The effects of bed topography and strength on stability of marine ice sheets

The "marine ice-sheet instability hypothesis", which states that unconfined marine ice sheets are unconditionally unstable on retrograde slopes, was developed under assumptions of negligible bed slopes. Realistic ice sheets, however, flow over beds which topographies have a wide range of bed slopes (for example, Thwaites Glacier in the Amundsen Sea sector, West Antarctica). Reexamining the original model of marine ice sheets proposed by Schoof (2007), and relaxing an assumption of negligible bed slopes, we find that a steady-state ice flux at the grounding line is an implicit function of the grounding-line ice thickness, bed slope and accumulation rate. Depending on the sliding conditions, the magnitudes of the ice flux at the grounding line differ by one-to-three orders of magnitudes from that computed with a power-law expression derived by Schoof (2007) under assumptions of the negligible bed slopes. Non-negligible bed slopes also result in conditions of stability of the grounding line that are significantly more complex than those associated with the "marine ice sheet instability hypothesis". Bed slopes are no longer the sole determinant of whether the grounding line is stable or unstable. We find that the grounding line can be stable on beds with retrograde slopes and unstable on beds with prograde slopes.