A history-matching analysis of Antarctic Ice Sheet evolution since the last interglacial

Lecavalier, Benoit S. (2024) A history-matching analysis of Antarctic Ice Sheet evolution since the last interglacial. 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 (16MB)

Abstract

One technique to explicitly quantify uncertainties of glacial systems is a history-matching analysis (HMA) of a model against a large observational database. This is achieved by ruling out simulations that are inconsistent with an observational constraint database. A comprehensive database (“AntICE2”) was compiled for state-space estimation of past Antarctic Ice Sheet (AIS) changes and to evaluate model reconstructions. This research applies a HMA on a 3D glacial systems model (GSM) for Antarctica against the AntICE2 observational constraint database. A HMA represents a crucial steppingstone towards a comprehensive Bayesian calibration. A HMA consists of identifying model reconstructions that are consistent with observations given uncertainties in the model and data. Our HMA extensively samples model uncertainties against fits to observational data through Markov Chain Monte Carlo methods using Bayesian artificial neural network emulators of the full GSM. This methodology produced several large ensembles exceeding 40,000 simulations that were evaluated against observational constraints. The terminal large ensemble consisting of 9,293 members represents the culmination of this research. The GSM simulation output is scored against the AntICE2 database to evaluate the model reconstruction. The HMA rules simulations as being broadly inconsistent with the AntICE2 database based on being within a 3σ or 4σ threshold of each various observational data type. The simulations from the full ensemble that are tentatively not inconsistent with the observational constraint database are classified as the not-ruled-out-yet (NROY) sub-ensemble. The HMA of the AIS since the last interglacial and the resulting NROY sub-ensemble addresses several outstanding research questions. Considering the extent to which uncertainties across the glacial system and data were incorporated in the HMA, the NROY sub-ensemble should approximately bracket the past evolution of the actual ice sheet. The NROY simulations have excess Last Glacial Maximum (LGM) volumes ranging between 9.2 to 26.5 meters equivalent sea level. This range has upper limits that are considerably higher than past studies and this addresses in large part inferential deficits in the LGM sea-level budget. Moreover, the NROY sub-ensemble represents an envelop of chronologies which can be used as input boundary conditions for general circulation models and glacial isostatic adjustment models to better understand past atmospheric and oceanic circulation, and sea-level change.

Actions (login required)

View Item