UC Berkeley

Access to reliable and affordable transportation shapes opportunities for employment, education, healthcare, and food. Yet for many urban residents, particularly those without access to a private vehicle, mobility remains a challenge. Public transit networks can fall short in terms of coverage, frequency, and flexibility. Carsharing has long been positioned as one strategy to address these gaps by offering short-term vehicle access without the costs and responsibilities of ownership. Since its early implementation in Zurich in 1948, carsharing has expanded globally and evolved into multiple service models, including roundtrip systems, where vehicles must be returned to their point of origin (e.g., Zipcar), and one-way systems, which allow for pickup and drop-off at different locations. Within the one-way model, some services are station-based, requiring users to pick up and return vehicles to designated stations (e.g., BlueLA), while others operate as free-floating systems that allow vehicles to be parked anywhere within a defined service area (e.g., Evie Carshare). These carsharing systems can offer flexibility and convenience, particularly in dense urban areas.

However, the benefits of carsharing have not been equitably distributed. Research has shown that users tend to be younger, higher-income, and multimodal travelers, while cost, carsharing access, and digital barriers have historically limited adoption among lower-income households. In response to these disparities and energy and environmental concerns, electric vehicle (EV) carsharing has re-emerged in recent years. EV carsharing combines the flexibility of shared mobility with the climate benefits of electrification. While EVs have been included in carsharing programs since the 1990s, advances in battery technology, expanded charging infrastructure, and targeted policy incentives have accelerated their integration. EV carsharing promises to reduce urban air pollution and greenhouse gas (GHG) emissions, but it also introduces new challenges, including infrastructure requirements, operational constraints, and questions around long-term service viability.

My dissertation focuses on one-way EV carsharing as a potential strategy for addressing both environmental and social equity goals. I investigate whether EV carsharing reduces energy consumption and associated GHG emissions, how different user groups engage with these services over time, and whether carsharing improves access to essential destinations such as grocery stores. These questions are especially relevant in underserved communities, where mobility barriers and transportation-related pollution are more pronounced.

Research GapsMy research identifies the following gaps in the EV carsharing literature: 1. Environmental Impact Measurement: Previous studies have extensively documented the environmental benefits of roundtrip and one-way carsharing; however, most of this work focuses on conventionally powered vehicles in higher-income, urban neighborhoods. The electrification of carsharing fleets has been recognized as a promising strategy for reducing emissions, yet studies evaluating EV carsharing’s energy and environmental impacts in underserved communities remain limited. Additionally, prior assessments estimate emissions based on VMT reductions and fuel economy assumptions, rather than directly measuring electricity consumption and the emissions associated with EV charging. 2. User Retention and Behavior Over Time: While previous studies have examined the sociodemographic profiles and tracked usage patterns and impacts such as VMT reduction and changes in vehicle ownership over time, there has been less focus on longer-term engagement and retention. 3. Grocery Access: Although access to a private vehicle is one determinant of food access, few studies examine the relationship between carsharing and grocery store access. Research on food accessibility has typically relied on static proximity metrics, such as straight-line distances or fixed travel times, which are unlikely to capture the accessibility experienced under real-world conditions from both a spatial and temporal perspective.

To address these gaps, my research is guided by the following five hypotheses:1. One-way EV carsharing reduces VMT and GHG emissions, with electrification further amplifying these effects. 2. One-way EV carsharing user activity patterns, including trip frequency, duration, and travel distances, differ between general and low-income qualified members, with low-income users demonstrating higher frequency of use, longer trip durations, and greater travel distances. 3. One-way EV carsharing membership type (i.e., general or reduced-cost for low-income qualified users) influences the duration of user active engagement and retention, with low-income users remaining active in the service for longer periods than general members. 4. Spatial-temporal accessibility from one-way EV carsharing stations to different grocery store types varies based on store type, traffic conditions, and trip departure times. 5. One-way EV carsharing users differ in their perceived grocery access and station usage patterns, with low-income users reporting greater improvements in access.

Research Approach and ContributionsTo test these hypotheses, I follow a mixed-methods approach that combines quantitative analysis of trip activity data with survey data and spatial modeling techniques. I focus on BlueLA, a one-way, station-based EV carsharing service launched in 2018 in Los Angeles, California. In a one-way, station-based system, vehicles must be picked up and returned to designated stations, but not necessarily the same one, allowing for one-way travel between fixed locations. BlueLA was designed to expand clean mobility access in underserved communities and operates under a tiered membership structure that includes both general population (Standard) and low-income qualified (Community) users. Community eligibility is determined by the operator (i.e., Blink Mobility) based on household income thresholds (e.g., $41,700 or less gross annual income for a single-person household) or enrollment in public assistance programs (e.g., SNAP, Medicaid, CalWORKS) . BlueLA’s location in Los Angeles—a city shaped by car dependence, sprawling development, and transportation equity challenges—serves as my empirical case. As a graduate student researcher with the Transportation Sustainability Research Center (TSRC) at UC Berkeley, I worked on a California Air Resources Board (CARB)-sponsored evaluation of BlueLA, which provided access to the data employed in this study.

My analysis is based on four main data sources. First, I use a trip activity dataset (n=59,112 trips from 2021–2022) provided by Blink Mobility in collaboration with CARB. This dataset includes detailed information on trip times, distances, station use, and vehicle charge levels across a two-year period. Second, I analyze a user survey that I co-designed with Professor Susan Shaheen and Dr. Elliot Martin. The survey (n=215 responses) captures information on vehicle ownership, travel behavior, BlueLA’s impact on access and mobility, and user experience. Third, I incorporate a service area resident survey (n=1,017 responses) co-developed with Professors Elizabeth Deakin and Susan Shaheen and Dr. Martin to capture insights from non-users. This addition offers a broader view of adoption barriers and the perceived value of EV carsharing among those living within the BlueLA service area. Lastly, I use several external datasets to support spatial and demographic analysis, including grocery store data from the United States Department of Agriculture (USDA) Supplemental Nutrition Assistance Program (SNAP) Retailer Locator (n=5,888 stores), U.S. Census and American Community Survey data, and routing and store attribute data from Google, Yelp, and Mapbox APIs.

My dissertation makes three primary contributions to the study of one-way EV carsharing and its environmental, behavioral, and equity implications: 1) addressing knowledge gaps by testing hypotheses related to EV carsharing’s impact on emissions, user retention, and accessibility, 2) methodological contributions, introducing novel approaches including a finite state machine model for measuring EV charging emissions, clustering and survival analyses for modeling user retention, and a dynamic accessibility model incorporating real-time travel conditions and store characteristics for assessing grocery access, and 3) empirical contributions, providing evidence on how different user groups engage with EV carsharing, the extent of its environmental benefits, and its potential to improve grocery store access for underserved households.

Addressing Gaps in Knowledge of CarsharingThis dissertation addresses the three research gaps including the: 1) limited understanding of EV carsharing’s environmental impacts in underserved communities and in contrast to the general population of users; 2) lack of research on user retention and engagement patterns across different socioeconomic groups; and 3) underexplored relationship between EV carsharing and access to grocery stores among underserved households. My dissertation contributes to the literature, addressing gaps by testing the five hypotheses noted above, within the BlueLA context.

Methodological ContributionsMy dissertation builds upon established methodologies in carsharing impact assessment, user retention analysis, and accessibility modeling while introducing new approaches to improve measurement precision and applicability. In Chapter 2, I extend the existing framework by Shaheen and Martin (Martin and Shaheen, 2011; Martin and Shaheen, 2016), which combines trip activity data with survey responses to estimate carsharing impacts on vehicle ownership and travel behavior. I contribute by developing a finite state machine model that infers vehicle charging activities from trip data. This approach allows for a direct measurement of electricity consumption, improving upon previous studies that estimated emissions solely based on VMT reductions and fuel economy standards. To calculate emissions, I apply time-variant and efficiency-adjusted emission factors, following the methodology developed by Zhang, Martin, and Shaheen (2025), which accounts for hourly fluctuations in grid emissions intensity and charging efficiency losses. In Chapter 3, I build on previous carsharing research that has largely relied on survey-based demographic analyses by applying clustering analysis to trip activity data, segmenting users based on actual usage patterns rather than just self-reported behavior. Additionally, I employ survival analysis , a technique widely used in engineering (e.g., machinery failure modeling) and consumer analytics (e.g., user retention in social media and ridehailing platforms) to quantify retention patterns and disengagement probabilities in EV carsharing. I also collaborate with Deakin, Shaheen, and Martin to implement a service area resident survey to examine the perceptions of non-users, providing a complementary perspective to the user survey and expanding the understanding of one-way EV carsharing adoption barriers. In Chapter 4, I advance existing accessibility models that rely on static proximity measures such as straight-line distances and cumulative opportunity counts, by integrating spatial-temporal accessibility modeling that accounts for real-world traffic conditions, store operating hours, and price levels. I construct a custom grocery access dataset by augmenting the USDA SNAP Retailer Locator dataset with store attributes from Google and Yelp APIs, allowing for a more realistic assessment of when and where grocery stores are accessible to BlueLA users. Collectively, these methodological contributions establish a replicable framework for evaluating one-way EV carsharing programs across environmental, behavioral, and accessibility dimensions.

Empirical Contributions Empirically, my dissertation provides data-driven insights into the role of one-way EV carsharing in promoting sustainability, accessibility, and transportation equity. Through the case of BlueLA in Los Angeles, I analyze how a publicly supported one-way, station-based EV carsharing program operates in a complex urban setting characterized by car dependence and transportation inequities. Each chapter focuses on a different dimension: environmental outcomes in Chapter 2, user behavior and retention in Chapter 3, and grocery store access improvements in Chapter 4. These empirical findings offer practical insights for carsharing operators and policymakers seeking to improve EV carsharing services.

The next section presents key findings from each chapter, followed by a discussion of limitations and directions for future research.

Key FindingsSystem Overview The BlueLA system is composed of 40 stations, each with five dedicated parking spaces equipped with Blink IQ 200 Level 2 alternating current (AC) chargers. The fleet currently includes 100 Chevrolet Bolt EVs, each with a battery capacity of 66 kWh, an estimated range of 260 miles, and a full charging time of approximately 8.5 hours. As of April 2025, Standard members pay $5 per month, and income-qualified Community members pay $1 per month. Standard members are charged $0.25 per minute (i.e., $15 per hour), while Community members pay $0.20 per minute (i.e., $12 per hour). These rates were increased in March 2024; prior to that, Standard and Community members paid $0.20 and $0.15 per minute, respectively, with a 15-minute minimum rental. The minimum rental period is now one hour, with a 12-hour cap. BlueLA also offers two promotional packages: a three-hour package priced at $28 for Standard members and $22 for Community members, and a five-hour package priced at $45 and $36, respectively. After the package period ends, per-minute rates apply.

From an operational perspective, we defined the BlueLA service area as the geographic region within a 30-minute walk from any BlueLA station, based on user-reported walking thresholds from our survey. While the City of Los Angeles currently has a total population of about 3.88 million and 1.46 million households (of which 12% are carless and 18% receive cash public assistance or food stamps/SNAP), the defined BlueLA service area—covering approximately 891,283 residents (23% of the city total) and 377,280 households—exhibits higher proportions of carless (22%) and SNAP-receiving (19%) households. Approximately 81% of residents in this area are age 18 or older and thus eligible to use BlueLA, although users must also hold a valid driver’s license and possess a credit or debit card. Stations are densely clustered in central neighborhoods such as Downtown Los Angeles, Koreatown, and Westlake, which are characterized by multi-story residential buildings and limited parking. In contrast, peripheral areas like Hollywood have sparser station coverage, suggesting that some neighborhoods benefit from more accessible station networks while others rely on fewer stations. Within a five-minute walk of a BlueLA station, 19% of total households and 17% of the population are covered, including 20% of SNAP-recipient households and 26% of carless households. At a 10-minute walk, the shares rise to 49% of households, 46% of the population, 52% of SNAP households, and 62% of carless households.Between 2021 and 2022, the system recorded 59,112 trips taken by 3,074 users. Community members made up 37% of the total user base, but they accounted for 51% of all trips. BlueLA experienced growth across both membership types: in 2021, there were 1,632 users (988 Standard, 644 Community), which increased in 2022 to 2,121 users (1,299 Standard, 822 Community). Trip activity data show that the average trip distance was 35 miles, with a median of 27 miles. Average trip duration was five hours, with a median of 4.8 hours. Despite being a one-way carsharing service, 36% of all trips were roundtrip (i.e., began and ended at the same station).

The user survey (n=215), which we conducted in December 2022, offered further insight into member demographics, motivations, and experiences. We distributed the survey to the entire BlueLA user base at the time (n=4,122), and it was available in both English and Spanish. Participants were incentivized by a raffle for one of forty $50 Amazon gift cards. We employed two types of anonymous user identifiers: one for linking survey responses to trip activity data and one for identifying membership status. We categorized respondents as “active” if they were subscribed at the time of the survey and “inactive” if they had previously subscribed but were no longer active. Of the 215 respondents, 60% were active members and 40% inactive; 56% were Standard members and 44% were Community members. Among all respondents, 71% reported not owning a personal vehicle. Income levels varied significantly by membership type: 74% of Community members reported household incomes under $50,000, while 68% of Standard members reported incomes over $75,000. Regarding educational attainment, 74% of Standard members and 54% of Community members held a bachelor’s degree or higher. This relatively high level of educational attainment is common in surveys and may reflect sampling bias toward more educated individuals. Age distribution among respondents was as follows: 7% under 25, 31% ages 25–34, 28% ages 35–44, 18% ages 45–54, 12% ages 55–64, and 4% age 65 or older. Language use at home was predominantly English (89%), with others reporting Spanish (6%), and the remainder including Arabic, French, Korean, and Tagalog.

Among all survey respondents (n=215), 67% reported using BlueLA for non-shopping errands, 65% for grocery shopping, and 62% for social or recreational activities. Other purposes included healthcare appointments (38%), commuting to or from work (22%), accessing public transit (11%), and attending school (5%). When asked why they joined BlueLA, 63% of respondents reported that they did not have a personal vehicle and joined to gain additional mobility. Other reasons included avoiding replacement of an inoperative personal vehicle (11%), changes in lifestyle requiring a vehicle (7%), the need for an additional vehicle for flexibility (6%), and the convenience of parking BlueLA vehicles (4%). Barriers to continued use were primarily identified by inactive users (n=79). The most common reason for no longer using BlueLA was the difficulty of finding an available BlueLA vehicle nearby (67%). Other challenges included preferring other transportation modes (20%), cost (15%), difficulty finding parking at drop-off (11%), charging station access (9%), loss of access to a personal vehicle (9%), and loss of a valid payment method (4%).

Environmental Impacts of EV Carsharing (Hypothesis One)In Chapter 2, we assessed whether the environmental benefits of traditional carsharing extend to one-way EV-based systems. Using trip activity data (n=59,112 trips from 2021–2022) and a user survey (n=215), we estimated changes in personal vehicle use and associated GHG emissions. I contributed to this work by developing a finite state machine model to infer charging behavior from trip records and applying time-sensitive, efficiency-adjusted emissions factors—based on Zhang, Martin, and Shaheen (2025)—to estimate actual electricity consumption and resulting emissions from EV charging. Our analysis focused on a subset of 81 active BlueLA members whose survey responses were matched with their trip activity data. Based on this sample, we assessed changes in vehicle ownership and driving behavior that respondents explicitly attributed to their BlueLA use. Respondents were asked whether they or their household had shed a vehicle due to increased mobility through BlueLA, whether they would have still shed that vehicle without access to the service, and whether they had postponed the purchase of a vehicle that would otherwise have occurred. We also asked respondents to estimate the change in their annual personal vehicle mileage due to BlueLA access. To account for potential sampling bias, we applied a weighting procedure based on frequency of their BlueLA use. Survey respondents and the broader user population were grouped by usage frequency, and we calculated weights to adjust for differences in representation across these groups. These weights were then used to scale the reported impacts to reflect the full BlueLA membership base (n=3,074 users). After weighting, we estimated that 4% of users shed a personal vehicle due to BlueLA, while 30% postponed a vehicle purchase they otherwise would have made. Additionally, 10% of users reported driving fewer miles annually in their personal vehicles as a result of using BlueLA. These behavior changes formed the basis for estimating system-wide VMT and emission reductions.

The results indicated that participation in BlueLA led to significant environmental benefits. Accounting for vehicle shedding, postponed vehicle purchases, and reduced driving, we estimated a net reduction of 460,845 VMT and 656 metric tons of GHG emissions over the two-year study period. Standard members contributed (n=1,931) 1,001,000 miles of BlueLA travel and Community members (n=1,143) 1,051,000 miles, yet the greater behavioral changes among Community members resulted in larger net GHG reductions (371 metric tons vs. 234 metric tons). On average, each BlueLA vehicle replaced 16 privately owned vehicles. We also compared the VMT and associated GHG emissions produced in the presence and absence of BlueLA. In the presence of BlueLA, we combined VMT and GHG emissions resulting from personal vehicle and BlueLA driving, and we deducted those avoided as a result of vehicle shedding, suppression, and reduced personal vehicle driving behavior. In the absence of BlueLA, we combined VMT and GHG emissions resulting from personal vehicle driving and those previously avoided but now realized from vehicle shedding, suppression, and reduced personal vehicle driving behavior. Comparing system-level scenarios, BlueLA reduced VMT by 34% and GHG emissions by 48% relative to a baseline without the service. Additionally, we compared the emissions produced by the BlueLA fleet to those of a hypothetical gasoline-powered fleet by translating the total VMT driven by BlueLA vehicles into GHG emissions using the EPA’s estimated real-world fuel economy for 2021 of 31.85 miles per gallon (MPG). When compared to a hypothetical internal combustion engine fleet, the BlueLA EV fleet produced 43% lower GHG emissions.

User Behavior, Adoption, and Retention (Hypotheses Two and Three)Chapter 3 examined how activity patterns and retention rates varied across user groups. We applied clustering and survival analyses to the BlueLA trip data and drew on both the user survey (n=215) and the service area resident survey (n=1,017) to understand motivations for use, challenges to continued participation, and differences between Standard and Community members. I contributed by developing the clustering and survival analyses for this study. In this context, retention refers to the length of time a user remained actively engaged with the service, while engagement was measured based on patterns in trip frequency over time. Specifically, we defined disengagement as a statistically significant increase in the inter-trip interval, using a Z-score threshold applied to a rolling window of one week. The analysis revealed distinct behavioral profiles. Community members had longer engagement (median of 449 days vs. 362 days for Standard members), traveled longer distances (27 miles on average vs. 15 miles), and used the service for longer durations (4.8 hours on average vs. 2.4 hours) compared to Standard members. Another segment, composed of both member types, exhibited frequent long-distance use (49 miles and 5.6 hours on average). Retention declined steeply after the first 500 days, but Community members stayed active longer overall. Survey responses showed that lack of vehicle access was the most common reason for joining BlueLA across both groups, while barriers to sustained use included vehicle and charging station availability, affordability, and parking constraints. These findings underscore the importance of accounting for socioeconomic differences in designing and managing EV carsharing systems. Understanding the retention dynamics of different user groups is essential for improving service longevity and equitable access.

Grocery Access and Accessibility Outcomes (Hypotheses Four and Five)In Chapter 4, we assessed whether one-way EV carsharing has the potential to improve access to grocery stores for households without personal vehicles using a spatial-temporal accessibility model, trip activity data, and user survey responses. I contributed by developing the accessibility model, constructing an enriched grocery store dataset using third-party APIs, and integrating these components with trip and survey data to evaluate accessibility outcomes. The findings showed that grocery shopping was a frequent trip purpose: 69% of Community members and 61% of Standard members used BlueLA for this purpose. Reported improvements in grocery access were higher among Community members (84%) than Standard members (71%). While Community members also reported greater perceived benefits in travel distance, speed, and scheduling flexibility, these differences were not statistically significant. The only statistically significant difference was in the ability to make late-night grocery trips, reported by 66% of Community members compared to 37% of Standard members.

Proximity to BlueLA stations was critical for underserved populations. For this analysis, we defined the service area to include all census blocks within a 30-minute walk of any BlueLA station, forming the boundary for our accessibility analysis. Within this area, 26% of carless and SNAP-receiving households were within a five-minute walk and 62% within a 10-minute walk of at least one BlueLA station. Trip activity data revealed that 11 stations accounted for half of all BlueLA trips. While all stations contribute to overall coverage, at least one of these high-volume stations was located within a 30-minute walk for 46% of carless households in our defined service area. More specifically, 11% of carless households were within a five-minute walk and 26% within a 10-minute walk of at least one of these highly active stations. However, differences in station-level activity between Community and Standard members were minor and statistically insignificant. Scenario-based modeling showed that smaller grocery outlets (e.g., convenience stores) were more accessible, while supermarkets and superstores required longer trips and incurred higher costs, particularly as access expanded beyond nearby stores. Under low-traffic conditions, reaching 1–5 of the nearest stores required about 11 minutes of roundtrip travel. Including an average in-store shopping duration of 46 minutes, the cost for Community members—who pay $0.20 per minute with a one-hour minimum—was approximately $12. Accessing 25% of available stores required 20 minutes of roundtrip travel, bringing the total trip time to around 70 minutes and cost to $14. Extending access to 50–100% of stores under higher traffic conditions raised the total trip duration to 79–100 minutes and cost to $16–$20. However, these estimates do not account for the additional time and effort required to walk to and from the nearest station or unload groceries, which may further increase the total trip time and user cost. The timing of grocery trips was also important. Store accessibility peaked between 10 AM and 6 PM, while late-night options were limited, often forcing users to rely on convenience stores with limited product variety. These findings highlight the value of aligning EV carsharing grocery trips with store hours and lower traffic periods to minimize travel time and rental expenses. To further enhance affordability for low-income users, carsharing operators could implement targeted incentives such as discounted short-term rentals or loyalty programs. In addition, grocery retailers could help address late-night accessibility gaps by extending store hours in locations that serve underserved households.

Limitations and Future WorkMy dissertation had several limitations that should be considered when interpreting the findings. One key limitation was the reliance on user survey data with a relatively small sample size, which introduced potential sample and self-reporting biases. While we attempted to mitigate selection bias by including a service area resident survey to capture responses from both users and non-users of BlueLA, the voluntary nature of survey participation meant that respondents may not fully represent the broader user population. Additionally, survey responses were self-reported, which could have led to recall errors or unintentional misrepresentation of travel behaviors. Although we applied weighting techniques to better align responses with BlueLA usage patterns, these adjustments may not have fully accounted for non-response bias or systematic differences between survey participants and non-participants. Future research could improve upon this by expanding survey sample sizes by offering increased incentives to encourage participation, using stratified sampling methods to target population subgroups of interest (e.g., oversampling low-income users in underserved communities), or incorporating alternative data collection approaches such as interviews or longitudinal studies.

Another limitation was the estimation of EV charging behavior and emission impacts. The finite state machine model we developed relied on assumptions about charging times, rates, and maximum charge levels, based on the known specifications of BlueLA’s Chevrolet Bolt EV fleet. These assumptions were necessary due to the lack of direct charging event data, but they introduced uncertainty in the precise measurement of electricity consumption and emission calculations. Additionally, while we incorporated time-variant and efficiency-adjusted emission factors to provide a more accurate representation of environmental impact, the real-world carbon footprint of EV charging can vary based on electricity grid fluctuations, weather conditions, and energy demand cycles. Future research could refine emission assessments by obtaining direct charging data from carsharing operators and integrating higher-resolution energy grid data to improve emission factor precision.

The timeframe of the study (2021 to 2022) also posed challenges in generalizing findings beyond this period. This timeframe coincided with the COVID-19 pandemic and its economic aftermath, which likely influenced travel behavior, car ownership decisions, and mobility preferences. While our analysis found that BlueLA usage remained stable or increased during this period, external factors such as pandemic-related economic conditions, fluctuations in fuel prices, and changes in remote work policies may have impacted both carsharing adoption and retention trends. Additionally, our findings on grocery store access improvements were based on a snapshot of user behavior and store access conditions, which may change over time. Future research could address these concerns by conducting longitudinal studies to track changes in mobility patterns, user retention, and grocery accessibility under varying economic and policy conditions.

The focus on a one-way, station-based EV carsharing model also limited the broader applicability of these results to other carsharing systems. One-way EV carsharing presents unique benefits and challenges compared to roundtrip or free-floating carsharing models, which may exhibit different usage patterns, cost structures, and accessibility outcomes. The findings suggest that Community members relied more heavily on BlueLA, but they also faced retention challenges due to cost and station availability constraints, highlighting the need for a comparative analysis of different carsharing models. Future studies could expand this research by evaluating how roundtrip or free-floating EV carsharing services perform in terms of emission reductions, retention, and grocery accessibility, as well as assessing policy interventions that might enhance carsharing accessibility for low-income users. Additionally, while we observed that low-income users remained active in the service longer than general population users, the reasons for this retention gap remain unclear. Possible contributing factors include the discounted pricing offered to Community members, as well as the greater reported mobility benefits among lower-income users. Future research could explore causal factors that drive longer retention among low-income users, including the role of affordability, unmet mobility needs, or broader lifestyle factors. Future work could also examine the potential of alternative policy interventions—such as providing eligible households with flexible transportation cards usable across public transit, ridehailing, carsharing, or bikesharing platforms—to support mobility and access. A comparative approach contrasting multi-modal integration versus service-specific strategies could reveal whether holistic access or targeted investments are more effective at advancing mobility equity in underserved communities.

The dynamic grocery accessibility analysis provided a more detailed assessment of food access than traditional static distance measures, but certain limitations remained. The accessibility model assumed that larger grocery stores provided greater product variety than smaller outlets, based on prior research, but it did not directly incorporate price or inventory data. The model relied on scenario-based routing and store-level attributes from third-party APIs, which may contain inaccuracies, temporal fluctuations, and incomplete information on store operations. Additionally, travel time estimates were limited to automobile trips and did not account for other transportation modes such as ridehailing or public transit. While the model integrated store operating hours and traffic conditions, it did not fully account for seasonal variations in grocery shopping behavior or day-to-day fluctuations in store availability. Additionally, because we did not have the data, the analysis did not include users’ home locations, limiting the ability to measure door-to-door accessibility. Lastly, while the model quantified access in terms of the number of reachable stores, it did not capture individual decision-making behavior or store choice preferences, where prior research has shown that presenting users with a large number of options can lead to choice overload, particularly in situations involving complex decisions. Future research could refine this approach by integrating residential location data, multi-modal travel scenarios, detailed inventory and pricing data, and behavioral models that account for consumer preferences, perceived effort, and decision complexity. Despite these limitations, my dissertation provides a detailed evaluation of one-way EV carsharing’s environmental, behavioral, and accessibility impacts, offering valuable insights for researchers, policymakers, and carsharing operators. By addressing these challenges in future research, scholars can further refine EV carsharing’s role in sustainable urban mobility, accessibility improvements, and emission reduction strategies.

This dissertation examines how transportation barriers impact community college (CC) students' ability to achieve their academic objectives. Although CCs serve as critical pathways to educational and economic opportunities for marginalized populations, transportation accessibility remains an overlooked variable in student success. The transition to a CC often signifies a meaningful achievement for a broad segment of the population. For some, securing adequate funding and initiating formal application procedures can seem like insurmountable hurdles. CCs, also known as two-year colleges or junior colleges, are a critical component of the U.S. higher education system. CCs offer an extensive range of courses that lead to associate degrees, certificates, or self-paced learning options, often allowing students to earn credits transferable to four-year institutions for bachelor's or advanced degrees (Kerr & Wood, 2022). CCs are critical to enhancing equitable access to higher education for first-generation college students, individuals from groups historically excluded from higher education, older learners, and workers seeking to enhance their skills (Community College Research Center [CCRC], 2021). An overarching goal of my dissertation is to bring awareness to the transportation challenges faced by CC students when accessing campus and to highlight the need for additional work in this area. There is some research that looks at how certain data points such as socioeconomic status, race, and access to a car are linked to CC access, but there is a lack of research on how access and associated barriers to it affects student completion, including reliable and affordable automobility, public transit, and other modes.

This abstract describes the research objectives and overarching methodology and highlights key results from these contributions and concludes with a summary.

These ongoing barriers remain prevalent among students in rural, suburban, and economically disadvantaged areas and can become the most significant determining factor in their academic outcomes. This research employs three complementary approaches: (1) a comprehensive literature review of transportation accessibility in higher education contexts, (2) spatial analysis comparing CC accessibility in Texas and California, and (3) case study analysis of hypothetical student experiences at Austin Community College.

Chapter 1 - Obstacles and Opportunities for Improving Transportation Access to Community College Education

The goal of Chapter 1 is to better comprehend the transportation access barriers faced by CCs and develop effective strategies to address them. I conducted an in-depth analysis of 88 studies examining transportation-related issues within higher education institutions. My findings revealed that while previous research has delved into the transportation challenges encountered by communities at universities, it has largely neglected the specific transportation issues faced by CC students. While some research exists in the context of CCs, there is a significant gap in empirical analyses that explore the modal preferences of CC students, examine the correlation between transportation barriers and the academic success of CC students, and evaluate the effectiveness of strategies aimed at improving access to CC campuses.

To investigate past work relevant to college access, I performed a systematic literature review on transportation to higher education campuses. Since the existing body of work focusing on transportation access to colleges is limited, a broader search scope was employed to analyze enough studies. From these studies, relevant insights about college access were extracted. Key articles were identified using a hub-and-spoke search process. The analysis section aims to achieve three main objectives: 1) provide summaries of the 88 studies included in the literature review, 2) identify significant themes in the past work on transportation to higher education, and 3) explain why these themes are crucial for studying transportation access to colleges.

Shaheen et al. (2017) introduced a framework called the STEPS to Transportation Equity Framework, which serves as a tool for analyzing the various transportation barriers that individuals may encounter during their travels. The STEPS framework revealed that long commute distances, conflicting transit and course schedules, and housing unaffordability can hinder transportation access to CCs. The literature underscores the use of public transit pass programs to assist CC students with transportation, but there's an opportunity to explore innovative approaches to meet the growing access requirements of CC students.

Building on these insights from the literature review, Chapter 2 moves from theoretical understanding to geographic analysis by examining two major community college systems.

Chapter 2 - A Tale of Two States: Exploring Transportation Accessibility to Community College Education in California and Texas

I delve into the extent to which students must travel to reach their nearest CC campus, and the variety of degree and educational programs offered at each CC campus. To achieve this, I selected CCs in Texas and California as case studies to explore how access to education is influenced by the distinct policy models prevalent in these two states.

This investigation is driven by a desire to comprehend how transportation access to the two largest CC networks in the U.S., Texas and California, which collectively encompass the two largest states in the country, impacts pathways to employment and transfer routes to universities. These two states play a significant role in shaping the economic, political, and social landscape of the United States.

To explore accessibility to CCs, I selected a method of measuring access to CC campuses. Since education is a pathway to employment opportunities and other benefits, the variety of degree and certificate offerings provided by a CC is a measure of the opportunities available. To measure the number of offerings available at CCs, I utilize data maintained by the Integrated Postsecondary Education Data System (IPEDS). I also examine the number of opportunities available at CC campuses that are accessible to individuals in their vicinity.

To gain insights into the characteristics of individuals with better access to CCs and those with limited access, I measure access proxy variables alongside three socioeconomic variables collected from the 2021 five-year American Community Survey (ACS) census tract estimates.

To investigate our research objectives, I selected statistical tools that enable us to draw inferences from the data used to measure transportation access to pathways to employment and education through CCs. The method I utilized to assess sociodemographic affluence influences access to CCs and their educational offerings is crosstab analysis. Crosstab analysis involves organizing all data points measured into bins, or crosstabs, based on the magnitude of the variable used to define the crosstabs.

When it comes to access to CC campuses and education, economic affluence plays a more significant role in California, while educational affluence holds a more prominent position in Texas. In contrast, mobility affluence has a minimal impact on access in both states. On average, educational offerings at the nearest CC campus tend to be higher in Texas than in California. This disparity can be attributed to the differences between California's centralized CC policy model and Texas' decentralized CC policy model. While one state may not clearly offer better CC campus access compared to the other, a comprehensive analysis of access provides a broad understanding of CC access in the two states. However, further exploration should be conducted through more detailed access case studies that are situated in both states.

While Chapter 2 provides a macro-level analysis of accessibility patterns across two states, Chapter 3 zooms in to examine the lived experiences of individual students navigating these transportation systems.

Chapter 3 - A Case Study Analysis of Transportation Accessibility to Community College on Public Transit

This study delves into the accessibility of public transit to CCs through a comprehensive case study analysis. The primary objectives are to:

1) Understand the impact of transportation accessibility challenges on the lives of CC students.

2) Show that transportation accessibility should be recognized as a significant barrier hindering CC students' ability to achieve their academic goals.

3) Develop a framework to comprehend how equity factors influence student access to CCs via public transit.

To address the real-world transportation accessibility challenges faced by CC students, I developed a qualitative research methodology based on hypothetical student profiles. This approach allowed us to gain a deeper understanding of their transportation burdens by presenting realistic barriers and challenges in a definable and descriptive manner. In the fall of 2019, StudentMoveTO, a Toronto survey, conducted the largest survey ever conducted to better understand student travel patterns, experiences, and preferences. The StudentMoveTO survey served as a proxy and framework for my approach to this methodology. Based on these assumptions, I analyzed the data and demographic characteristics of 200 students from the Toronto survey to create 10 hypothetical student profiles that served as a proxy for conducting our research with Austin Community College (ACC) students. For example, my analysis and coding of the student demographic characteristics in the sample revealed approximately 14 significant and repetitive elements (race, age, gender, income, enrollment status, employment, major parental status, transit time, frequency of travel, multiple campus transit, etc.) that correlated with transportation accessibility. Using this framework, I obtained publicly accessible data from ACC and selected a representative sample of approximately 200 students. I then followed the same methodology process as the Toronto survey to calculate and identify 10 hypothetical student personas/profiles that were most likely to matriculate at ACC for the purpose of this research.

After presenting qualitative evidence of the transportation accessibility challenges faced by each student profile, I drew some clear observations, themes, and takeaways that advance this discourse. I categorized these as the "Seven Top Barriers of Transportation Accessibility for CC Students."

The findings from these three chapters collectively illuminate critical aspects of transportation accessibility for community college students. Chapter 1 established the theoretical framework and identified gaps in existing literature. Chapter 2 provided insights on how different state policy models relate to transportation accessibility and access to educational opportunities. Chapter 3 provided qualitative insights into how transportation barriers manifest in students' daily lives. Together, these findings contribute to our understanding of transportation accessibility in several important ways, as outlined below.

This dissertation demonstrates that (1) transportation barriers significantly impact CC student persistence and success; (2) accessibility varies based on geographic location, economic status, and institutional policy models; and (3) addressing transportation equity is essential if CCs are to fulfill their mission of providing inclusive educational opportunities. These findings suggest that integrated transportation planning should be a core component of CC strategic initiatives.

In this research, I contribute to both transportation equity and higher education accessibility scholarship by establishing a clear connection between transportation barriers and educational outcomes at CCs, providing a methodological framework for assessing CC accessibility across different policy contexts, and identifying specific intervention points for improving student success through enhanced transportation access.

Future research should incorporate direct student surveys, real-time transportation data, and expanded geographic coverage.

References

Community College Research Center. (2021). An Introduction to Community Colleges and Their Students. Teachers College, Columbia University. [https://ccrc.tc.columbia.edu/publications/introduction-community-colleges-students.html]{.underline}

Kerr, E., & Wood, S. (2022, November 29). Everything You Need to Know About Community Colleges: FAQ. [https://www.usnews.com/education/community-colleges/articles/frequently-asked-questions-community-college]{.underline}Shaheen, S., Bell, C., Cohen, A., Yelchuru, B., & Booz Allen Hamilton, Inc. (2017). Travel Behavior: Shared Mobility and Transportation Equity (PL-18-007). [https://rosap.ntl.bts.gov/view/dot/63186]{.underline}

Shaw, K., Asher, L., & Murphy, S. (2023). Mapping Community College Finance Systems to Develop Equitable and Effective Finance Policy. Community College Research Center. [https://ccrc.tc.columbia.edu/publications/mapping-community-college-finance-systems-develop-equitable-effective-policy.html]{.underline}

Occasions: Poetry with Prose, 1836-1877 begins in an observation: that the reader of texts from this period is always encountering verse when it is least looked for. Misquoted in newspapers, tucked into the busy media environments of the emergent monthlies, deployed in friendship albums, journals, and sentimental novels, and woven into the tissue of the genre we now call literary prose, verse fragments are called on to perform a range of literary and social tasks that are easy for contemporary readings to miss. Though poetry of the period may reach toward the aesthetic independence now associated with “lyricization”, alongside such ambitions Occasions recovers a genre that invites, as often, its own decomposition and requotation. Much of the difficulty lies in how Ralph Waldo Emerson has been read. When foundational, but complex, manifestoes such as “The Poet” are read towards an “ideal” or “imperial” poet, the Transcendentalists’ perennially neglected verse may appear slight or stiff in comparison. But Emerson’s essay dwells also at length in the possibilities of minor verse, granted it be close at hand, just as he advocates in another essay for an “adventitious” poetics of “happy hits”, arguing verse gives “greater delight … in happy quotation than in the poem” (72). It is all these other, less mystified moments of American poetics—intermittent, immanently social, and prone to disappearance—that Occasions traces, among the Transcendentalists, and outward in nineteenth-century social and political life. In the aftermath of the Romantic revolution, “occasional verse” is a cloudy pejorative; Hegel dismisses such pieces as slight, even while recognizing the category may include most lyrics of distinction. To recover a firmer lexicon for Emerson’s “moments” and Henry David Thoreau’s “occasions” of poetry, I return to the rhetorical, commemorative, and oratorical cultures of the early nineteenth century, and to the vivid tradition of Revolutionary and early Republican topical satires derived from them. The influence of these throughlines on Emerson and Thoreau locates the occasion and the verse it produces, as the key to a poetic historiography of crisis and response in which the individual speaker is called toward his moment and audience. With this move, Occasions participates in the gradual erosion of what Rei Terada calls “the autonomy of the lyric object”—and, I contend, the lyric subject—that has anchored much twentieth-century literary theory. Indeed, in Thoreau’s radical appropriations of other’s poetic property, and in the extremes of anonymity he uses verse to create and court, I locate a transhistorical commons and a resistance to the cordon of single authorship. This anonymity, and this ability to make the past present, and vice versa, help clarify the explicitly social affordances that Emerson and Thoreau use verse to propose. One assertion of my readings of Emerson is that “The Poet,” as articulated in the 1844 essay of that name and elsewhere, is shadowed by another model figure of a poet significantly crosshatched by values such as wit, utility, conversation, lightness, and disappearance. Drawing on significant archival research, I suggest the close source of this figure is “Ellery” Channing. Though now a footnote, Channing was in many respects the most prominent exemplum of transcendental poetry in the 1840s; when Emerson withdrew his support of Thoreau’s verse, it was to transfer that endorsement and affection to Channing. I argue that Channing’s poetry and company remain important precisely because of the lightness, the responsiveness to social contexts, and the intermittence that Emerson finds in them. From this angle, the collaborative projects Channing assembles—the unpublished “Country Walks” (compiled from his, Emerson’s, and Thoreau’s journals), and his 1873 biography, Thoreau: The Poet-Naturalist—are striking for the way they figure the analysis of verse (in the former case) and occasional modes such as elegy (in the latter), as the both the materials and the mode of sociality and life narration. Occasions turns last to Thoreau’s inheritance of Emerson’s poetics. Thoreau is initially motivated by the imperial powers of the poet, and his sacralizing function with respect to American geography, as is seen in A Week on the Concord and Merrimack Rivers (1849); however, the sheer proliferation of verse and verse analysis in that text, which overwhelms its putative physical context, exposes the latent tension in Emerson’s system between the visions of the poet as a translator, and as a reorderer of nature. Walden (1854) marks Thoreau’s elaboration of the occasion into an epistemological principle that derives from, and involves poetry, while transcending verse form. I follow finally Walden’s sustained poetomachy with Channing, whose poetry Thoreau frequently and dramatically quotes; the central term of the debate is the cultural value to be accorded to the Poet. I argue that the peak of this debate, in the problem chapter “Baker Farm,” marks Walden’s rejection of ownership of land as of language, and its turn to a poetics of the commons, influentially modeling poems as malleable sites of linguistic reception and barter responsive to specific historical and local needs.

Much prior research has shown that people tend to view themselves as better than others, referred to here as better-than-other (BTO) effects. Much of that research has asked people to compare themselves to the “average” person. In this dissertation, I provide a more nuanced examination of this BTO effect. First, I measured self and peer perceptions in a field study of nine intact social communities, to test whether people hold more positive views of themselves than they do of each of their fellow community members on average. This examination replicated the general BTO effect, but found it to be weaker than effects typically found in prior work. Second, I tested whether the strength of the BTO effect depends on the target of comparison. Specifically, I assessed the social networks within social communities, and found that BTO effects were weaker in comparisons with close friends, and stronger in comparisons with others distant from the individual in the social network. Third, I measured BTO effects specifically along the Big Five personality dimensions, and found BTO effects were strongest for the most evaluative dimensions (i.e., agreeableness, conscientiousness, and openness), and weakest for the less evaluative dimensions (i.e., extraversion and emotional stability). In further exploratory analyses, I also found that closeness in the network interacted with personality judgment domain such that the effect of social closeness on the BTO effect was stronger for the more evaluative Big Five dimensions.

Computational models of cognitive processes have provided deep insights into specific mechanisms of human learning and decision-making. A natural corollary of research into the typical functioning of such mechanisms is to investigate how mental disorders might cause impairment of the underlying algorithmic processes. This line of research, frequently referred to as computational psychiatry, seeks to contribute to more personalized diagnoses and treatments of mental disorders by characterizing behavioral and computational profiles of psychopathology. In the current dissertation we focus on the role of latent dimensions of internalizing psychopathology in computational processes of learning and decision-making to provide insight into dissociable dimensional phenotypes. Chapter 1 of the dissertation provides a primer to computational characterizations of learning and decision-making, including recent and foundational research results in computational psychiatry which motivate the subsequent original research questions. In Chapter 2, we present an investigation of effortful decision-making in reward pursuit and threat avoidance. We use a detailed dimensional profile of depressive and anxious symptoms to characterize unique mechanistic impairments across frequently comorbid symptoms, with individual symptoms of depression relating to specific differences in effortful reward pursuit processes, while a dimensional characterization of anxiety relates to multiple mechanistic differences in effortful threat avoidance. In Chapter 3, we extend the space of psychopathological inquiry to include subclinical levels of mania in an investigation of decision-making under ambiguity in reward pursuit, loss avoidance, and threat avoidance paradigms. Here, individual differences in loss avoidance decisions under ambiguity show opposing effects of anhedonic depression and physiological anxiety in risk sensitivity, with depression showing increased risk sensitivity and anxiety showing reduced nonlinear risk valuation. Ambiguous decisions in threat avoidance reveal a significant dissociation between anhedonic depression and hypomania, with higher levels of hypomania associated with significantly less risk sensitivity in threat avoidance than anhedonic depression. We also find suggestive, although not concrete, evidence that sensitivity to threat magnitude may relate to comorbid variance between anxiety and depression, and we recommend additional work in this area of investigation. In Chapter 4, we present a detailed investigation of dimensional characterizations of anxiety in reinforcement learning impairment, using an experimental paradigm which differentially manipulates the relative load of working memory versus learning systems across task. Here we find that physiological, but not cognitive, anxiety is related to significant learning impairments which are contributed to both by slower learning processes and by increased rate of working memory decay. These results may have implications for a wide variety of extant reinforcement learning modeling studies in computational psychiatry by emphasizing the importance of an algorithmic account of supporting working memory processes in such investigations. Chapter 5 provides a discussion of the results in the framework of dissociating unique profiles of dimensions of internalizing psychopathology. As additional insights into the relationships of latent dimensions of psychopathology with algorithms of cognition are gained, the field as a whole is enabled to approach the idea of replicable computational phenotypes of psychopathology. The aim of this dissertation is not to operationalize such phenotypes at present, but rather to contribute to the growing body of knowledge within the field to enable and inspire more detailed investigations and replications of the findings contained herein.

The ongoing urbanization process is swiftly giving rise to megaregions, reshaping the urbanlandscape. Travel constitutes a vital aspect of urban life; hence, a comprehensive understanding of traffic dynamics is crucial for the efficient management of cities. Traffic dynamics refer to the patterns of movement exhibited by people and vehicles within a transportation system. These dynamics are typically characterized in terms of flow and speed, providing crucial insights into the functioning of the urban transportation system.

Traffic simulators are extensively used to analyze traffic dynamics in cities and regions.However, many traffic simulators used in regional studies have limitations. One primary limitation is the substantial data and computational requirements necessary to model a large urban region with high fidelity and speed. To tackle this challenge in traffic simulators, many researchers reduce the size of the road network, including only major arterials and highways; model a subset of the population; and/or aggregate travel demand to a smaller subset of network nodes, aiming to obtain rough estimates of travel volumes. However, the downsizing of the road network and the reduction/consolidation of demand lead to alterations in the characteristics and performance of the network. This approach can result in highly inaccurate predictions that fail to capture the actual dynamics and behavior of traffic. It can provide misleading information to agencies regarding the necessary investments for constructing, maintaining, or improving their roadway infrastructure, as well as decisions related to traffic management and incident response plans.

To address these gaps, this dissertation provides a detailed characterization of large-scaleregional traffic dynamics across diverse scenarios using a high-performance traffic simulator. It contributes to and leverages a scalable high-performance mesoscopic traffic simulation platform named Mobiliti, which incorporates various routing strategies to model real-world traffic conditions with high fidelity and speed. In the case of San Francisco Bay Area, Mobiliti can simulate 19 million vehicle trips on a road network with approximately 0.5 million nodes and 1 million links, representing 7 million drivers and 4 million truck trips in less than three minutes. Subsequently, using data analytics and machine learning models, we identify traffic patterns and characterize cities based on transportation-oriented typologies for large metropolitan regions.

Leveraging Mobiliti alongside real-world data sources, we delve into various facets of regionaltraffic dynamics, covering both novel and critical areas. Our investigation began by examining dynamic routing, a prevalent feature introduced by the widespread adoption of navigation apps. This feature introduces an additional layer of traffic control, thereby altering traffic dynamics on streets. Through modeling the varying penetrations of dynamic routing, we quantified its effects on the San Francisco Bay Area using metrics such as Vehicle Miles Traveled (VMT), Vehicle Hours of Delay (VHD), affected trips, and its impact on local roads, among others. Next, we recognized that different types of traffic routing, for example, prioritizing time savings versus prioritizing fuel efficiency, influence traffic dynamics in distinct ways. These dynamics, in turn, shape our cities and significantly impact quality of life. Consequently, we developed a framework to analyze these complex dynamics. We evaluated the impact of different routing strategies across multiple dimensions, examining their effects on neighborhoods, safety, environment, and more. Thirdly, network resilience is a critical factor in the San Francisco Bay Area, known for its interconnected bridges. Incidents in this region have the potential to disrupt multiple cities, impacting productivity and energy efficiency. Therefore, we focused on examining the vulnerability of the transportation system to such events and the cascading effects they may cause. A deeper understanding of these dynamics can assist in effective event response planning. The final two studies focus on cities within metropolitan regions. We begin by analyzing street network structures of cities and then proceed to incorporate additional transportation dimensions such as travel demand, traffic flow, and infrastructure. By employing clustering techniques, we identify city typologies and their respective characteristics. This analysis offers us the opportunity to reflect on past urban development efforts, learn from one another, and envision our future city-building objectives.

In total, the specific contributions of this dissertation are:

1. Analyzed the effects of dynamic routing and its varying penetration rates across avast metropolitan region using large-scale discrete event simulations, demonstrating its substantial influence on mobility metrics at a regional scale. Previous studies were constrained by geographic scale and a limited number of simulation runs, thus failing to capture the full impacts of dynamic routing. 2. Developed a novel multi-themed analytical framework called the Socially-Aware Evaluation Framework for Transportation (SAEF), aiding in comprehending how traffic routing and resulting dynamics impact cities within a region. Our framework’s indicators are carefully chosen to detect system changes when routing strategies are altered, with a focus on neighborhood-related indicators, often overlooked in existing frameworks. 3. Enhanced the evaluation of large-scale network disruptions by modeling dynamic route choices for travelers within a full-scale urban network, encompassing an entire day’s demand. Our method more realistically captures drivers’ behavior during incidents and we are able to capture the full impacts of incidents at scale, thus enabling the creation of better traffic management and response strategies. 4. Created city typologies for all cities within a metropolitan region based on street network structure, which can provide valuable insights into how drivers experience a city based on its street layout. To aid in this classification, we introduced a new metric for categorizing intersections that distinguishes between various types of 3-way and 4-way intersections based on geometric angles. By incorporating this geometric metric alongside existing centrality metrics, we have achieved better differentiation for improved typology generation, capturing the nuances between grid and other street typologies more effectively. 5. Developed transportation-oriented city typologies based on various dimensions including traffic flow, trip demand, multi-modal network, land use, and road network. These typologies serve as a foundation for facilitating the effective exchange of policies and resources, relying on a thorough understanding of traffic characteristics. We integrated metrics related to trip demand and traffic flow alongside commonly used metrics from road network, multi-modal network, and land use. This integration is crucial for capturing the travel behavior and traffic dynamics of cities, enabling the generation of meaningful and comprehensive typologies.

Each of the items mentioned above is described in more detail in the following paragraphs.

In the second chapter, we examined dynamic routing and its impact on large urban areasusing the Mobiliti traffic simulator. Over the last few decades, navigation apps have introduced a new level of traffic control and warranting study as they become pervasive and dictate street traffic flows. Previous work on dynamic routing has been constrained by limited geographic scale and a small number of simulation experiment runs, often requiring hours to complete a single simulation. This limitation poses a bottleneck for running multiple simulations and testing various what-if scenarios to identify the full range of rerouting impacts. We address this gap by utilizing high-performance parallel computing, large urban scale simulator Mobiliti, which can run a single simulation for the entire San Francisco Bay area in less than 10 minutes. We ran multiple simulations with varying penetration rates, revealing diminishing benefits of rerouting after a 70% penetration rate. We also found that dynamic rerouting effectively reallocates vehicle flows from heavily utilized highways and arterials to less congested neighborhood links, reducing overall system delay. Interestingly, the increased traffic volume on local roads does not always lead to congestion, as many links do not reach congested levels despite the increased flow. In summary, our analysis demonstrates, for the first time, the effects of varying penetration rates on traffic dynamics at a regional scale.

In the third chapter, we present an analytical framework called Socially- Aware EvaluationFramework for Transportation (SAEF), which assists in understanding how traffic routing and the resultant dynamics affect cities across a region across multiple dimensions. With the proliferation of real-time navigation routing apps, traffic dynamics in urban environments have changed, resulting in undesired effects that compromise safety and neighborhood health. Therefore, understanding these disparities in traffic distribution across various dimensions is crucial for decision-makers. While previous studies have created frameworks to assess the effects of wide-ranging transportation infrastructure changes or the adoption of smart city technologies, none have established a framework with indicators specific enough to capture the impacts of various traffic routing strategies on cities. Furthermore, existing frameworks and metrics lack translatability to identify the impacts of routing strategies, as crucial dimensions like safety or neighborhood considerations are not adequately addressed. Therefore in this work, our first contribution is developing a framework with a set of themes and indicators that can capture the impact of traffic routing holistically. We identified relevant indicators from the literature, organizing them into four themes: neighborhood, safety, mobility, and environment. When necessary, we developed new methodologies to calculate these indicators. A second contribution is the application of SAEF framework to four cities in the Bay area in the context of three different routing strategies - user equilibrium travel time, system optimal travel time, and system optimal fuel. The four cities were compared to understand how city structure and urban form play a role in the distribution of traffic dynamics. The results demonstrate that many neighborhood impacts, such as traffic load on residential streets and around minority schools, degraded with the system-optimal travel time and fuel routing in comparison to the user-equilibrium travel time routing. The findings also show that all routing strategies subject the city’s disadvantaged neighborhoods to disproportionate traffic exposure. With the widespread adoption of navigation apps, our intent with this work is to provide an evaluation framework that enables reflection on the consequences of traffic routing, allowing city planners to recognize the trade-offs and potential unintended consequences.

In the fourth chapter, we offer a set of evaluation tools designed to measure the impact of significanttransportation disruptions on a regional scale. We illustrate the application of these tools through a case study involving the closure of the Richmond-San Rafael Bridge in the San Francisco Bay Area. Evaluating the dynamics of transportation networks in the context of events can inform disaster plans and aid in traffic management strategies in preparation for or during an event. Existing research on road network disruptions often relies on short time frames and small-scale models, largely due to computational limitations that hinder the widespread adoption of large-scale urban simulation models. Consequently, smaller-scale micro-simulation models are commonly preferred for designing response plans, typically targeting selected highways and major arterials in close proximity to incidents. However, these studies face three key limitations. Firstly, they often rely on user-equilibrium assumptions for route choice, which fail to adequately reflect realistic driver behavior during incidents. Secondly, they use reduced road network representations due to computational constraints, typically focusing on small areas surrounding closures. Thirdly, they frequently extrapolate findings from peak periods to estimate daily impacts, potentially overestimating congestion due to differences in traffic dynamics between peak and non-peak periods. To address these gaps, our study employs a large-scale, mesoscopic simulation model with dynamic routing capability. This model enables us to simulate a full-scale urban network with an entire day’s demand, allowing for a comprehensive assessment of the regional traffic impact of the incident. Our findings indicate that the region experienced an additional 14,000 vehicle hours of delay and 600,000 vehicle miles due to the bridge closure. Furthermore, the median traffic volume on neighborhood streets in San Francisco, Vallejo, and San Rafael increased by more than 10%, highlighting the role of local roads in accommodating the traffic overflow, a factor often overlooked in prior studies. With large-scale modeling of a critical network disruption using dynamic rerouting capability, complete road network, and full demand, we provide valuable insights into the response dynamics of this specific event. In doing so, we demonstrate the value of such regional analyses to incident and disaster planning.

In the fifth chapter, we developed typologies to classify cities within a metropolitan area accordingto their street network characteristics. Spatial networks such as streets and transit lines influence urban dynamics and travel behavior. Analysing these patterns can also help identify how drivers experience city streets and understand the unique characteristics and challenges present in each urban environment. While previous studies have investigated global network patterns for cities, they have often overlooked detailed characterizations within a single large urban region. Additionally, most existing research uses metrics like degree, centrality, orientation etc., and misses the nuances of street networks at the intersection level, such as geometric angles formed by links at intersections, which could offer a more refined feature for characterization. To address these gaps, this study examines 94 cities in the San Francisco Bay Area, taking into account diverse road network features. We introduce a novel metric for classifying intersections, distinguishing between various types of 3-way, and 4-way intersections based on the angles formed at the intersections. Through the application of clustering techniques in machine learning, we have identified three distinct typologies - grid, orthogonal, and organic cities - within San Francisco Bay Area. Gridded cities are distinguished by their dense network of right-angled four-way and three-way intersections. These cities exhibit a compact layout with smaller link lengths and slower traffic speeds. On the other hand, orthogonal cities exhibit a street network configuration characterized by right-angled three-way intersections and longer street lengths. Organic cities represent a third typology, characterized by their irregular and non-grid-like street network. These cities feature long links with numerous dead ends and winding, circuitous roads. Our findings indicate that the integration of the new metric has improved our ability to distinguish between different types of cities, complementing the existing metrics. In gridded cities, the introduction of the new metric enhances the recognition of grid patterns by explicitly considering 90-degree intersection angles. Conversely, for non-gridded cities, a notable advancement is the ability to differentiate between various types of degree 3 nodes (3-way intersections). While many cities have a significant number of degree 3 nodes, the arrangement of these intersections can vary greatly due to angle variations, resulting in either 90 degree T intersections or non-T intersections. Our study showcases the effectiveness of the new metric in capturing these distinctions, facilitating the classification of cities with a high proportion of T intersections into orthogonal cities and those with non-T intersections into organic cities. The significance of this differentiation extends to how drivers navigate and experience intersections and streets within cities. Based on the angles, turns, and curves of the road network, driving experiences vary significantly. Therefore, understanding these nuances is crucial for optimizing traffic flow, enhancing road safety, and improving overall driving experiences for motorists.

In the sixth chapter, we expanded upon our previous city characterization work focused onnetwork structure by incorporating multiple transportation dimensions. As cities evolve and face shared challenges, the development of city typologies, rooted in a comprehensive understanding of traffic characteristics, becomes crucial for facilitating the effective exchange of policies and resources among them. Prior work on transportation based city typologies often fails to provide characterizations specific to a single extensive urban area, as it predominantly focuses on cities globally. Furthermore, these studies frequently overlook essential dimensions such as trip demand and traffic flow in their characterizations, despite their significant impact on street behavior and traffic dynamics. Therefore in this study, we develop a transportation-focused characterization for all cities within a large urban region, specifically the San Francisco Bay Area, California. We incorporate over 40 metrics across five transportation dimensions: road network, trip demand, traffic flow, multi-modal network, and land use. Using factor analysis and unsupervised machine learning clustering methods, we identified eight distinct typologies for the Bay Area: Live Work; Job and Activity Magnets; Anchor Cities; Multi-modal; Hyper-connected; Low-density residential; Mediumdensity Residential; Mixed-use residential. The results revealed that many clusters were characterized by features from travel demand and traffic flow dimensions, thus signifying their importance in typology generation. These typologies can serve as a basis to create discourse among Bay Area cities and determine if, through success/failure experiences, common strategies can be formed.

In total, the analytical framework and methods outlined in this dissertation provide detailedand nuanced insights into regional traffic dynamics, surpassing existing literature. By utilizing and contributing to the Mobiliti simulator, we modeled large urban areas with high fidelity and speed, enabling the testing of multiple “what if”scenarios for large metropolitan regions. Our investigation of dynamic routing and its varying penetration rate in Chapter 2 represents the first large-scale regional study examining the impact of real-time traffic routing. Furthermore, the SAEF framework presented in Chapter 3 of this dissertation represents the first analytical framework that captures the impact of traffic routing holistically. With the widespread adoption of navigation apps, this framework enables reflection on the consequences of traffic routing, allowing city planners to recognize the trade-offs and potential unintended consequences. The large-scale network disruption evaluated in Chapter 4 provides a suite of analytical tools for assessing disruptions at both regional and local levels. These tools enable the creation of enhanced traffic management and response strategies by capturing driver behavior more realistically. The typologies developed in Chapters 5 and 6 provide a comprehensive understanding of cities in a region, considering both network structure and overall transportation dimensions. The new metric introduced in Chapter 5 aids in quantifying the network more precisely, while the comprehensive use of various metrics from different transportation dimensions, particularly trip demand and traffic flow, facilitates a more thorough characterization of cities in Chapter 6. The identified typologies can catalyze dialogue among San Francisco Bay Area cities, facilitating the exploration of common strategies derived from shared experiences of success or failure. Ultimately, the findings presented in this dissertation contribute not only to enriching academic discourse on transportation dynamics but also carry practical implications for policymakers. They furnish invaluable guidance for crafting more effective and nuanced traffic management strategies for cities and large metropolitan regions, thereby shaping the future of urban mobility with precision and foresight.

The increasing complexity of global challenges demands a STEM-enriched approach to learning for all students, regardless of their future career paths. Challenge-Based Learning (CBL) is a pedagogical method to foster a STEM-enriched education, engaging students in the design of societally impactful, interdisciplinary solutions. To investigate the potential of CBL, specifically in the context of Undergraduate STEM Education (USE), it is crucial to assess students’ affective development such as their attitudes, beliefs, and self-perceptions related to STEM. This dissertation explores the impact of CBL on student affect through three interconnected studies centered on a large-enrollment Bioinspired Design course. Chapter 1 explores overall growth in measures of science connection—Science Identity (SciID), Science Self-Efficacy (Eff), and Internalization of Scientific Community Values (Val)—using the Tripartite Integration Model of Social Influence (TIMSI) framework. Results demonstrated significant pre/post increases in SciID and Eff across five semesters, with Val remaining stable. Item level analyses revealed specific impacts of CBL activities on these affective measures, particularly in developing students’ confidence in creating novel technologies. Chapter 2 investigates the equity of these affective growth outcomes across seven demographic variables. Results indicated that the observed increases in science connection were largely equitable across diverse student populations, with differences in SciID development based on STEM major status and class status. Chapter 3 introduces and validates a novel affective construct: Innovation Skills self-efficacy. Developed using the Berkeley Evaluation & Assessment Research (BEAR) Assessment System, this construct provides a more targeted measure of self-efficacy aligned with the Innovation Skills needed for the future STEM-enriched workforce. Results showed approximately one standard deviation of pre/post growth, with a large effect size in the context of educational interventions. Collectively, this dissertation showcases the potential of CBL approaches in USE to foster equitable development of science connection and Innovation Skills self-efficacy across diverse student populations through comprehensive, psychometrically robust assessments of student affect. This research underscores the importance of holistic approaches to STEM education that cultivate not only knowledge and skills, but also the attitudes and beliefs necessary for success in the known and unknown STEM-enriched careers of the future.

Wind energy is the leading renewable technology in the U.S., generating over 10% of utility-scale electricity in recent years. Rapid growth in wind energy installations has made modeling and prediction of atmospheric boundary layer (ABL) wind speeds and the associated turbulence critical for wind turbine siting, resource assessment, and operational power forecasting. A number of modeling challenges currently exist, such as representing the impact of terrain on wind turbine wakes and capturing small-scale turbulence in stably-stratified conditions. Many low-fidelity wind turbine simulation methods fail to incorporate topography and struggle to account for dynamic flow behavior. In this dissertation, results are presented using high-fidelity large-eddy simulation (LES), which captures the dynamic and turbulent behavior of ABL winds, providing a framework to simulate a wide variety of turbulent atmospheric phenomena with a wind turbine parameterization to understand turbine-airflow interactions.

First, high-resolution simulations of the 2017 Perdigão field campaign in Portugal are conducted. The Perdigão site consists of two parallel ridges with a wind turbine located on top of one of the ridges. Both convective and stable atmospheric conditions are simulated to understand how the wind turbine wake behaves in complex terrain in two representative flow regimes. For the convective case study, flow recirculation in the lee of the ridge occurred, thus deflecting the wake upwards. For the stable case study, the wake deflected downwards following the terrain due to a mountain wave that occurred. The vertical behavior of the wind turbine wake can be detrimental to downwind turbines; however, this vertical behavior is not accounted for in current wind farm design wake models. These case studies demonstrate the dependence of the wind turbine wake behavior on terrain-induced flow phenomena, which, in-turn, depend on the thermal stratification of the atmosphere.

The stable case study from Perdigão is then studied in more depth to better understand both the ambient and wind turbine wake turbulence characteristics. Novel derived measurements of the turbulence dissipation rate are available from the field campaign, providing an opportunity to further examine the spatial structure of turbulence predicted by the model. Additionally, in this study, the dynamic reconstruction model (DRM) LES turbulence closure is used to better represent smaller-scale turbulence. The DRM closure more accurately predicts turbulence metrics, including the turbulence dissipation rate, most notably upwind of the major topographic features. After the flow passes over the first ridge, the differences between the DRM and a standard eddy-viscosity closure are small close to the surface, although the DRM closure does better predict the turbulence dissipation rate in the upper atmosphere in this region. Because the DRM closure is not a standard eddy-viscosity closure, negative turbulence dissipation rate or the backscatter of energy from smaller scales to larger scales is predicted; however, backscatter cannot be derived from Perdigão measurements due to the experimental setup and analysis methods used, thus leaving validation of this aspect for future work.

Next, a range of idealized stable boundary layer (SBL) conditions are modeled in support of the American Wake Experiment (AWAKEN) field campaign to address: (1) the effect of wind turbines on SBL development, and (2) the effect of intermittent turbulence on wind farm performance. In weak SBL conditions, turbulence is continuous and easier to simulate. With the intermittent turbulence that occurs in strongly stable conditions, only the DRM closure can resolve realistic turbulence. For all SBL conditions simulated, the wind farm significantly impacts wind speeds and thermal structure well downwind (greater than 30 rotor diameters or 2.4 km) of the farm. Wind speeds in the wakes are reduced, and the increased mixing as a result of the wakes weakens the stable stratification in the boundary layer.

Finally, simulations are performed of a real case study of intermittent turbulence observed during the AWAKEN field campaign. The intermittent turbulence event is determined to be driven by a nocturnal mesoscale convective system (MCS). The MCS results in a cold pool, which radiates outwards as a density current. This density current perturbs the SBL, thus inducing gravity waves. The structure of the simulated gravity waves is found to be especially sensitive to the parameterization of cloud and precipitation processes (microphysics). The gravity waves have very strong effects on the flow in the upper atmosphere; however, closer to the surface where there is additional ambient turbulence and turbulence generated by wakes, the effect of the waves is more nuanced. Notably, the waves induce local wind direction variation, which leads to fluctuations in the power output as various turbines within the farm are subjected to the wakes of nearby turbines.

The findings presented in this dissertation provide insight into wind farm performance in a broad range of atmospheric conditions by incorporating both terrain effects and thermal stratification. Specifically, these conditions include dynamic turbulent phenomena that current wind farm design tools are unable to capture. The advances in this dissertation related to high-resolution LES reveal novel and complex relationships between wind turbines and the atmosphere that can significantly improve wind farm power predictions at large.

Slow electron velocity-map imaging (SEVI), a high-resolution variant of anion photoelectron spectroscopy, has proven to be a powerful and versatile spectroscopic technique capable of measuring the vibronic structure of a wide range of molecular and cluster species with exquisite detail. The extension of SEVI to study anions cooled to their ground vibronic state (cryo-SEVI) through the addition of a cryogenic ion trap dramatically improved spectral resolution and enabled larger species to be probed. This, however, also greatly limits the number of neutral states accessible via photodetachment. To overcome this, anions are resonantly excited to a selected vibrational state using infrared (IR) radiation prior to photodetachment. This new technique, dubbed IR cryo-SEVI, offers the potential to probe the vibrational structure of both anionic and neutral species to an even greater extent.The cryo-SEVI spectrum without vibrational pre-excitation was collected for the acetyl anion (CH3CO¯). This spectrum reveals a significant vibrational progression along the CCO bending mode with transitions up to Δv=11 clearly observed. The measured electron affinity for the acetyl radical is also used to calculate a refined value for the gas-phase acidity of acetaldehyde (CH3CHO). Cryo-SEVI with vibrational pre-excitation was first applied to the hydroxide anion (OH¯) by exciting the well-characterized R(0) rovibrational transition of the anion. The large rotational constant of the diatomic paired with the high resolution of cryo-SEVI enabled the rotational fine structure to be fully resolved. Upon excitation, depletion of the ground state features is observed along with new features that appear corresponding to transitions from vibrationally excited anions. Additionally, the IR absorption profile of the anion was measured by monitoring the growth of new features as the IR energy is varied, precluding the need for messenger-tagged species that perturb vibrational frequencies. IR cryo-SEVI was extended to polyatomic systems with the vinoxide anion (CH2CHO¯) excited along the CO (ν4) and lone CH (ν3) stretching modes. Excitation of the lower frequency ν4 fundamental results in photoelectron spectra that can be fully explained within the harmonic approximation. Excitation of the higher frequency ν3 fundamental, however, results in a more complicated spectrum consisting of several unexpected transitions. Theoretical considerations reveal the ν3 fundamental is anharmonically coupled to nearby vibrational states in both the anion and neutral manifold, leading to newly allowed transitions appearing in the spectrum. The nitrate radical (NO3) has been the focus of several theoretical and experimental investigations owing to its complex electronic structure arising from strong vibronic interactions between electronic states. IR cryo-SEVI is used to definitively settle a decades-long controversy over the fundamental frequency of the degenerate stretching mode (ν3), wherein the ν3 and 2ν3 vibrational states of the NO3¯ anion are accessed prior to detachment. Through comparison to theory, assignment of transitions from the 2ν3 state allow for unambiguous determination of the neutral ν3 frequency. Vinylidene (H2CC), a high-energy isomer of acetylene, represents a model system to study how isomerization affects the vibrational structure of molecular species. The small barrier for isomerization to acetylene (HCCH) as well as the asymmetric shape of the potential energy surface allows for low-lying vibrational states of vinylidene to interact with highly excited HCCH states, resulting in a complex vibrational structure. Cryo-SEVI spectra collected with and without vibrational pre-excitation probe vinylidene’s complicated structure with unprecedented detail, giving insight into which vibrational modes drive the isomerization reaction. The high-resolution of cryo-SEVI also reveals the nature of vinylidene-acetylene coupling in the energy region surrounding the predicted isomerization barrier height.

Chapter 1. The relevant background to the project is communicated in addition to the project hypothesis and strategy. Tetrylene-f-element bonded complexes are introduced as compounds of nearly unexplored chemical reactivity and bonding character. The strategy of combining tetrylenes with coordinatively unsaturated f-element precursors is briefly described.

Chapter 2. Novel uranium-tetrylene bonded complexes are synthesized by utilization of amidinate-supported silylenes. The solid- and solution-state structures of these compounds are examined by X-ray crystallography, absorption spectroscopy, nuclear magnetic resonance spectroscopy, and variable-temperature magnetometry. The nature of the uranium-silicon interactions is further elucidated by density functional theory methods.

Chapter 3. The reactivity of f-element-silylene complexes toward hydrogen gas is reported. Despite showing little evidence for bonding in solution, a uranium-silylene complex rapidly activates hydrogen to yield a dihydrosilane product. Lanthanide analogues to the uranium-silylene complex are much less efficient catalysts, while common main group Lewis acids show no catalytic activity. The mechanisms of both the actinide- and lanthanide-catalyzed reactions are deconvoluted through isotope labeling studies and kinetic and theoretical modeling. Investigation of the uranium-catalyzed pathway reveals that dihydrogen complexation by uranium is accessible and may underpin the particular efficiency of this catalyst.