Traditional hydrodynamic models face the significant challenge of balancing the demands of long prediction spans and precise boundary conditions, large computational areas, and low computational costs when attempting to rapidly and accurately predict the nonlinear spatial and temporal characteristics of fluids at the basin scale. To tackle this obstacle, this study constructed a novel deep learning framework with a hydrodynamic model for the rapid spatiotemporal prediction of hydrodynamics at the basin scale, named U-Net-ConvLSTM. A validated high-fidelity hydrodynamic mechanistic model was utilized to build a 20-year hydrodynamic indicator dataset of the middle and lower reaches of the Han River for the training and validation of U-Net-ConvLSTM. The findings indicate that the $R^2$ value of the model surpassed 0.99 when comparing the single-step prediction results with the target values. Additionally, the required computing time fell by 62.08% compared with the hydrodynamic model. The ablation tests demonstrate that the U-Net-ConvLSTM framework outperforms other frameworks in terms of accuracy for basin-scale hydrodynamic prediction. In the multi-step-ahead prediction scenarios, the prediction interval increased from 1 day to 5 days, while consistently maintaining an $R^2$ value above 0.7, which demonstrates the effectiveness of the model in the missing boundary conditions scenario. In summary, the U-Net-ConvLSTM framework is capable of making precise spatiotemporal predictions in hydrodynamics, which may be considered a high-performance computational solution for predicting hydrodynamics at the basin scale.

Land use monitoring by satellite remote sensing can improve the capacity of ecosystem resources management. The satellite source, bandwidth, computing speed, data storage and cost constrain the development and application in the field. A novel deep learning classification method based on feature augmentation (CNNs-FA) is developed in this paper, which offers a robust avenue to realize regional low-cost and high-precision land use monitoring. Twenty-two spectral indices are integrated to augment vegetation, soil and water features, which are used for convolutional neural networks (CNNs) learning to effectively differentiate seven land use types, including cropland, forest, grass, built-up, bare, wetland and water. Results indicated that multiple spectral indices can effectively distinguish land uses with a similar reflectance, achieving an overall accuracy of 99.70%, 94.81% and 90.07%, respectively, and a kappa coefficient of 99.96%, 98.62% and 99.76%, respectively, for Bayannur, Ordos and the Hong Lake Basin (HLB). The overall accuracy of 98.18% for the field investigation demonstrated that the accuracy of the classification in wet areas and ecologically sensitive areas was characterized by significant desert–grassland interspersion.

Main Functional Area Planning (MFAP) is a significant initiative in China, aimed at promoting coordinated socio-economic progress while ensuring resource capacity and environmental sustainability. However, there is a lack of quantitative assessments of China’s county-level agricultural development level (ADL) following the implementation of the MFAP. In this study, a coupled “agricultural product-agricultural space-agricultural population” evaluation index system which was based on plan requirements, remote sensing imagery, statistical data, and industry-specific information was proposed for assessing the development level of agricultural after implementing the MFAP, and we utilized the system to evaluate the ADL of 2850 counties across China from 2009 to 2015 at the county level. The results indicate that MFAP has played a positive role in driving agricultural development in China. From 2009 to 2015, the ADL of counties in China showed an upward trend, and the agricultural development within the “Seven regions & Twenty-three belts” reached a high level, with the proportion of the top three districts and counties in the agricultural development zone increasing from 86.78% to 88.72%. The spatial distribution of ADL ratings shows a central > east > northeast > west pattern, with the western regions exhibiting the fastest growth rate. Moreover, targeted policies were provided for pathway optimization and upgrading the level of agricultural development of regions with different levels of development within the seven main agricultural production areas and others.

Terraces play a crucial role in regulating the water cycle. The management and restoration of, and the conservation strategies and plans for basins rely heavily on the assessment of alterations in hydrological characteristics, whether natural or man-made, within these basin systems. This study centers on evaluating the impact of terraces on temporal variations in the hydrological regime within the Laixi River basin, within the context of climate variability. To achieve this, we employed a hydrological model and applied the Indicators of Hydrologic Alteration and Range of Variability Approach (IHA–RVA). The results show that, under the influence of terraces, the annual flow exhibited an overall decreasing trend. There was a noticeable decrease in streamflow from March to July, while from October to December, there was a clear upward trend, with increases ranging from 4.76% to 8.10% at the Guanshenxindi (GSXD) and Watershed outlet (WO) cross-sections. The indicators related to the minimum flow showed an overall increasing trend, whereas all indicators related to the maximum flow exhibited a clear decreasing trend. Both high and low flow pulse counts decreased, while high and low pulse durations increased overall. The overall trend of flow reversal counts also showed a decrease. All hydrological indicators exhibited low levels of alteration, with hydrologic alteration values lower than 33.33%. However, the influence of terraces on the upstream hydrological conditions was more pronounced than their downstream effect. The terraces in the study area have played a beneficial role in the flood regulation, drought mitigation, and water resource conservation of the Laixi River. They have contributed to stabilizing the daily average flow during high flow periods and have had a positive impact on the ecological base flow during low flow periods, thereby improving the aquatic ecological environment. This study provides theoretical support for river regulation within the framework of soil and water conservation measures.

The concept of synergistic water resources, water environment, water ecology remediation, and restoration (3WRR) is essential for addressing the interlinked challenges of water scarcity, pollution, and ecological degradation. An intelligent platform of remediation and restoration project optimization was developed, integrating multi-source data fusion, a coupled air–land–water model, and dynamic decision optimization to support 3WRR in river basins. Applied to the Jinshan Lake Basin (JLB) in China’s Greater Bay Area, the platform assessed 894 scenarios encompassing diverse remediation and restoration plans, including point/non-point source reduction, sediment dredging, recycled water reuse, ecological water replenishment, and sluice gate control, accounting for inter-annual meteorological variability. The results reveal that source control alone (95% reduction in point and non-point loads) leads to limited improvement, achieving less than 2% compliance with Class IV water quality standards in tributaries. Integrated engineering–ecological interventions, combining sediment dredging with high-flow replenishment from the Xizhijiang River (26.1 m³/s), increases compliance days of Class IV water quality standards by 10–51 days. Concerning the lake plans, including sluice regulation and large-volume water exchange, the lake area met the Class IV standard for COD, NH₃-N, and TP by over 90%. The platform’s multi-objective optimization framework highlights that coordinated, multi-scale interventions substantially outperform isolated strategies in both effectiveness and sustainability. These findings provide a replicable and data-driven paradigm for 3WRR implementation in complex river–lake systems. The platform’s application and promotion in other watersheds worldwide will serve to enable the low-cost and high-efficiency management of watershed water environments.

Aiming to address the limitations in accuracy and reliability of land use classification models when encountering complex and variable land use features, a novel method for land use classification (4D-SOMKS) is developed, which constructs a low redundancy four-dimensional (4D) spectral feature system with biophysical meaning of distinct land surface properties and organizes pixel-level features in a topology-preserving space using the Self-Organizing Map (SOM), further groups the SOM neurons into spectrally coherent clusters through K-means, and uses a small number of labeled samples only for semantic assignment of the resulting clusters rather than for pixel-level supervised model training. Empirical research in the Bayannur and Hong Lake Basins (HLB) have revealed the driving role of key spectral indices in classification. High overall classification accuracies were achieved, reaching 98.45% and 98.55% respectively, with robust performance across evaluation metrics including precision, recall, F1-score, and the Kappa coefficient. The results show that 4D-SOMKS achieves high accuracy and robustness while significantly reducing reliance on large-scale labeled data, providing an effective avenue to improve the accuracy and reliability of land use classification under spatiotemporal dynamic changes.