Estimating the glacial valley fill of the north-west Greater Caucasus Mountains using artificial neural networks

Fallon, Ciara (2025) Estimating the glacial valley fill of the north-west Greater Caucasus Mountains using artificial neural networks. MSc(R) thesis, University of Glasgow.

Abstract

Glaciation and subsequent glacier retreat has resulted in unknown quantities of sediments being stored in valleys throughout the Greater Caucasus Mountains. It is difficult to estimate the quantity of this sediment fill as a result of there being no direct measurements of the thickness of these sediments. In this study, valley cross-sections are extracted from a catchment area in the western Caucasus using a DEM. Training data comprised of valley widths and sediment depth estimates is obtained from these cross-sections based on assumptions about the geometry of glacially eroded valleys. This training data is used to train an artificial neural network to create a model which can be used to estimate sediment depths and associated volumes in other valleys in the Caucasus. This model was then applied to other valleys that were found to show evidence of glacial sediment fill. A volume of 42.05 ± 1.01 km3 is found from a combination of 7 valleys in the north-western and central part of the Caucasus which corresponds to a total mass of 8.83x1013 ± 8.67x1012 kg. The individual results from each valley showed that there is a relationship between the average width of the valleys and the maximum depth of sediment found within them. The results also highlighted that there are parts of the valleys that do not show evidence of glacial fill where they may be expected to. This presents implications for the glacial processes that impact the location of this sediment fill such as glacier motion and glacial erosion. The quantity and distribution of these sediments also have implications for isostatic adjustment associated with glacial retreat in the Caucasus as they demonstrate how mass has been transferred away from the mountains, which can contribute to the surface uplift resulting from ice mass loss.

• Item Type: Thesis (MSc(R))
• Qualification Level: Masters
• Subjects: G Geography. Anthropology. Recreation > G Geography (General); Q Science > Q Science (General)
• Colleges/Schools: College of Science and Engineering > School of Geographical and Earth Sciences
• Supervisor's Name: Eizenhöfer, Dr. Paul and Neill, Dr. Iain
• Date of Award: 2025
• Depositing User: Theses Team
• Unique ID: glathesis:2025-85601
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 21 Nov 2025 14:33
• Last Modified: 21 Nov 2025 14:33
• Thesis DOI: 10.5525/gla.thesis.85601
• URI: https://theses.gla.ac.uk/id/eprint/85601