Setting the stage to advance oil toxicity testing: Overview of knowledge gaps, and recommendations.

The Chemical Response to Oil Spills: Ecological Effects Research Forum created a standardized protocol for comparing the in vivo toxicity of physically dispersed oil to chemically dispersed oil to support science-based decision making on the use of dispersants in the early 2000s. Since then, the protocol has been frequently modified to incorporate advances in technology; enable the study of unconventional and heavier oils; and provide data for use in a more diverse manner to cover the growing needs of the oil spill science community. Unfortunately, for many of these lab-based oil toxicity studies consideration was not given to the influence of modifications to the protocol on media chemistry, resulting toxicity and limitations for the use of resulting data in other contexts (e.g., risk assessments, models). To address these issues, a working group of international oil spill experts from academia, industry, government, and private organizations was convened under the Multi-Partner Research Initiative of Canada's Oceans Protection Plan to review publications using the CROSERF protocol since its inception to support their goal of coming to consensus on the key elements required within a "modernized CROSERF protocol".

Field fluorometers for assessing oil dispersion at sea.

Oil dispersion by the application of chemical dispersants is an important tool in oil spill response, but it is difficult to quantify in the field in a timely fashion that is useful for coordinators and decision-makers. One option is the use of rugged portable field fluorometers that can deliver essentially instantaneous results if access is attainable. The United States Coast Guard has suggested, in their Special Monitoring of Applied Response Technologies (SMART) protocols, that successful oil dispersion can be identified by a five-fold increase in oil fluorescence. Here we test three commercial fluorometers with different excitation/emission windows (SeaOWL, Cyclops 7FO, and Cyclops 7F-G) that might prove useful for such applications. Results show that they have significantly different dynamic ranges for detecting oil and that using them (or similar instruments) in combination is probably the best option for successfully assessing the effectiveness of oil dispersion operations. Nevertheless, the rapid dilution of dispersed oil means that measurements must be made within an hour or two of dispersion, suggesting that one feasible scenario would be monitoring ship-applied dispersants by vessels following close behind the dispersant application vessel. Alternatively, autonomous submersibles might be pre-deployed to monitor aerial dispersant application, although the logistical challenges in a real spill would be substantial.

Advances to the CROSERF protocol to improve oil spill response decision making.

The Chemical Response to Oil Spills: Ecological Effects Research Forum (CROSERF) created a standardized protocol for comparing the toxicity of physically dispersed oil versus chemically dispersed oil to address environmental concerns related to the proposed use of dispersants in the early 2000s. Since then, many revisions have been made to the original protocol to diversify the intended use of the data generated, incorporate emerging technologies, and to examine a wider range of oil types including non-conventional oils and fuels. Under the Multi-Partner Research Initiative (MPRI) for oil spill research under Canada's Oceans Protection Plan (OPP), a network of 45 participants from seven countries representing government, industry, non-profit, private, and academic sectors was established to identify the current state of the science and formulate a series of recommendations to modernize the oil toxicity testing framework. The participants formed a series of working groups, targeting specific aspects of oil toxicity testing, including: experimental conduct; media preparation; phototoxicity; analytical chemistry; reporting and communicating results; interpreting toxicity data; and appropriate integration of toxicity data to improve oil spill effects models. The network participants reached a consensus that a modernized protocol to assess the aquatic toxicity of oil should be sufficiently flexible to address a broad range of research questions in a 'fit-for-purpose' manner, where methods and approaches are driven by the need to generate scientifically-defensible data to address specific study objectives. Considering the many needs and varied objectives of aquatic toxicity tests currently being conducted to support and inform oil spill response decision making, it was also concluded that the development of a one size fits all approach would not be feasible.

Adopting a toxic unit model paradigm in design, analysis and interpretation of oil toxicity testing.

The lack of a conceptual understanding and unifying quantitative framework to guide conduct and interpretation of laboratory oil toxicity tests, has led investigators to divergent conclusions that can confuse stakeholders and impede sound decision-making. While a plethora of oil toxicity studies are available and continue to be published, due to differences in experimental design, results between studies often cannot be compared. Furthermore, much resulting data fails to advance quantitative effect models that are critically needed for oil spill risk and impact assessments. This paper discusses the challenges posed when evaluating oil toxicity test data based on traditional, total concentration-based exposure metrics and offers solutions for improving the state of practice by adopting a unifying toxic unit (TU) model framework. Key advantages of a TU framework is that differences in test oil composition, sensitivity of the test organism/endpoint, and toxicity test design (i.e., type of test) can be taken into quantitative account in predicting aquatic toxicity. This paradigm shift is intended to bridge the utility of laboratory oil toxicity tests with improved assessment of effects in the field. To illustrate these advantages, results from literature studies are reassessed and contrasted with conclusions obtained based on past practice. Using instructive examples, model results are presented to explain how dissolved oil composition and concentrations and resulting TUs vary in WAFs prepared using variable loading or dilution test designs and the important role that unmeasured oil components contribute to predicted oil toxicity. Model results are used to highlight how the TU framework can serve as a valuable aid in designing and interpreting empirical toxicity tests and provide the data required to validate/refine predictive toxicity models. To further promote consistent exposure and hazard assessment of physically and chemically dispersed oil toxicity tests recommendations for advancing the TU framework are presented.

Impact of mixing and resting times on the droplet size distribution and the petroleum hydrocarbons' concentration in diluted bitumen-based water-accommodated fractions (WAFs).

The preparation of Water-accommodated Fractions (WAFs) and chemically enhanced WAFs (CEWAFs) are essential for evaluating oil toxicity. The Chemical Response to Oil Spills: Ecological Research Forum (CROSERF) method was widely adopted, with variables (e.g., mixing time, oil loading, etc.) being continuously changed among research groups, which limits the cooperation in this area. Herein, we conducted WAF and CEWAF experiments using two loadings of diluted bitumen (Dilbit): 1 g/L and 10 g/L. For the CEWAF, the dispersant to oil ratio was 1:20. We investigated the impact of three mixing durations (18 h, 42 h, and 66 h) and two resting times (6 h and 24 h) on the droplet size distribution (DSD) and accommodated oil concentration. This would be highly beneficial for analyzing toxicity from oil spills, especially when considering the toxic effect of both suspended oil droplets and dissolved hydrocarbons. The DSD results and oil chemistry analysis showed that at a low oil loading concentration (1 g/L), both WAFs and CEWAFs had the same DSD, with an average d50 (volume median diameter) of 3.38 ± 0.70 µm and 3.85 ± 0.63 µm, respectively. At a high oil loading concentration (10 g/L), the WAFs had an average d50 of 3.69 ± 0.52 µm, showing no correlation with mixing and resting time. The DSD of CEWAFs increased significantly at 42 h mixing and 24 h resting time, with oil concentration reaching equilibrium after 42 h mixing. Therefore, WAFs appears to require only 18 h mixing and 6 h resting, while it is recommended to have 42 h mixing and 24 h resting for CEWAFs at high dilbit oil loading concentrations.

Development of a dispersibility assessment kit for use on oil spill response vessels.

The use of dispersants can be an effective response tool for large offshore spills by applying dispersants on unemulsified slicks and treating as much oil as possible before it becomes too viscous. Assessing the dispersibility of an oil slick under actual environmental conditions is an important step in spill response decision-making. This research seeks to develop a new field kit that is quick and reliable and could be used by spill response personnel without scientific training. The resulting Dispersibility Assessment Kit (DAK) incorporates an automated mixing unit to standardize the applied energy, thereby eliminating the variability in "hand mixing" that is used in other dispersant field kits. The automated mixing energy was studied to determine the optimal mixing regime that correlates with ocean conditions and was incorporated in the DAK protocol. The DAK was validated against 14 oils and emulsions and was successfully tested by response personnel during at-sea demonstration.

Issues and challenges with oil toxicity data and implications for their use in decision making: a quantitative review.

Aquatic toxicity considerations are part of the net environmental benefit analysis and approval decision process on the use of dispersants in the event of an offshore oil spill. Substantial information is available on the acute toxicity of physically and chemically dispersed oil to a diverse subset of aquatic species generated under controlled laboratory conditions. However, most information has been generated following standard laboratory practices, which do not realistically represent oil spill conditions in the field. The goal of the present quantitative review is to evaluate the use of standard toxicity testing data to help inform decisions regarding dispersant use, recognizing some key issues with current practices, specifically, reporting toxicity metrics (nominal vs measured), exposure duration (standard durations vs short-term exposures), and exposure concentrations (constant vs spiked). Analytical chemistry data also were used to demonstrate the role of oil loading on acute toxicity and the influence of dispersants on chemical partitioning. The analyses presented here strongly suggest that decisions should be made, at a minimum, based on measured aqueous exposure concentrations and, ideally, using data from short-term exposure durations under spiked exposure concentrations. Available data sets are used to demonstrate how species sensitivity distribution curves can provide useful insights to the decision-making process on dispersant use. Finally, recommendations are provided, including the adoption of oil spill-appropriate toxicity testing practices.

Toxicity testing of dispersed oil requires adherence to standardized protocols to assess potential real world effects.

Recently, several researchers have attempted to address Deepwater Horizon incident environmental fate and effects issues using laboratory testing and extrapolation procedures that are not fully reliable measures for environmental assessments. The 2013 Rico-Martínez et al. publication utilized laboratory testing approaches that severely limit our ability to reliably extrapolate such results to meaningful real-world assessments. The authors did not adopt key methodological elements of oil and dispersed oil toxicity standards. Further, they drew real-world conclusions from static exposure tests without reporting actual exposure concentrations. Without this information, it is not possible to compare their results to other research or real spill events that measured and reported exposure concentrations. The 1990s' Chemical Response to Oil Spills: Ecological Effects Research Forum program was established to standardize and conduct exposure characterization in oil and dispersed oil aquatic toxicity testing (Aurand and Coelho, 2005). This commentary raises awareness regarding the necessity of standardized test protocols.