Use of Daphnia spp. for the ecotoxicological assessment of water quality in an agricultural watershed in South-Central Chile.

Because of the importance of surface waters from the Chillán River watershed (Chile) for recreation, agricultural irrigation, and the production of drinking water, local concern about river water quality has increased considerably during the last decade. Agricultural and forestry activities in the watershed, characterized by an intensive use of pesticides, are thought to play an important role in the generation of non-point-source pollution, whereas the discharge of urban wastewater from the city of Chillán constitutes a major point source of pollution. In the present investigation, acute and chronic laboratory bioassays using Daphnia spp. were conducted on surface water samples from 17 river stations located throughout the watershed. Sampling occurred on 6 occasions during a 16-month period (2000 to 2001) and included both high and low flow conditions. Almost all toxic effects observed in summer were directly related to the discharge of urban wastewater, whereas toxicity in rural areas was mainly detected during the winter period when rainfall and river flow are high. Toxicity test results were compared with measured physicochemical water-quality data. Mortality and alterations in reproductive success of Daphnia spp. were not consistently reflected in detected chemical pollution. With only one exception (atrazine), detected pesticide concentrations were below known toxicity levels. However, additive and synergistic effects of the presence of a mixture of pesticides could not be excluded as a possible cause of observed toxicity. At several stations, filtering of the water sample led to a strong decrease in toxicity, which suggests the presence of xenobiotics attached to the smaller sediment fraction. Inclusion of sediment chemical analysis and sediment toxicity testing in future work should therefore be encouraged. The presented approach provided information about the adverse effects of human activities on surface water quality in the watershed, not easily obtained from classical monitoring schemes. In specific cases, the approach may represent an economically attractive alternative to physicochemical analyses. Modifications to the proposed methodology should be introduced if the effects of intrastorm and interstorm variability of water quality are to be analyzed.