biological effects. For example, in the whole effluent toxicity (WET) testing,
nutrients such as nitrogen or phosphorus (usually enriched in municipal waters)
may stimulate the growth of algae in the bioassay, which often outweighs the effect
of other toxicants leading to “false negative” responses (Sandermann 2004). Similarly, variability of the composition of even standardized artificial soils affected the
ecotoxicity of PAHs and other toxicants by orders of the magnitude (Bielská et al.
2017; Hofman et al. 2014; Vašíčková et al. 2015).
Despite the promotion by international authorities including, e.g., EU, routine
application of bioassays in regulatory monitoring or field surveys is still at the stage
of research or ad hoc risk assessment studies. Nevertheless, some environmental
monitoring programs implemented toxicity testing with bioassay(s) such as whole
effluent toxicity testing in the USA or Germany (Escher and Leusch 2011) or
EU-supported effect-based monitoring programs (Tousova et al. 2017). Additional
flagship initiatives include, e.g., effect-based assessment of estrogen endocrine
disrupters in the effluents (Centre Ecotox 2018).
A number of open issues remain to be thoroughly addressed before broader
acceptance of bioassays in distributed monitoring, including:
• Validation of the bioassay responses for samples with different composition and
variable situations. Responses of the bioassay should be characterized across
gradients of environmentally realistic conditions (e.g., in water for concentrations
of ions, pH, OC levels, oxygen, ammonia). This should be done for both model
laboratory samples (factorial design) and testing diverse field samples of variable
composition.
• Quantitative characterization of the bioassay sensitivity and performance for both
model laboratory and field samples should be provided, i.e., characterization of
limits of detection, risks/frequency of false positives/negatives.
• Variability should be assessed and characterized by dedicated inter-laboratory
exercise(s) that should demonstrate the repeatability and the robustness of the
bioassay responses to standardized, yet chemically complex samples in different
laboratories and conditions.
• Definition of thresholds or trigger values for each bioassay is needed. For
example, it should be clearly defined what response would be considered significant or acceptable or what biological response would indicate non-acceptable
risk. Examples of such trigger values could be, e.g., quantitative inhibition—
percentage of immobilization of Daphnia magna in undiluted sample (or 5Â or
10Â diluted) or IC20 (IC50, etc.) expressed as dilution factor of the original
sample.
Unlike the effect-based monitoring, ecotoxicity testing of the contaminated
samples before, during, and after remediation provides invaluable information on
changes of the overall toxicity during the remediation process. The problem of the
matrix interference discussed above is less relevant because of the specificity to
certain site. Parameters of the soil or water matrix usually do not greatly change
during the treatment, which should allow for clear observation of changes in the
eventual ecotoxicity during the treatment process.
27 Ecotoxicology of Environmental Pollutants
557
nutrients such as nitrogen or phosphorus (usually enriched in municipal waters)
may stimulate the growth of algae in the bioassay, which often outweighs the effect
of other toxicants leading to “false negative” responses (Sandermann 2004). Similarly, variability of the composition of even standardized artificial soils affected the
ecotoxicity of PAHs and other toxicants by orders of the magnitude (Bielská et al.
2017; Hofman et al. 2014; Vašíčková et al. 2015).
Despite the promotion by international authorities including, e.g., EU, routine
application of bioassays in regulatory monitoring or field surveys is still at the stage
of research or ad hoc risk assessment studies. Nevertheless, some environmental
monitoring programs implemented toxicity testing with bioassay(s) such as whole
effluent toxicity testing in the USA or Germany (Escher and Leusch 2011) or
EU-supported effect-based monitoring programs (Tousova et al. 2017). Additional
flagship initiatives include, e.g., effect-based assessment of estrogen endocrine
disrupters in the effluents (Centre Ecotox 2018).
A number of open issues remain to be thoroughly addressed before broader
acceptance of bioassays in distributed monitoring, including:
• Validation of the bioassay responses for samples with different composition and
variable situations. Responses of the bioassay should be characterized across
gradients of environmentally realistic conditions (e.g., in water for concentrations
of ions, pH, OC levels, oxygen, ammonia). This should be done for both model
laboratory samples (factorial design) and testing diverse field samples of variable
composition.
• Quantitative characterization of the bioassay sensitivity and performance for both
model laboratory and field samples should be provided, i.e., characterization of
limits of detection, risks/frequency of false positives/negatives.
• Variability should be assessed and characterized by dedicated inter-laboratory
exercise(s) that should demonstrate the repeatability and the robustness of the
bioassay responses to standardized, yet chemically complex samples in different
laboratories and conditions.
• Definition of thresholds or trigger values for each bioassay is needed. For
example, it should be clearly defined what response would be considered significant or acceptable or what biological response would indicate non-acceptable
risk. Examples of such trigger values could be, e.g., quantitative inhibition—
percentage of immobilization of Daphnia magna in undiluted sample (or 5Â or
10Â diluted) or IC20 (IC50, etc.) expressed as dilution factor of the original
sample.
Unlike the effect-based monitoring, ecotoxicity testing of the contaminated
samples before, during, and after remediation provides invaluable information on
changes of the overall toxicity during the remediation process. The problem of the
matrix interference discussed above is less relevant because of the specificity to
certain site. Parameters of the soil or water matrix usually do not greatly change
during the treatment, which should allow for clear observation of changes in the
eventual ecotoxicity during the treatment process.
27 Ecotoxicology of Environmental Pollutants
557