The Beyond EPICA – Oldest Ice project (a European collaboration) drilled a 2.8 km (about 1.7 miles) ice core at Little Dome C in East Antarctica, reaching bedrock. It provides the longest continuous record of Earth’s climate and atmospheric composition yet recovered, extending back at least 1.2 million years (with some indications it could approach 1.5 million years in the oldest layers).

Ice cores are uniquely valuable because they provide direct measurements of past atmospheric composition (not just proxies). This one fills a critical gap and offers a longer “blueprint” for understanding natural climate variability, the role of CO₂ in temperature changes, and how the climate system behaves over very long timescales. It doesn’t rewrite fundamentals but adds high-resolution data for a key period in Earth’s recent geological history.

The Mid-Pleistocene Transition (MPT, ~1.25–0.7 million years ago) marks a profound shift in Earth’s glacial cycles: from relatively symmetric, lower-amplitude ~41,000-year (obliquity-dominated) cycles in the Early Pleistocene to the strongly asymmetric, higher-amplitude ~100,000-year cycles (with longer, deeper glaciations) that characterize the Late Pleistocene.

This occurred without any significant change in orbital (Milankovitch) forcing, making it a key puzzle in paleoclimatology. Orbital parameters (eccentricity, obliquity, precession) continued as before; the change involved internal Earth system feedback and a long-term cooling trend.

The article notes that preliminary data are impressive but much analysis remains. Results were presented at the European Geosciences Union assembly on 4 May 2026 (not yet peer-reviewed at the time of publication). Expert Edward Brook (Oregon State University) is quoted praising the ability to now examine CO₂ differences cycle-by-cycle in detail.

One leading hypothesis for the MPT is a drop in atmospheric CO₂ that amplified cooling and ice-sheet growth, but the exact drivers (and the changing role of greenhouse gases) are still under investigation. This core provides the direct gas and temperature data needed to test ideas.

The article is a news summary of ongoing project results — full scientific papers with detailed datasets are expected in the coming months/years as more samples are processed in labs across Europe.

_____________________________________________________________________________________

Ice core reveals longest-ever continuous record of Earth’s climate

The study referenced is not yet a single peer-reviewed paper but a set of preliminary results from the Beyond EPICA – Oldest Ice (BE-OI) project, presented at the European Geosciences Union (EGU) General Assembly in Vienna on/around 4 May 2026. The Nature News article (DOI: 10.1038/d41586-026-01523-7, by Davide Castelvecchi, 12 May 2026) summarizes these early findings.

Location: Little Dome C, East Antarctica (near the older EPICA Dome C site).

Depth: 2.8 km (2,800 m) to bedrock.

Age span: Continuous record back to at least 1.2 million years (potentially up to ~1.5 million years in the deepest, highly thinned layers).

Key contents: Trapped air bubbles (direct atmospheric samples for CO₂, CH₄, N₂O, and isotopes) + ice chemistry/isotopes for temperature, dust, and other proxies.

This extends the previous continuous record (~800,000 years from EPICA Dome C) by ~50%, crucially covering the full Mid-Pleistocene Transition (MPT, ~1.25–0.8 Ma).

The 2.8 km core from Little Dome C provides the first continuous, high-resolution direct record (via trapped air bubbles and ice isotopes) spanning the Mid-Pleistocene Transition (MPT, ~1.2–0.8 million years ago). This is when glacial-interglacial cycles shifted from ~41,000-year (obliquity-driven) to ~100,000-year (eccentricity-influenced, but amplified) periods with greater intensity: longer, colder glaciations and thicker ice sheets.

Key new CO₂ insights (from preliminary analyses, e.g., by Florian Krauss/University of Bern team):

• Around ~950,000 years ago (end of an interglacial), CO₂ spiked by ~50 ppm in just a few thousand years.
• It then plummeted to ~170 ppm — the lowest level in any continuous ice-core record (today’s level is >420 ppm).
• This low point aligns precisely with the onset of the first full 100-kyr cycle.
• Over the broader MPT, average CO₂ didn’t drop dramatically (~20–25 ppm lower), but variability between glacials and interglacials changed, with progressively deeper glacial minima.

Temperature proxies (e.g., δD/δ¹⁸O) track these CO₂ swings closely, reinforcing the tight coupling.

Implications for the MPT Mystery

Supports a major role for carbon cycle feedback: The sharp swings suggest internal Earth system changes (e.g., ocean circulation, carbon storage/release) amplified orbital forcing. Isotopic data (δ¹³C-CO₂) hint at shifts in deep-ocean carbon storage rather than massive outgassing.

Challenges the regolith hypothesis: This idea posited that ice sheets gradually eroded loose sediment until reaching bedrock, allowing thicker, more stable ice. Supporting evidence is weak here: no expected rise in CF₄ (from bedrock exposure), and the transition appears too abrupt.

Reconciles with other records: Aligns with marine sediment proxies (boron isotopes for past CO₂). Discontinuous “blue ice” records (up to millions of years old) show less variability because they often average glacial + interglacial periods.

Analyses are ongoing across ~14 European labs (gases, isotopes, dust, chemistry). Full peer-reviewed papers with detailed datasets are expected soon. The project also recovered bedrock fragments for absolute dating.