In comments at Climate Audit Boris (#38) seems to suggest the
A tropical tropospheric hotspot is expected from any warming
Claiming that a response to any warming is a fingerprint of anthropogenic warming is twice as silly.
Well, this is an interesting accusation! Because if we believe claiming the hot spot is a fingerprint is silly, it think by after reading my post, we will be required to conclude , the authors of the AR4 are not only silly, but twice as silly!
Lucky for us, in a comment explaining what an ensemble of models shows, Deep Climate provides figures to support Boris’s single sentence claims. The figures came from a Dec 2007 blog post authored by “the group” at Real Climate and discussed the results from one model, not an ensemble of models.
It turns out that back in Dec. 2007, “The Group” at Real Climate informed readers that the hot spot was expected for any source of warming. The tone of the RC blog post would suggest that this information is widely accepted and agreed on by everyone. They appear to support their contention using a single GISS model run using an increase in solar forcing not experienced by the actual earth.
Deviating from scientific cultural practices, they fail to suggest their readers refer to a recent widely respected consensus document that might seem to contradict their own views about the cause of the “hot spot” interpretation. (Interested readers can refer to section 9.2.2 of the WG1 report for the IPCC AR4. )
How is this “seeming” achieved? By inserting a discussion of non-uniqueness of the pattern of warming into a criticism of Douglas et al, the RC article would make it appear that the RC view is that their figures are meaningful to Douglas et al.’s allusion to the pattern. Those figures are not meaningful in this context. The Douglas discussion relates to predicted patterns one might hope to detect in the observational record. These would be patterns that arise given the specific temporal evolution of forcings that has occurred on the earth.
So, what’s the truth about the hotspot?
Who knows?
But we can answer easier questions like these: Who tells us they think the tropical tropospheric hot spot is a fingerprint of AGW? Who thinks it’s expected for any and all warming?
In their blog post, “The Group” at RC tells us the hot spot is expected but appears regardless of source of warming.
What does the AR4 tell us?
Chapter 9 of the WG1 report to the IPPC’s AR4 published in 2007 contains a number of figures which look somewhat different from those published at RC:
I have circled the plates illustrating the results for well mixed GHG’s and those for all sources of warming combined. As you see, according to the AR4– a consensus document written for the UN’s IPCC and published in 2007 — models predict the effect of GHG’s as distinctly different from that of solar or volcanic forcings. In particular: The tropical tropospheric hotspots appears in the plate discussing heating by GHG’s and does not appear when the warming results from other causes.
The figures are accompanied by a fairly lengthy the discussion of the fingerprints in the AR4. The first paragraph is shown to the below.
Note that the IPCC AR4 discussion of this “fingerprint” issue, and the hot spot, using Gavin’s favorite word: “Robust”.
Evidently, the scientists who’d examined the output of the many models used to support the AR4 thought, “The major features in figure 9.1 are robust to using different climate models.”
All this would appear to differ from the storyline provided by RC. What is one to think?
Maybe the AR4 authors use the term “robust” to mean “Found in most models, except when driven by forcings not experienced on earth which we have decided to ignore for the purposes of our discussing what we mean by the ‘fingerprints’ of global warming.”
Who knows?
Now, the answer to the question!
“Who expects tropical tropospheric hot spot from any and all sources of warming?”
Evidently, “The Group” at Real Climate, and those readers of Real Climate who either do not read or do not believe section 9.2.2 of the WG1 report for the IPCC AR4.
Who thinks the tropical tropospheric hot spot is a fingerprint of GHGs? People who read and believe section 9.2.2 of WG1’s contribution to the IPCC AR4.
It is, of course, a possibility section 9.2.2 of the WG1’s contribution to the IPCC A4 provides an entirely distorted, confusing or misleading impression of the consensus position on what models predict about the hot spot. Anything is possible, right?
Update
I added a clause to a sentence to better describe the argument at RC which relates to their switching from the IPCC discussion using real earth variations in forcing to discussing hypothetical forcing levels that are not possible for certain sources of warming. I added a paragraph to clarify my use of “seeming”.
Update
New comment thread here.
lucia,
The images you display above are based on simulations from the actual forcing from 1890 to 1999. *If* there was substantial solar forcing over that time period, then there would be a tropical troposphere hot spot from solar. Little solar forcing leads to little solar warming.
I believe there’s another image plate that shows the warming patterns for different types of forcings. I’ll see if I can find it…
JohnV–
Yes. But this is what the IPCC AR4 authors mean by a finger print. That’s how they use the term.
We don’t expect that “fingerprint” based on the amount of solar forcing we have actual experienced. Whether or not we expect it based on an increase in solar forcing that has not occurred is a red herring promulgated by RC.
lucia,
Your title for this article made me think you were asking if a tropical tropospheric hotspot would occur for reasons other than GHG forcing.
You them implied that the GISS model results were different than other model results, and that IPCC AR4 contradicted the GISS model results. You used Figure 9.1 as your evidence. However, Figure 9.1 shows the response to *actual* forcing and the RealClimate images were due to a hypothetical 2% increase in solar forcing.
Of course the results are different. The experiments are different.
John V–
Fair enough. The title is already long. in my opinion, “any and all” in the title can encompass amounts too (at least in titles.) But, the fact is, it’s difficult to craft titles that convey detail.
I agree on my sentence was poor. I edited the sentence you don’t like after I read your first comment and before I read your more recent one. I posted the update note.
By my reading, the RC article is criticizing those who adhere to the idea the hot spot is “the fingerprint” of global warming using the exact idea described in the AR4. To formulate their criticism, RC introduces a red herring, uses forcings outside the range seen on earth, and shows that if fictional things happened, we’d get the same result. That is the way they try to make the hot spot not a “fingerprint” of AGW.
Boris was criticizing Steve for using the term exactly as the AR4 uses it and suggesting Steves use is “silly”. If Boris, or Deep Climate wish to provide more detail and explain they think the use in the AR4 is wrong or silly because one should think about this in light of what would happen if we applied fictional forcing to the earth, they should say so.
The fact is: According to the IPCC, the models don’t predict a tropical tropospheric hot spot on the real earth absent the warming effect of GHGs. We don’t expect it based on the decrease in volcanic activity since the turn of the century or more recently. We don’t expect it based on the slight increase in solar activity since the 1800s. We don’t expect it based on any actual changes in forcings other than the theorized changes due to GHG’s. We don’t expect it from natural variability.
So, the hot spot is considered the fingerprint (according to the AR4.) It is expected to occur only as a result of the amount of warming attributed to GHGs.
lucia,
I agree with your comment above. However, your article still conveys that the GISS model results contradict the IPCC AR4 results. That is simply not the case. They are different analyses. You don’t think the RC analysis was pertinent or appropriate, but it’s still a different analysis than the one presented in IPCC AR4. The analyses do not contradict each other.
I understand that you disagree with the definition of “fingerprint”, but your article goes way beyond that.
I think we agree on two points:
1. Hypothetical solar forcing would lead to a tropical troposphere hotspot;
2. With real world forcings, only GHG forcing can account for a tropical troposphere hotspot;
Do those results contradict each other?
JohnV:
Agreed on the first paragraphs. I edited to:
lucia,
The part that I think over-reaches is this:
“Deviating from scientific cultural practices, they fail to suggest their readers refer to a recent widely respected consensus document that might seem to contradict their own views about the cause of the “hot spot†interpretation.”
JohnV–
I agree on 1 &2 with the caveat that I’d tag on “The model predict” or “theory predicts”. The reason for the tag on is I am an empiricist and until any prediction gets strong confirmation by data, I consider it a prediction. (It can be plausible one I anticipate will be confirmed, but still a prediction.)
lucia,
I agree with your “add-ons” to points 1 & 2.
By the way, I looked for an IPCC image equivalent to the RC image but couldn’t find one.
Give “the Group” a break. They are just trying to steer folks away from the “hot spot” issue, because nobody can find it and that is getting embarassing for the believers. Now, they have no hockey stick, and even the models are failing in many respects. What’s left to bolster all the fear? Somebody has no clothes.
Thanks, this is a very clear explanation. I did once ask about this on Tamino’s blog, and got the reply that the hotspot was not a critical test, and would happen no matter what the source of warming, but got no explanation of why not. Its a really important issue, because if it is a signature, it is pretty simple to verify whether or not GG warming is taking place, and you show very clearly that it is an IPCC view that it is a signature.
Also, if it is a signature, McKittrick’s suggested tax starts to make a lot of sense – that is, you tax people for emitting CO2 in direct proportion to the extent of the warming the hotspot shows it is producing. The worse the GG warming, obviously, the higher the tax rate should be.
Its a natural extension of your stuff on verification. At the moment you are checking whether the level of temperatures is in accordance with the forecasts. Its a natural extension to ask whether the distribution of temperatures is in accordance also.
Happy Christmas. Your blog has been a great addition to the climate discussion. A rare combination of good humor and forceful argument with technical sophistication. Looking forward to what it comes up with in the New Year.
Lucia – you have completely misinterpreted the discussion in IPCC AR4 WG1 section 9.2.2; the realclimate folks are correct on this, and you are quite wrong. Any source of (tropical) warming causes a tropospheric hot spot; it is essentially equivalent (though slightly more difficult to detect due to measurement constraints) to the question of whether the surface has warmed.
Has the tropical surface temperature increased on average since year x? If so, then under the straightforward assumption that the tropical atmospheric lapse rate is governed by the wet adiabatic lapse rate (i.e. convection is running at essentially its maximum rate while the air is, at least above a relatively low altitude, essentially saturated with water), the increased water vapor levels have the effect of decreasing the lapse rate, and thus raising the temperature at higher altitudes. Greenhouse gases do not stand in the causatory flow between warming and the tropospheric hot spot.
What the IPCC report was getting at in that section on signatures (I don’t believe it uses the word “fingerprint” at all) is the actual specific signature of greenhouse warming which has indeed been observed: cooling of the stratosphere. If the warming was from solar forcing, the stratosphere would warm just like the surface.
If you still have trouble believing this, reread that section of the report. I’ll go through it with you line by line if you’re really having trouble, it’s not that hard to understand. After you do understand it, you might want to seriously question the sources you have trusted in the assumptions you made in writing this post in the first place.
Lucia:
Since this “hot spot” is just what Pat Keating has argued is a probable error in GCM modeling (your Dec.3 blog), why not solicit some input from him for this thread?
JohnV–
You are entitled to your opinion, but I don’t think that sentence over-reaches. Here’s my view:
Did RC provide a reference to the AR4’s discussion of the fingerprints? No.
Is it the norm to provide these when criticizing Douglas et al. reference to those finger prints? Yes.
Are the AR4 usage and the Douglas usages the same? Yes.
Does the AR4 wording seem to suggest that people believe the hot spot is a finger print of AGW and GHG’s? Yes.
If RC wanted to provide a nuanced argument about the hot spot or follow normal scientific cultural practices, they would have pointed to the AR4. They did not.
An increase in the solar constant of 2% is at least an order of magnitude beyond the observed and possible in human time scales so any comparison of solar warming on that scale to ghg warming is a red herring so rotten it glows in the dark. Not to mention that a change in the solar constant of that magnitude would be readily and immediately observable. Stratospheric cooling is indeed a signature (and how is a signature different in principle from a fingerprint?) of increased stratospheric CO2 (and ozone loss). So the question is, if the stratosphere is cooling as predicted by radiative energy transfer calculations, why isn’t the upper troposphere warming? Could it be that the GCM’s don’t properly model deep convection and/or clouds? The tropospheric hot spot is not so much a signature of ghg warming as it is a test of the validity of the current climate models.
Arthur–
With all due respect, I do not believe I have completely mis-interpreted sect 9.2.2. Have you read section 9.2.2 recently?
On this:
“Any source of (tropical) warming causes a tropospheric hot spot;”
Sure. See comments above. Also, why do you think I’ve denied the adiabatic lapse rate?
Also, do you think section 9.2.2 says any source of warming causes a tropospheric hot spot? I don’t dispute that any source can cause the hot spot- but that doesn’t mean the IPCC authors discussed this in section 9.2.2. As far as I can tell, they did not do so– and so presumably, they didn’t consider that to be an important point with regard to their discussion of fingerprints or patterns.
I note that immediately after discussing the hot spot in section 9. 2.2, the document says this “but in contrast to greenhouse warming, the simulated solar-forced warming extends throughout the atmosphere”. The use of “in contrast too” would see to somewhat dissociate the solar warming from the “hot spot”.
The IPCC authors do discuss a small effect of tropospheric warming and cooling during the solar cycle, quantifying to emphasize the small ness of the solar effect.
The key is: The Boris, inspired by the RC article, suggested calling the hotspot a fingerprint of AGW “silly”. But the RC argument requires a hypothetical increase in solar warming. In contrast, when discussing fingerprints/patterns or signatures, the IPCC restricts themselves forcints that actually occur. In the IPCC usage the warming of the troposphere is a pattern/signature/ fingerprint of GHGs.
Whether or not a pattern would also exist under some fictional circumstances is a red herring. You can’t just change the subject to the counterfactual “what if fictional forcings existed” and t hen redefine terminology to apply only to the counter factual and decree we can’t use it the way the IPCC did. RC also can’t go around insinuating there is anything odd about Douglass, Christy, Singer and Pearson comparing data to the fingerprint.
Did you know you can search PDFs for terms like fingerprints?
And looky here, from page 702 in chapter 9:
You’ll find the term in Chapter 9. Appendix 9A has the title “Optimal fingerprinting”.
However, you are correct they don’t use the term “fingerprint” in section 9.2.2, they use the word “pattern”. Is the fact they vary terms important to anyone’s argument?
BTW: Have fun trying to find “adiabatic” in chapter 9. Had the IPCC authors wished to make some point about the effect of the adiabatic lapse rate, the term would likely appear.
fred–
Lack of a hot spot is a strong test.
Even under Arthur’s view that the hot spot happens for any and all warming, then we ought to see it if there is warming for any reason. (That is: Unless there is some problem with our understanding of the lapse rate. As Len pointed out, that’s what Pat Keating discussed.)
In contrast, if what Tamino, RC and Arthur say is correct (and I don’t dispute it) then finding the hotspot does not prove the source of warming and tells us nothing about attribution. No matter what the reason for warming we would see it. (Though…. I suspect if we start discussing volcanos, we’ll get a different view.)
And Merry Christas.
The tropical hotspot is not a signature of CO2 induced warming, it is a signature of the claimed H2O-mediated positive feedback magnifying the warming. (Although the hotspot itself is a negative feedback.) For people who agree that CO2 warms 1.1C/2x but dispute the 3x positive feedback postulated to turn it into a ‘disaster’, the hotspot is still a hot issue.
The models predict it because they reckon the warming of tropical oceans will lead to more water vapour – a powerful greenhouse gas itself – further warming the surface. The water vapour carries heat from the surface to the upper troposphere more rapidly, cooling the surface. The combination of the two counteracting effects is calculated by the computer models to be positive.
The absence of the hotspot is likely evidence that their model of massively positive water-vapour-based feedback is incorrect, and that water vapour is not increasing as much as expected. This does put a serious hole in the AGW dogma, but not quite in the place people think.
The IPCC show a hotspot only for greenhouse warming simply because their models say only greenhouse warming is generating the effect. (GIGO…) But what it is really a fingerprint of is increasing water vapour. A fingerprint that in reality is absent.
Stevo,
Let me put this in crude terms.
The tropical heat causes the formation of huge clouds/ thunderstorms which raise heat up towards the troposphere but that heat as it rises is cooled and turned into regular torrential downpours on a daily basis – but the models say the first process outweighs the effects of the second on the climate?
When and where has this actually been measured?
Lucia, you quoted the first paragraph of section 9.2.2.1 above, so let’s start there. What they specifically say there that’s important to this question of a signature or fingerprint is:
But being “robust to using different climate models” does not mean that every one of these “major features” is actually a “signature” that differs from one forcing to another. The “major features” in the temperature response refer of course to areas in the zonal/altitude plot that are on average expected to be warmer or cooler – but is the concern here the magnitude of that temperature change, or rather the *relative* response (and sign) of one region with respect to another? It becomes clear from further discussion in that section that the important issues are the *relative* responses, not the magnitudes – we’ll get to that in a bit (at least if I get a chance to get to that part before you cry uncle…)
So let’s look in detail at the “major features” with regard to *relative response* that can be found in figure 9.1, which you posted above:
First, note that (a) solar, (c) LLGHG, (d) ozone are all factors that warm the surface of the planet, while (b) volcanoes and (e) sulphate aerosols produce cooling. That is, the relative patterns associated with an increase in (a), (c) and (d) should be compared with their mirror images (in color), i.e. the negative of the changes in (b) and (e).
The major “regions” in figure 9.1 I’ll subdivide by “zone” into North polar region (60N-90N), tropics (30N-30S), and South polar region (60S-90S) (N, T, and S, respectively in what follows), and by altitude as surface (1000 hPa), mid-troposphere (200-400 hPa), and stratosphere (10-50 hPa) (Surf, Trop, Strat). To get relative numbers, let’s scale everything by the T-Surf values (obviously somewhat shaky given the limited color scheme of these graphs, but good enough for the overall patterns to emerge):
(a) solar –
T-Surf = 1 (actually 0 to 0.2, assume 0.1); N-Surf = 7 (0.6 to 0.8); S-Surf = 1
T-Trop = 3, N-Trop = -2, S-Trop = 0
T-Strat = 1, N-Strat = -2, S-Strat = -3
(b) volcanoes –
T-Surf = 1 (-0.2 to 0, assume -0.1); N-Surf = -3, S-Surf = 0
T-Trop = 1, N-Trop = 3, S-Trop = 1
T-Strat = -1, N-Strat = 3, S-Strat = 5
(c) GHG’s –
T-Surf = 1 (0.4-0.6, assume 0.5); N-Surf = 2.2, S-Surf = 1
T-Trop = 1.8; N-Trop = 0, S-Trop = 0.4
T-Strat = -2, N-Strat = -2, S-Strat = -3?
(d) Ozone –
T-Surf = 1 (assume 0.1), N-Surf = 3, S-Surf = 1
T-Trop = 2 (range 0 – 0.4), N-Trop = -5, S-Trop = -5
T-Strat = -2, N-Strat = -5, S-Strat = -5
(e) Sulphate aerosols –
T-Surf = 1 (assume -0.2), N-Surf = 1.5, S-Surf = 1
T-Trop = 2.5, N-Trop = 1.5, S-Trop = 1
T-Strat = -0.5, N-Strat = 2, S-Strat = 2
What are the patterns here?
(1) Stratosphere sign of change vs. surface:
Negative across all zones: (c) and (d).
Positive tropical stratosphere: (a) only
Positive polar stratosphere: (b) and (e)
(2) North pole amplification at the surface:
Positive amplification: (a), (c), (d), (e)
Negative response: (b)
No forcings see a small north pole response.
(3) Mid-troposphere polar response:
Positive in all zones: (b) and (e)
Negative polar response: (a) and (d)
Close to zero polar response: (c)
(4) Tropical mid-troposphere “hot spot”
Positive amplification in T-Trop: (a), (c), (d), (e)
For (b) (volcanoes) the ranges shown are too small to determine if there’s any amplification.
In other words, there *are* distinct patterns of response for “major features” 1 to 3, but “major feature” #4, the subject of this post, is in fact *not* a pattern that can be used as a “fingerprint” to identify one over another.
Supporting commentary in the text of the report (though perhaps not very well worded) to follow, if you really want me to.
Okay, round one is over… and it’s a good thing Lucia is wearing gloves and not brass knuckles because she clobbered Arthur. Arthur? I like a good fight… have your knees stopped wobbling enough to give some back?
of course Arthur jumped back in as I typed,,,
Round 2 has started. Arthur threw a wild left which missed it’s mark by about a metre.
Now, on another point, Lucia, you keep claiming that section 9.2.2 talks about the tropical mid-troposphere hot spot:
but that section, in addition to not using the word “fingerprint”, also never uses the term “hot spot”, and does not refer to the “mid-troposphere” specifically (it talks about warming of the troposphere as a whole, implying from surface to tropopause), and it also does not talk about the tropical troposphere or special warming in the tropics at any point.
So what makes you think section 9.2.2 has *any* support at all for the contention that the “fingerprint” is a tropical mid-tropospheric “hot spot”? I would love to see an actual full quote in context that supports this claim; I can find it nowhere in the text.
Did he miss by a meter or throw a punch in the wrong direction? And will Lucia catch onto what was the usual AGW response…
Arthur–
Who said it is?
I said the IPCC calls this a pattern / fingerprint / signature and associates it with GHG’s using whatever they consider relevant to decreeing an association exists.
If, for some reason, you or RC doesn’t like this definition of “associated”, because you think the term should be only applied to specific things that are unique in some other way, so what? People can still adopt the IPCC terminology and usage.
Arthur– Are you seriously complaining that refer to the bit red blog as a tropical tropospheric hot spot?
Does red indicate “hot” or “cold”? Where is the big red hot spot in the figure they discuss? What latitude? Do those latitudes correspond to the tropics? What elevation is the hot spot? Where is the troposphere?
DING DING DING…
Round two ends with Lucia catching Arthur’s attempt to redefine the arguement…THEN SHE BODY SLAMS THE STRAWMAN! Okay kids, no hitting below the belt… but you CAN body slam the strawman.
Lucia educates Arthur on those pretty colors on the graphs and what the X and Y axis stand for… but was she effective in countering Arthur’s arguement that some of the fingerprint is really there!
Lucia misses an opportunity on the “fingerprint” issue. It’s not used in 9.2.2, but in the intro (9.2).
But round two still goes to Lucia despite her kind and gentile (or weak) response in round two.
Mike– I am currently watching tv with hubby. If I spend more than 20 seconds on a response, he will howl. Gotta go!
Intermission…
http://www.youtube.com/watch?v=oKU0HPOWhUc&feature=related
I don’t know how we can assess whether the climate models and the theory are accurate if the goalposts keep getting changed.
Now the tropical tropospheric warm spot is not required. This is goalpost move number 43 so far. What’s next?
We need to be able to go back and assess whether the models are giving us results that we can rely on.
The warmers need to publish public forecasts (not spaghetti diagrams) like Hansen did in 1988 so we can all gain some trust or jointly come to a conclusion to start over. Goalpost moving is not going to get that done.
I’m confused.
Is Arthur saying that the world has warmed but it is not due to GHGs?
Is he saying that the lack of a hotspot should not be of concern to models projecting warming?
Arthur, what are you arguing, and why?
Are you carrying over a hobby horse?
http://arxiv.org/PS_cache/arxiv/pdf/0811/0811.4600v1.pdf
“I don’t know how we can assess whether the climate models and the theory are accurate if the goalposts keep getting changed.”
Bill—Gavin refuses Roger Pielke’s challenge to construct ANY goal poast or testable case to GCM’s. Don’t you know–any trend is consistent with the ensemble of GCM predictions!
Bob B-
Here is a some more interest games to play with “consistent with”. Earlier this year Real Climate wrote:
http://www.realclimate.org/index.php/archives/2008/02/antarctica-is-cold/langswitch_lang/sw
Just today Nature Climate Feedback reports:
http://blogs.nature.com/climatefeedback/2008/12/agu_2008_evidence_that_antarct.html
The author is RC’s Eric Steig.
What do you want to bet that Antarctic warming is also deemed to be “consistent with” what the models say? 😉
More fun here:
http://sciencepolicy.colorado.edu/prometheus/the-consistent-with-game-on-climate-models-and-the-scientific-method-4334
They do not call this a pattern/fingerprint/signature. The mid-troposphere hot spot is *not referred to in the text* of the IPCC AR4 WG1 in section 9.2.2. The only place you find any discussion of the issue of tropical lapse rates verbally is in section 9.4.4.4, and there it is in no way referred to as a pattern, fingerprint or signature.
Why do you assert “the IPCC calls this a pattern/…” when they never “call” it anything? The “tropical troposphere warming” is not the subject of section 9.2.2 anywhere. The only place you can tell anything about it is from figure 9.1 – so that’s what I just discussed in my comment above: figure 9.1 shows the same “pattern” of tropical mid-troposphere amplification of the surface trend for *all* of the forcings except volcanoes (where contrast is so low it is hard to tell), so it is not a distinctive pattern of one forcing, which is why it is completely ignored in the text. Look at the magnitudes I cited in my previous comment – all of them see an amplification of 1.8 to 3 in T-Trop. And you can obviously see it in Figure 9.1 itself.
Goodness, Lucia, they not only do not use the words “hot spot”, they don’t use the words “big red” either. Talk about putting words in the mouth of the IPCC. Where in the text when they “discuss” the figure do they specifically make mention of anything that can be positively identified as the “big red hot spot”? I don’t see it!
Look at the captions of the figure, and look at each of the individual images. Part (a), solar forcing, 200 – 400 mb, center of the figure, a bright yellow spot, with darker yellow above and below. Tropical mid-troposphere warming. Part (d) ozone, there’s a spot (small one) right in the tropical mid-troposphere. Part (e) aerosols, there’s a dark blue spot right there in the tropical mid-troposphere.
No, it’s not “big red”, but it is there in every image, at the exact right appropriate scale to follow the same amplification pattern as the GHG one in part (c).
Now, this gets to the question of whether absolute magnitude, rather than relative magnitude and sign, is a factor in what the IPCC is calling a signature, fingerprint, or pattern here. Normally in English one would think of “absolute scale” and “pattern” as two orthogonal components of the description of an effect, but let’s look at the text of the IPCC to be sure that’s what they mean:
Hmm, no mention of “big”, “red”, or of the significance of the absolute magnitude here – this is a discussion of *relative* effects, and their signs. Specifically it talks about the troposphere – the whole troposphere, not just the tropical mid-troposphere, and cooling in the stratosphere – i.e. a negative *-Strat, and “more warming near the surface in the NH” – i.e. an amplified N-Surf. Those are the relative patterns identified as specific to GHG’s (although from my comment above you can see the same patterns are there for the ozone component of forcing as well).
i.e. positive response in most places with north pole amplification, the relative patterns I mentioned above as specific to the aerosol pattern from the figure.
i.e. even more consistently positive response (with north pole amplification) expected in this case than from the sulphate aerosols (“throughout” the atmosphere vs. “throughout most”); IPCC here doesn’t seem to acknowledge the polar stratospheric negative response to solar forcing that you see in figure 9.1a, but perhaps that’s not a “robust” feature worth discussing. Again, the pattern in question is one of relative, not absolute magnitude.
The pattern is once again only about relative responses – enhanced either way!
– a discussion of temporal as well as spatial “pattern” here – but caused by a change in relativestrength of forcings resulting in a change in the spatial pattern of response (SH vs NH). Once again pattern is referring to relative, not absolute behavior.
Then we get to 9.2.2.3 and the perhaps confusing:
This talks about the magnitude – but this is about a fitting procedure, where the magnitude of different patterns is adjusted to try to fit observations. Magnitude and pattern are here clearly two independent, orthogonal variables, in describing temperature response.
Recall some old mathematics – vector spaces. The observations – actual values for the temperature profile of figure 9.1, T-Surf, T-Trop, T-Strat, etc. form a multi-dimensional space – 9 dimensions in this simplification. The “patterns” consist of a variety of vectors in that space, roughly the ones I listed in my comment above. We only have 5 distinct patterns to choose among (the 5 forcings shown), but we can do our best to try to reach any given observation by adjusting the magnitudes along the 5 vectors to get as close as possible. That would be the best fit.
The *direction* of these vectors is the pattern. The *distance* along these vectors in the match to observations is the magnitude. Figure 9.1 doesn’t normalize the different vectors, it shows them effectively as they would be for a fit to the past century’s warming. The RealClimate post referenced above was an attempt to normalize the vectors to the same magnitude, taking the magnitude piece out of the picture and leaving only the pattern/signature/fingerprint. That is why Lucia’s whole post here is so wrong-headed.
I will claim that chapter 9 in no way claims that the tropospheric hot spot is a GHG fingerprint.
1) I think it is pretty clear that in the sentence you underline from page 702 the fingerprint is a warming troposphere versus a cooling stratosphere: not, as you seem to suggest, a mid-troposphere versus surface difference.
2) On “hypothetical scenarios”. To use the fingerprint analogy: presume that I have two model fingerprints from two suspects, and one real data fingerprint. One suspect pressed down hard leaving a dark fingerprint, and the other suspect pressed down lightly, leaving a light fingerprint. My real data fingerprint is light. Does that mean that suspect two is the criminal? Hardly! In order to fingerprint, you aren’t looking at the magnitude (darkness) of the fingerprint, you are looking at the shape and swirls. It might be easier to compare if you take digital photographs and then scale all the images so they are the same average darkness. This is the equivalent of boosting solar forcing in the model so that you get the same total warming of the system (or perhaps, the same surface warming) and looking at how the patterns differ (boosting solar forcing might be the equivalent of assuming that the solar signal causes much larger response in the climate system than the GHG signal, for some unknown reason).
Mike points out that “fingerprint” is also used in the first part of 9.2. Let’s look at the quote: “In such studies, a climate model is used to calculate response patterns (‘fingerprints’) for individual forcings or sets of forcings, which are then combined linearly to provide the best fit to the observations. This procedure assumes that the amplitude of the large-scale pattern of response scales linearly with the forcing, and that patterns from different forcings can be added to obtain the total response.” And from Appendix 9.A “The vector a accounts for possible errors in the amplitude of the external forcing and the amplitude of the climate model’s response by scaling the signal patterns to best match the observations.”
Note the use of the word “scales”. Basically, this sentence is saying that we take various patterns, and if we are uncertain about the magnitude of response, we see if a different scaling of each individual pattern might better fit the sum. 9.2 does not say that the magnitude of the heating is a fingerprint, scaling solar up like RC does fits perfectly well with 9.2.
3) 9.2.2: The solar figure shows mid-troposphere warming faster than the surface. The GHG figure also shows that. The feature of mid-troposphere warming more than the surface is robust across a variety of models, but also across multiple forcings. Therefore, it can’t be a fingerprint for telling GHG vs. solar forcing, and 9.2 does not claim it is.
4) So, to sum up: cooling stratosphere versus warming troposphere: fingerprint of GHG vs. solar. (note that there might be other explanations: we haven’t ruled out a third suspect who happens to have that same fingerprint, but we also have more evidence pointing toward the GHG suspect than this one fingerprint feature). Lack of tropospheric hotspot? Not a fingerprint differentiating GHG and solar, but possibly a warning sign that we might be getting something wrong somewhere, but we don’t know what yet. And the rest of the RC post discusses whether the lack of hotspot is actually outside model bounds yet, given different observational datasets and modeled natural variability.
More quoting: “Greenhouse gas forcing is expected to produce warming in
the troposphere, cooling in the stratosphere…” next paragraph: “The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings described above. Solar forcing results in a general warming of the atmosphere (Figure 9.1a) with a pattern of surface warming that is similar to that expected from greenhouse gas warming, but in contrast to the response to greenhouse warming, the simulated solar-forced warming extends throughout the atmosphere”
Again. Fairly clear, in 9.2.2, that the “fingerprint” difference between GHGs and solar is the stratosphere/troposphere comparison, not the mid-trop./surface comparison.
Dan Klein:
I am trying to clarify what the IPCC said here, and what they did not say. Lucia is claiming the IPCC said things in AR4 WG1 section 9.2.2 which it clearly did not say, and seem to be based on misinterpretations that came from some outside sources – or else where did she get the idea this section discussed a “fingerprint” of a “tropical troposphere hot spot” when those terms are not even used in that section?
If you accept the IPCC reports, then clearly the world has warmed, and it can be almost entirely attributed to anthropogenic GHG’s. Since the IPCC said essentially nothing about the hotspot or lack thereof (it is very briefly discussed in section 9.4.4.4) it is clearly, as far as IPCC is concerned, not something of great concern to the analysis or models. The prediction of enhanced tropical mid-tropospheric warming is unequivocal, and as section 9.4.4.4 points out, amplification of short-term tropical temperature variations in the mid-troposphere is a very well-observed phenomenon (RSS satellite data in particular shows this clearly). There are some real long-term data calibration issues that have made it hard to tell what is really going on there though – if the data really get better and there’s still none of the predicted enhancement, then certainly it would be of more concern as far as our general understanding of the atmosphere goes.
I have seen no evidence that any climate scientist was ever touting the tropical mid-troposphere hot-spot as an essential “fingerprint” for GHG warming, so this is certainly not an issue of moving the goalposts. Some contrarians evidently misunderstood the issue or blew it out of proportion, but nothing has changed in the theoretical understanding of this issue at all in recent years. The changes have rather been in the continued evolution of the measurements and analysis.
When the measurements finally do prove the existence of the hot-spot, what will you folks do then, I wonder? Move on to some other red herring, I suppose. Oh well.
Round three begins with a post from Arthur which may be targeted to the colorblind… and it’s lengthy nature may be a clever tactic to keep Lucia in front of the TV where she does not want her hubby to howl (an inherent contradiction in some circles).
Arthur–
Honestly, where are you going here?
and
What are you talking about?
Figure 9.1 is discussed in the second sentence of the first paragraph of 9.2.2.1, the first subsection of 9.2.2 The section title indicates they are discussing “patterns” or response. This appears on page 672. The figure hows a a big red blob in the tropical troposphere is discussed in section 9.2.2 called . That big red blob represents a hot spot in the tropical troposphere, and is discussed in the first paragraph of 9.2.2.1 paragraphs and several other paragraphs. For example, paragraph 1 of section 9.2.2.1 says
“These figures indicate that the modeled vertical and zonal average signature should depend on the forcings.”
After this sentence introducing the general topic of “signatures”, the authors proceed to discuss specific “signatures”.
Paragraph 2 describes the singature for GHG thusly:
“Greenhouse gas is expected to produce warming in the troposphere….. (Figure 9.1c)”. The sentence specifically calls out the GHG panel 9.1 C and attributes the big red blob (hot spot) in the tropical troposphere to GHGs. Ozone– an anthropogenic issue– has is also mentioned later in that paragraph.
I don’t know how you can call this “not discussing” this feature or not identifying it as a signature. Is your concern that discussed the shape, relative warmth and location by referring to a speciric figure instead of using the specific word “hot spot”?!
Are you concerned with whether or not they give the feature a specific name? Call it Fred. Call it Ethyl. Call it the big red blob — spot what have you– indicating warmth in the tropical tropospheres. The authors are calling out warmth in the troposphere as a signature of warming by GHGs, and they do that in section 9.2.2!
After that, I have no idea where you are going. On this, which I previously quoted and you now quote:
Note: in contrast the solar warming extends throughout the atmosphere. You may be interpreting the yellow swiggle in figure 9.1a discussing solar as some sort of spot, but the IPCC authors text describes this warmth as a) different from the GHG pattern and b) extending throughout the atmosphere!
I’m curious about so many things in your comment! Why are you vehemently explaining the IPCC doesn’t don’t discuss the mid-troposphere in that section? Why do you think that rebutts anything I (or anyone) has said? And for much of the other stuff… is your point that the IPCC also discuss effects of other things in that section? No on has denied they discuss more than the effect of GHG’s in that section.
Based on the bulk of the post, the RC post was purportedly a critique of Douglass. Douglas compared some projections that anticipated the vertical profile would look more or less like that in figure 9.1.e and purported to show that profile did not match the data.
For some mysterious reason, RC decided to go off on a tangent explaining that that feature would be predicted even if warming was due to solar forcing provided solar forcing achieved an entirely fictional level no one believes possible. So?
And now, based on that RC post, and RC’s odd attempt to redefine the criterian the IPCC used to identify a signature, pattern or fingerprint of warming, people are trying to claim that the features the IPCC discuss as signature, patterns or fingerprints of AGW should not be called signature, patterns or fingerprints of AGW.
Marcus–
I’m not suggesting anything much different about the stratosphere.
In discussing the surface warming, I was engaging Arthur’s discussion of the similarities between solar and ghg warming in the IPPC figure. The text of the IPCC document tells us the solar warming affects the entire atmosphere, and so looks different from the ghg warming. My reference to the surface was intended to simply convey that we would be experiencing warming at the surface. I did fail to mention the solar shows no cooling in the stratosphere, but I didn’t mean to suggest anything about the stratosphere by not discussing it. Sorry if that wasn’t clear.
On the solar thing: you might what to look at figure 9.1a again: there is a red spot on the surface in antarctic. The entire southern hemisphere warms just as quickly as the the mid-troposphere. The stratosphere warms just as much as the surface. There some pretty darn warm areas far aloft in the northern hemisphere. The sideways “s” shape for the bright yellow warm region in that specific figure is distinctly different from 9.1c. However you interpret this, the IPCC used the term “in contrast” to highlight the fact that they think the fingerprint for solar and GHG warming are different from each other.
Of course, the IPCC may be wrong to say “in contrast”. They may have selected inappropriate figures. But even if you put an enhance colors filter on the solar figure (9.1a), it doesn’t have the same “pattern” as the GHG figure (9.1c). Or at least, the solar pattern doesn’t look like a dimmed version of the GHG pattern to me!
Arthur-
What? Huh?
You say, “I have seen no evidence that any climate scientist was ever touting the tropical mid-troposphere hot-spot as an essential “fingerprint†for GHG warming”
Then you say, “When the measurements finally do prove the existence of the hot-spot, what will you folks do then, I wonder?”
So there is no claim to a hot spot, but when the hot spot is observed, as you think it will, it’ll then prove what?
How weird.
The signature pattern continues… round 3 to Lucia… was close to a TKO. Arthur, after being soundly defeated in the first two rounds, is resorting to lawyerly word games (do they really call it a hot spot, or, is it really that hot?).
The issue here is why do models predict certain temperature variations in the atmosphere when the surface warms… but the observational evidence indicates otherwise. Is it because the models don’t have it all together yet? Since most of the terrible predictions of future death and destruction rest on evidence provided by these computer models, one might fathom the likelyhood that the models should accurately predict present day conditions. dun-dun-DUNNNNNNN. Of course, those whose reputations and careers rest on the accuracy of the models might be expected to play lawyer in order to defend them when they fail.
In the above refrenced RC post, Gavin shows the modeled results of surface warming using his GISS model. There are some differences in his predicted results if the warming were solar and if the warming were caused by GHG’s. Both show that familliar hot spot (or Ethyl, red blob or Fred if you prefer) The problem with what Gavin illustrates is that neither of his model predictions equal observation. Thus it is easy to conclude that Gavin needs to get back to the drawing board.
In the mean time… more intermission
http://www.youtube.com/watch?v=LMGpJX_Y7lU&feature=related
Arthur–
Then you say, “When the measurements finally do prove the existence of the hot-spot, what will you folks do then, I wonder?
Who is “you folks”?
Based on the text in the AR4, I would say observation of hot spot in the tropical troposphere with a similar configuration to that shown in figure 9.1c supported the model simulations which predicted it would arise as a result if increases in GHG. In otherwords: Because the hot spot is called out as a fingerprint using the definition in the AR4, and it looks different from the other fingerprints, an observation of the predicted fingerprint would support the theory that GHG’s specifically caused the warming.
If the hot spot is ever confirmed by observations, what will you say? Will you say the hot spot is also a fingerprint of warming due to solar forcing and is consistent with the theory solar forcing increased in some anti-maunder-minimum? Or that the hot spot is not inconsistent with any theory of warming including cosmic rays? After all, it’s just a consequence of the adiabatic lapse rate, right?
Now that you are calling it a hot spot, can I continue to do so? Or must I call it Fred or Ethyl or something?
Lucia: first, you are confusing the Southern Hemisphere and the Northern Hemisphere. Both a and c show Northern Hemisphere warming more than the Southern Hemisphere (due to more land mass, and more Arctic feedback). (Left is North, Right is South – the RealClimate figures are mirror imaged, and have North on the right).
And I apologize, I didn’t mean to imply that you were arguing against a stratospheric vs. tropospheric fingerprint. I meant to point out that the IPCC concentrates in the text on this troposphere vs. stratosphere fingerprint, not anything else. “in constrast” and “entire atmosphere” refers to the warming of both strat + trop, “in constrast” to GHGs which warm only trop and cool strat.
And I agree, 9a and 9c would not be identical even if they were scaled to some equivalent value, but I would argue that the slightly warm blob in the tropical mid-trop in 9a would scale to a nice red – thereby creating a solar induced “hot spot” that would be similar (not identical) to the GHG “hot spot”. Hypothetically, you could use other features to distinguish between the two, but “robust” features that could have been used in that way would hopefully have been highlighted in the text.
And I would argue that the authors of Ch. 9 never referred in the text to that “extra warm” mid-trop tropical temperature as a signature of GHG versus solar, but did mention several times the stratospheric difference between the two.
“When the measurements finally do prove the existence of the hot-spot, what will you folks do then, I wonder?”
Is there a plan in the works? Arthur, does your climate model come equipped with a crystal ball? Arthur, have you been talking to god again?
Arthur–
Huh? Some Douglas et all noticed that, as you say “The prediction of enhanced tropical mid-tropospheric warming is unequivocal”. The compared this unequivocal predictions to observations.
You agree the predicted mid-tropospheric hot spot is unequivocal but then tell us Dougless et al somehow misunderstood something related to that unequivocal prediction. Would that be anything that would minimize the impact of showing the the hot spot did not exists had they succeeded in doing so?
While the Douglas data comparison has flaws they compared the observations to a something that was actually predicted and which is described in the section of the AR4 discussing patterns and signatures– sometimes called fingerprints.
Prior to figuring out the real problem with Douglas’s post, the team at RC wrote a hurried post whose purpose appears to be to introduce a distraction. The distraction was to tell the world that some aspect of the feature tested by Douglas would be predicted even if the warming were due to utterly unrealistic increases in solar forcing.
Whether or not RC was correct on that red-herring issue and/or the IPCC entirely agrees with their interpretation, the RC discussion was simply a meaningless distraction diverting attention from the actual agument in Douglas which was: The hot spot, which is predicted to exist does not exist.
Showing a predicted feature is not there may not tells us what the flaw is but it points out that some flaw exists. Even if the models predicted the same thing with increased solar and the feature wasn’t there, then the models would still be wrong.
The problem with Douglas is not has nothing to do with whether or not anything is a fingerprint. The problem with Douglas is that didn’t account for the uncertainty in the measurements! So, they failed to show the hot spot does not exist.
FWIW: Your argument that section 9.2.2 doesn’t discuss hotspot in the troposphere and/or doesn’t discuss it as a fingerprint is nothing short of amazing.
Can anyone suggest what a plate might look like for surface albedo changes (it was not modelled in section 9 of the IPCC AR4)?
At a rough guess, I would say:
– surface warming, mainly in the Northern Hemisphere;
– stratosphere cooling, due to loss of heating from reflected radiation.
Marcus–
Yes, you are right on the North South thing. Dang conventions!
I don’t know… I still think “whole atmosphere” means “whole atmosphere”. There is very little preferential warming in the figure and the text says “whole atmosphere”.
I don’t see how you don’t read that section as telling us the extra warm red blob in 9.1f arises as a result of the GHG’s.
The way I read it, the red blob in figure 9.1f is not claimed to be caused by solar because it cannot arise with solar forcing of the magnitude that is possible on earth. That’s why people refer to the feature as being a fingerprint of GHG’s not solar.
Arthur and others can keep repeating that the blob could be caused by solar forcing solar forcing varied much much more than is possible. But solar forcing falls in a category of “any and all forcings” that simply cannot vary that much.
We know this. We have direct measurements of solar forcing– so, as a practical matter, the hot spot can’t be due to solar forcing. Saying a hot spot of that magnitude can be due could be part of a “pattern” , “sign” or “fingerprint” that the warming is due to solar forcing doesn’t make any sense to me.
Now, given the fact that I consider the possible magnitude of the pattern to be included in the idea of a “fingerprint”, I’ll admit the term “fingerprint” may not be the best possible term. I understand the use to be in the lines of “firm evidence that points to this specific cause”. I don’t see it as literally “a fingerprint that might be pressed very hard or very lightly”.
But, the fact that I don’t think of “light” fingerprints and “heavy” fingerprints just makes analogy with human fingerprints imperfect. But its imperfect in other cases too.
For example: Santer calls the rise in the tropopause a fingerprint of warming. If an observations of the rise mis-matched the predicted value by an order of magnitude, would you say you matched the fingerprint? Or not?
If there was a mismatch, I’d say the observed rise did not match the fingerprint! In this analogy, I think the magnitude of an effect associated with a level of expected forcing is part of the fingerprint/symptom/pattern or whatever word you prefer.
So, I don’t consider the hot spot a possible fingerprint of solar warming.
Lucia, the absolute magnitude is NOT part of the fingerprint, according to how it is commonly used by practitioners. Including magnitude would make fingerprinting superfluous. If we were to agree a priori that solar radiation were insufficient to produce the observed magnitude of the hot spot, then we would also agree that solar is insufficient to produce the observed change in global mean temperatures, and there would be no need for a fingerprint analysis. Arthur and Marcus have it right; there is no conflict between the RC entry and the IPCC.
The magnitude of the observed temperature change, and the inability of models to produce that magnitude with any forcing other than GHG, is one piece of evidence in support of AGW. The pattern of the observed temperature changes, with a warm troposphere, a cool lower stratosphere, and an even cooler middle stratosphere, and the ability of models to reproduce this pattern with GHG forcing and not with other forcings, and a firm theoretical basis for this distinctive GHG temperature pattern, is a separate piece of evidence in support of AGW.
JohnNg–
First I have little objection to anything after the first paragraphs, but they are irrelevant to explaining what is or is not a print. They are only discussions of whether or not we know that GHG’s cause warming.
But
1) Even if we knew solar didn’t cause the warming, we would need fingerprint analysis. There are people who say the warming is caused by “natural” variations, and whether you like their arguments or not, they are not necessarily claiming it’s solar. Consider Christy’s paper on cloud feedback. Or consider the discussions of the PDO. Or other cycles. Fingerprint analysis would be required to distinguish between external forcing and noise.
2)The absolute magnitude of the human fingerprints is not a parameter for identifying a person. But if you use the fingerprint detection techniques used as vectors pointed to by both Arthur and Marcus, it would seem it has to be.
3) There is a conflict between the RC and the IPCC in so far as RC is purporting to comment on Douglas’s decision to compare observations to predictions. The structure of their article would suggest that the results of that comparison wouldn’t tell us much about warming, and the reason it wouldn’t is the predicted hot spot might be due to solar. The IPCC attributes any exisiting hot spot to GHG’s. Their document indicates absolutely does not predict that spot for the level of solar forcing experienced by the real earth. In this sense– which at a minimum applies to discussing Douglas– a detectable hot spot is not and cannot be caused by solar (according to IPCC section 9.2)
If you don’t know the factor or proportionality between the strength of the a GHG print and it’s pattern or a Solar print and it’s pattern, how do you figure out the proportion of warming caused by GHG, Solar and how much is unexplained? Equally important, even if you scale by surface warming, you do you do any fiduciary checks?
The word “fingerprint” is used as an analogy with police detective work and it’s not perfect. I think the way it is used– as consistent with the vector additive arguments and methods of detection discussed in later parts of the document– with respect to the warming prints, the magnitude has to be part of the print. With humans it doesn’t.
(Oddly,–and irrelevant to normal police practice and to a large extent, this argument– I’m suspect human prints do have the equivalent of magnitude. With those electronic detection devices used to let you in building, the strength is an adjustable parameter. I know this because I have very light fingerprints. When they installed those things into a building at PNNL, the had us all enter the prints on a first round. Then people who couldn’t get in regularly went to a second round, and they changed the detection threshold for us, explaining we had light prints. FWIW, the group consisted of women only. Then they had a third round for the one person who needed it fiddled. That person was me! The grocery store got a fingerprint pay system. I decided to use that, and the operators said they never found someone who was harder to enter than me. So, since I can see the ridges on my thumb are deeper, I switched to the thumb. Presto! )
information for you..
=========================================================
Snow in Vegas??
Praise Al Gore!!!
I’m so happy that most of the world has listened to him because we now see the end of global warming. All of Al Gore’s hard work has paid off. Why, things have gotten so good that global warming has completely reversed itself and it is now snowing in Las Vegas.
And just so you don’t think it is only Sin City that has been wonderfully blessed by Al Gore’s pious efforts, it has snowed in Houston, New Orleans, and Mississippi. (OK, technically it was only sleet in Mississippi).
The world has been told for the last few years that we are all going to be drowning under the antarctic ice melt, but this year is the coldest year in the last decade. Of course a look at the graph shows a scary comparison between the last 2 decades and the 1800’s. Well, there was this little thing called the “little ice age” that ended in the 1800’s, which caused lower than average temperatures.
read more…
http://hernadi-key.blogspot.com
=========================================================
Yeah. I’ve made the north-south mistake a number of times in my research. It is annoying when different code uses different conventions…
I built a little chart that shows the all the temperature observations (GISS, Hadcrut3, RSS, and UAH) versus CO2 back to the start of the record for each temp series.
I think this one demonstrates that we are far off the predictions of the warming models.
http://img254.imageshack.us/img254/2626/tempobsrvvsco2ct4.png
Basically, the global warming model predictions line keeps getting pushed out farther and farther every time they put out a new series of forecasts (which incorporates the fact that actual temps to date are not keeping up with the trendline expected).
The global warming theory trendline in the chart is very close to the one Hansen used for Scenario B – the increase from 1960 in the line is 0.80C up to 2008, while Hansen had projected 0.85C to 2008.
Bill, I’m not sure if using the satellite record in that respect is appropriate since most of the record was created during a period of warm ocean cycles.
Now can we get back to the question of whether Ethyl is hot?
Mike C, I have done the exact same charts with the ocean cycle influences taken out. Since some people didn’t trust this method, I did it with just the unadjusted raw observations this time.
It actually doesn’t make a lot of difference, there is just less variation in the observation scatterplot.
Here is just Hacrut3 adjusted for the ENSO and the AMO.
http://img132.imageshack.us/img132/3930/hadcrut3scattervj2.png
Bill, it’s a little difficult to examine your graph with the dates like they are.
Since Bill brought up Scenario B, I figure this is as good a place as any to post this. I’ve updated Hansen’s figure from 2005 to include current data, including the land-only “climatological year” anomaly for 2008. (This particular rendition used only station data).
http://img201.imageshack.us/my.php?image=hansen2008qa4.jpg
Any changes from Dec 2008 for the 2008 data are likely to be small, or at least smaller than the size of my asterisk (don’t read that out loud).
(Note added in proof: Hmmm, that 2008 point looks a little lower than it did in my original. For the record, the D-N anomaly for station data is currently listed as 0.55 in the GISS database).
Mike C, I can give you a little zoom-in for the tropics only covering only the satellite record years, but its Lucia’s blog and I think she has to give me permission.
It does show why there is no tropical troposphere hotspot. There is not enough warming in the tropics to produce one.
John M, I’d like to see the Land-Ocean chart as well. Good stuff.
Bill– if it’s not pornographic, you can post images!
Yikes!
What happened to my link? I thought I was being safe by posting a link instead of using an html command.
Oh well, so much for the precautionary principle. 🙂
Bill, I’ve taken the lazy way out and just added asterisks to Hansen’s original plot. Not only did that save me time, but I can’t be accused of distorting his data treatment. I’ll have to find a version with the land-ocean plot and add to it as well. Shouldn’t be too hard.
JohnM– it’s back. I was trying to wrap it in html so the image would show. It must be htaccess blocked from hotlinking. The links shows again.
Here is the Tropics RSS (20S 20N) and Hadcrut3 (30S 30N) temperatures from 1979 to Nov. 2008 versus a model based on the ENSO, the AMO and ln(CO2).
http://img205.imageshack.us/img205/8008/rsstropicsnw0.png
http://img249.imageshack.us/img249/4659/hadcrut3tropicscg7.png
I think it is pretty clear the ENSO (at least) is driving these temp series. If you leave out the AMO, it is still a pretty good match but the AMO certainly improves the fit. (Note that the ENSO and the AMO index cover about 7% of the dataset for the tropics so the degrees of freedom are dropping rapidly here but it does explain up to 70% of the variation.)
When you pull out the ENSO and AMO influence, this is a warming signal left over (still noisy but the big 1998 El Nino bump is smoothed out as well as the 2007-08 La Nina drop.)
There is only 0.045C per decade warming in the tropics when you pull these influences out and the tropics are only heading for 0.66C of warming per doubling. No significant warming, no hotspot.
http://img212.imageshack.us/img212/5963/hadcrut3tropicswarmingwt1.png
http://img510.imageshack.us/img510/3549/rsstropicswarmingob9.png
Thanks Lucia, not sure what I did wrong over at imageshack.
Anyway, back to Bill’s question, didn’t take me long to end up here.
http://scienceblogs.com/illconsidered/2006/04/hansen-has-been-wrong-before.php
Take a good hard look at that figure borrowed from the silly wabbit. In particular, look at the anchoring points for those big a$$ squares for the station-only data especially for 2007. Here are the J-D and D-N from the GISS dataset (note that the value of 74 for J-D is what I’ve always had for the 2007 station only data.)
Global station data…….Land-ocean data
……..J-D….D-N……….J-D…..D-N
2005…76…..76…………62…….61
2006…66…..65…………54…….54
2007…74…..75…………57…….59
2008…NA….55…………NA……43
And I was worried about where my asterisks were landing.
Of course, I guess we have to assume all the symbols have the same anchoring points, including the Scenario B baby blue triangles.
It’d be nice to have Hansen’s actual data for his Scenario’s.
Anyway, be that as it may here is my ham-handed MS Paint adaptation of the Coby/Rabbit graph (from above link), with 2008 D-N added.
http://img201.imageshack.us/img201/4387/hansen2008allua4.jpg
Oh heck, that darn asterisk looks low again. Oh well, everyone has the data now.
Arthur and Lucia:
It seems to me that the focus should be specifically directed to the differences between the patch (spot, blob), from 30N to 30S and from 400 to 150 hPa (the upper tropical troposphere) –––and the patch from 30N to 30S from the surface (1000) to 400 hPa (the lower tropical troposphere).
Santer et al., in their most recent criticism of Douglass et al., Figure 6, show clearly that a few radiosonde data (IUK, RATPAC-A, and HadAT2) show relatively uniform temperatures from the surface up to 400 hPa, and then the temperatures steadily drop by about 0.2 degrees up to 150 hPa.
They (and the IPCC, Fig. 9.1c and f) show that the models give about 0.5 degrees higher temperature in the upper troposphere (the “hot spot”) than in the lower troposphere. The same kind of model trend (at reduced amplitude) is suggested for solar warming in Figure 9.1a.
Whether the latter differences are, or are not claimed to be “signatures” or “finger prints” of warming in general, or of GHGs, in particular seems much less important than the question of whether this discrepancy between observation (these sondes) and the models should receive some more scrutiny. Pat Keating apparently thought so, and seems to argue, with some force, that the discrepancy is probably due to modeling ‘errors’.
This surely deserves more attention.
JohnM,
The actual correct scenario should be between A and B after 1992 because of the date of the volcano vs the date of the simulated volcano in scenario B.
Lucia, thanks for the entertainment. Some of these guys should get up and adjust there underwear once in awhile. They’re wound a little too tight.
Either Lucia knocked em all out or they took their grant money to the bar.
What a great thread. Full of humour and interesting factoids.
I still don’t understand why it is so important to claim that this tropical tropospheric hotspot is not a signature for AGW. Surely it is a part of the whole pattern. I will start thinking about whether the troposphere part might be caused by solar when a) we find the hot spot and b) when TSI increases substantially.
Of course, if someone had had the foresight to take accurate measurements of temperatures, by latitude and altitude, in 1890 it would have been easier to verify the hindcast shown in Fig. 9.1 above.
The use of anomalies always seems to lead to linguistic curiosities like calling cold hot. Just like the hot spot in Siberia is actually damn cold, so the troposphere is really colder than the surface.
My thanks to all the participants for a delightful evening.
MikeC–Comments always drop on weekends. Plus, at some point, everyone is permitted to agree to disagree. You really can’t call wins and loses at blogs.
Though, there is one sort of blog spat that can be a win for a blogger. It’s getting trolls to leave without banning them. I’ve done that in the past but it doesn’t apply to this thread. 🙂
Jorge–
Yes. Even if the models predict a hot spot could be caused by the sun going super-nova as well as CO2 doubling, if either occurs and the hot spot does not materialize, that shows a deficiency in the models ability to simulate the full earth’s climate.
Would the deficiency be “important”? I guess that depends on what you wish to learn from models.
There isn’t. Best way to test this to everyone’s satisfaction would be to email one of the lead authors. Joyce E. Penner was the lead author from the USA and would be a good first choice. I won’t publish her email here, but it should be easy to find.
May I propose a bet? Loser donates to Malaria no More or some other charity?
Hmmm, looks like this is getting into an undecidable debate about “what the meaning of is is”.
Can we step back a bit and see if everyone can agree with this relatively simple language:
https://publicaffairs.llnl.gov/news/news_releases/2008/NR-08-10-05p.html
Boris–
There is a conflict in a sense that cannot be resolved by a lead author.
When addressing the issue in Douglas, the authors at RC decided to change the subject to an entirely unrealted issue. In Douglas, the authors are testing the types of signatures actually discussed in the report. Those at RC want to make it appear they must or are testing some other sorts of signatures.
What I am saying is the manner in which RC presents the information shifts the conversations in a way that would make their discussion seem inconsistent with the IPCC’s discussion of what is actually expected based on real honest to goodness forcings.
I have agreed multiple times in comments here that the lead authors of the IPCC’s article could presumably tell us there is no conflict with the signatures being as discussed in the chapter and those discussed for fictional forcings– as in the RC article. But that’s not the issue I have with RC, or what I make by saying the is a seeming contradiction.
The seemign contradiction lies in the tendentious framing of the discussion at RC. RC’s littel discussion of what would happen under hypothetical forcing is wedged into a criticism of Douglas et al. comparison of observations to model prediction.
When Douglas compares obsevations to model predictions, rationally, they must compare to predictions for forcings that exist. Obviously, whatever they discuss ought to be considered in the light of forcings that can occur on earth.
Model predictions basd on what can occur on earth
are discussed in the IPCC. These accord with the features discussed by Douglas and which Douglas compares to data.
For reasons known only to the mind of the RC authors, in their critique bring up the utterly irrelevant issue of what the models would predict under ficticious circumstances. The do this without neither linking to the IPCC report (which might cause readers to notice the diagrams that agree with features described by Douglas; readers might then ask and get clarification.) They discuss the irrelevant response to hypothetical forcing without reminding readers that Douglas is comparing to observations and observations must be compared to predictions using the same forcings. (If Douglas did otherwise, they would be making an apple to oranges comparison.)
So, through this sleight of hand, RC makes it appear that what they are discussing is relevant comparing observations to model projections and that Douglas is the who is somehow misunderstanding what the models predict.
The fact that RC surrounds their sleight of hand switch with some correct physical arguments, and correct information about what models would predict those features if the sun went supernova is irrelevant to the fact that in the context of their article the IPCC graphs are the approriate ones to use.
By suggesting that their own figures are relevant to criticsim of the Douglas article the RC information would seem to contradict IPCC information which provides the figures that actuall are relevant to the Douglas article.
If our disagreement were over the physics or the truth of the irrelevant information in the RC article, contacting the lead IPCC author might be meaningful. But that’s not the crux of our disagreement.
The appearance of disagreement with the IPCC discussion arises from RC
s article implying that the counterfactual, hypothetical response is relevant to the Douglas et al comparison of model projections to data. It’s not. The IPCC figures are the relevant ones, and they look entirely different from those at RC.
Lucia,
So you’re saying they didn’t do a Saturday night of drinking after a Friday night of beating?
Lucia,
Here is what Douglass said about his paper:
So, showing that the trop/trop hotspot is not a “characteristic fingerprint” of GHG warming, as Real Climate does, is entirely appropriate. Now, would you care to email a lead author and confirm that a trop/trop hotspot is not a “characterstic fingerprint” of GHG warming, or would you like to keep posting–what was it, “twaddle”?
Boris–
What’s your point?
Are you suggesting the image in figure 9.1 f cannot be referred to as a “characteristic fingerprint” associated with greenhouse warming? Of course it may. Are you suggesting features in that figure cannot be referred to a characteristic fingerprints? Are you suggesting there is only one fingerprint?
I have 10 finger and 10 fingerprints. Google fingerprints for global warming and you will find this term is applied to only one feature.
Correction to #7431:
“Santer et al., in their most recent criticism of Douglass et al., Figure 6, show clearly that a few radiosonde data (IUK, RATPAC-A, and HadAT2) show relatively uniform temperatures from the surface up to 400 hPa, and then the temperatures steadily drop by about 0.2 degrees up to 150 hPa.”
Should read: Santer et al., in their most recent criticism of Douglass et al., Figure 6, show clearly that a few radiosonde data (IUK, RATPAC-A, and HadAT2) show relatively small temperature-anomalies from the surface up to 400 hPa, and then the radiosonde temperatures steadily drop by about 0.2 degrees up to 150 hPa.
The hotspot is not characteristic of greenhouse warming. The cooling stratosphere is.
In other words, we have observed the distinguishing feature of an enhanced greenhouse effect. We have not observed a feature common to all types of warming.
This is as simply as I can put it.
lucia,
Down here in the comments you’re trying to be subtle and nuanced about your problems with the RC article. Apparently your issue is just that they weren’t fair to Douglass. It’s very chivalrous.
However, up at the top of the page you’re claiming that the RC results with a large (hypothetical) solar forcing contradict the IPCC results with a small (actual) solar forcing. Different experiments. Different results. No contradiction, despite your anger with Gavin. You don’t even mention Douglass until the 14th comment.
Now on to the issue of fingerprints. Yes, you have ten, but they all have the same pattern. Any one of them can be used to identify you. In this context, a fingerprint is an identifying characteristic. With that definition in mind:
Does the existence of a tropical troposphere (TT) hotspot by itself distinguish between sources of warming? No.
Is a TT hotspot expected for most sources of warming? Yes.
Is a TT hotspot a fingerprint of GHG-induced warming? No.
Boris–
1) Figure 9.1 C showing the effect of GHG’s shows the hotspot. Why you think this is not a characteristic of GHG’s when it arises from GHGs I do not know. One cannot have a hot spot without having cooling or at least non-heating on the opposite sides of the hot spot. So, stratospheric cooling makes the hotspot more of a “spot”. In contrast, solar warming does not have a “spot” because the warming is– according to the IPCC– spread more fully over the atmosphere.
JohnV– Why do you think my only issue is they aren’t fair to Douglas?
I said the structure of their discussoin would seem to contradict the IPCC. Due to it’s structure it does. Let’s change one your question inserting a word and answer the other ones.
Does the existence of a tropical troposphere (TT) hotspot on the earth itself by itself distinguish between sources of warming? Yes. It distigugishes between warming that is due to some claimed natural sources and those due to GHGs. For example: internal fluctuations due to weather.
Is a TT hotspot a fingerprint of GHG-induced warming? Yes. In particular, given the magnitude of forcings that the earth has experienced, the hot spot is expected as a result of GHGs as shown on Figure 9.1C.
So, as you see, I disagree with your answers.
lucia,
I think I *might* understand the point of disagreement. In your response to Boris you ask if a TT hotspot is a “characteristic” of GHG-induced warming. For something to be a “fingerprint”, it needs to be more than a “characteristic”. It needs to be a “distinguishing characteristic”. Is a TT hotspot a “distinguishing characteristic” of GHG-warming. Would the existence of a TT hotspot distinguish between GHG- and solar-induced warming?
yes, now comes the time Lucia plays word games. Yawn.
JohnV–
The term fingerprint is an analogy. As such, the term means that the term “Fingerprints of GHG” is similar to human fingerprints in some respects. Clearly, they do not agree is all respects: For example, there are no litteral fingers. There are no whorls.
Your argument seems to hinge on require the analogy to be overlap human fingerprints some specific amount you prefer, differs in other respects and but you, JohnV prohibit other differences in the analogy (for unknown reasons).
Fingerprints are identifying features. The term fingerprints is sometimes enclosed in “scarequotes” in chapter 9.
The existence of the TT hotspot would distinguishe between GHG expected at the level of GHG changes seen on earth and the amount expected as a result of solar induced warming.
I think I’ve said this several times. I think you understand my point which is associated with magnitudes. My answer to that question isn’t going to change. I’m pretty sure yours isn’t. I think at this point, we both understand each other– but we disagree.
Boris–
Word games? You posted a criticism about word usage. We are discussing word usage.
Yes, exactly Lucia.
Boris says:
“The hotspot is not characteristic of greenhouse warming. The cooling stratosphere is. In other words, we have observed the distinguishing feature of an enhanced greenhouse effect.”
Sorry, no it isn’t since stratospheric cooling is also “characteristic” of ozone depletion.
What a tangled web is being weaved here. Your patience, Lucia, is exemplary.
dover_beach – “Your patience, Lucia, is exemplary.”
I’m not so sure about that… Friday night she was landing blows on some heavy weights… today, it’s like she’s swatting flies
No. Ozone depletion cools the strat. at about 20km. Cooling from CO2 occurs at 40-50km. Bonus points for being both cocky and wrong.
I’ll see your “tangled web” and raise you “A little learning’s a dangerous thing….”
Re: word games
You changed my “distinguishing feature” to “characteristic.”
Having a leg is a characteristic of a woman.
Having an ovary is a distinguishing feature of a woman.
The tropical tropospheric hotspot is a leg.
The cooling stratosphere is an ovary.
We have not observed a leg.
We have observed an ovary.
Douglass’ claim that the troptrop hotspot is a “characteristic fingerprint” is wrong, wrong, wrong. Wrong.
point of clarification, here, Boris,
Is the 20 km altitude at the poles or at the equator?
Lucia,
This has been an interesting debate and recalls an article by Dr David Evans who WAS a modeller for the Australian Greenhouse Office — basicaly the non-appearance of a hot spot has led him to doubt the whole theory.
Here is the link:
http://www.theaustralian.news.com.au/story/0,,24036736-17803,00.html
Andrew
Boris nows says:
“No. Ozone depletion cools the strat. at about 20km. Cooling from CO2 occurs at 40-50km. ”
Whereas he earlier said: “The hotspot is not characteristic of greenhouse warming. The cooling stratosphere is”.
In other words, more word games.
dover_beach:
Clarifications are different than word games.
lucia:
I like to use the dictionary definition of fingerprint. From Dictionary.com:
=====
fin·ger·print (fÄng’gÉ™r-prÄnt’) Pronunciation Key
n.
1. An impression on a surface of the curves formed by the ridges on a fingertip, especially such an impression made in ink and used as a means of identification.
2. A distinctive or identifying mark or characteristic: “the invisible fingerprint that’s used on labels and packaging to sort out genuine products from counterfeits” (Gene G. Marcial).
3.
1. A DNA fingerprint.
2. A chemical fingerprint.
=====
Without “distinct or identifying” a “characteristic” is not a fingerprint. Was this whole thing a mix-up in word definitions?
For those who (unlike yourself) posit that 20th century warming was due to solar effects, the TT hotspot will not resolve anything vis-a-vis solar vs GHG. It is not a distinct or identifying characteristic of GHG-induced warming relative to solar-induced warming.
This is getting really disturbing. How is it possible for a non-native-english speaking person ever to discuss anything climate related topics if the debate is about minuscule nuances of word usage. Has climate research really come to this?
There are so many slightly different theories that even though I have studied natural sciences I just cannot keep up with this. And still there is some sort of consensus claimed by IPCC. But if IPCCs report could be understood in many ways how on earth I, a non-native-english speaking person, am able to understand what IPCC means *shakes head*
This kind of word games disintegrate science.
Boris–
Why are you switching to distinguishing feature from “fingerprint”? You specifically criticized Steve for using fingerprint saying his use was silly. Yet, it’s clearly the usage by the IPCC. I am re-posting conversations there here:
You provided this IPCC definition of “fingerprint” at Climate Audit:
(This is from the glossary http://www.ipcc.ch/glossary/index.htm )
I responded to you there. (See http://www.climateaudit.org/?p=4687#comment-316614 ) I explained the IPCC definition to you as follows:
They don’t say the response must be unique to the forcing. They say it is a response to that specific forcing.
See dictionary example:
http://www.answers.com/topic/specific
“a specific remedy for warts.” is a use of the term specific. This means the remedy specifically treats warts. You won’t need any additional treatment to get rid of the warts. Use this specific remedy by itself, your wart will go away.
The word doesn’t mean remedy treats nothing other than warts but that it will treat warts without the need to use any additional remedy.
A response to the specific forcing of ghg is whatever response arises as a result of ghg even if all other forcing were absent. It doesn’t mean it can’t arise from something else– it means it’s the response to ghgs.
If the IPCC authors wish to define fingerprint differently and to capture the “if and only if” aspect’ of the analogy to fingerprints, they need to get better copy editors. In the meantime, people who read the definition of “fingerprints” given by the IPCC will read it to mean what it says: It is the the pattern that arises when we increase GHGs even if we increase no other forcings. That’s what they show in figure 9.1 C of chapter 9.
The Douglas quote you criticized as using the definition fingerprint inaccurate. As it happens, they used “fingerprint” exactly as defined in the IPCC glossary!
Both at CA and here, you now want to discussed whether or not it’s distinguishing. Why? You can find the answer to the question of whether or not the hot spot distinguishes between sources, how and under what circumstances above.
Andrew–
Failing to find the hot spot will tend to make people doubt. That’s why the issue can be important to the debate. However, it does appear the data are too uncertain to permit us to conclude it’s not there. It may be there. We can’t tell!
That said: PatK’s theory says something different.
John V–
Please read the IPCC definition Boris provided. In their definition, they borrow distinctive, and the fact that that characteristic is specific to the forcing. But that doesn’t make it unique.
This is exactly like definition 2. A mark is distinctive, and intended to indicate the true item. This does not mean the mark will actually be unique. Counterfeiter may duplicate it. (That will be illegal, but they would do it.) However, if the mark is absent it’s not the true goods.
Oddly, similar things can happen with DNA fingerprints. My two brother’s in law are twins and so share identical DNA. But their DNA fingerprints are specific to them, in the sense that if you check the DNA sequence from my head, you can exclude theirs. It does not match the specific pattern that corresponds to them.
All those examples share the feature of being specific to a particular item or person. Most are also unique or nearly so. But two: the manufacturers mark and DNA are not necessarily unique. GHG fingerprints fits in, but would be an example that is specific, but not necesarily unique.
Finding it would not completely resolve the issue for those who do think the warming is caused by the sun.
But this doesn’t mean the spot isn’t a “fingerprint”. It doesn’t make Douglas’s or SteveM’s usages wrong. They use it as the IPCC uses the word.
It also doesn’t mean modelers expect the spot to arise as a result of solar. We all agree models predict it could arise from solar forcing if the sun went super-nova. But given the response anticipated as a result of the know level of forcing does not result in a “hot spot”. Those who predict the “hotspot” predict it to appear because it appears in the fingerprint for GHGs.
“Clarifications are different than word games.”
John V, not necessarily. Boris made a general statement, “The hotspot is not characteristic of greenhouse warming. The cooling stratosphere is”. I replied by suggesting that stratospheric cooling is also characteristic of ozone depletion and therefore not characteristic of GHG warming alone. He replied by calling me “cocky and wrong” even though my general statement is right and his initial general statement is wrong. Seen in this context, his ‘clarification’ was part and parcel of someone engaged in a word game that would shame a suburban lawyer.
Regarding the distinguishing characteristic of GHG-forced warming, it would have to be a combination of tropospheric warming and stratospheric cooling. Not one or the other.
Jonde, you are right, but there it is.
Jonde–
The difficulty with these word games and the sleight of hand is precisely why I object to RC introducing this red herring into a criticsm of Douglas et. al.
Douglas et al discussed the hot spot as a fingerprint of ghgs. It is a fingerprint using the definition in the IPCC glossary.
Even if they had mis-used the term, the hotspot is a feature that is predicted to arise as a result of elevated GHG levels we have seen on our actual planet. It is not anticipated to arise as a result of the variations in solar forcing we have experienced. Comparing observations to a predicted profile is a rational thing to do.
This sleight of hand is going to come back to bite the modelers because when the hot spot is unequivocal and rising at a rate that matches models, those who believe it’s the sun are now going to be fishing out those pictures and claiming it’s the sun. But, the answer to that will be: No. It’s not the sun because the solar forcing is too low.
Lucia, I’m curious to know the magnitude of the increase in solar output we would expect to see in order to get a solar forcing of approx. 2%. Can you help?
Lucia,
I’m bothered by both your and Arthur’s statements that suggest that the day will come that the hot spot will be found.
The hypothesis is straight forward; if the surface warms, the atmosphere higher up should warm faster. We have sonde and sat data that clearly demonstrate that no such thing is happening.
Folks need to deal with it, there is something in the system that is not understood and therefore the physics are not being applied properly. This is not unusual, especially in complex and chaotic systems such as the climate.
Mike C– I think it will likely be found. If it’s not, then I will have predicted incorrectly. But, bear in mind PatK has a theory that it won’t be found. He could be right too. The thing about predictions is they are predictions. We have more confidence in some and less in others. I’m more confident it will still be cold in Chicago tomorrow than in the prediction we will someday detect the hotspot.
If and when it’s found, I would say that lends support to theories that predict it. If it’s not found, that supports theories that predict it’s not there. The ability to predict data we have not yet seen is what makes some theories better and models better than other theories and models.
Dover_beach: The magnitude of the solar constant is discussed here: http://en.wikipedia.org/wiki/Solar_radiation Does that answer your question?
Lucia,
My understanding is that the HOT SPOT is actually an area that has a higher warming trend, not that the temp is necessarily hotter than the surface.
With a higher warming trend it would eventually develop higher temps.
Is this correct?
If so, the upper trop appears to have a negative to flat trend from sat data. After almost thirty years of sat data, and much more sonde data, this is an embarrasment to the IPCC, Modellers and warmers in general
Lucia,
you state ” However, it does appear the data are too uncertain to permit us to conclude it’s not there.”
So, what I am hearing is that they THINK we have warmed by about .6c over the last 30 years. That this is exceptional. That it will continue, if we do not reduce the CO2 in the atmosphere, and become dangerous to the environment and humans, yet, the TREND in the Hot Spot is still to small to be identified even though it should be about twice the trend of the surface??
Please explain this so that I can understand it. Apparently I am a little dim.
kuhnkat “My understanding is that the HOT SPOT is actually an area that has a higher warming trend, not that the temp is necessarily hotter than the surface. ”
yep
Not sure it does, Lucia, but that simply might be a failure on my part. What I’m concerned about is this figure of an increase in 2% of solar forcing that RC raised. What increase in TSI would be required in order to satisfy a 2% increase in solar forcing (since it appears to fall outside the bounds of current variations in TSI)?
Boris (Comment#7457) December 21st, 2008 at 7:08 pm ,
If that’s true, then why does the graph in my copy of A First Course in Atmospheric Radiation show in Fig. 10.6 on page 313 (2nd edition paperback) the shortwave heating from ozone at 30 km to be 2 C/day and increasing rapidly with altitude, and at 20 km to still be greater than zero? Then in Fig. 10.8 on page 317 it shows long wave cooling from CO2 at 30 km to be ~2.5 C/day and greater than 1 C/day at 20 km, i.e also increasing rapidly with altitude. You also get some long wave warming from ozone in the 20 to 30 km range too, so a decrease in ozone does have a cooling effect below 30 km as does an increase in CO2. Feel free to cite a reference that supports your contention.
Bill as I’ve told you before it’s inappropriate to use the baseline [CO2] as a free parameter in your fitting, in the two graphs here you use 306ppm as the baseline in one and 334ppm in the other.
lucia,
I’ve regressed UAH and RSS tropical (land only) monthly tmt to tlt anomalies for 1979 to 11/2008. In both cases the slope was less than one (0.94 and 0.91 respectively +/- 0.03 at 95% confidence) with an F statistics for the regressions on the order of 5,000. So the satellite data shows, if I did my sums correctly, that the middle troposphere anyway, always warms less rapidly than the lower troposphere over the 1979 to 2008 period. Is this analysis valid? Will autocorrelation cause a problem with this method? Comparing OLS trends of the time series calculated separately seems to me to be a very weak method as the error in the slopes calculated that way is large even before correcting for autocorrelation.
A cursory look at AMSU satellite data suggest that we may be going down again in global temps for the next 3-6 months like last year
http://discover.itsc.uah.edu/amsutemps/execute.csh?amsutemps
BTW WITHOUT a significant La NINA
re previous you have to go to main amsu page and then pick 400mb (hotspot area)
Lucia,
If Douglass had used the term “a” fingerprint instead of “the” fingerprint, most of this argument could have been avoided.
This link from a Google book page has an interesting discussion about fingerprinting.
http://books.google.es/books?id=BfjuQupHmQ4C&pg=PA92&lpg=PA92&dq=GHG+warming+fingerprint&source=bl&ots=RGvW9Ww83H&sig=e5YB3uGeFTT4L7iKd5b5Pir76SY&hl=en&sa=X&oi=book_result&resnum=9&ct=result#PPA92,M1
They talk about fingerprints as spatial patterns and then talk about efficient fingerprints and unique ones. They also mention the problem, identified by Hansen 1984, of distinguishing between GHG warming and solar warming at the surface and suggest that it would be better to look at all altitudes.
I am afraid that I do not buy the argument being put forward by Boris that the faster warming of the troposphere is not a part of this particular fingerprint.
If this tropospheric trend is not found by observation it is a serious problem involving the understanding of the physics of lapse rates.
Thank you Jonde #7463 for introducing a moment of sanity into this argument. This is not the first time I have seen the debate descend into Jesuitical hair-splitting over the precise meaning of words. I recall the discussion of Douglass et al on Climate Audit which had endless comments, from an AGW proponent, discussing the meaning of the word “consistent”.
My take on this is that the sceptics thought that a tropical tropospheric hot spot was a testable hypothesis arising from greenhouse theory. It appears now that it is not, but we are left with not much in its place, (except something debatable about stratospheric cooling, which appears to have stopped around 1995).
On the subject of tropical temperatures at Climate Audit, ‘cce’ said
“no forecast can be made, nor have they been made, for the tropical troposphere over a period as short as 30 years.”
When we will see a greenhouse theory that is falsifiable?
Lucia: Bill Illis’s comments about ENSO appear to me to have significance in the discussion. As the group noted at RealCimate, regardless of whether the warming is from an unrealistic increase in solar irradiance, or an unrealized increase in CO2, or an El Nino, which we have had an overabundance of over the last 30 years, there should be a tropospheric hot spot. Douglass et al says the hot spot doesn’t exist. Santer et al says it does, but did Santer et al attribute it to GHGs or did they simply say it existed?
“Why are you switching to distinguishing feature from “fingerprintâ€?”
Are you fingerprints not distinguishing features? Why do you think the IPCC does fingerprint studies? To identify distinguishing features–right?
Your defense of Douglass neglects that he uses the term “characteristic fingerprint,” which means “distinguishing feature” despite your attempts to cover for his ignorance.
You see, all you have is word games. You won’t admit that a TT hotspot is not a distinguishing feature of GHG warming. Frankly, you don’t seem to want to learn anything, but only to defend skeptics and their stupid arguments.
So will the twaddle continue or will you admit that Douglass was wrong? Will you ignore his use of “characteristic fingerprint”? Will you try to redefine it in some obscure way? How will Lucia get out of this one–stay tuned!
Dewitt,
Here’s my source:
http://www.atmosphere.mpg.de/enid/20c.html
I was (over?)simplifying things. There is definitely some overlap between effects. And , of course, there’s anthropogenic stratospheric water vapor to contend with as well. However, the fact that CO2 has more of an effect in the upper strat. and ozone depletion in the lower strat. is not, I think, in dispute.
To those interested – I generally have pretty busy weekends, especially this time of year. Last evening, for instances, our family was busy delivering cookies to friends and neighbors…
Lucia – you have not responded or addressed, as far as I can tell, the central issue that a *fingerprint* or *pattern” or “signature” is something that represents effectively a direction in a vector space, but not its magnitude. You complain about RealClimate’s estimate of the hot spot from a large solar forcing, but your comments fail to recognize that these responses are expected to be linear in forcing.
That is, on average (you would have to do longer averages for a low forcing, of course, to get reasonable accuracy) – the pattern of response is the same for a given forcing, independent of the amount of forcing. If we have 2 W/m^2 of GHG forcing we have essentially the same pattern as if we have 0.2 W/m^2 of GHG forcing, but the magnitude of the response in the latter case is 10 times weaker. The RealClimate graphs are just as valid for very small changes in solar forcing, or GHG forcing, and provide a comparison of the pattern, independent of the magnitude of the forcing.
And what those pictures prove, as do the graphs in Figure 9.1 of IPCC AR4 WG1, is that, exactly as Boris stated up front,
Nothing you have claimed here has disproved Boris’s original statement on this.
You also of course continue to be wrong about the text of section 9.2 – it nowhere identifies the mid-troposphere separately from the lower troposphere, nor does it specifically identify the tropics; the only mention of latitude is at the surface, with northern hemisphere amplification in some of the patterns. I know that the human eye is drawn to red spots, but these are scientific plots, and the red is not as important as the fact that the entire troposphere is yellow (or red) for GHGs, while the entire stratosphere is blue (in contrast to solar and some of the others). That is what the text in section 9.2 is discussing; I’m sorry, but it’s absolutely clear they did not highlight the “red spot” at all by any name or reference or allusion in the text.
And as far as the evidence for the hot spot being found in the future, it seems very likely, given as reported in section 9.4.4.4:
“except something debatable about stratospheric cooling, which appears to have stopped around 1995”
If your source is MSU TLT channel 4, then you should know that that channel only covers the lower stratosphere.
Stratospheric cooling seems to be caused much more by volcanoes than global warming. Another fingerprint which seems to have some confusing data.
KuhnKat—
I am referring to anomaly maps– so, yes, I mean hot relative to “without CO2”. Sorry if that’s confusing.
Arthur—
Boris’s claim which you quote, might have been fine had he edited his sentence to say precisely what he meant which was: We would expect a hot spot from any warming caused by forcings of identical magnitudes. But he didn’t say that. And he was specifically criticizing a conversation that was discussing warming due to forcings that have been occuring on the real earth. So, in context, if he meant to say what he now explains, his statement is incomplete. A more complete version would have changed the subject from the actual topic being discussed.
As the incomplete sentence stands, inserted into the context of a conversation where everyone is discussion warming as it would happen on the real earth, it is either a) a red herring, because it changed the subject or b) wrong because unless he changed the subject we must assume he means to be introducing a fact that makes sense in the context of the subject under discussion.
If red in context of the discussion, with the missing information filled in from the surrounding conversation, it suggests and idea that is wrong, and so Boris’s statement as it stood was wrong.
So, tell us which it is. Irrelevant red herring with missing ideas to make it appear wrong? Or just wrong because he really meant it in context?
On this:
Arthur– even if true, this would not make their inserting that idea into a criticism of Douglas valid in any way. My complaint– as I have stated over and over is not that the idea that large solar forcing could cause this. My complaint is the observation is irrelevant to discussion of Douglas, and was inserted as a red herring. Including irrelevant but correct information in a discussion is a known logical fallacy and used to distract from the real topic. In the case of their article the topic was what Douglas claimed, and where it was wrong.
I don’t know why you want to focus on whether or not the red herring contained true content. The information was irrelevant because it had nothing to do with Douglas. The little lecture also seems to have caused numerous people to conclude, working backwards, that “fingerprint of ghg” must point to something unique. That’s not the definition for fingerprints in the IPCC glossary.
Arthur (still)– in your last paragraph, you are rebutting things I did not claim. I said they said tropical troposphere. That’s what they say.
That said, I don’t know why you think a “spot” showing something that is an isotherm in anomaly space is not a a spot, and think that is tricking my eye, and yet you end your sentence by saying you think there is evidence they will find the hot-spot. If it’s not a hot spot, why do you think they will find it.
Boris–
If you want to discusss distinguishing features, fine. I have B+ blood. If you draw my blood it will be B+ not A- etc. You can use this information to distinguish my blood from that of my sister whose blood is O+ or my father who is B-, or my mother who is A+. You are trying to use ‘distinguishing’ as “uniquely distinguishing”.
While human fingerprints are both distinguishing and (as far as we can determine) uniquely distinguishing, the IPCC definition of fingerprint does not include the “unique” aspect. It may be a pity the modelers didn’t call these “climate blood types”. The analogy between the term and the degree to which the characteristics distinguish anything would be clearer, but they didn’t do that. That’s what happens in English. It’s inconvenient, but there you go.
Douglas used fingerprint the way the IPCC glossary defines the term. It may be that you have developed a stronger definition in you mind, but there is nothing wrong with using “fingerprint of GHG” following the IPCC glossary definition, particularly when writing a peer reviewed paper in climate science.
Jorge–Dang the link to the book reached it’s viewing limit. But yes, Douglas et al would have done better to say “a” fingerprint rather than “the”. Just as I have more than 1 finger, more than one thing canbe called a fingerprint.
Bill I don’t think Santer17 says we have observed the fingerprint. Do you mean another paper? The recent paper only says we can’t exclude its existence because the data are to uncertain.
Lucia, you are contorting yourself here. Just admit you were wrong, and it will all be a lot simpler.
But they do not say tropical troposphere. That phrase is nowhere in section 9.2.2. Once again, you are putting words in the mouths of the IPCC authors.
The only piece you have mentioned which they actually did say was troposphere. The troposphere extends from the surface to the stratosphere, and form the north pole to the south pole. For both GHG’s and solar forcing, the troposphere – all of it – warms up. You have presented no evidence that, in referring to the troposphere, the IPCC authors were implying anything less than the entire troposphere in this discussion. They were not emphasizing any more limited “spots”, except when they pointed out the amplification in the north polar region.
Despite the dramatic nature of the “red” spot on figure 9.1c, the actual contours are only 0.8 to 1 C anomaly, vs the 0.4 to 0.6 C anomaly at the surface. That’s an enhancement, but hardly worthy of much discussion, particularly since the same ratio of enhancement is there in the other plots as well!
You say:
(I don’t – did I ever say the spot wasn’t in the pictures? Really!) and yet earlier you said:
the yellow “swiggle” is yet another isotherm in anomaly space, showing a spot there for solar, just as for GHG’s, in that very figure 9.1a. And figure 9.1d shows an even littler yellow “swiggle” demonstrating another similar spot for ozone. 9.1e shows a blue “swiggle” demonstrating a cold spot for aerosols, the same kind of amplification effect. *Every* forcing has the same mid-troposphere spot. So the IPCC definitely expects a “Hot Spot from ANY and ALL sources of warming” – just look at those figures.
Why would you deny the clear contours shown in these graphs, while focusing solely on the one for GHG’s? The temperature differences between these “hot spots” are not even that significantly different – the solar “swiggle” contour is 0.2 to 0.4 C, while the GHG one is 0.8 to 1.0 C. Are you saying that an 0.3 C rise is not “hot” while a 0.9 C rise is?
But yet again you are reverting to the issue of magnitude when the meaning of fingerprint is independent of magnitude, it is one of a *pattern*, of relative temperature changes in response. And all of the forcings have the same hot spot – I certainly don’t deny it, it’s there for all of them. You titled this post to imply that the IPCC doesn’t believe they all share this feature in their “fingerprint”, but they do.
And don’t get diverted into what RealClimate or Douglass said or didn’t say – I really don’t care who said or meant what in that debate. The issue here is your misinterpretation and misrepresentation of the IPCC AR4 WG1 on the matter.
“uniquely distinguishing”? It just never ends with you and word games, does it? What do you think it means to distinguish between two different forcings?
You have ignored that Douglass used the term “characteristic fingerprint,” so even if he hides behind his misunderstanding of the IPCC (way to truncate their definition upthread, BTW) he still shows his ignorance by using the word “characteristic,” which means “revealing, distinguishing or typical of an individual.”
Once again, the presence/absence of a TT hotspot does not distinguish between forcings (impossible for some people to admit, I know). Just a reminder: the Douglass paper was about NOTHING but the so called absence of the TT hotspot.
Bill, volcanoes WARM the stratosphere.
“Bill, volcanoes WARM the stratosphere.”
And what will absence of volcanos do? Right.
lucia,
The term “fingerprint” is not some clever or new analogy. It is a well-accepted term in the english language. It’s in every dictionary I’ve checked. In this context it means a distinct characteristic. Period. You can’t pick your own definition.
As Arthur Smith says, it’s the pattern. The magnitude of the TT hotspot for a particular source of warming is proportional to the magnitude of the source of warming. Any and all sources of warming produce a TT hotspot proportional to the magnitude of the source.
As for your original question, consider this: *If* there was significant solar-induced warming, then would we expect a significant TT hotspot from solar warming? If the answer is yes, then consider your title question. If your answer is no, then I don’t even know where to begin.
Jorge (Comment#7479) December 22nd, 2008 at 5:50 am ,
If there is no hot spot, then it’s likely due to a failure to correctly model the humidity profile. Convection and clouds may be the weak links in the process. If the water vapor doesn’t get to the upper stratosphere for whatever reason, the lapse rate there won’t decrease and there won’t be a hot spot. To a first approximation there is no convection in the stratosphere so radiative transfer dominates and the expected effects of CO2 and ozone are observed.
I know what’s going on here. AGW Scaremongers can never be wrong. If they ever are, they have to give up a closely-held belief, which is very hard for anybody to do. That’s why the argument degenerates into word games.
They can’t be wrong.
Andrew ♫
*Every* forcing has the same mid-troposphere spot
If they are the same, why do they look so diferent? Not only by the amuont of temperature change, but also the extension, latitude, and height. And why would they bother to show all af them? Four repetitions of the same thing?
Looks like a quite particular reasoning to me.
when i was young i read somewhere that the goal, the ideal of a mathematician, was to write a mathematical treatise whithout words, only equations… (i think it was Euler) less words clearer the concepts and “harder” the science. how far from that ideal is the climate science!!!
Andrew:
Insisting that “fingerprint” means “fingerprint” is not a word game. On the other hand, pretending it’s some clever analogy that’s never been used and choosing your own definition is just silly.
LC:
Look at the *pattern*. The amplitude of the TT hotspot is proportional to the amplitude of the warming from a particular source. Since the sun had little warming effect in the 20th century, the TT hotspot from the sun has a small amplitude. *If* the sun was a significant factor in 20th century warming, then the TT hotspot from the sun would be more substantial.
BEGIN SARCASM
If only there was some way to look at the pattern without being confused by the magnitude. Hmmm…. Maybe we could use hypothetical solar forcing that makes the scale similar. Then we would look at the differences between the patterns of solar- and GHG-induced warming. I wonder what that would look like? Oh look, there it is on the RealClimate page that lucia linked above. What a handy image for understanding the differences in the patterns arising from different sources of warming.
END SARCASM
Excuse me, John V. I see something very different in the GHG spot. Its no only stronger, but it’s maximum strength is located very precisely in a place where in the other cases nothing of interest happens. Well, the maximum of the solar forcing is strong in the same place, but not only this place.
What’s the idea? With stronger solar forcing, the maximum temperature change would concentrate where the GHG forcing does in this plot? Or, with a lesser forcing, would the maximum of the GHG forcing extend to where the solar forcing does? Or maybe this plots are just meaningless aproximations.
I apreciate your answers. Just trying to understand.
LC,
I’ll do my best to explain it as I understand it. My 6-year-old is pulling on my leg, so I apologize in advance for any errors or omissions:
For a similar amount of solar forcing or GHG forcing, the hotspot in the tropical troposphere will be similar. (The “hotspot” is actually an enhanced warming trend — it’s still cold up there). Other parts of the warming pattern will be different. It is the other parts that can be used to distinguish between solar- and GHG-induced forcing. The hotspot in the tropical troposphere, on its own, says very little about the cause of warming. That is why the hotspot, on its own, is not a distinguishing characteristic (or fingerprint) of GHG-induced warming.
John V
Does that mean an increase in the height of the tropopause isn’t either?
If you look at the stratosphere temperature history, it seems pretty clear the volcanoes cause a temporary increase (1.0C) in stratospheric temperatures followed by a decline (of greater magnitude at 1.5C) to a new lower level at which temps stabilize and (perhaps) slowly build back up until the next big volcano.
http://hadobs.metoffice.com/hadat/images/update_images/global_upper_air.png
http://hadobs.metoffice.com/hadat/images/update_images/msu_timeseries.png
John M – no, an increase in the height of the tropopause is another distinguishing characteristic of GHG-induced warming (though it is also similarly influenced by ozone changes). In fact it is very closely related to the real essence of the greenhouse effect, the changes in thermal radiation from the different atmospheric layers.
Increasing tropopausal height – the observations that demonstrate it and some of the modeling – is discussed in IPCC AR4 WG1 section 9.4.4.2, and figure 9.14.
Arthur,
Now you’ve got me really confused. I thought the gist of the argument by Boris et al. was that it’s not a fingerprint if something else can cause it.
And on the subject of stratospheric temperatures, the graphs Bill Illis lnked to seem eminently “fingerprintable”, in the same sense that global surface temperatures over the past 100-150 years are supposed to be a fingerprint. Since volcanic aerosols, GHGs, ozone depletion, and solar effects are so well understood, it would seem those curves should be able to be modeled. Have any models been able to explain the seeming “rebaselining” of stratospheric temperatures after volcanic eruptions?
DeWitt,
I agree that clouds and convection are the likely culprits. These seem to be the least understood part of the thermal balance.
LC asks:
IPCC showed these graphs to demonstrate the differences in patterns – in particular stratospheric cooling for GHGs vs warming for solar, and north polar amplification, and some of the other differences for ozone, aerosols, and volcanoes. You’re quite right that the sameness in patterns (the “hot spot” in particular) was not a scientifically interesting aspect, which is why it was not discussed in the text that referred to figure 9.1 (section 9.2.2), essentially taken for granted. Despite the eye-catching redness in the GHG case, it was scientifically irrelevant to the discussion there.
Here are some coloured contour plots based on observations.
http://hadobs.metoffice.com/hadat/images/update_images/zonal_trends.png
I can’t seem to find the “red blob” but maybe someone else has better eyes.
Andrew:
Insisting that “fingerprint†means “fingerprint†is not a word game. On the other hand, pretending it’s some clever analogy that’s never been used and choosing your own definition is just silly.
Re: #7497
John V,
I’m basing my conclusion upon observation. It appears that AGW’ers have it backwards. They have the belief first that Man Is Causing Global Warming and they find fingerprints to support their beliefs. The evidence should dictate the conclusion, not the other way around.
I don’t have any choice but to be a skeptic based on the evidence, FYI. I’m simply pointing out that there are a significant number of people who retain the belief in AGW and the possibility exists that they may be wrong. These people just don’t want to face the fact that they may be wrong. It is too much for them to bear, personally.
Andrew ♫
Andrew, you have a lot of insight into the motivations and feelings of people who believe in mainstream science. I assume you do not include yourself in that group. I have to wonder how you have such insight into their psyches.
I have a number of ideas about the motivations of those with one-way skepticism against AGW, but I’ve learned that sharing those ideas is rarely productive.
Of course there is a *possibility* that AGW is not real — that is why IPCC conclusions are given in terms of likelihoods instead of absolutes.
John V
I’m just saying that there are patterns to be seen patterns in posts, that’s all.
Andrew ♫
Andrew
“AGW Scaremongers can never be wrong. If they ever are, they have to give up a closely-held belief, which is very hard for anybody to do”.
You are right.
Have you noticed how on sites like RC and Open Mind they can never accept the idea that Mann’s hockey stick has been,at least, severely dented? The mantra is always ‘this is old, but groundbreaking, science and it is time to move on’ – whilst at the same time they ‘defend it to the death’ against any criticism.
Boris, John V, et al,
I think you are managing to obfuscate simple facts.
A strong majority of GCM’s and simulations show tropical mid troposphere relative warming trends when greenhouse gas forcing is applied. The IPCC states this and graphically portrays it.
Present atmospheric measurements either do not confirm this effect, or the measurment uncertainties can admit some small amount of relative warming.
Whether some other forcing could cause this effect, which is presently nonexistent or at least not confirmed within measurement uncertainties, does not bear on the previous two statements. As Lucia has said repeatedly, it’s a red herring argument.
Clearly, there’s a problem with the models, the measurements, and probably both. Improve the models and the measurements.
I agree with Duane. For Arthur and Boris, keep it simple guys before running off on semantics. The IPCC figure shows what a model will predict is the change in temperature per century over the Earth though the atmosphere based on a number of theorised effects due to measured ‘assumed’ causes. It shows that the most prominent ‘signature’ is due to GHGs. So anyone reading that (scientific or not) can come away thinking the models predict that there will be the largest change in atmospheric temperature trends due to GHGs. Not due to the effects of the sun etc as observed to this date. The report is quite specific to point out that they are using real observations and attributing theorised forcings on them.
The report then goes into the predictions etc. But based on this graph the question you should ask is then: what is observed? Errors withstanding the answer is not what the models predict. That’s okay except that the hubris engine behind AGW has been churning out alarm for some time now such that the original scope of the exercise should be repeated for clarity. Not that suddenly the measurements showing something different was actually included in the models anyway. Maybe some humility and common sense with this graph in mind should be applied. The theory is not matching observations. If you then try and fiddle the errors bars and say there are larger you should look at your apparatus and method and improve them, not the results it generates, which you should ignore instead of herald.
“As Lucia has said repeatedly, it’s a red herring argument.”
The argument has been about about claiming that the TT hotspot is a “characteristic fingerprint” of GHG warming. I welcome discussions about improving models and observations. If “skeptics” would stop saying that the TT hotspot is a fingerprint of GHG, maybe we cold get somewhere. However, that is apparently impossible.
Wow… I go shopping for the day and look at this.
Boris– your’s is at the bottom: The reason you think the TT hotspot is not a fingerprint is that you refuse to accept the definition of fingerprint that appears in the IPCC glossary for the AR4. That is the definition that applies in climate change. Others will continue to use the IPCC definition when applying the term to climate change.
Now…. inlaws coming. So, carry on.
Just to be clear, a lack of a TT hotspot (if confirmed within the error bars) is definitely an issue for the models. The models predict that the hotspot should be there.
However, that does not make it a fingerprint. The fact is that warming from any source is predicted to cause a TT hotspot. The TT hotspot is a *characteristic* of GHG-induced warming, but definitely not a *distinguishing* characteristic.
It does not make the image at RealClimate contradict the IPCC.
It does not mean that solar warming (if it existed) would not cause a hot spot.
John V
Given that a rise in tropopause height can also be caused by decreases in stratospheric ozone, is tropopause height a *distinguishing* characteristic of GHG-induced warming?
JohnV–
No one said the lack of a hotspot is the feature that makes that hotspot a fingerprint. It is a characteristic that arises as the result of the specific forcing: GHGs. This meets the definition of a fingerprint as defined in the glossary for the AR4.
You are leaving out a word in what I said in order to try to rebut what I said. We have gone over this image itself standing alone, and some other context that is not pretending to criticize wht Douglas would not contradict the IPCC figures. The insertion of the image in a discussion of Douglas, which compares observations to model predictions under the assumption that forcings are as have occurred on the planet earth is what makes the RC discussion seem to contradict the IPCCs discussion. The IPCC discussion is relevant to Douglas. The RC one is not. However, the fact that the RC would cause an unsuspecting reader to assume it was intended to apply to the circumstance discussed in the IPCC document and Douglas would make it seem to contradict the IPCC discussion.
In reality, the RC discussion applies to circumstance that are irrelevant to Douglas. The only way to resolve the seeming inconsistency between RC and the IPCC document is to point out that the IPCC figures are relevant to Douglas and the RC figures are irrelevant to Douglas, which they purport to criticize.
It’s fine to disagree with my point of view. But what’s the point of dropping a word out of what I said so as to change the meaning and then rebutting the different idea?
John V, I congratulate you on #7516. That post clearly gets away from the lawyerly parsing of what “fingerprint” is, as if it actually matters to the debate. It also leads me to want to unpack this a bit. I accept, albeit reluctantly, the TT hotspot as not being a “distinguishing” characteristic of GHG warming, although to say it has not been promulgated that way is disingenuous at best. To have to go to otherworldly physics to make the point says plenty in and of itself.
Since plenty of GHG forcing has happened, and this is not in dispute, the absence of a TT hotspot would be trouble for more than the models, would it not?
lucia,
You said:
“No one said the lack of a hotspot is the feature that makes that hotspot a fingerprint.”
I don’t know what you’re talking about. Rather than try to parse, let me try again:
Do you believe that solar-induced warming (if it occurred) would cause a TT hotspot?
Do you believe that warming induced by other means (it it occurred) would cause a TT hotspot?
If your answers are yes, then how would you answer your title question? Who believes that a TT hotspot would arise from most/all sources of warming? Do you? I do.
(Note that solar in the 20th century was not a source of significant warming, so it is irrelevant to the question).
—
I’m not sure which word I droppped, and I certainly did not do s0 intentionally. You write so many words — how am I to know which one is the culprit? 🙂
Anyways, both the RC image based on a 2% solar increase and the IPCC image based on actual solar forcing have the same pattern. Yes the magnitude is different, but they in no way contradict each other. Going back to your light fingerprints (which you brought up many comments ago), does a light image of your fingerprint contradict a dark image of the same? The only difference is how hard you pushed (“forced”) when making the print.
In the future, if you want “fingerprint” to mean “characteristic” or “property” then you should just say “characteristic” or “property” and be done with it. That way those of us who use dictionaries for word definitions (vs IPCC glossaries that missed a word) won’t be misled.
The Douglass argument is a whole other thing. I can only assume that the insertion of the RC image was because of Douglass calling the hotspot “*the* characteristic fingerprint” or due to other context that is now forgotten.
Anyways, I’ll soon be out of the office for a week or two so may not reply. Don’t think that means you won. 🙂
Are these guys using the term fingerprint in a way that will pass muster with Boris, Arthur, John V etc?
http://www.springerlink.com/content/j84225w15l380x08/
“The greenhouse fingerprint used in this study may not be unique to the enhanced greenhouse effect and may be due to other forcing mechanisms. “
Peter,
The problems in these types of arguments is often that the scope of the argument is not tight enough. The discussion is about whether a TT hotspot can be called a fingerprint of GHG-induced warming. It’s not about Douglass, or Gavin, or model verification. One thing at a time.
Who’s doing the promulgating that the hotspot is a fingerprint? In my experience, it’s been the “doubters”.
I’m not going to hypothesize about the implications of a missing hotspot — this discussion is already too rambling. I will say this though. The implications are not all in the direction of less warming and lower climate sensitivity.
What a strange discussion.
This all could have been averted if Lucia had properly defined her terms at that start.
Lucia, you said:
“Yes. But this is what the IPCC AR4 authors mean by a finger print. That’s how they use the term. ”
Did you ask them?
I believe the models predict it would the forcing is higher than experienced by the earth. I believe the predict it would not if the forcing is the level that we have experienced on earth. (This is the same answer I have been giving.)
I believe the models predict this for levels of GHG experienced on earth.
Your questions are consistently incomplete. For this reason, I include the caveat in my answer. So, if the conversation either implicitly or explicitly involves the assumption that we are talking about levels of forcing that have occurred on earth (as in Douglas or at Steves blog) the answer to the title question is “No.” If the conversation either implicitely or explicity states we are assuming the same level of forcing, the answer is yes.
First: I think your question is incomplete as it does not include any statement about whether you are discussing “at levels of forcing actually experineced on earth” or “at sufficiently high levels of forcing, including some that are cannot occur.”
Second: I’ve never read anyone suggest the TT hotspot must arise if observed surface warming is due to El Nino. I don’t know if it must arise if surface warming is due to PDO or some of the longer time scale oscillations suggested by some. (I’m trying to remember the name of the guy who thinks it’a sll PDO. What’s supposed to happen if Christy’s theory of clouds is true. Must the TT hotspot form? I don’t know. Do you?
On the one hand, I believe quite a bit of the warming over the past century is due to GHGs and not those natural features. I expect we’ll probably see the hotspot eventually and the reason is because it is a characterict specific to GHGs. That is using the IPCC definition of fingerprints, It is a part of the fingerprint from GHGs.
But that only returns us to the question: Does the question imply only things that can happen on earth? Or does it include hypothetical forcings or hypothetical sources of warming?
I think I’ve answered these questions quite a few times now. The answers aren’t going to change by presenting the incomplete question over and over again. I’ll just answer the question adding the caveats related to what we mean by “any and all” forcings.
JohnV–
Are you suggesting the reason it’s wrong to call the hotspot a fingerprint is because doubters are the ones pointing this out? The difficulty is, notwithstanding who calls it a fingerprint, the hotspot fits the definition of a fingerprint as defined in the glossary of the IPCC AR4.
True statements are true statements. If doubters say 2+2=4, it’s just as true as if believers say 2+2=4.
Nathan–
I don’t see any reason to write them as the IPCC authors kindly provided a glossary with the definition of “fingerprint”. Both the definition and the link to the glossary are in comments above.
John M, that is damning evidence.
Lucia,
Again, if you had said you were using the term as defined by the IPCC rather than the colloquial use of the word this whole discussion could have been avoided.
Otherwise this whole post is little more than a poke at Boris for using the word fingerprint in a different way to that used by the IPCC.
“Are you suggesting the reason it’s wrong to call the hotspot a fingerprint is because doubters are the ones pointing this out? ”
Well they should define their terms when doing it. When non-scientists read about something being a ‘fingerprint’ they aren’t necessarilly using the IPCC definition. It’s important before any scientific discussion to define your terms, mostly to stop pointless bickering.
This whole discussion seems to come about from your claim that the red spot in Figure 9.1 is a fingerprint. You have decided that according to the IPCC’s definition of fingerprint that it is. Did you ask the authors?
Also if this is your definition of fingerprint:
“The climate response pattern in space and/or time to a specific forcing is commonly referred to as a fingerprint. ”
Then doesn’t that refer to the whole pattern, not just bits of it.
Dover_Beach
Thanks for responding.
I was beginning to think my mike was off. 🙂
Nathan–
I already answered that: The IPCC provided a glossary defining the term.
I don’t need to write the authors of the glossary or the document– it’s explicitly defined. I also don’t need to write Merriam Websters to tell me whether the really mean what they say when they define “foot” or “specific”. People don’t need to write explicit definitions of individual words in every paragraph. If the did, nothing would be communicated.
As you see in comments, several readers have found numerous usages of of climate fingerprint in peer reviewed literature that specifically point out that climate fingerprints are not unique. Those uses parallel the definition in the glossary of the AR4. So: they aren’t unique.
Of course you, a reader, or any individual reader may not immediately know the definition of “fingerprint” in the IPCC glossary. The fact that some lay people don’t understand a term of art doesn’t change the definition of the term. In any case, you can look it up– just as you can look up “catastrophic” or “somersault”.
Even if you don’t understand the term and Boris’s decree that those who do use the IPCC usage are being silly is absurd. If they are silly, then so is the IPCC.
FWIW.. this discussion did not start from my claim. That came up after the dicussion over the term “fingerprint” started.
(I’m actually mystified why this particular post has a record breaking number of comments. Is everyone home on vacation?)
John M,
I feel you are not giving an accurate summary of that paper. Here’s the abstract in full:
“As an example of the technique of fingerprint detection of greenhouse climate change, a multivariate signal or fingerprint of the enhanced greenhouse effect is defined using the zonal mean atmospheric temperature change as a function of height and latitude between equilibrium climate model simulations with control and doubled CO2 concentrations. This signal is compared with observed atmospheric temperature variations over the period 1963 to 1988 from radiosonde-based global analyses. There is a significant increase of this greenhouse signal in the observational data over this period.
These results must be treated with caution. Upper air data are available for a short period only, possibly too short to be able to resolve any real greenhouse climate change. The greenhouse fingerprint used in this study may not be unique to the enhanced greenhouse effect and may be due to other forcing mechanisms. However, it is shown that the patterns of atmospheric temperature change associated with uniform global increases of sea surface temperature, with El NinoSouthern Oscillation events and with decreases of stratospheric ozone concentrations individually are different from the greenhouse fingerprint used here.”
So you note that the authors begin by defining what their fingerprint is.
This is the problem with this post, Lucia expects everyone to be working from the IPCC definition first.
Lucia,
I think you may need to ask the authors.
They may have defined what they meant by ‘fingerprint’ but it is you that has decided that the Tropospheric hotspot is a fingerprint. That is what you need to ask the authors.
I think both. Patterns can contain subpatterns. For example, lowering of the tropopause by itself is decribed by Santer is a fingerprint. Cooling of the stratosphere by itself is called a fingerprint for ozone forcing. The way oceans warm or cool relative to the land can be included in some fingerprints and/or are t hemselve called fingerprint.
Google a bit and you’ll find all sorts of patterns that are sub portions larger portions called “fingerprints”.
Example
“A graph of global surface temperature over the past century shows a crooked line that can be thought of as a fingerprint. Any climate driver
fed into a model must match this fingerprint to qualify as a likely cause of observed warming.”
“The data show that the oceans have been warming from the surface downward and that heat penetration with depth varies from ocean to ocean, providing a fingerprint that drivers in a model must match. Modeling of internal variability alone did not produce temperature
profiles that matched this fingerprint, whereas combining internal variability with the effects of greenhouse gases did. ”
“Hence, the overall loss of ice is a fingerprint of global warming.”
“The existence of a trend of intensification in all six of the tropical cyclone-producing ocean basins thus represents a fingerprint of global warming, consistent with the enhanced greenhouse effect and not with natural variability alone. ”
The definition doesn’t say “whole pattern”– just a pattern. The term is pretty generic. If the IPCC provided Boris’s definition, many using fingerprint studies are using the term incorrectly!
lucia,
“Are you suggesting the reason it’s wrong to call the hotspot a fingerprint is because doubters are the ones pointing this out?”
No. No. No.
I didn’t say anything like that.
Nathan–
You have things a bit backwards. I expect Boris and to refrain from decreeing those who use that definition “silly”. I expect people to refrain from accusing those who use the IPCC definition of being “confused”.
Clearly, people ought to be permitted to use the word in the way defined by the IPCC without being called silly or confused. If Boris wants to use another word for patterns that are not unique when he writes his own paragraphs or expresses himself, that’s fine.
But others ought to be free to use the IPCC definition. That usage can hardly be called wrong or silly.
Lucia
You said you were using the IPCC definition. You were using IPCC figures. By their definition a fingerprint is
“The climate response pattern in space and/or time to a specific forcing is commonly referred to as a fingerprint. â€
And now you want to claim that ‘pattern’ also means parts of patterns?
This is actually beyond trivial now. You are now just defining terms according to how you want them. This is why you should speak to the authors and find out. Otherwise you are just making stuff up.
Why do you expect everyone to use the term the way the IPCC does? Fingerprint, as you point out, means a lot of things and careful authors will define it (like the IPCC). It doesn’t mean that different studies that use the term in different ways are somehow contradictory.
Nathan,
That’s fine. I’m as big a fan of Humpty Dumpty as the next guy.
http://www.leeandkristin.net/OldQuotes3.html
Don’t you think, that in a discussion about “climate change”, that the IPCC definition in the glossary ought to be the default position?
Look, gang tackling by AGW proponents on the fine points of the defintion of “fingerprint” is one reason folks become skeptical. People often ask me how I can be skeptical. I point out this kind of stuff and their jaws drop.
So I’ll ask you, is an increase in the height of the tropopause a “fingerprint” or a “distinguishing feature” of GHG-induced global warming, even though it can also be due to ozone depletion?
Lucia
“You have things a bit backwards. I expect Boris and to refrain from decreeing those who use that definition “sillyâ€. I expect people to refrain from accusing those who use the IPCC definition of being “confusedâ€.”
No, I don’t. Boris didn’t do either of those things.
“Clearly, people ought to be permitted to use the word in the [w]ay defined by the IPCC without being called silly or confused.”
You decided that the TT hotspot was a fingerprint using their definition, not the IPCC. And I pointed out above that the IPCC definition doesn’t really allow you to do that.
John M
No I don’t think the IPCC should be the default position. As long as it is clearly articulated what the meaning is it’s fine.
“So I’ll ask you, is an increase in the height of the tropopause a “fingerprint†or a “distinguishing feature†of GHG-induced global warming, even though it can also be due to ozone depletion?”
You can call it whatever you like, it depends on how formal you’re being. If you are attempting to quantify something it’s best to be formal (so that you can set limits).
Also, I think the Ozone depletion causes cooling of the stratosphere and is only significant over Antarctica, not the Tropics.
“I point out this kind of stuff and their jaws drop. ”
Happens often in science, especially when people don’t define their terms. That’s why the IPCC went to the trouble.
Nathan–
Are we going to argue “whats a pattern”? Say you enter a room. The floor is covered with two type of rugs laid like this:
ABABABAB
BABABABA
ABABABAB
BABABABA
Then, the “A” rugs is covered with paisley swirls and the “B” rugs are plaid.
Is the whole thing as a collection a pattern? Or is the alternating ABABA structure a pattern? Are the paisleys as pattern? Is the plaid a pattern?
I think all four things can be called patterns. (The whole think together is probably horribly ugly.)
Nathan,
I guess to no effect.
Lucia, when does the “Boris frequency of posting” filter kick in? I must be getting close.
Lucia
That’s a poor analogy. Look at figure 9.1, there are no patterns as you describe a pattern in your analogy. So if this is how you want to define pattern then again, the pattern must be the whole not the parts.
Again, this is where you need to contact the authors. That way you don’t have to guess, or make stuff up.
Nathan–
I didn’t say different authors use the terms in different ways. The various examples all describe patterns in space and time. All fit the IPCC definition, which only says “pattern”. Some patterns are fit inside other patterns.
You asked if the pattern had to be the whole thing or if it can be a subset. The IPCC definition says only that it needs to be a pattern. So, if it’s a pattern, that’s enough.
John M.
They defined it for their document, so that people reading their document would have an understanding of what they meant. It is to reduce ambiguity, not attempt to dictate word usage to the wider community. Some scientists may adopt their usage.
JohnM–
A person has to be hardwired into the “Slow Down Boris” filter (and as you know, he inspired it one evening, but is no longer coded in.)
I modified the filter and there is one person coded in. It’s not Nathan. That said… I’m going to bed now.
BTW: Good night. Don’t let the bedbugs bite. Don’t kill each other while I’m away.
Lucia,
So are you saying that they are all using the IPCC definition?
Ok, so according to you the TT hotspot is a fingerprint.
This is based on your undestandin of the IPCC definition of fingerprint, which says “The climate response pattern in space and/or time to a specific forcing is commonly referred to as a fingerprint. â€
Now, section 9.2.2.1 is called “Spatial and Temporal Patterns of Response”. Which sounds somewhat familiar? They perhaps could have shortened that title to “Fingerprints” – possibly it didn’t sound as fancy.
They then say, “these figures indicate that the modelled vertical and zonal average signature of the temperature response should depend on the forcings” – so the modelled response is the spatial and temporal pattern. That is, it is the whole thing.
“I think your question is incomplete as it does not include any statement about whether you are discussing “at levels of forcing actually experineced on earthâ€..”
Why is that important? You can’t even answer a simple question without lawyering up. Did you even bother to read the entire IPCC definition of fingerprint? A fingerprint study uses a forcing of a certain magnitude. For exmaple, Gavin did a small fingerprint study using ModelE and published the results on RC. It would be stupid to use historical forcings to determine a fingerprint (so much for your “figure 9.1 is a fingerprint” argument) because the whole point is to compare the expected response to the actual response. You seem to think we should compare the actual response to the actual response–what the heck would that prove?
Moreover, you avoid learning anything by claiming that the wet adiabatic rate is a “red herring,” which only makes me think that you don’t know what you are talking about and, by God, you’re not going to start now.
If only I was as nice as John V! Let me apologize for being brusque (now there’s a word we agree on in its application to me!)
Lucia – there is a good reason this post has received so many comments – you are not usually wrong, or at least not as obviously wrong, as you are in this case, and the issue is quite an important one since it has been widely promulgated by various people, many of whom are not scientists.
Please re-read and address the concerns I raised in my most recent comment – #7488; you have contradicted yourself and been wrong about what the IPCC says on this several times already, so please read my comments there carefully and perhaps we can come to a real resolution here.
Arthur–
I’ve read what you say, and quite honestly, I don’t know what precisely you are trying to is a misunderstanding of anything.
1) Yes. I mispoke when, in comment, #7487 I replied to you saying the words of the document say “tropical”. Prior to that, I had not mispoken in this way. Boris says tropical in the bit I quoted. The IPCC did only use the word troposphere.
What I had said before that comment– several times– was authors say they are discussing the graphs: The graph shows a red blob in the tropical troposphere. Real Climate identifies that spot as the tropical troposphere. Are you objecting to the idea that the authors, when pointing to the image are expecting the readers to see the red denoting a relatively large amount of warming in the region everyone knows is the tropical troposphere? (You were objecting that they weren’t discussing this before. But they were discussing the warming, and pointing to a figure that people recognize as at a minimum enhanced warming in the tropical troposphere or alternately the a hot spot in the tropical troposphere.)
*) “That’s an enhancement, but hardly worthy of much discussion, ” Not worthy of discussion? Yet RC also discussed it. And many discuss others it. So, notwithstanding your opinion that it’s not worth discussing, some people think it is worth discussing.
*) “Why would you deny the clear contours shown in these graphs, while focusing solely on the one for GHG’s? ” What do you think I am denying?
I described the contours you are describing. They contours in 9.1A differ from those in 9.1 C. The degree of warmth differs. If use the word “hot” to describe the warmest regions shown in graphs and “cold” to describe the coldest, and warm or cool for levels in between 9.1A has no hot regions.
There can be no “hot spot” when nothing is “hot”.
*
Of course. There is a factor of three difference; that’s a significant difference If we were discussing water, 30F above room temperature vs. water 90F above room temperature, wouldn’t one be only warm and the other hot? Also -90F below room temperature is much colder than -30 below room temperature.
I’m very puzzled by your suggestion the two temperature differences are insignificant. If 1C is not significantly different fro 0.3C,how do you see any patterns at all? After all if +1C is not significantly different from +0.3C, how is +1 C significantly different from -1C? If those aren’t different, the entire graph is should be interpreted as a constant temperature? If that is so, then all fingerprints vanish!
The way I see it, in the context of climate change a 1C change represents great warming. So, in this context I say “hot” for +1C. This usage isn’t idiosyncratic to me. So did Boris above. JohnV has been using that terminology. So do lots of people. I’m sure we’d all admit that +1C is not “hot” in other context of furnaces or oven temperatures– but if we can’t use “hot”, “warm”, “cool” and “cold” with a sliding scale, we will lose comparative terms.
4)
Actually, the title of the post doesn’t mention the IPCC. 🙂
That said, in the images selected by the IPCC to show what changes models predict to have occurred based on the forcings that have actually occurred on earth the IPCC does not expect what Boris referred to as the hot spot to have been caused by any forcing other than GHGs.
Your argument seems to be that it would arise if the solar forcing was larger than actually occurred.
Sure. But so? The IPCC does not not anticipated that hot spot for solar forcing with variations we experienced. That type of forcing (solar) at that specific level falls under the “any and all” designation. It isn’t expected to have created a “hot spot”.
With respect to predicting testable features that one might expect to have appeared on the earth over the past century, the IPCC predicts a hotspot could have arisen from ghg’s but not solar forcing. This is the topic they discuss.
I would suggest that the reason for the many comments to this post is NOT that Lucia is wrong. It’s because strongly held beliefs are being threatened.
Why do the models predict tropical mid troposphere warming with GHG forcing, and why does it not appear to be present to the extent predicted?
That is the real science question, and armwaving discussions about what other forcings might produce, even it technically sound, don’t answer that question.
Lucia, are you now deciding what is and isn’t a hotspot? Have you defined one? Why isn’t the yellow blob on solar forcing (Fig 9.1a) a hotspot?
“the IPCC predicts a hotspot could have arisen from ghg’s but not solar forcing. This is the topic they discuss.”
“Hotspot” is never mentioned in the document
So we have a fingerprint of GHG
“Greenhouse gas forcing is expected to produce warming in
the troposphere, cooling in the stratosphere, and, for transient
simulations, somewhat more warming near the surface in the
NH due to its larger land fraction, which has a shorter surface
response time to the warming than do ocean regions (Figure
9.1c).”
That is the spatial and temporal pattern, or fingerprint. The whole lot. No mention of hotspots in the tropical troposphere.
For Solar
“Solar
forcing results in a general warming of the atmosphere (Figure
9.1a) with a pattern of surface warming that is similar to that
expected from greenhouse gas warming, but in contrast to the
response to greenhouse warming, the simulated solar-forced
warming extends throughout the atmosphere”
That is the spatial and temporal pattern. No mention of hotspots here either.
No mention of hotspot’s by the IPCC, just your own interpretation of the figures.
Duane
“It’s because strongly held beliefs are being threatened. ”
What beliefs? That the definition of fingerprint isn’t widely known?
“Why do the models predict tropical mid troposphere warming with GHG forcing, and why does it not appear to be present to the extent predicted?”
Yes it is a scientific question that needs answering. But this is not the place you’ll find the answer. The problem is that science doesn’t yet have the ability to get that data. However, we have data that matches the rest of the fingerprint. We have the warming near the arctic and the cooling stratosphere.
I keep reading the statement over and over that the “hot spot” is a feature of any type of warming of the tropical surface, not just AGW … but I find that very difficult to believe. Why should any warming, regardless of the source, show the same pattern?
All I get from anyone on this question is that it has to do with the lapse rate. But that’s just a statement, not evidence.
Does anyone have a single citation of any study showing this is true? Because I’ve looked pretty hard without finding one.
w.
Boris says:
“Once again, the presence/absence of a TT hotspot does not distinguish between forcings (impossible for some people to admit, I know).”
Anthony and the rest of the loony warmers please take note of this.
The so-called hot spot is an indication of forcings and increased temperature due to these forcings. I believe most of us can agree on this.
If we have a so-called hot spot it is so minor that we can not be sure it is there. What does this say about the magnitude of the warming??
It says there is no warming due to these many and various forcings or that the warming is too small to worry about.
Thank you for your agreement. Can we all get on to more important things like how to get off this planet and explore the universe in person now??
Boris, do you get it yet?? The AGW physics says no hot spot no warming. PERIOD. That means GG warming ALSO!!!!!!!!!!!!!!!! Of course, this is why y’all have to be so argumentative over inconsequential issues. You are WRONG and don’t want to admit it!!!
By the way, I disagree with the statement that you can not tell the type of forcing by the hot spot characteristics, but, it is not worth arguing about when the real issue is its magnitude and whether it is getting hotter or NOT from ANY source.
IF we HAD a hot spot, THEN it would be worthwhile arguing over its other characteristics.
Lucia, any possibility that you would have the data handy as to what the actual temps of the hot spot should be, based on your average global surface temperature compared to the output of one of the 3C per century model runs?? Oh, and along with the temp of the missing hot spot, shouldn’t the models have a projection of strat temp and tropopause height also?? I think it would be quite interesting to be looking at more than just the projections of surface temps!!
By the way Nathan, when someone goes out of their way to make a particular area of a graph, signifying elevated temperature, bright red, you don’t think they want you to think of it as HOT?????
HAHAHAHAHAHAHAHAHAHAHAHAHA
Rotten red herrings
glow sulphurous in the dark
I pinch my nostrils
(apologies to DeWitt Payne (Comment#7378)
December 19th, 2008 at 2:48 pm;
I arrived late to this thread and found that early bit of imagery irresistible)
(and besides, it looks like you could all use some levity)
Nathan
““Why do the models predict tropical mid troposphere warming with GHG forcing, and why does it not appear to be present to the extent predicted?â€
Yes it is a scientific question that needs answering. But this is not the place you’ll find the answer. The problem is that science doesn’t yet have the ability to get that data. However, we have data that matches the rest of the fingerprint. We have the warming near the arctic and the cooling stratosphere”.
Nathan please explain to me in clear terms why science has the data for the cooling of the stratosphere but doesn’t yet have the ability to get the data for mid troposphere warming?
I’m all ears. Clue = confirmation bias & it certainly isn’t science.
Nathan
““Why do the models predict tropical mid troposphere warming with GHG forcing, and why does it not appear to be present to the extent predicted?â€
Yes it is a scientific question that needs answering. But this is not the place you’ll find the answer. The problem is that science doesn’t yet have the ability to get that data. However, we have data that matches the rest of the fingerprint. We have the warming near the arctic and the cooling stratosphere”.
Nathan please explain to me in clear terms why science has the data for the cooling of the stratosphere but doesn’t yet have the ability to get the data for mid troposphere warming?
I’m all ears. Clue = confirmation bias & it certainly isn’t science.
lucia; commendable patience; as one who has engaged Arthur I can appreciate your fortitude and attempt to maintain the focus of the ‘debate’; the THS is gone; it required an increase in SH and alteration to the moist adiabatic lapse rate; there has been no increase in atmospheric SH at the relevant levels for AGW theory;
http://wattsupwiththat.com/2008/06/21/a-window-on-water-vapor-and-planetary-temperature-part-2/
Nor has RH increased or remained constant as AGW requires. That this discussion has been primarily about semantics is both extraordinary and typical; the art of sophistry is not to reveal meaning but to obfuscate its absence; all that AGW has left now is stratosphere cooling; may I suggest a thread on that remaining AGW “fingerprint”? It would be interesting to see what verbal gymnastics that invokes.
Well in 186 posts I have learned that :
a) There is no fingerprint of GHG warming because anything that it might theoretically cause could also be caused by something else . This is a rather trivial statement for a complex interacting system out of equilibrium . Everything is caused by everything and reciprocally .
b) A well defined volume with characteristically higher temperature than its surrounding may under no circumstances be called a hotspot .
c) Computer games also called GCM (Trade mark : Physics Inside) include all existing and not yet existing natural laws so GHG forcing among others and produce a prediction featuring a well defined volume with characteristically higher temperature than its surrounding .
From b) we infer that this red blob can’t be a hotspot and from a) we infer that it is uninteresting . Therefore the silly detail that in the real world nothing such happens is irrelevant .
Lucia the time is the most precious thing we have been given on this world and you are really too patient with people who make us loose it .
I know why there are so many comments. Duane is right. When you get down to the bare bones, the argument for AGW is word games. That’s why this so interesting.
We’ve got scientists arguing over irrelevancies. What else is new under the sun? That’s what the AGW debate is about. Here we get to see it on parade. Fascinating from a human behavior perspective.
Andrew ♫
KuhnKat–
Good morning all. I’m sipping my coffee and skimming and responding to things I can quickly comment on:
SteveM was discussing temperature of the troposphere. This seems to have triggered the repetition of the now important “meme” that the hotspot is not a fingerprint. I think Steve is trying to explore how to access model data more easily. I don’t know where the reservoir of real world data is. You could read the Santer paper, get the references and hunt down the archives.
Nathan– No one is denying the stratosphere has cooled and that that is also a fingerprint for a few of the forcings.
Roger– I like Arthur and he makes good points. But yes, all the stratospheric cooling is left, and while the hot spot fails to appear, the talking points appear to be that somehow this is not a problem for the models even if it never appears. The “reason” seem to be non-uniqueness (shared by many things called fingerprints).
The idea that the hotspot not appearing would not be a problem for the models clearly wrong: If the data were sufficiently accurate, and the hot spot wasn’t there, it would be a problem for the models. Models and theories demonstrate their power by predicting things that were not seen before the data are available. The models predict that feature. Regardless of whether or not the feature is unique to ghg, it is predicted to arise as a result of ghgs. If it doesn’t appear, that would be evidence of some lapse in their predictive ability.
Roger Pielke Jr., they’ll need to tweak their Antarctic models to get a curve that’s consistent with the ERSST.v3 version of the Southern Ocean SST anomalies.
http://i35.tinypic.com/s3djds.jpg
Regarding Roger Pielke, Jr. (Comment#7399).
Our paper on Antarctic temperature change isn’t out yet, but news of it is getting out due to my AGU talk. We of course anticipated that people would — without having read the paper — complain that “first scientists say Antarctica is cooling, now they say it is warming”. Well, that’s right. They are both consistent. Our paper shows that over the last 50 years, Antarctica has been warming. If you look just at the post-ozone-hole era, East Antarctica (only) has been cooling. That’s very consistent with what the models show. Indeed, our paper shows that our understanding of the recent cooling of East Antarctica passes the falsifiability test beautifully. That understanding would have predicted that East Antarctic cooling didn’t start until the late 1970s or later. That’s exactly what we found.
.. and you can see this in the SST data posted by the previous poster. Note how the cooling starts right around 1980.
Lucia – you seem not to have even noticed up to this point that the RealClimate graphs showed warming considerably larger than the amounts in the IPCC’s figure 9.1 – over 2 C I believe in the central “hot spot”. So if 0.9 C in the IPCC’s figure is “hot”, then what are the RealClimate numbers, “scorching”?
The reason for the difference is that the RealClimate graph was calculated for a 2xCO2 situation, while the IPCC figure 9.1 is only for the 20th century rise in CO2, about 25%.
Meanwhile, the actual observations of the tropical troposphere don’t date back nearly so far; we have only 30 years or so of satellite date, during which CO2 increase has been about 15%. So the “hot spot” that would be expected from observations over those 30 years would be only about 0.5 C (assuming CO2 changes were the only factor of interest). While the surface warming over that period would be about 0.3 degrees, for a 0.2 degree differential warming of the mid-troposphere. Measurable? Not easily I think! And how does that small 0.2 C difference count as “hot” in your book, when the same 0.2 C difference that could be estimated from the solar image in IPCC’s figure 9.1a is not considered “hot”?
Lucia – you know better than this. Pattern and magnitude are orthogonal, and the IPCC, Douglass, and RealClimate discussions are all about *pattern*, not about the magnitude of the changes. And the *pattern*, from any source of warming, is the same in the tropical mid-troposphere. Where it differs is the important question for attribution, which is what IPCC 9.2.2 is about; features that are the same, like the hot spot, are not relevant, and not discussed there by the IPCC.
As I mentioned earlier, IPCC *does* discuss the hot spot (i.e. the change in lapse rate) in section 9.4.4.4, which is very informative, and I recommend to everybody who has questions about what the IPCC WG1 group really thought on the subject.
Arthur–
Why do you think I missed that point about the magnitudes. Are you suggesting I wouldn’t use the word scorching for those levels? Yes. RC showed graphs using levels of forcing that haven’t happened while criticizing a paper that was comparing observations to predictions of what would happen on earth.
I thought you said we were not to relate the issue of whether or not the hot spot is expected to the context of the RC post. I assume that’s permitted again? Real Climate’s post is about introducing a distraction to make it appear that people who are comparing observations to the predicted pattern of warming in the vertical direction of the troposphere is somehow confused.
If their point had been merely to say hot spot is difficult to measure because only 0.5 C is difficult to measure, they should make that point by discussing data uncertainty only. With respect to the issue of data uncertainty, introducing the any ambiguity of the cause (solar vs. ghg) is an red herring because that is irrelevant to the issue of detecting the spot.
RC should have cut that little lecture on the physics out of that article. Had they deemed the issue interesting otherwise, they could have written it up as a separate article totally unrelated to commenting on Douglas. Of course, they would only do that if the hotspot was worth mentioning in and off itself. Possibly, as you suggested before, it is not worth mentioning (unless someone is trying to insinuate something bizarre about Douglas trying to whether the predicted warming exists.)
Eric Steig: Oops! That graph is from the old data set. Smith and Reynolds updated it within the last few months. Sorry about the confusion on my part. The Southern Ocean cooling based on the ERSST.v3b version doesn’t start until 1993.
http://i44.tinypic.com/2uen29u.jpg
So, again, why is 0.9 C “hot” and 0.5 C also “hot” but 0.3 C (figure 9.1a) not hot?
To be precise, let me say what I believe everybody (except perhaps Lucia) means when they say “hot spot”:
Definition: the “tropical mid-troposphere hot spot” is a *pattern* where the tropical mid-troposphere warms at a faster rate than the surface does.
Under that definition, “ANY and ALL” sources of warming, when modeled, result in a “tropical mid-troposphere hot spot”, so everybody who is familiar with the models (and/or the lapse rate reasoning behind the result) does expect that hot spot.
Now, you clearly have a different definition, because you state in the main post:
Can you clearly state your definition so that it fits that comment, and also fits the discussion by Douglass on recent warming, which would have a significantly lower magnitude?
Eric-
I haven’t seen your paper, and I do look forward to it. Surely, science advances and sometimes old views are replaced by new ones.
However, if your paper does what was reported — “calls into question existing lines of evidence” — then it means that the new study is inconsistent with previous understandings.
Logically then it cannot be the case that models that were able to explain cooling are also able to explain that new research that calls that evidence of cooling into question and replaces it with evidence of warming.
So whatever your study actually says, given the very loud explanations that a cooling Antarctica was to be expected, you will need to come up with a far better explanation for why the models were consistent with the earlier evidence of cooling and are now also consistent with the new evidence of warming that overturns the earlier evidence. Otherwise, we are back in a situation where cooling warming and everything in between are judged to be “consistent with” the models.
But perhaps the reporting was wrong and the new evidence merely adds to the earlier evidence rather than calls it into question?
Arthur–
Earlier, you asked me how I could call 0.9C “hot” and not call 0.3C “hot”. I explained this: 0.9C is hot relative to 0.3C. Both are discussed in the same context: Changes in temperature associated with 20th century warming.
If you want to discuss changes in temperature associated in the measurement period 0.5C is hot relative to smaller values of warming that are only “warm”.
Are the changes associated with the fictional warming even warmer? Yes. I think you suggested “scorching”. I did not object to that word.
Arthur: My point is that on earth, if the a “hotspot” were detected in the troposphere, this would be attributed to GHGs, not solar. This is because the contribution of solar would be deemed too small. At currentl levels, relative to the contribution to GHG, solar can only provide warmish amounts.
As for the seeming contradiction– which appear to be your objection it goes like this:
1) RC placed their discussion in the middle of Douglas where the fictional forcings are irrelevant to the issue of comparing observations to the predictions.
2) Had they linked to predictions– or levels that are realistic to the earth, or shown such images like the explanatory ones in 9.2,2– those images would have appeared distinctly different from the images they chose to show.
3) Had readers clicked those links, the images are so different, they would seem to contradict the storyline at RC.
4) Puzzled readers would ask why the images look entirely different.
5) The answer is: The images RC was showing were irrelevant to Douglas. The images in the IPCC are more relevant to Douglas which is comparing observations to projections for things that might have actually happened, not to things that might happen under fictional levels of forcing.
So, the seeming contradiction can be resolved. But the resolution involves explainig the images in RC are irrelevant to the paper they are criticising, they are irrelevant to the question of testing the models and they whole discussion of the ambiguity is irrelevant.
A real physical hot spot is not expected to arise for any levels of solar forcings that could happen on the actual earth. It doesn’t matter what you call it, it doesn’t matter if you can run models with fictional forcing to play what if games: No one believes solar forcing can vary sufficiently to cause a that spot to be “hot”. GHG forcing might if we pump that amount of GHG’s into the atmosphere.
If you look at the monthly Southern Hemisphere temperature anomalies, there is a very unusual trend and a great deal of variation – quite a bit different than the northern hemisphere and the tropics.
http://img122.imageshack.us/img122/7797/shtempanomalybq1.png
I imagine a bit of cherry picking from this dataset and you get any trend you wanted. On the other hand, it does match the southern Atlantic SSTs pretty closely.
Eric: And here’s a graph of the combined land plus sea surface temps [ERSST.v3b] from the NCDC. It too shows the peak at 1993.
http://i42.tinypic.com/23r5kzb.jpg
From 9.2.3 “”Similarities between the responses to different forcings, particularly in the spatial patterns of response, make it more difficult to distinguish between responses to different external forcings, but also imply that the response patterns will be relatively insensitive to modest errors in the magnitude and distribution of the forcing. Differences between the temporal histories of different kinds of forcing (e.g., greenhouse gas versus sulphate aerosol) ameliorate the problem of the similarity between the spatial patterns of response considerably. For example, the spatial response of surface temperature to solar forcing resembles that due to anthropogenic greenhouse gas forcing (Weatherall and Manabe, 1975; Nesme-Ribes et al., 1993; Cubasch et al., 1997; Rind et al., 2004; Zorita et al., 2005). Distinct features of the vertical structure of the responses in the atmosphere to different types of forcing further help to distinguish between the different sources of forcing. Studies that interpret observed climate in subsequent sections use such strategies, and the overall assessment in this chapter uses results from a range of climate variables and observations.””
If the “response patterns will be relatively insensitive to modest errors in the magnitude and distribution of the forcing” and “(D)istinct features of the vertical structure of the responses in the atmosphere to different types of forcing further help to distinguish between the different sources of forcing”, in what way is Lucia incorrect? The 9.1 Figure definetly has spatial and temporal distributions shown. Section 9.2 does not contradict Lcia, but rather supports that a characteristc pattern is a “fingerprint”. The IPCC indicates that Arthur’s emphasis on magnitude is misplaced. In section 9.2.2.1, it is stated “”The net effect of all forcings combined is a pattern of NH temperature change near the surface that is dominated by the positive forcings (primarily greenhouse gases), and cooling in the stratosphere that results predominantly from greenhouse gas and stratospheric ozone forcing (Figure 9.1f). Results obtained with the CSIRO model (Figure 9.2) suggest that black carbon, organic matter and biomass aerosols would slightly enhance the NH warming shown in Figure 9.1f.”” The discussion indicates that the area showing altitudes, temperuture, and lattitudes is acceptable. They are describing an area much less in extant and at the same magnitude of temperature as the area Lucia/Douglas. The section would be difficult at best to understand without being able to use the graph as a visual aid to the discussions. In which case, in what way do Lucia/Douglas not use the graph and discuss as did the IPCC? I think Authur should answer this, since the magnitude is not a problem per the discussion in 9.2.2 on NH.
Lucia, you’re contradicting yourself within a single comment:
but
Solar forcing can vary – and it does vary – just on the solar cycle there’s enough variation to cause warming on the order of 0.1 C at the surface, and perhaps 0.2 C in the “hot spot”. Milankovitch cycles are triggered effectively by changes in solar forcing that lead to 5 C average changes at the surface. Between the Medieval Warm Period and the Little Ice Age there was a drop in average temperature of 0.5 to 0.8 degrees, believed due to a change in solar forcing. So such changes are certainly not out of the question.
And the issue with a *pattern*, the way I defined the hot spot in my previous comment, is that even for a very small temperature increase at the surface, the hot spot is there if that region has a larger temperature rise than the surface. So every one of those past solar forcing changes (and solar cycle-induced changes) should have seen a “hot spot” enhancement caused by the solar forcing. It most certainly *is* expected on the real Earth.
The pictures the RealClimate folks showed were for 2xCO2 forcing, and therefore showed a “scorching” hot spot. But the pattern for a 1.15xCO2 forcing, appropriate for a 30-year period, would have looked identical, when scaled to the appropriate surface temperature change. And the corresponding relative change of solar forcing is certainly within a plausible range (though obviously it is fictional given satellite data showing little change in solar constant).
Now why they focused on that issue I don’t know, certainly Douglass’ paper was, as you say, more about comparing models with observations, rather than talking about fingerprints (it doesn’t mention fingerprints or patterns at all).
But RealClimate’s faults don’t give Lucia the right to misrepresent the facts of the matter, either on what the IPCC said, or what really is expected associated with the “hot spot”.
Again, I’d like to see your own precise definition of the “hot spot” so we can perhaps come to a better understanding here.
I find these comments thoroughly entertaining as well as being (at least partially) enlightening. Seems to me that we are lacking any definition of what AGW signals are unique to AGW. Without a falsifiable signal,are we just supposed to take it on faith that all GW is attributable to AGW, and therefore, we are in big trouble from run-away global T (caused by positive feedback (unproven, and certainly unlikely). Another article of faith is what are we going to do about it. I’m unaware of any approach that will make any difference what-so-ever. I’m not in favor of turning the management of all energy sources over to the UN and just hope for the best.
Arthur–
I haven’t mistated what the IPCC said.
Ok… over very long times, yes solar forcing can vary that much. But not in recent periods and it would not be the cause enhanced warming that Douglas might or might not detect in the data set.
A “hot spot” is a region that is “hot” and has a “spot” like shape.
The IPCC figures indicate that models do not expect solar forcing of the levels we experienced to have created warming that results in a “spot”.
To be a spot things must look more or less “sort of dot” like. The word “spot” implies the region is fairly localized. Spots don’t have arms or apendages . They aren’t “S” shaped and splayed out all over the place. Spots must stand out from the surroundings and be noticable.
To be “hot” a spot must be noticably hotter relative to surroundings locations. Figure 9.1 a does not show a “hot spot” in the tropical troposphere because the region is a) barely warmed relative to the surrounding and b) not “spot” shaped.
Lucia – by your definition, Douglass et al could not have seen a “hot spot” then on both grounds, since even the models predict it would be “barely warmer than its surroundings” (0.2 C, about the same as figure 9.1 a), and Douglass et al don’t seem to have looked at the warming trends outside the tropics to determine whether any changes were confined there, or had “arms or appendages”.
On the other hand, if you grant that 0.2 C counts as “hot” enough, then figure 9.1 d clearly shows a “hot spot” in the ozone case. The real trouble is the resolution of these figures, which is terrible except in the GHG one, because they were all scaled to a common temperature scale. They were sufficient for the IPCC discussion there because they show the general patterns of overall warming and cooling which is all section 9.2.2 discussed. But if you had a higher-resolution plot for solar and the other forcings, you get much nicer plots with clear “spots” that, if 0.2 C is “hot”, are definitely “hot”.
I only just realized GISS has a whole collection of these plots based on a 2005 Hansen paper, here:
http://data.giss.nasa.gov/efficacy/
Take the “sol.irrad. 1880->2000,Lean” line in Table 3 there, click on the 100-yr response Lat-Hgt link, and click “Show Map”, and you get a big “hot spot” for solar irradiance changes for that 120-year period, of 0.2+ degrees. Resolution could be still better here too, but if you want a “hot” “spot”, there it is…
Arthur–
I didn’t define 0.2 C as “barely warmer”. It turns out that can’t be detected. So, sure, after examining the data, it turns out that 0.2 C is not enough warmer to detect.
Sure, ozone has a small hot spot detectable using the scaling to a common temperature scale selected by the IPCC, and it’s warm. But solar doesn’t. I don’t consier scaling to a constant temperture scale a problem. It’s the most appropriate as it lets the reader see which forcings are expected to have larger effects. Rescaling is deceptive because it would make it difficult for a reader to understand the order of magnitude of the effects.
I assume the IPCC decide to use the entirely reasonable choice for scaling because they wished people to graphs the order of magnitude of the various effects. To the extent any “hot spot” is visible in over the tropics, it’s due to GHG’s.
Also, note that Hansen made a different choice from the IPCC. In the GISS plots, the images for solar (Lean) use non-constant scaling. In the images for Lean, maroon is 2C; in those for GHG’s maroon is 9C. So, if you try to estimate the order of magnitude effects from that, you need to remember that the choice of temperature scaling means that much of the image in Lean would be indiscernable under the scaling for GHG’s. In otherwords.
The IPCC didn’t chose to commincate the information the way Hansen did. They chose the scaling in the figures I duplicated. So, it’s reasonable to say they were trying to communicate what they communicated using the scaling they chose. You only seem to be suggesting that had they wished to do so, they could have communicated some other idea, using some other sets of graphs — more like those GISS displays — that emphasized something other than what they chose to emphasize in their document.
Lucia,
Wow this is a bit of an onslaught isn’t it? Wonder why- attack is the best form of defence?
Seems to me some in the climate change world are a trifle worried that not all is going according to the recipe.
Have a good Christmas and New Year and may all your recipes turn out exactly right.
Dave– I don’t take the length of comments as evidence of that. A lot of people who previously lurked jump in on one side or the other. So…. maybe people are just getting chatty?
Arthur, Boris and I often go back and forth a lot. So, that’s just a disagreement. But if it were only use, the comments wouldn’t go over 40!
Well … this is getting kind of boring. I keep asking whether any of you good folks that keep stating so categorically that the “hot-spot” will arise from any surface warming have any evidence to back up your position.
I have asked here, and at Climate Audit, and somehow, curiously, everyone who has been most vociferous with the claims that the “hot-spot will come from any warming” suddenly falls silent. A deathly, boring quiet reigns.
Strange, that.
For example, Boris says:
A tropical tropospheric hotspot is expected from any warming
Which it may be … but how about providing some evidence to back up that sweeping claim? You do remember that “evidence” stuff, I’m sure. It comes from the real world, not from computers, and is generally comprised of measurements and observations of actual phenomena.
So, while (as they say) the absence of evidence is not evidence of absence … the prolonged absence of evidence is certainly an indication of the actual absence of evidence. And that’s what I’ve seen so far.
w.
Lucia, you haven’t defined what you mean by hotspot. If you can’t do that then this discussion, yet again, is pointless.
Are you a lawyer? You seem to have successfully driven this discussion well away from the original topic, which was whether or not the TT hotspot is a fingerprint of AGW.
I mentioned this before but it’s worth metioning again:
Ok, so according to you the TT hotspot is a fingerprint.
This is based on your undestanding of the IPCC definition of fingerprint, which says “The climate response pattern in space and/or time to a specific forcing is commonly referred to as a fingerprint. â€
Now, section 9.2.2.1 is called “Spatial and Temporal Patterns of Responseâ€. Which sounds somewhat familiar? They perhaps could have shortened that title to “Fingerprints†– possibly it didn’t sound as fancy.
They then say, “these figures indicate that the modelled vertical and zonal average signature of the temperature response should depend on the forcings†– so the modelled response is the spatial and temporal pattern. That is, it is the whole thing.
So in the IPCC we have a ‘fingerprint’ (that is the Spatial and temporal pattern) of GHG
“Greenhouse gas forcing is expected to produce warming in
the troposphere, cooling in the stratosphere, and, for transient
simulations, somewhat more warming near the surface in the
NH due to its larger land fraction, which has a shorter surface
response time to the warming than do ocean regions (Figure
9.1c).â€
That is the spatial and temporal pattern, or fingerprint. The whole lot, and no mention of hotspots in the tropical troposphere.
For Solar
“Solar forcing results in a general warming of the atmosphere (Figure
9.1a) with a pattern of surface warming that is similar to that
expected from greenhouse gas warming, but in contrast to the
response to greenhouse warming, the simulated solar-forced
warming extends throughout the atmosphereâ€
That is the spatial and temporal pattern. No mention of hotspots there either.
You are looking at the two spatial and temporal patterns (or fingerprints) and have seen one is yellow and the other red and decided that this difference is a fingerprint. However, by the IPCC definition of fingerprint this isn’t so.
Willis
“I have asked here, and at Climate Audit, and somehow, curiously, everyone who has been most vociferous with the claims that the “hot-spot will come from any warming†suddenly falls silent. A deathly, boring quiet reigns.”
Why are you asking here? Why don’t you use google scholar?
Nathan, thanks for the reply. I’m interested in what those who make these claims are using as their support. What google scholar reveals may or may not be what they are relying on, so it doesn’t help in this case.
In addition, it is common scientific practice that if someone says “Cite?”, it is the responsibility of whoever is making the claim to provide some support for the claim. Why are the “hot-spots come from any kind of heating” crowd so unwilling to follow that practice?
Having said that, I have in fact investigated the question on Google Scholar and elsewhere. I find things like this:
Tropical observations show convective activity increasing sharply above sea surface temperatures (SSTs) of around 26°C and then decreasing as the SST exceeds 30°C, with maximum observed SSTs of around 32°C. Although some aspects of this relationship are reasonably well understood, as of yet no theory has explained the decrease in convective activity above 30°C.
That’s interesting, and it fits with my experience of the tropical climate (I live in the deep tropics) but it hardly supports the claim that any warming will lead to a hot-spot. Instead, it supports the idea that climate is a complex beast, with lots of emergent phenomena which often work in opposition to each other, and which is neither linear nor particularly predictable. In such a system and with no evidence to support the claim, the idea that no matter how you heat the surface the tropospheric reaction will be identical is … well … let me say that to me, such a conclusion seems charmingly naive and quite premature, and leave it at that.
w.
“Lucia, you haven’t defined what you mean by hotspot. If you can’t do that then this discussion, yet again, is pointless. Are you a lawyer? You seem to have successfully driven this discussion well away from the original topic, which was whether or not the TT hotspot is a fingerprint of AGW.”
That is what is called an own-goal.
And BTW,the rest of your comment wasn’t worth mentioning before and thus wasn’t worth mentioning again. Let me give you one reason why. If, as the IPCC says, “Greenhouse gas forcing is expected to produce warming in the troposphere, cooling in the stratosphere, and, for transient
simulations, somewhat more warming near the surface in the
NH due to its larger land fraction, which has a shorter surface
response time to the warming than do ocean regions” then the TT hotpot is a part of the GHG-fingerprint. If you had everything else but no warming in the troposphere (and not merely warming in the troposphere but warming concentrated in the tropical troposphere; that is, a TT hotspot, or else figures 9.1c and f are misleading) then you would not have GHG-forced warming; you would in fact have something else. Yes, a pattern within a pattern; computers match fingerprints by comparing characteristic squiggles within a representative pattern of squiggles called a fingerprint.
Willis, don’t say I t don’t do anything for you:
http://arxiv.org/PS_cache/arxiv/pdf/0811/0811.4600v1.pdf
Something by your good friend Arthur Smith.
It’s not actually that hard to find information. The best place to start would be your local university, ask some physicists.
Dover Beach,
“That is what is called an own-goal.”
No, I was just summarising the discussion. It was Lucia who declared the TT hotspot a fingerprint.
“then you would not have GHG-forced warming; you would in fact have something else.”
You go too far. You made the last bit up.
And your fingerprint anaolgy is flawed. With only a partial fingerprint a match could still be made.
There is another complicating factor as you are demanding that the ‘fingerprint’ match exactly. Figure 9 is an illustration, it is not an exact replica or prediction.
So “The ability to distinguish between climate responses to
different external forcing factors in observations depends
on the extent to which those responses are distinct (see, e.g.,
Section 9.4.1.4 and Appendix 9.A).”
The problem as has been explained many times is that Lucia is loose with her definitions and then decides what the definition means as she uses it. If she had announced at the start what she was defining this whole discussion probably would never have arisen..
From p. 2 of Smith’s paper linked directly above:
“Whatever the measurements and theory sort themselves out to on this, note again that tropical mid-troposphere temperature trends are not a signature of greenhouse gases, and this whole argument has no bearing on” CO2 forcing.”
This is clearly wrong. From a paper attempting to correct errors of logic and interpretation, we should expect something more than this. If Smith had said, the TT hotspot was “not a signature of greenhouse gases [alone]”, he may have had a point, but the absence of a TT hotspot of course does have a bearing on “CO2 forcing” since it is one of all the forcings itself that may cause a TT hotspot and a TT hotspot is one of at least three characteristic patterns expected from GHG-forcing, the others being surface warming and stratospheric cooling. It also should be noted that the warming in the TT is expected to be more pronounced then at the surface, according to fig. 9.1 Chapter 9 AR4.
If, as Smith also writes, that “warming at the surface,
and cooling in the stratosphere… is observational proof that the sun cannot be behind recent warming” then the lack of a TT hotspot is observational proof that GHG-forcing cannot be behind the recent warming, at least for the time being.
One law for the lion and another for the lamb is tyranny.
There are other problems with his papers. He says, for instance, “But the surface and low-altitude warming (or cooling) patterns are essentially the same across all the forcings”. If patterns are not the same, they are different even if we like to think of them as “essentially the same” and these differences are noticable and therefore not essentially the same for historically observed forcings over the last century. They are not even “essentially the same” with respect to the troposphere at high latitudes for a doubling of CO2 and a 2% increase in solar forcing. In other words, we may in fact be able to distinguish between the two simply by the pattern of warming that occurs in the troposphere.
Willis, irrespective of what Nathan says, you will find no joy (evidence) in that paper either regarding your question. Lets just hope Santa is more generous.
Nathan, please don’t lecture me on how to find information. I asked what you and others here were using to buttress your claim that the hotspot comes from any warming. That information is not on Google, nor is it in any library or university, despite your glib attempt at an insult.
The paper you cite says:
The tropical mid-troposphere “hot spot†that Monckton highlights is not a “ï¬ngerprint†of greenhouse gases: it is well known to be a consequence of higher water vapor levels in a warmer world, whatever the cause of the warming. As warm air rises, it cools almost adiabatically – this is known as the “lapse rateâ€, and stability of the atmosphere ensures that temperatures fall no faster than this rate with altitude. When air holding water vapor rises and cools, some of the water condenses and releases heat, resulting in warmer air at a given altitude, and a lower lapse rate.
Unfortunate, once again all we have is his unsupported assertion, he provides no more citations for this claim than you do. He, like many people making the claim here, just says it is “well known” … well, perhaps it is well known, but is it based on any evidence? At present, like you, the good Arthur wants us to take his word for it … sorry, that’s not science, that’s anecdote.
In addition, he seems woefully ignorant about thunderstorms. Yes, when rising water vapor cools and condenses, it warms the immediate surroundings. But the immediate surroundings are not the troposphere. The immediate surroundings are inside the thunderstorm, so the net result is that air inside the thunderstorm is heated and rises up to near the tropopause. In the process, much of the heat is turned into work. Of course, the part of the latent heat that is turned into work doesn’t heat a damned thing. Instead, it performs work … but oops, Arthur forgot to consider that. Remember that the Hadley cells, and ultimately the entire atmospheric circulation, are powered by this work done by tropical thunderstorms, so we are talking about a huge amount of energy which does not heat anything at all, much less the troposphere.
Next, consider what happens to the air exiting the thunderstorm at the top. It has to sink (to replace the warm air sucked up at the base of the thunderstorm). Now, if we are constantly taking warm air from the surface, removing the heat (through condensation and work) inside the thunderstorm (without interacting with the troposphere) and turning it into cold air up by the tropopause, what will happen to the tropospheric temperature?
Well, in general, as the process continues the tropospheric air will be gradually replaced with cool dry air from aloft … perhaps you can explain how that will warm the troposphere?
In other words, his simplistic idea (that if water vapor condenses at a certain altitude it perforce must warm the troposphere at that altitude) is incorrect.
The author (Arthur Smith) is merely repeating your claim, that all warmings lead to the same hotspot … and just like you, he provides no evidence and no citations to back up the claim. You citing him is no better than you agreeing with say Boris. Since none of you are providing evidence or citations to evidence to support your claim, but all are just repeating some variation of the bromide that it is “well-known”, you saying that Arthur agrees with you means no more than if you say Boris agrees with you.
w.
Roger (nt#7573) wrote:
‘However, if your paper does what was reported — “calls into question existing lines of evidence†— then it means that the new study is inconsistent with previous understandings.’
That’s not me. That’s Nature Reports Climate change, writing about a paper they haven’t read or talked with me about. They can’t have talked with me about it, of course, since that would be breaking the embargo with Nature! (I’m not allowed to talk with reporters about this until the week it is in print, and I don’t know the date of that yet). Nature Reports really shouldn’t have done this, because it just leads to ill-informed speculation that I’m not really even aloud to comment on.
When the paper comes out, I suggest you read it. Indeed, it is usually a good policy to read a paper before writing about it.
Willis
“Nathan, please don’t lecture me on how to find information.”
You asked…
Truly, if you are going to understand this you need to find out for yourself. Why do you want others to point it out for you? The only reason you ask what I am using is so you can criticise it (as you did above).
“Yes, when rising water vapor cools and condenses, it warms the immediate surroundings. But the immediate surroundings are not the troposphere. The immediate surroundings are inside the thunderstorm, so the net result is that air inside the thunderstorm is heated and rises up to near the tropopause. In the process, much of the heat is turned into work. ”
Willis, where is the troposphere?
L
“Smith: I’ll go through it with you line by line if you’re really having trouble, it’s not that hard to understand”
“Supporting commentary in the text of the report (though perhaps not very well worded) to follow, if you really want me to.”
I’ve read more and …
I’ve worked it out: He’s a grammar teacher!!
What a superior tone!
Still I must admit he’s at least taken up the cudgel not like most “It’s already settled” “I told you.”
I must keep reading.
Oh dear Boris has started “This is as simply as I can put it.”
Sorry L. I gave up after Boris at 7452. How utter arrogant and condescending. Didn’t he learn from last time?
But you clearly don’t need me to defend you.
savo
Eric-
Fair enough.
So Nature Reports Climate Change has it wrong. As Nature is a source of record, you can then understand why there might be some confusion.
I look forward to seeing what your paper actually says.
So, the sonde data that goes back further is to thrown out because it doesn’t match the expected cooling.
How many forms of data does that make that the warmers have adjusted or thrown out because they are inconvenient??
Just about all of them I think. I guess when you have perfect models and KNOW the answer you do not need imperfect observations!!
Having read Nathan’s snarky “It’s not actually that hard to find information. The best place to start would be your local university, ask some physicists.” reply to Willis, I smiled and gleefully scrolled down in anticipation of Willis’ retort.
I was not disappointed. Nothing left but the feathers.
Nathan, you really should be a little more careful.
BillBodell
” I smiled and gleefully scrolled down in anticipation of Willis’ retort. ”
Yes BillBodell, that really showed me, didn’t it?
Actually I am sorry I posted here, I thought there’d be reasonable discussion. But it’s little more than cheerleading.
Anyway have a lovely Christmas and pleasant New Year
Nathan, are you saying it is “cheerleading” to ask you for some kind of reference, backup, study, evidence of any kind to support your claims? I hope not, as have not taken a position for or against. I’ve just asked people to back up a claim I find dubious (that all types of heating lead to the same “hotspot” result) with some evidence. I’m trying to have a reasonable discussion of the issue and to find and examine any evidence, exactly the discussion you say you thought you might find. I’ve given the reasons that I find it dubious … so it’s your turn in the reasonable discussion.
Now I can understand if you don’t have such evidence. I myself have made claims without checking my facts first, I’ve put my trust in someone’s claim that something is “well-known” … usually to my lasting regret, I might add, but I can’t deny having done it.
I could even understand if you have such evidence but don’t want to reveal it. I can think of a couple of legitimate reasons why someone might do that.
Heck, I can even see why you might not want to provide the evidence to me, on the basis that you don’t like the way I part my hair or how I phrase my sentences. Although it would be cutting off your nose to spite your face, I could still understand you doing it, I’ve done things that self-defeating more than once, simply because I was upset with someone.
However, when you decline to provide any evidence or support for your claim, on the basis that you don’t like “cheerleading”, your statement that you came here for reasonable discussion begins to sound a bit hollow. There’s always random noise on any blog, and this one is no exception — if you don’t like it, just ignore it.
And when you want to leave on that basis, without providing any support for your claim, and wish us a happy Christmas on your way out … well, I leave it to others to assess your possible motives for that. I can only look at the result.
No evidence.
w.
Rather than arguing about the semantics of the word hot, isn’t the point in question ‘what is meant by a spot?’
I find it interesting that the RC graphs use a linear scale to suppress the key difference between their two graphs, but the difference is still there – if you look for the specific difference which has been called out as a distinguishing factor.
Willis,
In this post:
http://www.realclimate.org/index.php/archives/2007/12/tropical-troposphere-trends/
Gavin runs modelE with similar magnitude solar and CO2 forcings. Both responses show the TT hotspot.
Of course, if you don’t trust RC, you could always download a model of your choice and perform the experiment yourself. There are some fingerprint analyses that might also show the hotspot in response to various forcings. K. Hasselmann appears to be a name to look into, but I haven’t read any of the papers.
In 9.2.2.1, the IPCC authors state about the figures and models used “These figures indicate that the modelled vertical and zonal average signature of the temperature response should depend on the forcings. The major features shown in Figure 9.1 are robust to using different climate models.”
They go on to say “Greenhouse gas forcing is expected to produce warming in the troposphere, cooling in the stratosphere, and, for transient simulations, somewhat more warming near the surface in the NH due to its larger land fraction, which has a shorter surface response time to the warming than do ocean regions (Figure 9.1c). The spatial pattern of the transient surface temperature response to greenhouse gas forcing also typically exhibits a land-sea pattern of stronger warming over land, for the same reason (e.g., Cubasch et al., 2001).”
Lucia has 9.1c circled at the top of the post. The volume under discussion by the IPCC is about 45N to about 90N, from about 0 to 4 kilometers, and 0.6 to 1.2 C. As I see it, the volume under discussion is 30N to 30S, from about 4 to 14 kilometers, and 0.6C to 1.0C. The 9.1C is contrasted by the IPCC by the following “The simulated responses to natural forcing are distinct from those due to the anthropogenic forcings described above. Solar forcing results in a general warming of the atmosphere (Figure 9.1a) with a pattern of surface warming that is similar to that expected from greenhouse gas warming, but in contrast to the response to greenhouse warming, the simulated solar-forced warming extends throughout the atmosphere.”
The IPCC also support (unsuprisingly) Lucia about the magnitude of solar by stating “The peak-to-trough amplitude of the response to the solar cycle globally is estimated to be approximately 0.1°C near the surface.”
In case there is any doubt about the claim of Lucia or the IPCC about pattern recognition, there is this “The net effect of all forcings combined is a pattern of NH temperature change near the surface that is dominated by the positive forcings (primarily greenhouse gases), and cooling in the stratosphere that results predominantly from greenhouse gas and stratospheric ozone forcing (Figure 9.1f).” The patten claimed here by the IPCC’s reasoning is the same claimed by Lucia. The IPCC is using a smaller volume to describe “the pattern of NH temperature change near the surface that is dominated by the positive forcings.”
By comparing as the IPCC do, 9.1a with 9.1c, 9.1f, the IPCC has identified an area AGW, as does Lucia/Douglass. This also shows that the previous comments about the magnitude by certain posters do not agree with the methodology and information of section 9.2.
Nathan–
In answer to your question, I am a mechanical engineer. My area is fluid mechanics, and I have published research articles in Physics of Fluids, International Journal of Multiphase Flow, The Journal of Fluids Engineering. I’m perpetually puzzled by those who bring up the idea that some people might be lawyers and/or demands that people provide credentials.
I agree with you that section 9.2.2 is about fingerprints, and the IPCC is discussing fingerprints. If you scroll back and read previous comments, you will see that has been my contention all along.
Based on the tone, and lack of substance of many of your later comments, I’m going to not engage most of the remaining stuff. I have discussed precisely all those quotes, and they appear in various forms already on the thread.
Eric—
Interesting. I have often thought a great deal of confusion is sown by Nature, PR departments.
Kuhnkat
Which sonde data were thrown out? I heard a lot of discussions of bad sonde data over the dinner table way…. way… back. It was bad. But I don’t know if it’s related to the sonde you are discussing:
http://ams.allenpress.com/perlserv/?request=get-document&doi=10.1175%2F1520-0426(2003)020%3C0117%3ADBAVIV%3E2.0.CO%3B2
Unfortunately, sondes sometimes had manufacturing problems, deployment problems or other problems.
John F. Pittman
Oddly enough, if we examine the arguments put forward, one person telling me I don’t understand that chapter thinks the problem is that that section doesn’t discuss fingerprints at all. The other thinks it’s because I failed to notice the chapter title matches the IPCC definition of fingerprints and insists the discussion is about fingerprints! (And he seems to think I didn’t discuss the section!)
“The patten claimed here by the IPCC’s reasoning is the same claimed by Lucia.”
John, you’ve made a mash of it. Nowhere does the IPCC say that a TT hotspot is indicative of GHG warming. Nowhere do they state that we should not see a TT hotspot with solar forcing.
The IPCC does not support Lucia. I will email the lead author after the holidays if necessary.
A Little Christmas Levity to Break Up The Monotony
I hope everyone takes it in the cheery spirit intended:
Little miss/mr., little miss/mr., little miss/mr. cant be wrong
Aint nobody gonna bow no more when you sound your gong.
Little miss/mr., little miss/mr., little miss/mr. cant be wrong
Whatcha gonna do to get into another one of these rock n roll
Songs?
Andrew ♫
Boris–
Are you suggesting GHG warming does not cause a TT hotspot at all?
Lucia claims that Douglass and others use “fingerprint” as defined by the IPCC. She (partially) quotes the IPCC definition of “fingerprint” (from the AR4 glossary):
So Lucia claims that Douglass et al were saying that the pattern through space and time was not observed. If we agree that Douglass was using the IPCC definiton of fingerprint then we can say that this is true– the entire response was not observed. However, if we assume that Douglass is using the IPCC definition of fingerprint, then we must conclude that his other statements are dishonest or ignorant (stay tuned–I’ll support this in a later post).
However, what about Steve McIntyre? And what about Lucia?
My original comment on climate audit was in response to McIntyre’s statement that
And in this very post, Lucia says:
They talk specifically about the TT hotspot. (In McIntyre’s case I assume he meant tropical tropospheric temperatures–but maybe he got it even more wrong.) Now Lucia wants to claim that they are using the IPCC definition of an entire response, but they are not. They are talking about part of the entire response. In fact, they are talking about a part of the response that is not characteristic or distinguishing of an enhanced greenhouse effect.
So even after word gaming and lawyering, Lucia is still wrong.
Finally, Lucia wants to claim that Gavin’s two runs of modelE for 2% increased solar and 2XCO2 are some kind of red herring to proper fingerprinting. But notice that the IPCC defines fingerprint as a run of a model to test the pattern against observations. This flatly proves that, by the IPCC definition, figure 9.1 is not a fingerprint analysis. Fingerprint analyses are compared to observations and not run with observations. However, Gavin’s brief experiment is a fingerprint analysis, and shows clearly that models predict a hotspot from both solar and GHG surface warming.
So claiming that the TT hotspot is a fingerprint (non IPCC def.) of CO2 warming is also clearly wrong.
No.
Boris–
Where does the IPCC says “entire” in the definition? A) They don’t. B) If they did, their usages would disagree with huge numbers of usages in the peer review literature. The fact that models are used to detect fingerprits does not imply that the fingerprints are “the entire pattern”. If use of modles inplied the word “entire”, then the actual presense of word “temporal and spatial” are both in there, you would need to use the entirel temporal and spatial pattern or response including the full 3D time response of the ocean, atmosphere, surface, ice sheets glaciars etc.” This is clearly not what they mean.
So, not only do you wish to include “entire”, but you mean “the bits, I, Boris find convenient to consider “entire”, excluding the bits I want to exclude, while still claiming entire”.
You can complain about lawyering all you want. But I don’t see why I can’t call you trying to decree someone used the word incorrectly by adding and subtracting words of your choice to the definition. The fact that lawyer a lawyer might not let you get away with this sort of redefinition doesn’t mean it’s ok for you to try to do it.
When you impose “unique” to the definition, it’s as if someone said: “strawberries are fruits” and you said “No. Because apples are fruits!”. When you include “entire”, it’s as though you said “No. Strawberries aren’t fruit. To use the word fruit, you must mean every single fruit that exists on earth!”
On the lawyer bit: Feel free to say “lawyer” all you want. It’s an odd sort of ad hominem some in climate circles have latched onto. Lawyers, scientists, engineers, mathematicians and many people use logic. To the extent that they do, arguments will sound similar. Why people in comments want to refute arguments with this bizarre contention that it sounds like a lawyer is just weird.
That’s not what I said. The isolated action of running model E is not the red herring. Since you retranslate what I said to something utterly different, it is clear you do not understand what I actually said.
Now…. happy christmas eve.
This has been a fascinating blogspot word fight. But blogs like this are read by people who talk to policymakers, and maybe even by policymakers. I know this because I talk to policymakers in the UK (we call them ministers). My colleagues do so on a regular basis because they have energy companies as clients, and they meet ministers on social bases as well ,because we’re Party members – so Party conferences, Party dinners and the like. What I want to confirm is the following:
1.Is a TT hotspot a natural consequence of GHG forcings and have we found one, (yet)? As far as I can tell from the postings, we haven’t yet.
2.What does this failure (or not) mean for the reliability of the climate models? (For example, If Lord Mandelson was asked to approve new build power stations how much attention should be paid to the climate modelstemperature forecasts and the impact on them of increased carbon emissions)? This is a real world question.
3. Lucia ran a comparison of temperatures actual against models forecasts and found the central forecast was 90-95% probability likely too high (should be 0.1C/decade not 0.2C/decade). Is Lucia right and should Governments be looking for a lower 21st Century temperature rise, and if not, why?
Thank you in advance.
PS I talk to treasury policy makers myself, although I have met Miliband at dinner and Mandelson in other roles.
tolkein,
As we have had various debates over the precise meaning of “TT hotspot”, I’ll say it this way: Models predict enhanced warming in the tropical troposphere at elevations indicated in figure 9.1C. Evidently, such warming, if it occurs, is not yet sufficiently large to detect. This may either be because the warming has not occurred or because the measurents are not sufficiently accurate.
We don’t know. We don’t know if the warming has occurred or not.
I’ve done more analyses, and I’m putting together a fuller paper with different time frames. That’s going to be more formal.
What governments do will depend on why the projections are wrong. In some discussions by email, the discrepancies may, in part, be due to a) a bug in one of the models b) a subset of models that are held in lesser esteem by some.
It may simply be that if the bugs and the “bad” models were eliminated, we’d get the same projection for temperatures in 2030, but have problems now. It may be that aerosol forcings are somehow not dealt with quite right. What would this mean? That would simply mean that, in future, modelers should be willing to screen models and not include information from those held in lower esteem.
I’d say generally: Policy makers should assume that collectively the ability of panels of climate scientist acting as a group and to predict the future is uncertain and so policy makers need to assume quite a bit of uncertainty.
I personally would recommend action to avert the worst consequences of warming. The reason is that even though I find models are on the high side now, we do have strong reasons to believe GHG’s cause warming. If models or modelers are missing something (or buggy or whatnot) they could be high now and low in the future.
The three words from Lucia’s post that say it all…
“reasons to believe”
Merry Christmas and Happy New Year to Everyone!
Andrew ♫
“the bits, I, Boris find convenient to consider “entireâ€, excluding the bits I want to exclude, while still claiming entireâ€
Huh? This is exactly what you are doing.
You’ve told me:
1. An identifying characteristic is not a fingerprint.
2. A modeled response through space and time is not a fingerprint.*
3. A tropical tropospheric hot spot is a fingerprint.
*Do you think they have a model of just the tropical troposphere? I don’t get your problem with “entire.” To clarify: the global response to the forcing in the model run is what I meant by “entire.”
By no definition is the TT hotspot a fingerprint.
Boris said ““The patten claimed here by the IPCC’s reasoning is the same claimed by Lucia.â€
John, you’ve made a mash of it. Nowhere does the IPCC say that a TT hotspot is indicative of GHG warming. Nowhere do they state that we should not see a TT hotspot with solar forcing.
The IPCC does not support Lucia. I will email the lead author after the holidays if necessary.”
Boris, my position is that the same methodology was used. In 9.x as quoted and observed in the graphs, the IPCC authors identify a smaller area and discuss the AGW signal. Using the same 9.1a, 9.1c, and 9.1f, that the authors used, one can see the TT “hotspot” as well as the NH that they discuss. They contrast the non-AGW and the AGW responses indicating that the models show “distinct” differences between the natural and anthropogenic forcing results. The IPCC authors claim this distinction.
In particular, they state that these “describe the pattern of NH temperature change near the surface that is dominated by the positive forcings.†These temperature changes are identified in the graphs, and can be visually determined, and compared to the text of the document. The “hotspot” for NH is identifiable, and the visual presentation matches the description in the text.
Boris–
Once again, you are modifying what I said. I did not say 1 or 2.
Of course they don’t have a model result of just the tropical troposphere. I made no such claim. They also don’t have a model result showing nothing more than a change in the height of the troposphere– which is called a fingerprint by Santer. That change in height is feature embedded in the more complete descriptions of spatial and temporal feature described by models. I would never tell Santer he can’t call the change in elevation in the troposphere a fingerprint because it’s a sub-feature.
lucia-
As a mechanical engineer specializing in fluid mechanics you have undoubtedly run into chaotic problems, such as turbulence, where it is extremely difficult to get a model to reproduce something for which you have very strong experimental evidence. I believe that it is generally accepted (no references-this is a statement of belief) that the atmosphere, oceans, landmass, etc. problem is chaotic,nonlinear, and highly complex. I also believe that all of the models artificially introduce some damping in one form or another to constrain the results so that they don’t diverge into unreproduceable (and unacceptable to the modelers) exponential positive or negative growth. I also believe that they always throw the negative results away. For policy makers, this should produce a very high degree of uncertainty. We don’t know what the worst consequences of warming are,or what methods will change GW. We sure don’t know whether AGW is significant or controllable. We do know that there are far fewer deaths due to increased heat than there are to the same change to colder conditions.We have no concept of the consequences of meddling with the climate, except that it will be very expensive and will thus lower the world standard of living. I believe that we should continue to study climate effectively (i.e. cut way back on modeling and fund those scientists who want to study real world data). Messing with complex systems is a good formula for disaster. If it ain’t broke, don’t try to fix it. Also consider the likelihood that any of the advocates of “fixing’ AGW will take any responsibility for the consequences. Remember impending global starvation ?
Feliz Navidad a todos.
Lucia,
I believe that’s an increase in tropopause height that’s the “fingerprint”.
My understanding is that he claimed it as a “fingerprint”, even though it also depends on stratospheric ozone and is not therefore “distinguishing”, because they were able to calculate the combination of both effects. Maybe the combination is distinguishing? So, my understanding is that the change in height can be caused by both heating from GHGs and ozone depletion (as Arthur Smith pointed out wa-y-y-y-y above), but they call it a “fingerprint” because they say they can account for the ozone depletion effect.
My opinion (and I freely admit it is just that), is that if the “hot spot” or “warming” or whatever you want to call it were to appear, we would suddenly hear that it is indeed a “fingerprint” because the solar forcing is not enough to account for it.
This whole fingerprint stuff is pretty amusing. In analytical spectroscopy, “fingerprints” are frequently referred to, not always consistently. What amazes me about climate science is how mean-spirited folks get over the “misuse” of vague terms.
I guess, if we follow Nathan’s lead, as long as each blog includes a glossary detailing the exact meaning of every term, we can call things anything we want.
JohnM– Yes. I think you’re right– increase not decrease. I should have checked.
With regard to Boris’s insistence that a fingerprint must be uniquely associated with a particular forcing, and also the entire response for the climate system, change in height of the troposphere violates both criteria. Yet, it is called a fingerprint by Santer. No one criticizes him for this.
Also, it is worth nothing that Santer’s rebuttal of Douglas did neither of the following: a) Claim the TT hotspot is not associated with GHG’s nor b) say it is not a fingerprint (which Douglas did say.) The paper describes the issue about the ratio of surface warming to warming in the enhanced warming region. The paper goes to great lengths to compare that to the observations.
But, for some reason, the claim that the TT hotspot is not a fingerprint for GHG has become a meme in comments at climate blogs. The claim tends to refer back to that RC article. I think this is becaue the structure of the article somehow caused a number of people to jump to the conclusion that there was something wrong with Douglas et al’s statement that the hotspot is a fingerprint of GHG.
This appear in “any and all” forcings issue also got introduced. This part of the meme refers back to the same RC post which results of hypothetical levels of forcing. So, it appears that in some people’s minds “any and all” automatically includes “provided we forget those cases where the forcing experiences the temporal variations in magnitudes of forcing and limit the discuss to the spatial structure in sensitivity studies”.
Boris, thanks for the heads up re Gavin’s models.
I have never said that the models don’t show a hotspot from any forcing. I’m looking for evidence, not the results of a computer model run. In a Playstationâ„¢ world, anything is possible. I’m asking about the real world.
w.
Nathan, you say:
Once again, I did not ask for general information on the subject. I have clarified this twice. I asked for what you and others here making the claim are depending on. This is different from saying “where can I find any information on hotspots”. The only way I can possibly know what information you, Nathan, are depending on is for you to point it out for me. Is my writing really that unclear?
Are you channelling the spirit of Phil Jones? He famously said, when Warwick Hughes asked for the station records underlying the HadCRUT dataset, “Why should I make the data available to you, when your aim is to try and find something wrong with it?”
Since neither you nor Phil seem to understand this curious thing called “science”, here’s the short version:
1. Someone has a brilliant idea about how the natural world works. They publish the idea, along with facts and evidence from the natural world supporting their idea.
2. Other people try to find something wrong with that idea or the facts and evidence used to support the idea. If they can find something wrong, they, in turn, publish what they think is wrong along with evidence to support their contrary view.
3. If nobody can find anything wrong with the idea or the evidence used to support it, it is provisionally accepted as a scientific “fact” (until somebody does find something wrong with it).
4. If someone can find something wrong with the idea or the evidence, then is is not accepted as scientific fact.
So yes, dear Nathan, if you do reveal what evidence you are using, I am sorry to say that I and other people will try to find errors in it … it’s called a “scientific discussion”, which is what you claim you wanted to find here.
Next, I had stated:
Rather than dealing with a substantive issue, such as the fact that much of the heat has turned into work, or that different kinds of warming mechanisms lead to different ratios of sensible to latent heat, you replied
I could tell you but you’d likely just criticize my answer … just kidding.
In fact, it’s not actually that hard to find that information. The best place to start would be your local university, ask some physicists.
w.
Lucia, thanks for coming up with a reasonably clear definition of what you’re talking about (#7582 – A “hot spot†is a region that is “hot†and has a “spot†like shape.) Upon reflection I think about half the essence of our disagreement can be boiled down to what people are usually referring to when they talk about the tropical mid-troposphere hot spot, or whatever it is they usually call it. What do RealClimate, Douglass, Santer, and so forth typically think is the definition of this “hot spot”?
Let’s phrase the definitions as precisely as possible, using relevant mathematical expressions. Say DT is the temperature anomaly, z is altitude (positive = up, surface = 0) and x is latitude (positive = north, 0 = tropics).
(A) (Arthur): dDT/dz > 0 for small z and small |x|, switching to dDT/dz < 0 for large z and small |x|
(B) (Lucia): there is a point near x = 0 for positive z where dDT/dz and dDT/dx are both zero, and the Hessian of second derivatives is negative-definite (i.e. this point is a local maximum in DT).
I’m satisfied with this form for my definition – are you agreed that this captures your definition?
Arthur– The short answer is no. At a minimum, by selecting zero as the thresholds, your definition omits the need for some threshold that makes things discernable. That is a point we have discussed. I haven’t thought further about trying to formulate a mathematical expression for the hot spot.
Unfortunately for this conversation, the sugar cookie dough that needed to be refrigerated for 4 hours is scheduled to be rolled into cookies at 2pm. Then, I transition to Swedish meatballs. Then, I will be going to seemingly non-stop X-mas parties, family enforced mass etc. for at a minimum 48 hours.
Well, I’ll look forward to a mathematical definition later then. Enjoy your Christmas, I’m sure I’ll be busy with family too for a while – take care!
Fair enough, WIllis. However, the whole point of skeptics has been: “models show a hotspot with GHG warming, so once we haven’t seen one, CO2 can’t have caused the warming.” In other words, what is at issue is what the models say. If you wonder why they say what they say, then that’s another argument entirely.
Okay, Lucia, now you bring in Santer. 250 comments in and you have entered yet another phase in your tortured definition of “fingerprint.” Is it any wonder that people can’t tell what you are trying to say?
But your definition of fingerprint is absurd. By your definition I could say “warming at the surface is a fingerprint of GHG warming.” and I could go around publishing this in op eds and on blogs and I would be perfectly honest and clear.
So try to explain to me why warming at the surface is not a fingerprint of GHG.
And merry Xmas by the way. I’ll be back on Friday. Maybe. I have the feeling Natasha will be hiding the laptop any second now 🙂
Boris,
Rather than accusing you of tortured language, I have to assume there is something I am obviously missing here:
Are you saying it is or is not reasonable to speak of surface warming as a fingerprint?
I’m glad (for Lucia’s sake) that she can get away for Christmas eve, day, etc. She has done a remarkable job rationally directing the flow of this discussion without loosing her cool. Quite a feat, considering some of the comments. In my view, she has won all the rounds, and therefore, the bout. She should be commended by all, even if you don’t agree with her. Have a fun time with your family and enjoy the holidays.
If some of the combatants still remain, will some of you tell me-What are some of the unique characteristics that pertain-only to AGW, so that we can distinguish between natural global warming (or cooling) NGW/C and AGW ? this seems like it should be a relatively simple question (with a relatively simple answer). Best to all.
Frank (#7647) – there’s nothing that’s completely “unique” about AGW in terms of the pattern of temperature changes at any given time – otherwise paleoclimate studies would be pointless, since they compare response of the climate to forcings that are definitely not anthropogenic. In fact, most forcings act in very similar ways, once scaled to the same forcing value (number of W/m^2 imbalance at the tropopause) – that’s why climate sensitivity can largely be split into a “forcing” and “response” component, where the response (surface temperature change per W/m^2 change at the tropopause) is pretty much the same for all the different forcings.
However, the primary contender in recent years for “skeptics” advocating an alternative source of warming is that the sun has been changing – recovering from the Maunder minimum, perhaps. And there are characteristics of the temperature response to LLGHG’s (and some other forcings) that are certainly quite distinct from the sun. In particular, GHG’s, almost fundamentally from the theory of the thing, have to lead to a cooling of the stratosphere while they warm the lower atmosphere. Solar forcing, similarly, almost fundamentally from the way it works, has to warm up the stratosphere about as much as the lower atmosphere. That sign difference is a strong discriminator between the two, so the observed cooling of the stratosphere is important in ruling out potential changes in solar forcing as the cause of surface warming.
The increase in the height of the tropopause is a similar effect expected from increasing GHG’s – tropopause height and GHG levels are closely linked also on a theoretical level, while there’s no such link for changes in solar input. So that would be another discriminator, although it’s less sharp than the stratospheric cooling one (since increased sunlight doesn’t decrease the height of the tropopause).
The most convincing reason to believe the human increase in GHG’s is responsible for the increase in surface temperatures, though, is not the spatial pattern, but the pattern in time: the 20th century increase is very close to what would be expected just from the GHG change alone, and including reasonable estimates of other forcings can be very well matched, temporally. And even that has only become convincing enough in the last decade or so, as the series of IPCC reports has shown increasing confidence in the attribution.
Differential temperature changes at a smaller scale, such as the “hot spot” we’ve been discussing here, even if they were useful in discriminating between different forcings (this particular small scale effect is not) since they involve even smaller temperature differences and finer scales and more complex averaging processes, are going to be much harder to find and attribute. As the discussion in the literature (Douglass, Santer, IPCC section 9.4.4.4) has shown to this point.
Twas the night before Christmas…
And what started out as a lop-sided boxing match turned into more of a wrestling match. Lucia the Hulk vs. the AGW Blind Man Tag Team may have been the most appropriate title. The parties to the brutal battle are currently enjoying a holiday-induced cease-fire.
Arthur returned from a day of gifting cookies. Proof of his alms came in the form of sugarcoated arguments and a tone indicative of a chocolate chip on his shoulder.
He continued to target the optically challenged, but his efforts to rely on the text rather than the colors on the graph were not compelling.
John V tagged in and pressed the issue of the common definition of words, rather than the appendix definition from the document being discussed.
Boris was the real heavyweight of the Blind Man Trio. Aside from the usually polarized language and tone, his was closest to the most effective of the AGW arguments; that the lack of a hot spot did not necessarily disqualify surface warming as a result of anthropogenic greenhouse gases.
The blind man tag team did however score a tactical victory; they stayed far enough away from the Hulkess to avoid being pinned by the realization, “if the worlds leading computer climate models cannot currently model the Earth’s atmosphere on such a basic issue, then how can we trust them on the future apocalyptic predictions that they are famous for?†That precedes questions about… ahem… the Sahara Jungle.
Despite expectations, the only hotspot detected in Lucia’s world harbors the soon to be devoured Holliday treats. Nevertheless, one question that still remains is who howls louder, the lonely hubby affront the television or the snowmen in the oven?
Willis Claus took a break from the lovely young skirt clad natives seated upon his lap and whispering expensive sweet nothings in Santa’s ear.
Even AGW heavyweight Eric momentarily stopped in before bailing out over Scott Base for another bout of spring melts equal catastrophic warming.
Lucia says:
I suggest you carefully read this paper: http://www.sciencemag.org/cgi/content/abstract/sci;309/5740/1551 It contains the basic answers to your questions:
(1) Yes, the amplification is expected for fluctuations in temperature…Not only is it expected but it is in fact observed. This is one of the interesting features that suggests the data are likely the problem: For the monthly to yearly timescales over which the data is reliable (should not have enough drift to affect the results), the “hotspot” is seen. It is only when one looks at the long-term trends over the multidecadal timescales, where the data is unreliable (any small drifts over time due to changes in radiosonde shielding or changes from one satellite to another will contaminate the results severely) that there has been a problem with observing the “hotspot”.
(2) The fact that this amplification is clearly seen for the fluctuations on monthly to yearly timescales severely limits the sort of theoretical explanations one can come up with to explain why it could be missing for the long term trends. In fact, the sort of hypotheses that Spencer and Christy are proposing all seem to involve effects that occur on timescales of days or weeks. So, I don’t see how such effects could nix the “hotspot” on the multidecadal timescales while not doing so on the monthly to yearly timescales! In fact, I have yet to hear of any reasonable hypothesis that seems like a likely candidate to explain this.
Boris–
But the temporal pattern of surface warming has been called a fingerprint. I mentioned that earlier in the thread. Here’s a longer quote
You’ll find more at this article compiled by Compiled by Jay Gulledge, PhD
Senior Research Fellow on Science and Impacts
Pew Center on Global Climate Change
Boris, thank you for your response. You say:
You are correct about what you are referring to. However, I was responding to another part of the discussion. The AGW supporters have said “We know the world has been warming. If there is no hotspot, the data must be wrong, because any kind of warming will cause a hotspot” In fact, this is in the title and the (original) theme of this thread.
When I heard that, it seemed dubious to me. For example, heating from GHGs goes on both day and night, while solar heating only occurs during the day. Since the atmosphere is (basically) stable at night and overturning during the day, it didn’t make sense to me that the two types of warming would have the same effect on the atmosphere. And it still doesn’t make sense.
Unfortunately, when I went to look for evidence of the claim that all warmings lead to the same hot spot, what I found over and over was some variation of the claim of Arthur Smith, viz:
The tropical mid-troposphere “hot spot†that Monckton highlights is not a “ï¬ngerprint†of greenhouse gases: it is well known to be a consequence of higher water vapor levels in a warmer world, whatever the cause of the warming.
What I have not been able to find is any evidence (or even a coherent theoretical exposition, as opposed to hand-waving) that this “well-known” claim is true.
You are correct that the models show similar responses to various kinds of warmings. However, anyone who considers model results to be evidence about the real world has never made a computer model of any complexity. A computer model can only reflect the beliefs of the programmers. Here’s a very relevant quote from today’s ScienceDaily:
In other words, despite the modeled world doing one thing, the real world doesn’t care, it does what it does regardless. All that the models are showing is that their programmers didn’t understand the atmospheric reality, and their assumptions (enshrined in the models) were incorrect. The models all agreed about the question … but that was not evidence of any kind. In fact, the evidence shows that the models are wrong.
Please note that agreement between an entire ensemble of models does not mean they are right. The majority of the models agreed about the troposphere/stratosphere exchange mentioned in the quote above, but like I said … the real world doesn’t give a sh*t if the models agree or not. All that model agreement means is that the programmers all made similar assumptions … and in this case, wrong assumptions.
So let me ask once again … do you believe that any kind of warming, no matter what the source, will lead to a TT hot-spot, and if so, on what evidence are you basing that belief?
My best to you,
w.
Thanks Joel– That’s interesting. I’ll read it after the whole Christmas rush.
Willis Eschenbach says:
First of all, since the question at hand is whether or not the “hotspot” is a distinguishing prediction of GHGs being the cause of the current warming, then I think the evidence from the computer models is very relevant. After all, it is these models that we use to draw out what we expect from our theories…and if the models show such a hotspot for all the forcings then not observing the hotspot would say something about the models but nothing about what is causing the warming. On the other hand, if the models show the hotspot is a unique signature for GHGs then if it is not observed that would tend to provide evidence against GHGs being the cause of the current warming.
Second, I am surprised that you feel that there has been no real-world evidence provided for the existence of this amplification for other warming mechanisms since we have discussed all this stuff before and I have referred you ad nauseum to the Santer et al (2005) paper that I gave in my post to lucia immediately above. In that paper, they show that, in fact, such amplification is seen in the temperature fluctuations that occur on monthly- to yearly-timescales as you go up in the tropical atmosphere. (It is seen in the satellite data analyses and in the radiosonde data analyses.) Surely, you don’t think that these fluctuations are due to GHGs, do you? (If you did, you would be way ahead of us AGW-ers in attributing all climate observations to increasing GHGs!!!)
Arthur Smith, you say:
To test this theory of yours, let’s take say black carbon as one forcing, and solar as the other. BC acts locally, solar acts globally. BC acts much more strongly in the Arctic (think of the effect of soot on snow) and weakly in the tropics, solar acts much more weakly in the Arctic and more strongly in the tropics. The sun shines on the whole world pretty equally, BC is concentrated in the Northern Hemisphere.
I could go on, but you see the point. There is absolutely no theoretical reason to assume that forcings act in similar ways.
Yes, some given forcing may act in a similar way to another given forcing. The effects of CO2 and methane will likely be similar … but to jump from that to “most forcings act in very similar ways” is a leap with neither theoretical nor evidentiary underpinnings.
w.
DeWitt Payne says:
Actually, such discussions like this of what could possibly be wrong in the models assuming the lack of a “hotspot” for the multidecadal trends is real (which I sort of doubt, but anyway…) are, I think, a more useful avenue for discussion. And, you might be right that, by making very different assumptions about convection of water vapor, perhaps you could get rid of the hotspot in the models. (It is not clear to me if this is the case, but it seems at least plausible given the understood physical origin of the hotspot.) However, your suggestion that perhaps it is a sign that water vapor isn’t getting into the upper atmosphere as predicted runs into a few problems with data that seems to contradict this:
(1) Soden et al. ( http://www.sciencemag.org/cgi/content/abstract/sci;310/5749/841 ) have indeed shown that there is a distinct radiative signature in the satellite data that strongly suggests that the upper troposphere is in fact moistening with increasing temperature as expected…and this is true both of the long terms temperature and moistening trends as well as the shorter term fluctuations. [There is other work by Dessler and some others that also argues the relative humidity is behaving about as expected, although I am not sure how high into the troposphere their results go to.]
(2) As I noted above, Santer et al (2005) have shown that fluctuations on monthly- to yearly-timescales do show the expected amplification as you go up in the tropical troposphere. So, any mechanism to explain away the hotspot for the multidecadal trends has to do so without explaining it away for those fluctuations. As near as I can figure, this is no easy task!
Mike C comments:
No. Arthur is targeting those who are able to look at a contour plot and understand what it is actually telling you (and not telling you) rather than those who just see pretty colors. In particular, one has to understand that if you don’t see something clearly on a contour plot that may be not because it isn’ there but rather because the contours have not been drawn with sufficient resolution to allow you to see it.
What the contour plot for solar forcing in the IPCC report chapter 9 shows is that there is an amplification of the warming as you go up in the tropical troposphere relative to the surface. However, because of the contour spacing relative to the strength of the warming due to solar, all that we can say is that the solar forcing at the surface is somewhere between 0 and 0.2 and that farther up in the troposphere is 0.2 to 0.4. Strictly speaking, that is compatible with an amplification factor of anywhere from a little more than 1 to infinity…and, in particular, is compatible with the amplification factor of about 2 or so that is seen in the plot for the GHG forcings.
However, admittedly this compatibility is weak in the sense that estimating the amplification as being between 1 and infinity doesn’t narrow it down very much. This is why Gavin ran the NASA GISS model with a larger solar forcing so that the pattern of the warming becomes more apparent.
Joel, good to hear from you. You say:
Joel, you are making Boris’s mistake about the question at hand, and thus creating a straw man to attack. The question is not whether the hotspot is a distinguishing prediction of AGW (a “fingerprint”). The question is also not whether the “models show the hotspot is a unique signature”. Look at the title of the thread. The question at hand is whether the tropical hotspot will form in the real world from any and all types of forcings, which is a related but very different question.
Also, you say “evidence from computer models” as if there were such a thing. See my post 7652 above. Computer models can provide insights and help our understanding. They cannot provide evidence.
Again, this is a straw man. I did not say there was no real-world evidence for “the existence of this amplification for other forcings”. As I said above, CO2 and methane are likely to have similar responses.
I said that there was no evidence that this amplification would perforce arise from any and all types of forcings, as you and many others have claimed.
Finally, Joel, you know that I take very little for granted, I am very suspicious of what is “well-known”, and I try to do the shovel work myself. While there seems to be agreement that there is tropical amplification on the shorter term (monthly), I don’t find it to be very evident. I just downloaded the UAH T2LT tropical dataset from
http://vortex.nsstc.uah.edu/data/msu/t2lt/tltglhmam_5.2
and the HadCRUT3 tropical dataset from
http://climexp.knmi.nl/data/icrutem3_hadsst2_0-360E_23.456–23.456N_n.dat
The overall correlation between the two datasets is 0.82, pretty strong as we’d expect … but the correlation between the monthly changes (∆T), on the other hand, is only 0.25. This is a very weak relationship, with r^2 only being 0.06. The correlation is three times greater in the long term than the short … what’s that about?
Next, out of the 353 month-over-month changes (∆T), in some 149 of them (42%) the surface anomaly cooled while the tropospheric anomaly warmed or vice versa … hardly what you’d expect if you are looking for “amplification”.
I encourage anyone who is truly interested in investigating this subject rather than speculating about it to download the data and run the analysis yourself. Then explain to me how, when in 40% of the months the tropical surface anomaly cools while the tropospheric anomaly warms or vice versa, we can come to any firm conclusions about any short-term amplification at all.
It seems to me that before we run around looking for mechanisms that can make the long term changes in surface and atmosphere make sense, first we have to understand why and how in the short term the tropical surface anomaly can warm and the tropospheric anomaly cool or vice versa. Once we understand that, we may be in a position to answer the longer term question.
My best holiday wishes to you, Joel, and to everyone
w.
Joel Shore Says:
“No. Arthur is targeting those who are able to look at a contour plot and understand what it is actually telling you (and not telling you) rather than those who just see pretty colors.â€
But Joel! It’s Christmas, a time for pretty colors!
Besides, what is there not to understand? For instance, look below the graph and you are given the magic key to understanding all of the dimensions of the graph. First, is the time frame. Then there is a bar of more pretty colors that tell you what the pretty colors on the graph represent. The it tells you that the little numbers on the graph represent the Lattitudes and the pressure or height. Not too difficuklt to figure in my book, but then… this is climate science.
Of course, I could have commented on Arthur’s odd rounding of the numbers when he tried to muddy the amount of warming with the dates, but that doesn’t involve pretty Christmas colors.
Then Joel says:
“This is why Gavin ran the NASA GISS model with a larger solar forcing so that the pattern of the warming becomes more apparent.â€
You gotta hand it to ol’ Gavin. The man was thinking ahead… all the way to Christmas of the next year, with his using the pretty colors on his graphs to represent what the difference would be for GHG and solar warming. But that was not Gavin’s real accomplishment. What Gavin did was to model the effects of solar on the atmosphere within months of the worlds leading climate scientists at the IPCC proclaiming that the level of scientific understanding of such was “Low.â€
Now, not to be Scrooge… we still have to give Boris a great deal of credit… not because of anything he has argued here… but because he always takes that lump of coal that Santa stuffs in his stocking every year (for you know why) and puts it in his fish tank rather than cooking some serious grillables, then eats dry ham and lumpy gravy in a sacrificial gesture to support his beliefs.
The moral of the story? Don’t fret too much about the silly comments by a commentator who is home alone slamming corrupted eggnog shooters after Christmas was cancelled because the family he was to celebrate with was quarrantined by the CDC after a member came down with Chicken Pox for the second time after being vaccinated. I’d go cry on Willis Clause’s shoulder but the native girls might not think too highly of it.
It’s really sad to me to see people with math skills and interest like Lucia and SteveM playing word games so insistently. It makes me wonder how much there is in our skeptic campaign. Then, when I see how they never write papers (Steve has thousands of posts, but has not produced a new paper in 4 years. Lucia has donut hole on papers…just interminable self-referencing posts with reading of comments even required to get all the info.) it makes me even sadder.
The only cheering thing is that the AGWers have a lot of mediocrity and blind spots on their side. Not the quality of the skeptics.
I think it’s important for people like SteveM and Lucia to look at things and pressure test them. But hoi polloi need to really be wary of what they say. They have a tendancy to be unfair to the people they are examining. Zorita and Berger are more guys that I respect, who are actually looking for insights and I feel would publish something whichever way it cut. Rather than just advocating a position.
Willis,
Since I can’t cry on your shoulder then maybe I’ll engage you for a moment. I recall there were several data sets available for this discussion. You chose one from UAH and compare it to one from Hadley. But UAH extrapolates the poles and Hadley does not. Is there any way to campare UAH and GISS since they do extrapolate and then compare RSS with Hadley since they do not? Also, if you have any access to sonde data, maybe a little of that too?
errrrr… assuming that would not be too much of a task since so much of this is lattitude and pressure dependent… would the satellite datasets even offer that data? I’m sure the sondes would
Mike C, “The Blackboard” is nothing if not a full service site. You say:
Whether they extrapolate for the poles is not meaningful for this discussion, since we’re just looking at the tropics and using nothing but tropical data.
I just went and got the GISS data from the same source (KNMI) and repeated the analysis using UAH and GISSTEMP. The results were:
Overall correlation: 0.85 (previously 0.82 unsing UAH/HadCRUT)
∆T correlation: 0.25 (previously 0.25)
Number of datapoints going in opposite directions (one up, one down) 142 (40%) (previously 149, or 42%)
In other words, using the GISS dataset instead of HadCRUT makes no significant difference to the results.
All the best,
w.
Willis… I’m a little weirded out by all of this… what you are saying is basically the same thing that Santer et al said in the study Joel linked (except on a slightly different time scale):
“On multidecadal time scales, tropospheric amplification of surface warming is a robust feature of model simulations, but it occurs in only one observational data set. Other observations show weak, or even negative, amplification.â€
Amazing! (given the list of authors):
B. D. Santer,1* T. M. L. Wigley,2 C. Mears,3 F. J. Wentz,3 S. A. Klein,1 D. J. Seidel,4 K. E. Taylor,1 P. W. Thorne,5 M. F. Wehner,6 P. J. Gleckler,1 J. S. Boyle,1 W. D. Collins,2 K. W. Dixon,7 C. Doutriaux,1 M. Free,4 Q. Fu,8 J. E. Hansen,9 G. S. Jones,5 R. Ruedy,9 T. R. Karl,10 J. R. Lanzante,7 G. A. Meehl,2 V. Ramaswamy,7 G. Russell,9 G. A. Schmidt9
1 Program for Climate Model Diagnosis and Intercomparison, Lawrence Livermore National Laboratory, Livermore, CA 94550, USA.
2 National Center for Atmospheric Research, Boulder, CO 80303, USA.
3 Remote Sensing Systems, Santa Rosa, CA 95401, USA.
4 National Oceanic and Atmospheric Administration (NOAA)/Air Resources Laboratory, Silver Spring, MD 20910, USA.
5 Hadley Centre for Climate Prediction and Research, UK Met Office, Exeter, EX1 3PB, UK.
6 Lawrence Berkeley National Laboratory, Berkeley, CA 94720, USA.
7 NOAA/Geophysical Fluid Dynamics Laboratory, Princeton, NJ 08542, USA.
8 Department of Atmospheric Sciences, University of Washington, Seattle, WA 98195, USA.
9 NASA/Goddard Institute for Space Studies, New York, NY 10025, USA.
10 NOAA/National Climatic Data Center, Asheville, NC 28801, USA.
And there goes that “robust” word again.
Mike C, it’s more complex than that. The Santer paper says:
What my results show, on the other hand, is that there is no consistent relationship on the monthly timescales. There is only a modest correlation (r^2 = 0.06) between monthly changes in surface and tropospheric temperatures, the scatterplot looks like a shotgun blast. More curiously, there are many months (~40%) when not only is there no amplification, but they are going different directions, with the surface warming and the troposphere cooling or vice versa …
If there were amplification, I’d expect that if the surface anomaly warmed by X degrees from say January to February (∆T), that the troposphere would warm by 1.2 times X or something like that. But I’m not finding that. In fact, linear regression shows that the monthly changes in the troposphere are only about half of the monthly changes at the surface. (This, of course, is not unrelated to the fact that often they have different signs.)
On an annual scale, on the other hand, I find a much greater correlation. In contrast to the month-over-month changes in the anomaly, the annual (year-over-year) change (∆T) of the troposphere and the surface have a correlation of 0.95 (monthly=0.25) , and in only two of the 28 years (7%, monthly = 40%) do they go in different directions. I also observe the amplification, with the annual MSU change being about 1.5 times the annual HadCRUT change.
I’m not saying that I understand why any of that should be so, I’m just reporting what the data says. It says there is amplification on yearly but not monthly scales. I’d be more than happy if someone checks my work (data sources listed above) and finds an error, because I don’t understand what’s going on.
w.
Joel-
We all understand what contour plots show us. We all understand how the choice of the temperature scale affects the detail we see.
If you read the conversation, you will see that the disagreement about what that graph tells us has nothing to do with misunderstanding or disagreement about what contour plots can and cannot tell us. It has to do with a) how we define when a hot spot exists and b) whether or not Gavin’s ability to create one when running the code using solar forcings that have not happened during the period of interest means these exist for “any and all” forcings.
I think when defining “any and all” forcings, it does not make sense to exclude the class of forcings that with temporal patterns that match the period of interest. So, the solar forcing pattern experienced during the observations record is “a” forcing. I think we define “hot” relative to the changes that are expected to arise as a result of everything that happens during that the period of interest (i.e. volcanoes, solar, ghg etc.) So, the solar forces don’t have any hot spots. They don’t even have any “hot” regions.
The choice of IPCC temperature scale is appropriate to discuss the relative contributions, the patterns that are actually expected to arise during the period of interest etc. That choice doesn’t “hide” anything. It reveals precisely the information that is of interest to understanding what patterns we might “see” in the field. “Spot” is a word that engages our sense of vision. To my mind an nearly invisible “spot” doesn’t make sense in the context of discussing what has happened.
Also, if simply being able to run model E with hypothetical forces tells us the spot happens for any and all, then presumbably, one could run model E adding electrical heating coils at the poles ramped up to ridiculous levels at the poles. We’d get surface heating without a TT hot spot. It’s a ridiculous “any and all”. But, this hypothetical but it could be done. (BTW: If you want a definition where extremely small effects count in the “any and all”, forcings, model E results seem to show contrails cause surface heating without a TT hot spot. So I don’t even need to get Gavin to run an idiotic electric heating coil realization. )
Lucia says:
Well, strictly speaking, I agree that he hasn’t rigorously demonstrated it for ALL possible forcings, but he has done it for two quite different forcings and he has presented the basic reason why such a structure occurs in the model…And, given that he works with these models, and that Santer et al have also shown the similarity of the structure in the tropical atmosphere to the predictions of moist adiabatic lapse rate theory, I think it is safe to at least tentatively assume that this indeed is what the models predict (and in particular, to note that the presence or absence of the hotspot is not a very useful diagnostic in trying to attribute the warming we have seen to a particular cause).
Will it in fact hold true in the real world? Well, obviously that is something that needs to be tested.
But, this is a bizarre definition! If you define things this way then it is basically just tautology that, if one forcing is expected to have been dominant over the others, then the structure that you see will be due to it and the other ones won’t make a significant contribution to that structure. You have created a definition that seems to be completely useless for drawing any sort of conclusions concerning attribution of warming to a specific mechanism.
At any rate, you can have whatever definition you want of what a hotspot is, just as long as you don’t do what many people do and make a claim that the (supposed) lack of the hotspot is evidence that the warming seen must be due primarily to some other cause than GHGs.
Joel–
It’s not a tautalogy. The question of “hot” depends on context. When discussing comparisons of models to observations (as for Douglas or as Steve M were doing), the issue of what causes the hot spot depends on the specific forcing set used. So, in that context we can talk about whether solar is expected to have actually caused any spot we could call “hot”– it isn’t.
With this same usage, we can say solar could not have caused the rise in temperatures since the 70s. It’s not that we couldn’t run models using fictional solar forcings alone and get that. We could– just as we can run models with hypothetical solar forcings and create a “hot” spot. When discussing the temporal pattern in temperature, people say solar “could not” have caused it all the time. Why can’t this usage apply to discussing hotspots?
MInd you, someone wants to change the subject to hypothetical forcing in isolation over entirely different time frames– the topic Gavin interjected in a discussion of a paper discussing a comparisons to observations– we can call that a hot spot for solar. But we would then need to say that the rise in temperature since the 70s could be caused by solar forcing (albeit, hypothetical forcing.)
Words and terms do get used in context.
I don’t see the current inability to detect the hotspot proof that warming is caused by something else. However, the idea that it might or might not be caused by something else is an uterly irrelevant observation. The reason failure to detect the hotspot is not proof warming is caused by something else, is that that failure seems likely due to poor data quality.
If the hotspot weren’t found, and the data were good, it would create big questions. After all, the more certain we are that it should appear based on fundamental physics unaffected by model parameterizaitons, the more surprised we should be if it doesn’t appear.
But currently, poor data seems a entirely plausible to me.
I hate to be the monkey in the wrench on Christmas… but if we can’t be so sure about the tropospheric hot spot then how can we be so sure about the stratospheric cooling?
Well, this story is certainly not getting any simpler.
To investigate whether the amplification is “robust” at various timescales, I calculated the amplification at all scales between one month and 340 months. Here is the result:
Figure 1. Amplification at timescales from 1 to 340 months. Amplification is calculated as the tropospheric trend over the time period divided by the surface trend over the same period. Trends calculated by linear regression, 95% CI = 1.96 * std. error of trend. Data from UAH MSU and Had
Several oddities about this:
1) Despite the claim by Santer that amplification exists at monthly and annual scales, in fact there is no amplification at timescales of three months or less, and at one month the “amplification” is only 0.5, much less than 1.
2) There is a peak of the amplification at around 24 years … why should that be?
3) At the longest timescales, the amplification disappears again.
As I said before, I’m happy to have anyone prove me in error. But if my figures are correct, the amplification is not robust at various timescales. Instead, it does not appear at either the shortest or the longest timescales available from our data.
Why? I haven’t a clue, I’m just reporting results I don’t claim to understand. But if nobody can prove me wrong, we have important questions to answer before we can possibly deal with the issue of whether a hotspot arises from any and all warming.
Regards and holiday wishes to everyone,
w.
To add to lucia’s points, I can’t see how Hansen and GISS can get the same tropical troposphere hotspots from solar forcing and GHGs.
From GISS’s modele 1880-2003 simulations page at:
http://data.giss.nasa.gov/modelE/transient/climsim.html
I see temperature change from solar irradiance to be this chart (for MSU channel 2 height) at approx 0.12C over the 123 years.
http://data.giss.nasa.gov/work/modelEt/time_series/work/tmp.5_E3SOaeoM20_12_1880_2003_1951_1980-L3AaeoM20A/global.gif
And this is the temperature change from GHG forcing at 1.1C over the 123 years (nearly 10 times more).
http://data.giss.nasa.gov/work/modelEt/time_series/work/tmp.5_E3WMGaeo20_12_1880_2003_1951_1980-L3AaeoM20A/global.gif
How could they produce the same hotspot?
(And they do at the GISS modelE simulations page but there can be no logical reason why.)
Solar hotspot:
http://data.giss.nasa.gov/work/modelEt/lat_height/work/tmp.1_E3SOaeoM20_1_0112_1880_2003_1951_1980_-L3AaeoM20A_lin/map.gif
GHG hotspot:
http://data.giss.nasa.gov/work/modelEt/lat_height/work/tmp.1_E3WMGaeo20_1_0112_1880_2003_1951_1980_-L3AaeoM20A_lin/map.gif
The more you play around with the model at this site, the more you realise there is just a lot of “forcing” of the code to get the results they want.
Willis, if relative temperatures dropped and stabilized relative to the time period 10 to 250, say starting around some future point, 350 for example, could it then have an amplification of, say, 0.5 or 1.0 tropos/surface trend? If it did, would this indicate an attractor state shift in a chaotic system? WIth highly variable pulses during the shift?
John Pittman, you say:
Um … er … well … John, not to put too fine a point on it, I’m farked if I know the answer to any of those questions. Like I said above, I’m reporting results that I don’t understand.
w.
I wonder if one of the frequency analyses would not be appropriate. There seems to be a small about 4.5 year and a larger about nine year cycle. I don’t know if there are enough data points or that it is actually a signal. There were a couple at CA who did some frequency analysis. I believe Leif was one. Could be wrong though. I could only try to follow the results of the analysis. The transition combined with the possible frequency reminded me of a signal pulse with a transition of state.
Willis,
Happy holidays to you too! Your graph of the amplification at various timescales is quite interesting…and I agree that I am not quite sure what to make of it. Of course, there are almost two orders of magnitude in timescales over which the amplification is robust but I agree that the breakdown at short timescales seems somewhat mysterious since these timescales still seem like they should be considerably longer than the timescales over which the convective processes operate.
(As far as I remember, Santer et al. had a different way of measuring the amplification factor…I think that they might have done it simply by measuring the standard deviation of the temperature, once for data computed monthly and again for data computed annually in order to investigate the different timescales. Your method, however, does seem better at picking out the different timescales. An even better method might be to look at the power spectrum…i.e., take the fourier transform.)
Lucia says:
Well, personally, I have always found the most convincing arguments against solar to be studies that showed that it doesn’t give the right temporal or spatial fingerprint no matter how you scale up the solar forcing. I.e., if solar would work if you just scaled it up, then I think there would be much more room for an argument in favor of something (like cosmic rays) that amplifies the solar forcing.
Well, I have seen lots of people who unfortunately make this argument. In fact, it seems to be second only to “the globe is cooling now” as the most popular argument that the data contradicts AGW. It is true that some of them will then retreat to a weaker argument (like that it just tells us that the models are not very good and thus can’t be trusted) if you confront them with overwhelming evidence that the amplification pattern in the tropical atmosphere is not expected to be any sort of unique signature for GHGs. But others continue to hold out and argue that it is the shining piece of evidence that GHGs cannot be responsible for most of the warming that we have seen.
I tend to agree…but a lot of “skeptics” will say that if you say this you are just ignoring data that is inconvenient because it points to the warming being due to some other cause. In fact, they will say that the only reason that you still accept AGW is that when confronted with evidence directly contradicting it, you say that the data is wrong and the models are right. So, I think it is important to point out additionally that the “hotspot” (in the sense of an amplification of the warming that occurs at the surface as you go up in the tropical atmosphere) is a very general expectation for warming caused by just about anything and, while its lack of existence would point to a significant hole in our understanding, it would not specifically say anything one way or the other concerning the cause of the warming that we have seen.
I’ve done the frequency analysis on the global temperature series and there is a really big repeating cycle at 25 years (300 months).
It is probably related to the solar cycle since Michael Mann undertook the same analysis before he got into tree rings and he found the same thing with other cycles at 9-11 years and 4.3-4.8 years. The numbers are expected to float around a little since the solar cycles are not regular.
http://holocene.meteo.psu.edu/shared/articles/MannLees1996.pdf
Mike C says:
Well a merry Christmas to you too! Seriously though, I think the big difference is that the stratospheric cooling is much larger in magnitude and thus much more robust to any corrections in the data. In fact, I think over at RealClimate, or somewhere, they said that the current corrections to the radiosonde data through the RAOBCORE re-analysis project do in fact raise the estimated temperatures all the way up through the stratosphere too but that this actually makes the scientists happy since it tends to improve the agreement with the models there too. I.e., if anything, the original data had the stratosphere cooling a little too much.
Bill could you do the analysis on Willis’s Amplification(tropos/surface trend)? I remember the cycling you found, and Mann’s paper. Though it would be intersting if it matched and somehow showed that the solar cycle was as Leif had it…very small temperature input. It would also be interesting as to why there is a change about 2002. Was it you that also found some changes about 2002?
John Pittman, I don’t think my skill set is as good as Willis. I do know I couldn’t figure out what to do with it before.
Mann seems to have given up after as well since he didn’t follow up that I am aware of (actually for this paper he tried to go back farther in time to see if the cycles continued in the past and to do it he needed to use proxy data and, hence, tree rings and the rest is made-up/PCA-weighting-gone-amock history).
Joel, you raise an interesting point when you say:
What I did was to look at all possible trends of length X in the data. This gives me many more data points. For example, if we look at annual data, there are 28 years available. But if we look at the 12-month trends over all possible 12-month periods, we have 342 data points. This gives a much tighter confidence interval.
A couple preliminary conclusions from all of this:
1) Our first task is not to understand the difference between the data and the models, or between the different forcings, but to understand the data itself.
2) Although Santer et al’s conclusion was that the radiosonde data was likely flawed because it did not show amplification on long term time scales, we find the exact same pattern in the satellite data.
Regards,
w.
Joel,
I do not share your observation of differing magnitudes.
It would involve the same arguement that Arthur was making but cool area instead of hot spot. There are different theoretical changes for different forcings, both warm and cool. I do not see how the same datasets with relatively simillar problems and magnitudes can have different uncertanties in different locations in the atmosphere.
Willis says:
I don’t think Santer disputed that…at least in the case of the UAH data. In their abstract, they state: “On multidecadal time scales, tropospheric amplification of surface warming is a robust feature of model simulations, but it occurs in only one observational data set.” The one data set that they are referring to is the RSS analysis of the satellite data.
In fact, they noted that the UAH data actually showed negative amplification (a slight cooling trend) in the tropics, which they noted that they found very suspicious. Shortly after their paper was published. the UAH group made a major correction to their data (correcting an error that the RSS group had found in it) and the trend became positive, although still notably smaller than the surface trend.
Well… I believe in AGW– even though some of my readers don’t. Also, I don’t disagree with most of what you say. (That said– I think there are problems with people saying things like the hotspot is not a fingerprint of GHG’s. It is a fingerprint of GHG. It’s not necessarily unique– but that’s not how the term is used.)
Still, if it isn’t found, there is something flawed in our understanding and it was to the extent that there is no hot spot, then our ability to project would be poor. The uncertainties on sensitivity would be greater.
Greater uncertainty doesn’t mean no warming. It also doesn’t neceesarily mean less or more– it means more uncertainty.
However, there must be at least some warming due to GHG’s. (Or if not, I can’t understand the physical mechanism to counteract this.)
BTW Pat Keating published an interesting analysis breaking the photons into three types of groups, and breaking the atmosphere into layers. He suggests phenomenologically based equations for each combination of layer/photon group, and gets a result that would suggest no hot spot. The proposed model is neither inconsistent with the types of arguments used to discusse the adiabatic lapse rate, nor with ghg’s causing warming. I haven’t had time to look at in detail, but the structure of the idea is interesting.
Mike C:
They don’t have to have different uncertainties. Let’s say the trend in the upper troposphere is +0.15 C +-0.15 C per decade and the trend in the stratosphere is -1.0 C +- 0.15 C per decade. These have the same absolute uncertainty…but the uncertainty in the upper troposphere means the actual trend may be anything from 0% of the central value to double the central value. By contrast, the trend in the stratosphere just varies by +-15% about its central value. (These numbers are just an example…not necessarily the actual numbers…although I tried to use numbers that I think are at least roughly realistic, as near as I can remember.)
Joel–
Also, it’s possible for design problems in instruments to cause errors that arise in some situations but not others. I don’t know what caused the measurements errors in sondes, but it’s at least hypothetically possible there were bigger problems in the troposphere.
Actually, here is a page that discusses the data for the cooling trend in the stratosphere: http://www.atmosphere.mpg.de/enid/20c.html and it looks like the number I chose in my example is high (in magnitude)…A better estimate of the data (post-1979 is what we are interested in) is a trend ~-0.6 C per decade (except for NOAA, which is an outlier at ~-1.1 C per decade), at least through 2000.
Nonetheless, my basic point still holds — corrections on the scale of, say +0.1 or +0.2 C per decade to the data will make a significant change to the question of amplification in the upper troposphere but will not change the trend in the stratosphere from cooling to warming.
Lucia (from Comment 7688)-
True enough. Although I think it would tend to make the people who are already suspicious of the corrections to the data even more suspicious if they were applied in a way that was highly spatially non-uniform in just such a manner that produced the hotspot in the upper troposphere while preserving the cooling in the stratosphere (even if there was a good justification for why the corrections should be this way)!
Fortunately, I don’t think we are in that situation.
Joel–
Yes. It’s always better all around if data are good in the first place. Collecting data, finding and discrepancy and correcting in the direction of a theory will always result in suspicion on the part of those who don’t believe the theory. And logically…. why shouldn’t they be somewhat suspicious? Or at least not as easily persuaded to change their minds in favor of the theory based on data that was corrected after it was found to disagree with the theory?
But I have to admit that when I read there were problems with sondes, my reaction was “Not a big surprise!”
Mostly– the sooner measurement are found and corrected, the better.
Joel, thanks for the heads-up. You say:
As is my style, I’ve now done the shovel work on the RSS dataset as well. Here’s the comparison of RSS and UAH:
As you can see, there’s very little difference between the two, and both of them show no amplification at the short and long ends of the scale. Like the UAH data, the RSS has an “amplification” of only 0.5 at the monthly scale.
Go figure … it is still very unclear to me exactly what is going on, but there appear to be different mechanisms working at the different scales.
w.
Joel,
The link you offered was the same one Boris used. It is unclear how much stratospheric cooling there is because they gave numbers for portions of the stratosphere (height wise) but what we are dealing with is also a problem that involves latitude as well (this does not include the problem of how the atmosphere and it’s components change in height by latitude). Your .6 figure also seems to have been estimated from a graph that they call “lower stratosphere.†WAKE UP MISTER!!! (LOL) The page you linked is also a little limited on time scale. The next problem that I noticed is that the 50 km cooling (where GHG’s are predicted to be responsible) is nearly split between similar cooling for both ozone and GHG’s with what I will call “counter warming†by solar and aerosols (REF the IPCC graph Lucia provided way, way above). So in the end, there was as much cooling predicted in that part of the stratosphere as was warming in the upper troposphere. So the .1 or .2 / decade for tropical troposphere vs. .6 / decade or whatever for upper stratosphere (50 km) does not add up.
But now that I’ve looked at it closer, I would bet the problem is the part of the prediction that involves aerosols… they have that one pooh-poohed real bad in all of the models.
Joel… kindly disregard this part since I was looking at pressure and not altitude… but the data between both sources was incomplete to get a handle on it either way.
The next problem that I noticed is that the 50 km cooling (where GHG’s are predicted to be responsible) is nearly split between similar cooling for both ozone and GHG’s with what I will call “counter warming†by solar and aerosols (REF the IPCC graph Lucia provided way, way above). So in the end, there was as much cooling predicted in that part of the stratosphere as was warming in the upper troposphere. So the .1 or .2 / decade for tropical troposphere vs. .6 / decade or whatever for upper stratosphere (50 km) does not add up.
There has been robust detection and attribution of anthropogenic influence on tropospheric warming…… in the fingerprint pattern of response.
From IPCC…
Table 9.4. A synthesis of climate change detection results: (a) surface and atmospheric temperature evidence and (b)
evidence from other variables. Note that our likelihood assessments are reduced compared to individual detection studies in
order to take into account remaining uncertainties (see Section 9.1.2), such as forcing and model uncertainty not directly
accounted for in the studies. The likelihood assessment is indicated in percentage terms, in parentheses where the term is
not from the standard IPCC likelihood levels.
Result – Tropospheric warming is detectable and attributable to
anthropogenic forcing
(latter half of the 20th century)
Region – Global
Likelihood = Likely
Factors contributing to likelihood assessment – There are observational uncertainties in radiosonde and satellite records.
Models generally predict a relative warming of the free troposphere compared to the surface in the tropics since 1979,
which is not seen in the radiosonde record (possibly due to uncertainties in the radiosonde record) but is seen in one
version of the satellite record, although not others (Section 9.4.4).
http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter9.pdf
Page 730
Merry Xmas
I didn’t say temporal pattern. That’s a different type of analysis than the one we are discussing.
So let me restate:
By your definition of fingerprint, I could point to the surface of the zonal mean plot in figure 9.1 and say “The surface warming in this plot is a fingerprint of AGW.”
You don’t find that absurd?
Willis it may be shovel work, but it reminds that when I whittle with a knife, I get wood shavings. Yours looks a bit like something real. I looked at the solar record. By eyesight, some points of interest with Leif’s active region count http://www.leif.org/research Most%20Recent%20IMF,%20SW,%20and%20Solar%20Data.pdf . However, it seems to line up with what appears to me as anti-phase points near the peaks of the solar cycles. But, I agree with Leif, the data is so noisy, eyeballs are always drawn to the pattern, but the patterns almost always fail the math test. Interesting graphs you have made. A nice Christmas present. Thanks
Frank Scammell:
I couldn’t agree more. Messing with a complex system is exactly what we’ve been doing for the past 200 years.
You were thrown off by Lucia here.
They produce the same hotspot because the forcings have close to the same magnitude. This illustrates that models show a TT hotspot in response to (at least) two different forcings.
No one would expect solar to cause a significant TT hotspot for the 20th century, because solar forcing was not very strong in the 20th century, especially compared with CO2 forcing.
MikeC
Which have you hear has aerosols “pooh-poohed”.
I’ve heard the same rumor from a pro-modeler person. I haven’t checked myself, but I’d like to know more. If the problem is confirmed, I’d like to know if it’s widely known.
Boris– No I don’t find that usage absurd. I don’t find usages like the change in elevation of the tropopause a “fingerprint” either. Lots of usages focus on sub-patterns embedded within the whole.
There are reasons some particular things specific sentences aren’t found– this is not because there would be a problem with the definition of fingerprint.
“And logically…. why shouldn’t they be somewhat suspicious? Or at least not as easily persuaded to change their minds in favor of the theory based on data that was corrected after it was found to disagree with the theory?”
Wait a minute.
Disagree with the theory? Since when are we allowed to do that?
I didn’t know such a thing was allowed. 😉
Andrew ♫
Boris-
Yes. This is precisely what I have been saying. How am I throwing anyone off by saying exactly what you are saying?
Solar cannot be a cause of a “hotspot” for the 20th century. Given the relative magnitude of the forcings, it can’t cause “hot” at all.
For similar reasons, solar is not the cause of the upswing in the temporal pattern– also called a fingerprint.
So, if climate scientists call the temporal pattern of surface warming a “fingerprint” of ghgs, when discussing patterns expected in the 20th century, clearly, we can call the hotspot a “fingerprints” of ghgs. Both could be caused by some other hypothetical forcing– if one bothered to exercise these forcings. In the context of comparing observations to models, people say solar forcing can’t cause that pattern. They mean can’t in the same way I mean solar can’t cause the TThotspot. It’s can’t in that context because solar forcing hasn’t varied that much.
For some reason, you want to emphasize the “can’t” for the 20th century when discussing the temporal pattern and then forbid it for the hotspot.
If you just want to make the point that it would be possible in some other context– fine. But jumping into conversations discussing 2oth century warming, and trying to prohibit a perfectly good usage of “can’t”, in context is silly.
I didn’t even talk about the temporal pattern of surface warming. You brought that up.
Look, since you acknowledge that I could point to the surface warming in the zonal mean and claim it is a fingerprint of greenhouse warming, then we will have to agree that we have vastly different definitions of “fingerprint.” However, yours does not conform to the IPCC’s definiton. nor does it conform to the dictionary definition. You can’t just make up a definition of your own.
In addition, your definition allows one to make statements that are flatly untrue and highly misleading if one considers the real definitions of the word. I haven’t read the Santer paper or the others you cite (I’d be willing to bet that you are wrong in your derivation of their use of the term); however, for the sake of brevity (Oh, sure!) I will stipulate that if those sources use such a vague definition of “fingerprint,” then they too are being misleading.
I refuse to use misleading terms just to win a point. Douglass and others apparently have no problem using a misleading definition, especially in non-scientific communications (where, of course, the dictionary definition of fingerprint is the expected one).
The fact remains that the TT hotspot is not uniquely associated with an enhanced greenhouse effect. Any claims that it is are untruths. Any documents or sources that would lead the average reader to believe that it is are misleading.
Speaking of aerosols, Hansen has a 0.3C reduction built in for aerosols for global temps.
http://data.giss.nasa.gov/work/modelEt/time_series/work/tmp.4_E3TAIaeo20_12_1880_2003_1951_1980-E3TADaeo20/global.gif
With a much bigger temp reduction for the tropical hotspot and especially the northern hemisphere and the Arctic (which are strangely both warming much faster than the southern hemisphere – the aerosols are a good thing since the Arctic ice would be much closer to melting completely if not for the -0.7C cooling effect provided by the aerosols in the Arctic.)
http://data.giss.nasa.gov/work/modelEt/lat_height/work/tmp.1_E3TAIaeo20_1_0112_1880_2003_1951_1980_-E3TADaeo20_lin/map.gif
Boris
Lucia, just in case you are cooking more cookies during the 10 minute wait:
Boris said “”The fact remains that the TT hotspot is not uniquely associated with an enhanced greenhouse effect. Any claims that it is are untruths. Any documents or sources that would lead the average reader to believe that it is are misleading.”” I don’t remember you claiming this “uniqueness”. In fact I thought 07:55 comment pointed out it was not unique, but that solar changes for 20th century could not be the cause.
Also, I thought your definition was the IPCC definition. I looked at how they used it. It agreed with you. I also posted text from the IPCC chapter 9 that showed they used it and discussed it as you claimed.
I do find it troubling that a comment should be made without reading the source per Boris 08:31.
On other parts perhaps not read: From 9.4.1.4 p2, on page 687 “”Fingerprint studies that use climate change signals estimated from an array of climate models indicate that detection of an anthropogenic contribution to the observed warming is a result that is robust to a wide range of model uncertainty, forcing uncertainties and analysis techniques”” … “”Multi-signal detection and attribution analyses do not rely on such agreement because they seek to explain the observed temperature changes in terms of the responses to individual forcings, using model-derived patterns of response and a noise-reducing metric (Appendix 9.A) but determining their amplitudes from observations. As discussed in Section 9.2.2.1, these approaches make use of differences in the temporal and spatial responses to forcings to separate their effect in observations.””
This conforms as well as I can tell to what you have stated. Or rather, I should say that what you have stated conforms to this, to be more correct.
Willis – There is something I find kind of strange about the results that you show. It just seems like too big a coincidence to me that the deviation at long timescales occurs so close to when we approach the end of the record. I.e., if we take your results as being completely real and uninfluenced by the length of the record, then they would seem to imply that the deviation from the amplification seen over a range of timescales has just occurred once the record approached 25 years long and wouldn’t have been detectable before that.
In particular, it would imply that back about 4 years ago when only ~25 years of the record existed, people looking at the trends over the entire time period available at that time would have concluded there was an even larger amplification of around two or so. And, that back around 2000, when they had just a little over 20 years of the record in existence, they would have not yet seen any significant deviation from the amplification seen over shorter timescales. I find this hard to believe…And, in fact, it would be easy to test simply by truncating the record in those years and re-doing your analysis. [You could do this by cutting off either the last N years of data or the first N years of data…Or some at the beginning and some at the end and seeing what happens in these different cases.] My guess is that you will find that the results near the end of your record are not very resilient to when you truncate. I am not sure if this is an indication of something erroneous in your analysis or an indication that it is only when one looks at trends that approach the entire record length that one gets a significant deviation in the amplification (which could be an indication of artifacts in the data that add a spurious long term trend, as has been suspected by many).
Boris-
My use conforms to the IPCC definition. The reason I bring up examples in peer reviewed literature is to show that they do conform to the IPCC definition, which is not as restrictive as you wish it was. There is no “unique” in the IPCC defintion. There is no “entire” in the IPCC defintion. Moreover, climate scientists do not restrict the term to patterns that are “unique” or “entire”. You want to restrict. One individual can’t do that to language.
In my previous discussion when I used “you” I should have said “climate scientists”. Unfortunately, in colloquial english around here, we sometimes way “you” when we mean “people in general”. You specifically did not forbid people from saying the temporal fingerprint of surface temperatures during the 20th century “can’t” have been caused by solar– but many people use “can’t” in this way. They mean “can’t” in context of what happened with solar forcing over the period. They use surface warming as a “fingerprint”. This is the way these words are used.
JohnM–
Actually, I don’t think climate scientists themselves get in an uproar about the usage. It’s climate blog devotees trying trying to rein in conversations discussing comparisons of model predictions of the TT Hotspot to the observations.
The TT Hotspot has not yet been observed. Not content with simply noting that this is due to poor data, some want to jump in and start insisting the TT Hotspot is “not a fingerprint”, discuss the adiabatic lapse rate and divert the conversation away from the discussion of the comparison. They want to bring up the irrelevant issue of whether or not the hotspot would occur with hypothetical amounts of solar without emphasizing they are changing the conversation from actual levels of solar to hypothetical etc. (They might admit it when pressed.)
Changing the conversation to distract from the issue at hand is called “a red herring”.
I think most of those introducing the red herring are doing so unintentionally. They may think discussing what patterns we would hypothetically see, and how we would attribute those pattern in entirely hypothetical conversations is important to the comparison of observations to predictions of is predicted to have happene on earth during the time period in question with known forcings.
The hypothetical issue of what would happen under forcing that have not happened on earth during the period being discussed has very little importance to the data comparison.
The reality is: According to models, if we do a data comparison for the 20th century or any portion of that time period, the solar component is nearly washed out by the GHG component. If we saw a TTHotspot, we wouldn’t attribute it to solar. For all practical purposes, solar is not thought to have caused any “HOT” spots during the 20th century.
If for some reason, a person wants clarification of how someone is using “can’t”, that’s fine. But no one gets to decree that words can not be used in context. When people like SteveM or those in conversation are clearly discussing comparisons of models to real earth observations for the period 1979-now, and then say “Y can’t cause X” it’s rather obvious the what they mean is “Y can’t cause X in the context of real earth forcings during the period of 1979-now”. When Douglas or anyone uses fingerprint in a sentences, he should certainly be permitted to use the IPCC dictionary definition (which he did).
Why people want to jump in and suggest that these usages are wrong is mystifying. These usages are fine.
Lucia,
I’m not sure how the Campos paper affected the modelers in general, but I imagine it opened some eyes given the prominence of the models they used… but before that, outright denial was the norm.
“My use conforms to the IPCC definition.”
It most certainly does not. You claim that it does, but it is clear that the IPCC is talking about runs of a GCM (when you don’t truncate the definition that is), which, as you must admit (for God’s sake don’t argue with me on this) are GLOBAL. I tire of the definition games that you play to try to turn the dneialists into real scientists.
Once again, you definition is so broad that is it ludicrous and serves no purpose whatsoever. If Real Climate were to post a piece claiming that the warming of the surface as shown in a zonal mean graph was a fingerprint of GHG warming you would post an apoplectic screed about how Gavin is a liar and the only real scientists post comments over at Climate Audit. Your double standards are evident even to TCO, a skeptic who sees this mess for what it is.
The fact remains that the TT hotspot is not uniquely associated with an enhanced greenhouse effect. Any claims that it is are untruths. Any documents or sources that would lead the average reader to believe that it is are misleading.
Boris, From IPCC. “There has been robust detection and attribution of anthropogenic influence on tropospheric warming…… in the fingerprint pattern of response. Tropospheric warming is detectable and attributable to anthropogenic forcing….Models generally predict a relative warming of the free troposphere compared to the surface in the tropics since 1979,…. (Section 9.4.4). http://www.ipcc.ch/pdf/assessm…..apter9.pdf
Page 730
The models generally predict Tropospheric Warming, as a fingerprint of anthropogenic influence. The reality is there is no Tropospheric Warming, and therefore the IPCC theory supporting that part of Anthropogenic Warming is WRONG.
MarkR:
Ah, Mark, you are aware that the troposphere includes the atmosphere right at the surface of the earth where the surface temperature record is measured aren’t you? There is indeed warming there. The satellite record also shows warming further up in the troposphere. What some of the analyses of the satellite record and radiosondes don’t show is an amplification of the surface warming as you go up through the tropical troposphere. But you might also note that the part you quoted from the IPCC report doesn’t discuss this.
Joel Shore. OK, FREE Troposphere it is. “Models generally predict a relative warming of the FREE troposphere compared to the surface in the tropics since 1979, which is NOT SEEN in the radiosonde record (possibly due to uncertainties in the radiosonde record) but is seen in one version of the satellite record, although not others” (Section 9.4.4). IPCC
As I nearly said, no actual free tropospheric warming equals no anthropogenic warming theory.
Can anyone explain why the Antarctic is bloched out in all of these simulations?
Mike–
The “Y” axis is pressure. When a region is mountainous, the surface is at a lower pressure already. For most parts of the world, this isn’t an issue. But if you go far enough south, nothing is at sea level and so nothing is at atmospheric pressure.
Lucia,
The Y axis is pressure on one side and elevation on the other… and since Antarctica is two land masses joined in the middle by a large chunk of ice, I’d say there is something missing on the graph. In the mean time, the models seem to predict warming there too, but it aint so unless you are Eric Steig.
Yes, the models generally predict this. Of that there is no doubt. However, it is not a prediction specific to the mechanism of the warming being from greenhouse gases. I.e., it is not one of the things that the IPCC notes is a “fingerprint” of the warming being due to GHGs. And, of course, as the rest of Section 9.4.4 discusses, there are serious problems with the observational data for the trends on the multidecadal timescales.
Joel, you raise a good issue. You say:
This was why I pointed out the data sources and invited people to do the analysis themselves, as I’m not really happy with the results either. Is there an error? I don’t know, I haven’t found one, but that means little.
I plan (when I get a few moments) to split the data in half and analyze the two halves to see what I find.
I was encouraged to see that the RSS data showed the same pattern as the UAH data, however, because that narrows down the possibilities. It also means that if we measured it for only a couple of months we’d get a different result than measuring it for a year.
In addition, I doubt if the lack of amplification (actually existence of “anti-amplification” that I don’t have a word for) in the short term is an artifact. It is the same result I got using another method. And if the 1 month response is different from the 1 year response, I see no theoretical reason why both might be different from what happens at the quarter century scale.
Again, thanks for the heads up.
w.
I have a post on this topic at my webpage as well. Unfortunately, anyone who believes tropical tropospheric amplification is unique to GHG’s is just wrong, and because this serves better for detection and not attribution, the only conclusion that could come out of some people’s logic is that there has been no warming at all. It is also possible much of our understanding of moist convection is wrong. This does not mean GHG theory is wrong so this meme should stop.
As for the IPCC models, don’t confuse what actually happened with what would happen given a more substantial influence from non-GHG factors.
MarkR,
Your link doesn’t work. Is that the TAR?
But I think “in the fingerprint pattern of response” supports my position–that the IPCC uses fingerprint to mean the entire pattern, and so referring to part of the pattern is incorrect usage.
The tropical tropospheric hot spot is a fingerprint of the model that produced it which has a whole set of inputs – GHG forcing, decrease in saturated adiabatic lapse rate as temperature increases, model result for region over which saturated adiabatic lapse rate is maintained etc. If the hot spot doesn’t actually appear it means that one of these inputs is not correct and not necessarily the one you prefer it to be. e.g. the model includes a presumption that the saturated adiabatic lapse rate decreases over a large range of tropical tropospheric altitudes. This assumption hasn’t been tested as far as I know. Until we know which one it is we certainly can’t say that the missing fingerprint is proof that there is no GHG forcing or that there is no response to GHG forcing.
Chris–
Absolutely correct.
Should the pattern never appear, we can’t say what a missing fingerprint would mean.
But, that’s different from decreeing the predicted pattern is not considered a “fingerprint”.
Anyway, for all we know, it may eventually be detected when better data are available.
Lucia, in responding to Chris O’Neill, I did not notice that you did not respond at all to his double negative nor his implied triple negative (until we know = we do NOT know presently). Perhaps or perhaps not, I did not miss something.
You have a warm and sugarly high from your Christmas cooking? You usually point out the inherrent problems with multiple negatives, rather than the letting the poster use multiple negatives, instead of using a straightforward statement.
Can’t blame you. I had a good Christmas, and I hope you did too.
Willis, perhaps the term that you require is “damplification”…
😉
lucia:
So someone considers it a fingerprint of AGW. Who?
Boris writes:
“But I think “in the fingerprint pattern of response†supports my position–that the IPCC uses fingerprint to mean the entire pattern, and so referring to part of the pattern is incorrect usage.”
Having earlier said:
“The hotspot is not characteristic of greenhouse warming. The cooling stratosphere is.”
But, of course, stratospheric cooling is itself only a part of the pattern, and not itself the entire pattern, that are illustrated in figure 9.1 of Chapter 9 of the AR4.
Bernard J., I love “damplification”.
Joel Shore, I stumbled across one reference which might be of interest in this discussion. It is:
Don’t know yet if this has been mentioned above, further commented on, explained, or overturned …
w.
Boris. http://www.ipcc.ch/pdf/assessment-report/ar4/wg1/ar4-wg1-chapter9.pdf
AR4 Page 730
“Fingerprint studies that use climate change signals estimated from an array of climate models indicate that detection of an anthropogenic contribution to the observed warming is a result that is robust” Page 697.
“Models and observations also both show warming in the lower part of the atmosphere (the troposphere) and cooling higher up in the stratosphere. This is another ‘fingerprint’ of change that reveals the effect of human influence on the climate.” Page 703
No actual free tropospheric warming means the models are wrong, and the theories behind them, are wrong.
I have been trying to wade through all this. I consider all the semantic statements obfuscation.
The world was presented with a policy makers report of the weighted calculations from several GCM s and the IPCC demanded action on the results of this study. The action was not required because of future studies or cogitations. It was based on 800 pages of documentation. One of these pages has figures of predictions for the tropical troposphere temperature behavior, clearly marked in fig.2 for each supposed “forcing”. There is a unique color code for all the plots shown in the figure. Only one shows the highest temperature deviation and that is the GHG plot ( used in preliminary versions to be labeled CO2 plot, currently one has to read the a, b, c etc words in the caption below to make sense of the plot).
Data do not support this excess of heat so, as far as the IPCC report goes, the models as presented in summary to the politicians are invalidated on that point, which means that new calculations are necessary.
The crux of the matter is that the world politicians are being stampeded/bamboozled according to this IPCC reports to take drastic actions on energy issues. They are not being told : wait, there are new calculations, because this would invalidate the “science is settled” mantra.
Now as far as I am concerned it is fine to recalculate these GCM so that the tropical troposphere results agree with the current data, but at the same time (concurrently, in parallel) they should show plots of the warming trend. I strongly suspect that once the models fit the tropical troposphere profiles, the warming trends will stop being catastrophic. Politicians should then be informed that they have ample time to think and plan in the next decadeS and not the next years of what to do about energy planning.
I agree anna. It does seem in the midst of the discussion that the issue of “high” confidence is missing in action. The other item that seems to be missing is that the IPCC claim that the fingerprint is warming versus cooling, as posted by MarkR “Models and observations also both show warming in the lower part of the atmosphere (the troposphere) and cooling higher up in the stratosphere. This is another ‘fingerprint’ of change that reveals the effect of human influence on the climate.†Page 703
Amother interesting aspect of page 703 is that it contradicts Boris’s statement at
“”December 26th, 2008 at 11:20 am
“My use conforms to the IPCC definition.â€
It most certainly does not. You claim that it does, but it is clear that the IPCC is talking about runs of a GCM (when you don’t truncate the definition that is), which, as you must admit (for God’s sake don’t argue with me on this) are GLOBAL. I tire of the definition games that you play to try to turn the dneialists into real scientists. “”
Apparently on page 703, when the IPCC authors said “”FAQ 9.2, Figure 1. Temperature changes relative to the corresponding average for 1901-1950 (°C) from decade to decade from 1906 to 2005 over the Earth’s continents,
as well as the entire globe, global land area and the global ocean (lower graphs). The black line indicates observed temperature change, while the coloured bands show the combined range covered by 90% of recent model simulations. Red indicates simulations that include natural and human factors, while blue indicates simulations that include only natural factors. Dashed black lines indicate decades and continental regions for which there are substantially fewer observations. Detailed descriptions of this figure and the methodology used in its production are given in the Supplementary Material, Appendix 9.C”” , apparently breaking the comparisons to smaller units than the tropical area of the earth, and did not find a problem with it. In fact, this continental brakdown is used to support global warming as real; and is used to answer the following question Frequently Asked Question 9.2 Can the Warming of the 20th Century be Explained by Natural Variability?
JohnFPittman…. On rebuking people for double or triple negatives…
I have to admit with so many comments appearing on xmas eve, xmas and the day after, I’m skimming a lot of these! In anycase, I rarely bother to rebuke people on their writing style. If it’s confusing, and someone wants clarification, they can ask the writer. If it’s confusing and everyone skims past, the writer will likely learn to be more clear. If it turns out people understood it, and engaged the writer, then the writer is free to continue using any number of negatives!
Anna v,
That is a nice summary and a very insightful comment. The UN has been laying down these 10 year drop dead warnings since the 80s. Why on earth anyone still believes these prophets of doom is beyond me.
Yes, the effect of GHGs is much greater in magnitude than the effects of the other forcings as we’d expect from the much greater magnitude of GHG forcing. Greater forcing -> greater effect.
The solar forcing produces a hotspot, it’s just a lot weaker than the hotspot produced by GHG forcing simply because the solar forcing is a lot weaker than the GHG forcing.
If they understand what they’re reading, they’ll think the tropical tropospheric hot spot is a fingerprint of GHGs AND assumptions that go into WG1’s model and that it’s silly to think that the hotspot is a fingerprint of GHGs on their own and does not depend on assumptions made in the models.
The impression it gives me is that the hot spot comes mainly from the effect of GHG forcing combined with the assumptions built into the atmospheric models. Different assumptions built into the atmospheric models won’t necessarily cause a hot spot with GHG forcing.
Dover_beach: Yep. Everyone, including Boris, calls portions of the entire patterns fingerprints at times.
The whole pattern may also be called fingerprint. I’ve been saying both uses are correct. At times, it appears Boris has tried to rebutt this by showing that some people use the entire pattern to mean fingerprint and/or some specific fingerprint studies use the entire pattern.. Of course they do: Both uses are correct. We can find both. We can find studies where part of the pattern is used, and others where the whole pattern is used.
Some words in English work that way: they are defined broadly, not narrowly.
Anna v good summary. And I also see what you mean now Boris. I believe that when Boris is saying that the ‘fingerprint’ is not unique to GHGs in the models, he means the pattern is not produced purely by the maths in the models that deals with GHGs, but that if other forcings are changed then a ‘hotspot’ appears. So its not a characteristic of GHGs but more of the model generic maths instead.
Using a set of real inputs, GHGs produce the hotspot and so this is the only data set we should be baselining as a reference. However in a more abstract sense the hotspot is not a unique characterstics of GHG relationships and forcings. I think that’s the gist.
However so what? Its just semantics and it is a red herring if used as a tautology. Only deal with the set of inputs that represent the real world as everybody else in science tends to do and change the models within that context, not with the sun becoming a K type star. That is what the IPCC did. They ran a model based an assumed forcing from CO2 and got those graphs.
I’m using the lay definition of fingerprint–which is the one that I have always used–meaning a uniquely identifying feature. (Of course the reason the IPCC calls these studies “fingerprint studies” is because they are looking for identifying features. Imagine that!) Lucia was the one who brought up the IPCC definition in her labyrinthine defense of McIntyre and Douglass. I accepted that definition and showed her to be wrong.
You see, the reason denialists want to broaden the definition of terms is to be able to mislead people whilst having an “out” to some vague definition they glean from some study or other (No matter if their interpretation of the paper is right–and it usually is quite wrong).
So when Douglass says the “characteristic fingerprint” of GHG warming has not been observed–and when he states this in a press release or interview–he is misleading the lay reader, who understands “characteristic fingerprint” to mean what it sounds like it means, something that would uniquely identify a cause for the warming. The reader goes away thinking “CO2’s fingerprint was not observed, so something else must have caused the warming.”
It doesn’t help that Douglass is either ignorant or lying when he says
The discrepancy between models and data proves no such thing. The TT hotspot is not linked in any way to the effects of GHGs, except through surface warming, which has undoubtedly occurred. So will someone here have the intellectual honesty to say that Douglass doesn’t know his stuff or that he is lying, or must we go round and round again? Wait, wait, don’t tell me…
“You see, the reason denialists want to broaden the definition of terms is to be able to mislead people whilst having an “out†to some vague definition they glean from some study or other (No matter if their interpretation of the paper is right–and it usually is quite wrong).”
Not at all. We denialists haven’t been shown any evidence that proves AGW, word games or no word games. In fact, the AGW hypothesis keeps changing, (every time a model is tweaked or any time the data is adjusted) so it’s not being tested in a valid way. The information surrounding it keeps changing as the sun rises everyday.
I hope I put that in understandable terms. 😉
Andrew
Boris–
The IPCC definition and the lay definition match. Fingerprints can be subportions of patterns under both defintions. It has always been used this way.
You are trying to change the definition for some reason unto yourself.
No matter what the word means, the pattern predicted by models has not yet been observed. I don’t know how saying something that would lead lay readers to understand that the predicted pattern is not observed would mislead them!
Neither have I. Proof is only important in mathematics and liquor.
You totally just blew my mind.
So since the TT hotspot is not an impression on a surface and is not distinctive or identifying, then it’s not a fingerprint. Well, case closed.
Speaking of fingerprints, if y’all haven’t read Twain’s Pudd’n’head Wilson, you should.
“Neither have I. Proof is only important in mathematics and liquor.”
Interesting. I take this as an admission that your position (a belief in AGW) is a religious one?
Andrew
Boris, a TT hotspot is an identifying feature of GHG-forcing. No TT-hotspot, no GHG-forcing. Simple as that. It is one of three other identifying features of GHG-forcing, the others being surface warming and stratospheric cooling. If any one of these three are missing it is very likely something other than GHG-forcing is occurring. And it is intellectually honest to question the magnitude of GHG-forcing if one of the three identifying features is missing.
None of this is at all difficult to grasp.
“The discrepancy between models and data proves no such thing. The TT hotspot is not linked in any way to the effects of GHGs, except through surface warming, which has undoubtedly occurred.”
Except that GHGs on their own only warm the atmosphere by approx. 1 degree C and require water feedbacks to get it to approx. 3 degrees C. You get GHG plus water feedback, you apparently get a TT-hotspot. There is thus a clear link between GHG-forcing plus water feedbacks and a TT-hotspot.
I’m happy to give up the TT-hotspot as an identifying feature of GHG-forcing if we all give up the 2 degrees C warming that purportedly arise from feedbacks.
Andrew says:
“Interesting. I take this as an admission that your position (a belief in AGW) is a religious one?”
Why not ask the Pope if he’s Catholic?
“Why not ask the Pope if he’s Catholic?”
Next time he’s in town. 😉
Andrew
Boris (#7751),
From my trusty Merriam Webster’s Collegiate Dictionary (10th Ed.)
Definition 2 (I’ll give you that def 1 doesn’t apply, it’s that digit thing):
Got any more?
Good morning to all. I’ve now, following on Joel’s excellent suggestion, tried to see what subsets of the data might show about the amplification (or “damplification” of the tropospheric temperature.
Unfortunately, this has generated more heat than light. The problem is that when we use the shorter datasets (half the length), the error bars get so wide that it is hard to say much about the situation, particularly at the longer end of the scale.
However, the shorter datasets (half UAH as well as half the RSU) still show the “damplification” (the opposite of amplification) at the shorter end of the scale. So the mystery remains.
This brings up a recurring problem that I have with climate science, which is the habit of climate scientists of avoiding providing underlying data or citations by simply saying that something is “well-known”, or “consistent with basic theory”. Whenever I see such claims, I run …
But I don’t run away. I run to the data to see for myself. In this case, despite Santer’s claim that the amplification is “consistent with basic theory”, he provides no reference, and now that I have gone to look for it, I can’t find it at either the short or the long time scale.
So before we get into the question of how many fingerprints can dance on the head of a pin, or whether those things dancing on the pin are really fingerprints, it would behoove us greatly to examine the data itself so we can see what is going on. The results of my analysis say that at both ends of the time scale, there is little or no amplification.
Now it is certainly possible that my analysis is flawed, wouldn’t be the first time … so once again I ask anyone who might be interested to download the data and repeat the analysis. Or do a different analysis. Or find a flaw in my analysis.
Unfortunately, folks seem much happier discussing things akin to exactly what the meaning of “is” is … which could serve as a metaphor for the whole field of climate science. All too often, people want to look at theory and models and definitions rather than data and observations and calculations. Before we get to the definition of “fingerprint”, before we start parsing how many whorls it takes to make a fingerprint and how many points of similarity we need to say two fingerprints are from the same source, how about let’s find out what’s actually happening in the real world?
w.
willis–
It would be interesting to see whatever graphs/ analyses etc. you posted. If you send me graphs, I can host them and then you can explain what you found in more detail.
Andrew said at Comment #7749:
And Boris replied at Comment #7751 with:
And Andrew responded with:
This shows a very common misconception amongst non-scientists, that the scientific method is about ‘proof’. Science is in fact about the converse: experimentation or data-gathering either support, an hypothesis or a theory, or they disprove said hypothesis/theory.
As Boris points out, ‘proof’ is the province of mathematics and distillation – in science all ‘possible’ knowledge is never in practice known, and this Rumsfeldian fact precludes us ever being able to definitively ‘prove’ any hypothesised condition. It is possible of course to find a refuting fact, amongst an infinity of facts, that will disprove an hypothesis/theory, but the current filibustering over the so-called hotspot does not qualify.
I concerns me that so many denialist discussions in the last several weeks and months have promulgated the myth of the routineness of scientific ‘proof’. Whilst some blogs (which shall remain nameless) are beyond any hope of redemption in their standards of scientific discourse, I would hope that this one rises above such rabble and is able to discern what science is, and what it isn’t.
I would hope too that all here would understand why the current debate about the hotspot is not sufficient to qualify as ‘disproof’ of AGW, no matter with which side of the debate one concurrs. More generally I would expect that most here realise why an absence of ‘proof’ is not the same as the adoption of ‘religious’ faith, and would thus refrain from confabulating these two modes of interpretation.
I would also hope that the quality of contributors here is sufficient that no definition is required of the difference between the lay understanding of a ‘theory’ and its scientific usage.
Heaven help the thread if such are not the cases…
Bernard J. (Comment#7759
I would hope too that all here would understand why the current debate about the hotspot is not sufficient to qualify as ‘disproof’ of AGW, no matter with which side of the debate one concurrs. More generally I would expect that most here realise why an absence of ‘proof’ is not the same as the adoption of ‘religious’ faith, and would thus refrain from confabulating these two modes of interpretation.
Not a disproof of AGW as a theory, because it has so many parameters that can be twiddled that it can become again consistent with the data. And when it is shown to be consistent with the data, minimum scientific integrity requires that there will be at the same time a publishing of the temperature trends from these same runs that have no hot spot. ( too much to expect? scientific integrity by politicized scientists?) I very strongly suspect that the temperature trends will no longer be predicting imminent catastrophy.
Certainly the lack of a hot spot is a clear disproof of the specific model runs published with 800 pages by the IPCC and rammed down the throat of politicians. One disagreement invalidates a theoretical proposal. Do not forget Newtons theories fell because of a disagreement of one number in the position of Mercury.
If the word is being stampeded into dangerous economic decisions because of a disproved run of a specific theory, the disproof should be loud and clear and that is what skeptics are shouting about.
Play with your AGW theories to your hearts content, but stop trying to push the world into an energy self immolation.
lucia:
If I write a signature while holding a pen with someone else, is that my signature? I don’t think so but that’s what denialists are saying, i.e. the upper TT hotspot comes from the decrease in lapse rate that is a theoretically expected response to the overall GHG warming. The signature is written by both the theoretical expectation and the GHG warming. Somehow this is transformed into a signature of just the GHG warming.
At least lucia doesn’t disagree that Douglass is either ignorant or lying in Boris’s quote.
dover_beach:
That should be: No TT-hotspot, no TT-warming. We know for a fact that the TT has warmed, so what went wrong? There are only two possible answers to this question:
1. There really is a hotspot (which relies on measurements made a long time ago).
2. The average lapse rate doesn’t actually decrease over a long period in response to a warming TT. In this case GHG-caused warming (or any other cause of warming for that matter) won’t produce a hotspot.
It is really very simple. If the models predict something that is not observed it reduces my strength of belief in other predictions from those models. The whole point of the models is that they claim to embody the physics that allows them to predict how the temperatures of the surface, troposphere and stratosphere will change in response to a different level of GHGs or other factors. All of the layers are coupled in the models and an error in one part will almost certainly have an effect on the others.
The objective part of science is essentially empirical. Sometimes there are equations that fit the factual evidence and then one can explore beyond the originally observed data to find how far those equations still hold.
The utility of this kind of work is mainly derived from the ability to make predictions. Our strength of belief in our predictions will normally increase every time that our predictions come true. However, strength of belief is always partly subjective and the passage from frequentist evidence to strong expectation is not along mathematical lines.
For me at least, a failure to find the TT hotspot would strike at the credibility of the current models and their predictions of all temperature changes from whatever cause. The models provide the main evidence for attributing the most recent rising temperature trend to GHGs. If the models are broken, they cannot provide evidence one way or the other.
anna v:
There’s no certainty about it because there’s no certainty that there is no hot spot. This “hot” spot is only about 0.3 deg C of difference over a period of 30 years after all. Some data sets say there is a hot spot anyway. We still don’t have accurate enough measurements of this trend difference to say one way or the other.
dover beach said:
“Boris, a TT hotspot is an identifying feature of GHG-forcing. No TT-hotspot, no GHG-forcing. Simple as that.”
You are absolutely wrong. An increase of solar of similar magnitude produces a TT hotspot in the models. Any surface warming would. Read the thread again.
On the effects of aerosol forcings, from the IPCC ch 9, page 671, p1 “”The combined forcing by greenhouse gases plus ozone is 2.9 ± 0.3 W m–2 and the total aerosol forcing (combined direct and indirect ‘cloud albedo’ effect) is virtually certain to be negative and estimated to be –1.3 (90% uncertainty range of –2.2 to –0.5 W m–2; see Section 2.9).””
Re: #7759
Bernard,
Between proof and pure speculation resides science, on this we agree.
However, short of proof the gaps are filled in with belief. Thats my point.
AGW is a belief system. A religion. You are just as religious as I am. 😉
Andrew
Bernard J. (Comment#7759)
I look forward to your letter of complaint to these guys.
http://www.eea.europa.eu/highlights/proof-of-climate-change-unequivocal
Or these guys.
http://www.livescience.com/environment/global_warming_041115.html
Or these guys.
http://www.panda.org/about_wwf/what_we_do/climate_change/problems/global_warming/scientific_proof/
I think you will also find that a surprising numbers of folks who comment here don’t need a lecture on the scientific method.
Boris. The game is up. An absence of TT hotspot must mean an absence of GHG forcing.
O’Neill. Thanks for saying you have no idea whether there is a real TT hotspot or not. Courageous.
Lucia, maybe you should add the following figure to the top in your discussion
http://icecap.us/images/uploads/GHGModsvsReality.jpg
This is also discussed in
http://icecap.us/index.php/go/joes-blog/response_letter_to_chairman_dingell_by_ross_mckitrick/
sometimes a plot is worth a thousand words, or repeats of “fingerprint” definitions.
Thamks Anna. From Anna’s link, Ross McKitrick wrote to US Congressional Committee re TT Hotspot Fingerprint:
“» The effects induced by greenhouse gases are so large relative to other forcings (positive and negative) that the total pattern is predominantly a reflection of the contribution of greenhouse gases.
» The tropical troposphere should have been heating up at a rate of at least 0.25 oC/decade over the past few decades in response to historical greenhouse gas emissions. A middle-range warming projection scenario in the IPCC report predicts warming of about 0.5 oC/decade should now be observable in the tropical mid-troposphere.”
http://icecap.us/images/uploads/Response.to.Dingell.EAQ.pdf
Annav–
I wanted to check the time frames for the figures. Here’s what Ross wrote on the page yo linked:
Boris, you say
I know I’ve said this before, but I’ll say it again, as I don’t seem to be getting through.
In a debate about whether or not the models are able to simulate simulate GHG warming, it does not add anything to the discussion to say “the models show a TT hotspot from any forcing”.
Whether or not they in fact show a TT hotspot from any forcing, since we don’t know if the models can simulate any warming correctly, it means nothing. It is a circular argument to say “the models say all warmings have the same outcome (TT hotspot), and therefore they must be right about GHG warmings”. We have no more information about whether the models are right about the response to the other warmings than we do about their response to the GHGs.
In fact, there are good theoretical reasons to think the warming patterns would be different. For example, GHGs warm both day and night, and solar only warms during the day. Since in general the atmosphere is stable at night and overturning during the day, we would a priori expect the warming patterns to be different.
In addition, solar forcing causes tropical thunderstorms to form, and GHGs do not. Since thunderstorms both route the heat around much of the troposphere and turn much of the heat into work, once again we would expect that ceteris paribus, the patterns would be different.
If that is the case, and I have found no one willing to argue the contrary, then any evidence that the models show a TT hotspot for all warmings would tend to show that the models are wrong in all cases, rather than right in all cases …
This is why I keep asking for evidence (as opposed to model predictions) about whether the TT will happen in response to other warmings. Unfortunately, all I’ve gotten from anyone is that it is a “well-known fact”. The continued failure of the AGW adherents to produce even a scrap of evidence to back up their position, or to produce a coherent argument why all warmings are created equal, suggests to me that they have neither.
However, Boris, I’m quite willing to examine any evidence or theories as to whether all warmings should affect the troposphere equally. If you have it, bring it on … you could start by explaining how a forcing that creates thunderstorms (solar) will heat the troposphere in exactly the same way as a forcing that does not (GHG’s). Since the mechanisms that move the heat aloft are totally different in the two cases, it seems extremely unlikely to me that the results would be the same, but since you obviously think different, I’d be interested to hear your logic.
w.
Willis, not that I agree with the following, but to be considered, that with the models depending on positive forcing due to water vapor, and with stratospheric enhanced cooling, there would be probably(?) too little difference to differentiate the scenarios with the current amount of CO2 if solar was modelled much greater.
This does not mean the models are correct, but rather consistant in how they handle water vapor, and transport (the ones that use a hyperviscous layer to handle the PDE’s, ODE’s, etc.)
Thanks, Lucia,
here is the completer reference for the plots and the color scale included in chapter 5 p116 of the US CCSP 2006 report
http://www.climatescience.gov/Library/sap/sap1-1/finalreport/sap1-1-final-chap5.pdf
John, I don’t actually know if the models show an identical TT hotspot from any forcing. They may. At the moment, it is of no interest to me. Why?
For me, I say again, the models are the last step in the process, not the first. For me, the first step is the data. I have shown that neither the RSS data, nor the UAH data, show any amplification at either the long or short time scales.
I tried shortening the dataset by about four years, to see if the lack of amplification at long time scales is an artifact. It does not show up in the shortened dataset, indicating (but certainly not demonstrating) that it is not an artifact.
Until we understand why there is no amplification at either the short or the long end of the data, there is no point in seeing what the models have to say. We don’t understand the data.
Having said that, since none of the models include thunderstorms, I would not be surprised if the models show the same results for all forcings. Thunderstorms route rising air and energy around the troposphere, by shielding it from any interaction (either by radiation or by mixing) with much of the troposphere. The lack of any representation of thunderstorms in the models is the gaping hole, the giant problem, the elephant in the room that none of the modelers want to talk about.
As a result, all the models can possibly tell us is what would happen in a world without thunderstorms … interesting, but hardly applicable to our world. It also makes it totally meaningless to me whether the models show the same result from all forcings.
w.
Boris writes:
“You are absolutely wrong.”
What part of:
” a TT hotspot is an identifying feature of GHG-forcing. No TT-hotspot, no GHG-forcing….It is one of three other identifying features of GHG-forcing, the others being surface warming and stratospheric cooling. If any one of these three are missing it is very likely something other than GHG-forcing is occurring. And it is intellectually honest to question the magnitude of GHG-forcing if one of the three identifying features is missing.”
don’t you understand?
BTW, I would have preferred the fifth sentence to read: “If any one of these three are missing it is very likely something other than GHG-forcing is the more significant forcing.
Boris also writes:
“An increase of solar of similar magnitude produces a TT hotspot in the models. Any surface warming would. Read the thread again.”
Firstly, if a TT-hotspot is an identifying feature of ALL forcings, then it is also an identifying feature of GHG-forcing. You cannot, as you seem quite happy to, say that a TT-hotspot is characteristic of all forcings but then suggest it is not characteristic of GHG-forcing. That would be nonsensical.
Secondly, we are very unlikely to observe a solar forcing of a magnitude that would produce a TT-hotspot this century considering that a solar forcing of 2% is at least eight times the increase in TSI observed over the last three hundred years. The solar forcing of 2% was and remains a red herring.
Willis, you continue to make an excellent point. The idea that the TT-hotspot should be identical for all forcings seems strange to me. Similar, may be; identical, very unlikely.
Looking at the graphs that anna posted I don’t see any thing like what the models predict. The hot spot is missing, the cool areas above high lattitudes, nothing. So far the only data that is in agreement is the stratosphere cooling, but the observations show more cooling than the models predict.
dover_beach:
Exactly the same thing can be said for surface warming, i.e. if surface warming is an identifying feature of ALL forcings, then it is also an identifying feature of GHG-forcing.
If surface warming says there is GHG warming and the TT-hotspot is obscured by noise, which observation are you going to base your conclusions on?
Chris said:
“Exactly the same thing can be said for surface warming, i.e. if surface warming is an identifying feature of ALL forcings, then it is also an identifying feature of GHG-forcing.”
Yes, no problem with that.
Chris also said:
“If surface warming says there is GHG warming and the TT-hotspot is obscured by noise, which observation are you going to base your conclusions on?”
Yes and no. You appear not to have noticed that I said “AN identifying feature” as opposed to “THE identifying feature” of GHG-forcing. Surface warming and the TT-hotspot are two of three identifying features for GHG-forcing, the other being stratospheric cooling. Like I have said previously, you need to establish the presence of ALL THREE identifying features to establish GHG-forcing. The presence of surface warming does not, on its own, establish GHG-forcing. But the absence of any one of the three identifying features suggests that GHG-forcing may not be the predominate radiative forcing.
Chris O’Neil
“If surface warming says there is GHG warming and the TT-hotspot is obscured by noise, which observation are you going to base your conclusions on?”
Probably not the one that is contaminated by trash incinerators, barbecues, paved surfaces, asphault roofs, car radiators, ac exhausts… and the list goes on.
This effort to cloud the observations in uncertanty and noise is just getting plain fake.
Mike C (Comment#7799) December 29th, 2008 at 12:02 am
“This effort to cloud the observations in uncertainty and noise is just getting plain fake.”
This summarizes the issue. If the observations on which the models are based are uncertainty and noise, then GIGO, the models are uncertainty and noise and that should be made loud and clear to the politicians.
If not , and there is a meaning to the observations then the model runs have to go back to the drawing board, even with one disagreement with the data,and this should be made as loud and clear to the politicians.
In either case the IPCC model runs go to the trash can.
dover_beach:
I didn’t get any acknowlegement of my previous point:
dover_beach’s point should be: No TT-hotspot, no TT-warming. We know for a fact that the TT has warmed, so what went wrong? There are only two possible answers to this question:
1. There really is a hotspot (which relies on measurements made a long time ago).
2. The average lapse rate doesn’t actually decrease over a long period in response to a warming TT. In this case GHG-caused warming (or any other cause of warming for that matter) won’t produce a hotspot.
i.e. either there really is a hotspot which establishes ANY type of forcing that causes surface warming or a hotspot doesn’t actually arise in response to ANY type of TT warming.
So what is the predominant forcing?
Mike C:
Awful lot of those out in the oceans, especially the Arctic Ocean.
A few obsevations from a retired physicist and someone new to this site. It seems to me there’s been a lot of irrelevant distraction here about semantics such as use of the word “fingerprint”. What’s wrong with good old-fashioned evidence? Looking at anna v’s comment#7778 and the figure: from my experience with computer models and the validation thereof – if someone had come to me with a computer model validation calculation as in the top of the figure and compared it to the data in the bottom of the figure, I would have said “time to throw away the computer model and start again. There’s got to be something wrong with the physics in the model”. End of story.
I’ve extended my Hadcrut3 Tropics reconstruction back to 1871 (when reliable Nino3.4 numbers become available).
It puts a little different spin on things. There is now more warming in the 1871-on period than in the 1979-on period.
I’ve noticed this in other reconstructions – there is less warming after the satellites came on stream in 1979 – perhaps keeping the temperature record honest finally – perhaps warming just slowed down afterward – there are other cycles in the warming rates – there is also a definitive trend to the rate of warming based on the logarithmic impact of CO2 – warming goes exponential from about 1955 to 1985 and then becomes linear from about 1985 to 2100 and then the line starts to flatten out afterward – this is just a characteristic of the log warming nature of CO2 – perhaps that is what we are seeing here instead.
But if you look at this Tropics temperature reconstruction, there is a very, very good match for something which is just a mathematical model – it actually looks better than it really is but nonetheless.
http://img509.imageshack.us/img509/5935/tropicshadcrut3mb0.png
http://img84.imageshack.us/img84/7530/tropicshd3warmingwf6.png
Bill Illis (Comment#7805) December 29th, 2008 at 11:14 am,
Can you put in a time delay for CO2 forcing in your model? If the system has a sufficiently long time constant caused by, for example, ocean heat storage, then the apparent sensitivity will be less than the actual sensitivity. Of course a really long time constant, 100 years or more, would be unrealistic. But a constant on the order of 30 years is not out of the question. The short lag in the Nino factor is not necessarily indicative of the time constant for a radiative imbalance forcing.
Willis quotes Boris [7770] who quotes dover beach:
[“Boris, a TT hotspot is an identifying feature of GHG-forcing. No TT-hotspot, no GHG-forcing. Simple as that.â€
You are absolutely wrong. An increase of solar of similar magnitude produces a TT hotspot in the models. Any surface warming would. Read the thread again.]
This is a bit embarrasing, but to put it a different way:
IPCC: The moon is made of green cheese
dover beach, Willis: The moon is yellow, so it isn’t made of green cheese
Boris: There is another theory that says that the moon is covered with (green) cress so the observation that the moon is yellow does not refute the theory that it is made of green cheese.
As I said, it’s a bit embarrasing but I think that’s the form of Boris’ argument.
To Willis,
The to-date log warming line (using the numbers from the 1871 reconstruction) will take 200 years to reach the 3.25C warming level if it continued at the same rate.
(The numbers from the 1979-on reconstruction never get there.)
The trends-to-date say there needs to be a significant up-tick in temperatures within the next 5 years or so or the warmers will have to go back to the drawing board.
“I didn’t get any acknowlegement of my previous point:
dover_beach’s point should be: No TT-hotspot, no TT-warming.”
No, it shouldn’t. According to the enhanced greenhouse theory, historical levels of GHG-forcing give rise to a TT-hotspot. That is what figure 9.1 (c) and (f) suggest. Deal with it.
“We know for a fact that the TT has warmed, so what went wrong?”
Do we? Some datasets show weak or negative amplification (or as some wit above called it damplification) which is contra conventional wisdom.
“i.e. either there really is a hotspot which establishes ANY type of forcing that causes surface warming or a hotspot doesn’t actually arise in response to ANY type of TT warming.”
The above quote needs to be clearer. The presence of a TT-hotspot would not establish that ANY type of forcing that causes surface warming would also create a TT-hotspot since we have to give consideration to historically-witnessed levels of X, Y or Z-forcings. And if the surface warming caused by GHG-forcing does not also lead to TT-warming that includes a TT-hotspot, then the models the IPCC includes in its reports are wrong.
Chris O’Neil
“Awful lot of those out in the oceans, especially the Arctic Ocean.”
Funny that the oceans are cooling, and that the Arctic Oscillation is in a warm phase.
Shouldn’t Schmidt and Hansen et al from GISS “consider their positions* as we say this side of the Pond?
Chris, regarding the measurement of sea surface temperatures, you have the following problems:
Buckets to engine inlets and back to buckets. Between canvas and wooden buckets. Between uninsulated and insulated buckets. Etc. So, no, not BBQs, but very definitely inhomogeneities and discontinuities in the SST record.
http://www.climateaudit.org/?cat=67
dover_beach:
Yes, it should. In the tropics, simple thermodynamics (as covered in many undergraduate meteorology courses) dictates that it should actually warm faster, up to about 1.8 times faster by the time you get to 12 km or so, regardless of the cause of surface warming. The fact that most of the surface warming is caused by GHG forcing is beside the point. Figure 9.1 (a) shows that solar forcing produces a hotspot in proportion to its forcing.
Yes we do know the tropical troposphere has warmed. Even satellite derivations produced by an AGW denialist show the tropical troposphere has warmed.
i.e. corrupted:
Go back and read the whole statement again and don’t just quote out of context and state your own opinion as if it is just restating what I said:
“dover_beach’s point should be: No TT-hotspot, no TT-warming. We know for a fact that the TT has warmed, so what went wrong? There are only two possible answers to this question:
1. There really is a hotspot (which relies on measurements made a long time ago).
2. The average lapse rate doesn’t actually decrease over a long period in response to a warming TT. In this case GHG-caused warming (or any other cause of warming for that matter) won’t produce a hotspot.
i.e. either there really is a hotspot which establishes ANY type of forcing that causes surface warming or a hotspot doesn’t actually arise in response to ANY type of TT warming.”
BTW, I didn’t get an answer to the question:
Chris, you are bordering on the ridiculous.
“Figure 9.1 (a) shows that solar forcing produces a hotspot in proportion to its forcing.”
Proportionate being the important qualifier. Figure 9.1 (a) shows no TT-hotspot; figure 9.1 (c) does; (a) does not and (c) does because in the first the actual solar forcing over the last 300 years does not produce a TT-hotspot while the actual GHG-forcing does. Simple as that.
“Yes we do know the tropical troposphere has warmed. Even satellite derivations produced by an AGW denialist show the tropical troposphere has warmed.”
Yes, I admit I could have phrased that more clearly, but the point is that although it has warmed it hasn’t warmed to the extent theory expects, at a level of 1.8 times surface warming.
“Even satellite derivations produced by an AGW denialist [charming] show the tropical troposphere has warmed”
Yes, but it shows “damplification”. We see less warming in the mid-upper troposphere than we see at the surface. According to Santer at al 2005, only one dataset shows something like the warming that is expected in the mid-upper troposphere.
“Go back and read the whole statement again and don’t just quote out of context and state your own opinion as if it is just restating what I said”
I didn’t quote you out-of-context. Neither did I state my opinion as if I was restating what you said. I simply responded to what I thought you had said; that is, I gave my opinion. If I had done what you thought I had I would have prefaced the first sentence with: “Chris O’Neill should have said….” which seems strikingly similar to what you had previously done to me, “dover_beach’s point should be:…” How very revealing.
“So what is the predominant forcing?”
I don’t know.
Wow, you take a few days off and things keep running on there own. Joel – thanks for stepping in, and pointing out the Santer paper with its details on these correlations.
Willis, unless I’m misunderstanding your graphs (comment #7692) both UAH and RSS are agreeing on a remarkably flat ratio of about 1.2-1.5 in the range from around 6 months to 250 months, between the tropical troposphere and surface temperature anomalies. That sure looks like a very convincing argument that the relationship (over up to 20 years time-span) is almost hard-wired by the physics of the system. The “short range” (1 month, 2 month etc.) numbers have huge error bars so I’d put those values as close to meaningless. The longest range numbers show disagreement between UAH and RSS (starting around 250 months) so it sounds like those are hard to rely on also.
I.e. what Willis is showing is exactly the mid-troposphere amplification I’ve been talking about all along here; looks like it’s pretty convincingly proved by these satellite observations, and is there whatever the cause of warming (much of the warming and cooling that leads to these correlations is not climate-change-related, but due to El Nino and other “weather” effects).
Now, this is still observational proof only of the one-dimensional definition of “hot spot”, that the temperature anomaly peaks in the mid-troposphere (for tropical latitudes). To demonstrate what Lucia seems to require as a hot spot (getting back to the mathematical definitions I tried to rigorously specify in comment #7640) would require some sort of two-dimensional set of measurements. But we don’t have Lucia’s precise definition of “spot” to compare with so it’s hard to tell exactly what measurements would satisfy it.
In any case, what Santer and Douglass have been looking at is clearly the question of whether the one-dimensional definition of “hot spot” is there or not, nobody seems to be concerned about the latitudinal “spotness” of the temperature anomalies. Lucia, do you agree that nobody has been actually trying to experimentally test what you believe is the actual definition of the “hot spot”?
Arthur,
While you cherry pick a 20 year time span from the middle of the graph, the other posters here and on CA are actually graphing this stuff out, something that no one on your side wants to do.
davidc (Comment#7807) December 29th, 2008 at 3:07 pm
Willis quotes Boris [7770] who quotes dover beach:
[“Boris, a TT hotspot is an identifying feature of GHG-forcing. No TT-hotspot, no GHG-forcing. Simple as that.â€
You are absolutely wrong. An increase of solar of similar magnitude produces a TT hotspot in the models. Any surface warming would. Read the thread again.]
This is a bit embarrasing, but to put it a different way:
IPCC: The moon is made of green cheese
dover beach, Willis: The moon is yellow, so it isn’t made of green cheese
Boris: There is another theory that says that the moon is covered with (green) cress so the observation that the moon is yellow does not refute the theory that it is made of green cheese.
As I said, it’s a bit embarrasing but I think that’s the form of Boris’ argument.
Indeed – what about
IPCC: The moon is made of green cheese
dover beach, Willis: The moon rocks (observations) we have examined are rock not cheese, so it isn’t made of green cheese
Boris: There is another theory that says that the moon is covered with rock so the observations that only examined the surface does not refute the theory that it is made of green cheese under the rock.
dover beach: How deep does your theory say the surface rock is?
Boris: That is obvious, go and work it out yourself.
Mike C:
Sure if you say so.
dover_beach:
Believe it or not people already know about this.
I note you guys are no longer arguing about whether the TT-hotspot is related to TT warming and have shifted to a different argument. This shows that you have lost the original argument. I have no interest in chasing your goal posts all around the paddock. Good bye.
Arthur–
I believe testing vertical is a test. If the spot is there, the vertical profile should show it. So, it’s a sufficient test (provided the data are of adequate quality).
More detailed tests would be useful. But if the data aren’t sufficient for the vertical profile tests, it seems unlikely they’d be any better if someone tried to test whether the lateral profile is there!
O’Neill. It was you and the other warmers who shifted the argument, as you always do. No TT Hotspot = no CO2 derived AGW = models all wrong=AGW theory all wrong. Time for resignations at GISS and the other modelers.
Mark–
Going from “No TT Hotspot” to “AGW theory all wrong” is going too far.
If no hotspot exists there is either a) no warming or b) some missing physics in models. Some missing physics would mean the model predictions of the magnitude of warming are quite uncertain… but that doesn’t mean GHG’s don’t warm at all.
The surface has warmed, and that’s in line with what is anticipated if GHG’s do impeded the flow of radiant energy from the earth’s surface. GHG’s have increased. So, there is at least some evidence that the warming is due to the addition of GHG’s. So, “all wrong” is way, way too far!
That said, there does seem to be a meme to try to suggest that failure to detect the hotspot presents no problem what so ever as far as our ability to predict. Some seem to want to suggest it would be no problem even if the data were good and we knew for certain it wasn’t there.
Obviously, if the hot spot never forms at all, or ultimately forms but is an order of magnitude weaker than predicted by models — or the simplified physics encompased in “it’s just the adiabatic lapse rate”, then that means people have missed something that is sufficiently important to affect the sensitivity. It would mean cast doubt on the predictive ability of models as currently constituted.
But right now, we don’t know. The hotspot may simply be undetectable because the data are not sufficiently good. (Or at least, that is the case if we limit ourselves to data up to Dec. 1999 and ignore data up to 20008, as that is all Santer discusses.)
Re: #7823
No matter what happens, the hypothesis still could be right/correct/accurate.
What falsifies this hypothesis?
Andrew ♫
I stumbled on this today from ClimateAudit and read it in its entirety. A bit tedious in spots, but interesting. It should be used as a text for debate students. Logical fallacy after logical fallacy. Arthur (et. al.) should visit the link below and see just how many of these they committed:
http://en.wikipedia.org/wiki/Fallacies
And so, lucia, you are fighting a losing battle. If winning this argument means convincing Arthur, et. al., the very fact that they are not willing to debate in a logical manner demonstrates clearly that you will not succeed. Their minds are closed.
However, on the bright side, you have convinced me of the following:
1. The models (at least some of them) used by the IPCC predict a zone of air in the troposphere – roughly centered on the equator with a vertical height of roughly 10km – that is warmer than the surrounding air when AGW forcings are sufficiently large to result in the official warming predictions of the IPCC. This is independent of whether natural forcings are included or excluded.
2. None of the models used by the IPCC to produce the official warming predictions demonstrate such a zone when AGW forcings are excluded (or are at a level below that required to produce the officially predicted warming).
3. Actual observations of troposphere temperatures do not show the existence of such a warm zone.
This means at least one of the following MUST be true (red herrings, composition fallacies, and semantics aside):
1. If AGW is occurring, the models used to produce the official warming predictions of the IPCC are flawed and the resultant predictions are highly likely to be inaccurate, especially given the sensitivity of the models with respect to parameter changes. (Whether this flaw is in the magnitude of natural forcings, CO2 warming, water vapor feedback, adiabatic lapse rates, or the color of my shoes is unknown and immaterial.)
2. If the models are NOT flawed, then AGW on the scale predicted by the IPCC is NOT occurring.
Either the models are flawed – and, hence, inaccurate – or AGW on the scale predicted by the models is not occurring.
All other points are moot.
Quick self-followup:
Before anyone misinterprets, the statement that the models are flawed and, hence, inaccurate is NOT equivalent to saying that AGW is not occurring. All it is saying is that the predictions of the models are likely to be unreliable.
Lucia. Sorry, but if current AGW theory requires a TT Hotspot (Models generally predict a relative warming of the free troposphere compared to the surface in the tropics since 1979 (IPCC).), and there isn’t a Hotspot, then the theory is wrong. It’s no good saying the measurements might be wrong etc..If the measurements are wrong (eg Urban Heat Island Effect) then no AGW theory can be proven or disproven.
Additionally, Willis has said above, “…despite Santer’s claim that the amplification is “consistent with basic theoryâ€, he provides no reference,… The results of my analysis say that at both ends of the time scale, there is little or no amplification.” (Current AGW theory requires Water Vapour amplification of CO2 effect). No amplification=current AGW theory is wrong.
You assert that “The surface has warmed.” http://rankexploits.com/musings/wp-content/uploads/2008/12/rss_nov_2008-500×341.jpg
doesn’t look like warming to me. Also, you must know that the accuracy of the whole of the current temperature data set is being called into question?
It’s all wrong, and heads are going to have to roll.
MarkR–
For AGW to be true, we need only two things: a) global surface warming and b) caused by something done by mankind.
The hotspot predicted when people make additional assumptions about leading order effects, and include those in models. (Or just point to a simplified analysis involving the adiabiatic lapse rate.)
So, not finding the hotspot doesn’t negate AGW in it’s entirety, but it would point to something missing (or wrong) in the physics as embodied in those theories that specifically predict the hotspot. But, it would not necessarily mean GHG’s cause no warming– or even that they cause very little warming. We wouldn’t know exactly what is wrong with the theory until we had a new one that explains the features we actually see (and which is later confirmed by predicting as yet unseen features.)
Mark, you and lucia are saying the same thing in different words.
Lucia’s: “If no hotspot exists there is either a) no warming or b) some missing physics in models. Some missing physics would mean the model predictions of the magnitude of warming are quite uncertain… but that doesn’t mean GHG’s don’t warm at all.”
is equivalent to:
Mark’s: “Current AGW theory requires Water Vapour amplification of CO2 effect). No amplification=current AGW theory is wrong.”
So you’re saying the same thing there. But what you can’t do is conclude that, because the CURRENT theory of interaction between anthropogenic GHGs and the climate is inaccurate that the concept of AGW is false. It MAY be false, but is not necessarily so.
Totally OT:
Uni-Hamburg still hasn’t posted their November ice data and JAXA hasn’t updated for two days. I need a data fix NOW.
Hi Lucia. I cannot disagree that there may be other AGW theories which may be correct, but I do say that the current AGW theory, as deliniated by the IPCC, is demonstrably wrong.
Hi Ryan, I’ll happily agree to agree with Lucia.
“For AGW to be true, we need only two things:
a) global surface warming and
b) caused by something done by mankind.”
So the ‘real’ hypothesis doesn’t have anything to do with C02 specifically?
Andrew ♫
Santer has moved beyond saying that temperature alone is the “fingerprint” for global warming. He summarizes his latest work as follows: “Our investigation was the first to use rigorous statistical
“fingerprint†methods to tackle the question of why water vapor has
increased.”
“Basically, “fingerprinting†involves searching for a computer
model-predicted pattern of climate change (the “fingerprintâ€) in
observed climate records. Fingerprint techniques allow researchers to
examine a change in some property of the climate system and make
rigorous statistical tests of the different possible explanations for
that change.”
Our key findings were as follows:
A) Despite the relatively short length (19 years) of the observed water
vapor data, we were able to identify a “fingerprint†of human activities
in this observational record.
B) Unlike most previous “fingerprint†work, our study used results from
virtually all of the world’s major climate models. We showed that our
identification of a human “fingerprint†in satellite-based water vapor
records was robust to current uncertainties in climate models.
C) The model results enabled us to “disentangle†the contributions of
different factors to the overall increase in water vapor. We found that
in climate models, this increase in water vapor was primarily due to
human-caused increases in greenhouse gases.
D) Bottom line: our results suggest that there is an emerging signal of
human activities in the moisture content of Earth’s atmosphere. The
climate system is telling us a consistent story. The observed changes in
temperature, moisture, and atmospheric circulation fit together in an
internally- and physically-consistent way. link http://www.yaleclimatemediaforum.org/2007/10/lawrence-livermore-scientist-ben-santer-sees-benefits-in-proactive-approach-to-news-media/
It seems that if the GCM’s can’t explain the temp hotspot “fingerprint” then global warming will be linked another “fingerprint” such as moisture content.
I thought the science on this was settled and everyone was in consensus about how climate worked and that’s why we cand predict the next 100 years so accurately.
Hi Mark . . . and hi lucia . . . I’ve read stuff here for awhile (got here from McIntyre’s blog) but never posted before those first couple of posts. I couldn’t help myself . . . the deliberate misinterpretation of lucia’s point was too much.
For me, the defenses of the IPCC’s version (i.e., cataclysmic) of AGW that I’ve seen come down to semantics and “what-if” games. It might be true, but I have a difficult time understanding how the models predicting catastrophic consequences can be so widely believed when there are obvious fundamental problems with the models themselves, a lack of understanding of many of the physical processes involved, and – most importantly – the uncertainties and problems with data sets that feed those models. In my opinion, far too few of the papers written on AGW question these things with the necessary rigor (to use Gavin’s favorite term) and it is truly astounding. Aren’t scientists supposed to be the skeptics? Where has “extraordinary claims require extraordinary proof” mantra gone?
Over the past 15 years, the concept of AGW has slowly ceased being a science and become a religion. It MAY exist, but before such a claim is made, shouldn’t the science actually support it?
Anyway . . . all finished with the off-topic pseudo-introduction. Carry on, folks! 🙂
Willis writes:
Willis, I agree. But I’m not saying that. I’m simply saying that the TT hotspot is not unique to GHG warming and is in no way a fingerprint of GHG warming. Let’s say we were trying to determine if a surface warming was caused by solar or GHG. We wouldn’t even consider the TT hotspot as an identifying feature. We’d most likely look at the stratosphere.
Now, if the models are wrong, it does not disprove GHG warming at all, which is what denialists like Douglass claim in their press releases. Rather, it would show the models don’t handle the saturated adiabat correctly, which could be problematic in some ways, but would very likely not affect the modeled forecast (Think about all the things the models get right–those aren’t suddenly invalidated.)
dover_beach writes:
If a feature is common to all forcings, then it wouldn’t be an identifying feature, now would it?
Ryan O. writes:
For someone who comes here to lecture on logic, you’d think that you would present a logically valid conclusion. No one disagrees that the models are flawed–all models of anything are flawed. But you present no evidence that this particular flaw (if it is indeed a flaw: the data aren’t conclusive) would affect global temperature predictions. There is also plenty of evidence that models are accurate (prediction of cooling stratosphere, GMST, increased humidity, etc. , etc.) despite the flaws.
Boris writes:
[quote]For someone who comes here to lecture on logic, you’d think that you would present a logically valid conclusion. No one disagrees that the models are flawed–all models of anything are flawed. But you present no evidence that this particular flaw (if it is indeed a flaw: the data aren’t conclusive) would affect global temperature predictions. There is also plenty of evidence that models are accurate (prediction of cooling stratosphere, GMST, increased humidity, etc. , etc.) despite the flaws.[/quote]
Hi Boris. The Models predict ABC and D, but ABC and D don’t happen, therefore the models are wrong. End of.
Ryan– The “A” in AGW is anthropogenic not “carbon dioxide”. 🙂
Of course, it’s believed they main way the “As” have caused the GW is by adding CO2 to the atmosphere, causing an enhanced greenhouse effect.
Ah, the logic professor reveals himself! Excuse me while I tithe to the great church of the Royal Society of the UK.
Please read the Bibl…er, the IPCC Fourth Assessment Report before making wild claims about the integrity of working scientists.
BTW . . . how do you do quotes here?
Boris writes:
“For someone who comes here to lecture on logic, you’d think that you would present a logically valid conclusion. No one disagrees that the models are flawed–all models of anything are flawed. But you present no evidence that this particular flaw (if it is indeed a flaw: the data aren’t conclusive) would affect global temperature predictions. There is also plenty of evidence that models are accurate (prediction of cooling stratosphere, GMST, increased humidity, etc. , etc.) despite the flaws.”
No, for three reasons:
1. Models are simplifications that are meant to make complex things calculable. In that respect, no model is a complete representation of reality. If that’s what you’re calling a flaw, sure. All models are flawed – by definition. But that is clearly NOT what my conclusion is talking about. My conclusion is that the simplifications made result in AT LEAST ONE calculation that is aphysical (the existence of a hot spot). That makes the other calculations suspect unless you can prove that the simplification that results in the incorrect calculation has not affected any other result.
2. The models are not meant to simply predict things qualitatively or relatively (“cooling stratosphere”, “increased humidity”, etc.). The models are advertised as QUANTITATIVE predictors. Simply because the models show physical TRENDS does not mean that the absolute values of the calculated parameters are physical. As an analogy, I can use Newtonian gravitation to qualitatively predict behavior in strong gravitational fields, but any quantitative statement I make may deviate from reality by a significant margin.
3. For someone who purports to be using logic, you again resort to a logical fallacy: shifting the burden of proof. If the model is shown to produce at least one aphysical result due to inappropriate/inaccurate simplifications or parameter assignments, the burden of proof is YOURS to show that the other results are NOT aphysical.
“Of course, it’s believed they main way the “As†have caused the GW is by adding CO2 to the atmosphere, causing an enhanced greenhouse effect.”
Here we go with belief again.
Is the AGW hypothesis about C02 or not? Or is it about both C02 and not C02? Can anybody say Word Games?
“Please read the Bibl…er, the IPCC Fourth Assessment Report before making wild claims about the integrity of working scientists.”
So the IPCC is where I should go for authoritative answers? Appeal to authority. Logical fallacy. 😉
Andrew ♫
Lucia: I know. 🙂
Boris . . . on your prized 4th Assessment: http://www.lifesitenews.com/ldn/2008/dec/08121803.html
And I quote: “The report quotes Ivar Giaever, a Nobel Prize Winner for Physics, saying: ‘I am a skeptic … Global warming has become a new religion.'”
Or, if you prefer a more academic source, try this:
http://scienceandpublicpolicy.org/images/stories/papers/reprint/CarterMyth/carter_myth.pdf
Besides, I’m neither a logic professor nor do I live in the UK.
But Ryan, the models have made good QUANTITATIVE predictions. You are making the perfect the enemy of the useful.
Andrew, if you want to learn, you can read the IPCC report. Most “skeptics” haven’t, so they get to make up what they are arguing against. How easy for them!
Boris: Yes, the models have made quantitative predictions – and at least one of those predictions is aphysical.
My problem with the IPCC is that the AGW is not presented in a manner that is Popperian falsifiable (at least in the short term). Every time a prediction is made that is later shown not to match reality, the modifications to the models are proposed (no observed hot spot = thermal winds!). So while I may (emphasize MAY) concede that over time the models are increasing in accuracy, the degree of accuracy is unknown.
Where is the test of mutual exclusivity? There are models that show significant AGW and ones that do not. Some of the models used by the IPCC can actually show cooling if the forcings/feedbacks are adjusted WITHIN THE OBSERVED UNCERTAINTIES (see the carter document above). So if there is no mutually exclusive test between the theories/models, how is it that you can select one over another as representing reality?
Even within the IPCC report itself, they admit knowledge deficiencies of many of the physical processes that govern the climate (like cloud cover and the ability of the biosphere to expand in response to the increased CO2 concentration). They wash their hands of this by simply postulating that these unknowns are less significant than the AGW effect and give some anecdotal evidence that this MAY be true. They then go on to produce models with assumed values/behaviors for these effects – most of which have not been experimentally confirmed. This is good science?
To me, this seems pretty basic.
Boris,
Thanks. I have read sections of it -sections of it that have been debunked by others.
Why don’t you quote a paragraph of the report that’s especially convincing to you? Then I’ll know exactly what you are talking about. I’ll be able to see for myself why you believe the way you do, ’cause let’s face it… this is all personal belief, as like I said earier, no proof has been presented.
Andrew ♫
I may need to put the word “religion” in the spam filter. . . (There must be some sort of dis-em-vowler plugin available. If not, I may need to write one. )
Chris O’Neil:
“Funny that the oceans are cooling
Sure if you say so.”
Nope, I don’t say so, the data does.
http://www.cru.uea.ac.uk/cru/data/temperature/hadsst2gl.txt
A third of the warming since 1900 has disappeared in the last 6 years.
Since I seemed to have diverted the discussion, maybe I should try to get it back to the original topic. 🙂
.
First, discussions about whether a result from a model is a fingerprint, characteristic, unique characteristic, must be placed in the proper context. When discussing models, this context is not simply the model itself but also the values chosen for the adjustable parameters in the model. Different initial conditions, feedback constants, forcing constants, and any other adjustable parameters can result in widely divergent predictions. If the object is to attempt to validate the temperature predictions of a model, then the settings of the adjustable parameters are a necessary to limit the discussion to the validity of the prediction.
.
In this case, the debate is whether the apparent non-existence of a hotspot brings into question the accuracy of the IPCC’s temperature prediction. The context, then, should be limited to the models and parameters chosen that yield the IPCC’s predictions. I will call this set of models and set of associated parameters “Model X”.
.
With this in mind, the semantic games about the meaning of “fingerprint” are meaningless. In the context of Model X, both considering AGW forcing alone and considering the combined effects of AGW and other forcings results in a tropospheric hot spot that is not seen when AGW is removed. In the context of Model X, it is a fingerprint.
.
If this hot spot does not exist (as it seems not to), then the validity of the remainder of the predictions of Model X should be investigated. This, I believe, is Lucia’s point.
.
Taking Model X and adjusting the solar forcing parameter (or using a different model entirely) removes the context. While it is true that the hot spot can be reproduced by forcings other than AGW, by changing those parameters, you have (using the definition presented above) essentially produced a new model – Model Y.
.
This is the fundamental logical fallacy of Boris’ argument. He has changed the context. He is not talking about whether the hot spot is a fingerprint of AGW in Model X; he is talking about whether the hot spot is a fingerprint of AGW in Model Y. There is a name for this. It is called a red herring, and Lucia nailed it.
.
The point is that Model X (the set of models and parameters used for the IPCC’s official predictions) demonstrates a tropospheric hot spot that is not seen if significant AGW forcing (within the context of Model X) is not present. If such a hot spot does not exist, then this calls into question the remainder of the predictions of Model X.
.
It is entirely irrelevent that the hot spot can be reproduced by other means. Lucia’s point was that the nonexistence of the hot spot brings into question the validity of Model X. This is entirely true.
.
This does not mean that AGW does not exist. It may mean that. It may also simply mean that the physical means by which AGW is postulated to occur in Model X is wrong. The global temperature predictions of Model X may even be accurate. Just because the model produces aphysical results for one feature does not necessitate aphysical results for other features.
.
But the existence of ANY aphysical results should, at least, give us some pause.
.
If I have misstated or misinterpreted your point, Lucia, I apologize. 🙂
Mike C,
The data you’ve provided are for sea surface temperatures, not for the oceans as a whole.
Ryan O,
You seem to think I’m arguing something that I’m not.
The denialists claim that the “fingerprint” of GHG warming has not been observed. As this thread has shown, the only way this is true is if we define fingerprint to mean anything we darn well please. So what is and isn’t a fingerprint becomes central to debunking this erroneous claim.
Boris: Again, you are taking the claim out of context. The context is that the models the IPCC uses to calculate future warming show a temperature anomaly in the troposphere that is characteristic of AGW. Within the context of those models, this IS a fingerprint because it is not shared by other forcings. Even with a strict interpretation of “fingerprint” as a unique characteristic this statement holds true.
.
Now if you run around changing the forcings, you can cause the “fingerprint” to appear from any number of forcings. But in doing so, you’re changing the context of the claim. The variable parameters in the model are part of the model. If you change them, you have a different model.
.
This has nothing to do with the definition of fingerprint. It has everything to do with the context. You are saying that the hot spot is not a fingerprint of GHG warming in a model with a substantial increase in solar forcing. Great. Wonderful. And irrelevant. I could conceivably show that by changing the methane forcing that cow farts could produce the same effect. It has as much relevance.
.
The IPCC’s claim is that the models they use to predict future global temperatures necessarily result in a tropospheric hot spot that is characteristic to GHG warming. They never claim that this fingerprint also extends to some random model with increased solar forcing that is not used for predicting future global temperatures. That would be nonsensical.
.
So if the claim by the IPCC is that their prediction of global temperatures is based on models that show a feature characteristic of GHG warming (i.e. “fingerprint”), then a demonstration that the feature does not exist is a demonstration that either the models got the physical mechanisms wrong, or, if they got the mechanisms right, then GHG warming is not occurring.
.
You are right in that many denialists ignore the first half of that conclusion (that AWG may indeed be occurring, albeit via a different physical process than used by the models) and focus on the “GHG warming is not occurring” alternative.
.
But the right answer is not to attempt to cloud the issue by changing model parameters outside of those used for the global temperature predictions. And if you do want to go down that route, you have assumed the burden of proof that the remainder of the results from such a model represent physical reality.
.
Two wrongs don’t make a right.
I would like to point out that pressure and altitude are only interchangeable variables if the temperature is constant. As temperature increases, the atmosphere expands and the altitude for a given pressure increases. This means that the rate of temperature increase at constant pressure will be less than the rate of increase at constant altitude. In the case of a 6 K/km constant lapse rate, the delta T at 400 mbar for a 3 K change in surface temperature from 298.2 to 301.2 K is only 2.54 K (253.28 to 255.82 K).
Boris:
“If a feature is common to all forcings, then it wouldn’t be an identifying feature, now would it?”
I’ll assume that English is your second language. A feature common to all forcings identifies them as forcings. You see, I said “an” not “the” identifying forcing; there is a difference. To distinguish between several forcings we normally require more than one identifying feature, i.e. surface warming, TT-hotspot, stratospheric cooling. I have said this previously though not in these words. May be you should pay closer attention.
http://rankexploits.com/musings/2008/who-expects-a-tropical-tropospheric-hot-spot-from-any-and-all-sources-of-warming/#comment-7795
Damn, missed the edit time limit. Correction to my fourth sentence; it should read: You see, I said “an†not “the†identifying feature; there is a difference.
Hm . . . wonder if Lucia’s going to put in a spam filter for new guys who post too much . . . 🙂
.
Because I am anticipating Boris’ response to my last, I thought about a way to simplify the argument to show his fallacy clearly . . . at least in my opinion, anyway. Haha. So for anyone who cares, here goes . . .
.
.
Situation: I. P. Seesee, Ph.D., develops a theory that a certain type of rock has built-in jet packs to make it accelerate faster when dropped. Dr. Seesee puts together a mathematical model to describe the physical processes involved: gravitation, rock cross-section and associated drag, the effects of tumbling on drag, the interaction of the rock (which happens to be magnetic) with the earth’s magnetic field, the postulated contribution of the jet packs, and other factors.
.
Dr. Seesee applies the model to a situation in which the rock is dropped from a 500 ft. tower. He selects various rock configurations and uses his model to calculate velocities at 400 ft, 300 ft, 200 ft, 100 ft, and impact. He plots these results and discovers (among other things) that rocks with jet packs display a velocity distribution at 400 ft that is different from rocks without jet packs. He calls this velocity distribution the jet pack “fingerprint”.
.
.
Enter Doug Lass, Ph.D. Dr. Lass is skeptical of the existence of jet packs on the rocks. He decides that the best way to evaluate Dr. Seesee’s theory is to measure the actual velocity distribution of the rocks at 400 ft and analyze the results for the jet pack fingerprint.
.
Dr. Lass performs his experiment and discovers that the fingerprint is absent. He then concludes that either Dr. Seesee’s model is physically inaccurate, or that jet pack rocks do not exist, or both.
.
.
Enter Dr. Bo Riss. Dr. Riss claims that Dr. Lass’ conclusions are in error because the jet pack “fingerprint” is not exclusive to rocks with jet packs. To substantiate his claim, Dr. Riss modifies Dr. Seesee’s model by boosting the gravitational coefficient and eliminating the jet pack forcing. He then shows that under these conditions, the resulting velocity distribution is identical to that calculated by Dr. Seesee.
.
Dr. Riss goes on to show that he can reproduce the velocity distribution at 400 ft by modifying the drag coefficient, by adding a large electromagnet beneath the tower, and also by adding a strong downdraft. This, he says, clearly demonstrates that the velocity distribution is not a “fingerprint” at all . . . at least, not unless you loosen the definition of “fingerprint” beyond all reason. The velocity distribution is common to all types of downward acceleration.
.
.
The error of Dr. Riss’ analysis, hopefully, is clear. He eliminates the uniqueness of the “fingerprint” by introducing modifications that are either aphysical (changing the gravitational constant) or are not present in the real system (adding the electromagnet). He has changed the context of the question.
.
Changing the model parameters to reproduce the original results from a different source is a red herring.
.
Furthermore, Dr. Riss’ analysis does nothing to show why the anticipated velocity distribution from the original model fails to match observations. Even if Dr. Riss’ modifications are reasonable, they serve only to reproduce a prediction that does not match reality. Rather than attempting to show how Dr. Seesee’s theory could still be accurate, Dr. Riss focuses his efforts on showing that “fingerprint” is a poor choice of word to describe the velocity distribution. While an interesting exercise in semantics, this does not contradict either Dr. Lass’ observations nor his conclusion that Dr. Seesee’s model does not properly predict the observed behavior of the rocks.
.
The semantics argument about “fingerprints” is a red herring.
.
Lastly, Dr. Riss does not continue his analysis to show that his modifications to the model produce physical results at impact, 100 ft, 200 ft, 300 ft, or any altitude other than 400 ft. If his modifications produce aphysical results at the other altitudes, then we could immediately conclude that the modifications were in error. It does not matter that they reproduce the “fingerprint” if they cannot match observed velocities at other heights because they cannot physically happen. Using them to support the statement that the “fingerprint” is not unique to jet packs is fallacious because the conditions required to remove the uniqueness of the fingerprint cannot occur.
.
The use of aphysical models is a red herring.
.
.
Of course, there are ways of seeing who is correct about the jet packs. We can take the rocks and put them in wind chambers to measure their drag. We can monitor wind conditions at the drop site and add the appropriate correction factors to our raw data. We can conduct experiments to determine the actual value of the gravitational coefficient.
.
In other words, we have the ability to falsify Dr. Seesee’s theory.
.
In the case of AGW, no such ability exists. Can we actually confirm the negative feedback of an expanding biosphere on CO2 levels? Can we actually confirm the postulated adiabatic lapse rates? Can we actually confirm the effects of cloud cover, or, moreover, how cloud cover will respond to changes in humidity? Can we actually confirm the level of solar forcing?
.
For the most part, climate models have to take educated guesses at these factors, the equations that will adequately model their effects, the relative contributions of each, and the interactions between each factor. The IPCC itself admits that many of these factors are poorly understood.
.
That is Boris’ get-out-of-jail-free card. Rather than attempt to show how these poorly understood factors and estimations result in models that are able to accurately predict future temperatures, he simply changes the parameters to achieve the desired result. It’s what the modelers have been doing for years.
.
Sounds very scientific and convincing.
Ryan
First– I do use SpamKarma, and it sometimes does get angry with new people. So, if your comments don’t appear, click the contact lucia link and ask why. So far, SpamKarma hasn’t gotten upset at you though.
More generally, I don’t mind people posting a lot, as long as they aren’t just talking to themselves. OTOH, if someone posts a whole string of comments and no one else is talking….. that’s a bit much.
I did write plugin to interfere with people who post over and over when no one responds. Should you ever be a target of that plugin, you will know. What happens is that somewhere after 3 comments in a row, by you, your comment content will be replaced with “Slow down ‘name’. Let other people talk!” Other people see “So and so is being given a time out. Their comment will appear is X minutes.” After “X” minutes, the comment is visible to all.
I am thinking of looking for something to disnitigrate the word “religion” though. The accusations of religion can get a bit much.
Hi Lucia – ok, so we’re agreed that nobody’s been trying to measure the latitudinal “spotness”, and the only actual measurement that all this discussion of a “hot spot” has engendered is looking at the one-dimensional claim I distilled my “hot spot” down to.
Now, we get to the heart of this half of the disagreement I have with you. Namely, as is evident from figure 9.1 of the IPCC AR4 WG1 report, every forcing that produces significant warming of the tropical surface also results in an amplification in the tropical mid-troposphere, a peak in temperature with altitude, which is what the measurements have been looking at. In that sense of hot spot,
(Lucia’s comments in green.)
This sense of “hot spot” permits spots that are not “hot” to be called hot.
it is most certainly expected from ANY and ALL sources of warming (of the tropical surface).
No. The only hot spot is from GHG’s.
That is I believe what everybody who has been claiming that, including Boris from the start here, is talking about. Not the latitudinal “spotness”, but the fact that there is a peak with altitude. It is indeed expected whenever you have tropical surface warming.
No. The latitudinal aspect is part of the spot. You are conflating a test for thing with the thing itself. (Though I’m not sure this is important.)
Now, there are, as has been pointed out by some people here, some forcings that would primarily affect only the polar regions – black carbon soot, perhaps. If it produces no warming at all in the tropics, then of course the expected amplification would be amplifying zero, and would still be zero. But if there’s any warming of the tropics at all, then you get the “hot spot”.
Yes. But this still leaves warming that doesn’t cause a TT hot spot. Or was everyone to understand that the interjection meant “any and all provided they warm the tropics and are warm enough, and any other requirements I forgot to mention?”
If you had a forcing that caused warming at the poles but cooling in the tropics, then you would get a cold spot. What primarily matters is tropical temperature change at the surface.
I also sense that you and some of the commenters are trying to force some sort of holistic notions (“one IPCC model of AGW, take it or leave it”)
Where have I said this? (I agree some commenter say this. You will note that I corrected some. But there have been too many comments all around xmas to interject on each and everyone.)
here when in reality the science is composed of many nearly-independent components that each have their separate roles, and which can in fact be evaluated largely independently of one another. Many of the different forcings, for example, have effects that can essentially be linearly added together, at least for small forcing levels, and the effects are close to linearly proportional to the forcing level. There is some non-linearity, but not a lot at the level of changes experienced so far, and of course even less when you limit the time period to 20 or 30 years. Of course there are many “models” that are considered both by the IPCC and even more by the climate science world at large. Some of those models have very different estimates of 20th century forcings, there is not just one set of numbers on that. Whether or not a particular choice of forcings is realistic or not is a largely independent question from whether the models that try to solve for Earth’s climate under those forcings are good, and each of these models is made up of many essentially independent components derived from basic physical principles, such as cloud responses, or water vapor feedback, or the lapse rate feedback, essentially the main topic of this discussion.
Each of those relatively independent pieces can be tested separately under various conditions – for example, Willis posted a graph that essentially tests the physical lapse rate condition associated with increased tropical evaporation across a wide range of time scales: theory predicts a horizontal line, and the best fit horizontal line is at amplification of about 1.4 or so. To explain a non-horizontal line you need to come up with some new theory, or evidence that the experiments are failing for some reason when there’s a deviation from horizontal. There are clear reasons that I and others have cited why the experimental results are dubious outside the flat regions – this is almost the opposite of “cherry picking” for several reasons – in particular there is statistics involved from both recent and early records in those numbers.
Not disagreeng with what you are saying here…But I’m not sure whose argument you are actually engaging. Maybe I skimmed them?
So there is good reason to believe in tropical mid-troposphere amplification of surface temperature anomalies. Is that a proof of climate models? No, it is merely evidence supporting one of the physical components that goes into many of the models. If for some reason this evidence for amplification disappears, would that be disproof of the climate models? Not at all, it merely takes away support for that physical component of the models – reducing the lapse rate feedback, in essence.
I’m not sure what your definition of “disproof” is. If the models as currently constituted predict features that do not exist, that will mean that some physical component in the model is wrong in some sense. If, after we collect precise accurate data we discover the models predict features that do not exist, then, to use the word proof colloquially, the models containing the flawed physical component will be prove wrong. If you don’t like “proven” wrong, then let’s just way it will provide strong evidence that that some physical components does not correctly describe the physical processes on the real earth.
In that event, people will try to replace the flawed physical compoenent with something that correctly describes the real honest to goodness physics.
But note that the lapse rate feedback is a negative feedback, that lowers surface temperatures, by allowing a lower temperature gradient from surface to tropopause. If the lapse rate feedback is absent, all else being equal, the models would see larger warming effects from a given increase in GHG levels. An absent tropical mid-troposphere hot-spot would give no comfort to those who think models are overestimating climate sensitivity!
So? Who has claimed otherwise? And if this is so, wouldn’t we want to know this?
Science is reductionist at heart – everything has a cause, and the seeking out and understanding of those causes is the key quest of the scientist. Climate models like any scientific analysis are inherently broken up into pieces that look at one or another aspect of the system – that’s why it’s perfectly natural to look at what happens when the forcing is purely of one type or another, as well as looking at what happens when everything’s realistically mixed together. It’s perfectly natural to look at hypothetical forcings, or hypothetical planets with some major feature different in some way.
No one has said it’s unnatural to look at hypothetical forcings or counterfactuals. We do this in engineering too. Heck, despite all the criticism of “lawyerly” discussions, lawyers do it too.
Looking at short-term (1970’s to present) or long-term (20th century) changes are similarly natural reductionist questions, there is certainly not one and only one way to look at climate responses, and what IPCC did with their figure 9.1 is definitely not the be-all and end-all of such analyses.
Who said those figures are the be-all and end all?
When you look at the underlying components separately there’s no question of one component washing out the others, you are treating them separately and independently. That’s your right as a reductionist. And that’s what I, Boris, and everybody else mean when we say the hot spot is indeed there for ANY and ALL sources of (tropical surface) warming – when treated independently, on an equal footing, in a linearized fashion, you always get the same amplification of the surface warming in the mid-troposphere.
Arthur– but you aren’t seeing the problem. These little interjections are dropped into conversations about comparisons of model predictions to observations. The context of the conversation already implies that we are not discussing counterfactuals. When comparing observations to predictions, we are comparing observations to predictions based on forcings that actually occurred.
In this context solar forcing cannot cause a hot spot. It can barely cause a warm spot.
And for some reason, some wish to proclaim that other people cannot include the idea of realistic levels of forcing in the group of “any and all” forcings. Rather, those interjecting appear to want to insist that the idea of doing separate effects testing is the only point of view permitted when discussing whether or not a hot spot can be created by a particular forcing.
This is a peculiar point of view if (as at the CA post that triggered my post) people are discussing comparing observations to model predictions of what could happen on earth. In this context, some are insisting that realistic forcings are subservient to the idea of hypothetical forcings. Those same people seem to be insisting that other people are not permitted to discuss what can or cannot happen when we consider realistic levels of forcing.
If, someone wants to dive into these conversations that are discussing comparisons between observations and models and change the subject to hypothetical forcings, the onus is on the person changing the conversation to point out that they are now discussing forcings that are irrelevant to comparisons with observations. (At this point, the others can say, “Well, you are changing the subject. That has nothing to do with comparisons to observations.)
Ryan O – if you have a specific example of one of my “logical fallacies”, I’d love to see it. I am quite familiar with the range of them and do try to avoid them when possible… Without any evidence, you yourself are making the “bare assertion” fallacy…
You know, there used to be a wikipedia page discussing the “Google” fallacy. Unfortunatelty, it’s gone. I don’t know if the fallacy was a) asserting something and then telling other doubters they could find the proof if they weren’t to lazy to google or b) asserting something is false because they couldn’t find it on google.
Lucia
Arthur–
I’m going to do something I hate– add inline comments above.
Arthur, you say:
“that’s what I, Boris, and everybody else mean when we say the hot spot is indeed there for ANY and ALL sources of (tropical surface) warming – when treated independently, on an equal footing, in a linearized fashion, you always get the same amplification of the surface warming in the mid-troposphere”
Arthur, you have a theory that says that the hot spot is there for all sources of warming. The hot spot is not there (accepting the observations, which you seem to do). Either a) Your theory is wrong, or b) there is no warming from any source. If a) then the presence/absence of a hotspot can distinguish between sources of warming and the IPCC models seem to be wrong; if b) there is no problem with warming in the tropics.
You obviously don’t like a), so do you like b)? Or have I missed something? Is there a c) out there that I haven’t seen?
Borris
“Mike C,
The data you’ve provided are for sea surface temperatures, not for the oceans as a whole.”
That’s okay, Boris, The IPCC also admitted that there has been upper ocean heat content cooling in the same time period. The reason I cannot give you that data is because the clown in charge of it will not release it… although he is convinced that all that missing heat is hiding down in the deep ocean somewhere below 9,000 feet.
<blockquote.Or have I missed something? Is there a c) out there that I haven’t seen?
c) enhanced tropical troposhperic warming is there, but we haven’t ‘found’ it yet.
Or perhaps:
c) enhanced tropical troposhperic warming is there, although we haven’t ‘found’ it yet for all approaches to the analyses (remembering that some timescales do demonstrate the phenomenon) for any of a number of thus-far unconsidered reasons.
It would appear so.
“I am thinking of looking for something to disnitigrate the word “religion†though. The accusations of religion can get a bit much.”
Is this a case of the word ‘religion’ acquiring some kind of pornographic definition? Or is this a case of some posters being adverse to the word ‘religion’ and are complaining about it?
I’m not sure what it is about the word that is unacceptable, other than the moderator has a personal problem being associated with the word.
Andrew ♫
MIkeC– Please don’t call people clowns. (Not even if they aren’t here.)
Andrew– The problem with the word “religion” at climate blogs is that it’s both insulting and vague. It tends to derail conversations. I wouldn’t have anything against the word if I posted something about the Albigensian or Zoroastrian heresies.
Steve, UK- We’re back below zero!
Lucia,
But nobody is talking about heresies in general. We’re talking about only one specific belief, AGW. And it is a belief. You have stated as much.
Now, you can try to run away from that or block things that make AGWarmers look bad, but then you would be actively supressing the truth about the matter. Is that the kind of blog you have or want to have?
Andrew ♫
Ryan O,
You have redefined fingerprint to mean a common feature–the same thing Lucia has done. But, again, it is an absurd definition, because then you must agree that the surface warming of the zonal mean plot is a fingerprint of GHG warming, which is absurd. Once again, the point of fingerprinting is to find markers that uniquely identify a culprit.
And also, in claiming that models are junk, you forget that they do predict things that have happened. You also seem to forget that the Douglass data analysis is not widely accepted. There is not convincing evidence that the models are wrong about the TT hotspot.
dover beach says:
We know they are forcings already. But think: two people do not share a fingerprint. You claim that two (or all) forcings share a fingerprint. Absurd.
Andrew–
You seem to be jumping to the conclusion that my saying some peopel believe something implies their belief is based on religion. Saying something is “believed” implies nothing about the reasons why someone believes something. Jurors can come to believe something based on preponderance of the evidence. People can also believe something based on faith.
I don’t think discussion of religion makes AGWers look bad. I think very very few people on either side base their belief for or against AGW on religion. So far, I haven’t read anything to suggest anyone does base their position on religion.
The arguments alluding to religious belief are generally poor arguments. They generally reflect badly on those who advance them. That’s why I don’t like them.
“…they do predict things that have happened. ”
Like the fortune tellers on TV? 😉
Andrew ♫
Boris–
Douglas used the data that were considered accepted at the time the paper was submitted. That’s all anyone can do. The data were later corrected. The papers describing the corrected data are cited in Santer, and were published after Douglas.
I have no reason to doubt the corrections. However, it’s unfortunate that the errors in the observations were not identified before observations appeared to conflict with the model predictions. These delays in addressing observational issues are bound to cause problems for those who want to convince skeptics that the observations really do support the models.
The other problem with Douglas was their technique– which did not account for data uncertainty even using the data available at the time of publication.
On your fingerprint defintion: Had the IPCC wished to use your definition they would have included “unique” in their definition of climate fingerprint. They don’t. Moreover, the general usage in peer reviewed papers does not require fingerprints to be unique. If you want to call the IPCC usage, or usage in peer reviewed papers, absurd, that’s fine. But people discussing climate are still going to adopt that usage.
@ Lucia: Apologize for causing a firestorm with the use of religion. It was meant as an analogy, not a literal . . . 🙂
.
Arthur:
Sure.
.
The discussion centers around the IPCC claim in 9.2.2.1 that the feature shown in figure 9.1 is (within the context of the models examined by the IPCC) characteristic of AGW and is robust between the models. In fact, the portion of the paragraph that Lucia underlined states that all of the shown features from the figure are robust between the models. So not only does this mean that the hot spot associated with AGW is robust, the lack of a similar feature among models when AGW is excluded is also robust.
.
Fallacy #1: Diversion.
.
.
How is it substantive to discuss whether Lucia uses the same terms as the IPCC? Does it matter if the IPCC specifically uses “fingerprint” or “hot spot”? Until the issue is confused by the semantics game, it is clear that Lucia is using the word “fingerprint” to indicate that the feature as shown in Fig 9.1 c) and f) is (within the context of the models the IPCC examined) characteristic of GHG warming and not of other types of warmings. This is the same thing as the IPCC claims. Note that the underlined portions of Chapter 9 are a clear claim that GHG warming produces a different temperature signature (both in magnitude and zonally) in the atmosphere from other types of forcings . . . again, within the context of the models the IPCC examined. The place that this is graphically shown is Fig. 9.1.
.
You can play the Derrida deconstructionist game if makes you happy, but it’s irrelevant. The models the IPCC examined produce a feature in the tropical regions of the troposphere that is distinctly different for GHG warming when compared to the other forcings considered. The name does not matter. The descriptor does not matter. The only thing that matters is that, within the context of the models the IPCC examined, GHG forcing produces this feature and other forcings do not.
.
Verdict? Diversion (red herring).
.
Fallacy #2: Genetic fallacy.
.
This seems to be your fallback policy for everything. Whenever a specific claim of a specific model/theory predicting AWG is shown to be false, your response is:
.
True. And irrelevant. You’re changing the context.
.
The context is that the models the IPCC examined – and these models necessarily include the types and magnitudes of the forcings/feedbacks the IPCC evaluated – produce a feature in the troposphere that is a characteristic of GHG warming and is not shared by other forcings. If the feature is shown not to exist, then the description of AWG within the models is incorrect or incomplete.
.
This, of course, says nothing about other models which do not show this feature. On that, I do not think anyone is disagreeing with you. But if the feature does not actually exist, it unequivocally shows that the models the IPCC considered and used for their temperature predictions do NOT accurately predict tropospheric temperature anomalies . . . and, by your own statements that tropospheric temperature anomalies are connected to surface temperature anomalies, they are likely to be inaccurate in their surface temperature predictions as well.
.
Instead of dealing with the above, you merely move the goalpost by saying that nonexistence of a hotspot does not undercut the validity of the IPCC models by introducing models that do not show the same feature. That’s equivalent to saying that just because you can’t find evidence of frame dragging that General Relativity is still correct because Newtonian gravitation does not result in frame dragging.
.
If the above sentence doesn’t make any sense . . . EXACTLY. Existence or nonexistence of frame dragging in Newtonian gravitation is unrelated to frame dragging in General Relativity.
.
Verdict: Genetic fallacy.
.
Fallacy #3: Composition fallacy.
.
Another of your favorites. First you give the false impression that the forcings can be independently tested and verified. They cannot, at least not to the degree required by the models. The IPCC itself admits that many of these factors – key factors – are poorly understood.
.
Second, even if they could be independently tested and verified, their behavior in isolation says nothing about their behavior together. Independent testing (or, in your terms, “evaluation”) of different forcings makes no statement whatsoever about their interactions with each other. This problem is rampant in the AGW discussions.
.
The claim that you can simply combine forcings within a model by adding, multiplying, or otherwise arithmetically/geometrically combining their effects is an assumption. It is not proven. The methods for combining them have not been tested, except insofar as the resulting models match historical temperature trends that, on a geological time scale, are ridiculously small.
.
If you can show me a robust design of experiments that quantifies not only the main effects of the forcings but also the interaction effects that is supported by observational data, I will retract the above statement. Until then, your assumption that the whole shares the characteristics of the parts is classic composition fallacy.
.
Verdict: Composition fallacy.
.
I could continue, and maybe I will later, but I will stop there for now.
.
The only arguments that have been presented that in are in any way relevant to the original discussion are the ones questioning Douglas’ techniques and data. If you want to question his conclusions, that is the right way to do so. The rest of the stuff is fallacious.
.
.
Boris:
.
Wha??? Did you actually read what I wrote? I said within the context of the models considered by the IPCC that the feature shown in Fig. 9.1 c) and f) was a unique characteristic of GHG warming. How in the world does that equate to “common feature?”
.
For being disparaging towards those who play word games, you do seem to play a lot of them yourself.
.
Besides, if you read Fallacy #1 above for Arthur and my previous responses to you, when it comes to the question of the hotspot in the context of the IPCC models, the definition of fingerprint is wholly irrelevant. It DOESN’T MATTER. I could call it “behoosephrit”. It changes nothing.
.
.
Along with models that do not include AWG. Besides, if the IPCC models cannot accurately predict tropospheric temperatures, and by your own admission tropospheric temperatures are connected to surface temperatures, why would we assume that they get the surface temperature right? Who cares if they correctly predict atmospheric humidity?
.
Besides, which predictions, specifically, are you talking about?
.
.
No. I have consistently used the word IF when discussing the existence/nonexistence of the hot spot. I have never once claimed that Douglas was correct (except in the context of my example to show how your other arguments are red herrings).
.
ALL of my statements about the hotspot have been with the IF qualifier. IF the hotspot does not exist, then the IPCC models are incorrect in some way, shape, or form.
.
I have not once made the claim that the IPCC models ARE incorrect or that Douglas WAS right. You assume that is my intent, but it is not. My intent is to show that your defense is, by and large, logically flawed.
Like I said, your definition would conclude that the warming at the surface was a fingerprint of GHG warming, which is absurd.
What happened in the 20th century is irrelevant determining a forcing fingerprint.
Because they pretty much have been right.
BernardJ,
I said to Arthur: “(accepting the observations, which you seem to do)”
Your c) is that the observations are wrong. Well, they might be, but that opens up a completely different line of argument. If “might be wrong” is acceptable in general then no theory can be refuted by observations. So I think you need to have a reason why “might be wrong” is reasonable in this particular case. I can see two possible lines of argument:
1) faulty instruments or algorithms for interpreting instrumental data (climateaudit and wattsup.. provide a template for how to do that with surface instrumental data); or
2) the IPCC models show such outstanding performance in explaining other observations that this particular set of observations is probably wrong even though we don’t know why.
Again, any that I’ve missed?
Boris:
“We know they are forcings already.”
Yes, we know they are forcings, but finding an identifying feature like surface warming enables you to diagnose a possible change in one or more radiative forcings. Further diagnosis might identify a TT-hotspot which might sway our diagnosis to a change in GHG-forcing considering forcings we are likely to historically observe, and identifying stratospheric cooling might confirm this diagnosis.
“But think: two people do not share a fingerprint. You claim that two (or all) forcings share a fingerprint. Absurd.”
No, I’ve said they may share an identifying feature. The ‘fingerprint’ is the pattern of features we observe in figures like those reproduced in figure 9.1. Patterns within the whole pattern may be common to two or more forcings or unique to one alone. The idea however is to identify more than one feature for each forcing and so reduce the chances of mistake or misinterpretation.
Boris:
“What happened in the 20th century is irrelevant determining a forcing fingerprint.”
This is tantamount to saying that the fingerprints of those alive when a crime was committed are irrelevant to current criminal investigations.
Boris:
Forgive me for saying wha???? again.
.
First, the definition of “fingerprint” is entirely irrelevant. In my previous posts, I have shown why it is irrelevant. To avoid further confusion, perhaps you could cut-and-paste this whole thread into Word and replace every instance of “fingerprint” with any other word of your choosing. I understand how you want to use “fingerprint” . . . and I have no problem using it that way.
.
The problem comes in you ascribing your definition to the way the IPCC uses it. The IPCC clearly does not use your definition of the word in section 9.1. So while it is the same word in print, the meanings are different. One thing you don’t get to do in an argument is change the meanings of words after the argument has begun. So if you are uncomfortable using “fingerprint” in the more colloquial manner in which the IPCC and several other posters use it in this thread, then do the cut-and-paste operation and the problem will disappear.
.
Second, I have stated repeatedly that the “characteristic signature” (better?) of GHG warming IN THE CONTEXT OF THE MODELS THE IPCC CONSIDERED is shown graphically in Fig. 9.1 c) and f). As a whole, it is distinct from the model results when GHG warming is not included.
.
Note the qualifier “as a whole”.
.
The logical fallacy of division runs as follows:
.
[OBJECT] has property A. [OBJECT] is made up of Parts 1, 2, and 3. Therefore, Parts 1, 2, and 3 have property A.
.
In your case, I claim that the graphs in 9.1 c) and f) have the property that (within the context of the models the IPCC examined) they are a unique characteristic of GHG warming. The graphs have a distinct temperature anomaly profile WRT pressure (Part 1), a distinct temperature anomaly profile WRT latitude/longitude (Part 2), and demonstrate surface warming (Part 3).
.
Indiviually, these features DO NOT have the property that they are a unique characteristic of GHG warming within the context of the IPCC models. As a whole, they do.
.
So your statement that I must be considering “surface warming” as a “unique characteristic” of GHG warming is simply an exercise in the logical fallacy of division.
.
Incidentally, Arthur has committed this one himself several times.
.
.
On what? The models are continuously revised because they have this strange property of diverging from observation after a while. Simply revising the model to show that now it fits past data is not an exercise in prediction. It’s an exercise in curve-fitting. Showing a model matches the past (and only the very recent past, at that) is in no way validation of of its predictive power.
.
Besides . . . what have they been pretty much right on? If you make that claim, you should be able to provide objective evidence that it is true. 🙂
Ryan O. Bravo. Happy New Year to all.
PS. Have you tried the Rabett?
http://rabett.blogspot.com/
He needs a good dose of logic.
Lucia,
is it OK to use the descriptive term “bull headed” here??
;>)
KuhnKat– Hmmm….. Depends on you use it. I like to avoid name calling because calling people names is not an argument.
The erroneous use of the term fingerprint is what started this whole discussion. Reread Lucia’s post.
The fact that you continue to misuse the term “fingerprint” is entirely relevant.
We’ve discussed the IPCC definition of fingerprint: a response in space and/or time in a GCM to a specific forcing. This fingerprint is then compared to observations. You do not use observations to create a fingerprint of zonal means, because then you are comparing observations to observations. This should be obvious to everyone. I don’t know why people are having trouble with this concept.
Well, here’s some proof you don’t know what you are talking about: The IPCC does not use the term “fingerprint” in section 9.1 So from where did you glean this entirely new imaginary definition?
Here are the quotes involving “fingerprint” in the entirety of chapter 9:
Hmmm…won’t see Douglass or Christy quoting that section 🙂
Seems obvious the IPCC is using the common lay definition of “fingerprint” here. Makes sense, since this quote comes from the FAQ.
Finally:
I suppose you could have fun with that sentence, all those big words to redefine to suit your own agenda.
Oh, dear. Someone needs to get away from the logic textbooks and read some research. This time, actually READ the IPCC report rather than pretending to have read it.
PAWPRINT, what’s this nonsense about fingerprint.
However, there are Pawprints, and pawprints. A real Pawprint allows you to say which beast went bump in the night, which is the difference between cooling throughout the stratosphere and warming in the upper troposphere.
wow!!! near two weeks and still going on!!!!
will the post number 500 get a cookie????
happy new york for all!!!!
It doesn’t seem that there’s much stratosphere cooling since 1996, though.
If you’re using RSS or UAH, you’re not getting the upper stratosphere numbers, where most of the cooling expected for GHGs is occurring.
Lucia,
Is your belief in AGW based on the preponderance (having superior power or influnence) of evidence?
If so, please humor me and quote/link what you think is the best evidence for AGW.
Andrew
Summary of the debate between me and Boris:
.
Ryan O: The definition of “fingerprint” is irrelevant to the conclusion that if observations of the troposphere do not yield Fig. 9.1 c) and f) that there is something wrong with the models used by the IPCC.
.
Boris: But your definition of “fingerprint” is wrong.
.
Ryan O: Irrelevant. (Presents reasons)
.
Boris: But the models got other stuff right.
.
Ryan O: Please provide examples.
.
Boris: Your definition of “fingerprint” is wrong.
.
Ryan O: Irrelevant. But if it makes you feel better, replace “fingerprint” with any other word of your choosing. I will cease using the word “fingerprint”.
.
Boris: But the models got other stuff right.
.
Ryan O: Please provide examples.
.
Boris: But your definition of “fingerprint” is wrong.
.
Did I miss anything?
Ryan O:
These are your words. If you don’t think the definition of fingerprint is relevant, then stop arguing about it. Now you are trying to rewrite even your own words. How logical of you.
Examples are in the AR4 that you pretended to have read.
Ryan O writes:
“First, the definition of “fingerprint†is entirely irrelevant. In my previous posts, I have shown why it is irrelevant. To avoid further confusion, perhaps you could cut-and-paste this whole thread into Word and replace every instance of “fingerprint†with any other word of your choosing. I understand how you want to use “fingerprint†. . . and I have no problem using it that way.”
Boris responds:
“The erroneous use of the term fingerprint is what started this whole discussion. Reread Lucia’s post. The fact that you continue to misuse the term “fingerprint†is entirely relevant.”
Ryan responds:
“The definition of “fingerprint†is irrelevant to the conclusion that if observations of the troposphere do not yield Fig. 9.1 c) and f) that there is something wrong with the models used by the IPCC.”
Boris responds:
“If you don’t think the definition of fingerprint is relevant, then stop arguing about it.”
The jig is up, Boris.
dover_beach,
Can you read what Ryan O wrote? I quoted it right above the part of my post you quoted. Or are we going to now debate what a definition is? God, that sounds like fun.
Boris:
.
Your quote of mine is from 5 posts prior to my latest post. In the intervening time, I have ceased arguing about the definition of “fingerprint” and have instead decided to show how the definition is irrelevant. You seem to be pretending to read my posts in almost as much detail as you accuse me of pretending to read AR4.
.
AFA AR4 goes, the ability of the models to “predict” is predicated on fits to historical data. The fits are in turn predicated on the assumptions made for each of the various physical processes involved, along with estimates of forcing parameters.
.
So when I see something like (page 666, right column, 1st paragraph):
.
.
and compare that to the modeled AGW forcing over the 20th century:
.
http://www.ipcc.ch/ipccreports/tar/wg1/fig9-13.htm
.
I find that the uncertainty in aerosol forcing alone can account for half of the modeled AGW forcing (half because the models assume a nominal value for aerosol forcing of -0.8 W m-2).
.
And that’s just one forcing. The hand-waving that the IPCC does that this is still somehow “consistent” is ludicrous. “Consistent” is when the uncertainty in a single parameter in a complex multi-parameter model can account for over half of the effect you are examining? Wow.
.
And then, seemingly oblivious to this problem, they go on to say that the models are not consistent with estimates of ocean heat content variability. Not only are they not consistent, but the degree of consistency cannot be presently calculated. If the degree of consistency cannot be presently calculated, then no upper bound on the effect of the oceans moderating/exacerbating climactic conditions can be made.
.
Also I note that their confidence levels are derived from the models themselves. Therefore, if the models are incorrect (whether it is from the forcing values used, the mathematical description of the effects, omission of physical processes, or any other reason), the resulting confidence levels for the prediction are bunk because the underlying mathematics producing the prediction are bunk.
.
I additionally note that the range of expected temperature increases over the next hundred years (Table 3.2, page 46) shows “likely” ranges for the scenarios at 1.1 to 6.4 degrees C. This is interesting, given that in the 2001 report they estimated the “likely” range to be 1.8 to 5.6 degrees C.
.
So how is it if the data and knowledge of the climate has improved, as the IPCC claims literally everywhere in AR4, that the 2007 version has a significantly larger uncertainty in the temperature prediction?
.
Not only that, but the IPCC intentionally and deliberately reduced the uncertainty to this level by leaving out model runs that resulted in cooling:
.
http://scienceandpublicpolicy.org/images/stories/papers/reprint/CarterMyth/carter_myth.pdf (page 10)
.
I also find it hilarious that, in discussing solar forcings on page 673, that they refer to insolation changes of 0.83 W m-2 as small when it is over half of the proposed AWG forcing in the year 2000. Not only that, when trying to reconcile models with periods prior to the industrial age, they make another hand waving argument and, in the same breath, state: “Therefore, the total annual and global mean radiative forcing during the LGM is likely to have been approximately –8 W m–2
relative to 1750, with large seasonal and geographical variations
and significant uncertainties (see Section 6.4.1).” Insert laughter here.
.
What’s the error on that one? Given the errors they’ve considered “small” so far, I’d estimate it at -1 to -16 W m-2 . . . meaning the window for the models is so large that falsifying them by looking at past data is impossible.
.
I also note that much of the quoted observational data used in constructing the models was published after the Third Assessment Report. The model runs in AR4 were done after adapting the models to updated data . . . and the result was a greater uncertainty in the future temperature projections than in 2001 and an overall lower mean temperature rise.
.
So if the models aren’t being continually revised to fit past observational data when they begin to diverge from observation, then why aren’t we simply using the 2001 models? Or Hansen’s original models? If they were right, they should still fit. If they weren’t right and needed to be modified, then my comment stands 100%.
.
So yah, I’ve read it, Boris. So I understand exactly what I’m asking you when I request objective evidence that the models have predicted something correctly, rather than being big, expensive curve-fitting exercises.
.
And the lack of any examples in your response pretty much fit my expectations. 🙂
No it’s not. Why do you keep saying things that are obviously false and then continuing on as if no one has called you on it?
So? The uncertainty for aerosols cuts both ways.
Yes, but it’s the most uncertain forcing by far.
Um, table 3.2 is on page 243–and it’s something else entirely.
Hadn’t heard this chestnut before, so I decided to check it out–skeptic that I am. Here’s what your source says:
Sort of a shocking accusation. Stainforth rang a bell, name wise, but I found the cite easily enough:
http://www.climateprediction.net/science/pubs/nature_first_results.pdf
Then I searched for “obviously wrong” to find the quote.
But I got nothing.
The quote does not appear in the paper.
So, choose your sources more carefully.
Well, you read, but you’ve got to read more carefully. 0.83 W/m-2 was the increase in winter insolation. Summer insolation decreased by 0.33 W/m-2. The mean seasonal insolation decreased by 0.4 W/m-2. Glad you thought it was funny, but clearly this is a small impact over 1000 years.
Anyway, if you keep reading things and see only what you want to see, then you’re bound to fool yourself. But keep thinking you know everything about logic and all these mistakes you are making mean nothing at all–in fact, they never happened.
Sorry, I couldn’t leave this unanswered. Models aren’t changed to fit temperature data, they are changed based on physical observations. For example: there have been changes in the southern annular mode (theorized to be due to ozone depletion). These changes were not anticipated by models, but the models do need to take them into account.
So, yes, models are revised based on the latest physical observations. But, no, those observations do not include surface temperature.
RC explains it thus:
HTH
Boris:
“Can you read what Ryan O wrote?”
I don’t think my reading skills are in question here.
BTW Ryan, I did enjoy your parody (summary) above.
Arthur, thanks for your comments. You say:
Whoa, whoa, whoa. The 1 and 2 month numbers do not have “huge error bars”, that’s nonsense. Huge error bars would include the amplification we find in the mid-range time periods of around 1.5. That’s why I put the error bars there, to show that even including the error it’s nowhere near the 1.5 amplification level. Remember, those are the 95% confidence intervals.
In addition, many of the longer term periods do not show 1.5 amplification, including the error bars.
So no, Arthur, I’m not showing “exactly the mid-troposphere amplification [you’ve] been talking about all along here” in the slightest, that’s absolute nonsense. Y’all have been claiming that amplification should be a feature at all time scales … which I don’t find at either the short or long end of the record. Including the error bars.
Please re-examine the data I submitted in post 7692, and stop trying to hand-wave it away with scare stories of “huge error bars” and specious claims that I’m agreeing with you. It just makes you look like you’re not paying attention.
w.
boris, thanks for your response. Unfortunately, you are speaking again of the model world, not the real world. You say:
Look, boris, I know that you are saying the TT hotspot is not unique. It’s almost all you say … but you don’t back it up. I keep asking you for a single study, a single scrap of fact, a cohesive theory that explains why every form of heating will lead to a TT hotspot. You give … nothing.
I have provided you with theoretical reasons this should not be so, good reasons why the heating effects should be different. These include the fact that the sun only shines in the day, while GHGs radiate 24/7. Why should those heat the atmosphere in the same way, especially since the atmosphere overturns during the day and stratifies at night? You have not provided a single response to that issue.
I have also discussed how thunderstorms provide a clear mechanism whereby the tropical surface heat can totally bypass the majority of the troposphere, so the surface can heat with minimal tropospheric heating. Your response? Silence.
I also pointed out that GHGs have the biggest effect at the poles and smallest at the tropics, while solar effects are exactly opposite. Why should they affect the tropical troposphere equally? You don’t comment about that either. You just keep repeating your mantra, that they’re all the same, they’re all the same.
boris, it is possible that you are right, that a very similar TT hotspot may arise from all forms of forcings. But you’re the one making the claim, and so far you have not provided a scrap, even the tiniest morsel of evidence or theory to back up your claim. I, on the other hand, have provided both data showing amplification is not happening at the shortest and longest time scales, and several theoretical reasons why all forcings would not have the same result.
So it’s now your move … but please don’t once again waste your move repeating “The hotspot is not unique to GHGs”, you have done that enough. We know your claim. What we await is your evidence.
w.
PS – I cracked up when I read your statement that “Think about all the things the models get right–those aren’t suddenly invalidated.” You might not see the humor in that … but anyone who has built more than a few complex computer models certainly will.
dover: Thanks. 🙂
.
Boris:
.
Me:
Boris:
.
You are correct. I apologize; the mistake was unintentional.
.
I’m not going to argue about whose definition is right. It doesn’t matter who is right. It’s an irrelevant argument – the reasons for which you have been entirely silent on. I can only assume this means you agree.
.
If you don’t like the way the IPCC uses fingerprint on page 702 (which is obviously in reference to Fig. 9.1 – section 9.1 was a typo – and is, by your own admission above, is the colloquial use) or the way many of the rest of us have been using it, do a search-and-replace on it for some other word of your choosing.
.
From the Stainforth letter:
.
The supplemental info link is here:
.
http://www.nature.com/nature/journal/v433/n7024/suppinfo/nature03301.html
.
.
Which, in shorthand, means he put a drift requirement on the models that ends up preferentially removing the cooling models (a careful look at Fig. 1 in the main article shows very clearly that virtually none of the warming models and almost all of the cooling models are rejected by this) and the subsequent quality check is simply to make sure that he got all of ’em. There is no check on the discarded runs to show that they all exhibit UNphysical cooling. Nor is there any corresponding check on the runs that show warming to ensure that all of them show physical warming.
.
AFA the six cooling models that made it past that wicket, he produced a graph of exactly one (the worst case) and dismiss all of them by saying (without any justification) that they are all due to the same effect despite the fact that they had stable controls.
.
This is interesting in light of the fact that Essex and McKitrick showed that all cooling models should NOT be dismissed out of hand. Stainforth does so anyway with no check on the discarded models to verify that they all were, indeed, unphysical (or even to check to see what percentage were unphysical).
.
I suppose Carter could have said all of that, but his main point is valid. They were dismissed as “obviously wrong”.
.
Nice try at a careful reading, though.
.
.
.
And what happens to the models, that, after being updated with the “latest physical observations”, no longer match the historical record? They are discarded as unphysical.
.
No matter which way you slice this, the historical surface temperature record is the first wicket any model must meet. Not only that, the unknown parameters (like aerosol forcings) are back-estimated from model deviations from the historical temperature record:
.
(page 670)
.
In other words, once the physics of the model are set (including the GHG warming), estimates are made of the uncertain parameters are made by fitting the model to the historical record.
.
Hm. Sounds a lot like curve-fitting. The models take AWG as a given, and using that given, many of the uncertain parameters (like aerosols) are back-calculated by fitting the curve.
.
Final note. Your number of a decrease of 0.4 W m-2 for insolation changes applies to the northern hemisphere only. The global insolation change was not given. And the point is not that the number in isolation is small. The point is that it alone is 25% of the supposed forcing due to AWG. How many 25%’s does it take to make AWG go away? Better yet, look at aerosol forcings (which were back-calculated using models that already presuppose AWG, so there’s no circular reasoning there at all *insert sarcasm*) with an uncertainty twice that of the insolation changes due to orbital variations alone, throw in the IPCC’s admission that estimates of past solar forcing is subject to “significant uncertainties”, an admission that the role of the oceans is poorly understood, another admission that the role of clouds is poorly understood and difficult to model, an admission that actual observations from volcanic eruptions are difficult to interpret . . . should I go on?
.
It should be no surprise or any great revelation that the only reason the models actually work with the historical data is because the uncertain parameters are limited by back-fitting the models to historical data once the physics of the models are established.
.
If that isn’t curve-fitting, I don’t know what is.
Boris,
You say in 7840.
“Now, if the models are wrong, it does not disprove GHG warming at all…”
I agree with you about that. On the other hand, if the models are wrong, neither can they be used to prove GHG warming.
You really have to accept that or I might accuse you of having double standards. 😉
“prove GHG warming”
Evidently, in Science, you don’t need proof. Proof is irrelevant. You can believe whatever you want. You don’t even have to respond reasonably to posts. Heck, you don’t even need to respond at all. Just make assertions and claim your opponents just don’t get it. You’re covered.
Andrew ♫
This is insane. Figure 9.1 appears 28 pages before 702.
So was your number. Are you even making an effort to understand here?
As for your reading of Stainforth–models are routinely examined to see if they have a stable climatology. Your claim that Stainforth made some decision with the intent of removing only cooling models has no evidence to support it. In fact, it sounds like you are accusing Stainforth of cooking the books, which, in the end, is all you denialists have.
If you’d care to carefully and honestly consider sources, I’m happy to discuss known issues with you. If you are going to continually misread cites and throw around conspiracy theories, then I have no interest.
Willis,
One of the problems I’m having is finding a good source online for the saturated adiabatic lapse rate. The upshot is that as a parcel of air rises, it cools at a steady rate. However, when that parcel is saturated, as it rises and cools, latent heat is added from the condensation of water, which effectively alters the lapse rate by compensating for some of the cooling. As a result the middle troposphere will warm slightly faster than the surface.
But the key is that the lapse rate doesn’t care how the surface warms, so, the origin of the warming is unimportant.
Here’s the best online description I could find:
http://www.theweatherprediction.com/habyhints/161/
I’d imagine a meteorology textbook might have a more satisfying description, but you’d have to ask someone who knows more than I do.
I think you’re overstating the differences between forcings here. The GHG increased effect at the poles is slight. The difference at the tropics is negligible.
I’m not at all convinced by your discussion. But if you think this is the case, why does your data show amplification at most intermediate timescales? But I’m not sure what’s going on at short timescales–so maybe there is some process that accounts for this. Again, I’d ask an expert on this topic.
But once again, your data show amplification at most time scales.
Boris–
The adiabatic lapse rate, its impact on radiative convective models, and the difficulties in predicting vertical temperatures and sensitivity using these methods in discussed in chapter 4 of “A Climate Primer”. The figures show real temperatures (not anomalies). The discussion does not specifically address how adding GHG’s affects the anomalies, but rather simply explains how one bundles up what we use what we know about adiabatic lapse to try to understand the vertical temperature profile.
I have an older version of the book by both McGuffie and Henderson-Sellers, but I assume similar discussions are in the newest version:
Correction– The book does discuss sensitivity to CO2.
I have a couple of questions re: polar amplification. Do the expectations of polar amplification arise primarily from the rapid radiation models in GCMs, an increase in advected heat expected from GCMs, changes in cloud cover expected from GCMs, or newer LBL transfer model calculations? All of the above, or? Also, is the amplification seasonal?
The reason I ask is I’ve read some of the lay-type arguments for polar amplification before, but can’t seem to remember the primary arguments right now, and while reading some of the comments about PA in this thread, I remembered something I found a little curious in Clough & Iacono 1995 JGR, so went back to it and read this again:
“Consequently, in the troposphere, increases in uniformly mixed gases have the effect of reducing the cooling rate associated with water vapor alone while at the same time increasing the downward flux at the surface.
It should be emphasized that for atmospheres in which there is effectively no overlapping of the carbon dioxide with water vapor, e.g., the subarctic winter atmosphere, the results are quite different.
For these situations, the introduction of carbon dioxide increases the cooling rate in the troposphere.”
Now, is this only a cooling in the upper troposphere while still increasing the downward flux at the surface? Perhaps the case of the summer subarctic troposhere has enough H2O to warm the trop. to a greater degree than the cooling in winter? (unfortunately I can’t seem to find this in the paper) I’m also wondering if there are more current LBL calculations with mo’ better approximated polar regions showing differing results?
Well, sorry for so many questions, Happy New Year to all..
Ryan O says
“..In other words, once the physics of the model are set (including the GHG warming), estimates are made of the uncertain parameters by fitting the model to the historical record.
.
Hm. Sounds a lot like curve-fitting. The models take AWG as a given, and using that given, many of the uncertain parameters (like aerosols) are back-calculated by fitting the curve.”
—
That is exactly the impression one gets when you go through all the different components of GISS’s ModelE simulation.
The GHG forcing starts immediately in 1880 and goes straight up in a log fit to GHG increases.
Then there is a straight line going down for aerosols and truly monumental negative volcanic impacts to keep the fit reasonably close (volcanoes hardly register in the temperature record at all – Pinatubo is the only one which comes close – temps actually spiked +0.5C in the year starting immediately after El Chichon.)
Boris, thanks for your comments. I’ll take them a bit at a time:
Say what? Dude, are you sure you are following the bouncing ball here? The IPCC says:
That’s three times the forcing at the poles, hardly qualifies as “slight” in my book. So let me say it again. The GHG forcing is greatest at the poles and least at the equator, solar is the reverse of that. Thus, they may have the same effect at the tropics … but they may not.
Other than a false claim (a “slight” difference?) and a brand new unsubstantiated claim, what you have written adds nothing. The new claim is that “the difference in the tropics is negligible” … the difference between what? Total solar forcing and total GHG forcing? Don’t think so. Daytime solar and daytime GHG forcing? Not hardly. What “difference” are you referring to?
And once again, the IPCC doesn’t seem to agree with you regarding different forcings having identical results. The TAR says:
Gosh, imagine that, different forcings giving different results. Yes, yes, I know they’re talking about meridional trends rather than altitudinal trends, but my point remains — there is no a priori reason to assume that different forcings will give identical results.
w.
boris, you comment on thunderstorms as follows:
This is unbearably typical of you, boris. So you’re “not at all convinced by [my] discussion”, but you don’t condescend to inform us poor mortals regarding:
1) Where you think I’m wrong, or
2) What you think I’m wrong about, or
3) What you think the right explanation might be in the parts where I’ve given the wrong explanation.
You just say, boris is not convinced.
Look, my friend, this is a scientific blog. This means that if you think someone is wrong, NOBODY GIVES A SHIT until you explain why and where they are wrong, and you cite studies or data or observations that buttress your position. I have given a cogent explanation of how thunderstorms route heat around the troposphere, by moving it upwards through a tube made of clouds that prevents it from interacting (either through radiation or mixing) with the troposphere. If you think thunderstorms don’t do that, bring it on …
As to why the data only shows amplification at medium timescales, boris, I’ve said it before, and I’ll likely say it again … for some reason, it seems to be the three hardest words for most climate scientists to say, you might practice them until you can say them as if you mean them … here come the three words explaining why the data shows that particular pattern:
I DON’T KNOW.
(I suppose it’s hard for AGW supporters to say because they’ve been arguing for years that the science is settled … and if it’s settled, well, we should know all the answers, right?)
So no, I don’t know why the data shows what it shows. I don’t think at this point anyone does. Might be that there are different phenomena that dominate the record at different timescales. Seems quite possible to me, indeed, of the possibilities, that seems most probable.
Might be that the record is too short to reveal the true complexity of the situation, amplification might return at longer timescales. Seems doubtful, but possible. Might be that the data is wrong, and that amplification in fact does occur at all timescales. Again, seems doubtful, but possible.
My point is simple. Until we can either understand what the data is doing, or show that the data is in error, arguing about fingerprints is not too meaningful …
w.
Boris:
.
If the passage on 702 is not referring to Fig. 9.1 and the associated explanation, Boris, then what, pray tell, is it referring to?
.
I will take you up on your offer to contact a lead author for clarification – with one caveat: I want to see the exact wording of the question you ask. No more word games.
.
It’s besides the point, anyway . . . my contention that if the feature shown in Fig. 9.1 c) and f) does not exist then the models are incorrect has not been refuted by anything you have said to this point. So far you’ve managed a continual series of diversions that in no way show how the IPCC models can be correct if the feature shown those graphs does not exist.
.
AFA the comment about the NH goes . . . I know. What were the SH values? No values are given, nor are any limits placed. Does solar irradiance in the SH not matter when talking about global temperatures? Hardly. Yet again another piece of data that needs to be estimated – by curve fitting or educated guesses or both.
.
AFA Stainforth’s rejection of models, again, you are not listening. No evaluation of the discarded models was done to see if they drift limit inappropriately cut out models that did not display the obviously unphysical results. None. Zero. By his own admission. So how is it that he can claim that only nonphysical simulations were rejected . . . especially in light of the fact that other researchers have produced physical cooling with those same models?
.
Did he deliberately “cook the books”? I don’t know. I don’t know him, and I can make no statements about his scientific integrity. But I can say that the results are suspect because the discarded models were simply assumed to be bad based on a drift cutoff that was not validated by a subsequent quality check.
.
If the methods are poor, doesn’t prudence dictate that you cast a wary eye at the conclusions?
.
But I note that you only cast your wary eye at the supposed “denialists.”
.
I also find it interesting that you have taken to call me a denialist when I have never once stated that AGW is not occurring. All I have stated is that the evidence I have seen so far is unconvincing. It may be occurring. It may not be.
.
That’s skepticism, not denial. Or do you want to play a word game with that, too?
boris, you give the standard explanation when you discuss the moist adiabatic lapse rate, viz:
As far as that goes, you are totally correct. However, it does not go far enough. It suffers from what I call the “simple physics” problem. Let me explain what I mean by the “simple physics” problem by way of an analogy.
Suppose we have a block of something, let’s say copper, that weighs about 170 pounds. We put one end of it in a bucket of hot water. What happens to the temperature of the other end? Well, simple physics shows that the other end of the block will warm up, the heat is transferred from the water to the substance and eventually warms the whole thing.
So, being a sort of direct fellow, I decide to test whether this is true for other substances. I try a block of steel … same outcome. A block of brass … same deal. Simple physics.
Finally, I decide to try one more test. I remember that I weigh about 170 pounds, so I stick my feet in the bucket of hot water, and I wait for my head to warm up … and I wait … and I wait …
This is what I call the “simple physics” problem, which is that when “simple physics” is applied to non-simple systems, the answer is often simply wrong.
So yes, you are correct, the lapse rate doesn’t care what warmed the surface. It may care, however, whether the warmth came at night or during the day. It may care whether the warmth is from the sun, which heats the ocean for tens of meters deep, or whether it came from GHGs, which heat only the top millimetre or so. It may care whether the heat is transferred upwards as sensible heat or latent heat. It definitely may care whether the heat is mixed upwards, or is transported upwards by thunderstorms. Or the lapse rate may be quite sensitive to the surface temperature … but only the lapse rate in the middle of the thunderstorms where the air is actually rising, and not in the bulk air around the thunderstorms where the air is sinking.
So you’ll forgive me if I am not impressed when somebody gives a “simple physics” type of explanation for complex phenomena. Anyone who thinks that the climate can be explained by simple physics has never looked at the complexity of the climate. For example, the air in between the tropical thunderstorms is sinking, not rising. Not only that, but the more heat the thunderstorms move aloft, the more cold dry air comes out at the top of the thunderstorms and settles downwards. What effect does the descending dry air have on the lapse rate? I can guarantee you one thing … you won’t get many climate answers through “simple physics”. The answer in this case involves two different emergent phenomena (cumulus and cumulonimbus) with different thresholds of emergence. Good luck with simple physics of theoretical lapse rates explaining thunderstorms and what their effect is on the lapse rate in the air between the thunderstorms …
w.
YOUR OWN DATA DOESN’T SUPPORT WHAT YOU SAY.
Your data shows an amplification at most timescales, which supports what I’ve been saying. It doesn’t at long timescales–where we have fewer and fewer complete trends to look at–and at short timescales, where I have admitted that other processes might be dominant.
But I DON’T KNOW much about thunderstorms and I have never claimed such a knowledge. You want to convince me, then you should “cite studies or data or observations that buttress your position.” You haven’t done so, and it seems bizarre that you should CHASTISE ME IN ALL CAPS for doing the same as you.
Where does the IPCC say that? (Also, you appear to have it backwards. The forcing in the subtropics is 3 times the forcing at the poles according to the quote you provide.) And you’ll want to provide a cite for your “solar has the opposite effect” claim.
Yes, regional forcings like sulfate aerosols will have differing effects from global forcings. Do you really think you’ve discovered something profound here? But anything that warms the tropical surface should also create a TT hotspot.
Well, you came in here claiming that you had decided the science of AGW was turning into a religion (this, when it’s clear you don’t understand even the basics). It’s a dead giveaway for denialism. Then you misread reports and insinuate conspiracies. You cite denialist literature. You certainly quack like a denialist, and I’ve seen a good many in my time.
I’m waiting for post #7966 to be deleted because it has the word ‘religion’ in it. 😉
Andrew ♫
Boris:
.
Apparently, you equate skepticism with denialism. Good for you. It’s exactly that attitude (irrational defiance to any challenge to AGW) that led me to make the comparison between the science of AGW and the not-to-be-mentioned word.
.
And instead of responding intelligently to the original question – that the IPCC models show a feature associated with GHG warming that, if not present, calls the models into question – you have finally resorted to name-calling and making accusations of misreading reports and spinning conspiracies . . . which is very scientific and logical.
.
Is the comparison I made really that surprising?
Andrew– I rarely delete posts. I wouldn’t delete for that. But I do think I’m going to go find a dis-em-vowler. That’s the best way to deal with these things.
Re: #7971
Lucia,
I’m just pokin’ a little fun… a little jab to distract from the Ribbon of Darkness (Reign of Global Warming Terror) hangin’ over us. We only have a few years left, by the VERY best of estimates. 😉
Andrew ♫
Can someone explain the difference between climate modeling of the future and financial modeling of the future?
Cheers…
John Sevic
The latter is met with the proper amount of skepticism.
Mike N (Comment#7955) January 2nd, 2009 at 12:32 pm
Polar amplification is not a result of increased radiative forcing at the poles compared to the tropics. In fact, radiative forcing is much lower, ~2 W/m2, at the poles for doubling CO2 compared to ~6 W/m2 in the tropics. Here is a graph from a paper by Hansen, et.al. that shows the geographical distribution of the forcing from 2x CO2. So it looks to me like increased advection of heat from the tropics to the poles is the most likely cause of polar amplification. There is already substantial advection that results in the high latitudes emitting significantly more outgoing long wave radiation than is received from incoming solar radiation with the tropics emitting significantly less OLR than incoming solar energy. The notch in the latitudinal forcing graph corresponds more or less to the intertropical convergence zone (ITCZ), I think.
As far as seasonal, the rule of thumb is that ghg’s raise the lows more than the highs, so it would be at least somewhat seasonal at high latitudes.
boris, you say:
Say what? My data shows no amplification at either short or long timescales. You and others say it should be present at all timescales. So yes, my data does support what I say, which is that we don’t really understand amplification. Other than that I have made very few claims, so I’m not sure what you mean. Please quote for me whatever it was that I said that you claim the data does not support (don’t paraphrase, quote it). Failure to produce such a quote will be deemed evidence that you are blowing more wind than a thunderstorm.
Regarding thunderstorms, well, I made no profound or mysterious claims. I was describing thunderstorms at a primary school level so anyone could follow along. However, it seems you couldn’t, so let me simplify it even further to see where your understanding stops. My claims were:
1. Warm moist surface air enters at the base of the thunderstorm.
2. It is carried aloft in the center of the thunderstorm column, where it neither mixes with nor radiates to the troposphere.
3. As the air rises, the moisture condenses, which further heats the rising column of air in the core of the thunderstorm.
4. The rising air performs work in the process, so some of the heat is turned into mechanical energy.
5. At the top of the thunderstorm, well above most of the troposphere, the rising air (which is now cool and dry) exits the thunderstorm.
6. From there, the cool dry air sinks down around the thunderstorm to replace the air sucked in at the bottom.
Please let me know which of these statements you do not understand or do not agree with, and we can go on from there. I do find it kind of astounding, however, that a man with so little knowledge of weather phenomena feels qualified to lecture us on climate …
w.
Mike N
This paper may answer some of your questions.
http://pubs.giss.nasa.gov/docs/2008/2008_Rind.pdf
In the GISS model the high latitude amplified warming seems to come from a changed albedo caues by melting ice!
WIllis,
No need to be snarky. I don’t have a problem with a basic understanding of thunderstorms. I have a problem with you thinking that thunderstorms are the reason a hotspot has not been observed or thunderstorms disprove the moist adiabatic lapse rate. Do you really think meteorologists need your expertise on thunderstorms?
boris, first you say:
Then, when I list a series of statements about thunderstorms and ask which ones you dispute or don’t understand, you say:
Then you want to berate me for not being clear about your understanding of thunderstorms … followed by misrepresenting what I said about them.
Are thunderstorms the reason a hot-spot has not been observed? I don’t know. I said about them what I said about the day/night difference between solar and GHGs. I said about them what I said about black carbon vs. GHGs. I said about them what I said about the troposphere being unstable during the day and stable at night. I said about them what I said about the other differences I listed.
What I said was that all of these provide me with separate possible theoretical reasons why not all heating would create a hotspot. They are reasons that I would not expect a hotspot from every form of heating.
Now, way up at the top of this post is a quote ascribed to you which says:
]
I asked you what evidence you have to support that. You provided nothing. I provided a host of theoretical possibilities why it might not be happening. Once again I asked for any evidence to support your position. Once again, you provided nothing. Instead, you tell me that you don’t understand thunderstorms … then you tell me you do understand thunderstorms … but you still have not provided a single scrap, a tiniest shred, a solitary word of evidence to back up your position.
Do you really think you are fooling people with all of this evasion? boris, if you don’t have any evidence to support your claim that a hotspot is expected from any warming, why not just tell us that you don’t? Your avoidance of the issue is unseemly.
Finally, you may recall that you claimed the data didn’t support what I said. Not knowing what you were talking about, I replied:
I’m still waiting … and the wind is rising …
w.
Willis, I told you I was referring to model results–that’s the denialist argument.
Then, I pointed you to info about the wet lapse rate, which you have ignored. If you actually want to learn something, then read the link I provided and, if that’s not good enough, or if you want more information about how the wet lapse rate can still be applicable in a world where thunderstorms exist, find a good meteorology text.
So read it or don’t. Learn or don’t. Whatever you do, stop whining. It’s unbecoming.
boris, having read more than my share of meteorology texts, I am well aware of the lapse rate, both wet, dry, and environmental, and both with or without thunderstorms. I am aware of the time-stepping method and the Newton-Raphson method for calculating the temperature equilibrium. I am also aware of limitations in reproducing actual temperature profiles from lapse rates, including the fact that they make no allowance for thunderstorms, clouds, descending cold dry air, or other real-life phenomena.
However, I still have not seen anything that says “here is exactly why and how the existence of various lapse rates (wet, dry, and environmental) means a TT hotspot will form from any warming”. That’s what I keep asking for, not hand-waving and repeating “lapse rate” like some magic mantra. Yes, the lapse rate exists … but it does not behave in the real world like it does on paper. For example, lapse rate calculations do not include the most important lapse rate, which is the lapse rate occurring within the thunderstorms. In fact, increased surface heating, by increasing the movement of cold dry air downwards from the troposphere, can change the lapse rate in the opposite direction. Find a good meteorology text, it will explain it to you.
So your giving me a citation to a general explanation of the lapse rate does not answer the question at all, it doesn’t support your claim that all warming will create a TT hotspot. I gave an example before about how “simple physics” often can lead us very wrong in complex systems … which you ignored. But it was specifically in response to simple explanations of the lapse rate.
For example, consider what the data shows. During 40% of the months in the data, the surface temperature and the lower tropospheric temperature go in opposite directions, with one rising and one falling. Now these are monthly averages of about 30 days of data from all around the globe, and still 40% of them show, not just small amplification, not “damplification”, but temperatures going in opposite directions.
When you can explain to me how the lapse rate calculations show that almost half of the time the two temperatures will go in opposite directions, I may listen to further lapse rate based arguments. Lapse rate calculations say that the troposphere will always warm when the surface warms … but since that is demonstrably not happening in 40% of the months, where does that leave your lapse rate argument?
And before you come back to say that local conditions may not always fit the lapse rate calculations or the like, remember that these are global averages of thirty days of data, and even at that level of collation and averaging almost half of the months don’t follow your vaunted lapse rate calculations … as I said, applying simple physics to complex systems often leads to simple errors.
Next, since you have not provided any backup for your claim that “YOUR OWN DATA DOESN’T SUPPORT WHAT YOU SAY.”, despite being asked twice, I’ll assume that it has no more substance than the rest of your claims.
Finally, “stop whining”? Say what? Dude, I’m asking for cites and references. Calling that “whining” is just more of your duck and weave to avoid admitting that you have no citations or references that specifically back up your claim that a TT hotspot will form from any warming. It only shows that you have swallowed the “lapse rate” hand-waving without actually understanding it or thinking about its limitations in the real world.
Still waiting for citations …
w.
PS – I asked for evidence. You reply:
Why should I care what the “denialist argument” is? You made a statement, that all types of warming lead to a TT hotspot. I’m just asking for evidence backing that up. Not model results. Evidence. If you have only model results and hand-waving to back it up, and you have no physical evidence and not even a clear theoretical exposition that your claim is true, just say so and we can move on.
Willis,
Looks to me that your long-range amplification data has to be an artifact. Have you thought of the surface temperature “adjustments”, with “hinges” at various points in the past. Maybe a task for Wattsup…
That’s what the models say, Willis, which has been my argument from the beginning. In addition, the models give this result in long range runs–those of nearly a century or more. The wet adiabatic theory explains why this is the case in the models, no matter the cause of the warming. You seem to want to claim that some transient response disproves the equilibrium or long term response. You should know that’s bollocks.
And, once again, your own data shows an amplification over longer time periods. It does not over the longest time periods, but look at it, a huge upturn, then a huge downturn at the end. You’re hanging your argument on that data? What would thunderstorms have to do with that?
And, yet, when more data is avaialable–Presto!–we see the amplification. How do you explain the amplification seen in your data, or shall you continue to ignore it?
At risk of unnecessarily prolonging an seemingly endless discussion…
In response to Lucia’s inline comments in green (#7887):
Exactly – my definition, which I assert is the meaning Boris, RealClimate, and Santer intended by the term, if not others (recall IPCC said nothing about “hot” in the first place), simply implies that there be amplification of tropical surface temperature changes with altitude. If the tropical surface temperature change is small, the amplified change is also small, though slightly bigger than at the surface. If the surface change is large, the amplified change is large too (and bigger than at the surface). This amplification criterion is precisely what Willis is plotting above – and one reason the one-month numbers are suspect is because the month-to-month changes tend to be much smaller, so the ratio of two small numbers can be much harder to tell than the ratio of two larger ones (I suspect the actual error bars at 1-3 months or so should be a lot larger than he shows).
On the other hand, I do find it hard to call a 0.3 degree difference “hot” by any measure, so the “hot spot” name is clearly a misnomer from the start. Why do you insist on “hotness” as the major criterion, when what everybody else seems to be measuring and talking about is amplification, not the absolute quantity of change?
I later said:
and Lucia responded:
but then further went on to claim:
You see, you are insisting on treating this “realistic” comparison as a whole, and not allowing reductionist analysis of the components that could lead to any actual understanding or logical conclusions. Together with your peculiar viewpoint on what “hot” means, this forces you to make statements like “solar forcing cannot cause a hot spot. It can barely cause a warm spot.” – but a “warm spot” is a “hot spot” according to the amplification definition which is what everybody (except Lucia) uses here. That is what we all mean, and meant from the start.
And the reason that definition is the only one that makes sense when discussing fingerprints comes back to this central question of pattern vs. amplitude. To a good approximation, we expect the response – in temperature at the surface and through the atmosphere – to a particular forcing to be close to linear in the amplitude of that forcing. That means a “hot spot” (amplification) is either part of the pattern, or not, for each forcing (no matter what the amplitude). Once you talk about fingerprints or patterns you immediately are going into a reductionist analysis mode where you factor out the actual forcing amplitude, and look at relative changes. That’s exactly what Willis did in his plot up above.
No, the IPCC didn’t do that with figure 9.1 – but any scientist looking at it and thinking of patterns naturally does the manipulation mentally, to some approximation. Since the surrounding text is mainly a discussion of patterns (not amplitudes) it would have been better to scale the individual images to a common forcing amount. But I guess they went with what they had – I’m sure they didn’t expect that image to undergo such excruciating exegesis.
And this whole section of the IPCC report is on attributing climate change, not on matching observations to known forcings. The stance of this section is in principle that the forcings are not known – some of them certainly aren’t known very well (aerosols for instance) – and to use this kind of pattern analysis to try to pin down, from observations, what kinds of forcings could be responsible for the observed changes. It is a use of fingerprints to determine the causes of changes, not some sort of “see I told you so” demonstration. Appendix 9.1A (p. 744) explicitly goes into the sort of linearizing reductionist attribution analysis I’ve been trying to get across here – that’s what “fingerprints” are useful for.
And the “hot spot”, “warm spot”, amplification, or whatever you want to call it, is not a useful discriminant in that kind of analysis at all, because the same amplification of tropical surface temperature change is there for all forcings. Which is all RealClimate, Boris, Santer, and I have been trying to say. I think this’ll be my last word on the matter here, I don’t think I can be any clearer!
Arthur: This is a question, not an argument.
.
The crux of your argument is (in essence) that the pattern of direction of change is what matters, not amplitude. Forcings other than GHG forcings result in the same pattern. When forcings other than GHG forcings are set to an appropriate magnitude (not necessarily the same magnitudes the IPCC used) a response similar to 9.1 c) and f) is seen.
.
Was this a correct summary?
Not the same pattern everywhere, but the same pattern between the tropical surface and tropical mid-troposphere, yes, once forcings are normalized to the same level (technically, identity in the pattern in the tropical surface and mid-troposphere is only there if you normalize to the same tropical surface temperature change). The patterns do differ for other latitudes and altitudes, as can be seen from the sign changes in some of the figures (yellow is positive, blue negative, and rescaling certainly can’t change a negative ratio (or any ratio, for that matter, but it’s clearest with a sign change)).
Arthur:
Why do you think this? Recognizing the actual defnition of the word “fingerprint” does not prevent anyone from restricting a mathematical analysis to those uniquely identifying fingerprints or scaling the strength of the pattern to the magnitude of the forcing.
It also doesn’t prevent any individual who writes a paper from narrowing the definition for the purpose of communicating an idea in his paper. Then, in the context of that individual paper and analysis the word can have a narrower definition. This is done all the time in language. You’ll see it in contracts, engineering articles, text books, philosophy texts etc.
But the fact that someone can narrow the definition for the purpose of a specific article doesn’t suddenly make the more general usages– which pre-existed that analysis– wrong. Notwithstanding your admiration of this particular method of using fingerprints, the word “fingerprint” is not defined that narrowly in more general contexts.
As we see in the discussion above, many people easily found quotes and exmples of climate scientists using “fingerprints” in the sense of what we see given realistic forcings. They found usages of “fingerprints” that are not uniquely discriminating. Why you belive I am the only one using it in context of realistic forcings is beyond me.
It appears you think this becuase the other usages don’t make sense to you. OK– for some reason you think that.
But other people do use fingerprint in the sense of what we see under realistic forcings. These people include climate scientists using the term in peer reviewed papers, magazine interviews etc.
As for your discussion of Appendix 9.1: That discusses a method. Go up through comments and you’ll find other methods discussed in the links and quotes. At a minimum, you will find that the temporal evolution of temperature in the 20th century is called a “fingerprint”. This method relies on what we expect to find using realistic forcings applied at the rate we think they were applied. So, you can’t just back the definition of fingerprint out of one method. If you try, you will suddenly be decreeing all climate scientists using other methods, and using the word “fingerprint” to discuss their results are wrong.
If this is all RC, Boris, Santer or you have been trying to say, then you should actually just say that and stop with the foolish decrees about how the word fingerprint or hotspot should be used.
But also, I once agains have to ask: Why are you or boris, or RC suddenly bring up this irrelevant subject changes in the middle of conversations about something else?
That is: why resort to these sorts of red herrings?
Since boris gets upset when I say “red herring”, I will explain the nature of the herring.
Real Climate’s introduced of this idea of discrimination into their criticism Douglas. However, whether or not the hot spot is a discriminant is utterly irrelevant to Douglas’s paper. Douglas’s paper is checking to see if the spot exists.
Whether the spot discriminates, it’s non-existance would be interesting and useful information. So, as a criticism of Douglas, it is utterly irrelevant to suddenly go off on a tangent discussing the issue of whether or not it could be caused by some hypothetically large amount of solar forcing.
ERGO: RED HERRING!
As for Boris– Once again Red Herring because the blog post where he jumped in to introduce the Mr. Language Person rule was not discussing whether or not the hotspot was a discriminant. SteveM was just discussing whether or not it’s there.
So, once again: red herring!
As for Santer— Actually, at least in the 17 co-author paper, he says nothing about the hotspot not being a discriminant. So, you might want to take him off the list of people who want to make sure we know it’s not a discriminant.
But it’s perfectly relevant to what Douglass said about his paper in a SEPP press release. Douglass claimed it was a discriminant, and that now we don’t have to worry about CO2. So your claim that RC invoked a red herring is not accurate.
Arthur: Okay. You answered the question part, now here comes the argument part. Same conclusion as Lucia, different tact.
.
As you say, analyzing a response independently of the magnitude of the response is important in determining the underlying physics. No argument. Same principle applies in engineering and I use it all the time.
.
However, that’s not what we’re talking about.
.
The IPCC uses a set of models and model runs to make a prediction about the future state of the climate. This prediction is not just the direction of from the present change; it also includes the amplitude of the change.
.
And if the amplitude of the predicted change in the troposphere is not reflected in observations then it means the underlying physics in the models is wrong. Period. And if the underlying physics is wrong, then the predictions about surface temperatures may also be wrong (after all – surface temperatures are linked in some fashion to troposphere temperatures).
.
Nothing you or Boris has said contradicts this. You say that it does, but nothing you say actually refutes this. We’re not talking about what the best method is to study the physics; we’re talking about how closely a set of models matches observation. You cannot validate a theory with direction of change alone. You also need the magnitude. If either does not match observation, then the theory is wrong.
.
Not only that, I’ve stated (and Lucia’s stated) repeatedly that it is the whole pattern we are referring to when we are talking about the supposed “fingerprint”. In your own words:
.
.
So by your own admission the patterns for other forcings may not match Fig 9.1 c) and f) in its entirety. That makes the argument about the definition of fingerprint utterly irrelevant. If other forcings cannot duplicate Fig. 9.1 c) or f) in its entirety, then even using your definitions, it is a fingerprint because it is unique to GHG.
.
Lastly, there is the whole context of “fingerprint”. If the context includes models not examined by the IPCC, then you are free to use your definition. No argument. But if the context is the IPCC claim (which it clearly is . . . from Lucia’s first point onwards) then you don’t get to restate the definition. The IPCC is clearly stating that within the set of models they examined the pattern in Fig. 9.1 c) and f) is unique to GHG warming.
.
Nothing you or Boris have said contradicts this.
.
In other words, if observations do not match the predictions shown in Fig. 9.1 c) or f) then the models are wrong. Furthermore, even if the same pattern of direction of change is found, if the amplitude is less than that shown, then the models are also wrong. How willing do you think the world would be to pursue some sort of carbon-neutral goal if the amplitude of the temperature change by 2100 is actually 0.3 C?
Boris: A tropical tropospheric hotspot is expected from any warming
As we’ve found that there is not, in fact, a TT hotspot, should we then conclude that there hasn’t been any warming?
Boris–
If RC wants to criticize a press release, they should quote that which they are criticizing and state what is wrong more specifically. But if we examine the press release, the issue of uniqueness of the hotspot to ghg’s rather than solar is utterly irrelevant. So, RC is still introducing a red herring.
Here’s a link to a nearly unvisted SEPP blog linked by RC.
Let’s for example look at what Christy wrote:
There is nothing in this that suggests he is saying the warming aloft is specific to ghg’s. He’s just saying models predict it, they predict it based on the level of ghg forcings that we have experienced, and we are not finding it.
Douglas says this:
Douglas is using “fingerprint” the way it has been used by many before him: A characteristic pattern predicted for a specified forcing. He makes no claim of uniqueness. He simply observes it’s not there.
If RC was criticizing the press release by introducing this whole non-uniqueness because of solar issue that was a red herring. The issue of what models might also predict if driven by hypothetical solar forcing is irrelevant to the points Christy or Douglas make in their paper and in their press release.
(Singer says a bunch of stuff never mentioning the hot spot or models.)
Had RC quoted instead of using their usual “here’s a link we suspect you won’t click” method of making arguments, people would easily see their discussion was irrelevant to what Douglas said.
boris, many thanks for your clear answers. You say that your statement is based solely and completely on model runs.
Which is fine, but it needs to be qualified as such. One of the problems in the field of climate science is that modelers and AGW advocates seem to forget that they are working with a model. They say things like
when what they mean is:
These are very different things.
Next, you ask about the data. My problem is that I don’t understand the data … and near as I can tell, neither does anyone else. Why should there be amplification in the middle length time scales, and no amplification at either end of the scale? I haven’t a clue. I was very surprised by these, in particular by the lack of amplification at the shorter scales. Why should the tropical surface warm for two months and the troposphere cool over those same two months?
You ask if I shall “continue to ignore” the amplification in the data … ignore it? You do realize that I’m the one (and the only one so far) who independently ran the numbers? I’m the reason you can say that the amplification is in the mid-range data? I have never ignored that, I’m the one who established that for the purposes of this discussion. How could I ignore it?
Nor do I ignore, or try to hand-wave away, the rest of the dataset, the short and long term, when there is no amplification. I have said clearly and frankly that I don’t understand those … but that doesn’t mean that we can safely ignore them or assume that they are errors.
Yes, the wet adiabatic theory explains what happens in the models, as you point out. However, please do not mistake that for reality. Whether it explains amplification in the real world, or is only part of the explanation, remains to be seen.
One possibility which I have not seen discussed in print is this. The tropical surface is at about 30°C, whereas the lower troposphere is at about freezing (0°C). Here is a thought experiment.
Assume that the surface and the troposphere are in radiation balance. Now, assume that there is an increase in forcing of say 1 W/m2 from the troposphere, which is matched by an increase in upwards surface radiation of 1 W/m2. So once again, they are in radiation balance.
Now, let’s examine what has happened to the temperature. Neglecting emissivity, the Stefan-Bolzman relationship says:
30°C = 478.87 W/m2
0°C = 315.64 W/m2
Adding one W/m2 to each one gives us
479.87 W/m2 = 30.16°C, a temperature rise of 0.16°C at the surface
316.64 W/m2 = 0.22°C, a temperature rise of 0.22°C at the surface.
In other words, a simple radiative balance model gives us an “amplification” of about 1.4 without any change in the lapse rate … why? Well, because although energy is conserved, temperature is not conserved, so a 1 W/m2 change in the balance gives a different temperature change depending on the starting temperature.
Now, is this the reason for the amplification seen at the middle scales? Heck, boris, I don’t know, and as far as I can tell, nobody knows. Thats the problem with building models of a system we don’t understand, whether those models are as simple as my basic radiation/convection balance model, or as complex as GCMs … we can’t tell whether we’re modeling it correctly because we don’t know what the expected behavior looks like. We could be getting the right answer for the right reasons, or the right answer for the wrong reasons, or the wrong answer for the right reasons, or the wrong answer for the wrong reasons … and because of our lack of understanding of the climate, we can’t tell one of these results from the others.
For example, the PDO (Pacific Decadal Oscillation) has been in the warm phase since about 1976. It has recently gone into the cool phase. What effect will this have on tropical amplification data from time spans which start in the warm phase and end in the cool phase (e.g., the longest time span of our amplification data)?
Well … not to put too fine a point on it, we don’t have any solid theoretical basis to even guess what the answer will be. I doubt, however, that it will have no effect.
My main point in this is simple — for many of these questions, we have insufficient theoretical understanding to say what we should expect to find. We don’t know if TT hotspots will form from any and all types of warming. We don’t know what effect the chance in the PDO will have on the vertical temperature structure of the tropical troposphere.
Now, I wrote my first computer program in 1963, on a computer that took up a whole room, and have been writing programs ever since. I have since written a host of computer models for different purposes, from business analysis to pump design to a simple global radiative-convective balance climate model. One thing I have learned for sure in that process is that model results are not reality. They are not evidence.
For example, I once wrote a killer model (in Stella, an excellent modeling environment) of a business, which said that the business would make scads of money. Is that evidence that the business would make money?
I wish it had been evidence … but in the event … no. Although the model was excellent at matching the past (just like our current climate models), it proved to be far less than stellar in predicting the future (also like our current climate models). The map is not the territory, the model is not the reality, and neither the map nor the model provide evidence about what is happening in the inconvenient, messy, and terribly complex real world.
For example, the GCMs do not model the thunderstorms. However, in the real world, it is the tropical thunderstorms which drive the atmospheric circulation. Thunderstorms also allow heat to pass through the troposphere without warming it … but in the models, although the thunderstorms can be partially “parameterized” to simulate them driving the circulation, there is no physical mechanism in the models whereby that process (allowing heat through the troposphere without warming) can be represented. So it is ignored. (And as an aside, all of them wildly underestimate the atmospheric circulation of energy … but I digress.)
That’s a huge hole in the models, and it explains why they show a TT hotspot from any heating. They lack a mechanism to model crucial real world phenomena, so every kind of heating may end up having the same effect … but in the real world, there’s a host of other complex, hard-to-model, poorly understood phenomena at play, from thunderstorms to the PDO.
So to sum up:
1) The observations (using either RSS or UAH satellite data) shows no amplification at either short or long timescales. We don’t know why.
2) The models (presumably, I haven’t been able to find the data to test this one, still looking, any hints accepted) are said to show amplification at all timescales and from all types of heating. Some people think they know why.
Given the simplistic and crude state of the models, my suspicions fall on them. It is certainly possible that the data is in error, but the appearance of the same pattern from both the UAH and the RSS satellite data reduces the odds.
Now, in any other field of science or industry, the modelers would suspect their models. But in climate science this is never, ever done. The AGW supporters have painted themselves into a corner with the absurd idea that “the science is settled” … settled? We don’t even understand the climate system, and the science is settled? Please …
But if you hold that the science is settled, then the models must be right … riiiight.
So as I said up-thread a ways, many climate scientists (particularly modelers) should practice these three simple words over and over:
WE DON’T KNOW
This reluctance to admit that climate science is a baby science in its infancy, this ridiculous claim that we understand climate and that the science is settled, is stifling the field. For example, you say:
I should know that’s bollocks? On what basis? Before we can “disprove” the equilibrium or the long term response, we have to know what the “equilibrium” or the “long term response” are … and we don’t, that’s the whole point. You view the science of the “equilibrium” and the “long term response” as being settled, when they are nothing of the sort. We don’t even know if an “equilibrium” exists, much less why or what it might look like. Does the “long term response” include the PDO, or is that just long-term random variation? We don’t know. Do the models accurately portray the “long term response”? They might, but we don’t know. Say it again and again, “we don’t know”, you’ll get used to it in time.
boris, I don’t “want to claim” anything except that we don’t understand the data. I’m not trying to “disprove” anything. I’m simply pointing out that we don’t understand the data, and that our models are absurdly simplistic. That’s all. I have proposed a variety of reasons that the data might look like it does … but I make no claims for them as being other than possible explanations for why the data looks like it does.
My best to you, boris, and my apologies for when I have gone over the line in this discussion. I have faced this mindset of “model = reality” over and over. I get tired of people saying “we know X will happen” when what they really mean is “my favorite model says X will happen, and we have no observational evidence to back it up.” So I tend to get curt when people claim we understand the “equilibrium” or the “long term response” when in fact we don’t have a clue … my bad.
w.
Lucia, also, in the article, Douglas makes it clear that the analysis and conclusions are restricted to the 20CEN forcing:
.
He also states that the purpose is to test the models against observations, not against various permutations of the forcings (that may or may not be physical):
.
He additionally states that it is not just the direction of change that he is analyzing, but also specifically the magnitude of the change:
.
In other words, the context of the Douglas’ (and the other authors’) statements are clear from the article. Boris and Arthur’s arguments are merely an attempt to remove the context from the statements in order for them to be able to claim Douglas is wrong without needing to present any proof to the contrary.
Willis Eschenbach (Comment#8102) January 5th, 2009 at 6:19 pm ,
It’s acceptable to ignore emissivity of the surface because it’s most likely nearly 1.0. It is most definitely not acceptable to ignore the emissivity of the atmosphere, which is anything but black or gray. The emissivity/absorptivity of a given volume of the atmosphere varies strongly with both wavelength and altitude. Even for a given mass of the atmosphere emissivity varies strongly with altitude because the scale height of water vapor is about 1/4 that of the non-condensing gases and the line width varies with pressure and temperature. Therefore the S-B equation does not apply to the atmosphere. You have to multiply the emissivity at a given wavelength by the Planck function for that wavelength and temperature and integrate over the full wavelength range to calculate actual emission.
I agree that the GCM’s do not model thunderstorms. IIRC, weather models on a finer grid don’t do a great job either. However, your conclusions do not necessarily follow from this. Perhaps this deserves its own thread, but what is your evidence that thunderstorms in particular drive the atmospheric circulation rather than being near the high end of a (power law) scaling of upward vertical convection events in general (tropical cyclones being at the top) in an atmosphere that over a sufficiently large scale (10 to 100 km) is in hydrostatic equilibrium? What is your evidence that convection in general, whether vertical or horizontal, is wildly underestimated in the models? AFAIK, Hadley-Walker cells are, for example, features of the output of GCM’s. This is not to say that the models get convection right. They almost certainly don’t, but wildly off?
Let me define hydrostatic equilibrium so you know what I’m talking about. Hydrostatic equilibrium means that the center of mass of a sufficiently large block of atmosphere does not move up or down except for small changes due to pressure or temperature. That is, any upward convection is balanced (to a first approximation) by downward convection within the same block. Free horizontal movement does not violate hydrostatic equilibrium because there is no change in gravitational potential energy for a horizontal move.
Against my better judgment – a necessary response…
Lucia quotes Douglass (Lucia – 2 s’s!):
Lucia – I can’t imagine how you can defend this statement! Douglass explicitly states there is climate warming (his paper indeed saw warming, just not enhanced warming). He states that his conclusions imply CO2 and other GHG’s make only a negligible contribution to that warming. Since the averaged surface temperature response to forcing is close to independent of the type of forcing, that means Douglass is claiming the forcing from CO2 and other GHG’s is negligible compared to some other forcing. He does not say what that other forcing is, but there is no way to have climate warming without some cause.
Therefore, implicit in Douglass’ statement on the matter, is:
(a) the 20th century forcing from human GHG’s is negligible
(b) some other forcing is much larger, and has caused the recent warming.
Therefore, Realclimate’s analysis of the response to other possible (large or small) forcings in this context is not in any way a red herring, but absolutely apropos.
Furthermore, Douglass is far from alone in making this sort of bold statement about the lack of the hot spot implying GHG forcing is small. Monckton’s “Physics and Society” paper was similarly explicit on it, quoting Douglass:
And now we have Joane Nova’s “Skeptic’s Handbook”, which states boldly in point 1:
Douglass’ grossly misleading statement is behind all this nonsense, none of this is at all defensible.
“Something else caused the warming” indeed. Lucia, please, come to your senses on this – you are being badly taken in by these folks. Realclimate was absolutely correct in the way they addressed this issue. No other cause of warming can explain the warming of the surface without creating the hotspot, so the conclusion that “something else caused the warming” is nonsense.
Lucia –
I don’t see how you can read that SEPP press release ( http://science-sepp.blogspot.com/2007/12/press-release-dec-10-2007.html ) and come to the conclusion that you did! Its very title is “Climate warming is naturally caused and shows no human influence”. Its first sentence says: “Climate scientists at the University of Rochester, the University of Alabama, and the University of Virginia report that observed patterns of temperature changes (‘fingerprints’) over the last thirty years are not in accord with what greenhouse models predict and can better be explained by natural factors, such as solar variability. ”
If they wanted to just say there is a discrepancy between global climate models and reality in their general pattern predicted in a warming world, why on earth did they use the term “greenhouse models” to describe them (a term not used in the scientific literature, in fact a term basically devoid of any real meaning)? And, why did they suggest that this somehow shows that the warming that has been observed can better be explained by solar variability when that is not the case? And, how did they come to the conclusion that the observed warming is naturally-caused based on their not seeing the amplification?
If they had simply said, “There seems to be a discrepancy between what climate models predict for the pattern warming in the tropical atmosphere and what has been observed this suggests that they might be missing an important piece of physics in the tropical atmosphere and makes their predictions of future warming questionable”, then I wouldn’t have a particular beef with this and we could move on to the other ways in which the Douglass et al paper was wrong (like not understanding the appropriate use of standard error vs standard deviation). But, in fact, they went well beyond that.
And, look at what Douglass himself said in that press release: “The observed pattern of warming, comparing surface and atmospheric temperature trends, does not show the characteristic fingerprint associated with greenhouse warming. The inescapable conclusion is that the human contribution is not significant and that observed increases in carbon dioxide and other greenhouse gases make only a negligible contribution to climate warming.” This statement is completely and utterly wrong and is exactly the kind of completely fallacious statement that Real Climate and Arthur Smith and Boris and I are laboring to correct.
The whole point of their press release is that their research somehow shows that the warming is not due to greenhouse gases when in fact, even leaving aside that they screwed up their analysis and that the observational data has known artifacts in it, this is not true at all. The most it would show, if it were not hopelessly flawed in those other ways, is that something is missing in the physics currently in the climate models that are used to understand past climatic events and predict future climate response.
As it stands, all that we can really say is that there are discrepancies between the general predictions for a pattern of warming in the tropical troposphere (irregardless of cause) in the climate models and the observations … but that there are known artifacts in the observational data that make us unable at this time to conclude whether or not there is actually any problem with the models.
[Note added: I see that Arthur has submitted a post at the same time making similar points to what I make here.]
Arthur, that’s B.S. To make that claim, you have to show, for example, that if you increase other forcings to the level required to explain the current warming that a hot spot also appears. You can’t just arbitrarily raise it to some level, show that a hot spot appears, and then make the claim that “something else caused the warming = nonsense”.
.
SHOW that the level of solar forcing required to explain the current surface temperatures also simultaneously creates the T.T. hotspot and you have an argument.
.
Until then, you have merely asserted that (somewhat) equal magnitudes of different forcings produce the same feature in the troposphere . . . and all you have is one model with an unphysical level of solar forcing to back that up. To use your own words to me, all you have is a bare assertion fallacy.
Ryan O
“Not only that, I’ve stated (and Lucia’s stated) repeatedly that it is the whole pattern we are referring to when we are talking about the supposed “fingerprintâ€.”
This isn’t exactly correct.
I attempted to describe to Lucia that the IPCC definition of fingerprint implied the whole pattern(way up near the beginning), and Lucia argued that it could also mean parts of the pattern, she was actually arguing that the TT Hotspot could be considered a fingerprint on it’s own.
Strange how things go round in circles.
Nathan: Haha. Okay, I missed that one.
.
Well, for my part, I was arguing about the WHOLE pattern.
Ryan O
I actually agree that is what the IPCC meant for their figures.
I think our disagreement is about whether it is important the observations match the ‘fingerprint’ completely. I don’t see it as a huge issue that they don’t.
I think it should be explored what it means that they don’t. What does it mean to the system if the TT hotspot is missing. Does this mean that convection is more ‘lateral’? Perhaps the TT hotspot is due to vertical convection, and the lack of hotspot means that heat is actually being transported north and south more than the models predict. This could actually mean that the models are underestimating the effects of AGW (is this what Boris said earlier?)
If Douglass is using a scientific definition of fingerprint, then he should probably make that clear in a press release intended for lay persons. Unless he wants to misrepresent–well, hey, bingo!
The second part of Douglass’ quote is pure stupid or a lie.
And it was not a red herring because a lay person reading the Douglass press release (or any of Monckton’s nonsense on this issue) would think that a fingerprint is an identifying feature. I guess they are morans.
It disturbs me that you have problems with RC explaining things so someone can understand and don’t have a problem with Douglass’ statement which is either ignorant or a lie. You really want to side with the denialists on this that bad?
Joel and Arthur said it better.
Willis, I appreciate your response. Sorry if I got a little ruffled too. But it would really be boring if we didn’t give a damn about this stuff. Imagine this thread with only 2 responses.
Now, before I get cranky again, I’m going to bed.
Nathan: And I agree with what you’re saying. IF (intentional emphasis) the pattern is not observed, then the physics in the models is wrong . . . but that’s not the endgame . . . all that means is that we should admit it and try to figure out what is wrong.
.
By the same token, if Douglass’ analysis is proven wrong, naysayers should admit that, as well.
Arthur–
Define “defend”.
Boris suggested that Christy and Douglas made claims about what would happen under this hypothetical solar forcing scenarios.
I didn’t say I agreed with the statement, nor that everything in the statement is correct nor that Christy’s conclusion is correct. I am simply pointing out that it does not contain the specific mistake Boris claims RC was criticizing.
If people want to criticize Douglas and Christy for what they actually said, that’s fine. But suggesting in those statements, Douglas or Christy said hotspot couldn’t be caused by hypothetical levels of solar forcing is simply incorrect. They don’t say anything of the sort.
For some reason, people want to manufacture these false errors and pin them on D&C. What’s the point of that? If you have the high ground on the science stick with the things they actually say wrong.
If you want to criticize statements like: “Something else caused the warming” do so directly. People need to stop trying to “prove” the statement is wrong be redefning “fingerprint”, pretending Douglas and Cristy are making positive statements about what would happen if we has hypothetically large amounts of solar forcing etc.
The issue related to hypothetical solar forcing has absolutely nothing to do with what’s wrong with Christy and Douglas’s statements.
The reason the solar forcing issue and the possible non-uniqueness of any predicted hotspot introduced by RC is a Red Herring is Douglas’s argument has absolutely nothing to do with the distinctiveness of the hotspot.
Near the ed of your comment, you now seem to be arguing that the reason one should say something is wrong is that a statement appears in a skeptic’s handbook. Well… newsflash: if it said 2+2 =4 in both the Devils math primer and God’s math primer, I’d say they are both right. If either says 1+4=3, I’d say that’s wrong. I don’t care who says it.
In response to your statement
I agree that failing to find the hotspot will not prove something else caused the warming. I have said so repeatedly above. I have not gone back on that.
So far the hotspot has failed to reveal itself. Should it fail to reveal itself, this will suggest the models have some problem– but not that ghg aren’t a problem. It would just leave doubt about our ability to predict climate response.
I think this needs to be said directly. It’s easy enough to state.
But, for some reason, people seem to want to take some bizarre tack related to redefining fingerprints, and talking about hypothetical solar forcing and non-uniqueness of hotspot predictions (if we includes hypothetical forcings.)
Trying to pretend that Christy and Douglas made some incorrect statements about what models would predict if hypothetically large solar forcing caused the warming when discussing Christy and Doglass’s papers and statements is a red herring. The don’t make this incorrect statements because they don’t discuss what would happen in this hypothetical situation at all.
Ryan O
” IF (intentional emphasis) the pattern is not observed, then the physics in the models is wrong . . . but that’s not the endgame . . . all that means is that we should admit it and try to figure out what is wrong.”
Absolutely. I would think though that this is all in progress, too. I don’t think that the models that featured in the last IPCC report are still being used as they were. Also note that trying to get precise models is a fools errand, they don’t and will never exist. It’s about getting models that give a reasonable estimation. I’ll leave it to others to decide what is reasonable.
Joel–
If your question is why Fred Singer calls the models “greenhouse models” rather than “climate models”, my answer is “beats me”. That Fred Singer writes polemics is undeniable. But if you strip the “greenhouse” out of this:
“Climate scientists at the University of Rochester, the University of Alabama, and the University of Virginia report that observed patterns of temperature changes (‘fingerprints’) over the last thirty years are not in accord with what
greenhouseclimate models predict”That bit is fine.
I disagree with their addition : “and can better be explained by natural factors, such as solar variability. ”
The work in Douglas work didn’t tell us anything about how anything can be better explained.
So, if you want to criticize Douglas et al for over reaching in an article authored by Singer. GREAT!
But, trying to do so by deciding they claimed that models don’t predict hotspot when forced using hypothetically high levels of solar is wrong. The mere appearance of “natural variability such as solar” does not amount to making any such claim. For some mysterious reason, you are reading that in.
In the rest of your comment: You are now focusing on statements you consider generally incorrect, but which are not specifically related to Douglas making claims about the genesis of the hotspot, or anyone making claims about the genesis of the hotspot. So, we are now moving onto a slightly different issue.
It is true that even if we find no hotspot, we can’t rule out ghg’s as the source of surface warming. But, that’s not the case RC made. They brought up this utterly irrelevant issue about non-uniqueness of predicted hotspots under hypothetical levels of forcing.
I agree with you that failure to detect the hotspot doesn’t mean it’s not there. But the fact that the hotspot has not yet been detected will be discussed until the hotspot is detected.
The fact that RC decided to try to divert attention from this by introducing a Red Herring has made things worse because their little excursion into hypothetical solar land diverte attention from the truth which is: The hot spot may well be there. The data aren’t good enough for us to tell.
I think the hotspot is likely to appear. But meanwhile, like it or not, if the hotspot doesn’t appear, that suggests a problem with models. It won’t prove ghg’s don’t cause warming– Douglas
Lucia:
You say,
On this, we agree. However, I do not see how you can then say that Christy and Douglass have said nothing wrong. The clear message in their statements is that the lack of a hotspot shows the warming is not due to GHGs. I suppose if you do backwards-somersaults and parse their statements to death, you can say things like “Well, they never specifically claim that models driven by solar forcing wouldn’t give the same hotspot.” However, while this may be technically true, it really seems to me to be a desperate argument to ignore the fact that the entire press release is predicated on the idea that their conclusions somehow show that the warming is not due to GHGs but is instead due to natural causes.
It seems like you are basically trying to get them “off” on a technicality…and it is worse than just trying to get them off, since we are not trying to find them guilty of something but are merely trying to set the record straight by arguing against an interpretation that a reader might make when they read their press release. (I use the word “interpretation” rather than “misinterpretation” since I think it is what any reader who read it would conclude and what they intended any reader to conclude. If you can find ANY reader who would read what they wrote in that press release and then come to the conclusion that you stated above [in what I quoted] rather than the conclusion that I claim they wanted to be concluded, then I would be quite impressed! I don’t think it is even ambiguous enough that it could be interpreted one way or the other [which would still justify setting the record straight, by the way]…Rather, it is written so that the reader will be led to the wrong conclusion.)
That is why it is very important to set the record straight by pointing out that, even if they hadn’t f-cked up their analysis royally and refused to entertain the idea that the data may have systematic problems (which it is in fact known to have), the lack of a “hotspot” would in fact simply imply what you said above and would not imply what they claim it implies.
Boris–
What do you mean by “siding”?
I discussed the year old RC post because their oddball red herring argument was regurgitated in a comment at CA. The red-herring has legs. If people stopped regurgitating the meme introduced at RC, I wouldn’t have discussed it.
In contrast, on the rare occasions people repeat what Douglas said– that the failure to see the hotspot means most the warming is natural– I disagree with them. I say the lack of the hotspot doesn’t mean that. It only means there are problems with the models ability to predict some features of climate. I haven’t experienced people providing links to longwinded blog posts with figures defending the incorrect claim Douglas made.
In anycase, I wouldn’t spend much time criticizing a single sentence in a paragraph in a press-release. These things generally suffer from serious snippage. In contrast, the team at RC wrote a rather lenghthy post and have no particular space constraints.
For what it’s worth, in newarticles that snip less of Christie’s quote, he says more or less what I say. The lack of a hotspot would point to problems with models. Or at least that seem to be the gist of what he says here.
I think he goes a bit too far too– I’d say disagreement between models and observations would tell us models have difficulties simulating reality. But disagreement wouldn’t necessarily mean they over-estimate warming. We could ultimately discover that whatever physics they miss could mean the models as currently constituted under-estimate warming.
Joel–
I didn’t say they said nothing wrong. I said they did not make the specific wrong claim Boris accused the of. These are different things.
What I’m telling you is: This is not the same as making the claim that hypothetically, solar forcing of sufficient magnitude would not cause a hotspot. Nor is it making the claim that models forced by high levels of solar energy don’t produce a hotspot.
I’m not trying to get Douglas “off” on anything. I think
A) You can correctly say that Douglas is wrong when they say that failure to find the hotspot doesn’t mean ghg’s don’t cause forcing.
B) One can’t correctly say they made any claim about whether or not a hotspot appears in models driven by hypothetically large solar forcing.
But, for some reason the RC post, and Boris, and Deep Climate and many others appear to want to repeat incorrect claim “B”. In the process, they forget to point out Douglas’s real mistakes.
Yes. Douglas was wrong with claim “A”.
And it’s also why it’s important for people to stop wasting time trying to accuse Douglas of making claim B. Douglas didn’t make that claim.
Joel, in the same paragraph in which you write:
“since we are not trying to find them guilty of something but are merely trying to set the record straight”,
you then write:
“Rather, it is written so that the reader will be led to the wrong conclusion.”
That by any reasonable standard is an accusation.
Willis Eschenbach (Comment#8102) January 5th, 2009 at 6:19 pm ,
That’s called adiabatic expansion and is a fundamental assumption of Physical Meteorology for any vertical air movement. A kg of air moves up a kilometer, gains 9,800 Joules of gravitational potential energy and loses 9,800 Joules of kinetic energy by expansion because it is assumed that there is no heat transfer from or to the surrounding air. For dry air that means a loss in temperature of the packet of about 9.8 K. For moist air, the loss of temperature is lower because the heat capacity of moist air is higher than for dry air. So the temperature of the rising air column in the storm cell is higher than the surrounding air at any given altitude, but there is little heat transfer except where turbulence at the boundary causes mixing.
If the condensed water droplets in the packet remain suspended, then the packet could be returned to ground level at the same temperature as it started as it will follow the same moist adiabat on the way down. However, if the water droplets fall to the ground and the packet is then returned, it will follow a dry(er) adiabat at a higher lapse rate and have a much higher temperature at the surface. This is the origin of hot mountain winds like the Santa Ana in Southern California.
Now here’s where it might get interesting. If the dry air returns to the surface slowly enough, it might have time to cool by radiation to space. While CO2 causes warming at the tropopause because the temperatures above and below are higher, it causes radiative cooling above and below the tropopause. Could increased CO2 cause the air that returns to the surface in a Hadley cell to be cooler than at lower CO2? Dunno. I don’t have even a one dimensional radiative convective model to play with.
Arthur, you say above:
“Since the averaged surface temperature response to forcing is close to independent of the type of forcing …”
Huh? Cite, please.
While waiting for that, consider that sunlight heats the ocean for a depth of tens of meters, while longwave IR (also called DLR, downwelling longwave radiation) from GHGs is completely absorbed in the top few millimetres.
The DLR energy is absorbed by and heats the top skin layer of the ocean There it is free to radiate and evaporate, and thus to cool. The solar energy, on the other hand, gets absorbed and trapped in the deeper layers.
Since DLR energy heats the skin layer, this increases evaporation. So a larger percentage of DLR energy is lost by evaporation, compared with evaporation from the solar heat absorbed by the deeper water. This effect is amplified by the small amount of water heated. One watt per square metre of sunlight is heating tonnes of water per square metre. One watt of DLR energy is heating kilos of water. This gives a much greater surface temperature, which increases both radiation (temperature to the forth power) and evaporation.
Surely the temperature response of the surface of the ocean, 70% of the earth, is different in those two cases?
This is particularly true because the stratification of the ocean is different during the day and the night. But it is opposite to the atmospheric stratification. During the day, the ocean thermally stratifies, hottest on top. At night, surface waters start to cool. At some point during the night, the surface waters are cool enough to sink, starting the nocturnal overturning of the upper mixed layer.
Only the DLR energy is present at night. During the night, the effect of the DLR warming of the surface is to delay the onset of overturning, and to make it overturn more slowly once overturning begins. By slowing the cooling of the top skin layer throughout the night, it makes the overturning start later in the night and go slower. This, of course, affects the surface temperature.
Remember that the sun has absolutely no effect on the rate of nocturnal oceanic overturning. So in this case, a 1 W/m2 change in GHG forcing is guaranteed to have a different affect on surface temperature compared with a 1 W/m2 change in solar forcing …
In other words … the ocean surface temperature response to downwelling longwave radiation is very, very different from the ocean surface temperature response to sunlight.
Best regards to all,
w.
Lucia. The modelers say that any increase in CO2 will produce an increased temperature anomaly, but the reality is that although there has been an increase in CO2, there has been no increase in said anomaly.
That is the reason Real Climate, Boris, and the Warmers get their knickers in a twist. That is the reason why their unheralded rebuttal was made, of what wasn’t explicit in Douglas et al.
By indirect means Douglas proved the Models wrong. Couldn’t be allowed by the many headed warmer meme hydra.
PS. The labours of Heracles describe the battle against the Warmers:
“Heracles covered his mouth and nose with a cloth to protect himself from the poisonous fumes from the hydra’s eyes and fired flaming arrows into its lair to draw it out. Upon cutting off each of its heads he found that two grew back, an expression of the hopelessness of such a struggle for any but Heracles. So Heracles cut off each head and Iolaus burned the open stump leaving the Hydra dead. Its last head was invincible to any weapon so Heracles ripped it off with his hands.”
Should be Tropospheric Temperature anomaly.
DeWitt, thank you for your most interesting post. You said:
You are 100% correct. In fact, it’s worse than that, because emissivity (like temperature) is an intrinsic rather than an extrinsic variable so it doesn’t have a clearly defined “average”. But wait … there’s even worse … much of the “greenhouse effect” is caused by cloud layers, which have an effective emissivity of 1 … and global average cloud cover is about 69% …
So pick a number to represent the average. If we say clear sky emissivity is 0.5 and clouds are 1 and cloud cover is 69%, that gives us an average of 0.85. But in fact, curiously, the lower the emissivity the larger the effect. I can show it mathematically if you’d like.
We are looking at the ratio ∆Ta/∆Ts, where “∆” is the “change in” operator, Ta is temperature at altitude and Ts is surface temperature. This is the “amplification factor”, the change in atmospheric temperature corresponding with a change in surface temperature.
Transforming the temperatures into radiation temperatures by Stefan-Bolzmann gives us
[ (∆Wa/stefan*Ea) ^ 1/4 ] / [ (∆Ws/stefan*Es) ^ 1/4 ]
where Wa and Ws are the radiation (W/m2) and Ea and Es are the atmospheric and surface emissivities.
Taking the fourth power and cancelling out the stefans gives us
(∆Wa/Ea)/(∆Ws/Es)
From that you can see that as atmospheric emissivity (Ea) decreases, the numerator increases and the “amplification” is greater.
So regardless of emissivity, one watt/m2 radiated from the surface causes a larger temperature change in the atmosphere than at the surface, because their temperatures are different.
It likely does deserve it’s own thread, but here we are. Regarding “wildly off” on the Hadley cell transport, I’d have to look up the reference, but it’s an analysis of the GCMs used in the IPCC TAR. They found the GCMs underestimated the amount of energy transported poleward by the global circulation by a factor of 4.
I did not say that vertical convection is “wildly underestimated”. It is wrongly represented. I said that the models contain no magical tube like a thunderstorm. Inside the thunderstorm, air rises without regard to or interaction with the troposphere. It does not mix with the troposphere. It does not radiate to or absorb from the troposphere. It is insulated from the troposphere. A thunderstorm is an insulated heat pipe that moves air and energy from the surface to the top without interacting with the middle.
GCMs, as far as I am aware, do not contain a physical mechanism that allows air to magically pass from the surface to the tropopause without interacting with the troposphere.
Tropical thunderstorms are generally accepted to be the major driver of the Hadley-Walker cells. Living in the deep tropics of the Pacific (9° South), I watch them form daily. You are right that they are part of a continuum of emergent phenomena which function as atmospheric heat engines. These have the common characteristics of cooling the surface and funnelling warm air upwards. They form spontaneously as discrete entities. They have a definite lifetime, with an initiation at a certain temperature-related threshold, followed by growth, maturity, decay, and dissipation. The smallest of these are the “dust devils”, which can become fairly large. The biggest are cyclones. The most numerous are thunderstorms, which do the heavy lifting for the Hadley cells. Note that these are very different from simple vertical convection as is modeled by the GCMs.
I do understand what you mean. It is the reason that after a parcel of surface air takes a trip upwards through the whirling core of a thunderstorm, the air emerges at altitude and immediately begins to descend. The upwards flow is matched by the surrounding downwards flow.
This is one reason that bulk measurements (which are measurements of the masses of air between the thunderstorms, not inside the thunderstorms) of variables such as tropospheric temperature do not behave as expected. The warm, moist air is concentrated and moved upwards in the cores of the thunderstorms. The more warm moist air that sneaks past the troposphere, hidden in the thunderstorm cores, the more cool, dry air comes out of the top of the thunderstorm. The cool dry air which exits at the top to sink downwards is the air in between the thunderstorms, where the bulk measurements are taken. Bulk measurements of that descending air temperature will not necessarily be related to either how much energy is being moved, or to what is happening with the surface temperature.
Best to all,
w.
boris, you say:
I always figured that a man who could apologize was a gentleman. And a man who could graciously accept an apology is was true gentleman.
My hat is off to you,
w.
DeWitt, a pleasure as always.
Yes, I understand all of that, it’s all true. My point is that inside the thunderstorm, there is no interaction between an ascending parcel of air and the troposphere. It does not undergo either eddy diffusion or frictional mixing or mass exchange mixing with the troposphere. It neither radiates to the troposphere nor does it absorb radiation from the troposphere. It does not cool by radiation. It is insulated from any interaction with the troposphere. That is the difference between what I’m talking about and normal vertical air movement.
Yes and no. Yes, the precipitation leads to a dryer adiabat. But my understanding is that the Santa Ana is a gravitational orographic wind, a downslope wind created when cold dense air spills over the Sierras and flows downhill, warming as it goes.
My point exactly. Rather than radiating at the surface, with lots of GHGs above it, the air comes out at top of the atmosphere where it is free to radiate to space.
w.
Dewitt–
You need at least 2 D to model Hadley cells.
It’s funny. I tried to explain the link between the Hadley cells and the deserts to jae a year ago. He had some theory that seemed to suggest that the enhanced greenhouse effect doesn’t work, and the proof seemed to involve showing that dry deserts were hot.
On your more substantive question– I’d guess the air that with more CO2, the air that descends in the down leg of the Hadley cell must always be warmer than with less CO2. But enhanced radiation in the upper troposphere would be a negative feedback (all other things being equal.)
The other thing to remember is that while the adiabatic lapse rate is important as a limiting condition, the observed lapse rate rarely actually matches it. The atmosphere is sometimes in stable conditions, sometimes in unstable conditions. The averaged observed lapse rate also does not matched the adiabatic lapse rate (for many reasons.)
I never said that.
MarkR
Do you mean the TT hotspot? The surface has warmed. If you mean the hotspot think it is more correct to say that we are unable to detect the hotspot using data through 2000. (The Santer paper didn’t present more recent data.)
The difficulty is the data are thought insufficiently precise. So, the hotspot may exist, or not.
I don’t know why the RC don’t want to stick to the simple, clear, and true story line. I don’t know why they wrote a blog post that would tend to lead readers to make a incorrect conclusions about the meaning of the hotspot relative to both the models and the role of ghg’s. But they did. And we know readers made the incorrect conclusions because we read people explaining these incorrect conclusions and telling us their basis for these ideas is that RC article!
PS. (The Heracles myth is evocative, but I don’t think it tells us much about the debate over climate change.)
Boris–
You said this
Note, in my recent comment I used the word suggest, not said. My basis for believing you suggested they made that claim is the bit I quoted above.
So, what did you mean to suggest when you said Douglas claimed the hotspot was a discriminant? You think he said it discriminates vs. what from what?
Also, recall that you said this in context of telling us what Douglas actually said makes the RC discussion about models predicting the hotspot under solar forcing, and that it’s not .
So, maybe you can clarify the point you were actually trying to make, what you think Douglas actually said, and explain how in the heck it justifies the red herring discussion in RC.
Willis (#8129) –
Sure – see equation 1 here – one thing Monckton actually got close to right. For a more coherent discussion, see Bony et al – “How well do we understand and evaluate climate change feedback processes?” Journal of Climate 19: 3445 (2006). Also this was discussed I think in the very first IPCC report, and the basic understanding goes back at least to Hansen’s 1984 comparison of solar and GHG forcing effects. The breakdown of analysis into separate “forcing” and “feedback” components makes no sense unless this is approximately true.
Lucia – I’ve had my say on why RC’s points were perfectly valid in addressing the misconceptions arising from Douglass and friends, and your argument to the contrary (“red herring”???) still makes no sense to me. I don’t see any point continuing to go around in circles on this.
The simple, true, storyline doesn’t exist for the earth’s climate. That’s the problem. The AGW’ers have chosen to try and perpetuate a storyline about something that doesn’t exist as three words. But that’s what they use.
They put three (count them) words together- ‘Anthropogenic Global Warming’ to make a grandiose but simple claim about something not simple. That something (the climate) takes volumes, computers and long-winded posts to try and begin to describe. 😉
When the sepcifics go bad, AGW’ers can always fall back to the general hypothesis. It’s just words. The hypothesis should be more scientific and specific. AGW is three words designed to get people to support certain politicians. That’s all it is.
Andrew ♫
Arthur–
In the context of the overall discussion, you major points make no sense to me either. I grant that we agree on many points of physics. But that those points of physics do not resolve the issues being discussed.
By associating their discussion of hypothetical forcing to Douglas, creating the false impression the discussion had something to do with Douglas, and enveloping it in a discussion of comparisons between observations and model projections, the authors of that RC post injected numerous misconceptions about the hotspot and into the discussion about whether or not it has been detected. (It hasn’t.) They also divert attention away from the true significance of our failure to detect the hotspot and try to confuse those who have never taken thermo by incanting “adiabitic lapse rate”.
The true significance: If the hotspot does not appear, then the models are missing something.
I don’t know why you want to try to make their attempts to sow confusion about the meaning of the failure to detect the hotspot ok. Sometimes some seem, the RC authors were motivated by Douglas and Christy making incorrect claims about the meaning of our failure to observe the hot spot.
Douglas did that. But it’s possible to explain that Douglas and Christy made incorrect claims. Instead of doing that, the RC authors wrote their own post that created entirely new misconceptions about what failing to detect the hotspot means about climate change and modeling.
So the mistake Douglas made does not justify RC’s not specifically addressing Douglas &Christies mistake in the press release but discussing an irrelevant issue instead.
Some seem to suggest that the RC post is ok because it happens to make some correct statements of what models would predict under hypothetical circumstances. That would make the post “ok” only if the post were not written as if these observation had something to do with Douglas et al. As it stands the structure makes it appear that these observations contradict something Douglas said. It doesn’t.
Creating that false impression in readers is not justified even though the RC authors correctly report what models predict under entirely hypothetical forcings. Throwing in irrelevant discussion to distract the audience from the main issues is a red herring.
The confusion sown by that article is such that it continues to be used as an excuse to derail any discussion where people compare the model predictions of the TThotspot to data. The article continues to distract people from what failure to find the TT hotspot would mean.
Making the comparison is useful. If the hotspot is ultimately found, that will tend to support models. If it is not found, that will point to a flaw in models.
RC wasted their opportunity to say this by veering off into an irrelevant discussion about the adiabitic lapse rate or what models predict under hypothetical amounts of forcing to understand this.
Of course, you are not required to continue to repeat your views of why introducing the hypothetical ideas into discussion of model-data comparisons is justified, or how RC made some correct observations within the context of their irrelevant discussion. I get every one of these in my inbox (so I can check for spam/ nastiness etc.). Obviously, I’m going to reply from time to time.
I said quite some time ago: I’m surprised we’ve repeated this as often as we have.
lucia,
Once again you’ve managed to divert the argument to Douglass vs Gavin. However, your original post was about fingerprints and RC contradicting IPCC.
Let’s back away from Douglass vs Gavin and get back to your original statement:
WHO Expects a Tropical Tropospheric Hot Spot From ANY and ALL Sources of Warming?
Before you answer, let me clarify a little bit. Let’s define “TT hotspot” as “amplification of tropical surface temperature anomalies in the tropical troposphere”. That’s how I’ve always understood it. I’m pretty sure that’s what Arthur Smith and Boris mean as well.
So, the question becomes:
WHO expects an amplification of tropical surface temperature anomalies in the tropical troposphere from ANY and ALL sources of warming?
I do. Do you? If not, would you agree if “ANY and ALL” was replaced by “MOST”?
—
I really hope this whole thing is a case of mistaken definitions. Instead of continually starting new fights and going deeper into the nuances of who-said-what, why not use words that everyone can agree on? Instead of insisting on “fingerprint” why not use “characteristic” if that’s what you mean? Similarly, Arthur Smith and Boris could use “identifying characteristic” (per English dictionaries) if that’s what they mean.
I can think of no valid reasons to insist on using “fingerprint” knowing that your definition is not well-accepted.
—
Please, please, please — slow down, stop trying to score points against Boris and Arthur Smith, and get back to concepts instead of arguing about words.
Actually, never mind…
Don’t bother responding to my last post.
JohnV–
Boris and Arthur bring up issues, I respond to those issues. Even if you think these points are off topic, I run a more flexible blog, and permit these issues. (I also don’t consider them off topic. In particular the blog post discusses the RC post , their claims specifically. So, how you can think discussing those in comment is a diversion is beyond me.)
Also, if you don’t want discussion about word smithing, why are you asking things like this:
In answer to your question, read the top of the blog post. I am responding to Boris’s insistance that those at CA can’t use fingerprint in a particular way. He is insisting it can’t be used the way the IPCC defines it!
Obviously, I was not going to edit his quote to use the word “characteristic” when discussing Boris’s desire to decree the IPCC usages incorrect.
What’s your definition of “well accepted”, particularly in the context of discussions of climate change? I use the definition in the IPCC glossary. People have posted numerous examples from the peer reviewed literature that use this definition. I should think my usage is well accepted.
Embedded in all your discussion of your views on word definitions, you complain we should stick to concepts. You ask a question and then immediately provide your definition. (That is, you insist on wordsmithing but wish to be the smithy.)
I’ll answer : When discussing what happens in the real world, if the amplification due to a particular forcing (i.e. solar) is so small it’s contribution is enterly dwarfed by another source of forcing (i,e, ghgs), then the smaller forcing (i.e. solar) does not create a “hot spot”.
If solar acted alone at some very low level, then I would not call a spot “hot” in any real application unless we could, at least hypothetically, determine that the solar forcing raised the level of warthm sufficiently to resolve the feature from noise. This would mean that as a practical matter, the whether or not a “hot” spot can be said to be predicted will depend on the level of forcing, the magnitude of heating anticipated, and the natural weather variability.
I don’t consider things that aren’t hot, that aren’t “spot” like and that are undetectable as “hot spots” to be “hot spots”.
I think that given the level of solar forcings experienced in the thermometer record models do not predict a hot spot based on solar forcing. Given realistic forcings, the models predict a hot spot based on GHGs.
I’ve said this many times. I think you understood the concept.
lucia,
I attempted to define the question precisely because in my previous attempts at asking you complained that the question was too vague. Precisely defining a question is not word-smithing.
As far as I can tell, you are using definitions of both “hotspot” and “fingerprint” that are not well-accepted. I’ve given up arguing that they’re wrong (although I think they are), buy you have to admit that they’re not well-accepted given the number of people who do not accept them. 🙂
Using your definitions of “fingerprint” and “hotspot”, you are probably correct. Using the dictionary definition of “fingerprint” and the “enhanced warming vs surface” definition of “hotspot”, Boris is correct. This whole thing is about word definitions.
Down here in the comments you’re willing to be precise in your definitions. Up in the article you’re happy to leave the impression that enhanced TT warming is a unique and identifying characteristic of GHG-induced warming (when the same enhanced TT warming would appear if the source of warming was solar). You also state explicitly and incorrectly that a model run of hypothetical solar forcing *contradicts* the IPCC image of real solar forcing when the only difference is scale.
Of course, a lack of enhanced warming in the tropical troposphere versus the surface (if confirmed) would definitely be an issue for the models. I’m not saying otherwise.
JohnV
When the entire issue relates to the question of who expects “A”, then the definition of what A is matters. When Boris jumps into conversations and decrees “A tropical tropospheric hotspot is expected from any warming”, he does not define what he means by this.
The conversation he jumped into was discussing what a model-data comparisons, and context was what can happen given realistic forcings and what is expected given realistic forcings.
You want to come in afterwards impose a definition does not fit the context of that discussion. This might be fine if you could find a body who has decreed precisely what the term “hotspot” means. However, there is no standard definition.
In contrast, the IPCC glossary does define fingerprint– and they define it the way I use it. But somehow, you don’t like the IPCC definition of “fingerprint” to be used in the context of climate science. So, yes, if you wish to decree the IPCC definition of “fingerprint” is wrong then you might be able to make boris’s statements correct.
Since when doesn’t distorting scales to make wildly exaggerate or minimize an effect not result in contradiction? If I say I have a million dollars in my pocket and you empty my pockets and find only 1 cent, who would not think the evidence in my pocket does not contradict my claim? If I claimed there was no contradiction, it’s only a matter of scale because 1cent and 1 million dollars are both money, people would laugh.
The issue of scale matters to the discussion of whether or not anything that could possibly be called a hot spot exists.
The discussion in the IPCC document and in my blog post both relate to what is expected given forcings that have arisen on the real earth. These are relevant to Douglas. So, yes, because of issue realted to scale, no hot spot is expected to arise as a result of solar forcings when the solar forcing is as small as experienced during the period relevant to Douglas.
The RC figure, which is thrown into a discussion comparing observations to model projections, gives the false impressoin that something on the scale illustrated in the figure is predicted by models. Since the conversation is grounded in a particular time period, and nothing of that scale is expected, the impression given by RC is false.
So, yes. By showing an image with a totally distorted scale, the image contradicts the image in the IPCC which has a scale more appropriate to the discussion in Douglas, and later at CA. Scale does matter.
Lucia:
You seem to be offended not that the RC post makes any incorrect statements but rather that it brings up a topic in the context of Douglass that you don’t think they should bring up. However, don’t you think that it is they and not you who get to decide what is most relevant for the point they are trying to make? And, given that Douglass and co-authors were issuing press releases in which they were drawing all sorts of conclusions in regards to what the (supposed) lack of hotspot says about the mechanism responsible for the observed warming, you seem to be the only one who thinks it is somehow irrelevant to address under what circumstances (i.e., for which warming mechanisms) the models would predict an amplification of the surface temperature trend as you go up in the tropical atmosphere.
Could you tell me what specific fundamentally-incorrect point you think people reading that Real Climate post would come away with? You make the statement, “The RC figure, which is thrown into a discussion comparing observations to model projections, gives the false impression that something on the scale illustrated in the figure is predicted by models. Since the conversation is grounded in a particular time period, and nothing of that scale is expected, the impression given by RC is false.” However, this seems silly to me for at least a few reasons:
(1) The RC post makes it very clear the hypothetical experiment that they carried out (doubling CO2 vs increasing solar by 2%) and subsequently they also discuss how this relates to the actual observations (“If this is what should be expected over a long time period, what should be expected on the short time-scale available for comparison to the satellite or radiosonde records?…”)
(2) I agree with you that the estimates of solar forcing, or other forcings, mean that the model predictions are dominated by AGW. However, this is exactly the point of contention with “skeptics” including Douglass et al. I.e., they are arguing that the observed warming is due to something else besides greenhouse gases. Hence, it makes a lot of sense to consider hypothetical scenarios for warmings caused by other forcings (or simply due to internal variability), particularly if you are trying to convince people who don’t already believe that the modeling assumption that the climate change is dominated by greenhouse gases is correct. The way you seem to want to discuss “hotspot” basically amounts to a tautology…I.e., of course it will be the greenhouse gases that are expected to produce the hotspot by your definition if they are the forcing that is believed to have dominated. Duh!!! That is hardly a very interesting conclusion. More interesting is to ask what the pattern of the warming would have looked like if different effects were dominant.
(3) RC and Santer et al. (in various papers) have sharpened the debate by making clear both the origin of the tropical tropospheric amplification in the models, by noting the problems with the data (and various analyses and re-analyses thereof), by demonstrating the timescales over which the models and data do seem to agree (at least roughly…modulo the issues that Willis has raised here), and by correctly showing the estimated uncertainty in the model predictions. By contrast, Douglass et al. have only confused the debate by falsely leading people to believe that the tropical tropospheric amplification is a feature in the models that is unique to the warming being due to greenhouse gases and hence its (supposed) absence provides evidence that the warming is due to some other cause. They have also confused things by failing to acknowledge the very real problems that exist with the data for the multidecadal trends…and they have also vastly underestimated the model variability by incorrectly considering the standard error rather than the standard deviation to be the correct measure of the uncertainty. How in this climate (pun intended), you can heap blame primarily on the former folks and much less on the latter is beyond my abilities to comprehend!
I’m not offended. I think they are free to bring up what they wish to bring up, and of course they get to decide what to post and did. I also get to decide what I think of what they wrote, and I think I get to comment on things posted at blogs.
But Douglas made no claim about what would happen under hypothetical levels of solar warming. So, yes, I think it is irrelevant to bring up what models predict under hypothetical warming when trying to counter Douglass claims in either their press release of their paper.
It would have been better for RC to limit themselves to countering claims Douglas actually made.
Based on comments at other blogs, people have come away with the idea that even the hotspot is never be detected, or if it turns out to be much weaker than predicted, this would not indicate some problem with models and/or our ability to predict the effects of ghgs on climate. The supposed reason is that the fingerprint is not unique, and might be caused by solar forcing. People are saying this in comments at blogs, and linking back to RC to explain their issue.
This may not have been the intention of the authors at RC, but it has occurred. It is now a meme.
Had RC limited themselves to countering the arguments Douglas actually made, this confusion would probably not exist.
The RC posts never points out that the 2% increase in solar is utterly irrelevant to what we would see in actual observations. They make no mention of the fact that, during the observational period considered in Douglas, the relative magnitude of warming due to both causes is such that only the ghg component could possibly cause detectable warming in the troposphere. The level of solar forcing in that figure is irrelevant to comparisons between observations and models.
So, while the RC discussion is long, and describes data uncertainty and what not, and tells us why we might not even be able to detect the level of warming even if had happened as a result of ghgs, it is deceptive in suggesting the contribution hypothetical solar forcing to any hotspot is something we need to worry about when comparing observations to model projections. It is deceptive in liking the idea that the hypothetical forcings are relevant to anything Douglas said in their paper or in their press releases.
There are three ideas here:
1) You agree with me about the model predictions being dominated by AGW.
2) You complain the skeptics agree with both of us.
3) You think that because the skeptics agree with us, they use the truth of 1 to argue that observed warming is not due to greenhouse gases.
I’m not sure what to say to this. What I think is: When people make incorrect arguments, you identify the specific error, and correct that. So, if #1 is true, you don’t claim 1 is not true (or just shut up about it) simply because skeptics agree it’s true. (Ok. I’m pretty sure you mean something else here. But… I’m trying to figure out what.)
How does the discussion of the hypothetical scenarios convince anyone who doesn’t already believe that modeling assumption is correct?
I think the opposite occurs. By making it appear that countering Douglas’s argument, and data comparison, you have to introduce a discussion of what models predict would happen under fictional scenarios, you lose the audience.
Why not stick to the real difficulties in Douglas: Failure to properly consider uncertainty?
RC and may have have clarified something in some venue somewhere . But in that particular article RC introduced an irrelevant idea and distracted from the discussion of problems with the data. Santer may have also clarified something somewhere. But that turn the injection of the irrelevant discussion into the criticism of Douglas “clarifying”.
Santer et all do correctly point out the issues with data quality in Santer et al. That is the major contribution of Santer et. al. Notably, though you credit them with discussing the origin of the tt amplification in models, that discussion about the effects of hypothetical levels of solar forcing falls outside the scope of their paper addressing problems in Douglas et al.
Unfortunately, RC did not reign themselves in as Santer did in his paper.
Here you are accusing Douglas of two things. They did one but not the other:
1) Douglas never said that hypothetically large solar forcings would not cause the hot spot. They did not lead anyone to believe that tropospheric warming is a feature in models that is unique to ghg’s. That claim is utterly absent from their argument.
2) Douglas does claim absence of the hotspot provides evidence the warming is due to cause other than GHGs.
But heres’ the thing: They don’t use 1) to justify 2).
While (2) is an incorrect claim (in my opinion), the reason you don’t need (2) to claim (1) is that… Get this… they just make claim 2 leaving a big leap with a hole between their analysis and the claim.
We could use our imagination and fill in all sorts of candidate arguments to fill in the gap. But the fact that we can imagine candidates for fill in arguments doesn’t mean that’s an argument advanced by Douglas. Douglas did not claim #1.
Sure. I’ve blogged about this. When I have, I haven’t had people coming in to explain that Douglas didn’t make these mistakes. If people defended Douglas in this way, I’d contradict them.
But this issue is separate from the issue of the uniqueness of the hotspot or whether it is anticipated given realistic forcings. Even though Douglas made data analysis errors, that doesn’t magnically cause models to predict the hotspot for solar forcing levels experienced on earth during the instrumental period. They don’t. It doesn’t mean they made false claims about what models would predict under hypothetical forcings. Their data analysis error doesn’t magically make the RC discussion of the prediction of the hotspot under hypothetical solar forcing levels relevant to Douglas.
The RC discussion was irrelevant in that context.
I’m not blaming RC for Douglas’s errors. I’m criticizing them for making their own errors and sowing their own confusion.
I criticize RC for when I disagree with them. I criticize Douglas when I disagree with them. As I noted: When I did say Santer was correct about the errors in Douglas, I didn’t find people showing up to defend the errors in Douglas. The result was very few comments and less repetition. This has nothing to do with my heaping more blame on one than the other.
Lucia
“If I say I have a million dollars in my pocket and you empty my pockets and find only 1 cent, who would not think the evidence in my pocket does not contradict my claim? If I claimed there was no contradiction, it’s only a matter of scale because 1cent and 1 million dollars are both money, people would laugh.”
I think people would laugh at your anaolgy. You need to tighten up the magnitudes to make a reasonable anolgy. The difference in “hotspot” magnitude is not anywhere near 10^8…
If the claim is that you “have 1 million in your pocket” and you don’t have it, it doesn’t matter what the order of magnitude of the actual money in your pocket is… unless its exactly the 1 million that jibes with your claim.
Any amount other than 1 million makes your claim wrong, wether it be one penny or 10 billion in bailout money. 😉
Make different claim about what is in your pocket that’s accurate/right the next time.
Andrew ♫
Andrew
Of course it matters if that’s your analogy.
The difference in the figures of the IPCC is around 0.3C ish for solar forcing and about 1.1C for GHG’s. Lucia is, for some reason, saying that 1.1C is a hotspot and 0.3C isn’t.
The claim that there is no hotspot has neither been confirmed or contradicted. At the moment we don’t know how much is in your pocket.
Nathan,
I still think the analogy is good. If your claim of what is in your pocket is off, that means you miscounted how much was in your pocket. That’s a problem. The process itself (you count only in 3’s, for example) is wrong or the execution of the process (you counted right but forgot you gave some money to your mother 😉 ) is wrong. Either way, something needs to fixed.
The question that arises about being “close enough” is subjective. It all depends.
Andrew ♫
Andrew
The analogy is poor because it in no way mirrors the issue.
I think as well you are implying that the models are being claimed to be ‘correct’, which is not true.
Nathan,
I’m not even going to ask what “the issue” is in light of the 500+ posts that have tried to “explain” it already. 😉
Andrew ♫
Nathan–
It appears you accept that magnitude does matter. So, if that is the case, one can then debate whether a factor of 3, 10, 100, 1000 or so makes a difference.
lucia,
Your analogy about the amount of money in your pocket is just silly. It would be more accurate and complete if it went like this:
I add $1 to my pocket for every minute that I work.
Then if you worked 1 million minutes you would have 1 million dollars. If you worked 1 minute, you would have 1 dollar.
To be explicit, the number of minutes that you work is equivalent to the forcing amplitude. The amount of money in your pocket is equivalent to the amount of TT warming.
So, if you work 1 minute and have 1 dollar, does that contradict your claim?
—
You’re all upset that other people have mis-interpreted the RC article, but you seem to be completely unconcerned that your own article is easily mis-interpreted. Whether you like it or not, the word “fingerprint” has a dictionary definition that many/most people will use. Similarly, the word “hotspot” means enhanced warming to many/most people.
The issue of the TT hotspot has been mangled and mis-used many times on the “skeptic” side — are you not concerned that your article will contribute to more mangling and mis-use?
If your goal really is clear and un-ambiguous writing that prevents confusion (an admirable goal), then simply define what *you* mean by “fingerprint” and “hotspot” in your article.
Discussion bumped to new thread – Lucia