Tests of size structured neutral theory on tropical forest data

Background/Question/Methods Do species coexist because of niche differences, or because of similarities? Neutral theory posits the latter and predicts the properties of communities shaped primarily by demographic stochasticity and dispersal limitation rather than species differences and stabilizing mechanisms. Its predictions provide a good fit to the relative abundance distribution of tree species in the tropical forest on Barro Colorado Island, as well as other data sets. But does it provide a good fit because it is a good model of assembly processes in that community, or because relative abundance distributions are insensitive to community assembly processes? Many argue the latter. In its current form neutral theory ignores size differences among individuals and their impact on fecundity, growth, and survival. A recently created size structured neutral model predicts the same species abundance distribution as current neutral theory, but makes additional predictions that may provide a better litmus test for community process. In particular, this model predicts the distribution of total biomass across species. Current studies of that model only deal with the case where fecundity is independent of size. Here we simulate to obtain the predictions for the case where fecundity scales with size, and for the parameter values that have previously been used to fit neutral theory to the Barro Colorado Island species abundance data, and compare these predictions with observations for this data set. Results/Conclusions We obtain the conditions on the scaling of fecundity, growth, and death with size for the size structured neutral theory to additionally provide a good fit to the distribution of biomass across species for this data set, and compare those conditions with available information about the size scaling of these demographic rates. We find that these conditions are unrealistic. This suggests that neutral theory might be more easily rejected for this data set based on the species biomass distribution. However, further work is needed to more generally explore the ability of size structured neutral theory to provide a good fit to both the abundance and biomass distributions of this data set for the same parameter values.