Register to reply

Playing Devil's advocate on climate

by Galteeth
Tags: advocate, climate, devil, playing
Share this thread:
D H
#55
Sep22-09, 05:06 AM
Mentor
P: 15,167
Quote Quote by sylas View Post
Quote Quote by vanesch View Post
Well, we tend to try to establish indeed some kind of "journalistic balance", but to get it "biased" towards the scientifically sound, we are rather strict (much more so than anywhere else) on the "peer review of sources" side.
This is an excellent solution, and a perfectly good resolution of the problem. It still leaves the field wide open to all the genuinely open questions on climate.
That would be an excellent solution if the science behind AGW was not suspect. It certainly looks suspect from the outside. More importantly, it looks suspect to some very well-qualified meteorologists and climatologists. Because PF does not have a supply of meteorologists and climatologists onboard, the PF solution is, as vanesch mentioned, a bit of a cowardly solution.
sylas
#56
Sep22-09, 05:22 AM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by D H View Post
That would be an excellent solution if the science behind AGW was not suspect. It certainly looks suspect from the outside. More importantly, it looks suspect to some very well-qualified meteorologists and climatologists. Because PF does not have a supply of meteorologists and climatologists onboard, the PF solution is, as vanesch mentioned, a bit of a cowardly solution.
Bingo. As it happens, I disagree with you... but it doesn't matter.

The whole reason for this solution is that it doesn't matter if you think it is the science behind AGW that is suspect, or the science of the critics that is suspect. In either case, the case can be found in the scientific literature, and moderators don't have to rule on the basis of choosing sides.

There are also well qualified meterologists and climatologists who think the science behind AGW is just fine. Not settled in every respect of course! There are plenty of open questions.

This is precisely why having an expert won't work. Which expert would be acceptable? The experts you are thinking of, who consider science behind AGW to be suspect? Or the ones I am thinking of, who think AGW is a well supported scientific hypothesis, confirmed to good confidence by a large amount of perfectly good scientific research and still subject to open questions in the details?

As I said previously, you will never find an expert who is acceptable to all sides -- even all sides within the staff!

You've said the science behind AGW is suspect. All I am really saying is that almost everyone can agree that there's a lot of suspect science involved... even if we disagree as to which is the suspect science!

The proper thing to do is to allow the various arguments to be considered, on all sides, and on their own merits... as long as those sides are presented in the scientific literature. That strips out a lot of really bad argument, consistent with the general physicsforums guidelines, and allows us to look at the arguments that have at least passed through the first level of check for credible scholarship. We do NOT rule on the basis of "consensus", or by some method of evaluating the "expertise" of authors, or by deciding which side the argument is on and either binning it or lauding it on that basis.

We look at the actual evidence presented, case, by case, by case.

I don't expect this process to resolve all the differences, but it sure as heck will let you learn a lot, if you are serious about looking at the arguments on their own merits. I've benefited enormously in learning about all the background physics as I've been working on this. Not just arguments for or against AGW, but about the underlying physics that lets me start to follow the details of the arguments being presented. And I'm still learning on that.

I especially disagree with you that this is "cowardly".

There's nothing cowardly about it. What would be "courageous"? To open up physicsforums to all arguments of any kind? Of course not. That would be a complete cop out. To step in and rule that one side or the other is disallowed? That would be just as bad.

So what? How do you decide... and WHO decides? You need something that lets you and I work together on this; not a ruling that rejects the experts you are thinking of or rejects the experts I am thinking of.

Cheers -- sylas
Galteeth
#57
Sep22-09, 03:14 PM
P: 320
"Physicsforums also deals with open questions. From the guidelines again:
There are many open questions in physics, and we welcome discussion on those subjects provided the discussion remains intellectually sound. It is against our Posting Guidelines to discuss, in most of the PF forums or in blogs, new or non-mainstream theories or ideas that have not been published in professional peer-reviewed journals or are not part of current professional mainstream scientific discussion.
In climate, there are many open questions, relating to sensitivity, cloud effects, paleoclimate, approximations in modeling, the carbon cycle, and on, and on. We aren't going to resolve them, but understanding the credible alternatives is well worthwhile.

There are also a lot of questions which are not actually intellectually sound open questions at all, but simply misunderstandings or even outright crackpottery. There's a lot that is said outside the professional mainstream which has been enthusiastically passed around in the public domain as being of equal standing to genuine science. In some respects, a lot of this public debate (not all of it!) bears a significant similarity to the whole "intelligent design" movement."


Yes! And it would be quite useful for a lay person to be able to distinguish between legitimate questions and crack-pottery. There is so much noise on this issue that's its hard to get to what's substantive. It is so frustrating when you have people convinced with no evidential basis that global warming is a hoax, then you have "scientists" who post things like "global warming is a 100% certainty" (this was a statement I saw posted on the NYT by a suppossed climatologist which made me raise my eyebrow).

A general overview of what is relatively certain, what's generally agreed upon, what's questionable, and what's controversial would be great. Perhaps I should start a new thread for this?
Andre
#58
Sep22-09, 03:57 PM
PF Gold
Andre's Avatar
P: 5,458
well as far as I'm concerned, there may be relative agreement about a BASIC climate sensitivity of about one degree celsius / Kelvin warming for doubling CO2.

So if you want to increase that to IPCC main figures of something like 2-4 degrees per doubling, you need predominance of positive feedback factors. That is questionable to controversial, like it or not.

Two approaches for that. On one hand statistics, investigating actual dynamic random walk temperature variation to be non-persistent, which suggest dominance of negative feedback, (several publications here). On the other hand direct measurements as suggested here.
Skyhunter
#59
Sep22-09, 04:19 PM
P: 1,409
Quote Quote by Galteeth View Post
then you have "scientists" who post things like "global warming is a 100% certainty"
Global warming is a fact. The instrumental record proves it to the very limits of scientific certainty.

Why should a scientist lend credibility to the global warming is a hoax crowd by giving a confusing nuanced answer about open questions and probability distribution curves?
sylas
#60
Sep22-09, 06:14 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by Galteeth View Post
A general overview of what is relatively certain, what's generally agreed upon, what's questionable, and what's controversial would be great. Perhaps I should start a new thread for this?
Yes, I think you should; but your topic is going to be too general.

We have different ideas about what the problem really is.

What is "generally" agreed upon is not the same as "universally" agreed upon. For someone on the fence, who is confused, and who doesn't want to have to learn a whole pile of background physics, it is a perfectly sensible question to ask about the "consensus", in the sense of what is the view of the majority of scientists working on the subject. And by majority I mean almost overwhelming majority.

If you want to know a wider consensus going beyond the working scientists, then the consensus is weaker, and in the general public it is gone.

Terms like 100% certainty are not useful, and only rarely used by the working scientists. AGW is not a single proposition, but a whole body of propositions, some of which are more certain than others.
  • That the planet is warming, significantly, on decadal scales, is a measurement about as certain as you can get. It's ongoing work to measure that more accurately, and also to find how the warming is distributed, and to separate the warming trend from stationary variations.
  • That human factors have a major contribution is about as certain as you can get. But it is also certain that they are not the only factor, and there's a lot of ongoing work to identify and measure the various factors involved.
  • In particular, that CO2 is rising rapidly and that this is by far mostly from human industrial emissions is a fact about as certain as you can get. That this necessarily has a significant impact on Earth's energy balance is basic physics, about as certain as you can get.
  • A big open question is the one identified by Andre. What is the sensitivity of climate... how much does temperature change in response to energy balance forcing? The best we have is a range of possible values, which as Andre notes is about 2-4 degrees per doubling. It is more usually quoted as 2-4.5, or 1.5-4.5, which is the IPCC range in the 4th AR. That's not a 100% certainty range; more like a 90% certainty range. There continues to be work published from time to time which proposes smaller values, and that is legitimately part of the whole scientific project. If you just want the majority view, then 2-4.5 works. If you want to really get into the debate, then you'll have to look at other proposals also, on their own merits. But there are very few.

That's a brief road map of some of the questions associated with AGW, as I see it. The two biggest open questions are the magnitudes of other forcings and sensitivity to forcings.

Other forcings include in particular cloud and aerosol effects -- and identifying causes for them as well. There are significant anthropogenic factors in these other forcings as well, and most of these are much harder to sort out than CO2, which is comparatively straightforward by comparison.

The sensitivity of climate is being better constrained as time goes by, but there is still a substantial range of likely values, and a handful of published arguments for radically different values.

The guidelines at physicsforums do not limit us to the majority or consensus view. We are free to consider any work published in the peer reviewed literature. That includes a very wide range of perspectives; and being able to examine those on their own merits is great. This thread speaks of playing "devil's advocate". That means actually knowing and understanding the arguments you don't agree with yourself; both for those who tend to accept the reality of strong anthropogenic effects on climate, and those who are skeptical of strong human effects.

Some of these arguments are very technical. Quite apart from anything else, getting to follow them is going to involve learning more about the background physics, and that's what PF is about.

Cheers -- sylas

postscript.
Quote Quote by Andre View Post
On the other hand direct measurements as suggested here.
There's no such thing as a direct measurement of sensitivity; and the paper you cite makes this clear as well. This is one of the legitimate handful of low estimates. It uses ERBE data. There's a fair bit of work been done with the ERBE data and sensitivity, and if you want majority views, then most of the ERBE based estimates are significantly higher than Andre's citation. To actually follow the details of how the estimates are made, either the low estimates in the paper by Linden and Choi, or the more conventional estimates from ERBE made by Wang, or Gregory and Foster, and others, you have to actually pull apart the details on their own merits. Just declaring one paper to solve the whole thing, or be a "direct" measurement, is foolish.
Xnn
#61
Sep22-09, 07:40 PM
P: 555
Within climate science, nothing is 100% certain. However, with respect to CO2 influence on the climate, the following can be said:

Equilibrium climate sensitivity is likely to be in the range of 2C to 4.5C with a most likely value of about 3C, based upon multiple observational and modelling constraints. It is very unlikely to be less than 1.5C.
page 88: http://www.ipcc.ch/pdf/assessment-re...ar4-wg1-ts.pdf

Notice:
Likely = 66% probability
Very Unlikely < 10% probability

In other words, it is more than very likely (>>90% probability) that climate sensitivity is > 1C, while the most likely value is about 3C.

So, while some may wish to argue that climate sensitivity is less than 1C, the odds are greatly against them being correct. However, to suggest that climate sensitivity may be less than 2C while an extreme statement is not out of the realm of possibility since the chance of it being correct is about 33%.

Probably the best thing to say is that there is a small, but not negligible, probability that climate sensitivity is less than 1C.
vanesch
#62
Sep23-09, 01:58 AM
Emeritus
Sci Advisor
PF Gold
P: 6,236
Quote Quote by Galteeth View Post
Yes! And it would be quite useful for a lay person to be able to distinguish between legitimate questions and crack-pottery. There is so much noise on this issue that's its hard to get to what's substantive. It is so frustrating when you have people convinced with no evidential basis that global warming is a hoax, then you have "scientists" who post things like "global warming is a 100% certainty" (this was a statement I saw posted on the NYT by a suppossed climatologist which made me raise my eyebrow).
You've got it. That's what makes this forum so hard to moderate. It is what I personally (I'm not talking on behalf of PF here) disliked most in some IPCC discourse and publications at a certain point: a certain lack of the typical self-criticism one usually finds in well-established science. When a supposed scientific discourse starts to sound like a commercial, or the program of a political party, I'm a bit lost.

BUT, but, how to express some reserves towards that, without opening the flood gates of crackpottery and low-level denial of basic established scientific fact ? That's the hard problem to solve here. And, as sylas pointed out, how can you solve it when certain experts have publicly shown not to show some "scientific self-criticism" ? Even a genuine renowed climate scientist wouldn't be recognized as an "objective expert" by "the other camp".
vanesch
#63
Sep23-09, 02:01 AM
Emeritus
Sci Advisor
PF Gold
P: 6,236
Quote Quote by Xnn View Post
Notice:
Likely = 66% probability
Very Unlikely < 10% probability

In other words, it is more than very likely (>>90% probability) that climate sensitivity is > 1C, while the most likely value is about 3C.
I really wonder how one gets to these "probability" numbers. I know that they are Bayesian estimators, but as we know, Bayesian estimators need a priori probabilities of our "belief", and only work under the hypothesis that our error probability model is assumed correct.

This looks to me like trying to say that "Newtonian mechanics has 95% chance to be correct", no ?
sylas
#64
Sep23-09, 02:38 AM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by vanesch View Post
I really wonder how one gets to these "probability" numbers. I know that they are Bayesian estimators, but as we know, Bayesian estimators need a priori probabilities of our "belief", and only work under the hypothesis that our error probability model is assumed correct.

This looks to me like trying to say that "Newtonian mechanics has 95% chance to be correct", no ?
That is an excellent question. Such "probabilities" or "likelihoods" are inevitably subjective. Technically, they are obtained using a Bayesian analysis with assumed prior distributions, as you say... and it is the choice of prior distributions where you can't avoid being subjective.

There is a thread about this very point I started just recently. See A low likelihood for high climate sensitivity. This is about a new paper just out by Annan and Hargreaves, who argue that it is reasonable to reject with high confidence the likelihood of sensitivity values above 4.5; higher confidence than suggested in the 4th AR. The whole basis for the paper is not any new data; but a more "reasonable" choice of priors in the analysis.

Annan and Hargreaves do not propose any new "objective" likelihood measure, but face up to the inevitable subjective aspects of quantified likelihood estimates for a certain unknown real world number.

We've mentioned the range 1.5-4.5 which is widely used, on the basis of the IPCC 4th AR. This is obtained using a "uniform" prior, in which unreasonably high sensitivities are given the same prior expectation as any others, in a certain range (0 to 10, for example). A uniform prior is sometimes taken as a way of avoiding results that incorporate information other than the data being applied; Annan and Hargreaves argue that this is dubious. From their abstract:
In this paper, we investigate some of the assumptions underlying these estimates. We show that the popular choice of a uniform prior has unacceptable properties and cannot be reasonably considered to generate meaningful and usable results. When instead reasonable assumptions are made, much greater confidence in a moderate value for S is easily justified, with an upper 95% probability limit for S easily shown to lie close to 4oC, and certainly well below 6oC.
It seems to me that this is a well grounded criticism, and that exceptionally high sensitivities are less likely than you would think from the 4th assessment report.

Note that the choice of priors has more effect on the upper bound than the lower bound; this is a consequence of the "long tail" of sensitivities, explained in the thread.

Felicitations -- sylas
vanesch
#65
Sep23-09, 04:07 AM
Emeritus
Sci Advisor
PF Gold
P: 6,236
Quote Quote by sylas View Post
That is an excellent question. Such "probabilities" or "likelihoods" are inevitably subjective. Technically, they are obtained using a Bayesian analysis with assumed prior distributions, as you say... and it is the choice of prior distributions where you can't avoid being subjective.
This is one of the criticisms I naively have, but apparently it can be dealt with.

However, there's a second, more fundamental criticism, and that is the assumed inherent correcctness of the error probability model.

Imagine I have an unknown system of which I have to make a prediction of response for a given input signal. Now, we can take our system as a black box, a grey box, or a white box, which means that for the black box, we only take former input-output relationships and try to fit a "general-purpose" behavioural black box model on it (for instance, a dynamical neural network), or for the white box, we can study the physics of what's in the box, and write down the supposed dynamics of the model, with or without free parameters, or we can be somewhat in between (grey box) where some physical modeling is done, and where some behavioural modeling is done based upon experimental data.

Now, in order to do any Bayesian estimation, I do not only need a correct model of my system, I also need a correct probability estimation of the errors on my model. If I have a biased model, and/or erroneous error estimations of this model, my Bayesian estimator will be totally off, concerning its probability distribution. Its expectation value might be still more or less right, but its probability distribution will be totally off.

In fact, it is much more difficult to have a correct probability distribution for a Bayesian estimator than to have a more or less good estimator expectation value.

This is why I always expressed my reserves towards these probabilities.
Now, maybe this is due to my naivety, but I've been working in system identification, and usually people are much more prudent when they do such things, than what I've seen from the IPCC.

Of course, there is a totally subjective way in which these numbers DO make sense: you can take them as "this is what we think, *to our best knowledge*. In other words, if I *had to bet my money* on the sensitivity, I would use this Bayesian probability distribution because it is "the best I can do with what I know".

But estimating "the probability that such event will happen to be 10% to the best of my knowledge", and claiming that the probability of that event IS 10%, is different. The first statement is: "ok, this is what I know, but I can be wrong". The second is "what's true".
sylas
#66
Sep23-09, 05:24 AM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by vanesch View Post
But estimating "the probability that such event will happen to be 10% to the best of my knowledge", and claiming that the probability of that event IS 10%, is different. The first statement is: "ok, this is what I know, but I can be wrong". The second is "what's true".
These are not probabilities, but indications of confidence. The report is not saying that there is a known probability of an event. It is quantifying their level of confidence in a proposition.

In other words, the IPCC report, as I read it, more than any other scientific report I have ever seen, attempts to be careful about saying "this is what we think", with explicit acknowledgment of limited confidence. That's just my impression.

It doesn't matter all that much. I tend to use the IPCC assessment report as a secondary reference anyway. I nearly always go back to the primary literature to get at the details. The IPCC report is useful to summarize the state of play from many diverse working scientists; but it's not gospel. It's a useful resource. For reference, then, if anyone would like to check against the report itself:
  • The Physical Science Basis is the section of the fourth IPCC assessment report that is relevant to physicsforums and the underlying science of climate. It is the report of "working group 1". From the link, you can download the whole report (996 pages) as separate chapters; also the Technical Summary and other materials.

In such a large report, it has been important to have a high level of consistency in how the report expresses its limited confidence in different propositions. A set of guidelines has been worked out for use throughout the report for expressing the level of confidence. These are described in box TS.1 (Technical summary, page 22-23), with additional pointers to other parts of the report that go into details of error and uncertainty handling.

In particular, the estimates of sensitivity are given on page 66 of the technical summary; with a brief overview on page 88 (quoted above by Xnn) and with more detail in the main report. See especially box 10.2, on page 798 (chapter 10). I quote here the tail end of that box:
Since the TAR, the levels of scientific understanding and confidence in quantitative estimates of equilibrium climate sensitivity have increased substantially. Basing our assessment on a combination of several independent lines of evidence, as summarised in Box 10.2 Figures 1 and 2, including observed climate change and the strength of known feedbacks simulated in GCMs, we conclude that the global mean equilibrium warming for doubling CO2, or ‘equilibrium climate sensitivity’, is likely to lie in the range 2°C to 4.5°C, with a most likely value of about 3°C. Equilibrium climate sensitivity is very likely larger than 1.5°C.

For fundamental physical reasons as well as data limitations, values substantially higher than 4.5°C still cannot be excluded, but agreement with observations and proxy data is generally worse for those high values than for values in the 2°C to 4.5°C range.
You could hardly be more thorough in avoiding a claim of "what's true"! They are explicit on the foundations of the estimate, and about the limits of their confidence. The values given of 2 to 4.5 are not probabilities of an event, but an indication of how confident they are about this range. The level here is "likely". For the "very likely" range, they give a lower bound only, of 1.5. The omission of an upper bound is deliberate, and reflects how hard it is to give a strong upper bound, given the nature of sensitivity and the "long tail".

In brief, they don't know the sensitivity. No-one does. The best they can do, as an assessment of the current state of knowledge, is that is it "very likely" more than 1.5, and "likely" to be in the range 2 to 4.5. You should read those as indications of how confident they are; not that this is a probability for an event.

In fact, the major criticism of the IPCC report, in this instance, is that it is too cautious in drawing conclusions. The paper by Annan and Hargreaves that I discussed is suggesting explicitly that the IPCC report could reasonably put an upper bound of 4.5, or even lower, in the "very likely" confidence level. It looks to me that they make a good case for this.

It doesn't bother me at all that people have different reactions to the IPCC assessment. One possible approach for moderating consistent with the guidelines would be to let the assessment be a guide to the range of views that are "mainstream", and rule out any argument that goes outside those bounds. I don't think that would be a good idea; but if you ever did decide to delimit boundaries for "mainstream" this would be your best option. But I think you have a much better approach already in place; which allows us to consider any view, from any individual, as long as they have taken that first step of getting it published in the peer reviewed literature.

So I don't mind at all that people differ in their reaction to the report. That's neither here nor there, and shouldn't be used to limit what propositions may be considered in the forum.

My own main interest is to help people understand better the physics of climate, and to learn more about it myself. It seems to me that a lot of debate on climate is based on visceral reactions rather than real physical comprehension of the various concepts.

Cheers -- sylas

---

PS. On reflection... I suggest that there is a problem in some quarters about how the IPCC report is used -- or indeed any scientific paper. Some people seem to treat the IPCC report as gospel, and that's dreadful. Same goes for treating any scientific paper as a final definition of what is known. Presenting a single paper as a final proof or disproof of a tricky scientific question is a misunderstanding of the nature of science. I don't personally think this is a fault of the IPCC report itself; which is more than usually careful about explicit recognition of limited confidence. It is rather a failure of understanding of a few people supposedly defending the report.

You may be able to identify in retrospect a major paper that has solved some problem; but that is generally something that is properly recognized some time after the publication, as other working scientists test and review its results. This is also why I would be against using the IPCC report as a limitation to define the mainstream for the purposes of moderation.
Xnn
#67
Sep23-09, 10:37 AM
P: 555
Consider the 3C/CO2 doubling most likely value.

CO2 levels are going up about 0.5% per year.
That works out to 0.015C/year of warming.

Now; look at the amount of observed global warming over the last 30 years.

It's about 0.016C/year.

In other words, observations are tending to support the most likely value.
Xnn
#68
Sep23-09, 10:41 AM
P: 555
Concerning criticism of the the IPCC; has anybody noticed that peppered throughout their report are the identification of areas that have a low level of understanding?

Ice Sheet dynamics come to mind. A fair number of people are concerned about just how fast they may melt, but the IPCC hasn't made a definiative statement regarding them and for good reason.
sylas
#69
Sep23-09, 04:31 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by Xnn View Post
Consider the 3C/CO2 doubling most likely value.

CO2 levels are going up about 0.5% per year.
That works out to 0.015C/year of warming.

Now; look at the amount of observed global warming over the last 30 years.

It's about 0.016C/year.

In other words, observations are tending to support the most likely value.
That's technically incorrect on many levels.

First, the doubling value of 3 C/2xCO2 is an equilibrium response; not the transient response.

There's a lot more changing than just CO2. Taken in isolation, CO2 is the largest forcing, but there are a number of other forcings, both positive and negative, and they make a difference.

An increase of 0.5% per year is log2(1.005) of a doubling, and this is 0.0072. At 3 degrees per doubling you would have 3*0.0072 = 0.021 C/year.

As it turns out, observations are consistent with the estimates of sensitivity, but the argument is a bit different. It is not sufficient to give a strong constraint on the sensitivity estimate. If you look at the section of the technical summary you quoted previously, the next paragraph deals with "transient response" as opposed to "equilibrium sensitivity". Chapter 10 deals with the difference between these concepts in more detail. Here's the extract (page 88, technical summary)
The transient climate response is better constrained than the equilibrium climate sensitivity. It is very likely larger than 1°C and very unlikely greater than 3°C. {10.5}
With a transient response 2°C per doubling, and with an increase of 0.5% per year CO2, and in the absence of all other forcings, you should expect a temperature increase trend of about 2*log2(1.005) = 0.0144 C/year. Of course, there are other forcings as well; but all told this is why transient response is better constrained. The available data for the immediate present allows you to constrain it, where equilibrium response requires an additional level of indirection to get from observations to a number.

Also relevant is a recent thread with a discussion of the cummulative carbon emissions response (CCR), which tends to reflect transient climate response rather than equilibrium climate response. There were two recent papers on this, which are discussed in msg #83 and following of thread "Estimating the impact of CO2 on global mean temperature".

Cheers -- sylas
Xnn
#70
Sep23-09, 06:54 PM
P: 555
Thanks sylas;

So, doing this in natural logs, the calculation is 0.016/0.0072 = 2.2 C/CO2 doubling; using the last 30 years of global observational data if other forcings are ignored.

Not sure if how significant other forcings are.
sylas
#71
Sep23-09, 08:40 PM
Sci Advisor
sylas's Avatar
P: 1,750
Quote Quote by Xnn View Post
Not sure if how significant other forcings are.
Neither is anyone else; there are significant uncertainties in other forcings. CO2 is the simplest forcing to evaluate and is known to quite high accuracy. Other forcings are known, but with larger bounds of accuracy. The bounds are sufficient to conclude with strong confidence that greenhouse effects are the largest forcing, and that CO2 is the largest contributor to that. But the lack of certainty in other forcings still means a large spread of uncertainty about total forcing.

I've given a widely repeated diagram in msg #66 of thread "Estimating the impact of CO2 on global mean temperature", outlining estimates and uncertainties for forcings. It can be found also as figure 2.20, on page 203 (chapter 2) of the 4th AR. The estimates are quantified, with 90% confidence bounds, in table 2.12, page 204.

Felicitations -- sylas

PS. As an added wrinkle... a forcing is a change from one time to another. The table I've shown is giving the forcings from 1750 to the present. But if you want to look at the last 30 years, then the picture changes again, generally making the greenhouse effects and CO2 in particular even more significant for the immediate rate of change. The immediate rate of change also is affected by changes in heat uptake in the ocean, which will impact rates of change in much the same way as a forcing.
Saul
#72
Sep25-09, 07:13 AM
P: 272
The other explanation for the 20th century warming is a decrease in planetary clouds rather than AWG, in particular CO2.

The observations do not show a steady increase in the base line planetary temperature about which planetary temperature oscillates. Planetary temperature has in fact cooled slightly post 1998.

http://www.woodfortrees.org/plot/had...09.8/normalise


http://wattsupwiththat.com/2009/09/0...of-0-4%C2%B0c/

As I noted further up in the this thread planetary temperature in the 20th century has strongly correlated with geomagnetic field change measured by the parameter Ak that is in turn modulated by solar wind bursts. The solar wind burst remove cloud forming ions by a process called electroscavenging. The sudden warming events in planetary temperature records and the longer term warming and cooling trends correlate with Ak.


http://www.atmos-chem-phys.org/5/172...1721-2005.html


Analysis of the decrease in the tropical mean outgoing shortwave radiation at the top of atmosphere for the period 1984–2000

All cloud types show a linearly decreasing trend over the study period, with the low-level clouds having the largest trend, equal to −3.9±0.3% in absolute values or −9.9±0.8% per decade in relative terms. Of course, there are still some uncertainties, since the changes in low-level clouds derived from the ISCCP-D2 data, are not necessarily consistent with changes derived from the second Stratospheric Aerosols and Gas Experiment (SAGE II, Wang et al., 2002) and synoptic observations (Norris, 1999). Nevertheless, note that SAGE II tropical clouds refer to uppermost opaque clouds (with vertical optical depth greater than 0.025 at 1.02μm), while the aforementioned synoptic cloud observations are taken over oceans only. The midlevel clouds decreased by 1.4±0.2% in absolute values or by 6.6±0.8% per decade in relative terms, while the high-level ones also decreased by 1.2±0.4% or 3±0.9% per decade in relative terms, i.e. less than low and middle clouds. Thus, the VIS/IR mean tropical (30_ S–30_ N) low-level clouds are found to have undergone the greatest decrease during the period 1984–2000, in agreement with the findings of Chen et al. (2002) and Lin et al. (2004).

http://news.yahoo.com/s/mcclatchy/20...latchy/3295216

WASHINGTON — Has Earth's fever broken?
Official government measurements show that the world's temperature has cooled a bit since reaching its most recent peak in 1998.

"It's entirely possible to have a period as long as a decade or two of cooling superimposed on the long-term warming trend," said David Easterling , chief of scientific services at NOAA's National Climatic Data Center in Asheville, N.C.

So far this year, the high has been 0.42 degrees Celsius (0.76 degrees Fahrenheit), above the 20-year average, clearly cooler than before.


Register to reply

Related Discussions
Debating a Flat-Earth 'Devil's Advocate' General Discussion 15
Devil's Advocate for an A.I. project Computing & Technology 12
The Devil's in the Algebra. Introductory Physics Homework 3
Devil's Advocate General Discussion 77