Green water accounts for two-thirds of precipitation, and the proportion could be even higher in dry years. Conflicts between water supply and demand have gradually become severe in the Hai River Basin (HRB) due to the socio-economic development. Thus, the exploitation and the utilization of green water have attracted increasing attention. By gathering the related hydrological, meteorological, and geographic data, the spatiotemporal distribution of green water in HRB and the impacts of land use types on green water are analyzed based on the SWAT (Soil and Water Assessment Tool) model in this study. Furthermore, three new indices are proposed for evaluation, including the maximum possible storage of green water (MSGW), the consumed green water (CGW), and the utilizable green water (UGW). The results show that (1) the MSGW is relatively low in plain areas and its spatial distribution is significantly associated with the soil type; (2) according to the evaluation results of CGW and UGW in HRB, a further improvement of utilization efficiency of green water could be achieved; (3) in general, the utilization efficiency of precipitation in farmlands is higher than other land use types, which means that the planting of appropriate plants could be helpful to enhance the utilization efficiency of green water. Our results summarize the spatiotemporal distribution of green water resource and provide a reference for water resources management in other water-short agricultural areas.

Due to the uncertainty of meteorological factors and the influence of human activities, the monthly runoff series often exhibit the characteristics of non-stationarity. The appropriate prediction model and the hyperparameters of the model are often difficult to determine, and this affects the model prediction performance. For obtaining the accurate runoff prediction results, a novel prediction model (KVMD-KTCN-LSTM-SA) is proposed. This hybrid model uses Kepler optimization algorithm (KOA)-optimized Variable Mode Decomposition (KVMD), KOA-optimized temporal convolutional network–long short-term memory (TCN-LSTM), and the self-attention (SA) mechanism. KVMD effectively reduces the difficulty of predicting the monthly runoff series, KOA helps to find the optimal hyperparameters of the model, TCN is combined with LSTM, and the SA mechanism effectively increases the performance of the model. Monthly runoff from three hydrological stations in the Hetian River basin and one hydrological station in the Huaihe River basin are predicted with the proposed model, and six models are selected for comparison. The KVMD-KTCN-LSTM-SA model effectively reduces runoff fluctuation and combines the advantages of multiple models and achieves satisfactory runoff prediction results. During the testing period, the proposed model achieves NSE of 0.978 and R² of 0.982 at Wuluwati station, NSE of 0.975 and R² of 0.986 at Tongguziluoke station, and NSE of 0.978 and R² of 0.982 at Jiangjiaji station. The proposed hybrid model provides a new approach for monthly runoff prediction, which is capable of better managing and predicting mid-long-term runoff.