The Modern Area of Ice

The area of Arctic sea ice passed a numerical milestone this week: into the 21^st Century. In the first week of August, with a month or more to go in the melt season, area has dropped below not just the year-to-date values of previous years, but the annual low points of any satellite-era year before 2007 (Cryosphere Today data). The current area is slightly below even the previous record holder, 2007, for this date, and is heading for a minimum down with the “modern” group of post-2006 values. Figure 1visualizes the comparison between area on August 4 and the lowest days of previous years — which ranged from August 31 to September 30. The median date for minimum area is September 8 or 9.

  

Figure 1

There is no trend toward earlier or later minimum dates, so we can’t yet predict when that will fall this year. While the melt season continues, I will be updating this graphic regularly for our Daily Graphs page.

+++++

UPDATE SEPTEMBER 2011

In the weeks after the post above was written (August 4 2011), Arctic sea ice area continued to decline. By the second week of September, it edged past the previous record low set in 2007. While Figure 1 above remains in its original form, I will be updating Figure 2 below to reflect the final minimum record for 2011.

 

Figure 2

Comments

[Chris Biscan]

CT dropped 113K today.

now under 4.0m km2.

The area is still lower than 2007.

[Seke Rob]

Chris, you've posted 113k twice, but from 1st to 2nd, the CT Arctic change was 128kkmsquare.

Et al, Me own little plots show means since Jan 1 till present with 2011 persisting on 77-78kkmsq less area than 2007. That's the average of the first 213 days of the year per CT record.

Kind of expecting that JAXA for it's shorter record since 2003, will see 2011 ytd average fall below 2006. That was sort of preprogrammed at the start of July. The JAXA YTD Average through Aug.3

Calendar Daily Avg.
Year From 1.1.yr RANK
'11 11718905 (2)
'10 11994575 (5)
'09 12214459 (7)
'08 12217999 (8)
'07 11820022 (3)
'06 11716853 (1)
'05 11934784 (4)
'04 12205809 (6)
'03 12506853 (9)
'02 N/A N/A

Do these ytd average mean anything... well, they tell that on the whole, whether a year has been either good, bad, or plain ugly... the ''not even wrong'' coming up.

JAXA being now ~415kkmsq for 2011 behind on 2007, but on the slush puppy theme, not going to be surprised even to see a daily decline record. Plz cover your keyboards prior to grabbing your coffee, would this actually happen.

[Philiponfire]

Seke Rob I think your speculation is right on the ball. we could easily see some mind boggling extent changes in the coming weeks if for example a steady wind blowing off of Siberia were to persist for several days. the ice that side really does look like slush. no discernible floes and holes doesn't begin to describe it. more like lace.

I have my popcorn ready.

[Neven]

I dare say this is currently the best graph on the Daily Graphs page. Great stuff, Larry( love the title)!

[Chris Biscan]

2011.5808 -1.7069381 4.1861873 5.8931255
2011.5836 -1.7150214 4.1205010 5.8355227
2011.5863 -1.7828821 3.9926519 5.7755342

Maybe you need to clear your Cache Neven

[Chris Biscan]

you said 1st to 2nd.

this is 2nd to 3rd

I think thats where we got mixed up and I was also off on the 113K number.

[Neven]

Chris, CT area runs one day behind IJIS extent, so the drop from -128K (to a total of 3.992652) was for August 2nd. This was recently re-discussed.

[logicman]

I'm expecting to see the extent take a nosedive.
The IJIS AMSR-E graphs show that area continues to decline fairly rapidly while extent declines only at a very slow rate. This must mean that as the ice is melting, it is spreading out.
If either the spread out ice is flushed into warmer waters or it is pressed back hard into the main pack, we will see a very rapid drop in extent.
Keep watching those extent graphs!

[L. Hamilton]

CT area dropped by 67k on 8/3, so it now stands at 3.93.

That is 111k below 2007 on this date. I've updated the graph above, and will do that one more time tomorrow, so the post stabilizes with a graph for 8/4 -- consistent with the date of the post itself.

[L. Hamilton]

More arithmetic:

Min 2007 area was 2.92. Min 2011 area would pass that if delta averaged -36k between now and September 1.

[Seke Rob]

With todays drop, the daily average since Jan 1 for all of the years since JAXA started, has now slid to lowest spot.

2011 has now to stay above 2007 by 475kkmsq for the next 45 days, to not become the poorest year at whenever the minimum occurs... domino day for the coming 6 weeks.

NB: Don't know how to let full size image pop up.

[crandles]

>"I dare say this is currently the best graph on the Daily Graphs page."

Well maybe I would think it was if the 2011 bar had different colour stacked bars with top colour starting at current data less minimum drop from last data day to minimum; next colour starts at current data less average drop from last data day to minimum; next colour starts current data less maximum drop from last data day to minimum.

Perhaps I am asking for too much.

Anyway it is a very nice graph as it is.

[Artful Dodger]

CAPIE Extrema:

CAPIE (Cryosphere Today Area Per IJIS Extent) is relevant to predicting Arctic sea ice melt because it quantifies directly the albedo flip effect on the energy budget in the interior of the pack ice.

In the chart above you can see 2011 is closing in on the all-time record low CAPIE value which set on Aug 13, 2008.

[George Phillies]

Someone has probably caught this already, but to judge from the Bremen maps the Northwest passage appears to have opened in the last day or so. Not the big passage, but a passage.

I believe that is ten days earlier than last year, if I remember correctly a post here. But I could be wrong there, too.

[AmbiValent]

Someone here said it, but I think the passage was officially declared to be open. Nor do I think the Northern Sea Route was declared to be open, despite looking that way to eyeballing on a false color map for about two weeks now. (Other maps still show ice remnants there, so it might not be safe for non-icebreakers yet)

[L. Hamilton]

As of Aug 7, CT area is about 138k below 2007 on this date, and 270k above the 2009 annual minimum (updated on graphs page).

[Artful Dodger]

Hi Larry,

In 2010, between Aug 12 and Sep 8 (the date of min SIA), the Arctic lost an additional 972K km^2.

If this rate of loss repeats in 2011, there will be just 2.372 M km^2 of sea ice area. This would smash the previous record of 2.919 M set on Sep 07 2007.

This would also be close to 2 Std Dev below the Gompertz trend line. If so, we may need to compare its goodness-of-fit to the exponential trend line.

Cheers,
Lodger

[L. Hamilton]

You're right, with current SIA at 3.3, my Gompertz prediction of 3.1 for a September mean now looks pretty high. That referred to NSIDC monthly means, but the prediction is only slightly lower (3.0) if the same model were estimated from CT data.

When I did those analyses, I compared Gompertz results with those from quadratic regression, which up to then was a more popular approach. Gompertz curves based on data through Sept 2010 predicted lower values for 2011 than the quadratic curves did. For example,

NSIDC Sept 2011 mean area prediction:
Gompertz -- 3.1 (in my April post)
quadratic -- 3.2

CT Sept 2011 mean area prediction:
Gompertz -- 3.0
quadratic -- 3.2

A similar pattern occurred with extent and volume: the Gompertz predictions were more aggressive than quadratic. But now, it may turn out with respect to area at least, they were not aggressive enough! We'll see.

[L. Hamilton]

Postscript added to the original post, with an updated graphic reflecting the new record minimum area set in September 2011.