A 2023 survey of 138 groundwaters from 14 aquifers in the Canadian Prairie found that old (thousand- to >10,000-year) groundwaters in organic carbon-rich strata carry up to 1.4×10^7 microbial cells per mL and contain substantial dissolved oxygen (0.52 ± 0.12 mg/L, mean ± SE, n = 57), with oxygen-isotope and mixing-model evidence that at least part of this oxygen is produced in situ ('dark oxygen') by microbes, and that hydrogen is the basal electron donor sustaining methane, sulfur, and nitrogen cycling. The inference, however, rests primarily on community composition and geochemical correlations rather than on direct source apportionment and rate measurements, leaving the central question unresolved: what are the origins, production rates, and geographic extent of the hydrogen and dark oxygen that fuel this productivity, and do they require revising global subsurface biomass estimates?

Background

Groundwater ecosystems — the microbial communities inhabiting water-filled pores and fractures of aquifers — are among the least explored habitats on Earth, yet continental subsurface waters are thought to hold a microbial biomass comparable in scale to that of all surface soils. A long-standing expectation was that microbial cell counts decline as groundwaters age and become geochemically 'evolved' (progressively stripped of electron donors and acceptors), so that deep, old aquifers should be nearly lifeless. Ruff and colleagues overturned this expectation for the Western Canadian Sedimentary Basin: analyzing 138 samples from 95 monitoring wells (<250 m depth) of the Alberta Groundwater Observation Well Network (GOWN), a provincial long-term monitoring program, they showed cell numbers increasing with groundwater age, reaching up to 1.4×10^7 cells per mL in old waters from organic carbon-rich strata such as coal beds and shales, and measured unexpectedly high dissolved oxygen in these nominally anoxic waters. Several specialist terms are needed to follow their evidence. 'Dark oxygen' denotes molecular oxygen (O2) generated in the absence of photosynthesis — here, in groundwater never exposed to sunlight. Candidate microbial mechanisms are 'dismutation' reactions, in which a single enzyme simultaneously oxidizes and reduces the same substrate to yield O2 as a product: chlorite dismutase splits chlorite (ClO2−) into chloride and O2, and nitric oxide dismutase (encoded by nod genes) splits two nitric oxide molecules into N2 and O2; both enzyme families were detected by metagenomics (sequencing of DNA extracted directly from the environment, summarized in metagenome-assembled genomes, MAGs, of uncultured organisms) in the studied wells. Abiotic O2 sources are also possible, notably radiolysis, the splitting of water molecules by ionizing radiation from natural radionuclides. The isotope evidence combines the O2:Ar ratio (argon being an inert tracer of air-equilibrated water) with the oxygen-isotope composition of dissolved O2 (δ18O-O2): air contamination or microbial respiration both shift these tracers in directions opposite to the observed trend, which instead points to mixing with an isotopically light (~−20‰) in-situ O2 source. On the electron-donor side, hydrogen (H2) is invoked as the basal energy source — DNA-based surveys found fermentative organisms and diverse hydrogenases (the metalloenzymes that consume or produce H2), plus hydrogenotrophic methanogens (methane producers), methanotrophs (methane oxidizers), and sulfate reducers. H2 itself was measured only at selected sites (samples GW121, GW223, and GW951) and its origins were never apportioned; its possible sources are numerous: microbial fermentation of buried organic matter, abiotic water–rock reactions (including serpentinization of iron-rich minerals), water radiolysis, or even corrosion of steel well casings. The authors emphasize that all metabolic conclusions rest on gene potential and geochemical correlations, that free-living cell counts omit biofilm (surface-attached) biomass on aquifer solids, and that 'global subsurface biomass may be underestimated, particularly in aquifers containing organic carbon-rich strata.' Adjacent work has since made 'dark oxygen' a prominent theme — abiotic O2 production at the polymetallic-nodule abyssal seafloor and nitrate-driven O2 production by deep-sea microbes — but none of it addresses groundwater.

Problem Statement

The source study inferred, from gene inventories and geochemical correlations, that hydrogen serves as the basal electron donor and that microbially produced 'dark oxygen' contributes to the dissolved O2 pool of ancient Prairie aquifers. The open problem is to convert this inference into quantitative, activity-based understanding, jointly addressing four linked questions that together establish the mechanism, rate, and global significance of this productivity: (1) Origin of the hydrogen — what are the relative contributions of microbial fermentation of buried organic matter, abiotic water–rock reactions, water radiolysis, and anthropogenic sources such as steel well casings to the H2 fueling these ecosystems? (2) Origin of the oxygen — what fractions of the dissolved O2 pool derive from microbial dismutation (chlorite, nitric oxide, or H2O2 pathways), from abiotic radiolysis, and from physical co-migration with younger recharge water? (3) Rates and responsible organisms — what are the in situ production and consumption rates of H2 and O2 (methods such as H2^18O-labeling incubations or enrichment cultures are illustrative, not mandatory), and do the organisms carrying dismutase genes actually produce O2 at environmentally relevant rates, so that genetic potential becomes measured activity for methanogenesis, methane oxidation, sulfate reduction, and oxygen production? (4) Geographic extent and biomass consequence — do other sedimentary basins with organic carbon-rich strata host analogous H2/dark-O2-fueled high-productivity communities (including the biofilm biomass on aquifer solids that free-living cell counts omit), at a scale sufficient to require revising current estimates of global subsurface microbial biomass? Resolving the problem means answering all four questions jointly — each is required in its own right, and no single question alone constitutes resolution — replacing gene-potential and geochemical-correlation inference with quantified, activity-based understanding of the origins, in situ rates, and geographic extent of the H2 and dark O2 fueling these ecosystems. Temporal resilience of these communities to extraction, recharge, or climate perturbations lies outside this problem's scope, which is limited to the origins, in situ rates, and geographic extent of the H2 and dark O2.

References

  1. S. Emil Ruff, Pauline Humez, Isabella Hrabe de Angelis, Muhe Diao, Michael Nightingale, Sara Cho, Liam Connors, Olukayode O. Kuloyo, Alan Seltzer, Samuel Bowman, Scott D. Wankel, Cynthia N. McClain, Bernhard Mayer, Marc Strous. 'Hydrogen and dark oxygen drive microbial productivity in diverse groundwater ecosystems.' Nature Communications 14, 3194 (2023). DOI: 10.1038/s41467-023-38523-4. https://doi.org/10.1038/s41467-023-38523-4
  2. S. Emil Ruff, Pauline Humez, Isabella Hrabe de Angelis, Michael Nightingale, Muhe Diao, Sara Cho, Liam Connors, Olukayode O. Kuloyo, Alan Seltzer, Samuel Bowman, Scott D. Wankel, Cynthia N. McClain, Bernhard Mayer, Marc Strous. 'Hydrogen and dark oxygen drive microbial productivity in diverse groundwater ecosystems.' bioRxiv preprint (2022). DOI: 10.1101/2022.08.09.503387. https://doi.org/10.1101/2022.08.09.503387
  3. Shengjie Li, Damon Mosier, Angela Kouris, Pauline Humez, Bernhard Mayer, Marc Strous, Muhe Diao. 'High diversity, abundance, and expression of hydrogenases in groundwater.' ISME Communications 4, ycae023 (2024). DOI: 10.1093/ismeco/ycae023. https://doi.org/10.1093/ismeco/ycae023
  4. Shengjie Li, Xiaoli Dong, Pauline Humez, Joanna Borecki, Jean Birks, Cynthia McClain, Bernhard Mayer, Marc Strous, Muhe Diao. 'Proteomic evidence for aerobic methane production in groundwater by methylotrophic Methylotenera.' The ISME Journal 19, wraf024 (2025). DOI: 10.1093/ismejo/wraf024. https://doi.org/10.1093/ismejo/wraf024
  5. Beatrix M. Heinze, Valérie F. Schwab, Kirsten Küsel, Stefan Schloemer, Andreas Roskam, Xiaomei Xu, Susan Trumbore. 'Microbial oxidation significantly reduces methane export from global groundwaters.' Proceedings of the National Academy of Sciences 122, e2508773122 (2025). DOI: 10.1073/pnas.2508773122. https://doi.org/10.1073/pnas.2508773122
  6. Xiao Wu, Jianxi Zhu, Hongmei Yang, Yiping Yang, Xiaoju Lin, Xiaoliang Liang, Mang Lin, Barbara Sherwood Lollar, Kurt O. Konhauser, Hongping He. 'Crustal faulting drives biological redox cycling in the deep subsurface.' Science Advances 11, eadx5372 (2025). DOI: 10.1126/sciadv.adx5372. https://doi.org/10.1126/sciadv.adx5372
  7. Rikuan Zheng, Chong Wang, Hongliang Wang, Haibin Qi, Chaomin Sun. 'Nitrate-driven dark oxygen production by diverse deep-sea microorganisms.' bioRxiv preprint (2025). DOI: 10.1101/2025.03.10.642362. https://doi.org/10.1101/2025.03.10.642362
  8. Andrew K. Sweetman, Alycia J. Smith, Danielle S. W. de Jonge, Tobias Hahn, Peter Schroedl, Michael Silverstein, Claire Andrade, R. Lawrence Edwards, Alastair J. M. Lough, Clare Woulds, William B. Homoky, Andrea Koschinsky, Sebastian Fuchs, Thomas Kuhn, Franz Geiger, Jeffrey J. Marlow. 'Evidence of dark oxygen production at the abyssal seafloor.' Nature Geoscience 17, 737–739 (2024). DOI: 10.1038/s41561-024-01480-8. https://doi.org/10.1038/s41561-024-01480-8
  9. Betzabe Atencio, Eyal Geisler, Maxim Rubin-Blum, Edo Bar-Zeev, Eilon M. Adar, Roi Ram, Zeev Ronen. 'Metabolic adaptations underpin high productivity rates in relict subsurface water.' Scientific Reports 14, 18126 (2024). DOI: 10.1038/s41598-024-68868-9. https://doi.org/10.1038/s41598-024-68868-9
  10. Betzabe Atencio, Stas Malavin, Maxim Rubin-Blum, Roi Ram, Eilon Adar, Zeev Ronen. 'Site-specific incubations reveal biofilm diversity and functional adaptations in deep, ancient desert aquifers.' Frontiers in Microbiology 16, 1533115 (2025). DOI: 10.3389/fmicb.2025.1533115. https://doi.org/10.3389/fmicb.2025.1533115
  11. Chunfang Deng, Jinyun Hu, Qian Chen, Shungui Zhou, Jinren Ni. 'Expanded global groundwater microbial diversity reveals bioprospecting potential.' Cell Reports 45, 116760 (2026). DOI: 10.1016/j.celrep.2025.116760. https://doi.org/10.1016/j.celrep.2025.116760
  12. Carolina González-Rosales, Maryam Rezaei Somee, Moritz Buck, Stefan Bertilsson, Maliheh Mehrshad, Mark Dopson. 'A global deep terrestrial biosphere core microbiome.' ISME Communications 5, ycaf176 (2025). DOI: 10.1093/ismeco/ycaf176. https://doi.org/10.1093/ismeco/ycaf176
  13. Rachel C. Beaver, Josh D. Neufeld. 'Microbial ecology of the deep terrestrial subsurface.' The ISME Journal 18, wrae091 (2024). DOI: 10.1093/ismejo/wrae091. https://doi.org/10.1093/ismejo/wrae091
  14. Zongzhi Wu, Daliang Ning, Matthew W. Fields, Terry C. Hazen, Paul D. Adams, Adam P. Arkin, Jizhong Zhou. 'Diversity, biogeography and assembly mechanisms of groundwater microbiomes.' Nature Reviews Earth & Environment (2026). DOI: 10.1038/s43017-026-00813-y. https://doi.org/10.1038/s43017-026-00813-y
  15. Martina Cascone, Guillermo Climent Gargallo, Flavia Migliaccio, Davide Corso, Luca Tonietti, Angelina Cordone, Ilaria Pietrini, Elisabetta Franchi, Giovanna Carpani, Almerinda Di Benedetto, Giuseppina Luciani, Giuseppina Anzelmo, Andrea Giovannelli, David Iacopini, Mariano Parente, Katriona Edlmann, Marco Moracci, Donato Giovannelli. 'Hydrogenotrophic metabolisms in the subsurface and their implications for underground hydrogen storage and natural hydrogen prospecting.' Earth-Science Reviews 281, 105620 (2026). DOI: 10.1016/j.earscirev.2026.105620. https://doi.org/10.1016/j.earscirev.2026.105620