|
FISH MONITORING 273
<br />Based on the statistical properties of 1989-1992 data, we suggest modifications to sampling protocols that
<br />may lead to a more reliable tool. Firstly, the number of runs should be increased to a level that will provide a
<br />representative sample in each zone (15 in the inflow zones). Secondly, critical evaluation of individual metrics
<br />should continue. Those found to be unreliable, such as FHAI, should be dropped. Other metrics may require
<br />refinement for more effective use. In fact, TVA scientists continue to test and refine these methods.
<br />Finally, a pressing need is for independent data at several levels. Multiple samples within a reservoir
<br />should be used to determine biological patterns within and among major reservoir zones and among sites
<br />that are influenced by different reservoir habitats and local land use. Patterns among reservoirs with different
<br />management regimes also deserve study. Data from other systems of reservoirs would also provide a larger
<br />sample of possible conditions for reservoirs, improving the empirical basis for deriving metric expectations.
<br />Comparative data from other systems and better data on anthropogenic impacts (point and non-point
<br />sources, localized and watershed scale habitat alteration, sport and commercial fish harvest) should improve
<br />our ability to discover pattern and to use it to verify index (RFAI) performance.
<br />Although the success of approaches such as IBI for streams is encouraging, no simple formula is applicable
<br />to every system; this is especially true of highly modified environments. Our efforts to develop a multimetric
<br />index to assess the conditions of reservoirs met with some success. Differences among reservoir regions were
<br />shown for this study and an earlier study using rotenone data (Dionne and Karr, 1992). Much remains to
<br />be done, however, to provide a reliable RFAI.
<br />Sources of variation in RFAI scores need to be more carefully documented with a more extensive data set
<br />and with validation in independent data sets. Most importantly, success will be possible only with a long-
<br />term commitment to the collection of quality data and analysis that extracts and accounts for the relevant
<br />biological patterns. Premature use of RFAI to interpret monitoring data will damage both the credibility of
<br />existing programmes and progress toward the development of new and more reliable tools.
<br />We especially urge caution against using labels such as good, fair and poor to classify reservoirs. The
<br />temptation to do so is great in a regulatory climate that wants to label places with respect to some threshold
<br />(e.g. nominal or subnominal). A major point of this paper is to encourage agencies to resist this temptation
<br />until data and analysis can stand up to rigorous evaluations that warrant the application of these labels.
<br />ACKNOWLEDGEMENTS
<br />This work was possible because of the Tennessee Valley Authority's broadly based, long-term monitoring
<br />project. Numerous TVA biologists contributed throughout the project. Special thanks go to Gary
<br />Hickman, Tom McDonough, Al Brown and Ed Scott. Billie Kerans provided useful input on numerous
<br />occasions. The paper was improved with suggestions from two anonymous reviewers.
<br />REFERENCES
<br />Adler, R. W., Landman, J. C. and Cameron, D. M. 1993. The Clean Water Act 20 Years Later. Island Press, Washington, DC.
<br />320 pp.
<br />Angermeier, P. L. and Karr, J. R.. 1994. 'Biological ntegrity versus biological diversity as policy directives', BioScience, 44, 690-697.
<br />Berkman, H. E. and Rabeni, C. F. 1987. 'Effect of siltation on stream fish communities', Environ. Biol. Fishes, 18, 285-294.
<br />Cochran, W. G. 1977. Sampling Techniques, 3rd edn. Wiley, New York. 428 pp.
<br />Davis, W. S. and Simon, T. P. 1995. Biological Assessment and Criteria: Tools for Water Resource Planning and Decision Making. Lewis,
<br />Boca Raton.
<br />Dionne, M. and Karr, J. R. 1992. 'Ecological monitoring of fish assemblages in Tennessee River reservoirs' in McKenzie, D. H., Hyatt
<br />D. E., and McDonald, V. J. (Eds), Ecological Indicators, Vol. I. Elsevier Applied Science, London. pp. 259-281.
<br />Efron, B. 1981. 'Nonparametric estimates of standard error: the jackknife, the bootstrap and other methods', Biometrika, 68, 589-599.
<br />Fausch, K. D., Karr, J. R., and Yant, P. R. 1984. 'Regional application of an index of biotic integrity based on stream fish communities',
<br />Trans. Am. Fish. Soc., 113, 39-55.
<br />Gelwick, F. P. and Matthews, W. J. 1990. 'Temporal and spatial patterns in littoral-zone fish assemblages of a reservoir (Lake Texoma,
<br />Oklahoma-Texas, U.S.A.)', Environ. Biol. Fishes, 27, 107-120.
<br />Goede, R. W. 1986. 'Management considerations in stocking of diseased or carrier fish' in Stroud, R. H. (Ed.), Fish Culture in Fisheries
<br />Management. American Fisheries Society, Bethesda. pp. 349-355.
<br />Goede, R. W. and Barton, B. A. 1990. 'Organismic indices and an autopsy-based assessment as indicators of health and condition of
<br />fish', Am. Fish. Soc. Symp., 8, 93-108.
|