Analysis of Seismicity Characteristics and Seismogenic Mechanism in Badong County of Three Gorges Reservoir Area Before and After Impoundment

We built a high-resolution Boussinesq water-load model for the Badong County section of the Three Gorges Reservoir by referring to Google Earth image data for different years. We also analyzed the focal mechanism of some significant earthquakes, Coulomb stress, coda spectrum, the precise location of small earthquakes, and the characteristics of seismicity migration. Based on these investigations, we explain the seismogenic mechanism of earthquakes (from 2003 to now) in Badong County and trace the evolution of earthquakes in the reservoir area before and after impoundment on a NE–SW transect. Some preliminary findings are listed as follows: (1) The calculation results show that the hydraulic or Coulomb stress caused by the direct water load in the reservoir area is not the main reason for the continuous occurrence of earthquakes in recent years; in other words, the direct effect of hydraulic stress only plays a secondary role. (2) The migration and evolution of seismicity, together with the variation of coda spectrum density, suggest that the characteristics of different stages of fluid infiltration are obviously different. With the influence of continuous infiltration, earthquakes have occurred from shallow to deep, from scattered to crowded, and have gradually approached deeper and larger faults. In addition, the high-frequency energy released by these earthquakes increased. (3) Compared with the Wenchuan MS 8.0 and Xinfengjiang MS 6.1 earthquake, no earthquakes larger than MS 5.2 have occurred in the Badong and Zigui areas in the past 17 years, which far exceeds the time lag between the two mentioned strong earthquakes and impoundment of the respective reservoirs. This finding indicates that the occurrence of strong earthquakes near the reservoir area is mainly controlled by the risk level of original strong earthquakes, while fluid infiltration only provides a certain induction effect. (4) More attention should be paid to the role of invisible water in the process of seeping into the strata on both sides of the river. We conservatively estimate that the amount of invisible fluid infiltrating on both sides may be 15.1% greater than the visible water in the study area. Therefore, more careful analysis of geological disasters (e.g., landslides and earthquakes) caused by fluid infiltration of a long, narrow, and sharply sloped river after artificial impoundment is indeed required.