Not statistics

Just a little bit of fun:

Washington — The global output of heat-trapping carbon dioxide jumped by the biggest amount on record, the U.S. Department of Energy calculated, a sign of how feeble the world’s efforts are at slowing man-made global warming...

In 2007, when the Intergovernmental Panel on Climate Change issued its last large report on global warming, it used different scenarios for carbon dioxide pollution and said the rate of warming would be based on the rate of pollution. Boden said the latest figures put global emissions higher than the worst case projections from the climate panel.

Now that we're in the middle of another global climate progress panic (I guess), I decided to do my own little climate study.  Now obviously, I don't have access to all the numbers that scientists do*, but I did manage to find record highs and lows for Topeka as provided by the National Weather Service going back 120 years for a little non-scientific study of my own.

I picked Topeka because, hey, what better stand in for the "northern hemisphere" in the famous hockey stick graph?  And I also picked highs and lows because besides immensely reducing the number of numbers we have to deal with, since we are looking for broad trends, these numbers might be all we need.  For it seems to me that if we are really getting warmer, then the number of days which are the warmest of that day ought to increase as we get closer to today. Right?  You can't have the "warmest decade ever" without a good number of those days being the warmest days ever**.

So I made a spreadsheet of the warmest and coolest days and sorted them by decade. The results are interesting:
1880/90s: 21 (all-time highs)
1900s: 8
1910s: 20
1920s: 14
1930s: 105
1940s: 26
1950s: 34
1960s: 16
1970s: 16
1980s: 24
1990s: 23
2000s: 35
2010s: 11

We should expect that latter periods will be a little bit overweighted***, and the number of days in which the record high was reached in the 2000s was on the high side. However, the 1930s had three times as many high record days as the so-called 'warmest decade on record,' and that after up to 80 years for statistical outliers to wipe them out. Given that the 1930s have but 11 record record low days (compared to 21 for the 2000s), it is very hard for me to buy a the idea that the 2000s constituted the warmest decade on record.

However, there is probably little doubt that we have the highest CO2 concentrations on (recent) record. To my unscientific little mind, that does not bode well for a CO2 temperature correlation.

* Especially the numbers just above the double line on a grant application.
** Well, I suppose that theoretically you could so long as the temperature increased while the spread of temperature decreased. In other words, if the weather got both warmer more placid.  Not a bad thing, all told.
*** After all, the 1890s have had a full century in which truly 'freak' days could wipe out their records. This year's days? Not so much.