Modelling water isotopes in polar ice sheets

Abstract

Concentrations of water isotopes in marine sediments are a key indicator of past temperatures to whom understands the storage capacity and dynamical behaviour of the ocean, atmosphere and cryosphere and their exchanges. The contribution of the latter reservoir remains poorly established because of the paucity of deep ice cores in Greenland and Antarctica and the difficulty of interpreting these cores. To obtain the water isotope composition of ice sheets and gain an understanding of their stratigraphy, I develop a tracer transport method that I combine with numerical models of ice dynamics to predict the fine layering of polar ice such that it is locally validated at ice core sites. This sets a novel method to constrain reconstructions of ice sheets' climatic and dynamic histories. Using the UBC 3D Ice Sheet Model (ISM) of Marshall and Clarke (1997), I predict the fine layering of the Greenland Ice Sheet and use the ice core records of GRIP, Dye 3 and Camp Century to better determine the minimal ice extent during the last inter-glacial period, 127 kyr ago. I suggest that 3. 5-4. 5 m of sea level rise could be attributed to melting in Greenland. Tracers are also applied to Antarctica with the LGGE-ISM of Ritz et al. (2001). The 3D model is compared to simple flow models at the deep ice core sites of Dome C, Vostok and Dome Fuji to test the hypotheses on depositional conditions and dynamical conditions. These studies lead to a well-constrained stratigraphic reconstruction of the Greenland and Antarctic Ice Sheets and allow me to produce the first-ever self-consistent prediction of their bulk isotopic composition, hence closing the global water isotope budget of the Earth.