Search Results

You are looking at 1 - 2 of 2 items for :

  • Author or Editor: R. F. Marsden x
  • Journal of Atmospheric and Oceanic Technology x
  • Refine by Access: All Content x
Clear All Modify Search
R. F. Marsden
and
Y. Gratton

Abstract

Minor alignment errors (∼1°) can contaminate acoustic Doppler current profiler measurements of vertical velocities taken from a ship traveling at speed. The signature is consistently large positive or negative vertical velocities, recorded at all depths. A technique is proposed to postcorrect contaminated data in cases of tidally dominated flow over a flat bottom. The eigenmodes of the cross-correlation matrix of the depth-averaged velocities are found, and the data are rotated to align along the eigenvector corresponding to the smallest eigenvalue that is equivalent to the physical vertical direction. An example of corrupted data is presented, and corrections for the pitch and roll directions are found. The corrected flow field is shown to be physically plausible over both flat and sloping bottom sections of the cruise track.

Full access
R. F. Marsden
and
R. G. Ingram

Abstract

Spatial homogeneity assumptions inherent in the conversion of directly measured acoustic Doppler current profiler (ADCP) beam to Cartesian coordinates for the Janus configuration are investigated. These assumptions may be adequate for large-scale flows, such as tidal currents and wind-forced upwelling. However, for high-frequency features, such as internal solitons and turbulence, the velocity fields may vary over scales comparable to the divergence of the acoustic beams. Equations are derived for beam spreading, and it is shown that a first-order correction can be applied to improve velocity measurement accuracy. Two cases are examined. First, the effects of the spatial and temporal convolution inherent in beam spreading from the Janus configuration ADCP are applied to a model internal solitary wave. It is shown that the corrected vertical velocities have deviations of less than 2 mm s−1 for distances up to 30 m from the transducer face and are approximately 3 times more accurate than the uncorrected velocities for distance up to 20 m from the transducer face. Next, under a “frozen turbulence” hypothesis, the method is applied to processing turbulence data. It is demonstrated that the horizontal longitudinal velocity can be markedly improved.

Full access