Trait patterns of aquatic insects across gradients of flow-related factors: a multivariate analysis of Canadian national data

Abstract

Large-scale comparisons of aquatic bioassessment metrics based on taxonomic composition are currently constrained by the biogeographic limitations of taxon occurrence. The use of species trait patterns offers a possibility to overcome this constraint. We examine the assertion that the trait composition of aquatic insect communities changes in a consistent manner along similar environmental disturbance gradients by evaluating relationships between traits and three flow-related variables (velocity, water temperature, and dissolved oxygen) in 13 independently collected Canadian data sets. Certain trait states such as low crawling rate, common occurrence in drift, short adult life span, erosional rheophily, medium size at maturity, and cold or cool thermal preference were consistently sensitive to all three flow-related factors, velocity in particular, despite biogeographic differences in faunal composition. Trait modalities exhibiting the highest mean correlation with velocity, temperature, and dissolved oxygen were identified, and the potential confounding effect of trait “syndromes” was addressed by restricting the selection of flow-sensitive traits to those with high evolutionary lability. The results of the study provide a basis for the future development of flow bioassessment metrics at the national Canadian scale and potentially at the international scale.

Incorporating traits in aquatic biomonitoring to enhance causal diagnosis and prediction

“…The potential for traits to be correlated may confound causal inference in biomonitoring applications, and it suggests that a limited number of possible trait combinations or ''syndromes'' may exist Horrigan and Baird 2008). Thus, the relative sensitivity of individual and correlated traits to different stressors needs to be evaluated more fully.…”

“…Thus, Statzner and Bêche (2010) argue that individual traits must not be used as indicators of many different stressors if reliable interpretations of trait responses are to be determined in multiple stressor environments. Using the trait modalities described by Poff et al (2006), Horrigan and Baird (2008) demonstrated that in a multistressor environment, selected trait modalities were influenced exclusively by changes in flow conditions and were not responsive to thermal and oxygen stress. Other traits were simultaneously responsive to the multiple stressors, and consequently had reduced diagnostic power.…”

“…Statistical methods used include cluster analysis , correlation (Horrigan and Baird 2008), ordination (Dray and Legendre …”

Elevated CO2alters leaf-litter-derived dissolved organic carbon: effects on stream periphyton and crayfish feeding preference

“…1), MTB approaches should focus on evolutionarily labile traits that have low statistical correlations (i.e., are phylogenetically independent) and are linked in a mechanistic way to a specific human impact to avoid confounding effects of trait syndromes (see also Horrigan and Baird 2008). This argument calls for constructing molecular (e.g., Ball et al 2005) or at least morphometric phylogenies of freshwater invertebrates and other groups to isolate the part of history that explains the presence of a given taxon in a given environment.…”

Trait-based surveillance of flood channel effects on the River Thames

“…For a mechanical, multispecies biomonitoring of human impact; across-taxa traits provide a robust tool independent of topographical, geographical, geological and seasonal change in species composition (Charvet et al 2000;Bêche et al 2006;McGill et al 2006;Horrigan & Baird 2008). Chironomid taxa data in this study have been translated into definable, long-term functional traits such as feeding groups or significant adaptations to river flow or chemical composition.…”