SUMMARY The interaction of the Arabian plate with the Eurasian plate has played a major role in building the young mountain belts along the Zagros–Bitlis continent–continent collision zone. Arabia's northward motion is considered to be the primary driving force behind the present‐day westerly escape of the Anatolian plate along the North and East Anatolian fault zones as well as the formation of the Turkish and the Iranian plateaux. In this study we mapped Pn‐wave velocity and anisotropy structures at the junction of the Arabian, Eurasian and African plates in order to elucidate the upper‐mantle dynamics in this region. Pn is a wave that propagates within the mantle lid of the lithosphere and is often used to infer the rheology and fabric of the mantle lithosphere. Applying strict selection criteria, we used arrival times of 166 000 Pn phases to invert for velocity and anisotropy in the region. Using a least‐squares tomographic code, these data were analysed to solve simultaneously for both velocity and azimuthal anisotropy in the mantle lithosphere. We found that most of the continental regions in our study area are underlain by low Pn velocity structures. Broad‐scale (∼500 km) zones of low (<8 km s−1) Pn velocity anomalies underlie the Anatolian plate, the Anatolian plateau, the Caucasus region, northwestern Iran and northwestern Arabia, and smaller scale (∼200 km), very low (<7.8 km s−1) Pn velocity zones underlie southern Syria, the Lesser Caucasus, the Isparta Angle, central Turkey and the northern Aegean Sea. The broad‐scale low‐velocity regions are interpreted to be hot and unstable mantle lid zones, whereas very low Pn velocity zones are interpreted to be regions of no mantle lid. The low and very low Pn velocity zones in eastern Turkey, northwestern Iran and the Caucasus region may be associated with the latest stage of intense volcanism that has been active since the Late Miocene. The low Pn velocity zones beneath the Anatolian plate, eastern Turkey and northwestern Iran may in part be a result of the subducted Tethyan oceanic lithosphere beneath Eurasia. We also found a major low‐velocity zone beneath northwestern Arabia and the Dead Sea fault system. We interpret this anomaly to be a possible extension of the hot and anomalous upper mantle of the Red Sea and East Africa rift system. High Pn velocities (8.1–8.4 km s−1) are observed to underlie the Mediterranean Sea, the Black Sea, the Caspian Sea, and the central and eastern Arabian plate. Observed Pn anisotropy showed a higher degree of lateral variation than did the Pn velocity structure. Although the Pn anisotropy varies even in a given tectonic region, in eastern Anatolia very low Pn velocity and Pn anisotropy structures appear to be coherent.

We use Pn phase travel time residuals to invert for mantle lid velocity and anisotropy beneath northern Arabia‐eastern Anatolia continent‐continent collision zone. The primary phase data were obtained from the temporary 29‐station broadband PASSCAL array of the Eastern Turkey Seismic Experiment. These data were supplemented by phase data from available stations of the Turkish National Seismic Network, the Syrian National Seismic Network, the Iranian Long Period Array, and other stations around the southern Caspian Sea. In addition, we used carefully selected catalog data from the International Seismological Centre and the National Earthquake Information Center bulletins. Our results show that low (<8 km/s) to very low (<7.8 km/s) Pn velocity zones underlie the Anatolian plateau, the Caucasus, and northwestern Iran. Such low velocities are used to infer the presence of partially molten to absent mantle lid beneath these regions. In contrast, we observed a high Pn velocity zone beneath northern Arabia directly south of the Bitlis‐Zagros suture indicating the presence of a stable Arabian mantle lid. This sharp velocity contrast across the suture zone suggests that Arabia is not underthrusting beneath the Anatolian plateau and that the surface suture extends down to the uppermost mantle. Pn anisotropy orientations within a single plate (e.g. Anatolia plate) show a higher degree of lateral variation compared to Pn velocity. Areas of coherent Pn anisotropy orientations are observed to continue across major fault zones such as the EAF zone.

S U M M A R YContinuous recordings of 17 broadband and short-period digital seismic stations from a newly established seismological network in Saudi Arabia, along with digital recordings from the broadband stations of the GSN, MEDNET, GEOFON, a temporary array in Saudi Arabia, and temporary short period stations in Oman, were analysed to study the lithospheric structure of the Arabian Plate and surrounding regions. The Arabian Plate is surrounded by a variety of types of plate boundaries: continental collision (Zagros Belt and Bitlis Suture), continental transform (Dead Sea fault system), young seafloor spreading (Red Sea and the Gulf of Aden) and oceanic transform (Owen fracture zone). Also, there are many intraplate Cenozoic processes such as volcanic eruptions, faulting and folding that are taking place.We used this massive waveform database of more than 6200 regional seismograms to map zones of blockage, inefficient and efficient propagation of the Lg and Sn phases in the Middle East and East Africa. We observed Lg blockage across the Bitlis Suture and the Zagros fold and thrust belt, corresponding to the boundary between the Arabian and Eurasian plates. This is probably due to a major lateral change in the Lg crustal waveguide. We also observed inefficient Lg propagation along the Oman mountains. Blockage and inefficient Sn propagation is observed along and for a considerable distance to the east of the Dead Sea fault system and in the northern portion of the Arabian Plate (south of the Bitlis Suture). These mapped zones of high Sn attenuation, moreover, closely coincide with extensive Neogene and Quaternary volcanic activity. We have also carefully mapped the boundaries of the Sn blockage within the Turkish and Iranian plateaus. Furthermore, we observed Sn blockage across the Owen fracture zone and across some segments of the Red Sea. These regions of high Sn attenuation most probably have anomalously hot and possibly thin lithospheric mantle (i.e. mantle lid). A surprising result is the efficient propagation of Sn across a segment of the Red Sea, an indication that active seafloor spreading is not continuous along the axis of the Red Sea. We also investigated the attenuation of Pn phase (Q Pn ) for 1-2 Hz along the Red Sea, the Dead Sea fault system, within the Arabian Shield and in the Arabian Platform. Consistent with the Sn attenuation, we observed low Q Pn values of 22 and 15 along the western coast of the Arabian Plate and along the Dead Sea fault system, respectively, for a frequency of 1.5 Hz. Higher Q Pn values of the order of 400 were observed within the Arabian Shield and Platform for the same frequency. Our results based on Sn and Pn observations along the western and northern portions of the Arabian Plate imply the presence of a major anomalously hot and thinned lithosphere in these regions that may be caused by the extensive upper mantle anomaly that appears to span most of East Africa and western Arabia.

A 29‐station temporary broadband PASSCAL network was operated from late October 1999 to August 2001 in eastern Turkey in order to decipher the geodynamics of one of the youngest continent‐continent collision zones in the world. This paper focuses on the hypocentral distribution of local earthquakes located during the operation of the network and provides new insights into the active faulting in the Anatolian plateau. A total of 1165 earthquakes were located and classified into four different categories based on the reliability of the locations as established by the data coverage. The accuracy of the locations ranked in the best two categories is estimated to be less than approximately 5 km. The results show that seismic activity in Eastern Turkey is higher than previously documented and there were no subcrustal earthquakes beneath the Arabian‐Eurasian collision zone or beneath the Anatolian plateau during our deployment. This result suggests no or very little underthrusting of the Arabian plate beneath Eurasia. Our results also suggest that the North Anatolian Fault zone extends farther toward the southeast, well beyond the Karliova triple junction, and that a number of unmapped active, seismogenic faults exist in the region. We also observed a possible difference in the seismogenic thickness of the East Anatolian fault zone (EAFZ) and the North Anatolian fault zone (NAFZ).

Gravity data and P-wave teleseismic traveltime residuals from 29 temporary broad-band stations spread over the northern margin of the Gulf of Aden (Dhofar region, Oman) were used to image lithospheric structure. We apply a linear relationship between density and velocity to provide consistent density and velocity models from mid-crust down to about 250 km depth. The accuracy of the resulting models is investigated through a series of synthetic tests. The analysis of our resulting models shows: (1) crustal heterogeneities that match the main geological features at the surface; (2) the gravity edge effect and disparity in anomaly depth locations for layers at 20 and 50 km; (3) two low-velocity anomalies along the continuation of Socotra-Hadbeen and Alula-Fartak fracture zones between 60 and 200 km depth; and (4) evidence for partial melting (3-6 per cent) within these two negative anomalies. We discuss the presence of partial melting in terms of interaction between the Sheba ridge melts and its along-axis segmentation