Use of Altimeter and Wind Data to Detect the Anomalous Loss of SVP-Type Drifter's Drogue

The study presented in this paper is motivated by the unexplained time dependency of the Ekman model recently computed by Rio et al. using a dataset of drogued drifting buoy velocities distributed by the Surface Drifter Data Assembly Center (SD-DAC). This suspicious behavior is found to be fully explained by the presence of undrogued buoys in the drogued drifters database (Grodsky et al.). Altimeter and wind data are used to extract the direct wind slippage from the total drifting buoy velocities over the period 1993-2010. The obtained wind slippage values are then analyzed to identify probable undrogued data among the drifting buoy velocities dataset. The proportion of probable undrogued data among the drogued buoy velocities distributed by SD-DAC is found to be 48%, with an important spatial and temporal variability. When averaged over the full 1993-2010 period, the wind slippage affecting these undrogued buoy velocities reaches up to 13 cm s(-1) in the Antarctic Circumpolar Current. A simple procedure is applied to produce an updated dataset, including a drogue presence flag as well as a wind slippage correction. The drogued drifter data from this new dataset are finally used to estimate a new, stable over time Ekman model for the global ocean by latitudinal band and by month.