But like many questions in science, this isn't a case of straightforward cause and effect. Many scientists accept the broad premise that a hotter climate likely contributes to some increase in hurricane strength, that this process is already underway, and that it will intensify. There's also unambiguous evidence that sea level rise, another product of climate change, will contribute to higher, more dangerous hurricane storm surges.
Beyond that, though, the science gets more speculative, as it's based on computer models tracking the complex dynamics of climate and weather and an at-times spotty record of hurricane data. There's a lot of uncertainty built in. Here's a look at what we know, and don't know, about global warming and hurricanes.
Hurricanes are, in effect, giant heat engines. They transfer latent heat energy from the ocean to the atmosphere, transforming some of it into mechanical energy in the process: the maelstrom of hurricane-force winds and giant waves. If you pump more heat into such a system, warming up the atmosphere and the ocean, it stands to reason that the venting will grow stronger. "High sea surface temperatures lead to the evaporation of moisture, which provides fuel for the storm. Then it gives up the latent heat: that is what powers the storm. Together they provide for stronger storms. The evidence is abundantly clear," says Kevin Trenberth, a senior scientist at the NOAA's National Center for Atmospheric Research in Boulder, Colorado.
Trenberth's advice? Just look at Sandy. Sea surface temperatures off the East Coast were about 3 degrees Celsius above average at the time Sandy approached. Perhaps 0.6 degrees of that, Trenberth says, is attributable to global warming. With each degree rise in sea surface temperature, the atmosphere holds 4% more moisture, which may have boosted rainfall by as much as 5% to 10% over what it would have been 40 years ago.
A Vexing Question
Scientists have wrestled with this question for decades, trying to understand the systematic relationships between climate, oceans, and hurricanes. MIT climatologist Kerry Emanuel first suggested a link between climate and hurricanes in a 1987 paper, which proposed a new method for measuring overall hurricane force. Normal measurements such as barometric pressure, maximum strength (based on wind speed), or size, are constantly changing, and thus don't tell you much about the storm's overall performance. Instead, he came up with an absolute baseline called the "power dissipation," roughly equivalent to the total amount of energy a storm expends over its lifetime, which can last weeks. (In physics, "power" is a measurement of energy expended over time.)
Emanuel's analysis of past storm data shows this has fueled the cumulative violence of cyclones, far beyond what his initial theories predicted.
Global temperatures have risen about 0.5 degrees C since the 1970s, sea surface temperatures in hurricane zones by about the same amount. Emanuel's analysis of past storm data shows this has fueled the cumulative violence of cyclones, far beyond what his initial theories predicted. The overall power dissipation of Atlantic storms rose by about 60%, according to a 2007 paper. The potential intensity, a measure of the upper limit of a storm's strength, had gone up by 10%. Both the average duration and the top speed of storms had also increased, the latter by 25%.
Some, but not all, of this extra heat is the result of growing amounts of carbon dioxide and other greenhouse gases, Emanuel says. Some is due to natural causes such as volcanic eruptions. And ironically, better environmental regulations also seem to be contributing to the problem. The Clean Air Act limited the emissions of sulfate aerosols, microscopic particles produced in the burning of fossil fuels. But these particulates had a hidden, beneficial effect: the haze they create reflects some sunlight back into space. Without them, the atmosphere traps more heat.
If Emanuel is correct, then can we see the footprint of global warming when a huge storm hits? Not necessarily. Landfalling storms may be the only ones that matter to most people, but they are only a fraction of the total, a transient snapshot.
"When we look at storms over their whole lifetimes, that's a whole lot more information than you get with high intensity storms at the point they are affecting land," Emanuel says. Hurricanes that hit land make up a small dataset with a lot of statistical noise, in which warmer temperatures are just one factor. So far, it doesn't show any climate signals. "One thing that makes it very complicated from the viewpoint of climate scientists: in the case of global mean temperatures, we have records going back to the 1800s showing warming as a significant trend. But if you look back at landfalling hurricanes, there's no trend we can identify," says Tom Knutson, a research meteorologist at NOAA's Geophysical Fluid Dynamics Lab in Princeton, N.J. It will take decades, Emanuel says, before scientists have enough data to establish the connection between climate and landfalling storms.
Evidence of a trend, or just a cycle?
Emanuel has his detractors, including Christopher Landsea, a meteorologist at the National Hurricane Center, who say that he and other modelers overestimate the likely boost in hurricane effects, in part because their data is incomplete. In papers and public talks, Landsea has suggested the global warming effect on hurricanes is virtually insignificant—an increase of about 1 to 2 mph in top windspeed, he says. Landsea believes that the flux in hurricane frequency and strength, including the past decade's more active hurricane trend, can be better explained by cyclic ups and downs called the Atlantic Multidecadal Oscillation. Emanuel doubts the AMO's existence: he believes it's the result of reading too much into ambiguous data. Their debate is as old as climate science itself, having roots in an ongoing split in perspective between climate modelers and meteorologists.
If coastal cities are still there in 2100, what kind of hurricane threats will they face? Climate modelers trying to answer those questions have wrestled for years with a difficult problem—climate models operate on a global scale, while the weather models used to predict hurricane strength and tracks are much more fine-grained. Putting them together to predict how climate will affect storms decades in the future risks compounding uncertainties in both models.
One of the biggest of those? The current understanding of hurricanes, as well as most of the hard data scientists input into their models, comes from satellite observation. But reliable weather satellite data goes back only about 40 years, a short time span on which to base your prediction of future trends. Yet as computing power has grown, the models have improved.
Last year, Knutson and other modelers at the NOAA Geophysical Fluid Dynamics Lab took established climate models and embedded an established hurricane forecasting model within them. This way they could successfully model past hurricane seasons and then extend those trends ahead to 2100, as sea levels rose and the atmosphere grew hotter. Different climate models spit out different pictures of the future, so they ended up with an assortment of scenarios. But a couple of trends were clear. Overall, the number of tropical cyclones fell, while the frequency of strong storms—Category 4 and 5 on the Saffir-Simpson scale—nearly doubled.
If coastal cities are still there in 2100, what kind of hurricane threats will they face?
There are a couple of reasons for this mixed result. Hurricane formation depends on a low level of windshear in the upper atmosphere—that is, the difference in windspeeds at adjacent levels. Pronounced windshear lops off the top of a forming storm and makes it dissipate. Warmer sea temperatures tend to increase windshear, leading to fewer storms overall. When big storms do form, meanwhile, they churn up cooler water from beneath the surface, which also tends to tamp down nascent storms that might otherwise form.
A dangerous future
These results are bad news because stronger storms are far more dangerous than weaker ones. A 2005 study that examined hurricane impacts from 1900 to 2005 found that Category 4 and 5 storms accounted for only 6% of U.S. landfalls, but caused 48% of all hurricane damage. Using this study as a starting point, and accounting for the projected mix of more bigger storms and fewer smaller ones, Knutson's team estimated that by 2100, the overall destructive potential of hurricanes may increase by 30%.
Hurricanes do most of their damage with high winds and storm surges. The global warming effect on the former is debated, but the latter isn't. Global sea levels have risen approximately 1.7 mm per year between 1950 and 2009, and at an accelerated pace of 3.3 mm from 1993 on. This is due to climatic changes. Warmer water expands, melting ice puts more water in the ocean, and rainfall patterns have shifted. Sea level rise is worse in some places than others. As fate would have it, the Northeast Atlantic coast is one of those unfortunate locations. One recent study showed that sea levels from North Carolina to Canada have been rising at three to four times the global average since 1950. By definition, higher seas mean higher storm surges. This means that storms that might once not have caused a problem are getting more dangerous. And huge storms, whether amplified by global warming or not, can go from destructive to catastrophic. The danger is compounded by the fact that most coastal fortifications were built when sea levels were lower, on the assumption that conditions wouldn't change.
So whatever uncertainties exist about the role of global warming in hurricanes, the dangers storms pose are worsening. Meanwhile, improving observations and modeling will produce new insights into the relationship between climate and extreme weather. Sometimes, for instance, warming in one part of the world has unexpected effects in others.
Hurricane Sandy, for instance, might have headed out to sea were it not for a blocking ridge of high pressure to the north that guided it into a low pressure system moving eastward, creating a giant hybrid storm. One recent study suggests that such blocking highs may be related to melting ice in polar regions, which contributes to a "wavier" jet stream, the air current flowing from west to east across the Atlantic. "When the jet stream is in wavy configurations those waves tend to move eastward more slowly," says Jennifer Francis, a Rutgers climatologist and one of the authors. "All meteorologists know that the reason that's important is, those waves are what create our weather so it increases the chances of having a stuck weather pattern." Once "stuck," this pattern might lead to drought, more rainfall—or create a hurricane highway. We won't know the outcome for certain until it happens, but when it does, we had better be prepared.