Identifying the drivers of stochastic Dansgaard–Oeschger scale variability in a GCM

The driver of the Dansgaard-Oeschger (DO) events remains uncertain, in part because many models do not show similar behaviour of a climate system tipped into a DO oscillatory state. Here we present results from glacial simulations of the HadCM3 GCM that show stochastic DO-scale variability. This is driven by variations in AMOC strength in response to North Atlantic salinity fluctuations. This represents a salt oscillator, driven by the salinity gradient between the subtropical gyre and Nordic seas. We give a mechanistic explanation of the feedbacks that drive this oscillator, particular the interplay between surface fluxes and advection. We identify that the key trigger that pushes the system into this oscillatory mode is the height of the North American ice sheet, which alters atmospheric and subsequently ocean circulation patterns. Our results highlight that glacial background conditions and ice sheet height act to push the system past a tipping point and into an oscillatory state on a timescale comparable to the DO events.