
For more complete details on the theory of operation of the CSAT3BH, refer to
7.1.1 Algorithm Version 5
Since the release of the original CSAT3 in 1996, various improvements have been made to the
algorithms used for signal processing and measurement output. Each time a significant change
has been made to these algorithms, a new version number has been issued. The CSAT3BH uses
algorithm Version 5. Version 5 maintains many of the advantages of signal recognition and
diagnostic sensitivity that were made possible by Version 3, while also adding the advantages of
performance during precipitation events made possible by Version 4. It also resolves Version 4
issues of speed-of-sound measurement errors in very high wind conditions as reported by Burns
et al. (2012; see
(p. 91)).
7.1.2 Effects of crosswind on the speed of sound
The speed of sound is found by combining the out and back time-of-flight measurements (see
Equation B-5 in
(p. 95)). While the parallel component of the wind
along the sonic axis does not affect the measured speed of sound, the perpendicular component
does. An online calculator can account for the effects of the perpendicular component of wind
using the measured components of wind and simple trigonometry, or manually by using the
methods described by Schotanus, Nieuwstadt, and de Bruin (1983) and Liu , Peter, and Foken
(2001; see
(p. 91)). The CSAT3BH corrects for the effects of crosswind on the speed of
sound.
The equations derived by Schotanus, Nieuwstadt, and de Bruin (1983) apply to sonic
anemometers that make speed of sound measurements from a single pair of transducers. Liu,
Peter, and Foken (2001) extend these equations to sonic anemometers that measure the speed of
sound on all three axes and then average the results to a single speed of sound, as with the
CSAT3BH. Liu, Peter, and Foken (2001) assume that the geometry of each individual three-
dimensional anemometer is ideal when given factors are derived, as found in Table 1 of their
publication.
NOTE:
Liu, Peter, and Foken (2001) recommend that CSAT3BH sonic temperature variances and
sonic sensible heat flux are corrected for the effects of crosswind. The CSAT3BH, however,
performs an online correction. Additional correction of CSAT3BH data for crosswind effects
will cause errors in the measured fluxes.
CSAT3BH Three-Dimensional Heated Sonic Anemometer
62