Under-ice chemistry that could stall a planet’s thaw
A new geochemical modeling study from the Earth-Life Science Institute (ELSI) at the Institute of Science Tokyo argues that Earth’s deep-freeze “snowball” episodes were not chemically dormant. Instead, rock-water reactions continued beneath thick continental ice sheets, quietly consuming atmospheric carbon dioxide (CO₂) and stretching the lifetime of global glaciations. The work repositions subglacial weathering from a curiosity to a climate feedback with system-scale consequences, adding a missing piece to how planets can both enter and escape hard‑ice states.
Researchers put a missing feedback on the table
Our results demonstrate that subglacial weathering represents a previously unrecognised feedback mechanism that could account for the dramatically different durations of Neoproterozoic snowball Earth events,
says Shintaro Kadoya, lead author of the study and a Specially Appointed Assistant Professor at ELSI, Institute of Science Tokyo.
Multiple globally extensive glaciations punctuated the Neoproterozoic, most prominently the Sturtian and Marinoan events. Geological evidence shows they occurred under broadly similar boundary conditions, yet their durations diverged sharply. The new modeling indicates that CO₂ drawdown beneath ice sheets can counter volcanic outgassing for long stretches, delaying deglaciation-a plausible reason the Sturtian episode endured far longer. By turning what was assumed to be a “paused” carbon cycle into an active, under‑ice sink, the work helps explain why broadly similar forcings can still yield very different climate outcomes.
How subglacial weathering keeps working through a snowball
Even when continents are entombed under kilometers of ice, geothermal heat and the insulating properties of ice generate basal meltwater. That water percolates through comminuted, freshly crushed rock at the glacier bed, enabling silicate and carbonate weathering reactions. The result is alkalinity production and net CO₂ consumption-chemistry that typically regulates Earth’s climate in ice-free times continuing in the shadows of an all‑ice world. In effect, a hidden layer of the climate system remains switched on long after the surface appears frozen solid.
Minerals that depend on continental weathering, such as dolomite, appear in parts of the rock record associated with these glaciations. That signal aligns with a picture of persistent under‑ice reactions and meltwater export to the ocean, with knock‑on effects for nutrients once ice margins retreat. The study suggests that, far from being a short-lived curiosity, subglacial weathering can leave discernible fingerprints in both basin‑scale sedimentary successions and the chemistry of post‑glacial seas.
Inside the modeling: what was actually simulated
The team constructed numerical geochemical models of water-rock interaction under thick continental ice, exploring the balance between meltwater supply and delivery of fresh, reactive rock by erosion. A stable chemical state emerges when that balance holds, largely independent of absolute fluxes-a structure that naturally explains why small hydrological or erosional shifts could swing climate outcomes.
- Environment: Subglacial settings at the ice-bed interface, with basal meltwater generated by geothermal heating and insulated by overlying ice.
- Core processes: Silicate and carbonate weathering kinetics; secondary mineral formation; evolution of pH, alkalinity, and dissolved species.
- Controls explored: Meltwater flux; erosion rate and grain-size effects; residence time; geothermal heat flow; bedrock lithology.
- System behavior: CO₂ consumption can approach volcanic outgassing under favorable conditions, materially slowing greenhouse buildup and deglaciation. In model runs where erosion and meltwater are sustained, the planet’s route out of global ice becomes significantly longer and more sensitive to small changes in boundary conditions.
Why this matters for today’s climate models and carbon markets
Earth system models calibrate long‑term climate sensitivity and carbon cycle feedbacks partly from deep‑time constraints. Strengthening the case for active under‑ice weathering changes how those constraints are read-especially the timescales needed to accumulate CO₂ to end a global glaciation. That, in turn, feeds into how policymakers and regulators interpret “natural” carbon sinks when setting long‑duration climate targets.
- Model parameterization: Baseline weathering “shutoff” assumptions during extensive glaciation may need revision in integrated assessment and Earth system models, including those used to inform national climate strategies and scenario work by central banks and financial supervisors.
- Policy relevance: Claims about the permanence and pacing of geochemical carbon removal, including enhanced rock weathering, depend on how weathering responds to cold, water‑limited regimes. As governments implement long‑term net‑zero pathways under the Paris Agreement, a more accurate view of low‑temperature weathering improves how “durability” is defined in standards for engineered removals.
- Infrastructure planning: Long‑horizon climate risk tools used by grid, water, and transport operators benefit from improved representation of low‑temperature geochemistry that governs background carbon sinks. For regulators overseeing critical infrastructure resilience, subtle shifts in baseline assumptions about Earth’s natural buffering capacity can change how extreme or long‑lasting certain tail‑risk scenarios appear.
Signals to watch in the rock record and oceans
- Bedrock provenance and mineral authigenesis: Assemblages consistent with sustained subglacial fluid flow, including minerals precipitated from long‑lived, chemically evolving meltwaters.
- Carbonate textures and isotopes: Fabrics and carbon-oxygen isotope signatures consistent with alkalinity delivery from under‑ice reactions, distinguishable from purely open‑ocean precipitation.
- Nutrient export: Phosphorus and silica fluxes from basal meltwater that can prime post‑glacial productivity pulses, potentially visible as sharp increases in organic carbon burial after ice retreat.
Co‑author Mohit Melwani Daswani adds, This finding challenges a central assumption of the classical snowball Earth hypothesis by showing that weathering can continue beneath ice sheets and significantly influence climate.
If borne out in the field, that influence would need to be folded into how next‑generation climate assessments translate deep‑time behavior into future risk envelopes.
Computational and data integrity considerations
- Reproducibility: Clearly versioned reaction kinetics, thermodynamic databases, and erosion-hydrology coupling are essential for independent verification, especially as these models increasingly feed into public climate assessments and advisory processes.
- Sensitivity analysis: Formal exploration of uncertainty in geothermal gradients, bed roughness, and meltwater residence times helps bound CO₂ drawdown estimates and prevents overconfidence in any single realization.
- Cross‑checks: Coupling to sediment transport and reactive transport models can reconcile modeled fluxes with measured mineral assemblages and trace‑element ratios, tightening the link between code, rock record, and policy‑relevant narratives about Earth’s climate stability.
Planetary science crossover
Subglacial weathering is not just an Earth story. The same hydrochemistry informs habitability assessments for icy worlds with basal water or brine flow. Understanding how rock-water reactions proceed beneath ice strengthens mission strategies that sample plume material or sub‑ice oceans, where life’s energy sources may hinge on similar chemistry. For space agencies and science advisory bodies weighing investments in missions to ocean worlds, this work adds another argument that icy crusts can host long‑lived, energy‑rich environments.
Neoproterozoic snowball timeline at a glance
| Event | Approximate timing (million years ago) | Notes |
|---|---|---|
| Sturtian glaciation | ~717-659 Ma | Older, markedly longer global glaciation under broadly similar boundary conditions; new work suggests sustained subglacial CO₂ sinks may have extended its duration. |
| Marinoan glaciation | ~651-635 Ma | Later global glaciation with substantially shorter duration, potentially reflecting different balances between volcanic outgassing, subglacial weathering, and surface reflectivity. |
What would validate the mechanism next
- Targeted field campaigns at paleo‑subglacial interfaces to recover mineralogical and isotopic fingerprints of active under‑ice chemistry, ideally across multiple paleocontinents.
- Coupled climate-ice-geochemistry simulations to test whether subglacial CO₂ sinks reproduce observed deglaciation thresholds and durations when embedded in full Earth system models.
- Quantitative comparison of modeled meltwater solute fluxes with sedimentary records across multiple continents, narrowing the plausible range of under‑ice weathering intensities.
Together, these lines of evidence would move the mechanism from “plausible explanation” toward a quantitatively constrained component of the long‑term climate toolkit that informs governance and risk management decisions.
Institutional context and where to read more
The Institute of Science Tokyo was formed on October 1, 2024, merging Tokyo Medical and Dental University with Tokyo Institute of Technology, with a mission to advance science and human wellbeing. ELSI operates as a World Premier International research center, drawing global collaborators to interrogate Earth’s origins and life’s co‑evolution. Details of the peer‑reviewed paper are available via the journal’s DOI, and information about the institute can be found through the Earth‑Life Science Institute (ELSI). As governments, regulators, and scientific bodies refine long‑range climate and space‑science roadmaps, this kind of deep‑time, under‑ice chemistry is quietly resetting expectations about how resilient-or fragile-planetary climates can be.
