Near-real-time estimation of hourly open biomass burning emissions in China using multiple satellite retrievals

Open biomass burning (OBB) is an important source of air pollutants and greenhouse gases, but its dynamic emission estimation remains challenging. Existing OBB emission datasets normally provide daily estimates based upon Moderate Resolution Imaging Spectroradiometer (MODIS) retrievals but tend to underestimate the emissions due to the coarse spatial resolution and sparse observation frequency. In this study, we proposed a novel approach to improve OBB emission estimations by fusing multiple active fires detected by MODIS, Visible Infrared Imaging Radiometer onboard the Suomi National Polar-orbiting Partnership (VIIRS S-NPP) and Himawari-8. The fusion of multiple active fires can capture the missing small fires and the large fires take place during the non-overpass time of MODIS observations. Also, regional-based fire radiative power (FRP) cycle reconstruction models and OBB emission coefficients were developed to address the large spatial discrepancies of OBB emission estimations across China and to promote the estimate to an hourly resolution. Using the new approach, hourly gridded OBB emissions in China were developed and can be updated with a lag of 1-day, or even near-real-time when real-time multiple active fires are available. OBB emissions in China based on this approach were more than 3 times of those in previous datasets. Evaluations revealed that the spatial distribution of the estimated PM_2.5 emissions from this study was more consistent with the ambient PM_2.5 concentrations during several episodes than existing datasets. The hourly OBB emissions provide new insight into its spatiotemporal variations, enhance timely and reliable air quality modeling and forecast, and support the formulation of accurate prevention and control policies of OBB.