Glacial Drainage Systems Characterization Using Inverse Modelling and Remote Sensing
Water flow on top, trough and at the bottom of glaciers exerts key controls on ice-dynamics. Moreover, water pressure at the bottom of glaciers plays a crucial role modulating the ice-bedrock coupling. Under today’s climate change scenario, predicting glacier responses to future climatic forcing is...
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English |
| Published: |
Université de Lausanne, Faculté des géosciences et de l'environnement
2020
|
| Subjects: | |
| Online Access: | https://serval.unil.ch/resource/serval:BIB_8AAEF504B03D.P001/REF.pdf https://serval.unil.ch/notice/serval:BIB_8AAEF504B03D |
| Summary: | Water flow on top, trough and at the bottom of glaciers exerts key controls on ice-dynamics. Moreover, water pressure at the bottom of glaciers plays a crucial role modulating the ice-bedrock coupling. Under today’s climate change scenario, predicting glacier responses to future climatic forcing is imperative. Consequently, significant scientific effort has been drawn to the study and modelling of different aspects of glacier hydrology and its feedbacks with ice-dynamics. This PhD thesis focuses on studying glacier hydrology components. First, the attention is drawn to subglacial hydrology. Recent advances in subglacial hydrology include numerical models that successfully incorporate most of the known physical processes. However, conditioning to data and uncertainty quantification are challenging mostly due to the complexity of subglacial systems and expensive computations as well as a high number of unknown parameters. Moreover, estimating uncertainties is key to evaluate predictions of glacier evolution into climate change scenarios. Then, the first aim of this thesis is to develop a framework to infer the spatial structure and hydraulic characteristics of the subglacial drainage system from field observations. This is contrasting with previous work where some of the hydraulic parameters and the spatial structure are an emergent result from a process-based model where physical equations are solved. In Chapter II, a method that combines geostatistical and physical processes is developed to generate subglacial channels. One of the main advantages of geostatistical models is that they are computationally inexpensive. Then, a probabilistic inverse framework is used to infer subglacial channels that honour observations and moreover to quantify uncertainty. The methodology is tested on a synthetic or idealized ice-sheet geometry. Next, the methodology is further adapted and improved to suit a real-world scenario. The Gorner Glacier is selected as it comprises one of the most complete data sets. It is shown that, ... |
|---|