Key points are not available for this paper at this time.
An Argos‐tracked surface buoy deployed off Vancouver Island in July 1984 reached Hawaii in July 1985. We present an analysis of the drifter track with particular emphasis on the equatorward leg seaward of the North American coast for the period July–December 1984. A low‐pass filter is used to separate motions into mean and fluctuating components. The mean track was consistently to the right of the direction expected according to the climatological geopotential topography relative to 1000 dbar, and the average southward drift speed of 0.20 m s −1 exceeded the mean geopotential flow speed by a factor of 2–3. On the basis of 1 year of drifter positions we estimate that surface waters complete one circuit of the North Pacific Gyre in 4–5 years. The amplitudes of the mesoscale buoy displacements about the mean path were approximately 43 ± 10 km, and the root‐mean‐square drift speed was 0.32 ± 0.18 m s −1 . Motions at tidal and near‐inertial frequencies were negligible. From August to October the buoy paralleled a major upwelling regime centered off northern California. During this period the track underwent five sinuous zonal perturbations with characteristic amplitudes of approximately 50 km. Two of the perturbations were associated with abrupt temperature drops of 3°C and maximum seaward currents of 0.70 m s −1 . National Oceanic and Atmospheric Administration advanced very high resolution radiometer thermal imagery for these times reveals that the buoy had encountered seaward protruding cold water filaments (“squirts”) associated with 100‐km‐length upwelling zones that constitute the large‐scale upwelling regime. Southward of the upwelling regime the equatorward drift speed of the buoy was only slightly greater than the equatorward advance of the surface isotherms, suggesting that advection rather than winter cooling might be mainly responsible for the shift in the temperature pattern.
Thomson et al. (Mon,) studied this question.