Limited by the number of ground observation stations, PM 2.5 retrieval from the remote sensing data is an effective complement to conventional ground observations and is a current research hotspot. The general principle behind the remote sensing retrieval of PM 2.5 is to first retrieve the aerosol optical depth (AOD) and calculate the PM 2.5 via the AOD-based statistical relationships. This method is likely to cause error propagation, which leads to instability in the retrieval model. In this paper, we propose a PM 2.5 remote sensing retrieval method via an ensemble random forest machine learning method to directly establish the relationship between the moderate-resolution imaging spectroradiometer (MODIS) images and ground observational PM 2.5 to avoid retrieval errors from the atmospheric aerosol optical depths and obtain PM 2.5 retrieval results with higher precision and spatial resolution. The proposed method first uses a random forest to train and validate the MODIS images and ground observation station PM 2.5 data; then, an optimal multimodel group, according to the determination coefficient R-square (R 2 ) index, is selected. Finally, the optimal multi-model group is used on the whole MODIS image to obtain the PM 2.5 retrieval result for the whole area. In an attempt to use machine learning technology to retrieve PM 2.5 , the experiments selected a substantial amount of MODIS image data during four seasons in Guangdong Province for validation and compared three performance indicators (R 2 , RMSE, and correlation coefficient (CC) to verify the superiority of the proposed algorithm.INDEX TERMS Ensemble random forest, machine learning, remote sensing based PM 2.5 retrieval, Kriging interpolation, aerosol optical depth.