[1] We investigate regional variations in the Lg-wave quality factor (Q) in Northeast China and its vicinity with a tomographic method. Digital seismic data recorded at 20 broadband stations from 125 regional events are used to extract Lg-wave spectra. Tomographic inversions are independently conducted at 58 discrete frequencies distributed log evenly between 0.05 and 10.0 Hz. We simultaneously invert for the Lg-wave Q distribution and source spectra at individual frequencies without using any a priori assumption about the frequency dependence of the Q model and source function. The best spatial resolution is approximately 1°× 1°in well-covered areas for frequencies between 0.4 and 2.0 Hz. The Lg Q shows significant regional variations and an apparent relationship with regional geology. We use a statistical method to investigate the regional variations of Lg Q and their frequency dependence. The average Q 0 (1 Hz Lg Q) in the entire investigated region is 414. Sedimentary basins are usually characterized by lower average Q 0 values (from 155 to 391), while volcanic mountain areas have relatively high average Q 0 values (from 630 to 675). Lg Q generally increases with increasing frequency. However, the frequency dependence has complex nonlinear features on a double-logarithmic scale, indicating that the commonly adopted power-law relationship may be oversimplified in a broad frequency band. The frequency dependence varies in different geological areas, with larger variations seen at lower frequencies.

We investigate the regional seismic signature of the 9 October 2006 North Korean nuclear test. Broadband regional data for the nuclear test and a group of earthquakes close to the test site were obtained between December 2000 and November 2006. Epicentral distances from the stations to the test site are between 371 and 1153 km. We first use these regional events to calibrate the Lg-wave magnitude in the network. Then the network is used to calculate m b Lg 3:93 for the North Korean nuclear explosion. Using a modified fully coupled magnitude-yield relation, the yield of the North Korean nuclear test is estimated to be 0.48 kt. Because of large uncertainties in the source depth, the estimate is preliminary. The P=S-type spectral ratios Pg=Lg, Pn=Lg, and Pn=Sn are calculated for the nuclear explosion and a group of earthquakes close to the test site. At frequencies above 2 Hz, the network-averaged P=S spectral ratios clearly separate the 9 October 2006 explosion from the regional earthquakes. Our result indicates that a single-blast explosion in the North Korea region shows different seismic characteristics from an earthquake. Any well-coupled single-blast explosion detonated in this region with yield similar to that for the North Korean nuclear test has a large probability of being identified by a regional seismic network such as the one adopted in this study.

[1] Ambient noise tomography has been becoming an important tool to image the shallow lithospheric structure of the Earth. Using 2 months of ambient noise data from 20 stations of the Himalayan Nepal Tibet Seismic Experiment, we investigate the upper and middle crustal structure in the central Himalaya and southern Tibet. About 120 interstation Rayleigh wave empirical Green's functions with sufficient signal-tonoise ratio are obtained and used for group velocity dispersion analysis in the period range 6-25 s using frequency-time analysis technique. The obtained dispersion data are then used to construct 2-D group velocity maps. At the short periods from 9 to 15 s, the distribution of Rayleigh wave velocities delineates several distinct low-and high-velocity zones separated mainly by geological boundaries. The high group velocity zone is located mainly around regions with plutonic rocks, and the low group velocity zone is located around regions with sedimentary or metasedimentary rocks. Finally, we invert for the shear velocity structure in the upper and middle crust along a N-S trending cross section at the longitude 86.5°E. We observe a clear low-velocity layer in the middle crust (about 10-25 km depth) distributed on both sides of the Indus Yarlung Suture zone. The existence of this midcrustal low-velocity zone suggests a mechanically weaker middle crust beneath the central Himalaya and southern Tibet, which might decouple the upper crustal deformation from that of the lower crust in the Tibetan-Himalayan orogenic processes.Components: 6760 words, 6 figures.