SCIENCE CHINA Earth Sciences © Science China Press and Springer-Verlag Berlin Heidelberg 2010 earth.scichina.com www.springerlink.com * (email: zhangyong@cea-igp.ac.cn); † Corresponding author (email: xuls@cea-igp.ac.cn); †† Equal contributor (email: chenyt@cea-igp.ac.cn)  NEWS FOCUS  September 2010 Vol.53 No.9: 1249–1251 doi: 10.1007/s11430-010-4045-5 Source process of the 2010 Yushu, Qinghai, earthquake ZHANG Yong * , XU LiSheng † & CHEN Yun-Tai †† Institute of Geophysics, China Earthquake Administration, Beijing 100081, China Received May 20, 2010; accepted June 24, 2010; published online July 30, 2010 Citation: Zhang Y, Xu L S, Chen Y T. Source process of the 2010 Yushu, Qinghai, earthquake. Sci China Earth Sci, 2010, 53: 1249 – 1251, doi: 10.1007/s11430-010-4045-5 On April 14th, 2010, at 07:49 am (Beijing Time) (April 13th, 2010, 23:49 UTC), an earthquake of M S 7.1 occurred in Yushu county, Qinghai Province, China. According to the latest report from China Earthquake Network Center (CENC), the epicenter of the Yushu earthquake was at (33.2°N, 96.6°E), 44 km northwestern from Yushu city, and the focal depth was 14 km. By May 30th, 2010, the Yushu earthquake caused about 3000 people killed or missing, over 10000 people injured, and a large number of houses and buildings collapsed. The Yushu earthquake occurred on Ganzi-Yushu Fault, a southeast-striking, left-lateral strike-slipping fault, which lies on southern boundary of the Bayan Har Block (Song- pan-Ganzi Block). Historically there has been high seismi- city on the fault [1–4]. The Yushu earthquake was the largest event on the northwestern segment of the fault in recent 100 years. For quick response to the earthquake, we obtained the source rupture process by inverting the seismic recordings and had it released about 2.5 hours after its occurrence. Af- terwards when more and more data became available, twice did we update the results about 5 hours and 2 days after the earthquake occurrence, respectively (http:// www.csi.ac.cn). In order to better understand the source rupture process of the Yushu earthquake, after the quick response activity, we retrieved the apparent source time functions (ASTFs) of the mainshock from Love wave data using the largest after- shock as an empirical Green’s function (EGF) event, and estimated the rupture velocity of the involved sub-events by further analyzing the ASTFs. Once again we inverted the carefully selected P waveform data to improve the results of the source rupture process. 1 The Love-ASTFs analysis According to a report from Qinghai Earthquake Administra- tion, an aftershock of M S 6.3 occurred about 1.5 hours after the mainshock, with epicenter at (33.23°N, 96.58°E) and focal depth of 10 km. The aftershock had almost the identi- cal hypocentral location and focal mechanism with the mainshock (Figure 1) [6]. Therefore this aftershock was used as the EGF event to retrieve the ASTFs of the main- shock. Then we analyzed the retrieved ASTFs for rupture velocity of the mainshock. In this study, we only used the Love waveform data owing to the poor signal-to-noise ratio of most of the P and/or S phases of the aftershock. We selected 24 stations of Love wave data (Figure 2(a)), and obtained the ASTFs (Love-ASTFs) using the Projected Landweber Deconvolution (PLD) method (Figure 2(b)) [7, 8]. As Figure 2(b) shows, the Love-ASTFs systematically vary with azimuths of the stations, with those at the stations with similar azimuths having similar appearances. The du- ration times of the Love-ASTFs at azimuths of 100°–150° are about 10 s while those at around 300° are about 30 s. It suggests that the Yushu earthquake is a unilateral rupture event and overall its rupture propagates southeastwards (azimuth 100°–150°). Two sub-events are clearly visible on the Love-ASTFs at stations of 230°–330° azimuths, but they are invisible on the others. Using the times when the peaks of the two sub-events show up on the Love-ASTFs, we calculated the spatio-temporal parameters for the two sub-events by con- structing and solving the following objective function: s cos( ) ( , ) min, i i i R RT T t V φ φ − Δ = − − = ∑ (1) where i is used to number the stations; R and T are the spa-