Antarctic Update

We hear lots about polar amplification.  Mears and Wentz recently updated their TMT (mid-troposphere) data for 60-70S, but didn’t update the more controversial TLT3 (lower-troposphere) data, which is current only to August 2006.  Given the pending IPCC 4AR, you’d think that Mears and Wentz would be completely up-to-date with their lower troposphere results. Is there a reason?  Well, there’s a rather high correlation between the TMT and TLT3 series (0.76 – a huge correlation in climate terms) so I used the TMT series to project the TLT3 series to the end of December 2006. I’m projecting that the December 2006 Mears-Wentz Antarctic TLT3 when reported will be one of the lowest on record.

I’ve indicated my TLT3 “forecast” in red.  Also note that regardless of this last little downtick, there is no observable trend in the 60-70S Mears-Wentz lower troposphere results. It will be interesting to see IPCC 4AR handles the non-“amplification” at the South Pole.

Top- TLT3 (lower-troposphere); TMT – mid-troposphere. Last 4 months in top chart (dashed red) estimated from TMT by regression model.


  1. Peter Hartley
    Posted Jan 20, 2007 at 9:47 PM | Permalink

    Even ignoring the final down tick, to my eye the top graph shows a declining trend. I recall reading a recent article on the cosmic ray/cloudiness hypothesis (I think referenced somewhere on this site) that argued that it was consistent with a different pattern of warming over the poles while increased CO2 was not. The basic idea, as I recall, was that since incoming insolation is quite low at the poles, the effect of clouds on reflecting outgoing long wave radiation is stronger than the reflection of incoming shorter wave solar radiation. At lower latitudes, the effect of clouds reflecting incoming solar radiatin and cooling the surface exceeds the insulating effect, which tends to keep the surface warmer. Hence, when we have more cloudiness the poles (and especially the south pole) tend to warm while the rest of the globe cools. Conversely, reduced cloudiness tends to warm most of the earth, but the poles (and especially antarctica because of its land mass) cools. By comparison, CO2 diffuses throughout the atmosphere and its insulating effect with regard to outgoing radiation is the same sign everywhere, and greater in magnitude at the poles because of the lower humidity in polar landmasses.

    Does anyone else recall seeing this article? It would be interesting to look at it again in light of the above graph.

  2. Steve McIntyre
    Posted Jan 20, 2007 at 10:21 PM | Permalink

    Given all the talk about “polar amplification”, you’d think that scientists would pay attention to the Antarctica data trend. And that IPCC 4AR would provide an explanation. But they avoid the question like the plague.

    The differences between the Arctic and the Antarctic trends are one of the more intriguing aspects to current climate change.

    One thing that I learned recently when I was researching the Arctic ice shelves is that the Arctic Ocean is “warm” beneath the surface – and has a reverse thermocline. There is a veneer of cold fresh water from Siberian rivers which together with Siberian sea ice flows across the North Pole (about 4-5 years) to cross. Underneath is warm Atlantic water that is a couple of degrees above freezing.

  3. Steve McIntyre
    Posted Jan 20, 2007 at 10:39 PM | Permalink

    Here’s a comparison of GISS (black) and satellite (blue) for SH high-latitude (top) and NH-high-latitude (bottom).  The SH data prior to 1960 looks ridiculous.  The difference in trend is so extreme that you’d think that even IPCC would discuss it. But no such luck.

  4. epica
    Posted Jan 21, 2007 at 2:35 AM | Permalink

    There is no physical reason to assume polar amplification in the interior of Antarctica. The first order feedbacks are stronger in the tropics (mainly water vapour) and the only reason for polar amplification is albedo feedback. For that you need sesaonal snow cover or sea ice which is not the case in central Antarctica. So where are the projection showing strong amplification in Antarctica for the next say 50 years?

  5. Willis Eschenbach
    Posted Jan 21, 2007 at 4:01 AM | Permalink

    epica, people used to say that the polar amplification comes, not from albedo, but from the physics of CO2. A change of 1 W/m2 at 0°C is a blackbody change of 0.22°C, while the same 1 W/m2 change at 28°C is a blackbody change of 0.16°C. Thus, the same change in CO2 will cause about 0.22/0.16 -1 = about a third more warming at the poles. Basic physics, as the AGW folks like to say.

    Now, however, this explanation seems to have fallen into disfavor, since it applies equally to both poles … whereas the change in albedo is stronger in the Northern Hemisphere.

    In any case, Polyakov et al., “Observationally based assessment of polar amplification of global warming”, makes a good case that there is no amplification in the Arctic regardless of the theorized cause. Their abstract says:

    Abstract. Arctic variability is dominated by multi-decadal fluctuations. Incomplete
    sampling of these fluctuations results in highly variable arctic surface-air temperature
    (SAT) trends. Modulated by multi-decadal variability, SAT trends are often amplified
    relative to northern-hemispheric trends, but over the 125-year record we identify periods
    when arctic SAT trends were smaller or of opposite sign than northern-hemispheric trends.
    Arctic and northern-hemispheric air-temperature trends during the 20th century (when
    multi-decadal variablity had little net effect on computed trends) are similar, and do not
    support the predicted polar amplification of global warming. The possible moderating
    role of sea ice cannot be conclusively identified with existing data. If long-term trends are
    accepted as a valid measure of climate change, then the SAT and ice data do not support
    the proposed polar amplification of global warming. Intrinsic arctic variability obscures
    long-term changes, limiting our ability to identify complex feedbacks in the arctic climate


  6. Steve McIntyre
    Posted Jan 21, 2007 at 8:23 AM | Permalink

    Here are a couple of interesting links on polar amplification from realclimate and Connolley. epica, the realclimate position seems to be different than yours: they say that polar amplification is expected poleward of 70S:

    realclimate: Observations and models indicate that the equilibrium temperature change poleward of 70N or 70S can be a factor of two or more greater than the global average. …Antarctic amplification only occurs if a model is run long enough so ocean heat uptake in the Southern Ocean does not damp the positive feedbacks and if trends in stratospheric ozone do not cause compensatory cooling. Predictions with climate models indicate that Arctic amplification will be significant (above the 95% confidence level) in one to two decades, while significant Antarctic polar amplification will take much longer. A lack of polar amplification over relatively short periods of time is therefore understandable and does not undermine the credibility of climate models or climate theory in any way.

    The first sentence quoted here from realclimate doesn’t seem right – observations surely don’t indicate the claimed temperature change poleward of 70S. I wonder how long their models take to show Antarctic polar amplification. They say that they woldn’t expect the effect to show up in 1-2 decades, but CO2 forcing has been going on a long time now; it’s not just 1-2 decades.

    Here’s an excerpt from Connolley’s post:

    Connolley: polar amplification is:
    1.Ice-albedo feedback
    2. Forcing in the tropics warms the globe. Forcing in the extra-tropics warms the extra-tropics only. Hence, global forcing warms the extra-tropics more.

    Of these two effects, the second would apply to both poles; but the first would have minimal effect Antarctic. So based on this analysis, there should be polar amplification at the Antarctic as well. In his commentary, Connolley says:

    Before answering why, I should point out that only the northern polar regions show general warming at present, and that this is exactly what the climate models “predict”. This is generally (and I would say correctly) ascribed mostly to the moderating influence of the vast southern ocean. But it may also be partly because there is very little scope for ice-albedo feedback over Antarctica: its snow-covered now, and most of it will stay that way under any plausible warming. Unlike, say, bits of Siberia or Alaska, which (if you warm them a bit) will lose some seasonal snowcover. The only bit of “antarctica” which can participate in ice-albedo feedback is the seasonal sea ice; and this has only been measured well since 1979….

    However, if the 2nd of the two listed effects applies, then the models would predict some Antarctic amplification – maybe less, but still some. So northern amplification is not “exactly” what the models predict according to this. He does focus on the ice-albedo issue that epica refers to.

    The ice-albedo difference is logical, but each article has some inconsistencies that may or may not be important. In addition to the ice-albedo difference, I would have thought that there were other relevant differences between the two poles such as the presence of “warm” Atlantic water in the Arctic; the ongoing formation of sea ice in the seas offshore Siberia from north-flowing rivers; the transpolar drift of sea ice back to the Atlantic to melt; the thin veneer of fresh cold water from the above and reverse thermocline. I would have thought that fluctuations in poleward heat transport through the Atlantic could have a substantial impact separately from ice-albedo effects.

  7. Jeff Weffer
    Posted Jan 21, 2007 at 11:05 AM | Permalink

    An interesting observation comes from the Open Mind blog.

    The arctic temperature changes from GISS are a huge proportion of the overall global temperature. I wonder what the global temperature chart looks like if we exclude the 64N to 90N temps?

  8. Jeff Weffer
    Posted Jan 21, 2007 at 11:07 AM | Permalink

    Sorry, I meant to put in one with a different scale as well.

  9. Jeff Weffer
    Posted Jan 21, 2007 at 11:16 AM | Permalink

    I thought I’d also post the longest actual temperature record we have for one location on earth and that is Hudson Bay Canada, not quite the arctic but close enough. The temps recorded show large variations and natural cycles over time (and match the 1881 to 2006 GISS temp record for the arctic.)

  10. Ian
    Posted Jan 21, 2007 at 1:53 PM | Permalink

    As Willis points out, polar amplification is not dependent on albedo. This is simply a false argument created by the AGW camp to explain otherwise inconsistent results. Global warming from CO2 clearly predicts both higher temperatures at the poles (polar amplification) and higher temperatures in the troposphere first. They are only getting one out of three (higher temperatures in the Arctic). Under normal science scenarios this would cause one to re-examine their assumptions/models, but not so with the AGW camp 😉 At the very least this issue should be acknowledged and there should be some attempt ( the handwaving of always entertaining Mr. William Connolley aside) to explain the inconsistency.

  11. Mats Holmstrom
    Posted Jan 21, 2007 at 6:04 PM | Permalink

    Re #1,
    Maybe this is the paper you were thinking of:

    The Antarctic climate anomaly and galactic cosmic rays
    Henrik Svensmark

    It has been proposed that galactic cosmic rays may influence the Earth’s climate by affecting cloud formation. If changes in cloudiness play a part in climate change, their effect changes sign in Antarctica. Satellite data from the Earth Radiation Budget Experiment (ERBE) are here used to calculate the changes in surface temperatures at all latitudes, due to small percentage changes in cloudiness. The results match the observed contrasts in temperature changes, globally and in Antarctica. Evidently clouds do not just respond passively to climate changes but take an active part in the forcing, in accordance with changes in the solar magnetic field that vary the cosmic-ray flux.

  12. Peter Hartley
    Posted Jan 21, 2007 at 8:33 PM | Permalink

    Re 11 Yes, thank you, that was the article. Skimming it again, I see he uses the GISS temperatures. In light of Steve’s graph in post #3, that makes one a little concerned.

  13. Peter Hartley
    Posted Jan 21, 2007 at 8:52 PM | Permalink

    Further to #12: It would be interesting to re-do Figure 4b in the paper for the satellite temperature series as plotted in post #3. The satellite temperatures appear to be showing more of a contrary movement for antarctica and the NH than is apparent in the GISS series Svensmark used.

  14. Willis Eschenbach
    Posted Jan 21, 2007 at 10:53 PM | Permalink

    Before buying into the GISS or HadCRUT polar “data”, you might want to take a look at my post here. Simply put, what they are calling instrumental data is nothing of the sort.


  15. epica
    Posted Jan 22, 2007 at 3:34 AM | Permalink

    #5 I never understood the comparison with the sensitivity of a black body. DOesnt that imply sensitivity to changes in received energy (ie solar)? Infact I think the correct no-feedback calculation must be done by a Modtran or a real complete line-by-line model. I guess (without having checked in detail though) that the big first-order difference of 1/3 between the pols and the tropics will diminuish for simple reasons: Changes in the radiative balance do not take place at the surface (which is saturated in water vapour) but at several kilometers. Tropical temperature gradients are steeper than in high latitudes. However (as I said) the no feedback response has to be computed with a Modtran Model.
    I was referring mainly to the water vapour feedback. Water vapour concentrations will follow more or less Clausius Clapeyron which is why the WV feedback is stronger in the tropics.
    I cant see any contradiction with what William said. It will take much more time before the austral ocean react and seeice albedo feedbacks have stronger effects in Antarctica (in agreement with climate model predictions for the next 50 years).

  16. epica
    Posted Jan 22, 2007 at 4:09 AM | Permalink

    #6 The sentence you are referring to is “Observations and models indicate that the equilibrium temperature change poleward of 70N or 70S can be a factor of two or more greater than the global average.” (clearly pointing to paleo data). And in fact there are lots of observations that there was polar amplification in ancient climates. The LGM for example was 7K cooler in Antarctica and more than dobble of this cooler in high northern latitudes with tropics about 3K cooler. SO that gives a polar amplification of between 2-4 in agreement with GCMs (see Masson JoC 2006).

  17. David Smith
    Posted Jan 22, 2007 at 6:26 AM | Permalink

    RE #14 Willis, GISS also revises historical data in odd ways. I’ll e-mail a chart to you.

  18. Willis Eschenbach
    Posted Jan 22, 2007 at 7:17 AM | Permalink

    epica, thanks for your post.

    1) I find nothing by Masson in JoC 2006 except an article entitled “A CGCM Study on the Interaction between IOD and ENSO”, which doesn’t seem to pertain to polar amplification.

    2) The changes that occur during ice ages are affected by a host of factors other than CO2, including changes in global humidity, changes in ocean currents, and changes in weather patterns. We are discussing the “polar amplification” of CO2 warming. I don’t see what the ice ages might or might not prove about that.

    3) I fear that citing GCMs to support your position doesn’t impress me. GCMs project warning for the next century in a range from 1°C to 11°C, and their regional changes are just as variable. Trust them if you will … I won’t. Let me make my position clear: the results of GCMs are not evidence. I have spent too much time programming computers to take their output as evidence for anything. This is particularly true for simplified models of complex, chaotic, non-linear systems, such as GCMs.

    4) I agree that first-order calculations of effects are not reliable, due to known and unknown feedback mechanisms. I only brought up the example because that used to be used by AGM supporters to explain the expected “polar amplification. I trust actual observations more than either “simple physics” or GCMs. Speaking of observations, did you read Polyakov’s study? He lives in Fairbanks, works at the International Arctic Research Center there … he says, based on observation, that “polar amplification” isn’t happening.

    5) Saying “there are lots of observations that there was polar amplification in ancient climates” won’t get much traction on this blog. If you want folks to pay attention, cite the studies, summarize their findings, tell us exactly how they show that CO2 will lead to “polar amplification”, and tell us why they should be believed.

    6) The observations that we do have indicate that the North Pole and the South Pole are experiencing different trends … what is your your explanation of that?

    All the best,


  19. epica
    Posted Jan 22, 2007 at 7:45 AM | Permalink

    1) Sorry:Masson-Delmotte, V., Kageyama, M., Braconnot, P., Charbit, S., Krinner, G., Ritz, C., Guilyardi, E., Jouzel, J., Abe-Ouchi, A., Crucifix, M., Gladstone, R.M., Hewitt, C.D., Kitoh, A., LeGrande, A.N., Marti, O., Merkel, U., Motoi, T., Ohgaito, R., Otto-Bliesner, B., Peltier, W.R., Ross, I., Valdes, P.J., Vettoretti, G., Weber, S.L., Wolk, F., and Yu, Y., 2005, Past and future polar amplification of climate change: climate model intercomparisons and ice-core constrains: Climate Dynamics, p. doi:10.1007/s00382-005-0081-9.

    2) The issue is polar amplification. The basic idea is tha snow albedo/seaice albedo changes work in both direction not neccessarily exactly the same amplitude but in a comparable way. That is for the LGM cooler/more snox/more ice/lower albedo/even more cooler or for the future the other way around.

    3) 1-11C? That’s the wrong reading of the the Stainforth article, isnt it? Anyhow my point was only that a polar amplification of 2-4 is in agreement with GCMs (see the article above)

    4) Somewhere I read of Polyakov. Could you please give me the full citation please.

    5) Well mainly and to start with of course all the ice core works in Greenland or in Antarcica. Further Borehole temperatures and noble gases from both Greenland and Antarctica. That gives the 7C for Antarctica and for Greenland about 20-25C cooler. However in the Greenland case this would most probably overestimate polar amplification (compared with the tropics 3C cooler (based on UK37, Mg/Ca/ Nobel gases in Groundwater, faunal distribution, tropical snowline etc etc)). I havent checked all the literature for high latitude cooling but I guess something like 15C cooler for high northern lats seems ok. But check for more infos in the Masson article.

    6) South Pole has trends in all direction. I wouldnt speak of a solid difference in trends for the moment.

    1. Vaughan, D. G., Marshall, G. J., Connolley, W. M., King, J. C. & Mulvaney, R. Devil in the Detail. Science 293, 1777-1779 (2001).
    2. Turner, J., Lachlan-Cope, T. A., Colwell, S., Marshall, G. J. & Connolley, W. M. Significant Warming of the Antarctic Winter Troposphere. Science 311, 1914-1916 (2006).
    3. Monaghan, A. J. et al. Insignificant Change in Antarctic Snowfall Since the International Geophysical Year. Science 313, 827-830 (2006).
    4. Doran, P. T. et al. Antarctic climate cooling and terrestrial ecosystem response. Nature 415, 517-520 (2002).
    5. Comiso, J. C. Variability and Trends in Antarctic Surface Temperatures from In Situ and Satellite Infrared Measurements. Journal of Climate 13, 1674-1696 (2000).

  20. Steve McIntyre
    Posted Jan 22, 2007 at 9:17 AM | Permalink

    epica, if the phenomenon that people have in mind only is relevant at the North Pole in present circumstances, why do they call it “polar amplification” as opposed to “North Polar amplification”? Isn’t the term “polar amplification” overselling?

    On the other hand, despite your first comment above that there is no physical reason to assume polar amplification in Antarctica, IPCC 4AR takes a different view saying:

    A robust feature of the response of climate models to increases in atmospheric concentrations of GHGs is the poleward retreat of terrestrial snow and sea ice, and the polar amplification of increases in lower tropospheric temperature.

    So I think that it’s fair to ask people to account for Antarctica.

  21. epica
    Posted Jan 22, 2007 at 9:41 AM | Permalink

    #20 Might be I havent explained that well enough. Ofcourse there is polar amplification in both hemispheres, but models expect a large offset for the Antarctic Interior (all relevant processes such as seaice melting (see eg need more time in the Antarctic). Moreover even at equilibrium we have found lower amplification in the Antarctic (see the icecores) in the past. So I think IPCC 4AR does infact correctly address this point.

  22. Ian
    Posted Jan 22, 2007 at 10:00 AM | Permalink

    It appears the AGW camp is very good at changing their arguments — if not their minds. 😉 Seriously, what evidence would it take to for the AGW camp to question their theory?? A decade of declining temperatures? I doubt that would be enough either (climatic variability 😉 ).

  23. Steve Sadlov
    Posted Jan 22, 2007 at 10:27 AM | Permalink

    RE: #2 – And the impacts of massive Soviet / Russian irrigation diversions since the 1940s are …..? 😉

  24. Posted Jan 22, 2007 at 10:47 AM | Permalink

    Seriously, what evidence would it take to for the AGW camp to question their theory?

    i.e. what kind of observations would show that the propositions or theories are false?

  25. Ian
    Posted Jan 22, 2007 at 11:13 AM | Permalink

    And to just preempt the obvious response “what kind of evidence/observations would be required for the skeptics to believe?” Well a decent agreement between model predictions and reality would be a good start:

    – polar amplification ( of both poles). Certainly NOT declining temperatures in the Antarctic.
    – a warming trend in the troposphere that exceeds and leads surface temperatures (like that predicted in the models)
    – an overall temperature increase in the satellite data (the only decent global temperature data we really have) beyond normal variability
    – a prediction of absolute global temperatures from GCMs that is in line with reality and does not have to continuously invoke aerosols, ocean heat sinks and other bogeymen after the fact to explain the earth’s lack of cooperation in heating up (if these are known why are they not included in the models to begin with).

    That does not seem like too much to ask. GCM’s have failed terribly in their predictions. Does this mean AGW does not exist? No, of course not. Does it mean we should be skeptical? Yes, obviously.

  26. Posted Jan 22, 2007 at 11:21 AM | Permalink

    Actually, I think I remember one falsifiable statement, didn’t Prof. Jones say something like ‘2007 will be the warmest year on record’?

  27. bruce
    Posted Jan 22, 2007 at 11:25 AM | Permalink

    Re #26:

    Careful what you say about Phil Jones. He is one person who can make it so!

  28. Andy L
    Posted Jan 22, 2007 at 11:55 AM | Permalink

    I’ll second #25; although this site is not explicitly anti-AGW, many of the readers (myself included) are. I think of myself as a reasonable person, but in the end, the evidence that would convince me, is, in my opinion, impossible. So maybe I’m not very reasonable.

    Specifically, I would need to see climate models that could effectively explain current climate and predict future climate (at a level that would satisfy Willis) on both a local and a global scale. It would also have to give good bounds on natural variability and show convincingly how human actions can affect climate, both in the sense of “we’re breaking it” and in the sense of “let’s try to fix it”.

    Frankly, I believe that the system is too complex to ever be modeled in this sense. Specifically, I don’t believe it is deterministic with any reasonable amount of initial condition information over any significant timeframe.

    I’m open to being wrong here. But since we don’t have a second earth that we can set up with the same initial conditions and verify the behavior, even if global warming does happen and civilization collapses, I don’t think we can ever be specific about the attribution.

  29. epica
    Posted Jan 22, 2007 at 12:04 PM | Permalink

    I’m reading this site only sporadically, but #28 was by far the most honest post ever written here or most probably on any climate discussion forum.

  30. Ian
    Posted Jan 22, 2007 at 12:13 PM | Permalink

    I think you may be placing the bar too high personally. While climate may be chaotic, that does not mean it is completely unpredictable. We can be fairly certain for instance, that there will be another ice age sometime in the future with possible timelines and probability bounds. I think we can be fairly certain that if the sun gets hotter the earth will also. I think the theory of AGW is a decent one, it makes sense, it seems fairly straightforward, however the earth is not cooperating and this is why I am skeptical — I’m an old-school kind of guy that believes in verification from empirical evidence. If the conditions I listed above were met I don’t think anyone would be arguing. Everyone would be nodding their head in agreement and figuring out adaptation strategies. However, as it stands now the GCMs have no skill and observations have not matched theory. This means we should be skeptical.

  31. Douglas Hoyt
    Posted Jan 22, 2007 at 1:26 PM | Permalink

    RE 15:

    Water vapour concentrations will follow more or less Clausius Clapeyron which is why the WV feedback is stronger in the tropics.

    This assumption was put in the models by Manabe many years ago. At the time he said it was “guess” to use his word. It is an assumption since it assumes that thermodynamics completely controls the amount of water in the atmosphere. It is more likely that dynamics is the dominate control of water vapor and the water vapor feedback as argued by Hugh Ellsaesser and several others. For example, the Sahara is very hot but has a low relative humidity because dynamics (downwelling air currents) dominate rather than temperature (as the CC equation would say). Dynamics dominates on the regional scale and very likely on the global scale as well.

    Dynamics includes winds, convection, temperature gradients, pressure gradients, and the gravitational field. None of thes factors are captured by the CC equation. Ellsaesser says the WV feedback may be negative. My own calculations indicate it will be very slightly positve in agreement with Minschwaner and Dessler (2004). Without a strong water vapor feedback, none of the high climate sensitivies favored by the IPCC are viable.

    Also it is rather strange that models have higher temperature leading to more WV and then leading to less cloud cover, giving a second positive feedback. More WV would lead to more clouds offsetting any positive WV feedback.

  32. Steve McIntyre
    Posted Jan 22, 2007 at 1:53 PM | Permalink

    I received an email from a specialist whose employer prohibits him from posting here who says that my “TLT3 plot looks very suspiciously to be not as you claim – lower troposphere, but is instead the tropopause region (channel 3).” I’ve posted up the script used to produce these figures here

    AS you see, the data is read from the following file which appears to me to be the appropriate file.

    I also doublechecked by plotting the corresponding results from MSU and got similar results.

    It doesn’t look to me like my graph is in error, but, if it is, I’ll correct it.

  33. Posted Jan 22, 2007 at 2:57 PM | Permalink

    #28 Let’s not get carried away – there is the greenhouse effect, and the enhanced greenhouse effect due to anthropogenic CO2. Climate seems to be driven by the solar/cosmic ray connection. The best guess is 2/3 solar and 1/3 anthropogenic for current warming. The anthropogenic component is not restricted to CO2. Indeed, according to Pielke Sr, land use change over the past 300 years may have contributed more to climate change than would a doubling of CO2, and will continue to contribute in the future, regardless of what happens to atmospheric CO2. There is no short or medium term solution to the levels of atmospheric CO2, which seems to stay in the atmopshere for 50 to 200 years, at least according to DEFRA. So, ‘Action’ on climate change, unilateral and restricted to CO2 alone, is really no action at all.

    Anyway, if this happens soon:

    and I think it will, CO2 will be shown to be the weak GHG that it really is, with a deep cooling despite rising CO2 levels. We already had a cooling ‘test run’ from the 1940’s to 1970’s, when there was a fall in solar activity, and CO2 continued to rise.

  34. Andy L
    Posted Jan 22, 2007 at 3:18 PM | Permalink

    Thanks for #29; I try.

    #30, I think I’m going just far enough. Chaotic, unpredictable, these are all umbrella terms to cover a great deal of ground. The non-determinism is where my quibble is (and where I am most likely to be proven wrong). Even a well-calibrated model, that could say “this is the distribution of expected future climate” and predict the distribution well (in the sense of “of all the times that it was 50% certain, was it right 50% of the time”) would go a long way towards convincing me. My own observations, as well as Willis’s very deep investigations, have indicated that this is not the case now. And I have a feeling (pre-conceived bias) that the distributions that this hypothetical model would produce would have variances so high as to have no real predictive power; because I believe that the underlying system lacks that level of determinism.

    #33, I don’t know how we can blithely say that “there is the greenhouse effect”. I think the physical explanation reaches the level of plausibility. Generally, in epidemiology (another area where statistics are used to make decisions), you require a plausible mechanism of action (i.e. drinking tainted water leads to cholera), a correlation (incidence of cholera is higher around a particular water supply), and, generally, a measured dosage induces a measured response (the farther away you go from the source, the lower the incidence). So we have the first two for climate; but estimating the third is a nightmare — we even have directly contradictory evidence (that is essentially universally accepted) leading into the 70s. Since we can’t control the experiment, and we don’t have another planetary baseline to compare to, we’re left in a bad place.

    And since I’m an engineer (or maybe other reasons) I see having accurate compute models giving us a lot more power here; there may even be active solutions to the problem (“put more junk in the atmosphere”) rather than just the passive ones (“stop burning things”) that we have proposed now. Our inability to predict currently rules out both potential solutions.

  35. Posted Jan 22, 2007 at 4:38 PM | Permalink

    Re #21

    If I read this correctly, a strong warming of Antarctica, parallel or superior to the Arctic, would have been considered by leading climate scientists as a falsification of their models and publicly presented as a contradiction of the AGW theory, wouldn’t it?

  36. Ian
    Posted Jan 22, 2007 at 5:04 PM | Permalink

    #35 — Are you serious? 😉 LOL I’m sure this will the ‘revised’ history of what was predicted from now on — until observation disagree again (Antarctica warms up or something) then history will change again. How soon we forget the dire warning of rising temperatures and of ice melting in Antarctica, etc — that are still going on by the way — My memory isn’t quite that short to be told that cooling in Antarctica was all part of the plan — unfolding as prophesied.

  37. Pat Frank
    Posted Jan 22, 2007 at 5:42 PM | Permalink

    #25 — “Well a decent agreement between model predictions and reality would be a good start

    The problem is that no one has published the results of a propagation of parameter uncertainties and measurement errors through a GCM, which means the quantitative uncertainty is not known for any GCM. That being true, there are no uncertainty bars on a GCM projection, and so there is no way to tell what, exactly, a GCM is predicting. How, then, can anyone tell whether physical reality validates or refutes a model if the prediction uncertainties of the model are not known? “Ensemble averages” merely measure calculational divergence, and are not estimates of the physical reliability of a GCM model.

    As I see it, this is the basic problem problem in IPCC-style climatology. If they they haven’t quantified the inherent uncertainty limits on their projections, then they don’t objectively know what they’re talking about. Until the uncertainties are quantified, the entire alarmist position is built on nothing more than a tendentious guess.

  38. Pat Frank
    Posted Jan 22, 2007 at 6:22 PM | Permalink

    #31 — Doug, in light of your discussion, can you comment on the significance of Brian Soden’s 2005 paper in science 301, 841-844, where he detected the sort of linear increase of clear air humidity with temperature? He claimed that validated the WV feedback assumption in GCMs.

    I sent an email to Robert Cess, who had an essay commenting on Soden’s result in the same issue of Science, asking his opinion. His reply was that there was “[no] room for ambiguity” about the correctness of “fixed relative humidity.”

  39. Douglas Hoyt
    Posted Jan 22, 2007 at 7:09 PM | Permalink

    I think Soden’s results appear valid because they cover a limited region of the Earth for a limited time period. It is a mistake to generalize these results to say they represent what the entire globe is doing. There are enough areas and time periods around that one is bound to find a result that appears to agree with the constant humidity guess.

  40. Willis Eschenbach
    Posted Jan 22, 2007 at 7:10 PM | Permalink

    epica, thank you for your prompt and detailed reply. A few comments:

    1) I’m sorry for omitting the Polyakov link, I had given the full citation above and I thought I had also given a link. It is available here.

    2) You say:

    South Pole has trends in all direction. I wouldnt speak of a solid difference in trends for the moment

    However, since the North Pole doesn’t have “trends in all direction[s]”, that’s a solid difference in trends right there …

    3) While a polar amplification of 2-4° is “in agreement with” GCMs, since the GCMs have such a wide variety of outputs and totally lack error estimates, it is very difficult to find any phenomenon that is not “in agreement with” some or all GCMs. This says nothing about the phenomenon, only about the GCMs.

    4) Over the period of satellite observation, North Pole sea ice has been decreasing. This has been ascribed in part to a positive feedback from ice albedo. But during the same period, South Pole sea ice has increased by about the same amount. Should we ascribe this to a negative feedback from ice albedo? And if not, why not? This is a very important question from the point of view of our understanding of climate “¢’‚¬? AGW folks seem to forget that proposing a mechanism which could explain a given phenomenon is very different from showing that the mechanism causes the phenomenon.


  41. welikerocks
    Posted Jan 22, 2007 at 8:08 PM | Permalink

    #40 Willis, I posted this link back in Sept in the NAS Ice Core thread.
    Does this have anything to do with any part of your questions? This model is considering orbital cycles and our understanding of Antarctic ice. It illustrates with the opposite senerio -but you comment made me remember it:

    “The reason the frequency is not observable in records is because ice volume change occurred at both poles, but out of phase with each other. When ice was growing in the Northern Hemisphere, it was melting in the Southern,” said Raymo.

    “New Model Suggests Antarctic More Dynamic Than Previously Believe”

  42. Willis Eschenbach
    Posted Jan 22, 2007 at 9:54 PM | Permalink

    Rocks, there is an interesting potential explanation for the difference in North and South polar temperatures here.


  43. Willis Eschenbach
    Posted Jan 22, 2007 at 10:32 PM | Permalink


    There is an interesting 2006 paper here about the Holocene temperature history of Baffin Island (between Canada and Greenland), which contains the following graphic:

    Note that the area was up to 5°C (9°F) warmer during the early Holocene than at present. It is also worth noting that there are polar bears in the region, who are currently threatened with extinction due to a forecast 2°C warming …


  44. Steve McIntyre
    Posted Jan 22, 2007 at 11:54 PM | Permalink

    We’ve heard that ice albedo is what is supposedly causing “polar” amplification in the Arctic. Did anybody wonder about whether there was actually a trend in ice albedo? Here’s an article Laine 2004 which considers the matter. No significant trend – with ice albedo increasing in many parts of the Arctic. Notice the surprising size of the area with increasing albedo relative to the area with decreasing albedo – surprisingly balanced. Hard to see how this effect could generate an amplification – or as the IPCC likes to say: polar ammmmmm-plifica-tionnnnnnnn.  But hey, it’s only data. This article – which deals with actual data – is not cited in IPCC 4AR, which, needless to say, does not neglect to mention what models say. Typical.

  45. Posted Jan 23, 2007 at 4:07 AM | Permalink

    Re #36

    Yes, the other day I also watched a TV documentary on Antarctica (they didn’t bother to specify which region of it) where penguins were reported to suffer from “global warming” causing their habitat to shrink, amidst many other alarming messages. Also, right now we are in the middle of the Antarctic expeditions period so no self-respecting voyage will fail to include some sort of “scientific” research project to further study the effects of GW in the endangered continent. Just a couple of weeks ago a Spanish sailing boat barely managed to reach the polar circle in the Antarctic peninsula due to the ice accumulation (this season has set record highs in Antarctic sea ice extent) but in their blog they alarmingly announced to have found much less ice than expected in some stretches.

    In the meantime, the few of us in the public who know that, in fact, Antarctica as a whole has actually cooled in the last decades and is likely gaining ice mass are being lectured by the experts that this is in fact in total agreement with their model projections.

    I strongly suspect that if a narrator of a Discovery Channel documentary said that the Arctic sea ice extent had increased lately these same lecturers would protest in deep anger but they remain silent when the distortions are made on the other direction. It’s difficult to understand what has brought us to this silly situation.

    A good link for a visual appreciation of Antarctic sea ice evolution since 1978:

  46. epica
    Posted Jan 23, 2007 at 4:20 AM | Permalink

    # 44 It’s a pity that Laine 2004 is not mentioned in the AR4. It seems an interesting confirmation of ammmmmplificationnnnnnn. Laine says that albedo is not controlled principally by sea ice concentration (decreasing by about 2-3% per year) but by winter snow fall and start of the melting period (both of course partly depending on temperature). The data confirm a warming trend in agreement with albedo changes. Both (T and albedo) trends are of course over such a short period of 12 years very small and therefore it’s even more intruiging that we see more than noise. SOmeone knows why the 2004 article stopped in 1998? The satellite went down?

  47. epica
    Posted Jan 23, 2007 at 4:41 AM | Permalink

    Just a detail: Polar amplification is given as a multiplicative factor, so it’s not 2-4° as you wrote but 2-4 times global warming.
    In the masson paper you will see that nearly all models show polar amplification. For northern latitudes there is only an old version of the GISS that doesnt show polar amplification. Therefore my conclusion is IF there would be no amplification in obs over appropriate time scales the models are wrong. For the moment the models pass the test quite impressively. Thanks for the Polyakov paper.

  48. Steve McIntyre
    Posted Jan 23, 2007 at 7:41 AM | Permalink

    #46. epica, I agree that the period was very short, but the increase in Arctic temperatures has taken place over a very short period as well. I’m not arguing that sustained Arctic warming will not be associated with a reduction in Arctic sea ice or even with a reduction in Arctic albedo – both effects seem quite plausible to me. My point was a narrower one: the ice albedo “trends” such as they are do not seem large enough or consistent enough across basins to constitute a relevant amplification in connection with the present warming – thus a different explanatory factor would seem to be involved.

  49. Willis Eschenbach
    Posted Jan 23, 2007 at 7:48 AM | Permalink

    epica, you haven’t commented on the fact that we haven’t seen “polar amplification” at the south pole, or on Steve’s most interesting map showing the lack of albedo change. You say “The data confirm a warming trend in agreement with albedo changes.”, but you neglect to mention which “data” confirms which “warming trend”, and which “albedo changes” you are referring to. As I said before, this habit of making general claims without supporting evidence won’t get you much traction on this blog.

    South polar ice is growing, as shown by that HadISST database. What is your theory about “polar amplification” that explains that?

    All the best,


  50. Steve McIntyre
    Posted Jan 23, 2007 at 8:56 AM | Permalink

    #45. This is a very interesting site. Here’s a graphic from this website showing quantitative measurements of Antarctic sea ice area showing an increase over the past 25 years. Again this seems to be real data and not output from a model. I wonder what IPCC 4AR says about this data set.  Sea ice increases seem like an unlikely accompaniment to polar amplification, but I’m sure that epica will set us straight by saying that this is exactly what the models predict as a necessary component of polar amplification.

  51. Steve McIntyre
    Posted Jan 23, 2007 at 9:13 AM | Permalink

    #50. IPCC:

    Satellite data indicate a continuation of the roughly 2.7 ± 0.7% per decade decline in annual mean Arctic sea-ice extent since 1978. … Similar observations in the Antarctic reveal larger inter-annual variability but no consistent trends.

    Trends in noisy data are somewhat in the eye of the beholder, but the “trend” in the above figure doesn’t look a lot less consistent than other trends which we see that are supposed to be “:strongly significant”. I wonder if there’s any statistical test that can distinguish this trend as being insignificant while still preserving say, a trend in SH temperature.

  52. Jeff Norman
    Posted Jan 23, 2007 at 9:47 AM | Permalink

    Re: #49 Steve,

    What do you mean when you say: “the increase in Arctic temperatures has taken place over a very short period as well”?

    What is a very short period?

    I recently finished playing with some of the weather station data from the GISS database. I selected for proximity to the North Pole, length of the record and endurance of the record (to 2004 at least). I ended up with: Bjornoya, Danmarkshavn, Eureka, Gmo IM EK, Barrow, Hatanga, Ostrov Dikson, Ostrov Kotel, and Resolute.

    The I used the monthly data and complied them (based upon a 1960 to 1999 average) into a regional trends and ultimately an annual trend that shows the warmest ten year period was 1936 to 1945. The warmest year was a tie 1943 & 2005. The second warmest year was also a tie, 1944 & 2006.

    According to this trend, the high Arctic cooled from ~1944 to ~1968 and then warmed from ~1978 to 2006. Are you referring to this later period?

    During this period, the greatest relative warming occurred in March, April, November and December, months I would not directly associate with changing albedo.

  53. Steve McIntyre
    Posted Jan 23, 2007 at 9:52 AM | Permalink

    #52. I didn’t mean to opine on the 1930s versus the present about which there are issues. I was simply referring to the temperature increase in the past few decades.

  54. Steve Sadlov
    Posted Jan 23, 2007 at 10:09 AM | Permalink

    RE: #52 – That’s fascinating. If one were to accept, say, some point in the early 00’s as the start of the seemingly emerging negative PDO phase, then both the warmest and second warmest years fall into the head end of a negative PDO phase (the previous negative phase being considered ~1940 – ~ 1976). I am increasingly of a mind that PDO transitions (operational definition proposed to be, the 5 years encompassing the change and its initial reverberations) are highly disruptive events, replete with drought, wild swings in temperature and extremes in temperature.

  55. Paul Penrose
    Posted Jan 23, 2007 at 10:12 AM | Permalink

    Your faith in the models is amusing, but not very scientific. You assume that the error bars in the model outputs are less than the observations you are comparing them to, however we don’t know what the error bars are for any of these models. It is equally possible that they are larger than the observed data, so it is meaningless, in a scientific sense, to make these comparisons. You are free to believe that the model error bars are small, and that the comparisons mean something, but belief in absence of data is the definition of faith.

  56. Jeff Weffer
    Posted Jan 23, 2007 at 12:44 PM | Permalink

    Re: #52 Jeff Norman – You are sitting on the most important temperature record in my opinion. I posted above that virtually ALL of the 0.6C increase in global average temperatures since 1880 are, in fact, the result of swings in Arctic temperatures.

    The Arctic was very cold in the 1880s to the early 1900s, influencing the low global temperature average of 1880. It warmed considerably from 1920 to the 1940s (3.0C) which again influenced the rise in global average temperatures (0.6C between 1920 and 1940).

    The arctic then cooled 2.0C (global average down 0.4C between 1942 to 1975.) Since 1975, arctic is up 2.5C (global average up 0.7C).

    I also posted the long temperature record of York Factory/Churchill Manitoba Canada which shows long swings in northern temperatures. (When the Hudson Bay Company set up a fur trading post on Hudson Bay in the 1750s, the fur traders reported back that it got extremely cold there for long periods of time and it was difficult to stay alive in some winters. The Hudson bay Company teamed up with the Royal Society of England to test out the temperatures with a relatively new device – the thermometer – to find just how cold it really was. They continued testing on on off for many years afterward.

    The long-term average since 1769 is virtually the same as today however.

    And given the lack of temperature increase or change in the Antarctic, it seems to me that Arctic temperatures just have large swings that last for decades at a time (and these swings are THE factor influencing the global average temperature record seen since 1880 – 0.6C)

  57. Jeff Norman
    Posted Jan 23, 2007 at 1:14 PM | Permalink

    Re#56: Jeff Weffer

    I wonder how many of us “waste” 😉 our time recreating these temperature trends looking for clues? We should create our own web site just for posting these trends and sharing our methodologies.

  58. Posted Jan 23, 2007 at 1:30 PM | Permalink

    #27 Data manipulation wouldn’t be fair 😉

    Actually, that prediction is quite bold, my Random Walk Climate model says there is only 16 % change that 2007 will be the warmest on record.

  59. Posted Jan 23, 2007 at 1:46 PM | Permalink

    “Sea ice increases seem like an unlikely accompaniment to polar amplification”

    Indeed. And all the more striking when you take into account that:

    a) One of the reasons for the lack of polar amplification in Antarctica so far is that the vast expanses of water around it are taking up the enhanced greenhouse heat (how can they be doing that and freezing more at the same time?).

    b) By the same token, the reason for the ice mass gain in Antarctica is alleged to be the increase in snow precipitation due to warmer water around it (idem).

  60. JP
    Posted Jan 23, 2007 at 6:49 PM | Permalink

    re: 45

    The Bellingshausen/Amundsen Seas appears to be the only area of the Antarctic that has lost ice. Why has this one area seen shrinking ice extents? Do the variations in the thermohaline circulation have anything to do with it?

  61. Steve McIntyre
    Posted Jan 24, 2007 at 8:53 AM | Permalink

    #32. The specialist who questioned whether I had goofed on my plot has agreed that my calculation was correct and sent me the following email (which he agreed could be posted):

    Thanks for looking in to this. It was probably a trick of the light and
    shows danger of eyeballing timeseries. It looked like TLT was cooling
    more than TMT which is at odds with what the RSS plots show on their
    website I believe. I was mainly querying the relative rates of change
    rather than the sign from a physical viewpoint and I’m sorry I didn’t
    make this clear. I don’t doubt that you are right though and that the
    RSS plots are misleading at this latitude.

    It wasn’t that there was cooling that I was primarily querying, but that
    the cooling was greater than for the higher T2 channel (look at the 4
    weighting functions on RSS site). If you deconvolve T2LT, T2, T3, and T4
    and try to get a single vertical trend profile that fits all these
    reported trends at this latitude then I’m 99% sure that it would require
    massive heating at about 300hPa and nowhere else to get this result. I’m
    not sure this is physically sensible! I guess that shows the instability
    of this instrument and the whole MSU dataset stuff. You could also
    trivially show the same result for UAH. RSS may look worse if you
    include channel 3 as thats an extra d.o.f. but I don’t think the basic
    error structure is that different between the datasets. Its just a
    generic issue with the MSU datasets … they are no panacea and not the
    “precise monitoring” originally claimed.

    One thing to note is that both RSS and UAH strongly distrust the last
    two years worth of data from the AMSUs. They have been moving in
    different directions and neither group knows why. There’s more on this
    at UAH on why they reverted back from 6.0p. So I wouldn’t trust the
    latest 24 months for observational reasons anyway.

  62. Willis Eschenbach
    Posted Jan 24, 2007 at 6:11 PM | Permalink

    One thing to note about the RSS weighting function for the TLT is that it unphysical. It ascribes a negative weight to stratospheric levels, and thus stratospheric cooling is interpreted as tropospheric warming. This may be the source of the oddity noted in the email to Steve above


  63. Steve McIntyre
    Posted Jan 24, 2007 at 6:57 PM | Permalink

    #62. Maybe they consulted with Mann on principal components.

  64. Posted Jan 29, 2007 at 12:23 PM | Permalink

    The new papers are not supposed to influence IPCC AR4, only papers published by summer 2006, do I understand it well? I guess that most people interested in Antarctica already know about it, but if you don’t, read about the Battle of Antarctica. 😉

  65. KevinUK
    Posted Jan 29, 2007 at 2:10 PM | Permalink

    #64 Lubos

    Thanks for that. I must have been a sleep recently as I’d missed this one. I wonder whether this is the ‘iconic statement’ that they’ll be announcing on Friday? Does the non-publication of the full report until May give them time to include this ‘new evidence’ for catastrophic sea levels rises due to ‘melting of the Gorillas’ in the final report? I love the way ‘Anubis’ refers to the IPCC as conservative. Not from where I’m sitting they’re not.

    Also I’m a little puzzled by the Observer article. On the one hand it mentions that the BAS want the IPCC 4AR to include the Larsen B evidence yet ‘Billy the wiki’ seems to be arguing against it. Is this because he’s a climate modeller? Unless I’ve read it wrong it looks like there isn’t even a consensus at the BAS let alone the IPCC.


  66. KevinUK
    Posted Jan 29, 2007 at 2:12 PM | Permalink

    #62 and #63 Willis and Steve M

    Are you going to let us in on the joke?


  67. Posted Jan 29, 2007 at 2:38 PM | Permalink

    Dear KevinUK #65, thanks & it was a pleasure to offer the link. Nice that you noticed that Connolley is essentially dismissing the director of his BAS – that’s pretty courageous. 😉 Finally, it wouldn’t surprise me if there were more controversy within BAS than within IPCC itself – because there could be many (but not all) Anubis’ people who always want the predictions to be higher than all previous ones. The IPCC people in average could be more reasonable and more cautious. In the previous sentence, it is important to realize that “more [adjective]” doesn’t imply “[adjective]”.

  68. Steve McIntyre
    Posted Jan 29, 2007 at 3:28 PM | Permalink

    IPCC 4AR in the text notes both warming on the Antarctic Peninsula and colling over the continent. This pattern is attributed to enhanced circumpolar westerlies at ~60S, citing an article coauthored by the Chairman of WG1.

    Temperatures over mainland Antarctica (south of 65°S) have decreased slightly since 1966 (Doran et al., 2002), but there has very likely been strong warming over the last 50 years in the Antarctic Peninsula region (Turner et al., 2005), see Figure 3.6.7.

    The corresponding enhancement of the near-surface circumpolar westerlies at ~60°S, and associated changes in meridional winds in some sectors, is consistent with a warming trend observed at weather stations over the Antarctic Peninsula and Patagonia (Thompson and Solomon, 2002; see also Sections and 3.6.5).

    The trend in the SAM, which is statistically significant annually, and in summer and autumn (Marshall et al., 2004), has contributed to Antarctic temperature trends (Thompson and Solomon, 2002 – Thompson, D.W.J., and S. Solomon, 2002: Interpretation of recent Southern Hemisphere climate change. Science, 296, 895–899.; Kwok and Comiso, 2002b; van den Broeke and van Lipzig, 2003; Schneider et al., 2004); specifically a strong summer warming in the Peninsula region and a cooling over much of the rest of the continent (Turner et al., 2005), see Figure 3.6.7.

    In the SH, SAM changes are identified with contrasting trends of strong warming in the Antarctic Peninsula, and cooling over the interior of Antarctica, and the increasing positive phase of the SAM has been linked to stratospheric ozone depletion and to greenhouse gas increases.

    In the SH, SAM changes, in association with the ozone hole, have been identified with recent contrasting trends of large warming in the Antarctic Peninsula, and cooling over interior Antarctica. [Sections 3.5, 3.6, 3.8.4]

    In the SH, where the ozone hole has played a role, it has resulted in cooling over the interior of Antarctica but large warming in the Antarctic Peninsula region and Patagonia.

    Given the observed cooling in the continent, one can see why they might tread with some trepidation on these matters.

  69. David Smith
    Posted Jan 29, 2007 at 4:05 PM | Permalink

    Re #64 Thanks for the link! It communicates well.

  70. Posted Jan 29, 2007 at 4:12 PM | Permalink

    From Steve’s citation:

    In the SH, where the ozone hole has played a role, it has resulted in cooling over the interior of Antarctica but large warming in the Antarctic Peninsula region and Patagonia.

    And here is the station record for Punta Arenas, the southernmost town in the world:

    Wow, what a warming!

  71. Posted Jan 29, 2007 at 6:00 PM | Permalink

    Steve #68: Thompson and these people have a lot of exposure there. What do they complain about? Also, it may be just my ignorance but why is it believed – apparently it is – that the importance of the peninsula is about 50% of the importance of the continent? It looks like 1-2 percent of the area. Is there some rational explanation why they care about it so much other than that it is more compatible with global warming?

    Has someone read some of these papers about enhanced circumpolar westerlies? Circumpolar westerlies are about the most ordinary weather events that can occur around Antarctica. Why are they enhanced and is the enhancement believed to last? Also, they link the non-uniform patterns to all possible things from CO2 to ozone. Is there actually some evidence for these links or is it again just a science based on no complaints seen, no complaints allowed, consensus?

    I am sure that I would have a better idea which of these things are answered and which of them are not had I read the papers but couldn’t there be a concise explanation that could be squeezed to the summary? There has been a lot of citations of science papers that are supposed to justify a statement even though they don’t – like the infamous Stern report. Does IPCC plan to construct something similar as the Stern report?

    There are just too many effects seen and too many possible explanations and neither of the recent new observations seems to have been predicted. Who has predicted that the oceans would cool down in 2003-2005? Who has predicted that the bulks of the ice continents would grow while the tiny boundaries would melt? Who has predicted that 2006 would be much cooler than 2005 and hurricane free? None of the observed weather patterns at timescales longer than 1 week seems to be predicted by anyone. Despite these simple facts, we still hear that the theories they use today are essentially the same as the theories they had a few years ago and they are fine. Are they?

  72. David Smith
    Posted Jan 29, 2007 at 10:17 PM | Permalink

    Re #71 Lubos, I started to write a long post about the Antarctic weather mechanism they propose but I’ll instead summarize it by saying that I can’t find clear evidence for it. I tried to match NCEP data with what they say is happening and the data doesn’t support key parts of the hypothesis.

    I think if there was clear evidence, we’d have heard about it.

    As an aside, while looking at the data, I cross-checked atmospheric thickness against the reported temperatures. The colder the air, the thicker it becomes. What I found is that the Southern Hemisphere, from 50S southward, has clearly gotten thicker (colder) since the radiosonde record began about 1960. Confirmation.

    Then, I looked at the Peninsula, which is shown here . (The plot indicates the thickness of the air below about 2.5 kilometers. The lower the value, the thicker (colder) the air.) It shows that the Peninsula air, too, has been cooling.

    Well, maybe there’s a problem with using thickness, so I looked at the area where John A provided the surface station data (above). What I found was a trendless plot ( here ), which is not much different from John A’s plot.

    It all makes me scratch my head about the reported Peninsula warming. I wonder if warm local ocean currents play a role in the reported warming.

  73. Posted Jan 6, 2008 at 8:23 PM | Permalink

    Update: 2007 didn’t produce much warming in either the northern polar or southern polar regions.

%d bloggers like this: