Due to the joint influence of climate change and human activities, the hydrological rhythm of Poyang Lake has changed in recent years, leading to an increasingly severe drought problem during autumn and winter in this region. Notably, the extreme drought that occurred in 2022 had profound impacts on shipping, water supply and the ecological environment of the wetlands in the Poyang Lake Delta, sparking widespread concern. Based on the historical hydrometeorological data of Poyang Lake, we used statistical models (such as Chow test, correlation analysis, etc.) to analyze the cause of the extreme drought in the Poyang Lake Delta from the perspectives of natural factors and human activity. Through correlation analysis, we found that the water level, discharge, and drought duration of the Poyang Lake Delta were all significantly affected by climate change, particularly rainfall in the Poyang Lake basin. Furthermore, combining the results of Chow test and correlation analysis, we also found that the operation of the Three Gorges Reservoir had a notable impact on the water level of the Poyang Lake Delta. Based on remote sensing images, ecological and environmental sampling monitoring, the damage inflicted by the extreme drought event on the Poyang Lake Delta was analyzed. The results show that the inundated area of the delta wetlands in the extreme-drought year (2022) decreased by 45.75% compared with that in a normal year (2017). In addition, the ecological environment of the wetlands deteriorated significantly. The water quality parameters (TN, TP, NH₄⁺-N) increased by 50.2%, 240% and 64.7%, respectively. The concentrations of TN and TP were 3.8 mg/L and 0.17 mg/L, respectively, while the context values in the delta were 1.2 mg/L and 0.075 mg/L. The density and biomass of algae increased by 87.2% and 557.9%, respectively. In contrast, the density and biomass of benthos decreased by 59.9% and 78.5%, respectively. The control strategy for the Poyang Lake Delta under extreme drought was studied through an experiment on the operation of hydraulic controllers. The results show that under extreme drought conditions, the newly built hydraulic controllers could raise the water level of the delta from 9.1 ± 0.7 m to 14.2 ± 1.8 m, thus effectively solving the water cut-off problem in the four branches of the delta. Furthermore, by adjusting the distributive ratio of the main, north, middle and south branches of the delta to 50%, 4%, 24% and 22% through newly built hydraulic controllers, the water area can be increased by 56%.

Predicting slope deformation is pivotal for reservoir safety management; however, quantitative attribution to hydrologic–temporal factors with interpretable and hyperparameter-robust models under multi-point temporal dependence is still rare. Hence, we develop an interpretable hybrid framework that couples a Light Gradient Boosting Machine (LightGBM) with a Newton–Raphson-based optimizer (NRBO) for hyperparameter tuning. Unsupervised clustering is first employed to capture intrinsic temporal associations among multiple monitoring points. Subsequently, the NRBO–LightGBM framework is proposed to enhance prediction accuracy and model robustness in slope deformation prediction. Finally, SHAP analysis is integrated to quantify the contribution of influencing factors, thereby strengthening the physical interpretability and credibility of the model. The proposed framework is validated using long-term deformation monitoring data from the Lijiaxia Hydropower Station. Comparative experiments indicate that the NRBO–LightGBM model achieves a 22.8% reduction in RMSE and an 11.4% increase in R² relative to conventional statistical models, improving prediction accuracy with a 21.5% lower RMSE and a 15.5% higher R² compared with the baseline LightGBM. Furthermore, SHAP interpretability analysis elucidates the internal predictive mechanism, revealing that deformation evolution is primarily governed by temporal accumulation and seasonal variations represented by the time variable t and periodic components. Overall, the NRBO–LightGBM model provides high-precision and interpretable deformation prediction for reservoir slopes, effectively bridging predictive performance with mechanistic understanding and offering actionable insights for landslide early warning and risk management.

The problem of low water level and uneven distribution of flow in the multi-branch channels at the tail of the Ganjiang River (GJRT) during the dry season has been affecting the local water supply, navigation, and aquatic ecological environment. In recent years, water conservancy projects have been built in each branch of the multi-branch channels at the GJRT. Finding a way to utilize the water conservancy project group to carry out joint regulation and meet the water level and discharge requirements of each branch is an important issue that urgently needs to be solved. This paper analyzes the hydrodynamic process and its impact on water supply, navigation, and ecology in multi-branch channels without water conservation projects through hydrological data analysis and numerical simulation. By conducting numerical experiments on joint regulation of water conservation project group, a multi-objective regulation strategy is proposed to meet the water level and discharge of each branch. The results indicate that the discharge at the GJRT has been continuously decreasing from 1 September. Due to the jacking effect of Poyang Lake, the water level plunges at the GJRT from 1 October, which occurred later than the decrease in water level. The disruption of water levels and discharge makes it difficult to meet the regional water demand. The optimal time to initiate regulation is 1 October, and the target water level of Waizhou Station is 15.5 m, located upstream of the Ganjiang River tail. When the water level before each branch project gate is uniform and exceeds 15.5 m, the water level of Waizhou Station satisfies the requirement. However, the discharge of each branch does not meet the demand. In contrast to a scheduling regulation strategy that maintains the same water level in front of each gate, adopting a strategy with different water levels before each gate can effectively adjust the diversion ratio and fulfill the discharge demand of each branch at the tail of the Ganjiang River.