4 Ecological Environmental Effects Assessment Methods for Deterministic Oil Spill Modelling

CA PDF Page 105 of 1038

Part B: Deterministic Modelling of the Ecological and Energy East Project Human Health Consequences of Marine Oil Spills Volume 24: Ecological and Human Health Risk Section 4: Ecological Environmental Effects Assessment Assessment for Oil Spills in the Marine Environment Methods for Deterministic Oil Spill Modelling

4 ECOLOGICAL ENVIRONMENTAL EFFECTS ASSESSMENT METHODS FOR DETERMINISTIC OIL SPILL MODELLING

4.1 Introduction

This section provides details for the approach to the deterministic assessment of physically or toxicologically induced changes in health of ecological receptors exposed to spilled crude oil, or chemical constituents of spilled crude oil. Results of the assessment are provided for three spill scenarios: (i) a grounding in Saint John Harbour; (ii) a collision in the Bay of Fundy; and (iii) a grounding south of Grand Manan) (see Sections 6, 7 and 8).

This ecological risk assessment (ERA) follows accepted risk assessment methodologies and follows guidance published and endorsed by regulatory agencies, including CCME (1996, 1997), Government of Canada (2012a) and the U.S. EPA (1989, 1998).

4.2 Ecological Risk Assessment Problem Formulation

Problem formulation is the initial information gathering and interpretation stage that focuses the assessment on areas of primary concern within the study area for marine accidents and malfunctions (SAMAM). The problem formulation defines the nature and scope of the work to be conducted, and enables practical boundaries to be placed on the overall scope of work, so the ERA is directed at the key areas and issues of concern. Key components of the problem formulation include:

• establishing spatial boundaries for the overall assessment • identifying crude oil products being assessed, which individual COPC are present, and mechanisms of release to the environment • identifying and characterizing representative ecological receptors • setting assessment and measurement endpoints for toxicological effects on ecological receptors • identifying exposure media and pathways by which ecological receptors may be exposed to COPC • defining benchmarks from which the magnitude of toxicological responses caused by exposure to COPC can be predicted

A conceptual site model (CSM) follows the problem formulation, which provides a visual depiction of the relevant pathways linking COPC in various environmental media and biota to the ecological receptors.

Energy East Pipeline Ltd. May 2016 4-1

CA PDF Page 106 of 1038

Part B: Deterministic Modelling of the Ecological and Human Health Consequences of Marine Oil Spills Energy East Project Section 4: Ecological Environmental Effects Assessment Volume 24: Ecological and Human Health Risk Methods for Deterministic Oil Spill Modelling Assessment for Oil Spills in the Marine Environment

4.2.1 Assessment and Measurement Endpoints

Suter (1993) defined assessment endpoints as explicit expressions of environmental values or characteristics to be protected at a site, reflecting societal and ecological values. In practice, assessment endpoints are usually broad statements articulating the overall goals of a risk assessment. For this risk assessment, overall goals are to determine the extent to which accidents and malfunctions that may arise from the Project, or shipping of crude oil as a result of the Project, have the potential to adversely affect marine plant, invertebrate or fish communities, marine birds or marine mammals. In this context, acute effects and acute toxicity are defined as those environmental effects that result from or appear after brief exposure (usually 96 hours or less) to a chemical stressor, often at relatively high concentrations. Chronic effects and chronic toxicity are defined as those environmental effects that result from or appear after prolonged exposure to a chemical stressor, typically at lower concentrations than those that cause acute effects.

The information needed to deal directly with these high-level assessment endpoints is often difficult to generate and rarely available. Therefore, measurement endpoints, which are simpler and more clearly defined responses to stressors and directly related to the assessment endpoints, are often used as practical metrics for ERA.

Measurement endpoints may be defined in terms of an unacceptable level of effect on ecological receptors, such as a certain relative percent decrease in survival, growth or reproduction of the test organisms used to represent ecological populations. As part of a weight-of-evidence approach, one or more measurement endpoints may be used for each assessment endpoint. Measurement endpoints can also be used as a starting point in the development of follow-up or environmental effects monitoring programs. The following are the measurement endpoints considered in this ERA:

• For marine community-level resources, including marine plants and fish, concentrations of COPC in water following a hypothetical spill should not exceed levels that could acutely or chronically impair the survival, growth or reproduction of a sensitive species. A sensitive species is defined as the 5th percentile species in a species sensitivity distribution. • For intertidal sediment community-level resources, including algae and invertebrates, exposure to COPC arising from environmental effects of spills should not exceed levels that could acutely or chronically impair community diversity, biomass or productivity. • For subtidal sediment community-level resources, including marine benthic invertebrates, concentrations of COPC in subtidal sediment following a spill example should not exceed levels that could chronically impair the survival, growth or reproduction of a sensitive species. • For mammalian, reptilian and avian receptors, exposures to COPC arising from the environmental effects of spills should not exceed levels that could acutely or chronically impair survival, growth or reproduction.

The goal is to identify potential risks to marine biota at the community or population level rather than at the individual level, with the notable exception being species that are afforded legal protection as individuals, or in respect of their residences or habitats, under federal and provincial laws (e.g., Species at Risk Act).

4-2 May 2016 Energy East Pipeline Ltd.

CA PDF Page 107 of 1038

The federal Species at Risk Act (SARA) includes a Schedule (Schedule 1) listing species in Canada that are deemed to be extirpated, endangered, threatened, or of special concern:

• Section 32 (1) states that “No person shall kill, harm, harass, capture or take an individual of a wildlife species that is listed as an extirpated species, an endangered species or a threatened species.” • Section 33 states that “No person shall damage or destroy the residence of one or more individuals of a wildlife species that is listed as an endangered species or a threatened species, or that is listed as an extirpated species if a recovery strategy has recommended the reintroduction of the species into the wild in Canada. • Section 36 applies similar protection to those provided by Sections 32 (1) and 33 to species that occur on federal lands in that province or territory, that are listed as endangered or threatened under provincial or territorial classifications. • Sections 58 and 61 extend protections to habitat that is designated as critical habitat in the recovery strategy or in an action plan for species that are listed as extirpated, endangered or threatened.

For the purposes of this ERA, any species listed as extirpated, endangered or threatened on Schedule 1 of the SARA, or listed as endangered or threatened under the New Brunswick Species at Risk Act or the Nova Scotia Endangered Species Act, will be considered to have particular status from a management perspective. For such species, harm to an individual or to its habitat will be considered to be an environmental effect at the population level. However, the listing of a species on endangered species legislation is not necessarily an indicator of the sensitivity of that species to exposure to spilled oil or hydrocarbon compounds as toxic substances. Therefore, the analysis of potential environmental effects of spilled oil on ecological receptors is not necessarily related to their status under species at risk legislation. The status of species under such legislation represents an additional factor to consider when evaluating the magnitude and significance of environmental effects.

4.2.2 Spatial Boundaries

The spatial boundaries of the SAMAM are shown in Figure 1-1 and include all of the New Brunswick and Nova Scotia coastlines of the Bay of Fundy. The SAMAM also includes Saint John Harbour to the Reversing Falls. In addition, although beyond the limits of Canadian territorial waters, parts of the United States (Maine) coastline from the Canadian border to Mount Desert Island are considered, as well as some of the waters of the Gulf of Maine.

4.2.3 Crude Oil Products

As discussed in Section 3, a variety of crude oils will be transported by the Project. These oils can be divided into three general grades, namely light, medium and heavy crude oils. Based on information from potential shippers, Energy East has identified three crude oils that are representative of these categories:

• as a light crude oil, a Bakken crude oil of Canadian origin (BAK) • as a medium crude oil, Husky Synthetic Blend (HSB), a synthetic oil of Canadian origin • as a heavy crude oil, Western Canadian Select (WCS), a diluted bitumen of Canadian origin

Energy East Pipeline Ltd. May 2016 4-3

CA PDF Page 108 of 1038

Physical and chemical properties of the representative crude oils are provided in Part A, Table 3-7 and Table 3-8 of the EHHRA.

4.2.3.1 Pseudocomponent Approach for Deterministic Oil Spill Modelling

Models used in the deterministic oil spill modelling (e.g., H3D and SPILLCALC, which are described in Part A, Appendix C) have been developed over many years to integrate relevant information in order to simulate the fate and environmental effects of oil spills in an accurate and realistic manner. However, there are limits to the complexity of processes (or chemical mixtures) that can be modelled. For the deterministic EHHRA, the “pseudocomponent” approach of Payne et al. (1984) is used, such that chemicals in the crude oils are grouped by physical-chemical properties, and the resulting component category behaves as if it were a single chemical with characteristics typical of the chemical group. These pseudocomponents are meant to be representative of analytes demonstrating similar properties. For example, toluene, ethylbenzene, and xylenes (TEX), which represent a relatively narrow range of molecular weights, vapor pressures, log KOW and solubility, have been grouped together as pseudocomponent AR2. Benzene, which is typically considered alongside TEX, was segregated into another pseudocomponent, AR1, due to its relatively higher vapor pressure. The pseudocomponents used for the oil spill modelling and toxicity assessment of BAK, HSB and WCS are presented in Table 4-1. Further details of pseudocomponent properties for each crude oil type are provided in Appendix B.

Table 4-1 Pseudocomponents Used in the Oil Spill Modelling

Pseudocomponent Canada Wide Standards Fraction1 PAH/Compound VOL Not represented by the Canada Wide Ethane (C2) Standard fractionation scheme. Propane (C3) Isobutane (iC4) n-Butane (nC4) Isopentane (iC5) n-Pentane (nC5)

Hydrogen sulphide (H2S) Carbonyl sulphide Methanethiol Ethanethiol Iso-propanethiol n-Propanethiol Dimethyl disulphide

AR1 Part of F1, C6 to C10 Aromatic Benzene

AR2 Part of F1, C6 to C10 Aromatic Toluene, ethylbenzene, xylenes

4-4 May 2016 Energy East Pipeline Ltd.

CA PDF Page 109 of 1038

Table 4-1 Pseudocomponents Used in the Oil Spill Modelling

Pseudocomponent Canada Wide Standards Fraction1 PAH/Compound

AR3 >C8 - C10 Aromatic Naphthalene 1,2,4-Trimethylbenzene 1,3,5-Trimethylbenzene

AR4 >C10 - C12 Aromatic Acenaphthene Acenaphthylene 1-Methylnaphthalene 2-Methylnaphthalene C1-naphthalene C2-naphthalene Biphenyl Dibenzothiophene

AR5 >C12 - C16 Aromatic C1-acenaphthene Acridine Anthracene Fluoranthene Fluorene C1-fluorene C2-fluorene C3-fluorene C3-naphthalene C4-naphthalene Phenanthrene Pyrene C1-biphenyl C2-biphenyl C1-dibenzothiophene C2-dibenzothiophene C3-dibenzothiophene C4-dibenzothiophene C1-phenanthrene/anthracene C2-phenanthrene/anthracene

AR6 >C16 - C21 Aromatic Benzo(a)anthracene Benzo(b&j)fluoranthene Benzo(k)fluoranthene Benzo(c)phenanthrene Benzo(a)pyrene Benzo(e)pyrene

Energy East Pipeline Ltd. May 2016 4-5

CA PDF Page 110 of 1038

Table 4-1 Pseudocomponents Used in the Oil Spill Modelling

Pseudocomponent Canada Wide Standards Fraction1 PAH/Compound

AR6 (cont’d) >C16 - C21 Aromatic (cont’d) Chrysene Perylene Retene C3-phenanthrene/anthracene C4-phenanthrene/anthracene C1-fluoranthene/pyrene C2-fluoranthene/pyrene C3-fluoranthene/pyrene C4-fluoranthene/pyrene C1-benzo(a)anthracene/chrysene C2-benzo(a)anthracene/chrysene C3-benzo(a)anthracene/chrysene C1-benzo(bjk)fluoranthene/benzo(a)pyrene

AR7 >C21 - C34 Aromatic Benzo(g,h,i)perylene Dibenzo(a,h)anthracene Indeno(1,2,3-cd)pyrene C4-benzo(a)anthracene/chrysene C2-benzo(bjk)fluoranthene/benzo(a)pyrene

AL1 C6 - C8 Aliphatic n-Heptanethiol

AL2 >C8 - C10 Aliphatic –

AL3 >C10 - C12 Aliphatic –

AL4 >C12 - C16 Aliphatic –

AL5 >C16 - C21 Aliphatic –

AL6 >C21 - C34 Aliphatic –

RES1 F4 (C34-C50 Hydrocarbons) – RES2 Not represented by the Canada Wide Polars Standard fractionation scheme. RES3 Not represented by the Canada Wide Asphaltenes Standard fractionation scheme.

NOTE: 1 From CCME (2007) assuming that the Canada Wide Standards (CWS) fractions include F1 to F4 representing aliphatic and aromatic compounds having between 6 and 50 carbon atoms.

4-6 May 2016 Energy East Pipeline Ltd.

CA PDF Page 111 of 1038

4.2.4 Oil Spill Trajectory and Fate Modelling

Details of the stochastic oil spill trajectory and fate modelling completed by Tetra Tech EBA can be found in Part A, Appendix C. For the deterministic assessment, see Appendix A of this Part B. Three hypothetical spill scenarios were selected based on consideration of the overall stochastic assessment results, the likelihood of accidental collision or grounding events, and the sensitivities of ecological receptors in the SAMAM. In each case, the credible worst case spill volume is defined as a weighted average of larger spill volumes used in the stochastic assessment. The weighting took into consideration the anticipated number of vessels belonging to each vessel class, as well as the predicted frequency of accidents leading to a loss of containment as a result of vessel collision or grounding. The accident scenarios and locations selected for deterministic assessment include the following:

• The grounding of a vessel in Saint John Harbour, resulting in the release of 10,595 m3 of WCS crude oil, at location SJ2. The cumulative frequency for drift or powered grounding accidents at this location is estimated to be 3.22E-04/year, making this the most likely single accident type and location that could lead to a credible worst case crude oil spill volume. The WCS oil and summer season were selected for this scenario because diluted bitumen is frequently identified as a particular concern due to the perceived potential for its diluent fraction to evaporate quickly, and for the weathered diluted bitumen to submerge or sink. Saint John has the largest population in the region, making this a worst- case scenario for potential human exposure to hydrocarbon vapours. In addition, the selected scenario results in spilled oil moving along the New Brunswick coastline toward Grand Manan and the Gulf of Maine, important areas for fisheries, sea birds and marine mammals. • The grounding of a vessel south of Grand Manan, resulting in the release of 12,436 m3 of BAK crude oil at location GM1 during the winter season. The predicted accident frequency at this location is 4.50E-06/year, and if combined with nearby locations GM2 and GM3 would be 5.07E-06/year. This location allows examination of the potential adverse environmental effects of a crude oil spill that affects the Grand Manan Archipelago and nearby areas of Passamaquoddy Bay, important areas for fisheries, sea birds and marine mammals. • The collision of a vessel with a loaded outbound crude oil tanker in the shipping lane south of Grand Manan, resulting in the release of 12,436 m3 of HSB crude oil, at location D during the summer season. Although the predicted accident frequency at this location (4.34E-07/year) is not as high as at some other locations near Saint John, this location allows an examination of the potential adverse environmental effects of a crude oil spill that drifts toward Digby Neck and becomes entrained into a longshore current running northeast along the Nova Scotia coastline and into the Minas Channel and Minas Basin. These are important areas for fisheries, sea birds, shorebirds and marine mammals.

Taken together, these three deterministic scenarios represent a risk-informed selection of hypothetical spill locations that considers the likelihood of the initiating accidents and malfunctions, as well as the presence of key ecological and human receptor groups associated with the Bay of Fundy and the Gulf of Maine.

Energy East Pipeline Ltd. May 2016 4-7

CA PDF Page 112 of 1038

4.2.5 Ecological Receptors

With the large number of habitats and potential wildlife species found in and around the SAMAM, it is not practical or necessary to assess each individual species. Rather, ecological receptors were selected based on the following considerations:

• receptor species should be indigenous to the area, and likely to have high exposure to COPC due to their habitat and feeding preferences and home range or residency • receptor species should be representative of various trophic levels and feeding guilds in the marine ecosystem • consideration should be given to receptor species having cultural (e.g., traditional use), economic (e.g., recreational or commercial harvest), or social (e.g., species at risk) importance • consideration should be given to acute and chronic environmental effects on receptors due to chemical and physical properties of the oils

For additional context, the highest concentrations of volatile hydrocarbons in air, and dissolved hydrocarbons in water, are typically observed within the first 24 to 48 hours following an oil spill. The length of time that oil can remain floating on the water surface, or is available to strand along shorelines is more variable, and scales according to the characteristics of the crude oil, as well as those of the receiving environment. During the acute phase of an oil spill, which may persist for a few days to 30 days or longer, there is potential for acute environmental effects due to direct contact with oil or temporary high concentrations of hydrocarbon constituents in air or water. After the acute phase has passed, concern shifts to chronic environmental effects that might arise from low-level exposure to lingering oil or PAH exposures. This chronic exposure assessment focuses on oral ingestion of hydrocarbons in food or other media (e.g., sediment) and the environmental effects that such chronic exposure may have on the health of exposed wildlife species.

With these considerations in mind, the following bird and mammal species were selected as representative ecological receptors for chronic environmental effects assessment in the SAMAM:

• spotted sandpiper • black guillemot • common eider • bald eagle • herring gull • great blue heron • double-crested cormorant • coastal-dwelling mink • harbour seal • harbour porpoise • minke whale

4-8 May 2016 Energy East Pipeline Ltd.

CA PDF Page 113 of 1038

Since the primary exposure pathway for some flora and fauna may be from direct contact with a single abiotic environmental medium (e.g., marine plants, invertebrates, or fish exposed to COPC in water or sediment), the following community-level groups were also assessed in the deterministic ERA:

• open water communities • shoreline and intertidal communities • shallow subtidal communities • deep subtidal communities

These receptors are described in detail in the following sections along with key characteristics used in the ERA model for the calculation of exposure and risk.

4.2.5.1 Birds and Mammals

SPOTTED SANDPIPER

The spotted sandpiper is one the most widely distributed birds in North America. It has generalist tendencies and wide food preferences and may occupy almost any habitat near a water source. The adult spotted sandpiper weighs approximately 40 to 50 g (Oring et al. 1997; NatureServe 2009; Government of Canada 2012b). It is not a gregarious bird, but tends to migrate singly or in small groups. This species demonstrates reversed sex roles; females are typically 20 to 25% larger than males, and males care for the young (Oring et al. 1997; NatureServe 2009). This species also demonstrates polygamous and monogamous breeding (Oring et al. 1997). The breeding range of the spotted sandpiper covers the entire continent of North America from east to west, including from the southern edge of the Arctic to the southern U.S., and altitudes ranging from sea level to 4,700 m above sea level. A successful breeder will attempt to return to breeding sites, and to sites with which it has previous experience. Hatchlings will attempt to return to hatch sites to breed. The spotted sandpiper winters in south-west British Columbia, the southern U.S., Central and South America, Bermuda, and the West Indies (Oring et al. 1997).

The spotted sandpiper will occupy a variety of habitats, ranging from beaches, to meadows and fields in agricultural areas, and forest (Oring et al. 1997; NatureServe 2009; U.S. EPA 1993). It will typically forage within 200 m of a water source, and nest within 300 m (typically within 100 m) of a water source (Oring et al. 1997). The spotted sandpiper is described as an invertivore (NatureServe 2009) or animal matter generalist (Oring et al. 1997), and will consume almost anything small enough to be eaten, including invertebrates from terrestrial, aquatic, and marine environments (Oring et al. 1997). It will occasionally eat crustaceans, leeches, molluscs, small fish, and carrion (Oring et al. 1997; U.S. EPA 1993). Some vegetation is thought to be consumed incidentally (Oring et al. 1997). The spotted sandpiper will walk or wade forward and probes, jabs, or stitches prey with its beak (Oring et al. 1997). It will often dip insects in the water before consuming them (Oring et al. 1997). The spotted sandpiper occasionally swims when it is feeding and may make shallow dives to escape predators (Oring et al. 1997; NatureServe 2009).

Energy East Pipeline Ltd. May 2016 4-9

CA PDF Page 114 of 1038

The mean home range of the spotted sandpiper during breeding is approximately 2,500 m2 (0.24 ha) (Maxson and Oring 1980 in U.S. EPA 1993). However, female territories increase in size with resource scarcity and predation. Territory sizes were reportedly on average 812.8 m2 in north-central Minnesota, but in areas with greater predation, territories varied from 12,700 to 14,500 m2, and in areas with high predation ranged from 3,000 to 20,000 m2 (Oring et al. 1997). Following Government of Canada (2012b), the average body weight of the spotted sandpiper is assumed to be 37.5 g. It is assumed to consume approximately 6.75 g of dry food per day, or 32.6 g of wet food per day. The diet is 90% composed of invertebrates obtained while foraging along shorelines and in the intertidal zone, as well as 5% small intertidal zone fish, and 5% marine plant materials. Based on its consumption of these foods, the spotted sandpiper is estimated to incidentally ingest 0.566 g of dry intertidal sediment per day.

BLACK GUILLEMOT

The black guillemot (Cepphus grylle) is a medium-sized alcid. Adult birds have black bodies with a white wing patch, a thin dark bill, and red legs and feet. They show white wing linings in flight. In winter, the upperparts are pale grey and the underparts are white. The wings remain black with the large white patch on the inner wing. The black guillemot averages 32 to 38 cm in length, with a wingspan of 49 to 58 cm, and weighs about 435 g (Winn 1950).

The black guillemot breeds along rocky shorelines, cliffs and islands of the northern Atlantic and eastern North America as far south as Maine, although they are also found in Europe and the west coast of North America. Nests may consist of shells, pebbles, seaweed, and bones, or the eggs may be laid directly on rock with no nest material at all. Nests are often located under overhangs or boulders, or in a crevice or cavity (Winn 1950).

Black guillemots are pursuit divers that propel themselves through the water using their wings and feet. Although they can theoretically dive to 130 m, most dives appear to be restricted to depths of 10 to 50 m. The species is primarily a benthic forager, since much of the prey consists of benthic fish (including sand lances, blennies, flatfish, sculpins and gadoids) and invertebrates, including crustaceans. This is modelled as including 50% small fish such as herring, and 50% krill and other crustaceans. Adults tend to consume a higher proportion of invertebrates than chicks (which appear to be fed primarily fish), and invertebrates may be of greater importance during the winter than during the summer.

Most studies of foraging behaviour indicate that black guillemots tend to remain fairly close to shore (i.e., less than 5 km and usually less than 1.5 km from shore), although they may forage at distances of up to 15 km from nesting colonies (Birdlife International 2015). Nol and Gaskin (1986) reported that black guillemots in the Quoddy region of the Bay of Fundy preferred areas with moderate current (30 to 68 cm/s) and shallow to intermediate depth (17 to 31 m). The birds appeared to avoid shallow and deep areas with fast- or slow-moving water, and preferred islands with extensive underwater ledges, presumably because these harbour sufficient prey and provide protection from fast-moving tidal waters. Other species of alcids, such as puffins, typically forage farther out to sea, and disperse farther from their

4-10 May 2016 Energy East Pipeline Ltd.

CA PDF Page 115 of 1038

nesting colonies during the winter months. Black guillemots often overwinter in their breeding areas, moving to open waters if necessary, but usually not migrating very far south (Birdlife International 2015).

Cairns (1987) reported the peak feeding rate of black guillemot chicks to correspond to a dietary intake of 780 kj/day (186 kcal/day). Assuming a caloric value of 1.25 kcal/g for the wet fish ingested, this would give an estimated feeding rate of approximately 149 g/day (as wet fish). This value is very close to the value estimated using Nagy’s (1987) equation for seabirds, as recommended in the Wildlife Exposure Factors Handbook (U.S. EPA 1993). Using Nagy’s equation, the dry food ingestion rate for a seabird weighing 435 g is estimated to be 35.65 g/day. Assuming 75% moisture content in the diet, this estimate converts to 142 g/day as wet fish. The wet-weight food ingestion rate of the black guillemot is therefore assumed to be approximately 145 g/day. Assuming also that the fish and invertebrates in the diet of the black guillemot contain approximately 1% dry sediment by weight, then the subtidal sediment ingestion rate of the black guillemot is approximately 0.341 g dry weight/day.

COMMON EIDER

The common eider is the largest duck in the northern hemisphere, weighing approximately 1,800 g, but may range in size from 850 g to 3,025 g, depending on race, age, sex, and season (Goudie et al. 2000, Canadian Wildlife Federation (CWF) and Environment Canada 2005a). It is a gregarious bird, travelling, nesting, and feeding in flocks ranging in size from tens to thousands (Goudie et al. 2000, CWF and Environment Canada 2005a). There are four races in North America, which differ in body size, breeding ranges, bill and frontal processes shape, and to a lesser degree, colouration (Goudie et al. 2000, CWF and Environment Canada 2005a).

Common eiders are dependent upon marine habitat, frequenting coastal headlands, offshore islands, skerries, and shoals within its range (Goudie et al. 2000; CWF and Environment Canada 2005a), and occasionally islands in freshwater lakes and deltas in proximity to marine waters (Goudie et al. 2000). This species nests in the vegetation and rocks of offshore islands, and frequently reuses the same nest site for several years. The common eider rarely leaves the water in the winter months (CWF and Environment Canada 2005a). Large numbers of common eiders on the Atlantic coast winter in Newfoundland and Labrador and Maine, and in the Aleutian Islands of Alaska on the Pacific Coast. However, some may travel as far south as Florida (Atlantic Coast) or Washington State (Pacific Coast). Common eiders typically migrate in spring and fall. The common eider also completes moult migrations, travelling up to several hundred kilometres north of their breeding areas. Migration routes may cover distances of up to 2,300 km (Goudie et al. 2000). However, some populations remain at their breeding grounds year-round (Goudie et al. 2000; CWF and Environment Canada 2005a). The home range of the common eider during the brood rearing period will vary with prey availability, predation, and environmental conditions such as prevailing winds. This is the lifecycle period when the foraging home range is most constrained due to flightlessness of the young. Therefore, the common eider was treated as a year-round resident of the SAMAM.

Energy East Pipeline Ltd. May 2016 4-11

CA PDF Page 116 of 1038

Adult common eiders feed on mussels, clams, scallops, sea urchins, starfish, and crabs, diving to bottom depths of up to 20 m (Goudie et al. 2000, CWF and Environment Canada 2005a). Small quantities of fish and fish eggs are also consumed; vegetation is thought to be consumed incidentally (Goudie et al. 2000). Young eat primarily aquatic invertebrates (e.g., crustaceans and small molluscs) in the first few weeks of life, and are capable of foraging by themselves (Goudie et al. 2000; CWF and Environment Canada 2005a). Females do not feed while incubating the eggs, which may last for more than 3 weeks, but frequently take drink breaks at the nearest source of fresh or salt water (Goudie et al. 2000; CWF and Environment Canada 2005a). Starting at a young age, common eiders have the ability to eliminate excess salt through supraorbital salt glands (DeVink et al. 2005). It is therefore able to meet dietary water requirements by consuming sea water (Goudie et al. 2000).

Allometric models indicate that a 1,800 g common eider will consume approximately 0.447 kg of wet weight food per day (U.S. EPA 1993). The common eider’s diet is modelled as including 95% marine invertebrates (primarily molluscs, crustaceans, and echinoderms). Minor consumption of fish and fish eggs (2.5%) and incidental consumption of aquatic vegetation (2.5%) were also accounted for (Goudie et. al. 2000). Based on its marine diet, the common eider is estimated to incidentally consume 6.17 g/day of dry intertidal zone sediment, and 1.73 g/day of dry subtidal zone sediment per day.

BALD EAGLE

The bald eagle (Haliaeetus leucocephalus) is the second largest bird of prey found in North America, and the largest in Canada (Stocek 1992). Adult birds are readily identified by their striking appearance, characterized by dark brown body plumage contrasting sharply with white head and tail plumage (Buehler 2000). The bald eagle’s range is restricted to North America, where it prefers sea coasts, lakeshores or riverine habitat that possesses suitable nesting trees in which to breed. Cape Breton and Newfoundland support most of the Atlantic breeding population (Stocek 1992), although they are also common throughout the SAMAM. In the autumn, central Canadian breeding populations migrate to the west-central and southwestern United States, returning in late winter or early spring. Coastal populations may remain in their breeding habitat year-round if their fishing areas do not freeze over (Stocek 1992).

Female bald eagles are up to 25% larger than males (Buehler 2000). The typical body mass of the bald eagle ranges from 3 to 6.3 kg (Buehler 2000), although masses of 7 kg have been recorded (Stocek 1992). Immature eagles grow rapidly owing to a voracious appetite.

Bald eagles are opportunistic feeders, taking live prey when available but preferring to scavenge carrion or to pirate freshly killed prey from other predators (Stocek 1992; U.S. EPA 1993). Their preferred food items include fish, aquatic birds and mammals. However, choice of prey is site-specific and may vary widely across their range (Buehler 2000; Government of Canada 2012b). Adult birds are more likely to hunt and kill prey, whereas immature birds are more likely to obtain food through scavenging and piracy (Stocek 1992). Coastal bald eagle populations are modelled as consuming 65% marine fish and 35% small mammals and birds (Government of Canada 2012b).

4-12 May 2016 Energy East Pipeline Ltd.

CA PDF Page 117 of 1038

For the coastal bald eagle, an average body weight of 4.7 kg is assumed (Government of Canada 2012b). U.S. EPA (1993) provides a daily wet food ingestion rate of 120 g/kg body weight. Based on a body weight of 4.7 kg this provides a total food ingestion rate of 564 g/day. Assuming 65% of this is marine fish, the daily fish ingestion rate is 367 g (wet weight). The balance of the diet is small mammals and birds, which are assumed to be of terrestrial origin (not contaminated with hydrocarbon residues). Associated with the ingestion of marine fish, there will also be ingestion of marine subtidal sediment. The marine subtidal sediment ingestion rate is estimated to be 0.888 g/day.

HERRING GULL

Herring gull populations are found both inland and near marine coasts. For this study, the herring gull is modelled as a coastal- dwelling species. The herring gull is a medium- to large-sized seabird, weighing approximately 1.1 kg (U.S. EPA 1993). It has the largest range of any North American gull (U.S. EPA 1993) and is one of the most widespread species in Canada (CWF and Environment Canada 2002). The herring gull is migratory with the exception of adult residents in the Great Lakes area, which are year-round residents (U.S. EPA 1993). Herring gulls always nest near a body of water, and may be found beside lakes, rivers, in grassy meadows, on garbage dumps, golf courses, islands, cliffs, and islands (CWF and Environment Canada 2002). In winter, herring gulls are most likely to congregate on beaches along oceans and other large bodies of water (CWF and Environment Canada 2002).

Herring gulls feed on almost anything, including fish, squid, crustaceans, molluscs, worms, insects, small mammals and birds, duck and gull eggs and chicks, amphibians, and garbage, with foraging home ranges from approximately 300 ha to 785,000 ha (U.S. EPA 1993). They will consume approximately 0.25 kg of wet weight food per day. The coastal-dwelling herring gull's diet is modelled as including 7.5% soil invertebrates, 15% terrestrial mammals, 7.5% marine invertebrates, and 70% marine fish, of which the marine invertebrates (18.75 g/day) and marine fish (175 g/day) may be contaminated by hydrocarbon residues in the unlikely event of an oil spill. Based on its consumption of these foods, the herring gull is estimated to incidentally ingest 0.343 g/day of dry marine intertidal sediment and 0.45 g/day of dry marine subtidal sediment.

GREAT BLUE HERON

Great blue heron (Ardea herodias) is a large wading bird (greater than 1 m tall), weighing approximately 2.3 kg (U.S. EPA 1993, Government of Canada 2012b). They inhabit primarily aquatic and marine areas, spending most of their time foraging for fish in shallow waters of lakes, rivers, streams, or estuarine and sheltered coastal areas. Of the herons, the great blue heron has the widest distribution in Canada, from the maritime provinces across southern Canada to the Pacific Ocean, and north along the entire length of the British Columbia. This heron breeds in all provinces except Newfoundland and

Energy East Pipeline Ltd. May 2016 4-13

CA PDF Page 118 of 1038

Labrador, wintering in Canada only on the British Columbia coast and in parts of the maritime provinces (CWF and Environment Canada 2005b).

Great blue heron feed predominantly on small fish that are half the length of its bill, or under 65 mm long. On occasion, it will also eat shellfish, insects, rodents, amphibians (mostly frogs), reptiles, and small birds (CWF and Environment Canada 2005b). Based on U.S. EPA (1993), its diet is here modelled as including intertidal zone fish (95%), and invertebrates (5%). Adults consume approximately 414 g (wet weight) food per day (Government of Canada 2012b). Based on its consumption of these foods, great blue heron is estimated to incidentally ingest approximately 1.32 g/day of dry sediment from the intertidal zone.

DOUBLE-CRESTED CORMORANT

The double-crested cormorant (Phalacrocorax auritus) can be found both in coastal and inland areas, and is widely distributed in North America, from the Aleutian Islands to Mexico, and from Florida to the maritime provinces of Canada. Double-crested cormorants are colonial water birds that seek water bodies large enough to support their diet of mainly fish. However, they may roost and form breeding colonies on smaller lagoons or ponds, and then fly a considerable distance to a suitable feeding area. In addition to fishing waters, cormorants need perching areas for the time that they spend resting each day. After fishing, cormorants retire to high, airy perches (often rocks, wires, or tops of trees) to dry off and digest their food.

The double-crested cormorant dives to find its prey. In addition to fish, it will also eat amphibians and crustaceans.

Cormorants are relatively large birds, ranging in weight from 1.2 to 2.5 kg. From an average body weight of 1,800 g, a daily food ingestion rate of approximately 97 g dry matter per day is estimated using Nagy’s (1987) equation for seabirds, as recommended in the Wildlife Exposure Factors Handbook (U.S. EPA 1993). This value is adjusted assuming that marine fish have a moisture content of 75%, to give a food ingestion rate of approximately 390 g wet weight/day. In the Bay of Fundy, the preferred prey is assumed to be small pelagic fish such as juvenile herring, and bottom-dwelling fish. The intertidal zone dry sediment ingestion rate of the double-crested cormorant is estimated from the food ingestion rate to be 0.179 g/day, whereas the subtidal dry sediment ingestion rate is estimated to be 0.773 g/day, assuming that its prey (small fish) have an average dry sediment content equivalent to 1% of their tissue dry mass.

4-14 May 2016 Energy East Pipeline Ltd.

CA PDF Page 119 of 1038

COASTAL-DWELLING AMERICAN MINK

The American mink (Neovison vison) weighs approximately 820 g (Government of Canada 2012b). It is a medium-sized member of the weasel family and is the most abundant and widely distributed carnivorous mammal in North America (U.S. EPA 1993). Mink are found throughout the continental portion of Canada (including Newfoundland), except in the most barren portions of northwestern Québec and eastern Nunavut.

Mink are active year-round and are associated with aquatic habitats such as rivers, streams, lakes, ditches, swamps, marshes, coastal shorelines and backwater areas (U.S. EPA 1993). Home ranges vary considerably but are in the range of 7.8 to 380 ha (U.S. EPA 1993). The mink feeds extensively on small mammals, fish, amphibians and crustaceans, as well as birds, reptiles and insects depending on the season (U.S. EPA 1993). Mink consume approximately 0.22 kg/day of wet weight food (U.S. EPA 1993). The diet of coastal-dwelling mink is modelled as including 25% small mammals and birds (not contaminated with hydrocarbon residues), 40% fish from the intertidal zone, and 35% marine invertebrates (e.g., crabs, molluscs and echinoderms) also from the intertidal zone.

Based on its consumption of a variety of foods, the coastal-dwelling mink is estimated to incidentally ingest approximately 1.63 g/day of dry sediment from the intertidal zone.

HARBOUR SEAL

Two subspecies of harbour seal are found in eastern Canada: Phoca vitulina mellonae, which is a freshwater seal of the Lacs des Loups Marins area of Québec’s Ungava peninsula, and Phoca vitulina concolor, which is found along the Atlantic and Arctic coasts, extending into Greenland to the north, and the United States to the south. In the SAMAM, Phoca vitulina concolor is present primarily in the outer Bay of Fundy, along the Maine coast and offshore islands such as Machias Seal Island, to Quaco Head in New Brunswick; and from Parkers Cove to Cape Sable Island, Nova Scotia (COSEWIC 2007).

These seals are typically found in nearshore waters and use both aquatic and terrestrial habitat. They haul out on rocky or sandy substrates, often on isolated rocks and islets. Because they lack elongated front claws, they cannot excavate holes in the ice and therefore rely on areas of permanent open water or spend the winter at the edge of the fast ice (COSEWIC 2007). Male adult harbour seals in eastern Canada are slightly larger than females. Males reach a length of about 150 cm and a mass of 100 kg compared with females that grow to a length of about 140 cm and have a mass of 70 kg. Males rarely exceed 100 kg and females rarely exceed 90 kg (COSEWIC 2007).

Energy East Pipeline Ltd. May 2016 4-15

CA PDF Page 120 of 1038

Harbour seals in the northwest Atlantic have a broad diet consisting of invertebrates and planktivorous and omnivorous fish, including sand lance, herring, pollock, cod, capelin and squid (COSEWIC 2007; DFO 2015a). The food ingestion rate is reported to range from 0.05 to 0.13 kg wet weight per kg body weight per day (U.S. EPA 1993). A value of 0.1 is assumed here, giving a total diet of approximately 8 kg wet weight food per day. This diet is assumed to comprise 75% fish and 25% squid or other molluscs. These foods (on a dry weight basis) are assumed to contain 1% dry sediment, giving an estimated subtidal sediment ingestion rate of approximately 55.7 g dry weight per day.

HARBOUR PORPOISE

Differences in the skull morphology of harbour porpoises from the Atlantic and Pacific oceans has led to subspecific separation. In Canada, members of the Northwest Atlantic subspecies (Phocoena phocoena phocoena) are found from the Bay of Fundy north to Cape Aston, Baffin Island, and south to North Carolina (Environment Canada 2006a). The Atlantic subspecies can be further divided into three subpopulations in Canada: in Newfoundland-Labrador; the Gulf of St. Lawrence; and the Bay of Fundy/Gulf of Maine (DFO 2008). Both the Atlantic and Pacific subspecies are designated as species of special concern by COSEWIC (P. p. vomerina is listed as a Schedule 1 species under the SARA). Incidental catch of the harbour porpoise in fishing gear (e.g., gill nets) is a major cause of mortality and the major reason for these designations.

The harbour porpoise is a small, sexually dimorphic cetacean. Although colouration may vary, the harbour porpoise typically has a dark grey dorsal side that transitions to a white ventral side at the mid- flank (Hammond and Masi 2000). A dark stripe extends from the mouth to the flippers (CMS 2003). Adult females have a mean length of approximately 1.6 m and weigh 60 kg, whereas adult males are approximately 1.45 m in length and weigh 50 kg (CMS 2003). An average body weight of 55 kg was assumed here.

The harbour porpoise is found on the continental shelves of temperate North America, generally inhabiting oceanic depths of less than 150 m (Environment Canada 2006b, IMMA 1998). Some seasonal migration is exhibited (following movements of prey) as the harbour porpoise moves north and inshore during the summer, and spends winters farther offshore in southern waters (CMS 2003). In the Bay of Fundy, animals may traverse areas up to 11,000 km2 over the course of 1 month. However, much of their foraging effort is usually focused on an area less than 300 km2.

The diet of the harbour porpoise consists mainly of small schooling fish (e.g., herring and sand lance) and squid (Santos and Pierce 2003, Environment Canada 2006b). Harbour porpoises consume these food items, assumed to comprise 95% fish and squid and 5% benthic invertebrates such as crab or lobster, at a rate of approximately 4.32 kg wet weight per day (estimated using the allometric equation provided by Innes et al. 1987). Based on its diet of pelagic fish and squid, the rate of dry marine subtidal sediment ingestion for the harbour porpoise is estimated to be 14.4 g/day.

4-16 May 2016 Energy East Pipeline Ltd.

CA PDF Page 121 of 1038

MINKE WHALE

The minke whale (Balaenoptera acutorostrata) is the smallest member of the rorqual family of whales (Balaenopteridae – whales with baleen, a dorsal fin, and throat pleats). The baleen of the minke comprises 280 to 300 yellowish-white plates, usually no more than 11 inches in length, on each side of its upper jaw. Its body is slender and streamlined. Like all rorquals, the minke has a series of ventral grooves, or pleats, that expand during feeding (American Cetacean Society 2015). Of the whales regularly seen in the Bay of Fundy and Gulf of Maine, most (e.g., the fin whale, humpback whale, sei whale and blue whale) are rorquals. The North Atlantic right whale (Eubalaena glacialis), also found in the Bay of Fundy, is a baleen whale belonging to the family Balaenidae, which lack the throat pleats of the rorquals.

Adult male minke whales average about 8 m with a maximum length of 9.4 m, whereas adult females average 8.2 m with a maximum length of 10.2 m. Both males and females typically weigh between 7 and 10 tonnes. Sexual maturity is reached at 7 or 8 years, and breeding occurs during the summer months. The gestation period is 10 to 11 months, and calving is thought to occur once every 2 years on average. Calves are approximately 3 m long at birth and weigh about 450 kg. Minke calves nurse for approximately 6 months. The lifespan of the minke whale is believed to be about 50 years (American Cetacean Society 2015).

Minke whales are widespread, ranging from subtropical to polar waters. They tend to be solitary animals, though sometimes they are seen traveling in pairs or in small groups. Where food is concentrated, it is common to find larger aggregations of feeding animals. They appear to segregate by age and sex more than do the other baleen whales. Females remain close to shore, while males are seen farther out to sea. Some minke whales migrate long distances, but others may move only within a restricted area. In some regions, minkes may be found year-round. In the Bay of Fundy, Minke have been observed in all seasons, however, their abundance appears to be lower during the winter than in other seasons. Therefore, it is assumed that many of the minke found in the Bay of Fundy during spring and summer may migrate to other areas during fall and winter.

Minke whales feed on a wide variety of fish and invertebrates. In the North Atlantic, they consume mainly krill, herring, capelin, sand lance, cod, haddock, and other species of fish and invertebrates (NAMMC 2015). The diet varies both by location and over time. In the Northeast Atlantic, krill dominate the diet in far northern areas, whereas capelin, herring and haddock become more important further south in the Norwegian Sea and along coastal Norway. Sand lance and mackerel become more common in the diet in southern areas such as the North Sea. In the Central Atlantic, capelin appears to make up a larger part of the diet, but herring, sand lance and cod are also important. Interannual variations in diet composition, probably reflecting prey availability, have been noted in the Northeast Atlantic and around Iceland (NAMMC 2015). For minke whale in the Bay of Fundy, where both crustaceans (krill and copepods) and fish (herring, and other species) are abundant, it is assumed that the diet is divided more or less equally between these two main components.

Energy East Pipeline Ltd. May 2016 4-17

CA PDF Page 122 of 1038

Hunter et al. (2000) provide comparative estimates of basal metabolic rate and feeding rate for a variety of whales and seals. Assuming a maximum length for the minke whale of about 10 m, the energy requirement is estimated to be approximately 130,000 kcal/day. Assuming also that the caloric content of both krill and herring is about 1.25 kcal/g wet weight (Färber-Lorder et al. 2009; Paul and Paul 1998), the food ingestion rate of the minke whale is estimated to be about 104 kg/day. Pinkerton et al. (2010) provide a higher estimate of food ingestion for minke whale near the Ross ice shelf (approximately 24.5 tonnes over a 3 to 4 month period, or roughly 245 kg fresh weight per day). Since the energetic requirements of whales near the Antarctic ice shelf are likely to be greater than for whales in the Bay of Fundy, it is assumed that the food ingestion rate for adult male or female minke whales is both greater than the basal metabolic rate, and lower than the rate estimated for Antarctic animals. A value of approximately 200 kg fresh weight per day was chosen, comprising approximately 100 kg/day as small fish, and 100 kg/day as crustaceans. Fish are assumed to have a dry weight equivalent to 25% of their wet weight, whereas krill are assumed to have a dry weight equivalent to 15% of their wet weight. On this basis, the average dietary intake is represented by 80 kg dry food per day. Assuming further that both fish and krill contain 1% dry sediment on a dry weight basis, a dry subtidal sediment ingestion rate of 470 g/day is indicated for the minke whale.

4.2.5.2 Community-Level Receptor Groups

Key community level receptors included in the deterministic assessment for the ERA include:

• marine species (including algae, invertebrates and fish) exposed to dissolved hydrocarbons in the water column, with one or more of non-polar narcosis, blue sac disease, and phototoxicity as modes of toxic action • birds, mammals or marine reptiles exposed to crude oil on the surface of the water, leading to harmful effects on these species as a result of either hypothermia (caused by loss of insulative characteristics of fur or feathers) or ingestion of crude oil as a result of grooming or other behaviours following such exposure • shoreline and intertidal communities (including eelgrass, rockweed and kelp beds) exposed to crude oil as a result of the stranding of such oil along the shoreline following an accident or spill, and subsequent oil spill response or cleanup activities, with resulting harm to algal, intertidal invertebrate, or intertidal fish species • benthic invertebrate and demersal fish species exposed to crude oil as a result of processes that indirectly or directly lead to the sinking of oil and its incorporation into subtidal or deepwater sediments

4-18 May 2016 Energy East Pipeline Ltd.

CA PDF Page 123 of 1038

4.2.5.3 Species at Risk

For the purposes of this ERA, species that are listed as extirpated, endangered or threatened on Schedule 1 of the SARA, or that are listed as endangered or threatened under the New Brunswick Species at Risk Act, or the Nova Scotia Endangered Species Act will be considered species at risk. For these species, predicted harm to an individual or predicted damage to their habitats will be considered equivalent to a prediction of an adverse environmental effect on the species at a population level.

Receptor organisms identified in the ERA are intended to be representative of multiple species belonging to similar ecological guilds or occupying similar habitats and ecological niches. Thus, the minke whale is considered representative of all baleen whales, including the endangered North Atlantic right whale; and the bald eagle, which is listed as endangered under the New Brunswick Species at Risk Act, is also representative of other common and widely distributed fish-eating raptors such as the osprey, which do not have this designation. Conclusions about potential environmental effects of crude oil spills, and the potential significance of such environmental effects, will be expressed in terms of species that are common and widely distributed, and where appropriate, will also take into consideration the status of species at risk.

A review of Schedule 1 of the SARA, as well as the Regulations enacted in support of the New Brunswick Species at Risk Act and the Nova Scotia Endangered Species Act, identified 27 species that historically had or presently have ranges that include the Bay of Fundy (Table 4-2). About half (14) of the species were identified through the SARA, with the remainder being introduced as a result of their status under New Brunswick and Nova Scotia legislation. These species should be considered as having legal protection as extirpated, endangered or threatened species. However, not all of the listed species can reasonably be considered likely to interact with crude oil in the unlikely event of an oil spill. For example, the Atlantic walrus, Northwest Atlantic population (listed as extirpated under SARA), historically had a range that included parts of the Bay of Fundy. However, this species has been extirpated from the Northwest Atlantic, the Mackenzie delta and the St. Lawrence River and is now limited to Arctic waters from the polar ice-sheet in the Arctic Ocean to the Bering Sea, James Bay and the Labrador coast. The likelihood of encountering an Atlantic walrus in the Bay of Fundy is so low that this species can be removed from further consideration. Similarly, the Eskimo curlew (listed as endangered on Schedule 1 of the SARA and under the New Brunswick Species at Risk Act) is a bird that once migrated along the eastern seaboard, between breeding grounds in the Northwest Territories, and wintering grounds in South America. Eskimo curlew have not been positively identified anywhere since 1866, and again the likelihood of this bird interacting with spilled oil in the Bay of Fundy is so low that this species can be removed from further consideration. Table 4-2 includes a review of the status of each species to determine which should reasonably be carried forward for additional consideration as species at risk within the ERA. As a result of that review process, the following species are identified:

• North Atlantic right whale

• red knot

• piping plover

• roseate tern

Energy East Pipeline Ltd. May 2016 4-19

CA PDF Page 124 of 1038

• harlequin duck

• bald eagle

• leatherback sea turtle

• Atlantic salmon (inner and outer Bay of Fundy populations)

• white shark (Atlantic population)

• striped bass (Bay of Fundy population)

• Atlantic cod (Laurentian South and Southern populations)

• cusk

• bluefin tuna (Atlantic population)

• American eel

• American plaice (Maritime population)

• Acadian redfish

• Atlantic sturgeon (Maritime population)

4.2.6 Exposure Pathways and Conceptual Site Model

The CSM developed for this assessment, presented schematically in Figure 4-1, represents the potential interactions between receptors and COPC via the identified exposure pathways. In the CSM, relevant exposure pathways are designated by arrows leading from the contaminant source media to each receptor. The pathway is considered to be complete (i.e., functioning) for a receptor when the exposure pathway box is marked with an “X”.

4-20 May 2016 Energy East Pipeline Ltd.

CA PDF Page 125 of 1038

Energy East Project Volume 24: Ecological and Human Health Risk Assessment for Oil Spills in the Marine Environment Part B: Deterministic Modelling of the Ecological and Human Health Consequences of Marine Oil Spills Section 4: Ecological Environmental Effects Assessment Methods for Deterministic Oil Spill Modelling

Table 4-2 Species Deemed to be “At Risk” within the SAMAM for the Purposes of the Ecological Risk Assessment

Identifying Relevance to Ecological Risk Assessment of Crude Oil Spills Species Legislation Characteristics Modifying Interactions With Spilled Oil in the Bay of Fundy

Extirpated Under Federal Legislation

Atlantic walrus 1 Based on its present status, interactions between this extirpated population and a release of crude oil in the Bay of Fundy are extremely unlikely. Low. This species is extirpated from the Bay of Fundy. (northwest Atlantic population)

Grey whale 1 Based on its present status, interactions between this extirpated population and a release of crude oil in the Bay of Fundy are extremely unlikely. Low. This species is extirpated from the northwest Atlantic. (Atlantic population)

Endangered Under Federal, New Brunswick or Nova Scotia Legislation

Blue whale 1, 2 The Atlantic population of blue whales lives in the waters off the east coast of Canada. During the spring, summer and fall, these whales can be Low. This species is a highly unlikely visitor to the Bay of Fundy. (Atlantic population) found along the north coast of the Gulf of St. Lawrence and the east coast of Nova Scotia. Presence in the Bay of Fundy is highly unlikely.

North Atlantic right whale 1, 2 The North Atlantic right whale travels along the east coast of North America primarily from eastern Florida to the Gulf of St. Lawrence and High. Adults and young feed in deep coastal areas such as the Bay Newfoundland. In particular North Atlantic right whales are observed feeding and socializing in the lower Bay of Fundy and in Roseway Basin on of Fundy and Scotian Shelf from summer to late fall. the western Scotian Shelf.

Eskimo curlew 1, 2 The known breeding range was located in the Northwest Territories. The species wintered in South America. During migration it was found in all Low. This species is a highly unlikely visitor to the Bay of Fundy. provinces except British Columbia. No positively identified eskimo curlew nests or birds behaving as if they had nests or young have been found since 1866.

Ivory gull 1 In Canada, the species breeds exclusively in Nunavut, where there is permanent drift ice and open water. Colonies are concentrated round Jones Low. This species is a highly unlikely visitor to the Bay of Fundy. and Lancaster sounds on southeastern Ellesmere Island, eastern Devon Island, and the Brodeur Peninsula of northern Baffin Island. One outlying colony exists farther west, on Seymour Island, off the northern coast of Bathurst Island. The wintering grounds of the Ivory Gull are poorly known, but they are thought to be along the southern edge of the pack ice in the North Atlantic and North Pacific oceans. The Ivory Gull spends the winter on the pack ice of Davis Strait, the Labrador Sea, the Strait of Belle Isle and the northern Gulf of St. Lawrence. It is occasionally seen along the eastern coasts of Newfoundland and Labrador, particularly the Great Northern Peninsula of Newfoundland, and on the Lower North Shore of Québec.

Red knot 1, 2, 3 Red knot (rufa subspecies) is a type of sandpiper that breeds in the central Canadian Arctic and winters in Tierra del Fuego at the southern tip of High. While migrating in spring and fall, these birds use coastal (rufa subspecies) South America. Currently, the most important areas for migrating red knot (rufa subspecies) in eastern Canada are along the north shore of the St. areas with extensive sand or mudflats to forage for crustaceans Lawrence River in Québec. Potentially some red knot could migrate through the Bay of Fundy. and bivalves.

Piping plover 1, 2, 3 The piping plover (melodus subspecies) breeds along the Atlantic coast from Newfoundland to South Carolina, wintering along the Atlantic coast, High. These shorebirds nest on wide sandy beaches with little (melodus subspecies) from South Carolina to Florida, and in the Caribbean (Cuba, Bahamas). In Canada, the melodus subspecies breeds on beaches of the Magdalen vegetation and a mix of substrates such as pebbles, gravel, shells Islands of Québec, New Brunswick, Nova Scotia, Prince Edward Island and Newfoundland. and sticks. While this habitat type is not abundant along the Bay of Fundy coastline, the potential exists for these birds to nest or occupy habitat within the SAMAM during the spring and summer months.

Roseate tern 1, 2, 3 The Canadian population of roseate tern breeds almost exclusively on a few islands off the Atlantic coast of Nova Scotia, although small numbers High. Roseate tern breed in the spring and summer on islands of birds also breed on islands in Québec and New Brunswick. The location of small colonies changes unpredictably from year to year and only two within the SAMAM, particularly around the southern tip of Nova such colonies, both in Nova Scotia, have maintained relatively large numbers of Roseate Terns since the 1980s. The birds are migratory and Scotia, but also historically at Brier Island and recently at Machias winter in South America. Seal Island.

Harlequin duck 2,3 Harlequin ducks of the eastern population mostly breed throughout much of Labrador, along eastern Hudson Bay, and the Great Northern High. The eastern wintering population of harlequin duck uses Peninsula of the island of Newfoundland. There are also known breeding populations along the north shore of the Gulf of St. Lawrence, the Gaspé established sites along the coast of eastern North America from Peninsula, northern New Brunswick, and southeastern Baffin Island in Nunavut. Many winter on the east and south coasts of Newfoundland, in Newfoundland to Virginia. These include Jericho Bay, Penobscot southeastern Nova Scotia, in southern New Brunswick, in Maine, and at a few locations south of Cape Cod. Bay, and the Isle au Haut regions of Maine. The Atlantic and Bay of Fundy coasts of Nova Scotia and New Brunswick also regularly have harlequin duck in winter.

Bald eagle 2 Ubiquitous year-round residents in New Brunswick and Nova Scotia. High. The bald eagle is year-round resident of coastal areas of the SAMAM.

Energy East Pipeline Ltd. May 2016 4-21

CA PDF Page 126 of 1038

Energy East Project Volume 24: Ecological and Human Health Risk Assessment for Oil Spills in the Marine Environment Section 4: Ecological Environmental Effects Assessment Methods for Deterministic Oil Spill Modelling Part B: Deterministic Modelling of the Ecological and Human Health Consequences of Marine Oil Spills

Table 4-2 Species Deemed to be “At Risk” within the SAMAM for the Purposes of the Ecological Risk Assessment

Identifying Relevance to Ecological Risk Assessment of Crude Oil Spills Species Legislation Characteristics Modifying Interactions With Spilled Oil in the Bay of Fundy

Leatherback sea turtle 1, 2 Leatherbacks are migratory sea turtles that breed in tropical or subtropical waters and move to temperate waters in search of food (chiefly High. Leatherbacks are occasionally sighted in the Bay of Fundy, jellyfish). Leatherbacks are often sighted on the east coast of Canada between June and October. The cold waters of the Bay of Fundy are at although principally in areas around the southern tip of Nova Scotia about the thermal limit for this species. (south and east of Brier Island)

Loggerhead sea turtle 2 Loggerheads found in eastern Canadian waters likely originate from the same nesting populations (i.e., peninsular Florida) as turtles found in Low. Most observations of loggerhead sea turtles in Canadian northeastern U.S. waters. At sea, Loggerhead Sea Turtles prefer water temperatures of 18°C and warmer. Smaller individuals find shelter from waters come from offshore waters. predators, food, and warmth in floating mats of seaweed in the open ocean beyond the continental shelf. Larger juvenile loggerheads, which are less vulnerable to predation, occupy shelf waters along the southeastern United States through to New England, and offshore waters of the North Atlantic. Mature Loggerheads mainly inhabit relatively shallow continental shelf waters from New York south through the Gulf of Mexico. The cold waters of the Bay of Fundy are at about the thermal limit for this species.

Atlantic salmon 1, 2 This population spawns in rivers of Nova Scotia and New Brunswick that drain into the Minas Basin and Chignecto Bay, as far south as the Black High. Inner Bay of Fundy Atlantic salmon must migrate through the (inner Bay of Fundy population) River in New Brunswick. After these salmon go to sea, they remain in the Bay of Fundy, at least until late autumn, but it is not known where they Bay of Fundy as post-smolts, and as returning adults, and they spend the winter. may remain feeding in the Gulf of Maine.

Atlantic salmon 2 This population breeds in rivers tributary to the New Brunswick side of the Bay of Fundy, from the U.S. border to the Saint John River. After High. Outer Bay of Fundy Atlantic salmon must migrate through the (outer Bay of Fundy population) entering the Bay of Fundy, members of this population undertake a lengthy feeding migration toward Greenland in the North Atlantic Ocean. Bay of Fundy as post-smolts, and as returning adults.

White shark 1, 2 The white shark is widely distributed in subpolar to tropical seas of both hemispheres, but it is most frequently observed and captured in inshore High. White shark records from Atlantic Canada consist largely of (Atlantic population) waters over the continental shelves of the western North Atlantic, Mediterranean Sea, southern Africa, southern Australia, New Zealand, and the incidental captures. Clustering of white shark records in Atlantic eastern North Pacific. Off Atlantic Canada, the white shark has been recorded from the Northeast Newfoundland Shelf, the Strait of Belle Isle, the Canada during late summer months suggests they may be St. Pierre Bank, Sable Island Bank, the Forchu Misaine Bank, in St. Margaret’s Bay, off Cape La Have, in Passamaquoddy Bay, in the Bay of correlated with a seasonal shift of the Gulf Stream. Fundy, in the Northumberland Strait, and in the Laurentian Channel as far inland as the Portneuf River Estuary. The white shark occurs in both inshore and offshore waters. Individuals in Atlantic Canada are likely seasonal migrants belonging to a widespread northwest Atlantic population.

Atlantic whitefish 1 The Atlantic whitefish is found only in the Tusket (within the SAMAM) and Petite Rivière (outside the SAMAM) watersheds in southern Nova Low. This species appears to be extirpated from the SAMAM. (Nova Scotia) Scotia. The last confirmed report of an Atlantic Whitefish in the Tusket-Annis system was in 1982. Historically, the species appears to have had a sea-run life history trait in both the Tusket and Petite Rivière systems, although some lake populations were entirely landlocked. Sea-run fish would be found in estuarine and coastal marine waters in the summer, with fish likely migrating into freshwater for the winter months to spawn, returning to the sea in the spring.

Striped bass 2 The Bay of Fundy population of striped bass occurs in the Bay of Fundy and Atlantic Ocean. There is one confirmed spawning population in the High. The Bay of Fundy population includes the spawning (Bay of Fundy population) Shubenacadie River, NS, and possibly another in the Saint John River, NB. A spawning population has been extirpated from the Annapolis River, population of the Shubenacadie River, NS, and a possible NS. These fish are known to move throughout the Bay of Fundy, in addition to occupying freshwater habitat. Striped bass spawn in fresh water. At spawning population in the Saint John River, NB. Striped bass also the juvenile and adult stages, striped bass use coastal, estuarine and saltwater environments. In the fall they enter estuarine and freshwater use the Annapolis River to feed. They move around and feed in habitats where they spend the winter. coastal waters of the Bay of Fundy.

Atlantic cod 2 Atlantic Cod inhabit all waters overlying the continental shelves of the Northwest and the Northeast Atlantic Ocean. In the Bay of Fundy region High. Cod spawning takes place over a protracted period, but for (Laurentian South and Southern there are two designated populations (Laurentian South and Southern), both of which are designated as “endangered” under New Brunswick southern populations likely peaks in May. Spawning occurs at populations) species at risk regulations. depths of tens to hundreds of metres. Very large numbers of eggs are released. The eggs are buoyant and rise to the surface water layers during incubation (around 40 days). Larval cod remain pelagic in the upper 10 to 50 metres of water for a time, then descend to the bottom.

Cusk 2 Cusk is a cod-like fish found in deep waters around the Canada-United States boundary in the Gulf of Maine and Atlantic Ocean. Its centre of High. On the Scotian Shelf, cusk spawn in the spring and summer. abundance in the western Atlantic is in the Gulf of Maine and the southern Scotian Shelf off southwest Nova Scotia, extending from the Fundian The eggs are buoyant, and undergo a developmental process Channel and Browns Bank to Emerald, Western and Sable island banks. Cusk spawn from April to July, with peak spawning in late June on the similar to that of cod. Scotian Shelf. Adults begin to reach maturity at lengths of about 50 cm, or 5 to 6 years of age. Cusk produce large numbers of eggs (up to one million or more per female). Eggs are buoyant, and hatch larvae of 4 mm, which develop in the upper water column, settling to the bottom when they reach lengths of 50 to 60 mm.

4-22 May 2016 Energy East Pipeline Ltd.

CA PDF Page 127 of 1038

Table 4-2 Species Deemed to be “At Risk” within the SAMAM for the Purposes of the Ecological Risk Assessment

Identifying Relevance to Ecological Risk Assessment of Crude Oil Spills Species Legislation Characteristics Modifying Interactions With Spilled Oil in the Bay of Fundy

Porbeagle 2 The porbeagle is a large coastal and oceanic shark found in the northwest Atlantic Ocean and globally. In the northwest Atlantic Ocean, porbeagle Low. Porbeagle occurs in pelagic, epipelagic or littoral habitats, but (Atlantic ocean) populations are found in waters off Greenland, Canada, the United States and Bermuda. In Canada, this species occurs in an area extending from is most commonly observed on continental shelves. In Canadian northern Newfoundland to the Gulf of St. Lawrence and around Newfoundland to the Scotian Shelf and the Bay of Fundy. Immature porbeagle waters, Porbeagle are most commonly observed in deep basins appear to mainly inhabit the Scotian Shelf, while mature individuals migrate annually along the Scotian Shelf toward the Newfoundland mating and on the edge of the continental shelf. grounds. Males seem to migrate in the spring and females arrive a little later. It is believed that mating in the northwest Atlantic occurs in the Grand Banks, south of Newfoundland and at the mouth of the Gulf of St. Lawrence. These sharks bear live young. Gestating females can be found from late September to December on the Scotian Shelf and in the Grand Banks area. However, little is known about the overwintering and pupping grounds of the porbeagle.

Bluefin tuna 2 Bluefin tuna occur on both sides of the Atlantic ocean and are seasonal migrants through Canadian waters. They are found in the summer and fall High. Bluefin tuna are summer visitors to the SAMAM, feeding (Atlantic ocean) on the Scotian Shelf, in the Bay of Fundy, Gulf of St. Lawrence, and near Newfoundland. As seasonal migrants, their occurrence and abundance primarily in the area of high productivity between Grand Manan and varies from year to year. Bluefin tuna belonging to the western Atlantic population spawn in the Gulf of Mexico. Brier Island.

Threatened Under Federal, New Brunswick or Nova Scotia Legislation

Least bittern 1 Least bittern breed from southern Canada to South America, mainly in the eastern United States. They winter along the Atlantic coast, primarily in Low. This bird prefers freshwater marsh habitats and would be Florida and along the Gulf of Mexico, and in the region extending from California, Texas and Florida to Panama. In Canada, least bittern have highly unlikely to be affected by an oil spill in the Bay of Fundy. been observed in every province, but most individuals occur in Ontario. The species breeds primarily in southern Ontario, and in southern Manitoba, Québec, New Brunswick and probably Nova Scotia. These birds prefer freshwater marsh habitat (particularly cattails) for breeding.

American eel 2 American eels spawn in the Sargasso Sea and eggs hatch within about 1 week. The larvae drift passively on surface currents of the Gulf Stream High. The glass eel and elver life history stages of the American eel (Atlantic ocean) reaching destinations on the Atlantic coast of North America. When larvae reach 55 to 65 mm length, they metamorphose into ‘glass eels’, a post- would potentially be exposed and sensitive to hydrocarbons in the larval stage characterized by a lack of pigment. As they approach coastal estuaries they become pigmented and are known as ‘elvers’. This stage event of an oil spill in the Bay of Fundy. The outward migrating lasts 3 to 12 months, during which they may migrate up rivers or remain in brackish or salt waters, eventually becoming ‘yellow eels’. The yellow silver eel life history stage would have lower exposure and stage marks the growth phase where the skin thickens and sexual differentiation occurs. Between 8 and 23 years of growth are required for susceptibility to spilled crude oil. ‘yellow eels’ to become ‘silver eels’, at which time they are physically and physiologically adapted to migrate back to their spawning grounds.

American plaice 2 In Canada, American Plaice are distributed contiguously from Georges Bank and the Bay of Fundy in the south, over the Scotian Shelf, into the High. Once fertilized the eggs become buoyant and rise up in the (Maritime population) Gulf of St. Lawrence, surrounding Newfoundland and Labrador and along the eastern coast of Baffin Island, Nunavut. The Maritime population of water column to float near the surface. Time to hatching is 11-14 American plaice is found in the Gulf of St. Lawrence and Scotian Shelf, extending into the Bay of Fundy. Adjoining the Canadian populations are days at around 4°C. The larval fish are also present near the populations along the west coast of Greenland, on and around the Flemish Cap, in the Gulf of Maine and on Georges Bank. Spawning surface initially. commences earliest (February) for fish in the Gulf of Maine and Georges Bank, but likely somewhat later in the colder waters of the Bay of Fundy. American Plaice exist as pelagic eggs and larvae for the first few weeks of life. Settled juveniles prefer depths of 100-200 m and fine sediments that they can use to partially or fully bury themselves.

Acadian redfish 2 Acadian Redfish are almost exclusively found within Canadian Atlantic waters. They live primarily along continental slopes and in deep channels, High. Redfishes are born as larvae, and the newly released larvae (Atlantic ocean) from 150 to 300 metres. Larvae prefer surface waters, where they feed on copepods and fish eggs, while adults live in cold, deep waters where are found primarily within the surface 10 m of water. Juveniles they prey upon other fish. Acadian Redfish are ovoviviparous, meaning that females keep their fertilized eggs inside their bodies until the larvae move deeper after reaching a length of about 25 mm. have hatched.

Shortfin mako 2 Shortfin mako are found around the world from temperate to tropical waters. In the northwest Atlantic they have been found both inshore and Low. In Canadian waters the shortfin mako is typically associated (Atlantic population) offshore, from Bermuda to the waters east of Newfoundland. In Canadian waters where they are considered at the edge of their range, they have with warm waters in and around the Gulf Stream. It has been been recorded from the Grand Banks off Newfoundland, along the Scotian Shelf and down to Georges Bank. Shortfin mako are highly migratory recorded from Georges and Browns Bank, along the continental with distribution apparently dependent on water temperatures which they prefer between 17 and 22°C. They migrate to the Atlantic coast of shelf of Nova Scotia, the Grand Banks and into the Gulf of St. Canada generally in the late summer and fall where they are usually associated with the warm waters of the Gulf Stream. While these fish might Lawrence. It is an unlikely visitor to the Bay of Fundy, due to cold be found in the Gulf of Maine, the waters of the Bay of Fundy would generally be too cool to attract this species. water temperatures.

Atlantic sturgeon 2 Atlantic sturgeon are large, slow-growing and late-maturing fish that occur in rivers, estuaries, nearshore marine environments and shelf regions to High. Atlantic sturgeon can be expected to be present in coastal (Maritimes population) at least 50 m depth along the Atlantic coast of North America. They range as far north as Ungava Bay, into the Gulf of St. Lawrence and Bay of waters of the Bay of Fundy, after they become large enough to Fundy, south along the Atlantic coast to Florida, and into the Gulf of Mexico. The Maritimes population is known to spawn only within the lower leave fresh water. They are generally a bottom-dwelling and Saint John River in New Brunswick (although potentially also in other rivers tributary to the Bay of Fundy). Spawning occurs in freshwater, and bottom-feeding species. juveniles likely remain in freshwater until they are 80 to 100 cm long.

Energy East Pipeline Ltd. May 2016 4-23

CA PDF Page 128 of 1038 CA PDF Page 129 of 1038

Potential Source of COPC Exposure Media Exposure Pathways Bald Eagle Herring Gull Minke whale Harbour seal Common Eider Black Guillemot Great Blue Heron Harbour porpoise Spotted Sandpiper Aquatic Community Sub-tidal Community Coastal-dwelling mink (Invertebrates and Fish) Double-crested cormorant Intertidal Shoreline Community (Algae, Invertebrates, and Fish) (Algae, Invertebrates, and Fish)

evaporation Air Inhalation XXXX

Surface water Direct Exposure X

Accidental Hydrocarbon Spills to Marine Plants Ingestion X X the Marine uptake Environment Prey Ingestion XXXX XXXXX (e.g., fish, invertebrates) uptake

Intertidal Shoreline Direct Exposure X stranding Sediment

uptake Ingestion X X XXXX

Prey Ingestion X X XXXX (e.g., fish, invertebrates)

Sub-tidal Sediment Direct Exposure X sedimentation

uptake Ingestion XXXX X XXX

Prey Ingestion XXX X XXX (e.g., fish, invertebrates)

Figure 4-1 Ecological Risk Assessment Conceptual Site Model

Energy East Pipeline Ltd. May 2016 4-25

CA PDF Page 130 of 1038

4.3 Ecological Receptor Exposure and Environmental Effects Assessment

The purpose of the exposure and environmental effects assessment is to evaluate data related to the crude oil products, ecological receptors and exposure pathways identified in the problem formulation, and to determine whether such exposures (usually expressed in terms of the daily ingested dose from all relevant media, or an ambient exposure concentration) may be in a range that can credibly be linked to adverse health effects. The exposure assessment culminates in predictions about the behaviour and distribution of COPC in the environment, and the extent to which ecological receptors would be exposed via exposure scenarios and pathways. The magnitude of exposure depends on the interactions of a number of parameters, including:

• concentrations of COPC in various environmental media following a hypothetical spill • physical-chemical characteristics of the COPC, which affect their environmental fate and transport and determine such factors as efficiency of absorption into the body and rate of metabolic breakdown or excretion • influence of site-specific environmental characteristics (e.g., shoreline and sediment type, bathymetry, ocean currents, and weather patterns on the COPC behaviour within environmental media) • physiological and behavioural characteristics of the receptors

Exposure assessments specific to each hypothetical spill scenario (i.e., a spill caused by grounding in Saint John Harbour; a spill caused by collision in the designated shipping land in the Bay of Fundy; and a spill caused by grounding south of Grand Manan) are provided in Sections 6, 7 and 8. The toxicity assessment, however, is common to all scenarios, and is described in the following subsections of this section.

4.3.1 Hydrocarbon Environmental Effects in the Water Column

Three assessments were carried out for hydrocarbon toxicity in the water column, as follows:

• a non-polar narcosis toxicity model is used to evaluate the potential for mortality of marine biota (including algae, invertebrates and fish) exposed to dissolved hydrocarbons in the water column • the potential induction of deformities or mortality in developing fish eggs and embryos (the Blue Sac Disease endpoint) is evaluated using a model based on the potential for exposure to dissolved PAH compounds in the water column • the potential for phototoxicity in fish and other aquatic life due to bioaccumulation of dissolved PAHs from the water column, in combination with exposure to light (particularly ultraviolet light in the UVA range) is evaluated using a model that considers exposure to PAHs at multiple depths in the water column, in conjunction with the UVA light intensity likely to be encountered at each depth

The technical basis for evaluating these endpoints is briefly described below, with additional detailed information to be found in Appendix C.

4-26 May 2016 Energy East Pipeline Ltd.

CA PDF Page 131 of 1038

Conceptually, exposure for fish, fish eggs, and benthic invertebrates is considered to be primarily to hydrocarbons present in dissolved form. Hydrocarbons are hydrophobic and partition strongly between water and other available non-polar media, including organic matter suspended or present in bed sediment, and living organisms. Uptake of hydrocarbons from water by living organisms is considered to be regulated primarily by equilibrium exchange processes between water and lipids, and to take place across permeable or vascular surfaces, such as gills or egg membranes. Dissolved hydrocarbon concentrations in water are generally not persistent following crude oil spills due to a variety of factors, including low solubility of higher molecular weight hydrocarbons, rapid volatilization and microbial degradation of those hydrocarbon types that tend to be relatively more water soluble and dispersion of dissolved hydrocarbons into the very large and often turbulent ocean water mass. As a result, exposures to toxicologically relevant concentrations of dissolved hydrocarbons tend to be brief, and to occur in the surface water layers in close proximity to freshly spilled surface oil (McIntosh et al. 2010).

Once inside an organism, hydrocarbons become part of the generalized lipid pool, and may or may not be metabolized. Because some invertebrates (molluscs) lack enzyme systems capable of rapidly metabolizing PAHs, it is assumed that hydrocarbons can be accumulated and retained by benthic invertebrates. In contrast, vertebrate species are capable of metabolizing and excreting most hydrocarbon compounds, and bioaccumulation is less pronounced and of shorter duration.

The toxicity of hydrocarbons to aquatic receptors considers three primary mechanisms of toxicity, including:

• screening for acute toxicity benchmarks for narcosis based on the Target Lipid Model (TLM) of Di Toro et al. (2000). This assessment is conservatively based on the maximum 96-hour average exposure to dissolved petroleum hydrocarbon compounds at any depth in the water column. • screening for potential blue sac disease endpoints. This assessment is based on the maximum 24-hour average exposure to total polycyclic aromatic hydrocarbons (TPAH) at any depth in the water column. It is conservatively assumed that potential for the earliest and least severe symptoms of BSD in embryos of sensitive fish species may occur at an exposure threshold TPAH concentration of 1 µg/L, and potential mortality of developing eggs and embryos may occur at an exposure threshold TPAH concentration of 10 µg/L. • screening for potential phototoxicity with the probability of toxicity calculated on the basis of the maximum dissolved PAH concentration (as anthracene equivalents) at depth intervals of 0 to 1 m, 1 to 5 m and 5 to 10 m below water surface, and the seasonally adjusted probability of exceeding a critical activating dose of UVA radiation at that depth in the water column.

Energy East Pipeline Ltd. May 2016 4-27

CA PDF Page 132 of 1038

4.3.2 Potential Narcotic Environmental Effects on Marine Mammals from Inhalation Exposure to Hydrocarbons in Air

Non-polar narcosis as a mechanism to explain the toxicity of hydrocarbon vapours is highly relevant to oil spills as it provides an integrated model that predicts the toxicity of a wide variety of volatile petroleum hydrocarbon compounds. Many quantitative structure-activity relationship (QSAR) models have been developed over the last few decades to predict non-polar narcosis endpoints, including lethality and sublethal environmental effects. However, prior to the work of Veith et al. (2009), these generally were not extended or refined for consideration of the inhalation pathway.

Veith et al. (2009) used QSAR-based experimental and reporting criteria to develop a model to predict non-polar narcosis endpoints for mammalian inhalation. This work included review of the inhalation toxicity literature and compilation of a database of inhalation endpoints that met the QSAR-based criteria, focusing on 4-hour exposure for a single species of rodent (i.e., rat). Central to the QSAR model developed by Veith et al. (2009) is the Ferguson principle, which states that the ratio of the partial vapor pressure needed to produce narcosis to that of the pure chemical should be the same for all non-polar narcotics. Therefore, under steady-state conditions, the partial vapor pressure required for non-polar narcosis should be directly proportional to the partial vapor pressure of the pure chemical. Where the partial vapor pressure required for non-polar narcosis can be converted into a concentration needed to cause lethality (i.e., the LC50), a relationship linking the LC50 to the vapour pressure of the pure chemical can be derived. Based on the compiled database of 4-hour inhalation LC50 values for the rat, Veith et al. (2009) derived the following linear relationship correlating LC50 to the vapor pressure of the pure chemical:

log LC50 = 0.69 log VP + 1.54

3 where LC50 is the chemical concentration in air (mmol/m ) causing lethality in 50% of exposed test animals over the defined exposure period of 4 hours, and VP is the vapor pressure of the chemical (mm Hg).

The potential for inhalation toxicity from crude oil spills is assessed for each pseudocomponent (Table 4-3) but the equation of Veith et al. (2009) was adapted for different units (i.e., VP in Pa and 3 LC50 in µg/m ).

4-28 May 2016 Energy East Pipeline Ltd.

CA PDF Page 133 of 1038

Table 4-3 Mammalian Inhalation Toxicity Benchmarks for Hydrocarbon Pseudocomponents

Vapour 2 3 Pseudo- MW Pressure LC50 component Description 1 (g/mol) (Pa) (μg/m3)

VOL Volatiles 70.8 9.98E+04 2.34E+08

AR1 Benzene 78.1 1.27E+04 6.23E+07

AR2 Toluene, Ethylbenzene, Xylenes 99.2 2.47E+03 2.55E+07

AR3 Aromatics >C8-C10 120.0 1.27E+03 1.95E+07

AR4 Aromatics >C10-C12 130.0 4.14E+00 4.07E+05

AR5 Aromatics >C12-C16 150.0 8.72E-03 6.68E+03

4 AR6 Aromatics >C16-C21 190.0 2.13E-05 –

4,5 AR7 Aromatics >C21-C34 240.0 9.16E-08 –

AL1 Aliphatics C6-C8 100.0 6.38E+03 4.96E+07

AL2 Aliphatics >C8-C10 130.0 6.38E+02 1.32E+07

AL3 Aliphatics >C10-C12 160.0 6.38E+01 3.31E+06

AL4 Aliphatics >C12-C16 200.0 4.86E+00 7.00E+05

AL5 Aliphatics >C16-C21 270.0 1.11E-01 6.98E+04

4,5 AL6 Aliphatics >C21-C34 390.0 2.59E-06 –

4,5 RES1 F4 (>C34-C50) 570.0 1.00E-10 –

RES2 Resins 825.0 1.00E-10 – 4,5

RES3 Asphaltenes 1,599.0 1.00E-10 – 4,5

NOTES: 1 The ranges in this column represent the number of carbon atoms in the hydrocarbon molecule (e.g., aromatics >C8-C10 would be molecules containing at least one benzene ring, with specifically 9 or 10 carbon atoms. 2 MW stands for molecular weight. 3 LC50 stands for lethal concentration (50%). 4 Did not readily evaporate in modelling completed as part of this study. 5 Non-volatile organic compounds according to Kelly et al. (1994).

Energy East Pipeline Ltd. May 2016 4-29

CA PDF Page 134 of 1038

The linear relationship developed by Veith et al. (2009) correlating 4-hour inhalation LC50 values for the rat to the vapor pressure of the pure chemical is expected to be applicable to other mammalian species.

According to Paterson and Mackay (1989) inhalation LC50 may be related to a critical concentration in tissue such as the brain. Volatile organic compounds in tissue exhibit both water and fat solubilities with solubility in water predominating for the alcohols and solubility in lipids predominating for alkanes (Paterson and Mackay 1989). As such, it would be expected that species with similar brain lipid content would demonstrate similar inhalation LC50 for non-polar petroleum hydrocarbon compounds. Literature has shown that brain lipid content is 11% for rats (Pratt et al. 1969) and 11.5% for the grey whale

(Varanasi et al. 1993). Therefore it is expected that the LC50 values presented above, would be applicable to the rat (original test species) and to other mammalian species of similar brain lipid content such as whales.

Using GIS, and the volatile hydrocarbon in air concentration data predicted in conjunction with the oil spill fate and transport modelling using the CALPUFF model, the predicted maximum 4-hour average hydrocarbon-in-air concentrations at the water-air interface in each cell of the model domain can be compared to these benchmark values. For each pseudocomponent, the predicted concentration in air is

divided by the predicted LC50 value, to calculate the number of toxic units (TU) present in air. The TU are then summed across the pseudocomponents to estimate the toxicity of the volatile hydrocarbon mixture in air. The potential for acute toxicity due to narcosis caused by exposure to volatile hydrocarbons is identified where the sum of the TU is equal to or greater than 1.

The model of Vieth et al. (2009) was developed using data for rats, and is here adapted to petroleum hydrocarbons and extended to apply to marine mammals. It is not explicitly applied to marine birds due to increasing uncertainty. However, as a first approximation, it is reasonable to conclude that where the potential for hydrocarbon vapour acute toxicity to marine mammals is identified there may also be a risk to marine birds.

4.3.3 Potential Environmental Effects on Marine Birds and Semi-Aquatic Mammals from Exposure to Surface Oiling

Potential acute environmental effects on wildlife (i.e., birds and furred mammals lacking blubber) in the spill-affected area are evaluated based on probability of encounter with floating oil, and the amount of oil likely accumulated on an individual animal. This analysis is based on the methodology outlined by French et al. (1996) as the CERCLA Type A Natural Resource Damage Assessment (NRDA) Model for Coastal and Marine Environments (under the auspices the U.S. Department of the Interior), and as subsequently reported by French McCay (2009). In the NRDA model, a slick thickness of 10 μm is assumed as a threshold thickness for oiling mortality. Marine birds and furred mammals (e.g., mink or river otter) lacking blubber are assumed to die if exposed to a slick thickness greater than 10 µm. This is not to suggest that exposure to thinner oil slicks may not also be harmful, rather, the probability of mortality is assumed to be lower. To be conservative, and taking into consideration that the Bay of Fundy is a cold water body (i.e., there is a heightened risk of hypothermia for oiled wildlife), and that oil is not likely to be uniformly distributed within model grid squares, the NRDA model threshold of 10 µm thickness for oil slicks causing mortality of exposed marine birds and furred mammals lacking blubber is here reduced to a threshold thickness of 2 µm.

4-30 May 2016 Energy East Pipeline Ltd.

CA PDF Page 135 of 1038

Marine mammals having blubber (specifically seals and whales) are less susceptible to external crude oil exposure than furred mammals lacking blubber, because their ability to thermo-regulate is less affected, or not affected, by exposure to an oil slick. For these animals, as well as sea turtles (which are cold- blooded and therefore still less susceptible to a failure of thermoregulation), external oil exposure may cause irritation of the epidermis but is not likely to rapidly result in death due to hypothermia.

4.3.4 Potential Environmental Effects on Intertidal Communities from Exposure to Shoreline Oiling

Although the Bay of Fundy is part of the Atlantic Ocean system, and far removed from the Gulf of Alaska in the northern part of the Pacific Ocean, there are many similarities between the Gulf of Alaska and Prince William Sound (PWS) ecosystem, and that of the Bay of Fundy and Gulf of Maine. Both ecosystems are cold water marine environments that are generally free from sea ice (although the Columbia glacier releases icebergs into PWS). Both PWS and the outer Bay of Fundy and Gulf of Maine have predominantly rocky shorelines, with a predominance of rockweed habitat in the intertidal zone. Although tides in the Bay of Fundy (particularly in the inner bay) are famed as being some of the largest in the world, tides in PWS are also large (typically around 3 m and greater than 5 m in places), and comparable to tides in the outer Bay of Fundy. Both ecosystems feature marine food webs dominated by pelagic fish (e.g., herring) and krill, and feature whales (baleen and toothed) at the top of their food webs. Both systems also have abundant marine birds (including but not limited to gulls, terns, sea ducks, cormorants, and various auks). The experience from the Exxon Valdez oil spill (EVOS) is therefore both relevant and appropriate as a basis for evaluating the potential environmental effects of hypothetical crude oil spills in the Bay of Fundy and Gulf of Maine. The technical basis for deriving environmental effects endpoints for the intertidal community is described below, with additional detailed information to be found in Appendix D.

Considering the environmental effects of shoreline oiling on intertidal flora and fauna in a north-Pacific environment as an analogue for the Bay of Fundy and Gulf of Maine, the intensity of oiling after the EVOS was spatially variable, ranging from heavy and wide slicks, to thin and discontinuous oiling. Most of the oil affected the upper intertidal zone, with much lower levels of oil deposition on sediments in the lower intertidal and subtidal areas. Environmental effects of oil alone caused mortality of algae and invertebrates, but not to the same extent as the damage associated with shoreline cleanup activities, notably high-pressure warm-water washing. Although the most serious effects of oiling were not well documented during 1989, most researchers documented rapid recovery, already underway in 1990, so that the intertidal zone was deemed to be largely recovered by 1991 or 1992. Effects that have been characterized as playing out over longer periods of time (such as subsequent simultaneous senescence of even-aged Fucus stands that colonized rock surfaces after aggressive cleanup efforts) appear to be largely attributable to remedial actions such as hot-water high-pressure washing that are not likely to be applied in the unlikely event of a crude oil spill in the Bay of Fundy.

Energy East Pipeline Ltd. May 2016 4-31

CA PDF Page 136 of 1038

EVOSTC (2014) acknowledge that:

…by 1991, in the lower and middle intertidal zones, algal coverage and invertebrate abundances on oiled rocky shores had returned to conditions similar to those observed in unoiled areas. However, large fluctuations in the algal coverage in the oiled areas caused a subsequent alteration in community structure. The Fucus canopy was initially eliminated in most of the areas that underwent extensive cleaning, thereby removing the protection from predation, desiccation and abrasion provided by this alga to intertidal organisms. This early eradication of Fucus led to instability of this alga's subsequent populations because the single-aged stands present after re-colonization of the habitat were susceptible to large synchronous die-offs. Until a broader distribution of mixed-aged stands is established, this cycle may continue for many generations. Meanwhile, full recovery of Fucus is crucial for the recovery of intertidal communities at oiled sites, because many intertidal organisms depend on the shelter this seaweed provides.

This evaluation makes it clear that the ongoing environmental effects of the oil spill post-1991 were associated primarily with cleanup effects and not effects of the spilled oil alone.

Initial environmental effects on the flora and fauna of the intertidal zone would occur at all tidal levels and in all types of habitats, although the heaviest oiling would be expected in the upper intertidal zone. Dominant species of algae and invertebrates, including rockweed, limpets, barnacles, mussels, periwinkles, polychaete and oligochaete worms, would be directly affected. The degree of injury would be correlated with the intensity of oiling (taking into consideration the amount of oil loading and the degree of coverage). Within 1 to 2 years on exposed shorelines, and within 2 to 5 years on protected shorelines, the algal coverage and invertebrate abundance on oiled shorelines would be expected to return to conditions similar to those observed in unoiled areas.

Based on the information reviewed from the EVOS (see Appendix C), the following biological effects classes and effect durations were assumed for shoreline communities (intertidal and subtidal algae and invertebrates) (see Table 4-4). At low levels of initial oiling (<10% of initial oiling capacity), effects on intertidal algal and invertebrate communities may be very light, and would be difficult to discern using statistical methods due to the high level of variability found in the intertidal zones. Such areas would be judged to have low levels of effect magnitude, and recovery would occur rapidly. At higher levels of initial oiling intensity, greater effect magnitude will be observed. Time to recovery may also vary by habitat type, with those habitats having greater potential to retain oil (generally, sites having low wave exposure and more porous substrates) having longer effect duration and time to recovery.

4-32 May 2016 Energy East Pipeline Ltd.

CA PDF Page 137 of 1038

Table 4-4 Biological Environmental Effect Magnitude and Duration Associated with Initial Oiling Intensity

Biological Environmental Time to Recover from % of Maximum Initial Oil Effect Magnitude Biological Environmental Site Exposure Retention Capacity (% Initial Loss of Community)* Effect (years)*

Low >90 to 100 90 2 to 5

>50 to 90 50 2 to 5

>10 to 50 25 2

>0 to 10 10 <1

High >90 to 100 90 2 to 3

>50 to 90 50 2 to 3

>10 to 50 20 1 to 2

>0 to 10 10 <1

NOTE: * Assumes appropriate remedial actions are taken, and inappropriate shoreline cleaning methods are not employed.

4.3.5 Potential Environmental Effects of Crude Oil Deposition to Sediment

There are few published regulatory guidelines for petroleum hydrocarbons in sediment. This is due in part to the complexity of hydrocarbon chemistry, which can significantly affect toxicity. Routine analytical laboratory detection limits for petroleum hydrocarbons in sediment typically range from 2.5 to 15 mg/kg, as follows:

• C6-C10 (less BTEX): 2.5 to 10 mg/kg

• >C10-C16: 5 to 10 mg/kg

• >C16-C21: 5 to 10 mg/kg

• >C21-

• >C34: 10 to 15 mg/kg • modified TPH: 15 mg/kg

On this basis (and assuming that crude oil constituents reaching sediment would tend to be weathered and depleted in BTEX and light aliphatic and PAH constituents), total petroleum hydrocarbon concentrations in sediment that are below about 15 mg/kg are typically around or at the detection limit, or are non-detectable. Marine sediments in urban areas typically have a measurable background concentration of petroleum hydrocarbons. This background is derived from a number of sources, including but not limited to runoff from urban land areas (e.g., runoff from roads and parking lots), losses of hydrocarbons such as fuels and lubricating oils used by marine shipping, and spills. In the vicinity of New York Harbour, the median TPH concentration was 1,360 mg/kg, with a range of 25 to 12,600 mg/kg,

Energy East Pipeline Ltd. May 2016 4-33

CA PDF Page 138 of 1038

and the predominant source appeared to be motor oil, presumably from stormwater runoff (Brownawell et al. 2007).

The technical basis for evaluating potential environmental effects of crude oil deposition to sediment is described below, with additional detailed information to be found in Appendix C. For this risk assessment, petroleum hydrocarbon deposition to sediment is estimated as a mass per unit area (i.e., g/m2). Assuming that oil deposited to sediment is initially mixed into the surface 1-cm layer of sediment, and that sediment in this layer has a bulk density of between 1.0 and 1.2, then the surface 1-cm layer of one square metre would have a volume of 0.01 m2, and a wet mass of 10 to 12 kg. The following relationships would then apply:

• oil deposition to sediment of <0.1 g/m2 would result in a surface sediment TPH concentration of <10 mg/kg, and would be non-detectable in routine chemical analysis • oil deposition to sediment of 1 g/m2 would result in a surface sediment TPH concentration of <100 mg/kg, and would be unlikely to result in detectable biological effects • oil deposition to sediment of 5 g/m2 would result in a surface sediment TPH concentration of <500 mg/kg, which could result in adverse environmental effects on a limited number of sensitive species, but could increase overall benthic community productivity due to the enrichment effect • oil deposition to sediment of 20 g/m2 would result in a surface sediment TPH concentration of <2,000 mg/kg, which would be expected to cause reduced benthic community diversity, biomass and productivity • oil deposition to sediment of greater than 20 g/m2 would be expected to cause progressively more serious reductions in benthic community diversity, biomass and productivity

4.3.6 Chronic Environmental Effects of Crude Oil on Mammals and Birds

Birds and mammals living in the marine environment may be chronically exposed to the effects of low levels of crude oil or PAHs following an oil spill. Such exposures, which are assumed to be dominated by oral exposure due to consumption of prey or sediment containing traces of hydrocarbons, are evaluated here at two points in time. The first corresponds to a point in time about 4 weeks after the spill, when there is very little oil remaining on the surface of the water, although oily residues on shorelines are still being responded to. At this time the oil is somewhat weathered, and hydrocarbon concentrations in some nearshore or shoreline environments may remain relatively high. The second represents a time 1 to 2 years after the spill, when the oil is more weathered, and residual hydrocarbon concentrations are lower.

The selection of an assessment time point that is 1 to 2 years after the spill reflects the fact that a hypothetical oil spill could occur at any time. Crude oil may be expected to weather more quickly during warm weather than during cold weather periods. However, a spill that occurred in the spring would undergo one winter and two summers within 16 months of the event, whereas a spill that occurred in the fall would undergo two winters and only one summer during an equivalent period of time. The selection of an assessment time of 1 to 2 years after a spill is based on 1 year of model weathering, but is intended to reflect the reality that oil will weather more rapidly during some periods of the year.

4-34 May 2016 Energy East Pipeline Ltd.

CA PDF Page 139 of 1038

For the hypothetical spills in the Bay of Fundy, the potential for chronic environmental effects on mammals or birds caused by chronic oral ingestion of crude oil or PAHs is evaluated starting about 4 weeks after the spill initiation, and again after 1 to 2 years.

4.3.6.1 Selection of Exposure Point Concentration Values for Chronic Environmental Effects Assessment

The following processes were followed in order to determine appropriately conservative exposure point concentration (EPC) values on which to base the exposure assessment for mammals and birds:

• For organisms that are exposed primarily to COPC in the water column, including pelagic fish and squid, and mussels attached to rocks or other substrates emergent from the sediment, the maximum 24-hour average hydrocarbon concentrations (at any depth in the water column) were identified for each grid cell in the marine oil spill fate and transport model. From these values, the 95th percentile of non-zero values is selected as the EPC for areas affected by spilled crude oil. These concentrations were used to estimate bioaccumulation of hydrocarbons by pelagic organisms. • For organisms that are exposed primarily to COPC associated with intertidal sediment, including bivalves, gastropods, crustaceans, and intertidal zone fish, the impingement of crude oil onto non- permeable beach substrates (including mud, sand, gravel and cobble beach units) is estimated, and the 95th percentile of non-zero values is selected as the EPC for areas affected by spilled crude oil. It is further assumed that oil spill response activities would result in cleaning of beaches if the 95th percentile value exceeded 1 L of crude oil per square metre of intertidal sediment. Therefore, the value of 1 L/m2 was used as a cap. A model (see Appendix D) is then used to simulate the weathering of crude oil in intertidal beach sediments, and calculate the residual concentrations of crude oil pseudocomponents and PAHs in sediment interstitial water, and on the sediment organic carbon fraction, in addition to whole sediment. • For organisms that are exposed primarily to COPC associated with subtidal sediment, including bivalves, crustaceans, and demersal fish, the deposition of crude oil onto subtidal sediment (g/m2, assumed to be silty or muddy sediment) is estimated, and the 95th percentile of non-zero values is selected as the EPC for areas affected by spilled crude oil. A model (see Appendix D) is then used to simulate the weathering and burial of crude oil in subtidal sediments, and the residual concentrations of crude oil pseudocomponents and PAHs in sediment interstitial water, and on the sediment organic carbon fraction, in addition to whole sediment, were calculated.

Concentrations of bioavailable pseudocomponents of the crude oil, and PAHs, in the tissues of exposed fish, molluscs and crustaceans were estimated following methods that are described below.

Energy East Pipeline Ltd. May 2016 4-35

CA PDF Page 140 of 1038

4.3.6.2 Toxicity Benchmarks for Exposure of Mammals and Birds to Total Petroleum Hydrocarbons

Toxicity reference values (TRVs) for the ERA are based on dose response studies, typically conducted with laboratory animals where the lowest observed adverse effects level (LOAEL) or no observed adverse effects level (NOAEL) has been quantified. The TRVs used in this risk assessment were determined from studies in which endpoints were derived from the administered dose, rather than the absorbed dose. This is a conservative approach because compounds are often administered in a more bioavailable form than would be found in the environment.

The preferred toxicity measure used for derivation of TRVs in the ERA is the LOAEL since this value corresponds to the onset of toxicologically induced responses in test animals. However, in the absence of a suitable LOAEL, more conservative NOAEL-based TRV values can also be used. Generally, LOAEL values used for TRV derivation are based on long-term growth or survival, or sublethal reproductive effects determined from chronic exposure studies. As such, these endpoints are relevant to the maintenance of wildlife populations, as the LOAEL represents a threshold dose at which adverse outcomes are likely to become evident. This threshold is considered an appropriate endpoint for ERA since TRVs are used as the denominator in the risk calculation, and hazard quotients (HQs) equal to or greater than one may be considered indicative of potential adverse environmental effects. Hazard quotients calculated with NOAEL-based TRVs are more conservative since NOAEL refer to a threshold at which no relevant toxicological effects from COPC exposure are observed.

Numerous sources were reviewed to obtain the most relevant TRVs for ecological receptors. Information sources reviewed include, but are not limited to:

• Oak Ridge National Laboratory Toxicity Benchmarks for Wildlife (Sample et al. 1996) • U.S. EPA Ecological Soil Screening (EcoSSL) documents • Agency for Toxic Substances and Disease Registry • Canadian Environmental Protection Act, Priority Substance List Assessment Reports • primary scientific literature

The potentially harmful environmental effects of chronic oral exposure of wildlife receptors to petroleum hydrocarbons are evaluated in several ways. For some compounds, which are recognized to be important components of crude oil (e.g., the BTEX compounds, certain PAH compounds, and the CWS petroleum hydrocarbon fractions), sufficient toxicological data exists to determine whether toxicity is likely to occur due to exposure to individual compounds. Some compounds (notably the F4 fraction under the CWS fractionation scheme, and asphaltenes) are assumed to have negligible bioavailability, and hence are often not considered to be active contributors to the potential toxicity of hydrocarbon mixtures. However, there are gaps in the toxicological database, and data are lacking for many of the identified petroleum hydrocarbon compounds. For this reason, the following approach has been taken to evaluate the chronic toxicity of petroleum hydrocarbons derived from crude oil.

4-36 May 2016 Energy East Pipeline Ltd.

CA PDF Page 141 of 1038

Mammals

Mammals at risk from oil-related injuries include those that frequent water, including river otter and mink. These animals spend much time in water, have high site fidelity, and rely on fur to maintain thermoregulation. As a result, the potential for death from hypothermia to occur as a result of external oiling is of greatest concern, although subsequent (acute or chronic) exposure to ingested hydrocarbons is also a concern. Other terrestrial mammals of lesser concern include those less intimately associated with water bodies and riparian habitat such as raccoons and foxes, as well as potentially deer or domestic animals such as dogs or sheep (if present on island or mainland pastures without barriers to beach access). These animals have the potential to be affected by the chronic consumption of oiled food items, as well as by direct contact and habitat degradation (Hugenin et al. 1996).

Cattle will voluntarily ingest large doses of petroleum substances (Coppock et al. 1996). In such acute poisoning cases the lung is the target organ. Chemical pneumonia results when droplets of oil are inhaled. Another cause of chemical pneumonia is aspiration of hydrocarbons during vomiting, regurgitation or eructation. Lung lesions have been reported following the voluntary ingestion of petroleum by cattle. Such lesions have been reported in cattle given 20 to 60 mL crude oil/kg body weight, and in sheep after a 1-day exposure to water contaminated with natural gas condensate, which also caused reddening and hemorrhage in the digestive tract. Kidneys can also be target organs of petroleum hydrocarbon toxicosis (Coppock et al. 1996).

CCME (2008) derive a value for the toxicity of petroleum hydrocarbons to livestock (with a focus on fresh crude oil). A lowest documented effects dose of 2.1 g fresh crude/kg body weight/day is reported. This is divided by an uncertainty factor of 10 to obtain a Daily Threshold Effects Dose of 210 mg/kg body weight/day, which is then used to estimate the allowable concentration of whole fresh crude oil ingested in livestock drinking water. However, it is noted that the value for weathered crude oil could be 3.7 times higher, due to the lower toxicity of weathered crude oil (CCME 2008). A value of approximately 0.78 g/kg body weight/day could then be appropriate for chronic ingestion of weathered crude oil.

Other studies evaluating the toxicity of fresh or weathered crude oil to mammals are presented in Table 4-5. These studies include cattle, rats and ferrets, and therefore represent herbivores, omnivores and carnivores. The results from these studies are highly consistent with respect to the dose ranges that have been shown to have adverse effects on mammals. Based on the available information it is concluded that mammals are generally quite tolerant of exposure to fresh or weathered crude oil. Adverse environmental effects are considered unlikely at dose levels less than 0.5 g/kg body weight/day.

There is a general lack of literature describing the oral toxicity of crude oil or PAH exposure to marine mammals, including seals, sea lions or whales. However, it is reasonable to assume that the sensitivity of those species groups is generally similar to the sensitivity of other mammals. The TRVs for mammals exposed to crude oil are therefore also expected to be in the range of 0.5 g/kg body weight/day.

Energy East Pipeline Ltd. May 2016 4-37

CA PDF Page 142 of 1038

Table 4-5 Toxicological Benchmarks for Mammalian Receptors Exposed to Crude Oil

Study Design Effects Benchmark Derivation Reference

Domestic cattle (8/group) were No cattle died as a result of being dosed with crude oil. Acute toxicity greater than Khan et al. (1996) administered single oral doses of Pembina Cattle treated with 16.7 mg/kg body weight showed minimal signs 67.4 g/kg body weight. Cardium crude oil at 16.7, 33.4 and of intoxication; cattle treated with higher doses exhibited tremors, 67.4 g/kg body weight. Cattle were nystagmus, regurgitation and vomiting, myoclonic seizures, sacrificed at days 7 or 30 for pathological depression, locomotory abnormalities and pulmonary distress. and histological examination, as well as measurement of a suite of enzyme On day 7, cattle in exposed groups showed alteration in enzyme activities. activities in liver, kidney and lung tissues. Cattle sacrificed on day 30 showed few statistically significant differences from control animals, and reduced differences in cytochrome P-450 activity.

Sprague-Dawley rats were given doses of No rats died as a result of being dosed with crude oil or diesel No sign of distress or Khan et al. (2001) 0.25, 0.50 or 1.25 mL/kg Pembina fuel, and there was no sign of distress of intoxication in the intoxication in rats Cardium crude oil, or 1.25 mL/kg exposed animals. exposed to crude oil at up commercial diesel fuel, on days 1, 3, 5 Dose-dependent changes were observed in levels of a suite of to 1.25 mL/kg body and 8 of the study, and were sacrificed on enzyme indicators (including EROD). weight. day 10. Tissue and blood samples were tested for a suite of enzyme activities, The only significant systemic change was a small increase in the hematology and blood chemistry, and liver somatic index of rats exposed to the highest dose of crude oil pathological examination. or diesel fuel.

The toxicity of naturally weathered Exxon No mortality of ferrets occurred as a consequence of being Subacute toxicity of crude Stubblefield et al. Valdez crude oil was tested in a battery of administered crude oil. No grossly observable signs of toxicity oil to European ferrets is (1995a) acute and subchronic tests using were noted. No effects on mean body weight were detected. No >5 g/kg body weight/day. European ferrets (Mustela putorius). grossly observable signs of toxicity were noted during postmortem Acute toxicity of three Smith et al. (1980) Young adult male and female ferrets were examination of the ferrets. Microscopic examination of tissues did unweathered crude oil administered oil at a dose of 0.5, 1.0 or not reveal lesions considered to be related to oil exposure. samples to mice ranged 5.0 g/kg body weight once daily for three Significantly lower mean spleen to body weight ratios and raw from >10 to 16 g/kg body days. Prior to the study and at termination, spleen weights were noted in all female treatment groups. No weight. blood samples were taken for chemistry other organ weight differences were noted. and enzyme testing, and the animals were weighed. At study termination (day 5) the With the exception of lower mean serum albumin concentrations animals were subject to gross necropsy in the 5 g/kg female dose group, no significant differences among examination and selected tissues were clinical chemistry parameters were noted. No significant taken for histological examination. differences in the hematological parameters were noted in any group.

4-38 May 2016 Energy East Pipeline Ltd.

CA PDF Page 143 of 1038

Birds

Alcids (the family of web-footed diving birds with short legs and wings that include the auks, murres, and puffins) are considered to be the most vulnerable of bird groups to oil, due to their tendency to form large flocks and to spend much time in coastal and offshore waters, where they are obliged to dive to capture prey. Large scale mortality to eggs is also possible due to their tendency to form large breeding colonies, making them vulnerable to oiling in the event of an accident, if parent birds return to the nest site with oiled plumage. Among waterfowl, populations of dabbling ducks are generally less exposed because they tend not to form large colonies. Therefore, both adult mortality and environmental effects on eggs are less likely (at the population level) due to dispersion. Direct mortality rates for shorebirds are generally lower because they spend less time in the water.

Wading birds and shorebirds generally experience low mortality because they wade in shallow, sheltered waters to feed. However, plumage can become contaminated due to wading through oiled vegetation or exposure to oil slicks. Indirect effects can also occur due to loss of prey resulting in starvation, or shifting to alternative foraging sites. Raptors become oiled primarily via consumption of oiled prey or carrion. Oil exposure can also cause reproductive effects if eggs are oiled, or if nests are disturbed as a result of shoreline cleanup operations (Hugenin et al. 1996). Rapid death due to hypothermia is most likely to occur in birds that occupy habitats on or in the water, due to their higher likelihood of oiling, and the rapid heat loss associated with water contact. Birds such as gulls that have a higher level of behavioural flexibility than auks, or that have less contact with water, are more likely to survive light oiling (e.g., Camphuysen 2011).

Chronic low levels of oil pollution may have adverse environmental effects on aquatic bird populations. Small amounts of oil applied to the external surface of bird eggs are toxic. Single oral doses of oil have been demonstrated to cause lipid pneumonia, gastrointestinal irritation, and fatty livers. Pathological responses of birds examined after fatal exposure to Bunker C oil included enteritis, hepatic fatty changes, and renal tubular nephrosis (Szaro et al. 1978).

Other studies evaluating the toxicity of fresh or weathered crude oil to birds are presented in Table 4-6. These studies focus on mallard ducks, waterfowl which would be highly exposed to hydrocarbons in the event of a spill. Based on the available information it is concluded that birds are generally quite tolerant of exposure to un-weathered or weathered crude oil. The lowest reported adverse effects on a reproductive endpoint are identified at a dose of approximately 0.2 g/kg body weight/day.

Risks to mammals and birds were evaluated on the basis of their exposure to the individual BTEX compounds, as well as to the CWS petroleum hydrocarbon fractions, and individual (un-substituted) PAH compounds. The toxicity of these individual components is considered to be additive. In addition, to capture the potential toxicity of alkylated MAH and PAH compounds, as well as other hydrocarbon substances not otherwise accounted for, the total hydrocarbon mixture is summed to represent the total exposure of wildlife receptors to weathered crude oil, and the toxicity of the summed fractions will be compared to empirically derived benchmark values.

Energy East Pipeline Ltd. May 2016 4-39

CA PDF Page 144 of 1038

Table 4-6 Toxicological Benchmarks for Avian Receptors Exposed to Crude Oil

Study Design Effects Benchmark Derivation Reference

The toxicity of naturally weathered Exxon Valdez No adult ducks died following single doses Weathered Exxon Valdez crude Stubblefield et al. crude oil was tested in a battery of acute and of 5 g/kg body weight; no grossly oil presented little potential for (1995a) subchronic tests using mallard ducks. Adult ducks observable signs of toxicity were noted, and acute toxicity to wildlife species were tested with an acute oral dose of 5 g/kg body there were no significant effects on feed from oral ingestion. Lethal weight and observed for up to 14 days following consumption or body weight. No treatment concentrations and no-observed testing (acute oral toxicity); Five-day old ducklings related abnormalities were noted on adverse effect levels were were tested by feeding them a diet containing postmortem examination. greater than the maximum tested weathered crude oil at a concentration of 50 g/kg No mortality or observable sign of toxicity doses (>5 g/kg body weight in diet (subacute dietary toxicity) for 5 days, followed was noted in ducklings fed crude oil in their the oral study, and >50 g/kg diet by a 3-day observation period on uncontaminated diet at 50 g/kg body weight. Food in the subacute dietary tests). ration; food avoidance was tested using ducklings consumption was not affected, and not offered diet containing 0, 1.25, 2,5, 5, 10 and significant differences in body weight or 20 g/kg diet for 5 days, followed by a 3-day growth were noted. Post-mortem observation period on uncontaminated ration; and examination showed no evidence of ducks (16 weeks old) were fed a diet containing systemic toxicity. 0, 10, 30 or 100 g/kg weathered crude oil for 14 days. Ducklings did not avoid food containing crude oil at up to 20 g/kg. No significant differences in body weight or growth were found, and no consistent grossly observable lesions were noted in postmortem examination.

No mortalities or grossly observable signs of LD50 values for refined Hartung and Hunt (1966) toxicity were noted in 14-day exposure to hydrocarbon products were as cited by Stubblefield dietary concentrations of up to 100 g/kg diet. reported to range from 7 to et al. (1995a). No significant treatment related differences 20 mL/kg body weight. in clinical blood chemistry was noted between treatment and control birds. No consistent or substantive differences were noted among the histological appearance of the kidney, thymus, brain or bone marrow of high dose birds when compared to control birds. Spleens and livers of high dose birds were found to show some minor changes when compared to control birds.

4-40 May 2016 Energy East Pipeline Ltd.

CA PDF Page 145 of 1038

Table 4-6 Toxicological Benchmarks for Avian Receptors Exposed to Crude Oil

Study Design Effects Benchmark Derivation Reference

A one-generation reproductive toxicity study and a No deaths of ducks occurred that were Weathered crude oil is Stubblefield direct eggshell application toxicity study were attributed to crude oil exposure. All surviving substantially less toxic to mallard et al. (1995b) conducted using naturally weathered crude oil birds appeared healthy throughout the ducks than unweathered crude obtained following the EVOS. Mallard ducks, study, and no signs of toxicity were noted. oil. 16 weeks of age, were exposed to dietary No statistically significant differences in Ingestion of a diet containing concentrations of 0, 0.2, 2 and 20 g/kg diet for growth of birds or food consumption were weathered crude oil at 20 g/kg 22 weeks. Eggs laid between weeks 12 and 22 of noted. caused reductions in eggshell exposure were incubated and hatched. Mallard Consumption of diets containing crude oil at thickness and strength, which eggs were also treated with either weathered 20 g/kg feed resulted in changes in clinical could result in reduced hatching crude oil or Vaseline (a non-toxic control) to chemistry parameters (i.e., serum success of ducklings. determine the extent of coverage causing reduced phosphorus, total protein, albumin, bilirubin viability. and calcium), reductions in eggshell thickness and strength (although the viability of embryos was not affected), and suggested liver and spleen weight changes. No significant effects were noted at dietary concentrations of 0.2 or 2 g/kg feed. Long- term ingestion of weathered crude oil at dietary concentrations of 20 g/kg feed may result in reduced egg fitness. Application of weathered crude oil to areas of up to 33% of the shell area had no appreciable effect on embryo survival, suggesting that not only is it substantially less toxic than unweathered crude oil, but that is not as effective as a shell sealant as Vaseline, which caused effects when 17% or greater of the egg shell was covered. However, the severity of effects based on dietary exposure to weathered crude oil was considerably less than has been reported in studies using unweathered crude oil.

Energy East Pipeline Ltd. May 2016 4-41

CA PDF Page 146 of 1038

Table 4-6 Toxicological Benchmarks for Avian Receptors Exposed to Crude Oil

Study Design Effects Benchmark Derivation Reference

Fresh South Louisiana crude oil was fed to Growth was depressed in birds receiving a Exposure to fresh crude oil over Szaro et al. (1978) mallard ducklings at concentrations of 0.025, 0.25, diet containing 5% oil but there was not oil- an 8 week period caused 2.5 and 5% of diet from hatching to 8 weeks of related mortality. Diets containing as low as impaired development of mallard age. 0.25% oil caused behavioural response. ducklings at a dose of 0.824 g/kg Liver hypertrophy and splenic atrophy were body weight/day. evident in birds fed 2.5% or 5% oil. Some biochemical effects were noted, and tubular inflammation and degeneration in the kidney were noted in birds fed the 5% diet. High concentrations of oil in the diet impaired development of the wings and flight feathers and caused stunting.

Fresh South Louisiana crude oil was fed to No birds died during the study, nor were Exposure to fresh crude oil over Coon and Dieter (1981) mallard ducks at concentrations of 0.25 and 2.5% body weights significantly depressed. a 26 week period resulted in of diet for 26 weeks. Oviduct weight was greatly reduced on reduced egg production. The necropsy in ducks on the 2.5% diet, and reduction was about 14% in was also significantly reduced in ducks on ducks fed 0.25% oil in diet, and the 0.25% diet. Egg production was lower in was accompanied by reduced ducks fed oil in the diet, however, the oviduct weight. The 0.25% diet hatchability of eggs was not significantly equates to a dose of different, and there was no effect on approximately 0.2 g/kg body eggshell thickness. No significant effects weight/day. were observed on liver weight, although spleen weight was reduced on the 2.5% diet.

4-42 May 2016 Energy East Pipeline Ltd.

CA PDF Page 147 of 1038

4.3.6.3 Toxicity Benchmarks for Exposure of Mammals and Birds to Polycyclic Aromatic Hydrocarbons

While toxicological data are available for a number of individual polycyclic aromatic hydrocarbons (PAHs) compounds (e.g., naphthalene, phenanthrene, benzo(a)pyrene), they are not available in explicit form for most alkyl PAH compounds. In addition, many of the studies that are available are of limited utility due to details of study design and other factors. The U.S. EPA (2007) conducted a screening evaluation of PAH toxicity to wildlife and other ecological receptors as part of the ecological soil screening level (EcoSSL) process. Within this process, PAHs were divided into low (3 or fewer ring) and high molecular weight (4 or more ring) classes. An extensive literature search identified 5,478 papers that contained PAH toxicity data potentially suitable for developing mammalian or avian TRVs. Of these, only 46 studies met acceptance criteria, and only two contained data for avian species. This was judged by U.S. EPA (2007) to be insufficient to support the development of a TRV for avian receptors.

Mammals

For mammals, Harwell et al. (2010) re-evaluated the mammalian data reported by U.S. EPA (2007). The EcoSSL report stated that the best use of the available data was to calculate total doses of low- or high- molecular weight PAHs. However, Harwell et al. (2010) were able to present the data in terms of TPAH exposure, and selected a geometric 95% lower confidence interval as the TRV representing exposures based on both the NOAEL and the LOAEL. From Harwell et al. (2010), the selected TRVs for TPAH exposure for the NOAEL and LOAEL are 51.8 and 83.6 mg/kg body weight/day, respectively, and these values are adopted for mammalian exposures in the present ERA.

Birds

Harwell et al. (2012) reviewed the U.S. EPA (2007) EcoSSL report for PAHs and concluded that the avian toxicity data were not suitable as a basis for deriving TRVs to evaluate the exposure of harlequin duck to residual PAHs from the EVOS. This was due to limitations regarding the species tested, as well as the limited chemical representation (singular low- and high-molecular weight compounds), which did not support extrapolation to a TPAH mixture. Upon further review, however, they concluded that the studies of Stubblefield et al. (1995a, b) could provide a suitable basis. Using the studies by Stubblefield et al. (1995a, b), Harwell et al. (2012) derived NOAEL-based TRVs of 2.00 and 2.14 mg/kg body weight/day and LOAEL-based TRVs of 19.56 and 22.01 mg/kg body weight/day, for male and female mallard ducks, respectively.

Harwell et al. further reviewed studies that had been identified by U.S. EPA (2007) to determine whether data were available that could be compared with the proposed TRVs. Studies appropriate for comparison included work involving Cassin’s auklet, black oystercatcher, Leach’s storm petrel, sanderling, common murre, wedge-tailed shearwater, black-legged kittiwake, herring gull and Atlantic puffin. Although these studies did not individually provide information from which a TRV could be derived, they did not contra- indicate the TRV derived by Harwell et al. (2012). On this basis Harwell et al. (2012) concluded that the NOAEL and LOAEL TRVs calculated using data for mallard duck exposed to weathered Exxon Valdez crude oil provide conservative estimates for TRVs that can be applied to other species. For the present study, the TRVs proposed by Harwell et al. (2012) for male and female birds are combined to provide avian NOAEL and LOAEL TRVs for ingestion of TPAH of 2.0 and 20 mg/kg body weight/day.

Energy East Pipeline Ltd. May 2016 4-43

CA PDF Page 148 of 1038

4.4 Ecological Risk Characterization

The purpose of risk characterization is to evaluate the evidence linking COPC with adverse environmental effects by combining information from the exposure and hazard assessments. The potential for adverse environmental effects was quantified by comparing the concentration or dose of a substance that can be tolerated, or below which adverse environmental effects are not expected (i.e., benchmark), to the expected EPC. The quotient of the two ([COPC concentration]/[benchmark]) is referred to as a hazard quotient (HQ). Acute and/or chronic hazard indices (HI) are derived for chemicals with similar modes of action and target organs by summing the HQ of individual COPC. An acute or chronic HI value less than 1.0 indicates that the exposure concentration is less than the threshold of toxicity for the chemical class evaluated. Given the conservative approach to the estimation of exposure and selection of benchmarks, a chronic HI less than 1 is not expected to result in adverse environmental effects and no further assessment is required. Since the acute to chronic ratio for toxicity of non-polar narcotic substances has a value of approximately 5 (Di Toro et al. 2000), a chronic HI value greater than 5 may also be used as an indicator of potential acute environmental effects.

4.4.1 Hazard Quotients and Hazard Indices An HQ is derived by dividing the EPC in environmental media (e.g., soil, sediment, water), or the dose of a COPC ingested by a receptor organism as a result of its exposure to foods as well as other environmental media, by a benchmark value representing a safe concentration or dose. If a hazard quotient value is less than unity, there is no meaningful risk present. Where chemical substances have additive interactions (as is generally assumed to be the case for hydrocarbon exposure), the hazard quotient values for two or more substances may be summed to estimate an HI for a mixture of substances having similar chemical structure, mode of toxic action, and target tissue or organ. If a hazard index value is less than unity, there is no meaningful risk present.

When used in risk assessment, toxicity benchmark values may be derived from regulatory standards, or may be derived from toxicological studies for individual compounds (such as benzene). In either case uncertainty factors or safety factors may be applied to results that have been selected to represent the lower range of concentrations demonstrated to cause a toxicological response in one or more receptor species. As a result, while it is reasonable to believe that an adverse environmental effect is not likely to occur as long as a particular benchmark value is not exceeded (or the HI/HQ value is less than 1), it does not follow that an adverse environmental effect is likely to occur if the benchmark value is exceeded (or the HI/HQ value is greater than 1). Rather, an HI or HQ value exceeding 1 should be interpreted as being indicative of the potential for an adverse environmental effect to occur. A variety of factors, including the nature of the adverse environmental effects identified in the derivation of the toxicological or regulatory benchmark value, as well as a more rigorous analysis of the nature and extent of conservatism built into the underlying components of the assessment, should be considered.

4-44 May 2016 Energy East Pipeline Ltd.

CA PDF Page 149 of 1038

4.4.2 Chemical Interactions Risk assessments are complicated by the fact that most toxicological studies are conducted using a single chemical whereas environmental exposures generally involve more than one contaminant. Calculating a HQ for exposure to mixture of COPC is problematic because all COPC do not have the same modes of action, target endpoints or magnitudes of toxicity. Chemicals in a mixture may interact in four general ways to elicit a response:

• non-interacting – chemicals do not produce a response in combination with each other; the toxicity of the mixture is the same as the toxicity of the most toxic component of the mixture • additive – chemicals have similar targets and modes of action but do not interact; the hazard for exposure to the mixture is simply the sum of hazards for the individual chemicals • synergistic – there is a positive interaction among the chemicals such that the response is greater than would be expected if the chemicals acted independently or in an additive manner • antagonistic – there is a negative interaction among the chemicals such that the response is less than would be expected if the chemicals acted independently or in an additive manner

There are chemical classes that have similar modes of action and target organs (i.e., they act in an additive manner), and in these cases, an appropriate characterization of risk is achieved by summing the HQ for each compound. HQ for BTEX and TPH are summed to derive a single HI. The PAH substances are also treated as a class, but are not included with BTEX and TPH since measures of TPH also include the specific PAH compounds.

Energy East Pipeline Ltd. May 2016 4-45

CA PDF Page 150 of 1038

4-46 May 2016 Energy East Pipeline Ltd.