Mean Surface Circulation of the Global Ocean Inferred from Satellite Altimeter and Drifter Data

Sea level anomaly from satellite altimetry and in situ near-surface drifter velocities, combined within a simplified equation of horizontal momentum, reveal peculiar dynamical balances in the strongest known currents of the World Ocean. In the Gulf Stream, this study demonstrates the importance of centrifugal force due to the curvature of the streamlines. After filtering out Ekman currents (using NCEP winds and improved parameterizations), and by blending the mean dynamic ocean surface topography (MDOT) on large scale using GRACE model, the data provide a global map at mesoscale resolution and unprecedented accuracy. This map, when used to revise the pattern of large scale circulation, reveals a new jet in the South Atlantic. It also reveals for the first time distinct, steady, quasizonal jets in the eastern parts of practically all oceans. The MDOT data is released to public. 1. HORIZONTAL MOMENTUM BALANCE OF THE UPPER OCEAN As shown by [1] drifter velocities can be successfully combined with the Aviso [2] satellite altimetry sea level anomaly maps to quantitatively analyze the balance of horizontal momentum. While many studies (e.g., [3]) assume geostrophy, [4] emphasize the importance of Ekman currents controlled by local wind, and [5] reveal marked contribution from the socalled centrifugal force to the balance of the Kuroshio Current south of Japan. Although [3] found that variability (or eddy kinetic energy, EKE) of nearsurface velocity measured by drifters in the Gulf Stream (GS) is systematically larger/weaker than variability of geostrophic velocity assessed from data of satellite altimeter (Fig.1), the difference was attributed to the difference in processing the two datasets. We suggest that the difference can be due to the real physics, namely, the centrifugal force, unaccounted in [3]. Figure 2 illustrates sea level shape and force balance in the case of round eddies of the same horizontal size but different senses of rotation. Geostrophic response of sea level on the Coriolis force Fcor due to eddy rotation (dashed lines) has same magnitude and different signs Figure 1. Difference between drifterand altimeterderived EKE (reproduction of Plate 8 in [3]). for cyclones (a) and anticyclones (b) of the same kinetic energy. Centrifugal force Fcentr = V/r has same magnitude in both eddies and is always directed outward. In the cyclonic eddy, it adds to Fcor and larger (compared to geostrophy) pressure gradient force Fpress (and, correspondingly, larger depression of the sea surface) is needed to balance the system. On contrary, in anticyclones Fcentr partly compensates Fcor , so that smaller Fpress is needed to close the balance. (In principle, an anticyclone might exist even without any signal in the sea level, but such eddy is known to be unsteady). Therefore, geostrophic velocities based on sea level data overestimate/underestimate actual velocities in cyclonic/anticyclonic eddies and one may expect that geostrophic EKE assessed from sea level will be larger or smaller than the EKE calculated from direct velocity observations in areas dominated by cyclones or anticyclones, respectively.