Abstract
Understanding the spatiotemporal distribution of land surface temperature (LST) is crucial for managing urban thermal environments and mitigating urban heat island (UHI) effects. This study addresses the challenge of quantifying the complex interactions among natural and anthropogenic factors driving LST variations, while leveraging advanced modeling techniques to predict future thermal risks in rapidly urbanizing regions. By analyzing the evolution of LST in Jinan city, China, from 2002 to 2022, and forecasts future trends using advanced spatial analysis and predictive modeling techniques. Directional shifts in LST were quantified using the quadrant azimuth method and the standard deviation ellipse method, both of which analyze spatial distribution and dispersion. To identify the key drivers of LST variations, 14 socioeconomic and environmental factors were assessed using the optimal parameter-based geographical detector (OPGD) model, which effectively handles spatial heterogeneity. Key findings include: (1) a significant northward shift in the LST centroid and a 26.64% expansion in high-temperature areas, with noticeable cooling effects in the city center. (2) A nonlinear relationship between LST and socioeconomic factors, particularly GDP, where cooling effects were observed when GDP exceeded 10,000 yuan/km(2). (3) Synergistic interactions, especially between topographic factors (such as the Digital Elevation Model, DEM) and land-use indices (e.g., normalized difference built-up index, NDBI; normalized difference vegetation index, NDVI), were found to significantly influence LST variations. (4) The Oscillating Sequence Grey Model (OSGM), optimized for handling oscillating data sequences, demonstrated superior predictive accuracy, projecting a 20.72% increase in extreme high-temperature zones and a 40.61% reduction in moderate-high-temperature zones by 2047. These findings offer actionable strategies for urban planning and climate adaptation, aiming to mitigate thermal risks and inform future policies for urban sustainability and resilience. This research underscores the importance of integrating spatial and predictive analyses to inform urban planning and climate adaptation strategies, contributing to the mitigation of thermal risks and the development of sustainable urban policies.