Data Model Reconstruction of the Cenozoic Climate – AMOR

The aim of this proposal is to investigate the carbon-climate-ice sheet couplings through the study of major global transitions between icehouse and greenhouse states that Earth already experienced. Although atmospheric CO2 level is generally proposed as the main driver of these climatic changes, processes controlling variations of atmospheric CO2 levels remain enigmatic and require a good deal of more thinking. Additionally, ice-sheet growth over Antarctica during the Eocene Oligocene Transition ~33.9 Ma (EOT) did not happen in one step but appears to have suffered from several melt and growth events till the Middle Miocene Climate Transition (MMCT, ~14.6 – 13.1 Ma) after which the ice volume in Antarctica may have stabilized. Climate ice sheet models applied to the EOT have shown that the threshold in atmospheric CO2 levels required to initiate the Antarctica glaciation lies between 840 and 700 ppmv depending on the model’s climatic sensitivity. More recent studies have focused on the Antarctica ice-sheet melting events after the EOT, in particular those occurring during the Miocene Climatic Optimum (16 Ma) and during the Pliocene (4 Ma) in a particularly low CO2 context (i.e. between 380 and 500 ppmv). How is it possible to simulate a full Antarctica glaciation at 700 ppm (lower limit for the EOT), while remaining consistent with partial deglaciation at 500 ppm during the Miocene? Does this imply that the continental configuration differences between the EOT and the MMCT result in a lower Earth’s sensitivity to ice-sheets initiation? What could cause the drop in CO2 at the EOT but also the subsequent rebound? Recent papers all point to a 100-kyr-scale CO2 variability across EOT and MMCT but quantitative models explaining these oscillations are still lacking. In order to address these issues, we propose to combine two methods of investigation. First, the acquisition of new data is needed to better quantify environmental changes occurring at the EOT and at the MMCT with a substantial spatial cover. Mollusks from coastal areas of Western Europe, Eastern USA and New Zealand have been targeted for their potential significance to decipher ocean atmosphere feedbacks at play during both key events. New Zealand outcrops are closed to Antarctica and will provide us with the local answer to ice growth while Western Europe and Eastern USA outcrops may inform us on the planetary answer. More importantly, data acquisition in USA and in Europe may reveal potential changes in ocean dynamics in the Atlantic Ocean. Second, the use of numerical tools including ocean atmosphere model, ice-sheet model and marine carbon cycle model is an invaluable source of information. Important advances to understand climate – ice sheet interactions have already been made on these periods. Conversely, the role of the interaction of the primary productivity with oceanographic-induced modifications in response to gateway changes but also to the ice sheet itself has been overlooked.
Three tasks were identified: 1) a D47-Temperature calibration for living mollusks which should lead to a wider utilization of the mollusks fossils as one of the limit was due to the large salinity fluctuations occurring in coastal areas and impeding the use of d180 measured on mollusks as a paleo-thermometer; 2) characterization of coastal temperatures including the seasonality for both EOT and MMCT using mollusks fossil record from three different localities and 3) several sensitivity studies with the IPSL Earth System model mimicking the suite of events (CO2, ice growth …) at the EOT and MMCT to constrain potential feedbacks on CO2 levels related to the marine carbon cycle. Climate simulations will also provide a physical framework to interpret geochemical data from task 2. Our integrated model and data approach should allow the complex web of interconnected processes associated with EOT and MMCT to be untangled.