You are correct when you calculate that the 50% probability of breaking the record is reached in 69 days, but this is only the same as the mean time if the probability mass function is symmetrical – which it is clearly not in this case. This is a binomial calculation: so the expected mean of 100 observations = np = 100*.01=1. (If we want to get more technical, we would need to adjust for autocorrellation as a binomial distribution requires independent events while daily temperatures are path dependent)

Granting for a second all of your point (I’m not a climatologist or an expert on this data set): it seems like they are really all the same point, that the data pre-1979 may be suspect, or at least not as good as that point on.

The graph from 1980 on still shows a consistent downward trend, most pronounced in spring and summer, but also in the annual, and certainly since 1995.

Of course, there is a wealth of graphs available at
http://arctic.atmos.uiuc.edu/cryosphere/IMAGES/

Although individuals at Cyrosphere tend to be a bit more accepting of AGW fears than skeptical, their data since 1979 are well accepted by both Skeptics and pro-AGW adherents. (I do recognize that some Cyrosphere’s revisions have raised a few eyebrows by those who monitor Cyrosphere’s output closely.)
Neverthless, I believe one can be skeptical of the 1900 to 2007 graph that you posted from them for several reasons. First, there was no legitimate / consistant way to monitor ice extent before 1979. Second, the stability from 1900 to 1970 is suspect given PDOs, AMOs, ENSOs, etc. Third (to my knowledge) the methodology to develop pre-1979 estimates is not widely accepted. Fourth, ancedotal observations tend to contradict this graph such as passages through the Northwest Passage and local descriptions of open water — see http://wattsupwiththat.wordpress.com/2008/03/16/you-ask-i-provide-november-2nd-1922-arctic-ocean-getting-warm-seals-vanish-and-icebergs-melt/

“Do you have a source for the Antarctic having gained ice?”

See the bottom of the Cryosphere Today front page.

It’s difficult to see because a lot of the color spectrum is shades of purple and blue, so visually it all blurs together. Also, you have arbitrarily picked two points in time for comparison, so you can’t see any larger trends at work.

This is a better picture from the same site.

while the Antarctic is up about a million, so the net world anomaly is about zero right now.

Do you have a source for the Antarctic having gained ice?

The oil tanker which attempted the NWP was the SS Manhatten. However both her trips were made with the assistance of Canadian Coast Guard icebreakers.

Cheers

Actually, a city should see its 100-year high temps broken more frequently than every 100 days.

In any given year, the probability of not breaking a 100-year record is 1-(1/100) or 0.99. So the probability of not breaking a 100-year record in any given n-year period is 0.99**n.

We should expect those 100-year records to be broken about every 69 years because 0.99**69 is 0.4998–less than one half.

Of course, this only strengthens Climate-Skeptic’s point.

You could start with ClimateAudit, who have a post on this subject up at the top of the page at the moment.

A quick glance at the Cryosphere Today graphs indicates that the North pole ice area ‘normally’ varies between 15m sq km in March down to 5m sq km in September. (Last year it went from a high of 13.5m sq km to 3m sq km.) So it would appear that you normally only get 30% of the ice surviving, and 70% of the ice is new every year anyway. This year the peak was about 14m sq km, but this time only 15% of that is old ice instead of 30%. There’s a difference of about 2m sq km, and if you take McIntyre’s figures and suppose 75%-30% = 45% more of that is going to melt, you can expect an additional 1m sq km of melt because of the ice’s age, relative to the long term average.

But if you put proper error bars on that, the number is probably worthless.

Do you have a cite for that? Because it sounds like a warmist epicycle to me.