New interpretable deep learning model to monitor real-time PM concentrations from satellite data.
Journal:
Environment international
Published Date:
Sep 10, 2020
Abstract
Particulate matter with a mass concentration of particles with a diameter less than 2.5 μm (PM) is a key air quality parameter. A real-time knowledge of PM is highly valuable for lowering the risk of detrimental impacts on human health. To achieve this goal, we developed a new deep learning model-EntityDenseNet to retrieve ground-level PM concentrations from Himawari-8, a geostationary satellite providing high temporal resolution data. In contrast to the traditional machine learning model, the new model has the capability to automatically extract PM spatio-temporal characteristics. Validation across mainland China demonstrates that hourly, daily and monthly PM retrievals contain the root-mean-square errors of 26.85, 25.3, and 15.34 μg/m, respectively. In addition to a higher accuracy achievement when compared with various machine learning inversion methods (backpropagation neural network, extreme gradient boosting, light gradient boosting machine, and random forest), EntityDenseNet can "peek inside the black box" to extract the spatio-temporal features of PM. This model can show, for example, that PM levels in the coastal city of Tianjin were more influenced by air from Hebei than Beijing. Further, EntityDenseNet can still extract the seasonal characteristics that demonstrate that PM is more closely related within three month groups over mainland China: (1) December, January and February, (2) March, April and May, (3) July, August and September, even without meteorological information. EntityDenseNet has the ability to obtain high temporal resolution satellite-based PM data over China in real-time. This could act as an important tool to improve our understanding of PM spatio-temporal features.