Dynamic modelling of future glacier changes: mass-balance/elevation feedback in projections for the Vestfonna ice cap, Nordaustlandet, Svalbard

Abstract

ABSTRACT. Future projections of the evolution of ice caps as well as ice sheets and consequent sealevel rise face several methodological challenges, one being the two-way coupling between ice flow and mass-balance models. Full two-way coupling between mass-balance models -or, in a wider scope, climate models -and ice flow models has rarely been implemented due to substantial technical challenges. Here we examine some coupling effects for the Vestfonna ice cap, Nordaustlandet, Svalbard, by analysing the impacts of different coupling intervals on mass-balance and sea-level rise projections. By comparing coupled to traditionally deployed uncoupled strategies, we prove that neglecting the topographic feedbacks in the coupling leads to underestimations of 10-20% in sea-level rise projections on century timescales in our model. As imposed climate scenarios increasingly change mass balance, uncertainties in the unknown evolution of the fast-flowing outlet glaciers decrease in importance due to their deceleration and reduced mass flux as they thin and retreat from the coast. Parameterizing mass-balance adjustment for changes in topography using lapse rates as a cost-effective alternative to full coupling produces satisfactory results for modest climate change scenarios. We introduce a method to estimate the error of the presented partially coupled model with respect to as yet unperformed two-way fully coupled results.

Sensitivity, stability and future evolution of the world's northernmost ice cap, Hans Tausen Iskappe (Greenland)

“…This approach allows for the SMB to be calculated at any time for any geometry. In this study the SMB model is coupled to the thermomechanical ice flow model once a year, avoiding potential problems related to a long coupling interval (see Schäfer et al, 2015). The SMB is also available from the RCM RACMO2.3 (Noël et al, 2015) for a fixed geometry, but it is not directly used as the PDD is more flexible for generating the SMB under an evolving geometry without the need to modify RCM 10 output for a different surface elevation (Franco et al, 2012;Helsen et al, 2012;Edwards et al, 2014).…”

“…The SMB is also available from the RCM RACMO2.3 (Noël et al, 2015) for a fixed geometry, but it is not directly used as the PDD is more flexible for generating the SMB under an evolving geometry without the need to modify RCM 10 output for a different surface elevation (Franco et al, 2012;Helsen et al, 2012;Edwards et al, 2014). The implications of using such correction methods to account for the interaction between surface elevation and SMB were investigated in detail by Schäfer et al (2015) for Vestfonna ice cap (Svalbard), suggesting that such approaches should not be used for simulations of more than several decades and under extreme climate change scenarios.…”

Sensitivity, stability and future evolution of the world's northernmost ice cap, Hans Tausen Iskappe (Greenland)

“…The SMB is also available from the Regional Atmospheric Climate Model (RACMO2.3) (Noël et al, 2015) for a fixed geometry, but it is not directly used as the PDD is more flexible for generating the SMB under an evolving geometry without the need to modify RCM output for a different surface elevation (Franco et al, 2012;Helsen et al, 2012;Edwards et al, 2014). The implications of using such correction methods to account for the interaction between surface elevation and SMB were investigated in detail by Schäfer et al (2015) for the Vestfonna ice cap (Svalbard), suggesting that such approaches should not be used for simulations of more than several decades and under extreme climate change scenarios.…”

“…This approach allows for the SMB to be calculated at any time for any geometry. In this study the SMB model is coupled to the thermomechanical ice flow model once a year, avoiding potential problems related to a long coupling interval (see Schäfer et al, 2015). The SMB is also available from the Regional Atmospheric Climate Model (RACMO2.3) (Noël et al, 2015) for a fixed geometry, but it is not directly used as the PDD is more flexible for generating the SMB under an evolving geometry without the need to modify RCM output for a different surface elevation (Franco et al, 2012;Helsen et al, 2012;Edwards et al, 2014).…”

“…Since then several modelling studies have been performed on individual ice caps in Iceland (Aðalgeirs-dóttir et al, 2005(Aðalgeirs-dóttir et al, , 2006Flowers et al, 2005Flowers et al, , 2007Flowers et al, , 2008 and on Hardangerjøkulen (southern Norway) (Giesen and Oerlemans, 2010;Åkesson et al, 2017) with models based on the shallow-ice approximation (SIA). Schäfer et al (2015) used a full-Stokes ice flow model with an englacial temperature parameterization for a study on the Vestfonna ice cap (Svalbard) with a particular focus on the SMB-elevation feedback. In a recent study Ziemen et al (2016) modelled the temporal evolution of the Juneau ice field (Alaska), which exhibits some similarities to ice caps.…”