17:15 〜 18:45
[SCG40-P07] Temporal slip slip increases during short-term SSE
キーワード：slow earthquakes、temporal slip speed changes、slow slips with durations of one-three hours
Slow slips with durations between several tens of seconds of VLFEs (very low frequency earthquakes) and a few days of S-SSEs (short-term slow slip events) are difficult to observe with geodetic data. We have been searching for one hour duration slow-slip events with the data from borehole-strainmeters installed by JMA and AIST and laser strainmeters. We investigated continuous strain data from 2016 to 2022, and detected strain steps. Strain steps were found by calculating correlation coefficient (CC) between stacked strain data and ramp functions of 1-2 hour rise time. Stacking of strain data were calculated with assuming source locations, and enhanced S/N of strain. There are a number of peaks in running CC. We picked up high CC peaks for further analysis.
Source locations were estimated for strain steps with the peak CC > 0.88. The value of 0.88 was selected because the number of positive (consistent with plate boundary slip) CC peaks exceeded that of negative (reverse direction of plate boundary slip) CC peaks. Source locations were estimated with distribution of coefficient of determination (DC) between observed strain steps of individual sensors and calculated ones. Some of strain steps showed concentrated patterns in distributions of DC. Because random noise in strain sensors should not make a concentrated distribution of DC, the concentrated pattern indicates that the corresponding strain step were due to plate boundary slips.
Then the detected strain steps were compared with accumulated numbers of tectonic tremors, and their distributions of DC were compared with tremor hypocenters. Only five strain steps shared their sources with tectonic tremors spaciotemporally from 2016 to 2022. They occurred during S-SSEs, and they were temporal slip speed increases in the S-SSEs. There were more strain steps which showed high correlation with accumulated number of tremors. However, we could not get coincidence of source locations for many of those. The detected strain steps seems to be related to successive occurrences of S-SSEs with migration.
Source locations were estimated for strain steps with the peak CC > 0.88. The value of 0.88 was selected because the number of positive (consistent with plate boundary slip) CC peaks exceeded that of negative (reverse direction of plate boundary slip) CC peaks. Source locations were estimated with distribution of coefficient of determination (DC) between observed strain steps of individual sensors and calculated ones. Some of strain steps showed concentrated patterns in distributions of DC. Because random noise in strain sensors should not make a concentrated distribution of DC, the concentrated pattern indicates that the corresponding strain step were due to plate boundary slips.
Then the detected strain steps were compared with accumulated numbers of tectonic tremors, and their distributions of DC were compared with tremor hypocenters. Only five strain steps shared their sources with tectonic tremors spaciotemporally from 2016 to 2022. They occurred during S-SSEs, and they were temporal slip speed increases in the S-SSEs. There were more strain steps which showed high correlation with accumulated number of tremors. However, we could not get coincidence of source locations for many of those. The detected strain steps seems to be related to successive occurrences of S-SSEs with migration.