Variability in aquatic insect traits: Implications for traits-based biomonitoring

Background/Question/Methods

Traits-based analyses of aquatic insects have increased dramatically. These approaches benefit biomonitoring by incorporating measures of ecosystem function, allowing consistent comparisons among biogeographical locations, and providing mechanistic explanations and earlier detection of environmental change. Yet, few programs use traits effectively. One major obstacle could be the use of oversimplifed trait database categories, which ignore individual ontogenetic variation and phenotypic plasticity. Understanding trait variation and its effect on community structure is essential for the development of effective traits-based metrics. Here, we investigate two interrelated questions: 1.) How sensitive are trait database states compared to measured traits values for the characterization of site and organism variability, and 2.) Can ecologically informative trait states be developed based on direct measurement data? We evaluated quantitative variation in aquatic invertebrate body size and shape, two key traits often linked to ecological function, for four orders of aquatic insects: Ephemeroptera, Plecoptera, Trichoptera, and Odonata. Invertebrate samples were collected from the Miramichi River basin (New Brunswick, Canada). Specimens were identified to lowest possible taxonomic level, and digitally photographed with a stereomicroscope. Calibrated photographs were used for size measurement and geometric-morphometric analysis of shape.  

Results/Conclusions

Concordance between measured specimen sizes and their corresponding trait states is poor, with 51% of taxa falling outside the size ranges for their database body size trait states. The body size distribution for each site differs significantly (X²=428.1, df=2, p<0.0001), with similar trends observed for individual body size distributions among the sites for several abundant taxa (Capniidae, Epeorus, Lepidostoma, Leptophlebiidae, Maccaffertium, Rhithrogena). The size distribution shifts, detected using measured data, are obscured when taxa are summarized by trait states. Thus, measured trait values distinguish subtle differences in community structure among our variable reference sites. Using morphological attributes of specimens from biomonitoring samples, we identify 3 body size classes that appropriately represent the size structure of each site. Preliminary inspection of shape shows differences among taxonomic groups with the majority of variation present in the head and thorax. Ongoing analyses will determine and test potential shape categories. Finally, we provide recommendations on how trait measurements could be incorporated into biomonitoring procedures. Direct trait measurement data may provide greater ecological information than trait states alone; improving our ability to determine ecological condition from traits-based biomonitoring and offering linkages to community dynamics and other ecological properties. Realizing the full benefits of traits-based biomonitoring depends on evaluating trait properties and incorporating natural organismal variability.