Extinctions in ecological communities and the assessment of extinction risk

Identifying populations with high threat of extinction is important for deciding how we target conservation effort for individual species and, because those species have different roles in ecosystems, for predicting the ecosystem consequences of species loss. The majority of what is known about mechanisms of extinction focuses on species in isolation. For example, larger bodied species can suffer greater extinction risk, due to lower reproductive rates slowing recovery, and also because they are less abundant. This information about what types of species are prone to extinction, and the underpinning ecological theory, has been extremely important for designing classification schemes and predictors of species extinction risk, such as those used in the IUCN Red List of Threatened Species. Many of these predictors reflect the vulnerability of individual species to effects of environmental change. Such impacts can be direct: for example a small population is more is at high risk of extinction through direct effects of environmental variation on birth and death rates. However, there are also indirect effects that result from the interactions between organisms (e.g. predation, competition, facilitation). Indirect effects can cause environmental impacts on one species to propagate through the others in an ecosystem. In the extreme, theory predicts that a single extinction, perhaps caused directly by an environmental change, can cause a wave of secondary extinctions to spread through a community. To date, a lack of basic research into the effects of interactions on extinction means ecological risk classification may be unable to take into account such effects. It is this problem, of understanding the role of interactions with other species for the extinction times of individual species, which we propose to research. Research into extinctions is difficult. A lack of data on the community, or population, prior to an extinction event, and on the species interactions in the system, hamper attempts to understand the causes of extinctions in natural systems. For this reason we propose an experimental study of extinctions using communities of microbes in the laboratory. Such systems have already contributed greatly to our understanding of extinctions, in part because they allow the observation of long term population dynamics and the replication necessary to quantify stochastic events like extinctions. They also allow close control over the environmental conditions the communities experience. Preliminary work in these systems indicates clear effects of interspecific interactions on species' risks of extinction, and suggests these are ideal experimental systems for investigating mechanisms of extinctions in communities. From this experimental research we will gain a greater understanding of the causes, direct and indirect, of extinction timing. Understanding the importance of the results for the design of ecological risk classification systems will be realised by the collaboration proposed with the CASE partner, the Zoological Society of London (ZSL). ZSL has played a key role in developing classifications of risk that are used by International Agencies such as the WWF, IUCN and CBD. Their evidence based development of risk classification systems has previously involved reference to theory informed by experiments such as those we are proposing. We propose to establish and follow experimental communities, under different environmental scenarios, over the course of several species extinctions, to test whether an established classification system correctly identifies those experimental species that actually went extinct, whether interspecific interactions reduce the predictive power of the classification system, and why poor performance might occur. This will inform us about the utility of existing risk classifications and how they may be improved.