Landfalling major hurricanes in the northeast U.S. are a rare breed, with only five having been observed since settlers arrived in the early 1600s — an average of just over one such storm per century. And despite warmer sea surface temperatures in the past 30 years, the northeast U.S. has avoided a major hurricane landfall since Carol in 1954. Yet research suggests these storms may become more likely as the climate continues to change.
Since accurate records began in 1851, only two other major hurricanes made landfall in the Northeast U.S.: the 1938 Long Island Express Hurricane and an unnamed 1869 storm that hit Rhode Island (though there’s reason to be skeptical of the second one; in an email, historical hurricane expert Cary Mock said, “I doubt it would intensify that far north as a Cat 3.”)
What about before accurate record-keeping began? Severe hurricanes that hit Long Island and then Connecticut in 1635 and 1815 were likely major hurricanes. A 2014 paper, Estimation of Hurricane Wind Speed Probabilities: Application to New York City and Other Coastal Locations, estimated that major hurricane wind speeds should occur about once every 300 years on Long Island and once every 700 years in New York City.
Note that by ”major hurricane,” we mean one with sustained winds of at least 111 mph, making it a Category 3 on the Saffir-Simpson Scale. A large, west-angling hurricane like 2012’s Sandy – discussed below – can deliver a catastrophic storm surge to the Northeast even if its winds decrease below the major-hurricane threshold just before landfall.
As the climate warms, at least four factors might increase the risk of a major hurricane in the northeast U.S.:
- Ocean warming is causing the strongest hurricanes to get stronger.
- A poleward shift in which hurricanes reach their peak intensity farther north (primarily because of warming sea surface temperatures)
- More hurricanes are forming close to the U.S. coast as a result of ocean warming.
- More Greenland blocking (a “stuck” jet stream pattern that increases the risk of a Northeast U.S. landfall)
One factor that might decrease risk: a tendency toward slower-moving storms, which would result in more weakening as they cross the cold waters offshore of the Northeast.
My hunch: these competing factors may act to modestly increase the major hurricane risk to the Northeast, so that a one-in-100-year storm of the 20th century would be roughly a one-in-75-year storm during the 2026-2050 period.
Global warming expected to make the strongest hurricanes stronger
As discussed in detail in our previous post, Global warming is making the strongest hurricanes stronger, one of the more confident predictions on how climate change will affect hurricanes is that the strongest hurricanes will get stronger. As a result, the odds of a major hurricane hitting the Northeast will increase. A 2022 study, A globally consistent local-scale assessment of future tropical cyclone risk, found for the near term (until the year 2050), climate change could be expected to increase in the maximum winds of a one-in-100-year hurricane for the Northeast by about 7-11 mph (3-5 m/s, Fig. 4). Hurricane Bob of 1991 hit Rhode Island as a Cat 2 with 100 mph (160 km/h) winds; if it were to hit this year, those top winds might well be 115 mph (185 km/h), making it a low-end Cat 3.
A poleward shift in hurricanes would increase the Northeast risk
Climate change is expected to cause a shift to the north in where hurricanes reach their maximum intensity, which would increase the major hurricane threat to the Northeast. For example, a 2024 modeling study, Robust future projections of global spatial distribution of major tropical cyclones and sea level pressure gradients, showed a significant shift northward toward the northeast U.S. for major hurricanes by the end of the century — primarily because warmer waters farther to the north will allow hurricanes to retain their strength as they head north.
More hurricanes may form near the U.S. coast, increasing the number of landfalls
As we detailed in a previous post, The future of Atlantic hurricane tracks, as the waters near the U.S. coast warm because of climate change, we might expect that an increasing number of hurricanes could form in waters that were formerly too cool. This would likely increase the Northeast hurricane risk, as exemplified by 2021’s Hurricane Henri.
Summertime Greenland blocking is increasing, amplifying the Northeast’s landfalling hurricane risk
An atmospheric phenomenon known as Greenland blocking — a “stuck” jet stream pattern over Greenland — can deflect hurricanes approaching the Northeast from following their usual recurvature path to the northeast, forcing them instead to the northwest, resulting in increased landfall risks to New England. This phenomenon occurred with Hurricane Sandy in October 2012. According to a 2024 paper, Greenland summer blocking characteristics: an evaluation of a high-resolution multi-model ensemble, summertime Greenland blocking episodes of the type that would cause increased hurricane landfall risks to the Northeast U.S. showed a “sharp increase between the 1990s and 2010s,” reaching record levels. Such an increase could be caused by Arctic sea ice loss from human-caused climate change, though natural decades-long weather variability may also be responsible.
Unfortunately, climate models have a tough time correctly predicting Greenland blocking, so we are in the dark on how this particular threat might evolve. A 2013 paper, Model projections of atmospheric steering of Sandy-like superstorms, argued that our best climate models project we should see a decrease in the type of steering patterns that brought Sandy to the coast at such an unusual angle.
A 2023 paper, The impact of changes in steering patterns on the probability of hurricanes making landfall in the New York City area, analyzed observed variations in hurricane steering currents over the North Atlantic from 1979 to 2022. The researchers found that the probability of tracks to the northwest toward a landfall into the Northeast, like Hurricane Sandy’s, increased from approximately 0.26% in 1979 (0.48 days) to 0.45% (0.81 days) per season since 1979.
Since Hurricane Sandy, there haven’t been any similar hybrid superstorms — a hurricane that merges with an extratropical storm, then gets slung to the northwest into the U.S. East Coast. However, there have been a few model runs that suggested that possibility for Hurricane Eta in 2020 and for Hurricane Melissa in 2025. If Sandy were to recur in a much warmer future climate (the very high-emissions A2 scenario), the storm would be about 10 mb stronger by the end of the century, according to a 2015 paper, Hurricane Sandy before 1900 and after 2100.
Rapidly moving hurricanes are the norm for the Northeast, but global warming could slow them down.
Rapid forward speeds at landfall are to be expected for Northeast hurricanes because the jet stream winds are typically strong that far north, propelling a hurricane quite fast. All five of New England’s major hurricanes were moving very rapidly at landfall, between 29 and 50 mph (Fig. 6). Storms with such a fast forward speed will cause relatively modest damage on their left (west) side, with severe wind damage and a high storm surge on their right (east) side. Damage from flooding rains is limited with fast-moving hurricanes, because they don’t stick around long enough to pile up prodigious rainfall amounts.
A slowdown in hurricane motion has been observed in recent decades, as well as an increase in the annual frequency of stalling behavior of about 1.5% per year over the satellite era, 1966-2020. In addition, a slower rate of decay over land has been occurring, with a typical hurricane losing about 50% of its intensity in the first day past landfall, compared to 75% in the late 1960s. Slower-moving hurricanes present a lower wind damage risk to the Northeast because such storms will usually weaken quickly as they cross the several hundred miles of cool ocean waters that lie between the northernmost portion of the warm Gulf Stream current and the coast of New England. For example, Category 1 Hurricane Henri of August 2021 moved unusually slowly for its northerly latitude (11 mph) as it approached the coast of New England, and the much cooler waters it had to traverse during its last 12 hours before landfall weakened it to a 60-mph tropical storm by the time it hit Rhode Island.
However, slower-moving hurricanes present an increased risk of heavy rains. Henri dumped widespread rainfall amounts of five to 10 inches over the Northeast, which fell upon soils already saturated by the remnants of Tropical Storm Fred a few days prior. The widespread flooding killed two and caused over $650 million in damage.
In some cases, a slower-moving storm may produce a long storm surge over multiple tidal cycles, increasing the chance that a damaging storm surge will arrive at high tide.
A 2018 paper by James Kossin, A global slowdown of tropical-cyclone translation speed, reported a 20% slowdown in storm forward speed since 1949 over land for Atlantic storms. His co-authored 2019 follow-up paper presented new data showing that the highest quality and longest-running storm speed dataset that we have for tropical cyclones — for those over land in the continental U.S. beginning in 1900 — has shown a significant slowdown of 17% from 1900 to 2017.
These two studies did not attempt to attribute the slowdown to human-caused climate change, saying: “there is not at present a clear mechanism explaining the observed tropical cyclone speed reduction.” But Kossin said in an email that “the vast majority of the models consistently predict a slowing of the tropical circulation due to human-caused global warming. So, in my opinion, given that tropical cyclones are somewhat passively carried along in these winds, a reasonable hypothesis is that they are slowing down with warming. This is fairly compelling evidence for a human fingerprint on the slowing we’re observing.”
Kossin’s results were supported by a 2021 paper, Acceleration of tropical cyclones as a proxy for extratropical interactions: synoptic-scale patterns and long-term trends. That study found that both the forward speed and acceleration of Atlantic hurricanes undergoing recurvature (because they are caught in the jet stream) have been slowing down in recent years. The authors speculated this may be from a poleward shift and weakening of the jet stream. The observed trend in slower-moving hurricanes can be expected to continue into the future, according to a 2022 modeling study, Tropical cyclone motion in a changing climate. The paper found that future human-caused global warming could shift the jet stream northward, leading to slower hurricane steering currents at midlatitudes and a significant slowing of hurricane motion, including in the Atlantic off the East Coast of the U.S.
A 2022 paper, New York State Hurricane Hazard: History and Future Projections, found that the coast could expect in the future to see more slowly moving hurricanes, with an increase in the number that would approach the coast at a more east-to-west angle, like Hurricane Sandy of 2012. Increased sea surface temperatures and reduced vertical wind shear would likely increase the intensity of New York’s future hurricanes, the modeling found.
A major Northeast hurricane would be hugely expensive
The Northeast is a population-dense region with several large coastal cities, including New York and Boston. Because there is so much infrastructure near the coast, hurricane damage costs can add up fast.
A 2022 paper, Normalized Hurricane Damage in the United States: 1900-2022, estimated that if Hurricane Carol of 1954 were to hit in 2022, its total economic damage would have been $41 billion. The 1938 “Long Island Express” hurricane would have cost much more: $102 billion.
It’s important to emphasize that total damages are much higher than insured damages. For Hurricane Sandy of 2012 — the most expensive hurricane on record for the Northeast — about 50% of its $91 billion (2025 USD) damage was insured. More generally, the fraction of damage from major U.S. hurricanes over the past 10 years that was insured ranged from about 20-60% of the total damage, using insured damage estimates from multiple insurance firms as summarized by beinsure.com and total damage estimates from NOAA/Climate Central.
In an email, catastrophe modeler Suz Tolwinski-Ward, a vice president at Verisk Catastrophe and Risk Solutions, said that were the 1938 storm to occur today, insured losses alone could exceed $60 billion (in 2025 U.S. dollars). Within Verisk’s models, the one-in-10,000-year loss storm for the Northeast (defined as New England states, plus Pennsylvania, New York, and New Jersey) arises from a Category 4 hurricane making direct landfall on New York City. Such an event could cause as much as $500 billion or more in insured damage in today’s dollars, given the heavy concentrations of coastal exposure existing there today. Total losses would be much higher.
Our other posts in this series
Bob Henson contributed this post.
Republish our articles for free, online or in print, under a Creative Commons license.
