Detecting Ecosystem Responses to Anthropogenic Stress

Abstract

Recent ecological work on aquatic populations, communities, and ecosystems is reviewed for advances which show promise as early indicators of anthropogenic stress in aquatic ecosystems. Work at the Experimental Lakes Area (ELA) in northwestern Ontario indicates that among the earliest of responses to stress are changes in species composition of small, rapidly-reproducing species with wide dispersal powers such as phytoplankton, and the disappearance of sensitive organisms from aquatic communities. Work elsewhere illustrates that the incidence of morphological abnormalities in benthic invertebrates is also highly sensitive to pollution stress. For several categories of pollutants, this sensitivity of benthic organisms may be due to the greater concentrations of pollutants in sediments than in the water column. Variables reflecting ecosystem functions such as primary production, nutrient cycling, and respiration, were not altered by eutrophication, acidification, or cadmium addition at ELA, and are relatively poor indicators of early stress. Species diversity of phytoplankton was also insensitive to low levels of stress. Mesocosm experiments appear to be fruitful for addressing chemical- or plankton-related problems, but are less useful for addressing community- or ecosystem-level questions. Among population-level approaches, life-table population studies of invertebrates appear to be the most sensitive early indicators of stress on ecosystems.Relative sensitivities of freshwater and forested terrestrial ecosystems exposed to airborne pollutants are compared. Primary production seems to be reduced at a much earlier stage of air pollution stress in terrestrial ecosystems than in aquatic systems. Soils, like lake sediments, tend to be sinks for pollutants. This may protect the pelagic regions of lakes from influxes of toxins that would occur if watersheds and sediments were unreactive, but cause additional stresses to the fauna and flora of soils and sediments. In extreme cases, high concentrations of toxins may inhibit the replacement of terrestrial producers.The importance of long-term monitoring in distinguishing natural from anthropogenic stress is discussed. It is suggested that paleoecological techniques be rapidly developed and calibrated with whole-ecosystem experiments to resolve certain inadequacies of past monitoring records.