`Assessing the biological quality of fresh waters: RIVPACS and other techniques', edited by John F. Wright, David W. Sutcliffe and Mike T. Furse. Published by the Freshwater Biological Association, Ambleside, June 2000. ISBN 0 900386 62 2. 400 pages. Price £40 softback, £60 hardback (including p. & p.).

Developing a RIVPACS Type Model

In order to develop a RIVPACS type model, a series of high quality sites (normally referred to as "reference sites") is required. These are short river stretches, carefully selected to try to include the full range of running water sites within the region of interest.

Although practically no river is unaffected by human activities, the reference sites are chosen with great care to represent sites with minimal impact which are the best examples of their type. Site selection is a critical phase and users must ensure they include rivers across the whole range of geological formations in the region. Individual reference sites covering a sequence from headwaters to downstream sampling sites must also be included.

At each reference site, macroinvertebrates and environmental data are collected using agreed standard protocols. Each site may be sampled several times over the year to ensure a complete picture of the macroinvertebrate fauna is acquired. For example, in the UK reference sites were sampled in spring, summer and autumn. The macroinvertebrate fauna are identified to the lowest practical taxonomic level (species or genus). The environmental data collated for each site should focus on features that would be unaffected by environmental stresses.

Statistical models are then developed to summarise the inter-relationships between the observed macroinvertebrate fauna of the reference sites and their environmental characteristics. First, the reference sites are classified into a series of site groups, based only on the macroinvertebrate fauna. Then the relationships between the environmental features and the faunal characteristics of the "reference" site groups are defined, and lead to the development of a predictive model. The model is then strictly validated and the quality of the reference sites is assessed.

The final validated predictive model enables the user to estimate the macroinvertebrate community expected at high quality sites from the information on their environmental and physical features. By measuring these environmental features for a new site, you can then predict the macroinvertebrate fauna you would expect to find at the site if it was also of high quality. Expected fauna for a site is referred to as its “biological reference condition” within the WFD.

If a macroinvertebrate sample is then taken at the new site, using the same standardised protocols as for the reference sites, the observed fauna can be compared with the expected fauna and discrepancies between the two can be used to assess the biological condition or "ecological status" of that stretch of river.

The following pages give more detail on each stage of the development of a RIVPACS type predictive approach to assessing freshwater quality using macroinvertebrates.