Back to search

FRINATEK-Fri prosj.st. mat.,naturv.,tek

Subglacial Water Impact on long term glacier Dynamics

Alternative title: Innvirkning av vann under isbreer på langsiktig isdynamikk

Awarded: NOK 8.3 mill.

The current warming over Greenland leads to an increase of the melt at the surface of the ice sheet. This means that a larger volume of water reaches the base of the glacier where it can greatly impact its dynamic. SWItchDyn will investigate the long-term impact of this increase in melt water volume on the dynamics of glaciers. At the base of glaciers, an increase in water pressure triggers an acceleration of the ice flow. Water pressure is however not directly related to the volume of available water. This is due to the fact that the subglacial drainage system can adapt depending on the provided volume of water. Under a low water input, the drainage system will be able to drain only a small amount of water with a relatively high pressure. As the volume of water increases, the pressure increases until a threshold is reached where the drainage system becomes more efficient. This shift from an inefficient to an efficient drainage system leads to a decrease of the water pressure and a reduction of the glacier's velocity. In order to better understand these interactions, SWItchDyn will combine two different approaches. A numerical model will be used to get a better understanding of the relations between the hydrological and ice-dynamical system. In parallel, observations of the surface velocities on glaciers will be analyzed to give a better idea of the current state of the subglacial drainage system. We will combine the results from these two different approaches to build a realistic simulation of the west coast of Greenland. This final simulation will provide an estimate of the effect of the melt water volume increase on the mass loss of the Greenland Ice Sheet. On the modeling side, we performed a range of perturbation experiment to study the effect of changes in the characteristics of the melt season. Our results show that longer and less intense melt seasons lead to faster velocity than short melt seasons with a higher intensity. Those conclusion only apply to annual mean velocities as a more intense melt season will still trigger higher maximum velocity. It is also worth noting that the response of the glacier is different at different elevation and it is a point that we will investigate further in coming works. Our last results have been aimed to clarify the importance of the way we inject meltwater in our subglacial hydrology models. The experiments show that injecting water in a few specified locations rather than at every point of the model have a large impact on the glacier velocities. These results point to the need of further studies to be able to better characterize the water input into the subglacial environment in order to produce more accurate simulations in the future. The work performed on the analysis of remotely sensed velocity data confirms that care should be taken in the selection of the timing and duration of the integration period used to compute ice speeds. Our analysis has also shown that one should be careful when averaging velocity trends in large region which generally present a large variability in their trends. Some preliminary result also seem to point that the relation between melt water availability and velocity in large regions of Greenland might have been overestimated. Further analysis of the remotely sensed velocities have confirmed our modeling results and shown that indeed the velocity variations in relation with water input were very much related to the altitude at which one is investigating those phenomenon. those observations warrant a better classification of altitudes when observing the effect of meltwater availability on glacier sliding in long term simulations.

The outcomes of the project have been targeted to two different sub-disciplines. SWITchDyn's PhD has work extensively on the remote sensing part of the project giving advice on the best possible ways to treat ice velocities of low magnitude and how to interpret the trend that can be derived from those. From this work it appears that one should take more care than what is currently done in the selection of image pairs used to derive velocities in order to get reliable trends when the changes are small. Concurrent to that it appears that most of the trends that appear in the velocities are spatially heterogeneous and that one should take care when deriving trends from area averaged values. When looking at trends it also appears that the impact of runoff volume on ice velocities is not straightforward and highly variable depending on the surface elevation. SWItchDyn's PI has mostly worked on the response of the subglacial drainage system on different melt water forcing. Form this work it appears that the runoff volume is not the only value to consider when looking at the feedback between runoff and ice velocities. The temporal intensity of the water recharge is also an important factor to consider with longer melt season with a lower magnitude leading to faster glaciers. It seems that the impact of recharge magnitude holds also when investigating the spatial distribution of the water recharge, with more realistic inject ion from a limited numbers of moulins giving slower velocities than recharging the system from a uniform water sheet. This last outcome urges the community to move away from simpler forcing and find a good solution to produce realistic forcing to subglacial hydrology models. There is large potential to build upon the results of both the modeling and remote sensing work produced in SWIthcDyn. In term of modeling, the work done in SWItchDyn just opened the door to more extensive studies either on parameters that have not been investigated here, looking at longer timescales or more realistic settings. Some of those studies were clearly not achievable at the beginning of the project and the constant improvement of the subglacial hydrology model and it's coupling to ice dynamics during the duration of the project now allow to investigate new subjects.The remote sensing work of SWItchDyn has produced a huge database of velocities all over Greenland all the produced data is freely available and there is a wealth of data that has not been analyzed yet and could give a new insight on the effect of melt water volume on ice dynamics or other studies. The technical work of the PhD also sparked some interest into using annual velocity data and some work is ongoing to potentially allow to get higher temporal resolution time-series from annual velocity data which would allow to retain the advantages of annually derived data with a better temporal resolution.

Modelling of the basal velocities of glaciers remains an important problem in the prediction of the future evolution of glacier dynamics. While the current approach to glacier sliding is mainly based on the inversion of surface velocity measurements, this procedure is not well suited for the projection of future ice velocity. SWItchDyn will improve our understanding of the physical processes that impact the basal sliding of glaciers. SWItchDyn's main focus will be to investigate the evolution of the subglacial drainage system and its impact on glacier sliding under varying meltwater supply triggered by a changing climate. SWItchDyn will employ a dynamical ice sheet model to investigate remotely-sensed velocities and assess the link between climate change, subglacial hydrology, and glacier dynamics. Base on the knowledge provided by SWItchDyn it will be possible to carry out dynamical simulations with a coupled model explicitly resolving the sliding of glaciers and provide improved predictions of future sea-level rise.

Publications from Cristin

No publications found

No publications found

No publications found

Funding scheme:

FRINATEK-Fri prosj.st. mat.,naturv.,tek