Geophysical Modeling in Eurasia: 2D Crustal P and LG Propagation; Upper- Mantle Shear Wave Propagation and Anisotropy; and 3D, Joint, Simultaneous Inversions

摘要

New tomographic models have been developed to obtain improvements to travel-time predictions for a variety of crustal and upper-mantle seismic phases, and we approach the problem with both independent datasets and joint, simultaneous inversion of diverse geophysical parameters. New Pg and Lg velocity models in Eurasia improve estimates of crustal travel times across the region. Inversion of Pg and Lg travel times improves RMS by tilde 30% compared to iasp91. Site terms for both phases show similar patterns and likely reflect thicknesses and velocities of sediments above the upper crust. Event terms are tilde 1 s, suggesting 5-10-km-average depth errors. Velocities are consistent with typical upper-crustal velocities set forth in the Department of Energy (DOE) Unified Model but suggest more geographic variability. Upper-mantle shear wave (Sn) velocity tomography, including Sn gradient within the upper mantle, has been performed both with and without anisotropy. We use events that are GT25 and better, with depths of less than 50 km and path lengths between 400 and 1700 km. The 14,645 events recorded at 2,399 stations yielded 144,933 raypaths. Most arrivals were obtained from the International Seismological Centre (ISC), Annual Bulletin of Chinese Earthquakes (ABCE), and Earthquake Data Reports (EDR) catalogs. Crustal leg corrections assume 45-km average crustal thickness, 3.6-km/s velocity, and 52-degree incidence angles beneath station and source. The RMS reduction after inversion is tilde 56%. Sn velocities mirror those found for Pn; however, the Sn gradient appears to be weaker. Solving for anisotropy removes some high-frequency oscillations; the anisotropy appears to follow that found for Pn.