Floods are among the most destructive natural disasters, and accurate flood susceptibility mapping (FSM) is crucial for disaster prevention and mitigation amid climate change. The Poyang Lake basin, characterized by complex flood formation mechanisms and high spatial heterogeneity, poses challenges for the application of FSM models. Currently, the use of machine learning models in this field faces several bottlenecks, including unclear model applicability, limited sample quality, and insufficient machine interpretation. To address these issues, we take the 2020 Poyang Lake flood as a case study and establish a high-precision flood inundation sample database. After feature screening, the performance of three hybrid models optimized by Particle Swarm Optimization (PSO)—Random Forest (RF), Extreme Gradient Boosting (XGBoost), and Convolutional Neural Network (CNN) is compared. Furthermore, the Shapley Additive exPlanations (SHAP) framework is employed to interpret the contributions and interaction effects of the driving factors. The results demonstrate that the ensemble learning models exhibit superior performance, indicating their greater applicability for flood susceptibility mapping in complex basins such as Poyang Lake. The RF model has the best predictive performance, achieving an area under the receiver operating characteristic curve (AUC) value of 0.9536. Elevation is the most important global driving factor, while SHAP local interpretation reveals that the driving mechanism has significant spatial heterogeneity, and the susceptibility of local depressions is mainly controlled by the terrain moisture index. A nonlinear phenomenon is observed where the SHAP value was negative under extremely high late rainfall, which is preliminarily attributed to the “spatial transfer that is prone to occurrence” mechanism triggered by the backwater effect, highlighting the complex nonlinear interactions among factors. The proposed “high-precision sampling, model comparison, SHAP explanation” framework effectively improves the accuracy and interpretability of FSM. These research findings can provide a scientific basis for smart flood control and precise flood risk management in basins.

Over the past 30 years, marine-type glaciers in the eastern Himalayas have been significantly affected by climate change, making the region a key area for global climate change research. Based on Landsat series remote sensing imagery, this study employs visual interpretation and the band ratio threshold method to analyze the area and variation characteristics of marine-type glaciers in the eastern Himalayas from 1990 to 2015. It further investigates the influence of debris cover, glacial lakes, and climate change on glacier ablation. The results indicate that (1) From 1990 to 2015, marine-type glaciers in the eastern Himalayas showed an accelerating retreat trend, with an average annual retreat rate of 0.48%/a. (2) A significant increase in temperature was the primary climatic factor driving the accelerated retreat of glaciers during the study period. In the future, changes in precipitation may surpass temperature as the dominant climatic factor influencing glacier variation in the study area. (3) Debris cover mainly played a mitigating role in glacier retreat in the eastern Himalayas. Within the 50°–60° slope range, debris-covered glaciers gradually became dominant, resulting in a lower average annual retreat rate during 2000–2015 compared to 1990–2000. (4) The impact of climate change on glaciers in the study area was greater than that of glacial lakes. This study reveals the mechanisms by which debris cover, glacial lakes, and climate change influence glacier ablation, providing scientific support for glacier disaster early warning and water resource assessment.