Ecotoxicological threshold levels of a mixture of herbicides (atrazine, diuron and metolachlor) in freshwater microcosms

Twelve indoor, plankton-dominated, freshwater microcosms (600 l) were used to study the effect of a mixture of herbicides on structural and functional aspects of these ecosystems. The EC50, 72 h values of the most susceptible standard test alga Selenastrum capricornutum (EC50, atrazine=54 μg l−1, EC50, diuron=15 μg l−1, EC50, metolachlor=56 μg l−1) were used as a starting point for the dosage applied in the microcosms (dosages: 0, 0.01, 0.03, 0.1, 0.3, 1× EC50). The microcosms were exposed to chronic levels for 28 days and subsequently monitored for 4 more weeks. The following effects were observed: (1) direct effects became apparent from an initial drop in photosynthesis efficiency, pH and oxygen concentration and a decrease in the abundance of several phytoplankton taxa at the 0.3 × EC50 treatment level and higher. (2) Fourteen days post application an increase in the abundance of several phytoplankton taxa (Chlamydomonas sp. and Stephanodiscus/Cyclotella) was observed; oxygen concentrations recovered while alkalinity, conductivity and total inorganic nitrogen were elevated. (3) Effects on fauna were minor. Daphnia galeata showed a decreasing trend and the cyclopoid copepods an increasing trend at the end of the experiment. Multivariate statistical analyses demonstrated no effects of any treatment level on the zooplankton community. Effects were reported for the phytoplankton community at dose levels of 0.3 × EC50 and higher. On species level the most sensitive taxon was Chlorophyceae coccales. For this taxon a NOEC at the dose level of 0.01 × EC50 was calculated. This effect however was relatively small in magnitude and merely based on an increase in numbers in the control and lowest treated microcosms rather than a decrease in numbers in all other treatments. The standards based on algal toxicity data, as adopted by the Uniform Principles, consist of a safety factors of 0.1 to be multiplied with the EC50. The NOEC of coccales was lower than 0.1 × EC50. All other observed variables in this aquatic ecosystem were sufficiently protected against the mixture of herbicides by the safety factor as proposed in the Uniform Principles. This revised version was published online in August 2006 with corrections to the Cover Date.