Improving Magnitude Detection Thresholds Using Multi-Station Multi-Event, and Multi-Phase Methods

Abstract

Public reporting burden for this collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing this collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information, including suggestions for reducing this burden to Department of Defense, Washington Headquarters Services, Directorate for Information Operations and Reports (0704-0188), 1215 Jefferson Davis Highway, Suite 1204, Arlington, VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to any penalty for failing to comply with a collection of information if it does not display a currently valid OMB control number. PLEASE DO NOT RETURN YOUR FORM TO THE ABOVE ADDRESS. (31-07-2008) REPORT DATE SPONSOR/MONITOR'S ACRONYM(S)AFRL/RVBYE SPONSOR/MONITOR'S REPORT NUMBER(S) AFRL-RV-HA-TR-2008-1121 DISTRIBUTION / AVAILABILITY STATEMENTApproved for Public Release; Distribution Unlimited. SUPPLEMENTARY NOTES ABSTRACTWe report on the results of both a semi-empirical analysis and a case study in Xiuyan, China, that demonstrated that a correlation detector can lower magnitude detection thresholds by over one full unit for similar events as compared to a standard STA/LTA detector. 90 out of 90 events (100%) were detected by cross correlation for the case study whereas a procedure like the pIDC employs detected only 11%. We found that less than perfect matches still provided useful detections due to location, focal mechanism, and magnitude differences. Events with magnitude differences as large as 2.3 and 3.3 were shown to produce detection spikes on the correlation traces. Further work applied the correlation techniques on a larger scale to 5,000 events at Parkfield, California, and 19,000 events in and near China. We are attempting to see how broadly applicable correlation methods can be applied to different tectonic settings and for what percentage of the seismicity. 111 million correlations were performed on Lg-waves for the events in China at 363 stations. Final results indicate two thirds of the 19,000 events can be detected by cross correlation using this relatively sparse regional network. The full magnitude unit reduction in detection threshold is confirmed on a large scale in China and Parkfield.

Semiempirical Statistics of Correlation-Detector Performance

“…Importantly, this capability is achieved with acceptably low false alarm rates as demonstrated by semiempirical synthetic runs. Similar results of this order of magnitude improvement in detection thresholds have been demonstrated by case studies using real seismic data (Gibbons and Ringdal, 2006;Schaff and Waldhauser, 2006;Gibbons et al, 2007). To estimate false alarm rates, however, synthetics must be used, or complete catalogs to lower magnitudes for denser networks on the case studies or statistical assumptions employed (Wiechecki-Vergara et al, 2001).…”

“…Detection capability will decrease as the underlying waveform similarity decreases due to increasing interevent separation distances, mechanisms differences, and source-time function complexities. Other work has shown, however, that even semisimilar events with less-than-perfect waveform matches still provide useful detections (Schaff and Waldhauser, 2006).…”

“…Subsequent work extended correlators to subspace detectors to allow for variations in waveforms for a given source region (Harris, 1997). Case studies are beginning to show more promise for correlation detectors in reducing magnitude thresholds for practical applications (Gibbons and Ringdal, 2004, 2006Schaff and Waldhauser, 2006;Gibbons et al, 2007). The case studies, however, are not able to estimate false alarm rates unless a denser complete local catalog is available.…”

Broad‐scale applicability of correlation detectors to China seismicity

“…It is assumed that magnitude completeness of the catalog is the reason a larger improvement is not seen for closer station distances. Therefore the order of magnitude reduction in detection threshold observed for smaller studies [ Gibbons and Ringdal , 2006; Schaff and Waldhauser , 2006; Schaff , 2008] comparing correlation detectors for similar events with STA/LTA type algorithms appears to hold on a broad scale, across diverse tectonic regions, for tens of thousands of events.…”

“…Semi‐empirical synthetic runs showed this improvement is obtained with acceptable false alarm rates of about one per day [ Schaff , 2008]. A case study of 90 events in the 1999 Xiuyan earthquake sequence in China further confirmed a full magnitude unit reduction in detection threshold is possible for a different dataset and region on five regional three component stations 500 to 1500 km away [ Schaff and Waldhauser , 2006]. A key question is how well will correlation detectors work for seismicity in general on a large scale and not just specific case examples, since it requires a priori information on the shape of the master waveform template?…”