Inverse Modelling at Recovery Glacier, Antarctica

Abstract

The future ice loss of Recovery Glacier will probably be the largest of the East Antarctic Ice Sheet over the next millennia. Its evolution can be predicted by models solving the equations of the momentum and mass balance. Ice dynamics are fundamentally driven by bedrock conditions underneath the ice, but these can not simply be measured yet. This thesis utilizes an inverse method implemented in the Ice Sheet System Model (ISSM) to acquire basal parameters. The technique minimizes the difference between horizontal surface velocities derived from remote sensing and computed by the model. False values in the observations can lead to uncertainties in basal parameters. In order to remove such false values, this thesis presents a new filtering method. Data gaps are filled comparing four different interpolation methods. A sensitivity analysis shows that the influence of filtering outliers and interpolation on basal parameters derived from inverse modelling is large in specific regions. The resulting basal parameters do not indicate the existence of the previously proposed subglacial lakes at the onset of Recovery Glacier.