In this study， the physical PM2.5 remote sensing (PMRS) method was used to invert the FY-4A observation data, and through the fusion with the surface station data, the near-surface PM2.5 gridding observation data in China were obtained. A grid map of the relationship between PM2.5 and PM10 was established under different conditions of clean, haze and dust background, and PM10 concentration in China was further estimated. The results showed that there is a similar correlation between columnar volume-to-extinction ratio of fine particulates (VEf) and fine mode fraction (FMF) in different regions. With FMF=0.4 as the boundary, two fitting equations of VEf are established respectively. The estimated correlation (r2) between VEf and AERONET observation is higher than 0.85. During the haze period, the PM2.5 mass concentration obtained by PMRS has a good match with the observation site distribution, which can basically reflect the high concentration area of pollution. In single point test, the remote sensing result has a certain correlation with the surface observation, with r2 reaching 0.39 in the region of Beijing-Tianjin-Hebei. However, to some extent there is still overestimation or underestimation in the value of inversion compared to surface observation data. In order to solve this problem, the study combines the satellite inversion grid results with the surface observation site, and finally obtains a 0.25°×0.25° national PM2.5 gridding observation data that are in good agreement with the surface observation data. The r2 between PM10 and PM2.5 is higher than 0.7 under different weather conditions. Among them, the value PM10/PM2.5>3 matches with the sand source area and the high dust area in China during the dust period. During the haze period, PM10/PM2.5 is close to 1, and the value is between 1 and 2 during cleaning period for most stations. Through the grid map of the relationship between PM2.5 and PM10, according to different weather backgrounds, the gridded surface PM10 in China is obtained, and the results can better reflect the dust process in Northwest China.