Interaction of ice sheets and climate on geological time scales

Since the inception of the Antarctic ice sheet at the Eocene-Oligocene Transition (~34 Myr ago), land ice plays a crucial role in Earth’s climate. Through the ice-albedo and surface-height-temperature feedbacks, land ice variability strengthens atmospheric temperature changes induced by orbital and CO2 variations. Quantification of these feedbacks on long time scales has hitherto seldom been undertaken. Limiting factors are insufficient computational power and lack of reliable CO2 data beyond 800 kyr ago. In this thesis, we use a zonally averaged energy balance climate model bi-directionally coupled to a one-dimensional ice sheet model. The relative simplicity of these models allows us to perform fully transient simulations of global climate and all major ice sheets over the past 38 Myrs. First, we run the coupled model in forward mode, forced by prescribed CO2 from ice core data over the past 800 kyr, for benchmarking purposes. We show that the model produces results in good agreement with a recent data reconstruction of the Last Glacial Maximum to pre-industrial temperature difference, as well as several temperature records from ice cores and marine sediment cores. Thereafter, an inverse routine is used to yield CO2 over the past 5 Myr from a benthic δ18O record. Using this inverse model, we simulate Pliocene (5 to 2.5 Myr ago) CO2 levels that are generally higher and more variable than expected from global mean temperature changes, because of reduced ice sheet variability. Our findings therefore also indicate a smaller Earth System Sensitivity. Using the same approach, we simulate the past 38 Myr. We obtain similar CO2 levels just before the Eocene-Oligocene Transition (EOT; ~35 Myr ago) and during the Middle Miocene Climatic Optimum (MMCO; ~15 Myr ago), because the forcing δ18O values are also at comparably high levels. However, proxy data from several sources show higher CO2 before the EOT than during the MMCO. We offer an explanation for this MMCO-EOT conundrum by considering erosion and/or tectonic movement of Antarctica, implementing evolving topography over time. The output of the model is further used to study the influence of ice sheets on the evolution of global temperature and polar amplification by comparing runs with ice sheet-climate interaction switched on and off. We find that ice volume variability has a strong enhancing effect on atmospheric temperature changes, particularly in the regions where the ice sheets are located. As a result, polar amplification in the Northern Hemisphere decreases towards warmer climates as there is little land ice left to melt under those conditions. Conversely, decay of the Antarctic ice sheet increases Southern Hemispheric polar amplification in the high-CO2 regime. Hence, polar amplification changes contrastingly in the Northern and Southern Hemisphere because ice volume changes take place in temperature regimes that are different. We obtain similar results if we use an intermediate complexity climate model. Finally, our results show that in cold climates the ice-albedo feedback predominates the surface-height temperature feedback, while in warm climates they are more equal in strength