Closure to “Turbulence Measurements with Acoustic Doppler Velocimeters” by Carlos M. García, Mariano I. Cantero, Yarko Niño, and Marcelo H. García

The contribution was a timely notice that acoustic Doppler velocimeters have intrinsic weaknesses and that their signal outputs are not always true turbulence measurements. Herein it is demonstrated that, in steady open channel flows, the sampling record must be larger than 5,000 samples to yield minimum errors on first and second statistical moments of the velocity components. Significantly longer records (more than 50,000 samples) are required for accurate determination of higher statistical moments (e.g. skewness, kurtosis), Reynolds stresses and triple correlations. Further ADV signal outputs are adversely affected by the proximity of solid boundaries, particularly when the sampling volume is located less than 30 to 45 mm from the wall. Recent field observations in a small estuary showed also that ADV records may be affected by various disturbances including wildlife and man-made interferences. Comparative analyses of long duration, high-frequency data sets highlighted the needs for advanced post-processing techniques. It is hoped that the authors' contribution and the present discussion will stress enough the needs to educate and adequately train technicians, engineers, scientists and researchers deploying ADVs in the field, including portable ADV systems.