Final Report Summary - GLADS (2D glacial drainage system model (GlaDS): application and coupling to ice sheet model)
Executive summary and summary description
The prediction of future sea level rise due to dynamic changes of the ice flow in Greenland and Antarctica is one of the great outstanding problems of climate science. One reason why ice sheet models struggle to provide this information is the lack of a realistic subglacial drainage system model, providing basal boundary conditions for ice flow models. The aims of this project are the following: 1) The development and application of such a glacier drainage system model (GlaDS-model), which incorporates both the channelised and distributed drainage system at the glacier bed in 2D. 2) Two way coupling of GlaDS to an ice flow model.
The following work was done during the project:
• development of the GlaDS glacier drainage system model
• model application to synthetic and mountain glaciers
• field measurements in Greenland
• model application to two Greenland catchments
• coupling for GlaDS to two ice flow models: Stokes model Elmer/Ice; higher order model VarGlaS
• dissemination of the results through publications and presentations at conferences and universities
Put into a wider context, this research expanded our knowledge of the behaviour of the glacier drainage system. There are numerous future appli- cations of this work, including:
• The link between the drainage system and ice dynamics is probably the biggest interest as it will help improve our predictions of future sea level rise.
• The coupling between ice sheets and ocean may be influenced substan- tially by subglacial discharge. The fresh water discharge from the ice sheet enhances the circulation of ocean water underneath ice shelves and in fjords which leads to enhanced subshelf melting. This has im- pacts on ice sheet dynamics and shelf stability.
• The subglacial drainage system is one of the primary controls of sub- glacial erosion. The GlaDS model can help to improve our under- standing of glacial landscapes. This has far reaching implication, for instance, for the design of deep geological nuclear waste disposal.
• The drainage of glacier dammed lakes presents one of the greatest and far reaching hazards related to glaciers. This work will lead to improved modelling of such floods.
Summary results
Model application to Alpine glaciers
The model was applied to Gornergletscher, Switzerland, and results were published in Journal of Geophysical Research. The results are of the model application showed good agreement with our general understanding of the glacier drainage system. A future application to the same glacier is planned once the coupling of GlaDS to an ice flow model is finished.
Fieldwork on Russell/Leverett glacier catchment in Greenland
The fieldwork on Russell/Leverett outlet glacier in Summer 2012 together with Jemma Wadham’s team from University of Bristol was successful. Col- lected data include stream flow measurements and tracer data. It is planned to apply the GlaDS model to Russell/Leverett use the data for model vali- dation during the Return Phase in Bristol.
Model enhancements: tracer transport and thermodynamics
It was decided that and implementation of a parallel version of the model was not necessary as the model runs fast enough in its serial version to be applicable to ice sheet catchment scale problems. A coupled GlaDS-ice flow model would profit from being parallelised but then it would be best to implement this within the framework of the existing ice flow model.
The following model enhancements were implemented: coupling to sim- ple sliding relation, englacial water storage, sediment transport and erosion, bedrock erosion.
Application to ice sheet catchments
The following two applications were performed:
1. Application of GlaDS to Russell/Leverett catchment (West Green- land). Mauro Werder supervised Alan Kennedy, a Master Student, in this work. Substantial progress was made, including steady state drainage system calculations, seasonal runs and comparison to mea- surement. This work was presented at the IGS Meeting in Chamonix, France (May 2014), a publication is in preparation.
2. Application to Jakobshavn Isbrae: ongoing work to study the dynam- ical implications of subglacial hydrology for fast flowing ice streams. Presented at American Geophysical Union Fall Meeting (December 2013, San Francisco) and at the "Subglacial Hydrology Workshop" at the Lamont-Doherty Earth Observatory (Columbia University, NY, USA).
Coupling to ice dynamics
Coupling of GlaDS to ice dynamics is ongoing work in two projects:
1. Work with Olivier Gagliardini (LGGE Grenoble) to couple GlaDS to Elmer/Ice. Simple test cases of a shallow shelf model coupled to GlaDS are running. However, the coupling to the Stokes ice flow code is still unstable. Olivier Passalacqua did his Master Thesis dissertation on this topic (Université Joseph Fourier, Grenoble, 2014).
2. Work with Jesse Johnson (University of Montana, USA) to couple GlaDS to the VarGlaS ice flow model. Implementation of GlaDS within the FEniCS finite element library was done during a visit of Mauro Werder in Montana in December 2013. Coupling to ice flow is pending some work to implement a L1L2 depth integrated ice flow model.
Related work: glacial overdeepenings
The Swiss National Cooperative for the Disposal of Radioactive Waste (NA- GRA, http://www.nagra.ch/(öffnet in neuem Fenster)) is interested in understanding the formation of overdeepened glacial troughs better, in order to evaluate the risks for a planned deep geological nuclear waste repository. For a workshop organised by the NAGRA,‘ the GlaDS model has been applied to glacial overdeep- enings which shows a profound eff on the glacier drainage system with implications for the formation of such overdeepenings. A publication is in preparation.
The prediction of future sea level rise due to dynamic changes of the ice flow in Greenland and Antarctica is one of the great outstanding problems of climate science. One reason why ice sheet models struggle to provide this information is the lack of a realistic subglacial drainage system model, providing basal boundary conditions for ice flow models. The aims of this project are the following: 1) The development and application of such a glacier drainage system model (GlaDS-model), which incorporates both the channelised and distributed drainage system at the glacier bed in 2D. 2) Two way coupling of GlaDS to an ice flow model.
The following work was done during the project:
• development of the GlaDS glacier drainage system model
• model application to synthetic and mountain glaciers
• field measurements in Greenland
• model application to two Greenland catchments
• coupling for GlaDS to two ice flow models: Stokes model Elmer/Ice; higher order model VarGlaS
• dissemination of the results through publications and presentations at conferences and universities
Put into a wider context, this research expanded our knowledge of the behaviour of the glacier drainage system. There are numerous future appli- cations of this work, including:
• The link between the drainage system and ice dynamics is probably the biggest interest as it will help improve our predictions of future sea level rise.
• The coupling between ice sheets and ocean may be influenced substan- tially by subglacial discharge. The fresh water discharge from the ice sheet enhances the circulation of ocean water underneath ice shelves and in fjords which leads to enhanced subshelf melting. This has im- pacts on ice sheet dynamics and shelf stability.
• The subglacial drainage system is one of the primary controls of sub- glacial erosion. The GlaDS model can help to improve our under- standing of glacial landscapes. This has far reaching implication, for instance, for the design of deep geological nuclear waste disposal.
• The drainage of glacier dammed lakes presents one of the greatest and far reaching hazards related to glaciers. This work will lead to improved modelling of such floods.
Summary results
Model application to Alpine glaciers
The model was applied to Gornergletscher, Switzerland, and results were published in Journal of Geophysical Research. The results are of the model application showed good agreement with our general understanding of the glacier drainage system. A future application to the same glacier is planned once the coupling of GlaDS to an ice flow model is finished.
Fieldwork on Russell/Leverett glacier catchment in Greenland
The fieldwork on Russell/Leverett outlet glacier in Summer 2012 together with Jemma Wadham’s team from University of Bristol was successful. Col- lected data include stream flow measurements and tracer data. It is planned to apply the GlaDS model to Russell/Leverett use the data for model vali- dation during the Return Phase in Bristol.
Model enhancements: tracer transport and thermodynamics
It was decided that and implementation of a parallel version of the model was not necessary as the model runs fast enough in its serial version to be applicable to ice sheet catchment scale problems. A coupled GlaDS-ice flow model would profit from being parallelised but then it would be best to implement this within the framework of the existing ice flow model.
The following model enhancements were implemented: coupling to sim- ple sliding relation, englacial water storage, sediment transport and erosion, bedrock erosion.
Application to ice sheet catchments
The following two applications were performed:
1. Application of GlaDS to Russell/Leverett catchment (West Green- land). Mauro Werder supervised Alan Kennedy, a Master Student, in this work. Substantial progress was made, including steady state drainage system calculations, seasonal runs and comparison to mea- surement. This work was presented at the IGS Meeting in Chamonix, France (May 2014), a publication is in preparation.
2. Application to Jakobshavn Isbrae: ongoing work to study the dynam- ical implications of subglacial hydrology for fast flowing ice streams. Presented at American Geophysical Union Fall Meeting (December 2013, San Francisco) and at the "Subglacial Hydrology Workshop" at the Lamont-Doherty Earth Observatory (Columbia University, NY, USA).
Coupling to ice dynamics
Coupling of GlaDS to ice dynamics is ongoing work in two projects:
1. Work with Olivier Gagliardini (LGGE Grenoble) to couple GlaDS to Elmer/Ice. Simple test cases of a shallow shelf model coupled to GlaDS are running. However, the coupling to the Stokes ice flow code is still unstable. Olivier Passalacqua did his Master Thesis dissertation on this topic (Université Joseph Fourier, Grenoble, 2014).
2. Work with Jesse Johnson (University of Montana, USA) to couple GlaDS to the VarGlaS ice flow model. Implementation of GlaDS within the FEniCS finite element library was done during a visit of Mauro Werder in Montana in December 2013. Coupling to ice flow is pending some work to implement a L1L2 depth integrated ice flow model.
Related work: glacial overdeepenings
The Swiss National Cooperative for the Disposal of Radioactive Waste (NA- GRA, http://www.nagra.ch/(öffnet in neuem Fenster)) is interested in understanding the formation of overdeepened glacial troughs better, in order to evaluate the risks for a planned deep geological nuclear waste repository. For a workshop organised by the NAGRA,‘ the GlaDS model has been applied to glacial overdeep- enings which shows a profound eff on the glacier drainage system with implications for the formation of such overdeepenings. A publication is in preparation.