Foreshock search over a long duration using a method of setting appropriate criteria
Abstract
Recently, small foreshocks have been detected using cross-correlation techniques (e.g., Bouchon et al., 2011) in which the foreshocks are identified when the cross-correlation coefficient (CC) exceeded a certain threshold. For some shallow intraplate earthquakes, foreshocks whose hypocenters were estimated to be adjacent to the main shock hypocenter were detected from several tens of minutes before the main shock occurrence (Doi and Kawakata, 2012; 2013). At least two problems remain in the cross-correlation techniques employed. First, previous studies on foreshocks used data whose durations are at most a month (Kato et al., 2013); this is insufficient to check if such events occurred only before the main shock occurrence or not. Second, CC is used for detection criteria without considering validity of the threshold. In this study, we search for foreshocks of an M 5.4 earthquake in central Nagano prefecture in Japan on June 30, 2011 with a vertical-component waveform at N.MWDH (Hi-net) station due to one of the cataloged foreshocks (M 1) as a template to calculate CC. We calculate CC between the template and continuous waveforms of the same component at the same station for two years before the main shock occurrence, and we try to overcome the problems mentioned above. We find that histogram of CC is well modeled with the normal distribution, which is similar to previous studies on tremors (e.g., Ohta and Ide, 2008). According to the model, the expected number of misdetection is less than 1 when CC > 0.63. Therefore, we regard that the waveform is due to a foreshock when CC > 0.63. As a result, foreshocks are detected only within thirteen hours immediately before the main shock occurrence for the two years. By setting an appropriate threshold, we conclude that foreshocks just before the main shock occurrence are not stationary events. Acknowledgments: We use continuous waveform records of NIED high sensitivity seismograph network in Japan (Hi-net) and the JMA unified hypocenter catalogs. This work is supported by MEXT of Japan, under its Earthquake and Volcano Hazards Observation and Research Program.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2016
- Bibcode:
- 2016AGUFMMR41B2702T
- Keywords:
-
- 3902 Creep and deformation;
- MINERAL PHYSICSDE: 7230 Seismicity and tectonics;
- SEISMOLOGYDE: 8118 Dynamics and mechanics of faulting;
- TECTONOPHYSICSDE: 8163 Rheology and friction of fault zones;
- TECTONOPHYSICS