Emanuel 2008:Global warming and hurricanes

A new peer-reviewed paper has been published in an American Meteorology Society journal that raises many more questions on the linkages between hurricane activity and global warming. Eric Berger at the Houston Chronicle (SciGuy) did the leg-work and is the first (and only) mainstream media outlet to report the findings of Emanuel et al. (2008) and get reaction from other scientists in the climate/hurricanes community. Andrew Revkin at the Old Gray Lady also has blogged reaction … NY Times DotEarth

Important update: 04/17 MIT press release New MIT study validates hurricane prediction Provides confirmation that climate change intensifies storms

Apparently the above blogs misconstrued, misinterpreted, or just plain flunked this lesson, because the above Press Release has some very different conclusions.

While the earlier study was based entirely on historical records of past hurricanes, showing nearly a doubling in the intensity of Atlantic storms over the last 30 years, the new work is purely theoretical.

“It strongly confirms, independently, the results in the Nature [2005] paper,” Emanuel said. “This is a completely independent analysis and comes up with very consistent results.”

Emanuel does discuss some of the uncertainties, which seem rather important (I think this summarizes about 100% of the contradictory views to North Atlantic + AGW causation):

There are several possibilities, Emanuel says. “The last 25 years’ increase may have little to do with global warming, or the models may have missed something about how nature responds to the increase in carbon dioxide.”


Another possibility is that the recent hurricane increase is related to the fast pace of increase in temperature. The computer models in this study, he explains, show what happens after the atmosphere has stabilized at new, much higher CO2 concentrations. “That’s very different from the process now, when it’s rapidly changing,” he says.

And the final conclusion:

In the many different computer runs with different models and different conditions, “the fact is, the results are all over the place,” Emanuel says. But that doesn’t mean that one can’t learn from them. And there is one conclusion that’s clearly not consistent with these results, he said: “The idea that there is no connection between hurricanes and global warming, that’s not supported,” he says.

…Emanuel employs a downscaling approach to the IPCC model scenarios using synthetic tropical cyclone seedlings to judge the impact of a warming world on future storm activity metrics (frequency, power dissipation). His method’s ability to use historical reanalysis products to fairly accurately reproduce past activity lends credibility to Emanuel’s technique, especially for future activity. Of course there are many caveats concerning model downscaling efforts using IPCC scenarios which have been discussed extensively at CA.

From SciGuy’s blog:

“The results surprised me,” Emanuel said of his work, adding that global warming may still play a role in raising the intensity of hurricanes but what that role is remains far from certain.”

In the new paper, Emanuel and his co-authors project activity nearly two centuries hence, finding an overall drop in the number of hurricanes around the world, while the intensity of storms in some regions does rise. For example, with Atlantic hurricanes, two of the seven model simulations Emanuel ran suggested that the overall intensity of storms would decline. Five models suggested a modest increase.

Dr. Curry from Georgia Tech University also is quoted,

“The issue probably will not be resolved until better computer models are developed…By publishing his new paper, and by the virtue of his high profile, Emanuel could be a catalyst for further agreement in the field of hurricanes and global warming …

Kerry Emanuel has provided a link on his homepage for the BAMS 2008 Article and while a little technical, the paper is a good primer on the current state of the “debate” and presents an even-handed examination of his findings. I encourage all to read it and post their own reviews for consumption by the gallery. Historically, Emanuel’s change in thinking represents one aspect of the evolution of climate science.

***Back in 1987, Time Mag reported upon one of the first articles in Nature to sound the alarm (Emanuel 1987).

a warmer climate could result in hurricanes packing up to 50% more destructive power. This could happen, he suggests, within 40 to 80 years, when some scientists think CO2 levels will have doubled and ocean temperatures will have increased by 2 degrees C to 3 degrees C.

Emanuel (2005) in Nature largely confirmed this hypothesis, almost 20 years later with the so-called climate shift of 1995 (Goldenberg et al. 2001 Science). Back in 1987, the computer technology, climate model development, and physical understanding probably was not there to adequately test the hurricane-warming hypothesis. In 2027, it will be interesting to see what “more work” has been done on the problem…

Flashback to July 31, 2005: Press Release MIT’s presser on Emanuel (2005)


  1. CharlieIliff
    Posted Apr 11, 2008 at 1:56 PM | Permalink

    There’s hope for science, yet. Hats off to a man who continues to examine his own work and publishes even surprising results.

  2. Sam Urbinto
    Posted Apr 11, 2008 at 2:36 PM | Permalink

    That’s actually a link to the (it says) Front page story from the Huston Chronicle.

    The blog is at


    Somebody ( IANVS ) there asks about responses from other leading hurrican climate scientists other than Judith. Cool! 🙂

  3. John Goetz
    Posted Apr 11, 2008 at 3:33 PM | Permalink

    Someone really should write Emanuel a letter and ask him to tone his conclusions down a bit.

  4. Andrew
    Posted Apr 11, 2008 at 4:43 PM | Permalink

    Interesting! Isn’t this the same guy who claimed to find increasing destructiveness in Atlantic Hurricanes in the first place?

    BTW, somewhat off topic, but have you heard about Briggs’ work on hurricane trends?

  5. Sam Urbinto
    Posted Apr 11, 2008 at 6:37 PM | Permalink

    Oh, now we have quite the bruahahaa between this story and the posts here. Might the tide be turning?


  6. Judith Curry
    Posted Apr 11, 2008 at 7:37 PM | Permalink

    here are some excerpts from the review i sent to eric berger,

    The motivation for Emanuel’s paper is that
    climate model simulations generally have resolution that is too coarse to
    actually resolve the tropical cyclones, in particular the intensity of the
    strongest storms. Some of the higher resolution simulations (e.g. Oouchi et al.
    2006; Bengtsson et al. 2007) do a credible job of simulating tropical cyclones:
    they show an overall increase in intensity, increased fraction of major
    hurricanes, and little change in total number of tropical cyclones. Personally,
    I think the Oouchi and Bengtsson papers are the most reliable analyses of what
    we might expect in a warmer climate, although these studies are by no means yet
    regarded as definitive.

    Because most climate models have resolution that is far too coarse to simulate
    tropical cyclones (such as the runs for the IPCC), there have been numerous
    attempts to identify proxy variables for tropical cyclones from the coarse model
    fields: the recent papers looking at wind shear, potential intensity, stability
    are examples of this. I have found these studies unconvincing for several
    reasons, but most notably that I don’t think the coarse resolution models do an
    adequate job of getting the tropical atmospheric dynamics correct to justify
    doing these proxy studies, and our understanding of genesis and intensification
    is too incomplete to credibly do such proxy studies.

    Emanuel’s new paper adds a different twist: he develops proxies by developing a
    large number of synthetic tracks from the coarse resolution model fields.
    Synthetic tracks are routinely used by catastrophe modelers in risk analysis,
    but haven’t been used previously in the context of climate model hurricane
    proxies. Emanuel’s production of synthetic tracks requires a very large number
    of assumptions, many of which are made with little credible defense . The
    results are compared with historical data and other model simulations (the
    comparisons with observations aren’t very convincing). Projections are then
    made out to 2200. There are so many assumptions in what has been done (at least
    a few of which that seem to me to be incorrect), that I don’t find the results
    to have much credibility.

    The significance of this paper in the debate on hurricanes and global warming is the change in emanuel’s position; emanuel is arguably the most prominent figure in this debate. The lesson to be learned is that one paper doesn’t really make a difference (we get the windshield wiper effect described by Andrew Revkin), and that assessments by a group of scientists are very important since individual scientists tend to place too much emphasis on their latest paper.

    The bottom line is that i have seen nothing published in the last 12 months that has moved the argument away from the the IPCC summary, which i quote below:

    “There is observational evidence for an increase of intense tropical cyclone
    activity in the North Atlantic since about 1970, correlated with increases of
    tropical sea surface temperatures. There are also suggestions of increased
    intense tropical cyclone activity in some other regions where concerns over data
    quality are greater. Multi-decadal variability and the quality of the tropical
    cyclone records prior to routine satellite observations in about 1970 complicate
    the detection of long-term trends in tropical cyclone activity. . . Based on a
    range of models, it is likely that future tropical cyclones (typhoons and
    hurricanes) will become more intense, with larger peak wind speeds and more
    heavy precipitation associated with ongoing increases of tropical SSTs. There is less
    confidence in projections of a global decrease in numbers of tropical cyclones.
    The apparent increase in the proportion of very intense storms since 1970 in
    some regions is much larger than simulated by current models for that period.”

  7. Andrew
    Posted Apr 11, 2008 at 7:49 PM | Permalink

    Based on a
    range of models, it is likely that future tropical cyclones (typhoons and
    hurricanes) will become more intense, with larger peak wind speeds and more
    heavy precipitation associated with ongoing increases of tropical SSTs.

    I’m sorry, but is anyone else bothered by the jump from “based on models” to “likely”? No? Must just be us deniers. Other than this, I think the IPCC’s assessment, which Judith qouted above, is fair. I think many would be surprised at the level of unsure-ness expressed here, given the certainty in public discourse.

  8. old construction worker
    Posted Apr 11, 2008 at 8:00 PM | Permalink

    Well Judy how about this.
    “The Forecasting Models Are Unreliable.Complex forecasting methods are only accurate when there is little uncertainty about the data and the situation (in this case: how the climate system works), and causal variables can be forecast accurately. These conditions do not apply to climate forecasting. For example, a simple model that projected the effects of Pacific Ocean currents (El Niño-Southern Oscillation) by extrapolating past data into the future made more accurate three-month forecasts than 11 complex models. Every model performed poorly when forecasting further ahead.

    The Forecasters Themselves Are Unreliable. Political considerations influence all stages of the IPCC process. For example, chapter by chapter drafts of the Fourth Assessment Report “Summary for Policymakers” were released months in advance of the full report, and the final version of the report was expressly written to reflect the language negotiated by political appointees to the IPCC. The conclusion of the audit is that there is no scientific forecast supporting the widespread belief in dangerous human-caused “global warming.” In fact, it has yet to be demonstrated that long-term forecasting of climate is possible.”


  9. Posted Apr 11, 2008 at 8:55 PM | Permalink

    Judy: I don’t understand why Kerry is changing his position and what caused him so much surprise, especially, as you point out, the assumptions necessary to make any credible conclusions from the UN IPCC scenarios. I like the novel technique of dropping nascent seedlings in the stream and watching them develop. I first saw results of his work in Greece at Elsner’s TC summit. He had a bunch of Category 5 tracks in the Arabian Sea/Persian Gulf, which was coincidently the same time as Gonu. So, I was immediately impressed with the technique.
    Is this perhaps more of a cumulative response to all the data issues, which have arisen AFTER the highly-publicized series of papers were published in 2005? What hasn’t been discussed on here much is Emanuel’s most recent J. Climate paper which deconvolves PDI into its components, using Kossin’s satellite derived intensities. This paper also is not too supportive of a link between historical hurricane intensity and warming SSTs.
    So, am I right in interpreting Kerry’s change of stance as the following: Since the future scenarios of hurricane intensity/frequency changes are not entirely robust or convincing enough, therefore the claims made about hurricane climatology changes in the past are therefore called into question?
    I confused about how this paper jives with any of the previous “slam dunk” papers that claim links between TCs and AGW like all of Elsner’s work, Greg Holland’s, Michael Mann’s, and the bunch of other cross-pollination papers.

  10. hswiseman
    Posted Apr 11, 2008 at 9:59 PM | Permalink

    I have been as snarky as anyone here about Emanuel’s findings, as I thought he was running a rigged game with a dubiously adjusted data set and self-fulfilling metrics. The triumphant certainty of his public comments was also pretty annoying. That being said, he obviously read Kossin. Knapp et. al. and went back to his research. Perhaps this was with a view towards falsifying Kossin (nothing wrong with that) by running some novel modeling. When the modeling didn’t support his own prior research in its entirety, he didn’t bury the work. He published. While I am loathe to congratulate people for doing exactly what they are supposed to do, I hereby retract anything I have previously posted that could be interpretated as an ad hominem or character slur. I too am man enough to repudiate a mistaken prior judgement. (I am sure that Dr. Emanuel is breathing a sigh of relief now that all is forgiven).

    It sounds like his approach is form of model-tuning where you keep turning the knobs until the backcast works. If multiple versions of the settings replicate the backcast, you run them all and look at ensembles. This seemed to make Dr. Curry a little unhappy judging from her commentary, but I am not sure it is any different from what is being done with the GCM’s. If you have enough factors and resolution built into the model to capture reality, and you manage to get the dials set right, (even if it is by experimentation rather than proving out each individual degree of freedom in the model), the result might be meaningful if the model as assembled has forecasting skill. Even in the best case scenario,however, the error bars get large, as any model initiation errors will compound quickly.

  11. Harry Eagar
    Posted Apr 11, 2008 at 11:38 PM | Permalink

    I thought that was a reasonably good (B-) piece of newspapering, except that it isn’t correct that hurricane data before 1970 aren’t very reliable. They aren’t even a little bit reliable. Sheesh.

  12. Janet S
    Posted Apr 11, 2008 at 11:56 PM | Permalink

    Perhaps brain fossilization is setting in.

  13. Philip Mulholland
    Posted Apr 12, 2008 at 12:07 AM | Permalink

    By Chris Mooney:-

    The 2007 Atlantic Hurricane Season: A Post-Mortem An Oddly Quiet Year for Hurricanes

  14. Posted Apr 12, 2008 at 5:46 AM | Permalink

    Dr. Curry, pray tell the difference between the many assumptions that Dr. Emmanuel has made and the many assumptions that go into the models the IPCC relies upon. Why are you so encouraged to hang your hat on the assumptions of the models that the IPCC chooses to rely upon? It really seems a bit of hypocrisy, does it not?

  15. Posted Apr 12, 2008 at 7:09 AM | Permalink

    Gaelin – The Emanuel technique is based on the IPCC models, so it includes all the assumptions therein except the hurricane-specific one that disturbances in the coarse models can be associated with details of hurricane characteristics. It also adds several additional assumptions, most importantly that changes in the climatology of initial disturbances are unimportant. Speaking for myself, the newest Emanuel work seems consistent with the IPCC report, and it’s a matter of personal taste whether one regards it as strong evidence that mostly supports what IPCC said or weak evidence that mostly supports what IPCC said, so no hypocrisy is present in Dr. Curry’s position.

  16. John Goetz
    Posted Apr 12, 2008 at 7:11 AM | Permalink

    The New York Times “Dot Earth” posted a piece on the paper. There is an almost overt admission in this post that the science is not settled.

  17. paminator
    Posted Apr 12, 2008 at 7:35 AM | Permalink

    The last two pages of this paper discuss the dramatic effects of assumptions made concerning moisture’s role in hurricane formation, intensity and duration. In fact, Figures 8b and 10 in the paper show the difference in hurricane frequency by basin for two different assumptions about water vapor. One shows a slight decrease, while the latter shows a strong increase. At this point it still looks to be an educated guess. It sure would be easier to model weather and climate if we didn’t have all that water vapor to keep track of…

  18. Francois Ouellette
    Posted Apr 12, 2008 at 9:06 AM | Permalink

    Tango, tango!

    Couldn’t we all see this coming? I find this all really pathetic. Climate science is a joke.

    Snipomatic time, Steve…

  19. Bruce
    Posted Apr 12, 2008 at 9:15 AM | Permalink


    There is observational evidence for an increase of intense tropical cyclone
    activity in the North Atlantic since about 1970

    Doesn’t the ACE index show 1971 to 1994 as “below normal”?


  20. hswiseman
    Posted Apr 12, 2008 at 10:48 AM | Permalink

    After reading the article, instead of merely the news report, my previous post is a pretty gross oversimplicifation of the Emaneul technique used. A lot of creative work went into the model before the tuning begins. There are plenty of nits to pick, but this is an interesting piece of work. Ignoring the African wave train and hanging fronts/stalled troughs in Western Atl. is a shortcoming that could have been overcome using a distribution model similar to the one used in track determination, or adding something like the Madden-Julian Oscillation into the development/intensity stew. I also think that PDO/AMO relationship impacts intensity more than ENSO.

  21. yorick
    Posted Apr 12, 2008 at 11:13 AM | Permalink

    Shorter IPCC response. “it’s our story, and we are sticking with it! We need the intense hurricanes for rhetorical reasons. Look at the cover art of AIT if you need proof of that. For that reason, no matter how ill-founded our assertion that Hurricane intesity increases are actually in the offing, we are sticking with it. If we didn’t, how can we possibly frighten the rubes?”

  22. Tony Edwards
    Posted Apr 12, 2008 at 11:37 AM | Permalink

    Over on Dot earth, I was rather amused/amazed to read the reply quoted from Dr Emmanuel, to whit “The models are telling us something quite different from what nature seems to be telling us. There are various interpretations possible, e.g. a) The big increase in hurricane power over the past 30 years or so may not have much to do with global warming, or b) The models are simply not faithfully reproducing what nature is doing. Hard to know which to believe yet.”
    Given the first sentence, I think it is obvious that both is the answer, he said it himself. Certainly (b) agrees with his initial statement. However, looking at
    http://www.climateaudit.org/?p=2471 shows that, while there is an apparent increase in the ACE values, I would hardly call it a big increase.

  23. Andrew
    Posted Apr 12, 2008 at 11:53 AM | Permalink

    Tony, Emmanuel is relying on his own work on some destructiveness index, not ACE. Search for:
    Emmanuel, 2005a: Increasing destructiveness of tropical cyclones
    over the past 30 years. Nature, 436, 686–688.

  24. Harry Eagar
    Posted Apr 12, 2008 at 11:57 AM | Permalink

    So, Phillip, I clickd over to Mooney’s piece, where he says the UK Met prediction was better than the others. The prediction was for 7 to 13.


    Given that the historical range starts, I imagine, around 7 (I don’t remember any year with so few myself) and the upper limit is around 25, that’s not a prediction. It’s a disguised way of saying ‘We have no idea.’

  25. Posted Apr 12, 2008 at 12:23 PM | Permalink

    #25 Mooney’s book seems out of date right about now. This is the type of subject that needs a thorough review by someone that has a background in climatology and tropical cyclone research. As a journalist with a vested interest in promoting global warming and hurricanes (by selling a book and giving paid lectures), Mooney exemplifies the prototypical liberal idealism, especially in his Daily Green blogs, which breathlessly attribute “weather” to climate change. Usually the posts end with the following disclaimer: of course one such event (tornado, hurricane, etc) on its own cannot be attributed to global warming, but it is consistent with what we expect to occur under a global warming scenario.

    This sets up an “informational cascade” which infects many scientific fields. Most journalists are flat out unable to grasp the subject of climate change because they are just not educated enough in the sciences to do anything else but follow the herd.

  26. Philip Mulholland
    Posted Apr 12, 2008 at 2:29 PM | Permalink

    Harry #25
    The source paper by Philip Klotzbach and William Gray, Summary of 2007 Atlantic Tropical Cyclone Activity and Verification of Author’s Seasonal and Monthly Forecasts is linked here

  27. Judith Curry
    Posted Apr 12, 2008 at 3:08 PM | Permalink

    Andy Revkin over at dotearth has an additional quote from emanuel. I guess my own perception of Emanuel’s stance on the issue comes from presentations at various meetings and personal discussions with him, as well as from his published papers. He had a high level of confidence in his PDI/SST analysis, with an implied extrapolation to the future. Based upon his recent paper, he is now stating there is much uncertainty in terms of what to expect in the future in terms of hurricane activity. I would say Emanuel’s most recent statements are closer to the IPCC than some of his previous statements.
    Emanuel changed his mind previously in summer 2005, originally he was coauthor on a Roger Pielke paper about hurricanes and global warming that said there was no link. His then current research caused him to change his mind.
    Scientists changing their mind based on new evidence is a good thing. But the media amplification of Emanuel’s stance following his 2005 paper makes this change of stance significant in the public dialogue on this issue.

    With regards to the specifics of emanuel’s study. Yes you take all the uncertainties inherent in climate model simulations, which are particularly high for hurricanes since they are small and poorly resolved by climate models, and then he makes a host of additional assumptions on top of the ones already made in the climate models. These additional assumptions increase the uncertainty of the analysis further.

    In terms of the journalists doing the most serious job of understanding the hurricane issue (which is very complicated), kudos go to chris mooney, eric berger, and andy revkin. Mooney’s book is still a good analysis, and his final chapter certainly leaves room for alot of uncertainty, so this book remains a good resource for people trying to figure this all out.

  28. bender
    Posted Apr 12, 2008 at 3:37 PM | Permalink

    I must say: it is satisying to see climate uncertainty regularly making it into the news and onto the agenda. It’s a problem, not a strategy.

  29. Judith Curry
    Posted Apr 12, 2008 at 3:38 PM | Permalink

    Ryan, a number of scientists have attempted a review such as you suggest.

    My mixing politics and science paper in BAMS (2006)

    Click to access Curry_BAMS87.pdf

    My congressional testimony last summer (2007)

    Click to access Curry_Energy.pdf

    Roger Pielke Jr. et al (2005)

    Click to access pielkeetalBAMS05.pdf

    Marshall Shepherd wrote a good summary paper:

    Click to access jms0701.pdf

    And there may be others? Any summary written by an individual scientist will bring their own personal scientific perspective, although i tried pretty hard in my testimony to present the range ideas (you can judge for yourself whether i was successful). If you were to take any other central character in this scientific debate, they probably would have written significantly different papers.

    Assessments are in principle better than reviews by a single person or self selected group. But an assessment of sorts was attempted as part of the CCSP Synthesis and Assessment report 3-2 Weather Extremes in a Changing Climate, which may not turn out so great based on comments i heard from some of the participants (there were some pretty contentious meetings so I hear), but we’ll see. The report should be available to the public sometime this summer.

    The Mooney book was really quite good in my opinion, and scientists on both sides of the debate seemed to like it. This book was more on the sociology of the science and scientists than a very detailed scientific discussion, but I think it is of substantial value in clarifying the issues to the public.

    The bottom line is that the hurricane/global warming issue is a rapidly evolving and not very mature area of scientific research, and personally i think we need to improve the theoretical foundations as well as the improve the data sets and models in order to make significant progress on this.

  30. Fran Manns
    Posted Apr 12, 2008 at 3:53 PM | Permalink

    Pleistocene history shows cooling compressed ‘climate zones’ toward the mid latitudes hence the 100 km long 100m high cemented seif dunes of the Saudi Desert. As a geologist, I infer from this that intense weather is the result of steep climate gradients. Winter beaches have high steep profiles and summer beaches are wide and fine. Google the Saudi desert for an awesome display of out-of-equilibrium dunes.

    Another aside: I observed a hurricane on Intellicast run the channel south of Cuba split the Gulf of Mexico and hit New Orleans directly. It stayed over warm water for the entire path; it never once made landfall before New Orleans; it reached category 5. And yet with all the time, and communication in the world, people refused to move. I call that immaturity, not bad science. It became an AGW myth.

  31. Posted Apr 12, 2008 at 5:05 PM | Permalink

    #30 Judy, I was not implying that such “reviews” had not been attempted. Of course I have read your lead-author BAMS article, and the previous go-rounds between Piekle Jr. and others just after the original 2005 highly-publicized papers. I will give you that Mooney’s book was a sociological masterpiece, just like his previous well-received tome, the Republican War On Science. I am not convinced he is an “honest broker”, but that is my biased opinion. The IPCC summary and its WMO cousin are the best that I could hope for, well balanced.

    The fact is that Emanuel’s SST/PDI link in the Atlantic is not a unique global phenomena. You can take the North Atlantic ACE/PDI and regress it upon SST (after the obligatory 1-3-4-3-1 or 1-2-1 smoothing, to increase correlation, remove high frequency) and “many” parts of the world light up with correlations exceeding r>0.9. This is simply a reflection of a slow warming trend during a period of enhanced El Nino / PDO phase roughly corresponding with the early 1970s. The SST off of New Zealand during austral winter hardly has much do with hurricane activity in the Atlantic — or does it? I am not sold on SST being a causal agent for the increased activity, but more of a co-factor or tracer, since MDR SST trends are also well correlated with SST trends over other parts of the globe. I have very strong evidence of such a “tracer” mechanism for SST that explains over 90% of the variance in total Northern Hemisphere ACE during the past 30 years, which I will present at the Orlando Tropical Conference.

  32. Judith Curry
    Posted Apr 12, 2008 at 6:32 PM | Permalink

    Ryan, I totally agree that the current SST story is too simplistic. I won’t be at the Orlando meeting (one of my graduate students will attend), i look forward to hearing more about your new work.

  33. Kenneth Fritsch
    Posted Apr 12, 2008 at 6:55 PM | Permalink

    After reading this paper [Emanuel et al. (2008)] and thinking about how it will be handled by the mainstream climate scientists, I think some of the statements here pointing to a potential change in their view on the matter of storm frequencies and intensities with SST are greatly exaggerated.

    The authors have made a valiant effort to avoid using any past (and possibly unreliable) historical TC data to calibrate their model. They have gone out of their way to make a dynamical model to predict 2200 TC intensities, frequencies and durations. Their effort does not, however, come without their having to resort to parameterization (see wind shear for example). To summarize this attempt, I will excerpt some material from the paper:

    In this excerpt the authors are comparing the 1981-2000 climate to that projected for the 2181-2200:

    We also calculated basin-averaged changes in sea surface temperature and potential intensity, defining the averaging areas to coincide roughly with conventional definitions of tropical cyclone main development regions. Curiously, there is no systematic correlation between changes in any of the metrics described in this section with changes in either sea surface temperature or potential intensity averaged over the individual basins. This contrasts sharply with the high correlation between variations in observed Atlantic tropical cyclone power dissipation and sea surface temperature, reported by Emanuel (2005a), and the relatively high correlation among year-to-year variability of power dissipation, sea surface temperature, and potential intensity in the simulations forced by reanalysis data in the years of 1980–2006. Deconvolving the various environmental factors responsible for the changes in the various tropical cyclone metrics described here is the subject of ongoing work and will be reported in the near future.

    In this excerpt the authors are comparing their model results (which were adequate for the NATL) to model simulated results for the 1980 to 2006 period in the four TC basin excluding the NATL:

    Encouraged by these results, we simulated 200 events in each of the other four regions in each of the years from 1980 to 2006 and compared them to the observed events. Somewhat to our surprise, there is essentially no correlation between the simulated and observed annual frequencies of tropical cyclones anywhere outside the North Atlantic. This is perhaps related to the fact that while tropical cyclone frequency is well correlated with sea surface temperature in the North Atlantic (Mann et al. 2007), there is no indication of such a correlation elsewhere.

    And finally this excerpt that makes a passing note to the Kossin reanalysis but makes no concessions to what this might imply for their model verification:

    The overall trend in simulated power dissipation agrees reasonably well with
    the trend deduced from best-track data in all basins except the eastern North Pacific, the only basin in which the best-track trend is downward. Over the 27-yr period, the simulated global power dissipation increases by 63%, versus an increase of 48% in the best-track data. Note, however, that outside the Atlantic, the simulated increases are not in agreement with the reanalysis of tropical cyclone power dissipation undertaken by Kossin et al. (2007), who found that power dissipation decreased everywhere, except in the North Atlantic, and, marginally, in the western North Pacific.

    Much evidence has been presented at CA and elsewhere that could refute the observations by Emanuel, but I doubt whether scientists in the Emanuel camp are ready to back-off their views of 1970 to present NATL TC occurrence and intensity and the correlations with SST – even though data back beyond 1970 would strongly indicate a decadal cyclical nature of TC frequencies and intensities that has nothing to do with SST trends. Emanuel et al. (2008) in this paper refer to the 1970 to present period TC trends without at the same place in the paper noting the cyclical nature of TC measures that could be detected by going beyond 1970 other than to note that that data is probably unreliable beyond that point. In fairness they do later note that TC measures can have decadal variations when they question (for economy of computer time) using the 20 year 1981-2000 and 2081-2200 periods for their model comparisons.

    A final insight can be made revealing the tenacity that we might expect from the authors to adhere to previous views, in my judgment, from the following excerpt:

    These results suggest potentially substantial changes in destructiveness of tropical cyclones as a consequence of global warming. But large model-to-model variability and natural multi-decadal variability within at least some of the models also suggests large uncertainty in such projections, reflecting the uncertainties of climate model projections in general and the influence of natural variability. Ideally, we would like to compare simulations driven by climate model data accumulated over periods long enough both to quantify and account for the effects of natural variability, but because such data are not currently available, this is left to future work.

  34. Judith Curry
    Posted Apr 12, 2008 at 7:02 PM | Permalink

    Kenneth, i doubt that Emanuel’s paper will change the minds of any of the scientists working on this problem. Nearly everyone agrees that future projections of hurricane activity are highly uncertain (the IPCC says this, WMO consensus statement said this, even i said this in my 2006 paper). Personally, I don’t think this paper is particularly newsworthy in itself, i regard it as an interesting intellectual exercise that raises more questions than it answers. But anything coming from Emanuel is deemed as newsworthy. And the angle taken by Berger (and also Revkin) is a healthy one, showing that scientists change their minds as they continue their scientific investigations. Although they make it sound like this is some sort of exceptional event, which it is not.

  35. Kenneth Fritsch
    Posted Apr 12, 2008 at 7:13 PM | Permalink

    Re: #32

    I am not convinced he is an “honest broker”, but that is my biased opinion. The IPCC summary and its WMO cousin are the best that I could hope for, well balanced.

    Ryan, I agree with you that after reading Mooney’s books and comments about the TC issue that his view is clouded with partisanship much as I agree with Judith Curry that Rush Limbaugh’s views are clouded with partisanship.

    On the other hand, I agree with neither you or Judith on the IPCC statement. The first part of it says nothing and the second part is word smithed to give conjecture the appearance scientific results.

  36. Andrew
    Posted Apr 12, 2008 at 7:19 PM | Permalink

    Although they make it sound like this is some sort of exceptional event, which it is not.

    This is because they see science as Kuhnian “normal science.” Papers and research that report “anomalies” are ignored, and papers that confirm or extend the current paradigm are the ones that are “important.” So naturally they are shocked when a paper deviates.

    Hat Tip, this post.

    BTW I was just in a book store (and my terrible memory can’t remember which) and while stopping to look at some of the books, I noticed Emanuel has a book. I wonder if he’ll release a new addition in which he changes any statements therein becuase of this? Well, I don’t know if he mentions GW at all in the book (I didn’t read it) but it would be hard to imagine him not doing so.

  37. Andrew
    Posted Apr 12, 2008 at 7:33 PM | Permalink

    Kenneth Fritsch, a person’s political views (their “partisanship”) shouldn’t matter, if they have something valuable to say, they contribute to public discourse by doing so. Of course, I would sense that someone writing a book on how a certain political party is “at war” with “science” is not actually adding to the debate, but probably writing a garbage political diatribe. On the other hand, if you feel that what Limbaugh says about AGW is just partisanship clouding his judgment, which would otherwise be enthralled with the idea, perhaps you should take that up with his climatologist? I suspect that he is better informed than you or Judith give him credit for.

  38. Judith Curry
    Posted Apr 12, 2008 at 7:55 PM | Permalink

    Again, this Emanuel’s paper is not an anomaly in the global warming/hurricane debate. There have been scientists like bill gray and jim obrien that say there is no link. Landsea says there is no credible observed link, and climate models are uncertain. And so on. So Emanuel’s paper is interesting because it represents a shift in his own thinking, bring it closer in line to the IPCC and WMO statements that include a fairly large level of uncertainty.

  39. Geoff Sherrington
    Posted Apr 12, 2008 at 8:11 PM | Permalink

    Re # 28 Judith Curry

    I know next to nothing about USA hurricanes, but I do know a little about etiquette in science.

    When you again take this line –

    I would say Emanuel’s most recent statements are closer to the IPCC than some of his previous statements.

    – as you often do, you are being discourteous to those who came to similar research conclusions before the IPCC was even a gleam in the eye of the UN.

    Surely the earliest primary author should be quoted, rather than attempting to hold up the IPCC as a gold standard when many of us know about pyrite.

  40. Kenneth Fritsch
    Posted Apr 12, 2008 at 8:35 PM | Permalink

    Re #38

    On the other hand, if you feel that what Limbaugh says about AGW is just partisanship clouding his judgment,

    Andrew, I am making a distinction between the views of one making an issue partisan and a political person having legitimate views uninfluenced by partisanship. In either case persons can make valid observations about an issue, but those making the issue a partisan one are the ones that I scrutinize more (even though the partisan content of their statements are often rather transparent).

    Actually my biggest complaint about Mooney in “Storm” was his portrayal of Bill Gray as the archetypical empirical climatologists and characterizing him as the bombastic buffoon when in fact Gray is currently not even a very active member of the climate science community. Mooney used Emanuel as the dynamical counter part to Gray and was portrayed as quite and brilliant and whose theoretical writings on SST and TC intensity represented a paradigm shift for climate science. His choices of characters, while perhaps puzzling to anyone familiar with climate science, I think fitted Mooney’s purposes to a tee. He then fills the book with accounts of TC and hurricanes of recent (and increasing frequencies and intensities) vintages and all the resultant damages that are incurred while occasionally reminding the reader that the science may not be settled on hurricanes. Of course, who is the reader to believe a buffoon or a brilliant scientist?

  41. Judith Curry
    Posted Apr 12, 2008 at 8:44 PM | Permalink

    The whole point of an assessment is that it is not what one person says, it is a review of the literature by a large group of people that go through a very extensive review process. Yes, I know Steve M has gripes about the process wrt the hockey stick. But no one has complained about the hurricane part. Although Chris Landsea resigned from the IPCC in a huff in 2005, I havent heard that he objects in any way to the statement that the IPCC finally made regarding hurricanes.

    There has been a wide range of people’s opinions on this subject, a year or two ago I actually ranked a number of scientists on a scale from 0 (all of the recent increase in activity is global warming) to 10 (all of the recent increase in activity is natural variability), maybe someone can find this and we can discuss how/if the scientists have shifted. So whatever the current thinking is on the subject (something between 3 and 5, say), you can find a number of individual scientists who expressed opinions previously in that range.

    Considering that the IPCC fourth assessment report began in 2003 (before anyone was talking about this subject very much), i do not find very relevant any opinions expressed prior to 2005, prior to the onset of rather vigorous research activity on this topic. And anyone with a credible opinion on this subject has been reading the large number of papers published since 2005 on this topic (Chris Mooney, who devoted 2 years of his life to researching the topic and reading papers and interviewing scientists and actually wrote a book, can hardly be compared to Rush Limbaugh who has probably not read any of the scientific literature or interviewed any of the relevant scientists).

  42. steven mosher
    Posted Apr 12, 2008 at 8:48 PM | Permalink

    re 39. I liked his nifty idea of seeding tracks in. We used to do something similiar
    with large combat simulations ( obviously not with huricanes but tracks of other things)

    In any case the hurricane part of climate science appears most interesting because
    it’s still squishy and surprising.

  43. Posted Apr 12, 2008 at 9:15 PM | Permalink

    Re #36 Kenneth, you’ve got mail.

  44. Roger
    Posted Apr 12, 2008 at 9:36 PM | Permalink

    To be fair, Emanuel’s 2005 Nature paper was largely about addressing deficiencies in consistently quantifying hurricans and the risks of increased exposure of property and people to hurricanes generally.

    See also his 2006 Statement On The US Hurricane Problem, which said

    “As the Atlantic hurricane season gets underway, the possible influence of climate change on hurricane activity is receiving renewed attention. While the debate on this issue is of considerable scientific and societal interest and concern, it should in no event detract from the main hurricane problem facing the United States: the ever-growing concentration of population and wealth in vulnerable coastal regions….


    We are optimistic that continued research will eventually resolve much of the current controversy over the effect of climate change on hurricanes. But the more urgent problem of our lemming-like march to the sea requires immediate and sustained attention…….”

    The difference is Al Gore and the media cherry picked the 2005 paper for sensational bits but ignored the 2006 Statement.

  45. Geoff Sherrington
    Posted Apr 12, 2008 at 9:42 PM | Permalink

    Re # 42 Judith Curry

    The whole point of an assessment is that it is not what one person says, it is a review of the literature by a large group of people that go through a very extensive review process.

    Yes, about 30 bureacrats so far as I can make out for the IPCC.

    I’m not naive. I’ve written to review authors who failed to review, read verbatim from silenced dissenting authors, etc. While the review process might be very extensive, it also has large holes. Anyone who relies on it without study of the real data believes in the tooth fairy.


    “i do not find very relevant any opinions expressed prior to 2005”.

    Bye bye Boltzmann, Clausius, Beer, Lambert, Reynolds …….

  46. Andrey Levin
    Posted Apr 13, 2008 at 5:07 AM | Permalink

    Re#30, Judith Curry:

    The bottom line is that the hurricane/global warming issue is a rapidly evolving and not very mature area of scientific research, and personally i think we need to improve the theoretical foundations as well as the improve the data sets and models in order to make significant progress on this.

    The bottom line is that people should stop to build cardboard McMantions in hurricane prone zones. All private houses in Israel (by far country not as rich as US) are built from reinforced concrete, for example. With mandated solar water heating and hardened bomb shelters.

  47. Francois Ouellette
    Posted Apr 13, 2008 at 8:48 AM | Permalink

    Dr. Curry tries to paint a picture of science that is about 400 years old. Yes, Aristotelian science is back! We all thought that Galileo had put an end to it, and shown that theories must agree with experimental measurements. But, no. Now we’re back to a bunch of people deciding among themselves that we should believe what their beloved models say. Never mind that not one of those well-educated people can tell us exactly how many hurricanes we’re going to have this year, or even explain why we had so many in 2005, and so few in 2007. Still, we should believe them when they tell us how many there will be in 2200! Now why should we believe them? Because they all have Ph.D.’s and publish in peer-reviewed journals. Just think about it for 2 seconds. That is the essence of Dr. Curry’s argument.

    Sounds simple: theory, prediction, test, confirmation. If you can’t do that, all the degrees and the journals in the world won’t help. It’s not science.

  48. Posted Apr 13, 2008 at 9:22 AM | Permalink

    January 1998 Paper by Emanuel, Gray, Landsea, Holland, Webster et al

    Historical perspective.

  49. Posted Apr 13, 2008 at 9:56 AM | Permalink

    Excerpts from the 1998 paper

    The very limited instrumental record makes extensive analyses of the natural variability of global tropical cyclone
    activities difficult in most of the tropical cyclone basins.

    A strong caveat must be placed on analysis of results from current GCM simulations of the “tropical-cyclone-like”
    vortices. Their realism, and hence prediction skill (and also that of “embedded” mesoscale models), is greatly limited by
    the coarse resolution of current GCMs and the failure to capture environmental factors that govern cyclone intensity.

    As pointed out by Holland (1981), the quality
    of the tropical cyclone databases can be highly variable.
    Different definitions, techniques, and observational
    approaches may produce errors and biases in
    these datasets that could have implications for the
    study of the natural variation of tropical cyclone activities
    and the detection of possible historical trends

    1) Tropical cyclones are currently devastatingly severe
    weather events.
    2) Human vulnerability to TCs is increasing because
    of increasing populations on tropical coasts.
    3) Tropical cyclone formation and intensity change
    are currently very difficult to predict.
    4) Costs of TC impacts are increasing because of increasing
    costs of infrastructure and increasing “responsibility”
    claims on private and public funds.
    5) The balance of evidence indicates that greenhouse
    gas emissions are producing climate change
    (Houghton et al. 1996).
    6) Concern about possible future changes in tropical
    cyclone activity relates to changes in (i) frequency
    of occurrence, (ii) area of occurrence, (iii) mean intensity,
    (iv) maximum intensity, and (v) rain and
    wind structure.

    1) How to predict TCs today: genesis, maximum intensity.
    2) How the environmental parameters that appear to
    be important for TC genesis will change.
    3) How the large-scale circulation features that appear
    Bulletin of the American Meteorological Society 23
    to be linked to TC climatology, especially the
    quasi-biennial oscillation (QBO) and El Niño–
    Southern Oscillation (ENSO), will change.
    4) How the upper-ocean thermal structure, which acts
    as the energy source for TC development, will

    It has been suggested that the Atlantic’s hyperactive
    hurricane seasons of 1995 (19 tropical storms, 11
    hurricanes, and 5 intense hurricanes) and 1996
    (13 tropical cyclones, 9 hurricanes, and 6 intense
    hurricanes) may be heralding a return to an active regime
    similar to that seen between the 1940s and the
    1960s (Landsea et al. 1996). Since the Atlantic hurricane
    activity observed during the 1970s and into the
    early 1990s was anomalously low compared with previous
    decades, a return to a more active regime is not

    The global cyclone frequency taken from the National
    Climatic Data Center Global Tropical Cyclone
    Data Set indicates that the number of tropical cyclones
    may have increased since 1970. However, this increase
    has arisen entirely from the more poorly observed regions
    of the Southern Hemisphere and the eastern
    North Pacific and cannot be differentiated from changing
    observing practices and slow, multidecadal oscillations
    in cyclone numbers.

    Since the production of the 1996 IPCC reports, our
    knowledge has advanced to permit the following summary.
    • There are no discernible global trends in tropical cyclone
    number, intensity, or location from historical
    data analyses.
    • Regional variability, which is very large, is being
    quantified slowly by a variety of methods.
    • Empirical methods do not have skill when applied
    to tropical cyclones in greenhouse conditions.
    • Global and mesoscale model-based predictions for
    tropical cyclones in greenhouse conditions have not
    yet demonstrated prediction skill.

    The IPCC “Science of Climate Change” report
    stated that “it is not possible to say whether the frequency,
    area of occurrence, time of occurrence, mean
    intensity or maximum intensity of tropical cyclones
    will change” (Houghton et al. 1996, p. 334). We believe
    that it is now possible to improve on this statement.
    In particular:
    • there is no evidence to suggest any major changes
    in the area or global location of tropical cyclone
    genesis in greenhouse conditions;
    • thermodynamic “upscaling” models seem to have
    some skill in predicting maximum potential intensity
    (MPI); and
    • these thermodynamic schemes predict an increase
    in MPI of 10%–20% for a doubled CO2 climate but
    the known omissions (ocean spray, momentum restriction,
    and possibly also surface to 300 hPa lapse
    rate changes) all act to reduce these increases.

    Acknowledgments. The process used to generate this state-ofthe-
    art review extended from June 1996 to March 1997. The 10
    members of the WMO/CAS/TMRP Committee (A. Henderson-
    Sellers, G. Berz, R. Elsberry, K. Emanuel, W. Gray, C. Landsea,
    G. Holland, J. Lighthill, S.-L. Shieh, P. Webster) submitted upto-
    date assessments. These were synthesized into a single paper
    by the rapporteur (Dr. H. Zhang) and the chairman (Professor A.
    Henderson-Sellers). This draft was circulated to all committee
    members and also reviewed by attendees at the ONR Symposium
    on Tropical Cyclones in December 1996. Eleven scientists (K.
    McGuffie, W. Gray, R. Elsberry, M. Lander, F. Wells, G. Holland,
    J. Evans, L. Avila, I. Ginis, C. Landsea, and R. Abbey) reviewed
    the document during a working session of the ONR symposium.
    The resulting final version was circulated to all the committee
    members for agreement. We are very grateful to all those who
    participated in this process.

  50. Kenneth Fritsch
    Posted Apr 13, 2008 at 11:59 AM | Permalink

    Re: #42

    (Chris Mooney, who devoted 2 years of his life to researching the topic and reading papers and interviewing scientists and actually wrote a book, can hardly be compared to Rush Limbaugh who has probably not read any of the scientific literature or interviewed any of the relevant scientists).

    Which means that Mooney should know better? Your description of Mooney’s mission makes him appear to be an individual doing IPCC type work with a human interest twist. Like the IPCC summary on TC activity, I think a lot of these writings need a discussion/analysis on the separate issues involved. Anyone of us can (and often do) make general statements giving our POVs which is nice to know, but, again, the devil is in the details.

  51. Andrew
    Posted Apr 13, 2008 at 12:36 PM | Permalink

    Well, I was going to reply to all this, and state my opinion that I am more than a little disturbed by the deep antipathy Judith seems to have for Rush, I think that we have all taken this to far and gotten way off the subject at hand: Emanuel’s (near) 180 on hurricanes. Berger has it right, I think: if this is based on models, as it evidently is, skeptics should take this only as seriously as they do certain other simulations…

    And just to show that I am not trying to be partisan, I must say that the reaction from the “usual suspects” is a bit of an over reaction, and puts a but of “spin” in the results.

  52. Posted Apr 13, 2008 at 2:45 PM | Permalink

    2005: Theory1 and modelling2 predict that hurricane intensity should
    increase with increasing global mean temperatures,

    2008: two of the seven model simulations Emanuel ran suggested that the overall intensity of storms would decline. Five models suggested a modest increase.

    Sounds a bit like a model 180.

  53. MarkW
    Posted Apr 13, 2008 at 3:29 PM | Permalink

    There is observational evidence for an increase of intense tropical cyclone
    activity in the North Atlantic since about 1970

    Yes, and they are called satellites.

  54. Posted Apr 13, 2008 at 4:43 PM | Permalink

    In the new paper, Emanuel and his co-authors project activity nearly two centuries hence…

    I couldn’t get past this part.

    Nail the weather forecast for the next 60 days, then I’ll be interested in reading more. And yes, I know weather and climate are not the same, but the reliability of long-term forecasting for either one seems on par.

    Why does anyone in the science community believe such complicated systems can be accurately projected for the next 200 years, and that we should base expensive public policy decisions on those projections?

  55. Kenneth Fritsch
    Posted Apr 13, 2008 at 4:56 PM | Permalink

    In gleaning Emanuel et al. (2008) for potential journalistic take aways that could be used to urge the public to action on AGW, I think that I would have to settle on the information from Figure 8. The authors of the paper have certainly not emphasized these points but an astute journalist, perhaps in the mold of a Chris Mooney, could.

    The figure shows the measures of TCs for the ratio of the changes in these measures from the period 1981-2000 to the period 2081-2200 as predicted by their computer models. The results are put into the form of 100*Log(F/P) where F is the measure of the 2081-2200 period and P is the measure of the period 1981-2000. The measures of TC are Power Dissipation, Change in Frequency, Change in Intensity and Duration. What is convenient about presenting the ratios in this form is that since Power Dissipation is a product of the other three measures and one can derive Power Dissipation by simply adding the bars (measures) in the graphs for the other three measures. Also the changes positive in the ratio greater and less than one are better represented in the log format, e.g. an increase in ratio by 100% has the same magnitude but different in sign from decrease in ratio by 50%.

    What can be lost in this representation are easily visualizing the differences between model predictions. To that end below I converted the log ratios back to ratios for Frequency Change and Change in Intensity. Since Power Dissipation is dependent on the other measures and Change in Duration is a relatively small factor I eliminated those measures from my calculations. I show the highest ratio of the six models used along with the lowest and the average of all models for the five TC basins. I eyeballed the graphs (in Figure 8) to extract the data for my calculations and assumed that Log meant log to the base 10. The last column in the table below converts Change in Intensity to a measure of wind velocity change since Change in Intensity depends on the cube of the wind velocity.

    Now look at NATL for the highest model estimations for Changes in Intensity and Changes in Frequency. Could not one extract from this a headline such as “Climate Models Used by Leading Climate Scientists Show a Possible 70 to 80% Increase in Tropical Storm Intensities and Frequencies”?

  56. Andrew
    Posted Apr 13, 2008 at 5:00 PM | Permalink

    Dogwood, climate scientists make a subtle distinction between prediction and projection which they would doubtless chastise you for not understanding (even though it is their fault for using the terminology). To wit, Kevin Trenberth:

    In fact there are no predictions by IPCC at all.
    And there never have been. The IPCC instead
    proffers ‘what if’ projections of future climate
    that correspond to certain emissions scenarios.
    There are a number of assumptions that go into
    5. The Rate of Sea-Level Rise Is
    Unlikely to Increase
    these emissions scenarios. They are intended to
    cover a range of possible self consistent ‘story
    lines’ that then provide decision makers with
    information about which paths might be more
    desirable. But they do not consider many things
    like the recovery of the ozone layer, for
    instance, or observed trends in forcing agents.
    There is no estimate, even probabilistically, as
    to the likelihood of any emissions scenario and
    no best guess. Even if there were, the
    projections are based on model results that
    provide differences of the future climate relative
    to that today.
    There is neither an El Niño sequence nor
    any Pacific Decadal Oscillation that replicates
    the recent past; yet these are critical modes of
    variability that affect Pacific Rim countries and
    beyond. The Atlantic Multidecadal Oscillation,
    that may depend on the thermohaline circulation
    and thus ocean currents in the Atlantic, is not set
    up to match today’s state, but it is a critical
    component of the Atlantic hurricanes, and it
    undoubtedly affects forecasts for the next
    decade from Brazil to Europe.
    The starting climate state in several of the
    models may depart significantly from the real
    climate owing to model errors. I postulate that
    regional climate change is impossible to deal
    with properly unless the models are initialized.

    Trenberth is, incidentally, an IPCC lead author of the TAR.

  57. Patrick M.
    Posted Apr 13, 2008 at 5:54 PM | Permalink

    Given RealClimate’s recent parsing of the words “scenario” and “forecast” as in:

    But this is no excuse for Pielke, almost twenty years down the track, to talk about “forecast” and “prediction” when he is referring to scenarios. A scenario tells us something like: “emitting this much CO2 would cause that much warming by 2050″. If in the 2040s the Earth gets hit by a meteorite shower and dramatically cools, or if humanity has installed mirrors in space to prevent the warming, then the above scenario was not wrong (the calculations may have been perfectly accurate). It has merely become obsolete, and it cannot be verified or falsified by observed data, because the observed data have become dominated by other effects not included in the scenario.

    link here

    How do we perceive Emanuel’s work; as a prediction or a scenario, (that should not be compared to actual data in the future)? Will this be falsifiable? Or has RealClimate created a language parsing loophole in climate science?

  58. Kenneth Fritsch
    Posted Apr 13, 2008 at 6:43 PM | Permalink

    Re: #58

    How do we perceive Emanuel’s work; as a prediction or a scenario, (that should not be compared to actual data in the future)? Will this be falsifiable? Or has RealClimate created a language parsing loophole in climate science?

    Emanuel et al. (2008) clearly state that they use the IPCC scenario A1b which calls, as I recall, for CO2 (or its GHGequivalents) increasing to 720 ppm by the end of 2100 and holding there through to 2200. Their predictions’ uncertainties are therefore subject not only to uncertainity in their models predictions given the inputs but also subject to uncertainties in the inputs. Emanuel was doing a scenario and with all the additional (if not actually mentioned in the paper) uncertainty that that implies.

    RC’s lecturing Pielke on this matter is laughable since it was Pielke, as I recall, who called into question the circumstances where one of Hansen 1988 Scenarios A, B and C could output something close to the actual for the wrong reasons, i.e the model could be wrong in handling the inputs and these inputs were also not according to the scenario, but the output was close to actual.

  59. Kenneth Fritsch
    Posted Apr 13, 2008 at 6:50 PM | Permalink

    Re: #56

    where F is the measure of the 2081-2200 period and P is the measure of the period 1981-2000.

    Should have been: where F is the measure of the 2181-2200 period and P is the measure of the period 1981-2000

  60. Patrick M.
    Posted Apr 13, 2008 at 8:11 PM | Permalink

    re Kenneth 59:

    So all the computer model jockeys are willing to stand by is that if you use the same inputs as they do and the same algorithm… you will get the same results. Whether the inputs or algorithm are correct or based in reality is irrelevant.

    Also notice the “almost twenty years down the track” comment. If one were to use less than 20 years I’m sure we’d hear complaints about “not enough data” or “it’s too soon to tell”.

    They generate “scenarios” for public consumption and then dodge scientific questioning by claiming it’s not falsifiable. It’s a shame because we really could use some insight into what is going on with the Earth’s climate.

    Sorry if this is drifting a little off topic. I’ll be quiet now.

  61. Posted Apr 13, 2008 at 10:45 PM | Permalink

    Yes, I know the difference between predictions and projections, although you wouldn’t think they know it by the doom and gloom in their press clippings.

    However, if they are merely “projecting hypotheticals” then we have absolutely no reason whatsoever to pay any attention to them at all. All they are doing is playing with very expensive toys (supercomputers) to produce something (climate model outputs) that has no relevance to the real world because they don’t understand how the climate works.

    Until they understand every aspect of how the climate works, especially the positive and negative feedbacks in the system, the climate model outputs are meaningless.

    So why does anyone pay attention to them? Other than they scream a lot in an attempt to scare people?

    Sorry, I just don’t get it.

    Guess I’m just a flat Earther.

  62. Andrew
    Posted Apr 14, 2008 at 9:44 AM | Permalink

    World Climate Report weighs in:

    NOTE: They discuss three papers, one of which is the paper we are discussing here. Thought you’d want to know.

  63. Kenneth Fritsch
    Posted Apr 14, 2008 at 10:20 AM | Permalink

    Re: #63

    Andrew, that link gives a good overview of some of the issues that have been discussed in this and related threads. Maybe it is just me, but I see a definite lack of cross-pollination of views and findings in this area of climate science. Perhaps the “growing” seasons are longer than I impatiently have expected.

  64. JP
    Posted Apr 14, 2008 at 10:44 AM | Permalink

    If Emanuel believes that El Nino and its attendent zones of verticle wind shear may play a bigger role in NATL Hurricanes, then it is incumbent for forecasters to get a better handle on ENSO cycles. As far as I know, the GCMS do not handle the change in state(s) of any of the globe’s teleconnections (ENSO,AMO,NAO, etc…). Another lingering unsaid assumption is that increased GHG concentrations will lead to more EL Nino events. In light of recent events, I think this assumption should be questioned.

    Maybe this paper will force some climatologists to turn thier gaze to the tropics. If one considers that changes to the Walker Cell have global climate implications, it is surprising that we dont’s have more people and resources devoted to cracking some of these mysteries. As long as the majority of research is consumed with just numeric prediction, I don’t see any big break throughs in understanding our atmosphere.

  65. Sam Urbinto
    Posted Apr 14, 2008 at 11:06 AM | Permalink

    Is anyone else having trouble grasping the subtle nuanced difference between a prediction, projection,
    scenario, “what if”, prognostication, divination, maybe, possibly and the like?

    So, computer simulations of a mass of variables, many of which are guessed at, changed and otherwise massaged, show that 7 runs show 5 up and 2 down. Think one of them may come out somewhat correct? 🙂

    Or as the IPCC said (paraphrased) The output of computer simulations using a number of plugged in fixed parameters of unknown skill suggest it’s likely future large-scale weather events will be more intense, with stronger winds and heavy rain. But we have little confience in “projecting” a decrease in their numbers. However, what looks like might be an increase in intense storms some places is larger than what was simulated during those periods.

  66. J. Peden
    Posted Apr 14, 2008 at 11:12 AM | Permalink

    The whole point of an assessment is that it is not what one person says, it is a review of the literature by a large group of people that go through a very extensive review process.

    I agree with what Francois Ouellette said so well in #48. I’m getting very tired of hearing people intimate that something like a vague consensus or a majority vote of a “large group of people” establishes the scientific validity of what the majority decides to conclude. I don’t even care “what one person says”, but rather why s/he says it, as shown by that person’s science.

    Likewise, the whole ipcc, enc., p-r campaign designed solely to hallow the ipcc’s conclusions and its Peers and their Peer Review process only indicates to me once again that the ipcc is not proceeding scientifically. The actual proof of this nearly complete failure is being found in the pudding – by CA, etc..

    The ipcc “scientific” process simply has not involved an attempt to disprove its own hypotheses and has not seriously involved trying to see what is wrong with what it is doing itself.

  67. Posted Apr 14, 2008 at 11:15 AM | Permalink

    JP #65, Is it conceivable that under global warming conditions, the threshold for vertical shear (~10 m/s) being destructive to hurricanes may actually increase? While vertical shear may become more anomalous at the high end, I have not come across any dynamical thinking about what impact this would have on hurricanes in the future — other than to say it would be deleterious nowadays. A small amount of vertical shear increase would seem to be easily mitigated by adding a 1-2 C to the SST, which would at the same time raise the climatological MPI. I note that the Vecchi and Soden paper so-often cited uses the same suite of models as Emanuel’s new treatise.

    We will simply get another iteration of all the same papers and ideas when the next generation of climate models comes out.

  68. Sam Urbinto
    Posted Apr 14, 2008 at 11:24 AM | Permalink

    Oh, and perhaps Rush pontificating upon AGW to tens of millions of listeners, and Dr. Spenser’s contribution as an “expert witness”, might be a reason to dislike Rush’s actions.

    Roy W. Spencer is principal research scientist at the Global Hydrology and Climate Center of the National Space Science and Technology Center in Huntsville, Ala. He is also U.S. team leader for the AMSR-E instrument flying on NASA’s Terra satellite.

    Cirrus disappearance: Warming might thin heat-trapping clouds

    PDF: Statement to the committee on oversight and governmnet reform of the United States House of Repesentatives

  69. Judith Curry
    Posted Apr 14, 2008 at 11:28 AM | Permalink

    Ryan, you touch on a critical issue. Much of the hurricane diagnostics related to things like SST, wind shear, potential instability, static stability, etc. are overly simplistic and are based on rudimentary understanding that is decades old. Taking wind shear for example, it makes a big difference whether the shear is directional or speed, and whether you are talking about genesis or intensification (this is well known from the literature, but the climate diagnosticians do not account for any of this). So as i see it there are two choices: the first is much higher resolution climate model simulations, and the second is to do some hard work (theory, observations, diagnostics, modelling) on getting a better understanding of the genesis/intensification factors and how they change. Personally, I think that trying to milk the coarse resolution GCMs for much about hurricanes using these simplistic parameters is misleading at best.

  70. Posted Apr 14, 2008 at 12:16 PM | Permalink

    #70 Judy, I presented a class project on this idea at the AGU meeting in San Fran last December which dealt with the changing nature of the background climatology. Currently, anomalies are compared to the 1971-2000 mean state of the atmosphere/ocean mainly from our current crop of reanalysis datasets. But, what isn’t always reported is the variance about that mean. It would be instructive to take the UN IPCC model results for monthly means and calculate the mean and sigma of the vertical shear (zonal and directional) during a couple time periods, say 2001-2030 and 2061-2090 to gauge how the background climo is evolving.

    Gray’s parameters could use an update with current reanalysis products. Sounds like a masters thesis that I don’t want to do.

    Judy, on a side note, when I was applying for grad schools in early 2002, I visited GaTech, prior to your arrival and the move to the new building. Yet, as Jim O’Brien was my master’s thesis advisor here at FSU, I look forward to hearing your war stories some day.

  71. Posted Apr 14, 2008 at 1:24 PM | Permalink

    ****Perhaps OT and snippable, but I would like to enter into the record the following: *****
    Sam, don’t forget that Rush knew and consulted with the late <strong>Dixy Lee Ray</strong>, who was, among other things chair of the Atomic Energy Commission, assistant secretary in the US Bureau of Oceans, former governor of Washington, and a university professor who wrote:
    “<em>Trashing the Planet: How Science Can Help Us Deal With Acid Rain, Depletion of the Ozone, and Nuclear Waste</em>”
    Link: http://www.amazon.com/Trashing-Planet-Science-Depletion-Nuclear/dp/0060974907

    ryanm …What’s this supposed to mean?

  72. Posted Apr 14, 2008 at 1:42 PM | Permalink

    Might be in response to: Sam’s “Oh, and perhaps Rush pontificating upon AGW to tens of millions of listeners, and Dr. Spenser’s contribution as an “expert witness”, might be a reason to dislike Rush’s actions.”

    There was a predecessor to Dr. Roy Spencer, it was Dr. Ray to be specific.

  73. VG
    Posted Apr 14, 2008 at 1:47 PM | Permalink

    Like all scientist who never discover anything, It seems that the AGW scientists who are most vociferous,play great attention to minutiae (small details) and miss the big picture (ie “models). Example in biology: Venter with DNA versus the HUGO team, Using his simple method they deciphered it (DNA) years ahead of the “official government” HUGO method. In the end, it will be proven that climate change (even recent) is due (most likely) to solar activity fluctuations (although we cannot prove which one at this time)

  74. Mark_T
    Posted Apr 14, 2008 at 2:17 PM | Permalink

    I came across this NOAA article a while back. The article talks about how sunspots may have played a role in the Titanic sinking. In the article, there is a graph of sunspots and sea surface temperature and they both correlate well to each other.

    Here is the link:
    And here is the graph:

    Below the graph, it says “(b) The globally averaged sea surface temperatures are plotted with the sunspot numbers (Reid; 1999). Both sunspot number and solar cycle length are proxies for the amount of solar energy that Earth receives. The similarity of these curves is evidence that the sun has influenced the climate of the last 150 years.”

  75. Mark_T
    Posted Apr 14, 2008 at 2:19 PM | Permalink

    So, going back to the graph of sunspots and sea surface temperature, how can we be sure that stronger hurricanes are from man heating the planet? According to this sst-sunspot graph, it appears that the sun could just as well be responsible for heating the oceans and causing stronger hurricanes.

  76. Patrick M.
    Posted Apr 14, 2008 at 2:44 PM | Permalink

    re 75 (Mark_T)

    Where’s 1985-Present?

  77. Mark_T
    Posted Apr 14, 2008 at 3:28 PM | Permalink

    Re 77:

    Don’t know. I wish it was available. Same with the sun’s TSi data from the NOAA which stops at the year 2000.


  78. MPaul
    Posted Apr 14, 2008 at 3:32 PM | Permalink

    I’m still struggling to understand how the atmosphere could overcome the thermal inertia of the ocean to produce meaningful ocean warming in a short period of time (~100 years) when delta T is so small and delta m (mass) is so large. Of course the term ‘surface’ is not well defined, and I could certainly see transient warming caused by the atmosphere down a few inches, maybe. But beyond that, I’m unconvinced. Someone’s got to show me. I’m no expert on hurricanes, but I suspect their warm water energy source needs to extend deeper that the few inches of true surface effect warming.

    This really is a sincere question, what am I missing?

  79. Kenneth Fritsch
    Posted Apr 14, 2008 at 4:25 PM | Permalink

    Ryan M’s post made me go back to the paper “Increased Tropical Atlantic Wind Shear in Model Projections of Global Warming” by Vecchi and Soden (2007) pull out what I judged to be pertinent excerpts that are presented below.

    There are common features in both papers as they both used 20 year periods for comparisons, forced their scenarios using the IPCC A1B scenario for CO2 emissions and Emanuel et al. (2008) used all 6 of the climate models that were include in the 18 climate models employed by VS (2007). The effects of wind shear were prominent features of both papers. The VS suggestion of using the wind shear for projections of long-term hurricane activity was evidently taken up in Emanuel et al. (2008). VS talk about wind shear not only impairing cyclo-genesis but in reducing storm intensities. I do not recall the Emanuel paper discussing the latter relationship.

    As an aside I was wondering whether these climate models show a faster warming mid troposphere than SSTs that I would guess would, by itself, inhibit TC activity. What we have measured in the tropics over the past 30 years is a mid troposphere temperature trend which is lagging SSTs and this should, I would guess, by itself, enhance TC activity – even though that differing trend rate goes against the temperature differences predicted by most climate models.

    The influence of these Vs changes should be incorporated into projections of
    long-term hurricane activity.

    We explore 21st Century projected changes in Vs over the tropical Atlantic and its ties
    to the Pacific Walker circulation, using a suite of coupled ocean-atmosphere models
    forced by emissions scenario A1B (atmospheric CO2 stabilization at 720ppm by year
    2100) for the Intergovernmental Panel on Climate Change 4th Assessment Report (IPCC23 AR4). Changes are computed between two 20-year periods: 2001-2020 and 2081-2100.

    Of particular importance is the vertical wind shear (Vs) which acts to inhibit tropical cyclone development [e.g. Pielke and Landsea 1999, Goldenberg et al 2001, Emanuel14 and Nolan 2004, Camargo et al 2007] and has a deleterious effect on the intensity of developed tropical cyclones [e.g., DeMaria 1996, Frank and Ritchie 2001]. The impact can be substantial for Vs>10ms-1, with one modeling study finding that “[s]trong shear of 15ms-1 literally tore an intense storm apart in about one day” [Frank and Ritchie 2001].

    The amplitude of the projected Vs increase is considerable, given the 1.5-3.5°C global-mean surface air temperature increase in these models by the end of the 21st Century [Held and Soden 2006, VS07]. These Vs changes are robust across the multi-model suite, with all but a handful of models projecting an increase in the 21st Century (Fig. 1.b). We define the tropical Atlantic region in which there is large increase of Vs in the ensemble mean (90°W-40°W

    All eighteen models had three-dimensional monthly-mean horizontal wind data available.
    Thus, for our analysis of vertical wind shear (Vs) we used: BCCR BCM2.0, CNRM CM3, CSIRO Mk3.0, GFDL CM2.0, GFDL CM2.1, GISS-AOM, GISS-EH, IAP FGOALS, INM CM3.0, IPSL CM4, MIROC Hi, MIROC Med, MPI ECHAM5, MRI CGCM2.3, NCAR CCSM3, NCAR PCM1, UKMet HadCM3, UKMet HadGem1.

    In addition to the potential impact on cyclogenesis, which can be estimated using
    quantities such as GPI [Emanuel and Nolan 2004], vertical wind shear can also adversely
    affect the intensification of an existing tropical cyclone [e.g. DeMaria 1996, Frank and
    Ritchie 2001]. A full understanding of the impacts of shear on tropical storm intensity
    should take into account possible nonlinearities in the response to shear, as well as
    nonlinearities in the response to both shear and other quantities (such as large-scale
    thermodynamic conditions). However, to the extent that the statistical relationship
    between storm intensification and shear described by DeMaria [1996] can be applied to
    the climate change problem, it can be used to estimate the potential impact of the
    ensemble-mean shear increase described above and in the main manuscript.

    Thus, it appears that – in the tropical Atlantic and East Pacific Oceans – the increase in
    vertical wind shear could partly mitigate the increased thermodynamic tendency towards
    increased storm intensity. However, it is important to note that it is only in the tropical
    Atlantic and East Pacific Oceans that there is a projected increase of shear during the
    local hurricane season. In the West Pacific and Indian Oceans, the models projected a
    long-term decrease in vertical wind shear through the 21st Century.

  80. Dave Dardinger
    Posted Apr 14, 2008 at 6:39 PM | Permalink

    re: #79

    I’m no expert on hurricanes, but I suspect their warm water energy source needs to extend deeper that the few inches of true surface effect warming.

    The big thing you’re missing is that wind drives a mixing of the surface waters down to about 10 meters minimum in the tropical summer or as much as 100 meters in colder winter waters. This means that temperatures are nearly constant quite a number of meters down. OTOH, there’s not mixing further down so that any surface warming won’t migrate to any great depth. Only ocean currents can move down and then only around the fringes of the polar ice where freezing ice makes the waters denser. But it that case the sinking water is quite cold to begin with so it can’t warm the bottom. See the thread on ice ages for a little more discussion of the subject.

  81. Posted Apr 14, 2008 at 7:05 PM | Permalink

    A few thoughts:

    (1.) When listing “future unknowns” that affect storm intensity in an AGW world it’s important to also include (a) mid-level wind shear (hard to detect and underappreciated), (b) the dryness and stability of the surrounding air and ( c) the presence/absence of outflow channels.

    These, and the ones listed earlier today by Judith, Ryan and Kenneth, are factors that might get stronger or might get weaker in an AGW world. Untangling these, including their interactions, so as to create a meaningful view of the future is problematic and possibly hopeless. For some reason I think of Paul Linsay at this point.

    (2.) For new people it’s helpful to think of a hurricane as a machine with moving parts. Think of it as a car engine which transforms fuel (ocean heat in this case) into motion (wind). AGW provides higher-octane fuel but it may also monkey with the moving parts (via wind shear, dryness, outflow channels, etc). That’s the question – how would the moving parts, the machinery, be affected by AGW? The higher-octane fuel from AGW could make the car faster, but not if the car machinery runs more poorly. This analogy has flaws but is in the ballpark.

    (3) We talk about the number of storms and the maximum wind of storms but the most important factor is storm size (the aerial extent of destructive winds). It seems reasonable to expect more-intense storms to have broader winds but that may or may not be true. It’s yet another factor to consider.

    (4) My conjecture on global tropical cyclone numbers and intensities is that tropical cyclones do things to the tropical atmosphere which hinder the development and intensity of other cyclones. A cyclone creates a negative feedback of sorts into the environment, not unlike biological systems that “stomp out” the competition.

    Specifically a cyclone pumps warm, dry, stable air into the upper and mid tropical atmosphere, “speeds” the upper circulation (= wind shear) over a large expanse and (I think) stabilizes the atmosphere near the tropopause. Incipient systems which are nearby in space and time struggle to overcome these negative factors placed in the environment by the mature cyclone. The negative environment also affects other mature systems, possibly even at a distance in space and time.

    Is this effect, this negative feedback, big or small? I dunno.

  82. Kenneth Fritsch
    Posted Apr 15, 2008 at 11:01 AM | Permalink

    I find a few ironic twists in the Emanuel et al. (2008) that might simply be attributable to my POV on the subject matter:

    1. Emanuel finds that his model is reasonably well verified at least in the NATL, if not for the rest of the globe, with regards to frequency and intensity, but then with that verification finds that the IPCC scenario A1B which raises temperature as much as 4.5 degrees C (with 2.8 degrees C most likely) does not increase TC activity for the highest predicting model porportionately much more than what has already been observed for a fraction of a degree C.

    2. Emanuel points to the 1970s to current NATL TC activity increase without mentioning the potential for a decadal cycle in activity and then appears to question the use of the short 20 year periods for comparison of late 20th versus late 22nd century TC activity.

    3. Emanuel uses wind shear in his models partly from his own theories (on at least the cyclogenesis part) that may have played a significant role in lowering what might have otherwise been increasing TC activity (with increasing SSTs) that his initial calculations seemed to project.

    On the 3rd point above my unsophisticated calculations as described below would seem to indicate that one can make a more consistent case for wind shear and 925 hPa variations affecting TC activity than one can for SSTs.

    In Post #663 in the YTD Hurricane thread, I looked at the Saunders and Lea variables as I did in the previous post, but this time for the correlations I used the David Smith Easy to Detect TC counts and the Bob Koss west of 60W ACE indexes. I did this in an attempt to analyze the correlations with TC measures that are less contaminated with the effects of changing detections capabilities. The results are presented below in the table. I included only the ASO monthly variable measures, but the AS measures were similar and as with the previous analysis was not sensitive to the monthly period used.
    The results again show that the correlations are consistent over time for the wind shear and 925 hPa variables but not for SST. With the easy detection TC measures it appears that the wind shear becomes a larger factor and more consistent with time.

  83. Posted Apr 15, 2008 at 12:12 PM | Permalink

    David Smith #82

    We talk about the number of storms and the maximum wind of storms but the most important factor is storm size (the aerial extent of destructive winds). It seems reasonable to expect more-intense storms to have broader winds but that may or may not be true. It’s yet another factor to consider.

    Yes, you are on to something, which was a Poster I presented at the AMS conference in New Orleans. I asked the question: How well does PDI/ACE estimate the overall Power Dissipation which takes into account the dimensions of the storm? There are storm dimension datasets available: including Kossin’s satellite derived global best tracks as well as Mark DeMaria’s Extended Best tracks. Each has the radius of max wind, max wind, and radius of 34, 50, and 64 knots. With that information and a crude Rankine vortex model (or, if you prefer, a fancier version from Holland 2008), you can easily calculate the true Power Dissipation and compare it to Emanuel’s PDI for the Atlantic — and say something about whether storms are indeed getting larger. The correlation between the two: PDI and PD (from Rankine and size best track) is pretty good on a low frequency, yearly integrated time period, however, it gets confusing when comparing storm to storm. Specifically, comparing Charley 2004 and Ivan 2004, or Wilma 2005 and Linda 1997 (from EPAC). Size matters — but, as I have conjectured in a manuscript, it only matters to a point. This is well-known because of the lack of correlation between storm intensity (max winds) and storm size.

    Emanuel et al. (2008) lacks this consideration, but it is probably not critical to his overall results/conclusions, since the resolution of his models are insufficient to credibly resolve inner-core processes.

    On a side note, I did compare the results of PDI to HRD HWIND analyses of the last few years of Atlantic storms, and found some interesting things, which I can elaborate on offline.

  84. Sam Urbinto
    Posted Apr 15, 2008 at 1:25 PM | Permalink

    I posted this over at Dot Earth in the page about Andrew Revkin’s article on the paper in response to some hand waving and misdirection from a couple of the posters.

    There’s no doubt and no denying any rise in energy levels are due to humans, nor denying that what the IPCC said in the TAR WG II technical summary is true; “Human activities—primarily burning of fossil fuels and changes in land cover—are modifying the concentration of atmospheric constituents or properties of the surface that absorb or scatter radiant energy. ”

    Yep. And?

    Is anyone denying the models are iffy? Does anyone deny water vapor, carbon dioxide, methane, nitrous oxide and CFCs absorb and emit IR, or that they react kinetically and chemically with other substances, including the other 98% of the atmosphere? Is anyone denying the global mean surface temperature anomaly trend is up ~.8 C over the last 127 years? Is anyone denying there’s an unsubstantiated claim that that carbon dioxide (or its equivalent) is the main culprit, rather than a phase changing, IR absorbing/emitting multi-process substance in much greater concentrations that’s part of the atmosphere, hydrosphere and biosphere?

    Tell you what; prove carbon dioxide or any of the other dry IR absorbing/emitting gases add a net increase in energy levels to the weather system, and therefore climate over time, with some empirical evidence, rather than asking others to prove it doesn’t. The process goes both ways.

    But here.

    ModelE shows clouds and water vapor alone can provide 85% of the given 33 C of warming alone, but carbon dioxide only 26%. If you remove carbon dioxide, you only lose 9%. But you lose 66% from removing clouds and water vapor. Clearly, all are unphysical and impossible, but hey.

    LINEPAK and HITRAN2004 have water vapor covering most of the absorption spectrum from 1-70 um whereas carbon dioxide only operates alone at 4.5 and overlaps at 2 and 12-20

    Considering radiative forcing alone, the IPCC has carbon dioxide estimated (“scenario”) in the WGI FAR SPM as providing a 1.5 to 1.8 W/m2 of the non-ozone non-stratospheric water vapor gases, and the others as .9 to 1.1 So at most, it’s only 2/3 of the 4 “long-lived greenhouse gases” And that’s only considering radiative forcing.

    Carbon dioxide has a calculated Cp of .84 Water vapor has a measured one of 2 (gas, solid) and 4 (liquid) Heck, asphalt has .92

    Besides a much larger heat capacity, water also changes phases and can hold and move much more energy. It takes 2270 kJ to turn it from water into vapor, which can then form clouds to block IR, absorb heat into the air and ground, or remove it out into space. That’s the equivalent kinetic energy in a liter of water that would move a 2 ton car 160 miles an hour or the potential energy to raise it 25 meters.

    The models used in “Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms”. show a decrease of pH by .075, to 8.104, over the last 243 years. By-products of pollution and carbon dioxide uptake, in what percentages? A .9% reduction of alkalinity over 234 years.

    You are also totally ignoring the rise in population from 1 billion to 6, all the urbanization and industrialization that goes along with it, as well as the established role of UHI in altering the heat going into the atmosphere and the effects upon weather it has many tens of kilometers down and upwind, and the role of particulates in the air and on the ground in this.

    So humans do influence their climate. I doubt anyone here is going to deny that organisms impact their environment. Although we might not agree with you how or why, and all you can do is back it up with conjecture and cries that we haven’t proven it doesn’t. Well buddy, you haven’t proven it does.

    {“It” being carbon dioxide making the anomaly rise et al}

  85. Judith Curry
    Posted Apr 15, 2008 at 3:03 PM | Permalink

    Ryan and David, tropical cyclone size is hugely important but neglected parameter. I have a graduate student working on this (will present some of this at Orlando meeting), have found there is some correlation between size and intensity, but it is not a strong correlation, it is a nearly independent variable but hugely important in terms of damage. There is a data set back to 1988 of atlantic basin TC size. My graduate student has created a data base of landfalling TCs back to 1930. I am predicting that more attention will paid to this in the near future, but it is hard work developing TC size data set. You can use satellite pictures or weather maps, we have used weather maps to determine radius of outer closed isobar (ROCI) as the main measure of size.

  86. Posted Apr 15, 2008 at 3:56 PM | Permalink

    I have read the correspondence with interest. To me the striking issue is the general acceptance of Emanuel’s model results by most of you who have been lambasting models for years. I guess it is a matter of what the results show! Hmmmm!

    But back to the KE paper. Whether the KE results stand up to scrutiny remains to be seen. The model is highly parameterized and appears to be very sensitive to changes in some quantities. For example (for those of you who have read or will read the paper as distinct from reading press releases or stories) check out the differences between his conclusions in the main body of the paper versus those obtained by changing the saturated vapor pressure (that’s the last figure in the paper). Small change, large differences although KE was honest to show this sensitivity. So there a number of things that have to be done such as sensitivity experiments. I have one question that I communicated to KE some time ago and that was his assumption of saturation in the middle troposphere which is not observed in nature.

    The trouble is that all of these tests is that they take time and when we have done them we will publish them in the peer-reviewed literature. No instant blog gratification unfortunately. Sorry!

    For what it is worth here is my take on the problem. I believe that there has been a substantial increase in major hurricanes since 1970 (satellite era). Even if the observations were in error by 50% (underestimated) in the 70’s and 20% wrong (overestimated) since 1995, there still is a significant trend. This trend is concurrent with an increase of SST. But the confounding issue is that the global number of storms worldwide has stayed essentially constant. Only in the Atlantic has both the number and intensity increased. However, I think that it is possible to reconcile these two seemingly paradoxical observations and explain physically (note physically: not a model result, but fundamental) why the number of storms will remain constant in the future but with moderate shifts (positive) in intensity. This is a paper submitted and, in addition, we are attempting to reconcile EKs’s results. Hopefully more on this later.

    The work we have done since our 2005 paper is based on the following premise: Determining the impact on global warming on the characteristics of tropical cyclones (number and intensity) is an important problem with significant societal implications. Unless we can explain physically the history of the number and intensity of TCs in the recent past, then determining the number and intensity of TCs in the future will be either an extrapolation of very poor data sets or a belief in incomplete and inexact models.

    I think it is easy fare to think of scientists who have reported that there appears to be a connection between warming SSTs and TC intensity as being rigid in their beliefs and that nothing that is said will change our minds. Sorry to say that it is not true. If we are proven wrong in 2005 paper then so be it. There would be an advance in understanding. We shrug, acknowledge and move on. Sorry if that sounds noble or corny. The people I work with all feel this way. We wait on convincing evidence. But I wonder if some of the people who write on this blog would ever change their minds. I think it is more fun to relax and see where all of this leads us! I know there are consequences……… No time for fun….

    I doubt anything that I have to say will have sufficient weight to change hardened opinions. But if our warm pool work does get published (we presented Steve with some of our early results when he visited Tech) then we are in for some new fun if getting hit on the head with two baseball bats is more fun that being hit with one bat! With our 2005 paper we received all types of criticism from the non-global warmers (is that a neutral enough term?). I think this time we had better duck more quickly as both sides will be taking shots at us!

  87. jeez
    Posted Apr 15, 2008 at 4:45 PM | Permalink

    Dr. Webster, you’re a funny guy–kudos.

  88. Andrew
    Posted Apr 15, 2008 at 4:59 PM | Permalink

    Peter Webster, I think you may have missed the comment above where I said:

    Berger has it right, I think: if this is based on models, as it evidently is, skeptics should take this only as seriously as they do certain other simulations…

    Perhaps you didn’t catch that I was applying this label to myself? Probably.

    I’m not sold on the results-or the way he got them-but I find them interesting. And I joked last night with some of my friends about the rather ironic nature of Emanuel’s conversion. 🙂

  89. steven mosher
    Posted Apr 15, 2008 at 5:43 PM | Permalink

    re 87: WTF

    “I have read the correspondence with interest. To me the striking issue is the general acceptance of Emanuel’s model results by most of you who have been lambasting models for years. I guess it is a matter of what the results show! Hmmmm!”

    I dont see “general acceptance of his results” I see people interested that he changed his mind. My comment, for example, was restricted to his methodology which I thought was interesting. But amuse me. Go through the 86 comments before yours and pick out the ones
    that you think show “general acceptance” of his results. That has to be easier than counting hurricanes. Frankly, as a former modeler, I dont trust a gosh darn thing any of them say, dirty secrets be told.

    But, I did like his approach. I thought it clever.

  90. Sam Urbinto
    Posted Apr 15, 2008 at 5:46 PM | Permalink

    Peter Webster;

    Is it too much to ask what parts of a computer simulation are less physical and real-world than others?

    I think that by and large, at least here, we are skeptical 🙂 of the models, but certainly are aware they can give us some idea of what’s going on, and that they’re getting better. I simply happen to think that the obvious fact that there is a lot we don’t know is nice to see being admitted. All I am interested in are the facts, backed up by data that is at least somewhat, shall we say reliable, and that known shortcomings are divulged.

    I personally don’t accept his model more or less so than any other just on the face of it. it simply seems more honest and more realistic. It could still very well be crap, whatever it’s showing.

    But if you’ve read what I written and will continue to write, I agree with the IPCC, but also take it all with a grain of salt especially given the writing style and focus. And I take arguments that anything other than water is a major player in the weather/climate as part of the hydrosphere and atmosphere both, fueled by the sun and moved by the wind, well, rather unconvincing (to say the least).

  91. Patrick M.
    Posted Apr 15, 2008 at 6:28 PM | Permalink

    re 87 & 90 (Webster & Mosher):

    I second Mosher’s remarks. Certainly there will always be people with their minds made up, but I think you’ll find fewer rubber stamps here at CA than at some other climate blogs.

  92. Harry Eagar
    Posted Apr 15, 2008 at 8:49 PM | Permalink

    I don’t understand Webster’s contention that knowing (the size of?) increases in storm intensity has societal implications?

    Unless he is contending that places that never saw big storms are going to see them in the future, the only societal implications that make much practical sense to me are things like: design for storm impact, don’t build in flood zones, plan for evactuations.

    All of these things need doing even if storm numbers and/or intensity are going to decline.

    Anyhow, I don’t greatly care, except in an intellectually curious sense, what the trend is over the past 38 years. Give me 380 and we can talk.

  93. david
    Posted Apr 15, 2008 at 10:00 PM | Permalink

    the key point is surely that there is no longer consensus and certainty

  94. Posted Apr 15, 2008 at 11:02 PM | Permalink

    #87 Thank you Dr. Webster for adding to our conversation on Emanuel’s new paper. I think we have come to a consensus on the Emanuel paper that it really clarified nothing, other than Kerry changing his mind. We look forward to seeing all of your sensitivity results, even if we have to wait until the iron-clad peer-review process completes.

    I believe that there has been a substantial increase in major hurricanes since 1970 (satellite era).

    I disagree and present a couple of many interesting coincidences.

    Must be coincidence: A simple check of my dataset for the Northern Hemisphere fails to show a significant trend in hurricane days since 1970. There is exactly zero trend in ACE for the NH from 1982-2007. Last year (2007), was a record inactive global TC year over a wide variety of metrics (presentation at Orlando TC conference). I do not believe the best tracks in the Southern Hemisphere back in the 1970s after all the imagery that Karl Hoarau has dredged up.

    Must be coincidence: When Eastern Pacific and North Atlantic ACE are combined, from 1982-2007, which is 26 years, and split down the middle at 1994-1995, representing the magnanimous climate-shift, you get the following numbers:

    1982-1994 ACE = 245
    1995-2007 ACE = 245
    Trend r = 0.00

    1982-1994 ACE = 590
    1995-2007 ACE = 571
    Trend r = -0.001

    Anyone on CA, please chime in, how can these coincidences be explained away? Aren’t you surprised that no one is reporting this other than a lowly graduate student living out of his car? (not really)

    Oh yeah, and I almost forgot, Global TC activity has crashed over the past 24 months to historic lows, continuing a decade long downward spiral. I fear that hemispheric TC activity may never recover and is decreasing even faster than we expected.

    But I wonder if some of the people who write on this blog would ever change their minds. I think it is more fun to relax and see where all of this leads us!


    Ryan’s Theory There is an operative correlation… (with tongue-in-cheek), as the number of papers published increases linking global warming with hurricane activity, the actual real-life (not mannequin) hurricanes disappear in an inverse proportion.

  95. Posted Apr 15, 2008 at 11:16 PM | Permalink

    Judy #86, I found that the r (PD,PDI) when storm size is taken into account from the Atlantic from 1979 – 2006 = about 0.70, for all TS+ observations. Of course when you integrate over each year and do some low-frequency smoothing, the relationship is essentially unity. However, what is more interesting, when using a modified Rankine vortex as a poor-man’s bogus to create a surface wind field, the relationship between PD and PDI linearly increases until the wind speed reaches about 100 knots and then it actually levels off and slightly decreases.
    So, the largest differences are of course found in the most intense hurricanes, as you can get Wilma or Katrina or Linda or Charlie all at Cat 4 or 5 but with a huge variety of sizes. Also the evolution of the PD (size included) during rapid intensification and/or weakening (or landfall in your grad student’s case), is actually pretty cool, especially with an animation of Kossin’s HURSAT images at the same time…

  96. Posted Apr 15, 2008 at 11:31 PM | Permalink

    Kenneth Fritsch #83: Scientists agree that shear is important for hurricanes. That’s the main mechanism for ENSO to influence Atlantic hurricane counts/ACE, hence your strong correlation, more robust over time. But the wind shear signal is mainly interannual, while the temperature signal is mainly interdecadal. There’s no consensus on tropical Atlantic wind shear trends under global warming, while everyone agrees temperatures will go up. So most studies until recently (Vecchi/Soden, Emanuel, etc.) have focused on the easier-to-deal-with long-term effects of temperature changes.

  97. MarkW
    Posted Apr 16, 2008 at 4:56 AM | Permalink

    Most everyone was expecting an increase in hurricane activity starting in the late 70’s due to nothing more than natural cycles.

  98. Michael Jankowski
    Posted Apr 16, 2008 at 6:25 AM | Permalink

    But I wonder if some of the people who write on this blog would ever change their minds.

    Some of the poeple who write on this blog already have changed their minds – swallowed the catastrophic AGW hysteria hook, line, and sinker initially, but then became skeptical after delving into the issue.

    As other posters have basically noted, much of what folks find “interesting” here in Emanuel’s work is (1) the apparent stance change and (2) the way the media and other climate scientists are not seizing onto his new results the way they did his previous works.

    I don’t remember reading anyone here accepting his results as defining the truth on the subject.

  99. Mike Davis
    Posted Apr 16, 2008 at 7:13 AM | Permalink

    I think more accurate observation and less SWAGS to base the models on will one day produce better results. Of course due to the cyclic nature of our global weather system it would probably require another 30 to 60 years to aquire this information and then trying to reconstruct the distant past will still be a best guess situation. The work that is being done to understand TS and TC is important. But the tendancy to cry WOLF at every turn is what is hurting the activists cause the most.

  100. Kenneth Fritsch
    Posted Apr 16, 2008 at 8:47 AM | Permalink

    Re: #97

    But the wind shear signal is mainly interannual, while the temperature signal is mainly interdecadal.

    John N-G, you might want to expand on your comment here. I always have problems with these types of observations and particulary when it is stated that the temperature is signal is interdecadal.

    Are you saying that temperature anomalies tend to cycle decadally or that there is an anomaly trend over decadal periods of time or that one cannot necessarily calculate a good correlation between annual temperature changes and TC activity but that by using say a 10 year moving average one does obtain a better correlation?

    I have a problem with the latter observation as it merely indicates a trend with temperature that is overwhelmed on an annual basis by other variables. Furthermore the MA correlations more readily leads to confounding temperature changes with other variables that might change in the same direction over the same period of time — as noted for detection capability changes.

  101. Kenneth Fritsch
    Posted Apr 16, 2008 at 9:22 AM | Permalink

    Re: #87

    I have read the correspondence with interest. To me the striking issue is the general acceptance of Emanuel’s model results by most of you who have been lambasting models for years. I guess it is a matter of what the results show! Hmmmm!

    I agree with others here that I do not myself or see others at CA making any new found concessions to these models other than the interest generated when the models appear to go against the prevailing view and the reactions it provokes.

    What is of most interest is the questioning and revealed uncertainties in the climate models that is evidently available on all sides of the AGW issue.

    The trouble is that all of these tests is that they take time and when we have done them we will publish them in the peer-reviewed literature. No instant blog gratification unfortunately. Sorry!

    As long as papers keep coming in on this subject matter at the rate we have seen over the past few years, sufficient material to analyze will be avaiable to keep these threads interesting enjoyable and a learning experience — for me anyway.

    For what it is worth here is my take on the problem. I believe that there has been a substantial increase in major hurricanes since 1970 (satellite era).

    Are you ruling out a decadal cyclical occurence rate and a potential confounding of SST trends and changing detection capabilities?

    I doubt anything that I have to say will have sufficient weight to change hardened opinions.

    On the other hand, did the recent KE paper change your mind concerning any of your public pronouncements and writings? Judith Curry has indicated that the work published from 2005 on is the only work relevant to this subject matter. I need to ask her about some very worrisome projections she made for future TC storm activity in the NATL and whether she feels the projections continue to be valid in her mind. I do not recall if the projection was published post 2005 or not.

  102. Posted Apr 16, 2008 at 9:42 AM | Permalink

    I suspect that Dr. Webster participates in a good deal of “bat-swinging” himself! If Webster is concerned about being caught in the middle without a bat, I have a Louisville Slugger I can loan him.

  103. J. Peden
    Posted Apr 16, 2008 at 11:18 AM | Permalink

    we received all types of criticism from the non-global warmers (is that a neutral enough term?).

    No, it’s not quite neutral enough for me, Dr. Webster, unless I’m misunderstanding your intent, in which case I apologize upfront. But, I think you have perhaps become inadvertently trapped within the Denier/non-warmer vs Believer/warmer schema which is really intended to turn a scientific question into a political or even a religious one, a tactic which is often used to try to stigmatize CA. or anyone else auditing or finding fault with the processes, conclusions, and recommendations involved in AGW/ipcc science – but which can equally be used to inappropriately stigmatize anyone willy-nilly seen as a “warmer”. [Still, if the shoe does occasionally fit, some people are just going to have to wear it.]
    But from what you say in your total post, Dr., I assume you don’t really see things according to that particular schema at all and are intent upon ‘simply’ doing good science, and therefore fit quite well here with the vast majority of the contributors and commenters – at least as far as I can tell from extensive lurking here.

  104. Kenneth Fritsch
    Posted Apr 16, 2008 at 12:35 PM | Permalink

    Judith Curry, since you and Peter Webster have stated that scientists working in the TC field can be flexible and change their views as more information becomes available, I would like to ask you directly whether your views as expressed in your April 2007 Congressional testimony have changed in the intervening year’s time – and particularly with regards to the Emanuel et al. (2008) paper under discussion on this thread. I have excerpted some of your testimony and a link to the entire testimony below.

    In my view it easy for scientists to express the general thesis that flexibility and open mindedness is operating in their scientific communities, but perhaps more difficult to express on an individual level. Peter evidently sees some hardheadedness at CA so being sensitive to this trait he might also want to jump in and let us all know how his views have changed.

    Judith Curry Congressional Testimony: Hearing on “Dangerous Climate Change”
    26 April 2007

    Click to access Curry_Energy.pdf

    To infer what the hurricane activity might look like in the coming decades, a simple statistical model is formulated that accounts for both global warming and natural variability to estimate the average conditions in the year 2025, using results from both the observational record and climate model simulations. In the year 2025, we assume that the tropical sea surface temperatures have increased by 1oF owing to greenhouse warming. Figure 6 suggests that an increase of 1oF is associated with an increase of 5 tropical cyclones, while the increased number of North Atlantic tropical cyclones projected by Oouchi et al.’s high-resolution climate model simulation is slightly less than 1 when scaled for an increase of 1oF. Hence, we bound the range of the expected increase in tropical cyclones for a 1oF temperature increase by 1-5 storms. Further, we assume that 2025 is near the peak of the AMO cycle. Different interpretations of the relative importance of the AMO versus global forcing of the surface temperature yield estimates of the magnitude of the impact of the AMO on the total number of tropical cyclones per year range from 0 (no effect) to 4 (the AMO explains the entire magnitude of the trough to peak variability in Figure 7); since we are halfway up the positive phase of the AMO, we infer a maximum additional contribution of 1 cyclone from the AMO by 2025.

    Based upon these assumptions of variability of the total number of North Atlantic tropical cyclones, consider the following simple statistical model for the projection of the average number of North Atlantic tropical cyclones in 2025. The average annual number for the past decade of North Atlantic tropical cyclones is 14 (Figure 7). We assume that the effects of greenhouse warming and the AMO are separable and additive. Adding the range of contributions from global warming plus the AMO to the base value of 14 tropical cyclones yields a range of projected average annual numbers of tropical cyclones in 2025 ranging from 15 to 20, the range accounting for the uncertainties in the impacts of both global warming and the AMO. Interannual influences, such as the El Nino-La Nina cycle will lead to some years being substantially lower, but others substantially higher, and future years similar to or exceeding 2005 must be expected. Thus, the combination of global warming and the elevated activity associated with the active phase of the AMO can be expected to result a level of tropical cyclone activity that is unprecedented in the historical record. In terms of the intensity of the storms, an increase in the number of category 4 and 5 hurricanes is expected, ranging from 3-4 per year.

  105. Dave Blair
    Posted Apr 16, 2008 at 12:51 PM | Permalink

    Here is my prediction: Emanuel’s change in opinion will be linked to Big Oil.

  106. ChrisZ
    Posted Apr 16, 2008 at 2:33 PM | Permalink

    Re: Dave Blair 97

    Certainly – by teleconnection if needs be (i.e. if no plausible connection can be established otherwise…)!

  107. Posted Apr 16, 2008 at 8:49 PM | Permalink

    Somewhat OT:

    One of the poorly-understood aspects of intense hurricanes (cat 4 and cat 5) is that they seem to transform from garden-variety tempests into severe storms quickly. Sometimes the word “bomb” is used to describe their sudden transformation.

    For illustration here are pressure-change plots of several recent cat 4/5 hurricanes:

    Each storm had at least one period where its central pressure dropped by at least 20mb in six hours (that’s very rapid).

    These are just examples but I think that someone (Klotzbach?) wrote a short paper on Atlantic hurricane characteristics that mentioned this observation as being characteristic of intense storms. It’s unusual to find a storm that made it to cat 4 by slowly intensifying. Cat 4 and cat 5 storms underwent a sudden change of some sort whereas weaker hurricanes rarely do so.

    What’s the point? Well, I think that “bombing” represents a structural transformation of some kind to the storm, perhaps in the upper levels. This transformation makes intense storms differ from garden-variety ones in both degree and in kind.

    My conjecture is that some atmospheric parameter allows, or promotes, the transformation. If this trigger can be identified and understood then maybe it’d be possible to model or otherwise reason how AGW would affect this trigger and thus how AGW would affect intense hurricanes.

  108. Marine_Shale
    Posted Apr 16, 2008 at 9:38 PM | Permalink

    Peter Webster (87)

    With our 2005 paper we received all types of criticism from the non-global warmers (is that a neutral enough term?). I think this time we had better duck more quickly as both sides will be taking shots at us!

    The implication from this comment is that there was general endorsement of your views (in 2005) by your peers, this as you know was not the case. Both “sides” were dissatisfied with your 2005 paper, in particular the special pleading on your part to disregard the the very large difference between the theoretical projection of a 10% increase in cyclone intensity and your nearly 100% observed increase. Dr Curry, Dr Mann and yourself have argued that attribution of cause has been established (AGM)and that the argument now is simply about the magnitude of the effect, a very curious scientific method.

    This was the consensus statement by your peers:

    Consensus Statements by International Workshop on Tropical Cyclones-VI (IWTC-VI)
    Participants: November 2006
    1. Though there is evidence both for and against the existence of a detectable
    anthropogenic signal in the tropical cyclone climate record to date, no firm conclusion
    can be made on this point.
    2. No individual tropical cyclone can be directly attributed to climate change
    3. The recent increase in societal impact from tropical cyclones has largely been caused by
    rising concentrations of population and infrastructure in coastal regions.
    4. Tropical cyclone wind-speed monitoring has changed dramatically over the last few
    decades, leading to difficulties in determining accurate trends.
    5. There is an observed multi-decadal variability of tropical cyclones in some regions
    whose causes, whether natural, anthropogenic or a combination, are currently being
    debated. This variability makes detecting any long-term trends in tropical cyclone
    activity difficult.
    6. It is likely that some increase in tropical cyclone peak wind-speed and rainfall will
    occur if the climate continues to warm. Model studies and theory project a 3-5%
    increase in wind-speed per degree Celsius increase of tropical sea surface temperatures.
    7. There is an inconsistency between the small changes in wind-speed projected by theory
    and modelling versus large changes reported by some observational studies.
    8. Although recent climate model simulations project a decrease or no change in global
    tropical cyclone numbers in a warmer climate, there is low confidence in this
    projection. In addition, it is unknown how tropical cyclone tracks or areas of impact will
    change in the future.
    9. Large regional variations exist in methods used to monitor tropical cyclones. Also, most
    regions have no measurements by instrumented aircraft. These significant limitations
    will continue to make detection of trends difficult.
    10. If the projected rise in sea level due to global warming occurs, then the vulnerability to
    tropical cyclone storm surge flooding would increase.
    The comprehensive scientific statement is available at: http://www.wmo.ch/web/arep/arephome.

    You also state:

    The work we have done since our 2005 paper is based on the following premise: Determining the impact on global warming on the characteristics of tropical cyclones (number and intensity) is an important problem with significant societal implications. Unless we can explain physically the history of the number and intensity of TCs in the recent past, then determining the number and intensity of TCs in the future will be either an extrapolation of very poor data sets or a belief in incomplete and inexact models.

    This is true and I think this following summary indicates where you have failed to convince your fellow scientists to date.

    SEPTEMBER 2007
    Scott Power and Karen Pearce (Editors)
    Tropical cyclones in a changing climate:
    research priorities for Australia
    Abstracts and recommendations from a workshop
    held on 8 December 2006 at the
    Bureau of Meteorology in Melbourne
    Detection and attribution of climate change effects on
    tropical cyclones
    Kevin Walsh
    School of Earth Sciences, University of Melbourne

    In climate change science, detection is the “process of demonstrating that climate has changed
    in some defined statistical sense, without providing a reason for that change” (IPCC 2001). This
    could involve anything from a simple statistical analysis of a trend of a chosen observed
    variable to a complex fingerprint analysis of a combination of variables. Additionally,
    attribution implies not only that a signal has been detected but that it is of the correct magnitude
    expected from projections of climate change due to anthropogenic forcing. These projections
    are typically made by an ensemble of climate models, and for successful attribution to be
    claimed, the detected change must fit within the envelope of the ensemble predictions.
    Moreover, it must also be shown that the detected change is not consistent with alternative,
    physically-based explanations.
    For these reasons, recent claimed trends in tropical cyclone characteristics are not successful
    examples of attribution studies. Recent results (Emanuel 2005; Webster et al. 2005; Sriver and
    Huber 2006; Curry et al. 2006) suggest that large trends are evident in parts of the observed
    tropical cyclone record. For this to constitute successful detection of a trend there must be
    confidence in the magnitude of both the signal and the noise. In the case of a climatological
    analysis of tropical cyclone trends, we therefore must be confident that there are no artefacts in
    the tropical cyclone data that would give either a spurious trend or a record that was incomplete.
    Unfortunately, it can be easily argued that the current tropical cyclone records are not free from
    data inhomogeneities. The most telling examples of this so far are the analyses of Kamahori et
    al. (2006) and Wu et al. (2006). They examine trends in severe tropical cyclone numbers in
    different competing “best-track” data sets in the north-west Pacific region, finding very large
    differences in trends even for the same tropical cyclones and even in very recent times, when
    the data should be best. They ascribe this result to the different analysis techniques used in the
    rival data sets. At present, it is not clear which data set best represents reality.
    Recent partial reanalyses of the tropical cyclone record have shown substantial corrections in
    trends compared with studies that have analysed existing best-track data. Kossin et al. (2007)
    have employed a reanalysis of global geostationary satellite data from 1983 to 2005 to remove
    time-dependent biases, finding that detected changes in tropical cyclone formation basins other
    than the Atlantic are less than in previous analyses. This is particularly relevant to Webster et
    al.’s (2005) near-global analysis, as their largest claimed increase in severe category 4 and 5
    tropical cyclone numbers occurs in the early to mid 1980s. A recent reanalysis of the record in
    the western Australian region (Harper 2006) has also found that increases in severe tropical
    cyclone numbers are less than previously estimated using best track data. Landsea et al. (2006)
    uses modern intensity estimation methods applied to satellite images of non-Atlantic tropical
    cyclones from the late 1970s and early 1980s to show that the intensities of the storms are likely
    significantly underestimated in the best track data that were compiled at that time.
    The region where tropical cyclone data is best is undoubtedly the Atlantic basin, where regular
    aircraft reconnaissance is combined with the latest intensity estimation techniques. Here,
    detected trends in tropical cyclone intensities are harder to dismiss. Emanuel (2005) and Mann
    and Emanuel (2006) argue that trends since the 1950s in a measure of tropical cyclone strength,
    the power-dissipation index (PDI), are highly significant and well correlated with observed
    increases in sea surface temperatures since that time, a phenomenon which is well known to be
    associated with anthropogenic warming.
    Even so, it is by no means clear that even the Atlantic data is free from inhomogeneities that
    would cause intensity data from the 1950s and 1960s to have biases compared with recent data.
    But even if the tropical cyclone trends in the Atlantic constitute a detection of a climate change
    signal, is it clear that they do not at present represent formal attribution of a climate change
    effect. This is because the projected size of the intensity change in the Atlantic to date due to the
    observed increase in sea surface temperature (SST) is considerably smaller than the claimed
    observed change in intensity. For the observed increase in Atlantic SST since 1960 of about
    0.4ºC, theories of tropical cyclone maximum intensity (Emanuel 1988; Holland 1997) predict an
    increase in maximum intensity of about 2%, corresponding roughly to a PDI increase of about
    10% (Emanuel 2006), not about 60% as observed. Thus since the magnitude of the response is
    not as expected, attribution is not possible at this time.
    The projections of expected climate change used in attribution studies are usually obtained from
    climate models. For tropical cyclone studies, the utility of climate models is currently limited by
    their relatively poor performance at simulating important aspects of tropical cyclones and by the
    wide range in their projections of climate change effects on tropical cyclones. This should not
    be a cause for pessimism, however; attribution studies for other climate variables have become
    widely accepted (e.g. Barnett et al. 2005), and as climate model simulations of tropical cyclone
    characteristics improve, it may one day be possible to ascribe any detected trends in the
    observational record of tropical cyclones to the effects of anthropogenic global warming. Other
    approaches, though, might also prove productive. The development of a quantitative theory of
    tropical cyclone formation, particularly severe tropical cyclone development, would enable
    attribution studies to be performed without the need for full fine-resolution general circulation
    model experiments. Refinements to the existing theory of maximum potential intensity may
    enable more confidence to be placed in the magnitude of its predictions. Thus a serious
    modelling effort needs to be combined with observationally-based theoretical studies and
    further efforts to homogenize the existing tropical cyclone data.

    I am also interested in your comment about understanding the recent past.
    I think it would be good to look at the distant past as well.

    Global climate change and the tropical cyclone
    palaeo-record in Australia
    Jonathan Nott
    School of Earth and Environmental Science, James Cook University, Cairns

    Attempts to model the longer-term natural variability of tropical cyclones are commonly based
    on extrapolations from the either the historical record (past ~ 100 years) or the instrumental
    record (past 30-40 years). Inherent in this approach is an untested assumption that the longerterm
    synthetic records should display the same statistical properties as the shorter record upon
    which they are based. These properties are stationarity and randomness of successive events.
    Statistical tests of these assumptions for numerous real or actual (non-synthetic) long-term
    records of natural hazards from around the globe (floods, droughts, thunderstorms, earthquakes,
    volcanic eruptions) along with a number of the causative mechanisms of some atmospheric
    hazards such as sea surface temperature (SST), ENSO and IPO show that both serial correlation
    and non-stationarity are common properties of natural records when they exceed 100-150 years
    in length (Nott 2006). This suggests that in the vast majority of cases the shorter historical and
    instrumental records display an apparent white noise signal which is in reality part of a longer
    term cyclic trend whose mean and variance can change at centennial scale periodicities. This
    work implies that we cannot use the short-term records of tropical cyclones as the basis for
    modelling the long-term variability without first testing the inherent assumptions of stationarity
    and randomness; the only way to do this is against long-term records of actual or non-synthetic
    In order to assess whether global climate change therefore is really having an impact upon the
    behaviour of tropical cyclones in a region it is critical that any changes in behaviour are
    examined against the natural variability and the only way to do this at present is to use longterm
    records of actual events.
    Review of current research activities
    Long-term sedimentary records of tropical cyclones in the form of coral shingle, shell and sand
    ridges and sheets register only the most extreme events at multi-century to millennial scales
    (Chappell et al. 1983; Chivas et al. 1986; Liu and Fearn 1999; Nott and Hayne 2001; Nott 2003,
    2004). The resolution of these records is too coarse to determine the natural variability of these
    events. High resolution (annual) records are needed for this purpose. To this end an 800-year
    long, high resolution (annual) record of tropical cyclones (AD 2004-1200) has been derived for
    the Cairns region of North Queensland. This record registers the isotopic signature of tropical
    cyclone rainwater within an annually layered limestone cave stalagmite. Comparisons of the
    first 100 years of the record against the Bureau of Meteorology’s best track cyclone database
    showed that every major cyclone and greater than 75% of all cyclones to track within 300 km of
    the site (130 km inland of Cairns) were registered within the stalagmite.
    This annual record suggests that there have been centennial scale variations in tropical cyclone
    landfall frequency over the past 800 years. The centuries from AD 1800 to 1600 and AD 1500
    to 1400 were very active, the period from AD 1200-1400 was exceptionally quiet and since
    AD1800 activity has been moderately quiet and especially so since European settlement of the
    region in the late 19th Century. These variations in tropical cyclone landfall frequencies appear
    to be best related to the behaviour of steering winds in the region, at least for the past 400 years
    rather than variations in sea surface temperatures. The heightened phase of cyclone landfalls
    (AD 1800-1600) also correlates closely with a 400-year long coral luminescence record of river
    discharge which shows heightened flood activity here during the period AD 1800-1600.
    Both the isotope and sedimentary records from this region suggest a different long-term
    climatology of tropical cyclones for northern Australia than that assumed from the short (30-40
    year) instrumented record.

    Apologies for the length of the post, snip away if you want to Steve

  109. Posted Apr 17, 2008 at 6:44 PM | Permalink

    With the update I posted up above with the MIT press release, I predict hundreds of media articles and press on the new BAMS article. There is no UNCERTAINTY in that prediction.

  110. Posted Apr 18, 2008 at 8:22 AM | Permalink

    Re #110 That’s an interesting press release. It’ll be up to Emanuel to clarify, of course, if he so chooses.

    Anyone interviewed by the press, even in-house press, will likely have times where they were surprised by what later appeared in press. The interviewee may not communicate well or carefully choose his/her words, or may be asked bad questions, or maybe the reporter doesn’t comprehend properly, or maybe the reporter has preconceived notions or bias/agenda, or perhaps an editor rewords things, and so on.

    But, having said all of that, I suspect that this in-house press release got Emanuel’s OK before release. The effect of this release is to cloud the issue and make it difficult for anyone to use “sound-bite” quotes to promote their side. That is probably a good thing for Emanuel on a personal level, lest he be accused of aiding the Dark Side. It should also allow his inbox and cellphone to cool, which I suspect have been warm in recent days.

    The bottom line is that he has the power to clarify, if he so chooses.

  111. Posted Apr 18, 2008 at 9:20 AM | Permalink

    David #111 — It seems the majority of lefty eco-journalists without a background in climate science or meteorology simply cannot ask the right questions or be objective enough to interpret the complex machinations of a high-level theoretical modeling study. So, what you get are the rightie-bloggers seizing upon sound bites to fit their agenda, equally distasteful.

    On a scientific note, did you notice that he is saying the last 25-27 years of observational records in the Atlantic are now independently confirmed by the theoretical modeling study ! I am clueless on this statement. First of all, the observational records in the Atlantic since 1980 or 1982 are satellite era — and you can get the ACE or PDI trend to about 95% by just counting the number of satellite pictures where you see an eye — and call it a hurricane and give it 64 knots. It is the frequency trend where you have to count Tiny-Tims or subtropical whirls that causes more uncertainty.

    I just showed up in #95 that the Atlantic time series is very closely anti-correlated to the Eastern Pacific, so, Emanuel should provide an explanation in his BAMS article showing the Eastern Pacific PDI/ACE collapse. Anyone find that for me?

  112. Gerald Machnee
    Posted Apr 18, 2008 at 9:22 AM | Permalink

    Re #110: Is there really “No Uncertainty”??
    **There are several possibilities, Emanuel says.**
    **While the earlier study was based entirely on historical records of past hurricanes, showing nearly a doubling in the intensity of Atlantic storms over the last 30 years, the new work is purely theoretical.**
    **Another possibility is that the recent hurricane increase is related to the fast pace of increase in temperature. The computer models in this study, he explains, show what happens after the atmosphere has stabilized at new, much higher CO2 concentrations.**
    I see two big problems – 1) It is purely theoretical (models) and 2) It is linked to warming by CO2, which raises questions.

  113. Posted Apr 18, 2008 at 9:39 AM | Permalink

    Gerald, you need to reread #110. We are all aware of the assumptions made in the model, but that wasn’t the purpose of my comment.

  114. Kenneth Fritsch
    Posted Apr 18, 2008 at 9:39 AM | Permalink

    Re: #110

    With the update I posted up above with the MIT press release, I predict hundreds of media articles and press on the new BAMS article. There is no UNCERTAINTY in that prediction

    I am relatively new to the TC world, but with my brief tenure I think I can safely observe that reactions of climate scientists working in this field are much easier to predict/forecast than the seasonal and longer term TC activity.

    David Smith, to this layperson who wants to learn more about the science of TC activity, your posts on areal extent of TC/hurricanes and abrupt pressure drops in Cat 4 and 5 hurricanes seemed like interesting topics for separate threads. The areal extent of TCs, I guess I had assumed was well covered in already available TC metrics.

    Ryan Ms observation that while the agreement of PDI and a metric with more direct measurements of the areal extent (??) can be poor when calculated storm to storm that it agrees well over many storms and time periods would seem to indicate that a 20 year to 20 year comparison that Emanuel et al. (2008) made would be unaffected by the alternative measurement method.

    David, can you show some comparisons of Cat 4 and 5 hurricane pressure drops with Cat 1, 2 and 3 hurricanes? The naïve layperson might observe that since the pressure drops in the Cat 4 and 5 storms are assumed to be greater than in the lower categories storms that the overall drop in time for the Cat 4 and 5 storms from ambient conditions must be greater. Could one also observe that if most of the higher Cat storms appear to form quickly via pressure drops, that it is a quick transition is necessary in order for a higher Cat storm to avoid the destabilizing effects that would more readily degrade a lower Cat storm.

    Re: #109

    Marine_Shale, why would think that your informative post would be snipped? Good stuff.

  115. Posted Apr 18, 2008 at 10:10 AM | Permalink

    Re #112 Ryan I agree it’s an odd statement and it needs elaboration if it is indeed a major point he’s trying to make. Emanuel has the power to clarify, if he so chooses.

    Your point on the ATL/EPAC anticorrelation is excellent and I will gin up a couple of graphs, for newer readers. If you happen to have some graphs handy that show EPAC/ATL trends, or lack thereof, please post and that’ll save me some time this afternoon for my honey-do list. It’s a interesting topic that has gotten little notice.

  116. Posted Apr 18, 2008 at 11:51 AM | Permalink

    Re #115 Hello, Ken. I’ll pull together something on pressure profiles of intense storms vs garden-variety storms. It’s a side-street but interesting. Sudden changes in intensity is a poorly-understood aspect of tropical cyclones and is probably the one that makes forecasters sweat the most.

  117. Kenneth Fritsch
    Posted Apr 18, 2008 at 4:33 PM | Permalink

    Re: #117

    David, I was simply thinking in terms of Poisson type occurrences of TCs and perhaps up to Cat 4 and 5 hurricanes. If these storms needed a very brief period of just the right conditions to either form or intensify after their cyclogenesis, it puts them more in the realm of Poisson occurrences, I think. When I did an unsophisticated modification of the Mann theory of TCs counts following a binary Poisson distribution and determined how well the counts fit a Poisson distribution after the years/seasons were divided into two categories with the AMM index positive or negative for Easy to Detect storms for the NATL, I obtained some good fits with chi square tests.

    I want to use a version of this model for predicting TC frequency, intensity and hurricanes for your contest, so I need to become an expert in a big hurry and any assistance will be most welcomed. Remember I do not use emoticons in my posts.

  118. Posted Apr 18, 2008 at 7:24 PM | Permalink

    Dear Marine Shale (Re 109),

    Can’t quite work out if this is your real name (your parents were geologists?) or you are hiding behind some strata. That was an attempt at a “funny”. That aside: I am glad you have never reviewed a paper of mine. What you write is twice as long as our infamous (and blessedly short) 2005 Science paper.

    I apologize, but at the moment (I am about to go to Bangladesh for 2 weeks where we have a non-TC project: Note non-TC project!) I cannot give your comments the full justice they deserve. I will get back to them in the next two weeks (there isn’t too much to distract one in the night life of Bangladesh, alas so perhaps I can say something more quickly) so forgive me for the response that is incomplete and perhaps a little glib.

    I gave a talk yesterday at University of Maryland where I spent most of the time talking about the changes in deep tropical convection during the last 50 years and what we might guess (artfully guess?) what might happen in the next 100 years. I think I mentioned this in my last post. But I did show the Webster et al 2005 cat 4+5 result although it was a minor point. I argued that if we were 50% wrong in counting cat 4-5 in the first decade of the satellite era and if all of the errors were underestimates AND if we were 20% wrong in the last decade of the satellite era and all of the errors were overestimates, one can still find a +33% trend between 1970-74 and 2000-2004. So my point was that it was difficult (poor data sets aside) to ignore the increase in intensity over the last 30 odd years.

    Given this statistic, I argued that one had a choice: One could reduce everything to “climate arithmetic” where one waits for the next # of TCs and argue (depending on your opinion) that the new number is either an overestimate or an underestimate or “the data is too bad to find a trend but good enough to find a “natural variation”. I apologize but I have become bored with this argument. I am happy to admit that the data sets are flawed. But I have tried to take a different tack.

    Make a couple of hypotheses. Lets consider two of these.

    (1) The SST has increased by x % over the last y years and, through Clausius-Claperyon arguments, this should lead to an increase in the intensity of TCs. Basis of this hypothesis given above. Observations tell me that there has been an increase in TC intensity.
    (2) There has been no increase in the intensity of TCs globally. The SST has increased but this has nothing to do with the intensity of TCs.
    (3) Data and models so bad can’t say anything….!

    It doesn’t matter what hypothesis you choose. The point is that you have to defend it! I prefer (1) and the argument that we have put forward and to which I alluded to in my last post was to do with the evolution of the warm pool (More later). But (2), as pushed by Landsea and Gray needs a defense as well. It can’t be left hanging that “it can’t be true”. There is an obligation to show that as SSTs have risen that something else has “conspired” to keep the intensity constant. Is it that shear or stability somehow work oppositely to SST increases? If so let’s have a convincing argument about how the SST can increase but TC intensity doesn’t change. All of us have to defend our stands and the importance of the question (TC intensity in coming years) makes this an imperative.

    The point that I am making is really a simple one. I think we can all agree that the data is rough. And one should question conclusions made with poor data sets. I believe that the true value of “climateaudit” is the raising of these questions. But, let’s not stop attempts in trying to understand climate change because the data is bad. Lets test (1)-(2). Are they physically consistent? The data is never going to get better. And it is boring to hear the arguments that in 1923 an obscure ship made a cat 2 into a cat 3 or in 1888 the reverse happened. Isn’t it more fruitful to develop a hypothesis, state the hypothesis and test it physically! At the same time make sure that the hypotheses developed with the poor data stand examination and scrutiny (climate audit?).

    I will get back to you after Bangladesh. Your points deserve a more thorough discussion Sorry to be so long-winded.

    Peter W

  119. Posted Apr 18, 2008 at 8:13 PM | Permalink

    Below are three plots, one of which is quite interesting, I think.

    These are based on Atlantic storms over five years (2003-2007). I used those five recent years because the measurement technology is adequate and consistent, and there were plenty of storms. Five years should be long enough for this demo purpose.

    I looked at three pieces of data:

    1. Maximum windspeed during the life of the storm (knots)
    2. Maximum 6-hr pressure drop during the life of the storm (mb/6hr)
    3. Number of advisories issued on the tropical system (issued every 6 hrs, including tropical depression stage, excluding any extratropical stage) (this is sort of the period of opportunity that a system has for strengthening)

    I did scatter plots of each combo.

    The first is max windspeed vs number of advisories. It seems reasonable that, the longer a storm system is around, the greater its opportunity to grow. The plot is here:

    It shows a respectable relationship between duration and max wind, as expected, though it also indicates that there’s lots more to max windspeed than just the time the storm exists.

    (Note: Hurricane Wilma of 2005 was an exceptional storm which I’ve marked, for reasons I’ll mention below. Also, a number of those dots in the lower left are our young friends the Tims.)

    Next is a plot of storm system duration versus maximum rate of pressure drop (max pressure drop is my current interest). It seems reasonable to expect that, the longer a storm system is around, the greater its opportunity to have a period or two of sharp pressure drop.

    It’s rather surprising: there seems to be little correlation between system duration and the system’s max pressure drop rate. It takes more than “hangin’ around” to have a vigorous growth spurt.

    (You can also clearly see the exceptional Hurricane Wilma, an outlier in so many ways.)

    Now for number three. This shows the maximum intensity during the life of the storm versus the maximum rate of pressure drop it experienced at any time in its existence:

    If Wilma is excluded then there is a whopping r-squared of 0.78 for the relationship, a high value in the world of climatology (and there’s no smoothed data in any of this, of course).

    I wish I had a clear and simple explanation for such a strong relationship, but I don’t. I may be overlooking something quite simple – all suggestions are appreciated. This is what I’m pondering at the moment. It seems plausible that the max rate of pressure drop is a good indicator of the health of the system and/or the condition of the nearby environment but that’s too general a statement.

    Back to the original point, the plot suggests that the intense storms didn’t just drift into severe strength, they “exploded” into it. The science needs to understand why that happens before much can be said about the future. My sense is that nature is keeping a secret from us, but I may be overcomplicating things.

  120. Kenneth Fritsch
    Posted Apr 18, 2008 at 8:53 PM | Permalink

    Contrast Posts #119 and #120. The devil is in the details.

    Thanks David for your follow up on max wind speed and max pressure drop. I find that last graph most interesting, that I have learned something and that there is meat there for further discussion and/or analysis.

    It could have been a learning experience had Peter Webster felt inclined to discuss some of the details of the debate concerning the TC historical record — and before 1970. And sincerely adressing the evidence for changing detection capabilities and the decadal cyclical properties of TC activity and the potential for confounding SST with the detection changes and cyclical tendencies. Instead of addressing these details he chose, in my view, to tell us what most of us here already knew — outside Peter’s apparent tendency to become bored.

    Maybe I do not fully comprehend the essence of what Peter was saying but it sounded like: I have my story and I am sticking to it.

  121. Posted Apr 18, 2008 at 9:02 PM | Permalink


    (3) Data and models so bad can’t say anything….!

    Actually, you’ve got this option a bit wrong. The issue with the data can be: the data are either so bad or limited that we can’t do decent hypothesis test to confirm or refute our hypothesis.

    This doesn’t mean you can’t make a hypothesis, argue that it’s correct etc. It just means that, if the data are insufficient in some way, you can’t confirm or refute empirically.

    You can still advance any hypothesis you like, or even say what you like.

    However, people will criticize your conclusions if you claim your hypothesis is proven or strongly supported by the data, when in reality the empirical support is weak. This type of criticism is entirely valid. Not only that, it should be seen as valid no matter what side of the “hurricane wars” one falls on.

  122. Kenneth Fritsch
    Posted Apr 18, 2008 at 9:12 PM | Permalink

    David, your max wind speed versus max pressure drop may be exhibiting a plateau whereas once the sudden drop in pressure gets a hurricane up to Cat 4 or 5 wind speeds, the wind speeds tend to be limited to a natural max limiting value (as I recall this is not necessarily the case) or other factors operate to take the winds to higher speeds. Or the max pressure drop effect has a plateau above which there is no increasing influence on wind speed.

    We need some hurricanes with max pressure drops between the field and Wilma.

  123. steven mosher
    Posted Apr 19, 2008 at 10:00 AM | Permalink

    ryanm: snipped

  124. steven mosher
    Posted Apr 19, 2008 at 10:58 AM | Permalink

    ryanm: snipped

  125. Posted Apr 19, 2008 at 12:05 PM | Permalink

    #125, we are trying to get Dr. Webster to provide us with this new knowledge so we don’t have to bore ourselves (or he for that matter) with climate arithmetic. Clearly Dr. Webster has a new set of talking points he is test-driving on us here, which are backed up by some as-yet unknown theory to us.

    He is still clinging to the Cat 4-5 script like a sacred text, while I am clinging to my NH ACE time series which shows no trend and the current period of global TC energy collapse. Then again, I just may be bitter.

  126. steven mosher
    Posted Apr 19, 2008 at 2:48 PM | Permalink

    re 126. I know, forgive my intemperance. thanks for snipping.

  127. Gerald Machnee
    Posted Apr 19, 2008 at 3:15 PM | Permalink

    Re Webster’s #119, He states:
    (1) The SST has increased by x % over the last y years and, through Clausius-Claperyon arguments, this should lead to an increase in the intensity of TCs. Basis of this hypothesis given above. Observations tell me that there has been an increase in TC intensity.
    (2) There has been no increase in the intensity of TCs globally. The SST has increased but this has nothing to do with the intensity of TCs.
    My observations of these two comments tell me the following: In (1) Webster claims that SST has increased> In (2) he notes that there has been no increase in the intensity of TCs globally. Therefore GLOBAL Warming has no effect on the intensity of TCs. And I do not need error bars for this conclusion.
    Further to that, when you are studying TCs from the 1970s onward, you are looking at half a cycle, something similar to looking at a temperature trend from January to June.

  128. Posted Apr 19, 2008 at 6:48 PM | Permalink

    Re 119: I was posing possible hypotheses!

    But I am intrigued with your “half cycle” you quote. What is the physical cause of a global oscillation? I remember reading one of the the Bell and Chellia (2006, I think). They used 55 years of NCEP reanalysis and found “half a cycle” of 100 years duration. That’s pretty good with such a limited 50-year data set. That was their explanation of a trend over the last 50 years. It is one thing to say that everything is a “cycle” but you have to back it up with physics. Gray attempted to argue that it was a simultaneous oscillation of the thermohaline circulation. But he had to use a couple of “stretches” such as evaporation decreases with SST and evaporation decreases with increasing wind speed.

    So if you have observational evidence of a 100 t=year global cycle or a physcial explanation of why one might expect one, I would (sincerely) be very interested.


  129. Posted Apr 19, 2008 at 9:58 PM | Permalink

    Re #123 Kenneth I checked the records for the last 30 years and (fortunately) found nothing approaching Wilma’s pressure drop. So, the data gap remains.

    During this check I noted that 85% of the intense (cat 4 and cat 5) storms had at least one 6hr period of a 2mb/hr or greater pressure fall, with half seeing 3mb/hr or greater. Those are respectable pressure drops.

  130. Jim Clarke
    Posted Apr 20, 2008 at 6:35 AM | Permalink

    Peter Webster said (#131):

    “It is one thing to say that everything is a “cycle” but you have to back it up with physics.”

    Well, that depends on what you are trying to do. If you are trying to increase human knowledge and advance science, then understanding the physics is very important. If you are trying to make projections of future climate change to help policy makers make good decisions, a prerequiste to ‘understand the physics’ becomes detrimental to the process.

    For example, we may not understand the physics that drives the PDO, but we know it exists. We also know that it has a large impact on global climate that appears to go beyond increased El Ninos in the warm phase and more likely La Ninas in the cool phase. Ignoring all of this, just because we do not understand the physics, means that we try to reconcile all observations with the physics we do know, mistakenly coming to the conclusion that humans must be to blame for the majority of recent climate change.

    It is irrational to ignore what we recognize yet do not fully understand as if our lack of understanding means that it does not really exist. Yet the consensus view of climate change rests on this obvious fallacy. Ironically, we do not fully understand the physics of clouds and watervapor, but the climate change community doesn’t seem to have any problem supplying the models with guesses on those processes, especially when it supports the initial hypothesis.

  131. Gerald Machnee
    Posted Apr 20, 2008 at 7:22 AM | Permalink

    Re #131 – A simple explanation of the approximately 70 year cycle is that there is a maximum around the 1940 area. So we are at another maximum now. But this is being ignored because we do not have “satellite data”, so we start at a low point in the 1970’s.

  132. JamesG
    Posted Apr 20, 2008 at 9:55 AM | Permalink

    Like others here, I find it odd when people talk about global warming when referring to regional phenomena. Talking about Bangladesh, when I looked at the climate data there, I found zero trends (ie flat lines) in everything – rainfall, droughts, cyclones, temperatures – since records began in the 60’s and ’70’s. Of course that shows only that the models are very poor at a regional scale and hence shouldn’t be used to predict regional effects (a practice which is unfortunately very prevalent), but a zero trend cannot actually prove or disprove any storm link to temperature. Of course one can conclude that any slight effects from any future slight temperature trend in Bangladesh would be dwarfed by the massive seasonal variations – That is, they need help now, not in 2050. But, while I’ve not looked elsewhere, I wonder if this lack of tropical warming is typical, since most of the “global” warming seems to be in the Arctic.

  133. Kenneth Fritsch
    Posted Apr 20, 2008 at 10:25 AM | Permalink

    Re: #131

    I was posing possible hypotheses!

    It is one thing to say that everything is a “cycle” but you have to back it up with physics.

    On the other hand, if one is promoting a physical theory that seems to fit (to some degree) a short 30 to 40 year period of data (and for one TC basin of the several existing globally) but one does not look beyond that time period (with perhaps a little more effort in constructing a reasonable re-analysis) what does that say about the theory and the promoter thereof?

    It is also not like a proposed relationship between SST and TC activity could not be rather easily confounded with a cyclical component of TC activity and/or the improvements in measuring capabilities of TC activity over time — a potential confounding that needs to be dealt with head on.

  134. Posted Apr 20, 2008 at 1:05 PM | Permalink

    While exploring the side-street issue of pressure drop and intense storms I plotted the locations where the cat 4 and cat 5 storms had their most-rapid pressure drop:

    The blue dots are the locations and the background colors are SST (the redder the warmer). I used the SST map from September 2007 because I had it handy and temperatures were near-normal.

    This is a crude approach with a need to adjust for storm frequency, among other things. Nevertheless I was surprised at the number of dots in the open Atlantic, where SST and oceanic heat content are not as high as in the western Caribbean or Gulf.

    Different subject: the NHC has a colorful map of tropical cyclone paths for the Atlantic and eastern Pacific:

    (Note that the Pacific portion covers only 60 years while the Atlantic covers 160, meaning that there’s a lot of missing red in the Pacific. I wish they’d do apples-to-apples comparisons instead.)

    It gives a sense of that the eastern Pacific and Atlantic are part of one Western Hemisphere storm basin, rather than two completely distinct regions. That’s an arguable point. There is some evidence that they act in opposite ways to things like ENSO, with an increase in activity in one being offset by a decrease in the other.

    A plot of the eastern Pacific (EPAC) decadal-scale activity is below:

    As Ryan noted, EPAC activity declined at the same time that Atlantic activity increased. The net change has been a wash.

    In case you’re wondering, the ACE and SST in the chart have an r-squared of 0.012, and that’s after using 3-year smoothing.

  135. Posted Apr 20, 2008 at 9:14 PM | Permalink

    A final plot regarding storm pressure:

    This shows the central (“eye”) pressure of the storms in the famous 2005 season. Red and orange denote the major hurricanes while the cooler colors represent the tropical storms and weak hurricanes.

    The lower the pressure, the stronger the storm (more or less). Normal pressure (no storm present) is about 1014 mb.

    Note how the systems are roughly equivalent for the first couple of days. Then, about day 3 or 4, the ones which are to become major storms, the killers, sort of break away from the pack and tend to “bomb”. It’s not a slow drift into severe strength but rather an uneven sprint.

    The weaker systems, on the other hand, tend to rock along in the 980-1010mb band.

    Science today doesn’t have a thorough grasp of what drives rapid intensification. That grasp has to occur before anyone can say much of anything about future storm severity.

  136. UK John
    Posted Apr 21, 2008 at 2:27 AM | Permalink

    After an unpromising start, (the first blog leader was misleading, but later corrected) I learnt quite a bit from this Thread.

    Still amazed that results obtained from experiments on models, are treated as something worthwhile, when there is loads of historical real data that still defies adequate explanation or understanding.

  137. John A
    Posted Apr 21, 2008 at 4:11 AM | Permalink

    Time for the “John A” cynical review:

    While the earlier study was based entirely on historical records of past hurricanes, showing nearly a doubling in the intensity of Atlantic storms over the last 30 years, the new work is purely theoretical.

    It might be right, but you can’t prove it wrong until well after I’ve retired.

    “It strongly confirms, independently, the results in the Nature [2005] paper,” Emanuel said. “This is a completely independent analysis and comes up with very consistent results.”

    “Independence” meaning in climate science “maximum non-independence” since it involves the same personnel, mostly the same model and the same assumptions about climate.

    There are several possibilities, Emanuel says. “The last 25 years’ increase may have little to do with global warming, or the models may have missed something about how nature responds to the increase in carbon dioxide.”

    Or I could be guessing. About everything. I really have no idea.

    Another possibility is that the recent hurricane increase is related to the fast pace of increase in temperature. The computer models in this study, he explains, show what happens after the atmosphere has stabilized at new, much higher CO2 concentrations. “That’s very different from the process now, when it’s rapidly changing,” he says.

    So how could a model which assumes stabilization of CO2 have any useful prediction of today’s situation of rising CO2 levels? It’s an academic mystery. Something that surely a responsible peer reviewer would have asked before publication – but no.

    In the many different computer runs with different models and different conditions, “the fact is, the results are all over the place,” Emanuel says. But that doesn’t mean that one can’t learn from them. And there is one conclusion that’s clearly not consistent with these results, he said: “The idea that there is no connection between hurricanes and global warming, that’s not supported,” he says.

    Translation: Our modelling is not robust to even very slight changes in a single parameter, leading us to conclude that model is chaotic (Edward Lorenz, RIP) and so your guess is as good as mine, but you’re not being paid to guess and I am.

    On the other hand, just because the results are all over the place doesn’t mean that our central assumptions in the model are wrong.

    Just in case you missed the key pitch, here’s Professor Curry to explain it:

    “The issue probably will not be resolved until better computer models are developed…By publishing his new paper, and by the virtue of his high profile, Emanuel could be a catalyst for further agreement in the field of hurricanes and global warming …

    Translation: Now that we’ve got the headlines for a model that predicts everything and nothing, has zero predictive power, is speculated to be radically incomplete in its base assumptions, and proves nothing and disproves even less, please please keep the downspout of federal funding flowing into the silicon crystal balls that we work on and which our grad students and post docs depend on for their livelihoods.

    Did I miss anything? Was there any frontier of science moved back even a fraction as a result of this non-result?

    Did Dr Emanuel give any key indicators of any future climate or meteorological events or time series which would disprove his model?


    Now back to your regular programming.

  138. Posted Apr 21, 2008 at 5:01 AM | Permalink

    120, 137, 138 (david Smith):

    There are several aspects to tropical storm intensity that may influence a correlation graph. SST’s are but one. Surrounding environment and the length of time in a quality environment plays a large role, as does proximity to land areas. In actuality, environment and land mass influences go hand-in-hand to some degree. Additionally, if you were to plot cat 4/5 storms by month, you will find correlations than if you take the season as a whole.

    If you take your graph in 137 showing rapid intensification against SST’s, you show that SST’s are not central to rapid intensification, but supportive of it. In fact, if you were able to plot surrounding environment and the amount of time the TC was surrounded by this environment, I would gather that the correlation coefficient would be quite high. Indeed since you show that high SST’s are not required for cat 4/5 rapid intensification, then why do we not have such storms in June??

    It is all in the environment, all things being equal.


  139. JamesG
    Posted Apr 21, 2008 at 6:09 AM | Permalink

    “Indeed since you show that high SST’s are not required for cat 4/5 rapid intensification, then why do we not have such storms in June??”
    I’d say that just like every other storm, it is likely to be a potential difference that is the cause not an absolute value of temperature. IMHO the SST tells us only where it is likely to happen but the pd is what makes it happen.

  140. Posted Apr 21, 2008 at 6:36 AM | Permalink

    142 (JamesG):

    It is likely to be a combination of both our statements. A TC must have good structure and a good surrounding environment in order to take full advantage of the fuel (SST’s). We often see TC’s with poorly defined structures and/or poor outflows not take advantage of the high potential from SST’s. IMHO, environment plays the most important role in TC development. Without a good environment, TC’s will not develop. With marginal environment, high SST’s can allow the storm to obtain minimal intensification in spite of the surroundings. A good environment can allow a TC in minimal SST’s to obtain some intensification, but with high SST’s, it will allow the potential for a cat 4/5. All someone needs to do is quantify different states of environments and SST quality, and cross analyze the combinations. I know that is overly simplistic, as other factors come in to play, such as internal structure of the TC, but it may allow for discussion to focus on the important aspects.


  141. Judith Curry
    Posted Apr 21, 2008 at 7:18 AM | Permalink

    Re the issue of surface temperature, Peter Webster’s new research (which hopefully he will discuss again in a few weeks when he returns from travel) is that it is not the absolute value of the SST, but rather it is the region of intense diabatic heating from convection that is associated with tropical cyclogenesis. The threshold temperature associated with this diabatic heating varies with climate state and the hemispheric circulation patterns. The canonical value of 26.5oC found by Gray in the 1980’s is really about 27.2C under current conditions in the NATL.

    The issue of rapid intensification to cat 4/5 is a very interesting one (not to mention important), the whole issue of rapid intensification has been the bane of hurricane forecasters. We tried to forecast this last season in the NATL, did well with Dean but totally blew Felix (which was a very small storm in horizontal extent, associated with dry air). My current thinking on this is that there conditionals rather than separate absolute conditions, e.g. if SST is warm, you have a good disturbance (e.g. easterly wave) and the large scale dynamics (e.g. stretching deformation and wind shear) are favorable, then humidity may not be an inhibiting factor.

    Also, in the literature, the factors that cause initiatiation of tropical cyclones are lumped in with the same ones that intensify an established hurricane. This is an over simplification, the dominating factors are likely to different at different stages of the storm life cycle. Looking at the transition to cat 4/5 (which is typically pretty explosive and happens on a short time frame), is a good way to start sorting this out. (we are working on the cyclogenesis part at the moment).

  142. Posted Apr 21, 2008 at 10:38 AM | Permalink

    When you examine the overall number of observations in the best track record since 1985 that are of greater than 115 knots, the number is not overwhelming.

    The black dots are 1980s, red dots 1990s, and blue dots 2000s. I don’t need cluster analysis to show the interdecadal changes. The ring current in the Gulf is a bullseye, the western Caribbean, and the area north and northeast of the Islands all have SSTs of > 29C, which is a rough minimum for intense Cat45 activity. The longest track is Ivan 2004 which managed to avoid land in an uncanny way.

  143. Posted Apr 21, 2008 at 7:02 PM | Permalink

    One of the concerns about a warming of the eastern tropical Atlantic is that it would promote earlier development of seedlings, producing more-developed (stronger) tropical cyclones approaching the western Atlantic. Perhaps so, but there may also be a tendency for these early bloomers to curve poleward and stay entirely at sea.

    Below is a map of the eastern tropical Atlantic (big blue box) and the entry-region into the western Atlantic (red box).

    The numbers show the percent of storms which formed in the small blue boxes which later passed through the red box during the last 30 years. For example, 44% of the tropical cyclones which formed (reached 35kt windspeed) between 45W-50W and 10N-20N eventually passed through the red box. The other 56% curved out to sea.

    As the storms form farther and farther east the suggestion is that they may be more likely to curve out to sea and miss land entirely. There are some meteorological reasons to suspect that an earlier transition from a wave into a cyclone would indeed support an earlier veer toward the north.

    This is simply something to ponder when considering the effects of a warmer Atlantic on hurricane behavior.

  144. Posted Apr 21, 2008 at 7:20 PM | Permalink

    Re #141, #143 Tom, I agree with all that you’ve written.

    Regarding June storms, I think Michaels et al showed that the major storms need +28C SST, something of a necessary-but-not-sufficient condition. That’s probably in short supply in June. June also has relatively high wind shear and a general environment which is probably drier and more stable than it’d be in August or September. I recall only Audrey (1957) as a bad June storm, one which surprised the forecasters (and killed about 500 people in Louisiana).

  145. Geoff Sherrington
    Posted Apr 22, 2008 at 5:15 AM | Permalink

    Re MPaul # 79

    Theoretically, a hurricane is a very brief transient event in a restricted location. If surface sea temperature is to have an influence, then one should use localised SST not global or hemispherical.

    I have not done the heat transfer equations, but intuitively I cannot see how a delta T of 1K in SST would provide the extra (localised) input to change either the intensity or frequency of the event. Is there a reference showing how the heat balances work and how deep the sea temp changes as the hurricane forms and progresses? I can only see local temperatures acting akin to a catalyst in a chemical reaction. Please convince me I’m wrong.

    Until then, I have to agree with MPaul.

  146. Kenneth Fritsch
    Posted Apr 22, 2008 at 10:30 AM | Permalink

    Re: #148

    Theoretically, a hurricane is a very brief transient event in a restricted location. If surface sea temperature is to have an influence, then one should use localised SST not global or hemispherical.

    Here is a link to Micheals reply to an Emanuel critique and references to his paper on using local SSTs to characterize TCs.

    Your reference to the transient and brief nature of TCs gets me back to my story that I am sticking with (in Peter Webster fashion although as a layperson without a reputation to defend) that NATL TCs fit a Poisson distribution if the changes in detection capabilities over time are compensated for and one uses a two separate distibutions depending on whether the AMM index is positive or negative. I do not see any data presented here in the past several posts or anyone, for that matter, refuting my view (which may also arise because would-be critcs are not going to waste their time addressing conjectures from a layperson).

  147. Sam Urbinto
    Posted Apr 22, 2008 at 1:47 PM | Permalink

    I don’t see the issue here. This is about the hypothesis of AGW intensifying hurricanes being reconsidered. It says nothing of what’s sure or not. Just the way the release is written it looks.

    As the first story wrote:

    global warming may still play a role in raising the intensity of hurricanes. What that role is, however, remains far from certain.

    Or as Vecchi said:

    “While his results don’t rule out the possibility that global warming has contributed to the recent increase in activity in the Atlantic, they suggest that other factors — possibly in addition to global warming — are likely to have been substantial contributors to the observed increase in activity,”

    And remember, the original report says only 28.6% of the model runs for the Atlantic showed a decline in intensity, or as Emanuel said:

    “There’s still a lot of uncertainty in this problem. The bulk of the evidence is that hurricane power will go up, but in some places it will go down.”

    I don’t see how this is at odds with

    In the many different computer runs with different models and different conditions, “the fact is, the results are all over the place,”

    And I certainly agree this doesn’t disprove anything.

    “The idea that there is no connection between hurricanes and global warming, that’s not supported,”

    Or in other words, we’re at the same point in the process; how a newspaper story or press release or blog determines how to spin “We don’t know, and nothing’s proven or disproven in either direction.”

    I’d be interested in seeing the full set of remarks from Emanual, rather than bits and pieces strung together in the way the author wants to present or paraphase them. Certainly it seems the MIT release is trying to downplay certain aspects.

    The proper headline for the MIT release would be more like: “Study fails to invalidate hurricane predictions” or “New computer simulations still a mixed bag in their output”

    As in some of the caveats earlier in the release:

    But the new work shows no clear change in the overall numbers of such storms when run on future climates predicted using global climate models.


    the new work also raises some questions that remain to be understood. When projected into the future, the model shows a continuing increase in power, “but a lot less than the factor of two that we’ve already seen” he says. “So we have a paradox that remains to be explained.”

  148. Kenneth Fritsch
    Posted Apr 22, 2008 at 4:07 PM | Permalink

    Sam, if you follow these stories and paper publications closely I think you can see the need to put them into context. This is Emanuel who is obviously vacillating on the relationship of SST and TC activity. Do not expect any clearer statements from the scientists’ mouths on these issues. Some of these statements from press releases from some climate scientists would make a politician green with envy. You have to read their analyses (minus their press releases) and do you own evaluations.

  149. Sam Urbinto
    Posted Apr 22, 2008 at 4:30 PM | Permalink

    Hardly surprising, this is climate science after all. 🙂

    But yes, make our own conclusions. So where are the original questions? It sounds like one interview chopped up, of course it’s unknown if the questions were prepped, on the fly according to answers, or instead something pre-written as discussion by Dr. Emanuel.

  150. Posted Jul 23, 2008 at 6:25 PM | Permalink

    GEOPHYSICAL RESEARCH LETTERS, VOL. 35, L14705, doi:10.1029/2008GL033950, 2008
    A 1,000-year, annually-resolved record of hurricane activity from Boston, Massachusetts
    Mark R. Besonen et al.
    Department of Geosciences, University of Massachusetts, Amherst, Massachusetts, USA


    The annually-laminated (i.e., varved) sediment record from the Lower Mystic Lake (near Boston, MA), contains a series of anomalous graded beds deposited by strong flooding events that have affected the basin over the last millennium. From the historic portion of the record, 10 out of 11 of the most prominent graded beds correspond with years in which category 2–3 hurricanes are known to have struck the Boston area. Thus, we conclude that the graded beds represent deposition related to intense hurricane precipitation combined with wind-driven vegetation disturbance that exposes fresh, loose sediment. The hurricane signal shows strong, centennial-scale variations in frequency with a period of increased activity between the 12th–16th centuries, and decreased activity during the 11th and 17th–19th centuries. These frequency changes are consistent with other paleoclimate indicators from the tropical North Atlantic, in particular, sea surface temperature variations.

  151. Posted Jan 16, 2009 at 2:02 PM | Permalink

    Here’s the citation for a recent final rule to the Code of Federal Regulations regarding global warming. It concerns the EPA’s ruling on alternative substitutes for use in the refrigeration and air conditioning, fire suppression and explosion protection, and foam blowing sectors.


    40 CFR Part 82

    [EPA-HQ-OAR-2003-0118; FRL-8758-9]

    RIN 2060-AG12

    Protection of Stratospheric Ozone: Notice 23 for Significant New Alternatives Policy Program

    AGENCY: Environmental Protection Agency (EPA).

    ACTION: Determination of Acceptability.

    SUMMARY: This Determination of Acceptability expands the list of acceptable substitutes for ozone-depleting substances under the U.S. Environmental Protection Agency’s (EPA) Significant New Alternatives Policy (SNAP) program. The determinations concern new substitutes for use in the refrigeration and air conditioning, fire suppression and explosion protection, and foam blowing sectors.

    DATES: Effective January 2, 2009.

    Information taken from http://www.cyberregs.com

  152. Posted Jan 21, 2009 at 7:15 AM | Permalink

    Thanks for great post. In Bangladesh, we are victim of global warming. The weather sounds terribly changed here.. Its more hot in summer and more cold in winter. But this was not the case before. We all need to be united to stop global warming.

5 Trackbacks

  1. […] http://www.climateaudit.org/?p=2994#comments […]

  2. By Robot Blogger on Apr 19, 2008 at 2:03 PM

    ryanm: snipped

  3. By Robot Blogger on Apr 19, 2008 at 2:57 PM

    ryanm: snipped

  4. […] the news when it supports the consensus, but not when it does not (e.g. Antarctic warming, hurricanes and Himalayan glacier […]

  5. […] Hurricane Katrina (p.21). Latest evidence is that there is most likely a link between global warming and hurricanes, but the nature is unknown. It may be temporary whilst […]

%d bloggers like this: