I'm trying to model the flow of an ice deposit on top of a fixed rocky unit using solid mechanics creep (attached). The base of the ice is fixed to the rock and material above the base moves outward, extending the boundary between the domains and making the scarp at the end shallower. If the model is run long enough (>2800 yr) the ice domain deforms to the point where it goes beneath the surface of the rocky domain (see attached figure). Clearly, this isn't realistic. How can this be prevented? Can you set the boundary between the two domains to always be fixed even as the boundary extends? I've thought of using a contact pair, but I'm not sure that this is the correct approach since the original base of the ice isn't ever moving with respect to the rock (I also haven't been able to get a solution using contact pairs yet).
↧