Abstract: In October and November 2018, two large-scale landslides occurred at the same location on the right bank of the Jinsha River in Baige Village, Jiangda County, Tibet Autonomous Region, resulting in the formation of a dammed lake. The backwater of the dammed lake and the flood caused by the dam break flooded many towns and villages upstream and downstream, and had a great impact on the construction of cascade power stations downstream. Remote sensing images and field investigation analysis reveal that a large number of potentially unstable rock and soil bodies remained at the periphery of the two landslides. Notably, in December 2022, a re-collapse of the residual body along the left rear margin generated large quantities of debris deposits. Observations indicate that the above rock and soil bodies still had the possibility of instability in the future, leading to landslide blocking and the formation of secondary disaster chains. To assess and predict the impact of residual instability and the chain disaster process on the Yebatan Hydropower Station, this study used the numerical simulation methods of RAMMS, DB-IWHR, and HEC-RAS to analyze the disaster chain of formation, dam-break, and flood evolution of different combination barrier lakes based on the development trend of residual bodies. The results show that, after the two landslide events, the residual rock and soil body around the landslide is still in internal stress adjustment and slow deformation, and has a trend of further instability. The latest geological prospecting and monitoring data show that the residual body in areas of K1-1, K1-2, K1-3, K2-1, K2-2, and K3 have a large risk of instability and river block. Composite 9, representing the most adverse condition, has a residual body size of 435.11×10⁴ m³. Simulations predict that the height of the overflow dam formed by the unstable landslide blocking the river is 48 m, and the peak discharge of the blockage body is 13079.35 m³/s. The peak discharge when it evolves to Yebatan Hydropower Station decays to 9873.03 m³/s, with slight delay in flood arrival time. This study provides a scientific basis for the risk prediction and emergency treatment plan of Baige landslide.