Estimating Regional Ground-Level PM2.5 Directly From Satellite Top-Of-Atmosphere Reflectance Using Deep Belief Networks

Almost all remote sensing atmospheric PM2.5 estimation methods need satellite aerosol optical depth (AOD) products, which are often retrieved from top-of-atmosphere (TOA) reflectance via an atmospheric radiative transfer model. Then, is it possible to estimate ground-level PM2.5 directly from satellite TOA reflectance without a physical model? In this study, this challenging work was achieved based on a machine learning model. Specifically, we established the relationship between PM2.5, satellite TOA reflectance, observation angles, and meteorological factors in a deep learning architecture (denoted as Ref-PM modeling). This relationship was trained with station PM2.5 measurements, and then the PM2.5 values of those locations without stations could be retrieved. Taking the Wuhan Urban Agglomeration as a case study, the results demonstrate that, compared with AOD-PM modeling, the Ref-PM modeling obtains a competitive performance, with sample-based cross-validated R-2 and root-mean-square error values of 0.87 and 9.89g/m(3), respectively. Also, the TOA-reflectance-derived PM2.5 has a finer resolution and a larger spatial coverage than the AOD-derived PM2.5. This work provides an alternative technique to estimate ground-level PM2.5, and may have the potential to promote the application in atmospheric environmental monitoring.