The evolution of isolated cavities and hydraulic connection at the glacier bed – Part 2: A dynamic viscoelastic model
Many large-scale subglacial drainage models implicitly or explicitly assume that the distributed part of the drainage system consists of subglacial cavities. Few of these models, however, consider the possibility of hydraulic disconnection, where cavities exist but are not numerous or large enough t...
Published in: | The Cryosphere |
---|---|
Main Author: | |
Format: | Article in Journal/Newspaper |
Language: | English |
Published: |
Copernicus Publications
2023
|
Subjects: | |
Online Access: | https://doi.org/10.5194/tc-17-4817-2023 https://noa.gwlb.de/receive/cop_mods_00069889 https://noa.gwlb.de/servlets/MCRFileNodeServlet/cop_derivate_00068257/tc-17-4817-2023.pdf https://tc.copernicus.org/articles/17/4817/2023/tc-17-4817-2023.pdf |
Summary: | Many large-scale subglacial drainage models implicitly or explicitly assume that the distributed part of the drainage system consists of subglacial cavities. Few of these models, however, consider the possibility of hydraulic disconnection, where cavities exist but are not numerous or large enough to be pervasively connected with one another so that water can flow. Here I use a process-scale model for subglacial cavities to explore their evolution, focusing on the dynamics of connections that are made between cavities. The model uses a viscoelastic representation of ice and computes the pressure gradients that are necessary to move water around basal cavities as they grow or shrink. The latter model component sets the work here apart from previous studies of subglacial cavities and permits the model to represent the behaviour of isolated cavities as well as of uncavitated parts of the bed at low normal stress. I show that connections between cavities are made dynamically when the cavitation ratio (the fraction of the bed occupied by cavities) reaches a critical value due to decreases in effective pressure. I also show that existing simple models for cavitation ratio and for water sheet thickness (defined as mean water depth) fail to even qualitatively capture the behaviour predicted by the present model. |
---|