Please support this Forum and Neven's Blog

Poll

What will the CT 2013 Arctic SIA daily minimum be?

More than 3.5 million km2
0 (0%)
Between 3.25 and 3.5 million km2
1 (1.5%)
Between 3.0 and 3.25 million km2
5 (7.6%)
Between 2.75 and 3.0 million km2
9 (13.6%)
Between 2.5 and 2.75 million km2
8 (12.1%)
Between 2.25 and 2.5 million km2
6 (9.1%)
Between 2.0 and 2.25 million km2
13 (19.7%)
Between 1.75 and 2.0 million km2
10 (15.2%)
Between 1.5 and 1.75 million km2
4 (6.1%)
Between 1.25 and 1.5 million km2
4 (6.1%)
Between 1.0 and 1.25 million km2
2 (3%)
Between 0.75 and 1.0 million km2
2 (3%)
Between 0.5 and 0.75 million km2
0 (0%)
Between 0.25 and 0.5 million km2
0 (0%)
Between 0 and 0.25 million km2
2 (3%)

Total Members Voted: 65

Voting closed: July 20, 2013, 10:38:58 PM

Author Topic: Cryosphere Today 2013 Arctic SIA daily minimum: July poll  (Read 51795 times)

slow wing

  • ASIF Middle Class
  • Posts: 500
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #50 on: July 13, 2013, 11:55:32 AM »
1.75-2.00 million km^2.

  That's the same bin as I picked 2 months ago, alternating with the bin below it that I picked last month as well as 3 months ago.

  I went up a bin this month as the melt has been a bit slow and the ice volume has gained relative to last year. Also, the broken up ice from PAC 2013 hasn't yet disintegrated into holes near the Pole.

  PIOMAS reconstructs more volume now than at the same time last year. However, the direct sunlight from the current HIGH on the Pacific side might be expected to put paid to that difference, and maybe then some.

  Then I'm guessing that this year should pull below 2012, all else being equal, due to the smashed up ice on the Eastern side of the Arctic and around near the Pole, as well as all the ice on the Pacific side that is beginning to show melt ponds on the MODIS display.


  As a predicted value and sigma, I would choose 2.0 +/- 0.7 million km^2. That is, about a 2/3 chance of the area ending up between 1.3 and 2.7 million km^2.


  I see that I am somewhat on the lower side of predictions. On looking at the comments, there seems to be a gap between, on the one hand, those who look predominantly at the current quantitative data for sea ice area and volume and, on the other hand, those such as myself who also give significant weight to more qualitative factors such as the ice quality and distribution and the emerging high pressure system.

« Last Edit: July 13, 2013, 12:02:40 PM by slow wing »

BornFromTheVoid

  • ASIF Middle Class
  • Posts: 929
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #51 on: July 13, 2013, 01:23:34 PM »
After another large drop of 148k, we're now within 1 million km2 of 5 previous annual minima.


AndrewP

  • ASIF Citizen
  • Posts: 129
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #52 on: July 15, 2013, 07:24:45 AM »
Chris, I'll say it again, it defies even the most basic common sense that expectations could change by over 1.5 million sq km in the space of 3 weeks.

The fact is with volume already running well above 2012, and 2012 having above average losses in area relative to the May volume, a prediction below 2012 was already completely unwaranted.

The starting point for any prediction should have been an average of years with similar volume. 2010, 2011 and 2012. Such an average was substantially above 2012. And the volume gap with 2012 had clearly continued to widen in the first 3 weeks of June.

There was no objective reason for predicting less area than 2012, when May 30th 2013 had greater volume than 2012, 2011 and was on par with 2010 and melt was slow through the first 3 months of June.

I would also point out that your June area regressions are being messed up by the fact that volume has dropped so much even in the years 2007-2012. An ANOVA analysis which included volume and possibly other factors would likely find more robust results and more of the variance at SIA minimum attributable to the June area variable.

Performing a regression 2007-2012 is just silly. June 2007 had the 2nd lowest SIA but far more volume than any of the other years. Given that small sample size, that year alone might actually cause you to find a reverse correlation with June SIA to minimum SIA. An ANOVA should help to sort out what of the variation is due to SIA and what is due to SIV. By sorting out the SIV component, it will likely find much more variation attributable to June SIA.

A correlation of only .34 on June 29th defies common sense. Clearly this result is being distorted by the small sample size and the incredible variation in volume between years which is clouding out any other correlations that exist.

It was pretty inconceivable that SIA would have caught up to 2012 in the space of 2 weeks.


I will try to perform a better regression tomorrow. Honestly though, the sample sizes are too small and the variations in volume so great that it will be very difficult to tease out any correlations to June SIA, even though there are sound reasons for these correlations to exist.

AndrewP

  • ASIF Citizen
  • Posts: 129
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #53 on: July 15, 2013, 07:27:51 AM »
Chris, I'll say it again, it defies even the most basic common sense that expectations could change by over 1.5 million sq km in the space of 3 weeks.

The fact is with volume already running well above 2012, and 2012 having above average losses in area relative to the May volume, a prediction below 2012 was already completely unwaranted.

The starting point for any prediction should have been an average of years with similar volume. 2010, 2011 and 2012. Such an average was substantially above 2012. And the volume gap with 2012 had clearly continued to widen in the first 3 weeks of June.

There was no objective reason for predicting less area than 2012, when May 30th 2013 had greater volume than 2012, 2011 and was on par with 2010 and melt was slow through the first 3 months of June.

I would also point out that your June area regressions are being messed up by the fact that volume has dropped so much even in the years 2007-2012. An ANOVA analysis which included volume and possibly other factors would likely find more robust results and more of the variance at SIA minimum attributable to the June area variable.

Performing a regression 2007-2012 is just silly. June 2007 had the 2nd lowest SIA but far more volume than any of the other years. Given that small sample size, that year alone might actually cause you to find a reverse correlation with June SIA to minimum SIA. An ANOVA should help to sort out what of the variation is due to SIA and what is due to SIV. By sorting out the SIV component, it will likely find much more variation attributable to June SIA.

A correlation of only .34 on June 29th defies common sense. Clearly this result is being distorted by the small sample size and the incredible variation in volume between years which is clouding out any other correlations that exist.

It was pretty inconceivable that SIA would have caught up to 2012 in the space of 2 weeks.


I will try to perform a better regression tomorrow. Honestly though, the sample sizes are too small and the variations in volume so great that it will be very difficult to tease out any correlations to June SIA, even though there are sound reasons for these correlations to exist.


jdallen

  • ASIF Upper Class
  • Posts: 2333
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #54 on: July 15, 2013, 07:52:46 AM »
With respect, almost everything about the ice defies expectations currently.

I don't think we have a good enough handle on all the variables, in particular, *actual* volume, and *actual* energy being applied to the ice, to effectively predict.

In the space of a week, we have almost caught up with 2012.  There's reasonable probability that by the end of next, we may be equal or be past 2012 at the same date.

This space for Rent.

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #55 on: July 15, 2013, 08:12:08 AM »
My replies in bold.

Chris, I'll say it again, it defies even the most basic common sense that expectations could change by over 1.5 million sq km in the space of 3 weeks.

Common sense is often wrong. I've explained my rationale.

The fact is with volume already running well above 2012, and 2012 having above average losses in area relative to the May volume, a prediction below 2012 was already completely unwaranted.

The starting point for any prediction should have been an average of years with similar volume. 2010, 2011 and 2012. Such an average was substantially above 2012. And the volume gap with 2012 had clearly continued to widen in the first 3 weeks of June.

There was no objective reason for predicting less area than 2012, when May 30th 2013 had greater volume than 2012, 2011 and was on par with 2010 and melt was slow through the first 3 months of June.

My original prediction was a judgment (heuristic) based on sea ice thickness and the relationship between end of season volume and CT Area, thickness as indicated by PIOMAS, ASCAT and the DAM done back in April or May, (forget which). I needed to hold until end June, and had said I needed to in May, before reassessing. To have changed my opinion before that data was in would not have been sensible

I would also point out that your June area regressions are being messed up by the fact that volume has dropped so much even in the years 2007-2012. An ANOVA analysis which included volume and possibly other factors would likely find more robust results and more of the variance at SIA minimum attributable to the June area variable.

Performing a regression 2007-2012 is just silly. June 2007 had the 2nd lowest SIA but far more volume than any of the other years. Given that small sample size, that year alone might actually cause you to find a reverse correlation with June SIA to minimum SIA. An ANOVA should help to sort out what of the variation is due to SIA and what is due to SIV. By sorting out the SIV component, it will likely find much more variation attributable to June SIA.

A correlation of only .34 on June 29th defies common sense. Clearly this result is being distorted by the small sample size and the incredible variation in volume between years which is clouding out any other correlations that exist.

I didn't assert that the correlations were statistically significant for 2007 to 2012 (after 9 July). However indicative of their being sound is the similarity of the curve to the correlations for the whole period 1979 to 2012 (after 19 June). I have enclosed in brackets the date that correlations start to be statistically significant at 90%.



The more rapid increase in correlations in June is to be expected given the higher rates of change of anomalies of CT Area (1980 to 1999 baseline).



Again common sense is often wrong. If you look at the June cliff the break point in correlations becomes apparent. Note that as I said in my earlier comment, I detrended before making correlations. ANOVA? Good idea why not do it.


It was pretty inconceivable that SIA would have caught up to 2012 in the space of 2 weeks.

Look at the size of the June 2012 CT Area cliff, which happened in a similarly short time. It would no have been sensible to write off chances of a massive loss in 2013 until I had seen how that loss would play out in 2013.


I will try to perform a better regression tomorrow. Honestly though, the sample sizes are too small and the variations in volume so great that it will be very difficult to tease out any correlations to June SIA, even though there are sound reasons for these correlations to exist.

To be clear it was correlation in ten day intervals from April to September with the sea ice area on day 250 (leap days omitted).

Bob Wallace

  • ASIF Upper Class
  • Posts: 1123
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #56 on: July 15, 2013, 10:18:02 AM »
With respect, almost everything about the ice defies expectations currently.

I don't think we have a good enough handle on all the variables, in particular, *actual* volume, and *actual* energy being applied to the ice, to effectively predict.

In the space of a week, we have almost caught up with 2012.  There's reasonable probability that by the end of next, we may be equal or be past 2012 at the same date.

And *actual* "quality".

We have no measurement of the physical characteristics of the ice.  Be it rock or fluff?

Shared Humanity

  • ASIF Upper Class
  • Posts: 1754
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #57 on: July 15, 2013, 04:28:17 PM »
CT update is out. Looking at the the sea ice concentration maps, is it normal to have so many areas of 60% to 70% concentration spread across the CAB?

Nightvid Cole

  • ASIF Citizen
  • Posts: 419
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #58 on: July 15, 2013, 05:16:46 PM »
CT update is out. Looking at the the sea ice concentration maps, is it normal to have so many areas of 60% to 70% concentration spread across the CAB?

No, it is not normal. It is a result of a mechanically weak ice pack that cannot keep itself solid. The extremely large percentage of first year ice allows huge stresses to build up and cause it to crumble.

This is why I don't expect an increase in SIA over 2012.

Peter Ellis

  • ASIF Middle Class
  • Posts: 501
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #59 on: July 15, 2013, 05:40:43 PM »
What's you're evidence, Nightvid?

http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=07&fd=12&fy=1981&sm=07&sd=12&sy=2013

Stepping through this comparing 2013 to each year in turn, 2013 doesn't seem particularly unusual.  For example, I've linked to 1980 - while the overall extent was obviously higher then, the amount of "red" within the pack is much higher than 2013.  Comparing to 2012, I'd say that 2013 looks better in every respect in the CT image - higher area and higher concentration.

Obviously other satellites are telling different parts of the story, but we mustn't be tempted to go beyond the data for any particular source.

jdallen

  • ASIF Upper Class
  • Posts: 2333
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #60 on: July 15, 2013, 06:35:23 PM »
What's you're evidence, Nightvid?

http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=07&fd=12&fy=1981&sm=07&sd=12&sy=2013

Stepping through this comparing 2013 to each year in turn, 2013 doesn't seem particularly unusual.  For example, I've linked to 1980 - while the overall extent was obviously higher then, the amount of "red" within the pack is much higher than 2013.  Comparing to 2012, I'd say that 2013 looks better in every respect in the CT image - higher area and higher concentration.

Obviously other satellites are telling different parts of the story, but we mustn't be tempted to go beyond the data for any particular source.


Summaries can be deceiving.

Year vs year.  2013.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?subset=Arctic_r04c04.2013195.terra

2012, same day.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?subset=Arctic_r04c04.2012195.terra

2013, complete.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?mosaic=Arctic.2013195.terra.1km

2012, complete.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?mosaic=Arctic.2012195.terra.1km

2011, complete.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?mosaic=Arctic.2011195.terra.1km

2010, complete.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?mosaic=Arctic.2010195.terra.1km

2009, complete.
http://lance-modis.eosdis.nasa.gov/imagery/subsets/?mosaic=Arctic.2009195.terra.1km


Pity we don't have 2007.

This space for Rent.

Peter Ellis

  • ASIF Middle Class
  • Posts: 501
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #61 on: July 15, 2013, 07:16:55 PM »
As I said, "obviously other satellites are telling different parts of the story".

CT doesn't give any indication that this year is more fragmented than other years.  MODIS does, but the historical record is very short.  Moreover, if you looked at different parts of the MODIS mosaic (e.g. r05c02 in the Beaufort, r05c03 in the northern Chukchi or r05c04 in the northern Laptev), you'd come to very different conclusions about this year relative to 2012.

jdallen

  • ASIF Upper Class
  • Posts: 2333
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #62 on: July 15, 2013, 08:07:00 PM »
As I said, "obviously other satellites are telling different parts of the story".

CT doesn't give any indication that this year is more fragmented than other years.  MODIS does, but the historical record is very short.  Moreover, if you looked at different parts of the MODIS mosaic (e.g. r05c02 in the Beaufort, r05c03 in the northern Chukchi or r05c04 in the northern Laptev), you'd come to very different conclusions about this year relative to 2012.

I *did* present the whole, for exactly that reason.  The question posed by night vid is not one of concentration, but one of ice quality.  You may disagree, but qualitatively, looking at 2013 extent, I find it significantly more fragmented over all, than previous years. The reduction of MYI and commensurate reduction of pack strength is a key factor in this.
This space for Rent.

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #63 on: July 15, 2013, 09:10:58 PM »
I'm with Peter on this.

Not to say that Peter agrees with the following: In my opinion the ratio of Extent/Area is the most objective way to determine the state of the ice with regards dispersion. As I've not blogged for a while I'll do a post on this issue.

Neven

  • Administrator
  • ASIF Governor
  • *****
  • Posts: 3264
    • View Profile
    • Arctic Sea Ice Blog
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #64 on: July 15, 2013, 09:38:44 PM »
Not to imply that this year's ice pack is radically different from previous years, but the CT comparison tool is about the most imprecise one could use for that. I only use it to see where the edges of the ice pack are.

In my opinion the ratio of Extent/Area is the most objective way to determine the state of the ice with regards dispersion.

Indeed. I'm still stunned that CAPIE is so much higher than last year (7.3% for July 13th):



Of course there was more divergence on the edges of the ice pack last year, but still, I would expect the difference to be smaller due to the current dispersed state of the pack's interior.
Il faut cultiver notre jardin

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #65 on: July 15, 2013, 10:21:35 PM »
Quite, Neven.

Except I prefer to do it upside down.  8)
http://dosbat.blogspot.co.uk/2013/07/how-fragmented-is-pack.html

Dispersion index for average of first two weeks of July (IJIS Extent / CT Area).



Also a graph of August average showing strong increase in dispersion (NSIDC Extent/CT Area, monthly averages - what I term August Dispersion Index).


ktonine

  • ASIF Citizen
  • Posts: 324
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #66 on: July 15, 2013, 10:44:17 PM »
While we have to use past data in our predictions, we also have to recognize that this year's melt pattern is something new.  We've never had a year with so much open water north of 80 degrees latitude for so long.  As such, past behavior may not be an indication of future behavior.  I.e., there's probably a lot more uncertainty involved than we think.

Per CT area, a month ago 2013 was 10 or 11 days behind 2012.  Today it is about 6 days behind.    I expect that gap to narrow over the next couple of weeks - that 2013 will inch even closer to 2012. But what really separated 2012 from other years was the amount of melt *after* August 1st.  Is the melt pattern in 2013 more or less conducive to melt in August?

In most years the CAB is a rather static number - it gets eaten away at the edges in the Beaufort, Chuckchi, and Laptev - but incursions and significant open water north of 82 degrees are pretty rare.  Not this year.  But this area is also going to be one of the first to refreeze - so gains there will also be 'lost' before the final minimum numbers are reached.

If we get a storm in August similar to last year's GAC - what will happen?  There is so much lacy, frothy, weak ice that large areas could simply disappear from one satellite image to the next.  I've been tracking some relatively large floes in the r04c04 Modis images and they're just getting beaten up.  Shapes are noticeably changing over the course of 3 or 4 days. 

I would still *not* be surprised if 2013 ended up with a record.  I'm fairly confident that 2013 will end up lower than any year other than 2012.  Forced to choose a number, I'd hazard a guess in the 2.45 +/- .5 range.


ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #67 on: July 15, 2013, 11:12:06 PM »
CT Area difference between 2013 and 2012 for the last four days.

0.669
0.652
0.861
0.815

The apparent closing between 2013 and 2012 was mainly caused by 2013 falling behind the average loss (rising anomalies). Now that local 2012 peak has been reached the gap between 2013 and 2012 is increasing as 2012 falls away once more. This turned out to be a local peak during a period of broadly level anomalies until the early August 2012 storm. In other words, if 2013 doesn't do something in the next two weeks the task of challenging 2012 will become even more difficult because 2012 will be racing ahead faster than 2013 has since the end of the June Cliff.

The best that can be said in favour of 2013 gaining on 2012 is that since 28 June 2013 the anomalies have been on a slight decline (more loss than average) falling about 0.3M km^2 by yesterday's data. However viewed in the context of anomaly variations of other years around the same time, this cannot be taken as evidence of a significant decline.

I really think people should use anomalies more, then they wouldn't get excited about century and double century breaks. These are the norm at this time of year.

jdallen

  • ASIF Upper Class
  • Posts: 2333
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #68 on: July 16, 2013, 12:10:47 AM »
Chris - how do you square that with Espen and Wipneus, whose values are about a third of yours? They are citing IJIS and Bremen, which produce different data sets, but still?

It would seem there is a bit of disagreement over the actual state of things, for certain. What might be the sensible reasons for their differences?
This space for Rent.

ktonine

  • ASIF Citizen
  • Posts: 324
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #69 on: July 16, 2013, 12:40:43 AM »
Even using the same dataset it depends on how each individual is handling leap years.  There is less than six hours difference between the summer solstice in any two consecutive years - but calendar-wise we flip over one day in every four years.  The solstice in 2012 was 6 hours earlier than 2013, but some people end up counting it as a whole day difference because on the calendar it was June 20th in 2012 and in 2013 it fell on June 21st.

For instance I see the most recent CT area anomaly as 2013.5315 equals -1.1226677
In 2012 I find 2012.5315 equals -1.8452662.  The difference is 0.7225985.  The reason my numbers disagree with Chris's numbers is probably due to the way we individually handle leap years - or one of us can't do simple subtraction.  We're both getting up there in age, but I'm hoping neither of us has fallen to that level yet :)

Jim Pettit

  • Global Moderator
  • ASIF Middle Class
  • *****
  • Posts: 854
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #70 on: July 16, 2013, 12:51:17 AM »
CT Area difference between 2013 and 2012 for the last four days.

0.669
0.652
0.861
0.815

The apparent closing between 2013 and 2012 was mainly caused by 2013 falling behind the average loss (rising anomalies). Now that local 2012 peak has been reached the gap between 2013 and 2012 is increasing as 2012 falls away once more. This turned out to be a local peak during a period of broadly level anomalies until the early August 2012 storm. In other words, if 2013 doesn't do something in the next two weeks the task of challenging 2012 will become even more difficult because 2012 will be racing ahead faster than 2013 has since the end of the June Cliff.

The best that can be said in favour of 2013 gaining on 2012 is that since 28 June 2013 the anomalies have been on a slight decline (more loss than average) falling about 0.3M km^2 by yesterday's data. However viewed in the context of anomaly variations of other years around the same time, this cannot be taken as evidence of a significant decline.

I really think people should use anomalies more, then they wouldn't get excited about century and double century breaks. These are the norm at this time of year.

Well, the four days you listed above--Days 192 - 195 inclusive--are the first of a five-day span last July that saw CT area decrease by 813k km2. So it's not that 2013 is "falling away", but that 2012 moved ahead during this brief spike.

The thing is, after the Day 196 (15 July) loss, 2012 went into a 15-day period that could almost be considered a "lull", with a total decrease over the period of just 867k km2, or roughly 58k per day. I noted the other day that if 2013 is to catch 2012, or nearly so, it'll likely happen between now and 30 July. That is, assuming an area loss today of roughly 150k (WAG based on IJIS extent), 2013 will be roughly 815k behind 2012 as of tomorrow. Assuming a healthy daily average over the next two weeks after that--an average, by the way, lower than what we've already seen this July--it's very likely that the 2012/2013 gap will be narrowed considerably.

I really can't wait to see the July PIOMAS numbers, and then the August ones. If minimum 2013 volume is higher than 2012's, I'll kiss Anthony Watts on the lips*.

* - DISCLAIMER- Not really.  ;)

Juan C. García

  • ASIF Citizen
  • Posts: 365
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #71 on: July 16, 2013, 02:17:48 AM »

If minimum 2013 volume is higher than 2012's, I'll kiss Anthony Watts on the lips*.


Auuuggg!!!   :P
Which is the best answer to Sep-2012 ASI lost?
50% [NSIDC extent vs 1979-2000] or
80% [Orig. PIOMAS volume vs 1979, 77.6% with corrections]

Volume is harder to measure than extent, but 3D is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC official trends underestimate the real speed of ASI lost.

jdallen

  • ASIF Upper Class
  • Posts: 2333
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #72 on: July 16, 2013, 02:33:12 AM »
From the other thread, Espen had IJIS at 8,069,063 or 282,657 above 2012.

If 2012  is in a 'lull' of only 60ish KM2 drops a day, that's a week for 2013 to catch up and pass, assuming a modest average of 100KM2/day.

Or am I missing something here?  This is also possible.
This space for Rent.

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #73 on: July 16, 2013, 07:59:22 AM »
Average losses from 15 to 31 July.

Baseline (1980 to 1999 average): -71.2k km^2/day
2011 85.8k km^2/day
2012 68.5k km^2/day

Hence anomalies for 2012 were near level over that period, for 2011 there was a drop in anomalies.

Chris - how do you square that with Espen and Wipneus, whose values are about a third of yours? They are citing IJIS and Bremen, which produce different data sets, but still?

It would seem there is a bit of disagreement over the actual state of things, for certain. What might be the sensible reasons for their differences?


Different data sets are different. IJIS and Bremen are extent aren't they (not aware of them doing area)? If so area and extent are totally different metrics.

Ktonine,

Because PIOMAS neglects leap years, and I needed a way to get all the data to hang on the same time framework, I neglect leap days. Hence the difference. Your numbers may not agree exactly but it doesn't affect the thrust of the argument.

Jim Pettit,

2013 wasn't falling away, it was 'falling behind'.

After the fall of anomalies 2012 anomalies remained flat until the storm hit. If we don't see substantial 2013 losses by then, the battle to reach 2012 becomes harder. 

EDIT CT Area anomalies, baseline 1980 to 1999, detail.

« Last Edit: July 16, 2013, 08:16:22 AM by ChrisReynolds »

werther

  • ASIF Middle Class
  • Posts: 696
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #74 on: July 16, 2013, 09:09:28 AM »
Well, Chris,
In my opinion the numbers remained 'flat', but the structure of the ice was prepared end of July. It was clear, a week before the storm came in, that two million km2 was ready to fade out. The storm wasn't the actor, it was just the pinch-hitter.

Richard Rathbone

  • ASIF Citizen
  • Posts: 317
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #75 on: July 16, 2013, 12:17:27 PM »
Chris,

Wipneus calculates both area and extent. His area is a lot closer to his 2012 area than CT is. His 2013 is running about 300k above 2012 while CT is running about 700k. Its a consistent anomaly between them. Regions Wipneus shows as following 2012 pretty closely show significant lag in CT. There's something different about the way Wipneus gets area to the way CT do and its responding differently under 2013 conditions than it did in 2012 conditions. I have no idea what is causing it though, and nobody had any clue when I asked in Wipneus' thread.

Pmt111500

  • ASIF Middle Class
  • Posts: 933
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #76 on: July 16, 2013, 12:51:11 PM »
154510,1 km2 loss reported today. this would be July 14th number in my count. Or day number 195. Or 0,5315 of the year. Or day number 12277 on the lunisolar calendar set to begin on dec 3rd 1979, the date can also be written as lunar years.month of cycle.date as 33.33.23 , to whomever it may be of interest. Still not voted.
 
A quantity relates to a quantum like camel's back relates to camel's _______ ? (back, vertebra, vertebral tendon, spinal disc, paralysis)

Neven

  • Administrator
  • ASIF Governor
  • *****
  • Posts: 3264
    • View Profile
    • Arctic Sea Ice Blog
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #77 on: July 16, 2013, 01:05:45 PM »
With just a 54K drop for the 14th (the 154K was for the 13th, preceded by a very low 31K for the 12th, back and forth, back and forth), the difference with 2012 is back to a huge 900K. Difference in CAPIE is now almost 9% between 2012 and 2013.

This I find strange, but I have no way of telling what measurement is (more) correct. I'll wait and see how this play out, but the discrepancies surprise me. What's so different about this year?
Il faut cultiver notre jardin

Neven

  • Administrator
  • ASIF Governor
  • *****
  • Posts: 3264
    • View Profile
    • Arctic Sea Ice Blog
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #78 on: July 16, 2013, 01:25:02 PM »
Not to imply that this year's ice pack is radically different from previous years, but the CT comparison tool is about the most imprecise one could use for that. I only use it to see where the edges of the ice pack are.


I just remembered writing about the comparison map three years ago: Inner conflict
Il faut cultiver notre jardin

Wipneus

  • ASIF Governor
  • Posts: 3147
    • View Profile
    • Arctische Pinguin
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #79 on: July 16, 2013, 02:02:00 PM »
I have no way of telling what measurement is (more) correct.


Apart from going there, the next best thing might be studying MODIS.

CAPIE says that on average there is 75% ice concentration within the ice pack (and thus 25% is water). The AMSR2/SSMIS calculation says it is a bit over 90%. Leif Ratinger on the ASI blog found something similar two year ago:
http://neven1.typepad.com/blog/2011/08/capie-hits-record-low.html?cid=6a0133f03a1e37970b014e8ad7886b970d#comment-6a0133f03a1e37970b014e8ad7886b970d

Not nearly MODIS shows 75% average ice concentration.

I think it is time to consider that CT is NOT area, although it still may be a useful indicator.

ktonine

  • ASIF Citizen
  • Posts: 324
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #80 on: July 16, 2013, 02:12:16 PM »
The reason scientists shy away from using area and instead use extent is the high uncertainties that go along with the area calculations.

This is due to the problem with counting open water around all the floes in a constantly changing ice pack when you only get partial images of it at one time.  These marginal ice zones are almost impossible to get good consistent numbers from on a daily basis.

Area counts are 'off' when there is a lot of ponding and when there is an increase in the marginal ice zones.

Perhaps if we set up something like SETI@home we might be able to give them the resources to use an algorithm powerful enough to do the job with more accuracy and less uncertainty.

Juan C. García

  • ASIF Citizen
  • Posts: 365
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #81 on: July 16, 2013, 02:17:36 PM »
I think it is time to consider that CT is NOT area, although it still may be a useful indicator.

Does CT explains the method in which they calculate the area? Is there a CT link that can explain the procedure?
Which is the best answer to Sep-2012 ASI lost?
50% [NSIDC extent vs 1979-2000] or
80% [Orig. PIOMAS volume vs 1979, 77.6% with corrections]

Volume is harder to measure than extent, but 3D is real, 2D's hide ~50% thickness gone.
-> IPCC/NSIDC official trends underestimate the real speed of ASI lost.

Peter Ellis

  • ASIF Middle Class
  • Posts: 501
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #82 on: July 16, 2013, 03:18:00 PM »
I have no idea what is causing it though, and nobody had any clue when I asked in Wipneus' thread.
We know from the buoys and webcams that there's still substantial snow cover and reduced melt ponding in the central arctic, and we know from Modis that there's more fragmentation than last year.  I would deduce therefore that the algorithm used by CT is more sensitive to melt ponds and thus shows 2012 as artificially low.

When considering CAPIE, it's also important to remember that IJIS adjusts their ice-calling algorithm on a seasonal basis in order to minimise the effects of melt ponds, whereas I believe CT uses the same algorithm year-round.  It might be worth moving WAPWE (Wipneus area per Wipneus extent ;-) as a better measure of fragmentation.  Even if still affected by melt ponds, at least you have the same effect in the numerator and denominator.

Neven

  • Administrator
  • ASIF Governor
  • *****
  • Posts: 3264
    • View Profile
    • Arctic Sea Ice Blog
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #83 on: July 16, 2013, 03:24:46 PM »
WAPWE...  ;D
Il faut cultiver notre jardin

Richard Rathbone

  • ASIF Citizen
  • Posts: 317
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #84 on: July 16, 2013, 05:55:09 PM »
I have no idea what is causing it though, and nobody had any clue when I asked in Wipneus' thread.
We know from the buoys and webcams that there's still substantial snow cover and reduced melt ponding in the central arctic, and we know from Modis that there's more fragmentation than last year.  I would deduce therefore that the algorithm used by CT is more sensitive to melt ponds and thus shows 2012 as artificially low.


I don't think its just melt ponds, because their regional area graphs both show the notch features that I associate with melt pond formation and draining.  CT does seem to be a lot more aggressive at moving concentration away from 100%. Its saying about 3 times as much melt so far in the CAB as Wipneus and there are similar sorts of aggressive starts in some other regions.

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #85 on: July 16, 2013, 06:59:31 PM »
Chris,

Wipneus calculates both area and extent. His area is a lot closer to his 2012 area than CT is. His 2013 is running about 300k above 2012 while CT is running about 700k. Its a consistent anomaly between them. Regions Wipneus shows as following 2012 pretty closely show significant lag in CT. There's something different about the way Wipneus gets area to the way CT do and its responding differently under 2013 conditions than it did in 2012 conditions. I have no idea what is causing it though, and nobody had any clue when I asked in Wipneus' thread.

Thanks Richard,

For the present I'm committed to CT Area until I have a metric of daily ice area from 1979 to present that I can examine to decide if I move to it. I'm not asking Wipneus to do that because I know how much work is involved. Having dealt with PIOMAS at only monthly time steps from 1978 to present, I am about to start work on NCEP/NCAR pressure; even over only the summer and with only 2.5deg resolution, using only north of 30degN - there's a lot of data to process and error checking is as big an undertaking as processing itself.

For others, if they prefer to use Wipneus's index I suggest they ignore my comments.

Anway, I often make it clear that CT Area is but one index, albeit my preferred index. That no index is reality itself. And all of my work including predictions is within the confines of that index.

PS It's worth noting that lags/leads and anomalous behaviour have been noted between the various indices for years. That Wipneus has found his index shows similar behaviour is in that respect unsurprising.

Wipneus

  • ASIF Governor
  • Posts: 3147
    • View Profile
    • Arctische Pinguin
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #86 on: July 16, 2013, 07:02:15 PM »
The reason scientists shy away from using area and instead use extent is the high uncertainties that go along with the area calculations.

There are quite a number of ice concentration maps/data made available. Area is only a trivial step away from concentration: the area is bad can only be caused by concentration bad.
 

When considering CAPIE, it's also important to remember that IJIS adjusts their ice-calling algorithm on a seasonal basis in order to minimise the effects of melt ponds, whereas I believe CT uses the same algorithm year-round.  It might be worth moving WAPWE (Wipneus area per Wipneus extent ;-) as a better measure of fragmentation.  Even if still affected by melt ponds, at least you have the same effect in the numerator and denominator.

It is always better because there are more choices that are guaranteed to be the same. I am thinking of the domain: is Baltic sea included, Great Lakes included. Is the false ice in coastal area's, caused by land contamination removed? etc.

Further fragmentation is not the same as reduced concentration. I see the very fragmented ice in places, yet coverage well near 100%.

And a new index is not needed, some scientists  know it as compactness:


Steven

  • ASIF Citizen
  • Posts: 329
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #87 on: July 16, 2013, 07:07:37 PM »
CT does seem to be a lot more aggressive at moving concentration away from 100%. Its saying about 3 times as much melt so far in the CAB as Wipneus and there are similar sorts of aggressive starts in some other regions.

CT appears to say more about the albedo than about the actual ice area, especially in June/July.  But that may still be useful information.  The regions with a high concentration in the CT maps may have a better chance to survive the melting season.  The daily values can fluctuate but the accumulated information from May/June/July should provide relevant information.

According to the maps from last year, the whole region between Beaufort and East Siberian started to disintegrate significantly around this time of the year, two weeks before the August storm:
http://igloo.atmos.uiuc.edu/cgi-bin/test/print.sh?fm=07&fd=20&fy=2012&sm=07&sd=31&sy=2012

Bremen shows more detailed images but the general pattern is more or less the same.  So even if CT was too aggressive in its estimates, it still was a good indicator of what was coming.

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #88 on: July 16, 2013, 07:29:23 PM »
Wipneus,

How is compactness calculated.

Wipneus

  • ASIF Governor
  • Posts: 3147
    • View Profile
    • Arctische Pinguin
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #89 on: July 16, 2013, 07:40:32 PM »
Wipneus,

How is compactness calculated.

Area/Extent

(It was Axel Schweiger who pointed this out on the blog)

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #90 on: July 16, 2013, 09:40:06 PM »
Thanks Wipneus,

In the first post of your thread on your calculations you posted an image showing NSIDC/IJIS/ and your work matching up, with offsets. So the summer peaking of IJIS/CT Area (inverse shown in your graph above) must be due to CT Area, not some peculiarity in Extent (the peaking is implied by monthly average NSIDC extent).

Yet again, the peaking seems to be telling us something about sea ice state, as August NSIDC Extent/CT Area shows an upward trend with ice recession, and a jump at 2007. And that's despite the alternate implication that CT Area is a poor metric.

What puzzles me is that people are telling me your data shows we're in for a crash this year, that is not supported by CT Area. Whereas in 2012 and 2011 CT Area was indicating a new record (2011 late summer tracked 2007 very closely even including a parallel anomaly climb before the minimum - CT Area). I've noted your comparison of your method and CT Area, it is interesting that CT Area falls away from your calculation from May. This is when the Extent/CT Area curves diverge.

I know you've processed at least some of 2012, have you processed into August? I suspect from what I read previously you've not got enough years to have a baseline to form anomalies. That may make the following question hard to answer: CT Area 2012 showed level anomalies (climatological losses) until the August cyclone, when anomalies started to fall rapidly. Turning a new record into a dramatic record. Can you tell if this was the case in 2012?

Sorry if you've covered any of this elsewhere, I've been busy recently.

James Lovejoy

  • ASIF Citizen
  • Posts: 144
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #91 on: July 16, 2013, 10:45:50 PM »
I'm going with 2.5 to 2.75.

We're way behind 2012, and I don't think we will catch up barring perfect melt weather.  OTOH I think we will close the gap enough to make it to second place barring perfect weather for the ice.

I'm looking with interest to see which happens.

Wipneus

  • ASIF Governor
  • Posts: 3147
    • View Profile
    • Arctische Pinguin
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #92 on: July 17, 2013, 12:53:10 PM »
Thanks Wipneus,

In the first post of your thread on your calculations you posted an image showing NSIDC/IJIS/ and your work matching up, with offsets. So the summer peaking of IJIS/CT Area (inverse shown in your graph above) must be due to CT Area, not some peculiarity in Extent (the peaking is implied by monthly average NSIDC extent).

Yes, not that I wouldn't like to know all those detailed differences, extent seems rather well defined.


Yet again, the peaking seems to be telling us something about sea ice state, as August NSIDC Extent/CT Area shows an upward trend with ice recession, and a jump at 2007. And that's despite the alternate implication that CT Area is a poor metric.

Let me state again, CTArea may well be a useful index, but still be a bad metric for area. Except that my compactness metric looks a whole lot more believable than CAPIE, I am not claiming anything about the usefulness or correctness of the AMSR2-area series. 

So if in this thread the question is raised "why is CTArea so high?", the correct answer does might not be that the area is higher.

Knowing the right reason is in my opinion a crucial part of using the CTArea index.

What puzzles me is that people are telling me your data shows we're in for a crash this year, that is not supported by CT Area.

I don't know what you call a crash. But what I agree with some people is that 2013 may still end up very close or below 2012. If you care to look at my regional charts it is clear that two things are required. First Beaufort needs to melt the coming 1.5 month at a rate comparable to last year one month earlier. Second the CAB will need to repeat exactly what it did last year, loosing 1M5 km2 in about one month, starting at the end of July. That is a crash, but the same as last year.

Whereas in 2012 and 2011 CT Area was indicating a new record (2011 late summer tracked 2007 very closely even including a parallel anomaly climb before the minimum - CT Area). I've noted your comparison of your method and CT Area, it is interesting that CT Area falls away from your calculation from May. This is when the Extent/CT Area curves diverge.

The divergence in May, begin of June is explained by CTArea not using the true grid cell area. It overestimates cells in low latitudes, underestimates high latitudes, is true at 70 degrees. The geography currently explains about 260k of the difference between the CTarea and AMSR2-area.

In the rest of June, the divergence grows somewhat bigger, to about 650k and that has stayed so. Comparing with 2012, the qualitative behavior is the same except that the "June bulge" is a lot bigger but also that much of that has disappeared by the beginning of September.

I know you've processed at least some of 2012, have you processed into August? I suspect from what I read previously you've not got enough years to have a baseline to form anomalies. That may make the following question hard to answer: CT Area 2012 showed level anomalies (climatological losses) until the August cyclone, when anomalies started to fall rapidly. Turning a new record into a dramatic record. Can you tell if this was the case in 2012?

Sorry if you've covered any of this elsewhere, I've been busy recently.

I have AMSR2 data only for 2013. The Uni-Hamburg people seem to have too much high priority work, so I don't ask for the 2012 data.
So I turned my attention to a SSMIS (or SSM/IS) timeline from the very same people, same algorithm. The extent/area results are very close to those of AMSR2 where the times overlap. So I use the SSMIS data for comparing with 2012.

Before 2012, there is only the SSMI (without one S) available (for 1991-2012). Unfortunately calculating area here gives results (for 2012) about 6% lower than SSMIS/AMSR2. So I would need to correct  for that, which is the current state of research.

I have adjusted the scales of the area and extent graphs to include the September minimum. They are included and updated in the top post of "home brew" discussion.

Wipneus

  • ASIF Governor
  • Posts: 3147
    • View Profile
    • Arctische Pinguin
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #93 on: July 17, 2013, 02:43:06 PM »
Maybe this helps: CT regional graphs with 2012 included (brown lines)
Does not work for Kara, which they changed the x axis.

(log in to see attachment)

ChrisReynolds

  • ASIF Upper Class
  • Posts: 1711
    • View Profile
    • Dosbat
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #94 on: July 17, 2013, 06:59:46 PM »
Thanks for the detailed response Wipneus.

Shared Humanity

  • ASIF Upper Class
  • Posts: 1754
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #95 on: July 17, 2013, 07:18:02 PM »
From these charts, it sure looks like the Beaufort could run out of time before it matches the 2012 results.

BornFromTheVoid

  • ASIF Middle Class
  • Posts: 929
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #96 on: July 18, 2013, 01:21:01 PM »
A drop of 121.9k puts 451k off the 07, 11 and 12 average.

It also leaves us like this compared to previous minima


Neven

  • Administrator
  • ASIF Governor
  • *****
  • Posts: 3264
    • View Profile
    • Arctic Sea Ice Blog
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #97 on: July 19, 2013, 01:02:54 PM »
Et voilà, a 272K drop reported for the 17th (I was wondering when CT SIA was going to start to catch up). 2013 will probably be in 4th spot as of tomorrow.
Il faut cultiver notre jardin

Jim Pettit

  • Global Moderator
  • ASIF Middle Class
  • *****
  • Posts: 854
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #98 on: July 19, 2013, 01:45:51 PM »
Et voilà, a 272K drop reported for the 17th (I was wondering when CT SIA was going to start to catch up). 2013 will probably be in 4th spot as of tomorrow.

I saw that. The largest one-day drop of the year, and in fact larger than any that occurred last year. It's also the 13th century or greater one-day decrease this month, which is more than the 11 that happened all of last July.

From 19 July through 30 July, 2012 SIA dropped by a modest 626k. 2013 would need to average 106k a day over the next 12 days to catch 2013. The average daily drop so far this month has been 122k, meaning that the daily average can slow from its current pace by 15% and still catch.

We'll see...

(FWIW, SIE had a huge break as well, nearly 200k. If that's not revised substantially, that may mean another relatively large drop is in store for SIA in the next day or so.)

wili

  • ASIF Upper Class
  • Posts: 1797
    • View Profile
Re: Cryosphere Today 2013 Arctic SIA daily minimum: July poll
« Reply #99 on: July 19, 2013, 02:38:50 PM »
I had predicted that SIA would break out of the main pack and approach record territory about now. I still think it will break below the other years with a couple weeks. There is just so much ice that is so thin that it will just be vanishing in a 'flash melt' in the next few days and weeks.

"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."