Monitoring urban dynamics using Google Earth and GeoAI

Urban areas face mounting pressure from increased space demand, degrading key environmental services. Thus, understanding Land Use/Land Cover (LULC) changes is vital. In order to offer a robust decision-support framework for urban planning and environmental conservation, this study presents an innovative measurement method based on Google Earth Engine and Unsupervised K-means Clustering of multispectral satellite images to map urban and vegetation shifts. The proposed method was applied in 15 southern Italian cities and the results were validated with ESA Land Cover dataset. Results show 167 hectares consumed from 2005 to 2021. The proposed unsupervised classification achieved favorable F1-scores, with 0.64 for urban areas and 0.92 for vegetation, demonstrating strong performance despite the challenges of classifying diverse 30 m Landsat land cover types. For these reasons, these results show the potential to make the proposed method a useful tool for aiding policymakers and urban planners in making informed decisions to mitigate the adverse effects of urban growth on the environment.