The 2015 melting season

[ChrisReynolds]

Wanderer,

I'm not expecting a record year, I just don't want a boring year like the last two. My prediction for September NSIDC extent is 5.15 million kmsq +/- 0.64 million kmsq. I am hoping for the low end.

I think you're at the high end, Chris.  Even an average seasonal loss - 9.5 million KM2 - would put you at your low end figure.  As the average losses in the last 4-5 years have been running over 10, and because of the running start we've had in may, I'm pessimistic about the future of the ice; it is not good.

I still put beating 2012 at about 1 chance in 3.  I think we're on track to meet or beat 2011, and probably 2007.  I put our missing one of those outcomes at less than 1/10.

My method will probably be biassed high by the high volume in the Central Arctic not being reflected in the periipheral seas. The same method using Beaufort/Chukchi/ESS/Laptev volume gives: 4.79 +/-0.73 M km^2. That's a range of 4.06 to 5.52M km^2. Note the substantially higher range on that.

The method I have used might be biassed high, but the hindcast is 80% successful from 1979 to 2014. Only 2007 and 2012 go below the hindcast for those years. I will be using the CT Area and June compactness to heuristically adjust at the end of June.

[Michael Hauber]

That is a fair size low that is being forecast on the Atlantic side, and for a few days at peak intensity it looks like being well positioned on the fringe of the ice to do a lot of damage.  The Laptev region is going to get blasted by both strong wind and lots of warmth from the south.  Kara and Barents should also see some serious impact.  At 7 days it weakens and moves closer to the pole, so might be more a slowing influence with more clouds and less wind.

[jai mitchell]

This http://earth.nullschool.net/#2015/06/05/1800Z/wind/isobaric/850hPa/equirectangular=-175.41,79.84,1193 makes me think that we will get a good head start on the CAB surface melt like what happened in 2012.  So if it pans out that way then we should start to see daily SIE deltas approaching 100K within this week.

[werther]

Melt ponding in the Beaufort.

GPS Coords are 76.5~ N -165
[/quote]

Those coordinates, if I read well, 76.5N 165W, point to a location 135 km into the Central Arctic Basin, North of Chukchi Sea.
As stated earlier, quite a remarkable meltponding event this early in the season.

[sedziobs]

Certainly a discernible shade of blue on the ice visible through the clear skies north of Wrangel.

[Yuha]

But no melt ponds.
The snow around 2015B was turning dark in May, but then it got a layer of new snow that still looks pretty white and reflective.

I thought the Beaufort section was headed for strong melt season after the May heat wave, but I'm not so sure any more. It might have been too little too early for northern Beaufort.  The Chukchi section looks more promising though.

Melt ponding in the Beaufort.

GPS Coords are 76.5~ N -165

I count that already as a Chukchi sector and one of the reasons for thinking that the melt is progressing better there. Another reason comes from looking at Worldview.

How about O-Buoy 10 / 2013F? It's showing substantial surface melt in Beaufort, even though the camera's resolution is awfully low...

Those could be early stages of melt pond formation but it's difficult say for sure. I think they haven't really changed in the last few days. If they turn into proper melt ponds within a week or so, then a complete melt of this area is a possibility. On the other hand, if the melt ponds don't appear until July as they did last year (watch the movie), there probably isn't enough time for a complete melt anymore.

[werther]

O-buoy 10 is almost 300 km out into the CAB, N of Beaufort Sea. It was not in the meltponding event as of 23 May. It is right in the April cracking-event zone, where big amalgated floes are still covered by snow.
At the colder side of the persistent high, there's not much indication of rapid meltpond formation out there.
GFS/Climate Reanalyzer/Wetteronline suggest main warmth on the shores of East Sib and Laptev Seas next 7 days. Expect ice conditions out there to worsen as incoming summer makes up on a 'climatologically normal' spring over there.
An interesting season until now. But still uncertain where it will go. As most of the CAB is still 'untouched', I suppose PIOMAS will keep reporting larger amounts of volume, as it has since last summer.

[jai mitchell]

I suppose PIOMAS will keep reporting larger amounts of volume, as it has since last summer.

I am guessing it will be around 22.85 +/-.10 10^3 km3

[jdallen]

Wanderer,

I'm not expecting a record year, I just don't want a boring year like the last two. My prediction for September NSIDC extent is 5.15 million kmsq +/- 0.64 million kmsq. I am hoping for the low end.

I think you're at the high end, Chris.  Even an average seasonal loss - 9.5 million KM2 - would put you at your low end figure.  As the average losses in the last 4-5 years have been running over 10, and because of the running start we've had in may, I'm pessimistic about the future of the ice; it is not good.

I still put beating 2012 at about 1 chance in 3.  I think we're on track to meet or beat 2011, and probably 2007.  I put our missing one of those outcomes at less than 1/10.

My method will probably be biassed high by the high volume in the Central Arctic not being reflected in the periipheral seas. The same method using Beaufort/Chukchi/ESS/Laptev volume gives: 4.79 +/-0.73 M km^2. That's a range of 4.06 to 5.52M km^2. Note the substantially higher range on that.

The method I have used might be biassed high, but the hindcast is 80% successful from 1979 to 2014. Only 2007 and 2012 go below the hindcast for those years. I will be using the CT Area and June compactness to heuristically adjust at the end of June.

Makes sense - those factors may suggest a smaller total annuL melt - 9m km2.

[Jim Hunt]

Perhaps unsurprisingly to regular readers, temperatures near the North Pole are now above "average" according to DMI:

[johnm33]

Wanderer, I think a record melt is within the bounds of possibility, mainly because I think the ice is in a different state. As I see it 15-25% of the ice is old fashioned 2-5m thick ice frozen from the bottom up 30% plus is rotten ice which started to freeze from the bottom up but was covered by snow and never completely shed it's salt and 30% plus is layer cake ice which I see as a crust of ice formed rapidly in the cracks, which have been constantly opening up, topped with snow drifting to fill the voids between the older slabs, then when new cracks open a layer of condensing mist forms a crust, before more snow covers it again, rinse repeat. So Piomas may be accurate about volume but actual mass is significantly down, apart from the obvious drop in the necessary energy to melt this there's also the increased susceptibility to wave action.
 The recent 'melt ponding' in Beaufort I saw [mainly thanks to Wipneus's home brew animations] as Pacific waters sucked in causing huge amounts of bottom melt, [which would not be possible with 'old fashioned ice'] lots of cracks opened, then evaporation and condensation onto the surface, which looked like melt but rapidly froze.
How confident am I, not very, but that's how I see it.

[DavidR]

Well, Jun 1st NSIDC extent is in at 10.545 M km^2,  250K km^2 below the previous record, and 2.523 M km^2 above the  July 1st record.

The smallest June melt in the past 5 years is 2.559M km^2, so 2015 is well positioned to be still at a record low at the end of the month.

[Buddy]

Perhaps unsurprisingly to regular readers, temperatures near the North Pole are now above "average" according to DMI:

THAT.....is definitely going to be an interesting one to watch over the next 2 - 3 months.

[Tor Bejnar]

Beware:  when this year's melt is ahead of previous years' melt, predicting the same amount of future melt may not always be supportable.  2015 melt is ahead partly because Hudson Bay melt is ahead.  Whether Hudson Bay melts out completely by July 1 or September 1 won't affect the annual minimum much (a little more heat on the south side of CAA and Baffin Bay - even with the water generally flowing southward), but will sure make the 'total Arctic' July 1 comparison with previous years look 'good'. 

[oren]

Beware:  when this year's melt is ahead of previous years' melt, predicting the same amount of future melt may not always be supportable.  2015 melt is ahead partly because Hudson Bay melt is ahead.  Whether Hudson Bay melts out completely by July 1 or September 1 won't affect the annual minimum much (a little more heat on the south side of CAA and Baffin Bay - even with the water generally flowing southward), but will sure make the 'total Arctic' July 1 comparison with previous years look 'good'.

Is there anywhere a chart similar to the NSIDC showing the ice extent for the arctic ocean only, compared to previous years and long-term decadal averages? I think this could give some insights as to how different this year is (or not) in the core regions.

[crandles]

Is there anywhere a chart similar to the NSIDC showing the ice extent for the arctic ocean only, compared to previous years and long-term decadal averages? I think this could give some insights as to how different this year is (or not) in the core regions.

Wipneus Home brew thread
http://forum.arctic-sea-ice.net/index.php/topic,382.0.html

I think Wipneus makes data available if you want to add regions and graph as you wish. It is extent according to high definition data but not a long record.

CT
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/recent365.anom.region.1.html
Area not extent. Also not sure the data is available to replot as you want.

MASIE

ftp://sidads.colorado.edu/DATASETS/NOAA/G02186/plots/r11_Central_Arctic_ts.png
Not a long record and not sure data is all available.

PIOMAS
Volume data (modelled with assimilation) not extent. Chris Reynolds has made some regional data available to save having to work with raw PIOMAS data.
http://dosbat.blogspot.co.uk/2013/12/regional-piomas-volume-data.html

Are there other regional analysis data?

[ChrisReynolds]

Ahhh, I need to update my PIOMAS data, the fact that nobody has complained shows what I suspected - nobody is using it!

I'll update the online stuff when I get May's data just in case anybody does want it.

I guess I had better also get Gice uploaded....

[epiphyte]

@chrisreynolds... I honestly was looking for it yesterday!

[Neven]

I'm back from my holiday and just finished reading up here. Thanks, everyone, for your comments and analysis. This way I don't miss anything. 

[seaicesailor]

First open water and melt ponds signs off East Siberian coast ( see bottom left of todays r06c04 MODIS tile)

[Nightvid Cole]

Ahhh, I need to update my PIOMAS data, the fact that nobody has complained shows what I suspected - nobody is using it!

I'll update the online stuff when I get May's data just in case anybody does want it.

I guess I had better also get Gice uploaded....

Or maybe one of them had been using it, but switched to Wipneus's maps when yours went too long without updating. 

[wili]

"Perhaps unsurprisingly to regular readers, temperatures near the North Pole are now above "average" according to DMI:"

Doesn't that put it in the range where salty first year ice would start to melt?

[jdallen]

First open water and melt ponds signs off East Siberian coast ( see bottom left of todays r06c04 MODIS tile)

Thank you for the image.  Those thinking melt in the ESS and Laptev is behind last year might consider just how thin the ice is because of earlier export and shifting g of ice south

[Jim Hunt]

Despite the slightly chillier weather off the coast of Alaska recently, buoy 2015A now reports 22 cm of surface melt plus a bit more off the bottom:

http://GreatWhiteCon.info/resources/ice-mass-balance-buoys/summer-2015-imbs/#2015A

[Sonia]

The Lena delta has clear sky today and images of the melt are beautiful.  Check your favorite imagery source and be sure to turn off annotations such as coastlines.

[Jim Hunt]

The Lena delta has clear sky today and images of the melt are beautiful.

A bit like this one you mean?

http://GreatWhiteCon.info/resources/arctic-sea-ice-images/summer-2015-images/#Laptev

Plain vanilla "true color"!

[Neven]

Anyone else surprised that CAPIE/compactness is relatively high? Here's my CT/IJIS graph, but Wipneus' graph that I just put on the Regional Graphs page shows the same:

[Jim Hunt]

Anyone else surprised that CAPIE/compactness is relatively high?

Having been following both closely recently, then not really. "IJIS/JAXA/Whatever you want to call it these days" extent  is at record lows, whereas CT area is not. Apples v oranges? QED?

[plinius]

wasn't there also some sensor change for the CT area? A bit difficult to find, since they do not seem to be too enthusiastic about providing clean and easily found documentation.

[Neven]

Anyone else surprised that CAPIE/compactness is relatively high?

Having been following both closely recently, then not really. "IJIS/JAXA/Whatever you want to call it these days" extent  is at record lows, whereas CT area is not. Apples v oranges? QED?

Naturally, compactness is high when area is high relative to extent. So maybe I should rephrase: Why is area not relatively as low as extent, relative to other years? I mean, that heat wave on the Pacific side of the Arctic has obviously caused some melt ponding (more than 2013 and 2014, I'd say) and the ice is quite fractured in places.

Is it simply because the Siberian side hasn't been playing along, and a lack of transport (like last year)? But still, I don't see why CAPIE/compactness should be so much higher than cold 2013 and 2014, given that this year seems to be quite a bit warmer.

If this means there is little melt ponding, we could be in for another slow year, masked right now by the Hudson decline.

[seaicesailor]

Anyone else surprised that CAPIE/compactness is relatively high? Here's my CT/IJIS graph, but Wipneus' graph that I just put on the Regional Graphs page shows the same:

CT map has been showing 100% over the whole central Arctic, and from Laptev to ESS. That is huge. It just has been very cold. Compare May 25 2012 map vs same day 2015 in CT site.
However yesterday's map showed first signs of melting over Central Arctic and Eurasian side.

Will area "catch down" extent, or extent "catch up" area? I believe there is more potential for the first. I expect extent record lows until July at least, since there are open fronts in Hudson, Kara, Beaufort, Chukchi, .... Laptev is opening up now, ESS maybe.

That is a big difference with '14 and '13. In 2014 I believe all were late, except that Laptev runaway. 2013 was all slow too, Beaufort practically didn't open up at all.

I'd be very surprised if a stall in extent decline happens during June, but who knows.

In central Arctic there is that huge thick cold mass of ice. July-August stall?

[plinius]

I do not think that ice temperature plays any significant role here, when it comes to more than a couple of days of melt. If I have my numbers right, 1K temperature difference on a 2m thick ice pack will balance out a little bit more than 1 cm of melt...
@Neven: Concerning CT area, we have no real information how much melt ponding biases this measure and in particular not how this bias has changed over the past years and in particular with their more recent data source changes. Do we also have any idea how the failure of their snow cover source affects their algorithm?

@Jim: Given the pictures of 2015A, I find the measurements real fun to look at, but would make the claim that the measurement is biased, since it is sitting deep in a meltpond.

[Neven]

I'd be very surprised if a stall in extent decline happens during June, but who knows.

In central Arctic there is that huge thick cold mass of ice. July-August stall?

Yes, that's what I'm getting at. But then again, I'm obsessed with melt ponds.

My feeling right now is that last year's volume rebound could be wiped out because of the early attack on all that MYI in the Beaufort/Chukchi, and Fram transport on the other side. But extent- and area-wise there is not enough melt ponding for this year to make it big.

But that's my feeling right now. Could change tomorrow, or after dinner tonight even. 

@Neven: Concerning CT area, we have no real information how much melt ponding biases this measure and in particular not how this bias has changed over the past years and in particular with their more recent data source changes. Do we also have any idea how the failure of their snow cover source affects their algorithm?

It's the bias we're after by calculating CAPIE/compactness, because it tells us something about the amount of melt ponds and how compacted/dispersed the ice pack is.

If it were only CT, the cause might be something with the algorithm or resolution or whatever, but like I said, Wipneus' graph shows the same (using more sophisticated and compatible data sets):

[Jim Hunt]

@Jim: Given the pictures of 2015A, I find the measurements real fun to look at, but would make the claim that the measurement is biased, since it is sitting deep in a meltpond.

You could legitimately argue that the reported surface melt is not necessarily typical of the area in general, and Wayne and I have been having a discussion about the accuracy of the top few thermistor readings when the sun is shining on them over on the blog. However, do you have any reason to doubt the temperature readings from the floe itself and the water under it?

[Nick_Naylor]

Robert Scribbler is expecting melt ponds to spike this week:

https://robertscribbler.wordpress.com/2015/06/02/arctic-heatwave-forecast-to-crush-northern-hemisphere-snow-cover-this-week/

[seaicesailor]

Based on today's NSIDC sea ice concentration data, I calculate the following CT changes:

Wed       10.072692
Thu -80.8  9.991925
Fri -161.5  9.830429


Thursday drop is inflated by the first-of-the-month effect. The big drop in Friday is genuine and helped by a -80k drop in the CAB.

From the sea ice extent and area thread, this is relevant to our discussion. It reflects the higher temperatures now. Just two days but a lot of potential for area to decline faster, I guess

[plinius]

@Neven: mea culpa, had not noticed that Wipneus has the same statistics from a fully independent source. I suspect though that some things are too entangled here: Broken ice and melt ponds do not have the same effect on ice melt, as with which they impact your measure. So you have a hidden third parameter that probably precludes predictions. And concernind broken ice: I am not really aware of a demonstration that broken ice is more vulnerable than a compact ice area with a large external rim. E.g. 2015B shows no significant bottom melt despite being adjacent to a very active and relatively large rift. I think it is far more important how much open ocean + melt ponds we have in total, which would just boil down to putting more weight in situation assessment on area than on extent, with the ratio just reflecting this, but not being an own predictor.

@Jim: Given that the ice itself absorbs some of the solar radiation (in particular when impurities are present) and that the surrounding patches are still covered in snow, I would have some mistrust against the internal temperature of the floe too (where we would probably be back to the short discussion on thermal conduction and temperature profiles). Cannot imagine that the water beneath is significantly biased due to lateral flows, so that should be safe.

[Yuha]

One contributor to the recent uptick in compactness is probably the melting of sparse ice in Chukchi Sea along the NW Alaskan coast causing a big drop in extent and a much smaller drop in area.

[cesium62]

It's the bias we're after by calculating CAPIE/compactness, because it tells us something about the amount of melt ponds and how compacted/dispersed the ice pack is.

How about:  Compactness is high because the wind has been blowing across the Hudson?

[plinius]

@Yuha, would add the on-ice winds dominating in the Barents, too. Everything went for compaction there.

@cesium: Think that's a smaller part of the problem, Hudson has only a tenth of the area. so an impact there would not be that huge?

[Laurent]

The grinding machine is approaching the Nares strait.

[sedziobs]

The Chukchi melt makes sense as a reason behind greater compaction, but I don't see it in Wipneus' regional graphs.  Extent and area there have dropped mostly in concert.

From those graphs, I would say the main reason for greater compaction is the lack of area decline in the ESS and Laptev.  Other years had seen area begin to drop by now with extent remaining high.  This year area has remained high up to now as well.

Forecasts for the next week over the Siberian side would change that.

[seaicesailor]

Further speculation, fun weather for the next few days. Hot in CAA Hudson and Laptev, over zero in much of the Arctic, and that storm ...

My 2c extent falls 300 k and area 700 k in one week, compactness starts to drop.

[wanderer]

Anyone else surprised that CAPIE/compactness is relatively high?

Not at all surprised. As of today we are even above 2014 CT area.
Maybe this will change in the next few days as Wipneus calculated some bigger drops!
Otherwise I expect that we have another slow (boring) melt year ahead...

[Nightvid Cole]

12z ECMWF shows the cyclone getting replaced by a high-pressure system at 192h and 216h. I know it's a bit far out (long-range), but just sayin'.

[epiphyte]

For what it's worth I don't think the larger area than last year signifies anything about the outcome of the season. I don't see any ice, anywhere that was ice-free on July 7th last year, that looks more than two weeks from melting out under current conditions. I'm not saying that it will, of course... just that I don't see why it shouldn't 

I'm still standing by my meta-prediction from last year, which was that the year the arctic becomes ice-free in September, anyone who judges the progress of the melt by the progression of area decline over the season will be very surprised, because it will all go at once, just when everyone thinks it's not going to happen because temperatures are starting to drop.

[Rubikscube]

Posted below is the average concentration map for 2. June (2002-2014) as well as a comparison to 2015. The third attachment is a direct comparison between 2012 and 2015.

Beaufort is still below average, but the lack of ice there no longer seems unprecedented. Very strange melting pattern in Hudson meanwhile, and I think Cesium62 is right that it is the most important reason why CAPIE/compactness is abnormally high. Laptev and Baffin is still lagging behind slightly, while Chukchi and Kara are quite far ahead at the moment.

The perceptive reader might notice the rather funny looking geographical shapes in the last map. This is because 2012 is only available in SSMIS, which is also the reason why that year is not included in my average maps. Though, interpreting this data from 2012 leaves me with the impression that a large portion of the difference in extent between 2012 and 2015 is because of Bering sea, and if I were to make a judgement exclusively based on this comparison (which is of course a highly dubious decision) then I would say the pack was in a worse shape three years ago.

[Vergent]

The arctic sea ice snow cover is about to be torched.



Here is the current snow cover.



Here is the forecast snow cover for the 12th.



Here was the snow cover on June 12, 2012.



Here is the snow cover on June 12, 2014



Here is June 12, 2013

My conclusion is that there will be high arctic melt pond formation even earlier than 2012. I think that this may be a really bad summer for the deniers and luke warmists.

[jai mitchell]

I concour,

I didn't want to say it on my last post but the current positive Pacific North America index is the absolute opposite of what I observed at this time in both 2013 and 2014.



looking at the predictions of massive heat coming off of Siberia and cyclonic activity moving significant ice past greenland and down into the atlantic bodes extremely poorly for this season's ice.

The post above that shows lack of snow cover is also indicative of the PNA index analysis where increased warmth, but reduced snowfall, has allowed the snow cover to melt.

Finally, it really looks like CAB central temperatures are going to overshoot even the early warmth that was observed in 2012

2012




My apologies to Gordon Lightfoot (and you all) but the ice floes of December will give up their masses when the melt ponds of June come early. . .



addendum
This shows that we might see rain on the CAB in the next few days  http://earth.nullschool.net/#2015/06/07/0300Z/wind/surface/level/overlay=temp/equirectangular=-7.39,89.61,1193

is that even possible???

[Vergent]

My apologies to Gordon Lightfoot (and you all) but the ice floes of December will give up their masses when the melt ponds of June come early. . .

This forum really needs a "like" button. 

Verg