Boulton, Geoffrey

Hulton, Nick

Hagdorn, Magnus K M

2004-03-10T12:34:49Z

2004-03-10T12:34:49Z

2003-12

The aim of this project is to improve our understanding of the past European and British ice sheets as a basis for forecasting their future. The behaviour of these ice sheets is investigated by simulating them using a numerical model and comparing model results with geological data including relative sea–level change data. In order to achieve this aim, a coupled ice sheet/lithosphere model is developed. Ice sheets form an integral part of the Earth system. They affect the planet’s albedo, atmospheric and oceanic circulation patterns, topography, and global and local sea–level change. In order to understand how these systems work, it is necessary to understand how ice sheets interact with other parts of the climate system. This project does this by simulating ice behaviour as part of the climate system and evaluating model behaviour in relation to evidence of past ice sheets. Ice sheet simulations can be treated with more confidence if they can be evaluated against independent data. A methodology is therefore developed that compares relative sea–level records with simulations of past sea–level which result from modelling past ice sheets with a dynamic, high–resolution thermo– mechanical ice sheet model coupled to an isostatic adjustment model. The Earth’s response to changing surface loads is simulated using both a regional, flat Earth approximation and a global, spherical self–gravitating Earth model. The coupled model is tested by initially simulating the past Fennoscandian ice sheet because of the simpler topographic framework and the quality of geological evidence of past fluctuations against which to evaluate model behaviour. The model is driven by a climatic forcing function determined so that the simulated ice sheet resembles the past Fennoscandian ice sheet as reconstructed from geomorphological evidence. The Fennoscandian climate driver is then transferred to the British Isles to simulate the past British ice sheet. Finally, a non–linear regression technique is used to construct future ice sheet drivers from future sea– level change scenarios to forecast sea–level change around the British Isles during the next glacial cycle. The data used for the inversion procedure is limited to southern Scandinavia. Outside this area, the simulation compares poorly with reconstructions based on geological observations. However, model fit within this region is good and the simulation is also in good agreement with features not used during the inversion process. This approach illustrates the benefit of using a model coupling realistic ice physics to a realistic Earth model to help constrain simultaneously unknowns of Earth rheology and ice thickness. Ultimately, relative sea–level data together with other strands of data, such as geomorphological evidence, and a coupled ice sheet/isostatic rebound model can be used to help infer past climates.

495592106 bytes

8065989 bytes

52670422 bytes

application/octet-stream

application/pdf

application/octet-stream

http://hdl.handle.net/1842/433

en

University of Edinburgh; College of Science and Engineering; School of GeoScience

Reconstruction of the Past and Forecast of the Future European and British Ice Sheets and Associated Sea–Level Change

Thesis or Dissertation

Doctoral

PhD Doctor of Philosophy