The effects of transverse bed topography variations in ice-flow models

A suite of ice-flow models is subject to a performance test designed to investigate treatment of flow over variable basal topography. Using transfer functions developed by Gudmundsson (2003) and numerical models of various complexity, this study demonstrates that a widely used flowband model has strong limitations associated with its underlying assumptions, and thus should be applied only in specific geometrical settings. Its performance can be significantly improved by using the laterally averaged basal topography instead of centerline topography. In applications where spatial variability of flow fields is important, hybrid ice-flow models can be a viable alternative to flowband models. In addition, analysis of horizontal distributions of the various ice-flow characteristics (e. g., surface elevation, velocity, and horizontal stress components) shows that field observations that focus on assessing these parameters only along the centerline of ice flow could be misleading. These results also suggest that such spatial variability needs to be taken into account when designing field surveys of ice flow.