The identification of land use/land cover (LULC) changes is important for monitoring, evaluating, and preserving natural resources. In the Kurdistan region, the utilization of remotely sensed data to assess the effectiveness of machine learning algorithms (MLAs) for LULC classification and change detection analysis has been limited. This study monitors and analyzes LULC changes in the study area from 1991 to 2021 using a quantitative approach with multi-temporal Landsat imagery. Five MLAs were applied: Support Vector Machine (SVM), Random Forest (RF), Artificial Neural Network (ANN), K-Nearest Neighbor (KNN), and Extreme Gradient Boosting (XGBoost). The results showed that the RF algorithm produced the most accurate maps of the three-decade study period, accompanied by a high kappa coefficient (0.93–0.97) compared with the SVM (0.91–0.95), ANN (0.91–0.96), KNN (0.92–0.96), and XGBoost (0.92–0.95) algorithms. Consequently, the RF classifier was implemented to categorize all obtainable satellite images. Socioeconomic changes throughout these transition periods were revealed by the change detection results. Rangeland and barren land areas decreased by 11.33 % (−402.03 km²) and 6.68 % (−236.8 km²), respectively. The transmission increases of 13.54 % (480.18 km²), 3.43 % (151.74 km²), and 0.71 % (25.22 km²) occurred in agricultural land, forest, and built-up areas, respectively. The outcomes of this study contribute significantly to LULC monitoring in developing regions, guiding stakeholders to identify vulnerable areas for better land use planning and sustainable environmental protection.

Changes in land use/land cover (LULC) are a substantial environmental subject with considerable consequences for human well-being, climate, and ecosystems. Innovative investigations for predicting LULC changes are essential for effective land management and sustainable development. This study used Landsat images and supplementary spatial factors to evaluate spatiotemporal LULC changes in Erbil Province, Kurdistan Region-Iraq. It predicts future changes in 2040 using four climates scenario-based Shared Socioeconomic Pathways (SSPs). The Random Forest (RF) model was used to classify and forecast LULC changes, which are crucial for effective land management and sustainable development. The RF model was assessed using performance metrics, such as the overall accuracy, F1-score, and kappa coefficient. The simulated LULC outcomes demonstrated the efficiency of the selected model, achieving an overall accuracy of 94.34%, a perfect agreement in the kappa statistic of 0.92, and a high F1-score between 0.71 and 0.93. The study revealed that agricultural land declines under SSP126 but increases under other scenarios, with SSP585 showing the highest gain (+209.98 sq. km, 23.32%). Barren land increased across all scenarios, whereas built-up areas consistently increased. Forest gains in SSP126 but declined in all other scenarios, with the most significant loss in SSP585 (−101.20 sq. km, −5.31%). The riparian zone gains in SSP126 but declines in all the other scenarios. Snow remained stable, but minor losses were observed in SSP245, SSP370, and SSP585. Water showed a slight increase in SSP126 but declined in all other scenarios. SSP126 showed minor changes, whereas SSP scenarios 370 and 585 show severe land transformations, forest loss, rangeland degradation, and urban expansion, indicating increased deforestation and degradation. This study highlights the importance of integrating a scenario-based RF model with hyperparameter tuning in remote sensing applications to improve LULC dynamics predictions, benefiting land-use planning, environmental management, and rational decision-making.