Evaluation of crop-water mutual suitability and optimization of water-adaptive cropping for dryland farming in Loess Plateau

The intensified impacts of climate change, declining water resources, and rising food demand pose significant challenges to rainfed agriculture in the arid and semi-arid regions of China. The Loess Plateau, as a representative area, has long suffered from water scarcity, low water use efficiency, and mismatch between cropping structure and water resource distribution. To address these issues and promote high-quality, sustainable agricultural development, this study evaluated the crop-water mutual suitability to explore the strategies for water-adaptive cropping optimization. The research employs the following approaches: 1) Cropping structure on the Loess Plateau (2017–2022) were identified using an integrated phenology and machine learning approach. The resulting maps demonstrated high overall accuracies (0.83), showing strong agreement with municipal statistical data (R² ≥ 0.76). 2) The effective water supply and crop water demand were quantified using remote-sensing-based water balance assessment tool (RWBAT). A Crop Water Suitability (CWS) index was developed to quantitatively evaluate crop-water suitability across the region. The analysis revealed suboptimal water suitability (CWS < 0.35) in the central double-cropping zones and the hilly-gully regions of the Plateau. 3) A tightly coupled multi-objective optimization framework, integrating RWBAT model, was designed to optimize cropping structures. Compared to 2022 baseline condition, the optimization results indicated 25.6% improvement in crop water use efficiency and 5.3% increase in net income from crop production. The research results provide scientific basis and operational solutions for crop water resources management in the Loess Plateau and even similar arid areas, and provide an important reference for coping with climate change and agricultural water resources crisis.