Gavin Schmidt and Michael Mann
Today we witnessed a rather curious event in the US Senate. Possibly for the first time ever, a chair of a Senate committee, one Senator James Inhofe (R-Oklahoma), invited a science fiction writer to advise the committee (Environment and Public Works), on science facts–in this case, the facts behind climate change. The author in question? None other than our old friend, Michael Crichton whom we’ve had reason to mention before (see here and here). The committee’s ranking member, Senator James Jeffords (I) of Vermont, was clearly not impressed. Joining Crichton on climate change issues was William Gray of hurricane forecasting fame, Richard Benedick (a negotiator on the Montreal Protocol on ozone-depleting chemicals), and David Sandalow (Brookings Institution). As might be expected, we paid a fair bit of attention to the scientific (and not-so-scientific) points made.
Many of the ‘usual suspects’ of half-truths and red herrings were put forth variously by Crichton, Gray, and Inhofe over the course of the hearing:
- the claim that scientists were proclaiming an imminent ice age in the 1970s (no, they weren’t),
- the claim that the 1940s to 1970s cooling in the northern hemisphere disproves global warming (no, it doesn’t),
- the claim that important pieces of the science have not been independently reproduced (yes, they have),
- the claim that global climate models can’t reproduce past climate change (yes, they can)
- the claim that climate can’t be predicted because weather is chaotic (wrong…)
and so on.
We won’t dwell on the testimony that involved us personally since the underlying issues have been discussed and dealt with here before, though we will note that comments from both of us pointing out errors in the testimony were entered into the Senate record by Sen. Barbara Boxer (D-California). Instead, we will focus on the bigger picture.
First, let’s be clear where there is agreement. Climate science doesn’t deal in certainties – it deals in probablities and the balance of evidence. We agree with Crichton’s statement that ‘Prediction is not fact’. That certainly doesn’t mean, however, that projections of possible future climate changes are not meaningful or useful, as Crichton claims.
Crichton seemed to imply that “prediction” (such as that provided by weather or climate models) is useless in the decision making process. (As an aside, we wonder how Gray, who is largely known for prediction of hurricane behavior based on (statistical) modeling, felt about this?). We fundamentally disagree. All science is about observation, understanding and prediction. When those predictions work, you make new predictions. When they don’t, you revisit the observations, attempt to improve your understanding of the underlying processes, and make a new prediction. And so on. In the case of climate models, this is complicated by the fact that the time scales involved need to be long enough to average out the short-term noise, i.e. the chaotic sequences of ‘weather’ events. Luckily, we have past climate changes to test the models against. Even more to the point, successful climate predictions have actually been made in past Senate hearings. The figure at the end of this comment by Jim Hansen demonstrates that projections of global mean climate presented in a 1988 senate hearing (17 years ago) have actually been right on the money
Others panelists attempted to combat the onslaught of disinformation. Sandalow sensibly suggested that the National Academy of Sciences be used to inform the Senate on where the consensus of the science is, and Benedick made some excellent points about how legislation can be successful in the face of scientific controversy and uncertain predictions. However, none of that provided as good theater as the other witnesses.
A highlight of the session was Gray making one particular statement that he may be asked to defend (at least financially): “I’ll take on any scientist in this field …. I predict that in 5 to 8 years the globe will begin to cool” (1:10:00 on the video). This would appear to be a direct call to those “global warmers” (see also here, here and here) who are trying to get contrarians to put their money where their mouths are (with very limited success). We eagerly await developments!
Inhofe ended the hearing by declaring his desire to ‘sit back and look at [this] in a non-scientific way’. We think he already has.
Re: 99
“Even as early as 1991/2 Jim Hansen was able to successfully predict the cooling associated with the Pinatubo eruption.”
You don’t need to be Jim Hansen to predict the short term effect of major volcano eruption on climate. This is completely off-topic.
“He also predicted that this year would come very close to a new record high global mean temperature, and is clearly going to be right. Just lucky guesses, you think?”
This is not exactly what he predicted but it was certainly a daring bet. However it also has nothing to do with the subject of predictability.
Generally, skepticism is a healthy attitude with respect to any science. A skeptic shouldn’t necessarily be expected to produce independent research with the opposite results. Challenging the facts, the statistics, the logic and the assumptions is sufficient.
re 97: “if you have no doubts, you cannot learn”.
actually, i would turn it around: if you believe nothing, then you cannot learn. Also, if you insist
Re #95:
It is important to address this issue of scientific concensus and look at this analogy to ulcers. While it is true that many scientists believed ulcers were due to stress and diet, and this was later proven wrong, as with many early medical beliefs, the believers never had a significant body of evidence or research to support this contention. Climate science does have this body of research. When a researcher presented evidence that the cause was bacteriological, the medical community embraced this evidence quickly (in spite of the WSJ’s implication to the contrary in today’s editorial). On the other hand, the skeptics have not presented any credible evidence that the AGW theory is wrong. They just “don’t believe”. So there is no analogy here.
Concensus is not proof that a theory is correct. But it certainly cannot be used, as skeptics do, the insinuate that a theory is wrong. The logical extension of this line of thinking is to consider every scientific text as “likely to be wrong” because a concensus of scientists agree with it – a ludricruous proposition.
Do GW skeptics believe in theory of relativity? If so, why are they NOT skeptical? And please don’t tell me it is because they’ve read Einstein’s / Rheimann’s math and proven it for themselves and found it to be correct. If they believe in relatively, it is because a *concensus* of scientists continue to assert that it is correct.
In fact, virtually everything that we believe that we know – that the civil war took place, that viruses cause disease, that gravity is proportional to the inverse of the distance squared – we “know” because other people, a concensus of experts, tell us it is so. So if you are going to doubt, as Chrichton does, simply because a concensus of so-called experts asserts that it is so, then what is it that you think you “know”?
Excellent comment Paul Emberger.
D
Re: #98, “The caveat in the end is essential. Of course, IF the statistical process is followed correctly then there is no trouble computing the averages. The problem is that this is not quite the case. Not sure what you meant by weather predictability. What time scales are you talking about?”
Well, weather prediction is much less certain than climate prediction, since even small “butterflies beating their wings in South America” can effect change in short-term atmospheric processes. Climate neutralises these small factors, however, so even, say a 0.02% decrease in CO2 concentration can be easily dealt with in climate models. These climate models are NOT the same as weather models, I might add, which is one of the lies spun by many climate skeptics to try to inject uncertainty into the debate.
Also, as it relates to statistical processes, every study that passes the peer-reviewed test undergoes rigourous examination and analysis to determine whether there are flaws in the study which jeopardise the study’s findings. If the flaw renders the study useless, the study will be rejected. If it passes the test, the study is published.
I have no idea how you could say: “The problem is that this is not quite the case.” This is patently false. You seem to be regurgitating Steven Milloy’s gobbledygook.
“You next paragraph I don’t understand at all. Even the best models routinely make huge errors in forecasting hurricane trajectories, even on short time scales.”
I seem to recall Katrina and Rita’s projections being quite accurate, even over a week in advance. If you checked out Stan’s projection prior to it reaching the Yucatan three or four days ago, you would be impressed on how well the path was forecast.
Why can’t you realise that the science is there? The IPCC is the largest gathering of scientific minds in the history of civilisation, which is remarkable, since a scientist’s challenge is to prove your hypotheses correct and to disprove others’. To have hundreds, even thousands, of scientists agreeing on hundreds and thousands of pages of documents is most definitely unprecedented.
Re: 101
It’s easy to day today that “the believers [of the ‘consensus’ ulcer theory – S.] never had a significant body of evidence or research to support this contention. Read the history about how the dissenters were treated by the majority (Marshall had to infect himself to persuade the stubborn establishment.)- the similarity with the climate debate is striking. Which is not to say the outcome will be similar, of course.
“Climate science does have this body of research.”
Sure but there is a number of unexplained facts. It’s not that I “just” don’t believe. I don’t believe in projections of an incomplete theory which is nothing more than a normal skepticism. Blindly believing in such projections just because the majority of experts say so is quite foolish.
[Response: I’m curious. what are these “unexplained facts”? – William]
re 96: I absolutely do not endorse Rorsch and Thoenes, thank you very much! I got kicked out of the climatesceptics forum because of supporting a critic who attacked this view…..
Gavin can testify, he is stil in…
Re: 103
Steven, it’s as if we’re speaking different languages. Sometimes I have no idea what you are talking about. E.g. “Climate neutralises these small factors, however, so even, say a 0.02% decrease in CO2 concentration can be easily dealt with in climate models.” Whatever … Also, I don’k know who Steven Milloy is. I cannot check Stan’s projections for the same reason. However I do check the hurricane forecasts on weather.com on a regular basis. The projected paths (after landfall) typically vary widely between the models and seldom match the reality more than 2 days forward.
“These climate models are NOT the same as weather models, I might add, which is one of the lies spun by many climate skeptics to try to inject uncertainty into the debate.”
Speaking of lies, would you care to enumerate the differences between climate and weather models?
“I have no idea how you could say: “The problem is that this is not quite the case.” This is patently false.”
You can go ahead and get the patent if you like. If the models are so perfect in following the weather statistics, how come we don’t know whether the next January will be cold or mild?
“Why can’t you realise that the science is there?”
Who told you that I don’t? I do. But I also realize that not all science is there.
I completely agree with comments #1 and #7. Anybody who have read Crichton’s novel “Prey” already knows his stance to science, and it´s not a friendly one.
Re #104 – “Well, weather prediction is much less certain than climate prediction, since even small “butterflies beating their wings in South America” can effect change in short-term atmospheric processes.”
1) Where has this “science” been documented?
2) What is your source?
3) Has this effect on Climate been measured and where?
“Climate neutralises these small factors, however, so even, say a 0.02% decrease in CO2 concentration can be easily dealt with in climate models.”
4) Where can I find this documented?
5? What models “easily” deal with this level of change and what is its effect on temperature change?
6) Or by “easily dealt with” do you mean ignored?
[Response: I’m surprised you’re objecting to this so violently, since its all well accepted. Some of the answers are at my pet blog – William]
Re #101 (Sashka): Please don’t dodge. As you well know the key point regarding Hansen’s Pinatubo prediction is that he got the *amount* right. Regarding this year, I’ll agree that was essentially a weather prediction. On the other hand, it wasn’t exactly a cool year, was it? While we’re on the subject, could you say how many more years of warming will be needed to convince you that the theory is correct? Just curious.
On the role of skeptics, sure it’s important to have them. I’ll even agree with you that any given skeptic doesn’t need to engage in original work to have some credibility. If, however, the collective of all skeptics/contrarians only ever engages in sniping, denial, palpably bad science, etc., *especially when the resources are available for them to do real science*, it’s hard to avoid drawing certain conclusions. As in, they’re probably wrong.
Re: #107, “Speaking of lies, would you care to enumerate the differences between climate and weather models?”
This is easily found on the NOAA pages and at climateprediction.net.
Weather models: http://www.nco.ncep.noaa.gov/pmb/nwprod/analysis/
Climate model explanation: http://climateprediction.net/science/model-intro.php
“You can go ahead and get the patent if you like. If the models are so perfect in following the weather statistics, how come we don’t know whether the next January will be cold or mild?”
Many climate models are generally satisfactory in predicting future climate patterns. That is AVERAGE weather conditions over a set time frame (such as one month) and not a single day’s weather conditions. It is easier to predict a month’s climate conditions for a general location three to six months in advance than it is to predict the weather conditions for a single day three to six months in advance. The climate models prove that, since they take teleconnections, atmospheric composition, etc. into account).
Re: #109, “1) Where has this “science” been documented?
2) What is your source?
3) Has this effect on Climate been measured and where?”
It’s chaos theory. One minute change in the atmospheric regime can slowly effect further changes in the system and the chain reaction can occur, which, in the end, could shift a jet stream a few kilometres and create conditions possible for storm intensification, which may result in the flooding of an area where the storm would not have been or would have been greatly minimised. It’s a bit convoluted, but such complexity is the nature of the atmosphere, though we chould NEVER take any chances with what it might do if even more significant changes occur (such as increasing the CO2 concentration in the atmosphere due to fossil fuel burning).
“4) Where can I find this documented?
5? What models “easily” deal with this level of change and what is its effect on temperature change?
6) Or by “easily dealt with” do you mean ignored?”
Please see the sites above. They may turn you in the right direction.
As for point 6, that is NOT what I meant. These small alterations are taken into account in climate models, with the average of all models (i.e. an ensemble forecast, a term you should know well as a former meteorologist), scientists (like those at the IPCC) can arrive at a sensible estimate of what we are likely to experience in the future.
Re # 111 – “Climate neutralises these small factors, however, so even, say a 0.02% decrease in CO2 concentration can be easily dealt with in climate models.”
You have not answered this. Climate models use a more coarse grid and will not handle fine details. In other words they become smoothed or “ignored”.
Chaos theory does not work well in practice.
Re: #112, they become smoothed, as in taken into consideration in the models. Why can’t you understand that?
Off topic: an absolutely excellent article on the Sierra Club website, an interview with esteemed climatologist Stephen Schneider:
http://www.sierraclub.org/planet/200505/hotornot.asp
Here’s an excerpt, which I found very prevalent:
“The fundamental mechanism of global warmingâ??that gases like carbon dioxide accumulate in the atmosphere and trap heat from the sun, creating a greenhouse effectâ??is widely accepted. Whether human activities such as burning fossil fuels is contributing to the greenhouse effect is where much of the debate lies. But who is stirring up the debate?
Not the thousands of scientists around the globe who have reached a consensus on the significance of global warming and the need to reduce the role of humans in it. Nor the hundreds of governments, from towns to nations, that have signed on to the Kyoto Protocol to limit or reduce greenhouse gas emissions.
A corporation like ExxonMobil may claim that its â??actions include investments and strategic planning that address emissions today, as well as industry-leading research on technologies with the potential to reduce greenhouse gas emissions in the future,â?? according to its corporate Web site. But that is hard to reconcile with evidence like the 1998 memo from an industry-wide project that considers the recruitment of â??a cadre of scientists who share the industry’s views of climate science and to train them in public relations so they can help convince journalists, politicians and the public that the risk of global warming is too uncertain to justify controls on greenhouse gases,â?? as described by the New York Times. Nor does it square with the existence of 40 or more organizations that work to undermine mainstream climate science and have received funding from ExxonMobil, according to Mother Jones magazine.”
Re: 111
Recent temperatures (high overnight lows) and hourly dewpoints have set new record high readings for October in Minnesota and Wisconsin.
Last night, the Twin Cities area had 3 inches of rain in 45 minutes. Overnight, areas in eastern Minnesota and northwest Wisconsin had rain amounts exceeding 6 inches. Some spots had 21 inches or more over the last couple weeks. The high intensity rainfall has led to significant flooding in many local areas in MN and WI. Rivers are rising.
Now that global warming is underway, how do government meteorologists and hydrologists use ensemble forecasting to account for increasing potentials for high intensity rainfall in fall and winter months?
Are the the assumptions used in the probabilistic river forecast products (link below) valid?
Probabilistic River Forecast Products
http://www.crh.noaa.gov/ncrfc/
Re: 114 Correction: Some spots had 21 inches or more over the last couple months (not weeks).
In my backyard (15 miles southwest of Minneapolis) rain gage, I measured large 24 hour amounts of 6.50 inches on Sept. 4 th and 5.30 inches this morning. High winds during part of the rain last night may have resulted in my measured 5.30 inches being too low by about 0.50 inches … lower than the total amount of rain that actually fell near my gage, due to turbulence around the gage. Gage underestimates of precipitation for snowfall events can be low by more than 200 percent for blizzard conditions. I am not aware of any studies on gage catch errors with rainfall events. The error would be less with larger gage openings like the 12 inch precipitation gages used by government agencies.
re 112:
Gerald,
Predicting weather and predicting climate change are two very different sciences. It is not an issue of “documenting” this. Weather, since it is chaotic, is highly sensitive to small changes in initial conditions. Therefore, predicting the weather on September 7, 2226, in Duluth North Dakota is impossible. It would involve looking at the current weather map for the planet and iterating forward, hour by hour, using weather forecasting models, for twenty years. Small errors in hourly forecasts get progressively larger, until, after a couple of weeks, the forecast bears no relation to reality. Predicting the net average temperature for the planet earth in 2226 uses a totally different method. In aggregate, the overall climate in non-chaotic, and non-chaotic systems are far more predictable. We know, for example, where the planet Neptune will be many centuries from now, because we understand the physics of gravity and momentum, and because the solar system is a non-chaotic system. Of course, the physics of climate is far more complex than the simple gravitational interaction of two bodies, so our climate predictions are far less sure than locating the position of Neptune 100 years from now. But the principal is the same. Unlike weather, the if we know the physics, know the inputs and their influences, it seems that we can make pretty good predictions, as has already been fairly well demonstrated (albeit not absolutely conclusively).
re 105:
Sashka,
“Blindly believing in such projections just because the majority of experts say so is quite foolish.”
First, I don’t blindly believe. I have read extensively about climate change and certainly allow for the possibility that some assumptions and / or findings could be wrong. I might note that blindly disregarding experts is every bit as foolish as blindly believing. Isn’t it? An example is the suggestion you made (I think it was you) that perhaps the increase in CO2 is not due to burning of fossil fuels. Well, actually, this one has been proven extremely convincingly, by carbon the carbon isotopes in the co2 in the atmosphere, versus the carbon isotopes in ice-core co2. The increase in CO2 in the atmosphere IS due to burning of fossil fuels. This is fact to the extent that anything in science is fact. Denying it is not wise or thoughtful or healthy. It’s nonsensical.
Moreover, using your degree of skepticism, which includes an inherent skepticism of anything agreed upon by a “consensus”, you have not explained why you believe anything. Why do you believe in relativity (if you do)?
Lastly, please, will one of the skeptics provide his or her own well-formed theory that explains how the increase in CO2 is NOT going to warm the atmosphere?
Re: 110
[some inflammatory remarks have been deleted -moderator].
“If, however, the collective of all skeptics/contrarians only ever engages in sniping, denial, palpably bad science, etc., *especially when the resources are available for them to do real science*, it’s hard to avoid drawing certain conclusions”
I disagree on all counts. The premise is incorrect which is easy to demonstrate by mentioning Lindzen. The logic is absent, too. In math, for example, someone who didn’t believe yet another proof of Fermat theorem wasn’t ever expected to show an alternative proof.
Re: 111
“Speaking of lies, would you care to enumerate the differences between climate and weather models?”
“This is easily found on the NOAA pages and at climateprediction.net.”
No it’s not easily found there. Please summarize the differences in your own words.
“how come we don’t know whether the next January will be cold or mild?”
“Many climate models are generally satisfactory in predicting future climate patterns.”
OK. So tell me: will the next January in NY be cold or mild?
[Response: The weather next January is… weather. Not climate – William]
“It is easier to predict a month’s climate conditions for a general location three to six months in advance than it is to predict the weather conditions for a single day three to six months in advance. The climate models prove that, since they take teleconnections, atmospheric composition, etc. into account”
How do the climate models take into account teleconnections and atmospheric composition? Do weather models handle these differently or not at all?
[Response: Teleconnections are taken into account in the models. Teleconnections are just atmospheric dynamics. Weather models may or may not include changes in atmos comp – over their timescales, the changes are small – William]
Re: 116
” Predicting the net average temperature for the planet earth in 2226 uses a totally different method.”
Could you describe this method and summarize the differences against the weather forecast?
“In aggregate, the overall climate in non-chaotic”
Do you have a proof?
“We know, for example, where the planet Neptune will be many centuries from now, because we understand the physics of gravity and momentum, and because the solar system is a non-chaotic system.”
First, it’s a bad example because the set of equations is entirely different. Second, solar system is many-body problem which becomes very quirky.
“if we know the physics, know the inputs and their influences, it seems that we can make pretty good predictions, as has already been fairly well demonstrated (albeit not absolutely conclusively).”
We don’t know all physics and it wasn’t demonstrated at all that we can make good predictions.
“I might note that blindly disregarding experts is every bit as foolish as blindly believing. Isn’t it?”
True. I’m not sure why you brought it up, though.
“An example is the suggestion you made (I think it was you) that perhaps the increase in CO2 is not due to burning of fossil fuels.”
Think again.
“Moreover, using your degree of skepticism, which includes an inherent skepticism of anything agreed upon by a “consensus”, you have not explained why you believe anything.”
I didn’t say any of the above. You did. In the future I’d appreciate if you stopped putting words in mouth. I hope the mod won’t consider it as an inflamatory remark.
“Lastly, please, will one of the skeptics provide his or her own well-formed theory that explains how the increase in CO2 is NOT going to warm the atmosphere?”
Do you realize that being skeptical to particular positions adhered to by the majority doesn’t equate to wholesale rejection of what climate science has produced? Or you can only see the world in black and white?
Re #100: I would disagree with your comments about progress and the need to hammer politicians.
Comment 1. “I think we have to ask why it’s been so difficult to get much traction on the non-draconian “no regrets” conservation, efficiency and alternative energy measures…”
In fact progress has been made in those areas and continues to be made. The rate of change may be slower than some want and uneven across the world, but it is nevertheless not zero.
Comment 2. “The cynical but probably realistic conclusion is that nothing very meaningful will be done about global warming until the politicians are being hammered over the head with the consequences.”
This is my point. They do not need a hammer, If the goal is creation of an effective, realistic, and affordable energy policy, they need a realistic assessment of the probabilities associated with the question of global warming: its liklihood, its causes, and the mitigations it may require. There are consequences associated with each of the following risk scenarios.
* global warming and its causes are exactly as currently described and humanity does nothing about it.
* global warming theories are wrong but governments enact mitigation policies that disrupt social orders and world and local economies.
* global warming is not significantly affected by human activity but governments expend resources and disrupt social order and economies to reduce human impact but make no provisions for dealing with the effects of warming.
(There are other risks, but these 3 illustrate my point.)
The risks need to be assessed and then there needs to be dialogue among nations to determine which is the greater risk (Global Warming is right or Global Warming is wrong) before jumping to solutions.Cuuerntly the noise generated by the current polarized debate ( see many comments above) obscures the evaluation of risks associated with both sides of the issue.
In comment #91, Dan Allan wrote, “Please explain what is pseudoscience.”
I’ve explained this and responded to your other comments here:
http://markbahner.typepad.com/random_thoughts/2005/10/responses_to_da.html
Just a picky note to Hans Erren’s 93. The solar spectrum does have an IR component (it mostly declines reasonably sharply from its peak at ~520 nm as any good blackbody spectrum would) and there are absorptions in the near IR from especially water overtones and combination bands, but also from CO2. There is even a tail in the solar spectrum that extends to the near IR. These absorptions have been measured in great detail. You can find them, and their cross-sections in the HITRAN database.
So yes, you can say that CO2 absorbs sunlight, just not very much of it.
Paul Emberger is wrong on multiple grounds when he states:
“Ultimately, however, the conclusions are based on correlational data and computer models.”
While you might argue that the best quantitative and detailed estimates come from computer models, the physics of the greenhouse effect are very basic indeed. This is shown by the fact that estimates of the effect on global temperature from a doubling of CO2 have not moved significantly (even as a function of the fraction of the total greenhouse effect) since the turn of the century. The basic physics is there and can only be ignored at our peril. The derivation of size the of the greenhouse effect is very simple, and takes about half a page in an atmospheric science text. Go look it up.
Secondly, conclusions are not only drawn from the basic physics and the correlational data and the computer models separately but from the ability of the models to track multiple data parameters in both time and space.
It is amusing to now see Emberger now say
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
when for over 15 years those of us who follow atmospheric trends have been advocating no cost and negative cost policies to deal with the global climate change while the Embergers of the world have been playing grasshopper. The problems are upon us, the easy solutions are no longer enough, and of course, the blame is being distributed to those who advocated early preventative action. We are indeed into the third stage of the problem.
Re: 122
“the physics of the greenhouse effect are very basic”
This is true. It’s the feedbacks that are not trivial.
“The derivation of size the of the greenhouse effect is very simple, and takes about half a page in an atmospheric science text.”
Also true. As long as you ignore everything else that happens in the atmosphere.
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
While I’d wholeheartedly support no cost and negative cost policies (should they existed), the quoted statement is right on the money.
Re: 115 Correction – The error would be less with larger gage openings like the 8 (the 12 in message 115 was incorrect) inch precipitation gages used by government agencies.
re #82’s point 2:
“the claim that climate can’t be predicted because weather is chaotic” is countered by by Stefan’s spoof. While the piece is funny it does nothing to contribute to the substance of the matter; I don’t see the point of linking it in this context. The presense of the annual cycle in the temperature signal in no way contradicts the chaotic nature of the climate system. I’m sure everyone here understands it. For example, we cannot predict when the next ice age could naturally start. Therefore the quoted statement is not exactly “wrong” as stated above. It’s not entirely valid either, of course, but it deserves a bit more than derision.
The claim that “climate can’t be predicted because weather is chaotic” is clearly false. It is being replaced here by the suggestion that “climate can’t be predicted because climate is chaotic”.
If the poster understands the distinction between weather and climate then the first claim should have been withdrawn and the second made separately. If the poster understands the technical meaning of “chaotic” then clearly the time scale of the sensitivity to initial conditions should be specified. Otherwise the assertion has no practical import for the sorts of questions taht appear to be the poster’s interest.
In fact, it is simply not the case that on can assert that climate is predominantly chaotic. Certainly, global mean temperature on a century time scale appears to be predominantly a predictable function of forcing. There is no reason to assert that chaos especially enters into it.
As for the idea that the glacial cycle is chaotic, that seems to be a confusion between “chaos” and “incomplete understanding”. In fact, one of the mysteries of the glacial cycle is its regularity, i.e., predictability.
I wish the discoverers of “chaos” had come up with a different name for it. Its confusion with the vernacular meaning has caused no end of confusion. I also thought Gleick’s book on the subject was awful. It left many people with a vague sense that they understood something deep and fundamental, whereas in fact they misunderstood something straightforward and modestly important.
On the other hand, #82’s point 1 is not without merit:
“the claim that global climate models can’t reproduce past climate change” is countered by the set of graphs http://www.grida.no/climate/ipcc_tar/wg1/figspm-4.htm In my view, these plots prove little if anything. All I see is that the models capture first order response to CO2 forcing. The observed temperature curve frequently gets out of the band which strikes me as a bad sign. Anyhow, until you select a metric (e.g. correlation) to quantify the similarity between observed and predicted, the statement “yes they can” is borderline meaningless. Moreover, the plots are produced with all the benefits of hindsight that allowed the authors to tune the model to achieve the best results. Forecasting 21-st century could require a different set of parameters.
In fact, the true test of climate models is paleoclimate reconstruction, in which they have been rather successful. I happen to know that a major oil company uses climate models of the deep past to inform their oil drilling operations, for instance.
I suspect that this obvious test of the validity of GCMs is not emphasized in public discussions because of the considerable influence of people who have an unshakable belief that the earth is 5000 or so years old, such that paleoclimate evidence is unconvincing to them. However, I would urge readers who are unconstrained by such beliefs to examine the paleoclimate literature.
Re: 126
“The claim that “climate can’t be predicted because weather is chaotic” is clearly false. It is being replaced here by the suggestion that “climate can’t be predicted because climate is chaotic”. If the poster understands the distinction between weather and climate then the first claim should have been withdrawn and the second made separately.
I didn’t make the first claim. However if the claim is so clearly false I (and everyone else) would appreciate if you can demonstrate it. I didn’t make the second claim either. What I said is that the notion is not disproved or ridiculous. What I meant is that the locals (mods) do a disservice to their cause by pretending that the issue doesn’t exist.
“If the poster understands the technical meaning of “chaotic” then clearly the time scale of the sensitivity to initial conditions should be specified.”
Before it could be specified it should be studied first, don’t you think? Currently, we haverather limited means for such studies.
“In fact, it is simply not the case that on can assert that climate is predominantly chaotic. Certainly, global mean temperature on a century time scale appears to be predominantly a predictable function of forcing. There is no reason to assert that chaos especially enters into it.”
Quite the contrary. There is a lot of interannual variability that cannot be explained by the monotonous forcing alone.
“In fact, one of the mysteries of the glacial cycle is its regularity, i.e., predictability.”
Too bad none of us will live long enough to check your prediction.
“I also thought Gleick’s book on the subject was awful.”
IMO, Gleick’s book is absolutely brilliant.
“In fact, the true test of climate models is paleoclimate reconstruction, in which they have been rather successful.”
Is there a model which can reproduce the last 1000 years worth of climate record?
I happen to know that a major oil company uses climate models of the deep past to inform their oil drilling operations, for instance.
Would you provide any details?
Re: 126
Even if a climate model is successful in paleoclimate reconstruction it may fail badly in its prediction for this century and beyond. People have changed the surface vegetation and water quality in drastic ways. The previous periods of global warmth had thick subtropical vegetation over large parts of the world. The northern Great Plains are known to have had subtropical vegetation just 50 million years ago (Fossil Butte National Monument in Wyoming), with crocodiles in the Arctic. The changed surface conditions of today will not allow plants to migrate or adapt with rapid global warming. In comparison to the large amount of CO2 that was removed from the atmosphere by vegetation during past warm episodes, the rate of CO2 removal by vegetation during the 21st century and beyond will be orders of magnitude smaller. I doubt that climate models used today account for the substantial differences in vegetation for the predicted period, versus vegetation for the past warm climate episodes used for paleoclimate reconstruction.
Re: #127, “”The claim that “climate can’t be predicted because weather is chaotic” is clearly false. It is being replaced here by the suggestion that “climate can’t be predicted because climate is chaotic”. If the poster understands the distinction between weather and climate then the first claim should have been withdrawn and the second made separately.
I didn’t make the first claim. However if the claim is so clearly false I (and everyone else) would appreciate if you can demonstrate it. I didn’t make the second claim either. What I said is that the notion is not disproved or ridiculous. What I meant is that the locals (mods) do a disservice to their cause by pretending that the issue doesn’t exist.”
You “didn’t make the first claim?” What about this: “There is a lot of interannual variability that cannot be explained by the monotonous forcing alone.”
“”In aggregate, the overall climate in non-chaotic”
Do you have a proof?”
He doesn’t need “proof.” The definition of climate is long-term weather patterns or long-term weather AVERAGES. An average eliminates all chaos and sorts it out into some semblance of order.
“”In fact, it is simply not the case that on can assert that climate is predominantly chaotic. Certainly, global mean temperature on a century time scale appears to be predominantly a predictable function of forcing. There is no reason to assert that chaos especially enters into it.”
Quite the contrary. There is a lot of interannual variability that cannot be explained by the monotonous forcing alone.”
Sashka, your response completely ignored the “Certainly, global mean temperature on a century time scale appears to be predominantly a predictable function of forcing” part of his argument. The previous message mentioned CENTURY TIME SCALES, not INTERANNUAL variability. Please read everything thoroughly before responding next time.
Sashka, please read into peer-reviewed journals and not the ExxonMobil Public Relations brigade’s bulletins. The authors of the former are actually climatologists, i.e. people who study the world’s climate, while the latter are not even scientists much of the time and most of what they say is completely inaccurate.
Re #128, CGMs do not currently include the carbon cycle, so your concern is not with the models as they now exist but with the uncertainties of the forcings which are applied to them.
That said, efforts are presently under way to add a complete carbon cycle to CGCMs. I actually suspect that these efforts are ill-advised for several reasons, some of which are related to Pat Neuman’s concerns.
While it’s not at all clear that Neuman’s intuition that biomass will decrease along with biodiversity is correct, I am concerned that coupled carbon cycle GCMs may be severely underconstrained. If so, they are, at great trouble and expense, unlikely to tell us much about the past and future trajectory of the carbon budget beyond what a simple desktop model might.
Michael Tobis writes, “In fact, one of the mysteries of the glacial cycle is its regularity, i.e., predictability.”
Sashka responds, “Too bad none of us will live long enough to check your prediction.”
I think there’s an outside chance that some people who are less than 20 years old today will live 10,000 more years. And I think there’s a better-than-50/50 chance that some people born after 2050 will live 10,000 more years.
But I don’t think anyone who is born or lives in the next 100,000 years will see an ice age. Human beings will have both the power and the will not to let the world slip into another ice age.
After all, contrary to what is implied by many on this blog, it’s pretty obvious that a world that is 2 or 3 or 4 or 5 degrees Celsius warmer is better than a world that is an equal amount *cooler.*
Sashka,
I don’t have the time or energy to respond to everything in your email (post 119), but I would like to take up a couple of points.
I wrote:
“Predicting the net average temperature for the planet earth in 2226 uses a totally different method.”
You replied:
“Could you describe this method and summarize the differences against the weather forecast?”
I did this in my email. I discussed the difference between a series of iterations based on initials conditions in weather science, versus a more linear equation that derives a climate at any point in time from a set of forecast.
I wrote:
“In aggregate, the overall climate in non-chaotic”
You replied:
“Do you have a proof?”
I’m not sure what would constitute proof of this. How does one prove anything is by-and-large non-chaotic? Climate does not have the attributes that one associates with highly chaotic systems – though there are some feedback loops, so it would be a mistake to claim that it is 100% non-chaotic.
I wrote: “Moreover, using your degree of skepticism, which includes an inherent skepticism of anything agreed upon by a “consensus”, you have not explained why you believe anything.”
You replied: “I didn’t say any of the above.”
Well, if you didn’t say it then my mistake. I’m somewhat bombarded here and lose track of who said what.
Your point in post 117 re fermat’s theorem:
Yes, a skeptic of the proof would not be required to propose an alternative proof. But he would of course be expect to show, with absolute precision, what was wrong with the proof that was published and independently reviewed. I don’t see that it is asking too much for climate skeptics to do the same:
(1) show what part of the theory is wrong, and why.
(2) Explain why temperatures are rising, if not due to increased C02, or if you agre that they are rising due to increased CO2, why, specifically, do you not expect that to continue?
There is a lot of interannual variability that cannot be explained by the monotonous forcing alone.
I think you mean “monotonic” here. Of course, only one of the major forcings is monotonic. I’m not claiming that there is no free variability. I’m simply saying it doesn’t dominate the global temperature record on time scales of longer than 10 years from all current indications.
Too bad none of us will live long enough to check your prediction. [re predictability of glacial cycle]
I think it’s generally agreed that the next ice age is cancelled or postponed as a consequence of the increased greenhouse gas inventory. Aspects of the recent geological past do appear notably unchaotic, though.
Is there a model which can reproduce the last 1000 years worth of climate record?
Nobody has done this as far as I know, but on that time scale, there would be little point. As we read here endlessly, the millenial record itself is not fully known, and is really quite unspectacular as far as variability is concerned. Further, the solar forcing is not known on that time scale. So you are asking to replicate bumps and wiggles that are both small and not well-known in the presence of forcing that is also small and not well-known. It would be hard in this instance to distinguish success from failure.
Simulations of the more interesting and better observed twentieth century have been extensively done, and it’s widely known that models can do very well with reasonable representations of aerosol and greenhouse forcing, but are not known to do as well in the absence of either or both.
For example http://www.cgd.ucar.edu/ccr/publications/meehl_solar.pdf
However, the best test of models’ utility in the dramatically changing future is how well they apply in the deep past. This presumes you are willing to accept the existence of a deep past, though.
Sashka is playing Why daddy? Usually this ends in tears.
Re: 129
He doesn’t need “proof.” The definition of climate is long-term weather patterns or long-term weather AVERAGES. An average eliminates all chaos and sorts it out into some semblance of order.
This is a misconception. Once you fix the “long” to be a particular time scale of interest (say 10 years or 100) you’ll see non-periodic oscillations on greater time scales. Generally, I believe it’s a known fact that fixed-length moving average of a chaotic trajectory is still chaotic. Certainly, the longer the time scale the more “semblance” of order you’ll see but at the same time you’ll lose the information.
Sashka, your response completely ignored the “Certainly, global mean temperature on a century time scale appears to be predominantly a predictable function of forcing” part of his argument.
I apologize for the omission and I’ll take the opportunity to comment on this now. If the statement above were true we’d never experience the Little Ice Age or Medieval Warming, not to mention Younger Dryas.
please read into peer-reviewed journals and not the ExxonMobil Public Relations brigade’s bulletins.
Thank you for your reading advice. I assure you I’ve read more peer-reviewed publications than you would be willing to believe. In turn I recommend exactly the same to you instead of spending time on the Sierra Club web site.
WRT 124:
ER1: “the physics of the greenhouse effect are very basic”
S: This is true. It’s the feedbacks that are not trivial.
ER1: “The derivation of size the of the greenhouse effect is very simple, and takes about half a page in an atmospheric science text.”
S: Also true. As long as you ignore everything else that happens in the atmosphere.
The fact that you CAN ignore everything else and get good agreement with the basic global surface temperature indicates that you CAN ignore everything else on a global scale. For the spherical elephant stage of any physics argument this is typical. (physicists define elephants as spherical to first order) Where you definitely CANNOT ignore it is for spatial distributions and small changes in forcings.
You would have a possible argument if the ghg perturbation were small, and indeed this was the case up until a few years ago, but as we move inexorable toward 2xCO2 it no longer is.
The only feedback which is NOT significantly balanced is the water vapor cycle, but, again on a global scale that is just driven by the Clausius Clapyron relationship. QED
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
S: While I’d wholeheartedly support no cost and negative cost policies (should they existed), the quoted statement is right on the money.
You are cordially invited to prove it.
FWIW no cost policies include more efficient lighting (moving away from incandescent bulbs), intelligent electric motor controllers, more efficient combustion engines, lighter cars, etc.
Re: 132
How does one prove anything is by-and-large non-chaotic?
In simple cases, one can find actual solutions and analyze stability using well-developed mathematical machinery. You can google for Lyapunov exponents, for starters. In more difficult cases, well, you have to do more work, I’m afraid. Like I said, this is not a trivial thing at all.
Climate does not have the attributes that one associates with highly chaotic systems – though there are some feedback loops, so it would be a mistake to claim that it is 100% non-chaotic.
But of course it does. Lorenz derived his equations from the same Navier-Stokes equations that any GCM is based upon (caveats apply).
Yes, a skeptic of the proof would not be required to propose an alternative proof.
Great to hear that I am relieved from one responsibility at least.
But he would of course be expect to show, with absolute precision, what was wrong with the proof that was published and independently reviewed.
I agree but let’s remember that climatology is not an exact science. It’s all about assumptions and probabilities. It’s hard to pinpoint anything because so little is certain. For example, I have no way to know what exactly each GCM is doing. They won’t even tell me what free parametrs the GCM is using.
All I’m doing at this point is discussing a couple of particular claims made on top of the page.
(1) show what part of the theory is wrong, and why.
(2) Explain why temperatures are rising, if not due to increased C02, or if you agre that they are rising due to increased CO2, why, specifically, do you not expect that to continue?
(1) I don’t know that the theory is wrong. It’s partly right and partly questionable. The part related to the quantitative predictions of the temperature rise is not well-grounded.
(2) I agree that the temps will probably rise for the next few decades, probably (mostly) due to CO2 but I don’t know what feedbacks could kick it later on (or earlier). I have little confidence in models predictions.
Re: 133
I’m not claiming that there is no free variability. I’m simply saying it doesn’t dominate the global temperature record on time scales of longer than 10 years from all current indications.
I don’t know whether it’s true. Even if it is, I’m not sure it equates predictability.
I think it’s generally agreed that the next ice age is cancelled or postponed as a consequence of the increased greenhouse gas inventory.
I’m not aware of such consensus. Doubt it.
Nobody has done this as far as I know, but on that time scale, there would be little point. As we read here endlessly, the millenial record itself is not fully known, and is really quite unspectacular as far as variability is concerned. Further, the solar forcing is not known on that time scale. So you are asking to replicate bumps and wiggles that are both small and not well-known in the presence of forcing that is also small and not well-known. It would be hard in this instance to distinguish success from failure.
The record may not be fully known but it is known that it contains centennial time scale variablity that the models won’t be able to reproduce. Why? Either we don’t know something about the physics or the models are deficient. In both cases, the projections into future are suspicious. The third possibility is the (solar) forcing. But there are no such results, as far as I know.
Simulations of the more interesting and better observed twentieth century have been extensively done, and it’s widely known that models can do very well with reasonable representations of aerosol and greenhouse forcing
In my initial post here I disagreed with precisely this statement.
Re: 136
The fact that you CAN ignore everything else and get good agreement with the basic global surface temperature indicates that you CAN ignore everything else on a global scale. For the spherical elephant stage of any physics argument this is typical.
No you can’t. If that were true, a 1D heat balance model would be sufficient but it isn’t. We need GCMs. But I’m glad you mentioned the elephant.
The only feedback which is NOT significantly balanced is the water vapor cycle, but, again on a global scale that is just driven by the Clausius Clapyron relationship. QED
Let’s see. For starters, how about a possibility of increased cloud reflectivity? Just Clausius “Clapyron”?
You are cordially invited to prove it.
To prove what? That before spending a trillion dollars it’s only reasonable to ask why?
FWIW no cost policies include more efficient lighting (moving away from incandescent bulbs), intelligent electric motor controllers, more efficient combustion engines, lighter cars, etc
Those are technologies, not policies, and they are not free. But these are all wonderful things.
>….”””The claim that “climate can’t be predicted because weather is chaotic””… I didn’t make the first claim. However if the claim is so clearly false I (and everyone else) would appreciate if you can demonstrate it.
The seasons and the climate of Mars are two clear examples Sashka….
BTW, its dead-set simple to determine whether chaos is important for the problem you are looking at. Simply perturb the initial conditions (or models components). This has been done to death in climate, and shows that it doesn’t change the global warming story. Just as butterflies flapping wings in Brazil don’t stop the coming of winter, similarly the flapping of butterfly wings don’t stop the coming of anthropogenically induced global warming…
Regards,
David
Re: [#130 > CGMs do not currently include the carbon cycle,]
Do the latest model simuations by The Max Planck Institute for Meteorology in Hamburg include the carbon cycle?
Excerpt:
02 Oct 2005
â??The Max Planck Institute for Meteorology is participating in the calculation of the IPCC scenarios with a coupled atmosphere-ocean model that is considered one of the best climate models worldwide. As scientists, we want to provide politicians with a decision paper that is as understandable as possible, and from which they can decide which measures ought to be politically implemented as urgently as possible,â?? said Dr Guy Brasseur, director of the Max Planck Institute.
http://www.earthtimes.org/articles/show/4155.html
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
S: While I’d wholeheartedly support no cost and negative cost policies (should they existed), the quoted statement is right on the money.
Eli: You are cordially invited to prove it.
To prove what?
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
Shall we go another round?
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
S: While I’d wholeheartedly support no cost and negative cost policies (should they existed), the quoted statement is right on the money.
Eli: You are cordially invited to prove it.
To prove what?
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
Shall we go another round?
re #141:
Do the latest model simuations by The Max Planck Institute for Meteorology in Hamburg include the carbon cycle?
Not to my knowledge. The quoted text gives me no reason to think otherwise.
Dynamic climate models (CGCMs) are typically run with prescribed atmospheric composition ot prescribed changes in atmospheric composition, commonly called “scenarios”. The main reason climate models run into the future are not really predictions is that it is very expesnive to run each scenario, and so there are far too many of them to run every plausible case.
So if nuclear power suddnly becomes popular ( good news ) or clathrate feedbacks start (bad news) or anything like that, it will affect which scenario is appropriate, but these events will be outside the issues dealt with by CGCMs, which are what is commonly meant by “climate models”.
In my opinion this is as it should be, but that’s controversial within the field.
[Response: HadCM3 can be run with a carbon cycle. Often it isn’t – William]
Re: #139, “‘The only feedback which is NOT significantly balanced is the water vapor cycle, but, again on a global scale that is just driven by the Clausius Clapyron relationship. QED’
Let’s see. For starters, how about a possibility of increased cloud reflectivity? Just Clausius “Clapyron”?”
Increased cloud cover would result in more reflectivity, but would also result in a greater percentage of heat trapped in the atmosphere. It is unknown, to me at least, how much of each would be the result, but there are many other impacts of cloud cover fluctuations than just a change in reflectivity.
“‘please read into peer-reviewed journals and not the ExxonMobil Public Relations brigade’s bulletins.’
Thank you for your reading advice. I assure you I’ve read more peer-reviewed publications than you would be willing to believe. In turn I recommend exactly the same to you instead of spending time on the Sierra Club web site.”
Well, to turn the tables on you, Sashka, which peer-reviewed publications have you read? “Energy and Environment” and “World Climate Report” are not peer-reviewed, for your information, so discard those from your list.
Also, being on the Sierra Club web site does not discredit my arguments, which have been made by several highly-touted climatologists in the past (some of whom post on this site from time to time). I was attracted to the Schneider interview because it was of Dr. Stephen Schneider, a climatologist with an impressive publication list. His site can be found here:
http://stephenschneider.stanford.edu/index.html
Re 123: Rabett proves my point about the focus of the Global Warming debate. Because I advocated for a careful risk analysis of the probabilities associated with global warming models and projections he immediately casts me as someone who has no interest in conservation or alternative energy sources. It is important to know the chances of being wrong or right to effectively respond. The stakes are too high to avoid the debate. Calling skeptics “grasshoppers” does nothing to further the goal of increased understanding and better predictions leading to better policies.
Rabett says “The problems are upon us, the easy solutions are no longer enough, and of course, the blame is being distributed to those who advocated early preventative action.” Blame is a completely irrelevant issue in this discussion.
Update on comment 17, my bet offer to Dr. Gray: I’ve heard nothing. I emailed him again today, requesting a response one way or another. I’d say give it another week without a response, and then we can guess what the answer is.
[Response: I mailed him too, today. Nothing yet, but I’ll be patient – William]
Re: 140
The seasons and the climate of Mars are two clear examples
I’ll let this hang without a comment.
BTW, its dead-set simple to determine whether chaos is important for the problem you are looking at. Simply perturb the initial conditions (or models components).
To trust this logic you have to trust the models in the first place, don’t you?
This has been done to death in climate, and shows that it doesn’t change the global warming story.
Surely you won’t mind posting first 5-10 references?
Re: 142-143
“Policies intended to deal with the predicted events are far reaching and impose great costs on social order, local and world economies, and the individual lives of billions of people. It is reasonable to challenge the conclusions that would drive such draconian mitigation strategies.?”
Shall we go another round?
Here: Kyoto is estimated to cost 5 trillion dollars.
http://www.aei-brookings.org/admin/authorpdfs/page.php?id=236
You agree that 5 trillion dollars is draconian, don’t you?
Re: 144
Do the latest model simuations by The Max Planck Institute for Meteorology in Hamburg include the carbon cycle?
Not to my knowledge. The quoted text gives me no reason to think otherwise.
Look for references to Archer and Maier-Reimer.