[M-IS04] Interdisciplinary studies on pre-earthquake processes

[MIS04-06] Application of probabilistic prediction model integrating regional stress to reservoir earthquakes

*Rui Wang¹、Peng Han¹、Jiancang Zhuang²、Miao Miao¹、Ce Zhao³、Cuiping Zhao⁴、Renqi Lu⁵、Hongfu Lei⁶、Mengdi Yao⁶ (1.Southern University of Science and Technology, Shenzhen, China 、2.The Institute of statistic Mathematics,Tokyo, Japan、3.Institute of Geophysics, China Earthquake Administration, Beijing, China、4.Institute of Earthquake Forecasting, China Earthquake Administration, Beijing, China、5.Institute of Geology, China Earthquake Administration, Beijing, China、6.China Three Gorges Construction Engineering Corporation, Chengdu, China)

キーワード：reservoir earthquake, seismicity change, regional stress, ETAS model

Epidemic-Type Aftershock Sequence (ETAS) model is one of the most widely used statistical models applied to earthquake prediction. As a statistical method, ETAS and its extension are all based on earthquake catalogs that may ignore the physical mechanism. This makes the method impossible in some cases, in which the region stress changes frequently, such as reservoirs, shale gas extraction, and wastewater injection. Coulomb Rate and State Mode (CRSM) is based on the physical mechanism of stress influence on earthquakes but wake on earthquakes self-triggering and depending on stress simulation. We propose a new kind of probabilistic prediction model integrating regional stress evolution, which can combine the advantage of the two models. We analyze the seismicity of three regions near the reservoir. We find earthquakes in two of them can’t be fitted by the stationary ETAS model and their background depends on time. We apply the probabilistic prediction model integrating regional stress evolution to the two regions and find it can achieve a good fitting effect. The results show that the earthquakes near the reservoir are induced by the reservoir filling and the probabilistic prediction model integrating regional stress evolution can be well applied to reservoir earthquake and similar case.Epidemic-Type Aftershock Sequence (ETAS) model is one of the most widely used statistical models applied to earthquake prediction. As a statistical method, ETAS and its extension are all based on earthquake catalogs that may ignore the physical mechanism. This makes the method impossible in some cases, in which the region stress changes frequently, such as reservoirs, shale gas extraction, and wastewater injection. Coulomb Rate and State Mode (CRSM) is based on the physical mechanism of stress influence on earthquakes but wake on earthquakes self-triggering and depending on stress simulation. We propose a new kind of probabilistic prediction model integrating regional stress evolution, which can combine the advantage of the two models. We analyze the seismicity of three regions near the reservoir. We find earthquakes in two of them can’t be fitted by the stationary ETAS model and their background depends on time. We apply the probabilistic prediction model integrating regional stress evolution to the two regions and find it can achieve a good fitting effect. The results show that the earthquakes near the reservoir are induced by the reservoir filling and the probabilistic prediction model integrating regional stress evolution can be well applied to reservoir earthquake and similar case.