Betting on sea ice: following the herd

Posted by William M. Connolley on December 3, 2007
(25)
More »

Just a post-of-record, following Brian Schmidt, that I have a $333 bet against Joe Romm, with me taking the “cold side” of:

at no time between now and the end of the year 2020 will the minimum total Arctic
Sea ice extent be less than 10% of the 1979-2000 average minimum annual Arctic Sea ice extent, as measured by NSIDC data or any other measurement mutually agreed-upon; provided, however, that if two or more volcanic eruptions with the energy level equal to or greater than the 1991 Mount Pinatubo shall occur between now and the end of 2020, then all bets are voided.

I’m not up to my limit on this one yet, and I know James isn’t, so there could be more on offer if there are any takers?

See-also a bet on next years ice.

(25)
More »

Comments

  1. #1 Eli Rabett
    2007/12/03

    What is that, the number of the half beast??

    [We only do things by halves here :-) -W]

  2. #2 crandles
    2007/12/03

    ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Sep/N_09_area.txt

    gives an average of 7.03mKm^2 for 1979 to 2000 average September ice extent. However I suspect from what you have written you are working on minimums (five day?) rather than average September.

    Where do I find that data?

    Does it indicate anything much different from:

    Required drop 4.28-.1*7.03= 3.58mKm^2 over 13 years
    = 0.023m Km^2 per month

    I calculated a rate for Nov 2003 to Dec 2006 of 0.0135m Km^2 per month which was about 4.5 times faster than 1979 to Dec 2000. 2007 does not make it look like that acceleration has come to an end, though 2007 probably is mainly synoptics. A further 1.7 fold increase in rate does not look outrageous on top of the 4.5fold increase in rate.

    My reaction is therefore this is too close to try to call it or want to risk my money on.

  3. #3 crandles
    2007/12/03

    ftp://sidads.colorado.edu/DATASETS/NOAA/G02135/Sep/N_09_area.txt

    gives an average of 7.03mKm^2 for 1979 to 2000 average September ice extent. However I suspect from what you have written you are working on minimums (five day?) rather than average September.

    Where do I find that data?

    Does it indicate anything much different from:

    Required drop 4.28-.1*7.03= 3.58mKm^2 over 13 years
    = 0.023m Km^2 per month

    I calculated a rate for Nov 2003 to Dec 2006 of 0.0135m Km^2 per month which was about 4.5 times faster than 1979 to Dec 2000. 2007 does not make it look like that acceleration has come to an end, though 2007 probably is mainly synoptics. A further 1.7 fold increase in rate does not look outrageous on top of the 4.5fold increase in rate.

    My reaction is therefore this is too close to try to call it or want to risk my money on.

  4. #4 Gareth
    2007/12/04

    I think I’d want to wait until next year before making or taking a bet of that sort. If William wins his bet about next year, but only just – in other words, there is no significant bounce back from this year’s record (which is what I expect), then the recent trend will be a lot steeper than for the full 79-08 period, and Joe’s position will look better.

    And crandles, what if the synoptics are the warming signal…?

    Someone just emailed me this NOAA current temp anomaly plot. It’s very warm up there…

  5. #5 crandles
    2007/12/04

    >”And crandles, what if the synoptics are the warming signal…?”

    There may well be a warming signal in the synoptics. But even assuming that, I doubt that the synoptics will be as extreme (or worse) than 2007 in most years. It seems more plausible that it is due to oscilations and I don’t think it likely that there is anything near certainty one way or the other whether there is a warming signal in the synoptics. Having said this, perhaps the ‘probably’ that I used did indicate too much confidence.

    Waiting til next years data before making such a bet does sound sensible though I wonder what odds you and William would want to bet on the 2009 mimimum being lower than the 2007 minimum?

    Your expectation will make Joe’s position look better. Hmm. what does that mean? Your expectation is different from what you think the typical expectation is? (room for a bet in this difference?) Or is it to do with skewed distributions of expectations? Or something else?

  6. #6 outeast
    2007/12/04

    Hmm. Brian Schmidt says that if the term were 2050, not 2020, he’d be on the other side of the bet. What about you, William?

    [I wouldn't bother, because I'll be senile by then -W]

    I might be interested in going in on this bet, though. Not with a great degree of confidence, I’ll admit…! So probably a smaller sum. I’ll have a think on it.

  7. #7 Gareth
    2007/12/04

    Your expectation will make Joe’s position look better. Hmm. what does that mean? Your expectation is different from what you think the typical expectation is? (room for a bet in this difference?) Or is it to do with skewed distributions of expectations? Or something else?

    The huge drop in ice extent this last summer is so far off the bottom of the long term trend line that it looks like a step change to me (and similar drops in recent modelling (Bitz et al) are steps not blips). William wins his original bet if 2008 is not a new record, but if 2008 (and 2009) show a return towards that long term trend, then it’s probably not a step change, and the “cold side” of the Romm bet looks good.

    I still think I have a good chance of winning my bet, however. Although the current extent at CT is slightly greater than at the same time last year, the area of first year ice is much bigger. Warm spells in winter won’t have much impact on extent because a positive anomaly of 20C (see link above) still leaves temperatures below freezing, but it will mean thinner ice. When the spring melt starts, it could go very quickly. For William to win, we will need a succession of months when the synoptics deliver colder than (recent) average weather. To get back to the long term trend, we’ll need more like years. My expectation is that that is unlikely.

    If I win my bet, then Romm’s position looks good. On my calc of the figures (from NSIDC), for Romm to win, there has to be a drop in extent of 0.275m km^2 every year to 2020. Over 2005-7, the drop was 0.645m km^2, but three summers do not a trend make.. ;-)

  8. #8 uBeR
    2007/12/04

    At $333 by 2020, with the fast decline of the dollar, it’s safe to say no one will be too worried about winning or losing this bet, except of course for the real-world ramifications of ice extent.

  9. #9 Steve Reynolds
    2007/12/04

    >I’m not up to my limit on this one yet, and I know James isn’t, so there could be more on offer if there are any takers?

    If there are any takers that want to go above your limit, I’ll take some on your side.

  10. #10 outeast
    2007/12/05

    Ok, William; I’ll bet you EUR 150 – indexed to the HICP (or whatever Eurozone consumer price index should supercede it in the meantime) so that the impact of inflation is minimized. Same bet as you have with Joe Romm (as detailed above), for simplicity’s sake: I figure that even odds actually gives you the better side of the bet, but what the heck. Up for it?

    [In principle yes. Are you verifiable? -W]

  11. #11 Nick Barnes
    2007/12/06

    I think you have probably got the right side of the bet on next year’s ice, for the reason you indicate, but if it’s still open then I’ll take £10 worth, just for interest.

    [OK -W]

    Your bet against a blue arctic by 2020 is a bit past my time horizon, but I might go for the warm side of a wager on melting by 2012.

    [You can if you like, but are you serious? An essentially ice-free summer Arctic in 5 years? And you want the warm side? -W]

    Over at RC, I tried to engage Jim Cripwell in a bet for money that the 2008 minimum would be lower than *2005* (4.01 Mkm^2): he plainly doesn’t believe it will be, but isn’t prepared to bet money so the stake is now just our reputations. Joe Duck took me up at 50 euros, though.

    http://www.realclimate.org/index.php/archives/2007/11/les-chevaliers-de-lordre-de-la-terre-plate-part-i-allgre-and-courtillot/

    Generally, the more outspoken deniers won’t place or take either side of any bet. Possibly they don’t believe their own rhetoric. Whatever the reason, this leaves the mainstream betting against the catastrophic warmers.

  12. #12 Brian Schmidt
    2007/12/06

    Nick, some Gristmill commenters to Romm’s companion piece claimed they’d bet on 2012 and 2015. You might want to go after them.

  13. #13 Nick Barnes
    2007/12/07

    W: you misunderstand me, because I express myself poorly. By “a wager on melting by 2012″, I mean a wager on the *amount* of melting by 2012. Say: that we see a sea ice area of less than 2.5 million square kilometres before the end of 2012.

    There are probably some who would put money on a blue arctic by 2012, but that’s too rich for me. 2008 will tell how rich.

  14. #14 Brian Schmidt
    2007/12/07

    “Generally, the more outspoken deniers won’t place or take either side of any bet. Possibly they don’t believe their own rhetoric. Whatever the reason, this leaves the mainstream betting against the catastrophic warmers.”

    It’s certainly true that there have been more bets among the non-skeptics than with the non-skeptics. Total money bet so far, though, is higher with the skeptics.

    And Nick’s also right that the tiny number of skeptics who will put their money where their mouths are, aren’t the really loud ones.

  15. #15 crandles
    2007/12/10

    I don’t understand why William isn’t trying to arrange a bet with Nick Barnes on the less than 2.5 million km^2 before end of 2012. Comparing with the two bets William is making, I would have thought this was less likely to be lost by William. To expect a rebound this year, rapid melt for 5 years then much slower melt for 8 years may not be impossible but to the extent involved does seem odd.

    Are there negotiations by other means or is Nick’s 2.5 million just an example rather than an offer?

    [Nick wasn't very pressing, and I didn't stop to consider it very carefully. I don't even know what fraction 2.5M would be -W]

  16. #16 crandles
    2007/12/10

    Compared to 7.04M per NSIDC for average Sept 1979-2000, 2.5M is 35.5% left 64.5% gone. Or, to compare melt rates .356M Km^2 per year compared to .275M Km^2 for your 2020 bet. Nearly 30% faster. I could understand expecting 30% faster for years 4 to 9 of 13 but not for years 1-5 when you expect a rebound this year.

  17. #17 Steve Bloom
    2007/12/11

    A couple of the AGU sea ice abstracts (pasted below) include some pretty bold talk by some pretty big names in the field. The first one says that yes indeed, it is a big step change and describes the mechanism, and the second, while it has a less exciting title, has am even more exciting conclusion. Note that the authors of the second paper have access to the still-secret portion of the ice thickness data gathered by U.S. subs. We are left to wonder as to the meaning of “near future.”

    C22A-03

    A Younger and Thinner Multiyear Ice Pack: Significance for Extreme, Accelerated and Sustained Losses of Arctic Sea Ice Cover

    * Maslanik, J A (james.maslanik@colorado.edu), University of Colorado, Colorado Center for Astrodynamics Research, University of Colorado, CCAR, 431UCB, Engr. Center, North Tower, Boulder, CO 80309, United States Fowler, C (cfowler@colorado.edu), University of Colorado, Colorado Center for Astrodynamics Research, University of Colorado, CCAR, 431UCB, Engr. Center, North Tower, Boulder, CO 80309, United States Stroeve, J (stroeve@kryos.colorado.edu), University of Colorado, Cooperative Institute for Research in Environmental Sciences, University of Colorado, CIRES, 449UCB, Boulder, CO 80309, United States Drobot, S (sheldon.drobot@colorado.edu), University of Colorado, Colorado Center for Astrodynamics Research, University of Colorado, CCAR, 431UCB, Engr. Center, North Tower, Boulder, CO 80309, United States Zwally, H J (zwally@icesat2.gsfc.nasa.gov), National Aeronautics and Space Administration, Laboratory for Hydrospheric Processes, NASA/Goddard Space Flight Center, Code 614.1, Greenbelt, MD 20771, United States Yi, D (donghui@icesat2.gsfc.nasa.gov), National Aeronautics and Space Administration, Laboratory for Hydrospheric Processes, NASA/Goddard Space Flight Center, Code 614.1, Greenbelt, MD 20771, United States

    A satellite-derived record of sea-ice is combined with ice thickness estimates from the ICESat satellite to investigate the role of large-scale ice transport in modifying the distribution of ice age and thickness within the Arctic Ocean. These data show that in addition to the well-documented loss of the perennial ice cover over the last several years, the amount of the oldest and thickest ice within the remaining multiyear ice pack has declined significantly, and is now found over a much smaller area of the Arctic Basin. Ice greater than 5 years old covers 56 percent less of the Arctic Ocean than in the early 1980s, and the majority of the remaining perennial pack now consists of ice 2 to 3 years old. Locations of thinning due to this transition to younger ice have shifted from the Siberian Arctic to the western Arctic Basin, consistent with large-scale ice transport patterns. Multiyear ice is no longer surviving the typical east-to-west transit through the Canada Basin. As a result, the clockwise circulation of the Beaufort Gyre, rather than allowing ice to age and thicken for several years within the Arctic Basin as in the past, is now contributing to the loss of the oldest ice. The resulting younger and thinner ice pack is predisposed toward rapid, extensive and persistent reductions in ice extent. This is consistent with the extreme retreat of the ice pack underway this summer that is unprecedented over the available record of sea ice data, with August ice extent 15 percent below the previous observed minimum.

    C22A-06

    Understanding Recent Variability in the Arctic Sea Ice Thickness and Volume – Synthesis of Model Results and Observations

    Whelan, J (jwhelan@nps.edu), Naval Postgraduate School, Department of Oceanography 833 Dyer Road, Monterey, CA 93943, United States * Maslowski, W (maslowsk@nps.edu), Naval Postgraduate School, Department of Oceanography 833 Dyer Road, Monterey, CA 93943, United States Clement Kinney, J L (jlclemen@nps.edu), Naval Postgraduate School, Department of Oceanography 833 Dyer Road, Monterey, CA 93943, United States Jakacki, J (jjakacki@iopan.gda.pl), Institute of Oceanology, Polish Academy of Sciences 55 Powstancow Warszawy, Sopot, 81-712, Poland

    We examine the diminishing sea ice thickness trend in the Arctic Ocean using results from the NPS 1/12-degree pan-Arctic coupled ice-ocean model. While many previous studies have analyzed changes in ice extent and concentration, this research focuses on ice thickness as it gives a better indication of ice volume variability. The skill of the model is evaluated by comparing its ice thickness output to actual sea ice thickness data gathered during the last three decades. This includes the model comparison against the most recently released collection of Arctic ice draft measurements conducted by U.S. Navy submarines between 1979 and 2000. Our model indicates an accelerated thinning trend in Arctic sea ice during the last decade. This trend is robust and independent of timescales for surface temperature and salinity relaxation. The validation of model output with submarine upward-looking sonar data supports this result. This lends credence to the postulation that the Arctic is likely to be ice-free during the summer in the near future.

  18. #18 Gareth
    2007/12/11

    An earlier Maslowski presentation (May 2006 – I’ve got the pdf but lost the link) suggested “within ten years”. That makes Nick’s bet look a little more likely – and Joe Romm should be very happy.

    The polar bears and mustelids might not be.

  19. #19 crandles
    2007/12/12

    Presumably William was aware of the May 2006 “within ten years”. Maybe he wasn’t aware of

    2013 “can argue that may be our projection of 2013 is already too conservative” (Professor Maslowski)
    http://news.bbc.co.uk/1/hi/sci/tech/7139797.stm

    2012 (Jay Zwally)
    http://www.cnn.com/2007/TECH/science/12/11/arctic.melt.ap/index.html

    Any advance on 2012 ;) ????

    Does that reduce the wondering about the meaning of near future?

  20. #20 Gareth
    2007/12/12

    Thanks for the Beeb link. There’s more over at Eli’s warren.

    From the AP story: Still to be released is NASA data showing the remaining Arctic sea ice to be unusually thin, another record. That makes it more likely to melt in future summers. Combining the shrinking area covered by sea ice with the new thinness of the remaining ice, scientists calculate that the overall volume of ice is half of 2004′s total.

    I believe the odds have just shifted significantly.

    [Does that mean you want to up your bet? -W]

  21. #21 Steve Bloom
    2007/12/13

    I’ll say it reduces the wondering! These guys went from reticent to voluble in record time.

    Just to note that the article quotes Mark Serreze to the effect that Maslowski’s results may be too reliant on the short-term trend, but then has Maslowski noting that his model didn’t even incorporate data for 2005-7 (thus the “too conservative” remark). It would be nice to see more specifics.

    The proprietor may wish to note that the Beeb article also quotes Peter Wadhams as predicting a new record for next year (although I suppose it would be hard to not predict a new record for each of the next six years if you think it’s going to be gone by then).

  22. #22 Nick Barnes
    2007/12/13

    crandles: my sketchily proposed bet is on sea ice area, not extent.

  23. #23 crandles
    2007/12/14

    >”crandles: my sketchily proposed bet is on sea ice area, not extent.”

    Oops missed that, sorry.

    Having 5 goes at reducing .27M Km^2 from 2007′s 2.77M Km^2 looks rather likely to me and I would also want to bet on those terms. It possibly gets confusing with the .31M Km^2 area not imaged; does the quoted figure have to get down to 2.19? Perhaps that is why most bets are using extent?

  24. #24 Steve Bloom
    2007/12/14

    Are those area and extent metrics necessarily going to hold together down into the 10% range? I don’t care much relative to my own bet since it’s just for fun, but others may wish to check with NSIDC and UIUC.

    cr, NSIDC has said that they think the area metric is shaky, which is why they use extent instead. I do know that UIUC has to make a lot of corrections, but I don’t know if that’s just in the nature of providing a daily product or because of something more fundamental.

  25. #25 William Connolley
    2009/11/08

    JR recently wrote http://climateprogress.org/2009/11/08/arctic-multiyear-sea-ice-nsidc-david-barber/, so I put in a comment asking if he wanted to increase his bet.