Sampling and comparing odonate assemblages by means of exuviae: statistical and methodological aspects

Freshwater species are in serious decline all over the world. Thus, monitoring of freshwater ecosystems and species is crucial to guide policy actions and dragonflies are generally considered to be good indicators for these systems. The aquatic life stage of the Odonata is inherently more susceptible to changes in water quality than the imago and therefore exuviae give better insights into site-specific effects. However, utilizing exuviae for monitoring purposes introduces a number of problems. For example, they often do not persist long in the environment. Some of these problems have been addressed. However, there are no published data that distinguish the influence of sampling frequency and total sampling effort on the faunal completeness. Also the number of exuviae necessary to define local assemblages has not been investigated. These questions were addressed by analyzing the data on exuviae collected in seven sites and we found that for any given total amount of time invested, it was always preferable to conduct more short surveys, rather than fewer surveys lasting longer. The study also showed that a sample size of 300 exuviae allowed us to reliably estimate the similarity of two assemblages from different sites. However, when collecting 40 exuviae or less, the reliability was low. Based on our findings we recommend sampling exuviae for a minimum of 5 days, evenly spread out over the entire season during which Odonata emerge, to sample each time for approximately 60 min and aim to collect not less than 300 exuviae in total.