A geospatial and statistical assessment of surface temperature response to land cover change and spatial drivers in Sekondi-Takoradi, Ghana

Rapid urbanization is profoundly transforming land use and surface thermal environments in West African cities, with implications for urban heat island (UHI) intensification and climate resilience. Unfortunately, the influence of urban growth and spatial determinants on surface temperature variability remains underexplored in Sekondi–Takoradi and across Ghana. This research addresses this gap by employing geospatial techniques and statistical models to assess the influence of land cover change from 1991 to 2023 and spatial drivers on temperature distribution in the Sekondi-Takoradi Metropolis. Landsat images from 1991, 2009, 2016, and 2023 were classified using the random forest algorithm to map LULC transitions, while spectral indices, land surface temperature (LST), elevation, and proximity variables were employed to assess the spatial thermal dynamics. Statistical modeling combined Ordinary Least Squares (OLS) and Geographically Weighted Regression (GWR) to evaluate the relationships between LST and explanatory variables, and urban archetype analysis was applied to examine the role of spatial drivers on temperature variability. The findings revealed a 32.91% (63.07 km2) expansion of built-up, a 1.09% (2.08 km2) decline in water, and a 32.82% (60.99 km2) decline in vegetation cover, leading to a 3.1 °C rise in mean LST. Regression results showed that GWR consistently outperformed OLS, highlighting the importance of spatial heterogeneity in explaining urban thermal dynamics. The distinct spatial factors defining each class revealed that archetype 6 recorded the lowest mean temperature, whereas archetype 1 exhibited the highest. These results highlight the dominant role of urban growth and spatial drivers in shaping UHI intensification. The study provides critical insights for sustainable land use planning, urban greening, and climate-sensitive infrastructure development, offering policy-relevant evidence to support Sustainable Development Goal 11 on sustainable cities and communities.