A family of (dis)similarity measures based on evenness and its relationship with beta diversity

In this paper, I propose a new evenness-based method for calculating plot-to-plot (dis)similarity coefficients. The method is very flexible, as (dis)similarity can be calculated for any kind of species abundance data (also including functional or phylogenetic differences between species), and can be easily generalized to multiple sites. To show how the proposed method works in practice, the behavior of two similarity coefficients based on Pielou's and Williams' evenness is examined with simulated data representing an ideal ecological gradient. Being derived from classical evenness indices, which have been used in ecology for decades, this new family of measures has a great potential for future research in community ecology and multivariate analysis.