Impact of COVID-19 pandemic restrictions in urban mobility and air pollution in Lisbon, Portugal

• Autor(a) principal: Juma, Alexandre Sadik Vieira
• Data de Publicação: 2021
• Tipo de documento: Dissertação
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10071/23881
• topic: Urban mobility; Air quality; COVID-19; Machine learning; Automl; Mobilidade urbana; Qualidade do ar

The present work reports the impacts on urban mobility and air quality in Lisbon, Portugal as a consequence of the imposed restrictions to curb the transmission of SARS-CoV-2 virus which causes COVID-19 disease. During the first national emergency period (18-03-2020 to 03-05-2020) the sharp reductions in anthropogenic activities, most importantly road traffic, resulted in generally reduced criteria air pollutant concentration when compared to a homologous baseline from 2013-2019 measured in the six air quality monitoring stations throughout the city. The most negatively impacted air pollutant was No2 with a reduction of 54.35% in traffic stations and 28.62% reduction in background stations. An exception to this trend was the observed O3 concentration increase of 12.89% in traffic stations which is potentially due to changes in the Nox:VOC ratio and reduced O3 titration by NO as a result of sharp decrease of NOx emissions in the usually most polluted city hotspots. This phenomenon raises the need of additional measures to mitigate O3 pollution increases as part of the Lisbon and Tagus Valley air quality improvement plan which aims to reduce NO2 concentrations, namely specific measures for VOC management. Google mobility indicator for local commerce was found to be the main anthropogenic activity indicator for Lisbon with a moderate and positive correlation with NO2 concentration (r=+0.54), whereas the average wind speed was the most relevant natural phenomena contributing to NO2 concentration with a moderate and negative correlation (r=-0.53). A regressor ML pipeline was trained to predict NO2 concentration with the available anthropogenic activity, weather, and air pollutant inputs from March/2020 to March/2021, achieving R2=0.925 on the test set and subsequent feature importance analysis uncovered that anthropogenic features contribute to 41.19% of NO2 concentrations and natural phenomena features contribute to 58.81%.