An investigation of the accuracy of current profile measurements from a glider-mounted ADCP operating in shallow water

A payload section has been developed to mount an Acoustic Doppler Current Profiler (ADCP) with the ability to “bottom track” (directly measure speed over ground when in range of the bottom) to a Slocum glider. Gliders are designed to profile the water column through buoyancy adjustments, and are trimmed to move through the water by shifting a battery pack forward to pitch down during descent and aft to pitch up during ascent. The pitches are significant, nominally ±26° for the Slocum glider as it operates in its sawtooth profiling mode. However, this particular glider was also one of the first Slocum gliders equipped with a propeller for operation while trimmed for level flight. The ADCP used here has four acoustic beams fixed at 90° azimuth and 30° off the ADCP centerline and, for various reasons, the ADCP operates optimally for both current measurement and for bottom tracking when it is oriented vertically. The discussion that led to the decision to build the payload section with the ADCP mounted with beam pairs pointed directly fore and aft (and port and starboard) and physically pitched forward 11° to optimize for current profiling and bottom tracking on descent yet maintaining acceptable symmetry for those measurements during level fight is presented. However, this geometry means that the ADCP is not actually vertical during either descent or in level flight, so an experiment was designed to test the accuracy of the current profiles and of the bottom tracking when the glider was oriented at the nominal descent angle and for level flight. Because co-location of measurement is always a problem when doing current measurement comparisons, a lobster boat was outfitted with mounts for the glider and for a Workhorse ADCP. The Workhorse ADCP also has four beams at 90° azimuth but its beams are oriented at 20° off the centerline. It was mounted to look downward and configured to gather both bottom track and current profiles as the reference measurements for the experiment. The glider with the ADCP payload section was also fix-mounted to the boat, but with a mount that allowed it to be oriented at the nominal 26° descent angle or for level flight. The experiment was originally intended for Cape Cod Bay, but heavy seas forced us to stay in Cape Cod Canal (for fear of applying to much heave-associated forcing to the glider). Cape Cod Canal has significant, though variable, currents and the lobster boat was driven at speeds that were predominately ~1 m/s and ~3 m/s during the experiment. The accuracy of the glider ADCP is shown to be excellent in level flight. In tilted flight there was some disagreement at higher speeds (that is, greater than 1.5 m/s), but there are extenuating circumstances that, if are convincing enough to allow their removal, the remainder of the tilted flight glider ADCP data also has excellent accuracy. Establishing the accuracy of the bottom tracking allows removal of the platform motion with confidence when the bottom is within range of the ADCP. This is the first step to testing of algorithms designed to remove the platform motion when the bottom is out of range, because those algorithms can now be tested by using them to remove the platform motion when the bottom is in range (but bottom tracking not used) to that same data when the platform motion is removed by using the bottom track measurements.