On sampling strategies and interpolation schemes for satellite-tracked drifters

The effects of reduced sampling schedules (duty cycles) on velocity statist ics derived from satellite-tracked drifters in the northeast Pacific Ocean are investigated. Continuous segments of the drifter records tin which all available satellite positions fixes are recorded and processed by Service A RGOS) are degraded to match the standard duty cycle used in the World Ocean Circulation Experiment-Surface Velocity Program, in which there are 48 h o f no data transmission followed by 24 h of received transmission (48-24 h). Also examined are duty cycles of 32-16 h and 16-8 h. It is found that the strong inertial motions prevalent in the drifter records result in signific antly biased statistics derived from the degraded series. Reproduction of t he original prime (mean and standard deviation) and rotary spectral statist ics requires an interpolation that takes into account the oscillatory compo nent of the drifter motions. Duty cycles having shorter but more frequent g aps (e.g., 16-8 h) are not sufficient to resolve the main features of the f low. The authors recommend that interpolations over duty cycle segments of drifter records be customized to account for the dominant modes of variabil ity observed in available continuous segments.