Iceberg distributions in Greenland’s fjords



Shiggins, Connor ORCID: 0000-0001-8249-921X
(2023) Iceberg distributions in Greenland’s fjords. PhD thesis, University of Liverpool.

[thumbnail of 201430044_May2023.pdf] Text
201430044_May2023.pdf - Author Accepted Manuscript

Download (6MB) | Preview

Abstract

Iceberg calving is currently responsible for half of the mass loss from Greenland’s marine-terminating glaciers and could increase well into the 21st century. While dynamically important for the stability of the Greenland Ice Sheet (GrIS), icebergs also have a major influence on fjord circulation and stratification through melting and the subsequent release of fresh and cold water. The ice-ocean interface remains a difficult to access zone and we consequently know little about iceberg distributions in the near-terminus regions of Greenland’s outlet glaciers. This is as a result of few observational studies existing which have sought to derive iceberg observations in the fjords of Greenland. This study utilises cloud computing techniques and digital elevation models (DEMs) to automatically detect icebergs around the GrIS and constrain their distributions to interpret the importance of iceberg sizes on calving and fjord dynamics. This thesis shows cloud computing is a powerful application to automatically identifying icebergs on large spatial and temporal scales using open access DEMs. The approach has been packaged into a graphical user interface to provide an iceberg detection tool for researchers wanting to derive a dataset of icebergs at Greenland’s ice-ocean interface. The utilisation of DEMs allows the derivation of a volume for individual icebergs and has allowed this PhD project to provide comprehensive iceberg area-to-volume convertors for studies working with two-dimensional satellite imagery. Greenland-wide analysis has revealed each sector of the ice sheet calves relatively similar iceberg sizes, with pronounced differences in the larger size classes exceeding 10000 m3. Currently, it is not possible to derive a direct relationship between iceberg sizes and source glacier bed topography, as well as translating a glacier’s solid ice discharge into a predicted iceberg volume size due to data availability. When scaling to the largest marine-terminating glacier on the southwest coast of Greenland, Kangiata Nunaata Sermia (KNS) we find a complex fjord environment where iceberg decay is dominated by melt based processes, rather than fracture. These findings have implications for both remote sensing and modelling studies, and suggest icebergs in this region at the ice front are not necessarily best described as the typically assumed power law, rather lognormal is the best suited distributional fit. Icebergs calved at KNS tend to be small in size, particularly in August (2013 to 2017) when runoff is prominent and subglacial channels form, causing localised plume upwellings and enhanced melting of only a few sections of the ice front. For the first time, this PhD has been able to provide an ice-sheet-wide dataset of three-dimensional iceberg outlines which can be used to constrain and infer glacier calving behaviour and the subsequent implications of iceberg sizes on fjord environments.

Item Type: Thesis (PhD)
Divisions: Faculty of Science and Engineering > School of Environmental Sciences
Depositing User: Symplectic Admin
Date Deposited: 19 Sep 2023 10:53
Last Modified: 19 Sep 2023 10:53
DOI: 10.17638/03172587
Supervisors: [error in script]
URI: https://livrepository.liverpool.ac.uk/id/eprint/3172587