Kevin Smith and Ryan Almeida take a look at how we can predict and manage extinction risk. 

Here’s a thought experiment for you: ask a few people the question, ‘What was it about the dodo that led it to go extinct’? Compare your answers.

Based on our experience, your friends and colleagues are likely to cite some of the dodo’s traits that they feel like must be bad for surviving the filter of natural selection. Being a flightless bird? Bad. Nesting on the ground? Bad. Being large and possibly slow? Bad. The list might go on and the details may vary, but the theme should be clear: the dodo disappeared from the planet because it was, in some way, maladapted to its environment. It was not fit, and therefore it did not survive. The traits, such as flightlessness, are the cause; extinction is the effect.

Darwinism and extinction

This Darwinian view of the dodo’s extinction, and of extinction in general, is common not only among biologists and ecologists, but also the public at large. It has a clear logical appeal: things happen for a reason and with extinction, that reason is natural selection. Ecologists refer to this kind of extinction as selective extinction or Darwinian extinction, owing to its foundation in the Darwinian logic of natural selection. And, just like natural selection, selective extinction is an important ecological process that has shaped Earth’s biodiversity. As just one example, consider the transition from the age of the dinosaurs to the age of mammals.

Viewing extinction as a selective process is of great practical significance. If managers and conservationists can identify the characteristics of selective extinction by understanding which species go extinct and why, then we’ve accomplished an important step towards conserving species in the future. In the dodo’s case, if being a largish, flightless, ground-nesting bird is what caused its extinction, then we may wish to prioritize conservation policies that will protect other largish, flightless, ground-nesting birds. This only works, however, if extinction truly is a selective process.

But is all extinction selective?

Answers to the ‘dodo’ question’ generally assume that selective extinction is the only, or most important, mechanism of extinction. In our paper, we remind the reader that we actually know that other mechanisms are important in the extinction process as well. In our study we focus on the importance of random chance – plain-old bad luck –  and show how important it is in the extinction process. This isn’t always an easy sell: Darwinian extinction (extinctions happen for a reason) and random chance (extinctions are bad luck) seem to be in direct logical conflict with each other.

Then again, take the dodo as an example. Knowing that the dodo only existed on a single island in the entire world may change our perspective. When a species’ entire population is small and isolated with no apparent refuges, it is no longer so hard to see how bad luck can play a role in the extinction process. Almost anything could cause the extinction of small isolated populations of species. In short, we know that population and range size matter in the extinction process, and the way it matters is that small populations are more subject to the influence of random chance in the extinction process. If a species exists at only one place in the entire world, being ‘in the wrong place at the wrong time’ suddenly becomes a very real possibility.

Improving extinction predictions

In our study, we use this general principle that small populations are more susceptible to random extinctions to develop better-informed predictions about extinction risk. We hope that managers and conservationists can use this approach to estimate extinction probability for rare species facing hypothetical threats – for example, a loss of half of their habitat when logging or a natural disaster occurs. By taking into account the fact that extinctions are sometimes caused by bad luck, we think we can do a better job of identifying the cases when extinctions are selective, which can lead to real improvements in predicting extinctions. In our view, the ultimate goal of conservation is preventing extinction, but we can best prevent extinction when we can predict extinction. And understanding the role that random probability plays in the extinction process in a critical step toward that end.

Read the full article, When are extinctions simply bad luck? Rarefaction as a framework for disentangling selective and stochastic extinctions, in Journal of Applied Ecology.