Variability of NO<sub>2</sub> concentrations over China and effect on air quality derived from satellite and ground-based observations

Abstract

Abstract. The variation of NO2 concentrations in mainland China is analyzed on different timescales, from decadal to weekly, using both satellite data and data from ground-based monitoring networks. TROPOMI (TROPOspheric Monitoring Instrument) data were used to study the spatial variations of tropospheric NO2 vertical column densities (TVCDs) over the study area during 16–20 weeks after the Chinese Spring Festival (25 January 2020). These data were used to select 11 regions for more detailed analysis of the variation of NO2 TVCDs on a decadal timescale. In this analysis, monthly and annual averaged NO2 TVCDs derived from OMI (Ozone Monitoring Instrument) observations were used for the years 2011 to 2019. The results show the NO2 TVCD trends for different regions, all decreasing in response to emission reduction policies but with a different onset and a possible halt of the decrease in recent years; trends and period in the south of the study area are different from those in the north. Variations of NO2 TVCDs on shorter timescales, monthly and weekly, were analyzed using TROPOMI data. In addition, the variations of weekly-averaged ground-based NO2 concentrations in 11 major cities were analyzed together with those for O3 and PM2.5. In particular these data were used to determine their effect on the air quality as expressed by the air quality index (AQI). For quantitative estimates, the use of weekly concentrations is more accurate than the use of monthly values, and the effects of long-term trends and their reversal needs to be taken into account for the separation of effects of the lockdown and the Spring Festival. Neglecting the possible reversal of the trends leads to overestimation of the lockdown effect in the south and underestimation in the north. The ground-based data confirm earlier reports, based on satellite observations, that the expected improvement of air quality due to the reduction of NO2 concentrations was offset by the increase of the concentrations of O3 and the different effects of the lockdown measures on PM2.5, as well as effects of meteorological influences and heterogeneous chemistry. The AQI seems to be mostly influenced by PM2.5 rather than NO2. A qualitative comparison between time series of satellite and ground-based NO2 observations shows both similarities and differences. The study further shows the different behaviors in city clusters in the north and south of China, as well as inland in the Sichuan and Guanzhong basins. Effects of other holidays and events are small, except in Beijing where the air quality in 2020 was notably better than in previous years. This study was undertaken for China, but the methodology and results have consequences for air quality studies in other areas, and part of the conclusions are generally applicable.