The study of the spatial and temporal evolution of annual mean PM2.5 concentration and population exposure risk in Hunan Province can further analyze the influence of the landscape pattern index on the changes in annual mean PM2.5 concentration and population exposure risk. The spatial and temporal evolution of the coupling relationship between people, air, and land at a fine scale and its spatial heterogeneity were investigated by the population exposure risk model, the moving window method, the spatial auto correlation model, and correlation analysis. The results show that: (1) from 2000 to 2020, the annual average PM2.5 concentration in Hunan Province showed a slow increase and then a significant decrease, and its spatial distribution was high in the northeast and low in the southwest. (2) From 2000 to 2016, the average proportion of the population exposed to high annual average PM2.5 concentrations (>45 μg/m3) in Hunan Province reached 74.24%, and from 2016 to 2020, the proportion of the population exposed to low annual average PM2.5 concentrations (<35 μg/m3) increased year by year. In 2020, there were regions with annual average concentrations of 15-25 μg/m3, and the regional concentrations were all less than 45 μg/m3. The spatial distribution is dominated by low-risk areas, with an overall scattered distribution, and high-risk areas are concentrated in the Chang - Zhu - Tan urban agglomeration. (3) Cultivated land and forested land, as the main land types, have relatively obvious characteristics of changes in dynamics and attitudes; water bodies and impervious surfaces are second; cultivated land has a high degree of patch aggregation, high edge density and low fragmentation, which increases the risk of annual average PM2.5 concentration and population exposure; forested land has a high degree of patch aggregation, low edge density and low fragmentation, which is conducive to reducing the risk of annual average PM2.5 concentration and population exposure; and water bodies and impervious surfaces have a high degree of aggregation, low edge density, and low fragmentation, which is conducive to reducing the risk of annual average PM2.5 concentration and population exposure. The higher the degree of aggregation, the higher the edge density, the more complex the degree of fragmentation of water bodies and impervious surfaces, and the higher the risk of annual average PM2.5 concentration and population exposure. The results of this study can provide a theoretical basis for mitigating air pollution, improving the human environment, optimizing the landscape pattern, and promoting ecologically sustainable development.