ERRORS IN UPPER-LEVEL WIND COMPUTATIONS

Abstract

The evaluation of upper-level wind data by means of balloon-tracking methods involves a number of determinative and random errors. These errors depend on the precision in locating the points of the balloon track, which in their turn are related to the accuracy of measuring quantities such as azimuth, elevation angle, slant range, balloon height, and time. Since the number of quantities needed for plotting the horizontal projection of the balloon's track by radar systems is supernumerary, the observations may be arranged in three classes. The purpose of this paper is to determine which of these classes yields the best precision of measure in wind speed and direction. Moreover the errors are considered as a function of balloon height, wind speed, rate of ascent and time interval between two successive measured points. For special cases the extreme values of the random errors dependent on the location of the balloon's trajectory with regard to the direction of observation is presented in the form of tables.