Paleogeographic Influence on Oceans and Climate in Past Warm Periods: Miocene and Pliocene

Abstract

The rapid warming of our climate, driven by human greenhouse gas emissions, is one of the greatest challenges of our time. Although today’s warming is unprecedented in its speed, the planet has experienced warm periods in the past under CO2 levels comparable to those of today. These past warm periods featured a range of climate states including reduced ice sheets, altered precipitation patterns and ocean circulation different from today. Two periods in the past that have been proposed as potential analogues for a long-term future climate are the Miocene Climatic Optimum (MCO), between 16.75 and 14.5 million years ago, and the mid-Pliocene warm period, which was approximately 3.2 million years ago. Both featured a similar continental configuration to today and CO2 concentrations around or above present-day levels. Despite these similarities to the present day, important differences in the paleogeographic boundary conditions remain, including smaller ice sheets, different vegetation and altered regional geography. In this thesis, we study the impact of paleogeographic boundary conditions, including ice sheet size, orography and vegetation, on the climate and ocean circulation during these past warm periods. Mid-Pliocene simulations from the Pliocene Model Intercomparison Project Phase 2 (PlioMIP2) show a stronger Atlantic Meridional Overturning Circulation (AMOC) compared to the pre-industrial. We find that reduced freshwater transport into the Labrador Sea and a lower surface freshwater flux lead to salinification of the subpolar North Atlantic and strengthening of the AMOC across models. This intensification enhances overturning-driven ocean heat transport and likely contributes to warmer subpolar North Atlantic sea surface temperatures, improving agreement between reconstructed and simulated SSTs. Using a single global climate model, we then show that this AMOC strengthening is caused by orographic and vegetation boundary conditions, particularly the closure of Arctic gateways. Paleogeographic boundary conditions also impact Southern Ocean conditions in mid-Pliocene PlioMIP2 simulations. We show that warmer and fresher surface waters in mid-Pliocene simulations increase the density stratification, leading to a slowdown of the global abyssal overturning circulation, though a stronger AMOC partly compensates for this weakening. Approximately half of the Southern Ocean warming and freshening in the mid-Pliocene simulations is a consequence of paleogeographic boundary conditions. Finally, we explore how paleogeographic boundary conditions influenced the MCO climate. We show that paleogeography is the main cause of warming in our simulations, which capture reconstructed MCO warmth at 560 ppm CO2. However, the model fails to simulate the reconstructed reduced meridional temperature gradient. An alternative bathymetry causes AMOC collapse and global heat redistribution, but does not significantly improve agreement with reconstructions. The results emphasize the critical role of boundary conditions in shaping the MCO climate and the limitations of current models in capturing its key features. Overall, we show that paleogeography significantly impacts the ocean circulation and climate in the mid-Pliocene and MCO. While these paleoclimates may share features with a future climate, certain boundary conditions, such as closed Arctic gateways, are not representative of future climate change scenarios. This limits the use of paleoclimates as analogues for a potential future warm climate.