Summaries Published in the April 2019 Issue
Timothy J. Canfield, USEPA; Christine Lehman, The DowDuPont Specialty Products; and Valery Forbes, University of Minnesota
There continue to be growing concerns about conflicts of interest and bias in chemical risk assessment. It has become widely recognized that all sectors relevant to chemical risk assessment – academia, government, business and NGOs – are each guilty of biases of different kinds. As scientists, we like to believe that data and facts guide our conclusions and keep us objective. However, as humans we have values, beliefs, morals and convictions that lead to biases – both explicit and implicit.
Effectively addressing bias will require action by individual researchers, funding agencies, journals, and public and private institutions. It is necessary to incorporate perspectives from all these groups. The goal of this session was not only to encourage debate on this important topic but to identify practical steps that can be adopted to minimize bias and increase confidence in the science underpinning chemical risk assessment.
This session built on previous SETAC North America sessions held during the last few years that have raised questions about the causes and kinds of bias, the pervasiveness of normative science, and the need for multi-sector collaboration to address these issues. The session presented case studies to demonstrate practical steps for addressing bias and identified elements of best practice.
The session covered views from researchers across academia, business and government.
Christine Lehman, DowDuPont Specialty Products, introduced the session by giving an overview of previous sessions from SETAC North America annual meetings in Salt Lake City and Minneapolis – Bias in Chemical Risk Assessment, Biases and Conflicts of Interest in Ecotoxicology, and Conflicts of Interest and Normative Science – and what these sessions recognized or accomplished. These sessions recognized that sources of bias include equipment or method bias, human bias in the forms of conflicts of interest, and normative bias. Conclusions from the previous sessions were that:
- Bias exists in all sectors of science
- Normative science is an issue in our field but has many drivers
- Bias can be conscious or subconscious
While value judgements are a part of science, researchers should be very transparent where these judgements are applied to the data.
Lehman stressed that we should all have a common goal to produce unbiased, informative and relevant data that will be used by other scientists and colleagues because they trust our scientific integrity and ability to produce quality data that is held in high regard regardless of affiliation. As this first talk was used to set up the rest of the session, Lehman concluded her presentation by outlining what the goals of the session were for those in attendance. She concluded her talk by making the call to continue to discuss and examine what types of bias may exist in our science and encourage the production of practical strategies to combat these issues in a tripartite framework that includes all sectors of SETAC.
Doug Fort, Fort Environmental Labs, focused his talk on looking at how study design and measurement endpoints can bias the interpretation of results. He presented the premise that modern-day toxicological bioassays use technically sophisticated measurement endpoints to support the interpretation of the results produced. These selected measures are heavily weighted to allow the rejection of the null hypothesis in support of the research hypothesis without the possibility of fully understanding the alternatives. Fort showed two case studies to make his points. One focused on behavioral assessments of test organisms exposed to a chemical and demonstrated the critical importance of having adequate assessment approaches of behavior. The second study focused on a reproductive toxicity assay and showed how inclusion of additional endpoints assisted in clarifying the results and conclusions derived from the study data. He presented several conclusions from his work:
- Interpretation bias can occur when measurement endpoint biases causes mis-interpretation of results
- The assessment of overt toxicity is traditionally considered secondary to specific measurement endpoints
- Incorporating behavioral endpoints in ecotoxicological studies is important to determine if and how behavior influences results
- Target system bias (i.e., chemicals can target reproductive organs but not necessarily be endocrine active) can lead to mis-interpretation of data
- Better directives and guidance for the assessment of overt toxicity in focused ecotoxicological studies are needed
Jill Awkerman, U.S. Environmental Protection Agency, started out her talk by asking the question, “How do you know when bias versus uncertainty get in the way of making a decision?” She discussed how science has several different lenses based on training, experience and discipline, which will influence your potential interpretation of the data. Additionally, the endpoints prioritized by different sectors or nations may be different in risk assessments. She discussed the types of uncertainty and bias typically seen in risk assessments and provided a figure that demonstrated the tradeoffs among costs and benefits of reducing uncertainty. Ackerman provided several standard approaches to addressing uncertainly, which included Bayesian belief networks, weight-of-evidence approaches, sensitivity analysis, scenario analysis, and decision analysis. She concluded with recommendations that would help manage bias and uncertainty:
- Interdisciplinary or inter-sector perspectives can minimize bias, reinforce assessments, and provide checks and balances
- Clearly defining protocols, assumptions and caveats ensures effective communication and evaluation of uncertainty
- There is a need for additional research to minimize epistemic uncertainty
Bottom line: We need more research, strong collaborations and clear communication to better manage bias and uncertainty.
Christopher Borgert, University of Florida, spoke on reducing bias through multi-sector collaborations. He stressed the point that scientific bias is different than personal bias. Whereas, scientific bias is related to how the research is conducted, interpreted and characterized, personal bias typically is due to financial, political, sociological, ideological or interpersonal biases. He noted that these personal biases can lead to scientific bias if not recognized and checked. Borgert stressed that science is about measurements and not about the scientists themselves. The scientific method, if followed, ensures that the focus is on the work regardless of who is doing the work. He advocated for the use of multisector collaboration when possible to help reduce the potential occurrence of bias in the science being conducted. To make this collaboration effective, he concluded his talk by proposing several steps that will help to make this process successful:
- Identify the motives of the participants
- Identify potentially conflicting goals
- Identify the types of outcomes that will benefit each sector
- Identify and agree on how disagreements will be resolved
- Identify how and when results will be communicated
- Identify ways to preserve the agreement irrespective of the outcome
Mark Hanson, University of Manitoba, presented information exploring the question of whether reporting significant effects of pesticides translates into greater citations in the future for researchers. He pointed out that there are several ways that bias can be introduced into a study depending on what a researcher may expect to find as expressed in their hypothesis. This bias can show up in several ways and may influence what data may or may not be selected, how they may be interpreted, or what statistical analyses may be selected to evaluate the data. He presented references that demonstrated that publication bias is well documented in the medical field. Through the work he presented, he showed that in the pesticide field there are more citations of those studies that show effects than no effects. Hansen found that higher-tier journals favor studies that report effects as opposed to those that report no effects. He presented some recommendations to combat this bias:
- Authors should take the time to examine the body of knowledge around the pesticide they are studying and evaluate the strength of the methods and not simply the results
- Peer reviewers should examine whether the authors acknowledge studies that don’t support their observations
- Editors should make it clear to authors that well-constructed and conducted studies which present no effects data are encouraged, and to the reviewers that no effects data are not a reason to reject a paper
Anne LeHuray, Chemical Management Associates, LLC, focused her talk on bias and issues arising from the governmental sector and how to gather information to address these issues. She laid the foundation for her talk by asking the question “What happens if the government gets the science wrong?” LeHuray proceeded to describe the various avenues to try and gather information on science from the government and potential road blocks that are frequently encountered when pursuing this information. She concluded that although the government purports to be open and transparent with their scientific information, if there is a question about the information and conclusions, gaining access to the data is often difficult.
William Goodfellow, Exponent, Inc., spoke on chemical risk assessments that focus on field investigations and how to identify and manage potential bias. He pointed out that emerging contaminants often come with heightened attention to evaluate the hazards and risks, which many times places pressures from regulatory agencies and other potential funding organizations on researchers to prioritize these studies. There are often times increased funding and publishing opportunities that draw researchers in but can also introduce potential elements of bias from these pressures. He cautions that researchers need to be mindful of the potential for bias creeping into their work in how they design studies, gather data, interpret the meaning of the data into results, and potentially overstate the impact or importance of the results. He cautions researchers not to succumb to the temptation of creating glitzy soundbites regarding their studies or developing information materials for media releases that are not supported by the results. He summarized his talk by reminding the audience of the need to ensure that any compromises that had to be made to conduct field investigations, such as pooling of data, selection of sampling sites, or assessment of impacts, do not provide unnecessary bias. In cases where there may be some bias, it is critical to understand how the biases are impacting the conclusions. He strongly cautioned that investigators need to minimize the predetermination of the cause of an observed effect without doing a comprehensive analysis. Goodfellow concluded by succinctly stating, “Don’t take shortcuts.”
Cheryl Murphy, Michigan State University, closed out the session speaking about the different forms of bias in academic research and giving suggestions for mitigating these biases. She talked about how diversity and the incorporation of under-represented groups can produce better research. She pointed out that academic institutions show bias in several ways: Institutional bias, social-implicit bias, study bias and data bias. Institutional bias comes about because of the way academic institutions reward their researchers. These rewards are presented in tenure and promotion approaches and in how researchers are accorded scholarly status by publishing in high-impact journals where there is a bias for showing significant effect results. This often garners media attention and brings recognition to the institution, which in in turn may lead to more opportunities in the form of grants, invitations to serve on review panels and special appointments that all benefit the institution as well. She showed that social-implicit bias can be due to the researcher’s age, social class, gender or sexual orientation, and race or ethnicity. While attempts have been made to combat these bias issues, Murphy stated that they are still a problem in the science review process and in academic institutions. She proposed that academic institutions should redirect their metrics of success and re-evaluate what a positive or desired impact should be. Murphy advocated that higher education institutions should be judged on their capacity to educate conscientious human beings who put their values in practice in positive and meaningful ways, and should develop and advance learning communities that are equitable, safe, transparent and trusting and that have true inclusiveness. She proposed moving away from traditional measures of success like grants and numbers of papers in high-impact journals to a model that focuses on scholarly teaching practices, outreach and extension, collaboration, student mentorship and community impact in the form of policy changes or adoption of recommended practices.
Conclusions from the Session
The attendance at this session was very high (between 450 and 500 participants) with many questions and opinions expressed through the lively and interactive discussion periods. As with previous sessions focused on the area of potential bias in science, the comments from the audience further reinforced that there is a strong, shared concern by many in SETAC that bias is very much an issue and that there is interest across all sectors to address it regardless of when and where it arises. While this is the third time that a session focused on bias and its many forms has been included in a SETAC North America annual meeting, there remains a clear desire to see more sessions focused on this topic and on how SETAC is identifying and reducing bias – in all its manifestations – to better our science.
Disclaimer: This document has been reviewed by the U.S. Environmental Protection Agency, Office of Research and Development, and approved for publication. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of the U.S. Environmental Protection Agency.
Donna Davila, AbbVie, Inc.; Jane Staveley, Exponent, Inc.; and Holly Zahner, U.S. Food and Drug Administration-Center for Veterinary Medicine
An assessment of the potential environmental impact of pharmaceuticals is required as part of the approval or authorization process for new drug applications in the U.S., Europe and other countries. The continued application of new technologies to drug development may prompt uncertainty in terms of requirements for environmental assessments. New guidance from various jurisdictions can address some of this uncertainty, although harmonization (within and between agencies, programs and countries) remains an issue. During this session, the approaches recommended in guidance for preparing environmental assessments for drugs were discussed by regulators and industry. Proposed regulatory amendments to the Food and Drug Regulations of Canada were explained and brought forward for consideration by the attendees. Members of the pharmaceutical industry recounted their experiences in the preparation of tailored risk assessments for veterinary or human pharmaceuticals and described the uncertainty that they encountered; they also provided suggestions for harmonization across sectors. This session highlighted the importance of continued cooperative dialogue among regulators and stakeholders in the development and refinement of guidelines. Early consultation with regulatory agencies during the development of environmental risk assessments was also brought forward as an essential component of navigating the regulatory environment, particularly when novel pharmaceuticals require a different approach that is not addressed in current guidance. Concerns regarding endocrine disruption are a focus of recent and upcoming regulatory guidance documents, and as a result, this session also included presentations on fish studies used for detection of endocrine disruptors.
There were two presentations on “What’s new at the U.S. Food and Drug Administration?” in this session. Holly Zahner of the U.S. Food and Drug Administration (USFDA) Center for Veterinary Medicine (CVM) focused on the recommendations in the guidance for industry (GFI) on how to conduct an environmental assessment (EA) for animal drugs. She noted that while companies complete the studies needed for the EA, the USFDA is responsible for the scope and content of the EA; therefore, she emphasized the importance of early communication between each company and the USFDA. Zahner also discussed the evaluation of a new animal drug that falls outside the scope of the GFI (e.g., watershed-scale modeling of endocrine disrupting compounds, potential risk for secondary poisoning through dietary exposure, direct discharge from aquaculture facility and the potential risks of animals with intentionally altered genomic DNA), and presented a comparison of approaches used by USFDA CVM and USFDA Center for Drug Evaluation and Research (CDER).
The second presentation of “What’s new at the U.S. Food and Drug Administration?” focused on human drugs and was given by James Laurenson. It included:
- Discussion of the recent increases in drug applications and thus the need for more environmental assessments and screening or weight-of-evidence approaches
- Implementation of the GFI, “Environmental Assessment: Questions and Answers Regarding Drugs with Estrogenic, Androgenic or Thyroid Activity,” which addresses the need for full EAs or supported claims of categorical exclusion for drugs within its scope
- The recent focus of USFDA on opioid drugs and the potential for changes in aquatic risks from USFDA’s promotion of responsible drug disposal, such as use of the “Flush List”
Jason Weeks of ERA and IEH Consulting, Ltd. discussed the European perspective on the challenges and improvements in assessing environmental risks of veterinary medicines. The centralized registration of veterinary medicines is the responsibility of the Committee for Medicinal Products for Veterinary Use (CVMP) of European Medicines Agency. The CVMP also receives advice from the Environmental Risk Assessment Working Party (ERAWP). The ERAWP is working to update existing guidance to address new topics such as pharmaceuticals in groundwater, assessments of persistent, bioaccumulative and toxic (PBT) chemicals, aquaculture, and anti-microbial resistance. The development of reflection papers, such as the “Reflection paper on antimicrobial resistance in the environment: Considerations for current and future risk assessment of veterinary medicinal products” by the ERAWP was also discussed.
Kyle Roush of Texas Christian University discussed the effect of gonadosomatic index (GSI) of female fathead minnows on the results of screening assays for the detection of androgens and anti-estrogens. Exposure to 17-beta trenbolone had differing effects on estrogen-responsive gene expression (such as hepatic vitellogenin) in females with a higher GSI versus female fish with a lower GSI. Use of fish with appropriate GSI classification was concluded to be an important component of tests to measure potential endocrine-disrupting properties of pharmaceuticals.
Julie Chateauvert of Health Canada discussed proposed regulatory amendments to Health Canada’s Food and Drug Regulations pertaining to environmental risk assessments of medicinal ingredients in drugs. These amendments have been designed to align with the drug approval process stipulated by the Food and Drugs Act and to harmonize with international jurisdictions. Data requirements will apply to active ingredients in human and veterinary drugs, as well as cutting-edge technologies such as human cell-based therapies.
Tamara Lunsman of Gradient presented their experience conducting environmental risk assessments for new human drugs under existing regulatory frameworks in the EU and U.S., and provided some suggestions for improvements such as harmonization of requirements across regulatory frameworks. Lunsman illustrated the need for additional guidance for drugs with unique physical or chemical properties, a unique mode of action, narrow indications of use or naturally occurring substances using previously prepared environmental assessments for an endocrine active and an anti-infective drug.
The environmental risk assessment that was submitted to the USFDA’s CDER for an endocrine-targeting compound was presented by Donna Davila of AbbVie. She discussed the specifics of the Medaka Extended One Generation Reproduction Test (MEOGRT) study, including peer review of gonadal pathology findings and evaluation of findings in light of the apical endpoints within the study.
Henry Krueger of EAG Laboratories discussed the integration of response variables into a “strength of evidence assessment” to distinguish between endocrine activity of a chemical and endocrine disruption, where adverse effects occur. He underscored the importance of evaluating statistical results in light of biological relevance, with a weighting of variables in considering their importance in an overall pattern of responses. Signature patterns of responses of known ER-agonists, AR-agonists or AR-antagonists can help avoid erroneous conclusions when evaluating endocrine disruption in fish studies.
The posters associated with this session echoed the need for harmonization of global regulations pertaining to pharmaceuticals in the environment and the need to understand the effects of pharmaceuticals on fish under differing environmental conditions. The poster by Norman Forsberg, Arcadis US, showed how regional approaches (EU and U.S.) for the derivation and application of Predicted No-Effect Concentrations (PNECs) can affect risk evaluations of active pharmaceutical ingredients (APIs) in wastewater discharge. As a demonstration, PNECs were calculated for a variety of APIs using the USEPA’s or the EU’s Water Framework Directive methods, and an analysis was performed to identify sensitive calculation input parameters. Forsberg also showed the effect of using U.S.- and EU-based approaches to evaluate the dilution capacity of receiving waters on the calculation of acceptable daily water discharges, and he provided recommendations to help reduce uncertainty in the risk assessment of API discharges. The poster by Gavin Saari, Baylor University, focused on the toxicity of a common pharmaceutical, diltiazem, under conditions of reduced dissolved oxygen concentrations typically found within urban estuaries of the Gulf of Mexico. The uptake of diltiazem in adult fish was doubled under conditions of decreased dissolved oxygen relative to normoxic conditions, with concomitant effects on physiological responses of affected fish. Saari concluded that the observed physiological effects together with measurements of tissue concentrations of the drug help inform the identification of adverse outcomes and demonstrate the utility of using model pharmaceuticals for both basic and applied research.
Conclusions from the Session
During this session, it was evident that the regulatory landscape has continued to transform in response to the experience of regulators and stakeholders. There is recognition of the need for harmonization of regulations as evidenced by the goals of draft regulations in Canada. Harmonization was also evident when studies requested by the USFDA for a tailored environmental program aligned largely with programs typically developed for European filings (EMA). The speakers discussed areas in which gaps are present and suggested that continued effort should be made to harmonize the environmental risk assessment process across sectors. This goal makes sense in a global economy and will contribute to resource stewardship across companies and regulatory agencies. Early consultation with regulatory agencies during development of tailored risk assessments was also brought forward as essential to a more efficient drug development program. Not only are the regulations continuing to evolve, but the studies used to measure environmental effects of pharmaceuticals continue to be refined. In this session, progress in the interpretation of data from fish studies used to measure endocrine disruption was brought forward.
Jone Corrales, Florida International University, and Marlo Sellin Jeffries and Leah Thornton Hampton, Texas Christian University
Immunotoxicology is gaining traction at SETAC! The SETAC North America 39th Annual Meeting in Sacramento featured the second consecutive platform session focused on immunotoxicology. The mean number of attendees per talk increased from 56 in Minneapolis to 69 in Sacramento, demonstrating a growing interest in immunotoxicity among the SETAC community. Presenters from Canada, Japan, Europe and the United States shared their work aimed at identifying adverse immunotoxic effects, developing novel approaches for immunoxicity and addressing existing challenges in the field. The presentations comprised a broad range of projects from in vitro mammalian studies to in vivo wild-caught fish studies, as well as reviews focused on various aspects of fish immunotoxicology. Jointly, the talks underlined the complexity of the immune system and promoted discussion of critical challenges facing the field. Key highlights from each platform presentation are as follows:
- James Stafford, University of Alberta, presented the results of an in vitro study aimed at evaluating the immunotoxic potential of whole oil sands process-affected water. This presentation provided an approach for evaluating the immunotoxicity of complex mixtures using a mammalian immune cell line.
- Maria Rodgers, University of Southern Mississippi, discussed immunocompetence in three commercially and recreationally important fish species following exposure to oil and the bacterial pathogen Vibrio anguillarum. The research presented illustrated differences in immunotoxicity not only between species but also across various life stages within the same species.
- Natacha Hogan, University of Saskatchewan, talked about developing an amphibian model to explore the mechanisms of immunotoxicity using benzo[a]pyrene (BaP), a model polycyclic aromatic hydrocarbon (PAH). This presentation included a discussion of the challenges associated with immunotoxicity research ranging from the complexity of the immune system to the difficulties associated with linking alterations in immune parameters to disease risk in wild populations.
- Helmut Segner, University of Bern, showed evidence of an immunoregulatory function of the transcription factor aryl hydrocarbon receptor (AhR) in teleost fish, demonstrating the potential for contaminants to alter immunity through unexpected and novel mechanisms.
- Jessica Leet, U.S. Geological Survey, described the results of a longer-term pond study aimed at uncovering the immuntoxicity of environmental estrogens and pesticides in largemouth bass. Leet noted the seasonal differences in baseline immune parameters and in the response of such parameters to chemical exposure, demonstrating the influence of abiotic factors on immunity and immunotoxicity.
- Marlo Jeffries, Texas Christian University, discussed research showing that sex and reproductive status influence baseline immune parameters and immune responses in fathead minnows, highlighting the need to account for these biotic factors in immunotoxicity studies.
- Kei Nakayama, Ehime University, provided an overview of the pitfalls associated with designing immunotoxicity studies, including the identification of appropriate host-pathogen systems and the failure to take into account the effects of contaminants on the stimulated immune system. He also stressed the need to develop standardized methods and minimum reporting guidelines for immunotoxicity assessments in fish.
- Charlie Rice, Clemson University, discussed the potential for using genomics as a tool for identifying molecular initiating events to be used in the development of adverse outcome pathways predictive of immunotoxicity. He also encouraged the aquatic immunotoxicology community to collaborate with mammalian immunotoxicologists, comparative immunologists and computation biologists to advance the field.
Following the session, a group of about 15 environmental scientists, led by the session chairs, gathered for an inaugural Immunotoxicity Working Group meeting, where they discussed lessons learned from the platform sessions and laid out a plan to bring the field of immunotoxicology into the spotlight at SETAC. Considering the discussions emerging during this meeting and in response to the growth of this year’s immunotoxicity session, we hope to capitalize on the momentum being gained by holding the third consecutive session focused on immunotoxicology at the SETAC North America 40th Annual Meeting, which will be held from 3–7 November in Toronto, Canada. In addition, we will hold our 2nd Annual Immunotoxicity Working Group meeting in Toronto. We call for all those working on projects that investigate immune function in model and non-model organisms using controlled laboratory approaches or wildlife studies to submit abstracts to the immunotoxicology session. Meanwhile, if you are interested in learning more about the goals of the Immunotoxicity Working Group, please contact Jone Corrales or Marlo Jeffries.
Dawn Perkins, University of Wisconsin-Madison; Amanda Foss, Green Water Laboratories; Jim Lazorchak, USEPA; Bryan Brooks, Baylor University; Meredith Howard, Southern California Coastal Water Research Project; and Joel Allen, USEPA
Harmful algal blooms (HABs) in freshwater and marine systems are defined as an overgrowth of photosynthetic organisms (eukaryote and prokaryote) that have the potential to cause negative health, ecological or economic impacts. These negative impacts are caused by mechanisms that include, but are not limited to, toxin production, decreases in light attenuation and dissolved oxygen, diurnal swings in pH, offensive tastes and odors, and impaired visual aesthetics.
The purpose of this session was to highlight and exchange information on chemical, biological, instrumental and sensing methods for the detection of HAB-related organisms and their associated toxins. Such approaches discussed included unique sample collection techniques (e.g., SPATT), remote sensing of HABs, modeling to anticipate toxin occurrence and HPLC-MS techniques for toxin analysis. The session had more than 75 attendees and featured eight platform and five poster presentations.
The session included research from the U.S. Geological Survey (USGS), U.S. Environmental Protection Agency (USEPA), the Southern California Coastal Water Research Project, as well as researchers from Baylor University and University of California Davis.
- Meredith Howard, Southern California Coastal Water Research Project, presented water sampling strategies used to monitor HABs in California, including combining Solid Phase Adsorption Toxin Tracking (SPATT) along with grab and benthic mat sample collection.
- Jennifer Graham, USGS, discussed the use of real-time, water-quality data to estimate cyanotoxin occurrence. Linear and logistic regression models to estimate probability of microcystin occurrence above relevant thresholds were developed and evaluated for lake, reservoir and river sites in Kansas and Ohio; some of these models performed relatively well over time and others performed relatively poorly. Graham presented four case studies and discussed the strengths and limitations of using these near-real-time notification tools.
- Guy Foster, USGS, addressed how extreme spatiotemporal variability in cyanobacteria and microcystin within waterbodies pose challenges to collecting representative samples. The USGS, in collaboration with the Kansas Department of Health and Environment, collected data from a combination of time-lapse cameras and fluorescence sensors in Milford Lake, Kansas. While continuous fluorescence sensors may be useful for informing public health decisions in lakes, site-specific models need to be developed, and best practices for using continuous water-quality monitors to inform management strategies need to be established.
- Samuel Haddad, Baylor University, discussed a newly developed method to detect multiple cyanotoxins in water and fish using liquid chromatography tandem mass spectrometry. Two separation techniques, hydrophilic interaction liquid chromatography and reverse phase liquid chromatography, were used to analyze water and caged fish study samples collected at a hypereutrophic impoundment.
- Joel Allen, USEPA, presented an overview and observations of cyanobacteria population and intracellular and extracellular cyanotoxin production and high-frequency (HF) monitoring data from 2015 to 2017 on a lake which has been experiencing annual HAB blooms for the past eight or so years. High-frequency data coupled with cyanotoxin analyses demonstrate the utility of HF data for tracking the cyanoHAB status of the reservoir. It is also apparent that cyanotoxin concentrations may potentially be underestimated if cyanotoxin sampling is not coupled with bloom status.
- Heath Mash, USEPA, presented findings from a study investigating the performance of passive samplers for detecting extracellular cyanotoxins in stream mesocosms, lakes and streams. Two types of passive samplers, SPATT and Large Format non-selective Passive Sampler Device (LF nsPSD), were deployed in 18 different stream mesocosms and two locations in Lake Harsha, Ohio.
- Melissa Bolotaolo, University of California, Davis, presented information on HAB monitoring efforts in the Sacramento–San Joaquin Delta that was outside the typical cyanobacteria summer bloom months in order to evaluate trends. Results from this project reveal a shift in cyanotoxin regime from prior summer bloom month measurements and suggests benthic environments in the Delta are at risk for microcystin exposure.
- Raegyn Taylor, Baylor University, discussed an alternative, top-down approach to identifying HAB toxin mixtures. Suspect screening of liquid chromatography mass spectrometry (LC-MS) data found three prymnesin-type compounds and further analysis showed a positive relationship between the LC-MS peak area of prymnesin-aglycone and acute toxicity for each algal culture. However, statistical methods and relative difference plots highlighted additional compounds (i.e., peaks) that strongly co-varied with toxicity, suggesting other potential toxins and demonstrating the benefits of using a top-down approach with non-target analysis to interpret toxicity of complex mixtures.
Poster presentations comprised both freshwater and marine HAB issues, including the evaluation of microcystins following upland placement of dredged material from HAB-impacted waters, sediment-bound microcystin-LR detections in California’s Stream Pollution Program, domoic acid in tissues of field collected crabs from California and Oregon, toxicity and bioaccumulation of anatoxin-a, and microcystin-LR oral toxicity in mallard ducks.
Conclusions from the Session
The session highlighted the complexity of HAB monitoring and provided audience members information on how different strategies have worked (or not worked) in specific environments.
Themes repeated throughout the session were the need to continue to improve the tools used to detect HABs, the need for optimized and verified analytical standard solutions, and the importance of real-time monitoring to protect public and environmental health. As approaches to address HABs expand and improve, it is necessary to continue fostering collaboration and the dissemination of knowledge through these platforms.
Another session on HABs is being proposed for the SETAC North America 40th Annual Meeting, which will be held from 3–7 November in Toronto, Canada.
Disclaimer: This document has been reviewed by the U.S. Environmental Protection Agency, Office of Research and Development, and approved for publication. The views expressed in this paper are those of the authors and do not necessarily reflect the views or policies of the U.S. Environmental Protection Agency.
Sarah E. Allan, Lisa DiPinto and Mary Baker, National Oceanic and Atmospheric Administration (NOAA)
Risk assessments, emergency response decisions and damage assessments for oil and chemical spills in aquatic systems rely on measured and modeled chemical concentrations in the water column to estimate exposure and injury to aquatic organisms. Models that describe the transport and fate of oil in the water after a spill are a critical tool. These models vary in their complexity, scale, resolution, data needs, outputs and uncertainties. Understanding the range of modeling tools and their intended applications and limitations is critical for model developers and end users.
This session provided an overview of different models available for characterizing chemicals in the water column following aquatic spills and presentations about processes that are determinant in the oil transport and fate described by the models. It provided a valuable opportunity for modelers and model end users to share information about capabilities and needs.
Lisa DiPinto, NOAA Office of Response and Restoration, set the stage for the session with an overview of needs and considerations for modeling oil concentrations in the shallow, mixed layer of the water column to answer operational and scientific questions for oil spill emergency response and natural resource damage assessment. This discussion of end user needs provided context for the technical platforms detailing modeling tools and processes.
Four models of oil transport in the water column were presented. They varied in their technical parameters, capabilities and intended applications, and they are described below in approximate order of increasing complexity.
- Amy MacFadyen, NOAA Office of Response and Restoration, presented the General NOAA Operational Modeling Environment (GNOME), which is an oil transport and fate tool developed to support oil spill emergency response. This is a relatively simplified, Lagrangian-approach model aimed at answering shorter-term surface and near-surface oil transport questions for responders that can be run very quickly with a small number of inputs.
- Andrew Ernest, Research Applied Technology Education Services, Inc., presented the Water Column Exposure Modeling Toolbox, which is a Eularian-approach model used to estimate water column concentrations of petroleum hydrocarbons (PH) and polycyclic aromatic hydrocarbons (PAH) resulting from a range of transport processes. Twelve modeling tools include the ability to estimate PH exposure in fish eggs.
- Deborah French McCay, RPS, presented the Spill Impact Model Application Package (SIMAP); a comprehensive, Lagranian-approach, oil fate and exposure model that has been extensively applied and verified. This model includes different soluble, semi-soluble and insoluble oil pseudo-components and an activity-based exposure and toxicity model for fish and invertebrates.
- Steven Bartell, Cardno, presented the Chemical/Oil Spill Impact Model (COSIM), which estimates concentrations of oil chemicals over space and time as well as lethal, sublethal and indirect biological effects of oil exposure. This model characterizes oil components with similar chemistry and toxicology, estimates the toxicity of mixtures of oil constituents, and can be integrated with the Comprehensive Aquatic Systems Model (CASM) for toxic effects modeling.
Other platforms in the session focused on processes that are determinant in oil transport, fate, and effects in the surface mixing layer.
- CJ Beegle-Krause, SINTEF, presented an analysis of how the dynamics of the surface mixing layer impact oil transport and exposure after a spill, concluding that models should include a wider range of vertical transport processes beyond wave mixing.
- George Kapellos, Massachusetts Institute of Technology, presented a model for the biodegradation of oil microdroplets, which is determinant for many of the factors included in the fate and effects models, such as droplet size, chemical dissolution, and degradation or weathering.
- Yasemni Atalay, Integral Consulting, discussed the transport, fate and toxicity of polar hydrocarbon compounds that may be naturally occurring in petroleum hydrocarbons or formed during degradation or weathering and are not always included in models.
Oil spill models are fit-for-purpose tools that can help answer questions or fill in gaps in field data regarding the transport, fate and effects of oil. With a range of modeling tools available, it is critical to understand the specific questions that models are designed to answer and their spatio–temporal resolution as well as their uncertainties and limitations. Issues such as complexity, transparency and accessibility may also be important to model users. This session provided an overview of some of the most widely used models for oil transport in the water column after spills and discussion of processes that could be considered for future model development.
Amy H. Ringwood, University of NC Charlotte; Annie Jacob and Ed Johnson, National Centers for Coastal Ocean Science, NOAA National Ocean Service
Bivalves have been used worldwide as valuable bioindicators of pollution exposure and effects. Mussel Watch and similar programs have focused on using tissue contaminant concentrations to characterize the extent of ecosystem contamination and confirmed the value of marine bivalves such as mussels and oysters. Cellular biomarkers and physiological responses have been used to further characterize the potential relationships between contaminant exposures and organismal health. While most of the chemical and molecular work for the last few decades have been conducted with marine bivalves, a variety of freshwater bivalves are being developed as valuable biomonitoring organisms for freshwater ecosystems. Characterizing tissue concentrations for the extensive suite of potential contaminants in sediments and water, including emerging contaminants such as pharmaceuticals and nanoparticles, through routine monitoring can provide valuable spatial and temporal trends.
Recent data on emerging contaminants as well as metals and organic pollutants indicate that organisms are exposed to a variety of toxins that can interact in additive or synergistic manners and that can affect similar cellular and metabolic pathways. While individual chemicals may not flag as highly problematic, the combined effects can affect organismal health or pollutant effects and can be exacerbated by climate change conditions. Single contaminant or controlled mixture studies in the laboratory are important to identify key cellular targets that can be diagnostic for more general stress responses, such as energy allocation, oxidative stress and DNA damage, or more targeted responses, such as acetylcholinesterase inhibition or specific genes and proteins. And, while laboratory exposure studies are essential for facilitating mechanistic interpretations of field studies, ultimately, organismal health assessments based on sensitive, sublethal responses related to cellular biomarkers and metabolic pathways are essential for characterizing potential toxicity in complex, real-world habitats. An appreciation of species-specific differences in sensitivity is also important as different species are required for marine and freshwater habitats and when there are issues related to the availability of a suitable species.
For this session, a variety of presentations related to established and emerging tools based on cellular, genomic and metabolomic responses related to oxidative stress, DNA damage and physiological changes in both freshwater and marine bivalve species. Exciting advances using state-of-the-art approaches for characterizing sublethal toxicity were also described.
Based on extensive spatial analyses conducted in the Chesapeake Bay, numerous contaminants of emerging concern (CEC), including antibiotics, hormones and UV-filters, were detected in water, sediment and oyster tissue. Generally, CEC were ubiquitous throughout the bay. The artificial sweetener, sucralose, was used to discriminate potential inputs from wastewater treatment plants with those from animal feeding operations (Ke He, University of Maryland Baltimore County). This presentation, along with other field studies, highlighted not only the potential challenges of identifying specific compounds that are driving toxicity to local populations and habitats but also the need for complementary chemical and biological analyses.
Many of the presentations were related to recent investigations in the U.S. Great Lakes using Dreissena spp. (zebra and quagga mussels), in collaboration with the NOAA Mussel Watch program. Using gas-chromatography-tandem mass spectrometry with isotope dilution, modified DNA bases related to oxidative damage were identified (Pawel Jaruga, National Institute of Standards and Technology). This sensitive technique yielded very high-quality data indicating that DNA damage levels were higher in mussels from polluted sites in the Great Lakes. Elevated DNA damage can serve as a valuable screening tool to trigger more in-depth forensic chemical analyses.
Global gene analyses, as well as targeted gene analyses (mRNA analyses for detoxification genes – PgP, GST, AHR and HP70), were used to compare Dreissena spp. that were caged in the Niagara River and its tributaries (Rebecca Klaper and Nicklaus Neureuther, University of Wisconsin). After five weeks, evidence of greater toxicity in the tributaries compared with the river was detected; global and targeted gene expression indicated a tendency for down-regulation, and some genes were significantly correlated with tissue levels of PAHs, PCBs and select CECs. Laboratory studies supported the field studies indicating genes are a good biomarker for exposure in the midst of multiple compounds present.
Ultimately, changes in gene expression and increases in DNA damage can adversely affect various metabolic pathways. Complementary metabolomic profile analysis of Dreissena spp. from sites in Lake Erie and Lake Michigan were conducted using nuclear magnetic resonance (NMR) of whole body polar extracts (Tracey Schock and Elena Legrand, National Institute of Standards and Technology). As an emerging approach, the importance of experimental design, technical variability, measurement reproducibility and data interpretation, such as seasonal and handling variability, were discussed.
Another exciting new approach was the use of near-infrared spectroscopy for characterizing the Cellular Energy Allocation (CEA), based on energy stores (proteins, lipids, glycogen) and energy consumption (ETS) (William Maher and Jill Bartlett, University of Canberra, AU, and Mathew Purss, Pangaea Innovations Pty. Ltd.). Chronic as well as acute pollutant effects result in significant impacts on energy allocation and energy budgets. Traditional energetics approaches are labor intensive. The results of baseline studies and Cadmium exposures in five Australian marine bivalve species (three oyster species, one mussel, one cockle) indicated the broad applicability of the CEA near-infrared technique across species. The tissue processing and analyses are rapid and straight-forward, the sensitivity to pollutants was confirmed, and access to the software (Mortilver) for analyzing the outputs is available from Pangaea Innovations Pty. Ltd.
Another presentation with multiple species involved two candidate freshwater indicator species, Dreissena spp. and Corbicula fluminea (Asian clams) (Tuan Phan, University of North Carolina – Charlotte). Field studies based on both species deployed in cages in the Maumee River of Lake Erie, which is heavily contaminated with agrochemicals, especially chlorpyrifos, were conducted in collaboration with the NOAA Mussel Watch Program to evaluate the efficacy of the neurotransmitter acetylcholinesterase (AChE) as an indicator of pesticide exposure, along with biomarkers of oxidative stress (glutathione and lipid peroxidation). Laboratory studies with C. fluminea exposed to AChE were also conducted. These studies indicated severely depressed AChE in bivalves from the Maumee River compared with other Great Lakes sites (and severely depressed GSH levels); and the field and laboratory results suggested that Asian clams were not quite as sensitive as Dreissena spp.
The relative sensitivity of different species to pollutants can be affected by antioxidant and detoxification capacity. Glutathione (GSH) is widely recognized as an abundant antioxidant in all organisms, and numerous studies have demonstrated that glutathione depletion potentiates toxicity. A review of the GSH responses of a variety of bivalves (marine and freshwater) from a variety of locations throughout the world indicated significant species differences (Amy Ringwood, University of North Carolina – Charlotte). The baseline levels of GSH in marine bivalves (oysters, mussels, scallops) were very similar and were much higher than those in freshwater bivalves (Dreissena spp., Asian clams, paper pondshells), approximately 10 times higher; and Asian clams had slightly higher GSH levels than the other freshwater species.
Bivalves, marine and freshwater species continue to be very valuable bioindicator organisms for characterizing ecosystem health that integrate responses to the complex mixtures of legacy and CECs, as well as climate change effects. They are widespread and are very tolerant of caging methods, facilitating the use of resident species or more controlled caging approaches for in situ field studies. Coupling cellular, molecular and physiological biomarkers to sediment and tissue residues can provide detailed information on spatial and temporal changes in organismal health. Persistent damage to DNA, oxidative damage and reduced antioxidant capacity, adverse effects on neurotransmitters, changes in gene expression, metabolic alterations and energetic stress will affect organismal health and fitness and population sustainability. Cellular biomarker assays of caged or resident bivalve species that capitalize on emerging technologies are necessary for detecting and diagnosing potential impacts, especially chronic effects, and are especially valuable for characterizing the effectiveness of remediation actions in a timely manner.
Summaries Published in the March 2019 Issue
Carys L Mitchelmore, UMCES Chesapeake Biological Laboratory, and Iain Davies, Personal Care Products Council
The session’s goal was to bring scientists and stakeholders from academia, business and government agencies together to discuss the emerging field of research regarding the environmental risk of UV filters in aquatic ecosystems. The session chairs reached out to the primary authors that had published papers on the environmental risk of UV filters to coral ecosystems given that this was a current hot topic of concern.
Ultraviolet filters are the active ingredients in sun protection products, such as sunscreens, although they are also present in other consumer products, for example plastics, paints and apparel. There are multiple point and diffuse sources of UV filters to the aquatic environment, including swimmers, wastewater treatment plants as well as run-off from land uses due to leaching from products.
Recently, there have been concerns over the impact of UV filters to corals, leading to legislative activity in Hawaii that resulted in the banning the sale of two UV filters, namely oxybenzone and octinoxate. This action was essentially driven by a study conducted in 2016, which showed the presence of oxybenzone at one of seven sites near coral reef locations in Hawaii and toxicological impacts to coral planula and in coral cell lines exposed to oxybenzone. As an emerging field of research, there are only a few studies that have reported on the concentration of UV filters around coral reefs and even fewer studies investigating toxicological impacts in corals.
Risk assessments require both of these data sets, including knowledge of the concentration of the chemical of concern in the aquatic environment and the concentrations that cause damage to the species of interest. Therefore, the current session brought together researchers, who have investigated the occurrence, fate and effects of UV filters in the environment, with industry representatives to discuss environmental risk and the potential trade-offs with human health concerns. The session aimed to advance our knowledge of the environmental risk that such materials pose and to initiate scientifically grounded discussions.
Key points from the seven presentations included:
- Iain Davies (Personal Care Products Council) introduced the session by setting the scene and giving an overview of the state of the science and challenges regarding environmental risk assessments of UV filters in the aquatic environment and highlighting numerous data gaps.
- Given the current concern regarding the potential for negative impact of UV filters to Hawaiian coral reefs, coupled with the limited amount of data available to conduct a risk assessment, especially regarding environmental concentrations of UV filters in Hawaii, Carys Mitchelmore (UMCES, Chesapeake Biological Laboratory) presented data from a recent monitoring program measuring the concentration of 13 UV filters in seawater, sediment and coral tissues at various sites in Oahu, Hawaii. She concluded with a summary of the studies relevant in relation to environmental risk by using published toxicological thresholds.
- Broadening the discussion of consumer chemical pollutant risks to coral reefs, Michael Gonsior (UMCES, Chesapeake Biological Laboratory) followed up on the previously described Hawaii monitoring project by demonstrating the utility of a non-target chemical screening approach that revealed a diversity of surfactants and their degradation products also present in seawater, which may also be of concern to coral health.
- Very little is currently known regarding the environmental fate of UV filters, and Sara Wolfson (Rutgers University) gave an overview of some potential transformation products of UV filters arising from microbial activity in wastewater treatment plant effluents, which is relevant to their impact to aquatic organisms receiving this effluent.
- Addressing the second component of a risk assessment, new UV filter toxicity studies were presented by both Ray Banister (Mote Marine Laboratory) and Marc Leonard (L’Oréal). The first study demonstrated mortality in corals after nine days exposure to 3 mg/L oxybenzone and chronic impacts at concentrations of 0.3 mg/L. The study also highlighted changes to the host microbiome at the highest concentration. The L’Oréal study was a longer-term chronic exposure study investigating changes to photosynthetic activity in corals exposed to an array of organic UV filters, the mineral UV filter Zinc oxide (ZnO) and other aquatic contaminants, for example, herbicides (Monuron and Diuron) and antifoulant paint (Tributyltin). Photosynthesis was impacted more by ZnO and especially the other contaminants used than the other organic UV filters, which showed no adverse effects on the symbionts or animals up to their water solubility limits.
- Summarizing the session to balance the discussion between environmental risk and human health consequences, Jay Sirois (Consumer Healthcare Products Association) highlighted and summarized concerns from the industry and health professionals regarding the implications and problems with removing organic UV filters from use.
Conclusions From the Session
The environmental risks of UV filters in aquatic environments is an emerging field of research, and there are numerous gaps in knowledge critical for risk assessments. Much more data is needed regarding the environmental concentration, fate and impact of these products to aquatic organisms, especially corals. Determining the risk of UV filters to coral reefs should be placed in context with other chemical contaminants also present in these ecosystems. Data in this session highlighted that a number of chemicals known to have an effect to corals were also detected in environmental monitoring programs. The risk to corals from individual chemicals and contaminant mixtures needs further identification and prioritization of the most harmful chemical contaminants. It is essential that academics, industry, government agencies and other stakeholders work together to increase the wealth of data available regarding the environmental concerns of UV filters. If toxicological effects are observed at environmentally relevant concentrations, then a UV filter may pose an unacceptable environmental risk, which then needs to be managed. We plan to continue the discussions from this session in 2019 at the SETAC Europe and North America annual meetings.
Lisa S. Ortego and Steven L. Levine, Bayer CropScience, and Joel Robert Coats, Iowa State University
The interest and use of biological materials in crop production is increasing globally at a rapid pace. Part of the interest is that these technologies are viewed as safer alternatives to conventional chemicals. The action of some of these materials is very pest specific; a great advantage for safety. While establishing the safety of these materials is as important as for conventional chemicals, there are important distinctions between them. For example, a rich literature base usually exists for many micro-organisms, which can be relied upon for important information and can reduce the amount of testing needed. Also, unlike conventional chemicals, micro-organisms must be evaluated for their pathogenic potential, not only in people but in ecological receptors, as well.
The United States was the first country to develop biopesticide-specific testing guidelines. Therefore, many countries look to the U.S. when developing regulations for biopesticides and microbials. Several countries have recently proposed or approved specific regulations for biological materials, including China, Brazil and Argentina. However, other regulatory authorities are making slow progress to develop a framework to address testing and risk assessment requirements for biopesticide and microbial products. In part, the reason for this is that some products can be technically challenging to test in ecotoxicity assays, and estimating environmental concentrations can be complex. Nonetheless, there are opportunities to harmonize current testing and assessment approaches so that safety requirements no longer vary greatly between jurisdictions.
We organized a biopesticide session for the SETAC North America 39th Annual Meeting. Presentations covered a variety of topics, including understanding soil and plant microbiota, U.S. and international regulations, considerations for when to develop testing and assessment strategies based on problem formulation, examples of testing and ecological assessment for specific products, challenges in regulatory testing and assessment, and recommendations for the future regulation of biopesticides.
There were eight platform presentations and two poster presentations. Following is a list of the presenters, titles and key messages from their presentations.
- Steven Bradbury (Iowa State University) – Future Products of Biotechnology and Needs for Risk Analysis Science: NAS Findings and Recommendations
- Bradbury summarized the forecast for biotechnology products and recommendations for regulatory governance and risk analysis based on the 2017 US National Academies of Sciences report.
- Johan Leveau (University of California, Davis) – Characterization and Manipulation of Crop Microbiota
- Leveau presented interactions of micro-organisms that colonize plants and soils, including how the interaction influences plant health, and quantity and quality of yield.
- Richard Garnett (CropLife International) – Global Survey of Microbial Pesticide Regulation and CLI’s Microbial Pesticide Safety Framework
- Garnett described the various regulatory requirements for biopesticides outside of the U.S. and presented the recommendations of CropLife International for biopesticide safety evaluations.
- Shannon Borges (USEPA) – Ecological Risk Assessment for Microbial Pesticides in the U.S.
- Borges reviewed the methods used for testing and assessment of microbial pesticides in the U.S., including an interesting discussion of some of the differences between chemical safety testing and microbial pesticide testing. The USEPA is involved with revision of the biopesticide testing guidelines and is working on this at the Organization for Economic Cooperation and Development (OECD).
- Joshua Fischer (Bayer) – Considerations for Ecological Testing and Assessment of dsRNA
- Unique attributes of dsRNA products, such as proposed use patterns, mode of action, activity spectrum and bioinformatics analysis, were used by Fischer to inform appropriate non-target organism testing, problem formulation and ecological risk assessment.
- Joel Coats (Iowa State University) – Dissipation of dsRNA in an Aqueous System
- Coats described an interesting study that suggested abiotic factors alone can facilitate the degradation of dsRNA in the environment. He concluded that dsRNA recovery was reduced to background by 96 hours post inoculation in aquatic systems with minimal movement to sediment.
- Henry Krueger (EAG Laboratories) – Challenges in Testing Biopesticides in Conventional Laboratory Ecotoxicology Studies
- Krueger discussed the challenges in biopesticides ecotoxicity testing, with a focus on the terrestrial requirements. His talk led to an interesting discussion of problems related to biopesticide toxicity testing in bees.
- Alison Fournier (Smithers-Viscient) – Safety Testing Strategies to Evaluate Biopesticides Used in Crop Protection on Aquatic Organisms
- Fournier presented an overview of the aquatic testing requirements, including identification of the test substance and necessary controls and modifications necessary for successful biopesticide testing.
There were also two posters included in the session.
- Anderson Abel de Souza Machado (Leibniz Institute of Freshwater Ecology and Inland Fisheries)– Current Environmental Regulations Might Not Protect Against Non-Targeted Effects of Microbial Pesticides: The Case of Dipel and Daphnia
- This poster described studies in Daphnia using Bacillus thuringiensis. De Souza Machado displayed non-monotonic chemical heterogeneity, which may have led to non-monotonic dose-response curves.
- Amber Tompsett-Higley (Bayer) – Environmental Safety Assessment for Agricultural Uses of Lipochitoligosaccharides as Plant Growth Regulators
- In this poster, Tompsett-Higley described a safety testing and assessment strategy for two lipochitoligosaccharide products.
Two key cross-cutting themes from the session were that the activity and characteristics of biopesticides are unique and that chemical testing and assessment methods are not applicable in many cases. Also, as the number of biopesticide-type products is expected to continue to increase, methods need to be thoughtfully re-evaluated and revised in order to address the specific nature of the materials.
The biopesticide topic is a new one for SETAC, and it was good to see a healthy level of interest. Attendance was above average for a SETAC session (range 49–85), which was driven by the quality of the presentations. We look forward to building on the interest in this session for future interactions at SETAC.
Marie DeLorenzo and Peter Key, NOAA National Centers for Coastal Ocean Science
Marine organisms in early life stages are especially vulnerable to contaminant exposure. They have greater capacity for chemical uptake due to higher surface area:volume ratios and faster metabolisms, and decreased capacity for chemical metabolism due to under-developed detoxification mechanisms. Moreover, the narrow windows of development in early life stages often leave marine organisms sensitive to contaminant disruption, which can lead to deformities. As marine and estuarine species develop and migrate across habitats, they deal with additional stressors, such as changes in salinity, pH, temperature, etc. Environmental toxicity testing with organisms in early life stages poses unique challenges and requires use of new methodologies to conduct these tests with these early life stage organisms. Topics covered in this session included mixture and multi-stressor toxicity, growth, development, and biomarker endpoints, novel exposures and novel test species. Fourteen presentations (eight platforms and six posters) discussed various attributes of working with organisms in early life stages, including the complexities inherent with use of early life stages and the necessity to use early life stages in contrast to using adult life stages to address difference in contaminant sensitivity of organism life stages. Presenters shared studies across five classes of contaminants using species from four animal phyla and six classes.
Kimberly Prince (University of Florida) examined polychlorinated biphenyls (PCB) concentrations in marsh sediment and the organisms living there. Her research found that mussels can biomagnify PCBs, leading to increases in PCB concentrations in surrounding sediment and a co-inhabitant, the marsh crab. This demonstrated how benthic invertebrate interactions could affect PCB integration into food webs. Kady Lyons (California State University, Long Beach) also presented on PCB toxicity but with the round stingray (Urobatis halleri). Little information is available on organic contaminant exposure or effects for elasmobranch species. Lyons noted a variety of negative impacts, including altered reproductive and osmoregulatory capacity in adults and decreased capacity for energy utilization in embryos.
Use of another novel toxicity test species, the glass eel (Anguilla anguilla), was presented by Catia Figueiredo (Marine and Environmental Sciences Center, Portugal). The compound studied was also less common, the rare earth element lanthanum, which is a contaminant of emerging concern due to its growing use in batteries. Figueiredo quantified bioaccumulation and observed sublethal effects on acetylcholinesterase activity, lipid peroxidation and antioxidant enzymes in juvenile eels at environmentally relevant concentrations. These results will inform further research with other rare earth elements and glass eels.
Two presentations provided new information on microplastics in the marine environment. Cheyenne Stienbarger (University of North Carolina-Wilmington) presented on the trophic transfer of microplastics from larval silversides to larval and juvenile black sea bass. Her research pointed to the need for examining trophic transfer to lend guidance to fishery management and concerns with human consumption. Dorothy Horn (Portland State University) examined the impacts of microfibers on reproduction and development of the Pacific mole crab. Ingesting this debris at environmentally relevant concentrations did alter development, which calls for further research. This was also a relatively novel test species.
Amy Ringwood (University of North Carolina-Charlotte) focused on biomarker responses in coral larvae. She determined that baseline levels of markers of oxidative stress in coral larvae serve as guidelines of sensitivity since little information exists for the larval forms. Larvae of a Hawaiian coral species were also exposed to several sunscreens to assess effects on reef repopulation, development and preservation. These new larval test methods will benefit coral research and conservation. Similarly, Paula Antunes (AquaTox Testing & Consulting) described a new test method with copepod embryos (Acartia tonsa). She discussed the development of a 48-hour embryonic marine copepod toxicity test. The initial testing aimed to assess the acute lethality of saline effluent discharged into marine or brackish environments from Canadian mining operations. Test development lasted two years, and Antunes shared its trials and tribulations, lending insight into the complexities of early life stage testing. Another new testing strategy, aimed at incorporating a marine copepod species indigenous to North America, was presented by Alan Kennedy (US Army Corps of Engineers – ERDC). The test species was the calanoid copepod Pseudodiaptomus pelagicus, and the method developed was a 48-hour acute exposure for sediment elutriate testing. Kennedy presented the initial results with sensitivity to copper, phenanthrene and ammonia.
With continuing interest in oil spill toxicity to early life stages, there were several presentations on various aspects of this topic. Two presentations from the National Marine Fisheries Service Northwest Fisheries Science Center (NWFSC) highlighted oil impacts to Pacific herring embryos. Julann Spromberg presented a population model tested for its ability to predict delayed mortality of herring stocks in Puget Sound. While the model showed limited ability to match observations in the field, it did indicate the intrinsic growth-rate tipping point for the population. Spromberg is conducting additional toxicity assessments with a focus on adverse outcome pathways to refine the model. Alysha Cypher presented some of these additional herring embryo exposures with Alaskan North Slope crude oil. She described the effects on larval cardiac abnormalities, bioenergetics and immune function. These results will be beneficial in determining delayed population impacts of oil spills. John Incardona (also from NWFSC) investigated impacts on Polar cod after embryonic exposure to Alaskan and Norwegian oil. Severe craniofacial malformations were found in hatched larvae from the higher exposures with significant long-term growth impairment in larvae from the lower exposure levels. Lipid levels were also reduced in exposed larvae, which could lead to reduced recruitment due to high overwintering juvenile mortality. Collectively, these studies point to potential for population-level effects in arctic species that are exposed to oil in early life stages.
Another study with fish embryos examined the multi-stressor effects of oil and hypoxia. Casey Lindberg (Duke University) presented on oxidative stress in zebrafish embryos. These embryos were exposed to a polycyclic aromatic hydrocarbon (PAH) mixture and low oxygen concentrations (hypoxia). While PAH alone induced activity of a stress biomarker (glutathione reductase enzyme), hypoxia alone did not. There were also no interactive effects on oxidative stress of PAH and hypoxia. Sechi Uno (Kagoshima University) also presented on a multi-stressor exposure with fish embryos. Uno offered new data on ultraviolet (UV) light-enhanced toxicity of PAHs in marine medaka embryos from sediment exposures. Specific oxygenated PAHs were found to cause cardiac arrhythmia and cardiac abnormalities. The use of very early embryonic stages for this testing can aid prediction of fish population effects. Peter Key (NOAA’s National Centers for Coastal Ocean Science) also presented research on UV light and oil interactions. He described results of 24-hour exposures to a thin oil sheen and UV light in grass shrimp larvae. The multi-stressor exposure reduced fecundity as the shrimp became sexually mature. The subsequent generational effects resulting from acute exposures of early life stages to thin oil sheens and UV light might lead to long-term impacts on grass shrimp populations.
The emerging themes from this session were the importance of using early life stages in toxicity testing to predict population impacts, the diversity of species and test methods being employed, and the growing interest in multi-stressor interactions. Estuarine ecosystems are an important nursery habitat that can be impacted by environmental and anthropogenic stressors, and this research serves as much needed background information. Some data gaps to be addressed with future investigations include refining early life stage testing methodology, incorporating adverse outcome pathways and following acute exposure effects on early life stages through development to evaluate chronic impacts.
John Parsons, University of Amsterdam, and Graham Whale, Shell
Biodegradation is a natural but still poorly understood process, whose outcome strongly depends not only on chemical structure but also on environmental conditions and the microbial diversity. A thorough understanding of the effects of these factors is required for a reliable assessment of the biodegradability and persistence of chemicals. The session was intended to present new developments in our understanding of biodegradation and to identify how existing test methods can be improved and new approaches be used to provide weight of evidence regarding the potential persistence of chemicals in the environment. The session strove to identify where there is need for further research in this area and explore how we can improve the assessment of the persistence properties of chemicals from both a societal and regulatory perspective.
The short session consisted of eight oral presentations and five posters, presenting work from both industry and academia.
- Kathleen McDonough (Procter & Gamble) started the session by discussing practical issues relating to using the OECD 301B test protocol to determine ready biodegradability of, for example, chemicals with low solubility and potential toxicity and chemicals that require adaptation of microbial populations to achieve ready biodegradation. A specific example was the effect of the physical form of polyhydroxyalkanoates on the kinetics of their biodegradation.
- Paul Tratnyek (Oregon Health & Science University) presented data on the biodegradation of 1,2,3-trichloropropane, a relatively poorly studied but highly toxic emerging contaminant that is widely used as intermediate and solvent. This chemical appears to be more persistent than many other chlorinated aliphatic hydrocarbons, and extreme conditions are required to achieve either biological or chemical degradation.
- Alistair Brown (University of Manitoba) presented new data on the behavior of pharmaceuticals and their human transformation products in wastewater treatment plants. The fate of these chemicals depends on operational parameters such as hydraulic retention time and suspended solids concentration. The transformation products are often present as conjugates, and they can undergo back-transformation during treatment.
- Yong Ran (Guangzhou Institute of Geochemistry) focused on chemical oxidation of benzo[a]pyrene using hydrogen peroxide and how it is affected by structure and properties of sediment organic matter. Their results indicate that aliphatic moieties and the stability and nanoporosity of organic matter have an important impact on the mineralisation of benzo[a]pyrene by H2O2
- Uta Hellmann-Blumberg (California EPA) described how transformation products of hydrocarbons can be addressed in the risk assessment of petroleum contaminated sites. For many contaminated sites, a better understanding of anaerobic biodegradation and transformation products is required in order to enable a comprehensive risk assessment.
- Graham Whale (Shell) presented on behalf of Amelie Ott (Newcastle University) the results of a study investigating improvements to the OECD 306 marine biodegradation screening test. The study focused on two potential improvements: increasing cell concentrations to better represent microbial biodiversity and extending the test duration to include extended lag phases. Implementing these improvements in a ring test involving 13 reference laboratories improved the reproducibility of testing outcomes for a set of reference chemicals, although some variability attributed to differences in microbial populations in different seawater samples were still observed.
- Baptiste Poursat (University of Amsterdam) presented the results of a study of the adaptation of microbial populations during long-term exposure to chemicals and the impact this has on the results of ready biodegradability tests. For some chemicals that fail these tests, long-term exposure results in enhanced biodegradation and reduced variability when using populations from different wastewater treatment plants. In some cases, however, extended exposure can lead to loss of the ability to degrade chemicals, indicating that a better understanding of population dynamics under these conditions is required.
- Finally, Graham Whale (Shell) gave a short summary of the objectives and outcomes of the CEFIC LRI-Concawe Workshop on Recent Developments in Science Supportive to the Biodegradation/Persistence Assessment held in September 2018 in Helsinki. This workshop was attended by many representatives of the regulatory community, who discussed the results of a number of recent and current projects. The objective of this workshop and the follow-up activities was to develop approaches to translate key knowledge from past and ongoing research into recommendations to improve regulatory assessment of persistence.
- In addition to the oral presentations, there were posters on the effect of petroleum hydrocarbons and of estrone and triclosan on soil microbial communities (Meijun Dong and Ezinne Osuji, Texas Tech University), the fate of emerging micropollutants and mercury in Capbreton Submarine Canyon sediment (Alyssa Azaroff, Université de Pau), the potential for lowering of the substrate concentration in the OECD 301B ready biodegradability test (David Riggs, Smithers Viscient) and the evaluation of PBT properties of surfactants (Pujeeta Chowdhary, Wood Environment & Infrastructure Solutions).
The research presented at the session makes it clear that while there are many shortcomings in our understanding of the factors that determine the rates and extent of degradation of chemicals in the environment, we are making progress, and there is a potential for improving the methods we use to assess biodegradability and persistence of chemicals. We hope that the SETAC community takes up this challenge.
Larry Kapustka, LK Consultancy, and David Ostrach, Ostrach Consulting, Session Co-Chairs
Sessions addressing multiple stressors have been a mainstay of SETAC North America annual meetings for more than a decade, and the topic continues to hold considerable interest. This year, the local connection was on exhibit through presentations that dealt with chemical and non-chemical stressors acting in the waterways of the California Bay-Delta.
Despite being the last day of the conference and competing with the adorable therapy dogs in the “Pup Rally” in the exhibit hall, attendance hovered between 80 to 90 people throughout the session. Those in attendance in the first half of the session were treated to excellent summaries of monitoring programs to manage chemical contaminants and nutrients. Tracey Collier of the Delta Independent Science Board focused on the need to use adaptive management in water quality programs, made 10 recommendations on areas of needed research, and discussed how these multiple stressors contribute to organismal and population declines in the Bay-Delta. This was followed by Shawn Acuña, Metropolitan Water District of Southern California, and his work on multiple stressors affecting delta smelt where findings suggested hepatotoxins, nutritional status as well as habitat type were adversely affecting reproductive status. The talk of Bethany DeCourten, University of North Carolina Wilmington, focused on effects of interactions of temperature and endocrine disrupting chemicals on estuarine fishes with results indicating adverse multi-generation effects of contaminant exposure. Marissa Giroux, University of California, Riverside, presented her work on exploration of the interactions of temperature and the fungicide bifenthrin on juvenile chinook salmon behavior and survival with results suggesting temperature was adversely affecting behavior with potential population level impacts. A break and the Pup Rally diminished attendance at the onset of the final presentations. However, uncharacteristically attendance increased during each of the last three talks of the afternoon, culminating with more than 90 in attendance for the last presentation of the meeting! The final presentations began with Marie DeLorenzo, NOAA National Ocean Service, discussing an interesting examination of the complex interactions of UV light, temperature and salinity on toxicity in estuarine species across multiple trophic levels. DeLorenzo’s results indicated that exposure to Louisiana Sweet Crude (LSC) oil and abiotic factors (e.g. UV light, temperature) increased organism sensitivity to the LSC adversely affecting survival. Peter Van Metre, U.S. Geological Survey, described the early results of an extensive monitoring program that is being conducting in the eastern United States representing agricultural and urban disturbance gradients. Preliminary results show marked differences in stream invertebrate fauna between areas that are predominantly in agricultural use versus urban settings. Mark Sandstrom, U.S. Geological Survey, presented preliminary results of a multi-stressor study measuring pesticides in 85 small streams in California. Insecticides were predicted to be important stressors to aquatic organisms in the streams sampled. Interestingly, greater insecticide uses in urban landscapes dominate the effects on the stream life. The session closed with a stimulating presentation by Wayne Landis, Western Washington University, in which he emphasized the use of tools, including Bayesian Networks capable of depicting causality, to achieve adaptive management of risks posed by multiple stressors.
Summaries Published in the February 2019 Issue
Guy Gilron, Borealis Environmental Consulting, and David DeForest, Windward Environmental
It’s now been several decades since the adverse effects of selenium on fish populations in Belews Lake, North Carolina, and bird populations at Kesterson Reservoir, California, were first documented. Since those early events, research on the fate and effects of selenium in aquatic environments has expanded substantially. The goal of this session was to summarize and highlight some of that research. The session was well attended, with presentations including retrospective evaluations of what has been learned, new research and regulatory challenges. Participants expressed interest through thoughtful questions and interactive discussion throughout the session.
Selenium effects on fish populations in Belews Lake and other early examples were observed in closed systems (e.g., reservoirs) receiving high selenium inputs. These events resulted in selenium toxicity studies with fish from the 1980s to the present, which have supported the development of fish tissue-based selenium guidelines and criteria in Canada and the United States. Guy Gilron, Borealis Environmental Consulting, presented a review of historical episodes of selenium affecting fish populations, which were associated with high selenium concentrations that were much greater than selenium toxicity thresholds. Based on this review, over the last two decades, selenium monitoring has identified water bodies in which selenium toxicity thresholds are sometimes exceeded, but without clear evidence of whether fish populations are being adversely affected. In many of these areas, fish population studies are often conducted, but studies are rarely published and often difficult to find. A primary outcome of Gilron’s presentation was a call for conducting and publishing fish population studies, which can support whether selenium guidelines or criteria are protective against population-impacts, and to facilitate information-sharing and support the development of future population studies.
Regarding the issue of impacts on fish populations, Bill Adams, Red Cap Consulting, provided a review of selenium effects on sturgeon, as well as the effects cadmium, copper and zinc had on the fish. Sturgeon are of particular interest for these elements because they are among the most sensitive species tested for selenium, copper and zinc, and some species and sub-populations are listed as threatened and endangered. Related to concerns of selenium and white sturgeon, Jennifer Sun, San Francisco Estuary Institute, presented a muscle-plug evaluation study conducted by the Regional Monitoring Program for Water Quality in San Francisco Bay. This study was developed in support of the North San Francisco Bay Selenium Total Maximum Daily Load (TMDL), which established a target benchmark of 11.3 µg/g dw in white sturgeon muscle tissue. The studies found that muscle plugs were good proxies for muscle fillets and identified sources of variability in selenium concentrations in white sturgeon, which is being used to design a long-term muscle plug monitoring program to assess trends and attainment of the TMDL target.
Three students from the University of Saskatchewan provided excellent presentations that continue to advance our understanding of selenium fate and effects in aquatic systems:
- Taylor Lane evaluated the maternal transfer of dietary selenomethionine (Se-Met), its effects on the F1 generation of fathead minnows (Pimephales promelas) and the use of embryo microinjection of Se-Met as an alternative method to simulate maternal transfer. Dietary Se-Met (highest dietary concentration was 29.58 µg/g dw) did not affect fecundity rates, hatchability or survival until swim-up. Deformity analysis revealed there was an increased frequency of deformities present with increasing selenium concentration in maternal transfer embryos and Se-Met-microinjected embryos; however, the microinjection exposure had more frequent deformities at lower doses suggesting different bioavailability between maternally transferred Se and microinjected Se-Met.
- Katherine Raes evaluated the bioconcentration of selenite and selenate by field-collected periphyton and trophic transfer to amphipods (Hyalella azteca). Selenite concentrations were found to increase in amphipods a dose-dependent manner, with trophic transfer factors (TTFs) ranging from 0.13 to 0.44 across all treatments. Conversely, selenate concentrations in amphipods fed selenium-exposed periphyton did not differ from control amphipods. This study highlights the importance of understanding selenium speciation in assessing selenium bioaccumulation in aquatic systems.
- Stephanie Graves conducted an in situ selenium bioaccumulation study in International Institute for Sustainable Development – Experimental Lakes Area (IISD-ELA) in northwestern Ontario. Selenite was dosed into littoral enclosures and total selenium concentrations were measured in water, sediment, benthic macroinvertebrates and female fathead minnows (promelas). Selenium enrichment factors ranged from about 7,400 to 12,000 L/kg dw and TTFs ranged from 0.6 for Gammaridae to 3.5 for Chironomidae overall treatments. The mean ovary selenium concentration in fathead minnows was 45.51 µg/g dw in the enclosure dosed with a selenite concentration of 9.58 µg/L, which exceeds regulatory guidelines and criteria. The study highlighted the increased bioaccumulation potential of a reduced selenium form (i.e., selenite) in a lentic system.
While the presentations summarized above focused on selenium fate and effects research, Sarah Skigen-Caird and Suzanne Pargee (GEI Consultants) gave a presentation on the regulatory challenges in implementing the U.S. Environmental Protection Agency’s fish tissue-based selenium criteria, finalized in 2016. Although the USEPA developed draft implementation guidance, the guidance has not yet been finalized and states have been hesitant to adopt the tissue-based criteria. They reviewed some of the issues and challenges arising from their discussions with state regulators and regional USEPA offices. These included fishless streams, underlying geology, resident species and a disconnect between water and tissue concentrations. Lessons learned and practical considerations for developing site-specific criteria were presented.
Lastly, Adrian deBruyn, Golder Associates, presented a study that evaluated the cause of increased selenium bioaccumulation downstream of a treatment plant that was reducing in-stream total selenium concentrations. Laboratory algal uptake tests were combined with field measurements of bioaccumulation, and selenium speciation analysis of treated effluent and receiving waters to develop a model of selenium bioaccumulation that explicitly accounts for speciation. The observed increase in bioaccumulation were quantitatively attributed to reduced species, including selenite, dimethylselenoxide and methylseleninic acid. A model was then developed to predict how bioaccumulation would be affected by a post-treatment oxidation step that returns effluent speciation to a selenate-dominated condition.
Overall, the session highlighted how much the understanding of selenium fate and effects has evolved over the last 40 years but also that important research continues. Although the importance of selenium speciation on bioaccumulation potential has been recognized for a long time, it is clear that more studies are focusing on how speciation affects bioaccumulation models, and this information is starting to be used to address real-world problems. In addition, although selenium regulations for aquatic systems have substantially progressed, it is clear that implementation of regulations which include both water- and fish tissue-based elements is still evolving and that site-specificity will always be an important consideration for evaluating the impacts of selenium.
Barnett Rattner, USGS – Patuxent Wildlife Research Center; Katrina Leigh, Ramboll; and John Elliott, Environment and Climate Change Canada
Wildlife ecotoxicology research and monitoring activities have long served to support conservation and regulatory decision-making. This session, organized by members of SETAC and The Wildlife Society, included eight platform presentations (with 50 to 120 attendees) and 10 posters that provided examples of how ecotoxicological data are useful for informed decision-making relevant to wildlife management or related regulatory actions at local, national and even global scales.
Mark Johnson, US Army Public Health Center, kicked off the session with a description of the status of wildlife ecotoxicology test protocols and knowledge, emphasizing the paucity of laboratory model species used in testing. This makes interspecific extrapolation challenging due to a myriad of toxicokinetic factors (e.g., exposure characteristics, gastrointestinal physiology, environmental weathering of compounds). The need to derive toxicity reference values for wildlife is paramount. Next, Phyllis Fuchsman, Ramboll, discussed the use of body weight scaling factors for chronic bioaccumulation studies of metals and polychlorinated biphenyls (PCBs). In the 1990s, some methods of scaling small animal acute toxicity studies produced implausible results for large animals, and it was concluded that without multispecies chronic toxicity data sets, such extrapolations should be dropped from risk assessments. However, some two decades later use of multispecies datasets from chronic bioaccumulation studies provided insight into such interspecific extrapolations. Using various exposure metrics (e.g., dietary exposure concentration, dose mg/kg body weight, mallard-equivalent doses determined by body weight scaling factors), it may now be appropriate to reconsider the current practice of toxicity extrapolations among species on a dose basis without accounting for differences in body weight. Brad Sample, Ecological Risk, Inc., then presented results of field studies that guide management of molybdenum exposure of mule deer and elk at a mining site in New Mexico. Diet composition of deer and elk grazing at tailing facilities was determined using plant chloroplast DNA in scat samples. Using molybdenum bioaccumulation factors for 17 species of plants commonly consumed by these herbivores, six plants (rubber rabbitbrush, blue grama, bottlebrush squirreltail, big sagebrush, western wheatgrass and common sunflower) were identified for a re-vegetation seeding mix that would minimize exposure and risk to these large grazing mammals. A provocative presentation by Larry Tannenbaum, US Army Public Health Center, followed, describing the importance of site fidelity when selecting mammals as receptors of concern for ecological risk assessments. Based on state-of-the-art GPS tracking data with gray fox, paralleling previous work with the white-tailed deer, the worthiness of larger mammal inclusion for typically small Superfund-type sites was challenged. Audience debate followed, with conclusions suggesting that, for various reasons (e.g., lack of site fidelity, large home range), it could be that for sites small in size, no mammals are appropriate when undertaking risk assessments.
Modeling impacts of chemical mixtures across landscapes is challenging due to spatial heterogeneity and complexity of sources, pathways and fate of multiple contaminants in the mixture. Using geographic information systems (GIS) and associated techniques, Kristin Eccles, University of Ottawa, and coworkers evaluated 1,100 samples from five biomonitoring projects of the Athabasca Oil Sands area in Canada in a spatial ecological risk assessment. Results from this study support the use of GIS to integrate data and assess complex exposures and dose-response relationships. Observed patterns of contaminants in the oil sands region demonstrate an increase in metal exposure in wildlife located around the upgrader region when compared with samples from other regions. Myra Finkelstein, University of California, Santa Cruz, then presented extensive stable isotope data demonstrating that ingestion of spent lead ammunition from feeding on terrestrial mammal carcasses is the principal source of lead in free-ranging California condors, while feeding on marine mammals is a significant source of p,p’-DDE. Lead poisoning is the primary threat preventing condor recovery in the wild. Population viability and management models were used to quantify the effects of contaminant exposure and guide management actions that foster population growth of the condor. These analyses led the State of California legislature to enact regulations banning the use of lead ammunition for hunting in California (effective 2019), and also to prioritize the use of other management resources to enhance condor population growth, principally through release of captive-bred condors. Stella McMillin, California Department of Fish and Wildlife, next described long-term state monitoring for anticoagulant rodenticides in wildlife. Interpretation of data has been challenging. While some samples (mountain lion, San Joaquin kit fox, fisher) have been collected through rigorous scientific activities other samples have inherent biases (e.g., carcasses from rehabilitation centers near human population centers). Difference in species sensitivity and variation in residue reporting limits have complicated the interpretation of findings. Despite 2014 regulations restricting use of the more toxic second-generation anticoagulant rodenticides, cases of intoxication from these compounds have continued to be detected. The closing presentation from Carolyn Marn, US Fish and Wildlife Service, reviewed the process of Natural Resource Damage Assessment undertaken in response to release of hazardous substances and the discharge of oil into navigable waters, and the requirements and challenges that must be met to seek compensation from responsible parties to restore injured wildlife.
Ten poster presentations accompanied the platform session on a variety of topics, some of which focused on use of wildlife toxicology data to support management decisions. These included new test data that could support the establishment of an early life stage avian toxicity test, use of immune function measurements in colonial waterbirds to rank areas of concern in the Great Lakes, and development of toxicity reference values for selenium in birds that incorporate trophic transfer factors, which may be more robust in predicting effects in free-ranging birds.
Clearly, this session covered myriad topics and examples of how the science of ecotoxicology has supported risk assessment and remediation activities at the local, national and international levels. Presentations were informative and intellectually stimulating to students and young professionals and also to well-established scientists and risk assessors.
Michael Anderson and Regina Donohoe, California Department of Fish and Wildlife, Office of Spill Prevention and Response
Three years ago, an oil pipeline ruptured adjacent to Refugio State Beach, Santa Barbara, California, and more than 120,000 gallons of crude oil flowed down a drainage ditch into the Pacific Ocean. In the hours, days, weeks, months and years following the spill, natural resource experts conducted studies to shed light on the short- and long-term effects of the spill on this scenic stretch of coastline, using a process known as Natural Resource Damage Assessment (NRDA). In this session, scientists from the California Department of Fish and Wildlife, Office of Spill Prevention and Response (CDFW-OSPR); National Oceanic and Atmospheric Administration, Office of Response and Restoration (NOAA-ORR); and U.S. Fish and Wildlife Service (USFWS) summarized key NRDA findings in eight platform and three poster presentations.
The Santa Barbara Channel has unique ocean currents and natural oil seeps that release oil from the seafloor. Michael Anderson, CDFW-OSPR, presented how NOAA trajectory models predicted the spilled oil would move down the southern California coast to Los Angeles, which was verified by oiling observations along the shoreline. Through careful chemical analyses, Greg Baker, NOAA, detailed how tiered forensic techniques distinguished the spilled oil from natural seeps in samples of stranded tar balls, floating oil and sandy beach invertebrate tissues. Oil exposure was quantified in a variety of media. On sandy beaches, polycylic aromatic hydrocarbons (PAHs) in sand crab (Emerita analoga), polychaete (Thoracophelia spp.) and talitrid (Megalorchestia spp.) tissues and beach porewater were elevated following the spill, declining with distance away from the spill and over time, consistent with oiling observations. In surfperch (Embiotocidae), biliary PAH metabolite levels were elevated at Refugio State Beach, compared with less oiled and natural seep areas, providing evidence of oil exposure in the nearshore environment.
Bruce Joab, CDFW-OSPR, reported on talitrid population monitoring data which indicated oil exposure was associated with declines in abundance of this important wrack-feeding amphipod. In the intertidal and subtidal habitats, discolored surfgrass (Phyllospadix torreyi) and algae were observed near the spill site. Natalie Cosentino-Manning, NOAA-ORR, described quantitative field survey results that related the degree of discoloration to the level of nearshore oiling in this productive habitat.
Dead fish, invertebrates and birds washed onto beaches adjacent to the release point immediately after the spill. To understand effects on fish and invertebrate early life stages, sand crab megalopae, inland silverside (Menidia beryllina) larvae, and blue mussel (Mytilus spp.) larvae bioassays were conducted by exposing the organisms to dilutions of a high-energy, water-accommodated fraction of the spilled oil. Bryand Duke, CDFW-OSPR, and Stephen Clark, Pacific EcoRisk, concluded sand crabs were the most sensitive organism tested, followed by the inland silverside, with six- to seven-day lethal concentrations to 50% of the test organisms (LC50) ranging from 41 to 76 µg total PAHs/liter. Regina Donohoe, CDFW-OSPR, presented surf water and porewater chemistry data which indicated water concentrations after the spill exceeded lethal levels for early life stages. Similarly, Laurie Sullivan, NOAA-ORR, summarized how grunion (Leuresthes tenuis) eggs collected from beaches that were exposed to oil had lower hatch rates in the laboratory, compared with baseline levels. For birds, a plausible connection between oil ingestion via preening and feeding and the increase in egg infertility in Western Snowy Plovers (Charadrius nivosus nivosus) at Coal Oil Point Reserve was demonstrated by Jenny Marek (USFWS). Additional assessments were conducted to determine whether marine mammal strandings were likely spill-related.
Overall, more than 50 studies were conducted to quantify injuries to wildlife, habitat and lost human uses for these resources. Through the NRDA process, restoration projects will be identified to both restore and compensate for these losses. Assessment results and proposed restoration projects will be provided in the Draft Refugio Beach Oil Spill Damage Assessment and Restoration Plan anticipated for release in 2019 at www.wildlife.ca.gov/OSPR/NRDA/Refugio.
Steve Bay, Southern California Coastal Water Research Project; W. Tyler Mehler, University of Alberta; Tamara Sorell, Brown and Caldwell; Lisa Nowell, US Geological Survey; Jay Gan, University of California, Riverside; and Jing You, Jinan University
Sediment continues to be a key medium to evaluate aquatic risk of environmental contamination. Tools to evaluate sediments have continually been refined and developed and now are addressing complicated issues including bioavailability, mixture toxicity and causal assessment. This session included descriptions of new tools for assessing bioavailability and causation, as well as refinements to existing tools, and examples of their application in the field.
The first four presentations were dedicated to understanding some of the complexity of sediment contamination assessment, namely bioavailability and toxicity endpoint choice. After the coffee break, the subsequent platforms discussed novel tools for causal assessment and described applications of some of these tools in monitoring programs.
Key messages from the presentations included:
- The use of passive sampling and other methods to assess bioavailability or bioaccessibility of trace organics in sediment continues to expand. The use of solid phase microextraction (SPME) to monitor contaminant flux through a sediment cap was presented by Jacob Williams, US Army Corps of Engineers. Jay Gan, University of California, Riverside, described the use of TENAX desorption to investigate the effects of sediment aging on contaminant bioaccessibility and bioaccumulation.
- Application of laboratory exposures using aquatic macroinvertebrates to assess contaminant bioavailability and impacts were described in two presentations. Isabelle Proulx, GHD, described the use of different species of Chironomus (each having different feeding modes) to examine changes in metal bioavailability in oxic vs. anoxic sediments. Progress in refining the use of sediment avoidance endpoints in toxicity testing was presented by Roger Yeardley, US Environmental Protection Agency. These tests have the potential for increased sensitivity compared with standard endpoints, but more development and consideration of other behavioral responses is needed to improve interpretation of the results.
- Advances in methods for Toxicity Identification Evaluation (TIE) and causal assessment were described in several presentations. Refinement of an in-situ TIE system to characterize toxicants in sediment porewater was described by Allen Burton, University of Michigan; improvements include addition of an electronic controller to increase reliability. The combination of TIE with effect-directed analysis (EDA) to diagnose the cause of sediment toxicity was described by Jing You, Jinan University. Toxicity of EDA fractions were evaluated using a novel high-throughput test method with Chironomus. David Gillett, Southern California Coastal Water Research Project, described progress in developing a screening-level causal assessment approach for benthic macroinvertebrates. A key step in the framework is the development of comparator sites to control for the effects of confounding habitat factors.
- The closing talk of the session by Bryn Phillips, University of California, Davis, presented a synthesis of 10 years of chemistry and sediment toxicity monitoring in California’s Stream Pollution Trends Program (SPoT). The SPoT monitoring program has documented temporal changes in contaminant distributions and toxicity and continues to evolve to include new methods and address contaminants of emerging concern.
This session was supported by the Sediment Interest Group (SEDIG), and we encourage those interested in the topics summarized here to join the SEDIG community online.
Valery E. Forbes, University of Minnesota; Ismael Rodea-Palomares, Bayer Crop Sciences N.A.; Amelie Schmolke, Waterborne Environmental; and Nika Galic, Syngenta Crop Protection, LLC
Chemical fate models are fundamental tools for understanding and forecasting the behavior of chemicals in the environment and potential exposures to them. Biological and ecological effect models are essential to understand potential environmental impacts of chemicals on individuals, populations and ecosystems. Both families of models can function as powerful tools for risk assessment, and their integration is particularly beneficial for higher-tier environmental risk assessment (ERA). Higher-tier refinements typically include assessing risks to non-test species, across large spatial and/or temporal scales and multiple exposure scenarios. All of these are too difficult to test empirically, due to logistical or ethical reasons. Modeling approaches are able to bypass these difficulties and integrate important information at appropriate scales for a more comprehensive and environmentally relevant risk assessment. Notwithstanding their usefulness, there are still challenges for the development and integration of fate and effect models including technical, practical, conceptual and communication challenges across modeling communities and users. A lack of guidance on model development and implementation, a lack of data for model parameterization, and a lack of case studies to demonstrate proof of concept are frequent reasons given for models not being used or accepted.
This session focused on recent advances in environmental fate and effect modeling related to model parameterization, robustness, consistency and transparency. The first part of the session showed how integrated fate and effect models can improve the understanding of risks associated with pesticides. Rafael Muñoz-Carpena, University of Florida, demonstrated how global sensitivity and uncertainty analyses (GSUA) can provide a powerful and flexible framework to identify important factors driving model sensitivity when a multitude of factors interact together to affect model output. He illustrated the use of GSUA to guide model parametrization and development for pesticide run-off modeling. In particular, he presented a series of case studies emphasizing how features of realistic field conditions, such us concentrated flow or the presence of a shallow water table, could influence the effectiveness of vegetative filter strips in reducing pesticide runoff and inform pesticide run-off mitigation strategies. In related work, Ismael M. Rodea-Palomares, University of Florida, currently at Bayer CropScience, showed how current regulatory higher-tier exposure models (PWC shell) could be linked with a mechanistic aquatic effect model (AQUATOX) to evaluate the efficacy of vegetative filter strips in reducing the long-term risk and effects of pesticide runoff to aquatic organisms. The work demonstrated the challenges involved in integrating the models and highlighted the nonlinear relationships between reductions in exposure and magnitude of effects for different vegetative filter lengths. An important take-home message is that integrated models which can deal with such complexity can greatly facilitate the assessment of risk.
Matthew Etterson, US Environmental Protection Agency, continued the theme of model sensitivity analysis and pesticide risk assessment by exploring the relative importance of three model parameter categories – life history, exposure, toxicity – in an integrated avian fate and effects model (TIM-MCnest). In the absence of exposure, avian reproductive output was most sensitive to mortality (egg, nestling and adult) across all metrics. However, with exposure, different sensitivity metrics conveyed different information about the consequences of perturbations of model parameters. These were not necessarily correlated with each other, implying that great care is needed when choosing such metrics.
One important task for which modeling is ideal is providing information to help guide where the appropriate spatial and temporal scale focus should be that are relevant for the risk assessment being conducted. Wenlin Chen, Syngenta Crop Protection, LLC, evaluated two watershed modeling approaches for their ability to predict aquatic pesticide exposure concentrations. One model was process based (PRZM-SWAT), whereas the other was based on a time-series regression (SEAWAVE-QEX). Outputs of both models were compared with high sampling frequency (near daily) monitoring data and shown to have somewhat different predictive capabilities and limitations. As a process-based model, PRZM-SWAT may be used for long-term forecast of potential exposures. On a retrospective basis, SEAWAVE-QEX is more useful for estimating potential exposures that might be missed in monitoring programs with sampling intervals larger than daily. Results of the presentation also highlighted the importance of data-informed modeling to improve model predictability of different magnitudes.
The second part of the session featured mechanistic effect models, demonstrating their value in extrapolating from toxic effects of chemicals on individuals to consequences for populations. Jill Awkerman, US Environmental Protection Agency, presented a systematic framework for population model development applied to anuran amphibians. The value of such a framework is to add consistency and provide guidance on model development while allowing flexibility to adapt models for specific risk assessment needs. Chiara Accolla, University of Minnesota, used an individual-based model of three trout species (Oncorhynchus mykiss, O. clarkia stomias and Salmo trutta) to show how the same degree of chemical impact at the individual level can have different consequences at the population level, even for closely related species. In this case study, differences in fecundity and mortality rates appear to be important contributors to the observed species-specific differences in population dynamics. Maxime Vaugeois, University of Minnesota, explored the effects of hypothetical stressors, simulated to impact different aspects of fathead minnow (Pimephales promelas) energy budgets, on minnow population dynamics. Not only did the population-level consequences vary depending on which individual-level response was impacted, but they varied markedly between populations limited by food (i.e., those under density-dependent control) and those controlled by predators (i.e., those under density-independent regulation). Amelie Schmolke, Waterborne Environmental, linked an individual-based model of the Topeka shiner (Notropis topeka) with an aquatic food web model (CASM) to explore potential direct and indirect effects of pesticides on this endangered species. This is a particularly important application of population modeling given the requirement that potential direct and indirect pesticide risks to all species listed as threatened or endangered under the Endangered Species Act need to be assessed, and population modeling has been recommended as a tool with which to do so.
By addressing one or more of the key concerns believed to impede the use of models in ERA, all of the presentations in this session contributed to increasing model implementation in the future. Individually and collectively, they provided guidance on model development and use, they presented concrete case studies showing how integrated fate-effect modeling can quantify risks in complex and realistic systems, and they demonstrated a wide variety of strategies for model parameterization and analysis. Presentations in this session demonstrated the importance that fate and effect modeling has reached as ERA tools in the last couple of decades, and emphasized ongoing progress by introducing new tools and strategies to help improve model development, analysis and applicability, as well as understanding and credibility.