Geophysical and geological characterization of the active structures and of the nature of the basement in the Eurasia-Africa plate boundary (SW Iberian Margin): Implications for regional geodynamics and seismic hazard assessment
Resumen Abstract Índice Conclusiones
Sara Martínez Loriente
2014-A
The dissertation of this PhD Thesis has been organized as a collection of scientific publications and includes four articles, three of them published and one submitted for publication in scientific journals of the Science Citation Index (SCI):
1. Martínez-Loriente, S., E. Gràcia, R. Bartolome, V. Sallarès, C. Connors, H. Perea, C. Lo Iacono, D. Klaeschen, P. Terrinha, J. J. Dañobeitia, and N. Zitellini (2013), Active deformation in old oceanic lithosphere and significance for earthquake hazard: Seismic imaging of the Coral Patch Ridge area and neighboring abyssal plains (SW Iberian Margin), Geochemistry Geophysics, Geosystems (G3), in press, doi:10.1002/ggge.20173.
2. Bartolome, R., E. Gràcia, D. Stich, S. Martínez-Loriente, D. Klaeschen, F. L. Mancilla, C. Lo Iacono, J.J. Dañobeitia, and N. Zitellini (2012), Evidence for active strike-slip faulting along the Eurasia-Africa convergence zone: Implications for seismic hazard in the SW Iberian Margin, Geology, 40 (6), 495-498, doi:10.1130/G33107.1.
3. Sallarès, V., S. Martínez-Loriente, M. Prada, E. Gràcia, C. R. Ranero, M. A. Gutscher, R. Bartolome, A. Gailler, J. J. Dañobeitia, and N. Zitellini (2013), Seismic evidence of exhumed mantle rock basement at the Gorringe Bank and the adjacent Horseshoe and Tagus abyssal plains (SW Iberia), Earth and Planetary Science Letters, 365, 120-131, doi:10.1016/j.epsl.2013.01.021.
4. Martínez-Loriente, S., V. Sallarès, E. Gràcia, R. Bartolome, J. J. Dañobeitia, and N. Zitellini (submitted), Seismic and gravity constraints on the nature of the basement in the Africa-Eurasia plate boundary: New insights on the geodynamic evolution of the SW Iberian Margin, Journal of Geophysical Research (Solid Earth), under review.
To accomplish with the Article 37 of the doctorate regulation 99/2011 from the University of Barcelona regarding the doctoral thesis presented as a compendium of publications, this PhD Thesis has been structured into four main parts:
Part I: It corresponds to the introductory section and includes Chapters 1, 2, and 3. Chapter 1 presents the interest and motivation of this Thesis and the main objectives. Chapter 2 presents the geological setting of the study area, the southwest Iberian margin. Chapter 3 is an overview of the different geophysical methods used to carry out this work.
Part II: It includes Chapters 4, 5, 6, and 7 corresponding to extended summaries of the results, discussion and conclusion of this Thesis. Chapter 4 includes the results of this Thesis, which have been divided into three blocks according to the location of the different structures analyzed and / or the methodology used. The discussion of this Thesis is included in Chapter 5, which is divided into four sections according to the different topics covered. Chapter 6 includes the main conclusions of this Thesis. Finally, Chapter 7 includes a series of suggestions (i.e. forward look) that may provide new light into questions that still remain open after this work.
Part III: References. It corresponds to the alphabetic list of bibliographic references quoted in this Thesis. A list of acronyms used in this Thesis is also added.
Part IV: Annexes. This Thesis includes three annexes. Annex I corresponds to the four articles as they appear on the respective scientific journals. Annex II corresponds to the uninterpreted time migrated multichannel seismic profiles SW01 to SW16, which have been used in this Thesis. Annex III includes the record sections of OBS 01 to OBS 30 from the NEAREST profile P1.
In this PhD Thesis I present a new interpretation of: 1) active structures implicating old oceanic lithosphere; 2) the nature of the basement; and 3) the distribution of the basement domains and the geodynamic reconstruction of the SW Iberian margin, a region that hosts the slow convergent boundary between the African and Eurasian plates. This interpretation is based on new geophysical data acquired, processed and modeled in the framework of this PhD work. The main findings of my study are the following ones:
1) Recently acquired high-resolution multichannel seismic profiles together with bathymetric and sub-bottom profiler data (SWIM 2006 survey) from the external part of the Gulf of Cadiz (Eurasia-Africa plate boundary) reveal active deformation involving old (Mesozoic) oceanic lithosphere [Martínez-Loriente et al., 2013]. This dataset shows active strike-slip occurring along the prominent lineaments North and South, imaging seafloor displacements and active faulting to depths of at least 10 km and of a minimum length of 150 km [Bartolome et al., 2012]. Seismic moment tensors show predominantly WNW–ESE right-lateral strike-slip motion [Geissler et al., 2010]. Estimates of earthquake source depths close to the fault planes indicate upper mantle (i.e., depths of 40–60 km) seismogenesis [Stich et al., 2010, Bartolomé et al., 2012], implying the presence of old, thick, and brittle lithosphere. Moreover, the SWIM 2006 dataset also reveals E-W trending dextral strike-slip faults showing surface deformation of flower-like structures, which predominate in the Horseshoe Abyssal Plain. In contrast, NE-SW trending compressive structures prevail in the Coral Patch Ridge and in the Seine Hills [Martínez-Loriente et al., 2013]. Although the Coral Patch Ridge region is characterized by subdued seismic activity, the area is not free from seismic hazard. Most of the newly mapped faults correspond to active blind thrusts and strike-slip faults that are able to generate large magnitude earthquakes (Mw 7.2 to 8.4) [Martínez-Loriente et al., 2013].
2) Combined seismic and gravity modeling along NEAREST profile P1 acquired in the external part of the SW Iberian margin, reveals the presence of a serpentinized peridotite basement flooring the Gorringe Bank and adjacent sectors of the Tagus and Horseshoe abyssal plains [Sallarès et al., 2013]. These three domains would be part of a wide ultramafic rock band [Sallarès et al., 2013], similar to the Zone of Exhumed Continental Mantle off Western Iberia [Pinheiro et al., 1992; Dean et al., 2000]. Furthermore, the basement velocity structure of the southeastern part of the profile (i.e., the Coral Patch Ridge and Seine Abyssal Plain) indicates the presence of a highly heterogeneous, thin oceanic crust (4-6 km-thick), similar to that described in slow/ultra-slow spreading centers, with local high-velocity anomalies possibly representing serpentinite intrusions [Martínez-Loriente et al., submitted].
3) The integration of the results from NEAREST profiles P1 and P2 that runs across the central Gulf of Cadiz [Sallarès et al., 2011], and previously existing data reveals the presence of three main oceanic domains offshore SW Iberia [Martínez-Loriente et al., submitted]: (a) the Seine Abyssal Plain domain, made of oceanic crust that would be generated during the first slow (~8 mm/yr) stages of seafloor spreading of the northeastern segment of the Central Atlantic (i.e. 190 Ma – 180 Ma) [Martínez-Loriente et al., submitted]; (b) the Gulf of Cadiz domain, constituted of oceanic crust generated in the Alpine-Tethys spreading system between Iberia and Africa, which was coeval with the formation of the Seine Abyssal Plain domain and lasted up to the North Atlantic continental break-up (Late Jurassic) [Sallarès et al., 2011]; and (c) the Gorringe Bank domain, made of exhumed mantle rocks that was probably generated during the earliest phase of the North Atlantic opening that followed the continental crust breakup (Early Cretaceous) [Sallarès et al., 2013]. During the Miocene, the NW–SE trending Eurasia–Africa convergence resulted in thrusting of the southeastern segment of the exhumed serpentinite band over the northwestern one, forming the Gorringe Bank [Sallarès et al., 2013]. These models indicate that the Seine Abyssal Plain and Gulf of Cadiz domains are separated by the Lineament South strike-slip system, whereas the Gulf of Cadiz and Gorringe Bank domains are bounded by a deep thrust fault system located at the center of the Horseshoe Abyssal Plain, which we refer to as the Horseshoe Abyssal plain Thrust [Martínez-Loriente et al., submitted].
These new findings are relevant for geohazard assessment in the region. On one hand, the presence of active deformation has been demonstrated in the external part of the Gulf of Cadiz, involving structures considered inactive [e.g. Zitellini et al., 2009] until the present work. On the other hand, the knowledge of the nature of the SW Iberian margin basement may provide valuable information into the process of seismogenesis, such as earthquake nucleation and velocity propagation. Both aspects will help to refine regional seismic and tsunami hazard assessment models.
Contents
Summary i
Organization of this Thesis iii
PART I: INTRODUCTION 1
CHAPTER 1. Objectives and scientific approach 3
1.1. Interest of the study 3
1.2. Objectives 7
1.3. Definition of general concepts 8
1.3.1. Basic concepts of seismic hazard 8
1.3.1.1. General concepts of seismicity 11
1.3.1.2. Identification of seismic sources 12
1.3.1.3. Historical and paleoseismic data 14
1.3.2. Basic concepts of plate tectonics 16
CHAPTER 2. Geological setting of the SW Iberian margin 23
2.1. Geodynamic evolution 24
2.2. Morphology 26
2.3. Main geological structures 29
2.4. Stratigraphy 42
2.5. Seismicity 44
CHAPTER 3. Methods 49
3.1. Data acquisition 49
3.1.1. SWIM 2006 cruise 49
3.1.2. NEAREST 2008 cruise 51
3.2. Geophysical methods used 51
3.2.1. Swath-bathymetry and acoustic backscatter 52
3.2.2. High-resolution sub-bottom profiler 55
3.2.3. Multichannel seismic data 58
3.2.3.1. Standard MCS processing sequence 61
3.2.3.2. Pre-stack depth migration (PSDM) 66
3.2.3.3. Criteria for MCS data interpretation 71
3.2.4. Combined seismic and gravity data modeling 73
3.2.4.1. Processing and phase picking 77
3.2.4.2. Joint refraction and reflection travel-time inversion method 81
3.2.4.2.1. Uncertainty of the velocity model parameters 85
3.2.4.3. Velocity-derived density modeling 87
3.2.4.3.1. Velocity-density empirical relationships 87
3.2.4.3.2. Calculation of the gravity anomaly 89
3.2.4.4. Velocity-derived serpentinization degree 90
PART II: RESULTS, DISCUSSION AND CONCLUSIONS 91
CHAPTER 4. Results 93
4.1. Seismic evidence for active strike-slip faulting along the Eurasia-Africa plate boundary (Zone 1) 94
4.1.1. Multi-scale seismic imaging of the SWIM Lineaments 95
4.1.2. Assigning recent earthquakes to the SWIM Lineaments 98
4.2. Acoustic and seismic imaging of active structures of the external part of the Gulf of Cadiz (Zone 2) 100
4.2.1. Morphology and stratigraphy of the Coral Patch Ridge and neighboring Horseshoe and Seine abyssal plains 101
4.2.1.1. Seafloor morphology 101
4.2.1.2. Seismostratigraphy 103
4.2.2. Tectonic structure of the Coral Patch Ridge and neighboring abyssal plains 107
4.2.2.1. Eastern Horseshoe Abyssal Plain 107
4.2.2.2. Coral Patch Ridge 110
4.2.2.3. Northern Seine Abyssal Plain 113
4.3. Combined wide-angle seismic and gravity modeling to characterize the external part of the Gulf of Cadiz (Zone 3) 117
4.3.1. Description of the velocity model of the NW part of profile P1 119
4.3.2. Gravity modeling of the NW part of the profile P1 122
4.3.3. Description of the velocity model of the SE part of the profile P1 125
4.3.4. Gravity modeling of the SE part of the profile P1 129
CHAPTER 5. Discussion 133
5.1. Tectono-sedimentary evolution and active deformation in the external part of the Gulf of Cadiz 133
5.1.1. Tectono-sedimentary evolution of the Coral Patch Ridge region 133
5.1.2. Synthesis of active faults in the Coral Patch Ridge area 139
5.1.3. Seismic potential of the largest faults analyzed: Implications for earthquake and tsunami hazard assessment models 141
5.2. Basement affinity of the external part of the Gulf of Cadiz 145
5.2.1. Nature of the basement in the Gorringe Bank and adjacent Horseshoe and Tagus abyssal plains 145
5.2.2. Nature of the basement in the Coral Patch Ridge and Seine Abyssal Plain 149
5.2.3. Boundary between the serpentinized peridotite basement in the northern part of the HAP to the oceanic crust in the CPR 152
5.3. Definition of the geological provinces in the SW Iberian margin and their plausible origin 153
5.3.1. Geological cross-section along profile P1 155
5.3.1.1. Mantle exhumation during the Mesozoic extension 155
5.3.1.2. Uplift of the Gorringe Bank during the Miocene convergence 157
5.3.1.3. Thin oceanic crust generated during the early-slow stage of seafloor spreading of the Central Atlantic 159
5.3.1.4. Tomographic expression of large-scale faults resulting from the Miocene convergence stage 160
5.3.2. Geological cross-section along profile P2 162
5.3.3. Classification of the geological domains off the SW Iberian margin 165
5.4. Geodynamic evolution of the SW Iberian margin 167
CHAPTER 6. Conclusions 171
CHAPTER 7. Forward look 175
PART III: REFERENCES 181
List of acronyms 183
References 189
PART IV: ANNEXES 215
Annex I: Scientific articles 217
Annex II: Uninterpreted time migrated MCS SWIM profiles 299
Annex III: OBS record sections of NEAREST profile P1 307
In this chapter we summarize the main conclusions that have been attained from the integration of a number of geophysical data and models in the SW Iberian margin in the framework of this PhD Thesis. These data and models have been used to characterize the structure and properties of the sediments and basement up to the uppermost mantle in the different domains that constitute:
1. Combined WAS and gravity modeling provides compelling geophysical evidence indicating the basement affinity of the different structural domains in the external part of the Gulf of Cadiz. Integrating all the observations we propose the first map of the basement affinity of the SW Iberian margin together with a plausible geodynamic evolution:
a) The basement in the GB and the adjacent segments of the HAP and TAP are mainly made of serpentinized peridotite. The basement is characterized by a strong vertical velocity gradient in the upper ~4–5 km, by a higher velocity in the underlying 5 km, and the absence of crust–mantle boundary reflections in most record sections. We propose that the GB and adjoining sectors of the TAP and HAP were originated by exhumation of a single, 150–180 km-wide mantle band similar to the ZECM of the IAP. According to plate tectonic reconstructions and rock dating, the basement was exhumed by tectonic mantle denudation during the initial phase of the North Atlantic opening in the Earliest Cretaceous (147–133 Ma).
b) The basement in the CPR and the SAP is constituted by a thin oceanic crust. The velocity structure is characterized by the presence of a thinner-than-normal oceanic layer L3 (0.5-3 km-thick), a high lateral variability with high- velocity anomalies and a discontinuous Moho that we relate with the presence of localized serpentinized peridotite bodies. We propose that the oceanic crust present in the CPR and SAP areas was generated during the early-slow (~8 mm/yr) stages of seafloor spreading of the northeastern segment of the Central Atlantic (i.e. 190 Ma – 180 Ma).
c) There is evidence in the WAS data of the presence of an abrupt boundary in the middle of the HAP between the oceanic crust of the CPR and SAP, and the basement made of exhumed mantle rocks of the northern part of the HAP, at the GB and at the southern TAP. The sharp limit between the two domains appears to occur at the HAT, a deep SE-dipping reflector with a dip angle of ~30º.
d) The SE-dipping low-velocity anomalies identified in the velocity structure of the GB and in the thin oceanic crust of the CPR and SAP, may be the tomographic expression of crustal-scale faults and fault-related rock fracturing, which may have favored rock alteration by fluid percolation along the fault planes. In the case of the GB, this may be the first evidence of the large-scale thrusting developed within the exhumed mantle rock band as a response to the NW–SE-directed Miocene convergence between Eurasian and African plates that uplifted the GB. In the case of the CPR and SAP, the low-velocity anomalies spatially coincide with the major thrust faults identified in the MCS data (i.e., the NCP, SCP, and SH thrust faults). The uppermost mantle shows low velocities that may indicate serpentinization at upper mantle levels, suggesting that these thrust faults cross the Moho and reach the upper mantle.
e) After a reassessment of the NEAREST profile P2, which runs from the south Portuguese Margin to the SAP, and considering kinematic reconstructions, we propose that the 150 km-wide segment of oceanic crust is actually composed of two different segments generated by different rift systems. The northern part (~80 km-wide) would correspond to the only remnant western Alpine-Tethys, generated by oblique seafloor spreading through a transform system that developed between Iberia and Africa at Early-Late Jurassic (180-145 Ma). The southern segment would have been generated during the first stages of seafloor spreading of the Central-Atlantic, as described in the CPR and SAP. These two domains are separated by the LS strike-slip system, the major of the inherited structures of the Jurassic transform zone that were reactivated during the Neogene convergence.
f) According to the new basement affinities interpreted on the NEAREST profiles and integrating previous results from other WAS and MCS data, rock basement samples, and location of magnetic anomalies, we propose that the basement offshore the SW Iberian margin is composed of three main oceanic domains: (1) the Seine Abyssal plain, made of oceanic crust generated in the NE Central Atlantic during Early Jurassic; (2) the Gulf of Cadiz domain, composed of oceanic crust generated in the Alpine-Tethys system and coeval with the formation of the Seine Abyssal Plain domain; and (3) the Gorringe Bank domain, made of exhumed mantle rocks and generated during the first stages of North Atlantic opening, just after the end of spreading between Iberia and Africa.
2. The combined interpretation of high-resolution SWIM 2006 multichannel seismic reflection profiles together with swath-bathymetry, sub-bottom profiles and sediment cores yield new insights into the tectonic architecture and crustal structure of the CPR area and surrounding abyssal plains:
a) The geometry of the seismostratigraphic units allowed us to characterize successive deformation phases in the outer part of the Gulf of Cadiz and to distinguish the syn-extensional, post-extensional and syn-compressional sedimentary sequences in each domain.
b) NE-SW trending thrusts (NCP, SCP and SH1-SH6) and WNW-ESE trending sub-vertical dextral strike-slip faults (e.g. LS, and SS1) occur in the old oceanic lithosphere of the HAP, CPR, and SAP, and are consistent with the NW–SE regional shortening axis between Eurasia and Africa. These structures cut, fold or show growth-strata configuration in the most recent sedimentary units of Holocene age, indicating that they are active.
c) The major thrust faults in the CPR and SAP probably propagated from the same detachment level located either at the Moho (~7-8 km depth below the seafloor), or at greater depths below the serpentinized area in the uppermost mantle, at ~12-13 km below the seafloor. Secondary structures probably also propagated from a shallower detachment level in the upper part of the oceanic crust (between 2.5 km and 4.5 km depth below the seafloor).
d) The NE-SW trending thrusts located south of the SFZ probably grew through weakened zones by fracturing due to the opening of the NE segment of the Jurassic Central Atlantic rifting. The WNW-ESE trending strike-slip faults concentrated in the HAP may correspond to a reactivation of inherited structures from a Jurassic transfer zone located across the Strait of Gibraltar.
3. As for the earthquake and tsunami hazard assessment, the strike-slip faults represent one of the largest clusters of seismicity in the Gulf of Cadiz (nucleating in the upper mantle, > 50 km), whereas a maximum earthquake of Mw > 8 could be generated by the LS and LN. Despite the low seismic activity recorded south of the SFZ, our data suggest that the thrusts are active and potential sources of large magnitude (Mw > 7) seismic events and associated tsunamis. Furthermore, the complex and large diversity of types of basement that floors the SW Iberian Margin gives new light into the characterization of the seismogenic and tsunamigenic sources in the region, which from now on will need to take into account the geological variability between domains (i.e. age, lithology, rheology) revealed by our new findings.