Changes in ice-shelf buttressing following the collapse of Larsen A Ice Shelf, Antarctica, and the resulting impact on tributaries

Sam Royston,G Hilmar Gudmundsson

doi:10.1017/jog.2016.77

Abstract

ABSTRACTThe dominant mass-loss process on the Antarctic Peninsula has been ice-shelf collapse, including the Larsen A Ice Shelf in early 1995. Following this collapse, there was rapid speed up and thinning of its tributary glaciers. We model the impact of this ice-shelf collapse on upstream tributaries, and compare with observations using new datasets of surface velocity and ice thickness. Using a two-horizontal-dimension shallow shelf approximation model, we are able to replicate the observed large increase in surface velocity that occurred within Drygalski Glacier, Antarctic Peninsula. The model results show an instantaneous twofold increase in flux across the grounding line, caused solely from the reduction in backstress through ice shelf removal. This demonstrates the importance of ice-shelf buttressing for flow upstream of the grounding line and highlights the need to explicitly include lateral stresses when modelling real-world settings. We hypothesise that further increases in velocity and flux observed since the ice-shelf collapse result from transient mass redistribution effects. Reproducing these effects poses the next, more stringent test of glacier and ice-sheet modelling studies.

Highlights

The Larsen Ice Shelf, located to the east of the Antarctic Peninsula, has exhibited a persistent, stepped retreat since aerial observations began in the 1950s, with significant retreat since 1986 (Fig. 1; Cooper, 1997; Ferrigno and others, 2008; Cook and Vaughan, 2010)
A metric of local buttressing attributable to the presence of the ice shelf is defined by Gudmundsson (2013), calculated as the difference between the pressure, which would be exerted by hydrostatic equilibrium and that pressure acting normal to the grounding line
We have demonstrated the importance of ice-shelf buttressing for tributary glacier flow by replicating the observed acceleration following the collapse of Larsen A Ice Shelf

Summary

Introduction

The Larsen Ice Shelf, located to the east of the Antarctic Peninsula, has exhibited a persistent, stepped retreat since aerial observations began in the 1950s, with significant retreat since 1986 (Fig. 1; Cooper, 1997; Ferrigno and others, 2008; Cook and Vaughan, 2010). The floating area of the Larsen A Ice Shelf reduced by half during the last week of January 1995 alone, and by 8 March 1995, covered only 18% of its previous extent (Rott and others, 1997; Skvarca and others, 1999). Observations following the Larsen A collapse confirmed the significant and rapid effect of this event on the flow dynamics of tributary glaciers, with far-reaching effects upstream (Rott and others, 2002, 2014; De Angelis and Skvarca, 2003; Shuman and others, 2011). Previous modelling studies of Larsen A considered the conditions and possible mechanisms for the break-up of the ice shelf (Doake and others, 1998; Scambos and others, 2000; Vieli and others, 2007; Albrecht and Levermann, 2014), but did not consider the whole ice-sheet ice-shelf system including observed changes in buttressing and velocity and the effects upstream of the grounding line within tributary glaciers. The collapse of Larsen A provides one of the few examples in glaciology, where model validation can be performed

Methods

Results

Discussion

Conclusion