Support the Arctic Sea Ice Forum and Blog

Author Topic: Arctic Refreeze Discussion: 2014/2015  (Read 201204 times)

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #100 on: November 14, 2014, 07:08:03 PM »
Guided by these, it is pretty clear what you were doing was speculating about the impact of CH4.  Where you went off the rails and found yourself being challenged was when you tried to use that speculation as the foundation for further argument.
JD, you persist, so I've made a new thread for your assumptions off this 'official' thread:

http://forum.arctic-sea-ice.net/index.php/topic,1065.0.html
[]

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #101 on: November 14, 2014, 08:53:46 PM »
Guided by these, it is pretty clear what you were doing was speculating about the impact of CH4.  Where you went off the rails and found yourself being challenged was when you tried to use that speculation as the foundation for further argument.
JD, you persist, so I've made a new thread for your assumptions off this 'official' thread:

http://forum.arctic-sea-ice.net/index.php/topic,1065.0.html

I suppose I should be gratified you consider me sufficiently important that you need a completely separate thread to excoriate me.  I'm sorry you misinterpreted the intent of my last note.  Be aware, beyond this reply, I've no intention of engaging in any further personal exchange with you.
This space for Rent.

Neven

  • Administrator
  • First-year ice
  • Posts: 9470
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 1333
  • Likes Given: 617
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #102 on: November 14, 2014, 10:42:30 PM »
Great. Back to the Arctic, everyone.
The enemy is within
Don't confuse me with him

E. Smith

mark

  • New ice
  • Posts: 59
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #103 on: November 15, 2014, 12:18:10 AM »
I wish he wasnt so touchy he brings up some good topics of conversation (hes cut off talking to me as well!).

On the november extent (its on your graph) and piomass data isnt it possible that there may not have been a melt but changing systems and wind directions caused the ice to compact and reduce extent over a short period. Personally I dont know how PIOMAS calculates the volume but being a model it may have equated loss of extent with a loss of volume when in fact thickness may have gone up.

As usual in speculation like this over a short period there are several ways of looking at the data and a safer bet is over a slightly longer period to see if the change was consistent. Without all the possible forcings on view taking data in isolation is risky if you want to confirm a point of view.

Cheer up Vid, I've been lurking along with this and other threads you post on and if you werent so touchy and convinced everyone is against you, you may have made a few friends here instead of foes!

Peter Ellis

  • Grease ice
  • Posts: 619
    • View Profile
  • Liked: 33
  • Likes Given: 14
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #104 on: November 15, 2014, 12:19:21 AM »
I'm not saying you confirming it amounts to evidence, Chris. The good Peter Ellis here rejects PIOMAS as actual data, and what I did was to look for volume reductions in the (whole Arctic) PIOMAS data, before you went ahead and did exactly the same in the regional PIOMAS data.

For a guy who rejects PIOMAS altogether as not actual data, this isn't good enough, but I figured «appeal to authority» might work in this case (your name rather than what you're saying).

It doesn't, and I'm still not convinced that there's been any melt in the Arctic this time of year.  I can see the white pixels marching backwards in the IJIS map, of course - and that will doubtless propagate through to the PIOMAS model. However, that could be ice compaction, differential snowfall making areas more/less visible to the satellite, cloud interference, wave interference, etc.

It's possible there was some melt, of course, but the evidence just isn't strong enough to stand up and say "here is an unprecedented event which we must explain".

If I understand your other posts, you're not claiming that the methane directly caused the putative melt (which is good, because as covered previously this just doesn't make physical sense), but that they  may be due to a common cause - e.g. an influx of warm air melting some ice and releasing some of the underlying hydrates.

To which I would point out that this water was ice-free a few weeks ago, and the waters were warmer then.  Right now, they are cooling and the ice edge is advancing. Even if that advance has stalled or reversed for a few hours, why would that suddenly destabilise the hydrates many metres below, when fully open water a few weeks ago did not?

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #105 on: November 15, 2014, 01:38:53 AM »
From my setbacks.php script, looking for days of melt in the refreeze season in PIOMAS volume data during the month in question:

Code: [Select]
October:

1981/277 (13380/13372)
2006/282 (9386/9383) 2006/283 (9383/9374)
2009/281 (7039/7038)

November (thoroughly discussed here):

2013/311 (9249/9245)

December:

2008/351 (15210/15155)

January:

1981/8 (24666/24634)
1984/11 (24414/24398)
2004/24 (20878/20864)
2007/29 (19639/19497)

February:

1989/46 (27882/27876)
1992/40 (26481/26476)
1998/57 (27364/27362)
2007/53 (21550/21521)

March:

1987/90 (31427/31410)
1989/66 (29057/29039) 1989/90 (29949/29945)
1995/72 (27852/27847)
2000/79 (26404/26402) 2000/80 (26402/26391)
2007/77 (23163/23122) 2007/88 (23663/23654)
2008/70 (23450/23420) 2008/88 (24669/24581)
2011/68 (21139/21073)
2014/89 (22604/22602)
[]

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #106 on: November 15, 2014, 08:16:50 AM »
.
[]

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #107 on: November 15, 2014, 09:58:26 AM »
For information of the board. A drop relative to the average of a past baseline  period does not necessarily mean there was an actual drop over that period. There might be an actual drop, but in the cas of a period of dominant increase the most likely explanation is merely that the drop relative to the past period is caused by an increase that is less than the average for the past period.

Richard Rathbone

  • Nilas ice
  • Posts: 1730
    • View Profile
  • Liked: 387
  • Likes Given: 24
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #108 on: November 15, 2014, 09:23:22 PM »
It would be interesting if the anomaly changes were all  in the same direction at a particular time of year (which was happening 2007-12, but has faded away with the 13-14 rebound). Sometimes up and sometimes down is what anomalies normally do and isn't remarkable.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #109 on: November 16, 2014, 05:59:43 AM »
.
[]

Neven

  • Administrator
  • First-year ice
  • Posts: 9470
    • View Profile
    • Arctic Sea Ice Blog
  • Liked: 1333
  • Likes Given: 617
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #110 on: November 16, 2014, 11:40:39 AM »
Now there's a nice little hot spot  in the Beaufort and Chukchi:



Is that all from Nuri?
The enemy is within
Don't confuse me with him

E. Smith

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #111 on: November 16, 2014, 02:15:03 PM »
.
[]

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #112 on: November 16, 2014, 05:51:43 PM »
Now there's a nice little hot spot  in the Beaufort and Chukchi

Is that all from Nuri?

Not any more; however, there is a strong high pressure system in SE Alaska and the gulf which established itself in Nuri's wake, which is pulling heat into the region. As you can see, the Arctic is as much above normal as the U.S. is below.

The cold we are currently seeing across the central part of the continent is the cold air displaced by that flow.

A pattern appears to be settling in where warm low pressure systems are being swept through the western an central Bering across Alaska followed by Gulf of Alaska ridging.
This space for Rent.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #113 on: November 16, 2014, 06:07:07 PM »
[]

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #114 on: November 16, 2014, 07:21:02 PM »
Neven, JD,

Do you have a source of current temperatures from NCEP/NCAR operational at various atmospheric levels?

Taking NCEP/NCAR reanalysis the latest data available is 13 November. Looking at atmospheric cross section north of 60degN, and between 180degE and 240 degE reveals the sort of low lying warming (15degC above average!) associated with open water or thin ice. However on that date the winds were from Alaska into the Arctic, whereas looking at the latest HRPT IR from Environment Canada shows winds from the Arctic Ocean to Alaska.



Notably, along with the low lying intense warming there is also warming aloft, this higher warming is indicative of warm air influx from outside the Arctic, and at 500/850mb there's a strong northwards flow over the Bering Straits.

plinius

  • Frazil ice
  • Posts: 403
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #115 on: November 17, 2014, 01:34:03 AM »
sure it is an open water signature? Goes pretty much inland (Barrow is at 71 degrees) and you will see the same anomaly when strong winds break up the inversion.
Wind turned quickly because of a high pressure ridge forming quickly north of the coast, also because of the WLA.


jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #116 on: November 17, 2014, 01:38:03 AM »
Neven, JD,

Do you have a source of current temperatures from NCEP/NCAR operational at various atmospheric levels?

Nothing that sophisticated, Chris; just watching NOAA and the U of Maine models like Reanalyzer and the like.

Your are correct to point out the different source of heat - the open water in the Beaufort and Chukchi.  Not just the circulation producing the increases in temperature.  All in all, a lot of heat to dump.
This space for Rent.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #117 on: November 17, 2014, 04:25:01 AM »
.
[]

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #118 on: November 17, 2014, 02:50:14 PM »
[]

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #119 on: November 17, 2014, 07:50:53 PM »
sure it is an open water signature? Goes pretty much inland (Barrow is at 71 degrees) and you will see the same anomaly when strong winds break up the inversion.
Wind turned quickly because of a high pressure ridge forming quickly north of the coast, also because of the WLA.

Plinius,

A reasonable point. I just understand the climatological inversion to be thinner than this warm feature. See attachment. So disruption of the inversion shouldn't IMO produce a low lying warming to 850mb. My meteorology isn't strong enough to decide whether an influx of warmer air breaking down the inversion would be low level biassed as seen.

Rather than open water I see thin ice.
http://www7320.nrlssc.navy.mil/GLBhycomcice1-12/navo/arcticictn/nowcast/ictn2014111212_2014111300_923_arcticictn.001.gif
However there seems to be greater extent of thin ice over Siberia/Atlantic sector, that there is no such low level warming there might support the interpretation of wholly atmosphere driven with no ice role. However the above image (HYCOM) suggests thinner ice in Beaufort/Chukchi which might go some way to explaining the difference.

That said now I'm starting to think it is a wholly atmospheric feature with negligible ice role. But I don't know enough to cut through the matter and provide a totally convincing answer.


ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #120 on: November 17, 2014, 08:19:03 PM »
Related to the above. In my most recent blog post I graphed air temperature (NCEP/NCAR reanalysis) at three different levels for an area north of 70degN and from 90degE to 240degE, an area preferring the peripheral seas of the Arctic Ocean.



I've just re-done that for November same area etc (had the numbers in a spreadsheet).



What's odd there is how in November the low level warming kicks off in around 1995 (the start of the recent period of rapid volume loss in PIOMAS - coincidence?), whereas in October there's more of an exponential warming behaviour which seems to be related to ice loss in the summer.

I'll update that graph when November data is in.

plinius

  • Frazil ice
  • Posts: 403
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #121 on: November 17, 2014, 08:31:29 PM »
hi Chris,

I am not certain that we could quantify that effect. Killing the inversion in my wider sense of use implied that you move towards a dry adiabatic vertical gradient, as usual for a breaking through Foehn/Chinook wind. As you can see from your plot, the climatology is at least near isothermal, so it is easy to produce a lower level hot spot of >5K, when such Foehn winds happen to blow north over the Alaskan mountain ranges. I do fully agree with you that the thin ice plays it's role (though cannot be all, as models now play temperatures below freezing, since the "Foehn" has stopped), but quantification would only be possible by looking at full spatial resolution and the dynamics.

Concerning last post - fully agree and this is also my main reason for doubt that an ice free arctic would easily return to a fully covered state nowadays. Large open water area would massively shield itself with WV and local heat bubble and reduce the number of freeze days. Don't trust that the current models can fully handle the changes in circulation playing in there.

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #122 on: November 17, 2014, 08:46:45 PM »
Well Plinius,

There's often two ways to look at things...

When I look at those graphs I see massive venting of ocean heat which will assist the sea ice in recovering from loss events and act as a negative feedback on ice loss.  ;D

Laurent

  • Young ice
  • Posts: 2546
    • View Profile
  • Liked: 13
  • Likes Given: 50
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #123 on: November 17, 2014, 09:00:32 PM »

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #124 on: November 17, 2014, 09:30:31 PM »
PIOMAS estimates medio November

Code: [Select]
2014 305   9.782
2014 306   9.883
2014 307  10.030
2014 308  10.092
2014 309  10.249
2014 310  10.344
2014 311  10.512
2014 312  10.672
2014 313  10.683
2014 314  10.705
2014 315  10.830
2014 316  11.000
2014 317  11.222
2014 318  11.230
2014 319  11.154
2014 320  11.191

And as corrected on November 19th:

Code: [Select]
2014 305   9.782
2014 306   9.889
2014 307   9.984
2014 308  10.070
2014 309  10.158
2014 310  10.239
2014 311  10.321
2014 312  10.403
2014 313  10.490
2014 314  10.583
2014 315  10.674
2014 316  10.761
2014 317  10.851
2014 318  10.934
2014 319  11.011
2014 320  11.076
2014 321  11.106
2014 322  11.168
« Last Edit: November 19, 2014, 01:36:10 PM by viddaloo »
[]

P-maker

  • Frazil ice
  • Posts: 389
    • View Profile
  • Liked: 72
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #125 on: November 17, 2014, 10:42:37 PM »
Chris & Plinius

Before both of you go off on various tangents, please consider actual observations from the North shore of Alaska - such as these: http://pafg.arh.noaa.gov/wmofcst_pf.php?wmo=FPAK51PAFG&type=public

It appears to me, that the dominant weather these days is "freezing drizzle".

Comparing these observations to the last three weeks of drizzle on end in this country, just makes me wonder wether we are now entering the new phase of equi-thermal NH autumns.

As far as I reckon, this type of drizzle weather is just the way a warm and moist atmosphere would most efficiently get rid of it's surplus energy before the winter. A slow condensation process in a weak southerly air flow regime would give drizzle on the ground and positive temperature anomalies aloft.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #126 on: November 18, 2014, 12:47:08 AM »
What's odd there is how in November the low level warming kicks off in around 1995 (the start of the recent period of rapid volume loss in PIOMAS - coincidence?), whereas in October there's more of an exponential warming behaviour which seems to be related to ice loss in the summer.

It seems if you're going to "lose weight" during the Arctic winter, November is the month to do it. 1996 is the all–time leader in this game, putting off an impressive 778 km³ in the course of the month. (That's growing slower, mind you, and not actual melt.)

I bet we'll see this 778 km³ record broken before the end of this decade.
« Last Edit: November 18, 2014, 12:52:49 AM by viddaloo »
[]

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #127 on: November 18, 2014, 09:25:37 AM »
I have to post this update, as the already biggest November drop in sea ice gain anomalies this century just keeps getting bigger, day by day. True, the volume figures are based on my own algorithm, so we won't know for sure how big the drop was till PIOMAS data arrive. But the smoothing algorithm I made after October's data were in actually makes the drop even more severe, so I think we've got something here.

You could of course say that because this appears to be a record for the new century, the odds are the real picture is somewhere between this graph and no drop at all. But the smoothing function that made the October estimates more precise, says otherwise, and predicts a more extreme drop than in this graph. Wish I could Fast Forward to December 8th–ish!
[]

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #128 on: November 18, 2014, 02:18:28 PM »
[]

Peter Ellis

  • Grease ice
  • Posts: 619
    • View Profile
  • Liked: 33
  • Likes Given: 14
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #129 on: November 18, 2014, 04:48:28 PM »
It seems if you're going to "lose weight" during the Arctic winter, November is the month to do it. 1996 is the all–time leader in this game, putting off an impressive 778 km³ in the course of the month. (That's growing slower, mind you, and not actual melt.)

I bet we'll see this 778 km³ record broken before the end of this decade.

Surely the causality is the other way round?  In recent years we tend to have faster re-freeze across the board in Oct/Nov, because we're starting from a lower base.  In 1996, the Arctic Basin was already pretty full of ice by the start of November, so there wasn't so much scope for growth.  Ergo slower growth in 1996 than in more recent years.

Accordingly I don't think we'll see the 778 km³ record broken ever again (well not until we get near to ice-free winters, by which time I doubt there'll be much human life around to comment! Rather, we'll keep setting ever higher and higher record for more ice growth in November.

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #130 on: November 18, 2014, 05:39:46 PM »

Surely the causality is the other way round?  In recent years we tend to have faster re-freeze across the board in Oct/Nov, because we're starting from a lower base.  In 1996, the Arctic Basin was already pretty full of ice by the start of November, so there wasn't so much scope for growth.  Ergo slower growth in 1996 than in more recent years.

Accordingly I don't think we'll see the 778 km³ record broken ever again (well not until we get near to ice-free winters, by which time I doubt there'll be much human life around to comment! Rather, we'll keep setting ever higher and higher record for more ice growth in November.

I agree we 'have faster re-freeze across the board in Oct/Nov, because we're starting from a lower base', at least so far.

But if there is much more heat built up over summer as we approach virtually ice free summer, might the heat being vented last into Nov so growth is then slow during October and at start of Nov before speeding up?



In addition to the mainly downward movement, the bottom is getting flatter and the sides steeper.
Hence the trend towards more ice growth in October certainly could reverse (quite soon?) but perhaps November is less clear and likely will take a long time before the trend could reverse but not quite as long as you suggest.

Steven

  • Grease ice
  • Posts: 957
    • View Profile
    • Arctic sea ice data and graphs
  • Liked: 481
  • Likes Given: 19
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #131 on: November 18, 2014, 06:03:27 PM »
I have to post this update, as the already biggest November drop in sea ice gain anomalies this century just keeps getting bigger, day by day. True, the volume figures are based on my own algorithm, so we won't know for sure how big the drop was till PIOMAS data arrive. But the smoothing algorithm I made after October's data were in actually makes the drop even more severe, so I think we've got something here.

I don't think so.

Your estimates of daily PIOMAS volume for November 2014 are based on IJIS sea ice extent.

Your assumption is that the volume to extent ratio follows basically the same trajectory as 5 years ago, i.e. November 2009.

So I looked at the November 2009 data.  It turns out that the volume/extent ratio has an unusual drop from 13 to 17 November 2009.  This drop propagated through your calculation and it resulted in a spurious drop in your calculated 2014 curve.

Peter Ellis

  • Grease ice
  • Posts: 619
    • View Profile
  • Liked: 33
  • Likes Given: 14
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #132 on: November 18, 2014, 07:31:39 PM »
In addition to the mainly downward movement, the bottom is getting flatter and the sides steeper.
Hence the trend towards more ice growth in October certainly could reverse (quite soon?) but perhaps November is less clear and likely will take a long time before the trend could reverse but not quite as long as you suggest.

I agree the bottom's getting flatter and wider, but that I think really only compounds the issue in November - a wide flat bottom to the curve through October only means that November starts from even lower, meaning there will be a faster catch-up freeze.

Even in Hudson Bay (much lower latitude, ice free all summer, and soaking up mid-day sun non-stop from ~July onward) shows an end to the trough and a steep upward climb through November.  To shift the curve so profoundly that November is in the shallow part of the immediate post-trough growth period, you'd need there to be substantial open water right across the Arctic for multiple months, right up to the end of October or nearly so.

i.e. you'd need the entire Arctic to have a yearly profile that looks much more like the Baffin sea or the northern Bering sea.  And by that point, as with those areas, you're into a regime where you have declining winter maxima as well as summer minima.  Not going to happen for multiple decades yet.  Winter in the land of 24 hr darkness is just too cold.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #133 on: November 19, 2014, 08:11:29 AM »
Steven,

So I looked at the November 2009 data.  It turns out that the volume/extent ratio has an unusual drop from 13 to 17 November 2009.  This drop propagated through your calculation and it resulted in a spurious drop in your calculated 2014 curve.

Edit: I've added a smoothing function for the IJIS *and* PIOMAS 2009 data. Hopefully that will yield more precise estimates.

Edit2: You are right, of course. There was a huge natural variation rise in ext/vol for November 13 to 17, resulting in an unnatural false drop in 2014 gain for those same days. I'm thankful for the correction, Steven. Attached you'll find the graph for real Nov 2009 ext/vol versus smoothed.

Edit 3: And today's updated abrupt drops graph with smooth data! :)
« Last Edit: November 19, 2014, 01:21:02 PM by viddaloo »
[]

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #134 on: November 19, 2014, 08:48:13 AM »
It seems if you're going to "lose weight" during the Arctic winter, November is the month to do it. 1996 is the all–time leader in this game, putting off an impressive 778 km³ in the course of the month. (That's growing slower, mind you, and not actual melt.)

I bet we'll see this 778 km³ record broken before the end of this decade.

Surely the causality is the other way round?  In recent years we tend to have faster re-freeze across the board in Oct/Nov, because we're starting from a lower base.

It's not that simple, I'm afraid. If you look at this graph for November:



then the yellow line being lower than the black and green line, tells you the trend is for lower and lower November ice gains. The black 2007–2013/14 line tells you gains are higher than for the 1979–1999 average, but the trend has since turned towards slower gains again.

For October it's a whole different story, as October's gains are more and more determined by summer melt:

[]

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #135 on: November 19, 2014, 05:29:39 PM »

It's not that simple, I'm afraid. If you look at this graph for November:


then the yellow line being lower than the black and green line, tells you the trend is for lower and lower November ice gains. The black 2007–2013/14 line tells you gains are higher than for the 1979–1999 average, but the trend has since turned towards slower gains again.


Sorry no, you are fooling yourself by looking at just a few comparisons in the last few years and deducing the trend for November is downwards.

PIOMASONDGains by crandles2011, on Flickr

The trend for the gain in each of the three months is similar generally upward. Looking at the last few years and saying the trend is downwards is highly likely to be noise giving a spurious result. (You can see the large amounts of noise in this graph.) Last couple of years having higher minimum against the trend also makes looking at just a few comparisons to deduce the trend dangerous.

The graph I posted earlier with discussion of main movement being downwards but also a wider bottom and steeper sides gives much better understanding of the situation.

Just because the trends are similar so far does not stop us understanding that it is October that is going to turn downwards first and it might not take much lower minimums to cause that to occur.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #136 on: November 19, 2014, 10:49:27 PM »
The trend for the gain in each of the three months is similar generally upward. Looking at the last few years and saying the trend is downwards is highly likely to be noise giving a spurious result. (You can see the large amounts of noise in this graph.) Last couple of years having higher minimum against the trend also makes looking at just a few comparisons to deduce the trend dangerous.

Not sure I agree with you there, if we are to let the numbers themselves decide.

There seems to be a sea change at or just before 2007, from which year the first half of the X to 2014 years will have more rapid ice gain than the second half. That looks like a trend to me.

That means November 2007–2010 will have more ice growth than November 2011–2014, 2008–2010 more than 2011–2014, 2009–2011 more growth than 2012–2014, 2010–2012 more than 2013–14 and finally 2011–2012 more rapid ice gain than 2013–2014.

If you include more than 8 years, the trend is the opposite (last half of the selection will have quicker growth).

The graph I posted earlier with discussion of main movement being downwards but also a wider bottom and steeper sides gives much better understanding of the situation.

Just because the trends are similar so far does not stop us understanding that it is October that is going to turn downwards first and it might not take much lower minimums to cause that to occur.

This doesn't seem to be based in the data, Crandles. The trend is for faster melt in May and June, slower melt in July, August and September, continuing in faster refreeze in October, turning to slower refreeze in November, then back again to faster refreeze in December and April (with January to March in the 'Undecided' category).

My guess is that January to March will be the first to tip over into predominantly slower growth than the earlier years in the data, while October will keep growing ever faster until the summer crash expands into October 1st. Data suggest August will crash before October, though.
[]

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #137 on: November 19, 2014, 11:23:06 PM »

Not sure I agree with you there, if we are to let the numbers themselves decide.

Trouble is this years data does help decide. I say the trend to lower minimums are causing faster refreeze. But this year has a high minimum therefore I expect low refreeze this year. You expect low refreeze this year because you think the trend is down. So if the refreeze is low we both claim the data supports our position. So we need to wait for a low minimum year to decide the matter and even then 2 or 3 low minimum years is better than one.

Quote
My guess is that January to March will be the first to tip over into predominantly slower growth than the earlier years in the data, while October will keep growing ever faster until the summer crash expands into October 1st. Data suggest August will crash before October, though.

Interesting maybe. For this year I expect the regrowth in Jan to March to be slow and that could well be by a larger amount than the trend which I agree could emerge as downward.

I don't think it will take minimums much below the current minimum for slow growth to extend into early October, as 2007 and 2009 have already had pretty slow growth in the early part of October.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #138 on: November 19, 2014, 11:57:05 PM »
I think I've shown in the Transition thread how the first half of October is trending towards lower refreeze (ending in zero refreeze somewhere in the 2020s when there is no ice at all before medio October). Here it is:

[]

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #139 on: November 20, 2014, 04:34:51 AM »
I think I've shown in the Transition thread how the first half of October is trending towards lower refreeze (ending in zero refreeze somewhere in the 2020s when there is no ice at all before medio October). Here it is:
To my eye, that has the appearance of serious volatility.  I suspect  that the refreeze in the peripheral seas is at the root of that.  What does the scatter look like when we isolate it by region?  The CAB might be less subject to the variations we're looking at.
This space for Rent.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #140 on: November 20, 2014, 02:30:49 PM »
New methane peak yesterday, 2495 ppb.
[]

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #141 on: November 20, 2014, 08:45:52 PM »
Using:
1) PIOMAS gridded data for the Arctic Ocean, the difference December and September volume is used to calculate the autumn volume gain.
2) NCEP/NCAR surface temperature for the region 90 to 240degE north of 70degN.
3) NSIDC sea ice extent on 31 August, known as Late Summer Extent.

A)


B)


C)


The conclusion I draw is that:

There is a moderate to strong relationship (R2=0.6) between sea ice volume gain in the autumn and the late summer extent of sea ice, such that as late summer extent goes down autumn volume gain goes up.

Temperature has a weak effect on autumn volume gain (R2=0.195), and would seem to act against what one would expect, as autumn warming occurs so autumn volume increase increases.

However temperature is seen to be moderately to strongly affected by late summer sea ice extent (R2=0.5), this can be explained by increasing open water leading to more heat being vented to the atmosphere in the autumn.

Therefore the increase in autumn volume gain is due to larger open water in summer and happens in spite of the heat gained in summer which is vented to the atmosphere in the autumn. This venting of heat does nothing to reduce autumn volume gains.

crandles

  • Young ice
  • Posts: 3379
    • View Profile
  • Liked: 239
  • Likes Given: 81
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #142 on: November 20, 2014, 11:46:33 PM »
Using:
1) PIOMAS gridded data for the Arctic Ocean, the difference December and September volume is used to calculate the autumn volume gain.
Is that difference between average of September and average of December?


Therefore the increase in autumn volume gain is due to larger open water in summer and happens in spite of the heat gained in summer which is vented to the atmosphere in the autumn.

Seems a sensible conclusion.

This venting of heat does nothing to reduce autumn volume gains.

I wonder if that is too absolute a statement from the evidence provided.

If you looked at piomas gains from say ~25 Sept to ~10 Oct, I think it may be possible to arrive at a slightly different nuanced conclusion: It might be that this brief initial period does see a delay but the amount is made up for by faster gains later in the Autumn in order to reach the results you have given.

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #143 on: November 21, 2014, 07:49:24 AM »
Crandles,

Yes, PIOMAS monthly, average is close to that.

Perhaps the suggestion that heat loss does nothing to stem volume gain is too strong. But I don't really think so. Yes there is a delay as heat is vented until the ocean cools, yet despite this the overall tendency is for lower summer extents to cause greater autumn volume gain.

OSweetMrMath

  • Frazil ice
  • Posts: 137
    • View Profile
  • Liked: 0
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #144 on: November 22, 2014, 03:05:32 AM »
Chris,

This is consistent with your previous evidence that ice grows fastest when it's thin. Years with low extent will start fast ice growth when the temperature falls below freezing. Also, it's easier for open water to transfer heat to the atmosphere, so it's not surprising that years with low extent have both higher air temperatures in the fall and greater volume growth in the fall.

Eventually the ocean will get warm enough to prevent ice formation, but that's a second order effect and is probably a long way out.

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #145 on: November 22, 2014, 06:08:54 AM »
Flat–out normalized: The past 2 Novembers have normalized the big post–2007 increase in November ice gain down to an anomaly of zero.

This could mean we can now have 5th lowest ice levels (annual average volume, or 4th for extent) and still have no higher ice growth in November than the 1979–1999 average.

It could also mean that a compensating "quicker growth to make up for a larger melt" is really a short–term effect, lasting only a few years, until things normalize on a new level.
[]

jdallen

  • Young ice
  • Posts: 3410
    • View Profile
  • Liked: 650
  • Likes Given: 244
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #146 on: November 22, 2014, 08:42:19 AM »
It could also mean that a compensating "quicker growth to make up for a larger melt" is really a short–term effect, lasting only a few years, until things normalize on a new level.

Indeed it could be.  We will have to see how the energy balance in the ocean plays out.

I've been watching the Bering, Chukchi and Beaufort with increasing alarm.  The region is still extraordinarily hot, and has been for weeks. 

Chris, where was it you were getting your degree day information?

-J
This space for Rent.

wili

  • Young ice
  • Posts: 3342
    • View Profile
  • Liked: 602
  • Likes Given: 409
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #147 on: November 22, 2014, 01:14:42 PM »
"I've been watching the Bering, Chukchi and Beaufort with increasing alarm.  The region is still extraordinarily hot, and has been for weeks."

That further tips the 'cold pole' towards GIS and CAA, doesn't it?
"A force de chercher de bonnes raisons, on en trouve; on les dit; et après on y tient, non pas tant parce qu'elles sont bonnes que pour ne pas se démentir." Choderlos de Laclos "You struggle to come up with some valid reasons, then cling to them, not because they're good, but just to not back down."

viddaloo

  • Nilas ice
  • Posts: 1302
  • Hardanger Sometimes
    • View Profile
  • Liked: 1
  • Likes Given: 0
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #148 on: November 22, 2014, 02:33:16 PM »
The drop in anomalies from the 79–99 average has ended, and the drop from 2002–2013 looks set to do the same in the near future. The first drop ended at –178 km³ and was the 3rd biggest such November drop since 2007, and the 5th biggest this century (with 2006' –323 km³ as #1). In the 20th century, 1981 was bigger with –343 km³ and 1996 with a drop of –778 km³ in November.

The volume deviation from the 2002–13 gain average is currently at –518 km³, but looks destined to narrow towards the end of the month. Final PIOMAS data in 2 weeks' time will hand us the official tally. Thereafter, December will be an interesting month to follow through these graphs, as 2013 volume shows a big rise and 2013 extent an even bigger drop, compared to the 2002–13 December gain average.



« Last Edit: November 22, 2014, 02:57:04 PM by viddaloo »
[]

ChrisReynolds

  • Nilas ice
  • Posts: 1764
    • View Profile
    • Dosbat
  • Liked: 20
  • Likes Given: 9
Re: Arctic Refreeze Discussion: 2014/2015
« Reply #149 on: November 22, 2014, 03:48:16 PM »
It could also mean that a compensating "quicker growth to make up for a larger melt" is really a short–term effect, lasting only a few years, until things normalize on a new level.

Indeed it could be.  We will have to see how the energy balance in the ocean plays out.

I've been watching the Bering, Chukchi and Beaufort with increasing alarm.  The region is still extraordinarily hot, and has been for weeks. 

Chris, where was it you were getting your degree day information?

-J

What? Freezing degree days?

Ncep/ncar reanalysis time series. I can post a link if needed, but am on my phone right now. That data does reveal such warnings so they're taken into account, but I was using the average for the ESS and haven't got round to dealing with daily gridded data.

Edit- just been looking at earth nullschool and GFS at Wetterzentrale. It is warm there but that's relative to climatology, i.e. Only a little above freezing but enough to explain what looks like late freeze in chuckchi/bering, not checked the numbers on that. What I don't get is that earthnullschool is showing winds blowing south through Bering. Whereas gfs pressure suggests a high ridge drawing air broadly northwards.
« Last Edit: November 22, 2014, 04:10:43 PM by ChrisReynolds »