Thinning sea ice weakens buttressing force of iceberg mélange and promotes calving

Abstract

At many marine-terminating glaciers, the breakup of mélange, a floating aggregation of sea ice and icebergs, has been accompanied by an increase in iceberg calving and ice mass loss. Previous studies have argued that mélange may suppress calving by exerting a buttressing force directly on the glacier terminus. In this study, I adapt a discrete element model to explicitly simulate mélange as a cohesive granular material. Simulations show that mélange laden with thick landfast sea ice produces enough resistance to shut down calving at the terminus. When sea ice within mélange thins, the buttressing force on the terminus is reduced and calving is more likely to occur. When a calving event does occur, it initiates a propagating jamming wave within mélange, which causes local compression and then slow mélange expansion. The jamming wave can also initiate widespread fracture of sea ice and further increase the likelihood of subsequent calving events.

Figure 1. Snapshots of iceberg positions and velocities in two different channel configurations.

Both snapshots are 11 days into simulations with steady terminus advance and no calving, with 2-m-thick sea ice bonds, which are not plotted here. Velocity of each iceberg element is indicated by filled colour. Terminus elements are filled grey circles. Channel sides are filled black circles. (a) Straight channel configuration. (b) Narrowed channel configuration.

Figure 2. Mélange buttressing force on the terminus as a function of sea-ice bond thickness.

Buttressing force is obtained by averaging over the terminus width and over 1 month of simulations run to a statistical steady state. Red lines are simulations with a straight channel configuration. Blue lines are simulations with a narrowed channel configuration. Solid lines are simulations that permit sea-ice bonds between mélange elements and between mélange elements and channel side elements. Dashed lines are simulations with sea-ice bonds only permitted between mélange elements. Grey shading indicates the mélange buttressing force that prevents penetration of fractures near terminus, as estimated in Fig. 9 of Krug et al. and consistent with the Amundson et al. estimates for prevention of iceberg rotation from a terminus at flotation (for θ=1°).

Figure 3. Spatiotemporal response of mélange without sea ice to a prescribed calving event.

(a,b) are simulations, and (c,d) are observations from Jakobshavn Isbræ reproduced from Fig. 3 in Peters et al.. (a,c) Mélange velocity (averaged across channel width) as a function of distance along channel length at different times following calving event. (b,d) Mélange velocity (averaged across channel width) as a function of time following calving event at different locations along channel length. Grey shading indicates duration of calving event. Black dashed line plots the area occupied by mélange normalized by pre-calving area.

Figure 4. Simulated response of mélange with 2-m-thick sea ice to a calving event.

Green solid line is mélange buttressing force (averaged over terminus width and not including calving iceberg). Orange dashed line is the number of sea-ice bonds as a function of time following calving event.