Abstract
Greenland ice cores offer a unique opportunity to investigate the climate system behaviour. The objective of this PhD project is to investigate isotope modelling of present- day conditions and conduct model-data comparison using Greenland ice cores. Thus this thesis investigates how the integration of model and data can be used to improve the understanding of climate changes. This is done through analysis of isotope modelling, observations and ice core measurements. This dissertation comprises three projects: (1) Modelling the isotopic response to changes in Arctic sea surface conditions, (2) Constructing a new Greenland database of observations and present-day ice core measurements, and (3) Performance test of isotope-enabled CAM5 for Greenland.
The recent decades of rapid Arctic sea ice decline are used as a basis for an observational-based model experiment using the isotope-enabled CAM model 3, isoCAM3. Results show that δ18O of precipitation is sensitive to local changes of sea ice concentration and sea surface conditions. It is found that the distribution of the sea ice cover and sea surface conditions is essential for the spatial distribution of the simulated changes in δ18O over the Arctic Ocean.
A comprehensive database is created based on ice core and weather station data from Greenland within the period 1890-2014. Present day annual and seasonal mean values are computed for 326 locations in Greenland. Parameterization of the spatial distribution of temperature and δ18O are used to create the first observational-based gridded map of δ18O of precipitation for Greenland and the first gridded map of Greenland temperature, where ice core borehole temperatures are included. The database and gridded maps create a framework for conducting model-data comparison of isotope-enabled GCMs.
The simulation of Greenland isotopes is tested for the isotope-enabled model CAM5. Here the importance of model resolution is investigated. A positive bias of 6-10 ‰ is found for the annual mean δ18O on the central part of the Greenland Ice Sheet. The amplitude of the seasonal cycle of Greenland δ18O is highly dependent on model resolution. This is attributed to a complex interaction of local and large scale processes associated with changes in the intensity and position of large scale atmospheric circulation and synoptic systems in the North Atlantic.
The recent decades of rapid Arctic sea ice decline are used as a basis for an observational-based model experiment using the isotope-enabled CAM model 3, isoCAM3. Results show that δ18O of precipitation is sensitive to local changes of sea ice concentration and sea surface conditions. It is found that the distribution of the sea ice cover and sea surface conditions is essential for the spatial distribution of the simulated changes in δ18O over the Arctic Ocean.
A comprehensive database is created based on ice core and weather station data from Greenland within the period 1890-2014. Present day annual and seasonal mean values are computed for 326 locations in Greenland. Parameterization of the spatial distribution of temperature and δ18O are used to create the first observational-based gridded map of δ18O of precipitation for Greenland and the first gridded map of Greenland temperature, where ice core borehole temperatures are included. The database and gridded maps create a framework for conducting model-data comparison of isotope-enabled GCMs.
The simulation of Greenland isotopes is tested for the isotope-enabled model CAM5. Here the importance of model resolution is investigated. A positive bias of 6-10 ‰ is found for the annual mean δ18O on the central part of the Greenland Ice Sheet. The amplitude of the seasonal cycle of Greenland δ18O is highly dependent on model resolution. This is attributed to a complex interaction of local and large scale processes associated with changes in the intensity and position of large scale atmospheric circulation and synoptic systems in the North Atlantic.
| Original language | English |
|---|
| Publisher | The Niels Bohr Institute, Faculty of Science, University of Copenhagen |
|---|---|
| Number of pages | 128 |
| Publication status | Published - 2016 |