Abstract
Firn is defined as snow that has survived a melt season and provides the link between the high-frequency variability of the atmosphere to the ”slower” reacting ice sheet.In this thesis, firn is described by a theoretical and statistical approach to accommodate the variability in observed firn compaction on ice sheet scales. The modeling objectives are multiple and aim at estimating the contribution from the firn to the observed volume change of the GrIS and to the diffusion of stable water isotopes. The firn modeling then provides crucial information on total mass balance of the GrIS and the paleo-temperature reconstructions retrieved from ice cores.The dynamical firn model developed in this thesis explains13 % of the observed volume change of the GrIS from 2003-2008, without contributing to the global sea-level rise. This emphasizes the need for well constraint firn-compaction models. Here, the validation efforts are conducted by developing a new method of assessing firn compaction over large areas using shallow radar data. The method provides new in-sight in the spatial variability of the firn-compaction rate and suggests the firn compaction to be site-dependent and cannot be parameterized by temperature and accumulation alone, as assumed previous studies.
Ice-sheet model initialization is necessary to constrain projections of future sea-level rise and assess the present state of the GrIS. In this thesis, the initialization of the GrIS is investigated through ensemble studies of present-day ice-sheet configurations formed by the variation of both internal-model parameters and external climate forcing. To investigate the importance of the validation, a multi-metric validation approach is applied to the ensemble members. The validation shows that the commonly used validation measures, such as the total ice-sheet volume, do not capture the full aspect of the present day state of the GrIS. The ensemble of ice-sheet initializations is then used to investigate how future projections of ice sheet evolution depend on the initialization. The further projections show a surprising linearity in the model response to the applied forcing, despite different initial states.
Together, the firn and ice-sheet modeling studies suggest that the GrIS is not in balance with the present climate, and further mass loss should be expected.
Ice-sheet model initialization is necessary to constrain projections of future sea-level rise and assess the present state of the GrIS. In this thesis, the initialization of the GrIS is investigated through ensemble studies of present-day ice-sheet configurations formed by the variation of both internal-model parameters and external climate forcing. To investigate the importance of the validation, a multi-metric validation approach is applied to the ensemble members. The validation shows that the commonly used validation measures, such as the total ice-sheet volume, do not capture the full aspect of the present day state of the GrIS. The ensemble of ice-sheet initializations is then used to investigate how future projections of ice sheet evolution depend on the initialization. The further projections show a surprising linearity in the model response to the applied forcing, despite different initial states.
Together, the firn and ice-sheet modeling studies suggest that the GrIS is not in balance with the present climate, and further mass loss should be expected.
| Original language | English |
|---|
| Publisher | The Niels Bohr Institute, Faculty of Science, University of Copenhagen |
|---|---|
| Number of pages | 211 |
| Publication status | Published - 23 Nov 2012 |