Another setback for Teh Global Warming™

Meteorologist Anthony Watts compiled the results of all the sources. The total amount of cooling ranges from 0.65C up to 0.75C -- a value large enough to erase nearly all the global warming recorded over the past 100 years. All in one year time. For all sources, it's the single fastest temperature change ever recorded, either up or down.

Snow cover over North America and much of Siberia, Mongolia and China is greater than at any time since 1966.

The U.S. National Climatic Data Center (NCDC) reported that many American cities and towns suffered record cold temperatures in January and early February. According to the NCDC, the average temperature in January "was -0.3 F cooler than the 1901-2000 (20th century) average."

Originally Posted by ShawnJ

Whew, for a second there I thought global warming deniers didn't exist anymore.

Originally Posted by franksargent

First the source of the dailytech hit piece;

January 2008 - 4 sources say globally cooler in the past 12 months

Wow, four sites have temperature data that agree with each other. D'oh!

Oh, and it's nice that this source has links to the underlying data, makes my job really easy!

So I took the HadCRUT3 data set (easiest to import into Excel), the time series goes back to January 1850 (as does the GISS data set), N (data points) = 1897.

I then took a 12-month lag window of the entire data set (reducing N by 12 to 1885) and sorted in ascending order of temperature data, turns out that 16 events exceed this one, so that this is approximately a 3 sigma event. Cool!

The entire 12-month lag window data set is, of course, randomly distributed with each month having 157 points (one month had 158 since 1885 isn't divisible by 12)

I then dusted of my extreme value theory tool box to take a closer look at the data set, by windowing the data to exclude the middle 83.3 (or 5/6) percent (i. e. only including the lowest 8.33 (or 1/12) percent and the highest 8.33 (or 1/12) percent). Thus, the reduced data set consisted of 314 extreme events.

The distribution of these extreme events had the following monthly distribution;

AUG = 10
SEP = 11
OCT = 15
NOV = 28
DEC = 43
JAN = 55
FEB = 48
MAR = 34
APR = 25
MAY = 22
JUN = 11
JUL = 12

So what does the above distribution indicate?

Well, that year-to-year extremes are much more likely to occur in the winter than in the summer, that the extremes are not randomly distributed, and that the peak of the distribution is centered on the DEC-FEB timeframe (n = 146), and the trough of the distribution is centered on the JUN-AUG timeframe (n = 33). D'oh! In other words, winters exhibit ~ 4.4 times the variability as do summers for temperature extremes.

We now have conclusive proof that summers are hot and have less variability of extreme events, and that the winters are cold and have greater variability of extreme events.

As if we didn't already know this.

Hey SDW10000BC do you have any more interesting AGW contrarian links that I can pick apart?

And while your looking, here's a link for you;

2007 Was Tied as Earth's Second-Warmest Year

Originally Posted by Harald

Yes, you muppet, weather and the ecosystem are complex interdependent systems. Which is why short-term patterns are hard to predict but long-term trends reveal themselves through data.

Your laughie at the end of your sentence shows you don't understand science. Not that you've failed to grasp this issue, but that you don't understand fundamental science. It makes you look ignorant. Not clever.

For example, the Pentagon in 2004 stated there was a good chance that 'global warming' could lead to the UK freezing as the changes to global oceanic heat conveyers stopped conveying heat to the UK. Which is more northerly then every major Canadian city.

And by the way, Mr. Scientist, the plural of 'anecdote' is not 'data' as the saying goes, but you're clevererererere then all the scientists so you probably knew that.

Bell end.