Spatiotemporal profiles of ultrafine particles differ from other traffic-related air pollutants: Lessons from long-term measurements at fixed sites and mobile monitoring

In the absence of routine monitoring of ultrafine particles (UFP, Dp < 100 nm), air pollution epidemiology studies often use other co-emitted pollutants as a proxy for UFP, with NOx (NO + NO2) considered a good choice. We use long term fixed site measurements along with extensive mobile monitoring data to evaluate the spatiotemporal correlation of UFP and NOx. We incorporate 6 years of hourly particle number (PN, an approximation of UFP) concentration data from multiple fixed sites across the San Francisco Bay Area that include near-highway, urban, suburban, and rural sites. In addition, we incorporate observations from a 32 month mobile monitoring campaign comprising >1000 h of coverage of a range of road types and land uses. Across all fixed sites, PN concentrations show prominent mid-day peaks during the summer-characteristic of new particle formation-which are not observed for other co-emitted pollutants (NOx, BC, CO). While we find moderate correlation in diurnal patterns of NOx and UFP at sites with high traffic, the correlation drops significantly for low traffic areas, especially during high insolation (e.g., summer daytime) periods. Mobile monitoring data yields similar results: NOx is observed to have weaker correlation with PN for non-highway roads during high insolation periods. The spatiotemporal profiles of UFP can differ strongly from other co-emitted air pollutants when new particle formation contributes a significant share of UFP. This journal is