Incorporating dynamic population activities in modeling exposure to urban air pollution: insights from COVID-19 lockdowns in three European cities

Abstract The COVID-19 pandemic in 2020 led to global lockdown measures, significantly changing population activity patterns and providing an unprecedented situation to study their effects on air quality. Previous studies primarily focused on pollutant concentration changes, often neglecting the influence of modified population activities on exposure estimates. This study aims to evaluate the impact of changes in time-activity patterns on population exposures to NO 2 , O 3 , and PM 2.5 in three urban European areas during the first lockdowns in March and April of 2020. A comprehensive hybrid exposure model was used, integrating urban-scale air pollutant dispersion data with diurnal population activity, accounting for both concentration and population activity changes due to lockdown measures. Population-weighted exposures and total time-integrated exposure levels were assessed for Hamburg, Germany, Liège, Belgium, and Marseille, France. The lockdown measures led to significant reductions in NO 2 and PM 2.5 concentrations while increasing O 3 concentrations. Adjusting for population activity changes showed additional hourly population weighted exposure reductions for NO 2 by up to 6% and for O 3 and PM 2.5 by up to 7%, while total time-integrated exposure was additionally reduced for NO 2 (up to 3%), O 3 (up to 8%) and PM 2.5 (up to 7%). These findings highlight the importance of incorporating dynamic population activity data for more accurate exposure and health impact assessments, especially in urban areas. The study highlights that exposure estimated at residential addresses likely underestimate exposure and related health effects.