« Passing the Passages | Main | Sea ice extent update 20: spread it! »


Feed You can follow this conversation by subscribing to the comment feed for this post.


Looks like this report was revised in early July rather than now !

Lord Soth

After reading all the reports, I concluded that the experts can't decide on the minimun.

It will be interesting to see what happens to the 5+ year ice which is now being exposed to the warming water of the beaufort sea.


I like Zhang's style (Jinlun Zhang, Applied Physics Laboratory, University of Washington) - two contradictory predictions.

He predicts 4.8 million sq km:

But he also helps his colleague Ron Lindsay with a prediction of 3.96 ± 0.34 million sq km:

Jinlun - he wants to win that case of beer and bragging rights:

Next year, he'll coauthor a few more predictions... hey, the Rules allow it :-)


As we speak both predictions by Jinlun Zhang seem to belong to "La la " land. At today's rate we may not reach the 7 million at the end of July. This means that even reaching 4.8 million in Mid september would mean an average daily loss above 70 k unyil the end of August. Not impossible I admit, but pretty unlikely unless the weather conditions change dramatically. In all previous year the rate of SIE reduction has been lower in August than in July

Artful Dodger

HEY! Linear Extrapolation Fans:

At July's rate of loss, 2010 Sea Ice Area will undercut the SIA on the same day in 2007 by August 15 (+/- 5 days).

Jim Dowling

We are now behind 2009 in ice-melt.
2009 had 7,101,719, we are at 7,122,813.
Weather outlook for next 10 days is not good for melt either:
Low pressure to dominate the arctic basin.

On a related note, there's not much activity at intrade.com for betting on whether the ice minimum for 2010 will be higher than 2009.

Artful Dodger

Jim: Sea Ice Extent is not "ice-melt". The fact that 2010 has significantly smaller Sea Ice Area means the the Ice Pack is further spread out than 2009. And since most experts agree that the ice is also thinner, the sea ice is currently at much greater risk of precipitous melt than in 2009.



Your predictions so far have turned to be correct.


Compared to the June prediction (http://neven1.typepad.com/blog/2010/06/search-september-sea-ice-outlook-june-report.html) we see:
Average (all): June 4.8 +/- 1.1, July 4.6 +/- 1.1
Average (all)*: June 5.0 +/- 0.4, July 4.8 +/- 0.6
Average (Heuristic): June 5.2 +/- 0.2, July 5.0 +/- 0.9
Average (Statistical): June 5.1 +/- 0.5, July 4.8 +/- 0.7
Average (Modelling): June 5.1 +/- 0.5, July 5.3 +/- 0.4
Average (Heur/Stat): June 3.9 +/- 2.0, July 3.7 +/- 1.6
Average (Heur/Stat)*: June 4.9 +/- 0.7, July 4.4 +/- 0.4
*less Wilson

Note - I've assumed that WIlson and Wellman (who are now listed as "public") used the same techniques as June.

Of course some people have not made new predictions, but new teams have come into play. But we lost a 5.5 (Stroeve et al - June) and gained a 4.74 (Maier et al - July). So comparing like-for-like, for the 13 teams that appear on both months we see:
Average (all): June 4.7 +/- 1.3, July 4.6 +/- 1.3
Average (all)*: June 5.0 +/- 0.5, July 4.9 +/- 0.6
Average (Heuristic): June 5.4 +/- 0.1, July 5.5 +/- 0.1
Average (Statistical): June 5.0 +/- 0.6, July 4.8 +/- 0.8
Average (Modelling): June 5.1 +/- 0.5, July 5.2 +/- 0.5
Average (Heur/Stat): June 3.9 +/- 2.0, July 3.7 +/- 1.8
Average (Heur/Stat)*: June 4.9 +/- 0.7, July 4.6 +/- 0.3
*less Wilson

So...the ensemble average has dropped, but heuristic and modelling techniques both conclude a higher minimum is likely. The drop in the average is off the back of statistical and heuristic/statistical techniques - for these, there are several big drops, while Tivy and Wilson at the extremes remain the same. Heuristic and modelling already favoured a high minimum in June while statistical tended towards the lower end. The July numbers increase this tendency - purple and red to the right, light blue and orange to the left.

It will be interesting to see who gets closer at then end - models and educated guesses, or forcing the numbers through hoops. I've messed around with stats enough to we aware of their predictive limitations if the rules of the game change. This year's conditions are so odd that I wouldn't be backing the statisticians, myself.

Artful Dodger

Phil: I'm curious, as a PhD and a lecturer on Tax Policy, are you or your Department currently involved in the debate over a Carbon Tax in Australia?

Jim Dowling

@Artful Dodger: I think most people who read this blog, myself included, know the difference between sea ice extent and volume.

Artful Dodger

Jim: Fair enough. I just wouldn't want an uncritical audience to read your comments and incorrectly presume a strong correlation between fluctuating daily extend measures and the underlying variable (the change in total sea ice). This is the reason NSIDC uses a 5-day moving average in it's daily graphs, and only reports monthly.

The current generation of satellites is NOT measuring "melt", although we humans like to infer it. Right now, the sea ice is spreading out north of Siberia. This can also imply it is melting. This situation messes with JAXA summary numbers because it becomes (loss of extent due to melt) + (increase in extent due to spread) = daily change in extent

Hopefully Neven will soon have an animation of the Northern Sea Route (NSR) maps. The area between Wrangel Island and New Siberian Islands is undergoing repeated cycles of Spread-Melt, Spread-Melt.

Oh, and thanks for your continued vigilance on the weather reports, Jim. I follow them continuously.


Good info. I am doing research on environmental tax policy (among other things) but unfortunately had no part in the public debate about the now doomed Emission Trading Scheme. Any way, the government seems to take little notice of what tax policy researchers have to say when it comes to environmental policy. Politics and powerful lobbies get in the way. In the meantime, the current opposition leader and wannabe PM thinks Climate Change is " crap".God forbids if he gets in, we might end up with Lord Monkton as climate policy adviser ( He visisted Australia las summer and got an interview with Abbott (oposition leader)).
I am very interested in climate issues and I am learning a lot from this site and the various links . My training as a researcher has taught to be rigourous and "critical" and this is why I tend to question things . But overall I am on the same wavelength as you guys.

Artful Dodger

Phil: Good on ya, Mate! Welcome to the forum.


This should interest you Neven.


"British north-west passage ship found in Canada

HMS Investigator was abandoned in 1853 trying to find earlier mission searching for north-west passage across North America."


Thanks a lot, marktime! I will mention this in today's SIE update.


I agree with Jim Dowling that stable high pressure systems over the Arctic are nowhere in sight for the coming week or so (ECMWF and Unisys seem to be in agreement on that). On and off they're there, but just not stable. Perhaps when the AO turns negative.


Hey Phil,

If Gillard gets in, maybe you can jag a seat at her "citizens assemby" do over and spend a year rehashing everything we already know we need to do....



Frank D,

I know this idea of a citizens assembly is ridiculous. It really says a lot about the government's commitment to do anything about CC. I am just hoping that the Greens get the balance of power in the Senate. That would be better than Senator Fielding surely!


Frank D

" Of course some people have not made new predictions, but new teams have come into play. But we lost a 5.5 (Stroeve et al - June) and gained a 4.74 (Maier et al - July). So comparing like-for-like, for the 13 teams that appear on both months we see:
Average (all): June 4.7 +/- 1.3, July 4.6 +/- 1.3
Average (all)*: June 5.0 +/- 0.5, July 4.9 +/- 0.6
Average (Heuristic): June 5.4 +/- 0.1, July 5.5 +/- 0.1
Average (Statistical): June 5.0 +/- 0.6, July 4.8 +/- 0.8
Average (Modelling): June 5.1 +/- 0.5, July 5.2 +/- 0.5
Average (Heur/Stat): June 3.9 +/- 2.0, July 3.7 +/- 1.8
Average (Heur/Stat)*: June 4.9 +/- 0.7, July 4.6 +/- 0.3"

Looking at the numbers fron JAXA for the last few days, I'd say the "Heuristics" have probably got it right: 5.5 million +/- 0.2



So your heuistic method lines up with the ensemble's heuristic method? I think the saying goes "its always nice to have your opinion confirmed by experts..."

I'm a noob, so my educated guesses are entirely free of education. I'm better at playing with numbers, and despite my stated reservations about the statistical predictions, they line up with mine - 4.8 +/- 0.5. (calculated in early June, before I decided that for someone like me, trying to figure a number was a complete snipe hunt)

Maybe it'll be 5.3 and we'll both be right!

Only time will tell....

Charles Wilson

Jim Dowling - - your MOST WELCOME site: http://weather.unisys.com/gfsx/9panel/gfsx_500p_9panel_nhem.html
...has HIGH PRESSURE = SUN RETURN on the 5th, ending the 5 week Arctic Low.
The Colors go from Blue to Green - - you may have thought that was only intermediate as Highs are Red further South -- but the Colors are for Sea surface Height, which is affected by BOTH Temp & Pressure. It won't be Red that far North. Pressure is denoted by Contour lines & those look Very strong - - like a bunch of century melts
- - maybe even Double century melts by the 10th.
ECMMF/N.Hemi: same: blue vanishes & so do pressures under 1000. http://www.wetterzentrale.de/topkarten/fsecmeur.html Thanx Neven.

Ice Extent: http://www.ijis.iarc.uaf.edu/seaice/extent/plot.csv
Comparing _______2007___ to___ 2010_____&____2009__
Ahead June 28______ no________ 679,531 Sq.Km___no__
Ahead July 28____594,844_________no__________21,094
Ahead July 30____580,625_________no___________ 3,281
Ahead July 31____534,532_________no____________no__
(2009 now 31,438 behind)
July28-29 ______ - 48,594 _____- 57,912
July29-30 ______ - 51,250 _____- 77,969
July30-31 ______ - 52,812 _____- 86,719 +534,556 31,43892,812
Aug1-to-2 ______ - 51,250 _____ ? ?
Note Large revision down of 6,031, inverse of the 29th, implies ~ 60K tomorrow ?
Next 11 days: 2009 had only 1 day over 58K ... but 2007 had 4 over 100K (Aug 2-5)
... Pips 2.0 Forecasts Ice Motion, now Fast, to Slow DOWN = less Ice Loss ... perhaps to reverse direction as the High moves in.

PS The Sea Ice Outlook only posted part of my submission -- possibly in error (did they just post the first Page ?) What was missed was a lot on Wayne Davidson - - all the Outlooks were submitted July 14 ...I figured his Theories meant a return to Clear Skies = SUN, but the La Nina CTI or Cold Tongue Index implied a 6 week & the ONI index an 8 week lag realative to 2007
= Early or Late August = SUNNY.
... So the 9-10 day FORECASTS imply CTI was the Good indicator. Unless Aug 4-10 = High Pressure is a "false Positive". The delay Downgraded the chance of 300-mph winds but I expressed INCREASED Certainty of a Near-total Melt, as now the Clouds/SUN are not a "Wild Card" (albeit delayed). If it is the 8-week, there is a Heavy "bar" of thick Ice that may remain across the West of the basin. But I felt SURE the Sun will resume.

After the Submission deadline, Piomas updated their -10,700 June 18 Anomaly, to July 17th's -10,600 -- hardly any change, albeit there was a dip in between. Remembering PIOMAS' September Mean is 13,400... so IF the melt rate is only at 1979-2009 AVERAGE from now on, it will end up 2700 km3 come September = still over half of 2007's Volume, but plenty of SUN will start the Anomaly dropping again.

The comments to this entry are closed.