Journal ofAquatic EcosystemHealth 1: 283--293, 1992. M. Munawar(ed.),Assessing Aquatic EcosystemHealth:Rationale, Challenges,and Strategies. © 1992 KluwerAcademic Publishers. Printedin the Netherlands. A review and evaluation of study design considerations for site-specifically assessing the health of fish populations Kelly R. Munkittrick Great Lakes Laboratory for Fisheries and Aquatic Sciences, Department of Fisheries and Oceans, 867 Lakeshore Road, P.O. Box 5050, Burlington, Ontario, Canada L 7R 4146 Keywords: fish population, growth, reproduction, assessment, health Abstract. Since there are not any well developed procedures for site-specifically evaluating the health of fish populations, most field surveillance programs have been restricted to collecting information on chemistry, toxicity, bioaccumulation, biochemical alterations or in situ benthic community structure. Identification of the mechanism and significance of contaminant effects on fish populations depends on the ability to identify changes and to distinguish changes in survival, food availabilityor food conversion efficiency, from coincidental changes associated with alterations in habitat or natural variability.Preliminary identification of the characteristics of adult fish can be used to cost-effectively focus financial resources on the alterations which are of relevance to interpretation of impacts and identification of causal factors. Proper interpretation requires that appropriate attention be given to monitoring level and strategy, selection of species and timing of sampling, sample size requirements and choice of reference site. 1. Introduction Techniques are not well developed for examining and interpreting the health of fish populations exposed to contaminants (Cairns et al., 1984; Hodson et al., 1990) and the diversity of whole organism responses recorded in field studies make it necessary to assess the significance of changes on a site-specific basis (Munkittrick & Dixon, 1989a). Identification of the mechanism and significance of contaminant impact depends on the ability to distinguish changes in survival, food availability or food conversion efficiency asso- ciated with contaminants from coincidental changes associated with alterations in habitat or natural variability. Most field collections are con- ducted for monitoring or surveillance purposes, and have generally been restricted to collecting information on chemistry, toxicity, bioaccumula- tion, biochemical alterations, or in situ benthic community structure. This data can seldom be extrapolated to predict or describe changes at the level of fish populations. It is more cost-effective to identify whether or not population-level changes exist, and then focus financial resources on the impact points to determine the causal factors. Growth, reproduction, and morphological parameters can be used to cost-effectively direct an assessment of the significance of changes, and to help design appropriate follow-up studies to elucidate the mechanism and cause of impacts. Proper interpretation requires that careful con- sideration be given to factors involved in the study design, such as choice of monitoring level and strategy, selection of species and timing to sam- pling, sample size requirements, and selection of an appropriate reference site. 2. Monitoring level and approach The choice of approach and monitoring level will affect the assessment process by influencing the ability to detect and trace the response, to estab- lish cause and effect, and to predict the conse- quences and ecological relevance. Within studies of fisheries health, there are two possible ap- proaches. The bottom-up approach attempts to initiate monitoring at the biochemical level, since it is at this level that direct damage by toxicants is initiated (Hodson, 1990; Niimi, 1990). Monitor- ing within the individual organism allows the rapid detection of responses, although the changes usually have unknown consequences at the popu- lation or community level. Therefore, the ability to estimate, evaluate, or judge the significance or