Water scarcity is a significant constraint in agricultural ecosystems of arid regions, necessitating sustainable development of agricultural water resources. This study innovatively combines Bayesian theory and Water Footprint (WF) to construct a Bayesian Network (BN). Water quantity and quality data were evaluated comprehensively by WF in agricultural production. This evaluation integrates WF and local water resources to establish a sustainability assessment framework. Selected nodes are incorporated into a BN and continuously updated through structural and parameter learning, resulting in a robust model. Results reveal a nearly threefold increase of WF in the arid regions of Northwest China from 1989 to 2019, averaging 189.95 × 10⁸ m³ annually. The region's agricultural scale is expanding, and economic development is rapid, but the unsustainability of agricultural water use is increasing. Blue WF predominates in this region, with cotton having the highest WF among crops. The BN indicates a 70.1 % probability of unsustainable water use. Sensitivity analysis identifies anthropogenic factors as primary drivers influencing water resource sustainability. Scenario analysis underscores the need to reduce WF production and increase agricultural water supply for sustainable development in arid regions. Proposed strategies include improving irrigation methods, implementing integrated water-fertilizer management, and selecting drought-resistant, economically viable crops to optimize crop planting structures and enhance water use efficiency in current agricultural practices in arid regions. This study not only offers insights into water management in arid regions but also provides practical guidance for similar agricultural contexts. The BN model serves as a flexible tool for informed decision-making in dynamic environments.

Recent studies on China’s arid and semi-arid regions, particularly the Tarim River Basin (TRB), have shown an increase in the intensity and frequency of extreme weather events. This research examines the link between meteorological droughts, as measured by the Standardized Precipitation Evapotranspiration Index (SPEI), and hydrological droughts, as indicated by the Standardized Runoff Index (SRI) and the Standardized Terrestrial Water Storage Index (STI), over various time scales. Historical data indicate that SPEI drought frequency (DF) was 14.3–21.9%, with prevalent events in the northern oases. SRI DF ranged from 9.0% to 35.8%, concentrated around the Taklamakan and Kumtag Deserts, while STI DF varied between 4.4% and 32.7%, averaging 15% basin-wide. Future projections show an increased DF of SPEI in deserts and a decrease in oases; SRI DF decreased in deserts but increased in oases. STI changes were more moderate. The study also found a higher risk of drought progression from SPEI to SRI in the southwestern and northeastern oases, exceeding 50% probability, while central and eastern TRB had lower risks. The western TRB and inner Taklamakan Desert faced higher risks of SPEI to STI progression, with probabilities over 45%, in contrast to the lower risks in the eastern and central oases. The concurrence of SRI/STI with moderate to extreme SPEI droughts led to a higher probability and area of SRI/STI droughts, whereas consistent SPEI types showed a reduced induced probability and extent of SRI/STI droughts. This study enhances the understanding of drought propagation from meteorological to hydrological droughts in the TRB and contributes to the prevention of hydrological drought to a certain extent.

Grassland ecosystems in arid regions increasingly experience resilience loss due to intensifying climatic variability. However, the limited interpretability of conventional machine learning models constrains our understanding of underlying ecological drivers. This study constructs an integrative framework that combines temporal autocorrelation (TAC) metrics with explainable machine learning, employing Random Forest and SHAP (SHapley Additive exPlanations) analysis. Time series of satellite-derived vegetation indices from MODIS (2001–2023), particularly the kernel Normalized Difference Vegetation Index (KNDVI), support the generation of TAC and its trend-based derivative δTAC. The framework assesses ecosystem resilience across seven representative grassland types in Xinjiang, capturing diverse responses to climate variability and vegetation dynamics. Results reveal pronounced spatial heterogeneity: resilience declines in radiation-stressed arid zones, while hydrothermally stable regions maintain stronger recovery capacity. Key drivers include temperature variability and vegetation dynamics, with divergent effects among grassland types. Meadow and Typical Steppe exhibit higher resilience under stable hydrothermal regimes, whereas desert and alpine systems show greater sensitivity to warming and climatic fluctuations. This framework enhances diagnostic transparency and ecological insight, offering a spatially explicit, data-driven tool for resilience monitoring. The findings support the formulation of targeted adaptation strategies and sustainable grassland management in response to ongoing climate change.

Water resources are a key constraint on sustainable development in arid regions, especially for agricultural production where water use is intensive. To assess the sustainability of agricultural water use in such environments, this study utilizes 2010–2020 agricultural data from the Turpan–Hami Basin and is among the first to integrate the water footprint (WF) theory with the DPSIR (driver–pressure–state–impact–response) model into a comprehensive framework for evaluating water resource sustainability in arid agricultural systems. The agricultural blue, green, and grey WF in the Turpan–Hami Basin were quantified for 2010–2020, followed by a sustainability assessment under the DPSIR framework using a comprehensive weighting method. The results showed a continuous increase in the WF, dominated by the blue WF (>60%), largely due to crops like cotton and grapes, intensifying regional water stress. Turpan experienced prolonged resource overload, while Hami exhibited slightly higher sustainability. DPSIR analysis revealed stronger policy responses in Turpan and significant ecological investments in Hami. Key influencing factors included the crop yield, WF modulus, per capita WF, and water quality shortage index. Overall, sustainability in the basin fluctuated between “Basically Sustainable (Level III)” and “Insufficiently Sustainable (Level IV)”, with slight improvement in 2020. The findings highlight the need for region-specific agricultural optimization, strengthened ecological governance, and improved water-saving strategies to enhance water use efficiency and sustainability in arid regions.