The neutral bomb
This is not a sex bomb, but who knows. A couple of weeks ago I was giving a talk at the Escuela Argentina de Matemática y Biología about neutral and non-neutral models in ecology. I was pointed to the paper of Clark 2012 The coherence problem with the Unified Neutral Theory of Biodiversity with the commentary: the neutral theory is uninformative.
And the paper basically says that neutral models or neutral theory is based on ignorance and that they are incoherent.
I think this is very interesting, but I disagree, here I review the last four points of the article:
(i) Incoherence: Neutral models assume all individuals are equal then go on and predict things like SADs and SARs, what is this telling us about the niche? The individuals are all equal, the species can have different effective probabilities to reach a place, different densities, but there is no niche. In this sense it can be thought as a null model for niches, other kinds of null models for niche can be constructed and maybe they are better, but I don’t see incoherence. Competition and niche are not the only things relevant to coexistence of species.
(ii) Ignorance: I am sure we are ignorant about nature, but assuming that all individuals are equal is not equal probability and equal probability is not a null model for sameness. Anyway the equivalence of individuals is certainly not ignorance is an assumption.
(iii)Victory: There are no victories here, but models or theories to explain nature patterns and we need to confront them with data in several ways, and much better in several ways at the same time.
(iv)Applying neutral models in conservation: you have to be careful to apply a model to conservation practices, neutral or not neutral does not matter at all. Models are not reality, always have simplifications, aproximations, or assumptions and we can select which is better for our problem (see Model selection in ecology and evolution).
The real issue I think is a matter of scale: which are the microscale processes that determine macroscale patterns. If niche models predicts the same patterns that neutral models and both are consistent with real data (see Niche and neutral models predict asymptotically equivalent…) a parsimonious interpretation is that we don’t need niche. But if real data does not match the model we do need to add more complexities.
I think that neutral models are usefull both intellectually and practically, more about this in Neutrality without incoherence: a response to Clark