The review of the ecotoxicology of oil and treated oil in the Arctic described by the authors in this section led to suggestions of further research which can reduce remaining uncertainties.  The more generic suggestions compiled from this review are summarized below while recommendations that are important for improving Arctic NEBA are listed separately.

1. Bunker oil and other fuel oils:  As the ship traffic increases in the Arctic, fuel oils are the most likely petroleum compound to be released.  The consequences of accidental discharges of bunker oil need to be better understood, particularly considering that currently limited technical solutions exist to recover bunker oil mixed with sea ice (communication by the coastal administration of Norway).
2. Chemical dispersants: a limited number of chemical dispersants have been tested.  Indeed only Corexit® 9500 (NALCO 2013) and Dasic Slickgone (DASIC 2013) could be found in the recent literature while many other dispersants exist on the market and could potentially be used during an oil spill response. In order to select the best response option, it is important to understand the performance characteristics of chemical dispersants in low temperature applications.
3. Ecotoxicological studies:
   1. To complement the existing lethal toxicity dataset, we recommend conducting toxicity studies to better characterize the effects of dispersed oil, including both acute and chronic exposures.  Where appropriate, longer-term studies should be conducted with treated and untreated oil focusing on ecosystem/species recovery.  Such long-term toxicity data can be used in combination with experiences from cold-water spills (e.g. Exxon Valdez) to identify critical factor for effects/recovery (See Ecosystem Recovery Section).
   2. Single PAHs:  Toxicity data with single compounds provide an important tool for comparing the toxicity of different OSRs, as well as providing model input.
   3. Herding chemicals:  Chemical herding agents are a promising enhancement for spill response amendment under certain conditions.  Examine the effects of herder chemicals that concentrate surface oil and are relatively unknown to date.
   4. Photooxidation: sea ice species living at the surface of the sea can be exposed to oil whose toxicity has been enhanced following exposure to UV, this could be addressed as part of a larger effort to resolve this issue for any marine spill (this issue is not unique to the arctic, and we do not have sufficient resolution for arctic or temperate exposures).
   5. Examine modes of action:  Knowledge of the quantitative effect of oil droplets and soluble components of oil on biota is also of great importance.  Examinations of toxicity modes of action need to address chemical uptake, body burdens and the narcosis related responses but also need to address the potential for the direct effects of fouling and implications of impacts of soluble or volatile compounds on epithelial tissues, especially those related to respiration.  
4. Community level assessments:  As the need for an ecosystem management approach has been growing over the last few years, there is a need to move from single species toxicity tests towards community studies to produce data with the highest ecological possible relevance with a specific focus on the structural and functional changes. One methodological approach as discussed earlier would be the use of in situ and other mesocosms and also to work with early life stage impacts that can be fed into population and possibly ecosystem models.

Photo 6-14: Pelagic Hyperiid amphipod Themisto sp (Bjørn Hansen).

The recommendations presented below indicate where increased knowledge of ecotoxicology would result in reducing existing uncertainties in NEBA assessments.  No prioritization has been made to the list; for some of the recommendations, surrogate data may be already available.

1. Toxicity studies of weathered and unweathered OSR residuals.  Given the remoteness of the Arctic and the inclement weather, it is possible that responders will be treating weathered oil.  The toxicity of oil weathered under Arctic winter conditions is less well understood.  Priorities for toxicity data production should be defined based on the need for input into existing impact assessment tools which will be used by the stakeholders.
   1. ECs coupled with VEC test species:  In non-pelagic environments (surface microlayer, intertidal, shallow subtidal, deep water species, annual and multi-year ice); Pelagic: weathered oil
      • Create exposures to constant droplet size and concentration of OSR residuals; determine biological responses (survival, growth, lipid content, possible fecundity measure); determine impacts due to dissolved fraction and risk to pelagic organisms
      • Conduct in-situ assessment of rate of benthic recovery from controlled oil exposure
   2. Endpoints:  fouling, mortality, uptake and narcosis, histopathology associated with measured chemical concentration in exposure media.
   3. Exposures:  spiked, low constant dose, multiple spiked to simulate re-oiling (e.g. Droplet generator studies).  Although LC50 are used for regulatory risk assessment, the ecotoxicological information that they provide is often not environmentally relevant since the amount of oil tested far exceeds potential environmental concentrations.
   4. Evaluate SSD curves   Evaluate SSD curves for weathered and fresh oils for various modes of action.
   5. Evaluate consequences of OSR strategies.  To date there are very few toxicity studies that have been conducted with certain types of OSRs, most notably ISB residues and OMAs.  A full evaluation of OSR alternatives will require additional toxicity data for targeted VECs, defined by the anticipated fate of the treated oil.Additionally, very few chemical dispersants have been fully evaluated for acute and chronic effects.
2. Augmentation of relative sensitivity dataset. The relative sensitivity assessment of arctic versus temperate species conducted by several authors indicates that the toxicity database for Arctic studies could be augmented with data from temperate species.  Further data are required for sublethal and chronic effects of oil and treated oil before the data can be extrapolated from temperate studies.  Data on benthic species and subtidal communities are sparse.  Further information is also needed for other Arctic VECs such as nearshore fish (e.g. capelin), sea ice, microlayer, and higher trophic groups (such as birds and mammals).
3. Improve model relationships
   1. Impacts due to volatile fractions of oil.  Volatilization of hydrocarbons exposes marine mammals and seabirds to respiratory impacts (greatest with natural attenuation, herder application, mechanical recovery, ISB); exposure needs to model concentrations near air/water interface.
   2. Encompass all modes of toxic action