Niche versus neutral theory

Community ecologists have long debated the role of functional diversity in sustaining the species diversity of plant and animal communities. The traditional view of species cohabiting by virtue of differences in life-history traits has been questioned with increasing vigour over the last 10 years by developments in ‘neutral’ modelling. Neutral models are built on the simplifying premise of all species having identical vital rates. They are gaining considerable attention because of the surprisingly high accuracy that they achieve in representing empirical data on species-abundance distributions and species-area relationships. Clearly species are not all identical in their ecologies, leading to the inference that their differences may have little influence on community assembly. One hundred years of traditional niche theory seems to be sidelined by finding a good fit to global community data from models that ignore species differences.

In a paper published in PLoS ONE, Dr Patrick Doncaster has constructed a simulation model capable of switching between neutral scenarios (all species with identical life histories) and niche scenarios (species with more or less strongly contrasted life histories). By allowing the modelled community to find its own equilibrium of species richness and total individuals, he has shown how niche and neutral models can produce very similar predictions for species-abundance distributions and species-area relationships. The reason for the similarity is that the equilibrium state of any community is effectively a level playing field where everyone has zero net growth. Species that are intrinsically very different to each other, such as dominant competitors and fugitives, appear very similar when cohabiting together.

Although neutral models may fit well to the coexistence states of dissimilar species, Dr Doncaster’s work identifies significant dangers in drawing conclusions from the fit. Firstly, a good fit by neutral models does not signify a reduced role for niches, because it is the niche differences that allow coexistence, and the coexistence that then gives species their character of similarity at equilibrium. Secondly, neutral models predict substantially lower total numbers of individuals in a community than is predicted by niche models. Neutral models up to now have overlooked this important difference because they generally prescribe total population size. The risk is that a neutral model may under- or overestimate the resilience of a community to changes in carrying capacity imposed by environmental change. This can be mitigated against by incorporating a realistic measure of interspecific competition into the model – in effect, by acknowledging differences between species.

Doncaster, C.P. (2009) Ecological equivalence: a realistic assumption for niche theory as a testable alternative to neutral theory. PLoS ONE 4(10): e7460. doi:10.1371/journal.pone.0007460.