Numerical ice sheet modeling of Heinrich Events

Hank, Kevin (2024) Numerical ice sheet modeling of Heinrich Events. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (31MB)

Abstract

Layers of Ice-Rafted Debris (IRD) found in sediment cores from the North Atlantic are attributed to quasi-periodic episodes of iceberg discharge from the Hudson Bay/Hudson Strait region. At least six such so-called Heinrich Events (HEs) have been identified during the Last Glacial Cycle (LGC). Due to the associated release of freshwater, HEs are inferred to cause climatic changes on a global scale. Several hypotheses for generating HEs, including an internally driven binge-purge model, an ice shelf buildup-collapse mechanism, and a hypothesis encompassing underwater melt modulated by Glacial Isostatic Adjustment (GIA), have been proposed in the literature. However, a comprehensive study identifying the role of individual system processes such as GIA is still missing. Here, I use 3D thermo-mechanically coupled ice sheet models, primarily the Glacial Systems Model (GSM), to identify the numerically most robust model configuration within a HE framework. Based on these results, I determine the importance of different system processes in a HE context. Model setups with varying complexity are used, including the first transient HE study covering the full LGC. Another major step forward is the first use of a fully coupled sediment model to determine the sediment discharge during HEs. To partly address potential non-linear dependencies of model results on ensemble parameters, all experiments are run with high-variance ensembles instead of a single parameter vector. Model results are then evaluated with respect to numerical uncertainties, a subset of structural uncertainties, and (revised) proxy data estimates (e.g., sediment discharge during a HE). The key takeaways of this study are the relevant physical and numerical sensitivities in a HE context. For example, including sub-temperate basal sliding can reduce the resolution dependence of surge characteristics. However, the surge pattern is highly sensitive to the poorly constrained geothermal heat flux in the Hudson Bay and Hudson Strait. Model results further indicate that sub-surface ocean warming and ice shelves in the Labrador Sea are unlikely to be the main driver of HEs. HE characteristics based on the sediment flux modeled with the most comprehensive GSM setup generally agree with proxy constraints. However, the mid-Hudson Strait ice flux shows only a weak correlation with the sediment flux and should not be used as a metric to identify HEs. This weak correlation indicates that Heinrich Layers do not necessarily record large-scale Hudson Strait surges and questions the previously inferred climatic implications of Heinrich Events.

Actions (login required)

View Item