To investigate technical issues associated with the particle-tracking numerical models frequently used to reproduce the behavior of objects drifting in the actual ocean, the trajectories of satellite-tracked drifters released in 2003, 2004, and 2007 were reproduced using a numerical model. In particular, the wind stress driving the surface currents which carried the drifters has been computed using satellite-observed QuikSCAT/Seawinds data provided twice daily in conjunction with in-situ Ieodo-station wind data. Although it is difficult to reproduce the trajectory of a single drifter using numerical models because of the uncertainty induced by random-walk processes, the similarity between the modeled particle and observed buoy trajectories is statistically significant, except for the experiment in 2007. In general, the satellite-derived wind field modified using in situ data is likely to be able to reproduce observed drifter motion. However, it is found that the model is unable to reproduce drifter trajectories in windy 2007. The numerical modeling result demonstrates that wind-induced leeway drift prevails in drifter motion in 2007, in spite of the wind-resistant drogue attached to the drifters, and that this drift shows non-negligible spatiotemporal variability, suggesting that leeway drift is not simply proportional to wind speeds, as in previous studies have maintained.
All Science Journal Classification (ASJC) codes