New York City Transit Accessibility Analysis

In New York City, the subway system is crucial for millions, supporting daily commutes and influencing urban mobility. This project uses Geographic Information Systems (GIS) for a detailed analysis of NYC subway coverage, examining how well the system meets urban demographics and accessibility needs. The study highlights transit equity gaps and sustainable development opportunities, incorporating theories of accessibility and land-use interaction that link transit accessibility disparities to socioeconomic and demographic factors (Levinson & Krizek, 2018; Martens, 2016; El-Geneidy et al., 2016).

Previous research has documented the crucial role of GIS in identifying and addressing transit deserts—areas with significant transit demand but insufficient service. Studies have shown that these areas often coincide with neighborhoods housing marginalized populations, exacerbating social and economic inequities (Levinson et al., 2015). Furthermore, GIS-based studies contribute to a nuanced understanding of transit accessibility by mapping service distribution and integrating this data with demographic analytics, thus providing a robust framework for policy development aimed at enhancing transit equity (Geurs & Van Wee, 2004; Fan et al., 2016).

This project hypothesizes that the distribution of subway services in NYC is uneven and correlates with demographic and socioeconomic disparities. By employing GIS to map and analyze these patterns, the study seeks to offer actionable insights that can inform urban planning and contribute to the development of more equitable and efficient public transportation systems.

The primary goal of this project is to provide a comprehensive analysis of transit accessibility in New York City, using spatial and demographic data to illuminate areas of concern and opportunities for enhancement. The objectives are designed to inform policymakers and urban planners with actionable insights:

• Integrate Demographic Data: Analyze the populations served by transit to determine the distribution of transit demand.
• Map Transit Accessibility: Conduct detailed mapping of subway station service areas at different walking intervals to identify geographic disparities in access to public transit.
• Evaluate Spatial Distribution of Transit Access: Utilize spatial statistics to assess the distribution of travel times to the nearest subway stations, identifying anomalous areas where transit access deviates from citywide norms.
• Assess Commute Efficiency During Peak Hours: Assess commute efficiency during peak hours by examining workplace transit accessibility within the transit system, focusing on the reachability of key employment centers during morning rush hours.
• Identify Areas for Potential Service Improvement: Based on the analysis of transit accessibility and demographic needs, identify blocks and neighborhoods where subway access is notably deficient and provide insights for areas that could benefit from transit development initiatives.

Analyze the demographic characteristics of transit-dependent populations and assess the distribution of transit demand across New York City.

Incorporate relevant demographic data into the GIS dataset, enriching the analysis of transit-dependent populations.

• Data: 2020 Census New York City Blocks

• Key variables:
  • Total Population: Examine overall population metrics.
  • Income Below Poverty Level: Assess economic vulnerabilities affecting transit reliance.
  • Owner and Renter Households without Vehicles: Analyze vehicle access among homeowners and renters to gauge public transit dependency.

Combine demographic data layers to generate a transit demand distribution map and ensure consistency and repeatability in the analysis.

1. Duplicate Layers: Copy the original result layer into four separate layers for individual processing and symbology.
2. Population Estimation:
   1. Calculate total population for each block group, assuming uniform distribution.
   2. Address variable block sizes (264 feet by 900 feet), and estimate population per 237,600 square feet area.
3. Raster Conversion and Sizing
   1. Convert the each feature layers to a raster format, setting the block size.
   2. Set the output cell size for the raster based on the square root of the block area, which is approximately 487.44 feet.
4. Focal Statistics: Estimate average density within a defined radius around each raster cell, targeting an area encompassing three city blocks.
5. Combine Rasters: Perform raster calculations to merge all four layers into a single comprehensive density map, which represent the transit demand.
6. Automate with Model Builder: Construct and automate the entire workflow within Model Builder to ensure consistency and repeatability across analyses.

• Network Dataset Configuration:
  • Transit Stations and Lines Data: Process GTFS data to create spatial features.
  • NYC Streets Data: Include one-way or two-way roads to enhance traffic and accessibility modeling.
  • Evaluator: Walking time (min) along street segments.

Identify the geographic areas served by the city's transit system by tailoring the network for pedestrian movement analysis, using walking speeds to simulate realistic travel times within the city.

• Solver Configuration:
  • Set three travel time cutoffs, 5, 10 and 15 mins to identify areas accessible within specified walking time from subway stations
  • Use an 'overlap' configuration to highlight intersections of accessible areas, producing detailed maps of coverage overlaps.
  • Apply a 'dissolve' configuration to merge overlapping areas, creating unified maps and highlighting regions where the cutoff time exceeds 15 minutes.

Assess the proximity of the nearest subway stations at the block level by measuring the walking time required from the centroid of each block to the nearest subway station. Then, categorize blocks according to their access levels, using specific travel time thresholds—5, 10, 15 minutes, and over 15 minutes—to evaluate transit accessibility.

• Dual-Scope Evaluation:
  • Citywide Analysis: Conduct a broad analysis across all of New York City to identify overarching disparities in transit service coverage.
  • Borough-Specific Analysis: Focus on localized transit accessibility within individual boroughs.

• Inferential Statistics:
  • Spatial Autocorrelation (Moran's I and Incremental): Analyze the degree of geographical dependency among travel times for each block, and determine whether the pattern is clustered, dispersed, or random.
  • Hotspots Analysis (Getis-Ord GI*) and Cluster and Outlier Analysis (Anselin Local Moran's I): Identify areas of high and low value clusters (hot and cold spots) and detect local patterns that significantly differ from overall trends.

• GTFS to Public Transit Data Mode: Converts multiple GTFS public transit datasets to a set of feature classes and tables that represent the transit stops, lines, and schedules.

• Connect Public Transit Data Model to Streets: Connects transit stops to street features for use in a transit-enabled network dataset.

• Feature Class Integration:
  • Stop: Designated access points for passengers, defining entry and exit points within the transit network.
  • LineVariantElements: Details variations in transit routes, capturing differences in train schedules and destinations.
  • Streets: NYC Streets.
  • StopsOnStreets: Connects transit stops to their precise street locations.
  • StopConnectors: Pathways that facilitate pedestrian movement from streets to transit stops.

• Evaluator: Public Transit Time (min) = Travel time along a public transit line segment + Walking time along street segments (Walking speed: 308 ft/min)

Assess workplace transit accessibility by evaluating job connectivity, measuring the total number of jobs reachable within a 40-minute travel time by transit and walking during the morning rush hour.

• Data Configuration and Analysis Settings:
  • Origins and Destinations: All blocks in NYC.
  • Network Data Source: Utilizes Transit Network Dataset to access transit routes and schedules.
  • Travel Model: Implements the Public Transit Travel Model to simulate commuter conditions during peak hours.
  • Time Frame: Focus on the morning rush time window, 7:00 AM to 9:00 AM, on a typical weekday, with one-minute increments for high temporal resolution.
  • Cutoff Time: Set at 40 minutes, based on average commute times to the office for New Yorkers.
  • Destination Weighting: Based on the number of jobs in each block, referring to New York LEHD Origin-Destination Employment Statistics (LODES) data.

• Analysis Insights:
  • Identify areas with optimal job accessibility by combining demographic data and transit accessibility coverage, as well as pinpointing those areas that require improved transit services.

The determination of population reliance on public transit as a measure of transit demand could be refined by including a broader range of factors. Current assessments may not fully capture the complexity of transit dependency. Future work should integrate additional variables such as socio-economic status, availability of alternative transportation modes, and temporal variability in transit needs to provide a more comprehensive understanding of transit demand.

The transit accessibility analysis relies on a street network dataset to calculate travel times from each block to the nearest station. However, missing data for some blocks highlights imperfections in the street network, due to the solver's inability to find valid routes. This lack of complete path data can skew accessibility measurements and does not accurately represent areas without direct paths to transit. Future work should focus on enhancing the quality and completeness of street network data to ensure more complete and accurate travel time estimations.

The current transit network dataset used to establish transit travel time costs does not account for several significant factors that affect total transit time. For instance, it omits the waiting times for transit and the time spent transferring between lines within stations. Future enhancements should include these time components to provide a more accurate representation of accessibility and better reflect the real-world experience of transit users.

Our Workplace Transit Accessibility study excludes commuters who live in New York City but work outside it, as well as those who live outside New York City but work within it. Consequently, the findings may not fully represent the transit accessibility issues faced by all workers connected to New York City.

Previous studies on New York City's transit accessibility have typically focused on broad assessments of transit coverage or isolated analyses of demand within specific neighborhoods. However, this project extends beyond these studies by systematically mapping transit demand gradients, accessibility in terms of travel time to subway stations, and job accessibility across all boroughs. Unlike earlier research that might use static data analysis tools like Excel or basic programming in Python, this GIS analysis integrates spatial data with demographic and transit service information, providing a more comprehensive overview of the transit landscape.

GIS technology allows for a dynamic visualization of data layers, which is crucial for understanding the geographic spread and interaction of variables such as population density, transit facility locations, and employment hubs. This spatial integration is something tools like Excel and non-spatial Python scripts cannot achieve, as they lack the capacity to effectively render and analyze geographic and spatial relationships.

The analysis reveals a gradient of transit demand, decreasing from high in city centers to lower in outer areas. Central Manhattan and Brooklyn consistently show the highest demand, with sporadic pockets of high demand also emerging in more remote areas.

The accessibility measured by travel time to the nearest subway stations indicates that while the New York City transit system broadly covers areas from urban cores to the outskirts, significant service gaps still exist. And, although the distribution of access times is relatively uniform across the city, narrowing the focus to each borough reveals outliers where certain blocks experience notably longer access times compared to their neighboring areas.

Job accessibility is highly concentrated in Manhattan and central Brooklyn and decreases outward. While most regions maintain a balance between population density and accessible job opportunities, the outer areas of the city show substantial gaps in transit access. These areas are marked for needed enhancements in transit services to improve job connectivity.

Our analysis highlights areas with high demand, poor job accessibility, and overall poor access, emphasizing the need for targeted interventions like extending subway lines, adding new stations, or improving other public transportation options to serve the city's diverse populations more effectively. By visually representing disparities in job accessibility and the locations of transit deserts, particularly in outer boroughs, this analysis provide urban planners and policymakers with actionable insights that can directly shape transit policies and infrastructure development.