Abstract

The Urban Heat Island (UHI) effect can intensify the impacts of climate change on human health, underscoring the need for accurate urban mapping to understand the influence of Local Climate Zones (LCZs) on the UHI phenomenon. This study investigates the application of foundation models for mapping LCZs and analyzing UHIs. Geo-spatial foundation models, pre-trained on vast datasets using advanced machine learning and deep learning algorithms, show significant potential for improving the accuracy of LCZ classification from satellite imagery. By utilizing these models' ability to process large and diverse datasets, this research seeks to enhance the detection of fine-scale urban variations, resulting in more precise UHI assessments. These insights can support policymakers and urban planners in advancing climate resilience and sustainable urban development, ultimately contributing to the mitigation of urban heat stress and fostering better climate adaptation strategies.