The presentation of precipitation information in televison broadcasts: what is typical?
Singer, Marc
A statistical analysis of precipitation values for Columbia,
Missouri was performed using monthly averages for the 1971-2000 period.
The data was acquired from the Missouri Climate Center. The analysis
demonstrates that, unlike temperature data, precipitation values are not
normally distributed, but are closer to a "gamma"
distribution. Precipitation data is typically presented as a cumulative
monthly total in television broadcasts, unlike temperature information.
Thus, a typical daily range for precipitation cannot be displayed as has
been recommended for temperature in order to incorporate simple
statistical information into a broadcast. Precipitation events result
from "randomly" occurring transients, thus, comparing daily
precipitation information to daily statistical precipitation information
is meaningless. Here it is suggested that historical monthly rainfalls
can be used to partition the data into quintiles following Guinan
(2004). These quintiles can then be used to compare cumulative monthly
precipitation to typical by displaying "stretching" quintiles
in order to provide the viewer with a better statistical context for
precipitation data.
* Hagen, T.E., J.M. Glisan, A.R. Lupo, and P.E. Guinan. Department
of Soil, Environmental, and Atmospheric Sciences, University of
Missouri-Columbia.
