Original paper

The development and testing of a macroinvertebrate biotic index for detecting the impact of acidity on streams

Davy-Bowker, John; Murphy, John F.; Rutt, Graham P.; Steel, John E.C. Furse

Abstract

A new macroinvertebrate biotic index for assessing the impact of acidity on streams and rivers in England and Wales is presented, the Acid Waters Indicator Community index (AWIC). Using a 1042 sample training dataset, the AWIC index was derived by firstly using canonical correspondence analysis (CCA) to identify independent physical variables that best accounted for the observed macroinvertebrate communities. Secondly, partial CCA was used to factor out the confounding influence of physical variables other than mean pH. Thirdly, the ranking of taxa along the first axis of the pCCA (strongly correlated with mean pH) was the basis for assignment of index scores. Ephemeridae and Physidae were among the most acid sensitive taxa (AWIC score of 6), while Chloroperlidae and Nemouridae were the most tolerant of low pH conditions (AWIC score of 1). Within the training dataset, AWIC was strongly correlated with mean pH (rs = 0.814, P <0.001). Testing on an independent dataset of 2710 samples, AWIC was also more effective at discriminating between samples of differing mean pH and gave rise to fewer Type I errors (false positives) than an existing biotic index derived by a classification approach. Our analysis suggests that at family level, macroinvertebrate communities are not characterised by a specific obligate acid-waters assemblage so that while most acidic sites primarily support families that are acid tolerant, these families are also commonly found in circum-neutral waters. The AWIC index distinguishes acid sites by the absence of any of a large number of acid sensitive families, any of which can shift the overall AWIC index towards one indicative of higher pH. The AWIC index should compliment existing family and species