Sea ice extent update 19: promising, but premature

I'm regularly writing updates on the current sea ice extent (SIE) as reported by IJIS (a joint effort of the International Arctic Research Center and the Japan Aerospace Exploration Agency) and compare it to the sea ice extents in the period 2006-2009. The IJIS graph is favoured by almost everyone, probably because it looks so nice compared to other graphs (like the one by Arctic ROOS, the University of Bremen and the Danish Meteorological Institute). All the years have a nice colour of their own which makes it easy to eyeball the differences between trends. Most of the betting on minimum SIE is based on the IJIS data. NSIDC has a nice explanation of what sea ice extent is in their FAQ.

July 25th 2010

2010 has still not decided to stay on the canvas and let the referee declare a knock-out. After the abrupt ending of a small century break series a few days ago, things seem to be picking up yet again. But the Beaufort Gyre still hasn't reversed to a clockwise motion and pressure areas still haven't firmly installed themselves for the Arctic Dipole Anomaly to be either strongly positive or negative. Except for low pressure areas continuing to dominate the Arctic - reflected by a positive Arctic Oscillation - things are overall still quite erratic.

The big drop after the first two century breaks of the month was followed by a slightly higher 54,532 square km reported for the 22nd. The 23rd saw the second upward revision of the season and ended at a respectable 84,375 square km. Today it was a bit better still, though this morning's reported century break didn't survive the revision: 93,906 square km.

2010 is still holding onto that second spot. If it manages to produce a few good melting days it will have succeeded in fending off 2009's attack, after which it is 2008's turn with its series of century breaks in the first week of August.

The current difference between 2010 and the other years is as follows:

• 2006: -217K (70,025)

• 2007: +523K (98,609)

• 2008: -391K (81,069)

• 2009: -49K (92,127)

The average daily melt for the month of July is between brackets. 2010's average daily melt for July is currently 61,751 square km per day.

And here's the IJIS graph:

The erratic behaviour is still apparent in the Cryosphere Today Sea Ice Area graph. It has been going up and down again in the past few days, oscillating around the 1.3 million square km anomaly mark. It currently stands at 1.292 million square km, with the large drops in the East Siberian Sea and Laptev Sea having halted for the moment. The Arctic Basin is still hovering on the same level it has for the past few weeks.

In the comment section of the last SIE update there has been some discussion of the Swiss cheese quality of several parts of the central ice pack's interior that have become increasingly visible on the MODIS satellite images. These areas seem to become a bit smaller now with the tentative shifts in atmospheric circulation and winds blowing more towards the Pole causing the ice to be pushed together and against Greenland and the Canadian Archipelago:

It seems this will continue for a few days, as can be seen in the PIPS ice displacement forecast for tomorrow. The arrows are still not pointing in a clockwise direction, but they are big and that should have an effect:

TIPS - Other interesting blog posts and news articles concerning the Arctic and its ice:

I've added a few last images to the ice-floe-as-big-as-Corsica animation. And there's a freshly updated NWP animation.

Dirk Notz, head of the research group “Sea ice in the Earth System” at the Max-Planck-Institute for Meteorology in Hamburg, has published this article on Real Climate.

Patrick Lockerby has posted the 4th Update of his Arctic Ice July analysis, discussing the state of Petermann and HumboldtGlacier.

More glacier stuff from Mauri Pelto on his GlacierChange blog concerning annual balance measurements on Alaska's Lemon Creek Glacier.

There's a great read on the Edmonton Journal website, tying Robert McClure's HMS Investigator to the disappearance of muskusoxen on Banks Island.

Here's a NASA feature called 'Cutting into Arctic Sea Ice' on the ICESCAPE oceanographic mission which is studying the physics, chemistry and biology of the ocean and sea ice within a changing Arctic.

Comments

[Mike]

Hi Neven, I'd just like to say that you're doing a good job here with lots of relevant material.

[Neven]

Thanks a lot, Mike!

[Neven]

Another miserable melt: 56,875 km2. 2010 is now in 3rd place.

[Artful Dodger]

I wouldn't worry, Nevin. The High pressure system now between the New Siberian and Wrangel Islands spread the pack substantially between the 24 and the 25. Tomorrow, it'll all melt out.

[Phil263]

2010 looks more and more like 2006 or even 2003. Similar weather pattern? If the melting pattern for the rest of July and August follows the same track, we would be looking at a minimum around 5.6 - 5.8 million towards the top of the range of predictions made in June..
http://www.arcus.org/search/seaiceoutlook/2010/june

[Artful Dodger]

Look at what the wind from the passing Low did to the Sea Ice in the Kara and Beaufort Seas. Down to 50% concentration to the East and a big open spot to the West.

Additionally, Pacific heat influx is now beginning. 7C water has moved North of the Bering Strait, and is entering the Kara Sea.

Remember, warm Pacific water flowing at 1 SV through the Bering Strait carries as much heat as an entire season of Sunshine on the Kara Sea (enough heat to melt 1 million km^2 of 1 m thick ice). It's beginning to look a lot like 2007.

[Gili1]

Hi Dodger,

If you look at the past 5 years, melt season weather has played a major role (cloud cover, winds). The Gulf stream is also clearly important, but I'm not aware of important Pacific flows. Will appreciate pointers to articles about this.

-Gili

[Artful Dodger]

@Gili1 | Hi Gil. Did your open the 3 links I placed in my post? The last link is a full pdf paper titled "The 2007 Bering Strait Oceanic Heat Flux and anomalous Arctic Sea-ice Retreat" published in Geophysical Research Letters in Jan 2010. You can also see the abstract at the author's site, here.

[Jim Dowling]

@Artful Dodger:
according to the paper, the Bering strait only pumps 1/3 of the heat of the atlantic through the fram strait.
Have you seen something I missed? Are heat flows suddenly increasing over the Bering strait?
Cloud cover looks like it will continue for the Arctic Basin for the next 10 days, so I fail to see how this looks like 2007. What am I missing?

[a_yeeles]

@Artful Dodger:
re warm water entering arctic through Bering Strait/2007, the ocean temperature anomaly graph from NOAA/NWS etc suggests nothing out of the ordinary this year:

http://polar.ncep.noaa.gov/sst/ophi/color_anomaly_NPS_ophi0.png

What is interesting are the values along the NWP which presumably help to explain the observed melt.

[Artful Dodger]

Hi Jim. Are you suggesting that 'only' an extra 4 terajoules of heat won't have an effect. Pacfic heat influx is what made 2007 extraordinary. Not winds, nor ice advection.

[Nathan]

Something strange may have occurred around the NOAA webcam 2. If you look at the photo
http://www.arctic.noaa.gov/npole/2010/images/noaa2-2010-0725-190401.jpg
and compare with the next one
http://www.arctic.noaa.gov/npole/2010/images/noaa2-2010-0726-005531.jpg

It looks like the puddle of water sloshed up and over on the left side. Was the area of ice the camera sits on collided into from behind and to the left?

And that's a wide piece of open water in the background!

[Artful Dodger]

@a_yeeles | The heat flux is just beginning. Instead of looking at yesterday's anomoly, try creating a 14 animation of the SST and watch the heat building Northward through the Bering Strait. This is a temporal pattern, not a static one.

[Artful Dodger]

Dirk Notz, head of the research group “Sea ice in the Earth System” at the Max-Planck-Institute for Meteorology in Hamburg, has published this article today on Real Climate.

He says, in part: "Currently, the ice covers an area that is slightly larger than the extent in late July of the record year 2007. However, this does not really allow for any reliable projections regarding the future evolution of Arctic sea ice in the weeks to come.

The reason for this is mostly that sea ice in the Arctic has become very thin. Hence, in contrast to the much thicker ice of past decades, the ice now reacts very quickly and very sensitively to the weather patterns that are predominant during a certain summer. This currently limits the predictability of sea-ice extent significantly."

[Lord Soth]

With regard to the North Pole Webcam, it appears some new snow has fallen, and the melt pond had a light covering of ice, and the new snow is covering the ice.

[Lord Soth]

Thanks for the interesting article by Dirk Notz.

I have really noticed how the day to day weather is having a dramatic impact on sea ice extent. Instead of depending on climate issues, we are looking at the right weather conditions for three to four months of the year, to make the knockout punch on the arctic ice cap.

2008 has proven to us that we can have significant ice loss in August. 2009 has also proven that unfavorable weather can turn a contender into a dud for sea ice loss, based on extent.

The important thing to remember is that volume is going down year after year, and 2009 despite its lackluster extent, did beat 2007 and 2008 for lowest volume. And it appears that 2010 will score a knockout punch in regard to minimun volume.

I would love to get some detailed info on cryosat-2, such as how long would it take to get full coverage of the arctic (even north of 80). Are we talking daily, weekly, monthly or seasonal post processing for sea ice volume graphs ?

[Phil263]

All of us would prefer a " healthy" arctic ice and the trend that has been happening over the last 25-30 years is really a concern. However, sea ice extent is what the media report. A minimum well above 2009 after a maximum not seen since 2003 will be news that will be exploited by those in the media who want deny that Climate Change is happening ( and there's plenty of them around, ask Rupert!) . I can hear it now after "the winter of Al Gore's discontent", we had the "summer when the sea ice "refused" to melt". Lord Monkton and his ilks will be crowing, may be Steven Goddard might even "come out"... In the meantime, CO2 will keep being pumped into our atmosphere!

[Nathan]

Yeah, you're probably right Lord Soth :)

We don't see much snow in Australia, so I didn't recognise it.

[Nick Barnes]

At the north pole, it is definitely a substantial snow-fall. Look at the big yellow buoy, or the instrument box (?) in the centre-left.

[Lord Soth]

Neven. Nice plug for your blog over at realclimate. You should receive even more blog input now.

[Kevin McKinney]

With all humility, Nick, I'm not so sure about the snowfall idea. You would think that you'd see some accumulation on top of the buoy--though admittedly it could have been blown clear. The buoy looks to me as though it has settled deeper into the ice, while I don't see much change with the instrument box at all.

Meanwhile, the melt pond in front of the cam looks to me as if it has drained--and there's even a feature at what appears to be the low point which could be an eroded "drain."

Hard to be sure--I could be completely wrong. But that's how it looks to me, tentatively. (And I do know snow!)

[Nathan]

Kevin

It looks weird because there are larger mounds (of snow?) around the melt ponds, it looks to me as if there's been some snow and perhaps enough wind to blow the snow into those mounds. Or there has been a small collision, which has sloshed the pools and the water has pushed the new snow (?) into mounds... The slooshing (if I can use a technical term) looks like it was on a line from bottom left to top right of the photo.

Does anyone else have any ideas on the mounds on either side of the long thin melt pool and on the left side of the big pool in the foreground?

[Nathan]

Also the buoy in the background moves a long way... It's moved in the past but not that much in 5 hours...

[Anu]

The sea ice area is still well below 2009:
http://www.ijis.iarc.uaf.edu/seaice/extent/AMSRE_Sea_Ice_Area.png
Ice floes drifting around at the margins causing extent to vary is not that important in late July.

The 'quick look' at sea ice extent for July 26, given by Universitat Bremen, shows a sharp dip below 2009:
http://iup.physik.uni-bremen.de:8084/amsr/ice_ext_n.png
This should show up tonight in IARC-JAXA data.

I think my prediction of the 2010 minimum being below 2009 (using IARC-JAXA sea ice extent data) is still very probable:
http://nsidc.org/images/arcticseaicenews/20091005_Figure3.png
The weather is hard to predict every summer, yet the end result of this sequence of annual weather creates a clear trendline.
There's a small probability 2010 overshoots 2009 for the summer minimum, but I'm going with the larger probability - 2010 less than 2009.

[Lord Soth]

The revision came out and now the ice lost for the 25th is 69,531 sq. km. Thats a 13K revision towards greater melt. So 2010 still officially holds second place for another day.

I say there is a good chance for a century loss tonight, with such a large revision.

[Neven]

Artful Dodger, thanks for linking to that article on RealClimate, I've added it to the TIPS section of this SIE update. And thanks goes also to Mauri Pelto for mentioning the blog. I'm linking to his latest glacier article in TIPS as well. I've already got some visitors coming via RealClimate. Better start writing something interesting again, so as not to disappoint them. ;-)

Dirk Notz writes in his RealClimate article:

Hence, in contrast to the much thicker ice of past decades, the ice now reacts very quickly and very sensitively to the weather patterns that are predominant during a certain summer.

This is exactly what we have been witnessing and discussing the past 2 weeks or so. The melt is extremely erratic. One day winds blow towards the (Siberian) coast, all the slush spreads out and we get a miniscule change in extent. The next day we get a shortlived high pressure area over the Beaufort Sea, the slosh moves back in, and we have a century break. In that respect Lord Soth hits the nail on the head:

I have really noticed how the day to day weather is having a dramatic impact on sea ice extent.

As Phil263 notes:

A minimum well above 2009 after a maximum not seen since 2003 will be news that will be exploited by those in the media who want deny that Climate Change is happening

This is indeed a problem. If ice really is as thin as they say it is it would be better if this shows through in extent figures. If the minimum ends above the 5 million square km mark, there will be no end to the spin. Combined with the developing La Niña pseudo-skeptics will have enough material to promote their beloved inaction for at least a year and a half.

With regards to sea surface temperatures: does anyone know if there's an archive for these NCEP images? I've been saving the images for a few days now for an animation, but archives make things go so much faster. I haven't been able to find. Maybe I should ask Robert Grumbine on his blog.

And finally, Lord Soth:

I say there is a good chance for a century loss tonight, with such a large revision.

I agree. I was expecting one today, but everything looks poised for a century break reported tomorrow.

[Jon Torrance]

Neven,

"If the minimum ends above the 5 million square km mark, there will be no end to the spin. Combined with the developing La Niña pseudo-skeptics will have enough material to promote their beloved inaction for at least a year and a half."

Assuming we're talking about IARC-JAXA extent figures, I don't think just above 5M sq kms for the minimum (i.e. less than the minimum for 2009) would represent a clear pseudo-sceptic propaganda victory since it will be a bit hard to spin that as continued recovery. Of course, there'll be no end of spin no matter what happens. I'm sure anything above the 2008 minimum will have them trotting out the no death spiral line.

Personally, I'll be okay with about 5M sq kms - I'll win my Intrade bets and Steven Goddard's prediction will be about 0.5M sq kms too high.

Phil263,

"2010 looks more and more like 2006 or even 2003. Similar weather pattern? If the melting pattern for the rest of July and August follows the same track, we would be looking at a minimum around 5.6 - 5.8 million towards the top of the range of predictions made in June.."

Ummm... if the IARC-JAXA extent decreases as much from yesterday to the minimum as it did in 2006 or 2003, the minimum will be 5.54 or 5.27 million respectively, not 5.6 - 5.8 million. Maybe that's too simpleminded a way of looking at it but you'll have to explain your view of what a similar track to 2003 or 2006 would look like further if you want me to understand your point.

[Neven]

Personally, I'll be okay with about 5M sq kms - I'll win my Intrade bets and Steven Goddard's prediction will be about 0.5M sq kms too high.

Same for me, except for the 20 EUR I'll happily send to William Stoat Connolley. Unfortunately I didn't understand the Intrade system.

In my view the shape of the ice pack is increasingly resembling that of 2007. With all of that grey ice in the Beaufort, East Siberian and Laptev Seas disappearing and just that arm of ice reaching for Novaya Zemlya.

Talking about spin: in his latest SIN he says "Looks like a Northwest Passage traverse is quite possible (by helicopter.)" There's still quite a bit of ice in there, but when it's pushed aside like it was for the past week you could actually get a ship through there already. No ice bridges anywhere.

I think it's a safe bet that the NWP direct route will be open this year. The only question is whether that long arm of ice reaching for Novaya Zemlya will reach all the way to the Siberian coast, leaving the Northern Sea Route open or closed.

[Anu]

@Neven | July 26, 2010 at 23:12
The only question is whether that long arm of ice reaching for Novaya Zemlya will reach all the way to the Siberian coast, leaving the Northern Sea Route open or closed

I think you mean the ice reaching towards Severnaya Zemlya and beyond to the peninsula ?
http://www.geographicguide.net/pictures/arctic/arctic-map.gif

I think that will melt through, opening the Northern Sea Route in August.

[Jon Torrance]

Neven,

At the risk of you competing with me for a limited supply of counterparties (once my most recent check clears with Intrade's bank), the Intrade systems doesn't seem to be intuitive for most but it isn't all that complicated and the property that the trading price can be interpreted as the market's opinion as to the probability in percent of the event in question happening is nice. Taking their current sea ice bet (2010 JAXA minimum extent > 2009) as the example, buying or selling at 50.0 is an even money bet. If I sold the contract at 25.0, I'd be giving my counterparty 3:1 odds in his favour, since I'd lose $7.50 per unit traded if he won the bet while he'd lose $2.50 if I won the bet (all ignoring the small Intrade commission taken out of the winnings). If I accepted the current high bid to buy at 10.0, I'd be giving my counterparty 9:1 odds in his favour (my $9 per unit traded risked against his $1).

[Neven]

Yeah, of course I meant Severnaya. My Russian isn't what it was. Thanks, Anu.

For people who didn't know it, I made an animation of that area, which I will update as soon as we see what is happening under those clouds.

[Neven]

Jon, I really appreciate your efforts, but I'm still not getting it. Maybe next year I'll ask you to be my bookie. ;-)

[Artful Dodger]

@LS: "I would love to get some detailed info on cryosat-2"

Hi, LS. Objects in Low Earth Orbit are described by NORAD Two-Line Element sets. Cryosat-2's International Designator is 2010-013B. Here are some relevant elements:

Inclination: 91.9198°
Revolutions per Day: 13.93092259

So CryoSat-2 passes within 2 degrees of the poles, and orbits every 103 minutes or so.

Data will be released to ESA partners. It will be up to these individual institutions to set the frequency of public updates.

[Phil263]

Jon
.....but you'll have to explain your view of what a similar track to 2003 or 2006 would look like further if you want me to understand your point.

The 5.6- 5.8 million figure is just based on a statistical inference from the pattern of decrease in extent until the end of the melt season as it happened in 2006. Of course, I have no idea of what the weather will be like in August, but July so far has seen an even lower average melt than 2006 (around 62 k vs 70k).So from a "statistical point of view" it makes sense to infer that conditions in 2010 may be similar to 2006, but I agree with you they may not be. ...

[Mark Shapiro]

While watching MODIS terra and aqua satellite mosaics for the ice melt, I've also watched the gorgeous cloud formations and huge forest fires. There has been a slew of fires in eastern and western Siberia (lower right and upper right portions of the MODIS mosaics).

Today, however, there is a huge fire just below a large pear-shaped cloud about 100 km SE of Archangelsk (I think). It's near the lower right corner of the mosaic. It looks like a large white tear-drop with a fat, grey tail coming out the bottom to the left.

Is there something besides weather and forest fire going on there?

[Mark Shapiro]

BTW, I think both the NWP and the northern passage (Siberian side) will be visibly open next week, and stay open for two months.

Of course, my definition of "open" may differ from yours, and from a freighter captain's.

[Artful Dodger]

Neven, archives of the NOAA SST's are kept here, but they are in GRIB format, not the derivative graphical product.

I wouldn't worry too much about recreating them. The Pacific heat influx is just beginning, so if you're archiving the graphs now, you'll soon have a useful animation.

[Neven]

Thanks, AD!

[Artful Dodger]

De nada.

[Steve Bloom]

Looking at the PIPS 2.0 thickness projections for 7/26 in 2007 and this year, the ice in the immediate vicinity of Severnaya Zemlya doesn't look all that different. OTOH, in 2007 ice was clearly being pushed into that part of the Arctic, and absent that effect we may see the the NSR open up all the way. Overall, the condition of the main pack this year remains much less organized and distinctly thinner than in 2007.

[Neven]

Again no century break reported today. I'm even more surprised than yesterday.

PIPS' ice displacement forecast is still showing big arrows, but pointing in the wrong direction. The ice seems to be getting spread out more and more.

[Jon Torrance]

I find it calming, at times like this, to look at http://www.ijis.iarc.uaf.edu/seaice/extent/AMSRE_Sea_Ice_Area.png and see that area has been dropping at a steady rate since early or mid-July, depending how much steadiness one insists on.

[Neven]

Thanks, Jon, that was just what I needed. ;-)

Total extent number was revised downwards again today. The ice must be so spread out that a whole team of Japanese pixel-counters are working in shifts to check the satellite data.

[Lord Soth]

The correction for today, has also resulted in an increased melt. 2010 lead over 2009 now stands for another day. Usually when the correction leads to addtional melt, its an indication of a potential century break. I have however been proven wrong two days in a row, and im not making a prediction for tonight.

Actually if you remove the blip between late june and early july, the downward slope in the AMSRE sea ice area graph has been quite consistent since the start of May. If things don't change, we would match the 2007 sea ice area numbers by the third week of August.

[FrankD]

A couple of thoughts/questions occurred to me while reading this and the latest RC thread.

I've been comparing MODIS pics for Day 207 for 2009 and 2010, and the contrast is marked. Not (just) the area but the condition of ice where there is >15% coverage. The pix http://rapidfire.sci.gsfc.nasa.gov/subsets/?subset=Arctic_r05c03.2009207.terra and http://rapidfire.sci.gsfc.nasa.gov/subsets/?subset=Arctic_r05c03.2010207.terra are probably the clearest examples, but its true of many areas. 2009 seems to have had either good ice or open water, and where it's >15%, it's mostly >90%. 2010, with a similar area, is completely pants - a lot of it is just brash, waiting to melt. In the Chukchi/Beaufort Sea pic above, perhaps 1/3 of the "extent" is good ice. Without a better qualitative aspect, I'm beginning to wonder how useful "extent" really is as a measure?

So what are we to make of volume? Well, of course there's a couple of months to go yet, but PIOMAS, on the daily volume graph, bottoms out at an average of 13,500 km^3 (text says 14,400 km^3. Their current volume anomaly is -10,500 km^3. It is climbing at the moment, but we would seem to be heading for between 3,000 and 5,000 km^3 by September. The lowest volume figure reported by PIOMAS is 5,800 km^3 (Sep 09), so a lowest volume record would seem to be a pretty solid bet?

A lot has been said about the weather (esp winds) of 2007 compared to this year, and the volume of ice exported through the Fram Strait. Fine - favourable winds export ice and reduce extent. But surely there are now large areas that are so thin that they will simply melt in situ, in contrast to years prior to say 2005? That's what I'm seeing aroung the NWP, anyway. Ice is sloshing back and forth like your seventh margarita, but its not really being exported anywhere. It is however disappearing and ever greater areas are opening up - it must be melting in place. It seems to me that in situ melting will continue to become more important while export will become less important in coming years. Bottom melting has always been a big factor for volume, but a lot of area is now thin enough that its melting from the bottom all the way to the top.

Now, I notice on CT that in 2009 some of their region graphs didn't bottom out until mid-late October (eg Laptev and Kara Seas). So it seems to me that even if the predictions of 5 million km^2+ are right (personally I think they're too high, but I don't have cash on it), that points to very thin averages. Allowing for a few thicker bits there would be huge swathes averaging less than half a metre thick. If this is in the form of the current brash, it will probably stay mobile into October, and if wind and currents favour export at that time, we may see a very late minimum...

Idle musings...

< /ramble>

[Neven]

I have however been proven wrong two days in a row, and im not making a prediction for tonight.

Yes, I'm shutting up too. For now. :-p

[Neven]

Great rambling, Frank.

It is however disappearing and ever greater areas are opening up - it must be melting in place. It seems to me that in situ melting will continue to become more important while export will become less important in coming years.

This is also the feeling I get when watching all the graphs and images. But it's the first time I'm watching the ice so closely, so I can't compare.

These are my main questions:
Will we get heavily decreasing extent through compaction when weather conditions reverse again in a significant way?
Will we see heavy in situ melting if weather conditions remain as they have this whole month?

[Jon Torrance]

My speculative question of the moment: what would it take for the large relatively concentrated mass of ice north of Svalbard between the NE tip of Greenland and Franz Josef Land to separate from the rest of the ice pack and drift southwards? Or is that totally implausible?

[Neven]

Jon, it's not as implausible as my first question on this blog if it was possible that the ice pack could move away from the Greenland coast with a lead opening that would make Greenland circumnavigable, but still implausible I think.

There would have to be some mechanism that pushes the relatively concentrated mass of ice north of Svalbard away from the ice pack, but at the same time keep the rest of the very mobile ice from following. That concentrated ice mass still isn't one huge floe that can be moved about uniformly.

[Jon Torrance]

Looking at the PIPS ice displacement forecast for tomorrow, I'd argue the mechanism to push that piece away from the icepack is anticipated to exist. And I'm picturing the relatively thick and solid ice north of Greenland (this is the expected last redoubt of the thick multi-year ice, right?) being restrained from following by being too strong to fragment and Greenland being too much in the way for it to move as a solid mass. So there's a certain simple mechanical plausibility to it but I have no idea how thin (unknown anyway with current data) and how unconcentrated the ice around the edges of the piece would have to get before it's something that could actually happen due to realistic winds and currents.

Mostly, I can't help thinking it would be frightfully dramatic for a mass of ice about the size of, eyeballing it, Finland to separate from the main pack and drift south.

[Neven]

Jon, I think it is possible that a combination of winds and currents might divide the ice pack with a part of it flowing towards and through Fram Strait, but I just don't think it will look like one big piece of healthy ice with the extra thick ice staying behind. How concentrated is the piece in the centre?

It would need a very positive Dipole Anomaly though, as I believe this is an important factor in the Transpolar Drift Stream. Or perhaps, you actually wouldn't want the TDS to kick in action because it will push much of that ice against Greenland, right? Just have those cyclone arrows point towards the East.

A good MODIS picture of how the ice looks under those clouds would be very instructive. It's been quite a while now since that happened.

[Charles Wilson]

An Experiment -- when JAXA extent is updated from Preliminary to Final, if it goes down, the next day is Down (recently) ... It was revised up a little -- by 1,719 -- a tenth the largest revision, true, but let's see if it goes up ....

[Kevin McKinney]

Modest melt once again--60k, close enough.

Nonetheless, FrankD's comment sums up what I keep seeing this season--concentrations are much lower, over much wider areas, despite the extent being only slightly lower than 2009.

[Artful Dodger]

Kevin, I have today's decrease in Extent at -70,469
(we do not know the melt).

[Phil263]

Just a question. Where do you get the jaxa sea ice area graph?

[Artful Dodger]

Hi Phil. AMSRE_Sea_Ice_Area. I don't think they make tabular data available. I'd like to track this statistic: Area/Extent. It would give us a sense of how 'spread out' the pack is at any point in time, an important predictor for melt.

[Phil263]

Thanks AD. Do you think that the computation of sea ice area is as rigorous as sea ice extent though? I understand that sea ice extent has its flaws, however its measurement is basically cut and dry. I am not so sure about sea ice area since the data about ice concentration seems to be a bit fuzzy. Am I wrong here?

[Charles Wilson]

JAXA Data (extent = edge of 15% ice) is at http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv
PS: it seems my Experiment is easy - - A WUWT poster got IJIS to explain they average 2 days together. So Tomorrow is already a part of Today, so to speak.

Other good Number lists: Monthly Satellite Temp Anomaly (with breakout for the Arctic OCEAN) http://vortex.nsstc.uah.edu/data/msu/t2lt/uahncdc.lt
Australian ENSO/ El Nino site http://www.bom.gov.au/climate/enso/ OR:
http://www.cpc.noaa.gov/products/analysis_monitoring/ocean/index/heat_content_index.txt

[Peter Ellis]

No, yesterday is a part of today. Otherwise you're implying that they make up tomorrow's data before it comes in! :-)

[Artful Dodger]

Phil: The area vs extent issue was discussed quite thoroughly by Lord Soth and Anu on July 13-14 in comments to Animation 7: Jakobshavn Isbræ.

[Phil263]

Thanks AD

[toby]

I am taken with the idea of a massive ice flow the size (Holland? Denmark? Ireland? Finland?) breaking off and floating away.

In the last war, the British dreamt up massive artificial ice floes as cheap aircraft carriers. They never followed up. Suppose it came true ....

[Kevin McKinney]

AD wrote: "Kevin, I have today's decrease in Extent at -70,469. . ."

And: "(we do not know the melt)."

[Kevin McKinney]

WRT to #1 above: Me, too, in the morning light!

And #2: I was speaking loosely there, of course. I wasn't the first, but maybe you're right that we should be more exact in this.

[Anu]

I bet the area decrease is just about all melt:
http://www.ijis.iarc.uaf.edu/seaice/extent/AMSRE_Sea_Ice_Area.png
Maybe a bit of ridging and stacking going on.

And I bet that sea ice area decline holds steady until at least August 3rd...

p.s. Arctic Amplification - more open water, more solar energy absorbed, warmer water, more ice melted (even if it's only thinning the ice - this preconditions the ice for the next summer melt).
Weather can have a big effect on a summer melt season, but global warming is pushing things in one direction, summer after summer after summer (slightly warmer water, slightly warmer air):
http://tamino.files.wordpress.com/2010/07/pred2010.jpg

[Anu]

Quick note on 2010 compared to 2009:
There are 4 more melt days in July, 31 in August, say about 14 in September (2007 had a minimum on September 24, but was basically flatlined since 9/15/2007, and the random noise gave a minimum on 9/24)

So, 49 more days of summer melt.
For 7,167,031 km² (todays extent) to reach 5, 249, 844 km² (2009 minimum) requires only an average daily extent decline of 39,127 km²

That puts the recent 70,000 km²/day declines in perspective.
Although, the decline rate slows down a lot in late August, for some years.

[Gas Glo]

>"For 7,167,031 km² (todays extent) to reach 5, 249, 844 km² (2009 minimum) requires only an average daily extent decline of 39,127 km²

That puts the recent 70,000 km²/day declines in perspective.
Although, the decline rate slows down a lot in late August, for some years."

An alternative (more realistic?) perspective would be to suggest linear decline in the rate of decline in extent so this requites 78,254 km² declining slowly to nil.

This shows that the current average
(whether you take last day or 70045 over last 7 days, 68851 over 14 days or 61555 over 21 days)

is *not* up to the required rate for the extent to decline as low as in 2009.

[Anu]

An alternative (more realistic?) perspective would be to suggest linear decline in the rate of decline in extent so this requites 78,254 km² declining slowly to nil.

That's not what happened in 2007 or 2008:
http://www.ijis.iarc.uaf.edu/seaice/extent/AMSRE_Sea_Ice_Extent_L.png

The trends make it probable, but not definite, that 2010 will decline lower than 2009:
http://tamino.files.wordpress.com/2010/07/pred2010.jpg

Luckily, we don't have long to wait to see how this all plays out.
ಠ_ಠ

[Gas Glo]

Anu,

The years 2007 and 2008 show sharp reductions between late July and the minimum compared to other years. 2009 is nearer the low end of extent reduction after late July so the point you make does have validity (even if using 2007 and 2008 seems to stretch the point a little).

However, if you are therefore expecting more extent reduction than this, that only makes the comparison of 70k rate of reduction linearly declining even more surprising low compared to the required rate of 78k to get down to 2009 level and an expected rate of more than 78k.

Doesn't this makes the comparison of 70k rate to your 39k average required rate less acceptable?

You aded qualification to your average rate that "the decline rate slows down a lot in late August". So I should probably add qualifications to my linearly declining required rate including that we haven't necessarily reached the peak rate of decline yet and if we haven't then the linearly declining required rate model is rather suspect.

Another way of putting 70k rate in perspective is to look at my updated JAXA Extent rate of change graphs:

http://farm5.static.flickr.com/4124/4837879315_30d3d89e45_b.jpg

Existing rate is lowest, second lowest, or third lowest of last eight years depending whether you look at 21 day, 14 day or 7 day average.

[Artful Dodger]

@ Anu | July 28, 2010 at 15:00

Thanks for posting this curve. About time somebody noticed that the decrease in extent is NOT linear. Is this your work? Did you use the JAXA data? Would you post your best fit equation please? Did you have correlation coefficients for linear best fit vs. your curve?

[Jon Torrance]

AD,

The curve is Tamino's work - go check out his blog An Open Mind for more details.

[Phil263]

GasGlo,

Excellent graphs.. You can also see that for all years, the rate of reduction in SIE decreases sharply after mid August. The graphs also indicate that the decrease in the SIE reduction rate is not linear.They show a more or less stable reduction rate for the next 20 days and then a exponentially decreasing reduction rate after that date.
.
I totally agree with you. Based on statistical patterns for the last 7 years, the probability that the minimumin SIE for 2010 will be lower than for 2009 is less than 50%.

[FrankD]

Tamino brings his statistical weapons to bear, and finds some interesting wrinkles in the data: http://tamino.wordpress.com/2010/07/28/sea-ice-curiosity/

Further to my musings/rambling/griping earlier, I'm interested in Peter Ellis' comment on that thread: Wouldn’t it make more sense to look at (area / extent)? That should be a reasonable approximation of the average ice concentration across the pack: i.e. how healthy the ice is.

Might have to do some number crunching to see if that shows anything...

[Anu]

AD:
Jon is correct, but I saw Tamino's graph because somebody here on Neven's blog pointed to a RealClimate article on the Arctic sea ice:
http://www.realclimate.org/index.php/archives/2010/07/an-icy-retreat/comment-page-2/#comments

Comment #94 is from Tamino, who links to his graph. I haven't seen his blog for awhile, but I did once read a funny article there about how WattsUpWithThat didn't realize that GISS had a different baseline than UAH in their global temperature plots, which is why their "temperature anomalies' were bigger - WUWT attributed the difference to some evil Dr. Hansen manipulations to get more grant funding or carbon trading profits or something...

http://tamino.wordpress.com/
Open Mind

http://www.drroyspencer.com/
Dr. Roy Spencer, entrepreneurial skeptic and shoddy scientist (20 years of "skeptic lectures" based on "UAH data shows the troposphere is not warming like the GCM's say it should, so global warming is bunk", until RSS found their data reduction errors, and now it's "whoops, ok, its warming, but not as much as RSS says - really, trust me")

[Anu]

Gas Glo:
The graphs are interesting - are the 7 day averages that day and 6 trailing days, or the midpoint of 3 trailing and 3 leading days ?

Yes, you can argue 'if 2010 is more like 2008, it can beat 2009, but if it's more like 2003, it will not decrease fast enough by mid September..."

But I still think the two most important keys to understanding the minimum extent probability for this summer are these:
http://tamino.files.wordpress.com/2010/07/pred2010.jpg
2010 is different from 2003 because the Arctic is warming each decade.

http://www.ijis.iarc.uaf.edu/seaice/extent/AMSRE_Sea_Ice_Area.png
There is more open water surrounding the Arctic Basin than at any time on record, except for 2007. I think every day that this open water is not covered by clouds puts more heat into the system, causing more bottom melt in the nearby sea ice:
http://rapidfire.sci.gsfc.nasa.gov/subsets/?subset=Arctic_r02c03.2010209.aqua
Those swirly patterns on the fringes of the sea ice are the last vestiges of melted ice floes.
Look at that black, black Arctic water soaking up the sun...

Of course, even during the Ice Age some summers in Europe were unusually warm - weather happens superimposed onto climate. That's why I talk of probabilities.

But unlike FrankD, I give the probability of having a lower minimum than 2009 more like 80%, not less than 50%.

[Artful Dodger]

Kevin: Don't PANIC when you see today's -46,250 net decrease in SIE. There was substantial spreading of the pack west of the New Siberian Islands. However, it's still over 15% concentration, so still counted in Extent based measures. Comfort yourself with the steep drop in Sea Ice Area, and the sure knowledge it'll melt soon. I hope you feel better!

[Phil263]

Anu,

"I think every day that this open water is not covered by clouds puts more heat into the system, causing more bottom melt in the nearby sea ice:"

Yes, but it looks like clouds are covering the water right now and they have been for a while. Look at today's SIE reduction - 46,250 sqkm. I agree with you this does not change the fundamentals but the weather is definitely superimposed and that can blur the overall message.

[Kevin McKinney]

What, me panic?

;-)

[Anu]

@Phil263 | July 29, 2010 at 05:25
You don't need the entire Arctic cloud free every day - here we see large sections of the Beaufort Sea, Canadian Archipelago, and Greenland Sea with clear skies:
http://rapidfire.sci.gsfc.nasa.gov/subsets/?mosaic=Arctic.2010209.aqua.4km

Here the Canadian Archipelago is cloud free, and big swaths of the Beaufort and Chuckchi Seas, plus over the ice in the Kara Sea:
http://rapidfire.sci.gsfc.nasa.gov/subsets/?mosaic=Arctic.2010206.aqua.4km

Here Baffin Bay is clear, and you can see extensive melt:
http://rapidfire.sci.gsfc.nasa.gov/subsets/?mosaic=Arctic.2010199.aqua.4km

When you can see black water in these satellite images, it is heating.
I saw some papers on the 2007 summer melt - the skies were unusually clear that summer, but that meant 70% cloud covered days, rather than the usual 80%, something like that - the normal summer is very cloudy anyway.

Artful Dodger is correct - the sea ice area continues to drop relentlessly, even as the ice floes are bunched up or spread out by the wind and currents. Only in September, when the final minimum is near, will the extent compaction or expansion be locked in.