Table
1: RMS of the differences for the zonal component U, the meridional
component V, the current modulus and the direction of the current, of
the independent drifting buoy velocities and the altimeter geostrophic
velocities obtained using MDT solutions CNES-CLS22, CNES-CLS18 and the
CNES-CLS22 first guess. The last two lines show the reduction/increase
in RMS of the differences between two solutions: CNES-CLS22 versus
CNES-CLS18 and CNES-CLS22 versus CNES-CLS22 first guess, in percent (a
negative percentage is a decrease in the RMS of the differences). These
statistics were generated using all available drifters up to 70°N.
Given the small number of independent drifters in Arctic, this area is
treated separately. Table 2 summarizes the RMS of the differences
between the geostrophic currents derived from drifters (degraded
treatment) and the geostrophic currents derived from ADTs calculated
with the CNES-CLS22, CNES-CLS18 and background CNES-CLS22 solutions. The
last two lines show in percent the reduction in RMS of the differences
between CNES-CLS22 and CNES-CLS18 MDT and its first guess. In the zonal
and meridional components of the current, the new solution reduces the
RMS of the differences compared with the drifters by 19% and 12%
respectively. This improvement translates into a clear improvement in
direction of 9%, which reduces the RMS of the differences from around
56° to 50°; but the RMS of the differences in current amplitude remains
very slightly better for CNEs-CLS18. Note that, as the coverage is not
quite identical, there are more points taken into account for CNES-CLS22
than for CNES-CLS18.