Over at Prometheus, Roger Pielke, Jr., has an interesting post taking Kevin Trenberth to task for discussing how global warming may have increased Katrina's total rainfall and thus caused direct damage to New Orleans. Pielke doesn't think Trenberth can justifiably say this, although previously (I forget the exact link) I seem to recall that folks at RealClimate had defended Trenberth's back-of-the-envelope calculation.
I don't know enough at this point to have an opinion about the validity of what Trenberth said. However, irrespective of Trenberth's claim, I do wonder whether or not it will be possible, at some point in the future, to directly attribute some fraction of a single hurricane's damage to global warming.
We all know we cannot say GW caused a specific storm (that's where Time magazine went overboard in its recent cover story). And in terms of GW increasing storm intensity, that would be an effect that could only be measured as an average increase. You couldn't detect such a change in any single storm.
But just suppose, for the sake of argument, that we find ourselves in a world where sea levels are 1 foot higher due to global warming. Could we not then say of a hurricane's storm surge that global warming had added a foot to it? I have a hard time seeing why not, but I am open to argument on this question.
Please excuse me for being a standard deviation or two off topic, but perhaps you can indulge me under the "hypothetical scenario" heading. Given the roles of increased temperatures and higher humidity in hurricane formation and other meteorological effects, has there been any research or speculation on what fuel cell-powered vehicles, should they become mass produced, will do to the atmosphere, as their by-product would be water vapor? It seems to me that with tens or hundreds of millions of vehicles being a new and significant source of water vapor being introduced into the atmosphere, the dynamics of global warming might be in for yet another shift.
The problem comes, I think, in trying to determine what the effects of a given storm would have been absent global warming.
"Given the roles of increased temperatures and higher humidity in hurricane formation and other meteorological effects, has there been any research or speculation on what fuel cell-powered vehicles, should they become mass produced, will do to the atmosphere, as their by-product would be water vapor? "
Water vapor : atmospheric residence time of days.
CO2: atmospheric residence time of 50-200 years.
That also ignores the fact that automobiles are already releasing large amounts of water vapour into the atmosphere. For every molecule of fuel burned, you get x amount of C02, and x+1 H2O molecules, where x is the amount of carbon in the fuel. So a single molecule of Octane (C8H18) would produce 8 carbon dioxide molecules and 9 water molecules.
Also, the content of water vapor in the atmosphere is dependant on the temperature. Cars operating in a Canadian winter will release condensed water (fog) which will freeze to the nearest surface as ice, and thus, nothing added to the atmosphere. Carbon Dioxide dosen't freeze at those temperatures though.
This thread is now totally off on a tangent. Just a warning: I may not publish off topic comments in the future.
Apparently hurricanes aren't interesting enough. (For Harris, in future you can easily get the answers to basic questions like that over at RealClimate or Wikipedia.)
Back on topic, Chris, I think you're right about being able to do an incremental attribution of the storm surge difference. It might still be hard to calculate the increment of damage from the additional water, though.
Regarding his take on Trenberth, RP Jr. is still smarting from being largely ignored 5-10 years ago when he first began advocating for a change in coastal land use policies to take into account the known natural variability of hurricanes (a point on which he was absolutely correct). To see the debate "moving on" to include consideration of AGW impacts is apparently a bit frustrating for him, but I think only because he chooses to characterize things that way.
Another aspect of this is that to the extent that climate scientists move more into the policy arena, this crowds into RP Jr.'s academic field.
Trenberth, IMHO, is absolutely correct to characterize the impact of GW on individual storms by way of a statistical attribution of damage. Even if GW were to become responsible for a majority of hurricane strength, it wouldm't be possible to say anything else. On the flip side, RP Jr. misses the fact that so long as we know that there is some amount of GW effect on hurricane strength, similarly it becomes unsupportable to ascribe any specific portion of that to natural cycles (for individual hurricanes). So without resorting to statistical attribution, we can't say much of anything.
Underlying all of this for RP Jr. is his belief that the upswing in strength is wholly ascribable to natural cycles. As you know, this is the NHC/Gray view as expressed in the new Bell+Chelliah paper. But as an amateur on the science, RP Jr. can't criticize Trenberth on that level and must resort to criticizing the way Trenberth expresses the impact.
Speaking of B+C, make sure to have a look at Peter Webster's comment near the end of the recent RC hurricane post. One has the feeling that B+C's conclusions aren't long for the world.
I'm dying to hear your impressions of the conference. Report soon, please!
Thanks for the comment. You'd have to ask him, but I'm not sure Roger would say that the "upswing in strength is wholly ascribable to natural cycles." I think he admits, as we all do, that at the very least there is now raging scientific debate about the GW influence on hurricanes.
I will not be reporting from the conference. I agreed that for journalistic purposes the event would be deemed "on background/off the record."
Just curious why you don't link to Prometheus on your Blog Roll. Is it because Roger has given quite valid criticism of your book? You are correct that it's a mis-characterization of his viewpoint to say he blames natural cycles entirely.
Chris - There was a piece in New Scientist recently about the problem of subsidence. I don't have it around any more, but for context on this question it would be worth tracking the story down and following up to see if the research supports the article's claims. Its basic point was that, for significant low-lying areas that have been intensely developed by humans (in other words, the precise places that are vulnerable in the scenario you sketch above) subsidence as a result of groundwater pumping and depletion of inflows of sediments to river deltas is much more significant than sea level rise. So Bangladesh, for example, is sinking far faster than sea level is rising, the article's author suggested.
As we've learned, blogging has its unique aspects. One is how some people who don't know you, can feel confident describing your motives and feelings and views. Sometimes it is OK simply to disagree about issues, it happens all the time in science an politics, and there is no need to make issues personal as Mr. Bloom seems to want to do.
Fortunately, my views on hurricanes and global warming do appear in the peer reviewed literature, and are quite clear for anyone with a good faith interest in understanding them. I do not belong to a "camp" as described by Mr. Bloom. I am not trained as a physical scientist and thus have no idependent basis for making judgments about whose views may eventually turn out to be more or less correct on this issue. There is in fact a debate, which will be played out in the peer reviewed literature over coming years.
But as I presented in Lamont, I am pretty sure that the debate - however it turns out - is not so significant from a policy perspective, that is even if one invents completely unrealistic scenarios for TC intensity increase, it would not dramatically change how we would evaluate alternative policies for hurricane damage mitigation - as I showed in my presentation and will be appearing soon in the literature.
Since I have repeatedly explained my views to Mr. Bloom at our own site, and he continues to completely mis-characterize my perspectives, I can only conclude that he is not interested in a good faith exchange of ideas and instead is up to something else, which is unfortunate. However, in my experience, the vast majority of people who comment on our blog are interested in exchanging ideas and learning from one another. Sorry to divert from your original question, which I will return to in a second.
My apologies, not my intent to derail the discussion.
On your main question the answer for damage in particular events is exactly the same as that which climate scientists correctly say about the occurence of particular events -- attribution of single events is not possible, but with enough events and with good enough data, one can make a statistical connection between changes in climate and damage.
This is exactly what we did in 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.
We were able to detect the effects of increasing precipitation in the US in the damage record. It is a small signal but statistically significant. Climate scientists are not able to attribute increasing preciptation to greenhouse gases, so we can't link the increase in damages to GHGs, but if such attribution were to be made, it would be a step in the direction of connecting aggregate damage trends with human caused climate change.
But even with such a linkage it would not allow one to say that, say, for the 1997 Red River flood, X% of damage resulted from GHGs, for the exact same reason that one cannot attribute the event itself to human caused climate change.
We've done a number of studies of this sort, as have others, in particular Stan Changnon.
what i think scares some folks isn't (only) the "it's gradually getting worse" responses in sea levels and storm intensities, it's the possibility that we may encounter other "tipping points" besides irreversible warming. these aren't only the global freeze possibilities, but qualitative changes in weather due to dramatic shifts in ocean currents and other mechanisms that might not be well understood.
this is after all and to borrow the words of the related exhibit at the Harvard Museum of Natural History, a "global experiment" we're conducting, and we don't know where we're headed.
I had just put Prometheus on the blogroll, even as you wrote that. I won't deny disagreements with Roger in the past, and maybe there will be more in the present, but I plan on trying to get along.
John: Good point on subsidence, that does indeed complicate matters.
Roger: Thanks for your comments. I'm still trying to wrap my mind around this, though. If global warming changes the earth in such a way that all storms are changed--i.e., they are *all* riding atop higher seas--would that not mean that global warming's effect was present in every storm? I agree that you might need statistics to actually detect the effect. But wouldn't it be safe to argue that it's there, and directly attributable to greenhouse gases (something that would not be safe to argue for intensity)?
I don't know if I'm making myself clear. In any case, sea level rise is not significant enough yet for this to become a matter of concern; this post and this comment are merely speculative and futuristic in nature.
Chris, you were perfectly clear. Roger was just avoiding your question. Of course it's the case that wetlands degradation, subsidence and sea level rise are all anthropogenic factors to which a portion of storm damage can be ascribed (where applicable), even though the calculation of the amounts of damage to ascribe to each would no doubt be tricky. I'm not so sure it's the case that sea level rise is a small factor even given the small amount we've had so far, BTW, although if we use Roger's example of a flood in the Red River valley it can be discounted completely.
The attribution of damage from increased storm intensity to global warming effects (aside from sea level rise) is a different discussion, of course, and I've already stated my view that Trenberth's approach is correct. I am completely confident that whenever Trenberth makes such a statement he places it into an appropriate statistical context. That part of the discussion is perhaps a little boring, though, which is probably why the NZ paper cut it out of the quote:
"This is not to say Katrina was due to global warming . . . There is an influence of global warming, something like an 8 per cent influence."
But giving Trenberth credit for that would have somewhat taken the punch out of RP Jr.'s argument.