How can we test neutral theory robustly in ecology?

Background/Question/Methods

Neutral theory has the potential to play an important role in ecology as a null model whose rejection indicates the importance of niches in allowing for the coexistence of competing species in communities.  However, so far most tests of neutral theory’s predictions of species abundance patterns have done little to provide clear evidence for the role of niches. In many cases our neutral model is flexible enough to explain observed patterns, but it seems that a niche model might do just as well.  Alternatively, when our neutral model fails to explain observed patterns, it is unclear if niches are really at play, or if some important details of demographic processes have been left out or misrepresented in our neutral model.  Some have suggested that ecologists should abandon any hope of gleaning the role of niches by testing neutral theory’s predictions about species abundance distributions.  Some have even suggested that species abundance distributions are virtually unaffected by the presence of niches.  However, other work, also to be presented at this conference, indicates that niches do indeed influence species abundance distributions.  Furthermore, alternatives approaches that have been suggested for gleaning the role of niches seem useful only in systems where experimental manipulation of whole communities is logistically feasible, or for which very long-term time series are available.  So how can we test neutral theory robustly in ecology? How might we design our tests so that we can more confidently distinguish niche from purely neutral in the wider range of systems ecologists want to study?  Is there any hope in using static patterns in the distribution and abundance of species?  

Results/Conclusions

In this talk we will suggest that there is.  We will describe and take inspiration from some of the ways in which evolutionary biologists test the neutral theory of molecular evolution.  We’ll point out that evolutionary biologists have found ways to deal with uncertainty about expectations in the neutral case, and discuss how we might do the same in ecology.  We’ll discuss two key approaches to testing neutral theory robustly:  1) reducing uncertainty about neutral predictions by fitting demographically complex neutral models to available data, and 2) focusing our tests on patterns where the expected outcome from niche and neutral models differs qualitatively regardless of uncertainties of the details of the demographic processes.