Sea Ice – End of Game Analysis

On July 2, I started this popular series of threads as follows:

For anyone who’s betting that 2008 meltback will exceed 2007 meltback, I think that you’ll be able to pretty much know where you stand by the end of this week and your chances are not looking good right now based on this week’s exit polls. Another Climate Audit first.

The 2007-2008 gap was then just over 500,000 km. I don’t pretend to have any expertise in this topic, but I think that my early July predictions on this stood up better than those of the official organizations.

Thanks to TAC for keeping an eye on things.

As observed by commenters, 2008 levels were still exceptionally low and certainly cannot be interpreted as disproving anything about long-term trends. Indeed, quite the opposite. The most that can be said in the opposite direction is only that the baby-ice models all predicted a huge decrease in 2008 relative to 2007 and considerably more baby ice survived than initially projected.

741 Comments

  1. bender
    Posted Sep 23, 2008 at 5:36 PM | Permalink

    Without some measure of the uncertainty on these ice extent observations it’s a bit hard to say, but 2008 ended up pretty close to 2007. Closer to 2007 than any of the other years. On the one hand 2008 melt didn’t exceed 2007. On the other hand it was so close to 2007 that the long-term trend (whether 5, 10, 20 years) dropped significantly in slope. Some will try to spin this by saying that 2008 is the first year of a reversal. Others will say that 2008 did not have us cross a “tipping point”. Such arguments require more than a simple sentence or paragraph to prove. Be wary of all one-liners. Some, will argue that this or that index should not be used. I think it is prudent to consider all indices (extent, area, volume, mass, changes therein) and compare inferences obtained using each of them. Be wary of any analysis that is not robust to a choice of index.

    Have at ‘er.

  2. Larry Sheldon
    Posted Sep 23, 2008 at 5:47 PM | Permalink

    I was reading at the CNN Science site a while ago that the polar bears are turning to cannibalism because of the continued decline in ice.

    Which, ignorant as I am, does not make any sense at all even if the “decline in ice” did.

    Any ideas?

    • Jeremy
      Posted Sep 23, 2008 at 7:19 PM | Permalink

      Re: Larry Sheldon (#2),
      Actually, polar bears are already known for cannibalism. The males tend to eat the young if they get a chance (i.e., they are unprotected by the mothers, or the mother is otherwise unable to keep them away from the hungry male). So, claiming the loss of ice is forcing them to cannibalism sounds rather silly since they already do eat themselves if given the chance.

    • Geoff Sherrington
      Posted Sep 23, 2008 at 7:25 PM | Permalink

      Re: Larry Sheldon (#2),

      Re polar bears, this eating is just a positive feedback to global warming.

    • bender
      Posted Sep 24, 2008 at 1:58 AM | Permalink

      Re: Larry Sheldon (#2),

      Polar bears resort to cannibalism as Arctic ice shrinks

      “The Arctic sea ice melt is a disaster for the polar bears,” according to Kassie Siegel, staff attorney for the Center for Biological Diversity. “They are dependent on the Arctic sea ice for all of their essential behaviors, and as the ice melts and global warming transforms the Arctic, polar bears are starving, drowning, even resorting to cannibalism because they don’t have access to their usual food sources.”

      Scientists have noticed increasing reports of starving Arctic polar bears attacking and feeding on one another in recent years.

      Rather than “move along”, perhaps it is possible to track down the source of the last statement? Clearly, scientists do track such things, as a commenter here has remarked. Is there a trend? Is it statistically robust? Or is it just “scarebear” “alarmism”?

      I mean, is this ClimateAudit or ClimateMoveAlong?

      • bender
        Posted Sep 24, 2008 at 2:07 AM | Permalink

        Re: bender (#50),
        staff attorney?

      • Bill Jamison
        Posted Sep 24, 2008 at 2:25 AM | Permalink

        Re: bender (#50),

        From LiveScience in 2006:

        ANCHORAGE, Alaska (AP)—Polar bears in the southern Beaufort Sea may be turning to cannibalism because longer seasons without ice keep them from getting to their natural food, a new study by American and Canadian scientists has found.

        The study reviewed three examples of polar bears preying on each other from January to April 2004 north of Alaska and western Canada, including the first-ever reported killing of a female in a den shortly after it gave birth.

        http://www.livescience.com/animals/060612_polar_bears.html

        The Center for Biological Diversity is infamous for alarmist propaganda. In fact they admitted that the polar bear was chosen as the “poster child” species in the battle against CO2 emissions:

        The environmental left candidly admits the importance of the polar bear as a cute, cuddly symbol of their cause:

        Both sides agree that conservationists finally have the poster species they have sought to use the Endangered Species Act as a lever to force federal limits on the greenhouse gases linked to global warming, and possibly to battle smokestack industry projects far from the Arctic.

        “All animals are equal, but some animals are more equal than others,” said Kassie Siegel, an attorney with the nonprofit Center for Biological Diversity. “And then there is the polar bear.”

        http://agw-heretic.blogspot.com/2008/02/polar-bear-poster-child-of.html

    • Mark H.
      Posted Sep 27, 2008 at 11:14 AM | Permalink

      Re: Larry Sheldon (#2), I should have scanned the thread before answering – however, I also read that claim. However, I have also read (from more than one source) that cannibalism among polar bears is not unusual. Large adult bears will kill the young for food.

  3. bender
    Posted Sep 23, 2008 at 5:48 PM | Permalink

    Andy Revkin, in an article titled “Arctic Ice Retreat Misses Last Year’s Mark” put it this way:

    Everyone I queried stressed that the retreat this year, while not a new modern record, does mean that the downward trend in ice cover is still on track to reveal an entirely open ocean in summers within a few decades. Of course, trend is not destiny.

    This seems to be a fair summary. Of course, that is not the entire article. And I purposefully left off the last sentence, which was fairly speculative. This is appropriately open-ended, indicating some uncertainty on the major drivers.

    Neither “alarmists” nor “activists” will be happy with my editing (and editorializing). So be it.

    • kim
      Posted Sep 23, 2008 at 5:59 PM | Permalink

      Re: bender (#3),

      Pielke Pere agrees it was fair.
      ==================

  4. Patrick Henry
    Posted Sep 23, 2008 at 5:56 PM | Permalink

    2008 will probably end closer to 2005 than it will to 2007.

    • bender
      Posted Sep 23, 2008 at 5:59 PM | Permalink

      Re: Patrick Henry (#4), Not if it’s ended already. (Also: recall also my point about uncertainty bounds?)

  5. Patrick Henry
    Posted Sep 23, 2008 at 6:04 PM | Permalink

    bender,

    It depends on whether you consider the end point to be the date of the 2007 low, or the date of the 2008 low. It looks like the 2007 inflection point isn’t for two or three more days, with a high probability of ice gain in the meantime.

    http://pafc.arh.noaa.gov/marfcst.php?fcst=FZAK80PAFC

  6. Bill Jamison
    Posted Sep 23, 2008 at 6:04 PM | Permalink

    Did anyone else see this little item on ICECAP from a NWS Sea Ice Advisory?

    Sea surface temperatures along the Alaska Chukchi and Beaufort Sea coasts are 2 to 8 degrees Celsius colder this year than at the same time last year. Significant ice will begin developing along the Alaska coast north of 70n within the next 10 to 14 days.

    Maybe we’ll see another year of record growth.

  7. Steve McIntyre
    Posted Sep 23, 2008 at 6:15 PM | Permalink

    I added the following sentences:

    As observed by commenters, 2008 levels were still exceptionally low and certainly cannot be interpreted as disproving anything about long-term trends. Indeed, quite the opposite. The most that can be said in the opposite direction is only that the baby-ice models all predicted a huge decrease in 2008 relative to 2007 and considerably more baby ice survived than initially projected.

    I wasn’t really trying to write a post (as I was working on other things), but the other rather was creating a new placeholder for an overburdened thread.

  8. mugwump
    Posted Sep 23, 2008 at 6:16 PM | Permalink

    Time to open a book on 2009.

    I bet 2009 meltback will not exceed 2008 meltback.

    • kim
      Posted Sep 23, 2008 at 6:26 PM | Permalink

      Re: mugwump (#10),

      Following van Loon’s mild winter and cool summer I’ll go with maximum extent less this winter than last and minimum extent greater next summer than this. I’m counting on the general cooling to overcome local effects.
      =========================

  9. Ernie
    Posted Sep 23, 2008 at 6:25 PM | Permalink

    Does anyone know of a plausible explanation why the 2007 arctic extent was way off the average? I realize that even though at the start of 2008 the winter extent was a good start, all that baby ice was going to have a hard time in summer as the graphs have shown, so any recovery from 2007 is a good effort. Has anyone graphed the average extent back to 1978 yet to incorporate this years data?

    – Ernie.

    • kim
      Posted Sep 23, 2008 at 6:30 PM | Permalink

      Re: Ernie (#11),

      That’s an excellent question I can’t answer. I see it as a last gasp from the lastest warming finally working its way North, but that would imply a lag that I’m just guessing at.
      ==============================================

      • Posted Sep 24, 2008 at 1:52 AM | Permalink

        Re: kim (#13),

        Kim said;

        “That’s an excellent question I can’t answer. I see it as a last gasp from the lastest warming finally working its way North, but that would imply a lag that I’m just guessing at.”

        Does the warming work its way North or trickle down from the North? My post in the other place (repeated below) mentioned the falling temperatures in 5 out of 6 seas surveyed by the European Marine Agency in 2007. If the first proposition is correct the Arctic sea temperatures seem set to drop. Also all the evidence suggests the Arctic is a pretty fragile environment and ice levels oscillate considerably. When we talk of low ice extent we should add ‘in the modern era-post1979.’

        My previous post read as follows;
        ” I thought this might give you something interesting to discuss. Its a document from an EU marine agency and relates largely to fish

        http://www.sesame-ip.eu/doc/MB_Climate_Change_VLIZ_05031.pdf

        If you go to page 3 of the Executive summary very close to the start of the document, it shows six graphs with average sea temperatures dating back to 1930.

        Interestingly 5 out of the six are showing temperature declines-the only one going up is the Barent sea. The document is from Feb 2007 and I have just got figures from Defra (a British agency) that show the Celtic Biscay shelf has lost around another half a degree from the figures shown. I don’t have figures from the others. Perhaps an early sign of cooling with the Arctic lagging behind? Or perhaps not.”

        Tony Brown

        • Posted Sep 24, 2008 at 8:56 AM | Permalink

          Re: Tony Brown (#49),

          Now polar bears are banned can someone comment on the scientifc implications of the reduction in sea temperatures surrounding the Arctic. If there are any conclusions to be drawn at all of course. The link to the graphs can be accessed via #49

          Tony Brown

    • Posted Sep 23, 2008 at 6:55 PM | Permalink

      Re: Ernie (#11),
      There are those who believe the increase in soot (decreasing albedo) is responsible for most, if not nearly all, of the ice melt.

  10. Harry Eagar
    Posted Sep 23, 2008 at 6:40 PM | Permalink

    Disaster averted! Film at 11.

    I have my 2009 prediction ready: Whatever happens, it won’t be as bad as Al Gore is going to say it will be.

  11. David Smith
    Posted Sep 23, 2008 at 6:48 PM | Permalink

    Ignatius Rigor’s website is worth a visit. Arctic ice variation involves ice flow/loss as well as summer melt conditions.

    The wintertime values of the Arctic Oscillation ( chart ) give a pretty good indication of the extent to which sea ice, including old ice, is flushed from the Arctic during the winter. This flushed ice then melts in the Atlantic. Keep an eye on the AO this winter to see if the winds are conserving ice (negative AO) or flushing ice (positive AO).

    • Shawn Whelan
      Posted Sep 23, 2008 at 6:59 PM | Permalink

      Re: David Smith (#15),
      Just of interest, the Resolute was abandoned in the Northwest passage in the 1850’s and was flushed into Baffin bay as a ghost ship where it was recovered by an American whaler. The wind has much to do with the ice in the Arctic.

  12. Shawn Whelan
    Posted Sep 23, 2008 at 6:50 PM | Permalink

    The reason the Arctic ice extent in 2008 exceeded the 2007 extent was the colder temperatures and the hard freeze in the 2008 Winter and the cooler Summer temperatures in 2008.

    While the open water at the end of the 2008 melt is low, it also allows the water to lose heat easier since there is not as much ice and snow insulating the ocean. And any snow that falls instead of insulating the ocean will remove heat from the open water. That is a natural balance. And at the start of the freeze there is less heat in the Arctic in 2008.

    Looking at the historical record the Arctic can change rapidly from a low level of ice to a huge increase. Now if the cooling from the PDO, low level of Sunspots and LaNina work as they historically have the freeze will be quick and hard and a large extent.

    • BarryW
      Posted Sep 23, 2008 at 8:53 PM | Permalink

      Re: Shawn Whelan (#16),

      The reason the Arctic ice extent in 2008 exceeded the 2007 extent was the colder temperatures and the hard freeze in the 2008 Winter and the cooler Summer temperatures in 2008.

      2008 did start from a higher value but it still lost more ice extent overall.

      Yr max min diff
      2007 13.94562 4.254531 9.691094
      2008 14.51688 4.707813 9.809062
      diff -0.57125 -0.453282 -0.117968

  13. Joe Solters
    Posted Sep 23, 2008 at 7:22 PM | Permalink

    What’s bender up to now? Telling people how to think and react to the arctic ice issue! This sounds like someone with no introspective capacity whatsoever. Warmies have a bad habit of actually telling other people how to think about climate change. Preaching is OK. The reflex is now so thoroughly entrenched it’s led to Hansen, under oath, with no introspective capacity whatsoever, telling people it’s OK to attack coal burning pplants.

    • bender
      Posted Sep 23, 2008 at 7:28 PM | Permalink

      Re: Joe Solters (#20),
      Interpreting data trends is telling people how to think and act? Interesting.

    • bender
      Posted Sep 24, 2008 at 12:54 AM | Permalink

      Re: Joe Solters (#20),

      What’s bender up to now? Telling people how to think

      This is a classic ad hominem attack. Nothing here to refute anything substantive I’ve said. But Joe Solters would never tell people what to think, isn’t that so? Hey Joe, I think that is what the blogosphere is all about – communicating different points of view without the tyranny of some autocrat controlling what can be said. The difference is that I’m listening and learning. So go on and tell me what and how to think. That’s what I’m here for.

      Hey Joe, where you goin’ with that gun in your hand?

    • bender
      Posted Sep 24, 2008 at 9:54 AM | Permalink

      Re: Joe Solters (#20),

      Warmies have a bad habit of actually telling other people how to think

      What’s a “warmie”? Two years ago I asked CA regulars to state their “priors” – what proportion of 20th century warming could be attributed to GHGs. Here were the results. The last figure is the pertinent one.

      B.Hurd November 22nd, 2006 at 9:07 pm 0.30
      pat November 23rd, 2006 at 12:56 pm 0.20
      Reid November 23rd, 2006 at 10:27 pm 0.02
      KevinUK November 23rd, 2006 at 6:09 am 0.30
      Willis November 23rd, 2006 at 11:12 pm 0.30
      bruce November 24th, 2006 at 1:21 am 0.05
      Chris H November 24th, 2006 at 2:14 am 0.30
      beng November 24th, 2006 at 10:43 am 0.20
      Dardinger November 24th, 2006 at 1:15 pm 0.30
      jae November 25th, 2006 at 12:46 pm 0.10

      average = 0.21

      And here are the “warmies”:
      Lee November 27th, 2006 at 3:19 pm 0.70
      Juckes November 28th, 2006 at 3:29 am 0.50

      average = 0.60

      Nobody in either group replied “zero”.

      Anyone wishing to have their guesses registered or revised is welcome to answer the question.

      I don’t tell people what to think. I collate thoughts. People can think for themselves.

      • SteveSadlov
        Posted Sep 24, 2008 at 8:46 PM | Permalink

        Re: bender (#75),

        0.3333333333….

      • DeWitt Payne
        Posted Sep 24, 2008 at 8:53 PM | Permalink

        Re: bender (#75),

        0.3 for me.

      • Posted Sep 25, 2008 at 1:10 AM | Permalink

        Re: bender (#75),

        Bender

        .15. Assuming that the logarithmic theory is correct a doubling of CO2 from pre industrial levels would presumably take us to around .20 (Whether theory on a computer model is the same as the real world figures is another thing.

        DeWitt Payne. Can we have an ‘OFFICIAL’ end of race figure in both extent and area when that point has defintely been reached?

        The butter to soft spread transition date in my household is definitely looking favourable for an unprecedently early October 5th.

        Tony Brown

        • DeWitt Payne
          Posted Sep 25, 2008 at 10:25 AM | Permalink

          Re: Tony Brown (#107),

          I need the Uni-Hamburg data for September, which probably won’t be available until after October 10, to wrap things up completely. We’re close to (and may already have passed) the peak in the Antarctic as well. Right now the maximum stands at 14.963 Mm2 on 9/19/2008.

          CT area update 9/25/2008
          region area(Mm2) difference anomaly
          Arctic 3.07 0.022 -2.077
          Antarctic 14.864 -0.055 -0.196

          The adjusted concentration went up slightly. If I’m correct about CT being five days behind for their digital data, then we should see some fairly large increases in area the next few days.

  14. SteveSadlov
    Posted Sep 23, 2008 at 7:25 PM | Permalink

    The whole premise of “new” and “old” ice is at least partially incorrect. The theory goes, “old” ice has incurred phase changes, phase zoning, and chemistry zoning, over time, that increases its resistance to melting. Of course such processes exist. But they are not the be all and the end all. Ice can be deformed by compression and extension. It can be brutally compressed to the point of compressive conjugate fracturing and a form of “thrust faulting.” It can be extended to open up leads. In the midst of this, there is also plastic behavior. Meanwhile, things get deposited on the top. Not only the many things besides water, some of which increase melting, but also snow, and other forms of frozen precip. Within one year’s time, a given parcel of ice might be made to resist melting for an eon. Or, it might incur a worst case scenario of non aqueous deposition, extension, and movement into more temperate areas, in the middle of summer, and be gone only months after forming. Who really understands this system?

    I certainly do not. I doubt anyone really does, just yet. There is so much more to learn.

  15. bender
    Posted Sep 23, 2008 at 7:30 PM | Permalink

    There is a big difference between predicting human behavior and prescribing it.

  16. Phil.
    Posted Sep 23, 2008 at 7:58 PM | Permalink

    Steve:

    As observed by commenters, 2008 levels were still exceptionally low and certainly cannot be interpreted as disproving anything about long-term trends. Indeed, quite the opposite. The most that can be said in the opposite direction is only that the baby-ice models all predicted a huge decrease in 2008 relative to 2007 and considerably more baby ice survived than initially projected.

    The area of melt was approximately equal to the area of freeze (new ice) so old ice had to melt in place of any new ice which didn’t melt. Consequently the average age (and hence thickness) of the Arctic ice must have gone down this year. This is notable particularly in the Beaufort Sea where the fragmented old ice present this spring melted and late in the summer the flow shifted to push more old ice into the Beaufort.

    • Phil.
      Posted Sep 23, 2008 at 8:24 PM | Permalink

      Re: Phil. (#25),
      Steve, the images got flipped, this is this year’s.

  17. Posted Sep 23, 2008 at 8:23 PM | Permalink

    Both NSIDC and IJIS are showing 2008’s ice extent(still not sure how that differs from area) several hundreds of thousands of sq.kms of sea ice extent. As far as I have seen, those two are the ones most referred to in this blog and the media. The ICDC site (Cyrosphere Today)is showing 2007 and 2008 ice area on par. As good or better is always, always better than worse.

  18. Patrick Henry
    Posted Sep 23, 2008 at 8:25 PM | Permalink

    Phil,

    Given that that 2008 is ending with more ice extent than 2007, there will obviously be more multi-year ice going into 2009 than there was going into 2008.

    • Phil.
      Posted Sep 23, 2008 at 9:09 PM | Permalink

      Re: Patrick Henry (#28),

      Given that that 2008 is ending with more ice extent than 2007, there will obviously be more multi-year ice going into 2009 than there was going into 2008.

      Not obvious at all, ‘ice’ is not proportional to ice extent, the extent measure includes the spaces between the ice floes.

    • Phil.
      Posted Sep 24, 2008 at 9:23 PM | Permalink

      Re: Patrick Henry (#28),

      Patrick Henry:
      September 23rd, 2008 at 8:25 pm
      Phil,
      Given that that 2008 is ending with more ice extent than 2007, there will obviously be more multi-year ice going into 2009 than there was going into 2008.

      Actually NSIDC results agree with my analysis in the report they posted today.
      This year we have exchanged first year ice for multiyear ice which suggests that we will enter next summer with a less stable arctic.

      • Maikdev
        Posted Sep 25, 2008 at 1:47 AM | Permalink

        Re: Phil. (#102),

        Not neccesarialy “we will enter next summer with a less stable arctic”

        All the multiyear ice that survives the summer not always survives the winter…

        http://meteo.lcd.lu/globalwarming/Nghiem/rapid_reduction_of_Arctic_perennial_sea_ice.pdf

        Mean values and standard deviations of the areas in million km2 of different
        sea ice classes for the month of March for entire Arctic Ocean
        Year Perennial ice Seasonal ice Mixed ice Total ice area
        2005 4.69 ± 0.10 6.28 ± 0.12 2.98 ± 0.14 14.24 ± 0.13
        2006 4.23 ± 0.09 6.54 ± 0.12 2.99 ± 0.10 13.99 ± 0.16
        2007 3.61 ± 0.13 7.01 ± 0.22 3.23 ± 0.20 14.15 ± 0.09

        Despite a larger summer extent in 2006 than in 2005, for March 2007 there was less MYice than in March 2006. Where was the lost amount of MYice? During autumm and winter it was pushed out from the Arctic through Fram Strait…
        In March 2008 the MYice amount was 1 million sq. km. lower than in march 2007: 2.61 million sq. km. ?? http://seaice.apl.washington.edu/Outlook/2008/
        Now we have 4.5 mil. sqkm of ice. Multiyear amount: 3 million sq. km.???
        If that multiyear ice doesn´t flow out during next winter, the arctic will start the 2009 melt season at least as stable as this year.

        • Phil.
          Posted Sep 25, 2008 at 11:28 PM | Permalink

          Re: Maikdev (#110),

          Re: Phil. (#102),
          Not neccesarialy “we will enter next summer with a less stable arctic”

          Which is why I included ‘suggests’ in front of the part you quoted!

          All the multiyear ice that survives the summer not always survives the winter…

          http://meteo.lcd.lu/globalwarming/Nghiem/rapid_reduction_of_Arctic_perennial_sea_ice.pdf

          Mean values and standard deviations of the areas in million km2 of different
          sea ice classes for the month of March for entire Arctic Ocean
          Year Perennial ice Seasonal ice Mixed ice Total ice area
          2005 4.69 ± 0.10 6.28 ± 0.12 2.98 ± 0.14 14.24 ± 0.13
          2006 4.23 ± 0.09 6.54 ± 0.12 2.99 ± 0.10 13.99 ± 0.16
          2007 3.61 ± 0.13 7.01 ± 0.22 3.23 ± 0.20 14.15 ± 0.09
          Despite a larger summer extent in 2006 than in 2005, for March 2007 there was less MYice than in March 2006. Where was the lost amount of MYice? During autumm and winter it was pushed out from the Arctic through Fram Strait…
          In March 2008 the MYice amount was 1 million sq. km. lower than in march 2007: 2.61 million sq. km. ?? http://seaice.apl.washington.edu/Outlook/2008/
          Now we have 4.5 mil. sqkm of ice. Multiyear amount: 3 million sq. km.???

          The values quoted above are areas not extents, current value is ~3Mm^2 and judging by Drobot’s data I’d think that no more than half is multi-year ice. Looks like less MY ice than March too.

          If that multiyear ice doesn´t flow out during next winter, the arctic will start the 2009 melt season at least as stable as this year.

          If there’s no more more MY ice loss during the winter (and currently the drift isn’t favorable) it will still be behind March 2008.

        • Maikdev
          Posted Sep 26, 2008 at 12:58 AM | Permalink

          Re: Phil. (#145),

          Sorry, area values.
          Now we have 3Mm2. 1.5Mm2 would be MY ice. The first year ice will be in March, by definition, multiyear ice (of second year, but MY). So, in autumm and winter the arctic could lose 0.4Mm2 of multiyear ice and still be above March 2008 (with thinner ice perhaps? Maybe, but not sure)

  19. Gerald Machnee
    Posted Sep 23, 2008 at 8:33 PM | Permalink

    Regarding the future, there have been two types of predictions – an ice free Arctic and open water at the north pole. I have saved a number of clippings from this year by scientists predicting an ice free Arctic by 2013. So I will be able to verify that one before my toast get cold (note that it takes longer to cool with the runaway warming).

  20. Brian Macker
    Posted Sep 23, 2008 at 8:51 PM | Permalink

    “I was reading at the CNN Science site a while ago that the polar bears are turning to cannibalism because of the continued decline in ice.”

    Bears are cannibalistic, as are many other carnovores, insects, even chimps.

    Here’s a study of black bear cannibalism from 1996, which you cannot get the text.

    Here’s another incident from 1962 from another study:

    “Older bears have also been victims of cannibalism. For example, an incident reported by Troyer and Hensel (1962) involved a trapped 170kg female brown bear killed and eaten by an adult female accompanied by a yearling.”

    This is merely a case of ignorance masquerading as evidence.

    Note that the adult polar bear reported eaten in the recent report was cornered in her den. Nothing unusual, move along.

    • bender
      Posted Sep 24, 2008 at 12:17 AM | Permalink

      Re: Brian Macker (#30),
      This does not refute the proposition that the rate of cannibalism has increased.

      • ianl
        Posted Sep 24, 2008 at 12:55 AM | Permalink

        Re: bender (#38),
        Nor does it support it.

        Maybe just another “scary-bear” story to grab headlines (pun intended)

        • bender
          Posted Sep 24, 2008 at 1:02 AM | Permalink

          Re: ianl (#42),
          Whether it is a scare story or not, the rate of cannibalism could be a useful and meaningful indicator of population stress.

        • ianl
          Posted Sep 24, 2008 at 1:06 AM | Permalink

          Re: bender (#43),
          And how will we determine this against unknown rates ?

        • bender
          Posted Sep 24, 2008 at 1:43 AM | Permalink

          Re: ianl (#44),
          By “unknown” you mean “previous”?

        • Dave B
          Posted Sep 24, 2008 at 6:03 AM | Permalink

          Re: bender (#43),

          Kind of like tree rings.

          Prove it.

    • bender
      Posted Sep 24, 2008 at 1:48 AM | Permalink

      Re: Brian Macker (#30),

      Nothing unusual, move along.

      The irony. You know where this line comes from?

  21. frethack
    Posted Sep 23, 2008 at 9:15 PM | Permalink

    Someone may have posted this link in the past (this is my first time to post here), but its been helpful to me.

    Comparative sea ice 1979 to present

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh

  22. crosspatch
    Posted Sep 23, 2008 at 9:40 PM | Permalink

    “I was reading at the CNN Science site a while ago that the polar bears are turning to cannibalism because of the continued decline in ice.”

    Ice is not a bear nutrient. A lack of ice means more seals have to come to the shore to rest and generally makes food easier to find. But it also forces the bears to remain on shore and reduces their foraging area. These bears are very independent and though not extremely territorial, forcing more of them in a smaller area might result in conflicts. Add to the reduced foraging area the fact that polar bear populations are at record high levels and you can begin to see that what some might be reporting as “cannibalism” might really be conflict between animals.

    What I find interesting about this story is that it first appeared in June and got wide press attention. Then suddenly the same story appears again in September and again gets printed by several outlets on the same day. I smell the work of a PR agency here.

  23. DeWitt Payne
    Posted Sep 23, 2008 at 9:52 PM | Permalink

    For the record, minimum extent (JAXA) was on 9/9/2008 at 4707813 km2. There was a secondary minimum on 9/17 at 4718594 km2. Minimum extent in 2007 was on 9/24/2007 at 4254531 km2. There was a near minimum on 9/16/2007 at 4267656 km2. The six year average of JAXA extent data reaches its minimum on 9/18 at 5544389 km2.

    I’m still not officially calling the CT minimum area until I see the exponentially smoothed rate go positive. U.Hamburg area and extent data for September is still about 3 weeks away. The smoothed JAXA extent rate went positive today with the preliminary JAXA data of 4.854844 Mm2 and a difference from the day before of 0.045625 Mm2. The smoothed rate anomaly is now positive so extent is increasing faster than average at the moment.

  24. Posted Sep 24, 2008 at 12:11 AM | Permalink

    David Smith, you have to love a website run by a guy named:

    “Ignatius Rigor”

  25. Tim Ball
    Posted Sep 24, 2008 at 12:14 AM | Permalink

    NASA attributed the reduction in ice of 2007 to changes in wind and consequent ocean circulation change not temperature. As I noted elsewhere it probably also contributed to the piece of ice shelf that broke off.

    http://www.nasa.gov/vision/earth/lookingatearth/quikscat-20071001.html

    • Ernie
      Posted Sep 24, 2008 at 3:59 PM | Permalink

      Re: Tim Ball (#37),

      Nghiem said the rapid decline in winter perennial ice the past two years was caused by unusual winds. “Unusual atmospheric conditions set up wind patterns that compressed the sea ice, loaded it into the Transpolar Drift Stream and then sped its flow out of the Arctic,” he said. When that sea ice reached lower latitudes, it rapidly melted in the warmer waters.

      That’s an interesting theory, the problem with 2007 occurring in winter.

      – Ernie.

  26. Posted Sep 24, 2008 at 12:33 AM | Permalink

    I found an interesting statement when looking for links to old Nimbus 1-3 data last week. It had to do with the formation of a polniya (sorry the spelling is probably off as this is by memory now). What this was was an opening in the Antarctic ice off the Weddell Sea that was quite large and was interpreted at the time as signaling a major shift in sea currents under the ice. As this was in the 60’s during one of the colder spells I was quite interested in this as they said that it only happened at this time, caught by Nimbus, and has not recurred.

    Anyone know anything about this?

  27. Maikdev
    Posted Sep 24, 2008 at 12:34 AM | Permalink

    Curiously, the difference between the NSIDC(SSMI) and IARC/JAXA(AMSR-E) about the sea ice extent minimum is growing each year, from 20.000 square km. in 2006 to 200.000 square km. in 2008. Why??

    NSIDC (SSMI) IARC/JAXA (AMSR-E)
    4.520.000 sqkm 4.707.813 sqkm (2008)
    4.140.000 sqkm 4.254.531 sqkm (2007)
    5.758.000 sqkm 5.781.719 sqkm (2006)

  28. Maikdev
    Posted Sep 24, 2008 at 1:06 AM | Permalink

    Curiously, the difference between the NSIDC(SSMI) and IARC/JAXA(AMSR-E) about the sea ice extent minimum is growing each year, from 20.000 square km. in 2006 to 200.000 square km. in 2008. Why??

    NSIDC/JAXA
    2008: 4.520.000 / 4.707.813 sq. km.
    2007: 4.140.000 / 4.254.531 sq. km.
    2006: 5.758.000 / 5.781.719 sq. km.

  29. AndyW
    Posted Sep 24, 2008 at 1:28 AM | Permalink

    #40 Do they use the same concentratrion level between themselves and all years ? I think they do but am not 100%.

    Thanks for this series of threads Steve by the way.

    I see we have already had one guess for max extent in early 2009 of a whoppa 15.5MM km2 ..wooaa. For fun I am going for a large refreeze but this being tempered by a low starting point so an average 14.0MM being the final score come Q1 2009. I have used exactly 1% science in that choice and 99% “feel”. :D

    Regards

    Andy

  30. CuckooUK
    Posted Sep 24, 2008 at 1:58 AM | Permalink

    The NSIDC show a graph indicating the average sea ice extent from 1979 to 2000, the sea ice extent for 2005, 2007 and 2008.

    I asked the NSIDC why they don’t show the classic climate period of 30 years to indicate the average sea ice extent and why they don’t show the years 2001, 2002, 2003, 2004 and 2006. Their reply was reasonable on the 22 year average (to keep it consistant with previous graphs, although I still don’t see why they couldn’t do a 30 year graph as well – a 30 year average would surely show a lower average sea ice extent than the 22 year average), but they didn’t give a reason why they didn’t show the other years.

    Anybody have an answer or excuse?

    • bender
      Posted Sep 24, 2008 at 2:02 AM | Permalink

      Re: CuckooUK (#51),
      This is the fourth of four threads on this topic. Why not read them and see if your answer is there?

  31. bender
    Posted Sep 24, 2008 at 2:06 AM | Permalink

    Start here and work backwards: sea ice thread #3

    • CuckooUK
      Posted Sep 24, 2008 at 6:09 AM | Permalink

      Re: bender (#53)

      Apologies, bender, that was really useful. The graphs posted pretty much what i thought they would

      thanks for your help

  32. crosspatch
    Posted Sep 24, 2008 at 2:19 AM | Permalink

    The polar bear’s “usual food source” is seals. If there is no ice, the seals must hang out near the shoreline because seals are mammals and need sleep and while they can sleep at sea, they are at risk from orcas and other predators if they do so. When there is a lot of ice, they surface and rest. If there is no ice, they go to shore. This nonsense about polar bears somehow needing ice to survive is nuts. If there is a lot of ice, the seals will be dispersed and the polar bears forage across the ice in search of them. When there is no ice, the seals come to the bears. The last interglacial was warmer than this one has been and there was less ice than there is now. In fact, there might have been no polar bears at all in the last interglacial and the current “polar bear” is simply a brown bear that has been selected due to coat color. The whiter the coat, the more seals they catch, the longer they live, the more cubs they produce in a lifetime, and the trait is selected in over time. That trait might get selected out when there is no ice and the bears need to be “browner” or patchy brown/white to better catch seals on shore.

    Again, ice is not a polar bear nutrient. If anything the lack of ice will result in a decrease in seal populations by keeping them closer to shore where more are eaten by bears.

  33. Perry Debell
    Posted Sep 24, 2008 at 3:40 AM | Permalink

    Re: Larry Sheldon (#2),

    “Polar bears often eat their cubs if they believe they are ill, though zookeepers have no clear picture if that instinct is what motivated the mother bear.”

    http://www.time.com/time/world/article/0,8599,1701828,00.html

  34. kuhnkat
    Posted Sep 24, 2008 at 3:59 AM | Permalink

    Bender,

    “Scientists have noticed increasing reports of starving Arctic polar bears attacking and feeding on one another in recent years.”

    Do you consider this an extensive study??

  35. Chillin'Jim
    Posted Sep 24, 2008 at 6:28 AM | Permalink

    I haven’t seen mention of this anywhere, but I don’t know that I’ve read all 4 of the threads bender referenced.
    And as I’m not a scientist by any stretch, if this is irrelevant, please snip.

    http://news.nationalgeographic.com/news/2008/06/080626-arctic-volcano.html

    It wasn’t clear to me exactly what the timeframe was, as both 1999 and 2007 were mentioned.

    Jim

  36. Kevin B
    Posted Sep 24, 2008 at 6:30 AM | Permalink

    Whilst we should all celebrate the survival of a large amount of baby ice this year, there is absolutely no cause for complacency.

    For a start, if the predictions for a cold arctic winter come to pass then there will be a large amount of new baby ice to worry about next year.

    Then there is the fact that this year’s baby ice will be entering the toddler stage, and we all know the problems that can cause.

    In addition, the amount of teenage ice will increase and teenagers do some incredibly stupid things to impress their peers.

    So, baby ice watchers, celebrate this year’s success, but be ever vigilant.

    (Bender, I apologise if this point has been made before, but I haven’t had time to read every comment on every thread on the site, though I do appreciate all those who maintain and contribute to its splendours.)

    • Gerald Machnee
      Posted Sep 24, 2008 at 7:44 AM | Permalink

      Re: Kevin B (#62),
      Having aging parents, we should also have a safe haven for senior ice. There has been concern lately about the senior shelves and islands breaking up and disappearing.

  37. Patrick Henry
    Posted Sep 24, 2008 at 6:43 AM | Permalink

    phil,

    The ice melts radially inward, so the extent at the end of the melt season is the key parameter. A sliver of first year ice surrounded by second year ice can’t be viewed in the same way as a region of thin first year ice around the periphery.

    Anyway, the 2007 low was primarily the result of a strong transpolar drift, which forced a lot of ice out into the warmer waters of the North Atlantic where it melted.

    • Phil.
      Posted Sep 27, 2008 at 1:08 AM | Permalink

      Re: Patrick Henry (#63),

      phil,
      The ice melts radially inward, so the extent at the end of the melt season is the key parameter. A sliver of first year ice surrounded by second year ice can’t be viewed in the same way as a region of thin first year ice around the periphery.

      Except your concept of how the melt progresses is flawed, this is what the ice looks like in the Beaufort. Fragmented ice like this surrounded by sea water has much more effective heat transfer than if it was a single unbroken floe.

  38. Ashby Lynch
    Posted Sep 24, 2008 at 7:20 AM | Permalink

    How do the GCMs handle the huge heat transfer that takes place in the Arctic. It must involve the specific heat of the water, the specific heat of the air, the temperature of the air, and the temperature of the water. Now, the air circulates and changes its temperature, and the water circulates and changes its temperature. Some areas are covered with ice, so there are three layers in the heat exchange calculation. Then the calculation has to be done for some sort of estimate of all of these changes. Anecdotal evidence seems to indicate that the changes are large and lack a pattern that can be reliably predetermined. What would be the boundaries of the annual heat transfer that takes place in the Artic: what is the year to year potential for differences in heat exchange? For half the year there is no incoming solar radiation, so does CO2 play any direct role when no sunlight is involved? For the rest of the year there is not much incoming sunlight per square unit of the earth’s surface even though the sun will shine continuously. It would be interesting to have someone explain verbally the significance of the Arctic heat transfer mechanism in the overall global heat budget, what factors are most important, and how variability is estimated.

    • DeWitt Payne
      Posted Sep 24, 2008 at 7:32 AM | Permalink

      Re: Ashby Lynch (#64),

      Somewhere on another thread I posted on calculating poleward heat transfer based on the difference between absorbed solar radiation and outgoing long wave radiation. There is definitely net heat transfer from the tropics to the poles. I have it from Gavin himself as co-author of Hansen, et.al. 2005 that direct ghg forcing is much lower at the poles than in the tropics (2 W/m2 compared to 6 W/m2 for doubling CO2, contrary to some conventional wisdom. I don’t know if there’s satellite data over a long enough period to look at variability.

      The calculation of atmospheric poleward heat transfer is probably very dependent on getting the vertical heat transfer in the tropics right. That seems to be a problem for the GCM’s based on the satellite tropical lapse rate data.

  39. Dave Andrews
    Posted Sep 24, 2008 at 7:32 AM | Permalink

    Re polar bear cannibalism,

    The killing of cubs to bring females into estrus (infanticide), or killing of cubs and adults for food is not uncommon in other species of bears and occurs occasionally in polar bears (Taylor et al. 1985; Derocher and Taylor 1994; Taylor 1994; Derocher and Wiig 1999; Dyck and Daley 2002; Amstrup et al. 2006).

    From, ‘Demographic & Ecological Perspectives on the Status of Polar Bears’ by Taylo and Dowsley http://scienceandpublicpolicy.org/images/stories/papers/reprint/taylor_polar_bears.pdf

    The paper also has a discussion of the effects of Arctic ice loss on polar bears and is critical of Kassie Siegel (Centre for Biological Diversity) ans testimomy she gave to Congress earlier this year.

    • mundus
      Posted Sep 24, 2008 at 10:45 AM | Permalink

      Re: Dave Andrews (#65),

      I´m not aware of any scientific paper which confirms a significant increasing of cannibalism among polar bears. As far as I know there are only some few people claiming that they have observed such incidences in the arctic region. It is not clear whether this is a consequence of melting ice. It could also be a result of over-population: I remember the information that the number of polar bears in the Arctic increased since 1970 from about 5.000 to 25.000 until today, mainly due to legal restrictions for bear-hunting. I have to check the source where I read this.
      BTW, the polar bear – like the „hockey stick“ – has become a prominent icon for the AGW-apologists. Pictures of sad looking white bears sitting all alone on a melting piece of ice in the arctic sea were used frequently by Al Gore (in his film “AIT”) and are still published by mainstream media to stoke public emotions and fear about the negative consequences of global warming. Therefore I would be very careful about eyewitnesses reporting cannibalism among polar bears – maybe they are following a biased political agenda.
      There is an interesting paper by Scott Armstrong, Kesten Green and Willie Soon (2008): “Polar Bear Population Forecasts: A Public-Policy Forecasting Audit”. They couldn´t locate “any papers that referred to scientific procedures for making forecasts of polar bear populations”. Armstrong et al. concluded:

      “Without scientific forecasts of a substantial decline of the polar bear population and of net benefits from feasible policies arising from listing polar bears, a decision to list polar bears as threatened or endangered would be irresponsible.”

  40. Steve McIntyre
    Posted Sep 24, 2008 at 7:47 AM | Permalink

    No more polar bears please.

  41. Chris
    Posted Sep 24, 2008 at 8:09 AM | Permalink

    So kind of an anti-climax then? Today’s the day we’ve been awaiting for months – I’m sure many of us would have paid good money earlier in the summer to know what the ice extent would be exactly 365 days after the 2007 minimum! So what’s the result?

    09,24,2007,4254531
    09,23,2008,4873125

    365 days on, the extent is 618,594 km2 or 14.5% greater.

  42. Chris
    Posted Sep 24, 2008 at 8:19 AM | Permalink

    “…So, you might wonder, “Area seems like a more intuitive measurement. Why do scientists generally talk about ice extent instead of ice area?”

    The answer is that scientists are cautious about summertime values of ice concentration and area. To the sensor, surface melt appears to be open water rather than water on top of sea ice. So, while reliable for most of the year, the microwave sensor is prone to underestimating the actual ice concentration and area when the surface is melting…”

    http://www.nsidc.org/news/press/2007_seaiceminimum/20070810_index.html

    I suggest that if you want to compare 2008 extent with 2008 area, the best way to do this is using graphs produced with the same concentration thresholds and data derivation methods, see e.g.

    http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic

    • Phil.
      Posted Sep 25, 2008 at 11:56 PM | Permalink

      Re: Chris (#70),

      “…So, you might wonder, “Area seems like a more intuitive measurement. Why do scientists generally talk about ice extent instead of ice area?”
      The answer is that scientists are cautious about summertime values of ice concentration and area. To the sensor, surface melt appears to be open water rather than water on top of sea ice. So, while reliable for most of the year, the microwave sensor is prone to underestimating the actual ice concentration and area when the surface is melting…”

      It depends on the exact system used, this year with the CT data if meltpools were a significant contributor to the summer melt then we’d expect a rapid increase in area as the air temperature drops, but instead we see it hold steady (last year too).

      • DeWitt Payne
        Posted Sep 26, 2008 at 8:45 AM | Permalink

        Re: Phil. (#148),

        It depends on the exact system used, this year with the CT data if meltpools were a significant contributor to the summer melt then we’d expect a rapid increase in area as the air temperature drops, but instead we see it hold steady (last year too).

        CT may be steady, but at least one other isn’t.

        Arctic ROOS SSM/I ice area chart as of 9/24

        Ice extent chart 9/24

        Both area and extent are rapidly increasing according to them.

  43. Bruno
    Posted Sep 24, 2008 at 9:19 AM | Permalink

    As a non climate-scientist, here’s what I learned (not that much though):
    – Some metrics are useful, others are not. E.g. ‘total melt’ does not make sense at all. Ice that is formed in January (no matter how much), will be gone in September. Only minimum to minimum or maximum to maximum comparisons make sense.
    – It’s colder than last year, ergo there is more ice. The ‘baby-ice’ theory has some merit, but is only one of the many factors others besides surface temperature are wind patterns, ocean currents, weather, autocorrelation,… . I still think the main factor in the longer run (several years) is ‘surface temperature’. Also: the 30year trend is down because the local temperature is up on a 30-year scale.
    – Since were talking about some local (and not global) thing, unpredictable factors (like local weather) have a lot of influence.

    • bender
      Posted Sep 24, 2008 at 9:32 AM | Permalink

      Re: Bruno (#72),

      unpredictable factors (like local weather) have a lot of influence

      It is possible that some semi-predictable factors (like climate response to GHGs) also have an influence. (Unless, of course, you think that effect is zero, or lower.)

  44. Shawn Whelan
    Posted Sep 24, 2008 at 9:37 AM | Permalink

    The August Arctic Sea surface temps are cooler than 2007.

    http://www.nsidc.org/arcticseaicenews/2008/090408.html

  45. Shawn Whelan
    Posted Sep 24, 2008 at 10:02 AM | Permalink

    These graphs from the NSIDC are interesting. They show a similar low level of ice in the past and a quick reversals from a low ice level to a high level and back.

    http://nsidc.org/research/projects/Barry_Eurasian_Arctic.html

  46. Aaron Wells
    Posted Sep 24, 2008 at 10:06 AM | Permalink

    Here are my graphs for yesterday’s ice extent data, based upon revised IARC/JAXA Ice Extent data. Yesterday’s JAXA arctic ice extent increase was so large, that it makes no sense to track the ice extent reduction anymore. So today will be the last of my daily graphs.

    Yesterday’s Julian day (or day of the year) 267, was the 2007 minimum ice extent based on JAXA data. On that day, 2008 ice extent area was closer to 2005’s extent area, due mainly to the fact that 2007’s minimum was later in the season, as was 2005’s, and that 2008 is clearly on a re-freeze tear.

    Also of note, 2008’s “drop-from-peak” value finished just below 2007’s drop from peak maximum (which was yesterday’s Julian day), although for the past few weeks 2008 was slightly higher.

    Daily Arctic Sea Ice Extent Reduction Day 267

    Arctic Sea Ice Extent Day 267

    2008 Arctic Ice Extent Differences From Previous Years Day 267

    Arctic Sea Ice Total Drops from Peaks Day 267

    I would like to give credit to http://www.imageshack.us for hosting all of the graphs that I have posted these past couple of months.

    • Gerald Machnee
      Posted Sep 24, 2008 at 1:53 PM | Permalink

      Re: Aaron Wells (#77),
      **and that 2008 is clearly on a re-freeze tear.**
      With 2008 refreezing rapidly, would that suggest that all that open water did not really absorb as much solar energy as has been suggested by some, which would contribute to runaway melting in the Arctic?

    • Ernie
      Posted Sep 25, 2008 at 10:40 PM | Permalink

      Re: Aaron Wells (#77), Aaron thanks for providing those graphs, I made a point of checking them out when you posted an update, especially the Arctic Sea Ice Extent Day ### graphs that shows the race pretty nicely. How did you make them, is it an “R” script?

      – Ernie.

  47. Jared
    Posted Sep 24, 2008 at 10:31 AM | Permalink

    Thank you for your graphs every day, Aaron, they were one of my favorite ways of tracking the ice.

    As some of us here predicted, it is apparent that the sea ice is now re-freezing faster than it did last year, likely in large part due to the cooler SSTs in the Arctic this year. I’m sure there will continue to be negative spin on what happened this year from some, but overall I’m encouraged by 2008’s recovery from 2007’s drastic melt. I think we will see 2008’s extent quickly catch up to 2005 and other years in the coming weeks.

  48. brendy
    Posted Sep 24, 2008 at 11:16 AM | Permalink

    Current posting on NSIDC seems to acknowledge the key role of wind patterns in producing last year’s minimum as a means of trying to explain away this year’s summer sea ice extent.

    Dispersive sea ice motion

    Sea ice motion also helps determine how the ice will fare each melt season. In 2007, a strong northward sea ice motion at the end of the melt season pushed ice floes together, compacting the ice. The tightly packed ice and high temperatures worked together to create a record-low extent.

    This year, the wind patterns were different, leading to a less compacted ice cover. This, paired with slower summer melt, helped keep the overall extent larger.

  49. brendy
    Posted Sep 24, 2008 at 11:32 AM | Permalink

    Seems intuitive that, all things being equal, a relatively low level of sea ice would follow the year in which there was a record low – particularly if last year’s lost ice was replaced with new “thinner” ice this past winter.

  50. DeWitt Payne
    Posted Sep 24, 2008 at 12:36 PM | Permalink

    I’ll keep this up until I’m positive the minimum has happened which I define as the exponentially smoothed (alpha=0.1) rate going positive. Current smoothed rate is -0.010 Mm2/day.

    CT area update 9/24/2008
    region area(Mm2) difference anomaly
    Arctic 3.048 0.012 -2.066
    Antarctic 14.919 0.055 -0.106

    I’ve recalculated concentration by lagging the JAXA extent compared to the CT area by five days. The plot is much more sensible looking with recent concentration remaining relatively flat and a big decrease in early August. Current concentration corresponding to calendar date 9/19/2008 is 64%.

    Arctic Ice Concentration

  51. Posted Sep 24, 2008 at 1:27 PM | Permalink

    Can I make some observations about Arctic sea ice having read about three thousand posts in this forum!

    I sit on a committee of the British Environment agency concerned with -amongst other things- sea flooding. Generally we protect against a 1 in 50 year or 1 in 100 year flood likelihood

    In the sea the key flooding factors are usually that there has to be an exceptionally high tide, a wind from a very particular direction, a low pressure system in a particular location allowing a storm surge to form, and a considerable volume of water coming down an estuary (where relevant) At least 3 and preferably four of these factors need to coincide for the greatest problems to occur-individually any one is unlikely to cause too big a concern.

    A separate but relevant observation is that sand dune systems are highly vulnerable to wind erosion but this would be reduced if water had damped the surface causing the grains to become heavier, and also that tides can have an awesome power.

    Studies cited here seem to indicate that snow laying on ice will insulate it and help to prevent the ice thickness increasing in winter, whilst the snow in summer acts as an insulation layer preventing it melting. 2007 was particularly (but not exceptionally) sunny. It was also particularly (but not exceptionally) windy. It may or may not have been abnormally snowy in winter preventing ice to readily form. Combine sun (disastrous for snow and ice as any skier knows) with strong winds (perhaps from a unusual direction) breaking up the ice or blowing snow insulation on to or off the ice at the ‘wrong’ times, and consider all the other possible factors including high tides dislodging ice platforms, and perhaps we have the 3 out of 4 or even 4 out of 4 factors needed to cause unusual melt.

    Unusual melt in one year is likely to lead to vulnerable ice the following year. We saw that in 1816/17/18 as we have in 2008.

    Just observations-not a theory- that a variety of external factors working together may have an effect that can considerably accentuate any existing warmth in the sea. During the last three months -from blog comments and my own research- I have been struck by the apparent fragility of arctic ice but the fact that many of the adjacent seas are cooling (page 3 of the executive summary)

    http://www.sesame-ip.eu/doc/MB_Climate_Change_VLIZ_05031.pdf

    suggests that the likelihood is that 2009 will build on 2008 levels, all things being equal with the various factors.

    Tony Brown

  52. Dave Andrews
    Posted Sep 24, 2008 at 4:12 PM | Permalink

    http://www.climateaudit.org/?p=3819#comment-300445

    Sorry Steve this is a polar bear comment, delete if necessary

    Mundus,

    I was trying to point out that occasional cannibalism in polar bear circles appears to happen and that the Centre for Biological Diversity and K Siegel are wrong. I agree with the rest of your post

  53. bender
    Posted Sep 24, 2008 at 4:18 PM | Permalink

    NO BEARS! If I can not comment on bears neither can anyone else!

  54. Bill Jamison
    Posted Sep 24, 2008 at 5:32 PM | Permalink

    I’m not sure this is really the appropriate thread, but did anyone else see this article claiming that huge amounts of methane are being released due to warming of the arctic and melting of the ice?

    http://www.independent.co.uk/environment/climate-change/exclusive-the-methane-time-bomb-938932.html

    The first evidence that millions of tons of a greenhouse gas 20 times more potent than carbon dioxide is being released into the atmosphere from beneath the Arctic seabed has been discovered by scientists.

    The Independent has been passed details of preliminary findings suggesting that massive deposits of sub-sea methane are bubbling to the surface as the Arctic region becomes warmer and its ice retreats.

    Is there such a thing as “under-sea permafrost”?

    • DeWitt Payne
      Posted Sep 24, 2008 at 5:39 PM | Permalink

      Re: Bill Jamison (#89),

      That would be methane clathrates. And the real evidence would be an increase in atmospheric methane, which isn’t happening, yet.

      • Bill Jamison
        Posted Sep 24, 2008 at 7:46 PM | Permalink

        Re: DeWitt Payne (#90),

        Thanks! I also found a paper published last year on the same subject, apparently this isnt’ really news at all or at least not the shocking news it is made out to be.

        http://www.cosis.net/abstracts/EGU2007/01071/EGU2007-J-01071.pdf?PHPSESSID=e

        Did anyone else notice the pic they used? It is a stock photo from Almay and has NOTHING to do with methane bubbling up but of course most people will assume the ice is actually foam from the methane.

      • Phil.
        Posted Sep 24, 2008 at 8:13 PM | Permalink

        Re: DeWitt Payne (#90),

        Re: Bill Jamison (#89),
        That would be methane clathrates. And the real evidence would be an increase in atmospheric methane, which isn’t happening, yet.

        Are you sure about that?

        • DeWitt Payne
          Posted Sep 24, 2008 at 8:52 PM | Permalink

          Re: Phil. (#95),

          Do the annual average as in this figure from the US EPA site. The seasonal variation obscures whether the trend for the last decade or so is flat or not. If increasing Arctic temperature produces more methane, then the trend should be accelerating not what looks like from eyeballing the monthly data, decelerating.

        • DeWitt Payne
          Posted Sep 24, 2008 at 9:17 PM | Permalink

          Re: DeWitt Payne (#97),

          That raises an interesting point. Considering the large seasonal variation in atmospheric methane, it’s possible it’s a proxy for at least NH temperature. If it indeed has been flat since 1998, then it’s independent evidence that the global average temperature has been flat too. Which is unfortunately OT for this thread.

        • Phil.
          Posted Sep 24, 2008 at 9:56 PM | Permalink

          Re: DeWitt Payne (#101),

          Re: DeWitt Payne (#97),
          That raises an interesting point. Considering the large seasonal variation in atmospheric methane, it’s possible it’s a proxy for at least NH temperature. If it indeed has been flat since 1998, then it’s independent evidence that the global average temperature has been flat too. Which is unfortunately OT for this thread.

          Actually that’s a proxy for OH concentration.

        • Gerald Machnee
          Posted Sep 24, 2008 at 9:06 PM | Permalink

          Re: Phil. (#95),
          **Are you sure about that?**
          That is six year old data and not necessarily representative of the globe.

        • Earle Williams
          Posted Sep 24, 2008 at 9:28 PM | Permalink

          Re: Gerald Machnee (#100),

          Hey, it was a timely argument in 2002!
          ;-)

        • DeWitt Payne
          Posted Sep 24, 2008 at 9:35 PM | Permalink

          Re: Gerald Machnee (#100),

          Here’s the most recent methane data that I could find from NOAA. It shows an increase starting in 2007, but it’s still relatively flat for the last ten years.

  55. crosspatch
    Posted Sep 24, 2008 at 5:58 PM | Permalink

    “the real evidence would be an increase in atmospheric methane, which isn’t happening”

    In fact, the latest data I saw showed atmospheric methane was stable to declining for reasons unknown.

    Chemist Isobel Simpson led the research examining samples from 1998 through 2005 and found that methane levels had practically stopped rising, reaching 1,772 ppbv in 2005. During this period some years did see rises while others actually saw slight decreases, according to the paper presenting the result in the November 23 issue of Geophysical Research Letters. By also measuring levels of ethane (C2H6) and perchloroethylene, or perc, (C2Cl4) the researchers determined that these pulses in methane levels during this period could be linked to major forest fires, such as the massive burn in Indonesia from late 1997 to early 1998. “All three of these molecules are removed by the same process–reaction with hydroxyl,” a radical formed from water in the atmosphere, explains Nobel Prize-winning chemist F. Sherwood Rowland, who participated in the research. “Both methane and ethane are produced in biomass burning, but perc is an industrial solvent. If biomass burning is the source, then perc [levels] should behave quite differently from the two hydrocarbons, and this is what we observed.”

    LINK

  56. bender
    Posted Sep 24, 2008 at 6:03 PM | Permalink

    I saw a study published recently in Science that showed that there are chunks of permafrost in the arctic that are over 700,000 years old. i.e. The arctic hasn’t been “permafrost free” for a very, very long time, despite two periods warmer than the current interglacial (two periods which the unmentionable polar bears also survived).

  57. MattN
    Posted Sep 24, 2008 at 7:24 PM | Permalink

    Very clearly according to Aaron’s graphs, 2008 minimum came MUCH earlier than 2007. At least 2 weeks earlier. I appologize for this question that I’m sure has been answered elsewhere, but what is 2008’s minimum vs 2007’s minimum? (I have a new job that has drastically cut my internet time…) Eyeballing his graph it appears to be ~400K.

  58. Michael Hauber
    Posted Sep 24, 2008 at 10:45 PM | Permalink

    Based on the official IPCC position i say that CO2 accounts for around 30% of total climate variation.

    http://en.wikipedia.org/wiki/Image:Radiative-forcings.svg

    Just reading of their climate forcings chart I read:

    Co2 1.7
    Halocarbons 1
    Ozone .4
    Stratospheric H2O 0.1
    Black Carbon 0.1
    Solar 0.1
    Land Use – 0.2
    Aerosols – 1.3

    Net 1.6
    Total +ve 3.3
    Total -ve 1.5
    (they don’t add up because I’m reading them off the chart)
    If percent Co2 = co2 / net influence then co2 is actually more than 100% of the climate change influence. However I prefer to divide Co2 by total _ve + total +ve = 4.8, to get a Co2 percentage of about 35%.

    Assuming that unforced variations due to ocean circulations play a small part as well (they are not an external forcing, but included (in theory) in the model dynamics), this leaves a value of 30% of climate change being due to Co2.

    On sea ice I find the fact that so much of the ice left is 1st year rather, meaning much less multi year ice to be somewhat disturbing. A cool year, less unfavourable winds + lots of first year sea ice led to a melt close to last year’s unusually high record melt. If we get a warm year + even more first year ice next year I suspect the melt record may be broken by a long way.

    (of course if it keeps getting colder in the future the ice will have to grow sooner or later)

  59. AndyW
    Posted Sep 25, 2008 at 1:29 AM | Permalink

    That Independent article made me laugh. Methane is “bubbling up” but they did not give a massive amount of actual figures on the subject. No doubt the kayak person farted but I doubt that has any effect either. Then they assign it to global warming. Why not the volcanic activity? We know seabed volcanic activity will not melt ice at the sea surface but it will have a better chance of warming frozen methane on the seabed.

    On another topic where has this idea of 2008 having a rapid refreeze compared to 2007 come from? We are still in September and the graphs are still shallow. It was 2007 that had the rapid refreeze in October. It’s too early to tell yet about the refreeze rate. All you can say is the minima happened early this year.

    Regards

    Andy

    • Bill Jamison
      Posted Sep 25, 2008 at 1:49 AM | Permalink

      Re: AndyW (#108), regarding 2008 refreeze rate…

      NOAA is predicting an early return of ice due to colder water and cold air moving in:

      WATER TEMPERATURES ACROSS THE CHUKCHI AND BEAUFORT
      SEAS ARE UP TO 5 DEGREES COLDER THIS YEAR THAN LAST YEAR. SEA SURFACE
      TEMPERATURES ALONG THE COAST FROM WAINWRIGHT TO KAKTOVIK HAVE DROPPED
      TO LESS THAN 2 DEGREES CELSIUS. WEATHER FORECAST MODELS ARE BRINGING
      COLDER AIR OVER THESE AREAS LATE THIS WEEK. IN 2007 SIGNIFICANT SEA
      ICE HAD FORMED ALONG THE ALASKA NORTH COAST BY THE 7TH OF OCTOBER. IN
      2008 SEA ICE WILL RETURN EARLIER TO ALASKA WATERS.

  60. AndyW
    Posted Sep 25, 2008 at 1:46 AM | Permalink

    Tony,.my wife feels the cold earlier than your butter. I think it is still fine but she is demanding we put the heating on. I pointed out where the jumper drawer was and told her to start walking.

    With current gas and electricity prices it’s time for her “extensive” wardrobe to be bought into play.

    Regards

    Andy

    • Posted Sep 25, 2008 at 5:01 AM | Permalink

      Re: AndyW (#109),

      Every sympathy-its what jumper drawers are for.

      It is not generally known that it is actually illegal to put on central heating before the butter/soft spread changeover day. As my datasets show it will be Oct 5th this year, tell your wife it is 10 days earlier than normal for which she should be thankful for the apparent reversal of AGW.

      With regards to Methane, someone ought to go and check ancient Russian folklore where igniting methane gas gave rise to such things as Will o the wisps-although I dare say they werent called that in Russian

      Tony Brown

  61. TAC
    Posted Sep 25, 2008 at 3:20 AM | Permalink

    The Next Generation of Baby Ice Comes to Life
    (nearly 100K upswing in past 2 days)

    9 23 2002 6.001563 0.015782
    9 23 2003 6.080469 0.033438
    9 22 2004 5.925625 0.058906
    9 23 2005 5.335156 0.020000
    9 23 2006 5.883281 0.031875
    9 23 2007 4.267344 -0.009375
    9 22 2008 4.809219 0.035469

    9 24 2002 6.047969 0.046406
    9 24 2003 6.101719 0.021250
    9 23 2004 5.994688 0.069063
    9 24 2005 5.377969 0.042813
    9 24 2006 5.972188 0.088907
    9 24 2007 4.254531 -0.012813
    9 23 2008 4.873125 0.063906

    Celebrating this moment of renewal begs for a poet’s touch. Where’s jeez?

  62. Michael Hauber
    Posted Sep 25, 2008 at 3:44 AM | Permalink

    Interesting that extent is on the way up whereas Cryosphere Today’s area is still at bottom. Usually area goes up first and then extent I think?

  63. Maikdev
    Posted Sep 25, 2008 at 4:18 AM | Permalink

    “Usually area goes up first and then extent I think?”

    I think so. A problem with the algorithm used by Cryosphere Today and NSIDC?

    http://seaice.bplaced.net/nsidc/

    ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/nasateam/

    http://arctic.atmos.uiuc.edu/cryosphere/iphone/iphone.currentarea.series.html

    Using the same satelite data (SSMI) but their own algorithm, fron Norway report that the both, area and extent are going up:

    http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic

  64. Nylo
    Posted Sep 25, 2008 at 4:51 AM | Permalink

    I have a question regarding the importance of the loss of the ice in the Arctic. It is often said by the alarmists that the loss of the ice is a positive feedback too because it implies a loss in Albedo, as the ice reflects more light than the water. But I have some doubts about this, mainly because the water absorption of light depends a lot on the angle with which the light comes.

    Anyone who has seen a sunrise or a sunset in the sea knows that the water shimmers with a lot of brightness if you look in the direction of the sun. Most of the light is being reflected upwards, as the sea acts almost like a mirror when the sun is under a certain angle. But the thing here is that, in the arctic, the sun is always low! And by the time we reach the ice minimum in early september, it is really low. How does this affect the albedo difference between the ice and the water? Do they become similar? Maybe the water albedo becomes even greater? Has anybody thought about this before?

    Thanks.

    • bender
      Posted Sep 25, 2008 at 5:24 AM | Permalink

      Re: Nylo (#115),

      in the arctic, the sun is always low

      It gets high enough in summer that insolation is relatively strong, albedo is relatively weak. “always” ignores huge seasonal variation.

      Has anybody thought about this before?

      Yes, people who are paid to build climate models for a living. In fact they do more than just think, they measure. Sea ice albedo has been measured over the course of a season. It drops significantly as the ice melts. Probably hits an overall minimum between summer solstice when sun angle is most direct and early September when sea ice extent in minimal.

      • Nylo
        Posted Sep 25, 2008 at 5:46 AM | Permalink

        Re: bender (#117),
        Sorry Bender, I wrote before I read your comment.

        I understand what you say. However I don’t have the details about how the albedo measures were taken. If the measures are taking only when the sun is at its highest, they will favour ice, no surprise. But they should be taken all along the day and integrated. Also the heigth of the instrument is important. It should be as close to the ice / water as posible, because the ice disperses the light more than the water. But what I would really need (but I don’t know if it exists) is a formula telling how much light the water reflects depending on the angle of incidence of the light, and the same for the ice, and then integrate it with the height of the sun during a day, for the different days of the year.

        I have no doubt that at the summer solstice, albedo is lost. What I am wondering is whether at the ice minimum in early september albedo could be gained because of the sun being already too low.

        • Mark
          Posted Sep 25, 2008 at 9:59 AM | Permalink

          Re: Nylo (#120),

          From this source:

          http://www.fas.org/spp/military/docops/afwa/ocean-U1.htm

          The percent of the Sun’s electromagnetic energy (all wavelengths) reflected by the surface of an object back into space is known as albedo. The albedo of water varies with the angle of incidence. The albedo for water is 100 percent when the Sun is very low on the horizon, and decreases to 2 percent when the Sun is directly overhead. This decrease is not linear. After the Sun is more than 25 percent above the horizon, the albedo is less than 10 percent. The angle of incidence varies with the latitude of the water surface, the season of the year, and the time of day. Generally, the albedo of water is quite low. The albedo of the ocean varies in some parts of the world due to prolonged periods of cloud cover, large areas of sea ice, and intense atmospheric pollution.

          Also check:

          At autumn equinox, when sea ice is at minimum, the peak sun angle is less than 24 degrees at high noon, so for half the day you’re looking at angles of 12 degrees or less at which point water is starting to get just as reflective as ice! Another key factor is melt water on top of ice – it’s albedo would be similar to open water so a higher area of ice melt doesn’t necessarily translate into less sunlight being reflected!

        • Mark
          Posted Sep 25, 2008 at 10:05 AM | Permalink

          Re: Mark (#128),

          Also note that the end of previous post was specific to a point on the Arctic Circle while any areas of marginal melt at Autumn solstice while be at even higher latidtudes thus meaning that albedo will tend to be even higher most of the day!

        • Mark
          Posted Sep 25, 2008 at 10:12 AM | Permalink

          Re: Mark (#129),

          And again, one most remember that on average only 1.4% of incident radiation strikes the earth within the Arctic Circle. So changes to ice levels within the Arctic area especially when the incident angle albedo effects are taken into account, cannot have major implications in terms of heating or cooling the planet as a whole.

  65. Nylo
    Posted Sep 25, 2008 at 5:30 AM | Permalink

    I found this in a website talking about albedo:

    Las nubes son las que mejor reflejan la radiación solar, mientras que las superficies de agua son las peores; sin embargo, hay que matizar que en caso de que el Sol se encuentre cercano al horizonte la reflexión del agua puede aumentar considerablemente

    I will translate: “The clouds are the ones which better reflect the sun light, while the water surfaces are the worst; however, we have to concede that in case the sun is close to the horizon, the water reflection can increase considerably“.

    This is what I was talking about in #115. In the arctic, the sun is low, and water albedo may not be very different from ice albedo.

    • bender
      Posted Sep 25, 2008 at 5:38 AM | Permalink

      Re: Nylo (#118),
      I answered your question. Water albedo is indeed very different from ice albedo which is very different from snow albedo which, yes, is very different from cloud albedo.

  66. Richard111
    Posted Sep 25, 2008 at 6:06 AM | Permalink

    Arctic sea water albedo for calm seas is higher than for rough seas.
    Is this worth taking into account and? I assume it is possible
    to quantify for a calm sea surface at low sun angles.

  67. Richard111
    Posted Sep 25, 2008 at 6:08 AM | Permalink

    Ooops… discussed above. Appologies. :-(

  68. AndyW
    Posted Sep 25, 2008 at 6:30 AM | Permalink

    Although NOAA are predicting a quick refreeze it has not happened yet. This summer should show that predictions are often flexible due to new data.

    Even if there is a rapid refreeze it seems, from the last few years, that the extent is similar going into Dec/January and it is only after this time that the various years spread out to have large or small maxima. Therefore I think it’s best to make an accurate guess on total winter 2008/9 extent around Feb 1st! All guesses upto then are just for fun and talk about rapid refreeze now as a guide is a bit pointless. That won’t stop me doing it of course though, cough :)

    Regards

    Andy

  69. crosspatch
    Posted Sep 25, 2008 at 9:40 AM | Permalink

    “Although NOAA are predicting a quick refreeze it has not happened yet.”

    I think it would be quite unreasonable to expect to see dramatic refreezing yet. Looking at the charts of the past several years, I think we have another week or so before we really start to see the amount of ice begin to curve upwards. What the slope of that curve is going to be and where it will top out is anyone’s guess but I do believe that a couple of expectations could be reasonable. 1: Since this year’s freeze is starting with more ice than last year’s. Given an “average” freeze, we would probably expect to see the ice top out higher than last year’s did. 2. Since next year’s ice will have more multi-year fresh ice and less first year “brine ice”.

    So I think a reasonable prediction would be a good probability of 2009 having a greater ice extent than 2008 at both maximum and minimum extents even if conditions are “normal” for the winter. Any predictions of major deviations from normal would be only for a week or two, possibly a month into the future. Weather predictions become less reliable as you look farther into the future. And my prediction of 2010 would be that I see it having equal probability of being either greater or less than 2009 at this time.

  70. crosspatch
    Posted Sep 25, 2008 at 9:42 AM | Permalink

    Oops: “Since next year’s ice will have more multi-year fresh ice and less first year “brine ice”. ” I chopped that of when I went for coffee.

    Meant: “Since next year’s ice will have more multi-year fresh ice and less first year “brine ice”, it would be reasonable to assume there will be more of it at the end of the 2009 melt season given a “normal” summer than there was in 2008.

  71. M. Jeff
    Posted Sep 25, 2008 at 9:57 AM | Permalink

    crosspatch

    Your comments suggest that you are knowledgeable on the issues under discussion. I recently quoted you in an e-mail, but the correspondent questioned the usefulness of information from an anonymous poster. Do you have professional credentials that might enhance your status to the correspondent in question?

  72. Tim Ball
    Posted Sep 25, 2008 at 9:58 AM | Permalink

    Re #85

    If you read it properly it says “a rapid decline in winter perennial ice” which means it is occurring in the spring and summer, it does not mean it was melting in the winter.

  73. Tim Ball
    Posted Sep 25, 2008 at 10:19 AM | Permalink

    We read reports about thinner ice and that is likely due to changing wind and current patterns carrying warmer water under the ice.

    This changes another factor rarely mentioned and that is the transfer of heat through the ice from the warmer water (it does not drop below -1.6°C) to the colder air, so the sea modifies air temperatures even when frozen. There is a rapid change in temperature regimes between the surrounding land and the Arctic Ocean because of the Ocean as a heat source. Unfortunately, as the Arctic Climate Impact Assessment (ACIA) report acknowledges (Chapter 1. Page 3. Figure 1.3), we have virtually no temperature data for air over most of the Arctic Ocean.

    • Shawn Whelan
      Posted Sep 25, 2008 at 10:35 AM | Permalink

      Re: Tim Ball (#131),
      Then is -1.6deg C the approximate freezing point of the Arctic Ocean?

      • Tim Ball
        Posted Sep 25, 2008 at 11:43 PM | Permalink

        Re: Shawn Whelan (#133),
        Sea water acts differently than fresh water as it cools. Fresh water reaches maximum density at 4°C with fusion at 0°C. Maximum density in sea water is close to the fusion point at approximately -2°C, although this varies with salinity.
        The Arctic Ocean is the least saline of the oceans due to low evaporation, another factor generally overlooked in studying the Arctic climate, and the influx of large volumes of freshwater from the massive rivers of Russia and Canada. So yes, -1.6°C is the approximate freezing point of the Arctic waters.

        • Shawn Whelan
          Posted Sep 26, 2008 at 6:00 AM | Permalink

          Re: Tim Ball (#146),
          With all the talk of uncovered ocean allowing the intake of heat and leading to more heat in the ocean one never hears of the opposite. The uncovered ocean will be able to more easily release it’s heat and cool quicker which would explain the rapid reversals of ice extent in the past.

          Looking at the history of Arctic ice it is ridiculous to feed inputs into a computer and make an irreversible trend line downward and expect that the ice cannot quickly reverse itself.

          Now this year with the cooling PDO, LaNina, low amount of activity from the Sun, the lower heat in the Arctic Ocean and the open ocean and an earlier freeze is starting in the Arctic. Gonna be interesting to see the result.

  74. brendy
    Posted Sep 25, 2008 at 10:50 AM | Permalink

    Re the Cryosphere data, here is there response to my inquiry on the subject.

    The NSIDC reports sea ice extent.
    They assume that any grid cell with more than 15% ice concentration is
    totally covered by sea ice. On the cryosphere today, we report ‘ice
    area’. Ice area weights the coverage by the concentration (fraction)
    of sea ice in the grid cell. So, if the satellite says a grid cell is
    50% covered by ice and the area is 4000 sq. km, we add 2000 sq. km to
    our area totals, NSIDC would add the total 4000 sq. km.

    There’s no right or wrong way to do it, each has its own merits. Just
    a different methodology.

    Thanks for your interest.

  75. Moptop
    Posted Sep 25, 2008 at 11:00 AM | Permalink

    The albedo of water varies with the angle of incidence. The albedo for water is 100 percent when the Sun is very low on the horizon, and decreases to 2 percent when the Sun is directly overhead.

    This brings us back to the Holocene Optimum, when the Sun was more directly overhead in the Arctic and the summers were longer. GRIP shows it was warmer on average in the Arctic at that time than today. Even then, the GRIP core smooths out peaks and valleys, so it seems reasonable to infer that there were years far warmer than any that have occured this century. Yet still, the polar bear survived. Mile thick glaciers melted, yet the methane was not released in massive quantities from the permafrost. Kind of makes you wonder.

  76. Posted Sep 25, 2008 at 11:15 AM | Permalink

    “Sea ice in the Arctic appears to have passed its minimum extent for 2008 without breaking last year’s record. “

  77. Mike Bryant
    Posted Sep 25, 2008 at 11:23 AM | Permalink

    Just for fun maybe some of the best prognosticators here could take a guess at arctic sea ice extent for October 1st, November 1st and December 1st.

    • bender
      Posted Sep 25, 2008 at 6:06 PM | Permalink

      Re: Mike Bryant (#137),
      And 2010, 2020, 2030?

    • AndyW
      Posted Sep 25, 2008 at 11:52 PM | Permalink

      Re: Mike Bryant (#137),

      Not sure about expert prognosticators, but if average prognosticators can have a basj, just for fun I would say

      Oct 1 = 5.5
      Nov 1 = 8.5
      Dec 1 = 10.0

      I’m not convinced the supposed cold weather will cause anything but a normal run up to January.

      Regards

      Andy

  78. crosspatch
    Posted Sep 25, 2008 at 3:46 PM | Permalink

    “Your comments suggest that you are knowledgeable on the issues under discussion. I recently quoted you in an e-mail, but the correspondent questioned the usefulness of information from an anonymous poster. Do you have professional credentials that might enhance your status to the correspondent in question?”

    I am an engineer by profession in a different field. My interests have been in various fields of science since I was a youngster and I do read on the subject from a variety of sources on nearly a daily basis. I am also not so much interested in taking a “side” of the issue so much as I am looking objectively at what data shows. So far I have learned that the data used by various agencies (GISS being apparently the worst) is of questionable quality and nobody has been able to prove to my satisfaction that recent warming has been either unprecedented or particularly rapid. In fact, it appears to me that we have undergone substantial “global cooling” recently. “Garbage In, Garbage Out” works no matter what field one is in and if the temperature data going in is skewed by urbanization and land use changes along with generally poor siting of the measuring stations, the data resulting from that input will also be skewed. I don’t think one needs any special training in climatology to understand that ice with less salt melts at a higher temperature or that as ice ages, the salt works out of it and it becomes denser due to compression. Those are basic principles of science that practically anyone can learn. It has been my experience that if one can not counter the argument, they attempt to attack the person making it. It is a shifting of focus from the actual argument being made to the person making it. It is like you saying that water is hydrogen and oxygen and someone questioning your credentials by asking if you are a chemist as if only a chemist could make that kind of statement with any certainly. It is an attempt to dismiss the argument by attempting to dismiss the person making it. In other words, your correspondent changed the subject from what I said to what qualifications I have to say such a thing. I am flattered. It means they had nothing to argue the point with. When in a debate with someone, then they go ad hominem, I generally take that as victory.

  79. M. Jeff
    Posted Sep 25, 2008 at 6:01 PM | Permalink

    crosspatch

    Thanks for the information.

  80. Mike Bryant
    Posted Sep 25, 2008 at 6:31 PM | Permalink

    No… the CCMs are probably already doing that.

  81. Mike Bryant
    Posted Sep 25, 2008 at 6:48 PM | Permalink

    I mean GCMs… sorry
    I thought it might be fun to see month to month who is the best seer… I guess not…

  82. Evan Jones
    Posted Sep 25, 2008 at 9:32 PM | Permalink

    Well, it wasn’t an “Incredible Journey” for Baby Ice. But at least it was like one of those zombie movies where–some–of the actors survived.

    Baby Ice, we hardly knew ye . . .

  83. Flanagan
    Posted Sep 26, 2008 at 4:38 AM | Permalink

    For those who are still interested in extents:

    09,25,2007,4265000
    09,26,2007,4297813

    09,24,2008,4878750
    09,25,2008,4865625

    Yesterday’s extent is 0.0131 less than the day before (not negligible, although preliminary) – the gap with 2007 being now 567k.

  84. Flanagan
    Posted Sep 26, 2008 at 6:39 AM | Permalink

    Shawn and Tim: I would not say we have such an early refreeze. It is actually very similar to what happened in 2004 (in terms of dates). Moreover the area (which is more representative of the amount of ice) does not increase a lot for the moment.

    What I still do not fully understand is why the antarctic extent is so low this year – an all (satellite-era) time lowest maximum or close to it. I know variability is high down there, but what were the leading factors this time? Does anybody have an idea?

    • Shawn Whelan
      Posted Sep 26, 2008 at 10:27 AM | Permalink

      Re: Flanagan (#152),
      Climate changes, it always has and it always will.

    • Phil.
      Posted Sep 27, 2008 at 12:34 AM | Permalink

      Re: Flanagan (#152),

      What I still do not fully understand is why the antarctic extent is so low this year – an all (satellite-era) time lowest maximum or close to it. I know variability is high down there, but what were the leading factors this time? Does anybody have an idea?

      It’s not that low, maximum area is bang on average.

  85. BarryW
    Posted Sep 26, 2008 at 9:44 AM | Permalink

    I haven’t seen this posted so I thought this might be of interest link

  86. DeWitt Payne
    Posted Sep 26, 2008 at 10:30 AM | Permalink

    CT area update 9/26/2008
    region area(Mm2) difference anomaly
    Arctic 3.133 0.063 -2.044
    Antarctic 14.807 -0.057 -0.256

    The CT smoothed Arctic rate has gone positive. Combined with the graph from Arctic ROOS and the CT area race is officially over. Minimum was 3.004 Mm2 on calendar date 9/13/2008. I posted the JAXA final extent results above.

    The Antarctic looks to have peaked as well at 14.998 Mm2 on 9/12/2008.

    Note that for CT, the calendar date may not correspond to the actual date. My best guess at the moment is that the actual date is 5 days less than the calendar date. I’ll have a more robust estimate when the Uni-Hamburg data for September is posted in about two weeks.

  87. Posted Sep 26, 2008 at 10:38 AM | Permalink

    Off topic but this is where all the brains hang out…

    Can anyone point me towards a graph (a real one not Dr Manns) showing the actual co2 content from say the beginning of the 20th century through to today with an overlay of world temperatures to the present day (not to 1998)Ive done a search but can’t find a match. The Keeling curve doesn’t seem to match the search and seems heavily adjusted, doesn’t go far enough back and never seems to include temp levels anyway.

    Thanks

    Tony Brown

    • BarryW
      Posted Sep 26, 2008 at 11:05 AM | Permalink

      Re: Tony Brown (#157),

      Try JunkScience.com

    • Posted Sep 26, 2008 at 2:06 PM | Permalink

      Re: Tony Brown (#157),

      Thanks Barry W and Flanagan. What a great site woodfortrees.org is. However neither did what I wanted.

      I have plotted co2 and temperature levels on graph paper and it seems to me that unless they are hugely adjusted (you’re a cynical man Shawn) the graphs dont have anywhere near the correlation they are supposed to have. I don’t have the knowledge to superimpose one graph on another electronically but someone must have done this before as surely it is the first thing that climate modelers would look at. (Isn’t it? Someone reassure me.)

      The search continues and further signposts are appreciated.

      Tony Brown

      • Phil.
        Posted Sep 26, 2008 at 2:48 PM | Permalink

        Re: Tony Brown (#163),

        Since the dependence on CO2 is log([CO2]) then that plot would only make sense if you plotted log([CO2]), “a huge adjustment”.

        • Posted Sep 27, 2008 at 1:35 AM | Permalink

          Re: Phil. (#164),

          Hi Phil

          I dont want to hijack this thread by going off topic but there are three or four people posting here who will know the answers

          We discussed the log efect on temp over on unthreaded some weeks back and all the charts were posted

          http://www.climateaudit.org/phpBB3/viewtopic.php?f=4&t=344

          It still doesn’t really clarify a number of related things

          *We know the CO2 levels over a long period
          *We know the temperatures over a long period
          *Others say that doubling CO2 (from pre industrial levels) will put 3 degrees c on the temp
          *We know what the temp should therefore be now (according to the actual levels of CO2 recorded) so it should be possible to plot actual co2 levels against what the temp actually is, and what the log effect says it should be.

          The trouble is, what is considered the start date for temperatures against which the 3 degrees rise will occur?? The preindustrial age (at 280ppm) covers most of mans history yet there are various times when even Al Gore says the temperatures were warmer than now. So does the co2/temp cslculation start in the 1730’s-same temp as now?
          The 1100’s much warmer than now? The 6th century -much warmer than now?

          What is considered the start point against which we should plot- and if we are starting from a low temp rather than a high one-why?.

          Thanks

          Tony Brown

        • See - owe to Rich
          Posted Sep 27, 2008 at 4:50 AM | Permalink

          Re: Tony Brown (#168)

          Tony, I should step in here and say that I followed a similar thought process to you a year ago, except I decided to add solar cycle lengths to the equation as well, with this graph as a result.

          I now have a slightly improved graph which I will post within a couple of weeks, including allowance for removal of bias suggested by McKitrick & Michaels. Anyway, the bumps and hollows in temperature are reasonably well fitted by the solar cycles, while I choose to attribute the upward trend to the log(CO2) effect. And over 150 years you can’t get a good fit with sensitivity as high as 3.0C.

          HTH,
          Rich.

        • Phil.
          Posted Sep 27, 2008 at 12:28 PM | Permalink

          Re: See – owe to Rich (#169),

          Did you include major volcanic eruptions?

        • Ernie
          Posted Sep 27, 2008 at 7:12 PM | Permalink

          Re: Phil. (#171), You have to be careful with incorporating volcanoes into a model, only the ones that emit large volumes Sulphur compounds have a major impact on blocking the sun, the ones that spit out mostly silicon do not. eg the big eruption in Chile this year was mostly silicon.

          – Ernie.

        • Phil.
          Posted Sep 27, 2008 at 7:36 PM | Permalink

          Re: Ernie (#174),

          Which is why I used the word ‘Major’, in the time-frame being considered, Krakatoa, Soufriere, Agung, El Chihon, & Pinatubo would all qualify.

        • See - owe to Rich
          Posted Sep 28, 2008 at 1:27 PM | Permalink

          Re: Phil. (#171) and major volcanic eruptions.

          Nah, that’s just weather noise. We’re talking big league climatic variations here, the Sun (think TGGWS) and CO2 (think AIT).

          Seriously, Pinatubo probably cooled Earth by 0.2C for 2 years, which out of my 11 year means is less than 0.04C. My standard errors for the 11-year mean are about 0.07C, so, yes, Pinatubo would have a bit of an effect, but less than the standard error. And my goal has been to distil climate to means of a decent span, with relatively few degrees of freedom (but smaller autocorrelation problems), and not try to fit too many variables.

          Against that background I am waiting for someone to beat my model. Actually, someone suggested a trend plus 60-year sine wave, which did beat it by a small margin, but which would have been useless for projections backwards from 1850, where it would predict high temperatures just where the Dalton Minimum was occurring.

          Can’t say too much here, as it’s OT and I’m busy with something else for next 2 weeks. More discussion on my BB thread might be a good idea.

          Rich.

          P.S. Aaron I’d like to see your graphs reach the end of their natural span.

        • Phil.
          Posted Sep 28, 2008 at 3:21 PM | Permalink

          Re: See – owe to Rich (#180),

          So what model do you use for solar insolation?

        • Posted Sep 27, 2008 at 1:17 PM | Permalink

          Re: See – owe to Rich (#169),

          That was a great graph-thanks. I see you thought about going back to 1790 for the solar cycle-did you intend to do that with temperature also?

          Surely results -and any correlation between CO2/temp-will depend on the start point? If you start in say 1900 with low temperatures immediately after the LIA, you will end up with different results than starting in say 1730 with similar CO2 levels to 1900 but much higher temperatures than that year. It would be interesting to see that graph extrapolated way back.

          Re: Phil. (#171),

          I had understood Volcanoes reduced temperature due to the particles obscuring the sun but do major eruptions actually put a significant measurable amount of CO2 into the atmosphere?

          Tony Brown

        • Phil.
          Posted Sep 27, 2008 at 3:57 PM | Permalink

          Re: Tony Brown (#172),

          I had understood Volcanoes reduced temperature due to the particles obscuring the sun but do major eruptions actually put a significant measurable amount of CO2 into the atmosphere?
          Tony Brown

          Precisely, so you wouldn’t be able to get a good match to all the bumps and hollows using only solar and CO2.

        • Gerald Machnee
          Posted Sep 28, 2008 at 8:09 PM | Permalink

          Re: Tony Brown (#168),
          **The trouble is, what is considered the start date for temperatures against which the 3 degrees rise will occur?? The preindustrial age (at 280ppm) covers most of mans history yet there are various times when even Al Gore says the temperatures were warmer than now. So does the co2/temp cslculation start in the 1730’s-same temp as now?
          The 1100’s much warmer than now? The 6th century -much warmer than now?

          What is considered the start point against which we should plot- and if we are starting from a low temp rather than a high one-why?.**
          For an interesting presentation on the CO2 effect, you can read David Archibald’s presentation at:

          http://www.warwickhughes.com/agri/Solar_Arch_NY_Mar2_08.pdf

          Steve had a post in January/08 on the calculation of the doubling effect.

        • Posted Sep 29, 2008 at 3:20 AM | Permalink

          Re: Gerald Machnee (#183),

          Thanks for the link to the interesting article. I must have read 500 articles over the last six months and basically I am more bemused than ever over this debate.

          There is no doubt whatsoever that temperatures have been warmer in the past, so humanity can not be to blame for those episodes. Personally I agree that there is a compelling argument that there is a very close (but not perfect) correlation with sun activity.

          However Usokov, Solanki et al believe they demonstrated this natural correlation broke down in the 1970’s, others say that we diverged from ‘natural’ temperatures several decades earlier.

          http://www.mps.mpg.de/dokumente/publikationen/solanki/c153.pdf

          The theoretical rise of 3 degrees from doubling Co2 is remarkably close to Tyndall and Shadurov’s estimates that a rise of 1% of water vapour could raise the global average temperature of earths surface more than 4 degrees Celsius. As water vapour is the overwhelming GHC-and as it absorbs more energy than Co2- it seems reasonable this is the culprit, not a minor trace gas where a doubling of concentration would seem unlikely to cause a rise of 3 degrees-unless scientits have now managed to create the missing formula-in which case Steve would have shut this site down!

          The surface area of all water sources (many in hot countries) has increased by something getting on for one quarter of one percent over the last 100 years and irrigation has also allowed more water to enter the lower atmosphere where a proportion will evaporate. Any natural increase in temperatures-which have happened throughout our history- would likely result in enhanced evaporation thereby raising temperatures further by a measurable amount. Steve doesn’t like us discussing pet theories so I am saying this is surely factual?

          This obsession with Co2 seems strange especially as enhanced levels are known to increase plant growth at a time it is needed to feed the growing population. I am sure this is highly un scientific but surely Al Gore fingered the wrong culprit back in 1992 in his rather good book ‘Earth in the balance’ but following his growing fame and fortune from the co2 industry has been unable to revisit his earlier thoughts. In the same way Dr Mann would surely have checked his figures and sources better if he had realised the furore the HS would cause. His HS directly contradicted entire Chapters of Gores 1992 book regarding past temperatures-did the two ever meet or was there ever a fuss about two icons of the CO2 science contradicting each other?

          Tony Brown

        • DeWitt Payne
          Posted Sep 29, 2008 at 9:04 AM | Permalink

          Re: Tony Brown (#185),

          I must have read 500 articles over the last six months and basically I am more bemused than ever over this debate.

          Then you need to read more. I suggest that concentrating on atmospheric radiation transfer physics would be a big help. Some Physical Meteorology wouldn’t hurt either. There are several textbooks available at on line book sellers. The primary literature can be difficult because the authors assume that their readers know the subject in detail. Textbook authors don’t. Some of the articles at RC can actually provide some insight. The signal to noise ratio in the RC article comments is generally too low to be worth reading, though. Then you can play with the MODTRAN program and look at the generated spectra and flux densities and see how they change with temperature and CO2 concentration.

          This is not to say that land use/land cover changes can’t have a significant effect on climate.

        • An Inquirer
          Posted Sep 29, 2008 at 9:34 AM | Permalink

          Re: Tony Brown (#185),
          Tony, I believe that you raise a valid question about water vapor which can increase without being induced by a CO2-induced feedback loop. Speaking against the signficance of such a phenomenon is the vast area of the globe covered by oceans and lakes. Manmade lakes and ponds would be a small, probably insignificant, addition to the sheer size of natural water surface area. Nevertheless, there have been ancedotal cases where measured local temperatures are linked to increased local water pracdtices. With increasing irrigation, watering of lawns & golf courses, and use of water features, it is possible for local surface temperatures to increase. Yet, one likely would not expect these increases to show up in satellite temperature trends.

        • Shawn Whelan
          Posted Sep 29, 2008 at 3:03 PM | Permalink

          Re: Tony Brown (#185),
          The catalytic convertor on the autos converts poisonous CO into CO2. I wonder if there is anny connection to CO2 becoming the bogeyman and the widesread use of the catalytic convertor?

        • Phil.
          Posted Sep 29, 2008 at 3:45 PM | Permalink

          Re: Shawn Whelan (#192),

          Not really, CO has a lifetime of ~2months in the atmosphere so the convertors are just hastening the inevitable.

        • Daryl M
          Posted Sep 29, 2008 at 4:20 PM | Permalink

          Re: Shawn Whelan (#192),
          In pre-catalytic convertor days, the amount of CO in typical exhaust was relatively small, on the order of 1-2%. Turning that into CO2 makes an imperceptible difference.

  88. Shawn Whelan
    Posted Sep 26, 2008 at 10:52 AM | Permalink

    Tony,
    Apparently you think the role of science is to collect the information and then disclose it without massaging the data to fit the previously decided results. You are sadly mistaken.

  89. Flanagan
    Posted Sep 26, 2008 at 11:46 AM | Permalink

    Shawn: did I say the opposite? I do agree climate is a varying thing but I was only wondering if somebody here knew what made the antarctic extent so low this year? A special wind pattern, or something like that?

    Tony : you can go to woodfortrees – though I think they only have CO2 starting in the 60s. They do however have sunspots activity from very early dates.

  90. hswiseman
    Posted Sep 26, 2008 at 12:06 PM | Permalink

    This is a photoshoped image ovelaying 9/25/2008 over 9/25/2007 Red Ice is surviving BABYICE. Blueish Ice is at least 2 summers old now, as we approach the minimum.

    • Phil.
      Posted Sep 26, 2008 at 12:46 PM | Permalink

      Re: hswiseman (#161),

      This is a photoshoped image ovelaying 9/25/2008 over 9/25/2007 Red Ice is surviving BABYICE. Blueish Ice is at least 2 summers old now, as we approach the minimum.

      A couple of points about your image:
      first you assume stationary ice and that the ice from 9/07 is still where it was which we know isn’t so, multiyear ice moved during the winter,
      second the bright blue ice on your image isn’t there any more it’s water, red ice isn’t necessarily new it could be older ice that drifted (i.e. in Beaufort).

  91. hswiseman
    Posted Sep 26, 2008 at 3:26 PM | Permalink

    Hi Phil

    I didn’t post for disputation, just thought it was an interesting image worth discussing. I did misstate as you point out. Blue ice with no red overlay is currently gone. Blue ice with Red overlay could be several years old (either original or drifted) or Babyice if last year’s ice drifted away/melted from its position exactly one year ago and was filled in with pure Babyice. Red Ice could be drifted old ice from the Beaufort as you say and not Babyice, drifted/drained towards the Greenland Sea in a typical summer high pressure regime. I am sure there is an animation around somewhere if someone were inclined to look for it.

  92. AndyW
    Posted Sep 28, 2008 at 10:43 AM | Permalink

    Just from eyeballing it looks like the Canadian side is freezing up quicker than the Russian side, perhaps because of all the dismantled subs and their nuclear powerplants lying off Murmansk :D

    Seriously though, is it as I see it?

    Regards

    Andy

  93. Shawn Whelan
    Posted Sep 28, 2008 at 11:00 AM | Permalink

    The NW Passage Parry Channel route is frozen solid.

  94. BarryW
    Posted Sep 28, 2008 at 12:09 PM | Permalink

    2008 extent gain is proceeding at about the average pace and extent is back over 5.0 mkm2,

  95. TheDude
    Posted Sep 28, 2008 at 3:18 PM | Permalink

    Moar acronyms plz:P

  96. tty
    Posted Sep 29, 2008 at 9:25 AM | Permalink

    Re 185

    “The surface area of all water sources (many in hot countries) has increased by something getting on for one quarter of one percent over the last 100 years”

    Very unlikely. At least in Europe, North America and the Middle East the area of standing water has decreased drastically during this period due to large-scale reclamation of wetlands and better draining of cultivated land.

  97. bender
    Posted Sep 29, 2008 at 9:58 AM | Permalink

    Why did the arctic not go permafrost-free in the last 700,000 years? Did arctic clouding serve to put a lid on past warming? Low cloud has an albedo comparable to that of snow. And vertical convection is not the sort of process that the GCMs are particularly good at simulating. I think there’s a crack there in the armor. This idea is consistent with Lindzen’s skepticism about negative feedbacks resulting from cloud formation (iris effect). I wonder what Lindzen would have to say about the arctic?

    • DeWitt Payne
      Posted Sep 29, 2008 at 12:00 PM | Permalink

      Re: bender (#190),

      Melting permafrost is somewhat akin to melting the Greenland or Antarctic ice caps. It requires massive amounts of energy that can only be transmitted by conduction from the surface, which is pretty slow. Sea ice, OTOH, is subject to the influence of warm water currents from below as well as air above. I think.

      • Reference
        Posted Sep 29, 2008 at 3:55 PM | Permalink

        Re: DeWitt Payne (#191),

        And slower than we all thought possible. Yukon permafrost older than previously thought

      • Schnoerkelman
        Posted Oct 2, 2008 at 2:56 AM | Permalink

        Re: DeWitt Payne (#191), Doesn’t this imply that it’s much easier to grow the Greenland and Antarctic ice through small amounts of additional precipitation than it is to shrink them by warming?

        • DeWitt Payne
          Posted Oct 2, 2008 at 9:53 AM | Permalink

          Re: Schnoerkelman (#200),

          Sort of. But you have to remember that ice under pressure is plastic and flows. What keeps the ice caps from increasing without limit is the flow of ice into the sea.

  98. Mike Bryant
    Posted Oct 1, 2008 at 9:05 PM | Permalink

    AndyW,
    Looks like October 1 will be about 5.3, you said 5.5.
    You are the closest without going over!!!

  99. Michael Hauber
    Posted Oct 1, 2008 at 10:29 PM | Permalink

    Well sea ice does seem to be increasing. Which is boring for alarmists. But the increase doesn’t seem to be particularly spectacular either. Which is boring for anti-alarmists (can I abbreviate that as AA?)

    • Daryl M
      Posted Oct 1, 2008 at 11:37 PM | Permalink

      Re: Michael Hauber (#197), But the increase doesn’t seem to be particularly spectacular either.

      I’m curious as to why you’d say that. Looking at the graphs for AMSR-E Sea Ice Extent for 2002 – 2008, I’d say that 2008 has turned the corner relatively quickly and is heading up at a rather impressive rate.

      Of course, if a certain person who shall remain nameless, but whose initials are AW, was still posting daily graphs like before, we wouldn’t have to look at those other sites. :)

      • Posted Oct 2, 2008 at 9:47 AM | Permalink

        Re: Daryl M (#198),
        I know, I’m lazy, slovenly and an all around worthless human being for not doing this myself. Once a week or so would be absolutely wonderful.

        • DeWitt Payne
          Posted Oct 2, 2008 at 10:05 AM | Permalink

          Re: Robert (#202),

          I’m planning on posting some data and links to graphs tomorrow. Right now, the smoothed JAXA extent rate is significantly above average. Only 2002 had a higher rate at this point in the year. CT area seems to be increasing at about the normal rate from a below normal level. The CT Arctic area anomaly is remaining relatively constant rather than plunging like it did in 2007. There seems to be a discrepancy between Arctic-ROOS and JAXA. Arctic-ROOS shows a pause in extent increase while JAXA is increasing rapidly. They use different satellite images and different algorithms though.

        • Posted Oct 2, 2008 at 10:20 AM | Permalink

          Re: DeWitt Payne (#204),
          Thank you so much. Totally off-topic, but this is my current focus http://www.gsebailouts.us/. I’m spending every spare minute with a sign handing out flyers. I wish I could spend more time here.

        • Posted Oct 2, 2008 at 12:57 PM | Permalink

          Re: Robert (#206),

          Robert

          I hope your flyers have got more information on them than the website-it doesnt show up

          Tony Brown

        • Robert
          Posted Oct 2, 2008 at 7:59 PM | Permalink

          Re: Tony Brown (#207),
          Well, it has the right link :-) http://www.gsebailout.us/. Unfortunately, the firewall at work blocks my innocuous little site so I could not check it.

  100. AndyW
    Posted Oct 2, 2008 at 12:19 AM | Permalink

    The NW passage is filling up quickly and I agree with DarylM that 2008 has had a couple or three large increases and has a good turn into the home furlong. However I still think it will be an average run up to Dec/Jan.

  101. mark in austin
    Posted Oct 2, 2008 at 7:59 AM | Permalink

    seriously…i just don’t know how to do it or i would. he tried to give me a “script” for extracting data from the data site, but i don’t really know how to use a script.

  102. DeWitt Payne
    Posted Oct 2, 2008 at 10:17 AM | Permalink

    For anyone new to the thread or who hasn’t bookmarked the URL’s:

    JAXA Arctic Sea Ice Extent

    Arctic-ROOS daily charts

    Cryosphere Today Recent NH area

    Uni-Hamburg main sea ice page

    Cryosphere Today main page

  103. tty
    Posted Oct 3, 2008 at 12:52 AM | Permalink

    Strange, the values for the last three weeks (since the minimum) has disappeared from the plot.csv file at JAXA. I feel an “adjustment” coming up.

    • MartinGAtkins
      Posted Oct 3, 2008 at 7:36 PM | Permalink

      Re: tty (#210),

      “Strange, the values for the last three weeks (since the minimum) has disappeared from the plot.csv file at JAXA. I feel an “adjustment” coming up.”

      Before suggesting wrong doing please read data guidelines.

      The method for area calculation.

      Usually, sea ice extent is defined as an average of several days in order to eliminate calculation errors by data deficiency. However, we adopt the average of two days in this site for the purpose of rapid release.

      The data is up to date.

      JAXA

  104. Posted Oct 3, 2008 at 10:30 AM | Permalink

    Arctic Ice Shrinks to Record Low

    Sea ice at the top of the planet has apparently reached the lowest volume ever recorded, say scientists, with conditions declining toward a point where the Arctic Ocean may soon be completely ice-free in summertime.

    See, same headline, just a change in terminology.

    • DeWitt Payne
      Posted Oct 3, 2008 at 11:00 AM | Permalink

      Re: Ryan Maue (#211),

      Note the apparently. In fact, they don’t know because there is insufficient thickness data both current and historical. That doesn’t mean they’re wrong though.

      Anyway. Update time.

      CT area has increased to 3.487 Mm2 from the low of 3.004 Mm2. JAXA extent has increased to 5.385 Mm2 from the low of 4.708. Ice concentration (CT area/JAXA extent) is up to about 70%. The absolute numbers are well below average but higher than 2007. The current exponentially smoothed rates are near average for this time of year. I’m not going to update again until the Uni-Hamburg data for September becomes available. That could be as early as next Friday, but I’m not holding my breath. Some of the Uni-Hamburg area data in the figures below has been modified to correspond to CT data by shifting the date and adjusting the value.

      Arctic Ice Extent Day 276

      Arctic Ice Extent Rate Day 276

      CT Arctic Ice Area 10/3/2008

      CT Arctic Ice Area Rate 10/3/2008

  105. DeWitt Payne
    Posted Oct 4, 2008 at 12:19 AM | Permalink

    Uni-Hamburg updated early, on 9/29 covering through 9/23/2008. Minimum extent was 3.91953 Mm2 on 9/18 and minimum area 3.52144 Mm2 on 9/2/2008 compared to 3.53623 Mm2 minimum extent on 9/17/2007 and minimum area of 3.25860 on 9/15/2007. Regressing CT Arctic area on Uni-Hamburg area shows the best fit has a lag of between 4 and 5 days. That may also imply some sort of moving average for the CT data with an even number of days (ten would give a five day lag, but seems too high). The 5 day lag fit was slightly worse than the 4 day lag, but the 5 day lag has a slope of almost exactly one, which is esthetically pleasing, so I’m going with a five day lag for comparing CT and UH. I’ll put up some graphs soon.

  106. tty
    Posted Oct 4, 2008 at 2:42 AM | Permalink

    Re 213

    Yes, the data is now up to date again, but it wasn’t when I wrote my first post.

  107. DeWitt Payne
    Posted Oct 5, 2008 at 10:52 PM | Permalink

    Here’s something I don’t understand. I can regress the JAXA average (2002-2008) on the same average for Uni-Hamburg ice extent and get a really good fit as shown here:

    Average Arctic Ice Extent

    But when I use the coefficients from the above regression to calculate a similar plot for 2008 data I get this:

    Arctic Ice Extent 2008

    The JAXA 2008 data and the adjusted Uni-Hamburg 2008 data (red and brown lines) are in near perfect agreement until about June 10. Then the traces diverge and Uni-Hamburg shows much lower extent than JAXA. The Uni-Hamburg extent from June 10 on is much closer to 2007 than JAXA. Both groups get their data from the same satellite images. Curious. Now I’m going to have to plot each year individually.

    • DeWitt Payne
      Posted Oct 6, 2008 at 12:42 AM | Permalink

      Re: DeWitt Payne (#216),

      Problem solved. I should have checked the data better when I imported it. I found all the places where data was missing in 2002-2007 and interpolated the missing data. But I didn’t look closely enough at 2008. There were 3 days missing in June and one in July. No more major discrepancy after interpolation and the difference between 2007 and 2008 in extent and area looks more like the JAXA and Arctic ROOS plots.

      Re: AndyW (#217),

      That isn’t exactly news, see #211 above and my reply in #212.

      • Phil.
        Posted Oct 6, 2008 at 2:15 PM | Permalink

        Re: DeWitt Payne (#218),

        Problem solved. I should have checked the data better when I imported it. I found all the places where data was missing in 2002-2007 and interpolated the missing data. But I didn’t look closely enough at 2008. There were 3 days missing in June and one in July.

        So does this deal with your 5 day lag? Re: DeWitt Payne (#214),

        • DeWitt Payne
          Posted Oct 6, 2008 at 8:49 PM | Permalink

          Re: Phil. (#222),

          It has no effect on the CT five day lag because I was comparing JAXA and Uni-Hamburg above. The lag there seems to be 2 days with JAXA behind Uni-Hamburg. JAXA uses a 2 day moving average. With CT vs. UH Arctic area the F statistic for the fit goes from 1289 at lag 0 to 2745 at lag 5. Lag 4 gives a slightly higher F, but the slope is 1.00 at lag 5. The wiggle match is quite good as shown by the 2008 and 2008 UH lines in the graph below. The UH data has been offset by the calculated regression intercept of -0.49 Mm2 and shifted right 5 days.

          CT Arctic Ice Area 10/6/2008

          I’m now convinced that the date tic marks on the CT graphs are, in fact, on the first of the month. The polar images are from the date listed, but the digital data is from an earlier date, possibly due to the use of a moving average or just the need to massage the data or a combination of both. I’ve asked William Chapman by email to confirm that there is an offset.

        • See - owe to Rich
          Posted Oct 11, 2008 at 11:23 AM | Permalink

          Re: DeWitt Payne (#223), back in August when I was wondering how to prove Phil wrong over his claims that CT area data was “current”, I looked carefully at the CT graphs on a daily basis, magnifying so I could see the pixels.

          I didn’t succeed in that objective, because one really needs an indpendent comparison like you are providing with Univ. Hamburg. But I did discover a few things which may interest you.

          1. The Iphone data was consistent with the graphs, given that the vertical resolution is 20 pixels for areas 1-2, 5-6, 8-11 etc and 19 pixels for the other intervals.

          2. If, as you say, the 1st of the month is the marked vertical line, then August 1st’s data was entered on August 5th.

          3. Tracking throughout the month is made awkward by the number of pixels not matching the number of days in the month. In fact, BIZARRELY, the number of pixels per month follows the pattern

          24 28 24 25 27 25

          for each 6-month period starting with January. Thus both February and August are 28-pixel months! This is actually visible by looking at the graphs in normal resolution.

          If anyone can explain this pattern then they are a better man (or woman) than me.

          Hope this helps in your researches, and that you manage to extract an apology from Phil ;-)

          Rich.

  108. AndyW
    Posted Oct 6, 2008 at 12:07 AM | Permalink

    Update from NSIDC if you have not seen it

    http://nsidc.org/news/press/20081002_seaice_pressrelease.html

    “In fact, preliminary data indicates that 2008 probably represents the lowest volume of Arctic sea ice on record, partly because less multiyear ice is surviving now, and the remaining ice is so thin.”

    Will be interesting to hear their predicitons for winter and summer 2009.

    Regards

    Andy

  109. Posted Oct 6, 2008 at 9:17 AM | Permalink

    I’m still of the opinion that this reflects only upon the brevity of record-keeping, indeed, of their being anyone describably as human to keep anything recognisable as a record. What’s happening around now, bit hotter, bit colder, isn’t really significant. Show me accurate records for a few billion years with a discernable trend therein and I’ll start to take an interest.

    BB

  110. DeWitt Payne
    Posted Oct 6, 2008 at 9:32 AM | Permalink

    A minor milestone was reached today. JAXA extent has increased by 1 Mm2 since the low on 9/9/2008, 5.73 Mm2 up from 4.71 Mm2. The smoothed rate of increase has been above average for most of that time. So much for a repeat of the relatively flat bottom seen in 2007.

    Re: The Housekeeper (#219),

    IIRC, for most of the last 500 million years, there were no polar ice caps. The Earth itself is only a few billion years old.

  111. DeWitt Payne
    Posted Oct 6, 2008 at 10:29 AM | Permalink

    Now that I’ve firmed up my calculation of approximate CT area from Uni-Hamburg data, I’ve revised the Arctic ice concentration (CT area/JAXA extent) graph. The behavior of the ice concentration in 2008 turns out to be close to the 2002-2007 average. The apples to apples comparison of Uni-Hamburg area/extent appears to show similar behavior (with the concentration being much higher).

    Arctic Ice Concentration 10/6/2008

  112. DeWitt Payne
    Posted Oct 6, 2008 at 8:53 PM | Permalink

    Phil,

    I misunderstood your question. The missing data came from before my CT data starts at the end of July so it has no effect on the fit as the calendar dates for the Uni-Hamburg data used for the regression were still correct.

  113. Mike Bryant
    Posted Oct 7, 2008 at 8:42 AM | Permalink

    I wonder when CT will update the seasonal sea ice extent graph (NH)?

    • Phil.
      Posted Oct 7, 2008 at 3:12 PM | Permalink

      Re: Mike Bryant (#225),

      William has told lucia that he will update it shortly (lucia asked him nicely because that was the metric for the betting on her site).

  114. AndyW
    Posted Oct 7, 2008 at 10:50 AM | Permalink

    Pretty impressive recovery

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    what does it mean in the medium term up to the extent in Q1 next year though?

    Regards

    Andy

  115. kim
    Posted Oct 7, 2008 at 12:20 PM | Permalink

    If ice volume at extent minimum the year after a record extent minimum gets close to the record extent minimum then volume at minimum will be lower than the year of record extent minimum. A more useful metric might be volume of ice melted during the season, and I’ll bet that was less for this year, than for last.
    =================================

  116. AndyW
    Posted Oct 7, 2008 at 2:28 PM | Permalink

    “If ice volume at extent minimum the year after a record extent minimum gets close to the record extent minimum then volume at minimum will be lower than the year of record extent minimum”

    Say what?

    Regards

    Sarah Palin

    • kim
      Posted Oct 7, 2008 at 3:39 PM | Permalink

      Re: AndyW (#228),

      Well since much of what melted this year was first year ice, more multi-year ice melted last year than this, making it possible that more volume was lost last year even though volume at end of year was less this year. Phil. is probably right though; I don’t have volume figures for min/max for last year and this.
      =====================================================

  117. SteveSadlov
    Posted Oct 7, 2008 at 2:32 PM | Permalink

    You ain’t seen nothin’ yet…

    Check back in on this ten years and three months from now.

    Prediction: The Arctic will not be ice free. In fact, we’ll have the opposite problem.

  118. Posted Oct 7, 2008 at 4:29 PM | Permalink

    Mike Bryant–
    When I explained that the winner of the ice bet gets a plate of brownies, William understood the urgency of getting numbers up. :)

    Since the CT page is based on the July-Aug-Sept average, it’s not surprising it’s not up yet. It also appears Chapman was in the habit of updating a full years worth of data in one fell swoop. The 2007 Oct-Nov-Dec data aren’t up.

    I’m thinking of announcing a bet on the winter maximums. But I may need to based it on JAXA data. That way, I can invite William to bet.

    • DeWitt Payne
      Posted Oct 7, 2008 at 4:58 PM | Permalink

      Re: lucia (#233),

      Uni-Hamburg has both extent and area data calculated from AMSR-E data which has higher resolution than the satellite data CT uses. They update monthly. The data is on their ftp server.

  119. J solters
    Posted Oct 7, 2008 at 4:56 PM | Permalink

    The warmies lost the argument on 2007 vs 2008 ice area/extent; but typically rather than admit to facts, the argument now shifts to ice volume, which is highly conjectural, as a basis to advance the catastrophic AGW-fingerprint ball down the field. Notice that it’s always a moving target for AGW testing. Nonetheless, if 2008 and 2009 fail to exhibit true AGW fingerprints, and annual average temp’s drop or stay flat, the AGW crowd had better shop around for some fallback positions beginning in 2010. Their credibility will be zilch. In the meantime , not even the most nit-wit, green governments will pass mandatory carbon trading laws in the face of worldwide economic meltdown. Not all the news is bad!

  120. Mike Bryant
    Posted Oct 7, 2008 at 7:08 PM | Permalink

    Lucia,
    Thanks for the info.
    Mike Bryant
    PS I love your site…

  121. STAFFAN LINDSTROEM
    Posted Oct 8, 2008 at 3:15 AM | Permalink

    #229 …Steve Sadlov…10 years and 3 months from now, ice-free arctic
    2019, January 7…Que?? Non comprendo nada?? MaslowskY possibly…!?
    …Just joking Phil….It seems as only 2004 [1.240.000 km2] had an early refreeze
    season rate higher than 2008…Hmmm!! 2002 was probably numero uno with
    some 1.4 million km2 from Sept 17+20 days, whereas 2003 same period
    less than 800.000 km2!! Any studies/papers on refreezing speed, related to thickness, Phil.?? I mean, there is BABY ICE and BABY Ice and BAby ice …??!!

    • DeWitt Payne
      Posted Oct 8, 2008 at 1:48 PM | Permalink

      Re: STAFFAN LINDSTROEM (#237),

      At the moment, the JAXA smoothed extent rate is exceeded only by 2002. One or two more days like today, though, and the rate will be (sarcasm) ‘unprecedented’ (/sarcasm).

      • BarryW
        Posted Oct 8, 2008 at 2:44 PM | Permalink

        Re: DeWitt Payne (#238),

        And it seems to be rapidly closing in on the average extent. Another month could see it exceeding the extent of all of the other years for this time of the year. I hope all of the non-icebreaking ships are out of the way of the baby ice!

        • DeWitt Payne
          Posted Oct 8, 2008 at 3:22 PM | Permalink

          Re: BarryW (#239),

          That would be the 2002 to 2007 average. The 1979-2007 average is still about 2 Mm2 higher than the current level. Don’t get too excited yet, the rate tends to run into a brick wall about now. 2007 and 2005 were exceptions.

        • BarryW
          Posted Oct 8, 2008 at 3:37 PM | Permalink

          Re: DeWitt Payne (#240),

          Should have been specific on what average I meant.

          Is the brick wall due to the limited sea area in the arctic constricting growth?

        • DeWitt Payne
          Posted Oct 8, 2008 at 4:06 PM | Permalink

          Re: BarryW (#242),

          Good question. Don’t know the answer. I’m just observing a feature of the rate plots for individual years from 2002 to 2007. With more data it could well be that the other years are the exception and large increases in the rate are common. Or as Phil pointed out above, rapid increase could be due to asymmetric extent/area loss leading to rapid increase at the high latitudes that are normally already frozen over by now. That’s potentially a testable hypothesis if someone wants to look up the images from 2002 and 2004 for the current dates and compare to other years that didn’t show such rapid increases in extent/area. I’m way behind in my TV watching so I probably won’t get around to it any time soon.

        • Phil.
          Posted Oct 11, 2008 at 10:29 AM | Permalink

          Re: DeWitt Payne (#243),

          Good question. Don’t know the answer. I’m just observing a feature of the rate plots for individual years from 2002 to 2007. With more data it could well be that the other years are the exception and large increases in the rate are common. Or as Phil pointed out above, rapid increase could be due to asymmetric extent/area loss leading to rapid increase at the high latitudes that are normally already frozen over by now. That’s potentially a testable hypothesis if someone wants to look up the images from 2002 and 2004 for the current dates and compare to other years that didn’t show such rapid increases in extent/area.

          If you look at last year’s curve alongside this year’s now that there’s a bit more data it’s clear that they both grow at the same rapid rate from the same starting extent/area, the difference is only when the upturn started. Since those are the only two years that had such large areas of open water that’s not too surprising, 2006 matches quite well over the latter part too. Based on that previous data we’d expect a turnover when extent/area reaches ~9.5/8.5 Mm^2.

    • Phil.
      Posted Oct 8, 2008 at 3:36 PM | Permalink

      Re: STAFFAN LINDSTROEM (#237),
      Staffan, the two areas where the ice is regrowing is the archipelago and the central basin, the first is ~0.04 Mm^2 ahead of last year, and the second is ~0.6 Mm^2 ahead, where there’s more northerly open water than in previous years.

  122. Mike Bryant
    Posted Oct 9, 2008 at 8:52 AM | Permalink

    Wow, big day for sea ice extent.

  123. mark in austin
    Posted Oct 9, 2008 at 8:54 AM | Permalink

    seriously! 200,000 is impressive. one more day like this and we start to move through previous years on the graph:

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    it will be funny indeed if we pass through the past 5 years this winter and next year’s drop off is far less than even this year’s was, establishing a trend.

  124. mark in austin
    Posted Oct 9, 2008 at 9:23 AM | Permalink

    is there a reliable measurement of ice thickness? if so, where can i retrieve it?

    • DeWitt Payne
      Posted Oct 9, 2008 at 9:53 AM | Permalink

      Re: mark in austin (#246),

      If you find out, let us all know. There are some buoys in the ice that measure thickness, but there aren’t all that many and it’s not at all clear that there is sufficient coverage for an accurate estimate. The submarines that cruise under the ice cap report back on the thickness of the ice along their course, but coverage is still an issue.

    • Shawn Whelan
      Posted Oct 11, 2008 at 8:36 AM | Permalink

      Re: mark in austin (#246),
      The people that would have the best record of ice thickness would be those that make the Arctic ice roads. They drive huge loads over the ice as soon as the ice is thick enough. Of course that is only for one small piece of the Arctic.

      Does anyone have access to the BBC story on the huge and quick recovery of the Arctic ice?

  125. DeWitt Payne
    Posted Oct 9, 2008 at 9:57 AM | Permalink

    This site has buoy data.

  126. tty
    Posted Oct 9, 2008 at 11:36 AM | Permalink

    I think the increase of the last few days is somewhat deceptive. Both off East Greenland and between Wrangel’s land and the New Siberian Islands the ice front has been advancing a couple of hundred kilometers per day which seems unlikely. Probably there was a lot of scattered ice in these areas which has just become dense enough to register. If one looks at the animation it rather looks like icefields that disappeared in late August are popping up now, in the same areas and the same shapes.
    So the increase is probably less impressive than it looks (and if so the melt in September was also less impressive than it looked).

    • DeWitt Payne
      Posted Oct 9, 2008 at 12:01 PM | Permalink

      Re: tty (#249),

      Are those figures from CT or AMSR-E (JAXA) images? CT doesn’t show ice if the concentration is below 30% (I think) and JAXA doesn’t show ice if the concentration is below 15%. Year over year comparisons are still valid, though, because the same rules are used for every year.

    • DeWitt Payne
      Posted Oct 9, 2008 at 1:41 PM | Permalink

      Re: tty (#249),

      I take it back. On further consideration, having a lower cutoff for ice will, I think, introduce a bias in any trend. A decreasing trend will be larger and an increasing trend will be smaller, I think. Or maybe they will both be larger. The higher the cutoff, the higher the bias will be. Somebody who has better math or programming skills than I do should be able to prove or disprove this.

      • BarryW
        Posted Oct 9, 2008 at 3:34 PM | Permalink

        Re: DeWitt Payne (#253),

        My quick guess is that an large increasing trend could occur very quickly. Consider a large area where the conditions are about the same and the ice area is 14%. The whole area would be considered as not in the ice extent. All it would take is a relatively small shift in temperature to freeze just 1% triggering the 15% threshold and adding that area to the extent. I’ve seen rivers ice over with skim ice overnight so I think that’s quite possible. I would think this would require calm conditions though. Anyone know what the weather conditions are?

        I would think that the inverse would not be as likely, but don’t know enough (anything) about melt rates. Winds and currents seem to have a large affect from what I’ve read on these threads, so they might cause a similar effect in extent reduction.

  127. Mike Bryant
    Posted Oct 9, 2008 at 11:50 AM | Permalink

    TTY,
    You mean someone hid the ice? I wonder if they put it the same place they put the heat that’s missing.

  128. Posted Oct 9, 2008 at 1:24 PM | Permalink

    This recovery is so impressive, I’ve proposed a brownie bet for November. With luck, William Chapman will tell us the winning number for the JAS avereage and we’ll know the winner for November at the same time.

    Brownie Bet Post

  129. mark in austin
    Posted Oct 9, 2008 at 9:35 PM | Permalink

    as long as they follow the same rules every year this is still very impressive!

  130. AndyW
    Posted Oct 10, 2008 at 12:20 AM | Permalink

    Why a 7 day average and why 1-7 Nov for this competition?

    Regards
    Andy

    • Posted Oct 13, 2008 at 4:06 PM | Permalink

      Re: AndyW (#256),
      Why not a seven day average? It’s just for brownies. There is no significance to the outcome. Just brownies.

      As “fun” bets go, there are reasons to avoid a 1 day reading. Every now and then, something goes wrong and JAXA does a -9999. I’m calculating the average dropping out the -9999. So, unless JAXA posts 7 -9999s we’ll get a reading. The other reason for the average of a week is so we can tally the average and watch it as the week rolls by. If we have enough betters, it’s likely that many bets will be stacked near the winning value. It will be a bit more like a horserace.

      The purpose Nov1 is to have a quick bet. It’s not pseudo-scientfic because it’s not scientific at all!

      If I wanted a remotely scientific bet, I’d do the Jan-Feb-March average or the winter minimum. I may do that later but running this bet now gets me partice figuring out how to deal with rule making. Also, I’ll be wanting to inplement a script when there is a more serious bet.

      • AndyW
        Posted Oct 13, 2008 at 11:59 PM | Permalink

        Re: lucia (#305),

        Ok, fair enough Lucia, the spread does make it more interesting over that week rather than a day, bowever if you do get -9999’s present the average is for 3-4 of them at one go so an 7 day average might be skewed and you will have to wear your umpires tin hat I fear :D

        I’ll have a guess on your (nice) web abode. I see someone has bet 11.75MM km2 .. snowball earth alert! I think I might see ice advancing down the English Channel come March if that is true!

        Regards

        Andy

  131. Flanagan
    Posted Oct 10, 2008 at 1:24 AM | Permalink

    As was mentioned earlier, the recovery is rapid but I would say it’s due to the presence of a lot of nearly-15% ice at the end of the melt season. This is why I typically prefer ice area, which also increases but at a relatively normal rate. In any case, we have more ice surface up there than last year, but what would be really intersting to have is the actual ice volume. So, again, if anyone knows a reliable source…

    • Daryl M
      Posted Oct 10, 2008 at 2:43 PM | Permalink

      Re: Flanagan (#257), the recovery is rapid but …

      As has been already stated, as long as the rules are the same for this year as they were for previous years, then your point applies equally to previous years, which makes this year’s recovery impressive relative to other (recent) years. By the time the JAXA graph is updated for today, 2008 will almost certainly have surpassed 2005 and could even be very close to surpassing 2006.

  132. mark in austin
    Posted Oct 10, 2008 at 8:55 AM | Permalink

    at any rate, it is making a VERY quick recovery. they just adjusted UPWARDS another 50,000 square kilometers.

  133. DeWitt Payne
    Posted Oct 10, 2008 at 9:28 AM | Permalink

    JAXA extent
    10/10/2005 6436406 km2
    10/10/2007 5248906
    10/9/2008 6352344
    2002-2007 average 10/10 6658932

    Re: AndyW (#256),

    Why not? The previous contest was a 3 month average (July, August, September) CT ice extent. Unfortunately, that data hasn’t been updated since 2007 so nobody knows who won yet. Not to mention that I think the Summer 2007 extent in the table is incorrect. It drops way more than JAXA and Uni-Hamburg extent for the same period.

    • Posted Oct 10, 2008 at 12:24 PM | Permalink

      Re: DeWitt Payne (#259),

      Have I missed your final results service? What was the % difference between 2007 and 2008 ice. I’m hoping to win the $50000 that I think was being offered…I just hope the prixe money was lodged in a secure bank.

      Tony Brown

      • Posted Oct 10, 2008 at 12:34 PM | Permalink

        Re: Tony Brown (#261),

        Sorry, ‘prixe’ should of course be ‘prize’ but it looks very much like ‘pixie money’ which is closer to the truth than we all like to believe I suspect.

        Tony Brown

      • DeWitt Payne
        Posted Oct 10, 2008 at 1:43 PM | Permalink

        Re: Tony Brown (#261),

        I didn’t calculate percent differences. I reported JAXA on 9/23 here. JAXA 2008 extent was 10.7% higher than 2007. Uni-Hamburg results were reported here. 2008 UH extent was 10.8% higher than 2007 and UH area was 8.1% higher than 2007. CT results are here. 2008 CT area was 2.9% higher than 2007. The big difference between 2008 and 2007 is that 2008 lost area and extent later than 2007 and is gaining area and extent earlier and faster (at the moment) than 2007. 2007 has a big jump in rate coming up in a few weeks. Combine that with a higher Spring 2008 maximum and the annual average area looks to be much higher in 2008 than 2007, possibly higher than 2005 and 2006.

        • Posted Oct 10, 2008 at 5:13 PM | Permalink

          Re: DeWitt Payne (#263),

          I forecast 16%. Do you need my bank details for the prize? I’m betting 13.6km2 for the extent this year.

          Tony Brown

    • AndyW
      Posted Oct 10, 2008 at 4:17 PM | Permalink

      Re: DeWitt Payne (#259),

      Why not?

      Because I don’t understand the rational behind it!

      When you are deciding a competition on increaae in ice extent, just for the jollies, why select a 7 day spread rather than a pure point in time, and why the start of Nov when typically things are starting to level off? Just seems to be on a whim to me.

      Better to say who will be the most accurate on Christmas day according to corrected http://www.ijis.iarc.uaf.edu values, at least that puts a purely fun prespective on it than some sort of pseudo scientific 7 day average.

      Regards

      Andy

  134. DeWitt Payne
    Posted Oct 10, 2008 at 10:15 AM | Permalink

    Since apparently I can’t post a link to a commercial image hosting site on The Blackboard, I’ll post it here.

    Summer Average Arctic Ice Extent CT, JAXA and Uni-Hamburg
    .

    Every year except CT for 2007 seems to have a constant offset.

  135. Chris
    Posted Oct 10, 2008 at 2:51 PM | Permalink

    The ice area in the Arctic ocean is amazingly close to that of [post- Pinatubo-cool-years] 1995

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=10&fd=10&fy=1995&sm=10&sd=10&sy=2008

    If summer 2009 were to turn out like summer 1996

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=09&fd=15&fy=1996&sm=09&sd=15&sy=1979

    then it would be interesting.

    [By the way I don't have an opinion on soot - that's another Chris]

  136. Chris
    Posted Oct 10, 2008 at 3:16 PM | Permalink

    #264 Daryl M

    Extent has already surpassed 2006 and area is damn close, on the Nansen graphs:

    http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic

    Negative anomalies are still quite large, but the ice coverage is already ~ halfway back to the 1979-2007 average (as compared with a year ago)

  137. DeWitt Payne
    Posted Oct 10, 2008 at 9:22 PM | Permalink

    I’ve made some overlay plots to compare some sources of Arctic ice area data. They’re my first attempt so they’re not too pretty. I should have deleted the axis labels in the overlay. I’ll do better next time.

    Arctic ROOS and Uni-Hamburg

    Arctic ROOS and CT (2008)

    Uni-Hamburg seems to track Arctic ROOS quite well except near the minimum. That is likely due to the lower resolution of the Arctic ROOS data causing small areas of low ice to be included in the count that are not included at higher resolution. CT, OTOH, is just lower. I had to shift the CT data seven days to the left for the wiggles to more or less match.

  138. DeWitt Payne
    Posted Oct 10, 2008 at 9:25 PM | Permalink

    In the first graph, the fat lines are Uni-Hamburg and the legend is in the upper right. The Arctic ROOS data are the thin lines with the legend in the lower left. In the second graph, the fat line is the CT data for 2008 that I have collected so far.

  139. Mike Bryant
    Posted Oct 11, 2008 at 7:47 AM | Permalink

    DeWitt, Do you think that the CT algorithms have been changed recently? or is it something else?

    • DeWitt Payne
      Posted Oct 11, 2008 at 9:32 AM | Permalink

      Re: Mike Bryant (#272),

      I don’t think CT has changed. They have an ice extent table going back to 1870 and the only number I question is Summer 2007. Their ice area number isn’t all that much lower than Uni-Hamburg. But that in itself is surprising. Lower resolution should result in a higher minimum. But maybe they have something in their algorithm that corrects for this. That doesn’t explain their high extent figure, though, unless they have adjusted the data so it is consistent with the pre-satellite data before 1979.

  140. Jared
    Posted Oct 11, 2008 at 12:30 PM | Permalink

    Yes, Phil, I agree that is the primary difference and reason we are seeing such rapid recovery this year…but don’t you think the fact that 2008 started its rapid growth way before 2007 did somewhat significant? A number of us here were predicting just that, because the Arctic overall is colder this year compared to last…both air temperatures and sea temperatures.

  141. crosspatch
    Posted Oct 11, 2008 at 4:08 PM | Permalink

    I find it interesting that when the amount of ice is below normal or the rate of decrease above normal, it doesn’t seem to be questioned. Yet when the rate of recovery increases, it must be mitigated with a lot of “yes, but” stuff. Why can’t it just be what it is? Are we to believe all the plots for the other years as presented yet provide caveats for this year’s because it is doing what some didn’t expect it to do? As long as the data were taken using the same criteria and methods for all years, then it doesn’t matter, it is what it is. And so far it looks reasonable to say at this point that 2009’s summer ice has a pretty good shot at being in the normal range. 2008 would be expected to be below normal as it had a higher percentage of yearling ice. 2009 won’t have that problem and will have more old ice than ’08 had. So it looks like we are coming out of a period anomalously low ice coverage. While I believe predicting 2009 to have greater ice than normal would be nothing more than crystal ball gazing, it seems reasonable to me to say that there are no indications that 2009 will vary significantly from “normal” given what we have seen to date.

  142. sean egan
    Posted Oct 12, 2008 at 1:28 AM | Permalink

    Crospatch
    “2009 won’t have that problem and will have more old ice than ’08 had.”
    That is the reverse of National Snow and Ice Data Center see it.

    http://nsidc.org/news/press/20081002_seaice_pressrelease.html

    Regular readers will also remember earlier in the years similar claims.

    http://nsidc.org/arcticseaicenews/2008/040708.html

    Note, they have not given a measure of volume in either graphcs, nor have they defined 1st Years and Multi year ice. Often it is just based on thickness – not age. If they could estimate volume I think they would have done so. Please no-one count pixels!

  143. crosspatch
    Posted Oct 12, 2008 at 2:12 AM | Permalink

    Sean, I can’t argue with you over different people’s crystal balls. All I can base my opinions on are the mechanics of it. The basic physical properties are not opinion. At the start of the 2008 melt, there was less multi-year ice than there was at the start of the 2007 melt. That is not opinion, that is fact. Older ice has less salt and is denser due to compression over its lifetime. Again, that is not an opinion, it is a physical reality of the ice. Denser, fresher ice is harder to melt and stands up over the summer better than saltier, less dense ice does. Again, not an opinion but a basic physical reality. So that 2008 ice did not stand up well in a relatively favorable year isn’t a surprise. In fact it is rather surprising that it stood up as well as it did and that was due mostly to favorable weather conditions.

    2009 *will* (not an opinion, but a fact) begin with more multi-year ice than 2008 had. That means that 2009’s ice will stand up better than 2008’s under a given set of conditions. Given that today’s ice coverage is equal to the ice coverage in 2006 and currently rising at a pace quicker than 2006 was makes it a reasonable to assume that ice coverage has a good chance of ending up in the average range.

    Assume a coverage area at the start of the 2009 melt season to be something close to what it was at the 2008 season. Given that 2009 ice is going to be able to withstand summer better than 2008’s ice did (because 2009 will have ice left over from 2007 and 2008 included) and adding a reasonable assumption that the anomalous wind conditions of 2007 that blew most of the old ice out of the Arctic Ocean and into the Atlantic where it melted in warming waters won’t be repeated in 2009, I believe one can feel pretty safe in concluding that the ice in 2009 will stand up better than 2008 did. So I believe that 2009 will show continued recovery from the 2007 anomaly.

    I find this comment interesting in the link you provided “The 2008 season strongly reinforces the thirty-year downward trend in Arctic ice extent.”. It fails to mention that the extent of the freeze preceding that melt was higher than recent years had been. So what we have seen in the past two years are both unusually high winter extend AND unusually low summer extent. It makes a big deal of one side but makes no mention of the other. And why does it matter anyway? There has only been ice at the poles for the past 2 million years or so. Before that it was ice free for a half a billion years. Having ice there at all is a climate anomaly over geological time scales. But most importantly, it appears to be a “made up fact” because looking here it appears that Arctic ice coverage has been very stable from 1979 until about 2000, so we are talking about a 7 year trend, not a 30 year. Just because someone issues a press release, it doesn’t mean it is true. Looking at the data, I can’t spot any 30-year trend. Sounds to me like someone writing a press release spouted off some “conventional wisdom” without checking it to see if it was true.

    I see more evidence to support a bottoming out of a 7 year trend and heading back up than of the opposite.

    • Geoff Sherrington
      Posted Oct 12, 2008 at 5:34 AM | Permalink

      Re: crosspatch (#280),

      Are we looking at a case of persistence, where the effects of an anomalous ice cover year will carry over for (you tell me how many) several more years because of the new ice/old ice melting differences. If we go back before observations were frequent, how can we model the persistence of ice of various ages and its possible disconnect with sea temps or air temps? Or are the effects too small to matter?

  144. Stephen Wilde
    Posted Oct 12, 2008 at 4:33 AM | Permalink

    I agree entirely with crosspatch.

    The supply of warm water to the Arctic from the exceptional run of El Ninos up to 2000 has now been cut off and the only way the Arctic ice can go is to increase.

    A lag of about 7 years to get warm water from the Pacific to the Arctic circle may well have been the primary driving force for the 2007 melt.

    • AndyW
      Posted Oct 12, 2008 at 7:28 AM | Permalink

      Re: Stephen Wilde (#281),

      I think Phil has put up this before

      http://jisao.washington.edu/pdo/

      to show how the PDO is actually going to increase water temps in the North Pacific.

      Although a cold PDO means cold and wet temps for USA and Canada the actual water temps between Alaska and Russia are up, so not cut off of water temps is probably actually true. Or is it? And will that matter if low PDO means cold Canada? Lots of competing factors here.

      On more physical grounds the rapid refreeze seems to be possibly taking a breather

      http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

      Regards

      Andy

      • DeWitt Payne
        Posted Oct 12, 2008 at 8:37 AM | Permalink

        Re: AndyW (#283),

        The AMO should be much more important than the PDO. When the AMO is high, the Gulf Stream can feed directly into the Arctic Ocean. The Bering Strait is too narrow to allow the passage of much warm water from the Japan current. The AMO should be in a declining phase as the NAO, which seems to predict the AMO, peaked some years ago.

        JAXA extent for 10/11/2008 is 26,406 km2 greater than 2005 on 10/12/2005. The current extent is also only 240,157 km2 less than the 2002-2007 average for this day of the year. The current extent rate seems to be peaking at about 100,000 km2/day, but that’s still way above the 2002-2007 average rate for this time of year. I expect that 2008 will fall behind 2005 again soon unless the 2008 rate picks up again.

        Arctic ROOS data predicts large increases in CT Arctic ice area for the next five days.

  145. Stephen Wilde
    Posted Oct 12, 2008 at 8:34 AM | Permalink

    Thanks Andy,

    However to my mind the main access of warm water into the Arctic Circle is around via the North Atlantic and the Gulf Stream. It takes a number of years to get there from the Pacific.

    The N. Pacific region seems more to be a ‘cut off’ region which just swirls around a bit and doesn’t make much progress into the Arctic Circle. That is why it tends to have an anomaly of an opposite sign to the PDO. It tends to reflect the residuals from the preceding PDO phase.

    Stephen.

  146. crosspatch
    Posted Oct 12, 2008 at 11:29 AM | Permalink

    Geoff Sherrington:

    Yes, I would expect there will be some “persistence” when you have a situation where a large amount of the old ice gets (literally) blown out of the Arctic ocean. In fact, had the weather pattern of 2007 repeated in 2008, the Arctic Ocean could well have had very little ice at all in 2008, but it didn’t. And this wasn’t due to any warming temperatures in the Arctic, it was due to wind patterns coupled with an unusually sunny Summer. How many years it takes to build up a good stock of old ice also depends on weather and wind. Ice drifts. Not much ice lasts more than 5-7 or so years anyway before it eventually gets circulated around to a point where it finds a path to warmer water.

    ’09 will see less multi-year ice than ’05 or ’06 had but more than ’08 had. So speaking only on what I believe is most probable based on on the physical state of things, the ’09 melt has a better chance of ending with more ice than ’08 did. If it actually does or not is going to greatly depend on weather conditions and predicting next year’s summer weather is a crystal ball exercise at best.

    • Posted Oct 12, 2008 at 4:21 PM | Permalink

      Re: crosspatch (#286),

      Whenever we say arctic ice is at unprecedentedly low levels we should also add ‘in the modern era’ that is since 1979 satellites. Researching sea ice and other historic climate change episodes I was struck by how fragile arctic ice actually appeared to be. This idea of it always being frozen solid seems to be a myth-it regularly has episodes of warming. The amount of this gets confused sometimes when say soot from a volcanic eruption might have had a warming effect-rather than increased temperatures- but a 40 or 50 year cycle as suggested by Chris in #289 seems possible.

      Tony Brown

  147. Chris
    Posted Oct 12, 2008 at 2:39 PM | Permalink

    Thanks crosspatch for your last couple of posts – well said.

  148. sean egan
    Posted Oct 12, 2008 at 3:33 PM | Permalink

    On the evidence provided by NSICD I linked to above I certainly can not infer that 2009 has started badly. Particularly as they used the same argument and data to say 2008 started badly, and there is no link to the full details of how the graphics were calculated. However they do, and so they are still briefing the press that 2008 supported thier view and by implication the “Ice free/ poor polar bear” headlines. Which natural helps the “consensus” view.

    The good news is this gives an excellent excuse to have another ice run!

  149. Chris
    Posted Oct 12, 2008 at 3:40 PM | Permalink

    Just one thing I might add: as far as I’m aware, none of the relevant models from the years before 2007 “predicted” 2007’s anomalous summer melt even at the pessimistic end of their projections. But the point has rarely been made that if the models can be wrong one way, why can’t they be wrong the other way too? In other words, given the longer term “trend”, isn’t it also possible that we could have a summer which suddenly and unexpectedly bucks the trend the other way…. and with knock-on effects opposite to those of summer 2007?

    Incidentally, that’s why I don’t tend to make too much of trends in global temperature since the turn of the century. They increasingly seem to be going below the projections of many climate models, but I’m aware that this needs to continue for longer for it to “matter” in a convincing way. Certainly, consensus AGW theorists don’t seem to think this cooling “noise” matters at all….. yet they are quick to label the recent low late-summer Arctic ice coverage as strong evidence of accelerating AGW. Yes, the Arctic has appeared to buck the models in a warming sense more than global temperatures have in a cooling sense. But the Arctic is a relatively small and volatile area in terms of climate, and summer ice extent is particularly volatile given the average summer temperature in the high Arctic, which to my knowledge is ~0C i.e. crudely speaking 1 degree higher freshwater sea ice melts a lot, 1 degree lower it doesn’t. Whereas the globe is, well, the globe. Which incidentally impacts the Arctic far far more than the other way round.
    (re: Stephen Wilde) In some ways I’d like to believe that anomalously warm water generated in the tropics years ago was the cause of recent Arctic warmth. But what about the La Nina period of ~1998-2001, what happened to all that colder water, and wouldn’t that have cooled the water below it more than the El Nino water warmed it? My opinion has been for some time that events in the Arctic in the last couple of years have been a lagged effect of the warmth of the last decade in general. As crosspatch mentioned, the average age of Arctic sea ice is rarely more than ~7 years – thus, seeing as the years 2000-2007 were undoubtedly warmer at NH high latitudes than any 7-year period in the 20th century (at least, given the data we have available) it’s hardly surprising that the period should culminate in anomalously low ice.
    Now, 2008 appears roughly equivalent to the years 1999/2000/2001 in terms of global temperatures. Therefore even 3 years of 2008 temperatures is unlikely to bring a major recovery beyond 2005/6 Arctic ice levels, especially if further warmer years then return (El Nino, next solar max, little bit more greenhouse gases etc.)
    The PDO cool phase has some way to go to become fully established and significant – c.f. quite strongly negative values betweeen 1998 and 2001

    http://www.jisao.washington.edu/pdo/PDO.latest

    and we’re not due a cool AMO phase for about 20 years based on 20th century cycles:

    http://en.wikipedia.org/wiki/Image:Amo_timeseries_1856-present.svg

    As for the sun, well, who knows…….
    So in summary, I agree that the AGW “case” in the Arctic has been overstated and I also think that there has not been enough consideration of the possibility that Arctic ice is cyclical (c.f. 1940s). But on the other hand I don’t see enough evidence (yet?) to allow me to predict with confidence anything more than a weak recovery in the next couple of years.
    Overall I would expect the unexpected, and unfortunately this could go either way.

  150. crosspatch
    Posted Oct 12, 2008 at 5:54 PM | Permalink

    Chris:

    Nobody predicted the 2007 melt because the unprecedented (in modern times) loss of ice wasn’t a result of temperature change. It was the result of an unprecedented (in modern times) wind pattern that they could not have foreseen the year before combined with an usually high number of cloudless days. The wind was the major player and simply blew the ice out past Greenland into the Atlantic where it met warmer currents and melted. And once such a large amount of old ice is lost in that way, it takes a while for it to accumulate again. It doesn’t matter how quickly the Arctic ocean re-freezes the following winter, the next summer you are still left with a smaller amount of older ice that survives summer conditions better. So it knowing that new ice melts easier would lead one to expect that it would take a few seasons to rebuild the stock of more mature ice that survives Summer conditions better.

    Of course it is just as possible for the opposite to happen. Wind conditions in any given year could be such that less ice than normal drifts out of the Arctic and could be coupled with an unusually high number of cloudy days and I don’t think anyone today could produce any reliable prediction of that happening next summer. So much depends on weather, not climate, and predicting weather gets less precise the farther into the future you try to forecast. To produce any kind of forecast with the scant history we have of the region that goes that far into the future is no more reliable than what Madame Sofia can do from her parlor down the street.

    • Posted Oct 13, 2008 at 1:40 AM | Permalink

      Re: crosspatch (#291),

      Madame Sofia? You must live in the same street as I do.

      Do you know of any proper study of weather conditions that have caused unprecedented (in modern times) arctic ice melt in specific years? Wind strength and direction, sun, cloud levels, exceptional tides? All must have a part to play to some extent or other in breaking up ice or stopping it forming. I read a report that said exceptional snow cover prevented ice freezing too deep in the first place in the winter, but acted as insulation against melt in the summer. So too much or too little snow could be added to the list.

      Tony Brown

      • Posted Oct 13, 2008 at 1:44 AM | Permalink

        Re: Tony Brown (#292),

        Partially answered my own question on the tidal aspect

        http://www.ims.uaf.edu/tide/

        I live by the sea with an exceptionally powerful tidal range and impact. If it can create and shift sand banks it could presumably do the same with ice as this report indicates.

        Tony Brown

    • Phil.
      Posted Oct 13, 2008 at 1:37 PM | Permalink

      Re: crosspatch (#291),

      Nobody predicted the 2007 melt because the unprecedented (in modern times) loss of ice wasn’t a result of temperature change. It was the result of an unprecedented (in modern times) wind pattern that they could not have foreseen the year before combined with an usually high number of cloudless days. The wind was the major player and simply blew the ice out past Greenland into the Atlantic where it met warmer currents and melted. And once such a large amount of old ice is lost in that way, it takes a while for it to accumulate again. It doesn’t matter how quickly the Arctic ocean re-freezes the following winter, the next summer you are still left with a smaller amount of older ice that survives summer conditions better. So it knowing that new ice melts easier would lead one to expect that it would take a few seasons to rebuild the stock of more mature ice that survives Summer conditions better.

      Not strictly true, the models of the ice extent showed just such a behavior, i.e. individual realisations showed a steady decline with fluctuations but frequently there would be a season with an unusually large drop. These events weren’t expected to occur as early as now but the decline has been faster than predicted so an earlier occurrence shouldn’t be a surprise.

      Holland showed that:
      “Holland discovered an abrupt transition that occurs in the CCSM runs early in the 21st century. During this interval, a rapid infusion of warm water from the North Atlantic melts large quantities of Arctic ice from below. The influx of warm water triggers a cascade of positive feedbacks, and the summer sea-ice minimum plummets within a decade from about 80% of modern-day coverage to about 20%. By the 2040s, the model average shows only fragments of September ice left along the far north coasts of Canada and Greenland. Otherwise, the Arctic is virtually ice-free each September through the year 2099.

      Each of the seven CCSM ensemble members depicts this rapid transition in the September ice cover. However, the timing and length of the abrupt transitions differ among the members, says Holland, with the onset occurring as early as 2015. Holland points to the unsettling rapidity of changes across the Arctic over the last few years as a sign of the region’s potential for quick transformation.”

      See here.

      Re: crosspatch (#299),
      Chris I think you’re making a bit of a jump in equating NH temperature behavior with the Arctic, reports from the Arctic did in fact indicate unusual weather in 07, notably abnormally clear skies during the summer.
      Also the trans-polar drift normally pushes ice out via the Fram strait, it wasn’t an ‘unusual pattern’, but stronger and more persistent than usual. It was the reason for heavy winter ice off the E Greenland coast last winter, it was notable that it stopped this summer leading to a retreat of ice there, including loss of fast ice.

      • Posted Oct 13, 2008 at 3:24 PM | Permalink

        Re: Phil. (#300),

        Not strictly true, the models of the ice extent showed just such a behavior, i.e. individual realisations showed a steady decline with fluctuations but frequently there would be a season with an unusually large drop. These events weren’t expected to occur as early as now but the decline has been faster than predicted so an earlier occurrence shouldn’t be a surprise.

        I’d like to freely translate Phil’s (or whoever) prediction.

        Tomorrow a strong cold front will pass through you area with heavy snow starting at 10 pm.
        Actually the snow started to fall at 5 pm, at rush hour, and the traffic get blocked.

        A really good prediction that one of mine, wasn’t it?

        • GeneII
          Posted Oct 13, 2008 at 3:41 PM | Permalink

          Re: Paolo M. (#302),

          A really good prediction that one of mine, wasn’t it?

          It’s those one-size-fits-all predictions, the kind Al Gore is fond of making. They’re adjustable. Just make a prediction then in pasage of time when you see what really happens you adjust you prediction so it fits what is really happening. After that you find all sorts of arguments that rationalize your adjustments.

          As long as no one looks in to what you’re saying, hey, everything’s great for you.

  151. Chris
    Posted Oct 13, 2008 at 7:28 AM | Permalink

    #291 crosspatch: “an unprecedented (in modern times) wind pattern that they could not have foreseen the year before combined with an usually high number of cloudless days.”

    I suspect you’re right about the “unprecedented (in modern times)” but can you be certain? Is it possible that there have been years with similar wind patterns/anticyclonic conditions in recent decades, but no one paid much attention because there was a thicker base of ice and lower background temperatures hence not such a pronounced impact on late summer ice extent?

    • Dave Dardinger
      Posted Oct 13, 2008 at 7:43 AM | Permalink

      Re: Chris (#294),

      Is it possible that there have been years with similar wind patterns/anticyclonic conditions in recent decades, but no one paid much attention because there was a thicker base of ice and lower background temperatures hence not such a pronounced impact on late summer ice extent?

      I think the point is that those who claimed that 2007 was definitely a harbinger of some iceless summers in the not-so-distant future were cutting corners and fear-mongering. Since it appears that an explanation of what happened is fairly clear, projecting the future, as far as arctic ice goes, should explicitly take such weather patterns, and their opposites, into consideration. Had there not been a measurable weather pattern to explain the 2008 melt then it would be more reasonable to imply an increased melt regime and therefore an increased likelihood of iceless summers.

  152. Chris
    Posted Oct 13, 2008 at 8:27 AM | Permalink

    #295 Yes, but still I’m not sure it’s a good idea to “explain away” summer 2007 in the Arctic in terms of its weather patterns. Any “unprecedented” local warming (anywhere in the world) is likely to be associated with particular weather patterns. Seeing as the ice is still substantially thinner than pre-summer-2007, then if the summer 2007 weather patterns repeat, the 2007 record minimum could be easily broken by some margin.
    So to be confident that this will not happen, I would suggest it is important to demonstrate a large degree of natural variability in the causation of the summer 2007 patterns, thus leaving a relatively small role for knock-on effects of NH warming. Otherwise, a continued relatively warm NH significantly increases the chances of another 2007.

  153. Chris
    Posted Oct 13, 2008 at 8:32 AM | Permalink

    I should add that the natural variability demonstrated should be cyclical (e.g. phase changes in Arctic Oscillation) and “on its way out”, otherwise it’s not going to help rule out further records!

  154. Stephen Wilde
    Posted Oct 13, 2008 at 11:40 AM | Permalink

    Chris,

    We had 30 years of strong El Nino so that isn’t going to be negated straight away by a couple of La Nina years.

    By now the La Nina conditions after 1998 probably are reaching the Arctic Circle and will have helped the trend up to 2007 to now (hopefully) go into reverse.

    It looks like a period of 7 or 8 years for a change in Pacific temperatures to feed into the Arctic Circle.

    Stephen.

  155. crosspatch
    Posted Oct 13, 2008 at 12:21 PM | Permalink

    Chris, there was no “unprecedented warming” in the NH in 2007 that I can find any record for. According to the MSU lower troposphere data measured by satellite (not prone to UHI or siting problems), January 2007 was the start of significant NH cooling that continued for the rest of the year. In January the anomaly was +0.762C for the Northern Hemisphere and by June it was down to +0.375C and in September was down to +0.240C. So there was fairly rapid Northern Hemisphere cooling through the entire year of 2007.

    Again, the loss of ice in 2007 was not a result of melting as much as the wind blowing much of the ice which is floating on water out into the Atlantic where it melted. It had nothing to do with any change in air temperatures in the Arctic which were steadily cooling (relative to “normal”) over that entire year.

  156. Chris
    Posted Oct 13, 2008 at 2:46 PM | Permalink

    #299 crosspatch:

    Re: MSU. 2007 NH figures for Jan-Mar equalled 1998’s “unprecedented” figure of +0.71C. And 1998 had the “excuse” of a strong El Nino (which began in 1997) skewing the figures strongly upwards at the lower latitudes of the NH (which cover the largest surface area in any event.) Whereas 2007 appears to have started out warm across the entire NH, which is borne out by the MSU figures for the highest latitudes (“NoPol” i.e. north of ~60N). Here’s the NoPol figures for 2007:
    Jan +0.63
    Feb +1.29
    Mar +0.53
    Apr +1.78
    May +0.04
    Jun +1.58
    Jul +1.35
    Aug +1.06
    Sep +0.00
    Oct +1.17
    Nov +0.82
    Dec +1.12

    http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt

    There was very little discernible cooling trend at these latitudes – although ocean currents did a good job of progressively cooling things at lower latitudes, as you suggest, see e.g.

    Incidentally I don’t see any sign here of the warm upwellings in the Arctic that Stephen’s theory ought to indicate, quite the reverse?

    #Phil: I’m not equating NH temperature behaviour with the Arctic; however, the latter inevitably is influenced very strongly by the former, winds from the tropics to the poles being the very basis of weather patterns.
    Out of interest, why do you think there were abnormally clear skies during the summer of 2007 in the Arctic? (i.e. what was the cause?)

    • Jared
      Posted Oct 14, 2008 at 2:33 PM | Permalink

      Re: Chris (#301),

      One thing is for sure: 2008 has been cooler than 2007 both in the Arctic and the NH as a whole.

      2007 Jan-Sep RSS Arctic lower troposphere anomaly: +.856
      2008 Jan-Sep RSS Arctic lower troposphere anomaly: +.602 (drop of about .25C)

      2007 Jan-Sep NH anomaly: +.629
      2008 Jan-Sep NH anomaly: +.445 (drop of about .2C)

      Of course, the drop for the globe overall has been a bit greater.

  157. Chris
    Posted Oct 13, 2008 at 3:38 PM | Permalink

    #302 Don’t encourage alarmism! Currently global catastrophe is forecast by many by ~2050 (10pm). By your reasoning, modellers and campaigners should start forecasting catastrophe by ~2025 (5pm) just to be on the safe side!! :)

  158. crosspatch
    Posted Oct 13, 2008 at 4:43 PM | Permalink

    “There was very little discernible cooling trend at these latitudes – although ocean currents did a good job of progressively cooling things at lower latitudes, as you suggest”

    I was looking at the aggregated NH data and not the data that is segregated by region. But in any case, no evidence of any warming either … which was the real point. I wasn’t so much making a case that there was cooling as much as I was disputing the notion that there was any “unprecedented in modern times” warming on 2007. In September when the minimum is generally reached, it looked like the temperature anomaly was a big fat 0. Average. In any case, NASA has said that it was wind, not temperature, that caused the great reduction of ice in ’07, so nobody needs to take my word for it.

    • Phil.
      Posted Oct 13, 2008 at 5:51 PM | Permalink

      Re: crosspatch (#306),

      I was looking at the aggregated NH data and not the data that is segregated by region. But in any case, no evidence of any warming either … which was the real point. I wasn’t so much making a case that there was cooling as much as I was disputing the notion that there was any “unprecedented in modern times” warming on 2007. In September when the minimum is generally reached, it looked like the temperature anomaly was a big fat 0. Average. In any case, NASA has said that it was wind, not temperature, that caused the great reduction of ice in ’07, so nobody needs to take my word for it.

      That’s a gross over-simplification no one has said that ‘it was wind, not temperature’.
      Here’s some actual data for the anomaly that refutes your argument.

      “As outlined in the August 14 and August 22 reports, the atmospheric pressure pattern over the Arctic has been unusual this summer. Sea-level pressure over the Arctic Ocean has tended to be fairly high, while pressure has been fairly low over northeastern Siberia. This has given rise to a pattern of winds bringing in warm air from the south over the coastal seas of eastern Siberian, fostering strong melt and tending to push ice from the coast into the central Arctic Ocean. Melt has been further enhanced by the fairly clear skies under the high-pressure area, known to climatologists as an “anticyclone.”
      This atmospheric pattern has persisted through late August and early September. Figure 4 shows temperature anomalies at the 925 millibar pressure level (roughly 2,500 feet) averaged for August 25 through September 4. Positive anomalies cover much of the Arctic Ocean, but are especially strong (up to 7 degrees Celsius, or 13 degrees Fahrenheit) along the Siberian coast. While consistent with the pattern of winds bringing warm air into the region, we are probably also seeing a second and quite different effect.”

      • Posted Oct 14, 2008 at 3:16 AM | Permalink

        Re: Phil. (#307),

        That’s a gross over-simplification no one has said that ‘it was wind, not temperature’.
        Here’s some actual data for the anomaly that refutes your argument.

        In a meteorological point of you, your refutation has no value.

        First of all, you have to see at surface anomaly since over the ocean or sea ice covered ocean you cannot simply transfer the aloft anomaly to the surface. Basic meteorology.

        Second, even if a surface temperature anomaly is present, you have to compute the energy transfer to the ice or water from air, at least the order of magnitude of that transfer.

        Third and the most important, the source of ocean and Arctic ocean heat is the Sun. If a stationary high pressure, with a lot of sunny skies, is above an ocean area, the water in that area gets warmer because of short wave radiation and not beacause of warmer air. Still basic meteorology.

        That said, I have not considered the fundamental importance of temporary and regional noises such as wind stress or heat coming from deeper water.

  159. Mike Bryant
    Posted Oct 13, 2008 at 8:27 PM | Permalink

    Probably…

  160. hswiseman
    Posted Oct 13, 2008 at 10:11 PM | Permalink

    October/November/Dec looking pretty mighty chilly in the Great White North according to Environment Canada

    http://www.weatheroffice.gc.ca/saisons/image_e.html?img=spfe1t

    and

    http://www.weatheroffice.gc.ca/saisons/image_e.html?img=s234pfe1t

  161. Geoff Sherrington
    Posted Oct 14, 2008 at 2:58 AM | Permalink

    Is polar bears eating each other an example of feedback?

  162. Stephen Wilde
    Posted Oct 14, 2008 at 12:08 PM | Permalink

    There was very little discernible cooling trend at these latitudes – although ocean currents did a good job of progressively cooling things at lower latitudes, as you suggest, see e.g.

    Incidentally I don’t see any sign here of the warm upwellings in the Arctic that Stephen’s theory ought to indicate, quite the reverse?

    Both those charts show anomalous warmth north of Iceland and especially east of Greenland in March 2007.

    How do they indicate the reverse of my theory ?

    Stephen.

  163. Chris
    Posted Oct 14, 2008 at 6:52 PM | Permalink

    #313 Stephen:
    “Both those charts show anomalous warmth north of Iceland and especially east of Greenland in March 2007.
    How do they indicate the reverse of my theory?”

    For each small warm anomaly in the far north Atlantic there was an equivalent cold anomaly. In any event, your theory as I understand it was that the Arctic melting reached a peak in 2007 because of the lagged arrival of El Nino warm waters. Thus you would need to see anomalous warmth relative to previous years, not just the average for previous decades. Firstly, this is not the case in the far north Atlantic if you look at previous maps.

    Secondly, in any event the far north Pacific was anomalously cold relative to the long term average, cancelling out relative warmth of the north Atlantic.
    Thirdly, equatorial Pacific waters have an incredibly long way to travel to reach the far north Atlantic.

    (crosspatch) GISS 64-90N data do not make pretty viewing for 2005-2007 in terms of avoiding the word “unprecedented”
    And what NASA said was essentially that reduction of perennial ice over previous winters, caused by “unusual” winds, preconditioned the Arctic for the summer 2007 record
    They didn’t address the shorter term causation of the record during the summer itself.

    (Paolo M) I don’t understand your arguments. It was anomalously warm both aloft and at the surface, huge amounts of ice melted implying plenty of energy transfer, the short wave radiation heated the water which heated the air (alongside the warm winds)

    (Jared) True. However, La Nina peaked in Jan/Feb, and typical lag re: global temperatures is up to ~6 months.
    See how much the oceans cooled (i.e. tended towards greater negative anomalies) between a year ago and the turn of the year
    and then how much warmth has returned to the NH oceans since.

    In fact NH SST anomalies have been warmer than in 2007 for the last 3 months.

    I sympathise with the sentiment behind the various arguments referred to, i.e. the AGW signal in recent Arctic trends has likely been exaggerated. But IMO it’s easy to get carried away with rather tenuous ideas. Best to keep it rigorous, or for one thing we might see Bender on the warpath again …. :)

    • Posted Oct 17, 2008 at 5:41 AM | Permalink

      Re: Chris (#315),

      (Paolo M) I don’t understand your arguments. It was anomalously warm both aloft and at the surface, huge amounts of ice melted implying plenty of energy transfer, the short wave radiation heated the water which heated the air (alongside the warm winds)

      My point is a very general one: if you want to look at the impact of warm air on ice melting (as Phil likes to do) you have to look at surface air.
      As you surely know, inversions are not so rare above cold surfaces and the lower layers can be decoupled from the upper ones. It’s not this the case.
      Instead, I don’t understand if you are suggesting that warm air melted the ice first and then the Sun heated the water.

  164. Stephen Wilde
    Posted Oct 16, 2008 at 2:42 AM | Permalink

    Chris,

    I think you make fair points but ignore the matter of timescales.

    All your points are based on timescales of just a few years whereas I’m considering periods of 30 years for each phase of PDO plus a lag of about 7 years to get water from Pacific to North Atlantic. I also anticipate modulating effect from changing solar activity during the decades involved although I accept that Leif disagrees with that.

    Of course there will be short term contra indications but over the time scales that I am considering I still believe that my interpretation of events is accurate.

    The 30 year positive PDO reached it’s peak effect in the Arctic Circle by 2007 and is now in long term decline whatever shorter timescale contra indications we might see for brief periods.

    If events in the future show me to be wrong then so be it but it take some time for the position to become clear either way.

  165. mark in austin
    Posted Oct 16, 2008 at 8:45 AM | Permalink

    in the mean time the ice is growing VERY rapidly men and women. take a look!

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    • BarryW
      Posted Oct 16, 2008 at 4:49 PM | Permalink

      Re: mark in austin (#317),

      Historically in (2003-7) the extent gain starts falling off over the next few days. Anybody know why? Running out of room?

      2008 hasn’t show a peak gain yet that exceeds some of the other years but the gains over the last couple of weeks is running above the average.

      • Filippo Turturici
        Posted Oct 17, 2008 at 3:54 AM | Permalink

        Re: BarryW (#320), Likely because the bigger increase amount is in the Arctic Sea: once that sea is full covered by ice sheet, the growth begin to slow down. Also, to tell the truth during winter time sea ice sheet has never decreased too much: the problem is a minor sea loss in peripherical areas, almost all on the curve line Jan Mayen-Svalbard-Novaja Zemlija and in the sea of Ohotsk, but there never was an ice-free winter in the Hudson bay, on Newfoundland coast or in the Bering Strait (while e.g. during the top of Middle Ages Optimum, Newfoundland coast was likely mainly ice-free during winter too).

  166. DeWitt Payne
    Posted Oct 16, 2008 at 9:25 AM | Permalink

    Uni-Hamburg seems to be spending their time updating their web site (and changing the URL) rather than posting new data. The new link is:

    Uni-Hamburg AMSR-E Ice Concentration

    I see a lot of white space at the top of the page with Firefox. YMMV.

    I’m also seeing a big difference between the area data for Arctic ROOS and CT. Both use SSM/I data, but Arctic ROOS area is higher and is increasing much faster than CT. I’ll post an updated area overlay plot after today’s CT data is posted.

  167. DeWitt Payne
    Posted Oct 16, 2008 at 10:50 AM | Permalink

    Here is the graph promised above:

    Arctic Sea Ice Area Arctic ROOS and CT 10/15/2008

    The thick lines are CT data. CT average is 1979 to 2000 and Arctic ROOS average is 1979 to 2007. The CT data was date adjusted to match wiggles in the 2008 data.

    The CT algorithm appears to attempt to correct for low resolution by reducing the area near the minimum compared to the Arctic ROOS algorithm. Compared to Uni-Hamburg area data, the CT algorithm significantly underestimated the area in 2007 (2.92 Mm2 vs. 3.26 Mm2). This may also be the reason that CT hasn’t updated their seasonal averages since Summer 2007. That point is clearly an outlier and their excuse for not having time for updating the numbers only makes sense if updating requires a significant change in the algorithm.

  168. mark in austin
    Posted Oct 17, 2008 at 8:49 AM | Permalink

    i may be way off here, but it seems like things freezing early bode well for a larger ice area and extent later. so having ice grow back as quickly as later gives more time for the peripheral portions to eventually freeze by march. it will be interesting to see what happens for sure.

    • DeWitt Payne
      Posted Oct 17, 2008 at 11:27 AM | Permalink

      Re: mark in austin (#323),

      It is possible to have quite long periods of little or no gain in extent and area. 2006 was like that. See: Arctic ROOS charts. The rate shows signs of peaking already.

      CT has (finally) updated their seasonal average extents through Summer 2008. Steve Mosher won that brownie bet. I’m now convinced that CT changed their extent algorithm at the end of 2006 in a way that makes the extent loss trend look much worse. Here’s my evidence, the year to year difference in extent for Summer average extent (July, August and September)

      year… JAXA….. CT…. UH (km2)
      2003 178104 157760 146137
      2004 -21795 -137900 27088
      2005 -543877 -383990 -544048
      2006 106515 46030 134330
      2007 -1003957 -2210700 -1072133
      2008 615274 642400 549470

      The difference in the CT extent from 2006 to 2007 is completely out of line with all the other data including the difference from 2007 to 2008, which is again comparable to the JAXA and Uni-Hamburg differences. You can see the same problem in the other CT seasonal averages as well as the annual average. There is a step change in 2007 that does not show up in JAXA or Uni-Hamburg data. For the Summer average, extent went from being about 1 Mm2 larger than JAXA in 2002 to 2006 to 0.22 Mm2 less than JAXA in 2007 and 2008.

  169. crosspatch
    Posted Oct 18, 2008 at 12:50 AM | Permalink

    Something like that must have happened, DeWitt, because otherwise there would be no way to explain why ice extent is about at 2005 levels but the anomaly shown on CT is 100,000km^2 more negative today than this day in 2005 according to their strip chart. They did something that is making the anomaly 100K more negative today for the same ice extent as 2005. I believe the anomaly should be about 100K more positive than it is currently showing on their site.

  170. crosspatch
    Posted Oct 18, 2008 at 12:53 AM | Permalink

    OOps, meant 1,000,000km^2

  171. crosspatch
    Posted Oct 18, 2008 at 1:04 AM | Permalink

    Here is the problem I am currently having understanding CT: Look at this comparison of today in 2005 to today in 2008. According to the anomaly strip chart 2005 was 1,000,000 below normal for the date and today is about 2,000,000 below normal. But looking at the comparison graphic it certainly doesn’t look like we have 1 million km^2 less ice than we had in 2005. What I see in this graph reflects more closely what I see in the CT comparison image.

    Something is completely borked with their anomaly strip chart calculation. It is off by a million square kilometers.

    • Posted Oct 18, 2008 at 3:44 AM | Permalink

      Re: crosspatch (#327),

      But looking at the comparison graphic it certainly doesn’t look like we have 1 million km^2 less ice than we had in 2005.

      This appears to be very true. Enlarge the screen about 150% and scroll down. This gives a clear straight line cut off of both dates. Overall 2008 actually appears somewhat greater.

    • DeWitt Payne
      Posted Oct 18, 2008 at 11:09 AM | Permalink

      Re: crosspatch (#327),

      The stripchart anomaly is area not extent. You won’t be able to estimate area by eyeballing the false color image. The CT images also don’t show ice concentrations less than 30% that may be contributing to higher area in 2005. In addition, the anomaly for 2005 at the same date looks to be about -1.5 compared to the current -1.88. You are comparing images on the same calendar date. 2008 is a leap year so the date in 2005 should be one larger than the date in 2008, 10/18/2005 = 10/17/2008. The area plots on CT are about four days behind the images, AFAIK.

      The change I think I identified at CT has to do with extent, not area. I have no evidence at this time that CT has changed the way they calculate area. I’ll see if I can do an overlay plot of UH area vs CT area for 2002 to present.

      • DeWitt Payne
        Posted Oct 18, 2008 at 11:31 AM | Permalink

        Re: DeWitt Payne (#331),

        An overlay area plot shows no evidence of any change in algorithm at either UH or CT. The pattern is nearly identical. I could clean it up and post it, but since there’s no significant difference, I don’t see the point.

  172. mark in austin
    Posted Oct 18, 2008 at 7:54 AM | Permalink

    any theories as to why the highest parts of the atmosphere which are reported are suddenly dropping?

    http://discover.itsc.uah.edu/amsutemps/

  173. Mike Bryant
    Posted Oct 18, 2008 at 8:12 AM | Permalink

    Has CT become an outlier? I guess I should be looking at JAXA and UH.
    We don’t even have the sea levels to trust anymore.
    Is the whole “scientific” community now dancing to those old
    “Al Gore Rhythms”?

  174. BarryW
    Posted Oct 18, 2008 at 2:25 PM | Permalink

    Anthony Watts has a blog entry on the sea ice. What caught my eye was the following quote from the National Ice Center:

    ponds of meltwater pooling on sea ice could fool certain satellite-borne instruments into interpreting ice as open water, “suggesting areas that have substantial ice cover as being sea-ice free.” The highlighted area is probably still impassible ice, including large amounts of thick old floes,

    So now it looks like the wilily old daddy ice was hiding with under the water!

    • DeWitt Payne
      Posted Oct 18, 2008 at 9:18 PM | Permalink

      Re: BarryW (#333),

      Melt pools would also freeze over quickly once the season turned leading to a rapid increase in area and extent, as seems to have happened this year but not last year.

      • Phil.
        Posted Oct 18, 2008 at 9:54 PM | Permalink

        Re: DeWitt Payne (#334),

        Certainly did happen last year, just about 12 days later. If you compared the area where there was a lot of melting according to the satellites, such as Beaufort Sea, with the ice analysis of the Canadian Ice Service there was a very good agreement, no sign of the pools.

  175. steven mosher
    Posted Oct 18, 2008 at 10:27 PM | Permalink

    324. I won fair and square. speaking of bets, hows the hurricane stuff looking

    • DeWitt Payne
      Posted Oct 19, 2008 at 1:34 PM | Permalink

      Re: steven mosher (#336),

      I won fair and square.

      Congratulations, btw, and I never said you didn’t. Are you waiting for the 10/25 deadline to place your next bet? Maybe previous winners should have to sit out the next contest.

      I’m just surprised that hardly anybody seems to have noticed or is curious about why the numbers are what they are, now that there is sufficient data for comparison.

      • Posted Oct 19, 2008 at 1:55 PM | Permalink

        Re: DeWitt Payne (#339),

        I’m just surprised that hardly anybody seems to have noticed or is curious about why the numbers are what they are, now that there is sufficient data for comparison.

        I’m sure lots of people are curious. But being curious, having a theory, and being able to support the theory are entirely different things. I think the increase in extent is impressive. What does it mean? That I do not know.

  176. crosspatch
    Posted Oct 18, 2008 at 11:26 PM | Permalink

    “The stripchart anomaly is area not extent”

    Could someone please explain how ice can extend to a place and not cover area? I am having trouble with the terms. In my experience, area is a measure of extent. An ice field that extends over 100 acres covers 100 acres in area. I am not understanding the semantics of the thing here. It is to me like saying something spans 9 feet but is only 3 feet in length.

    • BarryW
      Posted Oct 19, 2008 at 11:37 AM | Permalink

      Re: crosspatch (#337),

      My understanding is that the standard definition seems to be that a region that has at least 15% coverage is part of the extent. So a region of 100 acres that might have 15 acres of actual ice would be considered to have an extent of 100 acres but an area of 15.

  177. Jon
    Posted Oct 19, 2008 at 1:46 PM | Permalink

    BarryW:

    My understanding is that the standard definition seems to be that a region that has at least 15% coverage is part of the extent. So a region of 100 acres that might have 15 acres of actual ice would be considered to have an extent of 100 acres but an area of 15.

    Others have claimed that the area time-series use a cut-off. Were this true, 15% coverage is likely reported as ‘0’ area. On the other hand, I am certain that 70% coverage is report as 100% by extent and 70% by area.

  178. crosspatch
    Posted Oct 19, 2008 at 3:06 PM | Permalink

    “My understanding is that the standard definition seems to be that a region that has at least 15% coverage is part of the extent. So a region of 100 acres that might have 15 acres of actual ice would be considered to have an extent of 100 acres but an area of 15.”

    Ok, that sounds reasonable. But now the clue that the NYT published in their “greenspot” blog looms as even more important and that is that melt water /rain ponding on ice could be deceiving to satellite imagery where one might conclude that a large melt lake is actually open water when in fact it isn’t. But even looking at this comparison of 2008 with 2005, the amount of “100%” coverage seems to greatly exceed 2005 yet the anomaly in the strip chart shows a significant difference in the opposite direction that doesn’t seem to match what my eyeball wants my brain to believe from looking at the picture (no, I haven’t counted pixels).

    Well, 2009 is going to tell the story one way or the other. I believe we will end up with a September minimum something close to 2006’s. 2007 was an outlier year that is going to take a couple of years to recover from.

  179. tty
    Posted Oct 20, 2008 at 12:39 AM | Permalink

    It gets even curiouser if one looks at specific areas. The Laptev sea is now completely ice-covered and with coverage in the 60-90% range to judge from the maps. But http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.8.html
    shows the ice area as 250,000 km2 (about 35% coverage) and the negative anomaly increasing! Either there is something seriously wrong or the update lag is rather more than the usual 5 days at the moment

    • DeWitt Payne
      Posted Oct 20, 2008 at 8:50 AM | Permalink

      Re: tty (#343),

      I just clicked on your link and the area is 400,000 km2 with the anomaly rapidly approaching zero. Either your browser didn’t refresh or they were behind in updating the graph.

  180. Posted Oct 20, 2008 at 1:06 PM | Permalink

    Dewitt–
    Ahhh! Ok. I could ask Willam Chapman nicely. I could even try to bribe him with brownies.

    • DeWitt Payne
      Posted Oct 20, 2008 at 1:17 PM | Permalink

      Re: lucia (#346),

      Please do. You could also ask him about the apparent lag in the 365 day area graphs. I tried, but I’m not noted for my tact, to put it mildly, and maybe I didn’t phrase my question properly.

      BTW, (and completely OT) are your brownies chewy or more like cake? If I win, can I have pecans instead of walnuts?

  181. Flanagan
    Posted Oct 21, 2008 at 9:04 AM | Permalink

    There are here two possibilities: 1- there is a problem with NSIDC data or 2- there is a problem with arctic sea ice.

    Your opinion? Mine is that the problem lies with NSIDC but nevertheless announces a slowdown for the increase in ice extent.

  182. Murray
    Posted Oct 22, 2008 at 5:12 PM | Permalink

    Re 348 If the sea ice grows as fast this year as it did last year, by end Oct. it will be back to the 1979-2000 average. Could be 7+ years of decline recovered in one year. Also the snow cover this year is pretty amazing. Hmmm. Murray

  183. Mike Bryant
    Posted Oct 23, 2008 at 6:44 AM | Permalink

    Dewitt Payne, I think it would be great if you could do a guest post over on Wattsupwiththat dot com. That info you have needs to be heard.

  184. tty
    Posted Oct 23, 2008 at 9:22 AM | Permalink

    Re 344

    The problem that microwave measurements mistakes melt pools for open water is well known, and I have seen a figure of 10 % underestimate of area during the melting season quoted in the literature. CT did change their algorithm rather drastically early in 2007, though the have never explained exactly how or why. Could they have removed a correction for this error then? It would explain why the step change is largest in summer, smaller in spring and autumn and essentially zero in winter, and also why the results are now parallell to the others again, though at a lower level.

    • Nylo
      Posted Nov 6, 2008 at 6:08 AM | Permalink

      Re: tty (#351),

      The problem that microwave measurements mistakes melt pools for open water is well known, and I have seen a figure of 10 % underestimate of area during the melting season quoted in the literature. CT did change their algorithm rather drastically early in 2007, though the have never explained exactly how or why.

      (Bolds are mine)

      Hi TTY, do you have any links to the source for that information? I have been looking at CT but was unable to find any reference. Is it something that they admit having done, even if they don’t explain how or why?

      CU.

      • Posted Nov 6, 2008 at 11:43 PM | Permalink

        Re: Nylo (#420),

        I asked Cyosphere’s Bill Chapman about this via email some time back.

        Here is his reply:

        Hi Carl,

        If I remember correctly, during this period we found 3-4 cases of missing days of data scattered throughout the record. The effect of this was to offset the calculation of the anomalies so that we were subtracting the daily climatological mean from a day that was actually up to several days earlier or later than the corresponding actual date. The effect of these missing dates caused the error/offset to be more pronounced for the later period of the record (most recent years).

        We found another pair of missing days in 2007/2008 about a month ago and made a similar correction to the most recent few months.

        For what it’s worth, it looks to me like the correction you pointed out increased some anomalies as well.

        Thanks for your interest in the cryosphere. Let me know if you have any other questions.

        Sincerely,

        Bill Chapman

        On Mar 18, 2008, at 7:19 PM, Carl Smith wrote:

        Dear Sir.

        I wonder if you could please explain something to me.

        Archived images show a substantial adjustment was made to the Arctic Ice Extent data base in 2007 – see for example this pair of anomaly graphs from March and June 2007:

        What was the reason for such a large downwards adjustment being made some time between 11 March and 4 June 2007?

        Regards,
        Carl Smith.

  185. DeWitt Payne
    Posted Oct 23, 2008 at 10:48 AM | Permalink

    Today’s JAXA extent (8.088 Mm2) is greater than for the same day in 2006 and 2004 as well as 2007 and only slightly behind 2005. I’m worried that my brownie bet estimate for the 11/1 to 11/7 average (9.022 Mm2) is too low, but I’m sticking with it anyway.

    • BarryW
      Posted Oct 23, 2008 at 7:17 PM | Permalink

      Re: DeWitt Payne (#352),

      Don’t jump ship just yet. The rate of change is going to decrease significantly in the next few days if it follows previous years, and if it tracks near the other year’s rates it looks like you might be very close. 2007 closed rapidly up until about the time your bet is based on and then slowed down.

  186. Michael Hauber
    Posted Oct 23, 2008 at 10:07 PM | Permalink

    Rapid ice growth seems to be on the Arctic side of the Bering Straight. It will be interesting to see what the ice growth will be after this area is frozen solid, at which time I think the main focus of freeze will move to NE Canada area, and the freeze slow down, if my limited browing through Cryosphere Today history is a good enough guide.

    At ice minimum I was kind of guessing on a fast start to the season, and a slow finish. Based on my assessment that SSTs within the Arctic were colder than the last few years, but in the North Atlantic about the same, and in the NW pacific much warmer. However just as the freeze is approaching the Bering Straight and what I expected to be a warm water brick wall on that side the water in the NW Pacific seems to have cooled down quite a lot with some below average areas appearing where it was several degrees above normal.

    The other intersting aspect is that 2007 was a melt that was worse than the trend of the last few years. Then the 07/08 winter recovery was much stronger than the trend for the last few years. Then in my opinion the 08 melt was again worse than the trend. Now we are seeing early evidence that the freeze may be substantial. Perhaps something changed in 2007 (PDO?) to significantly increase the seasonal cycle in sea ice, making for both bigger melts and bigger freezes.

    • Posted Oct 24, 2008 at 1:39 AM | Permalink

      Re: Michael Hauber (#354),

      I posted a link in this thread some time ago from the EU marine agency that showed 2006 research that 5 out of 6 European oceans were cooling-the exception being the Barents sea. The currents seemed to suggest that would also eventually follow, which seems to be the case. Nobody at all commented on it

      As for the ‘warm water brick wall’ could the be related at all to a programme on BBC radio a few days ago and one item in particular which astonished me? This was a top person from ‘Science’ stating that there is no accepted theory as to why hot water freezes more rapidly than cold water-but it obviously did. Does ‘warm water’ also freeze faster? I don’t know, just asking the question.

      Tony Brown

      • Shawn Whelan
        Posted Oct 29, 2008 at 5:44 PM | Permalink

        Re: Tony Brown (#356),
        “Can hot water freeze faster than cold water?
        Yes — a general explanation
        Hot water can in fact freeze faster than cold water for a wide range of experimental conditions. This phenomenon is extremely counter- intuitive, and surprising even to most scientists, but it is in fact real. It has been seen and studied in numerous experiments. While this phenomenon has been known for centuries, and was described by Aristotle, Bacon, and Descartes [1-3], it was not introduced to the modern scientific community until 1969, by a Tanzanian high school student named Mpemba. Both the early scientific history of this effect, and the story of Mpemba’s rediscovery of it, are interesting in their own right — Mpemba’s story in particular provides a dramatic parable against making snap judgements about what is impossible. This is described separately below.”
        cont

        http://math.ucr.edu/home/baez/physics/General/hot_water.html

  187. Bill Jamison
    Posted Oct 23, 2008 at 11:16 PM | Permalink

    I’m getting the impression that my bet of 8,895,422 is low also. I’m thinking that it will be closer to 9.0 – 9.15 instead. But I’m sticking with my bet.

  188. AndyW
    Posted Oct 24, 2008 at 4:49 AM | Permalink

    I’m sticking with my 9.125 prediction though it may be low. Glad you pointed out on Lucia’s site that the JAXA figures get updated from the prelimonary value Dewitt, that should stop any people claiming victory slightly too early ;)

    Regards

    Andy

  189. AndyW
    Posted Oct 28, 2008 at 3:17 AM | Permalink

    The Antartic seems to be keeping it’s ice quite well after the return of Spring down there.

    Regards

    Andy

  190. Bob B
    Posted Oct 29, 2008 at 6:39 AM | Permalink

    I am cheering for the baby ice to accumulate more rapidly so I have a chance at 9.3X10^6 Km^2

    • DeWitt Payne
      Posted Oct 29, 2008 at 9:42 AM | Permalink

      Re: Bob B (#359),

      I’ve been using a projection method for the 11/1 to 11/7 JAXA average extent similar to what’s used on the Uni-Hamburg site to estimate average extent for their September average extent. The numbers are not all that different whether I include 2007 or not. They haven’t moved outside the range of 8.9 to 9.1 Mm2 since I started and seem to be converging on 9.0 Mm2. I’m pretty sure you’re out of luck with your bet. At least it’s in the realm of reality, unlike a lot of bets on the high end.

  191. stephen richards
    Posted Oct 29, 2008 at 2:15 PM | Permalink

    Tony

    If I remember my thermo physics correctly, and it is a long time ago, the rate at which warm water freezes is greater than cold because of the “pumping” effect of evaporation.

    As I recall, the ejection of H²O molecules from the surface caused by the heat of the water causes a pressure drop at the surface (hence pumping) and PV=RT donc pressure drop more rapidly, temp drops more rapidly.

    But it takes less time to freeze cold water because there is less thermdynamic energy to dissipate.

    The “cascade cooling” process was used in ancient times ( when I was studying physic) move down the temperature scale towards absolute zero. They pumped a liquid gas such as co² which reduced it surface pressure, hence temperature, which was then used to liquify another gas such as N², H, and eventually He3 and H4 to reach below 4°K. Not sure if I have remembered the sequence correctly but I think the principal is right.

  192. Mike Bryant
    Posted Oct 29, 2008 at 2:21 PM | Permalink

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=10&fd=27&fy=1997&sm=10&sd=28&sy=2008

    Is it my imagination or is greenland getting bigger? And IF it is, doesn’t that affect Arctic Sea Ice Area? It seems like since the snow was added, there is less area for the sea ice. Anyone?

    • adoucette
      Posted Oct 30, 2008 at 7:15 AM | Permalink

      Re: Mike Bryant (#362),

      Its not your imagination, but it is explained on the display:

      “Snow cover data is displayed only for most recent dates.”

      Arthur

    • adoucette
      Posted Oct 30, 2008 at 7:21 AM | Permalink

      Re: Mike Bryant (#362),

      No, the display points out that snow cover is only shown in the current image.

  193. steven mosher
    Posted Oct 30, 2008 at 2:35 AM | Permalink

    re 339 I put my bet in on the average from nov 1-7, but the methodology was suspect. I like what lucia did. I was going to recuse myself. If I win this time I’ll donate the brownies to charity. HEY, what happened to the 2008 hurricane season bet

  194. Mike Bryant
    Posted Oct 30, 2008 at 7:24 AM | Permalink

    Adoucette,
    I understand that, however snow cover should not make Greenland larger, which it obviously is.
    Also, if Greenland is made larger won’t it affect the Arctic Sea Ice Area?

    • Phil.
      Posted Oct 30, 2008 at 8:33 AM | Permalink

      Re: Mike Bryant (#367),

      Mike I think you’ll find it’s an optical illusion, all that white seems larger.
      I tested this by subtracting the one image from the other, the image is shown below, the extra white is just the snow on the land area of the first image.
      difference

  195. DeWitt Payne
    Posted Oct 30, 2008 at 9:36 AM | Permalink

    Below is a graph that shows what I’ve done to project the 11/1 to 11/7 average. The x axis is the extent for the current day, in this case 10/29. The y axis is the 11/1 to 11/7 average. So each previous year has a point on the graph. Fit a line to the data and use the equation with the 2008 extent for 10/29 (the red dot) and you have your projection, 9.041 Mm2 for 10/29 data including 2007. If I were really energetic, I would try to calculate an error bar for the estimate, but I’m not so I didn’t. The results excluding 2007 are not significantly different.

    Projected 11/1 to 11/7 Average Ice Extent

  196. tty
    Posted Oct 30, 2008 at 11:54 AM | Permalink

    Re 368

    the extra white is just the snow on the land area of the first image

    Thats only true for the West Coast. On the East coast the fjords from Scoresbysound north have really been “snowed over”, not that it is really significant, the area is probably not more than 100,000 sq km.

  197. Mike Bryant
    Posted Oct 30, 2008 at 7:21 PM | Permalink

    Hey 100,000 sq km here, 100,000 sq km there. Pretty soon you’re talking about alot of area.
    I think there are similar areas elsewhere, along with a few islands that weren’t around befor.(Not on this image though.)
    It just seems a little careless for an organization that knows that it is being closely watched.

  198. Mike Bryant
    Posted Oct 30, 2008 at 7:23 PM | Permalink

    Also, I know that it’s not an optiocal illusion. It appears the same when the images are swapped left, right.

    • Phil.
      Posted Oct 30, 2008 at 8:55 PM | Permalink

      Re: Mike Bryant (#372),

      Mike Bryant:
      October 30th, 2008 at 7:23 pm
      Also, I know that it’s not an optiocal illusion. It appears the same when the images are swapped left, right.

      That wouldn’t change anything, the white patch just looks bigger, as I showed when I took the difference of the two images they are the same size. As tty pointed out the fjords on the E coast once frozen over and covered with snow show up as snow.

  199. Michael Hauber
    Posted Oct 30, 2008 at 9:56 PM | Permalink

    Well if the fjords used to be counted as sea ice when frozen, and are now counted as snow when frozen, then this would result in a drop in reported sea ice area.

    Is there any evidence that the fjord areas which are now showing as snow used to be counted as sea ice? Or where excluded? Squinting at the CT picture I wouldn’t have a clue either way.

  200. Mike Bryant
    Posted Oct 31, 2008 at 3:25 AM | Permalink

    Oh, it only looks bigger because the folks at CT covered water with snow, so that is why it is an optical illusion. Thanks for the explanation.

    PS That is precisely the type of shape that creates an illusion depending on left/right orientation.

    http://wattsupwiththat.com/2008/10/30/chill-in-the-air-part-2-us-breaks-or-ties-115-of-cold-and-sets-63-new-snowfall-records/

  201. Mike Bryant
    Posted Oct 31, 2008 at 3:27 AM | Permalink

    Oh, it only looks bigger because the folks at CT covered water with snow, so that is why it is an optical illusion. Thanks for the explanation.

    PS That is precisely the type of shape that creates an illusion depending on left/right orientation.

    oops wrong site

    http://yeinjee.com/discovery/tag/optical-illusion/

  202. Mike Bryant
    Posted Oct 31, 2008 at 7:32 AM | Permalink

    In a real bit of irony, expanding ice shelves could make the sea ice area decrease, as temperatures plummet.

  203. Mike Bryant
    Posted Oct 31, 2008 at 7:57 AM | Permalink

    OK, the problem is not only with the fjords on the east coast of Greenland. In this comparison as you follow the coast line around the entire perimeter of the sea ice you will notice that many bays and inlets hve been “whited out”. There have also been a few ice islands added. I wonder if this was the adjustment that I’ve heard about.

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=02&fd=27&fy=1997&sm=02&sd=28&sy=2006

    • Posted Oct 31, 2008 at 2:05 PM | Permalink

      Re: Mike Bryant (#378),

      I took a thin piece of typing paper and a fine point pen. I traced the current image with all the snow, and overlayed over the older greenland image. I matched the west side. Indeed over 2/3rds of the entire east side in the older image showed sea ice over what currently shows snow. They are not the same size.

  204. tty
    Posted Oct 31, 2008 at 11:50 AM | Permalink

    Actually it’s wrong both ways. If you compare the two images in the link above and look at the russian coast from the White Sea east to Taimyr you will notice that in the older map the ice comes well inland. The Kanin Peninsula and Vaygach are more-or less obliterated. On the other hand the younger map, while less distorted is definitely a bit “snowed-over”, for example the whole Taz estuary which is about 250 km long and 40 km wide has disappeared completely.

    Also there is another error I’ve mentioned before. From the 80’s and up to about 2003 the maps definitely shows ice in areas where there isn’t any. This is easily seen as “ice” in the Baltic in summer, where there never is any. Probably the ice in the Arctic is also exaggerated. Look at this for example:

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=07&fd=01&fy=1990&sm=07&sd=01&sy=2006

    There is positively never ever any ice in the Baltic in July!

  205. Mike Bryant
    Posted Oct 31, 2008 at 2:06 PM | Permalink

    Thank you David.

  206. Mike Bryant
    Posted Oct 31, 2008 at 5:30 PM | Permalink

    You know what? It’s not just bays and inlets that have been snowed over. Every land mass has gained area, leaving less area for sea ice. We need someone with an extra large monitor to trace the area before and after snow. Then do it again with the maps switched so that there will be no monitor error.

    Talk about a “snow” job. I feel so used. icck…

  207. Soronel Haetir
    Posted Oct 31, 2008 at 5:47 PM | Permalink

    Tracing off a monitor is likely to produce other errors unless extreme care is taken. Pixels are generally vertically oriented rectangles not squares. Try the same thing from print-outs (I can’t see the images in question since I am now blind, but <shrug

  208. crosspatch
    Posted Oct 31, 2008 at 9:32 PM | Permalink

    Soronel Haetir:

    Of course there is likely to be some small amount of error but simply looking at the page it is quite obvious that the fjords have been snowed over and water that showed open or frozen is now represented as snow. The coastline has been basically smoothed. How much area in all that jagged coastline that is now gone? And it is more, as someone noted, than just Greenland but Greenland is where it is most obviously seen. So sure, there might be some error in tracing but it seems pretty clear that the error from tracing is going to be small compared to the amount of coastline that has now disappeared under snow.

    Now that in and of itself doesn’t mean that it has impacted their area calculation unless we can somehow see that things changed right when “snow” was added to the diagram. I recall seeing someone post an animated GIF at some point in the past that showed a retroactive “adjustment” downward in ice. Does that adjustment correspond with the addition of the snow? I have no idea but am certainly curious about it.

    Maybe the root of it is that there has been so much apparent deception and attempts to slip retroactive changes in quietly without any explanation that it is difficult to trust what one sees published anymore. I personally have a certain feeling that the entire community in that field of study lacks integrity and that calls their presentations into question and invites skepticism. Many researchers seem to have some kind of Tardis where they can jump back in time and modify the past. Try animating the GISS temperatures of, say, the decade of the 1970’s from data presented over the past ten years. It might be quite surprising at how much the temperatures of that decade are still changing today!

  209. Mike Bryant
    Posted Nov 1, 2008 at 8:13 AM | Permalink

    People generally believe what they can actually see with their own two eyes. If a graph represents one thing, you can actually check the data behind the graph. Of course if that data has been “adjusted” it becomes more difficult to arrive at the truth. To back up a lie even further, you could present actual photos that have been photoshopped to match the data that has previously been adjusted to prove your point.
    IF that is what is happening here, it is particularly pernicious since the lie has been amplified and subliminally inserted into our consciousness through our eyes.
    I also remember seeing the animated GIF that records the downward retroactive adjustment. Could someone please post that GIF?
    Thanks,
    Mike Bryant

    PS This reminds me of the photos the AP presented of the bombing in Lebanon. Of course those photos were very crude. The “photos” on CT are somewhat more skillfully done.

    “Are you going to believe me or your lying eyes?” In this case, your eyes don’t even realize they are lying.

  210. Mike Bryant
    Posted Nov 1, 2008 at 8:22 AM | Permalink

    It seems to me that GISS could learn a lesson here. What if they had a photograph of each thermometer in each surfacestation? A graphics artist could easily apply the GPS reference on each photo. He could also have alternate photos for all possible temperatures required. As the database is adjusted the appropriate photo could easily be attached to that bit of data. Of course there would be numerous steps to add to the veracity of the process, including the purchasing and distribution of the connected cameras.
    If done correctly this would add much credibility to the GISS database. Why believe a satellite when you can avtually see a photograph of the thermometer with a couple of clicks?

  211. STAFFAN LINDSTROEM
    Posted Nov 1, 2008 at 9:11 AM | Permalink

    #386 Mike Bryant… why not a web camera with stereo microphones, so
    if you hear a big truck in the background you understand why the
    temperature makes a jump upwards…and that jump is taken out. If
    you check Iceland Road Administration site “Vegagerdin” you can
    actually see how many cars pass by, some roads have 30 passing
    vehicles a day and with a little better resolution…It’s blowing
    like hell in the Icelandic mountains, hurricane strength in the
    gusts…30-35 m/s…

  212. Mike Bryant
    Posted Nov 1, 2008 at 9:19 AM | Permalink

    Steffen,
    That’s a great idea, you could show the temperature digitally in the corner 24/7. The reading could be adjusted by the large mainframe back at GISS to match the politically correct temperature of the moment!

  213. Mike Bryant
    Posted Nov 1, 2008 at 5:19 PM | Permalink

    Here is the GIF that shows the sea ice anomaly adjustments.

  214. Mike Bryant
    Posted Nov 1, 2008 at 5:28 PM | Permalink

    Here is the latest:

    They are still not through changing it. At the end 1996 there is a break. It appears there are two values represented for the same time.

  215. Vincent Guerrini Jr
    Posted Nov 1, 2008 at 5:58 PM | Permalink

    #389 Maybe cryosphere no longer serious or credible site/data source (a bit like GISS temp data.. Anthony Watts does not even refer to it anymore). BTW there have been previous “huge” adjusments by Cryosphere Today which invariable favour the AGW position see here

    http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html

    Also the graph data is usually 6-7 days old compared to Jaxa or others I wonder why?

  216. Mike Bryant
    Posted Nov 1, 2008 at 7:16 PM | Permalink

    From the Google Wayback Machine:

    http://web.archive.org/web/*/http://arctic.atmos.uiuc.edu/cryosphere/

    March 24, 2007 The last time the anomaly was “normal” late ’06:

    May 18, 2007 The last time the anomaly was “normal” beginning ’03

  217. John Lang
    Posted Nov 2, 2008 at 8:24 AM | Permalink

    You can find an old sea ice dataset from the NSIDC of daily sea ice extent from 1972 to 2002 here (before they adjusted the data without telling anyone what they were doing and how it was done.)

    ftp://sidads.colorado.edu/pub/DATASETS/seaice/polar-stereo/trends-climatologies/esmr-smmr-ssmi-merged/

    This dataset also used satellite data Sea ice extent has been declining slowly since about 1976 (which seems to pop up as a climate shift date in just about every climate dataset.)

    • Posted Nov 2, 2008 at 11:46 AM | Permalink

      Re: John Lang (#393),

      I often hear that claim regarding a climate shift as well. Looking at Scripps there is no notable spike in co2 and Hadley CET figures also show nothing extraordinary- other than we had two consecutive warm summers in 1975 and 1976 (happy memories!)both caused by hotter and sunnier than normal weather, but then it reverted to type.

      The most notable changes came in the decade starting the nineties. I don’t know where this idea of a climate shift came from- unless you count us in the UK getting two decent summers in a row as being one!

      Would exceptionally high levels of sun in 1975 and 76 -assuming the artic also enjoyed those conditions-had done lasting harm to the ice?

      Tony Brown

  218. Vincent Guerrini Jr
    Posted Nov 2, 2008 at 10:03 AM | Permalink

    On the basis of above numerous posting showing consistent blatant alterations/adjustments to data (always favouring AGW hypothesis), could we conclude that cryosphere today = not reliable/credible data

    • Phil.
      Posted Nov 2, 2008 at 12:24 PM | Permalink

      Re: Vincent Guerrini Jr (#394),

      You could make a similar argument concerning UAH MSU temperatures.

      • DeWitt Payne
        Posted Nov 2, 2008 at 3:20 PM | Permalink

        Re: Phil. (#397),

        But at least UAH maintains a readme file that documents when their changes were made and why they made them. No such animal exists to explain the step change between 2006 and 2007 in the seasonal extent data at CT.

  219. John Lang
    Posted Nov 2, 2008 at 12:17 PM | Permalink

    Tony Brown #394 – The mid-1970s climate shift is either a change to more El Ninos – the PDO shift (if you believe in climate cycles) or a reduction in NH aerosols (if you believe in the climate models).

    There was nearly a continuous La Nina from 1970 to 1976 (with a small El Nino in the middle) and there has been substantially more El Ninos since 1976 especially from 1986 to 2006.

    There was clearly a shift in the global climate in 1976. The Hadley Centre global temperature anomaly was -0.47C in March 1976 while it was +0.357C in August 1945. There was nearly a -0.8C decline in temperatures over the 30 years between 1945 and 1976.

    Of course, subsequent to 1976, temps have rebounded by about 0.9C. The Sept 2008 anomaly at +0.375C is virtually the same as August 1945 at +0.357C

    Hadley Centre global temp anomaly since 1850: column 1

    http://hadobs.metoffice.com/hadcrut3/diagnostics/global/nh+sh/monthly

    • Posted Nov 2, 2008 at 5:42 PM | Permalink

      Re: John Lang (#396),

      I was referring to CET figures

      1945 was 10.27 – one of out warmest years, however 1944 was 9.67 and 1946 9.45 with an average for that decade of 9.67

      The fifties had an average of 9.497
      The sixties 9.277
      the average mean temp in 1976 was 10.08 with an average mean for that decade of 9.539
      The eighties 9.518. Not a huge difference.
      In the 1990’s it shifted up a gear to 10.096

      There is no sign of a climate shift in 1976 other than a hot sunmmer which sadly reverted to type in the following year

      However even the summer of 1976 at an ave of 17.766 has to be put into perspective as that of 1826 was only fractionally behind at 17.60.

      Even the warm 1945 was beaten by 1733 1736 1779 1828 1834 and 1921

      The seventies as a decade was beaten by the 1730’s 1930’s 1940’s and 1990’s.

      The actual temperature records for the UK CET-which I was quoting- do not show this climate shift and I am always wary of global anything- let alone anomalies!

      The ‘substantially more’ El ninos from 1986 to 2006 ties in very well with rising temperatures in that period but not from 1976.

      Is there any plausible reason for more El Ninos? (other than AGW obviously).

      TonyB

  220. STAFFAN LINDSTROEM
    Posted Nov 2, 2008 at 12:36 PM | Permalink

    #397 Phil. GWS…

  221. John Lang
    Posted Nov 2, 2008 at 6:17 PM | Permalink

    Watch this 6 month animation of ocean temps (of course paying attention to the El Nino area along the equator). After it loads, hit faster several times and speed it up as fast as your computer will allow.

    You’ll see that the ENSO is a trade wind driven phenomenon. The trades (which blow east to west at the equator) occasionally speed up for sustained periods of time and blow the warm surface waters across the Pacific – and also bring up colder water from below. If sustained over a long enough time, we end up with a La Nina.

    If the trades are slowed for sustained periods of time, the surface warms up and we have an El Nino. Why there seems to be longer periods of sustained winds lasting decades or more is not clear but it could even be chance. I don’t think AGW has affected the trade winds.

    Watch the animation, in the early part, the trades have almost reversed and the La Nina of 2007 and 2008 ended by May or June. Now however, the trades have picked up again and you can see the “waves” of cold water appearing and being blown across the Pacific. This is a fascinating animation at high speed.

    La Nina is coming back.

    http://www.osdpd.noaa.gov/PSB/EPS/SST/anom_anim.html

    If you plot the Nino 3.4 region temp anomaly versus the Hadley Centre global temp anomaly, you will see a very scary correlation (with a 3 month lag.) The global temp anomaly = 0.15 times the Nino 3.4 region anomaly (with a 3 month lag). The 1997 El Nino peaked in Nov 1997 with an sea surface anomaly of +2.8C and the Hadley Centre global temp anomaly peaked (its highest ever month of course) in Feb 1998 at +0.749C. Throw in a little global warming of 0.08C per decade along with this formula for the Nino 3.4 region and you have a very good model of global temp anomalies at a monthly level – closer match than any global climate model has ever produced in any event.

  222. Mike Bryant
    Posted Nov 2, 2008 at 8:13 PM | Permalink

    The Nansen site has the Arctic sea ice area within a whisker of the historic normal.

    http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic

  223. crosspatch
    Posted Nov 2, 2008 at 10:29 PM | Permalink

    What I find interesting about those graphs is that if area is near the average but extent isn’t, then it must mean that the ice extent today has above average area. While I wouldn’t be prepared to say we have a greater 100% coverage area than average simply based on those graphs, it would be no surprise to learn that to be the case.

    In other words, if I had 10 units of extent and 5 units of area one year and 7 units of coverage but still 5 units of area the next, it would seem that the ice area that exists is more concentrated.

    • DeWitt Payne
      Posted Nov 3, 2008 at 9:45 AM | Permalink

      Re: crosspatch (#403),

      Yes the concentration of ice is currently somewhat above average, but not by all that much. What is actually measured is concentration and extent. Area is calculated from those numbers. So it isn’t possible for concentration to exceed 100%. Uni-Hamburg concentration runs about 95% when the ice is near maximum. They are the only source, AFAIK, that has digital data posted for both area and extent on a daily basis in an easily accessible format. Unfortunately they only update once/month in the middle of the month and only the data for the previous month. So they’re about two weeks behind at a minimum and more at the maximum. Currently they only have data up through 9/23 and I don’t expect them to post the rest of September and October data until next week some time.

      My real question is what’s going on at JAXA? They are stuck on 10/31 for their extent data even though they are posting the current Arctic image.

  224. John Lang
    Posted Nov 3, 2008 at 6:53 PM | Permalink

    Anyone know how I can post an excel chart online with a URL that others can access?

    • John M
      Posted Nov 3, 2008 at 7:08 PM | Permalink

      Re: John Lang (#407),

      Do you want just the chart image or the entire file?

      For just the chart image, you can try imageshack.us after creating a jpg or bmp in Paint or other image software. For the entire file, try myfreefilehosting.com. I’ve never uploaded to it, but have accessed spreadsheets others have posted on blogs. There seem to be a bunch of others. Google: free file hosting

  225. John Lang
    Posted Nov 3, 2008 at 7:34 PM | Permalink

    Okay, thanks for the info, I was trying different web hosting sites all day to no avail.

    So here it is for 1940 to 2008 on a monthly basis. The correlation coefficient is 0.765 which is either good or bad depending on your perspective but for a monthly data set that includes 725 data points, I think that is pretty good (maybe not scary until you see the chart.)

    Data is in the Data spreadsheet and is charted in the Model Chart tab.

    I don’t think you can look at the chart and not recognize how inportant the ENSO anomaly is to global temperatures. Too many of the models are smoothed over a year or over 5 years (60 months?) so that all the individual monthly information is missing.

    The ENSO directly impacts global temperatures (with a 3 month lag) almost continuously. I am surprised I have never seen this charted before and I would not interpret the climate as chaotic ever again.

    This model is not optimized. One could play with the coefficients to make it a little better. It does go off track by 0.2C to 0.3C for periods of time (although the monthly variation is still preserved) but I’ve noticed that adding the AMO index could probably improve it.

    I did this just for myself to show that not accounting for the increased El Ninos of 1986 to 2006 could mask the actual global warming trend (by simple math alone). When the ENSO is accounted for, the best global warming fit is only 0.08C per decade versus the models which predict 0.2C per decade.

    Someone else who is better with this kind of analysis should run with it and optimize it.

    http://d4.myfreefilehosting.com/d2/Basic%20Hadley%20Model.xls

    • Posted Nov 4, 2008 at 8:21 AM | Permalink

      Re: John Lang (#409),

      This doesnt seem a straightforward link-how do we access the information?

      Tony Brown

    • DeWitt Payne
      Posted Nov 4, 2008 at 1:05 PM | Permalink

      Re: John Lang (#409),

      The ENSO directly impacts global temperatures (with a 3 month lag) almost continuously. I am surprised I have never seen this charted before and I would not interpret the climate as chaotic ever again.

      But ENSO itself is chaotic, or at least nobody has ever developed a model that predicts it that is worth the paper its printed on, so in fact the climate is still chaotic. You’ve just moved the chaos back one step. If I had a clue as to how to do it, I would calculate the Hurst coefficient for ENSO.

      Steve: DeWitt, you can easily obtain a Hurst coefficient using the fracdiff program. Let’s be a bit careful in the use of the term “chaotic”, as much commentary on this seems muddied in terminology. ENSO seems to me to have 1/f properties as a time series, which is a type of very persistent red noise but not “chaotic” in the attractor sense, which is sometimes implied by this term. Hurst’s original Nile series has been argued to be a proxy for ENSO, so there’s potentially somewhat of a full circle.

      • John Lang
        Posted Nov 4, 2008 at 1:21 PM | Permalink

        Re: DeWitt Payne (#413),

        But ENSO itself is chaotic, or at least nobody has ever developed a model that predicts it that is worth the paper its printed on, so in fact the climate is still chaotic. You’ve just moved the chaos back one step.

        Yes, that’s true. I should not have put it that way. I meant the climate is impacted by more than global warming and simple white noise chaos.

        But noone can predict the ENSO cycles so at the end of the day, there is no predictive ability in this model (except for 3 months at a time) and the fact that temps are not guaranteed to increase by 3.0C per doubling of CO2.

  226. John Lang
    Posted Nov 4, 2008 at 9:05 AM | Permalink

    Try “Click Here to Download” at the bottom of this page.

    http://myfreefilehosting.com/f/ee8a33f896_0.42MB

    First time using this service so …)

    • Posted Nov 4, 2008 at 11:21 AM | Permalink

      Re: John Lang (#411),

      Thats better this time-before it just opened an ascii file (I assume) I tried to print but it showed I was just about to print over 200 pages!!It looks interesting but I cant look at it on screen very easily so would like to print out. Is there any sort of ‘executive summary?’

      Tony Brown

  227. John Lang
    Posted Nov 4, 2008 at 1:12 PM | Permalink

    It should just open as an Excel xls file. Two tabs – one for the data, the other is a chart.

    The executive summary is that the global temperature anomaly tracks the Nino 3.4 sea surface temperature anomaly (of 3 months prior) extremely closely. There is no smoothing and no accumulation required.

    It is not perfect of course. It is just a very simple model. It does go off track for periods of time. It overestimates the impact of the 1942-44 La Ninas and the 1999-2000 neutral conditions. The impact of the 1965-66 El Nino is overestimated quite a bit. Sometimes it appears the lag should be 2 months or 4 months rather than 3. But other than that, it is pretty close.

    Of course, there is an upward trend from global warming of 0.08C per decade (less than half of that predicted). Some would like to believe there is no global warming but it has always appeared to me there is definitely a GHG influence, it is just much smaller than the CGMs predict.

    Simple imperfect model – but nevertheless contains important conclusions. The biggest is that warmers are trying to take credit for the influence of the recent El Nino cycles and no one has been able to demonstate that AGW is causing more El Ninos.

    • Posted Nov 4, 2008 at 2:46 PM | Permalink

      Re: John Lang (#414),

      This is very interesting-and Enso is driven by trade winds? So presumably if we were to look up the records set by such ships as the Cutty sark and other clippers over a sustained period there would be a correlation to particularly active Enso and to Temperatures?

      tony Broown

  228. Vincent Guerrini Jr
    Posted Nov 4, 2008 at 1:32 PM | Permalink

    #397 I though the UHA RSS data changes FAVOURED AGW? Correct me if I am wrong.

  229. DeWitt Payne
    Posted Nov 4, 2008 at 1:33 PM | Permalink

    Now for something completely different: an on-topic post.

    The Arctic extent rate has fallen drastically the last few days, as expected for this time of year. There was even a day (11/2) of a very small loss of extent. Area as of 11/1 in the Nansen charts on the Arctic ROOS site was still increasing rapidly and was very close to the 1979-2007 average. The CT Arctic anomaly is now -0.769 Mm2 (from the 1979-2000 average), the Antarctic anomaly is 0.552 and the global anomaly is -0.335. The Arctic and Antarctic anomalies are for different days, the Antarctic is one day ahead of the Arctic so the global anomaly is not the sum of the currently posted Arctic and Antarctic anomalies. The Antarctic anomaly is high because the Antarctic is losing area at a slower than average rate while the Arctic is gaining area at well above the average rate for this time of year.

  230. Earle Williams
    Posted Nov 4, 2008 at 2:01 PM | Permalink

    Re #416

    Vincent,

    I doubt Phil. has any substantive complaint about UAH MSU that compares to the lack of transparency at CT. I do wonder if he feels the same snide commentary applies to RSS MSU. Given the disparaging nature of the prior comments perhaps any further discussion should itemize the specific complaints?

    Cheers,
    Earle

  231. tty
    Posted Nov 6, 2008 at 12:27 PM | Permalink

    Nylo:

    Compare these two versions from the Wayback Machine:

    http://web.archive.org/web/20070311234525/http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg (March 11 2007)

    http://web.archive.org/web/20070607031303/http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/current.anom.jpg (June 7 2007)

    I don’t think they have ever commented on the change.

    Incidentally according to IJIS the ice extent on Nov. 5 was larger than on the same date 2002-2007, so it’s “unprecedented” for at least seven years…..

  232. Posted Nov 6, 2008 at 2:44 PM | Permalink

    As of yesterday the global sea ice anomaly at cryosphere reached 0. Four weeks ago it was at a negative 2.6 million sq kilometers. This is the fasted move in the 30 year history. The sea surface temperature anomaly around the Antarctic remains strongly negative.

  233. SteveSadlov
    Posted Nov 6, 2008 at 5:26 PM | Permalink

    “Growing” continental areas, coastlines smoothed, adjustments and tweaks, surface ponds counted as open water – these are all tricks of the trade in the art of sea ice book cooking.

  234. Mike Bryant
    Posted Nov 6, 2008 at 11:48 PM | Permalink

    Very weak reply from CT… very, very, very weak.

  235. hswiseman
    Posted Nov 7, 2008 at 12:49 AM | Permalink

    There are several climate scientists who make me shake my head, roll my eyes and question motives. Chapman is not even close to being on that list. My sense is that Lucia also holds him in esteem, which is good enough for me. Bill Chapman answered the query nicely, and with a reasonable amount of detail. If you have some questions about the response, run some numbers yourself, or write Dr. Chapman. Is there anything whatsoever in his response to Carl that would indicate he wouldn’t respond politely and reasonably to a followup? No. There is not. Lastly, as a point of etiquette, I have had occasions to email climate researchers with a question or speculation. I have had some interesting responses, especially from NOAA staff. Under no circumstances would I post that correspondence here without expressly discussing it with my correspondent.I am alot more interested in exchanging ideas than playing gotcha.

    CT is a very good and useful website, and only scarcely/barely inconsistent with JAXA or Arctic-Roos. Yeah, remote sensing is not perfect. Maybe you should write Dr. Chapman about it. I bet he would answer.

  236. STAFFAN LINDSTROEM
    Posted Nov 7, 2008 at 5:42 AM | Permalink

    #425 Mike Bryant….Weak??????????? It sounds they just by chance happened to check what they post on their site… I KNOW nothing
    made by man is even close to serious…but… Just a little black
    today as that dirty, peabrained pigeon hurricane “Paloma” is destroying
    my NATL HS ACE prediction …So DIE DOVE DIE… ICE betting suggestion
    1st, 11th, 21th and last day of month… Starting today. Predictions
    at least 168 hours in advance GMT -5 HOURS…when writing this I start
    to figure the earlier the bet is done, the higher the score…
    [By reading Swedish papers you could believe Obama is already the executive
    President of the United States of America...IDTS...]

  237. Posted Nov 7, 2008 at 10:53 AM | Permalink

    Carl Smith–
    That sounds like a polite response to me. You asked a question, he answered to the best of his recollection. What you’ve learned in the exchanged is that when, during post processing they discover data were missing, they do some sort of adjustments. I don’t see how you can expect Bill Chapman to do more than that based on a casual email from someone he’s never met.

    FWIW, Bill keeps an archive, and my impression is the web site is a sideline. CT makes data available for those interested. Some data and images are available more or less in real time, others are not. If you were a researcher doing a deeper investigation, you could request more detailed sources and assemble your own data archive from the sources.

    Bill was quite nice to me. When I explained that brownies were riding on his CT data, he bumped up the priority of posting. What more could one hope for?

    Lucia

    • Phil.
      Posted Nov 7, 2008 at 3:25 PM | Permalink

      Re: lucia (#428),

      And very good brownies they were too. :)

  238. Posted Nov 7, 2008 at 3:48 PM | Permalink

    For the record, I agree that Bill Chapman was both courteous and forthcoming in his response to me, and I appreciate the fact that he is a busy man who had no obligation to even respond to me.

    He provided what he feels is an explanation for something I brought to his attention.

    Whether it is or is not an adequate explanation for the differences between the graphs is something which probably cannot be determined without someone accessing the data and running some kind of analysis on it, and opinions such as “Very weak reply from CT… very, very, very weak.” would seem to me to be premature prior to such an analysis.

    Personally, I am happy that he responded in such a straightforward and forthright manner, and appreciate his having done so – there are many others involved in climate science who would do well to take note of Bill’s example when dealing with queries from the public.

  239. Mike Bryant
    Posted Nov 7, 2008 at 6:36 PM | Permalink

    He obviously knows that many would like more transparency. Too bad he didn’t put an end to the speculation with a detailed response instead of only giving a snippet that answers nothing.
    Why won’t he lay out his case and put an end to the speculation?

  240. Vincent Guerrini Jr
    Posted Nov 7, 2008 at 6:38 PM | Permalink

    Re CT: At least BC provided an explanation and this will always enhance the standing of CT. The layout is great and very informative. I notice too, that all the images are up to date. On the other hand I have not seen any significant CT adjustments going against the “warming Principle” or any explanations for some major changes done to Antarctic ice as well

    http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html

    • Phil.
      Posted Nov 7, 2008 at 7:18 PM | Permalink

      Re: Vincent Guerrini Jr (#432),

      Look at the two graphs you show Vincent, the first of the two shows a vertical jump in April, in the second graph that jump has been filled with missing data, the last 6 months or so of data is the same just shifted later by the missing data. This is the same problem that Bill Chapman described in his reply to Carl.

      • Earle Williams
        Posted Nov 7, 2008 at 7:48 PM | Permalink

        Re: Phil. (#433),

        Phil., it’s good to see you have divorced yourself from your previous deep analysis of the discrepancy and acknowledge the distinct change in the two plots that heretofore was never explained by CT. I am less sanguine than you that the difference is easily explained by Bill Chapman’s email. Is your explanation consistent with the fact that this gap is from SH in mid-April whereas the two images that Bill Chapman responded show changes in the NH anomaly dating back to 1979? I personally am at a loss to see how inserting “missing” data from April, 2007, is going to affect past deviations from the 1978 – 2000 mean. I welcome any insight you may have into how that could work. It makes me wonder what other data is just laying around waiting to be rediscovered and inserted into the timeline.

  241. Mike Bryant
    Posted Nov 7, 2008 at 7:59 PM | Permalink

    A small note on CT with a link for a further explanation would snip speculation in the bud and enhance CTs reputation.

    What about the snow filling all the bays, fjords and inlets? Why not take the snow question out of play?

    • Phil.
      Posted Nov 7, 2008 at 10:20 PM | Permalink

      Re: Mike Bryant (#435),

      The sea/land boundary is defined by a mask, it isn’t calculated from the current image, the snow is shown as an aid for the non-scientific user, rather like the alternative images provided in the uni-bremen site (‘visual’ vs ‘nic’)

  242. Mike Bryant
    Posted Nov 7, 2008 at 10:40 PM | Permalink

    Reference?

  243. Mike Bryant
    Posted Nov 7, 2008 at 10:42 PM | Permalink

    Perhaps the non-scientific viewer would prefer to see the snow on the land masses and the sea ice in the sea. You know, maybe even scientific types might prefer that too.

  244. Vincent Guerrini Jr
    Posted Nov 8, 2008 at 3:02 AM | Permalink

    As an aside the 250mb and 400mb temps are the LOWEST ever (current) recorded since satellite data started 1998

    http://discover.itsc.uah.edu/amsutemps/

    click on the java program 400mb and 250
    Also looks like La Nina may be back

    http://weather.unisys.com/surface/sst_anom.html

  245. DeWitt Payne
    Posted Nov 8, 2008 at 9:04 AM | Permalink

    The brownie bet at The Blackboard on the 11/1 to 11/7 average JAXA extent is unofficially over. The average came in at 9.160 Mm2 and the closest to the average was Jared at 9.150 Mm2. This average is higher than for any previous year in the JAXA record (starting in 6/2002). The current extent, 9485000 km2, is also higher than any previous year in the record. Quite a come back for baby ice.

  246. AndyW
    Posted Nov 8, 2008 at 11:49 AM | Permalink

    Wasn’t it Jared who named it baby ice also ?
    :D

    Regards

    Andy

    • Jared
      Posted Nov 10, 2008 at 6:01 PM | Permalink

      Re: AndyW (#442),

      I’m the Jared who won Lucia’s brownies, and alas, no, I am not the one who came up with “baby ice”. :)

  247. Posted Nov 8, 2008 at 2:07 PM | Permalink

    Carl–
    Ok. I remember someone saying “weak analysis” for Bill Chapman. I guess I got it into my head you did. I should have scrolled up.

    Dewitt– Do note: Any previous year on record at JAXA. JAXA is fairly new. It will be interesting to watch the seasonal patterns for ice extent and area over the next few years at least.

    • DeWitt Payne
      Posted Nov 8, 2008 at 6:26 PM | Permalink

      Re: lucia (#443),

      Yes the JAXA record is short, but take a look at the NANSEN data (from SSM/I data) shown at Arctic ROOS. The current extent is just about within one SD of the 1979 to 2007 average. The area is almost at the 1979-2007 average. The annual average JAXA extent for 2008 is now very likely to exceed not only 2007 but 2005 and 2006 as well. The same pattern is observed with Uni-Hamburg data. Only the CT seasonal extent data has 2008 only slightly better than 2007 and way below 2005 and 2006.

      Arctic Sea Ice Extent Arctic ROOS

      Arctic Sea Ice Area Arctic ROOS

      • BarryW
        Posted Nov 9, 2008 at 7:49 AM | Permalink

        Re: DeWitt Payne (#450),

        2008’s extent appears to be running above all of the 2003-2007 extents, all of which maxed out below 15.0. The average max from the ROOS graph appears to be somewhat above 15. Any relationship relative to max vs min?

        • DeWitt Payne
          Posted Nov 9, 2008 at 9:21 AM | Permalink

          Re: BarryW (#453),

          Different algorithms produce different max and min extent and area. For example, I’ve posted links above to graphs showing the different average seaonal extents from CT, Uni-Hamburg and JAXA. There is usually very good correlation between the different methods, though, so that calculated linear trends have similar slopes but different intercepts. That’s how I noticed what appears to be a step change in the CT seasonal extents between 2006 and 2007. I have also been able to calculate correction equations that allow a more direct comparison between different data sets. I can convert JAXA to Uni-Hamburg extent or Uni-Hamburg area to CT area. The Uni-Hamburg to CT conversion is what I consider proof that the CT Arctic area data is several days, approximately 5, behind the other data sets.

          If you look at the longer term data, this plot for example, it looks like the trend of lower annual average ice may be the result of a relatively steady decrease in the minimum area. The maximum area, with the exception of a dip starting in 2004 that appears to have reached its minimum in 2006 and 2007, is more nearly constant.

        • BarryW
          Posted Nov 9, 2008 at 2:01 PM | Permalink

          Re: DeWitt Payne (#454),
          Thanks, one thing I noticed from your winter and summer graphs is that the trend for the winter average appears to have started increasing while the summer was still decreasing somewhat. Either the summers have warmed relative to the winters or it’s very hard to recover once your dealing with a lot of first year ice.

  248. John Finn
    Posted Nov 8, 2008 at 3:04 PM | Permalink

    Re: #440

    As an aside the 250mb and 400mb temps are the LOWEST ever (current) recorded since satellite data started 1998

    I’m not sure what this means particularly as temps at 900mb are the highest ever – by some distance.

  249. Chris
    Posted Nov 8, 2008 at 4:08 PM | Permalink

    #444

    The 900mb chart on amsutemps does not appear to be corrected for drift.

    Take e.g. Sep 2008: on the amsutemps chart it looks like at least the joint warmest ever (with 2006) for the LT. Yet RSS TLT final figures have Sep 08 at +0.194 anomaly, compared with:
    07 +0.281
    06 +0.338
    05 +0.403
    04 +0.250
    03 +0.362
    02 +0.324
    01 +0.207
    00 (+0.115)
    99 (+0.136)
    98 +0.494
    97 (+0.191)
    96 (+0.181)
    95 +0.325
    [88 +0.228]
    ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_2.txt

    Oct 2008 is a shade cooler/less warm :) at +0.181C

    So the statement in #444 “temps at 900mb are the highest ever – by some distance” is misleading. On the other hand, the statement in #440 “As an aside the 250mb and 400mb temps are the LOWEST ever (current) recorded since satellite data started 1998″ is much closer to the truth. See:

    ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tmt_anomalies_land_and_ocean_v03_2.txt

    At +0.008C, the Oct 08 anomaly for the TMT was colder than for every year since 1995, with the exception of Oct 1999, which at -0.005C was just over a hundredth of a degree cooler i.e. it is very plausible that daily 250mb and 400mb temps have recently reached some of their lowest levels since 1998.

    (Note Oct 08 LMT temps were also colder than Oct 91, 90, 88, 87, 83, 80 and 79.)

  250. Chris
    Posted Nov 8, 2008 at 4:13 PM | Permalink

    Last sentence in brackets in #445 should have read *TMT* temps

  251. Chris
    Posted Nov 8, 2008 at 4:36 PM | Permalink

    It’s also interesting to note that according to RSS TLT, Oct 08 in the Arctic/highest latitudes (60.0/82.5N) at +0.452C was cooler/less warm than Oct 1980 (+0.883C), Oct 1981 (+0.680C) and Oct 1987 (+0.620C). Even given the Arctic is now post-tipping point and 5C above “normal”. (Apparently).

    While around Antarctica (-70.0/-60.0S) Oct 08 at -0.696C appears to have been the coldest October in the satellite record (i.e. since 1979). No wonder the ice has been slow to melt there with the approach of SH summer.

    ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_2.txt

  252. Chris
    Posted Nov 8, 2008 at 5:30 PM | Permalink

    Also note from RSS the less than impressive progress of the NH in “recovering” from La Nina:

    0.0/82.5N:

    Mar-Aug 08: +0.27C
    Sep-Oct 08: +0.26C

  253. Chris
    Posted Nov 8, 2008 at 5:46 PM | Permalink

    And one final note: anyone wishing to be objective (re: recent Antarctic temps) should bear in mind that Sep 08 saw many stations on the Antarctic Peninsula trounce their records for highest mean Sep temperature.

    http://www.antarctica.ac.uk/climate/surfacetemps/

    (Sorry to those who were enjoying the unremitting “coolness” of my previous posts!)

  254. John Finn
    Posted Nov 8, 2008 at 6:43 PM | Permalink

    Chris

    So what you seem to be saying is that the 900mb temps haven’t been “adjusted” yet – and for what it’s worth I think you’re right. However the 250mb and 400mb temps (if they are raw temp readings) may be contaminated by stratosphere cooling – so they also may need adjustment.

    The point I was trying to make in my original post is that I’m not sure there are too many people outside of RSS and UAH who have sufficient background knowledge to interpret the readings correctly. But there are a number of ‘sceptics’ (UK sp) who are happy to choose whatever measurement suits their argument.

    There is currently no evidence to support the case that a global cooling trend is underway and it doesn’t help the arguments of responsible sceptics to suggest that there is.

  255. Chris
    Posted Nov 9, 2008 at 3:08 AM | Permalink

    #451

    I agree with most of what you say. With the last statement, though, I would have been more likely to agree if it had said that there is currently no *convincing* evidence to support the case that a *significant* global cooling trend is underway.

  256. Vincent Guerrini Jr
    Posted Nov 9, 2008 at 7:00 PM | Permalink

    Yes I posted before about this delay in graphed CT data. In fact we are now looking at what happened 5-6 days ago in CT NH graph. I think same for SH. Also NH image is usually 3-4 days late. Of course ther may be a very valid reason for this. If we keep criticizing the poor fellow he may decide to give up so lets give give him a break! LOL

    • DeWitt Payne
      Posted Nov 10, 2008 at 10:29 AM | Permalink

      Re: Vincent Guerrini Jr (#456),

      There may be some funny business with how the Arctic image comes up on the CT site when first accessed, but I’ve just looked at the current image dated 11/9/2008 and it appears to be nearly identical to the image at Uni-Hamburg dated 11/9-10.

      I’ll be interested to see if the next update at Arctic-ROOS confirms the significant decrease in Arctic area posted yesterday at CT.

      • Phil.
        Posted Nov 10, 2008 at 5:29 PM | Permalink

        Re: DeWitt Payne (#459),

        Re: Vincent Guerrini Jr (#456),
        There may be some funny business with how the Arctic image comes up on the CT site when first accessed, but I’ve just looked at the current image dated 11/9/2008 and it appears to be nearly identical to the image at Uni-Hamburg dated 11/9-10.

        Not surprising since they’re from the same satellite.

        • DeWitt Payne
          Posted Nov 10, 2008 at 10:57 PM | Permalink

          Re: Phil. (#460),

          Since CT uses SSM/I data for calculating area and Uni-Hamburg uses AMSR-E data, I wasn’t sure that the images on the site were in fact from the same satellite or which one it was. I’ll take your word for it though.

        • Phil.
          Posted Nov 11, 2008 at 10:00 AM | Permalink

          Re: DeWitt Payne (#465),

          Re: Phil. (#460),
          Since CT uses SSM/I data for calculating area and Uni-Hamburg uses AMSR-E data, I wasn’t sure that the images on the site were in fact from the same satellite or which one it was. I’ll take your word for it though.

          The High res image for the day’s ice is AMSR-E, the images you get when you do the comparison are at the lower resolution (whether they still use SSMI for that or convert AMSR to that resolution I don’t know but it shouldn’t matter). They use the SSMI equivalent resolution for their numerical data to maintain consistency over time.

  257. Mike Bryant
    Posted Nov 9, 2008 at 11:09 PM | Permalink

    I guess if you show the image of 4 to 6 days ago there will be less ice on the side-by-side comparison feature. Perhaps he might put a little note there so everyone will know to subtract the proper number of days from the right side image. Nahhhh.

  258. Mike Bryant
    Posted Nov 9, 2008 at 11:19 PM | Permalink

    I mean the left image.

  259. Mike Bryant
    Posted Nov 10, 2008 at 5:30 PM | Permalink

    Nope, Roos has ice continuing it’s ascent at the same pace. Besides CT, it seems that GISS is also an outliar, oops I mean outlier.
    October 2008 is apperently the warmest October ever:

    http://wattsupwiththat.com/2008/11/10/giss-releases-october-2008-data/

    I believe that this may be the Obama effect.

    • Jared
      Posted Nov 10, 2008 at 6:08 PM | Permalink

      Re: Mike Bryant (#461),

      The research is still ongoing, but it appears there may be some errors in the GISS data from Russia for October…it seems that many stations in Siberia/eastern Russia carried over their temperature numbers from September – which would explain that ridiculously huge swath of +4 to +13C anomalies over sparsely populated northeast Asia. ;)

  260. Mike Bryant
    Posted Nov 10, 2008 at 9:18 PM | Permalink

    Apparently The NOAA had the Russian September temps repeated for October. Of course, GISS should have checked the numbers given the huge anomoly. Especially since a plumber could check it on the internet. They are supposed to be the professionals.
    Strange that CT’s drop in the Arctic Ice Area might be considered by some as a confirmation of the GISS October number.

    Our public servants are not serving us very well.

  261. DeWitt Payne
    Posted Nov 10, 2008 at 11:10 PM | Permalink

    If you squeeze the Arctic ROOS image horizontally and vertically you can get a near perfect match of the 2008 curve with the recent 365 day plot at CT, but only if you shift the Arctic ROOS image forward about 5 days compared to the CT plot. That also shows that CT area will have to increase at least 500,000 km2 in the next five days. If you use Paint, you have to paste in the image from CT first because the plot background is colored and will block most of the Arctic ROOS image.

    JAXA 10 PM Eastern extent data for 11/10 was only 10,000 km2 larger than the final figure for 11/9. When the change is that small, the 9 AM update could be an increase or a decrease in extent.

    As for GISS October anomaly, look at the Arctic satellite images for 10/1 and 10/31/2008 on CT. It sure snowed a lot without melting during October for such a warm month.

  262. AndyW
    Posted Nov 14, 2008 at 2:07 AM | Permalink

    Looks like 2008 rebound is now tracking closely the proceding few years on the JAXA graph, will be interesting to see the value at the start of 2009 to give an indication of where the max will be in the Spring. Apart from this year that is normally a good indication.

    Congrats to Chris for spotting the Russia temp error by the way. His dedication is the way he wins all the discusions with layman Andy :D

    Regards
    Andy

  263. Erl Happ
    Posted Nov 14, 2008 at 7:51 AM | Permalink

    This diagram supports the observation of a climate shift in 1976-78 that ushered in a long series of El Ninos that have been responsible for the strong warming trend since that time. It relates to John Lang’s comments re tropical warming and the effect on global temperatures. It also relates to what is happening to the ice.

    Briefly, the air pressure in this zone is monitored as the Eastern end of the southern oscillation index, Darwin being the western leg. When 200hPa temperature rises (that’s 10km up) it weakens the downdraft over Tahiti and surface air pressure falls with important consequences namely.
    1. Cirrus density between 500hPa and the tropopause falls as 200hPa temperature rises allowing more sunlight to reach the surface of the ocean.
    2. The Easterly winds across the Pacific slacken as the pressure differential falls away.
    3. Warm waters from the west spread slowly eastwards and the centre of convection moves with these warm waters.
    4. Less cold water up-wells in the East Pacific.

    But, the critical question is what starts the process. It begins with a temperature increase at 200hPa. Why does this happen?

    The atmosphere above Tahiti has a strong downdraft and the air is very dry and very cold. It is straight from the tropopause and is rich in ozone. Ozone reacts strongly to two wave lengths. The first is UVB and the second is OLR from the Earth. You will notice that temperatures at 100hPa peak as temperatures at the other levels hit a trough. 100hPa temperature peaks also in August in the warmest part of the northern hemisphere summer. That is the clue to the temperature dynamics. Northern hemisphere summer temperatures are highly dependent upon the general level of atmospheric humidity and in the middle of a La Nina atmospheric humidity is minimal. The moisture has been rained out. So, under minimal humidity the northern hemisphere experiences a warmer summer of the sort just experienced. More cloud is lost and more sunlight gets through to the surface. OLR peaks strongly and 100hPa temperatures reflect the level of OLR.

    There is nothing gradual about the increase in 100hPa temperature in 1978. But the word gradual does apply to the decay of 200hPa temperature after 1978.

    The driver of surface warming is in part (not the re distributive component) the amount of cirrus cloud between 500hPa and the tropopause over Tahiti and in other ozone rich areas that have a cold upper troposphere. Cirrus density varies inversely with temperature because humidity at this level is relatively invariable. Above 500hPa atmospheric temperature responds to the level of UVB. 1978 marked the first big El Nino of the very strong solar cycle 21 that followed the weak cycle 20. The contrast between the period before and after 1978 was also in part due to stratospheric aerosols from volcanic activity producing cooling leading up to 1978.

    So, ENSO does have a cause and it is the sun. The current La Nina is due to a paucity of ultraviolet light and a weakness in the solar wind.

    Do you like this story?

  264. DeWitt Payne
    Posted Nov 14, 2008 at 9:30 AM | Permalink

    Uni-Hamburg still hasn’t posted October 2008 data. Maybe next week. Meanwhile, JAXA extent has been nearly flat the last two days with increases of 4541 and 3111 km2. CT area rate(~800,000 km2 increase in the last four days) should peak soon to reflect the recent extent rate drop.

  265. John Lang
    Posted Nov 14, 2008 at 11:19 AM | Permalink

    Erl, I’ve done a little more research into the 1975-76 climate shift. It was not the ENSO or the PDO, it was the Atlantic Multidecadal Oscillation (AMO) (or more accurately the AMO and ENSO together.) The AMO index reached a peak low of about -0.5C in 1975 and has since cycled up to about +0.5C in 2002 and has since been declining.

    The AMO seems to drive the longer cycles of the climate and ENSO drives the shorter swings in temps. There is much closer correlation of the AMO to arctic ice than ENSO.

    • Erl Happ
      Posted Nov 14, 2008 at 2:34 PM | Permalink

      Re: John Lang (#471),
      John, The data for southern and northern hemisphere is at http://i249.photobucket.com/albums/gg220/erlandlong/SSTglobal.jpg

      For what its worth this is what I think. What you see in SST at 0-10°N is the same as for 200hPa temperature in #469. The climate shift of 1978 is apparent in the behaviour of sea surface temperature at high latitudes in both hemispheres. One does not need to invoke an ‘oscillation’ manifesting in a particular NH ocean to explain it. What happens to Arctic ice depends on the trend in the bit of the graph circled. That in turn reflects what happens between the equator and 30° south. Because last southern summer was cold there is little warmth reaching that latitude and little warmth reaching the Arctic today.

      Looks to me like the big swing in temperature that saw a decline from 1880 to 1910 and an increase thereafter is reflected in southern hemisphere temperature in the near tropics and nowhere else. The most vigorous manifestation of the driver of that warming process is in the Eastern Pacific near Tahiti. Gilbert Walker worked out what was happening with ENSO a hundred years ago. What happened to ENSO in 1978 explains the shift from declining SST to increasing SST at high latitudes in both hemispheres and the northern hemisphere in particular. The gradual slackening in 200hPa temperatures after 1978 has allowed a build up of upper atmosphere cirrus cloud that is protecting the Earth from the suns radiation. That is most apparent in the south east Pacific.

      SST in the 0-10N latitude band is taking a dive and as it does Arctic Ice will grow. The Earth is like a big pot on a tiny gas ring and its a bit off centre. Its so large that little heat reaches the margins of the pot. Turn up the burner and a little more gets there to melt the Arctic ice from below.

      The ITCZ generates a lot of low cloud north of the equator. Most of the sea is in the southern hemisphere. Its the sea that is transparent and absorbs light energy storing it while increasing surface temperature very little at all. So, radiation does not increase when light falls on the sea.

      I will have no truck with NAO or PDO. Got to look at the big picture.

  266. Chris
    Posted Nov 14, 2008 at 12:09 PM | Permalink

    #468 AndyW:
    “Congrats to Chris for spotting the Russia temp error by the way. His dedication is the way he wins all the discusions with layman Andy”
    Thanks Andy! I’m a layman too, and I recall you’ve won a fair few as well. I didn’t do much re: Russia. I posted about this point on WUWT today, but I lost the post due to a glitch and didn’t have time to re-write it. Basically, a ten year old could have put two and two together given the WUWT post mentioning the surprising “heatwave” in Russia, and how simple it is to access the GISS station data.

  267. Chris
    Posted Nov 14, 2008 at 12:10 PM | Permalink

    AndyW:

    “Congrats to Chris for spotting the Russia temp error by the way. His dedication is the way he wins all the discusions with layman Andy”

    Thanks Andy! I’m a layman too, and I recall you’ve won a fair few as well. I didn’t do much re: Russia. I posted about this point on WUWT today, but I lost the post due to a glitch and didn’t have time to re-write it. Basically, a ten year old could have put two and two together given the WUWT post mentioning the surprising “heatwave” in Russia, and how simple it is to access the GISS station data.

  268. Stephen Wilde
    Posted Nov 14, 2008 at 1:58 PM | Permalink

    Re post 471.

    I’ve been pointing out for some time now that for global purposes we need to consider the net combined effect of ALL oceanic oscillations at any given point in time and then link that to solar variation and the average net position of the jet streams.

  269. DeWitt Payne
    Posted Nov 16, 2008 at 12:36 PM | Permalink

    Here’s a first cut at a guess for the 2009 maximum. Rather than guess the maximum itself, I’m going to guess the 3/1 to 31/2009 average extent. So here it is: 14.914 Mm2. It’s pretty much a SWAG, although there is some method involved.

    Current conditions: The Antarctic is losing area faster than average at the moment and Arctic is gaining area faster than average according to the CT data. The Arctic extent growth went through a pause recently and I expect the area data to follow suit with a five day lag. Uni-Hamburg still hasn’t posted October data.

  270. sean egan
    Posted Nov 19, 2008 at 5:08 PM | Permalink

    I am confused, from Chris #447 the arctic is not particularly warm.
    From http://nsidc.org/arcticseaicenews/
    “Over much of the Arctic, especially over the Arctic Ocean, air temperatures were unusually high. Near-surface air temperatures in the Beaufort Sea north of Alaska were more than 7 degrees Celsius (13 degrees Fahrenheit) above normal and the warming extended well into higher levels of the atmosphere. These warm conditions are consistent with rapid ice growth.”

    So
    a)is there a historically record to show when the arctic were warm at this time of the year, the ice growth was rapid?

    b) is the arctic warmer than usual, and if it is, is this a good sign for the 2009 minima?

  271. Chris
    Posted Nov 19, 2008 at 6:16 PM | Permalink

    #477 Sean

    “I am confused…” So am I :)

    These kinds of discrepancies make me extremely tempted to become what I would call a “skeptic” i.e. someone who just stops believing any claims whatsoever about rapid global warming.

    I’m afraid it would take me a long time to do justice to your questions, time I don’t have right now. But yes I think the trends of the last couple of months make it more likely the 2009 minimum will be well above the 2008 minimum. But who knows…..

  272. Dave:
    Posted Nov 19, 2008 at 10:29 PM | Permalink

    #477/478 – Sean & Chris

    Think of it this way: for the ocean to be freezing, heat is leaving the ocean and going into the atmosphere. (Because how does the ocean freeze? It freezes because it loses heat.) A record amount of freezing means a massive amount of heat loss – so there’s basically been big pipeline of heat going from the Arctic ocean, through the atmosphere and out into space. So it’s not that it was warm, and that caused it to freeze. The atmosphere was warm (or really, warmer than normal) because of the massive loss of ocean heat (which resulted in freezing).

    Also, to me it seems like this kind of anomaly isn’t likely to be repeated in November – once the freezing/heat loss has occurred, the atmosphere will no longer be warmed in this way. So subsequent temperatures should fall back towards the normal range (all else being equal).

  273. sean egan
    Posted Nov 20, 2008 at 12:51 AM | Permalink

    Dave,
    If more heat is transferring than usual, why? Either the thermal resistance is lower or the temp difference is bigger? Warmer air with the same thermal resistance at the water / air interface should slow the water cooling.
    So is there there a known reason for the thermal resistance to be lower?

  274. Chris
    Posted Nov 20, 2008 at 2:20 AM | Permalink

    Dave, my original post said “It’s also interesting to note that according to RSS TLT, Oct 08 in the Arctic/highest latitudes (60.0/82.5N) at +0.452C was cooler/less warm than Oct 1980 (+0.883C), Oct 1981 (+0.680C) and Oct 1987 (+0.620C). Even given the Arctic is now post-tipping point and 5C above “normal”. (Apparently).”

    ftp://ftp.ssmi.com/msu/monthly_time_series/rss_monthly_msu_amsu_channel_tlt_anomalies_land_and_ocean_v03_2.txt

    Now the following has also been published:

    http://climate.uah.edu/oct2008.htm

    The satellite data paint a somewhat different picture to that of the NSIDC, especially over the western Beaufort (lower troposphere colder than average, albeit with a more recent baseline i.e. 1979-1998)

    Of course there’s an effect from the latent heat of fusion. But the NSIDC are exaggerating: (i) the magnitude of this effect on temperatures within the Arctic region overall (probably a fraction of a degree at most)
    (ii) how “unusually high” the air temperatures were in the region in any event.

    Here’s what a month with genuinely “unusually high” temperatures over the Arctic looks like:

    http://climate.uah.edu/august2008.htm

    I’m not saying the satellites prove the NSIDC wrong. I’ve looked at all this a lot more thoroughly than that, and obviously the satellite data are only one bit of evidence which only implies so much. But with limited time, it’s the best way I can indicate that it might be worth questioning the NSIDC picture slightly more.

    • Phil.
      Posted Nov 21, 2008 at 8:56 AM | Permalink

      Re: Chris (#481),

      MSU satellite data aren’t particularly comparable with surface measurements in the Arctic due to the presence of a strong surface/near surface temperature inversion during the no-sun/low-sun seasons.

  275. Chris
    Posted Nov 20, 2008 at 3:38 AM | Permalink

    To be more precise, when I said “(i) the magnitude of this effect on temperatures…” in my previous post I meant the magnitude of the *extra* effect on temperatures in 2008 (compared with “normal”)

  276. MikeP
    Posted Nov 20, 2008 at 7:08 AM | Permalink

    I noticed from the Cryosphere Today animation that there’s a circulation inhibiting growth of sea ice in the Kara Sea and enhancing development along both sides of Greenland. Is this the circulation that normally moves some multi-year ice south into the N. Atlantic where it will melt next summer? Although the growth along Greenland seems about normal, the open water in the Kara Sea seems larger, i.e. the Kara Sea should be pretty much frozen by now. Does anyone have a take on what this will mean for next year?

  277. Dave:
    Posted Nov 20, 2008 at 7:55 AM | Permalink

    #480 Chris:

    Just to be clear, I’m a total layman – not an expert. But the reasoning makes sense to me. So, why did more heat transfer than usual? Because there was more heat to be transferred. The ocean was warmer (than average) – less of it was frozen. So it had a lot more heat content than usual, and so had a lot more heat it could lose. It’s like if you were in a cold room, and set a space heater to “warm”. You’ll get a certain amount of heat transfer – a certain rate of heat transfer. If you crank it to “hot”, you’ll get a greater rate of heat transfer. It’s the heat differential that determines the rate of transfer, and the ocean/atmosphere difference in October this year was greater than average (since less of the ocean was frozen to start with).

  278. Sam Urbinto
    Posted Nov 20, 2008 at 3:40 PM | Permalink

    Some comments from AR4 might be helpful here.

    TS.6.4.4 Sea Level

    Robust Findings:
    Sea level will continue to rise in the 21st century because of thermal expansion and loss of land ice. Sea level rise was not geographically uniform in the past and will not be in the future. {10.6}
    Projected warming due to emission of greenhouse gases during the 21st century will continue to contribute to sea level rise for many centuries. {10.7}
    Sea level rise due to thermal expansion and loss of mass from ice sheets would continue for centuries or millennia even if radiative forcing were to be stabilised. {10.7}

    Key Uncertainties:
    Models do not yet exist that address key processes that could contribute to large rapid dynamical changes in the Antarctic and Greenland Ice Sheets that could increase the discharge of ice into the ocean. {10.6}
    The sensitivity of ice sheet surface mass balance (melting and precipitation) to global climate change is not well constrained by observations and has a large spread in models. There is consequently a large uncertainty in the magnitude of global warming that, if sustained, would lead to the elimination of the Greenland Ice Sheet. {10.7}

  279. Sam Urbinto
    Posted Nov 20, 2008 at 3:46 PM | Permalink

    Or even more to the point on Arctic Ice specifically, in TS.6.2.2 Snow, Ice and Frozen Ground

    RF: Since 1978, annual mean arctic sea ice extent has been declining and summer minimum arctic ice extent has decreased. {4.4}

    KU: There are insufficient data to draw any conclusions about trends in the thickness of antarctic sea ice. {4.4}

  280. AndyW
    Posted Nov 21, 2008 at 2:06 AM | Permalink

    The way I looked at NSIDC’s comments was that the energy lost from the refreeze is lost to the environment and because the amount of refreeze was rapid a lot of energy was dumped quickly which helped warm the region because input into it was greater that radiation out. It was a rate of freezing that caused the temp rise, not the energy in the sea as such.

    I’m not sure their terminology is misleading though you could hardly wear a Hawaiian shirt fishing through your ice hole. Would it not better to say that this release of energy slowed the cooling down of the region rather than warmth?

    Regards
    Andy

  281. John Lang
    Posted Nov 21, 2008 at 8:03 AM | Permalink

    Speaking of sea level, the agency now running the Jason satellites, Aviso, has started updated the data again (now to the end of July).

    Looks like sea level has starting increasing again after a 2 year lull. This is the raw unadjusted data. July is usually the low month of the year so these numbers are probably a relatively big increase.

    This is what it looks like when they apply all their “adjustments”.

  282. tty
    Posted Nov 21, 2008 at 9:06 AM | Permalink

    Re 486

    “There is consequently a large uncertainty in the magnitude of global warming that, if sustained, would lead to the elimination of the Greenland Ice Sheet.”

    However, based on temperature data from the previous interglacial it requires a warming considerably larger than 5 degrees centigrade in Greenland.

  283. Sam Urbinto
    Posted Nov 21, 2008 at 12:27 PM | Permalink

    A slight reduction of base pH is acidification, a slowing in cooling is warming, a vague uncertainty about magnitude is disaster.

    I cry for baby sea ice.

    • Phil.
      Posted Nov 21, 2008 at 9:09 PM | Permalink

      Re: Sam Urbinto (#491),

      Do you have something against logs Sam?

      A slight reduction of base pH is acidification,

      An increase in the H+ ion concentration of 30% certainly qualifies.

      • John M
        Posted Nov 21, 2008 at 10:20 PM | Permalink

        Re: Phil. (#492),

        Phil.

        Is this the kind of variability one can expect from ocean pH data?

        http://www.wunderground.com/climate/acidoceans.asp

        If so, looks like +/- 30% and greater happens pretty regularly.

        • Phil.
          Posted Nov 22, 2008 at 1:24 PM | Permalink

          Re: John M (#493),

          Phil.
          Is this the kind of variability one can expect from ocean pH data?

          http://www.wunderground.com/climate/acidoceans.asp

          If so, looks like +/- 30% and greater happens pretty regularly.

          From that graph it seems that a std dev of ~0.02 in pH is reasonable, that’s ≃5% in [H+],
          the change in pH due to increased CO2 is said to be ~0.1 which is ≃25%. So a change of ±30% isn’t observed regularly and a systematic change in [H+] of that magnitude is cause for concern.

  284. DeWitt Payne
    Posted Nov 21, 2008 at 11:14 PM | Permalink

    NSIDC’s statement that rapid freezing warmed the air could easily be misread as a violation of fundamental thermodynamics. It’s very poorly phrased. This is clearly a case where some reference should have been made to the actual temperature rather than the temperature anomaly. Heat of fusion will maintain the temperature near the freezing point. Metal freezing points (and the triple point of water) are, for example standard reference points for the standard temperature scale because the temperature of the melt changes hardly at all over most of the range between completely molten and completely solid. Unless you can formulate a reverse Mpemba effect (warm water freezes faster than cold water), freezing water cannot warm the air above it, it can only keep it from cooling. If there isn’t any wind, a strong temperature inversion will form over land (or ice for that matter) when the sun is below the horizon. So the air above a solid, dry surface can be significantly warmer than very near the surface. But that won’t happen over open water because the thermal conductivity and heat capacity of water is much higher than land or ice. I’m afraid that still isn’t very clear, even to me, but here it is anyway.

  285. Richard Sharpe
    Posted Nov 22, 2008 at 11:04 AM | Permalink

    John M said:

    Phil.

    Is this the kind of variability one can expect from ocean pH data?

    Acid Oceans

    If so, looks like +/- 30% and greater happens pretty regularly.

    However, the article you linked offered no support, at least as far as I could see, for the claim that such increases and decreases were pretty regular. Indeed, it seemed to make the outlandish claim that we have not seen such increases on acidity for some 650,000 years and perhaps millions of years.

    • John M
      Posted Nov 22, 2008 at 12:51 PM | Permalink

      Re: Richard Sharpe (#495),
      I was referring to the scatter in the graphs (~0.06 – 0.08 pH units, or 15-20% if you like), despite the fact that

      The mean seasonal cycle was removed from the data

      I have been looking for a concise exposition (as Steve Mc. says) of the errors and uncertainties in ocean pH measurements. As a chemist with more years of industrial experience than I’d like to think about, I can imagine standing up in front of a plant manager with the sort of data I linked to and explain to him why he ought to spend even $500 to address the “alarming trend shown in the data”.

      I’ve seen lots of fuzzy language about buffering effects and troubling statements that equilibrium constants determined in “synthetic” sea water are somewhat different than those determined in “natural” sea water.

      Since my understanding is that ocean pH is generally determined indirectly by measuring CO2/carbonate/bicarbonate and then calculating pH based on “known” equilibrium constants, I really do have a big question with regard to uncertainties in stated pH values.

      I should note that I’ve noticed that skeptics and believers alike are sometimes under the impression that measuring ocean pH is as simple as sticking a pH probe in the ocean and writing down a value.

      Anyway, I note this is a sea ice thread. There is at least two ocean pH threads at the BB here and here where perhaps we should continue the discussion.

      • Phil.
        Posted Nov 22, 2008 at 1:46 PM | Permalink

        Re: John M (#496),

        Try this:
        Although potentiometric pH measurements are versatile and satisfactory for many applications, spectrophotometric pH measurement procedures have at least two important advantages that make them particularly desirable. Since spectrophotometric pH measurements can be determined via absorbance ratios, and the calibration of pH indicators is a laboratory exercise that establishes how each indicator’s molecular properties vary with temperature, pressure and ionic strength, spectrophotometric pH measurements are inherently calibrated and can be termed “calibration free” . Subsequent to careful laboratory calibration, spectrophotometric pH measurements do not require the use of buffers. Secondly, thousands of at-sea observations have demonstrated that the imprecision of shipboard spectrophotometric pH measurements is on the order of 0.0003 to 0.0004 pH units, approximately an order of magnitude better than potentiometric results. These advantageous attributes of spectrophotometric pH measurements have made spectrophotometric procedures valuable for not only observations of pH, but also for measurements of CO2 fugacity and total dissolved inorganic carbon.

        Spectrophotometric pH measurements have been increasingly utilized for measurements of pH in natural waters. Bellerby et al. developed a flow injection procedure for spectrophotometric measurement of seawater pH with a reported precision of 0.005 pH units and a sample frequency of 25 hr”1(Bellerby R. G. J.; Turner, D. R.; Millward, G.E.; Worsfold P.J. Analytica Chimica Acta 1995, 309, 259.). Tapp et al. described the use of a shipboard system for spectrophotometric measurements, of surface water pH with a reported precision on the order of 0.001 pH units and a 1- Hz measurement frequency (Tapp, M.; Hunter, K.; Currie, K.; Mackaskill, B. Mar. Chem. 2000, 72, 193.). Relative to discrete measurements however, observed discrepancies were as large as 0.02 pH units. Martz et al. described the construction of a submersible pH sensor with a 0.003 unit measurement precision and a measurement frequency of 6 hr”1 (Martz, T. R.; Carr, J. J.; French, C. R.; DeGrandpre, M. D. Anal. Chem. 2003, 75, 1844.)

        From http://www.wipo.int/pctdb/en/wo.jsp?IA=US2006013524&DISPLAY=DESC

        • John M
          Posted Nov 22, 2008 at 2:12 PM | Permalink

          Re: Phil. (#498),

          Thanks for the links.

          I’m familiar with the spectrophotometric techniques and had a discussion with “Jack” about them here a while back.

          I think, however, the majority of ocean pHs are still determined using CO2/carbonate/bicarbonate measurements, and measurements around the world do in fact vary routinely by tenths of pH units.

          Can you add anything to my concerns about the equilibrium constants?

          BTW, I’ve got no problem with “more CO2, lower pH”. Just as I’ve got no problem with “more CO2, higher temperature”.

          I just need to be convinced that “lower pH” means more than one or two tenths of a pH unit per century and “higher temperature” means more than 0.7 – 1 deg/century.

          Visually determined standard deviations on data that’s already been filtered by removing seasonal cylces don’t do it for me.

  286. Mike Bryant
    Posted Nov 22, 2008 at 6:34 PM | Permalink

    Here is an interesting take on CO2 and ocean acidity from an australian scientist:

    http://jennifermarohasy.com/blog/2008/10/not-enough-co2-to-make-oceans-acidic-a-note-from-professor-plimer/

    Here Freeman Dyson seems to think that controlling shoud be very easy. He calls it only a “land management problem”. It is in about the last thirty seconds or so of the video.

  287. DeWitt Payne
    Posted Nov 23, 2008 at 1:15 PM | Permalink

    Uni-Hamburg updated their data on AMSR-E extent and area for the Arctic and Antarctic through 10/31/2008 yesterday. I regressed the CT area against the UH area (n=99) and found the best fit required an offset of six days between the calendar date that CT posts their data and the date of the UH data. The F statistic for the regression was 8158 for an offset of five days, 13052 for six days and 9714 for seven days. The slope and intercept for the six day fit was 1.0465 and -0.6833. I was using the previous day’s date for the CT data in my spreadsheet, so that gave an offset of five days for the best fit. Before this update, the difference in the F statistic for the regression between an offset of four and five (now five and six) was not very large.

    The Antarctic is currently very close to the 1979 to 2000 average. The Arctic anomaly at -0.6 Mm2 from than the 1979 to 2000 average is well above the 2002 to 2007 average. There has been a significant loss of extent and area in the Kara Sea recently and the Barents Sea is not freezing as fast as normal. It looks to me like there may have been an incursion of warm water that has melted the ice there. The Chukchi Sea north of the Bering Strait is freezing much faster than 2007, though and has an anomaly near zero from the 1979 to 2000 average.

  288. AndyW
    Posted Nov 28, 2008 at 12:20 AM | Permalink

    Holy Artic Heatwaves batman

    http://www.meteo.uni-koeln.de/meteo.php?show=En_We_Wk
    :)

    Regards

    Andy

    • Dave Dardinger
      Posted Nov 28, 2008 at 10:36 AM | Permalink

      Re: AndyW (#502),

      Someone ought to tell them they have some of their labels wrong. I went to the snow cover maps for Europe first and in the legend it shows “land” and “ocean” backward. But when I looked at a couple of world maps they were correct. I suppose it was just sloppiness and perhaps just in the English version, but surely not every site has to have such errors?

  289. AndyW
    Posted Dec 5, 2008 at 1:33 AM | Permalink

    IARC JAXA extent graph for 2008 has now joined the “main sequence” for past years and indeed is dropping towards the lower bound. Still almost 1 month to go before we can see where it starts the important stage where max is determined.

    I’m going to take a punt and say max will be 13.8-13.9 and so be towards the lower end of the maximum extent for the past few years.

    Regards

    Andy

  290. Buddenbrook
    Posted Dec 5, 2008 at 2:03 AM | Permalink

    Roger Pielke Sr’s weblog on the subject: http://climatesci.org/2008/12/04/are-there-long-term-trends-in-the-start-of-freeze-up-and-melt-of-arctic-sea-ice/

  291. mark in austin
    Posted Dec 8, 2008 at 9:19 PM | Permalink

    wow….suddenly it is jumping up quickly! any thoughts?

  292. hswiseman
    Posted Dec 9, 2008 at 12:15 AM | Permalink

    As long as the cute girls believe in AGW, its an uphill battle to anyone to focus on the math and science.

  293. hswiseman
    Posted Dec 9, 2008 at 12:16 AM | Permalink

    http://www.boston.com/lifestyle/green/gallery/polarplunge/

  294. Hoi Polloi
    Posted Dec 9, 2008 at 3:52 AM | Permalink

    Interesting to see people risking pneumonia for the sake of AGW…

  295. Gerald Machnee
    Posted Dec 9, 2008 at 7:34 AM | Permalink

    Looks like a lot of “See” Ice.

  296. sean egan
    Posted Dec 15, 2008 at 12:55 PM | Permalink

    From IJIS

    Dec 11 2008 11678594
    Dec 12 2008 11681563
    Dec 13 2008 11662813
    Dec 14 2008 11640625

    Anyone know why has the snow stopped going up the last few days? Wind compacting?

    • Len van Burgel
      Posted Dec 15, 2008 at 6:44 PM | Permalink

      Re: sean egan (#512),
      The value for Dec 14 is a decrease to 11623438 km2.

      The daily readings fluctuate considerably. It may have to do with inherent inacurracies in estimating the daily values. Although the ice extent has now decreased for two days, the ice extent overall has still increased by 670,000 km2 over the last 10 days which is an average of 67,000 km2 per day.

  297. AndyW
    Posted Dec 16, 2008 at 1:05 AM | Permalink

    Definite “bobble” on the graph now pushing it from near the top of the band for previous years down towards the bottom. It will be interesting to see if it is smoothed out or how it progresses from here.

    Regards

    Andy

  298. Leonard Herchen
    Posted Dec 16, 2008 at 1:40 AM | Permalink

    The numbers have just been changed to give less ice. What gives? See

    http://wattsupwiththat.com/2008/12/13/something-is-rotten-in-norway-500000-sq-km-of-sea-ice-disappears-overnight/#more-4514

    I’m surprised its not been mentioned here yet.

    • DeWitt Payne
      Posted Dec 16, 2008 at 12:33 PM | Permalink

      Re: Leonard Herchen (#515),

      Arctic ROOS had an error in some of their charts that lasted for months. They have now corrected it. The ice didn’t disappear, it was never there. I mentioned that there was an error some time ago. If I weren’t so lazy and the internal search engine here didn’t suck so badly, I’d look up and link to the actual comment.

      The recent big run up in extent was the result of Hudson Bay freezing over, finally. This has been somewhat balanced by a lowering of concentration elsewhere so the area has not gone up as quickly as the extent. Uni-Hamburg should post their November data soon. More then.

  299. Joe Solters
    Posted Dec 16, 2008 at 8:32 AM | Permalink

    NASA is just now coming out with a new study (American Geophysical Conference, Dec. 18, 2008, San Fran.) which according to AP, claims that land ice melting since 2003 totals 2 trillion tons. This resulted in sea level rise of one fifth of an inch in that period. One fifth of an inch? WOW! Is Madoff building an ARK Fund I can invest in?

  300. BarryW
    Posted Dec 17, 2008 at 10:09 AM | Permalink

    I would think that the negative numbers must be due to compaction. Extent was increasing above average then moved below for the last week. It’s starting to move back to the average. The extent seems to be tracking 2005 right now.

  301. sean egan
    Posted Dec 19, 2008 at 1:18 AM | Permalink

    The extent has been static for a week now on IJIS. I do not see any other pauses this long in other years.

    • DeWitt Payne
      Posted Dec 19, 2008 at 2:23 AM | Permalink

      Re: sean egan (#519),

      I do not see any other pauses this long in other years.

      Then you’re not looking very hard. Look at 2006. There was a big flat spot in November and a smaller one in December about the same time of year as now.

  302. Nylo
    Posted Dec 19, 2008 at 1:41 AM | Permalink

    Considering the difference in the sea ice anomalies, when at the mininum, in the SH between Feb 2007 and Feb 2008, and the same for the NH between Sep 2007 and Sep 2008, how much baby-ice has become adult-ice this year?

  303. sean egan
    Posted Dec 19, 2008 at 12:54 PM | Permalink

    DeWitt Payne #521

    “There was a big flat spot in November and a smaller one in December about the same time of year as now.”
    Nov 2006 has missing days. Dec 2006 looks smaller.

    • DeWitt Payne
      Posted Dec 19, 2008 at 10:23 PM | Permalink

      Re: sean egan (#522),

      missing days

      Oh yes, four whole days and the extent at the beginning of the lacuna was 9377656 km2 on 11/16/2006 and 9363438 km2 on 11/21/2006. Even if you don’t interpolate, it still looks flat to me and resulted in a precipitous drop in the average rate that lasted for quite a while. It remains to be seen how the current situation will play out, but at the present it’s just weather and has had little effect on the longer term averages.

  304. mark in austin
    Posted Dec 19, 2008 at 1:09 PM | Permalink

    hmmm, so why is this happening in spite of it being so cold up there? is it because it fills up certain areas and now we are waiting until the cold reaches a bit further south and starts freezing up more?

  305. mark in austin
    Posted Dec 21, 2008 at 10:02 PM | Permalink

    i am still trying to figure out how the ice can drop while the temps are far below zero up there. any theories?

    • DeWitt Payne
      Posted Dec 21, 2008 at 11:04 PM | Permalink

      Re: mark in austin (#525),

      It’s warm water from the Gulf Stream that’s probably the cause of the melting on the Greenland side of the Arctic. The pattern of the ice, particularly around Svalbard, seems to me to be quite consistent with warm water flow increasing and decreasing from fluctuations in the position of the Gulf Stream at high latitude.

  306. AndyW
    Posted Dec 22, 2008 at 12:39 AM | Permalink

    Yes, on the Atlantic side the ice forms to the leeward side of the islands, from a current perspective rather than the wind.

    At this rate it’s going to go into the new year not much higher than 12×106 m2 which will be interesting for the maxima.

    Regards

    Andy

  307. Freezedried
    Posted Dec 22, 2008 at 10:27 AM | Permalink

    Probably residual effects from all those Soviet nuclear tests and the nuclear waste dumping around Novaya Zemyla during the cold war.

  308. curious
    Posted Dec 22, 2008 at 4:18 PM | Permalink

    Sea ice has stopped freezing this winter:

  309. Patrick
    Posted Dec 22, 2008 at 4:45 PM | Permalink

    Just as it was about to reach the normal average, it suddenly stops freezing. As I live in the Southern Hemisphere, I can only go by what I see on the news but it seems to me that you guys in the North are having a pretty diabolical start to winter and temperatures appear to be very very cool.

    How on earth then can the sea ice not only have stopped freezing, but also now be at a record low level for this time of year? It just doesn’t fit with the observations that come from looking out the window (or in my case, watching the news).

  310. curious
    Posted Dec 22, 2008 at 4:46 PM | Permalink

    re:curious at 529
    curious – there is another Curious?

    [yes.]

  311. iskanda
    Posted Dec 23, 2008 at 4:06 AM | Permalink

    @ Patrick:530
    over on watts, there has been mention of warm water halting sea ice growth. This thread http://wattsupwiththat.com/2008/12/15/northeast-siberia-braces-for-extreme-cold-of-60c/

    “The reason for the halt is warm water. See water temperature anomalies at http://www.osdpd.noaa.gov/PSB/EPS/SST/data/anomnight.12.15.2008.gif
    There’s a tongue of above-normal water sticking up between Labrador and Greenland and there’s anarea on the west side of Novaya Zemlaya Island. In addition, there are some way-above-normal hotspots…
    – on the northern tip of the Scandinavian Peninsula
    – on the north and west coast of Svalbard
    – on the north coast of Iceland

    As long as those warm areas remain, no more major freezing”. (Posted by Walter Dnes on 16/12/08)

  312. Posted Dec 23, 2008 at 6:34 AM | Permalink

    Areas of sea ice usually include both ice and open water. Wind blows the ice around in these areas. This wind-driven movement can cause the ice to bunch together (which may reduce the extent of the ice) or spread out (which increases the extent, though if the percent coverage falls below 15% then the area is no longer counted).

    I suspect that the recent pause in growth of the Arctic “sea ice extent” reflects wind behavior, where winds have caused ice to temporarily bunch together or spread so far apart that it falls below 15% coverage. Typically this is temporary and, as the winds shift, we may see a “spurt” in which the reported ice extent suddenly increases.

  313. John M
    Posted Dec 23, 2008 at 12:16 PM | Permalink

    RJP SR has more regarding the situation in the Arctic.

    link

  314. See - owe to Rich
    Posted Dec 23, 2008 at 12:37 PM | Permalink

    A bit more, or different, information from that linked at #534. The UK was very cold until mid-December, with a high in mid-Atlantic driving northerly winds down. These winds would have also slowed the Gulf Stream (?) and pushed ice south to increase extent. The conditions have now reversed and it’s been up to 12-13C in the UK recently.

    So it probably is just weather.

    Rich.

  315. Posted Dec 23, 2008 at 12:50 PM | Permalink

    As i have seen in the discovery channel last month,some of the scientist is covering some of the ice in the north pole to prevent the ice from melting so fast.
    Is there anyone seen that episode?

  316. Red Etin
    Posted Dec 24, 2008 at 11:34 AM | Permalink

    533 and prior posts

    Has anyone ever attempted a narrative description of the major oceanographic and meteorologic influences that have created this 30-year pattern of seasonal sea ice oscillation in the Arctic (or Antarctic)?

  317. Mike Bryant
    Posted Dec 24, 2008 at 1:34 PM | Permalink

    Steve Keohane posted this at WUWT…

    Steve Keohane (10:14:56) :

    Regarding another popular depiction of NH ice, I spent a little time on Cryosphere the other day and noticed something odd in comparing 12/20/80 to 12/22/80 NH ice extent. Hudson Bay and the outlet of Ob river in Russia, the boot-shaped inlet next to the arctic, appeared larger in the 1980 plat. I took the landmass/shoreline from 1980 and overlaid it on the 2008 plat, and got this: http://i44.tinypic.com/330u63t.jpg
    The Arctic and Bering Seas appear to have been encroached upon, ie. made smaller. The white adjacent to the shoreline is the current snow/shoreline, and shows the loss of available ‘sea’ area for ice. I think that the representation of the river Ob’s outlet is more realistic on the 2008 shoreline, but previously a much greater extent of ice was measured there. There are extensive areas off eastern Russia that used to be measured for ice extent but now are designated ‘snow’, as is much of the shoreline of the whole arctic. This next blink comparison is four images, 12/20/1980, 9/22/2008, 12/20/1980, and 12/22/2008. I chose 9/22 as a minimum ice & snow image to maximize the modern shoreline. This allows the examination of the old shoreline with the modern shoreline of 9/22 and the modern snowline of 12/22. I hope this is all a change in accounting for the ice to make it more realistic, but can’t help but wonder if it is only another in a long list of biases to exaggerate the effect of AGW. If the latter is true, the ice extent can never be what it once was. http://i39.tinypic.com/b7f4fc.jpg
    Please note that I retained the star background in all images, and used their pixels for image registration. At full size, I see no perturbation of those pixels from one image to the other, and therefore assume they are correctly registered. Merry Christmas to all from a deeply white western Colorado, 80″ so far this Dec. on Aspen Mtn., more to come tomorrow and the weekend.
    Expect to break all records.

    • Phil.
      Posted Dec 26, 2008 at 10:15 PM | Permalink

      Re: Mike Bryant (#538),

      What you have to remember is that the resolution of the SMMR images was 25x25km, the width of the Ob estuary is in the 25-60 km range and so won’t be well represented, the ASMR-E images use a pixel 12.5×12.5 km which will give a much better representation.

  318. Chris
    Posted Dec 24, 2008 at 7:19 PM | Permalink

    Happy xmas all.

    #535 It`s only been a temporary reversal in the UK. Turning colder again as we speak, with the possibility of another prolonged cold spell into January. If it hadn`t been for the mild spell, 2008 CET would have been the coldest since 1996 (despite some exceptionally mild months at the beginning of the year). Now probably going to finish just above 2001 (as 2nd coldest since 1996.)

    As for the ice extent, look where the ice is anomalously low on the NSIDC map, and look at SST anomalies.

    http://www.nsidc.org/data/seaice_index/daily.html

    http://weather.unisys.com/surface/sst_anom.html

    Looks like extent may start to catch up east of Russia quite soon with the intensifying pool of cold anomalies.

    As for the area NE of Scandinavia, the strong Greenland high looks to be preventing much of an incursion of milder Atlantic air for some time, though the stormy conditions may limit ice formation.

    http://www.coldswell.com/longrange.php

    Anyway, these are just a few thoughts – better get back to Xmas duties…..

  319. Nylo
    Posted Dec 26, 2008 at 2:41 AM | Permalink

    Is it just by chance that, at the same time that we have seen a strongly atypical reduction in sea ice area for this time of the year, by December the 16th we have also seen a sharp increase in the NH snow cover? Maybe they have started to consider some area that was previously sea-area as being land-area snip

  320. Mike Bryant
    Posted Dec 26, 2008 at 8:01 AM | Permalink

    I have sent these questions to Dr Chapman at CT.

    Dr William Chapman,
    Can you please explain a couple of things on the Cryosphere Today “Compare side-by-side images of Northern Hemisphere sea ice extent” product, please?
    Why does the snow in the more recent dates cover areas that were previously sea inlets, fjords, coastal sea areas, islands and rivers? (Water areas, most easily discernible in the River Ob inlet)
    Why does the sea ice in the older images cover land areas? (Land areas, most easily discernible in River Ob inlet)
    See this overlay:

    Looking forward to your answer,
    Mike Bryant

  321. Mike Bryant
    Posted Dec 26, 2008 at 11:15 PM | Permalink

    Phil.,
    What you must remember is that these comparisons are not in any real sense whatsoever “comparisons” then, are they? Would they still be presented in this way if the impression received were reversed? Do words like “comparison” or “misleading” mean what they used to mean? If these images cannot be made to represent a comparative reality, should they even be called comparisons?
    Phil., you know these trivialities about the size of the pixels and perhaps you are not misled, would you say that the average user of CT at least MIGHT be misled by these images?
    The first time I brought this up on this thread you said that the larger size of Greenland in the later dates was only an “optical illusion”.
    In post 368 you said:
    Re: Mike Bryant (#367),
    Mike I think you’ll find it’s an optical illusion, all that white seems larger.
    I tested this by subtracting the one image from the other, the image is shown below, the extra white is just the snow on the land area of the first image.

    However it is not an optical illusion. http://i44.tinypic.com/330u63t.jpg

    These images are misleading and should either be corrected or discarded with explanations for either course.

  322. Phil.
    Posted Dec 27, 2008 at 6:53 PM | Permalink

    Mike as I told you before the image overlay showed no significant difference between the coastlines (except for some minor snow over fast ice). I ran the same test this time with the same result, however your 1980 image shows significant distortions which leads to your result, that’s where your problem lies.

  323. Mike Bryant
    Posted Dec 27, 2008 at 7:03 PM | Permalink

    So are you saying that the 1980 image isthe only misleading then?

    • Phil.
      Posted Dec 27, 2008 at 7:12 PM | Permalink

      Re: Mike Bryant (#545),

      Yes, your 1980 image appears to be distorted wrt the images on CT, perhaps something happened when you rescaled it?

  324. Mike Bryant
    Posted Dec 27, 2008 at 8:15 PM | Permalink

    Well the overlay was done by Steve Keohane. This his explanation of the images:

    “Regarding another popular depiction of NH ice, I spent a little time on Cryosphere the other day and noticed something odd in comparing 12/20/80 to 12/22/08 NH ice extent. Hudson Bay and the outlet of Ob river in Russia, the boot-shaped inlet next to the arctic, appeared larger in the 1980 plat. I took the landmass/shoreline from 1980 and overlaid it on the 2008 plat, and got this: http://i44.tinypic.com/330u63t.jpg
    The Arctic and Bering Seas appear to have been encroached upon, ie. made smaller. The white adjacent to the shoreline is the current snow/shoreline, and shows the loss of available ‘sea’ area for ice. I think that the representation of the river Ob’s outlet is more realistic on the 2008 shoreline, but previously a much greater extent of ice was measured there. There are extensive areas off eastern Russia that used to be measured for ice extent but now are designated ‘snow’, as is much of the shoreline of the whole arctic. This next blink comparison is four images, 12/20/1980, 9/22/2008, 12/20/1980, and 12/22/2008. I chose 9/22 as a minimum ice & snow image to maximize the modern shoreline. This allows the examination of the old shoreline with the modern shoreline of 9/22 and the modern snowline of 12/22. I hope this is all a change in accounting for the ice to make it more realistic, but can’t help but wonder if it is only another in a long list of biases to exaggerate the effect of AGW. If the latter is true, the ice extent can never be what it once was.

    Please note that I retained the star background in all images, and used their pixels for image registration. At full size, I see no perturbation of those pixels from one image to the other, and therefore assume they are correctly registered. Merry Christmas to all from a deeply white western Colorado, 80″ so far this Dec. on Aspen Mtn., more to come tomorrow and the weekend.
    Expect to break all records.”

    Since Steve retained the stars in each of the visualizations, I think that any mistake that was made would have to be attributed to CT.

  325. Steve Keohane
    Posted Dec 27, 2008 at 10:13 PM | Permalink

    Phi, re: 546: There was no rescaling of any image, no change was made to the pixels, with the exception of tinting Greenland and a few islands blue so they would have contrast when overlaid.

  326. Steve Keohane
    Posted Dec 27, 2008 at 10:15 PM | Permalink

    Phil, sorry for omitting the ‘l’ above.

  327. Phil.
    Posted Dec 28, 2008 at 1:43 AM | Permalink

    Steve, I was puzzled by the discrepancy so I went back and rechecked. The reason I mentioned rescaling was because the size of your image was different than the original. My comparison with September 1980 showed that the coastline then matches the current winter line (snow or not). However looking at the winter images from 1980 shows the ‘sea ice’ encroaches onto the land. So the snow line is a good match with the actual coastline but the old coastline in the presence of sea ice isn’t. The actual numbers are derived from the designation of the individual pixels as ‘land’ or ‘sea’ and would be unaffected.
    So for some reason the older winter images have the distortion you showed, otherwise everything is compatible, the present hi res ASMR-E images are better but only have a limited history.

    • DeWitt Payne
      Posted Dec 28, 2008 at 12:08 PM | Permalink

      Re: Phil. (#550),

      Isn’t AMSR-E resolution 6.25×6.25 km? That’s the impression I get from the Uni-Hamburg comment that their data is on a 6.25 km grid.

      Now for the final nail in the coffin of the idea that CT and JAXA are synchronous:
      Assumed synchronous, CT area JAXA extent
      date area difference extent difference
      12/17 10.766 0.030 11.704 -0.028
      12/18 10.791 0.025 11.688 -0.016
      12/19 10.826 0.035 11.707 0.019
      12/20 10.887 0.061 11.727 0.020
      12/21 10.962 0.075 11.776 0.049
      12/22 10.952 -0.010 11.889 0.113
      12/23 10.932 -0.020 11.989 0.100
      12/24 10.912 -0.020 12.060 0.071
      12/25 10.949 0.037 12.125 0.065
      12/26 11.142 0.193 12.142 0.017
      12/27 11.339 0.197 12.075 -0.067

      Not very good agreement in the wiggles is there. But shift the CT data back five days so 12/27 is actually 12/22 and things look much better. Here’s a plot covering a longer time scale using exponentially smoothed rates with the CT data shifted five days. So expect the CT area to continue to increase for a couple of days and show a big drop in the area posted on 1/2/2009 which will actually be for 12/27/2008.

      The holiday season seems to have kept Uni-Hamburg from posting their November data on schedule.

      Using the average increase in extent from this date from JAXA for 2003 to 2008, the 2009 peak extent looks to be about 14.1 Mm2.

  328. Mike Bryant
    Posted Dec 28, 2008 at 10:41 AM | Permalink

    Phil. you said:
    “The actual numbers are derived from the designation of the individual pixels as ‘land’ or ‘sea’ and would be unaffected.
    So for some reason the older winter images have the distortion you showed, otherwise everything is compatible, the present hi res ASMR-E images are better but only have a limited history.”

    Are you sure that the individual pixels are unaffected since the resolutions are now different?

    SMMR images was 25×25km, the ASMR-E images use a pixel 12.5×12.5 km.

    Also since these resolutions are different shouldn’t an extraordinary amount of care be used in any “comparison” product to eliminate the charge of bias?
    As for the snow data, I believe it should not properly be included in a product that is primarily for comparisons of sea ice. Do you think that comparisons should include extraneous data that could easily be set aside?
    Mike Bryant

    • Phil.
      Posted Dec 28, 2008 at 3:33 PM | Permalink

      Re: Mike Bryant (#551),

      Mike Bryant:
      December 28th, 2008 at 10:41 am
      Phil. you said:
      “The actual numbers are derived from the designation of the individual pixels as ‘land’ or ‘sea’ and would be unaffected.
      So for some reason the older winter images have the distortion you showed, otherwise everything is compatible, the present hi res ASMR-E images are better but only have a limited history.”
      Are you sure that the individual pixels are unaffected since the resolutions are now different?
      SMMR images was 25×25km, the ASMR-E images use a pixel 12.5×12.5 km.

      The comparison images are SMMR, when I did the overlays I used SMMR images for consistency, on the CT site you can choose the resolution you want. (as I recall 12.5km is the highest resolution displayed).

      Also since these resolutions are different shouldn’t an extraordinary amount of care be used in any “comparison” product to eliminate the charge of bias?

      That’s why you have the option of choosing image resolution.

      As for the snow data, I believe it should not properly be included in a product that is primarily for comparisons of sea ice. Do you think that comparisons should include extraneous data that could easily be set aside?

      That’s the set of winter images (+ snow) that has the correct representation of the coastline, perhaps it would be better to use the correct mask on the early winter images?
      If CT feel that showing the snow is informative that’s their call.

  329. Mike Bryant
    Posted Dec 28, 2008 at 3:56 PM | Permalink

    “That’s why you have the option of choosing image resolution.”

    On the product called “The Cryosphere Today Compare Daily Sea Ice” I absolutely have no choice of image resolution.
    On the product that has been specifically designed to “compare any two dates” you take what you get. Unfortunately you get a skewed view. This product is obviously designed for people like me who want to click a couple of times and compare the past with the present. This product, The Cryosphere Today Compare Daily Sea Ice, not whatever you are talking about, is very misleading.
    Of course CT can do whatever they wish, but it is still misleading.

    • Phil.
      Posted Dec 28, 2008 at 6:10 PM | Permalink

      Re: Mike Bryant (#554),

      It’s not ‘very misleading’, it compares images from the same source. The only issue is that in images during winter in the early years the sea ice appears to run over the coastline in some places. Since the area can’t increase at that point it’s moot, the only points of interest is the expansion into the Atlantic and Pacific which isn’t affected. The current images in the Compare Daily Sea Ice have accurate coastlines.

  330. Mike Bryant
    Posted Dec 28, 2008 at 7:16 PM | Permalink

    “The only issue is that in images during winter in the early years the sea ice appears to run over the coastline in some places.”

    IT doesn’t appear to, it has been MADE to appear to. And sorry, but that is not the only issue. It is also an issue that in the later years the snow cover has been made to run over the coastline in the other direction.

    If this is not misleading then I might have to start speaking German, since English is clearly not precise enough to discuss anything whatsoever.

    • Phil.
      Posted Dec 28, 2008 at 10:29 PM | Permalink

      Re: Mike Bryant (#556),

      Mike Bryant:
      December 28th, 2008 at 7:16 pm
      “The only issue is that in images during winter in the early years the sea ice appears to run over the coastline in some places.”
      IT doesn’t appear to, it has been MADE to appear to. And sorry, but that is not the only issue. It is also an issue that in the later years the snow cover has been made to run over the coastline in the other direction.
      If this is not misleading then I might have to start speaking German, since English is clearly not precise enough to discuss anything whatsoever.

      No you need to understand what you’re talking about and not make accusations, something that Steve McI does not appreciate on this site!
      The snow accurately matches the coastline mask in every case that I’ve looked at including the one referred to by you in Mike Bryant (#557).

      As I’ve said several times before the ‘decreased size of the Bering Strait’ is the correct configuration and agrees with the coastline as seen throughout the database when there is no sea ice present. The only issue is that in the images without snow in the winter the sea ice encroaches on the coastline. Differentiation between coastline and sea ice is a known difficulty with this type of satellite imaging. E.g. “Land has an emissivity that is near to or above the sea ice [Parkinson, CL and Gloersen, P, 1993], and therefore its presence can produce a similar signal as sea ice does.”

  331. Mike Bryant
    Posted Dec 28, 2008 at 7:41 PM | Permalink

    Also… “the only points of interest is the expansion into the Atlantic and Pacific which isn’t affected.”

    If you go to Cryosphere Today, scroll down to the bottom and click on the “compare side by side” button, you will see a comparison of Dec. 27 1980 and Dec 27 2008.

    The apparent expansion into the Atlantic is definitely affected by the increased size of Greenland in the current depiction since it covers some of the sea ice. The Expansion into the Pacific is also
    affected by the much decreased size of the Bering Strait.

    This comparison product is misleading and inaccurate.

  332. Mike Bryant
    Posted Dec 28, 2008 at 11:09 PM | Permalink

    I am making no accusations, only pointing out shortcomings in a product that purportedly makes comparisons. The comparisons, however, are not good comparisons. No matter how difficult it may or may not be to differentiate between land and sea ice, it must be accomplished before true comparisons can be properly made. If this cannot be corrected, perhaps the best course for CT would be to rename the product. Perhaps they could call it, “Comparisons Between Correct and Incorrect Images.”

  333. Mike Bryant
    Posted Dec 29, 2008 at 5:37 PM | Permalink

    It appears that something is happening to the snow at CT. This is odd. See Christmas Eve compared to Christmas Day:

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=12&fd=24&fy=2008&sm=12&sd=25&sy=2008

  334. Posted Dec 30, 2008 at 9:15 AM | Permalink

    Mike

    think its more to do with CT than the snow. Have their pixels gone on holiday?

    tonyB

  335. AndyW
    Posted Dec 31, 2008 at 12:25 AM | Permalink

    Happy New Year All. Thought I would get that in early :) . Looks like JAXA graph will be going into 2009 at the bottom of the range for the last few years, maybe. I think they will be using black for 2009. My 13.9×10^6 km max is still looking good, but of course it will be hard to tell until Feb.

    The big question is of course what happens if we get a 13.3-13.5 max for the summer?

    Chris, what do you think about our run of cold weather at the mo in the UK? Ok, it is not that cold but it is going on for rather a long time. Those poor old parakeets will be wondering what has happened…

    Regards

    Andy

  336. Chris
    Posted Dec 31, 2008 at 8:36 AM | Permalink

    #562 AndyW – happy new year

    The current freeze in the UK is interesting. I’ve been following the discussions of model output etc on Netweather. As you will probably know, the models start to lose much of their reliability beyond 5 days, so its hard to know what will happen in January. A few days ago, the long range output was starting to show almost a 1963-style set-up….

    http://www.wacv.co.uk/index.php?option=com_content&view=article&id=78:the-winter-of-1962-63-part-1&catid=7:kevinsarea&Itemid=6

    However, they’ve now retreated to a less extreme pattern, still pretty cold for the next few days or so but with a (brief?) milder interlude (preceded by snow?) now more likely after that. And with the northeasterlies, which would have brought the coldest/snowiest weather, now being held at bay for the time being with the high retreating to our SE (rather than west/northwestwards and ridging into an intense Greenland high which is what would be more favourable for extreme cold!)

    http://www.coldswell.com/longrange.php

    (If you keep returning to that link you will see the longer term forecasts can change hugely in the space of as little as a few hours i.e. from run to run)
    http://www.theweatheroutlook.com/twodata/datmrfcompare.aspx – includes 850 HPa Ensemble lines for London etc

    Still, an exceptionally cold (as opposed to moderately cold) January overall still can’t be ruled out. I would be so proud if the UK were able to contribute some home grown sea ice to the Arctic totals LOL
    “…..In January 1963 the sea froze for 1 mile (1.6km) out from shore at Herne Bay. The upper reaches of the River Thames also froze over…..”

    http://en.wikipedia.org/wiki/1963_United_Kingdom_cold_wave

    When you say “our run of cold weather” I actually tend to think of the weather over the last few months in general. For the CET it can now be said with certainty that 2008 will have the coldest second half of the year since 1993. And obviously 2009 will start with some exceptionally cold daily values, so it looks unlikely that Jan at least will buck the trend (significantly)…..

  337. Chris
    Posted Dec 31, 2008 at 9:03 AM | Permalink

    And just a bit of fun with the CET:

    Summer 2008 was the 142nd warmest ever (209th coolest)
    Autumn 2008 was the 114th warmest ever (237th coolest)
    Dec 08 looks like finishing approx 125th warmest ever (226th coolest)

    All depends which way you look at it and spin it, but it could be said that three seasons in a row outside the top 100 warmest would be kind of disappointing for those looking forward to England’s climate accelerating towards its promised Mediterranean state under global warming :)

    http://hadobs.metoffice.com/hadcet/data/download.html

    As for the Arctic ice, I was particularly interested between mid-August and early October when its levels were exceptionally low. But for the other 10 months of this year including the present it’s been within ~normal bounds, at least for extent and area (obviously there’s the lasting legacy of the 2007 record melt in terms of thinness and hence volume). I will only get particularly interested in it again if levels go exceptionally low again in summer 2009 – which of course can’t be ruled out, or if levels in summer 2009 end up notably high (relative to previous years) in which case I would enjoy seeing some of the more extreme fearmongers being forced to do some backtracking……

    • Phil.
      Posted Jan 1, 2009 at 1:54 PM | Permalink

      Re: Chris (#564),

      As for the Arctic ice, I was particularly interested between mid-August and early October when its levels were exceptionally low. But for the other 10 months of this year including the present it’s been within ~normal bounds, at least for extent and area.

      “Within normal bounds”, hardly, according to Arctic Roos it’s not been within one standard deviation of the 79-07 average all year. Currently it’s about -2σ for the second year in a row.

  338. Alan Wilkinson
    Posted Jan 1, 2009 at 1:19 PM | Permalink

    Darn it, missed out on sending some of these this Christmas:

    http://store.theonion.com/merry-second-to-last-christmasbrholiday-card-set-p-151.html

  339. Chris
    Posted Jan 1, 2009 at 4:13 PM | Permalink

    #566 Phil

    In the Arctic, according to CT, area was > ~1 million km2 less than the 1979-2000 average in June, July, August, September and October 2008

    And in the Antarctic, area was > ~1 million km2 more than the 1979-2000 average in March, April, May and June 2008

    (I’m referring to CT because it’s easier to see the 1 million km2 threshold)

    Note of course that the NH was particularly affected by volcanic-induced cooling for several years in both the 1980s and 1990s, but not in the 2000s.

    Currently, Arctic ice area is 0.77 million km2 below the 1979-2000 average, and Antarctic area is 0.43 million km2 above the 1979-2000 average.

    http://arctic.atmos.uiuc.edu/cryosphere/iphone/iphone.currentarea.series.html

    Put another way, Arctic sea ice is covering ~3.2% of Earth’s total ocean area compared to ~3.4% on average at the same date in 1979-2000. and Antarctic sea ice is covering ~1.6% of Earth’s total ocean area compared to ~1.5% at the same date in 1979-2000.
    (Taking total ocean area as 360 million km2 – http://en.wikipedia.org/wiki/Ocean )

    This all seems pretty ~normal to me. One pole is up, the other is down by a bit more. Of course there’s statistical ways to make the situation seem particularly unprecedented. There always are.

    Insofar as it’s not ~normal, of course, the extra depth of the downward variation in the Arctic (beyond effects of +ve AMO etc) could well be explained by increases in greenhouse gas concentrations in recent decades – let’s say, forcing with no feedbacks. But I guess that wouldn’t be scary enough in some quarters……..

    • Phil.
      Posted Jan 1, 2009 at 5:18 PM | Permalink

      Re: Chris (#567),

      This all seems pretty ~normal to me. One pole is up, the other is down by a bit more. Of course there’s statistical ways to make the situation seem particularly unprecedented. There always are.

      Just like one way of dodging when caught in a mis-statement is to change the subject and prevaricate.

      I’ll remind you that what you said was: As for the Arctic ice, I was particularly interested between mid-August and early October when its levels were exceptionally low. But for the other 10 months of this year including the present it’s been within ~normal bounds, at least for extent and area . No mention of the Antarctic nor the other distractions you attempted to drag in!
      Most of this year the Arctic ice has been around -2σ which is a less than 2% shot, not what I regard as ‘within ~normal bounds’.

      • Daryl M
        Posted Jan 2, 2009 at 11:16 PM | Permalink

        Re: Phil. (#568), “Within normal bounds”, hardly, according to Arctic Roos it’s not been within one standard deviation of the 79-07 average all year. Currently it’s about -2σ for the second year in a row.

        I was away for a couple of weeks but I seem to recall the area was almost through the 1979-2007 monthly average and the extent was almost within 1 std of the average around the 2nd week of December. Is my memory failing me or did the graphs get “corrected”? If it’s the latter, what happened?

        • DeWitt Payne
          Posted Jan 3, 2009 at 8:15 AM | Permalink

          Re: Daryl M (#573),

          The graphs at Arctic ROOS were corrected. They were in fact wrong. Watt’s Up had an article on it and there were comments in this thread here for example.

      • Alan Wilkinson
        Posted Jan 3, 2009 at 8:51 PM | Permalink

        Re: Phil. (#568),

        Presumably sea ice extent is highly autocorrelated in which case I don’t believe you can calculate probabilities from individual deviations simply by direct normal distribution statistics.

        • Phil.
          Posted Jan 3, 2009 at 10:46 PM | Permalink

          Re: Alan Wilkinson (#576),

          Do you think there’s much correlation between the extent on any given day and on the same date on the previous year?

        • DeWitt Payne
          Posted Jan 4, 2009 at 9:46 AM | Permalink

          Re: Phil. (#577),

          Yes. There’s every reason to believe that the data are autocorrelated on an annual basis. For example, let’s take the seasonal extent data from Cryosphere today from 1870 to 2008 and calculate the lag 1 correlation coefficient.

          season lag 1
          Autumn 0.835939933
          Summer 0.903185717
          Spring 0.872381651
          Winter 0.706974329
          Annual 0.915153511

          Looks highly autocorrelated to me. Because there is a trend in the data, I suspect even an anomaly would be autocorrelated.

        • BarryW
          Posted Jan 4, 2009 at 11:24 AM | Permalink

          Re: DeWitt Payne (#578),

          Your comment brought another question to mind. What is the cross correlation between say winter and summer? In other words there’s been speculation on the max this year but does it have any bearing on the 2009 min?

        • Phil.
          Posted Jan 4, 2009 at 1:40 PM | Permalink

          Re: BarryW (#579),
          BarryW:
          January 4th, 2009 at 11:24 am
          Re: DeWitt Payne (#578),
          Your comment brought another question to mind. What is the cross correlation between say winter and summer? In other words there’s been speculation on the max this year but does it have any bearing on the 2009 min?

          Not really there’s not even a good correlation between May and September (see below). Last winter was the largest winter extent for several years and yet that was followed by a record loss and the second smallest minimum extent.

          http://www.nsidc.org/news/press/20041607_seaice.html#fig4

          With regard to DeWitt’s comment, by averaging over a 3-month period you see a correlation which isn’t seen when you look at a shorter period.

        • DeWitt Payne
          Posted Jan 4, 2009 at 5:29 PM | Permalink

          Re: Phil. (#580),

          by averaging over a 3-month period you see a correlation which isn’t seen when you look at a shorter period.

          I wasn’t using a three month running average, which does increase autocorrelation, but the same three months averaged each year, which was conveniently tabulated at Cryosphere Today. Using June 15 (or 6/14 in those years when 6/15 wasn’t available) data from NSIDC, which only covers 1979-2006, the lag 1 correlation is still 0.40. Besides, if the data weren’t correlated year to year, why would one bother to calculate an anomaly?

          Re: BarryW (#579),

          I was curious about winter and summer since the models appear to predict much more warming in winter than summer. See for example this presentation from GISS (which is an interesting read as it details a number of problems with regional climate modeling). So I averaged UAH NoPol anomalies for Jan,Feb,Mar and Jul,Aug,Sep. The correlation coefficient within year is 0.48. A more sensitive test, it seems to me, is to look at the difference between winter and summer anomalies, which should be increasing. It isn’t. There is no significant trend in the difference. It’s more than possible that the trend is too small to resolve from the noise and the data doesn’t cover a long enough time span. Still, when a GISS model gives an increase of ~4 C in the difference between summer and winter and a GFDL model shows a difference more than twice that from 2xCO2, you would think there would be something. The CT seasonal extent data does show an increase in the difference in extent between summer and winter over time.

        • BarryW
          Posted Jan 4, 2009 at 8:52 PM | Permalink

          Re: DeWitt Payne (#581),

          So the areal seasonal differences are remaining constant but the extent differences are changing? Interesting.

          It may be that the winter warming would not be enough to reduce the amount of freezing in the near term at a noticeable level. Once the temperatures fall below a critical level shouldn’t the refreezing occur at the same rate? There may also be limitations due to the limited size of the Arctic basin. I would think that the shape of the freezing curve would change first, with the melt starting earlier and the refreeze later.

          On another note, there hasn’t been any updates of the JAXA downloadable extent data this year, but I haven’t seen any explanation.

        • DeWitt Payne
          Posted Jan 5, 2009 at 3:42 PM | Permalink

          Re: BarryW (#582),

          The temperature anomaly seasonal difference is remaining constant, unlike model predictions as I understand them. I haven’t looked at the seasonal area differences, but I expect they will increase just like the extent does.

          It’s not just JAXA. Uni-Hamburg still hasn’t even posted November data and now Cryosphere Today is behind as well, or seems to be. I can’t access the pages for iTouch and iPhone that display digital as well as graphic area data and a lot of image data won’t display either. Arctic ROOS, otoh, is up to date.

  340. Chris
    Posted Jan 1, 2009 at 6:14 PM | Permalink

    #568

    Yes Phil but I didn’t originally define what I meant by “~normal”. You can’t define it for me, then accuse me of changing the subject and prevaricating when I start to elaborate on what I meant, including reference to the Antarctic and “the other distractions”.

    This is the main link I was referring to when making my original statement:

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    Within the spread of 2002-2006 for 10 months of the year. I consider that within ~normal bounds.
    Your -2σ, if it proves anything, is that in ~1993 there would have been a less than 2% shot (if i accept your back-of-the-envelope calculation) of 2008 Arctic ice levels occurring in that year. But who’s to say the middle or early years of a 30-year period are any more normal than the later years? 2002-2006 were a significant chunk of your reference period. And 2008 was clearly ~normal in relation to that 5-year chunk for 10 months of the year!

  341. Mike Bryant
    Posted Jan 1, 2009 at 11:44 PM | Permalink

    What is normal sea ice? If the sea ice is sufficient to moderate the temperatures of the earth, then is that normal? Perhaps the sea ice at present is a little too high since earth temperatures have been level or falling for a few years. Perhaps sea ice levels are a little too high, given the rest of the earth’s climate. I guess we must query Al Gore so that we may be very, very sure of what “normal” really is. If someone on this site could snap his, or her fingers and make the sea ice conform to the average of the last thirty years, would that then be “normal”? Or would that perhaps throw the earth’s temperatures into a downward spiral that would insure the next glaciation? Anyone who believes that they have all the answers to this and many other questions that pertain to climate are being less than truthful, since the science is in it’s infancy. Let’s find the answers without belittling and berating the ones who ask the questions….

  342. DeWitt Payne
    Posted Jan 2, 2009 at 10:30 PM | Permalink

    The JAXA average extent for 2008 was higher than for 2005, 2006 and 2007.
    year average extent(km2)
    2003 10839129
    2004 10657076
    2005 10344909
    2006 10220438
    2007 9965810
    2008 10460621

    Extent for days not reported were replaced by linear interpolation.

    Significant gains and losses are normal as extent and area approach the peak. It’s called weather.

  343. AndyW35
    Posted Jan 3, 2009 at 11:00 AM | Permalink

    I think arguing about figures here on standard deviations is misguided. The last 2 years have been anything other than standard and so any comparison with the past, either the short range JAXA or longer range NSIDC or similar should have a proviso attached.

    This thread is about the JAXA graph in generality and so we should try to keep to it’s general guideline of an interesting short term trend we can all watch for the coming months to see how it fairs. Personally I think that even with the PDO, the ENSO, the GPO and the YMCA we will have something knocking on the door of 2007 come summer 2009. yes I am an alarmist, but don’t tell my wife.

    Phil, don’t want to put a downer on you but for every 5 great posts of info you manage to put on here (and they are appreciated) you do sort of revert to confrontational ism at the merest whim as if you are on one of the more combative weblogs and feel the need to make things right. We should all be down the pub here arguing about things we all have in common rather than soap operas etc.

    Chris, thanks for your weather forecast from previous, we seem now to be in a zone of highs that don’t want to shift and although this will mean cold weather for the next few days the atlantic prevailing depressions are being forced once more along the Greenland-Iceland-Denmark Straight route so my daughters wish for snow is probably not going to happen any time soon.What we need is a moisture rich westerly to be held up over Kent by a Siberian high. And I won’t be able to get to work sadly.

    Regards

    Andy

  344. DeWitt Payne
    Posted Jan 5, 2009 at 7:38 PM | Permalink

    The CT web site is back up, but they are still displaying data that was originally posted on 1/3/09. So unless they skip ahead on the next update, they’re going to be at least seven days behind.

    • Phil.
      Posted Jan 6, 2009 at 3:23 PM | Permalink

      Re: DeWitt Payne (#584),

      DeWitt Payne:
      January 5th, 2009 at 7:38 pm
      The CT web site is back up, but they are still displaying data that was originally posted on 1/3/09. So unless they skip ahead on the next update, they’re going to be at least seven days behind.

      But their SSMI data is up-to-date.

      • DeWitt Payne
        Posted Jan 7, 2009 at 12:25 PM | Permalink

        Re: Phil. (#586),

        The images are up to date (finally), but the data on the iPhone/iTouch pages and all the recent 365 day graphs look like they are further behind than before the end of the year hiatus. The data posted on 1/6 does not appear to show a very big change. It looks like the data that should have been posted on 1/4. I’ll need more data to be sure, but it looks very much to me like the graphs are now 7 days behind.

        • DeWitt Payne
          Posted Jan 7, 2009 at 1:52 PM | Permalink

          Re: DeWitt Payne (#587),

          Today’s update at CT was a big jump. It looks like the two missing days have been skipped over and the data today is for 1/1/2009. I think there may have been a transcription error on the global data, though. The average for today would make much more sense if it were 16.986 rather than 16.896. Either that or I’m still a day off one way or the other.

  345. DeWitt Payne
    Posted Jan 6, 2009 at 12:32 AM | Permalink

    JAXA is up to date again (through 1/5/09 first cut) and the two days of previously missing data in December have been filled in as well in the csv file.

  346. Chris
    Posted Jan 9, 2009 at 7:34 AM | Permalink

    Here’s some food for thought (and a bit of fun.) These graphs are deliberately cherry-picked, but they perhaps illustrate the potential importance of 1998 re: Arctic ice as well as global temperatures?
    I think it’s interesting that it’s even possible to produce recent decadal upward trends in ice in parts of the Arctic, however slight and localised (hope that’s a sufficient health warning….. i know there are bloggers who would have a fit at the sight of these graphs LOL)

    OK here goes…….

    Using ice graph tool: http://ice-glaces.ec.gc.ca/IceGraph/IceGraph-GraphdesGlaces.jsf?id=11874&lang=eng (and using the widest spread of dates it covers)

    These are a couple of areas where the trend is up from 1998-2008:
    Western High Arctic – http://tinyurl.com/9vlgw3
    Hudson Bay – http://tinyurl.com/8xaqfl

    And these are some areas where the trend is up from 1998-2007 (hence is 2008 a confirmation of a downward trend, or could it conceivably be an interruption of an upward one……)
    Beaufort Sea – http://tinyurl.com/84lvjt
    Northwest Passage – http://tinyurl.com/8yxjnt
    Eastern High Arctic – http://tinyurl.com/a3ak5w

  347. mark in austin
    Posted Jan 15, 2009 at 9:51 PM | Permalink

    anyone seen the growth in ice the past 2 days???

    • DeWitt Payne
      Posted Jan 15, 2009 at 10:53 PM | Permalink

      Re: mark in austin (#590),

      Actually, the smoothed extent rate anomaly has been running about 20,000 km2/day above the 2003-2008 average for about a week now. Today looks to be another big day with total extent now greater than 2004. Extent is on track to be about the same or higher than 2008, which was higher than every year in the JAXA record except 2003.

      • Phil.
        Posted Jan 16, 2009 at 12:24 AM | Permalink

        Re: DeWitt Payne (#591),

        Over the last week there’s been a strong drift out of both the Bering and by Svarlbard. In the latter case there’s multi-year ice drifting out so the ice ‘upstream’ of the N Pole is getting ‘younger’. On the Atlantic side it looks stronger than it was last year at this time.

      • BarryW
        Posted Jan 16, 2009 at 9:27 AM | Permalink

        Re: DeWitt Payne (#591),

        The extent change for previous years starts to show some negative values over the next 30 days, so I would expect to see something similar in 2009. If 2009 kept the same rate of change difference with 2003 that it seems to have now it would catch 2003 about the end of the month by eyeballing it.

        What smoothing are you using and could you tell us what the smoothed value of the 2003-8 average is for the next couple of weeks? Thanks.

        • DeWitt Payne
          Posted Jan 16, 2009 at 11:05 AM | Permalink

          Re: BarryW (#593),

          I use exponential smoothing (Exponentially Weighted Moving Average) with an alpha of 0.1. Here is my current plot of the smoothed average rate as well as the data for 2008 and 2009.

          Through the end of January the daily smoothed average (2003-2008) rates in tabular form starting on 1/16:

          36712.76948 km2/day
          38841.49253
          36776.10328
          36517.23295
          34865.52966
          33344.55669
          31982.00102
          32261.92092
          31195.10883
          29034.97795
          29875.22015
          32009.57814
          32605.50032
          31366.81029
          30145.76926
          29815.57233

          I’m still waiting for Uni-Hamburg to post the rest of their 2008 data.

        • BarryW
          Posted Jan 16, 2009 at 4:31 PM | Permalink

          Re: DeWitt Payne (#594),

          Thanks, that’s helping me with the picture.

  348. Ernie
    Posted Jan 21, 2009 at 6:16 PM | Permalink

    Nature has just published a report that claims Antarctica has warmed by 0.1C/decade over the last 50 years. The authors are Eric J. Steig1, David P. Schneider, Scott D. Rutherford, Michael E. Mann, Josefino C. Comiso & Drew T. Shindell

    http://www.nature.com/nature/journal/v457/n7228/full/nature07669.html

    I don’t have an account with Nature so I can’t see the whole article, but I thought it was worth a mention here for those that follow the southern ice as well.

    -Ernie.

    Assessments of Antarctic temperature change have emphasized the contrast between strong warming of the Antarctic Peninsula and slight cooling of the Antarctic continental interior in recent decades1. This pattern of temperature change has been attributed to the increased strength of the circumpolar westerlies, largely in response to changes in stratospheric ozone2. This picture, however, is substantially incomplete owing to the sparseness and short duration of the observations. Here we show that significant warming extends well beyond the Antarctic Peninsula to cover most of West Antarctica, an area of warming much larger than previously reported. West Antarctic warming exceeds 0.1 °C per decade over the past 50 years, and is strongest in winter and spring. Although this is partly offset by autumn cooling in East Antarctica, the continent-wide average near-surface temperature trend is positive. Simulations using a general circulation model reproduce the essential features of the spatial pattern and the long-term trend, and we suggest that neither can be attributed directly to increases in the strength of the westerlies. Instead, regional changes in atmospheric circulation and associated changes in sea surface temperature and sea ice are required to explain the enhanced warming in West Antarctica

  349. mark in austin
    Posted Jan 27, 2009 at 3:24 PM | Permalink

    any thoughts on why the ice has been decreasing over the last week? strange considering how utterly cold it has been…..then again i noticed that in other years there was a downtick before a huge uptick that followed….

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    • DeWitt Payne
      Posted Jan 27, 2009 at 4:04 PM | Permalink

      Re: mark in austin (#597),

      As the average rate of change approaches zero, the normal fluctuations in rate become more obvious. OTOH, the six year average data shows higher variance about now. It increases beginning about the 15th, peaks and then comes back to about average by the 30th. The recent big drop in extent has reduced the smoothed rate to below average from a high nearly exceeding the 3 sigma level of the previous 6 years. Fluctuations in ocean currents that can melt a lot of ice are not always reflected in air temperature. There could also be some compaction going on. I’ll be curious to see how the concentration changes as the CT area data corresponding to the recent drop in extent comes in.

      Uni-Hamburg finally posted the rest of the 2008 data last week. I need to do a year end wrap up. It would be nice if CT also updated their extent charts to the end of 2008, but I’m not holding my breath for that one.

  350. Gerald Machnee
    Posted Jan 27, 2009 at 7:48 PM | Permalink

    Whether its warming or cooling, the ice has surrounded three ships including an ice breaker and a small cruise ship in the St. Lawrence River.

  351. mark in austin
    Posted Jan 31, 2009 at 7:23 AM | Permalink

    i am hoping that this year passes 2003 in ice extent and area….that would be quite a feat coming off the low point of just 2 summers ago….of course, it will be fun to see if this summer’s ice can stay ahead of 2008’s…

  352. Posted Jan 31, 2009 at 2:43 PM | Permalink

    Is it time for a new thread? Sea Ice – Getting Ready for Spring Training.

    • DeWitt Payne
      Posted Feb 1, 2009 at 9:34 PM | Permalink

      Re: Robert (#601),

      Right now it looks like 2003 is out of reach, but 2004 looks like a possibility. With one month’s data, the year-to-date average is ahead of 2005 through 2008 and just behind 2004.

      Has anyone been able to get the charts or anything else at Arctic ROOS to load? All I get is a “waiting for” message.

      • Daryl M
        Posted Feb 1, 2009 at 11:39 PM | Permalink

        Re: DeWitt Payne (#611) It’s loading for me. Based on JAXA, extent is almost equal to 2008 but still below 2003. The graph lines are too thick to resolve exactly.

        • BarryW
          Posted Feb 2, 2009 at 7:23 AM | Permalink

          Re: Daryl M (#612),

          Here’s the most recent JAXA they may change it later today.

          yr extent diff
          2003 14.08891 0.066718
          2004 13.48406 -0.019375
          2005 13.24641 -0.004532
          2006 13.13266 0.079375
          2007 13.42766 -0.014375
          2008 13.66656 -0.018593
          2009 13.69375 0.023125

        • DeWitt Payne
          Posted Feb 2, 2009 at 11:07 PM | Permalink

          Re: Daryl M (#612),

          Well file this under computer oddities. I went to dslreports.com and ran their tweak test on my TCP/IP stack after some hours researching the problem. After setting MTU to 1500 as recommended and turning on a couple of options that seemed like they might be useful, I can now access arctic-roos.org. Weird. Note that the tweak test is for XP and older OS’s. Not recommended for Vista.

  353. Daryl M
    Posted Jan 31, 2009 at 8:13 PM | Permalink

    One aspect of the sea ice issue that has been mentioned from time to time is an apparent visual discrepancy between the shore line in current imagery vs. in older imagery on The Cryosphere Today. I’m looking at a comparision of today’s northern hemisphere with the same date in 1979. Here is a link. There are significant differences in the shoreline which appears to indicate reduced area for ice in 2009 as opposed to 1979. If this discrepancy is has an effect on the computed ice area and extent, then there is a built-in anomoly that precludes ice area and extent from ever reaching levels reported in the past. I would be interested to hear a more rigorous discussion on this matter.

  354. Mike Bryant
    Posted Feb 1, 2009 at 12:58 AM | Permalink

    Daryl,
    This is rather odder than usual… I wonder why they are not showing the snow cover for January 31, 2009… I guess it will be updated tomorrow… This comparison is a joke… I wish someone could get to the bottom of this ridiculous “comparison product” which actually compares nothing at all worthwhile…
    Mike Bryant

    PS Apparently the will does not exist within the author of this site to make things right…

    • Daryl M
      Posted Feb 1, 2009 at 1:35 AM | Permalink

      Re: Mike Bryant (#603), I agree that its more strange than usual due to the missing snow cover, but even when the snow cover is depicted, the recent image clearly shows snow in areas which show up as ice in the earlier image. If they are doing a pixel count, then there would seem to be a blatent inherent bias that would result in less area for ice now than historically. Of course, such a bias would lend itself to claims that polar ice is at historic lows and would also explain why it cannot reach historic levels even in an unusually cold winter such as this winter and last winter.

  355. Mike Bryant
    Posted Feb 1, 2009 at 10:38 AM | Permalink

    Daryl,
    According to Phil. the recent depictions of the coastline are absolutely correct, with or without snow. He has also stated that the older depictions DO show sea ice in areas that are actually land. He said that the reason for this is that the satellites often have trouble distinguishing between land and sea ice. If that is true, then your supposition sounds correct, that the older data would include those areas as part of the record. I have not been able to get a definitive answer on this question, In fact I received no answer whatsoever. See comment #541 where I requested information from CT… No answer.

    • Daryl M
      Posted Feb 1, 2009 at 2:20 PM | Permalink

      Re: Mike Bryant (#605) Mike, I take whatever Phil says with a rather large sack of salt, but irrespective of that, the coastlines have been well known for years and it therefore should not be difficult for someone with the right software to calculate ice area and extent based on known shore lines. At least that would make a comparision of today’s imagery with that of earlier years “apples to apples” rather than what it is today, which is biased in favour of the AGW position. Considering the budgets available for producing this data, I find it rather surprising that this has not been done already. The skeptic in me thinks that if the bias was in the other direction, like in so many other situations, there would have been sufficient motivation to secure whatever funding was required to correct this deficiency.

  356. Mike Bryant
    Posted Feb 1, 2009 at 10:42 AM | Permalink

    Daryl,
    See this overlay:

    It was done by Steve Keohane.
    Mike

    • Daryl M
      Posted Feb 1, 2009 at 2:31 PM | Permalink

      Re: Mike Bryant (#606) Mike, Thanks for pointing that out. I re-read that portion of the thread. The image makes it very clear that comparing the historic ice area and extent with current data is rather dubious. I am very curious what the total area of the erroneous regions is.

  357. Richard Sharpe
    Posted Feb 1, 2009 at 2:39 PM | Permalink

    If you look at the latest comparison at Cryosphere it seems they decided to eliminate the snow cover.

  358. BarryW
    Posted Feb 7, 2009 at 10:30 PM | Permalink

    Thought I’d try to provide the JAXA data on a weekly basis. The data is missing for today however so it’s not totally up to date (used previous day’s value for extent). I’m also experimenting with some R functions for html table formatting

    year extent weekly change
    200314.05828 0.12734
    200413.75344 0.24172
    200513.23469-0.06719
    200613.35063 0.38750
    200713.44891 0.01891
    200814.04625 0.31688
    200913.72703 0.11719

  359. BarryW
    Posted Feb 10, 2009 at 1:56 PM | Permalink

    For those who are interested about the Arctic extent: 2009 is running above the recent average and is just behind 2003 and very close to 2008.

    2003 14.16203
    2008 14.06438
    2009 14.07609

  360. Robert
    Posted Feb 10, 2009 at 3:00 PM | Permalink

    Thanks, I’m still checking in

  361. Jim P
    Posted Feb 12, 2009 at 1:16 PM | Permalink

    #616 Yes – I’m interested. I continually follow ice via:

    Today it looks like just over 13 million km^2, so why are your values higher?

  362. BarryW
    Posted Feb 12, 2009 at 2:50 PM | Permalink

    I’m reporting extent instead of area. Extent generally seems to be based on any pixel where there is at least 15% coverage. Haven’t found a numerical source of historical area data that is kept up daily. U. of Hamburg has a monthly update after the fact for area I believe. Dewitt Payne has been following that. Extent data is from JAXA

  363. BarryW
    Posted Feb 13, 2009 at 11:11 AM | Permalink

    Here is the Arctic sea ice extent for this week based on JAXA.

    yearsextentweekly change
    200314.185160.12688
    200414.121410.36797
    200513.442970.20828
    200613.553440.20281
    200713.709370.26047
    200814.137340.09109
    200913.952810.18781

  364. BarryW
    Posted Feb 13, 2009 at 11:13 AM | Permalink

    Hmm, the table looked fine in the preview.

  365. BarryW
    Posted Feb 13, 2009 at 11:54 AM | Permalink

    Trying again

    years extent weekly change
    2003 14.18516 0.12688
    2004 14.12141 0.36797
    2005 13.44297 0.20828
    2006 13.55344 0.20281
    2007 13.70937 0.26047
    2008 14.13734 0.09109
    2009 13.95281 0.18781

  366. aurbo
    Posted Feb 15, 2009 at 8:29 PM | Permalink

    As of Feb 14th, JAXA has the ice extent over 14M k² at 14,023,438 km².

  367. DeWitt Payne
    Posted Feb 15, 2009 at 10:28 PM | Permalink

    Extent goes up and down a lot at this time of year. There was a big drop in extent in 2008 at about this same time. I’ve been making lots of graphs lately. Here’s a graph of exponentially smoothed Arctic area and extent rate anomalies. The area anomaly is calculated from the Cryosphere Today average data which is from the 1979 to 2000 average. The JAXA extent data is calculated from a 2003 to 2008 average. The UL and LL lines are really estimated error bars on the 2003 to 2008 data so take them with a large grain of salt.

    As I’ve pointed out many times before, the area data from Cryosphere Today is 5 days behind the JAXA data. The graph includes the preliminary JAXA data from February 15 which puts JAXA six days ahead for the moment.

  368. aurbo
    Posted Feb 15, 2009 at 10:43 PM | Permalink

    It seems this season that there have been a number of instances where the daily or weekly changes in ice cover differ markedly between NSIDC and Cryosphere Today. In almmost every case, the NSIDC numbers have slowed their rate of increase, or even reversed their rate of increase (as is the case today) while the CT analysis continues to show an increase. This is particularly noticeable in this current week’s numbers. Can someone explain what’s goiung on? In the meantime I’ll rely on the JAXA numbers which appear to be more consistent.

    There are a lot of unanswered questions re the daily presentations of Arctic ice according to NSIDC and the differences between their accounting and that at Cryosphere Today. There are too many ways to fudge the data in order to manage the results. With just a few changes the sponsors of the data could close down some of these avenues to manipulation. First, both systems use the measurement of total ice cover in km² to provide their principal comparative year-to-year data points. What they fail to provide us is the total area of the particular basins they are looking at. There is some evidence that the representation of the coastlines that bound these areas have have changed over the years making it impossible to meaningfully relate total area covered between one year and the next. Some basins have a finite size and as a result can never be more than 100% covered. Other basins have unrestricted boundaries…usually to their south…which can be covered in some years and not in others. In these areas total ice area would be more meaningful since there would be no way of knowing if there is an arbitraily derived number that represents 100% of such basins. If the basin areas are fixed in time, then a percentage of total coverage would add more useful information than simply using the total ice-covered area of an ill-defined basin size. The current system makes year-to-year comparisons much too amenable to subjective shenanigans.

  369. Flanagan
    Posted Feb 16, 2009 at 10:37 AM | Permalink

    Yes, the NSIDC arctic sea ice extent literally plunged down recently

    Can this be seen as a realistic measure?
    And what about the antarctic?

  370. Flanagan
    Posted Feb 16, 2009 at 11:12 AM | Permalink

    Well it’s kind of going down also, though the anomaly is certainly not as large as that

    Could it be a parametrization problem?

    • DeWitt Payne
      Posted Feb 16, 2009 at 11:40 AM | Permalink

      Re: Flanagan (#628),

      The NSIDC plot doesn’t look much like Arctic ROOS or JAXA. Where is the big dip in extent that was observed in 2008 in mid February? Nor has the extent in 2009 in any other data source that I know of dropped as far below 2008 as indicated in the NSIDC data. In fact, today’s JAXA extent is actually slightly higher than for the same day in 2008.

      • Mike Bryant
        Posted Feb 19, 2009 at 12:45 PM | Permalink

        Re: DeWitt Payne (#629),

        DeWitt Payne,
        I am very interested in your take on the recent problems at NSIDC. Is it possible that the corrections at Nansen were an attempt by them to line up with NSIDC? You seem to be the person that has studied the reporting agencies the most and I am sure that your comments would be the most helpful.
        Thanks,
        Mike Bryant

        • Phil.
          Posted Feb 19, 2009 at 2:01 PM | Permalink

          Re: Mike Bryant (#640),

          Looks like a much better match with JAXA:

          While they use the same sat. data as NSIDC they don’t use the same algorithm.

  371. Mike Bryant
    Posted Feb 16, 2009 at 10:02 PM | Permalink

    Arctic land mask information:

    http://polar.ncep.noaa.gov/seaice/support/land.html

    From the article:

    “The potential effects of different land definition policies are not trivial. Approximately 1.6 million km^2 can be moved into or out of the ‘land’ category, depending on how ‘land’ is defined. Much of this occurs in areas with complex coastlines, such as the Baltic Sea or the Canadian Archipelago. Since both can include an ice cover, this represents a significant fraction of the total Arctic ice pack (which reaches a maximum extent of approximately 14 million km^2).”

    Also:

    “In order to stabilize time series which may be derived from the NCEP ice products, we have recomputed the land masks all at the same time, and then recomputed the products. The hemispheric products have not included (in their data files) land, but the global fields do include land as part of the processing. Since it is also the global products which are most used for time-series purposes, this does require revisitation by those users.”

    I wonder if CT will use these masks to correct their Comparison Product and their other presentations.

    • Phil.
      Posted Feb 17, 2009 at 1:40 AM | Permalink

      Re: Mike Bryant (#630),

      I wonder if CT will use these masks to correct their Comparison Product and their other presentations.

      A bit late, they’ve been doing so for year’s!

  372. Mike Bryant
    Posted Feb 17, 2009 at 7:24 AM | Permalink

    Not in their Comparison Product, they haven’t…

    http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=12&fd=20&fy=1980&sm=12&sd=22&sy=2008

    Here is the overlay which definitely shows shortcomings in the sea ice mask:

  373. Fred Harwood
    Posted Feb 18, 2009 at 1:38 PM | Permalink

    http://nsidc.org/arcticseaicenews/

  374. brendy
    Posted Feb 18, 2009 at 4:05 PM | Permalink

    Re the link in #633 above, the NSIDC had this announcement today.

    “As some of our readers have already noticed, there was a significant problem with the daily sea ice data images on February 16. The problem arose from a malfunction of the satellite sensor we use for our daily sea ice products. Upon further investigation, we discovered that starting around early January, an error known as sensor drift caused a slowly growing underestimation of Arctic sea ice extent. The underestimation reached approximately 500,000 square kilometers (193,000 square miles) by mid-February. Sensor drift, although infrequent, does occasionally occur and it is one of the things that we account for during quality control measures prior to archiving the data. See below for more details.”

  375. brendy
    Posted Feb 18, 2009 at 5:18 PM | Permalink

    NSIDC Part II – Do you get my drift?

    An excerpt from http://nsidc.org/arcticseaicenews/

    Data error sources

    As discussed, near-real-time products do not undergo the same level of quality control as the final archived products, which are used in scientific research published in peer-reviewed journals. However, the SSM/I sensors have proven themselves to be generally quite stable. Thus, it is reasonable to use the near-real-time products for displaying evolving ice conditions, with the caveat that errors may nevertheless occur. Sometimes errors are dramatic and obvious. Other errors, such as the recent sensor drift, may be subtler and not immediately apparent. We caution users of the near-real-time products that any conclusions from such data must be preliminary. We believe that the potential problems are outweighed by the scientific value of providing timely assessments of current Arctic sea ice conditions, as long as they are presented with appropriate caveats, which we try to do.

    For several years, we used the SSM/I sensor on the DMSP F13 satellite. Last year, F13 started showing large amounts of missing data. The sensor was almost 13 years old, and no longer provided complete daily data to allow us to track total daily sea ice extent. As a result, we switched to the DMSP F15 sensor for our near-real-time analysis. For more information on the switch, see “Note on satellite update and intercalibration,” in our June 3, 2008 post.

    On February 16, 2009, as emails came in from puzzled readers, it became clear that there was a significant problem—sea-ice-covered regions were showing up as open ocean. The problem stemmed from a failure of the sea ice algorithm caused by degradation of one of the DMSP F15 sensor channels. Upon further investigation, we found that data quality had begun to degrade over the month preceding the catastrophic failure. As a result, our processes underestimated total sea ice extent for the affected period. Based on comparisons with sea ice extent derived from the NASA Earth Observing System Advanced Microwave Scanning Radiometer (EOS AMSR-E) sensor, this underestimation grew from a negligible amount in early January to about 500,000 square kilometers (193,000 square miles) by mid-February (Figure 2). While dramatic, the underestimated values were not outside of expected variability until Monday, February 16. Although we believe that data prior to early January are reliable, we will conduct a full quality check in the coming days.

    Sensor drift is a perfect but unfortunate example of the problems encountered in near-real-time analysis. We stress, however, that this error in no way changes the scientific conclusions about the long-term decline of Arctic sea ice, which is based on the the consistent, quality-controlled data archive discussed above.

    We are actively investigating how to address the problem. Since we are not receiving good DMSP SSM/I data at the present time, we have temporarily discontinued daily updates. We will restart the data stream as soon as possible.

    Some people might ask why we don’t simply switch to the EOS AMSR-E sensor. AMSR-E is a newer and more accurate passive microwave sensor. However, we do not use AMSR-E data in our analysis because it is not consistent with our historical data. Thus, while AMSR-E gives us greater accuracy and more confidence on current sea ice conditions, it actually provides less accuracy on the long-term changes over the past thirty years. There is a balance between being as accurate as possible at any given moment and being as consistent as possible through long time periods. Our main scientific focus is on the long-term changes in Arctic sea ice. With that in mind, we have chosen to continue using the SSM/I sensor, which provides the longest record of Arctic sea ice extent.

    • Daryl M
      Posted Feb 18, 2009 at 7:03 PM | Permalink

      Re: brendy (#635) What I’m wondering is does the degradation of the sensor have anything to do with the abrupt levelling off of the NSIDC sea ice extent graph in mid-January?

  376. Mike Bryant
    Posted Feb 19, 2009 at 4:54 PM | Permalink

    Thank you Phil. still like to see DeWitt chime in…

    • DeWitt Payne
      Posted Feb 20, 2009 at 4:52 PM | Permalink

      Re: Mike Bryant (#642),

      I think Nansen just made some sort of formatting error plotting the data and took a long time to correct it. The wiggles in the graph didn’t change much if at all, just the scale. It only affected the data starting sometime after the Arctic ice minimum and certainly before the degradation in the sensor was discovered. The odd thing was that they corrected the upper extent and area plots where the current data are compared to the two previous years quite a while before they corrected the lower two plots where the data are compared to the previous year and a long term average and variance.

      By and large, it’s possible to convert data from one source to that of another by a linear transformation. You can take the JAXA extent plot or the Uni-Hamburg area data overlay it on the Nansen plot, match the horizontal axis and stretch or squeeze the vertical axis and offset until the lines overlap for almost all of the data. The Nansen plots have a higher minimum, but the rest of the plot matches quite well. It’s also possible to regress Uni-Hamburg area data with Cryosphere Today data or the Uni-Hamburg vs. JAXA extent and get an excellent fit using a linear regression and a horizontal offset of five days for the Cryosphere Today Arctic data. Even the Cryosphere Today 1979 to 2000 average is very close in shape to the Uni-Hamburg 2003 to 2008 average, which isn’t really all that suprising. There is, of course, an offset because the means are different by ~1 Mm2.

      The only real oddity is the unusual shift in the Cryosphere Today seasonal extent averages for 2007 and 2008 compared to any other source I’ve been able to find. I’ll have more on that eventually.

  377. thefordprefect
    Posted Feb 19, 2009 at 6:22 PM | Permalink

    Just posted this over at WUWT
    Please have a look at this comparison between AMSRE (JAXA site) and SSM/I (NSIDC) This compares monthly averages back to 2002. In general the AMSRE data is smaller than the NSIDC and the only significant time it is larger is in Sept 2008 when it is 170,000 sq km larger. A difference of 3.5% larger compared to AMSRE area at that time.

    There seems to be no significant drift visible in earlier data. There is an obvious difference in the algorithms used to genereate the 15% sea ice areas.

    There are many warnings on all pages that reference the real time data e.g.
    “Daily images are also available. Monthly averages are considered more accurate indicators of overall trends. Please read Image Derivation and Interpretation Resources to understand the uses and limitations of these figures”

    To see a truly amazing video of sea ice download this AVI showing the whole polar cap in motion (Very big file but worth it!)

    http://polynya.gsfc.nasa.gov/animations/amsr/amsr_seaice_north_89v_hdtv.avi

    Mike

    • John F. Pittman
      Posted Feb 20, 2009 at 9:22 AM | Permalink

      Re: thefordprefect (#643), Thanks, Mike. I agree. I have dled major patches quicker than this. I note that the wind/current effect is visible, especially in the wake of certain islands. Most impressive visual work. I hope a paper with the corresonding science was done. If so, bet it was, with the video, an excellent piece.

    • Earle Williams
      Posted Feb 20, 2009 at 4:34 PM | Permalink

      Re: thefordprefect (#643),

      Is the image you’ve posted inline one of your creation? If so could you show the difference between SSMI and AMSRE over that time period?

      Thanks!

      • thefordprefect
        Posted Feb 20, 2009 at 7:34 PM | Permalink

        Re: Earle Williams (#655), Yes the plot was mine and was generated from monthly figures (all that I could find for SSM/I). The data for the (NSIDC)AMSRE (JAXA site) was daily data averaged over the month. The NSIDC data was only available up to end of December 2008 and I would assume it will now be a time before January’s average is produced!

        The graph at the bottom of the extent curves is the diference between sea ice extent SSM/I and AMSRE – scale is on right side of plot. It shows large discrepancies for most of the year (SSM/I is greater than ASMRE by about 800,000 km^2) but is just about the same at sea ice minima.
        Mike

  378. BarryW
    Posted Feb 20, 2009 at 1:12 PM | Permalink

    I keep trying to find a way to format these in a nice table, but no luck. Heres this weeks JAXA arctic extent

    years extent weekly change
    2003 14.55281 0.307813
    2004 14.28484 0.154063
    2005 13.45453 0.010781
    2006 13.43641 -0.116407
    2007 13.78641 0.064375
    2008 14.08078 -0.043750
    2009 14.09531 0.132500

    • Phil.
      Posted Feb 20, 2009 at 2:31 PM | Permalink

      Re: BarryW (#646),

      Have you tried setting up a table using html table tags?

      http://www.w3schools.com/tags/tag_table.asp

    • NeedleFactory
      Posted Feb 20, 2009 at 2:50 PM | Permalink

      Re: BarryW (#646) said:

      I keep trying to find a way to format these in a nice table

      If you apply the “Code” tag (one of the “Quicktags” visible when you post) to your table or other text, the font will be monospaced, e.g.:
      years extent weekly change
      2003 14.55281 0.307813
      2004 14.28484 0.154063
      2005 13.45453 0.010781
      2006 13.43641 -0.116407
      2007 13.78641 0.064375
      2008 14.08078 -0.043750
      2009 14.09531 0.132500

      • NeedleFactory
        Posted Feb 20, 2009 at 2:53 PM | Permalink

        Re: NeedleFactory (#649)
        Whoops Barry, I see what you mean!
        In the preview of my post above, the table was monospaced,
        but the font got lost upon pressing “Submit Comment”.
        Apologies to all…

  379. UK John
    Posted Feb 20, 2009 at 2:12 PM | Permalink

    UIUC Cyrosphere Today states in response to a Daily Tech article of Jan 1st 09.

    Global climate model
    projections suggest that the
    most significant response of the
    cryosphere to increasing
    atmospheric greenhouse gas
    concentrations will be seen in
    Northern Hemisphere summer
    sea ice extent. Recent
    decreases of N. Hemisphere
    summer sea ice extent (green
    line at right) are consistent with
    such projections.

    intuitively I would have thought the winter sea ice extent would be a better guide as that is free from solar influence.

    UOI don’t give any reference to a model or any explanation, has anyone any idea.

    • Dave Dardinger
      Posted Feb 20, 2009 at 3:00 PM | Permalink

      Re: UK John (#647),

      intuitively I would have thought the winter sea ice extent would be a better guide as that is free from solar influence.

      AGW is supposed to happen because solar input (converted to IR) is trapped by GHGs in the atmosphere. Without solar input, there’s no greenhouse effect to start with. That’s an oversimplification, of course, but basically winter isn’t the season to look for AGW in the arctic.

  380. Phil.
    Posted Feb 20, 2009 at 3:25 PM | Permalink

    Here’s an example using the table tag, I hope it works.

    years
    extents
    weekly change

    2003
    14.55281
    0.307813

    2004
    14.28484
    0.154063

  381. BarryW
    Posted Feb 20, 2009 at 5:04 PM | Permalink

    To those who offered advice on the table problem. Thanks for trying.

    I’ve tried a number of tricks including the Code tag and even tried Latex. The preview works fine but it gets trashed when it is committed. Code at least leaves a space, but even then the commit compresses spaces so you can’t spread things out.

  382. Steve McIntyre
    Posted Feb 20, 2009 at 5:11 PM | Permalink

    TRy:
    maketable=function(X,align=-999,widths=-999){
    N=nrow(X);M=ncol(X); out= “d:/temp/temp.txt”
    if (align[1]== -999) right=rep(TRUE,M) else right=align
    code=rep(“”,M);code[right]=” ”
    if(widths[1]== -999) pct= floor(100/M) else pct=widths

    write.table(“”,file=out,col.names=FALSE,row.names=FALSE,quote=FALSE)
    for (j in 1:N){
    write.table(“”,append=TRUE,file=out,row.names=FALSE,col.names=FALSE,quote=FALSE)
    for(i in 1:M){
    write.table( paste(“”,code[i],X[j,i],””,sep=””),append=TRUE,file=out,row.names=FALSE,col.names=FALSE,quote=FALSE)
    }
    write.table(“”,append=TRUE,file=out,row.names=FALSE,col.names=FALSE,quote=FALSE)
    }
    write.table(“”, append=TRUE,file=out,row.names=FALSE,col.names=FALSE,quote=FALSE)
    }

  383. Mike Bryant
    Posted Feb 20, 2009 at 10:39 PM | Permalink

    “The only real oddity is the unusual shift in the Cryosphere Today seasonal extent averages for 2007 and 2008 compared to any other source I’ve been able to find. I’ll have more on that eventually.”
    -DeWitt Payne

    Looking forward to your post.

  384. SidViscous
    Posted Feb 20, 2009 at 11:39 PM | Permalink

    The 2007-2008 gap was then just over 500,000 km

    Funny that number. It has turned up elsewhere.

    Arctic Sea Ice Underestimated for Weeks Due to Faulty Sensor

    A glitch in satellite sensors caused scientists to underestimate the extent of Arctic sea ice by 500,000 square kilometers

  385. SidViscous
    Posted Feb 20, 2009 at 11:42 PM | Permalink

    Oh sorry. I forgot to mention that in that article, yet again sensors that have a bias error showing more warming than there actually was, again, do not effect the results in any way.

  386. AndyW
    Posted Feb 26, 2009 at 1:31 AM | Permalink

    Always pays to shop around.

    Looks like 2008/2009 is going to split the previous years on JAXA and be “average”. My guess of 13.9Mm was a bit low. Wonder how a possible lower winter extent will effect this summers final result. Lowest year ever and can indeed paddle to the pole? :D

    Regards

    Andy

  387. BarryW
    Posted Feb 27, 2009 at 9:32 AM | Permalink

    Here is the JAXA numbers as of Feb 27. It looks like the max will occur around the second week in March based on the previous 2003-08 dates. Last year was about the 9th. Hopefully the table will display like it appears in the preview.

    &nbsp
    year           extent       weekly change
    2003         14.64516             0.07141
    2004         14.27281             0.00422
    2005         13.88875             0.39969
    2006         13.47969             0.03797
    2007         13.84297            -0.01844
    2008         14.41547             0.29531
    2009         14.17031             0.10984

  388. BarryW
    Posted Feb 27, 2009 at 9:54 AM | Permalink

    Just for reference here are the max extent values from JAXA. Mean is 14.25802.

                  maxice
    2003       14.844063
    2004       14.360313
    2005       14.098906
    2006       13.782344
    2007       13.945625
    2008       14.516875

  389. brendy
    Posted Feb 27, 2009 at 10:37 AM | Permalink

    NSIDC is back up with revised real time numbers and images. http://nsidc.org/arcticseaicenews/ Despite tracking higher than the figures posted earlier in February, they show only modest marginal increases from those figures.

    The explanation of the new data source, however, is fascinating – essentially “we are going back to the discredited F13 sensor that we abandoned a year ago as a questionable data source because, given the circumstances, even though it wasn’t all that reliable, we don’t think it has gotten worse.” Here’s the official line.

    “Before acquiring data from F15, NSIDC obtained sea ice data from the SSM/I sensors on the DMSP F8, F11, and F13 satellites. In March 2008, we switched from the F13 satellite to the F15 satellite because the F13 SSM/I sensor had started to have regular areas of missing data. The missing data were caused by malfunctioning data recorders on the satellite, not because of any problem with the sensor itself. At the time, we were concerned that the recorder problem would become more serious. However, the F13 data recorders have not degraded further.

    In comparison, the F15 sensor drift makes those data unreliable for our purposes, so we have switched back to F13.”

  390. AndyW35
    Posted Feb 28, 2009 at 8:36 AM | Permalink

    Hmm, making the best of a bad lot, springs to mind. I think I will give it a miss. I still like their comments though, I know they are not everyone’s cup of tea though.

    2009 is a spiky little feller in Feb isn’t he?

    Regards

    Andy

  391. mark in austin
    Posted Feb 28, 2009 at 11:06 AM | Permalink

    we have come a long ways from our 2007 summer….i think this summer’s low will be higher than last summer’s. will be fun to watch….

  392. Jim P
    Posted Mar 1, 2009 at 4:20 AM | Permalink

    Another projection of the summer ice end game:

    “Arctic ice modeller Professor Wieslaw Maslowski, a science adviser to the survey, hopes the data will enable him to refine his forecast of when the first ice-free summer might arrive.

    Currently, he has it down for 2013 – but with an uncertainty range between 2010 and 2016. ”

    http://news.bbc.co.uk/2/hi/science/nature/7917266.stm

    Steve will you start a new thread once we hit the 2009 max? Should be some time in the next two week?

    • Gerald Machnee
      Posted Mar 1, 2009 at 8:28 AM | Permalink

      Re: Jim P (#672),

      Renowned Arctic explorer Pen Hadow and two companions were dropped onto the ice by plane some 1,700km (670 miles) north of Canada on Saturday night.

      So where in cold hell is Canada? Were they dropped in Siberia??

  393. JLamsam
    Posted Mar 1, 2009 at 9:44 AM | Permalink

    Quite a bit of press recently on the Catlin Arctic Survey group (link)
    Whatever data they manage to gather still provides only a small part of the big picture. They should repeat this trip every 2 years for the next 50 years and give us a dynamic picture of the arctic sea ice extent.

  394. BarryW
    Posted Mar 2, 2009 at 9:20 AM | Permalink

    for you horseracing fans, 2009 has just moved into third in the extent race to the top so it has a chance to finish “in the money”.

    2004 &nbsp &nbsp 14.31500
    2009 &nbsp &nbsp 14.37203

  395. mark in austin
    Posted Mar 3, 2009 at 4:15 PM | Permalink

    i saw that barry….only 100,000 K’s from coming in 2nd place…i would guess that 2003 is a VERY longshot, but 2008’s recent record may be in grasp! of course, it would be fun to half the distance to 2003 and it is POSSIBLE as some years have a very late finish (as in march 21st or so)….

    other thoughts?

  396. mark in austin
    Posted Mar 3, 2009 at 4:15 PM | Permalink

    i saw that barry….only 100,000 K’s from coming in 2nd place…i would guess that 2003 is a VERY longshot, but 2008’s recent record may be in grasp! of course, it would be fun to half the distance to 2003 and it is POSSIBLE as some years have a very late finish (as in march 21st or so)….

    other thoughts?

  397. AndyW
    Posted Mar 4, 2009 at 11:58 PM | Permalink

    Steven Goddard on WUWT tried to claim that if the recent NDISC graph followed for another week we would be back at 1979-2000 levels so I said no way that was going to happen and to put a small wager on, but sadly he has not taken it up. Of course the last two JAXA days are on a decreasing trend so I expect that the NSIDC graph, which takes a 5 day average, will soon slope off and I will be theoretically victorious! Maybe.
    :D

    Regards

    Andy

  398. BarryW
    Posted Mar 6, 2009 at 11:20 AM | Permalink

    JAXA extent for March 6. Siting in second with a guaranteed third place! Only 2008 and 2003 are higher maxes.

    Re: BarryW (#668),

     
    year           extent       weekly change
    2003         14.68328            -0.00141
    2004         14.14734            -0.14500
    2005         14.05734             0.11250
    2006         13.64203             0.08078
    2007         13.87063             0.11016
    2008         14.38687            -0.04141
    2009         14.41281             0.15922

  399. AndyW35
    Posted Mar 7, 2009 at 5:19 AM | Permalink

    I wonder if it has hit the max Barry, probably. The 6th is around the average for the max to be reached I seem to remember NSIDC saying. It seems to be bobbling along at around the peak now.

    Regards

    Andy

    • BarryW
      Posted Mar 7, 2009 at 7:20 AM | Permalink

      Re: AndyW35 (#680),
      Re: mark in austin (#677),

      Lately it seems to be peaking around the 9th or 10th. 2003 drops down and then grows again peaking at the 21st. The trend looks like it will probably start dropping in about 5 days if it holds to the average. I don’t think it will make up 100K, but in any event it will be very close to 2008. I haven’t been looking at the NSIDC numbers but I just checked and they are showing higher values (14.84) for Feb but about the same relative difference (140K) from 2008.

      I wonder how thick it is.

  400. AndyW35
    Posted Mar 8, 2009 at 6:44 AM | Permalink

    Hi Barry,

    I’m happy it is tracking 2008 as it will give an equal start point for the summer melt of 2009, whereas is 2007 it started off from a lower maximum. Wonder how it will do compared to 2008? You would have to say it would haver a slightly greater thickness but this can depend on not only the temps but how much insulating snow has fallen which we don’t know.

    This is probably going to be a more interesting melt than last year considering. A lot of people have stated that it will be a return to the normal and as the maxima is similar to last years there is no hiding place with that if it is not. Going to be fun!

    Regards

    Andy

  401. mark in austin
    Posted Mar 8, 2009 at 11:20 PM | Permalink

    well, it is starting to drop now, so perhaps the max has happened, but we won’t know for sure for another week or so…..

    • BarryW
      Posted Mar 9, 2009 at 5:46 AM | Permalink

      Re: mark in austin (#683),
      If the max is past then 2009 is coming in third with 14.41281 vs 2008 at 14.516875 (diff = 0.104065).

      Re: AndyW35 (#682),
      I think you also need to factor in wind and current as to how much compaction might have occurred.

      I bet the CL (Chicken Little) position will be that because 2009 is below 2008 we’ll be ice free at the pole (baby ice or no). We’ll see.

      Barry

  402. mark in austin
    Posted Mar 9, 2009 at 7:51 AM | Permalink

    although 2009 is only slightly below 2008’s high and you have to think there was some thickening this year….it should all be very interesting.

    anyone know why cryosphere today is down?

  403. hengav
    Posted Mar 9, 2009 at 10:56 AM | Permalink

    I did some posts on unthreaded using the Aqua LST images to chart the plight of the Catlin survey. You can look at the daily images to get an idea of how much the Arctic ice has been pushed and pulled around.

  404. hengav
    Posted Mar 9, 2009 at 11:22 AM | Permalink

    I found this March 09/09 image almost directly over the north pole and including the path of the Catlin group.

    No open water anywhere to be seen. Looks like they are going ot be travelling a long time through a rubble field. If they are attempting to take ice measurements they may think about fast forwarding closer to the pole, where it appears there are less fractures to deal with.

  405. AndyW
    Posted Mar 10, 2009 at 11:57 PM | Permalink

    Is there going to be a new thread for 2009 soon ?

    The last few days have seen a large drop in extent, quite unusual, well quite unusual within our few years on the JAXA graph! Certainly not the uniform decline of last year.

    Regards

    Andy

    • Phil.
      Posted Mar 11, 2009 at 12:14 AM | Permalink

      Re: AndyW (#688),

      Not unusual according to the JAXA graph, for example about the same drop occurred at a similar time in 2004, 2005, 2006 & 2007.

    • BarryW
      Posted Mar 11, 2009 at 8:56 AM | Permalink

      Re: AndyW (#688),

      As Phil said the change in extent varies quite a bit. There may be some positive days to come but the trend is into the negative. Here’s a spaghetti graph of the changes for 2003-2009. The center is today with the dotted lines future dates. The black line is the average of 2003-2008.

      extent change

  406. DeWitt Payne
    Posted Mar 11, 2009 at 4:09 PM | Permalink

    Unless we get a repeat of 2003’s late freeze, we’ve seen the peak Arctic, minimum Antarctic ice extent and area and minimum global ice area for 2009. Arctic area peaked, so far, at 13.853 Mm2 on 3/2/2009,Antarctic area hit a minimum of 1.790 Mm2 and global area a minimum of 15.113 Mm2 on 2/23/2009 according to Cryosphere Today data. The Antarctic ice area is increasing faster than average for this time of year. The anomaly for today’s data (actual date 3/5/2009 or possibly 3/4/2009) is 0.387 Mm2. This rapid increase has reduced the global sea ice anomaly to near zero (-0.036 Mm2). Because CT is five days behind JAXA, the recent drop in Arctic extent means that Arctic area is likely to drop a lot in the next five days. Uni-Hamburg February extent and area data is not expected before next week at the earliest.

    • BarryW
      Posted Mar 11, 2009 at 9:21 PM | Permalink

      Re: DeWitt Payne (#691),

      DeWitt, assuming we are past the peak ( and I’m pretty sure we are), where does 2009 Arctic area fall compared to 2003-8? With JAXA it’s in third for extent.

      • DeWitt Payne
        Posted Mar 12, 2009 at 9:03 AM | Permalink

        Re: BarryW (#692), Re: BarryW (#692),

        where does 2009 Arctic area fall compared to 2003-8?

        I don’t have Cryosphere Today digital data for anything but the last 9 months or so. I could do an approximation with past Uni-Hamburg data. But I haven’t looked at the peak area data for past years yet. I’d actually rather wait until Uni-Hamburg posts their March data in mid to late April. I would be surprised, though, if the ordering for the area is different than for the extent.

  407. Flanagan
    Posted Mar 12, 2009 at 9:41 AM | Permalink

    Well it looks like the 2009 extent rightnow is just the same as in 2007 at the same date. It will be interesting to see what happens next…

  408. DeWitt Payne
    Posted Mar 12, 2009 at 9:45 AM | Permalink

    Transforming CT area to UH, it looks like 2009 will be fourth, slightly behind 2004 while extent is slightly ahead of 2004.

  409. AndyW
    Posted Mar 13, 2009 at 12:47 PM | Permalink

    Phil and Barry,

    I am talking about start of March to date, it’s an early drop and a big one. Add up all the days from 1st March to today on the spreadsheet and tell me which other years get close.

    I probably didn’t make myself clear.

    Regards

    Andy

  410. AndyW
    Posted Mar 13, 2009 at 12:55 PM | Permalink

    #691 Hi Dewitt, the Antarctic looks as interesting as the Arctic this year. Although it is above zero for area anomaly it is half the value of this time last year and last years maximum was a very sorry effort so I do wonder what we are going to get this summer NH wise.

    It’s going to be very interesting to watch. Any thoughts on NH summer minima? I went too low on winter freeze extent which probably indicates more my alarmist side than science side :D

    Regards
    Andy

    • DeWitt Payne
      Posted Mar 13, 2009 at 6:35 PM | Permalink

      Re: AndyW (#697),

      For a brownie pool bet, I would pick a number higher than 2008 for the Arctic minimum under the assumption that 2007 was a fluke and not an acceleration of the trend. So the annual minimum would then return to the long term trend over the next few years. If the AMO remains negative, we could even see some increase in the minimum and annual average.

  411. BarryW
    Posted Mar 13, 2009 at 2:02 PM | Permalink

    here’s this weeks exents, the mean extent change is falling negative so It looks like we’re past max for sure. I won’t keep posting numbers unless something unusual happens till we get closer to minimum.

     
    year           extent       weekly change
    2003         14.62844            -0.03766
    2004         14.33359             0.17469
    2005         14.08875            -0.01016
    2006         13.75703             0.09234
    2007         13.86859             0.02109
    2008         14.47906             0.08047
    2009         14.13391            -0.27297

  412. mark in austin
    Posted Mar 13, 2009 at 8:11 PM | Permalink

    i am guessing it will be a higher minimum too. you have to think we regained some thickness in the past 2 winters….

  413. DeWitt Payne
    Posted Mar 17, 2009 at 8:41 AM | Permalink

    I’m going out of town for a few days and may not have internet access. Would some kind soul please post the Arctic and Antarctic area, anomaly and average from the Cryosphere Today iPhone pages starting today. The link takes you to the Arctic data. Click on the Antarctic button and then current for that data. Thanks.

  414. AndyW
    Posted Mar 17, 2009 at 10:23 AM | Permalink

    NH = -0.751
    SH = 0.364

  415. mark in austin
    Posted Mar 18, 2009 at 9:33 PM | Permalink

    looks like we MAY still have a shot at a slightly higher maximum than we have reached. this is surely the last hurrah, but the graphs have turned up the past 2 days and today even went up during the daylight, so with another cold night, we may see some good ice over night. then again, the bottom may fall out tonight and this post may be in vain.

  416. AndyW
    Posted Mar 18, 2009 at 11:55 PM | Permalink

    NH = -0.737
    SH = 0.345

    That’s a possibility mark, look at 2003 for instance. Yesterday though it is not much of a gain. Alaska and Russia opposite had been quite warm, then got a lot colder and now are intermediate, I get the feeling it will bobble along at the 14×10-6 mark for a while.

    1st of April is interesting, the JAXA graph is split into 2 distinct groups, the lows and the highs. You have to think 2009 will favour the high and join 2003,2004 and 2008.

    Regards
    Andy

  417. mark in austin
    Posted Mar 19, 2009 at 8:24 AM | Permalink

    i have a theory that they post 2 times per day….once at around 9 or 10 central time and again 12 hours later. so while yesterday only went up 1,000 miles for instance, this morning it added about 40,000 more miles. do you think this is due to overnight temperatures and ice buildup? my theory was that if it gained 1,000 during the day yesterday, it may grow more radically overnight. that’s what happened, but i could be VERY wrong about why!

    • BarryW
      Posted Mar 20, 2009 at 11:46 AM | Permalink

      Re: mark in austin (#705),

      I think Andy is right. The preliminary numbers are put up that night and a final is put up about ten the next morning.

  418. AndyW
    Posted Mar 19, 2009 at 11:57 PM | Permalink

    NH= -0.713
    SH= 0.292

    Hi Mark,

    I think it is simpler than that, they put up the rough value as soon as they have it and then adjust it after further evaluation. That’s always been my assumption anyhow.

    Regards

    Andy

  419. Hu McCulloch
    Posted Mar 21, 2009 at 8:30 AM | Permalink

    Next on NOVA: “Extreme Ice”

    http://www.pbs.org/wgbh/nova/extremeice/

    Tuesday, March 24 at 8 p.m.
    (Check your local listings as dates and times may vary. Broadcast in
    high definition where available.)

    Remarkable time-lapse footage by James Balog, one of the world’s
    foremost nature photographers, reveals massive glaciers and ice
    sheets splitting apart, collapsing, and disappearing at a rate that
    has more and more scientists alarmed. This NOVA-National Geographic
    Television special investigates the latest evidence of a radically
    warming planet.

  420. AndyW35
    Posted Mar 22, 2009 at 3:01 PM | Permalink

    NH= -0.666
    SH= 0.412

    Might have missed a day there, went down to the coast! It was not frozen in the English Channel and the fish and chips were hot!

    Regards

    Andy

    • DeWitt Payne
      Posted Mar 23, 2009 at 12:08 AM | Permalink

      Re: AndyW35 (#709),

      Thanks. I’m back now. Very competitive race between Audi and Peugeot at the Sebring 12 hour race this year.

      • desmitty
        Posted Mar 24, 2009 at 4:06 PM | Permalink

        Re: DeWitt Payne (#711),
        Most Interesting is the fact that the Audi TDI cars finishing 1st & 3rd as well as the Peugeot HDI which finished 2nd were all diesels! The Audi diesels have been cleaning up at Sebring and LeMans for three years or so… You can buy a TDI powered VW from your local dealer as an ’09 Jetta sedan or Sportwagon… These are extremely clean-running (particulate filtered) and quiet cars and currently qualify for a $1300 federal tax credit… Just returned from a 3+Km trip and averaged between 38 -40 mpg without even trying. In several locales, diesel is now cheaper than RUG… AGWphiles should embrace this technology…. Torque is incredible!

        Dave

        • DeWitt Payne
          Posted Mar 24, 2009 at 5:54 PM | Permalink

          Re: desmitty (#718),

          Totally OT.

          Audi won at LeMans last year because the weather (rain) favored their cars. The Peugeot was much faster than the Audi in the dry. This year’s Audi was about equal to the Peugeot at Sebring with the leaders for both makes on the same lap for the whole race, excepting lead chnges from the pit stop cycle. A record was set for number of lead changes. The lead changed three times on one lap. Unfortunately, the pole winning Accura broke fairly early.

          MPG is not a perfect measure for the difference in efficiency between diesel and gasoline. Diesel has about a 15% higher BTU/gallon content because it’s denser than gasoline. The price of diesel wrt gasoline also varies during the year because diesel is also sold as heating oil and the refineries can only adjust their production ratio over a fairly narrow range. So diesel goes up in the fall and down in the spring while gasoline does the opposite. Last year was worse because the Chinese were buying mass quantities of diesel regardless of price. So the calculation of whether a diesel is worth a premium price compared to a gasoline engine is complicated.

  421. mark in austin
    Posted Mar 22, 2009 at 4:31 PM | Permalink

    andy, that makes more sense…..thanks!

    so far the ice is holding up fairly well for this time of year…..

  422. AndyW
    Posted Mar 23, 2009 at 12:39 AM | Permalink

    Didn’t a Honda Accord get on pole? I’ll have to check results but I guess from your post the new Audi won! Will be interesting at Le Mans this year.

    Regards

    Andy

  423. mark in austin
    Posted Mar 23, 2009 at 9:06 PM | Permalink

    ok..i am back to my original theory. if it really is just rough and then actual…then why is it being adjusted up? i mean, i am not saying you are wrong, but i am curious. if my theory is right it should be adjusted up from 14,238,750 tomorrow morning…

    • BarryW
      Posted Mar 23, 2009 at 9:12 PM | Permalink

      Re: mark in austin (#713),

      Good question, I haven’t tracked the two numbers to see whether there is any pattern to the adjustments. Maybe there is a manual correction for questionable areas.

      • DeWitt Payne
        Posted Mar 24, 2009 at 9:58 AM | Permalink

        Re: BarryW (#714),

        Think about the data itself. The satellite passes over the pole and images a swath that does not cover the entire area of interest. Ninety minutes or so later it passes over the pole again and images another swath that overlaps the previous swath. But the ice concentration changes continuously so the overlapped areas won’t match. The 11 PM post is for the data accumulated over the previous ~thirteen hours by assembling the swaths into a complete image using the most recent data for any given area, which may include data from the previous day. The 10 AM post is for the data accumulated over the next eleven hours and would not include any data from the previous day. Orbital mechanics are such that even then, it’s possible that not all the area of interest is covered. That shows up as odd shaped blank areas in the images.

        • BarryW
          Posted Mar 24, 2009 at 3:00 PM | Permalink

          Re: DeWitt Payne (#716),

          That brings up the question of how much “noise” is caused by the temporal variation in the “slices” that are being accumulated. We’re given a single value with no uncertainty identified. How much is real change and how much data collection error.

    • Phil.
      Posted Mar 24, 2009 at 6:49 AM | Permalink

      Re: mark in austin (#713),

      It’s not always adjusted up and only one number is recorded per day in their database.

  424. AndyW
    Posted Mar 25, 2009 at 12:05 AM | Permalink

    OT but I am sort of hoping Aston Martin do well this year at Le Mans (being british).

    This is great, 2009 is going into April at 2008 extent! Now we will have a nice back to back year to compare for the summer. I’m guessing that we will have an average year like last the summer minima will be between 2008 and 2005. I think the concencus here is along those lines from what peoople have posted.

    Regards

    Andy

    • Phil.
      Posted Mar 25, 2009 at 10:43 AM | Permalink

      Re: AndyW (#720),

      This is great, 2009 is going into April at 2008 extent! Now we will have a nice back to back year to compare for the summer. I’m guessing that we will have an average year like last the summer minima will be between 2008 and 2005. I think the concencus here is along those lines from what peoople have posted.

      Last year’s melt was a record not average, however the SST surrounding the ice is currently warmer than this time last year so a larger melt might be expected this time. Irrespective of what the extent is now the tendency over recent years has been for the extent to converge on ~12.5 Mm^3 around May 1st.

  425. mark in austin
    Posted Mar 25, 2009 at 3:13 PM | Permalink

    phil,

    imo, that’s a bit misleading. yes, last year was the largest melt off, but that is sort of expected because the ice had to be thinner coming off the 2007 melt. this year will be very interesting if the ice has indeed thickened because it will be more difficult to melt it….

    • Phil.
      Posted Mar 25, 2009 at 11:22 PM | Permalink

      Re: mark in austin (#722),

      And if the ice has not thickened (why would you think that it has?) it will not be more difficult to melt. Which part do you find misleading, that the surrounding seas are warmer than last year?

  426. AndyW
    Posted Mar 26, 2009 at 12:06 AM | Permalink

    Hi Phil,

    I meant averge weather last year rather than the ideal weather they had in 2007, sorry I did not make myself clear. It’s a good point though that last year more area was lost than in 2007.

    SST’s seem quite cold in the northern Pacific and warm in the Atlantic.

    http://weather.unisys.com/surface/sst_anom.html

    Regards

    Andy

    • Phil.
      Posted Mar 26, 2009 at 1:38 AM | Permalink

      Re: AndyW (#724),

      Here’s this month’s SST anomaly:

      and last March’s:

  427. Chris
    Posted Mar 26, 2009 at 11:33 AM | Permalink

    Bear in mind the bigger picture in the NH:

    http://www.cru.uea.ac.uk/cru/data/temperature/hadsst2nh.txt

  428. AndyW35
    Posted Mar 26, 2009 at 12:27 PM | Permalink

    Thanks for the images Phil. Certainly a lot warmer this year than last. Does this mean you expect the retreat to be amplified from April onwards? I would expect the SST’s to be most noticeable when the direct solar effects are not the driving force as they are around the summer.

    Regards

    Andy

  429. mark in austin
    Posted Mar 26, 2009 at 5:07 PM | Permalink

    i am making an assumption that 2 cold winters may have the ice a bit thicker than it was last year at the same time. needless to say, this may be a faulty assumption.

  430. Chris
    Posted Mar 26, 2009 at 6:28 PM | Permalink

    #728: Yes the problem with your assumption is the near-record melt season in between.
    However, there could also be a “First In Last Out” assumption……
    – During Oct 08 the ice was consistently “In” close to 2 weeks earlier than in Oct 07, and during Nov 08 close to 1 week earlier than in Nov 07.

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    – Close to the North Pole, surface temperatures this winter appear to have been reasonably similar to those of last winter (i.e. within ~0.5C), while around many of the peripheries they appear to have been significantly colder.

    http://tinyurl.com/ce62ko

    – Therefore *if other things were equal* it would seem logical that the ice will be “Out” later in Jun/Jul/Aug.
    Anyway, I didn’t mean to get drawn into this again. Believe what you will. Time will tell.

  431. Chris
    Posted Mar 26, 2009 at 6:41 PM | Permalink

    Not sure if my last link was right. I think it should have been http://tinyurl.com/c68tnw
    (slightly ambiguous when selecting years)
    This shows some “hotspots” in parts of the High Arctic relative to last year, but still an emphasis on cold as you get to the peripheries.

  432. AndyW
    Posted Mar 27, 2009 at 12:06 AM | Permalink

    Nice to see you back after your summer hibernation Chris ! Always good to hear your thoughts, especially the meteorology side of things.

    Regards

    Andy

  433. DeWitt Payne
    Posted Mar 27, 2009 at 11:18 AM | Permalink

    Not that it means all that much, but the global sea ice area anomaly at Cryosphere Today went positive today, 0.078 Mm2.

  434. tty
    Posted Mar 28, 2009 at 2:20 PM | Permalink

    I just had a look at Cryosphere and noticed something (in addition to a new and rather garish colour scheme). They are up to March 24 and on that day they show the southwestern part of the Gulf of Bothnia as being ice-free. It so happens that I flew across that area on that very day, and so I know that it was actually ice-covered. However it was covered by smooth new ice without any snow on it, perfect skating ice as a matter of fact. So it seems that smooth ice without snow is invisible to satellites (or at least the satellite Cryosphere is using). Interesting.

    • Phil.
      Posted Mar 28, 2009 at 8:16 PM | Permalink

      Re: tty (#733),

      Not just the satellites apparently!

  435. tty
    Posted Mar 29, 2009 at 3:52 AM | Permalink

    Re 734

    Actually the norwegian map is much better. I was referring to the orange area that the norwegians are calling “close drift ice 7-9 tenths”, which may well be true by now, if the new ice has broken up in the meantime.

  436. AndyW35
    Posted Mar 29, 2009 at 7:30 AM | Permalink

    Do they include ice in the Baltic or off the pacific coast of Russia or even the US Great Lakes in the calculation for Artic ice extent/area?

    I wonder if 2009 is going to drop like 2004 in the next week or two?

    Regards

    Andy

  437. DeWitt Payne
    Posted Mar 29, 2009 at 11:24 AM | Permalink

    Cryosphere Today has an interactive map that shows the regions they include in their Arctic ice area measure. The Sea of Okhotsk is included. The US Great Lakes aren’t. The color scheme isn’t the best. The Sea of Okhotsk to the East of the Bering Sea is the same color as the Pacific. You have to move the cursor over it to see where it is.

  438. See - owe to Rich
    Posted Mar 29, 2009 at 1:53 PM | Permalink

    The northern sea ice is no longer in end of game mode.

    It’s starting a new game.

    This thread is more than 6 months old and diabolically slow to load.

    Pse pse can we have a new thread?

    Rich.

    Steve: Yep

  439. mark in austin
    Posted Mar 30, 2009 at 10:54 PM | Permalink

    what is the link?

  440. randy
    Posted May 26, 2009 at 5:27 PM | Permalink

    Where’s the link?

  441. randy
    Posted May 26, 2009 at 5:28 PM | Permalink

    Nevermind, I found it now. It’s at:

    http://www.climateaudit.org/?p=5614

  442. Phil.
    Posted Feb 18, 2009 at 10:36 PM | Permalink

    Re: mark in austin (#636),

    mark in austin:
    February 18th, 2009 at 5:21 pm
    yeah….Phil, that graph was bogus. were you not suspicious? i guess if it fits your AGW beliefs….then go with it?

    No reason to think that it’s bogus at all since it’s consistent with the AMSR-E data. Since I always crosscheck between the sensors rather than rely on one your abuse is off the mark, and you know nothing about what I believe in anyway.

    The Arctic map however has had some problems recently due to problems with the SSM/I sensors, which I’d noticed and posted on elsewhere.

    Re: Daryl M (#637),

    What I’m wondering is does the degradation of the sensor have anything to do with the abrupt levelling off of the NSIDC sea ice extent graph in mid-January?

    Nothing at all, it shows up on graphs of data from AMSR-E sensors too, e.g.

    http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm

    It’s quite a common phenomenon around this time of year in any case.

  443. Dave Dardinger
    Posted Feb 18, 2009 at 11:07 PM | Permalink

    Re: Phil. (#638),

    The map at your last link in interesting. I notice that there’s a blip in at least several years data in early June, which corresponds to the remark that the data began in June 2002. Has anyone looked at code used to see whether there’s some sort of analysis which uses the data from previous years, resulting in the strange correlation?

  444. Daryl M
    Posted Feb 20, 2009 at 3:46 PM | Permalink

    Re: Phil. (#637) In #636 I wrote: “What I’m wondering is does the degradation of the sensor have anything to do with the abrupt levelling off of the NSIDC sea ice extent graph in mid-January?”, to which you replied, “Nothing at all, it shows up on graphs of data from AMSR-E sensors too [...] It’s quite a common phenomenon around this time of year in any case.”

    Apparently NSIDC does not agree with you. They released this graph which clearly shows an increasing divergence in sea ice extent between the SSM/I and AMSR-E sensors starting around the beginning of January, which is exactly what I was referring to in my post. SSM/I levels off but AMSR-E continues to increase.

  445. Phil.
    Posted Feb 19, 2009 at 12:14 AM | Permalink

    Re: Dave Dardinger (#638),

    I notice that there’s a blip in at least several years data in early June, which corresponds to the remark that the data began in June 2002. Has anyone looked at code used to see whether there’s some sort of analysis which uses the data from previous years, resulting in the strange correlation?

    That’s been commented on before, as I recall there’s a switchover around that time of year in the algorithm and microwave frequencies used in order to compensate for the formation of liquid water pools. Sometimes there’s a little ‘blip’ when they changeover. JAXA uses the NASA algorithm whereas UBremen uses ASI.

  446. Phil.
    Posted Feb 20, 2009 at 4:45 PM | Permalink

    Re: Daryl M (#654),

    In #636 I wrote: “What I’m wondering is does the degradation of the sensor have anything to do with the abrupt levelling off of the NSIDC sea ice extent graph in mid-January?”, to which you replied, “Nothing at all, it shows up on graphs of data from AMSR-E sensors too [...] It’s quite a common phenomenon around this time of year in any case.”
    Apparently NSIDC does not agree with you.

    But JAXA does, which is what I was referring to:

    Here’s the data for that period:
    01,15,2009, 13282969
    01,16,2009, 13364688
    01,17,2009, 13363125
    01,18,2009, 13385000
    01,19,2009, 13471250
    01,20,2009, 13439219
    01,21,2009, 13383594
    01,22,2009, 13390625
    01,23,2009, 13396563
    01,24,2009, 13387813
    01,25,2009, 13363594
    01,26,2009, 13413906

  447. Daryl M
    Posted Feb 20, 2009 at 6:13 PM | Permalink

    Re: Phil. (#656) Here’s the image generated by NSIDS showing the difference between the two sensors. It’s located at http://nsidc.org/images/arcticseaicenews/20090217_Figure2.png. If this doesn’t make it obvious that there is a discrepancy between the two sensors, I don’t know what does.

  448. Phil.
    Posted Feb 20, 2009 at 8:56 PM | Permalink

    Re: Daryl M (#660),

    Daryl M:
    February 20th, 2009 at 6:13 pm
    Re: Phil. (#656) Here’s the image generated by NSIDS showing the difference between the two sensors. It’s located at http://nsidc.org/images/arcticseaicenews/20090217_Figure2.png. If this doesn’t make it obvious that there is a discrepancy between the two sensors, I don’t know what does.

    What it shows is that there’s a discrepancy between NSIDC’s implementation of SSM/I and AMSR-E. While NSIDC has had major problems with their calculation ArcticROOS for example does not (see Phil. (#641)) Comparison between ArcticROOS and JAXA shows very good agreement and both show the flat spot in Jan. For whatever reason NSIDC has a problem handling the missing swathes which AROOS does not (at least so far). Even so the graph you show still shows a flat region in both datasets therefore even for NSIDC it is not due to the problems with SSM/I.

2 Trackbacks

  1. [...] Re CT: At least BC provided an explanation and this will always enhance the standing of CT. The layout is great and very informative Excerpt from [...]

  2. [...] we do not speak of looks at the arctic ice shrinkage that [...]

Follow

Get every new post delivered to your Inbox.

Join 3,297 other followers

%d bloggers like this: