Abstract
Antarctica is one of the largest potential sources of sea level rise under a warming climate. Ice sheet response to warming is governed by internal ice sheet dynamics, complex feedbacks with atmosphere and ocean processes and the bedrock on which it sits. The impact of these processes on ice sheet evolution is challenging to characterise and timescale- dependent. These challenges are a source of uncertainty in projections of ice sheet evolution under anthropogenic climate change, and reconstructions of ice sheet dynamics under past warm climates. This thesis will explore uncertainties in modelling the Antarctic ice sheet (AIS) under 21st century warming and a period of past climate change: the warm Pliocene, using the BISICLES ice sheet model.First, projections of AIS sea level contribution by 2100 are presented, under low and very high emissions scenarios, using a range of climate and ice sheet modelling choices. These find a range of plausible sea level contributions (-53 mm to 125 mm AIS contribution to sea level), dependent on choice of climate forcing and sensitivity of the ice sheet to ocean warming. Next, developments in BISICLES for simulating the AIS in the warm Pliocene are presented. Aspects of the model setup, including Glacial Isostatic Adjustment and hydrology coupled basal sliding, needed for multi-millennial palaeo-ice sheet simulations are added and tested in a 30-member perturbed parameter ensemble of the Pine Island Glacier catchment, under Pliocene forcing. This ensemble highlights the importance of a basal friction coefficient in rate of retreat in the Pine Island Glacier. Following this, uncertainties in climate model forcing and modern initial condition uncertainty are explored for five climate models and two initial ice sheet and bedrock conditions in simulations of the full Antarctic ice sheet. A scoping study of the influence of surface mass balance parameters is also carried out, with a 9 member one-at-a-time sensitivity study. This shows the strong influence of precipitation lapse rate on sea level contribution under Pliocene climate. It also shows that perturbing surface mass balance parameters can have a comparable effect with choice of climate model, when surface mass balance parameters are not perturbed.
Finally, simulations of the AIS during the warm Pliocene are presented that use the modelling approach developed across the rest of the thesis. This section explores a range of climate and ice sheet modelling choices, and compares results to reconstructions of Pliocene ice sheet and sea level change. Uncertainty in model parameters and climate is explored in a 120-member perturbed parameter ensemble. Uncertainty in choice of model initial condition is explored in a 30-member ensemble. Parameter uncertainty under a modern climate is explored in a 30-member ensemble. This work shows a broad range of simulated Pliocene ice sheets driving an overall sea level fall (up to 15.89 m) to large sea level rise (up to 28.27 m), with strong sensitivity to a basal sliding parameter and the initial ice sheet state. It shows strong sensitivity of simulated sea level contribution to parameter perturbations in the control ensemble, compared with sea level contribution relative to a single control simulation. A subset of simulations are able to reproduce sea level change compatible with Pliocene reconstructions, and a smaller subset show a regional retreat in the Wilkes basin indicated in geological records of AIS margin in the Pliocene.
Across this thesis, we show the vulnerability of the West Antarctic ice sheet to ocean melt-driven mass loss under warm 21st century and Pliocene forcing. However, we also find that warming-driven increases in accumulation in the interior can offset sea level fall, when sensitivity to ocean warming is low. Under some modelling scenarios, this can drive an overall sea level fall both over the 21st century and with millennium-scale Pliocene warming.
Date of Award | 1 May 2023 |
---|---|
Original language | English |
Awarding Institution |
|
Supervisor | Tamsin Edwards (Supervisor) & Lauren J. Gregoire (Supervisor) |