Basal Topography Simulation of the Tsanfleuron Glacier (Swiss Alps) Using Geostatistical Methods and GPR Measurements : Ice Volume Estimation and Uncertainty Calculation.
Abstract
In Alpine environments, changes in precipitation and temperature due to global climatic changes have modified the glaciers mass balance equilibrium resulting in the most likely disappearance of the majority of them in the next decades. With the rapid retreat of these ice sheets, monitoring of glacier thickness and their associated ice volume are essential keys for future water resources management, sediment production and slope stabilisation assessment. In 2019, Ground Penetrating Radar (GPR) lines were acquired on the Tsanfleuron Glacier (Vaud, Switzerland) to determine the thickness of the ice. This method provides thickness data only along the acquisition lines and therefore an interpolation step is required to produce a spatially continuous basal map. The Tsanfleuron Glacier lies on a homogeneous limestone bedrock with complex karstic geomorphological features, which can be delicate to model using classical interpolation techniques. We here investigate the applicability of Multiple Point Statistics (MPS) simulations to interpolate complex glacier basal topography using GPR lines as conditioning data and uncovered bedrock topography as training image. Variogram-based methods such as Sequential Gaussian Simulation (SGS) and Kriging estimation are also compared to the MPS approach, on both synthetic - with known topography - and real cases. These methods are compared using both pixel-based quality indicators and global volume estimations. Depending on the chosen interpolation strategy, the calculated ice volumes and the basal topographies can differ significantly. Using the modeled MPS basal topographies set, the ice volume for the Tsanfleuron Glacier was estimated to be 113.9 ± 1.6 Mio m3 in 2019. The MPS approach, unlike SGS and Kriging, produces not only an accurate volume estimation, with a robust uncertainty quantification, but also the most realistic basal topography, with complex karstic geomorphological features being well reproduced. Such structures reproduction are capital especially when trying to predict complex non-linear processes such as subglacial underground flow.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2021
- Bibcode:
- 2021AGUFM.C13B..02D