Monthly anomalies are defined relative to a climatological mean which includes a seasonal cycle [1]. These pseudo- Eulerian mean fields are calculated from all quality-controlled surface drifter observations on a 1 degree square grid.
For a particular month of kriged observations [2], drifter velocities are calculated from the drifter displacements and lowpassed at 5 days to remove high-frequency motions such as inertial oscillations and semidiurnal/diurnal tides. The amplitudes of the mean currents and SST are interpolated onto each quarter-day drifter position using standard Optimal Interpolation techniques [3], and subtracted from the total current and SST to generate a time series of the residual along each drifter's trajectory. Anomalous currents are represented in a coordinate system aligned with the mean current: a positive "mean current anomaly" indicates motion in excess of the climatological mean current speed, in the direction of the mean current. A positive "meander anomaly" indicates motion oriented to the left of the mean current.
Anomalies for individual drifters are often dominated by changes at scales smaller than the 1 degree resolution of the climatological means (such as fronts), and by mesoscale features such as eddies. In order to estimate larger-scale, climate-significant changes, the observations from many drifters must be averaged together at regional scales. For the AOML monthly anomaly plots, observations are averaged in bins which are 20 degrees (longitude) by 10 degrees (latitude). Averages are only shown in bins where more than 25 days of drifter observations were collected in the month.
Month-by-month SST anomalies in well-sampled regions such as the Labrador Sea demonstrate a clear relationship with the North Atlantic Oscillation, a well-known interannual climate signal [4]. However, in regions which have been sparsely sampled by the drifters until recently, mean values are poorly resolved and so anomalies may not be robust. The primary example of this is in the eastern tropical Atlantic, which has been seeded by drifters only in the last couple of years. Current anomaly maps indicate where climate signals may be present, but the anomalous speeds may also be biased by strong eddies on the edges of the 10x20 degree bins. More careful analysis, such as [5], is required to demonstrate climatic variations of currents from these observations.
REFERENCES:
[1] Lumpkin, R., 2003: Decomposition of surface drifter
observations in the Atlantic Ocean. Geophys. Res. Letters,
in press.
[2] Hansen, D. and P.-M. Poulain, 1996: Quality control and
interpolations of WOCE-TOGA drifter data. J. Atmos. Ocean.
Tech. 13, 900-909.
[3] Bretherton, F. P., R. E. Davis, and C. B. Fandry, 1976:
A technique for objective analysis and design of oceanographic
experiments applied to MODE-73. Deep-Sea Res. 23, 559-582.
[4] Hurrell, J. W., Y. Kushnir, G. Ottersen and M. Visbek, Eds.,
2003: The North Atlantic Oscillation: Climate Significance and
Environmental Impact. Geophysical Monograph Series 134, 279pp.
[5] Flatau, M., L. Talley and P. Niiler, 2003: The North Atlantic
Oscillation, Surface Current Velocities and SST changes in
the subpolar North Atlantic. J. Climate, in press.