Mass extinctions

A taxonomic group is said to go extinct when it vanishes from the fossil record.  The paleo-record has shown us that there have been five mass extinctions of species on Earth in the last 600 million years.  An event is categorised as a ‘Mass Extinction’ when over 75% of species become extinct in the wild.  A summary of these events can be found in table 1, click on the links for more details (synthesised from Purvis et al., 2000 and Barnosky et al., 2011). 

Table 1           Extinction magnitude of nine marine invertebrate groups at the five Mass Extinction events.  The fossil record of these groups is relatively complete. 

Event	Millions of years before present	Mean extinction severity (families extinct/extant)	Proposed cause	Casualties	Survivors
Late Ordovician	455	0.21 (65/310)	Episodic glacial and interglacial periods; marine transgression; CO2 sequestration	Crinozoans	Poriferans
				Cephalopods	Foraminifera
					Gastropods
Late Devonian	370	0.21 (68/322)	Global warming and cooling following CO2 drawdown caused (possibly) by land plant diversification.  Deep water anoxia.  Possible bollide impacts.	Cephalopods	Bryozoans
				Poriferans	Foraminifera
				Ostracodes	Gastropods
Late Permian	255	0.63 (220/348)	Volcanism; Global warming; Anoxic deep water spread; High marine and terrestrial H2S and CO2; Acidic oceans; Bolide	Crinozoans	Foraminifera
				Anthozoans	Poriferans
				Brachiopods	Bivalves
				Bryozoans	Gastropods
Late Triassic	220	0.10 (18/178)	Central Atlantic Magnetic Province activity elevated CO2; global warming and global marine calcification catastrophe	Cephalopods	Crinozoans
					Bryozoans
					Ostracodes
					Poriferans
Late Cretaceous	75	0.11 (46/425)	Bolide impact in Yucatan peninsula; previous volcanic activity and CO2 increase; techtonic uplift accelerates erosion and leads to ocean eutrophication and marine anoxia.	Cephalopods	Crinozoans
				Poriferans	Anthozoans

To define mass extinction, the background rate (i.e. extinctions per million species years) and magnitude (i.e. percentage of species lost) for a given time period needs to be established.  One of the greatest challenges measuring baseline extinction is the incompleteness of the fossil record.  Geographic sampling bias is a major contributor to the incompleteness.  For example, it appeared that diversity of marine invertebrates showed a 62 Ma cyclicity.  But, on closer inspection this repeated flux of diversity was a sampling artefact caused by periods of non-preservation of fossils.  On this temporal scale (i.e. not glacial/interglacial cycles of thousands of years), it’s been suggested that extrinsic geological or astrological drove these fluctuations, but none were statistically significant. 

Peters et. al. (2005) showed a strong correlation between diversity and sediment deposition, so that more species are found at times of great flooding rather than when sea level is retreating (figure 1).  Figure 1 c shows the relationship between the retreating seas and species diversity.  Two hypotheses are likely to explain this:  Biological (long term flooding of coastal areas increases the area of shallow, marine habitats and so promote the development of large, diverse marine communities which are preserved in the rock record) and Geographical (the amount of rock formed/preserved at times of sea level rise is greater than when the sea is retreating, thus more fossils are preserved during these times).  This has left some corners of the world with few fossil representatives of past biodiversity, with the tropics being particularly under-sampled. Conversely, North America and European records are disproportionately well documented (Smith, 2007). 

Figure 1  Global marine genus diversity v. rock at outcrop record for western Europe (France, England and Wales) since the late Triassic. (a) Log diversity of genera whose taxonomic duration is 45 Ma or less(b) Marine sedimentary rock at outcrop, based on number of 1:63 360 (UK) and 1:80 000 (France) geological maps (c) Detrended plots of genus diversity and rock at outcrop diversity (d) Second-order sequences in western Europe. Figure and legend from Smith (2007). 

There are other biases in the fossil record, and it hasn’t been agreed on which biases make the biggest difference.  I will discuss more of these biases in the next few entries, and then explore how they can be accounted for and corrected to help clarify the fossil record. 

Barnosky, A., et al. (2011), Has the Earth’s sixth mass extinction already arrived?  Nature, 471, 51-57.

Peters, S., (2005), Geological constraints on the macroevolutionary history of marine animals.  Proceedings of the National Academy of Sciences of the USA, 102, 12326-12331.

Purvis, A., Jones, K., Mace, G. (2000), Extinction.  Bioessays, 22, 1123-1133

Smith, A., (2007), Marine diversity through the Phranerozoic: problems and prospects.  Journal of the Geological Society, 164, 731-745