Functional diversity through the mean trait dissimilarity: resolving shortcomings with existing paradigms and...

Area of Science:

• Ecology
• Quantitative Ecology
• Biodiversity Research

Background:

• Estimating functional trait diversity in biological communities is crucial for ecological understanding.
• Existing indices like mean species pairwise dissimilarity (MPD) and Rao quadratic entropy are widely used but often misunderstood.
• A common misconception is that Rao quadratic entropy is a form of MPD, leading to potential misinterpretations and selection errors.

Purpose of the Study:

• To mathematically and biologically distinguish between MPD and Rao quadratic entropy.
• To correct the misconception that Rao quadratic entropy is an abundance-weighted form of MPD.
• To provide a practical tool for ecologists to accurately assess trait diversity.

Main Methods:

• Mathematical derivation of the relationship between MPD and Rao quadratic entropy.
• Analysis of the indices using both species abundance and presence/absence data.
• Development and testing of a new R function ('melodic') for trait diversity calculation.

Main Results:

• The study establishes a clear mathematical relationship: Rao = MPD × Simpson diversity.
• This relationship holds true for both abundance and presence/absence data, refuting the idea of Rao being a simple abundance-weighted MPD.
• Significant differences between MPD and Rao were observed, particularly at low species diversities.

Conclusions:

• MPD and Rao quadratic entropy are distinct indices with different mathematical underpinnings and biological interpretations.
• The misconception of Rao as a form of MPD is mathematically unfounded.
• The new 'melodic' R function offers a reliable alternative for calculating trait diversity, avoiding issues with existing functions.