IMPROVED WIND AND TURBULENCE MEASUREMENTS USING A LOW-COST 3-D SONIC ANEMOMETER AT A LOW-WIND SITE

Abstract

A year of data from sonic anemometer and mechanical wind sensors was analyzed and compared at a low-wind site. Results indicate that 15-minute average and peak 1-second wind speeds (u) from the sonic agree well with data de- rived from a co-located cup anemometer over a wide range of speeds. Wind direction data derived from the sonic also agree closely with those from a wind vane except for very low wind speeds. Values of standard deviation of longitudinal wind speed ( u) and wind direction fluctuations () from the sonic and mechanical sensors agree well for times with u > 2 ms -1 but show significant differences with lower u values. The most significant differences are associated with the stan- dard deviation of vertical wind fluctuations ( w): the co-located vertical propeller anemometer yields values increasingly less than those measured by the sonic anemometer as u decreases from 2.5 approaching 0 ms -1 . The combination of u over-estimation and under-estimation of  w from the mechanical sensors at low wind speeds causes considerable under- estimation of the standard deviation of vertical wind angle fluctuations (), an indicator of vertical dispersion. Calcula- tions of  from sonic anemometer measurements are typically 5° to 10° greater than from the mechanical sensors when the mechanical instruments indicate that  < 5° or so. The errors with the propeller anemometer, cup anemometer and wind vane, caused by their inability to respond to higher frequency (smaller scale) turbulent fluctuations, can therefore lead to large (factors of 2 to 10 or more) errors in both the vertical and horizontal dispersion during stable conditions with light winds. The sonic anemometer clearly provides more accurate and reliable wind data than the mechanical wind sensor with u < 2.5 ms -1 .