Aquatic Ecology

, Volume 41, Issue 3, pp 387-398

First online:

Application of clustering techniques for the characterization of macroinvertebrate communities to support river restoration management

  • V. AdriaenssensAffiliated withLaboratory of Environmental Toxicology and Aquatic Ecology, Ghent University Email author 
  • , P. F. M. VerdonschotAffiliated withAlterra Green World Research
  • , P. L. M. GoethalsAffiliated withLaboratory of Environmental Toxicology and Aquatic Ecology, Ghent University
  • , N. De PauwAffiliated withLaboratory of Environmental Toxicology and Aquatic Ecology, Ghent University

Rent the article at a discount

Rent now

* Final gross prices may vary according to local VAT.

Get Access


The European Water Framework Directive prescribes that the development of a river assessment system should be based on an ecological typology taking the biological reference conditions of each river type as a starting point. Aside from this assessment, water managers responsible for river restoration actions also need to know the steering environmental factors to meet these reference conditions for biological communities in each ecological river type. As such, an ecological typology based on biological communities is a necessity for efficient river management. In this study, different clustering techniques including the Sørensen similarity ratio, ordination analysis and self-organizing maps were applied to come to an ecological classification of a river. For this purpose, a series of sites within the Zwalm river basin (Flanders, Belgium) were monitored. These river sites were then characterized in terms of biotic (macroinvertebrates), physical–chemical and habitat variables. The cluster analysis resulted in a series of characteristic biotic communities that are found under certain environmental conditions, natural as well as human-influenced. The use of multiple clustering techniques can be of advantage to draw more straightforward and robust conclusions with regard to the ecological classification of river sites. The application of the clustering techniques on the Zwalm river basin, allowed for distinguishing five mutually isolated clusters, characterized by their natural typology and their pollution status. On the basis of this study, one may conclude that river management could benefit from the use of clustering methods for the interpretation of large quantities of data. Furthermore, the clustering results might enable the development of a cenotypology useful for efficiently steering river restoration and enabling river managers to meet a good ecological status in most of the rivers as set by the European Water Framework Directive.

Key words

European Water Framework Directive Multivariate analysis Ordination River typology Self-organizing maps Similarity ratio