This project aimed to improve understanding of the link between millenial- and orbital-timescale changes in Earth's climate. It was funded by the Natural Environment Research Council (NERC) with the grant references - NE/J008133/1 and NE/J009350/1 - which were led by Dr Stephen Barker (Cardiff University) and Dr Andy Ridgwell (University of Bristol).

Earth's climate varies on timescales ranging from decades to tens of millions of years. Once such mode of variability is related to changes in the Earth's orbit around the Sun. This is known as 'orbital-timescale' variability and has characteristic timescales of tens to hundreds of thousands of years, giving rise to the well known glacial cycles of the Late Pleistocene. Superimposed on this glacial-interglacial variability is another mode of climate change, known as 'millennial-scale' climate variability (characterised by changes on a timescale of hundreds to a few thousands of years). Both of these modes of climate variability have received significant scientific enquiry because they involve major changes in global climate and yet both remain enigmatic in their underlying mechanisms. However, studies have suggested that these apparently separate mechanisms may in fact be intimately related. As such, improving our understanding of one should promote understanding in the other. This project investigated the potential role of millennial-scale climate variability in the wider changes associated with glacial-interglacial climate change. Specifically they examined the effects that occur in response to abrupt changes in ocean/atmosphere circulation that may play a role in the transition from glacial to interglacial climate (such as the last deglaciation, which occurred between 20 and 10 thousand years ago).

It is thought that changes in ocean circulation and related atmospheric phenomena can give rise to dramatic temperature fluctuations such as those recorded by Greenland ice cores during the last glacial and deglacial periods. Of note is the corresponding temperature variations recorded across Antarctica, which suggest that the climate system may act like a sort of seesaw; when circulation is strong, Greenland (and north western Europe) is warm and Antarctica cools. A weakened circulation gives rise to cold conditions across Greenland while warming occurs across Antarctica. An important side effect of this so-called 'bipolar seesaw' is that atmospheric carbon dioxide appears to rise every time the circulation is in a weakened state. Of particular relevance to this project is the rise in carbon dioxide that occurred during the last deglaciation, which was associated with a distinct oscillation of the bipolar seesaw. Moreover, several other seesaw oscillations occurred during the last glacial period, which also gave rise to increases in carbon dioxide but did not lead to deglaciation.

The project aimed to find out why certain bipolar seesaw oscillations (terminal oscillations) apparently lead to deglaciation while others (non-terminal oscillations) do not. It asked the question: Is there anything special about these events or is their affiliation with deglaciation merely coincidence? In order to answer to this question they combined quantitative data analysis with state-of-the-art computer models of the climate system. The project analysed climate records spanning several glacial cycles in order to provide a statistical representation of 'terminal' and 'non-terminal' oscillations of the bipolar seesaw. They then used computer models to investigate how the seesaw operates under a variety of background conditions. The ultimate goal was to find out what, if anything, makes terminal oscillations special. In so doing they provided important constraints on the mechanism of deglaciation.

The overall aim of this project was to improve understanding of the link between millennial- and orbital-timescale changes in Earth's climate. The main objective was to quantify the mechanisms and global impacts of the so-called bipolar seesaw, and to determine whether those events associated with glacial terminations are in some respect unusual. In so doing they aimed to improve our mechanistic understanding of glacial terminations. The specific objectives (of equal importance) werere as follows:

(1) To quantitatively characterise terminal and non-terminal oscillations of the bipolar seesaw with respect to key boundary conditions (insolation, ice volume, carbon dioxide)
(2) To isolate those deglacial changes not associated with bipolar seesaw oscillations
(3) To identify the physical characteristics of bipolar seesaw oscillations under different climate states using fully coupled General Circulation Model (GCM) simulations
(4) To examine the carbon cycle response to these changes using a combination of GCM and Earth System Model of Intermediate Complexity (EMIC) experiments