From the Bay of Biscay to the High Atlas: Completing the anisotropic characterization of the upper mantle beneath the westernmost Mediterranean region
Introduction
The investigation on the anisotropic properties of the uppermost mantle is one of the best approaches to better understand the geodynamic processes affecting this depth range. The origin of the upper mantle anisotropy has been classically related to the strain-induced lattice preferred orientation (LPO) of the mantle minerals, in particular of olivine (e.g., Nicolas and Christensen, 1987) developed in response to tectonic flow. Even if the relationship between deformation and anisotropy properties is not straightforward, in tectonically active areas (mid-ocean ridges, rifts, subduction zones) fast polarization directions (FPDs) are expected to mark the direction of mantle flow, while in zones without present-day large-scale tectonic activity, FPD can be related to the strain from the last significant tectonic episode preserved in the subcrustal lithosphere, to dynamic flow in the asthenosphere or to the combined effect of both mechanisms (Savage, 1999, Silver, 1996, Vauchez et al., 2012).
Uppermost mantle anisotropy can be explored using different seismic methodologies, including surface wave scattering, Pn tomography, P wave travel-time azimuthal variation and shear-wave splitting, the latter being widely accepted as the most fructiferous approach. When traveling across an anisotropic medium, a shear wave will split in two waves orthogonally polarized and traveling with different velocities, which will arrive to the seismic station separated by a certain time delay. As this delay is smaller than the period of the teleseismic shear-waves, the polarization of these waves is not linear but shows a characteristic ellipticity. SKS waves, which travel as compressional waves through the external core and are converted again to shear waves in the core mantle boundary (CMB) grossly beneath the station are widely used to investigate anisotropy, as their waypaths assure that any anisotropic effect detected must be generated near vertically beneath the receiver station, hence providing a good lateral resolution. On the contrary, the method does not provide direct constrains on the vertical position of the anisotropic zone. During the last decades a large amount of contributions have analyzed SKS splitting in many tectonic settings, from the oldest cratons to active subduction zones (e.g., Long, 2013, Vinnik et al., 2012). As SKS waves have typical periods of 5–10 s, they lay in the middle of the microseismic peak, the zone of the seismic spectra with the largest background noise, mostly related to the interaction of oceanic waves. This makes difficult to get clear SKS arrivals, even for large magnitude events. Additionally, SKS waves need to be isolated from other phases traveling with similar apparent velocities, thus limiting the distance range of useful events to 85°–120°. The first contributions in the 1980s and 1990s (Silver and Chan, 1988, Vinnik et al., 1989) were based in the data from the scarce number of permanent broad band stations deployed worldwide. With the availability of an increasing number of portable broad-band seismometers, experiments focused on anisotropy started to be carried out at local scale using dense deployments recording data during weeks to months. Those studies have been typically focused in relatively small areas or devoted to study larger regions by means of high density linear profiles crossing their main structures. The EarthScope project, started in 2005 in the United States, marked the beginning of a new era in seismic exploration, as it involved the deployment of an homogeneous network of broad-band stations covering the contiguous USA with a regular grid of about 70 km × 70 km. Regarding anisotropic studies, the EarthScope project allowed to obtain very detailed results, including, for example, the observation of toroidal flow around the Juan de Fuca plate (Zandt and Humphreys, 2008). The Iberarray seismic network, integrated in the large-scale TopoIberia project (Díaz et al., 2009), has allowed deploying a similar network in Iberia and northern Morocco. Using more than 70 broad-band instruments and three different deployments, and integrating the permanent stations in the area, the experiment provided a final database holding more than 300 sites, forming a 60 km × 60 km grid and covering a region of approximately 600.000 km2. Therefore, the TopoIberia network is one of the first examples of high density regional scale seismic experiment providing information on large scale regions with unprecedented detail.
The anisotropic analysis of the first TopoIberia deployments, covering North Morocco and the Southern and central part of the Iberian Peninsula, have already been published by Díaz et al. (2010) and Diaz and Gallart (2014). The objective of this contribution is twofold; firstly, we will present the anisotropic results derived from the data recorded during the last TopoIberia deployment, covering the northern part of Iberia, from the Mediterranean Sea to the Atlantic passive margin (Fig. 1). Secondly, the results from the three deployments will be summarized in order to get, for the first time, a comprehensive image of the anisotropic properties of the westernmost Mediterranean region. In order to complete this image, we present also the anisotropic parameters derived from the analysis of broadband seismic stations in Portugal, including permanent sites and the stations deployed in the framework of the WILAS project, an independent experiment designed to complete the Iberia coverage provided by TopoIberia (see Custodio et al., 2014 for details). The retrieval of the anisotropic properties of this large area allows to further constraining the geodynamic interpretation of the region.
Section snippets
Tectonic setting of northern Iberia
The northern part of the Iberian Peninsula has been affected by two large compressional episodes, the Variscan and Alpine orogenies, separated by a period of significant extensional deformation in the Mesozoic.
The Variscan orogeny started with the closure of the Rheic Ocean and the collision between Laurentia-Baltica-Avalonia and the continental margin of Gondwana during the Carboniferous, giving rise to the building of the Pangea supercontinent (Matte, 1991). The western part of northern
Data acquisition and processing
The third Iberarray deployment started in fall 2010 and was fully operational as of spring 2011. Most of the stations remained active until the end of the project, in summer 2013, hence providing more than two years of available data. Data from the permanent networks have been analyzed for the period 2010–2013, thus gathering about 3 years of teleseismic recordings. This has resulted in a dataset comprising 98 stations to be added to the nearly 200 sites previously investigated by Díaz et al.
Geodynamic interpretation
In order to interpret geodynamically the anisotropic parameters, a good knowledge of the mechanism responsible for the anisotropy is needed. It has been established that the crust contribution to the delay times observed between split SKS waves is limited to few tens of second and hence that most of the anisotropy comes from the upper mantle, even if it remains unclear whether anisotropy is confined to the upper 200 km or it continues through the transition zone (Savage, 1999, Yuan and Beghein,
Conclusions
The last TopoIberia deployment covers the northern part of the Iberian Peninsula, from the eastern Pyrenees to the most hinterland parts of the Variscan belt. The results show an average fast velocity direction close to E–W. The origin of this anisotropy is globally related to the LPO of mantle minerals generated by mantle flow at asthenospheric depths. Delay times of around 1.0 s–1.5 s are observed in most of the stations, but lower values, not exceeding 0.8 s, are measured in the
Acknowledgments
We want to acknowledge all the field and processing Spanish and Moroccan teams of the TopoIberia-Iberarray experiment that made this large dataset available. We want also to acknowledge the institutions responsible for the maintenance of the different permanent seismic networks contributing data to this study (Instituto Geográfico Nacional, Real Observatorio de la Armada, Instituto Andaluz de Geofísica, and Portuguese Seismic Network). This is a contribution of the Consolider-Ingenio 2010
References (67)
- et al.
Mapping upper mantle anisotropy beneath SE France by SKS splitting indicates Neogene asthenospheric flow induced by Apenninic slab roll-back and deflected by the deep Alpine roots
Tectonophysics
(2004) - et al.
Evidence for slab rollback in the westernmost Mediterranean from improved upper mantle imaging
Earth Planet. Sci. Lett.
(2013) - et al.
Seismic anisotropy beneath Southern Iberia from SKS splitting
Earth Planet. Sci. Lett.
(2008) - et al.
From the North Iberian Margin to the Alboran Basin: a lithosphere geo-transect crossing the Iberian Plate
Tectonophys
(2015) - et al.
Cenozoic thick-skinned deformation and topography evolution of the Spanish Central System
Glob. Planet. Chang.
(2007) - et al.
Seismic anisotropy from the Variscan Core of Iberia to the Western African Craton: new constrains on upper mantle flow at regional scales
Earth Planet. Sci. Lett.
(2014) - et al.
Probing seismic anisotropy in North Iberia from shear wave splitting
Phys. Earth Planet. Inter.
(2006) - et al.
The structure and evolution of the lithosphere–asthenosphere boundary beneath the Atlantic–Mediterranean transition region
Lithos
(2010) - et al.
From mantle to crust: stretching the Mediterranean
Earth Planet. Sci. Lett.
(2009) - et al.
Mantle dynamics and seismic anisotropy
Earth Planet. Sci. Lett.
(2010)
Accretionary history and crustal evolution of the Variscan belt in Western Europe
Tectonophysics
Crustal structure of the external variscides in northwest Spain from deep seismic reflection profiling
Tectonophysics
Faults (shear zones) in the Earth's mantle
Tectonophysics
SplitLab: a shear-wave splitting environment in Matlab
Comput. Geosci.
Identifying global seismic anisotropy patterns by correlating shear-wave splitting and surface waves data
Phys. Earth Planet. Inter.
Seismic anisotropy changes across upper mantle phase transitions
Earth Planet. Sci. Lett.
Tertiary basins and Alpine tectonics in the Cantabrian Mountains (NW Spain)
Anisotropy patterns in the subducting lithosphere and in the mantle wedge: A case study—The southern Italy subduction system
J. Geophys. Res.
Lithospheric anisotropy beneath the Pyrenees from shear wave splitting
J. Geophys. Res.
Azimuthal seismic anisotropy constrains net rotation of the lithosphere
Geophys. Res. Lett.
The North and North-West Spanish continental margin: a review
Rev. Soc. Geol. Esp.
Multiple-frequency tomography of the upper mantle beneath the African/Iberian collision zone
Geophys. J. Int.
High-resolution imaging of the Pyrenees and Massif Central from the data of the PYROPE and IBERARRAY portable array deployments
J. Geophys. Res.
Constraints on lithosphere net rotation and asthenospheric viscosity from global mantle flow models and seismic anisotropy
Geochem. Geophys. Geosyst.
Ambient noise recorded by a dense broadband seismic deployment in Western Iberia
Bull. Seismol. Soc. Am.
Kinematics of the Western Mediterranean
Anisotropic features of the Alpine lithosphere in Northern Spain
Geophys. Res. Lett.
The IBERARRAY broadband seismic network: a new tool to investigate the deep structure beneath Iberia
Orfeus Newsl.
Mantle dynamics beneath the Gibraltar Arc (western Mediterranean) from shear-wave splitting measurements on a dense seismic array
Geophys. Res. Lett.
Shear wave anisotropy beneath the Aegean inferred from SKS splitting observations
J. Geophys. Res.
Mantle dynamics in the Mediterranean
Rev. Geophys.
Aportaciones al conocimiento de la litoestratigrafía y estructura de Galicia Central
Mem. Fac. Cienc. Porto
The Cenozoic geodynamic evolution of the Valencia trough (western Mediterranean)
Rev. Soc. Geol. Esp.
Cited by (29)
Azimuthal anisotropy of the westernmost Mediterranean: New constraints on lithospheric deformation and geodynamical evolution
2022, Earth and Planetary Science LettersCitation Excerpt :However, these observations refer to the NW part of Iberia, an area where our model does not need a second layer. Díaz et al. (2015), using data acquired by permanent and temporary deployments, show that although the mean fast velocity directions retrieved from SKS analysis around central Iberia are consistently oriented close to the E-W direction and do not require the presence of multiple anisotropic layers, several good-quality measurements do not adjust the general trend, suggesting the presence of second-order features beneath northern Iberia in an area including the zone with differentiated lithospheric anisotropy. Controlled source wide-angle reflection and refraction seismic profiles beneath the southwestern Iberian Massif have shown the presence of layered anisotropy in the mantle lithosphere, with fast velocity directions oriented NE-SW, clearly different from those retrieved from SKS splitting measurements (Díaz et al., 1996).
Can sub-slab low-velocity anomalies be an artifact caused by anisotropy? A case study from the Alboran slab area in the western Mediterranean
2021, TectonophysicsCitation Excerpt :3) Toroidal-domain: A toroidal flow domain behind (i.e. west of) the downdip entrained flow domain. SFDs in this domain are horizontal and with a strike that mimics the arcuate pattern of the Gibraltar arc and the FPDs of SWS observations (Buontempo et al., 2008; Diaz et al., 2010, 2015; Miller et al., 2013; Wüstefeld et al., 2009) (e.g. Fig. 2b). This domain represents the anisotropy produced by the rollback process, which presumably compresses the ambient mantle behind the slab (e.g. Faccenda and Capitanio, 2012, 2013; Foley and Long, 2011; Long and Silver, 2008).
Four decades of geophysical research on Iberia and adjacent margins
2021, Earth-Science ReviewsA common deep source for upper-mantle upwellings below the Ibero-western Maghreb region from teleseismic P-wave travel-time tomography
2018, Earth and Planetary Science LettersCitation Excerpt :We propose that the toroidal component of slab rollback-induced mantle flow entrained hot upwelling mantle material from the sub-slab domain (L3), directing hot material around the lateral slab edges to below the eastern Betics and eastern Rif. Fast polarisation directions of SKS-waves (Buontempo et al., 2008; Díaz et al., 2015, 2010; Miller et al., 2013) show a general trend of slab rollback-induced toroidal mantle flow in the Gibraltar region. Fully dynamic models of interaction between a slab and a sub-slab mantle upwelling have shown that the quasi-toroidal mantle flow driven by the slab can transport the hot material around the lateral slab edges even before the slab comes in contact with the upwelling (Mériaux et al., 2015b, 2016).
Evidence for mantle heterogeneities in the westernmost Mediterranean from a statistical approach to volcanic petrology
2017, LithosCitation Excerpt :Eruption of calc-alkaline and HK calc-alkaline magmas is in agreement with a mature subduction process (no retreating or active subduction), as proposed by Vergés and Fernàndez (2012) in stage 7. In northern Africa (Morocco and west Algeria), the sodic alkaline volcanic episodes following the typical subduction-related products can be mainly ascribed to three mechanisms: i) sublithospheric flow, as evidenced from mantle anisotropy analysis (Diaz et al., 2010, 2015); ii) slab break-off, as also proposed for the alkaline products of the intra-Apennines volcanism by Bianchini et al. (2008); or iii) splash-plume effect (Davies and Bunge, 2006) due to convective instabilities, as has been proposed by Beccaluva et al. (2011) for Sardinia and Calatrava (southern Spain) volcanism. Furthermore, the occurrence of anorogenic rocks in Tallante is additional evidence of a lateral tear responsible of uplift of the eastern part of the Betics (e.g., García-Castellanos and Villaseñor, 2011; Mancilla et al., 2015; Spakman and Wortel, 2004), as included in the model by Vergés and Fernàndez (2012).