Implementation of a bathymetric constraint in LADDIE and its evaluation with in-situ and remotely sensed data

Abstract

Accurate modelling of basal melting beneath Antarctic ice shelves is critical for predicting future ice sheet evolution and global sea-level rise. As current ice sheet models often rely on simplified basal melt parameterisations, LADDIE- a two-dimensional ice-ocean interaction model capable of resolving high resolution melt rates at low computational cost- offers a promising approach to bridge the gap between low-resolution ocean models and high-resolution ice sheet models. This study applies LADDIE to the Totten Ice Shelf (TIS). This drains the Aurora Subglacial Basin in East Antarctica, and is particularly sensitive to oceanic thermal forcing, exhibiting high basal melt and dynamic thinning (Rignot et al., 2013, Liu et al., 2015). Further, a novel bathymetric scaling based on water column thickness is implemented using two bathymetric datasets: BedMachine v3 from Morlighem et al. (2020) (BM) and the updated dataset from Vaňková et al. (2023) (V23). Model results are evaluated against in-situ autonomous phase sensitive radar (ApRES) observations and remotely sensed TIS sub-shelf melt rate products. The bathy metric scaling does not reproduce the sub-shelf melt asymmetry in the grounding zone detected by the ApRES measurements presented in Vaňková et al. (2023). Nevertheless, the fit of the model output to the ApRES data improves when BM is employed, in contrast to to when V23 is used. The results highlight the potential of bathymetric constraints to improve the physical realism of basal melt parameterisations. For a more robust implementation of LADDIE, including a bathymetric constraint, in ice sheet models, more advanced oceanic forcing and more temporally matched observations of ice shelf geometry and melt rates are required