Vertical water velocities from underwater gliders

The underwater glider is set to become an important platform for oceanographers to gather data within oceans. Gliders are usually equipped with a conductivity/temperature/depth (CTD) sensor, but a wide range of other sensors has been fitted to gliders. In the present work, we aim at measuring the vertical water velocity. The vertical water velocity is obtained by subtracting the vertical glider velocity relative to the water from the vertical glider velocity relative to the water surface. The latter is obtained from the pressure sensor. For the former, a quasi-static model of planar glider flight is developed. The model requires three calibration parameters, the (parasite) drag coefficient, glider volume (at atmospheric pressure) and hull compressibility, which are found by minimising a cost-function based on the variance of the calculated vertical water velocity. Vertical water velocities have been calculated from data gathered in the Northwestern Mediterranean during the Gulf of Lions Experiment, Winter 2008. Although no direct comparison could be made with water velocities from an independent measurement technique, we show that for two different heat loss regimes ( 0 and 400 W/m2), the calculated vertical velocity scales are comparable with those expected for internal waves and active open ocean convection, respectively. High noise levels due to the pressure sensor require the water velocity times series to be low-pass filtered with a cut-off period of 80 seconds. The absolute accuracy of the vertical water velocity is estimated at 4 mm/s