Theory 13/Cryosphere & Ice /Glaciology/3 min
Sliding Has a Pressure Threshold
More water is supposed to mean faster sliding. Sometimes it means the opposite.
Glacier sliding is controlled by effective pressure at the bed rather than by meltwater volume, so adding water can slow a glacier if it raises pressure, and speed-up events should follow drainage configuration rather than melt rate.
01The evidence
- The growth and collapse of the distributed subglacial hydrological system of Kennicott Glacier was shown to control basal motion, with the configuration of the drainage system mattering more than the water volume.Kennicott Glacier subglacial hydrology
- Subglacial water pressure is the variable that couples hydrology to basal motion, which is why the same water input can produce different responses depending on how it is stored.Water pressure overview
02The tension
Meltwater is treated as a lubricant: more melt, faster sliding, higher sea-level contribution. The relation is observed in some settings and reversed in others, and the reversals are usually attributed to local conditions rather than to a threshold.
03The theory
Sliding depends on effective pressure, the difference between ice overburden and water pressure at the bed. High water pressure lifts the ice and reduces friction; low pressure increases friction even if water is present.
A distributed drainage system can evacuate water efficiently and keep pressure low, so adding melt may increase sliding only until the system reorganises. A channelised system drains efficiently; a distributed system does not. The transition between them is the threshold.
The prediction is that speed-up events should coincide with drainage reorganisation rather than with peak melt, and that adding water to a well-drained bed should slow the glacier by increasing connectivity and lowering pressure. That is the opposite of the lubricant intuition and it is testable in the field.
The consequence for sea-level projection is that models driven by melt rate alone will mis-time speed-ups, and that the drainage configuration should be the state variable.
04What this is not
Those models make sliding a function of melt supply. This makes it a function of drainage configuration through effective pressure, which predicts slowdowns from added water in well-drained settings.
05Predictions
- P1Speed-up events should follow drainage reorganisation rather than peak melt.
- P2Adding water to a well-drained bed should slow sliding.
- P3Models with drainage state should outperform melt-driven models on seasonal records.
06What would refute it
Sliding tracking melt supply regardless of drainage state would restore the lubricant model.
07What it would change
It would change how glacier speed-ups are forecast, and improve sea-level projections.
08Evidence base
- Growth and collapse of the subglacial drainage system and basal motion, Kennicott Glacierhttps://scholar.google.be/citations?view_op=view_citation&hl=fr&user=MIBKAWwAAAAJ&citation_for_view=MIBKAWwAAAAJ:u-x6o8ySG0sC
- Water pressure — overviewhttps://www.sciencedirect.com/topics/earth-and-planetary-sciences/water-pressure