Most scientific research on the influence of climate change on hurricane activity has focused on the relationship between warm Atlantic Ocean conditions and tropical activity in the basin, but what re/insurers really want to understand is how climate change will affect the risk of U.S. hurricane landfalls.
Results from General Circulation Models (GCM) are finally getting to a resolution where storms are occurring with nearly the right frequency and intensity, but long-term simulations are still very time consuming to run. And extracting insight from the historical record is challenging given the large interannual variability, which makes it difficult to identify impacts from climate change.
So we can instead use our understanding of the physics and thermodynamics of how hurricanes form to identify the importance of warm ocean conditions and examine years with anomalously warm sea surface temperatures (SSTs). The data show that when the Atlantic is warm (as it is under the positive phase of the Atlantic Multidecadal Oscillation, or AMO), basinwide hurricane frequency is, on average, elevated. If we assume that history portends the future and that the Atlantic will continue to warm, we can also assume that basinwide hurricane activity will increase; yet we still have not answered the question regarding how landfalls will change.
But will there be more U.S. hurricane landfalls?
To address the question, we must determine the probabilities of tropical storms becoming hurricanes and then making landfall along the U.S. coastline. If these probabilities are not tempered by warm SST conditions, we can conclude that increases in basin activity translate into increased hurricane landfall risk.
Since 1948, there has been an average annual frequency of 10.6 storms per year. Of these, about 58 percent have become hurricanes; the long-term probability of a tropical storm both becoming a hurricane and making U.S. landfall at hurricane strength is about 14 percent.
Are landfall probabilities influenced by changes in climate now?
An analysis performed by AIR showed that the long-term probability of tropical storms becoming hurricanes is not climatologically different when the ocean is warm. Warm conditions generally lead to more tropical storms, but the rate at which storms become hurricanes has been fairly stable since 1948. So, at least in the last 70 years, the impact of a periodically warm ocean on the probability that a tropical cyclone will intensify to hurricane status is relatively stable.
Since 1995, the Atlantic Ocean has been warmer than average and most seasons have seen elevated tropical storm activity. The correlation between warm SSTs and landfall frequency is not so clear. The long-term hurricane landfall frequency is about 1.5 storms per year. Despite the elevated frequency of tropical storms in most of these years, only about half the years had a higher-than-average number of hurricane landfalls.
In fact, in only four of the last 12 completed seasons has the landfall proportion exceeded the long-term average. The highest proportion occurred in 2004, but there have been several years since with no landfalls at all. Variability in both basin and landfall activity leads to large fluctuations in landfall proportion from one season to the next. This contradicts the notion that landfall proportion is constant over time.
But wait there’s more …
If the only thing there was to know about Atlantic basin and U.S. landfall hurricane activity was how warm the oceans are going to get, we would have much more confidence in what the future would look like. But, other aspects of the climate will also likely change in ways that are not so easily understood. For example, the biggest driver of basin activity is the existence, strength, and flavor of El Niño events that develop; for the Atlantic, El Niño increases vertical wind shear over the main development region, which inhibits basin and landfall activity.
Some studies suggest that El Niños may become more frequent and more intense, thereby reducing basin—and possibly landfall—numbers. Climate change will also continue to weaken the pole-equator temperature gradient. Some studies have shown that this will cause major changes in atmospheric circulation; weather patterns may become more persistent and weather systems (e.g., fronts) may move more slowly [eastward]. How these changes will interact with tropical cyclones coming from the other direction is not well understood at all.
Still other studies have shown that these changes would be manifested as a narrower region of trade winds near the equator but larger high pressure cells (e.g., Bermuda High) responsible for steering hurricanes. These changes would result in many fewer weak storms and probably an increase in stronger storms, which would travel farther northward and remain strong, courtesy of the larger steering high and warmer underlying ocean. That could mean an increased frequency of hurricanes [making landfall] in the Northeast with intensity that once was not possible.
AIR's work in the area of climate and hurricanes is by no means complete, and conclusions reached thus far are tentative. More refined views of how hurricane risk shifts under a changing climate will come with more historical experience and more research.
Dr. Peter Sousounis is assistant vice president, director of meteorology in the Research and Modeling Division at AIR Worldwide. This article first appeared in AIR Worldwide's In Focus blog and is reprinted with permission.