A New Preprint by Wunsch on Abrupt Climate Change

Carl Wunsch has an interesting new preprint entitled: Abrupt climate change. An alternate view at his website.

C. Wunsch, 2005. Abrupt climate change. An alternate view states in the abstract:

Hypotheses and inferences concerning the nature of Dansgaard-Oeschger (D-O) events are reviewed. There is little concrete evidence that these events are more than a regional Greenland phenomenon. The partial coherence of ice core àƒÅ½à‚ⲱ8O and CH4 is a possible exception. Claims, however, of D-O presence in most remote locations cannot be distinguished from the hypothesis that many regions are just exhibiting temporal variability in climate proxies with approximately similar frequency content. Further suggestions that D-O events in Greenland are a generated by shifts in the North Atlantic ocean circulation seem highly implausible, given the weak contribution of the high latitude ocean to the meridional flux of heat. A more likely scenario is that changes in the ocean circulation are a consequence of wind shifts. The disappearance of D-O events in the Holocene coincides with the disappearance also of the Laurentide and Fennoscandian ice sheets. It is thus suggested that D-O events are a consequence of interactions of the wind-field with the continental ice sheets and that better understanding of the windfield in the glacial periods is the highest priority. Wind fields are capable of great volatility, very rapid global-scale teleconnections, are efficient generators of oceanic circulation changes and (more speculatively) of multiple states relative to great ice sheets. Connection of D-O events to the possibility of modern abrupt climate change rests on a very weak chain of assumptions.

Wunsch illustrates the conventional view as essentially being a commentary on the following well-known figure:

Wunsch Caption: Figure 1: àƒÅ½à‚ⲱ8O from the GISP2 ice core as measured in the Quaternary Research Laboratory, U. of Washington (Stuiver and Grootes, 2000). Note that time runs from left to right in the physics convention. The ratio of 18O to 16O concentrations is believed to track local atmospheric temperatures in central Greenland to within an approximate factor of two (e.g., Landais et al., 2004). Large positive spikes are called Dansgaard-Oeschger (D-O) events and are correlated with abrupt warmings– measured independently in some cases. Note in particular, the quiescence of the Holocene interval (approximately the last 10,000 years) relative to the preceding glacial period. The Holocene coincides with the removal of the Laurentide and Fennoscandian ice sheets. Age model is by layer counting back to about 30,000 years and is model-based prior to that time. The range of excursion is believed to correspond to about 15àƒ⣃”‚¬’€à‚⥃.

Wunsch states in his introduction:

Given the implications for modern public policy debate, and the use of this interpretation of D-O events for understanding of past climate change, it is worthwhile to re-examine the elements leading to the major conclusions. Underlying the now very large literature of interpretation are several assumptions, assertions and inferences including:(1) The àƒÅ½à‚ⲱ8O variations appearing in the record of Fig. 1 are a proxy for local temperature changes.(2) Fluctuations appearing in Greenland reflect climate change on a hemispheric, and probably global, basis and of large amplitude.
(3) The cause of the D-O events can be traced back to major changes (extending to “shutdown”) of the North Atlantic meridional overturning circulation and perhaps even failure of the Gulf Stream.(4) Apparent detection of a D-O event signature at a remote location in a proxy implies its local climatic importance.
The purpose of this paper is to briefly re-examine these assumptions and assertions, but with emphasis on (2) and (3). A summary of the outcome of the survey is that (1) is in part true; little evidence exists for (2) other than a plausibility argument; and (3) is unlikely to be correct. Inference (4) can only be understood through a quantitative knowledge of controls of local proxies, and is briefly likened to the problem of interpreting modern El NiàƒÆ’à‚Ⱟ signals. The paper ends with a discussion of how to move forward.

All in all, an interesting read.


  1. TCO
    Posted Aug 20, 2005 at 7:42 AM | Permalink

    Do you ever see interesting papers from the “other side”? Ones with new insights or that tighten experimental methodologies, while lending more strength to the AGW argument? Can you please bring them forward as well? I would think that you have spent so much time digging through the literature, and since you are a smart guy to start with, you would be able to do this type of detection easily.

  2. Steve McIntyre
    Posted Aug 20, 2005 at 8:28 AM | Permalink

    I’ll keep an eye out. The only literature that I know really well is the proxy literature. I posted up on Moberg a number of months ago. Take a look at that. If the multiproxy premise were correct, then tree rings in the 1990s would be world-record thicknesses: they aren’t. If I see a proxy study that seems to go the other way, I’m usually much more interested in it than say Mangini, which showed a strong MWP and I try to write up a blog note if I see one. I actually want to know what’s going on.

    However, I’ve noticed that the proxy arguments have now changed from tree rings to glacier retreat. I think that the silence from the tree ring people is deafening. For example, Hughes took new samples from Sheep Mountain bristlecones in 2002 – if these ring widths had been off the charts, do you seriously believe that Hughes wouldn’t have rushed into print with the results? Because the measurements are still being studied, I’ll bet that they aren’t off the charts. Same with Thompson’s Puruogangri. You learn this stuff in mining promotions.

    I’ve started collecting some information on glacier retreats. There’s certainly some evidence that, even in the Holocene, the present glacier retreat is not exceptional. I don’t promise to go at glacier retreat thoroguhly at present because I’m still working on tree rings.

    As to more general issues like climate models, someone needs to be going through them in detail, but I’m already at full stretch and can’t do it. But I will be even-handed about proxy literature.

  3. Jeff Norman
    Posted Aug 20, 2005 at 8:31 AM | Permalink


    Interesting article. I won’t pretend to understand much of it.

    I found Figure 12 which shows the retreat of the NH ice sheets complete ~8,000 years ago different from other sources which show Quebec to be still covered by ice at this time.


    There seems to be considerable uncertainty in knowing when the ice sheets actually retreated.


  4. Paul
    Posted Aug 20, 2005 at 10:18 AM | Permalink

    I can’t find the url, but a it’s been pointed out that the glaciers in the Alps all have trees embedded in them. Since trees don’t move, it means that they originated at higher elevations in areas now covered by ice. In other words, the glaciers have been much smaller than their current extant or even non-existent sometime after the last Ice Age. Either way, earlier times were quite a bit warmer than today. Using C14 to date the trees would provide some very interesting information.

  5. ET Sid Viscous
    Posted Aug 20, 2005 at 10:53 AM | Permalink

    This site is pretty much devoted to the (in)famous Hockey stick, which is the single most *interesting paper from the “other side”*. So you could say that this site is almost completely about an *Interesting paper from the “Other side*, with a dew dalliances into side issues. But the simple fact of the matter is that there are few interesting papers from the other side, should you want to examine said papers at Mr. Milloy’s site spends plenty of time linking to articles about them. The most recent, the UAH study, was the supposed “Smoking gun” (Yet again they find supposed smoking guns every two months or so, mostly so that they can say AGW is “definitely proven now, the skeptics can’t refute this” usually the papers are less then significant) When in reality yet again it was anything but a smoking gun. There numbers still didn’t go far enough to show the proper atmospheric temperature to ground temperature correlation required to be an indicator (far from a smoking gun) of AGW.

    So maybe Mr. McIntyre doesn’t find articles that are re-hashes of old data interesting. Or articles that say “Look it’s 2:00 it must be because of Global Warming” “Look I saw an otter, must be global warming” Look it’s 87.2 degrees outside, must be global warming”*.

    Maybe if the “other side” would actually put out some interesting papers.

    * These are my favorites. When we get record cold temperatures, or snow in May, something like that, AGW types like to say, That’ weather not climate. As soon as we get a record high temp day they start screaming about melting glaciers and Global Warming, and omg we’re all going to die.

  6. John A
    Posted Aug 20, 2005 at 10:55 AM | Permalink

    Re: #4

    We’ve already mentioned the “Green Alps” theory here

  7. Peter Hearnden
    Posted Aug 21, 2005 at 1:57 AM | Permalink

    Re #4. “I can’t find the url, but a it’s been pointed out that the glaciers in the Alps all have trees embedded in them. Since trees don’t move, it means that they originated at higher elevations in areas now covered by ice.”

    Does it? Why? Might the trees have fallen onto the ice as the glacier moved to a lower level?

    Does the idea of trees growing at the same time as an ice field make sense? Clearly not. So you’re saying the trees were there, died, didn’t rot but became embedded in the developing ice field? It’s possible, but it’s not the certain explaination for present day trees found in ice – surely?

    So, what’s the scenario you’re painting?

  8. John A
    Posted Aug 21, 2005 at 2:01 AM | Permalink

    Does it? Why? Might the trees have fallen onto the ice as the glacier moved to a lower level?

    It’s statements like these that cause me to wonder which planet Peter Hearnden is on. Really.

  9. Peter Hearnden
    Posted Aug 21, 2005 at 2:41 AM | Permalink

    Mr ‘A’, so glaciers don’t move downhill, glaciers in New Zealand and Patagonia don’t reach down into forests well before they melt? Do try to think about what I’m trying to ask (seeking clarification really, I asked several questions to this end) instead of reaching for the insult button *every* time – it’s pretty tedious and rather demeans you tbh.

  10. fFreddy
    Posted Aug 21, 2005 at 3:29 AM | Permalink

    Peter, I’m confused. Are you saying (#9) there are glaciers with living forests embedded in them ?

  11. Peter Hearnden
    Posted Aug 21, 2005 at 3:51 AM | Permalink

    Re #10, Honestly Ff’, what do you think? That’s an ad hom at it’s crudest. No, of course I don’t think forests grow in glaicers. Do you??? Of course you don’t! What I’m trying to get at is why and how people (I actually asked Paul…) think these dead trees get into the ice. That’s all.

    It was Paul who said ‘the glaciers in the Alps all have trees embedded in them’ why not get at him in the same way? I’m not making statements just asking questions about the how, and if the reason Paul gave is necessarily the only possible expalination.

    Now, thankfully, I have little time today, so anyone hopeing for a prolonged flame is going to be unlucky!

  12. fFreddy
    Posted Aug 21, 2005 at 5:07 AM | Permalink

    Peter, honestly, that was not an attack; I was just trying to work out what you were saying.
    The original reference to the “Green Alps” (by Paul in #4) was referring to the idea that the glaciers spend long periods at a much higher altitude than today – long enough that forests are able to grow up at the edge of the glaciers. When the glaciers expand, they over-run the forests, so there are (dead) trees embedded in the glaciers.
    Further, carbon dating seems to show very distinct phases of tree growth, which seems to me to be very good evidence of long-term cyclical variation in glacier length, and hence some function of precipitation and temperature.
    The Der Spiegel article which talks to the chap from the University of Innsbruck who is digging this out is here. There are some other interesting links in the post John A links to in #6. Seriously, when you have some time, go have a look, it is interesting stuff.

    With regard to your comment in #9 about “glaciers in New Zealand and Patagonia …[reaching]… into forests well before they melt” – well, I’m still confused.

  13. Jeff Norman
    Posted Aug 21, 2005 at 6:05 AM | Permalink


    I too do not understand your comment/question. Could you re-state it?

    Glaciers cannot pull dead trees up hill. If there are dead trees being revealed by retreating glaciers then they must have been pushed there when the glacier originally expanded. It is a rare tree that gets buried in a glacier. Most loose materials like plants, buildings and top soil end up in the terminal moraine. Most of the organic materials in the terminal moraine decays naturally. Only the material trapped in ice is preserved.

    The dead trees can be carbon dated to ensure they are not from some previous interglacial period prior to the Holocene.

    Sometimes these trees are sampled to get tree ring series. I have read about samples from hemlock trees recovered from retreating glaciers in the Cascades being used to push back the proxy record prior to 1,000 AD.


  14. Steve McIntyre
    Posted Aug 21, 2005 at 6:58 AM | Permalink

    In the Sierra Nevadas, the highest stumps date from the Holocene Optimum ~6000 BP. They last a really long time without rotting and I suspect that the same is true for conifers in the Alps.

    The stumps provide a date prior to the glacier expansion to the altitude of the stumps. For example, let’s suppose that glaciers developed in the Sierra Nevadas in the 23rd century to engulf the Holocene Optimum bristlecone and foxtail stumps (but not the lower modern trees),and then retreated, disgorging the stumps in the 30th century. Would the 30th century paleoclimatologist date the glacier to the Holocene Optimum?

    A related point: LIA moraines are the lowest since the Last Glacial Maximum (that’s why they called it the Little Ice Age originally, the term came from people studying glaciers). If a modern glacier retreat disgorges something from the Holocene Optimum, can we really conclude the glacier has retreated to its highest level since the Holocene Optimum? Perhaps the LIA glacial advance picked up something from the Holocene Optimum that had not been touched by previous Holocene glacial advances (picture a Holocene Optimum bristlecone or foxtail pine).

  15. Dave Dardinger
    Posted Aug 21, 2005 at 7:59 AM | Permalink

    I don’t like supporting Peter on anything, but I do think we have to be careful about the Green Alps and similar theories. Glaciers tend to run along the bottoms of valleys and there are slopes on either side which are likely to have trees on them. These trees may fall into the valley for various reasons including land-slides, rockfalls, erosion, etc. And most glaciers at the bottom are a combination of a number of separate streams of ice. Thus even if a tree is found in the middle of a glacier it doesn’t mean it can’t have been deposited from a slope far up the hills and still not indicate anything about whether or not it was colder or warmer than it is now when it began growing.

  16. Paul
    Posted Aug 21, 2005 at 8:45 AM | Permalink

    #15 Dave and Peter, fair enough, it’s a possible explanation of how the trees got into the glaciers. A measurement of the distribution of trees across the glacial front could decide the argument. Mostly at the edges you’re right. Uniformly across the front or peaked to the center, the rest of us are right.

    Let’s ask a similar question: How did the mastodons get frozen into the Siberian tundra? No glaciers moving downhill there and the animals were herbivores that wouldn’t have hung out where there was no vegetation, i.e., ice fields. Isn’t the simplest explanation a warmer climate very suddenly turning cold and staying that way?

  17. fFreddy
    Posted Aug 21, 2005 at 8:56 AM | Permalink

    Steve and Dave, yes, both fair points. At worst, the most you can say is that these trees that are being revealed originated somewhere uphill from here, although it would be interesting to analyse a map of where trees were discovered, and what their carbon dating shows.
    But it seems to me that the key point is that the carbon dating shows that the trees grew and died at (vaguely) regular intervals, indicating that there is some sort of local extinction event that keeps happening there. The idea that the glaciers keep expanding and contracting is only one hypothesis, but it strikes me as a pretty good one.
    Any obvious alternate hypotheses that I am missing ?

  18. John A
    Posted Aug 21, 2005 at 12:33 PM | Permalink

    My point about Hearnden was that trees do not grow above glaciers anywhere in the world. The only way that glaciers could transport trees would be if the trees grew up at higher elevations before glaciation came along.

  19. Armand MacMurray
    Posted Aug 21, 2005 at 1:54 PM | Permalink

    John A — you might want to check out photos like the following:
    Those look like trees on the bluff above the glacier. 🙂

  20. John A
    Posted Aug 21, 2005 at 2:20 PM | Permalink

    Re: #19

    I’ll have to take your word for it. I can definitely see trees silhouetted in front of the glacier terminus. However my point is, trees don’t grow above glaciers, and even if they did how would they manage to plant themselves the right way up underneath the glaciers if they (hypothetically) fell on to them?

    The point about the “Green Alps” theory is that the remains of entire forests that previously grew in situ have been revealed by glacial retreat high above the current treeline, indicating that sometime in the past the treeline was much higher than it is today and the glaciers did not exist in that area at that time the forests were alive.

  21. John G. Bell
    Posted Aug 21, 2005 at 4:04 PM | Permalink

    Could it be that the main impediment to the rise of civilization was that the usual situation
    exposed humans to radical climate change at a frequency that didn’t allow the evolution of a
    culture that could survive the change? That we have lived in a period unique for relatively
    continous mild weather? I find it hard to believe that humans 40,000 years ago were so
    biologically dissimilar to people of today that, had the climate cooperated, they couldn’t
    have put a man* on the moon back then. My world outlook makes it hard for me to understand
    the mindset of the climate change people. Is anyone else on this boat?

    *2.Any human being, regardless of sex or age: a member of the human race: a person.

  22. TCO
    Posted Aug 21, 2005 at 4:53 PM | Permalink

    Before running amuk with stuff that is at such a casual level of sophistication, might be useful to hear the more experienced glacialists and what they make of vegetation finds.

  23. JerryB
    Posted Aug 21, 2005 at 5:07 PM | Permalink

    FWIW: a brief article on a receding glacier in Iceland and uncovering old farmland:


  24. ET Sid Viscous
    Posted Aug 21, 2005 at 5:36 PM | Permalink

    I apologize, not on-topic, except in a larger sense. But I found some of the data extremely intersting and figured others here might too.

    I’ve always wanted to see an overlay of solar output and surface or atmospheric tempratures. This isn’t but for those that have seen enough 20th century temprature graphs a mediocre alignment should be possible.


  25. Paul
    Posted Aug 21, 2005 at 5:42 PM | Permalink

    If you look at http://www.klimanotizen.de/html/newsletter_4.html p13-15 they find that the Great Aletsch Glacier in the Alps has now retreated back to about where it was in the years 800 and 100. They use tree stumps(!) to mark advance and retreat. “Only roots, stumps and trunks of trees found in the original place of growth (in situ) are suitable for glacier fluctuations, because complex transportation paths from the place of growth to the place of reappearance are of no concern.”

  26. Peter Hearnden
    Posted Aug 22, 2005 at 1:21 AM | Permalink

    Re #15, well, of course, I don’t like supporting Dave on anything… but…he’s right. To see a glacier that terminate in warm temperate rain forest see here – http://www.forestpark.co.nz/glacier.asp . That said, I don’t dispute glaicers have advanced and retreated (I’ll have to transleante the report in post #25, but I also accept there is evidence Europe has been as warm (perhaps a little warmer, perhaps not) as now during the holocene), I just don’t think how tree parts get into glaciers is certainly just that given in post #4. I do think it’s clear glaciers are in the process of a massive globalwide retreat atm.

  27. stephan harrison
    Posted Aug 22, 2005 at 2:26 AM | Permalink

    Dear All
    re the arguments about how trees can become incorporated into glaciers. In Patagonia most of the glaciers run through valleys whose sides are covered with trees (and are therefore at higher altitude than the glacier surfaces) and tree trunks are regularly found on the glacier surfaces. Because of this, we only use in situ trees for dating purposes (ie trees which have been overridden by glacial ice whilst growing and which have been uncovered by ice recession).

  28. fFreddy
    Posted Aug 22, 2005 at 4:45 AM | Permalink

    Stephan, do you happen to know if the pattern of very discrete phases of tree growth, referred to in the Der Spiegel article, is replicated in Patagonia, or anywhere else outside Europe ? Or is there insufficient data ?

  29. Steve McIntyre
    Posted Aug 22, 2005 at 6:47 AM | Permalink

    Re # 27: Stephan, thanks for dropping by, I’ve read a number of your articles and found them very interesting. I’ve been to Chile quite a few times, but mostly to see mineral exploration properties around the Vallenar area. I haven’t been to Patagonia. I wish that there would be about 10 times as much attention spent on Southern Hemisphere paleoclimatology as Northern Hemisphere for a while. Aside from the discrepancy in accumulated data, it strikes me that the “noise” in the Southern Hemisphere is much less. Cheers, Steve

    For other’s here is Stephan’s webpage http://www.geog.ox.ac.uk/staff/sharrison.html.

  30. stephan harrison
    Posted Aug 22, 2005 at 6:47 AM | Permalink

    Dear fFreddy
    I’m afraid I don’t. Most of my work has been on recent glacier variations (Little Ice Age) or much older ones (outside the range of dendrochronology). In Patagonia, the topographic settings of the glaciers generally do not allow for the preservation of long dendro records. There is a lot of work from the Canadian Rockies (by Brian Luckman amongst others) which may be of interest.
    Best wishes,

  31. John A
    Posted Aug 22, 2005 at 6:59 AM | Permalink

    I take it back. Trees do grow on valley sides that have glaciers in them. Trees do not however, grow underneath glaciers.

    Oh, and Peter, the Franz Josef glacier in NZ is currently growing longer.

  32. TCO
    Posted Aug 22, 2005 at 8:01 AM | Permalink

    Steve: could you characterize how the use of vegative finds intersects with ice core dating and paleoclimatology in general? What corrective (or additive) value to proxy dating by glacialation does vegatative finds allow and is this accepted (or if debated, what is status of debate) in general scientific community?

    By in situ, do you mean “roots still in the ground”?

  33. stephan harrison
    Posted Aug 22, 2005 at 8:38 AM | Permalink

    Re #29. Thanks for your kind words. I agree with you about Southern Hemisphere climate and the relative neglect in comparison with NH….I guess the absence of proxy data is a problem, as is the difficulty of access to some sites. Ice core records from the Patagonian icefields are too short to be much use because of very high accumulation rates, and the proxy data that do exist contradict each other! (especially the pollen and glacier variation data).

    Re #32. I’m not sure if this is addressed to Steve McIntyre or me, but the use of vegetation remains for reconstructing glacier variations in certain areas of the world is generally seen as very valuable, if only because there isn’t much else to go on (especially for the late Holocene and in areas where there are no historical records). The problem with trees is that C14 dating doesn’t give you a refined enough date to really intersect with the ice core stuff….and is usually only of use from the middle part of the Holocene. in situ means that we can demonstrate that the tree has remained in its original growing place, and this usually means that it still has some roots in the ground.
    Hope this helps.
    Best wishes, Stephan.

%d bloggers like this: