On-highway vehicle emission factors, and spatial patterns, based on mobile monitoring and absolute principal component score

An important component of air quality engineering is quantifying in-use, fleet-average emission factors, and the spatial patterns of vehicle emissions. We report here that an absolute principal component score (APCS) analysis of on-road mobile measurements is a straightforward, efficient method for identifying the major contributors of traffic-related pollutants, deriving fuel-based emission factors, and mapping spatial patterns. Specifically, we applied the APCS model to on-highway measurements of nitrogen oxides (NO _X ), carbon monoxide (CO), carbon dioxide (CO ₂ ), black carbon (BC), and particle number (PN) obtained from a mobile platform deployed over a 5-day sampling period in Chengdu, China. Data were collected for (1) heavy-duty diesel truck (HDDT) plumes (“chase data”) and (2) the general on-road environment (“non-chase data”). The bootstrapped APCS model was used to estimate area-wide, fuel-based average emission factors and their respective 95% confidence intervals. Two components representing diesel trucks and gasoline vehicles were extracted from non-chase data, accounting for 67% of the variance of the on-highway concentrations. Two additional principal components extracted from HDDT chase data, representing normal and high emission features, further separating the emissions characteristics of HDDTs. The fleet-average emission factors for NO _X , CO, BC, and PN were 2.2, 50.3, 0.023 g/kg, and 0.32 × 10 ¹⁵ particles/kg for gasoline-powered vehicles, respectively; 33, 3.7, 0.19 g/kg, and 3.3 × 10 ¹⁵ particles/kg fuel for HDDTs’ normal emission feature, respectively; and 105, 29, 2.5 g/kg fuel, and 16 × 10 ¹⁵ particles/kg fuel for HDDTs’ high emission feature, respectively. APCS results for chase data revealed the existence of high emitters among Chengdu's HDDT fleet, with emission factors 3 to 13 times higher than the normal HDDT vehicles. Although the high emitters are a minority of the fleet, they disproportionately contribute to the overall emissions; emission control policies may wish to target such vehicles.