When addressing the use of fish for the environmental safety of chemicals and effluents, there are many opportunities for applying the principles of the 3Rs: Reduce, Refine, and Replace. The current environmental regulatory testing strategy for bioconcentration and secondary poisoning has been reviewed, and alternative approaches that provide useful information are described. Several approaches can be used to reduce the number of fish used in the Organization for Economic Cooperation and Development (OECD) Test Guideline 305, including alternative in vivo test methods such as the dietary accumulation test and the static exposure approach. The best replacement approach would seem to use read-across, chemical grouping, and quantitative structure–activity relationships with an assessment of the key processes in bioconcentration: Adsorption, distribution, metabolism, and excretion. Biomimetic extraction has particular usefulness in addressing bioavailable chemicals and is in some circumstances capable of predicting uptake. Use of alternative organisms such as invertebrates should also be considered. A single cut-off value for molecular weight and size beyond which no absorption will take place cannot be identified. Recommendations for their use in bioaccumulative (B) categorization schemes are provided. Assessment of biotransformation with in vitro assays and in silico approaches holds significant promise. Further research is needed to identify their variability and confidence limits and the ways to use this as a basis to estimate bioconcentration factors. A tiered bioconcentration testing strategy has been developed taking account of the alternatives discussed.

The ability to predict metal toxicity in sediments based on measurements of simple chemical parameters is not possible using currently available sediment-quality guidelines (SQGs). Past evaluations of available SQGs for metals indicated little difference in their predictive abilities; however, the scientific understanding of cause–effect relationships is progressing rapidly. Today, it is clear that they can be protective of benthic ecosystem health, but single-value SQGs will be ineffective for predicting the toxicity of metals in sediments. Recent exposure–effects models and the sediment biotic ligand model both indicate that a better approach would be to have SQG concentrations, or ranges, that are applied to different sediment types. This review indicates that significant improvements in laboratory and field-based measurements, better recording of parameters that influence metal toxicity in sediments, as well as quantification of the metal exposure routes and the relative contribution of dissolved and particulate sources to toxic effects are needed to improve the power of predictive models and the overall effectiveness of SQGs for metals. Simply exposing benthic organisms to contaminated sediments and reporting effects concentrations or thresholds based on particulate metal concentrations will provide little information to aid future SQG development. For all tests, careful measurement and reporting of concentrations of particulate metal-binding phases (e.g., sulfide, organic carbon, and iron phases), metal partitioning between porewater and sediments, and porewater pH are considered as minimum data requirements. When using metal-spiked sediments, much better efforts are required to achieve sediment properties that resemble those of naturally contaminated sediments. Our current understanding of metal toxicity indicates that considerably greater information requirements will be needed to predict sublethal and chronic effects of metals, because the toxic, metabolically available concentration of metals within an organism will fluctuate over time. Based on the review of exposure and effects models, along with improved measurement of metal exposure-related parameters, the measurement of the short-term uptake rate of metals into organisms is likely to improve future models.

A new regression-based copper toxicity model was applied in a case study of San Francisco Bay, California, USA, to demonstrate its utility in estimating risk and site-specific water quality criteria. This was accomplished using probabilistic techniques and a simple model relating dissolved organic carbon (DOC) concentrations with the toxicity of dissolved copper to embryos of the most copper sensitive taxon (Mytilus) in the US Environmental Protection Agency's (USEPA) water quality criteria database. Similar probabilistic techniques were applied to data developed for San Francisco Bay using the USEPA's water-effect ratio (WER) methods for comparison with the DOC-based method. Based on 595 site- and date-specific DOC model observations at 26 sites in San Francisco Bay, none suggested risk of chronic toxicity. Safety factors (1/risk quotient) on average across all sites ranged from 2.4 to 9.1. Comparisons were made between 1) estimates of site-specific criteria made using the DOC method, 2) estimates of site-specific criteria made using the WER method, 3) USEPA national and California Toxics Rule criteria, and 4) region-specific criteria recommended for regulatory implementation by the Clean Estuary Partnership. The DOC- and WER-based methods indicated that copper criteria for San Francisco Bay could be increased above USEPA and California Toxics Rule criteria and will retain the level of protection (≥97%) embodied in the USEPA copper saltwater water quality criteria. The DOC method overall was more conservative (i.e., implies the need for lower criteria in the Bay) than the WER method. The DOC method suggests that the region-specific criteria being recommended for regulatory implementation would be underprotective in some areas and yet could be increased and remain protective in other areas of San Francisco Bay.

Toxicity data for tropical species are often lacking for ecological risk assessment. Consequently, tropical and subtropical countries use water quality criteria (WQC) derived from temperate species (e.g., United States, Canada, or Europe) to assess ecological risks in their aquatic systems, leaving an unknown margin of uncertainty. To address this issue, we use species sensitivity distributions of freshwater animal species to determine whether temperate datasets are adequately protective of tropical species assemblages for 18 chemical substances. The results indicate that the relative sensitivities of tropical and temperate species are noticeably different for some of these chemicals. For most metals, temperate species tend to be more sensitive than their tropical counterparts. However, for un-ionized ammonia, phenol, and some pesticides (e.g., chlorpyrifos), tropical species are probably more sensitive. On the basis of the results from objective comparisons of the ratio between temperate and tropical hazardous concentration values for 10% of species, or the 90% protection level, we recommend that an extrapolation factor of 10 should be applied when such surrogate temperate WQCs are used for tropical or subtropical regions and a priori knowledge on the sensitivity of tropical species is very limited or not available.

Population-level assessments of the ecological risks of dioxin-like polychlorinated biphenyl (PCB) exposure to fish-eating birds in Tokyo Bay and its vicinity were performed to judge the need for risk management measures to protect aquatic wildlife from dioxin-like PCB contamination. Egg mortality risk and changes in the population growth rate (λ) in relation to the contamination levels of dioxin-like PCBs in eggs of gray heron (Ardea cinerea), great cormorant (Phalacrocorax carbo), osprey (Pandion halieaetus), and kingfisher (Alcedo atthis) were determined by integrating the results of exposure analysis, effect analysis, and a life-history model for each species. The egg mortality risks for the gray heron, great cormorant, osprey, and kingfisher populations were calculated to be 5.8, 6.8, 12, and less than 1%, respectively. The estimated λ for those populations were calculated to be 1.061, 1.405, 1.024, and 1.131, respectively. The percentage changes in λ for those populations were estimated to be 1.2, 2.0, 1.6, and less than 1%, respectively. Our results implied that the levels of dioxin-like PCBs observed in the Tokyo Bay area alone would not have significant population-level effects on the fish-eating bird populations. It is concluded that along with the trend toward decreasing dioxin and dioxin-like PCB levels in Tokyo Bay, no urgent need exists for risk-reduction measures to protect fish-eating bird populations against dioxin-like PCBs.

The spread of nonindigenous (nonnative) species introduced into the United States is a significant and growing national problem and results in lost agricultural productivity, increased health problems, native species extinctions, and expensive prevention and eradication efforts. Thousands of nonindigenous species have either become established or spread, and introduction of aquatic nuisance species (ANS) into freshwater lakes threaten aquatic biodiversity. Expanding global trade is likely to increase the number of species that are spread across the globe, so the need to develop an approach to predict potential ANS invasions is great. Risk assessments currently being used to assess ANS risk rely on qualitative or semiquantitative information and expert opinion; thus, such approaches lack transparency and repeatability. A more quantitative approach is needed to augment the qualitative approaches currently in use. A quantitative approach with the use of the traditional ecological risk assessment (traditional ERA) framework combined with decision analysis tools was developed for assessing ANS risks in which the causative ecological risk agent is an organism rather than a chemical. This paper presents a systematic risk assessment framework that includes structured decision analysis to help organize and analyze pertinent data, state assumptions, address uncertainties in estimating the probability of an undesired ANS introduction, or spread and integrate these outputs with stakeholder values. This paper also describes when and how decision analysis tools can be used in such assessments for ANS. This framework and methodology will enable risk managers to systematically evaluate and compare alternatives and actions supporting ANS risk management and thus credibly prioritize resources.

To evaluate the relative toxicity and the occurrence patterns of pesticide mixtures in streams draining agricultural watersheds, a 3-step approach was used. First, a landscape of interest was identified, defined, and isolated. Second, the relative toxicity of mixtures, on the basis of pesticide toxicity index scores, was compared with the relative toxicity of the highest individual pesticide, on the basis of highest toxicity quotient values. Third, occurrence patterns of pesticide mixtures were identified for use in follow-up mechanistic studies. The landscape of interest was identified as the corn and soybeans crop setting and concentrations of pesticides in streams within this crop setting were determined from US Geological Survey data. Pesticide toxicity index scores for individual samples were highest for the primary producers, Pseudokirchneriella subcapitata and Lemna gibba; with 95th percentile pesticide toxicity index scores of 4.7 and 1.9, respectively. The 95th percentile pesticide toxicity index score for Daphnia magna was 0.40 when a chronic sublethal endpoint was used. Pesticide toxicity index values were above the highest toxicity quotient values, indicating that consideration of mixtures does increase the estimated risk, but pesticide toxicity index scores were generally within a factor of 2 of highest toxicity quotient values, indicating that the increased risk is not large for most samples. Pesticide toxicity index scores tended to be dominated by individual pesticides and simple mixtures. Two different prioritization strategies were used to identify important mixtures for further study on the basis of potential effects on P. subcapitata. Both techniques decreased the complexity of the pesticide mixtures to consider by reducing the number of components within the identified mixtures as well as identifying a few specific combinations that constitute the majority of mixtures within the sample. Nearly all important pesticides for P. subcapitata were herbicides from 2 chemical classes: Acetanilide and triazine herbicides.

The object of this study was to estimate site- and region-specific dissolved copper criteria for a large embayment, the Chesapeake Bay, USA. The intent is to show the utility of 2 copper saltwater quality site-specific criteria estimation models and associated region-specific criteria selection methods. The criteria estimation models and selection methods are simple, efficient, and cost-effective tools for resource managers. The methods are proposed as potential substitutes for the US Environmental Protection Agency's water effect ratio methods. Dissolved organic carbon data and the copper criteria models were used to produce probability-based estimates of site-specific copper saltwater quality criteria. Site- and date-specific criteria estimations were made for 88 sites (n = 5,296) in the Chesapeake Bay. The average and range of estimated site-specific chronic dissolved copper criteria for the Chesapeake Bay were 7.5 and 5.3 to 16.9 μg Cu/L. The average and range of estimated site-specific acute dissolved copper criteria for the Chesapeake Bay were 11.7 and 8.3 to 26.4 μg Cu/L. The results suggest that applicable national and state copper criteria can increase in much of the Chesapeake Bay and remain protective. Virginia Department of Environmental Quality copper criteria near the mouth of the Chesapeake Bay, however, need to decrease to protect species of equal or greater sensitivity to that of the marine mussel, Mytilus sp.

Watershed management processes continue to call for more science and improved decision making that take into account the full range of stakeholder perspectives. Increasingly, the core principles of ecological risk assessment (i.e., the development and use of assessment endpoints and conceptual models, conducting exposure and effects analysis) are being incorporated and adapted in innovative ways to meet the call for more science. Similarly, innovative approaches to adapting decision analysis tools and methods for incorporating stakeholder concerns in complex natural resource management decisions are being increasingly applied. Here, we present an example of the integration of ecological risk assessment with decision analysis in the development of a watershed management plan for the Greater Vancouver Water District in British Columbia, Canada. Assessment endpoints were developed, ecological inventory data were collected, and watershed models were developed to characterize the existing and future condition of 3 watersheds in terms of the potential risks to water quality. Stressors to water quality include sedimentation processes (landslides, streambank erosion) and forest disturbance (wildfire, major insect or disease outbreak). Three landscape-level risk management alternatives were developed to reflect different degrees of management intervention. Each alternative was evaluated under different scenarios and analyzed by explicitly examining value-based trade-offs among water quality, environmental, financial, and social endpoints. The objective of this paper is to demonstrate how the integration of ecological risk assessment and decision analysis approaches can support decision makers in watershed management.

Selenium is increasingly an issue for a wide range of mining, industrial, and agricultural operations. Appropriate methods for evaluating the impacts of selenium in aquatic ecosystems are vigorously debated in the literature. Two common approaches include the use of tissue residue guidelines and reproductive toxicity testing using field-collected fish; however, each approach on its own does not provide sufficient evidence that wild fish populations are in fact impaired. The limitations of each method are discussed, and recommendations to improve the relevance of each line of evidence are provided. A 3rd line of evidence, field measurement of fish population dynamics, is proposed and also discussed. A framework, consistent with an ecological risk assessment methodology, for the design, application, and interpretation of selenium weight-of-evidence investigations is proposed.

Toxigenic cyanobacteria, commonly known as blue green algae, are an emerging public health issue. The toxins produced by cyanobacteria have been detected across the United States in marine, freshwater and estuarine systems and associated with adverse health outcomes. The intent of this paper is to focus on how to address risk in a recreational freshwater scenario when toxigenic cyanobacteria are present. Several challenges exist for monitoring, assessing and posting water bodies and advising the public when toxigenic cyanobacteria are present. These include addressing different recreational activities that are associated with varying levels of risk, the dynamic temporal and spatial aspects of blooms, data gaps in toxicological information and the lack of training and resources for adequate surveillance. Without uniform federal guidance, numerous states have taken public health action for cyanobacteria with different criteria. Vermont and Oregon independently developed a tiered decision-making framework to reduce risk to recreational users when toxigenic cyanobacteria are present. This framework is based on a combination of qualitative and quantitative information.

In Oregon's Willamette River Basin, health advisories currently limit consumption of fish that have accumulated methylmercury to levels posing a potential health risk for humans. Under the Clean Water Act, these advisories represent an impairment of the beneficial use of fish consumption and create the requirement for a mercury total maximum daily load. A percent load reduction for total mercury was determined by comparing mercury levels in surface water to a water column guidance value linked to the protection of specified beneficial uses. In this case study, we discuss how probabilistic (Monte Carlo) methods were used to quantify uncertainty in the water column guidance value, how they provided decision makers with knowledge as to the probability of any given water column guidance value affording human health protection for methylmercury, and how this knowledge affected decisions as to a mercury load reduction for the Willamette River Basin. Through consultations with stakeholders, a water column guidance value of 0.92 ng/L (a median for higher trophic level fish) was chosen from among a suite of values of differing probabilities. The selected water column guidance value, when compared with ambient total mercury levels, indicated that a 50% probability of achieving the tissue criterion would require a load reduction of about 26%. Having and working with an explicit knowledge of uncertainty was not easy for many decision makers or stakeholders. However, such knowledge gave them more informed choices, a better understanding of what a specific choice of water column guidance value could mean in terms of achieving protectiveness, and led to a lower load reduction than suggested by a purely deterministic analysis. Nonetheless, more attention must be given to developing management, communication, and regulatory frameworks that can effectively use the greater knowledge of uncertainty afforded by probabilistic methods.

The principal conclusion of a workshop in October 2005 at RIVM (Bilthoven, The Netherlands) on the assessment of field studies with pesticides for authorization is that the lack of a definition of acceptability of effects is recognized as a problem by all stakeholders: Industry, risk assessors, and regulators. Because of this lack of definition in the legislation, it is unclear what critical effect values should be assessed in field studies. Despite the extensive documentation on field study performance, the decision making is not based on justifiable scientific opinions or publicly shared values but on technical limitations of the test design instead. In the workshop, research was identified that should result in a scientific basis for value judgments applied in decision making.