Comparison of buoy-mounted 74-kHz acoustic Doppler current profilers with vector-measuring current meters

In December 1991, the National Data Buoy Center (NDBC) deployed two meteorological buoys in the Southern California Bight on a transect between San Diego and San Clemente Island. Each buoy consisted of a 10-m discus hull instrumented to measure a suite of meteorological parameters, and, for the first time in the NDBC buoy program, acoustic Doppler current profilers (ADCPs) were included to gather hourly current profiles beneath the two buoys. Moorings instrumented with seven vector-measuring current meters (VMCMs) were deployed adjacent to the NDBC buoys for several months and provided current observations for comparison with the ADCP measurements. When wave-induced buoy motion is not overly large, the observations of horizontal current made by the ADCP and the VMCM are highly correlated. Time series of differences between ADCP and VMCM measurements are characterized by a mean difference (bias error) of about 0.01 m/s and standard deviation of about 0.035 m/s for 1-h observations. Estimates of current spectra from ADCP and VMCM records suggest that the ADCP system can be characterized by a white noise level of 2 x 10(exp -3) sq m/sq s/cph. However, when the in situ environment is such that large surface waves are present (including breaking waves and whitecaps), erroneous current values are usually reported by the ADCP. Mean values of vertical velocities reported by the ADCP appear to be much larger than what could be physically expected and are therefore deemed unreliable.