Is global warming making hurricanes stronger?
Kerry Emanuel says yes and explains why.*
Kerry Emanuel is Professor of Atmospheric Science at Massachusetts Institute of Technology. Emanuel’s research suggests that our warming climate is already turning out fiercer tropical storms. Emanuel spoke with Earth & Sky’s Eleanor Imster in August 2005 about hurricanes in a warmer world.
Imster: When you say we are seeing more hurricanes, does that mean we seeing more storms altogether, or just the same number of storms that are bigger?
Emanuel: Let me be very specific about that. I very definitely did not say we are seeing more hurricanes. I was rather clear on that point and everyone else in the business will tell you the same thing. Globally, the frequency of tropical storms is about 90 per year, of which roughly ten occur in the Atlantic. There are a lot of fluctuations from year to year, and from decade to decade, but if you look at the global total, it’s rock steady.
What is changing, is the total amount of energy consumed by hurricanes over the course of their life. That’s the particular statistic I looked at. That statistic shows big swings from one year and one decade to the next, but if you look at the global statistics particularly, there is on top of that a general upward trend.
Imster: Does that mean we are seeing more Category Four and Five hurricanes, rather than Ones and Twos?
Emanuel: Yes, that has something to do with it. More of the storms are reaching higher intensities as measured by wind speeds, or as measured by their categories. But on top of that, they’re lasting longer. And that really does affect my statistic as well, that is they tend to maintain high intensity longer than they used to.
Imster: For your research, you looked back at history. What kinds of data did you use?
Emanuel: Hurricane data is collected differently in different parts of the world. And this is actually is the issue that one has to spend the most time on in doing research of this kind.
The best database in terms of quality, is in the North Atlantic, where we have been flying airplanes into hurricanes since the late nineteen forties. The very best measurements of wind speed and central pressures in hurricanes come from aircraft measurements. The Atlantic is fairly well measured. Another place where airplanes flew into hurricanes systematically was in the Western North Pacific. That also began in the later 1940s, but was discontinued in 1987. Elsewhere in the world, such as in the Indian Ocean and all of the southern hemisphere, we’ve never had systematic airborne reconnaissance of hurricanes. In those basins, the only other means to estimate the strength of hurricanes is from satellite data. But the method of estimating hurricane strengths from satellites wasn’t developed until the mid–seventies. It’s fairly new stuff, and it’s fairly subjective as well. It depends upon examining satellite images. Although the technique is fairly refined, it does depend very much on human beings looking at pictures and making decisions.
Imster: Are those NOAA satellites or NASA satellites?
Emanuel: Both. And European satellites and Japanese satellites. We use everything we can get our hands on. That sort of technique has been relied on in the whole Western North Pacific since 1987, and elsewhere outside the Atlantic since the mid–1970s when these techniques were developed. Of course, we didn’t have systematic satellite coverage of the world until about 1970, so it’s very difficult to go very far back in the historical record. Even in the Atlantic, although they started measuring with aircraft in the 1940s, the means of estimating wind speeds didn’t get really good until the late 1950s. And so we don’t have a very long record of this, and that’s what makes this kind of analysis difficult.
Imster: How do you learn about the times before 1940, or 1900?
Emanuel: You can’t. There is no way of doing that. Well, there is a new field developing called paleotempestology, which looks for proxies for hurricanes in the geological record. That’s actually a very exciting field. They do things like look at lakes that are separated from the ocean by a sandbar. When you get a hurricane, the storm surge will wash over the sandbar and wash sand into the lake, which normally has a muddy bottom. So, if you go into the lake and take a core, you see mud with interspersed sand layers. Geologists can date the mud, using radiocarbon techniques, and make a pretty good estimate of when the sand layer was put down. That’s one technique. The only problem is that so far this has only been applied in a handful of places around the world, not nearly enough to get a feeling for global activity. There is hope that if these techniques are applied much more widely than they are to date that we might be able to make some inferences about global climate change from these records.
Imster: Some scientist say that we don’t have enough data yet to say whether or not global warming is intensifying tropical storms, that these recent heavy hurricane seasons are not due to global warming, they’re just part of a natural 20 to 40–year cycle. If our hurricane data only goes back to the 1940’s, how could you even make that kind of statement?
Emanuel: Those are very good points. There are a few things we can say by looking strictly at the Atlantic record of landfalling storms, that is, not hurricanes while they’re over the open ocean, but at the point they hit land. Because people can observe from the land, we have records of those going back reliably, to some degree, from the middle of the nineteenth century. It’s those sorts of records that lead people to say that there are 20– to 40–year oscillations in the activity. One thing that’s important to keep in mind is that only about twelve percent of all the storms that occur in the world occur in the Atlantic, but the Atlantic gets by far the most press. So, when people say it’s dominated by 20– to 40–year cycles, they’re talking about the Atlantic. I’m talking about the world. And when you look at the whole world, you see far less of those 20– to 40–year cycles, because they tend to cancel in different parts of the world. So, the global trend is more evident.
Imster: Here in the North Atlantic, where are we in this cycle?
Emanuel: We started a big upswing in the mid–1990s. And that upswing had been predicted by a number of people who specialize in that. So we’re in the middle of a big upswing, and I think you get widespread agreement about that, especially from people who live in Florida.
Imster: We’ve had an intense active hurricane season here in the North Atlantic. Has that been true in other parts of the world as well?
Emanuel: No. In fact, it’s been fairly quiet in the Western North Pacific. We do tend to see, by the way, that when one ocean basin’s active, some of the other ocean basins may be more quiet than usual. That’s why when you look at the global number of storms per year, it only varies by about ten from year to year. It’s a pretty steady number. There may be a lot of storms one year in the Atlantic and a few in the Pacific, and vice versa. There’s kind of a seesaw there.
Imster: Many scientists theorize that global warming should increase storms’ intensity. What is it about a warming climate that would do that?
Emanuel: A hurricane’s powered by the amount of heat that can be transferred from the ocean to the atmosphere by evaporation of seawater. We know that as you raise the temperature of the water, the amount of water you can evaporate into the air goes up. That’s fairly well quantified, both theoretically and in global climate models. Adding greenhouse gas to the atmosphere means that, to maintain equilibrium, you have to evaporate ocean water a little bit faster, which requires that there be more potential for evaporation. This is a place where theory and models are in pretty good agreement, that as you raise the temperature of tropical oceans, the potential power of hurricanes goes up.
Imster: Do you think that in a warmer climate we could see not only more intense hurricanes, but more hurricanes?
Emanuel: That’s very good question. The research that we were talking about has to do with how strong storms can become once they’re there. But we just don’t collectively have much understanding about what controls the frequency of events. And there’s no evidence globally that it’s changing. Now, in the Atlantic there are these big swings. But when you look at the whole world, it’s pretty steady.
Imster: How big can hurricanes get?
Emanuel: Let’s be careful to make a distinction that we meteorologists make. When we talk about size, we mean the geometric size of the storm, whether it’s 100 miles across or 1000 miles across. When we talk about intensity, we means how strong the winds can get. We know that there’s an upper limit on hurricane size. It’s well defined. And it goes up with this thermodynamic limit on intensity. But it also goes up as you get closer to the equator. So, if you’re in the deep tropics, you can have a geometrically bigger storm than you can have at high latitudes.
Below this limit, hurricanes come in all sizes. There was a hurricane that completely destroyed the city of Darwin in Australia on Christmas day in 1974, called Tracy, and the eye of Tracy was only a few miles in diameter. It was literally about as tall as it was across. It was almost like a big tornado in size. It was a tiny storm, but very, very devastating. On the other hand, there have been typhoons in the Pacific whose eye diameter is about 100 miles across, really enormous storms. You can put all of Tracy, the whole circulation and everything, inside the eye of one of these big Pacific typhoons. And we don’t really know what determines the geometric size of hurricanes. Except that we know that they can only get so big. And that sort of bounding size does go up a bit when you make the ocean warmer.
Imster: And what about intensity?
Emanuel: Well, again, there is a very well–developed body of theory on intensity which says there is an upper limit. Now there is argument among professional researchers about dotting the “i“s and crossing the “t“s on that upper limit, and there are some coefficients there that aren’t very well known. But when we actually look at the data, we don’t ever see storms that exceed that upper limit. That’s one of the nice triumphs of theory: theory says you say you can’t go faster than this, and some storms get right up to the limit, but they don’t go faster than that. Now that limit is function of the climate. If you make the climate warmer, that limit goes up.
Imster: It sure would be nice to be able to look back in Earth’s climate history to see if that limit is higher during warmer periods.
Emanuel: We’re able to look at that limit for the last few decades, and in certain parts of the tropics, it is indeed has started to go up for the last couple of decades. But it’s very hard to go back much further than a few decades for that.
But in terms of the threats that hurricanes pose, most of us who work in the field think that the main problem isn’t global warming, or even necessarily these cyclical changes. It’s that huge numbers of people are moving to the coasts and building flimsy structures. And projections of damages from hurricanes are skyrocketing. But that’s not because of global warming, it’s because of demographics.
If you look at the US history of landfalling hurricanes and the damage they cause, you don’t see any trend in that going back for 100 years or so, once you make a correction for the fact that terrific numbers of people have moved into coastal regions. Of course, you have to correct for inflation too. Hurricanes hitting land are fairly rare phenomenon in absolute terms, particularly intense ones. I mean we have had only two or three Category 5 hurricanes hit the US coastline in recorded history. But on the other hand, if you look at damage just corrected for inflation and not for demographics, you see this huge upswing. And that’s all driven by the fact that people are moving to the coast in record numbers, and that’s really the problem.
