Polar Grid-based Nonparametric Trajectory Analysis for Improved Near-Source Air Pollutant Monitoring Applications.

Nonparametric Trajectory Analysis (NTA) combines continuous air pollutant concentrations with local-scale back trajectories estimated from wind speed and direction to quantify source area impacts at a monitoring site. Previously, NTA calculations used regular grid spacing and kernel smoothing which limited the spatial variability in NTA maps around monitoring sites located near sources. NTA was modified to be based on a polar grid to produce finer resolution close to the site. Polar NTA results were compared to the previous version using one year of continuous data from a near-road monitoring network site in St. Louis, MO. Measurements of wind speed and direction showed that how often and for how long the monitoring site was downwind of the roadway was highly variable, as were concentrations of nitrogen oxides (NOx), black carbon (BC) and particulate matter less than 2.5 μm (PM2.5), but PM2.5 remained elevated or lower for longer periods. Background subtracted PM2.5 was estimated from PM2.5 and had diurnal variability similar to the traffic related air pollutants. Polar NTA results indicated impacts from the roadway, with areas closest to the roadway contributing the highest average concentrations (1.4 μg/m3 for BC and 39 ppb for NOx). Results for background subtracted PM2.5 were similar but indicated additional contributions from nearby point sources. Polar NTA increased maximum estimated contributions by 22% for BC and 39% for NOx. Comparison of results for multiple air pollutants demonstrated the utility of Polar NTA for understanding relative impacts of different sources on local air quality in near-source monitoring applications.