Assessing Climate and Land-Use Change Scenarios on Future Desertification in Northeast Iran: A Data Mining and Google Earth Engine-Based Approach

Desertification poses a significant threat to dry and semi-arid regions worldwide, including Northeast Iran. This study investigates the impact of future climate and land-use changes on desertification in this region. Six remote sensing indices were selected to model desertification using four machine learning algorithms: Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting Machine (GBM), and Generalized Linear Models (GLM). To enhance the model’s reliability, an ensemble model was employed. Future climate and land-use scenarios were projected using the CNRM-CM6 model and Markov chain analysis, respectively. Results indicate that the RF and SVM models performed best in mapping current desertification patterns. The ensemble model highlights a 2% increase in decertified areas by 2040, primarily in the northwestern regions. The study underscores the importance of land-use change and climate change in driving desertification and emphasizes the need for sustainable land management practices and climate change adaptation strategies to mitigate future impacts.