 |
| Titel |
Toward more accurate basal boundary conditions: a new 2-D model of distributed and channelised subglacial drainage |
| VerfasserIn |
M. A. Werder, I. J. Hewitt, C. Schoof, G. E. Flowers |
| Konferenz |
EGU General Assembly 2012
|
| Medientyp |
Artikel
|
| Sprache |
Englisch
|
| Digitales Dokument |
PDF |
| Erschienen |
In: GRA - Volume 14 (2012) |
| Datensatznummer |
250070878
|
|
|
|
|
|
| Zusammenfassung |
| Basal boundary conditions are one of the least constrained components of today’s ice sheet
models. To get at these one needs to know the distributed basal water pressure. We
present a new glacier drainage system model to contribute to this missing piece of the
puzzle.
This two dimensional mathematical/numerical model combines distributed and
channelised drainage at the ice-bed interface coupled to a water storage component. Notably
the model determines the location of the channels as part of the solution. This is achieved by
allowing channels (modelled as R-channels) to form on any of the edges of the unstructured
triangular grid used to discretise the model. The distributed system is represented by a water
sheet which is a continuum description of a linked-cavity system and exchanges water with
the channels along their length. Water storage is parameterised as a function of the subglacial
water pressure, which can be interpreted as storage in an englacial aquifer or due to
elastic processes. The parabolic equation that determines the water pressure is solved
using finite elements, the time evolution of the water sheet thickness and channel
diameter are governed by local differential equations that are integrated using explicit
methods.
To explore the model’s properties, we apply it to synthetic ice sheet catchments with areas
up to 3000km2. We present steady state drainage system configurations and evaluate their
channel-network properties (fractal dimensions, channel spacing). We find that an
arborescent channel network forms whose density depends on the water sheet conductivity
relative to water input. As a further experiment, we force the model with a seasonally and
diurnally varying melt water input to investigate how the modelled drainage system evolves
on these time scales: a channelised system grows up glacier as meltwater is delivered to the
bed in spring and collapses in autumn. Water pressure is highest just before the formation of
channels and then drops. Conversely, the diurnal variations in discharge affect the drainage
system morphology only slightly. Instead they lead to large water pressure variations
which lag meltwater input and coincide with changes in the volume of stored water.
By incorporating an evolving R-channel network within a continuum model of
distributed water drainage and storage, this 2-D model succeeds in qualitatively
reproducing many of the observed and postulated features of the glacier drainage
system. |
|
|
|
|
|
|