I’m sure you have heard your T.V. meteorologist say, “We are expected to be 10 degrees above average today!” That sounds pretty impressive, right? But is it really? I have often seen T.V. meteorologists report the departures from normal or average for a particular day. But, what is really important is the distribution of maximum/minimum temperatures for that day. If the distribution is pretty spread out (meaning it has a high standard deviation), then being 10 or even 15 degrees above or below average may not be impressive at all.
I went to NOAA’s National Climatic Data Center online and retrieved daily temperature data for Houston, TX (Hobby Airport) from 1950 to 2012.
Below I plotted the density curve for the high temperatures for January 11 (from 1950 to 2012). The mean is about 61 degrees and the standard deviation is about 12 degrees. The NWS (weather.gov) is forecasting a high of 73 degrees for Houston’s Hobby Airport today. So this is 73-61=12 degrees above average. That sounds like a lot, but it is still just within 1 standard deviation of the mean.
What is even more interesting is that this density curve is bi-modal. And, 73 is pretty close to the second mode. This means that a high temperature of 73 occurs quite often – more often than the mean does.
Caveat: These data go back only to 1950. The data at weather stations probably go back to earlier. But, this is still a good example of how departures from “normal” may not be so impressive.