Numerical simulations of the ice flow dynamics of the Brunt Ice Shelf - Stancomb Wills Ice Tongue System

Ice shelves play an important role in determining regional ocean properties and in modulating ice flux from \nland to sea. Their dynamics are complex, however, and localised rifts and zones of weakness can have a \nsignificant but poorly understood effect on flow and ultimately on the integrity of the shelf. \nThe Brunt Ice Shelf (BIS)- Stancomb Wills Ice Tongue (SWIT) System, situated on the Caird Coast, Oates \nLand, Antarctica, is characterised as a thin, unbounded ice shelf with a highly heterogeneous structure. In \ncontrast to most ice shelves, icebergs calve along much of the grounding line but are trapped and subsequently \nbound together by sea ice. This calf-ice / sea-ice aggregate makes up a large part of the Brunt Ice Shelf in \nparticular, and this heterogeneity makes the BIS-SWIT a good test case for investigating the importance of \nweak zones in shelf dynamics. \nWe applied a diagnostic, dynamic/thermodynamic ice-shelf model to simulate the present flow of the ice \nshelf that results from the ice-thickness distribution, the influx at the grounding line and the surface and \nbottom temperature. We then compared the model results with flow velocities measured by Synthetic \nAperture Radar feature tracking. We found that our simulations were clearly improved by the use of a \nhigh- resolution ice thickness distribution on the heterogeneous ice shelf calculated from ICESat surface \nelevation data using an assumption of hydrostatic equilibrium. We then assessed the model’s sensitivity to \nice thickness, inflow velocities and a flow enhancement factor that parameterises the role of sea ice, whose \nmechanical properties are known to be significantly different from those of meteoric ice. \nWe found that the numerical simulations were improved by incorporating the detailed variations in shelf \nstructure. Simulated flow velocities on either side of rifts in the ice shelf became decoupled as we softened \nthe sea ice within the rifts. On a larger scale, we found that soft sea ice can lead to a decoupling of the \nmovement of the Stancomb-Wills Ice Tongue and the Brunt Ice Shelf. When we simulated a regime where \nsea ice was absent, ice shelf flow speeds increased along the western edge of the SWIT ice front, in general \nagreement with observations made in just such a sea- ice-free dynamic regime that occurred in