2) Mesocosm experiment 125flm consisted of 635 m³ cosms pulse-exposed to Trahalomethrin 5 times during 65 days. Because of rather high through-flow 90% of the pecticide was washed-out within 24 h after each dosage. Hence, the calculated total exposure concentration (=sum of each dosage) inevitably will grossly overestimate the actual concentrations (i.e. the experiment may not qualify for a true stagnant water experiment).

As shown in Table 4 the highest predictability (Q²(cum)) was obtained for the PLS model based on mesocosm experiments where both sediment and macrophytes were present in the test system. However, an almost as high Q²(cum) were obtained for the PLS models for mesocosm experiments with sediment but without macrophytes in the test system. On the other hand a much lower predictability (Q² = 0.481) was obtained when the PLS model was developed including all experiments carried out in stagnant water (i.e. including laboratory experiments). Hence, the larger similarity in the constituents of the different mesocosms (and closer resemblance to natural conditions) the higher is the predictability of toxic effects based on the various physical, chemical and toxicological properties of an experiment (see Table 2). Therefore, a PLS model for macroinvertebrates with an acceptable predictability needs to be based on mesocosm experiments containing sediments and preferentially macrophytes in the test system. The interpretation of the PLS model is therefore based on the PLS model for mesocosm data with both macrophytes and sediment present in the test systems, but excluding 2 experiments (76tll and 125flm) i.e. a total of 9 experiments.

5.4.1 Interpretation of the PLS model for macroinvertebrates

An overview of the model selected in the previous section is shown in Table 5.

Table 5.
Predicted variation of the significant axis of the PLS model selected for macroinvertebrates. Q²: Variation in the Y matrix predicted from the variation in the X matrix by the current axis. Q²(cum): Cumulative variation in the Y matrix predicted from the variation in the X matrix.

As shown in Table 5 the first PLS axis predicts 52.7 % of the variation in the Y matrix (i.e. the toxic response) from the variation in the X matrix (system characteristics and toxicological and physico-chemical characteristics of the pesticide). The second PLS axis predicts 20.7% of the variation in the Y matrix from the variation in the X matrix. The fact that Q²(cum) is lower than the sum of Q² for the two axis (i.e. 0.527 + 0.207) indicates some overlap between the predictions of the first and second PLS axis.

Figure 8.
Weights (loadings) of variables contributing to the first PLS axis for macroinvertebrates. Day number through OECD represent variables in the X-matrix while the responses (LOEC) of the different macroinvertebrate groups are shown at right. Weights with opposite signs can be interpreted as being negatively correlated (e.g. INTERVAL between pesticide doses and toxic response of either macroinvertebrate group), while weights with identical signs can be interpreted as being positively correlated (e.g. mesocosm Depth, Number of dosings and toxic response of macroinvertebrates).

Figure 9.
Weights (loadings) of variables contributing to the second PLS axis for macroinvertebrates.

As described in Section 4.2 the PLS axis was interpreted on the basis of the weights (loadings) of the variables to each PLS axis. The loading plots are shown in Figs 8 & 9.

	The first PLS axis showed positive loadings of almost equal magnitude in all groups of macroinvertebrates, which means that the four groups are equally sensitive to pesticides and the conditions in the mesocosms.
	For the second PLS axis a much lower positive loading is obtained for the infauna (i.e. borrowing animals) than for the other functional groups. Thus the two PLS axis seems to be indicative of different toxic responses of the macroinvertebrates according to their habitat.
	For the X variable longitude positive loadings are seen for both PLS axis, whereas for latitude, negative loadings are seen for both PLS axis. Thus all macroinvertebrate groups in the mesocosms seem to be most sensitive when the experiments are conducted at high latitudes and low longitudes in mesocosms. Therefore, toxic effects at lower concentrations are expected with increasing distance from Equator. A likely explanation is probably related to a slower turn-over of populations at high latitudes, i.e. fewer generations each year at lower temperatures. Therefore, recovery of populations affected by pesticide exposure takes longer time at northern latitudes. The positive loading for longitude suggests that organisms in experiments conducted in USA are less sensitive than the macroinvertebrates in experiments conducted in Europe. A possible explanation could be that most mesocosm experiments in USA, but not in Europe, are carried out with fish present in enclosures, which may override or mask the effect of pesticides.
	For the first axis the "Interval" between pesticide dosings correlated negatively with LOEC of the macroinvertebrates. Thus by increasing the interval between dosings a lower LOEC will result. This may be due to the relativly long generation time of most macroinvertebrates. Hence, recovery will be hampered if pesticides are dosed at intervals close to the generation time. Interestingly, the related variable "Number of pesticide Dosings" correlated positively with LOEC’s (especially on the second PLS axis), meaning that a low but persistent pesticide concentration will have a lower effect on the macroinvertebrates than a high but temporary pesticide concentration.
	High positive loading for the variable Depth on both axis could be related the fate of pesticides in mesococms. At decreasing mesocosm depth a larger fraction of the pesticides will end up in the sediment compartment and thus increase the exposure to the sediment living macroinvertebrates.
	For the X variables Log KD, Log KOW and Interval (between dosings) significant negative loadings are obtained for the first PLS axis. The loadings of these variables to the second axis were considered as insignificant. Thus the toxic response of the macroinvertebrates expressed by the first PLS axis are most pronounced for hydrophobic, adsorbable (high log KD) substances added to the mesocosms over a long time period (Interval). In effect, the toxic response of the macroinvertebrates expressed by the first PLS axis might be considered as a long term response probably involving sorption of the pesticides to particles, sedimentation of the particles and a subsequent exposure of the organisms to pesticides adsorbed to sediment particles.
	High positive loadings to the second PLS axis are obtained for the X variables hazard concentrations (HC5,50 and LC50/10 (i.e. OECD procedure)), while their significance on the 1. axis are considered insignificant. As stated above the loadings of the infauna to the second axis is insignificant (see Fig. 9). Hence, the toxic response of the macroinvertebrates expressed by the second axis can be considered as a short-term response attributable to a direct exposure through the water phase, which consequently do not affect the macroinvertebrates living within the sediment. As the hazard concentrations are calculated from standardised short term (48-96 h) toxicity tests the loadings (correlations) are expected.
	The variables day number (i.e. season) and volume are considered as insignificant (low loadings and of opposite sign).

A summary of the effect of experimental mesocosm and pesticide characteristics is shown in Table 6.

Table 6.
Summary of influences of mesocosm and pesticide characteristics and toxicology (extrapolated effect concentrations) on toxic response on macroinvertebrates. Ý Ý = major decrease in toxicity; Ý = minor decrease in toxicity (i.e. higher LOEC); ß ß = major increase in toxicity; ß = minor increase in toxicity (i.e. lower LOEC); - = no effect. See Table 2 for an explanation of system variables.

The arrows in Table 6 indicate if numeric increases in system variables (see Table 2) will decrease ( Ý ) or increase ( ß ) the toxic response in the different groups of macroinvertebrates. Double arrows denote that a system variable have the same significant influence in both PLS axes.

5.4.2 Predicting effect concentrations for the macroinvertebrates with the aid of the PLS models

The observed and predicted effect concentrations with associated 95 % confidence intervals for all the mesocosm experiments analysed by the PLS model appear in Table 7 and Figure 10. The asymmetric confidence interval is due to the logarithmic transformation of data before the PLS analysis. When the lower limit of the confidence intervals was below 0 (seemingly a bug occurring in the Simca program during log- and antilog transforming procedure) the lower limit of the confidence interval was set to 0.

Table 7.
Comparison of observed and predicted LOEC (µg l^-1) with associated 95 % confidence interval for the mesocosm experiment calculated with the PLS model for macroinvertebrates. Macroinvertebrates have been grouped according to mode of feeding (non-predatory/predatory) and habitat (epifauna = macroinvertebrates living on sediment surface; infauna = macroinvertebrates living within the sediment).

Figure 10.
Plots between observed (µg l^-1) and predicted LOEC by PLS models for macroinvertebrates allocated to different modes of feeding and their habitat. X=Y (stippled line) shown.

As depicted in Fig. 10 the LOEC predicted by the PLS models were in excellent agreement with the observed LOEC, irrespective of the chosen grouping of macroinvertebrates.

5.5 PLS models for zooplankton

An overview of raw data from the database used in the PLS models developed for zooplankton is shown in Table 8. The analyses are based on data from 31 different experiments with a total of 14 different pesticides.

The PLS models examined for zooplankton are summarised in Table 9.

The PLS model with the highest predictability (0.736) for zooplankton was obtained when the pesticides were added as single addition and the analysis was restricted to insecticides only (see Table 9). However, an almost as high predictability was obtained when the analysis included both insecticides and herbicides (0.655). For the remaining PLS models examined lower and more inconsistent predictabilities (Q²(cum)) were obtained. The interpretation of the PLS models was therefore restricted to the PLS models for mesocosm experiments with a single addition of insecticides (i.e. a total of 11 experiments).

Table 8.
Overview of raw data from the database for the PLS models developed for zooplankton. See Table 2 for abbreviations used. Sediment 1 refers to sediment present in mesocosm, 0 to no sediment; Macrophytes: 1 = present, 0 = no macrophytes, N= no information given on presence of marcophytes; Field/lab: 1 = Field study, 0 = Laboratory study. For all zooplankton groups the lowest effect concentrations observed for each taxonomic group in each mesocosm experiment were used as Y variables, expressing the toxic response of the organisms in the mesocosms (values shown in bold). L = Laboratory study at controlled temperature and light availability (hence latitude and longitude not relevant); F = flow-through study; - = no data. See Annex B for literature references.

Se her!

Table 9.
Predictability (Q²(cum)) of the examined PLS models for zooplankton. The procedure for removal of outliers is explained in Section 5.1.

5.5.1 Interpretation of the PLS model for zooplankton.

The prediction of the model selected in the previous section is shown Table 10.

Table 10.
Predicted variation of the significant axis of the PLS model selected for zooplankton. Q²: Variation in the Y matrix predicted from the variation in the X matrix by the current axis. Q²(cum): Cumulative variation in the Y matrix predicted from the variation in the X matrix.

As shown in Table 10 the first PLS axis predicts 73.6 % of the variation in the Y matrix from the variation in the X matrix. The second PLS axis predicts 0 % of the variation in the Y matrix from the variation in the X matrix. Hence, the second axis does not contribute to the overall predictability of the model and an interpretation of the second axis was therefore not carried out.

Figure 11.
Weights (loadings) of variables contributing to the first PLS axis for zooplankton. Day number through OECD represent variables in the X-matrix while the responses (LOEC) of the different zooplankton groups are shown at right.

From the loadings of the different variables (see Fig. 11) to the PLS axis it appears:

The axis primarily represents a "traditional" toxicity axis with positive correlations between hazard concentrations (HC_5,50 and LC50/10 = OECD) and LOEC obtained in the mesocosms. In specific:

	Loadings were most positive for the cladocerans, least positive for the rotifers and intermediate for the copepods. Hence, cladocerans semingly are the most sensitive zooplankters to insecticides followed by copepods and rotifers.
	Positive loadings were obtained for the variables Day number and latitude, whereas a negative loading was obtained for longitude. Thus insecticides seem to be less toxic to the zooplankton (i.e. high LOEC) if experiments are conducted in cold climates (high latitude) and/or during in the summer (high Day #). The effect of latitude is in contradiction to the effect of climate on macroinvertebrates, but could be due to a higher activity of zooplankters and thus exposure to pesticide at higher temperatures. On the other hand, the negative correlation between Day# and LOEC, does not support such relationship. The negative loading for Longitude suggests that zooplankton in experiments conducted in USA are more sensitive than the zooplankton in experiments conducted in Europe. This is in contradiction to the response of macroinvertebrates. As for macroinvertebrates the deviation between European studies and studies carried out in USA could be related to the stocking of fish in enclosures in USA.
	For the variables expressing hazard concentrations (HC5,50 and EC50/10 = OECD) positive loadings were obtained, whereas a negative loading was obtained for log KOW. Hence, as expected hydrophobic substances characterised by high single species toxicity seems to be most toxic to the zooplankters in the mesocosmos.

A summary of the effect of experimental mesocosm and pesticide characteristics on response of zooplankton is shown in Table 11.

The arrows in Table 11 indicate if numeric increases in system variables (see Table 2) will decrease ( Ý = high LOEC) or increase ( ß = low LOEC) the toxic response in the different groups of zooplankton.

Table 11.
Summary of influences of mesocosm characteristics, pesticide characteristics and toxicology (extrapolated effect concentrations) on toxic response on zooplankton. Ý = decrease in toxicity (i.e. higher LOEC); ß = increase in toxicity (i.e. lower LOEC);
- = no effect. See Table 2 for an explanation of system variables.

5.5.2 Predicting the effect concentrations for the zooplankton with the aid of the PLS model

The observed and predicted effect concentrations with associated 95 % confidence interval for the mesocosmos experiment calculated with the PLS model for zooplankton are shown in Table 12 and Figure 12. When the lower limit of the confidence intervals was below 0 the lower limit of the confidence interval was set to 0 (see section 5.4.2).

Figure 12.
Plots between observed LOEC (µg l^-1) and LOEC predicted by PLS model for zooplankton (Cladocera, Copepoda & Rotifera). X=Y (stippled line) shown.

Table 12.
Comparison of observed and predicted effect concentrations with associated 95 % confidence interval for the mesocosmos experiment calculated with the PLS model for zooplankton. -- = no observation. See Annex B for references.

Se her!

Generally, within the observed interval the effect concentrations predicted by the PLS model were in excellent agreement with the observed effect concentrations for both cladocerans and copepods, while the PLS model tended to over-estimate effect concentrations for rotifers. Such deviation could be expected, however, as effects of insecticides on rotifers primarily was of indirect nature (i.e. increases in abundance due to reduced competition from crustecean zooplankters).

5.6 PLS models for microalgae

An overview of raw data from the database used in the PLS models developed for microalgae is shown in Table 13. In total only 9 mesocosm experiments with toxicity data for microalgae were available from the data base. Thus it was only possible to consider the scenarios including either all experiments or all mesocosm experiments carried out in the field. Of these two scenarios the highest predictability (0.721) was obtained for the scenario of field mesocosm experiments (Table 14).

Table 14.
Predictability of the examined PLS models for microalgae

5.6.1 Interpretation of the PLS model for microalgae

For the selected PLS model with micro algae only one significant PLS axis was present (Table 15).

Table 15.
Predicted variation of the significant axis of the PLS model selected for micro algae. Q²: Variation in the Y matrix predicted from the variation in the X matrix by the current axis. Q²(cum): Cumulative variation in the Y matrix predicted from the variation in the X matrix.

From the plots of loadings and of variable importance the following interpretation of the PLS axis of the PLS model for micro algae is conducted:

The highest (and positive) loading was found for the X variable Interval (between pesticide dosings). Thus, pesticides added over a short period are most toxic to the algae in the mesocosmos. This is probably related to the short generation time of microalgae: frequent dosings will prevent microalgae to recover, while one or less frequent dosings will allow the microalgae to recover, when the pesticide dissipates.

Positive loadings were seen for the X variables expressing the extrapolated hazard concentrations (HC5,50 and EC50/10 = OECD procedure), whereas negative loadings were seen for the variables log KD and log KOW. Hence, hydrophobic and adsorpable substances with high single species toxicity were most toxic to the micro algae in the mesoscosmos.

Figure 13.
Weights (loadings) of variables contributing to the PLS axis for microalgae.

5.6.2 Predicting the effect concentrations for the micro algae with the aid of the PLS model

The observed and predicted effect concentrations with associated 95 % confidence interval for the mesocosmos experiment handled with the PLS model for micro algae appear is shown in Table 16 and Fig.14. When the lower limit of the confidence intervals was below 0 the lower limit of the confidence interval was set to 0 (see section 5.4.2).

Table 16.
Comparison of observed and predicted LOEC’s (µg l^-1) with associated 95 % confidence interval for the mesocosmos experiment calculated with the PLS model for microalgae.

Table 13.
Overview of raw data from the database for the PLS models developed for microalgae (phytoplankton & periphytes). See Table 2 for abbreviations used. For Sediment 1 refer to sediment present in mesocosm, 0 to no sediment; Macrophytes: 1 = present, 0 = no macrophytes; Field/lab: 1 = Field study, 0 = Laboratory study. For all microalgal groups tested the lowest effect concentrations observed in each mesocosms experiment were used as Y variables, expressing the toxic response of the organisms in the mesocosms (values shown in bold). L = Laboratory study at controlled temperature and light availability (hence latitude and longitude not relevant); - = no data. See Annex B for literature references.

Se her!

Figure 14.
Plots between observed LOEC (µg l^-1) and LOEC predicted by PLS model for microalgae (phytoplankton + periphytes). X=Y (stippled line) and linear regression equation shown.

Within the observed interval the effect concentrations predicted by the PLS model were in excellent agreement with the observed effect concentrations for microalgae.

5.7 Summary of PLS models

The amounts of data available for the different communities were quite variable and a direct comparison of PLS models should therefore be conducted with caution.

Macroinvertebrates

To obtain a PLS model with a reasonable predictability of the toxic effects to various macroinvertebrate groups, mesocosms should contain sediment and preferably macrophytes in the test system. Overall, the model developed was able to predict 63 % of the observed effects among macroinvertebrates.

In summary, the PLS analysis showed that

5. All macroinvertebrate groups in the mesocosms seem to be most sensitive when the experiments are conducted at high latitudes. Therefore, toxic effects at lower concentrations are expected with increasing distance from Equator, which may be due to a slower turn-over of populations at high latitudes, i.e. fewer generations each year at lower temperatures. Therefore, recovery of populations affected by pesticide exposure takes longer time at northern latitudes.
6. Macroinvertebrates living within the sediment (i.e. infauna) were less sensitive the pesticides than macroinvertebrates living on the sediment surface.
7. At a given total dose the effect of pesticides decreases with number of pesticide additions. Therefore, a low but persistent pesticide concentration will have a lower effect on the macroinvertebrates than a high but temporary pesticide concentration.
8. The toxic effects of pesticides are most pronounced in shallow mesocosms. At decreasing mesocosm depth a larger fraction of the pesticides will end up in the sediment compartment and thus increase the exposure to the sediment living macroinvertebrates. This interpretation is further reinforced by the inverse relation between Log K_D of pesticides and toxicity to invertebrates.

Zooplankton

The PLS model with the highest predictability for zooplankton was obtained when the pesticides were applied as single addition and the analysis was restricted to insecticides only.

The PLS analysis showed that

Microalgae

The highest predictability of pesticide effects to microalgae was obtained when only field mesocosm experiments were included in the analysis.

The PLS analysis showed that

6 Effect of pesticides in mesocosms

The previous PLS analysis was carried out at a rather high level of taxonomy and organism functionality to satisfy the requirement of data abundance within each group. In effect, detailed information on specific effects of pesticides and differences in sensitivity among different taxonomic groups have not been dealt with. In the following the specific effects (mortality, changes in abundance and sublethal effects) of individual herbicides and insecticides to different taxonomic groups within the major groups (microalgae, zooplankton and macroinvertebrates) are evaluated. In contradiction to the PLS analysis the evaluation has encompassed all mesocosm studies contained in the data base (see Annex B).

6.1 Phytoplankton and microalgae

The insecticide investigations contained in the data base have not demonstrated any directly significant effects such as reduced phytoplankton abundance at the prevalent insecticide concentrations. Therefore, the currently available data do not allow us to determine the maximum permissible insecticide-associated reduction that a phytoplankton population may suffer without becoming extinct. On the contrary it can be concluded that compared with zooplankton and benthic invertebrates, higher concentrations must prevail before a reduction in phytoplankton abundance occurs. This implies that for insecticides the various zooplankton and also invertebrates are affected before the phytoplankton community is directly affected.

A total of 193 records on effects of herbicides on algae (including phytoplankton, epibenthic microalgae and filamentous algae) were distributed between the following end-points LOEC: 83; NOEC: 101 and EC50: 9. We have not attempted to discriminate between phytoplankters and epibenthic algae as their environment (pelagic or benthic) especially in the shallow mesocosms will change rapidly according to mixing conditions. For filamentous algae the number of records was low which excludes a specific analysis. In line with zooplankton and macroinvertebrates structural parameters dominate the effect measures (abundance and biovolume by cell counts, biomass as fresh or dry weight for filamentous algae, chlorophyll a for microalgae). Primary production estimated by oxygen production or 14-C fixation was measured in two experiments (9 records).

In the data base direct effects of herbicides on algae were examined and quantified by relating the dosing of herbicide to changes in abundance relative to corresponding controls (without herbicide dosing). Mesocosm studies with herbicides ranged in duration between 14 and 373 days. Except for one study LOEC’s were recorded during or shortly after termination of herbicide exposure. Hence, most of the data presented below are from this initial period. In the following the relative sensitivity of algae to 3 different herbicides is visualised in diagrams showing LOECs and numeric changes in abundance (Figs. 15-17).

Figure 15.
Summary of effects of Alachlor on abundance of different microalgal species. Tests were carried out in recirculating flumes (175 l) in laboratory dosed at 5 concentrations (1-150 µg l^-1). Samples were taken 5 times during 3 weeks. Position of bars along the concentration axis refer to LOEC for the different groups/species. Numbers shown along bars denote decrease (-) or increase in abundance (in %) of corresponding controls. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Alachlor was 0.73 µg l^-1 (see Table 1).

Figure 16.
Summary of effects of Atrazine on different microalgae. A: recirculating microcosms in laboratory. B: primary production in phytoplankton in large mesocosms (470 m³) dosed once (regular samplings: 39-374 days). C: 1 m³ mesocosms dosed 3 times during 52 days. D: 120 m³ mesocosms dosed twice during 36 days. Biomass was reduced even 280 days after the last application. E: 120 m³ mesocosms dosed twice during 223 days. Periphytes recovered within 7 weeks; phytoplankton still reduced after 7 weeks. F: recirculating microcosms in laboratory. Numbers shown along bars denote decrease in abundance, concentration or primary production (in %) compared to corresponding controls. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Atrazine was 19.9 µg l^-1 (see Table 1).

Figure 17.
Summary of effects of Linuron on microalgae (abundance) and macrophytes (Elodea nuttalii, growth, EC50). Tests were carried out in laboratory mesocosms (600 l) dosed almost continuously through 28 days at 5 concentrations (0.5 – 150 µg l^-1). Numbers shown along bars denote decrease (-) or increase (%) in abundance or growth of corresponding controls. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Linuron was 19.7 µg l^-1 (see Table 1).

On the basis of these comparisons it is evident that:

1. The effect of herbicides on the different algal groups (and macrophytes) is very variable, with both reductions and increases occurring within one systematic group (Fig. 14). For Alachlor increases were observed only at the lowest test concentration (1 µg l^-1).
2. Atrazine as the mostly studied herbicide consistently led to reductions in algal biomass (Fig. 16). Generally, the effects were rather persistent in accordance with the slow dissipation of Atrazine, e.g. in one study primary production was impaired more than one year after application (Fig. 16 B).

In a study with a mixture of Atrazine, Diuron and Alachlor, some of the most sensitive species were Cyanophytae filaments and Monoraphidium sp. demonstrating inhibited growth. Cryptomonas sp., Chlorophyceae coccales, Diatoma sp. (single cell) and Scenedesmus sp. were less adversely affected, while the growth of Chlamydomonas sp. and Stephanodiscus sp. was stimulated. Several species were virtually unaffected by herbicides, e.g. pennate diatoms, Cyanophytae coccales and Anabaena sp.

Whether the phytoplankton can recover after a herbicide-related reduction is difficult to conclude from the mesocosm studies contained in the data base. In one study with Atrazine dosed at 100 µg l^-1, primary production was not fully recovered even one year after the application, while in a comparable study (80 µg l^-1) periphyton biomass and species composition recovered within 49 days. For Alachlor, almost full recovery was attained within 3 weeks for most algal groups except at the highest concentration tested (1000 µg l^-1). However, based on the dynamics of the phytoplankton communities observed in lakes, phytoplankton seem capable of recovering even after a pronounced reduction. It is a well-known phenomenon that some phytoplankton species may disappear from lakes for several years, only to reappear when growing conditions improve.

6.2 Zooplankton

Eighteen studies in the data base have examined the effects of herbicides. None of these studies have demonstrated direct effects on zooplankton.

The data base contains a total of 43 individual mesocosm experiments with zooplankton effect concentrations distributed among 18 different insecticides. A total of 1564 records on effects of insecticides on zooplankton were distributed between the following end-points LOEC: 706; NOEC: 817 and EC50: 14. The majority of records concern Copepods (674 records), Cladocerans (543 records) and Rotifers (279 records), while unspecified zooplankton records amounts to 32. Abundance of individuals is by far the most used effect parameter (1549 records), while species diversity and biomass are scarce at 13 and 2 records, respectively.

In the data base direct effects of insecticides on zooplankton were examined and quantified by relating the dosing of insecticides to changes in abundance relative to corresponding controls (without insecticide dosing). To include results from both single and multiple pesticide application experiments the average decrease in abundance within the period 3-14 days after the first application of insecticide was used. By this approach studies with single and multiple applications could be compared and bias due to different recovery was eliminated. In cases where sufficient data were available EC50 was calculated after probit transformation.

Figure 18.
Temporal variation in abundance (% of control) of zooplankton following a single dose of esfenvalerate in 5 different concentrations ( µg l^-1)to mesocosms (after Lozano et al. 1992 ).

In Fig. 18 is shown an example of the temporal variation in abundance of Cladocera, Copepoda and Rotifera after a single dose of esfenvalerate. Both the initial impact and the subsequent recovery were dependent on the dose. By combining 2-3 different mesocosm studies distinct dose-response curves can be established (see Fig. 19). Noticeable features are dose dependent decreases in Cladocera and Copepoda and increases in phytoplankton and Rotifera.

Figure 19.
Dose-response relations of plankters in mesocosms exposed to esfenvalerate. Mesocosms were shallow (0.5-1.1 m depth), had sediment and macrophytes and ranged between 25 – 1100 m³ in volume (from Fairchild et al. 1992;Lozano et al. 1992, Webber et al. 1992).

Overall, during the first 1-2 weeks after insectide application Cladocerans were more sensitive to insecticides than Copepoda, even though large variation were evident in the data (see Figure 20). Within the copepod population nauplii were more sensitive (» 10%) than adult copepods (see Figure 21). Besides, the variation in the plot was limited reflecting that the two groups probably represent the same species within each experiment.

Figure 20.
Scatter plot of decreases in abundance of Cladocera and Copepoda 3-14 days after insecticide application (Diflubenzuron; Methoxychlor, Hexazinon, Chlorpyrifos, Esfenvalerat, Deltamethrin, Permethrin, Bifenthrin). Regression line and X=Y shown (stippled). Decrease in Cladocera was significantly larger (i.e. Cladocera being more sensitive) than corresponding decrease in Copepoda (Kolgomorov-Smirnov test).

Figure 21.
Scatter plot of decreases in populations of Nauplii and adult Copepoda after insecticide application (Fenvalerat, Methoxychlor, Chlorpyrifos, Esfenvalerat, Deltamethrin, Permethrin, Bifenthrin, Trahalomethrin). Regression line and X=Y shown. Only reductions lower than 100% were included. Decrease in Nauplii was significantly larger (i.e. Nauplii being more sensitive) than decrease in adult Copepoda (Kolgomorov-Smirnov test).

The larval stage of the dipteran Chaoborus is an important pelagic predator in lakes and ponds. In 6 mesocosm studies the abundance of Chaborus was sufficient to calculate the impact of insecticides and compare its sensitivity to Cladocera and Copepoda (Figure 22). In these studies representing different classes of insecticides Chaoborus consistently was more sensitive than Copepoda, but had a sensitivity similar to Cladocera’s.

Figure 22.
Scatter plot of decreases in abundance of Chaoborus and crustacean zooplankton after insecticide application (Chlorpyrifos, Permethrin, Lindan, Methoxychlor). X=Y shown. Only reductions lower than 100% were included. Decrease in Chaoborus was significantly larger (i.e. Chaoborus was more sensitive) than decrease in adult Copepoda (g ) but identical to impact on Cladocera (u ) (Kolgomorov-Smirnov test).

Hitherto, effects have been evaluated at the Order level (e.g. Cladocera) as dictated by the level of taxonomy reported in the majority of mesocosm studies. This invariably will lead to variation in the aggregated data in case of interspecies differences in sensitivity (see Figure 20). Two studies allow extracting quantitative information on variability in sensitivity within Cladocera.

In mesocosms exposed to Azinphos-methyl LOEC ranged between 4 and 20 µg l^-1 while for esfenvalerate the observed range in LOEC was markedly wider at 0.01-5 µg l^-1 (Figure 23). In both studies Sida was the most sensitive genus and Pleuroxus the least sensitive. The difference may be related to size and habitat of species. For comparison the calculated Hazard Concentration of esfenvalerate to Cladocera is 0.18 µg l^-1 and 0.02 µg l^-1 for HC_5,50 and OECD₁₀, respectively (see Table 1). The range in LOEC for different species within Copepoda varied between 0.08 – 5 µg esfenvalerate l^-1.

Figure 23.
Lowest observed effect concentration for different Cladoceran species in mesocosms exposed to Azinphos-methyl (A) and Esfenvalerate (B); (*) all observations were identical, range of LOEC recorded during exposure.

In summary, mesocosm studies have demonstrated that zooplankters are very sensitive to insecticide exposure. At the group level:

1. Cladocerans and Chaoborus are the most sensitive followed by copepod nauplii and adult Copepoda.
2. At a given concentration the Cladoceran population on average will show larger reductions (20 %) than the copepod population (based on regression analysis).
3. Copepod nauplii on average will show 10 % larger reductions than the adult population and observed reductions in one group are very good predictors of the reductions of the other group.

The variation in sensitivity within each zooplankton group as demonstrated in mesocosm studies is considerable. For esfenvalerate LOEC varied 2.5 orders of magnitude for the different species among cladocerans. This variation is probably related to the size of the different species, their habitat and/or feeding mode.

6.2.1 Statistical power of impact of insecticides on zooplankton in mesocosm experiments

Overall, the statistical power in the mesocosm studies was rather low. The average reduction in abundance of zooplankters (i.e. excluding indirect effects) exposed to insecticides at recorded LOEC’s was 75.4 % (± 21.3 %; SD). The low power is due to low number of replicates, low number of and/or large range in test concentrations. The use of few test concentrations spanning 2-3 orders of magnitude invariably will lead to crude estimates of LOEC.

In Table 17 is shown the distribution of reductions in abundance of zooplankton at the various combinations of replicate number and number of test concentrations applied in the different studies. The different combinations are based on observations ranging from 6 to 124 in number and from 1 to 4 different studies carried out at different locations and using different mesocosms (volume, ± macrophytes etc.). Hence, conclusions drawn should not be too firm. Still, the data suggest that in order to obtain a sufficient resolution and sensitivity the experimental design should be a hybrid approach encompassing more than 4 test concentrations and at least two replicates at each concentration. As the size of experimental design usually is constrained by economy with a maximum number of units of 15-16 (see Table 17) based on the results shown in Table 17 they should be distributed between 5 (8) test concentrations each with 2 (3) replicates. Still, to achieve a sufficient sensitivity the range in concentrations applied should not be unduly large, i.e. less than 2.5-3 orders of magnitude.

Table 17.
Average reduction (%)± SD in zooplankton abundance at LOEC in mesocosm studies of different experimental design. Number of observations in brackets. – # observations below 5.

* 2 replicates in control and one replicate per test concentration.

6.2.2 Recovery of zooplankton populations

Recovery of zooplankton populations following insecticide exposure relies on reproduction from surviving individuals, hatching of resting stages (eggs) or immigrations from outside of the system. For the dipteran Chaoborus recovery may also take place by egg laying from imagos. Whether the zooplankton community may endure a 100% reduction depends on whether the resting stages of the various zooplankton groups are tolerant to pesticides, which remains to be elucidated.

To be able to examine recovery of zooplankters, mesocosms need to include sediment and in addition to be in operation for several weeks-months after pesticide dosing has stopped. However, in more than 50 % of the mesocosm studies where zooplankton were followed the post exposure period was too short and/or the doses of insecticides too high to observe complete recovery of zooplankton.

Based on the recovery studies, an attempt can be made at defining the lowest level to which zooplankton populations may be reduced as a consequence of pesticides without being at risk of extinction. For Cladocera the time elapsed for full recovery after the insecticide dosage varied between 10 and 120 days. In Figure 23 is shown a plot of the initial (and maximum) reduction in population size (relative to corresponding control) and the time elapsed after dosage had stopped until full recovery of the population. In mesocosm experiments where cladocerans had been reduced severely (i.e. > 95 %) it took more than 12-15 weeks for full recovery. Such lengthy recovery is probably the result of slow dissipation of the insecticide in the mesocosm and thus continued toxic effects after dosage was stopped.

Figure 24.
Scatterplot between initial reduction in abundance of Cladocera and time elapsed for full recovery of the population. The relation can be described by: R = 8.5 e^0.019x, r²=0.6, where 8.5 (y-axis intercept) indicate the generation time for non-affected populations. Only reductions below 100 % were included.

The relation between the initial reduction in population size and time until full recovery was met could be described by an exponential function (see Figure 24). At reductions below 80 % of the initial population size recovery was fast, i.e. less than 20 days. However, recovery time increased markedly if the initial population was reduced by more then 85 %. Still, even at population reductions close to 100 % full recovery of Cladocerans was observed in the mesocosms where the length of observation period was sufficient long.

For copepods an almost identical relation between initial decrease and recovery was obtained (see Figure 25). Fast recovery within Cladocera (usually analysed at Order level) as observed in numerous studies is likely to be governed by parthenogenetic reproduction. However, to maintain populations of cladoceran species sexual reproduction is essential at intervals. Therefore, recovery studies terminated successfully within 3-4 months may not be sufficient to describe the recovery on the long term.

Figure 25.
Scatterplot between initial reduction in abundance of Copepoda and time elapsed before full recovery of the population. Curve fitted by eye.

Without being at risk of extinction, a significant reduction of cladoceran and copepod numbers may, however, result in reduced species diversity and thus a decline of environmental quality. It may also be that a less diverse community/ecosystem is more sensitive to sudden outside influences such as increased nutrient input or additional pesticide inputs during the recovery phase. Unfortunately, the data contained in the data base do not allow examination of such relations.

6.3 Indirect effects of insecticides on plankton.

The most prominent indirect effect of insecticides on the plankton community includes increases in phytoplankton and rotifers. Following a decrease in population size of crustacean zooplankton, phytoplankton biomass generally will increase due to relaxation of grazing control. In addition, planktonic rotifers that are less sensitive to insecticides will increase in abundance due to increased food availability and reduced competition from crustacean zooplankton (see Figs. 18&19).

In Figure 26 is shown that the phytoplankton biomass increases, when the crustacean zooplankton becomes affected by insecticides. As expected being an indirect effect the scatter is substantial, however, the relation is highly significant. It seems that low impacts on the crustacean zooplankton will not result in increased growth of phytoplankton. If however, zooplankton becomes reduced by more than 50 % dramatic increases in phytoplankton (>100 %) must be expected (Fig. 26).

Figure 26.
Decrease in crustacean zooplankton (Copepoda & Cladocera) and corresponding change (increase) in phytoplankton biomass (Chla) in mesocosm experiments with insecticides (diflubenzuron, endosulfan, deltamethrin, esfenvalerate).

Planktonic rotifers constitute direct competitors to cladocerans and copepods. Reductions caused by insecticides in these groups generally will lead to increases within Rotifera. Due to high reproductive potential increases in abundance up to 3000 % have been observed. In Figure 27 is shown a scatter-plot of changes in crustacean zooplankton and corresponding observations in rotifer abundance in mesocosm experiments with insecticides. Note that the increase in rotifer abundance has been scaled to 100 % within each experiment. On average the decrease in crustacean zooplankton only explains about 20 % of the observed variation in rotifer abundance. Still, the inverse relation is highly significant. Despite increases in rotifer abundance the pelagic grazing control in insecticide affected systems become impaired and phytoplankton biomass will increase (Fig. 26).

Figure 27.
Decrease in crustacean zooplankton (Copepoda & Cladocera) and corresponding change (increase) in Rotifer abundance in mesocosm experiments with insecticides (methoxychlor, esfenvalerate, fenvalerate, cyfluthrin). Within each experiment the increase in Rotifera has been normalised to 100 %.

Table 18 shows an overview of recorded effects on plankton communities with 12 different insecticides in 19 different mesocosm studies. While direct effects on Cladocera and Copepoda are very consistent, indirect effects on Rotifera are more variable. In addition, it is striking that changes in phytoplankton biomass were observed in only 5 out of 19 mesocosm studies. Presence of non-eatable phytoplankters may be responsible

Table 18
Overview of direct og indirect effects of insecticides in mesocosm experiments. ß = significant and consistent decrease in abundance (direct effect), Þ = no effects, Ý = significant and consistent increase in abundance (indirect effect), Ý ß = both decrease and increase observed , - no records.

In conclusion, indirect effects of insecticides on the plankton community have been recorded in more than 50 % of mesocosm studies. In those studies the indirect effects were at least as sensitive as direct effects, e.g. a 75 % reduction in crustacean zooplankton on average will be followed by a 500 % increase in rotifer abundance and a 200 % increase in phytoplankton biomass (see Figure 26). However, indirect effects are very variable in both magnitude and direction and thus less robust compared to direct effects.

6.4 Effect of Insecticides on Macroinvertebrates

Macroinvertebrates generally are insensitive to herbicides. Hence, in only 3 out of 7 mesocosm experiments involving macroinvertebrates in the data base were effects on the macroinvertebrate community detected. They included reduced emergence of Chironomids due to food limitation (reduction of epibenthic algae due to Atrazine), increased drift in streams (Triclopyr-ester & Hexazinone). However, effect concentrations were above calculated hazard concentrations for these herbicides (see Table 1).

The data base contains a total of 41 individual mesocosm experiments with macroinvertebrates distributed among 19 different insecticides. A total of 935 records on effects of insecticides on macroinvertebrates were distributed between the following end-points LOEC: 424, NOEC: 491, EC50: 17 and NEC: 3. Dipterans were used in 29% of the experiments followed by mayflies, Ephemeroptera, which was used in 21% of the experiments. All other macroinvertebrate orders were followed in less than 10% of the experiments. In total, insects constituted 73% of all macroinvertebrates sampled and non-insects 27%.

The majority of recorded effect concentrations concern Dipteran (316 records) with the majority belonging to the family Chironomidae, Ephemeroptera (131 records), Amphipoda (93 records), Isopoda (75 records), Tricoptera (67 records), Hemioptera (58 records), Gastropoda (53 records), Coleptera (50 records), Oligochaeta (32 records), Odonata (29 records), Plecoptera (10 records) and Lepidoptera (4 records).

Abundance of individuals was by far the most used effect parameter (872 records), followed by drift (26 records), mortality (17 records), emergence (13 records) and survival (4 records).

In the data base direct effects of insecticides on macroinvertebrates were examined and quantified by relating the dosing of insecticides to changes in abundance relative to corresponding controls (without insecticide dosing). To be able to compare studies with different application schemes the average decrease in abundance within the period 28-56 days after the first application of insecticide was used. By this approach studies with single and multiple applications could be compared.

In Fig. 28 is shown an example of the temporal variation in abundance of Amphipoda, Chironomidae and Oligochaeta after a single dose of esfenvalerate. Both the initial impact and the subsequent recovery (Chironomidae and Oligochaeta only) were dependent on the dose.

Figure 28.
Temporal variation in abundance (% of control) of macroinvertebrates following a single dose of esfenvalerate in 5 different concentrations to mesocosms (after Lozano et al. 1992 ).

The sensitivity of different macroinvertebrate groups and effect parameters to insecticide exposure in mesocosms were evaluated by comparing corresponding LOECs and numeric reductions. In Figure 29 is shown an example of reductions in abundance of various macroinvertebrate groups exposed to esfenvalerate along with effect on the emergence of insects. These studies demonstrate the general pattern among macroinvertebrates: amphipods and mayflies being rather sensitive to insecticides, while gastropods, Odonata and oligochaetes are rather insensitive.

Figure 29.
Dose-response relations of macroinvertebrates in mesocosms exposed to esfenvalerate. Mesocosms were shallow (0.5-1.1 m depth), had sediment and macrophytes and ranged between 25 – 1100 m³ in volume (from Fairchild et al. 1992; Lozano et al. 1992, Webber et al. 1992).

In the following the relative sensitivity of macroinvertebrate groups to individual insecticides is visualised in diagrams showing LOECs and numeric reductions of macroinvertebrate abundance or alternative endpoints such as increase in drift in artificial streams. Only experiments with more than one group or two or more endpoints followed within one group are presented and discussed. Because of differences in mesocosm volume, season and latitude that all influence the measured toxicity of insecticides (see chapter 5) comparisons can only be evaluated within single mesocosm experiments.

On the basis of these comparisons it is evident that:

1. The sublethal effect drift in stream macroinvertebrates generally appears to be a more sensitive endpoint than changes in abundance (Figure 30AB). In stream ecosystems drift is a natural behaviour of crustaceans and insect larvae for dispersal and colonisation of substrate. When exposed to insecticides (and several other toxic substances) arthropod macroinvertebrates may leave the substrate and drift to avoid the toxicant. Hence, in the short term, drift and population size are reciprocal measures: increased drift invariably will lead to reduced abundance. The seemingly higher sensitivity of drift compared to abundance presumably is related to differences in sample size and stronger statistics in drift data.
2. The endpoint emergence of adult insects seems to be as sensitive as changes in abundance of larvae (e.g. Figure 30). Insecticides may increase the mortality of larvae and reduce growth rate. In effect, emergence will decrease or be delayed. Rate of emergence usually is assessed using float traps that integrate samples from a fairly large bottom area and therefore show less spatial variability than benthos samples. On the other hand, the timing of emergence in affected populations of insect larvae often will differ from the emergence in non-affected populations (i.e. controls) which may complicate sampling and interpretation.
3. The insect order Tricoptera consistently was the most sensitive macroinvertebrate group to insecticides (Figs. 32-34), followed by Plecoptera/Hemiptera/Ephemeroptera/Cole-optera/Amfipoda/Isopoda (no particular order). Chironomidae as a very diverse group (individual size, mode of feeding etc.) showed a rather large variation in sensitivity (e.g. Fig. 31).
4. Odonata, Gastropoda and Oligochaeta consistently were the groups with the lowest sensitivity to insecticides.

Figure 30.
Summary of effects of Lindane on drift, insect emergence and abundance of different macroinvertebrate groups. Experiment A&B (artificial streams) received Lindane continuously for 4 weeks, while in experiment C (1000 l stagnant mesocosm) Lindane was dosed only once. Numbers shown along bars denote the increase in drift (in percentage) or decrease in emergence or abundance of corresponding controls. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Lindane was 2.9 µg l^-1 (see Table 1).

Figure 31.
Summary of effects of Chlorpyrifos on macroinvertebrate groups in laboratory mesocosms (A & B), experimental ditches (C) and in artificial streams (D). In all 4 experiments reduction in abundance was used as end-point. Numbers shown along bars denote the reduction in percentage of corresponding controls. Experiment A-C received Chlorpyrifos as a single dose, while in experiment D Chlorpyrifos was dosed continuously for 21 days. The different colours denote different species within one group. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Chlorpyrifos was 0.04 µg l^-1 (see Table 1).

Figure 32.
Summary of effects of Lambda-cyhalothrin on abundance of different macroinvertebrates in mesocosms. In experiment A and C (25 m³) Lambda-cyhalothrin was dosed 4 times during 42 days, while B (450 m³) was dosed every 14 days through 147 days. Numbers shown along bars denote the reduction in percentage of corresponding controls. The different colours denote different species within one group. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Lambda-cyhalotrin was 80 ng l^-1 (see Table 1).

Figure 33.
Summary of effects of Diazinone on abundance of different macroinvertebrates in mesocosms. Numbers shown along bars denote the reduction in percentage of corresponding controls. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Diazinon was 0.03 µg l^-1 (see Table 1).

Figure 34.
Summary of effects of Fenvalerate on abundance of different macroinvertebrates in mesocosms. In experiment A (small recirculating flume) Fenvalerate was dosed once times and abundance was monitored after 30 days, while B was followed through 84 days. Numbers shown along bars denote the reduction in percentage of corresponding controls. The different colours denote different species within one group. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Fenvalerate was 50 ng l^-1 (see Table 1).

Figure 35.
Summary of effects of Esfenvalerate on abundance of different macroinvertebrates in 1100 m³ mesocosms. Esfenvalerate was dosed every week through 10 weeks. Only LOECs, but no numeric reductions were given in the report. The different colours denote different species within one group. As a comparison the hazard concentration (HC_5,50%) calculated from distribution based extrapolation of single species toxicity data for Esfenvalerate was 180 ng l^-1 (see Table 1).

6.4.1 Statistical power of impact of insecticides on macroinvertebrates in mesocosm experiments

The statistical power for effects on macroinvertebrates was comparable to the impact on zooplankton with an average reduction in abundance (i.e. excluding indirect effects) at recorded LOEC’s of 76.5 % (± 20.3 %; SD). In Table 19 is shown the distribution of reductions in abundance of macroinvertebrates at the various combinations of replicate number and number of test concentrations applied in the different studies. The different combinations are based on observations ranging from 7 to 103 in number and from 1 to 4 different studies carried out at different locations and using different mesocosms (volume, ± macrophytes etc.), which may set limits to conclusions drawn.

Overall, the general pattern resembles the data for zooplankton suggesting that a sufficient sensitivity may be obtained by a hybrid approach with more than 4 test concentrations and at least two but preferably 3 replicates at each concentration.

Table 19.
Average reduction (%)± SD in macroinvertebrate abundance at LOEC in mesocosm studies of different experimental design. Number of observations in brackets. – # observations below 5.

6.4.2 Recovery of macroinvertebrates after insecticide exposure.

Recovery is essential when evaluating effects of pesticides. In macroinvertebrates recovery may take place by invasion from non-affected populations outside the affected area (e.g. by drift in streams, reproduction) and reproduction by surviving individuals within the affected area. In order to evaluate recovery, mesocosm studies need to be carried out in the field (to allow flying insects to lay eggs) and should at the minimum extend a full life cycle of the organisms studied after insecticide dosage. And obviously, repeated sampling of macroinvertebrates will be necessary to follow changes in populations. In the data base not all of the mesocosm experiments included a time series. Furthermore, the majority of experiments in the data base were terminated within 150 days although a few experiments ran for a whole year.

Taking the general life-cycle length for macroinvertebrates into consideration (ranging from less than a month to several years), the experimental time frames in most mesocosm studies appear to be too short. This might partly explain why there are only very few examples of recovery in mesocosms contained in the data base, none of them being a full recovery (Fig. 36). There was no sign of recovery in 81 % of the observations. Signs of recovery were found in 13% of the observations and a moderate recovery in 6% only.

Chironomids (belonging to the order Diptera) and Isopoda were the most important taxonomic groups in the "slight recovery group" (Fig. 37 left) whereas Chironomids and Ephemeropterans dominate the "moderate recovery group" (Fig. 37 right). Both Chironomids and Ephemeropterans in general are considered as good colonisers with short life cycles and this probably explains why they show the most rapid recovery.

Figure 36.
Recovery of macroinvertebrate populations in mesocosm studies contained in data base. An observation includes changes found over time in a macroinvertebrate taxon. No recovery is defined as a less than 5% change (increase) after the initial decrease; slight recovery less than 25% change and moderate between 25 and 75% change.

Figure 37.
Percentage composition of macro-invertebrate orders that showed a slight recovery (left) or moderate recovery (right).

Overall, it is surprising that there is so little evidence of recovery within macroinvertebrates. This finding might reflect that most studies have been too short. One reason for this might be that studies in general involve other taxonomic groups such as zooplankters with shorter life spans and that the duration of the experiments are set to reflect their life span and not the macroinvertebrates. More specific studies targeted towards macroinvertebrates might be needed or the duration of the experiments should be increased when mimicking whole ecosystems. Generally, one should expect that organisms with limited ability for colonising, i.e. non-insect groups such as Isopods, Amphipods and Gastropods or insects with long generation times such as Odonata would be the slowest to recover following insecticide exposure. However, both Gastropoda and Odonata are among the least sensitive to insecticides and the recovery will only be an issue after excessive insecticide exposure. For Amphipods, however, the limited ability to recover can be very critical as these organisms also are among the most sensitive to insecticdes.

7 Comparison of extrapolated hazard concentration and observed effects in mesocosms

As shown in this study only a limited number of "high quality" mesocosm experiments (see Chapter 4 for selection criteria) examining the effects of pesticides in freshwater systems have been reported. As a consequence, an alternative approach using the results from numerous standardised single species tests has been developed. Hazard concentrations for ecosystems may be calculated from distribution-based extrapolation of single species toxicity data (EC50, LC50) using (slightly) different statistical methods (e.g. Wagner & Løkke 1991, Miljøstyrelsen 1994, Emans 1994). The widely used calculation of hazard concentration, HC_5,50% aims to protect 95% of the organisms in an ecosystem with a 50% probability. Others consider that a 90% protection of ecosystem species is adequate to avoid adverse effects on the natural ecosystem, i.e. HC_10,50% (Hall et al., 1998). A alternative approach adopted by OECD multiplies the lowest effect concentration observed among all standardised tests by 0.1 (application factor of 10) (see Chapter 4 for more details).

The major limitation of these approaches is the availability of single species test results, as they are biased towards dominance of cladocerans, planktonic algae and fish. Thus, data on effect concentrations of insect larvae are scarce or not available for several pesticides (see Table 3 and Annex 1). For macrophytes, no standardised test results were available for the pesticides included in the data base! As standardised tests usually are short-term (48-96 hours) they may fail to reveal long-term effects caused by pesticides accumulated in organisms. And relying solely on standardised single species tests’ extrapolation methods will never be able to account for behavioural effects and interactions between populations and trophic groups (i.e. indirect effects).

In Table 20 we have summarised a comparison of extrapolated hazard concentrations and the lowest observed effect concentrations in the mesocosm experiments contained in the data base. An extended version of the comparison including 66 mesocosm experiments is shown in Annex 3. Within single mesocosm experiments LOECs for different organisms can vary 2-3 fold (see Chapter 6 & 7), hence LOEC for one group can be NOEC for several others groups. Still, we have selected the lowest observed effect concentrations within an Order, genus or species and tested effects for significance (i.e. persistence) (see chapter 5).

We have used the ratio HC_5,50/LOEC or OECD/LOEC as a measure of the success of the extrapolated hazard concentration (HC_5,50 or OECD) to "protect the species" in an aquatic ecosystem. Table 20 only includes experiments, where the ratio HC_5,50/LOEC is above 1. With exception of experiment 57flm (that included 25 different taxons) the experiments in Table 20 included less than 20 taxons (range 1-13). Hence, in 13 out of 66 experiments the widely used approach failed to protect 95 % organisms in the ecosystem (see Annex 3). Even using the more conservative OECD approach the hazard concentration failed to protect 95% of the organisms in 5 experiments. In about half of the experiments contained in Table 20 (i.e. 8 experiments) NOEC was not be established for the most sensitive parameter, because sufficient low concentrations were not tested. Hence the ratio HC_5,50/LOEC calculated for these experiments represents a minimum.

It is noticeable, that the vast majority of examples of "failures" of extrapolated hazard concentrations to protect sensitive species are found in experiments, where LOECs were recorded for macroinvertebrates and insects, while LOECs for phytoplankton and zooplankton except for two occasions (see Table 20) result in ratios of HC_5,50/LOEC well below 1 (see Annex 3). Therefore, extrapolated hazard concentrations generally will protect the plankton environment in ecosystems, which hardly is surprising as the extrapolated values primarily rely on standardised tests with cladocerans and phytoplankton. On the other hand, extrapolated hazard concentrations are much less successful in protecting the macroinvertebrate community.

The importance of including macroinvertebrates in mesocosm experiments is further demonstrated by an ANOVA, where the "failure" of extrapolated hazard concentrations (HC_5,50) to protect the aquatic ecosystem was explained by number of organism groups monitored, inclusion of macroinvertebrates and the number of insecticide doses (Table 21).

Intuitively, one would expect that the chance of "failure" would increase with increasing number of organism groups monitored and when the insecticide was dosed several times. In the analysis neither variable was important. However, if macroinvertebrates were monitored in mesocosms the risk that extrapolated hazard concentrations would fail to protect the whole ecosystem was substantial (significance level – p = 0.017) (see Table 21). Average ratio HC_5,50/LOEC in experiments with macroinvertebrates was 9.0 (1.10 if the high value of 80 in exp. 76tll was omitted) but much lower at 0.26 in experiments without macroinvertebrates.

Table 20.
Comparison of extrapolated hazard concentrations and the lowest observed effect concentrations in the mesocosm experiments. NOEC: Yes = lowest test concentration were lower than the lowest effect concentration observed in mesocosm; No = effect was observed at the lowest test concentration applied. HC_5,50/LOEC = ratio between extrapolated hazard concentration (see Table 1) and lowest observed effect concentration. OECD/LOEC = ratio between hazard concentration (OECD₁₀ approach) and lowest observed effect concentration. HC_5,50/low = ratio between hazard concentration and the lowest test concentration applied. OECD/low = ratio between extrapolated hazard concentration (OECD₁₀ approach) and the lowest test concentration applied. Observed effects at lowest concentration: ¯ decrease (mostly in abundance); ­ increase.

* Only significant effects included (see chapter 4).

** 2,4 D are toxic to higher plants only, while extrapolated hazard concentrations were based on single species test with algae and zooplankton, only.

The failure of extrapolated hazard concentrations in protecting 95% of organisms and especially macroinvertebrates, against insecticides in the aquatic environment probably occurs because

	macroinvertebrates are the most sensitive organisms to insecticide exposure, probably related to the high Kd of most insecticides. Hence, under natural conditions the exposure of sediment-dwellers will be higher than plankters.
	macroinvertebrates are underrepresented in the single-species tests used for extrapolation of hazard concentrations
	the duration of standardised single species tests is too short to reveal the potential effects on macroinvertebrates as the maximum effects on macroinvertebrates are recorded 2-8 weeks after exposure start in mesocosm experiments.

Table 21.
Result of 1-way ANOVA for effect of number of organism groups (phytoplankton, periphytes, macrophytes, zooplankton, macroinvertebrates, fish) monitored, inclusion of macroinvertebrates (yes, no) and number of insecticide doses (1-10) during the experiment on the ratio HC_5,50/LOEC (HC_5,50/LOEC >1 ® value = failure; HC_5,50/LOEC <1 ® value = success). The analysis was restricted to experiments with insecticids, where NOEC was recorded for the most sensitive organism (n=23). Mean squared effect, mean squared error, F statistics and level of significance shown.

In conclusion, long-term (abundance and emergence) and short-term effects (drift in streams) of insecticides on macroinvertebrates are among the most sensitive effect parameters recorded in mesocosms. Such effects cannot be explained in sufficient detail by extrapolations based on calculations of Hazard Concentrations from standardised single species test. Therefore, the data bases used for extrapolation ought to be extended with tests on macroinvertebrates and preferentially the duration of these test should be increased. Alternatively, mesocosm experiments should be carried out. To arrive at environmentally realistic effect concentrations and protect the whole ecosystem, mesocosms need to include a benthic compartment encompassing a diverse fauna including important and sensitive taxonomic groups such as Tricoptera, Ephmeroptera and Amphipoda.

8 Conclusions and recommendations

The regulation of pesticide use and protection of non-target species primarily relies on evaluations based on single species tests. If a pesticide is evaluated to constitute a hazard to aquatic life, further and extended analysis must be carried out to show that the pesticide does not constitute a risk to the aquatic environment (EU directive 91/414). In line with several other countries Denmark relies on extended risk evaluations based on tests carried out under near-natural conditions at an ecosystem level by using experimental mesocosms of various size and design. Several guidelines describe protocols of how to carry out mesocosm experiments and what endpoints should be measured. Still, a general (uniform) procedure of how to interpret the results from mesocosm experiments and apply these results in a regulatory procedure has not been accepted at an international level.

In this study we have carried out a critical analysis of published results of mesocosm experiments, extracting and quantifying the influence of the experimental set-up on the sensitivity of organisms and the statistical power of observed effects, when and where the experiments were carried out, which taxonomic and functional groups were the most sensitive, and to what extent available single species test results can be used to protect the environment using various extrapolation procedures.

For a number of taxonomic and functional groups we have developed regression models using a PLS technique relating effects of pesticides to system characteristics and physico-chemical characteristics of the pesticides. The predictability of the models was rather high at 0.63-0.73. As the PLS models are based on all appropriate data in the database it is possible to develop a evaluation procedure taking all the available information into account, rather than on a restricted use of a single or a few mesocosm experiments for each pesticide. Thus with the aid of the PLS models it is possible to evaluate all mesocosm experiments with pesticides on a common basis.

The following presents an extract of the results of the analysis and thus constitutes a checklist for managers evaluating mesocosms in connection with the approval procedure.

Checklist to be applied when evaluating results from mesocosm studies. The left column contains general information and definitions. The right column contains the important results from the analysis with references to the appropriate sections in the report (in brackets).

9 Reference List

Aanes K. and T. Baekken. 1994. Acute and long-term effects of propiconazole on freshwater invertebrate. Norwegian Journal of Agricultural Sciences Supplement, 13: 179-193

Aldenberg, T., Slob, W. 1993. Confidence limits for hazardous concen-trations based on logistically distributed NOEC data. Ecotoxicology andEnvironmental Safety, 25: 48-63

Barry M.L.D. 1998. The use of temporary pond microcosms for aquatic toxicity testing: direct and indirect effects of endosulfan on community structure. Aqua. Toxicol. 41:101-124.

Boyle T., J.F. Fairchild, E.F. Robinson-Wilson, P.S. Haverland and J.A. Lebo. 1996. Ecological restructuring in experimental aquatic mesocosms due to the application of diflubenzuron. Environmental Toxicology and Chemistry 15: 1806-1814

Brazner J. and E.R. Kline. 1990. Effects of chlorpyrifos on the diet and growth of larval fathead minnows, Pimephales promelas, in littoral enclosures. Canadian Journal of Fisheries and Aquatic Sciences 47:1157-1165.

Breneman D.H et al. 1994. Stream microcosm toxicity tests: predicting the effects of fenvalerate on riffle insect communities. Tarrytown, N.Y. : Elsevier Science Inc. British Crop Protection Council (BCPC), 49 Downing Street, Farnham GU9

Brink, P. J. Van den, Hartgers, E. M., Fettweis, U., Crum, S. J. H., VanDonk, E., and Brock, T. C. M. 1997. Sensitivity of macrophyte-dominated freshwater microcosms to chronic levels of the herbicide linuron . 1. Primary producers. Ecotoxicology and Environmental Safety 38: 13-24

Brink, PJ van den; Donk, Ellen v., Gylstra, R., Crum, SJH & TCM Brock. 1995. Effects of chronic low concentrations of the pesticides chlorpyrifos and atrazine in indoor freshwater microcosms. Chemosphere 31: 3181 - 3200

Brock T. 1998. Assessing chemical stress in aquatic ecosystem: remarks on the need of an integrative approach. Aqua.ecology 32:107-111.

Brock T et al. 1992. Fate and effects of the insecticide dursban 4E in indoor Elodea-dominated macrophyte-free freshwater model ecosystems: 1. Fate and primary effects of the active ingredient chlorpyrifos. Arch. Environ. Contam. Toxicol. 23: 69-84.

Brock T et al. 1993. Fate and Effects of the Insecticide Dursban(R) 4E in Indoor Elodea-Dominated and Macrophyte-Free Freshwater Model Ecosystems .3. Aspects of Ecosystem Functioning. Arch. Environ. Contam. Toxicol. 25:160-169.

Brock, TCM, Roijackers RMM, Rollon R, Bransen F & L van der Heyden 1995. Effects of nutrient loading and insecticide application on the ecology of Elodea-dominated freswater microcosms: II. Responses of macrophytes, periphyton and macroinvertebrate grazers. Arch. Hydrobiol., 134: 53 - 74

Carder JP and KD Hoagland. 1998. Combined effects of alachlor and atrazine on benthic algal communities in artificial streams. Environmental Toxicology 17: 1415-1420

Day KE. 1986. The acute, chronic and sublethal effects of the synthetic pyrethroid, fenvalerate, on zooplankton in the laboratory and the field. Environ. Toxicol. Chem. 8:411-416.

Day KE, Kaushik NK a& KR Solomon. 1987. Impact of fenvalerate on enclosed freshwater planktonic communities and on in situ rates of filtration of zooplankton. Canadian Journal of Fisheries and Aquatic Sciences 44:1714-1728.

Day KE, Kaushik NK and KR Solomon. 1987. Impact of fenvalerate on enclosed freshwater planktonic communities and on in situ rates of filtration of zooplankton. Canadian Journal of Fisheries and Aquatic Sciences 44:1714-1728.

Dewey SL 1986. Effects of the herbicide atrazine on aquatic insect community structure and emergence in experimental ponds. Ecology 67:148-162.

Donk, E Van; Prins, H, Voogd, HM., Crum, SJH & TCM Brock. 1995. Effects of nutrient loading and insecticide application on the ecology of Elodea-dominated freshwater microcosms. I. Responses of plankton and zooplanktivorous insects. Archiv für Hydrobiologie 133 4: 417-439

Emans, HJB, Plassche van de EJ, Canton JH, Okkerman PC & PM Sparenburg.1993. Validation of some extrapolation methods used for effect assessment. Environ. Toxicol. Chem. 12: 2139-2154

Eriksson L, Hermens JLM, Johansson E, Verhaar HJM & S Wold. 1995. Multivatiate-analysis of aquatic toxicity data with PLS. Aquatic Sci, 57(3): 217-241

Faber MJ, Thompson DG, Stephenson GR & DP Kreutzweiser. 1998. Impact of glufosinate-ammonium and bialaphos on the zooplankton community of a small eutrophic northern lake. Environmental Toxicology & Chemistry 17: 1291-1299

Fairchild, JF, La Point TW, Zajicek JL, Nelson MN & FJ Dwyer. 1992a. Population, community, and ecosystem-level responses of aquatic mesocosms to pulsed doses of a pyrethroid insecticide. Environmental Toxicology and Chemistry 11:115-129

Fairchild JF, Point TWl, Zajicek JL, Nelson MK, Dwyer FJ & PA Lovely. 1992b. Population-, community- and ecosystem-level responses of aquatic mesocosms to pulsed doses of a pyrethroid insecticide. Environmental Toxicology and Chemistry 11:115-129.

Farmer D. I.R. Hill and SJ Maund. 1995. A comparison of the fate and effects of two pyrethroid insecticides (lambda-cyhalothrin and cypermethrin) in pond mesocosms. ECOTOXICOLOGY 4: 219-244

Fliedner, A. and Klein, W. 1996. Effects of Lindane on the Planktonic Community in Freshwater Microcosms Ecotoxicology and Environmental Safety 33: 228-235

Forsyth D., P.A. Martin and G.G. Shaw. 1997. Effects of herbicides on two submersed aquatic macrophytes, Potamogeton pectinatus L. and Myriophyllum sibiricum Komarov, in a prairie wetland. Environ. Pollut. 95: 259-268

Giddings, JM, Biever, RC, Helm, RL, Howick, GL & FJ de Noyelles. 1994. The fate and effects of guthion (Azinphos methyl) in mesocosms. In: Aquatic Mesocosm Studies in Ecological Risk Assessment. chap. 25: 469-495.

Giddings J, RC Biever and KD Racke. 1997. Fate of chlorpyrifos in outdoor pond microcosms and effects on growth and survival of bluegill sunfish. . Environ. Toxicol. Chem. 16: 2353-2362.

Giddings J, Biever RC, Annunziatio MF & A.J. Hosmer. 1996. Effects of diazinon on large outdoor pond microcosms. Environmental toxicology and chemistry. 15: 618-629.

Hall, L.W., Jr., M.C. Scott, W.D. Killen. 1998. Ecological Risk Assessment of Copper and Cadmium in Surface Waters of the Chesapeake Bay Watershed Environmental Toxicology and Chemistry. 17: 1172-1190

Hamer, MJ, Hill, IR, Rondon, L & A Caguan. 1994. The effects of lambda-cyhalothrin in aquatic field studies. In: Freshwater Field Tests For Hazard Assessment of Chemicals. pp. 331-337 Potsdam (Germany), Lewis Publishers.

Hamilton P., G.S. Jackson, N.K. Kaushik and K.R. Solomon. 1987. Impact of Atrazine on Lake Periphyton Communities, Including Carbon Uptake Dynamics using Track Autoradiography. Environ. Poll. 46: 83-103.

Hamilton, PB, Lean, DRS, Jackson, GS, Kaushik, NK & KR Solomon. 1989. The effect of two applications of atrazine on the water quality of freshwater enclosures. Environmental Pollution 60: 291-304

Havens K. 1995. Insecticide (carbaryl, 1-napthyl-N-methylcarbamate) effects on a freshwater plankton community: Zooplankton size, biomass, and algal abundance. Water, Air Soil Pollut 84: 1-10

Herman, D, Kaushik, NK & KR Solomon. 1986. Impact of atrazine on periphyton in freshwater enclosures and some ecological consequences Can J Fish Aquat Sci. 43: 1917-1925

Hill I., S.T. Hadfield, J.H. Kennedy and P. Ekoniak. 1988. Assessment of the impact of PP321 on aquatic ecosystems using tenth-acre experimental ponds. Proceedings of the ...British Crop Protection Conference-Pests and Diseases. 1988. v. 1 p. 309

Hoagland K. and R.W. Drenner. 1991. Freshwater community responses to mixtures of agricultural pesticides: Synergistic effects of atrazine and bifethrin. Texas Water Resources Institute Completion Report TR-151

Hoagland K., R.W. Drenner, J.D. Smith and D.R. Cross. 1993. Fresh-Water Community Responses to Mixtures of Agricultural Pesticides - Effects of Atrazine and Bifenthrin. Environmental Toxicology and Chemistry 12:627-637.

Hotelling, H. 1947. Multivariate quality control. In: Eisenhart et al. (Eds.): Techniques of Statistical Analysis. NY, McGraw-Hill.

Huber W. et al. 1994. Atrazine in aquatic test systems: An evaluation of ecotoxicological risks. Freshwater Field Tests For Hazard Assessment of Chemicals.

Huber, W. 1995. Effects of A-7504 C in aquatic outdoor microcosms Report Technical University Munic-Weihenstephan (from Danish EPA)

Johnson PC, Kennedy J, Morris RG, Hambleton FE & RL Graney. 1994. Fate and effects of cyfluthrin (pyrehtroid insecticide) in pond mesocosms and concrete microcosms. In: Aquatic mesocosm studies in ecological risk assessment. RL Graney, J.H. Kennedy, JH Rogers (Eds.). Society of Environmental Toxicity and Chemistry, Michigan: Lewis Publishers, pp. 337-371.

Jurgensen T. and K.D. Hoagland. 1990. Effects of short-term pulses of atrazine on attached algal communities in a small stream. Archives of environmental contamination and toxicology. 19: 617-623.

Jüttner I, Peither A, Lay JP, Kettrup A, Ormerod SJ. 1995. An outdoor mesocosm study to assess ecotoxicological effects of atrazine on a natural plankton community. Arch Environ Contam Toxicol 29: 435-441

Kaushik, N.K., Stephenson, G.L., Solomon, K.R., and Day, K.E., 1985, Impact of permethrin on zooplankton communities in limnocorrals: Can J Fish & Aquatic Scien. 42: 77-85.

Kettle W., F. DeNoyelles, Jr., B.D. Heacock and A.M. Kadoum. 1987. Diet and reproductive success of bluegill recovered from experimental ponds treated with atrazine. Bulletin of Environmental Contamination and Toxicology 38: 47-52.

Kreutzweiser DP & SS Capell. 1992a. A simple stream-side test system for determining acute lethal and behavioral effects of pesticides on aquatic insects. Environ. Toxicol. & Chem. 11: 993-999.

Kreutzweiser DP, Holmes SB & DJ Behmer. 1992b. Effects of the herbicides hexazinone and triclopyr ester on aquatic insects. Ecotoxicol. & Environ. Safety 23: 364-374

Kreutzweiser D., S.S. Capell and B.C. Sousa. 1995. Hexazinone effects on stream periphyton and invertebrate communities. Environmental toxicology and chemistry 14:1521-1527.

Kreutzweiser D., S.S. Capell, W.L. Wainio-Keizer and S.B. Holmes. 1993. Effects of a molt-accelerating insecticide on stream invertebrates. Canadian Technical Report of Fisheries and Aquatic Sciences 0

Kreutzweiser D & DR Thomas. 1995. Effects of a new molt-inducing insecticide, tebufenozide, on zooplankton communities in lake enclosures. Ecotoxicology 4: 307-328

Krieger K, Baker DB and JW Kramer. 1988. Effects of herbicides on stream Aufwuchs productivity and nutrient uptake. Archives of environmental contamination and toxicology. 17: 299-306

Larsen D, deNoyelles F, Stay F & T Shiroyama. 1986. Comparisons of Single-Species, Microcosm and Experimental Pond Responses to Atrazine Exposure. Environ. Toxicol. Chem. 5: 179-190

Liber K, Schmude KL & TD Corry. 1996. Effects of the insect growth regulator diflubenzuron on insect emergence within littoral enclosures. Environmental entomology. 25: 17-24.

Liess M. and R. Schulz. 1996. Chronic effects of short-term contamination with the pyrethroid insecticide fenvalerate on the caddisfly Limnephilus lunatus. Hydrobiologia 324: 99-106

Lozano S., S.L. Ohalloran, K.W. Sargent and J.C. Brazner. 1992. Effects of Esfenvalerate On Aquatic Organisms in Littoral Enclosures. Environmental Toxicology and Chemistry 1992, 11:35-47.

Ludwig G. 1993. Effects of trichlorfon, fenthion, and diflubenzuron on the zooplankton community and on production of reciprocal-cross hybrid striped bass fry in culture ponds. Aquaculture 110

Maund S., A. Peither, E.J. Taylor, I. Juttner, R. Beyerle-Pfnur, J.P. Lay and D. Pascoe. Toxicity of lindane to freshwater insect larvae in compartments of an experimental pond. Ecotoxicology and Environmental Safety 23: 76-88

Mayasich J. et al. 1994. Evaluation of the ecological and biological effects of tralomethrin utilizing an experimental pond system. In: Aquatic Mesocosm Studies in Ecological Risk Assessment. chap. 26:497-515.

Miljøstyrelsen. 1994. Technical guideline for hazard and risk assessment of industrial effluents. Environmental project report No. 256 Danish Environmental Protection Agency

Mitchell G., D. Bennett and N. Pearson. 1993. Effects of lindane on macroinvertebrates and periphyton in outdoor artificial streams. Ecotoxicol. Environ. Safety 25, 90-102

Neumann, Ch. 1995. Outdoor aquatic mesocosms study of the environmental fate and ecological effects Report; CGA 114900 EC 750 (A-7516A) Ciba-Geigy (from Danish EPA)

Peither A., I. Juttner, A. Kettrup and J.P. Lay. 1996. A pond mesocosm study to determine direct and indirect effects of lindane on a natural zooplankton community. Environmental Pollution. 93: 49-56.

OECD 1996. OECD Guidelines for Testing of Chemicals. Draft proposal for a guidance document - Freshwater lentic feld tests. July 1996. p.l2

Poirier D and GA Surgeoner. 1987. Laboratory flow-through bioassays of four forestry insecticides against stream invertebrates. Canadian Entomologist 119: 755-763.

Pusey, B., Arthington AH & J McLean. 1994. The effects of a pulsed application of chlorpyrfos on macroinvertebrate communities in an outdoor artificiel stream system. Ecotoxicology and Environmental Safety. 27: 221-250

Sarkar S. 1991. Effects of the herbicide 2,4-D on the bottom fauna of fish ponds. The progressive fish-culturist 53:161-165.

Scherer E. and R.E. McNicol. 1986. Behavioural responses of stream-dwelling Acroneuria lycorias (Ins., Plecopt.) larvae to methoxychlor and fenitrothion. Aquatic Toxicology 8:251-263.

Schroll H., C.L. Pedersen and P.H. Jespersen. 1998. Indirect effects of Esfenvalerate (Insecticie) on the density of periphytic algae in artificial ponds. Bull Environ Contam Toxicol. 60: 797-801

Sierszen M & SJ Lozano. 1998. Zooplankton population and community responses to the pesticide azinphos-methyl in freshwater littoral enclosures. Environmental Toxicology and Chemistry 17: 907-914

Spawn R, Hoagland KD a& BD Siegfried. 1997. Effects of alachlor on an algal community from a midwestern agricultural stream. Environmental Toxicology and Chemistry 16: 785-793.

Stephenson, Gladys L, Kaushik NK, Solomon KR & Kristen Day.1986. Impact of methoxychlor on freshwater communities of plankton in limnocorrals Environmental Toxicology and Chemistry. 5: 587-603

Tanner D & ML Knuth. 1996. Effects of esfenvalerate on the reproductive success of the bluegill sunfish, Lepomis macrochirus in littoral enclosures. Archives of Environmental Contamination and Toxicology 1996, 31: 244-251

Thompson D G, Homes SB, Thomas D, MacDonald L & KR Solomon. 1993. Impact of hexazinone and metsulfuron methyl on the phytoplankton community of a mixed- wood/boreal forest lake. Environ. Toxicol. Chem. 12:1695-1707

Tidou A. and e. al. 1992. Effects of lindane and deltamethrin on zooplankton communities of experimental ponds. Hydrobiologia 232:157-168.

Wagner, C. & Løkke, H. (1991) Estimations of ecotoxicological protection levels from NOEC toxicity data. Water Res. 24, 1237 – 1242

Ward S., A.H. Arthington and B.J. Pusey. 1995. The effects of a chronic application of chlorpyrifos on the macroinvertebrate fauna in an outdoor artificial stream system: species responses. Ecotoxicology and environmental safety. 30: 2-23.

Wayland M. 1991. Effect of carbofuran on selected macroinvertebrates in a prairie parkland pond: an enclosure approach. Archives of environmental contamination and toxicology. 21: 270-280.

Webber EC, Deutch WG, Bayne DR & WC Seesock. 1992. Ecosystem-level testing of synthetic pyrethroid insecticide in aquatic mesocosms. Environmental Toxicology and Chemistry, 11: 87-105.

Wijngaarden, RPA, Brink PJ van den, Crum SJH, Oude Voshaar JH, Brock TCM and P Leeuwangh. 1996. Effects of the insecticide Durban 4^®E (a.i. chlorpyrifos) in outdoor experimental ditches: I. Comparison of short-term toxicity between laboratory and field. Environ. Toxicol. Chem. 15: 1133 - 1142.

Yasuno M, Hanazato T, Iwakuma T, Takamura K & R Ueno. 1988. Effects of Permethrin on phytoplankton and zooplankton in an enclosure ecosystem in a pond. Hydrobiologia 159: 247-258.

10 Annex

10.1 Annex A

Single species toxicity data used for calculation of Hazard Concentrations. Effect parameter: BMS = biomass; PGR = population growth; IMM = immobilisation; MOR = mortality; Exp Typ (exposure type): S = static, F = flow through; NR = not recorded. Ref#: numbers denote reference number in Aquire data base; MST = data provided by the Danish EPA, Pesticide Manual.