Project at a Glance

Title: Mobile platform III: characterizing spatially inhomogeneous non-criteria pollutants in the Los Angeles air basin

Principal Investigator / Author(s): Paulson, Suzanne, and Arthur Winer

Contractor: UC Los Angeles

Contract Number: 09-357

Research Program Area: Health & Exposure, Emissions Monitoring & Control

Topic Areas: Field Studies, Mobile Sources & Fuels, Monitoring


A number of epidemiological studies have shown that exposure to elevated levels of fresh vehicular emissions causes a wide range of adverse human health effects. Fresh vehicular emissions contain a wide range of particle- and gas-phase species. Because such emissions are emitted and diluted together, their individual impacts are difficult to separate. Ultrafine particles (UFP) might contribute to the degradation of health associated with exposure to elevated levels of fresh vehicular emissions, but they are also an excellent tracer for fresh vehicle emissions, and are one of the foci of this study.

In urbanized areas, a large fraction of households are commonly located near freeways. In a highly urbanized area such as the California South Coast Air Basin (SoCAB), approximately 50% of the population lives within 1.5 km of freeways. An earlier mobile measurement platform study under the direction of Drs. Winer and Paulson demonstrated a large pollutant impact zone, extending beyond 2.5 km downwind of a freeway in Santa Monica, California during pre-sunrise hours.

The current study explores the variability of extended freeway plumes at several locations in Southern California. The ARB mobile measurement platform (MMP) was employed to measure vehicle-related pollutant concentrations on transects running upwind and downwind perpendicular to four freeway segments in the coastal, central and eastern areas (downtown Los Angeles, Paramount, Carson and Claremont) of the California South Coast Air Basin (SoCAB) during the pre-sunrise period (04:30-06:30).

Extended freeway plumes were observed for ultrafine particles, nitric oxide and particle-bound polycyclic aromatic hydrocarbons (PAH), at all four sites during stable pre-sunrise periods. Plume lengths were measured to be ~2 km or more with a dilution rate coefficient about a factor of ten lower than commonly observed for daytime. An average of 39 and 19% (񸁳%) of freeway plume peak concentrations of UFP remained 500 and 1500 m downwind from the freeways, respectively, for the four transects studied here. Because a large fraction of UFP and other vehicle related pollutants typically penetrate into indoor environments, and nocturnal surface inversions are widespread across the globe, our findings have significant implications for more extensive human exposures to vehicle-related pollutants than previously indicated based on daytime measurements of roadway plumes.

Factors controlling pollutant plume length downwind of freeways under stable conditions were background-subtracted peak concentration (which is a function of traffic flows and temperature) as well as meteorological parameters, such as wind direction and speed. Vertical stability (Richardson number) plays a minor role in dispersion coefficient variations within stable boundary layer conditions. A curve fit using a Gaussian dispersion model solution described excellently the observed UFP profiles both at the peak and far downwind (> 2 km) with R² ~0.9 or larger for all measurement sites.

The measurements of detailed plume shapes offer several additional insights about the details of pollutants near roadways. The geometry of the intersection of the secondary roadway (or, presumably, other components of the built environment) with the freeway geometry is an important parameter controlling the position of the plume peak concentration. The maximum concentration of the pollutant plume is closer to the centerline of the freeway if the freeway passes under the roadway, and it moves further downwind, by about 100 m, if the freeway passes above the secondary road. In the second case, the maximum concentration also tends to be lower.

Particle size is a factor in health impacts because it determines how much and where airway deposition occurs. As particles are diluted, they also undergo coagulation, deposition, evaporation and/or condensation. We explore the interplay of these factors, and demonstrate that as particle size decreases, UFP decay rates increase with distance from freeway.

Mobile measurement platform measurements are by their nature conducted at single places and times and thus are challenging to compare. With an aim toward making comparisons quantitative, we have developed an objective and systematic classification scheme of meteorological conditions affecting atmospheric primary pollutant levels in the (SoCAB). The method used is a classification and regression tree (CART) modeling approach. Previous CART approaches have been applied to secondary pollutants such as ozone. Here, we develop regression trees to predict the levels of traffic-related primary air pollutants such as NO and CO, based on combined upper air and surface meteorological conditions for 200709. The resulting regression trees perform well, providing excellent correlations between the regression classifications developed for different primary pollutant metrics, such as daily CO and NO maxima, as well as between monitoring sites. The spatial variations in primary pollutant concentrations between East-West monitoring sites in the SoCAB are more significant than those between North-South monitoring sites. The regression trees indicate these East-West variations are at least partly caused by Santa Ana Winds during winter and spring seasons. The meteorological parameters that determine the variability in primary pollutant concentrations, in approximate order of importance, are the mean surface wind speed, geopotential heights at 925 mbar, the upper air north-south pressure gradient, the daily minimum temperature, relative humidity at 1000 mbar, and vertical stability.

Here we apply the CART analysis to an inter-comparison of MMP measurements collected in several locations and times within Southern California. Daytime UFP concentrations in neighborhoods showed strong inter-community variations between West Los Angeles (1.110⁴ particles·cm⁻³), downtown Los Angeles (2.210⁴ particles·cm⁻³) and Boyle Heights (3.310⁴ particles·cm⁻³) in 2008. Intra-community pollutant variations were less intense but significant as an air mass experiences emissions from major freeways (I-405 and I-10).

Pollutant concentrations including UFP were highly elevated in close proximity to major freeways, as well as Santa Monica Airport. Impacts of high emitting vehicles on UFP distributions both on arterial roadways and in neighborhoods were also significant. About 70% reductions of UFP and PM2.5 were observed during the I-405 closure event (so called "Carmageddon") in 2011 with 20 - 85% decreases in nearby traffic flows in West Los Angeles.

Several lines of evidence point to the reduction in fleet averaged per-vehicle ultrafine particle emissions. By fitting freeway plumes using a formulation of the Gaussian plume dispersion model together with traffic data, we estimated a particle number emission factor of 7.510¹³ particles·vehicle⁻¹·km⁻¹, about one seventh of an estimate for nearby freeways made in 2001 in the published literature. For measurements in neighborhoods and on the freeways and arterials in West and downtown Los Angeles, ultrafine particle concentrations declined by between 10 and 70% between 2008 and 2011, depending on the location (neighborhood interiors, arterials, freeways etc.) and the contribution of high emitters. These comparisons demonstrate in-use motor vehicle UFP concentrations have significantly declined and suggest in-use motor vehicle emissions in general have declined as well.

For questions regarding this research project, including available data and progress status, contact: Research Division staff at (916) 445-0753

Stay involved, sign up with CARB's Research Email Distribution List