Melt/fluid Migration and the 2022 Luding Ms 6.8 Earthquake

Xingpeng Dong,Dinghui Yang,Weijuan Meng

DOI: https://doi.org/10.1360/tb-2023-0015

2023-01-01

Abstract:The main seismogenic fault of the 2022 Luding Ms 6.8 earthquake is the Moxi fault in the southeast section of the Xianshuihe fault zone, which has caused serious casualties and heavy economic losses. Historically, a destructive earthquake of magnitude 7 3/4 occurred on the Moxi fault in 1786, which caused landslides around the area to block the Dadu River and caused serious flooding. Therefore, studying the deep structure, thermal-rheological state and deformation characteristics of the Luding earthquake and its surrounding areas will help to deepen the understanding of the preparation and occurrence mechanism of large earthquakes in this region. The time scale of large earthquakes can be as long as hundreds of years or even thousands of years. It is not a simple elastic behavior, but involves a complex rheological process. Therefore, obtaining high-resolution and multi-parameter three-dimensional (3D) images of crust-mantle structure (such as velocity, anisotropy, Poisson's ratio, etc.) of the eastern Tibetan Plateau and discussing the seismogenic tectonic background of large earthquakes in this region are the current hotspots and important directions. To achieve this goal, it is necessary to develop a method that can maximize the use of all useful information (including travel time and amplitude) in seismic waveform signals. Full waveform inversion can simultaneously use phase and amplitude information in seismic records to constrain subsurface structures, with a resolution up to half of the shortest wavelength, which is superior to other traditional tomography methods and is a very suitable choice. In this study, we use the full waveform inversion method to obtain high-resolution seismic wave velocity, radial anisotropy and Poisson's ratio structures around the source area of Luding earthquake. In order to verify the reliability of the imaging results, we carried out a recovery test. A small low-velocity perturbation of SV wave is placed near the source area in the initial model and then the Frechet sensitive kernels of the corresponding SH wave and SV wave are calculated by using the adjoint method. The results suggest that the Frechet kernel pattern of SV wave is generally consistent with the shape of anomaly body and the tradeoff between V-SH and V-SV can be ignored. Since the Frechet kernel indicates the update direction of the model in full waveform inversion, the recovery test demonstrates that our imaging results are reliable. The focal mechanism solution shows that the Luding Ms 6.8 earthquake is a strike-slip type (www.globalcmt.org), indicating that it is mainly controlled by the differential motion between the Sichuan-Yunan block and the Bayan Har block. However, whether the earthquake was only controlled by horizontal motion still need to be further clarified. For this reason, we give the vertical profiles of the S-wave velocity perturbation, radial anisotropy and Poisson's ratio along 29.6 degrees N and 102.1 degrees E across the focal area of the Luding earthquake. The results show that the Luding earthquake occurred at the edge of the low-velocity perturbation. There is an anomaly region of low S-wave velocity, negative radial anisotropy (V-SH<V-SV) and high Poisson's ratio just below the source area of Luding earthquake. The formation of this region may be related to the fluids (including partial melting) within the deep crust and upper mantle. These melts/fluids probably migrate upward along the inclined Moho and/or intracrustal faults, causing the weakening and stress concentration of the seismogenic layer in the middle and upper crust, and leading to the occurrence of earthquakes. Our tomographic results indicate that melt/fluid migration might play an important role in the initiation and rupture of the Luding earthquake.