The predictive power of ice sheet models and the regional sensitivity of ice loss to basal sliding parameterisations: a case study of Pine Island and Thwaites glaciers, West Antarctica

Abstract

Ice sheet models use a wide range of sliding laws to define a relationship between ice velocity and basal drag, generally comprising some combination of a Weertman-style power law and Coulomb friction. The exact nature of basal sliding is not known from observational data, making assessment of the suitability of different sliding laws difficult. The question of how much this choice could affect predictions of future ice sheet evolution is an important one. Here we conduct a model study of a large sector of the West Antarctic Ice Sheet (WAIS), a particularly critical component of the cryosphere, using a range of sliding parameterisations, and we provide an assessment of the sensitivity of ice loss to the choice of sliding law. We show that, after initialisation, various sliding laws result in broadly similar ranges of sea level contribution over 100 years, with the range primarily dependent on exact parameter values used in each sliding law. Comparing mass loss from Thwaites and Pine Island glaciers and the neighbouring regions reveals significant qualitative geographical differences in the relationship between sliding parameters and the modelled response to changes in forcing. We show that the responses do not necessarily follow universal systematic patterns, and, in particular, higher values of the sliding exponent m do not necessarily imply larger rates of mass loss. Despite differences in the magnitudes of ice loss and rates of change in the system, all of our experiments display broad similarities in behaviour which serve to reinforce the decade-to-century-scale predictive power of ice sheet models, regardless of the choice of basal sliding.