Boston University
Modeling and observations of urban greenhouse gas and air pollutant co-emissions
Abstract
dc:description.abstractThe combustion of fossil fuels for transportation and energy emits both carbon dioxide (CO2) as well as many other pollutants including, but not limited to, nitrogen oxides (NOx) and particulate matter (PM). It is broadly understood that there are substantial future climate and health risks posed by these constituents. However, understanding the sources of emissions and their variability at an intra-urban scale, where there is the greatest potential for policy action, continues to be a challenge. This is due to sparse observational data coverage, especially of all co-emitted species, making it difficult to gain sector-level insight at time and spatial scales relevant to stakeholders. Uncertainties in bottom-up inventories are driven by varying estimates of combustion efficiencies, input fuels and the rapidly shifting magnitudes of anthropogenic source emission fluxes. Throughout this dissertation I apply open-source emissions datasets, meteorological models, and dispersion modeling techniques to establish methodologies that bridge these gaps in knowledge by linking ambient measurements to source/sector emission behavior and impact. This work provides a foundation for improving estimates of intra-urban scale source sector emissions in cities in the United States and abroad, and future applications of these methodologies that can be used to bolster emission reduction policy programs like Climate Action Plans (CAP) and Environmental Impact Assessments (EIA). Chapter 1 of this dissertation provides a general overview of these topics and their importance. In Chapter 2 of this dissertation (Radford et al., 2021) we set out to improve EIA practices in areas that are data poor and where most EIAs, if created, are publicly inaccessible. To achieve this, we conduct a modeling study to establish an open-source methodology for estimating health risks from fine particulate matter exposure (PM2.5) emitted from coal-fired power plants in data poor areas in developing countries. We find that open-source and globally accessible meteorological models and remote-sensing terrain data can provide the necessary surface and atmospheric information to accurately supplement incomplete local observations. The results from this open-source methodology are shown to be similar to proprietary and/or much more computationally intensive methods. In Chapter 3, we examine the representativeness of urban monitoring to capture emissions, and quantify sector-level influence urban CO2 and NOx variability. We convolve hourly 1km resolution anthropogenic CO2 emission inventories across 10 source sectors and Massachusetts specific NOx:CO2 source sector emission ratios with a Lagrangian transport model to simulate local source sector contributions to ambient CO2 and NOx enhancements during the wintertime in Boston, MA. We are able to reproduce observed diurnal weekday and weekend enhancements for CO2 and weekday enhancements for NOx, and find that these patterns are driven by different source sector predominance. We also find that the location of air quality monitors impacts their representativeness and is critical to constraining conclusions about source sector contribution from a single monitor and applying these conclusions to an entire urban area. Chapter 4 examines source sector influence during severe air quality events and investigates sources of major disagreement between predicted and observed ambient NOx:CO2 ratios. We apply the methodological framework established in Chapter 3 to model enhancement in the Salt Lake Valley during wintertime persistent cold air pooling (PCAP) periods. We find that the overall difference in NOx:CO2 ratio enhancement between PCAP and non-PCAP periods is not statistically significant, but that there is a statistically significant higher NOx:CO2 ratio during PCAPS at nighttime. Analysis of the emission inventories suggests on-road gasoline and diesel vehicles are the most significant source of NOx emissions in the Salt Lake Valley, but that NOx emissions, and NOx:CO2 ratios, may be overestimated by almost a factor of three. From our analysis we hypothesize that this substantial gap between inventory derived ratios and observed ratios arise from 3 factors: NEI overestimation of urban NOx emissions in the Salt Lake Valley region, persistent CO2 emission underestimation by inventories, and unaccounted for wintertime chemistry leading to shorter than expected lifetimes for NOx emissions during daytime. Chapter 5 summarizes the conclusions drawn in Chapters 2, 3, and 4 and provides concluding remarks and suggestions for future work.
Degree
thesis:*- Name thesis:degree_name
- Doctor of Philosophy
- Level thesis:degree_level
- doctoral
- Discipline thesis:degree_discipline
- Earth & Environment
- Grantor
- Boston University
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Radford, Arden Caroline
- Advisor dc:contributor.advisor
-
- Geddes, Jeffrey A
Subjects
dc:subject × 7Rights
dc:rights- Statement dc:rights
-
- Attribution 4.0 International
- Licence dc:rights.uri
- Language dc:language.iso
- en_US
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/2144/53705