The Blackboard

Just as a point of reference, the value for November 1-7 for 2002 was 9,088,000 km^2, and 2004 was 9,101,000 km^2 from the available data. (2003 was missing, and there’s a traffic jam in the area, so I stopped looking at the other years.)

I’ll be looking at the numbers and providing data for people to ponder. But I know loads of people interested in the ice have their own favorite methods. So, I’d advise those who seriously want to win brownies to look at the data, and also visit the ice thread at climate audit: http://www.climateaudit.org/?p=3819#comment-304208

Waiting for Oct 25th to roll around before predicting.

From my spreadsheet, the 11/1 to 11/7 averages are (km2):
2002 9088370.571
2003 11001138.57
2004 9101071.571
2005 9119397.571
2006 8886406.429
2007 8536629.714

I’ve linearly interpolated any missing numbers. I’d have to go back to the original data to see if that had an effect on the averages.

The entire week was missing in 2003 from the current text file DeWitt.

I’ll be happy to bet, but if I am to employ my favourite, the “GAS_STATION_LOCALIZATION_PRINCIPLE”, I need a number of others to bet first. It’s a post hoc strategy.
BTW, the steep curve fot ice growth is even more remarkable as the SST of the polar seas seems to be above average. I would expect it to level off soon.

Please note that the false coloring could easily be misinterpreted. I think the “standard” temperature for ice is – 1.7 deg.C.

Cassanders
In Cod we trust

Even worse, I used 12/1 to 12/7 for 2003. The interpolated average for 2003, FWIW should have been 9015941 km2. CT seasonal extent data doesn’t agree all that well with JAXA and Uni-Hamburg. Uni-Hamburg only goes through 9/23 at the moment. Regressing CT on JAXA gives 6866517 km2 for JAS 2008, but the error in the fit is on the order of 50000 km2 so there are still several potential winners. Atmoz has an outside chance, but the most likely are Lucia and Cassanders.

After plotting UH, JAXA and CT summer extents, I can see why CT to JAXA or UH fits are so bad. This site doesn’t take links to commercial hosting sites?

This increase is not unprecedented. Large extent increases occurred in 2003, 2005, and 2007 of over .25 mil km2. 2008 has been running over average for about a month.

Here’s a graph plus minus 30 days .

I bet 9.3X10**6
BTW I think the increase is so sharp because the baby ice melt count did not account properly for floating baby ice chunks which have quickly recombined

I just think that baby ice is an energetic little chap! (Alternatively that we have av 15-20 year cold record on Arctis, but I’m not so sure about that…)

Magnus– I think we need Jeez to tell us the tale of how the baby ice leader saved and hid the vulnerable ice.

Yup – overheard that…

OT,

But here’s Gavin on GCMs relating to new research on water vapour

” The data that are going to come out of this experiment really will allow us to go forward with confidence and know what it is we are modelling“

http://www.nature.com/news/200.....5714a.html

When are we ever going to get a serious in depth discussion of GCMs and their limitations? ( BTW this is not a dig at you Lucia just a general expression of frustration)

Count me in. If the winner, I’ll donate the brownies to someone more needy. ie; thinner.

I get a nearly perfect 2nd order polynomial fit (R2=0.9883) using:

2921.4×2 – 46593x + 5E+6

with y at x = 1 being the years low ice areal coverage.

My guess for the Average from Nov 1 – Nov 7 = 11,500,000 sq km. I took a little off, as a fudge factor for the apparently normal inflection in the curve around Nov 1.

And yeah, I know 3rd to 5th order polynomials fit a little better, but the numbers are HUGE.

This reminds me of a contest during my freshmen year at MIT many moons ago. The object was to predict the time when a block of ice placed in the main entrance on Mass. Ave. would be completely melted. I knew nothing about thermodynamics or heat transfer yet, so it was actually quite intimidating that I didn’t know how to approach it other then to guess.

This problem could be similar as to the intimidation factor, except that I am older and wiser now and not so easily intimidated and you know what? I feel some brownies coming my way.

Don’t anyone else bother trying to guess because I have a secret arctic energy transfer model that teleconnects with a GCM and a solar model (their secret too).

Those brownies are MINE!

Back on Sept 23 there was a tiny post on icecap.us:

Sea surface temperatures along the Alaska Chukchi and Beaufort Sea coasts are 2 to 8 degrees Celsius colder this year than at the same time last year. Significant ice will begin developing along the Alaska coast north of 70n within the next 10 to 14 days.

That seems pretty significant to me and the rapid increase in ice seems to reflect that.

As the ice extent reaches average levels I think it will will start to increase at a more normal rate.

My entry for the contest is 8,895,422 km^2

This is kinda silly but i was thinking of my old Grandpa today. He really loved those R/C model airplanes. He told me once, “You know, boy, that radio don’t control that model aeroplane, I do.” He would spend hours and hours building those models. He would make them in every shape and size you can imagine, circles, triangles, rectangles and octagons. It seems no matter what he built, it would fly. And while it was flying, he was so, so proud. Of course they would all eventually crash to the ground. Sometimes he would take the pieces and cobble it back together, and sometimes he would just start over. I really miss old Grandpa, but I could have done without those GCMs…
Grandpa Controlled Models…

I’ll go 8,650,000 km^2. I’d actually prefer something about 8,900,000. My assessment of current SST anomalies around about where the ice front has been in early November in recent years according to CT, suggests around an average amount of ice at this stage. I expect most people will be overpredicting ice extent due to the current fast freeze, so think I have a better chance of winning going on the low ice side. So with Bill’s guess I’ll just go even lower again, and hope most others go higher.

I’m in Australia so if I win please pass my brownies to the next closest guess.

11,300,000 km^2 I have near that much in frost this morning…

Excellent!
I think Michael has some good points WRT SST, but I think he was slightly too cautious. Hence I go for 8,651,000 km^2.
And of course I am then employing the “bensinstasjonslokaliseringsprinsippet” (The-Gas-Station-Localization-Principle).

Cassanders
In Cod we trust

I’ll go for 9,750,000 Km^2, although if the current slope continues it looks to be 10,300,000

I’ll play. I’ll use the BTSOMP (By the seat of my pants) WAG method to predict 10,153,864 square kilometers. While I live in the contiguous 48, I’ll gladly donate any brownies I might win to a food drive or shelter in Lucia’s vicinity. Everyone deserves a treat as Thanksgiving approaches.

The current trend will continue, so: 10,300,000 km2
Why not?

The first week in November has been a gathering point in recent years when all the trajectories cross so I’ll go for 8.9 Mm^2.

I’ll speak to jeez about the writing assignment

Can I change to 8,652,000?

Or maybe I’ll make a sockpuppet who will bid for 8,652,000

1/11/08-7/11/08

9,186,585

@Michael
If you are allowed to go for 8,652,000 I should be allowed to go for 8,653,000, and so on. If you increase to where I think we overshoot, I’d reduce my following bet. I do of course not need to use incements/reductions of 1,000 km^2 ( but would like to have some reasonable increment relative to the resolution of the measurements).

Anyway, I would think Lucia expect us to behave and stick to our original bets

Cassanders
In Cod we trust

No sock puppets! I plan to write a script that will let people auto enter. I’ll pre-enter everyone who has already voted. I think I’m going to let you change your bets up until Oct. 25, but in which case, I’ll also report how many time you changed your bet! Heh.

(If I get a script running, I’ll also use the register features so we can minimize sock-puppetry!)

For now, hang onto your votes.

[...] figure shown above will be updated on the actual “Ice Recovery Brownie Bet” post. The mystery numbers after each bettor’s name indicates the comment number and whether [...]

I will give the scientifically correct figures nex weekend, but may I, in Scandianvia, get paid via Paypal instead? Then You, millions of Americans, can – when I very likely win this – send me $2 each.

/humbly, Magnus

Magnus–
If I’m going to start sending money, I might need to force losers to send money to me! That would take the fun out of it.

Will you be using this money to buy Melort? Or Lutefisk? I wonder if I can get some of my father-in-law’s cousins to make Lussekatter for you!

For those wondering what Lussekatter are here’s an image and link to a recipe:

Ok, I succumb. The whole world (or at least a large part of the financial bit of it) seems to have been heavily engaged in betting on the future so why shouldn’t I have a go. I reckon it will be 11,750,000km^2.

If I am right please donate brownies to deserving local kids.

9,543,210.987654321 square kilometers. Or so.

Please delete…

Oh, no 2 million dollars… Lutfisk, no, but a semla would be fine. I may not win because of these natural changes which mess up science.

9.25 Million square kilometers, using Mk1 eyeball and hunch.

My bet is 1.07 e07. I actually thought it would be 1.06 but MikeB beat me to it. Darn. Well maybe SC24 will be slow enough on the upatke, I will win anyway.

lucia (Comment#5792) October 11th, 2008 at 6:14 am

“For now, hang onto your votes.”

I agree. If you allow people to keep to revising their bets your gonna make a whole lot of work for yourself. I think as this is the first sea ice extent derby we need to give better cookie odds to those who placed their bets early.
We could do this with a sliding scale algorithm with multiple smoothing averages and throw out anything that stuffs up the result we want.

Next file it in a directory marked “censored” and pretend it’s lost if anyone asks for the details.

Hey, Lucia, you didn’t include me on the graph on the second post. I’m still interested. No change in my prediction.

Keith (Comment#5774)
October 10th, 2008 at 9:51 am

I’ll play. I’ll use the BTSOMP (By the seat of my pants) WAG method to predict 10,153,864 square kilometers. While I live in the contiguous 48, I’ll gladly donate any brownies I might win to a food drive or shelter in Lucia’s vicinity. Everyone deserves a treat as Thanksgiving approaches.

Lucia, I suggest you cut-off the betting earlier, say maybe the 15th. As time gets closer those who wait till the last minute will have a better chance. But with that said –I’m feeling lucky–I can taste those brownies already!

Bob–
I started this one quickly due to the impressive appearance of the recovery. Obviously, those who hold off betting have an edge! That’s why I’m thinking about letting people update. Obviously, people who bet last week should be allowed to update.

But the way Mike is cutting off Cassanders isn’t right.

I was mulling over “rules” for updating. I was thinking, given the total spread of bets is 3 million km^2, an updated bet has to differ by at least 0.3 million km^2 from the previous bet. That would let people who realize they goofed update, but it wouldn’t let Mike come in cut off Cassanders!

But if I’m going to create complicated rules, I need to write a script that keeps track of values. Then, I can describe what the rules are. (And, of course, make the script available for V&V!)

[...] be posting all the bets and providing all relevant links by updating the brownie bet post. « Who has bet on the Nov. 1-8 NH Ice Extent so [...]

My guess: 9.15 million sq. km

SInce you posted MIkeR at 1.07. Give me 1.06 e07. It was what I wanted, but someone changed their mind it appears.

Hmmm I thought I had 10.6 but I’ll take 10.5…

The first 4 months of arctic winter look like one full SD to the negative temperature wise, and the water is already colder than average.

9.999E+06

http://www.weatheroffice.gc.ca.....=s234pfe1t

I’ll take a stab at 9.608 please.

Comparing the year-to-date annual average with the 11/1 to 11/7 averages and throwing out 2007 as an outlier, I’m going with 9.022 Mm2 (9.022E06 km2).

As far as changing one’s bet, how about only once per day. That’s probably too many. How about changing no more than two times.

Agree with dewitt, you should have one guess and then one Mulligan, to quote a golfing term. The second guess no later than October 25th.

My guess is, based on how 2008 matches 2004 spatially on pretty pictures

9.125E+06 Km2

None of this mathematicals for me, I’m an artiste!

Regards

Andy

PS Cookies are illegal in the UK so you will have to send beer, which is not.

I’d like to place a bet for 9.250 million km^2. Just a guess that we’ll be looking at higher ice extent than the last 6 years. Since I live in Alaska I’d be more than willing to subsidize shipping, or barring that I have a proxy recipient in mind.

I’ll take 9.6 million km^2 by the guesswork method. If I win since I’m in Aust (G’day Michael) please pass the brownies on to a homeless shelter or food bank as per Keith’s suggestion.

I’ll take 8.765 million km^2, anything over 9.2 is for fools! I’d take some M&M cookies

Hi Lucia

I think a Gremlin has been active. My bet should be:

MarkR (Comment#5805)
October 11th, 2008 at 6:41 pm

9.25 Million square kilometers, using Mk1 eyeball and hunch.

I’ll go for 10.5 x 10^6 please!

Less than 3.0M km2 to go and we’ve already knocked off the 05 trajectory! Its definitely getting dark early up there.

‘the way Mike is cutting of Cassanders is not right’

(in a whiny 8 year old voice)

But she cut me off first.

@Michael
(In a mature voice) To be fair, you are sort of right.
(Except that I’m a male)

I am sorry if I conveyed a “whiney” attitude to anyone. On the contrary, my comments are made with a grin. (I just find it tiresome to splatter smilies all around) .
Anyway, I belive our gracious hostess still will let us bet, so I think I work on a strategy through the weekend.

Tally ho!
Cassanders

Michael–
Yes. I agree Cassanders cut you off first! That’s one of the reasons I’m trying to develop better rules. I figure it will take me an iteration, and some thought. I have a list of ideas, but also a “to do” list for the next bet. (I’m planning a winter maximum bet!)

Things I’m thinking of:
1) Let each person enter directly into database.
2) Keep individual bets private until betting is closed. (Or maybe show in bins? Like a histogram?)
3) Let people change bets– but whoever gets a particular bet in first wins.
4) Permit people to pick their prize– brownies or chocolate chip cookies.
5) Maybe force people to register to avoid sock-puppet bets.
6) Post the averages and standard deviation of all bets on the fly.

But I find it’s best to have all the rules before I write the script!

Rule 7) You are only allowed to bet when you are smiling. If you’ve had an argument with your partner go make up with him/her first and then come bet.

Private bets might in some sense be farer. But half the fun is explaining your logic and reading other people’s reasons for their bets.

And Cassanders I did not find you to be whiny. Probably more whimsical.

Put me in for:

13.343 million km^2

An illustration of the power, and folly, of curve fitting and extrapolation.

As I expected, the rate of increase has decreased

I’ll stick with my bet of 8,895,422 km^2 even though Phil coming in at 8.9 didn’t give me much breathing room!

Bill Jamison (Comment#5915) October 20th, 2008 at 1:11 am

“As I expected, the rate of increase has decreased.”

I’m still happy with my early placed bet.

I think much more is not taking into account SST and currents.

9.1×10e6 based as are a bunch of others on the trusty Mk 1 eyeball

As I’m a resident of France, if I win you can donate the cookies to your favourite good cause

I’ll play – 9.06X10^6

My guess is 9.0210 ×10^6 based on a CA zipcode popularized on TV.

8.421E6

8930563

The current NH sea ice extent is now within normal STD
8,017,969 km2 (October 22, 2008) , antarctica is NORMAL. Prediction both will go ABOVE anomaly for 2009-2010 and maybe beyond=end of AGW LOL

ok, i will go with 9.523 million. i hope that is kosher.

My bet is 10,024,000 km2 NH sea ice extent for Nov 1-7.

The 11 PM Eastern time JAXA extent is a preliminary figure. It will probably be too low during a period of rapid increase and too high during a rapid decrease. You should wait for the 10 AM Eastern update the next day. For 10/22/2008 the final result is 8,087,969 km2, 70,000 km2 higher than the 11 PM estimate.

I’ll take 9.647 please.

lucia,

My foolish bet, based on the 2007 recovery curve, is 9,258,000 sq km

@Michael (&Lucia)
As we are still far from running into an infinite regression, the “GasStationLocalizationPrinciple demands me to second-guess for 8,653E+06 km^2

Cassanders
In Cod we trust

Given current trend in the JAXA data, plus predicted continued positive Arctic Oscillation (http://www.cpc.ncep.noaa.gov/p.....ensm.shtml), plus predicted day 8-10 500mB height anomalies (http://www.meteo.psu.edu/~gado.....tcomp.html), I expect a leveling off of the rate of increase by the Nov 1st timeframe. So my prediction is 9.25 Million sq.km.

Lucia, for chocolate cookies my bet is 9,050E+06mm^2

As a long time reader of this and ClimateAudit, but seldom a commenter, I offer a WAG:

11,200,000 km2

cheers,
Robert

5. Bets must be posted by Oct. 25, midnight, Chicago time!

We are under starters orders. Last bets please.

Commentary as the race progresses, I’m sure will be welcome.

This is the first Ice Extent Derby and any stuff ups will be blamed
on CO2.

We thank our sponsors Al Gore and the Goddard Institutes
James E. Hansen.

9527972…….argggg…..

no wait: 9245684 err 9.2456 x10^6. final answer.

I think (for no particularly good reason besides looking at the various years trends) that 08 will slightly exceed 05’s numbers.

Thus 9,226,000 km 2

Arthur

oops typo in my bet, try

10,200,000 km2

instead.

(Still hopelessly optimistic!)

Robert

Please put me down for 9.109 (x10^6). Now that I’ve typed this, I am looking over some other bids and see mine is scrunched in tightly between several others. Oh well, no time now to change. Lucia, thanks for all the fun.

Michael (#5788) 8,652,000
Cassanders (#5791) 8,653,000
Matt Edin (#5847) 8,765,000
Bill Jamison (#5759) 8,895,422
Phil. (#5778) 8,900,000
mccall (#5938) 8,930,563
Dan McCune (#5932) 9,021,000
DeWitt Payne (#5830) 9,022,000
Paul Demmert (#5967) 9,050,000
Bob North (#5931) 9,060,000
FrancisT (#5930) 9,100,000
chopbox (#6008) 9,109,000
AndyW (#5833) 9,125,000
Jared (#5825) 9,150,000
MartinGAtkins (#5790) 9,186,585
adoucette (#5998) 9,226,000
steven mosher (#5984) 9,245,684
Earle Williams (#5843) 9,250,000
MarkR (#5805) 9,250,000
Fred Nieuwenhuis (#5962) 9,250,000
Philip Mulholland (#5960) 9,258,000
Bob B (#5751) 9,300,000
mark (#5940) 9,523,000
Alan S. Blue (#5799) 9,543,210
AndrewK (#5844) 9,600,000
Mike N (#5829) 9,608,000
Jamie Dyer (#5946) 9,647,000
Steve Keohane (#5773) 9,750,000
hswiseman (#5828) 9,999,000
Donald Wilson (#5944) 10,024,000
Keith (#5774) 10,153,864
Robert Leyland (#6003) 10,200,000
Bruce Cobb (#5777) 10,300,000
Mike Bryant (#5827) 10,500,000
PaulM (#5849) 10,500,000
John F. Pittman (#5826) 10,700,000
DaveM (#5770) 11,300,000
Les Johnson (#5757) 11,500,000
Dave Andrews (#5798) 11,750,000
Dan Hughes (#5903) 13,343,000

Martin, you forgot our host!

Michael (#5788) 8,652,000
Cassanders (#5791) 8,653,000
Lucia (Post on 10/13) 8,728,000
Matt Edin (#5847) 8,765,000
Bill Jamison (#5759) 8,895,422
Phil. (#5778) 8,900,000
mccall (#5938) 8,930,563
Dan McCune (#5932) 9,021,000
DeWitt Payne (#5830) 9,022,000
Paul Demmert (#5967) 9,050,000
Bob North (#5931) 9,060,000
FrancisT (#5930) 9,100,000
chopbox (#6008) 9,109,000
AndyW (#5833) 9,125,000
Jared (#5825) 9,150,000
MartinGAtkins (#5790) 9,186,585
adoucette (#5998) 9,226,000
steven mosher (#5984) 9,245,684
Earle Williams (#5843) 9,250,000
MarkR (#5805) 9,250,000
Fred Nieuwenhuis (#5962) 9,250,000
Philip Mulholland (#5960) 9,258,000
Bob B (#5751) 9,300,000
mark (#5940) 9,523,000
Alan S. Blue (#5799) 9,543,210
AndrewK (#5844) 9,600,000
Mike N (#5829) 9,608,000
Jamie Dyer (#5946) 9,647,000
Steve Keohane (#5773) 9,750,000
hswiseman (#5828) 9,999,000
Donald Wilson (#5944) 10,024,000
Keith (#5774) 10,153,864
Robert Leyland (#6003) 10,200,000
Bruce Cobb (#5777) 10,300,000
Mike Bryant (#5827) 10,500,000
PaulM (#5849) 10,500,000
John F. Pittman (#5826) 10,700,000
DaveM (#5770) 11,300,000
Les Johnson (#5757) 11,500,000
Dave Andrews (#5798) 11,750,000
Dan Hughes (#5903) 13,343,000

Although technically, isn’t Phil.’s bet only 8,900km^2?
BTW, what does Lucia get if she wins? It doesn’t seem to have been resolved yet.

Martin and Mike N!
Thanks for the time saving work! That will make it easier for me to put in the spreadsheet. (I was going to do that today. Weekends are busy for me.)

This graph shows why the sun controls our climate
http://en.wikipedia.org/wiki/I.....umbers.png

Most here would probably dispute this but the 11-13 years more active cycles clearly are related to an AVERAGE increase since the 50’s: Ocean warms a bit more (even though there are ups and downs). We are probably on a down phase starting 2008 (now). It may be that the rapid increase in ICE and lower temps recorded lately show this.

Re previous AVERAGE global temperature increase

Another issue: Has it been established that the rising C02 is due to ocean release or human activity, or better can it be established?

It looks like the rate of increase has peaked more or less on schedule. Today’s increase was only 35,938 km2.

Mike N (Comment#6017) October 27th, 2008 at 1:51 am

“Although technically, isn’t Phil.’s bet only 8,900km^2?”

Yes..but it has been accepted by our host and is on the graph.
He made the bet early so I assume the typo has been overlooked.

Lucia (Post on 10/13) 8,728,000

I can’t find it but who’s gonna argue with the boss.

I presume there are no more objections or additions.
Thanks for your backup work.

MartinGAtkins (Comment#6034) October 27th, 2008 at 9:18 am
Mike N (Comment#6017) October 27th, 2008 at 1:51 am
“Although technically, isn’t Phil.’s bet only 8,900km^2?”
Yes..but it has been accepted by our host and is on the graph.
He made the bet early so I assume the typo has been overlooked.
Lucia (Post on 10/13) 8,728,000
I can’t find it but who’s gonna argue with the boss.
I presume there are no more objections or additions.
Thanks for your backup work.

Actually you’re wrong, according to the SI convention I’m the only one who’s close , technically everyone else is a million times too low! Thanks for pointing this out I can claim brownies on a technicality.

http://physics.nist.gov/Pubs/S.....html#6.2.1
“6.2.3 Inseparability of prefix and unit

The grouping formed by a prefix symbol attached to a unit symbol constitutes a new inseparable symbol (forming a multiple or submultiple of the unit concerned) which can be raised to a positive or negative power and which can be combined with other unit symbols to form compound unit symbols.

Examples: 2.3 cm3 = 2.3 (cm)3 = 2.3 (10-2 m)3 = 2.3 × 10-6 m3
1 cm-1 = 1 (cm)-1 = 1 (10-2 m)-1 = 102 m-1
5000 μs-1 = 5000 (μs)-1 = 5000 (10-6 s)-1 = 5000 × 106 s-1 = 5 × 109 s-1
1 V/cm = (1 V)/(10-2 m) = 102 V/m”

Phil,

Actually you’re wrong, according to the SI convention I’m the only one who’s close

Nope. My bet was in Mm2 as well. Specifically 9.022 Mm2 = 9.022E06 km2

Nope. My bet was in Mm2 as well. Specifically 9.022 Mm2 = 9.022E06 km2

Aha, I seem to remember you and I having this conversation at another place.

Actually, the scale on the graph is wrong. The number plotted is the actual value multiplied by 10^-6.

If the reported scale is assumed to be correct, a value from the graph, 8.0 for example, would read

Sea Ice Extent x 10^6 = 8.0. So, Sea Ice Extent = 8.0/10^6.

And the km^2 is another problem with the scale. It should read

Sea Ice Extent x10^-6 / (km^2). The example then becomes

Sea Ice Extent = 8.0 x10^6 km^2.

All assuming of course that the extent is indeed of the order of 10^6 km^2. I don’t know.

Phil. (Comment#6048) October 27th, 2008 at 12:09 pm
“Actually you’re wrong,”

I used the maxim “common sense prevails”

The intention of the poster was more important than the exact math.

It was difficult enough compiling the final bids without people using pedantic mathematics.

Is everyone happy with the final result?