9706152 Okal This research is to extend a highly successful project to recover seismic data from analog recordings of deep earthquakes prior to the digital Global Seismic Network installation. Prior work concentrated on earthquakes in the depth range of 300-700 km prior to 1976. This effort will concentrate on deep events between 125-300 km in depth. Thus, the Harvard University focal mechanism catalogue will be extended back in time to increase significantly the statistical sampling density in time and space. Focal mechanism inversions will be done on the back-projected data using the same Harvard algorithm. It is envisioned that about 200 new centroid moment tensor solutions will be obtained; this would double the size of the data set of reliable, quantitative estimates of earthquake sources in the relevant windows of depth and size. These new results will provide powerful new insight into problems such as the origin of faulting in intermediate depth earthquakes in relation to mineralogical transformations, their distribution in space and time in relation to local geological features, and improve our understanding of their recurrence patterns, an especially important topic, since several intermediate earthquakes have resulted in severe loss of life. This research is a component of the National Earthquake Hazard Reduction Program. ***
