Post-depositional geochemical processes in EPICA Dome C ice: implications for BE-OIC dust analysis

The Beyond EPICA–Oldest Ice Core (BE-OIC) project successfully recovered the oldest continuous Antarctic ice core, extending back to at least 1.2 million years. This landmark achievement provides an unprecedented opportunity to address long-standing questions regarding the mechanisms underlying the Mid-Pleistocene Transition (MPT) (Barbante & Beyond EPICA Team, 2025). Among others, this core could be used to study past changes in atmospheric aerosol composition and here, in particular the geochemical composition of mineral dust.

However, deep ice is increasingly recognized as a “geochemical reactor”, in which primary mineral impurities undergo post-depositional transformations into secondary phases such as jarosite (Baccolo et al., 2021; Lanci et al., 2025). These alterations pose a major challenge for extracting reliable paleoclimate signals from the analysis of mineral dust trapped into old ice. As such, to avoid misinterpretation of dust-related proxy records, we need to better constrain the nature and extent of deep-ice geochemical processes.

Here, we investigate post-depositional geochemical alterations in the EPICA Dome C (EDC) ice core through elemental analysis of ten sections (55 or 110 cm long) spanning the depth of 282 to 3137 m. For the first time, we apply single-particle inductively coupled plasma time-of-flight mass spectrometry (sp-ICP-TOFMS) coupled to the Bern continuous flow analysis (CFA) system to EDC ice core analysis. This approach allows the separate quantification of dissolved and particulate elemental fractions and enables the characterization of the chemical composition of individual particles. Our results reveal extensive dissolution of primary minerals (e.g. hornblende-like phases), accompanied by the precipitation of secondary insoluble and soluble sulfates (e.g. jarosite, alunite), and possibly other Fe-oxide phases. These transformations are likely driven by localized acidic and oxidative microenvironments that develop during the metamorphism of deep ice, depending, to first order, on the growth of ice grains.

Our findings provide new insights into post-depositional geochemical processes in deep Antarctic ice and are crucial for ensuring robust paleoclimate reconstructions from dust records in the oldest ice cores, including BE-OIC. Notably, significant geochemical alteration is observed in EDC sections at temperatures of approximately −15 °C and above, indicating that, at the conditions encountered at EDC, these changes emerge at around −15 °C and intensify under warmer conditions. Given that BE-OIC ice of comparable age to EDC is colder while exhibiting similar dust concentration, the BE-OIC ice core may preserve a less geochemically altered, and therefore higher-quality, dust archive for periods already covered by the EDC record (<800 ka). On the other hand, portions of the BE-OIC stratigraphy extend to substantially greater ages, implying longer residence times in ice and a higher potential for post-depositional alterations.