The accuracy of mid-Pliocene δ18O-based ice volume and sea level reconstructions

Understanding the sensitivity of the polar ice caps to a modest global warming (2-3 °C above preindustrial) is of paramount importance if we are to accurately predict future sea level change, knowledge that will inform both social and economic policy in the coming years. However, decades of study of the Pliocene (2.6-5.3 Ma), an epoch in recent Earth history characterized by atmospheric CO2 levels similar to today, have so far failed to provide definitive sea level and ice volume estimates for that time. Here we review the sources of uncertainty in the paired Mg/Ca-δ18O methodology used to estimate past sea level, ice volume, and ocean temperature, as well as discuss common assumptions that may bias our interpretation of ocean geochemical records including the LR04 benthic-δ18O stack, a global compilation of 57 oxygen isotope records that forms the standard template of climate change history for the last 5.3 million years.