Estimate hourly PM2.5 concentrations from Himawari-8 TOA reflectance directly using Geo-intelligent long short-term memory network

Abstract Fine particulate matter (PM2.5) has attracted extensive attention because of its baneful influence on human health and the environment. However, the sparse distribution of PM2.5 measuring stations limits its application to public utility and scientific research, which can be remedied by satellite observations. Therefore, we developed a Geo-intelligent long short-term network (Geoi-LSTM) to estimate hourly ground-level PM2.5 concentrations in 2017 in Wuhan Urban Agglomeration (WUA). We conducted contrast experiments to verify the effectiveness of our model and explored the optimal modeling strategy. It turned out that Geoi-LSTM with TOA reflectance, meteorological conditions, and NDVI as inputs performs best. The station-based cross-validation R2, root mean squared error and mean absolute error are 0.82, 15.44 μg/m3, 10.63 μg/m3, respectively. Based on model results, we revealed spatiotemporal characteristics of PM2.5 in WUA. Generally speaking, during the day, PM2.5 concentration remained stable at a relatively high level in the morning and decreased continuously in the afternoon. While during the year, PM2.5 concentrations were highest in winter, lowest in summer, and in-between in spring and autumn. Combined with meteorological conditions, we further analyzed the whole process of a PM2.5 pollution event. Finally, we discussed the loss in removing clouds-covered pixels and compared our model with several popular models. Overall, our results can reflect hourly PM2.5 concentrations seamlessly and accurately with a spatial resolution of 5km, which benefits PM2.5 exposure evaluations and policy regulations.