Why Is Iran Drier and Hotter than Its Neighbors? A Comparative Climate–Topography Analysis with Arizona USA Using Machine Learning

Core Problem:Iran, located in the subtropical dry belt of the Northern Hemisphere, experiences markedly lower precipitation and higher temperatures than neighboring countries, threatening water security, agriculture, and urban sustainability. Similar challenges exist in parts of the United States, particularly arid states like Arizona, where droughts, reduced snowpack, and extreme temperatures impact water resources and agriculture. Despite regional studies, the combined influence of topography, atmospheric circulation, and anthropogenic factors driving Iran’s extreme aridity has not been systematically compared with analogous regions in North America. Understanding these dynamics provides insight into dryland hydrology, climate adaptation, and sustainable water management globally.Approach:This study integrates climate and topographic data with machine learning models to analyze aridity drivers in Iran and Arizona. Data sources include ERA5 reanalysis, MODIS snow cover, precipitation and temperature records, land-use datasets, and digital elevation models (DEMs). Supervised ML models (Random Forest, Gradient Boosting, Long Short-Term Memory networks) quantify the relative contributions of topography, atmospheric patterns, and human activity to aridity. Model interpretability is enhanced using SHAP values and partial dependence plots.Key Findings:Results indicate that mountain-induced rain shadow effects, coupled with temperature anomalies and land-use changes, largely explain Iran’s low precipitation. Comparisons with Arizona reveal both regions share drivers of aridity, yet Iran’s latitude, elevation, and atmospheric blocking create more extreme dryness. ML models achieved high predictive performance (R² > 0.88) and identify areas most vulnerable to drought.Impact:These findings provide a transferable framework for water resource management, drought mitigation, and climate adaptation in arid regions worldwide. By integrating topography-driven analysis with machine learning, the study enhances understanding of extreme dryness, informing both regional policy and global scientific insights.