Considering thermal-viscous collapse of the Greenland ice sheet.

Abstract

We explore potential changes in Greenland ice sheet form and flow associated with increasing ice temperatures and relaxing effective ice viscosities. We define “thermal‐viscous collapse” as a transition from the polythermal ice sheet temperature distribution characteristic of the Holocene to temperate ice at the pressure melting point and associated lower viscosities. The conceptual model of thermal‐viscous collapse we present is dependent on: (1) sufficient energy available in future meltwater runoff, (2) routing of meltwater to the bed of the ice sheet interior, and (3) efficient energy transfer from meltwater to the ice. Although we do not attempt to constrain the probability of thermal‐viscous collapse, it appears thermodynamically plausible to warm the deepest 15% of the ice sheet, where the majority of deformational shear occurs, to the pressure melting point within four centuries. First‐order numerical modeling of an end‐member scenario, in which prescribed ice temperatures are warmed at an imposed rate of 0.05 K/a, infers a decrease in ice sheet volume of 5 ± 2% within five centuries of initiating collapse. This is equivalent to a cumulative sea‐level rise contribution of 33 ± 18 cm. The vast majority of the sea‐level rise contribution associated with thermal‐viscous collapse, however, would likely be realized over subsequent millennia.