Thursday, 18 December 2014
Thomas M Hearn, New Mexico State University, Las Cruces, NM, United States
Modern data from the China Bulletin and temporary network deployments has been used to update amplitude tomography using ML and MS seismic amplitudes. This work builds on the results of Hearn et al., 2008. ML attenuation estimates are much better resolved due to the inclusion of subnet data. We find that the trade-off between geometrical spreading and attenuation estimates are well constrained; however, both of these parameters have significant trade-off with the frequency dependence of attenuation. Maps of attenuation using the ML amplitudes are similar to those of Lg attenuation found by other authors suggesting that ML attenuation estimates form a suitable proxy for Lg attenuation estimates. We are now able to associate high attenuation directly with the Longmen Shan and the Qilian Shan mountains and also, where resolved, with the Kunlun Shan, Altyn Tag, and Tian Shan mountains. Grabens around the Ordos Platform also show high attenuation. Basins, however, do not in general show high attenuation. The main exception to this is the Bohai Basin. We conclude that the ML waveforms, like the Lg waveforms, interrogate the entire crustal column and are most sensitive to tectonically active structures and rapid changes in crustal structure. Data from MS data do not include subnet readings and do not have the resolution that was obtained with the ML data. Nonetheless, features are similar with the exception that basins appear more highly attenuative.