UC Berkeley

Air quality in California has undergone dramatic changes due to a variety of federal and state-wide regulatory efforts. The air pollutant monitoring network has evolved to achieve greater spatial and temporal coverage, incorporating novel measurement technologies such as the Beta Attenuation Method. Many key pollutants such as nitrogen oxides (NOx), black carbon (BC) and particulate matter with diameter less than 2.5 μm (PM2.5) have seen reductions in ambient concentrations. Controls on emissions from fuel combustion are a major contributor to these reductions, especially emission standards and reformulated fuel requirements for mobile sources. The goal of the research in this dissertation was to quantify changes in mobile source emissions and ambient pollutant concentrations as a result of emission regulations, using data from decades of air pollutant monitoring efforts.

Ocean-going vessels are an off-road source that contributes significantly to the emissions of sulfur dioxide (SO2) and PM2.5 in California, due to its long coastline and busy ports. In 2009, the state mandated that all ships operating within 24 nautical miles of the coastline burn fuels with lower sulfur contents, instead of heavy fuel oil. Ambient pollutant measurements near the Port of Oakland in the San Francisco Bay area were analyzed, before and after the lower-sulfur marine fuel mandate came into effect. Vanadium concentrations decreased by more than ten-fold, and 28 - 72% reductions in SO2 concentrations were observed at nearby measurement sites, with the magnitudes of reduction approximately proportional to the distance from shipping lanes. It was estimated that the change in ship fuel reduced ambient PM2.5 concentrations by about 3.1±0.6%, or 0.28±0.05 μg/m3. The site nearest to the port saw largest reductions in vanadium and SO2, whereas all sites experienced approximately the same magnitude of reduction in sulfate, a secondary PM2.5 component.

On-road mobile sources emit carbon monoxide (CO), NOx, and non-methane hydrocarbon (NMHC) from the burning of diesel fuel and gasoline in vehicle engines. These emissions can be estimated using the California On-Road Mobile Emission Model (EMFAC). Its predictions were evaluated using real-world emission measurements for on-road vehicles collected at truck weigh stations, in highway tunnels, and at the roadside using emission spectrometers, also known as remote sensors. EMFAC tends to overestimate the population of newer engines and underestimate travels by older engines, but the overall emissions estimates from EMFAC agree in general with fuel-based estimates developed in the present research. Predictions of diesel engine age distribution also capture the accelerated diesel replacement program currently being implemented in California. In recent years, diesel engines contribute more to statewide NOx emissions compared to gasoline engines, whereas gasoline engines are dominant for CO and NMHC emissions.

California has identified diesel particulate matter as toxic and has mandated that newer diesel trucks replace the existing older fleet by 2023. It is necessary to model the effects on air quality of changes in diesel emissions as well as in other sectors, such as gasoline burning and off-road sources, for the upcoming decade. Emission inventories for the South Coast Air Basin were modified to reflect changes in emission factors and population growth, and were used as input to the Community Multiscale Air Quality model (CMAQ). It was found that the accelerated diesel engine replacement program is effective in reducing ambient concentrations of BC, NOx, and PM2.5, but not of O3. Wintertime NOx concentration was predicted to decrease by 71±5% thanks to the universal use of newer selective catalytic reduction systems to reduce NOx. CMAQ aerosol chemistry was unable to depict summer daytime secondary organic aerosol formation in Southern California.

The PM2.5 ambient monitoring network in California has evolved to include the Beta Attenuation Monitoring (BAM), and the data from the network reflects the benefits of many on-road and off-road emission regulations. BAM is superior to the traditional filter methods because it minimizes evaporation of some components in PM2.5 and human errors. BAM measures higher than traditional filter-based sampling methods by 3 - 6 μg/m3 for annual averages. Since the 1990’s, all the sites examined in this study have shown clear downward trends, with some showing reductions of more than 50%. The use of BAM produces hourly concentration measurements every day, in contrast to only 24-hour averages that were available only once every 3 or 6 days with the older method. PM2.5 concentrations are higher in winter months in the San Francisco Bay area (+80±25%) and the San Joaquin Valley (+123±28%), but lower in the wintertime for inland sites in the South Coast Basin (-46±12%). Night-time PM2.5 concentrations reach maximum values due to the stagnant atmosphere and wood-burning. Weekend PM2.5 concentrations decrease by 8.9±5.9% for the San Francisco Bay area and 8.0±3.0% for the San Joaquin Valley, relative to the midweek values. This study shows the significance of making continuous measurements a part of the monitoring network.

This dissertation analyzed trends in observed air pollution concentrations and used an Eulerian Air Quality model to assess the impacts of major pollution control regulations in California. These regulations are effective in controlling primary pollutants and some secondary pollutants. Recommendations for future research include updating on-road emission inventories using real-world emission measurements, revising aerosol chemistry in CMAQ, and developing new methods to analyze hot spot pollution.