Bridging data gaps in urban heat forecasting: Remote sensing and image processing approach for Qatar’s climate

Traditional Land Use Regression (LUR) methods for environmental prediction are limited by costly and time-intensive ground data collection. This study introduces a novel, remote-sensing-based framework that eliminates the need for extensive in-situ measurements by leveraging Google Earth Engine (GEE) and Python image processing to model and forecast Land Surface Temperature (LST) across Qatar. Using a decade of MODIS satellite data (2013–2023), daily, monthly, and annual average temperature maps were generated, processed with Sobel gradient and binary thresholding algorithms to visualize spatial variability and compliance with Qatar’s Heat Stress Legislation. A pixel-wise temporal regression model predicted LST distributions for 2024, achieving a validation RMSE as low as 1.6°C and a maximum of 6.36°C under high-variability conditions. The workflow accurately filled missing data and identified persistent “cool islands” associated with vegetation and water bodies. This open-source approach establishes a computationally efficient alternative to LUR, capable of extending to pollutant or air-quality prediction, and offers a scalable tool for urban climate resilience and heat-stress management. This study addresses a practical gap in urban heat forecasting for hot-arid, data-sparse regions by introducing a transparent workflow that combines remotely sensed land surface temperature (LST), image-processing operations, and pixel-wise temporal regression to identify, interpret, and forecast urban thermal patterns in Qatar. Rather than replacing conventional urban climate models, the proposed approach provides a computationally accessible baseline methodology for engineering applications where dense in-situ data are limited. The added value lies in translating satellite-derived thermal information into interpretable spatial indicators and predictive outputs that can support heat-risk screening, urban environmental assessment, and climate-responsive planning in arid cities.