Urban heat island mitigation in Tehran: District-based mapping and analysis of key drivers

Urban Heat Islands (UHIs) result in higher urban temperatures than rural areas, increasing energy use for cooling, degrading air quality, and worsening heat-related health issues. Although previous studies have explored UHIs in Tehran, a comprehensive analysis remains lacking. This study adopts a macro-scale approach, combining linear, non-linear, and location-based methods to investigate key factors influencing UHI intensity (UHII) in detail. It uses remote sensing and machine learning to analyze UHIs in Tehran by examining Land Surface Temperature and Land Cover data within GIS. The study compares urban temperatures with those in surrounding rural areas and identifies key factors influencing UHIs through ordinary least squares (OLS), random forest (RF), and geographically weighted regression. The findings reveal substantial UHII variations across Tehran, with central districts showing temperature differences greater than 5 °C compared to rural areas. The OLS model identifies aerosol optical depth (0.38), population density (0.45), and Normalized Difference Built-up Index (NDBI) (0.17) as contributors to higher UHII, while relative humidity (-0.51) and terrain slope (-0.22) are linked to lower UHII. The RF model, with a Mean Squared Error of 0.33 and an R-squared value of 0.84, highlights population density, relative humidity, and NDBI as primary factors influencing UHII. Results can inform planning and design strategies to foster climate change adaptation and enhance urban resilience.