THE SEISMIC STRUCTURE OF THE ARABIAN PLATE AND THE SURROUNDING PLATE BOUNDARIES

Much of the Cenozoic geology and the present lithospheric and upper mantle structure of the Arabian-Eurasian collisional belt including the Anatolian and Iranian plateaus are the result of the collision and suturing of the continental Arabian plate to the Eurasian and the escape of the Anatolian plate to the west. This process of collision, suturing, and escape was strongly influenced by three active structures in the region: the Caucasus mountains, the Zagros mountains, and the Dead Sea fault system. In order to address upper mantle dynamics in this region we review the variations in the seismic wave velocity structure across the Anatolia plate and the African-Arabian-Anatolian plate boundaries. Recent techniques are giving us a new view of the seismic velocity structure across these important tectonic provinces and thus important insights into their evolution.

The northern portion of the Arabian plate appears to be relatively thin (~120 km) and in places relatively since there is circumstantial evidence to suggest regions of partial melting. Specifically along the Dead Sea Fault System we observe some evidence for a hot and possibly thin lithosphere. In general these anomalies coincide with Cenozoic volcanism. The adjacent Eurasian and Anatolian lithosphere, however, is even more anomalous. There appear to be large regions of the Anatolian plateau and Lesser Caucasus where there is very little, if any, lithospheric mantle. Using seismic records collected from 54 broadband stations located in the region were used to develop a new three dimensional shear wave velocity model we determined the fundamental mode Rayleigh wave phase velocities at periods between 20 and 145 seconds. We used the method of Yang and Forsythe (2006) that models the network variations in phase and amplitude of teleseismic Rayleigh waves to map the variations in phase velocity. We observe a relatively high velocity zone located in the upper mantle under the Kura basin and the western part of Caspian Sea that is continuous to the Moho. This high velocity may either be a subducting slab or a thick lithospheric mantle root beneath the basin. We observe a separate high velocity body beneath the Lesser Caucasus that appears only in our longer period (> 100s) phase maps. We believe that this may be the remnant Neo Tethys slab that broke off after the initiation of continent-continent collision between the Arabian and Eurasian plates. Also our images show very low velocities implying the existence of a partially molten asthenospheric material underlying a very thin lithosphere beneath most of the Anatolian plateau. Our surface wave results are broadly consistent with teleseismic S-wave receiver functions across the eastern Anatolian plateau and Lesser Caucasus that show a very shallow (60-80 km) Lithosphere-Asthenosphere Boundary (LAB). The shallow LAB, young alkaline volcanism, and recent plateau uplift are all consistent with a recent break-off of the Neo-Tethys oceanic slab.

In terms of shear wave splitting across the northern Middle East, nearly all stations in central and eastern Anatolia have a NE-SW fast direction and lag time similar to that observed from temporary broadband stations within the Anatolian plate, indicating that the anisotropic fabric may be relatively uniform throughout the upper mantle beneath the Anatolian plate. The extensive young basaltic volcanism, regional travel time tomography, and regional phase attenuation tomography all indicate that observed anisotropy is from existing finite strain, not a preserved lithospheric mantle anisotropic fabric. Furthermore, we have also obtained the depth extent of the azimuthal anisotropy beneath the eastern Anatolian plateau and Lesser Caucasus using Rayleigh wave phase velocities. We have found that beneath eastern Anatolia the fast directions do not vary significantly with depth. We have found that the magnitude of the anisotropy is relatively (small ~2%) and also does not vary strongly with depth down to depths of 200 km Unless exceptionally high (approximately 12%) anisotropy exists in the thinned lithosphere, the main contribution to the observed delay times (of order 1 s) must therefore be asthenospheric and thus reflect recent asthenospheric flow patterns. Furthermore, the lack of correlation between surface deformation and the observed shear wave splitting parameters would suggest that the measured anisotropy is primarily asthenospheric. The measured fast directions across the major fault zones in the region do not change significantly, clearly not reflecting the change in lithospheric or crustal strain [McCluskey et al., 2000]. The lone exception appears to be a change in the fast direction across a region of concentrated extension in western Anatolia. We observe a change in the orientation of the splitting to be somewhat consistent with the direction of crustal extension.