This is a big one: The weekly-average summer sea ice minimum bet! The rules for this one are:
- We use Jaxa Values posted here.
- On October 15, 2010, I’ll compute the 7 day average over 7 adjacent all adjacent days from January 2010 to October 15, 2010 to compute the running 7 day average of the most recent 7 reported values. (In case of drop outs, I still compute over 7 entries. This is the only way I know to deal with the possibility that Jaxa might fail to report for 7 days in a row.) Whoever gets closest to this value wins the quatloos. I anticipate this will be the minimum weekly average and it will probably be the minimum for the year.
- Whoever gets closest to that 7 day average wins these quatloos.
Here is a late May image from JAXA:
[sockulator(../musings/wp-content/uploads/2009/09/UAHBets2.php?Metric=NH Ice?Units=x 106 km2?cutOffMonth=7?cutOffDay=1?cutOffYear=2010?DateMetric=NH Ice Weekly Ave Minimum? )sockulator]
Results older than 2 months will not be displayed to avoid taking my site down.
The cutoff day is recorded as 6/30/2010, which I think means the script close the first second of 6/30/2010 using a clock based in California. I’m going to schedule a “mid range” forecast bet to appear in July 1, and a “long range” forecast to appear in August 1.
For newbies: If you make a mistake entering your bet, just re-enter. For Europeans: The script speaks American. It thinks 10.1 = “ten + 1/10th”; don’t write 10,1.
Discussion rules for Sea Ice Threads: Please keep your discussions focused on observations of, predictions of and bets on sea ice. If you want to debate what you consider to be unfair comment policies at other blogs, do so at the other blog.
So is this just the average for Oct 8-14, or an average of averages?
Mike– Thanks. I clarified.
I will load up all the data from Jan 2010- Oct 15, 2010. I’ll throw away all blanks, and then compute the running 7 entry average. So, this should catch the minimum in Septamber, but if for some mysterious reason, 7 Pinatubos erupt in June, and cloaking the earth in darkness and bringing on a cold September, it should catch the minimum whenever it occurs.
Why not just use the NSIDC published number for the sea ice minimun. They usually announce their result in the first week of October. They are generally considered the authority on the sea ice minimun, (well at least with the media) and they use a 5 day average for the 15% extent.
Soth–
Because I don’t like the way NSIDC’s site is organized for betting. The media can use and report what they like.
We just need a specific metric, and I’m going to use Jaxa.
Information for players:
Seven day moving average minimums for 2002-2009
5.659085
6.064464
5.828840
5.348951
5.823215
4.279241
4.733126
5.278840
I will be waiting till the end of the month to make my prediction.
The big question, is are we going to get a repeat of the 2007 weather pattern, that caused the steep decline starting in late June thru August. Only time will tell.
In 2008, the ice just kept melting fast in August, and almost caught up to the 2007 mimimun, but just could not close the gap before the cool weather set in.
2009 kept up the pace with 2008 until mid August, and then the curve just flatten out, as wind conditions led to spreading the ice out, rather than compacting it as in 2007.
Regardless, 2009 ended up having the smallest ice volume.
If we get the June and July conditions of 2007 combined with the August conditions of 2008, we could see numbers under 4 milion square kms. Regardless of the outcome, I still see the ice volume heading down.
Soth–
I don’t see any reason why the long term downward trend should end either.
Lucia: “I don’t see any reason why the long term downward trend should end either.”
How about these reasons:
(1) One does not believe CO2 is the dominant forcing.
(2) One recognises the cyclicality of Arctic sea ice, from references such as
http://alexeylyubushin.narod.ru/Climate_Changes_and_Fish_Productivity.pdf?
(3) One agrees with Pielke Sr “Of course, as I and others, including Kevin Trenberth, have repeatedly urged…we need to move to the use of the ocean heat content change as the metric to asses global warming and cooling.” And, there has been no warming of ocean surface temperatures for a decade.
http://pielkeclimatesci.wordpress.com/2010/05/25/lack-of-a-trend-in-the-ocean-surface-temperature-since-2000-its-significance/
(4) One therefore surmises that a change in the Arctic sea ice trend may be at hand.
One reason why the long term trend may slow is the loss of multi-year ice. In the early 00’s each year had less multi-year ice, and ‘other’ factors causing an increased melt each year. Now that we have very little multi-year ice left, the reduction in sea ice is driven by ‘other’ factors but not reduction in multi-year. So slower downward trend in Sea ice if ‘other’ factors trend in the same direction. The melt seasons of 08 and 09 give me enough confidence that seasonal ice is thick enough to not al melt at once just because the multi-year ice is gone or ‘rotten’.
And of course why is ice melt so far ahead of IPCC models? Are the models underestimating the impact of Co2 on ice, and the sea ice will continue to melt fast?
Or are the models missing/underestimating the impact of something else? Black Carbon perhaps. [Turn on speculative mode] Or the cooling PDO. What little I’ve seen in the scientific literature suggests that a cooling PDO is likely to result in an increase in transport of ice from the Pacific to the Atlantic side, where it is pushed into the warmer Atlantic and melted, and contributing to a reduction in overall ice extent. The PDO has been cooling since the early 80s, and this cooling accellerated in the 00s. It is expected to continue cooling for another 10 or 20 years maybe, and then it will probably warm again. So in 10 or 20 years time the downward trend in sea ice extent may slow down or reverse.
But in cooler conditions, as in summers of several decades ago, or in winters of today, the Arctic is cold enough that blowing the ice into the Atlantic has little effect and any loss is quickly replaced by refreezing of an ocean exposed to very cold air. And in the North Pacific where a lot of the ice edge is, a cool PDO makes it colder and so there is more ice. Hence the recent trend of more ice in winter, but strong melt in summer, and why the cool PDO prior to 1980 didn’t cause any reduction of sea ice (as far as was measurable).
Michael Hauber (Comment#44622) June 2nd, 2010 at 8:42 pm
And in the North Pacific where a lot of the ice edge is, a cool PDO makes it colder and so there is more ice. Hence the recent trend of more ice in winter, but strong melt in summer, and why the cool PDO prior to 1980 didn’t cause any reduction of sea ice (as far as was measurable).
You have this backwards, a ‘cool’ PDO results in a warm N Pacific.
http://jisao.washington.edu/pdo/
Michael
>>So slower downward trend in Sea…
Slowing is different from ending. I could, hypothetically, envisions reasons why the long term decline could slow down or speed up. Should we ever hit zero ice, the decline will obviously end. But right now, I don’t see any particular reason to see the long term decline end. I anticipate that for the near future (i.e. 20 or so years) we will see some “up” years and some “down” years, but overall, we will see a tendency for the amount of ice to decline. How fast will it decline? I have no idea!
‘You have this backwards, a ‘cool’ PDO results in a warm N Pacific.’
Hmm the typical PDO pattern shows warm in NW and cool in NE, but NW is where all the ice is, and much more warm than cool. Looking at trends on GISS for SST over the last 30 years, the arc of cooling water to the east and north east that seems to be associated with a cooling PDO extends a bit further west into the Bering Strait region than would be expected from the JISAO map. Perhaps something other than PDO is causing cooling of waters in that region.
http://data.giss.nasa.gov/gistemp/maps/
Select ocean, and trend instead of anomaly and period 1980 to 2009, on an annual basis.
A pedantic point: it’s hard to say the PDO causes anything, since there’s not much of a mechanism attached to it yet. The PDO index is just a numerical abstraction, and somebody decided to give values > 0 and < 0 the names warm and cool. At this point, I think the index value is more an effect than a cause. It doesn't even really deserve the O, since it doesn't have much regularity to it. Sometimes it's warmer than normal in the Alaskan fisheries, sometimes it's cooler. Beyond that, I'm not sure what the PDO really is – what sorts of circulations act to give those effects.
Based on the 2003-2009 average loss from this date, the projected minimum is 4.86 Mm2 with a three standard deviation range from 6.68 to 3.04 Mm2.
The AMO index at 0.505 is approaching the all time high of 0.562 in 1998. However, the UAH NoPol temperature anomaly is much higher now than in 1998. Three of the last four months have been above the 2007 high of 2.26 degrees. The high in 1998 was 1.16 degrees. I may have to lower my bet.
Lucia wrote:
” I’m going to schedule a “mid range†forecast bet to appear in July 1, and a “long range†forecast to appear in August 1.”
Shouldn’t that be “a short range forecast”?
I bet 3 quatloos on 5.1^6 km^2, because this would be annoying to both sides. It would fail to continue the recovery (being a bit less icy than 2009), but it would also fail to meaningfully continue the decline. It would go mostly sideways.
Incidentally, what are “quatloos”? (I’m new here.)
Oops, I meant to bet 5.1 *10^6 km^2, not 5.1^6 km^2.
Re: Roger Knights (Jun 15 21:28),
See this thread for information about quatloos.
5.1 Mm2 is probably going to be too high by a fair amount unless things get a lot better real soon now. Based on the past loss from this date the likely minimum is 4.7 Mm2 and I think that may be too high as well. The loss rate normally accelerates at this time of year and we’re already way above the recent average. The anomaly plot is about 2 weeks ahead of 2007 right now. A La Nina might help, but I don’t know if it’s going to happen soon enough to make a difference in the minimum extent.
DeWitt Payne (Comment#45415) June 9th, 2010 at 3:00 pm
Based on the 2003-2009 average loss from this date, the projected minimum is 4.86 Mm2 with a three standard deviation range from 6.68 to 3.04 Mm2.
The AMO index at 0.505 is approaching the all time high of 0.562 in 1998. However, the UAH NoPol temperature anomaly is much higher now than in 1998. Three of the last four months have been above the 2007 high of 2.26 degrees. The high in 1998 was 1.16 degrees. I may have to lower my bet.
Two consecutive -0.1+ Mm^2 days!
Re: Phil. (Jun 27 09:00),
And it only gets worse. 6/28 was -0.2 Mm2. The form’s gone. I want to change my bet to 4.15 Mm2. Given the current state of affairs, record lows for both area and extent are odds on now. Of course, the Antarctic is now fairly likely to set a record high. IMO, this is not a coincidence. The Arctic and Antarctic tend to move in opposite directions in the short term. The heat that is being dumped into the Arctic by the positive AMO has to come from somewhere.
DeWitt–Sorry. I was took out scripts from old posts to see if that’s what’s causing problems when we all post comments. I put the script back into this one!
I hit enter rather than mouse clicking the suggested name. DeWi is me. Can you fix this or do I need to enter again correctly?
DeWitt Payne (Comment#47474) June 29th, 2010 at 10:23 am
Re: Phil. (Jun 27 09:00),
And it only gets worse. 6/28 was -0.2 Mm2. The form’s gone. I want to change my bet to 4.15 Mm2. Given the current state of affairs, record lows for both area and extent are odds on now. Of course, the Antarctic is now fairly likely to set a record high. IMO, this is not a coincidence. The Arctic and Antarctic tend to move in opposite directions in the short term. The heat that is being dumped into the Arctic by the positive AMO has to come from somewhere.
Actually I doubt whether the Antarctic will set a record, it seems to oscillate between the max and min, the melting follows the average fairly closely but the refreeze goes faster early, hence a big anomaly, but slows down to the same value for the max between 15 and 16Mm^2.
arctic sea ice area anomaly took a serious dive.
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/sea.ice.anomaly.timeseries.jpg
Re: sod (Jun 29 16:40),
Wow (in a bad way).