A Python modelling framework for estimating tunnel-valley erosion under former and future ice sheets

Estimating the erosive potential of future glaciations is important for assessing the long-term safety of nuclear waste repositories. Of special interest is the formation of tunnel-valleys, glacial erosional features over 500 m deep carved by meltwater channels beneath ice sheets and glaciers. We present a Python modeling framework to estimate tunnel-valley erosion using the erosion and deposition model of Walder & Fowler (1994) in combination with several open-source models and libraries for ice sheet modelling, flexural isostasy modelling, image pattern recognition, and meltwater routing. We showcase how the model works and performs on a synthetic fluvial pre-glacial landscape with a fixed simplified ice sheet configuration. The model works by routing both surface meltwater through simulated moulins in the ice sheet and basal meltwater on the hydraulic potential surface under the ice. The meltwater flow modulates the locations for initial channel incision, wherein the erosion and deposition will be calculated iteratively as the erosion and deposition rate depends on the channel cross-section. The preliminary results of this model show that a steady flux of meltwater from a deglaciating ice sheet can incise tunnel valleys tens of meters deep and several hundred meters wide over a period of hundreds of years. We anticipate incorporating features such as catastrophic lake drainage and an erosion parameter controlled by lithology into the model, as these mechanisms are believed to play a crucial role in the formation of large-scale tunnel valleys.