Agricultural Land Suitability Evaluation in the Aksu Region Under Future Climate Scenarios Using the MaxEnt Model

Scientific and rational agricultural land suitability assessment is essential for improving land resource use efficiency and promoting sustainable agricultural development. In arid and ecologically fragile regions, agricultural production is jointly constrained by climatic conditions, water availability, soil properties, and existing land-use patterns. This study employed the Maximum Entropy (MaxEnt) model to evaluate the spatial distribution and future evolution of agricultural land suitability in the Aksu region from 2021 to 2080 under three future climate scenarios, namely SSP-126, SSP-245, and SSP-585. The evaluation framework integrated climatic variables, topographical factors, soil properties, land use, and hydrological conditions, while climatic indicators were further screened through correlation analysis. The results showed that the MaxEnt model achieved high predictive accuracy and stability across different climate scenarios and periods, with AUC values for both the training and testing datasets consistently exceeding 0.90, indicating good model applicability and discriminatory ability. Feature importance analysis revealed that land use, temperature-related variables (bio1, bio4, and bio10), precipitation seasonality (bio15), and soil factors were the main variables influencing agricultural land suitability. These findings highlight the important roles of land-use patterns, thermal conditions, precipitation stability, and soil properties in determining agricultural suitability in arid oasis regions. Furthermore, future climate scenarios did not substantially alter the overall spatial distribution pattern of agricultural land suitability, but more pronounced local improvements and degradation were observed under higher-emission scenarios. This study provides insights into future changes in agricultural land suitability and offers scientific support for climate-adaptive agricultural management and sustainable land resource utilization in arid regions.