1deg grid has oscillations in dyu

[aekiss]

dyu alternates between slightly smaller and slightly larger values as a function of j in /g/data/ik11/grids/ocean_grid_10.nc. This is also evident in dy in the supergrid, e.g. /g/data/ik11/inputs/access-om2/input_20201102/mom_1deg/ocean_hgrid.nc.

This alternation is most obvious in the equatorial region but also seems to be evident in the tripolar cap, so it may be present everywhere but hidden by larger gradients in other places.

dyu looks fine at 0.25deg, and dyt, dxu, dxt all look ok at both 1deg and 0.25deg.

• is this cause for concern?
• is it something to do with how the 1deg equatorial grid refinement was calculated?

[aekiss]

Possibly related: #279