Is the June cliff (mostly) caused by the geography of the Arctic Ocean?

[seaicesailor]

During April and May melt front advances through Bering, Okhotsk, Barentz, and so. The front as a whole is broken and limited in size.

It happens in June that ice along the shores of Beaufort, Chukchi, ESS , Laptev, and Kara ( the internal shores of the Arctic properly speaking) starts melting or compacting, and so opening new melt fronts.

If the opening happens in one or two weeks, suddenly a very long front from Beaufort to Kara is established that may support much faster melting during June, advancing thru the internal peripheral seas.

I think this is a good moment to share this idea since HYCOM shows for the coming week what seems big compaction caused by this anticyclone that is coming over Chukchi; a large melt front from Baeufort to Kara may be established.

I think this geographic effect might be a main contributor of the  acceleration of extent loss known as June cliff, although not the sole factor.

In May the front does not exist and in July it is well established (not fully otherwise there would be open route already); the area melting rate by an advancing "circular" arc of such dimensions can be very big even when the average front velocity is not so espectacular. My only supporting evidence.

[Nightvid Cole]

There are a couple of good alternate explanations for the June cliff:


(1) Melt onset requires some threshold amount of insolation - so that earlier in the season (especially April), the insolation gradient as a function of latitude is very steep - thus only a thin band has the correct level of insolation for surface melt onset (too low and it does not happen yet, too high and it already has happened). Since only a small portion of the surface at any given time is melting, this means a low rate of increase in area which has reached surface melt onset. By contrast, in late May and in June, the North pole is actually the point of maximum insolation under clear skies - so the high insolation sort of hits the whole Arctic "all at once".

(2) In order to have a lot of melt ponds form to create a "cliff", the ice surface must be able to support melt water without allowing it to percolate through, simply flow downhill and off the floe's edge, or otherwise get off the surface. Because areas outside the Arctic Ocean proper tend to consist of highly-broken-up ice cover without a good level surface, melt ponds have more difficulty actually forming and staying on the surface, so coverage is limited. By contrast in the Arctic Ocean proper, there tends to be a lot more flat-surfaced ice which is ideal for melt pond formation and retention.

[ChrisReynolds]

http://dosbat.blogspot.co.uk/2014/08/june-sea-ice-area-and-melt-ponding.html

Changes in the June cliff is examined in that post. I find that:

Melt ponds are not the main factor. The marginal zone behind the ice edge plays the largest role.

Loss of ice in other regions e.g. Hudson Bay, plays a non negligible role.

***

Hudson is undergoing what might be the earliest melt in record. So that might see strong anomalous melt in the next three weeks. Chukchi is showing very low compactness, and might be hit by clear skies over the next few days.

However it seems to me that the Siberian coast remains cool. Melt will start in June how rapid I don't know.

Anyway Michael, you hunch is right, the lengthening of the ice edge is a big player.

[DavidR]

It  is also  possible that the geographic boundaries currently restrict the decline in the winter maximum.  The winter maximum is currently only extendable across Bering, Okhotsk, Newfoundland and the North Atlantic.  Each of these boundaries is currently getting shorter as the winter extent boundary  moves north.  So the tipping point for rapid maximum extent declines will not occur until the boundary moves through the Bering Strait and freezing along the Arctic coastlines is no  longer 100%. One that  happens we can expect rapid winter extent  declines.

[seaicesailor]

There are a couple of good alternate explanations for the June cliff:
(1) Melt onset requires some threshold amount of insolation [...] in late May and in June, the North pole is actually the point of maximum insolation under clear skies - so the high insolation sort of hits the whole Arctic "all at once".
(2) In order to have a lot of melt ponds form to create a "cliff", the ice surface must be able to support melt water without allowing it to percolate through, simply flow downhill and off the floe's edge, or otherwise get off the surface. Because areas outside the Arctic Ocean proper tend to consist of highly-broken-up ice cover without a good level surface, melt ponds have more difficulty actually forming and staying on the surface, so coverage is limited. By contrast in the Arctic Ocean proper, there tends to be a lot more flat-surfaced ice which is ideal for melt pond formation and retention.

Melt ponds, being heat traps that cause thinning of the ice in-situ: in the very simplified picture I described, the more melt ponds, the faster the melt front travels. So there is one answer to my question: June cliff is not mostly caused by geography, though this may help. About the draining of the melt ponds, I thought (or heard) all melt pond heat was not lost if cyclically drain out since that means transport of warmer water to underneath the ice. 

http://dosbat.blogspot.co.uk/2014/08/june-sea-ice-area-and-melt-ponding.html
Changes in the June cliff is examined in that post. I find that

Melt ponds are not the main factor. The marginal zone behind the ice edge plays the largest role.
Loss of ice in other regions e.g. Hudson Bay, plays a non negligible role.
***
Anyway Michael, you hunch is right, the lengthening of the ice edge is a big player.

Your blog post is a thorough analysis. I see that the reasons for accelerated melt in June were different depending on what year. I completely forgot about Hudson Bay. And some seas out of Arctic may still be melting during June. Also it appears the "cliff" properly refers to area loss and not extent loss.

My comment is just a very simplified view. The amount of ice behind the ice edge (what I am calling melt front) is directly proportional to the length of the edge. So yes, my hunch is, I think once the melt front is established within the perimeter of the Arctic Ocean, that may help to speed things up. Insolation leading to warmer waters right behind the melt front, poor integrity of the ice, and melt ponds among other factors will be the catalysers, and clearly dictate the speed of propagation.

My name is not Michael 
Anyway, thank you John 

It  is also  possible that the geographic boundaries currently restrict the decline in the winter maximum.  The winter maximum is currently only extendable across Bering, Okhotsk, Newfoundland and the North Atlantic.  Each of these boundaries is currently getting shorter as the winter extent boundary  moves north.  So the tipping point for rapid maximum extent declines will not occur until the boundary moves through the Bering Strait and freezing along the Arctic coastlines is no  longer 100%. One that  happens we can expect rapid winter extent  declines.

I expressed something similar in your thread (plateau hypothesis), warmer global temperature may cause a winter ice edge that lies at higher latitude, may lead to lower maximum values in February-March, but then that edge takes a lot more time to really advance.
This also is a very simplified argument, but simple ideas can help us understand some things, I guezz

[ChrisReynolds]

Sorry Sea Ice Sailor,

I was posting on my phone while on a bus and with limited visibility misremembered you as Michael Hauber. 

Now I am on my laptop and typing isn't such a pain...

The overall loss of area during June is largely contributed to by drop in concentration for regions that start June with high concentration. That is the cause of the drop in compactness in June. But your question relates to the June Cliff, which I take to be the drop in anomalies, or difference from the long term average. That is when that table at the end of my post comes in; and it supports you with 63% of the increase in area loss coming from areas with concentration 0.8 to 0.4, which I estimate to be largely due to the ice edge.

However...

The plot of extent anomalies for the whole Arctic shows sharp drops in some years, or 'cliffs'.



But if I do the same plot for the Arctic Ocean the cliffs are less pronounced.



The difference is largely due to the drops in Hudson Bay.



This year my only reservation is the East Siberian Sea and Laptev. Beaufort, Chukchi, Barents, Hudson are all looking good for rather an exciting June loss. That's as long as we don't get hit with the crummy cold of the last two years...

In re-reading that post it strikes me that the anomaly plots in that post are in terms of extent - why did I do that??? It makes more sense in terms of area. Give me a while and I will re-work for area. Scrub that, they're actually area data, the graphs in my spreadsheet say 'extent' when they're for area data - Doh!

[seaicesailor]

All fine with the name, Chris, I supposed it was just that Speaking of names I am no longer sure if I was talking about what people call "June cliff" after all . . .

Nevertheless, a very long ice edge (or melt front) well established within the Arctic ocean may be an enabler of the June cliff in whatever sense. This front may "burn" all peripheral ice by the end of July if preconditioning (the catalizer ahead of the front) and open water albedo feedback (heat behind the front) establish a high propagation speed.

If the front is not created (if the mix is not sparked, like right now in ESS), combustion does not take place.

I dont know why but this analogy with flames kind of works. Two-states hysteresis, irreversibility, feedback, edge, activation energy . . .

[oren]

I thought the June Cliff in this context was the way the average extent graphs for each decade take a downturn increasing the rate of decline during June. Each decade seems to have this downturn a few days earlier. At least my imagination supplied this explanation when reading the first post.

[seaicesailor]

Yes I was referring to the downturn itself, the extent loss acceleration around June, without regard of it coming earlier or how anomalous. But regardless. I might change the title of the thread.

[ChrisReynolds]

I thought the June Cliff in this context was the way the average extent graphs for each decade take a downturn increasing the rate of decline during June. Each decade seems to have this downturn a few days earlier. At least my imagination supplied this explanation when reading the first post.

I'd presumed the 'cliff' referred to the anomaly plots since it is only in the anomaly that a sharp drop, or 'cliff' is visible. In the plots of area and extent it is just a rather steeper slope.

[seaicesailor]

Since this also pertains to this thread I repost
I attach figures of ice drift predicted for today, Jun 1 &  2 (HYCOM+CICE model). Also the model predicts what can be the opening of a large ice front along the Siberian Coast in one week (bottom image, sorry for the sizes!). Add up the already open front on Beaufort, Chukchi, and Kara.

The speed of this front during June should increase with preconditioning (high in Beaufort and Chukchi, nonexistent yet in ESS and Laptev), compaction, transport (will happen during the first week), insolation of open water, and weather (getting better in Eurasia, worse in America, then who knows).

[Nightvid Cole]

Sorry Sea Ice Sailor,

I was posting on my phone while on a bus and with limited visibility misremembered you as Michael Hauber. 

Now I am on my laptop and typing isn't such a pain...

The overall loss of area during June is largely contributed to by drop in concentration for regions that start June with high concentration. That is the cause of the drop in compactness in June. But your question relates to the June Cliff, which I take to be the drop in anomalies, or difference from the long term average. That is when that table at the end of my post comes in; and it supports you with 63% of the increase in area loss coming from areas with concentration 0.8 to 0.4, which I estimate to be largely due to the ice edge.

However...

The plot of extent anomalies for the whole Arctic shows sharp drops in some years, or 'cliffs'.



But if I do the same plot for the Arctic Ocean the cliffs are less pronounced.



The difference is largely due to the drops in Hudson Bay.



This year my only reservation is the East Siberian Sea and Laptev. Beaufort, Chukchi, Barents, Hudson are all looking good for rather an exciting June loss. That's as long as we don't get hit with the crummy cold of the last two years...

In re-reading that post it strikes me that the anomaly plots in that post are in terms of extent - why did I do that??? It makes more sense in terms of area. Give me a while and I will re-work for area. Scrub that, they're actually area data, the graphs in my spreadsheet say 'extent' when they're for area data - Doh!

It looks like in SOME years, melt ponds are in fact the main cause of the June cliff, and in others, it's the ice breaking up in Hudson Bay.

This raises the question of whether the "Arctic Ocean only" June cliff is a better predictor of September extent than is the June cliff taken on the entire CT domain, since we do not expect peripheral areas like Hudson Bay to matter since they fully melt every year.

[ChrisReynolds]

Nightvid,

Yes, considering only the Arctic Ocean alone would seem reasonable.

[seaicesailor]

An update: HYCOM predicts an almost complete melting front inside the geography of the Arctic by June 15:



Open water or very low concentration all within the perimeter of the Arctic ocean except part of the ESS. It looks like a big dive is on schedule (not use 'cliff' to avoid confusion/controversy)

PS. Granted that CAA and Greenland sections wont have significant open water in their coasts and so no melt front,... , for the time being

[ChrisReynolds]

Anomalies remain level at the moment, but I think it is reasonable to expect a fall in the next week or two.

[seaicesailor]

Update, I was expecting an ESS edge well open, and Laptev too by now. But no. The front did not advance in Beaufort (although that may be bad for the MYI that entered that sea), Chukchi stalled.

So according to my "melting front hunch", it is forming late and may spare ice from a big melting runaway.

The ice seems preconditioned and all ready to go if weather changed drastically though.

I repost here being relevant:

That is gtreat piece of knowledge Bruce, thx.

A few days ago it was also about winds transporting heat from sea sutrface in Chukchi into the Atctic ice, and as LMV showed us several times, that did not happen. Rather the opposite, the region has cooled down a bit. Saved some ice!!

For big July melting two big questions are: will finally ESS and Laptev will open up soon enough?? Once open, they can have a melt runaway and affect CAB with some sun still heating up.
And, will those nice MYI floes in Beaufort and the thick ice in Chukchi melt out? Things have slowed down so much that a really warm windy sunny weather had to come.

You seem to think that the melting of the ice pack proceeds from the outer edges inwards and that peripheral melting causes central Arctic melting.

But this is not the case - the ice melts from the top surface and from the bottom surface. You can indeed have melting on the lateral surface of ice floes, but the lateral melting from the edges of the pack is not what gets you to the September minimum - simple back of the envelope calculations suggest that the melting from action at the edges of the ice pack is orders of magnitude too slow.

In essence, all the melting is local. While you can certainly have short-range effects such as warm water from open areas undercutting ice elsewhere, the speeds of water and ice motion are too slow for the ice cover in the peripheral seas in July to impact the ice near the North pole by September.

You could literally delete all ice in the peripheral areas right now and the impact on the North Pole ice by September would be too small to detect.

All ice melt is local.

I believe that greatest portion of melting happens in the edges advancing inwards. BUT, I believe the speed of this progression is accelerated by prior setup of local melt (mainly surface ponds) . What makes the difference is open ocean being heated within this gradually rotten edge and behind  it, because it causes further top melting (if enabled by atmospheric heat transport, south winds) and bottom melting, especially in July-August. Hence the importance of this melt front being established during June, to really benefit from insolation peak. If later, then the runaway is not so much, or there is no runaway at all. Remember how fast Laptev bite advanced in 2014 (very early establishment of front), and how slow almost inexistent Beaufort advance in 2013, so late!. Note that all these effects are clear manifestation of albedo feedback.

Compacting drift also makes the front advance faster, which clears out even more open ocean water, which again, heats up and make ice melt faster and front advance faster.

I dont think lateral melting becomes important until the last stages of floe melt out.

And I don't agree, not all melting is local in the sense you seem to imply. Part of it, yes, but greater part of it needs substantial ocean open water close by. (My hunch, again)

There is much more in this than what I am describing as a simple conceptual model. compaction, export, divergence, upwelling ...