Abstract
The main objectives of this PhD study have been to quantify the large-scale hydrological and geomorphological processes and implications related to three large proglacial rivers draining into Kangerlussuaq Fjord from the Greenland Ice Sheet (GrIS). These rivers are Watson, Umivit and Sarfartoq River. Hydrological processes studied are: proglacial discharge volumes and timing of this, creation of runoff (i.e. melt), storage and release features such as ice dammed lakes (jökulhlaups lakes), the role of supraglacial lakes, storage and release inside or beneath the ice and the role of the so called firn zone are studied. Geomorphological processes studied are: Mass transfer from the glacial system and into the nearby fjord and delta systems. As a part of this glacial erosion rates are deduced. Extensive discharge and sediment transport measurements provided the basis for gaining insight to these processes.
It was shown that the specific sediment transport from the Watson River catchment (WRC) was 940, 470 and 1830 t km-2 yr-1 for average, minimum and maximum respectively which corresponds erosion rates of 0.3, 0.2 and 0.7 mm yr-1 when averaging the total sediment load over the entire catchment area. The role of jökulhlaups in this catchment was found to be less than 2% of the total annual discharge and sediment transport in Watson River (WR). Thus, despite their spectacular appearance, they do not play an important hydrological role.
Measurement campaigns carried out in the two other large glacially-fed rivers draining into Kangerlussuaq Fjord indicated similar values for sediment transport in Umivit and lower values in Sarfartoq River. The average annual sediment load provided from these three rivers would correspond to a sedimentation of ~44mm yr-1 if the sediment is assumed distributed equally over the deep inner basin of Kangerlussuaq Fjord.
Different melt models were applied to the MIKE SHE and MIKE 11 water routing models in order to simulate the proglacial water flows. The best melt model was a surface energy balance model. The routing model improved the modelled proglacial discharge and was applied to investigate detailed processes such as storage and release of meltwater from the glacial system. It showed promising first results, though better constrained input data is needed before firm conclusions on detailed processes can be drawn.
In July 2012, a several day long period of unprecedented high proglacial discharge rates was observed. The source to all this water was investigated in detail and it was concluded that less than 2% of the water originated from a release of stored water in lakes. The remaining volume was attributed to extreme surface melt caused by high temperatures and a low surface albedo that amplified surface melt. This unprecedented amount of proglacial discharge would only be possible if the runoff was generated from the lower accumulation zone, where runoff is usually buffered by the air-filled pore space in the firn. This means that the water melted in the lower accumulation zone of the GrIS is able to runoff directly and contribute to sea-level rise.
It was shown that the specific sediment transport from the Watson River catchment (WRC) was 940, 470 and 1830 t km-2 yr-1 for average, minimum and maximum respectively which corresponds erosion rates of 0.3, 0.2 and 0.7 mm yr-1 when averaging the total sediment load over the entire catchment area. The role of jökulhlaups in this catchment was found to be less than 2% of the total annual discharge and sediment transport in Watson River (WR). Thus, despite their spectacular appearance, they do not play an important hydrological role.
Measurement campaigns carried out in the two other large glacially-fed rivers draining into Kangerlussuaq Fjord indicated similar values for sediment transport in Umivit and lower values in Sarfartoq River. The average annual sediment load provided from these three rivers would correspond to a sedimentation of ~44mm yr-1 if the sediment is assumed distributed equally over the deep inner basin of Kangerlussuaq Fjord.
Different melt models were applied to the MIKE SHE and MIKE 11 water routing models in order to simulate the proglacial water flows. The best melt model was a surface energy balance model. The routing model improved the modelled proglacial discharge and was applied to investigate detailed processes such as storage and release of meltwater from the glacial system. It showed promising first results, though better constrained input data is needed before firm conclusions on detailed processes can be drawn.
In July 2012, a several day long period of unprecedented high proglacial discharge rates was observed. The source to all this water was investigated in detail and it was concluded that less than 2% of the water originated from a release of stored water in lakes. The remaining volume was attributed to extreme surface melt caused by high temperatures and a low surface albedo that amplified surface melt. This unprecedented amount of proglacial discharge would only be possible if the runoff was generated from the lower accumulation zone, where runoff is usually buffered by the air-filled pore space in the firn. This means that the water melted in the lower accumulation zone of the GrIS is able to runoff directly and contribute to sea-level rise.
| Originalsprog | Engelsk |
|---|
| Forlag | Department of Geosciences and Natural Resource Management, Faculty of Science, University of Copenhagen |
|---|---|
| Antal sider | 253 |
| Status | Udgivet - 2014 |