"Quantifying the precision of ecological risk: Misunderstandings and errors in the methods for assessment factors versus species sensitivity distributions".

The science of species sensitivity distributions (SSDs) is a blend of statistical theory, ecotoxicological testing, study reliability, and biodiversity. The utility of SSDs has been well reviewed and they are viewed as a high tier assessment tool in environmental risk assessment and other disciplines. SSDs seek to improve upon probabilistic extrapolation of laboratory (and sometimes field) collected ecotoxicity data for environmental protection by modeling the diversity of multiple experimental results in the form of a single statistical distribution which reduces or eliminates the need for extrapolation with deterministic assessment factors. SSDs thus depend heavily on both statistical and biological knowledge. In this commentary we review recently published literature identifying areas of improvement based on fundamental statistical theory or application in environmental assessment contexts. We reveal that sound application of SSDs relies heavily upon a grasp of probability distributions, how asymmetric confidence intervals are derived for distributions common to SSDs, the influence of sample size on parameter estimation, and how these are collectively applied across the myriad of regulatory systems globally. Statisticians and ecotoxicologists are inextricably bound together in the goal of actually improving hazard assessment using both probabilistic and deterministic methodologies.

Ecotoxicity thresholds for ametryn, diuron, hexazinone and simazine in fresh and marine waters.

Triazine and urea herbicides are two groups of photosystem II inhibiting herbicides frequently detected in surface, ground and marine waters. Yet, there are few water quality guidelines for herbicides. Ecotoxicity thresholds (ETs) for ametryn, hexazinone and simazine (triazine herbicides) and diuron (a urea herbicide) were calculated using the Australian and New Zealand method for deriving guideline values to protect fresh and marine ecosystems. Four ETs were derived for each chemical and ecosystem that should theoretically protect 99, 95, 90 and 80% of species (i.e. PC99, PC95, PC90 and PC80, respectively). For all four herbicides, the phototrophic species were significantly more sensitive than non-phototrophic species, and therefore, only the former data were used to calculate the ETs. Comparison of the ET values to measured concentrations in 2606 samples from 15 waterways that discharge to the Great Barrier Reef (2011-2015) found three exceedances of the simazine PC99, regular exceedances (up to 30%) of the PC99 in a limited number of rivers for ametryn and hexazinone and frequent (> 40%) exceedances of the PC99 and PC95 ETs in at least four waterways for diuron. There were no exceedances of the marine ETs in inshore reef areas. Further, ecotoxicity data are required for ametryn and hexazinone to fresh and marine phototrophic species, for simazine to marine phototrophic species, for tropical phototrophic species, repeated pulse exposures and long-term (2 to 12 months) exposures to environmentally relevant concentrations.

A Data Fusion Pipeline for Generating and Enriching Adverse Outcome Pathway Descriptions.

Increasing amounts of systems toxicology data, including omics results, are becoming publically available and accessible in databases. Data-driven and informatics-tool supported pipeline schemas for fitting such data into Adverse Outcome Pathway (AOP) descriptions could potentially aid the development of nonanimal-based hazard and risk assessment methods. We devised a 6-step workflow that integrated diverse types of toxicology data into a novel AOP scheme for pulmonary fibrosis. Mining of literature references and diverse data sources covering previous pathway descriptions and molecular results were coupled in a stepwise manner with informatics tools applications that enabled gene linkage and pathway identification in molecular interaction maps. Ultimately, a network of functional elements coupled 64 pulmonary fibrosis-associated genes into a novel, open-source AOP-linked molecular pathway, now available for commenting and improvements in WikiPathways (WP3624). Applying in silico-based knowledge extraction and modeling, the pipeline enabled screening and fusion of many different complex data types, including the integration of omics results. Overall, the taken, stepwise approach should be generally useful to construct novel AOP descriptions as well as to enrich developing AOP descriptions in progress.

Analysis of neurobehavioural data by chemometric methods in ecotoxicological studies.

Incorporation of chemometric tools in behavioural data management workflows allows for the early identification of most relevant endpoints complementarily to statistical confirmatory approaches. In this work, the effects of two model neurotoxicants, chlorpyrifos (CPF) and nicotine, exposures on behavioural profiles of adult zebrafish at three different times (2, 6 and 24h) were evaluated using open field test (OFT) paradigm experiments. Two chemometric methods like Principal Component Analysis (PCA) and Analysis of Variance-Simultaneous Component Analysis (ASCA) have been used to interpret the changes observed in the obtained behavioural data. A decreased of the locomotor activity, an anxiolytic effect and an altered exploratory behaviour were the most affected behavioural endpoints in the CPF exposures. However, an increase of the locomotor activity and an anxiogenic effect were observed in the nicotine exposures. Finally, an excellent correlation between the ASCA results and the results obtained using traditional statistical procedures for both compounds were encountered.

Accuracy Evaluation of Three Modelling Tools for Occupational Exposure Assessment.

OBJECTIVES: The objective of this study is to evaluate the accuracy and robustness of three exposure-modelling tools [STOFFENMANAGER® v.6, European Centre for Ecotoxicology and Toxicology of Chemical Target Risk Assessment v.3.1 (ECETOC TRA v.3.1), and Advanced REACH Tool (ART v.1.5)], by comparing available measured data for exposure to organic solvents and pesticides in occupational exposure scenarios (ESs). METHODS: Model accuracy was evaluated by comparing the predicted and the measured values, expressed as an underestimation or overestimation factor (PRED/EXP), and by regression analysis. Robustness was quantitatively described by the so-called variable 'Uncertainty Factor' (UF), which was attributed to each model's input: a higher UF score indicates greater model uncertainty and poorer robustness. RESULTS: ART was the most accurate model, with median PRED/EXP factors of 1.3 and 0.15 for organic solvent and pesticide ESs, respectively, and a significant correlation (P < 0.05) among estimated and measured data. As expected, Tier 1 model ECETOC TRA demonstrated the worst performance in terms of accuracy, with median PRED/EXP factors of 2.0 for organic solvent ESs and 3545 for pesticide ESs. Simultaneously, STOFFENMANAGER® showed a median UF equal to 2.0, resulting in the most robust model. DISCUSSION: ECETOC TRA was not considered acceptable in terms of accuracy, confirming that this model is not appropriate for the evaluation of the selected ESs for pesticides. Conversely, STOFFENMANAGER® was the best choice, and ART tended to underestimate the exposure to pesticides. For organic solvent ESs, there were no cases of strong underestimation, and all models presented overall acceptable results; for the selected ESs, ART showed the best accuracy. Stoffenmanager was the most robust model overall, indicating that even with a mistake in ES interpretation, predicted values would remain acceptable. CONCLUSION: ART may lead to more accurate results when well-documented ESs are available. In other situations, Stoffenmanager appears to be a safer alternative because of its greater robustness, particularly when entry data uncertainty is difficult to assess. ECETOC TRA cannot be directly compared to higher tiered models because of its simplistic nature: the use of this tool should be limited only to exceptional cases in which a strong conservative and worst-case evaluation is necessary.

(Q)SARs to predict environmental toxicities: current status and future needs.

The current state of the art of (Quantitative) Structure-Activity Relationships ((Q)SARs) to predict environmental toxicity is assessed along with recommendations to develop these models further. The acute toxicity of compounds acting by the non-polar narcotic mechanism of action can be well predicted, however other approaches, including read-across, may be required for compounds acting by specific mechanisms of action. The chronic toxicity of compounds to environmental species is more difficult to predict from (Q)SARs, with robust data sets and more mechanistic information required. In addition, the toxicity of mixtures is little addressed by (Q)SAR approaches. Developments in environmental toxicology including Adverse Outcome Pathways (AOPs) and omics responses should be utilised to develop better, more mechanistically relevant, (Q)SAR models.

Are engineered nano iron oxide particles safe? an environmental risk assessment by probabilistic exposure, effects and risk modeling.

Nano iron oxide particles are beneficial to our daily lives through their use in paints, construction materials, biomedical imaging and other industrial fields. However, little is known about the possible risks associated with the current exposure level of engineered nano iron oxides (nano-FeOX) to organisms in the environment. The goal of this study was to predict the release of nano-FeOX to the environment and assess their risks for surface waters in the EU and Switzerland. The material flows of nano-FeOX to technical compartments (waste incineration and waste water treatment plants) and to the environment were calculated with a probabilistic modeling approach. The mean value of the predicted environmental concentrations (PECs) of nano-FeOX in surface waters in the EU for a worst-case scenario (no particle sedimentation) was estimated to be 28 ng/l. Using a probabilistic species sensitivity distribution, the predicted no-effect concentration (PNEC) was determined from ecotoxicological data. The risk characterization ratio, calculated by dividing the PEC by PNEC values, was used to characterize the risks. The mean risk characterization ratio was predicted to be several orders of magnitude smaller than 1 (1.4 × 10-4). Therefore, this modeling effort indicates that only a very limited risk is posed by the current release level of nano-FeOX to organisms in surface waters. However, a better understanding of the hazards of nano-FeOX to the organisms in other ecosystems (such as sediment) needs to be assessed to determine the overall risk of these particles to the environment.

Comment on ET&C perspectives, November 2015-A holistic view.

In response to a recent collection of perspectives published in Environmental Toxicology and Chemistry, the authors argue that there is little value in revisiting and rehashing the well-documented issues around toxicity metrics, competing statistical paradigms, legitimacy of theoretical constructs for species sensitivity distributions, and a number of other unresolved (and perhaps unresolvable) attendant statistical issues that have occupied journal space for more than 30 yr. This is not to say that these matters are unimportant-they are; however, the discussion on these topics is mature, with very few new insights being offered. To move forward on some of these seemingly intractable issues, the authors suggest the ecotoxicological community would be better served by the formation of a subdiscipline of "statistical ecotoxicology," where professional statisticians and ecotoxicologists work in unison. As it currently stands, statistical developments in ecotoxicology are not necessarily undertaken or peer-reviewed by professional statisticians, a situation that has no doubt contributed to the lack of real progress on important recommendations such as the phasing out of no-observed-effect concentrations. Environ Toxicol Chem 2016;35:1337-1339. © 2016 SETAC.

Development and application of the SSD approach in scientific case studies for ecological risk assessment.

Species sensitivity distributions (SSDs) are used in ecological risk assessment for extrapolation of the results of toxicity tests with single species to a toxicity threshold considered protective of ecosystem structure and functioning. The attention to and importance of the SSD approach has increased in scientific and regulatory communities since the 1990s. Discussion and criticism have been triggered on the concept of the approach as well as its technical aspects (e.g., distribution type, number of toxicity endpoints). Various questions remain unanswered, especially with regard to different endpoints, statistical methods, and protectiveness of threshold levels, for example. In the present literature review (covering the period 2002-2013), case studies are explored in which the SSD approach was applied, as well as how endpoint types, species choice, and data availability affect SSDs. How statistical methods may be used to construct reliable SSDs and whether the lower 5th percentile hazard concentrations (HC5s) from a generic SSD can be protective for a specific local community are also investigated. It is shown that estimated protective concentrations were determined by taxonomic groups rather than the statistical method used to construct the distribution. Based on comparisons between semifield and laboratory-based SSDs, the output from a laboratory SSD was protective of semifield communities in the majority of studies. Environ Toxicol Chem 2016;35:2149-2161. © 2016 SETAC.

Don't be fooled-A no-observed-effect concentration is no substitute for a poor concentration-response experiment.

Renowned mathematician and science historian Jacob Bronowski once defined science as "the acceptance of what works and the rejection of what does not" and noted "that needs more courage than we might think." Such would also seem to be the case with no-observed-effect concentrations (NOECs) and no-observed-effect levels in ecotoxicology. Compelling arguments were advanced more than a quarter of a century ago as to why the use of a model to describe the concentration-response relationship was preferable to an isolated metric, with the NOEC singled out as a particularly poor toxicity measure. In the ensuing years numerous articles critical of the NOEC have been written, with some calling for an outright ban on its use. More recently, arguments have been made for the retention of NOECs, with supporters suggesting that this metric is particularly useful in situations where the concentration-response relationship is weak or nonexistent. In addition, it has been claimed that there are situations in ecotoxicology where suitable models are simply not available. These arguments are not correct, and they also have impeded the decades-overdue incorporation of numerous recommendations based on research that NOECs should no longer be used. In the present study the authors counter some of the most recent claims in support of NOECs and provide new insights for 1 class of problem claimed not to be amenable to such modeling. They are confident that similar insights will be developed as further original research in this area is undertaken. Environ Toxicol Chem 2016;35:2141-2148. © 2016 SETAC.

Predicting chronic copper and nickel reproductive toxicity to Daphnia pulex-pulicaria from whole-animal metabolic profiles.

The emergence of omics approaches in environmental research has enhanced our understanding of the mechanisms underlying toxicity; however, extrapolation from molecular effects to whole-organism and population level outcomes remains a considerable challenge. Using environmentally relevant, sublethal, concentrations of two metals (Cu and Ni), both singly and in binary mixtures, we integrated data from traditional chronic, partial life-cycle toxicity testing and metabolomics to generate a statistical model that was predictive of reproductive impairment in a Daphnia pulex-pulicaria hybrid that was isolated from an historically metal-stressed lake. Furthermore, we determined that the metabolic profiles of organisms exposed in a separate acute assay were also predictive of impaired reproduction following metal exposure. Thus we were able to directly associate molecular profiles to a key population response - reproduction, a key step towards improving environmental risk assessment and management.

Sensitivity analysis, dominant factors, and robustness of the ECETOC TRA v3, Stoffenmanager 4.5, and ART 1.5 occupational exposure models.

Occupational exposure modeling is widely used in the context of the E.U. regulation on the registration, evaluation, authorization, and restriction of chemicals (REACH). First tier tools, such as European Centre for Ecotoxicology and TOxicology of Chemicals (ECETOC) targeted risk assessment (TRA) or Stoffenmanager, are used to screen a wide range of substances. Those of concern are investigated further using second tier tools, e.g., Advanced REACH Tool (ART). Local sensitivity analysis (SA) methods are used here to determine dominant factors for three models commonly used within the REACH framework: ECETOC TRA v3, Stoffenmanager 4.5, and ART 1.5. Based on the results of the SA, the robustness of the models is assessed. For ECETOC, the process category (PROC) is the most important factor. A failure to identify the correct PROC has severe consequences for the exposure estimate. Stoffenmanager is the most balanced model and decision making uncertainties in one modifying factor are less severe in Stoffenmanager. ART requires a careful evaluation of the decisions in the source compartment since it constitutes ∼75% of the total exposure range, which corresponds to an exposure estimate of 20-22 orders of magnitude. Our results indicate that there is a trade off between accuracy and precision of the models. Previous studies suggested that ART may lead to more accurate results in well-documented exposure situations. However, the choice of the adequate model should ultimately be determined by the quality of the available exposure data: if the practitioner is uncertain concerning two or more decisions in the entry parameters, Stoffenmanager may be more robust than ART.

International scientists' priorities for research on pharmaceutical and personal care products in the environment.

Pharmaceuticals and personal care products (PPCPs) are widely discharged into the environment via diverse pathways. The effects of PPCPs in the environment have potentially important human and ecosystem health implications, so credible, salient, and legitimate scientific evidence is needed to inform regulatory and policy responses that address potential risks. A recent "big questions" exercise with participants largely from North America identified 22 important research questions around the risks of PPCP in the environment that would help address the most pressing knowledge gaps over the next decade. To expand that analysis, we developed a survey that was completed by 535 environmental scientists from 57 countries, of whom 49% identified environmental or analytical chemistry as their primary disciplinary background. They ranked the 22 original research questions and submitted 171 additional candidate research questions they felt were also of high priority. Of the original questions, the 3 perceived to be of highest importance related to: 1) the effects of long-term exposure to low concentrations of PPCP mixtures on nontarget organisms, 2) effluent treatment methods that can reduce the effects of PPCPs in the environment while not increasing the toxicity of whole effluents, and 3) the assessment of the environmental risks of metabolites and environmental transformation products of PPCPs. A question regarding the role of cultural perspectives in PPCP risk assessment was ranked as the lowest priority. There were significant differences in research orientation between scientists who completed English and Chinese language versions of the survey. We found that the Chinese respondents were strongly orientated to issues of managing risk profiles, effluent treatment, residue bioavailability, and regional assessment. Among English language respondents, further differences in research orientation were associated with respondents' level of consistency when ranking the survey's 15 comparisons. There was increasing emphasis on the role of various other stressors relative to PPCPs and on risk prioritization as internal decision making consistency increased. Respondents' consistency in their ranking choices was significantly and positively correlated with SETAC membership, authors' number of publications, and longer survey completion times. Our research highlighted international scientists' research priorities and should help inform decisions about the type of hazard and risk-based research needed to best inform decisions regarding PPCPs in the environment. Disciplinary training of a scientist or engineer appears to strongly influence preferences for research priorities to understand PPCPs in the environment. Selection of participants and the depth and breadth of research prioritization efforts thus have potential effects on the outcomes of research prioritization exercises. Further elucidation of how patterns of research priority vary between academic and government scientists and between scientists and other government and stakeholders would be useful in the future and provide information that helps focus scientific effort on socially relevant challenges relating to PPCPs in the environment. It also suggests the potential for future collaborative research between industry, government, and academia on environmental contaminants beyond PPCPs.

Selecting the best design for nonstandard toxicology experiments.

Although many experiments in environmental toxicology use standard statistical experimental designs, there are situations that arise where no such standard design is natural or applicable because of logistical constraints. For example, the layout of a laboratory may suggest that each shelf serve as a block, with the number of experimental units per shelf either greater than or less than the number of treatments in a way that precludes the use of a typical block design. In such cases, an effective and powerful alternative is to employ optimal experimental design principles, a strategy that produces designs with precise statistical estimates. Here, a D-optimal design was generated for an experiment in environmental toxicology that has 2 factors, 16 treatments, and constraints similar to those described above. After initial consideration of a randomized complete block design and an intuitive cyclic design, it was decided to compare a D-optimal design and a slightly more complicated version of the cyclic design. Simulations were conducted generating random responses under a variety of scenarios that reflect conditions motivated by a similar toxicology study, and the designs were evaluated via D-efficiency as well as by a power analysis. The cyclic design performed well compared to the D-optimal design.

NOEC and LOEC as merely concessive expedients: two unambiguous alternatives and some criteria to maximize the efficiency of dose-response experimental designs.

NOEC and LOEC (no and lowest observed effect concentrations, respectively) are toxicological concepts derived from analysis of variance (ANOVA), a not very sensitive method that produces ambiguous results and does not provide confidence intervals (CI) of its estimates. For a long time, despite the abundant criticism that such concepts have raised, the field of the ecotoxicology is reticent to abandon them (two possible reasons will be discussed), adducing the difficulty of clear alternatives. However, this work proves that a debugged dose-response (DR) modeling, through explicit algebraic equations, enables two simple options to accurately calculate the CI of substantially lower doses than NOEC. Both ANOVA and DR analyses are affected by the experimental error, response profile, number of observations and experimental design. The study of these effects--analytically complex and experimentally unfeasible--was carried out using systematic simulations with realistic data, including different error levels. Results revealed the weakness of NOEC and LOEC notions, confirmed the feasibility of the proposed alternatives and allowed to discuss the--often violated--conditions that minimize the CI of the parametric estimates from DR assays. In addition, a table was developed providing the experimental design that minimizes the parametric CI for a given set of working conditions. This makes possible to reduce the experimental effort and to avoid the inconclusive results that are frequently obtained from intuitive experimental plans.

Statistical reporting deficiencies in environmental toxicology.

Null hypothesis significance testing is one of the most widely used forms of statistical testing in environmental toxicology. In this short communication, the authors show that the reporting of statistical information when using null hypothesis significance testing is frequently inadequate in environmental toxicology research. The authors demonstrate this by analyzing the statistical information reported for papers employing t tests or analyses of variance in the Environmental Toxicology section of Environmental Toxicology and Chemistry in 2010, which comprised 68% of papers published by this journal in that year. Of these papers, 60% fail to report exact p values, 85% fail to provide degrees of freedom, and 90% fail to report critical effect sizes. Statistical power was reported in only <2% of the published papers. The insufficient provision of statistical information makes interpretation of study results by reviewers and readers difficult. Consistently reporting exact p values with degrees of freedom, considering and explicitly stating biologically relevant critical effect sizes, and reporting statistical power associated with nonsignificant results would be easy to implement and would promote scientific progress in environmental toxicology through increased statistical transparency.