RealClimate logo

Fact, Fiction, and Friction in the Hurricane Debate

Filed under: — group @ 18 August 2006

Michael Mann and Gavin Schmidt

Judith Curry and colleagues have an interesting (and possibly provocative) article, “Mixing Politics and Science in Testing the Hypothesis That Greenhouse Warming Is Causing a Global Increase in Hurricane Intensity” in the latest issue of the Bulletin of the American Meteorological Society (BAMS). The article provides a solid review of the recent developments in the science focusing on potential climate change impacts on tropical cyclones. However, the article is more novel in its approach than the typical scientific review article. For instance, it attempts to deal with the issue of how one should test hypotheses that reflect a complex causal chain of individual hypotheses. This is of course relevant to investigations of climate change influences on tropical cyclone activity, where one is attempting to connect a phenomenon (climate change) that is global in spatial scale and multidecadal in timescale, to a phenomena that is intrinsically “mesoscale” (that is, spans at most hundreds of kilometers) in space and lasts only a few days.

More unusually, the article also takes an introspective look at the role of scientists in communicating societally-relevant science to the public, and provides a critical review of how the science dealing with climate change impacts on tropical cyclones and hurricanes has been reported in the media, and how that reporting has occasionally deepened the polarisation on the issue. In doing so, the article revisits some of the “false objectivity” problems we have talked about before (see here and here). They also assess fairly the quality of the arguments that have been made in response to the Emanuel (2005) and Webster et al (2005) papers in the hope of focussing discussion on the more valid points, rather than some of the more fallacious arguments. The article is unapologetic in advancing their particular point of view, and while we generally share it, we imagine that some readers may disagree. We hope, as we suspect the authors do as well, that it will in any case generate a productive discussion.

243 Responses to “Fact, Fiction, and Friction in the Hurricane Debate”

  1. 151
    Stephen Berg says:

    Please change the upload link above to the following:


  2. 152

    #144 John, I know that AGW is a tricky thing to explain, and so we do, as best as we can. People in my own private very little surveys from two climatic zones are overwhelmingly aware that something is wrong with the weather, wrong or different. Their first questions are: “why?” Answers can be found everywhere else, from institutions to gossip papers, the disproportionate quantity of theories is healthy, but comes a time when they have to be narrowed down a bit, this is why I am in favor of fierce, sure to be very popular TV debates, along with data, super bright professors, especially including those tagged as the “regular contrarians”, in order to reduce the noise and increase the knowledge ratio. Looking forward to see authors of important papers such as those mentionned above. The more popular, the better, I am sure that Crichton can bring out his ideas on the fore, and then regain his reputation as a science fiction writer.

  3. 153
    Steve Bloom says:

    Re #143: Dave, you’re being a little fuzzy on most of what you say.

    “An unspoken assumption from Dr. Tricca?” Citation, please? In any case maybe he ought to speak.

    “That there is a one way associative cause and effect logic chain would appear to specify that increased SSTs are the result of increased ambient GW.” (I assume you mean GT here.) Huh? I thought I just finished telling you that wasn’t the case. Can you quite some language from the paper to the contrary?

    “(…) to claim a correlation where it is more likely an association is my primary concern.” I don’t know much statistics, so perhaps you could enlighten me on the formal distinction between a correlation and an association.

    Regarding the data set, you present no specific argument why earlier data is insufficient. If you look at the HURDAT documentation, Landsea seems to think there’s no fundamental problem. Do you have a quote from him or anyone else to the opposite effect?

    On the SST data issue, it appears you conceded my point.

    Generally you seem to throw in a lot of extraneous material, e.g. “(…) clear indication that we really may not understand the adiabatic character of water vapor nearly as well as had been indicated in the past(.)” Huh? Less of that and more citations/quotes of peer-reviewed work would be helpful.

  4. 154
    Steve Bloom says:

    Re #150: Harold, as I understand it the big advantage of sticking to the satellite observation era as the Webster group does is that it makes it more possible to have a consistent data set consistently applied through the whole record. From what Judy indicated at least part of that will be available soon, and I suppose the rest will closely follow. In the meantime the obvious presence of the inconsistencies you list doesn’t seem to have kept the reality-challenged crowd over at Climate Audit from engaging in yet another Triumph of Pure Statistics Over Applied Science by means of HURDAT. Let those guys loose with a data spreadsheet and they’ll overturn all of climate science in an hour or two! :)

  5. 155
    Steve Bloom says:

    Re #138: Global warming has various consequences. Some are useful as metrics (literally to answer the question of how much global warming do we have). Others are not. We can measure global surface temps and say we have .6C (or whatever the current number is) of warming. The same can be done with ocean temps. We can’t say, e.g., “We had 8 category 4 cyclones worth of warming this year” (or whatever). Well, maybe we could, but I hope that just by putting it in those terms I’ve made it obvious how it would be less than useful. There are other non-temp quantities that might be a bit more suitable, e.g. growing season length, but they too have obvious limitations.

  6. 156
    Harold Brooks says:

    Re #154: Steve-it makes it _possible_ to have a consistent data set, but unless there’s been a analysis of the satellite data to make the estimates consistent, the data set won’t be. Changes in practices over time and forecast center means that using the so-called best-track data won’t be consistent, unless those data have been reanalyzed for that purpose.

    As an example, Knaff and Zehr (2006, in press) have shown that the Atkinson-Holliday wind-pressure relationship that the Joint Typhoon Warning Center used to use to estimate winds resulted in an underestimation of winds in the western North Pacific from 1974-1987. (A different WPR was used before that from 1966-1973. It’s not clear what it was, but it also had underestimation problems.) Reanalyzing the west Pacific data, using a new WPR and environmental wind pressure data from the NCAR/NCEP reanalysis, Knaff and Sampson get 93 cat 4 or 5 storms from 1975-1989, instead of the 75 from the Webster et al. dataset (there’s an error in Table 1 of Webster et al. giving that number as 85). The fraction of all typhoons that are Cat 4-5 in 1975-1989 in the reanalysis is 39%, instead of 26% from Webster et al. Since the reanalysis doesn’t change the 1990-2004 value (42%), the increase in fraction of events that are Cat 4-5 is much smaller in the reanalysis.

    The 1970-1974 count of Cat 4-5 is 34, instead of 30. They also use data back to 1966, when the satellite era began in the west Pacific. They show a linear trend (the fit isn’t good) from 1966-2005 of an increase in Cat 4-5 events of 0.024 per year instead of 0.086 per year in the pre-reanalysis values.

    The limitation to satellite era also means that one can’t compare the current high-event frequency period in the North Atlantic to the previous ones, unfortunately. The dramatic changes in data content between those eras (satellite, more frequent recon, etc.) make those comparisons difficult to do in any case, so that it’s not at all clear what the best approach is. If we had satellite data or flight recon data on every event back for a century, the problem would be easier.

  7. 157
    Bryan Sralla says:

    Re: #155 Steve Bloom, on your point, it does not seem that we really have any disagreement here, but somehow I am not feeling the love.

  8. 158
    ike solem says:


    See my post at

    RE #141

    The issue of concern here (in my initial post) is what might happen to bottom water formation on a global basis due to anthropogenc global warming. Low-oxygen bottom waters mean that the microbial population of the deep ocean is limited in its ability to oxidise photsynthetically produced organic matter back to CO2; thus the sediments accumulate a higher proportion of organic matter. Those sediments may go on to form petroleum source rocks. This is very relevant in that it is a mechanism that relates biological photosynthesis to deep-water carbon burial, something that is thought to be of limited importance in the current ocean system. An introduction to studies like the Global Ocean Circulation Experiment (GOCE, HOTS, BATS, etc.) can be found at

    The role of the oceans in climate variability is due to their sequestration, transport and storage of carbon and heat. Reduced ocean circulation will mean less ability to absorb both carbon and heat, as well as lower levels of oxygen in the deep ocean, and yes, that will lead to conditions conducive to source rock formation.

    So, what is the problem with this? Why did you say that “You are so far off base on this, that it is really hard to even begin to take what you said seriously. Wrong tectonic setting, wrong source rock environment, wrong reservoirs, wrong structural setting. Try again! Global ocean stratification forming oilfields? I have never heard of this in my career as a petroleum geologist. There is way too much junk science on this website.”

    Not only did you attack my comment on the basis of my fairly offhand reference to hydrocarbon formation, (my intial comment primarily related to the dead zone off the Oregon coast and its possible causes and effects) you used my comment to claim that this website, which is one of the few places where unbiased scientific discussion of global warming appears outside of strictly academic circles, has “way too much junk science”.

    You are the same Bryan Sralla who heads Hewitt Minerals and has recently been involved in bidding on oil leases, are you not? If I’m wrong about that, please accept my apologies. In the interest of complete disclosure, I am very interested in seeing renewable energy companies take over the energy markets from the fossil fuel sector, though my personal financial involvement is nil, at this point. Cheers!

  9. 159
    Steve Bloom says:

    Re #157: If we don’t have any disagreement *now*, that’s nice.

  10. 160
    Steve Bloom says:

    Re #156: That’s very interesting to hear about, Harold. So it sounds as if, from what Judy and Chris Landsea (in his recent Science piece) mentioned, there are at least three separate reanalysis efforts in various stages of completion. I guess I will be so bold as to ask where you think all of this is headed; i.e., is it possible that these efforts will produce the additional 150 early strong storms Peter Webster says are needed to reduce the apparent trend to insignificance?

    This may be a stupid question, but wouldn’t it be possible to get a more or less internally consistent analysis going back to early satellite days by reducing the quality of more current data (photos, as I understand it) to the same level as that of the early stuff? In other words, if what you have from 1970 are grainy photos from certain angles at certain intervals, couldn’t more current data be processed into the same format and quality? The results might not be correct relative to modern data, but as long as they were consistent wouldn’t that be useful for identifying trends?

  11. 161
    Bryan Sralla says:

    Re: #158, Ike, your e-mail has not shown up yet. The reason for my junk science comments:

    You compared the modern environment off the Oregan coast to middle East source rock formation. In this regard, your statements were indeed junk science in the classic sense of the word. I have studied the Middle East source rocks and oil fields. Have you? In what area is your academic and professional training? Biology? Here is some homework Mr. Biologist.


    Ala, M. A., R. R. F. Kinghorn, and M. Rahman, 1980, Organic geochemistry of the Zagros petroleum province southwest Iran: Jour. Petroleum Geology, v. 3, p. 61-89.

    Ayers, M.G., Bilal, M., Jones, R. , Slentz, L., Tartir, M., Wilson, A., 1982 Hydrocarbon Habitat in Main Producing Areas, Saudi Arabia AAPG Bull, v.66, p.1-9

    Arabian American Oil Company Staff (Aramco), 1959, Ghawar oil field, Saudi Arabia : AAPG Bull., v. 43, p. 434-454.

    Dow, W. G., 1978, Petroleum source beds on continental slopes and rises: AAPG Bull., v. 62, p. 1584-1606.

    Dunnington, H. V., 1967, Stratigraphic distribution of oil fields in the Iraq-Iran- Arabia basin: Inst. Petroleum Jour., v. 53, p. 129-153.

    Kamen-Kay, M., 1970, Geology and productivity of Persian Gulf synclinorium: AAPG Bull., v. 54, p. 2371-2394.

    Kent, P. E., and H. R. Warman, 1972, An environment review of the world’s richest oil-bearing region–the Middle East: 24th Internat. Geol. Cong., Montreal, Proc., Sec. 5, p. 142-152.

    Kirkland, D. W., and R. Evans, 1981, Source-rock potential of evaporitic environment: AAPG Bull., v. 65, p. 181-190.

    Law, J., 1957, Reasons for Persian Gulf oil abundance: AAPG Bull., v. 41, p. 51-69.

    Lees, G. M., 1950, Some structural and stratigraphic aspects of the oil fields of the Middle East: 18th Internat. Geol. Cong., London, Proc., pt. 6, p. 35-44.

    Murris, R. J., 1980, Middle East: stratigraphic evolution and oil habitat: AAPG Bull., v. 64, p. 597-618.

    Murris, R. J., and K. de Groot, 1979, Oil habitat of carbonate provinces: Cong. Panam. Ing. Petr., Mexico, Proc., sec. 1, paper 5.

    Powers, R. W., 1962, Arabian Upper Jurassic carbonate rocks, in W. E. Ham, Classification of carbonate rocks: AAPG Mem. 1, p. 122-192.

    Powers, R. W., 1968, Saudi Arabia , in L. Dubertret, Lexique stratigraphique international, v. 3, Asie, f. 10: Paris, Centre Natl. Recherche, Internat. Geol. Cong. Comm. Stratig., p. 177.

    Steineke, M., R. A. Bramkamp, and N. J. Sander, 1958, Stratigraphic relations of Arabian Jurassic oil, in L.
    G. Weeks, ed., Habitat of oil: AAPG, p. 1294-1329.

    Tissot, B., et al, 1974 Influence of nature and diagenesis of organic matter in formation of petroleum: AAPG Bull., v. 58, p. 499-506.

    Waples, D. W., 1980, Time and temperature in petroleum formation: application of Lopatin’s method to petroleum exploration: AAPG Bull., v. 64, p. 916-926.

    Wilson, J. L., 1969, Microfacies and sedimentary structures in “deeper water” lime mudstones, in Depositional environments in sedimentary rocks: SEPM Spec. Pub. 14, p. 4-16.

    After you finish reading these, I have a few more. If you choose to complete this study, and still hold to your opinions, then we will talk. Otherwise, for professional reasons, my comments must stand. This will be my last public comment to you on this matter. You can continue to badger me and google search all you want, but I will not be responding further in this forum. My e-mail is always open. I certainly hope we can be friends.

    Best Regards, Bryan

  12. 162
    Hank Roberts says:

    Quick Study
    Hurricanes: Tempests in a greenhouse

    Greenhouse gases make Earth’s surface hotter than it would be if the planet were simply a blackbody radiator. That additional warming is an important driver of hurricanes.
    Kerry Emanuel
    August 2006, page 74

  13. 163
    L. David Cooke says:

    Ref #153

    Dear Mr. Bloom;

    First let me share that I do not believe I have been “fuzzy” in any of my communications. Based on the data references, I believe that I was quite clear, evidence for my statements are supportable within my list of references of which I will share a sub-set, (Note the references below we accurate as of April 2006; however, you will have to research your own answers as this is data and not a study that spoon feeds you.)

    The difference between an association and a correlation is twofold, The first characteristic of the data is that one variable should be independent and the other a dependent, meaning, one describes hypothesis the other supports the hypothesis. (This is where the one way truth statement comes into play.) Secondly, though two variables may be changing in the same direction at the same time does not mean there is a dependence relationship to the data.)

    As an example: Households in the US have a high percentage of multiple television sets; Households in the US generally enjoy a higher level of health compared to most third world households. (Note, two simple casual variables that are significantly provable.) However, to draw the conclusion that the multiple televisions are resulting in the greater general health of US households is likely a false conclusion.

    However, if your study was that there were a high amount of health related television shows observed in US households. And that there were multiple television sets in US households. To draw the conclusion that there was an increase in the general health of US households where these shows were observed as opposed to households that did not observe these shows would be a significant study and likely true.

    And finally, since we are discussing hurricane development here (You know a reference to the original issue rather than a discussion off the subject.) and the creation of Tropical Storms. As the development of a Tropical Storm relates to water vapor rising cooling and precipitating out. This transition of state of water relates to the basic physical processes of adiabatic cooling and heating of water vapor. In the CloudSat and Calipso data it is clear that in a Tropical Wave this process somehow does not apply. Funny, that even the saturated adiabatic process does not seem to operate well, at least based on the visable interior view of a tropical storm, I wonder what a IR view would show?

    I had debated if it was worth while to reply to your message for sometime, and when my system went south it appeared that the lack of response was decided. However, I was able to revive a fallback system and today have observed your process of engagement and though you appear to be confrontational you do do indicate an interest to learn. Hence, though I am not qualified to teach, I gladly share this basic sub-set with you so you can begin to learn on your own. If you are willing to apply yourself you might even be able to figure things out for yourself and can stop relying on someone elses studies.

    Note that the first three links under Atmospheric Temperature and Clouds relate to most of what I have shared to date. The rest are good references of participants I have observed as important in understanding how natural processes can effect weather, Good Luck!

    Aerosol Contributions

    [edited for conciseness]

    Dave Cooke

    [Response: Complete downloads of every site you’ve ever read is neither helpful nor welcome. If you want to discuss these kinds of issues here, the watchword is focus. – gavin]

  14. 164
    Harold Brooks says:

    Re #160: I’m not sure where things will end up. It’s enough outside of my specialty that I’d hesitate to predict. As far as I know, all of the people working in the field agree that the mid-70s-early 90s (with a couple of years on either side, depending on the details) had fewer tropical storms than the recent years. The questions are how much less, attribution, and what happened prior to that. Knaff’s work in the western Pacific suggests that the late 60s-early 70s max was about the same or slightly lower than the recent period. The trend from Knaff’s version of the west Pacific is pretty small (~0.024 major hurricanes/year, which might be more consistent with the prediction of several percent increase in wind speed for a 2 K SST increase from Knutson and Tuleya’s work-Webster et al. cite a 0.5 K increase, so I’d expect ~1-2% increase in the winds for that SST change, which would be hard to see in the storm record).

    Consistent data collection is probably more important, in some ways, than accurate data collection, so moderate quality satellite images might be OK. I don’t know the details of how good of an image is necessary to get a decent estimate from satellite. I know that the angle of viewing can be a problem.

    It would be nice to have a global reanalysis of satellite observations. It’s unfortunate that the time that I think would be the most interesting to know about in the north Atlantic (the last peak period) is prior to satellite. I’m not sure how much, if anything, can be said with a great deal of confidence about trends in events based on the observed record of events in databases such as HURDAT. There are great uncertainties in the estimates and I don’t see how to get around them. They also don’t include data on possibly important parameters, such as size of the circulation.

    My personal biases make me think that a project that identifies favorable environmental conditions and looks for changes in the historical distribution of those might be a good approach. We’ve had some success recently in trying to apply such a technique to large hail. The hail database is much worse in terms of quality than the tropical cyclone data base, but it is good enough to build fairly good relationships between necessary environmental conditions and events. Then, we can count how often and where those necessary conditions occur using either environmental observations or reanalysis data. The preliminary result shows a decrease in the occurrence of favorable conditions in the US from the late 1950s into the mid-1970s followed by an increase until the late 1990s of ~0.5-1.0% per year. The late-90s look a lot like the late 1950s. The big stumbling block on the hurricane problem for this approach, to me, is the question of initiation-I can think qualitatively about how to look at SST and shear, so that good environments should be countable, but the initiation question is hard. In general, because of my experiences with the severe thunderstorm databases, I’m much happier looking at environments that would support events of interest than counting raw reports. For hurricanes, that might not work as well because we might be observing the environments all that well. I think it’s a very hard problem and I’m glad other people are working on it and not me.

  15. 165
    Graham Dungworth says:

    Re#161,158 and authors Bryan Sralla and Ike Solem.
    Before I add my 30 minutes pennyworth about the degree of caustic comments,junk science attributes, ridicule etc in these threads let me present my own credentials. Google scholar key words “dungworth g geology geochemistry and/or heat flow” as I’ve published in these multidisciplinary fields and worked in the Petroleum industry for many years.
    When you debate source rock formation in the geological environment the timeframe is enlarged. Average depositional rates vary widely in the region of 0.1mm per year. Sedimentary marine source rocks that generate oil and gas were deposited in time frames of millions of years. We combust these carbon sources thousands of fold faster than their earlier deposition rates. The mass of coal deposits exceed oil prone components of marine rocks, the former seams, often 10m in thickness in SE Asia were formed more quickly, each seam over ca. 100000-20000years. Chemically reduced carbon has to be sequestered in the crust to generate the excess of oxygen in the atmosphere. What has been omitted is the role of sulphur S in the crust.Reduced S or sulphide and sulphur have to be sequestered to produce a gross gain of O2. Sulphur is half as abundant as Carbon in the Earth’s crust, mass 2.2exp10^25g. Half of this sulphur is the oxidised form sulphate, found in sabhkas(evaporites) for instance, whereas only ca one fifth of carbon is in the reduced form;the majority is carbonate ie. limestone. Similarly, iron which is far more abundant in the crust, thermodynamically and kinetically prefers to exist as ferric(Fe111) in our oxidising atmosphere unlike its ferrous form in the crust eg pyrite. Biogenic cycles account for there far from thermodynamic equilibria in the biosphere.
    Climate change deniers often express that conditions were far warmer in the past. That’s true, but they fail to express the much longer timescales involved. For instance, the absolute pressure of the atmosphere is never mentioned. Higher O2(50% and N2(?%) pressures would create a much warmer climate, but the rate of change is measured over millions of years, something the deniers conveniently forget.
    For the timescale of climate change, perhaps only the prevalence of biogenic gas(methane) as clathrates in permafrost or the cold bottom water sediments represent a significant contributor to catastrophic climate change. Ike’s review of the source rock scenario is basically sound even though he probably lacks the working credentials of petroleum or organic geochemistry, by itself a huge multidisciplinary field.Also, as I’ve pointed out above it is peripheral to the other even more hugely multidisciplinary field.
    In my time I’ve also published in biochemistry and even heat flow. If I want to keep up with developments in these fields where do I go? Even universities do not maintain comprehensive coverage of hundreds of journals, even if we live in a university town. takes 3 to 4 Internet coverage is only limited to abstracts. It takes 3 to 4 years to get a basic degree. That reference list is for post graduates. It would take 1 year for an MSc candidate to read it, think about it and then perhaps spend another 3 years to get a PhD. If you stay the course you still haven’t a working knowledge re- a post doc publication record. It can take another 5 years to gain an international reputation yet still the chances are one doesn’t have tenure anywhere. All these years are spent on a minimal salary. By the time I had tenure as a lecturer I was aged 32 and disillusioned by across the board cuts of grants, nationally. All of this is specific to a single discipline. How long does one need to publish in a multidisciplinary field? Let me further add that often vitriolic debate at conferences is the venue for a multidisciplinary endeavour; it’s common to demand credentials and infer vested interests as was asked of Bryan but it’s underhand even though it is common practice amongst AGW debaters.
    What is unique about is the forum for real scientific debate exists no where else other than the private science forums, unavailable to the interested public. There are other forums but none which attract such a wide diversity of opinion. I hope you both can settle your differences and feel a measure of pride that others appreciate your posts.
    Finally, I’ve known a couple of Nobel Laureates and many other international gurus who accused each other of promoting junk science,that they had nothing useful to contribute, you stick to astronomy and we will stick to oil generation, you have nothing worthwhile to contribute,he never said that;it’s a lie I have the paper here to prove it, let’s face it other than physics there are a huge number of things that you know absolutely nothing about, get cycling Otto, shut up!
    So finally finally, I wouldn’t worry about the level of too heated debate in a scientific field that has becomevery politicised.All the chicanery and sculldugerry is still as relevant today as it was in yesteryear; drat all 40 minutes worth of my pennyworth!

  16. 166
    m.j. says:

    global warming is in words from a 14 year old high school student bull sh*…is it though..idk because i cant find 1 good site with a debate about it…so help!

  17. 167
    Bryan Sralla says:

    Re: #165 Thank you for your reconcilitory words. You said: It takes 3 to 4 years to get a basic degree. That reference list is for post graduates.

    This is exactly why interdisciplinary discussion is so difficult. We all have a great deal of time, effort, and sacrifice devoted to our areas of specialty (skin in the game). It is hard to have that invalidated by offhand comments from either side of a debate. It hurts feelings, and ruffles pride.

    To Ike, I sincerely would like to agree to disagree and move on to more relevant discussion of climate change. Contrary to popular belief, I am not a denialist. Your comments about sapropel and the basic workings of source rock genesis were indeed valid. You are obviously very knowledgeable in this area and others. It was only your application that I disagreed with.

    To clarify one additional point. I am not the head of Hewitt Mineral Corporation (I wish). I am a lowly staff geologist. My opinions are my own, and in no way should be taken to reflect the views of my employer.

  18. 168
    L. David Cooke says:

    Ref #163 Dr. Schmidt Inline Comments

    Dr. Schmidt;

    Point taken, my apologies. The specific references are as follows:'Umberto%20Triacca
    “Global warming and carbon dioxide emission; An Empirical Study”, Dr. Linyan Sun and Dr. Muhong Wang, 1996, The Journal of Environmental Management, Vol. 46 Issue 4 Pages: 327-343

    I erred in trying to share the basis of a case for non-anthropogenic GW causes here. This will not occur again.

    Dave Cooke

  19. 169
    Steve Bloom says:

    Jim Elsner, Judy Curry and Kevin Trenberth talk about hurricanes, global warming etc. on NPR Science Friday today. Download available here after 6:00PM ET.

  20. 170
    Jerry Steffens says:

    After reading some of the recent comments, I decided to refresh my memory on the “comments policy”. In particular, these two seem to be relevant:
    (3)”Only comments that are germane to the post will be approved.”
    5) “No … you said/he said type arguments are allowed. This includes comments that (explicitly or implicitly) impugn the motives of others, or which otherwise seek to personalize matters under discussion.”

  21. 171

    Because there have been fewer tropical storms this year, the latest, Ernesto, has the initial letter E rather than the K for Katrina for the hurricane at the same time last year. However the area that is going to suffer from Ernesto could well be the same as suffered under Katrina. See:

    Ernesto is heading for New Orleans, and “the vulnerable Galveston/Houston area” as the hurricane chief called it. See: FEATURE-Worst is yet to come, US hurricane chief says

  22. 172
    Dan says:

    re 171. I understand your point but please keep in mind that the average hurricane forecast track error three days out is on the order of 300 miles (thus the “cone of error” indicated on NHC charts). So no one should be pinpointing locations now. At best this early, anywhere in the Gulf of Mexico is under a possibly threat.

  23. 173

    Re 172 As I recall the 3 day track that was forecast for Katrina was pretty accurate, and at present if Ernesto continues on this path which has held steady for at least the last 24 hours, then it will hit the northern coast of the Gulf of Mexico. That is where New Orleans and Galveston-Houston are. Let’s hope it misses them.

  24. 174
    Graham Dungworth says:

    Re#171 This prediction is premature. If you look at the loop for the Western Atlantic and toggle TFP at the top-
    and then go to the active predicted loop for ERNESTO-
    you will see that the live actual path has already deviated. If you substitute DEBBY in the url on the browser you can see how the actual path has deviated from the daily updated predicted paths. Also, Ernesto is still a tropical storm, not yet a hurricane so it is too early to predict landfalls. Additionally, Ernesto was spawned in the same locale just to the east of the Windward Islands as that for Hurricane Frances in 2004.
    Finally, if you want to see all former tracks of hurricanes globally visit-

  25. 175
  26. 176
    Graham Dungworth says:

    Re#71&73 Overnight Ernesto has battered Jamaica with up to 70mph winds . Cuba is next and the predicted course has already changed to put Florida at possible risk. How do you quantify risk of Hurricane damage when the Ernesto storm might fizzle out or never become a hurricane? Positions of possible landfall are highly unpredictable.

  27. 177
    L. David Cooke says:

    RE #176

    Hey Graham;

    I don’t know that the possible landfall is highly unpredictable unless you are considering the level of accuracy. If you remember the original suggestion was that Ernesto would reach Hurricane strength by Tuesday and almost overnight it reached this boundary.

    By the same token originally the path was to the upper Gulf Coast and based on the current air pressure patterns the path looks to be changing hourly. For instance based on the URL:

    seems to point to a significant change in the last 8 hours. A floor to ceiling cyclonic event over the Alabama/Georgia area seems to have moved south slightly and a upper level high pressure appears to be forming off the coast of Virginia and the Carolinas. At the same time there appears to have been a loss of the Western Pacific Hurricane replaced by a new Tropical storm forming off the Baja coast. The low pressure area over Oklahoma and Arkansas seems to be losing some of it’s moisute content. And finally a low pressure zone appears to be forming off the coast of the Yucatan. The question is the low pressure of the Yucatan actually a different event or the top of Ernesto which seems to have most of its energy pushed to the east and south?

    Based on the observation it appears that the Gulf high pressue is more likely to remain in place or possibly retrograde. If this happens then the possibility of Ernesto moving further west may not be likely. Outside of this view is the precipitative water moisture over the area of the southern Central American and northern South American areas. If there is a strong low pressure zone then the possibility of a low southern movement of Ernesto is likely. If as the indications of the prior 12 hours are accurate then it is more likely to expect a strong northward movement.

    The question is are you saying the direction of a storm is not predictable with any accuracy based on current observation techniques? Or are you saying that through out the life cycle of the storm, being able to predict with a level of +/- 5% the precise path is not discernable?

    As to strength would this not be a factor of the surrounding pressure zones. Meaning if there is a strong outlet of warm surface air from Ernesto should there not be a strong inlet from an adjacent high pressue zone. And would the strength of Ernesto not be co-dependent on the adjacent pressue zones?

    Dave Cooke

  28. 178
    L. David Cooke says:

    Ref #177


    My Error; The statement,

    “A floor to ceiling cyclonic event over the Alabama/Georgia area seems to have moved south slightly and a upper level high pressure appears to be forming off the coast of Virginia and the Carolinas.”

    Should have been:

    A floor to ceiling ANTI-cyclonic event over the Alabama/Georgia area seems to have moved south slightly and a upper level high pressure appears to be forming off the coast of Virginia and the Carolinas.


    Dave Cooke

  29. 179
    Karl Hoarau says:

    The most important parameter is the maximum surface sustained wind when you are analysing the tropical cyclones intensity with the satellite pictures. Remember that 879 hPa match with 155 kt in the pressure-wind relationship used in the Indian Ocean, the South Pacific and the western North Pacific. The surface pressure in the eye depends on the environmental pressure of a cyclone and depends on the size. Monica formed in April, at the time of the year when the environmental pressure is rather high. Moreover, Monica had a rather small central structure. It is why for a given sustained surface wind, you can get two different pressures for two cyclones. Remember the North Atlantic with Georges at 937 hPa/135 kt in 1998 and Floyd at 921 hPa/135 kt in 1999.

  30. 180
    Karl Hoarau says:

    RE #69
    As one of the authors of the paper with Chris Landsea, Bruce Harper and John Knaff (2006), I shall comment only the last part refering to the “global data since 1970”. I am afraid that there are more than anecdotal reports that some storms have been misclassified even if Dr. Curry told the contrary. She refered to a study made by Bruce in the area 90°E to 135°E. But she did not know that Bruce and the Bureau of Meteorology did not have the satellite pics before 1980 to reanalyze the cyclones. Moreover, [edit] the Cat 4& 5 TCs given by Webster et al.(2005) concerning the North Indian Ocean(NIO) and the South Pacific(SP), from 135°E and eastward, are completely wrong : only 1 for the 1975-1989 period and 7 for 1990-2004/ and 10 for 1975-1989 and 20 for 1990-2004, respectively. Actually, I made a reanalysis from the original satellite pictures :
    For the NIO : 7 TCs at Cat 4 & 5 for 1978-1989 and 9 for 1990-2004
    For the SP : 23 for 1979-1989 and 24 for 1990-2004
    As you can see, I did not have satellite data for three or four years period. But, we already lost the trend that Webster et al.(2005) thought to have found ! I already began a reanalysis for the western North Pacific (not completed) and I already found 99 TCs at Cat 4 & 5 for 1979-1989 while Webster et al.(2005) found 116 for the 1990-2004 period(only + 15% for the moment). I think that in the middle of 2007, my global reanalysis will be finished. But I can already tell that there was no near doubling trend for the TCs at Cat 4 & 5 between 1975-1989 and 1990-2004. Therefore, there is no inconsistencies in our group [edit].

    Karl Hoarau
    (response to comment 69)

    [Response: Comment was edited to preserve the information, but to remove personal comments. Dr. Hoarau, your comments are welcome, but please note that this topic is already excessively personalised and we won’t be adding to that here. -gavin]

  31. 181
    Graham Dungworth says:

    Hello Dave-sorry I’ve been out all day. I don’t know how to answer your questions. You have raised 5 conditional statements and then summarised with a supposition about conditions over the Yucatan.
    The direction of these storms/hurricanes is predictable.Last year a dozen hurricanes tracked beyond Cuba.In 2004 Florida suffered badly with several landfalls. Currently, today, Ernesto became a category 1 hurricane but its path has diverged greatly from the earlier 5 day forecasts.This year hurricane Chris, developed in the same area as Ernesto but dissipated just to the north of Cuba and Haiti. One cannot a priori predict the lifetime of a storm/hurricane. A 5 day prediction of the landfall of Chris or the accuracy of its path track is conditional ie. were it to evolve. The 5 day prediction is also updated daily. The unisys data concerns storms that developed into hurricanes and does not include storms that didn’t develop into hurricanes.
    As to the theory of the quantitative strength of hurricanes I have learnt much from the early posts on this thread and haven’t anything of value to add.

  32. 182
    Judith Curry says:

    Re #180: some clarifications on the Landsea et al. Science Perspective. It is my understanding that this was originally submitted to Science (with a total of 11 authors) as a Report, where it did undergo peer review. It was eventually published as a Perspective, with 4 authors. The analysis undertaken that Mr. Horau refers to is not anything close to adequate for establishing this version of the tropical cyclone data record for use as a climate data record. Rather, it points out (in an anecdotal way) that there are reasons to question the raw data used by Webster et al. Numerous efforts are underway to reanalyze the global tropical cyclone data; the studies that use the most rigorous and scientifically defensible methods, combined with thorough documentation, will eventually become accepted as a climate data record after careful scrutiny by interested scientists in the relevant fields. I suspect that we are a few years away from an accepted climate data record for global tropical cyclones, and I expect that Jim Kossin’s analysis since 1983 will be the first important word on this subject, but not the last word.

  33. 183

    Dear Mr. Hoarau,
    Your reanalyis will be most welcome. But it is important that it is undertaken in a manner that enures objectivity. Please rember that our initial analysis was undertaken in the spitit of skepticism regarding Kevin Trenberth’s comments in early 2005. Your reanalysis was motivated because I believe (as you have indicated in countless comments to the tropical storms list) you could not believe that we could be right although this sentiment was stated a little stronger. This is why we have advocated an indepenedent reanalysis. I think I have learnt from you that there are many ways of intepreting satellite data. The question is how do we establish this standard? That is where, I believe, all of us can contribute.

    One other issue that is often ignored. The SST is all of the ocean basins has shown a steady and statistically significant trend over the last 30-40 years (e.g., see Hoyos et al 2006). Vertical wind shear is associated with shorter interannual variability. SST, by the way, is an independent data set. Can I turn the question around to you. If the SST has increased almost lineraly since 1970 but there is no trend in any of the other “necessary conditions” how can there not have been a change in tropical cyclone characteristics given what we know about rudimentary physics of these storms? What physical process would you invoke to explain this?

    I would be interested in your response.


    Peter W

  34. 184
    L. David Cooke says:

    RE: 181

    Hey Graham;

    Thank you for the clarification of your concerns about the unisys data, in light of your more recent commment, your original post makes sense to me now. I too have tried to apply the data from various NOAA publicly available data bases and have not been able to develop a clean statistical analysis. (I have difficulty getting access to data that meets statistical test and probability requirements along with setting up linear regressions so that they are valid.)

    The final questions I asked were to try to determine if the observations I offered coincided with yours. I was not casting aspersions, it was a poor attempt to share my understanding and to request a clarification of what you were stating.

    (Part of my observations include the character of a healthy storm Eye diameter -vs- Eye pressures. Over the years I have observed that when the pressure reaches a certain point, it seems that storms with small “eyes” open up. The process of the increasing diameter of the eyewall seems to involve a secondary eyewall forming to support surface air necessary to reach the top of the storm. It is almost like when the SSTs are too warm or the Tropopause is too cool the rising air is so great that the storm seems to self destruct by “eating its own eye”.)

    Thanks for your clarification and your response.

    Dave Cooke

  35. 185
    Urs Neu says:

    Re 156,160,180

    Question concerning the reanalysis:
    As far as I understood from Knaff and Simpson, they used a new method from Knaff and Zehr (2006), which I couldn’t find, because it hasn’t been published yet.
    They did the reanalysis with this new method only to the dataset until 1987. However, they didn’t analyse the data from 1990-2004 with the same method. As far as I can see, Knaff and Zehr only analysed the N Atlantic and E Pacific, but not the W Pacific. So how can we be sure that there aren’t new inconsistencies in the analysis method? Aren’t we still left with two data sets for the two periods compared in Webster et al. which are analysed with different methods?
    A real improvement in estimating trends would be the analysis of the whole data set with exactly the same method.

  36. 186
    Karl Hoarau says:

    I met John (Knaff) at the AMS Monterey Hurricane Conference in last April. And we have had the occasion to speak about the reanalysis that he made with Ray Zehr. The method is a way to estimate the maximum surface sustained wind over one minute which is the most significant data of a cyclone intensity. To get the wind, the authors used the minimum sea level pressure recorded by the reconnaissance aircraft in the western North Pacific, the size of the cyclones and the environmental pressure. Therefore, a reanalysis for the 1990-2004 period is not possible with this method as the reconnaissance endded in August 1987 in the western North Pacific.

    This work is very important for me because amongst my skills, I know how to use the Dvorak Technique (based on the thermal infrared pictures) to estimate the intensity of tropical cyclones. And my results are very closed to those found by Knaff and Zehr (2006): John gave me the intensity (wind)that they found. As I have the pictures for the western North Pacific from 1979 to 1987, I could make a comparison : here, we have two independent ways to reanalyse the intensity and the results did not confirm the trend found by Webster et al. (2005).

    Who has said that we have inconsistencies in our group ?

    Karl Hoarau

    PS : I was initiated to the Dvorak Technique (1975, 1984 and 1995) since 1996 and even if english is far to be my native language, currently I know all the points of the Dvorak’s Technique by heart. [edit – please no personal comments!]

  37. 187
    Judith Curry says:

    Re #186: a few clarifications of why I am unconvinced by by Mr. Hoarau’s analysis:

    1) a credible reanalysis of the satellite TC record needs to start with a reliable data set. The satellite data set needs to be reprocessed, since the satellite data integrity has evolved with time also. This includes consistent calibration of the IR radiances.

    2) your reporting only of storms whose intensity should be higher (e.g. cat 4 and not 3), while not mentioning storms that went the other way (e.g. cat 4 to cat 3), particularly in view of Bruce Harper’s earlier analysis, lends your analysis subject to the suspicion that you have started with a conclusion and then analyzed and reported the data solely to support the conclusion, rather than presented an unbiased analysis.

    Note, in the BAMS article, my testimony, and in many public lectures, I have called for a reanalysis of the global TC data. The Webster et al. paper motivated such reanalyses, and now we can wait for the various reanalysis projects to be completed and to be assessed by a variety of different people. While many of us have been very concerned about Michael Crichton becoming a major spokesperson on the global warming issue, he does make one very important point. People collecting and processing data can interject bias (even inadvertently) into data sets, which is why the medical profession conducts double blind studies. We don’t have that particular option in our field, but we must make every effort to insure that our data sets are not biased by those creating them, and this can only be accomplished by multiple efforts evaluated by a number of different people. There is no “turf” to be had in climate data records; the best version of the eventual climate data record will be the one with the soundest, most objective and most transparent analysis method that stands up to scrutiny by a variety of different scientists.

  38. 188

    Mr. Hoarau,

    It is a well known tactic to deflect criticism or to avoid tough questions by being outraged that someone might think of you as a charlatan. I have no idea where you have got this impression as certainly no one in my group has ever inferred it. But it is a convenient deflector as here I am addressing it. In any event, it is absolutely irrelevant! The issue is whether or not a reanalysis can be considered impartial and independent of one’s mindset prior to the conducting of the analysis. It’s as simple as that!

    Yesterday, I posed a simple question which you decided not to respond. I noted the data (independent data not depending on the identification of the intensity of tropical storms) indicates that there is no trend in all but one of the necessary conditions of cyclogenesis (wind shear, large scale vorticity and etc: see Hoyos et al. 2006) and that these parameters change only on interannual time scales. On the other hand, the one necessary condition that does have a trend is SST. I posed the question of how there could not be a trend in tropical storm characteristics given what we know from maximum intensity theory and from models. If you are correct with your numbers then you have a real physical problem to explain. And an exciting one!

    Be that as it may: back to the reanalysis and why I feel I should be skeptical of the results you come up with. In the Landsea et al. (2006) paper of which you were a coauthor, you state that in the North Atlantic there has been no trend in hurricane characteristics from 1960 to the present; a statement I might add, deplete of numbers or graphs. Yet, if one compares the average annual tropical storm characteristics between the two periods 1944-1964 and 1995-2005 one finds 50% more named storms, 37% more hurricanes, 167% more category 4+5 storms, 55% greater ACE, and 63% greater PDI. So I worry considerably about a reanalysis from someone who is party to this complete misrepresentation of what occurred in the North Atlantic in the last 40 years.

    Further, you dismiss the Harper results (who found random error in the Southeastern Indian Ocean assessment of major storms and no change in the trends found in WHCC) by saying that that the Australian Bureau of Meteorology had no access to satellite data until 1980. This is, of course incorrect. The Australian BoM had satellite data that was used in the tropical meteorological centers after 1968 (see Holland, G. J., 1981: On the quality of the Australian tropical cyclone data base. Aust. Met. Mag., 29, 169-181. Here he looks at the impact of satellite data on the tropical storm identification and categroization after it became available around 1970. Greg or I can supply you with a copy if you would like). Perhaps you mistake when the Dvorak scheme was introduced: that was about 1980, and prior to that the NOAA NESDIS scheme was used. Irrespective, the Harper study suggests that it was irrelevant. But if one is going to reanalyze data that will be considered with credibility then one has to be accurate. In other words, get the Australian data situation correct. This will build up credibility in your efforts.

    Finally, you note that you are an expert in the use of the Dvorak technique. That is well and good although I do suggest you read the logical fallacy of “appeal to authority” in Curry et al. 2006. I do however, look forward to papers in the reviewed literature demonstrating your expertise which I have not been able to find. But please do not expect complete adherence to your reanalysis just because of your expertise anymore than I expect adherence to the conclusions in WHCC because I have written a lot of papers on tropical meteorology and climate. We are all fair game and so we should be! Skepticism makes the world go around. Despite the level of ad hominem attacks that have occurred in this discussion during the last year, getting to the bottom of the problem is the ring we should all keep our eye’s upon.

    Peter Webster

  39. 189
    Urs Neu says:

    Re 186

    Thank you for your comment. Thus we are stuck with two different methods for the two different time periods for the NW Pacific, and still have to wait for a homogeneous analysis.
    Another problem is that Knaff and Sampson ( find the strongest increase in cat.4/5 hurricanes compared to the best track data prior to 1973 , while Emanuel mentions in the supplement to his Nature article ( that the wind-SLP relation used for the best track data of JTWC, overestimated maximum surface wind speed prior to 1973 (Knaff and Simpson only report a “different” relation prior to 1973, without specifying). This seems rather contradictory. Any ideas?

  40. 190
    Karl Hoarau says:

    I was wondering why Mrs. Curry did not speak about the reliability of the data set when she was refering to the statistics of Bruce Harper concerning the reanalysis of a very small sample of australian tropical cyclones ? May be because this study matched with the conclusion of Webster and al. (2005) ? Sorry to tell that the point 1) is very irrelevant while the people who I am working with have checked every thing !!! Moreover, I am working with all the satellites plateform : GOES, Meteosat, GMS, Noaa and DMSP !! And my team checked the differences in comparing the data for the same time !!!
    Your points 2) and 3) are irrelevant too because I quoted figures showing differences with Webster et al. (2005) and I did not quote particular cyclones upgraded from Cat 3 to 4 and vice versa. But when I shall publish a very serious study, you will see that I shall show examples of downgraded or upgraded cyclones !!!
    Mrs. Curry has accused me to begin a study with a well known conclusion : â??you have started with a conclusion and then analyzed and reported the data solely to support the conclusion, rather than presented an unbiased analysisâ??. I was wondering how someone could tell me this as I am working on the tropical cyclones climatology since 1996 !!! And each time, I am telling something, I quoted figures and I shall write them when I â??ll publish results showing that Webster et al. (2005) results are wrong !!!
    If I had wrote the same thing than Mrs. Curry (accusation), I gained the impression that the moderators would have cut my sentence : I know that I am a french !
    But I have arguments stronger than that !!! Sorry, Mrs. Curry, I made my analysis and my figures are very telling and I am quite sure that all the reanalysis will never confirm your results (Webster et al.).
    By the way, I even found cyclones originally classified at Cat 2 which should have been classified at Cat 4 !!!
    Concerning the reanalysis of tropical cyclones by Kossin et al. (Cimss), I already can tell that the results will :
    – not show a near doubling (far from that) in the number of Cat 4 & 5 tropical cyclones !
    – show less Cat 4 & 5 cyclones : when you are reducing the satellite resolution pictures at 8 km, you cannot see the right eye temperature of a cyclone (colder eye temperature). As the intensity depends on the eye temperature and the clouds tops temperature around the eye, this should lead to an underestimate of tropical cyclones intensity estimate by at least 15 kt or 0.5 T-number on the Dvorak scale.
    Another feature to consider, the Cimss reanalysis is an automated estimate (Advanced Dvorak Technique) which is not fully reliable : the cyclones from 75kt to 100 kt are overestimated and the cyclones from 115 kt are underestimated !!

    Karl Hoarau

  41. 191
    Karl Hoarau says:


    Happy to know that you have some doubts about the raw data used by Webster et al.(2005)!!!

    It is very funny that you mentionned that we were 11 authors for our paper : this highlighted the deontology of the reviewers of our paper !

    Do not worry, you will see that I am implicated in official projects of tropical cyclones reanalysis !!!

  42. 192
    L. David Cooke says:

    RE: #183

    Dr. Webster;

    Regarding your thoughts on the SST increases, I have noticed several apparent changes in the period between 1970 and 1980, when I was a resident of Central Fl. Up through about 1969 there was a character of a Fall Tradewind (@16-24 knot) that would normally sweep through from about November to about mid-Deccember, as a general rule, about every other year. (Observations from 1960-1969 , location Melbourne, FL.)

    Based on a comment by FerdEngb. on Ukweatherworld, I think there is a possibility that wind and low humidity could have easily explained the cooling of the Winter Caribbean/Gulf Stream surface waters. This cooler start to the following tropical storm season might have been enough to drop the SSTs and coupled with an approximate 3% higher ITCZ cloud cover would have delayed the surface heating the following Spring.

    Since the late 1960’s through around 1980 this process seems to have been suspended. Is this just a matter of perception on my part? I have seen no studies of this characteristic nor it’s possible change since around 1980. If as Ferdinand seemed to suggest, that the spring SSTs could be dependent on the Southern Trades coupled with the lack of saturation what would have removed this characteristic? Along with this observation was an apparent reduction of the southwestern extension of the Sargasso Weed “Sea”, and an apparent stagnant and persistent Bermuda anti-cyclonic pressure zone that was much closer to the US Atlantic seaboard during this time period.

    If this change were due to a steadily increasing change to a natural process I would have expected to see a trend in which the character changed slowly with increasing periods of doldrums. If the character was an anthropogenic based that the change would be to demonstrate the change occuring faster then via pure natural variability would change. However, the observations I made seemed like a switch being flipped. Are my observations without merit or are they indicitive of a trend that has been missed in recent studies?

    Dave Cooke

  43. 193
    Steve Bloom says:

    Re #192: Dave, what’s the basis for your assumption that the rate of change of a given climate phenomenon is related to whether it has a natural or anthropogenic cause? I hadn’t understood that to be the case at all.

  44. 194
    Graham Dungworth says:

    Re#184 Dave
    Beneath the Sun’s hot ill tempered breath,

    Ernesto rests and gathers strength,
    for 12 hours brooding above the Hispaniola trench,
    the Atlantic Ocean’s deepest 8 km point and Neptune’s lair
    where brooding in the vast abyss,
    he seeths against the warring anger of Janus above and stirs the four winds with his waves.
    Ernesto restless with foe in sight
    oblivion for such Earthly mortals.

    Until we get the science right can we keep an open mind on our predictions(lol).

  45. 195
    Hank Roberts says:

    >192, 193
    I think the fast natural events are fairly obvious, aren’t they? A volcano, asteroid impact, are brief.
    Other than say a change in solar output, or the Solary System moving from clear to dusty space (as I think is expected in a few millenia) what else do we know of in nature that happens as fast and persists as long as the rate of increase of anthropogenic GH gases?

    Aside — thanks to all the climate scientists for participating here. It makes the whole process of doing science real, to see some of it as conversation (and argument). And thanks to the moderators for focus.

  46. 196
    Karl Hoarau says:


    > Yesterday, I posed a simple question which you decided not to respond.
    > I posed the question of how there could not be a trend in tropical storm characteristics given
    > what we know from maximum intensity theory and from models. If you are correct with your numbers
    > then you have a real physical problem to explain. And an exciting one !

    You know that I am not a guy who does go away in front of a debate !!! You will have a response about this very soon !!!

    Concerning the Atlantic basin, you should ask to Chris (Landsea) or John (Knaff) since they are the specialists for this area. Do not worry for me, I am very objective while I am doing my reanalysis. And other analysts could contradict me if they want as they have an access to the same satellite data !!!

    > Further, you dismiss the Harper results (who found random error in the Southeastern Indian
    > Ocean assessment of major storms and no change in the trends found in WHCC) by saying that that
    > the Australian Bureau of Meteorology had no access to satellite data until 1980.

    If I well remember, Bruce Harper already gave you a response on this subject on another forum ! But even if the number of Cat 4 & 5 had really doubled in the Southeastern Indian Ocean, you could not have told that this reflects a global trend (do not forget the part of this basin compared to the rest of the other basin) !
    I exactly said “that Bruce and the Bureau of Meteorology did not have the satellite pictures before 1980 to reanalyze the cyclones” !!! This meant that they could not reanalyse the cyclones intensity before this date as they did not keep the pictures in their archives ! I never said that they did not receive satellite pictures before 1980 !!!

    > Perhaps you mistake when the Dvorak scheme was introduced: that was about 1980, and prior to
    > that the NOAA NESDIS scheme was used.

    Concerning the Dvorak scheme, believe me, I am more informed that you could be !!! The Dvorak’s Technique from the visual satellite pictures has been introduced in 1972 and published in 1975 !!! The thermal infrared Technique has been published in 1984 and updated in 1995 !!!
    Before the Dvorak’s Technique, there was the Vincent Oliver classification and I am quite sure that you are not the right person to teach me how to estimate the cyclones intensity from the satellite pictures !!!

    I do read the Curry et al (2006) paper you are refering to and at the page 1028 (right top), you wrote : the satellite-derived dataset of WHCC showed a global increased in the number of category 4 and 5 hurricanes since 1970 ; I would love to know where you found the satellite data from 1970 and who has done the reanalysis in your team ?
    Do not worry, you will not wait for a long time before reading me on cyclones reanalysis !!! [edit]

    Karl Hoarau

    [Response: Karl, I do not want to have to edit your posts. Please stick to the science and do not cast aspersions on people’s credibility. Additionally, you might find you get a better response if you restrict the use of “!!!” to a minimum. -gavin]

  47. 197
    Karl Hoarau says:


    > I think I have learnt from you that there are many ways of interpreting satellite data.

    You did not catch me right but it is normal : I am not sure that many people are able to understand the Dvorak’s Technique vocabulary at the first time. Actually, you should have understood that, in the Dvorak’s Technique, there are three ways to lead to a T-number which represents the intensity.

    >Can I turn the question around to you. If the SST has increased almost lineraly since 1970 but
    >there is no trend in any of the other “necessary conditions” how can there not have been a
    >change in tropical cyclone characteristics given what we know about rudimentary physics of these

    I am not sure that someone can tell how a condition like the vertical wind shear has evolved since 1970 as there was not very accurate data in the 1970’s or 1980’s like the Cimss chart today !!! It is why I do not trust the results of Hoyos et al. (2006) as I do not trust the results of Webster et al. (2005) !!! But I have more than an idea about the response to your question and you should have a response about that in a publication in late 2006 or at the beginning of 2007.

    Karl Hoarau

  48. 198
    Karl Hoarau says:



    it is curious that Mr. Webster has the right to found my reanalysis “suspicious” and I am not allowed to tell that Webster et al. (2005) and Hoyos et al. (2006) used wrong data. It is not personal attack but it is a critic about the philosophy that the scientist use the data to publish paper.

    If this forum was in France (the first historic country of freedom which helped the USA to become free), no sentence would have been censored like here !!!

    Therefore, this will be my last email for this forum.

    [Response: Karl, you are at complete liberty to discuss any possible issues with the data or Webster’s interpretation of them. But as I said before, personal remarks are not on. One should think of forums like this as an invitation to someone’s home; discussion and even disagreement are allowed, but there are limits which the hosts (i.e. us) set in order to foster substantive discussion. If you don’t like it, feel free to set up your own forum where you can set your own rules. We have found that the rules we impose work reasonably well, and I’m surprised that you feel that you can’t work within them. It is of course completely up to you. (Ça n’a rien avoir avec la liberté, c’est une question de politesse, et ça, je pense, existe en France aussi qu’aux Etats Unis). – gavin]

  49. 199
    Urs Neu says:

    Hoarau presented in 2002 an analysis of satellite pictures of intense cyclones (more than 100 kts) over the south-west Indian Ocean for 1970-1999, stating that “the number of intense cyclones has a tendency to increase ; this is especially the case of the extreme cyclones (120 knots and more) …” (
    His analysis of satellite pictures (done in 1999) showed that frequency of extreme cyclones (120 knots and more, which corresponds more or less to cat. 4 and 5 cyclones) has doubled from 1970-79 to 1990-1999: 6 for 1970-79, 7 for 1980-1989 and 13 for 1990-1999. It is very likely that this trend also can be found for the periods 1975-1989 and 1990-2004, since there were more intense cyclones before 1970-1974 than from 1975-1979 and there were quite a number of extreme cyclones from 2000-2004 according to JTWC.
    ( Thus the trend at least in the South Indian Ocean seems to be robust, even after satellite reanalysis by K. Hoarau.

  50. 200
    L. David Cooke says:

    RE: #193, 194, 195


    The basis of my assumption regarding the rate of change is a reflection of the basis of assumption of the AGW community. A steady linear change input to the thermodynamic radiator should result in a steady linear output. The rate of change indicates the rate of forcing. Natural processes in our current environment, due to natural varibility, would have nearly as many positive as negative forcings to balance out with a slight trend change. Anthropogenic forcings would cause a stronger sign of trend change. And finally, sudden output or reaction as the result of sudden input or action. (Hence, if Neutonian physics is accurate you get out in an energy balanced universe what you put in. Even though there may not be a large like for like energy exchange, there is a remarkable like for like reaction for an action. Now if you are suggesting that Neutonian physics does not apply or that the observations of the AGW proponents are invalid I am interested in hearing about it.)


    In relation to the state of the science you have a valid point. However, there has been a great deal of data collected to point and you would have expected after 40 years that the science would be mature enough to exceed the capability of simple observations of a precipiable water image loop offering more insight then the current sophisticated computer models.
    (As to your prose, I do not know that the depths of resource has near as much to do with the event as the surface heat content. (By my observations, The “Tounge of the Caribbean” or the Hispanolia Trench is typically pretty cool.))

    Mr. Roberts;

    Other then my commments to Mr. Bloom, let me suggest if your concerns of solar input to the “black radiant body” were valid I am sure the solar observitories and satellites would have been “Johnny on the spot” in noting it. Even so the DOE’s ARM and NASA’s Dr. Hansen et al 2005, data does not indicate more then a possible +0.0045% deviation or increase in total reaching the surface. From all the data I have found, the current Satellite data indicates less then a TOA deviation of less then +0.003 deviation. (Every indication of the NOAA evaporation pan experiment deviations appear to have more to do with the change in the surface moisture saturation.)


    I apologize if my simple observations appear to run up against your views of issues. I do not credit my observations being much more then possible miss-preceptions. As to any critism regarding my observations that there is a possible bias to them, I admit I look for data that supports my preceptions; however, IMO that is no different then the character of many of the studies I have seen of late.

    As Dr. Curry suggested maybe a double blind study or take a lesson from the Security Community, with a distribution of data supporting a study being sent blindly for processing. Or as suggested during the 27 July 2006 Congressional Committee on Commerce and Energy hearing, that a central global data base be created and the work of all these recent studies be applied to try to complete a long term data table that represents the relevant data points.

    In the meantime, I wait to see the detailed analysis from the CloudSat and Calipso programs as it helps provide a possible 3-D visual view of the water vapor cycle. I hope that including a similar (rather then a LIDAR) a vertically sweeping IR view on the next “bird” would help with the understanding the heat distribution of the process. I just hope I am around to see it…

    Dave Cooke