Mobile air monitoring data-processing strategies and effects on spatial air pollution trends

Abstract

Abstract. The collection of real-time air quality measurements while in motion (i.e., mobile monitoring) is currently conducted worldwide to evaluate in situ emissions, local air quality trends, and air pollutant exposure. This measurement strategy pushes the limits of traditional data analysis with complex second-by-second multipollutant data varying as a function of time and location. Data reduction and filtering techniques are often applied to deduce trends, such as pollutant spatial gradients downwind of a highway. However, rarely do mobile monitoring studies report the sensitivity of their results to the chosen data-processing approaches. The study being reported here utilized 40 h (> 140 000 observations) of mobile monitoring data collected on a roadway network in central North Carolina to explore common data-processing strategies including local emission plume detection, background estimation, and averaging techniques for spatial trend analyses. One-second time resolution measurements of ultrafine particles (UFPs), black carbon (BC), particulate matter (PM), carbon monoxide (CO), and nitrogen dioxide (NO2) were collected on 12 unique driving routes that were each sampled repeatedly. The route with the highest number of repetitions was used to compare local exhaust plume detection and averaging methods. Analyses demonstrate that the multiple local exhaust plume detection strategies reported produce generally similar results and that utilizing a median of measurements taken within a specified route segment (as opposed to a mean) may be sufficient to avoid bias in near-source spatial trends. A time-series-based method of estimating background concentrations was shown to produce similar but slightly lower estimates than a location-based method. For the complete data set the estimated contributions of the background to the mean pollutant concentrations were as follows: BC (15%), UFPs (26%), CO (41%), PM2.5-10 (45%), NO2 (57%), PM10 (60%), PM2.5 (68%). Lastly, while temporal smoothing (e.g., 5 s averages) results in weak pair-wise correlation and the blurring of spatial trends, spatial averaging (e.g., 10 m) is demonstrated to increase correlation and refine spatial trends.