Derivation of predicted No-effect concentrations for lindane, 3, 4-dichloroaniline, atrazine, and copper.

Environmental risk assessment is a key feature of regulations controlling the placing of new, and the maintenance of existing, chemicals products in the market place. For example, European Commission Directive 93/67/EC on Risk Assessment for New Notified Substances and Commission Regulation (EC) No. 1488/94 on Risk Assessment for Existing Substances requires that risk assessments be carried out for new and existing substances in the European Community. The process of environmental risk assessment seeks to determine the balance of probability of species and communities being damaged by chemical releases. The process relies upon a valid estimation of a predicted environmental concentration (PEC) in relevant environmental compartments and a predicted no effect concentration (PNEC) below which the organisms present in that compartment are unlikely to be significantly affected. If the PEC exceeds the PNEC there is a potential for damaging effects to occur. This article focuses on the determination of PNECs for risk assessment. Methods for determining a PNEC described in OECD Monograph 26 (1989, Report of the OECD Workshop on Ecological Effect Assessment, Paris, France, have been applied to data derived for the four chemicals lindane, 3,4-dichloroaniline, atrazine, and copper in a series of collaborative research projects funded by the European Commission.

Development of methods for evaluating toxicity to freshwater ecosystems.

This article presents a summary of a collaborative research program involving five European research groups, that was partly funded by the European Commission under its Environmental Research Program. The objective of the program was to develop aquatic toxicity tests that could be used to obtain data for inclusion at Level 2 of the Risk Evaluation Scheme for the Notification of Substances as required by the 7th Amendment to EC Directive 79/831/EEC. Currently only a very limited number of test methods have been described that can be used for this purpose and these are based on an even smaller number of test species. Tests based upon algae (Chlamydomonas reinhardi, Scenedesmus subspicatus, and Euglena gracilis), protozoa (Tetrahymena pyriformis), rotifera (Brachionus calyciflorus), crustacea (Gammarus pulex), and diptera (Chironomus riparius) were developed. The tests encompassed a range of end points and were evaluated against four reference chemicals: lindane, 3, 4-dichloroaniline (DCA), atrazine, and copper. The capacity of the tests to identify concentrations that are chronically toxic in the field was addressed by comparing the effects threshold concentrations determined in the laboratory tests with those determined for similar and/or related species and end points in stream and pond mesocosm studies. The lowest no-observed-effect concentrations (NOEC), EC(x), or LC(x) values obtained for lindane, atrazine, and copper were comparable with the lowest values obtained in the mesocosms. The lowest chronic NOEC determined for DCA using the laboratory tests was approximately 200 times higher than the lowest NOEC in the mesocosms.

Design and construction of model stream ecosystems.

A series of 12 outdoor model stream mesocosms was designed to evaluate the effects of chemicals and mixtures on the biota in stream ecosystems. An integrated design plan incorporated physical, chemical, and biological factors as well as the specific experimental objectives and effects parameters to be evaluated. Analysis of biological assemblages such as macroinvertebrates and periphyton in the model stream mesocosms demonstrated the presence of diverse and sensitive taxa. These model stream mesocosms also included the ability to evaluate responses of sentinel fish species such as Pimephales promelas and Lepomis macrochirus as well as macrophytes. The design and construction of the stream are discussed in detail and a brief description of a typical experimental protocol is provided. Experiments, to date, have spanned 14-30 days pretreatment "colonization" with 30 days of treatment and 15 days of posttreatment observation. Precision and accuracy of test chemical delivery to the stream have been excellent and the overall experimental design has been useful for delivering threshold toxicity concentrations and evaluating ecosystem potency for the chemicals tested.

Hazard assessment in freshwater ecosystems.

Hazard assessment of chemicals in freshwater environments depends on comparing concentrations that are expected to occur in water and sediment, i.e. expected environmental concentrations (EEC), with those that are estimated to have no biological effects, i.e. the no-observed effect concentrations (NOEC). The difference between these two estimates is the margin of safety. The EEC can be estimated from data for chemical release rates, physicochemical properties and environmental parameters that affect transport and transformation. The NOEC can be estimated from the results of toxicity tests using aquatic plants, invertebrates and fish. When making these estimates it may be necessary to extrapolate from relatively limited laboratory data to the real world. Inevitably, this involves some degree of uncertainty. Such uncertainty can often be resolved by carrying out controlled field tests, using small, outdoor enclosures (microcosms), relatively large, outdoor ponds (mesocosms) and experimental streams. In this paper the advantages and disadvantages of various experimental approaches and systems will be reviewed.

An outdoor artificial stream system designed for ecotoxicological studies.

Six outdoor artificial streams were designed to simulate natural stream environments. Their primary intended use is to define acceptable threshold toxicity concentrations for chemicals and effluents in the aquatic environment and so provide data that can be used to define water quality criteria more accurately than is possible on the basis of data obtained solely from laboratory tests. Each stream was divided into pool and riffle sections that were colonized by communities of periphyton and invertebrates. They were operated as partly flow-through, partly recirculating systems. The design and construction of the streams are described in detail, and brief descriptions are given of methods of establishing aquatic communities, measuring water quality parameters, and treating the streams with chemicals and effluents. The performance of the streams in environmental research is illustrated by reference to typical data for primary production, for invertebrate diversity and abundance, and for chemical concentrations measured during an experiment of two weeks duration.

A method for evaluating effects of toxic chemicals on the productivity of freshwater ecosystems.

The finding that trout apparently consumed more invertebrates than were produced in the Horokiwi stream has been described as "Allen's paradox." One explanation for this is that invertebrate sampling techniques underestimate the standing crop. If so, trout growth might be a relatively sensitive indicator of the effects of toxic chemicals on the productivity of experimental ponds. In a replicated pond experiment, methyl parathion was applied at concentrations toxic to invertebrates but not to fish. Three 50-m3 ponds were treated at 40 micrograms liter-1 and three at 10 micrograms liter-1, and there were three untreated controls. Each pond was stocked with eight small, individually marked rainbow trout. The fish were removed by electrofishing 3 weeks after treatment. Mean growth rate of fish in control ponds was 6.3% per day, 4.3% per day in ponds treated with 10 micrograms liter-1, and 3.7% per day in ponds treated with 40 micrograms liter-1. Effects were significant at the 1% level. The standing crop of invertebrates was apparently insufficient to support the growth of the fish, an indication that the active, predatory rainbow trout is more efficient at sampling invertebrates than standard limnological procedures.

Fate of 2,5,4'-trichlorobiphenyl in outdoor ponds and its uptake via the food chain compared with direct uptake via the gills in grass carp and rainbow trout.

A field experiment was carried out to investigate the fate and the extent to which food chain transfer of a PCB congener (2,5,4'-trichlorobiphenyl; 3-CB) could affect its potential for bioaccumulation. Three 35-m3 ponds were each stocked with 25 rainbow trout (a carnivore) and 20 grass carp (a herbivore). 3-CB was supplied to each pond at a nominal concentration of 14 micrograms liter-1. Samples of water, sediment, grass carp, rainbow trout, aquatic plants, and invertebrates were removed at intervals from 0 to 28 days after treatment and residues of 3-CB were determined using gas-liquid chromatography with electron-capture detection. The fate of 3-CB in the ponds was determined by transport rather than degradation processes. Evaporation accounted for 86-87% loss and sorption onto sediment and biota for 11-12% loss of 3-CB from the pond water after 28 days. The kinetics of transport between water, air, sediment, and biota were studied after fitting the data to a three-compartment model. This model was used to calculate rates of evaporation (ke) and sorption. The calculated value for ke was in good agreement with predictions based on fundamental relationships between Henry's constant, windspeed, and ke. Residues of 3-CB in rainbow trout accumulated to a significantly greater extent (P less than 0.05) than in grass carp. This difference could not be explained by differences in growth rates, distribution of lipids, or by a difference in rates of metabolism. The most plausible explanation is that there was a difference in accumulation of 3-CB residues via the food chain. This experimental work supports the conclusion, suggested by a modeling approach of other workers, that food chain accumulation of PCBs can be an important route for uptake when environmental concentrations are very low.

Fate and biological effects of methyl parathion in outdoor ponds and laboratory aquaria. I. Fate.

Three outdoor ponds were treated with methyl parathion (MEP) applied beneath the water surface at a concentration of 100 micrograms liter-1. Laboratory aquaria containing either tap water, pond water, tap water plus plants, tap water plus sediment, or tap water plus sediment and plants were similarly treated. Samples of water, sediment, and fish were analyzed for residues of MEP. The rate of loss from water and concentrations found in sediment were compared with predictions based on a calculated rate of biodegradation and a sediment:water partition coefficient. The rate of loss of MEP from pond water isolated in an aquarium was similar to the predicted rate. However, the rate of loss from outdoor ponds, or from aquaria containing plants and sediment, was greater than predicted. MEP was not detected in sediment even though predicted concentrations far exceeded the limit of detection. These results are discussed and it is suggested that the rate of biodegradation in shallow bodies of water may be determined predominantly by bacteria attached to sediments and plants, rather than by planktonic bacteria. Bioaccumulation in fish was predicted from empirical equations based on the octanol:water partition coefficient. Observed values were in good agreement with predictions.

Fate and biological effects of methyl parathion in outdoor ponds and laboratory aquaria. II. Effects.

Methyl parathion (MEP) applied to three outdoor ponds at a nominal concentration of 100 micrograms liter-1 was toxic to some species of aquatic insects and crustaceans but not to fish. The spectrum of toxicity was similar to predictions based on a literature survey of data obtained from laboratory tests. Various secondary effects occurred that could not be predicted from laboratory toxicity tests. An increase in populations of Diaptomus in treated ponds was probably caused by mortality of predators and competitors. A bloom of filamentous algae which then collapsed, leading to severe depletion of dissolved oxygen and fish deaths, may have been triggered by mortality of herbivorous mayflies and daphnids. The growth of juvenile rainbow trout in treated ponds was significantly less than in untreated ponds. On the other hand their growth in laboratory aquaria was not affected when rainbow trout were exposed to higher concentrations of MEP than occurred in the outdoor ponds. It was concluded that growth of rainbow trout in the ponds was probably affected by mortality among aquatic insects and crustaceans on which they feed.

The effect of the molluscicide N-tritylmorpholine on transmission of Fasciola hepatica.

Field trials were carried out in north Devon to investigate the relationship between molluscicide treatment of pastures and control of liver fluke infection in sheep. Seven tracer lambs per plot were used to estimate the infectivity of 17 pairs of 0.20 hectare plots. One plot in each pair was treated with the molluscicide Frescon (N-tritylmorpholine). There was a highly significant difference (P less than 0.001) between the numbers of Fasciola hepatica recovered from lambs grazed on treated and untreated plots in the period after molluscicide treatment. The overall degree of snail control achieved by one application of molluscicide to 17 plots was about 90 per cent and this was matched by a comparable degree of liver fluke control. There was considerable variation between the plots and there was no simple correlation between snail numbers and liver fluke numbers (r = -0.03). The use of molluscicides is discussed in relation to the biotic potential of Lymnaea truncatula and environmental factors which limit its population growth.

Field trials to evaluate the effectiveness of the molluscicide N-tritylmorpholine in irrigation systems.

In field trials with the molluscicide N-tritylmorpholine (Frescon, WL 8008), a prolonged low-dosage technique has been developed for use in irrigation systems. In Tanzania a dose of 0.025 ppm applied for 30 days to the headworks of a 5000-acre (2025-ha) irrigation system gave effective control of Biomphalaria pfeifferi, the snail host of Schistosoma mansoni, for a period of 3-4 months. This was comparable to the effectiveness of other molluscicides and other dosage regimes but the technique had the advantages of simplicity, lower cost and lack of toxicity towards fish or other aquatic fauna. Similar trials have been carried out in Egypt and Southern Rhodesia. Theoretical and practical considerations involved in the choice of dosage regimes are discussed and a general formula is suggested for choosing dosage regimes in irrigation systems.

The rate of hydrolysis of the molluscicide N-tritylmorpholine.

N-Tritylmorpholine (Frescon, WL 8008) shows promise for the control of the snails that are intermediate hosts of trematodes. A knowledge of the rate of hydrolysis of the compound is important in connexion with its field use, and the rate of hydrolysis of (14)C-N-tritylmorpholine has been found to be dependent on the pH of the water and on the initial concentration of the compound. This rate is relatively high at low pH values (< 7.0), but decreases logarithmically as the pH increases.A prolonged low-dosage technique has been developed for the application of N-tritylmorpholine to irrigation systems. The present results indicate that, under these conditions, effective control of the snails is likely to be achieved in water of pH > 7.5, whereas in water of pH < 7.0 the compound is unlikely to penetrate very far downstream before losing its molluscicidal activity. In waters in the pH range 7.0-7.5 the use of higher concentrations and shorter application times may be desirable.