Modeling Roadway Link PM2.5 Emissions with Accurate Truck Activity Estimate for Regional Transportation Conformity Analysis

The impact of fine particulate matter (PM) on public health has long been a concern. The primary mobile sources of fine (PM) (PM2.5) are diesel trucks. In practice, accurate roadway link-based modeling of the truck emissions remains a major challenge because of aggregated and unreliable truck activity data. The advanced emission model MOVES has been recommended by the U.S. Environmental Protection Agency for estimating emission factors, but supplying accurate and detailed truck activity-related inputs has become another challenge. Daily truck traffic activity is usually not estimated accurately and cannot be disaggregated to hourly activity with traditional methods. To address this problem, two innovative econometric methods were successfully enhanced in this study to predict accurate truck activity-based inputs for emission estimation. The models for truck factor spatial panel and multinomial probit hourly vehicle miles traveled were improved and tested with regional traffic data from the greater Cincinnati, Ohio, area. The application of those models indicates that using MOVES default input data underestimates the regional PM2.5 inventory. The proposed methodology enables plotting the spatiotemporal distribution of PM2.5 emissions in a subarea. Such an integrated method provides a useful decision support tool for practitioners because they can also model PM2.5 emissions at a detailed level as required by project-level conformity analysis. The methodology presented is scalable and transferable and holds technical promise in its application to different regions and different pollutants.