Mass Extinction

Why do we study past biodiversity, and is it applicable to the current 'biodiversity crisis'?

It is no great secret that the Earth is on the brink of a large-scale extinction. The extinction of both species and populations is becoming a major concern in the scientific and policy domains for social, environmental and economic reasons. Although genuinely documented recent species or population extinctions are relatively limited (when compared to the Permian-Triassic mass extinction, in which an estimated 96% of species went kaput, around 252 million years ago), it is extremely likely that any figure is a vast under-estimate of actual extinction rates, as we simply do not have any rigorous estimate for how many species there are currently on this planet.

Marine biodiversity holds some of the most enigmatic
and beautiful forms we currently know. Image Credit: Richard Ling (Wikimedia commons)

Species don't just die out - new species are coming into being all the time, through evolutionary processes. If the rate at which species are becoming extinct is at least 75% higher than the rate at which new species are appearing (speciation), traditionally over a period of 2 million years or so, then palaeontologists call this a "mass extinction". Seeing as this is quite irrelevant to current processes happening in just tens to thousands of years, some would prefer to adapt this definition to something more relevant to modern biodiversity issues.

The IUCN (International Union for Conservation of Nature) have undertaken the most recent and thorough extinction status analysis, but this is limited to only 2.7% of the nearly two million named extant species (species we know to be alive today). This low figure calls into question any claims that we are entering a mass extinction, as the analysis simply doesn't cover enough species for us to be sure it gives a comprehensive picture of what is happening. 

Nevertheless, there are mounting concerns that environmental, ecological, and climatic factors are being perturbed in a way that will be substantially detrimental to global biodiversity, issuing the onset of a new, and 6th, 'biodiversity crisis' or mass extinction.

Past biodiversity – what do we know?

If we can produce a holistic picture of the forces that govern biodiversity, and use this as a basis to predict the future, this can provide grounds for conservation efforts and minimise the impacts of the way in which we are altering Earth's many natural systems.

There are many areas of biodiversity-focussed research in Palaeontology at the present, two of the most prevalent being in the Quaternary (a period of time spanning from 2.6 million years ago to the present) looking at the interaction of the expansion of early humans and climate on terrestrial mammals and selectivity of extinction, and the Mesozoic (250 to 65 million years ago), focussing on diversity patterns in marine and terrestrial vertebrates and the effect of anthropogenic and geological megabiases (e.g., heterogeneous sampling, amount of available fossiliferous rock). Both of these are variably focussed on organisms or groups of organisms that have direct modern ancestors (e.g., birds), or those which don't (e.g., marine reptiles such as plesiosaurs). What the biases tend to do is skew our observations of patterns of biodiversity over time. There are numerous methods of correcting or standardising biodiversity curves to compensate for this, and is very much an on-going part of statistical macro-scale palaeontological research.

Biodiversity curve during the last 542 million years, based on genera counts as a proxy for standing diversity. Uncorrected for sampling biases, the troughs in biodiversity (extinctions) are quite evident, as is the steadily increasing biodiversity towards more recent times. Image Credit: Wikimedia commons.

Many of the recent studies involving both Mesozoic to recent vertebrates and Phanerozoic (the past 541 million years) invertebrates, observe and interpret biodiversity at the genus level. However, the genus level is an arbitrary taxonomic unit, designed purely for classification purposes and with no biological meaning. The same applies to all higher level taxonomic units considered in this manner (e.g., Family). Species are the only true biologically meaningful unit of comparison, as there has been a considerable devotion of theoretical and empirical research into the biological meaning, albeit still largely unresolved from a wealth of numerous species concepts. One of the reasons for this is that most of the species concepts designed for and applied to modern species cannot be reconciled with fossil organisms due to the lack of relevant information fossils often possess (e.g., biological species concept, and reproductive isolation). What this does mean, therefore, is that despite the uncovering of much about biodiversity in the past, it may or may not be scalable to modern faunas and floras based on the assumption that taxonomy does not accurately reflect relevant biological properties.

Closer to the present day, there is a substantial body of research developing from the analysis of the 'charismatic megafauna' of the Quaternary period (including mammoths), and the impact that human evolution and expansion and characteristic climatic fluctuations had on them. Fossil-bearing deposits from this time are not perfect, but are arguably less influenced by the multitude of biases that otherwise are pervasive throughout the earlier fossil record. Additionally, the resolution of data in terms of both geography and time is much better than for any other period in the fossil record. This preserves the factors affecting the co-evolution of the Earth and its residents through this time with remarkable detail. One of the main issues with this research at present, is the difficulty in teasing apart human-caused and naturally occurring extinctions (e.g., due to climatically induced range size contraction). The advantage however, is that the patterns that can be deduced in this way are more useful for the management of future biodiversity.

Where have all the big mammals gone? Image credit: Mauricio Anton (Wikimedia commons)

Can past, present and future trends be resolved?

This is an extremely difficult question, as currently the data sets we need to compare are hypervariable within and between each other. Extant organisms are sampled unevenly within groups, and all from within one time slice, and fossil organisms are subject to the natural variations of the geological record and taphonomy (preservation), and anthropogenic sampling issues too. Some extant groups, such as birds, mammals and amphibians are extensively sampled, with extinction rates (if scaled) comparable to the previous 'big five' mass extinctions documented in the fossil record. These however only represent a minute fraction of the diversity of modern species, and as higher members of ecosystems are naturally more sensitive to ecological perturbations, and may not accurately reflect what is happening or could happen.

The current episodes of extinction are happening on a scale of years, decades, and centuries. What we're uncovering during the Mesozoic about patterns of speciation and extinction took place over significantly longer periods of time, and are thus largely irreconcilable without the use of proportional scaling methods, which by nature will inflict great uncertainties on datasets. This isn't a bad thing, however. Such studies can tell us about past biodiversity over long periods of time, and the macroevolutionary and macroecological trends throughout, and just because this is difficult to apply to anything meaningful today, does not mean it isn't relevant altogether. 

What can we do?

There is a rift between what sample groups have been studied in the fossil record and what is being assessed for extinction risk status in modern organisms. For example, marine bivalves, echinoids and gastropods have been extensively covered in the fossil record, especially at mass extinction intervals, but are extremely under-studied in modern forms (especially with respect to gastropods, in which there is a focus on terrestrial forms). One of the reasons for this is the preservation potential variation in the fossil record, which is skewed substantially towards near-shore shallow marine deposits, in which many of these organisms would have dwelled, died, and transformed into fossils. With such apparent discrepancies, how can we rise to meet the challenges of assessing extinction risk as combined and complimentary fields of research?

The more data scientists can collect, the better their models
and predictions can get.

One of the answers lies in the collation of large datasets – more data leads to higher resolutions, more rigorous assessments of patterns, and increased accuracy. The Palaeobiology Database or PanTHERIA archive are good examples of progression in this domain. Simply knowing about large scale patterns doesn't help though – future research needs to focus on what the biological and environmental factors are that correspond to extinction, speciation, species’ range contractions or expansions, and a host of other internal ecological dynamics, and how these have interplayed to drive and be driven by macroevolution and macroecology through time (extinction and speciation selectivity). These in turn can provide clues about future responses, if committed to the right temporal or taxonomic scale (e.g., corresponding feeding style based on tooth microwear in specific fossil fishes to a shift in their diversity, and then finding extant functional analogues to which these data can be applied).

We need to be able to convert what we know about extant well-studied and sampled groups into poorly-understood groups. Given the under-sampling and unevenness of sampling, utilising data in this manner is critical, as well as utilising the fossil record in that we might be able to make theoretical predictions, bypassing (or at least mediating) the need for statistical empirical data based on extant organisms. This will require the use of as much data as we can possibly get our hands on, so that we can draw conclusions based on numerous lines of evidence, and that our future conservation efforts are focussed and employed correctly. One of the biggest problems we face is interpreting the impact that humans will have on future ecosystems and biodiversity. We don't have any record of past impacts of the way in which we are currently perturbing Earth systems, and the number of unknowns makes forecasting impacts a nightmare. This will be the biggest challenge that scientists from all relevant fields, be they molecular ecologists, environmental geochemists, or even palaeontologists, have to combine to face as one.

This guest post is by Jon Tennant, currently undertaking a PhD in vertebrate macroevolution at Imperial College, London. Jon usually blogs over at Green Tea and Velociraptors, and co-hosts the Palaeocast  podcast. He can also be found tweeting as @protohedgehog. 

Suggested further reading
Alroy (2003) Global databases will yield reliable measures of global biodiversity, Paleobiology, 29(1), 26-29

Barnosky et al. (2011) Has the Earth's sixth mass extinction already arrived? Nature, 471, 51-57, doi: 10.1038/nature09678

Benton et al. (2011) Assessing the quality of the fossil record: insights from vertebrates, In: Comparing the Geological and Fossil Records: Implications for Biodiversity Studies, Geological Society of London Special Publications, 358, 63-94, doi: 10.1144/SP358.6

Pigot et al. (2012) Speciation and extinction drive the appearance of directional range size evolution in phylogenies and the fossil record, PLoS Biology, 10(2), e1001260, doi: 10.1371/journal.pbio.1001260

Sahney et al. (2010) Links between global taxonomic diversity, ecological diversity, and the expansion of vertebrates on land, Biology Letters, 6, 544-547, doi: 10.1098/rsbl.2009.1024

Smith (2001) Large-scale heterogeneity of the fossil record: implications for Phanerozoic biodiversity studies, Philosophical Transactions of the Royal Society London B, 356, 351-367, doi: 10.1098/rstb.2000.0768