Iron-Choked Oceans Slowed Life’s Recovery After Earth’s Worst Extinction

Roughly 252 million years ago, life on Earth came closer to total annihilation than at any other moment in the past half-billion years. The end-Permian mass extinction wiped out the overwhelming majority of marine species, and the ecosystems that followed were slow, stuttering, and strangely impoverished for millions of years. Scientists have long suspected that oxygen-starved seawater played a starring role in both the killing and the delayed recovery, but pinning down exactly where and when the oceans lost their oxygen has proven stubbornly difficult. A new study published in the journal Climate of the Past now offers one of the most detailed reconstructions yet, drawing on rock records from two sides of the ancient supercontinent Pangaea to reveal a striking story: after the catastrophe, the seafloor of shallow epicontinental seas became dominated not by the toxic, sulfidic conditions many researchers expected, but by an iron-rich, oxygen-free state that quietly starved the oceans of nutrients and throttled the return of life.

The research team, led by Fen Yang and Sen Li of the China University of Geosciences in Wuhan, together with Stephen Grasby of the Geological Survey of Canada, David Bond of the University of Hull, and colleagues, examined two exquisitely preserved rock successions that straddle the Permian-Triassic boundary. The first, at Chibi in South China, sat in equatorial waters of the Tethys Ocean during the Late Permian. The second, at Ursula Creek on the north shore of Williston Lake in British Columbia, lay some thirty degrees north along the northwestern margin of Pangaea, facing the vast Panthalassa Ocean that surrounded the supercontinent. By comparing these palaeogeographically distinct archives, the team could test whether the oxygen crisis was a local quirk or a global phenomenon, and whether its character changed as the extinction gave way to the hothouse world of the Early Triassic.

Deciphering the chemistry of ancient seawater requires proxies, and the study deployed three independent ones in concert. The first involves pyrite framboids, tiny raspberry-shaped clusters of iron sulfide that form in the water column before settling into sediment. Their size is a sensitive oxygen gauge: framboids forming in anoxic, sulfidic waters are typically smaller than five micrometers, while those forming under low-oxygen but not fully anoxic conditions grow larger, often between six and ten micrometers. Where bottom waters are well oxygenated, framboids are absent altogether. The researchers measured more than two thousand framboids from Ursula Creek using scanning electron microscopy, tracking how their size distributions shifted through the extinction interval.

The second and third proxies come from sediment geochemistry. Iron speciation separates the iron locked in pyrite and other reactive phases from the total iron pool, allowing researchers to distinguish between euxinic conditions, where hydrogen sulfide accumulates in the water, and ferruginous conditions, where the water is anoxic but rich in dissolved iron(II) and essentially sulfide-free. Meanwhile, the enrichment of the redox-sensitive trace elements molybdenum and uranium, and especially the ratio between their enrichment factors, provides an independent cross-check: strong enrichment of both metals with a high molybdenum-to-uranium ratio points toward euxinia, whereas moderate enrichment with a low ratio signals oxygen-depleted but non-sulfidic waters. Applying all three methods to ninety-seven samples for iron analysis and one hundred and three for trace elements gave the team an unusually robust picture of redox evolution.

The results tell a two-act story. During the late Changhsingian, the final stage of the Permian, bottom waters at both Chibi and Ursula Creek were predominantly dysoxic, meaning oxygen was present but persistently low. At Ursula Creek, abundant framboids with mean diameters between roughly six and ten micrometers, combined with low molybdenum-to-uranium enrichment ratios, painted a picture of a seafloor limping along under chronic oxygen stress. Intriguingly, an interval in the middle of the Late Permian succession showed intense bioturbation, trace fossils burrowed by organisms that need oxygen to survive, alongside a near-total absence of framboids, indicating that oxygenated refugia existed at least episodically even as the crisis approached. The picture was not one of uniform suffocation but of a patchy, oscillating ocean.

Then, at the extinction horizon itself, conditions at Ursula Creek deteriorated dramatically. The framboid populations shrank, with mean diameters dropping to between 5.3 and 6.5 micrometers, and molybdenum and uranium enrichment factors surged, with molybdenum concentrations in the earliest Triassic Grayling Formation reaching mean values of twenty parts per million, more than twenty-five times the Late Permian average. These signals, the authors argue, implicate seafloor anoxia as a key trigger for the marine extinctions recorded in this Panthalassan margin section. The killing mechanism, in other words, was not a slow decline but an abrupt intensification of oxygen loss synchronized with the biotic collapse.

The most surprising twist came in the aftermath. In the earliest Triassic, both sections recorded a shift to persistently ferruginous conditions, anoxic waters charged with dissolved iron but lacking hydrogen sulfide. At Ursula Creek, this showed up as a striking gap in the pyrite framboid record, a phenomenon the researchers link to sulfide-poor waters, combined with high ratios of highly reactive iron to total iron and elevated sums of oxide- and pyrite-bound iron that exceeded the threshold for ferruginous deposition. At Chibi, the Lower Triassic Daye Formation similarly lacked framboids in its basal beds and showed iron speciation signatures and molybdenum-to-uranium ratios well below one-tenth of modern seawater values, all consistent with iron-rich, oxygen-limited bottom waters. A global compilation of iron speciation data from sections spanning Panthalassa, the Tethys realm, and the Pangaea margin confirmed that this was no regional accident: ferruginous conditions expanded significantly across epicontinental seas in the earliest Triassic, even as euxinic and ferruginous states continued to trade places in a spatially patchy ocean.

Why did the oceans swing toward iron instead of sulfide? The authors propose a combination of mechanisms. Extensive deposition of Late Permian evaporites and intensified pyrite burial drew down the ocean’s sulfate reservoir, starving the microbial processes that generate hydrogen sulfide. At the same time, extreme greenhouse warming stratified the ocean, suppressing vertical mixing and nutrient upwelling, which curtailed primary productivity and reduced the organic carbon flux that fuels sulfate reduction in sediments. Finally, intense chemical weathering of iron-bearing silicate rocks under the hot, carbon-dioxide-rich Early Triassic atmosphere delivered a flood of reactive iron to the sea, tipping the balance decisively toward ferruginous chemistry.

That ferruginous aftermath may hold the key to one of the great puzzles of the extinction: why recovery took so long. Ferruginous waters are nutrient traps. Dissolved iron(II) binds phosphorus, the limiting nutrient for marine productivity, sequestering it into authigenic minerals such as vivianite and green-rust phases and locking it away in sediments. This reduces the recycled phosphorus flux back into the water column, weakening the feedback loop that normally sustains rich plankton blooms. The study points to covariation between redox state and productivity across the extinction interval: where ferruginous conditions prevailed, productivity collapsed, while euxinic settings sometimes sustained higher organic carbon fluxes. In South China, total organic carbon concentrations and accumulation rates plummeted across the extinction, and phosphorus recycling was restricted, even as enhanced continental weathering delivered more detritus to the sea than ever. The contradiction between abundant eroded material and starving seas is resolved, the authors suggest, by recognizing that thermal stratification, disrupted nutrient recycling, redox instability, and lethal heat stress on primary producers combined to keep the marine food web from restarting.

The broader implication is sobering. Oxygen depletion alone may not fully explain either the killing or the delay; instead, the end-Permian crisis may have been a compound disaster in which expanding ferruginous seas imposed simultaneous oxygen and nutrient stress on marine ecosystems. As the modern ocean warms and loses oxygen in response to rising greenhouse gases, the Permian-Triassic record offers a deep-time warning: deoxygenation does not merely asphyxiate marine life, it can rewire ocean chemistry in ways that sabotage the very nutrient cycles life depends on, turning recovery from a sprint into a marathon of millions of years.

Subject of Research: Ocean redox conditions and ferruginous anoxia across the Permian-Triassic mass extinction and recovery

Article Title: Permian-Triassic redox shift and its ferruginous aftermath in epicontinental seas

Article References: Yang, F., Li, S., Grasby, S. E., Bond, D. P. G., Pan, M., & Sun, Y. (2026). Permian-Triassic redox shift and its ferruginous aftermath in epicontinental seas. Climate of the Past, 22(8), 1481-1497. https://doi.org/10.5194/cp-22-1481-2026

Image Credits: AI Generated

DOI: 10.5194/cp-22-1481-2026

Keywords: end-Permian mass extinction, ocean anoxia, ferruginous conditions, pyrite framboids, iron speciation, molybdenum-uranium proxies, Early Triassic, epicontinental seas, phosphorus cycling, biotic recovery, paleoceanography, Climate of the Past