On the use of meiofauna in ecological monitoring: Who needs taxonomy?

Taxonomic problems have hindered the widespread use of meiobenthos for the purpose of pollution monitoring despite its potential usefulness. The question whether it is necessary to go down to species level in order to distinguish assemblages and stations was studied using data on meiofauna densities available from a GEEP-workshop held in Norway, August 1986.

Using the complete species list of harpacticoid copepod and nematode species, the six stations could be clearly separated by different multivariate techniques (TWINSPAN, DCA, MDS, Clustering). The abundance data of the genera, the families and even the orders (in the case of nematodes) still allowed a more or less clear separation of the stations, although their relative relationships were sometimes distorted. Thus, from a practical point of view, it may be sufficient to identify the animals to genus or family level.

It is proven that this result is not due to ecological similarities between congeneric or confamiliar species. The species groups typifying the stations in the TWINSPAN analysis did not contain more confamiliar species pairs than expected from random sorting. Randomly assembled ‘families’ allowed a separation of the stations as clear as that based on taxonomic groups.

We conclude that the information contained in species abundances was preserved, at least partially, when the species are grouped, even at random. The grouping in higher taxonomic categories was as good as any, and more convenient than all the others. However, within the (limited) range of habitats studied, ecological adaptation is achieved nearly exclusively on the species level.