Li, Aibing2022-06-18August 2022021-08August 202https://hdl.handle.net/10657/9317Precursory anisotropy change before a major earthquake remains a controversial topic in earthquake seismology. Japan is an ideal place to study this topic because it experiences numerous large earthquakes and has a dense seismic network. We analyze shear-wave splitting (SWS) data from foreshocks recorded at four Hi-net seismic stations (JUOH, ASGH, SKGH, and TYNH) for four major earthquakes in Fukushima, Shizuoka, Tottori, and Kumamoto. SWS parameters (fast polarization direction and delay time) are calculated using a semi-automatic algorithm. The average fast directions at the four stations are either parallel to the local maximum horizontal stress orientation or the fault strike, consistent with previous studies. Both SWS parameters do not show noticeable temporal variations before large earthquakes at station JUOH, ASGH, and TYNH. At station SKGH in Tottori, the delay time varies randomly within a small range for most of the foreshocks that occurred from 25 days to 1 day before the main earthquake. A sharp jump of delay time (0.03 s) is observed within two hours between two groups of events before the mainshock in Tottori. The earlier group of events are relatively far from the mainshock, while the latest foreshocks are very close to the mainshock, indicating the region near the major earthquake is more anisotropic and evidencing a precursory anisotropy change. Our observations suggest that it is possible to observe anisotropy changes before large earthquakes when there are favorable foreshock datasets, although the success rate is low. More SWS measurements from foreshocks need to be conducted to corroborate the findings in this studyapplication/pdfengThe author of this work is the copyright owner. UH Libraries and the Texas Digital Library have their permission to store and provide access to this work. Further transmission, reproduction, or presentation of this work is prohibited except with permission of the author(s).SWS, Earthquakes2022-06-18Thesisborn digital