Inferring basal melt rates underneath the Ross Ice Shelf using data assimilation

More than 80% of the grounded ice of the Antarctic ice sheet drains into the ocean through ice shelves. Loss of these ice shelves could cause an increase of the discharge of grounded ice which would lead to additional sea-level rise. Roughly half of the ice shelves’ mass is eventually lost through melting from the underside, where the ice gets in contact with warmer ocean waters. However, estimating these basal melt rates is notoriously difficult. Here, we present a novel methodology for calculating the melt rates by assimilating remotely derived estimates of surface velocities, ice sheet thickness, surface elevation changes and modelled surface mass balance using an ice sheet model (Úa). This methodology allows for a less noisy, physically consistent estimate of the ice mass divergence, and weighs each of the input data with their uncertainty. As a case study, we apply our method to the Ross ice shelf and find that the melt rates are highly spatially variable.