Editor’s Note: In March 2021, Sidewalk Labs retired the NYC Transit Explorer described in this blog post.
There’s no greater challenge facing New York City transportation than the L train East River tunnel shutdown that could start as early as 2019 and will last 18 months. The dreaded L-mageddon will upend commute patterns for many of the 250,000 people who travel that line between Brooklyn and Manhattan every day. Even residents who don’t use the L will feel the pinch, as extra travelers squeeze onto subway cars, buses, and roads across the city’s already-packed transportation system.
Recently the local advocacy group Transportation Alternatives reached out to see if Sidewalk Labs could demonstrate what impact the L train closure will have on daily commutes and general accessibility for people and businesses along the line. Today, on behalf of Sidewalk’s engineering team, I’m happy to release a new tool that helps do just that — and that visualizes accessibility via transit across NYC more broadly.
The tool is a map demo we’re calling the NYC Transit Explorer. It’s built on top of a transit router that uses the latest GTFS feeds for all MTA bus and subway lines, as well as the Staten Island ferry, to determine travel times and routes throughout the five boroughs. Traditional navigation routers direct people from Point A to Point B (call it one-to-one); our router adds the ability to show people how long it would take to get anywhere in the city via transit from a given origin (one-to-many).
Users can explore accessibility from a given origin based on a range of variables, including departure time, max walking distance, and transfer and mode preference. You can also compare accessibility from two different points (especially useful if you’re planning to move and deciding between two neighborhoods), or from the same point with different settings. Inspired by TransAlt’s request, you can see what happens in a world without L train stops in Manhattan. You can also see how access improved with the opening of the Second Avenue subway.
Let’s take the L train tunnel closure as an example.
First, have a look at what the shutdown means for general accessibility. We’ll start from Canarsie (the pin below) and use the default settings for departure time (8 a.m.), max walking distance (half a mile), transfers (included), and mode preference (none). The only difference between these two worlds is that one includes the L train stops in Manhattan, and the other doesn’t.
The impact is pretty striking. Suddenly it takes much longer for people in Canarsie to reach most of Manhattan, as well as the Bronx (areas shown in blue, above). To illustrate the difference further, I’ll use a potential commute for a nurse living in Canarsie to a hospital on the Upper East Side.
With the L train in full operation, the trip takes 72 minutes, transferring from the L at Union Square to the uptown 6 train. That’s already a haul, but without the L, that commute becomes much tougher: 94 minutes riding the B82 bus away from Manhattan before taking the Q train up the Second Avenue subway line. Keep in mind that 94-minute trip is probably a best-case scenario. This alternate route is based on normal subway conditions, but after the L shutdown, crowding on the rest of the system could lead to additional delays unless service increases to match demand. The actual new commute trip might take even longer.
That worries Transportation Alternatives. When we met to discuss the accessibility demo, Thomas DeVito, Director of Organizing for TransAlt, said such a change in commute time could nudge someone off the transit network and into a car, leading to more road congestion. Then again, not making the change means the nurse in my example would spend at least 44 more minutes a day commuting. In the end, everyone is worse off.
In the hopes of improving transit options during the shutdown, TransAlt has been exploring a workaround called the 14th Street PeopleWay — a corridor of rapid buses and bike lanes. Our router currently doesn’t include East River ferry service (it doesn’t seem to publish real-time feeds) or Citi Bike stations, though these modes could also help reduce the strain on the system. The MTA and NYC Department of Transportation are currently holding a series of public workshops to discuss alternative service plans.
Use cases and caveats
The L train tunnel shutdown is just one scenario to explore. The map also demonstrates the vital role that bus service plays in the city (especially in a place like Red Hook), and it reveals the challenges facing outer-borough transit use.
Going from Ridgewood to parts of Astoria by subway, for instance, takes 68 minutes and actually involves going into midtown Manhattan to make a transfer. (The bus-only route shown below takes 110 minutes, with two transfers and a longer walk.) By car the trip can be made in half an hour. These insights could help identify parts of the city where strong coordination between ride-hail services and public transportation might enhance accessibility to jobs, friends, and other destinations.
Even in a city with 24-hour transit service, time of day is another variable that makes a big difference for riders. Let’s say you’re a waitress living in Bushwick and working at a restaurant in the Upper West Side. Service changes and reduced frequencies make that trip significantly longer at 2 a.m. than during morning rush — 45 minutes longer in the example case shown below.
The demo is a work in progress, and we’d love to hear from developers and civic hackers who are interested in collaborating to help us improve it. As a cyclist, I’m hopeful the tool eventually integrates Citi Bike data into the mix. You could also potentially create speculative service changes along certain corridors to test their impact on accessibility, such as upgrading a regular bus route into Select Bus Service.
The NYC Transit Explorer has some limitations that bear mention. The biggest is that its underlying router measures walking distance as the crow flies instead of along actual blocks. We did this so the demo runs quicker, but using actual pedestrian routes would improve the router’s accuracy. The routes reflect a weekday schedule, so they can’t visualize the impact of weekend service changes. Whenever you’re using GTFS feeds, it’s important to remember that sometimes buses and trains don’t run on schedule for a thousand little reasons. And of course, right now the tool is only useful if you live in New York City.
So test it out! Explore! And let us know what you think. The more insight we gain into the challenges of access, the more options local agencies and companies can provide to help city residents overcome them.