Bias in weather-station measurements

E. Linacre


The inauguration of automatic processing of hourly temperature measurements at American airports has led to a consideration of the effects of instrument error at ten airports over a year (1). The conclusion was reached that differences between the old and the new instruments accounted for errors of about 0.2 K at night-time, but differences between the instrument sites within the same airport can cause biases of 1.0 K or more. The bias is also affected by the time of day and the kind of weather being experienced. The biases are largest during the daytime under calm and clear conditions.



(1) Guttman, N.B. and C.B. Baker, 1996: Exploratory analysis of the difference between temperature observations recorded by ASOS and conventional methods. Bull. Amer. Meteor. Soc., 77, 2865-73.