Now that the residual ice issue is (hopefully) solved, we have a request from @rgrumbine to clean up the initial condition files (gx3 in particular, but check the others too).
It's not clear how small would be "small enough", though, since the min concentration and mass parameters can now be chosen in namelist. We could take them all the way up to the big dynamics values we have been using (1.e-3, 1.e-2 respectively), which might be okay since these initial condition files are supposed to only be used for testing.