Simulating the internal structure of the Antarctic Ice Sheet – towards a spatio-temporal calibration for ice-sheet modelling

Abstract. Ice Sheet Models are a powerful tool to project the evolution of the Greenland and Antarctic Ice Sheets, and thus their future contribution to global sea-level changes. Probing the fitness of ice-sheet models to reproduce ongoing and past changes of the Greenland and Antarctic ice cover is a fundamental part of every modelling effort. However, benchmarking ice-sheet model data against real-world observations is a non-trivial process, as observational data comes with spatio-temporal gaps in coverage. Here, we present a new approach to assess the ability of ice-sheet models which makes use of the internal layering of the Antarctic Ice Sheet. We simulate observed isochrone elevations within the Antarctic Ice Sheet via passive Lagrangian tracers, highlighting that a good fit of the model to two dimensional datasets does not guarantee a good match against the three dimensional architecture of the ice-sheet and thus correct evolution over time. We show, that paleoclimate forcing schemes commonly used to drive ice-sheet models work well in the interior of the Antarctic Ice Sheet and especially along ice divides, but fail towards the ice-sheet margin. The comparison to isochronal horizons attempted here reveals, that simple heuristics of basal drag can lead to an overestimation of the vertical interior ice sheet flow especially over subglacial basins. Our model-observation intercomparison approach opens a new avenue to the improvement and tuning of current ice-sheet models via a more rigid constraint on model parameterisations and climate forcing which will benefit model-based estimates of future and past ice-sheet changes.