Abstract
Mesoscale eddies are a dominant source of spatial variability in the
surface ocean and play a major role in the biological marine carbon
cycle. Satellite altimetry is often used to locate and track eddies, but
this approach is rarely validated against in situ observations. Here we
compare measurements of a small (under 25 km radius) mode water
anticyclonic eddy over the Procupine Abyssal Plain using CTD and ADCP
measurements from 3 ships, 2 gliders, 2 profiling floats, and one
Lagrangian float with those derived from sea level anomaly. In situ
estimates of the eddy center were estimated from maps of the thickness
of its central isopycnal layer, from ADCP velocities at a reference
layer, and from the trajectory of the Lagrangian float. These were
compared to three methods using altimetric SLA: one based on maximizing
geostrophic rotation, one based on a constant SLA contour, and one which
maximizes geostrophic velocity speed along the eddy boundary. All
algorithms were used to select CTD profiles that were within the eddy.
The in-situ metrics agreed to 97\%. The altimetry
metrics showed only a small loss of accuracy, giving
$>90$\% agreement with the in situ
results. This suggests that current satellite altimetry is adequate for
understanding the spatial representation of even relatively small
mesoscale eddies.