The PlioMIP2 Mid-Pliocene Climate Simulations Using the NASA-GISS ModelE

The Mid-Pliocene Warm Period (MPWP) ca. 3.2 Ma provides an opportunity to explore an equilibrium climate state under pCO₂ and solar insolation conditions similar to those that may govern near-term future climate as per the IPCC’s SSP1 “Sustainability” scenario, which limits warming to +3.0 °C by the year 2100. Here we discuss mid-Pliocene simulations using the NASA-GISS-E2.1 coupled ocean-atmosphere model conducted for the Pliocene Model Intercomparison Project Phase 2 (PlioMIP2). We include our results using both modern and mid-Pliocene paleogeography, in accordance with PlioMIP2 protocols and provide comparisons to the results from PlioMIP1.

The PlioMIP2 simulations do not result in a significant change in global mean surface air temperature or global mean sea surface temperature compared to PlioMIP1. Global mean warming was +2.2 °C in PlioMIP1 and is slightly lower at +2.1 °C in PlioMIP2. However, there are notable contrasts in how temperatures respond regionally and the northern and southern high latitude temperature changes in PlioMIP2 show a marked contrast with PlioMIP2 producing less warming in the Arctic and more warming in the Southern Ocean compared to PlioMIP1. There is significant improvement in the simulation of Pliocene SSTs in PlioMIP2 in the region south of the Labrador Sea, where the PlioMIP1 simulation was anomalously cool. However, the region of greatest warming in the North Atlantic moves too far north, thus the SST warming in the Greenland Sea is only half what was seen in PlioMIP1 simulations weakening correlations to proxy data at key core sites in that region.

The GISS GCM has the strongest AMOC amongst the PlioMIP2 GCMs at 33 Sv, as was the case in the PlioMIP1 study, but the GISS model also has the largest increase in Pliocene AMOC strength compared to its preindustrial control run indicating that the intense overturning in GISS PlioMIP2 Pliocene simulations is, at least in part, due to changes in the response of the CMIP5 vs CMIP6 versions of the GCM. Similarly strong AMOC changes are found in the preindustrial control run comparisons of the GISS ModelE CMIP5 vs CMIP6 runs. As with other models in the PlioMIP2 study, the effect of AMOC on ocean heat transports is limited. Regardless, the increased sensitivity in the Pliocene in our model is impacted by changes in the freshwater flux and salinity distribution resulting from the PlioMIP2 paleogeographic changes, specifically the effects of closing the Bering Strait and the straits of the Canadian Archipelago connecting the Arctic Ocean to the Labrador Sea. Overall it is difficult to evaluate the impact of these substantial boundary condition changes in the Arctic gateways versus the effects of changes in model physics between CMIP5 and CMIP6, so additional simulations may be required to separate these factors. As with nearly all the PlioMIP models, the GISS ModelE continues to point to the potentially large impacts of CO2 on climate – and generally shows that the long-term sensitivity to increasing CO2 could be world altering.