|HOME SCHEDULE AUTHOR INDEX SUBJECT INDEX|
(PH215) A comparison of the sensitivity of fish, benthic, and periphyton multimetric indices with multivariate statistics to detect effects along a wadeable stream in Pennsylvania.
Thomas, J.1, Hall, T.1, 1 National Council for Air and Stream Improvement, Anacortes, WA, USA
ABSTRACT- The last decade has seen a major move by regulatory bodies to adopt multimetrics, such as indices of biotic integrity (IBIs) as assessment tools. The results of an IBI are simple and easy to communicate, but the method has been criticized as oversimplifying complex systems. This study compared the sensitivity and repeatability of published fish, benthic macroinvertebrate and periphyton IBIs, and their individual metrics, with multivariate statistical methods in their ability to distinguish spatial (upstream and downstream) differences along a wadeable stream in south central Pennsylvania. Results demonstrated that the total IBI scores for the fish and benthics were able to differentiate a cold water site and change in stream order, but were unable to identify upstream to downstream differences within a stream segment. The total score for the periphyton IBI found no significant differences, although there was a significant correlation with river distance, indicating a trend in higher scores upstream to downstream. Some of the individual metrics for all three communities were able to identify the broad categories of stream temperature and stream order. The results also suggested that caution should be taken when using non-site specific IBIs or with too few sample numbers. Multivariate statistical analysis of the same datasets allowed for more detailed identification of spatial differences and a greater ability to identify and quantify subtle to moderate spatial differences. Disadvantages to the multivariate method include that it does not provide a classification system nor result in one easy to communicate number. In addition, although multivariate procedures do provide the ability to identify subtle effects, these may or may not be ecologically relevant. An alternative approach is suggested, pairing multivariate analysis with individual metrics to provide a way of labeling the subtle to moderate differences found in the multivariate analysis with some specific community differences.
Key words: multivariate analysis, index of biotic integrity, fish, benthic macroinvertebrates
Internet Services provided by|
Allen Press, Inc. | 810 E. 10th St. | Lawrence, Kansas 66044 USA
e-mail email@example.com | Web www.allenpress.com
All content is Copyright © 2004 SETAC