WHAT CAN WE LEARN FROM AFTERSHOCKS

From the spatial distribution of aftershocks with respect to the regio ns of fault slip during the mainshock, information can be gained about the heterogeneous structure within the source volume and about the fr ictional properties of surrounding faults. Focal mechanisms of aftersh ocks reveal how stress is redistributed by large earthquakes. While an alog data have mainly been used for these studies in the past, a new d egree of data quality is obtained with currently available digital rec ording systems. In addition to the increased accuracy for the determin ation of purely kinematic data, waveform information can be utilized w ith digital data. With state-of-the-art signal processing techniques, the contributions of source, path, and site effects on the observed se ismic signals can be studied. Provided these effects can be separated, aftershock signals will potentially help us to learn about the proper ties of fault regions in unprecedented detail. The degree of resolutio n which can be achieved, however, will strongly depend on the number o f stations and the geometry of the network employed. Depending on the special scientific problems to be addressed, optimum station geometrie s may vary. Modern methods in optimization theory, such as simulated a nnealing, have been successfully used to find optimum station distribu tions for aftershock monitoring. An additional aspect which has to be addressed is the problem of managing and processing aftershock data. S ince high-quality digital data come at the price of huge data volumes, new strategies and concepts for data handling and signal processing h ave to be developed.