HR: 1340h
AN: S43B-1311    [Abstracts]
TI: Improved depth estimation using integrated small-aperture array and network processing
AU: * Tibuleac, I M
EM: ileana@seismo.unr.edu
AF: University of Nevada, Reno, Laxalt Mining Eng. Bldg., # 174, Reno, NV 89557, United States
AU: Anderson, J G
EM: jga@seismo.unr.edu
AF: University of Nevada, Reno, Laxalt Mining Eng. Bldg., # 174, Reno, NV 89557, United States
AU: Biasi, G P
EM: glenn@seismo.unr.eu
AF: University of Nevada, Reno, Laxalt Mining Eng. Bldg., # 174, Reno, NV 89557, United States
AU: Seggern, D v
EM: vonseg@seismo.unr.edu
AF: University of Nevada, Reno, Laxalt Mining Eng. Bldg., # 174, Reno, NV 89557, United States
AB: We are testing a new approach to estimate the depth of earthquakes reliably and rapidly. Accurate estimates of earthquake depth are very important for nuclear monitoring, as well as for hazard assessment. The use of teleseismic recordings to determine this parameter is particularly important for sparsely monitored regions. Depth can be estimated by measuring the time separation between the primary arrival (P) and associated depth phases (pP, sP). Depth phases, however, are difficult to identify and thus they are used infrequently by institutes monitoring global seismicity such as USGS, IDC, and USNDC. As an example, less than 15% of the events located by USGS have associated depth phases. Using small or medium-aperture (< 25 km) array processing integrated with network processing, we can recognize secondary phases that are not visible on individual stations. Our approach works because, for stations in the vicinity of each array, the arrival-time difference between primary and secondary phases varies slowly with increasing epicentral distance. We estimate P-arrival parameters at small-aperture arrays using crosscorrelation in the time domain. From these parameters, we derive a time-variable set of weights. We beam the weighted envelopes of autocorrelated waveforms from nearby network stations to form the integrated network beam (INB). The resulting time series should have a central pick (P) and symmetrical side picks (pP), similar to autocorrelation of waveforms with ghost arrivals in exploration geophysics. To estimate the pP arrival time, we apply an F-detector to the INB and its components. Using a dataset of well-located events recorded at calibrated arrays and the surrounding networks, we determine whether this depth-phase identification methodology is widely applicable at local, regional and teleseismic distances. We also determine whether the methodology will work for real-time processing, and whether it will provide reliable depth estimates for earthquakes as shallow as 10 km.
DE: 7200 SEISMOLOGY
DE: 7203 Body waves
DE: 7219 Seismic monitoring and test-ban treaty verification
DE: 7230 Seismicity and tectonics (1207, 1217, 1240, 1242)
DE: 7294 Seismic instruments and networks (0935, 3025)
SC: Seismology [S]
MN: 2007 Fall Meeting