Abstract: Rural areas have experienced significant changes in land use/land cover (LULC) during urbanization, resulting in alterations of the surface thermal environment. However, the relationship between rural land surface temperature (LST) and landscape patterns at different scales of analysis remains poorly elucidated. This study scrutinizes rural regions of the Fuchun River Basin by employing correlation analysis, spatial autocorrelation analysis, and geographically weighted correlation coefficient analysis. The investigation aimed to discern the impacts of land-use landscape patterns on LST during the summers of 1990 and 2020 across multiple scales of analysis (300, 600, 900, 1200, 1500, 1800, 2100 m, and township levels). The results showed that: (1) From 1990 to 2020, cultivated land and forest land were the predominant land use types. There was a substantial reduction in cultivated land area over 30 years, primarily due to the conversion to industrial land, urban construction land, and rural residential land. (2) Zones with moderate and severe temperature increases concentrated in urban construction land, rural residential land, and industrial land clusters along both banks of the Fuchun River. Furthermore, the distinctiveness of LST changes diminished as the scale of analysis units increased. (3) The degree of correlation between changes in landscape pattern indices and LST among various land types follows the order: urban construction land > industrial land > cultivated land > water bodies > rural residential land > forest land, exhibiting different trends with an increase in the scale of analysis. The percent of landscape (PLAND) index exerted the strongest impact on LST. The rise in local LST could be attributed to decreased area of cultivated land and increased area of urban construction land and industrial land. The increase of rural residential land had a relatively mild impact on LST. The change in forest land was not significant, but it alleviated the surface thermal environment. (4) There was a clear spatial autocorrelation and spatial spillover effect between landscape pattern index of urban construction land and industrial land and LST. These findings contribute to understanding the impact of LULC changes on LST in rural areas. Moreover, the findings furnish a theoretical foundation for mitigating the adverse effects of urbanization on rural environments.

Key words: scale effect, surface thermal environment, land use and land cover change, spatial spillover effect, sustainable development