Report: Evaluating Urban Mobility with Sustainability

New York City is actively implementing sustainable transportation policies to address air pollution. A key initiative is the  transition to electric and hybrid vehicles , evident in sectors such as taxis, school buses, and patrol vehicles. Particularly, the expansion of electric vehicle charging infrastructure supports this transition and is a crucial strategy for the city's response to climate change. The city’s approach includes incentives for EV adoption, investments in public transportation, and the integration of green technologies, all part of the broader climate action plan to achieve carbon neutrality by 2050.

In line with this, our project aims to assess the usage of various sustainable transportation modes such as electric vehicles, public transit, and New York’s bike-sharing system, Citi Bike, within New York City, and analyze how they support the community using ArcGIS pro. Through this analysis, we will explore which areas are utilizing sustainable transportation least effectively and how the accessibility of these modes correlates with race, income levels, or population density.

The goal of this project is to understand how New York City's diverse transportation policies and specific implementation plans interact with environmental and economic factors in different areas. Our analysis will provide a clear overview of the status of sustainable transportation infrastructure and offer data that can inform the development of effective future policies.

Recent research highlights significant disparities in the distribution of electric vehicle (EV) charging infrastructure, underscoring socio-economic and racial inequities in urban settings. Hsu and Fingerman (2020) observe access to public EV chargers is consistently lower in Black and Hispanic majority areas compared to white-majority areas, accounting for factors including proximity to highways and median household income. Khan et al. (2021), similarly, analyze New York City’s EV charging stations, finding an inequitable distribution that disadvantages low-income, predominantly Black, and disinvested neighborhoods. These two studies mentioned the importance for policy interventions that help build equity in the expansion of EV charging infrastructure.  

Time & Date : 4/29/2024 9AM

To assume the most extreme scenario, we set the time and date in the route solver to Monday at 9 AM, the busiest time. This will also be applied to the service area analysis.

PLACE 1 From : Dyker Beach Park To : Marcus Garvey Park

PLACE 2 From : Ocean Breeze Park To : Queens Farm Park

PLACE 3 From : John F. Kennedy International Airport To : Haffen Park

To conduct a detailed network analysis, Travel Mode 4 will be utilized, as it has demonstrated the closest alignment with the travel times provided by Google Maps across various routes.

This is the final map for our network analysis part, yet there are numerous variables that still require consideration to deepen our understanding. By incorporating demographic factors such as race and income into our analysis, we can assess the accessibility of the charging infrastructure more comprehensively. This approach allows us to view the infrastructure's impact on different segments of society and understand disparities in access. The forthcoming analysis will explore these socio-economic aspects in greater detail, aiming to provide a holistic picture of how the charging network serves the community.

After identifying areas where EVs take 8 or more minutes to find a charging station, which we define as underserved areas, we decided to focus on the distribution of sustainable transportation within these underserved areas using hot spot analysis. We also conducted a hot spot analysis for the whole NYC area to see the public transportation accessibility and the socio-demographic characteristics.

The two main parts to our hot spot analysis: one for the underserved areas and another for the entire NYC area, are both at the census tract level.

There are some areas that require extra attention. For example, in the right corner, the region is identified as cold spots at the whole New York City census tract level (left), but it is regarded as hot spots within the underserved area (right). This discrepancy is due to the different scales used for hot spot analysis. When the right corner is compared with the entire NYC, it appears as cold spots relative to areas like Manhattan. However, it is considered a hot spot when compared to other underserved areas, which is logical. This also aligns with our focus on the cold spots of underserved EV charging station areas. These steps ensure a thorough analysis of the underserved EV charging station areas, allowing us to examine the socio-demographic characteristics associated with the identified cold spots.

The two pie charts illustrate the racial composition within areas served and underserved by EV charging stations, highlighting distinct racial patterns. There are some notable observations: in the served areas, the percentage of White individuals is 39.02%, which decreases to 34.77% in underserved areas. Conversely, for Black or African American individuals, the percentage increases from 22.27% in served areas to 30.96% in underserved areas, suggesting a racial disparity in the distribution of EV charging stations. Note, all races analyzed here represent individual races only; the composition of mixed races is not considered in this study.

The median household income is highest in the underserved area recognized as cold spots, and it is lowest in the broader underserved areas. The population density per square mile is lowest in cold spots within underserved areas, yet it is highest across the entire New York City.

This data is unexpected, as typically, cold spots within underserved areas are considered with low median household income and high population density. However, the unexpected result arises because most of the cold spots are located in Staten Island, which has a relatively high median income and low population density. Then the large proportion of cold spots in Staten Island skews the overall data for cold spots.

1. High Accessibility in Manhattan: Most of Manhattan is well-served, with EV charging stations accessible within 4 minutes and high accessibility to public transportation.
2. Intermediate Accessibility: Manhattan and its surrounding areas, including Long Island City and most of Queens, are considered to have intermediate accessibility to EV charging stations.
3. Underserved Areas in Staten Island: Most of Staten Island is identified as underserved concerning EV charging stations.
4. Data Skew in Staten Island: Most cold spots within underserved areas are located in Staten Island, skewing the data on median household income and population density due to its high median income and low population density, which is atypical for underserved areas.
5. Racial Disparity: There is a noticeable racial disparity in access to EV charging stations in these underserved areas, underscoring the need for focused interventions to address these inequalities.

1. The accuracy issues arising from not using partial census tracts

After identifying underserved areas within New York City, we utilized the 'Select By Location' tool to determine which census tracts intersect with these areas. By this, we mean that if any part of a census tract falls within an underserved area, the entire tract is classified as underserved. We chose not to utilize partial census tracts, since if the partial area is very small, normalizing counts by such small areas can lead to extreme values that skew the input data for hotspot analysis. Currently, using complete census tracts is a remedial approach; however, it's not entirely accurate because parts of a census tract might be well-served. This means there is a risk of misidentification due to partial areas being underserved.

2. Race Composition within Areas Served and Underserved by EV Charging Stations:

This study focuses exclusively on individuals identified with a single race and does not take into account individuals identified with multiple races. This approach may limit the comprehensiveness of the study in understanding detailed racial patterns fully. In future research, it will be necessary to include mixed-race individuals in the analysis to better support demographics facing disparities in EV charging access.

1. Analysis of Electric Vehicle Registration and Charging Station Distribution through Rasterization

We aimed to determine the regional distribution of EV registrations and, by dividing this by the number of EV stations, calculate the number of vehicles that each station can cover per region. This would allow us to assess the adequacy of EV station placement relative to EV registrations and assign a corresponding score. However, the EV registration data was only available at the zip code level, whereas our primary geographic unit for other datasets was the census tract. Due to the lack of a conversion method to bridge these two geographic units, we were unable to perform this analysis. Nevertheless, we discovered that rasterization and subsequent interpolation could facilitate calculations by using pixels within the same rasterized area. This method can address the previously mentioned accuracy issues arising from not using partial census tracts. Consequently, we propose rasterization as a future work to achieve more accurate and detailed results.

2.Additional analysis using trip volume

Additionally, the area around JFK, which is one of the places with high foot traffic, also has many uncovered parts. Therefore, if buses and taxis in this area transition to EVs, the installation of charging stations should be considered a priority.

By counting the destinations of transportation modes, as shown on the left map, and using the most recent version of traffic volume data, we can more precisely identify which areas will need EV chargers. This analysis can be utilized along with underserved areas by using a spatial join.

This report presents a network analysis and hotspot analysis of EV facility accessibility in accordance with the implementation of major policies in New York City, along with an integrated analysis of other variables. This analysis identifies the initial inconveniences experienced by citizens and areas still lacking infrastructure in the bustling and crowded city of New York following policy changes. This was particularly noted between EV charging stations and other sustainable transportation modes, especially in Staten Island and the outskirts of Queens (Jamaica). These areas were identified as having poor accessibility to both EV stations and other transportation options, indicating a need for improvement.

This map ultimately illustrates regions deficient in sustainable infrastructure.We combined the results of the network analysis and hotspot analysis to analyze the underserved areas identified as cold spots using the Select by Location tool. Since there were no cold spots in Manhattan, we focused on the remaining four boroughs.

The analysis revealed that Staten Island accounts for 43% of the total area classified as Underprivileged Areas, making it the largest contributor, followed by Queens with 37%. Additionally, Queens has the highest population within the Underprivileged Areas at 39%, with Staten Island following. This confirms that Staten Island, despite its larger area, has a lower population density. These regional characteristics highlight distinct reasons for infrastructure requirements in each area.

Staten Island requires fundamental infrastructure to cover its extensive area, while Queens, with its high population density relative to its area, needs sustainable transportation infrastructure that can address both spatial coverage and population demands.The clear takeaway is that prioritizing infrastructure improvements in both Staten Island and Queens is crucial to establishing a more efficient and sustainable transportation environment.

Furthermore, by analyzing the correlation between eco-friendly transportation and socio-demographic characteristics in cold spots within underserved area, underserved area, and whole New York City, insights in finding additional EV charging locations can be gained. Through these GIS analyses and case studies, the expansion of eco-friendly transportation can be effectively supported.

 Borough boundaries 

 Bus stops 

 Census Tract shapefile 

 Citi Bike docks 

 Demographics by Census Tract 

 EV charging station 

 EV Registration  

 Median income by Census Tract 

 Road with speed limit 

 Subway station 

 TLC Electrification Report 

 Inequitable Access to EV Charging Infrastructure 

 Median Household Income for Each Borough in NYC 

 Median Income In NYC 

 Population Density for Each Borough in NYC 

 Public Electric Vehicle Charger Access Disparities Across Race and Income in California 

 Road Feature Class and Type 

 Speed Limit in NYC  

 Transitioning to Electric Vehicles 

 When Does the Electric Vehicle Low Battery Warning Come On?