Estimating submarine melting around the Greenland Ice Sheet

Submarine melting of ice in contact with the ocean at the calving fronts of tidewater glaciers is thought to be a key link in the climate system, coupling the atmosphere and ocean to the dynamics of the Greenland Ice Sheet. With increasing interest in coupling ice and climate models, necessarily at coarse resolution, simple parameterizations for submarine melting are highly desirable. Submarine melting is a complex process driven both by plumes and by wider fjord circulation, responding to variability in subglacial runoff, ocean conditions, fjord/glacier geometry and subglacial hydrology, and current parameterizations do not account for all of these factors.

Here we present a parameterization which for the first time accounts for melting both inside and outside of plumes, and for the first time incorporates all of the principal factors known to modulate submarine melting. We validate the form of the parameterization with a suite of numerical ocean simulations. We then apply our parameterization to tidewater glaciers around the ice sheet with input data from climate reanalysis, local and regional oceanographic data and fjord/glacier bathymetry, to provide a first ice-sheet wide assessment of submarine melting.

Key themes which emerge are (i) in-plume and out-of-plume contributions to total mass loss are comparable, so that out-of-plume melting cannot be neglected, (ii) there is substantial glacier-to-glacier variability in the fraction of ice flux removed by submarine melting, driven by variability in ice velocity and regional climate, and (iii) the fraction of the ice flux removed by submarine melting at Greenland's largest glaciers may be small, indicating that if submarine melting is dynamically important at the largest glaciers, it is due to shaping of the calving front and the promotion of calving rather than total mass removal. We also highlight the large uncertainty associated with estimates of submarine melting, such that spatial and temporal variability in submarine melting should be trusted more than absolute values.