Extrapolation of effects of pesticides on aquatic communities and ecosystems across different exposure patterns

Researcher: Mazhar Iqbal Zafar

Email the Author

Project runs from 20 September 2007 to 20 September 2011

Added on September 9, 2009

Introduction:

In Europe, environmental risks associated with appropriate pesticide use, is estimated using procedures described in the directive 91/414/EEC (EU, 1997). This directive uses a tiered approach: the first tier consists of well described rules, which are supposed to yield a worst-case risk assessment. When pesticides fail at the first tier, a higher tier risk assessment can be performed which consist of less well described procedures and is herewith more flexible to address the problems of concern. Herewith higher tiers, allow the inclusion of novel, state of the art experimental tools and simulation models, used to assess the risks more realistically. For the aquatic compartment, the protection aim of the risk assessment is not set at the individual level but at the population and community level. The directive states that no authorisation will be granted if the pesticide does not pass the first tier “unless it is clearly established through an appropriate risk assessment that under field conditions no unacceptable impact on the viability of exposed species occurs — directly or indirectly — after use of the plant protection product according to the proposed conditions of use.” Since no definition of “unacceptable” is provided in the directive, efforts have been put in defining scientifically sound higher tier studies to address this issue during workshops held during the last decade (e.g. HARAP (Campbell et al., 1999) and CLASSIC (Giddings et al., 2002)), attended by industry, academia and regulators. Although no clear criteria on acceptability could be set during these workshops, it is clear that short-term effects on populations might be acceptable if recovery is completed within a certain time-frame.

Pesticides used for crop protection in agriculture and horticulture may enter ditches, ponds, lakes and rivers in numerous ways such as direct overspray, spray drift, leaching to surface and ground water, run-off from land, and/or accidental spills (Goldsborough & Crumpton., 1998; Capri and Trevisan,1998 ). Consequently, these hazardous chemicals may affect the non-target biotic communities of such fresh water ecosystem (Hann & Goldsborough., 1997 ; McDougal et al.,1997). To protect the biological integrity of these waters it is important to determine the potential pesticide stress to aquatic ecosystems. Pesticides may affect aquatic life because they are designed specifically to kill organisms (both the noxious target organisms and other non-target ones) and they are released into the natural environment intentionally. It has been widely documented that pesticide concentrations in the natural environment are often high enough to kill certain organisms (Hatakeyama et al.,1994) and affect the structure and function of natural communities (Helgen et al.,1988; Hatakeyama et al.,1990; Liess et al., 2005). Pesticides exert their impact at multiple levels, including molecules, tissues, organs, individuals, populations and communities, and a variety of ecotoxicological tests have been designed to assess these effects (Cairns and Niederlehner,1995). However, this assessment is hindered by the fact that natural ecosystems are diverse and the effects are complicated.

 In the high tier , the risks of pesticides to aquatic ecosystems are often assessed by performing  experiments with man-made aquatic ecosystems, i.e. microcosms and mesocosms (cosms) evaluating a particular exposure regime (e.g. a pulse application), which do not necessarily correspond with the exposure for which the risk assessment is being conducted (e.g. multiple applications). To allow an appropriate linkage of the fate to the effects part of the risk assessment, the results of these cosm experiments, therefore, sometimes need extrapolation to a different exposure pattern than the one that evaluated in the cosm experiment itself (Boesten et al., 2007). This mismatch between exposure patterns observed in the field using chemical monitoring or predicted by models and the exposure regime used in experiments underpinning the effect assessment (e.g. cosm experiments) is one of the biggest challenges in contemporary ecological risk assessment. Therefore in 2007, two EU workshops ( ELink I and II) are organised to discuss how exposure and effects should be linked in ecological risk assessment procedures of pesticides under the European Plant Protection Products directive 91/414/EEC (Brock et al., 2009)  during of which the extrapolation of effects across exposure patterns is a major issue (elink-info.unicatt.it). Current procedures for aquatic risk assessment have not been able to adequately address some of the uncertainties arising from the time-variable surface water exposure profiles that are more often the rule rather than the exception in the field. Hence there is need to provide more information in understanding and addressing the uncertainties posed by time variable exposures.

At the population level much progress has been made recently by Ashauer and coworkers (Ashauer et al., 2006a,b; 2007).They present a model, called the Threshold Damage Model (TDM) that models the toxicokenetics and toxicodynamics of the pesticide in individuals to link exposure and effects. This 5 parameter model can be used to describe the uptake, elimination, damage, repair and threshold of death for a given species. Using this model effects at the species level can be predicted for any given exposure regime. At the community and ecosystem level, however, there is both a lack of theory as empirical data to perform such an extrapolation. This PhD project intends to establish empirical, experimental and modelling approaches to establish this extrapolation.

Aim:

This proposed  project  aims at gathering empirical data for extrapolation using the PERPEST data base of classified effects observed in cosm experiments and conducting experiments ourselves to test obtained hypotheses and improving the theoretical basis of extrapolation at the community level by upscaling of the TDM to the community level and testing its validity.To achieve this goal  we will make an attempt  to answer  the following research hypothesis :: (1) review relevant published cosm experiments to refine the exposure part of the PERPEST informatics model to allow an extrapolation of effects across exposure regimes using bioinformatics from which rules of thumb for extrapolation can be extracted, (2) perform experiments verifying or rejecting these rules of thumb, (3) to try to upscale the Threshold Damage Model(TDM) to the community level to allow extrapolation across exposure regime using modelling of toxicokenetics and toxicodynamics and (4) perform experiments to verify the outcome of this Community Threshold Damage Model.

Research:

In order to address our research hypotheses, first of all a literature review will be performed to describe and categorize the exposure patterns used in the cosm experiments. This review will consist of all those experiments that are currently  present in the data base underlying the PERPEST model (Van den Brink et al., 2002). The exposure will be grouped on the basis of their shape (peak versus constant), dissipation rate and number of applications (whether the exposure was resulting from a single or pulse application on the one side or multiple applications or chronic exposure on the other side). This data base is the basis of the bioinformatics PERPEST (Predicting the Ecological Risks of PESTicides in freshwater ecosystems) model that is used to predict the effects of pesticides using information across compounds and (if wanted) across mode of actions (Van den Brink et al., 2002; 2006b).­ The ELink I workshop addressed a need for such a categorization, which will be drafted before and discussed in the ELink II workshop in September 2007. We will use this categorization proposed by the ELink workshops as a starting point.

When the exposure patterns of all cosm experiment are categorized, their effects in terms of intensity of effects (as incorporated in the PERPEST model: www.perpest.wur.nl) and duration can be compared on the basis of nominal concentrations (expressed as toxic units, scaled using the median laboratory toxicity value based on the expected sensitive group, e.g. arthropods for insecticides). This comparison can be done for chemicals separately (i.e. compare the effects one addition of the insecticide chlorpyrifos with multiple additions of chlorpyrifos) and within toxicological mode of actions (i.e. compare the effects of one addition of the insecticide lambda-cyhalothrin with multiple additions of deltamethrin). The latter can be done because the nominal concentrations are scaled on the basis of their general toxicity to the most sensitive group. This comparison will yield rules of thumbs to extrapolate effects across exposure patterns and also whether these rules of thumb are unique for a chemical or can be generalised over chemicals with the same toxicological mode of action.

Secondly, experiments will be performed to verify these rules of thumb. To assess the spatial and temporal effects of pesticides on populations and communities, experiments in microcosms and mesocosms (cosms) will be performed evaluating different exposure patterns using different concentrations for  chemicals . By measuring physico-chemical parameters and sampling biological communities in time, the effects at different treatment levels can be followed in time. By doing so threshold values of effects can be set at the population, community and ecosystem level, recovery patterns can be extracted for populations and communities and indirect effect patterns can be evaluated (Van den Brink, 1999). Simple, small, indoor cylinders (Van Wijngaarden et al., 2005b) hosting a planctonic aquatic ecosystem will be used to achieve this number of experiments. Results will be evaluated using univariate and multivariate techniques to elucidate population- and community-level effects, respectively. Classification of effects on the basis of their intensity and duration will be done to compare the effects at the ecosystem level. It will be evaluated whether the results of these comparisons verify or contradict with the rules of thumb deduced from the empirical analysis using the PERPEST data base as a literature review. 

Finally experiments will be performed using ecosystems that mimic the simple modelled ecosystem. The ecosystem will consist of the same physico-chemical and biological components as present in the model described above. One experiment evaluating different exposure patterns for different concentrations of Chlorpyrifos will be used to test and adjust the model, another experiment evaluating different exposure patterns and concentrations to validate or reject the model.

External funding

Yes

External Funding Sponsors

NUFFIC- Netherlands organization for international cooperation in higher education; “water framework directive” research programmes of the Dutch Ministry of Agriculture, Nature, and Food Safety

Research Group

WUR - Aquatic Ecology and Water Quality Management

FTE

1.00

Extrapolation of effects of pesticides on aquatic communities and ecosystems across different exposure patterns

Researcher: Mazhar Iqbal Zafar

Email the Author

Project runs from to

Added on September 9, 2009

Introduction:

In Europe, environmental risks associated with appropriate pesticide use, is estimated using procedures described in the directive 91/414/EEC (EU, 1997). This directive uses a tiered approach: the first tier consists of well described rules, which are supposed to yield a worst-case risk assessment. When pesticides fail at the first tier, a higher tier risk assessment can be performed which consist of less well described procedures and is herewith more flexible to address the problems of concern. Herewith higher tiers, allow the inclusion of novel, state of the art experimental tools and simulation models, used to assess the risks more realistically.

For the aquatic compartment, the protection aim of the risk assessment is not set at the individual level but at the population and community level. The directive states that no authorisation will be granted if the pesticide does not pass the first tier “unless it is clearly established through an appropriate risk assessment that under field conditions no unacceptable impact on the viability of exposed species occurs — directly or indirectly — after use of the plant protection product according to the proposed conditions of use.” Since no definition of “unacceptable” is provided in the directive, efforts have been put in defining scientifically sound higher tier studies to address this issue during workshops held during the last decade (e.g. HARAP (Campbell et al., 1999) and CLASSIC (Giddings et al., 2002)), attended by industry, academia and regulators. Although no clear criteria on acceptability could be set during these workshops, it is clear that short-term effects on populations might be acceptable if recovery is completed within a certain time-frame.

Pesticides used for crop protection in agriculture and horticulture may enter ditches, ponds, lakes and rivers in numerous ways such as direct overspray, spray drift, leaching to surface and ground water, run-off from land, and/or accidental spills (Goldsborough & Crumpton., 1998; Capri and Trevisan,1998 ). Consequently, these hazardous chemicals may affect the non-target biotic communities of such fresh water ecosystem (Hann & Goldsborough., 1997 ; McDougal et al.,1997). To protect the biological integrity of these waters it is important to determine the potential pesticide stress to aquatic ecosystems. Pesticides may affect aquatic life because they are designed specifically to kill organisms (both the noxious target organisms and other non-target ones) and they are released into the natural environment intentionally. It has been widely documented that pesticide concentrations in the natural environment are often high enough to kill certain organisms (Hatakeyama et al.,1994) and affect the structure and function of natural communities (Helgen et al.,1988; Hatakeyama et al.,1990; Liess et al., 2005). Pesticides exert their impact at multiple levels, including molecules, tissues, organs, individuals, populations and communities, and a variety of ecotoxicological tests have been designed to assess these

Aim:

hhhhhhhhhhhhhhhh

Research: