From what I’ve read, these effects are actually pretty well explained by the AGW hypothesis. For one thing, greenhouse-gas temperature forcing has a greater effect over land, of which there’s more in the northern hemisphere. Also, the polar amplification effect doesn’t have to be symmetric between north and south; according to the current models, ocean heat uptake increases more in the Southern Ocean than near the north pole. So we’d expect the southern regions to have warmed less than the northern ones up to now.
As far as I can tell, this and your other objections are fairly standard climate-skeptic debating points that are addressed in the How to Talk to a Climate Skeptic FAQ and the references cited therein.
Regarding your uncited claim that Patrick Michaels was the most cited person
I had missed that, jshore, thanks for pointing it out in your normal gracious fashion. In addition to mis-remembering that it was about CNN and not the media in general, you also didn’t remember that is a single bloggers report of an undocumented claim made about a story that Peter Dykstra is alleged to have told at a meeting in New York … not what I’d call a substantial citation. Other than being picked up from that website by a few other bloggers, there is absolutely no mention of this on the web. To me, the story reeks of urban legend, especially this part:
Me, I fear I simply don’t believe this story as re-retold. It likely has a grain of truth in there somewhere, but where? For example, although Michaels is “right there in town” in DC, CNN in general and Peter Dykstra in particular are “right there” in Atlanta. We also have no indication of how good Dykstra’s search was, or who else he searched for, what the search terms were, or what time period he searched.
In addition, the link you provided approvingly cites another source which make the … umm … unusual … claim that in all of 2005, the total amount of TV time on climate change for all major networks was less than 4 minutes … 4 minutes? This doesn’t give me much confidence in the first blogger’s judgement.
intention: Well, I can’t personally vouch for the accuracy of that story regarding Michaels but Robert McClure, who posted it, is a bona-fide reporter for the Seattle P-I (you can find his byline there) and also a member of the Board of Directors of the Society of Environmental Journalists (SEJ). Here is a blurb about him at the SEJ website, noting among other things that he was a Pulitzer Prize finalist. So, while I wouldn’t bet my life on the story being true, I think it is certainly above the status of “urban legend”.
Thanks, jshore, you are likely right that the fact that Dykstra actually told the story is above the status of “urban legend”. However, the contents of Dykstra’s story (how he searched, what he searched, what he searched for, what time period, etc.) are not only not verified, they are not even mentioned.
And McClure approvingly cites, as I pointed out above, a reference that claims that in all of 2005, the major networks aired a total of 4 minutes of news about climate change, which doesn’t say much for his fact checking.
Unfortunately the supercomputer I would need to run a climate model seems to be missing from my desktop, so I can’t run the models. In the meantime, consider this, from Gavin Schmidt:
So merely by adjusting these few parameters, I can make the Arctic and Antarctic ice thinner or thicker, larger or smaller, more or less drift. And that’s just these few. With all of the available parameters, I could make the elephant wiggle his trunk.
The problem is that replicating the modeled past by using parameters means nothing about whether the model is accurate about the future. If we could predict the future that way, stock market models would be a dime a dozen … but they’re not.
I don’t really understand what you think is wrong with that. ISTM that that’s exactly how the process of developing mathematical models of physical systems is supposed to work: you “adjust the parameters” of the model and see how the changes affect the model’s predictions. When you find parameter values that produce predicted results that agree better with the observed results, then voila, you’ve got a better model.
AFAICT, this is essentially the exact same process by which earlier scientists built up accurate mathematical models of other physical systems, such as celestial mechanics. We started out (after Copernicus) with circular heliocentric orbits for the planets, then modified them to elliptical orbits to produce better agreement with more precise observational data. Then we starting adding on higher-order terms corresponding to more complicated theories of celestial dynamics to produce better agreement with even more precise data. And eventually, after decades and centuries of tweaking the models and adjusting the parameters, we wound up with a theory of celestial mechanics that’s reliable and accurate enough to let us shoot a tiny space probe onto a given spot on a tiny moon across millions of miles of space. Yay, parameter adjustment!
Well, they probably would be if stock markets acted in accordance with mathematically consistent natural laws governing physical systems, the way that celestial mechanics is. In physical systems, past behavior does tend to be a good predictor of future performance.
Yes, climate systems are currently much less well understood scientifically than celestial mechanics, which is now pretty much a solved problem. But it looks like a safe bet that global climate—being part of the macroscopic physical world and involving many physical processes that we do already have a fairly good understanding of—is a better candidate for accurately predictive mathematical modeling than stock markets are.
IMO, your criticisms about climate models are kind of trying to have it both ways at once. When current climate models don’t produce accurate predictions or hindcasts, you say, in effect, “See, the models are crap! That means the AGW hypothesis must be bullshit!” But if the models do reproduce the data accurately, you say “See, the models can be tweaked to give any answer you like! That means the AGW hypothesis must be bullshit!” It’s heads you win, tails the modelers lose.
I’ve read the Grist “How to Talk to a Climate Skeptic” FAQ, and I wasn’t impressed at all. It’s on the same level, or perhaps worse, than RealClimate.
For example, you say, based on the RealClimate article, that “according to the current models, ocean heat uptake increases more in the Southern Ocean than near the north pole” as though that settles the question. However, a recent study by the GISS folks themselves says (emphasis mine):
You seem to think the question of polar amplification is settled by the climate models results for the Southern Ocean, but the modelers themselves say that problems with the models provide inaccurate results with “far-reaching consequences” … funny how neither Grist nor RealClimate got around to mentioning that in their discussions of polar amplification.
Here’s another example, the Grist FAQ answer to the objection that “Models Don’t Take Clouds Into Account”:
Now that all sounds good, it’s very smooth and reassuring … until you look at the figures. The GISSE model , for example, says that the earth has 59% cloud cover, but it also says that in fact, observed cloud cover is 69%. This is a huge error, leading to a miscalculation of downward LW cloud forcing of over 12W/m2 … in a model you seem to trust to reveal the effects of a 4 W/m2 CO2 forcing …
Despite that, their modeled albedo (29.3%) is very close to the measured albedo (29.75%). How can that be, when clouds are the major albedo source? Well, heck, it’s easy … to quote the authors:
In fact, if you read the paper you will find that their saying that albedo is “to some account tuned for” actually means they calculate the surface albedo and just put in a number for cloud albedo.
So despite the soothing lullaby of the Grist “How To Speak To …”, it turns out that the original statement was correct, the climate models don’t consider the clouds, they just tune up the old albedo, let’r rip, and ignore the fact that their cloud figures are wildly incorrect … of course, since the albedo is tuned for rather than calculated, it is impossible for them to detect whether the net feedback effect of clouds is positive or negative. Their statement that the models “take the effects of clouds into account” is nonsense, and the modelers who wrote the models agree.
Short answer? If you believe anything in the Grist article, or from RealClimate, without carefully checking the actual facts and making sure you still have your wallet afterwards, you’re being quite foolish.
I think I see what the communication problem is here: basically, it’s that you think I’m claiming that various climate issues are definitively settled, when in fact there are still serious uncertainties about them.
But that’s not at all what I’m saying. I’m not trying to sweep climate model uncertainties under the rug or pretend that their problems have been fully resolved when they haven’t. I’m just pointing out that the models do address the issues you raise, and that identifying their flaws is not the same thing as convincingly refuting them.
Here I think you’re making the same error of interpretation as above: you assume that the Grist FAQ and realclimate.org are trying to get their general statements about climate models accepted as something definitively proven and not open to criticism. But I don’t think that’s really their aim.
Like it or not, there is no way to talk about climate science issues for a general audience without leaving out lots of details. That doesn’t necessarily mean that one’s argument is disingenuous.
For instance, I’m sure you didn’t mean to be disingenuous when you brought up the issue of observed asymmetry in polar warming while failing to mention that there’s an AGW hypothesis called “polar amplification” that does address that asymmetry. You wouldn’t want me to conclude from that omission that “Willis Eschenbach is dishonest about climate science and if you believe anything he says without carefully checking the actual facts, you’re being quite foolish”.
It’s a good thing to dig into the details, but nobody’s going to be able to fit every relevant detail into every statement. That doesn’t automatically mean that they’re being deliberately dishonest, or deliberately overstating the certainty of their conclusions.
Kimstu, I was responding to a post by jshore, who seemed to think it was impossible to tweak a climate model to give any global temperature trend desired. I was pointing out that, given the number of parameters, it’s not difficult. The price you pay for this tweaking, of course, is that various other model results (clouds, downwelling LW forcing, etc.) get way out of line … but nobody seems to care about that.
Next, you compare climate models with models of the solar system … you were doing pretty well until you said “after decades and centuries of tweaking the models and adjusting the parameters” … my friend, models of the solar system don’t have parameters. They are purely physics based. That’s the point. There’s not a single parameter in the lot, there’s nothing to adjust. Please compare that to the climate models, with dozens and dozens of parameters. If you don’t see the difference, I fear I can’t help you.
You say that the stock market and climate are different, because the climate is run by physical processes. However, I was discussing the fact that past performance is no guarantee of future profits (as they say in the stock market), which is equally true in the climate models. I can, and have, produced simple climate models that can hindcast global temperature trends with amazing accuracy. However, they don’t perform better than average when I include more years of historical data. Past performance is no guarantee of future accuracy.
Part of the reason for this inability to forecast is that our understanding of turbulent systems is in its infancy, and the climate is a turbulent system. Yes, we have the Navier-Stokes equations, but they don’t work well in discrete models. In order to make them work in climate models, you have to assume unrealistically high viscosities … just one more in the list of parameters.
Another part of the problem is that climate is an externally driven system. The main drivers seem to be solar flux, solar magnetism, cosmic rays, and perhaps interstellar dust. These are not inherently predictable.
The main problem with the climate models, however, is our lack of knowledge of the internal dynamics of the system we are modeling. There are dozens of forcings and feedbacks, both known and unknown, operating at different time and distance scales, both internal to and between the 5 major climate subsystems (atmosphere, ocean, lithosphere, cryosphere, and biosphere). We should not be surprised that (as I pointed out above) the models end up with answers that are at great variance with reality — even the IPCC says our scientific understanding of 9 of the 12 forcings they discuss is either “Low” or “Very Low”. Given the paucity of our understanding of even the known forcings, such inaccuracy is to be expected.
What is surprising is that, given the known problems with the models, and given the very large number of parameters, and given our low level of understanding of the system we are modeling, you keep citing their results as if they actually meant something. Grab a double dose of humility, Kimstu, we are nowhere near that point yet.
Sorry, but this doesn’t make sense. What do you call a “quantity in a mathematical model that is empirically adjusted to improve the model’s predictive accuracy”, if not a parameter? And the development of celestial mechanics certainly did involve decades and centuries of such adjustments. True, there’s little or nothing to adjust in the theory of celestial mechanics nowadays, because its details are so well understood, but that wasn’t always true. Parameter adjustment is a crucial part of the process of transforming poorly-understood models into solid theories, which was my point.
Are you sure that this isn’t an outdated or inaccurate factoid? Because AFAICT from the new IPCC report released this month, of the nine major radiative forcing components discussed there, only four have a LOSU (“level of scientific understanding”) rating of “Low”. Two have a LOSU of “High”, and the remaining three rate “Med” or “Med-Low” (see figure SPM-2 on p. 4 of the link). Admittedly, there’s still a lot of work to be done there, but those LOSU levels seem to be nowhere near as bad as you claim.
Well, to be frank, ISTM that the professional climate scientists’ assessment of the meaningfulness of their conclusions is probably worth more than yours. You seem like a perfectly nice guy, but you’re not a professional scientist, much less a climate scientist. And judging from the misinterpretations and errors of fact that jshore and others have detected in your arguments on these boards from time to time (hey, no shame there, we all make mistakes), you’re certainly not infallible.
So when you claim that the (admittedly rough and imperfect) theories accepted as a promising working hypothesis by thousands of active researchers in climate science are merely a big bag of uncertainty that means nothing at all, why should I believe you? I don’t tend to put much stock in conspiracy theories or contrarian ideas about “The Establishment” ruthlessly suppressing the Lone Voices of Truth. I think that if current climate science theories were in fact as worthless as you claim, a majority of professional climate scientists would probably have noticed it themselves by now. They wouldn’t need a retired construction manager, no matter how feisty and articulate, to point it out to them.
Now, of course I could be wrong about that. It’s certainly possible that you might actually be, not just a run-of-the-mill contrarian with a bee in his bonnet but a few valid points nonetheless, but rather a true misunderstood outsider-genius who’s been able to see an overwhelmingly important truth that undermines the whole structure of the currently-accepted scientific conclusions but which the vast majority of professional scientists have overlooked. It’s possible. But statistically speaking, run-of-the-mill contrarians with bees in their bonnets are way more plentiful than brilliant misunderstood outsider-geniuses. So, no offense intended, but I’m betting by the form book on this one.
I notice that you didn’t touch the total misrepresentation of the model treatment of cloud effects in the Grist FAQ. It’s hard to believe that is accidental, or just a “simplification” as you claim. The models ignore the clouds’ calculated albedo and simply tune the albedo, and their other cloud calculations are abysmal. Sweeping that under the rug is clearly deliberate, regardless of the motive. Clouds are definitely not accounted for in the climate models as they claim.
In any case, it’s not just foolish to blindly trust Grist FAQ or RealClimate, it’s foolish to blindly trust anyone’s claims. I believe very little of what I read, regardless of the source, until I go and read the underlying scientific papers or source documents, do the calculations myself, and make my own judgement.
On this list, however, it is far too common for people to just swallow false claims, like the idea that Pat Michaels gets more media time than any other climate scientist, without the slightest question. Me, I’m not just skeptical about the climate models … I’m skeptical about everything …
What misrepresentation? ISTM that they were perfectly frank about the cloud effects having lots of uncertainties and being hard to model:
They made no specific claims whatsoever about the mechanisms used to model cloud albedo, which is what you seem to be upset about.
Don’t get me wrong, I think it’s very important to look at details carefully and to point out places where popular presentations of the issues are scientifically incomplete. But I don’t see how it’s outright “misrepresentation” when a FAQ answer doesn’t get into all the details of the model. Remember, the answer in the FAQ is chiefly addressed to the climate-skeptic claim that clouds will actually stop warming and AGW theories somehow haven’t noticed that. The point is to make it clear that cloud effects are incorporated into the AGW models, even if admittedly in an inconclusive and unsatisfactory way.
Well, I’m happy to have been able to help you with your skepticism about the false claim that the IPCC estimates nine out of twelve major climate forcings as having a level of scientific uncertainty of “Low” or “Very Low”. You see, we all make mistakes from time to time and accidentally “swallow” claims that aren’t true. So far, I haven’t noticed that your self-proclaimed extreme skepticism has made you significantly more accurate or invulnerable to error than the rest of us average well-intentioned ignorance-fighters.
Are we talking about the same FAQ? The one where they say “All of the atmospheric global climate models used for the kind of climate projections synthesized by the IPCC take the effects of clouds into account,” when in fact they totally ignore the cloud albedo, one of the largest cloud effects, and replace it with a parameter? How is the cloud albedo “incorporated into the AGW models, even if admittedly in an inconclusive and unsatisfactory way”, when it is totally ignored and replaced with a parameter. It is not incorporated in any way, shape, or form, it is ignored.
Are we talking about the one where they say “There is no indication anywhere that any kind of cloud processes will stop greenhouse-gas-driven warming …”, when the models are incapable of providing this kind of indication because they ignore the albedo? Of course there is no indication, their representations of the clouds are junk, how could there be an indication?
Perhaps you consider that they have adequately described the situation by ignoring it … me I like a bit more truth quotient in what I read.
And I’m totally in mystery about how you have "been able to help you with your skepticism about the false claim that the IPCC estimates nine out of twelve major climate forcings as having a level of scientific uncertainty of “Low” or “Very Low”.
According to the IPCC , the greenhouse gases have an LOSU (Level of Scientific Understanding) of “High”, stratospheric ozone LOSU is “Medium”, tropospheric ozone is “Medium”, and the other nine forcings have an LOSU of either “Low” or “Very Low”. How is that a “false claim”, it’s right there on the page? What are you talking about?
“They”? Who’s “they”? I think you may be getting confused between the FAQ which gives a general answer about cloud effects, and the research article you cited (Schmidt et al.) which talks about the actual modeling. AFAICT from your link, that articledoes indeed consider cloud effects in its model, as in the following passage:
Could you please point to the exact passage in the article (page and column number) that you’re so unhappy about? At present, you seem to be so focused on one statement about one cloud feature in one article that you’re overlooking the basic fact that yes, climate models do include cloud effects, which is what the FAQ is saying.
Oh, I see, you were looking at the old graph in the 2001 report. Note the link to the 2007 report in my earlier post, where I asked if your information might be outdated.
Kimstu, my apologies, I missed this post entirely. You’re too fast for me.
Perhaps if you could give us an example of a “parameter” in the historical description of the solar system, it might be clearer. Remember that a physical constant like “G”, the force of gravity, is not a parameter.
Then, perhaps you could explain how putting the code into the GISS model that simply restricts the formation of melt pools to two months out of the year will somehow transform the model into a “solid theory”. (GISS code available on request)
Did you actually read the document?
To make things look better, they’ve split the previous one single solitary category with the “High” rating into two categories. Both, of course, have a “High” rating, so right there, we’ve doubled our scientific knowledge … now there’s a trap for true believers, and I do believe you fell in …
What else? Well, they’ve totally eliminated three of the categories previously rated “Low” or “Very Low” (biomass burning, sulphate, and mineral dust) … whoa, that definitely shows that scientific understanding has improved as well.
This is sleight-of-hand designed to impress the credulous, just omit some of the forcings that were Low or Very Low, split the one with a High rating in two, and presto! … instant scientific understanding, just add water …
While in most fields of science I would agree with you, Kimstu, we’ve been lied to and deceived too many times by climate scientists, guys with degrees, professionals, who have hyped meaningless results, made unsustainable claims, used incorrect methods, reviewed their own studies and surprisingly gave them top marks, hidden their data, omitted their error estimates, exaggerated their accuracy, minimized the unknowns, and refused to reveal their data and methods. At this point, anyone who seriously follows the field is very, very careful about what they believe.
They don’t just count the categories in the new report as you do, they pull back the Wizard of Oz’s curtain and see how they have gotten their shiny, new, higher level of scientific understanding. Divide a high category in two so you can count it twice? The IPCC should be ashamed that they do it, although given their history it’s not surprising. And given their history, you should at least be concerned that you took their chicanery at face value.
When have I ever asked you to believe me? Quite the contrary, I have strongly urged you not to believe anyone, me, professional scientists, the IPCC, or anyone. That’s why I try to provide citations for the statements I make.
I am neither a brilliant misunderstood etc., nor a bee-in-the-bonnet etc., nor a lone voice of truth etc. I am a climate researcher who has written three published contributions on climate, two of which were published in peer-reviewed Journals, one of which was Nature magazine. And there of plenty of climate scientists and researchers out here that think like me about climate “science”. But … so what?
Why all this concern about me? I haven’t asked you to believe me, or marry me, or take anything I say on trust. Why this sudden interest whether I’m a contrarian or a genius? What’s next, are you going to send me emails saying you’re worried about the length of my fundamental organ and you have herbs to lengthen it? It’s not about me, Kimstu, that’s an ad hominem argument, and absolutely worthless. Neither my credentials nor the length of my male anatomy has the slightest thing to do with whether I am right or not.
I do not ask you to trust or believe me. Instead, I strongly recommend that you take up a much more skeptical attitude. Read everything that comes before you with the idea that it just might total bovine excrement regardless of the source, you’ll be surprised how often you are right. Plus, it will allow you to avoid being conned into believing that splitting one “High” category into two “High” categories means that scientific understanding has increased …
Remember how the Grist FAQ said there was no evidence of negative cloud feedbacks in the model results? Remember I said that the models couldn’t give evidence of that, because they didn’t include the cloud albedo forcings, and their cloud cover was off by 15% in any case? Remember I mentioned how the models didn’t include a bunch of known and unknown forcings?
And oh, yeah, remember how they have increased the Level of Scientific Understanding by taking sulfates off the list, so they wouldn’t have remind us how low our understanding of sulfates actually is?
Now I know, Kimstu, that you don’t want to believe me when I say our understanding of climate is very poor and that as a result, the models are not very useful. And I applaud that exhibition of healthy skepticism, and could only wish you applied it more widely.
But it turns out there’s a new paper in the Internet that touches on all of the above questions:
What this means is that the hotter it gets, the more clouds and sulfate aerosols get produced over the ocean, which cools the planet down … can you say “negative feedback”? Gosh, I guess the Grist FAQ was right after all … in fact, the models don’t show that negative feedback, just like they said…
w.
PS - Oh, yeah, this paper was published in that noted pro-skeptic rag, the home of run-of-the-mill contrarians with bees in their bonnets … Science Magazine … I post this as a cautionary tale for those of you foolish enough to think that we understand the climate well enough to model it.
Well, I read pages and pages of the Grist FAQ, 'til my head ached from looking up references and verifying their “facts” … having done so, let me invite y’all to do the same, to read the other side of the story, a 10 part series of articles about other folks that don’t believe the revealed wisdom. They start here …
Am I saying that they are all correct, or that you should believe everything they say? You know me better than that … when you find a part you don’t believe, read up on the science underneath … and don’t forget to do the same when you find something you do believe …
The UK is pretty well covered by Sat Nav - although there are some pretty amusing dead ends.
We have the roads pretty well covered.
Now tell me the quickest route from London to Glasgow - in the year 2050
Not a problem, we’ll just get a few nifty computer modellers to ‘predict’ new roads, road closures, bypasses and whether we will actually have wheels on cars in 2050.
The data is all there, we have a complete road map for 2007, and we can model what planners will do and future technological developments.
With enough past data, it just takes the odd model to predict the future.
You see, we all make mistakes from time to time and accidentally “swallow” claims that aren’t true. So far, I haven’t noticed that your self-proclaimed extreme skepticism has made you significantly more accurate or invulnerable to error than the rest of us average well-intentioned ignorance-fighters.