Changning, an area in the southwestern Sichuan basin, is characterized by abundant resources of salt mine and shale gas. With rapid and large-scale exploitation of the shale gas, this area has suffered several moderate-to-strong earthquakes and numerous micro ones. The active seismicity means the seismic risk in the area is considerable.
For the assessment of seismic risk, the seismic parameter of b in the frequency-magnitude relationship plays an important role because it can reflect the change in underground stress. To obtain the b-value characteristics in Changning area, we first constructed a high-resolution earthquake catalog with a deep learning-based method including phase picking, event association and earthquake location. The data for constructing the catalog are continuous waveforms recorded by 107 dense stations in Changning from April 28, 2021 to June 30, 2022. Finally, we have located more than 240 000 seismic events and employed a maximum likelihood method to calculate b-values.
For the whole Changning area, b-value is 1.06 on average, larger than that of Sichuan Province (0.8) . The parameter b behaves differently in the northern Changning-Shuanghe brine mining area and in the southern Changning-Zhaotong shale gas production area. Their b-values are 1.25 and 1.01, respectively, indicating that earthquakes in the two areas have distinctly different seismogenic environments. Changning area is divided into elements with a size of 0.02°×0.02°. As for each element, if earthquakes within it are not less than 50, we calculate its b-value. The resulting b-value spatial map shows significant heterogeneity, which reflects the heterogeneity of the underground media to some degree. Most earthquakes with magnitudes above 4.0 occur within or on the edges of low b-value regions. From this we can conclude that felt earthquakes will tend to occur in such regions.
The temporal evolution of b-value from three largest earthquake clusters shows that the b-value is low or significantly decreases before a large earthquake, and increases rapidly after the event. The decrease in b-value before the occurrence of large events probably indicates the accumulation of stress, similar to large tectonic earthquakes.
These b-value spatio-temporal characteristics have implications for judging where and when large earthquakes will happen. Therefore, comprehensive analysis of the b-value spatial distribution map and its temporal evolution is helpful for assessing the seismic hazard of Changning area.