River habitat surveys and biodiversity in acid-sensitive rivers

✍ Scribed by P.A. Brewin; S.T. Buckton; S.J. Ormerod

• Book ID: 101277185
• Publisher: John Wiley and Sons
• Year: 1998
• Tongue: English
• Weight: 161 KB
• Volume: 8
• Category: Article
• ISSN: 1052-7613
• DOI: 10.1002/(sici)1099-0755(199807/08)8:4<501::aid-aqc290>3.0.co;2-w

✦ Synopsis

1. The distribution of aquatic invertebrates (102 sites) and river birds (74 sites) was assessed along acid-sensitive streams in upland Wales. River Habitat Surveys (RHS) and measurements of pH at the same sites were then used to ask the questions: (i) do RHS measures indicate habitat requirements for individual species? (ii) do RHS measures reflect patterns in community composition? (iii) are the relationships from (i) and (ii) modified by strong gradients in acid-base status?

2. The abundances of 13 out of 26 invertebrate species were significantly but weakly correlated with variables derived from RHS. Using discriminant analysis, abundance values for invertebrate species could be assigned to the correct category at around 50% of sites using RHS data alone, but results for acid-sensitive taxa were improved when pH was used as an additional discriminating variable. Moreover, assignment of abundance using pH and altitude was at least as successful as pH plus RHS. At the community level, RHS illustrated likely effects on aquatic invertebrates over and above those due to acidity.

3. RHS data also correlated significantly with the distribution of common sandpipers and dippers. For the former species, prediction using sub-sets of the RHS data gave better results than the whole data, while pH significantly improved models for dippers.

4. These results further support the wider use of RHS in assessing river biodiversity. In particular, predictive models combining RHS and chemical data have the potential to indicate pollution as well as habitat change. However, the data show also that simpler predictors of distribution sometimes give comparable results to RHS; that sub-sets of RHS variables sometimes describe distribution better than the entire RHS data set, and that spurious relationships with biota can sometimes arise. The latter problem results from the large numbers of variables involved in RHS, so that the use of RHS data will have to be judicious and well supported where possible by data on cause-effect links with ecological pattern. These will be important lessons as RHS is used increasingly in integrated river management for a wide range of uses.