Parsing the toxicity paradox: Composition and duration of exposure alter predicted oil spill effects by orders of magnitude.

Oil spilled into an aquatic environment produces oil droplet and dissolved component concentrations and compositions that are highly variable in space and time. Toxic effects on aquatic biota vary with sensitivity of the organism, concentration, composition, environmental conditions, and frequency and duration of exposure to the mixture of oil-derived dissolved compounds. For a range of spill (surface, subsea, blowout) and oil types under different environmental conditions, modeling of oil transport, fate, and organism behavior was used to quantify expected exposures over time for planktonic, motile, and stationary organisms. Different toxicity models were applied to these exposure time histories to characterize the influential roles of composition, concentration, and duration of exposure on aquatic toxicity. Misrepresenting these roles and exposures can affect results by orders of magnitude. Well-characterized laboratory studies for <24-hour exposures are needed to improve toxicity predictions of the typically short-term exposures that characterize spills.

Hot and Cold: Photochemical Weathering Mediates Oil Properties and Fate Differently Depending on Seawater Temperature.

Photochemical weathering transforms petroleum oil and changes its bulk physical properties, as well as its partitioning into seawater. This transformation process is likely to occur in a cold water marine oil spill, but little is known about the behavior of photochemically weathered oil in cold water. We quantified the effect of photochemical weathering on oil properties and partitioning across temperatures. Compared to weathering in the dark, photochemical weathering increases oil viscosity and water-soluble content, decreases oil-seawater interfacial tension, and slightly increases density. Many of these photochemical changes are much larger than changes caused by evaporative weathering. Further, the viscosity and water-soluble content of photochemically weathered oil are more temperature-sensitive compared to evaporatively weathered oil, which changes the importance of key fate processes in warm versus cold environments. Compared to at 30 °C, photochemically weathered oil at 5 °C would have a 16× higher viscosity and a 7× lower water-soluble content, resulting in lower entrainment and dissolution. Collectively, the physical properties and thus fate of photochemically weathered oil in a cold water spill may be substantially different from those in a warm water spill. These differences could affect the choice of oil spill response options in cold, high-light environments.

Recommendations for the advancement of oil-in-water media and source oil characterization in aquatic toxicity test studies.

During toxicity testing, chemical analyses of oil and exposure media samples are needed to allow comparison of results between different tests as well as to assist with identification of the drivers and mechanisms for the toxic effects observed. However, to maximize the ability to compare results between different laboratories and biota, it has long been recognized that guidelines for standard protocols were needed. In 2005, the Chemical Response to Oil Spills: Ecological Effects Research Forum (CROSERF) protocol was developed with existing common analytical methods that described a standard method for reproducible preparation of exposure media as well as recommended specific analytical methods and analyte lists for comparative toxicity testing. At the time, the primary purpose for the data collected was to inform oil spill response and contingency planning. Since then, with improvements in both analytical equipment and methods, the use of toxicity data has expanded to include their integration into fate and effect models that aim to extend the applicability of lab-based study results to make predictions for field system-level impacts. This paper focuses on providing a summary of current chemical analyses for characterization of oil and exposure media used during aquatic toxicity testing and makes recommendations for the minimum analyses needed to allow for interpretation and modeling purposes.

Bridging the lab to field divide: Advancing oil spill biological effects models requires revisiting aquatic toxicity testing.

Oil fate and exposure modeling addresses the complexities of oil composition, weathering, partitioning in the environment, and the distributions and behaviors of aquatic biota to estimate exposure histories, i.e., oil component concentrations and environmental conditions experienced over time. Several approaches with increasing levels of complexity (i.e., aquatic toxicity model tiers, corresponding to varying purposes and applications) have been and continue to be developed to predict adverse effects resulting from these exposures. At Tiers 1 and 2, toxicity-based screening thresholds for assumed representative oil component compositions are used to inform spill response and risk evaluations, requiring limited toxicity data, analytical oil characterizations, and computer resources. Concentration-response relationships are employed in Tier 3 to quantify effects of assumed oil component mixture compositions. Oil spill modeling capabilities presently allow predictions of spatial and temporal compositional changes during exposure, which support mixture-based modeling frameworks. Such approaches rely on summed effects of components using toxic units to enable more realistic analyses (Tier 4). This review provides guidance for toxicological studies to inform the development of, provide input to, and validate Tier 4 aquatic toxicity models for assessing oil spill effects on aquatic biota. Evaluation of organisms' exposure histories using a toxic unit model reflects the current state-of the-science and provides an improved approach for quantifying effects of oil constituents on aquatic organisms. Since the mixture compositions in toxicity tests are not representative of field exposures, modelers rely on studies using single compounds to build toxicity models accounting for the additive effects of dynamic mixture exposures that occur after spills. Single compound toxicity data are needed to quantify the influence of exposure duration and modifying environmental factors (e.g., temperature, light) on observed effects for advancing use of this framework. Well-characterized whole oil bioassay data should be used to validate and refine these models.

Counter-historical study of alternative dispersant use in the Deepwater Horizon oil spill response.

Recent completion of oil fate modeling and a mass budget of the Deepwater Horizon (DWH) oil spill allows for a counter-historical study using quantitative Comparative Risk Assessment (CRA) methodology. Novel application of subsea dispersant injection (SSDI) during the response reduced surfacing oil, volatile organic carbon emissions, and oil on shorelines. The effectiveness of that application, and potential alternatives had dispersant not been used or been used more aggressively, were evaluated by modifying and comparing the validated oil fate model under different SSDI strategies. A comparison of mass balance results, exposure metrics, and CRA scoring for Valued Ecological Components (VECs) shows the value of SSDI in achieving risk reduction and tradeoffs that were made. Actual SSDI applied during the DWH oil spill reduced exposures to varying degrees for different VECs. Exposures and relative risks across the ecosystem would have been substantially reduced with more effective SSDI.

Validation of oil fate and mass balance for the Deepwater Horizon oil spill: Evaluation of water column partitioning.

Model predictions of oil transport and fate for the 2010 Deepwater Horizon oil spill (Gulf of Mexico) were compared to field observations and absolute and relative concentrations of oil compounds in samples from 900 to 1400 m depth <11 km from the well. Chemical partitioning analyses using quantitative indices support a bimodal droplet size distribution model for oil released during subsea dispersant applications in June with 74% of the mass in >1 mm droplets that surfaced near the spill site within a few hours, and 1-8% as <0.13 mm microdroplets that remained below 900 m. Analyses focused on 900-1400 m depth <11 km from the well indicate there was substantial biodegradation of dissolved components, some biodegradation in microdroplets, recirculation of weathered microdroplets into the wellhead area, and marine oil snow settling from above 900 m carrying more-weathered particulate oil into the deep plume.

Oil fate and mass balance for the Deepwater Horizon oil spill.

Based on oil fate modeling of the Deepwater Horizon spill through August 2010, during June and July 2010, ~89% of the oil surfaced, ~5% entered (by dissolving or as microdroplets) the deep plume (>900 m), and ~6% dissolved and biodegraded between 900 m and 40 m. Subsea dispersant application reduced surfacing oil by ~7% and evaporation of volatiles by ~26%. By July 2011, of the total oil, ~41% evaporated, ~15% was ashore and in nearshore (<10 m) sediments, ~3% was removed by responders, ~38.4% was in the water column (partially degraded; 29% shallower and 9.4% deeper than 40 m), and ~2.6% sedimented in waters >10 m (including 1.5% after August 2010). Volatile and soluble fractions that did not evaporate biodegraded by the end of August 2010, leaving residual oil to disperse and potentially settle. Model estimates were validated by comparison to field observations of floating oil and atmospheric emissions.

Potential impacts of offshore oil spills on polar bears in the Chukchi Sea.

Sea ice decline is anticipated to increase human access to the Arctic Ocean allowing for offshore oil and gas development in once inaccessible areas. Given the potential negative consequences of an oil spill on marine wildlife populations in the Arctic, it is important to understand the magnitude of impact a large spill could have on wildlife to inform response planning efforts. In this study we simulated oil spills that released 25,000 barrels of oil for 30 days in autumn originating from two sites in the Chukchi Sea (one in Russia and one in the U.S.) and tracked the distribution of oil for 76 days. We then determined the potential impact such a spill might have on polar bears (Ursus maritimus) and their habitat by overlapping spills with maps of polar bear habitat and movement trajectories. Only a small proportion (1-10%) of high-value polar bear sea ice habitat was directly affected by oil sufficient to impact bears. However, 27-38% of polar bears in the region were potentially exposed to oil. Oil consistently had the highest probability of reaching Wrangel and Herald islands, important areas of denning and summer terrestrial habitat. Oil did not reach polar bears until approximately 3 weeks after the spills. Our study found the potential for significant impacts to polar bears under a worst case discharge scenario, but suggests that there is a window of time where effective containment efforts could minimize exposure to bears. Our study provides a framework for wildlife managers and planners to assess the level of response that would be required to treat exposed wildlife and where spill response equipment might be best stationed. While the size of spill we simulated has a low probability of occurring, it provides an upper limit for planners to consider when crafting response plans.

Partial Photochemical Oxidation Was a Dominant Fate of Deepwater Horizon Surface Oil.

Following the Deepwater Horizon (DWH) blowout in 2010, oil floated on the Gulf of Mexico for over 100 days. In the aftermath of the blowout, substantial accumulation of partially oxidized surface oil was reported, but the pathways that formed these oxidized residues are poorly constrained. Here we provide five quantitative lines of evidence demonstrating that oxidation by sunlight largely accounts for the partially oxidized surface oil. First, residence time on the sunlit sea surface, where photochemical reactions occur, was the strongest predictor of partial oxidation. Second, two-thirds of the partial oxidation from 2010 to 2016 occurred in less than 10 days on the sunlit sea surface, prior to coastal deposition. Third, multiple diagnostic biodegradation indices, including octadecane to phytane, suggest that partial oxidation of oil on the sunlit sea surface was largely driven by an abiotic process. Fourth, in the laboratory, the dominant photochemical oxidation pathway of DWH oil was partial oxidation to oxygenated residues rather than complete oxidation to CO2. Fifth, estimates of partial photo-oxidation calculated with photochemical rate modeling overlap with observed oxidation. We suggest that photo-oxidation of surface oil has fundamental implications for the response approach, damage assessment, and ecosystem restoration in the aftermath of an oil spill, and that oil fate models for the DWH spill should be modified to accurately reflect the role of sunlight.

Photochemical oxidation reduced the efficacy of aerial dispersants applied in response to the Deepwater Horizon spill.

Chemical dispersants are one of many tools used to mitigate the overall environmental impact of oil spills. In principle, dispersants break up floating oil into small droplets that disperse into the water column where they are subject to multiple fate and transport processes. The effectiveness of dispersants typically decreases as oil weathers in the environment. This decrease in effectiveness is often attributed to evaporation and emulsification, with the contribution of photochemical weathering assumed to be negligible. Here, we aim to test this assumption using Macondo well oil released during the Deepwater Horizon spill as a case study. Our results indicate that the effects of photochemical weathering on Deepwater Horizon oil properties and dispersant effectiveness can greatly outweigh the effects of evaporative weathering. The decrease in dispersant effectiveness after light exposure was principally driven by the decreased solubility of photo-oxidized crude oil residues in the solvent system that comprises COREXIT EC9500A. Kinetic modeling combined with geospatial analysis demonstrated that a considerable fraction of aerial applications targeting Deepwater Horizon surface oil had low dispersant effectiveness. Collectively, the results of this study challenge the paradigm that photochemical weathering has a negligible impact on the effectiveness of oil spill response and provide critical insights into the "window of opportunity" to apply chemical dispersants in response to oil spills in sunlit waters.

Oil spill impact modeling: development and validation.

A coupled oil fate and effects model has been developed for the estimation of impacts to habitats, wildlife, and aquatic organisms resulting from acute exposure to spilled oil. The physical fates model estimates the distribution of oil (as mass and concentrations) on the water surface, on shorelines, in the water column, and in the sediments, accounting for spreading, evaporation, transport, dispersion, emulsification, entrainment, dissolution, volatilization, partitioning, sedimentation, and degradation. The biological effects model estimates exposure of biota of various behavior types to floating oil and subsurface contamination, resulting percent mortality, and sublethal effects on production (somatic growth). Impacts are summarized as areas or volumes affected, percent of populations lost, and production foregone because of a spill's effects. This paper summarizes existing information and data used to develop the model, model algorithms and assumptions, validation studies, and research needs. Simulation of the Exxon Valdez oil spill is presented as a case study and validation of the model.

Development and application of an oil toxicity and exposure model, OilToxEx.

An oil toxicity and exposure model (OilToxEx) was developed and validated for estimation of impacts to aquatic organisms resulting from acute exposure to spilled oil. Because oil exposure is shorter than the time required for equilibrium between the organism and the water to be reached, the time and temperature dependence of toxicity is addressed. Oil toxicity is a function of aromatic composition and the toxicity of individual aromatics in the mixture. Lethal concentration to 50% of exposed organisms (LC50), as a function of octanol-water partition coefficient (Kow), and an additive model are used to estimate the toxicity of monoaromatic and polycyclic aromatic hydrocarbon mixtures in water-soluble fractions (WSF) and oil-in-water dispersions (OWD) of oil. The toxicity model was verified by comparison with oil bioassay data where the exposure concentrations of aromatics were measured. The observed toxicity in the bioassays could be accounted for by the additive narcotic effects of the dissolved aromatics in the exposure media. Predicted LC50s were compared to those calculated from measured concentrations after spills to verify the exposure model for field conditions. These results indicate that the additive toxicity and exposure model may be used to estimate toxicity of untested oils and spill conditions.