`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.).

RIVPACS - Reference Condition

Predicting the Reference Condition

The development of the RIVPACS type approach led to the whole concept of an expected fauna and Reference Condition, and to the idea of comparing the observed fauna at the site with its expected fauna all of which is now at the heart of the prescribed methodology within the WFD.

The crucial first step in the RIVPACS type approach is the selection of high quality unstressed reference sites upon which the prediction model will be based. At each reference site, we must then obtain biological data and environmental variables representing the perceived environmental drivers or at least correlates of the biota. It is vital that an adequate number of reference sites are sampled to ensure that there is sufficient coverage of the different stream types in the prediction model.

A lowland river

The RIVPACS type approach requires high quality sites for each type of site or at least sites which have been only marginally impacted which can be used to set a realistic target. The WFD describes a reference site as that with no, or only very minor, anthropogenic alterations to the values of the hydrochemistry and hydromorphology and with biota usually associated with such undisturbed or minimally disturbed conditions. The word "pristine" should never be used in connection with reference sites. In the UK, it is unlikely that any pristine sites exist. In North America, some ecologists argue that pristine should refer to conditions prior to colonisation by Europeans.

Another constraint is that the reference sites must be chosen subjectively beforehand. Initially for UK RIVPACS, these were chosen by consulting local biologists as to the location of the best quality sites, and over time the process aimed to cover the full range of environmental types and river systems. In reality this selection is an iterative process. Over time, some reference sites in earlier phases of the project have been removed when judged to be of insufficient quality and others have been added to improve the representation of some stream types. For example, the latest version of RIVPACS in the UK provides better coverage of small head water streams.

An upland stream

A RIVPACS type approach cannot define a reference condition for a site in the absence of any high quality or marginally impacted reference sites of "similar" physical type. However, nor can any other prediction system based only on relationships derived between observed biological and environmental field data, or at least not without unjustified extrapolation. From a practical point of view, if the best available sites are chosen then targets can be set to improve other sites to at least this level.