The Pleistocene glacial cycles: A dance of ice and climate

During the past few million years, the Earth underwent glacial cycles where large ice sheets intermittently covered parts of North America and Eurasia. The growth of these ice sheets was largely driven by changes in CO2 and long-term incoming solar insolation. At first, glacial cycles took 41 thousand years to complete, but roughly 1 million years ago, the periodicity shifted to on average 100 thousand years. This change in glacial cycle periodicity is the mid-Pleistocene transition (MPT). In this thesis, I investigate the glacial cycles of the past 1.5-million years by simulating the evolution of the ice sheets using a computer model. My goal was to improve the model’s ability to simulate glacial cycles, glacial terminations and ice-climate interactions. I used CO2 and insolation reconstructions to calculate temperature and ice volume changes. The ice-sheet model was able to capture the frequency change at the MPT. Before ~1 million years ago, CO2 was high enough to trigger a deglaciation every 41 thousand years. For the past ~1 million years, low CO2 concentrations were sometimes maintained through periods with strong summer insolation, thus preventing a glacial termination and increasing the length of glacial periods. If CO2 and insolation forcing are strong, the North American ice sheet can melt within 10 thousand years. This is facilitated not only by temperature increase, but also by the creation of lakes at the southern margin of the retreating ice sheet. These lakes accelerate the melt mainly due to the low friction underneath floating ice, increasing ice flow towards the South. More generally formulated: a successful termination depends on whether the climate (e.g., CO2 and insolation) and ice sheets can trigger a full collapse of the ice sheets. Otherwise, the ice sheet may survive until the next period with strong summer insolation and thereby increase the glacial cycle periodicity.