A double continuum hydrological model for glacier applications

Abstract

Abstract. The flow of glaciers and ice streams is strongly influenced by the presence of water at the interface between ice and bed. In this paper, a hydrological model evaluating the subglacial water pressure is developed with the final aim of estimating the sliding velocities of glaciers. The global model fully couples the subglacial hydrology and the ice dynamics through a water-dependent friction law. The hydrological part of the model follows a double continuum approach which relies on the use of porous layers to compute water heads in inefficient and efficient drainage systems. This method has the advantage of a relatively low computational cost that would allow its application to large ice bodies such as Greenland or Antarctica ice streams. The hydrological model has been implemented in the finite element code Elmer/Ice, which simultaneously computes the ice flow. Herein, we present an application to the Haut Glacier d'Arolla for which we have a large number of observations, making it well suited to the purpose of validating both the hydrology and ice flow model components. The selection of hydrological, under-determined parameters from a wide range of values is guided by comparison of the model results with available glacier observations. Once this selection has been performed, the coupling between subglacial hydrology and ice dynamics is undertaken throughout a melt season. Results indicate that this new modelling approach for subglacial hydrology is able to reproduce the broad temporal and spatial patterns of the observed subglacial hydrological system. Furthermore, the coupling with the ice dynamics shows good agreement with the observed spring speed-up.

A confined–unconfined aquifer model for subglacial hydrology and its application to the North East Greenland Ice Stream

“…For this reason, a new generation of 5 subglacial drainage models has been developed recently that is capable of coupling the two regimes of drainage and reproducing the transition between them (Schoof, 2010;Hewitt et al, 2012;Werder et al, 2013;De Fleurian et al, 2014). While these models produce remarkable results for spontaneously evolving channel networks, it is still a challenge to apply them on continental scale.…”

“…This leads to increased modeling complexity and high computational costs. An exception to this is the work of De Fleurian et al (2014), where both systems 15 are represented by Darcy flow through separate porous media layers. The layer representing the channels has its parameters (namely permeability and storage) adjusted to exhibit the behavior of an effective system.…”

“…We take this idea even further and only use a single layer of Darcy flow with locally adjusted transmissivity of the layer at locations where channels form while, at the same time, accounting for a decrease in storage. This means that we approximate the channel flow as a fast diffusion process similarly to work in De Fleurian et al (2014); however, a single Darcy flow layer 20 with spatially varying parameters (effective permeability) accounts for both drainage mechanisms. Similar approaches are known to have been applied to modeling of fracture networks in rock Van Siclen (2002).…”

Mass-conserving subglacial hydrology in the Parallel Ice Sheet Model version 0.6

“…Models have combined subsets of these different morphologies and processes (Flowers and Clarke, 2002a;Hewitt, 2013;Hoffman and Price, 2014;Van der Wel et al, 2013;Werder et al, 2013;de Fleurian et al, 2014). However, the completeness of the modeled processes must be balanced against the number of uncertain model parameters and the ultimate availability of observations with which to constrain them.…”

Determining Ice-Sheet Uplift Surrounding Subglacial Lakes with a Viscous Plate Model

“…Models of subglacial hydrological processes are becoming more sophisticated, with the ability to simulate 2D configurations of efficient and inefficient drainage systems (Hewitt, 2013;Werder et al, 2013;de Fleurian et al, 2014) and are converging toward integrated description of entire basal drainage networks. However, these models do not incorporate realistic criteria for ice flexure in response to changing basal water pressure.…”