Continuing from http://www.climateaudit.org/?p=6734
-
Tip Jar
(The Tip Jar is working again, via a temporary location) -
Pages
-
Categories
-
Articles
-
Blogroll
- Accuweather Blogs
- Andrew Revkin
- Anthony Watts
- Bishop Hill
- Bob Tisdale
- Dan Hughes
- David Stockwell
- Icecap
- Idsos
- James Annan
- Jeff Id
- Josh Halpern
- Judith Curry
- Keith Kloor
- Klimazweibel
- Lubos Motl
- Lucia's Blackboard
- Matt Briggs
- NASA GISS
- Nature Blogs
- RealClimate
- Roger Pielke Jr
- Roger Pielke Sr
- Roman M
- Science of Doom
- Tamino
- Warwick Hughes
- Watts Up With That
- William Connolley
- WordPress.com
- World Climate Report
-
Favorite posts
-
Links
-
Weblogs and resources
-
Archives
Climate Audit 
486 Comments
First on the new thread??
A rather big adjustment today, to 6,093,594 km2. Have the winds stopped (slowed down)?
Re: MikeP (#1),
date extent(km2) difference EWMA rate
10/9/2009 6093594 108438 40524
10/8/2008 6209219 210156 88004
So we have finally seen an increase of over 100,000 km2 in a day. Unfortunately, 2009 still lost 100,000 km2 to 2008. Cryosphere Arctic area actually decreases today. If all this were due to compaction, I would expect to see concentration going up. It’s not.
Re: DeWitt Payne (#5),
Well, it’s blinking cold in the southern US at the moment, so I suppose it has to be warm somewhere else. I think it’ll all average out over the next year or so when the mild el nino caused by oceanic heat release subsides.
Re: MikeP (#1), Now at 6,160,938 km2.
At the beginning of this year I did an analysis that appears to indicate a systematic drift between the NASA team SSM/I relative to AMSR-E IRAC JAXA. In fact the whole historic Arctic sea ice record seems to be somewhat suspect:
Click to access How_Fast_is_Arctic_Sea_Ice_Declining.pdf
I requested additional information about validation between data sets from Chapman and Walsh, but there was no response.
Re: Jonathan Drake (#2),
That’s a great piece of work.The recent paper by Syd Levitus et al regarding AMO and Barents Sea correlation may be of interest.
Click to access 2009GL039847.pdf
Could it be that a rise in north atlantic ocean heat content during the long run of high C20th solar cycles is the main driver behind arctic sea ice area and extent?
Re: tallbloke (#3),
Tallbloke, thank you for your kind remark. I haven’t read the paper yet, but in my opinion it would be surprising if the ocean circulation has no impact.
Re: Jonathan Drake (#2), That’s a great article. When do you intend updating it to 2009?
Re: Phillip Bratby (#9),
Phillip, thank you for your kind remark. I had thought about updating it, but since there has been very little reaction to it over the months, my interest in doing so waned. Even as it is, I think there is enough to raise serious questions about the Arctic sea ice records.
still claim to see the 1999 Gakkel Ridge eruption on an animation of the ice movement. The area was obscured by clouds, photos of which are hard to find, but a meteorologist ought to be able to distinguish whether those clouds are ordinary ones, or the clouds generated by open water, or by thin, warmed, ice. I know, supposedly there was too much mass of water and thermoclines to allow any effect to reach the surface, but that can’t be known if don’t even know the magnitude of the eruptions.
========================================
The Danish Meteorological Institute Arctic Temperaure graph has not been updated since October 7th. Does anyone know why?
http://ocean.dmi.dk/arctic/meant80n.uk.php
Sea ice growth has been rather slow lately.
I speculate that it has a lot to do with the very big transfer of cold air masses to continental Canada and USA that is now occurring. This has made room for warmer southern air masses to move up into the polar arctic regions.
A huge mass of polar air is also forecast to plunge down into western Europe this week and stick around for awhile. See the temp anomaly chart at this link:
http://wxmaps.org/pix/temp4.html
And the following state-funded institute (jobs program for scientists with nothing to do) delivers the long awaited “3rd lowest ever sea ice extent!” release.
http://www.damocles-eu.org/research/Arctic_sea_ice_extent_third_lowest_on_record.shtml
6,257,188 reports of slow growth look over stated.
Extent = Area + lots of open water. Cryosphere reports for 5 days ago over 2 million km square shortfall on that measure for the Arctic. And whatever the lalala on NASA/NOAA/NSIDC/GISS, long as there is extensive matching between the different data collectors and processors, it is really True, however disappointing that may be to the Aint Trueists. The North Eastern Passage is WIDE open. Nansen http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic shows currently drop for Area and Extent below 2008 levels. All do, including Jaxa. Some recovery.
Re: Sekerob (#15),
Cryosphere Today is now only two days behind JAXA. They had server problems some months ago and didn’t update for weeks at a time. When that was fixed, three of the five day lag had disappeared. Based on the JAXA extent increase, CT should show big increases in area the next two days. Ice has started to form in the Chuckchi Sea as well as Baffin/Newfoundland and Hudson Bay. There was a loss in area in the Arctic Basin and the Laptev Sea
Re: Sekerob (#15), Re: tty (#16), You were saying Sekerob?
Re: Michael Jennings (#18), I usually look at the Cryosphere Today picture of Arctic sea ice. It seems to show the NE passage “wide open”, but tty’s link tells a different story – barely open.
Which graphic do we believe?
Rich.
Sekerob:
It was wide open, have a look:
http://www.ijis.iarc.uaf.edu/cgi-bin/seaice-monitor.cgi?lang=e&mode=img&size=L&date=set&y=2009&m=10&d=10
Looks like..
source: Bob Tisdale
Okay, change
to: Sea ice HAD BEEN rather slow.
Maybe some historic perspective would be useful?
Ships logs are to be examined to assess past ‘climate’:
http://physicsworld.com/blog/2009/10/climate_science_aboard_hms_bea.html
Quote from Physics World:
At the CORRAL website some of the analysed data is here:
http://www.corral.org.uk/digitised-logbook-observations
Re: Jonathan Drake (#22), Re: Jonathan Drake (#22),
Wow!!! The CORRAL website is a real eye opener. Am I missing something – doesn’t this pretty much kill the hockey stick for good? Who would prefer to use tree rings in a small area than actual air and sea temperature measurements covering a much larger area to base their analysis of the global temperature over the past 100 years? What’s the conclusion can one come to using the CORRAL results? Global temperatures increased only very slightly, if at all over the past 100 years?
DMI still showing air temperatures in Arctic way above normal which is a cold US and Europe and slow ice growth IMO
Re: Colin Aldridge (#24),
DMI has not updated the temperature graph since 7 October.
Re: Colin Aldridge (#24),
Looks like DNI updated this morning. Not nearly so warm as it was. Looks like temperatures have begun falling again.
If the cold is not in the Arctic, then it is somewhere else.
Sekerob #15
The JAXA site shows that at the point of highest melt-off, approx mid September, 2009 has some 500,000 sq km more ice than 2008 and some 750,000 sq km more than 2007. As of today, 2009, 2008 and 2005 values are the same. If that is not a recovery from the 2007 low, what am I missing?
Now at 6,379,063 km2
Am I the only cynic worried that the UK Met Office has a significant role in the the analysis of the Royal Navy logs?
I look at the graphs from the logs, and the current temperatures are overlaid for comparison. The current temperature line actually looks like the average of the 19th Century readings, and we get the following “analysis” (from the Met Office representative no doubt)
“Detailed analysis demonstrates that, around Svalbard, the summer of 1818 not markedly colder than was typical in the late 20th century.”
Is it so difficult to avoid this use of weasel words? Why not just say “1818 temperatures appear very similar to current temperatures”?
How long before the Met Office has a mysterious fire which destroys all these embarrassing logs…
Re: Tony B (#30), This calls to mind a picture:
TonyB:
The MET has not yet adjusted the historic records you refer to. They still need to account for the “Known” differences between past and current measurement equipment which will dramatically cool the past.Also the Time Of Day adjustment has probably not been accounted for yet. After proper adjustment we could expect .7C trend for that period also unless it was supposed to have a flat trend before 1944 when it shows a dramatic upward slope to the present.;-)
@Mike Davis
Wonderful news! Adjustments – Not yet, but galloping over the horizon.
I don’t suppose there is any possibility that any of these adjustments will warm up the past, is there?
No of course not – silly me.
So this will mean that the embarrassment of the “Most Influential/Important Tree in the World” can be forgotten, because it will not actually matter…
Phew, that’s a relief.
Move on again.
http://nsidc.org/images/arcticseaicenews/20091005_Figure3.pngRe: Juraj V. (#19),
Not much of a correlation between ice extent: http://nsidc.org/images/arcticseaicenews/20091005_Figure3.png
and Tisdale’s graph of ocean heat content: http://i38.tinypic.com/9u417d.png
Adjusted figure for 10/12 is 6,473,438 km2 for a total increase of ~94,000
Cryosphere Today hasn’t updated for today yet. Looking at the images at Uni-Bremen, there’s been a big increase in area in the East Siberian Sea and a loss in the Beaufort Sea from 10/8 to 10/12. The images for 10/9 and 10/10 are still not in the archives.
Arctic 10/8/2009
10/11/2009
10/12/2009
Re Chris Schoneveld
Actually, the correlation is quite good, if you compare only the 1979-2009 part of the OHC graph. Second, Arctic minimum had been hit hard in 2007 by “unusual patterns of winds and ocean currents” (quote by NASA), which Arctic is still recovering from. Third, plotting only September minimum is not the best choice. Here is plotted AMO index vs Arctic ice extent; AMO is correlated very well with OHC, since positive AMO cycle pushes warmer water into the Arctic. The correlation is excellent, and the recovery is on the way, despite pronounced minimums for the last three years.
I find all this hubub about the arctic ice rather disingenous. The “Cap”, that is the water in the actual arctic circle fully freezes every year. The Winter anomolies are the result of looking at ice that, imo, is actually outside of the arctic.
Granted, the summer melt has been interesting, although, JAXA shows clearly that 2009 represents a two year growing trend. But, the winter interpretations are bunk! Cryosphere’s anomoly chart [the Tale of the Tape] makes it look as if 2009 had more melt than 2008, but this is all smoke and mirrors.
Re: Deanster (#36),
I pretty much agree. The only two places I fund interesting are maximum and minimum. The rest of the year when it is in the process of melting and freezing in summer and winter are sort of like the first three innings of a baseball game. It pretty much is what it is at this point and any slow freezing can be made up later or any fast freezing can also be changed later. About the only thing that matters now is how much we end up with in March.
And the anomaly chart only shows the difference between that day and “average”. Being so low right now just means that on this day in 2008 there was more ice than there is in 2009, which is accurate. It is about the proper interpretation of the graph more than the value it shows. Sure, one who doesn’t really understand what it is showing might be misled but that is pretty much the case with everything.
Re: crosspatch (#38),
The problem is that the current highs and lows are a lot lower than the 1972-2008 average. See this plot for example.
Re: DeWitt Payne (#39),
Two points of order.
Do you think the technology used to measure the sea ice has improved dramatically? I mean AQUA was launched in just 2002 for the express purpose of helping eek out each nook and cranny of vanishing ice. Hansen didn’t even hit the alarm bells until 1988. Way more scrutiny now, a very positive thing IMHO. All things being equal, I wouldn’t be surprised that the modest recoveries for the last 2 years are within experimental error.
Would you consider the lows of the past 8 years to be “unprecedented” in the last 130 years? My bet is that you don’t. If my hunch is right, would you consider qualifying your comment ?
Thank you for the info, with the recent catastrophic incidents which caused many lives, we are to be aware of the irregular climate changes.
Oh, and I was walking on a meter of artificial snow at our Canada Olympic Park in Calgary tonight. We’re into our second week of sub-zero temperatures after a wonderful El-Nino September. When we had the games here in 1988 during a serious “Chinook”. We all walked around in shorts. This new El Nino has no effect on this major and early winter blast.
Re DeWitt Payne
The problem is that the current highs and lows are a lot lower than the 1972-2008 average. See this plot for example.
Depends on which graph you use.
Re: Juraj V. (#43),
I consider the Arctic ROOS graphs to be useful for qualitative information only. They do not agree in detail with any other source of Arctic ice extent and area data. A year ago you could overlay JAXA or Uni-Hamburg on Arctic ROOS and with a little stretching or squeezing get all the years to line up pretty well. That isn’t true any more.
Lol whats going on with DMI arctic ice graph? It is suddenly heading steeply down 😮
http://ocean.dmi.dk/arctic/icecover.uk.php
Re: Juraj V. (#44),
DMI has looked messed up since the beginning of October.
Today’s figure is almost unchanged from yesterday at 6,474,844 km2 October 13, 2009Re: Juraj V. (#44),
Could be an algorithm adjustment?
Re: Michael Jennings (#45),
Posting the preliminary data from JAXA is nearly useless.
Revised data for 10/13/2009:
date extent(km2) difference EWMA rate
10/13/2009 6533594 60156 62895
10/12/2008 6695000 125625 99288
10/13/2007 5374219 74844 48407
10/13/2006 7048438 271094 50053
10/13/2005 6678594 135625 54063
10/12/2004 7549063 147500 74372
10/13/2003 7253281 170781 49160
10/13/2002 7806094 52656 80610
DOY 286 8244977 79535 69286
The final number is the NSIDC 1979-2000 average extent for the current Day of Year. 2009 continues to lose ground to 2008. Cryosphere Today is still having problems.
I knew it would come. the results of the Catlin Expedition
Today’s Australian newspaper
Sorry link may not have worked, its here at http://www.theaustralian.news.com.au/story/0,25197,26213406-601,00.html
DeWitt Payne [#47]
Have a look at WUWT earlier today on this topic. Watts makes a rather compelling argument that only the Jaxa data stands up to scrutiny. All the other “sources”, be they US, Danish, German or Norwegian, rely on a satellite that we know beyond discussion to have fundamental failures.
A delta of some 2,000,000 sq km between one source that has been consistent for 9 years and another that has had an obvious series of electronic heart attacks resulting in downward spikes, is hard to reconcile in the latter’s favour. Not?
Re: tetris (#52),
Uni-Bremen (images) and Uni-Hamburg (extent and area data) also use only data from the Aqua satellite (AMSR-E), but use a different algorithm for calculating ice concentration than JAXA. NSIDC before 2007 seems to be reliable as there is good correlation between JAXA and NSIDC for the period of overlap. The greatest error in passive microwave concentration determination is at the minimum because the ice surface is wet. A difference in the area or extent does not bother me if the difference is consistent from year to year.
DJA [#51]
Put enough window putty and hot sauce on a piece of cardboard and [for a quite a while] you will be able to convince the gross of the buyers that they are eating “pizza”. Catlin is merely frozen hot sauce.
http://news.bbc.co.uk/1/hi/sci/tech/8307272.stm The BBC are reporting ice free artic in ten years as well.
2009 is still losing ground to 2008, the freeze season is pretty slow starting this year. Previously it was increasing on the Canadian side but now has moved to the Russian side so it should pick up soon.
Regards
Andy
Re: AndyW35 (#56),
Overall the difference is very little, though. Also, looks like DMI has fixed whatever glitch that was.
Re: crosspatch (#59), If the problem is caused by the sensor, it’s far from fixed.
I may be about to look foolish, but here goes. Presumably the sea ice acts as an insulator, slowing energy flux to the atmosphere in winter. So if there is less sea ice then energy flux to the atmosphere will be greater, especially in spring and autumn. So is it possible that the effect of reduced sea ice will be to cool the arctic ocean, so there will be a negative feedback set up and we will actually not get to a zero sea ice situation?
Re: Paul (#57),
“So is it possible that . . . we will actually not get to a zero sea ice situation?”
Paul,
Not only possible, but also likely. At various times, I have offered wagers of up to $100,000 that we will not be ice free by 2013 — or 2016 which is the latest year of the range given by Maslowski and Wadhams. And I am willing to extend the years out farther out. And I have offered wagers that Polar Bears will be alive and thriving in 2050. Etc. So far, no takers.
It is my understanding that Climate Audit does not see itself as a forum for bets, so I do not pursue that issue, but I do not see a committed belief even amongst GW pessimists that we actually will be ice free.
Re: An Inquirer (#62),
“Not only possible, but also likely. At various times, I have offered wagers of up to $100,000 that we will not be ice free by 2013 — or 2016 which is the latest year of the range given by Maslowski and Wadhams. And I am willing to extend the years out farther out. And I have offered wagers that Polar Bears will be alive and thriving in 2050. Etc. So far, no takers.”
Probably because it’s not a bet with any sort of weight behind it rather than make a point with bluster? If you are that sure go and put it on at the bookmakers and show us the slip?
Regards
Andy
Adjusted figure for 10/14 is 6,632,813 km2 for an increase of ~100,000
Currently global sea ice is between 1.5 and 2 million sq km below average according to Cryosphere Today. Global sea ice maximum is due in the next month or so and if sea ice stays far enough below 1 million sq km below average a new record low for global maximum will be set.
The last couple of years Arctic anomalies have increased (more ice) significantly around this time of year. Also the Antartic ice seems to be quite volatile with plenty of up and down movement, so could easily jump up by half a million close to the global max, and even if only for a couple days or a week this would reduce the chance of a record low max as well.
Re: Michael Hauber (#61),
Cryosphere Today hasn’t updated for several days now (server problem) so the data there does not reflect the recent uptick in the Arctic. Unless the Antarctic has lost a lot of ice, I would expect that the global anomaly has become less negative even though Arctic ice is increasing at a below average rate.
Why not start a Catlin debunking thread, and lay out in clear and well reasoned arguments why their data should be ignored? Pen Hadow has pretty much been clearing the table; seemingly with no opposition.
Cheers!
Re: INGSOC (#63), I believe that has happened on WUWT at http://wattsupwiththat.com/2009/10/15/top-ten-reasons-why-i-think-catlin-arctic-ice-survey-data-cant-be-trusted/ Steve and Anthony tend not to replicate each other so that is probably your best bet for the conversation.
Re: realist (#67),
Point taken. But there is, how should I say this… CA has a more technical and restrained policy regarding comments that I believe would reduce the amount of vitriol that might occur with this subject. I am in no way denigrating Mr. Watt’s fine site! In fact, far from it as I enjoy more than the occasional visit there myself! I was thinking that perhaps the more stately approach evinced here might tend to bolster WUWT in what is perhaps a very important battle for “hearts and minds”. I would suggest that the two sites would rather effectively re-enforce each other, as well as provide broader coverage of this issue.
Your point of cross posting is well taken however. It is merely a suggestion. Lets just say that I feel that the supporters of the Catlin boondoggle (among others) have a decided advantage in peddling their nonsense, and further that we should be countering this baloney with every thing we believers in true science can muster.
Re: INGSOC (#63),
I posted a reply to you here earlier today but now see that it has been removed. I’ll try again.
It is a waste of time trying to debunk the Catlin story. The fact that the “science” is complete rubbish is immaterial. The Catlin caper is a product of alarmist “science by press release” that is enabled by the fact that most media no longer have journalists with any sort of scientific training/background to vet news wire feed or as in the case of e.g. the Economist, a science editor who has gone religious on AGW/ACC. In any event, there will be a slew of these alarmist “scientific” “preditions, “forecasts”, “projections”, etc., over the next weeks in the run-up to the Copenhagen climate confab.
To cap it all, most observers appear to be unaware of the key element in the Catlin caper, namely that it was a carefully planned publicity stunt, sponsored by the Catlin insurance syndicate, which -no extra points for surprise- specializes in “global warming” and “climate change” insurance. As I wrote yesterday to the editor of one Canada’s leading dailies, imagine for just one moment what would have happened to a “CO2-is-good-for-you” study that turned out to have been sponsored by a coal company. Seriously?
Picking up a bit now. Plus 112000 for 15th so far.
Final tally for 10/15 is 6,745,781 km2 which as mike worst pointed out, is a ~112,000 increase
DeWitt Payne, It seems that we have a sensor failure involved here. Details will come but we can discount anything from Cryosphere Today for now.
2009 falls further behind 2008 today.
date extent(km2) difference EWMA rate
10/16/2009 6759688 13907 65445
10/15/2008 7242344 178125 121964
10/16/2007 5640156 34062 58543
10/16/2006 7521875 75781 95955
10/16/2005 6941875 -16875 76763
10/15/2004 7722344 30938 79412
10/16/2003 7674063 52344 84668
10/16/2002 7972500 22656 73480
DOY 289 8477443 81158 71388
So the extent anomaly from the 1979-2000 average is -1.71 Mm2, second only to 2007, admittedly by quite a lot. However, in two days, 2007 extent starts to increase rapidly. Unless 2009 follows suit, it will start to lose ground to 2007 too.
Re: DeWitt Payne (#72),
What is the significance? Fact remains that at the [crucial] time of max melt [around September 15th] 2009 showed a 25% increase in extent over 2007 [and by all accounts thicker ice than anyone anticipated or any model projected. Ref the German/Canadian 2009 airborne survey].
Considerably more important – in terms of both ice rejuvenation and thickness – 2009 had absolute JAXA record levels around mid-February and for the entire mid-April through mid-May period. Whether the 2009 curve intersects 2008, 2007, 2005 or 2006 next week or the week thereafter is moot. 2009 was a record “max” year when it counted and a whopping improvement over the previous two year at the “min”.
Arguing anything different is like arguing that the DOW index recovering to over 10,000 after the 2007-2008 meltdown is not significant because that is still under the proverbial “1979-2000” average…..
Re: tetris (#73),
The significance should be obvious. The hypothesis that Arctic sea ice is stable or increasing can still be rejected at high confidence. Spring extent is fairly meaningless. That’s a period of rapid change so the numbers tend to be similar every year. 2009 is hardly a record max year. The minimum was only higher than 2007 and 2008 and below every previous year in the record. The 365 day moving average remains at 2005 levels and is currently decreasing. The year to date average is barely above 2008 and will lose ground every day the 2009 extent remains below the 2008 level for the same day. At this point, there’s no chance the annual average will exceed 2003 or 2004. The extent in April was about equal to the 1979-2000 average, a period of steadily decreasing extent. So for a month, the extent reached 1990’s level. Break out the champagne!
Re: DeWitt Payne (#74),
The 2009 mid-February and mid-April through mid-May period max values I referred to are in fact max values for the entire JAXA data period, and the ice extent during the April-May period was on or above the 1979-2007 average [ref Nansen:]. This means more ice and a later onset and lower melt early in the season, which probably has something to do with the thicker than expected ice measured by the German/Canadian aerial survey.
So maybe we have a somewhat slower onset of the re-freeze this year. But when the 2009 the max is highest for the 9 year data period and the 2009 min is 25% higher than the lowest value in that data series, it seems to me hard to argue that we are witnessing a catastrophic decline.
Re: tetris (#80),
JAXA April average extent
year extent(km2)
2003 13656865
2004 13104438
2005 13187755
2006 12981370
2007 13035979
2008 13492120
2009 13587025
mean 13292222
std 279374
The 2009 April average is barely one standard deviation above the mean and is less than 2003. The hypothesis that the variation is significantly above average cannot be confirmed. Averaging fewer than 30 days will only increase the variance as well as being obvious cherry picking. Do I need to calculate April averages for 1979-2003 as well? Reliance on the Arctic ROOS graphs for quantitative analysis is not recommended.
Re: DeWitt Payne (#82),
We know you don’t like the Nansen data. For all we know you don’t really like the JAXA data either. The JAXA graph doesn’t show anything on a month-to-month basis, it’s continuous. Your choice of month over month comparison to make your point as opposed to my pointing out mid-February and April-May 2009 absolutes in the JAXA graph tells us very little.
In the 8 year JAXA data series, mid-February and mid-April through mid-May 2009 show the highest ice extent in that period and September 15th 2009 shows a 25% increase over the series absolute low.
You refer to 1979-2003? [1979-2000? or 1979-2007?] I thought it was either 1979-2000 or 1979-2007. Why?. Because it’s by convention, based on the satellite era? There is quite reliable [Danish, Norwegian, Russian, Canadian, American, etc. ] data available from before 1979. So how do you think current JAXA data would measure up to, say, 1899-2009 in terms of showing us a purported continuous and downward trend in Arctic ice extent?
Re: tetris (#83),
Your cherry picking a few points by eyeball from a graph tells us nothing significant in the statistical sense as far as trends. You have to crunch numbers. The question is still has the trend changed. The answer is we don’t know. In other words, we cannot yet reject the hypothesis that the trend hasn’t changed. Is it important that the minimum was higher than 2007 and 2008? Certainly. If 2009 had been lower than 2007 there would be no question that the trend was the same or even more negative. A 25% increase seems large, but in the end, the 2009 minimum was still below the 2002-2008 JAXA average and only the third from the bottom of the eight years in the JAXA record. It’s also still well below the 1979-2006 trend line (graph). And for both the April and May monthly averages (easily calculated if you download the csv file from the JAXA site into Excel), the increase in 2009 is still not outside the expected range. I’ll get excited when I see a point more than 3 standard deviations above the trend line or a run of 8 years of steady increase.
Re: DeWitt Payne (#84), We cannot yet reject the hypothesis that the trend HAS changed. So? You still haven’t justified the use of trends anyways.
Re: MikeP (#88),
What do we have without trend analysis? IMHO Mr. Payne’s analysis is reasonable and intuitive. If 1998 marked a local high and the trend has changed for a time, then it there would be a lag in that trend change getting reflected in ocean heat content and thence in the sea ice. Whther it is true long term (30 year) trend change is an open question. We can agree that it is short term trend change (probably on a 5 year basis) and agree to wait for the long term conclusion.
What I can take away is that we will not have a global warming crisis anytime soon.
Re: Robert (#89),
I do follow DeWitt Payne’s reasoning. That said, when the ice extent minimum in the JAXA series is decreasing and the maximum is increasing to absolutes for a key period in the ice cycle, it seems to me hard to argue that nothing is changing.
As I asked in my previous posting, has anyone looked at the entire 20th century [for which acceptable data exists] and then compared that the 1979-2008 period? If so, what does that tell us?
Re: tetris (#90),
Lucia does very good tests of significance. http://rankexploits.com/musings/
Go back in time to the last statistical significance test that she did for the IPCC projections. She does this hypothesis test so well that it is pointless for me to try to explain it.
Re: Robert (#92),
Here is one:
http://rankexploits.com/musings/2009/july-uah-anomaly0410-c-2nd-warmest-july-in-the-uah-record/
Re: Robert (#92),
I am familiar with Lucia’s posting on this. Very interesting.
Meanwhile did a little digging on the subject of long term data and the issue of trends and found the following paper which touches on some of the key issues here, including the issue of measuring “trends”:
http://www.frontier.iarc.uaf.edu/~igor/research/data/sat slp.php . This Polyakov, et.al’s paper [American Meteorological Society, 2003] “Variability and trends of air temperature and pressure in the maritime Arctic, 1875 – 2000”
Some of their observations:
“Arctic air temperature and pressure display substantial variability on time scales of 50-60 years. The mulitdecadal variability [LFO] is evident in various isntrumental and proxy records for the Northern Hemisphere. This variability appears to originate in the North Atlantic and is likely induced by slow changes in oceanic thermohaline circulation. However, SAT records demonstrate stronger multidecadal variability in the polar region than in lower lattitudes. This may suggest that the origin of this variability may lie in the complex interactions between the Arctic and North Atlantic. Associated with the LFO, SAT record shows two periods of highest temperatures in the Arctic: in the 1930-40s, and in recent decades. In contrast to the golbal and hemispheric temperature, the maritime arctic temperature was higher in the late 1930s-early 1940s tha in the 1980-90s.”
“…over the entire record the warming trend was 0.094C/decade, with stronger spring- and winter-time warming [fig 3]. The Arctic temperature trend for the twentieth century [0.05C/decade] was close to the Northern hemsipheric trend [0.06C/decade]. The oscillatory behavior of arctic trends results from incomplete sampling of the large-amplitude LFO. For example, the arctic temperature was higher in the 1930-40s than in recent decades, and hence a trend calculated for the period 1920-present actually shows cooling.”
“We speculate that warming alone cannot explain the retreat of arctic ice observed in the 1980-90s. Also crucial to this rapid ice reduction was the low-frequency shift in the atmospheric pressure pattern from anticyclonic to cyclonic….. .”
They concluded as follows: “The complicated nature of arctic temperature and pressure variations makes understanding of possible causes of variability, and evaluation of the anthropogenic warming effect most difficult”.
As has been evidenced ad nauseum in the AGW/ACC story, calculating trends from an an arbitrarily but deliberately chosen point [aka “cherry picking”] will generate the desired trend, but in actual fact proves nothing. Interstingly, following the 2007 ice extent minimum [“unprecedented” in the minds of some] a number of analyses supported precicisely what Polyakov, et. al. argue, namely that a shift in wind and oceanic patterns was the cause. If that is correct, in may also explain the increases in 2008 and 2009 s these shifts disappeared. As I have suggested before, a 25% increase over two seasons is not something that can simply be pooh poohed away. Imaging for a moment of it had gone the other way.
Contrary to what some would have us believe, the ice extent situation in the Arctic over the past three decades is not without precedent and the NW Passage was truly open [not sporadically as was the case in 2007, 2008 and this year] during the early 1940s, as evidenced by the two way passage made by the Canadian Coast Guard cutter St Roche between 1940 and 1942.
To the extent that it is at all relevant, I think we are a ways off seeing data points 3 SDs above present values for a decade or so as deemed necessary by Mr Payne.
Re: tetris (#90),
There’s this at Cryosphere Today. I have some problems with their data, though. I’m pretty sure the drop in 2007 is too big. They also haven’t updated since Fall of 2008.
Re: See – owe to Rich (#97),
Aqua (AMSR-E) isn’t having problems. Data are available from JAXA and daily maps and graphs can also be found on the Uni-Bremen page. The East Siberian and Laptev Seas seem to be where the action is at present. Antarctic ice seems to now be below average as well as Arctic ice extent, so the global ice anomaly may be in record low territory.
Cryosphere Today hasn’t updated for over a week now. They had a disk fail on their server and the RAID array didn’t rebuild when the disk was replaced. There’s been no response to my recent request for a progress report.
I keep waiting for 2009 to take off, but it hasn’t yet. In fact, it’s tracking the 1979-2000 average rate right now. That’s not a good thing, though, because the current extent is over 1.6 Mm2 less than the average for this date. 2007 gains over 1 Mm2 extent in the next seven days.
Today’s data:
date extent(km2) difference EWMA rate
10/19/2002 8126563 74844 68014
10/19/2003 8046406 163437 86747
10/18/2004 7852656 76875 65686
10/19/2005 7317500 187500 73134
10/19/2006 7743906 102656 87479
10/19/2007 5991719 202031 76132
10/18/2008 7517813 81307 113504
10/19/2009 7004375 73125 69911
DOY 292 8647027 52352 67440 (1979-2000 average)
I’m using the same baseline average period as Cryosphere Today. There doesn’t seem to be a standard. Arctic ROOS uses 1979-2006 and Uni-Bremen uses 1972-2008.
Re: MikeP (#88),
If you don’t understand why you can use linear trends for hypothesis testing even if the shape of the underlying trend is unknown, this isn’t the place to explain it.
Re: tetris (#83),
“In the 8 year JAXA data series, mid-February and mid-April through mid-May 2009 show the highest ice extent in that period and September 15th 2009 shows a 25% increase over the series absolute low.”
Mid February 3 other years were higher on the Jaxa graph. For mid April to mid May 2009 was high due to cool air from Siberia. However, given that you stated to Dewitt that the current slow freeze rate is not significant compared to around minima time in Sept I wonder why you think a slow melt is significant in April – May compared to maxima in March? That smacks of favouring a period to suite your argument and leads you to be inconsistent.
I guess you’d rather brush the slow refreeze under the carpet as it is annoying 😉
Regards
Andy
Re: AndyW (#102),
I don’t see why the current slow re-freeze should be “annoying”. Data is never annoying. It is what it is. All that is required is enough statistically significant and verifiable data. I am certainly not sweeping any data points under the ice, as it were.
So far, the 2009 re-freeze is not dissimilar from 2005, and until we start seeing sustained evidence of something resembling 2006, I don’t think there is any grounds to suggest that the 2008 and 2009 increases over 2007 are in the process of being reversed.
Re: DeWitt Payne (#100),
Years of unexplained “adjustments” of Cryosphere data [a la GISS] aside, how anyone with a straight face can reference as credible the now clearly corrupted satellite data provided by Cryosphere or any other site using that same data source is beyond me. The satellite has for some time now experienced obvious and probably terminal sensor issues and no data has been made available for the past 10 days. And as you note, they haven’t updated since the Fall 2008. I too have “some problems with their data”. As in fatally flawed and not worth discussing any more. Let’s keep our eyes on the JAXA data, because that is probably the only reliable source we have right now.
And by the way, I thought that one of the stated objectives at CA was to provide others with insights they may not have. So my suggestion is that you explain to Mike [#88] and me and others why it is meaningful to use “linear trends for hypothesis testing even when the shape of the underlying trend is unknown”, so that we all may understand what we ought to understand. That would certainly be better than posting a snarky throw-away.
Look forward to your explanation.
Re: tetris (#106),
Suppose your hypothesis is that ice extent is increasing. How are you going to test that? One way is to take the data you have, calculate a linear trend so you can detrend the data and minimize serial autocorrelation. Then when you add a new data point you can see if it falls within the range of previous data or not. Your null hypothesis is that the trend hasn’t changed. You can only reject that hypothesis if new data is sufficiently beyond the range of the old data that the probability of it happening by chance is small. The confidence limits have to be adjusted for the number of data points, autocorrelation and other possible sources of bias. If you have enough data points, and you need a lot, you can also test the distribution of the detrended data to see if a linear trend is a reasonable approximation. Note that if the actual trend in the data isn’t linear, the data will eventually exceed the confidence limits, so it’s not important what the pattern is for the purpose of the test.
For example: I’ve calculated the May average extent for each year from 1979-2006 using the NSIDC bootstrap data. Then I adjusted the NSIDC data to the scale of the JAXA data using a linear transform calculated from the overlapping daily data from 2003 to 2006. R2 for that regression was 0.9993. After subtracting the mean of the 28 data points, the linear fit has a slope of -30251 km2/year and an adjusted R2 of 0.44. The residual plot shows no obvious trend and the normal probability plot is close to a straight line. A linear fit to the probability plot has an R2 of 0.97 so a linear trend for the data cannot be rejected. The standard error, s, for the fit is 273334 km2. So now I can use the linear fit to detrend (calculate residuals from) new data points. If those data points fall within +/- 2s, then the hypothesis that the trend is constant cannot be rejected with 95% confidence. We have data points from 2007, 2008 and 2009.
year lower limit residual upper limit
2007 -547000 -84978 547000
2008 -547000 128408 547000
2009 -547000 362088 547000
The 2010 JAXA May average will have to exceed 12,471,000 km2 to achieve significance. That’s more than 150,000 km2 more than this year.
I haven’t tested for serial autocorrelation. If it exists, it would expand the confidence limits because autocorrelation reduces the number of degrees of freedom, the denominator in the calculation of standard error. There are other tests as well: the longest run above or below the mean and the number of points in a run of continuously increasing or decreasing values. For a single data point, though, the uncertainty in the calculation of the standard error would make me prefer a difference from the mean of 3s. Or it would take two or three points in a row more than 2s different from the mean before I would state with confidence that the trend had changed. I’ve done the same thing for the September average and the results were similar. The hypothesis that the trend has not changed cannot be rejected. Because the September average is still below the trend line, the 2010 September average would have to go up a lot. Because 2007 was so low, the trend line and standard error must be recalculated and I haven’t gotten around to that. Steve M and Lucia at The Blackboard do this sort of thing much better than I.
Re: DeWitt Payne (#107),
It appears to me you are getting lost in the very nuts and bolts of your argument. Please have another look at Polyakov’s example as provided in #98: it all depends on where you start. If you start your trend analysis for say temperatures in the 1940s you can show beyond much discussion that there has been a cooling trend in the Arctic, while the data since 1900 shows a slight warming [let’s face it, 0.5C over 100 years is not something to get all hot an bothered about].
My sense is that as we get further into the next decade, it will likely become clear that -even if long term temperature data were to show a continued slow increase consistent coming out of the last Ice Age – the decline in sea ice extent between the late 1970s was in fact cyclical and due to natural variability factors [such as oceanic and air variables].
Re: tetris (#106),
I’ve been collecting Cryosphere Today data for more than a year now and I have seen no evidence whatsoever of adjustments of the area or extent data. The correlation to Uni-Hamburg area data, who use the Aqua satellite, continued to be excellent up to the current hiatus, which is a server, not data, problem. Please document your charges with specific examples or retract them.
Re: DeWitt Payne (#108),
Care to comment on the point made in #110? And also the NIDC Antarctic data, which shows increases in Antarctic sea ice extent for the past three years and the possibility of an absolute record max value? Because if there is an increase in Antarctic ice extent, and 2009 Arctic lows are 25% higher than the record 2007 low, you would be hard put to support your rather over the top conclusion that “global ice anomaly is in record territory”.
Re: DeWitt Payne (#108),
I refer you to my two questions in #115. I’m sure others like me, look forward to your comment.
Re: tetris (#128),
And I’m still looking forward to you answering mine in #117
“How do you come to that conclusion when the average daily increase in 2005 from mimima is 70% greater than in the daily increase in 2009 up to now?”
Regards
Andy
Re: tetris (#128),
Your critique of my statistical analysis consists purely of hand waving. If you don’t like it, do your own and post it for comparison.
So CT did make an adjustment. Arctic ROOS has made adjustments as large or larger more recently. The difference is that CT data correlates with independent sources like Uni-Hamburg while Arctic ROOS doesn’t. The comparison to GISS is inapt. GISS adjusts some past data at least every month. This appears to be more like what UAH did in 2005 when they were advised of an error in their algorithm.
As far as global sea ice, did you even look at the graphs I linked? Antarctic extent is now well below average and so is Arctic extent. With CT down, I don’t have access to current Antarctic data, so I won’t know for sure until Uni-Hamburg updates next month.
Today’s data:
date extent(km2) difference EWMA rate
10/22/2002 8450625 110469 79031
10/22/2003 8388594 120625 97232
10/21/2004 7977813 76719 58701
10/22/2005 8023125 220937 114758
10/22/2006 8003438 57813 91014
10/22/2007 6677500 172812 116506
10/21/2008 7930781 180468 120712
10/22/2009 7385781 90937 85180
DOY 295 8878371 88474 70059
Recovery continues to be weak. 2009 extent is now about half way between 2008 and 2007. From the lack of ice in the Kara and Barents Seas as well as the eastern Greenland Sea, it looks like warm water is still flowing into the Arctic from the Atlantic. I suspect the AMO index for October will be even higher than September. The area around Svalbard is nearly ice free, which is unusual for this time of year.
Re: tetris (#106),
“So far, the 2009 re-freeze is not dissimilar from 2005, ”
How do you come to that conclusion when the average daily increase in 2005 from mimima is 70% greater than in the daily increase in 2009 up to now?
You still haven’t answered why the slow start to the melt season in 2009 is more important than the slow start to the freeze season? That was worth mentioning to you, this isn’t apparently. I still think you are being inconsistent because of how you lean, rather than just looking at the data.
Regards
Andy
Re: AndyW (#117),
The later the on-set of the melt, the thicker is the residual ice, which has an impact on so called “second” and “third” year and older ice. This important because we are asked to accept as fact that not only is there less ice, but that it’s thickness is declining rapidly. The latter was proven to be incorrect by the German/Canadian aerial survey earlier this year, which show Arctic ice to be twice as thick as anyone had predicted.
And please stop your insinuations about where I might stand on the matter. I just happen to think that what we are told to believe and the facts do not necessarily mesh.
Re: Michael Hauber (#118),
It is beyond much dispute that GISS data is “adjusted” only in one direction. Only when they are caught flat-footed [as here at CA] will they [quietly] rectify the error, never, heavens forbid, admitting to anything untoward, of course.
Re: tetris (#123),
“The later the on-set of the melt, the thicker is the residual ice, ”
That’s no true, where the ice is melting at the start of the season it has completely melted by the summer, so there is no residue. So it’s no great shakes, indeed 2009 got back all that and more from the slow start and it was actually the poor August weather that scuppered it being as low as 2008.
However, slow refreeze around the edges may reduce thickness come maxima, so perhaps it is more important? However there are other factors involved here so it would be too simplistic to call that either.
Regards
Andy
Re: AndyW (#117),
Let’s have another look at this around mid-November. Since for the entire JAXA series most of the spaghetti bundles up around that time before going their various ways through the winter we may get a clue then where 2009 is heading for the remainder of the year.
From a recent press release from the UK Met office:
CYA? Ice-free is still coming .. just later than we thought. 😉
“reaching an all-time low” – does “all-time” mean all time?
“widely reported” – not by CA bloggers!
“this was caused by man-made climate change” – does this mean that there was not other reason for the reduction?
“due, in part, to unusual weather patterns” – Och!
“enough data to detect a human signal in the 30-year trend” – so what is the “signal”?
“Arctic ice has not yet reached a tipping point, if such exists.” – eh?
“If the rate of global temperature rise increases then so will the rate of Arctic sea-ice decline.” – and vice versa?
My #76 refers to the Met Office press release mentioned by RomanM #75
#77
Although the religious leaders mentioned in this article stopped short of “asking God to reverse the thaw”, maybe others did ask!
http://www.reuters.com/article/environmentNews/idUSL0793503820070907
From What I can see at DMI and Cryosphere and problems with the satellite, I would discont any ice data from 3 weeks ago. Its probably increasing so much thay cant allow it to be seen.. my guess
Re: Vg (#79),
If the growing cold snap in Northern USA is going to settle in then I would buy Gas and Grain futures. (just musing)
regards
From Jaxa 2009 currently is 1.16 million sq kilometres more ice than 2007, 148k less than 2005, and 615k less than 2008. For the last 30 days freeze for 2005, 2007 and 2009 have been all fairly close to 50k/day average. From the 18th onwards both 2005 and 2007 jump up to melt rates of about 175k/day.
So if 2009 continues its slow freeze it would start to open up a big lead on 2005 and start to gain on 2007. However I note models are suggesting that the pattern of a strong low in the NE Atlantic pusing southerlies into the Barents/Kara sea area of the Arctic appears to be ending, and that high pressure with clear skies and lighter winds will dominate the Arctic for the next week, so I guess 2009 looks likely to continue to parrallel 2007 and 2005 with a strong acelleration in freeze. The coming week shows a big low in the North Pacific pushing a decent warm plume north, but I think it won’t make it far enough north to make much of a difference.
Even if sea ice continues to rise by half million km2 per year, 30-year linear trend will be going down for several years more. You can not recover 30 years of gradual decrease by few years increase of similar rate.
What is the second glitch on DMI graph?
http://ocean.dmi.dk/arctic/icecover.uk.php
The glitch is what it say below the graph:
“Due to repeting data fall-out since 1st of October, the sea ice extent calculation can be unreliable. We are working on solving the problem!”
Probably not solved until the Copenhagen meeting is over. I guess…..
True. Model it on a sinusoidal curve with a period of 60 years and of course this is true. This affixation with linear trends for something where there is a lot of evidence for it being cyclical is, well unscientific.
I’ll go even further. My relative lack of concern for the 30 year warming trend is that it is imposed on a millinial cooling trend since the Holocene maximum.
http://en.wikipedia.org/wiki/Holocene_climatic_optimum
The trend is your friend and the trend is cooling.
My point was that you need a physical basis on which to expect a linear change before linear trend analysis makes any sense. Otherwise you run the risk of your analysis unintentionally becoming self satire. Much like Mark Twain’s linear analysis of the length of the Mississippi River. Linear trends were a new tool back then and much abused, especially as a tool to get the public to “do something”. Linear trends still are abused today IMHO, both in the fields of climate and finance.
Some people complained at fitting a higher order curve through the UAH data, because it was “misleading”. It suggested a downward turn to surface temperatures which at that point could not be shown as statistically significant. IMHO linear trends over the time frames we’re talking about are also very misleading.
MikeP: Agreed, It is not possible to apply a linear trend to a chaotic system in the short term. Perhaps not in any term.
Re: mike worst (#95),
I did notice that Hadley took down their 5 year trend shortly after it rolled over.
I am concerned about the continuing lack of satellite data. Are there any other sources which could be used to compare with (say) last year? For example photographs from aeroplanes, reports from Canadian and Russian authorities on the icing in the NWP and NEP?
Rich.
Re: See – owe to Rich (#97),
I have been using this:
http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic
Anyone else getting an error message (Forbidden) trying to access the Cryosphere website?
Re: Michael Jennings (#103),
Yes. Still no progress report from Chapman.
Re: DeWitt Payne (#104),
Site works for me now but no new data since 11 Oct
Being in the data collection business I have a higher trust for data that is adjusted and corrected, and which has slight differences from other sources. The size of the adjustments and the differences from other similar data sources is then a good estimate of how accurate the data is. If the data is never adjusted, and is exactly the same as another similar source then all you know is that when errors are discovered they are ignored and swept under the carpet instead of being dealt with (unless of course the worlds first perfect data collection system has been invented). And that if the data is exactly the same between two different data sources it means that the data has been deliberately fudged to be exactly the same.
Of course if large adjustments or differences crop up then you know you have a problem.
Re: Michael Hauber (#109),
I would appreciate your views on a situation where, in mostly subtle increments, the data is consistently “adjusted” with one particular bias [down in the case of Arctic ice extent].
Contrary to the assertion in #108, there is good evidence [#110 being just one of them] of this being the case with the data published on the Cryosphere site. Just as with another source of data held up as proof of AGW/ACC, namely the GISS land surface temperature database [in that particular case the “adjustments” are always upwards], as has been demonstrated on CA and elsewhere.
Re: Michael Hauber (#109),
A further two questions: what in your view is the significance of the fact that [nothing to do with Arctic sea ice, per se] the original, non-adjusted, UK based CRU/Met Office long term land temperature data, which all along was held up as on of the pillars for the AGW contention, was erased because of purported storage space “issues”. This leaves the scientific community with only the “adjusted” data without the means to verify/understand the basis for the “adjustments”. What is the value of the adjusted data set in practical/scientific terms?
Tetris: You luckyperson A record has been kept of Cryosphere dalliances LOL
http://mikelm.blogspot.com/2007/09/left-image-was-downloaded-from.html
Re: Vg (#110),
William Chapman was kind enough to send me the data from the hiatus caused by their server problems. I asked him about the ‘adjustment’ in September 2007. Here is his reply:
So apparently no data, other than the incorrectly calculated anomalies, were actually changed. I stand by my contention that Cryosphere Today does not routinely adjust data up or down like GISS.
I was going to avoid this issue, but since you brought it up tetris….
The data is not continually adjusted in one way. It is adjusted both ways. For instance GISS was adjusted for a cooler trend when the ‘Y2k’ effect was uncovered by Steve, and it was adjusted for a cooler trend after the September Russia data had been accidently repeated in October. Likewise the NSIDC ice cover has definitely been updated upwards this February after sensor failure caused areas of ice to not be counted.
When data is adjusted for a warmer temperature trend or faster ice decline, critics make the accusation that all the adjustments are biased in the same direction. When data is adjusted to reduce the warming trend or reduce the ice decline rate, the critics will then complain about the bias in the data before it is adjusted.
So where do you think the bias is then?
Re: Michael Hauber (#114),
Michael,
My point entirely: the GISS data was “adjusted” back only after they were shown up, twice, on their underlying upward “adjustment” bias.
Since we are now [none of your doing, mea culpa] in uncomfortable territory, and given that CA and WUWT/surfacestation.org have demonstrated holes in the ground-based GISS and NOAA networks one could drive a semi through without hitting anything, how to explain away the persistent and growing discrepancies between the GISS and the UHA and RSS satellite data sets. Same thing goes for SST data, and why the discrepancy between the Cryopshere and JAXA data?
Tetris,
Uah, GISS, HADCRUT and RSS trends from 1992 agree quite closely. Trend for Uah from 1979 is about 20% lower than HAD,GISS and RSS, which agree very closely. In 1992 the satellite instrument changed and there is a clear step change in the difference between UAh and RSS, so there is a fair possibility that the difference in trend betwen Uah and GISS for anything prior to 1992 is due to this instrument change.
And isn’t arguing about GISS and UAh in this thread a bit off topic? But I don’t want to let what I consider your highly misleading statements pass, and you don’t want to let what you consider my highly misleading statements pass so I think a moderator needs to snip the whole lot about temps.
Well maybe CA should have a closer look at the ice data. LOL
BTW I think DMI Ice is very reliable and that is why they have pulled it because they dont trust current data (for about 4 weeks now)
Re: Vg (#120), How can you say that DMI is reliable? They are pretty clearly having problems as is evident from the discontinuities in the graph since the beginning of October.
Huge increase yesterday and the re-freeze seems to mean business now. It will be interesting to see if (and possibly when) it crosses 2008 again.
Re: Michael Jennings (#122), Don’t read too much into one day. I want the freeze up to happen just as fast as the next guy but one day doesn’t make a trend and 2009 is still second last place and a long way behind third last place. When 2009 joins the rest of the pack there will be reason to blow your horn, but not today.
Re: Daryl M (#125),
Did not realize I was blowing my horn with words like “seems”, “if”, “interesting” but to each his own interpretation I guess. The fact is that the last two days indicate an increase in what has been a rather slow re-freeze to date and that was my point. I will be sure to check with you as to when I should make any further suppositions though.
Daryl That is why they are reliable they dont put up trash data duh…
On the 17th of Oct 2009 had 615,468 sq km less ice than 2008, and 1,1157,813 more ice than 2007.
As of the prelinary data for the 22nd, 2009 has 753,281 sq km less ice than 2008, and 657,188 more ice than 2007.
2009’s slow freeze has been in the Barents and Kara sea sector where 2009 appears to have less ice than any other year for this date in IJIS (2007 is close). Persistent low pressure in the North Atlantic has been feeding southerlies in this sector. Models suggest that this pattern will continue, but as the North Eurasian continent starts to get rather cold, air temperatures in this sector look set to dive a fair way below freezing.
I’ve also noticed that the Baffin Bay area SSTs seems to be getting fairly cold compared to past years, after having a very early melt, and being quite warm through summer. This area had gained some significant ice cover by the end of October in 07 and 08.
Consider a time series of independent, identically distributed data with a mean of zero and a standard deviation of 1. If a new point has a value of -2, then the next point will probably show a large positive difference from that point because the expected value of the next point is still zero. That makes the expected difference plus two. If the data is autocorrelated, however, the difference will be smaller because each new value has some dependence on the previous value. But one would still expect new values to move toward the mean by fairly large steps unless there has been a change in the process. That’s why the increases in the minimum extent in 2008 and 2009 don’t mean all that much other than to confirm that 2007 was something of an outlier.
snip – please stop editorializing and taunting
“Antarctic extent is now well below average”
According to NSIDC, on Oct 22, the Antarctic sea ice extent is above the 1979-2000 average.
The Uni Bremen data does seem to be the odd one out in showing below average Antarctic sea ice. It seems to show that generally the sea ice has been close to or below average the last few years. Climate scientists and pro AGW blogs seem to accept this increase as real as they have responded by offering explanations for what may be causing this instead of questioning the data. Eg http://psc.apl.washington.edu/zhang/Pubs/Zhang_Antarctic_20-11-2515.pdf
If the trend on uni Bremen for the last couple weeks is accurate I guess CT would likely be showing quite close to average for today, and with the slow freeze in the Arctic I would guess that the chance of breaking the 30 year record low for maximum global ice area later this year is becoming significant. The last updated anomaly figure for CT global is around 0.5 million sq km lower than what is required to break this record, but the Arcic 30 year trend is for a faster than average freeze in Autumn as summer min is decreasing faster than winter max.
Re: Michael Hauber (#136), Re: DeWitt Payne (#132),
What is interesting in all of this, is that SSTs, as measured by the satellite data published by UHA for the area between 60N and 60S as a whole, have been falling for the past 6-7 years. If that is correct, than whatever is going on the Arctic must have something to do with what Polykov, et. al. suggested in their 2003 paper. Variations in air temperature just by themselves can not explain variations in either ice extent or thickness, were it only because by its very nature sea ice has 85% of its mass in the water, which is where the calorific exchange takes place. As was surmised by two NASA studies last year, variations in wind and current are likely at issue in the what we have been observing.
Re: DeWitt Payne (#132),
I do not engage in hand waving. I come from several decades of dealing with loads of “bum fluff” [often with my neck on the block at the highest levels, which I might add in passing, tends to focus the mind], and both former and present colleagues would tell you that I have very sensitive “BS detectors”. As I pointed out to Andy [#117], what we are asked to believe based on the AGW/ACC story line in a broad number of areas on the one hand, and what the data tells us on the other, does not mesh. Full stop.
What a lot of folks [conveniently] tend to forget, is that when [a growing body of] data does not support the hypothesis, you modify the hypothesis, not the data, and if the hypothesis can not be modified adequately, you scrap it and move on..
Collaterally, it is not only easy, but many find it quite convenient to get themselves lost in the proverbial “green washers, blue bolts and assorted red nuts” of an argument, quite simply because it spares them the intellectual hardships of digging into and dealing with the synthesis, which is where it all comes together.
Fact is, we have a rapidly growing body of data and resulting understanding, including on crucial Arctic and Antarctic sea ice issues, that not only no longer supports the AGW/ACC hypothesis, but falsifies it. Time then, to look for other explanations for what is going on.
Re: tetris (#138),
You must define hand waving differently than most since you assert that you don’t engage in hand waving and then post two paragraphs of nothing but.
After a fun weekend at the races at VIR, things have deteriorated significantly on the Arctic sea ice front with an average gain of less than 40,000 km2/day for the last three days.
date extent(km2) difference rate(EWMA)
10/25/2002 8636875 42969 73909
10/25/2003 8605938 33907 98090
10/24/2004 8251406 125156 63070
10/25/2005 8360469 130313 121347
10/25/2006 8358125 114219 94552
10/25/2007 7039531 101250 117212
10/24/2008 8307344 131250 121788
10/25/2009 7497656 -30000 70747
DOY 298 9035536 42455 65225
A continuation of the current trend will put 2009 below 2007 by next Monday. That is an observation, not a prediction. The Kara, Barents and Greenland Seas are still not freezing. It looks like the Beaufort and Chuckchi Seas are behind schedule as well. Apparently, warm water is still entering the Arctic from both the Atlantic and Pacific side with the Atlantic side being where most of the damage is being done.
Re: Michael Jennings (#141),
To which sea ice map are you referring? Sea ice maps are often produced from exactly the same satellite data that is used to calculate extent and area. That is certainly the case with Uni-Bremen. If it is a satellite data map then a sensor failure would produce missing swathes of data in the map. I seriously doubt anyone’s ability to detect by eye a difference of +/- 30,000 km2 out of 7,500,000 km2.
Re: DeWitt Payne (#142),
The situation represented in this graph seems to show the freeze still progressing at a greater than “average” rate.
Re: crosspatch (#144),
And by that comment I mean that the red line appears to be converging with the black line.
Re: DeWitt Payne (#142),
By all means keep on crunching the numbers, if that to you means “no hand waving”. Doing what you do may provide you comfort, but day-to-day, or month-to-month calculations on Arctic ice extent are as meaningful as the same thing on the DOW index. They show or “prove” nada. As the cattle farmer would poetically say: “as useless as teats on a bull”.
You keep on focusing on the what [the comfort of the “nuts, washers and bolts”]. As I asked before: ever considered looking a bit harder at the why? The what without the why is useless: data without the context of a hypothesis. Hypotheses without the context of a theory, and a theory without the context of a paradigm. Have a look at Michael Hauber’s thoughts at Re: Michael Hauber (#139), and crosspatch at Re: crosspatch (#140), or Re: crosspatch (#144). Or maybe you would care to respond the argument at Re: Michael Jennings (#146), and what to make of vg’s comment at Re: vg (#147)
Looking for coherent data and arguments vs. “story line” is not hand waving. Maybe it is to you, but it certainly isn’t in the real world. Understanding comes not from [myopic] data crunching, no matter how brilliantly done, but from the ability to synthesize from the conclusions. How do I know? Coming up for four decades of working with often very gifted people, learning to do just that.
Re: tetris (#148),
I don’t have the data to properly formulate and test a hypothesis about why the ice is doing what it’s doing. Show me where I can find data on the flow of water from the Atlantic and Pacific into the Arctic and then we might be able to begin to talk about why. Without that data it’s all just conjecture and speculation, hand waving if you will, as far as mechanisms. A long wave radiative emission to space time series would also be nice.
In the category of “it’s about time”, 2009 extent almost kept up with 2007 today. Cryosphere Today is indeed back up and if you examine the global sea ice area anomaly you will see that it was significantly below -2 in mid-October. It didn’t quite reach the -2.6 Mm2 of 2008 and it certainly didn’t stay as low for as long, but I think that qualifies as being in record low territory as I stated above.
Today’s Arctic extent data:
date extent(km2) difference rate(EWMA)
10/27/2002 8729063 41875 68582
10/27/2003 8686875 45000 86098
10/26/2004 8393281 42968 72241
10/27/2005 8548125 73437 121441
10/27/2006 8461250 34687 93711
10/27/2007 7463281 191562 134995
10/26/2008 8436094 35938 110595
10/27/2009 7804219 188438 86780
DOY 300 9146458 40921 63224
Re: crosspatch (#149),
Did I miss the sarcasm tag? You can’t possibly be serious. Arctic ROOS uses SSMI data. We know there is a problem with SSMI data. QED. Their graphs only look good because they’re heavily smoothed.
If you have a problem with sources that adjust their data all the time, how about this and this. Or are you okay as long as the adjustment is in the “right” direction.
Re: DeWitt Payne (#157),
You are right, a stand corrected. What made me think it was from a different source is that it apparently hasn’t suffered from the recent dropouts that other sources have.
The direction of adjustment is what it is. What is going on at this time of year is of little interest to me personally. What is most interesting to me is where it ends up in March and September, what it does along the way doesn’t really matter.
I still don’t see anything that would lead me to believe that we will have less ice in Sept. 2010 than we had in 2009. The inventory of old ice has increased. The way I look at it, maximum is an indicator of conditions this season. Minimum is generally an indicator of the conditions over the last 5 seasons or so barring extreme anomalies such as the wind event in 2007.
Sorry its getting like RC here I wont poost any further thank you
Preliminary figures for 25th show an 80k loss of ice. Is this another sensor failure?
Or it could relate to some ice that formed in Baffin Bay a couple days ago and may have been blown apart by a low pressure system over that area. A strong low is currently retarding ice growth on the Pacific side of the Arctic, and a weak low affecting the NE Atlantic corner, which accounts for all 3 areas currently suitable for sea ice formation.
If the current seven day trends were to continue, 2009 will overtake 2007 as lowest for date ice extent in JAXA in just over five days, although the 2007 freeze rate slows a little in coming days, and it would seem unlikely that 2009 can maintain a freezing rate this far below normal.
If current 30 day trends continue, 2009 will take another 16 days to overtake 2007.
Re: Michael Hauber (#139),
“Preliminary figures for 25th show an 80k loss of ice. Is this another sensor failure?”
Very likely. Air temperatures are heading down to where they belong. Seeing such a huge melt figure at this time of year would be rather surprising.
Re: Michael Hauber (#139),
Has to be a sensor problem Michael. Go to the map and toggle back and forth between the 24th and 25th and watch the ice change and just eyeballing it seems to me it is certainly not less on the 25th than the 24th
We’ve had the occasional very low or high value before and then next day an opposite very large/smaller value has been put it and the graph keeps it’s normal movement, so I think it is not a sensor failure and more a glitch. If we get a massive jump tomorrow then that’s the explanation.
Regards
Andy
Dewitt Payne said
“To which sea ice map are you referring? Sea ice maps are often produced from exactly the same satellite data that is used to calculate extent and area. That is certainly the case with Uni-Bremen. If it is a satellite data map then a sensor failure would produce missing swathes of data in the map. I seriously doubt anyone’s ability to detect by eye a difference of +/- 30,000 km2 out of 7,500,000 km2.
The JAXA map and yes I, and most people, can tell whether there has been a 30,000 decrease rather than an undeterminate increase for the day. It is impossible to put an accurate number on it + or – but the overall trend for the day is most definitely discernible, or do you disagree?
Well I hope I wont be snipped for providing the truth http://ocean.dmi.dk/arctic/icecover.uk.php
DMI IS back and clearly from October 1 data is silch, they say so… its probably following 2005 or greater and will be shown in weeks to come
After being considerably above normal for the past few weeks, temperatures in the arctic are now diving toward normal and are now within 5C of “average”. I would not trust any data coming from the satellite being used by CT at this point. Even DMI still has their caveat: “Due to repeting data fall-out since 1st of October, the sea ice extent calculation can be unreliable.” so I am going to treat those numbers as just that … unreliable, until they get the problem sorted out. In the meantime, Arctic ROOS seems to be the most reliable data at hand.
Jaxa continues it’s rise again, the small plateau looks similar to the one that happened in 2003 in early October so it is not out of the ordinary.
I actually disagree with the claim that people can look at complete Arctic maps of ice extent and be able to judge the ice increase or decrease. It’s be shown time and time again that people can’t by peoples comments on this thread. Certainly I can’t, I’ve tried in the past and not been able to with any sort of accurracy.
By the way the ice edge actually retreated to the north of Svalbard over the last week or 2 I think.
Regards
Andy
Re: AndyW (#150),
I guess we will just have to agree to disagree here Andy. I am not claiming any clairvoyance here, just that by studying the images side by side or toggling one day to the next, I think I am able to see if there has been an overall decrease/increase without knowing the precise amounts. But hey, I could be wrong.
Well I toggled the 2 days images and thought I saw a decrease. Perhaps our respective belief systems were leading us both to see what we expected to see…
Instead of toggling, why not difference the images? With different colors for + and – bands? If you could mentally allow for the effect of projection on area, you could estimate increase or decrease with a nice steady image to look at.
You’d also have to be able to detect and disregard “clutter”, this might be easy if it is 1000km away from the next bit of ice, but hard if it is next to it. With eyeballing it is hard to remove any favouritism as Michael Hauber says, you tend to notice what you want to notice.
Regards
Andy
We may have witnessed, dare I say it ” a weather event” in the Arctic. Anyone who watched the sea ice earlier this year during the melt year saw it hang inside the 1st SD for a while and drop to near record low extent a bit later.
Neither were important in terms of the trend for the year. I suspect this recent event is also fairly meaningless and the ice will trek its way back to toward 2005.
Cryosphere Today is back up.
How do you know that the inventory of old ice has increased? (not that I have a specific opinion either way)
If you take a look at the Quickscat satellite shots, it looks like the persisent low pressure system in the Barents sea has taken a significant area of older ice from that region and pushed it into the Atlantic Ocean east of Greenland where it will continue to move south and sooner or later melt. The ice left at minimum in September does not stay still and automatically become next years old ice.
http://manati.orbit.nesdis.noaa.gov/cgi-bin/qscat_ice.pl
Re: Michael Hauber (#160),
“How do you know that the inventory of old ice has increased?”
Simply because the season this year ended with more ice than last year. That ice will be 2yo ice next season. There will be more 2yo+ ice at the start of 2010 than there was at the start of the 2009 ablation.
Re: Michael Hauber (#160),
In addition, the winds this year seems to have pushed most of the ice against the North American continental mass during this year’s ablation which would have served to preserve more ice than if the ice had been blown more toward Europe. But in any case, it will be what it is, we shall have to wait and see.
De witt note taken I am a fair person, as long as persons provide a reason no problem. I still say current NH is 100% unreliable and should not be posted as DMI does (they do not post trash data until they are sure the data is realiable)
I have a naive question regarding the pole hole. Since we currently can’t make measurements of sea ice extent or area in the pole hole, why not just orbit closer to the pole? Or is it that the satellite is an old satellite launched with an orbit for other reasons than measuring ice?
7,932,031 for October 28th, pre-adjustment.
I had to get the number by downloading the csv file. 10/28/2009 7,987,344 km2 a difference of 183,125 from the previous day. 2009 gained 100,000 km2 on 2008 but still lost 20,000 km2 to 2007. The smoothed rate for 2009 has finally exceeded the 2003-2008 average. 2007 has some slow days coming up. The year-to-date average for 2009 is still slightly higher than 2009 (2008 10,419,779 km2, 2009 10,447,365 km2). The ice concentration (CT area/JAXA extent) looks just like 2008.
Looking at Cryosphere Today global ice charts, we are close to the normal time of the annual global sea ice max. We are currently about half a million sq km below the previous record in 2002, and have had a small reduction in global sea ice the last day or three. It is of course far too early to call a definite maximum.
October JAXA Average extent
year extent(km2)
2002 7932924
2003 7665877
2004 7683513
2005 7201573
2006 7382636
2007 5998695
2008 7209061
2009 6828594
OLS 2002-2009 trend is still negative, but somewhat less so than 2002-2008, -177236 km2/year compared to -207387km2/year. I haven’t done the 1979-200x trend for October but I will be very surprised if it doesn’t look a lot like September.
That list really shows how 2007 was such an amazing year.
Regards
Andy
Re: AndyW (#169),
November could be interesting. 2006 had a big flat spot in November where the extent was quite a bit lower than any previous or following year in the JAXA record. Was that a portent of what happened in 2007? Insufficient data at the moment. Let’s see what 2009 does.
Hi Dewitt, that plateau is interesting isn’t it, I wonder what caused it? NSIDC didn’t do monthly updates in those days. I wonder if it is connected to events in 2007, I think Anthony Watts had an interesting graph mentioning this feature. Talking of 2007, 2009 has for most of the year tried to ape 2008, but now it’s copying 2007! 🙂
NSIDC have their monthly update up http://nsidc.org/arcticseaicenews/
Quite an interesting read this month with some info on research that shows more storms with low sea ice extent and also mention of more Autumn snowfall in years with lower extent.
Unless we have an episode like 2006 the nxt 2 months up to January are pretty boring, it’s only in the New Year when you can start to tell where the max will be.
Regards
Andy
Well it will be interesting to see if 2009 continues the general recent trend of most years being almost the same this time of year. Or whether it copies 2006, or does something else surprising.
The record for global ice concentration is looking fairly safe now. Just slightly below it and still going up at a decent rate. Would probably require Antarctic melt to get a serious move on for the record to be threatened.
JAXA 2009 extent fell below 2007 today. NH area is probably also below 2007. Global sea ice extent is also falling again. Looking at the ice around Svalbard, it looks like another pulse of warm water has entered the Arctic. That’s probably good for global ocean heat content, though.
date extent(km2) difference EWMA rate
11/5/2002 9200625 71406 62678
11/5/2003 9056879 40938 58684
11/4/2004 9123125 180937 68415
11/5/2005 9191094 60313 90041
11/5/2006 8931094 31719 68739
11/5/2007 8672813 84844 132274
11/4/2008 9106719 117656 86870
11/5/2009 8667656 27187 86510
The arctic temps are running higher than normal, have been for a while.
Re: realist (#174),
Looks like they have taken a nosedive in the past day.
Here is what October looks like:
October (month end averages) NSIDC (sea ice extent)
30 yrs ago
1980 Southern Hemisphere = 18.9 million sq km
1980 Northern Hemisphere = 9.5 million sq km
Total = 28.4 million sq km
Recorded Arctic min yr.
2007 Southern Hemisphere = 18.6 million sq km
2007 Northern Hemisphere = 6.8 million sq km
Total = 25.4 million sq km
Last yr.
2008 Southern Hemisphere = 18.1 million sq km
2008 Northern Hemisphere = 8.4 million sq km
Total = 26.5 million sq km
This yr.
2009 Southern Hemisphere = 18.5 million sq km
2009 Northern Hemisphere = 7.5 million sq km
Total = 26.0 million sq km
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/S_198010_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/N_198010_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/N_200710_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/S_200710_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/S_200810_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/N_200810_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/S_200910_extn.png
ftp://sidads.colorado.edu//DATASETS/NOAA/G02135/Oct/N_200910_extn.png
Another blogger asked me if I thought this year’s seasonal ice recovery (extent 1st month ice growth) appears sluggish or lagging? I thought the following might help answer the question:
1980 Northern Hemisphere (Oct.) = 9.5 million sq km
1980 Northern Hemisphere (Sept) = 7.8 million sq km
Delta Sept\Oct Extent = +1.7
2007 Northern Hemisphere (Oct.) = 6.8 million sq km
2007 Northern Hemisphere (Sept) = 4.3 million sq km
Delta Sept\oct Extent = +2.5
2008 Northern Hemisphere (Oct.) = 8.4 million sq km
2008 Northern Hemisphere (Sept) = 4.7 million sq km
Delta Sept\Oct Extent = +3.7
2009 Northern Hemisphere (Oct.) = 7.5 million sq km
2009 Northern Hemisphere (Sept) = 5.4 million sq km
Delta Sept\Oct Extent = +2.1
1979-2000 mean (Oct.) = 9.3 million sq km
1979-2000 mean (Sept) = 7.0 million sq km
Mean Delta Sept\Oct Extent = +2.3
So in the Arctic, the first month freeze-up ice growth (+2.1) is comparable to the mean first month growth (+2.3). Beyond this… we just have to wait and see. GK
Yow! Looks like Arctic temps have REALLY taken a nosedive according to DMI. They are now (just barely) below average for the fist time in over a month. Looks like current temp is below -20. I would expect to see a rebound in ice recovery over the next several days.
I agree crosspatch, though SST’s are not going to assist the cold air temps.
Talking of which, the SST’s are looking a lot higher in the Pacific equator area link. I guess you global temp people would say that is an El Nino yes? Which makes global temps tend to go higher? So, will that effect our JAXA graph and if so is there a lag, ie will it impact maxima or minima next year or both?
Andy
This is a very odd el nino. We will just have to see how it plays out. Things like this almost have a mind of their own and temperatures are going to depend a lot on the trade winds. What the winds will be doing a month from now is crystal ball territory.
2009 Arctic ice extent is still below 2007. Area increase has been faster, which is normal for this time of year. But the average concentration cannot continue to increase at the current rate so the area rate is likely to fall back to the extent rate in the next week or so. Unless we see a repeat of 2006, we’ve seen the peak extent rate for the year. I went back and looked at the images on Cryosphere Today and since the record started in 1979, there has never been this little ice in the Kara and Barents Seas this late in the year.
date extent(km2) difference rate(EWMA)
11/11/2002 9465156 87187 56696
11/11/2003 9356250 55312 54134
11/10/2004 9582656 74218 77146
11/11/2005 9464844 33125 70233
11/11/2006 9189063 51875 56342
11/11/2007 9185625 68750 110647
11/10/2008 9646094 35625 84862
11/11/2009 8992500 27344 71207
The year-to-date average for 2009 will drop below 2008 tomorrow or the next day.
Re: DeWitt Payne (#180), Assuming that this significant delay in ice formation is due to warm water, does this provide any significant increase in heat flux from the still un-covered ocean? There seems to be mention of this, but I’ve not seen any significant analysis. Regardless of the delay, I’m not sure that the final extent in these areas will be much affected. The bulk of the areas must ‘obviously’ freeze, any variation will just be along the periphery. I’m suggesting that the max. extend is an equilibrium condition wrt seasonal temperature, rather than a rate limited approach to equilibrium before melting resumes.
Re: Sean Houlihane (#185),
Unless the open water is covered by clouds, there should be much higher heat flux to space from open water than from ice. Ice is a fairly good insulator. Once ice is formed, the surface cools rapidly and in the absence of strong wind, a temperature inversion forms.
As far as peak extent, it’s very likely that it will be about the same as always. The problem is that the later the ice forms, the thinner it will be when melt season comes around again. Total ice volume would be the ideal metric. Unfortunately, ice thickness measurements are spotty so there isn’t good data on ice volume. 2007, 2008 and 2009 have shown unusually high peak loss rates compared to previous to years in the satellite record (see this plot for example). That suggests that ice thickness has indeed decreased.
Re: DeWitt Payne (#180),
2009 Arctic ice extent below 2007? We must be looking at different graphs/data sets because that is certainly not what the red spaghetti in the JAXA graph tells us about 2009 vs 2007. And what are we meant to extrapolate from your multi-year same-day comparison?
Re: tetris (#207),
The JAXA graph clearly shows 2009 below 2007 and all other years in early to mid-November. The specific dates are 11/7-15/2009. Here’s the data from the JAXA.
11/7/2007 8829219
11/7/2009 8788594
11/15/2007 9276719
11/15/2009 9260313
Re: DeWitt Payne (#208),
My point entirely. 2009 was over 2005 at a particular point in early October. It is now getting close to the mid section of the entire spaghetti cluster which includes the 2002-2003 highs of the past decade. So what?
As I have tried to suggest to you before on this thread, staring oneself blind on various specific points on a moving graph is bound to lead all the wrong conclusions. Several weeks ago I referred to a seminal 2003 article by Polyakov, et. al. which very clearly shows that temperatures in the Arctic were considerably higher in the 1940s than today, and by extension, if one wanted to extrapolate a trend from that particular point until today, it would be negative. The Arctic is not melting before our eyes and Antarctic sea ice extent has been hitting absolute record values for three years now.
And in passing, the very notion that there is somehow a meaningful measure of the amount of combined ice on both hemispheres is scientific nonsense. Same nonsense as the notion of a “global” surface temperature – the core of the “global” warming meme conveniently morphed into “climate change” – which is as meaningful as calculating the average, mean or any other statistical value that can be derived from the constantly changing NYC/Paris/London/Shanghai, etc Yellow Pages.
Re: tetris (#209),
If you think this thread is pointless, why are you reading and commenting? I’m doing this because I like to play with the numbers. At present, I haven’t made any conclusions. If you have a substantive point, make it. I’m not interested in debating philosophy.
Re: DeWitt Payne (#210),
I would be very hesitant to pass on the link to cryopshere’s estimate of Arctic ice beginning in 1900. The documentation for the graph itself points out that its data sources could be considered unreliable. And I think that is putting mildly. The graph is at contradictions to numerous reports of ice conditions earlier in the 20th century and seems to be created with an agenda in mind rather than scientifically.
Re: An Inquirer (#215),
I agree that the CT seasonal extent chart has multiple problems including the data for 2007 and 2008. But I don’t consider tetris’ reference to be definitive either. I can’t analyze data that doesn’t exist so I concentrate on what I have. As far as starting points for trend analysis, I’m not modeling ice extent so I don’t much care about the magnitude of the trend, I’m looking at whether the trend is significantly different from zero and whether new data is consistent with the trend of the old data, somewhat like a control chart. That makes the starting point irrelevant to me. Because I am using linear regression, I need to keep the time scale of the data small enough that linearity is a good approximation. Going back to the 1940’s, if the data existed, would almost certainly violate that assumption.
Looking at Antarctic ice extent, I think the September monthly average is a better measure of maximum extent than an individual day. Using the NOAA extent data, the maximum September average was in 2006 at 19.360 Mm2. Also, a regression of the time series from 1979-2009 gives a slope that is not significantly different from zero at the 95% confidence level and only one data point is outside the 2 sigma range, 1986 at 17.970 Mm2. I didn’t check for serial autocorrelation, but if it exists, the error bars get wider.
Re: DeWitt Payne (#210),
I am hardly engaging in a philosophical discussion. I am questioning your assertion that 2009 ice extent is below 2007, which it isn’t no matter much you “like to play with numbers”.
I thought I had made my reservations regarding Cryosphere well known, and I am certainly not alone in that. See Re: An Inquirer (#215), An Inquirer’s comment.
Re: tetris (#217),
Sea ice extent according to JAXA for 2009 was below 2007 for the following dates: 11/7,8,9,10,11,12,13,14,15/2009. That is not an assertion, it is a fact that you could have easily checked. It’s above 2006 and barely above 2007 today rather than being in the middle (11/20/2009 9,676,094 km2, 11/20/2007 9,663,594 km2) as can seen more easily in an anomaly plot.
JAXA Arctic Extent Anomaly Plot
How is showing that you were incorrect about 2009 never being below 2007 “my point entirely”?
October 2005 average 7,201,572 km2
October 2009 average 6,828,594 km2
Every day in October 2005 had a larger extent than 2009 with the smallest difference being 69,531 km2 and the largest 862,813 km2.
I believe that makes you the one making unsupported assertions.
Don’t bother to answer, you’ve managed to make my very short troll list. *plonk*
Re: tetris (#209),
” Antarctic sea ice extent has been hitting absolute record values for three years now”
have you got a link to this? That would be out of step with the area values as posted on the Cryosphere web page.
Regards
Andy
Re: tetris (#209),
Also you said
“2009 was over 2005 at a particular point in early October”
which is not true, just wishful thinking.
Andy
Re: AndyW (#212),
Have another look. Same goes for late September. Fact [not wishful thinking] remains that overall 2009 numbers are significantly over both 2007 and 2008.
Re: tetris (#218),
I did have another look, no days in October 2009 had a larger extent in the data set, unless I have misread it, can you tell me which days do?
Andy
Can anyone get Bremen AMSR-E images currently, I get invalid page.
Andy
I get server reset errors from the Uni-Bremen page so it’s likely a temporary problem with their server. The current Arctic and Antarctic images from AMSR-E can also be found at the Uni-Hamburg sea ice page here: http://www.ifm.zmaw.de/forschung/fernerkundung/meereis/amsre-sea-ice/ . There’s a lot of white space at the top of the page so scroll down to get the maps.
Thanks DeWitt that’s just missing the coloured concentration maps, good enough to be going on with until Bremen is back up. Cheers.
Andy
Current, JAXA AMSR-E Sea ice extent.
The latest value : 9,260,313 km2 (November 15, 2009)
The largest areas of ice formation from now until maximum will no longer be within the Arctic Ocean proper (i.e. between Arctic Islands, Greenland Alaska and Russia, but in the North Pacific and SW of Greenland. These areas have had little or no sea ice so far, so conditions could be different. I get the impression that the coldest part of the Arctic (in anomaly terms) has been NE Canada, which is the area where the most ice would form between now and then end of an average winter.
I suspect Ice extent will move closer to average as we approach maximum.
One would expect overall for the thickness to have decreased after 2007 but the trend from 2008 on would tend to increase a thickening.
In 2007 an anomalously large amount of ice was blown into the warmer Atlantic water where it melted. That left the inventory of older, thicker ice at a low point. 2008 saw no such wind pattern and there was greater ice coverage at the end of the ablation season than in 2007 and 2009 saw even more than 2008. So it seems reasonable to expect that ice volume in 2009 was greater than 2008 which was greater than 2007. The current trend, I believe, would be a thickening overall of ice since 2007.
From looking at the pictures and comparing to weather maps, there has been a fairly strong flow of air from the South across Iceland and to the East of Iceland. This will move at some point and the ice will expand Southward. Yes, later forming ice may result in less thickness but I am not convinced that ice increases in thickness at exactly the same rate at all times. We will just have to wait and see what happens.
The last two days of increase, +111,406 on 14 Nov and +156,719 on 15 Nov are quite large for this date compared to past years. It is something of a rarity to have two consecutive days of that much growth in mid-November. In fact, I can’t find one going back to 2002 but I only checked the record manually.
Another big day. November 16 comes in at 9,394,219 km*km +133906 for the day (pre adjustment). Three consecutive days well over 100K at this point in the season is noteworthy.
Re: crosspatch (#189),
The catch up area has to be mainly the Barents and Kara Seas North of European Russian and Scandinavia. The coverage of ice there, as I noted in #180 above was way below anything previously seen. The increase should be showing up on the regional maps at Cryosphere Today over the next few days.
Crysphere general graph has flattened off rather oddly.
It’s completely flat, so I am not sure if there is some issue there
Andy
Re: AndyW (#191),
They haven’t updated their pictures for a week, either.
Re: AndyW (#191),
Cryosphere Today is approximately two days behind, so the flat spot corresponds approximately to the 11/11-13 where the net gain in JAXA extent was less than 30,000 km2 for the three days. The area rate has also been way above the extent rate. That’s normal at this time of year as average concentration increases. But we’re at the point where concentration doesn’t change very fast so area and extent rates should synchronize more or less. But I don’t have a lot of data on concentration calculated from CT area and JAXA extent. I only have about 15 months of numeric CT data.
Re: AndyW (#191),
Some charts here:
http://arctic-roos.org/observations/satellite-data/sea-ice/Arctic-ice-concentration-maps-from-SSMI-and-AMSRE
Extent looks to be headed back to a more normal slope. Increases are about 100,000 km per day for the last three days. It’s hard to tell whether these are real events or measurement errors. Data drop outs and bad information seem to crop up a couple of times a year.
Thanks crosspatch for the link.
Andy
Looks like another day >100K. That’s four in a row and unusual for this time of year. The wind must be settling down.
“Looks like another day >100K”
Current JAXA AMSR-E Sea ice extent.
The latest value : 9,515,781 km2 (November 17, 2009)
Bremen is back .. hooray! Yes, finally 2009 has rejoined the main sequence again after a burst of large increase days.
Andy
Does any of you know whether we just beat the lowest global sea ice as recorded by cryosphere today?
Re: Flanagan (#200),
Probably take a couple of days to find out for sure but there will be loud trumpeting by the Copenhagen crowd if it did. As for 2009, it looks to have passed back over 2007, 2006 and pretty equal to 2003 so far.
The global anomaly graph bottomed well above the past two years. So I suspect that, no we didn’t just beat the lowest. In the money at third, just like many other measures this year. As mentioned above by MJ, the deafening silence is confirmation of this.
I realize it’s not sea ice – it’s land ice. But it’s close to the sea ice of question in this thread. It’s showing a pretty good trend of melting over the past decade or two…
http://www.sciencemag.org/cgi/content/abstract/326/5955/984
Re: bent-out-of-shape (#203),
This paper received a good bit of media attention in Europe last week, because of the hype about massive sea level rises and the media’s need for something to moan about in the run-up to the fizzling Copenhagen confab.
Until we have a proper understanding of their real value, Grace data are best taken with a healthy pinch of salt.
That’s Grace data, not direct measurement. Grace data is still a “work in process” and estimates have and will continue to change. I wouldn’t be surprised if, in the end, they even got the sign wrong.
“That’s Grace data, not direct measurement. Grace data is still a “work in process” and estimates have and will continue to change.”
Yeah,
Well it appears they got the melt data wrong for the Antarctic.
West Antarctic ice loss overestimated by NASA sats
http://www.theregister.co.uk/2009/10/20/antarctic_ice_loss_overestimated/
Re: Marian (#205), I notice that the article repeatedly refers to ice loss across Antarctica but the work appears to be based in West Antartica and the peninsula.
The NSIDC has shown Antarctic sea ice to be well above the 1979-2000 mean for most of the three previous Antarctic winters. Each of the past three maximums have been at or near records highs for extent.
What is this about Antarctic sea ice maximum being near record last 3 years? Judging by CT 2007 was record, 2009 is high (roughly top 5), but 2008 was below average. NSIDC also shows Antarctic sea ice maximum below average in 2008.
Re: Michael Hauber (#219),
Data generated by the NSIDC [hardly a hotbed of AGW skepticism] shows that 2007, 2008 and 2009 are all above the 1979-2000 average. Antarctic sea ice extent has been stable with a slight long term growth trend and is not collapsing before our eyes as some would have us believe. Even the IPCC acknowledges this.
At the risk of being told that I am “philosophizing” [as in #210 above], I remain less interested in the “what” than in the “why”, and think it is useful to bear in mind that no matter what AGW/ACC proponents want us to accept as proven fact, the role of humans in the observed changes in NH and SH sea and land ice, including glaciers, remains an open question. Tellingly, not only did by far the largest portion of glacier retreat observed at Glacier Bay, AK occur between the 1790s and the early 1900s, some of those glaciers are currently expanding again.
Re: tetris (#220),
Did you read my post? I was not trying to claim that Antarctic Sea Ice was collapsing, or that it was not generally above average the last 3 years.
I was asking about data to support the claim that 2007, 2008, 2009 were near record maximum. In particular I was responding to your post 209: ‘Antarctic sea ice extent has been hitting absolute record values for three years now’, and Realist in post 214: ‘Each of the past three maximums have been at or near records highs for extent.’
The NSIDC data which you appeal to shows that around September 2008, the time of Antarctic Sea maximum, the sea ice was clearly below average according to NSIDC.
The only number that apparently matters is the minimum. What it does on the way down and on the way back don’t seem to matter. Apparently maximums don’t matter either (except for when they do by validating someone’s meme).n Once you get to January, the spread between years becomes fairly small. There is quite a lot of spread during ablation and recovery and the amount of ice a maximum during any given year has little relation to the amount that will be left at minimum.
So far I still see no reason to change my call that next year’s minimum will be greater than this year’s was as being the most likely outcome with next year probably looking closer to 2006 than 2007 barring unusual wind conditions. But that can change. It would appear to me that the old ice inventory is still recovering from the 2007 event which I expect to take 5 years to fully recover from.
Re: crosspatch (#223),
I made a similar point earlier on on this thread. Somehow DeWitt Payne doesn’t like the reasoning and keeps on throwing day-to-day data points our way to prove otherwise. On top of that I’m now on his short “troll list”. Better go do some penitence.
Current, JAXA AMSR-E Sea ice extent.
The latest value : 9,953,281 km2 (November 23, 2009)
Nice uptick today. Almost into the territory of 2003. Arctic temperatures are still well above normal on DMI. I wonder how much of a factor that has been on the ice growth this year.
Seems like our thread is getting back up to speed. Those dodgy scientists in East Anglia causing Steve’s server to suffer 😀
Andy
November 23rd has been revised to 9,994,219.
Finally made it over the 10 million km2 Mark. It does appear to be in around 2003 range for this time of the year.
The latest value : 10,082,656 km2 (November 24, 2009)
Re: Marian (#229),
I see 10,163,750 for 25 Nov. Not too shabby. So it is bunching up with several other years. Looks like a fairly “average” season so far for the period since 2002.
Re: crosspatch (#231),
Looks like 25 Nov was adjusted up to 10,174,531 from 10,163,750
Unless anything really unusual happens, I will see you all in February.
Have a nice break crosspatch :waves:
Looking at the maps, is there normally that much free water north (NE) of Svalbard at this time of year? I’m trying to think back. I’ll have to check the ice video for the years when I get home. Temps in Siberia now getting properly chilly. I can’t imagine what it must be like at -44C. It must hurt your eyes and breathing. What’s the coldest people have been here? Mines -19C in Norway.
Andy
Thanks for a great thread. As a total layman I have nothing to contribute. Until now:
– 31, northern sweden. No problems for eyes or breathing, its your hands and toes…;)
AndyW: you can still use the cryosphere comparator to get a rough idea
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=11&fd=26&fy=2008&sm=11&sd=26&sy=2009
in 2008 at least there was much more ice there.
#234
Svalbard and Barrow, AK, but it was only around -5 to -10C. We experienced -25 in Glasgow around Jan 1997, when I just returned from Oman (+25C at the time, though I experienced +50C there).
-30C in my part of China and yes fingers and toes suffer. Although there was one night where it was -28C with a 20 knot wind blowing. Chill Factor probably put the temp at -50C and my ears hurt after 5 mins. What made it worse is that I wear glasses and I had to take them off, The cold metal on my face really did hurt. 🙂
Does anyone know why JAXA is updating some much earlier lately? I realize we switched away from DST, but it seems that the updates are considerably more than an hour earlier in the day than they were not that long ago.
Having been originally from Edmonton, I have experienced the -40s many times. I think the coldest official temperature was -47C, not counting the wind chill. I would imagine that counting the wind chill, the effective temperature would be well into the -50s during a windy storm. It is very unpleasant. Exposed skin can freeze in minutes. Fingers, toes, ears and nose are all very susceptible to frost-bite. It is painful to breathe and any facial hair (even eye lashes) builds up frost. The strange thing is that after a spell of -40s or even -30s, when the temperature comes back up to the -20s, people are walking around with their jackets open as if it’s warm. Thankfully I don’t live in Edmonton any more and if I never have to experience -47C again, I won’t be disappointed in the least.
Re: the recently released Copenhagen Diagnosis and sea-ice.
“The observed summer-time melting of Arctic sea-ice has far exceeded the worst-case projections from climate models of IPCC AR4.” (The Copenhagen Diagnosis, p. 31)
When I look at the graph with plots of the models and actual sea-ice measurements (fig. 13, p. 32), I note that data from actual observations have only come close to the model mean once in the last thirty years. Worse than that, going back to 1980, the observed measurement has been outside the ‘worst-case’ projection 19 times, with only 10 years where the measurements and models intersect.
Perhaps someone in this discussion can help me answer this question:
When the model seems so consistently wrong (even when modeling the past I assume), why compare actual measurements to the ‘model’?
(This is an honest question)
Oh, re:240
I should include this link to the data I am refering to. It is missing the last few years of data, but I think most here know what it looks like:
Copenhagen Diagnosis:
http://www.copenhagendiagnosis.org/
Re: Dan (#241),
does this curve exist with a longer extension into the past? The version you link to looks somehow like it starts “conveniently” near a local maximum (and thus the 2000s might still be nothing unusual in the long term).
Re: ChrisZ (#242),
Well actually, my second question is where did the observed data come from on the figure (fig. 13, Copenhagen Diagnosis) pre-1979? It seems to be the only data that fit the models.
Dan,
The source for Figure 13 is apparently Crysophere’s estimate for Arctic Ice from 1900 to 1979. The fact that Copenhagen Diagnosis use this data source undermines any credibility that the Cop Diagnosis might have. Cryosphere in its graphic caption points out that it used data points that might be questionable. Also, well-documented actual observations are at odds with the estimates.
Regarding the deviation of observations vs. model results, there is an easy explanation — the model is inappropriate! It is like shouting that the volcano’s lava flow is worse that we thought! Our models say that we need to sacrifice more virgins!
November data is in for JAXA. Missing data in early years was interpolated.
year average(km2)
2002 9904932
2003 9662230
2004 9870563
2005 9736620
2006 9254391
2007 9337073
2008 9861724
2009 9390198
graph
November 2009 was barely above 2006 and 2007 and was below the 2002-2008 trend. The hypothesis that Arctic ice extent is still declining cannot be rejected.
Re: DeWitt Payne (#244),
My apologies for “trolling” in.
If you provided us with same 2002-2009 comparison for all the months of the year, you would no doubt see a somewhat different picture about purported declining Arctic sea ice. Then again, the JAXA graph already does that.
I know you have your misgivings about Nansen Roos, but their graphs do show 2009 summer ice extent above the 1979-2007 average and current extent numbers to be within 1 STD.
Re: tetris (#245),
“I know you have your misgivings about Nansen Roos, but their graphs do show 2009 summer ice extent above the 1979-2007 average”
Can you link to this graph or data?
Also, can you link to your claim that Antarctic ice extent has been hitting record values for 3 years now? I requested before as did Michael Hauber.
Currently your record is of making claims and then not being able to back them up. Same as you claimed “2009 was over 2005 at a particular point in early October” and then couldn’t back that up with evidence either.
Are you just making these claims up? If so then please desist as it not adding anything to the thread. If you do have links to data then please put it up as it will be interesting as it seems to clash with other sources charts. Thanks.
Andy
Re: AndyW (#246),
a quick googling comes up with this:
http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic
but these graphs don’t seem to show what tetris claims. During May 2009 only, they show a very slight “above average” level. Currently, we seem to be still slightly below one SD of the average (although markedly above 2007 which really seems to be a negative outlier rather than a “tipping point” or the beginning of a trend).
Since there is no 2005 data at all in the graphs I am linking to however, I suspect tetris is looking at yet another page, which I still hope he will have the kindness to inform us about soon.
My post after 12/3 got lost. 2009 was above 2005 for a brief time in September not October. You can see it in the JAXA graph. However, the September average for 2009 was lower than 2005. Today’s number, 11,037,813 km2 puts 2009 just above 2007 and below all other years in the JAXA record for this date.
Current JAXA AMSR-E Sea Ice Extent
The latest value : 10,583,125 km2 (December 2, 2009)
The himalayan glaciers are melting too …..not
http://news.bbc.co.uk/2/hi/south_asia/8387737.stm
Thanks for the link to the bbc article, vic.
“The himalayan glaciers are melting too …..not
http://news.bbc.co.uk/2/hi/south_asia/8387737.stm”
Lovely quote from the article:
“The IPCC relied on three documents to arrive at 2035 as the “outer year” for shrinkage of glaciers… Incidentally, none of these documents have been reviewed by peer professionals, which is what the IPCC is mandated to be doing.”
This would be funny if not so sad, as the IPCC (in 2007) uses non- peer reviewed material to make alarmist claims. Does anyone really think the Himalayan glaciers will be gone by 2035?
A simple calculation helps to set out the picture. Himalayan glaciers are estimated to contain 12,000 cubic kilometers water. They are said to feed the large Indian rivers. Brahmaputra (610), Ganges (380) and Indus (90, a small river) have a cumulative annual runoff of slightly above 1000 cubic kilometers water per year. So, if these rivers were fed by melting glaciers alone, they would have depleted the total ice store in just 12 years.
We, however, know that people have lived in these river valleys for thousands of years. This means that both the runoff AND the standing ice store in the glaciers are maintained by continuous ocean-to-land atmospheric moisture transport. If this transport process halts, glaciers will disappear and rivers will run dry in a few years even at unchanged or decreasing global temperature, simply (in the limit) at the expense of non-zero melting rate at zero precipitation.
The water-stressed people of India should, in my view, be most concerned about understanding the nature of atmospheric flows that deliver moisture to their part of Eurasia.
Some posts seem to have gone missing on the location move and now the Himalayas are now in the Arctic .. Steve what have you done? 😀
Looks like that was a pretty unusual couple of days of ice extent loss for this time of year. 2006 and 2007 are now lower in extent, the other years higher. It will be interesting to read about it in the NSIDC monthly report, however the instrument they use, and/or the concentration level they use and/or the fact that they use a 5 day average for this graph
means that graph has not picked it up, though it is heading more horizontal.
Andy
I do hope this is the current see ice thread. I have seen elsewhere claims that the antarctic ice mass is declining based on the GRACE satellite pair. Does anyone know if there is data online, or any papers online. I can find articles, but not papers and as we know the deveil is in the detail.
Cazenave, et. al. 2009 is primarily about sea level budget, but it does have GRACE information on Antarctic and Greenland ice.
This does seem to be current but some October comments seem to have dropped to the end – probably due to the server move.
“Pardon me for speaking to your face” (J.Jimenez, 1960) but can some ‘expert’ here kindly explain the general conceptual relationship between summer sea ice melt and sea level rise. Please follow that concept with a description of mid-winter, refrozen sea ice and its impact on sea levels. If in fact mid-winter sea ice levels are roughly constant, is the related sea level issue primarily a summer-time problem?
Sea ice has almost nothing to do with sea level because it’s floating in the water. Melting of land based ice like the Greenland and Antarctic ice caps or mountain glaciers will raise sea level however. The GRACE data has to do with total ocean and land based ice mass and their changes over time.
These old posts need to be pruned from the bottom.
The atlantic side is very interesting at the moment, here is the latest SST’s
http://weather.unisys.com/surface/sst_anom.html
The anomaly is actually getting bigger in that region in the last month or so.
Andy
The North Atlantic positive anomaly does not surprise me in the least, given the weather pattern in November. There was a very persistent depression west of the British Isles which spread southerly winds over the region and on upward to Iceland, which will have kept the sea (relatively) warm. It also produced northerlies in the north-west Atlantic which led to negatives there.
The pattern changed a few days ago, with high pressure near our Isles. I would expect the anomalies and freeze patterns to change therefore. But what do I know? I’m not putting any money on it…
Rich.
The error bands in the plots of Arctic sea ice extent and area at the Arctic-ROOS site are misleading. Assuming they just averaged the raw data for each day, the distribution of the residuals from the mean is far from normal because there is a trend in the time series. That trend also increases the calculated sd. Also, including recent years in the average pulls the average down and makes it look like recent data is not all that far from the average when there is in fact a much larger change from the beginning to the end of the record.
JAXA extent today is below even 2007. Some may say this isn’t important because it isn’t the max or the min. If you think this, ask yourself honestly, would you still think it was not important if the extent today were above all previous years?
I think the fact that TODAY the 2009 ice extent is equal to or slightly below 2007 and 2006 is mere hand waving. To take one day, one week, or even one month, as opposed to the maximum and minimum end figures, and try to draw a conclusion is foolish. 2009 had been substantially above 2006 and 2007 for a while before taking a dip this past week and I saw no posts claiming that was an indication of anything important and I feel the same way about this brief drop. There will be fluctuations weekly from here until the maximum is reached in March and if it continues all during that time to remain below 2006 and 07, then maybe, just maybe it will have some significance. Until then it means squat as long as it doesn’t go several standard deviations lower from other years with a precipitous
Oh, but it isn’t just one month or one day. The thirty year trends for both the March, September and annual averages are negative and significantly different from zero. The annual average for 2009 is on track to fall right on its trend line.
<a href="http://i165.photobucket.com/albums/u43/gplracerx/NOAAMonthly.pngGraph.
I think there’s a real possibility over the next decade or two for the trend to change sign. UAH NoPol temperature may be trending down. The AMO index may be going negative. But there is, in fact, no sign that this has affected the trend in Arctic ice extent yet. Current fluctuations are just that, fluctuations about a longer term negative trend.
Forgot to close the link:
or Graph
I can’t help but agree with DeWitt. How can you argue with the AMSR-E numbers? When the 2009 minimum came out ahead of 2007 and 2008 I was very hopeful that the 2009 would make stronger gains through the freeze up and make a convincing statement that the trend has become less negative, even if it did not reverse. It has obviously not been doing that to any significant degree. I wonder if the temperatures have anything to do with it. If COI is to be believed, the temperatures have been oscillating from 5-10 degrees above the average. It will be interesting to see what happens over the next few days. Will 2009 stall like 2008 did or will it pull ahead and catch up with 2002-2004?
After a strong few days 2009 has pulled in to 4th place behind 2004 according to AMSR-E. If it can keep up this pace for a couple more days, it will surpass 2004.
I’ll go out on a limb, based on wishful thinking, not science, that 2009 will finish the calendar year in 3rd place.
There have been a number of news articles about ice bergs, such as the huge one that may hit Austrailia, and some seen around New Zealand. Anyone have any data or reference to numbers of bergs vs temps or sea ice extent?
Seems to me that the greater the ice extent the higher the number of bergs generated since the ice would be pushed farther from the coast of Antarctica and more likely to calve off bergs. Secondly, the colder the sea temps the longer the bergs would survive letting them drift farther north. In essence these reports would imply that the SH is in a strong cooling phase rather than warming
I think you are right Barry. The more the ice increases, the weight of the outer shelves becomes too heavy for staying intact, causing them to break off into huge icebergs. With the PDO now entering it’s cold phase and the AMO to follow in the next 5-10 years, look for a reversal of the warming we have seen in the past 30 years. Also, the figure for ice extent from JAXA yesterday does not look quite right to me (if the figure for 12/12 is correct) because at 11,297,969 km2, it shows an unusual 170,000+ increase for the date.
A three day gain of 400,000? Something seems amiss for such large gains this time of year
They got stuck on 12/10 for a number of days, briefly updated to 12/13, went back to 12/10, then finally 12/13 followed by 12/15. So the increase was not over 3 days, it just seemed that way because of how they updated.
Not to belabor the point but on 12/12 the total was 11,126,000 and on the 15th it is 11,526,000 so in three days it increased 400,000 unless I misread the figure for the 12th. The total on the 11th was 11,064,531 so the 12th figure looks accurate.
The value on 12/10 was 11,047,… The graph doesn’t look out of line with other years, though. I think the numbers on specific days can be questioned, since they had some difficulties updating and thus may have wound up doing kind of backwards fill-in.
They interviewed an Australian scientist about these icebergs on BBC radio a while ago and it seems to be due to increased winds flowing around the Antarctic and or associated increased currents rather than it being colder or warmer. This pushes larger icebergs around the Southern ocean at a faster pace so they go further before they melt.
The increased wind patterns are also the cause for the positive anomaly of Antarctic sea ice growth from March onwards
along with the geography of the land.
Andy
It seems we have lost a lot of postings unless they are somewhere else!!
So to pick up the thread
11,648,750 km2 (December 16, 2009)
whoops – now why didn’t I see those posts marked Dec 16th?? Blindness is setting in I suppose 🙂
Retospective nesting of comments has destroyed the flow of this post… making it look like last post was Oct 23rd… is there anyway to reverse the nesting? There must be lots of other strange consequences throughout the blog making it much harder to read.
Steve: there are many aspects of standard WordPress that are less satisfactory than “old” climateaudit.
Now using Firefox instead of Explorer and “upgraded” with CA Assist. Comments now appear in correct order and nesting acceptable. Nice to have the Quick Tags back as well. Huge improvement! Thanks, Steve!
The big jump in extent and area recently is because Hudson Bay finally began to freeze over. That’s now almost complete. Meanwhile losses in the Baffin/Newfoundland Sea are balancing gains in the Kara and Barents Seas. We should see a flattening of the trend soon. Cryosphere Today, being two days behind JAXA, should show higher than average increases for the next two days.
Firefox with the Greasemonkey add-on running the CA Assistant script is a big improvement. Getting Quicktags back alone is enough to justify installing it.
Siberia is slightly warmer than previously, also the area to the E and NE of Hudson bay seems rather warmer, due to the southerly winds due to the H over Greenland perhaps?
As pointed out above the topology of the land means that soon we will be into the Atlantic v Pacific side remaining to fight it out on max extent this year. Last year the Atlantic side was still filling up with ice in late Feb and March so one of the reasons for the slow start to the melt season.
I still have October posts at the end of December, I am using Google Chrome… not sure how to get rid of them, or if it is indeed possible?
Andy
Andy
For those of you who like to track ice in other places, I find this graph from the Canadian Ice Service for the Great Lakes to be interesting. Their methodology isn’t so important as is the comparison from one year to the next. As long as they use the same method from year to year, there should be at least a relative idea of how one year differs from the next.
It updates weekly through the season. Ice on the last count was about twice the median amount for this date.
Here are a few notes on the reported changes in Arctic sea ice extent around 1979.
Cryosphere Today offers a sea ice extent time series stretching back to the Nineteenth Century:
The period I’m interested in is 1978-79, which seems to take a step down at that time.
Satellite observations of ice began at about that time, which offered better (or at least supplemental) data to the analysts. My question is, was that 1978-79 change real or just an artifact of a change in analysis?
A 1987 McGill review of sea ice ( http://www.geog.mcgill.ca/gec3/wp-content/uploads/2009/03/Report-no.-1987-8.pdf ) doesn’t seem to show the bump (annual extent anomaly plot is towards the back and is reproduced here http://davidsmith1.files.wordpress.com/2009/07/1226094.jpg ).
A plot of the National Ice Center ( http://www.natice.noaa.gov/index.htm ) averaged biweekly values (averages of the 27 values shown for each year) ( http://davidsmith1.files.wordpress.com/2009/07/1226096.jpg ) shows nothing odd about 1978-79.
So, why the bump in the Cryosphere Today data? Dunno. Perhaps someone familiar with ice analysis and the data used by CT and NIC could comment. I asked CT about this some months ago but heard nothing back. I may be inadvertently comparing apples and oranges.
Does it matter? No, it does not if the time series in use begins in 1979. But, some analyses stretch to the 1950s (see http://davidsmith1.files.wordpress.com/2009/07/1226093.jpg from http://www.ucar.edu/news/releases/2007/seaice.shtml ) and the 1ate 1970s change seems to play a role, at least in appearance.
Hi David,
There is a jump down there on the Cryosphere plot, but it doesn’t look that significant to me from just eyeballing it?? The main change seems to be in about 1952 when the chart seems to start it’s decent.
Getting back to 2009 looks like we will be entering 2010 amongst the lower 4 traces on the JAXA graph, but that is no guide to where the maximum will be of course, considering this year and 2008.
Andy
Must be an algorithm adjustment going on because it is stuck on the 29th
Cryosphere Today has put new graphics on their site (including numbers) which makes it much easier to read, check it out.
Yes, the 2 yearly graph is better in length, not sure I am happy about the thickness and colour though from a aesthetic viewpoint. However the value on the graph is a good addition.
JAXA now jumped to 31st Decemeber, perhaps they are still deciding on which colour to have 2009 in now that 2010 will go red 😀
Cold here in the UK at present which will last through the week. Met office long range forecasts are as dismal as their short range are good.
Andy
Now that JAXA is finally up to date, here’s the year end report for 2009:
annual average (JAXA)
year extent(km2)
2003 10839129
2004 10657076
2005 10344909
2006 10220438
2007 9965810
2008 10460820
2009 10430927
Fall average (Oct,Nov,Dec)
year extent(km2)
2002 9909654
2003 9741750
2004 9790748
2005 9486211
2006 9318382
2007 8903869
2008 9573534
2009 9235566
December average
year extent(km2)
2002 11890953
2003 11894577
2004 11820741
2005 11528518
2006 11316054
2007 11389814
2008 11659113
2009 11492893
Max extent
extent(km2) date
14844063 3/21/2003
14360313 3/10/2004
14098906 3/6/2005
13782344 3/11/2006
13945625 3/10/2007
14516875 3/9/2008
14412813 3/5/2009
Min extent
extent(km2) date
5646875 9/9/2002
6032031 9/18/2003
5784688 9/11/2004
5315156 9/22/2005
5781719 9/14/2006
4254531 9/24/2007
4707813 9/9/2008
5249844 9/13/2009
There is still no confirmation that there is a change in the twenty year trend of steady decline in extent and area. Two years of increase in the minimum extent and area from an unusually low year is not sufficient.
Thanks for that DeWitt. You can see the effect the slow start to the melt season had on the yearly average, ie kept it relatively high. It will be interesting in 2010 if the summer minimum continues to walk back to the general decreasing trend from the exceptional outlier year of 2007 or not.
Considering the AO is negative at the moment I wonder if that will have any effect on maxima figure. Presumably it is warmer up there if the colder weather is coming out and south?
Andy
Re: AndyW35 (Jan 6 01:11),
I’m of the opinion that the AMO has more influence on sea ice extent than the AO and it’s been positive since June 2009.
Looking at the 2009 year end Jaxa sea ice data (See post of Jan. 5, 2009) along with the previous 6 years does not exhibit much fluctuation from year to year or across the time span shown, except minor blips in minimums. Annual data shown in the other charts appear within 5% which is probably within instrument accuracy. If nothing much has happened over the past 7 years in (1)annual, (2) fall, and (3) december average, as well as (4) max extent, why all the arm waving over the sea ice issue?
Re: Whatsamattame? (Jan 6 19:50),
The record low minimum in 2007 got a lot of people’s attention. The Chicken Little’s claimed the Arctic would be ice free at the minimum by 2013. Cooler heads thought 2007 was something of a fluke and that extent would either recover to the long term trend or possibly even start to increase again. However, the data are noisy and it will take years or even decades before we know for sure. There’s also the problem that there really isn’t any reliable data from before the 1970’s on total extent and area so we don’t really know whether the current situation is highly unusual or not.
The other thing is that models (and logic) predict that temperatures rise and fall faster at the poles than at the equator so sea ice is thought to be something of a canary in a coal mine. If sea ice weren’t decreasing, then it’s hard to believe that global temperature is rising.
Finally, some of us just have too much time on our hands and picked this subject. Speaking of too much time on one’s hands, check this out (completely OT).
At this link you will find all 30 years of data plotted on a single graph – the 5th plot from the top of the thread.
Where did my last comment go?
Andy
The 2010 plot seems a lot less wiggly than most previous years.
I was assuming that summer minimum next year might be assisted by the warm conditions leading to less thick ice over winter, however checking the latest monthly update from NSIDC they say that multiyear ice finds the current wind conditions favourable for keeping a lot of this. Given that I therefore expect there may be a large melt on the Russian side next year where the ice is mainly young and less so on the Canadian side. So, first prediction, the Northern Passage will be open again but not as convinced about the NW passage.
Andy
Re: AndyW (Jan 16 05:51),
Give it time. We only have two weeks of data in 2010 so far.
The AMO index was, as expected, still positive in December. That’s going to keep ice in the Barents Sea low and cause early melting on the European/Russian side of the Arctic as long as it persists.
True on not much of 2010 as has transpired but even so only 2004 approaches the smoothness. Not that you can read too much into it apart from an interesting observation though.
Any thoughts on Maxima extent yet? I think the current conditions you stated and also the fact that it started off 2010 fairly low in value means it won’t be much more than 14 million dead and perhaps lower.
Andy
Re: AndyW (Jan 17 04:59),
Based on the JAXA 2003-2009 daily averages, the projected maximum today is 14.13 Mm2 compared to 14.41 in 2009, ahead of 2005 (barely), 2006 and 2007. The recent high was 14.84 in 2003 and the low was 13.78 in 2006. The NSIDC 1979-200 average scaled to JAXA maximum is 14.66 Mm2. The NOAA March trendline, not scaled to JAXA, predicts 14.85 Mm2 for 2010 compared to 15.16 Mm2 in 2009.
I’m betting that the AMO index for January is going to still be positive. Ice is disappearing in the Barents and Kara Seas.
Rather an unusual body of fresh water behind Novaya Zemlya size wise it must be said. Also on the Pacific side the ice in the Bering sea is odd shaped as well, I seem to recall it is normally more hemispherical but currently it is favouring the Alaskan side for growth.
Andy
The decrease over the past two days merely shows how little air temperatures have to do with Arctic Ice. It is much more influenced by ocean currents and amount of sunlight it receives.
Plus wind and SST’s. I’d be interested to know whether SST anomalies have greater effect in summer or winter relative to the other factors such as air temp.
Andy
I haven’t seen an update in a while with everyone concentrating on climategate. How’s sea ice going? Huge build-back this year as everyone predicted???
my bad, I scrolled to the bottom of the thread and it has two posts from oct 2009. I thought no one was updating it, but I was wrong.
ATTN webmaster: the thread is messed up.
Re: bent-out-of-shape (Jan 25 13:02),
The thread looks a lot better if you use the CA assistant Greasemonkey script for Firefox. That adds back the html quicktags buttons and allows viewing either as linear order or nested. Look in the upper right corner of the page for the link to the instructions.
Re: bent-out-of-shape (Jan 25 11:01),
Not so much. Today’s JAXA extent is 13,104,688 km2. That’s below all years from 2003 on except 2005 at 13,058,906 km2 and 2006 at 12,691,719 km2. The anomaly from the 2003-2008 average is -200,338 km2. The end of 2009 and the beginning of 2010 have been less than impressive. So far it looks like a return to the thirty year trend of -50,000 km2/year at the peak rather than a recovery.
I would not normally comment on this thread but try to keep my eye on the latest data & comments for my own interest.
so, can anyone advise me on :-
1 I use xp i.e. & since the CA move find it difficult to read/display comments (found it easy on old site),can i use CA assistant?
2 find it difficult to get to this thread & for example ‘unthreaded’ if not in ‘recent comments’, how do you get there?
any tips appreciated
It would help if Steve would create a new thread, say Sea Ice 2010, at least those posts from October would then not be at the bottom. I use IE as well and it’s borked.
Andy
I’m getting the feeling the relatively large maximum extent values of the last 2 winters will not be repeated this year.
Andy
Re: AndyW (Jan 28 00:51),
At this point, that’s a pretty good bet. Current projected maximum based on 2003-2009 average behavior is 14.03 Mm2 compared to 14.52 and 14.41 for 2008 and 2009.
I agree, it will be something like a 2006-ish year.
JAXA January average extent:
2003 13532601 km2
2004 13201099
2005 12819158
2006 12650172
2007 12903604
2008 13062258
2009 13121810
2010 12876810
2010 is almost exactly on the trend line for 2003-2009. The change in slope was from -50915 km2/year for 2003-2009 to -47670 km2/year for 2003-2010, which is not significantly different. So there is still no sign of an end, or even a slowing, of the decrease in Arctic sea ice.
Thanks Dewitt. 14.03 estimate for maxima looks pretty good, or there abouts.
Interestingly, due to less ice so far in the Arctic as we head towards the winter maxima the overall global ice here
shows a fairly big negative anomaly. So much so that once Antarctic starts to freeze over, at the more rapid pace it has shown in recent years, this may not be able to drag the total anomaly positive. If that happens then 2010 might be the first year since this record began that it’s negative all year.
Andy
The satellite seems to be having problems. The JAXA graph has not updated since Feb 1st and looking at the maps at Bremen last update was 2nd and that was incomplete
http://www.iup.physik.uni-bremen.de:8084/amsr/amsre.html
There has been some solar activity recently with another bout to hit us on the 10th, but nothing out of the ordinary to cause problems. Routine maintenance or some issue??
Andy
The guys at Bremen say they are not getting data from JAXA and they are still trying to find out why, as JAXA is not updating it seems neither are Jaxa. They sent me this link as well for another ice movie, it’s pretty good.
http://www.youtube.com/user/KlimaCampus#p/a/u/0/ro0-7U8UtvI
but more interesting is the other IPCC estimate in relation to the Arctic
http://www.youtube.com/user/KlimaCampus#p/a/u/2/a2NLxlG4VsA
Andy
Re: AndyW (Feb 5 05:47),
Almost all the SRES are a complete joke. The assumptions defy logic. A1B has annual fossil CO2 emission almost 50% larger in 2100 than today. The major assumption is that there are vast amounts of natural gas available somewhere because energy use from natural gas in 2100 will be twenty times what is today. A1B is conservative compared to some of the other scenarios. About 19 of the 40 scenarios have oil consumption in 2100 higher than today. Some of those scenarios, though have oil use today at only a fraction of what it actually is. So even if the models were correct, GIGO still applies.
Yes, that is the sort of average middle safe one, but the main issue nowadays for us non IPCC specialists is what produced in the 2007 report was scientifically peer reviewed or not? And if it was peer reviewed was it peer reviewed without being subjected to bias as the CRU leaked emails suggest might have been going on?
Very tricky. Hence why I prefer to stay just looking at data in the Arctic rather than trying to make more of it than perhaps is worth.
The problem is you can give people these facts and they still want to put some spin on it. I put some of your figures on another blog after someone said there was a massive rebound after 2007 and they called me an idiot and emotionally wanting no rebound!
How can you converse with both sides of the fence when they just don’t look at the data except with eyes that want to read it one way or the other. Very sad. I’m glad on here, in general, we just look at the data and try to learn a bit more about what is happening up there and over the years try to ” get a feel for it”
Slightly off topic
http://news.bbc.co.uk/1/hi/scotland/8499931.stm
Fantastic; if they can make another batch from that after all these years I would buy a bottle. Can’t imagine any modern expedition taking 5 crates of whisky down there. Men were real men in those days. 😀
Andy
NOAA is in for January, 13.78 Mm2 Arctic extent and 11.59 Mm2 area compared with 14.07 and 11.91 for January 2009. The extent for 1/2010 is slightly below the thirty year trend line for January.
Maybe next year when we aren’t having an El Nino. The RSS TLT 60-82.5 latitude anomaly for January was above 1.5 degrees for the second month in a row. That’s still below the record high of 2.58 for January, 1981 but about the tenth highest for all months in the record.
Here is the latest on the Arctic ice from Canada:
Guess what! It’s worse than we thought! You’ve gotta believe it!!!
My favourite part:
I figure that it works out to a little over $1.5 million a paper. Makes me proud to be a Canadian (taxpayer) 😦
Arctic oscillation for January is still strongly negative, -2.587. El Nino seems to be easing from a visual look at SST’s rather than anything else.
Interesting info about that Canadian expedition, will be interesting to see the papers they present, not much meat in the article.
NSIDC monthly update available for Feb now
http://nsidc.org/arcticseaicenews/index.html
Andy
The JAXA graph has now resumed but with a gap from 2nd to 5th May or so, they must have lost all information for those days.
Andy
For the first 10 days of Feb the average gain has been less than half of 2009. This may bring the projected maximum down to below 14×10^6 km2
Andy
Re: AndyW (Feb 11 01:15),
A maximum less than the 13.78 Mm2 in 2006 is not out of the cards. A calculation based on the 2003-2009 data has an expected value of 13.95 Mm2 with a 2.16 standard deviation range of 14.30 to 13.61 Mm2. Second lowest is 2007 at 13.95 Mm2.
Interesting, even the graph looks pretty flat at the moment. The Barent sea area really hasn’t frozen up well this year up to now, but last year it had a late spurt. If it doesn’t have a late spurt then it’s going to crumble rapidly on when melt season starts.
Temps in Siberia have now dropped, those in north Canada risen.
As an aside I saw Phil posting on WUWT recently, shame he never comes on these threads anymore.
Andy
Re. No Phil. Contributors here know Phil’s track record. Instant, intense skepticism follows cherry-picked data and argument over here. Same will happen at WUWT. Give it a little time.
I am a contributor here and know his track record,it’s nothing like you suggest. He can be very scathing and doesn’t suffer fools gladly but when it comes to facts he does tend to come up with the goods, it’s not often you can say he is talking rubbish when it comes to facts rather than the politic.
Andy
We may see some posting activity around the time of the maximum. Then I don’t expect things to pick up again until June or later when speculation about the minimum will start.
So much for a record low. Today’s JAXA early number is 13,790,156 which is above the 2006 max already. An increase of 190,000 km2 in four days at this time of year is unusual. Projected max is now 14.14 Mm2, above 2005,6 and 7. Since this is somewhat like technical market analysis, I should add a disclaimer on the ‘projection’. Past performance is not really a very good indicator of future performance for sea ice.
I am expecting 2010 ablation to go pretty much along the lines of 2006. Except for a couple of areas of note, it has been a cold winter. Ice will be pretty thick, the drift this year is interesting in that there isn’t a lot of movement. If the wind patterns remain favorable, I see nothing that would indicate that this years ablation will end with less ice than last year. Considering that there is more old ice going into this year’s ablation than last year and that this year’s temperatures over a large part of the ice pack were colder than last year, the amount of ice at minimum would seem likely to increase again this year.
If you use 30% as DMI does extent rather than 15%, this year’s ice pack is looking pretty healthy. There is more ice right now than any year since 2005 except 2008 and it is very close to 2008 extent at 30%. But starting at about right now, 2008 put on a burst of growth for about a weeks or so to push it ahead of any of the other recent years.
http://ocean.dmi.dk/arctic/icecover.uk.php
I think 30% is a better indication of real ice extent, in my opinion. 15% tells you more about how stormy it has been and how much ice has been bashed up and scattered by wind.
A higher 30% extent and lower 15% extent might just mean calmer conditions with fewer storms. To clarify, if a date shows higher 30% extent than recent years but lower 15% extent, it might just mean things have been calmer.
I don’t see any indications of impending disaster and my expectation is that the recovery from 2007 will continue as it has the past two years.
And you can not tell where a year is going to end up by looking at the maxium. If you look at the DMI graph, you will see that on June 1, all the years are bunched up in a very narrow range no matter where they topped out in March. Even looking at July 1 you can see that the rankings on that date do not correspond to those in September. 2005 had the most ice and 2006 was next to last. By September, 2006 was the greatest of the bunch. That would tend to indicate to me that weather conditions were good for ice that summer. The wind did not blow so much of it out into the Atlantic. 2007 was another story and the wind blew like the devil itself. The ice was shipped in huge quantity to the Atlantic where ice goes to die.
Looking at arctic sea ice data provided here (tx to D. Payne), indicates very small changes in winter measurements from 2003 through today. Summer measurements fluctuate a bit, (2007).Based on actual measurement, the conclusion is that nothing significant has happened to arctic ice over the past 8 years to support AGW theory.
It’s the summer that the change is happening
as explained by the scientists
http://nsidc.org/arcticseaicenews/index.html
“While Arctic sea ice extent has declined in all seasons, the downward trends in winter ice extent are much smaller than in summer. Polar darkness and low temperatures mean that the ice generally refreezes to about the same boundaries each winter”
All your conclusion proves is that you will grasp any straw that tries to defeat the evil AGW. AGW doesn’t have much sway on this thread, it’s all about trends and current data and not reading too much into it apart from interest in the current events up there.
Andy
Andy: Read my comment; all I said was that according to data provided on this blog, nothing significant has happened to arctic sea ice over the past 8 years – snip
Steve: the word “conspiracy” is forbidden here.
And it looks like that for today, at least, 2010 ice extent at 30% has surpassed 2008 but as I mentioned earlier, 2008 put on a burst of growth at about this time of year.
“Atlantic. 2007 was another story and the wind blew like the devil itself. The ice was shipped in huge quantity to the Atlantic where ice goes to die.”
That’s not true. The record loss in 2007 was actually due to thin ice at the start of the year, lots of clear sunny days, warm air from the south melting the ice (not ablating, melting) and those same winds from the south pushing the ice north, away from the Siberian coast so reducing the extent. Nothing to do with shipping it to the Atlantic. Read the NSIDC reports from the time.
This “winds caused the whole of 2007 blowing it to southern climbs” is a mantra being slowly spread as an excuse by AGW skeptics. It may wash on the WWW but it doesn’t wash here where people know the details.
Andy
Andy
“‘winds caused the whole of 2007 blowing it to southern climbs’ is a mantra being slowly spread as an excuse by AGW skeptics”
Actually, that is not true according to NASA on October 1, 2007
That winds are some AGW skeptic mantra is really misinformation spread by the AGW “believer” community.
Touche on your last comment! 🙂
Nghiem’s comments are a subset of what NSIDC is saying, ie it was compressed northwards, but Nghiem doesn’t mention the other factors involved such as the warmer air compressing it (which causes melt)and also the sunny days.
In fact the rapid drop in 2007 corresponds in time with the start of the sunny clear days shows that to be also one of the main factors, if not the main factor. That keeps getting forgotten though :whistles:
Andy
Andy
And a new update.
The latest value : 13,840,938 km2 (February 16, 2010)
Hmm, I am showing 13,861,094 km*km
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
Re: crosspatch (Feb 16 22:35),
JAXA seems to have changed their update pattern. Rather than 10PM EST for the initial and 9AM the next day for the final, the initial comes out around 7PM and the final around 10 or 11PM the same day. The 13,840,938 km2 extent number was the initial number for 2/16, so you’re both correct.
Conintiues the good increases seen recently, seems to mainly be on the Pacific side.
Andy
I found this on WUWT. Looking at the pattern of ice in the Barents and Kara Seas, a shift in Atlantic currents as the main driver in this area has been obvious for years. A similar thing happened in the early twentieth century starting in the late teens and early twenties and dying out in the late thirties.
If that’s the case then it is no wonder the Northern (NE) passage is becoming more often open for passage. My bet is it will be open again this year.
Andy
Awwh… C’mon Andy. You know that arctic ice ain’t AGW, it’s just weather! So early a prediction? Was this an attempt to get a new thread form Steve?
I am completely flabbergasted that you would claim that the NWP would be open AGAIN. When was the last time? Do you think it was last year? If no, when were the last 5 years in which you think it was? If yes, lets let you lay down some general parameters for what you consider open. I’ve got 100 quatloos on you being wrong this year despite the funky (and humid) El Nino we have going on this winter.
Arctic sea ice at 14 million km2 ! Following snip advise here, I’ve checked into Pathological Skeptic Syndrome Therapy. It may not be working yet. I still can’t find the AGW fingerprint in the 14mm number.
Hi Hengav,
I’m talking about the Northern Passage, NE Passage as it is also known, not the NWP in my post, sorry if it was not clear. If it helps Steve make a new thread though then that’s great 😀
And J Solters, I was a bit heavy handed in my earlier reply to you, my apologies, I must have had a hard day at work or a bee in my bonnet that night. I sometimes just bash away at the keyboard.
Andy
Sorry Andy, it was my bad. I just read and assumed it was the NWP you were referring to.
Brad
Hi,
I think the NW passage is going to be tricky again, more so than last year, as a hunch. I think it is getting to the stage where the NE passage is going to more navigable than the NW in years to come, so some change. We will see though, I wouldn’t bet my house on either passage, either way.
Certainly the NW passage is not getting easier it seems with all these record melts, certainly not to get commercial stuff through, it’s still a pipedream.
Andy
Current JAXA AMSR-E Sea Ice Extent.
The latest value : 13,943,594 km2 (February 18, 2010)
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
Here is the updated figure for the 18th Marian, 13,966,406 km2 February 18, 2010
Forgot to say, the Arctic Oscillation figure for Jan is still very low, at below -2.5. It’s very cold in Siberia because of this currently, -55C.
Andy
WUWT on 2007
Of course the bridges or arches in the Nares straight completely collapsed in 2009 and yet there was less summer melt. So showing this was not the reason for the massive melt in 2007. This is yet more “wind driven rather than melt” wishful thinking. It doesn’t wash I’m afraid.
Andy
“Of course the bridges or arches in the Nares straight completely collapsed in 2009 and yet there was less summer melt. ”
They collapse every summer. They form in winter. They did not form in the 2006-2007 winter for some reason that is not clearly understood because temperatures that winter were below average. So it apparently wasn’t due to temperature, it was probably due to some unusual wind conditions during the winter that caused the arches not to form. And again, 2007 did not see any particularly great melt in the arctic, it saw an unusually large amount of old ice transported into the Atlantic. The inventory of old ice has been recovering quite well since.
Crosspatch, the collapse in 2009 was so great they had a running set of pictures of it at the Canadian ice service web site. It was so large you could actually see it on the AMSR-E maps at Bremen, and yet 2009 was less than 2007.
As I mentioned before, in 2007 the onset of the greatest amount of melt happened when clear sunny skies appeared in July.
Comparing 2008 to 2007-
http://nsidc.org/news/press/20081002_seaice_pressrelease.html
“This summer’s weather did not provide the “perfect storm” for ice loss seen in 2007: temperatures were lower than 2007, although still higher than average (Figure 5); cloudier skies protected the ice from some melt; a different wind pattern spread the ice pack out, leading to higher extent numbers. Simply put, the natural variability of short-term weather patterns provided enough of a brake to prevent a new record-low ice extent from occurring.
NSIDC Research Scientist Julienne Stroeve said, “I find it incredible that we came so close to beating the 2007 record—without the especially warm and clear conditions we saw last summer. I hate to think what 2008 might have looked like if weather patterns had set up in a more extreme way. ”
Just blaming it on wind only is going against what the scientists are saying. It played a part, but it was not the whole story.
Andy
My guess for date of maximum is March 9th and with a max of 14,225,000 km2
I am not going to venture a guess because maximum depends on factors quite distant from the pole. Mostly it depends on storms. The fewer storms, the greater the extent will be, generally speaking. We are already above the 2006 and 2007 maximums and just about at the 2005 maximum. The pattern is suggesting that it will come in somewhere between 2004 and 2005 but there can be amazing changes in very short periods due to conditions that have nothing to do with temperatures.
Current JAXA AMSR-E Sea Ice Extent.
The latest value : 14,009,844 km2 (February 24, 2010)
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
Yep, just crawled past 14, will be interesting to see if it has another spurt like it did last year at this time, or whether it plateau’s for a while.I’m not hazarding a guess 🙂
The area graph at Cryosphere shows same of course.
almost at minima for the Antarctic
Andy
Re: AndyW (Feb 25 01:09),
The rate of increase of Arctic extent has been way above average recently. That’s usually followed by a rate below average, but who knows. 2009 took off at this same time of year after running a rate way below average for a week or two before.
Yes, who knows. 2003 was an interesting year looking at the graph.
Extent is down slighty, if we have just seen the peak then it would be a rather early peak than most years so my guess it’s just having a breather and the peak will be later.
In other matters
http://www.abc.net.au/news/stories/2010/02/26/2830801.htm
I was listening to an Australian polar scientist saying this berg is large enough to affect both the local currents and stop bottow welling, so reducing food for wildlife. At least this cannot be painted as climate change .. hooray 🙂
Andy
Re: AndyW (Feb 26 01:21),
It’s kind of a big breather. You have to open the archive link to get yesterday and today’s data. It’s down over 100,000 km2 over the two days. We’re back at the 2007 level.
Won’t matter. Maximum has no correlation to minimum. Also, if you look at 30% extent rather than 15%, 2010 is currently higher than any year in the past 5 except 2008. In other words, save for 2008, 2010 is higher than any year going back to 2005 in 30% extent.
Just means that the ice is more packed together and concentrated this year and there isn’t so much broken up ice blowing around in the wind (less than 30% concentration but greater than 15% concentration).
That the 30% concentration is so high would indicate that there is a larger area of consolidated ice this year than in years past. That would indicate a lower rate of ablation might be expected this year. 15% concentration doesn’t mean a lot to me. I don’t want to count areas that are 85% water to be “ice”. The amount of ice at less than 30% concentration is going to be very volatile and not really having a lot to do with temperatures. It would have more to do with storms and winds.
Re: crosspatch (Feb 27 01:54),
Are you sure you aren’t comparing apples and oranges? The average concentration at this time of year is ~95% and it doesn’t vary much. You have to compare 15% and 30% cutoff using the same algorithm to calculate the extent and area from the individual pixel concentration data. I’ve looked at the individual pixel concentrations from the ASI algorithm (Spreen, et al., 2005 and Spreen, et al., 2008 ) and the difference between a cutoff of 15% and 30% wasn’t very large. Then there’s the question of which satellite they’re using. If it’s the same source as that used by Arctic-ROOS then it doesn’t mean very much.
From the DMI page:
Re: DeWitt Payne (Feb 28 21:53),
That should have been: you have to calculate area and extent with 15 and 30 % cutoff using individual pixel concentration data produced by the same algorithm.
February average JAXA extent results:
year extent(km2)
2003 14362322
2004 14069806
2005 13528214
2006 13438041
2007 13698203
2008 14117705
2009 13998137
2010 13710726
2010 is slightly below the trend line for 2003-2009 so the trend changed from -29527 km2/yr to -34415 km2/year, not significantly different. The 1979-2009 NOAA trend for Arctic ice extent is -44100 km2/year. Based on the JAXA data, the NOAA result should also be very close to the trend.
As of today, at 30% concentration, there is greater ice extent in the Arctic than there has been in any year since 2005:
http://ocean.dmi.dk/arctic/icecover.uk.php
Air temperatures remain below average though that can change literally overnight.
At 15% concentration, this year lags behind both 2008 and 2009.
Sure they are apples and oranges. 30% is not 15%. Greater coverage at 30% concentration implies that there is more consolidated ice this year with less breaking up, blowing around due to storms which would make sense if the polar jet is farther South this year keeping the storms farther from the ice edge.
It has also been indicated that so far this season there is less drift South than normal of older ice.
So the indications are that this year is more likely to end up with more ice than last year at the end of the ablation season than it is to end up with less ice.
1. More old ice at the start of the ice accumulation period.
2. Less drift of old ice out of the arctic preserving the inventory.
3. Colder air temperatures than recent years.
4. More consolidated ice than in recent years.
5. Cold temperatures continuing late into the accumulation season.
Basically the bottom like for me says that indications are favorable that ice continues to recover in the Arctic.
Re: crosspatch (Mar 5 12:24),
You’ve missed the point. You assume that the DMI extent numbers, if calculated at a 15% concentration cutoff, would look like the JAXA numbers. There is no evidence that this would be the case and good reason to believe it wouldn’t be. One of the good reasons being that there can’t possibly be much area containing ice concentrations in the 15 to 30% range when the average concentration is 95% including area in the 15 to 30% range.
AndyW,
See my comment above on how much faith I put in calculating the maximum (or minimum for that matter) by using the average of past behavior. And it’s 14,314,375 km2 as of today. That’s up over 99,000 km2 from the previous day.
Meanwhile Antarctica appears to have reached, or is very close to reaching, the minimum and is higher than the long term average by a little bit. http://www.iup.uni-bremen.de:8084/amsr/ice_ext_s.png
I suspect it will climb rapidly, more than the average, once again too. This has been explained as result of wind patterns and the geogpahy of the continent.
Meanwhile the Arctic is bumbling along just under the 14×10-6 level.
What are peoples thoughts on Hudson Bay melting more quickly this year after the warmer weather in that region over Winter, or was the warmer weather more to the Eaat?
Andy
Current JAXA AMSR-E Sea Ice Extent.
The latest value : 14,037,500 km2 (March 1, 2010)
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
I’m calling the minimum for Antarctic ice area (Cryosphere Today version) at 1.961 Mm2 on 2/20/2010. That’s about a week earlier than average, so not all that unusual. The 1979-2008 average minimum is 1.870 Mm2 (probably). Global sea ice area is also going up again. The minimum was 14.895 Mm2 on 2/7/2010 (Cryosphere Today). The 1979-2008 average minimum is 15.853 Mm2 on 2/17.
Current JAXA AMSR-E Sea Ice Extent.
The latest value : 14,159,531 km2 (March 3, 2010)
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
Nice late spurt going on currently!
Andy
Looks like it is putting on a spurt.
Current JAXA AMSR-E Sea Ice Extent.
The latest value : 14,215,000 km2 (March 4, 2010)
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
It will be interesting to see if this late rush of ice making will be followed by equally quick melting once past maximum due to being created so close to the “turn”
Dewitt’s statistical extrapolation to get maximum is now no longer accurate. Note- that is not any critisism of course, just pointing out how much this late spurt has increased the total.
Andy
It will be interesting to see if this late rush of ice making will be followed by equally quick melting once past maximum due to being created so close to the “turn”
***
Andy,
Good thought but how would this differ signifigantly from all other years?
Ian
Woops….If you look really hard you will see that’s a ‘c’ and not a ‘g’ in significant 🙂
Re: Ian (Mar 5 07:25),
Actually I thought it a rather elecant typo.
That made me smile, very clever.
Ian, my thought is that if it suddenly appears just before it start getting warmer then it’s not going to be as thick, so given an average spring we may find a fast melt unlike last year where spring was very slow to sprung.
Canada seems to be warm over winter in the east so I wonder if Hudson bay will melt faster than normal?
Andy
Not sure this figure today is accurate because it is a very large increase for this time of year 14,314,375 km2
Re: Michael Jennings (Mar 6 09:30),
14,350,938 km2 for 3/6/2010. The peak dates for 2004-2009 range from 3/5 to 3/11, so pretty much any day now.
Impressive late burst! Have to agree, think we might have seen the peak, or at least not much increase from now on.
Andy
AMSRE data:
For the period covered by the satellite 2002 to 2010 the change in extent on a particular day of the year is plotted as a linear trend.
It is interesting that just before the summer melt (May) the area increases by 5000sqkm over the 9 years.
There is then a sharp fall to 20,000 sqkm loss per year in october.
The area change over the remaining year seems to be about 5000sqkm/year
Is the rise from April to May due to delayed melt? or is it spreading of melting ice?
Mike
Hi Mike
Slow melt in 2009 and the high winter max in 2003 are pushing the value up for that time period. I don’t think you can read into it any trend.
Andy
I find this interesting: since the AMSR-E measurements started in 2002, the 2nd, 3rd, and 4th highest maximum Arctic ice extents occurred in the last 3 years — 2008, 2009, and 2010. (Of course, other systems were used in previous years with higher measurements, but that does not detract from this recent trend.)
With today’s decrease to 14,353,438 km2, it appears likely that March 8th will turn out to be the max for the season.
Sea ice at 14.35 million km2! No fingerprint here-move on. BTW, where’s the “Flying Fickle Finger of Fate” award (Laugh In ?) when we need it (them)? I see many worthy recipients in the near term when these fingerprint issues finally dissolve.
Re: thefordprefect (Mar 9 07:51),
My guess would be that it’s an artifact of the algorithm for converting the microwave emission intensities to ice concentration. Ice with surface water has different emission properties compared to ice with no water, for example. There’s a known glitch in early July and again in mid-October when JAXA switches to a different algorithm to take this into account and then switches back.
Re: Michael Jennings (Mar 10 08:06),
The long term average has a small dip and a second peak around 3/20, but the first peak is usually slightly larger. If 3/8 isn’t the maximum, the actual maximum probably won’t be very much larger.
There are some interesting graphics on Cryosphere Today. I randomly chose 14th Feb 1981-14th Feb 2010. It is possible to compare daily sea ice from 1979 to 3rd March 2010 for any given dates.
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=02&fd=14&fy=1981&sm=02&sd=14&sy=2010
Not sure if it is spurious but there seems to be a large body of clear water in Hudson bay currently. Temps are warm to the south but the Canadian ice service shows it fully covered. I would suggest it is an artifact on AMRS-E.
Andy
What happened to ‘sea ice’ data? Another thread here, or did it switch to another site? It’s been very interesting watching this winter’s progress. Help!
I can never seem to find the latest comments on this thread, unless there is one on recent comment sidebar at home page for me to follow.
perhaps the regulars can point out where I am going wrong. I use IE8.
any help/suggestions appreciated
Re: dougie (Mar 24 15:23),
The order of old threads is a problem unless you are using the Firefox browser with CA Assistant. Then you can set the reading properties of the thread to your preference.
The Cryosphere Today data archives for Arctic, Antarctic and Global sea ice area starting on 1/1/1979 are now available thanks to William Chapman. The columns aren’t labeled, but the order should be obvious: date (decimal year format), anomaly, area, 1979-2008 average area (Mm2).
I don’t like the decimal format for the year, but it is good to see him put them up.
The level of extent is gamely hanging in there, I thought it would have dropped by now. The AO has gone positive so the temps are cooler at those lattitudes of the edge of the sheet I think, helping prolong the level.
Andy
Re: AndyW35 (Mar 26 00:56),
Paste it into Excel and add a date column in whatever format you like. I set day 1 to 1/1/1979 then added 1 to the previous cell in the cell below and then extended that. There may even be an Excel function to do that directly, but I tend to go for the brute force solution rather than spending time exploring the help file or googling for a more elegant solution.
2010, 2009, and 2008 continue to have the largest Arctic ice extent since 2003 according to ASMR-E. 2010 is within 0.1% of taking the lead from 2008 (since 2003). Also 2010 Arctic ice area is now almost equal to the 1979-2006 mean. (I wonder how much closer it would be to the mean close it would be if 2007 was included in the mean?) Of course, these observations proof nothing except to suggest that nothing alarming is happening to Arctic Ice.
Re: An Inquirer (Mar 25 10:29),
March started a little slow but is finishing strongly. I suspect the AMO index for March will be much lower than for the previous 8 months and may wander into negative territory. A lot of the recent warming has been concentrated in the Arctic, as the cold air from the Arctic that made for such a snowy winter in the US and Europe was replaced by warmer air from lower latitude. Now that I have all the Cryosphere Today data, I’ll probably start some control charts on monthly averages.
Continued Deafening Silence on latest arctic sea ice data here. Anybody know why?
Re: J Solters (Mar 29 08:30),
I don’t think anyone cares much about the behavior at the maximum. It’s the minimum that gets all the attention. If you look at the behavior of the ice around Svalbard, you can see why the maximum has been extended. The flow of warm water from the Atlantic has obviously slowed drastically compared to January and February. Svalbard is now nearly surrounded by ice. There’s more ice in the Barents Sea as well. Whether this has anything to do with AGW is not at all obvious. Note that at about the same time as the Arctic extent has been closer to the average than since 2003, Antarctic ice formation has dropped below average. As a result, the sum of Arctic and Antarctic area has remained well below average. It’s been a lot closer to average at this time of year in past years.
The last arctic sea ice extent number was posted on Mar. 10, 2010 as 14.35 million km2. That’s 19 days old and counting. All I’m asking is for a current number. (I promise NOT to say anything about a high max number’s relationship to AGW or arctic cooling, oceanic or atmospheric.) After all, it’s just a number, not a long-term trend.
Re: J Solters (Mar 29 15:04),
Something ate my post and I don’t feel like retyping it. Since I’m annoyed about that, I’ll take it out on you. Get the data yourself. It’s readily available.
something ate my post twice.
14.30 on March 28.
It is frustrating to see four of my posts eaten and lost, but here is another try, because the latest update is truly noteworthy:
14.36 million sq km on March 29.
J Solters, this is from ASMR_E for Arctic Ice extent which I believe is consistent with your request.
Thanks for ice update. True to my word, I’m saying nothing about possible impact of this miserable little uptick in arctic ice extent on AGW theory.
According to JAXA, the current extent is higher than any year since 2003 and it’s on the verge of a double peak that could take it above 2009 and 2008.
Big increase today. This is getting more interesting than most maxima. Looks like colder weather in the Baltic region is adding ice there. SST’s still warmish so not them to blame.
Might surpass 2003 at this rate, excellent!
Andy
14.39 million sq km for extent on March 30.
There is some speculation (especially in AGW communities) that this recent uptick is due to ice patches flowing south earlier than usual. In such a scenario, Arctic ice levels would plummet faster than usual later in the year.
The thickness of the ice through the Framm Straight and down the East coast of Greenland has increased, but the big gains are really in the Baltic since it got colder. However I will be interested to read what the NSIDC update says in early April.
Expect WUWT to have a party if the value is higher than 2003 shortly, which I expect it to be. 😀
Andy
Still darn cold here in Japan where we have endured the coldest winter in decades. No dancing (and drinking) under the falling cherry petals yet…..not in my area.
Ian
Revision to 14.41 for March 30.
For March 30, 2003, it was 14.53 — the highest for ASMR-E which started in 2002.
“There is some speculation (especially in AGW communities) that this recent uptick is due to ice patches flowing south earlier than usual. ”
I would buy that if the increase was in only the 15% ice extent, but the 30% extent matches the behavior of the 15%. So either those are some really huge “patches” or it is baloney. We are just going to have to wait and see how it plays out but I don’t see anything that would indicate that we won’t see continued recovery at this year’s minimum. I am still expecting it to come in sort of 2006-ish.
Still holding at 14.41 for March 31. Still a bit less than 2003’s record level of 14.43. No doubt many eyes will be looking at the April 1 number!
Not much a decrease for April 1.
14.40 for 4/01/10
Getting close to the ADMR-E record-setting 2003 pace of 14.41.
It will be interesting how it turns the corner, quick drop or nice gentle curve. If the curve then it will be well above all previous years on the JAXA graph.
Andy
Yes, 2010 is clearly in the lead for all years of ASMR-E data. 14.31 for 04/03/2010 compared to 14.25 for 03/03/2003. No doubt many interested eyes will continue to see whether 2010 remains in the lead and surpasses the May record that was set in 2009. Both skeptics and believers alike would do well to remember a recent admonition from Dr. Serreze: “we must be more careful in not reading too much into one event”
“Both skeptics and believers alike would do well to remember a recent admonition from Dr. Serreze: “we must be more careful in not reading too much into one event”
Indeed!
Andy
“Let us not be tempted to read too much into one event”. I am tempted to follow this commandment into the 2007 “one event”. Just a blip. No trend here; move on, now!
Well 2010 is very high up so it will be doing some big drops to get down the the crowd just before the June correction. Talking of which, JAXA have said previously they will be trying to better smooth out this correction in June for the melt water ponds, so it will be interesting to see if they actually do it this year.
Andy
Well it was a very warm March according to the satellite temps but the Arctic ice is still very high in extent. It will be interesting to see how it drops now, a slow rate like last year or more of a plunge like 2004.
Andy
Your observation points out that Arctic ice extent is affected by many factors, not just GMT nor only Arctic temperatures. No doubt in the next 4 months, optimism on Arctic ice will wax and wane several times. Nevertheless, so far there doesn’t seem to be much support for the AGW argument that high March values for extent was due to ice flowing prematurely out of the Arctic which would soon lead to rapid decline of extent #s this spring. 2010 values are a remarkable 2.2% above 2003 values. 2009 is now in 2nd place for AMSR-E measurements (which started in 2002).
Could anyone tell me what has happened to Arctic and Antarctic graphs for Cryosphere?
At http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/
the graphs current.365.jpg and current.365.south.jpg used to compare daily ice area to a base period for Arctic and Antarctica. But no new #s have been added to them since December 31, 2009. There are other graphs for ice area, but I do not see one that compares current Arctic to a base — nor for Antarctica.
global.daily.ice.area.withtrend.jpg has the comparison for global ice.
Yesterday saw the first near 100,000 drop to date but I expect that will be fairly commonplace from now on.
It rained on the Catlin Survey team today which they are alleging to be proof of climate change (it should have snowed).
Well, it snowed a couple of days later in Vermont.
Also, some snow has fallen in the Dakotas, and apparently Ottawa has been “plagued” with April snow.
Interstate 5 was closed just north of Los Angeles yesterday due to snow. I can’t ever remember that happening anywhere near this late in the year before. (But I’m not alleging it to be proof of anything.)
As an interested but novice kibitzer to the climate thing, I’m often puzzled by some of the data I see. I generally interpret this as due to my ignorance. One of these puzzles is found on http://nsidc.org/arcticseaicenews/ Near the bottom of the page is a nice graphic called “Arctic Sea Ice Change 2009 to 2010.” I see that from September 09 to March 10, areas that had no ice in Sept become populated by>2 year ice by March. Other areas that had >2 yr or 1-2 yr ice become first-year ice or are downgraded by a year during a period where every accounting (e.g. Cryosphere Today) shows an increase in ice. I’d appreciate an explanation of this by someone in the know.
I think the answer to your question is that ice flows. Ice in the Arctic does not stay in one place, but moves as ocean current carries it. Eventually, ocean currents carry existing ice far enough south to melt.
I would speculate that over 80% of people think that Arctic ice just stays in place.
The flow of ice has some interesting implications. If we set up a monitoring station at the North Pole, it is no longer at the North Pole in a few days. Also, sometimes Arctic explorers can struggle all day to trek north, and at the end of the day — they might have lost “ground.”
Thanks. That likely explains the changes noted.
When a change in age happens at a given point, it is tempting to interpret this as more ice forming at that spot.
The question then arises as to how this is determined. Age is listed not ice amount. One could imagine tracking the motion of ice areas noting how long each chunk persisted. Or is thickness the determinant of age?
Local paper (Victoria Times Colonist) reported a 3-minute rain squall experienced by some “researchers” in the Arctic, who credited said squall as evidence of warming. Perhaps they haven’t looked at the sea-ice figures. Or, perhaps these things happen regularly. Or perhaps they are unrelated. Were they coring trees at the time?
Does look like the rate of ice loss has picked up now in the las week.
Lots of open sea areas (polynas) havwe appeared too.
Andy
Nenana ice clasic terminates on the trend line (earlier by 1/2 day per year)
//img693.imageshack.us/img693/1796/nenanaiceclassic2010.png put an http: in frnt of this for the link.
Sea Ice By plotting the slope of straight line curve fit to each day since 2002 you get this plot:
Note that the plot shows an increasing ice area up until early may followed by a rapid decline to minimum. This years so far seems no exception. Is this due to thin ice break-up increasing area (only 15% coverage needed)and then rapidly melting?
I do not yet have a good handle on the blue line in the graph you linked: “Area change per year vs. day of year”
What is the formula for “Area change per year?” Is it a statistical term? Or is it a mathematical term? Or what?
Take all Jan 1st for the available years and do a simple straight line fit through the areas (“slope” command in excel)on that day.
Repeat for all days of year. Plot the resulting slope values vs day of year.
Change in area per year is this slope.
Sea ice area has been
decreasing at a “fixed” rate from january to april
increasing more rapidly from april to may
decreasing more rapidly from may to nov
decreasing at a fixed rate from nov to jan
As the plot says only years 2002 to present are available so not many points in the slope calculation.
Mike: There is not enough data to find a trend. The history of the classic shows a variable nature of increasing and decreasing trends. Your evidence is only of picking dates to prove some purpose. If memory serves this has been repeatedly discussed and the classic only provides evidence of variability.
Re: thefordprefect (Apr 30 08:20),
I’m no statistician, but these statistics seem quite suspect!
In the Nenana plot, what is the basis for choosing a single breakpoint at 1965? It would appear that the general real-world trend is visible in the graph, as long as one ignores the overlaid arbitrary straight-line “fit.” I.e., there was a warm (low) point in the 1930’s, and another in the 1990’s. We’re now recovering from a warm potentially similar to what was seen 3/4 of a century ago.
Simple straight-line fits obviously bear little relationship to the real world, which hardly ever does anything in straight lines.
2010 now below 2009 and dropping at a fairly steep gradient. Cold SST’s still in Bering and Okhotsk with SST’s on atlantic side cooling.
Andy1
I have been following the email string on this site, and its predecessors, for several years. The commentary, with a few exception is of extraordinarily high quality. I started as an agnostic concerning whether warming was anthropogenic or not. I am convinced today that CO2 is not an important cause if we are warming. I say “if” we are warming because I am also convinced that station temperature data is so corrupt that it cannot be used as evidence of warming. However, it is clear that CO2 is increasing in the atmosphere and that ocean CO2 content is rising as some portion of the increased CO2 is dissolved in the oceans. Here and there I read that a side effect of the increase of CO2 in the ocean may be to cause dissolving of limestone reefs. I am surprised tht the global climate change types don’t argue more strongly that we should reduce fossil fuel usage to reduce CO2, regardless of the effect on global warming, to reduce damage to the oceanic environment. Maybe they have and I have just missed it. Can anyone point me to some good articles/blogs on the projected effect on the oceans of increasing CO2. I have a PhD in organic chemistry so I think I can handle technical discussions of this kind better than I have been able to comprehend some of the more technical commentary that I have enjoyed on this site.
Coral reef ~ Calcium Carbonate ~ Dissolved CO2. I’m 40 years from Engineering Chem, so can someone explain the relationship? thanks.
If my recollection is correct, it goes like this: CaCO3 is soluble in inorganic acids. CO2 in water forms H2CO3,an inorganic acid which dissolves limestone/CaCO3. Reduce the CO2 in water and the pH goes up making the limestone less soluble. Increase the CO2 and the pH goes down, increasing the solubility of limestone.
The extent is plunging now at a healthy rate and I think 2010 will end up below 2008 and 2009 when all is said and done in September. The big question is whether it can fall to 2007 levels and that will be the main topic here for the next several months as it takes a run at it.
It’s certainly not having a slow melt like 2009 in spring. A lot of the ice on the Atlantic side was only put on late so is fairly thin, so melts fast. Canada had a warm winter so I am expecting Hudson Bay to melt quickly too.
As for the summer minimum I have no idea where it will be, but I am guessing it will be below 2009 due to the continuing warm global temps.
I hope Steve gives us a new thread soon, what is his email to ask?
What happened to Dewitt as well? I miss his info!
Andy
Andy
2010 has rejoined it’s brothers on the JAXA graph and has now got into step with them. As time goes on I wonder if this means 2010 will be at the bottom of the pack come the June correction due to the warm conditions over the winter months making the ice less thick in the main melt zones at that time?
Andy
Soon, we will not be able to say that the three years of largest Arctic ice extent (per ASMR-E) are the last three years. 2010 is barely staying ahead of 2003, and within a week 2009 will be below 2003.
2010 has had a steep downward gradient, but I do not evidence for the AGW claim that the late 2010 extent surge was due to southward spread existing ice, setting up 2010 for historic lows.
I would not speculate on where 2010 will end up. On one hand, we have more multi-year ice at this point in time, and it is unlikely that we will have a repeat of 2007 wind and cloud conditions. But on the other hand, we still have Asian soot to spur melting.
Also, Hudson Bay has the lowest ice in it for the last 7 years at this time of the year
thanks to the warm winter there. Warm southerly winds are still hitting it so I expect the 2010 plot to continue downwards towards the bottom of the group, as long as the Pacific side warms up a bit too.
Andy
Yes, we can expect Hudson Bay to be ice-free earlier than in the recent past so 2010’s steep gradient likely will continue for a while. However, when Arctic ice is at its minimum, the Hudson Bay is always ice-free, so I do not expect Hudson Bay’s condition to play a role in the 2010 minimum.
JAXA have finally smoothed the graph out to remove the corrections done for melt water, looks pretty good, though they have not updated. Cryosphere area graph has plunged downwards so I expect JAXA to follow.
Andy
The 2010 plunge continues. Only 2005 and 2006 had lower values at this moment.
Interesting! Joe Bastardi is forecasting continuation of rapid Arctic ice melt in 2010. He believes that the at 2007 minimum could be challenged and that the minimum will definitely come in below 2008 and 2009. My impression is that Joe Bastardi has a better record in forecasting near term results than CAGW sites such as the Met office. His belief seems to be driven by sea temperatures to date, but he forecasts that sea ice will recover nicely in 2011 and 2012 with solid recovery by next decade.
Sea temps normally are better at making the winter minimum smaller not the summer maximum larger, so that is interesting. I think he is right this year, not sure why he thinks it will swing the other way in 2011 “the Inquirer” ??
It seems this thread only has me and you on it nowadays, a very sad end to what was the best informed thread on the Arctic melt. The owner of the blog seems not to care and everyone has left. I’ll be jumping ship too soon, to the WUWT thread which has more posts but less interest in the pure data and more erring towards political points 😦
Andy
Andy,
I’m still here, sort of. The problem, which I had meant to post about on the Roadmap thread, is that since the change to WordPress comments to active threads do not stay around long enough at the home page links. There must be some settings which could alter that, like “show the 3 most recent comments for each thread that has been active in the last 7 days”. I say that because I don’t notice the same problem on WUWT. I don’t think Steve realizes how many readers and writers he has lost since last autumn when the blog management software changed. I’ve hardly seen ‘bender’ here since, and he used to be a stalwart.
Speaking of bender, he and I noticed from a graph of Tisdale’s that NH sea ice tends to lag ENSO related temperatures by about 2.5 years. However, that should make 2011 worse than 2010, so maybe it’s a rubbish theory.
Are we going to need some more plaintive songs for Baby Ice this summer?
Cheers,
Rich.
A new thread is needed for sea ice 2010. The rate of melt the last few months appears to have been anomalously high. Anyone intersted in computing the monthly melt rate statistics and comparing historical versus current?
[Rich: My absence has nothing to do with the blog software. The climategate thing was a major distraction that generated noise in all threads. That appears to have calmed considerably.]
Interesting point Rich, I bookmark this page so always go straight to it, but I have to admit I prefer the old software too.
Andy
Netweather forums in England and Peak Oil forums have discussions on sea ice that are somewhat active.
Both discussions seem to have much ‘sea ice is at record low volumes and is certainly going to be amazingly low this summer’, ‘no AGW is a scam’
2010 now below 2006.
Regards
Andy
Right now the positive Antarctic ice anamoly is higher than the negative Arctic one giving us a net gain in sea ice
Yes, I was wondering whether it would be below zero for the first time ever but no, the Antarctic ice gain is now spurting ahead quite considerably. Will be interesting to see what they maxima will be down there, it often gets overshadowed by the minima in the northern hemisphere.
I’m going to stop posting here after September 2010 if nothing is done about this once exccellent resource on the melt/ freeze, the title of this thread says it all, the owner no longer cares.
Andy
Steve – yes, a new thread on this long overdue. Sorry – been following other things, but impolite not to have rethreaded. WIll do.
I understand your frustration Andy but Steve is only one man and has been quite busy with several matters for the past several months and deserves our patience and respect.
Thanks Steve.
Michael, hope it didn’t sound if I was getting grumpy, just stating facts on the ground, obviously with climategate etc Steve has been busy, but people have stopped posting on this apart from a few diehards so the thread suffered because of that.
Andy
JAXA graph has taken a downward turn
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
maybe the value of 155k loss for yesterday will be revised though. That’s a big loss
Andy
What happens now with ice has no bearing whatsoever to the minimum just look at the graph cheers
The massive loss is obviously due to thedanged wind simultaneously compacting the ice and blowing it out of the Fram/Nares strait:
Compacting:
[http://] wattsupwiththat.com/2010/05/28/the-great-2007-ice-crunch/#more-19963
“Simple – the ice was compacting to the north as it was pushed by southerly winds. It lost area – while it gained thickness.”
Blown out of strait:
[http://] wattsupwiththat.com/2010/02/19/jpl-missing-ice-in-2007-drained-out-the-nares-straight-pushed-south-by-wind-where-it-melted-far-away-from-the-arctic/
“This fits right in to what I’ve been blogging about for two years. the 2007 record minimum ice extent was wind driven not melt driven. A significant portion of the ice did not melt in place. It was pushed south by the wind where it melted”
Agree with Stephan, I was merely commentating on the large loss.
I’m not sure what 2007 has got to do with the current downward trend? The current losses are more to do with the melt in Hudson bay and other periphery locations which is due to the warm temps as well as wind etc.
Andy
I got around to looking at WUWT for this ice being blown out of the Nares straight in 2007 causing it to be a low year, as per the headline
jpl-missing-ice-in-2007-drained-out-the-nares-straight-pushed-south-by-wind-where-it-melted-far-away-from-the-arctic/
If you look at the figures quoted in the scientific article they state just under 90 000 square km of ice was lost, double the normal amount. However that is still a lot less than the 2 million extra km ^2 lost that year. So it’s not a reason for the loss. I see Steve Goddard uses another theory, it was all pushed north where it piled on top of each other making the thickness high but the extent low. It seems WUWT will plump for anything apart from warm temperatures causing melting as that does not fit in with “the message”.
Andy
looks like there may be an unprecedented U turn in DMI ice as reality of much thicker ice concentration and la Nina starts to kick in on AMSU temps. DMI NH temps at record LOWS
stephan, I am not sure of your logic. I do not know whether la Nina will affect Arctic temperatures. Ice concentrations are a matter of dispute, and at least to some degree, a person’s perspective on CAGW seems to influence what model of ice concentrations he follows. Speaking of degrees, the measured temperature north of 80 degrees hit the melting point degree on “schedule.”
Too many people will be glued to computer monitors this summer following estimates of Arctic ice rather than being outside and enjoying this comfortable summer.
Shame JAXA have not been updating, perhaps I’ll have to go outside and enjoy the sunshine at this rate ! 😀
Andy
JAXA has updated today. Nothing significant about the trend in the last four days, except no change in pace yet to suggest that it will get back to the lines of previous years in the near future.
I am pretty skeptical about the whole AGW but in this case I agree with Inquirer. I still think that you will get a shock when minima comes in this year due to increased ice concentrations
Is this relevant in a “global” context?
Mr. Mushroom, Although your question may be a good inquiry worthy of discussion, it is not an issue to be discussed in this thread. If we go down that route, every thread tends to degenerate into an argument of whether Gore is a demon or whether deniers are worse than Nazis.
What minima extent value are you guessing Stephan?
Andy
looks like sea surface temps are in free fall
http://discover.itsc.uah.edu/amsutemps/execute.csh?amsutemps be prepared for some big falls in mean surface temps in coming months
will get back to you re minimna after I look at graphs cheers
NB in AMSU go to sea surface temps botom of menu list and click
http://discover.itsc.uah.edu/amsutemps
The sea surface was .06F cooler than last year.
Are you predicting a “free fall” in the sea surface temperature sometime soon – or are you describing a .06F degree drop in a yearly range of roughly 1 F as free fall?
reality check for AGW
re Anndyw….predicting ice minimum to be above 2009
Stephan,
Those overall SSTs could affect the Pacific side, but the Atlantic side is still warm (relatively).
I think you are brave to predict a higher value than 2009 given how much ground it would have to make up. But maybe the Arctic ice thickness really is there and that is going to save it.
Personally, for *area* minimum, I am predicting an amount a tiny bit higher than 2008 (which was a tiny bit higher than 2007, but a lot less than 2009).
Bender, it’s good to see that reports of your disappearance were grossly exaggerated.
Rich.
No comment check for yourself. Its going 2005-2006 as I predicted or even higher
http://ocean.dmi.dk/arctic/icecover.uk.php
Stephan,
Well, we’ll see – it’s currently on the 2009 curve from your DMI link, but in any case I was referring to area, as measured by Cryosphere Today. Perhaps it could be win-win and we’ll both be right!
Cheers,
Richard.
Yes if you believe the thickness then it maybe that 2010 takes a turn like 2006, otherwise it is looking more like 2008. I’ve also gone for about 2008, but a hedged bet might be better and that is between 2008 and 2009.
Andy
Has the turn started? A different slope?
http://www.ijis.iarc.uaf.edu/en/home/seaice_extent.htm
There are those who believe such a turn should be expected once Hudson Bay is done melting.
It is a shame that this thread has lost so many posters. I’m glad there are still a few here giving updates, but I really enjoyed all of the anlaysis and graphs from the last 2 years.
I think Phil, DEwitt payne and all the AGW NH crowd are going to get a very rude shock this July, August Sept and Oct LOL
http://ocean.dmi.dk/arctic/icecover.uk.php
The pole has been exceptionally cloudy this year. It is pretty much all up to the wind and water temperatures at this point. In two weeks time the temperatures begin their drop toward winter. There are about two more weeks possible where solar influence can melt the ice. A look at the NOAA2 cam shows cloudy skies and skim ice forming over the melt pools.
I still see no reason why this season won’t finish somewhat like 2006 did. It will probably be a bit less because there is still a lack of ice > 5 years old compared to 2006. Ice formed in the 2007/2008 winter won’t become 5yo until 2012 so we can’t be back to “normal” conditions (normal mix of ice of various ages) until at least 2015. If the wind stays as it is now and if the pole remains cloudy and if the temperatures remain below average as they have so far this season, then it is possible we could end the ablation season with more ice than 2006.
Anthony Watts has created a great sea ice page over at his site.
http://wattsupwiththat.com/sea-ice-page/
We can use this link for comments.
Thanks Steve for giving us sea ice junkies a place to comment.
Well it ain’t looking good for the theory
http://ocean.dmi.dk/arctic/icecover.uk.php
BTW the beauty of these posts is that they are a record of what has happened…
All warmista sea ice prediction for 2010 have been wrong where are Phil and DE-Witt Payne now? Silence is deafening LOL
Stephan, I agree that Phil always seemed to be a bit of a “warmista”, but to my mind DeWitt Payne was pretty equinanimous and fair.
It’s a shame we don’t get to see the day by day blow by blow accounts of relative melting any more. The graphs at WUWT are handy but don’t provide quite the same excitement as “it’s another sub 100K melt day”. OTOH it does allow us to “get a life” instead!
Rich.
Back on May 7th I made a prediction 2010 would drop below 2008 and 2009 at the end but stay ahead of 2007 so I still have a chance of being right!