Multi-temporal remote sensing and geospatial analysis for urban ecosystem service dynamics: A three-decade assessment of land surface transformation in Jeddah, Saudi Arabia

Abstract

The rapid transformation of Earth's surface through urbanization presents critical challenges for ecosystem sustainability and climate resilience. This study employs advanced remote sensing and geospatial technologies to monitor and assess the spatiotemporal dynamics of urban ecosystem services in Jeddah, Saudi Arabia, over three decades (1993–2023). Using multi-temporal Landsat imagery analyzed through Support Vector Machine algorithms in Google Earth Engine, we achieved high-accuracy (>85%) land use/land cover classification to quantify urban expansion patterns and their impact on ecosystem service values (ESVs). The analysis revealed significant urban intensification, with built-up areas expanding by 145.70 km², resulting in the conversion of 130.74 km² of barren soil, 7.14 km² of vegetation, and 7.71 km² of water bodies. This transformation led to a substantial reduction in ESVs totaling $477.48 million, with the most significant impacts on hydrological regulation (-$114.25 million), waste treatment (-$97.05 million), and biodiversity protection (-$72.78 million) services. The spatiotemporal analysis demonstrated clear patterns of ecosystem service degradation, particularly in the city's central regions. Our findings provide crucial insights for achieving UN Sustainable Development Goals (Target 11.1, 6.6., 13.1, 3.9 and 15.1) by quantifying the environmental costs of rapid urbanization and informing evidence-based urban planning strategies. The study's innovative integration of remote sensing, machine learning, and ecosystem service valuation offers a robust framework for monitoring urban ecosystem dynamics and supporting sustainable urban development in rapidly growing arid regions.