Estimating vehicular emission factors and vehicle-induced turbulence: Application of an air quality sensor array for continuous multipoint monitoring in a tunnel

In this study, air quality monitoring nodes (equipped with pollutants (NO, NO2, CO, and PM2.5), temperature, and humidity sensors) were deployed at seven locations inside a road tunnel for continuous multipoint moni-toring of traffic emissions. At the same time, in-tunnel turbulence was measured with a 3D-sonic anemometer in the middle of the tunnel to estimate vehicle-induced turbulence (VIT). The emission factors (EFs) of all pollutants estimated from conventional two-point measurements varied significantly among arbitrarily chosen pairs of measurement locations (with a range of 53-193% compared to the mean value). These findings suggest the need for multipoint monitoring for more generalizable EF estimates. Our results indicate that installing a sensor array in a tunnel could enable long-term continuous multipoint monitoring of traffic-related pollutants with limited resources. The EF estimates obtained in this study were higher (similar to 7 times) than those calculated from an emission inventory based on actual traffic data, particularly for CO and wintertime NOX. One possible explanation for this inconsistency is an underestimation of high emitters in the emission inventory. We also estimated VIT as a function of the total traffic flow rate and the fraction of heavy-duty diesel vehicles in a fleet. We expect our results to be used as input data for local and regional air quality models, which can improve the model's per-formance. We also expect that long-term continuous monitoring of air pollutants in tunnels with multiple sensor nodes will aid evaluations of the effectiveness of air pollution reduction policies by tracking changes in traffic emissions.