Shopping for Ecological Indices? On the Use of Incidence-Based Species Compositional Similarity Measures

Abstract

β-diversity has been under continuous debate, with a current need to better understand the way in which a new wave of measures work. We assessed the results of 12 incidence-based β-diversity indices. Our results of gradual species composition overlap between paired assemblages considering progressive differences in species richness show the following: (i) four indices (β-2, β-3, β-3.s, and βr) should be used cautiously given that results with no shared species retrieve results that could be misinterpreted; (ii) all measures conceived specifically as partitioned components of species compositional dissimilarities ought to be used as such and not as independent measures per se; (iii) the non-linear response of some indices to gradual species composition overlap should be interpreted carefully, and further analysis using their results as dependent variables should be performed cautiously; and (iv) two metrics (βsim and βsor) behave predictably and linearly to gradual species composition overlap. We encourage ecologists using measures of β-diversity to fully understand their mathematical nature and type of results under the scenario to be used in order to avoid inappropriate and misleading inferences.