Explore the vast range of titles available.

Seismic phase picking is a fundamental task in seismology that is crucial for event detection and earthquake cataloging; however, manual analysis is impractical given the scale of modern seismic networks. We present SegPhase, a novel seismic arrival time picking model designed to efficiently process large-scale seismic data recorded by dense seismic networks in Japan. In contrast to conventional convolution-based models, SegPhase employs a hierarchical vision transformer structure that utilizes multi-head self-attention to dynamically focus on important waveform features, such as P- and S-wave onsets, noise, and coda waves. Compared to PhaseNet, the most widely used deep learning model, SegPhase improved arrival time match rates by ~ 11% and detected ~ 15% more events in continuous waveform tests, particularly enhancing the detection of small-magnitude events. Benchmark evaluations demonstrated that SegPhase achieved high classification performance in identifying P- and S-waves. We also examined the threshold of the output probability values when applying SegPhase to continuous waveforms for which the optimal threshold was unknown. By lowering the threshold to 0.1, we observed an increase in the number of detected events without noticeable changes in the hypocenter location error and observed–calculated discrepancies. This was achieved by more effectively utilizing high-probability picks, which further improved phase association. Based on these results, we recommend a threshold of 0.1 to enhance event detection while maintaining accurate arrival times. Our findings demonstrate that SegPhase enables robust arrival picking across diverse datasets and supports high-resolution seismic monitoring. Graphical Abstract

To clarify the relationship between earthquake occurrence and fluid, we analyzed data from a dense aftershock-observation network with 69 high-gain short-period seismographs installed immediately after the mainshock occurrence (October 21) in the aftershock area of the 2016 Central Tottori Prefecture earthquake. We determined the hypocenters and focal mechanisms of the aftershocks very precisely in the period from October 22 to December 15. We then investigated the temporal changes in the spatial distributions of hypocenters and T-axis azimuths of focal mechanisms. The distributions of aftershock hypocenters and T-axis azimuths are basically temporally stable, except those in limited portions in the shallow layer near the western edge of the aftershock area, where rapid decrease of aftershocks with T-axis azimuths of WSW to west was observed. If fluid rises from the lower crust due to fault rupture, the locations of aftershocks and focal mechanisms may change over time, especially in the deepest part of the aftershock region. However, the temporal change in these parameters was not apparent at depth. These observations suggest that the aftershock activity of the Central Tottori Prefecture earthquake was controlled mainly by stress concentration rather than strength reduction due to high fluid pressure.

We conducted high-resolution reflection analysis of data from 168 seismic stations with an average spacing of about 5 km, in northern Kinki district, southwestern Japan. Reflection analysis has previously been conducted in this region, assuming a homogeneous horizontal structure, resulting in an inclined planar zone of high relative reflection strengths (S-wave reflector). However, if the reflector is actually inclined, the location of the S-wave reflector differs from that of an assumed homogeneous horizontal structure. Hence, this study conducted high-resolution reflection analysis to determine the accurate location of the S-wave reflector. We confirm the previously reported S-wave reflector (reflector W). Furthermore, we detected the accurate location of the S-wave reflector and obtained more detailed results that revealed a second S-wave reflection structure (reflector E) to the east of reflector W, in an area that has not be imaged by previous studies. The northern edges of reflector E and reflector W are located near different hypocentral areas of deep low-frequency earthquakes (DLFs). Reflector W exists along the Kyoto Nishiyama fault zone, and its position appears to change along the fault zone as it deepens. Similarly, reflector E exists along the Hanaore and Biwako Seigan fault zones and its position appears to change along these fault zones. The reflector W and reflector E are imaged as separate S-wave reflectors in deeper regions, but they coalesce in shallower regions. According to previous studies, crustal fluid by dehydration from the Philippine Sea plate exists near these epicenters and we infer that this crustal fluid causes DLFs and forms S-wave reflectors.

After a large earthquake, many small earthquakes, called aftershocks, ensue. Additional large earthquakes typically do not occur, despite the fact that the large static stress near the edges of the fault is expected to trigger further large earthquakes at these locations. Here we analyse ~10,000 highly accurate focal mechanism solutions of aftershocks of the 2016 Mw 6.2 Central Tottori earthquake in Japan. We determine the location of the horizontal edges of the mainshock fault relative to the aftershock hypocentres, with an accuracy of approximately 200 m. We find that aftershocks rarely occur near the horizontal edges and extensions of the fault. We propose that the mainshock rupture was arrested within areas characterised by substantial stress relaxation prior to the main earthquake. This stress relaxation along fault edges could explain why mainshocks are rarely followed by further large earthquakes.