Linking nested and non-nested patterns of species turnover: Exploring the limits of the relationship

Background/Question/Methods

Patterns of species turnover have the potential to implicate what processes drive community dynamics across scales. However, a ‘best’ metric of turnover does not exist and it is unclear how redundant different metrics of turnover are.  Therefore, the objective of this study was to define an analytical framework for harnessing information gained from nested and non-nested turnover metrics simultaneously.  We developed analytical proofs that define the pattern of species occupancy that maximize and minimize non-nested turnover for a given level of nested turnover in a system of four equal area quadrats. We applied our analytical framework to plant community data from a mixed-hardwood forest, a tallgrass prairie, and an arid shrubland. For each community, we calculated a common metric of nested turnover – the slope of the species-area relationship, z, and a common metric of non-nested turnover – the average Jaccard’s dissimilarity, T_J.  We examined the empirical turnover values with respect to the mathematical constraints on the relationship between T_J and z defined in our framework. Lastly, the empirical communities were compared with a model of scale-invariant turnover, T_J = 2-2^1-z. 

Results/Conclusions

Due to the geometry of the constraint space, the potential difference between nested and non-nested metrics of turnover was largest (≈ 30% difference) at intermediate values of nested turnover (0.2 ≤ z ≤ 0.5). As nested turnover increases or decreases to 1 or 0 respectively the potential difference of non-nested turnover decreases to 0%. Although the empirical T_J-values were in the lower half of the constraint space, they were typically larger than expected by the model of scale invariant turnover.  Also the magnitude of T_J for a particular z was negatively correlated with variance in occupancy. Our study demonstrates, in an assemblage composed of four quadrats, that the degree of nested turnover imposes strong mathematical constraints on the magnitude of non-nested turnover.  Therefore, depending on the value of nested turnover, one may predict the potential range of non-nested turnover values.  Furthermore, by examining where empirical communities fall with respect to the mathematical constraint space we can gain a better understanding of how and why a community is structured in a particular way.