The importance of retaining a phylogenetic perspective in traits‐based community analyses

Abstract

1. Many environmental stressors manifest their effects via physiological processes (traits) that can differ significantly among species and species groups. We compiled available data for three traits related to the bioconcentration of the toxic metal cadmium (Cd) from 42 species representing the aquatic insect orders Ephemeroptera (mayflies), Plecoptera (stoneflies) and Trichoptera (caddisflies). These traits included the propensity to take up Cd from water (uptake rate constant, k u ), the ability to excrete Cd (efflux rate constant, k e ) and the net result of these two processes (bioconcentration factor). 2. Ranges in these Cd bioaccumulation traits varied in magnitude across clades (some clades had a greater tendency to bioaccumulate Cd than others). Overlap in the ranges of trait values among different clades was common and highlights situations where species from different lineages can share a similar trait state, but represent the high end of possible physiological values for one clade and the low end for another. 3. Variance around the mean trait states differed widely across clades, suggesting that some groups (e.g. Ephemerellidae) are inherently more variable than others (e.g. Perlidae). Thus, trait variability/ lability is at least partially a function of lineage. 4. Akaike information criterion comparisons of statistical models were more often driven by lineage than by the other tested potential biological or ecological explanations. Lineage-driven models generally improved with increasing taxonomic resolution. 5. Together, these findings suggest that lineage provides context for the analysis of species traits and that failure to consider lineage in community-based analysis of traits may obscure important patterns of species responses to environmental change.

Assessment of water quality in urban streams based on larvae of Hydropsyche angustipennis (Insecta, Trichoptera)

“…Our findings confirm that this species is capable of accumulating certain amounts of heavy metals even if their concentration in water is low (Girgin et al 2010), as it has been detected in other Hydropsyche species (Cain and Luoma 1998;Barata et al 2005). For previous studied Hydropsyche spp., the metal uptake and efflux were on the similar level irrespective of studied species (Evans et al 2002;Cain et al 2006;Poteat et al 2013;Poteat et al 2015). Taking into consideration the kinetics of metal presence in larvae, the biological half-life of kind of heavy metals, e.g., Cd or Zn, is relatively short (on the order of several weeks) for these caddisflies.…”

On the importance of trait interrelationships for understanding environmental responses of stream macroinvertebrates

“…Plecoptera are known as sensitive taxa predominantly present in low order, cold-water forested habitats, while many Odonata species also occur in warmer, sun-exposed lentic habitats (Kalkman et al, 2008;T€ ornblom, Degerman & Angelstam, 2011). This example shows that when using traits to understand the patterns in habitat use in stream macroinvertebrates, the order that taxa belong to can be a relevant proxy for 'hidden traits' shaped by their evolutionary history (see also Resh et al, 1994;Usseglio-Polatera et al, 2000;Verberk et al, 2013;Poteat, Jacobus & Buchwalter, 2015).…”

Improving cross-species extrapolation of chemical sensitivity

“…Finally, we argue that the models could not achieve a better model performance because, regardless of data availability, trait data alone may be insufficient in explaining species sensitivity. Several studies show that complementing trait data with data on phylogenetics greatly enhances model performance, and that both traits and phylogenetic indicators explain a distinct part of species sensitivity (Poteat, Jacobus et al 2015, Pilière, Verberk et al 2016. Other attempts combining phylogenetics and physiochemical properties into predictive models have also proven successful (Guénard, von der Ohe et al 2014, Malaj, Guénard et al 2016, although phylogenetics is still impossible to include in an inclusive approach as we performed in this study.…”

“…We find that combining traits with taxonomic information results in models with increased predictive power, although only marginal (Table 3.2). Previous studies likewise emphasize the importance of complementing traits approaches with taxonomic approaches (Poff, Olden et al 2006, Poteat, Jacobus et al 2015, Pilière, Verberk et al 2016. For example, Pilière and colleagues (2016) used boosted regression tree modelling to assess the environmental responses of single traits, orders and trait profile groups.…”

“…Furthermore, some studies have demonstrated that indeed traits and phylogeny (or other measures of relatedness between species) both explain an unique part of the sensitivity process (Poteat, Jacobus et al 2015, Pilière, Verberk et al 2016. However, phylogenetic approaches do not unravel any concrete mechanisms of sensitivity, and are therefore more susceptible to overfitting on the training data.…”