Northern ice sheets may have started the ice ages by rerouting the ocean’s carbon

One of the oldest puzzles in climate science is why ice ages deepen: atmospheric carbon dioxide dropped during glacial periods, and the ocean must have swallowed it. A University of St Andrews team led by Dr. Madison Shankle now proposes that expanding ice sheets in the far north may have set that process off by cooling the North Atlantic and so changing how the Southern Ocean stores carbon. The idea comes from a model simulation, and it is offered as a plausible mechanism, not a measured fact.

Save more, waste less: The Cool Down’s newsletter sends smart, simple changes that help you and the planet in your everyday life.

Subscribe to The Cool Down →

We may earn a commission if you sign up.

The study, with Graeme MacGilchrist, Andrea Burke, Alice Marzocchi, Malte Jansen and James Rae as co-authors, appeared in the Journal of Climate in 2026 as “Isolated Northern Hemisphere Cooling in an Idealized Model Promotes Ocean Carbon Storage and Glacial-like Ocean Features.” Shankle works in the School of Earth and Environmental Science at St Andrews, and the paper is the explainer’s anchor: it shows a route by which northern ice could reorganize the ocean’s carbon budget.

The chain reaction from Arctic ice to Antarctic water

The proposed sequence has five links, as the ScienceDaily summary of the St Andrews work sets them out. Ice sheets growing over the far north cool the nearby North Atlantic. That colder water circulates south toward the Southern Ocean. If the cooling is strong enough, it changes ocean circulation around Antarctica, and the ocean then holds more carbon dioxide beneath its surface. Less carbon dioxide stays in the air, the planet cools further, and the glacial state may intensify.

The point of the chain is that the carbon moves without anything dramatic happening at the surface where the carbon dioxide is stored. Water is rerouted, ventilation changes, and the deep ocean becomes a larger reservoir. Rerouting is the key word, because the carbon in the system stays the same while its address changes from the atmosphere to the abyss.

What an idealized model can and cannot show

The team used an idealized ocean circulation model and cooled only the northernmost latitudes, mimicking the presence of a large ice sheet. That regional cooling produced 74 percent of the carbon dioxide drawdown seen in a separate simulation that cooled the entire planet. Shankle called the result a plausible mechanism connecting two far-flung pieces of the ice-age puzzle, which is careful language for a finding that comes from simulations and not from direct observation.

The word “idealized” carries real meaning. An idealized model strips out much of the real ocean’s complexity so a single cause can be isolated, which is useful for testing a mechanism and weak for proving that it happened. Whether northern ice sheets actually started any particular ice age, as opposed to being able to, remains open, and the route to answering it would be proxy records from sediment cores, the kind of evidence Shankle used in earlier work.

The Southern Ocean’s place in the glacial carbon story

Shankle’s 2025 doctoral thesis, held in the St Andrews research repository under its own DOI, argues that Southern Ocean carbon storage in glacial periods may have removed more than 800 petagrams of carbon from the atmosphere. Its modelling chapters explore links between the northern and southern high latitudes, and one finds that keeping northern temperatures cool while the rest of the ocean warms limits the atmospheric carbon dioxide rise by about 73 percent, close to the 74 percent in the new paper. The thesis is restricted until November 2029.

The same group has pursued other pieces of the puzzle. A St Andrews news item from September 2025 reported that more vigorous Pacific circulation during the ice ages could have resulted in 50 percent less carbon escaping, again in model simulations. The new Journal of Climate paper adds a trigger candidate to that picture of storage and release.

Why should a model with a cooled northern cap matter for anyone thinking about the climate today? The ocean is the largest active store of carbon at the surface of the planet, and the pathways that move carbon in and out of its deep water are the same ones that decide how much of today’s emissions the sea can absorb. Studying how a glacial ocean locked carbon away tests the physics of circulation, ventilation and upwelling that modern climate models also depend on. A mechanism that explains how a modest cooling in the north can alter storage thousands of kilometers south is a check on whether those models respond the way the real ocean does. Shankle’s work sits within a broader effort at St Andrews to quantify how much carbon each ocean basin held during glacial periods, and each new simulation narrows the range of explanations that remain consistent with the ice-core record of atmospheric carbon dioxide.

The unresolved question is timing: the model shows that northern cooling can pull down carbon dioxide, but sorting out whether it came first or followed other changes depends on geological data that the idealized simulation cannot supply.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

Save more, waste less: Subscribe to The Cool Down →

We may earn a commission if you sign up.