Abstract
Velocity profile measurements at high spatial and temporal resolution are required for the detailed study of solute and momentum transfer close to the sediment–water interface. Still, not many devices allow such measurements in natural systems. Recently, a bistatic acoustic current profiler has become commercially available that allows the recording of profiles at down to 1-mm resolution with a maximum frequency of 100 Hz and a profile length of 3.5 cm. This study tested the ability to characterize the turbulent flow of this profiler in a laboratory flume and in a run of the river reservoir. The tests showed that average velocities were reliably measured in the upper 2.5 cm, while the flow statistics were affected by Doppler noise and signal decorrelation. The latter is caused by the decreasing overlap between the individual beam signals. Doppler noise can be estimated and accounted for by established correction procedures, but currently there is no method to quantify the influence of signal decorrelation. Both error sources mainly affect the measured variances of the velocities, while the Reynolds stresses are reliable as long as there is no interference with the solid bottom. In the field application, most problems arise because of the necessity of coordinate system rotation, since a perfect alignment of the profiler with the current is not possible. Also, because of the coordinate system rotation, the Reynolds stresses become contaminated by noise, which can be removed by low-pass filtering. Still, this filtering results in loss of the turbulent signal, which was estimated in this study to be between 2% and 10%.