Year: 2019
Type: article
Abstract: The objective of this study was to examine the relationship between air quality and traffic and weather parameters. With air quality measurements spanning over 11 months, the authors attempted to gain better understanding of the near-freeway air pollutant concentration, traffic speed, traffic flow, and weather parameters. The authors applied both multiple linear regression (MLR) and multivariate adaptive regression splines (MARS) models to examine the relationship among the weather conditions, traffic states, and near-freeway air pollutant concentrations. Both MLR and MARS showed that all weather parameters (e.g., relative humidity, temperature, wind) were significant variables. For the State Route 60 air monitoring station (AMS), MLR gave the adjusted R² as 0.077 and 0.264 for fine particulate matter (PM₂.₅) and nitrogen dioxide (NO₂), respectively, and MARS gave the R² as 0.19 and 0.53, respectively. For the Interstate 710 AMS, MLR gave the adjusted R² as 0.035 and 0.324 for PM₂.₅ and NO₂, respectively, and MARS gave the R² as 0.11 and 0.62, respectively. Generally, NO₂ concentration can be better explained by the selected variables than can PM₂.₅. The test of traffic speed segmentation indicates that the traffic speed has a considerable influence on near-road pollutant concentrations. When applying MLR and MARS models for winter months, the prediction performance for PM₂.₅ improves significantly at both AMSs, but the improvement effect is moderate for NO₂. The authors recommend that controlling seasonal weather variables can significantly improve PM₂.₅ prediction performance.
Source: ROSA P
Authors: Ji Luo, Ayla Moretti, Guoyuan Wu
Language: English
Subfield: Environmental Engineering
Field: Environmental Science
Domain: Physical Sciences
Open Access status: green
Award: G2652295918
OpenAlex ID: W3131657455
