Assessing urban heat island risk during heatwaves with satellite data and fuzzy clustering techniques

This is the authors' abstract. We don't add key points for this paper.

This paper presents a methodology for mapping urban heat island risk generated by heat waves, based on Landsat 8 satellite data and unsupervised artificial intelligence classification techniques. The integration of the land surface temperature (LST) raster and the vegetation index (NDVI) allows the identification of urban areas with high thermal exposure (hot spots), the most resilient areas (cold spots), and the estimation of the resident population at risk. The tests were carried out on the city of Naples during heat waves that occurred in the last three years from 2022 to 2024. The results show that the proposed framework can provide an operational tool for decision makers to monitor heat islands and plan climate-proof solutions in the most critical urban contexts. The proposed methodology is portable and transferable to other urban contexts.