As one of the most crucial indices of sustainable development and water security, water resources carrying capacity (WRCC) has been a pivotal and hot-button issue in water resources planning and management. Quantifying WRCC can provide useful references on optimizing water resources allocation and guiding sustainable development. In this study, the WRCCs in both current and future periods were systematically quantified using set pair analysis (SPA), which was formulated to represent carrying grade and explore carrying mechanism. The Soil and Water Assessment Tool (SWAT) model, along with water resources development and utilization model, was employed to project future water resources scenarios. The proposed framework was tested on a case study of China’s Han River basin. A comprehensive evaluation index system across water resources, social economy, and ecological environment was established to assess the WRCC. During the current period, the WRCC first decreased and then increased, and the water resources subsystem performed best, while the eco-environment subsystem achieved inferior WRCC. The SWAT model projected that the amount of the total water resources will reach about 56.9 billion m³ in 2035s, and the water resources development and utilization model projected a rise of water consumption. The declining WRCC implies that the water resources are unable to support or satisfy the demand of ecological and socioeconomic development in 2035s. The study furnishes abundant and valuable information for guiding water resources planning, and the core idea of this model can be extended for the assessment, prediction, and regulation of other systems.

Water environmental capacity (WEC) is an essential indicator for effective environmental management. The designed low water flow condition is a prerequisite to determine WEC and is often based on the stationarity assumption of low water flow series. As the low water flow series has been remarkably disturbed by climate change as well as reservoirs operation and water acquisition, the stationarity assumption might bring risk for WEC planning. As the reservoir operation and water acquisition under climate change can be simulated by a water resources allocation model, the low water flow series outputted from the model are the simulations of the disturbances and often show nonstationary conditions. After estimating the designed low water flow through nonstationary frequency analysis from these low water flow series, the WEC under the nonstationary conditions can be determined. Thus, the impacts of water resources allocation on WEC under climate change can be quantitatively assessed. The mid-lower reaches of the Hanjiang River basin in China were taken as a case study due to the intensive reservoir operation and water acquisition under the climate change. A representative concentration pathway scenario (RCP4.5) was employed to project future climate, and a Soil and Water Assessment Tool (SWAT) model was employed to simulate water availability for driving the Interactive River-Aquifer Simulation (IRAS) model for allocating water. Water demand in 2016 and 2030 were selected as baseline and future planning years, respectively. The results show that water resources allocation can increase the amount of WEC due to amplifying the designed low water flow through reservoir operation. Larger regulating capacities of water projects can result in fewer differences of WEC under varied water availability and water demand conditions. The increasing local water demand will decrease WEC, with less regulating capacity of the water projects. Even the total available water resources will increase over the study area under RCP4.5. More water deficit will be found due to the uneven temporal-spatial distribution as well as the increasing water demand in the future, and low water flow will decrease, which further leads to cut down WEC. Therefore, the proposed method for determining the WEC can quantify the risk of the impacts of water supply and climate change on WEC to help water environmental management.

Hydro-meteorological datasets are key components for understanding physical hydrological processes, but the scarcity of observational data hinders their potential application in poorly gauged regions. Satellite-retrieved and atmospheric reanalysis products exhibit considerable advantages in filling the spatial gaps in in-situ gauging networks and are thus forced to drive the physically lumped hydrological models for long-term streamflow simulation in data-sparse regions. As machine learning (ML)-based techniques can capture the relationship between different elements, they may have potential in further exploring meteorological predictors and hydrological responses. To examine the application prospects of a physically constrained ML algorithm using earth observation data, we used a short-series hydrological observation of the Hanjiang River basin in China as a case study. In this study, the prevalent modèle du Génie Rural à 9 paramètres Journalier (GR4J-9) hydrological model was used to initially simulate streamflow, and then, the simulated series and remote sensing data were used to train the long short-term memory (LSTM) method. The results demonstrated that the advanced GR4J9–LSTM model chain effectively improves the performance of the streamflow simulation by using more remote sensing data related to the hydrological response variables. Additionally, we derived a reservoir operation model by feeding the LSTM-based simulation outputs, which further revealed the potential application of our proposed technique.