Abstract
In a recent paid, the author reviewed methods for estimating the mean and the standard deviation of wind direction. Some comments on this subject are now added. If large sporadic eddy fluctuations do not occur in a sampling duration, we can exactly calculate an arithmetic mean and standard deviation of wind direction by using an approach suggested by Mitsuta. This is via a single-pass procedure without considering the discontinuity in wind direction scale. By using the Mitsuta approach it is found that Mardia's method provides a good estimator of the standard deviation of wind direction even in the case when the range of fluctuations of wind direction is wide.
