The influence of the sensor design on wind measurements with sonic anemometer systems

Responses of Kaijo Denki TR-61A, TR-61B, and TR-61C; Solent Research/Gill; and METEK USA-1 sonic anemometer systems have been examined in a wind tunne l investigation. To determine their characteristics the anemometers were tu rned for 360 degrees and tilted for up to +/-8 degrees. With a small Pitot tube the modification of the wind field by a Kaijo Denki TR-61B sensor is e xamined within its measuring volume. Extra measurements were executed to an alyze the influence of turbulent wakes behind sensor parts windward of the measuring volume. Measurements from the Echival Field Experiment in Deserti fication Threatened Areas of 1994 are used to compare the measurements of K aijo Denki TR-61C, Solent Research/Gill, and METEK USA-1 sonic anemometers in the atmospheric boundary layer. Struts and transducers are leading to de creased mean wind velocity, deviation, and higher variances depending on th e probe geometry and dimension. Best results can be achieved with the Solen t Research and Kaijo Denki TR-61B sensors. The Solent Research/Gill calibra tion procedure improves the mean horizontal wind velocity and direction sig nificantly, but it should be used with caution because it increases varianc es especially at incoming flow directions where sensor-induced turbulence i s at its highest. The TR-61C is still a usable instrument for the measureme nt of turbulent fluxes as long as the vertical sound path is not placed in the turbulent wake of the sensor foot. The direction characteristic of the TR-61A reduces its operational range but supplies most precise vertical win d velocity measurements. The METEK USA-1 has an interesting sensor geometry and user interface but needs further improvements in its electronics.