Runoff redistribution

[aekiss]

Splitting off from #208 (comment):

1. do we need river runoff redistribution horizontally or vertically (rivermix)?
2. what's done if the JRA55do runoff is located on land on the model grid - is it teleported to the nearest wet cell?
3. do the area-integrated friver and ficeberg MOM6 diagnostics match the area-integrated fluxes in the RYF.friver.1990_1991.nc and RYF.licalvf.1990_1991.nc JRA55do source files? (may require different remapping and time interpolation - see River and iceberg runoff mapped to incorrect locations #208 (comment))

[aekiss]

Related: ACCESS-NRI/dev_coupling#32

[anton-seaice]

@kieranricardo reports you can't use the vertical mixing (RIVERMIX) with USE_REGRIDDING=True (which I think is important / fundamental)

[aekiss]

Thanks, if that's the case, we can't use RIVERMIX

[aekiss]

GFDL OM5 doesn't use rivermix

[aekiss]

ACCESS-OM2 used YATM to do online horizontal redistribution of runoff (sec 3.5.5. in the tech report).

I'm not sure what was done in ACCESS-CM2 but I gather it used a pre-calculated redistribution, perhaps created with https://github.com/COSIMA/runoff_tools