From COVID-19 to Future Electrification: Assessing Traffic Impacts on Air Quality by a Machine Learning Model

The large fluctuations in traffic during the COVID-19 pandemic provide an unparalleled opportunity to assess emission control efficacy. Here we develop a machine-learning (ML) model, based on the large volume of real-time observational data, to predict surface-level NO2, O3, and 20 fine particle concentration in Los Angeles. Our ML model exhibits high fidelity in reproducing pollutant concentrations in the Los Angeles basin and identifies major factors controlling each species. During the strictest lockdown period, the traffic reduction led to decreases in NO2 and PM2.5 by –27.8% and –17.5%, respectively, but a 6% increase in O3. Heavy-duty truck emissions contribute primarily to these variations. Future traffic emissions controls are estimated to impose 25 similar effects as observed during the COVID-19 lockdown, but with smaller magnitude. Vehicular electrification will achieve further alleviation of NO2 levels.