The pairwise approach to analysing species co‐occurrence

Abstract

The analysis of species co-occurrence patterns continues to be a main pursuit of ecologists, primarily because the coexistence of species is fundamentally important in evaluating various theories, principles and concepts. Examples include community assembly, equilibrium versus non-equilibrium organization of communities, resource partitioning and ecological character displacement, the local-regional species diversity relationship, and the metacommunity concept. Traditionally, co-occurrence has been measured and tested at the level of an entire species presence-absence matrix wherein various algorithms are used to randomize matrices and produce statistical null distributions of metrics that quantify structure in the matrix. This approach implicitly recognizes a presenceabsence matrix as having some real ecological identity (e.g. a set of species exhibiting nestedness among a set of islands) in addition to being a unit of statistical analysis. An emerging alternative is to test for non-random co-occurrence between paired species. The pairwise approach does not analyse matrix-level structure and thus views a species pair as the fundamental unit of co-occurrence. Inferring process from pattern is very difficult in analyses of co-occurrence; however, the pairwise approach may make this task easier by simplifying the analysis and resulting inferences to associations between paired species.

Species co‐occurrence analysis: pairwise versus matrix‐level approaches

“…The use of paired statistical tests has been advocated over community-wide alternatives to examine patterns of species co-occurrence (Veech, 2013(Veech, , 2014Griffith et al, 2015). The idea is to perform tests for each of the S(S − 1)/2 possible pairs that can be formed with S species and tally the number of cases in which the observed number of co-occurrences is significantly lower or higher than expected.…”

“…The F-E algorithm homogenizes the number of species per site, creating spatial patterns that are very different from the real-life system, and consequently generating an inflated number of statistically significant cases of species aggregation, as shown with the example of Darwin's finches. Veech (2013Veech ( , 2014 argues that his pairwise approach differs from what he considers to be a traditional approach based on the randomization of whole presence-absence matrices and the analysis of matrix-level measures of co-occurrence. He presented the analyses associated with the controversy about Diamond's assembly rules (Diamond, 1975; as examples of what he calls the matrix-level approach.…”

“…However, Connor et al (2013) have emphasized that both the conceptual and the methodological frameworks in those analyses are in fact based on pairwise comparisons. Moreover, many of the studies that Veech (2014) cites as examples of pairwise comparisons (Sanderson, 2000;Sfenthourakis et al, 2006;Sanderson et al, 2009;Collins et al, 2011) are based on randomization procedures that involve the whole matrix, such as Sanderson's (2000) knight's tour algorithm or the swap algorithms that mimic SIM9 procedures (Gotelli, 2000). In other words, they are examples of the broad-sense pairwise approach, not of the restrictive Veech approach.…”

“…To address that question, Veech (2013Veech ( , 2014 advocates the use of a 'pairwise approach' to examine patterns of species co-occurrence. In the broad sense, the pairwise approach consists in performing tests for each pair of species in an assemblage, recording the number of cases in which the observed co-occurrence rate is lower or higher than expected.…”

On the operative use of community overlap in analyzing incidence data

“…The analysis of incidence data has a long history in studies ranging from community ecology to biogeography (Veech 2014). Species incidence among sites can be interpreted in two complementary ways (Arita 2015): in analyses by species (Pielou 1977) and in analyses by sites (Koleff et al 2003).…”