Support the Arctic Sea Ice Forum and Blog

Poll

When will the Arctic Extent dip below 1,000,000 Km^2

2018-2019
12 (17.9%)
2020-2025
21 (31.3%)
2026-2030
13 (19.4%)
2031-2040
15 (22.4%)
2041-2060
2 (3%)
2061-2080
0 (0%)
2081-2099
1 (1.5%)
2100-beyond
3 (4.5%)

Total Members Voted: 61

Voting closed: July 27, 2018, 07:46:32 AM

Author Topic: When will the Arctic Go Ice Free?  (Read 494787 times)

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #150 on: September 08, 2018, 02:27:19 PM »
I believe (and correct me if I am wrong) that this exercise is just curve fitting to the best possible line.  What we lack (me included) is an explanation as to why any particular fit should be chosen.  Oftentimes, a linear fit can approximate a particular data curve over a specified range.  As the data approaches zero, I would expect this to fail, although in which direction is debatable.  In an unknown situation, oftentimes a moving average can best define the trend, as it incorporates recent data, without bias.  Not that it is any more accurate, but it tends to smooth out the data, removing variations which may mislead the eye.  Lastly, extrapolated beyond known parameters is always highly speculative, as we do not know how that affects the environment.  In short, it is a guess.  But we strive to make it the most educated best possible, and your guess may be different than mine, depending on which parameters we each feel might predominate in a future scenario.

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: When will the Arctic Go Ice Free?
« Reply #151 on: September 08, 2018, 04:00:12 PM »
I believe (and correct me if I am wrong) that this exercise is just curve fitting to the best possible line.  What we lack (me included) is an explanation as to why any particular fit should be chosen.  Oftentimes, a linear fit can approximate a particular data curve over a specified range.  As the data approaches zero, I would expect this to fail, although in which direction is debatable.  In an unknown situation, oftentimes a moving average can best define the trend, as it incorporates recent data, without bias.  Not that it is any more accurate, but it tends to smooth out the data, removing variations which may mislead the eye.  Lastly, extrapolated beyond known parameters is always highly speculative, as we do not know how that affects the environment.  In short, it is a guess.  But we strive to make it the most educated best possible, and your guess may be different than mine, depending on which parameters we each feel might predominate in a future scenario.

Yes it is highly speculative.

>What we lack (me included) is an explanation as to why any particular fit should be chosen.

I try to refer to the models and if they almost all show a Gompertz like shape, then why would you use a different shaped curve to fit the data? A better fit might be one reason but if this gompertz shape does pretty well at reducing the RMSE that seems additional reason to go with it.

Use too many parameters and you can get a better fit and send the extrapolation off in any direction you choose. Hence it is necessary to try to minimise the number of parameters used. This can be considered to be a form of Occam's razor - if you don't need extra complexity to explain the data then that extra complexity is likely just wrong and should be omitted.

These considerations (particularly considering the physics which is what the models do) tends to act to place some limits on where the extrapolation goes.

Yes, it is highly speculative and the more so the further you go away from the known data. But at least it is based on something. If the alternative is making stuff up off the top of your head, and the people doing that sort of thing tend to be concerned about the issue and therefore likely to give a biased view towards catastrophic effects in order to motivate action,.... guess which I think should be preferred?

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #152 on: October 04, 2018, 09:11:24 PM »
It is time for the monthly update of my extrapolation when the extent [Extent], volume [Volumen] and thickness [Dicke] will reach zero. The extrapolation occured linearly and by a logarithmic function; the latter one almost constantly resulting in earlier times. September value now includes 2018, and BOE seems to appear later than extrapolated last year, mostly due to a slight increase in volume 2018 compared to 2017.
See attached table.
It is too late just to be concerned about Climate Change

kassy

  • First-year ice
  • Posts: 8235
    • View Profile
  • Liked: 2041
  • Likes Given: 1986
Re: When will the Arctic Go Ice Free?
« Reply #153 on: October 05, 2018, 01:49:27 PM »
Are any of these three measures ´better´ then the other two?

We measure extent directly (and then express it in a SIE way) while thickness and volume are further model extrapolations?

Just curious about this since the extrapolated dates diverge. The first to hit zero in the real world would mean that the others are then zero too.
Þetta minnismerki er til vitnis um að við vitum hvað er að gerast og hvað þarf að gera. Aðeins þú veist hvort við gerðum eitthvað.

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #154 on: October 05, 2018, 03:15:16 PM »
The real value to look at is the volume. If you imagine a block of ice 1km*1km wide and 1 m thick, melting on its edges, say 1m/year, will almost not change its area and you need 500 years until it is gone. But if you melt 10 cm/year (only 10% of the edges' value) from below at the same time, it only needs 10 years to be completely gone. This is the "secret" behind the diverge of the data.
It is too late just to be concerned about Climate Change

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #155 on: October 05, 2018, 05:03:10 PM »
The real value to look at is the volume. If you imagine a block of ice 1km*1km wide and 1 m thick, melting on its edges, say 1m/year, will almost not change its area and you need 500 years until it is gone. But if you melt 10 cm/year (only 10% of the edges' value) from below at the same time, it only needs 10 years to be completely gone. This is the "secret" behind the diverge of the data.

Not sure I agree with this logic.  Melting is really only occurring in two dimensions; above from sunlight and below from sea water.  Hence, extent seems to better reflect real world conditions. 

Pmt111500

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #156 on: October 05, 2018, 05:28:13 PM »
The real value to look at is the volume. If you imagine a block of ice 1km*1km wide and 1 m thick, melting on its edges, say 1m/year, will almost not change its area and you need 500 years until it is gone. But if you melt 10 cm/year (only 10% of the edges' value) from below at the same time, it only needs 10 years to be completely gone. This is the "secret" behind the diverge of the data.

Not sure I agree with this logic.  Melting is really only occurring in two dimensions; above from sunlight and below from sea water.  Hence, extent seems to better reflect real world conditions.
Well, ice conducts heat like any solids.

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #157 on: October 05, 2018, 07:10:06 PM »
My explanation was not to describe exactly what happens when an ice floe on sea water is melting but to illustrate the much more important volume decrease in comparison to an extent (or area) decrease, when it comes to define which one of the three measures (area/extent, volume or thickness) plays the major role in the time until the first BOE occurs.
It is too late just to be concerned about Climate Change

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #158 on: October 05, 2018, 08:20:05 PM »
My explanation was not to describe exactly what happens when an ice floe on sea water is melting but to illustrate the much more important volume decrease in comparison to an extent (or area) decrease, when it comes to define which one of the three measures (area/extent, volume or thickness) plays the major role in the time until the first BOE occurs.

Yes, but if you examine the charts, the first dimension to fail is thickness, which reaches a BOE in 2053.  The combined length and width does not occur until 2073.  How could volume decline to zero, before any of the dimensions?  Mathematically, volume will always decrease faster than any one- or two-dimensional measurement, and then slow as less volume is present.  Hence, I question the use of volume to make predictions.

oren

  • Moderator
  • First-year ice
  • Posts: 9805
    • View Profile
  • Liked: 3584
  • Likes Given: 3922
Re: When will the Arctic Go Ice Free?
« Reply #159 on: October 05, 2018, 08:44:26 PM »
Extent is deceptive. If the arctic has a typical thickness of 2.5m at max, and each year's melt is 2m, what does extent tell you? That all is well. When typical thickness drops to 2.3m at max, and annual melt increases to 2.1m, extent still looks almost fine. But when thickness at max drops to 2.2m and annual melt to 2.3m, you get a BOE all of a sudden.
Extent is showing a trend because at the edges the ice is thinner and annual melt larger than at the heart of the pack. But volume trend is much more indicative.

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #160 on: October 06, 2018, 02:16:26 PM »
Extent is deceptive. If the arctic has a typical thickness of 2.5m at max, and each year's melt is 2m, what does extent tell you? That all is well. When typical thickness drops to 2.3m at max, and annual melt increases to 2.1m, extent still looks almost fine. But when thickness at max drops to 2.2m and annual melt to 2.3m, you get a BOE all of a sudden.
Extent is showing a trend because at the edges the ice is thinner and annual melt larger than at the heart of the pack. But volume trend is much more indicative.

I thick you are misrepresenting the situation.  The thickness varies across the Arctic, and does not melt uniformly.  What you are describing is a small inland lake, which will freeze overnight in the winter, and then slowly melt in the spring, until one day it all disappears. 

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #161 on: October 06, 2018, 06:37:08 PM »
Extent is deceptive. If the arctic has a typical thickness of 2.5m at max, and each year's melt is 2m, what does extent tell you? That all is well. When typical thickness drops to 2.3m at max, and annual melt increases to 2.1m, extent still looks almost fine. But when thickness at max drops to 2.2m and annual melt to 2.3m, you get a BOE all of a sudden.
Extent is showing a trend because at the edges the ice is thinner and annual melt larger than at the heart of the pack. But volume trend is much more indicative.

I thick you are misrepresenting the situation.  The thickness varies across the Arctic, and does not melt uniformly.  What you are describing is a small inland lake, which will freeze overnight in the winter, and then slowly melt in the spring, until one day it all disappears.
oren has just explained the possibility of a "surprising" BOE. His figures are completely valid, independent how thick the ice is and independent how high the melting rate is. It is just an example.
I am very convinced that volume is the measure that has to be looked at if you want to extrapolate to the first BOE. oren's example shows that the extrapolation of area or extent gives much too high values, because if the ice gets very thin, then the floes' area will reduce with an accelerating rate. This rate will once meet the volume reduction rate in the graph close before the BOE will take place.
It is too late just to be concerned about Climate Change

El Cid

  • Young ice
  • Posts: 2507
    • View Profile
  • Liked: 923
  • Likes Given: 225
Re: When will the Arctic Go Ice Free?
« Reply #162 on: October 06, 2018, 08:57:14 PM »
As far as I understand, volume figures are just estimates, and very rough ones at that. We do however have measurements of temperatures which indirectly tell us how thick the ice is. Wintertime temperatures should be higher as the ice gets thinner and more fragmented, since ice is an insulator between air and sea. Thinner ice = higher temperatures.

Indeed, that is happening, and quite rapidly in the past few years. Attached is a chart of 70-90N temps with some notable outliers (winter of 2006/7, and 2011/12). The past 3 years have seen higher than ever temps up North. This probably means that the ice is in a bad shape, actually worse than ever (since we have measurements I mean).

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #163 on: October 06, 2018, 10:00:49 PM »
What once was declared as outliers (2006/07 and 2011/12) is now below of what we saw the last three winters...
It is too late just to be concerned about Climate Change

GoSouthYoungins

  • Nilas ice
  • Posts: 1427
    • View Profile
  • Liked: 150
  • Likes Given: 109
Re: When will the Arctic Go Ice Free?
« Reply #164 on: October 06, 2018, 11:19:27 PM »
2019 or 2020
big time oops

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #165 on: October 08, 2018, 03:49:00 AM »
Extent is deceptive. If the arctic has a typical thickness of 2.5m at max, and each year's melt is 2m, what does extent tell you? That all is well. When typical thickness drops to 2.3m at max, and annual melt increases to 2.1m, extent still looks almost fine. But when thickness at max drops to 2.2m and annual melt to 2.3m, you get a BOE all of a sudden.
Extent is showing a trend because at the edges the ice is thinner and annual melt larger than at the heart of the pack. But volume trend is much more indicative.

I thick you are misrepresenting the situation.  The thickness varies across the Arctic, and does not melt uniformly.  What you are describing is a small inland lake, which will freeze overnight in the winter, and then slowly melt in the spring, until one day it all disappears.
oren has just explained the possibility of a "surprising" BOE. His figures are completely valid, independent how thick the ice is and independent how high the melting rate is. It is just an example.
I am very convinced that volume is the measure that has to be looked at if you want to extrapolate to the first BOE. oren's example shows that the extrapolation of area or extent gives much too high values, because if the ice gets very thin, then the floes' area will reduce with an accelerating rate. This rate will once meet the volume reduction rate in the graph close before the BOE will take place.

If you are convinced that a BOE will occur when the ice gets very thin, the perhaps you should choose thickness as the measure to be looked at if you want to extrapolate.  The only reason that the two-dimensional measures give too high a value, is you think it should be lower.  I think that is a rather biased reason.  But, to each his own.

DavidR

  • Grease ice
  • Posts: 740
    • View Profile
  • Liked: 36
  • Likes Given: 3
Re: When will the Arctic Go Ice Free?
« Reply #166 on: October 08, 2018, 09:35:47 AM »
You are convinced that a BOE will occur when the ice gets very thin, the perhaps you should choose thickness as the measure to be looked at if you want to extrapolate.  The only reason that the two-dimensional measures give too high a value, is you think it should be lower.  I think that is a rather biased reason.  But, to each his own.
That's an odd argument if you ask me. The thickness is estimated by dividing area or extent by  volume. There is a lot of thin ice at the start of the season and a little bit of thick ice regardless of the overall extent.  As volume and extent decline the average thickness doesn't vary much. Thickness is probably a worse indicator than extent or area.  Volume seems a much more likely  measure because it measures the amount of ice lost each year.

Volume declines based on the difference between the ice formed in the freezing  season and the ice melting in the melting season. This has meant an average loss of about about 320 km^3 per year since 1988 but about 500 Km^3 per year since 2000 this appears to  be increasing exponentially  as warmer waters move into the Arctic.  With only  5000 Km^3 at minimum this year it seems highly likely  that within 10 years we will see a BOE in September. Extrapolating from extent,  thickness or area gives a much later date.
Toto, I've a feeling we're not in Kansas anymore

El Cid

  • Young ice
  • Posts: 2507
    • View Profile
  • Liked: 923
  • Likes Given: 225
Re: When will the Arctic Go Ice Free?
« Reply #167 on: October 08, 2018, 10:02:04 AM »
I'm amazed that people can still take extrapolations seriously. If you had extrapolated in 2007, 8, or 12 based on either extent or volume or whatever, we should already be ice free. We are not, which proves the point: extrapolation does not work. It doesn't work, beacuse it is a nonlinear system, and melting the shelf is not the same as melting the central pack. Therefore extrapolations are rather a waste of time.

Even if you wanted to extrapolate I would urge to disregard anything other than the central pack, as most everything else melts out anyway, and extrapolate just the central pack going down to zero. Even this is probably close to useless but much better than using the total volume or extent numbers.

El Cid

  • Young ice
  • Posts: 2507
    • View Profile
  • Liked: 923
  • Likes Given: 225
Re: When will the Arctic Go Ice Free?
« Reply #168 on: October 08, 2018, 10:18:48 AM »
So this is the rolling average past 12 month piomas volume of the CAB (based on Wipneus' file, attached). You can see that we lost cca 5000km3 from 1995 to 2011 (annual average loss of cca 300+ km3, which lineary extrapolated should have pointed to 0km3 remaining by around 2030), but it has been moving sideways in the past 7 years. The central pack is a hard nut to crack. We might see another 10 yrs of sideways movement or we might see a huge decline even the next few years. It is impossible to say. But extrapolation will not help.

I also attach an extrapolation from 2010 to 2018. It was such a nice fit, and still  it was a huge miss! Based on 2010 extrapolations, we should be around 5000 km3 average volume, whereas in realitywe are at 7400

 

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: When will the Arctic Go Ice Free?
« Reply #169 on: October 08, 2018, 02:54:43 PM »
which proves the point: extrapolation does not work. It doesn't work, beacuse it is a nonlinear system, and melting the shelf is not the same as melting the central pack. Therefore extrapolations are rather a waste of time.

Seems a bit sweeping to me. It certainly doesn't work when a new trend is emerging or when you use the wrong curve type or when trying to predict a long way outside the data range. Using exponential when the models suggest something more like gompertz was asking for big miss and even more so when the trend was beginning to show a levelling off and there were explanations in the literature to explain an acceleration and deceleration.

You pick out two reasons: non linear and central vs shelf. Certainly it is much easier to extrapolate with a linear system and I also agree re central vs shelf. But I would suggest that there are other reasons and maybe the two you picked alone wouldn't automatically mean that extrapolations will always fail.

El Cid

  • Young ice
  • Posts: 2507
    • View Profile
  • Liked: 923
  • Likes Given: 225
Re: When will the Arctic Go Ice Free?
« Reply #170 on: October 08, 2018, 03:33:14 PM »
which proves the point: extrapolation does not work. It doesn't work, beacuse it is a nonlinear system, and melting the shelf is not the same as melting the central pack. Therefore extrapolations are rather a waste of time.

Seems a bit sweeping to me. ...

You pick out two reasons: non linear and central vs shelf. Certainly it is much easier to extrapolate with a linear system and I also agree re central vs shelf. But I would suggest that there are other reasons and maybe the two you picked alone wouldn't automatically mean that extrapolations will always fail.

Crandles, you are absolutely right. I consciously oversimplified the question just like extrapolation is IMO a very serious oversimplification of such an amazingly complex system. Nonetheless, we are definitely progressing towards a (more or less) ice free Arctic, and I just wanted to emphasize that forecasting the timing of a BOE (or any other state) is nigh impossible, no matter how much we would like to...

Archimid

  • Young ice
  • Posts: 3511
    • View Profile
  • Liked: 899
  • Likes Given: 206
Re: When will the Arctic Go Ice Free?
« Reply #171 on: October 08, 2018, 03:40:46 PM »
Speaking of useless extrapolations...


The first attachment uses maximum volume and the complete data set of the melting season volume loss to estimate the first ice free arctic.

The second attachment shows the maximum volume to volume loss ratio. There is a very good evidence that after 2007 there was a step change in the Arctic. This is supported by significant increase of the average losses. 

The third attachment uses the losses only from 2007-2018 with the justification that the system changed.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

Pmt111500

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #172 on: October 08, 2018, 05:16:03 PM »
Yay Archimid! That looks like if the system change of 2010-2014 is the real change, and all other measurements are in error ice will never go away! Nice example of real problems in selecting anything other than a linear function for an extrapolation.

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #173 on: October 08, 2018, 05:53:42 PM »
Yay Archimid! That looks like if the system change of 2010-2014 is the real change, and all other measurements are in error ice will never go away! Nice example of real problems in selecting anything other than a linear function for an extrapolation.

Curve fitting can sometimes be an exercise in futility.  It looks nice, but does it describe the true situation.  Hard to imagine maximum ice volume approaching zero in an environment that is 20C below freezing.

Rodius

  • Nilas ice
  • Posts: 2123
    • View Profile
  • Liked: 645
  • Likes Given: 46
Re: When will the Arctic Go Ice Free?
« Reply #174 on: October 09, 2018, 01:08:02 AM »


Curve fitting can sometimes be an exercise in futility.  It looks nice, but does it describe the true situation.  Hard to imagine maximum ice volume approaching zero in an environment that is 20C below freezing.
[/quote]

You say this, yet the Arctic has been ice free multiple times before

Sam

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #175 on: October 09, 2018, 01:38:48 AM »
Interesting plots and extrapolations.

Yes, it is always important to be careful with extrapolations. That said, these two extrapolations lend at least some credence and support for my informal projection that we may see an ice free Arctic year round by 2035, and certainly not later than 2050.

As far as the idea that at -20 C the Arctic cannot be ice free....

That presumes without basis the idea that the Arctic system will remain as it is today despite the absence of the ice. Clearly it won't. How will it behave? We simply do not know.

We know the Arctic has gone ice free through quite a long part of geologic history. We know even that it has been quite warm, supporting subtropical plants and creatures as far north as Ellesmere and southern Greenland.

So how is this possible? That is a vastly better question, than the presumption contrary to known history that it will remain the same as now, and therefor cannot go ice free. That argument is circular and clearly wrong.

So, how?

We don't know. But we are about to find out. One plausible answer is that with the loss of the cold pole (no ice), and the dramatic shifts in atmospheric circulation, that at least two new things happen. 

First, the warm oceans and land support higher humidity that supports deep clouds through the Arctic winter insulating the pole from radiative heat loss.

Second, that the single circulation band and increased height of the tropopause drive heat migration northward via the atmosphere in a slower flow pattern.  The current three cell system acts as a block preventing that. With the loss of the three cell system, the circulation will likely look more like what we see on Venus.

Conditions there are so radically different from those on Earth, that we likely cannot learn too much from Venus as an exemplar. It does though give us a crude framework to consider.

Sam

SteveMDFP

  • Nilas ice
  • Posts: 2476
    • View Profile
  • Liked: 583
  • Likes Given: 42
Re: When will the Arctic Go Ice Free?
« Reply #176 on: October 09, 2018, 01:48:19 AM »
The current three cell system acts as a block preventing that. With the loss of the three cell system, the circulation will likely look more like what we see on Venus.

Conditions there are so radically different from those on Earth, that we likely cannot learn too much from Venus as an exemplar. It does though give us a crude framework to consider.

Sam

Thanks for that, Sam.  I didn't know that we knew the atmospheric circulation patterns on Venus.  Can you elaborate?  It might indeed be relevant, and I don't know that the subject has ever come up on this forum.

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #177 on: October 09, 2018, 02:59:35 AM »


Curve fitting can sometimes be an exercise in futility.  It looks nice, but does it describe the true situation.  Hard to imagine maximum ice volume approaching zero in an environment that is 20C below freezing.

You say this, yet the Arctic has been ice free multiple times before
[/quote]

Yes, during those times when the temperature was warming, preventing freezing.  Simple physics.

Sam

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #178 on: October 09, 2018, 04:47:48 AM »
Steve,

I've not tried posting links before. We will see how this goes.

Here are a couple of "simpler" articles on the subject.

https://www.lpi.usra.edu/vexag/chapman_conf/presentations/limaye.pdf

http://lasp.colorado.edu/~espoclass/ASTR_5835_2015_Readings_Notes/Notes_presentations/Schiff_Nov17-superrotation_presentation-expanded.pdf

And a more intense General one on exoplanets generally, especially others bodies in our system.

https://www.lpl.arizona.edu/~showman/publications/showman-etal-exoplanets-review-revised.pdf

Venus is an attractive case because it is our near twin. It is also has a series of problems for trying to relate it to Earth. Venus lacks a moon (and tidal forces). It has a tremendously dense hot and very different atmosphere. And it has a very slow retrograde rotation. Th atmosphere on Venus rotates in the direction of the planet, but much faster (super rotation).

Venus has massive storms over both poles that are persistent structures similar to those on Saturn.

Those might be suggestive of what our future may hold in store (Earth's distant past too).

Sam

Rodius

  • Nilas ice
  • Posts: 2123
    • View Profile
  • Liked: 645
  • Likes Given: 46
Re: When will the Arctic Go Ice Free?
« Reply #179 on: October 09, 2018, 04:48:23 AM »


Curve fitting can sometimes be an exercise in futility.  It looks nice, but does it describe the true situation.  Hard to imagine maximum ice volume approaching zero in an environment that is 20C below freezing.

You say this, yet the Arctic has been ice free multiple times before

Yes, during those times when the temperature was warming, preventing freezing.  Simple physics.
[/quote]

What do you think is happening at the moment?
Are you suggesting the Arctic is not warming?

Sam

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #180 on: October 09, 2018, 04:58:52 AM »
Steve,

Here is a bit clearer explanation. Skip to page 26-28. Lots of fascinating data and information.

https://spaceflightsystems.grc.nasa.gov/SSPO/SP/VenusUpper/Presentations/Crisp_1_What_We_Know_Today_Venus_STIM_20130124.pdf

Sam

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #181 on: October 09, 2018, 02:40:17 PM »


Curve fitting can sometimes be an exercise in futility.  It looks nice, but does it describe the true situation.  Hard to imagine maximum ice volume approaching zero in an environment that is 20C below freezing.

You say this, yet the Arctic has been ice free multiple times before

Yes, during those times when the temperature was warming, preventing freezing.  Simple physics.

What do you think is happening at the moment?
Are you suggesting the Arctic is not warming?
[/quote]

Not at all.  Should the Arctic temperature increase another 5C, then complete melting of the ice is certainly possible.  That is the estimated temperature the last time the Arctic was ice-free.

mostly_lurking

  • Frazil ice
  • Posts: 372
    • View Profile
  • Liked: 19
  • Likes Given: 18
Re: When will the Arctic Go Ice Free?
« Reply #182 on: October 09, 2018, 02:56:12 PM »
...snip


Not at all.  Should the Arctic temperature increase another 5C, then complete melting of the ice is certainly possible.  That is the estimated temperature the last time the Arctic was ice-free.

In the summer?

binntho

  • Nilas ice
  • Posts: 2193
    • View Profile
  • Liked: 878
  • Likes Given: 235
Re: When will the Arctic Go Ice Free?
« Reply #183 on: October 09, 2018, 02:57:06 PM »
Should the Arctic temperature increase another 5C, then complete melting of the ice is certainly possible.  That is the estimated temperature the last time the Arctic was ice-free.

Well, there are different opinions as to when the arctic was last ice-free, at least seasonally (i.e. during summer). Quite possibly the last time was during the holocenene maximum when temperatures were not significantly higher than they are now.

https://www.sciencedirect.com/science/article/pii/S0277379113004162
because a thing is eloquently expressed it should not be taken to be as necessarily true
St. Augustine, Confessions V, 6

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: When will the Arctic Go Ice Free?
« Reply #184 on: October 09, 2018, 04:13:33 PM »
You are convinced that a BOE will occur when the ice gets very thin, the perhaps you should choose thickness as the measure to be looked at if you want to extrapolate.  The only reason that the two-dimensional measures give too high a value, is you think it should be lower.  I think that is a rather biased reason.  But, to each his own.
That's an odd argument if you ask me. The thickness is estimated by dividing area or extent by  volume. There is a lot of thin ice at the start of the season and a little bit of thick ice regardless of the overall extent.  As volume and extent decline the average thickness doesn't vary much. Thickness is probably a worse indicator than extent or area.  Volume seems a much more likely  measure because it measures the amount of ice lost each year.

Volume declines based on the difference between the ice formed in the freezing  season and the ice melting in the melting season. This has meant an average loss of about about 320 km^3 per year since 1988 but about 500 Km^3 per year since 2000 this appears to  be increasing exponentially  as warmer waters move into the Arctic.  With only  5000 Km^3 at minimum this year it seems highly likely  that within 10 years we will see a BOE in September. Extrapolating from extent,  thickness or area gives a much later date.
Volume is the key metric, as it is the primary expression of over-all heat exchange and total enthalpy in the system.
This space for Rent.

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #185 on: October 09, 2018, 05:03:15 PM »
You are convinced that a BOE will occur when the ice gets very thin, the perhaps you should choose thickness as the measure to be looked at if you want to extrapolate.  The only reason that the two-dimensional measures give too high a value, is you think it should be lower.  I think that is a rather biased reason.  But, to each his own.
That's an odd argument if you ask me. The thickness is estimated by dividing area or extent by  volume. There is a lot of thin ice at the start of the season and a little bit of thick ice regardless of the overall extent.  As volume and extent decline the average thickness doesn't vary much. Thickness is probably a worse indicator than extent or area.  Volume seems a much more likely  measure because it measures the amount of ice lost each year.

Volume declines based on the difference between the ice formed in the freezing  season and the ice melting in the melting season. This has meant an average loss of about about 320 km^3 per year since 1988 but about 500 Km^3 per year since 2000 this appears to  be increasing exponentially  as warmer waters move into the Arctic.  With only  5000 Km^3 at minimum this year it seems highly likely  that within 10 years we will see a BOE in September. Extrapolating from extent,  thickness or area gives a much later date.

Not sure where you are getting your determinations from, but volume is the estimated parameter, and calculated by multiplying the thickness by the area.  Thickness is determined by averaging the modeled thickness in each gridded cell.  Hence volume has thickness plus area uncertainty incorporated into its calculation.  Also, if both area (2-dimensions) and thickness (1-dimension) are decreasing, the volume (3-dimension) must decrease faster.  Hence, volume will always show a faster rate of decline than either area or thickness, but will slow down as the other parameters catch up.

Tor Bejnar

  • Young ice
  • Posts: 4606
    • View Profile
  • Liked: 879
  • Likes Given: 826
Re: When will the Arctic Go Ice Free?
« Reply #186 on: October 09, 2018, 05:56:41 PM »
Quote
volume will always show a faster rate of decline than either area or thickness, but will slow down as the other parameters catch up.
I'm not sure this is always true, but I'm not a thermodynamics anything (not an expert nor even particularly knowledgeable).  If a significant cause of ice melt is bottom melt (which will likely increase as thickness decreases, as solar input into this process increases with ice thinness), then declining thickness will accelerate as zero volume approaches, while area remains fairly unchanged.  If area decreases 'at all', it might be a 'fact' that volume will decrease at a lower rate than does thickness (although faster than the area decline).  But I think this doesn't require volume declines to actually "slow down".
Arctic ice is healthy for children and other living things because "we cannot negotiate with the melting point of ice"

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #187 on: October 09, 2018, 07:43:00 PM »
Quote
volume will always show a faster rate of decline than either area or thickness, but will slow down as the other parameters catch up.
I'm not sure this is always true, but I'm not a thermodynamics anything (not an expert nor even particularly knowledgeable).  If a significant cause of ice melt is bottom melt (which will likely increase as thickness decreases, as solar input into this process increases with ice thinness), then declining thickness will accelerate as zero volume approaches, while area remains fairly unchanged.  If area decreases 'at all', it might be a 'fact' that volume will decrease at a lower rate than does thickness (although faster than the area decline).  But I think this doesn't require volume declines to actually "slow down".

It is a case of simple mathematics.  The current minimum sea ice extent is ~60% of the extent measured in 1980.  The minimum volume is ~30% of 1980.  The thickness is ~50% of 1980.  Volume is just a product of these two; and since both are decreasing, the volume must decrease faster.  As extent and thickness decrease, the volume loss will slow because there is less volume to lose. 

This is different from the typical ice cube in a glass experiment, whereby all sides melt equally.  The large disproportionate dimensions leads to a difference in the physical melting properties.  The larger extent (or area) dominates the melting characteristic. 

If one were to assume that volume is the key parameter, and volume continues to decrease at its current rate, then the decline in thickness would need to double and the decline in extent more than triple, so that all three parameters reach zero simultaneously.  That is not what we are experiencing currently.

Sam

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #188 on: October 09, 2018, 08:16:11 PM »
Klondike Kat,

I fear that you may have fallen victim to a variation of Zeno's paradox. (Ice volume decrease slow down)

Common experience suggests just the opposite. As anyone who has every drunk anything with ice in it can attest - the ice lasts and lasts and lasts, until near the end. As the volume of ice approaches zero, the exposed surface area increases and the melt rate increases with it, or ----- it at least seems to. I suspect the latter is the truth.

Rather than slowing down, I suspect that the volumetric loss rate would remain the same, all other factors remaining constant. But they aren't remaining constant. The heat input to the Arctic waters is increasing through direct solar absorption, through warmer fresh water inputs from rivers, and through intrusion of warmer waters from the Atlantic and Pacific. This is to some degree countered by increasing flows of cold (above freezing) water from Greenland melt.

At the same time, with the loss of ice, wind and wave action is increasing. This sloshing and stirring has to also be increasing the melt rate. Ditto for mixing of the Arctic Ocean bringing warmer and more saline waters into greater contact with the ice. Both increase the melt rate.

I fully expect the ice melt rate to increase as the ice volume nears zero. This I suspect will show up as increasing thinning, loss of multi year ice, shattering and dispersion of the ice, and formation of more vulnerable saline first year ice. And that is what we have seen.

Unsurprisingly, this is causing the extent to at least appear to increase compared to expectations due to the shattering and dispersion of the ice pack, coupled with the arbitrary rule that any area of Ocean that is covered by 15% or greater of ice is counted as being 100% covered.

That rule worked reasonably to smooth the edge of the intact ice sheet. It does not work at all well with a shattered ice sheet. To the contrary, it lies to us and causes us to falsely believe that conditions are no where near as bad as they really are.

And all of that leaves us vulnerable to seeing all of the ice seeming to go "poof!" over a short time span and making everyone involved look silly.

Sam

Ned W

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #189 on: October 09, 2018, 09:25:04 PM »
I fully expect the ice melt rate to increase as the ice volume nears zero.

Except that doesn't seem to be what's happening:


So perhaps intuition, and analogizing the Arctic to ice cubes in a drinking glass, aren't all that helpful?  Reality sure looks more like one of crandles's Gompertz curves.

Credit: Wipneus's monthly PIOMAS regional volume data.

Sam

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #190 on: October 09, 2018, 10:59:06 PM »
Quote
Except that doesn't seem to be what's happening:

snip...

So perhaps intuition, and analogizing the Arctic to ice cubes in a drinking glass, aren't all that helpful?  Reality sure looks more like one of crandles's Gompertz curves.

More than perhaps. We all of us are to a good degree - whistling in the dark. The dynamics of the current and future conditions are unlike anything we have good evidence or models for. There are far too many uncertainties to have high confidence in predictions.

However, we can be sure that the difference in consequences is important in how we plan for our future. The consequences of an ice free Arctic to atmospheric and oceanic processes are enormous for humans and all creatures on Earth. Hoping or planning for less consequential outcomes runs the risk of triggering horrific outcomes.

Also, though at least in the short term the model projected ice volume appears to be following something like a Gompertz curve, it is important to be mindful of a couple of things.

First, the PIOMAS volume, useful as it is, is a modelled result. It may be wrong.

Second, within any data set like this there is statistical noise as well as potential systematic errors that have escaped recognition. The statistical noise can lead to misleading projections. Time will tell.

Third, there may be self referential (circular) errors involved where the fundamentals of how PIOMAS works may be misleading.

Sam

Archimid

  • Young ice
  • Posts: 3511
    • View Profile
  • Liked: 899
  • Likes Given: 206
Re: When will the Arctic Go Ice Free?
« Reply #191 on: October 09, 2018, 11:23:34 PM »
There is a hiatus in the minimum but there is no hiatus in the maximum volume. To me this means that the hiatus of the minimum will continue until the maximum is low enough to push the minimum down.

I think 2030 is the  most the Arctic has left. The last few KM3 will probably puff out of existence.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

DavidR

  • Grease ice
  • Posts: 740
    • View Profile
  • Liked: 36
  • Likes Given: 3
Re: When will the Arctic Go Ice Free?
« Reply #192 on: October 10, 2018, 09:36:38 AM »
Quote
volume will always show a faster rate of decline than either area or thickness, but will slow down as the other parameters catch up.
I'm not sure this is always true, but I'm not a thermodynamics anything (not an expert nor even particularly knowledgeable).  If a significant cause of ice melt is bottom melt (which will likely increase as thickness decreases, as solar input into this process increases with ice thinness), then declining thickness will accelerate as zero volume approaches, while area remains fairly unchanged.  If area decreases 'at all', it might be a 'fact' that volume will decrease at a lower rate than does thickness (although faster than the area decline).  But I think this doesn't require volume declines to actually "slow down".

It is a case of simple mathematics.  The current minimum sea ice extent is ~60% of the extent measured in 1980.  The minimum volume is ~30% of 1980.  The thickness is ~50% of 1980.  Volume is just a product of these two; and since both are decreasing, the volume must decrease faster.  As extent and thickness decrease, the volume loss will slow because there is less volume to lose. 

This is different from the typical ice cube in a glass experiment, whereby all sides melt equally.  The large disproportionate dimensions leads to a difference in the physical melting properties.  The larger extent (or area) dominates the melting characteristic. 

If one were to assume that volume is the key parameter, and volume continues to decrease at its current rate, then the decline in thickness would need to double and the decline in extent more than triple, so that all three parameters reach zero simultaneously.  That is not what we are experiencing currently.
The key  point  you  are missing is that the thickness is an average measure over the entire ice cap.  The area that is less than 20cm thick at maximum is 20 - 30  times greater than the are that is 2 m thick for the same volume. The simple formula V=A*T  only  works for a single cell.

Over the entire cap the  AverageThickness = sum(( T  * N ) / Total(N))  for all thicknesses where N is the number of cells of a given thickness.  As the ice melts the ratios between the various N's doesn't change much  so the average thickness doesn't change either.

Simple example; there are 100 cells at thickness 0.1m and 10 at  thickness 1m;  after massive melting the number of cells has reduced to 10 at 0.1m and 1 at 1m.  The ratio hasn't changed and the average thickness hasn't changed despite area declining by 90%. 

Toto, I've a feeling we're not in Kansas anymore

johnm33

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #193 on: October 10, 2018, 10:28:50 AM »
I think there's a threshold that needs forcing through, whilst there's enough ice around the pole to preserve an ice desert the ice will always regenerate and expand. The ice around the pole has to be accelerated to force it south, 0kph at the pole @175kph at the northernmost tip of Greenland, any wind powerful enough to do that also serves to enhance the ice deserts regenerative ability. The warm currents entering the Arctic either fall into the deeps or have too much inertia to move towards the pole, mostly both. We may have to wait until the temperature  of the ocean itself has the energy to cause bottom melt in the CAB.
 Looking at the various animations this year [thanks everyone] it seems pretty clear that the ice is expanding from the center out towards either the exits or to be destroyed by wave action, with the possible exception of ESS where there was thick ice left over from last season in sufficient quantity to almost serve as it's own ice desert. At the moment I can't see how a seperate area of thick ice could be established, so we may see the CAB under siege on all fronts. 

Pmt111500

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #194 on: October 10, 2018, 01:49:05 PM »
As the thickening of ice on the late freezing season stops so does the generation of thick first year ice which is essential to the production of two-year/old ice. Maybe the year this happens could be estimated. Last three winters the ESS-CAB assisted by CAA channels have done this stuff. Possibly assisted, paradoxically, by the diminished export to Fram made by the sub-polar gyre expansion to former arctic realm.

Klondike Kat

  • Grease ice
  • Posts: 842
    • View Profile
  • Liked: 40
  • Likes Given: 56
Re: When will the Arctic Go Ice Free?
« Reply #195 on: October 10, 2018, 02:04:23 PM »
Quote
volume will always show a faster rate of decline than either area or thickness, but will slow down as the other parameters catch up.
I'm not sure this is always true, but I'm not a thermodynamics anything (not an expert nor even particularly knowledgeable).  If a significant cause of ice melt is bottom melt (which will likely increase as thickness decreases, as solar input into this process increases with ice thinness), then declining thickness will accelerate as zero volume approaches, while area remains fairly unchanged.  If area decreases 'at all', it might be a 'fact' that volume will decrease at a lower rate than does thickness (although faster than the area decline).  But I think this doesn't require volume declines to actually "slow down".

It is a case of simple mathematics.  The current minimum sea ice extent is ~60% of the extent measured in 1980.  The minimum volume is ~30% of 1980.  The thickness is ~50% of 1980.  Volume is just a product of these two; and since both are decreasing, the volume must decrease faster.  As extent and thickness decrease, the volume loss will slow because there is less volume to lose. 

This is different from the typical ice cube in a glass experiment, whereby all sides melt equally.  The large disproportionate dimensions leads to a difference in the physical melting properties.  The larger extent (or area) dominates the melting characteristic. 

If one were to assume that volume is the key parameter, and volume continues to decrease at its current rate, then the decline in thickness would need to double and the decline in extent more than triple, so that all three parameters reach zero simultaneously.  That is not what we are experiencing currently.
The key  point  you  are missing is that the thickness is an average measure over the entire ice cap.  The area that is less than 20cm thick at maximum is 20 - 30  times greater than the are that is 2 m thick for the same volume. The simple formula V=A*T  only  works for a single cell.

Over the entire cap the  AverageThickness = sum(( T  * N ) / Total(N))  for all thicknesses where N is the number of cells of a given thickness.  As the ice melts the ratios between the various N's doesn't change much  so the average thickness doesn't change either.

Simple example; there are 100 cells at thickness 0.1m and 10 at  thickness 1m;  after massive melting the number of cells has reduced to 10 at 0.1m and 1 at 1m.  The ratio hasn't changed and the average thickness hasn't changed despite area declining by 90%.

Which is another reason to be skeptical of the volume calculations.  Using this average thickness to calculate volume, introducing more error into the equation.  Hence, the reason that I prefer extent to the other parameters.  Sure, it is not perfect.  However, it is the most consistent and reliable parameter currently available. 

Stephan

  • Young ice
  • Posts: 2649
    • View Profile
  • Liked: 758
  • Likes Given: 459
Re: When will the Arctic Go Ice Free?
« Reply #196 on: October 10, 2018, 04:47:35 PM »
Just a question for us all.
We already have seas that melt out completely (like Hudson or Kara). Would these seas be an example for a BOE? And has anyone ever monitored the rates of volume, thickness and area decline close before the end and compared these rates with the rates at about 30, 50, 70 % of ice coverage? This analysis (also the relation of the rates) could help to answer the questions that arose in this thread.
It is too late just to be concerned about Climate Change

GoSouthYoungins

  • Nilas ice
  • Posts: 1427
    • View Profile
  • Liked: 150
  • Likes Given: 109
Re: When will the Arctic Go Ice Free?
« Reply #197 on: October 10, 2018, 05:07:55 PM »
Just a question for us all.
We already have seas that melt out completely (like Hudson or Kara). Would these seas be an example for a BOE? And has anyone ever monitored the rates of volume, thickness and area decline close before the end and compared these rates with the rates at about 30, 50, 70 % of ice coverage? This analysis (also the relation of the rates) could help to answer the questions that arose in this thread.

The seas are too different to be very informative. The central arctic (and greenland) is/are the Northern Hemisphere's ice cap. It is the lynch pin of half the world's weather. The dynamics as it starts to disappear are very odd and very complex and everyone is just guessing with varying levels of educatedness.
big time oops

magnamentis

  • Guest
Re: When will the Arctic Go Ice Free?
« Reply #198 on: October 10, 2018, 05:27:02 PM »
Just a question for us all.
We already have seas that melt out completely (like Hudson or Kara). Would these seas be an example for a BOE? And has anyone ever monitored the rates of volume, thickness and area decline close before the end and compared these rates with the rates at about 30, 50, 70 % of ice coverage? This analysis (also the relation of the rates) could help to answer the questions that arose in this thread.

a very interesting approach and to a certain extent a comparison might be valid. one thing that we see quite often is that some remaining ice is very persisting (reluctant to melt) and ultimately caves in within a few days.

i suspect that the preparation for a BOE is very similar and that below a certain amount of remaining ice we gonne be caught by surprise how quickly the rest will go.

all guesswork of course, but i think things gonna unfold somehow along such a line.
« Last Edit: October 10, 2018, 08:11:08 PM by magnamentis »

Dharma Rupa

  • Frazil ice
  • Posts: 493
    • View Profile
  • Liked: 52
  • Likes Given: 26
Re: When will the Arctic Go Ice Free?
« Reply #199 on: October 10, 2018, 05:29:57 PM »
We already have seas that melt out completely (like Hudson or Kara). Would these seas be an example for a BOE?

I kind of doubt it, with the main distinction being that they have the CAB next to them and the CAB has nothing (other than a bit of Greenland).