The long-term instability of glacier dynamics due to a sudden change of basal lubrication

Abstract:

A sudden change of subglacial hydrology, e.g., basal lubrication by meltwater routing, can trigger seasonal speedups and short-term glacier collapse. It is, however, still unclear about whether there exist some glacier geometries that are susceptible to the disruption of subglacial hydrology, and how this disruption poses long-term changes in glacier dynamics. Here I develop a modified 1-D physical framework designed to relate the dynamic vulnerability to basal lubrication. I start from the equations of ice thickness perturbations but assume that the initial perturbation comes from a sudden change of basal friction instead of a change of ice thickness. The elevation change rate (dh/dt) derived from this modified framework is composed of two parts. The first part is a constant over time (dubbed J₀), and its value depends on initial ice thickness, surface slope, and glacier speed. The second part describes a competing mechanism between advection and diffusion. The relative strength of advection to diffusion can be assessed by the Péclet number (P_e), which depends on the same physical quantities used for calculating J₀. To test J₀ and P_e as indicators of dynamic vulnerability, I use the glacier velocities, surface elevations, and basal elevations from the Ice Thickness Models Intercomparison eXperiment (ITMIX) data set. Both J₀ and P_e are calculated for each marine-terminating glacier basin in Austfonna Ice Cap, Svalbard, and Vavilov Ice Cap, Severnaya Zemlya, Russia. The long-term glacier instability for each is estimated by the difference of glacier velocity between the ITMIX (1996) and NASA ITS_LIVE (2014–present) data sets. My results show that J₀ and P_e correlate with the increase of glacier velocity, suggesting certain combinations of glacier geometry and speed that are more vulnerable to basal lubrication than the others. Since a rise in glacier speed increases glacier instability in my model by changing J₀ and P_e, short-term basal lubrication can possibly lead to a long-term shift in dynamic vulnerability, as already indicated by some large glacier destabilizations in the Arctic.