Congruence of biodiversity measures among larval dragonflies and caddisflies from three Canadian rivers



This article is corrected by:

  1. Errata: Corrigendum Volume 59, Issue 4, 885, Article first published online: 5 March 2014

  • Reproduced with the permission of the Minister of Environment.

Colin J. Curry, Canadian Rivers Institute and Department of Biology, University of New Brunswick, PO Box 4400 Fredericton, New Brunswick, Canada E3B 5A3.


1. Scientists tasked with collecting taxon richness and assemblage variation data for conservation purposes have identified biomonitoring studies as potential sources of information. This approach assumes that biodiversity patterns revealed by biomonitoring reflect those of the wider community, an assumption not thoroughly tested in riverine ecosystems.

2. We compared patterns of taxon richness and assemblage variation in an important biomonitoring group (Trichoptera) with a group with high conservation significance (Odonata) at 34 sites across three fifth-order catchments. We also explored the effect of abundance on observed patterns by rarefying the larval Trichoptera data set.

3. Our results indicate that Trichoptera do not fully reflect site-scale taxon richness or assemblage variation in Odonata. The magnitude of odonate assemblage variation was much greater than that of Trichoptera for one of the catchments. Odonata and Trichoptera richness was moderately correlated in two catchments, while assemblage variation was strongly correlated in another pair of catchments. However, comparisons based on rarefied data eliminated differences in the magnitude of assemblage variation and strengthened correlations in richness and assemblage variation, suggesting the lack of congruence in these measures might be due to differences in abundance among groups. Further, incomplete taxonomy may mask additional assemblage variation, particularly in Trichoptera.

4. Conservation planning in riverine ecosystems based on proxies derived from biomonitoring data should proceed cautiously until we understand how well the resulting information reflects biodiversity patterns in under-sampled taxa and habitats. Future studies of biodiversity congruence should consider both richness and assemblage variation as each provides valuable information for conservation-related decisions. The taxonomic resolution and relative abundance of comparison groups can potentially impact the strength, direction and statistical significance of patterns. Researchers should employ species-level taxonomy and account for differences in abundance among groups through rarefaction where at all possible and DNA-based taxonomy methods can support this.