Ice-free Arctic

[oren]

The arctic ocean will become seasonally ice-free, same as the peripheral seas today, long before it becomes ice-free in winter. Very long. Can't say much more than that, as too much is unknown.

[crandles]

and it will refreeze.

I disagree with you here.  Open water is subject to physical processes that ice covered water isn't.  It is subject to mixing by wave action.  The arctic ocean has a low salinity top layer that limits heat available to reject to the atmosphere.  Mix that layer in with the warmer lower water and there is plenty of heat to keep it from refreezing.

I am unsurprisingly with Daniel B.

If the temp was likely to stay at -10C or higher throughout winter and add in lots of strong storms to stir up the water and you might get severely reduced ice formation. (There is quite a bit of shallow water that loses any heat throughout the water column so you can't stir up heat from 50m down because it is shallower than 50m.)

From 1970 to 2017 temperatures in winter have increased from somewhere about -28C to about -23C so at that rate it is about 100 years before we reach -10C. (rough figures by eye from http://ocean.dmi.dk/arctic/meant80n.uk.php )

In addition, there is bound to be some periods and areas between storms where it is still wavy but not enough to stir heat up from 50m down. If the temp is below -10C plenty of tiny ice platelets (frazil ice) form and enough of these will damp down the waves allowing further stages of ice formation to be reached.

Even if your mechanism did work one year, so much heat would be lost from below 50m deep, the next winter there would be less heat at this depth where it is needed for this to occur that it is hard to see it being sustainable for more than a couple of winters and when it fails there would be less upward heat flux such that more ice than normal would form.

[Gray-Wolf]

I think it was last year that one of us approached NSIDC over the 15% cutoff and the issues it places on the measure of a highly fragmented 'inner pack'?

The reply was along the line of 'we know there is an issue.The measure evolved to better capture the peripheral ice and was not set up to measure internal ice'.

Basically , in the past, the ice beyond 80N would have been contiguous ice. Now it can be 16% ice covered yet be measured as it used to as 'contiguous ice'. This is a nonsense!

The increasingly open pack in the high Arctic ( since 2012) means the final ice figures are increasingly incorrect as we move down to the 15% cut off.

This notion of " when the ice goes it will go pretty fast" is even more true if we see areas 'blink out' that have been reducing down to the 15% cut off at seasons end for years.

[crandles]

The trend in volume melt has been upward. Once we reach completely ice free in summer every year, the volume melt has to decline as maximum volume declines.

The immediate natural assumption is that as max volume goes down the thickness goes down and this changes to low albedo faster so more volume melts, thus the maximum volume melt occurs as we reach seasonally completely ice free.

However, there is the possibility that a max volume melt was reached in 2012. The FYI that melts out every year is now getting thinner so there is less volume melt in these areas as time goes on. This is to some extent compensated by earlier melt further into the pack. However the area that is further into the pack is getting smaller. As this gets smaller, it might be possible that the extra melt volume in these area further into the pack is less and less able to compensate for the less melt of the FYI that melts out every year.

The trend in melt volume looks upward so there is little if any trend data to support the idea that volume melt max has passed. However there were some experiments with models reducing the ice thickness to 1m. While a lot melted, there was still some ice left at the minimum. This potentially supports the idea that the max melt volume might well be before we reach seasonally ice free.

Reactions to this? (probably not hard to guess)

[oren]

Reactions to this? (probably not hard to guess)

Actually makes a lot of sense. Lack of significant MYI and low winter maxima mean less ice available for melting. Indeed the most probable conclusion is that max volume melt will be in the first year of a BOE (if it wasn't in 2012).

[Archimid]

Considering that the minimum has not declined over the past four years, it is hard to say.  The minimum is not a direct function of the maximum.  If it were, the past four years would be the lowest, and not 2012 and 2007.

If initial conditions include Arctic sea ice, then minimum volume and maximum volume are both mostly functions of the tilt of the Earth and its orbit around the Sun.  Ice volume is a direct function of available heat in the system and most of the heat in the system is ultimately provided by the sun. The earth's climate generates the noise. That noise provides us with ceilings and floors for the total heat available to melt ice or freeze water.

Losses and gains of annual volume are a direct representation of all the heat available to the ice after the noise of the climate system.  The low and high bounds for the heat available to Arctic are given by the Minimum and Maximum of the gains and losses. Everything in between is the climate. Here is the breakdown of the noise:

 The first attachment has a chart for volume losses and gains from 1980-2006. Here are the stats:

                 Gain         Loss
Average   16.14   16.43
Median   16.3           16.33
Maximum   17.62   18.31
Minimum   14.71   13.93
SD           0.71           1.01


 The average volume loss was 16.43 and the average gain was 16.14. Gains and losses have similar variation, both relatively low, with the gains having the lowest noise.

Then the minimum collapsed in 2007. See the second attachment for a chart with volume loses and gains from 2007-2017. Here are the stats:

                Gain         Loss
Average   17.69   18.09
Median   17.73   18.24
Maximum   19.66   19.69
Minimum   14.87   16.25
SD       1.72   1.27

After 2007 both gains and losses increased significantly and the losses increased more than the gains.  The lower bound for the gains remained about the same and the upper bound of gains increased significantly. The lower bound for the losses increased significantly,  matching the old average losses.

Variation increased slightly for the losses, but look at the variation of the gains. The variation of the gains more than doubled.

 

The ultra low minimum in 2012, was surrounded by relatively high maxima both prior to and following the summer minimum (high for the past fifteen years).  While intuitively, it would makes sense that the minimum would follow a falling maximum, in reality, that is not occurring.

2012 is preceded by a near record low maximum, 2007 and 2016 are preceded by record low maximums. At current volumes a mere average loss puts the Arctic at record low levels. Maximum losses crash the minimum to never before seen levels.  See the third attachment.

[Archimid]

The arctic ocean will become seasonally ice-free, same as the peripheral seas today, long before it becomes ice-free in winter. Very long. Can't say much more than that, as too much is unknown.

I think that ice volume will follow the path suggested by the following visualization by Jim Pettit



Once volume reaches 0 the loops become very small for a few more roundabouts and then for all intent and purposes the ice will be gone. It will remain that way until we remove all that extra CO2 and methane from the atmosphere.

[Archimid]

If the temp was likely to stay at -10C or higher throughout winter and add in lots of strong storms to stir up the water and you might get severely reduced ice formation. (There is quite a bit of shallow water that loses any heat throughout the water column so you can't stir up heat from 50m down because it is shallower than 50m.)

That is the expected outcome. After the last bout of healthy growth in 2013-2014 much of the endogenous cold of the Arctic was used up. Since then growth has been anemic and matched by extremely high temperatures.

From 1970 to 2017 temperatures in winter have increased from somewhere about -28C to about -23C so at that rate it is about 100 years before we reach -10C. (rough figures by eye from http://ocean.dmi.dk/arctic/meant80n.uk.php )

That is extremely misleading. Most of that increase happened just in the last three years. for the rest of the datase the DMI temperatures are very close to the average. So when yo project into the future, you have 36 data points with almost normal temperatures and the last 3 with all the heat. Can you please make a sanity check  to that statement? 

In addition, there is bound to be some periods and areas between storms where it is still wavy but not enough to stir heat up from 50m down. If the temp is below -10C plenty of tiny ice platelets (frazil ice) form and enough of these will damp down the waves allowing further stages of ice formation to be reached.

Why this doesn't happen now? All growth of ice that I have seen comes from the ice edge. not from random cold spots above the ocean.

Even if your mechanism did work one year, so much heat would be lost from below 50m deep, the next winter there would be less heat at this depth where it is needed for this to occur that it is hard to see it being sustainable for more than a couple of winters and when it fails there would be less upward heat flux such that more ice than normal would form.

You speak as if the heat loss by the ocean magically disappears. That heat lost by the oceans goes into resulting in warmer atmosphere , which results in less FDD's which results in less ice.

Look at the Chukchi right now. Where was that extra growth that you claim it will happen? We saw the exact opposite. Anemic ice growth and high atmospheric temperatures. 

[Archimid]

The trend in volume melt has been upward. Once we reach completely ice free in summer every year, the volume melt has to decline as maximum volume declines.

Completely wrong. If we take the whole data set the rate of volume melt is increasing. If we take the trend since 2007 until the end then the trend is slightly decreasing but the lowest lows are equal to the average of the losses in the past. and the average loses are higher than the average of the past. Melting is outpacing freezing.

However, there is the possibility that a max volume melt was reached in 2012. The FYI that melts out every year is now getting thinner so there is less volume melt in these areas as time goes on.

 

Yes but the available heat in the system is the same. If it doesn't go to melt ice it must go somewhere else. Right now it looks like a good chunk of it is going into the atmosphere above the Arctic

This is to some extent compensated by earlier melt further into the pack. However the area that is further into the pack is getting smaller. As this gets smaller, it might be possible that the extra melt volume in these area further into the pack is less and less able to compensate for the less melt of the FYI that melts out every year.

Earlier melt means more heat over less ice, meaning there is more heat available to melt the center of the pack.

The trend in melt volume looks upward so there is little if any trend data to support the idea that volume melt max has passed.

Even if a peak in melt(losses) was reached, Max volume is still dropping. Right now an average volume loss for the last 11 years puts us at near record low volume while a repeat of a Max volume loss puts us at record low levels. If such event happen, will the gains recover to 2012 levels? The warming experienced in the Arctic over the last three years indicate no.

However there were some experiments with models reducing the ice thickness to 1m. While a lot melted, there was still some ice left at the minimum. This potentially supports the idea that the max melt volume might well be before we reach seasonally ice free.

It doesn't matter.

[Archimid]

Reactions to this? (probably not hard to guess)

Actually makes a lot of sense. Lack of significant MYI and low winter maxima mean less ice available for melting. Indeed the most probable conclusion is that max volume melt will be in the first year of a BOE (if it wasn't in 2012).

Yes but if there is less ice to melt, where all the heat that was supposed to melt that ice goes?

[oren]

Reactions to this? (probably not hard to guess)

Actually makes a lot of sense. Lack of significant MYI and low winter maxima mean less ice available for melting. Indeed the most probable conclusion is that max volume melt will be in the first year of a BOE (if it wasn't in 2012).

Yes but if there is less ice to melt, where all the heat that was supposed to melt that ice goes?

Eventually, when a BOE is reached and no more summer ice is available to melt, SSTs will rise.
But regardless of all this, even when the arctic is a real ocean with no grain of ice in sight, you will still get winter ice growth at some point for a long time to come. Currently ice typically grows from the vicinity of more ice (not always!) because that vicinity has fresher surface water, colder SSTs, less waves, and colder air temps. But during the long polar night, even with waves and warm autumn SSTs and whatever, a calm cold day will come at some point with -30 or -25 or -20 degC, and the surface will freeze into ice. That is just plain weather.

[crandles]

From 1970 to 2017 temperatures in winter have increased from somewhere about -28C to about -23C so at that rate it is about 100 years before we reach -10C. (rough figures by eye from http://ocean.dmi.dk/arctic/meant80n.uk.php )

That is extremely misleading. Most of that increase happened just in the last three years. for the rest of the datase the DMI temperatures are very close to the average. So when yo project into the future, you have 36 data points with almost normal temperatures and the last 3 with all the heat. Can you please make a sanity check  to that statement? 

If it was just the last three years I would be writing that off to natural variation. It takes place over much longer than that. However you do have a point, I just said 1970 as well known date that global temperatures started to rise after 1940-1970 aerosol induced hiatus. 2008 and 2004 were as low as the early years. Perhaps 1985-1989 might be the lowest 5 years average but then 1990 was surprisingly warm. So they are a bit all over the place. It would be nice to see lines for decadal averages rather than one long period average. Not sure if someone found the data to do that at some point. My impression is the cold years generally get less frequent over time.

Perhaps I should only be attributing that 5C change to last 25 years rather than 37 and maybe there is some acceleration, but it seems hard to tell. Maybe someone has a better data set?

Re "After the last bout of healthy growth in 2013-2014 much of the endogenous cold of the Arctic was used up."

Not sure what you are saying on a few counts:
1) Of the last 10 years, only the last years volume freeze was below average. 2013 and 2014 were high but how can you attribute this to anything other than random variation?
2) What do you mean by endogenous cold?
3) "used up"? Huh? There is no sunlight every winter. The temperature rises come from a) heat stored in the ocean built up over summer, b transported by winds & currents, and c more GHG in atmosphere reducing rate of emission from atmosphere to space. I don't see any reason to think any of these will do anything other than slowly increase over long periods of time.

>"Why this doesn't happen now? All growth of ice that I have seen comes from the ice edge. not from random cold spots above the ocean."

Why would you get a cold spot above the ocean, unless there was wind coming from a direction where there was ice or snow. The further it gets from that ice or snow the more it has warmed up and is less able to cool the water sufficiently. We don't really have resolution to see small amounts of ice forming and match it to openings in clouds which allow faster radiation loss to space.

>"You speak as if the heat loss by the ocean magically disappears"
A small amount goes direct from ocean to space. Most is from ocean to atmosphere and then atmosphere to space. No idea why you are thinking I am thinking something other than that.

[Archimid]

Oren I 100 percent agree with you. Ice will form in winter no matter what. The Arctic night is very long. And regardless how warm it gets there is Greenland there providing low albedo and fresh water.

However, the amount of ice will be large enough to make any difference when the sun comes back.

After the ice is gone for the first time, I bet growth starts north of Greenland and grows out from there.  Because the bathymetry of the Arctic and the cold arctic night it is likely that ice grows very fast over the arctic basin but then slows down due to very high Arctic temperatures.

At the end of the freezing season there will a very small amount of ice in the Arctic, literally all of it thin first year ice. When the Sun comes back and the melting season begins that ice will be gone record fast and all the heat that is normally in the system that didn't get to melt ice stays in the Arctic, warming it and delaying the formation of ice the next winter.

While the Greenland ice sheet is there, there will be at least some form of ice in the Arctic. Hopefully, that's a long time.

[crandles]

Reactions to this? (probably not hard to guess)

Actually makes a lot of sense. Lack of significant MYI and low winter maxima mean less ice available for melting. Indeed the most probable conclusion is that max volume melt will be in the first year of a BOE (if it wasn't in 2012).

Yes but if there is less ice to melt, where all the heat that was supposed to melt that ice goes?

Yes it warms ocean during melt season. Come winter, it gets vented to atmosphere warming atmosphere. Then if there is more heat in the atmosphere there is a greater temperature difference to space's 4K temp so there is a slightly higher rate of heat loss to space. With this higher rate of heat loss there is plenty of time in fall & winter to radiate it to space and then start forming ice. OK forming ice gets delayed slightly and the ice ends up slightly thinner as a result but no one is denying this.

[Daniel B.]

Oren I 100 percent agree with you. Ice will form in winter no matter what. The Arctic night is very long. And regardless how warm it gets there is Greenland there providing low albedo and fresh water.

However, the amount of ice will be large enough to make any difference when the sun comes back.

After the ice is gone for the first time, I bet growth starts north of Greenland and grows out from there.  Because the bathymetry of the Arctic and the cold arctic night it is likely that ice grows very fast over the arctic basin but then slows down due to very high Arctic temperatures.

At the end of the freezing season there will a very small amount of ice in the Arctic, literally all of it thin first year ice. When the Sun comes back and the melting season begins that ice will be gone record fast and all the heat that is normally in the system that didn't get to melt ice stays in the Arctic, warming it and delaying the formation of ice the next winter.

While the Greenland ice sheet is there, there will be at least some form of ice in the Arctic. Hopefully, that's a long time.

Good.  Something else upon which we agree.  The ice north of Greenland and around the Canadians islands will remain as long as the glacial ice remains on those landforms (hopefully a very long time).  Hence the term ice-free refers to less than 1 million sq. km., and not zero ice, as scientists readily admit that this ice will not melt easily.  The ice is likely to spread rapidly each winter, due to the absence of sunlight, and is likely to melt rapidly, due to its fragility.  This could lead to large inter-annual variability, as small changes in incoming solar radiation, cloudiness, ocean temperature and currents, etc. could have large implications. 

For these reasons, neither ice volume nor extent are likely to ever reach zero.  In reality, they will probably fall to some value, dependent on the adjacent land ice.  We should not get overly concerned about the 1 million number, as it is somewhat arbitrary.  The actual value may be several hundred thousand higher or lower, but represents the ice grounded to the adjacent land glaciers plus some surrounding floating ice.

[Archimid]

1) Of the last 10 years, only the last years volume freeze was below average. 2013 and 2014 were high but how can you attribute this to anything other than random variation?

I attribute that growth to the ice being very thin and the atmosphere almost as cold as average.

2) What do you mean by endogenous cold?

For example the temperature of the water under the Chukchi sea or the Beaufort. For most of the record those areas were almost completely covered in ice  year round keeping them nice and cold. I imagine (can't find data) that since the 2007 cover changes the areas with significant cover changes are becoming warmer. That effect was barely noticeable from 2007-2014, but it accumulates By 2015 those cold spots lost their "endogenous cold" and no longer help with the freezing. The same happens with hot air intrusions. Once the endogenous cold of the Arctic kept the Arctic dry and the heat from the hemisphere away. But as the arctic warms, warmth is allowed in, warming it more.

3) "used up"? Huh? There is no sunlight every winter. The temperature rises come from a) heat stored in the ocean built up over summer, b transported by winds & currents, and c more GHG in atmosphere reducing rate of emission from atmosphere to space. I don't see any reason to think any of these will do anything other than slowly increase over long periods of time.

a) heat stored in the ocean built up over summer:

Like the Chukchi this year. It was hot and releasing heat to the atmosphere until very late in the freezing season. Eventually it covered in ice but I bet there is still extra heat underneath the ice. From the looks of thing this year the pattern will be reinforced.

b)  transported by winds & currents

And that has increased over the years. The less ice, the more humid and the more warm air intrusions.

 c) more GHG in atmosphere reducing rate of emission from atmosphere to space

I think that can best be appreciated in the baseline of winter temperatures in DMI N80. Even in winter temperatures seem to have a new baseline. that new baseline is induced by higher arctic water vapor and clouds who are bound to increase even more as more ocean remains open for longer. Hot air intrusions also add to the new baseline.


>"Why this doesn't happen now? All growth of ice that I have seen comes from the ice edge. not from random cold spots above the ocean."

Why would you get a cold spot above the ocean, unless there was wind coming from a direction where there was ice or snow. The further it gets from that ice or snow the more it has warmed up and is less able to cool the water sufficiently. We don't really have resolution to see small amounts of ice forming and match it to openings in clouds which allow faster radiation loss to space.

>"You speak as if the heat loss by the ocean magically disappears"
A small amount goes direct from ocean to space. Most is from ocean to atmosphere and then atmosphere to space. No idea why you are thinking I am thinking something other than that.
[/quote]

Because if you are citing high ice growth rates because the oceans will vent heat. However high growth rates are also dependent on atmospheric temperatures. If the ocean vents heat to the atmosphere, the atmosphere will warm precluding ice formation. Also much of that heat will be in the form of water vapor, which serves to keep the warmth in during the winter night.

I guess that what I'm trying to say is that the fast growth you are counting on won't happen because it will simply be too warm for rapid ice growth.

[Archimid]

Yes it warms ocean during melt season.

It warms the ocean and the atmosphere.

Come winter, it gets vented to atmosphere warming atmosphere.

and that venting gets slowdown by the already warm and humid atmosphere, prolonging the time the atmosphere is warm.

Then if there is more heat in the atmosphere there is a greater temperature difference to space's 4K temp so there is a slightly higher rate of heat loss to space.

only if there are few clouds. High humidity might keep much of that heat in.

With this higher rate of heat loss there is plenty of time in fall & winter to radiate it to space and then start forming ice. OK forming ice gets delayed slightly and the ice ends up slightly thinner as a result but no one is denying this.

And that's the whole point. You start the freezing season late, from 0 and with a much warmer atmosphere. By the time the sun comes back it will eat all the ice very quick, allowing even more sun in the following year. Warming the arctic even more and further delaying the onset of freezing.

[Archimid]

Hence the term ice-free refers to less than 1 million sq. km., and not zero ice, as scientists readily admit that this ice will not melt easily.

I don't think that assumption is correct. Bottom melt is real, so the whole Arctic basin is vulnerable to disappear. It hasn't dissapeared yet because the thickest ice is there and the heat of the whole melting season is wasted melting the periphery. As the periphery weakens, the heat keeps encroaching the basin.

I believe that when we get a large loss year the CAB will poof out of existence very quickly. Thi sis suggested by the fact that volume hits 0 much earlier than extent, suggesting that the last bit of extent will poof out.

  The ice is likely to spread rapidly each winter, due to the absence of sunlight, and is likely to melt rapidly, due to its fragility.

 

The ice is likely to spread rapidly after a delay that could last months. Even when it will be extremely fast at first, even fast enough to pull a reverse poof, because the hot Arctic temperature ice growth will eventually slow to a crawl.

This could lead to large inter-annual variability, as small changes in incoming solar radiation, cloudiness, ocean temperature and currents, etc. could have large implications. 

We already have increased annual variability. after the ice s gone the variability will decrease in terms of there being a lot less ice made and a lot less ice melted.

For these reasons, neither ice volume nor extent are likely to ever reach zero.  In reality, they will probably fall to some value, dependent on the adjacent land ice.  We should not get overly concerned about the 1 million number, as it is somewhat arbitrary.  The actual value may be several hundred thousand higher or lower, but represents the ice grounded to the adjacent land glaciers plus some surrounding floating ice.

A few rounds after the first BOE there will be literally 0 ice in the Arctic and temperatures of 20C maybe even 30C. However the winter night will definitely keep dropping below 0, most likely below -10. Some trivial ice will form every year.

[Shared Humanity]

The arctic ocean will become seasonally ice-free, same as the peripheral seas today, long before it becomes ice-free in winter. Very long. Can't say much more than that, as too much is unknown.

I agree. If we use the same standard for ice free as during melt season (1 million sq km) then an ice free winter is way out into the future. This is not to say that the winter ice, at max, will look anything like 30 years ago or even today but the ocean in the polar night will freeze.

[Shared Humanity]

I think it was last year that one of us approached NSIDC over the 15% cutoff and the issues it places on the measure of a highly fragmented 'inner pack'?

The reply was along the line of 'we know there is an issue.The measure evolved to better capture the peripheral ice and was not set up to measure internal ice'.

Basically , in the past, the ice beyond 80N would have been contiguous ice. Now it can be 16% ice covered yet be measured as it used to as 'contiguous ice'. This is a nonsense!

The increasingly open pack in the high Arctic ( since 2012) means the final ice figures are increasingly incorrect as we move down to the 15% cut off.

This notion of " when the ice goes it will go pretty fast" is even more true if we see areas 'blink out' that have been reducing down to the 15% cut off at seasons end for years.

And this is absolutely relevant to when the Arctic becomes seasonally ice free but not year round ice free.

[Shared Humanity]

After the ice is gone for the first time, I bet growth starts north of Greenland and grows out from there.  Because the bathymetry of the Arctic and the cold arctic night it is likely that ice grows very fast over the arctic basin but then slows down due to very high Arctic temperatures.

We are already seeing this in the Arctic. The drastically reduced ice, huddles near the CAA at minimum and, during the freeze season, grows rapidly outward. We also see ice form along land across Siberia even though it is essentially ice free at the end of the melt season. Meanwhile, we have vast expanses of open water soaking up heat during the melt season. This behavior will continue IMHO and the Arctic will continue to freeze during the polar night for a long time into the future. Yes, it will be mainly 1st year ice and we may, after the 1st BOE, see this occur often at minimum.

[Shared Humanity]

I think that ice volume will follow the path suggested by the following visualization by Jim Pettit



Once volume reaches 0 the loops become very small for a few more roundabouts and then for all intent and purposes the ice will be gone. It will remain that way until we remove all that extra CO2 and methane from the atmosphere.

I agree this chart could be a very good predictor of how the ice will behave once we hit zero ice but the chart does not support your conclusion. This chart shows that, as the ice has approached zero, we see an acceleration in the freeze and growth in volume. We also see an acceleration in the melting as 1st year ice which makes up more and more of the volume melts more rapidly. The resulting dimple in the chart will remain.

My guess is volume will reach at least 9000 cubic km after the 1st BOE.

[Daniel B.]

I disagree that there will be literally zero ice in the Arctic.  Even if all the ice were to disappear one summer, it would reform readily the following winter.  In every other body of water on this planet which experiences complete summer ice melt, the ice reforms in the winter.  Check out the Baltic and Okhotsk seas, which are significantly warmer than the Arctic. 

I disagree also with your temperature predictions.  According to the CRU temperature series, the average Arctic temperature has risen 2C over the past half century, all occurring during the winter months.  There has been no overall temperature change during the summer, during which temperature barely rise above 0:

[dnem]

Daniel B., there are times when I question your sincerity.  You seem well informed enough to know why summer temps are pinned to 0 C.  In the absence of ice, they will no longer be pinned.

[TerryM]

Are there records that show increasing mist, fog, low clouds, in the high arctic? I recall Dr. Francis complaining of constant fog as she was steaming toward the pole very late in the season in (2012)?


Lots of water vapor has been assumed to account for the fossil alligators, primates, and rhinos found as far north as Ellesmere Island. It was a long time ago - but the sun wasn't shining there for months.


Lots of water vapor = lots of GHG keeping a cap over the Arctic.


Terry

[El Cid]

Daniel B., there are times when I question your sincerity.  You seem well informed enough to know why summer temps are pinned to 0 C.  In the absence of ice, they will no longer be pinned.

If the ice would completely melt out by June, then the water would have enough time to warm up and then air temps would rise significantly above zero.

However, during the "first phase" the ice will be gone only in August and by that time there will hardly be enough insolation to warm up the water, so even though we will lose the ice, water temps will probably not be much above zero in August and September, so ait temps will also be around zero. And then the polar night comes.

So, I agree with Daniel B. Even after we first lose the Arctic ice, summer temps will be pinned to zero for quite a long time.

[Daniel B.]

Daniel B., there are times when I question your sincerity.  You seem well informed enough to know why summer temps are pinned to 0 C.  In the absence of ice, they will no longer be pinned.

If you read my previous posts, you would see that I do not believe that there will be a complete absence of ice.  Hence, the temperatures will not rise significantly above 0. 

[Telihod]

If the ice would completely melt out by June, then the water would have enough time to warm up and then air temps would rise significantly above zero.

However, during the "first phase" the ice will be gone only in August and by that time there will hardly be enough insolation to warm up the water, so even though we will lose the ice, water temps will probably not be much above zero in August and September, so ait temps will also be around zero. And then the polar night comes.

So, I agree with Daniel B. Even after we first lose the Arctic ice, summer temps will be pinned to zero for quite a long time.

I remember hearing P. Wadhams in a youtube video (or maybe it was in his book, I'm not sure) that the surface temperature of the water was 17C somewhere over east siberian arctic shelf, and methane was coming out of the ocean, when he was on a ship. So  in those parts of the arctic where there is no ice in the vicinity, it can rise relatively high.

[dnem]

Daniel B., there are times when I question your sincerity.  You seem well informed enough to know why summer temps are pinned to 0 C.  In the absence of ice, they will no longer be pinned.

If you read my previous posts, you would see that I do not believe that there will be a complete absence of ice.  Hence, the temperatures will not rise significantly above 0.

That's fine, and I agree.  It seemed you were responding directly to this: "A few rounds after the first BOE there will be literally 0 ice in the Arctic and temperatures of 20C maybe even 30C."  When there is "literally 0 ice in the Arctic" temps will no longer be pinned.  Personally, I wouldn't be surprised to see some excursions above the 0 C line on the DMI 80 N plot during the summer over the next decade, even before all the ice is gone from the last refuges.

[Daniel B.]

Daniel B., there are times when I question your sincerity.  You seem well informed enough to know why summer temps are pinned to 0 C.  In the absence of ice, they will no longer be pinned.

If you read my previous posts, you would see that I do not believe that there will be a complete absence of ice.  Hence, the temperatures will not rise significantly above 0.

That's fine, and I agree.  It seemed you were responding directly to this: "A few rounds after the first BOE there will be literally 0 ice in the Arctic and temperatures of 20C maybe even 30C."  When there is "literally 0 ice in the Arctic" temps will no longer be pinned.  Personally, I wouldn't be surprised to see some excursions above the 0 C line on the DMI 80 N plot during the summer over the next decade, even before all the ice is gone from the last refuges.

Neither would I.  Large, sustained temperature rises, I cannot foresee.

[Archimid]

  Even if all the ice were to disappear one summer, it would reform readily the following winter.

After the first BOE, the ice will struggle to return.

I disagree also with your temperature predictions.  According to the CRU temperature series, the average Arctic temperature has risen 2C over the past half century.

all occurring during the winter months

And it will remain that way until sufficient ice is gone.

[Sciguy]

After the first BOE, the ice will struggle to return.

I've pasted links and excerpts of several peer-reviewed scientific papers that refute this argument.  Do you have any peer-reviewed studies that support it?

Why would the Arctic behave differently than other bodies of water that melt out and then refreeze seasonally?  It's much colder and in the dark much longer than they are.

[be cause]

My why's would include the Arctic being substantially different to those other bodies of water . b.c.

[Tealight]

The Arctic reaching 20-30C after a few blue ocean events, you guys lost it 

The Arctic will still be a deep ocean and not a shallow lagoon. Currently all heat goes into the top 1-2m of the ocean melting the ice. Even heat from subsurface water ends up into the top 1-2m. After the ice is gone the bottom heat source becomes a heatsink and you have to heat several tenth of meters of water to get a significant temperature increase. The heat loss to space also increases by the power of four for every degree more. With your imagination the always ice-free waters around Iceland would already be a tropical paradise.

[Shared Humanity]

Thank you.

[Archimid]

I've pasted links and excerpts of several peer-reviewed scientific papers that refute this argument.  Do you have any peer-reviewed studies that support it?

This is going to be weird, but I will give you a reference and tell you why I think their model is wrong. I think similar arguments apply to any model predicting the ice gone by 2070 or growing back regardless of CO2 forcing. This paper is well written and I think someone quoted it before, maybe you.

Recovery mechanisms of Arctic summer sea ice

https://agupubs.onlinelibrary.wiley.com/doi/full/10.1029/2010GL045698

FTA:

All our experiments start from sea‐ice free conditions on 1st July. As expected, the Arctic Ocean remains ice‐free for several months, and significant sea‐ice cover does not develop before November. However, sea ice then grows very rapidly, since the growth rate for thin ice is much higher than for thick ice, which acts as a negative feedback on thickness during the growth season

Here is the illustration of the fast growth their model assumes:



This makes no sense. The first September have 0 ice. That means that growth started around November (according to the model) and then grew to such level that the extremely thin ice survived the next melting season with several million kilometers of ice to spare? Sorry but no. That is not realistic at all. I have never seen growth like that. Has anyone?
In fact the anemic growth of 2016, 2017 and 2018 prove that such fast growth is not necessarily true as the models assume.

 Let’s keep going

For SAT a large positive anomaly occurs between October and February after the initial perturbation, with a peak of almost 11 K in November (Figure 2). After February, there are no further SAT anomalies stronger than natural variability.

We are already halfway there with the temperature anomalies lasting well into March.  The extra heat the Arctic accumulated so far from albedo changes and wavy jet streams is nothing compared to a few months of full insolation without ice + atmospheric and ocean currents going crazy.

And the model expects super record extent growth with enough thickness to have ice left over by September? Considering the last three years, this is not plausible.

But wait, there is more:



The paper calls this graph “Energy Budget of the Arctic Ocean Domain”. First let me say that it is a great graph, and the approach they took is elegant. I like the idea of the Arctic energy unit. However, the assumptions seem wrong after 2015. From September to March they have a net output of heat from the Arctic out to the planet. It seems to me that after 2015 that arrow was either reversed or significantly reduced.

On top of that, that arrow is not free. That energy goes back into the climate system and some of it ends up right back in the Arctic. After 2015 vapor plumes straight from the tropics are entering the Arctic.

And then there are clouds. A warm arctic is a humid arctic. And then there is methane pockets that will release quite a bit of methane when the Arctic warms. Anyway. The model fails a reality check.

Why would the Arctic behave differently than other bodies of water that melt out and then refreeze seasonally?  It's much colder and in the dark much longer than they are.

 The Arctic Ocean will behave exactly like the other bodies of water. Ice will grow from more ice when the temperatures and salinity requirements are met. When there is no ice, the coasts make a good proxy for ice and from there it expands.

Ice also forms in the open water when it is very close to ice (hundreds of miles), but it has to be very cold for that. The very cold times in the arctic are already rare. This year we had like 15 days of average Arctic cold.

[Archimid]

The Arctic reaching 20-30C after a few blue ocean events, you guys lost it 

 Sunlight is not enough to get the Arctic to 20C-30C but warm air transported over warm, ice free ocean is.  Please keep in mind, Arctic sea ice keeps the air near the surface cold. That cold air meets the warm air from the continents and keeps the continents cool during the summer. When the is gone, the continents will warm. 30C during summer is not very rare over the continents during the summer.  That air will transport into the warm ice less Arctic.

The Arctic will still be a deep ocean and not a shallow lagoon. Currently all heat goes into the top 1-2m of the ocean melting the ice. Even heat from subsurface water ends up into the top 1-2m. After the ice is gone the bottom heat source becomes a heatsink and you have to heat several tenth of meters of water to get a significant temperature increase.

And all that heat that used to go to melt ice now remains in the system for longer.

The heat loss to space also increases by the power of four for every degree more.

I'm not sure about this. How do you figure that increase?

With your imagination the always ice-free waters around Iceland would already be a tropical paradise.

Yeah, but with such long days imagine the sunburns!

hmm. Something occurs to me, what is the impact on temperatures of not having a night cycle?  The rest of the world gets at least some hours of night during summer. The ice less Arctic doesn't get that break.

[crandles]

This makes no sense. The first September have 0 ice. That means that growth started around November (according to the model) and then grew to such level that the extremely thin ice survived the next melting season with several million kilometers of ice to spare? Sorry but no. That is not realistic at all. I have never seen growth like that. Has anyone?
In fact the anemic growth of 2016, 2017 and 2018 prove that such fast growth is not necessarily true as the models assume.

So that response to peer reviewed science is basically 'I don't believe it'.

But wait, there is more:



The paper calls this graph “Energy Budget of the Arctic Ocean Domain”. First let me say that it is a great graph, and the approach they took is elegant. I like the idea of the Arctic energy unit. However, the assumptions seem wrong after 2015. From September to March they have a net output of heat from the Arctic out to the planet. It seems to me that after 2015 that arrow was either reversed or significantly reduced.

Do you spot the triangle delta sign? That means difference in forcing so what they are saying is that when arctic is cold lots of heat gets transferred to the arctic. Now that it is warmer less heat gets transferred in so the difference is a net outgoing.

So this seems a case of you thinking peer reviewed science contains major mistakes when it is actually you failing to understand.
[/quote]

[crandles]

I disagree completely. This seems a case where yet again, peer reviewed science is not, has not been communicated to the public in a way that the public could ever understand!

I disagree completely.

Public outreach is important. However papers are not the way to do outreach. In papers the (only?) purpose is to write concisely in a way peers will understand.

[Daniel B.]

I disagree completely. This seems a case where yet again, peer reviewed science is not, has not been communicated to the public in a way that the public could ever understand!

I disagree completely.

Public outreach is important. However papers are not the way to do outreach. In papers the (only?) purpose is to write concisely in a way peers will understand.

Agreed.  Research publications are intended for other scientists, or those interests in the field.  Hence, the paper is often outside the comprehension of the general public. 

[Daniel B.]

Yeah, but with such long days imagine the sunburns!

hmm. Something occurs to me, what is the impact on temperatures of not having a night cycle?  The rest of the world gets at least some hours of night during summer. The ice less Arctic doesn't get that break.

Perhaps we should look at real examples.  Located above the Arctic circle, and receiving 24 hours of sunlight for several weeks every year, Utqiavik (a.k.a. Barrow, Alaska) has experienced summer ice-free conditions for several years now, and has warmed considerably.  Using the most recent annual temperature and sea ice data (2017), we see that the ocean off the coast was ice-free from July through October, yet the temperature anomaly for that period was a mere 2C above average.  The temperature anomaly during the winter months was 10C above average.  The average annual increase was ~5C.  Granted this is not a direct comparison to what may occur over the open ocean.  However, one would expect some sort of correlation.

[Archimid]

So that response to peer reviewed science is basically 'I don't believe it'.

I don't care if an article is peer reviewed. It it contradicts my perception of reality I'll challenge it. This models predicts extraordinary growth that no one has ever witnessed before.

 I'll tell you exactly the mistake of the model.

The model assumes that after November, Arctic temperatures return to average. The average temps remain average even while generating record ice growth until April.

That's the opposite of the observations. So far the surface temperature anomalies last until April and ice has not grown at the incredible rate required by this model. There is something seriously wrong with it. The part of my reply that you didn't address explains why, let me repeat their assumption:

For SAT a large positive anomaly occurs between October and February after the initial perturbation, with a peak of almost 11 K in November (Figure 2). After February, there are no further SAT anomalies stronger than natural variability.

We are already at 5-8 K winter anomaly that last until April.  Yet you want me believe that after the first BOE all that extra heat will be vented out to of the arctic system? That makes no sense.

Do you spot the triangle delta sign? That means difference in forcing so what they are saying is that when arctic is cold lots of heat gets transferred to the arctic. Now that it is warmer less heat gets transferred in so the difference is a net outgoing.

huh? I'm not contradicting those basic physics. I'm, saying that even if the temperature gradient is smaller, leading to a net heat output, hot air intrusions from the oceans and continents have so far overwhelmed the cooling forcing of a smaller temperature difference. 

Of course this model, like most models I've seen, ignore the changes in atmospheric currents.

[Archimid]

we see that the ocean off the coast was ice-free from July through October, yet the temperature anomaly for that period was a mere 2C above average.

Because there is a huge chunk of ice a few thousand miles to the north interacting with southern heat. When there is no ice, the anomalies will change drastically.

[Archimid]

Frankly I'm becoming more convinced that the global models are slightly overestimating global temperatures because they are highly underestimating sea ice loss.

[Daniel B.]

we see that the ocean off the coast was ice-free from July through October, yet the temperature anomaly for that period was a mere 2C above average.

Because there is a huge chunk of ice a few thousand miles to the north interacting with southern heat. When there is no ice, the anomalies will change drastically.

And what if Barrow is not in fact a correlation to the whole Arctic Ocean with under 1 m2klms of ASI but is in fact an outlier variant?

Some supporting evidence and solid reasoning for such a claim would be helpful.

The temperature anomaly in Barrow is not an outlier.  The mean temperature north of 80 degrees has increased significantly in every winter over the past decade, while exhibiting a very slight decrease during the summer.   

http://ocean.dmi.dk/arctic/meant80n.uk.php

[Archimid]

The temperature anomaly in Barrow is not an outlier.  The mean temperature north of 80 degrees has increased significantly in every winter over the past decade, while exhibiting a very slight decrease during the summer. 

And the summer temperatures will not increase significantly until the ice is gone.

[crandles]

As far as I am concerned, advancing the science is almost everything and while there are a few other things like public outreach these are likely best done outside of the papers aimed at advancing science.

Sorry but just because you want public outreach to be treated as a top priority, doesn't necessarily mean you deserve it.

[Archimid]

As far as I am concerned, advancing the science is almost everything and while there are a few other things like public outreach these are likely best done outside of the papers aimed at advancing science.

The more easily understood is your research, the more outreach you will get. However at the higher levels of science, like peer reviewed literature precision and accuracy are a higher priority than understandability.

Peer review is important, but what happens when peers are wrong? Science advances.

I will like to say that the assumptions made in the paper I quoted were extremely good at the time they were made.

[timeTravelTo(2011)]

Yes 2007 was a record low volume but look at that growth rate! Volume is growing like it never has. Better yet, look at the winter surface temperatures anomalies. The heat disappears by November and then is just record ice growth.



[timeTravelTo(now)]

The same argument could have been made all the way to 2014. The model was accurate for the knowledge of the time.

Then the last three years (and ongoing) happened. Record ice growth ceased. Temperature anomalies are not "venting" the heat out. By November the Arctic doesn't return to normal.

[Alexander555]

Perhaps we should look at real examples.  Located above the Arctic circle, and receiving 24 hours of sunlight for several weeks every year, Utqiavik (a.k.a. Barrow, Alaska) has experienced summer ice-free conditions for several years now, and has warmed considerably.  Using the most recent annual temperature and sea ice data (2017), we see that the ocean off the coast was ice-free from July through October, yet the temperature anomaly for that period was a mere 2C above average.  The temperature anomaly during the winter months was 10C above average.  The average annual increase was ~5C.  Granted this is not a direct comparison to what may occur over the open ocean.  However, one would expect some sort of correlation.


I think it will be pretty hard to defend your point Archimid, 20 to 30 degree C at the Arctic. Even at the Equator its not 30 C today, above the sea. I think Daniels example gives a good indication. Something like 2 C warmer if the sea becomes ice-free in summer. And like you say, if all the ice is gone, the wind will become a little warmer. And it's hard to say how much, because that wind will have to travel a long distance. Otherwise he want be that warm.  But lets say another 1 to 2 degree C. And the average temperature today is only something like 1 C. That makes 5 degree C. Lets add an extreme heatwave in Europe or Africa, moving in the right direction, and you maybe have 6 degree C. And i think everything above that will be a long and slow proces. Because you still have a long winter without sunlight. So almost everything has to come by air or sea from the rest of the planet.

[Archimid]

I agree that 20C-30C is hard to justify. My first instinct was 10C-20C but then I checked out max temperatures over Siberia and decided to go with the higher range. There will definitely be segments of the Arctic that may hit daily 30C temps occasionally, but not Arctic wide and not for very long.