ISSN:
1573-1472
Source:
Springer Online Journal Archives 1860-2000
Topics:
Geosciences
,
Physics
Notes:
Abstract Even though propeller anemometers are found to give outputs which deviate from the desired cosine relationship by an amount which varies with wind speed, their overall performance is consistent with many atmospheric requirements. Their output per unit wind speed is a function of angle of attack, such that when used as sensors of the vertical or horizontal cross-wind components in the atmosphere, calibration factors may differ by as much as 30 % from those obtained in a normal wind-tunnel calibration procedure (in which wind velocity is parallel to the anemometer shaft). These characteristics are sufficiently important that great care should be taken in using these devices inu-v-w orthogonal arrays. For use in eddy-correlation equipment, it appears that it is best to vane-mount the horizontal sensor to ensure that the appropriate calibration factor is employed. The response lengths of propeller anemometers also vary with angle of attack. Nearθ=0 °, the axially-referred response length appears to depend linearly on cosθ, but nearθ=90 ° a dependence on cos1/2 θ fits the data. No strong effect of wind speed is found. Due to their limited response characteristics, these anemometers give rise to underestimates of the Reynolds stress measured near the surface. The extent of the loss is about 8 % when anemometers in good condition are employed at a height of 5m. Operation at a greater height would allow this error to be reduced. After exposure in the atmosphere for some time, the anemometers tend to respond more slowly and greater losses (of the order 25 %) can occur. Some improvement in performance is possible by the choice of a suitable spatial separation of the sensors.
Type of Medium:
Electronic Resource
URL:
http://dx.doi.org/10.1007/BF02033920
