[MITgcm-devel] Cube diagnostics

[Chris Hill]

JM etc...,

 I was trying to decide whether it made sense to include an online
conservative diagnostic option for the cube which interpolates back to
regular lat-lon coords and, for the flow, does an elliptic solve to ensure
div_3d(U) = 0. What do you think, we have the technology to do this and
because diags are not every timestep computational cost is less of an issue?
I'm sure we would screw up some conservation properties but it might be a
nice feature for some work. We would still put out the native fields too as
these are needed for offline, term balances etc...

Chris