AGU 2025

Coupled hydro-thermomechanical model run on T2 sample domain over Thwaites Eastern Shear Margin (fast end). 1 frame = 270 days. 20 frames ~ 15 years.

Ice flow is toward the left and out of the page toward the viewer (-x, -y direction wrt the labeled axes). Note vertical exaggeration is x10.

Shows high feedback between processes: particularly the melt - > slip -> reduced friction heat -> less melt -> less slip -> more friction heat - > more melt, has a periodically evolving effect over the whole shear zone. 

The Hydrologic boundary conditions of the above model are derived from an updated Amundsen Sea Sector regional hydrology model using the Glacier Drainage System model (GlaDS). Nearly identical to Dow 2022, we introduce two key changes. we connect the catchment domain of upper and lower pine island glacier to allow the possibility of water flow in that direction, rather than down Thwaites. Secondly, we update the creep closure parameter from 3.5*^-25 Pa-3 s-1 -- the Cuffey and Patterson value used in the original Werder 2013 paper, to 24*^-25, the Cuffey and Patterson value suggested for temperate ice (of which the bed is over the vast majority of the domain. 

The main change in the result is what we like to call PIG Piracy. in nearly all parameter ensemble runs Pine Island routes a decent portion of the melt water that ends up in the Byrd Subglacial Basin. Only under temperate ice with an enhancement factor >3 (note this is the creep closure enhancement, so it would have to be anisotropically speaking, a vertical single maximum, or vertical girdle crystal orientation at the bed) do the channelized systems turn and route into Thwaites. 

Because of the results above, we can see dynamic shifts between flow and slip on sub annual scales, it could still be within reason - especially if we consider sediment transport, that there is dynamic routing of water that alternates between basins. This exercise we will leave to my future postdoctoral work.