I’m trying to model how a crater in ice viscously relaxes. I expect for the floor of the crater to rise and velocities to be less than a micron/yr. However, this behavior is not seen because the entire domain is falling with vertical velocity on the order of cm/yr. I think this is because I am essentially starting gravity suddenly in the whole domain. What I really want to model is what happens to a crater that forms in this ice that has already been experiencing gravity and is in equilibrium.
To do this, I have tried making two components, one with a block of ice without a crater and another with the block of ice with a crater. Then I create two time dependent steps in my study. The first one solves component 1 without the crater and the second solves component 2 with the crater and I set initial values of dependent variables to be from the last time step of the step 1 solution.
However, this doesn’t seem to have any effect on the solution with the crater. I’d appreciate any insight on what I could be doing wrong.