Surface air composition in Moscow based on observations data and CAMS reanalysis

The composition of the atmosphere in large cities directly affects the quality of life and health of the population. In this paper, we investigated the seasonal and daily variations in the mass concentration of trace gases in the atmosphere (NOx, CO) and aerosols (PM_2.5, PM₁₀) in Moscow in the period from 2005 to 2024 based on direct measurements at the Air quality monitoring system (AQMS) MosEcoMonitoring (MEM) and model calculations of CAMS. Differences in the average seasonal course of the concentration of the studied impurities were revealed based on the results of observations and numerical modeling for most of the AQMS MEM. The medians of the NOx and CO concentrations according to the measurement data exceed the model values ​​by 0.01–0.02 mg, and the upper percentile (P90) is 2–3 times higher than the calculated ones. On the contrary, the model values ​​of the upper percentile of PM_2.5 exceed the measured ones by 2 times. The PM₁₀ measurement data from most AQMS MEMs are generally comparable with the model calculations. However, the CAMS reanalysis does not reproduce the summer peaks in PM₁₀ and PM_2.5 concentrations, which is due to the model underestimating the contribution of biological aerosol. Against the background of a long-term seasonal variation, the median values ​​of the concentrations of the studied impurities during the 2020 lockdown during the COVID-19 coronavirus pandemic, on average, 0.01 mg lower, both according to the AQMS MEM data and according to the results of model calculations. The CAMS reanalysis poorly reproduces the decrease in gas and aerosol emissions during the lockdown period, but at the same time repeats the maximum PM₁₀ concentration recorded by the AQMS MEM in October 2020, caused by the long-range transport of dust aerosol to Moscow from the Northern Caspian regions. Thus, the CAMS reanalysis data satisfactorily reproduce the average seasonal variation of NO, PM₁₀ and PM_2.5 obtained from ground-based measurements, while significantly underestimating the upper percentiles of the mass concentration of gas impurities and overestimating the PM_2.5 concentration values. The numerical modeling also does not take into account the contribution of biological aerosol in the warm season to the growth of the mass concentration of aerosol particles in the urban surface air.