Abstract: Fine particulate matter (PM_2.5) constitutes a significant component of air pollution.Investigating the impact of land use changes on PM_2.5 concentrations holds considerable importance.This study employs spatial autocorrelation,Pearson correlation analysis,and geographically weighted regression （GWR） to examine the relationship between PM_2.5 concentration fluctuations and land use transformations from 2000 to 2020 in Gansu Province.The results indicate that the predominant land use categories were cultivated land,grassland,and unused land.Over the two decades,there was a notable reduction in the area of cultivated and unused land,while water bodies and construction land areas expanded.The spatial distribution of PM_2.5 concentrations exhibited alternating peaks in the northwest and southeast regions.Temporally,the annual average concentration initially increased before decreasing.PM_2.5 concentrations responded significantly to land use changes,with grassland,forest land,and unused land exerting more substantial impacts compared to cultivated land.Specifically,the influence of different land use types on PM_2.5 concentrations,ranked by degree,is as follows:forest land,unused land,grassland,and cultivated land.Notably,increases in forest and grassland areas positively contributed to reducing PM_2.5 concentrations.