Anthony Watts has written about some research that shows that when weather observers estimate wind speed, they usually overestimate it. In April this year a team, led by Paul W Miller of the University of Georgia, published a paper in the American Meteorological Society Journal of Applied Meteorology and Climatology. The researchers reported that "As a general rule of thumb, humans overestimated nonconvective wind GFs [gust factors] by approximately one-third."In the USA, the scientists said that automated weather stations were relatively sparse, so weather observers apparently typically estimate wind speed. By comparing estimates made by observers with instrumentally recorded wind speeds in the GHCN network, the researchers concluded that the estimates were typically too high.
That's interesting. But wait. There's more. Anthony uses this research to accuse the thousands of volunteer weather observers of fudging temperature data. That's right. This is data that is read, not estimated. This is the "raw data" that deniers usually staunchly defend.
