Consistent With Chronicles, Antarctic Edition
January 21st, 2009Posted by: Roger Pielke, Jr.
[Update: In the comments Eric Steig kindly stops by and offers some thoughts, his comment begins: "I have to admit I cringed when guest writer Weart wrote the article on RealClimate, which I didn’t get a chance to read first. . . " Please have a look.]
An new paper is out in Nature that argues that the Antarctic continent has been warming. In an AP news story, two of its authors (one is Michael Mann from the Real Climate blog) argue that this refutes the skeptics and is “consistent with” greenhouse warming:
“Contrarians have sometime grabbed on to this idea that the entire continent of Antarctica is cooling, so how could we be talking about global warming,” said study co-author Michael Mann, director of the Earth System Science Center at Penn State University. “Now we can say: no, it’s not true … It is not bucking the trend.”
The study does not point to man-made climate change as the cause of the Antarctic warming — doing so is a highly intricate scientific process — but a different and smaller study out late last year did make that connection.
“We can’t pin it down, but it certainly is consistent with the influence of greenhouse gases,” said NASA scientist Drew Shindell, another study co-author. Some of the effects also could be natural variability, he said.
Of course, not long ago we learned from Real Climate that a cooling Antarctica was “consistent with” greenhouse warming and thus the skeptics were wrong:
. . . we often hear people remarking that parts of Antarctica are getting colder, and indeed the ice pack in the Southern Ocean around Antarctica has actually been getting bigger. Doesn’t this contradict the calculations that greenhouse gases are warming the globe? Not at all, because a cold Antarctica is just what calculations predict… and have predicted for the past quarter century. . .
. . . computer models have improved by orders of magnitude, but they continue to show that Antarctica cannot be expected to warm up very significantly until long after the rest of the world’s climate is radically changed.
Bottom line: A cold Antarctica and Southern Ocean do not contradict our models of global warming. For a long time the models have predicted just that.
So a warming Antarctica and a cooling Antarctica are both “consistent with” model projections of global warming. Our foray into the tortured logic of “consistent with” in climate science raises the periennel question, what observations of the climate system would be inconsistent with the model predictions?
January 21st, 2009 at 2:28 pm
Ha! I just pipped you with this one, although to be honest, I remembered reading the RC ‘consistent with’ quote on Prometheus a while back.
Here’s my take:
http://climateresearchnews.com/2009/01/hockey-team-plays-in-antarctica/
January 21st, 2009 at 5:13 pm
LOL. It will be fun to see if RC tries to explain this inconsistency. Maybe I can explain it for them: both conclusions rely on GCMs?
January 21st, 2009 at 5:17 pm
Oh, I’m sure they can explain it, post hoc. It’d be nice to have a clear discussion of “consistency” with models _before_ an analysis is done.
For a more thoughtful discussion of consistency, see this gem of a post by Lucia L.
http://rankexploits.com/musings/2009/year-end-trend-comparison-individual-model-runs-2001-2008/
January 22nd, 2009 at 12:59 am
The publication of this paper on Obama’s first full day in office looks like a set up to me.
January 22nd, 2009 at 2:20 am
In the linked RC article, Spencer Weart (a guest contributor) wrote that “indeed the ice pack in the Southern Ocean around Antarctica has actually been getting bigger. ” That’s something else than saying it is cooling. The “consistent with” referred to increased snowfall which was both observed in reality and in the models. In the Nature paper (and news story) it refers to the increased temperature.
Where’s the inconsistency?
January 22nd, 2009 at 2:32 am
I have a real problem with computer models. They can be tuned to produce almost any result one wants, which gives a lie to the notion that climate is a coupled non-linear chaotic system, the most chaotic known to mankind, if consistent results are obtained it suggests that either Chaos Theory is just plain wrong, or that the models are wrong. Also note that warming & cooling being “consistent with models” suggests a Heads I Win Tails You Lose scenario!
January 22nd, 2009 at 7:54 am
bverheggen–
The nature article quotes Steig, and RC team member saying
“East Antarctica, which scientists had long thought to be cooling, is warming slightly when yearly averages are looked at over the past 50 years, said Steig.”
On the one hand, you are correct to note that Weart does not literally say that the continent is known to be cooling. He says denialists point out that the continent is cooling (which, according to Steig’s current quote, was was scientists believed.)
He in no way contradicts the idea the continent is cooling, and lists a bunch of observations that one might think go along with cooling.
One overall impression conveyed by the Feb 2008 RC article was that, at that time, scientists thought the continent was cooling. The other impression and that this cooling was not only consistent with models, but that cooling in the antarctic was not only consistent with models, but just what we’d come to expect based on years of repeated predictions.
Now, I will grant you that it’s possible to parse the Wearths words to mean other things. For example, now that the new paper suggests it’s warming slowly, it’s possible to say that what models predict is *slow* warming.
But, overall, the careful word parsing, which seems to convey one idea when data say “A”, but can be claimed to not mean that when the data say “B”, is practice that will breed distrust among readers.
If over decades models were predicting *very slow* warming, it would have been better for Weart to say that explicitly. If the data were uncertain due to lack of sensors, Wearth should have said that explicitly.
As things stand the various posts at RC gives the appearance that the authors wish to convey an impression that data that say “A” actually supports models, while maintaining plausible deniability their words fostered that impression when the data are suddenly found to say “not A”. This particular practice looks like marketing.
January 22nd, 2009 at 8:24 am
Roger (and commenters),
I have to admit I cringed when guest writer Weart wrote the article on RealClimate, which I didn’t get a chance to read first. I’m not sure what models he was talking about that said Antarctica should be cooling. A review of the literature would show you (see e.g. Shindell and Schmidt in GRL) that models have been predicting warming. It is also true that models have predicted than in the Arctic, and of course that’s what we find.
But there is a larger point here that you seem to be missing. The entire question of whether models are “consistent” with data is a poor question. A good question is in what ways are the models inconsistent with the data. It is in comparing good data with model output that we learn more about what is important in the climate system. In this case, I think what we’ve learned is that influence of ozone depletion on Antarctic climate — which is certainly real, and certainly contributes to cooling — may be less dominant than we thought.
For those of you that are interested, you really should look at the summary of Antarctic models by Connolley and Bracegirdle in GRL, 2007. The 19 models they look at all give different answers. It happens that the model that they consider the best — as judged by whether it compares well with observations — looks pretty much like our results. Yet Connolley and Bracegirdle weren’t using our results. The point is that our results (which are partly statistical) don’t violate the model physics. That’s good.
Keep in mind that when we talk to journalists, they put words in our mouths. I was asked about 20 times this week whether our new data were “consistent” with the models, and not one journalist was willing to take “that’s not a good question” for an answer.
Eric Steig
January 22nd, 2009 at 9:10 am
correction to above: I mean to say:
It is also true that models have predicted LESS WARMING in ANTARCTICA than in the Arctic, and of course that’s what we find.
January 22nd, 2009 at 12:41 pm
Eric-
Thanks much for these comments and clarifications. In particular, I strongly agree with you when you write:
“The entire question of whether models are “consistent” with data is a poor question. A good question is in what ways are the models inconsistent with the data.”
At the same time f I had a dollar for every time a modeler talked about models being “consistent with” data (e.g., see Shindell’s verbatim quote) . . .
January 22nd, 2009 at 2:08 pm
To say, “I have a real problem with computer models. They can be tuned to produce almost any result one wants…” is a bit like saying “I have a real problem with thinking…” It is true in principle that a nonsensical linear model with enough variables can replicate any desired data. However, if this were relevant in practice, climate models would fit the observational data much better than they actually do. In the real world, it’s actually quite hard to get a model to fit data if you impose the additional constraints that variables have sensible interpretations, conserve physical quantities, etc. Increasing size makes it harder to fit data, because you can’t hill-climb to find a good parameter set with a model that takes days to execute. Instead, more of the parameters must be established based on measurements at lower levels of detail.
The ability to replicate the features of some chaotic behavior is also not an automatic disqualification of models. With an instance of the Lorenz model I can easily replicate the general characteristics of a run from some other instance, without knowing the exact parameters used. That is, I can replicate the general shape of the attractor without replicating a specific trajectory. That doesn’t mean the Lorenz system isn’t chaotic, or that my model somehow constitutes a trick. Similarly, the fact that weather is chaotic and measurements are sparse and noisy does preclude a climate model from replicating observations exactly (though no one’s claiming that), but it doesn’t preclude a model from simulating weather with similar descriptive statistics.
January 22nd, 2009 at 9:36 pm
Eric Steig -
I’m surprised that you “cringed” when you saw the article by Weart. Over at the web site you contribute to, someone asked virtually the same question as the one Roger posed here. In the comments section your co-author unloaded on the guy with the spirited invective for which he is justly famous. Apparently your co-author didn’t “cringe” over the Weart piece.
No doubt you regret not issuing a gentle corrective to Mr. Weart at the time and not asking “what models he was talking about”. I have a related question about “what models [people] are talking about. You see, we are in the middle of a tough winter here in Minnesota, and whenever someone makes a wry comment about the need for some global warming, an alert warming disciple in the vicinity will jump on the poor soul and point out that tough winters are “what the models predict”, and that AGW is leading to temperature extremes at any given location. This idea is very widespread. Can you tell me “what models they are talking about”?
January 23rd, 2009 at 5:32 am
Mr. Steig -
if you “cringed” when you read the article at RC, why is there no comment from you on the thread?
January 23rd, 2009 at 8:44 am
In the 1965 movie A Thousand Clowns, Murray and his 12 year old nephew Nick carry on a conversation as George Washington and Thomas Jefferson, claiming to their audience that they sound exactly like the Founding Fathers. Of course, they are mocking their audience, because no one really knows what George Washington and Thomas Jefferson sounded like. While we may have a tremendous amount of anecdotal evidence that George and Tom did not sound exactly like Jason Robards and young Barry Gordon, we can not PROVE that the actors impersonations are inconsistent with the voices of those historical figures.
Climate science today is a lot like claiming to sound just like George Washington. While the scene in the movie is obviously silly, climate scientists want us to talk their similar arguments quite seriously.
“We do not really know that much about historical climate (particularly in Antarctica), but trust us, what we are observing now is consistent with the theory that man-made CO2 changes are the primary driver of global temperature changes. Pay no attention to preindustrial warm periods. They were just regional (much like the current warming, but just ignore that). And pay no attention to mid 20th century cooling. That was just aerosals. Hey, aerosals…that’s the ticket. And pay no attention to the lack of Antarctic warming…Presto Chango…it is warming! I also happen to sound exactly like John Quincy Adams!”
Frankly…I feel like I am being mocked!
January 23rd, 2009 at 10:05 am
[...] Prometheus » Blog Archive » Consistent With Chronicles, Antarctic Edition [...]
January 24th, 2009 at 12:06 pm
All. I should clarify my point. When I said that “I cringed” I don’t mean that I thought there was anything wrong with Spencer’s article. I meant that I thought he wasn’t clear enough that he was referring to the models show a slower warming in Antarctica than e.g. in the Arctic, which was and remains the correct assessment of what the model show. And I suspected that his article would be used in exactly the way Roger Piekle Jr. has used it; to give the impression that scientists are being careless and inconsistent. But as I said above, this is a red herring.
As for why I didn’t make this point at the time, I have a day job. I can’t spend all my time worrying about how blogs on RealClimate may get mis-used and misrepresented by others.
By the way, I cannot help but point out that Roger was obviously poised to make this point of apparent contradiction. Yet he actually failed in this case to find a good example to fit his preconceptions. The quote from Drew Shindell doesn’t actually refer to the models, and whether they are consistent with the data, at all.
Roger: Now that you have strongly agreed with me that the entire question of whether models are “consistent” with data is a poor question, perhaps you can now move on to something more productive?
January 24th, 2009 at 1:20 pm
Eric-
There is no point in getting mad at me for either Weart’s unclear posting or Real Climate or your failure to clarify it at that time.
As far as Shindell speaking on the consistency of your recent findings with models, here is another statement of his on the findings:
“They’re really consistent with the general warming that we get from greenhouse gases. So we now see that warming is taking place on all seven of the Earth’s continents in accord with what models predict as a response to greenhouse gases.”
http://www.abc.net.au/pm/content/2008/s2472499.htm
Seems pretty clear he is talking about the models.
January 24th, 2009 at 2:42 pm
Eric–
It doesn’t appear the post at RC is being misused by Roger. Spencer Weart wrote an article which may not have included any strictly untrue statements but the conveyed a distinctly incorrect impression about what models predicted about temperature trends in Antarctica. That article was, and still is, posted at RC.
Now, you have published a paper, Weart’s article is still published at RC, and still seems to have at least on team member extolling the articles supposed virtues.
Of course this state of affairs is going to give people the impression that there is some newspeak going on at RC.
It’s true that Roger points out the issue in his “consistent with chronicles.” But Roger is just posting ideas other have already formed even before they read his blog. Roger’s posts on this have legs only because what he says resonates as true.
January 25th, 2009 at 12:31 am
[...] the CCTimes will not tell you (because it doesn’t fit their liberal agenda) is that just a couple of years ago we were being told that a cooling Antartica was in fact caused by global warming. So now we have a [...]
January 25th, 2009 at 10:17 am
“It is in comparing good data with model output that we learn more about what is important in the climate system.”
Wait, but didn’t we have actual data showing cooling? Isn’t statistical interpolation inferior to actual measurement?
January 25th, 2009 at 12:58 pm
[...] Consistent With Chronicles, Antarctic Edition [...]
January 25th, 2009 at 2:21 pm
Eric Steig writes, concerning Spencer Weart’s post on Real Climate, “I’m not sure what models he was talking about that said Antarctica should be cooling.”
Eric Steig should read more carefully, lest he be fooled by the sophistry of Real Climate!
The headline for the post is, “Antarctica is Cold? Yeah, We Knew That”
This is typical of the smarmy and bogus writing at Real Climate. Weart conflates “cold” with “cooling.”
The conflation of “cold” and “cooling” is emphasized both by the headline and the conclusion to the post:
“Bottom line: A cold Antarctica and Southern Ocean do not contradict our models of global warming. For a long time the models have predicted just that.”
As several commenters have noted here, it’s interesting (but not surprising) that Eric Steig didn’t submit even 1 of the 449 comments on Weart’s post.
P.S. Here is a comment I made to Real Climate, pointing out the conflation (misrepresentation):
“The headline for this post is, ‘Antarctica is Cold? Yeah, We Knew That.’
I’m curious…why does Spencer Weart make such a blatant misrepresentation of opponents’ arguments? I’m also curious…why did Real Climate publish such a blatant misrepresentation, or not correct it?
Or do you all not think that headline is a misrepresentation?
Based on my understanding of the English language (well, at least the U.S. version), there is a substantial difference between “cold” and “cooling,” (or “cold” and “not warming”). Do you all not think there is a substantial difference between those words? Or can any of you point to anyone who has ever said that Antarctica should not be cold?”
Of course, “Real Climate” didn’t publish the comment. That’s also typical of “Real Climate”.
January 25th, 2009 at 10:01 pm
[...] global warming theory demands that Antarctica cools. But, as Roger Pielke Jr notes, when a new study comes out showing it’s warmed instead, hallelujah: Challenging warming [...]
January 26th, 2009 at 5:05 am
I’m new to this blog, generally reading Climate Audit as my window onto the /magical/ world of climate ideas and fantasies. The background to this thread is really quite an amusing bit of word-twisting. I watch the comments and RC’s wrigglings and smile to myself. What an odd set of ideas they are fixated on. It must be getting increasingly difficult to explain away the sorts of things that are now being acknowledged by more realistic commentators on the scene
Clearly I must put in some time reading Prometheus!
January 26th, 2009 at 6:58 am
re: Mark Bahner January 25th, 2009 at 2:21 pm
And at well South of -50 C, it’s going to be Cold there for a very long time, no matter how long ‘not cooling’ occurs on the remainder of the planet.
January 26th, 2009 at 10:35 am
“re: Mark Bahner January 25th, 2009 at 2:21 pm
And at well South of -50 C, it’s going to be Cold there for a very long time, no matter how long ‘not cooling’ occurs on the remainder of the planet.”
Yes, that’s why I was so appalled by the headline:
“Antarctica is Cold? Yeah, We Knew That”
…it’s a blatant misrepresentation. The critique was that measurements seemed to show the Antarctic was cooling; no “sceptic” claimed that the Antarctic shouldn’t be cold!
And this misrepresentation is repeated in the summary:
“Bottom line: A cold Antarctica and Southern Ocean do not contradict our models of global warming. For a long time the models have predicted just that.”
The models have predicted “A cold Antartica and Southern Ocean…for a long time…”!
“Well, whoopteedo, Edit’!” (As Archie Bunker would say.)
Then, what was orders of magnitude more appalling and disgraceful (but typical for Real Climate) they refused to correct the misrepresentation, or even print my comment about it.
Many people have represented Real Climate as not simply a legitimate climate science blog, but as a very good climate science blog. But their disgraceful behavior on just this one post should be enough to show that they aren’t a good climate science blog.
January 29th, 2009 at 3:04 am
[...] up. A very amusing detail around the critical discussion about that study was pointed out by Roger Pielke Jr. One of the scientists involved in the study came out rather triumphantly declaring: “Contrarians [...]
January 29th, 2009 at 3:06 am
[...] up. A very amusing detail around the critical discussion about that study was pointed out by Roger Pielke Jr. One of the scientists involved in the study came out rather triumphantly declaring: “Contrarians [...]
February 10th, 2009 at 2:24 am
On the Australian Fires:
“The flooding in the northeast and the combustible conditions in the south were consistent with what is forecast as a result of recent shifts in climate patterns linked to rising concentrations of greenhouse gases, said Kevin Trenberth, a scientist at the United States National Center for Atmospheric Research.”
http://www.nytimes.com/2009/02/10/world/asia/10australia.html?hp
February 10th, 2009 at 5:13 am
re: #29
The headline for the article: Australia Police Confirm Arson Role in Wildfires
I think it is safe to say that if the fires had not occurred, the article might not have even been written. There might have been some discussions about the recent combustible conditions, but those very likely would not be as visible.
On a Climate Science issue, aren’t 30 years of data required in order for changes in Climate patterns, in contrast to changes in weather conditions, to be verified? Have the conditions responsible for the combustible conditions been present for 30 years?
The same questions obtain for the flooding; wetter conditions.
February 10th, 2009 at 6:52 am
Oh, I forgot. More Climate Science.
Can anyone point me to the specific peer-reviewed papers in which these changes in conditions in these regions of Australia have been shown to have been predicted/projected/forecast/what-if’d/whatever. Maybe the info is in the IPCC AR4 somewhere.
In the absence of connections between observed conditions and a prior published peer-reviewed papers, do we get to simply pick-n-choose our validation metrics?
Thanks