A proxy-based model of glacier thermal characteristics

Dia Martinez Gracey , Gwenn Elizabeth Flowers, Mylène Jacquemart

This work presents a spatially lumped, computationally efficient model for estimating glacier thermal characteristics at a regional scale. We formulated proxies for the major englacial heat sources (meltwater refreezing and vertical shear strain heating) and heat transport mechanisms (downstream advection and vertical diffusion). We subsequently calculated these proxies for all glaciers in Alaska (RGI-01) and Svalbard (RGI-07) for which sufficient input data are available (26,798 total). To relate glacier thermal characteristics to the computed proxies, we built a decision tree that classified the glaciers into five groups based on similar relative proxy values. We then cautiously interpreted each group based on its dominant thermal proxies and correspondence with glaciers of known thermal structure. Based on these interpretations, temperate glaciers would represent most of the glacierized area in Alaska (62\%), whereas cold-layer polythermal glaciers would represent 73\% of glacierized area in Svalbard. One unexpected thermal group emerged, which appears to comprise many small glaciers and some large glaciers (8\% of glacierized area) in Alaska with high-elevation accumulation areas, resulting in a thermal structure with cold ice in the accumulation area and temperate ice in the ablation area. This structure has not yet been documented in Alaska, and may warrant further attention.