Relevant to the point made in post #17, concerning how "science reporting" often garbles the original works, I'd like to remind folks that abstracts of almost every scientific paper published in the past decade or two are available online, and the full text of a substantial fraction of the papers are available as well.
It took me all of three minutes with Google to track down the paper described in the article which started this thread, and it turns out that the whole paper is available online:
"Did a gamma-ray burst initiate the late Ordovician mass extinction?", A.L. Melott, B.S. Lieberman, C.M. Laird, L.D. Martin, M.V. Medvedev, B.C. Thomas, J.K. Cannizzo, N. Gehrels, & C.H. JackmanWhen in doubt, consult the original source.Abstract: At least five times in the history of life, the Earth experienced mass extinctions that eliminated a large percentage of the biota. Many possible causes have been documented, and gamma-ray bursts (GRB) may also have contributed. GRB (Mészáros, 2001) produce a flux of radiation detectable across the observable Universe. A GRB within our own galaxy could do considerable damage to the Earth's biosphere (Thorsett, 1995; Scalo & Wheeler, 2002; Dar & DeRújula, 2002). Rate estimates (Thorsett, 1995) suggest that a number of such GRB may lie within the fossil record. The late Ordovician mass extinction shows a water-depth dependent extinction pattern that is a natural result of the attenuation of the strong ultraviolet radiation expected to result from a nearby GRB. In addition, a GRB would trigger global cooling which is associated with this mass extinction.
(The paper is in PDF format ["Portable Document Format"], as are most scientific papers and many other documents online. If you don't already have it, you'll need to install the free Adobe Acrobat Reader.)
The key passage which describes the mechanism producing global cooling is:
The resulting highly reactive products of N2 dissociation form various oxides of nitrogen. Nitric acid greatly exceeding anthropogenic levels is a probable product (Thorsett, 1995). Global cooling is expected (Reid, 1978) from the absorption of visible light by NO2.The paper cited in support of global cooling is:
Reid, G.C., McAfee, J.R., Crutzen, P.J. (1978) Effects of intense stratospheric ionization events. Nature, 275. 489-492.I can't find that full paper online (if anyone's curious they can always go look up the back issue of Nature in any well-stocked library), but I found a couple of synopses:
A supernova remnant shell (SNR) is used to explain the mass extinctions of the past. Shows how a SNR would increase cosmic ray levels reaching the earth, (100-1,000 times present level for several hundred years), increase atmospheric nitrogen levels, and deplete the ozone. Believes the Cretaceous-Tertiary (K-T) extinctions were caused by this phenomenonAnd:
Speculates that the Earth must have passed through the remnant shell of ancient supernovae which would have resulted in periods of harsh environmental conditions. Specifically increased UV radiation, a cooler, drier climate, and reduced photosynthetic activity.Not much detail there, but related Googling turned up this more detailed article on the same recent paper:
Water would protect marine organisms from the heat of a GRB, but not from its other effects, argues Melott's team. Its gamma-rays would convert some nitrogen and oxygen in the atmosphere into nitrogen dioxide, the brownish gas present in urban smog.That helps clarify the short line in the paper under discussion, "Global cooling is expected [cite] from the absorption of visible light by NO2."Nitrogen dioxide would filter out sunlight, turning the skies dark. The cooling effect could trigger an ice age - there is evidence of widespread glaciation 440 million years ago. Nitrogen oxides also cause acid rain and destroy the ozone layer, exposing Earth to more of the Sun's harmful ultraviolet rays.
So there you go. The upper atmosphere would block and absorb a lot of the Sun's energy, preventing it from heating the surface where life resides. And much of the absorbed upper-atmosphere heat would be reradiated back out into space, or directly lost as more higher-energy molecules "leaked" off into space.
Sounds reasonable to me as a mechanism, but of course the remaining question is how much NO2 would have been produced, and how hefty would the Sun-blocking effect have been?
From this good article on the same paper, we get:
Melott estimates that a burst would produce enough of the gas to darken the sky, blotting out half the visible sunlight reaching the Earth.Yeah, that'd ruin your whole millennium...
It also included this handy graphic:
I ran across references to a few other cheery papers on the same topic:
Terry, K. D. & Tucker, W. H. 1968. Biological effects of supernovae. Science 159:421-423. (Estimate that the Earth is exposed to a radiation dose of 500 r [Roentgens -- Ich.] once every 50 m.y. and a dose of 1500 r once every 300 m.y. as a result of supernovae. It goes on to note that most lab animals die after exposures between 200-700 r).The Ordovician extinction being studied in the paper which started this thread was a huge one: 60% of all marine invertebrate genera and 25% of all families went extinct. This also implies that far more than 60% of marine species went extinct, and *way* more than 60% of individual organisms died off (like 90+% of species and 98+% of organisms). Whatever caused that extinction, it was *enormous*, and nothing like a minor climactic change or continental shifts or period of increased volcano activity would qualify. Something like the cosmic gamma ray blast described in the paper is one of the few things big enough to bodyslam the Earth's biosphere like that. (List of mass extinctions)Ruderman, M. A. 1974. Possible consequences of nearby supernova explosions for atmospheric ozone and terrestrial life. Science 184:1079-1081. (Suggests that more than 90% of the Earth's ozone may have been destroyed by supernovae radiation at least a few times during the Phanerozoic).
Clark, D. H., McCrea, W. H. and Stephenson. F. R. 1977. Frequency of nearby supernovae and climatic and biological catastrophes. Nature 265:318-319. (argues that the Earth should (on average) pass close to (within 3262 light years) a supernovae once for each galactic revolution. This pass would alter the ozone layer, reduce greenhouse warming and possibly initiate an ice-age).
Meanwhile, I tripped over this article which describes a paper (full text) which may have identified a "smoking gun" tying a specific cluster of recent supernovae (20 supernova explosions over the past 20 million years) to a specific (different) mass extinction in Earth's history. The Scorpius-Centaurus OB cluster made its nearest approach (130 light-years) to Earth during the period 1.5-3.5 million years ago. And the Pliocene-Pleistocene marine extinction occurred 1.8 million years ago. Further confirming the connection is a layer of the rare isotope of iron, 60Fe in the Pliocene-Pleistocene boundary, which is a) clearly of "recent" origin since 60Fe has a half-life of 1.56 million years and none would be left from Earth's origins, and b) 60Fe is formed in supernova explosions, and c) the amount found on Earth is consistent with a supernova explosion around 100 light years away around 2 million years ago... Finally, the amount and type of marine extinctions at the PP boundary are what one would expect from the amount of Ozone-layer destruction such a supernova would cause. So supernova involvement in Earthly extinctions seems like it might be more common than previously thought. Fascinating.
I looked to see if there was a related cooling of the Earth around 2 million years ago, but since it occurred in the middle of the existing Pleistocene ice-age, it's hard to say. It might have made the existing ice age deeper or more prolonged, but that wouldn't be as obvious as the start of a temperature dip in the middle of an otherwise non-ice age period.
Yet another related find: Supernova poised to go off near Earth .