Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

The problems of the isotope paleo thermometer

  1. Sep 10, 2012 #1
    In some threads around here I have hinted to problems with the use of 'water' isotope ratio's (δ18O, δ2H aka δD) as thermometer of past temperatures. This could become a dense thread, so I'll give the conclusion first:

    The isotope ratio's in ice cores and other records of meteoric water are mainly proxies for absolute humidity at the water source.

    That's a bummer since the isotopes are used a prominent tool to reconstruct paleo climate. Also it took me years to find this out, while the actual principle is so simple.

    No this is not a new theory, it's just testing an existing theory, looking how paleo climatology and hydrography compare to basic meteorology. And actually combining some threads in the first page of this forum also gives this result.

    The theorie for using isotopes as temperature proxies is wrapped up in Jouzel et al 1997.

    Actually we could also keep it quite simple. The most important sentence in Jouzel et al is (page 46,481):
    Cuffy et al 1995 state (page 455):
    For the moment I don't think we have to dig deeper, because I have not found any isotope -temperature study that elaborates on cloud temperature. But we have a thread here about that.

    To be continued soon
  2. jcsd
  3. Sep 10, 2012 #2
  4. Sep 10, 2012 #3
    Meanwhile, here is a comprehensive textbook about the isotopes in the water cycle. Well worth reading if you have a few hours to spare.

    So how about the clouds?

    But that temperature is the dew point

    The plot is tightening. Well I'm just saying that to encourage you all to crush this stuff, because the outcome is important.

    to be continued.
  5. Sep 11, 2012 #4
    So there we have it:


    So what is causing....
    That's the absolute humity that fixes condensation and hence cloud temperature.

    So if we look at Jouzel et al (1997) fig 4

    https://dl.dropbox.com/u/22026080/jouzel-fig4.png [Broken]

    We are looking at the comparison of snow accummulation and isotope-reconstructed temperatures of the GISP-2 ice core in Greenland. It reflects a most interesting timeframe from right to left the (cold or warm?) mystery interval until 14.5 thousand years, the (warm?) Bolling Allerod until 12,7 thousand years, the (cold?) Younger Dryas until 11.6 thousand years and finally the (warm?) Holocene.

    But with such an extreme tight correlation, shouldn't the warning flags be out? Aren't we looking at twice the same? Snow accumulation being a function of absolute humidity and Isotope-cloud temperature being a function of absolute humidity?

    In such cases it may help to consult other data
    Last edited by a moderator: May 6, 2017
  6. Sep 11, 2012 #5
    Hey, I'm challenging a major fundamental idea in paleo-climatology here, discussion is allowed.
  7. Sep 12, 2012 #6
    So, if you'd accept that the Greenland Ice core graph just reflects absolute humidity both by direct precipitation and isotope dew point reconstructions, the Younger Dryas was just very dry/arid and cloudless skies in combination with an approaching solar summer insolation maximum caused mild temperatures, while during the wet Bolling Allerod which preceded it, it was also cool due to the more prevailing cloud cover. So all of sudden there is no anomaly at all and there is no need to model problems away.
    Last edited by a moderator: May 6, 2017
  8. Sep 14, 2012 #7
    Another highly detailed record of the same period can be found in the pollen counts of the sediments of the Meerfelder maar volcanic pond/lake in central Germany (Lücke and Brauer 2004)

    The abstract:

    This is the pollen diagram they reconstructed:

    https://dl.dropbox.com/u/22026080/meerfelder.gif [Broken]

    As trees and grasses usually use wind for pollination, their pollen is most abundant. We see a distict shift in both total pollen count and tree/grass ratio throughout the Younger Dryas. Apparantly the trees reduced during the Younger Dryas, while the grassed (gramineae) held up better. It should be noted that both facts could be explained by aridity. Steppes are generally dry, forests are usually wet. And under arid conditions the pollen production may reduce, (ask the hay fever sufferers)

    Especially interesting are the water plant/algae fern swamp pollen Potamogeton, Botrychium and Pediastrum suggesting swampy conditions, or a low stand of the lake, consistent with aridity while it could not really have been arctic.

    But the relative abundance of meadow flower species like Helianthenum is interesting. Since these are not polinated by the wind, these pollen are naturally rare. These species are still common there today, moreover there are no clear arctic pollen around like the Dryas octopetala, famous name giver for the period.

    So it looks like these species are a good reason for the authors to talk about warm summers during the younger dryas, but also more arid as suggested by pollen shift and abundance.

    How about the winters? Not a lot to say about that except that the Greater Burnet (Sanguisorba Officinalis) is limited to winter hardiness zone 4 and Potamogeton to zone 5. Cold yes, maybe, but not too extreme.

    So it looks like these pollen reflect clear changes in moist-arid conditions but an extreme temperature swing like this cannot be considered supported here, imo.
    Last edited by a moderator: May 6, 2017
  9. Sep 15, 2012 #8
    This thread could go on for a lot more information, we could go for evidence of cold events in the allegdly warm Bolling Allerod and the onset of the Holocene, the Preboreal. We could also dig deeper in the isotope physics, and see if we can find more evidence in Meteoric Water Lines and Deuterium Excess

    And after all that we could ask questions about the consequences of all this for the reconstruction of the geologic past.

    But is it useful?
  10. Oct 2, 2012 #9
    I drafted a little study about all this. I don't think I should post an open link to it yet. But anybody who wants to have look at it, is welcome, Just drop me a pm.
  11. Oct 3, 2012 #10
    The study is work in progress, draft number 0.2.

    I found some pretty convincing support recently, for instance this:


    Suggesting that the timing may be right to discuss things.

    Meanwhile this is the abstract:

    Just to make sure:
    Last edited: Oct 3, 2012
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook