« North Hole 2 | Main | Animation 16: temps »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

Evilreductionist.blogspot.com

Neven,

With those warm SSTs, it is possible that it will end and then restart again, if only briefly. If we get the average for the last five years, we should still see 100,000 more extent loss and end the melt season on or around 16 September. But there might be a few more surprises left in store ...

Neven

ER,

Sure, don't take my word for anything. It's huge what I'm still learning after these weeks of intently looking at graphs and maps, trying to figuring out how things work. Besides, the weather forecast can always change again (though they have been pretty consistent in the past 2-3 days).

But for this year to go below 2008, well, that would really say something about the state of the ice. I don't think it will, but if it does, I'll be flabbergasted.

PS this was my 101st post. Hurray, me, and big thanks to all commenters.

Kevin McKinney

Congratulations, Neven!

It hardly seems possible. (Like many other things this melt season! ;-) )

Steve Bloom

Hurray, you. :)

Steve Bloom

Re the (CT) area, considering that as of the 6th there was a 190K gap between 2010 and 2007, and that there've been two days of possibly pretty good melt since then, even just two or three days more could be enough to put it close. But regardless, it will be instructive to see what those lows (how stormy? -- I don't know how to read that chart) plus the warm SSTs do. And I suppose it's possible that the dipole will be back soon enough to make a difference.

davidp

G'day Neven, As we are approaching the end of melt, it would help give context to the "If 2010 loses as much sea ice extent as..." figures if you included the final extent for each of the past years.

Thanks for the series. I have been following it with interest for the last few months.
David

Neven

Hi, Davidp. I thought about adding those final extents, but figured it would be information overload. People can eyeball the graph, and quickly calculate the differences from what numbers are there.

Steve,

(how stormy? -- I don't know how to read that chart)

I don't know how stormy exactly, we'd have to go there some time and experience it personally. But if I've understood correctly, the winds are blowing along the isobars. So when there's a lot of isobars...

And I suppose it's possible that the dipole will be back soon enough to make a difference.

It is possible, but it's also possible that the DA disappears for 6 weeks like it did in July and the first half of August. If it hadn't 2010 would have gone under the 4 million square km mark, and not the 5 million one.

Artful Dodger

How do we know when the minimum Sea Ice Extent has been reached? It's easy to be fooled by the 'slosh' this time of year, as winds can either compact or spread the ice pack, and warm water replace solar isolation as the main cause of melt.

Extent measures can increase while melt continues, or Extent can go down after freeze-up begins. The pros at NSIDC handle these confounding variables by reporting monthly averages, and waiting for a period to declare the Minimum.

So let me offer an operational definition for the end of Melt Season: It is when the 31-day moving average for Daily Change in sea ice Extent (DCE) goes from a negative value to a positive value.

In the following graph, look at where the DCE trend line crosses the X-axis. Note that to make the graph more readable, I've multiplied each year's value by an incremental multiple of the Year. This insures the trend lines only overlap once, near the Date of minimum SIE.

Looking at the trend lines, it's clear that the melt season (by my definition) has ended in each of the last 8 years in a 3 day range between Sep 10 - Sep 13. A second observation is that each year, the date of the Minimum advances a small amount.

One advantage of this method of measurement is it compensates for the daily 'slosh' where a Year teases a minimum, only to retreat and then advance once again. Meanwhile, the underlying Trends are revealed and magnified. Neat, huh? The downside is we don't have a reliable declaration of the Minimum for 2 weeks after it occurs.

I have more to say on this topic, but for now I would like to hear your comments. Enjoy!

Artful Dodger

I know that Neven has quite a few Aussie followers, so I'd like to post this update to a comment I made on July 21. On that date, 2010 Sea Ice Extent fell to a level equal to the land area of the entire continent of Australia (about 7,617,930 sq.km). This represents half of the Winter maximum Extent, which is normally twice the size of Australia.

In the 7 weeks since July 21, we have lost nearly the area of the State of Western Australia, 2,645,615 sq.km. In fact, we are only -9,521 sq.km short of that mark (before the IJIS revision). While 9.5 K may sound trivial to veteran Arctic Watchers, that is the same area as the City of Melbourne, PLUS 6 Darwin's!

In these 49 days since July 21, we have averaged a loss in SIE of -53,798 sq.km per day. This is the equivalent area of 10 Perth's, each and every day.

Tomorrow, the losses continue.

Heraclitus

Using the average melt up to the minimum (in brackets after the difference from 2010, near the top of the post) is not so useful as previous months - 2008 in particular, if melt on 9th had been about 10000 less or melt on 17th 10000 more the average would have been about 18000 rather than 38000. Makes the current melt look even more exceptional.

Mike

Hi All, I'd like to propose a slightly opposite view, and that is, how is this going to effect the NH weather in the coming winter? The water is warm, I would think the ocean swell will slow surface wind and transfer energy to the water. Just a thought that there must be an effect there. Will this amplify the NAO and make it more likely to be positive. I know that James Overland has a theory on this but I don't understand weather enough to have a real opinion about this. I would think that this would also slow the winter ice growth in some places and cause a colder NH winter. maybe I have a bit of a theory!

me.yahoo.com/a/Xs1LkmQn0_XQko.cH1qfXW.Bvo0cvQ--

Just a quick question: Looking at the IJIS sea ice extent graph, it seems like the low point of 2005 and 2007 was close to the end of September. However, in other places, you refer to the date of the minimum as being closer to the middle of September. Why is there a difference?

Gas Glo

AD,

Nice and smooth progression to later dates :-)

However have you averaged away the interesting detail? Late in September there are always rapid rises and there are usually rapid falls at end of August early Sept compared with mid mis Sept. Therefore your date is effectively constrained to happen within a few days of Sept 12.

eg 2005 had a late surge in extent reduction but your graph has 2005 with earlier date than 2004 suggesting early minimum not the late minimum that people are aware of.

Fine if the smooth progression of dates is what you want but I think there is some interesting detail lost. Depends what you are interested in.

Artful Dodger

Gas Glo: The 31-day moving average for this chart takes in data from the range Aug 16 to Oct 15, so it is not constrained by the Sep monthly endpoints. Any large move within 15 days of a data point is reflected in it's value.

Using moving averages smooths out weather effects and reveals climate trends (few weather systems persist more than 2 weeks). The purpose of this particular chart is simple: it is designed to show Climate norms for the Date of the Annual Minimum SIE.

Nick Barnes

But Lodger, your chart does *not* measure the date of the annual minimum, because of the shape of the curve to which you apply your 31-day smooth. The day on which the 31-day smoothed mean extent change is zero is simply the day which falls exactly half-way between two dates, 31 days apart, which have equal extent (31-day average extent change = 0, therefore 31-day total extent change = 0). Because the curve is not symmetric (the fall is less rapid than the rise), the dates given by your chart will be consistently earlier than the actual minimum.

L. Hamilton

I noticed that both NSIDC and Uni-Bremen extent plots appeared to cross the 5.0 line some days ahead of IJIS; and Uni-Bremen at least shows 2010 now within a hair's breadth of 2008, rather than almost 200k above it as IJIS does.

Anyone have thoughts about these apparent differences, or am I over-reading the graphics?

Gas Glo

AD re Nick Barnes comment:

IJIS extents minimums:

09/09/2002
18/09/2003
11/09/2004
22/09/2005
14/09/2006
24/09/2007
09/09/2008
13/09/2009

average date 15th Sept with an almost negligable trend towards later dates.

me.yahoo...
Area minimum usually occurs before the extent minimum. The area is still going down 3.088m km^2 today per CT. Area drops perhaps getting smaller now while extent drops are still considerable for time of year.

Some freezing near pole has little effect on extent as the areas are fully counted for extent anyway but for area it is relevant. So earlier area minimum where area freezing counteracts areas lost by melt, compaction and spreading out to less than 15%. Whereas extent minimum is where melt etc ceases.

Neven

how is this going to effect the NH weather in the coming winter?

Mike, that's an excellent question. I might do a blog post on lake-effect snow, because that could be a side effect of all that warm sea surface releasing moisture and heat in the air. I'm not sure but it might cause some more of those severe blizzards we experienced last NH winter.

Looking at the IJIS sea ice extent graph, it seems like the low point of 2005 and 2007 was close to the end of September. However, in other places, you refer to the date of the minimum as being closer to the middle of September. Why is there a difference?

Me, is your question addressed to Artful Dodger or do you mean something I did in this SIE update? Ah, I see now Gas Glo commented on this. Yes, dates for area and extent minima differ in each given year. It has to do with the way both are defined. When the water starts to refreeze you immediately notice this in area, because every percent above the 15% sea ice concentration threshold is recorded. But for extent it doesn't matter if a grid cell is covered with 20% or 80% ice, it is still counted as 100%. So it takes a while for extent to start picking up. I hope I explained that correctly.

Anyone have thoughts about these apparent differences, or am I over-reading the graphics?

Larry, doesn't this just have to do with different logarithms employed by different agencies, etc?

L. Hamilton

Larry, doesn't this just have to do with different logarithms employed by different agencies, etc?

Hah, no, I bet they all use the same logarithms! -- Sorry, Neven, that was too good to pass up :-)

Chartsgraphs.wordpress.com

Lodger

I get different Arctic Sea ice Extent minimum dates. This chart shows that 2007's minimum was on 9/24/07, outside your Sept 10-13 range.

Kelly

L. Hamilton

But on a less jokey note, I'm sure you're right that the various extent measurements reflect differences in their data sources and algorithms. In this case we're assigning symbolic importance, at least, to where 2010 stands with respect to 2007 and 2008, so I thought the variations were interesting.

Neven

Hah, no, I bet they all use the same logarithms! -- Sorry, Neven, that was too good to pass up :-)

Well, don't expect any smarter answers from me. That's the best I could do. :-D

Neven

But on a less jokey note, I'm sure you're right that the various extent measurements reflect differences in their data sources and algorithms. In this case we're assigning symbolic importance, at least, to where 2010 stands with respect to 2007 and 2008, so I thought the variations were interesting.

They sure are. I was planning on doing a post soon on a comparison of all the darned area and extent graphs I can find.

Neven

Hmmm, the ECMWF weather forecast has changed quite radically since the last time I checked. Now they are expecting the Arctic to be divided up by broad fronts of highs on the US/Canadian side and lows on the Siberian side, followed by a conquest of high-pressure areas. Hmmm....

Neven

Big upward IJIS revision of 12K, no new PIPS ice displacement forecast map for tomorrow yet. Is this the end of the slide?

Gas Glo

I am sure you could do better now replacing logarithms with algorithms ;o)

I am interested in the differences between NSIDC and IJIS. Thinking about targets now, there is:
1. Area 3.004 m km^2 (2008 level) CT
2. Area 3 m km^2 CT
3. Area 2.919 m km^2 (2007 level) CT
4. Extent 4.719 m Km^2 (2008 level) IJIS
5. Sept average extent 4.935 m Km^2 (Stoat fails to win bets) NSIDC
6. Sept average extent 4.735 m Km^2 (Stoat loses bets) NSIDC
7. Sept average extent 4.68 m Km^2 (2008 level) NSIDC
8. Sept average area 2.93 m Km^2 (2008 level) NSIDC

How many of these are at risk?

For 5 and 6 I have taken IJIS numbers applied movements from years 2003 to 2009. Then assuming a fixed difference between IJIS and NSIDC applied difference on 4th Sept between IJIS number and an estimated from graph posted here by Jack? 4.97m km^2. The numbers I ended up with are:

4846656 4891599 4744880 4779328 4685932 4913401 4814474

1 under 4.735 rest in push territory.

However the assumption of a fixed difference may be suspect. Perhaps the difference will reduce?


Lord Soth

Well it it does end now, 2010 will be in a strong third for extent, a virtual tie for second on area, and taking first place be a huge margin on volume.

With the update, we have stayed under 5 million sq. km for extent, and we have ended the upward trend in 2008 and 2009 on extent.

If we didn't have six weeks of poor weather and a reversal of the beaufort gyre, we would not just be looking at a hole at the pole, but clear sailing to the pole.

Also of note, in 2008 the minimun extent date would have been Sept 17, if it could just lose another 10K sea ice loss. So maybe we should not look at the date of the actual minimun, but the date of the last decline before sustained recovery, when predicting if the date is getting latter every year.

Neven

I am sure you could do better now replacing logarithms with algorithms ;o)

*turns red in the face*

Now I get it...

Well, at least I didn't say "goalrithms" as the avid soccer/football-fan that I am.

5. Sept average extent 4.935 m Km^2 (Stoat fails to win bets) NSIDC
6. Sept average extent 4.735 m Km^2 (Stoat loses bets) NSIDC

I've bet Stoat EUR 20 on that one. I forgot it was based on NSIDC monthly average. NSIDC extent tends to be lower than IJIS, right?

Jon Torrance

"How do we know when the minimum Sea Ice Extent has been reached?"

Hindsight.

Seriously, I doubt there's any way to be sure until after the extent has climbed significantly on each of several days in succession.

Gas Glo

>"NSIDC extent tends to be lower than IJIS, right?"

Daily NSIDC numbers tend to be lower than IJIS daily numbers, but last years monthly average 4.36 was higher than IJIS daily minimum of 4.25. IJIS average for Sept was 4.38 so there was very little difference over Sept last year but there was a clear difference on 4 Sept this year.

L. Hamilton

Watching the movie on CT, it seems the Bering side of the ice pack is still melting, wispy patches of "blue" ice vanishing like smoke in the wind. Meanwhile on the Barents side, the ice front expands southward, but does that by drawing concentration down farther north. So these two areas of change exert opposite force on extent, but both are still pulling ice out of the central Arctic.

L. Hamilton

Near as I can tell from the NSIDC graph for Sep 8, extent is about 4.8. And that should be a 5-day average, yes?

For comparison, NSIDC mean Sep 2008 extent was 4.68. Other recent Sep means are:
2000, 6.32
2001, 6.75
2002, 5.96
2003, 6.15
2004, 6.05
2005, 5.57
2006, 5.92
2007, 4.30
2008, 4.68
2009, 5.36

Gas Glo

My eyeballs plus tape against screen suggested 4.75m and a difference similar to on 4th Sept with NSIDC around 220k lower than IJIS.

comparing NSIDC monthly average to IJIS daily minimum and IJIS monthly average:
2009 5.36 vs 5.25 vs 5.38
2008 4.68 vs 4.71 vs 4.84
2007 4.3 vs 4.25 vs 4.38
2006 5.92 vs 5.78 vs 5.91
2005 5.57 vs 5.31 vs 5.53
2004 6.05 vs 5.78 vs 5.95
2003 6.15 vs 6.03 vs 6.13
(not doing 2002 because too much missing data)

Just speculating really but:
The algorithms could be the same but the grids of different sizes and/or positions but I doubt this would make much difference.

Another possibility is the algorithms are the same but the calibrations of satelite signal to coverage % is different. If this is the reason for the difference then the difference above could be likely to form a trend whereby the greater the areas of low concentration ice the larger the lower NSIDC becomes compared to IJIS.

Does that sound plausible?

Patrice Pustavrh

Well, I've checked DMI and Arctic Roos (SSMI) graphs today. Both of these graphs, which were above 2009 before are now below. I think this fact has removed another (and one of the last except for PIPS 2.0 pixel counting) argument Goddard has used that ice is actually getting stronger this year comparing to 2009.
Well, nevertheless, even if his arguments are wrong, I thank him anyway. He showed some mechanisms and to be true, even his wrong and at first glance correct claims helped me to understand the process of melting slightly better.
But, I would also like to show my greatest respect to Neven and Patrick Lockerby. Guys, you do fantastic work. Yet after this year is over I think, that everyone here including me and deniers even more, have forgotten one very very important issue: Change in nature (I may not be precise due to my not so good English, but I mean climate or geological or iceological changes) are extremely slow. And the changes we are seeing in one decade are enormous given such amount of time. Even if prediction that Arctic won't be ice free between 2020 is correct, the amount of loss of ice from 1980 till now is extremely fast for such a big system such as Arctic is.
But anyway, it is fun to watch, it fun to learn and it is fun to watch debate and sometimes express your point of view and do a really decent debate.
And, now back to work: Is ice melting even faster than previously expected ? I think yes. But don't be surprised that even with this trend even ice volume may increase from single year to year. But, it's going doooowwwwnnn HIIIIIGGHHHWAAAAAAAYYY (think of AC/DC here).

Kevin McKinney

Patrice, that's a good point that you make about timescales: many interactions I've had with deniers turned upon the question of timescales in one way or another.

It may be rate of (climate) change as you mention, it may be difficulty really understanding the difference between "warmest in 1500 years" (as per Mann et al. 2009) and "warmest in history (which itself is ambiguous, possibly meaning "human history" and possibly meaning "earth's history".) Or it may be a matter of explaining why a prediction about hurricanes that referred to a timespan up to 2100 CE really can't be said to have been falsified in 2010 CE, no matter what the storm data say so far!

Steve Bloom

An IMHO illuminating point about deniers is how they manage to completely ignore the rapid and fundamental changes in atmospheric circulation that have already taken place. These include a host of changes driven by the expansion of the tropics, including a rise in the tropopause height, the poleward shift of the climate zones and the major components of the atmospheric circulation, including the Hadley and other cells, the climate zones, the storm tracks and all but the southern polar jet (although the Antarctic Circumpolar Current has accelerated), and the appearance of the Arctic Dipole Anomaly and a major shift in ENSO (El Nino modoki).

Of course, it's not just the deniers. Humans just don't seem to be wired to comprehend the seriousness of these changes, and the fact that the lagged response of the system means that these changes and their effects are going to get much, much worse.

Steve Bloom

So for CT area, a drop of 22k from the 6th to the 7th, close to the average decrease for the prior four days. This leaves 84K between 2010 and 2008, and 168K between 2010 and 2007.

Steve Bloom

Joe Romm has a new post featuring Julienne Stroeve's current view. It's thin ice and SSTs, she thinks.

BTW, Neven, I will never forgive you for ruining chances for any sort of new record by referring in the header to a continuing slide. "The seesaw continues" would have done the job nicely, but no... :)

Christoffer Ladstein

Steve Bloom: Haha..so you believe there will go "troll in words", as we might say in Norwegian!
Neven is powerful, by making this impressive blog he proves that, but why not allow the nature to surprise us, over and over again...

Though i am truly thrilled by the effort, energy and knowhow some of you are showing, what profession are you actually into?

Keep up the good work, the less skilled, but just the same devoted, are bedazzeled by your comments and the situation of the Arctic. (
Poor Ice-Bears, we have to build some fake icebergs for them in a few years ahead!)

Christoffer Ladstein

A closer look into the IJIS chart, comparing september 1. and 8., you'll see that the compaction rate north & northeast of Svalbard has NOT decreased, which would be natural due to the last week iceflow across the NorthPole bassin towards Fram Strait. This will hencefort lead to a stronger compaction potential, and the Iceflow will not stop that soon just because of a new weatherregime. A Supertanker is not that easily to make come to a halt! Therefore i support those of you believing in sep. 16. (or later!), as a most likely day of minimum.
So such a slide never witnessed before in September, will bring us below 4,8, I dare say!

adelady

I just did my customary CT 30 year comparison for 6 September, 1980 & 2010. Steve, if you want a pretty picture that's also pretty startling to show a denier friend, this would be a good one.

It is interesting and even exciting, but these comparisons bring me back to earth with a nasty bump.

logicman

Seesaws and slides. In Britain we often say that what you gain on the swings you lose on the roundabouts. We have ourselves quite a playground here, don't we? ;-)

Neven: I just finished my latest update and have linked to this new update of yours. I've been lurking while I get my blogging energy back, so nowhere near as many comments as I would like.

Keep up the good work, everyone. Neven's Arctic blog can rightly be called a science 2.0 blog. Spread education, knowledge and scientific discovery freely through collaboration. Nice!

http://www.science20.com/chatter_box/arctic_ice_september_2010_update_1

My best to all here,
Patrick Lockerby

Artful Dodger

No need to use screen and tape when estimating the current SIE value from NSIDC: for Windows Users, when viewing the chart, hit 'PRTSC' (PrintScreen) and then paste the clipboard into MS Paint. Then use the Line tool to draw vertical and horizontal line to the 2 axis. Hint: Hold the 'SHIFT' key down while drawing forces MS Paint to draw a perfectly horizontal or vertical line. Then estimate the SIE from your new tick mark on the Y-Axis.

The estimate of 4.8M upthread was pretty good! NSIDC announces their values to 2 decimal places, so for today it'd be rounded to 4.79 M sq.km. Here's an example:

Artful Dodger

Steve / Jack / Neven: You've all received Shout-Out's (and links) from Joe Romm on the Climate Progress: (kudos)
http://climateprogress.org/2010/09/09/julienne-stroeve-nsidc-scientist-arctic-sea-ice-melt/#comment-295422

NSIDC's Julienne Stroeve also says:

"All the old ice that was transported into the Chukchi Sea over the winter is gone and about 50% of the old ice in the Beaufort Sea remains.

Arctic-wide there has been a 60% reduction in ice ages 5+ years from the end of April to the end of August."

Steve Bloom

A commenter at CP posted a link to this interesting animation of radar images of the basin from 9/1 through current. It's a bit psychedelic-looking, but more photo-realistic than the UniBremen graphics.

JackTaylor

Artful Dodger - Steve Bloom, if my cheap server hosting package exceeds bandwidth for the month I will be posting the animation(a) on a different server, my political site.

I lurk in Joe Romm's ClimateProgress .Org blog, but have chosen to post only in Neven's blog because the political and Ad hominem climate is minimal.

Was late coming on scene, so perhaps next year will see if animations can be massaged to fit the page space width - height restrictions of blogs, send to Neven and perhaps Joe and Patrick Lockerby "logicman", another blog I lurk in, and if they want to use them so that viewers do not have to go off blog-site, will be testing the water(s).

Updating animations on a daily basis, unless "honey-do's" have precedence.
IARC-JAXA image usually updated by 2400-UTC (my local 8:00-PM EDST USA)
http://www.polk-nc.com/agw/arcA15end.swf online usually within 30 minutes.
Uni-Bremen does not remove their degree pointers until about an hour or so later so usually UTC next day at 0200 when http://www.polk-nc.com/agw/breA26end.swf is available updated online.

Evilreductionist.blogspot.com

A smaller drop today, although still decent for this time of year. We are now on 4,965,781. On average, we should see about 90,000 more go, with 16 September the minimum.

It should be noted that we have lost 363,594 already this September, which is the most ever (this includes the small gain in extent on 1 September).

me.yahoo.com/a/nSjChi4X3vr8X3DRw93GkY1.cerja.8nvWk-

A great MODIS view of the ice in the Arctic basin breaking up and extruding through the gaps into the archipelago:
http://rapidfire.sci.gsfc.nasa.gov/realtime/single.php?2010252/crefl1_143.A2010252233000-2010252233500.250m.jpg

Charles Wilson

ALL the indexes on 1 page.... Long numbers are actual DATA, rest Graphs. Sources below.

Year:_MINIMA_(07-09):___2007_____ 2008 ___ 2009 ___ 2010: Spt1 ___ Spt 8

AREA AMSR Cryo/ijis__2.9194391_3.0035558_3.4265975__3.2442__3.0876937
30%Ext.SSMI DMI______ 3.06_______3.41_____3.84______ 4.00_____3.80
AREA AMSR U.HAMBURG___3.50330___3.81714__4.22780__ 3.98389(Aug31)
Next 4 are all from Bergen U.:
AREA AMSR @ Topaz___???.?______???.?____4.40______ 4.07_____4.89 T
AREA SSMI Norsex_____ 3.62* ______3.87*____4.61*_____ 4.30_____4.20
AREA SSMI Nansen_____3.293737 ( ~ NORSEX*see note)__4.3507__4.120683
AREA SSMI @ Topaz____???.?______???.?____4.82_______4.49_____4.16 T
15%Ext AMSR HAMBURG___3.85724___4.27124__4.80018__ 4.77826
15%Extent_AMSR_Bremen_4.32*_____4.64*____5.31*______ 5.00_____4.67
"Extent SSMI NSIDC______4.17______???.?____5.10 ______ 5.2______4.78
"(a Blogger=NSIDC data)___4.24 ______4.64_____5.24__(ends)
"Extent JAXA/ I J I S____4.267344 __4.707823__5.291094__ 5.332344_4.965781
"Ext.AMSR @ Topaz______???.?______???.?___5.47______ 5.42_____5.07 T
"Ext.SSMI NORSEX/ROOS_4.74______5.23_____5.88______ 6.17_____5.79
"Ext.SSMI @ Topaz_______???.?______???.?___6.04______ 6.26_____5.87

Notes:
(1) Bergen U. added 10% to Older Data to match a Newer Algorithm ( = why the Nansen/Norsex Discrepancy is ONLY in 2007). NSIDC kept the Old Algorithm; Bremen adjusted by 1 of the methods... maybe added 7 or 8 % ?? NSIDC explanation is ~ 60% down the page at http://nsidc.org/data/docs/noaa/g02135_seaice_index/index.html#acquisition_proc
...(2) SSMI = DMI, NSIDC, & Norsex/Nansen ... JAXA, Bremen, Cryo, use AMSR.
(3) I have assumed TOPEX's legend has SWITCHED AMSR & SSMI, as AMSR typically gives Lower values elsewhere - - I COULD BE WRONG.
(4) Topex excludes East Greenland Sea & Okhotsk Ice.
Sources:
Cryo AREA: http://arctic.atmos.uiuc.edu/cryosphere/timeseries.anom.1979-2008
DMI 30% Ext.@ http://ocean.dmi.dk/arctic/icecover.uk.php
Bergen University:
_Nansen AREA @ http://www.nersc.no/main/index2.php (bottom of page)
_Topex @ http://arctic-roos.org/forecasting-services/topaz/topaz-model-forecast
_NORSEX/ROOS @ http://arctic-roos.org/observations/satellite-data/sea-ice/ice-area-and-extent-in-arctic
HAMBURG Area & Ext. data @ ftp://ftp-projects.zmaw.de/seaice/AMSR-E_ASI_IceConc/area-extent/
BREMEN AMSR @ http://iup.physik.uni-bremen.de:8084/amsr/ice_ext_n.png
NSIDC: http://nsidc.org/data/seaice_index/images/daily_images/N_timeseries.png
JAXA Extent @ http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv
... "Cryo" Data is actually from JAXA/ijis calculations.

Charles Wilson

Does anyone know why these NOAA maps show SO little Ice ? http://www.osdpd.noaa.gov/ml/mspps/seaiceprd.html
A quarter of the Center is just a gaping Hole ! From the color chart, they do not show Ice of under 25% concentration. So, like DMI, it will be less but it is Far less than DMI - - & DMI uses a 30% cutoff !

me.yahoo.com/a/Xs1LkmQn0_XQko.cH1qfXW.Bvo0cvQ--

Thanks for the answers (I posted somewhere around 17 or 18 in the blog.) I think I understand it now: when everyone is estimating that the ice is about finished decreasing based on the date (middle of September), they are generally referring to area and/or volume while extent is more variable due to compaction and is not necessarily as meaningful. In any event, all three are dependent on the particular 'logarithm' use to calculate the values. Meanwhile I enjoy the blog and am not intentionally logging in as 'me', that's what yahoo gives me as a log in and I don't know how to/don't think I am here enough to figure out how to change it.

FrankD

For those who followed the saga of Lockerby Island (or Petermann Ice Island, if you prefer), check out this:
http://ocean.dmi.dk/arctic/images/MODIS/Kennedy/201009091604.ASAR.jpg

It ran aground on an island at the mouth of Petermann Fjord and after several days a getting stressed finally snapped clean in two.

So I guess its now Petermann-A and Petermann-B (B is pretty screwed I'd say, smaller thinner and on the way south)

Neven

Poor ice island, sniff.

Thanks, Frank!

Gas Glo

AD,

"NSIDC scientist Julienne Stroeve to explain what is going on. I did, and she replied:

We’ve dropped to 4.76 today."
from
http://climateprogress.org/2010/09/09/julienne-stroeve-nsidc-scientist-arctic-sea-ice-melt/#comment-295422

So my 4.75 was nearer than both 4.8 and 4.786 :-P

Gas Glo

>"A smaller drop today, although still decent for this time of year. We are now on 4,965,781. On average, we should see about 90,000 more go, with 16 September the minimum."

I get 112k or 98k depending whether I include 2002 or not (looking at individual years then averaging rather than averaging to get minimum date) but am not too worried by that difference, my question is whether we should expect it to be more because of:
a) the steep declines recently seen
b) area still reducing
c) other factors like weather forecasts, PIPS displacement ....

dorlomin

For those who followed the saga of Lockerby Island (or Petermann Ice Island, if you prefer), check out this:
http://ocean.dmi.dk/arctic/images/MODIS/Kennedy/201009091604.ASAR.jpg
= = = = = = = = = = = = = = = = = = = = = =
Might it make it into the Atlantic?

Phil263

It is interesting to compare the Uni Bremen map on 9 September between 2008 and 2010 .
On 9 Sept 2008, SIE reached a minimum of 4,707,813 which was almost 260 k lower than today. However, the configuration of the ice extent was very different from this year. There was much more 100% compactness, but more importantly, apart from a tongue poking out towards the Canadian North West Territories, most of the ice was confined to within 80N. In contrast, the 2010 map shows a narrow peninsula of loose ice extending south of 75N towards eastern Siberia. That ice is less likely to be affected by an early freeze up and therefore we might see further melt before the minimum is reached.
DMI's 30% SIE is still pointing sharply downwards, and Unibremen's SIE curve is showing no uptick at this stage, so we should still expect further losses tomorrow.

me.yahoo.com/a/nSjChi4X3vr8X3DRw93GkY1.cerja.8nvWk-

The other 'me', I had the same problem logging in via yahoo, it wouldn't use my name despite my efforts! Interesting developments with the ice island which should now be able to pass down the Nares strait without too much trouble.

Christoffer Ladstein

"Might it make it into the Atlantic?"

Dorlomin: I would fear more for oil platforms outside New Foundland (Hibernian), than a new Titanic incident! Evacuate!!!
More seriously speaking, what time span are we talking until it will drift down to those latitudes?

Gas Glo

Smaller PIPS arrows particularly in Fram Straight

JackTaylor

Phil263 | September 10, 2010 at 13:29
"2010 map shows a narrow peninsula of loose ice extending south of 75N towards eastern Siberia. That ice is less likely to be affected by an early freeze up and therefore we might see further melt before the minimum is reached"

Phil I see a lot of INSTABILITY in your peninsula with opinion it is a contributing factor to not as much decrease in extent for the past day or so by spreading out on the 9th, am expecting it to melt and compact significantly. There must be a lot warm water allowing the ice to shift about so much. A lot of "albedo effect" heat absorbed in the water during the heavy June melt about three (3) weeks before & after of the Summer Solstice resulting in a lot of latent water heat remaining to dissipate. See:
http://www.polk-nc.com/agw/philsTongue.html
with expanded - magnified view at
http://www.polk-nc.com/agw/philsTongue.swf

Steve Bloom

Thanks, Jack. Is it possible to track how many hits you got from that? I'd be curious to know how many people are payingattention to that degree. I agree that the "peninsula" looks delicate.

Steve Bloom

A 15K CT area drop from the 7th to the 8th, leaving it 69K behind 2008 and 153K behind 2007.

r w Langford

The one unknown in much of this years ice melt has been deep water temperatures. Maslowski states that sixty percent of the melt can be attributed to water temperature. Deep water heat influx must be having a major influence on the central basin for us to see the open water near the pole at this time of year. Here is the reference. http://soa.arcus.org/sites/soa.arcus.org/files/sessions/1-1-advances-understanding-arctic-system-components/pdf/1-1-7-maslowski-wieslaw.pdf Water temperatures may continue to push this seasons melt beyond what winds and weather might indicate.

JackTaylor

Steve, I'm not running "hit counters" on any of those animations,but will look at the server stats request for each *.swf animation to see what's there. Prior to your link in ClimateProgress (Joe Romm's), 90% of referral requests were from Neven's Blog (here) because it was the only site I had linked them from, but that doesn't account for people with bookmarks or E-mail links... Will get back to you on it... I'm retired and the stuff I avoid may take me a couple of days.

Today's (09-10) IARC-JAXA images shows our delicate peninsula having some ice growth area????? http://www.polk-nc.com/agw/arcA15end.swf

r w Langford

There is a nice visual for the Arctic Ocillation here. http://www.cpc.ncep.noaa.gov/products/intraseasonal/z500_nh_anim.shtml There is a repeat of the high over the Beaufort happening now that we saw at the end of August. Not being a climatologist someone else will have to interpret it for me.

Steve Bloom

Jack, we are talking about the one sticking into the ESS, right? If so it looks like continued melt/retreat to me.

FrankD

Christoffer: More seriously speaking, what time span are we talking until it will drift down to those latitudes?

Comparing my original link to the latest image:http://ocean.dmi.dk/arctic/images/MODIS/Kennedy/201009101713.ASAR.jpg
and assuming that it doesn't run aground somewhere else (it has crossed to the west side of the channel, which makes that less likely anyway), the answer is probably "not long"

When it was stuck in the fjord, it was moving forward at 2 to 6 kilometres per day. But in the 25 hours between those images it has moved nearly 40 kilometres down Kennedy Channel. At that rate, it should clear Hans Island (the big island in the satellite pix) late tomorrow and Cape Jackson late the following day, which will take it into Kane Basin, the next part of Nares Strait. It's more likely to slow down once it gets into Kane Basin, which is several times as wide as Kennedy Channel, but it could be entering Baffin Bay proper in about 10-15 days.

But I don't think it will threaten anything in Newfoundland this year. Baffin Bay is 1500 km's long, and it will take a couple of months to traverse that. I think it will get frozen in before it can escape - maybe around mid-November, near the southeast corner of Baffin Island - and survive the winter. That stretch of coast starts freeing up around early to mid-July, at which point it might become a hazard to shipping.

Again, all that assumes it doesn't go aground anywhere, that it doesn't just fragment etc etc.

Petermann-A (the bigger part) is still hooked up on Joe Island, and doesn't look to be going anywhere. If anything, it seems to wedging itself more firmly in place.

JackTaylor

Steve: "we are talking about the one sticking into the ESS, right? If so it looks like continued melt/retreat to me"

About 170E and 75N thats the one.

The growth area is "wisp" about 176W, across the Prime Meridian, 76N - 77N,
but it also appeared around 08-31 through 09-02 our first uptick.
My Uni-Bremen "animation" has 09-10 as of a few minutes ago.

r w Langford

Jack and Steve. Cryosphere today shows dramatic ice loss in the area you are watching. It may be gone in a few more days if the water is warm enough. I will paste the URL but it fails to run for me in the preview for some reason. http://arctic.atmos.uiuc.edu/CT/animate.arctic.color.0.html

Kevin McKinney

IJIS has 4,948,750 for the 10th's prelim, so I guess we haven't hit the bottom of the slide yet. (About 240K above 2008's minimum, and about 120K above September 10, 2008. NB--the 2008 minimum came on the 9th!)

Artful Dodger

Here is an IJIS Western Arctic map for Sep 01-10, 2010:

logicman

Tip of the hat to FrankD for the heads-up on the ice island breaking in two!

New article:
http://www.science20.com/chatter_box/petermann_ice_island_now_there_are_two

Enjoy!
Patrick.

Artful Dodger

Well apologies My first animation seems to run backwards!. Photobucket seems to have cached the old animation, even after I've uploaded a corrected animation.

If you can't see the 'Date' in the bottom right corner of the animation above, then follow this link for the updated one:

Neven

Link to animation fixed, not running backwards anymore. That was very confusing, Lodger! :-p

Thanks for the link, Patrick. I'll put it under Tips straight away.

Steve Bloom

Jack, I draw the same conclusion from Dodger's animation. There's a lot of sloshing about, but the overall effect seems clear.

JackTaylor

"Cryosphere today shows dramatic ice loss in the area you are watching"
r w Langford | September 11, 2010 at 05:12

Yes, more than a few of us have "zoomed in" on that area, and as of about 29-AUG there was considerable area/extent in that quadrant, if the peninsula completely disappears soon the numbers should dive like a rock. Also, I've been fascinated (obsessed?) by the 'whole' eastern half, Asian side, of the Arctic for weeks and it probably shows in my animations.

Artful "Lodger" Dodger, if that is your 1st animation you've been holding back on me.
Excellent work with the "cropping" (no jumping about) & hitting the same pixels exactly.
I look forward to seeing more of your work as "ani-gifs" are very good in Blogs without having to go off-site to view.

Steve, "sloshing about"
Agree, and until it hardens in place the Fat Lady will not want to sing.

Artful Dodger

The most striking thing in the W. Arctic animation is how the same Ice comes and goes 3 times between Sep 4 to 9. I'm referring to the inside of the big clam shape straight North of the Chukchi Sea (centre left of the animation). This is not 'slosh', but sea ice that's teetering around 15% concentration. This area seems to finally be gone by Sep 10, but we'll see tomorrow... this is why it's important to look at IJIS maps if you're a Daily watcher of IJIS Area and Extent figures.

r w Langford

Jack. Your isthmus is very visible in yesterdays Modis image. There is hardly anything left of it. http://rapidfire.sci.gsfc.nasa.gov/subsets/?subset=Arctic_r05c04.2010253.terra.250m

Steve Bloom

CT area shows an increase of 29K from the 8th to the 9th. It had a larger single-day increase five days prior to that but then went on to decline more, so we shall see what tomorrow brings.

JackTaylor

Steve, re: your September 10, 2010 at 21:28 about hits from ClimateProgress Link,
116 for 69.6MBytes to 2200 UTC 20100911.

My 1.0-GigaByte transfer may be safe if requests taper off as Joe's topic becomes older.
It's no problem for me to FTP files to another host, just have to go through reposting URL's here if people seem to maintain interest, but as the month proceeds I expect usage to decrease.

Artful Dodger

Neven: I've updated the IJIS Western Arctic animation, now spanning Sep 01-11. Could you please adjust the link above to this latest version?

Neven

Could you please adjust the link

Done. :-)

Greg Wellman

I guess we're back on the see-saw. Preliminary 10k increase. Adjustments have been smaller than that for the most part. But Sept 10 would be a little early for a minimum, so there's a good chance we'll drop a bit more in the next few days.

Artful Dodger

As of Sep 10, 2010 has set new September Records for Average Daily Loss, and Total Decrease in SIE for the Month:

Year   SepAvg  Sep Dec
2010 -37,125 371,250
2007 -14,688 352,500
2005 -15,263 335,782
2008 -35,781 322,031
2003 -13,898 250,157
2009 -15,180 197,344
2006 -13,784 192,969
2002 -20,885 187,969
2004 -12,131 133,437

FrankD

New IJIS Area/Extent chart here: http://img408.imageshack.us/img408/9739/areaoverextent100910.png

Bottomed out on 2/9/10 at 78.2% (2nd lowest after 2007). Climbing now, but still tracking WAY below other years. In essence, it would have to compact ~200,000 sq kms without losing any area to be tracking in the same range as the last couple of years. Still very loose and mobile - we'll see what that brings...

The thought occurs that comparing extents from the sources that cutoff at 30% with those that cut of at 15% should give similar results (ie that 2010 is the lowest as of now).

In other news Petermann-B has passed Franklin Island (wrongly referred to in my previous post as Hans Island, which is a small island further up the channel). Not without incident, it looks like it grounded on the coast of Ellesmere before bouncing off back into the middle of the channel.

me.yahoo.com/a/nSjChi4X3vr8X3DRw93GkY1.cerja.8nvWk-

Peterman-A (main piece) has appeared to release from Joe Island and looks like it's swinging into the mainstream of the Nares strait.
http://rapidfire.sci.gsfc.nasa.gov/realtime/single.php?A102551120

Phil.

L. Hamilton

From Thorleif's blog on the Northern Passage today (9/12):

We are still surprised and worried about the high water temperature. At the moment we are registering around 7 to 8 degrees Celsius, which according to the experts is far higher than normal.

The “Northern Passage” is a happy ship at the moment. She is now entering the waters above the continental shelf in the southeastern part of the Beaufort Sea. Depth has decreased from 2000 to 80 metres in just the last two hours. On the sea chart we can observe some circular features described as pingos. This is submarine ice, columns of permafrost columns rising 20–30 metres above the seafloor …

http://www.ousland.no/blog/

Neven

As of Sep 10, 2010 has set new September Records for Average Daily Loss, and Total Decrease in SIE for the Month:

Well, what do you know? I wasn't even aware of that. Thanks, Lodger.

There was a huge revision by IJIS today, making the extent bounce up 33,593 square km. That's not a seesaw, that's a trampoline!

Bottomed out on 2/9/10 at 78.2% (2nd lowest after 2007). Climbing now, but still tracking WAY below other years.

Thanks a lot for that, FrankD! My CT area/IJIS extent graph also/still has 2010 tracking way below the other years as well, as you can see in the graph in this SIE update.

Peterman-A (main piece) has appeared to release from Joe Island and looks like it's swinging into the mainstream of the Nares strait.

Thanks, Phil.! (so you're one of the 'Yahoo-mes', I wasn't aware of it, good to see you)

JackTaylor

huge revision by IJIS today, making the extent bounce up 33,593 square km. That's not a seesaw, that's a trampoline!
Yes, and according to IJIS on 09,15,2007, 4267813 with a bounce up of 42,500 to 09,20,2007, 4310313 before ending at 09,24,2007, 4254531 means I'm not ready to turn up the volume on my hearing aid to understand the Fat Lady Singing.

Made a new animation starting on 7th September when the last over 50K decrease to run daily until minimum extent is declared. Is it only going up or will it come down some more?
http://www.polk-nc.com/agw/extS07up.html
Caution: if you use the swf version it becomes so 'pixelated' detail is blurred.

toby

Like eveyone else, I've taken note of teh uptick.

However I notice on the University of Cologne weather chart that High Pressure seems to be on the way back into the Arctic, and the PIPS displacement arrows are increasing again.

Does this mean a few decent days of compactification coming up?

Neven

Toby, if only the PIPS ice displacement maps weren't stuck at 9-11 we'd have a better idea of what's currently happening.

WRT to the high: according to the ECMWF forecast (big) highs will be entering the Arctic Basin again, but because they aren't stably positioned over the Beaufort Sea, it really looks as if the Arctic Dipole Anomaly is gone. So what happens now is anybody's guess. The steady decrease we witnessed in the past week were because conditions were very stable. They aren't anymore.

Christoffer Ladstein

Not able to tell the significance of it, but the AIR temperature north of the 80 degrees, have certainly been dropping fast the last 10 days, a TRUE Rollercoaster dive! http://ocean.dmi.dk/arctic/meant80n.uk.php
In recent years, there has always been a hickup upwards after such a downhill, but will this year prove different?
Harsh weatherconditions will lead to a fast "false" freeze up thanks to a thin crispy slush covered with heavy snowfall? Which, of course also might be ripped apart again before the Fat Lady truly sing a hymn to King Winter!

Steve Bloom

CT area shows an ~11K decline from the 9th to the 10th after the previous day's sharp (for this time of year) rise, putting things close to where they were three days prior.

Phil263

Over the past few months, the way the IJIS revisons went was usually a sign of the next day change ( may be to do with the averaging). As we had a 23k upward revision last night, it is very likely that we will have another big increase for Sept 12. possibly pushing SIE over 5 million. Even if we see further ( small) daily decreases in the next few days, I doubt very much that we will get near 4,95 million again. My own guess: the fat lady did sing on the 10th and we didn't hear her!

Jon Torrance

Does anyone else get the feeling Phil263 is trying to counteract Neven's alleged jinxing power?

Aug 21 - "SIE only 16k down on Jaxa today.... Am I wrong or are we already reaching the minimum?"

Aug 26 - "I would be suprised if we see any considerable reduction in extent from now; 200- 300 k may be, but we should still be able to end up under 2009."

Aug 30 - "An increase of nearly 10,000 sqkm is quite significant for this time of year. Nothing similar was ever recorded in August in Jaxa's records (going back to 2002). I note that Arctic Roos also show a significant uptick. Maybe the fat lady sung early this year?"

Sep 1 - "IMHO, on the basis of previous years patterns, it now seems unlikely that SIE will drop below 5 million, which would mean a further drop of 300 k over the next two weeks (?). Something between 5 and 5.2 milionlooks more plausible., and this assumes that melt hasn't already stalled out!"

Sep 2 - "are the conditions in 2010 similar to 2005? Isn'it possible that we already passed the SIE minimum on August 31 when considering the weather forecast for the next 10 days?"

Relax, Phil. Even the professional scientists can't be sure of identifying the extent minimum the moment it occurs. The 10th may have been it or the weather may turn sharply in favour of compaction and bring it lower again despite increasing sea ice area, as has happened in the past.

Phil263

Hi Jon

OK, I tend to see things a little bit on the grey ( lets say cautious!) side. To be fair I have also posted this on Sept 9:
"That ice is less likely to be affected by an early freeze up and therefore we might see further melt before the minimum is reached. DMI's 30% SIE is still pointing sharply downwards, and Unibremen's SIE curve is showing no uptick at this stage, so we should still expect further losses tomorrow." ....
which turned out to be an incorrect prediction. BTW.

I am definitely an amateur, and the only basis for my "predictions" are previous trends and a look at the maps.
I may be close this time however. If you look at the Unibremen map for today and for Sept 5th , it ssems to me that a) the ice everywhere is getting more compact (i.e less susceptible to melt) b) extent is gaining in Fram Strait and North of Svalbard We are also getting more advanced in the season and temperatures in the arctic are seemingly dropping very fast (see Christoffer's post above).
Yet 2010 has been full of surprises, i really did not think that we would see a drop of 350k at the begining of September, so who knows?

Artful Dodger

FrankD | September 12, 2010 at 15:08
"...comparing extents from the sources that cutoff at 30% with those that cut of at 15% should give similar results"

Hi Frank. I think we were getting close to a discover a few days ago during the discussion on why CAPIE ratios based on either CT or IJIS Area have differently shaped curves.

During melt season, it must be that any Sea Ice that is destined to melt must cross two finish lines: 30% concentration, and then 15%. So it follows that if a LOT of melt is imminent, the proportion of total Extent that is in the 15-30% concentration range MUST go up.

So if we had Area data at the 30% concentration cutoff, we could predict melt. I would LOVE to see this graph on my daily rounds!

Comments, idea, suggestions?

Artful Dodger

Bah, meant to say "if we had Area data at the 15% concentration cutoff" (CT uses 30%, of course...)

Verify your Comment

Previewing your Comment

This is only a preview. Your comment has not yet been posted.

Working...
Your comment could not be posted. Error type:
Your comment has been posted. Post another comment

The letters and numbers you entered did not match the image. Please try again.

As a final step before posting your comment, enter the letters and numbers you see in the image below. This prevents automated programs from posting comments.

Having trouble reading this image? View an alternate.

Working...

Post a comment