Portuguese 
English 
Spanish 
4 min of reading 
0 comments 
Study with data from 10 New York stations shows that subway passengers reduce trips unevenly during extreme weather events, with heavy rains affecting evening peak more and intense cold mainly impacting off-peak travel
New York subway passengers reduce trips differently depending on the type of weather event, station, and time of day, according to a study published in the journal npj Sustainable Mobility and Transport. The research analyzed hourly records at 10 major stations between 2023 and 2025 and identified stronger effects of heavy rain during the evening peak, while extreme cold caused smaller, more concentrated drops outside of peak hours.
The analysis was conducted by researchers from NYU Tandon, the University of Louisville, and the University of Hong Kong. The goal was to observe how demand behaves collectively within the network, rather than just measuring the isolated reaction of each station to storms, cold snaps, or other severe conditions.
To do this, the team used a statistical technique called vine copula modeling. The method allowed them to examine how the number of passengers varies between different points of the system under extreme weather conditions, revealing response patterns that repeat according to the type of event.
-
Hidden cave under a castle in Wales reveals 120,000 years of history, with fossils of extinct animals, stone tools, and rare clues about ancient humans
-
Brazil launches an autonomous and electric tractor for coffee farming that works 9 hours without fuel or an operator, costs R$ 340 thousand, has 80 horsepower, uses GPS and Starlink, and can allow a single supervisor to control up to five machines on the farm.
-
Metsul warns of an intense extratropical cyclone in the South Atlantic with rain, winds above 100 km/h, and the advance of a cold air mass over Southern Brazil, dropping temperatures to as low as 5°C in Rio Grande do Sul and Santa Catarina.
-
“Starlink Farm” draws attention in Brazil: Elon Musk’s internet uses dozens of antennas in the Amazon, requires multiple subscriptions (1 account for up to 2 devices), ignores the minimum distance of 1 meter, and redistributes the signal via fiber.
Subway Passengers Reduce Trips More Significantly During Heavy Rains
Heavy precipitation appears as the factor with the greatest impact during the evening peak. Average reductions vary widely among stations, ranging from nearly 29% at Columbus Circle to less than 8% at Grand Central.
Flushing–Main Street, located in a peripheral neighborhood, also recorded a significant drop on days of heavy rain. The reduction reached almost 26%, reinforcing that the impact is not uniformly distributed across the network.
Researchers explain that evening commutes tend to be more flexible than morning ones. Many passengers can leave earlier, wait for the storm to subside, or cancel non-essential appointments, which amplifies the drop during this specific period.
Even so, rain does not completely eliminate travel. Most people can still get home, but the intensity of the storm changes timing, reduces optional trips, and alters station usage patterns.
Extreme Cold Less Affects Essential Routines
Intense cold showed a different behavior in the study. Even during the morning peak, when the effect appears strongest, the drop in passenger numbers was considered modest, generally between 1% and 2.4% at the analyzed stations.
This result shows that essential routines tend to be maintained even when temperatures drop sharply. Work-related travel and mandatory appointments continue to occur, even under uncomfortable conditions.
The greater effects of cold appear outside of peak hours. During this period, less necessary trips are more likely to be canceled, postponed, or replaced by another mode of transport.
Joseph Chow, associate professor at NYU Tandon Institute and one of the article’s authors, stated that passengers maintain their routines even with sharp drops in temperature. The change is clearer among those who would make optional trips, such as going to a restaurant or visiting a friend.
Nearby Stations React in Very Different Ways
The study also identified strong differences between relatively close stations. Columbus Circle appears as one of the most sensitive to heavy rains, while Grand Central, less than three kilometers away, experiences much smaller drops.
This variation indicates that neighborhood location alone does not explain each station’s resilience. Infrastructure, spatial design, connectivity, and surrounding land use appear to play an important role in how subway passengers respond to bad weather.
The comparison between stations shows that the network does not react as a single block. Each point can concentrate different vulnerabilities, even when subjected to the same type of weather event.
This behavior is important for public transport planning. A storm can strongly affect one station, cause a moderate drop in another, and barely alter a third, depending on the combination of local factors.
Model Helps View the Network as a System
Omar Wani, assistant professor at NYU Tandon and also an author of the article, stated that the method offers planners a way to observe the entire network responding to a storm or heatwave. The proposal is to go beyond the isolated reading of each station.
Modeling also allows generating plausible demand scenarios under extreme weather conditions. These scenarios can aid decisions on how to prepare the system for moments of greater instability.
The authors highlight, however, that the analysis has limitations. The study focused on 10 stations with high passenger demand, and extreme weather events are relatively rare within the evaluated data.
To address this limitation, the model generates probable patterns based on observed relationships between stations. Therefore, the results should be read as estimates of possible responses, and not as simple averages of past storms.
Differences can amplify overloads between neighborhoods
Even with the limitations, the research points to a clear pattern. Heavy rain impacts subway usage more during peak hours, while extreme cold weighs more outside these periods.
Differences between stations also appear as consistent data, not as random variations. This reinforces the need to look at each area of the network with its own specific attention.
The implications are not restricted to the daily operation of the system. Neighborhoods that rely more on public transport may face greater overloads when severe weather events reduce the number of trips or hinder access to stations.
With the increasing frequency of severe weather events, understanding when and where subway passengers stop traveling can help agencies plan more targeted responses. The research shows that rain, cold, and location affect the system in distinct ways, creating different challenges within the same network.
Be the first to react!