Recently, significant temporal variations in local earthquake coda decay rate (coda Q-1) have been observed associated with (1) the long-term migration of seismicity in China, (2) the Round Valley earthquake of 1984, (3) the North Palm Springs earthquake of 1986, (4) the Misasa earthquake of 1983 and (5) an eruption of Mt. St. Helens. Some of these variations have different patterns depending on the frequency of coda waves and the depth of earthquakes. The understanding of these phenomena is urgently needed because of their potential use as a much wanted reliable precursor for intermediate-term earthquake prediction. In order to promote the understanding, we propose to: (1) Continue work on systematic determination of spatial variation in coda Q using the USGS Central California network and the USGS/Caltech/USC southern California network. (2) Perform case studies of temporal variation in coda Q in California, such as the San Jacinto fault zone, where focal depths range from 0 to 20 km (similar to the North Palm Springs area).
