Detection of long-term and short-term slow slip events by the network filter using GNSS data in Shikoku, Japan
Abstract
Three are two types of slow slip events (SSEs) on the plate interface between the subducting Philippine Sea plate and overriding Amur plate along the Nankai trough subduction zone in Japan. One has duration over several months and the other one has duration of several days. The former is called long-term SSE (L-SSE) and the latter is called short-term SSE (S-SSE). S-SSEs are associated with low-frequency earthquakes and ground displacements of several mm. The detection of L- and S-SSEs by the GNSS network in Japan is successfully reported by Monte Carlo Mixed Kalman filtering (MCMKF) (Sakaue and Fukuda 2019). In our study, we applied the network filter which was originally proposed by Segall and Matthews (1992) to the data in Shikoku Island, Japan, from 2019, when L- SSE was occurring in the Bungo channel area, Japan (Ozawa et al., 2019). We used position time series at 120 GNSS sites operated by the Geospatial Information Authority of Japan for the period between 2019 and 2020. Annual and semi-annual components estimated for the period between 2017 and 2020 and a linear trend estimated for the period between 2017 and 2018 are removed from the position time series. The results of the network filtering show major short-term SSEs in March, August, September, November 2019, February, June 2020 associated with low-frequency earthquakes together with L-SSE in the Bungo channel, Japan, although there were several false detections of S-SSEs. The estimated moment magnitude of S-SSEs ranges from Mw5.9 to Mw6.1. Some of the detected S-SSEs show spatial propagation, although we cannot find regular patterns. The major detected short-term SSEs are consistent with the detected events from the strain meters by the Meteorological Agency of Japan. Although most of the detected SSEs are associated with low-frequency earthquakes, the slip area in some cases centered on an area where few low-frequency earthquakes occur, similar to some cases of ordinary earthquakes in which the major slip area shows few aftershocks. Since the ordinary network filter is computationally less burden than MCMKF, we apply it to the GNSS data and monitor SSEs in Shikoku Island, Japan.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2020
- Bibcode:
- 2020AGUFMT003.0009O
- Keywords:
-
- 7215 Earthquake source observations;
- SEISMOLOGY;
- 7230 Seismicity and tectonics;
- SEISMOLOGY;
- 8118 Dynamics and mechanics of faulting;
- TECTONOPHYSICS;
- 8164 Stresses: crust and lithosphere;
- TECTONOPHYSICS