misleading weather statistics

I heard on the radio a couple of weeks ago that the hottest part of the year is usually the third week in July. Then I saw a graph on the internet, or maybe it was the Atlanta newspaper, of average highs and lows for each month, and the bars for July and August looked the same, which would go along with the turning point being near the end of July or beginning of August.

But it didn't seem right to me. It seemed to me that the hottest time was usually in August. Well, now August is here, and I see what the situation is. The highs and lows may be similar (although the lows still seem to be getting a warmer). But the big difference is that it is staying warmer much later at night. So while the average of the highs, and the average of the lows, might be similar for the two months, if you take an average of hourly temperatures, I bet August would be clearly hotter.

And it's getting worse:

http://www.msnbc.msn.com/id/14165332/

America in recent years has been sweltering through three times more than its normal share of extra-hot summer nights, government weather records show.

But it is not surprising because climate models, used to forecast global warming, have been predicting this trend for more than 20 years, said Jerry Mahlman, a climate scientist at National Center for Atmospheric Research and a top federal climate modeler.

One reason global warming is suspected in summer-night temperatures is that daytime air pollution slightly counteracts warming but is not as prevalent at night, said Bill Chameides, a climate scientist for the advocacy group Environmental Defense.

---