Chapter Selection 7.3 - Future Research Considerations Next Chapter Previous Chapter

Add notes to:

Or add a reference to download later

Note added.

7.0 POPULATION EFFECTS MODELING

Executive Summary

Photo 7-1. Young Arctic cod congregating under ice
Photo 7-1. Young Arctic cod congregating under ice

Assessing the transport of oil and its components and predicting the toxicity of the mixture to exposed organisms is relatively well developed. However, for a NEBA asessment the initial impact of oil and OSR techniques on individual organisms is only a small part of the overall evaluation that needs to be made. Species sensitivity together with parameters like exposure potential, population growth rates, reproduction capacity, population elasticity, and recovery protential will determine the overall population resiliency. Resiliency of the most important Arctic VECs should be further studied and assessed making this a crucial input parameter for NEBA analysis. Population effects modelling can help to study processes that determine populations resilience. Areas needing further development include the following: 

  • Develop resilience metrics for key Arctic populations 
  • Examination of other modes of action including fouling, inhalation exposure and respones, and epithelial tissue exposure and disruption will add to model accuracy. 
  • Extrapolate predicted toxicity effects to populations of key ecosystem components, requiring an assessment of population dynamics (e.g., age class distribution, sensitivity of different age classes, fecundity measures, distribution and connectivity of populations of key species). 
  • Compare natural annual or multi-year variation in populations of VECs to the projected impacts of oil spill residuals on VEC populations. 
  • Comparative evaluation of each oil spill response option (OSR), particularly in terms of on resilience of the environmental compartment (EC) and valuable ecosystem components (VECs). 

7.1 Introduction

Modeling the potential impact of oil release on marine Arctic communities requires an integrated assessment of the transport and fate of oil, alterations created by application of various oil spill response (OSR) technologies and predictable natural population dynamics of valuable ecosystem components (VECs) responding to an added stress from potential oil exposure.  The Arctic, as in other oceans has an extensive list of species that occupy different portions of that ecosystem.  Conducting evaluations of the potential impacts of oil spills on all species or age classes within each of these compartments is not feasible so a selection process needs to occur.  For this review we examined the different food web structures and selected representative species that are either valued by the inhabitants of the Arctic or are key faunal community members (See chapter 2).  Once the VECs were selected, the food web components that represent key links to those VECs and were tested under controlled laboratory experiment were chosen for toxicity assessment.  In the Arctic, key VECs include fish, seabirds, marine mammals, and polar bears; deep water corals and kelp represent important habitat-producing species.  Phytoplankton, copepods, krill and amphipods provide good examples of important prey species leading to the higher trophic level VECs.  Large pelagic carnivores excluded from food web assessments are jellyfish which are a dominant but poorly studied group of pelagic organisms. 

Over the past decade Arctic studies have been performed that improved the ability to model the effects of oil spills in the Arctic to bird and mammalian species, with more limited improvements in modeling impacts on fisheries resources; the health of these marine fishery resources also directly affects the subsistence cultures.  Enhancing the ability to predict the direct influence of petroleum exposure to populations of key food web components is the focus of this review.  There remain important areas of research that will improve the ability of applying toxicity data to the population dynamics models that are needed to address ecosystem impacts and resiliency of populations to recover from petroleum exposure.

In Section 2 of this report marine and estuarine aquatic environmental compartments (ECs) and corresponding VECs were identified that provide key food web connections to the resource species.  These environmental compartments were: shallow and deep pelagic waters, deep water and nearshore sediment and hard substrate environments, surface layers, nearshore lagoons and estuaries, as well as ice communities.   A number of the species selected as VECs are found in multiple ECs but often at different size and age classes with different sensitivities to oil.  Additionally, the various ECs support varying population levels of these key VECs; individuals within the populations have different natural mortality rates and potential for replacement of resources through reproduction after exposure to oil.  All of these factors need to be considered when developing models to assess resiliency of populations and compare the projected effects of oil spill response actions.

The selected VEC species include phytoplankton, copepods (primarily C. glacialis and finmarchicus), euphausiids (Thysanoessa inermis), amphipods (Lysianassids and Hyperiidae), fish species including the Arctic cod (Boreogadus saida), Polar cod (Arctogadus glacialis), capelin (Mallotus villosus), sculpin of the genus Myoxocephalus and Pacific and Atlantic Herring (Clupea spp), squid, ice seals, toothed and baleen whales, Polar bear and seabirds (Table 7-1).  The quantity and quality of toxicity data for each VEC varies considerably; available toxicity studies are summarized in Section 6.  For copepods, amphipods, and some fish (in particular Arctic cod, sculpin, and capelin) there is toxicity data with multiple age classes for oil and treated oil, as well as individual PAH constituents.  Toxicity data for other water column VECs is less well developed and may only include exposure to oil or specific PAHs. Information on the toxicity of oil to Arctic 
species that live on or create hard substrate (e.g. cold water corals) and those burrowed into the sediment is sparse.  A comprehensive study on Arctic benthic and epibenthic invertebrates was conducted with 2-methyl-naphthalene (Olsen et al. 2010).  Inhabitants of the surface microlayer (SML) and deep water species (>200m of water depth) are the least represented in ecotoxicology research not only in the Arctic but other environments throughout the world.  However, there is indication that Arctic pelagic and benthic species are similarly sensitive to oil and treated oil compared to species found in non-Arctic waters. This increases the data available for predicting Arctic effects for different life stages and reduces the need for Arctic specific effect data (DeHoop et al. 2011, Olsen et al. 2010).

Determining areas of seasonal population aggregations of VECs is important to inform NEBA decision making. However, these efforts require information on life-history and presence/absence data for each VEC.  Such analyses have been done for certain portions of the Arctic including the US and Canadian Beaufort.    An example of such an effort is the Beaufort Regional Environmental Assessment (BREA) program in the eastern Beaufort.  Seasonal movements of VEC species have been overlain with traditional hunting grounds and other data to create VEC vulnerability profiles that indicate specific locations and time periods where the population may be sensitive to certain OSRs (Trudel 2013).  As an example, data has been summarized for White whales (D. leucas) as they enter the eastern Beaufort in early June, with adults and young congregating at the mouth of the Mackenzie River delta (Figure 2-10).  During July, population densities are highest close to the mouth of the river, in areas used by indigenous fisheries.  In September and October, all whales leave the area.  Based on the vulnerability analysis, the White whales would be most vulnerable close to the mouth of the river during July. This approach has been used in different portions of the Arctic; a pan-Arctic compilation of such data would be useful to OSR strategizing.

 

Table 7-1.  Valuable Ecosystem Components of Arctic Communities

Valuable Ecosystem ComponentsAssociated Communities
PelagicBenthicSea-IceDeepwater

Phytoplankton

 

 

Sympagic copepods

Gammarus wilkitzkii

Apherusa glacialis

Onismus spp.

   




 

Calanoid copepods

Calanus hyperboreus

Calanus glacialis

Calanus finmarchicus




 






Euphausiids

Thysanoessa spp


 


 

Hyperiid amphipods

Themisto libellula


 


 

Cephalopods

Gonatus fabricii



 


Pelagic Fish

Arctic cod - Boreogadus saida

Polar cod - Arctogadus glacialis

Capelin - Mallotus villosus

Myctophids

Gonostomids
















Clams

Serripes groenlandica

Macoma sp.




   

Benthic and Epibenthic Amphipods

Ampelisca sp.

Anonyx nugax

Eurythenes gryllus

 




 




Decapod crustaceans

Shrimp -  Pandalus borealis

Crab -  Chionoecetes spp.




   

Echinoderm

Urchin - Strongylocentrotus droebachiensis

 


   

Epibenthic Fish

Sculpin – Myoxocephalus spp.

Eelpout – Lycodes spp.

Greenland halibut  - Reinhardtius sp.







 




Mammals

Ringed seal – Phoca hispeda

Walrus - Odobenus rosmarus

Narwhal – Monodon monoceros

White whale – Delphinapterus leucas

Bowhead whale – Balaena mysticetus

Polar bear – Ursus maritimus
















 

Seabirds

 

 

● Integral component of the community

℗ Prey item or predator, but not an integral component of the community

However, as mentioned earlier, the sensitivity of individual organisms as determined in toxicity tests, is only one parameter of many that are required for the assessment of population resiliency.  Population dynamics modeling studies require a strong understanding of life histories and reproductive cycles, as well as data regarding the sensitivity of different age classes to oil and treated oil.  For Arctic systems, the best candidates for population modeling studies are Arctic cod, Calanoid copepods, amphipods, and seabirds.  Local population abundance and biomass for these taxa have also been determined over short time frames in multiple locations within the Arctic.  Some of these locations have also collected significant information on the daily natural mortality rates that have occurred during different years and seasons.  The fish mortality rates are calculated using daily growth ring comparisons from otolith age-related data establishing population levels while the invertebrate estimations of natural mortality are built around size frequency distribution analyses.  The strength of these data can be enhanced by producing more information on the dynamics of local populations of these species that develop in different environmental compartments.  The importance of population data is illustrated by the lower daily mortality rates for Arctic cod obtained from the Canadian Beaufort near the Mackenzie Delta compared with collections from the Alaskan Beaufort (Gallaway et al. in review).  The population sizes were larger in the Mackenzie delta area than the population in the Alaskan Beaufort, and there also appears to be different daily mortality rates for subpopulations of Arctic cod that spawn in the later fall/early winter during annual ice formation compared to those that spawn to take advantage of the plankton blooms that occur shortly after ice breakup (Gallaway et al. in review).  Additionally, there are annual differences in the population sizes and natural daily mortality rates within the same location.   The ranges of differences in daily mortality rates are up to 10-fold, indicating a need for site-specific assessments.  All of these distinguishing attributes are important for developing models to assess the relative impacts of oil exposure to various life stages.

The effects of oil exposure to a number of pelagic components and life stages (copepods, euphausiids, amphipods, Arctic cod, sculpin and herring) can be modeled based on available data.  A presentation of the requirements for biological characteristics needed to perform population modeling for the effects of oil exposure to crustaceans and fish is addressed in the following sections.  Refinements in evaluating the potential effects of oil exposure on VEC species and life stages need to concentrate on expanding the toxicological database for the selected VECs and age classes.  It is also important to begin evaluating the connections of VECs to different environmental compartments that have been less studied (cold water corals, nearshore kelp, ice algae, hydrocorals, jellyfish, surface microlayer life stages).  The transport and fate models used to describe the movement of oil into various environmental compartments resulting from oil spills and the use of alternative response strategies are well developed for open water pelagic and ice-infested systems.  Evaluation of population level impacts needs to address not only the dilution and initial dispersion processes that occur after an oil release which drives the initial exposure regimes but also needs to examine areas where oil components may re-concentrate (e.g., current convergence zones, pycnoclines, upwelling or downwelling of water masses, shoreline stranding or concentration at air-water, ice-water, or sediment-water interfaces).  These concentration processes and the occupation of these interfaces by VEC food web populations need to be included in the evaluation of the potential effects of spilled oil and response actions. 

The level of realism achieved by well-designed population effect models can be, theoretically, further increased if food web-and ecosystem models are also incorporated in risk assessment procedures (Pastorak et al. 2001; De Laender 2007). Ecological models can be used to include more ecology in environmental toxicology. The need for this has been argued by Chapman (2002), Calow and Forbes (2003) and Van Straalen (2003).  Studies focusing on translation of individual effect levels to population consequences have resulted in recommendations for model use and data needs but have not led to a widely used generic protocol for ecological risk assessment based on population responses (Forbes et al. 2008). However, increasing complexity of models often goes together with an increased data demand. Alternative models which predict population level responses from LC50 data can be considered in cases with low data availability (Hendriks and Enserink 1996; Hendriks et al. 2005).  Incorporating population models in food chain or community models allow for a quantitative evaluation of interactions through competition and trophic relationships (Traas et al. 2004; Baird et al. 2001; De Laender et al. 2008b and 2008c). However, such models rarely include important density-dependent processes which are necessary to include if realistic results are to be obtained. Unfortunately, these processes are typically poorly quantified.

The following sections highlight two VEC component types that can be used to describe the input parameters required for many modeling characterizations.  The selected species represent crustaceans (Calanus spp) and small fish (Boreogadus saida), important components of the Arctic food web.  The types of information required for each component is similar for various species that might be selected for other compartment VECs.  Population models such as those described below can be used to extrapolate the data on the incidence of affected individuals obtained from laboratory exposures to estimated population responses (e.g. expressed as the population growth rate). Population models that relate individual level responses to changes in population size and structure have the potential to add more realism to ecological risk assessment (Forbes et al. 2008; Barnthouse et al. 2007).

7.2 Knowledge Status

7.2.1 Parameters Needed to Assess Potential Responses of VECs to Environmental Stressors

7.2.1.1 Transport and fate / exposure potential

Models have been developed that adequately address the transport and fate of surface oil spills (e.g. ASA 2011 and McKay 2009).  These models take into account the effects of wind and surface currents and are useful predictions of the movement of oil at the surface and into the water column.  They predict the concentrations of oil within a volume of water which permits the comparison of those exposure values to toxic effects based measurements of adverse effects observed in laboratory studies (Gallaway et al., in review). When oil is released at the surface of the water or at depth it begins to spread, thin and dilute into 2 or 3 dimensions.  For surface spills the oil begins to spread primarily in a horizontal direction based on currents and wind-driven transport with the more soluble components diluting into the third or vertical dimension, which includes the water under the surfaced oil and release of the more volatile components to the atmosphere. Subsurface releases generally have an initial momentum that provides rapid dilution by entrainment of surrounding waters during the initial rise of the fluids from the bottom, driven by any jet turbulence from the simultaneous release of natural gasses and the natural buoyancy of the oil.  This rapid dilution is followed by continued vertical transport based on the specific gravity or density and size of oil droplets created during the initial release including entrained diffusion of the more soluble components into the water column accompanied by horizontal dispersion by subsurface currents.  Other processes that occur and which may influence the bioavailability of oil components include biodegradation processes and physical/chemical changes that may occur due to pressure changes (e.g. clathrate formation) or emulsion formations by incorporation of water into the oil mass.  In addition to the physical contact of organisms with oil (e.g. fouling on birds) organisms also respond to the bioavailable chemical components that are present in water and food.

Another component of the fate evaluation for models is the change in chemical composition and bioavailability of oil that occurs during physical weathering and biological activity, mineralization and biodegradation.  The relative contribution and total concentrations of oil compounds changes through time and the rate of change varies depending on the environmental compartment in which the oil resides.  Surface oil that emulsifies undergoes less efficient biodegradation and appears to be less available for biological uptake into tissues than oil that is dispersed into the water column as small droplets.  Oil that is stranded on beaches or within interstitial waters between grains of sediment also undergoes slower rates of biodegradation as demonstrated by the ‘lingering oil’ effects seen in Prince William Sound decades after its stranding.  These important concepts are further developed in Sections 3 and 5 (fate of oil and biodegradation processes, respectively).

The transport and fate of a released quantity of oil has the potential to affect the population abundance, standing crop/production, increase daily mortality coefficients, and response to stressor exposure within different environmental compartments.  Variation in these attributes for each environmental compartment are expected and assessments can be generated based on assumptions that are meant to provide a maximum level of effect using the most conservative estimates or to best characterize the distinctions between the compartments.

7.2.1.2 Oil toxicity evaluations / sensitivity

The bioavailable fractions of oil derived from the soluble compounds that are present during the initial release of the oil or in locations where oil becomes re-concentrated (e.g. at convergence zones, pycnoclines, shorelines, air/water and ice/water interfaces).  As dilution of these compounds and biodegradation processes occur the relative and total concentrations may be reduced.  Eventually, the concentrations of the soluble compounds decreases to the point that the acute effects may be replaced by more subtle, sublethal doses resulting in longer term exposure to lower but more constant concentrations of the remaining compounds.  Measurements made during accidental and experimental spill events have shown that the higher total petroleum hydrocarbon (TPH) concentrations observed in the upper 3 m of the water column range from ~2 to 37 mg/L for physically dispersed oil and between 36 and 527 mg/L for chemically dispersed oil (Trudel et al., in press; Brandvik et al. 1995).  The chemical dispersants introduced more oil into the water column but the effects based concentrations for measured petroleum compounds were less per unit of oil for the chemical dispersions (Gardiner et al., 2013).  These field evaluations also showed reduced TPH concentrations of 50% within an hour and approximately 2 orders of magnitude differences in concentration within the upper 10 m of the water column for both physically and chemically dispersed oil. 

Empirical studies of the acute toxicity of oil and its components or modeling the effects of complex mixture exposure and uptake are alternative methods for estimating the effects of oil in different compartments.  For acute responses to oil exposure, including mortality and avoidance behaviors toward soluble components, toxicity testing is appropriate.  The exposure regimen for this type of assessment has been developed to emulate field conditions observed during spill events and controlled experimental spill events that were described in the previous paragraph.  The methods employ spiked concentrations of oil followed by dilutions with clean water to simulate the dispersion/dilution processes that occur in the field.  There are a significant number of experiments that have been performed using this protocol and it provides a good characterization of the acute effects that might be expected from dynamic exposure concentration from an oil spill at the surface or a depth (refer to Section 6 on Ecotoxicology for further toxicological information). The longer term, chronic and sublethal responses of organisms can be modeled by using constant exposure to lower concentrations of oil that might occur with organisms drifting with the oil after the initial dispersion or those inhabiting contaminated sediments.  There are also models that have been built to estimate the potential effects of the uptake of the bioavailable components into the tissues of organisms resulting in chronic responses to oil components.  These models are based on the relative narcosis effect of many PAH compounds (Di Toro et al. 2007).

7.2.1.3 Population distributions, stressors, and mortality rates

Modeling population responses that can occur as a result of exposure to oil should include a representation of the temporal and spatial distribution of age classes of the VECs.  The spatial evaluation needs to take into account the vertical and horizontal distribution of the abundance of each VEC and its relationship to key environmental components.  Examples of environmental components include river discharges, lagoon occupation, annual and permanent ice, propensity to cluster at zones of convergences (e.g. currents, pycnoclines, upwelling and downwelling) and different types of shorelines and bottom types for each of the VEC age classes.  The populations of VECs undergo different natural stressors within each of these habitats ranging from environmental exposure conditions, inter and intra-annular differences in environmental conditions, differences in spawning success and variable predation rates.  As a result, the influence of an additional stressor will influence populations at different rates.  These natural and ongoing impacts produce natural mortality rates that can be determined.  For fish the daily mortality rates can be determined by examining daily growth rings on otoliths and the relative abundance of individuals at comparable ages.  Invertebrates generally lack these types of structures and their mortality rates are based on the relative abundance of different size classes within a population.  Once the natural mortality rates are estimated, the impact of another stressor can be added to determine what the combined mortality will be. 

For example, a recent assessment of the acute effects of chemically dispersed oil into Arctic pelagic open ocean, ice-free waters on larval Arctic cod (Boreogadus saida) was modeled to determine the adult female equivalents that would be lost as a result of a large oil spill (Gallaway et al., in review).  The fisheries model obtained daily mortality rates from otolith assessments and applied toxicity information obtained by Gardiner et al. (2013) and assumed that a large oil spill would be at the center of the maximum population sizes observed.  Comparing the increased mortality rates of the oil indicated that number of adult female equivalents impacted by the oil were a fraction of the natural range in abundance of this species.  Based on these results the resiliency of the population of Arctic cod would be sufficient to recover during the following spawning year.  Had the oil not been treated with dispersants it would have been transported toward shore where juvenile Arctic cod are present in very large schools numbering in the millions, and the Arctic cod population would be subjected to greater stress. 

The following sections include technical information on copepod, amphipod, and fish population ecology and discuss the types of information necessary to assess the potential population effects of oil on two groups of species, copepods and fish. 

7.2.2 Copepod Population Ecology

Copepods are a subclass of small crustaceans found in marine and freshwater habitats throughout the world, including Arctic and sub-Arctic waters.  Copepods feed on phytoplankton or other zooplankton making them the dominant secondary producer present in the water column.  Their health and abundance can act as a bottom-up control on entire ecosystems, including specific effects on commercially important fisheries such as cod (Beaugrand et al. 2003) as well as populations of marine birds and mammals.  This section presents a broad overview of some aspects of the population ecology of copepods in the Arctic and sub-Arctic including: preferred habitat, life cycle and development, species present, and their respective populations. Each of these descriptions can be used to evaluate the potential impacts to copepod populations after an oil release.

7.2.2.1 Copepod Growth and Development

Copepods are divided into ten orders, but only cyclopoida, poecilostomatoida, and calanoida are common as plankton.  The order calanoida includes some of the most dominant species found in the Arctic in terms of abundance and biomass.  Calanoid copepods all follow the same life stages from egg to adult organism as follows:

  • Egg: Newly laid eggs are small capsules that sink to the bottom and swell until they become spherical.  The number of eggs laid by females of different species is variable.
  • Naupliar stages: The postembryonic development stages are referred to as naupliar stages.  Six such stages (Stages I through VI) exist for calanoid species.  The naupliar stages are characterized by the use of the antennae for swimming and the appearance of a first, unpaired eye.
  • Copepodite stages: A copepodite larva has two pairs of swimming appendages and develops a hind body comprised of the thorax and abdomen.  Five copepodite stages exist (Stages I through V). Molting of the outer exoskeleton marks the transition between each stage.
  • Adult: The adult stage is reached with the development of gonads in males and ovaries in females, with no further molts.

There are common factors such as water temperature and food availability that dictate the growth rate of all copepods.  There are also specific habitat preferences and adaptations that result in varying growth rates for the different species.  Three representative calanoid species, Calanus finmarchicus, Calanus glacialis, and Calanus hyperboreus, are used to illustrate how habitat preferences and seasonal variations in Arctic conditions influence copepod life cycles. Understanding the duration of the copepod life cycle and how their populations shift vertically in the water column in response to food availability can help to inform when a population might be more vulnerable to an oil spill incident.

All species depend on the seasonal production cycle.  The increase of light intensity and melting of sea ice in the Arctic spring produce nutrient rich waters that establish the conditions for a large bloom of phytoplankton.  The carbon fixed by photosynthesis is rapidly stored as lipids by the three Calanus herbivores.  At higher latitudes this cycle of primary production becomes shorter and delayed farther into the summer because of less intense light and differences in ice coverage.  These differences can be significant.  In ice free areas of the SW Barents Sea, the primary production period extends from April through July.  This is followed by continued, but diminished, phytoplankton production into November.  By comparison, the sea ice may not fully melt until July in the in the Kara Sea.  The primary production period extends until mid-September, with reduced production continuing for another month.  Even these time frames are approximations, as there are huge variations in ice cover over time (Falk-Petersen et al. 2009).

Spawning time depends on the species, but occurs before or during the phytoplankton bloom.  Once the eggs have hatched the nauplii rapidly progress through the six stages into copepodites.  For Arctic species the annual bloom ends before copepodites develop into adults.  Copepodites of each species overwinter at depth in a form of hibernation referred to as diapause.  Figure 7-1 presents a summary of the each species response to the spring bloom and corresponding winter diapause.  The transition through the copepodite stages and into adult male and female organisms is marked on the figure.  The following paragraphs detail some species-specific developmental characteristics.

Calanus finmarchicus:  The geographical range of this species is centered in the Norwegian and Labrador Seas as well as in the Barents Sea south of the polar front.  However, it is frequently observed in the Arctic Ocean as an expatriate species (Kosobokova et al. 2011).  C. finmarchicus has a one year life cycle where it spawns at the maximum of the phytoplankton bloom.  Copepodites develop into lipid rich stage V by June-July and then descend to 200 to 1600 meters to undergo diapause.   The stage V copepodites begin to develop into males and females around January.  This process continues through into March when they ascend to the surface waters (Falk-Petersen et al. 2009).

Calanus glacialis:  This is a shelf species that lives in waters that can be ice covered into the summer months.  C glacialis typically has a two year life cycle (Figure 7-1) that begins with spawning during the spring of its third year.  Spawning takes place before or during the Arctic bloom, and likely requires additional energy input from ice algae or fecal pellets of ice dwelling organisms (Sampei et al. 2009). 

Figure 7-1. Upper and lower lines span the depth distribution range. (Source: Falk-Petersen et al. 2009)
Figure 7-1. Upper and lower lines span the depth distribution range. (Source: Falk-Petersen et al. 2009)

Copepodite stage III is reached in the first summer.  Development into stage IV and V and then to adult females require significant lipid reserves and is less likely to occur in the first year.  Instead, the stage III copepodites enter diapause and resume growth with the following bloom.  Once stage V is reached, the copepodites overwinter and develop into females for the following seasons spawning (Falk-Petersen et al. 2009).

Calanus hyperboreus:  These are a large copepod species and one of the Arctic’s key grazers.  Spawning occurs in winter and is fueled by internal lipid reserves.  After the phytoplankton bloom, eggs develop rapidly into stage II and III copepodites.  C. hyperboreus enter diapauses at depths of 800 to 1500 meters as either stage III, IV, or V copepodites.  Over the next two summers, the copepodites grow rapidly through stage IV and into stage V. 

After reaching stage V, the copepodites develop into adults and spawn.  Adult females and stage V copepodites overwinter at different strata.  Females are shallower at 200 to 500 meters where they shed eggs.  Life spans of 1-2 years and 4-6 years have been suggested for C. hyperboreus depending on geographic region and food availability (Falk-Petersen et al. 2009).

In addition to the seasonal migration associated with diapause, copepods can also exhibit a diel migration in an effort to avoid predation.  There is some debate about the extent of this behavior in the Arctic, given both the ice cover and long summer days (Fortier et al. 2001).  However, even when it does occur, the diel migration is not of such a large distance that it significantly changes the vertical position of a copepod population in the water column.

7.2.2.2 Summary of Arctic and Sub-Arctic Copepod Species

Though thousands of copepod species exist throughout the world, relatively few are present in Arctic environments.  Perhaps the best summary of Arctic zooplankton (including copepods) is presented by Kosobokova et al. (2011).  In this study, the authors compiled the results of zooplankton surveys conducted throughout the Arctic Ocean from 1975 through 2007.  A total of 134 locations were sampled at depths ranging from 0 to 3,000 m.  Altogether, 174 different metazoan plankton species were identified from these samples, 91 of which were copepods.  Each species was listed along with their presence or absence in various basins of the Arctic Ocean and their preferred depth range in the water column.  When possible, reproductively active females, eggs, and individuals in the naupliar and copepodite stages were also classified by species.  If some combination of the former were identified, it was considered evidence of a reproducing population in the Arctic.  Species present but not part of reproducing populations were considered expatriates.    Almost 20 percent of Arctic copepods were expatriates.  Six species entered from the Atlantic Ocean, four from the Pacific Ocean, and eight were neritic (Kosobokova et al. 2011).

Arctic copepods were also sorted by their geographic home ranges as a means of determining which species were endemic.  This geographic breakdown is presented in Figure7- 2. Thirty one percent of the identified species were endemic to the Arctic Ocean, including the cryopelagic and newly described Arctic copepods.  The cryopelagic copepods were Jaschnovia tolli, J. brevis, and Eurytemora richingsi.  These three species are most abundant in association with the ice, but occurred with enough frequency at stations where ice was absent for them to be counted as pelagic transients (Kosobokova et al. 2011).

Nineteen percent of the Arctic copepod species could also be found in the North Atlantic compared to only one percent from the North Pacific.  A larger percent of species were present in the North Atlantic and the North Pacific (10 percent) as well as species that were more widely distributed throughout the globe (28 percent). 

The shared distributions of species from the North Atlantic and North Pacific, as well as the presence of expatriate species, are important because it demonstrates the large scale movement of copepods into the Arctic Ocean.  Pacific copepods can enter the Chukchi Sea through an influx of species via the shallow Bering Strait.  In the Atlantic, transport occurs from the Norwegian and Greenland Seas through the Fram Strait, over the shelf of the Barents Sea, and into the Arctic Ocean (Kosobokova et al. 2011).  Although there may be shifts in specific species, copepods in general serve a functional role in the Arctic trophic web (Figure 7-2).

Figure 7-2. Percent of copepods from various geographical ranges out of a total of 91 species. AO = Arctic Ocean, NA = North Atlantic, NP = North Pacific [Source: Kosobokova et al. 2011]
Figure 7-2. Percent of copepods from various geographical ranges out of a total of 91 species. AO = Arctic Ocean, NA = North Atlantic, NP = North Pacific [Source: Kosobokova et al. 2011]

7.2.3 Copepod Populations

The Kosobokova et al. 2011 species summary does not include the abundance of individual copepod species or a spatial distribution of copepod abundance throughout the Arctic Ocean.  A rough estimate from their study implies an abundance of ~1,100 individuals per m3 in the surface layer (0-25 m).  The overall abundance of zooplankton declined exponentially with depth from the surface layer down to the deepest layer (2000-3000 m).  Species diversity, as measured by Margalef’s richness, continually increased to a depth interval of 500-1000 m, and remained elevated to the deepest layer.

Community sampling conducted off Franz Josef Land (FJL) found a similar abundance of copepods at 1,300 individuals/m3 (Table 7-2; Dvoretsky and Dvoretsky 2011).  FJL is typically covered by ice year round, but recent warming trends in Arctic waters had made this region periodically accessible.  Two particularly warm years were 2006 and 2007.  Copepod sampling was carried out in each of these years with the intent to measure both abundance and biomass (as total carbon).  Differences in these measurements between the years were statistically evaluated and explained by inter-annual variations in salinity and temperature.  

Table 7‑2. Mean abundance and biomass with standard error of the six most represented copepod species caught off Franz Josef Land in 2006 and 2007.

SpeciesGeographic Range1AbundanceBiomass
(Individuals/m3)(mg Carbon/m3)
2006200720062007

Calanus glacialis

Abundant on Arctic shelf areas deeper than 50 m.

740 (± 190)

150 (± 38)

46 (± 11)

7.4 (± 1.6)

Calanus finmarchicus

North Atlantic species; expatriate to Arctic.

17 ( ± 5)

4.4 (± 1)

0.82 (± 0.23)

0.23 (± 0.045)

Calanus hyperboreus

Abundant in Arctic central basins.

2 (± 1)

7.7 (± 2)

1.4 (± 0.45)

4.7 (± 0.6)

Pseudocalanus minutus

Shelf species of boreal and Arctic waters.

160 (± 38)

65 (± 19)

0.71 (± 0.18)

0.2 (± 0.068)

Metridia longa

North Atlantic/Arctic species present on deep shelf areas

75 (± 27)

58 (± 32)

1.2 (± 0.45)

1.2 (± 0.59)

Microcalanus pygmaeus

Widely distributed species

33 (± 6)

29 ± (8.6)

0.023

0.068 (± 0.023)

Oithona similis

Abundant throughout Arctic; bipolar species.

130 (± 20)

530 (± 190)

0.011

0.011

Total

 

1300 (± 270)

1300 (± 320)

51 (± 12)

14 (± 2.4)

Notes:  Statistical significant (P<0.05) differences between 2006 and 2007 are bolded.  1Arctic Ocean Diversity Database http://www.arcodiv.org/watercolumn/copepod.html [Population Data Source: Dvoretsky and Dvoretsky 2011]

The five species with the highest abundance and the five species with the highest biomass are presented above.  A brief summary of the geographic range of these species is also included for reference. Seven species are actually listed because of the extreme differences in size between the copepods.  Even though Oithona similis was the second most abundant copepod, it contributed little to the total biomass.  Conversely, few Calanus hyperboreus were observed, but their larger size ranked them in the top five for total biomass.

There were significant differences in copepod abundance between the two years for several of the species, ranging up to a factor of five difference for Calanus glacialis.  Even with the difference in annual abundance, C. glacialis was the largest contributor to total biomass.  It comprised 91 percent and 53 percent of total copepod biomass in 2006 and 2007, respectively.  Total abundance was similar for both years, but there were statistically significant differences in total biomass due to the reduced abundance of the large C. glacialis and increased abundance of the smaller species, O. similis.

Interannual variability in copepod populations of this magnitude is not uncommon.  Long term monitoring of two species in the Gulf of Maine demonstrated that copepod abundance can vary by up to a factor of eight between years (Ji et al. 2012).  Frequent swings in abundance may impact food availability for the VECs relying on copepods as food source and these should be taken into account when modeling population impacts from oil spills.

7.2.4 Arctic Fish Population Ecology

Arctic waters are home to a variety of fish species, yet relatively little is known about their total diversity and abundance across the polar region.  In Canadian Arctic waters alone, 189 species representing 115 genera and 48 families have been documented.  An additional 83 species are known in Arctic waters adjacent to Canada (Coad and Reist 2004).   Taken together, this represents about 270 species, which compares well with other estimates (Mecklenburg and Mecklenburg 2009; Mecklenburg et al. 2011).

While there are several factors unique to the Arctic Ocean that contribute to these population uncertainties, the primary reason is the regions inaccessibility.  The seasonal and even continuous ice cover makes it difficult for scientists to employ standard sampling techniques, such as trawling.  The same limitation has prevented commercial fisherman from targeting the Arctic, resulting in no harvest records or by-catch data that could otherwise provide a rough characterization of fish stocks.  What is known is that the fish are a key component of the Arctic ecosystem, serving as consumers of the secondary production of zooplankton, and in turn becoming an important source of prey for marine birds and mammals.  Commercial fisheries exist in Barents Sea, which is warmer and more productive than higher latitude waters.  Even at these higher latitudes, Arctic fish have been harvested by humans. This fishing was historically conducted on a local scale at rates greater than the catches that were reported to the respective governments of the region (Zeller et al. 2011).  However, the overall removal rates from human fishery pressures were still low and fish stocks in Arctic coastal waters are thought to be relatively intact.

Climate change combined with increased anthropogenic presence and the additional associated environmental stresses in the region adds to the existing uncertainty in predicting the response of Arctic fish populations to a potential oil release.  The following section provides details on the population abundance, life cycle, and development of fish species that are of particular importance to the Arctic.  Additional details on the most representative species including geographic range, life span, and age at spawning are provided in Table 7-3. 

Table 7-3. Representative fish species of the Arctic Ocean.

Common NameScientific NameFamilyGeographic RangeSizeBiology

Arctic cod

Boreogadus saida

Gadidae

Arctic, circumpolar

40 cm

  • Cryopelagic or demersal down to 1383m
  • Occasionally occurs in large schools which attract predators; key food source for marine mammals and birds
  • Main consumer of offshore plankton
  • Mature at 2-6 years, spawns once per lifetime averaging 12,000 egg per female

Arctic cisco

Coregonus autumnalis

Salmonidae

Alaska, Atlantic, Eurasia

64 cm

  • Nerito/pelagic
  • Anadromous
  • Eats invertebrates and small fishes
  • Locally important fisheries and commercial fisheries

Arctic char

Salvelinus alpines

Salmonidae

Alaska, Atlantic, Eurasia

101.6 cm

  • Nerito/pelagic
  • Anadromous
  • Eats crustaceans and fishes
  • Important in native food fisheries, as well as commercial and sport fisheries

Greenland halibut

Reinhardtius hippoglossoides

Pleuronectidae

Circumpolar

120 cm

  • Epibenthic, occasionally pelagic. Surface to deep waters down to 2000m
  • Eats crustaceans, fishes, and squid
  • Increasingly important as a commercial fishery in the Eastern Arctic

Capelin

Mallotus villosus

Osmeridae

Circumpolar, Arctic but not present at high latitudes

25 cm

  • Nerito/pelagic shallows to 725m
  • Eats plankton, worms, and small fishes
  • Commercially important in areas
  • Major source of prey for Pacific cod, marine birds, and mammals
  • Mature at 2-6 years, high mortality rate after spawning

Glacial Eelpout (representative of eelpout)

Lycodes frigidus

Zoarcidae

Arctic Basins

69 cm

  • Endemic to Arctic
  • Deep-sea species, endemic to Arctic basins from 475 to 3,000m
  • Eats fishes, cephalopods, mollusks, and other abyssal fauna
  • Spawning likely occurs at great depths, fall or winter spawning likely

Arctic staghorn sculpin

Gymnocanthus tricuspis

Cottidae

Arctic, circumpolar

30 cm

  • Benthic, shallow water close to shore down to 450m
  • Eats polychaetes, gastropods, euphausiids, amphipods
  • Prey for other fishes including Arctic and Atlantic cod
  • Matures at 5-6 years, spawns in late autumn to winter
  • Females produce 2,000-5,500 eggs

Arctic alligatorfish

Ulcina olrikii

Agonidae

Alaska, Atlantic, Eurasia (not present in Barents Sea)

8.6 cm

  • Benthic, at depths of 7-520m
  • Eats amphipods, polychaete and nemertine worms, ostracods
  • Prey for halibut and other bottom fishes

Sources: Coad and Reist 2004; Mecklenburg and Mecklenburg 2009

7.2.4.1 Arctic Fish Species Diversity

The most representative fish species are those with the widest geographic range and greatest abundance.  In many cases the geographic range is better understood than abundance for Arctic fish.  Determining the relative size of fish populations from the literature can be a difficult task.  First, there is not a lot of data available.  Second, when data is available, there are often inconsistencies between studies in how the absolute numbers of fish are reported and what the numbers represent.    For example, catch records indicate the tonnage of fish caught are very different from those converted to Catch Per Unit Effort (CPUE), which in turn are different from studies that simply report the total number of individuals caught.   Given these discrepancies, the fish species mentioned in this section were selected based on qualitative and quantitative data available from the reviewed literature. 

The most comprehensive list of Arctic fish species reviewed was the “Annotated List of the Arctic Marine Fishes of Canada” by Coad and Reist (2004).  This document was a preparatory step towards a book describing the Arctic marine fish species of Canada.  The geographic area covered by this document is extensive, encompassing nearly all types of Arctic marine habitat. This document was the product of an extensive compilation of published and unpublished literature as well as museum collections and field studies.  All species present in this report were ranked by number as a means of determining the relative prevalence of a given fish in the Arctic.  The numbers do not represent counts of fish, but rather the number locations where a species has been observed.  Rankings extend from rare (1-2 records) to abundant (500+ records) and very abundant (2,000+ records). 

For the purposes of this review, the most abundant fish species are the most relevant, since they are most likely to be impacted by an oil release.   A total of fourteen species were listed as abundant, while only four were considered very abundant.  The species are listed below.

  • Abundant: Pacific herring (Clupea pallasii), Capelin (Mallotus villosus), Rainbow smelt (Osmerus mordax), Cisco (Coregonus artedi), Arctic cisco (Coregonus autumnalis), Lake whitefish (Coregonus clupeaformis), Least cisco (Coregonus sardinella), Roughhead grenadier (Macrourus berglax), Greenland cod (Gadus ogac), Arctic staghorn sculpin (Gymnocanthus tricuspis), Shorthorn sculpin (Myoxochphalus scorpius), Arctic alligatorfish (Ulcina olrikii), Gelatinous snailfish (Liparis fabricii), Arctic flounder (Pleuronectes glacialis)
  • Very Abundant:  Arctic char (Salvelinus alpines), Arctic cod (Boreogadus saida), Fourhorn sculpin (Myoxocephalus quadricornis), Greenland halibut (Reinhardtius hippoglossoides)

While Coad and Reist (2004) was very comprehensive and provides a sufficient list of representative species, the following paragraphs highlight studies conducted in areas outside the Canadian Arctic and provide additional background about Arctic fish populations. 

Historical fish harvests in the Arctic have been limited to small scale fisheries.  The fish caught typically went towards feeding indigenous peoples or sled dogs prior to the advent and adaptation of snowmobiles.  Zeller et al. 2011 reconstructed catch data for fishing communities throughout coastal areas of the Arctic for the years 1950 through 2006.  Their goal was to create a more accurate record of fish catches than was available from government reported data.  As such, this study is biased towards shallow and nearshore areas and the anadromous fish that inhabit these areas. 

Reconstructed catch records were reported for Russia, the United States, and Canada.  Russia was the largest harvester of fish during the selected period.  Areas surveyed in Russia included the Kara, Laptev, and East Siberian Seas.  Within these seas, the reported tonnage of the sardine cisco and Arctic cisco dominated the totals, with additional contribution from other Coregonus spp.  Catches of the long lived and slow growing Siberian sturgeon (Acipenser baeri) were recorded early in the survey timeline for the Kara Sea, but declined by the late 1960’s (Zeller et al. 2011).

Fish catches were also evaluated for Alaska and Canada.  The primary species caught along Canada’s Arctic archipelago and into Hudson Bay was the Arctic char, which accounted for around 90 percent of the total catch for both areas.  Within the Beaufort Sea, combined Alaskan and Canadian catches were dominated by Coregonus spp. and Dolly varden (Salvelinus malma).  Alaskan catches in the Chukchi Sea were primarily chum salmon (Oncorhynchus keta) with contributions from sheefish (Stenodus leucichthys) and Dolly varden (Zeller et al. 2011).

A demersal trawling survey conducted in 2008 in the U.S. portion of the Beaufort Sea offers a broader look at some of the species present in the Arctic (Rand and Logerwell 2011).  This was the first such study in the Beaufort Sea since 1977.  Twenty two successful trawls were completed at bottom depths ranging from 40 to 470 meters.  Overall, fish represented six percent of the total weight captured in the trawls, with the remainder consisting of invertebrates.  Arctic cod was the dominant fish species, comprising 92 percent of the total number of fish and 80 percent of the total weight.  The CPUE was in the range of 25.8 to 58.6 kg/ha across the longitudinal range of the deeper water survey area, indicating consistent density of this species.

The next most abundant demersal fishes of the Beaufort Sea included the eelpouts (Lycodes spp.), Bering flounder (Hippoglossoides robustus), and walleye pollock (Theragra chalcogramma).  The presence of walleye pollock and to a lesser extent Pacific cod (Gadus macrocephalus) suggests a possible extension of the known range of these species.  However, it should be noted that fish of spawning age or size were not found in this survey.  Both are commercially valuable species more commonly associated with the Bering Sea to the south.  While it is possible that their range is expanding, the limited time series of data available for the region was not sufficient to make this determination.

The Barents Sea is the southernmost extension of the Arctic Ocean.  Its southern location, geomorphology, and relatively warmer water inputs from the Atlantic Ocean have resulted in populations of commercially and ecologically valuable species not present in the high Arctic.  These species include haddock (Melanogrammus aeglefinus), saithe (Pollachius virens), Atlantic cod (Gadus morhua), Atlantic herring (Clupea harengus), and capelin.  The Barents Sea contains habitat for all life stages of these species, including spawning, nursery areas for larvae and juveniles, and feeding grounds for adults (Olsen et al. 2009).

A series of 257 demersal trawls conducted in the year 2000 reveal that while Atlantic cod, saithe, and haddock were the most abundant species in the Barents Sea, a wide diversity of non-commercial fish were also present (Byrkjedal and Hoines 2007).  A total of 58 species were identified in the trawls and grouped by family and species.  Excluding the cod in family Gadidae, species from Scorpaenidae, Cottidae, Stichaeidae, and Zoaridae were among the most abundant.  Representative members from these three families include the deep water redfish (Sebastes mentella), mustache sculpin (Triglops murrayi), snake blenny (Lumpenus lampretaeformis), and eelpout (Lycodes spp.), respectively.

Using statistical analysis, Byrkjedal and Hoines 2007 evaluated the geographical observations of the identified species against salinity and temperature.  Not surprisingly, they found that the distribution of most species was related to the location of the polar front, the zone where warm water from the south meets subzero polar water.  Gadidae and Scorpaenidae were found in warmer waters, while Cottidae, Zoaridae, and Stichaeidae species were found from the polar front northwards (Byrkjedal and Hoines 2007).

7.2.4.2 Representative Fish Species

Selection of representative Arctic fish species is based on a variety of factors including importance as commercial species, importance to the Arctic ecosystem, and overall abundance.  However, inclusion in the list presented in Table 7-3 does not necessarily mean the species is suitable for population effects modeling of an oil release.  Not all of the listed species would be equally affected by an oil release.  Nor would a population decline of some of the species have the same effect on the higher levels of the Arctic ecosystem.  The best species to use in a population effects model should meet a series of criteria:

  • The species should be abundant in the Arctic and be present over a wide geographic range.
  • The species should have an important position in the Arctic food web as both a main consumer of plankton and a major food source for marine birds and mammals.
  • The species would preferably occupy various zones of the water column.
  • The species should spawn and develop in the Arctic Ocean, where the more sensitive and less mobile eggs and larvae are exposed to oil.

Of the species listed in Table 7-3, Arctic cod best meets these criteria for use in modeling.  Capelin are similar to Arctic cod in many respects in that they have a wide circumpolar distribution and similar life history, but are not present at the high Arctic latitudes.   Anadromous species are only present in the Arctic as adults, and could more easily avoid contaminated habitat.  The remainder of the species are mostly demersal, and aren’t as important of a food source to marine birds and mammals.  In addition to these environmental factors, a large body of oil and dispersant toxicity data is available for Arctic cod, making them a frequent key species in models (Gallaway et al. in press). 

7.2.5 Application of Population Models

This section provides several examples of population models that have been used to predict population level effects combining life history and toxicity data. 

The effects of oil releases on Barents Sea stocks of Atlantic cod were evaluated (Johansen et al. 2003; Hjermann et al. 2007).  Atlantic cod have eggs and larvae that are concentrated in the upper 10 m.  Population mortality was estimated based on oil spill simulations for different scenarios and different toxicities of the dissolved oil. Mortalities were predicted for 4 km2 grids or cells over a portion of the Barents Sea using oceanographic and spill simulation data and life history data for the respective cells, as well as no effects concentrations for the eggs and larvae of 900 ppb based on a review of laboratory studies of fish, invertebrates and larvae.  Johansen et al. (2003) also used a safety factor of 10, to set the PNEC at 90 ppb.  Based on simulations, Johansen et al. (2003) predicted losses in the region of 5 %, and in some cases up to 15 %, for a blowout lasting less than 2 weeks, while very long-lasting blowouts could give losses of eggs and larvae in excess of 25 %.  A 20 % loss in recruitment to the cod population is estimated to cause a 15 % loss in the cod spawning biomass and to take approximately eight years to recover fully. The population impact is to a large degree dependent on whether high oil concentrations in the water column were coincident with high egg and larvae concentrations, which are generally only be concentrated at the surface during calm seas over a period of weeks or months.  Hjermann et al. (2007) further evaluated the life history of the Barents Sea fish populations and oceanographic models to refine effects estimates using Monte Carlo simulations.  Data for Barents Sea stocks of Atlantic cod, herring and capelin were used, as well as data on the variation of oceanographic and ecological conditions.  Hjermann et al. (2007) found that the variability within the system overwhelmed the models ability to predict long-term effects.

For the Beaufort and Chukchi systems, population models were coupled with field-collected data on the distribution, abundance, natural mortality and fecundity of Arctic cod (B. saida) to predict population level effects associated with modeled exposure concentrations (Gallaway et al. in prep).  Exposures were simulated using the Integrated Oil Spill Impact Model System (SIMAP) to allow for different spill and response scenarios to be evaluated.  Toxicity data from laboratory exposures of Arctic cod to WAF and CEWAF, field collected data from the western Beaufort Sea, and the SIMAP model simulations to predict potential population effects from physically and chemically treated oil in open water. A simulated release of 100,000-ton of Alaska North Slope oil (ANS) treated with dispersants resulted in a release of oil into the pelagic waters, creating a toxic volume of 266 million m3 as opposed to a volume of 71 million m3 for a 100,000 ton spill of ANS when not treated with dispersants. Dispersants rapidly transports large quantities of the oil from the surface into the water column. A 100,000-ton spill treated with dispersants resulted in estimated losses of about 2 million Arctic cod larvae. On the order of 87 million eggs would have been required to produce 2 million juveniles. This level of egg production represents the reproductive output of about 7,300 adult females. The population of Arctic cod age 1+ has been estimated to range between 2 and 4 billion fish. These and other data presented herein suggest that the abundance of adult Arctic cod (age 3) in the Alaska Beaufort number in the 10’s to 100’s of millions. For the time and place evaluated (mid-shelf Alaska Beaufort Sea, mid-August) even a large spill dispersed within the maximum concentrations of larval cod appeared unlikely to have the potential for major impacts on Arctic cod populations.

Integrated modeling systems allow life history and effects data from multiple trophic levels to be incorporated into predictions of population effects.  The SYMBIOSES model is an example of a complex predictive system connecting seven individual models into a computational framework contained within a supercomputer (www.symbioses.no).  It is designed to simulate marine ecosystem components in three dimensional space and time and to predict impacts from selected combinations of sea based activities, i.e. fisheries and petroleum operations. The SYMBIOSES modeling system targets selected marine ecosystem components and maritime activities and allows prediction of population level changes in key commercial fish species linked to the Lofoten/Barents Sea region off of Norway. Ecological models are used to simulate the distribution and behavior of different life stages of important commercial fish, zooplankton, zooplankton prey, and phytoplankton species. Ecotoxicology models simulate chemical uptake and effects on growth, mortality, and reproduction of marine organisms. An oil transport and fate model is used to trace chemicals in the marine environment and a physical model simulates the hydrodynamic features of the region. The SYMBIOSES aim is to become a vital aid to impact and risk analysis, management, decision-making and stakeholder communication for the Lofoten/Barents Sea region.

7.3 Future Research Considerations

The review of population effects modeling described by the authors in this section led to suggestions of further research which can reduce remaining uncertainties.  The following areas of work will improve the ability to predict long term consequences of OSR residuals and reduce the total ecological effect of an oil spill.  The more generic suggestions compiled from this review are summarized below while recommendations that are important for improving Arctic NEBA are listed separately. 

  1. VECs.  Obtain a better understanding of the vertical and horizontal spatial patterns of VEC species/age class abundances.  Baseline surveys conducted as part of environmental impact assessments at oil and gas production sites in the Arctic will help fill this knowledge gap on a regional basis.  It will be important to synthesize this information and augment currently available data with full season baseline information.  It is also advisable that the data output become open-source materials.
    1. Develop spatial and reproductive population metrics for VECs (other than polar cod and copepods).  Expand information on species/populations that have minimal population/age structure/reproductive information available. These efforts should, for instance, include the capelin (Mallotus villosus), the Arctic Cisco (Coregonus autumnalis) for near shore environments and the Pacific and Atlantic herring (Clupea spp) for rocky intertidal and open ocean SML environments.  Other phyla such as mammals and seabirds should also be investigated, especially in terms of resiliency and recovery potential.
    2. Seasonality:  Establish seasonal distribution patterns of principal VECs including seabirds and marine mammals; concentrate on their seasonal use of ECs
      • Increase understanding of the offshore feeding and resting habits of birds.
      • Identify EC usage by VEC populations on a seasonal basis – use the EC as a surrogate for the potential presence/absence of VECs
      • Determine population dynamics information for EC usage leading to better quantification of population dynamics for marine mammals.
      • Augment ESIs for nearshore and offshore marine ecosystems along the global Arctic areas of interest.
  2. ECs.  Identify key ECs and features of the Arctic that provide more valuable habitat for these VEC species/age classes.  Concentrate on key fish and crustaceans that we already have useful data including the Arctic cod, Boreogadus saida, the sculpin, Myoxocephalus spp, and the calanoid copepods C. glacialis and C. finmarchicus.
  3. OSR residuals.  We have significant information on the toxicity of dispersants and physically and chemically dispersed oil in open water, pelagic environments that indicate that Arctic species have similar acute responses to measured concentrations of oil; however, there is less information on the toxicity of oil residuals from in-situ burning processes.  OMA treated oil greatly reduces the quantity of oil on the surface of the sea but the physical and chemical effects that might occur on bottom communities is less well known.

7.3.1 Priority Recommendations to Enhance NEBA Applications in the Arctic

The recommendations presented below indicate where increased knowledge of population modeling 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. VECs. Obtain more information on population dynamics and age-related characteristics of populations that are key components of food webs leading to VECs. 
    1. These characteristics need to be refined by the seasonal EC occupancy and include changes in mortality coefficients, age/size population structure, variation in abundance or standing crop of populations and the relative potential for recovery based on population or species resiliency characteristics.
    2. The key invertebrates for the under ice environmental compartment include the autochthonous ice fauna amphipods, Gammarus wilkitzkiiApherusa glacialisOnisimus nanseni and O. glacialis.  They are also important VECs for seabirds and marine mammals that forage in this part of the Arctic.  Although there is significant information on the distribution of these species in part of the Arctic, seasonal reproductive patterns, natural mortality rates and toxicity studies on different age classes would provide the information necessary for an assessment of the potential impacts on this important environmental compartment.  
  2. Evaluate oil encounter rates by VECs.  Obtain further information on direct effects of OSR residuals as they relate to resource encounter rates; this information will improve population-level impact assessments via the use of population models.  Compare the population abundance information for VEC species within each compartment to determine the potential exposed population. 
    1. Then compare the impact based on toxicology information on the exposed populations within each compartment (acute and chronic lethal and sublethal effects). 
    2. Relate the impact on these key species to the potential impact at the next link in the food web. 
  3. Formulate population and environmental compartment resiliency scores.  Develop a resilience scoring method that provides a semi-quantitative characterization of VECs ability for population recovery within various ECs and determination of forecast health/sustainability of the various ECs
    1. Factors to consider are: reproductive potential, species sensitivity, exposure potential (encounter rates), exposure reductions (biodegradation, evaporation, dissolution, etc.)
    2. Determine whether a 20% reduction in adults or juveniles and larvae is detectable within the natural variation of populations (i.e. would maintain a sustainable population size)
    3. Resiliency attributes may include:  fecundity, immigration/emigration of species or populations, population sizes, structural and functional community measures
    4. For example, for commercial fisheries assessment an indicator of sustainable harvest of adult females from a population is based on harvests that are less than recruitment overfishing thresholds that provide for harvests of ~20% of these adults (Roberts 2007).
  4. Modeling Improvements.  Augment modeling parameters to achieve overarching sustainability assessment (affected species/populations ability recover from damage).
    1. Incorporate population and compartment specific information for OSR residuals and comparison to resiliency and recovery scores.

7.4 Further Information

Authors Dr. Benny Gallaway (LGL), Dr. Jack Word [ENVIRON], Will Hafner (NewFields), Dr. James Clark (HDR/EM&A), Dr. Ivar Singsass (SINTEF), Dr. Oleg Titov (PINRO) 

References

Applied Science Associates (ASA)

2011

SIMAP software

On-line: http://www.asascience.com/software/ simap/index.shtml.

Avery DE,  Dam HG

2007

Newly discovered reproductive phenotypes of a marine copepod reveal the costs and advantages of resistance to a toxic dinoflagellate

Limno. Oceanogr.  52(5), 2099-2108

Baird DJ, TCM Brock, PC De Ruiter, ABA Boxall, JM Culp, P Eldridge, U Hommen, RG Jak, KA Kidd, T DeWitt

2001

The food web approach in the environmental management of toxic substances

In: Baird DJ, Burton GA,(eds), Ecological Variability: Separating Natural from Anthropogenic Causes of Ecosystem Impairment. SETAC, Pensacola, FL, USA, pp 83–122.

Barnthouse LW, Munns WR Jr, Sorensen MT (eds)

2007

Population-Level Ecological Risk Assessment

CRC, Boca Raton, FL, USA

Barrett RT, Bakken V, Krasnov JV

1997

The diets of common and Brṻmach’s guilemots, Uria aalge and U. longvia in the Barents Sea region

Polar Research 16:73-84

Beaugrand, Brand, KM, Lindley JA, Souissi S, Reid PC

2003

Plankton effect on cod treatment in the North Sea

Nature 426:661-664

Brandvik PJ, O Knudsen, M Moldestad, and PS Daling

1995

Laboratory Testing of Dispersants under Arctic Conditions

The Use of Chemicals in Oil Spill Response.  American Society for Testing and Materials; Philadelphia, PA.

Briceno J, JW McKee, JR Clark, DD Whiting

1992

Relative sensitivity of Gulf of Mexico species and national test species in acute toxicity tests with dispersants

In Thirteenth Annual meeting, Society of Environmental Toxicology & Chemistry; Cincinnati OH

Byrkjeda I, Hoines A.

2007

Distribution of demersal fish in the south-western Barents Sea

Polar Research 26:135-151

Calow P, Forbes VE

2003

Does ecotoxicology inform ecological risk assessment?

Environ Sci Technol 37:146A-151A

Chapman PM

2002

Integrating toxicology and benthic ecology:  Putting the eco back into ecotoxicology

Mar Pollut Bull 44:7-15

Coad BW, Reist JD

2004

Annotated list of the Arctic marine fishes of Canada

Canadian Manuscript Report of Fisheries and Aquatic Sciences 2674

Cross WE

1982

Under-ice biota at the at the Pond Inlet ice edge and in adjacent fast ice areas during Spring

Arctic 35:13-27

De Laender F

2007

Predicting effects of chemicals on freshwater ecosystems: model development, validation and application.

PhD thesis, University of Ghent, Belgium

De Laender F, De Schamphelaere KAC, Vanrolleghem PA, Janssen CR

2008c

Do we have to incorporate ecological interactions in the sensitivity assessment of ecosystems? An examination of a theoretical assumption underlying species sensitivity distribution models.

Environ International 34:390-396.

De Laender F, De Schamphelaere KAC, Vanrolleghem PA, Janssen CR

2008b

Validation of an ecosystem modelling approach as a tool for ecological effect assessments.

Chemosphere 71: 529-545.

De Laender, F., K.A.C.De Schamphelaere, P.A. Vanrolleghem, C.R. Janssen

2008

Comparison of different toxic effect sub-models in exosystem modeling used for ecological effect assessments and water quality standard setting

Ecotoxicology and Environmental Safety

Di Toro DM, JA McGrath, and WA Stubblefield

2007

Predicting the toxicity of neat and weathered crude oil: toxic potential and the toxicity of saturated mixtures

Environ Toxicol Chem. 26(1):24-36.

Duval WA, LA Harwook, RP Fink

1982

The sublethal effects of dispersed oil on an estuarine isopod

Environment Canada; Technol Devel report, EPS-4-82-1.  Ottawa, Canada

Dvoretsky VG, Dvoretsky AG

2011

Copepod communities off Franz Josef Land (northern Barents Sea) in late summer of 2006 and 2007

Polar Biology 34(8):1231-1238

Eiane K, Ohman MD

2004

Stage-specific mortality of Calanus finmarchicus, Pseudocalanus elongatus, and Oithona similis on Fladen Ground, North Sea, during a spring bloom

Marine Ecology Progress Series Vol. 268: 183-193

Falk-Petersen S, Mayzaud P, Kattner G, Sargent JR

2009

Lipids and life strategy of Arctic Calanus

Mar Biol Res  5:18-39

Forbes VE, Calow P, Sibly R

2008

The Extrapolation Problem and How Population Modeling Can Help

Environ Toxicol Chem 27:1987-1994

Fossheim M, Nilssen EM, Aschan M

2006

Fish assemblages in the Barents Sea

Marine Biology Research 2:260-269

French-McCay D

2009

State-of-the-art and research needs for oil spill impact assessment modeling

Proceedings, 32nd AMOP Technical Seminar on Environmental Contamination and Response Environment Canada, Ottawa

French-McCay DP

2011

Oil spill modeling for ecological risk and natural resource damage assessment

Proceedings, 2011 IOSC, Am Petroleum Inst. Washington DC May 2011  Paper 415

Fucik KW, KA Carr, BJ Balcom

1995

Toxicity of oil and dispersed oil to the eggs and larvae of seven marine fish and invertebrates from the Gulf of Mexico

In The use of chemicals in oil spill response; Lane P (ed); ASTM, Amercian Society for Testing and Marerials, PA. p 135-171

Gjertz I, Lydersen C

1986

The ringed seal (Phoca hispida) spring diet in NW Spitsberfen, Svalbard

Polar Research 4:53-56

Hardie DC

2004

Population genetics, life history, and ecology of Arctic marine fishes

Arctic;  57(4):444-448

Hendriks AJ, Enserink EL

1996

Modelling direct response of single species populations to microcontaminants as a function of species size with examples for water fleas (Daphnia magna) and cormorants (Phalacrocorax carbo)

). Ecol Modell  88:247-262

Hendriks AJ, Maas-Diepeveen JL, Heugens EH, van Straalen NM

2005

Meta-analysis of intrinsic rates of increase and carrying capacity of populations affected by toxic and other stressors

Environ Toxicol Chem 24:2267-2277

Hildebrandt N, Bergmann M, Knust R

2011

Longevity and growth efficiency of two deep-dwelling Arctic zoarcids, and comparison with eight other zoarcid species from different climatic regions

Polar Biology 34:1523-1533

Hjermann DO, A Melsom, GE Dingsor, JM Durant et al.

2007

Fish and oil in the Lofoten-Barents Sea system: synoptic review of the effect of oil spills on fish populations

Mar Ecol Prog Ser  339:283-299

Hu, Q., Petrik, C.M., Davis, C.S

2007

Normal versus gamma: stochastic models of copepod molting rate

Journal of Plankton Research 29(11)985-997

Hunt G, Megrey BA

2005

Comparison of the biophysical and trophic characteristics of the Bering and Barents Seas

Journal of Marine Science 62:1245-1255

Ji R, Stegert C, Davis C

2012

Sensitivity of copepod populations to bottom-up and top-down forcing: a modeling study in the Gulf of Maine region

Journal of Plankton Research (in press)

Johansen O, K Skognes, OO Aspholm, C Ostby & KA Moe

2003

Accidental oil spills - consequences in the water column.  (In Norwegian)

SINTEF Rept No. STF66 F03028/ULB 7-c Foundation SINTEF, Trondheim, Norway 72 pp

Kosobokova KN, RR Hopcroft, H-J Hirche

2011

Patterns of zooplankton diversity through the depths of the Arctics central basins

Marine Biodiversity 41:29-50

Liu H, Hopcroft RR

2008

Growth and development of Pseudocalanus spp. In the northern Gulf of Alaska

Journal of Plankton Research 30(8)923-935

Logerwell E, Rand K, Weingartner TJ

2011

Oceanographic characteristics of the habitat of benthic fish and invertebrates in the Beaufort Sea

Polar Biology 34:1783-1796

Lonne OJ

1992

Distribution, abundance and trophic role of the sympagic macro-fauna in Svalbard waters

Thesis – Norwegian College of Fisheries Sci.  Univ of Tromso

Lonne OJ, Gabrielsen GW

1992

Summer diet of seabirds feeding in sea-ice-covered waters near Svalbard

Polar Biololgy 12:685-692

Mecklenburg CW and T Mecklenburg

2009

Saffron cod: Eleginus gracilis.

In: Arctic Ocean diversity http://www.arcodiv.org/Fish/Elegi nus_gracilis.htm

Melnikov IA

1992

The Arctic sea ices

Gordon and Beach, Amsterdam.

Melnikov IA, Lulikov AS

1980

Cryopelagic fauna of the central Arctic basin

In: Vinogrudov, ME and IA Melnikov, ed.  Biology of the central Arctic Basin (Canadian Transations of Fishery and Aquatic Sci. 4910) Nauka, Moskow 97-111.

Miller CB, Lynch DR, Carloti F, Gentleman W, Lewis CVW

1998

Coupling of an individual-based population dynamic model of Calanus finmachicus to a circulation model for the Georges Bank region

Fisheries Oceanography

Moll A, Allen I

2011

D2. 10 Sub-model Copepod Structured population model including user guide

University of Hamburg

Nilssen Kt, Potelov HT, Timoshenko V

1995

Feeding habits of Harp seals (Phoca groenlandica) during summer and autumn in the northern Barents Sea

Polar Biology 15:485-493

Olsen E, Aanes S, Mehl S, Holst JC, Aglen A, Gjosaeter H

2010

Cod, haddock, saithe, herring, and capelin in the Barents Sea and adjacent waters: a review of the biological value of the area

Journal du Conseil 67(1):87-101

Pastorak RA, Bartell SM, Ferson S, Ginsburg LR (eds)

2001

Ecological Modeling in Risk Assessment

CRC, Boca Raton, FL, USA.

Pike D, Welch HE

1990

Spatial and temporal distribution of sub –ice macrofauna in the Barrow Strait area

Can J. Fish Aquatic  Sci.  47:81-91

Polterman M

1998

Abundance, biomass and small-scale distribution of cryopelagic amphipods in the FranzJosef land area (Arctic)

Polar Biol.  20:134-138

Polterman M

1997

Biology and ecology of cryopelagic amphipods from Arctic Sea ice

PhD Thesis, University Bremen Ber Polarforsch, Bremerhaven.  Pp. 1-170

Rand KM, Logerwell EA

2011

The first demersal trawl survey of benthic fish and invertebrates in the Beaufort Sea since the late 1970s

Polar Biol  34:475-488

Roberts C

2007

The unnatural history of the sea.

Island Press/Shearwater Books.  435 pp

Sampei, M., Forest, A, Sasaki H, Hattori H, Makabe R, Fukuchi M, Fortier L

2009

Attenuation of the vertical flux of copepod fecal pellets under Arctic sea ice: evidence for an active detrital food web in winter

Polar Biology 32:225-232

Skarohamar J., Reigstad M, Carroll J, Eliane K, Riser CW, Slagstad D

2011

Effects of mortality changes on biomass and production in Calanus spp. Populations

Aquatic Biology 12:129-145

Slagstad D

1981

Modeling and simulation of physiology and population dynamics of copepods. Effects of physical and biological parameters.

Modeling, Identification, and Control, Vol. 2. No. 3, 119-162

Souissi S., Ginot V, Seuront L., Uye S

2004

Using multiagent systems to develop individual-based models for copepods: Consequences of individual behavior and spatial hetergeneity on the emerging properties at the population scale

CRC Press Page 523

Swaim ZT

2008

Lipid metabolism by right whales: Using fecal samples to assess assimilation of copepod triacylglycerols and wax esters

Thesis submitted to the University of North Carolina

Terui T, Kishi MJ, Ueno H

2012

Lagrangian ensemble model of Copepoda (Neocalanus cristatus) in the northwestern subarctic Pacific

Journal of Oceanography 68:727-Jnl Oceanogr 68:727-741

Traas T, Janse JH, Van den Brink PJ, Brock TCM, Aldenberg T

2004

A freshwater food web model for combined effects of nutrients and insecticide stress and subsequent recovery

Environ Toxicol Chem 23:521-529.

Trudel BK, Belore RC, Mullin JV, and A Guarino

2010

Oil viscosity limitation on dispersibility of crude oil under simulated at-sea conditions in a large wave tank.

Mar Poll Bull. 60(9):1606-1614.

United States Testing Company

1994

Untitled4

Report No. 066234

van den Bosch, F. and W. Gabriel

1994

A model of growth and development in copepods

Limnol. Oceanogr. 39(7) 1528-1542

Van Straalen NM

2003

Ecotoxicology becomes stress ecology

Environ Sci Technol 37:324A-330A.

Werner I

1997

Ecological studies of the Arctic under ice habitat colonization and processes at the ice-water interface

Ber Sonderforsch ung bereich, Univ. of Kiel.

Zeller D, Booth S, Pakhomov E, Swartz W, Pauly D

2011

Arctic fisheries catches in Russia, USA, and Canada: baselines for neglected ecosystems

Polar Biology 34:955-973