Correlations between broad‐scale taxonomic and genetic differentiations suggest a dominant imprint of historical processes on beta diversities

Abstract

Aim Dispersal limitation, environmental selection and drift are known to influence both taxonomic similarity between communities and genetic similarity between populations. However, disentangling the relative roles of these processes on spatial patterns of differentiation—whether regarding taxonomic differentiation (TD) between communities or genetic differentiation (GD) between populations—is challenging. Investigating whether spatial patterns of TD and GD are correlated (β‐SGDCs) is a promising approach to address this issue. Here, we investigated β‐SGDCs over broad spatial scales and 22 freshwater fish species to elucidate the processes shaping TD and GD between drainage basins. Location Global scope, data mainly from Europe and North America. Taxon Actinopterygii and Petromyzontiformes (freshwater fishes). Methods We used Mantel tests to investigate the raw correlation between TD and GD. We carried out multiple regressions to characterize the effects of geographic distance (proxy of dispersal limitation), environmental distance (proxy of environmental selection) and pairwise harmonic mean area between basins (proxy of drift) on TD and GD taken separately. We then analyzed the correlation between the residuals of these two regressions with Mantel tests. Finally, we investigated whether the variation in β‐SGDC between species was related to species traits. Results We detected an overall positive β‐SGDC, both for the raw and the residual correlations. This implies that, unmeasured effects, other than geographic distance, environmental distance and harmonic mean area, influence the β‐SGDC observed. Values of β‐SGDCs greatly varied between species, but this was not explained by any species traits. Main conclusions Independently of the effects of geographic distance, environmental distance and harmonic mean area, the β‐SGDC we observed suggests that historical processes strongly shaped the patterns of TD and GD between basins. Consequently, TD may be an appropriate proxy to explain the influence of historical processes on GD.

Spatial genetic differentiation correlates with species assemblage turnover across tropical reef fish lineages

“…Consequently, one might expect weak β-SGDCs for marine organisms. Our results parallel those observed for North American freshwater fishes that are strongly limited in their dispersal across drainage basins (Fourtune et al, 2016;Robuchon et al, 2019) Riginos & Nachman, 2001;Selkoe & Toonen, 2011). It is also worth noting that divergence times for most species belonging to the studied families (e.g., Gaboriau et al, 2018;Hodge et al, 2014) pre-date the period of establishment of the Monsoon Drift (circa 13 million years ago) in response to the onset of the Indian Monsoon (see Betzler et al, 2013).…”

“…Only Mantel test results are presented in the main text as the Procrustes analysis provided similar results. To test the overall significance of β‐SGDCs over the 20 species, we compared the observed mean p ‐value of the 20 β‐SGDCs (Mantel test or Procrustes analysis) to a null distribution of mean p ‐values obtained by sampling randomly 99,999 times 20 p ‐values in a uniform distribution in [0, 1] (see Robuchon et al, 2019). We calculated the p ‐value associated with the combined test as the frequency at which null mean p ‐values were below the observed mean p ‐value (Supporting Information Figure S5, Table S7).…”

“…The influence of these neutral processes manifest through the well-known correlations between α-diversity and habitat area or isolation at both interspecific and intraspecific levels (Fourtune et al, 2016;Messmer et al, 2012;Papadopoulou et al, 2011;Vellend & Geber, 2005). Similarly, positive β-SGDCs have been demonstrated to occur as a result of differential connectivity among habitats associated with dispersal processes for freshwater fishes (Fourtune et al, 2016;Robuchon et al, 2019). In a pioneer study using haplotype data for multiple species of tenebrionid beetles of the Aegean archipelago, Papadopoulou et al (2011) also identified a positive β-SGDC, caused by stochastic dispersal processes as expected by neutral theories (Antonovics, 1976;Hubbell, 2001).…”

“…Using simulations and empirical data, they showed similar spatial turnover rates at both haplotype and species levels if dispersal limitation processes dominated, while different spatial turnover rates were found if both environmental constraints and dispersal limitation influenced species distribution. However, as pointed out by Robuchon et al (2019), the observed variations of SGDCs among species has never been formally tested and the role of species traits in shaping the strength of the correlation remains understudied.…”

Species and genetic diversity relationships in benthic macroinvertebrate communities along a salinity gradient

“…Considering that salinity was surprisingly not found to be associated with species diversity in this study, including more study sites across the large spatial range was likely needed to capture the full range of variation in community species diversity in the Baltic Sea. Previous studies considering multiple species revealed positive SGDCs over large geographical scales than we studied [53][54][55][56] that were driven by larger scale extrinsic forces such as isolation, habitat area, climate and historical processes. Another difference in those studies compared to our study is that genetic diversity was measured in different species from the same taxonomic class and order.…”