PIOMAS July 2017

Another month has passed and so here is the updated Arctic sea ice volume graph as calculated by the Pan-Arctic Ice Ocean Modeling and Assimilation System (PIOMAS) at the Polar Science Center:

Last month we finally received some good news from PIOMAS, and the good news continues this month. With 2017 losing a below average amount of sea ice volume during June (compared to the average of the last 10 years), and a couple of years losing a great deal of volume, such as 2010, 2011 and of course, record smasher 2012, the gap has effectively been closed. At the end of May the difference between 2012 and 2017 was 1481 km3 (a month earlier it was 2412 km3 even), and now it's just 131 km3.

Here's how the differences with previous years have evolved from last month:

And here's Wipneus' version of the PIOMAS graph that shows even more clearly how large the gap has been since the start of the year, with the 2017 trend line now finally converging with its followers:

And although the trend line on the PIOMAS sea ice volume anomaly graph is still extremely low, at least it hasn't completely crashed to the point where it leaves two standard deviation territory (it probably will next month):

So, things look a lot better than they have for quite some time, but of course, that doesn't mean it's good. The weather hasn't been extremely conducive to melting so far, but it's no 2013 or 2014 either, where clouds and low temperatures saved the Arctic from falling in the hole that 2012 had left behind. This relative slowdown in melting is probably caused by the excess snow that has kept the thin ice white enough for melt pond formation to come later than it otherwise would have (as described in this bonus blog post on melting momentum I wrote a few weeks back).

The question now is whether this will be enough to keep 2017 from a reaching a new record low sea ice volume come September. Of course, only weather can prevent this from happening, and it will probably take some cold and cloudy weather. But only if the current trend follows the 2014 scenario, can a new record be avoided, as visualised by this Jim Pettit graph, showing all possible trajectories based on historic data:

As Jim says in the PIOMAS thread on the Arctic Sea Ice Forum: The average July-through-minimum melt of the last ten seasons would render a 2017 September minimum of 3.18 thousand km3, or nearly 500 km3 below the 2012 record.

This same thread has various graphs and maps on display, such as a comparison map in which PIOMAS data wizard extraordinaire Wipneus compares where this year the ice is either thicker (red) or thinner (blue) compared to a particular year. I've taken out 2010, 2011 and 2012:

As I wrote myself on the PIOMAS thread:

Very interesting to have the gap with 2012 close, but at the same time see this huge discrepancy, an almost diametrical opposition, on the comparison map. Sure, 2012 will catch up with the dark blue in the Chukchi and ESS where the ice has disappeared quickly this year, but the same goes for 2017 with all that red on the Atlantic side, especially that big blotch north of Svalbard and of course Baffin Bay. And then there's that last zone of red, in the CAA, that has received plenty of sunshine during the past few weeks.

It implies that it should be more than possible for 2017 to keep pace with 2012.

So, if the snow no longer plays such a big role, as satellite images and the NSIDC compactness graph seem to suggest, and weather and sea surface temperatures remain around average, 2017 will end up at least second lowest on record for sea ice volume (and if it does, other metrics will follow).

We'll know more next month, keeping in mind that 2012 didn't repeat June's volume scorcher, meaning the gap could possibly widen again.

-----

Only now do I notice that I had forgotten to post the average thickness graphs. Here's my PIJAMAS graph (a crude calculation of PIOMAS volume numbers divided by total JAXA sea ice extent), followed by the Polar Science Center thickness plot:

Comments

[iceman]

Yes, less alarming than it looked a couple of months ago. Yet volume anomaly still hit a record low during June. And as others have noted on the Forum, the picture appears worse if you give any credence to inferring volume from Hycom.
Further to Jim P's projection, one question is what path volume might follow to a new record minimum. Based on the thickness distribution, it looks like July would run ahead as the Atlantic-side melt picks up pace, followed by a slower August (that is, volume anomaly would turn up more steeply). OTOH if low-pressure systems predominate, they would give the central pack some measure of protection during July (which looks to happen in the coming week or so). But a deeper low during August would intensify wind-and-wave melting.
I'll refrain from guessing on this one: depends on the weather.

[DavidR]

It's worth observing as well that the gap between 2017 and 2016 increased by almost 200 km^3 over the month.

It is only really an extraordinary closing of the gap between day 158 and day 173 that saw 2012 approach 2017. 2011 the other big mover in June got 500 km^3 closer but still sits 750 km^3 behind in third place.

For the rest of the melting season 2012 had a fairly average melt by volume. As indicated a fairly average melt would see a record low volume and probably low extent and area. Even last year's area came within a couple of hundred 1000 km^2 of the 2012 record low.

[Neven]

Dr David Schröder from the University of Reading has sent me the modelled melt pond distribution maps for June. I've posted them as well as my thoughts (and weather forecasts for the next week) on the 2017 melting season thread, over on the ASIF.

Dr Schröder writes:

Based on melt pond fraction in May+June we predict a mean 2017 September ice extent of 5.1 (4.6 to 5.6) mill km2 (within the range observed during last 4 years). The likehood for a new record minimum is below 1% [my emphasis; N.]. While melt pond fraction has been above 2006-2015 mean values in the western parts of the Arctic, less ponding and melting occurred in the eastern part (see anomaly figures attached) due to more snow and relatively cold temperatures. In the past the regions - where melt pond fraction is low in 2017 - were more important for September ice extent (local pond fraction in areas enclosed by thick contour line show a negative correlation with mean September ice extent of R < -0.3) than e.g. the Beaufort Sea. Consequently, we predict the September ice extent to be quite large in spite of the lowest Arctic ice volume in recent months.

[David Appell]

Your posts have gotten very formulaic. Your heart isn't in them anymore.

Not a big deal -- it happens.

[Neven]

Which is also one of the reasons I'm on a sabbatical (except for these monthly PIOMAS updates and when there's something I cannot not write about).

But the 'problem' isn't that my heart isn't in it anymore, it's that my heart wants to do more than just describe the spectacle of Arctic sea ice loss (which inevitably becomes formulaic, as there are only so many ways you can describe this process).

[P-maker]

I believe the correct adjective is: laconic

as c.f. the Oxford Dictionary:

‘His laconic intellect and twinkling eye will never be forgotten by those who knew him.’

[Lennartvdl]

Or maybe 'iconic', as in: 'his iconic blog'.

[NeilT]

I'd say more like Heartbroken that Neven doesn't have the time to fully embrace the defining events of our lives and communicate it to us all in his excellent style.

It is soul destroying to see what is happening but be unable to keep this community engaged and talking and visible to the world; so that everyone can understand what is going on.

My view? Regardless of what is happening and how important this is, family comes first. After family there is time to worry about everyone else.

This is slow enough moving with such huge inertia behind it, that it will still be here in a few years when family pressures have eased and there is more time in life.

[Wayne Kernochan]

I'd like to delurk to add my two cents' worth (and I do mean two cents).

Neven, I strongly urge you to start a new blog about new technologies for building sustainable, minimum-carbon-footprint homes.

Your experience in that area is quite valuable, no one is putting the pieces together across the world from a practical-experience point of view, and it allows you to be entirely positive about one small part of the future.

Do it in parallel with Arctic Sea Ice. I for one would be fascinated by both.

[Hans Gunnstaddar]

Neven's doing what he can, which is what we are all doing. I'm just glad the site is still up and thankful to him for that. The posters on here are well versed on the topic and when there's something to hash out, we've got somewhere to toss our ideas back and forth.

That being said, I think it's too early to call this melt season. If it doesn't break any records that's good because it buys more time for the transition to renewables, which seems more promising with ramping up of the Tesla 3 production in the coming months. If enough charging stations and EV's are on the road, the price of oil drops due to a lack of demand and puts more pressure on those huge corporations to add wind and solar to their energy mix, while providing incentive to leave a lot of FF in the crust.

[Conrad777]

Newbie here. Why the fixation on September? The arctic has had 8 straight months of less ice volume EVERY DAY than any time in human history. Shouldn't that be more important?

[Elisee Reclus]

September is always the month of minimum sea ice in the North, and the sun is still fairly high in the sky. The longer the sun shines in September on blue water the less energy is reflected back to space and the more is absorbed by the wine-dark sea. The result is a positive feedback effect, making for a later start to the winter freeze, and less ice cover to melt the following spring. Because of this asymmetry, the lack of ice in the summer months has an exaggerated effect on the icecap in particular and the polar weather in general.

The ice reflection-temperature-sun angle interaction is inherently unstable: it doesn't tend to attenuate or damp out. All else being equal, it is more likely to initiate more and more violent oscillations, each more extreme than the last. There is even the possibility that if enough ice melts early enough the water will warm enough, deep enough, so that it will not freeze over again the following winter.

In other words, summer melt has a much greater effect on climate than at any other time of the year. And it is also the most extreme melt of the year. The plot of September ice minima for the years of the satellite era is the most dramatic, and least ambiguous, evidence we have of global warming. About 40% of the summer ice has melted away since 1979, and it looks like the situation will continue for the foreseeable future.

See NSIDC.org for the actual data. That's what convinced me.

[Robert S]

I certainly hope that Dr. Schroeder is right... but his analysis seems to me to put too heavy a focus on a single factor - an important factor, but only one of many... mind you, I'm often guilty of the same thing...

Based on my extremely qualitative estimate, I think that 5.1 million km2 is well out on the high arm of the bell curve - I'd put the middle of the curve at around 4 million km2, taking into account ice conditions, melt ponds, weather forecasts, etc... There's an awful lot of pretty "rubbly" ice out there...

[Robert S]

It seems to me that there are two primary "melt accelerating" ice conditions. One is coherent ice with melt ponds - coherent ice being ice that still has enough broad structure to develop 50+ km long leads. The albedo impact here is primarily from the melt ponds. The second condition is incoherent ice - ice consisting of bergs and bergy bits with no visible structure at a multi-km scale, where the albedo impact is from the open water between the bits, (as well as probably some local melt ponding) and where there is an additional ongoing fracturing and increase of surface area from wind/current effects.

Right now much of the CAA is in the first condition - coherent ice covered in melt ponds. Much of the western arctic ocean and a surprising percentage of the rest of the arctic ocean is in the second condition. The amount of coherent ice without extensive melt ponding appears to be quite limited.

[D-Penquin]

Off topic but may be of interest to some contributors.

"A one trillion tonne iceberg – one of the biggest ever recorded - has calved away from the Larsen C Ice Shelf in Antarctica...The iceberg, which is likely to be named A68, weighs more than a trillion tonnes. Its volume is twice that of Lake Erie, one of the Great Lakes."

All the details at:-
http://www.projectmidas.org/blog/calving/

[D-Penquin]

Events in the Antarctic appear to be running somewhat contrary to those in the Artic at the present time, in particular the melt days in Greenland are well below the levels of recent years.

[Rob Dekker]

There is something puzzling about the Sea Ice Volume Projection graph as shown by Jim Pettit :

http://neven1.typepad.com/.a/6a0133f03a1e37970b01b8d29304f5970c-pi

Notice that the recent years (2010-2016) show significantly LESS volume melt than the earlier years (80's or 90's line).

Does anyone have a physical reason for that odd difference ?

[Rob Dekker]

And may I note that the volume loss line with the LEAST volume loss is 2012, which happens to be the year with the MOST extent loss.

[Rob Dekker]

Sorry guys, I'm not sure about that last claim. I'm not sure which year had the least volume loss since the start of July through September.

Yet the first claim (that volume loss seems to decrease over time) still stands.

[Neven]

I would say it's because there's less 'easy ice' on the periphery.

[AmbiValent]

And this year, there's more ice on the Atlantic periphery than 2012. I think it would tend to be 'easier' than ice in the CAB.

[John Christensen]

Hi Rob,

I would agree with Neven: In recent years area and extent has gone so low that it could cause a decrease in volume melt rate for remaining ice. Reflecting a drawn out end to summer ice rather than a sudden crash possibly..

A separate observation for this season is an apparent reduction in north-bound sea current in the eastern Fram Strait: We have in recent years seen wide open water to the north-west/north of Svalbard, but this area got filled with south-bound sea ice early in the year and has stayed ice-filled, which to me seems to indicate a reduction in north-bound sea current, but also could be due to an increase in south-bound sea ice flow. A reduction in sea current would better explain the slight increase in sea ice in northern parts of Barents as well, so is where I am leaning, but have seen no news on Fram current.

[D]

There is less easy ice on the periphery in the first half of the year also, but volume loss has skyrocketed in that period. Recent years have had significantly more overall volume loss than years before 2007. Most likely the explanation is that ice is becoming steadily thinner. This allows insolation to increasingly dominate the melting season. More ice will melt at the peak and early in the season. Less ice will melt late as insolation decreases. It's possible that less available ice to melt will also become a factor at the end of the year.

http://forum.arctic-sea-ice.net/index.php/topic,2098.0.html

[Wade Smith]

This is not really the right place to put this, but I don't know where to put this. I really need to find an Entomology expert to discuss this.

[snip; I'm sorry, Wade, but this is the wrong place to post all of this stuff. I've copied your comments, so mail me if you want the text; N.]

[fryingpan136]

It seems there is a blue streak through the middle of the arctic basin.

https://go.nasa.gov/2ufJQtu

[fryingpan136]

Looks like shadows from clouds.

[Rob Dekker]

Excellent points, D.
Thinner ice could indeed imply increased sensitivity on insolation.
That could explain the rapid decline of volume during June of the recent years that is so clearly visible in Wipneus' volume anomaly graphs :
https://sites.google.com/site/arctischepinguin/home/piomas

But that still does not explain why volume loss in the second half of the melting season (July to September) is actually less than in the 80's.

[NeilT]

Rob, would that be due to the fact that there was more peripheral, weak, thin, ice out there later in the season in the 80's than now? We have seen all the peripheral ice melt out in the last decade, only leaving CAB ice behind. This was not true in the 80's.

As the ice reaches total failure point, volume loss, in the latter part of the season, must start to fall in the charts because there is simply less surface area, everywhere, to absorb heat, because there is less area and extent. The more compact the ice the less area too.

Then, of course, we have to wonder about rotten ice. It gives every signature of MYI, but has the strength and heat absorption capability of FYI or less. We know the satellites can't detect rotten ice so, I wonder, how do the models compensate.

If the ice is melting from within, but not giving a signal of melting, then, would we not, see something similar to this slowdown?

I did find an abstract about the heat sequestration and transport of melting snow cover on ice, I requested the full article and received a reply of "when the authors deliver it". I think, in the case of this year, it could be important too.

Right now we are almost at solar minimum, Nino is neutral, there does not seem to be any single,"normal" driving force for what we are seeing today. Yet, we're bumping along the bottom of the lowest charts for Extent and Area and off the bottom for Volume.

This would fit the hypotheses of a cycle, where the Arctic responds in a certain way throughout the cycle. Influenced by strong weather, certainly, but without strong influence it continues on the path.

The CAB is in a very poor state today, peripheral areas to the CAB are likely to melt out, the NSR and the NW passage look likely to be open in the next 2-3 weeks and we still have August and the storms it might bring to come.

The last 3 days has seen Extent drop by an average of 132k per day which will drop 2017 much closer to 2012. 2012 only averaged sub 100k over those same days. However the overheads show the peripheral ice changing colour and likely to disappear faster, not slower.

Nothing can be forecast, right now, from observations because they are not that consistent.

That being said, we can make some assumptions. My assumption will be that this melt will continue to exceed expectations in the periphery and the CAB will continue to show a weak and broken state.

That then leaves us in the shooting gallery of August storms as to whether there is a record or not.

I do, however, expect it to drop significantly below 2016.

[P-maker]

D, Rob & NielT,

Simple mechanisms could explain this apparent change in late season volume decrease from the 80’ies and 90’ies to the current decade:

Back then, a larger proportion of June ice would consist of MYI with a little snow on top. As soon as the snow was gone, it would be a hard struggle for the sun to get rid of the rest because the ice was so clean, dense and cold.

June ice nowadays is mainly thin FYI with lots of dirty snow on top, hence June melt progresses quickly (around 20-25 % of the total volume at the end of May is now lost in June).

Back then, volume of FYI + MYI at the end of June was still significant, hence average melt from summer solstice to the minimum in September would be in the order of 13 – 15,000 km3.

Late summer melt nowadays is only in the order of 11- 13,000 km3, but percentage wise the melt over the late summer has gone up from around 50 % of the annual total loss back then to around 60-70 % of the annual total loss today. Hence, the June cliff may be the cause of the late summer decline in total loss (simply less ice available).

In order to explain the lower absolute melt rates during late summer nowadays, one could either speculate in harder, denser, fresher MYI (hardly likely), or one could speculate in the socalled Atlantification process over the Arctic Ocean leading to more clouds, higher humidity, more snowfall and stronger winds, which would eventually lead to larger extent (relatively) and less volume loss (absolutely), just as we have seen.

[NeilT]

Hence, the June cliff may be the cause of the late summer decline in total loss (simply less ice available).

Oddly I thought that was what I'd said.

[P-maker]

Neil,

Less ice does not necessarily explain less melt. It depends both on the quality of the remaining ice and the character of the remaining summer, as I tried to explan in my last paragraph.

Your description of the peripheral ice may very well be valid as well. I was not discussing any kind of cycle.

[Rob Dekker]

I have another theory, and one we can possibly test :

Notice that the reduction in volume loss since the 80's happens mostly during July (see Wipneus' "Arctic Ice Volume Anomaly" graph here) :
https://sites.google.com/site/arctischepinguin/home/piomas

And remember that in recent years, July is when the melt starts to eat seriously into the Arctic Basin, which during the 80's the melt line was still outside the Basin.

So if PIOMAS underestimates the thickness of ice in the periphery of the Arctic Basin (and possibly overestimates the thickness of ice outside the Basic), then PIOMAS would record less ice loss during July in recent years, compared to earlier decades.

The estimate does not need to be much : just a 20% error in thickness would explain the (PIOMAS recorded) reduced ice volume melt during July.

Does that make sense ?

[Rob Dekker]

P-maker, back in the 80's, in June the melt line would be outside the Arctic Basin, where there is no MYI.

[NeilT]

P-maker. I know you weren't discussing a cycle. Only a few of us do that.. :-)

Apart from the fact that pretty much all the peripheral ice melts out nowadays and the peripheral ice is mainly very thin, I would assume that less surface contact with heat transfer would generate less melt. It doesn't matter quite so much if there are 100,000 sq miles of warmer water or 1M sq miles of warmer water, if only 80,000 sq miles of water is in contact with the ice whereas before there was a full 100,000.

Yes there is a larger "warm sink" but heat transfer through water is not very fast and also the melting ice creates a barrier between the ice and the warm water surrounding it. At least for a while.

[Hans Gunnstaddar]

"The last 3 days has seen Extent drop by an average of 132k per day which will drop 2017 much closer to 2012. 2012 only averaged sub 100k over those same days."

NeilT, good call, as we can see from the graph below, 2017 extent changed direction and is dropping towards 2012. http://nsidc.org/arcticseaicenews/

[Hans Gunnstaddar]

https://seaice.uni-bremen.de/data/amsr2/today/Arctic_AMSR2_nic.png

Take a gander at the Pacific side. Just since yesterday the CAB perimeter melt has increased noticeably.

[Jim Dowling]

@Hans - i noticed the same thing. A big swing. If you look at the sea ice thickness here, http://ocean.dmi.dk/anim/index.php, you can see that a big chunk of the Pacific side is going to melt out before August.
Today was the first time i became convinced that 2017 finish below 2012 - driven by the Pacific side. I don't even think it needs August storms now - it will melt out in place.

[Hans Gunnstaddar]

The link didn't work but so glad someone else noticed too, Jim. The process as you know is to ck. these graphs daily to try and detect changes, big and small.

This was a very big change and fast! I mean the day before it looked like the ice in that area was putting up a good fight, holding ground with coloration as seen on Bremens at a high percentage, then suddenly the next day, the one I posted a link to, just must have given way on a massive scale to then show ice bound to melt far into the Summer melt. I mean that's how fast it changed; from daily minor variations to sudden retreat (so to speak) on a perimeter probably around a thousand miles long X two-three hundred miles deep into the CAB. It actually game me a scare. Think about how much energy it took to do that, that fast.

I have no idea when all the ice battered in that major shift will all melt, but just to make a move like that was amazing. Now, it might not maintain that speed of momentum. We'll find out tonight. I think we're going to see some amazing moves like this moving forward trough mid-Sept. How often I don't know, but there is a confluence of metrics coming together this melt season, in spite of some that work the other way.

I highly suggest looking to the perimeter of the Cab on the Pacific side tonight after the graphs update. I like the false color Bremen as it paints a picture of not only the condition but how flows and ice conditions are changing.

[Hans Gunnstaddar]

https://seaice.uni-bremen.de/data/amsr2/today/Arctic_AMSR2_nic.png

Maybe I'm just posting to myself at this point, but the change from yesterday is a large pocket of ice melted out with more likely to melt, so it wasn't just a one day fluke. Seems most likely it's occurring due to a warm influx of tidal water from the Pacific side through the Chukchi. Will be interesting to see if extent continues to drop towards 2012 levels when the graph gets updated to the 17th.

[Robert S]

I'm starting to consider the buttressing effect, similar to that found with the Antarctic glaciers, for the arctic ocean ice. My theory is that the peripheral ice buttresses the core, slowing its drift, and more the peripheral ice melts out, the easier it is for storms and normal circulation effects to start to move the core ice north of Ellesmere and Greenland. If that ice gets moved out of that area over the summer (and I think we're already seeing some of this happening), the result may even be a short term increase in extent, due to drifting MYI flows, but this area now becomes likely to melt out each year, which would accelerate the arctic toward ice free...

[Neven]

Mid-month PIOMAS update over on the ASIF.

[NeilT]

Looking at the way the Atlantic side is disintegrating, on the Arctic Mosaic overheads, I'm quite interested to see how close the daily Extent on NSIDC gets to 3M

There is now significant damage on the Pacific side, significant damage on the Atlantic side and the periphery is finally succumbing to the relentless heat and rain.

Volume is also showing that it's starting to dip more sharply again, hardly surprising given the amount of area and extent that has been vanishing in the last two weeks.

NSIDC running 5 day average has 2017 just above 2012 but closing as the higher rates of loss in 2017 compare to the lower rates of loss in 2012 at this time.

Regardless of what happens now, it looks like 2017 will drop well below 2016 and I'm looking forward towards the next freezing season and how it evolves.

Personally I see the next two freezing seasons and their attendant melting seasons following as the final story in the potential for a 5 year cycle.

[Sam]

On Worldview, almost all parts of the Northwest Passage shattered today. That has to end up with significant impacts to all of the usual measures in the near future (area, extent, volume).

Sam

[Hans Gunnstaddar]

Good summary, NeilT.

http://nsidc.org/arcticseaicenews/

For those that haven't seen the latest extent and sea ice concentration, here's the link: It's updated through the 17th.

2017 extent and 2012 running neck and neck. It's a horse race. What I'm seeing is great variability in the consistency of the ice across the CAB as a whole. It looks like in the coming weeks, instead of simply contracting from the outside inward - wide open gaps may appear. I've been anticipating that point in time when the forces applied lead to open ocean pockets in random areas of the CAB, which if on a large enough scale could begin to disintegrate it's cohesiveness, leading to a break up.

[AnotherJourneybyTrain]

We, it seems to me Hans, are talking about serious advection of heat: are we not?

"Can we observe such expected(from my point of view) advection of heat taking form in any quantified way", becomes the question I suspect!

[AnotherJourneybyTrain]

Hey, today could finally be the day that 2017 passes 2016 for the first time since mid-April! Also, current extent is now lower than the minimum extent for 1980, and will probably pass most of the rest of the 1980s' miniimums over the course of the next week.
<<
Gregcharles made an interesting comment over on the forum!

So, "...the seasons seem to have moved by approximately a month in the Arctic over 37 years...", or is that an immature and inappropriate statement!

[NeilT]

We, it seems to me Hans, are talking about serious advection of heat: are we not?

I'd say we have already had serious advection of heat in the form of rain, all over the season. The main reason it has not really registered heavily so far, in terms of overall ice melt, I'd guess, is the very heavy snow cover.

But that heat is there and it's not going away. More heat just increases the impact on an already weakened system.

This seems to me to be one of the drivers that has driven the state of the CAB where it no longer resembles a pack.

[Hans Gunnstaddar]

"We, it seems to me Hans, are talking about serious advection of heat: are we not?"

AJT, I concur!

[Hans Gunnstaddar]

AJT & NeilT, this article just out;

https://robertscribbler.com/

"Remove a significant portion of the global sea ice, as we have done, and you’ll end up with oceans that both draw in more heat during the warmer months and bleed out more stored ocean heat into the atmosphere during the winter."

[NeilT]

I was recently looking back through the Chartic extent.

Roughly:

2006 reached the 2003 low
2011 reached the 2007 low

But 2016 only really equalled 2011. It didn't reach 2012.

So, in a way, I would not be surprised if we only equalled 2012 or went under it slightly. Of course if we went significantly under it, I would not be surprised either.

However, for me, I'm still looking forward to 2018/19 to see whether we continue with the losses of 2016/17 or whether we go back to a cycle of very limited re-growth for 2 years.

What is looking most interesting, to me, is the swathe of ice in the 80n to 85n from Svalbard to the Chuchki.

That looks vulnerable to a storm or two which could cause a very significant change in the ice.

Or not.. That is the whole thing about watching the Arctic.

For me I have come to the conclusion, in my quest to understand why the ice does what it does, that if we have 5 year cycles, then each year within that cycle is a different melt season. 2017 is peak melt so should act like peak melt. 2018 will be peak growth and should act like peak growth.

It has been and will continue to be very interesting to watch.

[AnotherJourneybyTrain]

Personally, I like to watch the 50 day forecast on The Great White Con pages even if I have been told they aren't exactly accurate.... it's a 50 day forecast so it deserves some credit just for it's bold nature IMHO!

You are trying to see 5 year patterns,... that's fair enough: you are trying to understand the topic by closer and closerr observation just like everyone else. Aslong as we keep looking and learning and start nutting out a few answers until we all learn what needs to be done as a species it's all fair game.

https://pbs.twimg.com/media/DFHAHLBXsAABqCk.jpg

....it's all pointing in one direction: but solid numbers are necessary for big change to take place.

[NeilT]

Not quite AJbT. I used to put every day under the microscope, I used to try and understand why each day and week were doing what they were doing.

None of it made any sense because every year seemed different.

It was only when I took a step back and looked much further out and did comparisons way beyond day, week, month or year that some things seemed to make sense.

But, yes, I'm still trying to learn.

I didn't use to have much patience for RealClimate's assertion that you could only see what was happening on a multi decadal scale. Especially when things were moving so fast and accelerating so fast. I have moved much more towards their way of thinking over the last few years.

[Hans Gunnstaddar]

https://seaice.uni-bremen.de/data/amsr2/today/Arctic_AMSR2_nic.png

Pacific side still getting battered vs. the Atlantic side which has settled back down.

Here's a question for whomever would like to try and answer it for this point in the melt season; What percentage of melt on the Pacific side is warm water from that ocean and what percentage of melt is from the weather above the ice?

[AnotherJourneybyTrain]

Hans, firstly: we need to answer if the rain is actually over the ice or just open ocean!

[VCAC]

End of melt season and a lot of rain 23/09 +/- 3days?

[Susan Anderson]

I suggest connecting events in the northern Pacific with events further south. For several weeks I've been watching a developing necklace of cyclones and tropical storms, starting south of Mexico and extending over to Japan as you can see here:
https://earth.nullschool.net/#current/wind/surface/level/overlay=temp/orthographic=216.15,53.36,443

Fernanda which is now fading out by Hawaii has been churning at Category 2 or 3 for weeks. The enthusiasts and experts who comment at Wunderground at saying over by Japan there might be Fujiwhara effect; one of them just made a loop with itself. You can see that it is shoving a lot of energy through the Bering Sea.

The whole thing looks to me like it's powerful enough to be sending some of the energy across the land into the Arctic area as well.

I'm normally a little hesitant about my detours here, but this grouping has been pretty intense and I think the connections are there.

[Hans Gunnstaddar]

Fascinating viewpoint from that link, Susan. Thanks for sharing.

"You can see that it is shoving a lot of energy through the Bering Sea."

That explains it, i.e. why the sudden melt on the Pacific side of the CAB, which started about a week ago. I knew it had to be something rather dramatic because it changed so fast from not much melt to a great deal of melt and is continuing on a torrid pace. 2017 extent remains close to that of 2012 due in great part to that influx because the rest of the CAB is melting much slower.

Now maybe I'm reading too much into it, but if you look at the Atlantic side (by grabbing the globe and rotating it), there is a flow from the Atlantic but also from over the top of Greenland. Correct me if I'm off on this, but even if the Atlantic air is warm, once they converge and mix its cooled and that's probably why we aren't getting as much melt on the Atlantic side.

[AnotherJourneybyTrain]

VCAC, was that a 70ish-day prediction?

[NeilT]

I noticed the same thing Hans. Where it is melting on the Atlantic side, it's coming around the coast. Where it is not melting in the CAB, it's coming over Greenland and rotating over the CAB.

What is more interesting is the state of the ice on the Atlantic side, now showing on the Bremen concentration maps as areas of open water N of 85N.

Also the bottom melt continues but is relatively invisible at this time.

It's going to make August rather interesting.

[Susan Anderson]

Thanks, I'm glad others can see it. It's only one piece of the puzzle. I've had to learn about scale and time following the cluster of related phenomena that add up to "global" warming. It's all too easy to reduce it to one's own scale. The earth/ocean is pretty big, and time vastly bigger. It's tempting to generalize from the particular.

One thing that this does demonstrate, though, is that without the modifying effect of a much larger and more stable ice cap, these disruptions are more disruptive. When it's hotter all over (particularly at night), there's more likelihood of the heat being pushed north.

[AnotherJourneybyTrain]

Neil, how do we know the bottom melt is continuing?

[NeilT]

AJbT, because we know the SST's in the areas where it is melting and we know that the currents draw that warmer water under the CAB.

Prior Buoy data over the affected areas has given us a wealth of stats which show a range of bottom melt during the season.

We also know the thickness of the ice and we know that FYI suffers from solar penetration and insolation of the solar energy in the water below. Also causing bottom melt.

FYI is predominantly salty ice which has not built up a store of snow and frozen water from rain which did not escape. Salty ice melts at -1.8C.

So we know bottom melt is going on, we know roughly how much will melt, in a normal season and we know how thick the ice is.

That is leading to some concerns as to the retention of the ice this melting season.

[Jim Hunt]

AJbT - Several years history of buoy measurements of sea ice thickness are available at:

http://GreatWhiteCon.info/resources/ice-mass-balance-buoys/

NeilT - If I may I'll quibble slightly with your "FYI is predominantly salty ice... Salty ice melts at -1.8C". See for example:

https://www.cambridge.org/core/journals/journal-of-glaciology/article/salinity-variations-in-sea-ice/0976B42E5B21CAE25BE26A0DD811513F/core-reader

[NeilT]

Thanks Jim, I should have said tends to be saltier. That would have been better and more accurate.

[AnotherJourneybyTrain]

http://nsidc.org/arcticseaicenews/files/1999/07/asina_N_iqr_timeseries.png

Late June - early July, 2013, seems to have the fastest rate of sea ice extent decline on record.

This years mid-July update at NSIDC says:

Tendency for warmer winters is increasing

A new study published this week in Geophysical Research Letters by Robert Graham at the Norwegian Polar Institute shows that warm winters in the Arctic are becoming more frequent and lasting for longer periods of time than they used to. Warm events were defined by when the air temperatures rose above -10 degrees Celsius (14 degrees Fahrenheit). While this is still well below the freezing point, it is 20 degrees Celsius (36 degrees Fahrenheit) higher than average. The last two winters have seen temperatures near the North Pole rising to 0 degrees Celsius. While an earlier study showed that winter 2015/2016 was the warmest recorded at that time, the winter of 2016/2017 was even warmer.

--> Reflecting back to the first graph: 2014 seemed to have the same sort of rapid decline in the same period(as 2013 late June to early July), whereas 2015 didn't show quite the same rapidness. 2016 we all know looked very scary for a large part of it's melting process and even though 2017 hasn't really reflected large rates of change it was characterised by it's low volumes to reveal, I think, to us all that very real trends seem to be at play and that the record of past years seem to be playing an increasing importance toward the expectation of the immediate following year.

(Yes, I stand ready to be humbled by the Arctic... but observations are there to be made are they not?)

[Jim Dowling]

Just looking at the GFS forecasts for the next 2 weeks, and it looks like there's a storm coming (no surprise there).
But there are also signs of a reverse dipole becoming established where the high pressure is over Siberia and the low pressure over the Beaufort. I say reverse dipole, because typically we see a dipole where the High is over Beaufort and the Low is over Siberia.
I've been observing the Arctic Sea Ice for about 5 years now, and it's the first time I've seen this in the middle of summer. Maybe somebody has seen this before?

Just to speculate on the a reverse dipole would have at this stage:

* ice pushed away from Svalbard towards the warmer waters on the pacific side
* possibly an ice-free Nord Pole!
* better for retaining ice than a classic dipole that pushes ice to its death in the Fram Strait.

https://ibb.co/gL3B4k

[Hans Gunnstaddar]

https://robertscribbler.com/

'Study Finds 21 to 25 Percent of Northern Permafrost Will Thaw at Just 1.5 C of Warming'

"Perhaps more concerning was the fact that the study found that this 1.5 C temperature threshold was reached by as early as 2023 under the worst case fossil fuel burning scenario even as it was held off only to 2027 if rapid fossil fuel burning reductions were achieved."

[Jim Dowling]

The first storm of the summer appears on it's way. August 3rd/4th. This pic is for Aug 4th.
It looks like a low with a center of around 975mb.

https://ibb.co/i40wpk

[Werther]

Jim, hi,

Saw that too, in my case on ECMWF. Still a week out, formation of 'dipole' starting on Sunday/Monday. ECMWF storm not that close to CAA, more ESS-CAB boundary. That could crush the pack in the Pacific sector.

[Hans Gunnstaddar]

https://seaice.uni-bremen.de/data/amsr2/today/Arctic_AMSR2_nic.png

Remember an earlier post of mine suggesting ice free pockets in the CAB, well, there's a big one between the NP and Laptev with other areas appearing like they may open up soon.

2017 extent is greater than 2012 at this point, but August and early Sept. will be the tale of the tape for this melt season and there are sizable areas with lots of melt potential. Sizing up to be an interesting period, especially if that soon to take shape dipole has a big impact.

[tony rickerby]

Hi,newbie here,i have a question.
Watching the weather via ventusky i see a lot of cloud in the arctic north.is this normal for summer?
Many thanks to you all (especially neven) for this discussion.

[tony rickerby]

https://www.ventusky.com

[NeilT]

How about a slightly different question. If, as we believe, the cycle has years predisposed to melt as well as years predisposed to ice growth, then it would take exceptional weather to stop the melt this year.

That is somewhat different than believing that it will take exceptional weather to create a surprising melt this year.

[NeilT]

Let me quantify. "We believe" was Hans and me as I was replying to his question.

[AnotherJourneybyTrain]

I don't mean to be distasteful but the link below may be important to discuss.

https://www.theguardian.com/environment/2017/jul/31/suicides-of-nearly-60000-indian-farmers-linked-to-climate-change-study-claims

We bring the world into being through language so obviously no one wants such things to be used for possibly alarmist ends. So the question becomes whether the above linked article is justified?

[Hans Gunnstaddar]

I just saw a YouTube video on that topic last evening, AJT. Very disturbing but probably going to get much worse as water supply continues to decline.

On another front;

http://nsidc.org/greenland-today/

Greenland melt was below average earlier in the Summer but has recently increased far above the average. Worth a quick peek.

[AnotherJourneybyTrain]

Yep I was just there, but for a large part of that time series it was below normal.

[John Christensen]

Welcome to you Tony!

To your question: Yes, clouds and general low pressure is the norm for the Arctic during summer months, primarily due to cold surface temps compared to surrounding continents - quite similar in fact to sea clouds appearing off the cost of Oregon due to cold Pacific surface temps.

Specific to the Arctic, Neven put together some great info here:

http://neven1.typepad.com/blog/2013/06/on-persistent-cyclones.html

[Kevin McKinney]

AnotherJourneyByTrain:

Your question is completely unrelated to Arctic Sea Ice. You could try asking it elsewhere, say RealClimate:

http://www.realclimate.org/

However, in the spirit of saving you some time there, I will gently suggest that you ask it more clearly than you have asked it here. I think many will be quite curious as to just what you mean by "possibly alarmist ends." Remember, "alarmed" calls a different world into being than "alarmist."

[tony rickerby]

Thanks john christensen,i'll read the link.

[AnotherJourneybyTrain]

Kevin, I'm asking is the 60 000 deaths actually justifiably linked or not: I am not out of order in my questioning of that!

[Actually you are, because I've told you before not to ask so many questions, especially if they're off-topic, so let's leave it at this; N.]

[Daniel Bailey]

"I'm asking is the 60 000 deaths actually justifiably linked or not?"

You could actually read the linked article. Then you would have noted the link there to the published research upon which it is based.

"possibly alarmist ends?"

A word to the wise: when you use the argot of denial, some will react accordingly.

[John Christensen]

Still 5-6 days out, but ecmwf is forecasting an intense low just north of the CAA.

[Wade Smith]

Fortunately, it takes a lot more CO2 than we currently have in order to melt down the arctic. Curvature of the Earth is really applying a negative feedback to the trend line for average Sea Ice thickness. It may take even longer to deplete than I previously suspected.

The next 5 months should be interesting.

Will we verify last year's ridiculous non-recovery for Global Sea Ice Area?

Will the Antarctic Sea Ice Area begin a permanent negative trend now?

How big can the biggest ice berg calving event be?

[Daniel Bailey]

"it takes a lot more CO2 than we currently have in order to melt down the arctic"

Out of curiosity, what's your evidence for this claim? Under current emissions scenarios, we're looking squarely at 5°C global rise by 2100, with much higher temperatures in the Arctic. Such are incompatible with summer ice anywhere in the Arctic, long term.

[Hans Gunnstaddar]

https://earth.nullschool.net/#current/wind/surface/level/overlay=temp/orthographic=250.92,73.64,443

Take a look at:

1) the area of wind circulation out in the orange color coded area of the Pacific with a wind speed of 64 km/h & temp. of 27.2C that then feeds into;

2) an area of circulation in the Bering Sea (close to the Chukchi), with wind speeds up to 30 km/h at 8.2C feeding;

3) directly across the CAB, and into another area of circulation right above the north pole, with wind speed up to 28 km/h with a temp. of 0.0C.

Temp. in those 3 locations goes from 27.2C to 8.2C to 0.0C.

[Jim Hunt]

Hans - There's more winds forecast for Sunday that should follow that with 2 meter plus waves heading for the ice edge in the Chukchi/Beaufort Seas:

http://GreatWhiteCon.info/2017/08/facts-about-the-arctic-in-august-2017/

[Sam]

Wade,

The arctic is already in full meltdown headed for catastrophic melt. That much is clear for anyone to see looking at the data. We do not need more CO2 for that. Yet we will have much more released before we are done. The complete loss of the artic ice and disruption of both the atmospheric and the oceanic circulation is a done deal.

And we haven't yet even begun to see the biospheres response added to what we have done.

Sam

[Hans Gunnstaddar]

Should be interesting to see the effect of those waves - thanks for the link, Jim.

[AnotherJourneybyTrain]

Is it just me or is wade smith asking too many questions:

[Actually, it's you asking too many questions. Snipped the rest, as it's no use. Leave these things to me, okay; N.]

[AnotherJourneybyTrain]

[snip; N.]

[Bill Fothergill]

Sam said... "The arctic is already in full meltdown headed for catastrophic melt. That much is clear for anyone to see looking at the data. We do not need more CO2 for that. Yet we will have much more released before we are done. The complete loss of the artic ice and disruption of both the atmospheric and the oceanic circulation is a done deal."

On that depressing theme, a group of scientists from the Max Plank Institute of Meteorology published a paper in the Journal of the American Meteorological Society last year. Its title rather says it all: "On the Potential for Abrupt Arctic Winter Sea Ice Loss".

http://journals.ametsoc.org/doi/full/10.1175/JCLI-D-15-0466.1

N.B. The CO2 trajectory assumed in said paper is RCP8.5

There is some discussion of this in the "Year-round ice-free Arctic" thread on the ASIF.

[Bill Fothergill]

@ Neil T & Hans G

I've just posted a little chart pertaining to global sea ice extent on the ASIF's 2017 melting season thread. (See #3562)
http://forum.arctic-sea-ice.net/index.php/topic,1834.3550.html

Notice anything about the periodicity with which the humps seem to appear?

[Sam]

Bill,

I had not seen that paper yet. Thank you for pointing it out. It is terrifying in its implications. However, it suggests things we really need to be aware of. And it applies directly to ice volume.

Caveat: the paper details and contrasts what the models suggest. And, the models are just that models, not the reality they are modeling. That they all show the same behavior suggests that we seriously need to pay attention. They can however mislead us. Most importantly, they may under represent the rate of change because of the large number of missing feedback mechanisms (clathrate collapse ...) and missing unknowns (cloud formation in winter ...).

Still, what they show is a roughly 30-35 year transition from a seasonally ice free Arctic summer to a perennially ice free Arctic! Yikes! And this comports well with recent studies of the PETM transition.

We now appear to be within five years of the first ice tree Arctic summer (<1 million km^2). This then suggest that we are within 35-40 years of a perennially ice free Arctic (so about 2048-2057).

Personally, I am of an age such that I may live to see that. And I find that personally terrifying. With the added feedbacks and unknowns, it may occur sooner than this.

My larger concern is that as we approach this transition, that the dynamics of the oceanic circulation and of the atmospheric circulations are likely (will undoubtedly) change in dramatic ways, ways that our models are not likely to be good at predicting.

These changes may (likely will) have major impacts first on ice thickness and volume, then next on temperatures and precipitation patterns and seasonality all over the northern hemisphere. This has immense implications for crop production, drinking water, power production, drought, deluge, pestilence, war, and other feedbacks.

What this all points out is how critical it is that we never allow conditions to reach the point that we see an ice free Arctic summer. Yet, that is already baked in. We cannot stop that. So now what do we do? What can we do?

Sam

[Sam]

Clarification: the PETM transition is a very different kind of transition. It applies to how rapidly the state of the world changed with with CO2 and methane changes. The ice models have similar timing for their transition.

[Hans Gunnstaddar]

Neil T & Hans G: "Notice anything about the periodicity with which the humps seem to appear?"

Yes, the undulating 5 year repeating cycle and with the 4th year, 2016, being as low as it was, I figured a new record low extent for this melt season, 2017. At present 2012 is lower and dropped quite a bit during August/Sept., so starting to look like it might not set a new record extent low, but it's not done yet as 2 meter waves and storms will make their mark.

[Wade Smith]

Bill,

It stands to reason that as summer time minimum sea ice continues to decrease, then further albedo feedbacks will continue to make weaker and weaker recovery seasons. Although the sunlight penetrates all the way down to 700 to 900 meters, therefore doesn't all go to warming the surface, this would imply that over multiple decades following the first summer meltdown it might be possible to warm the SST enough to totally prevent refreezing. The continental ice of Greenland and Alaska may have a lot to say about that though, because you can only warm the ocean so much before thermodynamics requires the heat to start going into melting the land ice more and more.

I'm pretty sure Greenland is already toast over the long term, but I doubt humans will be dumb enough to burn enough carbon to totally melt down Antarctica.

[Wade Smith]

I expect Greenland to experience about a 5% to 10% meltdown by the end of this century. Then after that, albedo feedbacks on land are going to greatly accelerate the loss of ice in the following century. So if you see 2.5 to 3 feet of sea level rise by the end of this century, you will probably see a further 10 feet or so of sea level rise in the following century, due to self-reinforcing feedbacks. This should be especially in Greenland and also especially the Himalayas, since they actually get much more sunlight during the year.

The change in the summer time daytime high temperatures for both Greenland and the Himalayas is going to be radical by the time this process is finished, because we're talking an albedo change from 0.8 or 0.9 all the way down to 0.2 or 0.3. So we're not talking about a 5 degree regional temperature change once this happens. We're probably looking at a 15 to 20 degree change in daytime high for these regions by the time they completely melt down. It might even be more than that. Like Everest will be ice-free eventually in a couple centuries.

[Wade Smith]

So basically in both Greenland and the Himalayas, and all the streams and rivers fed by them, there is going to be major biome changes over the next few centuries. As the glaciers melt, you'll have trees, grasses, and shrubs move in, either planted by humans or else seeded by the winds and birds. However, these streams and rivers are not going to be fed in the same way as they currently are fed. Once the glaciers are depleted enough, there are going to be major dry spells in these watersheds which currently don't occur. Obviously, this is going to greatly effect wildlife and fisheries, not just the human populations downstream.

Also, this is eventually going to lead to increase in Earthquakes in these regions, as mass loss of that much ice is going to cause isostatic rebound, and both regions are close to tectonic plate boundaries. This could be an especially big problem for the Himalayas region, since it can already have major earthquakes to begin with.

[navegante]

Wade Sunlight doesn't go all the way down to 700 meters. Most of the radiation is absorbed within the few first tens of meters, a region which in most oceans is well mixed, and the heat distributed. Only by whatever downwelling presrnt can that heat reach abyss.
Please Wade don't make specifics up, it is not really needed for your nice apocalyptic narrative.