by Stefan Rahmstorf, Michael Mann, Rasmus Benestad, Gavin Schmidt, and William Connolley
On Monday August 29, Hurricane Katrina ravaged New Orleans, Louisiana and Missisippi, leaving a trail of destruction in her wake. It will be some time until the full toll of this hurricane can be assessed, but the devastating human and environmental impacts are already obvious.
Katrina was the most feared of all meteorological events, a major hurricane making landfall in a highly-populated low-lying region. In the wake of this devastation, many have questioned whether global warming may have contributed to this disaster. Could New Orleans be the first major U.S. city ravaged by human-caused climate change?
The correct answer–the one we have indeed provided in previous posts (Storms & Global Warming II, Some recent updates and Storms and Climate Change) –is that there is no way to prove that Katrina either was, or was not, affected by global warming. For a single event, regardless of how extreme, such attribution is fundamentally impossible. We only have one Earth, and it will follow only one of an infinite number of possible weather sequences. It is impossible to know whether or not this event would have taken place if we had not increased the concentration of greenhouse gases in the atmosphere as much as we have. Weather events will always result from a combination of deterministic factors (including greenhouse gas forcing or slow natural climate cycles) and stochastic factors (pure chance).
Due to this semi-random nature of weather, it is wrong to blame any one event such as Katrina specifically on global warming – and of course it is just as indefensible to blame Katrina on a long-term natural cycle in the climate.
Yet this is not the right way to frame the question. As we have also pointed out in previous posts, we can indeed draw some important conclusions about the links between hurricane activity and global warming in a statistical sense. The situation is analogous to rolling loaded dice: one could, if one was so inclined, construct a set of dice where sixes occur twice as often as normal. But if you were to roll a six using these dice, you could not blame it specifically on the fact that the dice had been loaded. Half of the sixes would have occurred anyway, even with normal dice. Loading the dice simply doubled the odds. In the same manner, while we cannot draw firm conclusions about one single hurricane, we can draw some conclusions about hurricanes more generally. In particular, the available scientific evidence indicates that it is likely that global warming will make – and possibly already is making – those hurricanes that form more destructive than they otherwise would have been.
The key connection is that between sea surface temperatures (we abbreviate this as SST) and the power of hurricanes. Without going into technical details about the dynamics and thermodynamics involved in tropical storms and hurricanes (an excellent discussion of this can be found here), the basic connection between the two is actually fairly simple: warm water, and the instability in the lower atmosphere that is created by it, is the energy source of hurricanes. This is why they only arise in the tropics and during the season when SSTs are highest (June to November in the tropical North Atlantic).
SST is not the only influence on hurricane formation. Strong shear in atmospheric winds (that is, changes in wind strength and direction with height in the atmosphere above the surface), for example, inhibits development of the highly organized structure that is required for a hurricane to form. In the case of Atlantic hurricanes, the El Nino/Southern Oscillation tends to influence the vertical wind shear, and thus, in turn, the number of hurricanes that tend to form in a given year. Many other features of the process of hurricane development and strengthening, however, are closely linked to SST.
Hurricane forecast models (the same ones that were used to predict Katrina’s path) indicate a tendency for more intense (but not overall more frequent) hurricanes when they are run for climate change scenarios (Fig. 1).
Figure 1. Model Simulation of Trend in Hurricanes (from Knutson et al, 2004)
In the particular simulation shown above, the frequency of the strongest (category 5) hurricanes roughly triples in the anthropogenic climate change scenario relative to the control. This suggests that hurricanes may indeed become more destructive (1) as tropical SSTs warm due to anthropogenic impacts.
But what about the past? What do the observations of the last century actually show? Some past studies (e.g. Goldenberg et al, 2001) assert that there is no evidence of any long-term increase in statistical measures of tropical Atlantic hurricane activity, despite the ongoing global warming. These studies, however, have focused on the frequency of all tropical storms and hurricanes (lumping the weak ones in with the strong ones) rather than a measure of changes in the intensity of the storms. As we have discussed elsewhere on this site, statistical measures that focus on trends in the strongest category storms, maximum hurricane winds, and changes in minimum central pressures, suggest a systematic increase in the intensities of those storms that form. This finding is consistent with the model simulations.
A recent study in Nature by Emanuel (2005) examined, for the first time, a statistical measure of the power dissipation associated with past hurricane activity (i.e., the “Power Dissipation Index” or “PDI”–Fig. 2). Emanuel found a close correlation between increases in this measure of hurricane activity (which is likely a better measure of the destructive potential of the storms than previously used measures) and rising tropical North Atlantic SST, consistent with basic theoretical expectations. As tropical SSTs have increased in past decades, so has the intrinsic destructive potential of hurricanes.
Figure 2. Measure of total power dissipated annually by tropical cyclones in the North Atlantic (the power dissipation index “PDI”) compared to September tropical North Atlantic SST (from Emanuel, 2005)
The key question then becomes this: Why has SST increased in the tropics? Is this increase due to global warming (which is almost certainly in large part due to human impacts on climate)? Or is this increase part of a natural cycle?
It has been asserted (for example, by the NOAA National Hurricane Center) that the recent upturn in hurricane activity is due to a natural cycle, e.g. the so-called Atlantic Multidecadal Oscillation (“AMO”). The new results by Emanuel (Fig. 2) argue against this hypothesis being the sole explanation: the recent increase in SST (at least for September as shown in the Figure) is well outside the range of any past oscillations. Emanuel therefore concludes in his paper that “the large upswing in the last decade is unprecedented, and probably reflects the effect of global warming.” However, caution is always warranted with very new scientific results until they have been thoroughly discussed by the community and either supported or challenged by further analyses. Previous analysis of the AMO and natural oscillation modes in the Atlantic (Delworth and Mann, 2000; Kerr, 2000) suggest that the amplitude of natural SST variations averaged over the tropics is about 0.1-0.2 ºC, so a swing from the coldest to warmest phase could explain up to ~0.4 ºC warming.
What about the alternative hypothesis: the contribution of anthropogenic greenhouse gases to tropical SST warming? How strong do we expect this to be? One way to estimate this is to use climate models. Driven by anthropogenic forcings, these show a warming of tropical SST in the Atlantic of about 0.2 – 0.5 ºC. Globally, SST has increased by ~0.6 ºC in the past hundred years. This mostly reflects the response to global radiative forcings, which are dominated by anthropogenic forcing over the 20th Century. Regional modes of variability, such as the AMO, largely cancel out and make a very small contribution in the global mean SST changes.
Thus, we can conclude that both a natural cycle (the AMO) and anthropogenic forcing could have made roughly equally large contributions to the warming of the tropical Atlantic over the past decades, with an exact attribution impossible so far. The observed warming is likely the result of a combined effect: data strongly suggest that the AMO has been in a warming phase for the past two or three decades, and we also know that at the same time anthropogenic global warming is ongoing.
Finally, then, we come back to Katrina. This storm was a weak (category 1) hurricane when crossing Florida, and only gained force later over the warm waters of the Gulf of Mexico. So the question to ask here is: why is the Gulf of Mexico so hot at present – how much of this could be attributed to global warming, and how much to natural variability? More detailed analysis of the SST changes in the relevant regions, and comparisons with model predictions, will probably shed more light on this question in the future. At present, however, the available scientific evidence suggests that it would be premature to assert that the recent anomalous behavior can be attributed entirely to a natural cycle.
But ultimately the answer to what caused Katrina is of little practical value. Katrina is in the past. Far more important is learning something for the future, as this could help reduce the risk of further tragedies. Better protection against hurricanes will be an obvious discussion point over the coming months, to which as climatologists we are not particularly qualified to contribute. But climate science can help us understand how human actions influence climate. The current evidence strongly suggests that:
(a) hurricanes tend to become more destructive as ocean temperatures rise, and
(b) an unchecked rise in greenhouse gas concentrations will very likely increase ocean temperatures further, ultimately overwhelming any natural oscillations.
Scenarios for future global warming show tropical SST rising by a few degrees, not just tenths of a degree (see e.g. results from the Hadley Centre model and the implications for hurricanes shown in Fig. 1 above). That is the important message from science. What we need to discuss is not what caused Katrina, but the likelyhood that global warming will make hurricanes even worse in future.
_____________________
1. By ‘destructive’ we refer only to the intrinsic ability of the storm to do damage to its environment due to its strength. The potential increases that we discuss apply only to this intrinsic meteorological measure. We are not taking into account the potential for increased destruction (and cost) due to increasing population or human infrastructure.
References:
Delworth, T.L., Mann, M.E., Observed and Simulated Multidecadal Variability in the Northern Hemisphere, Climate Dynamics, 16, 661-676, 2000.
Emanuel, K. (2005), Increasing destructiveness of tropical cyclones over the past 30 years, Nature, online publication; published online 31 July 2005 | doi: 10.1038/nature03906
Goldenberg, S.B., C.W. Landsea, A.M. Mestas-Nuñez, and W.M. Gray. The recent increase in Atlantic hurricane activity. Causes and implications. Science, 293:474-479 (2001).
Kerr, R.A., 2000, A North Atlantic climate pacemaker for the centuries: Science, v. 288, p. 1984-1986.
Knutson, T. K., and R. E. Tuleya, 2004: Impact of CO2-induced warming on simulated hurricane intensity and precipitation: Sensitivity to the choice of climate model and convective parameterization. Journal of Climate, 17(18), 3477-3495.
Roger says (#49):
Ah, that makes things clearer. I had previously read your statement as meaning that any claims of a linkage between global warming and future hurricane impacts was premature given the current level of understanding. If you are only referring to the attribution of past (current) impacts, then it seems much less contentious.
Some questions:
Did Katrina have an unusually large area (diameter)?
Is the average size (diameter) of hurricanes increasing?
Should hurricanes in the future be larger due to global warming?
What effect should an increased tropopause height have on hurricanes?
Re: #51. James- Yes.
This was a very informative post. RC is doing a good job of providing information on current topics that are big in the news and replying to reader�s comments. The connection between climate change and hurricanes is a very appropriate discussion topic.
Popularizing climate change science is a worthwhile undertaking. IMO when science has political implications, popularizing science becomes very important. A source of accurate scientific information is highly valuable so people can learn about the science and hopefully better understand the political issues. Its good to see that RC does not stay away from controversial scientific topics that might draw some negative attention from those who are concerned about the policy questions that revolve around the science.
As far as rising ocean temperatures are concerned, I know studies (like Barnett et al 2005) demonstrated there is an anthropogenic signal (if I am using the term correctly, meaning a discernable increase in ocean temperature that is due to human activities) globally. I wonder how the level anthropogenic warming varies in different parts of the oceans and how this affects TCs.
There are groups that are helping Katrina victims. Help if you can.
http://www.charitynavigator.org/index.cfm/bay/content.view/catid/68/cpid/310.htm
http://www.networkforgood.org/topics/animal_environ/hurricanes/
http://www.redcross.org/
The link between heat and Hurricane intensity is unquestionable, as atmospheric water vapour density increases with higher temperatures, the energy source is likewise augmented, same goes for cyclones. Ever heard of a dry cyclone? Again stronger cyclonic activity is also found everywhere in the world, including in the Arctic, currently experiencing greater open water, with one Canadian High Arctic storm generating winds in excess of 115 KPH last winter (very uncommon), a precursor, one may say of Katrina and others. The logic is simple and is applicable worldwide. The point of view that Global Warming can’t trigger greater Hurricanes is a weak argument yet to be explained with clear words, but of course it is brought out to discredit that there is any Global Warming in the first place.
Re#5,
Could someone do it theoretically? Well, they could try to correlate AGW to SST to hurricane strength and come up with a number. But what about all the other factors? If AGW is real and influential, couldn’t that have had something to do with the path that hurricanes take? Should AGW sometimes get credit for pushing a hurricane away from Miami, New Orleans, Houston, etc, and towards unpopulated areas (or just back out over the Atlantic)?
We have a decent grasp of forecasting tropical storm and hurricane paths a few days in advance. But I don’t think we can model what Katrina or any other hurricane would’ve done without the existence of AGW to any degree of accuracy.
Look at it this way: you’re on a road trip across Texas. You’re a safe driver and are trying to save on gas, so you drive 45 mph. But at some point several hours into your trip, you fall asleep at the wheel, drive off the road, and hit a tree. You escape with just a broken arm and some bruises think to yourself, “Thank goodness I was only driving 45 mph instead of 55, 65, or 70!” But then you realize that had you been driving 55, 65, or 70 (or whatever speed) the whole trip, you would’ve reached your destination and therefore never fallen asleep at the wheel. Or if you had averaged just 0.001 mph faster the whole trip, you would’ve been further up the road and driven relatively harmlessly into an open field rather than hitting a tree and injuring yourself.
Looking at Katrina from birth to landfall, it seems to me that any significant AGW could have had a number of influences along the way. I think it’s a drastic oversimplication to try to attribute x% of the damage of a hurricane to AGW.
I used to not think humans had much to do with Global Warming, but then I zoomed in (MapGlobalWarming.com) on these satellite images of 785 U.S. facilities that emitted more that 100,000 tons of CO2 during 2004. Some of these locations released more than 20M tons during one year.
The web site takes a few seconds to load, but once it loads, the speed is fast as you click on the icons in the right-hand column and the map zooms in and gives you a satellite image of these smoke stacks.
RE # 34 & 40, something strange (according to our local weatherman) happened when Hurricane Emily came to us (its eye was about 70-80 miles south of us): There were lots of large hailstones that came down around Brownsville, TX. Our weatherman said he was unaware that had ever happened in our area during a hurricane. Seems that would perhaps indicate some interesting or extreme temperature difference was involved.
Now, I just saw “Day After Tomorrow” again (for a paper I’m writing on fictional accounts of GW), and it featured deadly hailstones raining down on Tokyo & elsewhere. I know the science was pretty much trumped up on the film, but maybe there’s something beyond normal expectations in the Emily hailstones, maybe a hint of some GW phenomenon, or maybe this is all just quite normal/natural.
[Response:The ‘science’ in “Day After Tomorrow” was plainly wrong: for starters, it would be impossible for a downsurge of air aloft on that scale, and such thin air would not be able to freeze the ground which in comparison has an enormous heat capacity (it would take time for the heat to drain away…). Furthermore, for thin air high in the atmosphere to ‘fall’ down to the ground, you would need to create a region with near vacuum conditions (people would suffocate and ‘explode’ because of lowered pressure, rather than freeze to death). In other words, it contained some very silly aspects. Even the several-degree cooling of the currents off Greenland while still being open sea, would be impossible, since the temperatures here are just above freezing anyway (the waters first would have to freeze, which would require extra release of heat due to phase change) and then the ice would have to freeze further (you would also take into account heat conduction in ice…). The aspect about hail storms, could be some of the more realistic part of the film, though, but – personally – I would say that this film is purely entertainment (as well as being a critique of politics?) and shouldn’t be drawn into scientific debates. -rasmus. P.S. a descent – air sinking – will lead to a compression of the air in the real world, which would again warm the air. The storm surges also constituted to some of the very unrealistic features of the film…]
RE 56. Or you could have hit the tree at 70 and died, and killed your wife and kids. Sort of like NO. Care to tell us the advantages of faster winds in hurricanes for those of us on the ground?
Re 47
I could find no such headline on the site of Der Spiegel. Could you provide a link?
There is a an article though quoting the chancelloer as saying that Katrina should have consequences for energy policy.
http://www.spiegel.de/politik/deutschland/0,1518,373480,00.html
I am not a climate scientist, just a chemical engineer working on renewable energy.
I do think that there is a broad consensus that there likely is an effect from increased SST, but that so far it is not sufficiently significant to be visible in the impact record, while lots of other things are, including the well known natural decadal cycles affecting hurricanes, flood defenses and increasing population and property values.
Roger also points out on his site that it is not just SST that are affected by climate change, I read this to mean that land use changes, local changes in climate forcings and so forth may counteract the expected increase from SST locally.
Which I take to mean that it is possible that the overall increase consists of a few decreases and many more increases. If that’s the case, it’s not just that landfalling hurricanes occur rarely and therefore statistical power to see a trend is poor, but that for US landfalling hurricanes there truely may be no net effect from global warming so far (I don’t think whether a consensus on the likelihood of that is possible, I guess it would be of the order of 10%).
Finally, I think what the media really want to know when they ask about Katrina and global warming is whether the connection is “significant”, and I don’t think there’s consensus there, partly as Gavin says, that’s a matter of semantics, partly it’s a genuine disagreement as to how important global warming is where hurricanes are concerned compared to other factors, and how “significant” an argument the influence of global warming on hurricanes is for deciding energy policy.
Further to my posting [35] – sorry about the web-link regarding increasing storms over the UK. I hope this one works better (and gives more details, too):
http://www.guardian.co.uk/uk_news/story/0,3604,1103539,00.html
I completely agree with Stefan – I cannot understand at all why people think that climate change would only affect future generations (which, for many of us are our children anyway). I, too, hope to be around by 2050, and when I see nature and wildlife around me change and diminish at rapid speed right where I live and already now, then I don’t know if “hope” is the right word. I found this good summary of people’s perception that climate change is real but will not affect them personally:
http://www.stabilisation2005.com/posters/Lorenzoni.pdf
It is one of the reports submitted to the climate stabilisation conference in Exeter this February, and it gives a link to the most recent public poll results. It makes it clear that people see climate change as some vague threat, rather than relating it to any specific impact that they can see. When I read the media insisting that hurricane Katrina, or the Spanish drought, or repeat flooding in the UK and Central Europe are part of a normal cycle and cannot be linked to global warming, then I can vaguely understand where those believes come from.
Re 37:
Heiko repeatedly equates Kyoto with action on climate change. I am quite aware that this web-blog does not comment on different policy approaches. But we must not use Kyoto a synoymous with ANY action on climate change! Yes, we can all wonder what Kyoto could have achieved, but then we can also all wonder where we would be if the world had adopted a Dutch proposal in 1989 to slash CO2 emissions by 20% at that time, with further cuts to follow. Or we can look at the TAR emissions scenarios or the papers about stabilising the atmosphere at the 2005 Stabilsation Conference instead. And if you are not sure about the benefits of mitigation – I recently went to a talk by an Indian scientist who spoke about the melting Himalayan glaciers and his fear that, if we don’t stabilise the atmosphere, they will go and stop feeding those major rivers that provide water and make agriculture possible for 1-2 billion people. Saving those people must be a pretty big benefit, no matter what costs you calculate.
Re: 60 On hurricanes and SSTs, I am simply repeating what I hear from the hurricane experts, e.g., Emanuel (2005) writes, “only part of the observed increase in tropical cyclone power dissipation is directly due to increased SST.” I am not suggesting anything more (not my specialty).
Look for a very significant paper by Webster et al. coming out in Science that adds considerable nuance and important data to this discussion. It has some interesting data and is consistent with Trenberth (2005), Emanuel (2005) and Pielke et al. (2005). If you focus on recent peer-reviewed studies, rather than quotes in the media, there is a clear consensus on hurricanes and global warming among the scientific community, despite characiatures to the contrary. The debate, such as it is, appears in strident statements made to the media by some scientists (e.g., Trenberth, Gray) and scientifically-incorrect assertions by political advocates (e.g., in connecting Katrina with global warming OR assertions that global warming can have no effect on hurricanes). The reality is more subtle, and the implications for policy more pedestrian.
RE#59,
If you’d been driving 70, you wouldn’t have fallen asleep at that spot, hence you couldn’t have hit that tree. It doesn’t mean you wouldn’t have hit another tree somewhere else (except for the fact that in my example, you would reach your destination prior to falling asleep and that therefore never would’ve fallen asleep and had an accident). The point is that just equating AGW to % destructiveness based on the theoretical increase in wind speed is a major oversimplification.
If you want to relate that analogy to New Orleans: it would be more like knowing full-well you were likely to have a serious accident on your trip but not bothering to tell the wife and kids to put their seat belts on until right before the accident when it’s too late, then being unable to administer life-saving first aid to your dying wife and children because you threw your kit out the window a few minutes earlier (you counted on the ambulance and hospital to fix everything).
I never said there were advantages to higher wind speeds in hurricanes, simply that it seems to me that significant AGW would affect more than wind speeds. For example: which would be worse: (1) Category 4 Katrina hitting landfall just east of New Orleans, (2) a Category 5 Katrina never making landfall, (3) a Category 5 Katrina hitting the FL panhandle or (4) a Category 5 Katrina hitting southern TX? Case #2 is obviously the best, and I would argue that Cases #3 and #4 would have far less devastating effects than what #1 did. So the worst case scenario in my example is actually the weakest hurricane. And it seems to me that AGW, if it is that real and significant to hurricane wind speeds and SSTs, could produce significant changes in hurricane paths. That could be generally good (less landfalls), generally bad (more landfalls), or simply a case-by-case basis. I’m no hurricane expert, and I don’t know if there’s anything out there on the relation of hurricane paths to GW/AGW/SSTs/etc. I just find it hard to believe that the path would’ve been identical in a significantly warmer (or cooler) world. If – and it’s a big if – we get higher wind speeds but far fewer landfalls, couldn’t that be a good thing? Obviously, higher wind speeds and the same/more landfalls is a bad thing. Is there any such modeling or studies out there on the subject?
Roger,
In your article [Pielke et al. (2005), Hurricanes and Global Warming], it states:
… “it is exceedingly unlikely that scientists will identify large changes in historical storm behavior that have significant societal implications, though scientists may identify discernable changes in storm behavior.” …
http://sciencepolicy.colorado.edu/admin/publication_files/resourse-1766-2005.36.pdf
Earlier, I asked realclimate if the average size (diameter) of hurricanes is increasing? I think an increase in the size of hurricanes would be a change which would have significant societal implications. Do you agree?
Another point made in your article: “The views expressed are those of the authors, and for the four co-authors employed by the U.S. government, do not necessarily represent those of the National Oceanic and Atmospheric Administration.” And earlier in realclimate you wrote: … “I doubt that NOAA has a policy on climate change, and I wouldn’t co-author a paper whose publication required approval by a government official.”
For me, will you ask your co-authors with NOAA if the article was discussed and given a go-ahead by those up their chain of command?
Thanks.
I appreciate your thoughtful and science-based debunking of obfuscators like Steven Milloy. His work strikes me as dangerous and pernicious in the extreme. Bravo to the scientists on this site who are taking on the corporate apologist who themselves masquerade as debunkers of junk science in order to push their political and economic agendas.
RE response in #58, I know DAT was way off scientifically. My main question had to do with the large hailstones that actually came down during Emily. I wouldn’t be asking, but our weatherman said he had never heard of that happening before, so I was wondering if they might at all be related to any GW phenomenon, or perhaps contrary to GW science, or just normal.
Re #66: Lynn, this is more of a weather question than a climate question, so I’ll take a stab at it based on my sole qualification as a former resident of tornado alley. Generally, there are numerous reports of tornados and thunderstorm activity associated with hurricanes. Hail is intimately associated with thunderstorm activity. What surprises me is that your weatherman would have been surprised by this. Is thunderstorm-associated hail (actually the only kind I’m aware of) not common near Brownsville? I wonder if he was talking about hail closer to the center of Emily, which for all I know might be unusual. My memory of basic meteorology is that thunderstorms require hot and cold air masses to form, which I suppose would cause hail to not be expected very near the eye of a hurricane even if the hurricane was pushing into a cold air mass. Finally, I doubt that the size of the hailstones is very meaningful since large hail could be expected from energetic thunderstorms and hurricane-associated thunderstorms should be on the energetic side.
Have not heard it yet… Is good to say that GW has no or very little effect on Hurricanes, but would appreciate an explanation….. That would be better. Contradicting simple atmospheric physics demands no less; like world wide increase in water vapour density triggered by greater SST’s and plain warmer air causations. Proposing it has no effect on cyclones, rings like a slogan not science. An explanation aside from vague unimpressive assertions would be appreciated. Hurricanes are not that dissimilar to cyclones (they die as typical cyclones), so lets say that you are not an expert on hurricanes, then you may have an opinion on lesser meteorological entities? The question is simple: does warmer moist air create stronger cyclones? I rather think the answer is positive, and that answers about GW questions are simple, just like the meaning of glaciers disappearing everywhere.
Pat- (#64)
1. A change in the size of storms, just like changes in frequency or various measures of intensity could certainly have societal implications. The question is whether we can observe the effects of changes in the behavior of storms in the impacts record (e.g., we do see pronounced differences in impacts according to the phase of ENSO, El Nino/La Nina). So far, we haven’t seen evidence of significant changes in the impacts record. We may see them in the future, but not yet.
2. I can tell you definatively that our paper was not subject to any approval up the “chain of command.” Though we should be so lucky that the views expressed would be reflected as official policies ;-) Again, our paper is consistent with Emanuel (2005) and Trenberth (2005) and a new study soon to be published. The “debate” over this issue exists only in media statements and advocacy efforts, and not in the peer-reviewed literature. Clearly, Gray and Trenberth, for example, have different expectations about what future research will show, but until that research is conducted and published, I think our paper is an accurate assessment of the current state of knowledge on this topic as reflected in the literature. Sooner or later it will be out of date, as new work is published, and then it’ll be time for an updated assessment.
1. Francis (2004) and Katrina (2005) were unusually large, right? Does anyone know of plans to study what goes into determining the diameter of hurricanes, and if size has increased on average recently? We won’t see any evidence of change unless we study this.
2. I still think some kind of discussion and nod of approval from the higher ups had to be given, or else it could be that the higher ups concluded that the authors involved in your article have a record indicating they’d lean toward a position of no evidence shown in the historical record for a connection between hurricanes and global warming.
Pat (#70)-
You’ve mischaracterized our paper. We cite Emanuel (2005) and describe it as suggestive of evidence of a connection of hurricanes and global warming.
This may be totally off, since I know squat about climate science & the high-powered statistical programs used, but when there is a dearth of data in the social sciences (not enough to give low enough p values on correlations & regressions, simply due to small numbers of data, which is sometimes due to loading in too many control variables), we sometimes turn to chi-square & log-linear analysis to see if actual data reveal patterns incongruent with expected patterns.
In this vein, could, say, log-linear analysis be used to see if there were a lack of match between the natural decadal patterns (since we have a good idea about them) and the actual patterns now occurring currently (since we have the actual data on them).
Forgive me if I’m way off on this.
Stefan-
I have just come across your comments above.
Re: #47 – I can’t read German (I wish I could!), so if you’d be kind enough to email to me a translation of the quote, I’d be happy to respond. I did speak to the reporter for quite a while, so I’d be disappointed to be misquoted, but it happens.
[Response: Headline (issue 36, p. 124): “Roller-coaster of monster-storms – number and strength of hurricanes is increasing, but this has nothing to do with global warming”. The text says:
(This is my imperfect translation. The first sentence, before the direct quote, uses a tense which I think cannot be easily put into english, which makes it an indirect quote from you, rather than a statement of the journalist.)
This indirect quote, picked up in the headline, is plain wrong, as you will agree. At most one could say we are not certain whether or not hurricane activity is linked to global warming. The direct quote is technically correct in that nothing is ever definitely “proven” in the Earth sciences. But one should be clear that there is reasonably good evidence for such a connection, both from models and data (see the two figures in our post above). -stefan ]
Re: #49 – Emanuel’s point and my point are one and the same — Simply put, his PDI metric is not a measure of “destructiveness”. While there may indeed be changes going on in the characteristics of storms themselves, this does not mean that those changes are at all meaningful from a policy perspective. Think of it this way — if there is no chance of observing the effects of changes in PDI in the global impacts record for 50 years (Emanuel’s estimate) and over that same time period we expect damage to increase in real terms by up to a factor of 32 (a real doubling in damages every 10-15 years), then I think that it is safe (and also responsible) to assert that over that time period the only policies that can have a discernable effect on tropical cyclone damage around the world will necessarily be adaptive.
[Response: Emanuel’s estimate means that globally, the power of hurricanes has increased by ~75% while tropical SST has increased by ~0.5 ºC in the hurricane season, and these two are linked by physical theory, models and by statistical correlation. This remains a real increase and there is no reason to assume that it is not linked to a similar increase in damage, even if this cannot be proven from data because they are too noisy. Bad signal/noise ratio does not mean there is no signal – as Emanuel shows in his example calculation, bad signal/noise is expected here despite a large signal, simply because the data are very noisy. From current understanding, it is reasonable to assume that further warming will contribute to a further increase in hurricane damage, and not by a small amount. In my opinion, this is one of many reasons why it is wise to stop the warming before it goes too far. Putting more people and values into harms way by settling on a highly vulnerable coastline is also likely to increase hurricane damage. I think it is a pointless discussion what is worse – neither adding a lot more greenhouse gases to the atmosphere, nor allowing a lot of development along such a coast without proper hurricane protection measures, seem wise policies to me. -stefan]
RE#70 – http://hurricanes.noaa.gov/prepare/structure.htm states, “Typical hurricanes are about 300 miles wide although they can vary considerably…size is not necessarily an indication of hurricane intensity. Hurricane Andrew (1992), the most devastating hurricane of this century, was a relatively small hurricane.”
Frances was huge, but I can’t find a number. I’ve found links saying “the size of Texas,” which would put it in the 700 mile range. Katrina was about 500 miles wide at some point. By comparison, Andrew was just 60 miles wide.
I think it’s worth noting that Katrina was a Category 1 when it struck south Florida. If she had veered northward at that point, we wouldn’t be talking so much about hurricanes right now. Katrina would’ve been a relatively minor Florida hurricane and probably mostly forgotten.
Re: #66
Thunderstorms in hurricane environments typically don’t have strong updrafts. The temperature profile is close to moist adiabatic and, as such, there’s little convective available potential energy. Also, since the environments are generally warm through the depth of the hurricane, there’s a long distance between the freezing level and the ground. Hail is generally rare with hurricanes and, in fact, so is lightning (in comparison with mid-latitude continental thunderstorms.)
Re: #70 (2)
I’m not one of Roger’s NOAA co-authors, but I am a NOAA author of other papers. I’ve never had to get a “nod of approval” from higher-ups for the content of any paper I’ve published, even though some of them have questioned NOAA policies and/or future directions.
Re 71, Roger, I don’t think I mischaracterized your paper. I’ll take another look. BTW, did your co-authors spend government time doing work on the paper?
Re 73, thanks for the info. When Katrina struck south Florida, was it nearly as wide as when it hit Louisianna/Mississippi? Did Katrina come ashore with widespread heavy rain like Francis?
Re 75, Did you use government time to work on your papers? When (date, yrs) did you work on these papers. Did you include disclaimers on the papers? I think the subject of global warming/climate change is a whole different ball game.
I would like to comment on the argument in the initial posting in this thread that one can compare the importance of decadal variability in the Atlantic and greenhouse gas increases for hurricanes by simply comparing the increases in SSTs that they cause. (This is also a continuation of my comment in #40). Suppose for the sake of argument that the Atlantic internal variations are uncorrelated with the Indo-pacific, but that the latter, becuase it provides the deepest, most intense convection on average, controls the tropical tropospheric temperatures above the boundary layer (which we think of as being uniform horizontally). Then, when the internal variability of the Atlantic creates warm SSTs, it also creates a more unstable troposphere. In contrast, greenhouse gases warm both the Atlantic and the Indo pacific, and therefore, they warm both the Atlantic SSTs and the tropical troposphere, hopefully, with no change in stability to lowest order. (By the same argument, ENSO, by warming the troposphere, with only small effects in Atlantic SSTs, would stabilize the Atlantic.) Whether one thinks about buoyancy in this simple way, or one uses the entropy-base arguments for potential intensity of hurricanes, the sense is the same — internal variability should be more important for hurricanes than global warming of comparable SST amplitude.
[Response: Isaac, I don’t think anyone is claiming that SSTs alone are determinant (since as you correctly point out, tropospheric temperatures and the stability of the atmospheric profile are obviously important). However, to the extent that SSTs are influential it makes sense to assess the possible contribution of a global warming signal to the SST variability. Where your argument might break down is in the assumption that GW would cause uniform warming across the tropics. Given the much greater internal variability in the Pacific and the arguments of Cane et al, that doesn’t necessarily hold true, though it would be worth examining further. -gavin]
[Response: Dear Isaac, you’ve got a potentially valid point, that hinges on the question which we have not really clarified yet: how important are differences in temperature, and how important is the absolute value of temperature (which does come in due to all the non-linearities, e.g. Clausius-Clapeyron, which are not just small second-order effects). One preliminary test: if your argument is right, the correlation of SST and hurricane intensity should be much higher in the Atlantic, compared to the Pacific or the global (tropical) average. That’s not the case: see Kerry Emanuel’s second-to-last graph here. -stefan]
To Pat Neuman and Roger Pielke, while I cannot speak for NOAA, most of the national labs I am familiar with have an internal review policy for scientific papers which functions to insure quality (personal axes to grind are another story). Among other things that they look for is that no confidential material is being disclosed and certain other policies are adhered to. Here is one example http://www.ncnr.nist.gov/WERB.html. Here is a general discussion of the issue for national labs. It appears on first glance that NOAA relies on external peer review http://www.gao.gov/archive/1999/rc99099.pdf.
To Michael Jankowski, it appears that you never saw a headline that says man falls asleep, plows into tree at 70, dies. Please, you have constructed a stawman. There are enough people who die in high speed crashes when they fall asleep. At slower speeds they are only mangled.
Re 76:
Yes, it was on government time. I worked on the papers from 1992 to the present and continue to do so. No disclaimers were put on them. I have seen no difference in any climate-related papers. A quick glance at NCDC climate change papers shows no disclaimers either on the first several I looked at on the AMS archive and they were all supportive of significant climate change in recent years.
Re 78
After reviewing your linked refererences, I think many of the agencies or offices which are under the NOAA umbrella have their own internal review policy for scientific papers. This will likely be my final post here on this subject.
Re#78,
You continue to twist my analogy immensely, ignore my points, and argue against points I have never made.
I am sure we agree that, assuming all other things are equal, a hurricane with higher wind speeds is more destructive than a hurricane with lesser wind speeds.
But I am not sure that you agree, however, that all other things are not necessarily going to be equal. It seems reasonable to me that if AGW can significantly affect wind speeds in hurricanes and weather patterns throughout the world, then other factors determining the paths and destructiveness of hurricanes can also be affected by AGW. Some may be generally bad, some may be generally good, and some may be either good or bad on a case-by-case basis. Does this not seem reasonable to you?
Re: #73
Stefan- Thanks for these comments, we may simply agree to disagree on these points. Some additional thoughts.
1. You write, “This remains a real increase and there is no reason to assume that it is not linked to a similar increase in damage, even if this cannot be proven from data because they are too noisy.” If you can’t see a signal in a noisy record then by definition that signal is not large with respect to other relevant factors. If I can’t see a climate signal in the damage record, I agree with you that this does not mean that a signal does not exist, but it does lead one to conclude that the signal has not thus far been important from an impacts/policy perspective. Policy deals with (or at least should deal with) causing detectable effects, not addressing effects that cannot be detected. And again, Emanuel himselves estimates 50 years before any signal can even be detected.
[Response: “…then by definition that signal is not large with respect to other relevant factors”. Nope, not quite logical. That signal then is not large compared to the purely random, stochastic elements. Kerry’s example shows we could not remotely prove even a 75% increase from landfall data. If global warming has caused a 75% increase in hurricane damage, then this is definitely policy-relevant, no matter whether your damage statistics can attribute it or are just too noisy for that. I suspect you are thinking: if my statistics don’t show the effect, then it can’t be a serious real effect. But that’s not true. Remember the loaded dice: assume your friend rolls it 10 times and gets 4 sixes. Assume without the dice being loaded, he would have gotten 2 sixes. So loading the dice had a huge impact. Yet, by analysing the statistics of those 10 rolls, there’s no way you could have demonstrated that the dice is loaded. You couldn’t tell whether the 4 sixes occurred just by pure chance. But if for each six you were losing $ 1000 to your friend, this would be highly policy-relevant for you: you would stop playing with him if you had good reason to think his dice were loaded. You would stop playing if you had some other, indirect evidence that the dice are loaded (say, like Kerry Emanuel telling you after having x-rayed the dice). You then wouldn’t play on until your loss statistics unequivocally prove that the dice were loaded, unless you’re stupid. Do the maths: how many rounds would you have to keep playing with doubled odds for losing until you can prove at the 95% confidence level that you’re being cheated? Would you really keep playing all those rounds, brushing aside Kerry’s x-ray pictures with the argument: I don’t see statistically significant losses yet, my losses are large but still within the even larger bounds of random chance? I wouldn’t. – stefan]
2. You write, “I think it is a pointless discussion what is worse [climate change or societal vulnerability].” I’m surprised at this statement. Policy has to be based on an assessment of alternative courses of action, and justifications for action that lend themselves to political accountability. If one has a goal of addressing the future impacts of extreme events on human society, then the most important question from a policy perspective must be “What options do we have to act, and what are our expectations for the consequences of those alteratives?” Understanding the reasons underlying undersirable outcomes must be part of our research portfolio, lest we act on assumption, hunch and political predispositions, none of which, in my view are as effective a basis for action than science. I agree with you 100% that we should address both greenhouse gases and societal vulnerability, but we should not pretend (and yes, until it is backed by science it is pretending) that these issues are one and the same. They are not. There are good justifications for reducing greenhouse gas emissions, but having an discernable influence on future hurricane dmaages is not among them.
[Response: Of course they are not the same. But you have to be clear that the two effects multiply – they do not just add up. Above (#73) you juxtapose a 32-fold increase in vulnerability due to population etc., to a non-discernible climate effect – say a 2-fold increase in hurricane power (roughly as seen in Fig. 2 of the post above), suggesting the latter thus must be unimportant. That’s not the case. If your 32-times larger values are struck by a twice-as-damaging hurricane, the resulting damage would be 64-times larger. In numbers: assume that decades ago, the damage from hurricanes would have been ~3 billion $. Now, after 32-fold increase in values, the same hurricanes would have caused ~100 billion $. But if hurricane power had doubled, that would rise to ~200 billion $. So even though doubling sounds like a lot less than a 32-fold increase, this is highly deceptive. It still remains that in this theoretical example, half the $-damage is due to climatic effects, not population! And that remains true even if the data are too noisy to prove the twofold increase statistically – you still have to pay this bill, see the example with the dice above. Perhaps I’m biased because I’m a physicist not a statistician, but to me it is clear that physical evidence, like our physical understanding of how CO2 affects SST and how SST affects hurricanes, gives you warning of the loaded dice long before you can prove it from the damage statistics. Kerry, hand me those x-ray pictures – I’m not gonna keep playing this game, ignoring the physical evidence, waiting until the losses are so huge that even very noisy statistics become unequivocal! -stefan]
RE #75, thanks for your response. It was mid-July; we’re a subtropical area (lat 25.9N), and we were in the throws of El Cunecula (regular, but intensified summer heat wave) when Emily with hailstones came. So, I guess our weatherman was right, hailstone are unusual in our hurricanes.
RE #79, great, so glad at least some of my taxes are going in the right direction. I think all branches of government should be addressing GW in their own ways – afterall, it IS “global,” which includes the U.S. I keep writing my reps “What are you doing about this,” and they keep responding “The science isn’t in yet,” and I keep responding, “Yes, it is — more than enough for policy decisions.” Of course, they get their science from the WSJ. So, it is indeed good that a few in gov are actually into the science of GW. Just too bad those that make life & death decisions for us are not.
And before the contrarians who claim science should be kept clean of policy & politics jump on this, I agree with you – I want science to strive after objectivity, let the GW loaded dice fall where they may. This post is not violating that principle, rather it is critiquing the situation of policy & politics keeping clean of science, which they should NOT do, because that amounts to the blind leading the blind into very troubled waters.
As a social scientist, I am interested in this perspective. Our Enlightenment mindset of analysis (cutting up, compartmentalization, div of labor, pulling apart the clock to see how it works) – great for most science, not so great for all of life. I think in science & life we now need some synthesis, bringing together, putting the clock back together, connecting science to policy, bringing in more variables into science, an integration of the physcial sciences with the social/behavior sciences (perhaps with the humanities), the university with laypersons. United we stand, divided we may fall.
Part of this problematic Enlightenment mindset is the either/or focus. I liked Stefan’s response to #73 – Why can’t we do both, (1) better prepare for the effects of GW AND (2) try to reduce GW. The money we save from 2 could be plowed back into doing 1.
#73: “There are good justifications for reducing greenhouse gas emissions, but having an discernable influence on future hurricane dmaages is not among them. ”
Is a statement which defies logic, the very source of energy creating powerful hurricanes is greater moisture generated by more heat. I rather like an explanation, are hurricanes not heat dependent? Never mind destruction statistics, which vary according to hurricane paths. The implication of this statement reads like GW has no disercnable influence on hurricanes, like GW has nothing to do with warmer air.
Re: #84
It is possible that GW has a discernable influence on hurricanes and yet has no discernable influence on hurricane impacts (however measured). For an undersatnding of how this can be the case see this paper:
Pielke, Jr., R. A. and D. Sarewitz, 2005. Bringing Society back into the Climate Debate, Population and Environment, Volume 26, Number 3, pp. 255-268.
http://sciencepolicy.colorado.edu/admin/publication_files/resourse-1688-2005.25.pdf
And I agree with Lynn (#83), we can and should be doing both sorts of things, but to do so effectively means not conflating them as one and the same.
I think Roger Peilke Jr is asking a hybrid climate science / regulation question. On the scientific side climate modeling and physics show a potential increase in TC intensity that correlates to global warming, and now there is evidence that this is actually happening. This increase in TC intensity has not had a measurable socio-economic impact and possible further intensity increases might not have a large socio-economic impact. The question then is what kind of regulatory action should be based on this information. I think it’s a reasonable policy question.
There is some good info about TCs and climate change
Pew has a release
http://www.pewclimate.org/specialreports/katrina.cfm
MIT’s Robert Korty, a scientist specializing in climate and hurricanes wrote a short article
http://grist.org/comments/soapbox/2005/09/07/korty/index.html
RE #84-86, as mentioned before while natural causes may go into creating the largest part of hurricane intensity, we have to figure GW’s part as an add-on at the upper level – and it is that last bit of intensity that might do the most damage, more than the first incrememts (assuming the hurricane make landfall in a populated/built area). So while GW’s part may be small, its effects could be very large. I think that should be more our focus from a policy standpoint than, “GW’s part in hurricanes is small, and its effects may be negligible.”
As Roger points out, there are many other (even better) reasons for reducing GW. However, I think we can add “reducing hurricane intensity” to our list, even if it doesn’t have as much results as in reducing other GW effects (I’m thinking of droughts in Africa & glacier melt in the Himalayas, which could put 40% of Chinese & South Asians at severe water shortage risk – no water to drink, much less to irrigat crops).
I can see the problem: If we focus solely on hurricanes, and not the entire list of effects, we might lose some of the “intensity” in our pleas to get people to reduce GHGs. And some people even get confused and bored with the long list of GW harms, which just keeps getting longer.
Dear Michael, NO. Why? Because the first order cause of hurricanes is temperature differences between the surface and the upper atmosphere. My experience over 30 years of work in the natural sciences and the unnatural (engineering) is that balancing becomes important ONLY for secondary issues. The reason is that if you have an major cause, it takes a whole lot of secondary effects all going in one direction to restore the balance. Mostly the secondary effects go in both directions and balance themselves out. A good example is how the climate sensitivity to 2x greenhouse gas forcing has remained pretty much the same, with pretty much the same variation between estimates for a whole lot of years, model sophistication and generations of Moore’s law.
In a lot of ways your argument is like the student who shows up before the final with a passel of F’s on tests and asks if they can get an A in the course. The stars would have to align, the second coming be announced and then maybe.
A good reason why it is dangerous to take up your argument, is that it is equally likely that the secondary causes go against you. Plan for the worst. Pray for the best. Don’t trust economists. They have no clue about assigning costs.
Re 81 & 88. I think you two actually agree. One thing I was told in no uncertain terms in an undergrad conservation biology course was that analogies are very dangerous when you want to explain science in a politically-charged realm. The analogy breaks down and people tend to focus on parts that really aren’t most relevant to the point being made. For example, who says that driving a long time was the main contributor to falling asleep (maybe it’s the engine sound at 70 or some other thing)?
But I enjoy analogies and I think Michael’s was very persuasive in suggesting that going back and looking at what damage hurricanes like Katrina would have done without AGW or with more is an exercise that could be done for interests’ sake only — knowing with any certainty would require a lot more information (I’d say an impossible amount) about how AGW affected other aspects of the storm. On average higher windspeed = bad, but what was proposed was not to look at things on average but to look at a specific case. I don’t know if things could have been much worse than they were with Katrina, but if they could not, then it’s quite possible that more AGW would have lessened the damage done (if not the inherent destructive force). Disagreeing with the forensic approach does not necessitate an attitude against assigning blame in general or against policy prescriptions to deal with future likely impacts, so I don’t think Michael was advocating that people plan for the best.
I’ll stop putting words at your mouths now.
Re 85
Thanks for the paper
Since hurricane paths don’t always hit the same city, having an identical population, likewise buildings and economic infrastructure, it is incorrect to make hurricane financial impact assessments while looking at a long time spans, an analysis without nearly identical landfall scenarios doesn’t mean much. However, if you shorten the time span, compare it with nearly identical demographics at landfall, you may find a better trend aside from the ‘grass” talked about above. Forecasting future hurricane impacts is done with clarity provided in Hurricane historical met records as proposed at the top of this page. Intensity is a key issue, which will continue to increase along with GW, however Cat 3-4-5 hurricanes/typhoons are not confined to landfalls, but intensity with a few other met parameters readily show the impact of GW (as showed above) . It is perhaps a matter of time before someone demonstrates a clear and proper hurricane destruction cost comparison, I wonder if it is possible given very significant temporal urban demographic disparities.
Re: #82
Stefan- As much as I like to engage in abstract philosphical arguments about dice they are unnecessary when we have a good quantitative, scientifically rigorous, empirically-based literature on climate impacts. Some points to ponder and literature to consider:
1. The same hurricane damage database that is too small a sample to show a correlation with the Atlantic basin, annual total PDI is perfectly able to show a very strong statistical relationship with another climate index, the annual ASO Nino 3.4 temperature anomaly index. See this paper:
Katz, R.W., 2002: “Stochastic modeling of hurricane damage.” Journal of Applied Meteorology, 41, 754-762.
http://www.isse.ucar.edu/HP_rick/damage.pdf
So why is it that we can see the effects of ENSO in the damage record but not, for example, the annual Atlantic basin total PDI, given the exact same randomness in impacts? Perhaps one index has a stronger relationship with damage than another, no?
2. Similarly, we have been able to show a statistical relationship between various measures of precipitation and flood damage, even though the spatial and temporal area that experience floods are far smaller than the regions over which precipitation is measured. We do get the strongest relationships at the basin level, of course where precipitation is integrated, but we also see a correlation at larger scales. See this paper:
Pielke, Jr., R.A., and M.W. Downton, 2000: Precipitation and Damaging Floods: Trends in the United States, 1932-97. Journal of Climate, 13(20), 3625-3637.
http://sciencepolicy.colorado.edu/admin/publication_files/resourse-60-2000.11.pdf
Again, some metrics have a stronger relationship with damage than others. There are good physically based reasons for this.
3. Let’s completely set aside the issue of trends, Emanuel’s PDI (for the Atlantic basin, annual total) is not correlated with historical damage. The ASO Nino 3.4 index is strongly correlated with historical damage. This tells me that the ENSO index is a very good predictor of damage and the PDI simply is not (FYI, nor is the ACE index used by Gray et al.). There are however climate indicies that are related to damage, such as the decadal AMO, see this paper:
Landsea, C. L., R. A. Pielke, Jr., A. Mestas-Nuñez, and J. Knaff, 1999: Atlantic Basin Hurricanes: Indicies of Climate Changes. Climate Change, 42, 89-129.
http://sciencepolicy.colorado.edu/admin/publication_files/resourse-76-1999.10.pdf
So long as some climate indicies prove to be better than other indicies for explaining historical damage, it seems unsatifactory to claim than “randomness” can explain why one particular index has no correlation with damage. A different explanation to consider is that there is in fact not a strong relationship between PDI and damage and projections of future changes in the PDI provide little basis for asserting any effects on damages. But these are questions that can resolved through rigorous research (e.g, IPCC WGII-type research). They won’t be resolved through philosphical arguments! ;-)
Re 82
To stick to Stefan’s example. I think Roger would argue that there are other factors we have more control over, which would be decisive and statistically detectable, such as whether to play the game in the first place (not so much settlement in vulnerable areas), or whether to incur damages of $10 or $1000 (build better dikes).
You earlier gave the example of the European heat wave. Now, climate change so far, has loaded the dice there, as you suggest. Said loading is roughly equivalent to living 100 km further South, which has no discernable socio-economic impact. That’s not just because of noise hiding an underlying trend. It’s because adaptation to the higher temperature is such an important variable here that it can make the effect of the higher temperature on mortality / life expectancy go away entirely.
I don’t doubt that the heat wave caused excess death, particularly among already frail people, and when compared to a more moderate previous August,
but I also think that the small temperature increase so far over Europe’s had no net effect on mortality or life expectancy, not just because of the noise being more important than the small change, but because the noise interacts with the small change (a little bit more air conditioning here, a little bit less heating there, people moving for retirement to Spain etc.).
[Response: Heiko, I’d agree with your argument if climate change were smooth and gradual. Then, the 1 ºC warming we’ve had in Germany over the past century would have hardly any human health impact, and could indeed be compared to climate zones shifting by x km. It would be easy to adapt to. The reason why things turned out otherwise, and we got 30,000 casualties, is non-linearity and surprise. June temperatures in 2003 were 7 ºC above the long-term average – 5 standard deviations! And this came by surprise. Studies have shown that it is highly likely that global warming made a big contribution to this extreme event, but clearly it cannot be explained just by a smooth warming trend of 1 ºC per century. The more we interfere with the climate system, the more likely it is that it will respond in surprising and unforeseen ways, that we are poorly prepared for. It is not a linear system. Wally Broecker once said: the climate system is like an angry beast, and we are poking it with sticks.
By the way: our own infrastructure also induces big non-linearities. Suppose you have a system of dykes that protects against hurricanes of category three. Perhaps global warming might add 10% of energy to a hurricane, and it’s just that extra 10% that make the dykes break. That would hugely increase the damage – would you attribute that to the warming, or just to tough luck? -stefan]
Re#88 Eli,
Interesting – you say “NO,” that it is not reasonable to assume AGW could significantly affect hurricanes in other ways, then you start talking about “secondary issues.” If you are open to the possibility of secondary issues, whether or not they are good, bad, or “cancel out,” then your answer should have been “yes.”
I do like the “stars would have to align” portion of your analogy. Metaphorically speaking, that’s along the lines of what it takes for a hurricane to hit a particular spot. 0.5 deg more of warming (or cooling) doesn’t affect that?
Re#89 Steve,
Thanks.
I am intrigued by your site and have a couple of questions.
1) If, as you state, the EU has the following goal;
“This is a completely moot point. Kyoto or any other measures are not designed to make a difference within 5 years, which is obviously impossible given the timescales of the system. There is a world of difference, however, between having 2 ºC warming by 2100 (which is the policy goal of the European Union) or having 4 ºC or 5 ºC or even more warming then, and this is what we can decide.
What model is used to effect a 2 or 3ºC reduction in planned warming? And what model predicts a 4 ºC or 5 ºC rise by 2100? Given 95 years at fractional reductions, I would like to understand the algorithm.
2) In the Emanuel 2005 study, what was the nature of the PDI equation? Were hurricanes measured on size and intensity? Does the lack of comprehensive data from earlieir periods account for the VERY significant spike in PDI?
3) Finally, for the climate novice (that would be me) can you recommend some well vetted texts or articles to get a good grip on the current state of thought?
[Response: Check the science links sidebat: AIP for history; IPCC report for the current state – William]
Thanks, I appreciate a well written scientific site!
Re 92
My reading of the consensus is that it is very likely that climate change will be “smooth and gradual” over this century.
Not to dwell too much on semantics (on what smooth and gradual exactly means in the context of a hardly constant weather), as far as temperatures over Europe are concerned, my understanding is that night time winter temperatures will rise most, and summer maximum temperatures least and that otherwise the variability of temperature won’t be affected much.
You seem to be saying that the heat wave wasn’t just made half a degree C warmer by climate change (as per previous sentence a bit less than the global average), but several degrees warmer.
This is completely contrary to what I thought the consensus on climate science said.
I know that sudden, unpredictable climate change is an issue with a potential policy impact.
I think it’s small for the following reasons:
1. My understanding of climate science is that all truely serious scenarios of sudden climate change are quite unlikely.
There are known low likelihood dangers, such as a major meteorite strike, and the likelihood for these does feature in determining the scale of our mitigation efforts, even though the potential consequences could include the Earth becoming uninhabitable.
2. We’ve changed lots of other things, a little extra warming may not only cause sudden climate change (The Gulf Stream is a frequently cited example), but help to prevent it or attenuate its effects.
3. While climate may be a beast, in many respects it’s an exceedingly slow one with lots of inertia. As you’ll be well aware, surface air temperatures can easily change by 10 C within a day, for large bodies of ice to melt, or sea water to warm would take centuries. The kind of instantaneous climate change depicted in a certain rather well known movie is, as you also will be well aware of, physically impossible.
In other words, there are some (policy relevant in my opinion) limitations on how bad climate change can get, even when considering exceedingly low likelihood worst case scenarios.
If India gets the climate of Saudi Arabia, say, while climate models aren’t predicting that, it appears to be a pretty horrendous outcome.
However, it’s actually one that can be adapted too, eg through large water projects, the importation of grain and air conditioning.
Of course, if India becoming a desert was a certain outcome of climate change, or even a likely outcome, I would deem this to be of major policy relevance. It’s the combination of low likelihood and limitations in the impact that I think makes it a weak argument for influecing policy.
After all, if there was an even 1% chance of a meteorite wiping out humanity over the next century, we’d probably gladly spend 5% of GDP to prevent it. We aren’t willing to spend 5% of GDP on a 1 in a million chance of that, and even less willing to spend that kind of money on an unlikely event that is adaptable to without major loss of life, just major economic (and social disruption and ecosystem) cost, should it strike.
Climate change, unlike many other risks (such as new diseases, very large meteorites, nanobots etc..), does not have the potential to wipe out humanity, and “India turning into a desert” type worst case scenarios are, at least that’s my understanding of the current state of climate science, exceedingly unlikely.
Finally, regarding the dike example, well global warming would raise the likelihood of it getting breached. We could try to reduce that likelihood by making the dike cat 5 proof (at a cost), or by changing energy policy.
We could also use engineering works (it’s not only natural wet lands that can do this job) to break/divert part or all of the storm surge, so that even a cat 5 would only yield a storm surge at the dike that previously a cat 3 would have done, thereby reducing the non-linearity introduced by what happens when the dike does get breached.
Regarding the question of trends in hurricane intensity, there is another analysis that I think is worth mentioning.
In a recent paper Saunders and Lea (2005) reported that July wind anomalies (averaged between heights of 925 and 400 mbar) are well correlated with the US Accumulated Cyclone Energy (ACE) an index of intensity of land-falling hurricanes in the US for August through October. They stated that “the wind anomalies in these regions are indicative of atmospheric circulation patterns that either favour or hinder evolving hurricanes from reaching US shores.”
Additionally, and as yet unpublished, Mark Saunders (University College London, Benfield Hazard Research Center) presented a paper (Tropical Cyclones in a Warmer World) in October 2004 at a Workshop entitled “Natural Hazards in a Warmer World” Workshop for Under 35s Reinsurance Group.
http://forecast.mssl.ucl.ac.uk/shadow/docs/BHRCWorkshop14102004.pdf
In that workshop, reporting on trend analysis in ACE in the Atlantic and Pacific Oceans, Saunders concluded that “No significant change in ACE index or ACE index variance is observed anywhere, either over the full period or, for the U.S., using the recent 1978-2003 period compared to prior 26-year periods. The change in ACE index due to natural climate variability is everywhere an order of magnitude greater than that which may be attributable to trend in the historical record. Global warming has, at present, had little or no increasing impact on northern hemisphere tropical cyclone activity.”
Saunders, M.A., Lea, A.S., 2005. Seasonal prediction of hurricane activity reaching the coast of the United States. Nature 434, 1005-1008.
re: 2nd reponse to #82 above
Stefan-
I much appreciate the continuing dialogue.
Lets ground this discussion in science and not abstraction. The IPCC SAR suggested a 10% increase in MPI of TCs by 2050. Knutson and Tuleya (2004) suggest a smaller increase, 5% by 2080. There are no studies that I am aware of that provide a clear picture on frequency (see Pielke et al. 2005 for a review). Emanuel (2005) offers no projections.
I am aware of no reason in the scientific literature to believe that the destructive power of hurricanes will double, as you suggest, over the next 50 years. Were that to be the case it would be unavoidably evident in the damage record. That is an extremely large signal (e.g., of similar magnitude to what we can easily observe in the damage record between warm and cold phases of ENSO). If science supports such claims I will readily accept them. Right now the conclusions of the IPCC lead one to conclude that the effects of societal changes will exceed climate changes by a ratio of 22 to 1 to 60 to 1. If we were instead to use the more recent work of Knutson and Tuleya, then these ratios would be something more like 44 to 1 to 120 to 1. What this sensitivity analysis tells you is where, from the standpoint of action, you have the greatest ability to affect future impacts. See Figure 6 in this paper:
http://sciencepolicy.colorado.edu/admin/publication_files/resourse-1688-2005.25.pdf
So which conclusions of the IPCC do you reject? The effects of climate change on hurricanes? Trajectories of population and development? It seems to me that, logically, if one accepts the views presented in the IPCC, then one also has to accept the analysis I have presented here.
Don’t get me wrong, we should be taking more effective actions to decarbonize the global energy system. But repeated claims that such actions will materially affect disaster losses in coming decades are simply inconsistent with current scientific understandings. I enjoy discussions about loaded dice as much as the next person, but when we have rigorous research available on climate impacts, it seems to me that we should focus our discussions there.
[Response: Roger, I feel condescension and insinuations about my arguments not being grounded in science are inappropriate. I was using the dice analogy as an everyday and non-political example, since such analogies can often help to understand some of the basic concepts. But my argument is essentially based on Kerry Emanuel’s data (which of course post-date the last IPCC report), and I was talking about the trends observed by him for the past, not about future projections. Kerry’s study has found that the power dissipation index has increased globally by 75% and in the Atlantic it has doubled over the past decades, that’s why I was using those numbers. If those numbers are correct, and the PDI is closely related to potential destructiveness (see our footnote to the post above), then there is no way your ratios above could be correct. Any increase in societal vulnerability, no matter how large, would be multiplied by 2 due to the increase in storm power, so the latter would always be the larger contribution to the damage. We can debate how robust Kerry’s data are, they are quite new and surely there will be further analysis of the observed trends – but I think we can say at this stage that Kerry’s data directly contradict your conclusion about the small contribution of climatic trends to the damage. -stefan]
Not that I am a big fan of economic impacts, I would suggest looking at Camille vs Katrina, or likewise hurricane landfalls as far as truer comparisons, and of indirect evidence proving Global Warming. Much more interesting would be discussion of hurricane diameters for instance Andrew vs Katrina and the climate creating both of them. Those claiming that predictions of greater damage to come are exagerated, should look no further than Katrina’s impact, heard is about $200 billion US now, certainly the biggest one in history despite other monster hurricanes in the past.
Re # 88 Michael, perhaps I was not clear enough. If you have a primary cause that is responsible for 50% of an effect, and three causes that are each responsible for 13%, each, all three secondary causes would have to break in the same direction to compensate for a change of the same magnitude in the primary cause. Admittedly this is a simplification, but it is a useful thing to keep in mind for most yes but arguments. A lot of people miss this point, but it is also why simple models that isolate primary causes work pretty well.
The problem with making economic arguments about disaster impacts is that a very small number of disasters account for most of the costs. Katrina going 30 miles in one direction can change the cost of storm damage in 2005 by an order of magnitude without changing the intensity of the storm at all. On the other hand, if nature increases the intensity of large hurricanes in the South Atlantic and Carribean by a factor of 50%, sooner or later one of them is going to cause massive damage. Worse, there is some probability that a Katrina or a Katrina^1.5 would precipitate recession or worse. How would you account for that cost? Sort of like playing with Dirty Harry, just how lucky do you feel?