Detection and decision-making in environmental effects monitoring

The majority of environmental effects monitoring (EEM) frameworks that have been proposed compare selected indicator variables as a means of assessing whether significant changes in stressed ecosystems have occurred. Most are deterministic in nature and do not appropriately account for the natural variability and dynamics within the systems being comapred. This suggests that the comparative procedures should be statistically based and immediately raises the issue of whether the selected comparative procedures are to be used as decision-making tools or conclusive procedures. Conclusive procedures require a significant body of evidence before rejecting the postulated null hypothesis. The costs and time involved in environmental data collection accordingly bias action toward the maintenance of a status quo approach to environmental management. if, however, EEM is treated as a decision-making procedure, risk functions that include consideration of type I and II statistical error may be developed and combined with costs to select a minimum expected loss strategy for environemental management. Examples of the interpretative difficulties and conclusion reversal phenomena caused when EEM is used as a conclusive procedure are presented. In addition, risk functions appropriate for environmental management within an EEM context are constructed and applied. Only when such tools are fully developed and applied can EEM expect to have significant impacts on minimizing environmental degradation.