Comparison of Aanderaa RCM-9 and MAVS3 Current Meters
Measuring Surface Currents in the Gulf of Maine

James D. Irish and Albert J. Williams, 3rd

Applied Ocean Physics and Engineering
Woods Hole Oceanographic Institution
Woods Hole, MA –02543

A Nobska MAVS3 3 axis and an Aanderaa RCM-9 2 axis current meter were deployed for comparison just below the Jordan Basin mooring buoy in the Gulf of Maine. In addition to providing surface current information needed by the Gulf of Maine Ocean Observing System (GOMOOS), the comparison is valuable for evaluation of the RCM-9. The RCM-9 current meter measures the Doppler frequency shift remote from the flow disturbance of the mooring and buoy. However, it requires sufficient scatterers to provide returns for processing, and uses only the “up-Doppler” shifted signal to prevent wake effects from contaminating the observations. It is not clear how this technique affects the measurement of a mean velocity in an oscillating wave field. The MAVS3 is a travel-time measurement of three components of velocity in a small volume directly under the buoy. The travel-time observations should not have any difficulty with wave oscillations as the MAVS3 samples rapidly enough to average out the waves leaving the mean currents. However, the travel-time technique may lose data due to reflection/refraction due to bubble presence in the surface layer. Zero offset drift could bias the measurement with exposure to marine growth or to changes in temperature

Hourly telemetered data from June through October 2002 were used in a preliminary comparison until the entire mooring with the full data record will be recovered in October. The current meters both sampled at 1 Hz for 1 minute each hour (dictated by limited Aanderaa sampling options). A visual comparison of time series indicated that both current meters observed similar currents. The Aanderaa was mounted 1.7 m below the surface and the MAVS 2.7 m. A regression of the speeds showed that the Aanderaa readings were about 3 cm/sec higher on the average than the MAVS. The slope of the regression showed the Aanderaa speeds about 3-4% higher than the MAVS. This might be expected if the velocity were wind driven and decayed with depth. There is a visual correlation between the wind speed and current meter difference in speed implying this may be the cause of the sensitivity difference. The standard deviation of the difference was 5-6 cm/sec and the regression scatter appeared to be larger than would normally be expected in deep water current meter comparisons. This may be in part due to the difficulty in measuring the currents in the surface layer. This may also be due to the short 1-minute burst sampling. The directions also agree reasonably. The mean difference is 4.8º, and the standard deviation of the direction is 27.1º where MAVS components were used to calculate the MAVS direction, and all 360-degree difference points were corrected for wraparound.

In October the mooring will be recovered, the full data recovered and compared, and these initial results will be updated with double the data points and accompanying data (tilts, etc.) As the current meters are both providing realistic currents, and the differences will be studied in detail to see which would be best in this application.