A variable resolution surface wave dispersion study of Eurasia, North Africa, and surrounding regions

This paper presents the results of a large‐scale study of surface wave dispersion performed across Eurasia and North Africa. Improvements were made to previous surface wave work by enlarging the study region, increasing path density, improving spatial resolution, and expanding the period range. This study expands the coverage area northward and eastward relative to a previous dispersion analysis, which covered only North Africa and the Middle East. We have significantly increased the number of seismograms examined and group velocity measurements made. We have now made good quality dispersion measurements for about 30,000 Rayleigh wave and 20,000 Love wave paths and have incorporated measurements from several other researchers into the study. We have improved the inversion from the previous study by adopting a variable smoothness with the conjugate gradient method for the group velocity tomography. This technique produces higher‐resolution models where the concentration of data allows it without producing artifacts. The current results include both Love and Rayleigh wave inversions across the region for periods from 7 to 100 s on a 1° × 1° grid and at resolutions approaching 1° under some conditions. Short‐period group velocities are sensitive to slow velocities associated with large sedimentary features such as the Caspian Sea, West Siberian Platform, Mediterranean Sea, Bay of Bengal, Tarim Basin, and Persian Gulf. Intermediate periods are sensitive to differences in crustal thickness, such as those between oceanic and continental crust or along orogenic zones and continental plateaus. At longer periods, fast velocities are consistently found beneath cratons, while slow upper mantle velocities occur along rift systems, subduction zones, and collision zones such as the Tethys Belt. We have compared the group velocities at various periods with features such as sediment thickness, topographic height, crustal thickness, proximity to plate boundaries, lithospheric age, and lithospheric thickness, and we find significant correlations. We have developed an empirical relationship between long‐period group velocity and lithospheric thickness that works well for both oceanic and continental lithosphere. We do not find any similar correlation between the longest‐period surface waves and hot spots.