2866 Environmental Toxicology and Chemistry, Vol. 18, No. 12, pp. 2866–2874, 1999  1999 SETAC Printed in the USA 0730-7268/99 $9.00 + .00 EFFECTS OF A PULSED TREATMENT WITH THE HERBICIDE AFALON (ACTIVE INGREDIENT LINURON) ON MACROPHYTE-DOMINATED MESOCOSMS. II. STRUCTURAL RESPONSES GERBEN J. VAN GEEST,² N ATHALIE G. ZWAARDEMAKER,² R ENE ´ P.A. VAN WIJNGAARDEN,‡ and JAN G.M. CUPPEN*² ²Department of Aquatic Ecology and Water Quality Management, Wageningen Agricultural University, Ritzema Bosweg 32a, 6703 AZ Wageningen, The Netherlands ‡DLO Winand Staring Centre for Integrated Land, Soil and Water Research (SC-DLO), P.O. Box 125, 6700 AC Wageningen, The Netherlands ( Received 14 September 1998; Accepted 25 March 1999) Abstract—Outdoor mesocosms (experimental drainage ditches) were treated repeatedly with linuron at nominal concentrations of 0, 0.5, 5, 15, and 50 g/L. After each of three treatments (with 4-week intervals), the systems were kept static for 1 week. Subsequently, they were flushed with nonpolluted surface water with a residence time of 5 d until the next treatment. Species composition and abundance of macrophytes, periphyton, phytoplankton, zooplankton, and macroinvertebrates were examined at prefixed time intervals. During the treatment periods, significant decreases in oxygen concentration and pH were observed, whereas the systems recovered during the flushing periods. The effects of the pulsed treatments on the structural endpoints (biota) were rarely significant and consistently were difficult to interpret. We conclude that under these experimental conditions, pulsed treatment with linuron causes only minor and negligible effects on the abundance of organisms in mesocosms, even when maximum permissible concentrations are exceeded by about 80 times. Risk assessment based on the more sensitive physiologic and functional endpoints suggested that the safety factor of 0.1 times the acute EC50 of the most sensitive standard alga (Scenedesmus acutus), as adopted in the Uniform Principles of the European Union, is sufficient to protect the community of the mesocosms in an exposure regime that was realistic for Dutch drainage ditches, using a photosynthesis inhibitor as the benchmark compound. Keywords—Mesocosms Ecosystem structure Linuron Risk assessment INTRODUCTION To determine if current standards for pesticides as set by the European regulatory authorities adequately protect aquatic ecosystems, we conducted an experiment with the herbicide Afalon (Hoechst Holland N.V., The Netherlands) (active in- gredient linuron) under realistic field conditions. Linuron was chosen as the benchmark compound for this purpose, because photosynthesis-inhibiting urea-based herbicides are widely used in agriculture. At the chosen treatment levels, we ex- pected the most pronounced (direct) effects on functional and structural endpoints of the photosynthetic community mem- bers. These expectations were based on results of a previous experiment with linuron [1,2]. However, expected direct ef- fects can lead to less predictable indirect effects on, for ex- ample, the animal community [2,3]. We therefore studied flo- ristic and faunal endpoints. This article is the second of two dealing with the impact of a repeated application of linuron on ecosystem structure and functioning. The first article of this series focused on functional parameters of the ecosystem [3]. This second part considers structural endpoints. The fate of linuron in the me- socosms has been reported in Crum et al. [4]. Laboratory data on the toxicity of linuron to algae estab- lished an EC50 of 6 g/L for the green alga Scenedesmus acutus [5]. The maximum permissible concentration of this herbicide, according to lower-tier criteria of the Uniform Prin- * To whom correspondence may be addressed (jan.cuppen@aqec.wkao.wau.nl). ciples [6], is a factor of 10 lower (i.e., 0.6 g/L). In addition to describing the effects of temporary exceedances of the max- imum permissible concentration for linuron on the structure of the aquatic communities in the mesocosms, we elected to focus on the following question: Does the current maximum permissible concentration for linuron adequately protect aquat- ic ecosystems, in the event of repeated applications of the herbicide in a realistic hydrologic regime? We also discuss the relation between functional and structural ecosystem responses in the herbicide-stressed experimental ditches. Ditches are sig- nificant ecological receptors of agricultural chemicals in The Netherlands. MATERIALS AND METHODS The assessments of herbicide effects were made by use of a combination of field sampling (phytoplankton, zooplankton, macroinvertebrates, and macrophytes) in outdoor mesocosms and in situ bioassays with Myriophyllum spicatum and pe- riphyton communities. Experimental design of mesocosm study Ten experimental drainage ditches (mesocosms) were used for this experiment. Each mesocosm had the following char- acteristics: length, 40 m; width at the water surface, 3.4 m; water volume, 60 m 3 ; and mean water depth, 0.5 m. The aquatic community in the mesocosms resembled that of numerous oth- er macrophyte-dominated drainage ditches. The herbicide Afalon was chosen as a benchmark chemical because of its propensity for entry into aquatic ecosystems.