cjvogl
commented
7 years ago The more I think about this, the more I'm ok with it. There is still continuity of flux as the flux out of one cell is then divided among sedimentation and evaporation... however, this should still be treated as a sink instead of something that happens at cell interfaces.
The use of "dum1" for cloud ice and "dum" for cloud liquid are meant to incorporate evaporation of these elements outside of a cloud (and should enter back into the equations elsewhere), but cause a discontinuity in the numerical flux across cell interfaces.
A better approach would be to treat this dynamic as a sink and incorporate it accordingly.