15 Sep 2010

Exploring the Links Between Hurricanes and Ocean Warming

One of the more contentious issues facing climate scientists is whether rising ocean temperatures will cause more frequent and powerful hurricanes. In an interview with Yale Environment 360, Kerry Emanuel, of the Massachusetts Institute of Technology, says that amid the uncertainty, one thing seems likely: an increase in the most potent — and destructive — storms.

In 2005, just weeks before Hurricane Katrina devastated New Orleans, MIT meteorologist Kerry Emanuel published a paper in Nature saying that the wallop packed by hurricanes was increasing as sea surface temperatures climbed. His research was cited by many as evidence that the world could expect more Katrinas in the 21st century. But in 2008, writing in the Bulletin of the American Meteorological Society, Emanuel seemed to say, "Whoa. Not so fast." While hurricanes might become more powerful, Emanuel said, there will likely be fewer hurricanes overall this century, and hurricane trends would vary widely around the globe.

Kerry Emanuel
As the 2010 Atlantic hurricane season winds down, Emanuel spoke with Yale Environment 360 senior editor Fen Montaigne about his current thinking on the issue of hurricanes and climate change. The bottom line, said Emanuel, is that sea surface temperatures in the tropical waters where hurricanes and typhoons form have increased by nearly 1 degree F in the past century. Higher sea surface temperatures are likely to lead to more powerful and destructive hurricanes — categories 3, 4, and 5 — while, paradoxically, a warming atmosphere will probably reduce the number of less powerful storms.

But Emanuel stressed that much remains to be learned about hurricanes, including why it is that roughly 90 tropical cyclones — about a dozen of which are in the Atlantic — seem to form every year. Models of future hurricane behavior, says Emanuel, tend to be “all over the place.” And natural feedbacks — hurricanes, for example, churn deeper, colder water to the surface, lowering sea surface temperatures — complicate the science of forecasting future hurricane trends.

One thing, however, is certain, says Emanuel: Government policies encouraging people to live in vulnerable, low-lying coastal zones — think South Florida — are folly. State and federal subsidies and bailouts of coastal home and business owners essentially mean that people living in less exposed inland areas are paying for the risk taking of coastal residents, who are frequently well-off. “We have the situation of hard-working people in factory jobs and farmers subsidizing the landowners of Palm Beach,” said Emanuel. “That’s crazy.”

Yale Environment 360: The topic of hurricanes and what might happen to them as the world warms has been a subject of great interest to you, and your position has evolved. I wanted to see what your thinking is now.

Emanuel: Let me turn the clock back to 1987 when we published the first paper on this subject, which made use of a recently developed theory that looks at the energetics of hurricanes.
The vast majority of hurricane damage has been done by a small handful of intense storms.”
And it was basically that the supply of energy is evaporation of ocean water, which transfers heat from the ocean to the atmosphere. And the drain on energy is friction between the winds and the ocean surface. And that allows you to make an estimate of how strong hurricanes could get…It gives you an upper bound of hurricane wind speed, and the upper bound turns out to be a very good predictor of how strong hurricanes can be. And the upper bound is such that you could easily calculate how climate change of any kind affects it. We realized way back in 1987 that CO2-induced warming would increase that speed limit on hurricanes, if you will. And that was also published in Nature [in 2005], and the increase was modest but by no means negligible. The main point of that paper is that there was this fantastic relationship between ocean temperature and hurricane power…It surprised us how much power increase you got with just a little bit of increase in the sea surface temperature. . . .

In the 2008 paper, we had developed a technique for basically inferring hurricane activity from a coarse representation of the climate, such as you might get from a climate model…The main point of the paper is to show the models are all over the place. [But] there is a general consensus that on the global scale there probably will be a decrease in the frequency of storms [hurricanes].

e360: And why would that be?

Emanuel:That’s because if you go to the tropics where hurricanes form, everybody knows it's pretty humid down where we live in the boundary layer of the atmosphere. But if you go two or three miles up, it's actually quite dry…When you try to form a disturbance in the tropics or anywhere else for that matter, if you try to get some thunderstorms together for example, the moist air from near the surface of the ocean rises into the atmosphere. This is what you see when you look at a cumulus cloud or a cumulonimbus cloud. But, when it's dry in the middle atmosphere, some of that condensed water in the form of cloud or rain will re-evaporate into the dry air and cause a downdraft. And that downdraft brings cold, dry air down to the surface. And, it's like pouring cold water on a fire. You know, it just basically wipes out everything that is trying to happen…There is a lot of dry air in the middle atmosphere that tends to suppress their formation…

e360: And in a warming world you would expect more of that suppressive dry air in the middle atmosphere?

Emanuel: Yes, the middle atmosphere, in a relative sense, dries out if you warm it…That's what causes there to be a decline. But there is something fighting against that, which is that the energy available for hurricanes, as we predicted in 1987, also goes up.

e360: Because of increasing sea surface temperatures?

Emanuel: Yes, basically…Now, the thing that we were very careful to point out in this paper is [that] the change in [hurricane] activity depends very much on whose climate model you look at. And the second thing we said, which is terribly important, is that it varies a great deal from place to place around the world. So, it may be that the global average [of hurricanes] is declining, but the Atlantic is sailing upward…

The other thing you should be very careful to understand if you look at hurricane damage over the last 100 years, the vast majority of damage has been done by a handful of storms that happened to be very intense when they made landfall. Category 3, 4, and 5 events have caused 80 percent of the damage in the United States.

e360: Hurricanes like Katrina?

Emanuel: Right, Rita, Andrew and so forth. The vast majority are still 1’s.
Hurricane power has been going up a lot in sync with the temperature of the tropical Atlantic.”
The Category 1’s and 2’s are far more numerous than 3’’s, 4’s, and 5’s, but they hardly do any damage at all. So, when you count events, if you say the frequency is going up, you're basically saying the frequency of 1’s and 2’s is going up because they dominate the frequency. So it doesn't really matter. What you really ought to care about are the 3’s, 4’s, and 5s. In just about every model simulation the frequency of those goes up in the global warming.

e360: Can you explain why?

Emanuel: So we have a bell shaped curve, and the peak of the curve is at a wind speed of 70 or 80 knots, and there's a little tail that sticks out, at 150 knots or something. Now, what happens in climate change is that bell gets a little bit squatter but it also gets fatter. And so, if you're looking at the middle of the distribution, you actually see a decline in the number of events. But, if you're looking way out on the tail you see an increase in the number of events…So it follows that the number of 160-knot storms goes up.

e360: Just so I understand, because I realize it's not as simple that as sea surface temperature goes up hurricanes become more frequent and more powerful. But, if you have warmer sea surface temperatures, more radiation, more heat going into the surface of the ocean, is the reason that you are likely to get more storms of that 3, 4, and 5 category simply because that greater radiation creates greater evaporation and larger storms?

Emanuel: You almost have it right. The key is to recognize that the strength of the hurricane is related to what we call the evaporation potential of the surface…In other words, that's the heat balance of the upper ocean. The ocean gains heat by radiation both from the sun and infrared radiation from the atmosphere or the clouds within it. But it loses heat mostly by evaporation. And so, the greater the amount of radiation coming into the sea, the greater the rate of evaporation has to be to balance that…That means that you have to have a greater evaporative potential. It's that potential that dictates hurricane wind speeds . . .

e360: Have you observed more energy in storms in recent decades?

Emanuel: Yes, especially in the Atlantic. We get into some nasty problems here of detection. And, the problem is that the Atlantic only contains about 11 percent of all the tropical cyclones on the planet, [but] is the only place where we routinely fly airplanes into hurricanes . . .

But when you look at hurricane power, it really has been going up a lot in the last 20 years or so in the Atlantic in sync with the temperature of the tropical Atlantic, which by the way is at an all-time record high right now.

e360: What's the high right now, and how much has it gone up in recent decades?

Emanuel: Well, we have to be very specific about when and where. But in the place where hurricanes form in the tropical Atlantic — we call it the main development region — just averaged over August, I think, you're talking about an average temperature of around 28 degrees Celsius…It’s about a half a degree centigrade [.9 Degrees F] warmer than it was a hundred years ago.

e360: And is that significant in terms of potential storm formation?

Emanuel: Here again, there's a problem because there is a constant tendency to think, well, it's just a function of the ocean temperature.
If you believe the global models, you’re talking another two degrees at least of warming in the tropics.”
But, as we discussed a minute ago, the ocean temperature is just a co-factor. It's not the cause but on the other hand, it's a very tempting thing because we can measure it and do measure it quite well. But the 2005 [Nature] paper showed that you get a lot of bang for the buck if you insist upon correlating hurricane power with ocean temperature. You know, a little bit, two-tenths of an increase, gives you a big increase in power.

e360: And globally are scientists measuring [ocean temperature increases] in the tropical regions where hurricanes form?

Emanuel: Oh, yeah. I mean, the temperature just about everywhere in the tropics is going up.

e360: By a half a degree over the last century?

Emanuel: Yeah, by a roughly similar amount over a hundred years.

e360: And I know scientists like Susan Solomon and others are saying, `You ain't seen nothing yet because a lot of this heat in the atmosphere is going to start transferring into the ocean.’ What is your thinking about what kind of sea surface temperature increases we could be looking at in the 21st Century?

Emanuel: If you believe the global models, you're talking about another two degrees at least of warming in the tropics.

e360: Wow, two Celsius?

Emanuel: Yes, but there are a lot of caveats…Hurricanes may matter themselves for the climate. In other words, there may be some feedback. We think there actually is a feedback of hurricanes on climate. And, the way that they feed back principally is by their effect on the ocean. You see, when you get a hurricane one of the things it does is churn the cold water out from the depths up to the surface, and it thereby keeps the surface of the tropical ocean cooler than it would otherwise be. Now, if you did start to see an increase in hurricane power, it's possible that that would be negative feedback on ocean temperature. It would keep the temperatures from going up quite that much. We don't really know how effective that might be. There is just beginning to be some work on this, but if we don't include that effect, you're looking at around two degrees [Celsius] additional warming.

e360: Which, according to your research and your models, has got to lead to an increase in more powerful storms?

Emanuel: Yes, most of the models in the theory suggest that the more powerful storms become more numerous, and those are the ones you worry about…I think the thing that people should see as interesting is the frequency of intense events. And that sort of goes up everywhere, but it goes up very unevenly. The reasons why it doesn't go up uniformly is because the climate is a complicated system, and there are changes in atmospheric and oceanic circulation which cause some regions to warm up more than others . . .

e360: If you talk to climate scientists. there is an expectation of a sea level rise in the next century of two feet, maybe five feet, maybe six feet. When you as a hurricane specialist contemplate those kinds of sea level rises in a place like Florida, with increased hurricane intensity, that has got to be really bad news, right?

Emanuel: Yeah, it is bad news because the two big killers in hurricanes are the storm
Most of the models suggest that the more powerful storms become more numerous.”
surge — which is basically like a tsunami but created by the winds of the hurricane rather than by an earthquake — and freshwater flooding. If you're talking about delta regions — and there are lot of delta regions around the world like in Bangladesh that are heavily populated — then you get both. You get increased water coming from the sea and increased rainfall from hurricanes, water coming down rivers. It's bad news.

e360: I would like to get into the issue of government policies that encourage intense coastal development. Given what you're looking at with the likely increase in intensity and power of storms and sea level rises, what needs to be done now to start planning for this?

Emanuel: Even if the climate weren't changing, if we don't want to have more Katrinas, we need to stop subsidizing people to live in dangerous places. The United States is one of the few places in the world that does that, and it does it heavily. So we are basically paying people to move to places where they're at risk. And, it doesn't make any sense…

We have situations where people who are not living in terribly risky places are paying maybe 10 percent too much premium on insurance policies for their houses, whereas people living in dangerous coastal regions are paying way, way less than the risk actually warrants. We in this so-called free market economy do radical things like cap the premiums that insurance companies are allowed to charge. And this is all done because people who live on the coast — not so much on the Gulf Coast, but the East Coast — are usually wealthy and therefore well-connected politically. And they get legislators to pass regulations saying you can't raise my insurance rate, and so the system adjusts in very complex ways. You want to keep insurance companies in business. You can't force them to take huge losses, so you quietly allow them to overcharge people living inland. And, it's a very unfair system. It's a net transfer of wealth from poor to rich, and the worst aspect of it is it promotes huge coastal development. And, therefore, huge damages when even ordinary storms make landfall in the United States.

If you go to other developed countries…I was just in Taiwan. It's got a very hurricane-prone coastline. The east coast of Taiwan gets battered every year. If you look at it, it's a much more sensible arrangement than we have. There are a handful of fortresses that wealthy people build so that they'll take a Category 5. But everything else are sea shanties. Every farmer and fisherman has got a plywood thing down on the beach where they go for a couple of weeks with their family in the summer. And every two years it blows away. So what? They rebuild it. It has no value. There is no insurance involved. That's very sensible. It's also very democratic, and they don’t have Katrina disasters in Taiwan. We have Katrina disasters. They're political disasters. People call them natural disasters to direct attention away from where it belongs. They're political disasters, and nobody in my field will tell you anything different, by the way. This isn't controversial. It's just that nobody knows how to change the system.

e360: And when you talk about this system of subsidies that has led to this rush of coastal development, you're talking about not only in private insurance, but are you speaking as well about government disaster aid?

Emanuel: Yes, the three big ones are capping insurance, federal flood insurance, which pays for storm surge damage, and federal disaster relief. And there are solutions to this.
Even if the climate weren’t changing, we need to stop subsidizing people to live in dangerous places.”
One is if you're talking about disasters in some sense they're predictable. Like if you build your house in the flood plain or you build your house in a hurricane- prone region, even if it is declared a disaster you should be forced to pay something, right? This is for the same reason that in a rational [insurance] system, if we have two identical people [and] one smokes, we charge the smoker because it's voluntary behavior, and if he wants to smoke, that's fine. It's a free world, but don't force your neighbor to pay your medical bills, please.

e360: Such a vulnerable stretch like Miami and the South Florida Coast. When you look at the intensity of development — and they're right at a few feet above sea level on the beach — what is going to happen as the 21st, 22nd centuries roll around in places like that?

Emanuel: I think we know what's going to happen. There will eventually be a hurricane there, and it will do far too much damage for the state to cover. You know, the state basically has become the insurer of property in Florida. Everybody knows that a relatively small hurricane will bankrupt the [state hurricane insurance] plan, and they will all go with hat in hand to the federal government. So the rest of us will bail them out. And so we have the situation of hard-working people in factory jobs and farmers subsidizing the landowners of Palm Beach. It's crazy.


SHARE: Tweet | Digg | | Reddit | Mixx | Facebook | Stumble Upon


If we look at the data we find that Northern Hemisphere hurricane activity has been declining to the levels last seen in the 1970s. When the Southern Hemisphere hurricane data is added and analyzed to create a global value, we find that Global Hurricane Energy has fallen to a three decade low. That empirical evidence should be enough to falsify the claims that warming would lead to more storm activity and greater damage.

The one thing that we should get out of the scaremongering is the unreliability of the models that are bring used by the AGW people. The models are clearly not very good and their predictions are not any better than those one would get from flipping a coin. It now looks as if Dr. Landsea was right in his criticism when he resigned from the IPCC for using their platform to make claims that were not supported by the empirical data. If the AGW people are not careful they will make astrology look good in comparison.

Posted by Vangel Vesovski on 15 Sep 2010

This report starts out with:In 2005, just weeks before Hurricane Katrina devastated New Orleans... Folks the very opening line is simply wrong. New Orleans was not destroyed by a hurricane in 2005. What happened was the Corps of Engineers built some poorly designed floodwalls. These flood walls were supposed to protect the city from flooding. They were supposed to be designed such that water would rise up against them and they would hold that water back. What happened was when the base of those walls got wet, they fell over. If those floodwalls had done what they were supposed to do no one would even remember of the Hurricane that missed New Orleans on August 29, 2005. The water never reached to the height of those walls. Imagine, if they had just done what they were supposed to do! Please get it right and refrain from making such incorrect statements as this opening line...

Posted by Roy Arrigo on 15 Sep 2010

Katrina did not directly hit the city of New Orleans. Neither did it devastate it as the author reports. The devastation was due to the poor construction and maintanence of the levee system surrounding the city. The failure of this levee system is what caused the destruction, not nature.

Posted by Thomas Spitzfaden on 15 Sep 2010

Thank you for a fascinating article with unusual (and greatly appreciated) technical depth, but it runs into the shallows when you classify Katrina as just another "natural disaster."

Shallowness and not wind was the problem in 2005 for New Orleans: shallow floodwall
foundations that allowed surge to seep and scour out the bases, and shallow elevations on
structures. Katrina's wind speed was Category 1 when it made landfall. What remained was the storm surge, and this came in well below the design specifications for New Orleans levees as mandated by Congress in 1965.

If the levees had performed up to the specification, the city would have suffered nothing worse than "wet ankles", in the words of hurricane researcher Ivor Van Heerden.

If you're going to make policy recommendations using income levels and political connections,
don't leave out the well documented need to reexamine how the Corps of Engineers and other
public trusts execute their charters to protect their constituencies.

Posted by W. S. Lob on 16 Sep 2010

Re: Explorig the Links Between Ocean Warming and Hurricanes, 15 Sep 2010

As a scientist of Earth History, I've followed the Global Warming issue with keen interest. I'm happy that this article does not spend a lot of energy on the anthropomorphic aspects of the debate. I find it irrelevant and counterproductive to a cleaner and healthier planet. I am quite disappointed however, that a couple of high powered academicians at Yale and MIT believe that
"In 2005, just weeks before Hurricane Katrina devastated New Orleans..." Please, the destruction of New Orleans AFTER Katrina was an engineering disaster of biblical proportions and falls square on the shoulders of the US Army Corps of Engineers. A quick fix to the quoted sentence would be to replace "New Orleans" with "Biloxi".

I love this one - it's aways good to hear what other folks are thinking about you (me). "One thing, however, is certain, says Emanuel: Government policies encouraging people to live in vulnerable, low-lying coastal zones — think South Florida — are folly." Interesting comment. I never felt that I'd been encouraged by the federales to live in coastal La but it's a valid point. And I like the details illustrated further down in the article. The theory is not exactly on point however for a couple reasons. First, humans have preferentially lived near water way before government subsidies and indeed way before governments even existed. It's what we do. Besides the obvious that carbon based organisms (including us) need water to exist, it's where the richest assemblages of fauna and flora exist - think food chain. Second, and here's the political/demographic spin, it's not just coastal communities that live under the threat of flooding and/or take advantage of federal flood insurance. Indeed, over fifty percent of the American population lives within the protection of Federal levee systems. Yes, even "... hard-working people in factory jobs and farmers subsidizing the landowners of Palm Beach. It's crazy."

It's all about the water. It's what we do.

Posted by A E DuVernay III on 16 Sep 2010

A fascinating article, but I kept waiting for the big question: How valid are these models?

The answer is that they are still in the infancy stage. They cannot accurately predict the number in a season, nor even if there will be more or less than average.

This is important work, and I hope Mr Emanuel continues to get funding, but the author needs to ask tougher questions and do some fact checking (like the Katrina issue that posters have mentioned).

Posted by dan houck on 17 Sep 2010

Sorry, I find it a little difficult to fathom when we are in pretty much a 30 year low for hurricane activity. Could there be an increase well yes but will it be due to AGW. Sorry I am not convinced... not even closely. As far as Katrina goes I agree with the above comments about the levee system that let them down Katrina mostly missed.

Posted by Dave on 19 Sep 2010



Trouble in Paradise: A Blight Threatens Key Hawaiian Tree
The ʻohiʻa is Hawaii’s iconic tree, a keystone species that maintains healthy watersheds and provides habitat for numerous endangered birds. But a virulent fungal disease, possibly related to a warmer, drier climate, is now felling the island’s cherished 'ohi'a forests.

Climate Change Adds Urgency To Push to Save World’s Seeds
In the face of rising temperatures and worsening drought, the world’s repositories of agricultural seeds may hold the key to growing food under increasingly harsh conditions. But keeping these gene banks safe and viable is a complicated and expensive challenge.

How Satellites and Big Data Can Help to Save the Oceans
With new marine protected areas and an emerging U.N. treaty, global ocean conservation efforts are on the verge of a major advance. But to enforce these ambitious initiatives, new satellite-based technologies and newly available online data must be harnessed.

A Delicate Balance: Protecting Northwest’s Glass Sponge Reefs
Rare and extensive reefs of glass sponges are found only one place on earth – a stretch of the Pacific Northwest coast. Now, efforts are underway to identify and protect these fragile formations before they are obliterated by fishing vessels that trawl the bottom.

An Up-Close View of Bristol Bay’s Astonishing Sockeye Salmon Runs
The first runner-up in the 2015 Yale Environment 360 Video Contest captures stunning images of the abundant sockeye salmon runs in Bristol Bay, Alaska, and tells the story of a 70-year-old project that has been studying the millions of salmon that annually pour into the region’s rivers to spawn.


Donate to Yale Environment 360