Recently I’ve been paying a lot of attention to the daily low temperatures reported… and it turns out you have to be really careful with them because the definition of the low temperature used is inconsistent and potentially very weird.
For example my phone uses weather channel data, I think, and over the weekend was reporting the low temperature for Monday as -17 (west suburbs of chicago). However I looked at the hourly forcasts and that was totally incorrect. The low for monday was actually expected to be -10 monday afternoon it was going to be around -5 (but windy) at 6-7am when the low temperature usually occurs. -17 was the temperature for 6-7am tuesday! For tuesday it showed the low temperature at -10 which was actually going to happen in the wee hours of wednesday. Turns out the definition of “low for the day” used was “the lowest temperature between 7pm of the day and 7am of the following day”. Which is ridiculous as the low temperature one would experience for a particular day, in the winter here anyway, is usually right around sunrise. A cutoff of 2 or 3 am might be defensible, but I don’t think many people expect 6am wednesday to be reported as tuesday.
I read a post by a tv weatherman explaining his lows. It sounded like he was doing a better job reporting "morning lows’ and “overnight lows” rather than just “lows”.
