From COVID-19 to future electrification: Assessing traffic impacts on air quality by a machine-learning model
Jiani Yang1, Yifan Wen2, Yuan Wang3,4
1Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125.
Summary
COVID-19 lockdowns revealed vehicle emission control effectiveness. Reduced traffic lowered nitrogen dioxide and fine particle pollution, but increased ozone, with heavy-duty trucks being key contributors.
Area of Science:
- Environmental Science
- Atmospheric Chemistry
- Transportation Studies
Background:
- The COVID-19 pandemic caused significant traffic volume fluctuations globally.
- These fluctuations offer a unique opportunity to study vehicle emission control impacts.
- Understanding pollutant dynamics is crucial for urban air quality management.
Purpose of the Study:
- To develop and validate a predictive model for air pollutant concentrations (NO2, O3, PM2.5).
- To assess the impact of traffic reduction during COVID-19 lockdowns on air quality in Los Angeles.
- To identify key factors influencing pollutant levels and evaluate future emission control strategies.
Main Methods:
- Development of a random-forest regression model.
- Utilized real-time observational data from the Los Angeles megacity during the COVID-19 pandemic.
- Analysis focused on predicting surface-level nitrogen dioxide (NO2), ozone (O3), and fine particulate matter (PM2.5).
Main Results:
- The model accurately reproduced pollutant concentrations in the Los Angeles Basin.
- Traffic reduction during strict lockdowns decreased NO2 by 30.1% and PM2.5 by 17.5%.
- Ozone (O3) concentrations increased by 5.7%, primarily influenced by heavy-duty truck emissions.
Conclusions:
- Vehicle emission controls similar to COVID-19 lockdown effects can improve air quality, albeit with reduced magnitude.
- Targeting heavy-duty truck emissions is critical for mitigating NO2 and PM2.5.
- Widespread vehicular electrification is projected to significantly reduce NO2 levels.
