Improvements in air quality in the Netherlands during the corona lockdown based on observations and model simulations

Guus J M Velders; Saskia M Willers; Joost Wesseling; Sef van den Elshout; Eric van der Swaluw; Dennis Mooibroek; Sjoerd van Ratingen

doi:10.1016/j.atmosenv.2020.118158

Improvements in air quality in the Netherlands during the corona lockdown based on observations and model simulations

Atmos Environ (1994). 2021 Feb 15:247:118158. doi: 10.1016/j.atmosenv.2020.118158. Epub 2021 Jan 9.

Authors

Guus J M Velders^{1

2}, Saskia M Willers³, Joost Wesseling¹, Sef van den Elshout³, Eric van der Swaluw¹, Dennis Mooibroek¹, Sjoerd van Ratingen¹

Affiliations

¹ National Institute for Public Health and the Environment (RIVM), PO Box 1, 3720, BA, Bilthoven, the Netherlands.
² Institute for Marine and Atmospheric Research Utrecht (IMAU), Utrecht University, the Netherlands.
³ DCMR Environmental Protection Agency, Schiedam, the Netherlands.

Abstract

The lockdown measures in response to the SARS-CoV-2 virus outbreak in 2020 have resulted in reductions in emissions of air pollutants and corresponding ambient concentrations. In the Netherlands, the most stringent lockdown measures were in effect from March to May 2020. These measures coincided with a period of unusual meteorological conditions with wind from the north-east and clear-sky conditions, which complicates the quantification of the effect of the lockdown measures on the air quality. Here we quantify the lockdown effects on the concentrations of nitrogen oxides (NO_x and NO₂), particulate matter (PM₁₀ and PM_2.5) and ozone (O₃) in the Netherlands, by analyzing observations and simulations with the atmospheric chemistry-transport model EMEP/MSC-W in its EMEP4NL configuration, after eliminating the effects of meteorological conditions during the lockdown. Based on statistical analyses with a Random Forest method, we estimate that the lockdown reduced observed NO₂ concentrations by 30% (95% confidence interval 25-35%), 26% (21-32%), and 18% (10-25%) for traffic, urban, and rural background locations, respectively. Slightly smaller reductions of 8-28% are found with the EMEP4NL simulations for urban and regional background locations based on estimates in reductions in economic activity and emissions of traffic and industry in the Netherlands and other European countries. Reductions in observed PM_2.5 concentrations of about 20% (10-25%) are found for all locations, which is somewhat larger than the estimates of 5-16% based on the model simulations. A comparison of the calculated NO₂ traffic contributions with observations shows a substantial drop of about 35% in traffic contributions during the lockdown period, which is similar to the estimated reductions in mobility data as reported by Apple and Google. Since the largest health effects related to air pollution in the Netherlands are associated with exposure to PM₁₀ and PM_2.5, the lockdown measures in spring of 2020 have temporarily improved the air quality in the Netherlands. The concentrations of the most health relevant compounds have only been reduced by about 10-25%.

Keywords: Covid-19; Nitrogen dioxide; Ozone; Particulate matter; Random forest.