Attenuation Of LG Waves in the Eastern Tibetan Plateau

摘要

In the past year I have been processing a large amount of regional/ teleseismic data from various broadband seismic stations in eastern Eurasia. Fourier spectra of Pn, Lg and Pg waves were computed for many events and paths to study path attenuations. Among the data collected and processed are Lg spectra collected from the 1991-1992 Tibetan Plateau Passive Experiment. Using these spectra and a standard two-station method that virtually eliminates source and site effects, I obtain a model of Q0 = (126+/- 9) and eta = (0.37 + -0.02) in a frequency range between 0.2 and 3.6 Hz, where Q0 and hare Lg Q at 1 Hz and its power-law frequency dependence, respectively. The estimated Q0 value is among the lowest ever reported for continental areas; it qualitatively supports the observation by McNamara et al. (1996) that Lg cannot be observed inside the plateau beyond about 700 km, a limiting distance that is much shorter than those in the other low Q0 ((caret) 200) regions, such as Iran and the western U.S. Quantitatively, the estimated Q0 value is lower by a factor of 3 than the values of 366 estimated by McNamara et al. (1996), who used data from the same experiment. Since there are several differences in the data processing and inversion procedures used in this and the previous studies, I investigated the effects of these differences on the Q estimates. I conclude that the most probable cause of the discrepancy is in the different inverse methods used. This is so because the previous inversion allowed the source and site terms to be free parameters solved for. Since 20 events, 8 stations and 5 frequency bands were used, the unknown source and station terms should be more than 100. In this study only two free parameters (Q0 and h) are solved for, thus avoiding the instability caused by parameter trade-offs.