10:00 AM - 10:15 AM
[SCG55-05] Postseismic Displacement of the 2022 Chihshang Earthquake in Eastern Taiwan Detected by InSAR and GNSS
Keywords:InSAR, GNSS, earthquake, fault, Taiwan, postseismic deformation
Taiwan is located at the collision boundary between the Philippine Sea Plate and the Eurasian Plate. Two large-scale active faults consume a large portion of plate convergence: the longitudinal valley fault (LVF) and the Central Range Fault (CRF). The Guanshan earthquake (Mw 6.5) occurred on the CRF on September 17, 2022, and the Chihshang earthquake (Mw 7.0) occurred about 17 hours later on September 18. The Chihshang earthquake also induced a slip on the LVF (Lee et al., 2023; Tang et al., 2023), but it is unclear how the stress accumulated on these faults and how they were redistributed by the earthquake. In this study, we used InSAR and GNSS data to detect the postseismic displacement of the Chihshang earthquake in detail with high accuracy and to clarify the process of stress accumulation and redistribution on the CRF and LVF.
First, we investigated the displacement field of each station for one year after the Chihshang earthquake using GNSS daily coordinates obtained by Precise Point Positioning (PPP). The result (Figure a) shows a displacement discontinuity along the LVF (a westside uplift of up to 10 cm and a left-lateral displacement of about 10 cm). This GNSS displacement pattern is clearly different from the steady-state convergence of about 2 cm/yr and slight westside subsidence observed along the LVF before the earthquake. Since the hypocentral area is located in the mountains with limited GNSS data, we therefore used InSAR data to increase the spatial resolution of surface displacements. To penetrate the dense vegetation in eastern Taiwan, we used the L-band SAR data acquired by ALOS-2. We then applied the Split Spectrum Method (Gomba et al., 2016; Wegmüller et al., 2018) to mitigate the strong influence of the ionospheric disturbances. The tropospheric disturbances were also corrected using GACOS (Yu et al., 2018). After these corrections, we stacked 39 interferograms (e.g., Figure b) created from 17 ScanSAR data taken from September 2022 to December 2023. The results show that (1) displacement discontinuity across the LVF, (2) satellite line-of-sight (LOS) shortening over the vast area to the west of CRF, (3) lack of the LOS length change at the main rupture area of the Chihshang earthquake estimated by Tang et al. (2023). The second characteristic (2) would be mainly due to the postseismic slip of the CRF and partly by the inelastic flow of the surrounding rocks. These characteristics provide the keys to quantitatively understanding the stress distribution before the Chihshang earthquake and the kinematic interaction between LVF and CRF. In the future, we will construct a kinematic model that explains the temporal evolution of GNSS and InSAR data, by which we quantitatively understand the stress build-up and re-distribution on LVF and CRF.
First, we investigated the displacement field of each station for one year after the Chihshang earthquake using GNSS daily coordinates obtained by Precise Point Positioning (PPP). The result (Figure a) shows a displacement discontinuity along the LVF (a westside uplift of up to 10 cm and a left-lateral displacement of about 10 cm). This GNSS displacement pattern is clearly different from the steady-state convergence of about 2 cm/yr and slight westside subsidence observed along the LVF before the earthquake. Since the hypocentral area is located in the mountains with limited GNSS data, we therefore used InSAR data to increase the spatial resolution of surface displacements. To penetrate the dense vegetation in eastern Taiwan, we used the L-band SAR data acquired by ALOS-2. We then applied the Split Spectrum Method (Gomba et al., 2016; Wegmüller et al., 2018) to mitigate the strong influence of the ionospheric disturbances. The tropospheric disturbances were also corrected using GACOS (Yu et al., 2018). After these corrections, we stacked 39 interferograms (e.g., Figure b) created from 17 ScanSAR data taken from September 2022 to December 2023. The results show that (1) displacement discontinuity across the LVF, (2) satellite line-of-sight (LOS) shortening over the vast area to the west of CRF, (3) lack of the LOS length change at the main rupture area of the Chihshang earthquake estimated by Tang et al. (2023). The second characteristic (2) would be mainly due to the postseismic slip of the CRF and partly by the inelastic flow of the surrounding rocks. These characteristics provide the keys to quantitatively understanding the stress distribution before the Chihshang earthquake and the kinematic interaction between LVF and CRF. In the future, we will construct a kinematic model that explains the temporal evolution of GNSS and InSAR data, by which we quantitatively understand the stress build-up and re-distribution on LVF and CRF.