Fossil fuel, mercury and the end-Palaeozoic catastrophe

Siberian flood-basalt flows in the Putorana Plateau, Taymyr Peninsula, Russia. (Credit: Paul Wignall)

The end of the Permian Period (~252 Ma ago) saw the loss of 90% of marine fossil species and 70% of those known from terrestrial sediments: the greatest known extinction in Earth’s history. In their naming of newly discovered life forms, palaeontologists can become quite lyrical. Extinctions, however, really stretch their imagination. They call the Permo-Triassic boundary event ‘The Great Dying’. Why not ‘Permageddon’? Sadly, that was snaffled in the 1980s by an astonishingly short-haired heavy-metal tribute band. Enough bathos … The close of the Palaeozoic left a great many ecological niches to be filled by adaptive radiation during the Triassic and later Mesozoic times. Coinciding with the largest known flood-basalt outpouring – the three million cubic kilometres of Siberian Traps – the P-Tr event seemed to be ‘done and dusted’ after that possible connection was discovered in the mid 1990s. Notwithstanding, the quest for a gigantic, causative impact crater continues (see: Palaeobiology Earth-logs, May, September and October 2004), albeit among a dwindling circle of enthusiasts. The Siberian Traps are suitably vast to snuff the fossil record, for their eruption must have belched all manner of climate-changing gases and dusts into the atmosphere; CO2 to encourage global warming; SO2 and dusts as cooling agents. There is also evidence of a role for geochemical toxicity (see: Nickel, life and the end-Permian extinction, June 2014). The extinctions accompanied not only climate change but also a catastrophic fall in atmospheric oxygen content (see: Homing in on the great end-Permian extinction, April 2003; When rain kick-started evolution, December 2019). Recovery of the biosphere during the early Triassic was exceedingly slow.

Research focussed on the P-Tr boundary eventually uncovered an element of pure chance. Shales in Canada that span the boundary show major, negative δ13C excursions in the carbon-isotope record that coincide with fly ash in the analysed layers. This material is similar in all respects to that emitted from coal-fired power stations (see: Coal and the end-Permian mass extinction, March 2011). The part of Siberia onto which the flood basalts were erupted is rich in Permian coal measures and oil shales that lay close to the surface 252 Ma ago. The coal ash and massive emissions of CO2 may have resulted from their burning by the flood basalt event. Now evidence has emerged that this did indeed happen (Elkins-Tanton, L.T. et al. 2020. Field evidence for coal combustion links the 252 Ma Siberian Traps with global carbon disruption. Geology, v. 48, early publication; DOI: 10.1130/G47365.1).

The US, Canadian and Russian team found large quantities of burnt coal and woody material, and bituminous blobs in 600 m thick volcanic ashes at the base of the Siberian traps themselves. They concluded that the magma chamber from which the flood basalts emerged had incorporated sizeable volumes of the coal measures, leading to their combustion and distillation. This would have released CO2 enriched in light 12C due to isotopic fractionation by biological means, i.e. its δ13C would have been sufficiently negative to affect the carbon locked up in the Canadian P-Tr boundary-layer shales that show the sharp isotopic anomalies. The magnitude of the anomalies suggest that between six to ten thousand billion tons of carbon released as CO2 or methane by interaction of the Siberian Traps with sediments through which their magma passed could have created the global δ13C anomalies. That is about one tenth of the organic carbon originally locked in the Permian coal measures beneath the flood basalts

Another paper whose publication coincided with that by Elkins-Tanton et al. suggests that environmental mercury appears to have followed the same geochemical course as did carbon at the end of the Palaeozoic Era (Dal Corso, J. and 9 others 2020. Permo–Triassic boundary carbon and mercury cycling linked to terrestrial ecosystem collapse. Nature Communications, v. 11, paper 2962; DOI: 10.1038/s41467-020-16725-4). This group, based at Leeds and Oxford Universities, UK and the University of Geosciences in Wuhan, China, base their findings on biogeochemical modelling of the global carbon and mercury cycles at the end of the Permian. Their view is that the coincidence in marine sediments at the P-Tr boundary of a short-lived spike in mercury and an anomaly in its isotopic composition with the depletion in 13C, described earlier, shows an intimate link between mercury and the biological carbon cycle in the oceans at the time. They suggest that this synergy marks ecosystem collapse and derives ‘from a massive oxidation of terrestrial biomass’; i.e. burning of organic material on the land surface. Their modelling hints at huge wildfires in equatorial peatlands but also a role for the Siberian flood-basalt volcanism and the incorporation of coal measures into the Siberian Trap magma chamber.

Verneshots (huge volcanic gas blasts) ten years on

One of the most daring hypotheses of modern geosciences: is that of the ‘Verneshot’ reported by Earth Pages in 2004.  Jason Phipps Morgan and colleagues explored the possible consequences of a build-up of volatiles in plume-related magmas at the base of thick continental lithosphere beneath cratons, prior to the eruption of continental flood basalts. They suggested that pressure would eventually result in an explosive release at a lithospheric weak point, followed by collapse above the plume head that would propagate upwards, at hypersonic speeds. Modelling the forces involved, the authors of the novel idea considered that they would be sufficient to fling huge rock masses into orbit.  Verneshots might neatly explain the circumstances around mass extinctions, such as their coincidence with continental flood basalt events; large impact structures, most likely at the antipode of the event; global debris layers containing shocked rock, melt spherules; unusual element suites and compounds (including fullerenes); and enough toxic gas to cause biological devastation.

Ten years on, Verneshots are back, again in the prestigious journal Earth and Planetary Science Letters, and this time among the co-authors are Morgan père et fils (W. Jason a founder of plate tectonics, and Jason P. who launched the idea). This time the yet-to-be –accepted hypothesis comes with evidence of an extremely unusual and fortuitous kind (Vannucchi, P. et al. 2015. Direct evidence of ancient shock metamorphism at the site of the 1908 Tunguska event. Earth and Planetary Science Letters, v. 409, p. 168-174). The origin of the paper lies in an attempt to verify reports of shocked quartz in samples collected close to the centre of the 2000 km2 devastation that resulted from what is now accepted to have been a comet or asteroid air-burst explosion in June 1908 in the Tunguska region of Siberia. Apart from a disputed 300 m crater in the area, the Tunguska Event left no long-lived sign: it ‘merely’ knocked over millions of trees. However, its epicenter lay in a 10 km depression ringed by hills, that has been suggested to be a volcanic centre associated with the end-Permian Siberian Traps.

Trees knocked down and burned over hundreds of square km by the 1908Tunguska Event (credit: Leonid Alekseyevich Kulik deceased)
Trees knocked down and burned over hundreds of square km by the 1908 Tunguska Event (credit: Leonid Alekseyevich Kulik deceased)

The reported shocked quartz locality turned out to associated with an isolated occurrence of quartz-rich sand and rounded clasts of quartzite that contains sedimentary structures. The occurrence is surrounded by basalts of the Siberian Traps, yet is situated topographically above them. The quartzite is thought to be Permian terrestrial sandstone that commonly underlies much of the remaining extent of Siberian Traps.

Quartzite clasts do indeed contain shocked quartz, together with pseudotachylite glass veinlets, quartz and feldspar crystal growth on sedimentary grains and silica-rich glassy spherules. These features are not uniquely diagnostic of shock metamorphism, but are oddly absent from the surrounding Siberian Traps nearby, which suggests that whatever formed them predated the final eruptive stages of the end-Permian large igneous province. Indeed it would be unlikely that airburst of some extraterrestrial bolide in 1908 could produce the metamorphic features of the quartzites without setting ablaze the trees that it felled. A second possibility, that the Tunguska Depression is a Permo-Triassic impact crater and the quartzites being part of an associated central uplift runs into the unlikely coincidence of lying less than 5 km from the 1908 epicentre.

A third hypothesis is that the Tunguska Depression is a massive diatreme associated with a Verneshot. Another odd association lies 8 km to the south of the epicentre, a carbonatite that is one of many, along with smaller pipe-like structures all possibly linked to magmatic gas escape. The Tunguska Event, a mighty puzzle in its own right, may perhaps be eclipsed. Will silence return as it did after the original Verneshot hypothesis was published? Quite possibly, but another quirk about the Siberian Traps was reported by Earth Pages in mid-2014. In a contribution to a link between this massive end-Permian volcanic effusion and the Permian-Triassic mass extinction it was noted that in the Chinese sedimentary repository of evidence for the extinction there is an isolated spike in the abundance of nickel  that is almost certainly of volcanic origin, but only the one when repeated flood basalt events perhaps ought to have led to a series of nickel anomalies. One huge volcanic gas release as the Siberian Traps were building up?