PhD defense by Mikkel Langgaard Lauritzen

Title: Modeling the Greenland Ice Sheet’s Flow and Mass Loss From Past to Present

Abstract:
The Greenland Ice Sheet is currently losing mass at a rate that threatens to alter the state of the Arctic climate system and force coastlines around the world to retreat. This mass loss rate is expected to change in the future, as it has changed since the Last Glacial Maximum (LGM). During that period, the Greenland Ice Sheet was much larger, extending far beyond its present-day boundaries, while global sea levels were about 130 meters lower than they are today. Computational ice flow modeling provides a method to explore the evolution of the ice sheet across past, present, and future climate scenarios. These simulations are crucial due to their high societal relevance, such as providing accurate sea level rise predictions to secure coastlines effectively. This thesis contains three studies using the Parallel Ice Sheet Model (PISM) to improve the understanding of the Greenland Ice Sheet’s dynamics and mass loss under different climate forcings. 

In the first study, we investigate the influence of inter-annual temperature variability, which is expected to increase in the future, on the Greenland Ice Sheet mass loss. To do this, we force PISM with different realizations of the variability observed in the reanalysis data from NOAA-CIRES from 1851–2014. We find that including the inter-annual temperature variability causes the simulated steady-state ice sheet volume to decrease by 1.9±0.4 cm of sea level equivalent and by 11.5±1.4 cm when the variability is doubled. The sensitivity is most significant in the northern basins, where 40% of the total mass reduction takes place. Our results stress the need to include temperature variability in projections of future mass loss.

In a second study, we calibrate a positive degree day (PDD) model of PISM to the surface mass balance (SMB) of the regional climate model RACMO for the 1960-1989 climatology over Greenland. Using a Markov chain Monte Carlo algorithm we find that the PDD model successfully captures the Greenland-wide integrated SMB using uniform PDD parameters of fs = 7.35±1.61 mm K−1 d−1 and fi = 10.21±2.65 mm K−1 d−1 for snow and ice respectively. The inferred probability density function of PDD parameters is useful for long paleoclimatic simulations, where the mass balance varies with the ice sheet’s topography.

In a third and final study, we investigate the Greenland Ice Sheet evolution from the LGM and throughout the Holocene period to the present day. Measurements of oxygen isotope ratios reveal substantial thinning at four different ice core sites throughout the last 11.7 ka comprising the Holocene period. We use these surface elevation histories to constrain an ensemble of 841 model simulations where we vary critical flow and climate forcing parameters to provide confidence in the modeled ice sheet evolution. We find that the Greenland Ice Sheet has contributed 5.3±0.3 m to the global mean sea level rise since the LGM. Furthermore, we find that the ice bridge that connected the Greenland Ice Sheet to the Innuitian Ice Sheet collapsed 4.9±0.5 ka ago and that the ice sheet is still responding to its Holocene deglaciation. This ongoing response has increased sea levels by 23±26 mm SLE ka−1 in the last 500 years, which should also be considered when making future mass-loss projections.
 

Committee: Associate Professor Anders Svensson (chair), Professor Nanna Bjørnholt Karlsson, Professor Heiko Goelzer

Supervisor: Professor Christine Schøtt Hvidberg