Twenty-first century ocean forcing of the Greenland ice sheet for modelling of sea level contribution

Abstract

Abstract. Changes in ocean temperature and salinity are expected to be an important determinant of the Greenland ice sheet's future sea level contribution. Yet, simulating the impact of these changes in continental-scale ice sheet models remains challenging due to the small scale of key physics, such as fjord circulation and plume dynamics, and poor understanding of critical processes, such as calving and submarine melting. Here we present the ocean forcing strategy for Greenland ice sheet models taking part in the Ice Sheet Model Intercomparison Project for CMIP6 (ISMIP6), the primary community effort to provide 21st century sea level projections for the Intergovernmental Panel on Climate Change Sixth Assessment Report. Beginning from global atmosphere–ocean general circulation models, we describe two complementary approaches to provide ocean boundary conditions for Greenland ice sheet models, termed the “retreat” and “submarine melt” implementations. The retreat implementation parameterises glacier retreat as a function of projected subglacial discharge and ocean thermal forcing, is designed to be implementable by all ice sheet models and results in retreat of around 1 and 15 km by 2100 in RCP2.6 and 8.5 scenarios, respectively. The submarine melt implementation provides estimated submarine melting only, leaving the ice sheet model to solve for the resulting calving and glacier retreat and suggests submarine melt rates will change little under RCP2.6 but will approximately triple by 2100 under RCP8.5. Both implementations have necessarily made use of simplifying assumptions and poorly constrained parameterisations and, as such, further research on submarine melting, calving and fjord–shelf exchange should remain a priority. Nevertheless, the presented framework will allow an ensemble of Greenland ice sheet models to be systematically and consistently forced by the ocean for the first time and should result in a significant improvement in projections of the Greenland ice sheet's contribution to future sea level change.

The future sea-level contribution of the Greenland ice sheet: a multi-model ensemble study of ISMIP6

“…Compared to earlier ad hoc approaches (e.g. Goelzer et al, 2013;Fürst et al, 2015;Calov et al, 2015;Beckmann et al, 2019), the advantage of the technique used (Slater et al, 2019(Slater et al, , 2020 is its empirically based and transparent implementation. Nevertheless, large uncertainties are attached to this part of the projections and leave room for considerable improvements in the future.…”

“…ometry using a technique developed specifically for that purpose (Goelzer et al, 2020b). The standard approach for ocean forcing is based on an empirically derived retreat parameterization for tidewater glaciers (Slater et al, 2019(Slater et al, , 2020 that is forced by MAR run-off and ocean temperature changes in seven drainage basins around Greenland. The forcing is illustrated as Greenland-wide average of prescribed tidewater glacier retreat in Fig.…”

“…The retreat parameterization is designed to be used in the wide variety of models under consideration. Uncertainty in the parameterization is translated into a set of three ocean sensitivities (medium, high, low) covering the median, 75 % and 25 % percentiles of sensitivity parameter κ that controls the amount of retreat given ocean temperature change and ice sheet run-off (Slater et al, 2019(Slater et al, , 2020. Results are explored with the last two core experiments (Table 3).…”

“…For ensemble statistics we report mean (µ) and the 2× standard deviation (2σ ) range to quantify the uncertainty unless stated differently in the text. (Slater et al, 2019(Slater et al, , 2020. The shading gives the range of ocean sensitivity sampled with two more experiments in MIROC5-RCP8.5-high and MIROC5-RCP8.5-low.…”

Experimental protocol for sea level projections from ISMIP6 stand-alone ice sheet models

“…The ISMIP6 Greenland ocean focus group (Slater et al, 2019(Slater et al, , 2020 proposed two methods for implementing oceanic forcing: the "retreat implementation" and the "submarine melt implementation." The two approaches are needed in order to maximize participation from current stateof-the-art ice sheet models.…”