Satellite-Based Long-Term Spatiotemporal Trends in Ambient NO 2 Concentrations and Attributable Health Burdens in China From 2005 to 2020.

Despite the recent development of using satellite remote sensing to predict surface NO 2 levels in China, methods for estimating reliable historical NO 2 exposure, especially before the establishment of NO 2 monitoring network in 2013, are still rare. A gap-filling model was first adopted to impute the missing NO 2 column densities from satellite, then an ensemble machine learning model incorporating three base learners was developed to estimate the spatiotemporal pattern of monthly mean NO 2 concentrations at 0.05° spatial resolution from 2005 to 2020 in China. Further, we applied the exposure data set with epidemiologically derived exposure response relations to estimate the annual NO 2 associated mortality burdens in China. The coverage of satellite NO 2 column densities increased from 46.9% to 100% after gap-filling. The ensemble model predictions had good agreement with observations, and the sample-based, temporal and spatial cross-validation (CV) R 2 were 0.88, 0.82, and 0.73, respectively. In addition, our model can provide accurate historical NO 2 concentrations, with both by-year CV R 2 and external separate year validation R 2 achieving 0.80. The estimated national NO 2 levels showed a increasing trend during 2005-2011, then decreased gradually until 2020, especially in 2012-2015. The estimated annual mortality burden attributable to long-term NO 2 exposure ranged from 305 thousand to 416 thousand, and varied considerably across provinces in China. This satellite-based ensemble model could provide reliable long-term NO 2 predictions at a high spatial resolution with complete coverage for environmental and epidemiological studies in China. Our results also highlighted the heavy disease burden by NO 2 and call for more targeted policies to reduce the emission of nitrogen oxides in China.