Summary of Presentations
The objective of this session is to identify scientifically robust indices that can be used to detect change in the state of ecosystems (or their component parts) relative to reference points in space or time. The authors present a variety of criteria for assessing ecosystem structure and function and discuss the strengths and limitations of each. The concept of scale is a recurring theme, particularly variability across time and space and across levels of biological organization (i.e. biosphere - region - ecosystem - community - population - individual - physiological - cellular -molecular).
Gray portrays the 1990’s as a period of increasing public awareness of environmental issues. Likewise, he notes that anthropogenic effects on ecosystems are being detected at very large scales, and new techniques suggest that even the open oceans may show evidence of human impact. Gray also presents case studies to demonstrate that indices of impact applied to the same system can differ in non-obvious, yet significant ways. For example, estimated impacts of the Ekofisk oil platform on benthic community structure ranged from an area of 3 km2 to 27 km2, depending on the type of statistical analysis applied to the data. An important theme of his presentation was the need to place greater emphasis on designing statistically sound sampling programs and on incorporating statistical power analyses into assessment schemes.
An array of techniques is rapidly being developed to detect responses of molecular, cellular and physiological systems to human impact. These so-called biomarkers are specifically suited for detecting change in response to chemical pollutants. Vasseur presents examples of the types of molecular biomarkers (e.g. changes in the structure and function of DNA) that can be used to predict carcinogenic risk and discussed the limitations inherent in such measures.
Although impacts of human activities may be detected at the ecosystem level of organization, it is generally difficult to identify cause-effect relationships at this level. Dethlefsen outlines a suite of changes that have occurred in the North Sea on regional and local scales and at many different levels of the biological hierarchy. These include decreases in the abundance of planktonic diatoms and benthic bivalves, increases in the abundance of planktonic flagellates and benthic polychaetes, increases in the frequency of toxic blooms and fish diseases, and malformations of pelagic fish embryos. He stressed the difficulties in estimating the extent to which such changes have arisen in response to anthropogenic versus natural factors and suggested that indices of ecosystem change may be useful as qualitative, but not quantitative, measures of human impact.
For the purpose of understanding the mechanistic bases of ecosystem change, linking responses of ecosystem components occurring at different levels of spatio-temporal and biological organization is critical. For the purposes of assessing or predicting system change, the elements and processes upon which studies are focused should be determined by the objectives of the study. Finding mechanistic links to higher or lower levels of organization may not be as important as ensuring that a suite of indicators is applied which are able to detect change at several levels. However, because we need not only to estimate the potential for change but the seriousness of the change as well, linkages across scales of measurement are ultimately necessary.
While it is recognized that changes in ecosystem components vary in sensitivity and information content, such differences remain to be explicitly quantified. For example, an important assumption of the application of various molecular, cellular and physiological biomarkers is that these parameters are more sensitive (i.e. respond after a shorter duration of impact or to a lower degree of impact) than changes at the community or ecosystem level. However, this assumption has not been rigorously tested.
Whereas ecologists recognize that critical ecosystem elements and processes operate at different scales, we know very little about how linkages across scales vary among ecosystems. Costanza discusses a new research program that he and coworkers designed to quantify linkages among ecosystem components across spatial and temporal scales. The Mutltiscale Experimental Research Center consists of artificial ecosystems built in a range of sizes and incorporating different degrees of complexity. Models predicting ecosystem performance are constructed for each system; model predictions are tested within and across systems as well as in natural ecosystems. This program aims to provide an explicit understanding of the role of spatial and temporal constraints on ecosystem performance and to determine the extent to which linkages among scales are ecosystem dependent.
It is widely agreed that indices of ecosystem change have been far too heavily concerned with chemical measures to the detriment of important biological criteria. Karr provided convincing evidence that the narrow focus on chemical water quality in attempts to fulfil mandates of environmental legislation has not succeeded in protecting ecosystem elements and processes. He attributed this to a historically narrow focus of environmental legislation on human health and to the fact that chemical parameters can often be precisely measured. Karr emphasizes that because ecosystems are multivariate, a multivariate approach for ecosystem assessment is required. He also argues that the standards against which ecosystem impacts are assessed should be regional, rather that universal, because systems vary geographically. Karr has applied these principles in developing an index of biotic integrity. The index, which was originally developed to evaluate human impacts on American midwestern streams, is calculated as a sum of 12 biological attributes (i.e. species types and numbers, trophic position of resident species, and condition) of a resident fish assemblage. This approach is being expanded to other assemblage and ecosystem types and provides a substantial improvement over chemically-based water quality measures.
Unable to display preview. Download preview PDF.