Improving solid ice runoff to ocean

Motivtation / Agenda

runs with coupled model have crashed due to negative salinities
this problem was traced back to very large solid runoff fluxes around Antarctica
in turn, this was found to be due to the runoff map for Antarctica being incorrect
we want to improve on this so that the snowcapping scheme works right for Antarctic
an image to give an idea of the actual ice flux (observed) is included below
one possible option to consider would be to supply the runoff routing scheme with routing directions based on the observed or modeled ice flow

ice / ocean team will provide river runoff team w/ ice velocity (vectors) for routing of solid runoff (Stephen Price)
vectors to be provide on lat / lon coords.
consider interpolating directly from MPAS-LI grid to 1/2 deg. runoff grid (need scrip file for the latter)
land ice folks need to figure out how best to provide velocity component information (e.g., because the components will be in x,y polar stereo)
liquid runoff CAN remain as is (using river channel network)
to do for runoff team: need ocean domain for matching up of coastline cells (this will be specific to ocean grid)