Spatiotemporal neural network for estimating surface NO2 concentrations over north China and their human health impact.

Atmospheric nitrogen dioxide (NO2) is an important reactive gas pollutant harmful to human health. The spatiotemporal coverage provided by traditional NO2 monitoring methods is insufficient, especially in the suburban and rural areas of north China, which have a high population density and experience severe air pollution. In this study, we implemented a spatiotemporal neural network (STNN) model to estimate surface NO2 from multiple sources of information, which included satellite and in situ measurements as well as meteorological and geographical data. The STNN predicted NO2 with high accuracy, with a coefficient of determination (R2) of 0.89 and a root mean squared error of 5.8 µg/m3 for sample-based 10-fold cross-validation. Based on the surface NO2 concentration determined by the STNN, we analyzed the spatial distribution and temporal trends of NO2 pollution in north China. We found substantial drops in surface NO2 concentrations ranging between 9.1% and 33.2% for large cities during the 2020 COVID-19 lockdown when compared to those in 2019. Moreover, we estimated the all-cause deaths attributed to NO2 exposure at a high spatial resolution of about 1 km, with totals of 6082, 4200, and 18,210 for Beijing, Tianjin, and Hebei Provinces in 2020, respectively. We observed remarkable regional differences in the health impacts due to NO2 among urban, suburban, and rural areas. Generally, the STNN model could incorporate spatiotemporal neighboring information and infer surface NO2 concentration with full coverage and high accuracy. Compared with machine learning regression techniques, STNN can effectively avoid model overfitting and simultaneously consider both spatial and temporal correlations of input variables using deep convolutional networks with residual blocks. The use of the proposed STNN model, as well as the surface NO2 dataset, can benefit air quality monitoring, forecasting, and health burden assessments.

Variations of Urban NO2 Pollution during the COVID-19 Outbreak and Post-Epidemic Era in China: A Synthesis of Remote Sensing and In Situ Measurements

Since the COVID-19 outbreak in 2020, China’s air pollution has been significantly affected by control measures on industrial production and human activities. In this study, we analyzed the temporal variations of NO2 concentrations during the COVID-19 lockdown and post-epidemic era in 11 Chinese megacities by using satellite and ground-based remote sensing as well as in situ measurements. The average satellite tropospheric vertical column density (TVCD) of NO2 by TROPOMI decreased by 39.2–71.93% during the 15 days after Chinese New Year when the lockdown was at its most rigorous compared to that of 2019, while the in situ NO2 concentration measured by China National Environmental Monitoring Centre (CNEMC) decreased by 42.53–69.81% for these cities. Such differences between both measurements were further investigated by using ground-based multi-axis differential optical absorption spectroscopy (MAX-DOAS) remote sensing of NO2 vertical profiles. For instance, in Beijing, MAX-DOAS NO2 showed a decrease of 14.19% (versus 18.63% by in situ) at the ground surface, and 36.24% (versus 36.25% by satellite) for the total tropospheric column. Thus, vertical discrepancies of atmospheric NO2 can largely explain the differences between satellite and in situ NO2 variations. In the post-epidemic era of 2021, satellite NO2 TVCD and in situ NO2 concentrations decreased by 10.42–64.96% and 1.05–34.99% compared to 2019, respectively, possibly related to the reduction of the transportation industry. This study reveals the changes of China’s urban NO2 pollution in the post-epidemic era and indicates that COVID-19 had a profound impact on human social activities and industrial production.