UC Berkeley

The widespread utilization of vehicles in urban regions is a significant contributor to the pervasive air quality challenges faced by many cities globally. Among the pollutants emitted from vehicles, nitrogen oxides (NOx = NO + NO2) play a pivotal role in influencing air quality. NOx emission reduction has been a primary focus in air quality improvement efforts given the substantial contribution of NOx to the formation of secondary pollutants with environmental and human health consequences, notably ozone and particulate matter (PM2.5). While emissions have decreased significantly over the last few decades, difficulty constraining vehicle emissions has led to discrepancies among emission estimates derived from field campaigns, satellite observations, and models developed and used by government agencies. This research seeks to develop, assess, and implement a high-resolution emission inventory for motor vehicles. The broader objective of this work is to improve the understanding of spatial and temporal trends in vehicle emissions which is critical to developing effective air pollution control plans.

In alignment with these objectives, Chapter 2 assesses long-term (1990-2020) trends in vehicle emissions for the United States as a whole and more specifically for California. Unlike previous studies, which only considered emissions from light-duty gasoline and heavy-duty diesel vehicles, this research introduces a new category: light- and medium-duty diesel vehicles. Analysis of roadside spectrometer measurements reveals that this vehicle category exhibits the highest NOx emission factors and the slowest rate of reduction over the past three decades. Assessment of on-road NOx emission trends indicates a consistent pattern, with both the US and California experiencing ~70% decreases in NOx emissions between 1990 and 2020. The relative reduction in diesel NOx emissions has been larger in California (48%) than nationally (32%) since 2010. This is attributed to the rapid turnover of the diesel truck fleet in California, with older engines being replaced with more modern engines equipped with advanced emission control equipment. Emission estimates for California indicate convergence between gasoline and diesel source contributions to NOx emissions, whereas the majority of on-road vehicle NOx emissions in the US still come from diesel engines. Analysis of emission factor trends suggests diminishing returns in reducing gasoline emission factors, emphasizing the potential benefits of improving emission control for light- and medium-duty diesel vehicles to maintain the downward trend.

Chapter 3 builds upon the emission trend analyses of Chapter 2 to assess the impact of emission control advancements on spatial patterns of NOx emissions and concentrations in the Los Angeles metropolitan area. This chapter develops a high-resolution 1.3 km gridded motor vehicle emission inventory, characterized by higher overall NOx emissions and a more pronounced drop-off on weekends attributable to the inclusion of the light- and medium-duty diesel vehicle category with higher emission factors. The new high-resolution inventory is used as input to the Weather Research and Forecasting model with Chemistry (WRF-Chem). Model results are evaluated using aircraft and TROPOMI satellite measurements of nitrogen dioxide (NO2). This evaluation focuses on model skill in reproducing large observed decreases in atmospheric NOx levels on weekends. The model showed comparable performance on weekdays and weekends, indicating appropriate day-of-week scaling and sector-based emissions distribution. This inventory suggests that on-road vehicles are responsible for 55-72% of NOx emissions in the South Coast Air Basin, higher than the current amount (43%) attributed to vehicles in the management district planning inventory.

Chapter 4 explores the future impacts on secondary particulate matter (PM2.5) resulting from vehicle-sourced NOx emissions and analyzes how reductions in NOx will influence the magnitudes and spatial distributions of PM2.5, NOx, and associated human health impacts. This chapter projects vehicle emissions from 2021 forward in time to 2030 and 2040 and expands the previously developed high-resolution NOx emission inventory from Los Angeles to encompass the entire state of California. Estimated emissions are used to determine resulting patterns of decrease in NOx and secondary PM2.5 using the Intervention Model for Air Pollution (InMAP). The resulting emissions are 62% higher in the fuel-based inventory in comparison to EMFAC in the baseline year 2021, with the largest discrepancy occurring in the light-/medium-duty diesel category. Predicted NOx impacts remain concentrated near major freeways, while secondary PM2.5 impacts extend further from high-emission areas, likely due to transport and spatially variable atmospheric reaction rates. Significant decreases in secondary PM2.5 exposure are observed, particularly in the San Joaquin Valley and San Diego County. There is no clear pattern in the reduction of PM2.5 exposure by race, indicating that while reducing vehicle emissions may benefit all communities, additional measures are needed to address existing exposure disparities among different populations.

Key findings include the importance of addressing light-/medium-duty diesel engines as a significant NOx emission source, given their persistently high emissions and slow rate of decrease. The fuel-based inventory performed well against observational data and estimated higher on-road emissions in comparison to regulatory inventories. This research provides a structure for how fuel-based emission inventories can be used to understand and target emission reduction efforts. Future studies should leverage new measurement technologies like the geosynchronous TEMPO satellite, model impacts on tropospheric ozone formation, evaluate impacts of reductions in primary PM2.5 emissions, and broaden the scope of the vehicle emission inventories developed in this work to understand source contributions to air pollution more broadly across the US and in other countries as well.