————————–
Here is my take on how Holland/Webster represented some of points in the literature that Pielke brought up. I agree with some, disagree on some, and am just confused on others:

Roger Pielke, Jr. wrote:
> Mischaracterization #1. Misreporting the literature.
> You [Holland/Webster] write: “Landsea et al. [2006] also state
> unequivocally that there is no trend in any tropical storm
> characteristics (frequency or intensity)after 1960, despite this
> being established in earlier papers by Emanuel (2005) and Webster
> et al. (2005), and more recently by Hoyos et al.(2006).”

Landsea et al. (2006) did indeed state that there was no trend in Atlantic tropical cyclone activity. This was based upon the analysis of Atlantic PDI from 1949 to 2004 shown in Landsea (2005, Nature). We went on to say: “However, differing results are obtained if large bias corrections are used on the best track databases (Emanuel 2005), although such strong adjustments to the tropical cyclone intensities may not be warranted (Landsea 2005)”. Bringing up Webster
et al. (2005) here seems a bit odd as everyone agrees that there’s been a big increase in Atlantic activity between the 1970s and the last ten years. The real issue is whether there’s a trend between the mid-20th Century (the last active era) and the last decade. Webster et al.
didn’t address that because it did not (by design) go back far enough in time. The Hoyos et al. paper probably shouldn’t be mentioned here at all, as it did not further the analysis of TC trends (frequency/intensity) beyond that shown in Webster et al as it was focussed upon changes in SSTs, shear, stability, etc since the 1970s. Holland/ Webster – Please clarify that Landsea et al. weren’t disregarding other papers, but that there is legitimate controversy in how the bias-removal for the Atlantic should (or should not) be done.

> Mischaracterization #2. Mischaracterizing the literature.
> You [Holland/Webster] write: “The overall Landsea et al. [2006]
> analysis is curious and is based on the premise that the data must
> be wrong because the models suggest a much smaller change in
> hurricane characteristics relative to the observed SST warming
> (e.g., Henderson-Sellers et al 1998).”

This is a mischaracterization. Nowhere in Landsea et al. (2006) did we state or imply that simply because there’s an inconsistency that the observational studies must be wrong. We did point out the inconsistency exists between the Emanuel/Webster et al. papers and the theory and modeling, as it is very important to resolve why
this is the case.

> “In contrast,Michaels, Knappenberger and Landsea (2005) argue the
> opposite, that the models must be wrong because they do not agree
> with the data.”

This is also a mischaracterization, as nothing opposite was being argued. In Michaels et al., we suggested that the 5% increase in TC intensity by 2080 due to greenhouse gas warming was overstated
because Knutson and Tuleya (2004) utilized “an unrealistically large carbon dioxide growth rate, an overly strong relationship between sea surface temperature and hurricane intensity, and the use of a mesoscale model that has shown little to no useful skill in predicting current-day hurricane intensity.” Regardless of whether there would be a 5% increase in intensity by 2080 or substantially less than that if Michaels et al. are correct, either scenario still has a huge discrepancy with the Emanuel (2005) and Webster et al. (2005) studies. The inconsistency would just be larger if Michaels et al are right.

> Mischaracterization #3. Selective reporting.
> You [Holland/Webster] ignore relevant work that discusses
> the relationship of models, theory, and observations that includes
> Landsea as an author (if that is your criterion here).

I’m not fussed here. If Holland/Webster decided not to reference Anthes et al. on this point, then it’s fine with me to also not reference the Pielke et al. paper.

> Mischaracterization #4. Misrepresenting the literature.
> You [Holland/Webster] write: “Of greater concern is that
> the conclusions in the Landsea et al. (2006) paper are at odds
> with several previous publications that include the same authors
> (e.g. Owens and Landsea 2003, Landsea et al. 1999), without
> introducing any additional evidence.”

I guess i am confused here as to the point that Holland/Webster are making. Landsea et al. (2006) didn’t address when the starting date for accurate records began in the Atlantic. (Landsea et al. were mainly addressing global TC databases for assessing trends in extreme – Category 4 and 5 – TCs.) Holland/Webster – I suggest just using
Neumann et al., Goldenberg et al., Owens and Landsea, and Landsea et al. as justification for using 1944 as the starting point for having accurate/reliable Atlantic TC frequency/intensity. Landsea et al. (2006) just isn’t relevant for this point.

For those that may be interested in seeing some of the papers mentioned:

http://www.aoml.noaa.gov/hrd/Landsea/landsea_bio.html

best regards,
chris
**********************************************************************
Chris Landsea
Science and Operations Officer
NOAA/NWS/NCEP/TPC/National Hurricane Center
11691 S.W. 17th Street
Miami, Florida 33165-2149
———————————

Thanks much for engaging on the substance! (Note: Judy and I both have the benefit of seeing Chris Landsea’s reactions on the TS list. He has given me explicit written permission to discuss and post them in full here, which I will do soon. Chris has asked me to hold off until the TS List policies are updated.)

A few replies to your comments:

1. As you know, Landsea has a bit less of a problem with this first assertion than I do, if the starting date is 1944. Clearly, it all gets to what date you start with to assess the trends. As you well know none of Emanuel (2005) and Webster et al. (2005), Hoyos et al. (2006) go back to 1960 (each starts after 1970) so to say that these three papers establish trends since 1960 is wrong. Landsea just yesterday stated that he accepts trends since 1970. You are a co-author of Webster et al. (2005) how can you assert with a straight face that the paper shows trends since 1960?

2. On models versus data, you continue to confuse parameterizations with data as input to models. As you know Landsea, agrees 100% with my interpretation here. The passages you cite are identical to the ones I have in the original post. You are not reading what they are saying – what do you think “driven by real-world observations” actually means? My critique is correct.

3. Your quote from Landsea et al. 1999 is incomplete. Did you not read what follows or just selectively report what that paper says? You conveniently neglect the following sentence, which also is footnoted as follows: “This [adjustment to the data] is only a temporary solution, however. What is needed is a “reanalysis” of all available data – primarily aircraft reconnaissance – with today analysis techniques to create an updated data set.” Why reanalyze the data if it is “reliable and accurate”? Landsea et al. 1999 clearly say that the data needs work.

As you know Landsea agrees with most of my critique, and he does disagree on a few points. He has shared these views with the original authors. If they yet decide to make any changes in their text we’d be happy to report that here and compliment them for their positive response.

I do note that even in your response you misrepresent your own work as showing trends from 1960 when your analysis does even begin until after 1970! Webster, Holland, and you are too good of scientists to be so sloppy in your work and how you present it to the public.

Please give the “appeal to motive” a rest. I asserted that Greg Holland’s interest in evaluating Bill Gray’s seasonal forecasts was related to the hurricane-climate debate. How did I know this? Well Greg Holland said so openly at the February, 2006 NSB workshop in Boulder where we both spoke. Further, I note that the issue of seasonal forecasts/global warming and arises again in the Holland/Webster paper we are discussing, so the fact that the two issues are related seems by now absolutely unimpeachable.

You seem to think that you and your colleagues are somehow above critique. I have posted my views of your colleagues work here, backed up with evidence. You may not like the visibility of blogs or the fact that I have critiqued their paper. Scientists should not publish if they do not want their work critiqued.

I said in the main post, that Both Drs. Holland and Webster are widely published and respected scientists with admirable track records. And I repeat that here. This is one reason why it was so surprising to see such a sloppy review of the literature in a paper that is sure to be widely read.

Thanks again!

Neither can I understand how anyone can complain about you making these corrections to their work.

I must admit that I do not understand how these corrections may change the overall view of climate science, if at all. I will say if the authors are refusing to make the changes you propose because they are trying to protect some “conclusion,” then they are no longer scientists.

Here is specific documentation that supports my previous statements that your accusations of misrepresentation of the scientific literature by Holland and Webster are completely unsupported. Below are the relevant citations from Landsea et al. (2006), Michaels et al. (2005) and Landsea et al. (1999).

Holland/Webster: “Questions have been raised over the quality of the NATL data even for such a broad brush accounting. For example, a recent study by Landsea et al (2006) claimed that long-term trends in tropical cyclone numbers and characteristics cannot be determined because of the poor quality of the data base in the NATL even after the incorporation of satellite data into the data base. Landsea et al. also state unequivocally that there is no trend in any tropical storm characteristics (frequency or intensity) after 1960, despite this being established in earlier papers by Emanuel (2005) and Webster et al. (2005), and more recently by Hoyos et al. (2006).”

relevant text from Landsea et al. 2006: “Data from the only two basins that have had regular aircraft reconnaissance flights – the Atlantic and Northwest Pacific – show that no significant trends exist in tropical cyclone activity when records back to at least 1960 are examined.”

Holland Webster: “Figure 1 shows a strong statistically significant trend since the 1970s similar to that found by Hoyos et al. (2006) and Curry et al. (2006). The overall Landsea et al. analysis is curious and is based on the premise that the data must be wrong because the models suggest a much smaller change in hurricane characteristics relative to the observed SST warming (e.g., Henderson-Sellers et al 1998).”

relevant text from Landsea et al. 2006: “Recent studies have found a large sudden increase in observed tropical cyclone intensities, linked to a warming sea surface temperature that may be linked to global warming. Yet modeling and theoretical studies suggest only small anthropogenic changes to tropical cyclone intensity several decades into the future [an increase on the order ~5% near the end of the 21st century.”

Holland/Webster: “In contrast, Michaels, Knappenberger and Landsea (2005) argue the opposite, that the models must be wrong because they do not agree with the data.”

relevant text from Michaels et al.: “In a signal paper on atmospheric chemistry, Elsaesser (1982) asked the question: “Should we trust models or observations?” . . .
When the real (rather than modeled) hurricanes are examined, the correspondence between intensity and SST drops by a factor of 5. . . But when driven by real-world observations rather than unrealistically parameterized and constrained model conditions, the prospects for a detectable increase in hurricane strength in coming decades are reduced to the noise level of the data.”

Holland/Webster: “Of greater concern is that the conclusions in the Landsea et al. paper are at odds with several previous publications that include the same authors (e.g. Owens and Landsea 2003, Landsea et al. 1999), without introducing any additional evidence. These papers state clearly that the author’s considered that the period of reliable and accurate NATL records commenced in 1944 with the implementation of aircraft reconnaissance.”

relevant text from Landsea et al. 1999: “For the whole Atlantic basin reliable intensity measures exist back to the commencement of routine aircraft reconnaissance aircraft in 1944 (Neumann et al. 1993), but even these data have been arbitrarily corrected to remove an overestimation bias in the winds of intense hurricanes during the 1940’s through the 1960’s (Landsea 1993).”

As clearly demonstrated above, there has been no misrepresentation of the literature by Holland/Webster. They have appropriately cited the references needed to support their argument, which is associated with the evolution in thinking on the quality of the data and the reliability of models versus observations. You may have preferred personally to make a different argument and use different references. But Holland/Webster make a legitimate argument and cite the appropriate references in support of that argument. Yes different people read different things into the literature. But such nuances of interpretation do not warrant accusations of misrepresentation of the scientific literature.

Your use of the inflammatory wording of “misrepresentation of the scientific literature” smacks of scientific misconduct. You have falsely accused Holland and Webster of misrepresentation of the scientific literature, and by implication of research misconduct. Such false accusations (so publicly made on your blog) unfairly impugn the integrity of Holland and Webster and, more seriously, unfairly impugn the public perception of the credibility of hurricane and climate research.

This false accusation, when combined with the appeal to motive attack that you made against Holland http://sciencepolicy.colorado.edu/prometheus/archives/scientific_assessments/000860another_problem_wi.html
associated with the AGU Workshop constitute an unprofessional personal attack on Holland. You owe Holland and Webster a public apology

In my days as a government economist reviewing regulatory impact analyses, I frequently encountered statements supported by references that either did not provide support or supported a contrary statement. For the first few years of my employment, I was able to (i.e., I had political support) reject any analysis that contained such material errors — the essence, if you will of effective peer review. Once I sent the signal that error of this form would not be tolerated, the practice diminished. It resumed almost immediately after political support stopped.

OMB now has statutorily-based information quality guidelines that prohibit federal agencies from doing this. If you find this kind of error in a government document, it can be administratively challenged. Someday soon it will be challenged in court, and a judge will decide that material errors of this type are arbitrary and capricious.

I cover these issues on my blog.

Cheers!

RBB

I think you do a disservice to the community (that you are trying to protect) by shooting the messenger and not addressing the substantive points that Roger raised. If you’re worried about PR, then you should be aware that your response comes across as someone who has something to hide… Transparancy is critical to science, but you choose to ‘circle the wagons’ to fend off what appear to be reasonable critiques. Based on the evidence presented, and your avoidance, I can only concur that Roger is right. I’d be more interested, however, in seeing a more thoughtful and engaged dialogue about the facts rather than feelings, hence your disservice to the community.

cheers
e.

Well, let’s see.

Greg Holland wrote, “In contrast, Michaels, Knappenberger and Landsea (2005) argue the opposite, that the models must be wrong because they do not agree with the data.”

Roger Pielke Jr. responded, “Michaels et al. (2005) do not say that “the models must be wrong because they do not agree with the data.” They say that if you run the models with different inputs you get different results.”

Looking at the Michaels et al. paper, regarding the increases of CO2 used by Covey et al.: “However, a 1% per year increase in atmospheric carbon dioxide is not a realistic scenario for the future evolution of the atmospheric concentration of greenhouse gases—neither for carbon dioxide alone, nor in combination with other greenhouse gases. A warning to this effect is made by Covey et al. (2003) when summarizing CMIP results.”

Sooooo…Roger Pielke Jr. is absolutely correct. Greg Holland clearly misrepresents the Michaels et al. paper. The Michaels et al paper is objecting to the model input of a ridiculously high rate of increase in CO2.

So how is it that this point has “absolutely no merit”?

I agree with Judith from a PR perspective, the best thing to do is respond behind closed doors not just posting online, that is the collegial thing to do. However Judith I fail to see your point because it appears that Roger has done exactly what you asked, however the behind closed doors meeting didn’t solve his problem and was dismissed by Holland.

Being that Roger obviously wants things to be as accurate as possible he thought it best to get the conversation started (perhaps it won’t solve anything, however it will allow people to know that there is a dissent).

It appears Landsea hasn’t said anything on Holland, or Roger’s critique of it, therefore using him to bolster your case is a wash.

In the absence of a statement on this issue Landsea could side with either Roger or Holland; we don’t konw. Also any of the other co-authors could take a side, or they could sit on the fence and allow Roger to make their points for them, and see how things play out. (not everyone is the confrontational type, some are rather reluctant, Roger doesn’t appear to be one of them Being that Roger is the only author that has spoken up he is the only reference point till someone else decides to join the conversation.

From the excerpts that Roger has posted he appears to have a point, the statements are at odds with each other.

Greg’s response that I quoted was to pielke@colorado.edu (my personal email), and he chose to copy the list. He has not expressed any concerns to me about the quotation.

Again, do you really want this debate to be about who is allowed to speak on these issues?

In posting the paper on the listserv, Greg Holland was requesting comments and suggestions. He has received several. This is very different from what you call “complaints” and I call an attack. He is responding to and considering comments with merit. Apparently he has found one or two of these comments to be good suggestions. Holland has stated several times that he finds your “complaints” to be without merit.

Your quotation of Greg Holland’s email that was posted on the tropical listserv is apparently a violation of the listserv policies.