<![CDATA[This study aimed to analyze Land Use/Land Cover (LULC) dynamics in the Da River basin, approximately 52,000 km2, during the period 2015–2025 and to predict future changes up to 2035. Landsat-8 imagery processed on the Google Earth Engine (GEE) platform was used to generate multi-temporal LULC maps. A total of 6 LULC classes were classified using the Random Forest (RF) algorithm. The classification results reported high accuracy, with Overall Accuracy ranging from 91% to 94% and Kappa coefficients between 0.86 and 0.90. Furthermore, an integrated Cellular Automata–Artificial Neural Network (CA-ANN) model was used to simulate future LULC change up to 2035. This model integrated spatial influencing factors such as distance to roads, distance to rivers, elevation, slope, and neighborhood effects. Validation suggested excellent model performance with training/validation error < 0.02. The results showed that there was a continuous urbanization trend, with the built-up area projected to increase from 5.35% to 9.15% of the basin area by 2035, while Forest and Agricultural land were expected to decline. Bare land, Water bodies, and Other land exhibited relatively minor changes. These provided important information for sustainable land-use planning, watershed management, flood-risk mitigation, and environmental protection in the Da River basin. Moreover, CA-ANN model performed better than traditional statistical approaches because the concept captured nonlinear spatial processes and neighborhood interactions. These results supported land-use planning and River basin management, particularly for flood risk mitigation and environmental protection in the Da River basin.]]>
