The 2014 Melting Season

[ChrisReynolds]

domen:

indeed, for an appropriate comparison with the first graphic you responded to the snap shot for end of august 2013 using that model is this one:

http://www7320.nrlssc.navy.mil/GLBhycomcice1-12/navo/arcticictn/nowcast/ictn2013083012_2013083100_502_arcticictn.001.gif

Although the others still illustrated that there has been significant melt/breakup of the CAA thick ice in the past few years at least.

EDIT:

That can't be right! The sea ice edges of the two models' graphics for the same date don't match up at all?!

I think there are two versions of HYCOM being used.

I use stuff from here:
http://www7320.nrlssc.navy.mil/hycomARC/arctic.html

I note that others are using plots with the GBL suffix - if I recall correctly they started nesting HYCOM with another atmosphere model recently, I suspect this is the GBL type of plot.

[ChrisReynolds]

Got it.

So there's one HYCOM here:
http://www7320.nrlssc.navy.mil/hycomARC/arctic.html
And another HYCOM here:
http://www7320.nrlssc.navy.mil/GLBhycomcice1-12/arctic.html

Both of which give different results:




Which is best?

PS - I know some people call this ACNFS. I am not convinced it is: https://hycom.org/

[jonthed]

Thanks Chris,

My query was that the two HYCOM models had totally different sea ice edges for the same date last year; 20130831,

leading me to think that the archive wasn't correct. it seems fine for recent dates.

[zworld]

take any one week and cloud variations will likely swamp the effect of even 96,000ppm of methane because cloud blocks IR (and emits back to the surface) at the peak of surface emission temperatures, whereas methane's effect is on the declining upward end of the emission spectrum.

If I've not been clear here, just ask me to explain.

Chris, I hope Im not being a bother cause alot of what you present goes over my head, but wouldn't that great an increase in atmospheric methane overwhelm the ability for the system to degrade the methane to water vapor, and not produce the same cloud effect as it normally would. I can see that a certain amount would still be converted in the lower atmosphere to water vapor, but would the outcome be the same.

Just like 7 day weather modeling which earlier posters have pointed out repeatedly can now no longer be considered even slightly valid as things change, wouldn't this be true for methane dispersion as well. With concentrations running 10 times background at the moment, and spikes consistently at 2+ ppm and growing, at what point do we say that the old way of analyzing emissions, degradation and dispersion are no longer valid.

I guess the question is, just like weather modeling, can any of the old way of looking at things apply, or do we need to consider new ways of understanding the processes at work.

Either way, thank you for your input. Though hard for a non scientist like me to understand, it is obviously well thought out and informative.

[ChrisReynolds]

Thanks Chris,

My query was that the two HYCOM models had totally different sea ice edges for the same date last year; 20130831,

leading me to think that the archive wasn't correct. it seems fine for recent dates.

Yes I noticed it too - sorry I don't know why, I have one paper on HYCOM but haven't had the chance to read further.

[ChrisReynolds]

Chris, I hope Im not being a bother cause alot of what you present goes over my head, but wouldn't that great an increase in atmospheric methane overwhelm the ability for the system to degrade the methane to water vapor, and not produce the same cloud effect as it normally would. I can see that a certain amount would still be converted in the lower atmosphere to water vapor, but would the outcome be the same.

Globally it would, locally I am not sure. OH radicals disasociate methane through a complex set of reactions energised by UV (IIRC), locally high methane wouldn't necessarily deplete the wider reserves of OH radicals. In winter anyway the lack of UV means methane builds over the Arctic so it wouldn't be a factor then. But clouds would still have a stronger effect over short timescales.

Just like 7 day weather modeling which earlier posters have pointed out repeatedly can now no longer be considered even slightly valid as things change, wouldn't this be true for methane dispersion as well. With concentrations running 10 times background at the moment, and spikes consistently at 2+ ppm and growing, at what point do we say that the old way of analyzing emissions, degradation and dispersion are no longer valid.

I guess the question is, just like weather modeling, can any of the old way of looking at things apply, or do we need to consider new ways of understanding the processes at work.

Sorry I have never bought such arguments. I've never payed much attention to forecasts over 5 days, except where stable conditions like blockings occur. And I am not convinced by the idea that we're now in an unpredictable regime. I am not aware of evidence for a decline in prediction accuracy in the Arctic and haven't noticed it myself.

Variation in extent at minimum has grown, but it has grown for mechanistic reasons, not because we're near some sort of 'tipping point'. Weather and ice state explain 2007 and 2012, likewise 2013 and 2014.

Methane at 2ppm? That's really rather low, it will climb in the centuries to come, but I still see it as a chronic issue, not a catastrophic issue. i.e. a cause for concern and to check up on it every few years, not a cause for alarm.

[deep octopus]

The 00z GFS and 00z ECMWF models are still in agreement, and it's bona fide horrible for the foreseeable future. The entire Arctic basin is forecast to be dominated by a strong high pressure system. This is a very 2008 situation. Or 1999 with strong high pressure dominance and very heavy August losses. We could probably expect several consecutive 100k loss days with this.

[Phil.]

Just a follow on to NeilT's questions about methane. The statements about methane's greater 'potency' can be confusing.

There are around 220 CO2 molecules for every methane molecule in the atmosphere. And roughly speaking the GH effect contribution is logarithmic for each gas so it is a warming per doubling. So (again roughly) adding 1 methane molecule has the same impact as adding 220 CO2 molecules since it is the contribution to doubling for that species that matters.

logarithmic for CO2 but square root for CH4

In fact the warming impact of methane is still given as less than this multiple because methane absorbs in a part of the earth's long-wave spectrum where there is less energy available to work with - CO2 and H2O are active at the high point in the spectrum.

If CO2 and Methane were both at 400 ppm then adding a methane molecule would have less effect than adding a CO2 molecule because of where they work in the spectrum.

[Frivolousz21]

Cice is far more realistic than arcus.

It matches piomas volume.

[Frivolousz21]

What really needs to be said?

[ChrisReynolds]

Cice is far more realistic than arcus.

It matches piomas volume.

Huh?

ARCUS = Arctic Research Consortium of the United States.

[pearscot]

Wow!

if that is real, it's quite shocking.  Looks like there isn't much left of the center ice pack.  Assuming the 1.5-2 meter areas survive there really won't be much left of the central ice pack.  What do you guys think?  I wonder if this recent warmth will be enough to really melt what's left in the central area.

[zworld]

I am not convinced by the idea that we're now in an unpredictable regime. I am not aware of evidence for a decline in prediction accuracy in the Arctic and haven't noticed it myself.

Thanks Chris. Still dont totally understand but its getting clearer. Sort of. The thing that still nags at me is the concept of an unpredictable regime.

Example, NASA has found a distinct correlation between noctilucent clouds in Antarctica and winter air temperatures in North America. A teleconnection they call it. It has been theorized that the increase in these clouds is related to methane overabundance making it's way into the upper atmosphere, and so far I have seen no one dispute this hypothesis. They also found that the strato winds in the Arctic affect noctilucent cloud formation in the Antarctic on a two week time scale.

Im think Im beginning to believe in the GAIA concept. Affect a local area that produces change in another area that then produces change locally in a few weeks timescale. So the concern isn't necessarily just with temps increasing locally under a methane blanket, but that the blanket may be causing other effects that come back to the localized area.

I know this is all a big unknown, but I give it as a possible example of how we may need to question what is happening outside the context of all that has happened before.

As an investigator my job is to take the pieces of a puzzle and try and form a larger picture, and AGW influenced by methane more than anything else keeps coming back into the picture. And since we have never calculated the full emission scope of methane, especially from offshore and onshore O&G drilling operations that experience a gas blowout, or fugitive emission rates, the question of how much of the CO2 in the atmosphere started out as methane release still lurks.

And as stated, I believe the Arctic is the place to learn the answers to these questions. Im probably wrong but must still keep an open mind to the possibilities.

Hope I havent been too confusing. And thanks again for all of your input.

[Frivolousz21]

Cice is far more realistic than arcus.

It matches piomas volume.

Huh?

ARCUS = Arctic Research Consortium of the United States.

There is a graphic about 5-6 pages back showing ARCUS volume about double piomas at the minimum while CICE is almost perfectly aligned with it.

[Frivolousz21]

**** is hitting the fan

[ChrisReynolds]

Cice is far more realistic than arcus.

It matches piomas volume.

Huh?

ARCUS = Arctic Research Consortium of the United States.

There is a graphic about 5-6 pages back showing ARCUS volume about double piomas at the minimum while CICE is almost perfectly aligned with it.

Here?
http://forum.arctic-sea-ice.net/index.php/topic,778.msg32221.html#msg32221

That's ANCFS. Confusion sorted?

[Frivolousz21]

Thanks it seems I confused the acronyms of Arcus and Ancfs.

The point still stands about CiCE

[Frivolousz21]

The Euro is flat nasty.  The ESS is in for a rough ride which is it already receiving.

[epiphyte]

It troubles me that so much weight is given to the rate of decline in area+extent as an indicator of the progress of the melt season. I think that this is becoming less and less relevant as the volume distribution becomes more even every year - I.e. you would expect that the year there is an ice-free September, there might still be a very high percentage of the April area/extent in early August.

Here today, gone tomorrow. I'm not saying it's this year - but on the year it happens, it will happen fast - and IMO anyone tracking progress by comparison with past years CT area decline will be very surprised to see it go.

To illustrate what I'm reaching for at I followed the melt progression of a large, high-albedo ice floe under favorable conditions. The image below is a series of snapshots of a ~300km^2 ice floe on the Laptev/CAB boundary  (~79N , 110E) taken from worldview between 01-jul and 01-aug. (see images below). For each image, I counted the number of pixels with > 50% luminosity on a 140x140 pixel array centered on the floe, cut from that day's 7-2-1 Terra/MODIS image.

As Peter Ellis pointed out earlier on the thread - anything you can see on EOSDIS is not slush. The floe we are looking at here is a contiguous, high-albedo (snow-covered?) ~300km^2 chunk of ice at the edge of the pack. It remains so from 01-26 jul. On Jul 27, it becomes a little less bright and loses 10% of its area. Three days later - poof. It's gone.

[Frivolousz21]

Dude thanks that is a hell of a post.  That is what I think could happen to the ESS. 

On face value the ESS looks super solid compared to years past. But is it that thick?  No.

Now a 5-10CM a day melt pattern with major compaction is upon us.  I don't think it survives as well as it looks right now.

[Frivolousz21]

It's not taken very long to get here.  We are embarking on a pattern that is the extreme opposite of how 2013 went thru the first half of August.

[domen_]

Chris and jonthed, thanks!

[Buddy]

Here today, gone tomorrow. I'm not saying it's this year - but on the year it happens, it will happen fast - and IMO anyone tracking progress by comparison with past years CT area decline will be very surprised to see it go.

ABSOLUTELY.........

[jdallen]

To illustrate what I'm reaching for at I followed the melt progression of a large, high-albedo ice floe under favorable conditions. The image below is a series of snapshots of a ~300km^2 ice floe on the Laptev/CAB boundary  (~79N , 110E) taken from worldview between 01-jul and 01-aug. (see images below). For each image, I counted the number of pixels with > 50% luminosity on a 140x140 pixel array centered on the floe, cut from that day's 7-2-1 Terra/MODIS image.

Magnificently done!

A "poster child" for what happens to ice when it wanders into water 2C+ above freezing.

[jdallen]

Globally it would, locally I am not sure. OH radicals disasociate methane through a complex set of reactions energised by UV (IIRC), locally high methane wouldn't necessarily deplete the wider reserves of OH radicals. In winter anyway the lack of UV means methane builds over the Arctic so it wouldn't be a factor then. But clouds would still have a stronger effect over short timescales.

I would love to be able to quantify this.

While I'm absolutely skeptical of sudden methane release's impact over the melt season, persistent increased methane over the freezing season may be relevant.  The timing would reduce outgoing heat flow and constrain the formation of new ice.  Systemically, it strikes me as a much worse time for methane concentration to increase than the melt season.

[oren]

To illustrate what I'm reaching for at I followed the melt progression of a large, high-albedo ice floe under favorable conditions. The image below is a series of snapshots of a ~300km^2 ice floe on the Laptev/CAB boundary  (~79N , 110E) taken from worldview between 01-jul and 01-aug. (see images below). For each image, I counted the number of pixels with > 50% luminosity on a 140x140 pixel array centered on the floe, cut from that day's 7-2-1 Terra/MODIS image.

As Peter Ellis pointed out earlier on the thread - anything you can see on EOSDIS is not slush. The floe we are looking at here is a contiguous, high-albedo (snow-covered?) ~300km^2 chunk of ice at the edge of the pack. It remains so from 01-26 jul. On Jul 27, it becomes a little less bright and loses 10% of its area. Three days later - poof. It's gone.

Very impressive!

[Neven]

It troubles me that so much weight is given to the rate of decline in area+extent as an indicator of the progress of the melt season. I think that this is becoming less and less relevant as the volume distribution becomes more even every year - I.e. you would expect that the year there is an ice-free September, there might still be a very high percentage of the April area/extent in early August.

Here today, gone tomorrow. I'm not saying it's this year - but on the year it happens, it will happen fast - and IMO anyone tracking progress by comparison with past years CT area decline will be very surprised to see it go.

To illustrate what I'm reaching for at I followed the melt progression of a large, high-albedo ice floe under favorable conditions. The image below is a series of snapshots of a ~300km^2 ice floe on the Laptev/CAB boundary  (~79N , 110E) taken from worldview between 01-jul and 01-aug. (see images below). For each image, I counted the number of pixels with > 50% luminosity on a 140x140 pixel array centered on the floe, cut from that day's 7-2-1 Terra/MODIS image.

As Peter Ellis pointed out earlier on the thread - anything you can see on EOSDIS is not slush. The floe we are looking at here is a contiguous, high-albedo (snow-covered?) ~300km^2 chunk of ice at the edge of the pack. It remains so from 01-26 jul. On Jul 27, it becomes a little less bright and loses 10% of its area. Three days later - poof. It's gone.

Very, very nice, epiphyte. An epiphany almost. 

Can I use this for the blog?

[Frivolousz21]

Winds are favorable for spreading the ice out in the CAB the next couple of days.

Extent may not drop very much until later in the week.  Unless the Kara, ESS, and Laptev really take some huge hits or the Chukchi opens up which it is about to bust wide open.

Winds are already turning very unfavorable for the ESS blowing waves and warmth and warm water right into the ice from the Chukchi water and ESS open water and this pretty much goes on for the foreseeable future.

[Michael Hauber]

The weather looks well suited for strong melting, but it did for the last week or so.  I was definitely wrong about previous speculation that the favourable melt weather would only last a few days before reverting.  But I'm also surprised at how mediocre the melt stats have been.  Only one century break.  We are still above 2013, and I had seen 2013 as a bit of a freak event not likely to be repeated.  And the weather this year seems clearly better suited to melt than 2013.  We have had a mix of poor melt and strong melt patterns - this isn't the first time Friv has got excited, and some of the other events Friv was excited about did come out similar to forecast (new graph for the sea ice page - Friv use of words such as OMG etc per day?)  We also had a very mild winter even compared to the last few years.  And NSIDC data shows that surface melt onset was early this year, more similar to 2012 and in marked contrast to 2013. 

So what is going on?  I continue to suspect that something needs explaining and that there may be cooler waters than previous years under the ice.

Or perhaps last weeks mediocre melt results were partly a momentum effect and the next week's melt figures will be more spectacular.  There is a significant area in the Chuckchi region that has quite quickly turned into mostly water, so maybe this is a sign of things to come.

[greatdying2]

Michael,

I think some possibilities have just been explained (very well!) by Epiphyte. What exactly makes you think that melting this year is not significantly stronger than in 2013 (or indeed that 2013 was so weak)? Area and extent may tell a very limited story.

[epiphyte]

Very, very nice, epiphyte. An epiphany almost. 

Can I use this for the blog?

I would be honored. At one point I thought I'd wasted a whole morning on that... but I've definitely changed my mind now

[deep octopus]

So what is going on?  I continue to suspect that something needs explaining and that there may be cooler waters than previous years under the ice.

Or perhaps last weeks mediocre melt results were partly a momentum effect and the next week's melt figures will be more spectacular.  There is a significant area in the Chuckchi region that has quite quickly turned into mostly water, so maybe this is a sign of things to come.

I suspect freshwater flux is partially at play. The GIS was hammered during the same year as the record low extent and volume in the sea ice (that is, the sordid outbreak of climate change-in-action that was 2012.) Rivers such as Mackenzie are also seeing larger discharges over time as ice breaks free earlier and precipitation increases, which during summer can amplify melt as the water is warmer as it enters the ocean; but I also figure that increased freshwater is temporarily altering the mixing in the ocean, and this can also drive negative feedbacks in the future. I would need to look at more studies to quantify how much such negative feedbacks may be at work. But it's a Faustian bargain, at that: short-term increases in sea ice at the expense of lost ice sheets. This is what investors would call a "dead cat's bounce" or a "bear rally." Ostensibly, years like 2013 are seen as "recoveries" by deniers when no such thing, in terms of overall heat budget, is happening.

Still, what we're seeing is interesting, and if 2014 performs worse (i.e. increased melt) than 2009/2013 or even 2010, it's indicative that negative feedbacks are either unable to hold back the long-term downward trend (an outcome predicted, of course, in climate models), or the negative feedbacks have not been fully realized yet. I tend to think it's the former. I simply do not think the sea ice is going to get much help until much later down the century when the GIS really gets wrecked—as solar energy channels elsewhere amidst ice-free summers—and freshwater flux again boosts sea ice, but for a relatively short period. At that point, it's salt on the wound anyway. Climate change will be in far more urgent, severe stages by then. The water is warming too fast. Epiphyte's great analysis demonstrates this destructive power that lurks. We also have trends such as an increasingly warmer and saltier Atlantic that contribute to the wreckage on the Atlantic-side ice as it did last year. Much is at play, but I think any uptick in the ice is an aberration at best, and compensated elsewhere in the world. And 2014 is still going to look bad on paper.

[Frivolousz21]

The weather looks well suited for strong melting, but it did for the last week or so.  I was definitely wrong about previous speculation that the favourable melt weather would only last a few days before reverting.  But I'm also surprised at how mediocre the melt stats have been.  Only one century break.  We are still above 2013, and I had seen 2013 as a bit of a freak event not likely to be repeated.  And the weather this year seems clearly better suited to melt than 2013.  We have had a mix of poor melt and strong melt patterns - this isn't the first time Friv has got excited, and some of the other events Friv was excited about did come out similar to forecast (new graph for the sea ice page - Friv use of words such as OMG etc per day?)  We also had a very mild winter even compared to the last few years.  And NSIDC data shows that surface melt onset was early this year, more similar to 2012 and in marked contrast to 2013. 

So what is going on?  I continue to suspect that something needs explaining and that there may be cooler waters than previous years under the ice.

Or perhaps last weeks mediocre melt results were partly a momentum effect and the next week's melt figures will be more spectacular. There is a significant area in the Chuckchi region that has quite quickly turned into mostly water, so maybe this is a sign of things to come.

And on Jaxa it's still all being counted in the extent.


Area has had 4 large drops in a row and is about -100K below 2013 now.

That gap will increase substantially over the next week or two.  Way to much emphasis is put on ice metrics to yield immediate results.

[Michael Hauber]

Michael,

I think some possibilities have just been explained (very well!) by Epiphyte. What exactly makes you think that melting this year is not significantly stronger than in 2013 (or indeed that 2013 was so weak)? Area and extent may tell a very limited story.

Its the arctic so you can never be sure of anything (except that as long as an ounce of ice remains the Arctic will perpetually be in recovery according to certain sources lol).  Do you think that the melt this year really is stronger than 2013 or any other year since 2010?  CT Area, IJIS extent and PIOMAS volume are all above 2013 by latest values. 

PIOMAS had shown every year from 2009 to 2012 having a lower volume than the year before, and extrapolation of the trend by quadratic showed ice free around 2016.  This had originally been speculated on in 2008, and up to 2012 the melt was maintained at the required rate.  I would allow that 2013 may have been a freak outlier and that the trend towards 0 in 2016 may have continued.  But 2014 has cemented the 2013 result as some type of 'recovery'.  At the minimum ice free in 2016 no longer looks reasonable, and we have recovered from a power dive straight to zero and may be on some type of slower trajectory. 

Or perhaps even stronger the recovery may be due to the reversal of a multi-decade variability such as PDO or AMO.  Perhaps such a factor may have caused most of the faster than predicted by the models melt.  Or perhaps the models need fixing due to failure to accurately capture an important process such as soot-albedo effects, and once they are fixed they will show a melt rate similar to what has been observed.  If the model-reality difference is due to multi-decade variability its possible that we may see several decade or more of increasing sea ice until thermodynamic inevitability catches up and the trend towards 0 resumes.  Modelling and recent history shows that global temperature can 'pause' for periods of up to about 20 years.  An individual region of the globe should be noisier and capable of bucking the long term warming trend for an even longer period.  Look at whats been happening in the Antarctic. 

[Pmt111500]

this: http://forum.arctic-sea-ice.net/index.php?action=dlattach;topic=778.0;attach=9260;image
On the Baltic Sea the last of the ice often disappears in this fashion usually accompanied by a warm front crossing the area of remaining ice. Most times the warm front is covered with warmer clouds than in winter or earlier spring so quantifying the speed of the effect like Epiphyte did here isn't too easy. Whether the last ice disappears on this front or the next isn't easily predictable and depends on the remaining thickness of the ice during the warm front attack on ice. Agree that it'll go 'fast' when it does go. Some august in the near future, August is still very much summer though the lower latitudes are getting some cooling during the nights.

On to see if the August temperature record here is broken today. 32 to 33 C predicted (~90F).

[notjonathon]

Michael--

Point 1: two seasons doth not a trend make (sounds Shakespearean, anyway).

Point 2: the Antarctic increase in sea ice is largely a result of the increased flow of fresh water from the decrease in land ice--the Antarctic is still contributing to sea level rise, I believe.

[cesium62]

But 2014 has cemented the 2013 result as some type of 'recovery'.   

I think we should at least let August play out, if not September, before we decide what tale 2014 is telling us.

[jdallen]

Modelling and recent history shows that global temperature can 'pause' for periods of up to about 20 years.  An individual region of the globe should be noisier and capable of bucking the long term warming trend for an even longer period.  Look at whats been happening in the Antarctic.

Michael;

There. Is. No. Pause.

Antarctic ice behavior is not the result of cooling in the southern hemisphere. 

While ice on the surface may be increasing in extent during the southern hemisphere winter, ice on the continent itself is actually weakening to the point where the West Antarctic ice sheet may disintegrate.

[Buddy]

....recent history shows that global temperature can 'pause' for periods of up to about 20 years.

Recent history doesn't show a "pause"......unless you ignore the 92% of the global warming that takes place in the oceans.

If you want to say that the atmospheric warming has slowed down it's rate of increase over the last 20 years.....then I would "buy" that.  But that is DEFINITELY not the same thing...

[Frivolousz21]

How fast extent drops will depend on how fast this area in the Chukchi becomes open water on the extent charts.  This is a pretty large area that goes all the way back thru over half the CAB but the worst spot is right here.  And it's spreading towards the ESS.  Which you can see on the right.

It has already been above 0C in this area for the last week.  Probably why the ice is now taking on the swirls look which means a lot of it is likely down to under .25M.


The HP that has formed over the Chukchi is sliding towards Alaska and arrives off the N. Alaska coast tomorrow.  Winds veer out of a Southerly direction with skies clearing out and vigorous warm air advection as well as warm water advection.


From what I have already seen I expect this area to open up and the ESS to quickly disintegrate over the next 5-10 days.

[Jim Hunt]

The point still stands about CiCE

No it doesn't. ACNFS and the new (not yet operational) GOFS 3.1 both use CICE 4. In due course GOFS 3.5 will use CICE 5.

Still confused?

[Frivolousz21]

The point still stands about CiCE

No it doesn't. ACNFS and the new (not yet operational) GOFS 3.1 both use CICE 4. In due course GOFS 3.5 will use CICE 5.

Still confused?

What is up with the pretentiousness?


The top one is realistic the bottom one is garbage.





Congrats on memorizing the labels of these graphics. 

[cesium62]

The point still stands about CiCE

No it doesn't. ACNFS and the new (not yet operational) GOFS 3.1 both use CICE 4. In due course GOFS 3.5 will use CICE 5.

Still confused?

I am.  I guess the original question stands:  if everything is using the same CICE model, why are there two different pictures of ice thickness, and which, if either, is generally believed to be more accurate?

[Pmt111500]

I am.  I guess the original question stands:  if everything is using the same CICE model, why are there two different pictures of ice thickness, and which, if either, is generally believed to be more accurate?

AFAIK, and there was some discussion last year about this but can't be bothered to find out where, the two graphs would be
 
one for ships (max thickness possible on the area, ARCwhatever) and
one for calculated average thickness (GLBwhatever), so

the latter one would be nearer to the true value. my guess is add a bit to that one to get near correct thickness. Please correct my memory if incorrect. Add some talk about a man with two clocks, and skip it since even pretty thick ice can be near isothermal at melting point and vanish from sight in three days.

[Michael Hauber]

Michael;

There. Is. No. Pause.

Antarctic ice behavior is not the result of cooling in the southern hemisphere. 

While ice on the surface may be increasing in extent during the southern hemisphere winter, ice on the continent itself is actually weakening to the point where the West Antarctic ice sheet may disintegrate.

Just to clarify my position a little further on the pause and the Antarctic connection.  Hansen in his landmark paper on the effects of Co2 in the early 80s stated that natural variation could cancel out co2 warming for periods up to 20 years.  If I remember right Gavin Schmidt on realclimate more recently said 15 years.  The figure should get smaller as warming accelerates.  Negative trends exist in UAH from 2005 to 2014, and HADCRUT from mid 1997 to late 2013, but in both cases adding the figures for the first part of 2014 pushes the trend positive.  The fact that the total ocean heat content has continued to go up during these periods is a worthwhile fact, but not at all relevant to my argument - that natural variation should in theory be able to cancel out warming for a region over a longer period than over the entire globe.

And on Antarctica - the situation there is complicated, and every paper I read explaining the trend seems to have a slightly different view, so i would not at this stage say that there is a explanation of the Antarctic trend that can be stated as settled science.  What we do have is well thought out working hypothesis that are probably mostly correct but may change in detail due to further study.  Whether Antarctic increase is due to winds, cooling, fresh water or whatever, the point still holds that perhaps the Arctic could do the same thing.  Unless we find a reason why it cannot - the two pole are quite different after all.  For instance if we confirm that the trend is due to melting land ice perhaps we will next be able to work out that there just isn't enough ice in the Greenland ice sheet to match the effects of the much larger Antarctic ice sheet.

[GeoffBeacon]

Chris Reynolds reply #2753

Chris

Thanks for that very informative reply.

It is getting me closer to understanding my question.

Geoff

[Neven]

Last year the weather was less conducive to melting, but it had the highest amount of first-year ice on record. This year the weather has been slightly better for melting, but a lot of multi-year ice is forming a wall on the Pacific side of the Arctic. On top of that, regional weather as well as SSTs in the Barentsz and Kara (important regions for the 2011 and 2012 melts) haven't been conducive to melting at all. And thus you get similar area/extent numbers.

The lesson from 2012 and 2013 seems to be confirmed this year: the start of the melting season is extremely important. As long as volume remains in the same ball park, a year that's off to a flying start and has the right ice thickness/age distribution, has a very good chance of breaking records. Despite the slow transition.

PS I've posted epiphyte's stuff on the ASIB: Poof, it's gone

[Frivolousz21]

The heat is building fast over the Pacific side and this is just getting started.

EC shows 8C in the Southern ESS.  2-3C up near the ice edge in the Chukchi/Eastern ESS.  4-5C over the Northern Laptev.

Even 4C right next to the ice off the NW coast of Alaska. 



https://weather.gc.ca/data/analysis/351_100.gif

[Peter Ellis]

I am.  I guess the original question stands:  if everything is using the same CICE model, why are there two different pictures of ice thickness, and which, if either, is generally believed to be more accurate?

AFAIK, and there was some discussion last year about this but can't be bothered to find out where, the two graphs would be
 
one for ships (max thickness possible on the area, ARCwhatever) and
one for calculated average thickness (GLBwhatever), so

the latter one would be nearer to the true value. my guess is add a bit to that one to get near correct thickness. Please correct my memory if incorrect. Add some talk about a man with two clocks, and skip it since even pretty thick ice can be near isothermal at melting point and vanish from sight in three days.

It wasn't last year, it was two weeks ago, and it was posted up here by Jim Hunt
http://forum.arctic-sea-ice.net/index.php/topic,778.msg32619.html#msg32619
... from a discussion on the HYCOM forum here
https://groups.google.com/a/hycom.org/forum/#!topic/forum/Rqa_ao0bGQE

AlanW (who posted here to tell us about the new model) is presumably the same Alan (@hycom.org) who is posting in the HYCOM forum, so he knows what he's talking about. The difference is largely that when they set up the new model, they had better initialisation parameters for ice thickness.

The stuff about "max thickness for ships" is as far as I can tell something that was spitballed as a possible explanation by a non-scientist on some arctic ice forum a few years ago and has since percolated into the ASI blogosphere consciousness without any substantiation.  If anything it derives from St*v* G*dd*rd's constant claims that the Navy knows best because they use the data to protect their ships.  If you go back and read the original Navy validation papers (which I have), it's very clear that the main aim is to predict the ice edge, and that thickness/volume is very much a secondary concern.

[Peter Ellis]

To illustrate what I'm reaching for at I followed the melt progression of a large, high-albedo ice floe under favorable conditions.

This is a beautiful post.