RealClimate logo


Technical Note: Sorry for any recent performance issues. We are working on it.

North Pole notes (continued)

Filed under: — gavin @ 22 August 2008

This is a continuation of the previous (and now unwieldy) post on the current Arctic situation. We’ll have a proper round up in a few weeks.


638 Responses to “North Pole notes (continued)”

  1. 251
    pat n says:

    Re: 238

    Hi Wayne,

    Temperatures at Alaska climate stations through July this year have been below historical (1971-2000) averages. I’ll check the NWS website next week to see how August 2008 compares to recent 30 year averages.

  2. 252
    crandles says:

    re #248
    Robert Matlock

    How good is your late inflection point argument? 2004 had a late inflection point and after the inflection point there was very little reduction in extent. Not very reliable would be my reaction. Also, I would rather trust when the minimum has been rather than when the inflection point might indicate the minimum will be.

    What is special about 9 August? Perhaps I expressed myself badly; I should have said high not good. Anyway just wanted to say that I was deliberately rotten-cherry-picking. If I picked say 1 August as the starting rate then the required rate would have risen from 37.5% to 111% faster than 2007 rate instead of 65% to 111%. The data would be better but it might cause people to wonder if I was cherry picking so making clear I was rotten-cherry-picking seemed appropriate.

    There may well be reason to think the minimum will be later than 2007 and that would be a problem for my simple required rate calculations. There may well be reason to say a new record is more likely than not. 90% sounds overconfident to me. Would you really offer odds of 8:1 or better for a bet?

  3. 253
    Timothy Chase says:

    Re: pat n in 235, crandles in 243

    I must admit that the melt isn’t quite keeping up with my own projections at the moment: it is about a day behind where I would expect it to be. However, there is a great deal of day-to-day variability. I will want to see where we are at on the 16th of September.

  4. 254
    CobblyWorlds says:

    90% chance of a new record?

    From my reading of the situation I agree with Crandles, however my main reason is I don’t see how anyone can predict what is to come due to the state of the ice, this is now in the hands of the weather. So probabilities like 90% seem gross overstatements of confidence to me.

    At present NSIDC is showing a slight flattening, that doesn’t sway me either way.

    #244 Mark,

    Pessimists or realists?

    If we are in a rapid transition to a seasonally ice-free state (I now think we are), then that implies substantial warming of the Arctic. Which will have secondary impacts:

    1) Climatic shift.
    If you warm the Arctic, and (crucially) increase humidity due to higher temperatures and more open water over 0degC, then you will change the relationship between the polar and tropical regions. This changes what happens in between the pole and tropics, i.e. climate in the northern hemisphere. Such changes would not be in the small-scale physics but in the large scale relationships, such as jet stream tracks affecting precipitation. Thinking of the atmosphere in the simplest possible terms – a heat engine – shows how inevitable this is. The impacts are not known but the worst are likely to be changes in timing and amounts of precipitation. But what practical impact will that have overall, on both humans and the Biosphere/Lithosphere/Oceans?

    2) Clathrate outgassing.
    A greater area of open ocean warms the oceans, just look at the
    JAXA EORC plots you can look back to see that warming. Furthermore removing the ice increases storminess (due to water vapour) and storms assist vertical mixing, with less ice this happens over a wider area. There is already evidence of clathrates being potentially unstable and there being more risk of rapid outgassing than land permafrost (Shakhova). But how much methane will be released and how quickly?

    3) Greenland and Sea Level Rise.
    Remove the sea-ice from the North Greenland coast and you will warm the northern flanks of Greenland, which have been cooled by the expanse of ice across the Arctic Ocean. The ice cap has been large enough in the summer to effectively develop it’s own climate, keeping it markedly cooler than in the wider Arctic Basin. Warmer temperatures in the north of Greenland are sure to lead to a greater contribution to sea level rise. How by how much and how fast will sea level rise?

    There are multiple permutations of the further/wider impacts of 1,2 & 3, that may enhance or offset the overall impacts.

    All of the above were risks of climate change anyway, at some future point. If this is a rapid transition then they have just been pegged in our lot: They are no longer possible risks in the future, they are now things that are commencing and once started will proceed.

    Were we looking at a seasonally ice free Arctic in 2050 we’d have time to learn more adapt and perhaps reduce our emissions severely enough to slow the process. If as I now fear we’re looking at an ice-free Arctic by 2018 we have barely any time to do anything before consequences hit. The only attainable option we may have now is to try to avoid a catastrophe by massive emissions reductions, reductions we will have to achieve as we struggle to cope with an unfolding disaster. And that’s ignoring the possibility of wider climatic destabilisation caused by impacts secondary to the Arctic’s 3 secondary impacts above.

    It is quite possible that we have just run out of time.

  5. 255
    Mark says:

    Good for you Gareth.

    Hopefully you’re not too tall for a pony.

  6. 256
    Lauri says:

    # Gareth Says:
    30 August 2008 at 5:43 AM

    > Are there satellite sensors in orbit that can detect methane >blooms, or keep an eye out for the surface signatures?

    >There’s the Sciamachy Envisat data, but that wasn’t showing anything >last year. It would be interesting to see an update.

    >And Mark: if wishes were ponies, I’d have enough to mobilise a tribe.

    I am not downplaying the problems with methane. I only note that the methane emissions will show up in the atmospheric concentration, measured by different organizations. Increase in concentrations pretty much stopped around 2000. We’ll see later this year if the observations on blurbs have any global importance at this time.

  7. 257
    Gareth says:

    Lauri,

    Last year, atmospheric methane began to increase again. The BBC reported NOAA’s 2007 figures here, back in April.

  8. 258
    Robert Henson says:

    Re #232 (Gavin’s reply to my post):

    We’ll fix this in the next printing of the Rough Guide. Thanks for clarifying.

    And kudos to CobblyWorlds (#254) for an excellent summation of what we ought to be thinking about over the next few years as the Arctic melt continues.

    For what it’s worth, I’ll lay 40% odds on this year’s melt surpassing 2007′s . . .

    –Bob

  9. 259
    Hank Roberts says:

    > secondary impacts:

    Don’t forget to get in touch with the biologists who were talking with Dr. Bitz back around the time of her thread here, who know some and expect to rapidly learn more about what part of primary ocean productivity depends on the sea ice cycle.

  10. 260

    #251, Thanks Pat, follows the trend, cooler air less ice!

  11. 261
    Timothy Chase says:

    CobblyWoods wrote in 254:

    90% chance of a new record?

    From my reading of the situation I agree with Crandles, however my main reason is I don’t see how anyone can predict what is to come due to the state of the ice, this is now in the hands of the weather. So probabilities like 90% seem gross overstatements of confidence to me.

    At present NSIDC is showing a slight flattening, that doesn’t sway me either way.

    I agree with your focus on the long range, the methane hydrates and so on, and I think that we could easily focus too much on what happens this year. But at the same time, I think we have entered a new regime of sorts. There haven’t been the big dips and turns that normally occur each year with sea ice extent. So much of it is new, thin ice. So what dips and turns we have seen have been fairly small. And there has been a trend of sorts for the past three months — one which is quite simple — the simplicity of which may be the result of this new regime.

    Here is actual vs. projected:

    Sea Ice Extent Since May 31, 2008
    http://img337.imageshack.us/my.php?image=seaiceextentsince200805cs2.jpg

    The data I based it off of was from IJIS and included their May 31 through August 27, 2008:

    IJIS Web Site: Data of Sea Ice Extent
    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    The last date of actual plotted is August 29, 2008 — which stands out as I omitted August 28, 2008. (However, that isn’t the reason why I omitted the August 28: including it would have meant downloading the data file again.)

    Starting with June 1, 2008, I calculated the sea ice loss for each day relative to the preceding day, then I did a quadratic trendline for the sea ice loss. Then for each day, I project a sea ice extent equal to the total projected sea ice loss since May 31, 2008 beginning with the sea ice extent of May 31, 2008. There is no smoothing of the real data on my part — and the two curves are nearly on top of one another since early June. Based upon this projection, this year’s sea ice extent should fall below that of last year on September 16, 2008 and should reach this year’s record on September 30, 2008: 4,062,663 km^2. (Note: this is a slight correction from my earlier projection — where I had begun with June 1 rather than May 31 sea ice extent — while also beginning with the sea ice extent loss of June 1 — which requires one to begin with the sea ice extent of May 31.)

    Now when I make that projection I am only being half serious. It isn’t based upon any physics, any analysis of the actual conditions, etc. and strictly amateur. But it is a very simple equation that rather accurately captures the evolution of sea ice extent for the latter half of this year so far. The weather has been the tiny wiggles. And the trend has been clear. Now at some point the trend will change, but then the fact that it changes will itself mean that we have learned something new.

  12. 262
    Lawrence Coleman says:

    Re 221 Kent Guy..Exactly wahat I was waiting for..and waiting..and waiting..A bit of ‘Anger’ and fear from the academic community..I think they are at last realizing the full implications of what is happening. Instead of treating this as just another academic exercise as they have done all through college and university..I feel they have crossed the threshhold. With directed anger you can acheive anything..you can put real pressure on respective governmental agencies if there enough of you who feel..rightly ‘terrorified’ of what the numbers you are numbly crunching actually imply for ALL of us. Now it is up to the scientists amongst you to do some forcing for yourselves..actually convey your fear and extreme concern and put this emotion into a logical and irrefutable protocol to present to your relevant authorities. A breath of fresh air came in the words of Amanda Eldridge..we can learn for that..humans are by nature ‘emotional’ creatures. The greatest oratories that have shaped our history have come from a deep feeling of ‘fear’ and ‘anger’ and concern. It is high time for all of to express our fear..to plagerize and modify an iconic saying..”I’m as ‘scared’ as hell and I’m not going to take it any more”!!!

  13. 263
    Peter Williams says:

    Yes Nukes. Now.

    There is no perfect solution. If you are not considering nuclear, you are not seriously considering how desperate the situation is. You are not thinking rationally. You are putting idealism ahead of pragmatism.

    Conservation has to be part of the solution. So does wind, solar, etc etc. However, we need additional baseload power that does not produce CO2. Wind and solar don’t cut it. You can’t build a power grid out of them. Sorry. Can’t do it. They can supplement the grid, but you still need baseload.

    Most of the problems with nuclear are problems of policy, not Nature. The cost of nuclear is dominated by things like compliance with radiation limits that put exposure orders of magnitude below natural background levels, based on the absurdity of a linear-no-threshold (LNT) model of biological risk to radiation exposure, or other policy absurdities like the abandonment of waste reprocessing, which adds huge expenses to nuclear energy in the form of the overhead of completely unnecessary waste disposal of perfectly useful fuel.

    Far more people have been killed just by coal trains alone than have ever been harmed by Western nuclear reactors, nevermind the thousands of people made ill by coal combustion products leading to asthma, cardiovascular problems, mercury poisoning, and nevermind the CO2 that coal puts out.

    It is true that Atoms of Peace was an unmitigated disaster; it backfired like so many other well-intentioned programs. However, I would love somebody to explain to me how a change in policy stance by the US or EU towards more civilian nuclear power, or a resumption of spent fuel reprocessing by the US, has any impact whatsoever on global nuclear security. It is quite obvious that Iran is trying to build a nuclear weapon, but I fail to see why in the world that should have anything to do with whether we decide to get more electricity from civilian nuclear plants. Of course Ahmedenejad will grandstand on anything, but that’s not really news.

    It is always the case with technology – always – that it can be used for good or evil purposes. You can use mills and lathes to make farm equipment, or you can use them to make weapons of war. That’s not a reason not to build mills and lathes.

  14. 264
    Peter Williams says:

    PS Oops I meant Atoms for Peace, not Atoms of Peace.

  15. 265
    CobblyWorlds says:

    #261 Timothy,

    Perhaps I am being too harsh regards the amount of confidence possible.

    That’s a very neat agreement, I’d been following the fairly constant lag between 2007 and 2008. I’d not tried fitting a curve. The match prior to August doesn’t surprise me, that it carries on thereafter does!

    If it’s not too much trouble: Can you update and re-post when the minima has been declared, it’ll be interesting seeing how it plays out.

    #258 Robert Henson,
    I’m done making any kind of guess about this year, I just don’t know. But thanks for your comment.

  16. 266
    LG Norton says:

    Well he is really doing it. Lewis Gordon Pugh (The guy that did the 1KM swim at the north pole last year) is kayaking to the north pole.

    Here is the blog of his progress.

    Lewis Goron Pugh Kayaking to North Pole”

    What I can’t figure out, is why he is attempting this from Svalbard. The ice is compressing in this part of the arctic. He should had started from one of the Russian islands in the Laptev Sea, maybe the Anzhu islands. The ice is more broken up on this side, and he would have stood a much better chance.

  17. 267
    crandles says:

    Re261 Timothy Chase

    A very good fit so far. The minimum dates per IJIS are 2003 18 Sept, 2004 19th, 2005 22nd, 2006 14th, 2007 24th. Losses from 30 Aug to minimum are 2003 293k km^2, 2004 123k, 2005 347k, 2006 172k and 2007 361k.

    So it seem a little strange that your fit of past data should be so aggressive in predicting minimum as late as 30 Sept and a further loss of over 1000k km^2. Have you tried seeing how you fit does with previous years and does this normally show that September is rather flatter than the quadratic fit prediction you have used?

  18. 268

    Whoever is organizing Polarstern?…. they are having the right idea, they are right next to a rapid melt zone, Some answers will come from them. It is also important to show the world this wide open huge seasonal sea, the Kayak guy is going to have a rough rough time, a mix of rotten
    and old ice awaits him. Its an extreme adventure, which will show how bad the ice has degraded,
    but Polarstern has a view of the better picture, wide open water when there should be nothing but ice. From what I gather the ice continues to melt, the 15% graph is a little confusing, when thin broken ice spreads out it gives the idea that the melt has stopped.

    [Response: Polarstern is run out of the Alfred Wegner Institute in Bremerhaven. - gavin]

  19. 269
    CobblyWorlds says:

    LG Norton,

    Re Lewis Gordon Pugh’s Team.

    The main reason I can see for trying from Svalbard is proximity and visa’s/permission. The current 30 Aug thinning from some of the Russian islands is transitory.

    This NASA Aqua image you can see from Svalbard to the Pole (distorted at the edge where the pole is).

    Through the cloud it’s possible to get an idea of the state of the ice. Given that the maximum resolution is 250m per pixel it might be feasible there’s enough open water in the cracks for a Kayak to get through. However even if the cracks are big enough it’ll be like navigating a constantly changing maze. Give it a few years and he may be successful on a second attempt.

  20. 270
  21. 271
    Hank Roberts says:

    A bit of history:

    Issue 20, Winter 1997
    U.S.-RUSSIAN ATLAS OF ARCTIC OCEAN
    “… On January 14, Vice President Gore announced a new atlas of Arctic oceanographic information in a press conference at the National Geographic Society. NSIDC is distributing this atlas. More than 1.3 million individual temperature and salinity observations on the Arctic Ocean collected over the period 1948-1993 from Russian drifting stations, ice breakers, and airborne expeditions were used to develop the atlas. Approximately 70% of the observations for the Arctic Ocean and shelf seas in this atlas are derived from Russian archives of formerly restricted data with 30% from comparable sources in the U.S. An article about the data in the February issue of National Geographic (“Arctic Breakthrough”, volume 191(2), p. 36-57) describes, in a forward by Vice President Gore, the negotiations with Russian Prime Minister Viktor Chernomyrdin that led to the release of the data…..”

    http://nsidc.org/pubs/notes/20/

    Following links associated with that story shows a huge amount of information, particularly on ocean currents, was being tracked by the Russian and US Navy submarine fleet. It probably still is. I wonder what they know now?

  22. 272
    Timothy Chase says:

    CobblyWorlds wrote in 265:

    If it’s not too much trouble: Can you update and re-post when the minima has been declared, it’ll be interesting seeing how it plays out.

    Already planned on doing so — no matter how foolish it might make me look. I figure the least I could do would be to share my confusion.

  23. 273
    Brian Dodge says:

    I was just wondering; with the shift to a seasonally ice free arctic, could the presence of (relatively) warm, moist, less dense air over open water instead of ice in the summer cause net rising air transport and reversal of the polar cell circulation, the elimination of the Ferrel cell, and sinking, stable, drought producing flow in the temperate zones? If this did happen, it would have what Frank Luntz might call “significant impacts” on food production.

  24. 274
    Timothy Chase says:

    crandles wrote in 267:

    A very good fit so far.

    You might not say that if you looked at the quadratic fit of the daily ice loss by date. The actual is rather jagged and the projected as smooth as a parabola — oddly enough. But the actual doesn’t remain on one side of the projected for very long, and in terms of ice extent it all seems to balance out rather quickly.

    crandles wrote in 267:

    The minimum dates per IJIS are 2003 18 Sept, 2004 19th, 2005 22nd, 2006 14th, 2007 24th. Losses from 30 Aug to minimum are 2003 293k km^2, 2004 123k, 2005 347k, 2006 172k and 2007 361k.

    So it seem a little strange that your fit of past data should be so aggressive in predicting minimum as late as 30 Sept and a further loss of over 1000k km^2.

    The big question is, “How long will the same quadratic remain a ‘good fit’ for the ice loss by date?” If ice melt continues until September 30th, then I see no reason why the quantity of projected melt shouldn’t be of the magnitude projected, give or take a little. But can it continue until September 30th? Well, the minima for 2003, 2004 and 2006 were on the 18th, 19th and 14th, but the minima for 2007 was on the 24th. The 24th is 6 days after the 18th. That was a new record — although only by 2 days when compared with 2005. So I would argue that if we could reach the minima as late as the 24th, there isn’t much reason to think that the next minima couldn’t occur 6 days later — even if that will mean another record for lateness and extent. Anyway, the lateness would be the more remarkable of the two in my view.

    crandles wrote in 267:

    Have you tried seeing how you fit does with previous years and does this normally show that September is rather flatter than the quadratic fit prediction you have used?

    I haven’t tried previous years as of yet. Then again, it doesn’t look like previous years haven been this smooth before, either. But it would be something worth trying. Easy enough to do. The data from IJIS goes back to 2002, and while it has some holes to it, one doesn’t need complete data to lay down a trendline in Excel — not even a quadratic one. At the same time, I don’t expect it to work quite as well. We have much less old ice this year. Old ice helps to anchor the formation of new ice. And the new ice has been being blown to-and-fro by the wind this year much more so than previous years — or so someone else observed earlier in this thread I believe.

    I will look into it.

  25. 275

    Gentlemen,

    I have been watching the breakup of that “landfast ice” on the north-northeastern coast of Greenland for some months now via the ENVISAT images on the Danish Institute of Technology’s website.

    I have a few questions.

    1. Why is it not called an “ice shelf”?

    2. Does anyone know if it has ever entirely broken away before?

    3. What is the name of the glacier, there, that is no longer being buttressed by the “landfast” ice?

    4. Is anyone studying the flow of that glacier? My impression is that it is indeed flowing out now, but this is pure conjecture on my part based on the ENVISAT images.

    5. Have y’all been watching the ice break up to the north of that position in an area that would seem “unmeltable” due to its position just south of all of the multi-year ice flowing by? And, note that the temperatures there in the north have been pretty low over the last month.

    6. As regards the MODIS images — is there a schedule available of when new images of that area of Greenland become available? I don’t really know how to access the images.

    7. The Jakobshavn glacier seems to be really going to town. When will the year’s data be released on flow rates, ablation of the ice sheet, and so forth?

    I would mention that once I ran across a great link (unfortunately, I lost it) to google images of the topography of Greenland, and they were on the scary side of awesome because that topography looks like it is just made for ice to slide out, as these paths of least resistance have been carved out over the millenia.

  26. 276
    pat n says:

    Re: 254

    Based on the on data plot trends on the nsidc website it is not a gross overstatements of confidence for me to say that 2008 extent will likely drop below the 2007 record low.

    Furthermore, it is not a gross overstatements of confidence for me to say that the low in 2008 will likely occur later in the season than the historical average.

    90% is a numerical confidence level I used to define the term “likely”.

    NWS includes probability values in issuing their flood outlook guidance at river stations in the Midwest. However, the NWS hydrologic modeling and outlook methodologies are flawed because NWS has refused to even consider that climate change has been changing rainfall intensities, the timing of snowmelt runoff, transpiration rates and evaporation.

  27. 277
    Hank Roberts says:

    For Tenney’s question #1 (and perhaps for the Glossary reference collection:

    “An «ice shelf» is a thick and extensive sheet of floating glacier ice, …
    Sea ice which forms and remains attached to the coast is termed «landfast ice»; …”
    http://www.pc.gc.ca/progs/amnc-nmca/systemplan/gloss_E.asp

    (Found by searching Google: landfast ice shelf )

    For #6:
    You could ask at the MODIS/AQUA page. I find:

    “Terra’s orbit … passes from north to south across the equator in the morning, while Aqua passes south to north over the equator in the afternoon. … Terra MODIS and Aqua MODIS are viewing the entire Earth’s surface every 1 to 2 days …” http://modis.gsfc.nasa.gov/about/

    http://rapidfire.sci.gsfc.nasa.gov/realtime/ shows the satellite tracks and time ticks for both satellites. Those have helped me look for a particular realtime image by seeing about when the satellite crossed my area of interest. No guarantee.

    For other than the ‘realtime’ gallery selections, there’s a lag: “The MODIS snow cover and sea ice products from the Terra satellite…. processing and reprocessing schedule for all MODIS data determines the lag …”
    http://nsidc.org/data/modis/faq.html

  28. 278
    Hank Roberts says:

    Yeek!
    http://climatechangepsychology.blogspot.com/2008/08/climatologist-dr-jason-e-box-byrd-polar.html

    _______excerpt follows_______

    Climatologist Jason E. Box, Ph.D., with the Byrd Polar Research Center, has spent the last 14 years monitoring Greenland’s massive ablation.

    “Estimates of sea level rise are now known to be significantly underestimated,” said Box. “The IPCC [Intergovernmental Panel on Climate Change] projected sea level rise of around one-and-a-half feet. But this did not take into account the profound ice sheet sensitivity now documented [in Greenland]. Sea level rise could be double, or more [than these predictions]. My best guess is a sea level rise of between three and six feet by the end of this century.”

  29. 279
    Phil. Felton says:

    Re #275
    Tenney Naumer Says:
    31 August 2008 at 12:13 PM
    Gentlemen,

    I have been watching the breakup of that “landfast ice” on the north-northeastern coast of Greenland for some months now via the ENVISAT images on the Danish Institute of Technology’s website.

    I have a few questions.

    1. Why is it not called an “ice shelf”?

    2. Does anyone know if it has ever entirely broken away before?

    3. What is the name of the glacier, there, that is no longer being buttressed by the “landfast” ice?

    4. Is anyone studying the flow of that glacier? My impression is that it is indeed flowing out now, but this is pure conjecture on my part based on the ENVISAT images.

    5. Have y’all been watching the ice break up to the north of that position in an area that would seem “unmeltable” due to its position just south of all of the multi-year ice flowing by? And, note that the temperatures there in the north have been pretty low over the last month.

    6. As regards the MODIS images — is there a schedule available of when new images of that area of Greenland become available? I don’t really know how to access the images.

    Tenney
    I don’t know the answers to most of your questions but I’ve been following that area for the last month or so and the break up has been spectacular!

    I look for the Modis pictures on this site.
    http://rapidfire.sci.gsfc.nasa.gov/realtime/
    I go to the little Terra map on the top left to see when the satellite crosses that region (usually between 1900 and 2100) and then look at the images indexed by time. Often they’re too cloudy but you find the odd gem!
    Here’s last night’s, I think you’ll enjoy it, try 250m res.

    http://rapidfire.sci.gsfc.nasa.gov/realtime/single.php?2008243/crefl1_143.A2008243190000-2008243190500.2km.jpg

    It seems to fit the definition of an ice-shelf, at least up to a week or so ago.
    Based on Google earth a Dane called Jørgen Brønlund is buried near there, the glacier is ~12miles wide so that’s a big piece of ice that’s gone. I think it is Nioghalvfjerdsbræ & Zachariae Isstrøm and the triangulular island between them is Lambert Land and the one to the Nth with clear water behind it on this image is Hovgaard, the little side stream to the Nth is Spaltegletscher.
    “GREENLAND” By ANKER WEIDICK http://pubs.usgs.gov/prof/p1386c/p1386c.pdf describes the ice as being in a “semipermanent condition of fast ice”.
    For a study see here:
    http://www.sciencemag.org/cgi/content/abstract/276/5314/934?ck=nck
    http://www.scienceonline.org/cgi/content/abstract/311/5763/986
    Hope that helps?

  30. 280
    sidd says:

    Tenney Naumer,12:13 pm, 31 Aug 2008
    2)re: name of glacier: Petermann ?
    http://geology.com/news/2008/petermann-glacier-breaking-up.shtml

    re:greenland topo:
    http://membrane.com/sidd/greenland.html

    toward the end, there is a link to an animated rotating view of the bedrock as well.

    sidd

  31. 281
    Timothy Chase says:

    Actual vs. Projected Sea Ice Extent for 2007

    Looking at 2007…

    I did two charts.

    The first is done exactly the same way as how I arrived at the projection for 2008.

    I calculate daily sea ice extent loss for every day from June 1, 2007 to August 27, 2007. Then I fit a quadratic trendline to the daily sea ice extent loss and use that to calculate a projected daily sea ice extent loss. I then calculate a running sum of projected daily sea ice extent loss and subtract that from the sea ice extent for May 31, 2007 to arrive at a projected daily sea ice extent.

    http://img530.imageshack.us/my.php?image=firstfit2007gg8.jpg

    For anyone familiar with last year’s sea ice extent record, at least some of the results aren’t that surprising. The projection is a poor fit. After the first 30 days, the projected is considerably above the actual and remains that way for the next 30 days — throughout all of July. It then reaches its minimum on September 1, 2007: 4722767.9. The actual minimum was on September 24, 2007: 4254531. As such the projection underestimates the duration of sea ice extent loss by 23 days and 468,236.9 square kilometers.

    The second chart was as follows…

    Now if you will remember, the bottom fell out back in June of 2007, so I decided to try and avoid having that throw off the quadratic trendline by beginning with July 4, 2007 as day zero, constructing my trendline using only the daily sea ice extent loss from July 4 through August 27 of 2007. I then used the equation from that trendline to calculate a daily sea ice extent loss for the preceding days, then calculated a projected sea ice extent as the sea ice extent of July 3, 2007 minus the running total of projected loss (but plus the running total of projected daily sea ice extent loss for the days preceding July 3, where July 3 would have projected sea ice extent equal to the actual).

    The results? Better after July 3.

    http://img530.imageshack.us/my.php?image=secondfit2007pj3.jpg

    In fact, it fits well for a stretch of about 70 days. The projected sea ice extent bottomed out on September 16, 2007 at 4552757.6. The second projection underestimates the duration of sea ice extent loss by 8 days and 298,226.6 square kilometers.

    Hurricane activity, perhaps? If hurricanes in the Atlantic and cyclones in the the Pacific have a significant impact on poleward oceanic advection, they would act to prolong the melt.

    For comparison, here is the chart for 2008 again:

    Sea Ice Extent Since May 31, 2008
    http://img337.imageshack.us/my.php?image=seaiceextentsince200805cs2.jpg

    Will the actual and the projected begin to diverge before September 30, 2008? Perhaps, but if 2007 is any indication, it would appear that I am actually underestimating the duration and extent of the melt, not overestimating it.

    In any case, once the old ice is gone, there is less to anchor new ice for new ice formation. The is less ice to dampen the wind-driven waves. And there should be more mixing of the surface layer of the ocean with deeper layers. All of this combined with infrared radiation that has remained fairly stable (even as the sun drops towards the horizon and visible light fades with the passing of the season) seem to be prolonging the melt in recent years. Meanwhile, hurricane intensity has been increasing.

  32. 282
    Hank Roberts says:

    You’d want to look at the ice to determine whether it formed from seawater as sea ice, or from snowfall that became glacial ice.

    Might see if there are any ice cores on record from that location that would distinguish its origins.
    ___________________
    “books whole-heartedly”

  33. 283
    Timothy Chase says:

    Correction on Second Fit for 2007:

    There was a nasty inflection point which I hadn’t noticed (somehow!) that existed in extending 2007 back from July 3, 2007. It indicted I hadn’t used an absolute anchor in a formula refering to July 3, but was instead adding a running total by means of a relative reference to the previous day (relative to whatever day I was projecting).

    New chart at:

    Correction on Second Fit for 2007
    http://img244.imageshack.us/my.php?image=correctedsecond2007wp8.jpg

    Anyway, with a suitably oversimplified climate model, it might actually be possible to derive what sort of trendline the melt should follow.

    *

    A few last notes:

    1. To have increasing melt that slows and is followed by freezing, you need a trendline that is at least quadratic.
    2. By fitting the trendline to the daily sea ice extent melt rather than a trendline to the sea ice extent itself, one is able to keep the trendline quadratic over a fairly long period.
    3. With higher degree trendlines you would be able “fit” any melt. Therefore they are not uniquely determined by the data.
    4. The quadratic is uniquely determined by the data for the period from which it is derived — no matter how poorly it may fit due to either the weather (including hurricanes) or large scale structural changes in old ice.
    5. Just as one may fit a linear trend which will be “realistic” over a suitably short period of time (assuming no inflection points), one should be able to fit a quadratic over a longer period, cubic over an even longer period, etc.. But I have avoided the cubic because it is not uniquely determined.

    *

    Is the trendline for 2008 realistic? So far, yes. But it has to break down at some point. And personally I expect it to break down when the internal dynamics nears the projected minima — due to outside forces that overwhelm those dynamics: hurricanes that extend the melting season with poleward oceanic heat advection.

    Thank you for your patience in my rather amateurish exercise in curve-fitting.

  34. 284

    Dear Hank,

    Thanks! You know, Dr. Hansen has been talking about 2-4 meters sea level rise for a long time. Lemme c, when was the last time he was majorly wrong?

    And while we can see that Greenland is starting to look pretty iffy, Dr. Hansen has always said that the main concern is the WAIS.

    Temperature anomalies down there have been in the startling +20 degrees C for months (and it was winter down here).

    You can see a 30-day animation of the temp. anomalies at this link (takes a while to load):

    http://www.cdc.noaa.gov/map/images/fnl/sfctmpmer_01a_30frames.fnl.anim.html

  35. 285

    Dear Phil,

    Thanks for those links, I have been trying to download the pdf file of the book, which is great, but I keep getting error messages halfway through. And that 250-m resolution image is fab.

    And, it looks like Gavin may have been holding out on us — hmmmm!

    http://www.sciencedaily.com/releases/2008/08/080831151346.htm

  36. 286

    sidd, thanks!

    I love your maps. The glacier in question is on the east side at the 80th parallel.

  37. 287
    Jim Galasyn says:

    Good times.

    Past evidence boosts concern for Greenland icesheet: scientists
    by Richard Ingham Sun Aug 31, 1:38 PM ET

    PARIS (AFP) – Scientists Sunday said they could no longer rule out a fast-track melting of the Greenland icesheet — a prospect, once the preserve of doomsayers, that would see much of the world’s coastline drowned by rising seas.

    The researchers found that the great Laurentide icesheet which smothered much of North America during the last Ice Age melted far swifter than realised, dumping billions of tonnes of water into the ocean.

    The discovery raises worrying questions about the future stability of Greenland’s icesheet, for the Laurentide melt occurred thanks to a spurt of warming that could be mirrored once more by the end of this century, they said.

    “The word ‘glacial’ used to imply that something was very slow,” said climate researcher Allegra LeGrande of New York’s Columbia University.

    “This new evidence from the past, paired with our model for predicting future climate, indicates that ‘glacial’ is anything but slow. Past icesheets responded quickly to a changing climate, hinting at the potential for a similar response in the future.”

  38. 288
    crandles says:

    Re 283 Timothy Chase
    “it would appear that I am actually underestimating the duration and extent of the melt, not overestimating it.”

    I think I expected your method to be close for 2007 and 2005 and to overestimate for 2006 2004 and 2003 but I could be wrong.

    Your method doesn’t use any data from past Septembers. A different approach might be to use the average of 3 days either side of each day for 2003 to 2007 to get the daily pattern from past years. Then use 31 August as a correct value and scale the daily pattern to get a best fit for May to August. Continue with the same scale of the daily pattern for a prediction.

    What I am trying to say is why use your quadratic fit when you could use data from past years?

    Doing this I get a minimum on 18 Sept of 4.79 million km^2.
    If I minimise the errors for August then I get minimum on 19 Sept of 4.72 million km^2.

    Due to the state of the ice I think the minimum will be lower than this. However note that this does build in the faster melt rate in August through the scaling factor.

    Your prediction may turn out to be better than the above calcs but would this just be a fluke?

    For 2003 my calc of the minimum was 47K km^2 out and for 2004 99k out. Note however I used data from 2003 to 2007, so I am using some data from these years so not a genuine attempt at a prediction without data from that year and I definately expect to be further out in 2008.

  39. 289
    Chris says:

    Updates from someone with a different perspective.

    The average daily ice extent reduction for the last 4 days was 34,375 km2, compared with 80,469 km2 for the previous 4 days.

    Temperatures at 6 GMT (Sep 1st) were -11C within ~3 degrees latitude of the North Pole, -9C to the N of Greenland, -6C to the NE of Greenland, -5C to the N of Spitzbergen, and -3C to the N of the Canadian Archipelago. It was -3C and snowing at ~81N on the Beaufort Sea side of the Arctic Basin, and -1C and snowing at ~78N on the Chukchi Sea side of the Arctic Basin.
    http://www.uni-koeln.de/math-nat-fak/geomet/meteo/winfos/synNNWWarctis.gif

    The SST anomalies (as of the last couple of days) continue to be colder than on the same date in 2004 on the approaches of both Pacific and Atlantic oceans to the Arctic:
    http://weather.unisys.com/archive/sst/sst_anom-040829.gif
    http://weather.unisys.com/archive/sst/sst_anom-080831.gif
    (Of course the open water in 2008 that was previously ice in 2004 is warmer! What matters is what happens at the ice boundary, which is further north on average than in 2004, and further from the warmer waters on the periphery of the Arctic)

    The SOI continues to lurch back towards La Nina, which suggests that there is an increasing chance the world will stay relatively cool or even become even cooler over the next year, which could be good news if you want the Arctic ice to recover.
    http://www.longpaddock.qld.gov.au/SeasonalClimateOutlook/SouthernOscillationIndex/30DaySOIValues/

    The buoy set in first year ice at ~88N on April 20th continues to show thickness of ~1.3m despite southward drift to ~83N (and the perfectly normal IR levels shown by its co-located PMEL Met Station – though incidentally the non-anomalously higher IR levels of the last two weeks couldn’t have anything to do with reflection from the two weeks of ~ 1-1.5m thick snow they have coincided with, I presume?)
    http://www.crrel.usace.army.mil/sid/IMB/2008E.htm

    The multiyear ice on the Beaufort Sea/Arctic Basin boundary continues to be >3m thick
    http://www.crrel.usace.army.mil/sid/IMB/2008F.htm
    and the multiyear ice a couple of degrees to the north of the Canadian Archipelago continues to be significantly thicker (~3.4m) than when the buoy concerned was deployed on Sep 9 2007 (2.8m)
    http://www.crrel.usace.army.mil/sid/IMB/2007J.htm
    The first year ice in the Central Arctic is doing just fine, at ~2m thick i.e. exactly the same as when the relevant buoy was deployed in April:
    http://www.crrel.usace.army.mil/sid/IMB/2008B.htm

    The less good news is that while sea ice extent reduction has slowed down in the last few days, sea ice area has seen further significant losses, now down to 3.239 or only ~11% higher than the 2007 minimum. Some might say that this suggests the previous slowdown in area reduction was an illusion caused by melt ponds re-freezing at the Pole. But remember that the melt pond issue could cut both ways: firstly, melt ponds can temporarily re-melt, and secondly, the temporary melting of the first winter snows (on the surface of the ice) can create the illusion of more open water than is actually the case.

    “No matter where we stand at the end of the melt season it’s just reinforcing this notion that Arctic ice is in its death spiral,” said Mark Serreze, a scientist at the center [NSIDC]. The Arctic could be free of summer ice by 2030, Serreze said by telephone.
    http://www.reuters.com/article/marketsNews/idUSN2745499020080827

    Clearly it is justified for people to have a *notion* that Arctic ice is in its “death spiral”, and that the Arctic *could* be free of summer ice by 2030.

    But we still have significantly more ice than this time last year (re: both area and extent), so I don’t think it’s unreasonable for me to wait until 2008 ACTUALLY surpasses 2007 before joining those on this thread who are already confidently predicting doom.

  40. 290
    John L. McCormick says:

    RE # 254

    CobblyWorld, you said:

    1) Climatic shift.
    If you warm the Arctic, and (crucially) increase humidity due to higher temperatures and more open water over 0degC, then you will change the relationship between the polar and tropical regions. This changes what happens in between the pole and tropics,

    We all know what exists between the polar and tropical regions of Western North America; it is called the world’s grain basket.

    When am I going to hear that the National Science Academy or any science body announcing the beginning of an exhaustive research to understand how the Arctic ice meltback will impact temperature and precip patterns in that vital part of the world?

    Gavin, do you have some insight here?

    John McCormick

  41. 291
    dhogaza says:

    Clearly it is justified for people to have a *notion* that Arctic ice is in its “death spiral”, and that the Arctic *could* be free of summer ice by 2030.

    But we still have significantly more ice than this time last year (re: both area and extent), so I don’t think it’s unreasonable for me to wait until 2008 ACTUALLY surpasses 2007 before joining those on this thread who are already confidently predicting doom.

    The scientist, of course, is placing this second-lowest ever minimum in the context of a statistically significant trend that spans the entire satellite era when he makes his claim.

    While you’re harping on 2008 vs. 2007 as though it’s meaningful. It’s not.

    However, since we’re playing …

    Ice area is not significantly higher than 2007.

    As far as ice extent … well the jet stream shifted the last few days, bringing an unseasonal cold friend blowing into the pacific northwest from the northwest.

    A high is bringing summer weather back to us this week. Not sure what implications there are for winds shifting up north, but if area is decreasing while extent is increasing, it would seem that wind’s doing it, and a shift could start compacting the extent again.

    All guessing but that’s all you’re doing, too … but I only do it for fun, because I know that the trend, not 2008 vs. 2007, is significant and even if 2007 is “only” the second-most minimum in area and/or extent since 1979, it will add to that trend.

  42. 292
    Chris says:

    #291 Dhogaza

    I’m as concerned as anyone else about the Arctic ice melting and its possible implications.

    The reason I’ve been following events in the Arctic this summer, reading this thread, and making posts, is because I want to understand exactly what is happening, and if I have any potential insights then I want to share them.

    I try to avoid responding to posts that tend to make broadly dismissive comments while avoiding specifics. However, on this occasion I will make a couple of comments.

    “While you’re harping on 2008 vs. 2007 as though it’s meaningful. It’s not.
    However, since we’re playing …
    Ice area is not significantly higher than 2007″

    The reason I’m “harping on 2008 vs 2007″ is because this is what many posters on this thread have been doing (since well before I joined in) and making assumptions such as that ice volume decrease this year is greater, or that the August melt in 2008 means the minimum is 90% likely to be lower than in 2007 – with obvious implications for the continuance of the longer term trends you mention. I’ve simply been joining in the debate. I’m certainly not “playing”.

    I don’t think I overstated my point re: area if you read all that I wrote: “…sea ice area has seen further significant losses, now down to 3.239 or only ~11% higher than the 2007 minimum.” Your link is the graph which represents the numbers I quoted. And clearly if the falls of the last couple of days continue then yes the area will no longer be (arguably) significantly higher. (Although the one/two/three etc month averages will continue to be definitely significantly higher – if one prefers longer-term to shorter-term trends….)

  43. 293
    Timothy Chase says:

    crandles wrote in 288:

    Your method doesn’t use any data from past Septembers. A different approach might be to use the average of 3 days either side of each day for 2003 to 2007 to get the daily pattern from past years. Then use 31 August as a correct value and scale the daily pattern to get a best fit for May to August. Continue with the same scale of the daily pattern for a prediction.

    What I am trying to say is why use your quadratic fit when you could use data from past years?

    Well, what this assumes is that each year is basically repeating the same pattern at the same time as the previous years — give or take some noise. But there are some obvious changes taking place. As a matter of how the Arctic responds to global warming, we expect the minima in future years to continue to take place later and later in the year. But that is just the general trend.

    More specifically, last year was a record not simply in minimum sea ice area or sea ice extent, but for the lateness of the minima itself. And as I have pointed out, there is very little multi-year ice left, the ice is thinner, meaning that it dampens the production of waves much less, waves tend to break apart ice, exposing it to the water. Larger waves will also tend to mix upper layers of the ocean with lower layers, and lower layers tend to be warmer. All of this tends to shift the minima further into what has traditionally been outside of the melting season. Your three days either way rule wouldn’t take this into account. It also wouldn’t take into account how drastically things changed with the destruction of so much multi-year ice in 2007, or for that matter, the fact that new ice is saltier and therefore will freeze or melt at lower temperatures.

    Of course my method doesn’t explicitly take into account any of this either, but it at least has the virtue of not making any assumptions about the timing of the minima which clearly no longer hold. It is extremely simple, and by means of a quadratic fit to earlier data in this season uniquely predicts a simplicity of behavior which quite closely matches what has actually taken place — for a period of time in which the Arctic itself has become much simpler — given the flushing out of fresh water in previous years (which tended to protect the sea ice from the warmer layers of salty water below) and the loss of so much multi-year ice. It appears that the law of large numbers is playing a much more conspicuous role this year.

    We will see what happens. The projection had tended to be slightly above actual sea ice extent, but recently fell slightly below. Now I notice that there was a slight bulge in the sea ice area graph but that sea ice area is beginning to fall back to its “mean behavior” for this year’s season, so I should see actual sea ice extent drop back down to the projection over the next couple of days.

    In any case, I think the projection is an improvement upon the straight trend lines some were trying or impressions of what should happen simply based on what had happened in previous years on the same dates.

    crandles wrote in 288:

    Doing this I get a minimum on 18 Sept of 4.79 million km^2.
    If I minimise the errors for August then I get minimum on 19 Sept of 4.72 million km^2.

    Can you graph it? Could you make available the equations you used? (Might not work that well here — but a link to a graphic shouldn’t be that difficult.)

    It sounds like a good projection — given the past behavior of the Arctic, and assuming things haven’t changed.

    Anyway, the nice thing is that we will soon know one way or the other.

    *

    captcha fortune cookie: average escape

  44. 294
    Schmert says:

    I’m as concerned as anyone else about the Arctic ice melting and its possible implications.

    It’s interesting looking at Tide heights, (excluding atmospheric pressure deviation, and weather assistance) but they certainly appear to have been getting higher over the passed few years.

    Although this does make getting a boat back into the water that little bit easier, it can also play havoc with one’s loafers.

    Not to mention a few low lying , propserous and often quite useful bits of the planet.

    With the certain inevitabitlity that, as the ice caps melt, the melt will increase due to the rise in temperatures rise in temperature !!!!!!! no one really seems to be talking about any kind of active steps to do anything about it.

    Guess it depends on whos feet get wet and wether they’ve managed to get to Mars or not, before contemplating going to the expense of refreezing it.

    or just another case of ‘Nero Syndrome’.

  45. 295
    Hank Roberts says:

    Biology that lives on the sea ice in the springtime:
    Biology associated with sea ice (this study done in Antarctica, in December — around midsummer):

    doi:10.1016/j.dsr2.2007.12.019
    Deep Sea Research Part II: Topical Studies in Oceanography
    Volume 55, Issues 8-9, April-May 2008, Pages 1024-1039
    Ice Station POLarstern (ISPOL): Results of interdisciplinary studies on a drifting ice floe in the western Weddell Sea
    Short-term biogenic particle flux under late spring sea ice in the western Weddell Sea

    “… two sediment traps were deployed at 10 and 70 m water depth under a drifting ice floe in December 2004. The amount and composition of the vertical particle flux under sea ice were determined during a period of 30 days in order to investigate the influence of biological processes in sea ice and on its underside on the flux…. A strong increase with time of the flux of chlorophyll equivalents, biogenic silica, and faecal material was recorded during the observation period, coincident with the increase in the concentration of chlorophyll a in the bottom ice layer above the trap array. The latter suggests a concomitant increase in the amount of food available for grazers, such as krill, in the bottom ice layer and on the underside of the ice floe, resulting in an increased downward transport of ice-algal material into the water column….”

  46. 296

    Cryosphere today places 2008 almost tied with 2007 extent, the difference is so small, consider it a tie.
    Now this is interesting, SST’s and surface temperatures are not favorable at all for melting ice. Yet
    its happening, a clue, 850 mb data shows a greater warm air than on surface near the archipelago, inversions are not always at all at the same height every where over the arctic ocean, it is extensively cloudy as well a huge cloud area over the entire Polar region. Surface temperatures are generally much below zero over the Arctic ocean, yet the ice vanishes nevertheless. Its quite warm at 850 mb over NE Greenland where ice is unusually scarce at that location.

  47. 297
    NeilT says:

    #292, I don’t think that many of us expect 2008 to surpass 2007 in melt or extent this year. That does not mean that the people who are trusted with watching and predicting the events in the Arctic feel any better about it.

    It is, today, massively lower than either 2005 or 2006. 2005 was considered a major event anomaly, 2006 a near miss. 2007 a major event anomaly and it’s looking like 2008 will reinforce that trend.

    What you are missing is that 2008 is tracking 2004 SST and atmospheric temperatures with >1M Sq Km less ice. Woods and trees come to mind.

    This year the ice has become detached from almost all land, another first. Just how many firsts do we have to have before people understand that it is now out of control?

  48. 298
    crandles says:

    re 293

    >”Can you graph it? Could you make available the equations you used?”

    http://www.boincforum.info/boincuser/Crandles/ExtentDailyPatternPredictionGraph.JPG

    http://www.boincforum.info/boincuser/Crandles/Extentdailypattern.xls

    >”It sounds like a good projection — given the past behavior of the Arctic, and assuming things haven’t changed.”

    That is exactly what it is trying to do.

    I hear a lot of qualitative arguments for retreat greater than average or greater than 2007 which was a record but very little quantitative. How much faster 50%? double? tripple? There are some vage record is likely but is there any evidence that it will be tripple rather than the 50% I have build in based on August rate of retreat?

    “my method doesn’t explicitly take into account any of this either, but it at least has the virtue of not making any assumptions about the timing of the minima”

    Well look at the last graph of
    http://arctic.atmos.uiuc.edu/cryosphere/sea.ice.minimum.2007.html

    The trend isn’t large and is swamped by weather noise. Is building in an ability of my model to have a later minimum worthwhile? It would probably just be another fiddle factor allowing overtuning.

    If this virtue allows crazy dates then I would prefer something more reliable.

    >”how drastically things changed”

    But is this known for sure in terms of its effect on trends? One way to see is to try to get a reliable prediction method based on past only and see how large the errors now are.

    OTOH I don’t see much point in a method that appears less reliable, makes a prediction which is much more aggressive than the past would indicate and frankly (sorry!) appears to rely on a wing and a prayer that apparent changes are drastic such that they just happen to be about the size required to balance out the aggressiveness compared to the past. Simplicity is a nice virtue but only if reliable. Having said this, your prediction will probably turn out to be very accurate.

  49. 299
    Lawrence Coleman says:

    Been following the uni-bremen arctic ice extent site for a number of weeks now and what I am seeing is a rapid accelleration in the rate of ice melt. I do not know the exact current figures of the area of arctic ice at the moment but I am willing to say it has now reached a new all time record, in the last two days there is now a band of sea water separating greenland’s northern most coastline with the ice shelf, the ice is also rapidly breaking up to the north estern corner of the country. More worrisome in the last two days the NW and NE open sea passage is now very clearly defined. I can not beleive that soo much ice has melted in just 48 hours. The whole pack ice area around 85N, 165E and 145W seems much more fractured and eroded over the last few weeks. I am interested in the fate of the ice at 75N, 105W. I would say based on current melt rates that ice will be completly gone by mon next week.

  50. 300
    John L. McCormick says:

    re # 297

    nEILt,

    YOU ASKED:

    [Just how many firsts do we have to have before people understand that it is now out of control?]

    I say, tell it like it is: Arctic sea ice crossed its TIPPING POINT (maybe 2005–no matter when…it did)

    The fact that future September satellite images of Arctic sea ice will never again look like those of 1979 is the reality 6.6 billion people will have to accept.

    It is no longer about record-breaking. It is about adapting to what we do not know is coming next.

    John McCormick


Switch to our mobile site