Support the Arctic Sea Ice Forum and Blog

Author Topic: The Summer Acceleration  (Read 8122 times)

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
The Summer Acceleration
« on: May 23, 2013, 09:06:16 PM »
I've just blogged about the summer acceleration of sea ice area.
http://dosbat.blogspot.co.uk/2013/05/summer-acceleration.html

This summer acceleration is that summer decline in area has been faster than winter decline in area in recent years.



Basically I argue that before this decline started both summer and winter were declining at similar rates. So why is summer declining faster than winter?



This is seen by comparing the decline in ice volume of PIOMAS grid boxes over 2m thick with the CT Area daily minimum, both curves track each other closely.

It is the decline in volume that is driving the decline in area. This has significance because April average thickness is now below 2m.



And it is around 2m that the percentage of melt to open water as a function of April thickness begins to increase rapidly.



We will see increasing volatility in summer melts in the years to come. And unless autumn growth of ice tempers the loss of April thickness, we will see this volatility manifest itself as a series of crashes.
« Last Edit: May 23, 2013, 09:41:58 PM by Neven »

Pmt111500

  • Guest
Re: The Summer Acceleration
« Reply #1 on: May 24, 2013, 06:22:12 AM »
So simplisticallly we might assume that inverting the average (gompertz-looking) line on the thickness loss graph we'd get that once there's an 80 % loss in area from the maximum of 5Mkm2 = 1Mkm2 the melt should slow down somewhat because FYI will form most of the ice that stays through summer? Posted a link to your blog on my facebook wall. Not sure if I get to it but this looks like a fine example of citizen science and I might want to reblog the images (and some text) at least.

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #2 on: May 24, 2013, 07:39:03 AM »
Quote
So simplisticallly we might assume that inverting the average (gompertz-looking) line on the thickness loss graph we'd get that once there's an 80 % loss in area from the maximum of 5Mkm2 = 1Mkm2 the melt should slow down somewhat because FYI will form most of the ice that stays through summer?

Actually I don't think so. The profile above is calculated from historic simulated (PIOMAS) ice state, so some thin ice is being protected by a bulk of thicker ice. If all the pack were to be less than 1m thick in April, I have no doubt it would all melt out.

Bob Wallace

  • Young ice
  • Posts: 3855
    • View Profile
  • Liked: 41
  • Likes Given: 5
Re: The Summer Acceleration
« Reply #3 on: May 24, 2013, 08:10:06 AM »
And all that FYI is going to be salty ice.  It will start melting sooner and continue to melt when temps drop to 0.

TerryM

  • First-year ice
  • Posts: 6002
    • View Profile
  • Liked: 893
  • Likes Given: 5
Re: The Summer Acceleration
« Reply #4 on: May 24, 2013, 12:38:00 PM »
FYI also stirs the top 200 meters of the ocean because of the high brine content bringing warmer water to the surface to melt more ice.


Terry

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #5 on: May 25, 2013, 07:56:13 AM »
Bob,

Quote
And all that FYI is going to be salty ice.  It will start melting sooner and continue to melt when temps drop to 0.

It is also flatter than when the pack was mainly MYI, so it will generate and keep melt ponds for longer, which will lower albedo.

Terry,

Quote
FYI also stirs the top 200 meters of the ocean because of the high brine content bringing warmer water to the surface to melt more ice.

Surely MYI melt will be more disruptive as it's is less salty. Although I suspect that both are so much less salty than sea water that the difference isn't great. What counts more is the greater volume of spring melt in recent years which dumps more fresh water into the ocean.

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: The Summer Acceleration
« Reply #6 on: May 25, 2013, 03:23:13 PM »
What's the average thickness of the thick category in April?

The ratio between the April volume and September area is 5m which ought to be that average thickness but I'd have guessed it at 3 to 4 rather than 5m

Shared Humanity

  • Guest
Re: The Summer Acceleration
« Reply #7 on: May 25, 2013, 03:38:45 PM »
SIE has been holding up fairly well as SIA loss accelerates. Isn't this unusual? Doesn't SIE loss usually occur in front of SIA loss?

Whit

  • New ice
  • Posts: 37
  • Standing in a river, waving a stick
    • View Profile
  • Liked: 0
  • Likes Given: 4
Re: The Summer Acceleration
« Reply #8 on: May 25, 2013, 04:03:17 PM »
I really wish we had more information about what is going on below the ice, and the state of submerged ice.

Like Wayne has showed there is an active melt-refreeze zone of porous ice. The more porous the ice is, the higher the effective area vs. volume is. And the higher the effective area vs. volume heat is allowed to work on, the faster the melt will be. If the ice is thin enough this porous zone will also contribute through decreased albedo.

From this laymans perspective the low quality of the 2012 ice could be a significant part of the explanation behind the rapid and extensive reduction last season.

If the quality of the ice is worse than last season 2013 might prove to be even more of a surprise than last year, especially if you just consider volume and area, without considering the quality of the ice itself.

If some of that warm atlantic water upwelled during the winter (and it probably did in Chukchi at least) we could have huge areas of porous low-quality ice, primed for accelerated melt when insolation starts doing it's thing.

I know most of this is pure speculation, but I just can't get past the similarities between the way a high alpine lake behaves at ice-off and the way the Arctic sea ice has behaved the last 5 years.

I just skimmed Ice Tank Experiments Highlight Changes in Sea Ice Types. Does anyone know if a similar experiment has been done in reverse?
Is it progress if a cannibal eats with a fork?

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #9 on: May 25, 2013, 04:32:21 PM »
What's the average thickness of the thick category in April?

The ratio between the April volume and September area is 5m which ought to be that average thickness but I'd have guessed it at 3 to 4 rather than 5m

I'm a bit confused by your question Richard.

April vol and Sept area will give unreal results because ones at max vol and ones at min area. If you don't have Excel there's breakdown data here:
http://dosbat.blogspot.co.uk/p/piomas-olume-breakdown-by-grid-cell.html
That's 15cm thickness bands broken down into volume contribution from each thickness band for each month 1978 to Mar 2013. There's also a spreadsheet available from the owning page for that (see link on that page).

With regards dividing PIOMAS volume by CT Area; Over the peak months when concentration is high the error is a few percent. But by summer the error is as great as 30% due to lower concentration of sea ice.

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #10 on: May 25, 2013, 04:35:40 PM »
SIE has been holding up fairly well as SIA loss accelerates. Isn't this unusual? Doesn't SIE loss usually occur in front of SIA loss?

No it's what is expected.

Sea ice area is the total area of each of the grid boxes taking into account the percentage ice cover in each grid box.

Sea ice extent is the total area of each of the grid boxes for which the percentage of ice cover is more than 15%.

So as concentration at the end of summer goes down (the pack becomes more fragmented), area goes down, but extent doesn't go down as much.

DaveHitz

  • New ice
  • Posts: 31
    • View Profile
  • Liked: 9
  • Likes Given: 1
Re: The Summer Acceleration
« Reply #11 on: May 25, 2013, 05:32:50 PM »
Quote
Surely MYI melt will be more disruptive as it's is less salty.

Chris,

I believe the FYI has brine pockets inside of it. The brine is dense. When FYI melts, the brine drops through the water causing mixing as it descends. (Potentially bringing up heat from below?) The MYI doesn't contain brine, or not as much, so it doesn't create this mixing effect.

Can't remember where I read about this. Either here or somewhere linked to by here, no doubt.

Dave

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #12 on: May 25, 2013, 05:49:43 PM »
Dave,

You are right. But what I'm saying is that I suspect the difference in salinity between both MYI and FYI melt and the sea water below is much greater than the difference in salinity between MYI and FYI.

So the greatest factor would be melt of a greater volume of ice rather than the MYI/FYI balance. That's for recent years as compared to climatology (for which I use 1980 to 1999).

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: The Summer Acceleration
« Reply #13 on: May 25, 2013, 10:53:27 PM »
What's the average thickness of the thick category in April?

The ratio between the April volume and September area is 5m which ought to be that average thickness but I'd have guessed it at 3 to 4 rather than 5m

I'm a bit confused by your question Richard.

April vol and Sept area will give unreal results because ones at max vol and ones at min area. If you don't have Excel there's breakdown data here:
http://dosbat.blogspot.co.uk/p/piomas-olume-breakdown-by-grid-cell.html
That's 15cm thickness bands broken down into volume contribution from each thickness band for each month 1978 to Mar 2013. There's also a spreadsheet available from the owning page for that (see link on that page).

With regards dividing PIOMAS volume by CT Area; Over the peak months when concentration is high the error is a few percent. But by summer the error is as great as 30% due to lower concentration of sea ice.

If the area left in September is the thick ice from April, then the area in September should be the same as the area of the thick ice in April, and from your graph its April thickness would be 5m.

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: The Summer Acceleration
« Reply #14 on: May 26, 2013, 08:22:28 AM »
I see what you mean, thickness=10/2, but a lot of the area of >2m melts out come the summer. You'll find a series of plots of sea ice thickness from PIOMAS here.
https://drive.google.com/folderview?id=0B3pB-kdzoLU3bENnNkxPZkRGLWM&usp=sharing
Let me know if it doesn't work.

You'll also note that in recent years the blue line in my final plot of the lead post shows that more than average of thicker ice melts out during the summer.