Seasonal and Multiscale Associations Between Surface-Water Fraction and Daytime Land Surface Temperature in the Wuhan Urban Agglomeration, China

Surface water is an important component of urban land-cover systems, but its relationship with land surface temperature (LST) is difficult to interpret in lake-rich urban agglomerations where water bodies, shorelines, vegetation, built-up land, and terrain are closely interwoven. This study examined the association between surface-water fraction and daytime LST in the Wuhan Urban Agglomeration, China, across six benchmark years from 2000 to 2025. MODIS daytime LST, Landsat-derived water maps, GHSL built-up data, vegetation indices, and terrain variables were integrated on a 250 m grid. Surface-water fraction was measured within each grid cell and surrounding windows from 250 to 2000 m. Summer and winter models were compared, and between-location differences were separated from within-location water-fraction changes. Spatial validation, city-specific models, and robustness checks were used to assess consistency. Higher surface-water fraction was associated with lower daytime LST in both seasons. The strongest standardized association occurred at 500 m in summer, whereas the winter association strengthened toward 2000 m. Between-location differences were much larger than within-location changes, and the negative association was retained across all cities. These results suggest that surface-water fraction should be interpreted as a scale- and season-dependent land-cover indicator rather than as a fixed cooling measure.