The 2014 Melting Season

[Frivolousz21]

The GFS has backed off towards the euro.  I was thinking about this compared to 2013 as May comes to a close but then I decided to see how it compares to the 2007-12 period and the results were very surprising.

Anyways.  I was expecting most years to have big high pressures sprawling over the arctic but that is not the case at all.

I posted the images over at americanwx.com.  But every year except 2012 was "favorable" for the ice.

I won't repost all of the images here but here is 2007-2011.  And the last image is 2014.

[F.Tnioli]

...

... And now, if 2014 will be similar or worse than 2012 (in terms of ice cover - which apparently is quite likely, reading this topic), then i'd say this graph will have a new high this year. And this means alot to sea level rise at very least, feeds back to Arctic ice cover dynamics (Greenland melt waters are very low salinity, i believe?), etc etc. ...

Quoting oneself is a bad manner, but this here case, i do it to underline the date i said that piece above - because, just 3 days after i said that, this got published: http://phys.org/news/2014-05-greenland-greater-contributor-sea.html . Surprise, surprise... I've been saying for years already that they'll find more and more new mechanics which accelerate and increase Greenland melt, and it seems, now they've warmed up alright. I guess they'll get to the idea of large meltwater pulses - and their consequences, - in some 3...5 years. Under "they" here, i mean those nice smooth-paper journals, of course...

[werther]

Thanks, F.Tnioli, your post seems better suited for the Greenland-thread though.
On the developments in the Laptev Sea, Jim Hunt continued it on the 'area and extent calculations'-thread, posting a fresh day 143 acqua image centered on the Bhuor Kaya Gulf in front of the Lena delta.
Melt is progressing fast out there.

[Nightvid Cole]

Is it just me or has very "smoky" air from Siberian fires now been pulled into the central Arctic?

http://rapidfire.sci.gsfc.nasa.gov/imagery/subsets/?mosaic=Arctic.2014143.terra.4km

Note the yellowish tinge in the central Arctic compared to either the Alaskan or Svalbard sides of the Arctic.

[Jim Hunt]

Sorry Werther. Here it is again. The Aqua 7/2/1 view of the Lena Delta area this morning:

[jdallen]

Sorry Werther. Here it is again. The Aqua 7/2/1 view of the Lena Delta area this morning:

The ice there should be getting attacked vigorously both top and bottom, I expect. Outflow should be pooling just under the ice, and possibly across the top of it in some places.  We can then add top melt and ponding from unusually warm air and insolation.  It may be losing 10CM/ day in some places.

[wili]

NV wrote: "Is it just me or has very "smoky" air from Siberian fires now been pulled into the central Arctic?"
That seems to be what RS is saying here: http://robertscribbler.wordpress.com/2014/05/18/arctic-heatwaves-rise-to-threaten-sea-ice-as-lake-baikal-wildfires-reignite/

Here's a consequence of Arctic Sea Ice Melt that I, at least, hadn't thought about before:


Trillions of Plastic Pieces May Be Trapped in Arctic Ice

Humans produced nearly 300 million tons of plastic in 2012, but where does it end up? A new study has found plastic debris in a surprising location: trapped in Arctic sea ice. As the ice melts, it could release a flood of floating plastic onto the world.

Scientists already knew that microplastics—polymer beads, fibers, or fragments less than 5 millimeters long—can wind up in the ocean, near coastlines, or in swirling eddies such as the Great Pacific Garbage Patch. But Rachel Obbard, a materials scientist at Dartmouth College, was shocked to find that currents had carried the stuff to the Arctic.

http://news.sciencemag.org/earth/2014/05/trillions-plastic-pieces-may-be-trapped-arctic-ice

Thanks to COBob at RS's site for the link.

[Jim Hunt]

NPEO webcam 1 seems to have come a cropper:

[AbruptSLR]

The attached SSTA image for May 23 2014 from cci-reanalyze (see link) shows the warm Pacific Ocean water continuing to pass north through the Bering Strait:

http://cci-reanalyzer.org/DailySummary/index_ds.php#

[Jim Hunt]

Another clear view of the Laptev Sea and the Lena Delta this morning:

[ChrisReynolds]

Posted under the correct thread this time...

Comparing 24 May 2012



and 24 May 2014.



I can't help but get excited at the prospects for the early part of this melt season due to conditions in the Siberian sector.

[Frivolousz21]

Area and extent have slowed in dropping.

However The Hudson, Baffin/St. Lawrence sea way are above normal.

[Neven]

Area and extent have slowed in dropping.

However The Hudson, Baffin/St. Lawrence sea way are above normal.

Indeed, and if melt ponds are so crucial this month, I would say that the melting season hasn't been broken (like 2013), but not exactly made either, ie maybe a new record has already been made impossible. But it also depends on where all the melt ponding takes place. With that reverse dipole and the high over the Siberian part of the Arctic (instead of the American side), it looks like most of the melt ponding will occur there in this final phase of the month of May.

The Bard says: Nothing is either good or bad, but thinking makes it so.

[idunno]

I agree with both ChrisR and friv21, above; my own immortal words, copied over rom the blog...

"I think that the fast melt off Alaska is a delayed symptom of the heatwave in Alaska in January and February, when temps were actually above freezing for some time.

Looking through the previous years plots, it seems to me that in Chukchi, ESS and Laptev, the closest parrallel to this year is neitherr 2012, nor 2013, but 2007.

2014 differs from 2007, in that the Atlantic side looks much weaker than then, similar to all recent years.

FWIW, I think that we are going to see a spectacular melt all along Eurasia, with the Northern Sea Route open for a record long period; and the possibility of clear blue water at the Pole.

OTOH, it has been abnormally cold for recent years everywhere in the quarter from the Pole to due South and due West. I think that the Hudson Bay will clear late; the melt in the Baffin Bay will be slack; and the NW Passage through the Canadian Archipelago may not clear at all.

And if none of the above happens, I shan't be terribly amazed, and I'll blame the weather. Sea ice is, I find, a very disobedient substance, which very rarely sees fit to do as it's told."

... not quite Shakespeare, I fear. But immortal in the sense that the internet never forgets.

P.S. ...though you can occasionally edit things, so I can add the following equally irrelevent...

The Moving Finger writes; and, having writ,
 Moves on: nor all thy Piety nor Wit,
Shall lure it back to cancel half a Line,
 Nor all thy Tears wash out a Word of it.

The Rubaiyat of Omar Khayamm

[jbatteen]

I suspect that the Hudson Bay ice is a consequence of the frigid North American winter and spring, and that it's not of much consequence to the greater Arctic sea ice scenario.  Once it melts out, area/extent numbers will be back down quickly.

[jdallen]

Posted under the correct thread this time...

Comparing 24 May 2012

<snippage>

and 24 May 2014.

<more snippage>

I can't help but get excited at the prospects for the early part of this melt season due to conditions in the Siberian sector.

While the Navy Hycom models have weaknesses when used to quantify the state of the ice, watching them for the last two years I sense they are useful in predicting the quality and behavior of the ice.  For example, the models did a good job of predicting the shattered state of the pack around the pole as it evolved last August.

Used as a reference against itself, year over year, I similarly sense again at a qualitative level, it does a reasonable job of identifying relative weakening and strengthening of the pack, and as you note, the news is not hopeful.

While the Kara appears to have thickened marginally by 20-30CM, and similarly in Hudson Bay, it implies most of the pack is 50 to 100CM thinner.  Across the greater portion of the Laptev and ESS, it shows as much as 150-200CM thinner.  Even if this is to generous by a significant margin, it indicates a huge vulnerability in the pack, which is about to be smashed by conditions which could chew away as much as 5CM of ice a day.  Even if conditions become much more favorable as they did last year, I do not think this is something this portion of the pack can recover from.  The disappearance of the ice will have an amplifying effect on melt elsewhere, regardless of weather, and will be a dynamic which I don't believe has been seen this early.

Considering idunno's comment, in *spite* of comparatively colder conditions over the CAA and NE North America, in fact over the historic record, they were not that extraordinary.  I think we would find the apparent cold stands out only by way of recent context, where we have become more accustomed to milder winter season weather over all.

Even if we see slack melt conditions in Baffin Bay, there is less ice over all there to melt out; further the ice in Hudson bay is not notably thicker that I can tell as compared to past years.  I don't see ice there lingering much longer than usual, if at all.  If conditions agreeable to ice retention persist in those areas, I think it bodes ill for the ice in others; for them to stay cold, cold conditions will need to be displaced from elsewhere.

I do concur though, that even if there are cool conditions, there will likely be spectacular melt all across the Eurasian margins of the Arctic, from end to end.  Too little ice is going to be exposed to too much heat for too long, too early.  It is yet to be seen, but that may undermine relatively cooler conditions elsewhere in the pack, with or without melt ponds. 

If we see the kind of massive melt out we are expecting across the ESS/Laptev/Kara region, combined with record melt out that has *already* transpired in the Bering and Chukchi,  I think we are still in serious danger of breaking the 2012 record.  There remains very little good news. Far too much heat is entering the region from multiple sources.

[ChrisReynolds]

JD Allen,

Quite right, I take HYCOM in a qualitative sense as a reasonable guide to conditions. The main problem I see with it is overestimation of MYI thickness, in recent years with a lot of FYI it's been tying to around 2m thick for FYI in March/April fairly well, I presume we can take the scaling between 0 and 2m as reasoably accurate.

Idunno,

Using April PIOMAS gridded data 2012 is similar to 2014, in the Siberian sector this year April was more like 2011. I have noted before that it may not be a good idea to get too fixated on small differences in thickness. But seeing HYCOM now I suspect that by May PIOMAS too will show 2014 being one of the poorest ice states in the Siberian sector.

[AmbiValent]

I think if one would know one area of the Arctic would melt out early and another late, but could select which does which, I think an early peripheral and late central melt would be better for long-term ice survival. After all, the periphery will always melt out completely, and then the water will just warm up - and later radiate that warmth into space. On the other hand, in the central Arctic, ice will either melt or not, which influences the future ice volume. And if water warms up there, it would be cooled again when it meets and melts neighboring ice (and that also impacts future ice).

Or is my thinking wrong on this?

[idunno]

Thanks Chris,

That changes my mind about the similarity of 2007/2014. On CT both years have anomalous open water in the Laptev/ESS area, and a rapid retreat in Chukchi, whereas 2012 actually had high early season ice area in both Chukchi and Bering. But I agree that those 2007/2014 thickness plots are different.

Sticking with my view that Hudson will melt later - it will melt, but later - and skew the early to mid season comparison with data from recent years, when very early Hudson melt has previously caused an early exaggerated negative anomaly.

[ChrisReynolds]

idunno,

I think the differences in open water formation at the start of the season that you outline are due to weather. Which alone shows that weather is critical. As I've found in trying to make predictions based on April thickness/volume, April thickness/volume defines the long term downward trend in extent/area, but there is significant 'noise' due to weather. This can make two years look the same in terms of extent/area even when the thickness fields are very different.

This year I suspect that May thickness off the Siberian coast may be enough to cause a rapid retreat through June, weather allowing.

Ambivalent,

I've recently posted a blog post outlining why due to the Central Arctic region tending to a ~2m thick April state we may see a reduction in winter volume loss rates and an ice free state in the late 2020s, rather than around 2020.
http://dosbat.blogspot.co.uk/2014/04/what-is-future-of-arctic-sea-ice-part-2.html
Whether or not this will be enough to delay transition to a seasonally sea ice free state I can't say. I do think we're seeing the start of a reduction of loss rates due to more of the pack becoming first year ice (~2m thick by April), but I cannot say how thinning in the Central Arctic will continue to respond. In the 1990s Central Arctic summer melt was around 4k km^3, by 2012 this had increase by 50% to around 6k km^3, for summer melt to be almost as much as the volume implied by an equilibrium thickness of around 2m requires an increase in summer melt to over 9k km^3, approximately a further 50% increase.

Given the high albedo of ice much of the sunlight hitting it is reflected so doesn't go into making more melt. I wouldn't rule out that a further 50% increase in absorbed sunlight could lead to a further 50% increase in melt volume (or thinning) during the melt season. That is all that is needed in the Central Arctic to melt most of the ice by September.

[Lord M Vader]

The ice in ESS and Laptev are quickly weakening now. Soon we'll have a huge opening there. What it'll mean for the rest of the melt melt season I've no idea about but it'll ceertainly be very interesting to see.. This will allow the SSTs to spike quickly then... Not to forget that the opening will have about 4 months to warm up.
 
f one look closely at the ESS at Hamburg pics one can see that the coast line in ESS is on track to connect with the Beaufort Sea in a while. Will the ice in ESS and Laptev be gone by May 31? THAT would be exciting!! I think that JAXAs numbers for SIE is too high given what's going on in ESS, Laptev and between Svalbard and Franz Josefs land..

[Frivolousz21]

Numbers are still high because the Baffin, Hudson, Greenland sea, and st lawrence sea way area are all above normal.

[AmbiValent]

Chris,
you might be right that - looking only at the Arctic itself - the ice loss could slow down once warmed water there radiates out into space when the ice is concentrated in the most central section of Greenland, Archipelago and North Pole. Whether other, positive, feedbacks would be stronger, I don't know. However, to keep the ice alive it still needs decent winters where the Arctic is really cold, not those that allow massive exchange between polar and temperate areas. And I fear we're getting more of those.

[Neven]

And how about increased ice mobility?

[crandles]

And how about increased ice mobility?

My WAGs'

I think this is likely to continue the fall in average ice age and therefore there is less time for mechanical processes to thicken up MYI.


I doubt it will lead to an increase in ice volume export. If the volume export continues to stay approx constant then it becomes a bigger proportion of the ice that is left and maybe same volume export means more thinner ice so there is more open water formation in Central Arctic as a result of export changes.

However I think volume export is more likely to decline than increase. For next two or three years I suspect volume export is most likely to stay roughly the same.

[Frivolousz21]

the models are trending back worse again.

[idunno]

Sorry, Chrisses, (Reynolds here and Biscan on the Blog), I don't completely buy the 'it's all down to weather' line, having got my inocculation during the 2012 melt, when parties off tried to explain the record minimum entirely by reference to one weather event.

The initial starting conditions of the health of the ice exert huge influencing limits on what weather can do, imho. Otherwise, a strong cyclone in August 1980, say, could have caused a massive crash in ice area and extent back then.

Yes, it's all down to how the pieces move in the course of the game; but also the layout of the board has changed. Due to climate change.

In this year, it seems to me that white (ice) has a strong defensive position in the SouthWestern quadrant (Lincoln, Atchipelago, Baffin, Hudson); black (water or heat) is confronted with pathetically weak defences everywhere else.

[ChrisReynolds]

Ambivalent,

How other feedbacks will play in - frankly I don't know. But if one takes the winter loss of volume in the Central region and extrapolates then it implies ice free winters by 2040 (i.e. in the 2030s), I can't see how that can happen, based on what is happening now (a pre-requisite for trend extrapolation). One issue that is relevant is that much of the central Arctic volume loss in the past has been from thinning of MYI, with FYI volume throughout the whole Arctic Ocean actually increasing. Something has to give. But what will happen to summer losses - I don't know.

Neven, Ambivalent,

Increased sea ice mobility hasn't, as far as I know, led to increased volume export through Fram, because the ice exported is thinner volume export has remained variable but without a large increase. That leaves the Beaufort Gyre, as Crandles (I think) pointed out a couple of years ago ice entering the central arctic can go one of two ways - Fram or Beaufort Gyre. If Fram export hasn't changed much then that leaves Beaufort Gyre. There is a paper that found about 1/4 to 1/3 (I forget) of MYI export (by extent) from the central Arctic is into the Gyre, where in summer (post 2007) most/all of it melts out - a major source of loss of increased MYI loss is through this path. In the past ice would have been cycled and aged before ending up back in the transpolar drift and up against the CAA and northern Greenland. I should add that the paper I just mentioned concludes by observing that some of the loss of MYI extent is probably due to mechanical compaction off the CAA.

So, and sorry for being long-winded getting around to this, what is needed is not just cold winters. During the Holocene Climatic Optimum (~6k year BP) the summer Arctic was virtually ice free. This was driven by insolation changes which caused an increase in summer insolation, with a commensurate decrease in winter insolation. Winter insolation decrease wouldn't affect the pack - it's dark in winter, but it would be expected to reduce warmth coming in from surrounding continents with reduced winter insolation. Even if I'm wrong on that, the point remains that with just summer increase in insolation (unlike AGW) the Arctic Ocean was seasonally (virtually) ice free.

I suspect that key to this is the Beaufort Gyre, without a reduction of melt out during the summer from Beaufort to the ESS there is no hope of a recovery of sea ice. Because for a recovery MYI has to start to survive and age until it fills the 60 to 80% of the Arctic basin that it did in the past. Even with colder winters it's hard to see how ice is going to thicken up enough to start surviving what has become regular melt outs in the peripheral seas around the central Arctic. I've been trying to find a better graph but can't locate it - the one I wanted was a plot of thermodynamic thickening as a function of temperature. The only one I can find is a plot of heat flux through ice (vertical axis) as a function of thickness (horizontal axis) at a series of temperature differences between the atmosphere at the surface of the ice and the ocean at the base of the sea ice - image attachment.

The point in showing this is that it shows that thickness is a far more dominant factor than temperature over heat flux through the ice, and heat flux through the ice is the factor that drives accretion of new ice at the ice/ocean interface. So I don't think that cold winters are what's needed because alone they cannot regenerate the sea ice, they have a relatively small impact on ice thickness at the end of the winter. What's needed to regenerate the pack are cold years - and short of a massive eruption (e.g. flood bassalt) or a run of massive plinean volcanic eruptions, or other similar force majeur - that just isn't going to happen.

[ChrisReynolds]

The initial starting conditions of the health of the ice exert huge influencing limits on what weather can do, imho. Otherwise, a strong cyclone in August 1980, say, could have caused a massive crash in ice area and extent back then.

No need to apologise -  I agree with you.

It's just that thinner ice and the resultant increase in open water formation efficiency with resultant increased sensitivity to weather forcing is such a part of my 'mental furniture' I don't feel the need to say it every time.

Maslanik et al 2007 "A younger, thinner Arctic ice cover: Increased potential for rapid, extensive sea-ice loss"
http://sites.fas.harvard.edu/~scia30/2008/SOURCEBOOK/2007_Maslanik_GRL.pdf

The replacement of older, thicker ice by younger, thinner ice over much of the Arctic Ocean, combined with cumulative effects of warming, unusual atmospheric circulation patterns, and resulting intensification of the ice-albedo feedback, contributed to this large and abrupt loss of ice. Taken together, these changes suggest that the Arctic Ocean is approaching a point where a return to pre-1990s ice conditions becomes increasingly difficult and where large, abrupt changes in summer ice cover as in 2007 may become the norm.

2007 - seems an age ago....

PS - there are two or three volume loss events as big as 2007 and 2010 in the 1980s and 1990s - but the ice was thicker then so they didn't affect area/extent.

[iceman]

One unusual feature about this early melt season is the open water at high latitudes: in the Chukchi Sea, northern Baffin Bay, Laptev and Barents:

While the area is relatively small, it could have a disproportionate effect on the Arctic heat budget owing to a counterintuitive (to me, anyway) feature of insolation: the polar regions receive the most energy near the solstice.  Though Tamino didn't provide a key for this graph (from his 2012 blog post on Sea Ice Insolation), the trace I infer to be one month before the NH summer solstice - second from top on right side - indicates that locations north of about 70* will have a high and increasing insolation for the next several weeks.  (The graph is probably for top of atmosphere, hence some unverified assumptions about insolation at the surface.)


What does that mean for the pockets of open water?  In the Chukchi Sea it's centered on about 67*N, near the latitudinal dip in the graph - though insolation here will increase quite a bit from now to solstice.  Northern Baffin Bay is higher on the curve at 75*N.  However, ocean currents don't seem to promote much effect on ice to the north from excess heat in this location.  Heat advection into the core ice region is more likely from Laptev and Barents, where the open areas are close to 80*N: just about maximum insolation.

Question remains of how much sunlight will actually hit that open water in coming weeks, and on that subject my crystal ball is cloudy.
Net: one more indicator on top of others that point toward an early melt on the Siberian side, and an extremely lopsided distribution of ice at the September minimum.

[Frivolousz21]

[TerryM]

ChrisR
I think that if you read any of Dr. English's studies of driftwood on Ellesmere Island you will be convinced that during the Holocene Optimum summer ice was still extant and rafting Siberian logs to Canada. That would not be possible under what most would consider to be virtually ice free conditions.
While Fram Strait's ice volume may be relatively constant the opening of Nares and in 2012 the opening of the CAA to MYI heading south might make a difference to retained volume.
Keeping my eye on Nares ever since PII2012 headed for warmer climes.


Terry

[ChrisReynolds]

Terry,

I've not read all of England's work, I'm basing that claim on the following paper....

Jakobsson et al 2010. "New insights on Arctic Quaternary climate variability from palaeo-records and numerical modelling" http://www.sciencedirect.com/science/article/pii/S0277379110003185



And from the abstract.

The combined sea ice data suggest that the seasonal Arctic sea ice cover was strongly reduced during most of the early Holocene and there appear to have been periods of ice free summers in the central Arctic Ocean

England is referenced as showing that there was no land fast ice prior to 5500 year before present.

England, J.H., Lakeman, T.R., Lemmen, D.S., Bednarski, J.M., Stewart, T.G., Evans, D.J.A.,
2008. A millennial-scale record of Arctic Ocean sea ice variability and the demise
of the Ellesmere Island ice shelves. Geophysical Research Letters 35.

[jdallen]

And how about increased ice mobility?

My WAGs'

I think this is likely to continue the fall in average ice age and therefore there is less time for mechanical processes to thicken up MYI.


I doubt it will lead to an increase in ice volume export. If the volume export continues to stay approx constant then it becomes a bigger proportion of the ice that is left and maybe same volume export means more thinner ice so there is more open water formation in Central Arctic as a result of export changes.

However I think volume export is more likely to decline than increase. For next two or three years I suspect volume export is most likely to stay roughly the same.

I think we may need to define what constitutes export.  While we focus primarily on the Fram and Nares, this year I've watched a steady march of ice into the Barents Sea through the Svalbard/Franz Joseph/Nova Zemlya gaps.  Pre-2000 or so, ice sent through there would simply stack up in the Barents which previously was much colder.  Now, with SST's as much as 1 or 2C even in the depth of winter, that ice gets chewed up promptly.  Put this way, the ice we see currently in the Barent's is not the same ice we were looking at a week ago.  That ice got pushed south and was replaced.  Good evidence for this can be seen in the drift "shadow" - gaps in the pack on the leeward side of Franz Josef land and Svalbard.

I think this implies a dramatic increase in export of of the CAB, possibly as much as is being exported through the Fram.  I think also, we can draw a direct relation between the ice exported here, and the gaps we currently see opening across the Laptev, ESS, possibly even the Chukchi.  Exported MYI made space for much thinner lead ice to form.  Most recently, that export has permitted leads to open, which have *not* been recovered by fresh ice.  The ice in those regions hasn't melted in place; it has shifted.  The new export into the Barents I think is the proximate cause of the open water.

[ChrisReynolds]

Iceman,

I agree with you, but am not sure how 'lopsided' the distribution of ice will be in the context of recent years. Ice in the peripheral Siberian seas (ESS to Barents) has typically melted out in most post 2007 years. If it does melt out early it holds out the possibility of significant inroads into the Central Arctic.

Looking at the last ten days....

There is already an anomaly warming over ESS/Laptev where the ice is receding.

Lower left quadrant of that image.

This warming is seen to be low level, but is part of a warm/cold split higher in the atmosphere.

Despite the activity higher up I think it is safe to assume that the low level warming above average in ESS/Laptev is due to the reduced sea ice and ice albedo feedback in the area.

The SLP plot shows that a dipole formed between low pressure over the Atlantic Sector and high pressure over the Pacific sector has drawn the ice away from Siberian coast.


Looking at 500mb geopotential height (GPH) (heights of the atmosphere where pressure is 500mb) the pattern reflects SLP.


For the same period in 2007, which preconditioned for the crash of that year, GPH is similar.


However 500mbGPH for the whole of May so far is not so similar to the same period in 2007. So the recent similarity may not mean anything in terms of the season to come. However the current set up seems to support aggressive ice albedo feedback in the Siberian sector provided the weather continues to move ice away from the coast and low pressure remains over the Atlantic sector, with high pressure over the Pacific.

[F.Tnioli]

User-friendly color ice thickness maps compare for this May, posted yesterday in Arctic News:



By the way, i've read in a few places that general trend for % of cloud cover for increasing temperatures - is reduction (i.e., in general, the higher air temperature is, the less clouds there would be, on average). I wonder, does it apply to Arctic summer conditions (much open water and the Sun being abovethe horizon 24/7): all the extra evaporation sounds like a possible source for more moisture in the air, possibly forming more (and not less) clouds. How often it would overcome increasing moisture "capacity" of warmer air, i don't know. Any thoughts on this, i'd read with much interest.

[ChrisReynolds]

Dr Slaters' latest 50 day lead time prediction was run on 24 May 2014, 50 days from that is 13 July 2014. From the graph here:
http://cires.colorado.edu/~aslater/SEAICE/
The prediction for 13 July 2014 is about 8.2M km^2, however this is in terms of NASA Team algorithm which is not the same as NSIDC Extent. NASA Team algorithm data is available here:
ftp://sidads.colorado.edu/pub/DATASETS/nsidc0192_seaice_trends_climo/total-ice-area-extent/nasateam/
Owning page is
http://nsidc.org/data/smmr_ssmi_ancillary/area_extent.html#smmr_ssmi

The problem is that (as often happens for me with NSIDC data) both IE9 and Chrome are failing to download the data - so I can't say how the prediction by Dr Slater compares with previous years. If I have more luck later I'll post - but maybe someone else will have more luck getting that data to download.  The dataset I'm trying to get is this one:
ftp://sidads.colorado.edu/pub/DATASETS/nsidc0192_seaice_trends_climo/total-ice-area-extent/nasateam/gsfc.nasateam.daily.extent.1978-2012.n

[Frivolousz21]

Huge huge model changes on both the GFS and Euro.

WoW.

[crandles]

If I have more luck later I'll post - but maybe someone else will have more luck getting that data to download.  The dataset I'm trying to get is this one:
ftp://sidads.colorado.edu/pub/DATASETS/nsidc0192_seaice_trends_climo/total-ice-area-extent/nasateam/gsfc.nasateam.daily.extent.1978-2012.n

no problem at moment - if you are struggling try
https://drive.google.com/file/d/0BzpGniYbi4anQnhQT2pSWHlGOEE/edit?usp=sharing

[Neven]

Huge huge model changes on both the GFS and Euro.

WoW.

If this comes about, it will be fireworks. But I thought the same last week, and then things flipped.

[iceman]

Despite the activity higher up I think it is safe to assume that the low level warming above average in ESS/Laptev is due to the reduced sea ice and ice albedo feedback in the area.

Good insights, Chris.  If ice albedo feedback is a primary cause, we could expect anomaly warming near the Siberian coast to become a recurring feature of the early melt season in coming years.

btw this early-season open water at high latitudes seems to be a corollary of the open water formation efficiency that you investigate on Dosbat.  The actual location of open water will be more random than the %OW trend itself, as influenced by less predictable factors such as
   ice mobility;
   changing export dynamics as described by jdallen upthread;
   atmospheric pressure/circulation patterns.
When the open water does appear at high latitudes, though, it's a positive feedback especially around the summer solstice.

[ChrisReynolds]

Thanks Crandles,

Using that data and keeping in mind that Dr Slaters prediction for 13/7/14 is 8.2M, the NASA Team Algorithm extents for 13 July from 2000 to 2012 are:

2000   9.93
2001   9.25
2002   9.61
2003   9.43
2004   9.80
2005   9.07
2006   8.67
2007   8.20
2008   9.12
2009   8.81
2010   8.37
2011   7.76
2012   7.95

So that puts the central point of Dr Slater's prediction between 2010 and 2011.

Actually comparing your copy of the NASA Team data to NSIDC Extent shows they're pretty close most of the time.

NSIDC Extent
2000   9.77
2001   9.25
2002   9.61
2003   9.43
2004   9.72
2005   9.08
2006   8.67
2007   8.20
2008   9.04
2009   8.81
2010   8.37
2011   7.76
2012   7.91
2013   8.28

And post 2010 - i.e. starting from 2011 - 2011 and 2012 show drops below previous years, which ties in with my obsession about the 2010 PIOMAS volume loss event - I don't recall spotting that before in NSIDC data. 2013 is obviously higher for reasons already established.

But Dr Slater's prediction for 13 July 2014 puts this year more in line with 2013 and the pre-2010 years, not ahead of 2013. Not sure if that sways my opinion of the prospects this year but it's something I'll be bearing in mind. My prediction based on April PIOMAS volume and my assessment of the state of the ice there still persuade me we'll see a large melt out with a lot of ope water off Siberia by mid July - more open water than Dr Slater's plot of estimated probability of ice. Between the two of us it's probably my assesment that's wrong, not his numbers.


Iceman,

I've not had the time to look for similar warmings in past years. I'm halfway through doing an analysis (per grid box) of the first date that concentration drops below 15% off the Siberian coast, I suspect that when I finally get round to finishing it I'll find earlier open water in that region in recent years. If I do (and my suspicions are more often wrong than right) examining low level warmings may be an interesting extension to the study.

[DaddyBFree]

Significant breakup of fast ice is occurring off the coast of Alert today. For anyone interested in seeing the difference between yesterday and today: https://earthdata.nasa.gov/labs/worldview/

[ChrisReynolds]

Neven, Friv,

ECMWF from WetterZentrale now gives ensemble middle (I suspect this is the average of the ensemble) and standard deviation.
http://www.wetterzentrale.de/topkarten/fsecmeur.html

i.e.

500hPa,SLP - Ensemblemittel - seems to be the ensemble average (for the lurkers see below re ensemble)

500hPa Ensmittel und Standardabw - seems to be the ensemble middle in contours and I think "Standardabw" is short for "Standardabweichung" which is German for 'standard deviation'.

This additional data isn't available for GFS, only for ECWMF.

The 'ensemble' is a set of model runs, they are started with identical (or near identical) initial conditions and run forward to see how much deviation occurs with time between the different runs. This then tells how chaotic or stable the atmosphere is (or how sensitive to small variations in initial conditions the atmosphere is) - I'm not sure exactly how GFS do it.

Anyway, if the standard deviation is small in the region around the area of interest then things will probably pan out as the model is predicting because all ensemble members are producing closely the same thing. As time moves on in the dates of prediction the scatter of the model always widens - you can see that already in the plots of standard deviation for the coming week. But if the standard deviation doesn't widen too much (compared say to the differnce of height between a trough and ridge making a dipole) then dipole like behaviour, if indicated to persist probably will.

Another useful trick is simply comparing GFS and ECWMF - seeing if different models say the same thing.

[Frivolousz21]

Both the GFS and Euro now go full on ridging by day 3 and then explode it as we go along.

Book it.  Without a cold core PV it's going to happen.

The cold core in summer will die without a cyclone to produce massive clouds.

[seattlerocks]

frivolous, probably somebody said it before, but I really like the passion you put in your forecasts.

[jdallen]

Huge huge model changes on both the GFS and Euro.

WoW.

If this comes about, it will be fireworks. But I thought the same last week, and then things flipped.

The Sumo match continues...

[jdallen]

Using that data and keeping in mind that Dr Slaters prediction for 13/7/14 is 8.2M, the NASA Team Algorithm extents for 13 July from 2000 to 2012 are:
<snippage>
So that puts the central point of Dr Slater's prediction between 2010 and 2011.

Actually comparing your copy of the NASA Team data to NSIDC Extent shows they're pretty close most of the time.
<more snippage>
But Dr Slater's prediction for 13 July 2014 puts this year more in line with 2013 and the pre-2010 years, not ahead of 2013. Not sure if that sways my opinion of the prospects this year but it's something I'll be bearing in mind.

I'll bring up my stock market metaphor again.  While weather is modestly more predictable than human behavior, a lot of energy is spent trying to forecast markets on information which while based on previously observed behavior, is founded more on correlation than cause.

With respect to Dr. Slater, I am suspicious of these projections.  The dynamics of the system have changed massively over the last decade, and as the previous two years demonstrated, are amazingly volatile.

The "buffer" - the volume of sea ice in the system - has diminished to the point that relatively modest changes can shift events quite radically.  The prediction is far more of a dice roll now.

My prediction based on April PIOMAS volume and my assessment of the state of the ice there still persuade me we'll see a large melt out with a lot of open water off Siberia by mid July - more open water than Dr Slater's plot of estimated probability of ice. Between the two of us it's probably my assesment that's wrong, not his numbers.

Dr. Slater's numbers demonstrate a great deal of thought, and by extension definitely are representative of previous behavior.

However, that's an awful lot of open water at high latitude Right Now, being exposed to weather highly conducive to capturing more heat.

I continue to think your arguments and assessment are more persuasive, and are more reliable in that they are based on current conditions, rather than extrapolations which are based on past behavior where ice states and weather have only gross similarity to those we see now.

I think the once we go past 2011, we exponentially decrease the utility of past ice conditions in predicting the future.

[Meirion]

I think 2014 is quite confusing - can see Northern Sea Route opening up earlier than ever yet temps north of 80 North after warm winter show signs of another cold summer like 2013 - and no melt-out at the Pole

[werther]

Fantastic break-up in the Lincoln Sea visible on MODIS today!