On the identification of an Early Pliocene time slice target for data-model comparison

Following the first Pliocene Model Intercomparison Project (PlioMIP), data-model comparisons showed that regionally there were significant discrepancies between predictions of climate. Uncertainties in the prescribed climate model forcings and in proxy data were shown to limit our ability to truly evaluate the models. In part this uncertainty was due to the proxy reconstructions being based on a time slab (of ~300,000 years). Haywood et al (2013) identified a time slice within the Piacenzian that presented a target for modellers and the data community for reconstructing a mid-Pliocene climate and environment. This time slice centred on 3.205 Ma (3.204–3.207 Ma) is a warm interval characterized by a negative benthic oxygen isotope excursion (0.21–0.23‰) centred on marine isotope stage KM5c (KM5.3). It occurred during a period of orbital forcing that was very similar to present day. This time slice became the key target for the second Pliocene Modelling Intercomparison Project (PlioMIP2) and data reconstructions led by the USGS PRISM Group and the PAGES PlioVAR group. 

As part of the planning for PlioMIP3, the community is considering whether a simulation in the Early Pliocene (5.3 – 3.6 Ma) would be a useful target for modelling groups. To this end, we show plausible time slices during the Early Pliocene that have similar to modern orbital forcing.  As derived from the astronomical solution (Laskar et al 2004), we use the annual insolation pattern at the top of the atmosphere to determine which time slices have the most similar to modern orbital configuration.  We consider 10 different solutions between 3.6 and 5.0 Ma alongside the Lisiecki and Raymo (2004) and Ahn et al (2017) benthic oxygen isotope records and the record of magnetic reversals (Gradstein et al (2020)).  We identify two negative isotope excursions that could provide a target for proxy data collection to compare to a PlioMIP3 Early Pliocene simulation.