Dynamic Greenland ice sheet driven by pCO2 variations across the Pliocene Pleistocene transition

Ning Tan,Paul Bachem,Christophe Dumas,Jean-Baptiste Ladant,Gilles Ramstein,Eystein Jansen

doi:10.1038/s41467-018-07206-w

Abstract

It is generally considered that the perennial glaciation of Greenland lasting several orbital cycles began around 2.7 Ma along with the intensification of Northern Hemisphere glaciation (NHG). Both data and model studies have demonstrated that a decline in atmospheric pCO2 was instrumental in establishing a perennial Greenland ice sheet (GrIS), yet models have generally used simplistic pCO2 constraints rather than data-inferred pCO2 evolution. Here, using a method designed for the long-term coupling of climate and cryosphere models and pCO2 scenarios from different studies, we highlight the pivotal role of pCO2 on the GrIS expansion across the Plio-Pleistocene Transition (PPT, 3.0–2.5 Ma), in particular in the range between 280 and 320 ppm. Good qualitative agreement is obtained between various IRD reconstructions and some of the possible evolutions of the GrIS simulated by our model. Our results underline the dynamism of the GrIS waxing and waning under pCO2 levels similar to or lower than today, which supports recent evidence of a dynamic GrIS during the Plio-Pleistocene.

Highlights

It is generally considered that the perennial glaciation of Greenland lasting several orbital cycles began around 2.7 Ma along with the intensification of Northern Hemisphere glaciation (NHG)
This 50 kyrs glaciation was followed by the wellestablished Mid-Pliocene Warm Period (MPWP) from 3.3 Ma to 3.0 Ma for which numerous observational and modelling studies performed in the framework of the PlioMIP project[19] have demonstrated that the warmer conditions led to a reduced GrIS20–22 (Supplementary Fig. 1)
For pCO2 levels above 320 ppm, the ice sheet extent remains limited across the whole 3–2.5 Ma interval. The results of these constant simulations are in good agreement but go one step beyond previous transient experiments at constant pCO237 in that we demonstrate that the Greenland ice sheet (GrIS) possesses a dynamism on orbital timescales across the PPT for only a narrow range of atmospheric pCO2 concentrations

Summary

Introduction

It is generally considered that the perennial glaciation of Greenland lasting several orbital cycles began around 2.7 Ma along with the intensification of Northern Hemisphere glaciation (NHG). It marks the beginning of a low pCO2 world with perennial ice sheets in both hemispheres, an infrequent occurrence in the Earth’s history[12], thereby creating specific geologic and climatic conditions allowing the development of glacial/interglacial cycles. Lunt et al.[23] demonstrated, based on a series of fully coupled GCM snapshot experiments with different forcing factors, that pCO2 decline was the major driver of the GrIS glaciation This result was obtained from equilibrium simulations at 2.7 Ma and the underlying GCM simulations included a pre-existent Pliocene GrIS21. We apply various reconstructions of pCO2 evolution across the PPT from both proxy records and inverse modelling studies (e.g. ref. 27,28,32–35), as well as constant pCO2 evolutions

Methods

Results

Conclusion