A general method for comparing species abundance distributions from different communities

It is often desirable to compare the species abundance distributions from different communities. To date, there are been few rigorous approaches to the problem, particularly when there is no underlying theoretical model upon which to base comparisons. Assuming a very general sampling model, it is possible to develop an object, model selection approach based on Akaike Information Criterion (AIC) comparisons. The basic approach is to develop a set of alternative models that encapsulate the research hypothesis being tested. Next, a log-likelihood is calculated for each alternative model using an unconstrained multinomial sampling distribution. From these, AIC values are obtained for each model. The model with the lowest AIC is chosen as the “best” model, and all other models compared to it. Several examples from different taxa are used to illustrate the interpretation of the model selection process.