Support the Arctic Sea Ice Forum and Blog

Poll

If we define Arctic sea ice free as less than one million of square kilometers measured with the official NSIDC sea ice extent (monthly average), when do you believe that the Arctic will be sea ice free?

2013-2016
44 (45.4%)
2017-2020
41 (42.3%)
2021-2030
9 (9.3%)
2031-2040
2 (2.1%)
2041-2060
1 (1%)
2061-2100
0 (0%)
Later than 2100
0 (0%)

Total Members Voted: 96

Voting closed: April 01, 2013, 10:12:01 PM

Author Topic: Arctic sea ice free (extent)  (Read 93148 times)

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #100 on: April 06, 2013, 02:11:10 PM »
Over on Skeptical Science, a moderator pointed out a follow-up on Eisenman & Wettlaufer, 2008

John Wettlaufer and Ian Eisenman have separately authored two follow-on papers to their seminal 2008 study:

Moon, W., and J. S. Wettlaufer. "On the existence of stable seasonally varying Arctic sea ice." arXiv preprint arXiv:1202.5506 (2012).

Eisenman, Ian. "Factors controlling the bifurcation structure of sea ice retreat." Journal of Geophysical Research: Atmospheres (1984–2012) 117.D1 (2012).
Cheers!
Lodger

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #101 on: April 08, 2013, 01:57:42 PM »
I'm in the camp that believes the FIRST ice free Summer will last 30 days. That's not to say the first Summer <1M km2 SIE. But once ALL the ice is gone in the CAB
MORE idle talk unless it's backed with some real science! Here it is: ;)

Eisenman, Ian. "Factors controlling the bifurcation structure of sea ice retreat." Journal of Geophysical Research: Atmospheres (1984–2012) 117.D1 (2012).

A PDF is freely available (yes, this is the same Ian Eisenman who with John Wettlaufer wrote the seminal 2009 paper "Nonlinear threshold behavior during the loss of Arctic sea ice").

Panel C of Figure 8 of Eisenman2012 (attached below) clearly shows that the first time sea ice touches zero, that year remains sea ice free for over 70 days (~20% of a year).  :o

Further, Panel C shows that 1st sea ice free day occurring around Day 240, which is August 28th in a normal year. Freeze-up would not begin until Day 310, or November 6th:-[

Figure 8. Poincaré maps and ~E(~t) trajectories. (a) Poincaré map with default parameter values... (b) Same as Figure 8a but with... initial conditions slightly warmer and colder (c) One-year trajectories of E (Surface enthalpy) associated with the green circles in Figure 8b.

Eisenman2012, page 11, Figure 8:
« Last Edit: April 08, 2013, 03:24:32 PM by Artful Dodger »
Cheers!
Lodger

TerryM

  • First-year ice
  • Posts: 5213
    • View Profile
  • Liked: 395
  • Likes Given: 5
Re: Arctic sea ice free (extent)
« Reply #102 on: April 08, 2013, 07:48:19 PM »
Lodger


Haven't read the paper yet, but could you synopsize what they're postulating would be sufficient to keep things open so late in the year?


I apologize for commenting before reading, but it's one of those days with a lot of things backing up & the subject is fascinating.


Terry

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #103 on: April 08, 2013, 09:55:37 PM »
...could you synopsize what they're postulating would be sufficient to keep things open so late in the year?
yes, latent heat going into the sea ice quickly switches to sensible heat going into the ocean after the last of the sea ice is gone. It takes time to give back all that heat.
Cheers!
Lodger

TerryM

  • First-year ice
  • Posts: 5213
    • View Profile
  • Liked: 395
  • Likes Given: 5
Re: Arctic sea ice free (extent)
« Reply #104 on: April 09, 2013, 05:07:23 AM »
Thanx


I'll try to get the paper read tonight.


I've been very interested in the release of sensible heat when the ice is gone, but suspected that it wouldn't  be a factor until the year after the first melt.


Terry

Richard Rathbone

  • Grease ice
  • Posts: 529
    • View Profile
  • Liked: 36
  • Likes Given: 7
Re: Arctic sea ice free (extent)
« Reply #105 on: April 09, 2013, 06:40:34 PM »
I think it depends how early in the season it hits zero the first time it does. If it just makes it there in late September, its not going to stay ice free long. So if it only goes to zero the first time this year because there was an unusually favourable September for melt, its not going to be ice free for 70 days. However, a couple of years of unfavourable weather, followed by a high rate melt season, and it could easily go ice free at the beginning of August and pick up enough heat to stay that way through October.

If you look at the trend lines for how fast each month is heading to zero, there are several not far behind September. September takes a year longer than trend, or they take a year less, and you could have a 100 day ice free Arctic the first time it happens.

This year I'd say if it happens its probably only for a few days and people will argue about whether it was technically "ice-free" during those days. If it doesn't happen until 2015 or 2016 it could easily be 100.

fishmahboi

  • Guest
Re: Arctic sea ice free (extent)
« Reply #106 on: April 09, 2013, 06:46:14 PM »
Is it possible for the Arctic Ice to his zero possibly during June or July and if not, what conditions would be needed for such an occurrence to ensue?


TerryM

  • First-year ice
  • Posts: 5213
    • View Profile
  • Liked: 395
  • Likes Given: 5
Re: Arctic sea ice free (extent)
« Reply #107 on: April 09, 2013, 07:02:03 PM »
Lodger


They're using a 50 m mixing layer in the model. In 2011 a British group found mixing down to 200 m under FYI. Would the much larger mixing layer act as a damper on seasonal variations even though it would result in higher temperatures at the ice/water interface?


I'd assumed that the first melt out year would be similar to the previous year, but that in the following year sensible heat would be released when there was no more ice to melt.


Terry

ChrisReynolds

  • Nilas ice
  • Posts: 1714
    • View Profile
    • Dosbat
  • Liked: 1
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #108 on: April 09, 2013, 07:57:01 PM »
Is it possible for the Arctic Ice to his zero possibly during June or July and if not, what conditions would be needed for such an occurrence to ensue?

This June/July?

Not at all possible.    

Monthly Average Volumes for 2012.

April 21.73
Loss of 6.16
July 14.70
Loss of 10.33
September 4.36

Therefore it is not going to happen, and you can rest assured that anyone telling you that it will, or even could, doesn't have the vaguest brush with the data or the physics.   

 ::)

fishmahboi

  • Guest
Re: Arctic sea ice free (extent)
« Reply #109 on: April 09, 2013, 08:56:17 PM »
Is it possible for the Arctic Ice to his zero possibly during June or July and if not, what conditions would be needed for such an occurrence to ensue?

This June/July?

Not at all possible.    

Monthly Average Volumes for 2012.

April 21.73
Loss of 6.16
July 14.70
Loss of 10.33
September 4.36

Therefore it is not going to happen, and you can rest assured that anyone telling you that it will, or even could, doesn't have the vaguest brush with the data or the physics.   

 ::)

Would the fracture event not have an effect on that (I probably should have included this in my initial question) in the sense that there is much more open water and the Multi Year Ice on the north of Greenland is quite mobile this year.

ChrisReynolds

  • Nilas ice
  • Posts: 1714
    • View Profile
    • Dosbat
  • Liked: 1
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #110 on: April 09, 2013, 09:29:45 PM »
Between April and July the loss in 2012 was 6.16.

Between July and September the loss in 2012 was 10.33.

Will the fracturing mean that 10330 cubic KILOMETRES of ice just go 'puff'?

Will it f.....

 ;D

anonymous

  • Guest
Re: Arctic sea ice free (extent)
« Reply #111 on: April 10, 2013, 01:20:37 AM »
Will the fracturing mean that 10330 cubic KILOMETRES of ice just go 'puff'?

Chris, don't spoil the game. Actually, I would find it interesting if someone calculates the energy needed to melt 10³km³ ice, and put this in context. I'm sure you don't need sun going supernovae. Last in Greenland a cloud cover with just the right thickness forced a lot of melting. Also, there is enough warm water below 200m to melt all the ice. The question itself is not wrongful, if there is a plausible cause for such a melt I'd like to know, even if it is not much likely to happen.

slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #112 on: April 10, 2013, 02:27:32 AM »
Hi Arcticio,

If you only need plausible then, sure, I will wave my hands...  ;D

The area of the Arctic Ocean is 14 million km^2.

Let's suppose the ice volume is 14,000 km^3 at the start of July, but it is all broken up into chunks and spread throughout the Arctic Ocean, with gaps everywhere exposing water to the sun.

Let's define the effective critical albedo as the fraction of the incident sunlight required to melt all the ice in the month of July.

In mid-summer, the sun shines 24 hours/day at the pole. I seem to recall that the incident sunlight energy/day there is more than at any other place in the world. I don't remember exactly, but lets assume 20 MJ/(day.m^2) x 31 days in July ~ 600 MJ/m^2 in July.

We know the average ice thickness is 14,000 km^3 / 14,000,000 km^2 = 1m. Therefore, we need to melt 1 tonne of ice per m^2.

The latent heat of fusion for ice is 334 J/g = 334 MJ/tonne.

Putting it altogether,

effective critical albedo ~  334 MJ/tonne x 1 tonne/m^2 / 600 MJ/m^2 ~ 0.56.


So, given the roughness of the calculation, it seems that about half the incident sunlight in July  would have to go towards melting the ice in order to melt it all that month.


But then we could move to a more plausible model than that - where the ice has all broken up by the beginning of June and then has both June and July to all melt out. The effective critical albedo in that case might only be about 0.3 - which doesn't look crazy to me if there is a lot of water exposed.

Then we could look at where the ice has all broken up by the beginning of May, where maybe we are down to an effective critical albedo of 0.2...

Then we can wave our hands a bit more about the currents, the wind and layer mixing bringing in extra heat as well, and maybe some extra export through the straits...


So I say the scenario is unlikely but it is not physically impossible for the ice to all melt out by the end of this July, due to it all breaking up into small chunks.




« Last Edit: April 10, 2013, 02:37:54 AM by slow wing »

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #113 on: April 10, 2013, 02:57:57 AM »
Slow Wing,

The actual figure is a little over 500 W/m2 at peak If you use 450, you get 39 megajoules/day. So your argument is on the conservative side.



V

slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #114 on: April 10, 2013, 03:14:58 AM »
 Cheers, Vergent. Wow, that's nearly a factor of 2 higher than I had assumed!  So the scenario becomes less implausible so far as I can tell.

  A question though to people who know about sea ice: is it plausible that it might continue to break up throughout much of the Arctic until it reaches small chunks of typically only a metre, or a few metres, on a side?

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #115 on: April 10, 2013, 03:24:57 AM »
Cheers, Vergent. Wow, that's nearly a factor of 2 higher than I had assumed!  So the scenario becomes less implausible so far as I can tell.

  A question though to people who know about sea ice: is it plausible that it might continue to break up throughout much of the Arctic until it reaches small chunks of typically only a metre, or a few metres, on a side?



It did last year

http://icefloe.net/Aloftcon_Photos/index.php?album=2012

V

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #116 on: April 10, 2013, 07:12:17 AM »
Therefore, we need to melt 1 tonne of ice per m^2... The latent heat of fusion for ice is 334 J/g = 334 MJ/tonne.
Hi slow wing,

Um, no. Sea ice contains brine and air bubbles, both of which decrease the density and latent heat of fusion for sea ice. First year sea ice (FYI) is saltier than multiyear ice (MYI). Winter 2013 has a record high ratio of FYI to MYI.

These are empirical issues. It's helpful to provide a reference when you quote figures. For instance, a trivia Google search finds this paper:

Ohshima, K. I., Mizuta, G., Itoh, M., Fukamachi, Y., Watanabe, T., Nabae, Y., ... & Wakatsuchi, M. (2001). Winter oceanographic conditions in the southwestern part of the Okhotsk Sea and their relation to sea ice. Journal of oceanography, 57(4), 451-460.

Searching within the PDF for "latent heat of fusion for sea ice" finds this, on Page 6:

Quote
the density of sea ice and the latent heat of fusion for sea ice are taken as 900 kg m–3 and 0.276 MJ kg–1, respectively

So the latent heat of fusion of FYI is about 276 MJ/tonne. And since 1 M3 of FYI has a mass of 900 kg, the heat of fusion of 1 cubic meter of sea ice is 248.4 MJ. The figure you use is 34.5% higher. ;)

Note that this study was conducted in the Southwestern part of the Okhotsk sea, which guarantees that the values obtained are for FYI.

BTW, searching the ASI blog is productive. This issue has been discussed extensively. The important metric we call CAPIE is an estimate of the ratio of open water to sea ice within the pack ice. Insolation is discussed there as well.

Wipneus posted this chart and this comment on the ASI blog: (quoted below)



Quote
Here is my version of the graphic (100 days high summer): So indeed in high summer, the north pole receives more energy than a place on the polar circle. The difference for these 100 days is about 7%.

Note: Day 120 is April 30, and Day 219 is Aug 07 in a normal year. This is the 100 day 'top-melt season' centered around the Northern Summer solstice. On the ASI blog, we've come to refer to the lower insolation days of late summer as the 'bottom-melt season'.
« Last Edit: April 10, 2013, 09:52:14 AM by Artful Dodger »
Cheers!
Lodger

fishmahboi

  • Guest
Re: Arctic sea ice free (extent)
« Reply #117 on: April 10, 2013, 09:17:37 AM »
I recognise that it is indeed possible for the Arctic to melt out in June or July depending on the amount of open water and the size of the Arctic Ice as a whole, it cannot survive it there is a large extent, but it mainly consists of broken ice, and thus I have to wonder about how long the Arctic would stay ice free if the ice did melt out as early as June.

I would ask about the impacts, but that's for the consequences subforum.

anonymous

  • Guest
Re: Arctic sea ice free (extent)
« Reply #118 on: April 10, 2013, 09:42:44 AM »
Slow wing, great. Let me throw in data from buoy 2006C, showing bottom melting of ~1m per month happened in summer 2007.


slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #119 on: April 10, 2013, 09:44:56 AM »
Hi Vergent and Dodger,

  Yes Vergent, the state of the ice in that photo should do the trick.

Dodger, thanks for information in the references. It was just a plausibility argument so I didn't worry too much about 10-20% effects. That factor of 2 in the insolation needed to be fixed though.

So, re-running the argument with the factors rounded to 1 significant figure:

Ice is all broken up by the end of May: could it melt out by the end of July? Assume:

1 tonne/m^2 of ice left at end of May (this is about 15,500 km^3 at a density of 0.9 tonnes/m^3)
300 J/g = 300 MJ/tonne = heat of fusion
Insolation is 40 MJ/(day.m^2) (question: does that allow for clouds?)
60 days (OK, is 61)

critical absorbtion fraction to just melt the ice =

300 MJ/tonne x 1 tonne/m^2 / (40 MJ/(day.m^2) x 60 days) = 1/8

So, in this simple model and neglecting other heat sources and sinks...

 If only 1/8th of the solar energy goes into melting the ice then the Arctic Ocean could be ice free by the end of this July.

So the scenario is at least plausible imo.











slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #120 on: April 10, 2013, 09:48:45 AM »
Slow wing, great. Let me throw in data from buoy 2006C, showing bottom melting of ~1m per month happened in summer 2007.
Wow! That's a lot!

Am I right that the general consensus was that about twice the normal amount of heat to melt the ice flowed in through the Bering Strait that year? It could be carnage if it happens again this year.

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #121 on: April 10, 2013, 10:30:48 AM »
Am I right that the general consensus was that about twice the normal amount of heat to melt the ice flowed in through the Bering Strait that year? It could be carnage if it happens again this year.
Hi slow wing,

Yes, twice the normal heat influx in Summer 2007 is about right.

Woodgate, Rebecca A., Tom Weingartner, and Ron Lindsay. "The 2007 Bering Strait oceanic heat flux and anomalous Arctic sea-ice retreat." Geophysical Research Letters 37.1 (2010): L01602.

This level of heat influx seems to be about a once-in-20-years event. Dr. Wieslaw Maslowski of the Naval Postgraduate School provides more data in Table 1 of this paper:

Maslowski, Wieslaw, et al. "The future of arctic sea ice." Annual Review of Earth and Planetary Sciences 40 (2012): 625-654.
Cheers!
Lodger

anonymous

  • Guest
Re: Arctic sea ice free (extent)
« Reply #122 on: April 10, 2013, 11:47:59 AM »
I think two more effects happened in 2007 to the floe carrying buoy 2006C. There was open water, clear sky and mostly southerly wind. The surface water on the luv side was heated first and then driven below the ice. But my pet theory is once all the surface water got transported away it was replaced by upwelling with warm water from below, also then blown under the ice. All three effects together lead to turbo melting, so watch out for long lasting Arctic Dipoles.

ael

  • New ice
  • Posts: 23
    • View Profile
  • Liked: 5
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #123 on: April 10, 2013, 03:44:58 PM »
Um, no. Sea ice contains brine and air bubbles, both of which decrease the density and latent heat of fusion for sea ice.

I am confused how brine would decrease the density of sea ice.  Is not salt water denser than ice?

TerryM

  • First-year ice
  • Posts: 5213
    • View Profile
  • Liked: 395
  • Likes Given: 5
Re: Arctic sea ice free (extent)
« Reply #124 on: April 10, 2013, 07:25:22 PM »
Wow


I had no idea that latent heat for sea ice was different than for fresh ice. Since reading Lodgers post I've done some reading that indicates that not only is latent heat much less with sea ice - one of the figures I got was 246 kJ kg-1 at -2C - but also that there was a 2% difference between freezing and melting latent heat, with freezing being lower than melting.


David N. Thomas, Gerhard S. Dieckmann 2009 - page


Back to the drawing board - with a big eraser.


Terry






crandles

  • Young ice
  • Posts: 2509
    • View Profile
  • Liked: 91
  • Likes Given: 47
Re: Arctic sea ice free (extent)
« Reply #125 on: April 10, 2013, 07:26:22 PM »
I am confused how brine would decrease the density of sea ice.  Is not salt water denser than ice?

If you exclude the weight of the brine but include the volume that it takes up then you end up with a lower density. While that may seem to be not comparing like with like, it is presumably what you would do for working out the latent heat needed to melt the block of ice that contains some brine.

ChrisReynolds

  • Nilas ice
  • Posts: 1714
    • View Profile
    • Dosbat
  • Liked: 1
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #126 on: April 10, 2013, 07:55:06 PM »
Let me repeat.

April 21.73
Loss of 6.16
July 14.70
Loss of 10.33
September 4.36

In other words there would have to be a melt of 267% of ice volume from April to July for July to end up near zero.

Slow Wing,

In June 2012 most of the ice was just under 2m thick, with thinner ice in Barents and Beaufort. Peak volume loss is in July, at a typical 7k km^3 loss between June and July, 12k km^3 loss between August and June. You cannot move that loss forward because it is insolation driven and insolation is fixed. Vergent's figures for insolation are for top of atmosphere insolation, due to the low angle of incidence figures hitting the surface are lower, and due to low angle of incidence albedo is higher due to reflection of oblique rays off the surface.

If you're getting calculated results that show that a loss of 267% of ice volume from April to June is feasible with only 1/8 of insolation your assumptions are wrong. For a start, it comes before the bulk of the region's annual volume loss.

Here's the profile of volumes and losses on the first day of the month for 2012

4, 21.65, 1.757
5, 21.447, -0.203
6, 17.809, -3.638
7, 10.879, -6.93
8, 5.718, -5.161
9, 3.455, -2.263

i.e. for 6 (june) the negative number is the change from 1st May to 1st June.

After July there is a total of ~7.3 thousand cubic kilometers of ice. To get ice free in July you must melt this before July. Bearing in mind that the loss to the end of July from 1 May is ~10.7k km^3.

Frankly, I am staggered I'm even spending time replying to this.

Vergent,

I seriously doubt next year will see a melt out by July for the reasons I've already given. Sea ice in the Arctic basin will still grow to 2m thick over winter even if all the MYI goes, so net volume for the Arctic ocean will still exceed about 20k km^3. Until we see significant warming over winter this will continue to be the case. So against such a gain of volume over winter the question becomes how much can we reasonably expect the ice to loose during the melt season? The answer is that to get a sea ice free Arctic the total loss must be over 20k km^3. Here are the volume losses for the 2000s, but these include ice outside the Arctic Ocean.

2000 16.342
2001 15.574
2002 16.746
2003 17.079
2004 15.93
2005 17.022
2006 16.198
2007 17.407
2008 18.087
2009 18.189
2010 18.974
2011 17.944
2012 18.662

Now the peak volume for the last three years* has been around 21.9k km^3, which gives an indication of the start volume for these losses. *Assuming this year will be around that.

I said above I was staggered I was bothering to reply to this. I've decided it's not a useful expenditure of my time. Go ahead and dream up what you all want, I'll leave the ice to give the answer.

anonymous

  • Guest
Re: Arctic sea ice free (extent)
« Reply #127 on: April 10, 2013, 09:08:24 PM »
Regarding turbo melting, didn't that happen at a regional level in 2011? The famous Laptev Bite? IIRC two weeks with ~20°C offshore wind was the reason. And I think Chris is right: Insolation has a well known upper level. Let's hope the other factors behave like the years before.

ChrisReynolds

  • Nilas ice
  • Posts: 1714
    • View Profile
    • Dosbat
  • Liked: 1
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #128 on: April 10, 2013, 09:16:27 PM »
"Could you use your superior skills with data and physics to estimate how much ice an inrush like this could melt?"

When has such an inrush occurred in reality?

Really you're just chasing ghosts. You want to believe in them fine, it's your mind. But don't expect me to waste my time on it.

Do you know what conformation bias is?

SATire

  • Frazil ice
  • Posts: 496
    • View Profile
  • Liked: 25
  • Likes Given: 5
Re: Arctic sea ice free (extent)
« Reply #129 on: April 10, 2013, 10:08:11 PM »
Dear colleagues,

I would like to ask you to calm down a bit. The topic is interesting, the changes are steep - but please get cool a bit. The worst thing here is to get called "Alarmist" without reason - so reason ist important. Errors are often made, like big ones by myself - but agree, when you missed something important or calculated in stupid way. That is normal and nothing bad - bad is only, if you are the only one not able to see, when you are wrong...

Something in the discussion above is misleading by exaggeration and/or misunderstanding - that could be ruled out easily. E.g. End of July is not 1. July. Ice free July is not ice free after July. Next year July is not end of this years June, a picture of ice condition in Sept. 2012 does not indicate albedo of this years May...

I think the strength of this site is the interdisciplinarity - experts discuss with noobs and both learn and are surprised again and again by daily novel pictures. Please take the comments serious in any case - there is no truth anymore but todays observations. Risks should be rated, but not hyped because that is dangerous for credibility.

I am sorry if I disturbed again. I just love to read all the comments here because of their style without bickering. 


Espen

  • Young ice
  • Posts: 3251
    • View Profile
  • Liked: 177
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #130 on: April 10, 2013, 10:28:18 PM »
SATire;

I appreciate your comments, and hope it will be read by all our colleagues and friends, and yes it is one of very few blogs with important issues without bickering, and that is what is making it so special  and important!!
« Last Edit: April 10, 2013, 10:45:56 PM by Espen »
Have a ice day!

ChrisReynolds

  • Nilas ice
  • Posts: 1714
    • View Profile
    • Dosbat
  • Liked: 1
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #131 on: April 10, 2013, 10:40:05 PM »
No bickering from me SATire, I'm not bothering, not worth my time.

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #132 on: April 11, 2013, 11:01:29 AM »
I am confused how brine would decrease the density of sea ice.  Is not salt water denser than ice?
Hi ael,

Compare pure water ice to brine ice. With pure ice, 100% of the mass is ice. With FYI, perhaps 90% (I'm choosing an arbitrary number here for the explanation) is ice, and 10% is liquid.

Now it still occupies 1 cubic meter of volume, but only 90% of the mass is ice. So the density of the ice is decreased, because it contains water.

It's not that hard really. Just ask yourself why FYI floats. It's less dense than the sea water that formed it.

HTH  ???
Cheers!
Lodger

Wipneus

  • Citizen scientist
  • Young ice
  • Posts: 3996
    • View Profile
    • Arctische Pinguin
  • Liked: 619
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #133 on: April 11, 2013, 11:24:40 AM »
Thanks Lodger, that makes even less sense than before.

So when we are talking about volume of sea ice we include the inclusions of brine.
And when we are talking about the mass of sea ice, we take out the inclusions of brine.

And on another day, we can just reverse and say hey, density increases.

Quote
It's not that hard really. Just ask yourself why FYI floats. It's less dense than the sea water that formed it.

Well excluding the mass of the brine does make the ice float higher than it does. Just ask Archimedes.


Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #134 on: April 11, 2013, 11:30:24 AM »
If it does not melt out this summer, it may melt out in late December!

The Met Office global model had a nightmare last fall. It envisioned about 1 Mkm^3 of warm Atlantic water flowing into the Arctic Basin between -50m and -300m. In December this water began convection to the surface. In early January the pole was ice free. Around the 10th of January they substituted in new historically normal water and ice layers. They have since shut down the model access.

Hi Verge,

You're on to some good stuff in this line, which IIRC has not been previously discussed here or on the ASI blog (pls correct me if i missed it).

Could you tell us where you got the model outputs? Are they still available online? Can you post a link, and perhaps provide some background?

Thanks in advance!  8)

BTW, your calculations are interesting too, but you need to re-examine your units for volume, area, and thickness. Showing your work is always appreciated. Clearly the model run ran the numbers, but we should be able to calculate the mean easily enough by hand!
Cheers!
Lodger

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #135 on: April 11, 2013, 05:28:19 PM »
Lodger,



All estimates, approximations, absolutes, and global negatives are, by definition, false(including this one).

V

edit

Yes, that is a white crow.
« Last Edit: April 11, 2013, 05:33:50 PM by Vergent »

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #136 on: April 11, 2013, 05:53:57 PM »


Our mighty MYI. As last seen.

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #137 on: April 11, 2013, 07:51:00 PM »

Hi Verge,


Could you tell us where you got the model outputs? Are they still available online? Can you post a link, and perhaps provide some background?

Thanks in advance!  8)

BTW, your calculations are interesting too, but you need to re-examine your units for volume, area, and thickness. Showing your work is always appreciated. Clearly the model run ran the numbers, but we should be able to calculate the mean easily enough by hand!

http://data.ncof.co.uk:8080/ncWMS/godiva2.html

The link broke down in late January. Totally bummed me out. It was like losing google earth.

I stopped saving screenshots when I got banned for impiety at AmWx. The most recent I have was Dec. 1413(saved on 12/14).

DISCLAIMER: This did not happen, this is a model that diverged with reality.
 
« Last Edit: April 11, 2013, 08:12:29 PM by Vergent »

Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #138 on: April 12, 2013, 09:38:25 AM »
Statistician George E P Box, in "Science and statistics", Journal of the
American Statistical Association 71:791-799
Quote
All models are wrong, but some are useful
Hi Verge,

That's why i don't hang out on denier sites, there's nothing there to learn ;)

Your model 'screenies' are more entertaining than a Charlie Sheen meltdown, and twice as eye-opening. This is how i envision the final collapse of the Arctic.

BTW, do you have any associated depth/salinity profiles for the big event? I'm curious what the model thought happens to the halocline.

"Impiety"? That's when you ate too much PI, wot?

Laters! and have a good'un!

« Last Edit: April 12, 2013, 09:45:42 AM by Artful Dodger »
Cheers!
Lodger

Vergent

  • Grease ice
  • Posts: 573
    • View Profile
  • Liked: 2
  • Likes Given: 1
Re: Arctic sea ice free (extent)
« Reply #139 on: April 12, 2013, 03:38:14 PM »
Lodger,


Quote
The background color shading depicts a zonally averaged density profile from observational data.
http://www.eoearth.org/article/The_Potential_for_Abrupt_Change_in_the_Atlantic_Meridional_Overturning_Circulation

I think depth/density is what is needed. The Fram outflow is limited to the in flow, which is restricted. Take away the restriction on inflow... Well its not much more complicated than a toilet flushing. Density seeks its level. 

The arctic basin has an area of 4.2Mkm2(CT). the dense depths are from -500m down. The Fram is 2,500m deep. So there is about 8.4Mkm3 poised to flow out.

V
« Last Edit: April 12, 2013, 04:42:44 PM by Vergent »

johnm33

  • Nilas ice
  • Posts: 1282
    • View Profile
  • Liked: 80
  • Likes Given: 40
Re: Arctic sea ice free (extent)
« Reply #140 on: April 12, 2013, 05:09:15 PM »
Another scenario for a complete meltdown is imo strongly related to the cracks.  Last year a large anomoly showed up in Mackenzie bay, at its most extreme IIRC  there were 12 bands each indicating 2deg. implying 24deg.C on the actual temperature map. This was the implied temperature of the ocean closest to the shore, although it wasn't sustained at this level for more than a few days the anomoly persisted past the date of the max. The only other feature that showed any similarity to this was the anomoly on the western shore of greenland, it wasn't so extreme but it too had a gradient, was warmest by the coast and isolated from any similarly warm water.This seems to be best explained by the restratification of waters by eddy kinetic energy, caused by the currents returning north having traversed the east coast of greenland.  A number of papers are available search EKE, eddy kinetic energy, labrador sea, I've previously linked a couple on ASIB.
 The only likely supply of kinetic energy to effect the anomoly in Mackenzie bay is the Pacific. Looking at the tidal ranges at Point Barrow and Tin City http://www.americantides.com/ It's clear that Arctic tides [.4m] have about a third the range of the Pacific [1.1m] ones in this area, implying that whenever high tides occur in the north Pacific an extended pulse of Pacific water will enter the arctic. If you look further south in Alaska http://www.americantides.com/tide-predictions/black-rock-walrus-islands-alaska where the range is more than 4m some sense of the inertia in this dynamic can be grasped.
My contention is that once this "Pacific" water passes through the strait, it is lost to the pacific due to the diference in rotational speed between its origin and where it now finds itself. Going from 60N to 75N effectively halves the distance to the rotational axis and the local speed of the earths surface, a large fraction of that energy is posessed by the inflowing water which would be more energetically comfortable further south, but is compelled eastwards by the excess of kinetic energy carried with it through the strait.
Although this is the warmest/least saline of Pacific water it is nevertheless denser than the ice/fresh water floating atop the Arctic ocean and consequently dives below it. My guess is that before 2007 what it found there was an endless inverted landscape of ponds and small freshwater lakes formed by the mechanically impacted ice, and whatever damage/melt it caused was confined to these or merely spilled into the next [inverted] pond, and should this fresh water spill over and reach the surface that too was largely contained within the arctic by the extensive permanent ice, like an inverted freshwater sea contained by an ice mountain  range. Ready for the next freeze.It may have happened before but I think there was a breakthrough for that fresher water towards the end of July last year.
You'll have to form your own impression of the action of EKE but my veiw is that  as it rolls like a tidal wave across the inverted landscape, the 'surf' before it is the heat both of its own motion and that carried to the surface from the warm saline Atlantic waters at depth, and just like a wave when it meets an obstacle it waits until the trough is filled or the barrier overcome before  proceeding. Thus wherever a gap appears in the ice warm water will be brought to the surface and wherever the ice changes thickness a similar build up will enhance the probability of cracks forming as the wave swells before moving on. In time however the landscape will erode to a much leveller plane. Whilst this process has been happening since at latest 2007 I think the cracking this year indicates a more 'mature' inverted landscape beneath the ice, and the likelyhood of all the bottom/meltwater with nowhere to hide, being shunted out of Fram or through the archipelago by the energetically charged influx coming twice a day from the Pacific.
All the above is compounded by the fact that last year over vast areas the air temps were too high for ice to form but cold enough for snow to settle, yet it appeared this snow was being counted as ice in some of the models, well over the winter no doubt some of it became ice of sorts, probably the sort easily melted by increasingly unimpeded Pacific water on its way to Fram/Baffin.
 Just two last points, prior to 2007 I'm sure the kinetic energy was expended before it reached across the arctic, after the breakthrough I expect the flow slowly accelerated, just as it will this year, if theres anything at all in this, and as it does the ice disappears never mind the weather.
If this is indeed what's happening, and i think it is even though what evidence i see is somewhat ambiguous, we need a north wind through Bering because without it the fresh water lens will disappear south leaving nothing for the Atlantic water to dive beneath and consequently nothing to stop the sudden evacuation of the extreme saline basal waters through Frams basement etc.etc.
« Last Edit: April 14, 2013, 01:45:19 AM by johnm33 »

PhilGChapman

  • New ice
  • Posts: 12
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #141 on: April 12, 2013, 05:11:39 PM »
I haven't seen it mentioned; if it has been and I missed it apologies.

A February 2013 paper in Geophysical Research Letters by Overland and Wang "When will the Summer Arctic be Nearly Sea Ice Free?", updating their earlier work (that I know has been mentioned), compared what they categorise as the three approaches to answering the question. I've uploaded the full paper here for anyone interested.

Jim Williams

  • Frazil ice
  • Posts: 398
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic sea ice free (extent)
« Reply #142 on: April 12, 2013, 05:55:30 PM »
I haven't seen it mentioned; if it has been and I missed it apologies.

A February 2013 paper in Geophysical Research Letters by Overland and Wang "When will the Summer Arctic be Nearly Sea Ice Free?", updating their earlier work (that I know has been mentioned), compared what they categorise as the three approaches to answering the question. I've uploaded the full paper here for anyone interested.

I was doing fine until I realized it was an apologia for modelling.  I will agree that EVENTUALLY modelling is the answer.  The problem is that eventually may not be at all soon.

slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #143 on: April 13, 2013, 05:29:08 AM »
...
Slow Wing,
...Vergent's figures for insolation are for top of atmosphere insolation, due to the low angle of incidence figures hitting the surface are lower, and due to low angle of incidence albedo is higher due to reflection of oblique rays off the surface...
Hi Chris,

Thanks for your comments and particularly the point quoted above.

It makes sense to look up the actual Summer insolation at the North Pole in Summer.
The appropriate insolation values can be obtained from https://eosweb.larc.nasa.gov/:

Monthly Averaged Insolation Incident On A Horizontal Surface (kWh/m2/day)
Lat 90
Lon 0   
22-year Average   
Jan 0.00
Feb 0.00
Mar 0.17
Apr 2.93
May 6.78, x 3.6 MJ/kWh = 24.4  MJ/(day.m^2)
Jun 7.49, x 3.6 MJ/kWh = 27.0  MJ/(day.m^2)
Jul 6.70, x 3.6 MJ/kWh = 24.1  MJ/(day.m^2)
Aug 4.03
Sep 0.82
Oct 0.00
Nov 0.00
Dec 0.00


In my first hand-waving post I simply guessed mid-Summer insolation of 20 MJ/(day.m^2),  x 1 kWh/3.6 MJ = 5.6  kWh/(day.m^2). This is seen to be a bit less than the measured value at the North Pole, but actually fairly close when averaged over the Arctic Ocean - try out a few different latitudes and longitudes on that website above. So it was a reasonable guess.



 In the second post I doubled the value based on Vergent's number, which turned out to be for the top of the atmosphere. That was a mistake and too optimistic. Going back to my original guess of 20 MJ/(day.m^2) gives...


critical absorbtion fraction to just melt the ice =

300 MJ/tonne x 1 tonne/m^2 / (20 MJ/(day.m^2) x 60 days) = 1/4

So, in this simple model and neglecting other heat sources and sinks...

 If 1/4 of the direct solar energy goes into melting the ice then the Arctic Ocean could be ice free by the end of this July.


That much is now pushing it but the scenario is still at least plausible imo if the ice has already broken up some more - which is all that I was trying to address.


Artful Dodger

  • Frazil ice
  • Posts: 453
  • The traps have got him, and that's all about it!
    • View Profile
  • Liked: 2
  • Likes Given: 72
Re: Arctic sea ice free (extent)
« Reply #144 on: April 13, 2013, 11:41:08 AM »
critical absorbtion fraction to just melt the ice =

300 MJ/tonne x 1 tonne/m^2 / (20 MJ/(day.m^2) x 60 days) = 1/4

So, in this simple model and neglecting other heat sources and sinks...

If 1/4 of the direct solar energy goes into melting the ice then the Arctic Ocean could be ice free by the end of this July.
Hi slow wing,

You're getting closer with the math. Several points:
  • you're still neglecting sea ice thickness, your calc above implies about 1.11 m average thickness for the whole Arctic ( 1000 kg m-2 / 900 kg / m-3). What does PIOMAS tell you about average thickness (initial conditions)? What is the thickness distribution (ice class)? You need to account for these (ie: 3 classes of ice would require an equation with 3 terms).
  • Saying "1/4 of solar energy goes into melting" implies that CAPIE averages around 0.75 over your specified 60 day period*. Because of the albedo flip effect described by CAPIE, that assumption doesn't work well. Instead, there is a more accurate way to account for this effect (see 'Solar Gain' below)
Last year, CAPIE reached 0.75 on about day 177 (or Jun 25, 2012 and about 5 days after peak insolation). CAPIE bottomed out at around 0.57 from day 214 to 222 (Aug 1 to Aug 9, 2012)



The difference in solar heat absorbed each day as reflective ice turns to dark water means that the melt curve is asymmetrical about the solstice. Even though there is less insolation after the solstice, the peak in heat absorbed didn't come last year until about day 200 (July 18, 2012 or fully 4 weeks after the solstice)! That's the Peak of Summer for the Arctic ocean. :o

I created a chart in Excel last year for the ASI blog showing solar melt potential as the relationship between insolation vs CAPIE, a little metric I dubbed "Solar gain":



* Note: the actual fraction of solar heat absorbed is slightly different than 1 - CAPIE. The albedo of open sea water is approx 0.07 while sea ice albedo varies from 0.8 for dry thick sea ice down to 0.6 for wet thinning sea ice as the melt season progresses ( constant of 0.7 used as approximation). Thus, the solar heat absorbed fraction in the pack ice region as used in the chart above is calculated like this:  8)

( total absorption ) =
( Ice absorption ) + ( Water absorption ) =
((1- Ice albedo) x (Ice fraction)) + ((1 - Water albedo) x Water fraction)) =
((1-0.7) x (CAPIE)) + ((1-0.07) x (1-CAPIE)) =
( 0.3 x CAPIE ) + ( 0.93 x (1-CAPIE)) =
( 0.3 CAPIE + 0.93 - 0.93 CAPIE)) =
( -0.63 CAPIE + 0.93 )

Note that there is a linear relationship between CAPIE and solar heat absorbed within the main ice pack (always a mix of ice and water). Here's an important milestone for the coming melt season: when CAPIE equals 68% then about 50% of the solar energy reaching the surface is absorbed by the pack ice.  :o

Using chart values for top melt (due to insolation), and given daily sea ice area and total volume melt from PIOMAS history, you should then be able to solve for 'bottom melt' (due to heat from external ocean currents). We got SkyNet by the balls now, don't we? Let's book! ;)

« Last Edit: April 15, 2013, 06:54:57 PM by Artful Dodger »
Cheers!
Lodger

slow wing

  • Grease ice
  • Posts: 792
    • View Profile
  • Liked: 145
  • Likes Given: 398
Re: Arctic sea ice free (extent)
« Reply #145 on: April 14, 2013, 12:08:12 PM »
Hi Lodger

  Thanks for the information. Yes, the simple model is too optimistic in assuming that the ice is evenly spread throughout the Arctic Ocean as a mish mash of ice chunks and water gaps. In reality this Summer, there are going to be areas of thicker and thinner ice and also large areas of exposed water where the ice has melted out completely.

That is why the CAPIE value is too optimistic for this prediction. In regions where there is no ice nearby, the sun's energy cannot melt ice even if it is absorbed and heating up the water. And where the ice is thickest, it won't melt out by the end of July even if 25% of the insolation in that region goes into melting the ice.

  Stepping away from the July question for a moment, it is noted that the ice cover this year is more even than it was last year. Although the volume is the same for March according to PIOMAS, there is less ice thicker than 3 metres and on the other hand the ice is not so thin as last year on the Russian side. That more even spread appears to favour more melting this year than last as it is closer to the situation in the simple model.



johnm33

  • Nilas ice
  • Posts: 1282
    • View Profile
  • Liked: 80
  • Likes Given: 40
Re: Arctic sea ice free (extent)
« Reply #146 on: April 15, 2013, 12:35:12 PM »
Further to my comment above, whilst the cracks, upwelling heat between them and the transport of warmer/fresher water through to Baffin are all implied, and may have other causes, further signs are anticipated. The disconnection of the ice from the coast, at the grounding line, should occur in Mackenzie bay on Axel Heiberg and the north of greenland in Lincoln. A repitition of the Mackenzie anomoly should be followed by similar anomolies in both the above mentioned areas, unless there's a massive breakthrough to Baffin at NWP and through the archipelago generally, sufficient to allow the dissipation of the Pacific waters south, this would lead to a large anomoly of sea temps through the islands and in northern Baffin.

Juan C. García

  • Nilas ice
  • Posts: 1494
    • View Profile
  • Liked: 501
  • Likes Given: 503
Re: Arctic sea ice free (extent)
« Reply #147 on: June 19, 2013, 09:12:01 PM »
NSIDC announces a change on the baseline climatological period for Arctic Sea Ice News and Analysis: from 1979-2000 to a 30-year reference period, from 1981 to 2010:

Quote
This July, NSIDC plans to change the baseline climatological period for Arctic Sea Ice News and Analysis and the Sea Ice Index, the data set we use for our sea ice analysis. We are making this change to match the comparison time frames used by other climate research.

Until now, we have used the 22-year period 1979 to 2000 when comparing current sea ice extent to past conditions. When NSIDC first began to monitor and analyze sea ice extent, a longer period was not available. Since the satellite record is now extended, we are choosing to move to a more standard 30-year reference period, from 1981 to 2010.

http://nsidc.org/arcticseaicenews/2013/06/updating-the-sea-ice-baseline/

Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.

ghoti

  • Grease ice
  • Posts: 766
    • View Profile
  • Liked: 12
  • Likes Given: 15
Re: Arctic sea ice free (extent)
« Reply #148 on: June 19, 2013, 10:34:39 PM »
Which makes anomalies seem smaller and less frightening. "A new normal"

Juan C. García

  • Nilas ice
  • Posts: 1494
    • View Profile
  • Liked: 501
  • Likes Given: 503
Re: Arctic sea ice free (extent)
« Reply #149 on: June 20, 2013, 12:14:20 AM »
That's right. The new standard will have lower values. But I believe that the difference will not be so much, in regards to the lower values of 2005-2012.
Which is the best answer to Sep-2012 ASI lost (compared to 1979-2000)?
50% [NSIDC Extent] or
73% [PIOMAS Volume]

Volume is harder to measure than extent, but 3-dimensional space is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC trends [based on extent] underestimate the real speed of ASI lost.