Station Inundation
Gauri Bahuguna, James Piacentini, Adam Susaneck, Alex Wong
Our project aims to demonstrate the vulnerability of the New York City Subway system through the simulation of future flood events and their impact on MTA infrastructure. This project seeks to achieve that aim by modeling the number of commuters affected by a potential future flood event of varying intensities.
After 2012’s Hurricane Sandy inundated several Subway stations and left lasting damage (still being felt to this day), it became painfully obvious that New York’s transit system was not prepared for the chaotic climatic future that awaits. Not only was the majority of lower Manhattan — the nation’s largest job center and one of the single most economically productive hubs on the planet — knocked out, but many of the aging tunnels connecting the island to Brooklyn were also flooded with corrosive seawater. This has resulted in countless headaches as repairs continue on the 120-year-old tunnels — most notably the Canarsie Tube carrying the L Train’s 225,000 daily commuters.
While the MTA has undertaken token efforts at building climate resiliency since Sandy, the system remains woefully unprepared. By overlaying the National Oceanographic and Atmospheric Administration’s flood prediction data over Manhattan geography we can see what stations are most at risk.
In our simulation when a station is flooded it becomes “disabled.” Commuters are randomly generated at stations and travel towards other stations randomly. If a commuter’s journey traverses a disabled station the commuter is terminated and counted towards the number of disrupted journeys.
By tallying up the number of disrupted commutes it is possible to gauge the magnitude of a future flood event with regards to transit infrastructure damage. This simulation allows us to predict the severity of the transit crises that await and hopefully encourage those in positions of power to do something about it.
Future iterations of this project will allow for many flood scenarios to be tested as well as taking into account the (scant) steps that the MTA has taken to harden particular stations against floodwater. In addition, commuter behavior will be adjusted to reflect the approximate ridership of each line and compare that to the line’s vulnerability (for instance, south of Chambers Street Station the 1 Train carries far fewer commuters than the 2). In the meantime, the code for our virtual commuters’ behavior can be seen below:
