Abstract: Under the positive human land use regulation (e.g. afforestation), understanding the impacts of landscape pattern changes on urban and rural thermal fields can provide reference data for future construction and optimization of urban afforestation projects and bring new insights for future landscape patterns and thermal fields environment related research. Under the context of “One Million Mu Plain Afforestation Project” in Beijing plain area, we applied remote sensing technology and geographic information system and used transfer-matrix method and coupling analysis to evaluate the impacts of plain afforestation on thermal field environment and to explore the correlation between changes in landscape patterns resulted from urban afforestation projects and urban land surface temperature. Our results showed that the percentage of landscape (PLAND) of forests and parks increased by 7.6% and 0.5% after the afforestation project in Beijing plain area, respectively. The heat island intensity decreased by 19.2% of entire plain area, and by 23.3% of suburban area, however, it increased by 23.5% of city area. Spanning urban to urban-rural interface, the PLAND of forests and parks, the mean patch size (MPS) of forests and wetlands were significantly negatively correlated with the changes in landscape surface temperature (LST). In city area, the area-weighted mean shape index (AWMSI) of forests was significantly positively correlated with the changes of LST. In suburban area, the MPS and AWMSI of forests showed significantly negative correlation with changes of LST, which was contrary to parks. After the implementation of afforestation project in Beijing plain area, the increased urban forests and urban parks had mitigated thermal fields to some extent. The scale of mitigation was limited and most areas with decreased heat island intensity were concentrated in newly planted sites in suburban area. The more connected the forests and wetlands, the better the relief of the thermal fields in the plain area. Impacted by the grain size, time, topography, human activities and other factors, the change in landscape patterns caused by the control of land use structure showed non-linear correlation with the change of LST.