HR: 1350h
AN: U23C-1447    [Abstracts]
TI: Back to the Future: Long-Term Seismic Archives Revisited
AU: * Waldhauser, F
EM: felixw@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory, Columbia University, 61 Rt 9W, Palisades, NY 10027, United States
AU: Schaff, D P
EM: dschaff@ldeo.columbia.edu
AF: Lamont-Doherty Earth Observatory, Columbia University, 61 Rt 9W, Palisades, NY 10027, United States
AB: Archives of digital seismic data recorded by seismometer networks around the world have grown tremendously over the last several decades helped by the deployment of seismic stations and their continued operation within the framework of monitoring seismic activity. These archives typically consist of waveforms of seismic events and associated parametric data such as phase arrival time picks and the location of hypocenters. Catalogs of earthquake locations are fundamental data in seismology, and even in the Earth sciences in general. Yet, these locations have notoriously low spatial resolution because of errors in both the picks and the models commonly used to locate events one at a time. This limits their potential to address fundamental questions concerning the physics of earthquakes, the structure and composition of the Earth's interior, and the seismic hazards associated with active faults. We report on the comprehensive use of modern waveform cross-correlation based methodologies for high- resolution earthquake location - as applied to regional and global long-term seismic databases. By simultaneous re-analysis of two decades of the digital seismic archive of Northern California, reducing pick errors via cross-correlation and model errors via double-differencing, we achieve up to three orders of magnitude resolution improvement over existing hypocenter locations. The relocated events image networks of discrete faults at seismogenic depths across various tectonic settings that until now have been hidden in location uncertainties. Similar location improvements are obtained for earthquakes recorded at global networks by re- processing 40 years of parametric data from the ISC and corresponding waveforms archived at IRIS. Since our methods are scaleable and run on inexpensive Beowulf clusters, periodic re-analysis of entire archives may thus become a routine procedure to continuously improve resolution in existing catalogs. We demonstrate the role of seismic archives in obtaining the precise location of new events in real-time. Such information has considerable social and economic impact in the evaluation and mitigation of seismic hazards, for example, and highlights the need for consistent long-term seismic monitoring and archiving of records.
DE: 0520 Data analysis: algorithms and implementation
DE: 3270 Time series analysis (1872, 4277, 4475)
DE: 7219 Seismic monitoring and test-ban treaty verification
DE: 7230 Seismicity and tectonics (1207, 1217, 1240, 1242)
DE: 7290 Computational seismology
SC: Union [U]
MN: 2007 Fall Meeting