DESIRABILITY OF USING A FAST SAMPLING RATE FOR COMPUTING WIND VELOCITY FROM PILOT-BALLOON DATA

Abstract

Advantages of using a fast sampling rate for providing raw data for computing wind velocities from pilot-balloon observations are discussed. Two independent approaches show that (for a given sampling rate) the errors in the mean wind computed through a layer tend to be smaller the larger the layer. However, if many observations are used to compute the mean wind through a layer, the computed wind is more accurate than that obtained by using just two observations, one at the bottom and one at the top of the layer. Graphs are presented which give relative estimates of wind errors as a function of sampling interval, sampling rate, and mean wind speed.