The
memory
of ice
Why we are hunting
for Antarctic ice
over a million years old
Earth has natural climate archives that preserve information about past environments.
These include tree rings, cave deposits, corals, ocean sediments – and ice cores.
Ice cores are cylinders of ice drilled from deep in polar ice sheets and glaciers.
Although difficult to get, ice cores are among the most useful and highest-quality climate recorders.
Air bubbles in a slice of ice core (photo: © Tas van Ommen/AAD)
Air bubbles in a slice of ice core (photo: © Tas van Ommen/AAD)
Falling snowflakes trap dust and trace chemicals in the atmosphere at the time. As snow gets buried and compressed, it forms layers of solid ice with trapped air bubbles.
The layers correspond to years and seasons past, with the youngest ice at the top and the oldest ice at the bottom of the core.
With their archive of entombed air bubbles, particles, and trace chemicals, ice cores are a powerful tool to determine how the Earth’s climate has varied in the past.
When accurately dated, this information provides a long-term context for modern climate change and a testbed for climate models.
The oldest ice on the planet is found in East Antarctica where the ice sheet has persisted for millions of years.
The oldest continuous ice core goes back 800,000 years and was drilled by a European consortium at Dome Concordia.
This timespan offers views across eight glacial cycles (ice ages) and a clear picture of abrupt changes that occurred during the last ice age.
Existing ice core records show a close coupling of carbon dioxide (CO2) and global climate cycles over the last 800,000 years. The ice cores reveal a tight connection between temperature and greenhouse gas levels through the ice age cycles. (figure: Ben Henley and Nerilie Abram)
Existing ice core records show a close coupling of carbon dioxide (CO2) and global climate cycles over the last 800,000 years. The ice cores reveal a tight connection between temperature and greenhouse gas levels through the ice age cycles. (figure: Ben Henley and Nerilie Abram)
However, between 1.2 million and 800,000 years ago, something in Earth’s climate system shifted.
The global climate was pushed into a new regime where the ice ages became longer, and the Northern Hemisphere ice sheets became larger. The shift is called the “Mid-Pleistocene Transition” (or MPT for short).
The MPT marks a shift from ice ages with around 40,000-year durations to ice ages with 100,000-year durations. The cause of the MPT is a vexing and long-standing problem in paleo-climate studies, partly because it’s just out of view from the 800,000-year-old ice core.
We’re hunting for an ice core extending well beyond one-million-years-old to unravel the secrets of the MPT. This oldest continuous ice record will enable us to infer from past interglacial conditions what might happen in the future, as the world warms, and the implications for climate, sea-level rise, and ice-sheet stability.
Below: Three million years of climate records from ice cores and marine sediments show gradually cooling and the Mid-Pleistocene Transition (MPT), or the shift from 40,000-year to 100,000-year ice age cycles. CO2 data with relatively high uncertainties, inferred from ocean sediment proxies, suggest higher concentrations in the 40,000 year world. Atmospheric concentration data from air bubbles in ice cores from the Allan Hills blue ice area are sparse and discontinuous. The Million Year Ice Core aims to complete the record at least to 1.5 million years continuously and extend existing continuous ice core records across the transition.
(figure: Andy Menking)
Why do we care
about past climate?
Compared to Earth’s full history, the length of time where we have direct measurements of climate is extraordinarily short.
For example, we only have direct measurements of carbon dioxide (CO2) in the atmosphere since 1958. Since then, CO2 concentration has risen by 37%, from 315 parts per million (ppm) to 430 ppm (as of April 2026).
This is useful for looking at recent changes, but we need to understand if that rise in CO2 is abnormal in the context of Earth’s past.
Were there times in the past when CO2 was this high? Is CO2 concentration normally stable, or does it rise and fall naturally? And why?
To answer these questions, we can ‘reconstruct’ past climate from the atmospheric ‘time capsules’ in ice cores. These reconstructions allow scientists to study examples of climate change before humans had any influence. This allows us to characterise ‘baseline’ conditions that existed before the intensification of fossil fuel combustion during the Industrial Revolution.
Ice cores also capture natural changes in climate that occurred in the past, like the transitions between ice ages and warm periods. These transitions offer valuable examples of climate change that provide insights into the current changing climate.
Drilling into the last Ice Age
Understanding the processes that drive climate change is crucial for building good models that predict future trends. It is also key to unravelling how much humans are contributing to modern global warming, relative to natural processes.
Extending the ice core record to over one million years ago will make it possible to study what caused the Mid-Pleistocene Transition and help validate climate and earth system models in different background states.
Old ice holds a memory of the ancient atmosphere. The Million Year Ice Core Dome C North borehole will eventually extend over three kilometres deep beneath the ice sheet to the Antarctic bedrock. (photo: © Joel Pedro/AAD)
Old ice holds a memory of the ancient atmosphere. The Million Year Ice Core Dome C North borehole will eventually extend over three kilometres deep beneath the ice sheet to the Antarctic bedrock. (photo: © Joel Pedro/AAD)
Ice core in drill (photo: © Joel Pedro/AAD)
Ice core in drill (photo: © Joel Pedro/AAD)
How do we collect a
Million Year Ice Core?
MYIC drill camp at Dome C North, 2025 (photo: © Damien Beloin/AAD)
MYIC drill camp at Dome C North, 2025 (photo: © Damien Beloin/AAD)
Dome C North drill site for the Million Year Ice Core is 1200km inland from Australia's Casey station in East Antarctica (figure: Andy Menking)
Dome C North drill site for the Million Year Ice Core is 1200km inland from Australia's Casey station in East Antarctica (figure: Andy Menking)
Travelling at a top speed of just 13 km/hr, the convoy of six tractors and two snow groomers takes up to 18 days to travel between Casey station and Dome C North (photo: © Damien Beloin AAD)
Travelling at a top speed of just 13 km/hr, the convoy of six tractors and two snow groomers takes up to 18 days to travel between Casey station and Dome C North (photo: © Damien Beloin AAD)
Million-year-old ice exists very deep in the Antarctic ice sheet. Getting a core involves overcoming significant logistical challenges. The Million Year Ice Core (MYIC) project, led by the Australian Antarctic Program, is drilling an ice core to 3,000 metres depth in three-metre segments, and returning the cores in pristine condition to Hobart, Tasmania.
To do this, the Australian Antarctic Program has developed new tractor-traverse capabilities to tow equipment 1200 km from Casey station to the drill site, and adopted new drilling technology.
The project has also invested in the design and manufacture of specialised insulated boxes to keep cores frozen during transport.
Scientists have worked both domestically and internationally to select the optimal site where there is a high chance for recovery of old ice – at Dome C North. In the summer of 2024-25, the MYIC team established a deep-field camp at the site. The field camp includes the drilling infrastructure, accommodation for researchers and traverse staff, and storage for equipment and ice cores.
Drilling operations commenced in early 2025 with the successful extraction of the first 150 metres of ice, and widening of the borehole, ready for installing a casing. Deep drilling commenced in the 2025-26 season and reached 400m depth at the end of that season.
Due to the extreme cold and remoteness of the Antarctic plateau, drilling can only continue during the Antarctic summer from late November to January. The team expects to reach bedrock in the 2028-29 season.
Australia is not alone in hunting for the oldest ice. Other efforts include the European Beyond EPICA Oldest Ice project, the Japanese Third Dome Fuji Project and the American Center for Oldest Ice Exploration. There are also similar projects proposed by the Chinese, Korean, and Russian Antarctic research programs.
These efforts are mutually cooperative under the International Partnerships in Ice Core Sciences, a SCAR (Scientific Committee on Antarctic Research) expert group. Multiple records are important to replicate and verify the climate and atmospheric records from the very oldest ice.
There are places in Antarctica with non-continuous layers of ice that are several million years old. These give useful complementary data that are snapshots of past climate, but are difficult to date and do not provide the time-continuous record that our three-kilometre ice core is targeting.
What will
we measure?
There has been significant investment by the Australian Antarctic Program to improve laboratory capabilities in advance of obtaining the Million Year Ice Core. The core laboratory is operated as a partnership between the Australian Antarctic Division (the Department of Climate Change, Energy, the Environment and Water) and the University of Tasmania.
It specialises in ice core analysis, using state-of-the-art instruments that measure different chemical components in the ice. This state-of-the-art lab, supported by the Australian Antarctic Program Partnership, is one of only a few of its kind in the world.
Expected data products include:
- A continuous, comprehensive, and high-resolution record of atmospheric greenhouse gases including carbon dioxide (CO2), methane (CH4), and nitrous oxide (N2O) to understand links between climate and the carbon and nitrogen cycles, including the role of greenhouse gases in forcing the Mid-Pleistocene Transition.
- Carbon isotopes of CO2, which are different versions of the same element and useful tracers to identify sources and sinks in the carbon cycle and screen for microbiological contamination of CO2 measurements.
- Climate-sensitive indicators such as water isotopes that create unique ‘fingerprints’ for water sources.
- Major ions, trace elements, aerosols to characterise various aspects of atmospheric composition including ash from volcanic eruptions.
- Dust particles to quantify Southern Hemisphere dust fluxes and sources.
- Records of other atmospheric gases like nitrogen, oxygen, and their isotope ratios to understand the rate at which snow transforms into glacial ice and traps bubbles at Dome C North, date the ice core and study productivity and water cycling in the terrestrial biosphere.
- Noble gas ratios to quantify changes in mean ocean temperature.
The full suite of data products will allow us to study the role of greenhouse gases in forcing the Mid-Pleistocene Transition (MPT), as well as changes in Antarctic climate, the atmosphere, and biogeochemical cycles in the 40,000-year versus 100,000-year worlds.
Andy Menking in the ice core gas lab with the isotope instrument (photo: © Wendy Pyper/AAD)
Andy Menking in the ice core gas lab with the isotope instrument (photo: © Wendy Pyper/AAD)
Andy Menking and Daniel Baggenstos sublimating a Law Dome ice sample with known gas concentrations in the final phase of testing before measurements commence with the Million Year Ice Core. Sublimation is the process where ice samples are turned directly into water vapour without melting into a liquid first. In this way, the gases trapped in the ice core air bubbles can be extracted for analysis. (photo: © Mark Horstman/AAPP)
Andy Menking and Daniel Baggenstos sublimating a Law Dome ice sample with known gas concentrations in the final phase of testing before measurements commence with the Million Year Ice Core. Sublimation is the process where ice samples are turned directly into water vapour without melting into a liquid first. In this way, the gases trapped in the ice core air bubbles can be extracted for analysis. (photo: © Mark Horstman/AAPP)
What will a
Million Year Ice Core
tell us?
The Million Year Ice Core will provide key information about Antarctica, global climate, and the stability of ice sheets and global sea level, beyond the window viewable with existing ice cores. The MYIC will:
- Improve our understanding of the natural climate variability that led to our current climate state.
- Enable better assessment of the likely course our climate will take in the next few centuries to millennia in the absence of human interference.
- Quantify numerous examples of the natural (pre-Industrial Revolution) relationship between greenhouse gases and climate, allowing us to deduce the underlying physical rules.
- Resolve key questions about the timescales and processes that control the exchange of carbon dioxide (including excess CO2 from human activities) between ocean, land, and atmosphere reservoirs.
A 1.5-million-year ice core record will shed new light on the remaining uncertainties in understanding the ice ages – namely, how and why glacial cycles end, the role of ice sheet instability, and the relationship between climate and the carbon cycle.
Australia’s effort is one of several to get a continuous ice core through the MPT. The international ice core community has long recognised that we need more than one record to extract the best information and confirm results.
Drill containing ice core rises from borehole (video: © Chelsea Long)
Drill containing ice core rises from borehole (video: © Chelsea Long)
Contributors: Andy Menking, Daniel Baggenstos, Joel Pedro, Mark Horstman, Wendy Pyper
Produced by Mark Horstman, AAPP Communication and Impact Manager
Thanks to the Australian Antarctic Division and MYIC team for imagery
© Australian Antarctic Program Partnership and Australian Antarctic Division, 2026
|
|
|
|
Australian Antarctic Program Partnership
Institute for Marine and Antarctic Studies
University of Tasmania
PO Box 367, Hobart Tas 7001
Email: aapp.enquiries@utas.edu.au
aappartnership.org.au
Built with Shorthand