J.D.M. Belgers

The species sensitivity distribution approach compared to a microcosm study: a case study with the fungicide fluazinam.

We assessed the sensitivity of freshwater organisms (invertebrates and algae) to the fungicide Shirlan (active ingredient fluazinam) in single-species laboratory tests and in microcosms. Species sensitivity distribution (SSD) curves were constructed by means of acute toxicity data for 14 invertebrate species, since algae were much less sensitive. The EC(10)-based SSD gave a median HC(5) value of 0.6microgL(-1) and a 90% confidence interval of 0.1-1.9 microgL(-1). The EC(50)-based SSD gave a median HC(5) value of 3.9 microgL(-1) (90% confidence interval: 0.9-9.9 microgL(-1)). The microcosms were treated four times with Shirlan (concentration range: 0.4-250 microgL(-1)). Responses of the microcosm communities were followed. The 2 microgL(-1) treatment was the no-observed-effect concentration (NOEC(microcosm)). The 10 microgL(-1) treatment resulted in short-term effects on a few zooplankton taxa. Clear effects were observed at 50 and 250 microgL(-1). The responses in the microcosms were in line with the toxicity data for the tested lab species. The median EC(10)-based HC(5) and the lower limit EC(50)-based HC(5) were lower, and the median EC(50)-based HC(5) was slightly higher than the NOEC(microcosm). This is consistent with other studies that compared SSDs with responses in model ecosystems that received repeated applications of pesticides.

Can time-weighted average concentrations be used to assess the risks of metsulfuron-methyl to Myriophyllum spicatum under different time–variable exposure regimes?

We tested the effects of the herbicide metsulfuron-methyl on growth of the submerged macrophyte Myriophyllum spicatum under laboratory conditions using different exposure scenarios. The exposures of each scenario were comparable in the concentration × time factor, viz., the same 21-d time-weighted average (TWA) concentrations but variable in peak exposure concentrations (ranging from 0.1 to 21000 ng ai L⁻¹) and exposure periods (1, 3, 7, 14 or 21 d). To study recovery potential of the exposed M. spicatum plants we continued the observation on shoot and root growth for another 21 d in herbicide-free medium so that the total observation period was 42 d. Non-destructive endpoints, length and number of new shoots and roots, were determined weekly from day 14 onwards. Destructive endpoints, dry-weight (DW) of main shoots, new shoots and new roots, were measured at the end of the experiment (t=42 d). Metsulfuron-methyl exposure in particular inhibited new tissue formation but was not lethal to main shoots. On days 21 and 42 after start exposure, EC₁₀/EC₅₀ values for new tissues expressed in terms of peak concentration (=measured concentration during exposure periods of different length) showed large differences between exposure scenarios in contrast to EC₁₀/EC₅₀ values for days 21 and 42 expressed in terms of 21-d and 42-d TWA concentrations, respectively. At the end of the experiment (day 42), 42-d TWA EC(x) values were remarkably similar between exposure scenarios, while a similar trend could already be observed on day 21 for 21-d TWA EC(x) values. For the macrophyte M. spicatum and exposure to the herbicide metsulfuron-methyl the TWA approach seems to be appropriate to use in the risk assessment. However, the data from the toxicity experiment suggest that on day 21 also the absolute height of the pulse exposure played a (minor) role in the exposure - response relationships observed.

Effects of four fungicides on nine non-target submersed macrophytes

We tested the sensitivity of nine submersed macrophyte species to the fungicides chlorothalonil, pentachlorophenol, fluazinam, and carbendazim. Endpoints determined 3 weeks after the start of the treatment were based on shoot and root growth in water. Carbendazim proved not or only moderately toxic to these macrophytes. Pentachlorophenol and chlorothalonil were more toxic than fluazinam. Taking all endpoints into consideration, toxicity levels differed very substantially. Although root endpoints reflecting root growth were in some cases more sensitive than shoot endpoints, shoot growth endpoints like relative growth turned out to be more reliable than the root growth endpoints. Due to the large differences in the type of mode of action between fungicides, it is very difficult to predict their potential effect in the environment or even to predict whether non-target organisms like macrophytes are likely to be sensitive. Ideally, therefore, the registration of fungicides requires an extensive risk-assessment procedure, which also covers non-target groups like macrophytes.

Effects of linuron on a rooted aquatic macrophyte in sediment-dosed test systems.

Effects of linuron on the sediment-rooted aquatic macrophyte Myriophyllum spicatum L. were studied in sediment-dosed test systems following a proposed guideline with extended test duration. Sediment, pore water, overlying water and macrophyte shoots were sampled weekly for chemical analyses. Linuron was stable in the sediments. Sediment and pore water concentrations were in equilibrium after 48 h. Overlying water concentrations increased over time, but did not reach equilibrium with pore water concentrations and were 100 times lower. Mass balances showed a rapid uptake of linuron by macrophyte roots. Known pathways and the compounds properties support the conclusion that Myriophyllum takes up linuron from pore water directly through the roots. Hence, effects on macrophytes in this type of sediment toxicity test should be expressed in terms of pore water concentrations. Pore water concentration is the most relevant parameter for describing effects on macrophytes.