcai at re Ca G

0

Saharan Platform

Rharb Prerif Guercif n s s oca Atla r seta r a e n l o M t a a n M set A NMAF har Ora le Sa Me d d i M Moulouya

Dors

de

30N

35N

n Rifia

ui

A

M ag

Mesorif Prerif

al

Cam Gib po de ralt ar Intrarif

ride

es

id

Corsica

Ligurian

15E

Cenozoic basins

B

Calabria

Sicily

Basement

Tyrrhenian Vavilov Marsili Sardinia Channel

Cornaglia

a

bt Se

rride

l ca en v o Pr

rsic

ma

Alpuja

ylia

ab

rK

e ess

Peloritanian

Alpine Corsica

Ap

Gho

IS

Calabrian

Corsica

10E

es

N

IU ES

IP

do e va rid Ne ilab F

Overthrust

Subduction/ancient subduction

5E

Co

tic Be r s a l Do e od mp ltar Ca ibra G

EP

Young oceanic crust

Nevado - Filabride (s.l.)

Alpujarride (s.l.)

Malaguide (s.l.)

Dorsalian (s.l.)

Betic/Alboran units

Iberian foreland

IS = Internal Subbetics

ES = External Subbetics

IU = Intermediate Units

IP = Internal Prebetics

EP = External Prebetics

Present day shore line

in

Predorsalian

n en

Ma ghr ebi an

86

Iberian Margin

40N

45N

Figure 4.3 Panel A: ~Middle Miocene reconstruction of the Alboran/Betic block and surrounding margins. Division after Sanz de Galdeano [1990] and Fontboté and Vera [1983]. Panel B: Basins and structural domains in northern Africa and the central Mediterranean basin.



Chapter 4

The general timescale used is the Cenozoic timescale by Berggren et al. [1995]. The paleo-position of the different crustal blocks/tectonic plates and their motion through time are based on the database of the Ocean Drilling Stratigraphic Network (ODSN), which can be found at http://www.odsn.de/odsn. The rotation poles on which these reconstructions are based are included in the ODSN website. Although one might argue about the correct position in this reconstruction of the many minor crustal blocks that have constituted the Mediterranean region through time, a reference database on the block movements has been chosen to enable understanding of direct implications of future improvements and new insights on the tectonic evolution of the area. In this chapter results of the compilation are presented for the Paleogene, the time interval during which the Pyrenean collision was the major plate tectonic event in Iberia. In 7 compilation maps (for 65, 54, 42, 36, 30, 27 and 24 Ma), the major trends and significant changes with respect to previous time slices will be discussed in detail. The many references on which maps are based allow for more detailed regional information. Names used in the text for regions, faults, basin and other features are depicted in a few regional maps: Figure 4.1 for the Iberian Peninsula, Figure 4.2 for southern France, Figure 4.3 for northern Africa, the western Mediterranean and the Betic/Alboran region. Figure 4.4 serves as legend for all the plates of the reconstruction.

65 Ma, KT-boundary, Figure 4.5 General Along northern Iberia and southern France, inversion of former Mesozoic normal faults occurs and subduction or underthrusting is active. The Pyrenean subduction system was linked with the Alpine subduction system [Stampfli et al., 2000]. Only the western part of the present Iberian Peninsula was emerged and had a planar and low relief, because the Mesozoic development of the western Iberian margin had not caused important rift shoulder uplift [Stapel, 1999]. Large parts of central Iberia were around sea level, so subtle eustatic sea level changes caused significant shifts in the position of the coastline.

Detail Western margin The Algarve is general emerging and a diapiritic phase is active [Terrinha et al., 1990]. In the Lusitanian area, the northern continental platform is about 500-1000m deep from Paleocene-Lutetian, sea covered northern Lusitania [Azevêdo, 1991]. Further north, onshore Galicia, peneplane surfaces are developing [Pagés Valcarlos & Vidal Romaní, 1998]. Northern margin General thermal subsidence affected the northern Iberian margin [Espina et al., 1996b], while subduction of the newly formed oceanic crust of the Bay of Biscay was active since Campanian [Ziegler, 1988]. The margin had a N120 trending structure and major sedimentation flux [Boillot & Malod, 1988], with erosion along the present-day coast and S.Cantabria [Rincon et al., 1983]. Central Iberia Since the Mesozoic, marginal environments of the Iberian Basin extended into Central Iberia [De Bruijne, 2001], gentle sea-rise or fall inundated or emerged large areas. At the transition towards the Tertiary, the central part of the Iberian Chain emerges finally [Adrover et al., 1983], shallow sea/lagoons and lowlands covered the region. The area was tectonically relatively quiet; a general (thermal?) subsidence was maintaining marine environments in parts of the region [Capote, 1983]. In the Almazan basin a stable shallow carbonate platform of L. Cret. Is topped by brackish to freshwater coastal swamp deposits of Danian age. A very similar situation was observed in the Sierra de la Demanda with the onset of nonmarine Tertiary deposition [Bond, 1996]. Large parts of Central Spain (e.g. the Loranca Basin [Diaz Molina & Lopez Martinez, 1979]) were covered by lagoon-beach environments [ITGME, 1990b] or just

87

Chapter 4


above sea level as for example in the northern Central System resulting in erosion or non-deposition [ITGME, 1991d]. S. Pyrenees and Ebro In the Ebro/Pyrenees area a large basin was present. The western basin was filled with marine (flysch) sediments, in the eastern part the ‘Garumniense’ (the continental equivalent) was deposited [Riba et al., 1983]. N. Pyrenees and SW France The southern border of Eurasia at this time, showed the first signs of the Pyrenean collision: the Aquitane basin slowly subsiding (onset of the flexural foreland basin) [Desegaulx & Brunet, 1990] while in the Provence compressive structures were general [Vially & Trémolières, 1996]. The same situation is valid with regard to the Alpine foreland due to the Briançonnais entering the subduction of the incipient Alps. The resistance to subduction of this continental block resulted in stress transmission into the Helvetic margin (northern part Valais Trough), where lithospheric deformation, uplift and erosion occurs [Ziegler et al., 1998] and even more distal in the European foreland (North Sea/Channel) [Ziegler et al., 1998]. The Briançonnais resisted underthrusting until M-L. Eocene [Stampfli et al., 1998], on its northwestern part ongoing sedimentation has been documented for the Paleogene [Schmid et al., 1996]. Catalan-Sardinian margin Stacking of ophiolitic thrust nappes ‘westward’ onto the continental margin of Corsica and Sardinia [Carmignani et al., 1995]. A subduction system dipping under the Iberian plate was activated along its eastern and southeastern margin during the latest Cretaceous and Paleocene [Ziegler et al., 2001] S. Iberia The southeastern margin of Iberia was under compression [Vera, 2001], resulting in erosion/nondeposition in the southern Prebetics [Kenter et al., 1990]. At the end of the Mesozoic, the Mulhacen complex (Internal Betics) reaches a peak in metamorphism (eclogite facies), increasing from east to west [Nieto Liñan, 1996]. N. Africa Carbonate platforms in a deeper marine basin covered large parts of northern Algeria. The stress field was N-S compression, resulting in E-W folding, NE-SW trending sinistral and NW-SE trending dextral strike-slip faults [Aris et al., 1998]. In the S.C. High Atlas more or less continuous subsidence and sedimentation occurred, but more to the west and south the first signs of uplift are documented [Görler et al., 1988]. Between the Middle and High Atlas, the Moulouya region is being uplifted [Morel, 1989]. The Oran Meseta constitutes an uplifted platform between the Atlas Troughs from Triassic and Miocene [Giese & Jacobshagen, 1992] and in the Tellian-Riffian margin erosion of the present-day foreland of Morocco and Tunisia [Wildi, 1983].

54 Ma, L. Paleocene - E. Eocene (Ypresian), Figure 4.6 General A change along the active plate boundary at chron 18 [Roest & Srivastava, 1991] is leading to large-scale deformation along the northern plate boundary. Clockwise rotation of Iberia continues in conjunction with Pyrenean-Cantabrian subduction, rapid convergence between Iberia/Africa with respect to Eurasia [De Jong, 1990] and the onset of sea-floor spreading between Greenland and Norway [Ziegler, 1988]. Collision occurs in all of the Pyrenean, Sardinia/Corsican, and Betic domains, and coincides with the full development of northwestward subduction of Tethyan oceanic crust under the Betic/Alboran and Corsica/Sardinia domains.

Detail Western margin In the northern Lusitanian margin a 500 –1000m deep marine basin persisted until Lutetian [Azevêdo, 1991]. Minor inversion off shore in the Lusitanian Basin can be correlated to onshore erosion of ~300m of U. Cretaceous sediments [Rasmussen et al., 1998]. Tectonic activity is observed in the Algarve/Lusitanian margin (NS directed compressional strike-slip [Lepvrier & Mougenot, 1984]) and southern Portugal as well. In the latter where alluvial fans are interfingering with lacustrine carbonates, first Tertiary movement of the Plasencia fault is inferred: the southeastern block upthrown, the Sado Basin down [Pimentel & Brum da Silveira, 1991]. Emplacement and uplift of the Gorringe Bank can be related to the change in plate boundary setting [Le Gall et al., 1997]. Northern margin

88


Chapter 4

The Asturian-Cantabrian margin is the locus of active southward underthrusting of the Bay of Biscay margin under Iberia, creating a fold-and-thrust belt in the Danois Bank area. Deep-sea sediments and basement are deformed and uplifted to even above sea level [Boillot et al., 1979]. Within the Cantabrian Range the first sedimentation occurs in the Oviedo Basin. Conglomerate is supplied from the north and onlapping to the south. The basin is endorheic at least for the Paleogene [Alonso et al., 1996]. Along the present-day Cantabrian coast (Santander) Alveolina limestone (shallow marine) is being deposited [Riba et al., 1983]. Central Iberia During the Paleocene, the Hesperic Massif (as the basement in western and Central Iberia is called) in western Iberia is being uplifted due to ENE-WSW extension with perpendicular compression, as inferred from the Duero Basin [Santisteban et al., 1996b]. Alluvial fans in the western part of the basin grade into littoral sedimentation in the eastern part [Corrochano & Armenteros, 1989], indicating that the eastern extreme of this basin was the coastline. The Sierras Obarense (S. Cantabrian) was an environment of marine-transitional-continental [Pol & Carbeillera, 1983], the Rioja area, a little more to the south, was at around sea level. Although the latter was still connected to the north with marine deposits, from the L Mesozoic onwards marine sedimentation never occurred! Paleocurrents of detritus in the Rioja area come from the south [Jurado & Riba, 1996], where gentle deformation of the Cameros Basin is testified by unconformities in a Paleocene regressive sequence in combination with open folding [Platt, 1990] and erosion of the Cameros massif [Muñoz Jiménez & Casas Sainz, 1997]. Increasing deformation of the Cameros Basin/de La Demanda can be observed towards the Almazan Basin (southwest of the Cameros) as well: 4 episodes of fluviatile sedimentation, coming from the NE (Massif De la Demanda) are evolving to more proximal [Pol & Carbeillera, 1983]. In the north side of the Almazan Basin, fluviatile sedimentation occurs in the synclinal Arganza Basin during the Ypresian [Floquet et al., ]. The rest of the Almazan was still under non-deposition or erosion [Bond, 1996]. The paleogeography of the Central System was completely different from its present-day configuration. Two separated basins existed: one in the Zamora/Salamanca/Avila/Toledo/NW Madrid and another E/SE of Madrid. The SCS did not exist as such, but another minor relief existed in between both basins [Portero Garcia & Olivé, 1983]. The continental basin in the western sector was wide and slowly subsiding distal flood plain with (continental) salt lakes [Portero Garcia & Olivé, 1983] in an open landscape without major relief nearby [Martín Serrano et al., 1996]. Alluvial systems put sediments into this basin from several sides. In the Penaranda-Alba area proximal alluvial systems have a southern provenance [Pol & Carbeillera, 1983], coming from the Guadarrama region where erosion or small scale alluvial fans, extending over the present-day mountains is documented for this period [ITGME, 1991d]. Related to the uplift of the Hesperic Massif, conglomerate & sandstone of an alluvial fan system entered the western side of the basin (Zamora/Salamanca) from the northwest, and in the southwestern region (Ciudad Rodrigo Basin) erosion or non-deposition occurred [Jiménez Fuentes, 1983]. The second basin in the Madrid area, in the present-day eastern Madrid and Loranca Basin extending east/southeast ward, was the locus of marine and coastal sediments during the Late Mesozoic -Early Tertiary. This implies that the eastern areas have experienced more pronounced uplift during the Alpine orogeny [Santisteban et al., 1996b]. Towards the southeast (Loranca Basin), unstable basins of transitional and/or continental deposition [de Torres Perezhidalgo et al., 1983] with coastal environments in the Loranca basin [Muñoz Martín, 1993], connection to shallow marine is not well known. In the Iberian Chain, detritic sedimentation occurs, but is not widespread [Adrover et al., 1983]. This can be related to active transpressional deformation [Ziegler, 1988]. The Cabriel basin infill changes from detritic to marls and gypsum [Adrover et al., 1983]. SE Iberia Generally, the southern an eastern margin of Iberia was emerged or became emergent during the L. Paleocene. Erosion is documented in the southwestern Valencia Trough with an estimated total of 5km erosion from 60-40Ma [Fernàndez et al., 1995], for the External Prebetics (in relation to deposition of red conglomerate) [Fontboté & Vera, 1983] and the Internal Prebetics, which was emerged during all of the Paleogene [Fontboté & Vera, 1983]. New topography was being formed to the south of the margin, related to extension? [Kenter et al., 1990]. The coastline was located near the Subbetics testified by a (few dispersed outcrops!) pink micritic limestone, called ‘Capas Rojas’ [Fontboté & Vera, 1983]. Towards the Eocene, the coastline [HNPC, 1992] shifted northward [Fontboté & Vera, 1983] leading to deposition of shallow marine limestone and sandstone in the Internal Prebetics, further south deepening to a ~1000m deep basin [De Ruig, 1991]. Water depth increased to 1000-2000m in the Sierra Alamedilla [Lu et al., 1998]. Betic realm In the Internal Betics, the Malaguide was being emplaced upon the Alpujarride leading to uplift and crustal thickening [Balanyá et al., 1997], related to the Balearic orogeny. To arrive at this important crustal thickening (up to 35 km [De Jong, 1990]) in the upper plate of the Corsican/Sardinian subduction system, a southeast ward shift of the subduction front might have occurred. Erosion of the Malaguide is

89

Chapter 4


evident from calcareous turbidites that enter the Predorsalian/Mauritanicas, which contain Malaguide fragments [Durand Delga & Olivier, 1988] such as limestone with Microcodium [Fontboté & Vera, 1983]. The Malaguide (including the Tellian-Rif arc) remains being eroded or an area of non-deposition until transgressive E. Eocene biogenetic sandy limestone is being deposited [Fontboté & Vera, 1983]; [De Jong, 1990], which can be correlated to the start of an extensional phase [Balanyá et al., 1997]. SE of the eastern Malaguide a carbonate continent with low relief was present. On this continent red continental deposits fill in erosional troughs [Martín Martín et al., 1998]. A distensive period is inferred from PTt data for the Mulhacen complex [Nieto Liñan, 1996] and the Ghomaride, which is a distensive margin until the L. Eocene [Maate, 1996]. S.Pyrenean and Ebro In the Pyrenean region, southward submarine emplacement of the Upper Thrust Sheets occurs [Muñoz et al., 1983] under high rates of south-directed shortening and widespread marine foreland deposition [Vergés et al., 1995]. In the S. Pyrenees/ Cantabrian Cordillera contraction deformation is active [Muñoz Jiménez & Casas Sainz, 1997], due to rapid convergence [De Jong, 1990]. In the western Basque area a deep marine Eocene siliciclastic flysch trough develops [Riba et al., 1983] in a WNW-ESE trending direction extending as far east as the Pamplona fault. This may be the eastward progression of downwarping of the N. Iberian Margin [Ratt, 1988] or the complex zone where the flip in vergence occurs (northward ‘subduction’ of Iberia in Pyrenees versus southward in Cantabrian). The Boixol and upper Pedraforca thrusts are fossilized prior to 55Ma, but the lower Pedraforca thrust starts its motion [Verges & Burbank, 1996] and in front of these thrust sheets a foreland basin starts to develop, most of which is presently overthrusted by the Pyrenees. Sedimentation patterns help to reconstruct the development of the foreland basin. In the S. Axial Zone of the Pyrenees Paleocene platform limestone are topped with Ypresian turbidites [Teixell, 1996] and in the present Eastern Ebro [Villena et al., 1996]. In the Eastern Pyrenees/Sierras Marginales L. Paleocene Garumniense (continental) deposition is followed by an E. Eocene transgression depositing Alveolinas limestone [Muñoz et al., 1983]. Both areas indicate progressive deepening or southward migration of the foreland basin, which can be related to rapid rising of the Pyrenees in the E. Eocene [Ziegler, 1988]. During the middle Lutetian the eastern foreland shifted from restricted marine sedimentation in the Ripoll basin to a broader and shallower basin [Verges & Burbank, 1996]. The present-day central Ebro basin (area of Zaragoza/Lerida) forms the bulge related to this foreland basin: Paleocene or Eocene sediments are hardly being deposited here [Riba et al., 1983]. Platform carbonates in the southern Ebro basin indicate that this area was the southern margin of the foreland basin. Finally, the southwestern border of the proto-Ebro basin (between the CCR and the Baleares) is active as a left lateral fault system [Ramos-Guerrero et al., 1989], leading to a sudden input of conglomerates into the southeastern Ebro basin [Verges & Burbank, 1996] and to collision between the Alboran/Betic and Corsica/Sardinia. N. Pyrenees and SW France The northern Pyrenean foreland is still relatively quiet with deposition of shelf platform carbonates [Vergés et al., 1995], although subsidence in the Aquitanian Basin, after slow subsidence during the Paleocene, is now increasing (water depth 0-500m) with a major phase in the eastern basin [Desegaulx o & Brunet, 1990]. ‘Pyrenean’ N020 compression that might be related to this increased subsidence is inferred from variations in sediment thickness in syndepositional folding [Rocher et al., 2000]. Just north of the developing Pyrenees, a syntectonic breccia is being deposited in front of the Montpellier thrust [Sérrane et al., 1995], related to a maximum of compressional deformation in the Gulf of Lions [Vially & Trémolières, 1996]. More external, the Marseille Basin/ Provence is an area of erosion or no deposition, sediments are absent until ~late Lutetian [Stampfli et al., 1998], for the Limagne/ Massif Central/Bresse emergence is inferred from erosion of the Mesozoic cover [Bois, 1993]. Massif Central: Alkaline volcanism [Bois, 1993]. In front of the Alps, the Briançonnais has subducted entirely and Valais oceanic crust enters the subduction zone around E-M. Eocene [Schmid et al., 1996]. In the present-day western Alps area, pre-Eocene west verging folding and thrusting occurs south of Pelvoux under NE directed horizontal extension and strike-slip against the edge of the Valais Zone [Coward & Dietrich, 1989]. The present-day Bresse Rift is uplifted and eroded in relation to this early Alpine compression [Bois, 1993], that is due to the incorporation of the Briançonnais in the accretionary wedge of the Alps [Ziegler et al., 1998]. In the Alpine foreland, flysch deposition starts [Coward & Dietrich, 1989]. Catalan-Sardinian margin Just as the southwestern border of the Ebro Basin, the CCR area is under NW-SE compression, but the Iberian Chain still has not been formed [Guimerà, 1984]. Deposition of Garumniense and conglomerates (red beds with high carbonate content, paleosols) point at erosion [Capdevilla et al., 1996]. The entire eastern margin is uplifted and eroded, [Fernàndez et al., 1995] inferred up to 5km of erosion during the Paleocene to M. Eocene in the southwestern Valencia Trough. Paleocene uplift and gentle deformation of Mallorca (no sedimentation) [Ramos-Guerrero et al., 1989], the Baleares in general (erosion [Fontboté et al., 1983]) or the Valencia Trough region is inferred from the fact that Eocene sub-crop map shows different ages [Ramos-Guerrero et al., 1989].

90


Chapter 4

The inferred start of the northwestward subduction of Tethys ocean floor under ‘Iberia’ (~55Ma [Zeck, 1996]) is contemporaneous with collision of the Calabrian/Corsica-Sardinia blocks [Ziegler, 1988], leading to stacking of ophiolitic nappes onto the continental margin. Ypresian sediments seal this tectonic activity [Carmignani et al., 1995]. Most probably related to this, southern Sardinia was emerged while the rest was in shallow marine (~200m?) environments, forming the southern extreme of the southern Pyrenean foreland basin [Carmignani et al., 1989]. Alpine Corsica: conglomerate deposition [Egal, 1992]. The south-dipping Alpine subduction system is linked with the north to northwest dipping Balearic-Corsican subduction zone via a left-lateral transform fault separating the Adriatic block from the western ensemble of blocks. N. Africa: The northern African Margin is in a relatively quiet setting; limited dextral motion of Iberia is documented by Brede et al. [1992]. An epicontinental sea covers large parts of the margin, in the Tellian part reeflimestone is being deposited [Aris et al., 1998] and in its Moroccan part shallow water carbonate deposition occurs [Uchupi, 1988]. The hinterland of this epicontinental sea, in the present-day foreland of Morocco and Tunisia, such as the Oran Meseta and a paleo relief north of M. Atlas [Herbig, 1988] and two separated blocks in N. Algeria [Wildi & Huggenberger, 1993], form emerged non-depositional or eroded areas. The development of this couple of epicontinental sea and emerging hinterland might be related to the onset of northwestward subduction of the oceanic crust of Africa under Iberia [Zeck, 1996]. o o More distal in the foreland minor activity of N040 thrusts (Middle Atlas) and N070 dextral strike-slip (High Atlas) is documented [Brede et al., 1992]. In the M. Atlas, the southern block of the North Middle Atlas Fault (NMAF) is continuously upthrown from ~Cenomanian to at least the Upper Eocene. This is maintaining minor relief that is being eroded [Morel et al., 1993], while on the northwestern, down thrown block siliciclastic sediments are being trapped [Herbig, 1988]. Even further south, in the central High Atlas, shallow marine to littoral/lagoon environments prevail on the southern block during the PaleoceneEocene, indicating both low relief and low erosion rates on the bordering Saharan Platform [Görler et al., 1988].

42 Ma, M. Eocene (L. Lutetian -E. Bartonian), Figure 4.7 General The Azores-Gibraltar fracture zone becomes active around chron 18 (42Ma), but relative motion between Africa and Iberia is limited until 36Ma [Roest & Srivastava, 1991]. Until the amalgamation of Iberia to Eurasia along the Pyrenean suture, Iberia moves as an independent plate from 42-24Ma [Roest & Srivastava, 1991]. The Kings Trough – Pyrenees boundary is a compressional active plate boundary from 44Ma until 25Ma [Srivastava et al., 1990]. In addition to the formation of the Pyrenees and its conjugate foreland basins, the Balearic orogen developed between Iberia and Adria/Magrheb [Butterlin et al., 1986] along the SE margin of the Iberian plate.

Detail Western margin The northern part of the western margin is affected by activity along the Bay of Biscay-Pyrenean subduction. Both offshore and onshore, uplift is documented. In the Galicia Interior Basin an important, sometimes erosional unconformity between M. and L. Eocene sediments is related to uplift of the Galicia Bank (estimated to be of the order of several kms) [Murillas et al., 1990]. Deformation decreases southward, moving away from the active boundary. Onshore, in NW Galicia, uplift is detected by the relative descent of the base levels of rivers [Pagés Valcarlos & Vidal Romaní, 1998]. Northern margin Along the northern margin, oceanic crust of the Bay of Biscay is subducting southward under Iberia, creating an accretionary wedge. Subduction is continuing into the Early Miocene; the M. Eocene is the principal episode of this process [Murillas et al., 1990]. This leads to NNW-SSE compression [Lepvrier & Martínez-García, 1990] in the Iberian northern margin, documented by a couple of NE verging folds and thrusts and inverse faults and thrusts verging SW developing in a shallow marine environment along the northern border of the Ebro Basin [Muñoz et al., 1983]. The Amorican margin experienced minor compression in relation to the Eocene activation of the subduction [Ziegler et al., 1995, 1998]. Central Iberia The stresses related to the ongoing subduction are not restricted to the northern margin, but are transmitted to the Iberian mainland as well. In the Duero basin NNE-SSW compression with a large

91

Chapter 4


perpendicular extension component favored fault-related lowering towards the northeast [Santisteban et al., 1996b]. Alluvial fans from nearly all edges enter the basin, the border facies pass into lacustrine environments in the central part of the basin [Jiménez Fuentes et al., 1983]. Alluvial fans enter the northern Duero from the Cantabrian Range [Alonso et al., 1996], in the northwestern Duero (Zamora) distal sediments of 2 alluvial systems with NW and SW provenance are deposited and in the western Duero (Salamanca) paleocurrents indicate southwestern and southern provenance [Jiménez Fuentes et al., 1983]. In the southwestern Ciudad Rodrigo Basin conglomerates and alluvial sandstones [Jiménez Fuentes, 1983] are deposited in fans with paleocurrents towards E/NE [Santisteban et al., 1996a]. Along the southern border non-deposition, alluvial facies (Penaranda-Alba) [Corrochano & Carballeira, 1983b] or a series of conglomerate with no relationship to the present-day Guadarrama [ITGME, 1991b] is observed. The high activity of the alluvial fans indicates uplift of many of the borders of the Duero Basin. Further foreland deformation related to the activity along the northern plate boundary of Iberia is found along the Sierra de La Demanda thrust front that becomes active, overthrusting the Rioja trough northwards. In the front of this thrusting, at the connection between the Duero and Ebro basins, the Montes de Oca develop as a 'high' [Pol & Carbeillera, 1983], separating the two basins. The Rioja area remains at around sea level and connected with marine deposits in north [Muñoz Jiménez & Casas Sainz, 1997], paleocurrents show a southern (Cameros) provenance of the clastic input [Jurado & Riba, 1996]. To the south of the active chain, in the Almazan Basin, the start of clastic sedimentation is the response to NW-SE compression and uplift of Iberian Range [Bond, 1996]. The amount of uplift is still limited as shown by the bordering eastern Duero Basin where lacustrine environments prevail and only limited clastic input from the Iberian Range and Sierra de La Demanda is observed [Pol & Carbeillera, 1983]. Both AFT data (1700 m uplift between 45-30Ma, cooling 2.5-5 degrees/Ma) [Sell et al., 1995] and sedimentary/structural data (mass flow conglomerates deposited in N. Madrid Basin [De Bruijne et al., 2001]) suggest general but limited uplift of at least the NE-SCS. In the Central System erosion or nondeposition supports this uplift. The trend of the SCS is not at all identical to the present-day geomorphology: alluvial fans and localized small basins in the Sierra de Guadarrama (Turegano/Segovia) cross over the present-day mountains [ITGME, 1991d]. The Loranca Basin shows a period of quietness just before activation SE part of the basin [Muñoz Martín, 1997] during which lagoon environments are deposited or non-sedimentation/erosion creates an unconformity between Lower Tertiary and Upper Eocene sediments [de Torres Perezhidalgo et al., 1983]. S.Pyrenean and Ebro A period of major convergence between Eurasia and Iberia, with rates up to 6mm/year [Vergés et al., 1995]. As a consequence, the Pyrenean belt is thrusting southward onto the Iberian foreland, the first important relief of the Pyrenees is being formed [Teixell, 1996] as documented by AFT (start of exhumation at ~50Ma) [Fitzgerald et al., 1999], an increasing thrust rate [Puigdefàbregas et al., 1991] and breakback thrusting, deforming the Pedraforca thrust sheet [Verges & Burbank, 1996]. The stacking pile of the Sierras Obarense (S Cantabria) is about to break through water surface [Muñoz Jiménez & Casas Sainz, 1997]. Around the Pamplona the south dpping Bay of Biscay subduction system stops and is replaced by the north dipping Pyrenean subdcution system [Engeser and Schwentke, 1986]. The foreland basin is marine, turbeditic and widening under the advancing load, motion along the Vallfagona thrust was initiated at around 43.5Ma [Verges & Burbank, 1996]. Together with the migration of the thrust sheets and foreland basin axis in southern direction, the related bulge invokes retraction of carbonate platform along the distal southern margin that is located in the central Ebro basin as shown by absence of deposition in this region [Villena et al., 1996]. In the western, marine Jaca basin only small alluvial fans enter from the south [Vincent & Elliott, 1996]. The sediments derived from the advancing imbricated western Pyrenean thrust belt [Millan Garrido, 1995] are shelf and slope marl and sandstone in the western External Sierras [Teixell, 1996] and major fans [Muñoz Jiménez & Casas Sainz, 1997] in front of the arising belt. Catalan-Sardinian margin NW-SE compression [Guimerà, 1984] in northeastern Iberia is thrusting the CCR over the southwestern border of the Ebro Basin, which evokes increased subsidence in the SE part of the basin [Vergés et al., 1998]. Under the same compression, the CCR hinterland, the later Valencia Trough, experiences NWverging thrusting that is creating local relief and deposition of red molasse type sediments [Martínez del Olmo, 1996]. Erosion of this uplifted hinterland creates huge conglomeratic wedges, e.g. the St Llorenç del Munt, active from ~50Ma and Montserrate, active from ~46Ma [HNPC, 1992]. The conglomerates enter a marine Pyrenean foreland basin and therefore are being deposited in a deltaic environment. Marine sandstone and conglomerate dominate the deltaic fronts but sometimes include fringing reefs and prodeltaic platforms [Capdevilla et al., 1996]. The elevation of the catchment area for the St Llorenç del Munt is estimated at around 700-1250m, Marzo, Oliana guide_7, which means that the highest summits in the area might have been well over 1500m. Mallorca forms the southwestern most part of this elevated region; a transgression towards the northwest is leading to the first Tertiary sedimentation in

92


Chapter 4

large parts of the Balearic domain. While northern Mallorca is still under continental conditions indicated by some fluvial intercalations, its southern domain is dominated by near shore platform sediments containing Nummulites [Ramos-Guerrero et al., 1989]. Sardinia is still connected to Iberia. The marine conglomerate of the southwestern Sardinian Cixerri formation contains clasts from Mesozoic levels that are restricted to the Provence and Iberia [Cherchi, 1979]. Moreover, northwestern Sardinia forms part of the southern (in this region continental) foreland of the Pyrenees [Stampfli et al., 1998]. Southeast of Sardinia and Corsica the Penninic accretionary prism develops, due to the northwestward subduction [Stampfli et al., 1998]. The thrust belt that develops in the Sardinia Channel area (until latest Oligocene) is just the prolongation of this system. Marine flysches are being shed from these accretionary zones towards the east/south (Alpine Corsica) [Egal, 1992]. N. Pyrenees and SW France The Pyrenean collision has its effect on the northern (European) foreland as well, the increased loading leads to major subsidence and deposition in the Aquitanian Basin, especially in central and western parts, water depth remains at ~0-200m [Desegaulx & Brunet, 1990]. Eastward, the Pyrenean belt was continuing, which can be observed in the Gulf of Lions [Vially & Trémolières, 1996], attested by overthickened and elevated crust [Sérrane et al., 1995] and in the Camarque Basin, where high topography (1000-1500m) [Sérrane et al., 1995] is estimated based on erosion of the Mesozoic and Paleogene succession. In the Gulf of Lions rocks similar to the Pyrenean axial & northern crystalline zones thrust over Permian and Mesozoic [Bois, 1993] and [Mauffret et al., 1995]. The northern foreland shows important compressional deformation as well, such as E-W trending folds/thrusts and broad synclines in the Camarque and Provence areas [Mauffret & Gorini, 1996], syntectonic continental sedimentation in the Alès Basin (in front of Montpellier thrust?) [Sérrane et al., 1995] and the Northern Provencal Cover Block where limited detritic continental deposits are documented [Villeger & Andrieux, 1987]. The compression is further recognized in the Languedoc [Ziegler, 1988], Nimes Basin (NS), [Villeger & Andrieux, 1987] and Ardeche area (~NE-SW) [Bonijoly et al., 1996]. Deposition of middle Eocene limestone in the Languedoc [Sérrane et al., 1995] and Marseille basin area is considered to be part of the northern foreland sequence [Stampfli et al., 1998]. In general, southeastern France and the complete southwestern European margin are under N-S compression [Roure & Coletta, 1996], although the first signs of rifting in southern France are observed [Sissingh, 2001]. S. Alps Pre- L. Eocene deformation (folding and thrusting) is predating the intrusion of Adamello. Local submarine fans deposit from Maastrichtian until L. Eocene marls in the Lombardian foreland [Bernoulli et al., 1989]. SE Iberia and Betic realm The effects of the Pyrenean collision and possibly activity of the Betic-Balearic orgeny can be observed in the southern part of Iberia as well in the Valencia-Alicante region. A Late Paleogene tectonic phase that can be related to early Pyrenean collision is inferred from uplift and erosion of the region creating an unconformity between Maastrichtian and Eocene sediments [De Ruig, 1991b]. The southern Iberian margin is deepening southward and probably thrust loaded by Internal Betic units [Vera, 2001]. The External Prebetics are being eroded, the Internal Prebetics are shallowing [Kenter et al., 1990] and the isolated outcrops in the Subbetics show marly facies with frequent turbidites, locally olistostromes [Fontboté & Vera, 1983]. Coastline SE Iberia: [HNPC, 1992] The Alboran units do not seem to show any evidence of compressional deformation related to the Pyrenean collision, which shows that for this epoch, the Alboran was not attached to/had not yet collided with mainland Iberia. The end of the emplacement of the Malaguide over the Alpujarride is followed by subsidence under vertical shortening of the Malaguide/Alpujarride pile [Balanyá et al., 1997]. This subsidence is documented by a general transgression depositing biogenetic sandy and marly limestone on the Malaguide [Fontboté & Vera, 1983]. The transgression cannot have affected all of domain, because the sediments contain clasts of the Mesozoic cover of the Malaguide, so parts must have been emerged and eroded [Fontboté & Vera, 1983]. Crustal thickening still occurs in the Maghrebide (Greater Kabylia) in relation to oblique collision [Saadallah & Caby, 1996]. In the Dorsalian and nearby domains tilting of blocks created considerable relief [Durand Delga & Olivier, 1988] and combined with the aforementioned transgression over the Internal Dorsalian and Ghomarides (Malaguide) lead to laterally strong varying depositional sequences. From shallow water limestone deposition [Durand Delga & Olivier, 1988] to polygene conglomerates in the Predorsalian [Fontboté & Vera, 1983]. In the External Dorsalian marl with pelagic foraminifers, turbiditic biocalcarenites and conglomerates of local or more internal origin were deposited [Durand Delga & Olivier, 1988] and in the Mauritanicas conglomerates, conglomeratic limestone and marls dominated [Fontboté & Vera, 1983]. To the south of these units, south of Lesser Kabylia, subduction-related LT-HP metamorphosis occurs [Fontboté & Vera, 1983]. N. Africa The northern part of the margin shows evidence for limited compressional deformation: folding of the Algerian [Aris et al., 1998] and Moroccan Prerif with related conglomerates deposited on the Riffian and

93

Chapter 4


Tellian margin [Wildi & Huggenberger, 1993]. Moreover, in the external western Tell a compressive phase is documented for the L. Lutetian [Wildi & Huggenberger, 1993]. Further south, in the High Atlas a last subsidence stage [Görler et al., 1988] occurs and the western and central parts are flooded by sea [Giese & Jacobshagen, 1992]. Northern Africa is being deformed and uplifted, this is the main period of inversion of the Atlas Trough related to the start of collision with the Alboran units [Ziegler, 1988], compression reached into the Shara platform: a main phase of inversion of the Saoura-Ougarta chain is observed [Ziegler, 1995], contemporaneous with the inversion of the Triassic-Jurassic Atlas troughs and emplacement of the internal Kabylian and ultra-Tellian nappes [Vially et al., 1994].

36 Ma, L. Eocene (Priabonian), Figure 4.8 General The Pyrenean collision reaches its peak and the Pyrenees are very rapidly uplifted. Deformation related to this collision starts to migrate westward and into the interior of Iberia and results in uplift of the northern border and isolation from the sea of the Ebro basin. Reactivation of basement faults occurs in the Iberian Range, Catalan Coastal Range and the Central System. The Rhine Graben starts to form and links up with the graben in SE France. A fully developed NNW-ward subduction system is active south of Corsica/Sardinia and the Betic/Alboran. Major inversion in the Atlas ranges and even the Saharan Platform and subsidence along the northern African margin are related to the onset of (or accelaration of) subduction along the Corsican/Sardinian/Balearic margin [Frizon de Lamotte et al., 2000].

Detail Western margin In the Sado Basin area, NNW-SSE compression inverts NW-SE trending faults, creating 3 alimentation areas. Marine environments prevail, but the basin is close to sea level [Pimentel & Azevêdo, 1994]. Northern Lusitania (north of Nazare) undeepens the marine basin to very reduced depth ( 7001250m) than in the central (500-800m) and southern (500-700m) part [Roca pers.]. The area od the future Valencia Trough is folded and emerged as well, 200-500m elevated above the L.Eocene sea level, which equals around 350-650 above present-day sea level [Morgan & Fernandez, 1992]. Sediments are being transported north and northwestward from this elevated region. SE Iberia and Betic realm The southern margin of Iberia is stable and deepening southward. The External Prebetics are emerged and eroded just as the Internal Prebetics are largely emerged until the early Oligocene [Fontboté & Vera, 1983]. The coastline of SE Iberia [HNPC, 1992] is located over the south of the Internal Prebetics: in the Intermediate Units and the Subbetics marl and marly limestone dominate with turbiditic intervals in the oriental sector [Fontboté & Vera, 1983]. In the Gulf of Cadiz, the first compressional features related to African-Iberian collision develop [Maldonado et al., 1999]. Alpujarride/Malaguide: emplaced as a complex over the Nevado-Filabride [Balanyá et al., 1997] at least before 22Ma: undeformed granites of that age and Oligocene-Miocene sediments unconformable over the Malaguides and Alpujarrides contact [Martínez-Martínez & Azañón, 1997]. These sediments are being deposited in a transgressive basin in the Internal Betics, to its eastern part deepening until hemipelagic environments[Fontboté & Vera, 1983]. Stacking of the Internal Betics crustal segments to climax from L. Eocene-M. Oligocene [Monié et al., 1994], an intermediate PT metamorphic peak is documented for the Mulhacen complex in the EoceneOligocene [Nieto Liñan, 1996]. This over-thickening resulted in extensional deformation of the Internal Betics [Durand Delga & Olivier, 1988]. Until the L. Eocene, the Ghomaride was a distensive, passive margin. From now until L. Burdigalian, compression dominates, culminating in the Aquitanian [Maate, 1996]. In the zones fringing the Alboran/Betics/Kabylia Block, sedimentation suggests activity of the block as well. In the Mauritanicas this consists of marine conglomerates, limestone and conglomeratic marls, in the western part of the Campo de Gibraltar fine detritic limestone, marls and limestone and in the Predorsalian red marl with sandstone beds are being deposited [Fontboté & Vera, 1983]. Block tilting in the Dorsalian considerable relief [Durand Delga & Olivier, 1988] and both the Internal Kabylian and the Ultra-Tellian nappes show deformation [Ziegler, 1988]. Catalan-Sardinian margin The St Llorenç del Munt conglomerate fan is basinward passing into more fully marine environments [Capdevilla et al., 1996]. On Mallorca smaller conglomeratic wedges prograde to the SE, showing tectonic activity (ext??). An ongoing transgression to the N-NW leads to passing of the conglomerate wedges into coastal clay-rich marls and sandy limestone [Fontboté et al., 1983]. The clastic input is

95

Chapter 4


associated with renewed relief in N. Mallorca [Ramos-Guerrero et al., 1989]. This northern highland was connected with the CCR and the European mainland, as derived from mammal occurrence [Pomar Goma, 1983]. On Corsica/Sardinia a possible start of transpressive tectonics that is active until ~ Aquitanian [Carmignani et al., 1995], related to the start of continental collision associated with thrusting of part of Corsican continental margin with obducted oceanic crust onto Eocene/Oligocene syntectonic deposits [Carmignani et al., 1995]. In Alpine Corsica active northwestward thrusting is documented with a northwards increasing displacement [Egal, 1992]. N. Pyrenees, SW France and N. Alps Generally, the French Alpine foreland is under near N-S compression [Bergerat, 1987]. In the Aquitanian Basin a major phase of subsidence is observed especially in the central and western parts [Desegaulx & Brunet, 1990], while synsedimentary folding occurs in the southern basin [Rocher et al., 2000]. This is related to the rapid build up of the Pyrenean chain [Vergés et al., 1995]. In the western part of the basin open marine environments prevail, towards the east an evaporitic event occurs, which is related to the Cardona salt in the Ebro foreland [Puigdefàbregas & Souquet, 1986]. The development of the foreland basin in the east has reached its maximum width: in front of the Northern Pyrenean Front the foreland sequence stops onlapping the Montagne Noir [Roure & Coletta, 1996]. In the northern foreland near the Pyrenean collision zone compression still dominates, as for example (NW-SE directed) in the Ardeche area [Bonijoly et al., 1996] and in the Languedoc [Ziegler, 1988]. Moving away from active belt the deformation style changes. The Durance and Cevennes faults were reactivated in a transpressional way during the Eocene [Roure & Coletta, 1996] but in the Les Matelles Basin left lateral pre-Oligocene strikeslip causes down throw of SE block which points at extensional strike-slip [Sérrane et al., 1995]. The Bresse Rift starts opening (Bergerat90), in the Alès Basin extension starts (earlier than in the rest of Languedoc) [Sérrane et al., 1995], as well as in both the Apt basin (105 directed) and the Manosque Basin (120 directed) [Hippolyte et al., 1993]. These are the preludes of the extensional regime that is invading southern France from the Rhine graben area, where E-W extension is active [Bois, 1993]. In the present-day Jura, extensional faulting is inferred [Guellec et al., 1990], no sediments are being deposited [Bois, 1993]. A continental marine incursion enters the Bresse Graben and even the Rhine Graben through a marine connection along the southern Jura [Sissingh, 2001]. In the western Alps, the foreland basin sedimentation shows a rapid deepening to flysches. The Helvetic margin is being incorporated in the deformation as shown by Helvetic erosion products in these flysches, dated at 37-34Ma [Stampfli et al., 1998]. Now the Valais oceanic crust is subducting to the south, at 35-33Ma slab detachment of the Alpine root is inferred. Closure was most likely not complete, since flexure in the Provence is not important, not forming a large foreland basin [Stampfli et al., 1998]. S. Alps - Adriatic domain To the south of the Alps a second foreland basin, the Piemonte Basin witnesses its very start of continental sedimentation [Bersezio et al., 1993]. Uplift of the south Alpine margin is inferred from the erosion of upper Eocene limestone [Bernoulli et al., 1989]. N. Africa The last marine sediments are being deposited in the Middle Atlas (shallow marine limestone) [Herbig, 1988]. In the Central Constantinois, NE Algeria folding of the Tellian domain occurs under E-W compression (090-120) [Aris et al., 1998]. Uplift of Central Morocco is evidenced by the fact that the eustatic E. Oligocene transgression did not inundate the region [Brede et al., 1992]. A suite of alkaline volcanism develops in the junction area of the Middle and High Atlas. Local conglomerate fans with High Atlas clasts [Brede et al., 1992] demonstrate the first uplift related to overthrusting of the High Atlas over its southern margin. Late Eocene compressional structures are unknown in the Tell-Rif external zones [Frizon de Lamotte et al., 2000]

30 Ma, M. Oligocene (Rupelian - Chattian), Figure 4.9 General Major E-W trending directed extension invades southeastern France. This process is breaking up the former easternmost Pyrenees and starts limited rotation of the Corsica/Sardinian block. Compressional deformation is still occurring in the Pyrenean range and its southern foreland basin towards the west. Crustal thickening and related metamorphism in the Betic/Alboran units reaches a maximum at around 25Ma. The inversion of the former Atlas Trough is culminating.

96


Chapter 4

Detail Western margin The present-day physiography of Kings Trough is formed around 32 Ma from intraplate volcanism o [Srivastava et al., 1990]. N040 directed strike-slip compression is documented [Lepvrier & Mougenot, 1984] in the Portuguese margin from the Lisboan area to the Algarve. Offshore S Portugal lows and highs start to develop and are interpreted as the first compressional features in the eastern Atlantic related to the African-Iberian collision [Torelli et al., 1997]. Under the same stress field the Lower Tajus Basin develops [Ribeiro et al., 1990] and is correlated to a post-Eocene but pre-Aquitanian unconformity on- and offshore [Lepvrier & Mougenot, 1984]. The large extensional component in the strike-slip zone offshore causes pure extension: the offshore part of central Portugal subsides by normal faulting in the outer shelf/slope area [Rasmussen et al., 1998]. The Sado Basin is emerged during large parts of the Oligocene but close to sea level. An erosional surface develops on its Eocene marine deposits [Pimentel & Azevêdo, 1994]. In northwestern Portugal a similar Paleogene erosional surface is creating planar topography [Cabral, 1989]. Tectonic activity in Galicia is demonstrated by normal faulting until M. Chattian in the As Pontes Basin [Huerta et al., 1996]. Northern margin Along the Asturian coast N-S compression is documented. A reorientation of the stress field from NW in Eocene to NE in the Oligocene is inferred [Lepvrier & Martínez-García, 1990]. Under this compression, ENE-WSW trending Hercynian thrusts are being reactivated in and around the Oviedo Basin and thrusting southward [Teixell, 1996] and conglomerate is being deposited in the Oviedo Basin [IGME, 1973]. Tectonic activity along the southern border of the Cantabrian Cordillera –Ubierna Fold Belt- is inferred from growth synclines that developed before the late Oligocene [Espina et al., 1996a]. Parts of the Cantabrian coast are under marine conditions [Lepvrier & Martínez-García, 1990]. Central Iberia Compression related to the Pyrenean collision is transmitted into the central part of Iberia. In the Duero Basin NNE-SSW compression and perpendicular extension generates a NE-SW trending horst and graben structure with major uplift of the borders and source areas. Oligocene arkoses are restricted to the S and E of the line Zamora/Salamanca, which implies uplift of the hills on the northwestern side of this line [Santisteban et al., 1996a]. Further evidence for reactivation of the surrounding relief from the western and southwestern Duero is the influx of coarser detritus (conglomerate and sandstone) and retraction of the basin edge. The source material of these deposits is Paleocene deposit of earlier stages [Jiménez Fuentes et al., 1983]. In the SSW-Duero (Penaranda-Alba) no sediments are known of Oligocene age, which suggests reactivated borders as well. In the southern Duero Basin tectonic activity of the Guadarrama Mountains can be deduced from a fining up sequence of breccia and conglomerate that is deposited discordantly over Paleogene sediments [ITGME, 1991c]. At least partially, the SCS was formed progressively [Portero Garcia & Olivé, 1983], the northeastern SCS N-S strike-slip compression [De Bruijne et al., 2001] is documented. The former Cameros Basin is thrusting northwards onto the Rioja Basin and southwestwards over the Almazan [Platt, 1990]. For the latter, this rejuvenation of deformation of the northeastern basin margin is related to N-S to NNE-SSW compression. Uplift of the border causes an increase in erosion rates, which results in building out of alluvial fans and fluvial sedimentation with a northern provenance, parallel to the axis of the basin [Bond, 1996]. To the eastern side of the Madrid Basin inversion of the Iberian Basin starts [Muñoz Martín, 1997], although very limited and not leading to deposition. Main activity of the border of the Iberian Range and the Loranca Basin is dated as middle to lower Oligocene. This resulted in an important angular unconformity between uppermost Eocene-lower Oligocene and upper Oligocene sediments Oligocene [de Torres Perezhidalgo et al., 1983]. The lower sequence of detritic sediments shows paleocurrents without any relation to the present-day bordering chains and remainders of Mesozoic relief influence sedimentation patterns [Diaz Molina & Lopez Martinez, 1979]. SE Iberia and Betic realm Stresses related to the Pyrenean collision and Tethys subduction benath the Balearic system cause deformation as far in the foreland as the Valencia area, where folding and SW thrusting under ~050 compression is observed [De Ruig, 1991]. In the External Prebetics tilting and folding is inferred. No sedimentation occurs until the E. Burdigalian and these sediments onlap different Eocene units [Kenter et al., 1990]. Erosion of the Prebetics is sourcing the marly and turbiditic facies of the Subbetics [Fontboté & Vera, 1983] that demonstrate a northeastern provenance [Geel, 2000]. The northern margin of the eastern Internal Betics starts to rotate while foreland basin sediments are being deposited [Allerton et al., 1993]. In the eastern Malaguide an E. Oligocene transgression deposits conglomerate, during the M. Oligocene fan deltas develop. Both show proximity of Sardinia [Geel, 1996]. The transgression is related to the onset of SE-NW extensional deformation and heating [De Jong, 1990]. Crustal thinning starts to invade the region south of the Balearic domain, entering the eastern Internal Betics. To the west, compression prevails and nappes are being emplaced, especially in the Ghomaride/Malaguide

97

Chapter 4


domain. Erosion of this internal basement follows upon its gradual uplift [Durand Delga & Olivier, 1988]. The maximum stacking of the Malaguide/Alpujarride complex onto the Nevado-Filabride domain [Balanyá et al., 1997] resulted in a peak in metamorphism [Monié et al., 1994]. The emergent landmasses establish a land connection between Africa and Iberia [Pomar Goma, 1983]. Detritus from essentially the emergent Ghomaride/Malaguide zone is deposited in the Dorsalian and Predorsalian [Durand Delga & Olivier, 1988], as up to 1200m Oligocene flysches in the Mauritanicas [Fontboté & Vera, 1983] and as deep water Numidian clastics in narrow basins parallel to the incipient Rif and Kabylian fold belts [Ziegler, 1988]. To the west in Campo de Gibraltar red clay with some limestone levels is sedimented [Fontboté & Vera, 1983]. S. Pyrenees and Ebro The linking zone between the Pyrenean and Cantabrian ranges is actively being deformed and uplifted. In the Basque Basin molasse sediments become more and more conglomeratic [Ratt, 1988]. The first occurrence of continental deposition in the synsedimentary folded Miranda- Urbasa syncline is related to major uplift of its southern border from 35-32 Ma [Muñoz Jiménez & Casas Sainz, 1997]. However, the southern Sierras de Obarense front has not yet reached its present-day position, terrigenous sediments in the Rioja still show provenance from the south (Cameros) only [Jurado & Riba, 1996]. The development of the External Sierras in the Western Pyrenees during the Rupelian, results in an influx of terrestrial sandstone and shale into the Ebro Basin [Teixell, 1996]. Rapid exhumation of the axial zone (Maladeta) occurs from 35-30Ma, as inferred by AFT-data [Fitzgerald et al., 1999] and maximum denudation rates of 240mm/kyr occur during the same period [Morris et al., 1998]. Under N-S compression, the Iberian Range is thrusting oblique onto the southern Ebro basin [Guimerà, 1984]. At the same time the CCR is thrusted onto the SE margin of the Ebro Basin. Catalan-Sardinian margin The Valencia Trough area forms a peneplane with very little relief at the time of an E.Miocene transgression [Martínez del Olmo, 1996], therefore erosion is likely during the Oligocene. Fracturing, subsidence and rift shoulder uplift indicate the very onset of rifting [Fontboté et al., 1983], developing a NE-SW trough bounded to the NE by relieves at the present-day CCR, Garraf and Malgrat highs. This continental basin forms the northwestern margin of an Oligocene marine basin [Roca & Deselgaulx, 1992]. The basin forms within a still compressive setting for the CCR [Roca & Deselgaulx, 1992], which is obliquely thrusting onto Ebro foreland, under NS-compression [Guimerà, 1984]. This suggests a piggyback type of basin between Baleares and CCR, verging to NNW [Roca & Deselgaulx, 1992]. Southeastward prograding conglomeratic wedges are building out along the Mallorca margin forming a regressive cycle. In the southern part of Mallorca, marine conditions are prevailing [Ramos-Guerrero et al., 1989], while in the north lacustrine sediments with important detritic influx are being deposited, which evidences unroofing at large scale of nearby elevated regions (active tectonics) [Fontboté et al., 1983]. Rift shoulder uplift could be the process causing the Mid-late Oligocene erosive denudation of a massif situated NW of Mallorca [Pomar Goma, 1983] and renewing of relief in the NE [Ramos-Guerrero et al., 1989]. The Baleares are in communication with Africa during mid.-upper Oligocene [Pomar Goma, 1983]. Ongoing active subduction of the African Plate under Sardinia is demonstrated by 29Ma calcalkaline volcanism in Sardinia [Hippolyte et al., 1993]. Alpine Corsica is folded under continuing shortening, producing N-S and NE-SW trending folds [Egal, 1992]. Pyrenees and SW France In the northern Pyrenean foreland, the last deformation features related to the Pyrenean compression are formed [Viallard, 1985]. In the Aquitanian Basin subsidence of the ~200m deep basin continues, but is of minor importance [Desegaulx & Brunet, 1990]. The loading of the European foreland by the Pyrenean belt has ended and a limited part of the foreland is incorporated in the deformation. In the Londres and Lodeve syntectonic basins, northward thrusting (E of Cevennes fault) or backthrusting of Montagne Noir (west of Cevennes fault) starts erosion of the flexural sequence in the northern Pyrenean foreland [Roure & Coletta, 1996]. South of the Marseille Basin basement erosion documents a paleohigh [Hippolyte et al., 1993]. An estimate for the elevation of this paleohigh comes from the Provence, where the pre-rift compressional relief is estimated to have been of the order of 1000-2000m, at least south of the Camarque basin [Sérrane et al., 1995]. The dominant deformation style in the region becomes extension, which is invading the area from the north, where in the Limagne and Bresse Rift shale with sand layers document active rifting [Bois, 1993]. With an overall extension direction in SE France of N110 [Roure & Coletta, 1996], the extension is migrating towards the Valencia Trough area. In the Gulf of Lions alluvial/lacustrine basins evolve under NW-SE extension [Bois, 1993], where the first marine sediments occur in the southeastern sector, pointing at a marine connection to Tethys [Roca et al., 1999]. Rifting begins in Ligurian Provencal region [Hippolyte et al., 1993] resulting in the first, very limited, formation of oceanic crust in the Provencal Basin [Lonergan & White, 1997], although full development will not start before the E. Miocene [Roca, 2001]. (CONTINUED ON PAGE 114)

98


Chapter 4

Structural data

Sedimentary environments Deep oceanic

Marine fan/ turbidite

Thrust

Fold axis

Open marine

Alluvial fan

Normal fault

Trajectory of Shmax

Shallow marine

Provenance

Continental/littoral Emerged land

Lacustrine sediments

Stress directions deduced from active structures

Conglomerates

Paleostress datapoint

Olistostromes

Alkaline volcanics

Calc-alkaline volcanics Figure 4.4 Explanation of colors, symbols and annotations used in Figures 4.5 – 5.8.

40N

Strike slip fault

20W

15W

10W

+ -

Uplift Subsidence

5W

0

35N

30N

25N

20N Figure 4.5 (65Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the Cretaceous-Tertiary boundary (65 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description. Based on reconstruction by the Ocean Stratigraphic Drilling Network (www.osdn.de). 99

Chapter 4


20W

45N

15W

40N

10W

5W

-

+

+ +

+

-

35N + - +

+

+

30N

25N Figure 4.6 (54Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at L. Paleocene - E. Eocene (Ypresian, 54 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

100

0


20W

45N

Chapter 4

15W

10W

5W

+

+

40N

+ ++

-

0

+

+ +

+

-

+

+ +

+

_

+

+

35N

+

_

_

30N

25N Figure 4.7 (42Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at M. Eocene (L. Lutetian -E. Bartonian, 42 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

101

Chapter 4


20W

15W

10W

5W

0

45N -

+

-

+

40N +

+ + +

+ +

+

35N

30N

+

+

+

Figure 4.8 (36Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at L. Eocene (Priabonian, 36 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

102


45N

20W

15W

Chapter 4

10W

5W

0 + +

+

40N

+

+

+

+

+

-

-

-

+

+

35N + +

-

+

30N

25N Figure 4.9 (30Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at M. Oligocene (Rupelian – Chattian, 30 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

103

Chapter 4


20W

45N

15W

10W

5W

0

+ +

40N

+ +

+

35N

+

+

30N

+ +

25N Figure 4.10 (27Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the L. Oligocene (Chattian, 27 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

104


45N

20W

15W

Chapter 4

10W

5W

0

+

40N

+ -

+ +

35N

30N

25N Figure 4.11 (24Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at L. Oligocene - E. Miocene (L. Chattian - E. Aquitanian, 24 Ma). See Figure 4.4 for explanation of symbols and colors and text for detailed description.

105

Chapter 4


20W

45N

15W

10W

5W

0

+ +

40N

+ +

35N -

-

30N +

25N Figure 5.1 (21Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the E. Miocene (L. Aquitanian - E. Burdigalian, 21 Ma). See Figure 4.4 for explanation and text for detailed description.

106


45N

20W

15W

10W

Chapter 4

5W

0

5E

40N

35N

30N

25N Figure 5.2 (18Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the E. Miocene (L. Burdigalian, 18 Ma). See Figure 4.4 for explanation and text for detailed description.

107

Chapter 4

45N

20W


15W

10W

5W

0

40N

35N

30N

25N Figure 5.3 (15Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the M. Miocene (L. Langhian - E. Serravallian, 15 Ma). See Figure 4.4 for explanation and text for detailed description.

108

5E


15W

10W

5W

Chapter 4

0

5E

45N

40N

35N

30N Figure 5.4 (12Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the M. Miocene (L. Serravallian, 12 Ma). See Figure 4.4 for explanation and text for detailed description.

109

Chapter 4


15W

50N

10W

5W

0

5E

+ -

45N

+ +

+ +

40N

+

+

-

-

+ +

35N +

30N

-

-

-

+

-

+

+ Figure 5.5 (9Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the L. Miocene (Tortonian, 9 Ma). See Figure 4.4 for explanation and text for detailed description.

110


15W

10W

5W

Chapter 4

0

5E

+

45N

+

40N

+

+

+ +

+

35N + - -

30N

+

+ Figure 5.6 (6Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the L. Miocene - E. Pliocene (Messinian – Zanclean, 6 Ma). See Figure 4.4 for explanation and text for detailed description.

111

Chapter 4


10W

5W

0

45N

5E

+

+

+ +

40N

+ + + + +

35N

+

+

30N

Figure 5.7 (3Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the L. Pliocene (3 Ma). See Figure 4.4 for explanation and text for detailed description.

112

10E


10W

5W

Chapter 4

0

10E

5E

45N

40N

6mm/yr

35N

-

4mm/yr

6mm/yr 5mm/yr

4mm/yr

30N

Figure 5.8 (0Ma) Paleo-tectono-geological reconstruction for the Iberian Peninsula and the western Mediterranean at the Holocene (0 Ma). See Figure 4.4 for explanation and text for detailed description.

113

Chapter 4


Several, presently onshore, extensional basins start their development with very limited sedimentation. In the Herault basin active rifting starts in small fault bounded basins, now buried under later synrift alluvial plain Aquitanian sediments [Sérrane et al., 1995]. During all of the Oligocene, in the Marseille Basin shallow fresh water deposits or non-deposition occurs under N125-105 extension [Hippolyte et al., 1993]. Similar extension direction (N120) is documented in the Nimes/Camarque Basin as well [Villeger & Andrieux, 1987]. Lacustrine sediments dominate in the E. Oligocene in the Manosque basin [Roure et al., 1992]. Under the increasing Alpine crustal wedge, the northern Alpine foreland basin subsides rapidly and expands northwards [Ziegler et al., 1996]. Large detritic fans start to enter the basin (Reuss and Sernf fans) [Sissingh, 1997]. Communication between the marine Alpine foreland basin and the Rhine and Bresse graben exists through two ‘inlets’, the Jura becomes an isolated island [Sissingh, 2001]. S. Alps - Adriatic domain The Bergell intrudes in the Southern Alps under E-W orogen parallel extension [Schmid et al., 1996], marking the end of the earliest stage of post collisional shortening related to slab detachment [Ziegler et al., 1996]. To the south the Gonfolite basin develops [Stampfli et al., 1998] and infill of the Piemonte Basin is turbiditic [Bersezio et al., 1993]. N. Africa In the Atlas ranges, first compressional deformation and related uplift is widespread. Relief is maintained in the Middle Atlas by continuous uplift of the southeastern block of the North Middle Atlas Fault from ~Cenomanian until ~Oligocene [Herbig, 1988]. For the South High Atlas first uplift [Herbig, 1988] related o to N-S to N020 compression [Fraissinet et al., 1988] is assumed.

27 Ma, L. Oligocene (Chattian), Figure 4.10 General The plate boundary running from Kings Trough to the Pyrenees remains active from 44Ma until 25Ma [Srivastava et al., 1990], deformation along this plate boundary is waning. WNW-ESE extension is dominant in the Provencal-Ligurian basin and starts to propagate into the Valencia trough. The onset of drifting apart of the Baleares, Sardinia/Corsica and Iberia. Rotation of Corsica/Sardinia results in collision with ‘Alpine Corsica’. Extension starts to invade the southern crustal segments of the Alboran block. Subduction of Tethys oceanic crust has ended in the west, after most of the oceanic crust has been subducted. The Alboran/Betics block is still not mechanically coupled to the southern margin of Iberia.

Detail Western margin The generation of a post-Eocene but pre-Aquitanian unconformity on- and offshore in the Arriba/Lisboa area is correlated to NW-SE directed compression [Lepvrier & Mougenot, 1984]. In northwestern Portugal the formation of a major Paleogene erosion surface results in deposition of clay-sandstone in planar topography [Cabral, 1989]. The Castelo Branco basin is the locus of deposition of conglomerate and sandstone [Dias & Cabral, 1989]. Northern margin Activity along the northern Iberian margin is renewed [Boillot & Malod, 1988] resulting in the development of horst and graben parallel to the margin [Boillot et al., 1979] under NE-SW compression, as documented in the Asturian Basin [Lepvrier & Martínez-García, 1990]. Along the Asturian coast marine environments are present in the San Vicente de la Barquera and Santander areas [Lepvrier & MartínezGarcía, 1990]. The As Pontes basin is being formed. Strike-slip activity along a curved fault plane leads to active thrusting and normal faulting from 28.7Ma [Huerta et al., 1996]. Central Iberia The first appearance of conglomerates with a northern (Pyrenean) source area in the Rioja [Jurado & Riba, 1996] show that the Cantabrian Cordillera is close to its present-day position. Activity and renewed uplift of both sides (Cameros and Sierras Obarense is dated at 27-26 Ma [Muñoz Jiménez & Casas Sainz, 1997]. For the rest, central Iberia is rather quiet. Deformation of the Iberian Range is inferred from the lateral restriction of the last phase of sedimentation (orange marls) and observed in the Maestrazgo/Serriana de Cuenca and Montes Universalès [Adrover et al., 1983]. The monotone alternation of canalized sandstone and lutite developing into gypsum [Diaz Molina & Lopez Martinez, 1979] deposited in the neighboring Loranca Basin, interpreted as the distal front of a wet alluvial fan [de

114


Chapter 4

Torres Perezhidalgo et al., 1983], does restrict any deformation of the Iberian Range to very limited. For the northern and northeastern SCS deformation possibly related to activity of the Iberian Range [De o o Bruijne et al., 2001] is inferred [Portero Garcia & Olivé, 1983]. N045 -055 directed compression results o o in sinistral reactivation of N060/090 trending and dextral reactivation of N140 /160 trending faults. These movements form two small basins (Corneja and Ambles) in the Sierra de Gredos [Babín Vich & Gómez Ortiz, 1997]. SE Iberia and Betic realm Erosion products of the External Prebetics arrive in the Internal Prebetics as detritic red continental series [Fontboté & Vera, 1983] and in the Subbetics where detritus has a NE provenance [Geel, 1996]. The coastline is situated over the Internal Prebetics [HNPC, 1992]; its eastern part is dominated by shallow marine limestone [Fontboté & Vera, 1983]. Within the Internal Zones of the Betics crustal thickening culminates resulting in a peak metamorphism before 25 Ma, coinciding with the onset of extensional thinning [Monié et al., 1994]. The Greater Kabylian Naciria massif shows a low temperature event (Ar-Ar), which suggests an extensional phase [Monié et al., 1988], before it starts overthrusting to the south at around 25 Ma. Extension of the Mulhacen complex from 28-23Ma is shown by PTt data [De Jong, 1991]. This extension, inferred from PTt modelling as well [Platt & Whitehouse, 1999], leads to breaking up of the Internal zones into several blocks, the western one of which becomes the Betic/Riffian or Alboran Block (bounded to the north by the North Betic fault, to the south by the Jebha fault) [Durand Delga & Olivier, 1988]. The extension within the Malaguide domain creates basin conditions that vary with respect to local tectonic activity. Series that are (still) submerged in parts of the domain are being eroded under continental environments in other parts [Fontboté & Vera, 1983]. This (L.Oligocene – E. Aquitanian) extensional rifting is only exposing the Malaguide realm because the developing grabens are filled with Malaguide detritus exclusively [Geel, 1996] (Alozaina formation). In the Eastern realm, similar sediments are deposited (Ciudad Granada). To the west of the Internal Zones, in the Internal Dorsal the transgressive series of "arenisca de Horca" (clay rich marl) indicates erosion of the Mesozoic cover and basement of the Malaguide and maybe even Alpujarride [Fontboté & Vera, 1983]. Towards the south and southwest, a deepening of the marine environment is suggested by the lateral change from Predorsalian microbreccia-limestone to up to 1200m Oligocene flysch in the Mauritanicas [Fontboté & Vera, 1983]. S. Pyrenees and Ebro The southern front of the Pyrenees is active: the External Sierras develop, leading to erosion (end of sedimentation cycle) of the Jaca Basin [Teixell, 1996]. The exhumation of the Maladeta slowed down significantly [Fitzgerald et al., 1999]. All along the border with the Ebro Basin series of progressive unconformities develop (last major tectonic?) [Muñoz et al., 1983] and the southern foreland starts to be filled in with conglomerates [Fitzgerald et al., 1999]. Towards the south in the Ebro Basin, these conglomeratic series pass into fluvial redbeds and lacustrine sediments [Teixell, 1996]. In the southern o o part of the Ebro Basin evidence for N020 -030 compression is widespread [Guimerà, 1984]. Although less pronounced, deformation related to this compression can be observed in the CCR as well, where fractures are being reactivated obliquely under sinistral shear of the CCR boundary faults [Alsaker et al., 1996]. N. Pyrenees and SW France Normal faulting in the Jura/Alpine foreland parallel to the Alpine belt and its foreland basin [Wildi & Huggenberger, 1993]. Infill of the Limagne and Bresse Rift with shale and sand layers [Bois, 1993] is related to active rifting. Active extension migrates southward through the Alpine foreland. Although Pyrenean compression is still observed in the Languedoc [Roure & Coletta, 1996] and folding in the Aquitanian basin [Viallard, 1985], contemporaneous ~N110 directed transtension is widespread in the same region [Roure & Coletta, 1996]. Closer to the Alpine front and in relation to this, more disperse directions of extension (155-015) occur in the Marseille Basin [Hippolyte et al., 1993]. In the Narbonne basin, the St Chinian thrust is reactivated extensionally [Roure et al., 1994] and active rifting in the Herault basin results in small fault bounded basins, now buried under later synrift alluvial plain Aquitanian sediments. Along strike to the north, the Alès Basin is still active as well [Sérrane et al., 1995] and just as well, the Valensole/Manosque basin (around sea level) documents extension [Roure et al., 1994]. S. Alps - Adriatic domain The southern Alpine domain is tectonically very active, leading to steep gradients of the Southern Alpine wedge. The Bergell experiences rapid uplift and erosion due to backthrusting of the Central Alps over the Southern Alps [Schmid et al., 1996]. Erosion of Bergell is witnessed by turbidites in the Gonfolite Basin with a clear Bergell Provenance [Bersezio et al., 1993]. The entire Lombardian fore deep experiences deeper water environments [Ziegler et al., 1996].

115

Chapter 4


Catalan-Sardinian margin Extension is entering the Valencia Trough area, but the Baleares are still linked with Eurasia (communication of mammals) [Pomar Goma, 1983]. The Sardinia Basin is being generated under N130140 extension [Vially & Trémolières, 1996] and volcanic activity starts [Monaghan, 2001]. For northeastern Alpine Corsica the last compressive tectonics is being documented [Egal, 1992]. Alpine Corsica is now attached completely to the Corsica/Sardinia. N. Africa The south central High Atlas is uplifting [Görler et al., 1988], but for the rest rather quiet [Frizon de Lamotte et al., 2000].

24 Ma, L. Oligocene - E. Miocene (L. Chattian - E. Aquitanian), Figure 4.11 General The active plate boundary is gradually relocated from the north (Kings Trough and the Pyrenees) to the south of Iberia (Azores-Gibraltar system) after 25Ma [Srivastava et al., 1990]. A major sedimentary break in the Iberian Neogene basins is observed [Calvo et al., 1993] and might well be related to the plate boundary reorganizations. Convergence rates between Eurasia (of which Iberia is a part now) and Africa change between 25-20 Ma from fast to slow [Lips, 1998]. Therefore, activity along the new plate boundary is limited and with a dextral wrench component [Ziegler et al., 1996]. Extension starts to invade the internal zones of the Betics/Alboran, breaking the Kabylian and Calabrian/Peloritan blocks from the Balearic/Betics/Alboran.

Detail Western margin Along the Portuguese margin, the Sado Basin is emerged, stable and eroded [Pimentel & Azevêdo, 1994], while in the Lower Tajus basin a transgressive series is being deposited [Azevêdo, 1991]. Deformation off shore Galicia [Murillas et al., 1990] and strike-slip activity onshore Galicia leading to the first sediments in the As Pontes Basin (L. Oligocene until ~E. Miocene) indicate N-S compression [Santanach Prat, 1994]. From geomorphological evidence, a new erosive period (due to compression?) [Pagés Valcarlos & Vidal Romaní, 1998] would provide theses sediments. Erosion of the western edge of the nearby Duero Basin supports this as well (see Central Iberia). Northern margin Last important compressional deformation of the Asturian margin took place until ~25Ma, resulting in an offshore horst-graben structure controlling the Cenozoic sedimentation [Boillot et al., 1979] and the main inversion of the Penas Trough (between Le Danois Bank and Asturian Massif) [Ziegler, 1988]. The top of this Danois bank is very close to sea level during the E. Miocene [Boillot et al., 1979]. The Asturian margin is the locus of transgressive sedimentation [Boillot et al., 1979], and along the Asturian/Cantabrian coast compressional deformation Nummulitic Eocene limestone documents N-S compression, own data. The As Pontes basin shows strike-slip and thrust activity [Huerta et al., 1996]. Central Iberia The sedimentation pattern in the Duero Basin suggests a major tectonic phase. In the W/SW Duero E. Miocene sediments are absent, which suggests a further retraction of the basin edge [Portero Garcia et al., 1983]. Along both the northern and eastern border conglomerate/coarse sandstone are being deposited discordantly over lacustrine carbonates [Mediavilla et al., 1996], in the E. Duero alluvial fans are infilling from the Sr. De la Demanda that is located to the east of the basin. More evidence for o tectonic activity comes from the nearby Almazan Basin, where N030 compression is demonstrated [Maestro et al., 1997]. The entire Duero basin shows a gentle sinking towards N and W accommodated by small slip along normal faults under EW-extension [Santisteban et al., 1996b]. This suggests limited foreland basin type development of the northern Duero Basin related to increased loading by southward thrusting of the Cantabrian Cordillera [Marín et al., 1995]. The relative uplift of the southwestern edge was leading to incision of the Duero with paleocurrents towards the west for the first time in this area [Santisteban et al., 1996a]. The present-day drainage patterns starts establishing [Santisteban Navarro, 1998], progressively changing a larger part of the Duero Basin into exoreic. Evidence for uplift of the southeastern border of the Duero Basin (the Spanish Central System) comes from 120m arcosic fans [Portero Garcia et al., 1983] and the region of Penaranda-Alba where E. Miocene coarse conglomerate

116


Chapter 4

and sandstone proceed from the SE [Corrochano & Carballeira, 1983b]. At least the northeastern SCS is o o tectonically active due to N050 -090 directed compression. This results in strike-slip deformation along the southern border fault [De Bruijne, 2001] and limited thrusting and related strike-slip deformation of o the Mesozoic/Paleogene cover [Sanchez Serrano, 1991]. In the central SCS (Sierra de Gredos) N050 directed compression is forming the Ambles and Corneja Basins [Babín Vich & Gómez Ortiz, 1997]. The Iberian Range shows major activity [Adrover et al., 1983] and is thrusting over the southern edge of the Ebro Basin [Guimerà, 1984]. A widespread erosion surface that is covered by L. Oligocene deposits is being deformed thoroughly in this tectonic active period. After that no significant deformation, erosion or uplift occurs (another, Early Miocene erosional surface covered by Lower Miocene sediments is only gently deformed) [González et al., 1998]. The Neogene basins within the IC are formed as small grabens with limited sedimentation. At the other side of the Iberian Range, progressive unconformities are formed at the border with the Loranca Basin [Muñoz Martín, 1993]. Large alluvial fans enter the Loranca Basin from the SSE (Tortola) and E. Partly related to this activity, the Sierra Altomira experiences its first uplift [Muñoz Martín & De Vicente, 1998]. Further south, the Guadiana, Jucar and southern Tajo basin are the locus of lacustrine limestone deposition [Adrover et al., 1983]. S. Pyrenees and Ebro In the western Pyrenees the External Sierras stop being developed [Teixell, 1996] just as in the southeastern Pyrenees thrusting terminates at around ~25Ma [Morris et al., 1998], followed by a southeast shift of major denudation. A relatively quiet period occurs: a Partial Annealing Zone is developed in the Maladeta [Fitzgerald et al., 1999]. Along the entire northern border of the Ebro Basin the last stage of the progressive unconformities takes place [Muñoz et al., 1983]. In the Jaca/Gauss basins the Huesca and Tremp/Gauss fans start entering the basin [Vincent & Elliott, 1996]. Catalan-Sardinian margin The northern Valencia Trough is starting to rift actively [Roca & Deselgaulx, 1992]. Calc-alkaline volcanism related to the rifting and the northwestward subduction is active in the area until Late Burdigalian [Bois, 1993]. A marine transgression is entering the Valencia realm [Roca & Deselgaulx, 1992]. In the Barcelona/Sant Feliu half grabens synrift sediments are deposited, restricted to deeper parts of the block faulted troughs [Roca et al., 1999]. Throughout the entire Valencia Trough a major unconformity separating 2 main tectono-sedimentary stages is observed [Martínez del Olmo, 1996]. In the Catalan Coastal Ranges extension starts [Gueguen et al., 1998], leading to a considerable decrease in relief. Estimates of elevation are of the order of ~300m, Roca pers. The eastern margin of Menorca is rifting into a NW-SE trending basin [Rehault et al., 1984] and leading to breaking of Kabylia from Menorca at ~23Ma. Syntectonic conglomerates are being deposited in this region [Pomar Goma, 1983]. Early volcanism occurs along the NW-SE oriented North Balearic Fracture Zone [Mauffret et al., 1995]. Related to this breaking apart, Mallorca/Menorca show a 20-degree clockwise rotation since timing of magnetization (prior to Oligo-Miocene) but before upper Miocene. Related to active thrusting on Mallorca [Freeman et al., 1989] unstable platform sediments, irregular topography, a non -linear coastline and mixed shelf sedimentation characterize Mallorca [RamosGuerrero et al., 1989]. Between northern Sardinia and the Alps, a complex pattern of deformation is observed: sinistral strikeslip between Corsica and N. Apennines [Dewey et al., 1989] along the major transfer zone linking the Alpine and Tethyan subduction fronts. Sinistral strike-slip along near E-W trending faults in eastern Sardinia/Corsica is related to transpressive tectonics [Carmignani et al., 1995]. The rifted basin that developed in central and south Sardinia under N130-140 directed extension [Vially & Trémolières, 1996], is now the locus of important post-rift sedimentation [Monaghan, 2001], continental environments prevail [Carmignani et al., 1989]. In the meanwhile, to the south of Sardinia, a fold-and-thrust belt is active in the Sardinia Channel because the Sardinian margin collides with the Maghrebian-Sicilian block [Catalano et al., 1995]. SE-verging imbricated thrust units shed arcosic turbidites to the SE [Catalano et al., 1989]. Most likely this belt is related to an accretion process in the ongoing NNW ward subduction. Related calc-alkine volcanics are observed in a volcanic arc running through Corsica/Sardinia [Bois, 1993], the Nice area and the Western Alps. SE Iberia The major rifting phase in the Valencia area [Roca & Deselgaulx, 1992] results with perpendicular ~170 extension in the development of NNW-SSE and ENE-WSW fault systems in the Prebetics [De Ruig, 1991]. Detachment between Iberia and the Internal Betics (Alboran) is still not active, as inferred from the fact that no compressional deformation that would be the effect of collision, is observed yet! Moreover detritus is arriving in the Subbetics with NE-provenance only [Geel, 1996]. Block faulting in the area leads to related complex sedimentation patterns. In the External Prebetics both continental redbed, conglomerate and sandstone deposition [De Ruig, 1991] and discordant shallow marine deposits [Fontboté & Vera, 1983]. In the eastern Internal Prebetics Aquitanian shallow marine shelf deposits onlap transgressive over red detritic continental Oligocene deposits, while in the western part, shelf – and slope deposition continued without a major sedimentary break [Geel et al., 1992].

117

Chapter 4


Betic realm Extensional systems are thinning the Malaguide/Alpujarride stack [Lonergan & White, 1997], breaking the Internal Betics up into several blocks [Durand Delga & Olivier, 1988]. Synchronous with the opening of the W. Mediterranean, WNW-directed extensional shear is thinning Greater Kabylia at ~25Ma [Saadallah & Caby, 1996]. The Malaguide starts a strong rotation (up to 200 degrees to L. Miocene) [Allerton et al., 1993] while the Alpujarride starts to be exhumed rapidly from below the Malaguide complex (until 19Ma) [Johnson et al., 1997]. In the meantime, a peak in compressional deformation is inferred for the Ghomaride [Maate, 1996]. Separation between Kabylia and Alboran (Rif/Betics) [Wildi, 1983] cannot yet have advanced far: erosion products of Kabylian basement are observed in mass flows in the Velez Rubio corridor [Geel & Roep, 1998]. Erosion of Kabylia can have been produced by emersion related to shoulder uplift in the extensional setting in the Algerian Basin [Wildi, 1983]. Evidence for the end of subduction under the Alboran comes from PTt data [De Jong, 1991] and coincides with the estimate for slab detachment under the Alboran shortly before 22Ma [Zeck, 1996]. A decrease in the convergence rate between Eurasia and Africa occurs [Lips, 1998]. This suggests that ‘subductable’ oceanic crust has been consumed and the Alboran continental block is about to start overthrusting the northwestern margin of Africa. To the south and southwest, flysch depocenters are filled in rapidly. In the Campo de Gibraltar a flysch trough develops that is deepening to the west and in the Predorsalian a sequence of marl and clay with Numidian sandstone (flysch-type) is deposited [Fontboté & Vera, 1983]. In the Dorsalian the same type of sedimentation dominates, forming the clay-rich marl of the 'arenisca de Horca' in the internal part and sandstone and clay in the external part [Fontboté & Vera, 1983] N. Pyrenees and SW France Even though a new start of contraction is documented [Rocher et al., 2000], the North Pyrenean Fault becomes inactive [Roure et al., 1989] and even close to the Pyrenees deformation is transtensional or pure extensive in a general ~110 direction, rotating more NW-SE towards the east. In southwestern o France, compression directions rotate clockwise to approximately N060 , perpendicular to the western Alpine front [Bergerat, 1987]. These facts show that the remnant compression in southwestern France is related to the Alpine collision and waning of the Pyrenean collision results in general extension in north eastern Iberia/Gulf of Lions. The graben of Rosselo develops along the Tet and Tec faults with a continental infill interrelated to small marine fluxes [HNPC, 1992], and in the Narbonne basin extensional reactivation of St Chinian former frontal thrust occurs [Gorini et al., 1991]. On the basinward part of the margin of the Gulf of Lions, synrift Oligocene-Aquitanian sediments are deposited in continental environments [Sérrane et al., 1995]. A thick sequence of synrift alluvial plain Aquitanian sediments is accumulating in the Herault basin [Sérrane et al., 1995]. More to the east, the former Pyrenean relief has been reduced; generally low relief is observed with sedimentary basins close to or at sea level. Infill of the Valréas basin even becomes marine due to thermal subsidence of the Oligocene rift basin [Roure & Coletta, 1996], the Camarque Basin is filled with synrift sediments close to sea level [Sérrane et al., 1995]. Marine environments (evaporite) are dominant in the Aix area during the latest Oligocene but in the Marseille Basin shallow fresh water deposits are accumulating during major subsidence of 1.4m/1000 year. This basin is a large graben perpendicular to the 155 extension in this region [Hippolyte et al., 1993]. And last, the Valensole/Manosque basins witnessed not much topography either, where around sea level and related to extension along Durance fault [Roure & Coletta, 1996], in the Manosque basin evaporites dominate [Roure et al., 1992]. The shale with sand layers deposited in the Limagne and Bresse Rift form the last sequence for the Limagne, the Bresse rift will experience late subsidence in Miocene-Pliocene [Bois, 1993]. The northern Alpine Foreland Basin is deepening eastward. Nondeposition in the Jura [Wildi & Huggenberger, 1993] changes to freshwater molasse in the western parts and even marine environments towards the eastern extreme [Andeweg & Cloetingh, 1998]. Major alluvial fans enter the basin from the rising Alps in the south [Sissingh, 1997]. S. Alps - Adriatic domain In the Bergell area (Western Alps) an end of the rapid uplift is observed [Schmid et al., 1996], which led to a regional break in sedimentation in the Gonfolite Basin [Bersezio et al., 1993]. A general deep marine unconformity is being formed, with the largest amount of erosion in the west [Bernoulli et al., 1989]. N. Africa In the North African realm, gentle compressional deformation is observed. The North African Flysch Trough shows south-directed contraction [Martínez-Martínez & Azañón, 1997]. The northern Moroccan Meseta is emerged and Numidian flysch from Algeria are arriving from the S [Wildi, 1983]. In the PrerifMesorif a syntectonic turbiditic sandstone (Zoumi) shows S-SE provenance [Morley, 1987], indicating uplift of the South Central Atlas. This uplift is related to subsidence in the southern foreland basin of the High Atlas, the Ouarzazate basin that is still in an initial marine stage [Görler et al., 1988].

118


0Ma 21Ma

Chapter 5

CHAPTER 5 - NEOGENE TECTONIC EVOLUTION OF THE IBERIAN PENINSULA AND WESTERN MEDITERRANEAN In this chapter the geological evolution of the study area from the Neogene up to present is presented in the same way as the Paleogene evolution has been described in the previous chapter, in maps for every 3Ma. The major tectonic events during the Neogene in the region were, after an initial period of tectonic quiescence, the opening of the Western Mediterranean basin and the collision of Africa and Eurasia (figures can be found in the color section, pages 99-113).

21 Ma, E. Miocene (L. Aquitanian - E. Burdigalian), Figure 5.1 General Iberia is in a tectonically quiet epoch and based on the available data, it is impossible to reconstruct the stress field. Tectonic activity is concentrated around the onset of seafloor spreading in the Provençal basin (lasting until about 16.5 Ma), and the onset of opening of the Algerian basin. The first limited compressional deformation of the Iberian foreland related to the collision with the Betics coincides with the first signs of major collision of Africa with the Alboran block. Contemporaneous with the outward thrusting of the Alboran block, its internal parts are under extension.

Detail Western margin Along the western margin the Sado Basin is stable and subjected to erosion, in the Lisboan area Miocene sedimentation starts [Pimentel & Azevêdo, 1994] by a transgressive series entering the Lower Tajus basin [Azevêdo, 1991]. Northern margin Although major deformation offshore Cantabria is ending during the Aquitanian, minor compressional deformation is affecting up to Lower Miocene sediments [Ziegler, 1988]. The Danois Bank is very shallow; the “inner Basin” is deeper [Boillot et al., 1979]. Deformation onshore Galicia is inferred from sedimentation in the As Pontes Basin, but this is the last episode of tectonics affecting the basin [Huerta et al., 1996]. Within the tectonic setting it seems very likely that the inferred N-S compression [Santanach Prat, 1994] has been active until the Late Oligocene. Alluvial gravels and limestone in the Bierzo Basin indicate a source area far away to W [Martín Serrano et al., 1996], which might be the same source area as for the sediments in the As Pontes basin. Central Iberia In the Duero Basin E-W extension is inferred from its westward sinking favored by N-S trending normal faults [Santisteban et al., 1996b]. But still in the western and southwestern basin no E. Miocene sediments are preserved [Portero Garcia et al., 1983], due to an increasing part of the basin that drains towards the W through Portugal [Santisteban et al., 1996a]. Along the SSW-Duero (Penaranda-Alba) E. Miocene conglomerate and sandstone, proceeding from the SW, suggest ongoing uplift of the southeastern basin border [Corrochano & Carballeira, 1983b]. The Rioja area is relatively quiet [Muñoz Jiménez & Casas Sainz, 1997]. A folding phase is occurring in the Loranca Basin and a westward prograding sequence of detritical sediments enters the Loranca basin from the Iberian Range [Diaz Molina & Lopez Martinez, 1979]. S. Pyrenees and Ebro Apart from the active uplift of the Aragonese Pyrenees (Jaca) [Arenas & Pardo, 1996], the only prominent feature is the inferred start of uplift of the entire Pyrenean region due to isostatic reequilibration [Roure et al., 1989]. SE Iberia and Betic realm Fold development related to N-S to NNW-SSE compression in the Eastern Prebetics [Beets & De Ruig, 1992] is the first sign of the collision of the Betics with the southern Iberian margin. In the same domain, Aquitanian- E. Burdigalian sediments consist of algal limestone, calcarenite and to the east o conglomerate [HNPC, 1992] under ~N060 compression [Montenat et al., 1996]. The Middle Burdigalian is discordant on top of the former serie, which is another indication of active tectonics [Fontboté & Vera,

119

Chapter 5

0Ma 21Ma


1983]. A transgression over the Internal Prebetics might be related to the onset of foreland basin development with typical first stage marine deposits: limestone, sandstone and marl, to the south less detritic, more limestone [Fontboté & Vera, 1983]. In the Subbetics (dispersed outcrops of) bioclastic shallow marine limestone and sandstone are observed, the eastern part of the Subbetics starts its clockwise rotation (up to 130 degrees presently) [Lonergan & White, 1997]. The collision of the Internal and External zones (Subbetics and Malaguide) is revealed by the sedimentary evolution in the Espejos Basin. First, the Aquitanian Solana formation, which contains submarine fans, is in close connection with the Malaguide (where major thrusting occurs) but still far removed from the Subbetics [Geel, 1996]. Whereas, in the E. Burdigalian Espejos basin (a deep basin on the suture) detritus from Subbetics, Malaguide AND Alpujarride shows the start of collision of Internal and External zones [Geel, 1996]. The Intermediate Units, the units in between the Internal and External zones, show a foreland basin setting with deposition of sandy limestone. Bathymetry is deepest in these Intermediate Units and in the southernmost part of Internal Prebetics [Fontboté & Vera, 1983]. The internal Alboran experiences a first rifting episode [Platt & Whitehouse, 1999], associated subsidence led to a westward transgression over emerged lands, depositing the first marine sediments [Comas et al., 1992]. This rifting is parallel to the regional axis of shortening (WSW) [Martínez-Martínez & Azañón, 1997], further proven by the intrusion of a basaltic dike swarm at 23-22 Ma [Priem et al., 1979] suggesting E-W trending rift (back arc spreading) [Torres-Roldan et al., 1986]. The subduction stopped shortly after this intrusion [Zeck et al., 1992]. The start of rapid cooling of the Alpujarride [Lonergan & Johnson, 1998] and the surrection of relieves in the Malaguide and related syntectonic deposits (Alazoina, Ciudad Granada formations) [Fontboté & Vera, 1983]. Compression in the Ghomaride is ceasing [Maate, 1996]. The Dorsalian is emerged and forms a relative narrow and probably discontinuous ridge, separating the new Mediterranean from a southern basin [Durand Delga & Olivier, 1988]. The "brechas de Nava", discordant polygene continental breccias deposited in the External Dorsal indicate an earlier phase of deformation [Fontboté & Vera, 1983] that led to the development of this ridge. During the E. Burdigalian marly pelite with silex is dominant in the Predorsalian [Fontboté & Vera, 1983]. Catalan-Sardinian margin While in the CCR synrift stage sedimentation in Valles Penedes [Roca et al., 1999], and in the Valencia Trough foreland limited lower Miocene continental sedimentation indicate an extensional setting [Martínez del Olmo, 1996], compressional deformation is active in the Baleares [Bakker, 1988]. On Mallorca northwestward thrusting results in unstable platform deposits [Ramos-Guerrero et al., 1989], littoral sediments and conglomeratic wedges on Mallorca. The Mallorca/Menorca block rotates 20 degrees clockwise since the timing of magnetization (prior to Oligo-Miocene) but before upper Miocene. This rotation might be related to the thrusting on Mallorca [Freeman et al., 1989] or the opening of the Liguro-Provencal Basin. Burdigalian marine deposits cover a paleorelief [Pomar Goma, 1983]. The rotation Corso-Sardinian Block occurs in a short time span between 20.7 and 18.6 Ma [Hippolyte et al., 1993] or 21 and 16.5 Ma [Roca, 2001]. Calc-alkaline volcanism related to the retreating northwestward subduction, enabling the rotation of Corsica/Sardinia, is observed in the Valencia Trough [De Ruig, 1991] and along the edges of the Sardinian extensional basin [Bois, 1993]. In between the subduction front and counterclockwise rotating Sardinia, the Sardinia/Maghrebian Chain develops to the southeast of Sardinia. Progressive foreland basin sediments are being deposited in front of an E-SEward advancing chain, from time to time overriding its own marine foreland sediments [Catalano et al., 1995]. On Corsica/Sardinia the last transpressive (??) tectonics (up to Aquitanian sediments in pull apart basins) are observed [Carmignani et al., 1995] but in Alpine Corsica mesoscopic normal faults lead to formation of clastic sedimentary basins [Egal, 1992]. A new active rifting phase is documented for the Sardinian rift basin [Monaghan, 2001]. N. Pyrenees and SW France In southern France, the Late Oligocene active rifting induced subsidence is diminishing, but late stage subsidence lowers many of the littoral basins to below sea level. The southernmost structures still show limited active extension: the graben of Rosselo is still developing along Tet and Tec faults, filled by continental and small marine fluxes [HNPC, 1992] and in the Narbonne basin extensional reactivation of St Chinian frontal thrust is still active [Roure et al., 1994]. For the rest of the basins, however, post-rift marine sediments are being deposited over the synrift sequences. In the Herault basin slight erosion is probable: Burdigalian marine sediments unconformable rest on top of the synrift deposits [Sérrane et al., 1995]. In the Valréas Basin marine infill is related to post-rift thermal subsidence as well [Roure & Coletta, 1996]. In the Nimes/Camarque Basin an erosion surface is formed between synrift Aquitanian and onlapped by post-rift Burdigalian marine sediments [Sérrane et al., 1995] and a little more distal: in the Valensole plateau and Manosque the last episode of sedimentation occurs (extension ~stopped) [Roure & Coletta, 1996]. In the German part of the northern Alpine foreland basin, continental molasse deposition or even erosion occurs [Andeweg & Cloetingh, 1998].

120


0Ma 21Ma

Chapter 5

S. Alps - Adriatic domain In the Gonfolite basin an Early-Middle Burdigalian erosional hiatus is observed [Bersezio et al., 1993], related to active thrusting [Ziegler et al., 1996]. N. Africa The Kabylian units are seperating from the Baleares and approaching the African margin and even obducted as nappes onto the Tellian Margin [Wildi, 1983], [Ziegler, 1988]. This emplacement results in the start of the development of foreland type basin, in which more than 1500m of Neogene deposits will accumulate. The entire Northern African Flysch Trough experiences south directed contraction [Martínez-Martínez & Azañón, 1997]. The Rif of Morocco starts to develop as a paleohigh, depositing the Acilah sandstone with southern provenance to north of it. Compressional tectonic activity is reported for the S. C. High Atlas as well. The Zoumi sandstone unit is deposited to the north [Morley, 1987] and southward overthrusting of the chain results in uplift with associated subsidence of the Ouarzazate (foreland) basin (becoming less open marine) [Görler et al., 1988]. Uplift of the Subrif chain is evidenced by AFT data [Azdimousa et al., 1998].

18 Ma, E. Miocene (L. Burdigalian), Figure 5.2 General Extension in the Ligurian and Algerian basins is coming to an end, but is still active along the basin borders in the Valencia Trough area and near the margins of Corsica and Sardinia. In the Alboran region contemporaneous internal extension and frontal compression are active. Along the western margin of Iberia inversion structures are common, which indicate plate boundary activity near Gorringe Bank. The North African foreland basin, formed by the emplacement of Kabylian units onto the northern margin, is incorporated rapidly in the evolving Tellian fold-and-thrust belt in the Langhian.

Detail Western margin The western margin of Iberia is under compression. Indirect evidence comes from a marine transgression that is ending in the Sado/Lusitanian Basin, which might be related to tectonic uplift [Pimentel & Azevêdo, 1994]. More direct evidence comes from the same Lusitanian Basin, where inversion of the basin under NW-SE compression [Lepvrier & Mougenot, 1984] eroded ~1000m Tertiary sediments [Wilson et al., 1989]. The northern border fault of the Lower Tajus Basin (Cercal fault) is activated under ~NS compression [Curtis, 1999]. The timing of this inversion is offshore well dated by relatively undisturbed Langhian sediments unconformable over a Burdigalian sequence with clear NScompression [Rasmussen et al., 1998]. Along the southern part of the margin, Arrabida is partly active causing significant erosion of the Mesozoic sequence [Ribeiro et al., 1990] related to the NW-SE compression [Lepvrier & Mougenot, 1984]. Offshore, Gorringe Ridge is popping up, overriding to the south the subsiding W Horse Shoe plain [Torelli et al., 1997]. In the entire Atlantic offshore SW Iberia, first Tertiary deformation starts during latest Oligocene-E. Miocene [Torelli et al., 1997]. Northern margin In Kings Trough extensional subsidence and rifting occurs between about 20 and 16 Ma [Srivastava et al., 1990]. During Miocene (not dated more precisely) the Inner Basin develops on the Asturian Margin, isolating and deepening the Danois Bank [Boillot et al., 1979]. Onshore, an end comes to the strike-slip activity in Galicia: in the As Pontes Basin sedimentation continues until ~E. Miocene related to N-S compression [Santanach Prat, 1994]. Central Iberia In the Duero Basin E-W extension and westward sinking continues, favored by N-S trending normal faults [Santisteban et al., 1996b]. The absence of E. Miocene sediments in the W/SW Duero suggests further retraction of basin edge [Portero Garcia et al., 1983]. In front of the Cameros thrust, a blind thrust is activated [Molinero Huguet & Colombo Piñol, 1996]. Subsidence in the Almazan basin progresses along ENE trending normal faults [Bond, 1996] and in the Loranca Basin alluvial fans enter from the SE and west, where the Sr. Altomira is actively thrusting westward [Muñoz Martín, 1997]. The first Miocene sediments are being deposited in the southern part of the Madrid Basin [Sanz Montero et al., 1992]. SE Iberia and Betic realm A general transgression affects the Prebetic related to foreland subsidence in front of the Betic thrust sheets, while other parts within the Subbetics/Prebetics emerge due to folding. This creates NE-SW trending highs and lows with pronounced relief. The basins are up to 100-200m deep [Kenter et al.,

121

Chapter 5

0Ma 21Ma


1990]. In the central part of the Prebetic of Alicante sudden subsidence occurs in a trough between two emerged areas, of which southern one is being uplifted, as attested by a southern provenance of detritus and angular unconformities [Geel & Roep, 1998]. The foreland basin is deepening from the External Prebetics, where Burdigalian-Langhian marl and calcarenite and to the east sandstone intervals are being deposited [HNPC, 1992] on the Internal Prebetics. Here marl with levels of turbiditic sandstone fill in a marine basin of up to ~1000m deep [Fontboté & Vera, 1983]. In the Intermediate Units sandy limestone deposition is the result of active deformation in the Subbetics [Fontboté & Vera, 1983], thrusted southward over the Espejos Basin (suture external-internal zones) [Geel, 1996]. Both the Subbetics and the Internal zones were located at a position offshore Alicante, 100km more eastward than their present-day location [Geel & Roep, 1998]. The western foreland basin (incipient Guadalquivir Basin) is witness of active tectonics in the approaching Internal Zones by the deposition of collapse deposits like the "Arcillas con Bloques" (clay with olistostromes) and (Campo de Gibraltar area) the "Areniscas de Aljibe" (1000m sandstone of the middle part of deep marine fans) [Fontboté & Vera, 1983]. The Internal zones in the western Alboran are actively thrusting westward [Durand Delga & Olivier, 1988], having incorporated the Dorsalian. The transgressive Las Millanas serie is sealing the frontal thrust of the Malaguide/Alpujarride complex over the Dorsal Units [Fontboté & Vera, 1983]. Active NNW-SSE (perpendicular to axis) extension [Martínez-Martínez & Azañón, 1997] in the Internal Betics is ceasing at 19-18Ma, but a last stage of cooling is inferred for 18-16Ma [Andriessen & Zeck, 1996]. Uplift is continuing, but the rapid stage comes to an end [Zeck et al., 1992]. Exhumation of the Internal Betics [Lonergan & White, 1997] and the western Alpujarride under uniform fast cooling at 1920Ma [Monié et al., 1994] are related to tectonic unroofing [Lonergan & Johnson, 1998]. Synchronous olistostrome/ marine conglomerate deposition occurs over the basement of the Western Betics [Comas et al., 1992]. Catalan-Sardinian margin The elevation of the CCR is reduced considerably by normal faulting; most of these normal faults become inactive in the Burdigalian although subsidence continues (post-rift) [Roca et al., 1999]. Marine conditions spread over the present-day coast; the first marine influence in the Valles Penedes is documented by transitional sedimentation [Batrina et al., 1992]. All along the northwestern margin of the Valencia Trough marine environments are spreading progressively over the trough margins [Roca et al., 1999]. Active opening of the Valencia Trough is waning, so the subsidence of the northwestern margin might be flexural, related to the contemporaneous northwestward thrusting at the Baleares [Gueguen et al., 1998]. This thrusting of the most frontal Betic related deformation stopped short of the axis of the Valencia Trough during the early Langhian [Geel & Roep, 1998]. The same is observed at Mallorca, where the last NW-wards thrusting and major folding of the Sierra de Tramuntana occurs in latest Burdigalian-Langhian [Fontboté et al., 1983]. Syntectonic turbidites with paleocurrents towards the NNNE into the Valencia Trough are shed from the emerging Baleares. Locally olistostromes occur as precursors of these turbidites, indicating very active tectonics [Ramos-Guerrero et al., 1989], which is finished at L. Burdigalian - E. Langhian [Bois, 1993]. In the Langhian, a maximum transgression submerges part of the Baleares; Ibiza is submerged [Pomar Goma, 1983]. The southern parts of Mallorca stay emerged due to the former thrusting [Fontboté et al., 1983] while in Menorca marine sedimentation is combined with syntectonic conglomerate fans [Pomar Goma, 1983], indicating tectonic activity as well [HNPC, 1992]. Throughout the entire Miocene, the Sardinia Rift is filled with shallow marine sediments, with reefs on paleohighs, indicating depths of several hundreds of meters [Carmignani et al., 1989]. At this stage active extension is ending and post-rift sediments accumulate in the basin [Monaghan, 2001]. On Corsica Miocene sedimentation starts, reaching full marine conditions in the Langhian [Sartori & Capozzi, 1998]. Southeast of Sardinia, in the Sardinia Channel emplacement of the Kabilo-Calabrian units along the Drepano Thrust onto the Sicilian-Maghrebian predate the end of the Langhian [Catalano et al., 1995], while the Calabrian and Peloritan blocks are being separated from Sardinia (start of opening of the Tyrrhenian) [Gueguen et al., 1998]. Deep-sea arcosic turbidites are being deposited in the Sardinia Channel, in front of the thrust stacks [Carmignani et al., 1989]. N. Pyrenees and SW France Active rifting has ceased completely in southern France, but a period of strong post-rift subsidence follows. The basinward part of the Gulf of Lions margin, during the late Burdigalian still a continental basin, is rapidly passing into marine series, implying strong post-rift subsidence [Sérrane et al., 1995]. In the Herault Basin post-rift marine sediments (unconformable over synrift) are onlapping the pre-rift basement [Sérrane et al., 1995], but the marine environment does not reach the Alès basin. A transgression affected the Nimes Basin as well during a tectonic pause [Villeger & Andrieux, 1987], the marine sequence was onlapping an erosional unconformity [Sérrane et al., 1995]. To the east, the Manosque Basin is being incorporated in the Valensole Basin [Roure & Coletta, 1996]. The Northern Alpine thrust nappes reach their present-day position, in the German foreland the Upper Marine Molasse is deposited, connection to the sea is located towards the east [Andeweg & Cloetingh, 1998].

122


0Ma 21Ma

Chapter 5

S. Alps - Adriatic domain In the Southern Alpine domain the foreland thrust wedge starts to be deformed [Schmid et al., 1996]. Massive sandstone and conglomerate (Como Formation) is deposited, related to renewed uplift of the N. Alps or development of the Southern Alps [Bernoulli et al., 1989]. N. Africa A marine transgression enters the Tellian margin, due to foreland basin development. In the Central Constantinois (NE-Algeria) the first documented internal deformation occurs under ‘Alpine’ NNW-SSE compression [Aris et al., 1998]. The emplaced Kabylian blocks are still submerged: olistostromes are being deposited onto them, indicating high tectonic activity [Wildi & Huggenberger, 1993]. In both the Algerian and Tunisian foreland calc-alkaline volcanic activity is increased [Dewey et al., 1989]. The uplift of the S.C. High Atlas is accompanied by the associated Ouarzazate basin subsidence [Görler et al., 1988] and uplift of the Subrif is documented by AFT data [Azdimousa et al., 1998].

15 Ma, M. Miocene (L. Langhian - E. Serravallian), Figure 5.3 General A decrease in convergence rate between Africa and Eurasia [Lips, 1998] results in a tectonically relative quiet period in the western Mediterranean. Extension in the developing Alboran Basin and radial thrusting outward of the Rif/Gibraltar/Betics are active contemporaneously. From 18-15Ma Iberia and Africa are connected, Kabylia is colliding with Africa [Frizon de Lamotte et al., 2001]. African mammals spread over the Peninsula [Calvo et al., 1993] and this connection did not occur through France and the Eastern Mediterranean [Geraads, 1998]. Extension in the Valencia Trough area and near Corsica/Sardinia is waning.

Detail Western margin The Sado basin is emerged, stable and a transgressive series enter in between the Lisboan and Sado areas [Pimentel & Azevêdo, 1994]. Central Iberia In the Duero basin, the E-W extension with related westward sinking favored by N-S trending normal faults [Santisteban et al., 1996b] is waning. In the central part of the basin the first dated Miocene deposits (Duenas facies) [Portero Garcia et al., 1983] occur as distal occasional alluvial canals coming from the NW. In the northern part small amounts of more proximal alluvial fan sequences are being deposited [Portero Garcia et al., 1983]. The short-lived Sierra de Altomira becomes inactive, attested by Langhian sediments onlapping the structure. In contrary, the SCS starts to pop-up [Muñoz Martín, 1997]. Deposition of the first sediments in the western Tajo, which have northern provenance [Junco Aguado, 1983], indicate the onset of this activity of the SCS. SE Iberia and Betic realm In the Eastern Internal Prebetic wrench related pull-apart basins develop with abnormal high sedimentation and subsidence rates and asymmetric infill [Geel, 1996], related to NNW-SSE compression [Montenat et al., 1996]. Deposition in the basins is dominated by marl with turbiditic sand passing into marl with marine fauna in the Intermediate Units [Fontboté & Vera, 1983]. Further southeast, in the Subbetics olistostromes are being shed of the advancing Internal Zones, especially in the western part of the ‘mobile belt’ of North Gibraltar Strait [Sanz de Galdeano & Vera, 1992]. At the border between the External and the Internal zones new basins form during the L. Burdigalian-Langhian, the onset of strike-slip deformation causes shoaling and uplift [Geel, 1996]. The Internal Betics display no important relief apart from several places near active faults [Sanz de Galdeano & Vera, 1992]. In the Sierra Alhamilla area during the Upper Langhian continental or shallow marine sandstone and conglomerate rich in Alpujarride detritus are deposited, while during the Early to Middle Serravallian marl to turbiditic open marine to pelagic environments prevail [Martínez-Martínez & Azañón, 1997]. The internal parts of the Alboran experience the first main episode of extension from 1715Ma [Comas et al., 1999], but the direction of this extension is far from uniform. Ranging from N-S in the eastern sector [García-Dueñas et al., 1992], NNW-SSE (perpendicular to axis) in the Alboran Domain [Martínez-Martínez & Azañón, 1997] to NW-SE extension in the western Betics and in the eastern Alboran (E of 3.2W). In the latter area extensive mud-diapirism and volcanics occur [Comas et al., 1992]. No sedimentation at this time is observed at ODP967: paleohigh until at least Serravallian [Comas et al., 1999].

123

Chapter 5

0Ma 21Ma


Catalan-Sardinian margin The marine influences spread further onto the margin of the CCR, related to thermal post-rift subsidence [Torné et al., ]. Major onshore marine influences are documented in the Valles Penedes, where major fan-deltas and coralgal platforms on highs develop. Towards the Serravallian the setting becomes more regressive [Batrina et al., 1992] resulting in an increasing restriction of carbonate sedimentation and the development of a shelf talus progradational system [Roca et al., 1999]. Along the northwestern margin of the Valencia Trough the same is observed from Late Langhian to present-day. Erosional breakdown of the CCR and progradation of the Castellon delta into the Valencia Trough starts in Serravallian [Ziegler, 2000]. Off shore in the Valencia Trough rifting of the W-NW margin is leading to volcanics, uplift of the Baleares and onlapping of the margin with the first open marine sediments (in combination with a sea level rise) [Martínez del Olmo, 1996]. Following this open marine stage, a platform-talus-basin system derived from a siliciclastic coastal belt develops [Martínez del Olmo, 1996]. On Mallorca, the youngest thrust movements occur in the Sr. Tramuntana [Ramos-Guerrero et al., 1989]. A sequence of turbidites with possibly the start of a regressive serie towards top has been observed on northern Mallorca [Ramos-Guerrero et al., 1989], contemporaneous, in the central part lacustrine and evaporitic sediments are being deposited in actively subsiding basins [Pomar Goma, 1983]. On Ibiza, the last Neogene sediments (marine) are being deposited [Pomar Goma, 1983]. Sea floor spreading has come to an end in the Provençal basin. Alkaline dikes intruded eastern Corsica [Carmignani et al., 1995], indicating extensional setting. To the north and east of Sardinia, areas of subsidence and sedimentation point at the onset of Tyrrhenian extension as well [Sartori & Capozzi, 1998]. To the southeast of the island, isolated calc-alkaline magmatic intrusions (until ~Tortonian age) occur in the Sardinia Channel, related to Nwards subduction of African lithosphere [Tricart, 1994]. N. Pyrenees and SW France Infill of the Rosselo Graben is shallow marine [HNPC, 1992]. The first compressional deformation after the Oligocene rifting in observed in SE France in both the Ardeche (~ENE-WSW compression [Bonijoly et al., 1996]) and the Nimes Basin. In the latter, marine sand shows indicators of tectonic activity under o N080 compression [Villeger & Andrieux, 1987]. This compression is related to the emplacement of the Western Alps accommodated by lithospheric strike-slip along the line Nice-Corsica-Tyrrhenian [Stampfli et al., 1998]. The Vercors experiences minor uplift [Butler, 1987]. In the German Alpine Foreland the Upper Freshwater molasse is deposited, with sediment sourcing from the east [Andeweg & Cloetingh, 1998]. S. Alps - Adriatic domain The foreland thrust wedge of the Southern Alps is building up [Schmid et al., 1996]. N. Africa In the Tellian foreland, gravitational nappes are being emplaced in the foreland basin [Vially et al., 1994], with contemporaneous alkaline volcanism in the Kabylian. This type of volcanism becomes active in the Middle Atlas as well from ~15-6Ma in a SW trending zone [Giese & Jacobshagen, 1992]. At the junction of the Middle and High Atlas, start of sedimentation in the Moulouya region is inferred [Morel et al., 1993]. The southern border of the High Atlas is deformed under N-S compression creating progressive unconformities [Fraissinet et al., 1988], gravity sliding and folding of the Southern Atlas marginal zone, closing the Ouarzazate basin from marine environments. From now on uplift of more than 1000m is shown by marine Miocene that is exposed at 1000m at present-day [Görler et al., 1988]. Uplift of the Subrif Chain is coming to an end at ~13.9Ma, as is inferred from AFT data [Azdimousa et al., 1998].

12 Ma, M. Miocene (L. Serravallian), Figure 5.4 General A period of major intraplate activity in the Iberian Peninsula. The southern margin of Iberia collides with the Internal Betics. Deformation related to this Alpine collision is not restricted to the nearby foreland, but major inversion is recognized throughout the entire Iberian Peninsula and even as distant as the western part of France, south England and in the Atlantic [Ziegler, 1988]. Collision is reactivated along the Northern African Margin as well, in particular in the Rif. Extension starts to develop in the Tyrrhenian region.

Detail Western margin Inversion is documented in the Algarve Basin, forming anticlines with a structural relief of over 1000m [Ribeiro et al., 1990] and southward thrusting of the Ponsul fault over the Castelo Branco basin, creating

124


0Ma 21Ma

Chapter 5

accommodation space for sedimentation [Dias & Cabral, 1989]. Paleocurrents in the developing Guadiana basin are towards the west (Lusitanian)[Moya Palomares, 1999]. The structural depression of the Eastern Horse Shoe Plain develops [Torelli et al., 1997]. Northern margin Limited information is available that the collision at the Betic side of the Peninsula even affected the northern margin. Offshore Cantabria, the Asturian Basin is being eroded [Riaza Molina, 1996] and limited deformation is affecting up to L. Miocene sediments [Ziegler, 1988]. The Cantabrian front is actively thrusting southward, tilting the earlier Tertiary strata along the northern edge of the Duero basin and depositing small amounts of proximal facies of alluvial fans [Portero Garcia et al., 1983]. Deposition of proximal (SW) to distal (NE) sediments in the Bierzo basin is followed by a fracturing phase [IGME, 1982c]. This fracturing is coinciding with the inferred reactivation of the western border of the Duero Basin and the individualization of the Bierzo basin [Corrochano & Carballeira, 1983a]. Central Iberia In the Duero Basin a maximum of lacustrine environments occurs, but alluvial fans are fringing the basin from the edges. Distal alluvial sediments and alluvial plains with provenance from S/SE, NW and NE are recognized [Portero Garcia et al., 1983]. The torrential alluvial deposits in the Penaranda-Alba (SSW Duero) attest activity of the southern basin border, the SCS [Corrochano & Carballeira, 1983b]. This is confirmed by active thrusting of the SCS over both its northern (Duero) and southern (Tajo) foreland basins [De Vicente et al., 1996b]. AFT data for the Guadarrama suggest uplift from 12-10 Ma at a cooling rate of ~6-7 degrees/Ma [Sell et al., ]. Basin subsidence in the southwestern Duero [Santisteban et al., 1996b] can be related to this thrusting stage. Subsidence in the Almazan Basin and activity of the Cameros thrust, shedding a coarsening up sequence of sandstone to conglomerate [Muñoz Jiménez & Casas Sainz, 1997] show that the compression related to the Betic collision is transmitted to these regions as well. This compression with a ~N140-155 directed Shmax remains dominant from now until recent day [De Vicente et al., 1996b]. SE Iberia and Betic realm The western part of the External Prebetics is submerged, while in the eastern part (Alcoi) marine sedimentation persists until Early Tortonian [HNPC, 1992]. Marl with marine fauna in the Intermediate Units [Fontboté & Vera, 1983] show a southward deepening in front of the approaching Betics. The northern Subbetics are incorporated in the active belt recorded by a phase of folding and fracturing [Fontboté & Vera, 1983]. In the eastern sector, the Subbetics are thrusted in N/NW ward direction over the Prebetics from the Middle Serravallian, while in the western Subbetics olistostromes are being deposited in the broad Proto-Guadalquivir (North Betic Strait), especially in western and central part [Sanz de Galdeano & Vera, 1992]. This different behavior between east and west is most likely related to difference in crustal strength between the eastern (weak) and western (strong) Iberian foreland. The strong western sector forms a broad foreland basin, while in the east the foreland basin is extremely narrow and incorporated in the advancing thrust fronts [Andeweg & Cloetingh, 2001]. During the Late Serravallian the Subbetic thrust slices break up due to dextral strike-slip faulting [Geel & Roep, 1998]. While the frontal parts of the Internal Betics are thrusting over the Iberian and African foreland, extension invaded the contractional areas. The Gibraltar thrust is being inverted, generating SSE directed low angle normal faults in the western part of the Betics, in the Rif these faults are NE directed [MartínezMartínez & Azañón, 1997]. Extension directions therefore, are varying from one place to the other. In the western Betics NW-SE dominates [Comas et al., 1992], in the eastern Betics a NE-SW direction is prominent [García-Dueñas et al., 1992] and in the Alboran domain the extension is WSW, parallel to the former orogen [Martínez-Martínez & Azañón, 1997]. But even within the Alboran domain, differences exist between west and east. In the first small deep faulted depressions bounded by paleohighs without sedimentation [Comas et al., 1993] occur while regional subsidence starts [Campillo et al., 1992], in the latter subsidence is localized in small E-NE/W-SW rift grabens [Comas et al., 1992]. This generalized extension is leading to the absence of important relief in the Internal Betics apart from near active faults [Sanz de Galdeano & Vera, 1992]. For example backthrusting in the Betic stack occurs, being thrusted over the External zones [Fontboté & Vera, 1983]. Rotation of the Sierra Espuña ceased [Allerton et al., 1993]. Deposition in the Neogene basins occurs in shallow marine environments [Martínez-Martínez & Azañón, 1997]. Through time, these basin have developed from open marine to pelagic during the Serravallian, to very shallow (end synrift?) towards the end of the Serravallian [Martínez-Martínez & Azañón, 1997]. Catalan-Sardinian margin The oceanic crust in the Algerian and Alboran Basin has been formed completely [Lonergan & White, 1997]. Development of progradational terrigenous shelf-slope complex with an outbuilding fan delta continuous in the CCR region [Roca et al., 1999], filling up progressively the former active rifts and onlapping paleohighs [Batrina et al., 1992]. Water depths in the Valencia are shallower than present-day [Roca & Deselgaulx, 1992]. In the Trough, accommodation space created by the waning active extension is filled by large amounts of sediments from the continent [Martínez del Olmo, 1996]. Uplift of the

125

Chapter 5

0Ma 21Ma


Baleares [Martínez del Olmo, 1996] can be related to active listric normal faults on Mallorca [Gelabert Ferrer, 1997]. Extension occurs to the east and southeast of Sardinia as well [Tricart, 1994]. S. Alps - Adriatic domain Growth and uplift of the foreland thrust wedge of the Southern Alps continues [Schmid et al., 1996]. o N050 directed compression is observed in the central and external belts of the Northern Apennines [Boccaletti & Sani, 1998]. To the back of the Northern Apennines, 3000-3500m Premessinian sediments in Carnaglia basin (E. of Sardinia) indicate an important contemporaneous tensile event [Catalano et al., 1989]. Western Sicily is a foreland of the Maghrebian/Sardinian belt during the E. Langhian-Tortonian [Catalano et al., 1989]. N. Pyrenees and SW France Limited folding in the Aquitanian basin [Dercourt et al., 1986] indicates that NW-SE compression [Rocher et al., 2000] is still active at Iberia-Eurasia boundary and can be related to the major collision of the Betics in the south of Iberia. More eastward, the first N-S directed compression in the region after the Oligocene rifting is documented by a N170 directed compressional event in the Nimes Basin [Villeger & Andrieux, 1987]. In areas closer to the Western Alpine front, the compressional direction is rotating towards perpendicular to the Alpine front, e.g. ENE-WSW compression in the Ardeche [Bonijoly et al., 1996]. This is resulting in inversion of the Valensole/Manosque basins due to activity of the Durance fault and molasse sedimentation in the Valensole area [Roure & Coletta, 1996]. In the German part of the Northern Alpine Foreland Basin the Upper Freshwater Molasse is filling the basin from the east and forms the last stage of molasse sedimentation [Andeweg & Cloetingh, 1998]. Major fans enter the basin in the Swiss part [Sissingh, 1997]. N. Africa Along the Tunisian shelf the Numidian Flysch nappes are being emplaced as gravitational nappes [Vially et al., 1994] or are thrusting to the south, onto the margin [Tricart, 1994]. In the Algerian Margin, the o same thrust sheets are emerged [Aris et al., 1998]. In the Moroccan foreland of the Middle Atlas, N040 o 060 oriented strike-slip faults are being reactivated due to emplacement of the Rif. Related alkaline volcanism is active [Brede et al., 1992] and migrated rapidly from the Valencia Trough area to northern o o Kabylia [Vérges & Sàbat, 1999]. N160 compression leads to N070 trending thrusts and sinistral o o motions along the N040 /060 oriented border faults of the Middle Atlas [Jacobshagen, 1992]. On the internal side of the Rif, extension is active and resulting in growth faults and bathyal claystone [Chalouan et al., 1997].

9 Ma, L. Miocene (Tortonian), Figure 5.5 General A change in direction of African-Eurasian plate kinematics is inferred and a NW direction for the convergence between both is documented for at least the central Mediterranean [Mazzoli & Helman, 1994]. Intraplate deformation in Iberia continues under similar tectonic conditions as in the Serravallian but in lesser magnitude. Changes in morphology of the seafloor in the North Atlantic are related to the Alpine orogeny [Tucholke & McCoy, 1986]. In the Betic/Rif tectonic uplift of massifs is common, limiting the connection between the Mediterranean and Atlantic progressively.

Detail Western margin The compressional deformation is waning, but still present in the SW corner of Iberia. In the Algarve, basin inversion and active southward thrusting of the Arrabida belt are observed [Ribeiro et al., 1990], related to renewed NW-SE compression [Lepvrier & Mougenot, 1984]. The northern borders of both the Lower Tajo Basin [Curtis, 1999] (Extremadura thrust) [Lepvrier & Mougenot, 1984] and the Castelo Branco Basin (Ponsul fault) [Dias & Cabral, 1989] are thrusting southward over the basins. Northern margin In the Cantabrian Range, the most recent thrusting movements of the Sierras de los Obarenes have been dated Tortonian [Jurado & Riba, 1996], although compressional deformation even extends into at least the Pliocene [Cortes Gracia & Casas Sainz, 1997]. Similarly, compressional deformation is recognized offshore affecting up to L. Miocene sediments [Ziegler, 1988]. Central Iberia The borders of the Duero Basin seem to be active still, or witness the importance of the Middle Miocene compressional event. Towards the east, in both the Almazan and the Rioja area, ongoing compressional

126


0Ma 21Ma

Chapter 5

deformation is documented. In the first, small alluvial fans continue to build out with paleocurrents from the NNE, only forming a basin margin until ~early Pliocene [Bond, 1996]. The borders of the Rioja Basin (Cameros and SW Pyrenean thrust fronts) reach their present-day position [Muñoz Jiménez & Casas Sainz, 1997]. Taking into account the active thrusting of the Sierras Obarenes and the occurrence of large proximal alluvial fans along the northern edge of the Duero Basin [Portero Garcia et al., 1983], it is most likely that the Cantabrian Range is actively deforming as well. Along the southern edge, large proximal alluvial fan sediments [Portero Garcia et al., 1983] and in the SSW (Penaranda-Alba) even torrential alluvial deposits [Corrochano & Carballeira, 1983b] indicate activity of the southern border (SCS) as well. This is confirmed by AFT-data of the SCS that show a renewed cooling episode from ~10Ma at 6-7C/ma [Sell et al., ]. For the central parts of the Duero basin a lacustrine salt lake basin is restricted to the Eastern sector and the Almazan basin [Corrochano & Armenteros, 1989]. Large parts of the Duero basin becomes exoreic, related to a sinking towards the west by near radial extension [Santisteban et al., 1996b] and capture by the head ward eroding Duero river. The Iberian Range reaches its topographic maximum (1800m) [Guimerà & González, 1998] and within the Chain, the Teruel Basins start to development along N-S normal faults [Alvarado, 1978]. The Cabriel Basin (to the south of the Iberian Range) is filled with up to 400m alluvial and lacustrine sediments, and related to a large listric basement fault [Anadon & Moissenet, 1996]. The Southern Tajo Basin is filled with Tortonian sediments, evolving from conglomerate into alluvial fan deposits [Adrover et al., 1983]. SE Iberia and Betic realm From about 11Ma, the Eastern Prebetics (Alicante) are emerged [Geel, 1996]. A major unconformity documents an orogenic phase that created a paleorelief that mirrors present-day topography. Around arising belts, coarse conglomerates accumulate in basins up to 200m deep [Kenter et al., 1990]. The compression direction in this region changes to N-S towards the Messinian [Montenat et al., 1996; Jonk & Biermann, 2001], leading to a new suite of strike-slip related basins in the Subbetics (e.g. Lorca) [Geel, 1996]. The North Betic Strait connection from the Atlantic with the Mediterranean is being interrupted, the last olistostromes are being deposited in its western sector at around ~E. Tortonian [Sanz de Galdeano & Vera, 1992]. The Internal Betics show an irregular topography with rising anticlinal mountain ranges, e.g. the Sierra Alhamilla [Weijermars et al., 1985], surrounded by complex basins capturing the erosional products. In the Eastern Internal Betics during the early Tortonian continental conglomerate basins occur [Martínez-Martínez & Azañón, 1997], in the area of the future Sierra Alhamilla a basal layer of red conglomerates (continental and marine) rich in Nevado-Filabride detritus is deposited. This layer forms the first post-rift sediments and is covered by a transgressive serie with turbidites [Martínez-Martínez & Azañón, 1997]. Parts of the Nevado-Filabride must have experienced spectacular uplift [Sanz de Galdeano & Vera, 1992] and the first Nevado-Filabride clasts are recognized in sediments. AFT data for the Sierra Nevada show a first phase of cooling related to tectonic denudation [Johnson, 1997], south of o the Sierra Nevada N070 trending strike-slip faults are very active [Sanz de Galdeano et al., 1985]. In the eastern Betics, E-W trending fault systems are activated under a NW-SE oriented maximum principal axis of stress [Huibregtse et al., 1998]. In the southern Betics, the coastline extended north of the present-day coastline, as witnessed by interconnected basins [Soria et al., 1999] in which shallow marine deposits accumulate [Sanz de Galdeano & Rodriguez Fernandez, 1996]. Tortonian marine sediments are the first to be deposited in the Granada/Guadix Basin [Fernandez et al., 1996]. Marine influences are found in transgressive deposits in the entire western Alboran. Normal faulting is still active along the eastern and western borders of the western Alboran Basin [Campillo et al., 1992] but active rifting is ceasing [Comas et al., 1999]. At least the eastern Alboran basin shows an interruption of rifting [Comas et al., 1992]: the base of the Tortonian sediments forms a generally unfaulted or only poorly faulted horizon and is onlapping previous structured basement highs [Comas et al., 1992]. Tectonics in the western Alboran basin is still active. In the southwestern Alboran active extension along normal faults ceased during Tortonian, somewhat later than in the northern and eastern Alboran, leading to deposition of bathyal claystone and coarse turbiditic sandstone [Chalouan et al., 1997]. On the west side of the Gibraltar arc, the Horse Shoe structural low is being formed in the middle-late Miocene and filled with giant endo-olistostromes during late Miocene from the structural highs surrounding the area [Torelli et al., 1997]. S. Pyrenees and Ebro The Cerdanya Basin and Conflent Basin develop along E-W normal faults related to dextral transtensional activity of the NE-ENE trending Tet Fault [HNPC, 1992]. The infill of the Emporda Basin with littoral sediments [HNPC, 1992] suggests tectonic activity as well. Renewed exhumation of the Pyrenees starts [Fitzgerald et al., 1999], related to this, excavation of the Ebro Basin leads to increased sediment supply in the Valencia Trough (Castellon delta). N. Pyrenees and SW France Along the Alpine Chain, the last molasse sedimentation occurs in the Valensole plateau [Roure & Coletta, 1996], and generally, in the Northern Alpine Foreland Basin, where 8 Ma dates the end of foreland sedimentation [Schmid et al., 1996]. In the German foreland basin (until Tortonian, freshwater

127

Chapter 5

0Ma 21Ma


molasse), uplift after the Upper Marine molasse amounts as much as 700m [Andeweg & Cloetingh, 1998]. In the front of these molasse basins, compressional deformation of the foreland continues inversion of the Manosque basin [Roure & Coletta, 1996] and development of the Jura Mountains [Bois, 1993]. Folding of the Jura commenced at ~11Ma and continues at least to the Pliocene [Ziegler et al., 1996] and is synchronous with active thrusting in the Vercors (10-6Ma) [Butler, 1987]. Uplift occurs in the foreland of the Molasse basin and Massif Central and major volcanic activity in the latter [Bois, 1993]. The Alps become evidently a climatic divide [Jelen et al., 1997]. Catalan-Sardinian margin Offshore CCR minor compressional inversion of normal faults has been observed, that could be related to the Betic compression [Roca, pers.]. Even although limited extension is active, rapid progradation of the Ebro Delta occurs in the Valencia Trough that is the locus of high sediment supply. This high sediment influx is related to erosion of the Ebro Basin. The first deep-water facies are being deposited on the Spanish margin [Martínez del Olmo, 1996]. Mallorca/Menorca experience an additional 025 (+-15) clockwise, propably local rotation since the upper Miocene [Freeman et al., 1989] and are affected from the Serravallian by an extensive phase, creating listric normal faults [Gelabert Ferrer, 1997]. S. Alps - Adriatic domain Extension is active in the area east of Sardinia. The initiation of extension in the N. Apennines is dated at 8-12Ma [Mazzoli & Helman, 1994] and at the Baronie Ridge (east off shore Sardinia), the start of extension at ODP 654 causes sub-aerial deposition [Sartori & Staff, 1989]. In the N. Apennines in the internal belt sedimentation is ending before a large unconformity develops, in the central and external belt, a L. Miocene unconformity is present [Boccaletti & Sani, 1998]. The Sardinia Basin that has been a depth of some hundreds meters throughout the entire Miocene, now is shoaling to evaporitic and lagoon environments close to sea level [Carmignani et al., 1989]. This indicates uplift, as is documented for Corsica and Sardinia [Bois, 1993]. Exposure of the N. Apennines and the Corsica/Sardinian block suggests rift shoulder uplift on both sides of the incipient rifting of the Tyrrhenian. In the Sardinia Channel en-echelon NW-SE dextral strike-slip accommodates this limited southeastward opening [Catalano et al., 1989]. N. Africa The northern margin of Africa is very active during the latest Miocene. In the east (Atlas Range of Central Tunisia), the first major contraction corresponding to a major post-nappe shortening event of the Tell Mountains [Tricart, 1994], does not yet cause compressional uplift in the southern Tunisian foreland. In Algeria the final emplacement of the Flysch nappes occurs under N-S compression [Aris et al., 1998]. The pattern of maximum horizontal compression is trending NNW-SSE in Algeria towards N-S in the Oriental Rif [Ait Brahim & Chotin, 1989] and a general NE-SW trend in the Moroccan foreland [GalindoZaldívar et al., 1993] is more or less perpendicular to the Riffian thrust front. The Rif is thrusting over the Mesorif foreland basin, and the accretionary wedge in the Prerif experiences rapid extensional collapse and deepening. Deep-water sediments are being deposited over shallow water sediments [Flinch, 1996] and a transgression inundates the Guercif foreland Basin [Zizi, 1996]. In the Subrif marl and marine conglomerate sedimentation [Ait Brahim & Chotin, 1989] is happening in a WNW-ESE extensional setting [Ait Brahim & Chotin, 1989]. The Moroccan foreland is being deformed. The emplacement of the o o Rif is reactivating the N040 -060 trending border faults of the Middle Atlas. Related to the sinistral o movement along northern border faults [Herbig, 1988], alkaline volcanism is active and N070 trending thrusts develop. At the junction of high and Middle Atlas, sedimentation continues in the Moulouya Basin [Morel et al., 1993]. Even more distal the south central High Atlas experiences a slow to vanishing uplift, lakes and swamps develop in a closed basin between the High and Anti Atlas [Görler et al., 1988].

6 Ma, L. Miocene - E. Pliocene (Messinian - Zanclean), Figure 5.6 General Northward thrusting of the Betics blocked the North-Betic connection with the Atlantic Ocean, while in the Rif the southern connection was closed as well. The Mediterranean Basin was closed completely from marine waters and started evaporating, leaving an accumulation of salt at the basin floor (the so-called Messinian Salinity Crisis). The severe base level lowering caused strong incision of rivers in the margins. Within the course of only a few million years, strike-slip deformation in the Gibraltar Arc reopened the connection with sea, flooding the basin in short time.

128


0Ma 21Ma

Chapter 5

Detail Western margin The marine environments that entered the Sado Basin towards the end of the Miocene return to terrigenous in the Pliocene [Pimentel & Azevêdo, 1994], in the Castelo Branco basin tectonic stability and continued erosion is documented [Dias & Cabral, 1989]. Central Iberia In the Duero Basin, a last system of lacustrine sedimentation (Paramo 2) occurs along a central axis (Valladolid/Palencia/Burgos). In the connection between the Ebro and Duero basins no sedimentation is occurring [Pineda Velasco, 1996]. In large parts of central Iberia ‘Raña’ is being deposited throughout the lower areas. In the northeastern Madrid Basin NNE and NNW basement discontinuities are being reactivated extensionally and control the distribution of L. Miocene alluvial systems. In the basins in the Iberian Range, the Teruel basin infill finishes and the Jiloca basin starts to develop [Guimerà, 1997]. The Cabriel Basin develops in relation to extensional subsidence enhanced by Triassic floored basement, and is filled in with alluvial deposits, predominantly coming from the NE [Anadon & Moissenet, 1996]. The western Valencia area is developed in the same sequence of deformation. SE Iberia and Betic realm Thrusting of the Prebetic onto the southeastern Iberian foreland blocks the North Betic Strait [Sanz de Galdeano & Rodriguez Fernandez, 1996]. The coastline in the Guadalquivir Basin is shifted westward and is relocated near Cordoba [Sierro et al., 1996]. In the Gulf of Cadiz active basin subsidence is ceasing [Sanz de Galdeano & Vera, 1992]. In the Gibraltar area, the Rif-Betic arc blocks the strait [Weijermars et al., 1985]. Within the eastern Betics several anticlinal ranges develop rapidly in relation to o o N020 -N040 trending strike-slip zones, possibly related with the rotation of the stress field to a N-S direction [Jonk & Biermann, 2001]. Rapid uplift of the Sierra Nevada is being revealed by AFT data related to a second phase of cooling by folding and accompanied by differential erosive denudation [Johnson, 1997]. Thermal activity and deep fractures in the area [Sanz de Galdeano et al., 1985] imply that caution should be taken when interpreting these data. However, independent data show similar features: Tortonian marine sediments in the Sierra de Carrascoy (eastern Betics) are presently elevated to ~1000m [Sanz de Galdeano et al., 1998], and in the Sierra Nevada these can be found at elevations of over 1850m [Sanz de Galdeano & López-Garrido, 1999]. Southeast of the Sierra Nevada, the Sierra Alhamilla rises up as well. These ranges can be explained by large scale folding in restraining bends of o o N020 -040 trending sinistral strike-slip faults [Andeweg & Cloetingh, 2001]. But, not only the mountain ranges are being uplifted, the whole Betics experience an important stage of uplift. Most of the numerous basins in the area change from basins with reefs to lakes in the L. Tortonian and progressively shallow further during the Messinian [Sanz de Galdeano & Vera, 1992]. Except for the basins close to the present-day coast, this uplift caused withdrawal of the sea and disconnection from marine Mediterranean or Atlantic waters [Soria et al., 1999]. Bisection and uplift of Guadix/Granada basins after L. Tortonian is related to the indentation of the Sierra Nevada [Andeweg & Cloetingh, 2001]. The general pattern of Messinian to present-day progressive shallowness in Betic basins is only interrupted by an E. Pliocene transgression. A distensive phase occurring in the Eastern Betics is associated with calc-alkaline volcanics [Goy & Zazo, 1986b]. In the eastern and central part of the Alboran basin, Messinian deposits record compressional tectonics, particularly along margins and structural highs. Active N-S contraction from the L. Tortonian causes folding, strike-slip faulting and inversion of previous faults [Comas et al., 1992]. Towards the end of the Messinian transcurrent movements occur along ENE faults in E-W to ESE-WNW extension [Campillo et al., 1992]. Catalan-Sardinian margin Paleo water depth in the northern Valencia Trough is approximately equal to today, in the southern part slightly shallower (~800m maximum)[Roca & Deselgaulx, 1992]. The Ebro delta is continuing to build out after the Messinian lowstand [Roca, 2001]. Volcanism is active in the Olot area, the Columbretes and southeastern Iberia. Corsica and Sardinia are uplifted [Bois, 1993], the Sardinia Basin finally shoals to continental environments from the Middle Pliocene [Yilmaz et al., 1996] leading to the first continental deposits since L.Oligocene. Regional extension invades the Sardinia Channel and the Tunisian shelf experiences (rift shoulder?) uplift [Tricart, 1994]. N. Pyrenees and SW France Folding of the Jura continues at least to Pliocene [Ziegler et al., 1996], the Jura allochton overrides the Bresse Rift partly [Roure & Coletta, 1996]. The West-Alpine external massifs are rapidly rising due to strong mechanical coupling of the orogenic wedge and its foreland. Related to this, the Bresse Rift experiences several hundreds of meters of late subsidence as the flexural depression in front of the advancing Jura. Uplift of the Massif Central [Bois, 1993] is in relation to significant volcanic activity. With the Nimes Basin NNE-NE (020) compression continues [Villeger & Andrieux, 1987]. S. Alps - Adriatic domain

129

Chapter 5

0Ma 21Ma


The Southern Alpine thrust wedge is inactive demonstrated by a sealing Messinian unconformity o [Schmid et al., 1996], but compressional deformation is still active in the N Apennines. ~N020 directed compression is active in both the internal and external belts [Boccaletti & Sani, 1998]. Contemporaneous extension along the Sardinian margin shifted southeastward [Spadini, 1996], leading to a water depth of about 1000m in the N. Carnaglia basin (E. Mess.) [Sartori et al., 1989]. N. Africa During the latest Miocene, a minor compressional event deforms the Tunisian shelf, corresponding to a second post-nappe shortening in the Tell mountains [Tricart, 1994]. In the Central Constantinois (NE Algeria) extensional Pliocene troughs develop, Mesozoic faults are being reactivated under NW-SE extension, forming fault bounded continental basins [Aris et al., 1998]. Extension is present in the o Oriental Rif and Rif foreland as well, directed ~N030 [Groupe, 1977] approximately perpendicular to the regional compression. Inversion of Subrif basins due this NNE-SSW compression [Ait Brahim & Chotin, 1989], results in accentuation of border relieves and closes the connection between the Atlantic and Mediterranean in N. Africa [Ait Brahim & Chotin, 1989]. Just as in the eastern Betics Guadix/Granada basin, renewed uplift in the S.C. High Atlas related to strike-slip activity results in separation of the Ouarzazate (east) and the Ait Kardoula region (west) [Görler et al., 1988].

3 Ma, L. Pliocene, Figure 5.7 General During the E. Pliocene, vast plains of carbonate cemented conglomerate caps (raña) developed in central Iberia. A L. Pliocene general uplift of Iberia results in uplifted beach sediments along many of the Iberian margins and the start of erosion of the vast raña deposits. The western Alboran basin is being deformed.

Detail Western margin Deformation off shore Galicia [Murillas et al., 1990] during the Pliocene is inferred. Sedimentation of fan conglomerate in the Castelo Branco [Dias & Cabral, 1989] is another clue for tectonic activity. In the southwestern corner of Iberia, compressional deformation is active. In the Sado basin paleocurrents towards NW [Pimentel & Azevêdo, 1994] suggest renewed uplift of the hinterland of the basin [Pimentel & Brum da Silveira, 1991]. The River Guadiana, that demonstrated paleocurrents towards the Sado basin during the Late Miocene and now drains for the first time towards the south Moya-Palomares, proves uplift of this region [Roca, pers.]. A Pliocene abrasion plateau in SW Iberia is now elevated at ~500-700m. Along the southwestern Iberian margin NW-SE compression and perpendicular NE-SW extension is observed. During the Quaternary local extension forms NW-SE trending normal faults in the region [Flores Hurtado, 1994]. S. Pyrenees and Ebro Apart from the margins and Central Iberia, a generalized uplift affects the entire Pyrenean mountain range [Muñoz et al., 1983]. In Navarra (S. border of Sierra de Cantabria) compressional structures extend to at least the Pliocene [Cortes Gracia & Casas Sainz, 1997] and in the Ebro Basin generalized fracturing and normal faulting are resulting from N-S to NNE-SSW compression with perpendicular extension [Arlegui Crespo, 1996]. Catalan-Sardinian margin In many of the littoral basins along the eastern margin of Iberia evidence for N-S compression and E-W extension has been documented [Santanach et al., 1980]. In general, the present-day emerged zones of the CCR are under non-marine conditions by now [Batrina et al., 1992]. The small basins more to the north, like the Emporda Basin still are under marine influences [HNPC, 1992]. Littoral sediments are being deposited in central Mallorca as well [HNPC, 1992]. Continued elevated sediment supply into the Valencia Trough is evident by rapid progradation [Martínez del Olmo, 1996]. Extension in Tyrrhenian shifted further southeast: the oceanic crust of the Vasilov Basin developed before 3.5 Ma [Spadini, 1996]. Central Iberia Within the Iberian Range, alluvial sediments fill the Jiloca basin [Guimerà, 1997]. The extensive Pliocene continental deposits high in mountain ranges (up to 1100m in the SCS) are being eroded during the Quaternary. A change of sedimentation to erosion occurs in the Duero, Ebro and Tajo Basins. Raña deposits (lime-cemented caps) are limited by N-S normal faults, indicating NS-compression [Martin Escorza, 1977]. An important NE-SW fluvial network developed in the eastern Madrid Basin [IGME, 1976], which indicates an active uplift of tilt of this part of the Madrid Basin.

130


0Ma 21Ma

Chapter 5

SE Iberia and Betic realm o o Ongoing compressive strike-slip activity of the N020 -040 trending faults in the eastern Betics [Andeweg and Cloetingh, 2001], e.g. the Serrata fault [Boorsma, 1993] and the Alhama de Murcia fault [Martínez Díaz & Hernandez Enrile, 1992]. The latter is causing continued uplift of the anticlinal sierras as the Sierra de Carrascoy [Sanz de Galdeano et al., 1998]. In general, an uplift of 125-700m is inferred for the Eastern Betics and the Prebetics [Janssen et al., 1993]. After uplift of the Central Betics (radial?) extension reorganized many older basins and maybe opened the Strait of Gibraltar [Sanz de Galdeano & Vera, 1992]. Small pull-apart basins in the western Alboran however, point to strike-slip along the Strait of Gibraltar as the governing mechanism of opening [Campillo et al., 1992]. But in the E. Pliocene in the western Alboran Basin dip-slip to oblique High Angle Normal Faults are active as well [Comas et al., 1992]. In the central to eastern Alboran basin, compressional deformation is abundant. The anticlinal structure of the Alboran Ridge is formed [Mauffret et al., 1987], along the SE bordered by a NW dipping reverse fault that lines up with the Jebha fault located to the SE [Meghraoui et al., 1996]. Along both edges of the Alboran Ridge positive flower structures develop [Campillo et al., 1992], pointing to an important strike-slip component in the deformation. Between the Alboran Ridge and mainland Algeria, o the Al Hoceima pull-apart basin is created (a N070 trending synclinorium with north dipping axial planes) [Chalouan et al., 1997] by dextral motion along two E-W faults that form the prolongation of the Yussuf Ridge [Meghraoui et al., 1996]. N. Pyrenees and SW France A climax of alkaline volcanic activity in the Massif Central is associated with (after L. Pliocene) regional basement uplift [Bois, 1993]. The Rosello Graben in Southern France is still in marine environments [HNPC, 1992]. N. Africa In the Tunisian Atlas E. Pleistocene synsedimentary reverse faults indicate that the deformation front has reached the northern boundary of the stable Saharan Platform [Tricart, 1994]. In the northeastern Algerian foreland a NW-SE extensional phase [Aris et al., 1998] is documented. In the Tellian belt, near N-S compression is active contemporaneously, especially east of Algiers. This combination of facts suggests a reactivation of the Saharan flexure [Vially et al., 1994]. Moving westward, in both the Oriental Rif and (Guercif) foreland (NNW-SSE) and the Western Rif (ENE-WSW) compression dominates [Groupe, 1977]. The Rharb and the Cheliff flexural basins in front of the Rif wedge are strongly subsiding and filled with up to 3000m marine Plio-Quaternary sediments [Meghraoui et al., 1996]. Active emplacement of the Rif is related to sinistral strike-slip movement along the southern border fault of the Middle Atlas [Herbig, 1988]. At the junction of the High and Middle Atlas, NW compression is deforming all sediments older than Pliocene in the Moulouya region [Morel et al., 1993]. The entire High Atlas region is being uplifted rapidly, the basins as well. A second phase of folding indicates that this uplift is most likely related to compressional deformation [Görler et al., 1988].

0 Ma, Holocene, Figure 5.8 General Intraplate deformation is still very active and accommodates the internal deformation of the Iberian Peninsula, which is squeezed between approaching Africa (convergence rates between Africa and Iberia based on NUVEL-1 [Argus et al., 1989]), Eurasia and the opening Atlantic. A generalized uplift affects Iberia and NW Africa. The stress trajectories in Iberia are according to SIGMA [1998].

Detail Western margin Tectonic activity is abundant along the Portuguese coast. Uplift and normal faulting (N-S and E-W faults), possibly related to NNW-SSE trending strike-slip fault is observed [Granja, 1999]. In the Sado Basin NNE/SSW faults are being reactivated under NW-SE compression, offsetting and tilting raña’s [Pimentel & Azevêdo, 1994]. Uplift of the Grandola basement block (west of Sado) separates the Sado basin from littoral environments and causes renewed subsidence [Pimentel & Brum da Silveira, 1991]. In the Castelo Branco: renewed SE-thrusting Ponsul fault is observed [Dias & Cabral, 1989]. GPS data show uplift of the northern border and subsidence of the Lower Tajus Basin. In the Algarve uplift is even dated since 1755 [Hindson et al., 1999]. NE-SW compression with perpendicular extension [Flores Hurtado, 1994] resulted along the Huelva coast in E-W trending normal faults and NNW-SSE and NW-SE faults that conditioned the deposition Holocene sediments [Zazo et al., 1999]. Off shore southwestern Iberia

131

Chapter 5

0Ma 21Ma


dextral slip occurs along an ENE-WSW directed basement fault along S. Gorringe Bank and Faro to the Guadalquivir Basin [Maestro et al., 1998]. Deformation of quaternary sediments in the area is observed [Tortella et al., 1997]. Northern margin Uplift is evident in Galicia as well: marine quaternary sediments are now up to 55-60m above sea level [Vidal Romaní, 1989] and the rivers in the Cantabrian Range are erosional, no sedimentation [ITGME, 1990c]. Helium isotopic ratios reveal important seismic activity in Galicia [Pérez et al., 1996], NS-trending normal faults cause significant seismicity [SIGMA, 1998] and recent movement of faults in the Bierzo basin affect Plio-quaternary sediments [IGME, 1982a]. Central Iberia In the Tajo Basin the last ‘basin’ sediments are of about 2-2,5Ma, sedimentation afterwards only occurs in river terraces. A large amount of terraces exists along the rivers, even up to 20 along the Henares. A tectonic control for their development is evident [Capote & De Vicente, 1989]: 1) the terraces are all located at one side of the river, while it is eroding the other side [Pérez González et al., 1989], and 2) paleoseismites have been dated at ~300.000 years in the area of River Jarama This fits with the observation that the area was struck in the Middle to Lower Pleistocene by increased tectonic activity [Giner Robles, 1996]. SE Iberia and Betic realm In the eastern Betics, left lateral activity along NNE-SSW faults (Carboneras, Palomares, Alhama de Murcia [Martínez Díaz & Hernandez Enrile, 1992]) indicates the activity of compressional tectonics [Bousquet, 1979]. Vertical motions observed by leveling demonstrate tectonic activity of several fault zones in the eastern Betics [Giménez et al., 2000]. In the Eastern Prebetic (Alicante) E-W folds are observed in Lower Pleistocene deposits. Even younger normal faulting along N-S faults is documented. Marine quaternary sediments are now at 45m [Goy & Zazo, 1986a]. In the west-central Betics continental subduction of Iberia under Alboran could have accommodated up to 150 km of convergence from the Middle-Late Miocene [Morales et al., 1999]. Uplift of the Alboran Ridge [Comas et al., 1999] is suggested by a basinward shift of the depot centers at a higher rate than the lowering of the sea level could produce [Campos et al., 1992]. In the central Alboran structural inversions in the Alboran Ridge and formerly formed pull-apart basins in the Strait of Gibraltar occurs with a dextral component along NNW trending border faults in the W. Alboran Basin [Campillo et al., 1992]. The Yussuf Ridge in the central Alboran shows the development of a tilted block bounded to the south by a normal fault with up to ~400m throw [Mauffret et al., 1987]. This deformation pattern in the Alboran and Betics can be correlated to activity in the Atlas and be related to a large sinistral shear zone [Andeweg & Cloetingh, 2001]. S. Pyrenees and Ebro Southern margin Pyrenees: recent activity of anticline Barbasto-Balaguer resulting in seismicity and deformed quaternary sediments, indicates N-S compression, still activity of southernmost Pyrenean feature. Ebro Basin: general fracturing predominantly N-S trending develops under E-W distension [Arlegui Crespo, 1996]. Normal faulting causes uplift in the Emporda Basin [Vergés et al., 1996] and alkaline volcanism migrated to this region [Vergés & Sàbat, 1999] Catalan-Sardinian margin Along the eastern margin of Iberia, N-S compression and E-W extension is observed in many littoral basins [Santanach et al., 1980]. NS-normal faults are still active in the Valencia Trough [Maillard et al., 1992]. In northeast Iberia recent uplift along buried frontal thrust of CCR and La Selva Basin has been detected by high precision leveling [Giménez et al., 1996]. Bisection of the Valles basin is related to neotectonic movements as well [de Mas Canals, 1984] and an extensive tectonic event is derived from seismicity [Masana Closa, 1996]. The Baleares are being uplifted as well: Pliocene marine sediments on Mallorca are now at 150-200m [Roca & Deselgaulx, 1992]. N. Pyrenees and SW France The stress directions derived from focal mechanism solutions in the Pyrenean domain remain unresolved [Delouis, 1993], due to difficulty of defining the stress field in a highly deformed collision zone. In the west, the North Pyrenean Fault is still active, while in the central and eastern part of the Pyrenees activity is related to the Tet fault and volcanic activity [Souriau & Pauchet, 1998]. The Massif Central is uplifted and affected by transtensional deformation, leading to ENE trending blocks [Ziegler, 1994]. The Jura foreland is still deforming, inferred by leveling [Jouanne et al., 1995]. Stress patterns in mainland France according to [Rebaï et al., 1992]. S. Alps - Adriatic domain N Apennines: stress data according to Montone et al. [1999] N. Africa N. Tunisia: stress data from WSM. Northeastern Algeria is under ~N130-150 compression [Aris et al., o o o 1998], resulting in N060 folds, inverse faults trending N050 -070 and conjugate strike-slip faults deforming the Plio-Quaternary basins in the Tellian Atlas [Meghraoui et al., 1996]. Alkaline magmatism migrated into the Tell region [Vérges & Sàbat, 1999]. The mean state of stress in northeastern Morocco

132


0Ma 21Ma

Chapter 5

is NNW-SSE directed compressional strike-slip, derived from shallow seismicity (up to ~17km) [Medina, 1995]. This direction is in good agreement with the patterns of stress directions inferred from fault slip data compilations [Galindo-Zaldívar et al., 1993]. All along the northern African Margin (from Tanger to Algiers) marine terraces are being uplifted at rates of 0.2mm/yr [Meghraoui et al., 1996]. Both structural indicators & recent earthquakes document sinistral motion along the Transalboran fault (TAF) in Morocco [Jacobshagen, 1992]. Compressive strike-slip as is shown along the southern border of the Middle Atlas, which is thrusting southeastward over Pleistocene conglomerates [Giese & Jacobshagen, 1992]. This tectonic activity is accompanied by a phase of alkaline volcanics [Giese & Jacobshagen, 1992]. Between the High and Middle Atlas, the Moulouya Basin is under N-S compression [Morel et al., 1993]. Within the High Atlas no present-day tectonic movements are being observed along the Southern High Atlas Fault (SHAF) east of the Tizi n’Test fault. To the west of this junction, the SHAF lines up with the TAF [Jacobshagen, 1992]. The southern High Atlas is being uplifted and eroded under NNW-SSE compression [Fraissinet et al., 1988], just as the Anti Atlas [Görler et al., 1988]. The Ouarzazate basin is subsiding slowly [Fraissinet et al., 1988].

133

Chapter 5

134

0Ma 21Ma



Chapter 6

CHAPTER 6 - FINITE ELEMENT MODELLING OF CENOZOIC STRESS FIELDS IN THE IBERIAN PENINSULA Based on the reconstructions presented in Chapter 4 and 5, geometry and boundary conditions can be obtained to construct numerical models of the (paleo)stress fields for different stages in the evolution of the Iberian Peninsula. Periods for which numerical models were constructed have been selected based on the following criteria: 1. completeness of the reconstructed stress field: To enable validation of the models, comparison of the results obtained by numerical modelling with the reconstructed stress field should be possible. For several periods it turned out to be impossible to construct stress fields for large parts of the studied region, either due to lack of sufficient good data or tectonic quiescence (e.g. 18 Ma). Numerical modelling has only been carried out for those periods for which a rather well constrained stress field could be reconstructed. 2. availability of first order paleotopography estimates: A first order estimate of paleotopography is required, because topography is used to calculate crustal thickness (Chapter 2). As for the present, the paleo topography is filtered with a low-pass filter with a boundary wavelength of 100km to simulate regional isostasy and to remove local effects. Panel A in Figure 6.1 demonstrates the effect of filtering the present-day topography and shows the low level of required accuracy for paleotopography, the estimates have error bars of over 500m. This is rather convenient: estimating paleotopography is a difficult process. Paleotopography can be generated to this accuracy for only a few of the time slices of Chapter 4 and 5 (see Figure 6.1.1 panels B-E). 3. representation of the major tectonic intervals in the Iberian Peninsula: The major tectonic intervals in the western Mediterranean have been outlined in detail in Chapter 4 and 5 and can be related to a) the Pyrenean collision, b) opening of the western part of the Mediterranean Basin, and c) collision in the Betics. Based on these three requirements, 5 time intervals were found to be suitable for numerical modelling: 54Ma, 36Ma, 24Ma, 12Ma and present. After a short general description of the general modelling procedure, the model setup, boundary conditions and results of the mentioned time steps will be discussed.

6.1 General model description Finite element calculations were performed using the ANSYS© software package. The models consist of triangular (type: SOLID92) elastic shell elements. Element thickness is 100km and typical element edge length in the area of interest is 100km. The models were constructed in a spherical coordinate system with longitude, latitude and Earth’s radius as xyz-coordinates. This is done to avoid errors in the geometry and direction of forces related to ridge push in the higher latitude regions. A frame was added around the free boundaries of the model to minimize edge effects, following Bada [1999]. Poisson’s ratio (ν) is kept constant at 0.25 for all of the models. To simulate differences in material properties, Young’s Modulus has been varied for the different crustal types as defined in the models for the calculation of stresses induced by density differences. Basically, two material settings were applied: (a) a constant Young’s modulus throughout the entire model or (b) different Young’s modulus for five areas: (1) young oceanic crust, (2) old oceanic crust, (3) thinned continental, (4) elevated continental and (5) surrounding 135

Chapter 6


frame. The elements in the mesh are assigned to be of any of the five types, depending on their location. The limit between oceanic and thinned continental coincides with the boundary between both as adopted by Stapel [1999]. Several routines were developed to calculate the magnitude of forces that have to be applied to the nodes of the models based on two parameters: the general value of the force per unit length for the specified boundary segment and the orientation of the force at every single node. The model results using the assumed geometries and activity at plate boundaries as derived from paleogeographical studies will be compared with the reconstructed stress fields for the different time slices. This comparison provides a quantitative test of the numerous qualitative paleogeographical concepts proposed for the tectonic evolution of the Iberian Peninsula and surrounding western Mediterranean. The models for 12Ma, 24Ma and 36Ma will show the important consequences on the stress field when choosing different tectonic scenarios. At first, an attempt was made to obtain a first order fit using the geometries resulting from the reconstructions in Chapter 4, always starting with the inferred boundary conditions. In the first runs, ridge push forces were calculated using the formulas described in Appendix A. To the best fitting scenario, stresses induced by lateral density variations are added. Their effect is discussed in the light of the changes they produce with respect to the first set of results.

6.2 Present-day stress field The best constraints on any of the (paleo)stress fields in the Iberian Peninsula are available for the present. Therefore, it is the starting point of the modelling procedure used to calibrate the applied techniques and concepts. The general trends in the trajectories of Shmax and regional state of stress within the Iberian Peninsula (see Figure 6.2.1b) have been determined by SIGMA [1998], based on a combination of focal mechanism solutions, fault-slip data and borehole breakout data. Generally, Shmax is oriented ~N140 in the southern to central peninsula, with rotations to more EW-directions along the western margin and more N-S in the northeastern corner of the peninsula. The orientation of the principal stresses and the principal stress difference ratio along the southwestern boundary of the Eurasian plate with the North American and African plates has been determined by De Vicente et al. [2000] by inverting focal mechanisms of earthquakes (see Figure 6.2.1a). The extension along the Mid Ocean Ridge (30º to 65º N) ranges from E-W to ESE-WSW, showing a triaxial extension. The state of stress changes to strike-slip regime in the transform fault zones, and the strike of Shmax rotates clockwise to NW in the dextral strike-slip faults and counterclockwise to the NE in the sinistral ones. Along the margin between the Eurasian and African plates (from the Azores triple junction to Algeria), Shmax keeps a constant NW-SE strike but the stress ratio values range from triaxial extensionin the west, to uniaxial compression in the east, passing through a strike-slip regime in the middle zone. With a NW-SE oriented Shmax, strike-slip and extensional stresses prevail in most of the Iberian Peninsula, whereas southwards it is dominated by uniaxial compressive stresses.

Observations, model geometry and boundary conditions The distribution of seismicity in the Iberian region from 1980 to present-day (as shown in Figure 2.4.1.4) has been used to determine the location of the active plate boundaries of

136


10W

5W

Chapter 6

0

10E

5E

10W

5W

0

5E

10E

B

A -

-

45N

45N +

+

+

+

+ +

-

+

+

+

+ +

+

-

+

+

-

40N

+

+

+

-

+ +

-

-2

+

-

+

0 50 -1

00 -1

+

+

-

35N

40N

+

+

+ +

-

-

00

0

1000 + +

-

35N

30N 30N

-2000 -4000 10W

5W

0

5E

-200 -1000

500 0m

1500 1000

2000

10E

10W

C

5W

0

10E

5E

D

45N

45N

+

20W

15W

10W

5W

0

E

+

0

45N +

~1500

1500-2000

~1500 ~1500

40N

0-500

+

300

1000

+

+ -200

700

40N

-200

0

~500

35N

-1000 -500

100-500

40N

-1000

35N

35N

30N

30N

+

-4000 30N

25N

Figure 6.1.1 (see www.geo.vu.nl/~andb/iberia for full color, large size) (Paleo) topography based on the reconstructions in Chapters 4 and 5 (accuracy at maximum 500m) filtered with a low pass filter applying a boundary wavelength of 100km to simulate regional isostasy. For the present-day situation (panel a), both the real topography (gray) and the filtered data set (color) are shown to indicate the accuracy level of the paleotopography. b) Filtered paleotopography for 12Ma, c) Filtered paleotopography for 24Ma, d) Filtered paleotopography for 36Ma and e) Filtered paleotopography for 54Ma.

Iberia. The western boundary is formed by the Mid Atlantic ridge from the Azores until north of Spitsbergen, aligned with shallow seismic activity. Ridge push forces have been calculated using the formulas described in Appendix A and are equivalent to 3.0x1012N/m along the southern part diminishing northwards to 1.5x1012N/m near Iceland and 0.8x1012N/m near Spitsbergen. The southern boundary of the model is located along the spreading Azores and Terceira ridge, which show up by superficial (