Update: Can a supernova affect the Earth System?

Earth-pages asked this question in August 2020 because it had been suggested that at least one mass extinction – the protracted faunal decline during the Late Devonian – may provide evidence that supernovas can have deadly influence. The authors of the paper that I discussed proposed mass spectrometric analysis of isotopes, such as 146Sm, 235U and  244Pu  in sediments deposited in an extinction event to test the hypothesis. In the 14 May issue of Science a multinational group of geochemists and physicists, led by Anton Wallner of the Australian National University, report detection of alien isotopes in roughly 10 million-year-old sediments sampled from the Pacific Ocean floor (Wallner, A and 12 others 2021. 60Fe and 244Pu deposited on Earth constrain the r-process yields of recent nearby supernovae. Science, v. 372, p. 742-745; DOI: 10.1126/science.aax3972).

Many of the chemical elements whose atomic masses are greater than 56 form by a thermonuclear fusion process known as rapid neutron capture – termed the ‘r-process’ by physicists. This requires such high energy that the likely heavy-element ‘nurseries must be events such as supernovas and/or mergers of neutron stars. The iron and plutonium isotopes  detected at very low concentrations are radioactive, with half-lives of 2.6 Ma for 60Fe and 80.6 Ma for 244Pu. That makes it impossible for them to be terrestrial in origin because, over the lifetime of the Earth, they would decayed away completely. They must be from recent, alien sources either in our galaxy or one of the nearby galaxies. In fact two ‘doses’ were involved. The authors make no comment on any relationship with marine or continental extinctions at that time in the Miocene Epoch

Supernova at the start of the Pleistocene

This brief note takes up a thread begun in Can a supernova affect the Earth System? (August 2020). In February 2020 the brightness of Betelgeuse – the prominent red star at the top-left of the constellation Orion – dropped in a dramatic fashion. This led to media speculation that it was about to ‘go supernova’, but with the rise of COVID-19 beginning then, that seemed the least of our worries. In fact, astronomers already knew that the red star had dimmed many times before, on a roughly 6.4-year time scale. Betelgeuse is a variable star and by March 2020 it brightened once again: shock-horror over; back to the latter-day plague.

When stars more than ten-times the mass of the Sun run out of fuel for the nuclear fusion energy that keeps them ‘inflated’ they collapse. The vast amount of gravitational potential energy released by the collapse triggers a supernova and is sufficient to form all manner of exotic heavy isotopes by nucleosynthesis. Such an event radiates highly energetic and damaging gamma radiation, and flings off dust charged with a soup of exotic isotopes at very high speeds. The energy released could sum to the entire amount of light that our Sun has shone since it formed 4.6 billion years ago. If close enough, the dual ‘blast’ could have severe effects on Earth, and has been suggested to have caused the mass extinction at the end of the Ordovician Period.

Betelgeuse is about 700 light years away, massive enough to become a future supernova and its rapid consumption of nuclear fuel – it is only about 10 million years old – suggests it will do so within the next hundred thousand years. Nobody knows how close such an event needs to be to wreak havoc on the Earth system, so it is as well to check if there is evidence for such linked perturbations in the geological record. The isotope 60Fe occurs in manganese-rich crusts and nodules on the floor of the Pacific Ocean and also in some rocks from the Moon. It is radioactive with a half-life of about 2.6 million years, so it soon decays away and cannot have been a part of Earth’s original geochemistry or that of the Moon. Its presence may suggest accretion of debris from supernovas in the geologically recent past: possibly 20 in the last 10 Ma but with apparently no obvious extinctions. Yet that isotope of iron may also be produced by less-spectacular stellar processes, so may not be a useful guide.

There is, however, another short-lived radioactive isotope, of manganese (53Mn), which can only form under supernova conditions. It has been found in ocean-floor manganese-rich crusts by a German-Argentinian team of physicists  (Korschinek, G. et al. 2020. Supernova-produced 53Mn on Earth. Physical Review Letters, v. 125, article 031101; DOI: 10.1103/PhysRevLett.125.031101). They dated the crusts using another short-lived cosmogenic isotope produced when cosmic rays transform the atomic nuclei of oxygen and nitrogen to 10Be that ended up in the manganese-rich crusts along with any supernova-produced  53Mn and 60Fe. These were detected in parts of four crusts widely separated on the Pacific Ocean floor. The relative proportions of the two isotopes matched that predicted for nucleosynthesis in supernovas, so the team considers their joint presence to be a ‘smoking gun’ for such an event.

The 10Be in the supernova-affected parts of the crusts yielded an age of 2.58 ± 0.43 million years, which marks the start of the Pleistocene Epoch, the onset of glacial cycles in the Northern Hemisphere and the time of the earliest known members of the genus Homo. A remarkable coincidence? Possibly. Yet cosmic rays, many of which come from supernova relics, have been cited as a significant source of nucleation sites for cloud condensation. Clouds increase the planet’s reflectivity and thus act to to cool it. This has been a contentious issue in the debate about modern climate change, some refuting their significance on the basis of a lack of correlation between cloud-cover data and changes in the flux of cosmic rays over the last century. Yet, over the five millennia of recorded history there have been no records of supernovas with a magnitude that would suggest they were able to bathe the night sky in light akin to that of daytime. That may be the signature of one capable of affecting the Earth system. Thousands that warrant being dubbed a ‘very large new star’are recorded, but none that ‘turned night into day’. The hypothesis seems to have ‘legs’, but so too do others, such as the slow influence on oceanic circulation of the formation of the Isthmus of Panama and other parochial mechanisms of changing the transfer of energy around our planet

See also: Stellar explosion in Earth’s proximity, eons ago. (Science Daily; 30 September 2020.)

Can a supernova affect the Earth System?

The easy answer is yes, simply because chemical elements with a greater relative atomic mass than that of iron are thought to be created in supernovae when dying giant stars collapse under their own gravity and then explode. Interstellar dust and gas clouds accumulate their debris. If the clouds are sufficiently dense gravity forms clumps that may become new stars and the planets that surround them. Matter from every once-nearby supernova enters these clouds and thus contributes to the formation of a planet. This was partly proven when pre-solar grains were found in the Murchison meteorite, some of which are as old as 7.5 billion years (Ga) – 3 Ga older than the Solar System (see: Mineral grains far older than the Solar System; January 15, 2020). Murchison is a carbonaceous chondrite, a class of meteorite which likely contributed lots of carbon-based compounds to the early Earth, setting the stage for the emergence of life. It has been estimated that a near-Earth supernova (closer than 1000 light years) would have noticeable effects on the biosphere, mainly because of the effects on atmospheric composition of the associated high-energy gamma-ray burst. That would create sufficient nitrogen oxides to destroy the ozone layer that shields the surface from harmful radiation. There are reckoned to have been 20 nearby supernovae during the last 10 Ma or so from the presence of anomalously high levels of the isotope 60Fe in marine sediment layers on the Pacific floor. Yet there is no convincing evidence that they coincided with detectable extinctions in the fossil record. But supernovae have been suggested as a possible cause of more ancient mass extinctions, such as that at the end of the Ordovician Period (but see: The late-Ordovician mass extinction: volcanic connections; July 2017).

Diorama of an Early Devonian reef with tabulate and rugose corals and trilobites (Credit: Richard Bizley)

The Late Devonian is generally accepted to be one of the ‘Big Five’ mass extinction events. However, unlike the others, the event was a protracted decline in biodiversity, with several extinction peaks). In particular it marked the end of Palaeozoic reef-building corals. Some have put down the episodic faunal decline to the effects of species moving from one marine basin to another as global sea levels fluctuated: much like the effects of the ‘invasion’ of the coral-eating Crown of Thorns sea urchin that has helped devastate parts of the Great Barrier Reef during present-day global warming (see: Late Devonian: mass extinction or mass invasion? January 2012). Recently, attention has switched to evidence for ultra-violet damage to the morphology of spores found in the strata that display faunal extinction; i.e. to the possibility of the ozone layer having been lost or severely depleted. One suggestion has been sudden peaks in volcanic activity, hinted at by spikes in the abundance of mercury of marine sediments. Brian Fields of the University of Illinois, with colleagues from the USA, UK, Estonia and Switzerland, have closely examined the possibility and the testability of a supernova’s influence (Fields. B.D. et al. 2020.  Supernova triggers for end-Devonian extinctions. Proceedings of the National Academy of Sciences, v. 117, article 202013774; DOI: 10.1073/pnas.2013774117).

They propose the deployment of mass-spectrometric analysis for anomalous stable-isotope abundances in the sediments that contain faunal evidence for accelerated extinction, particularly those of 146Sm, 235U and the long-lived plutonium isotope 244Pu (80 Ma hal-life). They suggest that the separation of the extinction into several events, may be a clue to a supernova culprit. A gamma-ray burst would arrive at light speed, but dust – containing the detectable isotopes –  although likely to be travelling very quickly would arrive hundred to thousands of years later, depending on the distance to the supernova. Cosmic rays generated by the supernova, also a possible kill mechanism, given a severely depleted ozone layer, travel about half the speed of light. Three separate arrivals for the products of a single stellar explosion are indeed handy as an explanation for the Late Devonian extinctions. But someone needs to do the analyses. The long-lived  plutonium isotope is the best candidate: even detection of a few atoms in a sample would be sufficient proof. But that would require a means of ruling out contamination by anthropogenic plutonium, such as analysing the interior of fossils. But would even such an exotic discovery prove the sole influence of a galactic even?