Abstract
Consideration of the inertial dissipation method for routine wind stress estimation suggests that the most significant errors are likely to be changes in height of the airflow before reaching the anemometers, and errors in estimating the true wind, due either to flow distortion-induced errors in the relative wind estimate or errors in estimating the ship's speed relative to the water. The results from four anemometers—Solent sonic and Kaijo Denki sonic anemometers, and R.M. Young propeller-vane and bivane anemometers—mounted on the foremast of a research ship were compared. The mean bias between the four anemometers in the friction velocity estimates was less than 3% (rms scatter 6%–12%). In contrast the bias and scatter for the drag coefficient was 17%–27% due to flow distortion-induced errors in estimating the true wind speed. It is concluded that, with a reasonably well-exposed anemometer, wind stress can be determined to 5% or better by the dissipation method whereas errors in the bulk aerodynamic method are likely to be between 20% and 30%.
The data from the two sonic anemometers showed the best correlation; the Solent sonic, a relatively new instrument, was comparable in performance to the Kaijo Denki. Comparisons of the two propeller anemometers typically showed twice the scatter compared to the sonic values. Overcorrection for the propeller response at low wind speeds resulted in spuriously high drag coefficient values for wind speeds below 10 m s−1. In contrast, the sonic anemometer data showed no change in the slope of the drag coefficient to wind speed relationship at low wind speed.
