Support the Arctic Sea Ice Forum and Blog

Poll

When will be the first year that daily ice volume per PIOMAS goes below 1000 km^3?

2018
10 (8.5%)
2019-2021
20 (16.9%)
2022-2024
32 (27.1%)
2025-2028
30 (25.4%)
2029-2033
15 (12.7%)
2034-2039
3 (2.5%)
2040-2049
2 (1.7%)
2050-2059
1 (0.8%)
2060 or later
5 (4.2%)

Total Members Voted: 118

Voting closed: May 27, 2018, 05:31:41 PM

Author Topic: First year PIOMAS volume below 1000km^3 poll 2018  (Read 10875 times)

Daniel B.

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 30
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #100 on: April 14, 2018, 06:36:16 PM »
Binntho, no there is no confusion.  The declines at minimum and maximum cannot be directly comparable precisely due to the geographic constraint you mentioned.  The only areas of potential expansion (contraction) are the Bering Sea and North Atlantic.  Hence, maximum extent has decreased only 10% since 1979.



The ice maximum appears to have decreased more in the past few years, presumably tied to recent winter warmth.  The minimum has decreased less, possibly tied to summer cooling.  The summer minimum does not appear to be that closely tied to extent or thickness at maximum.  The lowest minimum occurred in 2012, but the previous and subsequent maxima were relatively high.  Similarly, the last two maxima have been the lowest, but the intervening minimum was relatively high (only 8th lowest, and 4th highest since the 2007 record loss). 

I think we agree that one metric cannot reach zero before the other.  If volume were to continue a liner decrease, extent losses must accelerate to keep up, resulting in an ice-free Arctic ~2030.  However, if extent continues to decline linearly, then volume will continue on its current deceleration, resulting in an ice-free Arctic around 2050.  The thickness is an average across the entire Arctic.  In many areas, thickness goes to zero in the summer, resulting in open water.  Other areas are much thicker, retaining ice year-round.  Only that ice which is very thin will experience large melt each year, the rest will not.  This non-uniformity will extend the timeframe until an ice-free Arctic much longer than expressed by others here.  But I admit, that this is just my opinion also, and the Arctic will do whatever it does, regardless of what we say it will.

oren

  • ASIF Upper Class
  • Posts: 2288
    • View Profile
  • Liked: 122
  • Likes Given: 259
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #101 on: April 14, 2018, 07:15:08 PM »
There is a lot of micro-level data from mass balance buoys that can shed light on the discussion. Where to find it and how to analyse it?

Archimid

  • ASIF Upper Class
  • Posts: 1135
    • View Profile
  • Liked: 17
  • Likes Given: 29
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #102 on: April 14, 2018, 07:17:53 PM »

I appreciate folks posting statistics about Arctic ice thickness/volume/extent/area and drawing projections from them.  I'd much rather see statistics about the CAB.

The other day I rediscovered Wipneus great regional PIOMAS spreadsheets. I've been playing with the CAB for a few days. I've been mulling the two attached graphs but I'm not quite sure what to make of them. Since you ask for them I'll take this opportunity to post them.

The first attachment is CAB Monthly Minimum as a percentage of the Total Monthly Minimum. In 1979 the CAB represented only 65% of the total ice in the Arctic at minimum. From 1979 to 2006  the CAB increased it's representation in the total minimum to 80%. After the big crash of 2007 the CAB now makes up about 95% of the total minimum.

The second attachment is the CAB monthly Maximum as a percentage of Total Maximum. This is a whole different story. Notice how regardless of the tremendous changes that happened in the Arctic over the last 30 years the CAB wants to be around 46% of the total maximum. I'm still trying to understand why this is.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #103 on: April 14, 2018, 07:43:24 PM »
"My theory, that belongs to me " (pace Monty Python) is that the linear decline of sea ice volume and the linear decline of sea ice extent will continue for a few years and then the remaining ice will fall apart, on 27th August 2018 WHOOPS!! 2028 (Thankyou Daniel B) at 3.15 pm (GMT).

And the data available is insufficient to say that this is nonsense. And that is the problem.

The more we find out what data we need to know the more we find out that the data is not there or inadequate. Sod's law says that we will get the data the day the ice is gone.
« Last Edit: April 14, 2018, 07:59:29 PM by gerontocrat »
"Para a Causa do Povo a Luta Continua!"

Daniel B.

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 30
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #104 on: April 14, 2018, 07:51:20 PM »
"My theory, that belongs to me " (pace Monty Python) is that the linear decline of sea ice volume and the linear decline of sea ice extent will continue for a few years and then the remaining ice will fall apart, on 27th August 2018 at 3.15 pm (GMT).

And the data available is insufficient to say that this is nonsense. And that is the problem.

The more we find out what data we need to know the more we find out that the data is not there or inadequate. Sod's law says that we will get the data the day the ice is gone.

Based on the current data, your guess is as good as anyone's.  However, you may have made a typo in your year.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #105 on: April 14, 2018, 07:57:14 PM »
"My theory, that belongs to me " (pace Monty Python) is that the linear decline of sea ice volume and the linear decline of sea ice extent will continue for a few years and then the remaining ice will fall apart, on 27th August 2018 at 3.15 pm (GMT).

And the data available is insufficient to say that this is nonsense. And that is the problem.

The more we find out what data we need to know the more we find out that the data is not there or inadequate. Sod's law says that we will get the data the day the ice is gone.


Based on the current data, your guess is as good as anyone's.  However, you may have made a typo in your year.

You are correct, it is a typo - but what if it turns out to be correct?
"Para a Causa do Povo a Luta Continua!"

Daniel B.

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 30
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #106 on: April 14, 2018, 10:10:48 PM »
That would be very scary.

Dharma Rupa

  • ASIF Citizen
  • Posts: 271
    • View Profile
  • Liked: 14
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #107 on: April 14, 2018, 11:20:25 PM »
The thinner the ice the faster it melts.

I agree with this with one important qualification.  Below some threshold thermal conductivity takes hold and the thinner the ice the faster it melts.  Not sure but I believe it is just under 1 meter.

Once the ice is thin enough that ocean heat and insolation have merged effects the ice will melt faster in Summer and thicken faster in Winter.  (Not sure about that thickening faster in Winter, really.)

« Last Edit: April 14, 2018, 11:41:14 PM by Dharma Rupa »

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #108 on: April 15, 2018, 08:26:45 AM »
The second attachment is the CAB monthly Maximum as a percentage of Total Maximum. This is a whole different story. Notice how regardless of the tremendous changes that happened in the Arctic over the last 30 years the CAB wants to be around 46% of the total maximum. I'm still trying to understand why this is.
Thanks for the graphs Archimid. But isn't the 46% figure the proportion of CAB area of the Arcic Ocean + some external areas, all of which freeze over almost completely (or to more or less the same extent) each winter?

The Central Arctic Basin is of course the most important, but as far as I am aware, the term Arctic Ocean is used to cover more or less all the open ocean constrained by Siberia, Alaska, Canada and Greenland. Which is significantly bigger than the CAB.

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #109 on: April 15, 2018, 08:34:16 AM »
The thinner the ice the faster it melts.
I agree with this with one important qualification.  Below some threshold thermal conductivity takes hold and the thinner the ice the faster it melts.  Not sure but I believe it is just under 1 meter.

I agree with you both, my original statement should perhaps have had a caveat. But I had an inkling of this so that's why I stated that 2m thick ice looses volume as fast as 1m thick ice - which I'm sure is not totally correct, but near enough for that discussion.

Thinner ice lets more sunlight through to be absorbed by the underlying ocean, that's pretty clear.

And thinner ice is also presumably warmer  - although I'm not sure why that would be - won't the ice have an internal temperature that is a gradient between the ocean underneath and the air above?
« Last Edit: April 15, 2018, 09:58:03 AM by binntho »

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #110 on: April 15, 2018, 09:57:43 AM »
Binntho, no there is no confusion. 
I'm not so sure - I feel confused all the time ...

The summer minimum does not appear to be that closely tied to extent or thickness at maximum. 
I'm sure there is no year-on-year predictive value in looking at winter extent + winter thickness. Last year showed that very decisively. But I did find it intriguing that in the longer run, the decrease in winter extent (10%) + winter thickness (35%) = 45% which is the decrease in summer extent over the same period. But these are percentages of values eyeballed off some graphs, so I won't make any claims to them being particularly correct, valid or significant.

Having said that, I do think that winter thickness (which to all purposes is strongly linked to winter volume) is the most important metric when looking at what might happen if things continue as they are - which they might not, of course!

In my last post I talked as if all the ice melted at the same rate irregardless of thickness (which I think is close enough for our purposes) and location (which is obviously wrong). Ice further south melts faster, ice closer to warm waters melts faster, ice closer to the periphery melts faster. And no doubt there are other factors to consider as well. This is a very complex 4 dimensional system, no single metric is going to tell us what the future might hold.

Only that ice which is very thin will experience large melt each year, the rest will not.  This non-uniformity will extend the timeframe until an ice-free Arctic much longer than expressed by others here.

I'm not sure I follow your logic here. The "non-uniformity" you speak of has been there all the time, why should it change the projected outcome?

Thinner ice obviously has a higher chance of melting out than thicker. When has that not been true? But there is so much more thin ice than thick ice now, that is a change that has been ongoing for some time now. So shouldn't we conclude the time frame will be shortened rather than extended?

But all this talk of linearity and trends assumes a resilient system undergoing slow but steady change. And so far the Arctic has behaved like that, and the increasingly chaotic weather is, at least not yet, the decisive factor that it may become.

Personally I feel that the resilience is being chipped away, eventually to crumble into chaos. Using warlike metaphors fitting for this weekend, the ice is like a fortress that rebuilds its defenses each winter, only to suffer an increasing onslaught each summer. So far the ice has managed to retreat, and pulling its defenses along with it. But there is only so far it can go before the defenses crumble and the foes manage to destroy the rest. And this might happen quite suddenly.

The best defense of the sea ice is it's sheer size - even at minimum it's huge, maintaining a frigid atmosphere above and a cold-water lens below. Even as the ice becomes more fractured and mobile, the icepack maintains it's integrity, i.e. it is basically continuous at minimum, and landfast.

Going back to our original starting point of the landfast ice north of the Canadian Archipelago - this ice has so far mostly moved along the shore but can't very well move away from shore, mainly because of resistance from the ice further out. But as extent drops the whole area may break up and start drifting away from the shore. Two million km2 of ice that is drifting freely in the Arctic for a few weeks will really take a battering, the most serious consequence perhaps being that wave action and dispersion will break up the protecting underlying cold water lens.

We saw an example of how this might happen this winter, when a strong southerly wind managed to push the ice tens or even hundreds of kilometers away from the shore of Greenland, in an area of thick ice that does not melt out during summer (although it is pushed towards the Fram strait to eventual doom).

So my feeling is that irregardless of current linear trends, when we get closer to some threshold, all metrics may well plummet quite fast, as the normal resilience of the Arctic Sea Ice gives in to encroaching chaos. Logically one might then guess that the sea ice might vanish before the first of the linear trends reaches zero - but perhaps not. Perhaps we will see a period where e.g. volume flattens out while extent continues its linear decrease.

But then again, a minimum extent of 4 million km2, which is what we may expect at present, is already very fragile as the extreme melt of 2012 shows, when a combination of clear skies in early summer and very strong storms towards the end of summer caused extent to plummet by 2 million km2 in just over one month.

But I admit, that this is just my opinion also, and the Arctic will do whatever it does, regardless of what we say it will.

So true ...

Archimid

  • ASIF Upper Class
  • Posts: 1135
    • View Profile
  • Liked: 17
  • Likes Given: 29
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #111 on: April 15, 2018, 03:12:58 PM »
Thanks for the graphs Archimid. But isn't the 46% figure the proportion of CAB area of the Arcic Ocean + some external areas, all of which freeze over almost completely (or to more or less the same extent) each winter?

I eliminated the Sea of Okhotsk, Hudson Bay and St Lawrence from the maximum and minimum. The justification is that all those area start their freezing from land, not from sea ice. Similarly, their melt has no bearing on any connected sea.

I then performed the same calculations for CAB volume as a percent of total volume. There is no discernible change in volume minimum as expected. Volume maximum now hovers at about 50% of the CAB. Like before it wants to remain at a constant amount of 50%.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #112 on: April 15, 2018, 09:53:53 PM »
We have plotted volume against extent. The pattern looks like a straight line that does not intersect the origin. However, a more likely scenario, a curve through the origin can easily look like a straight line when well away from the place where it curves.

Now where was that old discussion? hmm. ???

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #113 on: April 16, 2018, 03:38:46 PM »
Found this

Setting a power function to origin of zero, and fitting to scatter plot of CT Area as a function of PIOMAS minimum volume, gives a function:

CTArea = 1.287*PIOMASVol^0.5066.

Essentially a logarthimic curve.

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #114 on: April 16, 2018, 04:57:19 PM »
Found this

Setting a power function to origin of zero, and fitting to scatter plot of CT Area as a function of PIOMAS minimum volume, gives a function:

CTArea = 1.287*PIOMASVol^0.5066.

Essentially a logarthimic curve.
Might be fun to try! But what is meant by CT Area? Google isn't sure if it is Connecticut or something to to with Cthulhu (All hail!).

Neven

  • Administrator
  • ASIF Royalty
  • *****
  • Posts: 5342
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 101
  • Likes Given: 83
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #115 on: April 16, 2018, 05:55:45 PM »
CT, or Cryosphere Today, was one of the first website on the Internet offering all kinds of graphs, maps and daily data to the public.
Compare, compare, compare

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #116 on: April 16, 2018, 06:20:34 PM »
CT, or Cryosphere Today, was one of the first website on the Internet offering all kinds of graphs, maps and daily data to the public.

Well about arctic sea ice anyway.  ;) :P

also found




binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #117 on: April 16, 2018, 06:56:23 PM »
So "thickness" may well be a predictive metric ...

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #118 on: April 16, 2018, 06:57:19 PM »
Found this

Setting a power function to origin of zero, and fitting to scatter plot of CT Area as a function of PIOMAS minimum volume, gives a function:

CTArea = 1.287*PIOMASVol^0.5066.

Essentially a logarthimic curve.
Should presumably have been "PIOMAS maximum volume"

Steven

  • ASIF Citizen
  • Posts: 415
    • View Profile
  • Liked: 34
  • Likes Given: 8
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #119 on: April 16, 2018, 08:46:00 PM »
Found this

Setting a power function to origin of zero, and fitting to scatter plot of CT Area as a function of PIOMAS minimum volume, gives a function:

CTArea = 1.287*PIOMASVol^0.5066.

Here is an update of that formula.  The power function in the graph below is calculated from data for 1979-2017 (so there are 39 data points in total):


gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #120 on: April 16, 2018, 10:15:57 PM »

Here is an update of that formula.  The power function in the graph below is calculated from data for 1979-2017 (so there are 39 data points in total):



Impressive R2.
It looks like a linear formula would also give a good fit?
"Para a Causa do Povo a Luta Continua!"

jdallen

  • ASIF Governor
  • Posts: 2665
    • View Profile
  • Liked: 24
  • Likes Given: 59
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #121 on: April 16, 2018, 11:21:07 PM »

Here is an update of that formula.  The power function in the graph below is calculated from data for 1979-2017 (so there are 39 data points in total):



Impressive R2.
It looks like a linear formula would also give a good fit?
I think the relationship is highly linear until we drop down to very small values of volume at which point thicker ice shed from Greenland's and other glaciers skew the thickness distribution.

Ice thickness has a lot of moderating feedbacks which have tended to keep it around a median value.  Increased heat in the system has slowly but steadily pushed that symmetry point down, such that in the last two decades it has become possible for normal seasonal variation in thickening and loss to threaten a blue Arctic Ocean event.

The recent winter warmings I think are far more crucial to the changes we are seeing, ant that if we look back, we may actually see th first hints of it starting to manifest 20 or more years ago. It is that more than the melt which I think has dropped our median volume.  My rationale here is, year over year insulation doesn't really vary that significantly, and it's influence is greatly dependent on other factors - albedo being most key.

However, increased heat import from the Atlantic and water vapor, along with increased cloudiness have steadily reduced the ability of the ice to recover in winter.

It is quite possible we could have a repeat of 2013 or 2014 with clouds and cool conditions. But even with that, the steady export of warm Atlantic and now Pacific water will reach the basin steadily increasing the total energy available to the system, effectively lowering the pendulum another few notches. 

Eventually, nothing more than natural variation will cause it to scrape bottom.
This space for Rent.

kiwichick16

  • ASIF Lurker
  • Posts: 37
    • View Profile
  • Liked: 0
  • Likes Given: 1
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #122 on: April 17, 2018, 12:06:33 AM »
if the gulf stream continues to weaken ......or shutdown ......won't that significantly reduce warm water transport to the arctic?

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #123 on: April 17, 2018, 12:07:40 AM »

Here is an update of that formula.  The power function in the graph below is calculated from data for 1979-2017 (so there are 39 data points in total):



Impressive R2.
It looks like a linear formula would also give a good fit?
I think the relationship is highly linear until we drop down to very small values of volume at which point thicker ice shed from Greenland's and other glaciers skew the thickness distribution.


I think that is unlikely. Consider lots of individual ice floes, a relationship where the bigger the area, the higher the thickness. Some melting occurs, and some small ones disappear and all become less in area and less in thickness. So after a bit of melting, all the pieces are smaller but are essentially similar in their ratio of area to volume.

Try such a exercise with linear relationship of area to volume and you run into problems: The ice suddenly goes poof as zero volume is approached, but why, shouldn't the smaller pieces disappear before the bigger pieces?

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #124 on: April 17, 2018, 12:10:43 AM »
if the gulf stream continues to weaken ......or shutdown ......won't that significantly reduce warm water transport to the arctic?

Doesn't 'continue to weaken' imply that it already is slowing down and if it continues to weaken at the same pace, doesn't that mean this is already in the historical data which show sea ice declining?

jdallen

  • ASIF Governor
  • Posts: 2665
    • View Profile
  • Liked: 24
  • Likes Given: 59
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #125 on: April 17, 2018, 12:55:13 AM »
I think the relationship is highly linear until we drop down to very small values of volume at which point thicker ice shed from Greenland's and other glaciers skew the thickness distribution.


I think that is unlikely. Consider lots of individual ice floes, a relationship where the bigger the area, the higher the thickness. Some melting occurs, and some small ones disappear and all become less in area and less in thickness. So after a bit of melting, all the pieces are smaller but are essentially similar in their ratio of area to volume.

Try such a exercise with linear relationship of area to volume and you run into problems: The ice suddenly goes poof as zero volume is approached, but why, shouldn't the smaller pieces disappear before the bigger pieces?
I think our views are more similar than different.  I think to get "poofs" we will be at those low volume numbers to begin with - we will have to be.  I'm thinking sub 1000KM3 is where the relationship will start to break down.

Until then, sudden area losses will still I think have corresponding linear volume losses.
This space for Rent.

kiwichick16

  • ASIF Lurker
  • Posts: 37
    • View Profile
  • Liked: 0
  • Likes Given: 1

jdallen

  • ASIF Governor
  • Posts: 2665
    • View Profile
  • Liked: 24
  • Likes Given: 59
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #127 on: April 17, 2018, 02:28:32 AM »
@ crandles  # 124

https://cleantechnica.com/2018/04/15/its-the-gulf-stream-stupid-climate-scientists-warn-tipping-point-is-near/
Sorry, not impressed. Fluffy article high in histronics and low on substance, which misinterpretes what is happening.
This space for Rent.

kiwichick16

  • ASIF Lurker
  • Posts: 37
    • View Profile
  • Liked: 0
  • Likes Given: 1

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #129 on: April 17, 2018, 06:26:21 AM »
I've seen a couple of people mentioning glacier ice from Greenland and other places as being a factor in Arctic Sea Ice. In my opinion this is extremely unlikely. The glaciers facing the Arctic Ocean can't be calving more than a few square kilometers each year, and this ice gets grabbed by currents and pushed out the Fram strait.

The huge icebergs typical of Greenland glacier calving float southwards, but has anybody ever seen any significant number of them in the Arctic? I don't think so.

Juan C. García

  • ASIF Middle Class
  • Posts: 663
    • View Profile
  • Liked: 41
  • Likes Given: 35
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #130 on: April 17, 2018, 09:16:55 AM »
I've seen a couple of people mentioning glacier ice from Greenland and other places as being a factor in Arctic Sea Ice. In my opinion this is extremely unlikely. The glaciers facing the Arctic Ocean can't be calving more than a few square kilometers each year, and this ice gets grabbed by currents and pushed out the Fram strait.

The huge icebergs typical of Greenland glacier calving float southwards, but has anybody ever seen any significant number of them in the Arctic? I don't think so.

One of those people is me and I recognize that I don’t have a formal education in any science related to this topic. I had following the ASI for a decade and Neven’s Forum from the beginning (2013).

Making a summary: After 2012, some of us were waiting to have a collapse of the ASI in the following years. 2013 and 2014 were good years for increasing the ice, but 2015-2017 were the contrary. In 2016, the ASI extent was very low on the first semester, but at the end, nothing happened. Similarly, on 2017 the ASI volume was terrible, but at the end, nothing happened, for the second time. So, I started to think that there is a negative feedback that we were not contemplating. And I believe that this negative feedback is the ice bottom melting on Greenland.
As I said before, I could be wrong. I just believe that I am not wrong, but with not formal studies in this subject.

I made a topic covering this subject and answering your question, I believe the following answer:

Juan, check the currents around Greenland, they are all flowing southward, so Greenland melt is surely affecting the Atlantic ocean but I doubt it has much to do with low ASI melt in the Arctic.

In fact, all the Greenland currents are flowing southward, but then they clash with the Atlantic meridional overturning circulation (AMOC), maybe slowing the AMOC and turning northward.

The topic covering this ideas is:
"Ice melting (Antarctica, Greenland, etc.) as a short term ASI negative feedback"
https://forum.arctic-sea-ice.net/index.php/topic,2171.0.html
« Last Edit: April 17, 2018, 02:39:08 PM by Juan C. García »
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.

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #131 on: April 17, 2018, 09:44:31 AM »
I've seen a couple of people mentioning glacier ice from Greenland and other places as being a factor in Arctic Sea Ice. In my opinion this is extremely unlikely. The glaciers facing the Arctic Ocean can't be calving more than a few square kilometers each year, and this ice gets grabbed by currents and pushed out the Fram strait.

The huge icebergs typical of Greenland glacier calving float southwards, but has anybody ever seen any significant number of them in the Arctic? I don't think so.

One of those persons is me and I recognize that I don’t have a formal education in any science related to this topic. I had following the ASI for a decade and Neven’s Forum from the beginning (2013).
You make a valid point, and I'm sure your scientific grounding is no worse than mine. But my point was on the actual floes themselves (the icebergs) constituting a significant proportion of Arctic Sea Ice - which I seriously doubt.

But it's true, we do tend to underestimate the resilience of the ASI - looking at the various metrics we get the feeling every year that it's all about to collapse, but then it doesn't. My feeling is that the biggest source of this resilience is the extent itself - the minimum 4 million or so km2 is still a huge area and it isn't reached until the end of the melting season, and for a very short time.

Weather is probably by far the biggest contributor to noise in the various metrics, so a year like 2012 (where an unlikely combination of weather effects caused a major melt out) becomes an outlier, but in the longer run the systems underlying the various metrics just chug along on their gentle downward slope.

Until perhaps one year, when the average minimum extent is considerably lower than now (between 2 and 3 million km2?) with another usual combination of weather events, and the whole thing might just collapse, with the contiguous ice breaking up and floating away from the Canadian Arctic Archipelago and melting rapidly in a totally new Arctic Ocean.

If my feeling is correct, and the resilience is somehow proportional to, or rather, dependent upon, extent staying above a certain threshold, then all bets are off and our metrics will all nosedive, and any trend lines become obsolete.

oren

  • ASIF Upper Class
  • Posts: 2288
    • View Profile
  • Liked: 122
  • Likes Given: 259
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #132 on: April 17, 2018, 11:09:26 AM »
Making a summary: After 2012, some of us were waiting to have a collapse of the ASI in the following years. 2013 and 2014 were good years for increasing the ice, but 2015-2017 were the contrary. In 2016, the ASI extent was very low on the first semester, but at the end, nothing happened. Similarly, on 2017 the ASI volume was terrible, but at the end, nothing happened, for the second time.
Just to defend 2016's honor, although min extent wasn't that impressive (2nd but near 2011 and 2007) due to high dispersal, in terms of area it was quite far from all other years, and not very far from 2012. And it too had a GAC. 2016 had a terrible winter and spring, but then June/July were lucky, otherwise I believe the 2012 record would have been broken.


Steven

  • ASIF Citizen
  • Posts: 415
    • View Profile
  • Liked: 34
  • Likes Given: 8
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #133 on: April 17, 2018, 11:25:43 AM »
Impressive R2.
It looks like a linear formula would also give a good fit?

A linear fit would give basically the same R2 as the power function (about 0.92).  But the (extrapolated) linear fit doesn't pass through the origin (x=0, y=0).  In contract, the power function passes through the origin, expressing the fact that volume and area can only be zero simultaneously.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #134 on: April 17, 2018, 12:17:12 PM »
Impressive R2.
It looks like a linear formula would also give a good fit?

A linear fit would give basically the same R2 as the power function (about 0.92).  But the (extrapolated) linear fit doesn't pass through the origin (x=0, y=0).  In contract, the power function passes through the origin, expressing the fact that volume and area can only be zero simultaneously.

But is using the power function a better indication of what will happen? My speculation is, and always has been, that volume and extent/area will decline in a more or less orderly linear manner until volume/thickness becomes low enough for a 2012-type summer to cause a disorderly collapse of much of the remaining ice. i.e. both functions only work to a lower limit of n km3, where n is unknown but less than 4,000 and greater than 1,000.

What happens the following winter / summer? Not a clue. But this poll is about a one day low of < 1,000 km3. I am sticking with 2028.
"Para a Causa do Povo a Luta Continua!"

Archimid

  • ASIF Upper Class
  • Posts: 1135
    • View Profile
  • Liked: 17
  • Likes Given: 29
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #135 on: April 17, 2018, 01:05:50 PM »
Quote
A linear fit would give basically the same R2 as the power function (about 0.92).  But the (extrapolated) linear fit doesn't pass through the origin (x=0, y=0).  In contract, the power function passes through the origin, expressing the fact that volume and area can only be zero simultaneously.

This implies that ice thickness will increase as the ice reaches 0? I don't see why. The CAB already makes up 95% of the volume at minimum. When volume reaches 0 extent will definitely be 0. However when volume is at 0+epsilon area will be relatively large. The only way to avoid that is by thickening the ice as extent disappears.  That makes no sense.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #136 on: April 17, 2018, 01:10:58 PM »
Wonder if anyone wants to update this graph



Is there still a sweet spot somewhere that neither curves up nor down?

Note even if there is, there must be some good chance that melt from [31 May or whenever] will start curving one way or the other and we just haven't seen it yet.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #137 on: April 17, 2018, 02:46:45 PM »
I've seen a couple of people mentioning glacier ice from Greenland and other places as being a factor in Arctic Sea Ice. In my opinion this is extremely unlikely. The glaciers facing the Arctic Ocean can't be calving more than a few square kilometers each year, and this ice gets grabbed by currents and pushed out the Fram strait.

The huge icebergs typical of Greenland glacier calving float southwards, but has anybody ever seen any significant number of them in the Arctic? I don't think so.

One of those persons is me and I recognize that I don’t have a formal education in any science related to this topic. I had following the ASI for a decade and Neven’s Forum from the beginning (2013).
You make a valid point, and I'm sure your scientific grounding is no worse than mine. But my point was on the actual floes themselves (the icebergs) constituting a significant proportion of Arctic Sea Ice - which I seriously doubt.


GREENLAND MELT - EFFECT ON ARCTIC SEA ICE
I found this :-
https://link.springer.com/content/pdf/10.1007%2Fs40641-017-0070-1.pdf
Glacier Calving in Greenland

Extract follows:-
Quote
Contribution of Calving Glaciers to Greenland’s Mass Budget

Dynamic ice loss is a major factor in the mass balance of the Greenland Ice Sheet. In 2000, the total discharge from all of Greenland’s tidewater outlet glaciers was 462 ± 6 Gt [1].

Fifteen glaciers make up 50% of the total, and only five (Jakobshavns Isbrae, Kangerdlugssuaq, Koge Bugt, Ikertivaq South, and Helheim) account for > 30%. Total discharge rate increased by 18% to 546 ± 11 Gt/year between 2000 and 2012, but inter-annual variability is high, particularly in SE Greenland . Dynamic ice losses were roughly equal to surface melt and runoff in the period 2000–2008, but the share was somewhat less during the exceptionally high-surface melt years of 2009–2012.

Taken together, dynamic ice losses and runoff of surface meltwater exceed snow accumulation over the Greenland Ice Sheet, so the overall mass balance of the ice sheet is negative.
Analysis of surface elevation and mass changes from CryoSat-2 and GRACE satellite data indicates that the Greenland Ice Sheet had a net annual mass balance of − 269 ± 51 Gt/year from Jan 2011–Dec 2014.

Dynamic ice loss from marine outlet glaciers makes a large contribution to this deficit, particularly from Kangerdlugssuaq in the east, Jakobshavns Isbrae, Upernavik Isstrøm and Steenstrup
Glacier on the west coast, and Zachariæ Isstrøm in the north-east.

From this my best guess is:-
- As Greenland is drier and colder to the North and East, and much wetter and warmer to the South and West, surface melt (May to August) must be greatest into the Greenland Sea,
- most glaciers are melting into the Nares Strait, Baffin Bay and the Greenland Sea.

If Greenland mass balance loss continues to increase, the direct effect on Arctic ice must be minimal, but the effect on the AMOC could be significant - all that warmer, less dense freshwater.
"Para a Causa do Povo a Luta Continua!"

Steven

  • ASIF Citizen
  • Posts: 415
    • View Profile
  • Liked: 34
  • Likes Given: 8
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #138 on: April 17, 2018, 03:16:37 PM »
This implies that ice thickness will increase as the ice reaches 0?

No, it doesn't.  Quite the contrary.  Note that the exponent in the power function is 0.5136, which is very close to 0.5.  So it's basically a square root function, i.e. area behaves like the square root of volume (multiplied by some unimportant constant).  This implies that volume goes to zero much faster than area.  For example, if area is halved, then only one fourth of the original volume remains.  So, based on that power function extrapolation, the average ice thickness would continue to decrease as the ice goes to zero.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #139 on: April 17, 2018, 03:21:46 PM »
And here is a map showing where Greenland is losing mass from this article :- http://www.pnas.org/content/109/49/19934
"Para a Causa do Povo a Luta Continua!"

binntho

  • ASIF Citizen
  • Posts: 154
    • View Profile
  • Liked: 5
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #140 on: April 17, 2018, 03:42:27 PM »
- As Greenland is drier and colder to the North and East, and much wetter and warmer to the South and West, surface melt (May to August) must be greatest into the Greenland Sea,
- most glaciers are melting into the Nares Strait, Baffin Bay and the Greenland Sea.

If Greenland mass balance loss continues to increase, the direct effect on Arctic ice must be minimal, but the effect on the AMOC could be significant - all that warmer, less dense freshwater.

I think you are quite right. Annual melt is in the region of 500 km3 and approximately half of which is due to calving, according to the extract. The vast majority of those calvings happen along the east and west coasts of Greenland and do not enter the Arctic Basin. But even those few calving that happen along the north coast eventually get swept away towards the Fram or Nares Straits.

Archimid

  • ASIF Upper Class
  • Posts: 1135
    • View Profile
  • Liked: 17
  • Likes Given: 29
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #141 on: April 17, 2018, 04:53:11 PM »
This implies that ice thickness will increase as the ice reaches 0?

No, it doesn't.  Quite the contrary.

I see it now. Thanks.
I am an energy reservoir seemingly intent on lowering entropy for self preservation.

DavidR

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 1
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #142 on: April 18, 2018, 10:59:21 AM »
Wonder if anyone wants to update this graph
<snip>
Is there still a sweet spot somewhere that neither curves up nor down?

Note even if there is, there must be some good chance that melt from [31 May or whenever] will start curving one way or the other and we just haven't seen it yet.

This is not quite up to date but you get  the general impression. All trendlines are simply polynomials extrapolated out to 2025.

I have added the Max and total melt numbers to clarify the picture further. 
Toto, I've a feeling we're not in Kansas anymore

DavidR

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 1
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #143 on: April 18, 2018, 11:22:13 AM »
An alternative view is to look at the gains and losses each year. This graph shows that although some years have greater gains than losses, on average the losses are larger than the gains by about 300 km^2 per year. This gap appears to be accelerating slightly probably as a consequence of Arctic waters warming up.
Toto, I've a feeling we're not in Kansas anymore

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #144 on: April 18, 2018, 12:18:59 PM »
Thank you very much.  :D

(31 May volume - 15) looks like it still works pretty well.

Maybe marginally curving up, but at some point I would expect it to curve downwards like the 30 June to Min curve.

DavidR

  • ASIF Middle Class
  • Posts: 659
    • View Profile
  • Liked: 8
  • Likes Given: 1
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #145 on: April 19, 2018, 11:00:24 AM »
Thank you very much.  :D

(31 May volume - 15) looks like it still works pretty well.

Maybe marginally curving up, but at some point I would expect it to curve downwards like the 30 June to Min curve.
The linear trend is an increase of  +15 km^3 (edited) per year and the polynomial trend is currently increasing. However if current trends hold we are only a few years away  from it being impossible to lose more than 15K after 31 May. The figures are only relevant until Sept volume reaches 0 which I and many others are suggesting will occur in about 5 years.
« Last Edit: April 20, 2018, 10:22:42 AM by DavidR »
Toto, I've a feeling we're not in Kansas anymore

litesong

  • ASIF Citizen
  • Posts: 111
    • View Profile
  • Liked: 5
  • Likes Given: 0
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #146 on: April 19, 2018, 04:52:41 PM »
I have been using the minimum volume, but I suspect that is not much difference from the September average.  In either case, it comes down to whether you feel that the higher volume loss in the first decade of this century or the slower volume loss since will predominate. 


Like you, I use Volume loss, thinking Volume is a better measure of conditions that are reducing Arctic sea ice. Recent studies on Antarctic sea ice with underwater drones indicate Southern Ocean warming currents are cutting large undersea "upside down canyons" in bottoms of Ice Shelves. Warming Arctic Ocean currents are helping cause Arctic Volume losses also. 

crandles

  • ASIF Upper Class
  • Posts: 1963
    • View Profile
  • Liked: 5
  • Likes Given: 9
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #147 on: April 20, 2018, 04:06:27 PM »
Thank you very much.  :D

(31 May volume - 15) looks like it still works pretty well.

Maybe marginally curving up, but at some point I would expect it to curve downwards like the 30 June to Min curve.
The linear trend is an increase of  +15 km^3 (edited) per year and the polynomial trend is currently increasing. However if current trends hold we are only a few years away  from it being impossible to lose more than 15K after 31 May. The figures are only relevant until Sept volume reaches 0 which I and many others are suggesting will occur in about 5 years.

Here are polynomial and 3 parameter gompertz fit to 31 May PIOMAS volume. These trendlines are going well below reality for last 5 years. Data doesn't justify 4 parameter gompertz (inflection point not til 2024). Maybe some other curve function is needed. Maybe the last 5 years values are oddly all above trend.

I think I should concede that the trend in May 31 volume isn't as leveling out as much as I expected and it may not be all that long before May 31 Volume reaches 15,000 km^3. (2020 - 2025 being plausible)

Even 31 July to minimum doesn't seem to be curving downwards much if at all. Obviously it has to curve downwards when there isn't that much left but it seems entirely possible that it could curve downwards before then but there is no evidence for that from what I have looked at.

gerontocrat

  • ASIF Upper Class
  • Posts: 2305
    • View Profile
  • Liked: 128
  • Likes Given: 3
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #148 on: April 20, 2018, 05:41:08 PM »
"Linear Regression rule, OK?"
Sorry, non-scientific interjection
"Para a Causa do Povo a Luta Continua!"

Stephan

  • ASIF Lurker
  • Posts: 50
    • View Profile
  • Liked: 11
  • Likes Given: 5
Re: First year PIOMAS volume below 1000km^3 poll 2018
« Reply #149 on: May 08, 2018, 10:02:58 PM »
I have listed all extent and volume data monthly from 1979 to now. As extent and volume come from different sources, the calculated thickness may not be realistic or true. But, anyway, I have treated the data both with linear and logarithmic trends and calculated, when the September average will reach zero. It makes a huge difference whether you use extent, volume, or thickness. The kind of fit (linear or logarithmic) is much less important, nevertheless the log fit always gives an earlier year, when zero is reached. Here are the results:

Extent: zero (lin) 2076 --- zero (log) 2066
Volume: zero (lin) 2032 --- zero (log) 2025
Thickness: zero (lin) 2050 --- zero (log) 2045

From these data it is clear that the volume is the big issue when it comes to the first blue ocean event. Of course, the behaviour of an overall very small ice thickness in the CAB in some years, which probably leads to a faster melt-out if weather conditions are the right ones, cannot be included in the historical data.