Using Isotope-Enabled Climate Simulations to Investigate CO2- and Orbitally- Driven Oxygen Isotope Variability in the Early Eocene

Periods of extreme warmth and elevated atmospheric CO2 levels, like the Early Eocene, are often used as an analog for future climate change. Paleoclimate reconstructions of the Early Eocene can provide important data constraints on the climate and hydrological cycle under extreme warm conditions. Available terrestrial oxygen and hydrogen isotope records of the Early Eocene have been primarily interpreted as the enhanced hydrological cycle associated with the large-scale warming induced by the high atmospheric CO2. However, the orbital-scale variations in these isotope records have been difficult to quantify and largely overlooked, even though orbital changes can contribute to the distribution of solar irradiance and therefore impact temperature and the hydrological cycle. In this study, we fill this gap using water isotope-climate simulation to investigate the relative impact of CO2 and orbital changes on the Eocene climate and water isotopes. We conducted a suite of fully coupled, isotope-enabled Community Earth System Model (CESM1.2) simulations of the Early Eocene with various atmospheric CO2 levels and orbital configurations. The Early Eocene simulations include cases of 3x and 6x preindustrial (PI) CO2 values (~285 ppm), with each case including 4 orbital sensitivity runs: modern orbit, maximum Northern Hemisphere seasonality, maximum Southern Hemisphere seasonality, and minimum global seasonality. We analyze the relative difference between climatic changes resulting from a doubling of CO2 and changes resulting from orbital differences, as well as compare our modeled δ18O and δD seasonal range to various terrestrial δ18O and δD records. Additionally, we use a point-by-point comparison to find the simulation that best matches a proxy record we used for the warmest interval within this time range, the Paleocene-Eocene Thermal Maximum (PETM), to determine the climatic context for this global warming event. Our conclusions consider relevant background into the proxy record, including the preservation state of the proxy, analytical uncertainty, and the relationship between δ18O or δD and environmental context, and illustrates the significance of fully coupled, isotope-enabled climate models when interpreting proxy records in times of extreme warmth.