Unlocking the IBX’s Transit Potential

The proposed Interborough Express is an opportunity to improve Brooklyn and Queens residents’ mobility and connectivity, the city’s most populous and largest boroughs, but also to transform the existing 14-mile corridor into a vibrant, livable and walkable urban landscape. The best way to maximize its potential is to implement comprehensive improvements that complement the existing neighborhoods, including upzoning, densification, mixed-use development, improved bus service and street redesigns. Such a holistic intervention would ideally begin before construction starts. These improvements will attract more residents and businesses to the corridor; thus, increasing ridership beyond current projections once it opens, saving thousands of travel minutes for local residents, reducing car-dependency, improving health and air quality as well as creating opportunities for new housing construction.

While the construction of the IBX can deliver all of these benefits, the MTA’s success will ultimately be measured by the access it creates and ridership it generates. But how can the city support IBX so that New Yorkers adopt it into their daily routines for traveling to work, school and other destinations? Transit ridership is correlated with population density, availability of jobs, mixed land use, existing transit service, and walkability conditions around stations. The busiest MTA subway stations are in neighborhoods with high residential and/or commercial densities, serve as important job destinations, connect to various modes of transit, and are walkable environments replete with amenities. We have quantified these characteristics around the proposed IBX stations, in an attempt to understand how supportive the current conditions are for generating ridership and how they can be improved.

We developed 10-minute isochrones, representing a 10-minute walk, around each of the 19 proposed IBX stations in order to collect data on residential and job density, land use, transit-service level and walkability. We also created isochrones and collected the same data for all of the 468 existing subway station complexes^[1] in New York City.

We refer to Transit Oriented Development (TOD) research to evaluate how our quantitative analysis results compare to recommendations from city-agency reports and academic literature. TOD studies have found that certain levels of residential and employment densities, mixed land use, transit service quality and walkability is essential to attract sufficient ridership to justify the large capital expense. IBX current cost estimate is north of $5 billion. Cost estimates early in a transit project, prior to any significant design and engineering work have been completed, are unreliable; however, they are useful as we consider benefits versus costs.

Our maps reveal how areas within a walkable distance from the proposed stations compare to others around the city and which areas have the potential to be developed and improved to cultivate transit-friendly conditions. Below we present our analyses of residential density, job density and mixed land use, transit service level and walkability conditions along the corridor. We wrap up by offering strategic recommendations to the city on how to maximize the value of the investment in building the IBX.

Population and Built Area Density

The IBX corridor has the potential to establish a robust rider-base, especially if the low-density neighborhoods along it are upzoned to allow for more residential development. The existing research on transit-oriented development (TOD) suggests that high-residential densities support high transit usage. Queens and Brooklyn have much lower population densities than Manhattan.^[2] Along the IBX corridor, however, census tracts near the proposed Roosevelt Avenue, East 16th Street, 8th Avenue and 4th Avenue Stations have population densities comparable to bustling Manhattan neighborhoods like the Columbus Circle, Greenwich Village, Chelsea or Alphabet City.

Population density around the IBX stations range between 11,800 and 132,800 residents per square mile with a median of 47,657.^[3] This is almost double the density that has been found to ensure a 40% mode share for public transit in US transit districts (Renne and Ewing, 2013). Renne and Ewing (2013) found that districts with 10,000-25,000 residents per square mile saw transit ridership drop to 14%; and those with 4,000 and 10,000 residents per square mile, to 5%. The developable land within each IBX isochrone has an average population density above this 25,000 threshold. However, upon closer study, there are at least 20 census tracts in the corridor with residential densities below 25,000, with a land area adding up to 2.5 square miles- more than double the size of the Central Park- presenting an invaluable opportunity for additional development.

Population density map highlighting density below 25,000

Along the IBX corridor, residential unit density ranges from 3 to 72 units/acre , with a median density of 26 units/acre. Academic researchers and city agencies have adopted between 15-25 housing units per acre to be a target residential density in an urban neighborhood to support transit use^[4] (SWEEP 2024; VISION 2040, 2022; Renne and Ewing, 2013; MARTA p44, 2010; Cervero, 2007). IBX will likely attract substantial ridership from existing residents along the corridor. However, 69 census tracts out of 2199 that intersect with the IBX isochrones remain below a 20 housing units per acre threshold, concentrating around Grand, Eliot, Metropolitan and Utica Avenue stations. A closer inspection of these areas reveal many lots that are underutilized, which could easily generate multiple housing units with the construction of even low to mid-rise residential buildings.

Residential units density map, highlighting density below 15 units/acre

Jobs and Mixed Land Use

With the exception of a few census tracts near the New Utrecht, Avenue I and Flatbush Avenue Stations, the job density along IBX is insufficient to support transit ridership. Only a quarter of the census tracts within the station isochrones have job densities higher than 20 jobs/acre, and only 15 have more jobs than 50 per acre; 20-50 being the range within which travel mode share shifts from single-occupancy vehicle use to transit use and walking (Frank and Pivo 2014). Precisely speaking the IBX corridor would need to add 222,485 new jobs to increase its overall job density to 35 jobs per acre, or the median of the range identified in the literature.

Residential density below 15 units/acre and employment density below 20 jobs/acre

The best way to introduce more housing and jobs along the IBX corridor would be to cultivate mixed-use. Mixed-use buildings sustain robust ridership throughout the day by attracting riders with different purposes and schedules, while also improving walkability by encouraging short, within-neighborhood trips (Ewing et al., 1994; Frank and Pivo, 1994). The dominant land use along the corridor is residential, with mostly single to two family, or multi-family walk-up buildings. These present an opportunity to densify the area by upzoning to allow 4-to-5-story^[5] mixed-use buildings, as well as by building larger-footprint buildings through land consolidation. We also found that the large, low-density and predominantly industrial census tracts with single or double-story warehouses often accompanied by large parking lots don’t actually provide significant job densities. Several such lots, highlighted in blue in our map, can be built up to accommodate many times as many square feet of commercial or mixed uses. We show the land use along the corridor and focusing on some of these low-density lots below.

Predominant land-use per census tract map (source: MapPluto 2023)

One & two family lots near Flatbush-Nostrand Station

Low-density Industrial uses near Remsen and Linden Avenues

Low-density residential, auto-repair and other industrial uses near Utica Avenue Station

Looking at the combined density of population and employment reveals another critical aspect of the IBX’s right of way: it acts as a demarcation line  between the high-density, mixed use areas to its north-west and the predominantly low-density residential neighborhoods to its south-east. This can partly be attributed to the right of way itself, acting as a physical barrier between its two sides. Some of this barrier effect can be minimized through design interventions, which should also be used to address walkability issues along the corridor to improve conditions around the stations. But another factor exacerbating the problem is that both subway and bus service declines on the south east of the line. The combination of dwindling transit and increased distance from Manhattan significantly increases travel time and accessibility to the largest job centers in Manhattan.

Residential and employment density

Transit Service

New transit lines benefit residents living within walking distance to their stations and also connect trips for passengers who can conveniently access them through other means of transit. In a densely populated urban context like New York City, renowned for its comprehensive public transit network, the subway and bus services have the potential to significantly expand the accessibility of a new line, extending its influence well beyond the walking distance of the stations. However, the geographic extent and frequency of service of the connecting transit lines, in other words, the convenience of access to the stations by public transit, has an impact on whether people will opt to make the extra trip. This is why TOD studies incorporate measures like public transport accessibility level (Kamurzzaman et al., 2014), node index (Bertolini, 1999; Reusser et al., 2008) or reachability (Zemp et al., 2011), which quantify the convenience of transit-service based on indicators like the diversity of transit options, frequency and reliability of service, number of directions served and number of stations within a given travel distance.

How much will the current transit network support the IBX by providing convenient connections to its stations? The IBX has already been planned to connect with 16 subway lines as well as the Long Island Rail Road. To determine the combined effect of the bus and subway lines along the corridor, we calculated the total number of weekday trips and unique routes that serve each station area based on the GTFS schedule data. As a reference for comparison, we visualized this information for every subway station complex in New York City.

Number of bus and subway trips that serve the 10-minute catchment area of subway stations

Among the newly proposed IBX stations, Flatbush Avenue (isochrone) receives 6,481 weekday bus trips, the highest among the IBX stations and Eliot Avenue receives 615 bus trips, the lowest among them all. Looking at subway service, Roosevelt Avenue (isochrone) receives 1,682 trips per weekday, the highest number of subway trips in the corridor, whereas Grand, Eliot, Utica, Remsen Avenues and the Brooklyn Army Terminal (isochrones) are not served by any subway lines at all. For comparison, the station isochrone served by the highest number of subway trips in all of New York is 42nd street – Bryant Park^[6] receiving 6,498 trips while Parsons Blvd Station in Jamaica Center receives 18,854, the highest number of bus trips per weekday.

Top and bottom ranking IBX isochrones for bus and subway service

The IBX will bring in 330 new weekday trips to all the station isochrones^[7], increasing Roosevelt Avenue’s rail trips by 20% and introducing 330 new trips to the stations that currently have no rapid rail service. This should boost ridership for all connecting lines to the IBX. But also, increasing bus service to the low-service stations like Grand, Eliot, Utica, Remsen Avenues and the Brooklyn Army Terminal stations would help support IBX ridership.

Walkability

IBX’s right of way cuts through several city blocks with dead ends where it meets most residential streets, and with trenches or overpasses at wider streets. The designs of the new station connections with the street network should reverse this interruption around 19 locations along the 14 mile corridor. Increased foot traffic will also enliven these areas and attract some commercial development near stations. However, without a well-planned set of urban design interventions to accompany densification and mixed-use development, many walking routes to stations will go through wide multi-lane traffic arteries, large parking lots surrounded by fences, blind facades enclosing industrial facilities, and frustrating detours in order to avoid dead ends, hitting but not providing access to the right-of-way.

IBX’s right of way cuts through city blocks with trenches or overpasses at wide streets.

Large parking lots, industrial facilities and dead ends where the streets meet the IBX

Walkability, both a part of TOD indicators and also studied independently, shares its primary indicators with TOD including density, mixed-use and transit-accessibility. Since the quantitative measures of TOD and walkability overlap , we did not conduct a walkability analysis of the IBX at the census-tract or block-group level. However, we did utilize Space Syntax, a street-network analysis method, and Convex and Solid-Voids, a 3d, street-level analysis method to evaluate walkability.

We plan to publish the details of these two analyses in separate posts. But our findings indicate that there are pockets of walkable neighborhoods around the IBX with well-enclosed, human-scaled, dense and mixed-use streets^[8] that are disconnected from each other. On the other hand, several large motor-ways cut through the corridor without bringing in the density that could enhance the enclosure and add mixed-use to activate these streets. With densification, introduction of mixed-uses and street redesigns, the walkability of the corridor should be improved to support the accessibility of the newly proposed transit stops and enhance ridership.

Below is an example from the Convex and Solid-Void measures: Height/Width of streets, which indicates the level of enclosure provided by buildings, walls, fences and other vertical elements bounding the street spaces. Where streets are wide, and surrounding buildings are low, as in the case of wide boulevards or highways without much density around them, this value drops. The pockets of more walkable neighborhoods with narrower streets and multiple-story buildings around them are seen in darker colors near Roosevelt Boulevard, Myrtle, Wilson, 8th and 4th Avenue Stations. Where we see low enclosure levels due to the widths of streets or low-rise development around them, lane-reductions, and addition of vertically bounding vegetation elements like tall trees with leafy canopies can be recommended to accompany densification strategies.

Street-Voids height/width measures

Construction of the IBX, if supported by land use changes, street-redesigns and strategies to attract residential and commercial development along the corridor, can connect 600,000 existing residents living within a 10-minute walking distance from its stations, and, hopefully, thousands of more new residents to job centers in Manhattan, Brooklyn and Queens. The City should also view building the IBX as an opportunity to encourage the construction of much needed housing, in a location already well connected to the city’s transit network and to provide access to many of Brooklyn and Queens’ underserved neighborhoods.

Conclusion

IBX is a promising project that will attract a strong rider base from Broolyn and Queens. However, even though its population density can be considered sufficient when compared to residential TODs in the literature, many of its stations, especially, Grand, Eliot, Metropolitan, Linden, Remsen and Utica Avenue Stations have low population densities and provide an opportunity to encourage development to match the multi-billion dollar investment that is being made. Development that will bring new employment opportunities to the corridor should also be encouraged. This is the perfect moment to bring mixed-use development to the corridor, which should ideally be initiated without waiting for the IBX to break ground, as a partial solution to the city’s  housing crisis.

The building of the IBX is especially critical for the many transit-deprived areas along the corridor; we found that out of the 19 proposed stations’ isochrones, 15 received fewer bus trips than the median number of bus trips received by subway station isochrones, and 13 received fewer subway trips than the median number of subway trips received by subway station isochrones in the city. While IBX will bring 330 more transit trips to each isochrone, bus service should be improved around the stations to boost ridership.

Finally, walkability conditions should be considered along the corridor, so that small pockets of pedestrian-friendly neighborhoods are connected, and the reach of the stations are extended beyond their 10-minute isochrones through an attractive, safe and accessible street network.

[1] Transfer stations are considered single station complexes. I.e. Times Sq-42 St/Port Authority Bus Terminal, Lexington Avenue-59th Street, Herald Square-34th Street etc.

[2] Manhattan has about 72k residents per square mile whereas Brooklyn and Queens have 39k and 22k respectively.

[3] At the census tract level, ignoring the census tracts that significantly overlap with the existing right of way and others that are not developable.

[4] About 12,500-13,000 units per square mile.

[5] Department of City Planning’s City of Yes Housing plan includes a “Transit Oriented Development” proposal where up to 5 story apartment buildings can be built in R1-R5 zoned if the lots are within Greater Transit Zones, are large enough and on the short sides of blocks.

[6] This isochrone covers an area including Times Square, the Grand Central, the 42nd Street Port Authority, Herald Square 34th Street and 47th-50th Streets Rockefeller Center Stations.

[7] This is assuming  24 hr service with 5 min headway during morning (6-9am) and afternoon (4-8pm) peaks, 10 min headway during the day, and 20 min headway between 11pm and 6am.

[8] See Ewing and Handy, 2009 https://www.tandfonline.com/doi/abs/10.1080/13574800802451155