AUTHOR=Zhou Cong , Fu Lei , Shi Kexu , Zeng Xiangzhi , Zhang Pei TITLE=Velocity changes after the 2021 MS 6.4 Yangbi earthquake based on passive image interferometry JOURNAL=Frontiers in Earth Science VOLUME=10 YEAR=2023 URL=https://www.frontiersin.org/journals/earth-science/articles/10.3389/feart.2022.1072017 DOI=10.3389/feart.2022.1072017 ISSN=2296-6463 ABSTRACT=

An MS 6.4 earthquake occurred in Yangbi, Yunan Province, China, on 21 May 2021. The epicenter was on the blind branch fault in the west of the Weixi–Qiaohou–Weishan fault, but no surface rupture was obvious. In the present study, the continuous vertical component of waveforms that were recorded in six nearby permanent stations was collected and the noise cross-correlation and autocorrelation techniques were utilized to investigate velocity changes that were induced by the Yangbi Earthquake. Velocity changes based on the single-station autocorrelation method reveal mainly coseismic declines, and a maximum of .09% was recorded in the EYA station. Results from the cross-correlation technique show both positive and negative velocity changes, and these lasted for approximately 3 months. The volumetric strain that was generated by the Yangbi Earthquake at a depth of 5 km exhibits an obvious four-quadrant distribution. Station pairs in the dilatation region (e.g., EYA–HEQ) mainly display a decrease in velocity, whereas those in the contraction region (e.g., BAS–TUS, TUS–YUL, and LUS–TUS) show an increase in velocity. Based on the depth sensitivity of scattered waves, velocity changes that were obtained using the noise cross-correlation involve the highest weight coefficients near the related two stations. Regarding stations of one station pair in different stress loading regions, the static stress of the station that is nearest to the epicenter exerted a greater impact on the velocity change. The observed velocity changes are likely attributed to a combination of near-surface physical damage and static stress changes. The validation of clock errors with magnitudes of seconds that were obtained using the noise cross-correlation and effects of these errors on measured velocity changes are also discussed.