Interesting presentation by Bill Gray posted up here . I’ve just scanned through it but I’m sure that it will cause much controversy. Please also consider the critique of Bill Gray at realclimate here . The only areas that I have personally examined data and methods are for proxy data, where obviously I feel that the Hockey Team has used slipshod methods. I do not endorse anything in Gray’s paper and obviously many scientists feel that Gray’s methods are also slipshod. It is beyond my resources to examine everything in the world, though I’m looking a little bit at hurricane data. I’d be happy to post any critical analyses of Gray’s paper from anyone that’s looked at it in detail.
-
Tip Jar
(The Tip Jar is working again, via a temporary location) -
Pages
-
Categories
-
Articles
-
Blogroll
- Accuweather Blogs
- Andrew Revkin
- Anthony Watts
- Bishop Hill
- Bob Tisdale
- Dan Hughes
- David Stockwell
- Icecap
- Idsos
- James Annan
- Jeff Id
- Josh Halpern
- Judith Curry
- Keith Kloor
- Klimazweibel
- Lubos Motl
- Lucia's Blackboard
- Matt Briggs
- NASA GISS
- Nature Blogs
- RealClimate
- Roger Pielke Jr
- Roger Pielke Sr
- Roman M
- Science of Doom
- Tamino
- Warwick Hughes
- Watts Up With That
- William Connolley
- WordPress.com
- World Climate Report
-
Favorite posts
-
Links
-
Weblogs and resources
-
Archives
Climate Audit 
463 Comments
Had a problem opening it Steve – any chance of putting it in another format?
I didn’t have any problem accessing the document: it’s in MS Word format and I was able to read it with Open Office.
Opened fine for me using OpenOffice. If you don’t have it, it can be downloaded free from here. http://www.openoffice.org/
Large download though.
Thank you for the link and yes it also works here with OpenOffice 2.0. For those who believe in open source models and science, try open source tools!
Gray’s paper is a delight to read, lucid and uncompromising; to quote:
It is irresponsible to claim that the scientific debate on global warming is settled. A true scientific debate on this topic has not yet taken place. The debate that has occurred has been conducted largely by the media, the environmentalists, and the scientists receiving federal grant support to supply evidence of human involvement in global temperature rise. Most warming skeptics have been purposely ignored. Federal research funding for scientists skeptical of the human-induced global warming hypothesis has not been available.
There is little doubt that this presentation will get plenty of coverage. This is what I posted on my usual weather/climate forum, Netweather.tv:
This is a most peculiar piece by a man with a considerable reputation.
In it, as well as dealing with the issue of GW and hurricane prediction, Gray lays into the science community for jumping on the AGW bandwagon, and expresses the opinion that human induced GW will cause a further increase of ~0.3C, before the climate naturally enters another (slight) cooling phase, in the next ten years or so. There is no doubt that this part of his opinion will be heavily hyped by skeptics and heavily hammered by warmers.
The reason I say it is a peculiar piece is that, along with some sound material, especially about hurricanes, are a number of assumptions about GW which, to be frank, are simply repetitions of the well-known, existing skeptic criticisms of the theory. Most prominent among these are; the GCMs have no skill (his reasoning here is suspect); the GCMs overestimate GW by a factor of 5-10 times because they exaggerate the effects of clouds; the current warming phase is a function of the THC (this one really got me, personally; even from my own, non-scientist pov, what he says about the THC is in contradiction to the findings of several pieces of research).
But most of the presentation is, effectively, an op-ed style piece about how the governments of the world have conspired to produce a common enemy with which they can cow the populace into mindless obedience, how the science community and funders have ignored skepticism because it isn’t fashionable, how the GCM teams have suppressed uncertainty because they are afraid of losing credibility; basically, how AGW is all some kind of huge conspiracy. (my apologies for all the paraphrasing, but it is a 42 page document).
What is irritating is that Bill Gray’s piece is likely to provoke more reaction than it probably deserves, simply because of who he is. I am not pretending that I know more about weather forecasting or climate science than he does, but if even I can pick holes in his material, then when the ‘heavies’ get their teeth into it, there will be all kinds of a furore, stimulated, I am sure, by American media response and claims from the skeptics that this justifies their position, because it omes from such an eminent source.
To add to this: The discussion of climate modelling versus weather modelling has been discussed in detail on both Climate Science and RealClimate: RP Sr. & Gavin Schmidt (with assistance from commentators) have different views on whether climate models can be compared to weather models, and why the appication of chaos theory to weather forecasting is relevant to climate modelling. Gray assumes that the two models are the same, and says that, as the weather cannot be forecast more than 5-10 days out, we can have no confidence in climate models. This is, to me, a suspect argument from the outset.
Gray also claims that the overestimation of warming in the GCMs is largely a function of errors in cloud parameterisation which are multiplied many times over. The uncertainties about cloud function are much discussed and I do not know of any fixed conclusions about them either way; for Gray to say that this is the ‘achilles heel’ of GCMs is simply a repetition of existing skeptic claims which are, as yet, unproven.
On the THC, Gray claims that much of the observed warming of the past thiry years can be attributed to a strong THC and natural variations in upwelling, and that this multidecadal cycle will reassert itself in a few years, after which we will see more (slight) cooling. In justification of this, he compares the current situation to the previous changes of the twentieth century, which saw a warming up to the 1940s, followed by a cooling to the 1970s.
There should be enough in here, from a non-scientist, to draw your attention to the fact that this presentation has several problematic elements, none of which, of course, is going to change the fact that it will be discussed ad nauseam for the next couple of weeks.
Regards,
Thanks Bob K for that, not quite sure why my system threw a hissy fit but the openoffice works fine.
It does read a bit like a rant and he does contradict himself on more than one occasion, which rather detracts from the whole document.
Im becoming increasingly sick of being bashed over the head about global warming conscensus. Every night on the media, im served up with a fait acompli science saying global warming has caused this and that.
Climate Audit is the closest thing that i can find that at least continues to discuss the science, and its current limitations. It is only the science that matters and nothing else.
My qualifications are not in the sciences, but many of the subjects studied gave me some basic grounding in the topics discussed. Consequently i grasp many of the issues discussed.
To me the numbers and data dont add up and certainly dont show irrefutable proof of human induced global warming.
Crichton is spot on with this comment, but take a few minutes to look at the general media, forums or the net and you will be told that the science is settled. The conscensus is that global warming is caused by humans. Again and again and again.
These days you have to be either one of the small minority that understand climate science, very brave or very stupid not to accept the global brainwashing that is going on.
Ill end this post with Bill Grays final comment.
Please keep yelling gents. Many of us depend on the sceptics, because i have no doubt that sooner or later the skeptics will be proven to be correct.
Fergus, I’ve never seen any explanation of how, as you claim, weather models are different from climate models. I do not see any fundamental difference between the two.
Each takes an opening set of variables, representing the current state of affairs. Each then steps one increment of time, and uses various physical laws and parameterizations to predict the next state of the climate. Repeat ad lib.
What is the difference?
The climate modelers say that they’re not trying to forecast the weather, they’re forecasting the climate. But since the climate is the average of the weather, how can you forecast the climate without forecasting the weather?
w.
The most telling point is that the modelers are willing to make projections 50 to 100 years in advance but refuse to produce predictions for next year or the year after.
Re. 11: Willis, the discussion of this on http://www.realclimate.org/index.php/archives/2006/08/short-and-simple-arguments-for-why-climate-can-be-predicted/#more-318
is far more substantial than I could manage. Alternatively, there’s a response on Climate Science (page 4) which you may prefer to read, in which RP Sr. takes Schmidt to task. As far as I get it, the argument is whether, like the weather (excuse the homonym), the climate is chaotic in nature; it may boil down to an issue of definition rather than of mathematics, though.
Hope this helps, as any effort I make to explain would be strewn with error and probably not much help.
Regards,
Wow, it is easy to see why the global warmers hate Bill Gray so much. Great presentation.
I think it is clear the debate whether global warming has increased hurricanes is over. Hurricane and tropical storm numbers are cyclical and are affected by a number of meteorological conditions unrelated to global warming. It is over.
It is now time to take on the hydrological link to increased CO2 which forms the real basis of the global warming theory. Increased CO2 leads to increased water vapour which provides the positive feedbacks (5 to 10 times) that global warmers need in order to make their forecasts of 1.5C to 4.0C temperature increase work.
Bill Gray points out that water vapour actually adds a neutral to dampening effect on temperature changes due to CO2. I think Willis has been making this point consistently on this board for awhile.
If people really understood this is what the global warmers are assuming (without having the science to back it up) it might make a big difference in the debate.
And did he tell the guy about the positive feedbacks in the GCMs? Did he tell him about the ‘flux adjustments’ in the GCMs?
With a few exceptions (like fFreddy) I think it is futile to have debates with under-grads on global warming. They have already made up their mind that global warming is entirely down to evil fossil fuels. I blame this (in the UK at least) on our education system which (with some exceptions) is populated by eco-theological teachers who push their own mindset onto our impressionable kids. By the time they leave senior school (high school) they have already been so indoctrinated into the faith that it will sadly in all too many cases take the rest of their lifetimes to un-indoctrinate them.
KevinUK
Re #13, KevinUK
Umm, sorry, are you saying that I am an under-grad ?
The difference between weather forecasts vs long-term climate forecasts is a valid one.
How about this analogy? Many of us invest heavily in mutual funds in retirement plans assuming they will make us a lot of money over the next 30+ yrs at 10% avg annual rate of return. We accept that there will be upswings and downswings but that they will generally trend upward. But as we approach retirement, we shift our retirement money into more stable bond and money market funds. We can no longer accept the risk of a stock market downswing, and we don’t know if one is going to happen.
(warning: long comment including a boring personal anecdote!)
I personally liked the presentation. I don’t know if his arguments are valid or not, but I like that he tries to give a simple physical explanation of what is going on. It may be the old way of doing science, what with hyper-complex GCM’s and all, but if you can’t simplify things it’s often because you don’t fully understand them. My own fear with regards to GCM’s is that we put too much faith in their results, without understanding how they come about.
A complex model, in my opinion, should be a way to help understand the physics of a complex system. It cannot just be a number crunching tool blindly spitting out results that you take as an article of faith. I get the feeling (maybe I’m wrong) that that’s how GCM’s are used. When you see the predictions going from 1.5 to 4.5 C or even more (climateprediction.net goes as high as 11 degrees!), and you’re being told that it’s all because of “positive water vapor feedback” effects, it seems to me that if we really understood these effects, we should be able to narrow down the temperature range substantially. That we still can’t do it after almost 20 years of intense model development tells me that we should maybe change our approach, and put a little more emphasis on a good physical understanding instead of just improving the resolution or getting rid of “flux adjustments”.
Let me just give an example taken from my own scientific experience that will illuminate my feeling about numerical models. Not many physical systems studied nowadays have what you call “analytical solutions”, where the system is described by a set of equations that can be solved with a pencil and paper approach. Thank goodness we have computers that can help us. The climate system is but an extreme example. My first scientific piece of work as a young graduate student was to study some physical effect that happens when you use lasers to do spectroscopy. Lasers can be very powerful so when the light intensity is high enough, you depart from a “linear” regime. The spectroscopic technique I was investigating was already what is called a “nonlinear” technique, but was usually described by taking the “exact” equations describing the interaction of light with matter, to a higher degree of approximation (in this case the third order; the first order being the “linear” approximation, and the second order being zero from a symmetry argument). So we wanted to know what happened if the laser intensity was even higher, and since the equations could not be resolved analytically, I decided to solve them numerically on a computer.
The numerical integration proved tricky, and I had to learn a lot about numerical techniques. You want an algorithm that converges easily and is not prone to instability. If I wasn’t careful, there were some features in the simulated spectrum that were just artefacts. In the end, I was pretty much convinced that my program was good enough. Still, one of those artefacts refused to go away! Eventually, I had to consider the option that maybe this could be a real physical effect, but how was I to know? The only solution was to go back to the equations, and try to solve them analytically up to the fifth order, by which time you ended up with many, many terms in your solution. Then I realized that with a simple argument, you could show that most of these terms gave negligible contributions. The handful of terms that remain could be shown to generate the feature that I got from my numerical program! From there I could develop a “diagrammatic” illustration of those higher order processes, where the physical explanation became evident.
Now this was all theoretical! The effect that I had found had never been observed. I got two papers out of that work, but then moved on to other things for my Ph.D. It’s only about 5 years later that I saw a poster at a conference where they were showing results that proved that my predictions were right!
What is the lesson to be learned here? It is that there is not much you can do with model results if you can’t extract a better physical understanding, and then simplify the problem. You should be able to devise a simple physical explanation of what the model tells you.
I’ve read many paper where the GCM results don’t even agree with the observations, and the authors don’t know why! How can you trust such a model? GCM’s can be great tools to help us understand the physics of the Earth’s climate. But trust them to make forecasts? I’m not sure we’re there yet. In this sense, I agree with Dr. Gray.
RE: #5 – There is a continuum between weather models and GCMs. The most ambiguous area is forecasts in the range between 72 hours and 90 days. At one end of that range, 72 hours is clearly close enough to t=0 to be more of a small perturbation analysis versus current conditions. however, 90 days is 2.5 percent of decade scale and is aguably not a weather forecast by any stretch of the imagination. But where do I find the 90 day forecasts here in the US? On weather server web sites. Point being, if we are honest, the same model should be used weather t=1hr, 1 day, 1 year, 1 decade or 1 millenneum. If a model is correct enough to be of true and lasting value, it needs to predict accurately at all time scales. If it diverges more than the acceptable accuracy requirement as t increases, then the model has innate flaws that are unacceptable in my book.
There’s an Einstein parable in there somewhere. IIRC, along the lines of “100 scientists think you’re wrong” and his reply “why, it only takes ONE to prove me wrong.”
Mark
RE: #12 – one of the articles of faith of the climate science orthodoxy (aka “warmers”) is that any net accumulation of energy in the tropics or even mid latitudes absolutely must be transported poleward at some point, minus some supposedly inconsequential losses. Yes indeed, the Hadley Cell model that is the basis for this was and continues to be a nice first order approximation of the grand circulation patterns of the atmosphere. However, let us contemplate the real world. When a major convection event occurs lower than some given latitude under some given synoptic pattern, the “excess” engergy that presumably led to the event, may well never reach above a certain given latitude and might, instead, be completely consumed by a combination of in situ kinetic and electromagnetic energy as well as transported as vertical flux into the Stratosphere and eventually thereafter, outer space. The GCMs do not explicitly treat these mechanisms and instead attempt to account for them via parameterizations, tweaking of flow tensors and other cheats. Does any one wonder why any attempt to discuss what I have laid out here so enrages the climate science orthodoxy?
RE: #13 – identical situation here in the US.
#18 Mark perhaps you were thinking about this quote from Einstein:
“No amount of experimentation can ever prove me right; a single experiment can prove me wrong.”
Allow me to offer another quote from Einstein for those of us who are statistically challenged:
“Do not worry about your difficulties in Mathematics. I can assure you mine are still greater.”
And finally (pun intended):
“Any man whose errors take ten years to correct is quite a man. J. Robert Oppenheimer (1904 – 1967), speaking of Albert Einstein
For the view on Gray’s paper from climate scientists that actually conduct research on the subject, i refer you to the RC post “Grays’ Muddy Thinking”.
http://www.realclimate.org/index.php/archives/2006/04/gray-on-agw/
RC only scratches the surface in identifying the flaws in this paper, IMO
I will pick up this issue on the GT report card thread (later tonite or tomorrow, v. busy today), but one of their comments was that climateaudit never says anything nice about a paper, only criticizes papers. I was just accosted in the hallway by a totally astonished student from the hurricane class (one of 3 still monitoring the blog), who told me i must urgently go to climateaudit, could not believe that climateaudit liked bill gray’s paper
re: #5 Fergus,
I very much dislike your supposed “paraphrase” of Dr. Gray’s presentation. But rather than pour out invective and generalities I’ll bring out one specific error in your message. You say that Dr Gray
This is not at all what he says. He says that a doubling of CO2 by the end of the century would result in warming of less than about 0.3C. He did not say this would happen before his projected slight cooling phase. If you’ll look at his Figure 4 this should be obvious. What you’ve really done here is concatinate two separate predictions and created a total non-sequitor. It’s similar to a “syllogism” by Douglas R Hofstadter in “Godel, Escher, Bach.”
So, to repeat what the Tortoise replied, “Now wait a minute…”
#19 Steve, while you may be right, I think that what “enrages” the “orthodox” scientists is that a lot of the sceptic’s critique is negative. It’s all very well to point out to this and this deficiency in the models or the observations, but what is the alternative proposed by the sceptics? For a scientist who is working full time on trying to improve GCM’s, to simply be told by some amateur commentator that his model is crap isn’t quite good enough, and can be infuriating. The criticism may well be valid, but it’s not constructive. At some point, for the scientists, it’s just an annoying background noise. Sorry to say so, but a lot of what we read here on this blog (the comments, not the posts) is of that nature, and really is of no value to the scientific debate. But then, can you expect more from a blog?
Bill Gray’s proposal to have a fund dedicated at researching “alternatives” has some merits, however. Allan Mazur and others have, in the past, proposed the idea of a “science court”, where, instead of trying to extract a consensus from the scientific community, you would have scientists arguing both sides of an issue, with an independent judge to draw the appropriate conclusions. The existence of the IPCC, and the emphasis on the consensus has, in my opinion, distorted the natural process of the scientific enquiry. One side (the majority side, I concede), has come to dominate not only the scientific debate, but the scientific “establishment”, i.e. grant agencies and journal editors. I agree with Gray that it may have become difficult for a researcher to make research proposals that explicitly go against the current orthodoxy. It doesn’t have to be true, only that the scientists themselves “feel” that it might be true, and that would be enough to discourage them from making such proposals (why spend the time and effort making a proposal that you feel might be rejected, better go with the flow…). There have been a number of instances in recent years where papers proposing alternative views (e.g. the solar papers, or, say, Lindzen’s “iris” effect) have been unjustly attacked in the litterature. The field of climate science has to find a way out of this, if it is to retain its credibility.
You all had better watch your backs. The Consensus Police may just swoop down oppon this blog and arrest you all. And Bush is the one who is silencing dissenters and destroying civil liberties??? Jesus Friging Christ!! (is it safe to type that?)
When the leftist brownshirts came for the CEO’s,
I remained silent;
I was not a CEO.
When they locked up the social conservatives,
I remained silent;
I was not a social conservatives.
When they came for the free trade advocates,
I did not speak out;
I was not a the free trade advocates.
When they came for me,
there was no one left to speak out.
Note to me – proof read and spell check next time.
#21
“No amount of experimentation can ever prove me right; a single experiment can prove me wrong.”
That is the “Popperian” view of science. While philosophically, it may be founded, in practice it’s not necessarily the way it happens.
Since you take Einstein as a witness, take the example of general relativity. Einstein published the more or less final version in 1915. But since it deals with tremendously small effects, one had to wait until 1919 for Einstein to be “proven right”, when Eddington went to South America to observe the solar eclipse, and the predicted deflection of light by the Sun. The idea is that the light from a star located behind the Sun would have to pass through the Sun’s large gravitational field, and would be deflected. Only during an eclipse can you observe that effect, since otherwise the light from the Sun drowns the tiny signal from the star.
Eddington did claim to have observed the predicted deflection. However, the true story is that his measurement error was large enough to overlap the alternative explanation. Most people don’t know about it, but the Newtonian theory also predicted a deflection, only half as large as Einstein’s. Some have even said that Eddington, intentionnally or not, narrowed his error bars to make his results look more convincingly in Einstein’s favor. In any case, for most people, this was considered as a “proof” that Einstein was right, and it was enough for most physicists to adhere to the new paradigm, even though the evidence wasn’t quite so convincing.
Re #22, Judith Curry
Gray’s paper is dated October 11th, 2006.
You are linking to a RealClimate post that is dated in April 2006, and whose last comment is dated in May 2006.
Re: #15, #17,#23. Thank you for your responses. Taking Michael & Steve’s separate posts together, it’s easy to see why I don’t want to go either way on this discussion; one agrees, the other disagrees. As I said before, this smacks more of a problem of definition than of strict comparison; the point Steve makes later about the 90 day forecasts is, however, perfectly valid. I believe, though, that these are products of weather models, not climate ones, so there are considerable differences in both input and output which would need to be considered. Whether the climate models should be able to offer predictive output at the 90 day scale is open to discussion, and I follow RP Sr.’s discussions on model output with great interest.
On Dave’s comment; you got me! quite right, I messed up on that part of the paraphrase and concatenated two separate statements; my apologies; an error of haste rather than intent. Why you should take such exception to my post is, otherwise, beyond me. What do you take exception to? I did my best to cut down a long presentation to a ‘bite size’ chunk (but where I did it on my regular forum, I also offered a link to the article). I got one bit wrong. Wherefore the invective, Dave? I didn’t think what I said was especially controversial, was it?
Regards,
#27 Francois Ouellette
At the risk of straying further OT let me quickly say that Popper’s concise proposition that a falsified prediction implies a false theory is at the core of the scientific method.
A central problem of AGW is that its predictions are not immediately falsifiable and therefore one could argue unscientific. The intriguing case of Eddington’s accidental or otherwise statistical mistakes are irrelevant as they were not falsifying GR predictions merely confirming them..
RE: #24 – Francois, here is my proposal to improve the models:
– Conduct an extensive characterization of model behavior – including a full blown error factor analysis. For each error, extensive root cause analysis along the lines described by you earlier in this thread.
– Attempt to rethink the very essence of the models from first principles, using an approach not unlike object oriented, structured software development strategies. For example, model the extrapolated effects of each first principle element, etc
– Somewhat related to the foregoing item, investigate with diligence a finite element approach, similar to what is used for commerical modelling software used in the electronic equipment industry to design thermal management strategies, such as Flowtherm. This is not to suggest that something like Flowtherm could be scaled but is to suggest that certain aspects of finite element approaches might lend themselves to the achievment of a reliable GCM, which is capable of dealing with losses, non linear behavior and turbulance more accurately.
This first step to corrective action is recognition of failure modes.
I am not going to critique Gray’s paper, it is beyond rational critcism, i will save technical comments for such an unlikely event as any of this actually ever gets published. Bill Gray is not a player in the scientific debate, his ideas reflected in the paper referred to at RC are so flawed that they are unpublishable. Bill Gray does not enrage the scientists, he simply isn’t a player in the scientific debate on global warming. However, he is a HUGE figure in the public debate on global warming. I am not alone in judging Gray to be “off the spectrum”, Richard Lindzen said something similar in an interview with Joel Achenbach (most climateauditors wont’ like the achenbach piece, but i believe the quote from Lindzen to be accurate:
http://www.washingtonpost.com/wp-dyn/content/article/2006/05/23/AR2006052301305.html
“Of all the skeptics, MIT’s Richard Lindzen probably has the most credibility among mainstream scientists, who acknowledge that he’s doing serious research on the subject. ”
“When I ask Gray who his intellectual soul mates are regarding global warming, he responds, “I have nobody really to talk to about this stuff.”
That’s not entirely true. He has many friends and colleagues, and the meteorologists tend to share his skeptical streak.
I ask if he has ever collaborated on a paper with Richard Lindzen. Gray says he hasn’t. He looks a little pained.
“Lindzen, he’s a hard guy to deal with,” Gray says. “He doesn’t think he can learn anything from me.” Which is correct. Lindzen says of Gray: “His knowledge of theory is frustratingly poor, but he knows more about hurricanes than anyone in the world. I regard him in his own peculiar way as a national resource.”
In the Klotzbach thread, I characterized Bill Gray as “off the spectrum”. He is all alone out there. This isn’t to say that his views aren’t widely “popular” in certain nonscientific circles.
p.s. In the hurricane class on Tues, i asked the question “how many students think i am wasting my time blogging at climateaudit”. 8/14 raised their hands. They are now telling me they are changing their vote, and that i need to straighten them out on Gray’s paper. I am not going to bother, but they are working on Peter Webster, maybe he will post his critique.
Re: #28
Dr. Curry what/who/where else would you reccomend a non-scientist person such as me and even for a scientist such as my husband to read online for an in depth honest analysis of the FLAWS in any of the the climate models/statistics/data written by OTHER scientists not the authors (like at RC) you approve of? Also husband would like to know if you’ve read any of Dr. Mann’s papers?
#14 fFreddy,
Humblest apologies. I must have mis-read one of your previous posts. Are you a student? If so, at what level? High school, mature student?
KevinUK
RE: #33 – it might be productive to search literature in Computer Science. The best critiques of models and algorithms may come from outside of so called “climate science.”
Uh, criticism is what science is all about, Judith. Without criticism, hypothesis remains that, and never becomes theory. Should a paper survive criticism (relatively intact), then it rightly becomes theory.
The history of this site is based on criticism, for sure, but many of the supposedly “pre-conceived notions” that some of us have here are based on a history of fundamental errors made by those that constitute the “consensus.” It is in this light that the views of many of your students that we “automatically assume any GW theory is flawed” are hypocritical. I.e., they have only looked at us for a very short time, and are drawing conclusions based on a limited sampling of what has been discussed, while comparing that to what is currently discussed at RC. While certainly a monumental task awaits anyone attempting to understand where everyone in here bases their opinions, it is required if the goal is to put the last couple months into perspective.
Mark
#30 I think a better representation of Popper’s view is that a theory is “scientific” if it makes falsifiable predictions (“demarcation” criterion). In practice, and that is more Kuhn’s argument, the fact that one prediction may be falsified never falsifies an entire theory. Paradigms can live for a long time, even in the face of falsifying evidence (e.g. the Mercury perihelion etc.)
#31 Steve, I thought GCM’s WERE based on some finite difference scheme ?! Isn’t their biggest drawback the difficulty of adequately representing processes that take place on a sub-grid scale, like clould dynamics, as explained by Gray? On the other hand, I agree that there has to be a thorough investigation of the model’s behavior as compared with observations. I think that a model that gives a correct “global” picture but an incorrect local one has got some fundamental flaw. The inverse is also true! I don’t know if building a new kind of model from scratch is realistic!
I am eagerly awaiting Dr. Curry’s comments (hoping she avoids the f** word). As I said, I don’t know if his arguments are scientifically solid (more references to the litterature might have helped), but I like his “philosophy”. Unfortunately, I can’t read RC. Makes me want to throw up every time.
Same here. The mere fact that they suppress dissent is reason enough to avoid them. We may be uber critical here, but RC is nothing more than a rah rah site for the fanbois, i.e. they are the antithesis of criticism (and falsification). At least here you get both sides.
Mark
RE: #37 – ” I think that a model that gives a correct “global” picture but an incorrect local one has got some fundamental flaw. ”
To wit:
http://climatesci.atmos.colostate.edu/2006/10/12/a-new-paper-that-evaluates-the-accuracy-of-the-ipcc-models-to-assess-regional-weather-patterns/
I think we actually agree regarding finite element approaches. The issue I was getting at was one of scale. Current grid scale may not comprise of sufficiently granular elements to capture the effects that are at least some of the set of root causes of errors. Combine that with revisitation of first principles, for example, the overuse of the Hadley Cell paradigm, cloud physics, E&M mechanisms, etc, etc, and I think performance levels might be vastly improved. Put another way, a Six Sigma / TQM type of approach, using either a DMAIC or even perhaps, if time and money would allow, DMADV, juxtaposed on a very rigorous and conservative hard scientific approach.
#35 Yep you are probably right.
But if you are going to bring up flaws on CA to prove your point, you are going have to answer to all of them first to prove your point IMHO. My husband had to stand in front of a panel of his peers and argue his thesis before it was accepted!
re 22 RC on Gray
I went to the RC article. Applying the word “juvenile” to tit would be an insult to juveniles. An example:
From the article:
“He claims flatly and without supporting evidence that models cannot simulate the THC properly, neglecting the fact that the models employed in the IPCC reports yield a rather wide variety of different possible THC behaviors, ”
Or in other words, our models are so good that they can model any sort of THC behavior including behaviors that are not done. How is this different from what they purport that Gray is stating.
Juvenile ad hominem statements are not convincing
Judith, I wish you could find some time to comment Gray’s section 7 (FLAWS IN THE GCM SIMULATIONS OF GLOBAL WARMING) and also the interesting paper linked in #39 (Ruti et al., 2006).
I also just finished reading your paper:
Rinke, A., K. Dethloff, J.J. Cassano, J.H. Christensen, J.A. Curry, P. Du, E. Girard, J.E. Haugen, D. Jacob, C.G. Jones, M. Koltzow, R. Laprise, A.H. Lynch, S. Pfeifer, M.C. Serreze, M.J. Shaw, M. Tjernstrom, K. Wyser, M. Zagar, 2006: Evaluation of an ensemble of Arctic regional climate models: spatiotemporal fields during the SHEBA year. Climate Dyn., 26 (5): 459-472.
which I also recommend for all people here interested in reliability of the (regional) climate models. I have a question: based on your results, what makes you believe (I suppose so) in the predictive value of the current climate models?
Re #34, KevinUK
Oh, no offence taken, just curious as to why you thought that. To answer your question, I graduated a couple of decades ago and haven’t been near academia since.
Re#22
What was his/her evidence that “climateaudit liked bill gray’s paper?” Do you actually agree with this student’s assessment?
I see a handful of posts prior to yours which agree with one or some of Gray’s points. Just as many posts are critical.
#32, Judith
Thank you for the link.
“I am not alone in judging Gray to be “off the spectrum”, Richard Lindzen said something similar in an interview with Joel Achenbach (most climateauditors wont’ like the Achenbach piece, but I believe the quote from Lindzen to be accurate”
I’ve just read it all and contrary to your expectation, I liked it very much and couldn’t see anywhere within it where you have got your “off the spectrum” quote from.
What I did find however were these quotes from non other than your very self.
“I thought he was playing politics. But, damn it, he was right.” Yourself in talking of Hansen 1988 congressioanl testimony. And also
“Curry, who believes the skeptics have mounted a “brilliant disinformation campaign,”. Not sure if you meant that as a compliment or not but it perhaps reveals your motives in posting on this web site and perhaps explains the comments of your students on this blog given your position of influence on them.
Would you care to comment on either of these two quotes particularly the last one?
I also found the following quotes interesting
“The modelers are equation pushers.” by Bill Gray on climate modelers.
“There’re people like [Lindzen] in every field of science. There are always people in the fringes. They’re attracted to the fringe . . . It may be as simple as, how do you prove you’re smarter than everyone else? You don’t do that by being part of the consensus,” Held says.”. Issac does that mean since I’m said to be on the fringe by the ‘warmers’ that I’m smarter than you? And would you also like to comment on the following quote
“But Held argues that the models are conservative. For global warming to be less of a problem than is currently anticipated, all the uncertainties would have to break, preferentially, toward the benign side of things. Held”. Issac could you explain to me how uncertainties can break preferrentially to either the benign or harmful side of things?
“He’s simply sick and tired of squishy-minded hand-wringing equation-pushing computer jocks who’ve never flown into a hurricane!” by the author when quoting Bill Gray. Judith have you even flown into a hurricane?
“When you’re looking for a signal in a very noisy record you do as much averaging as possible.” by Kerry Emanuel. Does this quote say something about the depth of KE’s statistical knowledge?
“The Web site criticized Gray for not adapting to the modern era of meteorology, “which demands hypotheses soundly grounded in quantitative and consistent physical formulations, not seat-of-the-pants flying.” quote from RC. They must be referring to those highly skilled (from a predictive capability perspective) climate models then?
And last but certainly not least the following quote
“The executives don’t understand “resource economics.” They lack faith in the free market to solve these issues. And they go to cocktail parties and find out that everyone thinks they’re criminals. Or their kids come home from school and say, ‘Daddy, why are you killing the planet?'”
Now its confession time. I now confess that amongst other reasons one the things that caused me to leave the nuclear industry for the rail industry was peer pressure. I got fed up being a social outcast. Following an exemplary safety record in the nuclear power generation industry, I joined the rail industry as a safety and risk management to subsequent live through the Southall, Paddington, Hatfield and Potters Bar rail disasters. I gave up S&RM after being expected to produce reports that justfied not fitting TPWS on the UK rail network and took up software development instead full-time as a career.
KevinUK
From section 4 on Global warming and hurricanes
“Atlantic hurricanes go through multi-decadal cycles. These cycles have been observationally traced back to the mid-19th century and inferred from Greenland paleo ice-core temperature measurements that go back thousand of years.”
Does anybody know whoes work this refers to? I do not know of any work that uses the Greenland ice-core parameters as a proxy for hurricanes.
I’ve added some comments to the thread header to indicate that I am not endorsing Gray’s paper, as I haven’t studied the data matters involved. Speaking from personal experience, realclimate comments on proxy topics are not a reliable source of information. Indeed, quite the opposite – they are biased and make many misrepresentions and mischaracterizations. Perhaps they are more reliable in other climate areas.
Gray’s paper contains many tendentious statements about politics. I called Hansen’s PNAS paper a hodge-podge considering that discussions of the Framework Convention fitted uncomfortably in an academic publication seeking to show how modern instrumental SSTs could be spliced to Mg/Ca G ruber proxy information. Should I make a similar comment about Gray? In one way, what’s sauce for the goose is sauce for the gander; but Gray’s presentation here is to a Washington public meeting, and is not in PNAS. I’m sure that editors would take out his political ruminations; I’d agree with that. I wonder why Hansen’s ruminations were allowed to remain in his PNAS article.
Re #28: Gray has said all of this before, going back at least a year. For that reason, I expect that it won’t get much attention. I don’t think there has been a meaningful change in the material the RC post was responding to.
Re #46: The language is a little unclear. The inference is that Atlantic climate cycles (which can be traced in the cores) in turn drive hurricane cycles. The connection between the two is in some dispute, however.
Re #32, Judith Curry
I really think you should. Given your confusion about dates, as mentioned in #28, anyone might think that you hadn’t actually read Gray’s paper, and that you were just having a pavlovian reaction to the guy’s name. I’m sure that isn’t the case, of course, but you wouldn’t want to leave that impression, would you ?
Dr. Curry, I think your contribution to the discussion on Gray so far has been what you criticize in some skeptics’ posts here and that is its lack of relevant content. I personally do not see where any significant numbers of posters here have taken a stand on Gray’s paper. They have rather reiterated their own personal, and in most cases previously stated, views on the arguments for AGW.
I think that the general subjects that many of the skeptical and not so skeptical posters here want to discuss with the experts in the climate field are their views on the issues of measuring uncertainty in published results, data snooping data and curve fitting, their opinions/judgments of the current state of climate science involving these issues and what processes they would suggest for containing/eliminating them.
You seem to deliver messages here through your students’ comments and they seem to have reached the level of aggravation that those undergraduate students of Kenneth Blumenfeld’s reached in their admonishment of RC to get it on with CA.
re: #29
Ok, since you were willing to admit a mistake, that goes a long way in itself in making you worth debating with. Here’s a second problem:
First of all, being repetitions of well-known, existing skeptic criticisms doesn’t make them wrong unless you’ve already dismissed them. In that case you’ve moved beyond attempting just to paraphrase him to what amounts to poisoning the well, i.e. setting up the reader to dismiss his points when you finally do address them. Your “his reasoning here is suspect” is of the same sort.
Finally you say he’s claiming that Gray says that GCMs exaggerate the effects of clouds, but that’s a distortion too. It isn’t that GCMs exaggerate the effects of clouds, it’s that they don’t even necssarily have the sign right. Now, it’s true that low clouds and high clouds have different effects on warming, high ones tending to enhance it and low ones tending to be a negative feedback. Dr. Gray claims that a proper analysis of the sub-grid effect of higher CO2 would be to reduce the amount of high clouds because of increased precipitation efficiency.
So your lumping a fairly detailed discussion under the rubric “clouds” and in partularly claiming that Gray claims an exaggeration, implies that clouds in general are also part of the positive feedback systems when this is clearly not the case.
#47 Please, Steve, this is an opinion piece, not a peer-reviewed paper, and it should be judged as such. Whatever Gray’s scientific views would be, they are not stated the way one would do in a journal article. Not enough detail, not enough references, etc. This was meant for the specific audience of the Marshall Institute, whoever they are.
Hansen’s paper is a bona-fide peer-reviewed paper. As such, it meddling with political issues should be a no-no. But hey, this is Jim Hansen… (I think he’s expecting a Nobel or something)
Re: #52, #53: Dave, as you wish to take me up on it: the ‘to be frank, are simply..’ comment is meant to show my disappointment at the familiar nature of the material, coming as it does from a respected meteorologist, rather than an implication that the arguments themselves are to be rejected. I haven’t ‘already dismissed’ all skeptic claims about GW, but the ones which are covered in the presentation are all well-covered in the literature, on both sides. My observation that ‘his reasoning is suspect’ on the subject of the GCMs and skill is that he appears to conflate weather forecasting and climate modelling; this is suspect for the reasons mentioned in above posts. It is not a claim that he is wrong, just that there is cause for closer scrutiny.
With respect to the comment on clouds; you will have to accept that this was simply an attempt to reduce down the argument, not to misrepresent the basic statement, which I do not believe I have done. Whether his comments on the rubric proves to be correct or not, I point out, a little later in my comment, that ‘I do not know of any fixed conclusions about them either way’; the implication being that his case is not (yet) corroborated by decisive scientific conclusions.
As far as I recall, Gray does claim that the GCMs’ treatment of clouds produces an exaggeration of the effect of CO2, (or perhaps it was an exaggeration of the future temperature trend).
At the start of my post, I pointed out that it was a cut & paste job from a web forum I contribute to; its readers are, in general, intelligent and interested in climate change, but unlikely to want to pursue lines of enquiry as far as some of us, so I made it as simple as possible and, insofar as I was able, as unbiased as possible; I always try to encourage readers to make their own minds up. I am sorry that I have had to defend what was never intended as a critique of the paper, and I accept that my use of language may sometimes imply judgements which have, otherwise, been avoided carefully.
In summary, then, it isn’t that I dislike or disapprove of the paper because it is skeptical about GW (otherwise, why would I be reading CA and Climate Science?), but I find it peculiar because it doesn’t really add anything to the debate; as it was a presentation to the Marshall Institute, perhaps I was expecting too much of it. I’ll let others, with a better science pedigree, debate the validity of the contents, apart from my comment about the THC, about which I have several papers recently, none of which mentions the phenomenon that Gray refers to, and a couple of which, I believe, are in more or less direct contradiction to his comment here.
Francois: It is probably useful to know more about the George C Marshall Institute: wikipedia should do the trick. Let’s just say that Gray knows his audience.
#45,
How about these comments from Judith as well…
Gray has “brain fossilization,” Curry told a Wall Street Journal reporter a few weeks ago, and “nobody except a few groupies wants to hear what he has to say.”
Whether Bill Gray is right or not, given the history between the two, I doubt I would use Judith Curry’s comments as a valid critique of Gray’s speech.
re: #54
Well you’ve backed away from you statements about as far as it’s safe to do without looking behind you to make sure you aren’t near a cliff. Let me just suggest that in the future if you’re going to post here you actually quote from the person you’re discussing and skip the paraphrase stage. It doesn’t seem to be your forte.
And I also think that the people here are not going to be impressed by a post written to, shall we say, less than totally informed induhviduals.
Re #54; I didn’t know this was an exclusive club. I won’t waste my time, thank you.
#54 Don’t worry, I know what the Marshall institute is. That’s what I said: don’t read this as a scientific paper, read this as a presentation to the Marshall institute.
If anyone has a good scientific critique of his THC explanations, I’d like to hear it. I mean, a polite and reasoned explanation for the layman that I am of why and how he would be wrong, with no ad-hom or insinuation about his motives or integrity or the state of his brain. References and links to key papers supoprting the argument would also be appreciated.
BTW I just finished reading the Lucarini paper that Pielke Sr. pointed to, which compares various GCM’s, and it’s highly interesting.
Judith, I was sad to see your comment:
I was sad because your posts are not a waste of time for me. I read them, and think about them, and value them, as avidly perhaps as any of the students in your class.
Which is why I was so disappointed in your response to Gray’s paper. So far, in stark contrast to your postings on other subjects, you have not said one substatantial thing about his paper. You merely declare that it is wrong, and refer us to a RC post, which discusses three of Gray’s “claims”. So, against my better judgement, I went there … here’s the first claim:
They ridicule this “claim”, saying:
Umm … say what? The Hadley cell circulation definitely cools the planet. It moves energy from the tropics to the pole, where it is removed by radiation. Water circulation in your car engine definitely cools the engine, because it moves energy to the radiator where it is removed by the air. Thunderstorms definitely cool the planet, because they move heat to the upper troposphere, where it is above most of the GHGs and is free to radiate into space.
Why is it “a scientific absurdity” that THC circulation could have the same effect? I haven’t a clue whether it is doing so or not … but simply declaring it to be a “scientific absurdity” doesn’t change the fact that circulation of a hot fluid definitely can cool the fluid. Perhaps one of your students could explain to us why that is “scientifically absurd”.
Again from RC:
About which they say:
All of the GCM models, following this logic, forecast that the atmosphere will warm more than the surface. However, it has not done so. Given this glaring flaw in the model forecast atmosphere-surface balance, I would be very careful about putting too much weight in a modeler’s claim such as the one that RC is making. The observed pattern of troposhere-surface warming may not support Gray’s claims … but it doesn’t support RC’s claims either.
Here’s some problems with their simplification. The greenhouse system, like all physical systems, has parasitic losses. These tend to cool the planet down. Evaporation is one of these losses. If we reduce evaporation as in their example above, the surface heats up more. But the atmosphere does not cool, because they have neglected the increased radiation from the hotter surface to the atmosphere and the need for global energy budget balance. As the surface heats, it radiates more to to atmosphere, which warms the atmosphere.
Since the total system must be in balance, in a perfect system (no radiation losses), the net result is shown in the following diagram:
Figure 1. Greenhouse planet with a perfectly absorbing atmosphere. S – Sun. E – Earth. W – Watts/m2 delivered by the sun. LH – Latent heat loss due to evaporation. Gray circle – atmospheric greenhouse “shell. a) Planet with latent heat loss.
Planet with no latent heat loss.
As you can see, removing the latent heat loss does not cool the atmosphere. The temperature of the atmosphere doesn’t change at all with changes in the latent heat loss, only the surface temperature.
Now the Earth’s atmosphere, of course, is not a perfect system. There are also radiation losses. Here is the situation including radiation losses:
Figure 2. Greenhouse planet with an imperfectly absorbing atmosphere. S – Sun. E – Earth. W – Watts/m2 delivered by the sun. LH – Latent heat loss due to evaporation. Lr – radiation loss. Gray circle – atmospheric greenhouse “shell. a) Planet with latent heat loss.
Planet with no latent heat loss.
Note that the atmospheric temperature does not depend directly on the latent heat loss, only on the net radiation loss. The problem is that both the radiation loss and the incoming energy “W” depend very sensitively on evaporation. First, evaporation increases the water vapor in the air, which decreases the radiation loss, the dread “positive feedback” the modelers correctly point out. However, evaporation also increases clouds, which reduce the incoming energy “W”, increases snow, which also increases the incoming “W”, and moves heat to the top of the atmosphere, increasing radiation loss. These effects are not correctly represented by the models.
The net effect of evaporation on surface temperature, atmospheric temperature, and radiative losses is one of the most hotly disputed areas in the so called “settled”, “consensus”, of climate science. Gray’s basic claim is not that “evaporation causes global warming”, as RC says. That is a straw man argument. His claim, which I believe is correct and which they ignore entirely, is that evaporation is overall a negative feedback to the climate system. To dismiss this claim out of hand by saying “in a perfect system, as the earth warms, the atmosphere cools” as they have done is both simplistic and incorrect.
The third of Gray’s “claims” that they attack is:
They say he did the calculations wrong, and they know how to do them right. If that is so, why is the GISS model provide such a poor match with the recorded ocean temperature? See here for how badly their understanding of the question has led them astray. If that is their best shot, I’d say no matter how he did the analysis, Gray’s claim that “ocean heat storage is inconsistent with CO2 as a cause of warming” has been proven by the RC folk’s own results …
I would certainly invite any of your students to “straighten [us] out on Gray’s paper.” However, linking to an error-filled page on RC won’t be enough, so they need to come prepared to do what we do here … take the actual statements in Gray’s paper and show them to be incorrect.
Again, Judith, let me say that I value your participation here, which is usually filled with interesting information.
w.
Damn it, teachers should teach kids how to think not what to think!
Re 59, errata, should be … “increases snow and thus snow albedo, which
the incoming “W”” …
w.
Following the 2006 season? It’s over already?
The paper is premature…but, maybe because the season has been a dud, might as well call it off.
Yeah, I was disappointed with Dr. Curry’s comments too about the Gray paper. And to cite RC as a source? What gives?
#59,
Posts like these are what I come here to read. Willis, thanks for another great post. Very informative, well-reasoned, easy enough to comprehend.
I need some help to understand Gray’s reasoning about the thermohaline circulation and global temperature:
Gray says, “When (a weaker thermohaline circulation) happens, the Southern Hemisphere oceans upwell less water into their upper mixed layer, and the globe gradually warms”.
He suggests that this weakness is what happened from 1970-1994.
It seems like less cool water upwelling in the Southern Hemisphere would warm the Southern Hemisphere 1970-1994. But, I think the Southern Hemisphere temperature has remained about the same in recent decades. If there was less Southern Hemisphere upwelling, why didn’t that hemisphere warm, and warm more than the North?
And, his statement about higher rainfall leading to more evaporation and cooling is a head-scratcher. I understand evaporation and localized cooling, but that evaporated water has to go somewhere and have its heat removed by condensation, which warms some other part of the atmosphere. Is he saying that IR is removed more easily at higher altitudes or latitudes?
Willis, I just spotted your post #59 on evaporation, which I am now reading.
Sorry, i am not going to delve into Gray’s paper. I am sticking either to published research or interesting arguments. 6 months ago there were a lot of critiques flying around of Gray’s paper, I am not interested in rehashing all that. Peter Webster posted a piece on RC (under the muddy thinking thread), that you may want to look at.
Gray’s paper is very significant in the public debate (but not the scientific debate). Gray got a standing ovation at the National Hurricane Conference (described in Achenbach’s article); note this is not a scientific conference. At the American Meteorological Society Conference on Tropical Meteorology and Oceanography a few weeks later, his presentation was absolutely hammered by the audience. A number of the GT hurricane students were in this audience, and we had also critiqued the paper thoroughly in class, which explains the reaction of the students to the climateaudit post.
Gray’s paper is rhetorically effective, and touches on issues that the broader public cares about (the same is true of JJ OBrien’s unpublished but widely circulated paper). This is in contrast to much of the published research with can be rather arcane and often seems irrelevant to the public, and is rarely rhetorically effective. In my BAMS article, I tried to capture this same rhetorical appeal and relevance in the context of rigorous scientific arguments that clearly identify the uncertainties. It is a very big challenge for science to get the research credibly and effectively communicated to the public.
I have read all of the papers that Gray has posted on his web site. Mostly the same story, with one interesting twist: he he “flip flops” on the sign of how the atlantic thermohaline circulation influences hurricanes.
The causes of the cycles in tropical storm intensity in the Atlantic is an open question.
But what we should know (according to Judith) is that they are not caused by changes in the Atlantic Thermohaline Circulation (as Bill Gray sometimes says.)
And we should know by now that they are not caused by global warming (despite all the published papers on the topic.)
So what causes the cycles in tropical storm intensity in the Atlantic?
The relatively low tropical storm season of 2006 is now being blamed on the development of El Nino conditions in the eastern Pacific. Why a person would say El Nino in the Pacific when the Gulf Stream in the Atlantic doesn’t work is beyond me.
It seems to me that there is no explanation for tropical storm intensity in the Atlantic then.
A chaotic climate which produces its own cycles is the only answer then (which is the topic of the board.)
#68: Jeff, the question of the effects of AGW on hurricanes is an open question, with ongoing research trying to answer it. If you had bothered to read the Gray paper you whould have known that the 2006 season was average but that most storms recurved in the Atlantic, before they had a chance to make landfall in the US. There is however, a well known link between El Nino conditions and a lack of TCs in the NATL. This is mainly due to changes in atmospheric circulation, El Nino has a global effect on climate. You might want to read El Nino in History by Cesar N. Caviedes for a good discussion of the effects of ENSO.
Bill Gray’s paper provided me with the word I was looking for: McCarthyism. But let’s back up a bit. On 23rd June, Steven Bloom posted this:
Re #368: Yes. Science moves on. Expect to see a whole lot more on foramins as a proxy for recent climate.
Comment by Steve Bloom “¢’¬? 23 June 2006 @ 10:15 pm
In hindsight, it is evident that Steven Bloom had a briefing on Mann’s recent foram paper three months in advance. It would have been a “keep the faith” briefing to the inner sanctum following the NAS panel, and it therefore follows that Steven Bloom is deeply embedded in the AGW camp. Then why does he have so much interest in Climate Audit? Could it be that he is the Sierra Club/Pew Foundation/whatever man assigned to Climate Audit, with the job of running interference?
Along comes Judith Curry saying that, although she is a professional climatologist, she hadn’t heard of Climate Audit before and stumbled across it by accident. After the Senate hearings, the NAS panel and everything else? She is quoted elsewhere that she believes that non-AGW views should be suppressed. So is she really interested in the full and frank free exchange of ideas at Climate Audit? Her post above was along the lines of,”My students are begging me not to, and this is against my better judgement, and I know I will regret this, and I have better things to do, and this is beneath me, but this is my dump on Bill Gray.”
It seems that the AGW camp has put another body onto Climate Audit. Steven Bloom has been deemed to need help.
This analogy really seems faulty.
First, we “know” the stock market goes up over many decades based on experience, not on a computer model.
Second, we know that the global temperature goes in cycles. It never just keeps going up. If so, the planet would be hotter than Venus. We know that someday there will be a new global ice age. We don’t expect the Dow Jones to fall to its lowest point in history on a cyclical basis, again, based on experience.
It would be instructive to see if the computer modelers could make a model of the stock market and the weather and make money off of it. I wonder how many modelers, for example, are making money off the wheat futures market right now?
Since the modelers were all predicting a big hurricane season, they would have lost their shirts this year if they had put their money where their mouths are. A lot of people did buy their prediction of increased hurricane intensity and frequency, and those people got hammered in the energy markets. Do the modelers feel guilty about this? Do they every feel any serious doubts about their theory.
Afterall, AGW is JUST a theory which can never be proven, and exists only for certain in a computer model which NOBODY understands.
Re #68 Jeff, I think what happens is that a warm Pacific (El Nino) creates hot air. That hot air rises and puts “extra air” into the upper troposphere. That extra air has to go somewhere, and it joins the normal upper air winds, which blow towards the Atlantic. The extra air speeds up those upper winds. This faster upper flow blows the tops off of storms in the Atlantic. A storm whose top is blown off stays weak. That’s how weather in the Pacific affect hurricanes in the Atlantic.
David
65, David:
One more note on El Nino. Here is a plot of the 2006 “wind shear” in the tropical Atlantic. When the upper winds from the Pacific get strong, the wind shear gets strong. Strong wind shear means weak, or no, hurricanes.
You can see that, in the last week or two, wind shear has gotten strong. It is stronger than the heavy black line, which is the climatological average wind shear for the date. I believe this is “the breath of El Nino”, so to speak.
Interesting that you would note that since Bill Gray and the National Weather Service and the National Hurricane Centre has been promoting the idea for many years. (But then they don’t anything about hurricanes do they?)
I’ve been following the El Nino / La Nina phenomenon for many years. Just when it seems like a strong El Nino or a strong La Nina is developing, the system reverses and nothing comes out of it.
You cannot tell if there will be El Nino until after the fact and the warm water conditions have existed into February for example.
Here is a good animation of the El Nino area over the past year up to October 1, 2006. It looks like El Nino is hanging strong (but last year’s very mild winter for North America happened during a La Nina event.)
http://www.cdc.noaa.gov/map/clim/sst_olr/sst_anim.shtml
Re #73 I don’t know. I would agree that the earth radiates more IR if warmth is transported towards the poles and dry air travels south (polar air is dry and aids radiation in the lower latitudes). Storms aid the transportation. But there are other aspects of precipitation and air parcels that I’m still trying to sort through in my head.
#70 Whoa. I wouldn’t say you are right but I have to say something is fishy.
And meanwhile back in the real world..the news just announced the price of orange juice is going up. The affects of Katrina, coupled with, get this: “a cold spell this summer” has produced a lagging crop of oranges this year. Are juice futures active when the wind blows and the frost forms? I wonder if you could use any of that data as a proxy in climate models? lol
#68,
Based on what I’ve read, El Nino hasn’t really played a role in this hurricane season. See the following post by Dr. Jeff Masters from Wunderground. Quite a few details about the hurricane season through Mid-August. And this one in mid September. El Nino doesn’t really appear to have played a role in any hurricane supression this year. Instead, it appears to be lower SST in the Atlantic, dry sarahan air over the Atlantic, a series of cold-core lows near the Bahamas that introduced significant wind shear in areas favorable for development, etc.
Speaking of RC’s take on Dr. Gray’s earlier presentation, I was a bit confused about whether he’s changed his position a lot; i.e. moved on in team-speak, or whether they’ve had his position wrong. Certainly whay they said about Gray’s take on the THC doesn’t seem to match what’s in his new talk at all. RC was talking about Dr. Gray saying that cold water mostly rose in the tropics whereas they actually rise in the southern ocean, but that’s just what is in the figure in the presentation. Can anyone help me?
Dr Curry, did you really say that? Regardless, given that your field is hurricanes, I wonder what you think of Richard Linzden’s statement (which you quoted without comment) that:-
Re #81, fossilization took place on 2nd February, 2006 in the Wall Street Journal. A larger quote is:
“Dr. Curry, in an interview at her Georgia Tech office, said Dr. Gray has “brain fossilization.” She added: “Nobody except a few groupies wants to hear what he has to say.”
Re: #82 — Judith Curry’s quote in the Wall Street Journal was reported in a Denver Westword, June 29, 2006 story about Bill Gray and the controversy surrounding him. The story is listed in LexisNexis. I’ve bolded Judith’s quote, and what Bill Gray remarked about it.
The original story is a bit long, so I cut it down to the relevant context, keeping the byline and source to show credibility. I have the entire story as a Word file if anyone wants it.
Here’s the bit.
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
2006 New Times, Inc.; Denver Westword (Colorado); June 29, 2006 Thursday
“The Skeptic; Celebrated and shunned CSU’s Bill Gray is taking heat in the global-warming debate.”
By Alan Prendergast
“”Had the last two years not had such bad hurricane damage we wouldn’t be talking about [them] now,” [Gray] says.
“Gray was hardly alone in his dissent. NOAA officials denied that global warming had anything to do with current hurricane patterns which were “due to natural fluctuations and cycles.” Chris Landsea, a former Gray student who now works as science operations officer at NOAA’s National Hurricane Center in Miami, re-crunched the numbers and found no increase in the number or intensity of Category 4 and 5 hurricanes globally over the past fifteen years; by his calculations even a four-degree increase in ocean temperature by the end of this century would have only a small effect on hurricane intensity. CSU’s Klotzbach zeroed in on the past twenty years of storm data, considered more reliable than that of previous years, and found a large increase in major North Atlantic storms and a large decrease in Northeast Pacific ones, but no global trend that would support the notion that global warming is whipping up stronger hurricanes.
“The rebuttals have prompted a flurry of additional papers and responses; Georgia Tech’s Curry for example, has suggested that Klotzbach is “cherry-picking” his data. But the most biting exchanges have involved Gray’s critiques and the counter-volleys. Gray has “brain fossilization,” Curry told a Wall Street Journal reporter a few weeks ago, and “nobody except a few groupies wants to hear what he has to say.”
“Stung Gray has responded in kind. “I’ve always liked her, but she doesn’t know a damn thing about tropical storms,” he says. “They want to put me on the fringe, sure. My brain is fossilized. I’m an old curmudgeon who doesn’t change with the times. They use anything they can against you.“”
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
Gray right or wrong, you were very unfair, Judith. On Roger Pielke jr.’s blog, you decried ad homs directed against you. But it appears you have an anti-Hermetian approach to them. They operate one way for you, but not the other.
And if people who think as I do are mere groupies, I’d like to see you produce the confidence limits of a GCM after propagating the parameter uncertainties and measurement errors through a centenary calculation.
If 1 sigma is 4 W/m^2 or less, I’ll change my mind about the A-part of GW, and magically transform myself from groupie into serious thinker. If 1 sigma is 20 W/m^2 (or more), then I wonder if you’ll have the scientific strength of mind to change your view and become a groupie before your erstwhile peers.
After all, if a typical GCM 1 sigma confidence limit is 20 W/m^2 (or more), then what could a good scientist credibly believe about the effect of a 4 W/m^2 forcing?
John A or Steve M., why did you remove my request to Carl that he post the 95% confidence limits after running the parameter uncertainties and measurement errors through a centenary GCM calculation? Isn’t that relevant to climate science?
The limits of predictive confidence are, after all, critically relevant to the credibility of theory-based modeling in every other branch of physical science.
Re #32 – “In the hurricane class on Tues, i asked the question “how many students think i am wasting my time blogging at climateaudit”. 8/14 raised their hands. They are now telling me they are changing their vote, and that i need to straighten them out on Gray’s paper. I am not going to bother, but they are working on Peter Webster, maybe he will post his critique.”
Another so-called consensus instead of asking the students for an analysis or critique.
Referring to RC for a critique is not science either.
At the risk of further upsetting Judith Curry’s students, I must confess that I also enjoyed reading Gray’s presentation. I don’t know if he is correct, partly correct or dead wrong. But perhaps this is not so terribly important. I value the fact that he dares to speak out his mind against the mainstream current.
Contrary to post #5, I forecast that virtually no-one here in Europe will hear a word about Professor Emeritus Gray’s speech in Washington. However, we did have politician Al Gore giving a speech in Spain the other day, so the “Science” sections of all Spanish newspapers (left or right wing) have been filled with his ominous predictions.
I think that it’s fair to say that a good half of all Spaniards now must know that if Greenland melts there will be 20 million climate refugees in Europe. And that a large portion of them has no reason not to believe that Greenland is actually melting very fast because of our CO2 emissions. Nobody will inform them that, under IPCC scenarios, it would take millennia for this to happen. Under these circumstances, is a sensible debate on CO2-emissions possible in countries like Spain?
Now, Meteorology Professor Gray did offer some valuable explanations as to how this nonsense has come into being, whether his scientific points are better or worse founded. And he’s daring to speak out, as opposed to those who silently watch this vast misinformation campaign.
Dr. Gray’s global warming section doesn’t have a lot that can be audited. One thing that can be checked is his statement that mid-level Southern Hemisphere zonal flow (west-east winds) weakened beginning in 2003. I checked the database, and I do not see the weakening he reports. I will look for the Lyman paper to see if it offers details.
What I see is a leveling-off, and perhaps that’s what he meant, but I struggle to connect leveling with the start of a cooling cycle.
I will look later today for the increased blocking he mentions and see if I can audit that.
I struggle to grasp his physical model and how its parts connect. That’s the same problem I have with Emanuel’s thermodynamic explanation in his 2005 paper. I’m an engineer and I guess I just need to understand how Gray’s GW (and Emanuel’s hurricane) model parts connect, and work, before I accept anything.
Mikel,
I have read Dr Gray’s presentation and I will reserve my comments on it, until Dr Curry confirms or denies that she said that:-
Dr Curry has been received with fawning good grace by many participants of this blog. To date, I have been uneasy about it, but have not doubted such a reception.
However, I have been somewhat disturbed about the ‘robust’ comment which she supposedly made about Dr Gray – viz. ‘…Gray has “brain fossilization,”…’
I should point out that Dr Gray’s and Dr Curry’s BSc are approximately 20 years apart. This, to me, indicates two things. First, Dr Gray was well on the way to being qualified before Dr Curry was born.
Second, if Dr Gray’s brain is fossilized and he is a mere 20 years Dr Curry’s senior,…well, I’ll leave it to her to explain what might happen to a brain in 20 years.
Now, I do not know Dr Gray. Nor do I know what he might have said to/about Dr Curry over the years. From what I have read of him, he’s opinionated, fiery and brash. He might well have attacked Dr Curry (or anyone else) with gusto in the past.
This may be the reason why she expresses such a strong opinion about him. Nevertheless, I would certainly be offended if I found that one of my colleagues had dismissed me or my contribution to my profession in such a flippant and derogatory way.
That is the reason for my #81: because I can’t believe that a scientist like Dr Curry could say such a thing about a colleague, unless there was some extreme prejudice between them.
Dr Curry, yours is a brave experiment. You have not only participated in the discussion on this blog, but you have also directed your students to it. Your have voiced strong opinions on this site and you have evidently expressed equally strong opinions about topics and colleagues in other places, as well.
You must be supremely confident about your research, your opinions and your peers.
I envy you your certainty.
BradH, you need to lighten up a little.
James,
No, I don’t. If I’d had a drunken rage in a pub about X or Y, I might need to lighten up.
Dr Curry has been quoted as making a most disparaging remark about one of her colleagues. If true, I don’t think Dr Gray would consider it to be “light” and entertaining.
He’s not here. I’m not defending his science. I don’t know whether or not he or Dr Curry is right. Nevertheless, I’d be a little miffed if someone told the Wall Street Journal that my brain was fossilized.
All I’ve done is invite Dr Curry to explain the comment.
Re: 90
Wall Street JournalWashington Post.re: 91
Whether originally in the Washington Post or the Wall Street Journal, it’s always hard to be sure exactly what the context of a remark was. Reporters like to soften an interviewee up with some light talk before or after the main interview so if they were running through a list of the major players in the field and when they got to Gray I could easily see her saying, “Oh his brain is fossilized” meaning that he wasn’t up on recent stuff. But even if she said that he was also a fount of wisdom, it’s the “fossilized” byte which would be quoted.
In the same way, if I said Dr. Curry was brainwashed about AGW, meaning she’d been around the AGW true believer crowd too much, and someone were to quote me, you can bet it’d be the word “brainwashed” that would be in the quote.
RE: #70 – “In hindsight, it is evident that Steven Bloom had a briefing on Mann’s recent foram paper three months in advance. It would have been a “keep the faith” briefing to the inner sanctum following the NAS panel”
I can’t rightly discount that. Mann is a UCB alum and maintains contact with many here in the Bay Area. He spoke at another UC campus, Santa Cruz, last year. Unlike the 1980s, when there was still a bit of distance between the mainstream scientists at UC campuses and the more radical fringe elements constituting the hard core environmentalist group (think “Earth First” for example) today, the 1960s – early 1970s generation now run the departments and the “climate” within the sciences has become radicalized. Therefore, the boundaries between the “climate science” orthodoxy and environmentalist community have grown quite blurred. Oh dang, I seem to be heading down the road of a social networking analysis, slap my hand, shame on me!
RE: #76 – that makes sense. The whole set up is determined by the ability of the trades and hence the main Equatorial Current to maintain the warm pool’s mass (or lacks of such abilities thereof). A relaxation in the trades makes it seem like an El Nino is imminent but then if the trades ramp back up before the pool can fully collapse, then no deal.
RE: #88 – What you allude to is a generational upheaval which has impacted, among other parts of society, the scientific community. I was fortunate, as a child, that my parents were part of that community. So I got to see the “old school” first hand, up close and personal. There has DEFINITELY been a sweeping change. Scientists, who used to overtly separate their political lives and work, no longer do this as strongly. Furthermore, the overturn of the socially and academically conservative campus environment, which prevailed until the 1960s, and nurtured the old schoolers, was gone by the time the current generation of scientists were undergrads. Something much needed and beneficial has been lost, maybe forever.
#45, Judith
Now that even Steve B has actually answered one of my questions could I also repeat the couple of questions I ask in a previous post.
“I thought he was playing politics. But, damn it, he was right.” Yourself in talking of Hansen’s 1988 congressianl testimony. So you think Jim Hansen WAS right? Given your use of past tense then does that mean that you no longer think that he IS right?
“Curry, who believes the skeptics have mounted a “brilliant disinformation campaign,”. As many of us on this blog are skeptics then a) are you saying that we are brilliant? If so thanks for the compliment b) are you accusing us of spreading disinformation on climate change on this blog? If so then I assume that you are in agreement with The Royal Society’s recent attempts to silence those who present an alternative explanation for global warming?
KevinUK
Re: 95 I’m beginning to think that climate science would benefit if reviewers were named up front, like my husband’s masters thesis review was. The reviewers had to sign their name to it as well. And if authors remained a secret until the review was finished-that would be even better. The lines are blurred. And my husband is part of the environmental community so we know first hand what can go on.
Re# 96
here’s another interview I found for Dr. Curry:
http://pubs.acs.org/subscribe/journals/esthag/40/i01/html/010106interview.html
re #97: Did you come accross this REUTERS story:
Are those quotes accurate, and if so, is it a scientist or an activist Curry speaking? What about those uncertainties, where were they when those statements were given?
#98,
Or this Reuters story…
How does this stuff even get coverage? Of course, since it’s primary sponsor is “Friends of the Earth”, they clearly would have no agenda to push. Yet anyone even remotely skeptical of these wild claims is somehow an industry hack or just too damn ingorant to “get it”.
RE: #98 – “The coastal regions are in jeopardy”
I would rewrite this as “Due to ongoing tectonic subsidence at the passive margins of the Continental crust of the North American Plate, combined with the trailing edge of Holocene scale sea level rise due to the trailing edge of the great melt of Pleistocene continental ice, combined with compaction of uncolidated alluvium and extraction of water and other liquids, the coastal regions along the Gulf / Atlantic Coast are in jeopardy.”
RE: #100 – “uncolidated” should have been unconsolidated…..
#96
Do you really need an example of sceptics using disinformation? Perhaps you recall the CO2 commercials from this summer? (These were decidedly un-brilliant.) I’m sure she didn’t name you by pseudonym, did she?
Another interesting aspect of climateaudit is the tendency of some posters to latch on to theories that go against AGW, but not to necessarily worry about unifying those theories. (Gray, for instance, does not implicate solar forcing in the observed warming.) I think this goes to what Dr. Curry says about hole-poking trial lawyers in the ranks here (And of course, I do not mean everyone, or even a majority). Anything against AGW is lauded by these posters or at least not held to the fire, without regard to how it fits the big picture. (Witness the posters who liked Gray’s presentation and do not know or care if he’s right [!])Well, the last part might apply to most here–no interest in unifying any theory, only auditing, debunking, and hole-punching. (Sorry for all the parentheticals [not really].)
#97, rocks thanks for the link
Some more questions for Judith based on a quote from that link.
“But you can’t use hurricanes to prove that there is global warming. What you can do is show an unambiguous link between the increase in hurricane intensity and the warming sea surface temperatures. And if you look for why the sea surface temperatures are warming since the 1970s, you don’t have any explanation other than greenhouse warming”
Judith I certainly agree with your first sentence. Just as hurricanes have been around since the earth developed an atmosphere so has global warming and of course global cooling (I assume you meant to say ‘prove that there is anthropogenic global warming?). Now could you do us all a favour and tell Al Gore that as he seems to think that you can.
Do you think that anyone has as yet shown an UNAMBIGUOUS link between increases in hurricane intensity and the warming SSTs? Can I ask why you think that “… you don’t have any explanation other than greenhouse warming”. Again I presume you meant global warming or perhaps anthropogenic global warming? Now I don’t think anyone on this blog disputes that the climate is changing. It’s been changing for hundreds, millions, trillions of years. There are perhaps very few on this blog who would disagree with the statement that the earth is warming just as it has similarly in the past also cooled (as recently as the 1940s to the 1970s).
What many of us on this blog dispute and challenge is the ‘A’ prefix added to GW and specifically the claim that the currently observed increases in mean global surface temperature are as a direct result of man’s continued unchecked use of fossil fuels. Some of us (myself included) object to the use of this natural phenomena (climate variability) as a means of enforcing changes to our way of life by certain individuals who have a clear eco-theologically inspired agenda, particularly when the claims by these individuals are based on proxy temperature reconstructions and the predictions of demonstrably tuned (to exaggerate the contribution of CO2 to the greenhouse effect through scientifically unsupportable positive feedback mechanisms) climate models.
KevinUK
#102 — “Well, the last part might apply to most here–no interest in unifying any theory, only auditing, debunking, and hole-punching.”
The basic point is that there is no unifying climate theory, Boris, but no one in climate science has the nerve to say so. No one understands why Earth climate is warming now, or why it cooled after ca. 1200, or why it was warmer prior to that. Climate theory is far too incomplete. It offers no definitive explanation for climate periods. In the face of that, demanding an alternative “theory” to AGW is identical to demanding one hand-waving explanation as opposed to another.
The only honest answer to climate change is, “We don’t know (but we’re working on it).” But the AGW extremists won’t allow a confession of ignorance because guilt-inducing accusation is a tactical necessity. And the politicized climate scientists don’t have the nerve to admit an over-riding ignorance; especially not, after making extreme public pronouncements of their own.
RE: #103 – The “tragedy of the commons” scenario of AGW which serves as the current norm of the orthodoxy, has been a godsend (er, maybe, godesssend?) for the early 1970s ecology green “e” crowd who have become embedded in the mainstream as they reached middle age. This version of AGW is the linchpin of the proposed program to try and implement globally the notions described in Ernest Callenbach’s books. Conveniently, that particular utopian urge coincides uncanily with a rising din of noise from deeply anti Western enclaves who desire to trim back Western countries as well as non-Western old line industrial countries such as Japan, South Korea and Taiwan. So here we have “constructive interference” between a Western intellectual utopian movement and geopolitical disturbances. To be fair, I reckon there are a good many amiable type personalities in the scientific community who’ve jumped on the AGW bandwagon without really giving it much thought. These may even be a significant majority.
re #99: That ain’t nothing 😉 , see this story from down under:
#102 you say we have : “no interest in unifying any theory”
No the real question is in my mind is, why are scientists and un-scientist alike exhibiting frenzied interest in unifying this theory?
What other scientific theories can you name that everyone and their mother should be interested in uniting just like this one? Is there only one? Surely this is a dangerous, hard and suffering world full of all kinds of kind hearts and big brains who know other things to save us.
Go on, name one other. Because I would be interested in if the scope of it in the media was the same. I’d be interested in the subject and the cause/reason behind it too. Does it get the same attention? Are there Heros and Villians? Do people die? Doomed? Hurt? Helpless?
Think about it- from MTV to Congress, to the UN to every weekend on the Discovery Channel, the news papers and blogs, across the Pond, across the Globe and to the Poles and inbetween..to an ex-President and ex-Vice onto Hollywood and beyond. Can you name any other Scientific Theory on the books that is that “important” and is filtered into our lives just about every single day?
#100 🙂
Yes, Climatetologists should come back down to Earth once in awhile.
Re #102 Boris, have you noticed those posters who don’t buy the global warming hysteria they are being force-fed and are happy to see a few scientists speaking against it but will also not blindly buy the latter’s science (for the same reasons)?
#98, Jean S
Thanks for the link and the quote. Judith care to comment?
“We have a 10-year window to do something about greenhouse gases” this quote sounds eerily familiar? Judith is Curry your maiden name and your married name Hansen?
And what do you think of this quote from Frank.
“Scientists who believe human-induced global warming is linked to hurricane formation and strength rely too heavily on numerical models, Frank said.
These same numerical models that I can’t put faith in for a two-week forecast, we’re told can be accurate out 200 years,” he said. “Ridiculous.”
KevinUK
Pat, you miss the point Boris is making. It’s not that there has to be any unifying theory, it’s that criticisms are not being made thoughtfully (are just fluff…stuff that jae comes up with…like percent CO2) AND that the criticisms are self-contradictory at times, but are met with acceptance regardless. It’s like the excuse that the village elders gave Napolean for not announcing his arrival:
1. Bellman is sick.
2. Bell is broken.
3. Don’t have a church.
Capisce?
#106,
Well, apparently my scare story isn’t nearly as scary as your scare story. lol…I guess my question for Dr. Butler is since those things are currently occurring, what do we blame them on now? Still global warming, or some other reason. If some other reason, why would the reason change from that other reason to global warming. Would global warming made me do it make a better excuse than we hate freedom loving westerners? I don’t know, just a guess.
#100 T “I am banned Timmy” C O,
You are in it for the drama.
All that matters is the data, and it stinks. I know of several scientists who will tell you so, and they don’t post on blogs or speak to the media. They are actually out there working and cleaning up the earth of all its troubles too, not “publishing” and becoming famous or contributing to the BS.
So this blog being here or not, doesn’t matter to the A/GW theory or to unifiying it for that matter. The data should speak for itself and it doesnt’. Never did and hasn’t yet.
re “brain fossilization”. It is interesting to see that the climateauditors are prepared to believe whatever they read in newspapers, but not in the scientific journals. Climateaudit is as good a place as any to set the record straight on this. There is a very interesting story surrounding the WSJ article in question, and some of this was obliquely discussed in my BAMS article:
“The media has played a significant role in inflaming this situation by reporters’ recitations of what people on the other side of the debate are allegedly saying. One reporter manufactured a personal conflict between the first author of this paper (including an egregious misquote) and a scientist on the other side of the debate that have had no personal contact in several years.”
The journalist who wrote the article, Valerie Bauerlein, appeared to be writing a serious article on the scientific controversy, and both Kerry Emanuel and I worked with her over a period of weeks. She came to Georgia Tech to interview both Peter Webster and myself. During the course of the interview, she kept bringing up the acrimony between WHC&E and Bill Gray, and she had heard this had resulted in the cancellation of a debate at the American Meteorological Society. We kept trying to steer her away from that topic (which we thought unfortunate, but essentially irrelevant in the overall scheme of things) and back to science. When I saw the published article with “brain fossilization” in the headline, I said “wow, I wonder who said that”. You cannot imagine my astonishment when these words were attributed to me.
The scientists quoted in the article (plus RP Jr) were all involved in a lengthy exchange of emails immediately following publication of the article which lasted several days (this was sort of an email “group hug”). All of the scientists participated in the exchange with the exception of Bill Gray; RP Jr followed up on this and found out that Bill Gray did not want to discuss this with us. I excerpt some of the statements from emails that I wrote as part of this exchange:
“I have read the WSJ article and I am apalled and astonished by it (Peter Webster and Greg Holland are as well). I thought that I was speaking to a responsible journalist in the prestige (and hopefully responsible) press. I am further appalled and astonished by what has happened over the past 6 months in terms of this debate. I am trying to understand it.
I spent a considerable amount of time with Valerie Bauerlein talking about how the scientific debate and even the scientific process was being confused and inflamed by the media, and that this is to the detriment of everyone, especially to our field. I further stated how we were concerned by NOAA scientists being restricted in their public statements and even in their research directions on this topic. I further stated my concern that owing to this inflammation that debates at professional societies are being cancelled, and the public debate is occurring the media and on blogs, which is to the detriment of our field. The peer reviewed journals will eventually record the responsible scientific debate, but the peer reviewed journals are way too slow at this point given the public interest in this. In fact we even publicly stated all this in our presentation at the AMS meeting (which is attached), and I believe that valerie bauerlein was even at our talk. I believe that these are important and valid points for science, journalism, policy makers, and the public. Instead, she chose to tell a much different story, playing on the inflammation to make it worse.
I have personally worked hard to promote civility in this debate and to try to understand the broader forces of what is going on here. In the beginning, i assumed that “us scientists” would agree to disagree and continue with our research, and see where it led, looking occasionally in amusement at the media as they tried to sensationalize this. That is not how it has played out.”
“The sad thing is that i think almost all of us are alot more civil to each other than what has been portrayed, i think this is a terrible example of sensationalist journalist magnifying a few words. In responding to the question raised by Bill about how did the experts who actually know something about hurricanes miss this trend if it is really there, I stated something like it often takes an outsider to look at something with fresh eyes. The hurricane community had its mind made up on this, and they weren’t even looking. I’m sure I stated something like Bill Gray’s opinions on this subject were set in stone,
and I didn’t see anything that would change his mind. Whether or not the word “fossilization” ever got used, neither Peter or I remember this. (added note: if you look at the sentence in the WSJ article, it says Judy Curry says Bill Gray has “brain fossilization”, with quotes only around “brain fossilization.” There is no way I said that sentence, it simply isn’t how I talk) But if it was, it was in the context of opinions set in stone. Late Wed afternoon, I got a call from Bauerlein asking how old I was. I was busy with something else, I just told her without thinking. This should have made me suspicious. I am just appalled that she took my sentiments stated above and turned them into an ageist ad hominem attack.”
For the record: during the last half of 2005 there was arguably a “pissing match” (direct quote from Jeb Bush during our meeting with him) between WCH&E and Bill Gray, although the various initials had different levels of involvement in this. I note that this was just with Bill Gray and did not involve Landsea, Michaels, OBrien or any of the other scientists on the other side of the debate. The problem that we had with Bill Gray can be found all over the media (ad hominem attacks, appeal to motive) and of course the Achenbach article with “Gore as Nazi”. These quotes published in the media were confirmed by Gray’s emails sent to the scientists involved and also posts to the tropical listserv (there was no question here of Gray being “misquoted” by the media.) One such email (one of the ones that Valerie Bauerlein referred to) resulted in the disinvitation of Gray to participate in a panel debate (after this disinvitation, one of the scientists on Gray’s side of the scientific debate said “Bill you are pissing in your own soup.”)
The pissing match is OVER (RP Jr, take notes), at least to the extent that two sides are needed for such a match. As an aside, in spite of my characterization in the WSJ article as “chief pisser”, I was barely involved in the pissing match, since I am not really a member of the tropical meteorology community and Bill Gray thought I was pretty irrelevant (he was mainly going after WH&E) and I have had no personal interaction with Gray in many years. After the “Gore as Nazi” statement, Gray was reprimanded by the higher administration at CSU, and he has cleaned up his act since then. We do not regard Bill Gray as a player in the scientific debate since he hasn’t published anything, and hence we ignore him. We are all a lot savvier than we were early on (Peter Webster and I have undergone “media training”; see also section 4 of my BAMS article; you have no idea how difficult it is to deal with the media; by contrast with blogging you can get whatever message across that you like). What we find unfortunate is that Gray is such a big player in the public debate on the issue, in spite of having not published anything on this topic.
That said, I fully understand the value of Bill Gray’s expertise, and agree with Lindzen’s comment but I think it needs elaboration. Bill Gray is an “old school” meteorologist, which takes great stock in the experience of the forecaster. In the 50’s-70’s, before there was “reliable” numerical weather prediction, the skill of a forecaster rested in the ability to see “analogues” of the current weather pattern with past patterns and so anticipate that the weather system would evolve like the previous analogue. As time went numerical weather prediction models far outstripped the capabilities of an individual forecaster using the analogue method, and the “experience” of the forecaster became less of an issue. However, in hurricane forecasting, the analogue method is arguably still of some use since numerical weather prediction models still do not do a fabulous job with hurricanes. Bill Gray’s unique contribution to all this was to extend the analogue approach (combined with some statistics) to seasonal hurricane forecasting. Gray’s 50 years of experience in watching hurricanes makes his knowledge of hurricane analogues unique in the world. However, I would argue that this particular expertise does not translate into value in the global warming debate, and its value is becoming increasingly questionable in the seasonal forecasts of hurricanes (a post on this to follow).
With regards to my personal relationship with Bill Gray: I do not know him well, but I always enjoyed my interactions with him (and his wife) when I lived in Colorado and we ran into each other a few times per year. I have not had any personal communication with him since I left Colorado over 4 years ago. As part of the email “group hug”, I did apologize to Bill for the statements that were attributed to me. I do not intend to have any future interactions with him unless he apologizes to WHC&E for his behavior. I will pay attention to any paper that he manages to publish, but I am no longer paying attention to his media statements or unpublished opinions. Bill seems not to mind that he has burned his bridges (as per quotes from the Achenbach article).
End of story. I am still amazed that this ended up on the front page of the WSJ (probably the journalist was told that if she spiced it up, it would make the front page).
I expect this is the bit Dr Curry hates the most:
According to Gray, hurricanes don’t vary with average temperature. So we can add hurricanes to the same bin as Bristlecones etc as proxies for global warming.
RE: #113 – Judith, while a bit sad, the inaccuracy of the WSJ’s portrayal of you is hardly surprising. You can count on the main stream media to behave that way.
Re # 98: Yawn. For a group who claims to be interested only in rigorous science, you spend a huge amount of time on on stuff in the popular media and attacking people rather than arguments. Trust me, if i ever say anything at all controversial in the media, RP Jr calls me on it. Roger caught that quote and asked me about it (i didn’t even notice it until roger pointed the article out to me), I looked for the relevant email and can’t find it (Roger was sufficiently convinced that there was nothing in that statement and he didn’t post anything about it over at prometheus). A misquote. I remember discussing the general issue with the reporter, but i did not make such a statement.
Many reporters already know what they want you to say before they talk to you. I have no idea how to avoid being misquoted, but i have taken to writing things in email rather than talking on the phone. This seems to help. The media training helped somewhat.
This is part of what the BAMS article is all about. I suggest that those of you who couldn’t get past figure 1, go back and read the rest of the article.
I am through clarifying statements attributed to me in the media. If you want to know where i stand on policy stuff related to AGW, check out my statements over at Prometheus (and Roger’s characterization of them). Roger refers to me as a “policy agnostic”, which is not entirely accurate. I do engage in the policy process, but do not advocate for specific policies.
Tell me, Dr. Curry, do you lambast Dr. Mann and his associates equally for all of their ad-hominem attacks?
Mark
Dr. Curry thanks for the dissertation on the “fossilization” misquote. Unfortunately it does not get the discussion back to the science content of the matter posed by Gray’s speech.
What I have found a bit disconcerting was that when Pielke and Gray (see excerpt from Gray below) refer to the number of land falling hurricanes and damage in constant dollars not changing significantly with SST over the decades, Emanuel agrees, but does a statistical analysis that says the increases in hurricanes and evidently their intensities would (will??) not necessarily change, from a statistical perspective, the land fall damage or frequency.
The question arising out of these statements would obviously be can we agree that Emanuel’s statistical analysis of land fall probabilities are valid in this matter and, if we do, than what should our increased concern (if any) be, given the Emanuel and others’ evidence of a trend for increased hurricane frequency and intensity that apparently can be shown not to effect land fall frequencies or damage.
Steve, I propose a new thread “Auditing Gray’s statistical seasonal forecasts of North Atlantic Hurricanes”.
Information on what goes into the Gray/Klotzbach forecasting scheme is described on Gray’s web page. Off the top of my head (insufficient time to check owing to frivolous posts re the media), it is something like this. They have a statistical model which is a regression equation with something like 6 predictors (SST, wind shear, etc). Then there is an analogue forecast (say, the conditions look like 1967, therefore we can expect x), then the final forecast is mysteriously different than either the statistical or analytical forecast (guess? Voodoo? Divine intervention? Seriously I have no idea what determines the “actual” forecast). For example, if memory serves me correctly, for the 2006 seasonal forecast in June, the statistical forecast said 10 named storm, analogue said 13, then the actual forecast said 17 (interesting that the statistical forecast of 10 looks about right for 2006).
Greg Holland did an evaluation of the skill associated with Gray’s forecasts (which have been made since 1984), which I cited in my congressional testimony (Figure 9) at
Click to access GT%20-%20Curry%20Testimony.pdf
Holland’s analysis was presented previously at an AGU Workshop on Hurricane Katrina, and RP Jr did not like this one bit
http://sciencepolicy.colorado.edu/prometheus/archives/scientific_assessments/000860another_problem_wi.html
http://sciencepolicy.colorado.edu/prometheus/archives/climate_change/000877follow_up_on_critici.html
It seems like people don’t want to hear what Holland has to say about Gray’s forecast owing to the pissing match factor (is this nuts or what). Holland’s analysis was discussed on the tropical listserv also, but this mainly focused on what is skilful vs what is useful, and sticking with prescribed skill scores (skilful if it does better than just picking the 50 yr mean).
Independent auditors needed! I am calling on Steve/Willis/Bender to take a look at this
This is really important, and it is low lying fruit, no messy data problems, you can probably get a paper ready for publication in a week.
Key issues (I’m sure you can think of more):
1) is the interpretation of Hollands analysis in my testimony correct/appropriate?
2) What kind of skill score is appropriate in the presence of a moving average?
3) Does it make sense to use 50 years of data to build a regression based forecast model in the presence of a large 70 yr cycle
4) what do you make of the combined statistical/analogue/guess scheme?
And just think, if you figure out a better way to do this, you can sell your forecasts (or keep them secret and use them in energy trading, etc) and support your blogging activities.
I think I have the data on the “actual” forecast somewhere, but I suspect that Phil Klotzbach would be happy to provide you with whatever info you might ask for (need to get the full data set on the statistical, analogue, and actual forecast). I think focusing just on the June forecast is sufficient.
Yeah, we could audit it. I have no doubt that Gray can’t predict the hurricanes that well. I don’t trust the Farmer’s Alamanac either. Jude, are you sure there’s not just a little bit of pissing contest motivating you to show what I already suspect? 🙂
Re #119 Ken, for a plot of landfalling named storms since 1851, see figure 4 of my testimony. you can see a strong 70 yr cycle in the landfalls, ostensibly associated with the AMO. the most recent data is slightly higher than the previous peak activity ca. 1950, but certainly no statistically significant increase from 1950-2005 (at least this is my inference from eyeball analysis). However, if the next peak of the AMO is ca. 2020, then the “analogue” for the decade 1996-2005 is 1926-1935. This would imply an increase in the past decade (actual statistical sorting out of all this is needed).
But lets say there is no increase in the number of landfalls, and the intensity doesn’t really show much of a difference either. However, damage increases have been huge. RP Jr. says all the increase in damage is owing to demographics, although his competitor Mills claims a signal from AGW. I have no idea which one is correct, but i don’t think it matters.
Our concern for the future is that the combination of increasing coastal development and the combination of marching up the AGW cycle plus AGW should result in unprecedented damage in coming decades.
Re: #120
Dr. Curry you are sounding more like some of the hardliners here. Your multiple references to pissing contest certainly has a TCO edge to it. Before you go beyond the point of no return you might want to reconsider your students’ urgings and also answering my query about concern for trends in land fall hurricane damage given Emanuel’s analysis.
Re #121 actually this is a “test” for climateaudit to see if you are prepared to audit the work of an anti AGW scientist.
I think that Gray does reanalysis of his work to determine skill level. I have never looked at that.
We posted simultaneously so disregard my answering request. Of course, I should have included the increase in coastal development in factoring the land fall hurricane damages.
Given that climatologists were willing to state or present statistical analysis that would mininmize the constant dollar/ constant population hurricane damage even with increasing total hurricane frequecy and intensity as would be the historical case (given Emanuel/others and Gray/Pielke) then what you point to as future problems for damage would be more an insurance issue, where the true risk of coastal living is taken into account (without government subsidized insurance or government subsidizing the uninsured or uninsurable), and not really as much of a climatology concern.
Bill Gray website
Looks like this contains their current forecast as well as electronically-stored forecasts back to 1999.
In glancing at a recent forecast, they have a section and report on their skill-check,where they see how their forecast actually did against climatology.
Their skill method looks rather simplistic to me.
Judith, I’ll be glad to offer a critique of their forecast methodology, once I finish my final comments on this storm-intensity issue. I will read Holland but I doubt I’d have any opinion on ways to measure forecast skill, if that’s what his piece is about.
David
And while we are checking on Dr. Gray, is anyone going to do an analysis of RC Hurricane forecasting skill? Or don’t the experts forecast hurricanes?
Judy-
Re: #120
The problem of course with what “Greg Holland has to say” is that no one (besides yourself apparently) is privy to his analysis, which has already been used as the basis for a consensus AGU report and your Congressional testimony. As you know the AGU report has been contested by all of the seasonal forecast teams. When asked for a copy of his paper on the TS list Greg declined. The IPCC won’t cite work before being peer-reviewed; it is a good policy. It is quite diffciult to accept claims that are counter to the existing peer-reviewed literature when they are not made available to examine, even if the claims are made by a a brilliant scientist. Be careful to distinguish a “pissing match” from simply asking people to follow widely accepted assessment practices 😉
With respect to your question #1, can you make Holland’s analysis available for people to examine? (Otherwise how can anyone weigh in?)
Also, before sending folks here off on a wild-goose chase you might also have noted that Phil Klotzbach shared with you, me, and the rest of the TS list a spreadsheet will all of the Gray team forecasts and skill baselines. An “audit” of this information would probably take all of about 15 minutes. Landsea, Saunders have provided similar information, including references to peer-reviewed literature. You know all of this of course, so why not just share it here?
Re: #122
Can you cite one peer-reviewed paper showing an AGW signal in the hurricane damage record? You cannot, because there is no such paper. You even have written that it will be 10 more years before the debate on hurricanes/AGW is likely to be settled. I believe that you know all of this as well, so your comments here are intriguing.
As an observer of all of this, it is interesting to note how the disucssion here has morphed away from science and data analysis. Any reason not to finish what you all had started? Looked promising . . .
re: Judy’s recollections in #113 and #116 — all of this is exactly as I recall these events as well.
I don’t think I’ve ever called anyone a “policy agnostic” however, what Judy might be looking for is “stealth issue advocate” — someone who claims (often sincerely) only to be focused on science or “policy relevant” science, but who is in fact advocating, or being used to advocate, specific policies.
Judy and I got off to a bit of a rough start with the ACS EST interview, but since then I think we’ve decided to agree to disagree on certian matters in a cordial way. She has alwauys been a welcome and valuable contributor on our blog, even though we do not always agree.
Thanks.
WOW!
Jude, I got that. I’m too steps ahead of you. No problem on passing that simple child-like test. Let’s really get into it, though. Let’s rip into Landsea!
We all appreciate these comments. But for the record Dr. Gray’s current forecasting page says:
So he’s been making these forecasts since the 80’s. The 50-70’s thing was a tad leaning toward the fossiled brain theme again. 😉
I guess we can look forward to a more indepth look at the methods.
PS.”anti AGW scientist” very insightful, that label is, we think.
#107:
How would you expect people to act when they believe AGW is real? Hands in pockets? Even if they think there’s a 50% chance of AGW, they should be out talking about it and doing somehting about it.
And here’s your analogue: cancer and smoking. Let’s check your criteria. I’d say the scope is the same, or close to the same in the media. People died. There were certainly heroes and villains. Hurt? Check. Helpless? Check. So there you go.
I will take a few minutes to clear up misconceptions re my previous posts.
Re #129 Roger, it takes anyone 15 minutes to reproduce Greg Holland’s analysis. You plot 20 numbers times 2. then you calculate bias and RMSE for each 10 year interval. This is why i suggested that the climateauditors check this. Re Greg Holland, I understand that he prepared a manuscript that contained this analysis in it. He said he would make the paper available once it was accepted for publication. I know nothing more about the paper. But it is a trivial analysis that is easily replicated. Re Holland not providing you with what you asked for, arguably a pissing match situation after you made that accusation on your blog.
Re AGW and damage, I am referring to the paper by Mills, perhaps you could provide the reference for this. I have no idea about whether any of these damage analyses are actually useful (although the insurance companies certainly want them). I think these analyses are naive since they do not account for such factors as changes in building practices, improved farecasting and warning/evacuation practices, etc., I have mentioned this to you before. I have stated in my BAMS article and in countless publications that i believe that the best explanation that we currently have for the increase in hurricane intensity involves AGW. I have further stated that projecting anything into the future about hurricanes is very dicey, and I have stated that we won’t see a clear and unambiguous signal from the observations esp in the NATL for another decade. There is nothing inconsistent in these statements. Does all this mean we don’t have to worry about hurricanes? @#$%^ no!
Re #128 Climate researchers are looking at the European seasonal forecasts using coupled atm/ocean models. These forecasts just went operational this past year but are not made public outside of the EU (this has been slow since it involves cooperation between three different govt meteorological agencies). The first paper on this is being writting by Frederic Vitart. I hope that next season these forecasts will be available operationally in the U.S. Climate researchers are not yet in the business of making seasonal hurricane forecasters, but I am sure that it is on the verge of happening. We need good statistical/dynamical forecasts that are based on coupled atm/oce simulations (e.g. without the ocean coupling, we are just guessing about whether there will be El Nino or not)
Re #133 Yes Gray has been making the seasonal forecasts since 1984. To do this, he uses data since 1950. Gray has been in the hurricane forecast business since the 1950’s (but not making operational SEASONAL forecasts since then. This is not an ageist slight by me, this is something Gray is very proud of, that he has been doing this stuff for 50 years (see the achenbach article)
Phil Klotzbach is now taking over the forecasts. For a graduate student, he has shown a huge amount of poise and maturity in dealing with controversy and the media
Yawn. Can we please get on with the science. I am finished on this thread unless someone digs into Gray’s forecast or brings up an interesting scientific issue to discuss
I compiled some of Gray’s forecasts going back to 2000. There re links that reference forecasts back to 1994 but they appear broken.
I have looked at analysis of economic forecast performance in the past. One of the easiest rudimentary ways to take a first look at perofrmance is by refernce to Mean Absolute Errors (MAE).
I took Gray’s one year ahead forecasts, which he makes in Decemer each year and compared with the “actual” out-turns he publishes at the end of the year.
In order to extend the limited data by on year, I used his latest forecast for 2006 (published this month) as a proxy for 2006 actuals.
If you just look at the MAE for for his year ahead forecast for the number of hurricanes, this comes to 3.0. On average he is out by 3 – although a look at the data shows this is skewed by a single large outlier in 2005 (13 hurricanes).
What does that mean? Well not a lot on its own. We need to compare it to some form of naive forecast. A look at the statistical properties of hurricane numbers give a guide to what form of niave forecast you could employ. If you plot a histogram of the data you see what looks gaussian (yes, there are only a few data points). So a single fixed forecast would be a decent starting point. Using the fame “eyeball” technique, 8 would appear a decent forecast. And if you had forecast 8 every year from 200 to 2006, you would have obtained a MAE of 1.9 – significantly superior to Gray.
However, that isn’t exactly fair, because that would not have represented an ex ante forecast. So, instead, we could use what Gray calls the “climatological” (I assume that means long-term average) of 5.8 to guide our niave forecast. So I could have forecast 6 each year from 200 to 2006 and this would have delivered a MAE of 2.4 – still superior to Gray’s 3.0.
This doesn’t shout out forecast skill to me. The caveat is the limited amount of data that precludes any statistical inference.
Of course the next step is to get a bit cute and deploy some of the additional properties, such as those discovered by Bender that show a significant 5 year partial correlation, to build an even better niave forecast, but in this case it doesn’t seem necessary.
I might extend this to the other variable that Gray forecasts, such as the number of tropical storms etc. But given the fact that these will not be indpendent events the additional information this would give with regard to forecast skill would be limited.
#116: Yawn. I’m rather tired of you “grouping” and “categorizing” us all the time. To my knowledge, I’ve never met anyone here personally. Only three persons (Steve M., Mark T. and Eduardo Z.) are supposed to know my true identity as I’ve been in contact with them offline. Even they know basicly nothing about my attitudes to AGW-controversy or politics in general. I think my situation applies to most of people writing here.
I was not attacking you, I simply linked to an article where there were quotes supposingly coming from you. I found those quotes rather strange considering that you have said here that you are concerned about policy, communication, and uncertainty issues related to your field. And furthermore, it took only few minutes to google that article (try your name and google search for the news groups), which I did because I was inspired by welikerocks (#97) to see what comes up with your name. I did that after checking if you wanted to have a real scientific discussions on the issues I was ineterested in (as I proposed in #42) and noticing that apparently (#67) you were unwilling to discuss those issues. For that (reading Gray’s presentation, and Ruti et al and Rinke et al papers) I did spend some time (apparently for nothing from the discussion point of view).
I note that Judith Curry uses Bill Gray’s 1984 study tying the El Nino ENSO wind shear to Atlantic hurricane intensity as a reference in her 2005 Science article.
http://www.sciencemag.org/cgi/content/full/309/5742/1844
http://ams.allenpress.com/perlserv/?request=get-abstract&issn=1520-0493&volume=112&page=1649
She also uses his 1968 study GLOBAL VIEW OF ORIGIN OF TROPICAL DISTURBANCES AND STORMS as a reference in the same study.
He is also a coauthor in another reference.
http://www.sciencemag.org/cgi/content/abstract/293/5529/474?ijkey=dd43ffcd7e46342173d9246086b3cefdc8326e2e&keytype2=tf_ipsecsha
It is interesting to note that Bill Gray was publishing about this topic many years before any one else was. It is interesting that he is still publishing as a coauthor in 2001. (I thought someone said he just likes to use the popular press and doesn’t like publishing.)
He might be a fossil but maybe his vast experience has allowed him to see the issues with the right perspective.
#120: Judith, I’ve been following from the side all these hurricane threads, and personally I have a problem starting to “audit” these papers (including Gray’s “forecast schemes”): from my point of view the methods used in all of these papers are “outdated”. I simply would not model those things the way they are doing. IMO there are much more advanced methods to attack those kind of questions available in the literature. In fact, I have a few models in my mind how I would try to predict hurricanes and trying to figure relative factors contributing to hurricanes. So instead of auditing few papers, would you be willing to put some effort on actually improving the models used in hurricane papers? If so, I can be contacted at jean_sbls@yahoo.com
#107
Hi. I would expect scientists to act like scientists, not politicians or public leaders. A 50% chance of being correct is not enough in any field of science to publish or present a theory. You’d be laughed at.
I guess smoking/cancer is an ok example, even a movie about the tobacco industry. 🙂 I won’t get in detail, and they are trying to tax it again in California too… but we are off topic now so I am being short sorry!
Judy-
Your cavalier dismissal of work outside of of your own expertise is pretty surprising coming from you. This is now a second time that you have criticized or mischaracterized my work on this site, only to then follow up that you will have no more to say once I have responded to your mischaracterizations.
Be aware that your views on factors responsible for hurricane damage trends are about as solid as Bill Gray’s ruminations on the “water vapor feedback”. You probably shouldn’t criticize him for making scientifically unsupportable statements in public if you are going to do the exact same thing.
For instance, you write, “I think these analyses are naive since they do not account for such factors as changes in building practices, improved farecasting and warning/evacuation practices, etc.”
Can you cite any research on the role of these factors in the damage record? Or how did you arive at these conclusions? Guess? Voodoo? Divine intervention? (Joke! Couldn’t resist!;-) I’ve discussed these factors in my work and (a) building practices have led to more not less damages, and (b) forecasting/evauation saves lives but has no meaning effect on damage. If you’d like to discuss the science of attribution of damage, let’s indeed do that. If you’d like to simply dismiss my work without engaging or understanding it, well don’t then crticize those who do the same with respect to your own work.
Bottom line: If there is no long-term trend in US landfalling storms (frequency/numbers or intensity/PDI, a point on which Landsea and Emanuel agree), then how can climate factors (regardless of cause) explain any part of the dramatic increase in U.S. damages?
2. For anyone interested in the evaluation of seasonal forecasts:
Lea, A. S. and M. A. Saunders, How well forecast were the 2004 and 2005 Atlantic and U.S. hurricane Seasons? in Proceedings of the 27th Conference on Hurricanes and Tropical Meteorology, Monterey, USA, April 24-28 2006.
Click to access AMS2006Forecast.pdf
Saunders, M. A. and A. S. Lea, Seasonal prediction of hurricane activity reaching the coast of the United States, Nature, 434, 1005-1008, 2005.
Click to access Saunders_Lea2005.pdf
From Klotzbach/Gray:
http://tropical.atmos.colostate.edu/Includes/Documents/Publications/seasonalskill.xls
http://tropical.atmos.colostate.edu/Includes/Documents/Publications/augustskill.xls
http://tropical.atmos.colostate.edu/Includes/Documents/Publications/septemberskill.xls
http://tropical.atmos.colostate.edu/Includes/Documents/Publications/octoberskill.xls
Owens, B. F., and C. W. Landsea, 2003: Assessing the skill of operational Atlantic seasonal tropical cyclone forecasts, Weather and Forecasting, Vol. 18, No. 1, pp.45-54.
Click to access operational.pdf
The community had this same debate on the skill of ENSO forecasts a while ago. The debate revolves around how rigorous a naive baseline to use as the metric for “skill”. Given that there are multiple possible baselines (e.g., 50 year climatology, 10 year, 5 year, 2 year, etc.), it is easy to search for one that proves a pre-determined point of view. According to Klotzbach and Saunders their forecasts are skillfull according to these simple baselines.
At some point the calculation of a baseline goes from being naive to being sophisticated. Where this boundary exists is of course subject to legitimate debate.
Judy-
Interesting (emphasis added):
You write: “Re Greg Holland, I understand that he prepared a manuscript that contained this analysis in it. He said he would make the paper available once it was accepted for publication. I know nothing more about the paper.”
In your congressional testimony you cite Holland’s paper, reproduce a paragraph of findings (on your p. 10) and a figure from it (your figure 10).
Click to access GT%20-%20Curry%20Testimony.pdf
Sounds to me like you have far more than just an “understanding that he prepared a manuascript.” Let’s engage in good faith, and not play games.
Thanks!
#137 Jean, OK, I will go back to the regional modelling issue, that is a legitimate one. In my scanning of the site, I miss plenty of comments, sometimes i choose not to reply to a post and other times i simply miss the post. I fully understand that there is a huge distribution of people (motives, expertise, opinions, etc). I view you as voice of reason on this site (was particularly suprised that you went to the effort to dig up that article and post it).
I would argue that global climate models do a credible job of reproducing the major features of the atmospheric and oceanic circulation and the surface climate, and that the “attribution” simulations (natural forcing only, AGW + natural, etc) are credible to the extent that the external forcing is credible (there are still some issues arguably with the solar forcing but this doesn’t seem to be a huge factor, there is definitely an issue with actually quantifying the magnitude and distribution of the big aerosol forcing mid 20th century). Projections into the future that are well outside the current climate regime (say > 1C warming) are uncertain owing to the “tunings” that have been done for the current climate; i.e. the sensitivity of the climate models in the doubled CO2 regime is questionable.
Now onto the simulation of regional climate. At the coarse resolution that most climate models are run (the resolution of the GISS GCM is particularly coarse), detailed regional climates simply are not produced. combine this lack of resolution with uncertainty in forward projections, then I agree with RP Sr that projections of regional climate change in doubling CO2 scenario is highly uncertain (the IPCC says this also).
There has been considerable efforts to develop regional climate models. Filippo Giorgi has been the leader in this. The idea is that regional models can be run at high resolution, but they need to be forced on the outer boundaries by results from climate model simulations. Arguably they should look at an ensemble of outer boundary forcing, but they are usually limited by computer time and spend the computational wad on resolution. There have been a number of simulations for the U.S.; regional climate models have done a pretty good job in the U.S. (for detailed critique of all this, see RP Sr’s blog).
Now on to the Arctic. People are starting to use regional climate model simulations for the Arctic to look at details of what is going on up there. I initially started the ARCMIP project to see if looking at high resolution regional simulations could help us understand how to better parameterize the physical processes in the arctic like clouds, radiation, boundary layer etc, since the climate models were not doing a fabulous job on these in the arctic. The first ARCMIP intercomparison project described by Rinke did not really accomplish these goals owing to problems with the regional models that participated in the project. Some of the models had been used mainly for weather forecasting rather than climate, and for example their radiative transfer models were a joke (with one exception, not of the caliber of what you would find in global climate model). All of the models had a problem with the upper boundary condition; without going into details, global climate models take care of this issue, and this issue doesn’t arise in lower latitudes away from the pole. Etc. So the bottom line is that the regional models participating in ARCMIP are performing for the most part worse than the global climate models in this region. This study has motivated substantial effort to improve regional climate modelling in the Arctic, see http://www.awi-potsdam.de/atmo/glimpse/ for a big EU funded project that is taking this on. BTW, most of my funded research is to work on improving parameterization of physical processes in the arctic towards getting the climate sensitivity correct in that region. Much work to be done.
but back to the main point. Errors in simulating regional details does not invalidate climate model simulations, which are designed to capture the larger scale features. There may be important small scale regional feedback process that end up being important to the global climate; this is arguably part of the uncertainty that i cited in future climate projections that go far outside the current climate regime.
re #141: Thanks Roger for interesting links! You say
I have a question. In that debate (or any other similar debate) did it ever come up to check what other fields are doing in order to improve these prediction models? I mean prediction is fundamental to say, e.g., econometrics, and can be also viewed to be a basis of a part of whole scientific field (Information Theory/compression).
re #142 Roger, puhleeze, this site does not need to get involved in the Pielke-Holland pissing match. I did receive a copy of Holland’s manuscript. I asked him if I could use one of the figures. he said yes. My statement “I know nothing more about this paper” referred to the fact that i know nothing more about the fate of this paper in terms of where it is at in publication process.
re #143: Thanks, Judith. I’ll comment those later (tomorrow?), but now I’m offline rest of the evening.
Steve, could you raise Judith’s comment (#143) to an own post, as I think there are plenty of people here interested in that topic? So we would have an own topic for GCMs and uncertainties therein.
Judy-
If you (mis)characterize my work here as “attacks” and “naive” (and miscite the literature, e.g. Mills) then I will respond substantively to correct these errors.
This has nothing to do with Greg Holland, other than the fact that you have asked for help “auditing” his work and his work is not available. If it takes 15 miuntes as you suggest then why audit? I’ve got nothing against Greg, he is a good scientist. Your repeated invocations of a “pissing match” are completely unfair.
How about this as a resolution – you don’t attempt to characterize my work on climate impacts or climate policy unless you are serious about engaging in the substance, and I’ll return the favor with respect to your climate work? Deal?
Thanks!
Dr. Curry,
puhleeze be aware that science isn’t the only method of looking at the world around us, and this “animal in the zoo” is using a few methods of her own to discern much of this and what’s going on. (For one thing, I’d like to know if you made any attempt to correct these apparently huge mis-quotes attributed to you in the media? And you were not specific about any of what you did say, and I like specifics-you got minus points there from me 😉 Also the label “anti-AGW scientist” is very telling.
#138
I am feeling a deliberate though slight negative inflection about old data here too. I wonder, could we do the same for the HS and it’s “other temperture records” which are probably just as “old fasioned”? I doubt it. ..those double standards rearing up their ugly head again we point out [over and over] on ClimateAudit? 😉
Re #141 Roger, some statement or evaluation from you regarding the uncertainties in your study would be appreciated. I have gone to great pains to point out the uncertainties associated with my papers and research. As I have stated to you numerous times before, I am not interested in debating your paper on hurricane damages.
The issue of past trends in U.S. hurricane damages does not tell us anything about whether global warming has influenced hurricane characteristics. Looking at U.S. landfalling hurricanes statistics does not tell us anything about whether global warming has influenced hurricane characteristics. An excerpt from my BAMS article on this:
The sampling errors associated with using statistics on U.S. land falling hurricanes to infer statistics on global hurricanes or the cause of changes in hurricane intensity can be assessed directly by using the same global data set employed by WHCC for the period 1970-2004. Atlantic hurricanes comprise only 11% of global hurricanes, and U.S. land falling hurricanes comprise only 25% of North Atlantic hurricanes and therefore only 3% of global hurricanes. While regional data records are useful for documenting regional change and supporting some aspects of regional decision making, understanding the causes of the variability requires that global, regional, and local processes all be analyzed to interpret the mechanisms responsible for the change.
Until we have a better understanding of the relative roles of AGW, AMO, etc we can’t definitively say whether or not AGW is contributing to increased U.S. hurricane landfalls.
You have put forward a hypothesis. I think there are too many uncertainties in such an exercise to draw definitive conclusions. But this is not one that I personally care about sufficiently as a scientist to spend much time on it.
Jude:
1. Are you satisfied with the evaluation of Gray’s forecast skill or do we need to do more to pass teacher’s test?
2. Just ruminating: on the GCMs, I really don’t see why a weather model should be expected to predict climate forcings well:
-given all the parameters and assumptions
-given that it does not skillfully predict one year out climate (versus naive baseline)
-given that it has a regional structure, but does not well predict regional effects, in fact sometimes produces rediculous regional effects.
3. I get the impression that there is sort of a broader community of paleo guys, modelers, experimenters and general climate predicters. And that within that social scope, it is ok to acknowledge that individual fields have work in progress, but probably not ok to fundamentally question the basic skill. But paleo recons are pretty sketchy…and so seem GCMs. They’re the kind of field where so much work has gone on that it’s easy for the workers to convince themselves that they must have something. But from the outside, one still sees fundamental questions, which leave the possibility that there are MAJOR problems with the fields.
rocks: settle down. Don’t get into an alpha female match.
No TCO I won’t, I have teachers and principals repeating those mis-quoted words to my children as fact.
#140:
Your 50% statement is absurd. So if the weather forecast calls for a 50% chance of rain, you shake your head and say “I will not prepare unless they are 100% certain.”? Me, I carry an umbrella.
If an asteroid is discovered to be 50/50 for collision, we should just sit back and wait till we’re positive? You’d be laughed at.
As for what scientists should or should not be doing, look to those against AGW for ample examples of people using the media for their goals.
See…you’re projecting.
#153 no it isn’t in scientific standards in regards to a theory. Your feelings are exactly why this theory, huge with errors is the perfect propagana tool and you are falling for it.
#143
I realize this will be deemed off topic but it seems to be the practice to avoid questions and consistently misdirect the debate. (It may underline the call for a model thread). It also seems to be the practice to claim there are more important things needing attention. (Going to the bathroom seems to be another important urge.) Since arctic climate models were used as a misdirect and apparently deemed more important I will ask: What station records are used to create the grids and parameterization for the arctic? Do they even come close to the World Meteorological Organization requirements for density of coverage? How large is a region? How do the models define arctic? Since this is a regional study and the point was made that regional models are still dependent upon and limited by the input from the periphery and this information is provided by GCMs, how much validity is there? The peripheral areas include mostly Canada, Russia, Greenland and Alaska in an almost circumpolar embrace and none of these areas has adequate density or length of records. How much data is available about heat energy coming through the ice from the much warmer ocean water beneath? I realize the comment was made that much work needs to be done, but I fail to see how adequate historic data can be generated to meet requirments. As Jones and Wigley said in 1994 “Many of the uncertainties surrounding the causes of climate change will never be resolved because the necessary historical data are lacking.”
155: And the editors of Science have fallen for it. And all major scientific organizations. And Shell Oil. And BP. And the Governator.
What is your theory, rocks? Where are your observations? Do you agree with Bill Gray’s work? Or with solar? Or with Lindzen’s Iris Effect? Or with Idso’s disbelief in CO2’s physical properties? Or is it all of the above? Does it matter to you that even the small number of non AGW scientists can’t agree on anything? Does it matter that their theories contradict one another? Is it just fashionable to go against the grain? You said you love specifics, and then you become curiously vague.
Your feelings are exactly why gainsaying this theory without any coherent argument is the perfect propaganda tool, and you have fallen for it.
Re: #153
Boris, your examples lead nowhere since they provide incomplete information and that is part and parcel to the AGW issue and what if anything is done about it.
You would first need to know what the odds of occurrence are along with the uncertainties of predicting it. You would have to know not only the statistics of the occurrence but the specific and overall effects of that occurrence. You would have to have knowledge of predicted costs of mitigating the occurrence and the uncertainties and unintended consequences attendant with the plan(s). Finally you would have to deal squarely with the question of whether all the potentially involved people are ready to make the recommended changes.
Umbrella example has little consequences of being wrong and involves individual choices — hardly analogous to the AGW issue.
The example with an asteroid discovered with a 50/50 chance of collision with earth has implications that we know the physical laws and can make some rather precise calculations about its trajectory very unlike the current case for AGW. Assuming this is a sufficiently large asteroid, we would know precisely the consequences of the occurrence, i.e. end of life as we know it on earth and again very much unlike the case for the consequences of AGW and for that matter the level of the effect of AGW. The mitigations for the asteroid collision are no doubt much more clearly known, understood and with attached costs than is the case for AGW.
AGW has to be handled as it is, not by some other references to examples that on more than surface analysis can be shown to be very different.
Boris, the Hockey Stick is broken. All the “falling for it” is about money and politics and getting votes.
My husband is an environmental geolgist, think about what that means, because he’s probably just as smart as any of the big players here. The data being presented to him (and then me too) for AGW is not up to his standards as a trained scientist, and he wouldn’t be allowed to work if he operated under those same standards. Putting it your way: There’s a 50% chance or more it’s all natural climate change if you look a the geological record.
So when anyone acts certain about any of this at this moment in time it is only going to be social, political, and ideolgical agenda driven.
It’s not up to us to disprove a theory, it’s up to the theory makers to produce a valid one, within scientific standards.
Thanks Ken. 🙂 Nicely put.
Ken,
I understand your point, and I was, of course, being simplistic. I fully recognize the differences in my examples.
My major point stands: uncertainty is not sufficient reason for inaction. Some of the things you list can never be known with 100% or even 80% certainty. And since they can’t, many will use them as an excuse to do nothing.
sheesh sorry I keep dropping my “o”s in the ology-logical words!
#143
That seems to be a basic philosophical assumption that is made by the modeling community, but I’d like to know if it is, or even can be, based on a rational argument. It sure looks like an act of faith.
The problem I see is that we certainly can capture the larger scale features, and GCM’s are obviously a great achievement. But then we seem to be using them to make fine forecasts. It would seem that the accuracy of a forecast would be limited by the scale of the features that are simulated by the model. The assumption is that the errors on, say, grid scale, average out when looking at the global picture. That’s OK if the errors are random. But if there are systematic errors, this assumption breaks down. I don’t know if what I’m saying makes any sense. Maybe this has been discussed in the litterature before. If so, I’d like to have references.
Boris: maybe you’re on the wrong blog. This group is not generally anti-AGW. We are just anti-crappy science. The reason the blog may appear to be anti-AGW is that we are having a hard time finding credible scientific papers that support AGW. You got any?
Dr. Curry, Dr. Pielke, others:
re: not trusting the media, I can only offer yet another brilliant essay by Dr. Charles Doswell.
Media training is one option, asking them to pay for your service is another.
BTW, it looks like the tropical research folks are due for another group hug. 😉
rocks,
More and mere hole-poking.
Why don’t those against AGW relaize that uncertainty has no politics, that it can be positive or negative? Anytime uncertainty is talked about it seems as if a lot of people here say “too much uncertainty! Everything’s cool!” But what if the effects are worse than projections? What if CO2 has more of an effect than we understand now?
Anyway, my point is that many posters here grasp at any article that doubts AGW, and furiously attack any article that agrees with AGW. If that’s not ideologically driven, then I don’t know what the phrase means.
It’s very easy to sit back and make claims about the hockey stick being broken. It’s harder to put together a theory of warming that goes beyond magic natural cycles.
You have to do better than a sound bite view of the AGW issue to effectively argue the case and potential mitigations. It is a very complicated issue and the least that one can do is spell out some of the uncertainties involved in the predictions, mitigations, cost of mitigations, and of course, in the end whether generational transfer of costs has a chance of approval given the unknown uncertainties. You might start with the Kyoto accords and the real world application of them.
Re: #165
Kenneth B thanks for the link which in one sentence quite clearly points to a problem that even the casual reader must see — if they want to.
#164 Yep. And to argue fractions of C° is mind-boggling to my husband also.
#166
Natural cycles are magic and the HS being broken is just an illusion?
Ok I’ve heard everything now. LOL
Judy-
Re: #149
1. You write, “As I have stated to you numerous times before, I am not interested in debating your paper on hurricane damages.”
My response: then why do you repeatedly critcize my work in public? Or is it that you’d rather to criticize and then just not have me respond? 😉 (Note, there are about a dozen relevant papers on the subject that i’ve participated in, not just one paper.)
2. You write, “some statement or evaluation from you regarding the uncertainties in your study would be appreciated. ”
My response: I thought that you said you’d didn’t want to debate this issue? 😉 On uncertainties in disaster damage data as collected by the US NWS, please see this paper:
Downton, M. and R. A. Pielke, Jr., 2005. How Accurate are Disaster Loss Data? The Case of U.S. Flood Damage, Natural Hazards, Vol. 35, No. 2, pp. 211-228.
Click to access resource-796-2005.31.pdf
3. You said, “You have put forward a hypothesis. I think there are too many uncertainties in such an exercise to draw definitive conclusions.”
How is your uniformed speculation on this subject any different than Bill Gray’s musings on “water vapor feedback”? It is quite easy to come to quick conclusions about areas of research in which one has little expertise or experience (sounds like some of your students critiques of CA;-). For an expert panel’s (more than 30 researchers from 13 countries) consensus conclusions on this subject see:
Click to access workshop_report.pdf
Full report will be out next week on our WWW site.
4. You write, “The issue of past trends in U.S. hurricane damages does not tell us anything about whether global warming has influenced hurricane characteristics.”
Of course it doesn’t (a strawman?). You have the interesting line of inquiry exsactly backwards. Understanding the role of hurricane climatology in the historical damage record does tell us something about how future changes in storm charcteristics due to AGW might influence future patterns of damage. Hence understanding trends and causes in damages is directly relevant to understanding the policy relevance of the hurricane-climate debate. It turns out that the debate over hurricanes and global warming is of very little practical policy relevance because, as you have written, the “main hurricane problem facing the United States [is] the ever-growing concentration of population and wealth in vulnerable coastal regions.”
So long as people continue to use the prospect of future hurricane damages as a justification for actions on energy policies, then studies that place the hurricane science debate into policy context will be quite important as a means to evaluate the policy merit of such proposals.
Re: 144
Jean S.-
You ask:
“In that debate (or any other similar debate) did it ever come up to check what other fields are doing in order to improve these prediction models?”
There has been in my view far too little of such cross-disciplinary efforts to understand how to evaluate predictions.
That being said, the atmospheric sciences (in particular weather and hydrology) have made tremendous contributions to forecast evaluation. In particular the work of Alan Murphy is seminal. We presented a comparative assessment of predictions in policy across the earth sciences in this book:
Sarewitz, D., R.A. Pielke, Jr., and R. Byerly, Jr., (eds.) 2000: Prediction: Science, decision making and the future of nature, Island Press, Washington, DC.
http://sciencepolicy.colorado.edu/homepages/roger_pielke/prediction_book/
On the hurricane-climate issue, Bill Gray has drawn a lot of attention, but there are other groups providing seasonal, and in one case 5-year, forecasts of activity. My sense is that at some point the critics of existing seasonal forecast methoids will eventually have to move beyond criticism and put forward their own forecasts!
rocks,
You’re the one who falls back on “natural climate change” (BTW a beautifully non-specific term) as if it can magically explain away the warming. What are those changes? Solar? Volcanic? Oceanic? Orbital? Gremlinic?
Who cares, right?
I’ll take the NAS opinion on the hockey stick.
#172 boris
what exactly do you propose to do that will have any measurable effect?
I for one find Judith’s references to “pissing contests” quite tedious. Trying to claim moral superiority based on one’s gender doesn’t reflect well on Dr Curry. It also suggests that Dr Curry has had a sheltered life, if she would care to study the delights of UK nightlife, she would see examples of female “pissing contests”.
It really doesn’t add to the debate Judith and it debases your arguments.
#172 Boris,
If you want to explain the Little Ice Age, the Medieval Warm period, the Younger-Dryas event, the Ice Ages in general, you’ve got to resort to “natural” causes. Obviously, to generate such a variety of different climates, the “natural” causes can be many. I think that it’s only once we have fully understood ALL these causes that we’ll be able to state with confidence that the recent warming can ONLY be attributed to GHG, because “there are no other explanations”. This “we can’t think of a better explanation” reason is often given as a “proof” of AGW. It seems to be the foundation of Dr. Curry’s adherence to the theory. Fine, but then you can’t just dismiss, for example, the “cosmic ray” forcing, saying that it’s a “long, long way”. There has been a ton of scientific research focused on the GHG explanation, and a minuscule amount on all sorts of solar-related forcings, this despite the fact that the Sun, volcanoes, and orbit fluctuations have, historically, been the MAIN driver of our climate, and that we still, to this day, have a very poor understanding of how they caused its historical variability.
Francois,
Yes, but unamed and undiscovered “natural forces” cannot be a valid reason for ignoring CO2 increases and CO2’s physical properties.
#176 Boris,
But we don’t actually know whether the CO2 increases are detrimental, insignificant or benificial.
#176 Boris,
Sure, but you should know that, by itself, the “radiative” forcing of CO2 is relatively minor. Then you have the so-called “water vapor feedbacks”, which are not “feedbacks” per se, but amplifying factors that increase the “pure” CO2 forcing by a factor or … (put preferred number here). And then you have the “cloud feedback”, which nobody seems to agree on even what sign it is. Funnily enough, playing around with the cloud feedback value can generate about any prediction you want.
So why is it so outrageous to think that a cosmic-ray modulated cloud formation mechanism could potentially affect the values that we attribute to those feedbacks?
The fact is: up to the “radiative” part of the GHG forcing, AGW is pretty solid. Once you get into the water feedback, it gets a little shakier. But when you add the cloud feedback, then we’re on very, very soft ground. That a lot of climate scientists convince themselves that they’ve got the right answer anyway doesn’t mean we should follow them blindly.
#176: Boris, certainly, but climate sensitivity to CO2 is still unknown with a high degree of certainly. Francois’ point is entirely valid – attributing the error variance to a politically correct cause, while quite common in many fields, nevertheless is not science and is unacceptable.
> the “radiative” forcing of CO2 is relatively minor
Well, exactly.
We add, at an unprecedented rate of change, to the natural background a volume of a gas, a relatively minor forcing, with a long atmospheric residence time, and the natural sinks aren’t collecting it as fast as we’re adding it.
As Winston Churchill once said:
“Madam, we’ve already established that. Now we are haggling about the price. … ”
We’re lucky we didn’t create an energy system and technology that caused comparable releases of methane.
Imagine if the Victorian era had begun by dredging up methane clathrate for steam-driven technology, instead of mining coal — burning a little and wasting the rest. That’s a major forcing. We were lucky.
Like we were lucky with chlorofluorocarbons, relatively weak catalysts of ozone breakdown –as Molina pointed out in his Nobel Prize lecture, bromine compounds were the competitive alternative that industry could have chosen:
nobelprize.org/nobel_prizes/chemistry/laureates/1995/crutzen-lecture.pdf
http://www.atmos.washington.edu/~davidc/ATMS211/Lecture8-slides-PDF.pdf.
Dr. Curry, you refer to “#121” above, which is this posting:
http://www.climateaudit.org/?p=857#comment-51954
as a challenge from your students to the auditors.
Note — your challenge posting is now #120. It’s not safe here to refer to postings by number. Whenever a posting has been deleted, all after that get renumbered, and if a lot have been deleted it becomes impossible to figure out what’s meant.
Instead — right-click on the timestamp for the posting you want to mention, “copy link” and paste that into your cite. Those should always work.
Re #180: This is a good time to mention that many of the “anti-AGW scientists” (e.g., Sallie Baliunas) got their start in the skeptic business by questioning the effect of CFCs. A fine track record, IMHO.
Ee #182: Which, like Acid Rain, turned out to have been much to do about little.
182 bloom
I supose then that we should discount all Mann et al efforts today because of hockey stick mistakes in 98?
Mann’s problem wasn’t that he made mistakes but that he hid data and methods.
This Bill Gray thread has me a bit confused when I go back and look at what it is we are attempting to accomplish in the framework of the more general debate.
If Bill Gray’s predictions are shown to be good or not good what does that have to say about the points that he has attempted to make about the weaknesses in computer modeling? If the predictions are not good it might reinforce the argument that climate/weather forecasting is a task too complicated to do with any certainty at least at the current time. If they are good does that mean Bill Gray’s methods lend weight to his criticism of computer climate modeling? Or can we surmise he was simply lucky? Or that his predicting abilities do not translate to criticisms of computer modeling?
Another feature of the thread is what do all these predictions and attempts to find a trend with SST and hurricane intensities have to do with government policy. It would appear to me that an honest assessment of the situation would reveal that we are very uncertain of hurricane trends and intensities and particularly when applied to land fall hurricanes. Even data used to gauge past hurricane activity is in significant doubt. Past damages by hurricanes are subject to much uncertainty as a quantitative measure and particularly so when all the factors of demographics and reporting are taken into consideration.
What is the one significant element of hurricane damage that we can probably all agree upon and what do we see in terms of policy to mitigate it? Without any significant increase in hurricane activity and intensity we can all agree that damages will continue to climb at what appears to be an exponential rate. And what is government policy in this regards? Has it been to insure that the risks are evident to those living in the affected areas and that the market place is allowed to estimate the risks in terms of insurance rates and in property values? The answer to that question has to be a flat no. The policy has yielded to politics. And what does that portend for a policy to mitigate AGW effects?
We were told that Katrina was a category 5 hurricane at land fall in New Orleans and the Army Corp of Engineers had warned that the flood systems would only withstand category 3 hurricanes. We found later that the land fall category was a 3 and that the levees were flawed in design and construction. Has the national media diligently gone back and corrected the original misconception? Not that I have seen? What does this portend for policy involving the Army Corp of Engineers or for that matter to mitigate AGW?
Re #185 Ken, here’s a Katrina story. Perhaps the most-important canal break was the 17’th Street Canal. One of my relatives routinely jogged down that canal levee before the storm. He knew exactly where the break was, because the levee had eroded so much at that point that it was hard (too steep) for joggers to jog. Bottom line: neglect.
The US Federal Government took the heat for the disaster but the truth is that widespread incompetence at the local and state level is the real story.
(I’ll try to make up for this off-topic by posting on Gray later tonight.)
My first check on Gray was to confirm that what he says are actual storm season figures (count, storm-days, hurricane-days and intense hurricane days) agree with the Unisys database.
I checked the 1999 and 2002 seasons. All numbers agreed, except that the 2002 intense hurricane-days are 3.0 according to Unisys and 2.5 according to Gray. I consider that to be a minor difference. Also, the higher number (3.0) would actually make his performance look better, so he has no motive to report a smaller number.
My second check is to confirm that what he says he forecasted are what he actually did forecast. I compared his seasons-end numbers for 1999-2005 and compared them to what was actually published earlier that season (circa June 1 forecast). All agree.
Simple spot checks, confirming that the actual data do agree with the actuals reported by Gray. So far, so good. Next, forecast accuracy.
Judith, thank you for a marvelous challenge. I agree that this needs a new thread.
The challenge is to audit, and possibly beat, William Gray’s hurricane predictions, and along the way answer your questions, viz:
Let me answer one question now, the others I’ll answer in a following article showing how it is possible to beat Bill Grays’ forecast, at least in the short term … and as an added bonus, I’ll give you my forecast for 2007 …
First, is Hollands analysis correct? It is neither correct (bad data, and more important, bad math) nor meaningful. You say of Holland’s analysis:
I cannot replicate these numbers. I used the numbers from the graph you showed, and I get the same average and RMS for the 1984-1994 period. I also get the correct average 1998-2005. However, I do not get the RMS error of 5.2. I believe the correct answer is an RMS error of 4.2 … see, this is why we need auditors.
However, whether we use your numbers or mine is immaterial, as the analysis contains no error bars. If it had had them, it would have been clear to both Holland and yourself that there is no statistical difference between the early and late periods of the analysis.
The 95% confidence interval of the bias error of the first period is from -1.6 to +0.2.
The 95% confidence interval of the bias error of the second period is from -6.2 to 0.0 (using my figures, which give a smaller interval than yours).
Since the confidence interval of the later period almost totally envelops that of the first period, we cannot say that they are different. In addition, neither of them are significantly different from 0.
The problem is that the time periods are just too short, which makes the error bars too wide. I fear that Holland’s analysis is useless.
w.
PS: Finally, you say “no messy data problems” … but Gray and Holland can’t even agree on how many Atlantic hurricanes there are in any given year. Some of the differences I understand. Gray does not count:
1. The final storm of 2000, which didn’t receive a name except “SUBTROP” meaning subtropical storm, and did not attain hurricane strength.
2. The first storm of 1997, same reason.
3. The first storm of 1992, same reason.
4. The first storm of 1984, same reason.
5. The final storm “Zeta” in 2005, I assume because it didn’t attain hurricane strength until 2006.
6. Hurricane Otto in 2004, I assume because (per Wikipedia, which for this kind of thing is invaluable … but don’t believe their climate “science” articles)
In other words, although it was named, upon re=analysis it was demoted.
Now, counting subtropical storms as “Named Storms” seems a bit flakey to me. Could you ask Holland why he did that? He’s included four storms in his analysis that are not named storms, then used bad math to blame Bill Gray for not predicting them …
My very best to you, Judith. Your scientific spirit continues to shine.
On subtropical storms, Gray’s articles clearly state that he is forecasting “named storms”. If a storm did not get a name, it is not part of Gray’s forecast. So, Holland should not include subtropical storms in any evaluation of Gray’s forecasts.
The effect of this subtropical issue should be small, though.
Better measures of forecast skill for Holland to use are storm-days, hurricane-days and intense (cat3,4,5) storm days. These are not so affected by the short-lived, minor “trash storms” and they give a better indicator of threat.
(Here are some examples of how “storm-days” are calculated. If a named storm (depression,tropical storm or hurricane) exists for 24 hours, then that is a storm-day. If it existed for 18 hours, then that is 0.75 storm-days. If two storms exist on the same day, each for 24 hours, that counts as 2.0 storm days. And so on.)
How does Gray do on storm-days? Here are the 1999-2005 storm-day actuals and Gray’s forecasts:
1999 forecast storm-days: 75
1999 actual storm-days: 77
(Historical average: 49)
Pretty good.
Here are other recent years:
2000 forecast:65
2000 actual: 66
2001 forecast: 60
2001 actual: 59
2002 forecast: 55
2002 actual: 54
2003 forecast: 70
2003 actual: 71
Those are excellent numbers. To continue,
2004 forecast: 60
2004 actual: 89
2005 forecast: 75
2005 actual: 115 (unsure how Zeta, which extended into 2006, was handled)
In 2004, 2005 and 2006, Gray’s accuracy in forecasting dropped. But I am impressed at how accurate his 1999-2003 forecasts were. Gray’s methodology has some skill.
The link to pre-1999 reports at Gray’s website did not work for me. If the link starts to work, I’ll look at his older forecasts.
Willis, I look forward to reading about your prediction approach.
OK, as promised, answers to questions, and forecasts.
I used a three step method to achieve skill in the forecast. I defined doing better than Gray as achieving both lower average error, and lower standardard deviation, in predicting named storms. Gray’s results were actual less forecast 1984 – 2006, average error 0.87, std. dev. of error, 4.01.
My requirements were that I could not use any information not known at the time of the forecast, including long-term averages or regressions, etc.
The three steps of the method are:
1. Remove the trend.
2. Remove the autocorrelation.
3. Remove the residual error.
I started by removing the trend from the number of storms. This is very similar to the end-point problem in using a smoothing filter. We want to determine what is the best guess for the smoothed curve, either at the end-point of the data, or a certain number of periods beyond the end point.
For this purpose, I use a truncated gaussian average of the past data. This uses half of a gaussian smoothing filter, which has its highest value at the last data point. While not different conceptually from a linear trend, it has the advantage of being more responsive to higher frequency changes in the data. The result of the trucated filter (which is always one year behind the analysis) are shown in FIgure 1. I used the filter width which worked best in the pre-1984 data.
Figure 1. Results of Truncated Gaussian Filter. Filter width (3 standard deviations each side of center) is 25 years, but only one half of the filter is used. Residuals are the original data less the lagged, filtered data.
Next, autocorrelation. Looking at the pre-1984 data only, I determined that the largest autocorrelations were at lag(1) (negative), and at lag(5) and lag(10) (positive). A linear regression on this information with the pre=1984 residuals of the truncated gaussian filter showed that adding the lag(5) data to the regression only made a trivial difference. Using just the lag(1) and lag(10) data, the regression between the autocorrelations and the residuals was:
I used this to remove autocorrelation noise from the data.
Finally, to remove the average error, I simply added the average error from the previous nine years of my forecasts to my new forecast.
My forecasts vs. Gray’s forecasts, along with the data, are shown in Figure 2.
Figure 2. Named Storms (Gray’s data), Gray’s forecast, and my forecast.
Finally, I told Judith I’d answer the rest of her questions, which were:
What kind of skill score is appropriate in the presence of a moving average?
In the presence of a moving average, I have not found anything to beat a truncated gaussian average to determine the underlying trend. Here’s a comparison of the long-term trend from the start of the data (1948, in this case) extended to each current year, the truncated gaussian average, and Gray’s forecast.
Figure 3. Alternate measures of skill, along with an actual forecast.
There is a fourth measure in the drawing, which is a trailing average-adjusted truncated gaussian average. Compared to the truncated gaussian average, this one usually has a lower average, but a higher standard distribution. To choose between these two in answer to your question, Judith, requires that we specify the nature of the analysis. Do we want to compare against a trend which does better on average but is scattered, or one that does worse on average but whose errors are clustered?
Gray’s forecast is much more skillful than the moving trend, and is also more skillful than the truncated gaussian average, whether average adjusted or not.
Does it make sense to use 50 years of data to build a regression based forecast model in the presence of a large 70 yr cycle
What do you make of the combined statistical/analogue/guess scheme?
Gray’s purpose with his forecasts, it seems to me, is to get them right. I have done my forecast using a combination of truncated gaussian, autoregressive, and trailing moving averages. My purpose was to do better than he did, and I was able to do so. Gray’s results were:
Actual less forecast 1984 – 2006, average error 0.87, std. dev. of error, 4.01.
My results were
Actual less forecast 1984 – 2006, average error 0.73, std. dev. of error, 3.94.
I take no great joy in this, as it may fall on its face next year. It’s a brute force method. About the only thing it does show is an 11 year cycle (solar?) in a short hunk of data.
Gray’s analysis has sounder scientific underpinnings, but also contains other elements. Like all forecasts, or like science in general for that matter, the key question is, does it work?
Gray at least is putting it on the line, and making that key step for science to succeed, which is making falsifiable statements and predictions. This is where the climate modelers haven’t got the … … nerve … … to step up to the plate and say something like “an ensemble of 11 climate models predict 16 named storms next year” …
If their models are so great, if they have as much faith in them as they claim, surely they can forecast that better than an old guy sitting in Colorado, no matter how brilliant he might be.
So let me join Gray here. Since I am working only with data that is one year old, I can make my forecast now, at the end of the hurricane season. So I now stand up and say that my projection for the number of named storms in 2007 is … goes back to Excel … ok, 18 named storms.
Now of course, this may be way wrong … but it’s more than the modelers are willing to put on the table. So in answer to your final two questions, Judith, I can only say, if his method works, that’s a good thing.
It’s an underlying problem with predicting chaotic systems such as hurricanes. Since we have only a fragmentary understanding of the forcings, factors, feedbacks, and theoretical underpinnings of the system, we often have to resort to heuristic, multi-discipline methods that may include analogue, signal analysis, correlation, and human intuition.
Take another example, Theo Landscheidt made brilliant predictions of the last few El Niños, years in advance, by his own, very strange, idiosyncratic solar/planetary method. Does that make sense? I don’t know … but it works, and he’s the only one who’se ever been able to do it …
w.
PS – Judith, you said somewhere that this was a test to see if we would “audit” Gray’s work, as he is anti-agw. However, I don’t understand what you are asking. His work is “audited” every year, in that people look at his forecasts and see if they are right. So I’m confused here. What exactly are you suggesting that we audit?
Re # 178
‘The fact is: up to the “radiative” part of the GHG forcing, AGW is pretty solid. Once you get into the water feedback, it gets a little shakier. But when you add the cloud feedback, then we’re on very, very soft ground.’
Following the same line of reasoning: Assuming that the cloud feedback will be negative and will cancel AGW, then we’re on very, very soft ground or don’t you agree on that? At least, I’ve not come across any solid science that supports this. The bottom line: AGW is a serious possibility, wether or not variations in solar forcing or cosmic rays play a significant role as well, and needs serious attention.
Re # 172.
Volcancoes and orbit fluctuations do not seem to play an important role in the current climate variations. Regarding the influence of the sun. Is it possible to construct a credible physical model that attributes a low sensitivity to CO2 (negative feedback by clouds, small influence on the amount of water vapour in the amtmosphere) and at the same time has a high sensitivity to variations in the solar radiance? I have a hard time believing that.
gb,
it would be worth your while to have a look at the work of Henrik Svensmark and colleagues at the Danish National Space Centre.
#187 David Smith thanks for openly posting your audit activity, it’s helpful to see your process. One suggestion: when you cross check data please give the source being used to validate the one in the paper.
From Willis: PS – Judith, you said somewhere that this was a test to see if we would “audit” Gray’s work, as he is anti-agw. However, I don’t understand what you are asking. His work is “audited” every year, in that people look at his forecasts and see if they are right. So I’m confused here. What exactly are you suggesting that we audit?
Willis, the issue is this. Every year, Gray’s forecasts are “wrong” (this does not mean that people don’t want to hear more of them). In my testimony and in an AGU workshop, a diagram and analysis from Greg Holland plotted the forecasts against the actual data, and did a simple statistical analysis. Bottom line is that Gray’s forecasts are worse than just forecast the average for the last 5 or 10 years. RP Jr has challenged us to provide a better forecast, and we could do that by simpling forecasting the average for the last 10 years (boring, but it would be a better forecast). Holland’s analysis is not yet published. RP Jr on Prometheus doesn’t want to hear from Greg Holland regarding Bill Gray’s forecast owing to p***** match factor (and this even made it into the media, accusing Greg Holland of personal bias against these forecasts)
The reason i suggested this to climateaudit is that we could use someone outside our community to evaluate the statistical part of the forecast, as well as the actual forecast (which looks at analogues and then also includes an apparent “guess”).
There are also a variety of other groups making similar statistical forecasts (NOAA and some private companies), but as far as i can tell, they are all typically close to each other.
I am not sure whether this is of sufficient interest to climateaudit, but it would certainly be a useful public service to formally evaluate these forecasts.
Re #194 Certainly. The source for storm data is the Unisys database. It is user-friendly compared to the NHC database.
What I did was to count the number of six-hour periods in which a named storm was of depression, tropical storm, hurricane or cat45 hurricane strength; put those numbers into the appropriate “bucket”; and divide by four. Then, do this same process for all the other named storms in a season. Then, add together all the storms of a season to give the season number.
The “buckets” are:
Storm-day = depression + tropical storm + hurricane strength
Hurricane-day = hurricane strength
Intense hurricane-day = cat3,4,5 hurricane strength
Re verifying or “auditing” Gray’s forecasts. Are any others such as Emanuel doing forecasts? If you are going to “audit” Gray, then all the others should be included for comparison. Maybe someone can find the forecasts from France. Otherwise it is only a witch hunt.
You also have to realize that hurricane forecasting still has many variables such as day to day or extended weather forecasting.
Francois,
If the effects of clouds are unknown, then they _could_ mitigate the warming, or they _could_ excerbate it. This is the nature of uncertainty. It can go both ways.
Kevin,
Spare me the poltically correct canard. Why do you assume uncertainty means nothing is happening?
Chris H.,
Even if I agree with you (and short answer, no, I don’t) uncertainty in this regard, given the consequences, is no excuse for hands in pockets.
Willis thank you for your analysis. A few questions, apologies if i am misinterpreting something. Can you please clarify what your trend forecast is? How is this different from the truncated gaussian average? (is it possible that red and black curves are misidentified in Fig 3?) It is the trend that Holland was comparing to (and your trend “forecast” looks different from taking say the previous 10 year average. Can you pls provide “skill” score for the trend and gaussian truncated forecast also.
Holland also focused on the degradation of the forecast in latter half of the period relative to the early half of the period, do you have any comments on this? (this was the main point that I made in the context of my testimony).
Regarding the “modelers”, i have mentioned this several times that 3 European weather forecasting groups are doing ensemble seasonal forecasts of hurricanes using coupled atm/ocean models, these forecasts are not yet available outside the EU, i hope that they will be available for next season. When i was at ECMWF last month, i specifically asked the Director of Climate Predictions about this. He said the forecasts were available on the web site. I said no they weren’t, and we went to the website. Apparently various forecasts on the website are accessible only to certain communities; since i was outside EU i couldn’t get access. The DIrector was surprised at this, he inferred that these were not publicly availble yet owing to the bureaucracy of trying to coordinate the multi-model ensemble forecasts and how they should be presented across three different agencies. The bottom line is that such model forecasts are a much more complex undertaking requiring much more effort and time than a statistical forecast
The number of NATL seasonal forecasts out there probably exceed a dozen; most are being prepared by the private sector for profit. Gray makes his forecasts publicly available. I suspect that some of the forecasts prepared by the private sector may be performing better than Gray’s (otherwise, why would anyone purchase them, but i could be wrong).
Good morning, Judith! In #195, you say that, “Gray’s forecasts are worse than just forecast the average for the last 5 or 10 years”. I’m not sure that the data supports that. I assume that the actual measure to which you refer is more complicated than that but, for blog purposes, you simplified.
Willis, excellent work, including the clarity of your writing! Even I could follow what you did at 7AM, before my third cup of coffee.
I think that one non-data point that Judith may be making is that “something has changed” in the atmosphere, such that traditional models of the atmosphere do not work as well as they used to. The example chosen was Gray’s model. Indeed, Gray’s 2006 articles mention that correlation with rainfall in Africa no longer work so well in his model, and he has dropped that factor from his model. It apparently worked prior to 1995, but not so well since.
To me, that’s interesting. It would be interesting to know if the value of African rainfall to his model declined slowly or all-at-once in 1995. And, more importantly, why? This is the fun part of all of this for me.
On a slightly firrent subject, a while back it was mentioned that the global climate models may begin to predict seasonal hurricane activity. All I’m familiar with are the big models (GFS, ECMWF) which predict weather perhaps 10 days out. While they are good at spotting general atmospheric patterns, they are usually wrong, wildly wrong at times, about tropical cyclone formation patterns. Perhaps the other GCMs are better. If they are not, then about the best they can do is predict atmospheric patterns (SST, sea level pressures, wind shear, El Nino, etc) and correlate that with historical seasonal data.
re #171: Thanks, Roger! I’ll try to get my hands on the book at some point. However, I was able to find the chapter I was most interested in (PartI.3/Thomas R. Stewart) from the web. This, together with the paper links you provided (and references therein) strengthened my conception that there would be a lot of things this field could (IMHO should!) learn from other fields.
Mainly, from all the papers I’ve seen, it seems to me that climate science has complete missed the remarkable development of the prediction/model selection research that has happened in other fields (information theory, statistics etc) during the last 40 years or so. It seems that the main problem of this research area is yet even to be realized within the climate community! Of course I may be completely wrong as I’m basing my view only on a rather few papers I have read.
When one is doing predictions of any kind, one is (implicitly or explicitly) selecting a probabilistic model for one’s data. Now, having a model and a limited number of observations, the fundamental question is “how do I know my model describes the important features of data (observations) best?”. If you are able to capture the essential features contributing to your observations, then you are also able to make the best predictions for the future. In other words, given a set of prediction models (giving different “fits” to existing data), which one of the models captures the best the essential features of data needed to make best predictions? This is the model selection problem, which IMO climate scientist have completely missed so far.
There are few competing approaches to the problem, but I only link here to my own favorite one 😉 That is, I highly recommend the recent book: Advances in Minimum Description Length — Theory and Applications, MIT press, 2005. The first two chapters are available here, and they should serve as an excellent introduction to the problem field. I recommend reading those (especially chapter one) also for all the people who could not understand a word of my bad explenation above 🙂
David, actually the ECMWF model is showing some pretty amazing skill even with their 10 day forecasts, they are actually picking up the TC genesis. Unfortunately, it is not a big objective of these centers to publish the details of all this in journals (they have their own internal publication series) and getting access to these forecasts outside the EU is costly. So “documentation” of this is slow in coming, but from what i have heard and seen, it is pretty amazing (much better than what NOAA/NWS is producing).
Judy, you sure that this is just a test…and not part of a pissing contest? You sure that you are not trying to make a big deal of Gray’s failed forecasting…to try to use that to discredit his evaluation of the AGW-entranced climatology field? Just asking. 🙂
P.s. WRT climate modelers making yearly predictions: I’m not used to seeing models for long term predictions (AGW studies) evaluated by how they perform where they can be observed. HAve the main ones, been calibrated this way?
TCO, I am really interested (as a scientist) in the evaluation of Gray’s predictions (would appreciate a separate evaluation of his statistical model also), since I am hypothesizing that these forecasts are increasingly biased low owing to the AGW effect on NATL TCs. I have been using this argument (a relatively new one for me, post BAMS article, but it appeared in my testimony) and I would like a better understanding of the the actual skill, bias etc of the forecast and how it is evolving over time. Note, my interest in this subject is different from greg hollands, who is more interested in the utility of the forecasts.
Jean, our field does have its gurus that think deeply about the issues you raise. Chief among them is arguably Lenny Smith (Oxford and the London School of Economics). I refer you to one of his more readable papers (they can get pretty theoretical tho). The best one is actually in one of the internal ECMWF reports that I alluded to earlier.
Title: What might we learn from climate forecasts?
Author(s): Smith LA
Source: PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA 99: 2487-2492 Suppl. 1, FEB 19 2002
Abstract: Most climate models are large dynamical systems involving a million (or more) variables on big computers. Given that they are nonlinear and not perfect, what can we expect to learn from them about the earth’s climate? How can we determine which aspects of their output might be useful and which are noise? And how should we distribute resources between making them “better,” estimating variables of true social and economic interest, and quantifying how good they are at the moment? Just as “chaos” prevents accurate weather forecasts, so model error precludes accurate forecasts of the distributions that define climate, yielding uncertainty of the second kind. Can we estimate the uncertainty in our uncertainty estimates? These questions are discussed. Ultimately, all uncertainty is quantified within a given modeling paradigm; our forecasts need never reflect the uncertainty in a physical system.
Judy-
I am once again going to respectfully ask you to stop writing things like the following:
“RP Jr on Prometheus doesn’t want to hear from Greg Holland regarding Bill Gray’s forecast owing to p***** match factor.”
Greg Holland has an open invitation to participate in our disucssions — both you and he know this. The only pissing match that I can see in the one that you repeatedly seem to be trying to start with me via inflammatory accusations and mischaracterizations. Please, once again, just stop, OK?
Judith:
I can answer all of those questions in simple mathematical form:
Re #202 I hope they succeed and soon. That would be a big step forward to the benefit of everyone.
I took a look at what the ECMWF forecasted for my part of the world ( attached , I hope). This is the ECMWF forecast from 10 days ago, for today. It shows dry, pleasant weather here with no evidence of tropical activity in the Gulf of Mexico.
The reality is that it is raining heavily here, connected to a developing semi-tropical low pressure area over the western Gulf of Mexico. Pressure are 1005mb in the Western Gulf, which is low. I wonder if this will become the tenth named system of 2006.
Now this is one data point, and means nothing, but it does illustrate why I stay a bit skeptical of the current models to predict events 10 days away, much less three or four months away. Anyway, we’re getting rain, which is a good thing!
Roger, I would be delighted to declare a truce. My conditions are that you stop ascribing to scientific researchers motives that you know nothing about. No, you do not engage in “ad hominem” attacks, but you do attack the motives of scientists (which IMO is much worse). When you cease doing this to climate researchers (particularly to me and my colleagues), I will cease “jerking your train”.
Some of my favorites from Prometheus:
One of the lead authors of the AGU assessment has been in a public feud with Bill Gray and is a strong advocate of a human role in recent hurricane activity. It is not unreasonable to think that the AGU assessment was being used as a vehicle to advance this battle under the guise of community “consensus.”
I hypothesized that the issue of seasonal hurricane forecasts had been caught up in the “hurricane-climate wars” between Bill Gray and Greg Holland. Holland and Peter Webster (who were both involved with preparing the AGU report) took serious issue with my even raising this hypothesis (how dare I!!), flatly denying any such relationship between the AGU report’s criticism of Gray’s seasonal forecasts and the global warming debate. However, In today’s hearing (that I participated in) Judy Curry’s testimony completely vindicates my raising this issue (Curry is a collaborator with Holland and Webster). Here is the relevant excerpt from her testimony (note if you’ve done your homework, you’ve all read this testimony)
Accusing me of an “ad hominem attack” for being reported as saying
“Well, he’s a prolific writer, but he is really a policy person [who] has a qualitative understanding of climate science. He is not a climatologist.”
And this is not to mention numerous attempts to smoke out stealth motives that I am “really” motivated by trying to limit CO2 emissions or something
re #205: I quickly scanned through his publications from here. Smith seems to have focused mainly on physical, dynamical models (like GCMs or weather models) evaluation, which is somewhat different on the issues I’m talking about. Still, although admittably he seems to be a clever guy, there is something striking about his publications (the same thing I’m seeing all over the climate field): almost no references outside the climate field! Again, although I only have a superficial understanding of those things, there are other fields researching very closely related things. Econometrics and mathematical biology comes immediately to my mind. If you read articles from those fields, they DO have a plenty of cross-referencies.
Jean, lenny is a nonlinear dynamicist by training, he works on a bunch of different things (including brain waves in epileptics or something like that) and also economics. He is a superb statistician, and his main contribution to the climate modelling field (he doesn’t build climate models) is the combination of model output and statistics to obtain a forecast.
189, Willis:
I find it very interesting that these pesky solar cycles keep showing up everywhere, especially that well-known 11 year cycle. Hmmm…Wish Steve would put up a solar thrread.
re #211: I believe that. And I also believe he’s somewhat following what’s going on, e.g., in those fields I listed. But still, the (low) number of references outside his own publications (and climate field) is striking. And if he is one of few contributors in that area, the situation is IMO almost alarming. I’m sure co-operation between climate model builders, and, e.g., biological model builders would be extremely beneficial (is there any?).
Judy-
Please understand that I am a political scientist. A discussion of the incentives, values, and indeed motivations of living, breathing, emotional human beings is a central part of my discipline. You may not like it, but political scientists discuss and write about these things, particularly about important, influential people involved in the policy process. Scientists, for better or worse, are such people. These discussions are not “attacks.”
For someone with self-characterized “think skin” it seems odd that you would object so strenuously to my simply raising the hypothesis that Holland/Webster/Curry have found a new interest in Bill Gray’s seasonal forecasts as part of their larger dispute over global warming and hurricanes. Seems a pretty obvious connection to make, and indeed has been raised a few times here at CA. At a minimum, for those of us trying to interpret what is coming out of the science community, it is important to understand the social dynamics. HWC are not a disinterested party in the Gray forecasts (e.g., why haven’t you focused on Saunders?). Sorry, but it is true. Is this an “attack”? No, absolutely not.
You must agree with the importance of understanding such human dynamics, as you have frequently charcterized the possible motives of others involved in hurricane science, e.g., such as your recently expressed concern about the NHC leading a hurricane reanalysis — which you are absolutely correct about — it also involves a real conflict of interest! Is this an “attack”? I did not interpret it as such, just an open recognition of the real-world incentives and motivations that exist in the community.
You may not like being at the center of an important science-policy-politics issue, but given that you are there you can expect the attention of people interested in scientists in policy and politics. Responding to this attention by disparaging my work through mischaracterization or trying to “jerk my train”;-) will certainly provide more data for our work, but is unlikley to reduce tensions!
I will continue to discuss the factors important to understanding scientists in policy and politics, and given my long-standing interest in hurricanes and climate change, I will continue to focus on that area. For my part I will do my best to keep the converstations professional. In return I would hope that you might gain and express some appreciation for the fact that a serious discussion of the motives, interests, and values of scientists in politicized settings is not the same thing as a personal attack.
As a peace offering I promise to send you a copy of my forthcoming book on science in policy and politics, in which you might be glad to hear that the focus of attention is not at all on climate science or hurricanes;-)
#198 Boris,
I do not want my children growing-up in the “atmosphere” you create on top of everything else they have to face in this life either. The worst thing you can do to any human being or child is to have low expectations of them.
On top of that you are still arguing over a senerio of “warmth” from a computer, displayed on graphs in fractions less than 1 C with data that stinks!-huge time scales, with huge error margins representing a planet we don’t fully understand with a climate that has never been stable- coming out of an age of ice and an during heavily polictical era ! Your tone is either scared, worried or guilt ridden. What gives? If you could tax the hell out of me some more would you be happy? Get it over with then! Or tell us something useful!! Sheesh!
And attention all scientists-putting data into a computer and creating a “model” of the world and sharing it with the media? Great idea! Gee thanks guys!! LOL 😉
rocks,
Notice how I do not ascribe motives to you. This is done because I do not know you. So your commments on taxes are unfounded and nonsensical.
You give me absolutely no reason to reject the NAS, Sceince, Nature, the AGU, etc etc etc etc etc etc etc etc etc etc etc etc etc etc etc etc etc.
All you give me is vague vagueness. Heck, you’re even vague about the “atmosphere” you don’t want you’re children growing up in.
You’re the one bringing politics into this, taxes and such.
Boris,
Holy cow, I think the data is crappy from my own investigation and observation, and I am not alone.
the “atmosphere” I do not like is the one of fear, and guilt or negative expectations for the future from data that stinks.
etc etc etc etc. I do not have to give you a thing.
That’s about as blunt and to the point as I can be dude.
I am becoming more confused with this thread and what it is that we are attempting to accomplish. I have not followed all the details of the post inputs but I do know that there is no way in heck you can compare a prediction of someone like Gray who predicts than is evaluted out-of-sample and evidently every year. A prediction scheme delivered currently and evaluated with historical data is subject to overfitting and data snooping and in no way can be evaluted as Gray is out-of-sample.
What one needs for fair comparison purposes is alternate predictions with out-of-sample results — pure and simple. I believe that Steve M had auditors’ suspicions of some modeller (in Europe??) who others had claimed had predicted the generally unexpected hurricane activity in 2005 and had a prediction published for the 2006 season prior to its start. I know that there was a long delay in getting acknowledgment and I would ask Steve M now if he were able to confirm that the predictions for 2005 were made and published prior to that season and if he was able to obtain the same information for the 2006 season.
Where is the out-of-sample results for alternative/competitive predicting schemes?
>> ascribing to scientific researchers motives that you know nothing about. — JC
>>> I am a political scientist — RJPJr
Wait. This is two very different things.
A “political science” researcher would have the same dispassionate and evenhanded approach to his subject field as any other scientist has to hers, right? And would study motivation academically and publish studies in the relevant journals.
A “political” scientist is someone claiming to do empirical, verifiable science, but spinning the results due to his personal politics.
You’ve _got_ to get the quotation marks right.
#218. Ken, I have never received any documentarion to show that the European predictions for the 2005 season were made in the spring of 2005. My efforts at obtaining corresponding 2006 predictions were unsuccssful.
I recall a nice comment from Paul Samuelson when congratulated on predicting a recession – he archly observed that he had predicted 5 out of the last 2 recessions.
You know, even if the predictions had been made, you would need to worry about selectively dramatizing the one of several predicitons that happens to be right. Need to have all the drama and such BEFORE the test.
TCO, you make an important point (copied below) that I should have included in my previous post (comment #218). And of course when all is said and done, even a good record has to be evaluated on the basis of the odds that it could have performed that well by chance.
#221. Yeah, one wonders about the Vegas tout phenomenon.
ok boris,
take your hands out of your pocket and tell us what you want to do, what it would cost, and what measurable effect (good and bad) it would have on the world
#217:
Fair enough, but pardon me if I don’t fall in line behind your observations and theories (especially the self-contradicting ones). I am a good deal less alone.
#224
charles, Glad to hear you’re interested in mitigation. Lots of places out there to find info on that. Google away.
Back to the seasonal hurricane forecasts. Willis did an excellent analysis, but still didn’t quite provide me what i need to answer the two questions:
1) Are the statistical forecasts biased low for the past decade? can we attribute this low bias to AGW?
2) Are the seasonal forecasts “useful”.
I personally began thinking about these issues at the National Hurricane Conference last April, as a result of Bill Grays talk and a conversation I had with Jim Elsner. Jim Elsner also makes seasonal hurricane forecasts (5 and 30 years in advance!), and his perception was that the statistical forecast schemes were no longer working (p.s. Elsner forecast 9.6 +/- 2.4 named storms for 2006, not bad), it seemed like “something else” was going on. The second thing that surprised me was in the Gray/Klotzbach presentation they counted their 2005 forecast as a “skilful” forecast (they forecast 15 in June, compared with 27 observed) since the observations and forecasts were above the 50 year mean of 9.6. This immediately called to mind an article in the popular media that i read at the end of 2005 that discussed the worst forecasts for 2005, and gray’s hurricane forecast was #1.
Greg Holland, who is working in the trenches with engineers and disaster managers in New Orleans, is very concerned about what is actually a “useful” forecast. His thinking (I hope I have this straight) is that a forecast scheme should be able to perform credibly against the 5 year average, and should also have a deviation of the correct sign for those years where the number is a standard deviation above/below the mean (i.e. at least get the sign of the deviation correct in years where the observed number is very high or very low).
Re point #1, a purely statistical scheme like Elsners would be a better way to explore this, but Gray is the only person actually making forecasts for several decades. I am also thinking that Gray’s cumulative wisdom about hurricanes is tied up in his seasonal forecasts, and untangling what is going on in his forecasts can help us understand what is going on in nature (this may be a hopeless endeavor). If his statistical scheme continues to be biased low and and analogues are also low, is the deviation of his actual forecasts from the statistical or analogue scheme is increasing with time, what does that mean? Maybe these are some suggestions for Phil Klotzbach’s thesis 🙂
Roger, my skin is not at all thin where your comments are concerned. I simply find them unprofessional. And your comments on prometheus regarding Greg Hollands “motives” were actually picked up by the media, further inflaming the public perception of the “hurricane wars” when the principals were no longer at war. We were all pretty apalled by this. You are the one with the apparent thin skin with regards to any criticisms of your papers or your blog. The motives of groups of people should be of interest to a political scientist, but publicly attacking a piece of research owing to your perception of the scientists motive is unprofessional behavior, what is expected of yellow journalists not scientists. As you continue to play “jiminy cricket” regarding my public statements, I will continue to call you out when you inappropriately attack the motives of a scientist.
#225 Hi Boris, Now I have to give pardon? LOL
Your need here is far more interesting then mine. You could explain instead of brushing off questions like you did in #224.
Like # 218 on topic I am confused here. We tried to look into the Gray/Judy beef and she denies the quotes to the media, and we don’t see what’s so bad about Gray or his methods yet or what exactly was said in this presentation to critique.
Yep TCO- all about the drama! 😉
RP Jr comments are interesting and I thank him for coming in at that angle, I think it matters. Sounds like an interesting book.
Apologies to SteveM for this low level unproductive side discussion.
Cheers!
Roger, do you recall the political science book “Prometheus Bound” by Zieman, of a decade or more ago? I wonder if this paper on his thesis describes something akin to your idea of your position as a political scientist?
Discussion here:
http://www.fu-berlin.de/ffu/akumwelt/bc2002/files/boehmer_christiansen.pdf.
HTML here: http://scholar.google.com/scholar?hl=en&lr=&q=cache:V3gzA_rjIj4J:www.fu-berlin.de/ffu/akumwelt/bc2002/files/boehmer_christiansen.pdf
Climate Audit – Bill Gray’s Presentation
This thread started out as a crit of Bill Gray’s presentation. It seems to have been sidetracked into a review of cyclone prediction methods among other things.
To get back on track, the thrust of Bill Gray’s paper was an appraisal of AGW in general and GCM support of AGW in particular. Not only does he cast serious doubt on the effectiveness of the GCMs, he puts forward some interesting ideas about why they may be going wrong. I found his arguments compelling. Shouldn’t they form the substance of this discussion?
Judith Curry – if you are so keen to develop the critical abilities of your students perhaps you should turn them loose on IPCC TAR Chapter 8. This too is an appraisal of GCMs. IMHO it is a travesty of statistical reasoning. Given the data they present the only conclusion a reasonable person could draw is that none of the models actually work at all. Sure they nearly work but that is like nearly remembering a ten digit telephone number – if you don’t get all the components correct the whole thing is worthless.
See my posts #67 and #74 on the Truth Machines thread
I am getting some ideas here for what to do in classes at Georgia Tech. Next fall for the climate seminar, we have decided to do the “hockey stick”, everything from proxy data, to statistical analysis techniques, assessment of the assessments (MM, North, Wegman), impact on the overall case for global warming, dueling blogs, etc. And of course an update on the latest scientific results. Kim Cobb and Peter Webster will coteach the course. I am teaching the last week of Peter Webster’s climate and global change course, where we talk about IPCC and policy. Ch 8 will certainly be part of what we talk about.
In looking at Gray’s statistical model predictions, it’s important to note that it’s not one model, but rather a series of models. It looks like the parameters measured, and the weight given to each, changed from year to year in search of better skill.
I’m going to take a look at Gray’s models’ prediction of intense hurricanes later tonight.
As a side note, I glanced at the opening part of Gray’s reports, where it lists team members. Here is 1999’s team:
Gray
Landsea
Mielke – Professor of Statistics
Berry – Professor of Statistics
Half the team was statisticians!
(They disappeared from the credits in 2002. Perhaps that was the downfall…)
Cobb is cute.
Judy- So be it. At least you have openly admitted that your actions here are baed on your intentionally trying to “yank my chain” and protect the honor of your colleague who was caught abusing an assessment process (for whatever reason). Now that you’ve admitted as much, I am happy to let your continued name calling and deliberate mischaracterizations of my work just speak for themselves! I won’t further engage you here, but I will continue to watch you play these interesting games with the CA folks.
Hank- Not familiar with that work, sorry!
All- There is no single “Gray methodology” — it has changed, as often as yearly over the time period that he has been doing his forecasts. No one to my knowledge has evaluated how each approach would have done if “frozen” in time. Looking for a evidence of AGW in seasonal hurricane forecasts seems like quite a stretch to me. The best place to look is in hurricane data itself, not a _very_ short prediction record based on many factors which changes up to yearly. You will have essentially no statistical power to demonstrate anything. Holland’s effort to suggest an AGW-related underforecast conclusion based on 7 years of forecasts is nonsense, from a statistical basis alone (as I think someone conclusively showed above).
Compare this paper about AGW signal detection in short data records (but much longer than Gray’s track record!):
http://www.agu.org/pubs/crossref/2006/2006GL027552.shtml
Evaulation of forecasts is quite important from the standpoint of their usefulness and value, and the same goes for Gray’s work. But in this case, trying to evaluate Gray’s track record with respect to saying this or that about AGW, as Curry and Holland have suggested, is a scientific wild goose chase. But it does have some PR value I suppose.
September 2006 Press Clips: “The words ‘global warming’ provoke a sharp retort from Colorado State University meteorology professor emeritus William Gray: ‘It’s a big scam.’ And the name of climate researcher Kevin Trenberth elicits a sputtered ‘opportunist.’ At the National Center for Atmospheric Research, where Trenberth works, Gray’s name prompts dismay. ‘Bill Gray is completely unreasonable,’ Trenberth says. ‘He has a mind block on this.’ ”
Link: http://www.ucar.edu/news/pressclips/archive/06/09.jsp
That points to a newspaper article, and doesn’t comment on the quotes. Maybe newspaper writers (or editors) are playing the old game of “Let’s you and him fight” by stirring up such comments, or inventing or misattributing them?
#231 Judith,
Kim Cobb and Peter Webster will coteach the course.
Why not bring in someone from the Georgia Tech Statistics Department? Looking at the roster, Brani Vidakovic seems to be a possible candidate. Do you know him?
Fly in Steve for a guest lecture. It would be a hoot.
Roger, you really should read Zieman, and the article linked above that extends Zieman’s thesis. You could almost be “channeling” them!
—- excerpt—-
“Answerism would assert that research for the sustainability transition should enable “policy-makers’ – who precisely and how? – to manage the global environment, which is surely little more than a statistical abstraction and perhaps a legal concept that has been politically defined. With reference to my knowledge of “global warming’ science and policy, I will argue that sustainability science is in danger of following a “bad’ or failed model, one that it both too simplistic in its science and too complex in its human implications to be implementable. The IPCC model does not give “relevant’ answers to the real global problems but rather tends to support the search for funding by an expanding group of natural sciences. In this search environmental alarmism and appeals to government, play a major role …”
He and Zieman to me read as the kind of political scientists who — than study dispassionately how politics works — see their task as freeing Prometheus. You may have rediscovered or invented the idea independently.
I looked over the college and they don’t have a formal statistics department. The issue of formal and theoretical statistics is one that goes back to Hoteling (anyone interested should read what he wrote in the 40s, as it is subtext to the Wegman remarks and even to his critics who see him as pushing for his field).
228 glad I made you LOL. Suppose you’ve shown me a thing or two about blowing off questions.
Of course, charles isn’t serious, it’s just another delay tactic, argument by exhaustion. If one argument is breaking down, retreat and set up defenses somewhere else. Unless, charles, I’ve read you wrong and you accept the science?
Boris, Boris, PLEASE read Bjorn Lomborg’s book, “The Skeptical Environmentalist,” specifically the chapter on Global Warming. Lomborg goes along with the AGW concept and shows how “taking your hands out of your pockets” is stupid. After you have read that, please tell us what is wrong with his logic!
Interesting sylabus here: http://shadow.eas.gatech.edu/~kcobb/seminar/multidecadal.htm
jae: stop making us look bad. I order you.
Roger, let me give you a counter scenario that will hopefully help you understand how damaging and inappropriate your behaviour is in this regard. I hereby “accuse” you of having stealth motives in being a supporter of the Gray/Landsea camp, and therefore defending them (Landsea, a student of Gray’s, is a collaborator of yours) by trying to discredit research that questions Gray’s forecasts by attacking the motives of the scientist that did the research. You are further motivated by trying to build readership over at Prometheus and bring media attention to you personally by making sensationalist accusations.
Did I get it wrong (who knows, maybe i didn’t)? This is what it looks like to a lot of people. Did you not like it? You are far too close to this situation, you want to be included in the hurricane and climate communities as a participant, you collaborate with many of the players, but you also expect credibility when you attack motives of scientists on the other side of a debate as your collaborators. Chris Mooney, a serious science journalist that is writing a book on hurricanes and global warming (not just the science, but the people), arguably has the requisite objectivity for such a task. You arguably do not, and you definitely do not have the appearance of objectivity in this owing to your collaborations with scientists on one side of the debate.
I am actively participating in climateaudit which is not particularly friendly to many of my views. But no one here has attacked my motives, other than generic attacks on the motives or warmers, and those have mostly stopped. Only you are attacking my motives. This is a complex situation that we find ourselves in. I am doing considerable reflection on what constitutes integrity in science and in the personal behaviour of scientists. I hope that you are doing the same. We may not come to the same conclusions. As long as you continue attacking the motives of me and my colleagues, i will continue to call you on it. Such behavior does not enhance the credibility of the useful contributions that you do make.
240 Your questions to me and the things you want me to say are contained in the contents of this blog examining the data Boris, why don’t you read it? And everything else you want me explain can be found first off by studying the earth’s history. If you look at any of the subjects here where I or my husband made a statement or opinion about the science ask me in that topic I will happily debate. Leave this alone for now would ya? Otherwise you are a troll.
231 I am glad you are thinking about that kind of class for the students. My husband had a climate course during his masters education that looked at the very first UN/ IPCC report. Do I have to say the prof and class were not impressed and neither was he so they didn’t spend much time on it? That was before blogs, the polictics/ media frenzy and CA. He’s been watching it turn into what it has become. If you add a geolgist to your list of teachers/speakers even better.
Cheers!
sorry- geologist! I keep dropping my o! 🙂
243: ?? Please be more specific, TCO. I’ve got Boris pegged as a precocious 15-year-old. He reminds me of you.
#231. Judith, here’s a post on Kim Cobb’s corals suggesting that the medieval dO18 that she observed could be plausibly explained by a more northerly medieval ITCZ as opposed to a cool medieval Pacific. Any thoughts?
I think that one could do an entire statistics seminar on the hockey stick -The Little Shop of Horrors – it’s a nice practical example of how to do almost everything possible wrong. Once you understand the issues, your students will have a useful sense of practical statistical issues. It’s much livelier than dry examples.
LOL. That ain’t gonna happen at Georgia Tech.!! Judith would be tarred and feathered and run out of town.
#238, Hank, Thanks for the reference, I’ll have a look!
240 boris,
My view is future increasing co2 and warming (driven by co2 or not) is most likely to be moderate and a net plus for the earth (like the last 100 yrs).
If the warming is greater than moderate and driven by co2 I still can’t think of any co2 related action where the net benefits would be worth the costs. Can you?
I didn’t think so.
Thus my hands remain in my pockets.
Hey, TCO, do you have a job? Just curious.
Here are some numbers on Gray’s forecasts of intense hurricanes. Intense hurricanes (cat3,4,5) are the probleme du mois and they are the hardest to forecast (my opinion), so let’s see how Gray’s statistical models performed.
(Note: I rounded any fractions in Gray’s June statistical forecast of storm count to the nearest whole number. If Gray gave several model results, without stating a preference for any particular one, I averaged the group.)
1999: 3 intense storms forecast by Gray’s models, 5 intense storms actual
2000: 2 forecast, 3 actual
2001: 2 forecast, 4 actual
2002: 2 forecast, 2 actual
2003: 3 forecast, 3 actual
2004: 3 forecast, 6 actual
2005: 2 forecast, 7 actual
2006: 3 forecast, 2 actual
Over the eight years, Gray’s models forecast 20 intense hurricanes. The reality was 32 intense hurricanes. Climatology (1950-2000) indicates that 18 storms would be expected.
To my (non-statistical) eyes, Gray’s models appear to have no particular skill at predicting intense hurricane count during 1999-2006. The models appear to have stayed close to the climatological rate of 2.25 intense storms per season.
The key question is, how good were the model(s) at forecasting pre-1999 intense hurricanes? If they used to have skill but no longer have it, then perhaps something has changed in the way that weather works. However, if the intense-storm forecasts have always been poor, then not much can be said. I haven’t yet found pre-1999 data but will look.
Gray’s skill at forecasting total storm count are a different story, as Willis showed (above).
Re: #113
My goodness! Go away for a couple of days and find you’re ovr 100 posts behind!
Dr Curry, thank you for that explanation and the good grace with which you provided it. I don’t think I could be much more cynical of journalists than I am, already. Your experience doesn’t surprise me in the least.
Judith, thank you for your comment. You say:
Judith, as I pointed out above, Holland’s analysis is both mathematically incorrect and uses the wrong data. If you are depending on him for your analyses, you should stop doing so immediately, he’s steering you wrong.
For example, your claim above that “Bottom line is that Gray’s forecasts are worse than just forecast the average for the last 5 or 10 years.” is absolute nonsense. Here are the results:
Length___________________________ Gray____ 10 Yr. Trailing Avg.__5 Year Trailing Avg.__1950-previous yr Avg.
Last 22 years Average Error____________0.9_______________1.1_______________1.5_______________2.3
Last 10 years Average Error____________1.0_______________1.1_______________2.0_______________4.0
Last 5 years Average Error_____________1.4_______________1.6_______________2.6_______________5.6
So if Holland told you that, he’s wrong. The trailing average that performs the best is about a 20 trailing average, but it doesn’t beat Gray either. You need to seriously rethink your position on Gray’s forecasts. You cannot beat his forecast by “simply forecasting the average for the last 10 years (boring, but it would be a better forecast)”, that’s a fantasy. I know. I tried all of the schemes shown just above, along with many others, before I could beat his forecast.
Also, note that while Gray’s average errors have increased with the shorter term record, so have the trailing average “forecast” errors.
Finally, can we read anything into the increasing error in the more recent forecasts? Absolutely not, for the reasons I detailed above. Because the records are so short, there is no statistical difference between the 22 year, 10 year, and 5 year forecasts for either Gray or any of the trailing average “forecasts”.
I hate to speak strongly like this, but you and Holland are making claims that are simply not true.
w.
A video is worth a lot of pictures, to see the beautiful complexity of hurricanes try this GSFC video of all 27 named 2005 Atlantic storms – there’s an accompanying “explanation” and music. Enjoy!
Errata – “The trailing average that performs the best is about a 20 trailing average” should be “The trailing average that performs the best is about a 20 year trailing average”.
w.
Thanks, Willis. Just a quick question (I’m too lazy to go through all these h. threads), is there any published work which describes Gray’s method or is it something “secret”?
Jean, background on Gray’s work can be found here.
The 1992 paper ‘Predicting Atlantic Seasonal Huricane Activity 6-11 Months in Advance’ discusses the methods used then, including the use of statistical methodology. I have not looked at all the papers, this was a random pick that seemed to have a good discription of the methodology used during a ‘skillfull’ period.
We are glad you finally said something willis.
It’s also not crazy to suppose a person studying such things for 50 years could develope a certain “6th sense” about such things. The brain is a computer after all. And it happens all the time: it’s even natural. That much experience should be at least respected [from the tone of some comments] a bit more if you consider that, even if it’s not scientific. (We see the youngins’ butting heads with the oldest in the herd more then a pissing contest here)
re #259: Thanks Cliff! So it’s based on simple LAD regression … anyone in the climate field ever heard of Information Theory?!??
Willis-
Re: 255
http://www.climateaudit.org/?p=857#comment-52342
This is a good analysis and reinforces that which has been presented on the TS list (on which Holland and Curry are members) a while ago by Mark Saunders, Chris Landsea, and Phil Klotzbach and available to anyone at the links I provided earlier.
By conventional atmospheric science metrics of “skill”, Gray’s forecasts have demonstrated skill. One could come up with a different metric of skill (a la what landsea/Knaff did on ENSO), but this turns the debate into about mainly about skill metrics, not Gray’s forecasts.
Some time ago we painted a not-too-optimistic picture about the utility of such forecasts. Interestingly, they have value, but are not terribly useful.
Pielke Jr., R. A., and R. A. Pielke Sr., 1997: Vulnerability to Hurricanes Along the U.S. Atlantic and Gulf Coasts: Considerations on the Use of Long-Term Forecasts, Ch. 8 in H. Diaz, and R. Pulwarty (eds.), Hurricanes: Climate and Socioeconomic Impacts, Springer-Verlag: Berlin, pp. 147-184.
Click to access 1997.10.pdf
Re #259 Jean, the Hoyos et al. paper (referred to many times on this site) used information theory.
Steve: Hoyos paper is online here http://webster.eas.gatech.edu/Papers/Webster2006c.pdf
Willis, the only claim I am personally making about Gray’s forecast is that the bias in the period 1995-2005 is greater than the bias in the period 1984-1994. Did you check this? If my statement is incorrect, then I will stop using this argument.
The thing about Gray having a 6th sense about hurricanes is exactly correct. Any scientist working for substantial period of time on the same subject will develop such a sense. Climate modelers have a 6th sense about what they are doing, paleoclimatologists have a 6th sense about what they are doing. Hence their insights within the scope of their expertise should be respected (there is a tendency however to claim 6th sense authority beyond the range of their expertise). This is not all reducible to statistics (which does not imply that the statistics do not need to be done currectly)
#248 Steve, Re Kim Cobb’s study. The whole issue of the ITCZ is very interesting and receiving increased attention. However, I don’t your argument can explain away the absence of MWP if it was a global phenomena and not just NATL/NAMER/WEUROP.
BTW, Kim Cobb previously pointed out the climateaudit thread on her paper. She chose not to get involved in the discussions owing to the sexist nature of the post and some of the comments. If this sort of thing can be minimized, there is a better chance of attracting more climate researchers to the site.
Re #264: In other words, hurricane “prediction” and climate modeling are as much art as science. I thought the whole point of science was to get beyond such intuitive approaches! [Smile]
Re: #244
Dr. Curry, I believe that I am sufficiently advanced in years that I can dispense with the usual deferential approach to members of academia and tell you that you are wasting bandwidth with your discussions with RP Jr. You say that you are participating here to learn something of what CA has to contribute to the AGW debate and the people who post at this site. In turn you must be aware that posters and readers here are scrutinizing what you have to say on various subjects and how you choose to join the discussions.
Can we please stay on the subject of this thread which was the skill of Dr. Gray’s TS predictive powers and the implications that skill has on his views of AGW? I think what is clear is that Dr. Gray like a lot of investment strategists probably adjusts his strategy each year (think Value Line) or at least it is never fixed for any length of time. His particular annual strategy can only be evaluated in that case from the fixed period of time or over the full time as “the evolving Gray strategy”.
In the bigger picture would such predictions (as I understand them) have much practical value? I would think not and for the very reason that I put forward previously for the work of looking at historical data and looking for increasing trends in frequencies and intensities. You can look at the numbers and intensities of the many sea going storms and not know much about land falling storms and particularly where they may come onto land.
The best way to view the Gray’s predictions would be whether they impart better understanding of the weather patterns that could potentially generate TSs and determine their intensities and after all that perhaps comparing them to computer modeling variables and parameters. Otherwise, the whole exercise here becomes a statistical numbers game that is played without the benefit and limitation of a specific a priori hypothesis or goals.
Willis, thanks again for your analysis, this is very helpful. This analysis convinces me that participating in climateaudit is useful (although sometimes it seems like we need to get through alot of noise to get to the useful part).
Re #265
Dr Curry, I am unfamiliar with current trends in academic communications and was wondering: if one wishes to insult an audience and/or host, is using a third party quote the current acceptable method or is it just your personal style?
RE: #120 – If suc a thread gets started, here would be one of my lines of enquiry. Namely, Gray’s overuse of the THC paradigm. That he has not developed an analytically rigorous model of how it factors in is what suggests this.
re #263: Judith! Well, that’s not exactly part of Information Theory I had in my mind … but … after reading the Hoyos paper… what can I say except:
a) did it every occur to you that there is a huge literature about mutual information related things? I don’t want to discuss about the technical issues in that paper more instead I say
b) PLEASE, use the option in #139 if you are interested in another (rather quick) publication in Science… 😉
This thread seems to be going nowhere. Personally, I don’t see the point in “auditing” Bill Gray’s forecasts. Dr. Curry proposes to do this to see if Climateaudit is also interested in auditing the work of “anti-AGW”.
But Gray’s forecasts are not peer-reviewed papers. As such he can publish whatever he wants, with whatever method suits him (including voodoo!). It’s really up to the people using his forecasts to decide whether they are of value or not. If the aim of this exercise is just to discredit whatever Bill Gray has to say on global warming on the basis that his forecasts have no skill, then it’s not up to the standards that Steve M. seems to have set for Climateaudit.
Jean, thanks much for pointing me to #139, I will be in touch. The statistical methods used are very outdated, the real skill seems to be the analogues and gray’s 6th sense.
Time for a summary. After the initial negative reactions of my students to having a thread on the topic of Gray’s global warming ideas, I think this thread turned out to be interesting and productive. Here is what I think the takehome message should be (my spin of course)
1) Much of this thread focused on the uniqueness of Gray in terms of the extremity of his position on AGW as well as his unique 6th sense expertise in hurricanes. I respect and value his 6th sense with respect to hurricanes; however in the absence of published research on this topic, relying on his 6th sense does not translate well into understanding multidecadal climate dynamics and global warming
2) The thread got sidetracked into the “hurricane wars”. While I was starting to agree with Ken that I was using up too much bandwidth on this, I have now changed my mind. I have received several off-blog emails from “lurkers” (not sure if this is the correct blogospheric term) who greatly appreciated the insights into the hurricane wars (and esp the brain fossilization dissertation) and have come to an improved understanding of how to sort things out in terms of what to think about this subject and how ad hominem and motive attacks can really derail progress.
3) a combinatin of #1 and #2 led to a focus on Gray’s seasonal forecasts, which IMO is the more valuable contribution of Grays than his AGW opinions. I am personally trying to figure out the 6th sense aspect of Gray’s forecasts (this thread helped my articulate that) and how this differs from the statistical modelling, and what all of this might mean in the context of the elevated hurricane activity in the past decade. Owing to the excellent work of Willis, we have identified that Gray’s forecasts are skilfull relative to one of the metrics being used by Greg Holland (contrary to what Greg Holland has been reporting). The argument that I have been starting to develop re statistical forecast models performing more poorly the past decade is not supported by using Holland’s analysis of Gray’s actual forecasts (but there wasn’t much influence of the statistical model on these forecasts anyways). Digging into the statistical part of his forecast might provide further insights into this.
4) Statistical forecasts of hurricanes arguably aren’t terribly useful (if not as lacking in “skill” as inferred from Holland’s analysis). There is a need for better forecasts. Maybe Jean has some good ideas on how to do better statistical forecasts. I think the future of seasonal hurricane forecasts lies in application of the ensemble seasonal forecasts from the European models, combined with statistical methods.
Judith, as pointed out in #259 above:
Bill Gray’s original 1984 “statistical” model is published here.
Click to access gray1984-2.pdf
Bill Gray’s updated 1994 model is published here.
Click to access grayetal1994.pdf
Thanks Jeff. I am familiar with Gray/Klotzbach statistical model. The part i am trying to figure out is why the actual forecast (of 17 as of june 06) is so different from the statistical (10) and analogue (13) forecasts. where did 17 come from? this is the 6th sense part i would like some insight into. this doesn’t seem to be documented in the otherwise complete documentation that is provided on Gray’s site.
Re #276: I faintly recall having seen a remark from Gray (not sure where) to the effect that a “sixth sense”-type adjustment is made based on what actually happened in the previous year or two. It does sound as if something along those lines has to be the answer, but maybe somebody could just email Gray or Klotzbach and ask.
Re #272
How many scientist does it take to forecast hurricanes anyway? 🙂
Do the new hurricane experts really want to learn to make better forecasts or is it they want to find the A in AGW? Otherwise why wouldn’t they be speaking, talking and sharing with a 50 yr vet about forecasting hurricanes no matter how he does it?
Willis E, I have some serious misgivings about the Gray hurricane prediction models and how they should be properly evaluated. The schemes for his predictive models published in the early 1980s and again in the early 1990s are similar to investment schemes that I have seen developed for mechanical type investing. If he can explain all the physics behind his models than I would have no qualms about the overfitting that one can do to match a model to historical data.
His models have the advantage of being adjustable each year based, I am guessing, on the trends and the presence of auto correlations. To truly evaluate what he has predicted would not one have to factor out those properties? He cannot be evaluted on the basis of a strategy fixed at the initial starting point for out-of-sample results.
Often after a few years of out-of-sample results investment strategists go back and fine tune their models based on intervening data and more overfitting. Gray evidently made some significant changes to his model(s) in the early 1990s.
Another factor that bothers me is that his predictions become more useful when they can predict the peaks (particularly, so the peaks) and valleys in the occurrences of annual storm activites. His results show me that outside of following rather conservatively a trend and the last year’s activity his predictions miss the sudden increases and decreases in activity. In my mind the ability of the model to predict extreme years is not only the most practical application of these predictions but also a true test of their predictive skill.
#274
I frankly do not see the logic behind that statement. So someone who has developed an intuitive/empirical understanding of hurricanes, and who does not publish in peer reviewed journals, by definition, cannot understand the physics of global warming?
So who will be the next victim? Hey, let’s audit MM03 ! Surely, being a retired mining industry executive and not having a long publication track record does not translate well into undertanding paleoclimate statistics?
For Judith’s students, re comments emailed but not posted — a good recent summary on process:
http://www.useit.com/alertbox/participation_inequality.html
“On any given user-participation site, you almost always hear from the same 1% of users, who almost certainly differ from the 90% you never hear from. This can cause trouble for several reasons …”
Re #278
I would think you would seek the council of a 50yr expert given that the individual has produced results justifying them being labeled as an ‘expert’. I guess the question at hand is has Gray’s forecast technique proven to be a reasonably accurate forecast?
Judy, I take back the comment about how cute Kim is, in case that is what is preventing her from participating. I kinda wonder if that is the only thing stopping her, though. Maybe she does not like intense science debate, particularly if it is skeptical and mathematically heavy and on her work. Mike Mann doesn’t like that either.
I think it’s interesting how wimpy “the students” and “the lurkers” are. Why do they have to go running to mommie Judy up in arms when they see something here. It’s the frigging internet. Just post. Sheesh. I’m not sure how much of this is personal lack of courage, how much weakness in position, and how much weakness in intellect. At least Judy has the guts to come in here and take on the “zoo animals” when they question her 5 year binnings and such.
Oh…on 283, it is a shame, since her paper was interesting and the corals seem to have a great (much better then tree ring) correlation to sea temp. But we won’t beg. We have (some) pride.
Too much, perhaps. You weren’t the worst. Think how offputting it is to see a thread begun with the blogger’s personal opinion of the researcher’s f*-quotient, eh? I’ve seen worse, but not since the 1970s — and never from putatively conservative hosts. It’s not the sense of pride that’s the problem, it’s the lack of the sense of tact.
On a science site it’s just ugly.
“In an excellent statistical analysis, fat, balding, and ugly Dr. Jekyll has once again shown the world that a scaly scalp and wandering eye are no impediment to doing good science. We hope he will participate here ….”
Re #286
Hank Roberts, since you failed to quote Steve M’s grave PC offense, so I thought I would help out:
My advice, to anyone that is offended by the above quote, is to seek profesional help.
re 265, 283 – 286:
I missed that original thread. I deal with this garbage wayyy too much. SteveM, you’re the blog owner, and you set the tone. Your initial post led the way with an inappropriate comment about Kim, and TCO (true to form) took it further. TCO has little discipline about such things and he needs to grow up. But Steve, you bear more responsibility for this one than does TCO.
Don’t like being dissed by Mann et al? The comments made set a similarly poisoned framework for conversation. The blog equivalent of hanging a “babe” poster on the wall of your cubicle (and expecting professional women to ignore it.)
And TCO, this is not elementary school. “I take back” on a hurtful comment doesn’t undo the damage. Real contrition is about more than the words.
Sadly, this exchange illustrates one aspect of what I wrote in another thread: one way the online world “maximises offense” is the reduced barriers to inappropriate talk. We say things online that we’d never say face to face. (Yes, that can be used for good but more often it is not.)
Enough. As Judith noted, minimize the existing damage, avoid such garbage in the future, and you might get better participation all around.
MrPete
-=-=-=-
PS: to make nice short “re” links as here:
1) Right click on the time-link at the bottom of the post in question (i.e. for #286, right click on “2:44 am”.
2) Choose “Copy Link Location” (FireFox), or “Copy Shortcut” (Internet Explorer)
3) Write the linkable word in your comment (e.g. “286”)
4) Select the linkable word (if only one word, just double click on it)
5) Press the “link” button above the comment box
6) (Ctrl-V) Paste the saved link url
7) Press Enter (or click “OK”, same thing)
Easy to do, takes much longer to read these instructions than to do it 😉
re #287
Cliff, you don’t get it. Sure, “she’s very pretty” is not an offensive comment in the right context, and it may well be true. This is not the right context for such comments.
It’s no different than how a team maintains civility in various kinds of meetings: keep the conversation focused on the topic and on the work people do, without attaching comments to the person. It’s bad enough that the general commentary of professional science blogs so often praises/attacks people (rather than the science). Taking it down another rung to comment on physical attributes just degenerates the conversation even more.
Once upon a time, I was part of a team that needed to work on this kind of thing.
We performed a (painfully difficult) experiment for a few weeks: nobody was allowed to use words such as “you”, “he” or “she” during work-review meetings. Everything said had to refer to the work itself, rather than the people who did the work.
That was SOOOooo hard! But it helped us learn to stay focused on the work.
I’d love to see this community learn to stay focused on the science, the policy, the various publications.
I agree with #287 because as a women I find that excuse lame for not participating.
Women I know and respect would take the compliment, laugh, or give it right back, ie: take up the challange. Even saying “blogs are just weird” would be better reason for not participating then that hype.
I like when SteveM illustrates and colors his personal commentary with descriptions. Sheesh, this is life. Enjoy it while you try to figure it out.
This science blog straddles a line between personal blog and science forum. Not sure that it makes sense to apply the rules of scientific collegiality to the blogosphere. If you do, you’re bound to get offended.
Re #288,#289,#290
MrPete, you are addressing the superficial form, not the basic principal. Civility depends on respect for other human-beings.
Steve M related meeting the author, of a paper that he found interesting and made the very human observation that she was pretty. In no way did Steve M treat the the review of Kim Cobb’s paper differently than other papers he has reviewed on this blog. Where is the disrespect? Are you suggesting that Kim’s work was devalued by the comment and she was rendered less than the professional that she is?
You on the other hand ride to the rescue of Hank Roberts, who equates Steve M with “In an excellent statistical analysis, fat, balding, and ugly Dr. Jekyll has once again shown the world that a scaly scalp and wandering eye are no impediment to doing good science.” – nice show of civility there, but in your world that doesn’t count. Why is that? Is SteveM not a human-being that deserves your respect?
You equate Steve M’s comment to “The blog equivalent of hanging a “babe” poster on the wall of your cubicle (and expecting professional women to ignore it.)”- this I find ridiculous, unless of course you found the comment in some way titillating (in that case seek help). Further, ‘serious professionals’ of either sex, that would reject the contents of this site for a comment that someone is pretty, are neither serious nor professional.
The major lack of civility that I find on this site is guests, like yourself, that refuse to grant your host the right to be human. Somehow, because Steve M has created this fine place for science to happen and allows you to be a guest, you have the right to criticize him for being even slightly less than perfect. To my mind, Steve M is a gracious host that has earned the level respect that forgives the fact that he is also human. If you can’t grant him that respect, where is your civility?
I took it back…I took it back. Come on out and play, now!
Not sure where to post this:
Danish National Space Center
Didn’t know the Danes had a space research center, but top marks for thir research.
re: various
So Mr. Pete is into political correctness. Why does this not surprise me?
Now, unlike the claim above, Steve M does not appear to be a politica or cultural conservative, though a lot of the skeptics here are. But even if he were, why would anyone say that they never run into conservative hosts who aren’t politically correct? Don’t you read Ann Coulter, or David Horowitz or even Jonah Goldberg? It’s true that the use of extreme politeness is more prevelent among conservatives than liberals (in US uses of the terms), but it’s hardly universal.
Re: #279
I would very much like to hear a reply to my concerns in #279 concerning Gray’s model skills in predicting abrupt changes year over year for named storms (using Willis E graphed data from 1984 through 2006 for actual versus Gray) and, given his annual adjustments for predicting with the benefit of past trends and autocorrelation, the proper method of evaluating that skill.
To make my point more clear I calculated an average year over year differential for the Gray prediction for named stoarms from 1984 through 2006 and compared it to the actual year over year differentials.
For Gray, the average differentials (using absolute values) = 1.95
For the actual, the average differentials (using absolute values) = 4.59
I then took a std dev knowing that the distributions may not be normal but simply to compare the variation from the average for Gray and actual and calculated:
For Gray, std dev = 1.36
For actual, std dev = 4.91
I continue to say that from both practical and theoretical standpoints a predictive model should be finding the extreme years results. Otherwise it would appear to be merely making a conservative adjustment for a trend that appears to tend to return to a trending mean.
I am as skeptical of the skill in predicting hurricane activity as I am for the skill of investment schemes and particularly so after a quick reveiw of how they are developed.
I would go one step further here and challenge Dr. Curry to put forth raw data from hurricane predictions with computer modelling for comparison. The results must be out-of-sample and explained as to whether and how many adjustments are made over the annual hurricane season.
The contents of this thread to date would appear to indicate just how little is actually known about the variables in hurricane formation and the grasping for straws that occurs when predictions fall on their faces, i.e. Saharan dust.
I heard a fascinating intervies this morning with Richard Lindzen (Charlotte, NC Public Radio). It was a bit of a cream puff interview (no Diane Rheams type probing questions), but gave Lindzen plenty of opportunity to speak . The interviewer started by reading Lindzen’s WSJ op-ed last April. Since this is the Gray thread, I will focus on comparing and contrasting Lindzen’s and Gray’s view on AGW (mostly relative to what is included in Gray’s paper). Lindzen and Gray are arguably the two iconoclastic AGW skeptics, but as pointed out in the Achenbach interview, they don’t interact or really march to the same tune. If anyone is unfamiliar with Lindzen, he is a Prof at MIT and a member of the NAS. Heavy credentials.
Point 1. Lindzen was asked when global warming alarmism began. Lindzen said it started in the 1970’s by the international diplomacy crowd and enviro advocacy groups, who needed a new cause after the air pollution issue was (sort of) brought under control. This is in contrast to Gray, who says the issue in the 1970’s was “global cooling”. Lindzen said that during this period AGW was studied by scientists, and it was regarded merely as an interesting research topic. He views the beginning of AGW “alarmism” in 1988 with Jim Hansen’s testimony. Gray’s issue of global cooling in the 70’s was a minor issue in the popular media, not an issue in scientific circles.
Point 2. Re the “establishment” and the IPCC, Lindzen was mainly critical of the summary for policy makers, said it didn’t reflect the science in the main part of the document. He said that the IPCC report correctly characterized the model simulations as highly uncertain on regional scales. He views the models as hugely uncertain with the negative feedbacks not correctly treated, but it is my understanding he is mainly referring to water vapor feedback and the part of this that is influenced by tropical convection. (Gray cites the water vapor feedback thing as the main uncertainty also). Lindzen’s water vapor feedback ideas have been debunked in the literature and are not generally accepted, although there are arguably some uncertainties in aspects of this. Lindzen stated that without the feedbacks (of either sign), the warming from doubling CO2 would be 1C (which is arguably still large enough to be concerned about, IMO)
Point 3. Re funding. Lindzen made the point (correct IMO) that the quickest way for scientists to lose their funding is to declare consensus, “the debate is over”. He said research funding is needed to address the uncertainties, and that more research was needed to quell the alarmism. He did say that people outside the climate community were jumping on the AGW bandwagon to try to get funding for their research in ancillary areas by connecting it to AGW. This view is in contrast to Gray’s, who seems to be accusing climate researchers of sounding the alarm to get more funding for their personal research.
Point 4. Lindzen was asked whether the dispute was about the observations, or about the interpretation of the observations. Lindzen said there wasn’t much dispute about the observations, yes there is uncertainty, but they are what they are. He said the dispute is about understanding how the climate system works. Gray tends to dispute the observations. (p.s.: Steve, I am not sure if it is appropriate to infer that Lindzen accepts the hockey stick observations?)
Point 5. Lindzen was asked about the role of industry and the government in skepticism. He made an interesting point here. He said that the industries that were against AGW didn’t want people to think about the issue, and were trying to ignore it and/or obfuscate it. He said the govt on the other hand, wants to understand it (he was not buying the government suppression of scientists stuff). He claimed he had never received any industry funds (not entirely correct)
Point 6. Lindzen was then asked about how we should be responding to the threat of AGW, given the uncertainties. He didn’t think there was any need for action now, he urged the politicians to be patient and not rush to action. He thought limiting emissions wasn’t going to work, and no one knew how to make it happen and whether it would have any effect. He thought Kyoto was disastrous as a policy, it wouldn’t have done anything to reduce CO2 and would have cost the developing world a lot of money. He thought mitigation strategies were much more sensible, you know how they will work. Unlike Gray, Lindzen does not apparently view limiting emissions as crippling the energy companies. He said energy companies don’t care if they are forced to change to alternative energies, they will just pass on the costs to the consumers and still make money (I think I am interpreting this correctly, was sort of a complex argument).
I just realized that I am playing “reporter” here. Given all of my criticisms of journalists and the media, I hope this report is accurate (I didn’t take notes during his interview). All in all, the interview was much more interesting than the WSJ oped.
RE: #298 – RE: The initation of alarmism – Let us not forget the “eco pop culture” fiction books during the 70s such as “Ecotopia,” “The Monkey Wrench Gang,” and “Greenhouse” (among others of this genre). Numerous people in college during the late 70s and early 80s were exposed to these books and embraced them as part of “the cause.” Mann is right in there, among others, in terms of the timing and the demographic.
#298. Thanks for the report.
It’s interesting to look back at the start of AGW as an issue. Much of the early research was sponsored by nuclear labs – the Carbon Dioxide Information Center sponsored Phil Jones for example; Ben Santer is at a nuclear lab. I chatted about this with a former diplomat in Sweden who said that the environmental groups such as Greenpeace were not active in the early stages (pre-1985) because of the implications of AGW for nuclear power.
In terms of economic impact, it seems to me that the most important practical economic consequence is not on the oil companies, but between coal and nuclear as means of producing electricity. If people are actually concerned about increasing CO2, the #1 issue right now is the development of new coal-fired electrical capacity in India and China, which will lock in consumption patterns for 1-2 generations. If people want to apply carbon credits to support nuclear construction in India and China, that would seem to me to be a rational program to deal with biggest topic. Everything else is nibbling around the edges.
Climate scientists are much too quick to blame critics as being stooges without examining the reasons for the criticism.
Re 298:
It looks like the interview will be re-broadcast this evening at 9:00 eastern time on WFAE in Charlotte. You can find a link to the audio stream here.
Alternatively it appears that the station makes podcasts of the show (Charlotte Talks) available the next day and you should be able to find it here.
#298 – the hockey stick “observations” – I’m not sure what you mean by observations in this context. do you mean the observation that bristlecone site chronologies show more growth in the 20th century than in the MWP?
Re #298
I’m not sure what Gray might be “accusing” people of; however Lindzen’s view is not in contrast to mine, which is that sounding the alarm gets climate impacts researchers more grant money (i.e. it’s not the climatologists themselves that are sounding the alarm in order to get more money). Conflating the two groups (climate dynamics vs climate impacts) leads to a flawed analysis of the relationship between scientific uncertainty and funding levels.
A clarification on something interesting that Lindzen said. He differentiated “industry stooges” as a separate category, people who were interested in obfuscating the issue towards supporting their own agenda, as opposed to people that are interested in the scientific truth. This is an important distinction, separating the Marshall Institute type reports (many of which are of the stooge nature), vs the more credible scientific scepticism. The challenge is for a bona fide skeptic to steer clear of being associated with stoogedom. Lindzen (and SteveM) IMO has done a better job of this than Michaels, Gray, OBrien. Once a scientist is associated with stoogedom, then attempts are made to discredit their science because of the association with stoogedom (e.g. some people tried to discredit Pat Michaels hurricane paper for stoogedom reasons than scientific reasons). Gray’s use of the 1970’s global cooling thing is one of the favorite arguments of AGW stoogedom and doesn’t hold up under any analysis (I suspect Gray came up with this on his own, independently of stoogedom PR). So stoogeism is arguably making the job of the real skeptic more difficult. The reverse is also a true. The enviro groups do sort of the same thing but with a somewhat different strategy (although arguably not as effectively), and close alignment with the more extreme enviro advocacy groups can hurt the credibility of pro AGW scientists. As a result of the policy relevance of climate research, all this definitely does get in the way of sorting out the “truth” and uncertainty. This whole issue presents a huge challenge to scientists working on relevant problems (as well as to the public who is trying to make sense of it all).
Judith,
Since you want to spend time here, perhaps you’d like to spend some actual science rather than the he said/she said/they said guff which may fascinate sociologists but no-one else. You made complaint that a discussion of people’s opinions bored you and then at the next breath start making exactly those tedious observations.
In terms of what Lindzen said and what Gray said about global warming alarmism, it’s really a matter of semantics when it began: most people including Gray, first heard of global warming as a serious issue with James Hansen’s seriously overhyped testimony to Congress in 1988 which contained many statements in it which were since proven to be wrong.
But alarmism about changing climate is very old and usually associated with cooling periods. Certainly the obsession with cooling and possibly being on the cusp of a new Ice Age was discussed all through the 1970s, especially by alarmist books published in 1976 ("The Genesis Strategy" by Stephen Schneider and "The Cooling" by Lowell Ponte) warning of dire consequences of continued cooling which had happened more or less continuously since 1940 – and this despite carbon dioxide rising steeply, natch.
The transition from global cooling to global warming alarm happened in 1977, even as the cooling trend was coming to an end (although no-one knew this at the time, of course) with a convention held in 1977 in Washington DC. This is commented on in a remarkable book called "The End – The Imminent Ice Age and how we can stop it" by Larry Ephron
Ephron continues that the new orthodoxy was already making its presence felt through ways that Steve McIntyre and readers of this blog have become very familiar. He quotes from the Sanctuary, the magazine of the Massachusetts Audubon Society from September 1984 quoting Orie Loucks, then director of Holcomb Research Institute at Butler University, Indiana:
Ephron describes the 1981 and 1982 Congressional Hearings under the banner "Carbon Dioxide and Climate: The Greenhouse Effect", co-chaired by Representative James Scheuer of New York and a certain Albert Gore, Jr., of Tennessee.
Yep, that Al Gore…
Larry Ephron again:
Also testifying at those hearings was someone we all know and love:
So you takes your pick and you makes your choice, Judith. Either way, the key characteristic was climate scientists chasing the dollars from government and blocking publication of papers which did not conform to a political orthodoxy. That, I strongly suspect, is the motivation of a lot of climate scientists today, including you because you won’t get published and you won’t get funded otherwise.
Plus àƒ⦡ change……
Source: "The End" by Larry Ephron, 1988 ISBN: 0890875073
JohnA, re the global cooling myth, In the 1970’s, we were just starting to understand what caused the ice ages on time scales of tens of thousands of years. We were also starting to understand in a quantitative way the impact of both carbon dioxide (warming effect) and pollution aerosol (cooling effect). Climate models were in their infancy, and the National Academies of Sciences reported in 1975 that “…we do not have a good quantitative understanding of our climate machine and what determines its course. Without the fundamental understanding, it does not seem possible to predict climate.” Credible scientists were not predicting an imminent global cooling, in spite of what appeared in the popular press and what a few scientists might have said.
Thanks for the advice, Dr. Curry, but I would assume that as thinking mature adults most of us are aware of the perils of stoogedom as much as we are of consensusdom. Can we get back to the more substantive part of the discussion — or is that all there is?
Broad brush on …. the American masses have, since the 1930s (as a result of the Great Depression and subsequent approach of FDR and his many successors) tended to be suspicious of big business, envious of the wealthy and never fully accepting of the political Right. As a consequence, stoogism has a much more negative impact on mass acceptance than association with radical enviromentalists. This state of affairs appears to be at a current peak, due to the combination of the macro social factor I’ve allueded to and the Boomer demographic being currently in power. Mainstream US culture is decisively more to the Left than it was a mere 40 years ago, and the rebel and the radical hold a sort of hero status for many, whether overtly or subconsciously. We now see all but the most Rightist fringe politicians, corporate people far and wide, and of course, academia and government social servants, embracing the point of view popularized in the film “An Inconvenient Truth” and as a result, creating substantial peer pressure to conform to AGW alarmism.
#298 (point 3) Judith Curry – Lindzen and Gray may be different ends of the same stick. Indeed declaring a consensus that the science is settled would probably reduce funding for examining the foundations of that science, yet it would not inhibit research into the consequences. Their views are not that contrasting, both may be valid.
Conform or be cast out …
This is my last comment/post/blog (whatever the terminology) to climateaudit. It has been an interesting experience. I would like to thank all of those that engaged in stimulating discussions on the hurricane threads. I have collected my data, and decided that it is time for me to move on. As we all move forward in attempting to understand how the climate system works and the social dynamic associated with the economic and policy implications of global warming, I hope that you will keep an open mind, continually challenge your own biases and prejudices, and work to make this blog a place where substantive discussions on climate change research can take place. Challenge your ideas and reasoning by looking at a variety of different sources with different viewpoints. And do not ignore the sociology of climate research, skepticism, and even the blogosphere. Ignoring the sociology of the situation has resulted in unnecessary elevation of conflicts that are counterproductive to scientific progress. It is not coincidence that many of even the more serious statisticians in the group choose not to comment under their real name and that very few climate researchers comment here. One of the commenters on the Georgia Tech thread used the word “blood sport” and this has stuck in my mind as an apt characterization.
To those of you that I have developed an off blog dialogue with, I look forward to continuing to communicate with you on hurricane research. Progress in climate science can be enhanced by meaningful dialogue and collaboration with statisticians. But the desire for “blood sport” is frankly not very conducive to this process. Best wishes to all of you.
Bye. Don’t let alarmism infect your science.
Judith,
At least try to not rewrite history in front of our eyes and credit us with a least some intelligence. This tedious and patronizing (or is it matronizing?) approach only antagonizes. The Global Cooling hysteria of the 1970s was no myth unless of course you’re a propagandist determined to send such embarassments into an Orwellian memory hole.
And please, this weasel distinction between "credible scientists were not predicting an imminent global cooling in spite of what appeared in the popular press and what a few scientists might have said". This will not wash.
What is the difference between a credible scientist and a mere scientist? A: It depends on what Judith Curry thinks.
Perhaps Judith would like to tell us which are the credible scientists and which are the mere "scientists" from the following:
–I.S Rasool and Stephen Schneider "Atmospheric Carbon Dioxide and aerosols: Effects of Large Increases on Global Climate" Science (July 9, 1971), 138
–Nigel Calder, editorial in New Scientist, 1975
–H.H. Lamb, quoted by Kendrick Frazier "Earth’s Cooling Climate", Science News (Nov 15, 1969)
–Reid Bryson, introduction to "The Cooling", 1976
–Kenneth E.F. Watt on air pollution and global cooling, Earth Day (1970)
Now of course, you might well say that climatology was in its infancy in tbe 1970s, and many people had not committed themselves one way or another (perhaps they felt no ideological or financial pressure to do so?), but to claim that there was no "Global Cooling" hype in the 1970s is a real denial of historical evidence. People like Stephen Schneider did not advocate one world government and control of most of the commanding heights of the economy unless he felt the need was sufficiently dire. Of course he advocates the same in the light of "catastrophic global warming" so maybe he’s learned to change horses without changing destination.
You know Steve, I wonder…because most of the people I’ve communicated with who are AGW skeptics have been socially and politically moderate or center left, with one or two notable exceptions. Their views on climate science have politically isolated them largely as a result of the actions of political extremists. What should be a purely scientific opinion on the validity of scientific evidence is conflated into an extreme Right Wing view that most if not all of the people I’ve met do not share.
Since when?
China is a big producer of aerosols yet shows warming lately. Shouldn’t China be in an aerosol-induced cooling phase?
I think there is also a filiation from the debate on supersonic super-transporters that raged in the 70’s, to the ozone hole debate, to AGW.
In the early 70’s Boeing was planning to develop a supersonic plane, and some scientists thought that the exhaust gases would deplete the ozone layer. There was no “consensus”, and there was a sound debate in the literature, but that’s when the need for models to determine the anthropogenic influence on climate became pressing. Boeing’s plans were eventually shelved, but all this attention to the ozone led to the ozone hole debate. More modeling, and this time, an international convention: big success for the environmentalists. So their next target was naturally AGW, since by then the cooling scare was over.
Finally, a word about “stoogedom”. Dr. Curry says:
Is Henrik Svensmark associated with stoogedom ?
Are Nir Shaviv and Jan Veizer associated with stoogedom ?
Is John Christy associated with stoogedom ?
Is Steve McIntyre associated with stoogedom ?
Yet they suffered multiple attempts to discredit their science. Their only “sin” is to explore theories, or present data that do not conform with the orthodoxy.
#306 said; “Credible scientists were not predicting an imminent global cooling, in spite of what appeared in the popular press and what a few scientists might have said.”
And who/what were the souces of information for ‘popular press’ reports? I suspect one source could have been press releases from ‘scientific’ organizations. I think most reporters do not study the scientific journals.
Additionally, some early (pre-mid 1970s) modeling calculations sometimes did what-if sensitivity studies about return to a snowball earth
Hey Judy: I actually agree with all your comments about the need to be skeptical of skeptics, to keep an open mind, to value science. I do think that the concern over “blood sport” tone is silly both from social and intellectual cowardice/courage standpoint. Actually, I kind of like the blood sport. It adds some spice to the dry mathematics.
I realize that not everyone is a Flame Warrior like me, so I can understand the gentler flowers being turned away (see, I can’t help it :)). However, even if you DISLIKE the blood sport personalization, you ought to hang here if you want to have your thinking pushed and push other people’s thinking. You ought to suck up the brutishness so that you get the benefit of the (well, ok, Steve’s) brilliance. I mean look at U of C economics department seminars. Those guys would (figuratively) rip each other new butt-holes. At Harvard, the economists always made sure to be cordial and made sure to cite the great men in the audience in the manner typical of the more styalized academics. Lot more Nobels came out of U of C, Judy. Isn’t that aggressive questioning attitude what we say we want to teach students in grad school, ug and even high school? To think for themselves?
-yer favorite zoo critter
Judith said in #24
I knew it wasn’t about the science.
#306. If you look at how short the interglacials were according to deep sea dO18 and ice core information, why wouldn’t scientists in the 1970s draw attention to the fact that the Holocene was already a pretty long interglacial?
If anyone could direct me to an article from the period explaining why the present interglacial could be expected to last much longer than past interglacials, I’d be very interested.
#319. Many executives of large corporations don’t like questions from shareholders (shall we call them “amateur executives”?) about how the company is being run either.
#320 I was thinking something along those lines also..
And I don’t get this! Can someone tell me?
1. are skeptics [amatuer or not] supposed to come up for another reason their theory is right?
when they demand “what’s the alternate?” or
2. Do they want skeptics to come up for another reason [on demand] why the earth may be warming? [as if we have all the answers about the way our planet behaves if we claim they don’t have all the answers]
sorry that was really disjointed! Do you see what I am asking?
It seems sometimes GW believers want another reason their theory is right if they don’t like the reason we say it is crappy.
Does that make sense?
Sorry, it still doesn’t make sense what I am asking. LOL
Here, what does this mean?????
“it’s all very well to point out to this and this deficiency in the models or the observations, but what is the alternative proposed by the sceptics?”
Rocks said:
I too find it a lame excuse for not participating. What I’m saying is separate from that… 1) in my experience it’s a prejudicial comment, witness the lack of ‘appearance’ comments about other kinds of folks; 2) such comments easily lead to an unhelpful subthread; see what followed in that thread.
Agree with that too! Lots of room to be colorful.
Someone else suggested that I’m “politically correct”. ROTFLOL!!!!
PC folks want to change the vocabulary we use; that’s not at all my desire. Yes I’m sensitive, particularly in international public online forums like this one, to situations when women are gauged on how they appear but men not at all.
You’ve gotten me motivated to go to the gym now, Petey. Got to get in shape if I’m going to hang with this crowd.
ABSOLUTELY!!! Yet forgiveness implies acknowledge-and-let-go. Ignoring it is different.
Why was I strong about it? A simple demonstration that I hope will be helpful for the future of the climateaudit community. (After all this is more than just a statistical science lab — it’s also a wonderful experiment in online collaboration!)
This logic conflates two very different situations. temporary face-to-face with permanent public screen-to-screen communication.
Face to face, the scenario would almost certainly work out ok:
* A makes minor ‘appearance’ comment about B to C
* If B is present, B acknowledges somehow (“thanks!”, or a smile, or a blush if not really appropriate in the context…)
* C takes note, perhaps says something in return.
* Episode over. Few if any others take note. Forgotten in 30 seconds.
Screen-to-screen, it is different:
* A makes minor ‘appearance’ comment about B to C
* Comment is public on public blog.
(Try googling this: mcintyre very pretty kim cobb)
* A variety of elements magnify the initial minor comment:
a) Friends of A take note; comment import is magnified
b) Friends of B take note; comment import is magnified
c) Others follow up with magnified observations (‘very pretty’? Nawww, let’s go for ‘hottie’!)
d) Comment is emailed to B; import magnified
e) B wonders what the fuss is, shows up, reads the thread and realizes that she’s got a lot more than a scientific paper review to address here.
Result: minor comment from A becomes a permanent subcontext of the conversation, a subthread that will be with B forever. Anyone who ever reads that thread will know the personal “minor” things said. Not to scare you, but “taking it back”, even through admin edits, is not a reliable fix. This blog is no longer logged in archive.org, but we don’t know who has archived it.
Barring catastrophic change, public web conversations will be read by our great grandchildren. A “minor” comment is not completely forgotten 30 seconds later. [For this reason, part of my research is to discover ways to make online conversation less permanent.]
This is why Steve benefits greatly as he learns to use an extra measure of tact.
I respect Steve very highly, which is why I took the risk of saying something. Otherwise, I would have said nothing.
#319 Rocks,
It’s me you’re quoting, not Dr. Curry! It’s not a critique of skeptics per se, they have the right to their opinion! It’s just my explanation of how they are perceived by full time scientists, and why. I should know, I was one. Nothing so annoying as the brother in law who explains to you how lasers work! No professional likes to be criticized by amateurs (or executives by shareholders). As we say in French:
La critique est facile, l’art est difficile!
(quick, to your dictionnary!).
Personally, I enjoy your comments! CA would be a lot more boring without them.
TCO I do get that, but since when is “The criticism may well be valid, but it’s not constructive” true in scientific terms?
“express yourself this way only people” a way to discover pure scientific fact?
Since when do you have to make a scientist “feel good” before they share their data and methods when they publish?
Francios!
LMAO I am so sorry to you and Dr. Curry, and SteveM!
Nevermind! What an I Love Lucy moment Sheesh!
I really attributed that comment to her the whole time in my mind.
Nothing else was said later to change my mind either. My gosh. So sorry!
I like your comments too, because they are balanced but Dr.Curry gave me some other impressions so that made your comment there add to my feelings [when i thought it was hers!] Sorry sorry.
You can all slap me now!
Still, I’d like to know the meaning of “what is the alternate?”
Maybe it’s to not share your theory/model until you get it right?
Totally understood Mr. Pete truly.
I can go further then that. I come from a school that says tolerance is tolerance, and you don’t get to decide what you get to be tolerent over. I try and usually stick to that, but I do believe in -healthy- bounderies. 🙂
TCO is going to the gym and I am going to go climb under a rock after this whole thing! LOL
Cheers!
Re:#327
MrPete, why do you quote from my post and then proceed to address welikerocks post #291? Why do you ignore my questions in that quote?
Welikerocks stated that she, and women that she respects, would take the comment in the spirit in which it was offered. You proceed to spin a scenario involving a chain of idiots that have access to the internet, that somehow converts an innocent comment into a crime against humanity recorded forever on the internet – and suggest that somehow Steve M is (or should be) responsible for that chain of idiots and the crime produced.
I asked: “Are you suggesting that Kim’s work was devalued by the comment and she was rendered less than the professional that she is?” You did not answer that question, but your reply suggests that your answer is yes, by the agency of a chain of idiots. Why would any thinking person grant that power to idiots on the internet?
Kim Cobb has a very public web site at GT, on which she has posted pictures of herself and her merry crew hard at work. Does this personal exposure diminish her standing as a serious scientist? After all, some GT undergrad might visit her site and post a ‘she’s hot’ comment on a blog at GT – does this diminish GT’s standing as a serious school?
This sub-thread was started by Judith Curry in troll mode (as opposed to Dr. Curry the scientist), by way of her third party paraphrase ad hominem method, suggesting that Climate Audit was a hotbed of sexism. She implied that the source was Kim Cobb. I expect that Kim would be surprised to learn that she as been painted, by Judy, as insecure and narrow minded. Are you not applying the same brush by suggesting that Kim (and by extension other professionals) would reject this site based on your chain of idiots? Having seen Kim Cobb’s work, I have no doubt that she would have little trouble sorting the idiots.
One last comment and then I am done with this subject (apologies to Steve M and everybody else for the waste of space). IMHO, that your comments on this subject are nothing less than back-handed sexism that promotes a gap between people. If Steve M was female or Kim Cobb was male, this tempest in tea pot would have never occurred. What does that say about your logic of civility? Are you not taking a sexist position?
p.s.
If you are sincere in wanting to improve the status of Climate Audit, why are you publicly supporting and granting credit to what is clearly a series of ad hominem attacks on Climate Audit and Steve M? Also, if you want to give Steve M the benefit of your wisdom, why not in a private e-mail, instead of creating more noise here?
Re:#331
Welikerocks, get back on top of that rock! Stand tall, metaphorical warts and all. 🙂
I’ll probably set MrPete off again, but I do think that Mr. Welikerocks is a lucky guy.
Thanks Cliff! I find all these revelations rather freaky this morning, but I am up on that rock and commenting again. Mr. Welikerock’s is not surprised at all by Dr. Curry’s behavior, not as a women, but as a heavy player in climate/environmental sciences- in this arena- at this time. Pretty sad that is.
#330 Rocks,
To be confused with the eminent Dr. Curry, oh my gosh!… But I accept your apologies!
“The alternative” is really just to have a sound scientific debate, by scientists, in the scientific litterature. Quite a few papers present results that do not accord with the alarmist AGW view. But it is never stated quite clearly because, I believe, the authors are intimidated and fear to be branded as part of the “stoogedom”. They should not be afraid to take a stronger position.
Now, what would really help is to stop this madness about press releases every time someone publishes a paper! The true scientific contents of the paper ends up being totally distorted by the media spin. And I suspect that quite a few scientists, whatever their complaints about being misquoted, nevertheless enjoy the media exposure, and that this distortion is not always innocent. I propose an embargo on press releases about climate for the next 5 years!
Re the Kim Cobb exchange – any professor that posts up a picture of herself in a wetsuit on her website is obviously aware of her looks. I would venture a guess that professors, male and female, posting images of themselves in wetsuits on the internet are, on the average, fitter than professors who do not post images of themselves on the internet.
The blog has evolved a bit over the past year. A year ago, Beyonce could be referred to. Or if I wanted to tease the conservatives, I’d set a few clues knowing that they would google the lyrics to Black-Eyed Peas or even Lil Kim’s Big Momma Thang (Lil Kim not Kim Cobb). But I guess those days are past.
As others have observed, I provided sensible and reasoned comments on Cobb’s cool medieval Pacific.
While I agree with most supportive comments, on balance, Mr Pete’s comments are fair enough in this particular context and I will edit the remarks in the original thread accordingly.
I also don’t see that this particular exchange contributes to the topic on this thread and I will export the discussion to Road Map where it can remain temporarily.
Looks like “Under Armour” apparel to me – not technically a wetsuit, but it doesn’t change your point.
Well, I think Dr. Curry has given you something to think about in her stoogdem vs. science dichotomy, and I think this reflects on my earlier post about how some sceptics tend to grasp at any and every theory to bolster a weak (at least in terms of numbers) position (i.e you want your Gray and your Lindzen too). JohnA’s post on the global cooling myth is a fine example of making much of nothing. A quote about the importance of determining how long a cooling trend will last is a very sound and exploartive vanatge to hold at the time. His arguments are one step away from the “scientists once believed the earth was flat” jab.
Another examples of the “jump on science that sipports my view” is illustrated by MarkR, who gives top marks to the cosmic ray research, but surely he has not audited it. And Realclimate’s post on the issue seems to be acceptable to one of the co-authors (in the comments section).
So, anyway, Dr. Curry’s admonishment to reexamine biases might be good advice for some here. I’ll do so myself as well.
Re: #311
I think many here were very much interested in welcoming a noted climatologist to come on board to lend content and “expert” information to the discussions here generally and particularly in the case of the then hotly debated issue of hurricanes. While I think there was some clarity contributed by Dr. Curry, in the end, the tendency to engage in “blood sport” by that climatologist appeared to overwhelm the beneficial aspects of her visit here.
My view of the visit was that it was from someone who had something to say about stoogedom (fitting I guess that it ended with that note) and judged what better place to do it than here and all the while having ample opportunity to point out the symptoms of stoogedom. Fortunately, I think that the misconception that skeptics are or must necessarily be in the grasp of the big bad industrialists (that Dr. Curry seemed to imply and is implied at RC) was easily dispelled by the reactions of many posters here during her visit. Whether one wants to believe one’s eyes is another issue.
What I find of general interest in this matter is the very obvious and increasing frustrations of some of the more “urgent” AGW theoreticians in not having their message taken seriously by the public despite having the backing of the main stream media in getting “out the message”. I think that their blame and frustrations are misplaced when aiming it at the skeptics and the industrial “stoogedom”. As difficult as this always is, a good hard and continuously long look in the mirror might be good beginning towards finding some answers.
The urgent advocates of AGW and particularly those from academia seem to take an ivory tower approach that never really deals with the issues of what do we specifically do about our predictions — other than alarm. Hence the source of further frustrations emerges.
I’m not going to let that comment pass. Several of the more serious statisticians have written to me offline and identified themselves. The reasons that they don’t identify themselves is that they are afraid of employment repercussions for daring to challenge any statistical aspect connected with climate science.
Boris:
It is being audited, through physical experiments. And I can “audit” the correlations close enough by just eyeballing the graphs. No statistical problems like overfitting or cherry-picking of proxies here, just simple plots and correlations between raw data. One could audit the data, I guess. The thing I find so interesting about this research is that it provides a very sound physical basis for warming/cooling, and it can be falsified. Unlike the “evidence” for CO2-caused warming. Do you dismiss the Sun as a likely cause of warming/cooling?
“sceptics tend to grasp at any and every theory to bolster a weak (at least in terms of numbers) position”
That statement is also true for non-skeptics if you practice un-biased eye and you are speaking of scientific “numbers” like statistics and proxies of the HS.
if you mean who holds the popular position or fan base well…I don’t care about that at all, and I don’t speak for anyone else.
“And Realclimate’s post on the issue seems to be acceptable to one of the co-authors ”
The co-author is a student, so time will tell and RC hasn’t really audited the paper either, maybe they will, but I see no math,data or alternate graphs posted for the readers, just an opinion and a rather sarcastic one at that.
Sorry to come to this string soooo late, but RE: #76 – Jeff your comment “You cannot tell if there will be El Nino until after the fact…” take a look http://www.john-daly.com scroll down to the section on “Recent Guest Papers” [just beneath the Tasmanian Devil picture] and read Dr. T. Landscheidt paper. This website is no longer kept up to date since John Daly death a few years ago, but his was one the first websites that brought me into the AGW debate.
re #342 Sorry wait I take that back, they do show a graph, I didn’t click on the “more” link.
But they do go on don’t they :
“Will these results be a spur to future research? Possibly. But the ridiculous spin put on this paper is liable to continue to put off mainstream scientists from pursuing it”
What does that mean? It’s not “cool” enough to pursue if RC says so?
RC speaks for all “mainstream” scientists? Do “mainstream” scientists always follow media spin and become “put off” by it so much they stop being interested in research? And what is a “mainstream” scientist anyway? Is that another branch of science I haven’t heard off? Sort of like an “anti-AGW” scientist. Do they have masters programs and classes for these titles?
Re#338 Boris it was RealClimate who jumped all over the Danish report, just as they jump on anything that might conceivably disturb their cosy worldview.
The co-author you refer to very gently IMHO corrected the Real Climate mis-understandings.
PS Boris
How many climate models take into account that “the released electrons significantly promote the formation of building blocks for cloud condensation nuclei on which water vapour condenses to make clouds.”?
#344:
I think the meaning is plain and clear. The press release claimed that the experiment proved that cosmic rays affect our everyday weather. The RealClimate post shows the areas needed for further study before the experiment shows any large real world effect.
The spin of the press materials is obvious, and someone who may be interested in the expermient might see the press release and dismiss it altogether.
This is exactly what I’m talking about. Like it or not, those against AGW are going to get lumped together. (Of course, you do tend to self-lump to an extent.)It would be wise to not embrace everything that goes against the consensus, but to apply the same amount of rigor to this Danish study as to the latest from an RC contributor. Else, the biases become apparent and credibilty becomes an issue.
Boris, as the proprietor of this blog, I have never mentioned the Danish study. It has been mentioned by posters here. I agree that rigor should be applied to the Danish study.
#347 IMHO Unless you show me evidence of the embracing and accepting of this specific paper, by those you elude to as ” those against AGW” to me you are blowing hot air and lumping at the same time.
You might be right about the media hyping it, and it just came out right? So far all I can find is that one press release and RC is picking it a part on their blog.
Physicians heal thy self! is what I say about that.
If you have a link to more than that please share it.
#348, I realize that and did not mean to imply that you have mentioned it.
#349, My comments are both general and specific. To the Danish paper, MarkR has clearly decided it is valid and worth adding to the models, though even a co-author feels more study is needed. This is also apparent in some posters’ response to Gray, which we have discussed already.
The general comment is that some posters here who are critics of AGW (is that a better terminology?) respond differently to papers based on how those papers come down in terms of AGW. Now, not everyone here does this, I dont even know if you do it or not. So that’s my hot air. It’s meant to be constructive in terms of the credibility of the discussions here, but if you disagree fair enough.
Boris,
I wouldn’t be taken in by criticisms of the Danish National Space Centre’s press office. All press releases have an air of exaggeration. If you read the press release for Jim Hansen’s latest paper, you see liberties being taken there as well.
As for the Danish research in itself, its an interesting paper but like all papers, it needs to undergo the scientific process of replication and falsification as well as auditing of its figures. It is a hypothesis at present, nothing more.
What I would add is that there has been little coverage of the paper in the media generally (certainly in the UK) so I’m not quite sure what RC are getting at.
Re #350
Boris: to some of us, not fortunate enough to live in a country where open debate of ideas is as highly valued as in America, the AGW theory is already costing us billions with no prospects of improvement for the future. Will you please excuse us if we’re not as concerned about the validity of alternative theories as much as we are about that particularly costly one?
Of course, that is not to say that one should embrace anything that goes against AGW but I don’t think that your demand of balance in this respect is fair or realistic.
Personally, I don’t feel any urge to find an alternative paradigm. In spite of its unfortunate acceptance by the media and most politicians, AGW is untenable. It leaves too many things unexplained and the failure of other theories is not going to change that.
Pre-emptive strike, out of fear of collapse of public support for “consensus”?
re: 353 RC did it to this paper as well:
“A critique on Veizer’s Celestial Climate Driver”
and the author got wind of it and posted:
Comment 19
Comment by Jan Veizer “¢’¬? 21 Jun 2005 @ 11:46 am
which seems to have been edited ??
“Dear colleagues, I believe it would be only fair to add the folloving to the “critique” of Veizer as his response:
It would be a relatively easy task to compile a lengthy list of model deficiencies and failings, but because I appreciate the fact that progress in science is incremental, I prefer not to do it. It is my firm belief that the climate debate will ultimately be decided by the merits of the most logical case, not by the demerits of the alternatives or by personalisation of the discourse and advocacy. I only ask that those researchers that wish to arrive at their own judgement of what is it that I am really saying can do so by reading the entire article in its context (available under supplementary material). Please note that is my personal preference to confine any further discourse to scientific ways and means.”
– and rasmus has a large response inline.
Judith, thanks as always for your contributions. You say:
While the claim about the increasing error in Gray’s forecast is true, it is also meaningless because it is not statistically significant. It’s equivalent to saying:
The problem, of course, is not global warming, but inadequate sample size. In climate science this problem is more prevalent than it first appears, because of autocorrelation. With autocorrelated data sets, a larger sample size is needed in order to have statistically significant results.
Even without autocorrelation, however, it takes un-intuitively large samples to tease out small effects. For example, Gray has made 22 forecasts. For the first decade, his error was 0.2, standard deviation 2.3. For the last decade, his error was five times as large at 1.0, standard deviation 5.3.
Now, let’s assume that he maintained those same averages and standard deviations, not for ten years in each period, but for each half of a longer time. How long would he have to be forecasting before the difference between the first and last halves of his forecast became 95% significant?
a. 21 years.
b. 42 years.
c. 208 years.
d. 416 years.
Surprisingly, with the differences in averages and standard deviations in Gray’s forecasts (quoted above), the final answer is “d”. He’d have to be forecasting 416 years (208 in each half) for the difference to be statistically significant.
This is why you may not be getting much traction with your claims based on a 10 year sample size, and it is also why you should fire Holland if he is your statistical guru and is letting you make these claims.
w.
I had the impression she was impressed with Bender and Steve’s math brains. I would think that she would have access to super stats guys herself. I just sorta worry that the field is populated with guys who are smart in terms of knowing some fancy techniques and linear algebra and the like, but not smart in terms of noodling out fallacies. And fallacies are very easy to come up in stats work. Even “basic” diff e q word problems can have tricks that can trouble you if you make the wrong assumptions and modeling. Heck even algebra word problems can be prey to these issues.
That said, I do worry about the skeptics cherrypicking the cherrypicking. Gotta watch them…
PS – Sorry for the delay in answering Judith, the earthquake here in Hawaii has knocked out my internet. I’m writing this at the local Starbucks …
w.
Boris,
About the Svensmark paper: I have not seen it myself. I hope I can get hold of it soon. But one thing I know is that it’s an experimental work, and as such, it has been performed under the constraints of the experimental apparatus. I am myself an experimental physicist (mostly), so I know what I’m talking about here much more than climate or stats. If they have obtained the results that I’ve heard and read about, it’s already a great achievement, because experiments like that are not easy to perform. But it’s obvious to me that this is really only a small first step towards a better understanding of the role of cosmic rays on cloud formation. One can’t dismiss a theory on the basis of incomplete first results! Science would go nowhere if that were the case.
Forget the press release. Forget EVERY press release, including the millions that hyped uber-warming results. If you want to talk about science, read the paper and comment on it.
Now here’s one for you. That paper is entitled “Effects of an assumed cosmic ray-modulated low global cloud cover on teh Earth’s temperature”, by J. Ramirez et al, published in “Atmosfera”, vol. 19, pp. 169-179. The abstract says:
No press release for this one.
When deciding how much effort to devote to auditing a particular paper it is worth looking at the profile of the journal. What is Atmosfera’s “Impact Factor”? I’d never heard of it until now. This is not an ad hom dismissal. Just a recognition that when time is limiting it is fair to ask why a paper was published in the journal in which it is found, and not some other.
And by extension…
Re #359: Here‘s the paper. The pub seems legit, BTW (Elsevier). At a quick glance, I’m a little less sure about the paper itself; in particular the use of a 40+ year old model seems odd. It may be meaningful that nobody on either side of the solar-cosmic ray discussion has cited it.
Dismiss my analyses if you like. I’m not going to publish in the specialist literature due to (1) the immense amount of time it takes, (2) the small appetite among journal editors for healthy skepticism, and (3) the rapid rate at which warmers “move on” to the next data set, QED. Got other things to do right now. Jour. of Clim. Aud. will have to do.
Judith has learned. She can pick up the cause. Hey, she even runs a Mac. So get her to run the SSA MTM toolkit to reconstruct the MBH99 confidence intervals (or whatever the heck it is UC and Jean S are doing). Let’s see if she’s as good at tree rings as I am at hurricanes.
Re #361
I know it’s “legit”. That’s not the question. What’s its impact factor? You think I didn’t read the paper before commenting?
Wimp.
Yeah, I agree. How hard can it be to download and run that thing anyways?
#363 Bender,
The “impact factor” is the biggest fraud in science today. It has been criticized by almost everyone. Please do not judge a paper by the impact factor of the journal in which it has been published. Just think: Nature has one of the largest impact factors.
I’m not commenting on whether the results are good or not: how could I? I was posting this just for the benefit of the audience.
There’s an aspect to tropical SST that Gray has mentioned but which has not received a lot of attention. It is the relationship between tropical sea surface temperature and the strength of the trade winds.
The idea is that the stronger the trade wind, the greater the evaporation. And the greater the evaporation, the greater the cooling of the sea surface. So, stronger trade winds lead to cooler SST. And weaker trade winds lead to higher SST.
Here is a time plot of SST in the Atlantic “box” used by Emanuel. (I use July-Sept instead of August-October for rather complicated reasons I won’t get into here, unless someone is curious and asks.)
Note the sideways temperature movement until about 1970, then a drop. Since the drop, the trend has been upwards for the last 35 or so years.
Here is a time plot of something called “zonal wind”. Zonal wind is the east-to-west component of the trade winds in Emanuel’s box. Calm is zero, so that the higher the red line on the chart, the closer the wind is to calm.
Note the sideways movement until about 1970, then an increase in wind speed. Since 1970 the east-west trade wind has progressively slowed down for the last 35 or so years.
Here is a time plot of something called “meridonal wind”. Meridonal wind is the north-to-south component of the trade wind in Emanuel’s box. Calm is zero, so that the higher the red line on the chart, the closer the windspeed is to calm.
Note the (mostly) sideways movement until about 1970, then an increase in wind speed. The wind speed stayed steady until the early 1990s, at which time the north-to-south wind began to slow. It has slowled dramatically in the last 15 years.
(Now, the best presentation of this data would be to combine zonal and meridonal wind speeds into a vector and present that, but I’m not that adept. You’ll have to mentally vector.)
So, in a broad way, SST and trade winds remained steady until circa 1970, at which time there was an increase in trade wind speed and a decrease in SST. then, for the last 35 years, trade winds have steadily declined while SST have risen.
My point is that SST and windspeed sure seem to share a pattern. There is a causal mechanism (evaporation) that connects the two. I don’t contend that the speed of the trade winds accounts for all, or even most, of the Atlantic tropical SST fall and rise over the last 50+ years, but I’d bet a virtual dollar, or euro, that it has played a role.
Interestingly, last July the tropical Atlantic SST dropped from significantly above normal to near normal for several weeks. The meteorologists, including Jeff Masters, attributed the drop to a temporary increase in trade winds. I don’t know what the climatologists thought but most meteorologists still believe there is a link between the two.
Re #367 Here is the data engine that generated the time series, courtesy of either bender or Willis.
I don’t have a great faith in the fine detail of historical temperature and wind analyses over remote stretches of ocean, but it’s the best we have.
Perhaps it is my imagination, but I think I can see a multi-year decline in windspeed in the late 1980s which corresponds in time to an increase in SST. And, in the early 1990s, I think I see a multi-year increase in windspeed which corresponds in time to a decrease in SST. But I leave it to my statistical cousins to find any significant relationship.
If the link in #368 does not work, try this .
Credit Willis. The guy’s a Starbucks-fuelled machine.
A final note on #367. The meridonal (north-to-south) wind component may seem minor compared to the zonal (east-to-west) component, but the north-to-south wind likely brings cooler northern water into the tropics.
If that component weakens, then less cool water is pushed south into the tropics and the sea surface warms even more. Meridonal wind is important, in my opinion.
Note also the shift to a strong decrease in the meridonal wind component in the 1990s, about the time of the change in SST patterns and increase in hurricanes.
RE: #367 – Someone should look to see if lower speed trade winds correlate with a more southerly “horse latitudes” position / sutropical Highs being further south in position (or, alternatively, sub tropical Highs having larger extent). That could get real interesting in terms of linking what went on with the mid latitude jets during the same time periods the trades have been weak.
David,
Interesting. Can you give details of the “box” used, and other variables, so that we can all reproduce the graphs? Thanks!
Re #373
The box is 6N to 18N, and 20W to 60W.
Here is how I used the link :
Choose the NCEP/NCAR reanalysis monthly means, which takes you to another screen. There, select –
Variable = “SST” or “Zonal wind” or “meridonal wind” (pick one)
Analysis level = “surface”
Choose “season average”
First month = “July”
Second month = “September”
Choose “yes” on area weight grids
Choose either plot or raw data, depending on your purpose
It would be great if someone skilled could look at the data statistically, to see if the Atlantic warming over the last 50 years can be explained by changes in the speed and direction of the trade winds, rather than by radiative forcing.
I’ll look at Pacific data and some pressure patterns later today. I am wondering if the Pacific shows similar patterns, and if the Atlantic decline could be explained by a northward shift in the ITCZ. A northward shift of the ITCZ is important to hurricanes, because it puts the seedlings in a region where the Coriolis effect is stronger.
Re #367 David, your analysis is useful but seems to beg several questions. Part of the reason for ignoring other potential answers is created by the AGW focus on excluding external forcings, especially the sun, and even there focussing only on electromagentic radiation. Labitzke and van Loon have published on the relationships between El Nino/ La Nina oscillations and sunspot activity. The evolution of theory about El Nino/La Nina is based on the assumption that the zonal winds ( zonal and meridional are terms long applied to the changes in flow of the circumpolar vortex) reverse to create the Southern Oscillation. As everyone knows surface ocean currents are primarily a function of wind, so to reverse the currents the wind must reverse. The question then is what reverses the wind? Classic theory would say it is due to changing pressure gradients but nobody to my knowledge then explains what causes the pressure differences. Again classic theory would say it is due to differential surface heating but I have never seen that explored or explained either. You provide two scenarios, zonal with easterly winds and calm, and meridional. As you mention in #371 the degree of meridionality is an important factor in changing weather patterns. This is certainly true for circumpolar meridionality and has led to the theory of MPHs presented by Marcel Leroux. It seems to me you need to have a westerly flow of the Easterlies to create the reverse flows of the ocean currents. I discuss this and a possible explanation for this reversal.
I continue to be confused about the measurement of skill in forecasting named storms for the NATL. There are several scenarios for predicting that to my mind would require different measurements of skills:
1. Seasonal predictions are made using the same model, with not only the output published but the included workings of the model so the prediction results could be repeated and fixed at the beginning of the out-of-sample period.
2. Seasonal predictions are made with a model that can change annually and without publication of the workings for repeating the results and/or some subjective tweaking can be applied to the model results.
3. Seasonal predictions are made as noted in 2 above except that the totals for the seasons are not from the beginning of the season but the numbers for each month that can be adjusted monthly and then added together.
Under conditions in 1 above, the proper applications of tropical storm physics, either as directly applied or as applied in the form of sensible model correlations, would be given the test of out-sample-testing for the whole period — and by rather standard methods.
Under conditions in 2 above, the model prediction is given the added flexibility of subjective and/or unpublicized inputs that can account for reversion to the mean and any short term trends which certainly detracts from the use of straight climate physics and the sensible correlations in the model. If the model changes than the out-of-sample test period would require adjustment.
Under conditions in 3 above, the model has all the flexibility as in 2 above except now intra year adjustments can be made to make the totals come closer to the actual by, for example, predicting higher numbers for a coming month when it appears the past months’ predictions have been too low. The only proper way of handling that possibility would be to evaluate the predictive skill for each and every predicting period and not for the season.
Can anyone described for me the procedures used by models/services in predicting the seasonal number of named tropical storms or am I beating the horse, on which I thought Dr. Curry galloped into this thread and might it now be considered dead? I would think this information would be required in a complete audit of any storm predicting claims.
Thanks for the link, Dr. Ball, i’ll read it this evening.
Here is a plot of zonal wind in the eastern Pacific, along the equator. The big El Ninos of 1983 and 1998 are easy to spot on the plot. Even in those big El Ninos, the wind continued to blow from the east, but at a reduced rate.
Interesting to me is that the zonal flow seems to be on a downtrend – it has become less windy along this stretch of the equator over the last 50 years, which presumably means less evaporation and less cooling.
I also find it interesting that there seems to be a noticeable increase in the rate-of-change in the 1970s. Many atmospheric parameters seem to have shifted in the 1970s, some rather quickly. Interesting decade.
David
David,
I’ve made some interesting plots and analysis, including the vectorial amplitude. There is a reasonably good correlation between wind (zonal, meridional, or vectorial) and SST, but the correlation is much better if you take the data 1948-1987. R2 is then 0.5. Between 1988 and 2006, the correlation breaks down and SST seems to be almost independent of wind amplitude, with R2=0.014. SST’s are also generally higher (between 26.5 and 27, vs 24.5 to 27 for the rest of the data; all the 1988-2006 SST’s are above the trend line for 1948-87). If there was a shift in behavior, it seems to have occured in the mid- to late 80’s.
Did they change the wind measurement method around 1988? The data sure look completely different from that year on. There is a discontinuity in both values and trend.
RE: #377 – Australia, New Guinea, Indonesia, the Philippines and the peninsulas of continental SE Asia form a trap / “capacitor” relative to the Equatorial Current / trades. This is exacerbated by persistent low atmospheric pressure in the area. One mechanism I’ve seen proposed is that the warm pool is sufficiently elevated versus the zero geopotential surface, that if there is any relaxation of the trades, the potential gravitation energy of the mass is sufficient to exert a countercurrent. That might get kind of interesting given the existing bracketing countercurrents on either side of the main equatorial. When the warm pool collapses, and some of it makes it across to the Americas, we get an El Nino. The trades do not need to change direction, only lessen. Obviously I’ve oversimplified the mechanism.
Re 367 David Smith says
There is another, possibly stronger, mechanism. Stronger winds generate greater turbulence which deepens the mixed layer. More deeper, colder water is entrained which cools the mixed layer.
Re #381 John, my two cents is that you are right. I don’t know the relative strength of the two effects, but both are factors.
Re #379 Francois, I don’t know if there was a change in the late 1980s. The stretch of ocean is remote, with no fixed buoys until the 1990s (I think) and even those are few in number and tend to be in the western portion. Beyond buoys, I imagine that ship reports were used, even though they were few in number, and then satellite measurements came into play in recent decades. I imagine there is a lot of interpolation across large regions. To cap it off, the temperature differences being found are in fractions of a degree C. It may be that all we can say is that the observed trade wind variation is consistent with the observed SST variation.
Draw your own conclusions, but I find the following rather telling:
ISTM that proponents of AGW can “move on” when someones paper is shown to have significant errors, and that it “doesn’t matter” because the rest of the papers (many of which have not yet been “audited”) put together a compelling case.
And yet, if one were to put together all the papers and views that seem to indicate that the A in AGW is maybe not quite as much as was thought, this does not, apparently, represent a compelling case.
For instance the CR paper – ISTM that RC say this doesn’t really show anything about climate. And neither do the papers that link sunspot activity to ENSO events and other climate altering oscillations, seemingly because there is no known physical mechanism to link the two. Yet together, they certainly make a compelling case that, at the very least, there is something going on here that it seems very few climatologists have considered. It “hangs together” in the same way that the AGW & CO2 hypothesis does. I have no idea which is right or wrong – hell, maybe they’re *both* right to some degree and the answer is a little from both.
I also find it incredible that someones funding source is relevent, but apparently *only* if they are against AGW! If you work for “oil”, you are suspect, but if you work for “greenies” you are not?
Finally, ISTM that someone like Dr Grey can be dismissed because he’s changed his mind about a correlation he’s noticed over many years, yet we cannot dismiss those who have changed from “coolers” to “warmers”.
It appears to me, from the above, that there is *way* too much politics involved in the AGW hypothesis and not nearly enough science.
Neil, that is AGWer double-standard #5. [Yes, I’m counting. Yes, I keep a file. Yes, I’ll post it … when the time comes.]
#382 David,
I’ve e-mailed my graphs to Steve M., since I can’t upload anything. If he manages to post them here, or e-mails them to you, you will see what I mean. Or simply plot the data in Excel, say of SST vs wind, but don’t use lines in between the points, and plot data points of a different color for the periods 1948-1987 and 1988-2006. Then have a trend line for each period. Those two sets of data cannot have been obtained the same way, and of the sets has to be wrong. Between 1948 and 1987, there is a clear correlation between SST and wind, especially zonal wind. For the next period, there is just no correlation.
Those are not raw data, as I understand it, but reanalysis. No indication on the web site of how the raw data were obtained.
Gotta go, it’s CSI time…
Re #385 Thanks, Francois. I’m doing some reading this evening on data collection and reanalysis, the heat removal capability of wind and the lag between wind and detectable cooling. This may go nowhere but it’s a chance to learn.
Francois’ xls sheet is online here http://data.climateaudit.org/data/hurricane/analysis SST wind.xls
It seems like just a frigging chaotic system, but I keep thinking there has got to be some physical explanation for all these quandries. I still think the Sun has something to do with it. Perhaps through the formation of clouds, which change the delta T between the atmosphere and the ocean? Somebody will figure it out, probably soon.
#387 Steve, Thanks but the link does not work!
#389 to make it work: copy and paste the entire address, up to .xls, and paste in the address bar.
I’ll try wind data from other “boxes” to see if they show the same pattern.
Comments?
Gray modelled his TS predictions based on long term historical data and started predicting in 1984. He wrote a paper indicating changes to that basic model in the early 1990s. Any connections with what Francois O has found.
The car salesman did it. Well so did the priest, but only once.
Francois, those are intriguing charts. It indeed looks like something happened to the database, perhaps changes in methodology. My reading has, so far, turned up no clear history of methodology.
However, it is apparent that there are many issues and that reanalysis is difficult. Some are subtle, like the average height of anemometers on ships. As ships got bigger and taller over the years, the wind measurement devices were mounted higher and higher on the ship, which makes a difference. So does the wind disruption caused by the ship profile. Imagine trying to sort through records to figure all of that out.
Even buoy wind device technology changed over the years, with early anemometers having restriction and pluggage problems.
If you’re looking for anothe “box” to check, you might try 5N to 15N, 130E to 180E. that covers a lot of the western pacific storm breeding ground.
Is this the area from which your plotted .xls data came Francois O? I was attempting to track back in the thread to find it.
Relative humidity of the Atlantic tropical box is shown in the link.
Evaporation, and thus the SST/wind connection, is affected by humidity. If humidity rose dramatically, then evaporation would suffer and the SST/wind relationship would change.
However, the plot shows no increase in humidity in the tropical Atlantic box. So, humidity is not an explanation for the post-1988 breakdown of SST/wind correlation observed by Francois.
Yep, I’m a klutz on links. try this for relative humidity in #394
According to Jan Veizer’s paper, the Earth’s albedo decreased greatly (i.e., less cloudiness) from 1985 to 2000. Could this somehow have an impact on the breakdown of SST/wind correlation observed by Francois? Less cloudiness might affect the delta T between the surface and the atmosphere?
RE: #392 – Since I am somewhat of a West Coast hurricane geek / storm chaser, I thought I’d add my two pence. I’ve noticed a few tropicals cyclones that had (at least) two lives. The first being life as an Acapulco / Guerrerro coast storm, working it’s way NW or NNW, then, in a manner never seen in the Atlantic, turning West just as it hits waters below 75 deg F or so somewhere around 23 N. By then, if not before, they have become mere depressions. I’ve seen a few of these then dawdle along toward the toward the Westsouthwest only to have a second life, forming a “new” storm in that very box you have noted. Some of these go on to be bona fide typhoons west of the IDL.
My apologies for going off-topic, but may I post here a “backup copy” of a comment I’ve sent to RC? This is not the intended audience but there’s an interesting discussion going on there which somehow ties in with Gray’s lack of confidence in GCMs. Mann is not missing the chance to disdain non-experts and judge their contributions as “nonsense”.
Thanks in advance.
————-
I’m just a humble observer but it seems to me that the main points Nicola and Ferdinand are making are quite reasonable.
Let’s put it this way: it is assumed that sensitivity to 2xCO2 is around +3 C. And we know that, since the late 19th century, CO2 has risen x1.35 and global temperature approximately +0.7 C. Well, if we come to the conclusion that half of that T rise was solar-induced or that the negative effect on T of sulphate aerosols over this period was much lower than previously estimated, we’ll have to revise our 2xCO2 sensitivity estimation downwards.
(As a matter of fact, we might well have already experienced the equivalent of a 1.5xCO2 increase when we take into account the rest of AGHGs for a total T increase of only +0.7C, but let’s not complicate matters).
#393 Yep, that’s the box. I have looked at the Pacific box suggested by David, but there is no visible trend or correlation between SST and wind in that one, and no sudden change in trend for the wind. I only see two explanations : either there was a real sudden shift in climate, or the data are wrong. In the first case, one has to find a physical explanation why SST continues its upward trend, while the wind trend changes direction. From 1948-87, less wind means high SST. After that, more wind means high SST. Mind you, it’s also possible that those two variables are just independent from each other, in which case their correlation pre-1988 would just be spurious. Is SST really that dependent on surface wind amplitude, or is it determined by other factors such as, as suggested, insolation, or ocean currents ? One might want to look at cloudiness, for example. It’s still puzzling that the SST trend is intact over all the period. Doesn’t even show the 1940-1975 cooling that is, on the other hand, apparent in the Pacific SST’s.
Actually, I did read that paper. And I noted that he very correctly predicted that La Nina conditions would develop from April 2004 to April 2005 and that El Nino conditions would develop starting in July 2006. This is from a paper published in 2003. So I more-or-less assumed it was a fluke even though it was very accurate. I also wanted to see if Madam Curry knew there were accurate and inaccurate models out there.
Don’t forget the animals..
“Marine Life Stirs Ocean Enough To Affect Climate, Study Says”
http://tinyurl.com/yapzaw
http://tinyurl.com/yh9ozr:
“Methane Devourer Discovered In The Arctic”
#383
I would suspect that they can “move on” because the errors, in their minds (and the NAS) do not alter the results of the study. In order to dismiss the other studies, keep in mind you must assume errors are there to be found.
As to the CR paper, the double standard is more evident here, as some have leaped to the conclusion that this paper is valid without an audit (or evaluation) and without specifically showing what is wrong with RC’s interpretation of the paper. This is even more important since a co-author on the paper has apparently posted at RC and agrees to the shortcomings in the CR study. Keep in mind also that RC makes the most hay out of the paper’s misleading press release, which they claim (and I strongly agree) hurts the chances of the paper being taken seriously by climatologists. This would be a very good example of too much politics in the debate.
As to funding sources, perhaps you are right. I would say oil has more incentive to buy science that the “greenies,” but I don’t expect that view to have much sway here.
As to Gray, has he changed his mind because of observations? Surely any “coolers” who have become “warmers” did so because of new data. What new data has Gray observed that causes the THC to act in an opposite way as he has previously claimed? In order for your double standard to be correct, the “coolers” must have claimed that CO2 caused cooling, then changed it to warming when the globe showed a warming trend. This is not even an aspect of the “global cooling myth” as earlier discussed.
So my own non-AGW double standard count might be available soon as well.
Boris, careful what you doing here. The double standards of those advocating for AGW and against AGW begin to show weaknesses in the whole of the climatology community. Skeptics on AGW, like myself, are simply saying that the evidence is not there and that climatologist have not made cases their either way. One hundred years from now it could be + 5 degrees centigrade or -5 degrees centigrade or something in between.
OK, I got as far as this from JC and I lost all semblence of tollerance. Apologies for missing out anything I’ve not bothered to read on this thread so far, but it seems a fairly rambling thread, so I’m not too bothered.
Firstly, I think you’re being a little too dismissive of the ‘students’ I learnt more about the philosophy of physics from my fellow students (doing that specific course) in my first year. I learnt no stats in my degree, all that is relevant to this subject is one practical I did with equipment which (I guess) had been tampered with to intentionally produce bad results. We never reconciled our results with theory…
I’ve since learnt to be very specific with my (usually on the vitriolic side of damning) criticism of other people’s results (you say 1 + 1 = 10 for example) so I can check what I understood when I turn out to be wrong.
To say f(1995-2005) > g(1984-1994) is simply meaningless, even if you ascribe the same word to f(x) and g(x). It is imprecise and can’t be proven wrong or corrected. How can a professional scientist justify rambling like this? It’s the sort of thing I’d expect to read in the news! It suggests to me a 2nd hand re-use of someone else’s analysis…
p.s. The 1919 experiment was novel in that it was understood to be the first opportunity to PROVE that the theory was wrong. It’s agreement with the theory proved nothing. I understand that we’re still looking for more and more precise ways to find errors in that specific theory, even though the concensus is? that the theory is wrong.
#403: Not really counting, just a comment in line with #384.
I wanted to review the Dr Gray NATL tropical storm predictive skill vis a vis Dr. Curry’s contention that last decade’s degradation of these predictive skills is an indication of AGW effects on tropical storms that Gray’s models and predictions had not taken into account, i.e. SST. I do this as much for my own edification as providing a summary for the more casual readers of this blog, so bear with me.
In my mind the major issue here is that Dr. Curry and other climatologist with like positions on AGW have found Dr Gray’s pronouncements on AGW and his purported skill in predicting tropical storms as a major distraction in getting their message out.
I went back over all the posts and most of the references originating from those posts on this thread in preparation for this summary posting. I intend to editorialize as I summarize.
On thinking about the proposition that Dr. Curry was putting forward I came to the conclusion that the comparison she was attempting was of apples and oranges. Dr. Gray uses model predictors to predict ahead of time the number of named storms and hurricanes that will occur in a hurricane season. This is similar but certainly not the same as creating a model that uses variables that can explain the variability in seasonal tropical storm activity. Obviously the predictors values at the time of a tropical storm are not as well know in months’ advance of the storm as they eventually are when they form.
One can readily see this by observing the increase in predictive skill that occurs with Dr Gray’s forecasts for June versus August as measured by Owens and Landsea. See link here.
Hard as one may try it is difficult to connect Gray’s predictive skills and recent changes in them to making a case for AGW. Having said that, I find the claiming of predictive skills fascinating, and particularly so in light of past experiences in studying the analyses of stock picking strategies. Willis E has already presented an exercise in evaluating Dr. Gray’s predictive skills and made several cogent comments in the process.
The main criteria, in my mind, in evaluating those skills is in knowing how the predictions are made. In Dr. Gray’s case the predictions for the number of seasonal named storms and hurricanes have been made since 1984 in June of the year being predicted and again in August. There are two predictions for each period: one called the statistical prediction and one called an adjusted prediction. The statistical one is made using a statistical model only while the adjusted one uses that input plus others that are determined, for example, by looking at a past year with similar storm activity.
As Pielke Jr. and Willis have alluded to in posts in this thread, the choice of a benchmark or naàƒ⮶e prediction for comparison with the predicted values can be critical in the determination of predictive skill. Owens and Landsea used two benchmarks in their evaluations of Dr Gray’s predictive skills with one being the 50 year average preceding the prediction and called climatology and the other being the past year,s observed value and that being called persistence. I find the use of the 50 year average not to be consistent with the fact that the models could be adjusted on a yearly basis and thus something more recent in trending needed to be used. The past year’s result in my mind is much too naàƒ⮶e and particularly susceptible to sudden changes from one year to the next.
Klotzbach and Gray evidently use a 5 year previous average for the benchmark and that value makes more sense to me. See link here:
When I took these results for the seasons named tropical storms from 1984 through 2006 and compared the observed values to those predicted by the June Adjusted Prediction (which by the way are the values used by Dr. Curry by way of the unpublished and evidently yet to be disclosed study of skill by Greg Holland and those used by Willis E in this thread) and to the previous 5 year average (naàƒ⮶e benchmark), I found the following:
For the entire period I found the Gray prediction out performed the naàƒ⮶e benchmark13 of 23 years with 1 tie. For the first 13 years the Gray prediction out performed 10 times and in the last 10 years out performed in only 3 years and tied in 1.
The nonparametric Wicoxon Signed-Rank test for the entire period did not show that the Gray prediction outperformed the naàƒ⮶e benchmark.
Doing the same analysis for the August Adjusted Gray prediction I found:
The Gray prediction out performed the naàƒ⮶e benchmark in 15 of 23 years with 1 tie. For the first 13 years of the period the Gray prediction out performed 12 times and for the following 10 years the Gray prediction outperformed only 3 times with 1 tie. Applying the Wilcoxon Signed-Rank test over the entire 23 year period showed that the Gray prediction out performed the naàƒ⮶e benchmark.
Sean,
1 + 1 = 10 is a perfectly valid expession in binary. Though I wouldn’t want to hold a conversation that way.:-)
Re #405
Comment #402 not at all “in line” with #384. You’re bluffing, whereas I’m not.
Fact is a double-standard is more damaging to the ruling orthodoxy than to the skeptic outsiders. So go ahead, count away. When you’ve caught up to me, let’s post our files back to back.
Re #402
If I write a purely analytical paper that makes no predictions but instead shows only a correlation, then anyone who duplicates my work is simply checking my maths, and in my mind anyway, these checks are not supporting any hypothesis I may present. Yet it certainly appears to me that those who are suggesting that errors in MBH9x “don’t matter” because others have duplicated the work are falling into the trap of believing that such duplication supports the hypothesis.
Re #409
Of course, nobody has yet managed to duplicate the work.
Re: #385
Francois O, I just wanted to make a comment or two about your analysis before this part of the thread is abandoned.
The correlation for wind versus SST has an R^2 (assuming a linear relationship) of 0.50 and of 0.02, respectively, for the periods of 1948 to 1987 and 1988 to 2006. For the entire period this correlation becomes 0.25.
Getting carried away with the Excel capabilities, a second order polynomial yields a relationship over the entire period with R^2 = 0.35.
My question is: Does this analysis really go anywhere without some a priori indication that a change in data handling occurred near the 1988 time period. Cannot one slice and dice these scatter plots to show all kinds of changing relationships with time? Look at the periods from 1965 to 2006 and 1948 to 1964 for the relationship in question. R^2 = 0.07 for the later longer period with a slight upward slope (left to right) while the earlier shorter period yields R^2 = 0.14 with steep downward slope .
#411 Ken, the way I found this is, first, visually. The SST vs Wind graph had what seemed to be a linear trend. After plotting the trend line with Excel, it was obvious that there was a bunch of data points in the upper part of the graph, that didn’t seem to follow the trend. So I asked myself if those points belonged to succesive years, or were just random over the period 1948-2006. So I calculated the R2 vs end year, always started at 1948, and found that R2, apart from the initial drop due to too few data points, started increasing, but reached a maximum at 1987. Plotting the post-1987 points with a different color (red), lo and behold!, the outlier data points were all red! Then I plotted SST and Wind vs year, and sawclearly saw the two different trends in Wind, but the same trend in SST.
So the best correlation is for the period 1948-87. After that, it totally breaks down. If you calculate R2, say, from start-year until 2006, you always get a low R2 because the 87-06 lack of correlation screws the R2 anyway.
Now, I don’t know why! I looked at other boxes but there is not the same effect. Maybe it’s a real, i.e. physical, effect. Please note that I have a lot of experience at looking at data and trying to find significant patterns. I remember waking up in the middle of the night with a flash of how to fit my data. Just part of my training. Sometimes it means something, sometimes not.
I made my observation in hopes of keeping you from being accused of “statistical special pleading”, although even a reasonable a priori will not always save you from that.
Seriously, I find by plotting the entire period from 1948 to 2006 for wind amplitude versus year a gently decreasing trend is seen with R^2 equal to 0.13. It is not something that hits you square in the head like when looking at cyclone activity over time periods where you see a step function transition from 1970 to 1980 and then a leveling off.
If you plot wind amplitude by year for other boxes, what kinds of trends do you see?
link to Rocky Mountain News
Re # 414 It is a beautiful,unlosable argument: If the trend is up it is most definitely global warming , if it is down it is natural variability.
El Nino did not play a significant role in 2006 until late September. El Nino does not explain the quiet June, July, August or the above-average September.
Also, this 2006 El Nino, so far, is weak.
Trenberth should have blamed Sahel dust or, more truthfully, say “we just don’t know why”.
Speaking of “we just don’t know why”, here is a chart from a state-of-the-art predictor of hurricane formation. It takes into account key large-scale factors (wind shear, humidity, vorticity, instability, etc) that affect storm formation and gives an indication of the chances that a tropical storm will form in the next 24 hours. This model combines the best knowledge of the tropical Atlantic atmosphere into a single model.
The linked chart is for 2006. The higher the wiggly line, the riper the atmosphere is for storm formation. The solid line is “climatology”, which indicates a normal season.
As can be seen, the state-of-the-art indicator consistently said that 2006 would be a busy season. It blew it.
Recognize that this is not some pre-season predictor, but rather a real-time predictor, forecasting only 24 hours out.
The truth is that 2006 was average somewhat because (1) of dust and somewhat because (2) of El Nino but mostly (3) for reasons that we just don’t fully grasp yet.
Re #416 – **The truth is that 2006 was average somewhat because (1) of dust and somewhat because (2) of El Nino but mostly (3) for reasons that we just don’t fully grasp yet. **
Was the upper shear greater in the earlier months?
The reason you don’t know why is there is no discussion of the complexity of electrics. This is criminal in this day and age.
RE #417
Here is a plot of tropical Atlantic wind shear for 2006. Wind shear was mostly normal in the early summer. Several other Atlantic regions, like the Western Caribbean, were somewhat below-normal (= favorable for storm formation).
Here is the link to the website that provides these charts, as well as many others on current and historical Atlantic conditions.
The Western Caribbean was inactive the year following Gilbert just like it was this year following Wilma.
Both storms were low impedence events and changed the magnetics of the earth.
A little weather info: wind velocity data is now available from satellites. This is called “Quikscat”. It estimates wind speed and direction over the oceans, based on how wind affects the ocean surface. I believe it bounces radar waves off the ocean, and the greater the scatter, the higher the wind. It is not perfect, and can be fooled by raindrops, but it’s still a neat tool for estimating storm winds.
Here is a recent Quikscat image for a typhoon in the Western Pacific. You can see the swirl of high winds at the typhoon center.
I was one Talkin Tropics last night with Mike Watkins discussing electrics and cloud behaviors. We talked about this thread on the show and Judith Curry’s view of William Gray. Watkins is not a fan of Curry and there were some more interesting fireworks going on. The broadcast is online at podweather for the November 2, 2006 show.
Mike Doran
This interview with Bill Gray is worth a read. The interesting part are his comments about funding, and how his funding dried up back in the 1990s. Gray seems to attribute that reduced funding to his publicly-skeptical view of AGW.
http://www.podweather.com/index.php?id=1599
Here’s the link to the Talkin Tropics Show. I am 20 minutes in. Gray’s problem is that he hasn’t articulated why tropical storms should be a part of the debate–and they should be, because tropical storms are the low impedence event on earth. And because tropical storms are organized electrically by decarbonation of the oceans. That is why they become disorganized over land-land is 5,000 more resistive than the oceans which contain the organizing carbonation. Judith Curry’s problem is that the ‘iris’ paper and then the papers that show a positive relationship to cloud wieghted SSTs can ALSO be explained by decarbonation–specifically the SOI index. All heat and no light because both miss the complexity of electrics.
I’ve been trying to make sense of Bill Gray’s thermohaline (THC) hypothesis. In a nutshell, what it says is that changes in the strength of the THC account for a major portion of the observed global warming of the last hundred years. Not all the warming, but the major part.
Now, I have had a hard time figuring out the details of Gray’s hypothesis, in part because what I’ve read is generalized and in part because his hypothesis seems to vary over time. I have tried to piece it together.
First, here is a nice writeup on the THC, for reference.
The THC involves salty, warm surface water flowing to the vicinty of Greenland, where it cools and sinks. It then travels along the bottom of the ocean to the Southern Hemisphere, where it rises. It then travels along the surface back to the Atlantic and Greenland. A round-trip is thought to take about a thousand years.
Gray argues that the THC oscillates: sometimes the flow is strong , sometimes it is weak. If it is strong, then there is a lot of cold, upwelling water in the Southern Hemisphere, which over time flows northward and cools the ocean surfaces, cooling the Earth. If it is weak, then there is less upwelling and so the oceans, and the Earth, warms.
(As best as I can guess, Gray also suggests that, when the THC is strong, there is greater net heat loss, because more tropical water has flowed to northern latitudes, where it can radiate away heat easier than it could in the tropics. I think this. or something like this, is necessary to maintain an oscillation.)
How can one tell if the THC has oscillated in the 20’th Century? Well, there is a famous 2005 paper which measured Atlantic flows and said the THC has weakened. (It got a lot of publicity, out of fear that the THC might be collapsing and making Europe colder.) Then there are two 2006 papers which measured Atlantic flows and found no change in the THC. (Those papers got little publicity.) The fact appears to be that the ocean flow measurement techniques have too much error to allow reliable calculation of the THC flow.
An alternate is to look at sea surface temperature (SST) patterns. If the THC is strong, then there will be extra warm water near Greenland (where the water sinks) and extra cool water in the cooler parts of the Southern Hemisphere oceans, especially south of Africa. If the THC is weak, then Greenland waters are cooler than normal and the Southern hemisphere waters are warmer than normal.
What do the SST patterns show, with regards to Gray’s statements?
Gray says that the THC weakened from 1930 to 1975. That would mean that Greenland waters should show cooling, due to a longer residence time near Greenland, and Southern Hemisphere high-latitude waters should show warming, due to reduced upwelling.
Map One is the SST change map, comparing 1970-1975 average SST with 1928-1933 average SST. ( The map is from GISS. which some refer to as the House of Hansen, I believe, so it should be credible data.) It shows a clear drop in SST in the downwelling region near Greenland and warmer SST in the high-latitude Southern Hemisphere. It looks consistent with Gray’s hypothesis.
From 1975 to 1992, the weakening continued, per Gray. Map two is the map for 1987-1992 vs 1970-1975. It shows that the THC weakening trend has slowed, with little temperature change in the key regions. (As an aside, note the cool bullseye in the North Central Pacific, versus the warm bullseye in the 1930-1975 chart. I believe that’s a fingerprint of the PDO.)
From 1992 to the present, Gray says we are in a strengthening THC. That should mean warmer SST in the Greenland downwelling area and cooler SST in the high-latitude Southern Hemisphere. Map Three is the map for 2000-2005 versus 1987-1992. It shows definite warming in the downwelling (Greenland) regions and weak cooling in the upwelling regions. This is consistent with Gray’s hypothesis.
One of Gray’s beliefs is that there is a lag, for some reason, in the upwelling, and it takes a decade or longer for the cooler upwelled waters to have a strong influence on global temperatures. Map Four is a map comparing the most-recent years (2004,2005) with 1987-1992. Note that the Southern Hemisphere cooling and upwelling is becoming more evident.
(Footnote: Another noticeable thing about the maps is the temperature of the tropical Atlantic. Maps One and Two, weak THCs, shows coolness in the tropical Atlantic and warmth near the US. To me, this is consistent with a strong Atlantic “mid-ocean” clockwise circulation (“mid-ocean” round-and-round circulation is not the same as THC circulation). The flow of cool water from the North Atlantic to Spain to the tropics is strong, cooling the tropics. The flow of warm water along the US coast is also stronger, warming it. Compare that to maps Three and Four, which are strengthening THC and show warmth in the tropics and neutral along the US coast, which would be consistent with a weakening mid-ocean circulation.
Perhaps a strong THC brings surface water from the South Atlantic into the North Atlantic, interfering with, and weakening, the mid-ocean circulation pattern, thus bringing less cool North Atlantic water into the tropical Atlantic, thus warming the tropical Atlantic. My conjecture.)
In summary, what I see is that Gray’s thoughts on THC behavior in the 20’th Century have some amount of evidence backing them. The key thing missing is strong, cool SH upwelling in the present time but, per Gray, that is “about to begin”. “The check is in the mail.” I will watch.
I neither agree with nor dismiss what Gray has proposed, due in part to me not fully understanding what he is hypothesizing. Also, ocean flows are very complex and I have nothing more than a layman’s understanding of them.
I’ll have some more comments on other aspects of Gray later.
Re # 425
Thanks, David S. for an excellent review of THC. I appreciate your efforts to present factual information at this blog with your own added insights and analyses.
David, you say that “A round-trip [of the THC] is thought to take about a thousand years.”
Since a round trip is something on the order of 20,000 miles, this is about 20 miles a year. As this is about 0.0002 miles per hour … we can understand why observations have large error estimates.
Also, it is interesting to note that the thermohaline circulation is not driven by thermo or haline, but by wind and tides …
w.
Part 2 on Gray’s global warming hypothesis. This is a look at the last 15 years, which appear to be a period of strengthening THC and global warming.
According to Gray, as best as I can tell, the strengthening THC initially creates a period of global warming, followed about 10 to 15 years later by global cooling.
The global (mostly Northern Hemisphere) warming is the result of anomalously warm water flowing into the far North Atlantic, warming the Arctic and then entire Northern Hemisphere temperatures.
The global cooling, which comes later, is the result of cold water upwelling in the traditional THC ocean areas south of Africa and South America. That cold water then spreads northward through the Atlantic, cooling it.
(Implicit in this is that there is a decadal lag between increased downwelling and increased upwelling, which seems odd to me. But, maybe there are density aspects of the oceans which elude me.)
I am also conjecturing that Gray believes that a strong THC surface flow from the South Atlantic into the North Atlantic interferes with, and weakens, the North Atlantic Subtropical Gyre, which I’ll call the “gyre”. This is a clockwise current around the Atlantic. It brings cool water southward by Spain and into the eastern tropical Atlantic, cooling the tropical waters. It also brings warm water from the western tropical Atlantic northward by the US and into the north Atlantic. This gyre includes the famous Gulf Stream but, importantly, it is not the same thing as the THC. The Gulf Stream splits into two, with 80% continuing in the gyre and 20% flowing to Greenland as part of the THC. I hope that makes sense.
Here goes. I will use Hadley SST and I choose 1991 as my base year. The year 1991 was about the start of the strengthening THC. Hadley is readily available and reputable, though I have found significant differences among various SST sets.
Here is 1995 SST compared to 1991. Waters have warmed near Greenland, a downwelling region, indicating increased flow of warm water into that region and consistent with a mildly strengthening THC. Waters have warmed in the eastern Atlantic and cooled near the US coast, indicative of a weakening gyre. (Ultimately, a weakened gyre warms the tropical Atlantic and (possibly) shifts the ITCZ northward, both of which favor hurricanes.)
Here is 1998. It shows a definite warming in the downwelling regions near Greenland, indicative of stronher THC. It shows growing warmth near Spain and in the tropics, and cooling along the US coast, indicative of a weakening gyre. there is no clear sign of cold upwelling south of Africa.
Here is 2002. It shows even more warming of the downwelling regions, indicative of a strong THC. The gyre still appears weakened, though not as much as 1998. The ocean south of Africa and South America shows little evidence of enhanced cold upwelling.
Here is 2005. There is expansive warming of the downwelling, and adjacent, regions. The gyre continues to show evidence of being weakened. There is a slight hint of increased upwelling south of Africa.
(I’ve looked at SST trend charts from NCEP for these regions, and can say that those charts show definite cooling in the far South Atlantic in recent years, while the GISS charts don’t show as strong a trend.)
Conclusions? There is evidence of a stronger THC and a weaker gyre (which favors hurricanes). But, if enhanced cold upwelling is part of this hypothesis, then nature needs to get going and do some major cold upwelling in the Southern Hemisphere or else the hypothesis crumbles.
Also, Occam’s Razor would not like this hypothesis, though maybe climate science is just too hairy to be shaved.
RE: #428 – Sea ice was lacking in the N. Atlantic and Barents Sea in winter 2005 – 2006. Look at it thus far this year. Much more extent than there was the same time last year. If the hypothesis you’ve rendered is true, and we have reached the peak, then this is precisely what I would expect.
David, I like this. Is that your own line? Go Tigers.
It’s mine, Shiner Bock-induced.
I think the Gators will make it to Glendale January 8. And, they’re preseason #1 in basketball. Heavens, leave at least one sport for the rest of the country to have a chance of winning a championship. Women’s softball? Synchronized swimming?
David, your analysis of Gray seems well thought-out. The recent Lyman paper,
http://climatesci.atmos.colostate.edu/2006/09/29/the-lyman-et-al-paper-recent-cooling-in-the-upper-ocean-has-been-published/
shows recent cooling in the SH oceans (this includes water temps to some depth, not just ocean surface temps). Whether that could be attributed to increased upwelling is a question.
Gray and Klotzbach have published their 2007 Atlantic hurricane forecast . They call for an active season (14 storms versus the normal 10 storms).
They’ve changed their basis for this long-range forecast. It looks like they are now focused on whether the SST will be warm, whether El Nino will be around and the fact that we’re in an active phase. It’s a rather vanilla approach but long-range forecast of hurricanes is extremely difficult, unless one uses Hansen’s GCM.
I await Steve M’s forecast, but I think he’s wise enough to wait a while, at least until the Europeans make their forecast. Otherwise they may cut-and-paste (smile).
A British firm has issued its early forecast for 2007 Atlantic hurricanes and Australian typhoons, located here . Looks like sixteen storms (normal is 10) in the Atlantic.
In the odd-weather category today is that buoy 42007, located in the northern Gulf of Mexico about 20 miles (35km) from shore, recorded freezing temperatures this morning. That’s unusual for Gulf surface waters that far from shore.
#433. Aren’t we still waiting for the Europeans to present last year’s “forecast”?
Average of the past ten years is 14.4 named storms.
Re #434 – I had to dig a bit deeper to get the 2006 forecasts. They continued the abvoe normal hurricane forecast into the August, 2006 forecast issue. So in other words, they they blew the 2006 forecast.
Re #433. The European groups (ECMWF, METEO France, UKMO) will go operational with their seasonal hurricane forecasts for the 2007 season. Their forecasts for the past two seasons have been experimental, have been made available only to researchers (and possibly to their paying customers). They have also conducted hindcasts for seasons back to 1993. The results are impressive. The public can judge the forecasts once the papers are published and they go operational next year.
Re #438 Have the European groups publicly discussed their methodology? I am wondering if it is the use of predictors of things like Caribbean mean sea level pressure, or is it the use of models to count computer-generated storms.
The out-of-sample results for forecasts and hindcasts are what impresses me, not the after the fact hindcast or forecast results. My simple statistical analysis of Gray’s forecasts left me less than impressed with the results. Computer model forecasts are of such recent vintage that they will not have sufficient out-of-sample results for making statistically significant conclusions for sometime to come. In the meantime one would hope that what we get is truly out-of-sample results and not some adjusted results or results from an ever changing model.
I have only found this page tonight so forgive me if I appear to have lost the thread. It seems to me, from reading many of the responses, that all ponts have been made and that the chat is now going round in circles.
For me the debate should deliberately eliminate all character assinations and concentrate on facts and in climate research there appears to be very few of them. Most papers I have read on the subject remind me of cookery books where guesses are promulgated as if they are facts and no scientific work is done to verify the old wives tales.
Now, if a climate model or weather model is ever to be taken as fact then surely it must give factual results whether you run it forwards or backwards. Am I being simple here? Develop a model, run it backward in time, if it fits well then you can reasonably use it to predict the future over a finite period. But hey, weather models for all their complexity are rarely correct over more than 3 days are are incorrect often enough over 12 hours to be worrisome
You would have to live in cloud cuckoo land to think that you can model the climate (read planet) over any period greater than ~50 years.
I believe that both parties are wrong (skeptics and GWs) because neither of you have yet proven your assumptions, conclusively.
Never the less there are some greta discussions here so thanks to you all
During the storm season, two things that correlate well with Atlantic storm count are ENSO (El Nino and its opposite, known as La Nina) and mid-Atlantic sea level pressure.
If a year has La Nina and low mid-Atlantic pressure during the storm season, then the chances are good that there will be many storms. If it’s El Nino and high pressure, then few storms.
The problem is how to predict El Nino and Atlantic sea level pressure.
Many agencies have tried to predict El Ninos, but there are limits as to how far in advance they can be predicted (see 2006). Mid-Atlantic pressure is more predictable, especially since it seems to get stuck in multi-year “ruts”.
Also, Mother Nature throws curves in the form of Sahel dust, anomalous upper-level lows, TUTT persistence, etc that can frustrate forecasters.
All in all, I think it is possible today to do what the long-term temperature and precipitation forecasters do: forecast “below-normal”, “normal” and “above-normal” seasons in advance, but accurate forecasts of storm numbers and intensities will be elusive until our knowledge grows, and maybe not even then.
Ken, for fun I used the ESRL correlation-mapper to see if I could spot any high correlations for 1993-2005. I found one. For that period, I found that the sea level pressure in March-May near Tasmania correlates (r=0.8) with subsequent Atlantic storm count. Unfortunately, that correlation completely disappears prior to 1993 and in 2006. Oh well, such are the dangers of correlations based on just 1993-2005.
I am very interested to learn if the Europeans will be using broad indicators (ENSO, SLP, heights, etc) or are using long-range computer map runs. If it is the latter, and they prove to be accurate, then hats-off to their skills. I’ll be quite surprised, though, if that’s the case.
#441 –
Your comment doesn’t make any sense:
“I believe that both parties are wrong (skeptics and GWs) because neither of you have yet proven your assumptions, conclusively.”
The AGW crowd are generally stating that the evidence exists and it either conclusively proves, or the great weight of evidence points very strongly to man-made warming. Politicians like Al Gore make the most cavalier statements; those of scientists seem more measured – if not deliberately slippery.
If you don’t think this has been shown to be true, and are waiting for more evidence to make up your mind, or believe the AGW crowd are excluding other evidence [e.g. – solar], then you are by definition a skeptic.
The skeptics don’t need to prove anything, other than that something is missing in the AGW arguments.
SOME skeptics may have specific competing models for climate change they are promoting, but none of those need to be proven right for specific AGW arguments to be shown to be lacking in merit. If ONE skeptic has a quack theory about conspiracy and anti-matter, that doesn’t pollute other skeptics genuine concern that public policy is being driven by bad science.
Part 2 of a search for natural oscillations which can partly account for the late-20’th Century warming: The Warm Pool.
The Warm Pool is the biggest thing on Earth that you’ve heard next-to-nothing about. It is important to climate but remains obscure. Here is some background:
* Its formal name is the Indo-Pacific Warm Pool (IPWP), and as the name indicates it sits in two oceans. The length stretches nearly 25% of the way around the globe. The area is almost as large as North America.
* The Warm Pool’s heat content is immense. The heat is transferrable via both sensible means and by latent (water vapor) means. Since the water is 28C+ it has a lot of moisture to offer and the temperature is such that convection into the atmosphere is enhanced. Think “steam bath”.
* This presentation describes the Warm Pool as “THE major source of heat for the global atmosphere”. Perhaps that’s a bit of hyperbole, but it’s not far off.
* Surprisingly, the Warm Pool (defined as 28.5C+ SST) can vary dramatically. The Indian Ocean portion of the Warm Pool can double in size, or completely disappear, for decades. The Pacific portion can vary by plus or minus 40%.
* The Warm Pool transfers much of its energy via Hadley-Walker circulations, which can impact climate far from the Warm Pool. By changing global atmospheric flows, it can affect the ocean/atmosphere’s ability to shed heat (my hypothesis).
How has the Warm Pool behaved in recent decades, temperature-wise? Here is a plot of Warm Pool surface temperature for the portion north of the Equator, for the cool months (November-April). Notice the warming since 1976, occurring in a remarkable jump followed by 15-20 years of flatlining (give or take an El Nino and occasional volcano) then another remarkable jump circa 2000. How can SST jump like that, if the cause is radiational forcing?
How has the Northern hemisphere air temperature behaved? I use something called “geopotential height”, which is (my opinion) a good proxy for the warmth of the troposphere. Warming makes the atmosphere expand and the geopotential height rise. Cooling does the opposite. Using this allows me to avoid all of the urban heat effect, satellite drift, sunshine on thermometer mess and allows a good look at pre-satellite periods. And, it matches well (pattern-wise) with the satellite data. Anyway, the plot of Northern Hemisphere 500mb geopotential height is here . Note the remarkable jump circa 1976, then a flat-to-slight increase, then another jump circa 2000. Looks familiar. Maybe there’s a connection between the NH Warm Pool and NH temperature(more to come on that, later).
But what can cause large changes in Warm Pool temperature? There are what I call “highways” of deep water passing beneath the various regions of the Warm Pool. Some deep water flows from the North Pacific through Indonesia and into the Indian Ocean. Some deep water also flows from the southern polar regions (possibly North Atlantic water) beneath the Indian Ocean. There is scientific opinion that these cold deep waters mix with the Warm Pool to varying extents and alter the Warm Pool’s heat content.
Could the warming of the Warm Pool in recent decades be due to less mixing with the cold deep water? Possibly, but why would that be?
Well, I made an ESRL correlation map of the strength of the Warm Pool versus global SST. The model won’t let me show the map, but here are the parameters to plug in to the model (SST,surface,Jan-Dec,1975 to 2005, seasons,Pacific Warm Pool, tropical display). What I get is a correlation (r=-0.5 to -0.7) between the Warm Pool strength and the SST between South Africa and Antarctica. In other words, when there is a lot of cold, upwelling water near Antarctica, the Warm Pool is strong. Perhaps something is “robbing” the Warm Pool of its normal intake of cold deep water and the cold deep water is instead mixing near Antarctica.
What could be causing such a “robbery” of cold water that usually goes to the Warm Pool? I don’t know. It could be wind-driven (stronger Antarctic wind would help mixing and upwelling, and winds indeed have been increasing in recent decades), it could be salinity and density considerations, it could be some combination of factors. Complex, plus data is sparse.
So, is there a plausible cause-and-effect natural oscillation, involving the Warm Pool, that could be impacting global temperatures in recent decades? Maybe. It would be something like: “X” causes cold deep water to increasingly upwell off Antarctica and not in the Indian Ocean, which warms the Warm Pool, which puts greater heat into the (especially Northern Hemisphere) atmosphere which directly heats and puts the atmosphere into a less-efficient heat removal state.
Could this be part of what Dr. Gray is proposing? Possibly.
Would this withstand Occam’s razor? Well, it depends on finding a plausible “X”.
Next: the paper on the IPWP Hadley-Walker cell
RE: Pacific Hadley-Walker Cell
This satellite image is worth a few seconds look, even though it may not make sense at first. The image is of the Pacific Ocean (Asia on the left, North America on the right) and is of atmospheric water vapor.
The water vapor detected by the satellite is mainly in the mid and upper troposphere. White means high water vapor and black means low water vapor.
Notice the high water vapor and white dots (thunderstorms) near the Philippines, associated with the eastern edge of the Warm Pool. The ITCZ is also slightly visible, snaking along the equator.
Of greater interest are the two dark regions near the west coast of Mexico and of Peru. These are the descending regions of the Walker-Hadley circulation, where air is descending. The air is dry, having had its moisture removed when it ascended over the Warm Pool and ITCZ. It has to “come to earth” somewhere, and that somewhere is in the eastern regions of the subtropical oceans.
Why is this important to climate discussions? Well, there is a recent study which notes that IR escapes more readily in these dry, descending regions of the atmosphere. Could it be that a warmer earth results in more ascending moist air (= more clouds and tropical rains) which descends elsewhere in the tropics, creating a larger area of good-IR heat loss? The cloud cover may be radiation-neutral (neither cooling nor warming the atmosphere) but the greater descending region may be atmosphere’s way of relieving the excess heat. Sort of like the “iris effect” someone had proposed several years ago.
Also FYI on satellite images, here is the current water vapor image of the Atlantic. This is the “live” image, so it will change from the time I type this.. If you look at it, you should see clumps and rivers of white, moist air surrounded by regions of brown or black-colored dry air. We temd to think of the tropics as a giant region of moist air but in reality the tropical atmosphere is lumpy. Tropical storms and hurricanes formed in the white lumpy regions. The brown and black regions are typically dry, sinking air and/or dry air from the Saharan Desert.
To us stormheads these types of images are almost like watching a ballet as the regions flow and snake and shrink and grow across the atmosphere. (Slight mental illness, yes.)
Re: #442
While those correlation mappers can present some valuable and fascinating pictures, they are ripe for data snooping. I used to read a web site where one of the talented participants put together a rather sophisticated program that could evaluate extended screening methods going back into history for long times to pick groups of stocks and evaluating their performances — but, of course, for periods that were almost entirely in-sample. Some of the screens that were developed had phenomenal performances (historically, not out-of-sample) and with a gaggle of true believers who were convinced that those performances should continue out-of-sample. And of course the 1 of 20 or so that did perform well out-of-sample (as predicted by pure chance) were well publicized and the losers soon forgotten.
David, thanks for your interesting comments on the Pacific Warm Pool above. I do not like to use short time series such as you have used, as they lack historical perspective. Here is a longer term look at the November-April Warm Pool temperatures, from the HadSST2 sea surface temperature database, for your contemplation:
Note that the warming in the Warm Pool (0.6°C) is approximately the same as the global warming during that period, and less than the NH warming.
w.
Re #447 Willis, thanks for the graph, and is there a link to a site that can generate plots like yours? The site I use only goes back to 1950.
The northern portion of the Warm Pool, post 1975, is my focus. There is evidence that changes in the Indo-Pacific Hadley circulation in the mid-1970s altered where the Hadley sent its heat, especially in December-February. More wintertime Hadley outflow goes northward post-1975, which has altered NH winter circulation patterns. (This is per Quan et al (2004), which I plan to write about later.) So, pre-1975 may be a different “climate regime”.
David, thanks for your reply. You say:
I made the graph myself, extracting the data from the HadSST2 sea surface temperature dataset. I have posted the relevant Warm Pool monthly data here for your use.
Best regards,
w.
Re #449 Much appreciated, thanks. You may want to check your longitude sign.
A general note on the Warm Pool and the THC: it may seem like where the cold Deep Water surfaces makes little difference, but it does, especially in the tropics. A cooling of the Warm Pool by 1C makes a significant difference in the vapor pressure of water and in the ease of convecting that moisture upwards (= fewer thunderstorms and a weaker Hadley/Walker circulation). In the subpolar regions, the impact is much less.
David, I’m using +West, -East as my sign convention.
w.
RE #451 Got it!
A news release from several years is given here . This is from NASA and concerns the Pacific Hadley-Walker cell. It is easy reading.
What was found is that the atmosphere has been shedding increased amounts of IR to space in the downflow regions of the Hadley-Walker cell. (There is a nice drawing in the article that shows the cell flow, with descent in the eastern Pacific.) This IR increase amounts to 4 W/m2 over the entire tropics, about 30% of the globe’s surface, and is obviously much higher in the area of descent.
At the same time, they detected a nearly-offsetting decrease in sunlight being reflected, meaning that more sunlight was making it to the earth’s surface and being absorbed.
What they decided, after examining atmospheric data, is that the Pacific Hadley circulation has been increasing on average in recent decades (not a surprise, given the tropical SST increase) and that the downflow regions now have lower humidity and, presumably, cover larger areas. This equals greater IR loss to space.
At the same time, there is less cloudiness, which allows more sunlight to reach Earth.
Now, I see no obvious reason that the two values (higher IR, higher surface absorbtion) have to offset one another. Clouds in the downdraft regions are typically low ones, like stratocumulus, that operate somewhat independently of what is happening in the air above them. Regional climate shifts, resulting in greater or fewer clouds, could affect the net heat loss (or gain) in these key regions. This is not a trivial amount of IR. Interesting stuff.
453: David, did you notice how poorly the models fared in that analysis? I wonder if the models have been “tweaked” to incorporate this information.
Re #454
I saw that, jae. Frankly, I’ve gotten to where I ignore much of the model talk. Maybe some day it’ll be better.
#455 — Reporters and gravely-concerned-press-conference-giving NCAR scientists don’t ignore model talk, however, such as here just today. No doubt, there’ll be pleasure cruises to the North Pole by 2040, and the polar bears will be either extinct or else look like gigantic naked mole-rats. Sunburnt ones.
Here is an article by Jeff Masters on why the 2006 Atlantic hurricane forecasts failed (except of course for Steve M’s, who rose to the rank of “Storm Sage of the Seven Seas” with his August persistence forecast).
Masters offers possible explanations but you may sense some uneasiness on his part. In both the dust explanation and the El Nino explanation he says that there are things we just don’t yet understand, and he’s right. The mechanisms are poorly-understood.
RE: #456 – If I am still alive in 2040, I’ll have to head up north to go take photos of all the idiots coming up there. I am willing to bet that after the first few waves of them experience the typical weather of the Arctic coast “summer” and get in a few scrapes with Polar Bears, they will never want to return. Would be really funny if some of them got iced in in mid July or early August (which can happen, given the right synoptic pattern). My bet is, 2040 and there will still be a very harsh, truly polar climate in the Arctic.
Further on Hadley cells, here’s a satellite image of the current (12 Dec) water vapor in the atmosphere.
* The intermittent white line (dots and clumps) near the equator is the “ITCZ” (Inter-Tropical convergence Zone). The white dots and clumps in the ITCZ are thunderstorms.
* The dark regions north and south of the equator are the dry downflow regions of the Hadley circulation.
* The white, wet upflow regions tend to be in the warm western parts of the oceans, like the Warm Pool, while the dark, dry downflow regions tend to be in the cooler eastern parts of the oceans.
* Here is the circuit: warm, wet air rises in the western ocean, where it rains, removing excess moisture. This drier air then moves eastward at high altitude to the eastern ocean, where it cools and sinks. The sunken air then completes the circuit by flowing westward at the surface, forming a thing called a “trade wind”.
* The trade wind humidies and warms as it flows back to the western ocean, where it rises and rains, completing the circuit.
The complete circuit is called the Hadley-Walker circulation.
A 2004 study (Quan et al) found that, circa 1976, the Hadley outflow shifted, such that more outflow now goes into the Northern Hemisphere (NH), especially in winter and during El Ninos.
Now, the dark areas where the air sinks has a low humidity, so that IR radiation is enhanced (good for removing heat from Earth). But, this dry air may also reduce cloud cover nearby (probably bad for removing heat from earth, especially the loss of low clouds). A change in where the air sinks can affect heat loss from earth, depending on how it affects nearby cloud cover.
So, when more outflow shifted into the NH in 1976, was the effect on clouds near the dry downflow NH region the same as the effect on the SH regions? Maybe, maybe not. Recall that the NH has a lot more land in the downflow regions than does the SH. Land clouds may not be the same as ocean clouds.
And, when the greater Hadley outflow to the NH began in 1976, did it affect the strength/reliability of the Pacific trade winds such that El Ninos became more common? Maybe.
And did the greater NH outflow speed up the NH winter winds, making it more difficult for frigid air to plunge southward? There is evidence of this. My hypothesis is that maximum heat removal occurs when maximum air mixing occurs, so a net reduction in cold/warm air mixing means that there is less heat removed (= higher net temperatures).
There is no punchline, only questions. This issues are open, the science is not settled. It’s important stuff, because it affects our knowledge of just how much GW is due to A.
The WMO press release from 11 December 2006 states:
The full report is a PDF at link.
A pretty good atmospheric energy balance for the Pacific Warm Pool is given here . It’s a moderate-difficulty read.
This section reviews the overal energy flow:
And, Figure 7 (near the bottom of the article) shows a nice schematic of the energy flow.
Several things of interest:
1. 1. The oceanic energy balance shows a net 150W/m2 input from radiation, balanced by 120 W/m2 loss via convection/evaporation to the air and 30 W/m2 via mixing with cooler oceanic water. The mixing effect is bigger than I had thought. The Warm Pool itself is a huge energy-processor (heat exchanger).
2. The Hadley-cell atmosphere above the Warm Pool is an even bigger heat exchanger. I had not recognized the amount of energy inflow via the trade winds.
The behavior of the Warm Pool heat exchangers are quite important to global temperature.
One note (not in the article): there is something of a “magic temperature” (28C) in tropical water. As the SST exceeds 28C, evaporation becomes greater and convection (rising air and thunderstorms) becomes easier. A seemingly small change in Warm Pool temperature around 28C can have a sizeable effect on the movement of heat into the atmosphere above it.
RE #461 If I recall correctly, mankind’s to-date CO2 and methane impact (excluding water vapor effects) equates to about +2 W/m2. Yet, for the important Warm Pool, oceanic cooling (mixing with water that was cooled at higher latitudes)removes about 30 W/m2.
Suppose that Warm Pool oceanic cooling varies naturally on decadal timeframes. A 5% increase (1.5 W/m2) in cooling via mixing would almost counterbalance the 2 W/m2 CO2 effect, masking it. On the other hand, a 5% reduction ( 1.5 W/m2) in cooling via mixing would add to the CO2 impact, raising Warm Pool surface temp well beyond that due to CO2 alone.
Now, the common practice is to attribute all warming since 1976 to man-made CO2 and methane. For that to be true for the important Warm Pool, decadal-scale variation in ocean mixing in the Warm Pool would have to be zero. If ocean mixing has varied in recent decades, it may be masking, or exaggerating, the effect of CO2.
So, has there been variation, and in what direction?
For your consideration, here is a 2002 NOAA news release . It is well-written, including
Equatorial cooling via mixing does vary, including a major change circa mid-70s. Perhaps it’s not 100% CO2 after all.
An easy-to-read paragraph and map of the most important ocean current you’ve (probably) never heard mentioned: Kuroshio Extension Current .
Size matters:
Atlantic Gulf Stream flow (current famous for its impact on Europe): 30 Sv (1 Sv = 10 to the 6’th cubic meters/sec)
North Atlanitc Deep Water formation (key part of the well-known thermohaline circulation): 13 Sv
Kuroshio Extension (obscure): 140 Sv
Extension matters:
The Extension (known as “KE” or “KOE”) carries 140 Sv of high-energy water which affects both the Pacific atmosphere and the Pacific ocean thermocline water. The thermocline water affects the subtropical Pacific ocean temperature.
The KOE varies. There is evidence discussed (obscurely) that KOE variation plays a major role in the phenomena known as the PDO (Pacific Decadal Oscillation). Some scientists are working, quietly, to untangle the relationship between KOE and Pacific (and thus global) climate (see here ).
Steve,
Re your post #47 on Hansen’s recent work. One of the things that I didn’t like about about that paper was that it was refereed by Cicerone, the President of the NAS. Since Hansen is an NAS member, this seems to me like asking a person in Department A of University B to referee a paper by someone else in Department A. There’s a risk of getting a biased report.
Roger Bell