It's easy to think of mass extinctions only in terms of the impact on animal species but of course plants suffered too.
Stephen Hesselbo of Oxford's Department of Earth Sciences reports in this week's Science on his research studying fossils to see how plants fared in a major extinction event 200 million years ago.
I asked him about what these plant fossils can tell us about extinctions, biodiversity and climate change.
OxSciBlog: Why are these Greenland fossils of particular interest to those studying climate/biodiversity?
Stephen Hesselbo: The East Greenland fossil plant beds represent a uniquely detailed record of floral change across one of the 'big five' mass extinctions that characterize the history of life over the last 500 million years - that which occurred at the Triassic-Jurassic boundary, 200 million years ago.
The plant beds, which represent a warm temperate community, have been known about for about a hundred years, but have previously been worked on principally from the point of view of taxonomy and evolutionary relationships.
This new study takes advantage of the fact that a series of plant beds leading up to the mass extinction horizon were laid down under similar conditions on the banks of rivers in flood - changes in plant assemblages can therefore be interpreted in terms of changing ecosystems rather than vagaries of plant preservation.
A series of expeditions have shipped back more than a tonne of fossil material and so the statistics of the present study are based on large numbers.
OSB: What do they tell us about interactions between CO2 levels and plant life in the Triassic?
SH: Previous work on the density of stomata [the holes through which CO2 diffuses to the site of photosynthesis] on the fossil leaves has indicated increases in CO2 over the Triassic-Jurassic boundary interval. These are estimated to have risen from a starting value of about 600 ppm to a maximum of about 1800 ppm through the extinction event.
One effect of the flux of CO2 into the atmosphere is to leave a record in the carbon-isotope ratios of organic matter formed at the time.
Although the Triassic-Jurassic extinction was dramatic amongst animals, hitherto the effects on plants appear to have been much less intense. However, the present study provides evidence for very marked collapse in ecosystem diversity in the run up to the extinction.
OSB: What do we think may have caused the sort of dramatic loss of biodiversity seen in this period?
SH: The Triassic-Jurassic boundary coincides with a period of volcanism on a massive scale - think Iceland but scaled up a hundred to a thousand times. This happened as the super-continent Pangaea broke up (initiating formation of the Atlantic ocean in the process).
As well as directly contributing gasses such as CO2 and SO2 to the atmosphere, it is likely that molten rock baked older deposits of organic-rich sediments and salt to create large fluxes of CO2 and SO2 and other gases such as halocarbons. Other mass extinctions similarly coincided with periods of hugely enhanced volcanic activity.
OSB: Can we draw any lessons from this period about how rising CO2 emissions may affect plant/animal life in the future?
SH: To a certain extent yes - the present day flux rates for carbon are higher than anything that was likely at the Triassic-Jurassic boundary, but the total amount of carbon emitted is considerably less, so far.
Further complications are the unknown quantities and effects of the the additional gases produced during the mass extinction event.
Nevertheless, these so called 'flood basalt eruption' episodes do represent a natural pre-run of exactly the kind of release of fossil carbon, sulphur and other elements that we are currently subjecting the planet to.
OSB: How might further research in this area help us to better understand the impacts of climate change?
SH: This study shows how plant ecosystems responded to environmental change in the interval leading up to a mass extinction. Further work should tackle also the recovery interval, after the peak of the mass extinction, as ecosystems returned to normality.
Additionally we need to better constrain the changes in atmospheric carbon-dioxide that went alongside floral change, and also document similar changes that must have been occurring on different parts of the ancient Earth at the same time.
Finally an improved 'age model' (i.e. chronology of events) would be very useful. With these pieces of information we will be in a better position to describe the deterioration of plant ecosystems in response to greenhouse-gas forcing.
Professor Stephen Hesselbo is based at Oxford University's Department of Earth Sciences. The research was conducted with co-authors Jennifer McElwain of University College Dublin and Peter Wagner of the Smithsonian Institution.