the Air Vent

Because the world needs another opinion

Antarctic Trends, Graphics and Discussion

Posted by Jeff Id on December 9, 2010

On Climate Audit, in response to a request by Hu McCulloch, Ryan posted the following improved graphic.

O'Donnel et al. Temperature trends of the antarctic. Click for larger image.


WordPress likes to help out with resizing images, so for the really big one go here.

Compare that to the Steig et al. reconstruction below which was used as the cover for Nature.

Back in the day when this work was still fresh in blogland, we made the point repeatedly that the warming in the peninsula was being spread around the continent.  On searching for a better jpg of the original Nature cover I ran across this reminder from Gavin Schmidt.

James Martin says:

I read today a claim that in the paper published recently by Dr Steig et al. in Nature regarding the Antarctic warming trend, there is a weighting problem. They claim that most of the weighting comes from the peninsula stations, which represents a relatively small part of the continent.

I was wondering if this is in fact the case? It doesn’t seem likely, but could you comment on this at all? If these assertions are left unchecked, before you know it they’ll be taken as fact.

[Response: The point of the Steig et al paper was to use spatial correlations in recent data to look at how under-sampled parts of the continent likely changed over longer time periods. Those correlations will necessarily weight different stations differently as based on the physical characteristics. The analysis you saw is simply a fishing expedition, an analysis of what the calculation is doing (fair enough), combined with an insinuation that the answer is somehow abnormal or suspicious (not ok). But how is this to be judged? What would be normal? No-one there can say and they would prefer simply to let people jump to conclusions. It’s kinda of typical of their tactics, but not a serious scientific point. – gavin]

My bold of course, while it is funny looking back at the usual RC commentary, there is a point yet to be made here.  These methods will always produce that kind of bleed of trend in the result.  The trick is to minimize it.   The spreading happens because they work by infilling missing data based on correlation of surface and satellite datasets, both of which contain noise.  This has the advantage of distributing information according to real world weather patterns rather than according to spatial location with the disadvantage of a bit of information spreading.  There were a number of mathematical improvements made in our recent paper which Ryan will explain in time, but one of the main ones was that our method used surface station trends only and spread them across the continent using the – trend noisy but spatially reasonable satellite surface skin temperature data.  Yet no matter how improved the method is, it will always have some of this problem.

My own contribution to this work was fairly limited.  Some commentary on methods, some minor editing and the original version contained some area weighted reconstructions similar to those presented here and below.  It was later removed from the SI after the first two reviews required us to completely change methods well away from Steig’s.  You can see that the pattern of trends is reasonably well matched to the published version above even though the image below used a different set of surface stations.  The difference though that catches my eye is the containment of the peninsula (image below) just to the finger region whereas there is a bit of trend bleeding into the mainland created by the more sophisticated methods.

A visual comparison of the above with our final results below can give a better idea of what the blending causes.  The peninsula region warming spreads slightly past the thumb into the mainland.  However it is far more defined than Steig 09 and much closer to the actual surface station trends shown in the closest station figure immediately above.

This next image was posted here also in the August timeframe and while the method is similar to what was published, it isn’t the same. The image from the published paper above does a far better job containing the peninsular trends than the one below.  However, despite the fact that different stations and methods were used, you can see that they are all  quite similar, except for Steig et al.

In all I would guess that hundreds of tweaked methods were used during the learning process.  What is learned when so many methods are tried is that the result is robust (ouch) to the method used and the basic result is something you can be confident represents the actual data.  On a final note, it was amusing in the review process where we were pushed to make conclusions in comparison to climate models and to support the conclusions  of Steig 09 as conditions of publication.  I’ve read a few papers on the models of the Antarctic but know very little on the topic.  I was and still am very much confused as to why the ‘models’ mattered at all when solving a math problem and found it rather interesting that climatologists would ask engineers, mathematicians and physicists to interpret a result in terms of weather patterns of a climate model.

 


67 Responses to “Antarctic Trends, Graphics and Discussion”

  1. Sonicfrog said

    Excellent post. Really displays the arrogance of the climate science establishment. I’m going to post this on my blog.

  2. timetochooseagain said

    “support the conclusions of Steig 09 as conditions of publication.”

    I was positively floored by this. Holy cow.

  3. hunter said

    The range of changes is 1oC?!?!?!?!?!?!
    And they are delcaring that
    1- this is an accurate representation of the temps of this vast area
    and
    2-that this is significant?
    What a load of utter crap.
    Tax payers should consider storming the Bastille of institutional science over this sort of bs.
    What an insult to common sense. What a demonstration of hubris on steroids.

  4. Jeff Id said

    #3 I’m not sure if you are referring to our work or Steig’s work but the methods are reasonable imo, Steig’s was just a little messed up while ours is less so.

  5. curious said

    That is a great headline image (including scale and reference!!) from Ryan – I think it would look good on a Christmas card…perhaps with Gavin’s quote and the previous version alongside?..🙂

  6. Ryan O said

    One thing to note with the “bleed-over” is that the ground station data doesn’t really support the relatively high trends our reconstructions show in Ellsworth Land. So the method is not perfect . . . which elicits a question about resolution. At what scale can our reconstruction be judged as “reasonably accurate”? The unfortunate answer is that we do not know, and the scale is likely to be dependent on, among other things, regional cloud cover (due to the cloud masking of the AVHRR data).

    So I think there’s still a good deal of improvement that can be made on our results as well.

  7. I agree with Hunter.

    The next revolution may well begin with “Tax payers” “storming the Bastille of institutional science over this sort of bs.”

    I am furious with myself for not recognizing government science as bs decades ago when the late Dr. Darka Das Sabu and I first went to Washington, DC in April of 1976 to present a paper showing that all primordial helium was labeled with excess Xe-136 at the birth of the solar system.

    Now, 34 years later, I belatedly recognize the government tyranny described in George Orwell’s book “1984” in the evolving climate scandal.

  8. TomRude said

    So this one will be the cover of JofC that will be on every MSM story?😉

  9. WhyNot said

    I want to be a smart %&$ for a moment, so a smart *^% I will be. A question I am sure that has not been asked much less answered is, did the O’Donnell et al. group take any MatLab classes, specifically any taught by Gavin? If I remember correctly, Gavin had a very cocky attitude and suggested that Jeff ID take one of his classes. It would appear that Gavin et al just got schooled.

    Frankly, I am tired of everyone trying to be “nice”, to appease the consensus crowd. We must as a nation call a kettle black when indeed it is black, lest we muddle our way into an indigent nation. The O’Donnell et al. paper (a great accomplishment and of course a very sound mathematical/statistical/scientific paper)has shown with a great deal of clarity how poor the pro-AWG crowd understands the analysis of noisy “point cloud” data.

    People have suggested that this paper relies on the S09 paper, and I say BS. If the O’Donnell et al. group would have been first to analyze the data, they would have came to the same conclusions as submitted, they would never have made the same blunders as Steig et al. I will call the kettle black here, and state that the O’Donnell et al. paper is professional while the Steig et al. is amateurish.

    One last statement as I am still a bit pissed off, does any one think Gavin has big enough balls to come over here and make another cocky statement like the one above? I personally think not.

  10. Layman Lurker said

    #9

    did the O’Donnell et al. group take any MatLab classes, specifically any taught by Gavin? If I remember correctly, Gavin had a very cocky attitude and suggested that Jeff ID take one of his classes.

    It was Eric Steig who made that suggestion, not Gavin.

  11. Layman Lurker said

    I believe that suggestion was made after Eric claimed that those who were requesting S09 archived code were “disingenuous”, as he had posted a link to his “code” (an online manual of matlab RegEM functions).

  12. Sonicfrog said

    Anyone see Eric’s response posted on RC? It’s pretty hilarious… and not in a good way.

  13. Biota said

    I do a bit of ecological modelling and test my methods against known data sets. Surely there is somewhere on earth where there is a good set of temperature data so that the pattern of the isolated Antarctic peninsular data can be replicated. The Steig and RO methods could be applied and compared against the known data in the extrapolated areas. Too simplistic?

  14. Jeff Id said

    #13, See the graph 3rd from the bottom for your answers.

  15. Jeff Id said

    Sonicfrog,

    I just read it, thanks for the heads up.

  16. Steve Fitzpatrick said

    “I was and still am very much confused as to why the ‘models’ mattered”

    I’m not at all confused… they don’t matter in this analysis; political considerations only.

  17. WhyNot said

    I stand corrected, thanks.

  18. Steve Fitzpatrick said

    “support the conclusions of Steig 09”

    Anybody who believes this paper supports the conclusions of Steig(09) has had one beer too many. The paper does support the general methodology of Steig(09)… as Ryan said, it was a very clever approach, but that doesn’t amount to supporting the conclusions of Steig (09).

  19. Brian said

    Did you see the map at RC and how much hotter they say your work makes the Antarctic trends?

  20. Jeff Id said

    Brian,

    We all did. Our unpublished work didn’t have that -.4 to 0.4 scale either. It also looks like the satellite mask may have changed too but I’m not sure. This was not our plot.

    Amazing, considering they claimed not to have a copy of our unpublished work.

    Here’s a comment we should save from the RC thread.

    MapleLeaf says:
    9 December 2010 at 1:01 PM

    Could someone please explain to me how O’Donnell et al. is a refutation of Steig et al.? Reading Ryans’ comments and the paper’s abstract I do not get that impression at all. To me, this look s like science advancing and building upon and improving and earlier (and seminal) work.

    I bets this is how that paper came to be. SM: “Guys, we have to attack the “team” again”. SM faithful: “But how, oh wise master?”. SM:”How about we look into that Nature paper by Steig that is getting so much attention and showing inconvenient things?”. SM faithful “Oh master thou truly are wise”.

    So they start all filled with hope and glee that Eric et al. messed something up big time. Nope. OK, dig deeper, and deeper and deeper still…until, voila we found some issues that were not adequately or properly addressed the first time round! Phew, good, science and peer-review still work.

    Now the WUWT crowd will spin and milk this for all it is worth, and use it to make attack the integrity of Eric and his co-authors, and then extrapolate that to all those scientists involved in climate research.

    What scared me upon looking at some of the images is how much more warming they found over large portions of the WAIS (especially the Peninsula) than Steig et al. originally did. Yikes, things are not looking good for the PIG. I’m sure Anthony will ignore that fact. And look at that warming over parts of the EAIS…fortunately that warming does not seem to be statistically significant just yet.

    And why did WUWT show an image that appears to have less warming than the one shown here by Eric? Sorry but I have to fault you both there..the figures should show for what season they are valid, or if they are for annual temperatures.

    [The figure here shows O’Donnell’s et al.s reconstruction for the same time period as our Nature cover image. These are annual mean estimates. I cannot speak to WTF WUWT has done.–eric]

  21. Layman Lurker said

    #18 Steve Fitzpatrick

    The paper does support the general methodology of Steig(09)…

    I think this is even a bit of a stretch Steve. I find this whole business of spinning OLMC(10) as somehow affirming S(09) as bizzare. How can OLMC(10) be supportive of S(09) methodology when the whole point of the paper was to improve upon it and correct it? Being generous, I suppose one could say that S(09)methodology was ‘in the ballpark’, but it was modelled in a manner which not only severely biased the 50 year trend, but also it’s temporal and spatial distribution. If this ‘doesn’t matter’ for climate science then perhaps I should become more cynical. However, I think the hints which Ryan and Jeff have given us regarding the favorable reviewer comments (and how they saw this work wrt Steig), along with the Weaver’s comment means maybe the team and RC are perhaps a little more isolated in their interpretation.

  22. John F. Pittman said

    It makes a really nice windows background. Hey JeffID, I gotta ask. Just how many times did you have to say “Oh, Master?”😉 Since some at RC show little ability to count, I gotta know if they got the number correct. Fewer still, it seems, have good reading comprhension, or memory about how this paper developed and why. Though perhaps, the truth is that they do not read the Air Vent. Though if true, I would note that they still manage to talk about persons whom posted, as far as I know, mainly here when it comes to the paper. I do wonder where or what sources of information were available other than here.

  23. Nic L said

    Brian, Jeff – #19 and #20
    I submitted a comment at RC over an hour ago pointing out the scale difference in the RC graph made the warming shown in our reconstruction look greater than in the original version. The comment is still “awaiting moderation”. I doubt it will actually get through moderation and be posted, but I live in hope of fair treatment!

  24. Hu McCulloch said

    Thanks, but you left off the color bar on the Steig version that accompanies it at http://earthobservatory.nasa.gov/IOTD/view.php?id=36736 . The large version there has the color bar embedded in the same jpg as the map, even if the smaller one does not. Note that their color bar only goes up to .25dC/decade, and that their cherry red is equivalent to your orange. You are getting far more warming in the Peninsula than they do anywhere.

  25. Steve Fitzpatrick said

    I tried pointing out (very respectfully) to Eric at the RC thread that the scale of the graphic had been expanded, and that the graphics for Seig (09) and RO(10) both used a very different scale. My comment appears to have ‘disappeared’ into the RC black hole of undesirable comments. Those guys just can’t bring themselves to admit some of their published work gave a less than accurate result. I give up trying to comment here; they are never going to change.

  26. Steve Fitzpatrick said

    Nic L #23,

    ‘Awaiting moderation’ is better than what I got.

  27. Nic L said

    Steve, #25
    The comment I submitted at RC has actually now been posted there! With a response from Eric Steig. They are as follows.
    “In response to MapleLeaf’s question in #6, the reason why WUWT showed an image that appears to have less warming than the one shown here seems to be that the scale has been altered on the RealClimate image, covering the range -0.4 to +0.4 rather than a range of -0.6 to +0.6 degrees C as used in the original and reproduced at WUWT. With the colour range used being much the same in both images, that obviously makes the warming trend appear greater in the image shown here. I can confirm that the continental 1957-2006 trend per our reconstruction, at 0.06 degrees C per decade, was only half the 0.12 level shown by the Steig et al. 2009 reconstruction.”

    Nicholas Lewis

    [Response: Yes. I never said otherwise. The point very simply is that Antarctica is not cooling, no matter how much some people try to make it so.-eric]

  28. Layman Lurker said

    Jeff, how would S09’s map look with RC’s color scale?

  29. Steve Fitzpatrick said

    Layman Lurker,

    I am suggesting that the general ability of using infilling to generate a useful reconstruction was affirmed, even if the Steig et al implementation was badly done. The two papers came to very different conclusions of the distribution of warming due to difference in implementation, not the general approach.

    Of course, singe Nic L is here, he could better comment on this than me.

  30. Steve Fitzpatrick said

    Nic L,

    I guess if you get a paper published, ‘disappearing’ your comments related to that paper is beyond even what RC is willing to do.

  31. I will be away for a few days for a much needed visit to the the Blue Mountain Center of Meditation:

    http://www.easwaran.org/index.php

    Here’s a nice video summary (parody) of Climategate.

  32. Jeff Id said

    Hu,

    Thanks for the link, the one I found was from a newspaper and didn’t have the scale. You are right that we found more warming in the peninsula than Steig did anywhere. You can see it in our side by side plot above. Not much we can do about the data tho right!

  33. hunter said

    Jeff Id,
    My dismissive post of this (#3) was in reference to Steig, Mann and gang using anything like this to support the idea that we are facing a CO2 driven climate apocalypse.
    The work that refutes it is brilliant- it shows (to my eyes) a large piece of the world where a sign could be hung saying: ‘nothing special happening here for the past 10,000 years’.

  34. Layman Lurker said

    #29 Steve Fitzpatrick

    the general ability of using infilling to generate a useful reconstruction was affirmed, even if the Steig et al implementation was badly done.

    I would agree with that.

  35. Nic L said

    Steve #30.

    Maybe, but I think Jeff may not have found so.

    Incidentally, Eric has now added the following sentence to his response to my comment:

    “Oh, and West Antarctic is still warming, even if you try to call parts of West Antarctica “the Peninsula”.

    This is referring to our paper using somewhat different geographic definitions of the various Antarctic regions from those used in Steig et al’s 2009 paper. The boundary of the Peninsula given in Wikipedia supports our definition.

  36. Nic L said

    Steve #29, Layman Lurker #34.

    I concur that the general ability of using infilling to generate a useful reconstruction was affirmed, although that ability is dependent on there being sufficient reasonable quality data to work from.

    In this case, the AVHRR satellite data, although unreliable as to trend, provided detailed spatial correlation information and enabled a useful reconstruction to be generated despite generally sparse ground station temperature data.

  37. Kenneth Fritsch said

    Hey bloggers can we discuss more of this:

    I concur that the general ability of using infilling to generate a useful reconstruction was affirmed, although that ability is dependent on there being sufficient reasonable quality data to work from.

    In this case, the AVHRR satellite data, although unreliable as to trend, provided detailed spatial correlation information and enabled a useful reconstruction to be generated despite generally sparse ground station temperature data.

    And less of this:

    “Oh, and West Antarctic is still warming, even if you try to call parts of West Antarctica “the Peninsula”.

    Nic L, I can see that if the AVHRR data were not stable over time but stable over space at any given time what you say is true. I am wondering in your paper whether you were able to show/estimate how stable the spatial correlations were over time – as I think you have to assume that that realtionship was stable to project the AVHRR spatial correlations back in time before the AVHRR measurements were made.

    Let us know when Steig et al says anything substantive about your paper – and Steig niggling does not count nor does his attempting to tell me the disposition of something that I can well decide for myself.

  38. Kenneth Fritsch said

    [Response: Yes. I never said otherwise. The point very simply is that Antarctica is not cooling, no matter how much some people try to make it so.-eric]

    That answer seems a bit over simplified in my eyes. Not statistically significant cooling does not imply statistically significant warming. Better and more in line with a disinterested scientist (not an advocate) might be something like if we use thus and thus data and methods we see statistically significant warming for 1957-2006, but if one comes forward approximately a decade we see no significant warming.

    Then the disinterested scientist would point to the O’Donnell data and note that others have found the warming by method A to be 0.06 +/-0.08 and by method B to be 0.04 +/-0.06 which shows that there has been no statistically significant warming in the Antarctica continent over the period 1957-2006. At that point, if he disagreed with the O’Donnell results, he could show where they went wrong – and would not merely continue to say that the Antarctica has warmed.

  39. Geoff Sherrington said

    When this theme was developing, there was discussion about the ability of satellite returns to discern cloud cover; and discussion anbut interpolation of temperatures under cloud. Were these resolved to your satisfaction?

    There are still some mismatches which you might not consider significant in the big picture.

    Australia maintains bases at
    Casey 66 17S, 110 32E
    Davis, 68 35S, 77 58E
    Mawson, 67 36S, 62 52E
    Macquarie Island 54 30S, 158 56E.

    Flipping through old notes and graphs, I recall that Macquarie Island was excluded from the data set. For interest, here is its Tmean from BOM –

    Different adjusters give different results. Here is BOM versus GISS for Macquarie. The differences are about as large as the total variation you found on your final map.

    I have not charted Casey, but here are the Tmean annual temps from Mawson and Davis, with Excel lines fitted (a practice I do not encourage).

    The themes are twofold. These bases are located where your map shows roughly 0.15 deg C per decade warming, yet that is not evident from the BOM data. Second, there is a much dependence on the sources of the data and I’m not sure that all sources would produce similar graphics.

    This is not meant to put a damp squid on your efforts. Your improvement of statistical methods was excellent, but I have residual worries about the integrity of the start data, for which you cannot be blamed.

  40. Geoff Sherrington said

    Sorry, was distracted in QQ and some of the lats and longs are wrong on the graphs.

  41. Re: Nic L (#35) about the boundaries of the Antarctic Peninsula.

    This seems a bit pedantic, but might be significant. The Steig and Wikipedia definitions look quite similar. BTW, Wikipedia is not an authorative source, and IMO citing it for critical definitions does not match the professionalism of your work.

    Steig 2009 definition (precise!): “Antarctic Peninsula as westerly longitudes north of 72S.”

    Wikipedia: “It extends from a line between Cape Adams (Weddell Sea) and a point on the mainland south of Eklund Islands.”

    A not very precise def from the Britannica:

    “Antarctic Peninsula, also called Palmer Peninsula, Graham Land, or Tierra de O’Higgins, peninsula claimed by Britain, Chile, and Argentina. It forms an 800-mile (1,300-kilometre) northward extension of Antarctica toward the southern tip of South America.”

  42. Nic L said

    Re Fabius Maximus (#41) about the boundaries of the Antarctic Peninsula.

    “Wikipedia is not an authorative source”

    I know Wikipedia isn’t an authoritative source, but I’m not sure that there is a fully authoritative source that precisely defines the boundary between the Peninsula and the West Antartica mainland. Very few climate science papers on Antarctica actually define that boundary, Steig et al. being an honorable exception. FWIW, a Google search on the exact Wikipedia definition brings up over 50 references to a diverse range of sources, all giving that definition of the boundary.

    “The Steig and Wikipedia definitions look quite similar.”

    Steig obviously doesn’t think they are that similar, or he wouldn’t have made his comment about us trying to call parts of West Antarctica “the Peninsula”! We also give a precise definition, which in fact in line with the Wikipedia definition.

  43. Nic L, thanks for the reply about this small point.

    Citing Wikipedia as a source for an alternative definition to the one used in the paper you were testing gives strength to Steig’s reply. It supports his effort to position you as amateurs (that’s the point of Wikipedia), without him neededing to describe the effect of the different definitions.

    Citing the volume of Google hits proves nothing. In this case, most of the Google hits appear to blogs and such copying the Wikipedia definition (which does not cite a source, but is probably copied from somewhere). But you can get many Google hits searching for proof of creationism, perhaps even for the flat earth theory.

  44. Ryan O said

    There is little precedent for Steig’s definition of the Peninsula (which is that portion of Antarctica from 72S and northward). Several papers give definitions closer to ours, and some include even more of “West Antarctica” in the Peninsula. Regardless, the definition used in our paper is based on the definition of the US Geological Survey:

    The major peninsula of Antarctica, extending from Prime Head in the north to a line between Cape Adams and a point on the mainland coast south of Eklund Islands. The first sighting of Antarctic Peninsula is contested but it apparently occurred in the 1820’s. Agreement on this name by the Advisory Committee on Antarctic Names (US-ACAN) and UK Antarctic Place-names Committee (UK-APC) in 1964 resolved a long-standing difference involving use of the American name, Palmer Peninsula, and the British name, Graham Land, for this feature. (Graham Land is now restricted to that part of Antarctic Peninsula northward of a line between Cape Jeremy and Cape Agassiz; Palmer Land to the part southward of that line.)

    http://geonames.usgs.gov/pls/gnispublic/f?p=gnispq:5:4594964652227490::NO::P5_ANTAR_ID:489

    Others include even more of “West Antarctica” in the Peninsula. For example, the North Dakota State University Geosciences Department defines the Peninsula as that portion northward of 75S, which includes part of Ellsworth Land bordering the Bellingshausen Sea:

    http://www.ndsu.nodak.edu/subantarctic/antarctic_peninsula.htm

    Regardless, Eric will have a hard time arguing that our definition is incorrect in any formal sense. While he may convince RC readers (who do not seem to need much convincing anyway) that our definition is wrong, given that ours is the official definition of the Peninsula, his influence will not extend much beyond the Ray Ladburys of the world.

  45. Ryan O said

    Geoff,

    In our reconstruction, Mawson is located in grid cell 5196, which demonstrates a 1957 – 2006 trend of -0.008 deg C/decade. Davis is in cell 5334, which demonstrates a 1957 – 2006 trend of 0.054 deg C/decade. Given that there is not 1957 – 1970 data for Davis, and most of the warming in East Antarctica occurs during this period, I believe our reconstruction is very consistent with the raw ground station data.

  46. Nic L,

    Thanks for the explanation. As rebuttals go, citing a USGS definition is the gold standard! Have you posted a note about this at RC (or rather, attempted to post)?

    Two other points:

    (1) Is there a central place for discussion of your paper? The debate seems to be spread out among several websites. Perhaps you could recruit some volunteers to copy everything (including RC) to one place.

    (2) The key aspect of publicity about your work vs. Steig 09 seems to be obscured in the debate (although mentioned many times): the relevance of both papers to the ice melting & sea level rise.

  47. Ryan O said

    Fabius,

    Yes, I have. It is “awaiting moderation”.

    As far as the key aspect of our work goes, I have a different impression. While I think that our results are useful for the physical processes you provide, the key point is that the details of the method matter. Eigenvector retention via thumbrules needs to be discouraged. Proper calibration needs to be performed for regression analysis. If physical weighting constraints are available, they need to be used. These are the important points, as failure to incorporate these concepts will lead to future results that may miss the mark even more than was done here.

  48. Jeff Id said

    #47, Sorry for the delay.

  49. AMac said

    Re: Ryan O (Dec 10 11:56),

    Can I bold Ryan O’s statement?

    the key point is that the details of the method matter. Eigenvector retention via thumbrules needs to be discouraged. Proper calibration needs to be performed for regression analysis. If physical weighting constraints are available, they need to be used. These are the important points, as failure to incorporate these concepts will lead to future results that may miss the mark even more than was done here.

    That’s a very succinct summary, accessible to a broad scientifically-literate audience!

    Perhaps you could highlight it in the body of a post on the subject?

    The important discussions that should be happening here, at RC, and elsewhere, are (1) whether and to what extent those criticisms of Steig09 have merit, and (2) whether and to what extent your paper’s methods improve upon those of Steig09.

  50. Nic L,

    There is no one master “key point” to anything.

    I’ll defer to you that the “the key point {to scientists} is that the details of the method matter. Eigenvector retention … Proper calibration … If physical weighting constraints …”

    That’s hardly the “key aspect of the publicity about your work”. None of these things were cental to the news stories about Steig 09, which focused on the consequences of the results Steig et al presented. Focusing only on the techical details cedes another round to the AGW advocates, who skillfully used Steig 09 to shape opinion both in the public and Washington. Public policy implications are one dimension of this debate, and IMO over a decadal time frame is the most important.

    Whether you or other knowledgable people work this aspect of the debate doesn’t matter, so long as it is not lost in the clutter.

  51. Ryan O said

    Fabius,

    If that is your concern, then you are likely to be disappointed. The likely interpretation of our results is that there is an increased danger of depletion of the West Antarctic Ice Sheet, since we show more warming in the Pine Island area. Although there is considerable debate on the effect of air temperature warming on these glaciers (and the newest research indicates that air temperature increases have small or negligible effect), you can rest assured that publicity concerning that particular result will fall into the “it’s worse than we thought” category.

  52. Bruce Friesen said

    Indeed, Ryan, AMac, Fabius, this has been a fascinating process. And it appears the strategy adopted by Steig et al is to shift the “key point” to one of their choosing.

    First, the authors of O’Donnell et al 2010 post up two figures showing geographic distributions of temperature trends, of which the Ozzies would say “Blind Freddie and his dog can see they are different.”

    Next, the authors of the newer work modestly and politely describe their work as an improvement, an advance in the science while giving credit to S09 as the foundation. I will presume to guess they assumed “Blind Freddie and his dog” could see visually there had been a change, and that blog readers would be interested in the science, the mechanics of the change.

    Then Dr. Steig shows up to point out (on the Doing it Ourselves thread) “Given that you guys have apparently demonstrated much greater warming that we found (!), in the most critical area glaciologically (the Amundsen Sea embayment), I’m afraid you may well find that the media you complain about will indeed emphasize the red even more than last time. The areas that are blue in your results are too cold to matter even if they did warm up a bit. No one has ever though the Ross Ice Shelf area (blue in your results) was sensitive to increasing temperatures. This is something I pointed out to you in an email or two some year or so ago. Nature has a funny way of conspiring against those that are trying to tell you that nothing interesting is happening!” He appears to be defining the import of S09 as of interest to glaciologists working to predict the behaviour of the Pine Island Glacier, or perhaps several adjacent glaciers.

    So to me it appears the two sets of authors are talking at cross purposes – one set talking about methods for reconstructing temperature trends for a continent, and the other talking about providing data to a specific group of specialists. Either that, or Dr. Steig has chosen to try to focus Blind Freddie on just about the only patch of Antarctica for which the two figures that started all this are similar.

  53. Layman Lurker said

    #51 Ryan O

    The likely interpretation of our results is that there is an increased danger of depletion of the West Antarctic Ice Sheet, since we show more warming in the Pine Island area.

    This would be a great topic to discuss from a technical POV – ie does the OLMC10 method allow for an ‘it’s worse than we thought’ interpretation for the PIG area? I would concede that the pixel color score would lead some to conclude that.

  54. Ryan O said

    Layman,

    As I replied to similar queries during the review process, while I may have opinions on this subject, the only thing on which I feel qualified to make any authoritative statements relates to the method, not the physical interpretation of the results. We specifically refrained from any such commentary in our paper. I agree that such discussion would be interesting; however, I would not be able to competently present a case for any proposed scenario.

  55. Layman Lurker said

    Agreed Ryan. I realize a physical interpretation of OLMC goes beyond the math to some degree, and my technical interest is confined to the method as well. One thing we discussed a lot during the blog deconstructions is how temperatures were distributed over time and space. IMO this is an important point when interpreting a finding of 50 year warming in the PIG or any other region. Greater warming, in the context of greater inherent variability wrt S09, may not be saying very much.

  56. Ryan O said

    I’m planning on doing an “uncertainties” post at some point in the future, which should prove somewhat enlightening (I hope). That may start the discussion you want to see. 😉

  57. Bruce Friesen said

    On reflection, I should have been more pointed in my comment of 1:07 p.m.

    Fenton Communications had lots of time to prepare for the release of OLMC10. It appears to me their strategy is two-fold:
    – to draw attention to the portion of the continent of Antarctica for which the two reconstructions are most similar, and say “it doesn’t matter”; and
    – to draw the discussion into minutia of geography and responses to warming, and away from the strengths of the authors of OLMC10.

    I was trying to encourage Ryan, Nic, Jeff and Steve to maintain the focus on the science and math of temperature reconstructions, areas of expertise in which they have, to my satisfaction at least, demonstrated the ability to “improve” on S09 methods.

    However, from reading Ryan’s post of 1:17 p.m. it is evident he at least is way ahead of me!

  58. Geoff Sherrington said

    Re – Ryan O said
    “In our reconstruction, Mawson is located in grid cell 5196, which demonstrates a 1957 – 2006 trend of -0.008 deg C/decade. Davis is in cell 5334, which demonstrates a 1957 – 2006 trend of 0.054 deg C/decade. Given that there is not 1957 – 1970 data for Davis, and most of the warming in East Antarctica occurs during this period, I believe our reconstruction is very consistent with the raw ground station data.”

    Thank you for the clarification. My comments should properly have followed a study of the final paper, rather than the gradual colour scheme in the region shown on the continental map. My apologies for jumping the gun.

    BTW, the data I have show Casey station also show cooling since 1970.

    Seeing you caveat about speculating (at 1.17 pm above), would you at least be asking yourself how the findings fit with the classroom concept of global warming by GHG? You appear to have shown that warming is not globally uniform and this it is hard to explain why it is patchy.

    Finally, one wonders if your quantitative results will be used in GCMs and the like; and whether they will impact on past model estimates of future global change.

  59. Ryan O (#6):

    Thanks for the explanation! It puts in perspective the comments of you and most of your co-authors about Steig 09. That is, his paper was largely correct. Considering the comments on skeptic websites like this, no wonder Stieg expresses annoyance.

  60. Geoff Sherrington said

    59 Fabius Maximus

    When you follow this Antarctic saga more closely, you will realise that Eric Steig was not largely correct, any more than your second spelling of his name is. When the authors of O10 maintain that their findings are more correct than those of Steig, such a statement is comparative. It will remain forever unknown which paper was closer to the number that could have been obtained by a dense coverage of stations over all those years since 1957 or so. It can be argued correctly that the later paper contained more suitable mathematics than the former and so is likely to be more correct.

    Why should Eric Steig be annoyed when a group of people showed him a better way to do the calculations? I would be delighted to have my concepts improved by volunteers. Why do you take the trouble to write about annoyance? Do tell us what upset you.

  61. BPW said

    Ryan O @51,

    You mean publicity like this?…

    http://rabett.blogspot.com/2010/12/melting-hopes.html#comments

  62. BPW said

    Or perhaps I should have linked the story rather than just the comments, though they are good as always…

    http://rabett.blogspot.com/2010/12/melting-hopes.html

  63. #60 Geoff Sherrington:

    I am going by what the authors are saying. Can you provide any expert analysis to support your assertions, overriding the authors’ comments above and elsewhere.

    “any more than your second spelling of his name”

    OMG, a typo in a comment! That’s a powerful rebuttal.

  64. Doug Proctor said

    The heating pattern suggests a warmed current on the west Antarctic peninsula. Only by coming at the image by a conclusion of AGW would one say it supports AGW.

    It is remarkable how much circular thinking there is to AGW. This image by itself explains why “climate disruption” is proposed to replace AGW: a consistent large scale effect is difficult, if not impossible, to find right now.

  65. curious said

    Does anybody have wikipedia editing status? Their Antarctic temperature trends text and graphic needs updating:

    Section: Effects of global warming

    Page: http://en.wikipedia.org/wiki/Antarctica

    Current text:

    “Some of Antarctica has been warming up; particularly strong warming has been noted on the Antarctic Peninsula. A study by Eric Steig published in 2009 noted for the first time that the continent-wide average surface temperature trend of Antarctica is slightly positive at >0.05 °C (0.09 °F) per decade from 1957 to 2006. This study also noted that West Antarctica has warmed by more than 0.1 °C (0.2 °F) per decade in the last 50 years, and this warming is strongest in winter and spring. This is partly offset by fall cooling in East Antarctica.[84] There is evidence from one study that Antarctica is warming as a result of human carbon dioxide emissions.[85] However, the small amount of surface warming in West Antarctica is not believed to be directly affecting the West Antarctic Ice Sheet’s contribution to sea level. Instead the recent increases in glacier outflow are believed to be due to an inflow of warm water from the deep ocean, just off the continental shelf.[86][87] The net contribution to sea level from the Antarctic Peninsula is more likely to be a direct result of the much greater atmospheric warming there.[88]

    In 2002 the Antarctic Peninsula’s Larsen-B ice shelf collapsed.[89] Between 28 February and 8 March 2008, about 570 square kilometres (220 sq mi) of ice from the Wilkins Ice Shelf on the southwest part of the peninsula collapsed, putting the remaining 15,000 km2 (5,800 sq mi) of the ice shelf at risk. The ice was being held back by a “thread” of ice about 6 km (4 mi) wide,[90][91] prior to its collapse on 5 April 2009.[92][93] According to NASA, the most widespread Antarctic surface melting of the past 30 years occurred in 2005, when an area of ice comparable in size to California briefly melted and refroze; this may have resulted from temperatures rising to as high as 5 °C (41 °F).[94]”

  66. I desire to learn
    more issues approximately it!

  67. For the reason that the admin of this web site is working, no doubt very
    shortly it will be well-known, due to its feature contents.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

 
%d bloggers like this: