AMBIENT NOISE in the KITIMAT FJORD SYSTEM by Eleanor Irene

AMBIENT NOISE IN THE KITIMAT FJORD SYSTEM

Eleanor Irene Heywood

Dr. Douglas Nowacek, Advisor

April 28, 2016

Masters project submitted in partial fulfillment of the requirements for the Master of Environmental Management degree in

the Nicholas School of the Environment of

Duke University

Table of Contents

EXECUTIVE SUMMARY ...... 3 INTRODUCTION ...... 5 SHIPPING AS A SOURCE MECHANISM ...... 6 IMPACTS OF NOISE ON MARINE MAMMALS ...... 7 STUDY SITE ...... 8 PREVIOUS ACOUSTIC RESEARCH ...... 10 SPECIES AT RISK ACT (SARA) AND CRITICAL HABITAT DESIGNATION ...... 11 METHODS ...... 14 ACOUSTIC RECORDING EQUIPMENT ...... 14 FIELD PROCEDURES ...... 17 RESULTS ...... 21 DISCUSSION & RECOMMENDATIONS ...... 27 ACKNOWLEDGEMENTS ...... 30 REFERENCES ...... 31

For more information contact: [email protected]

EXECUTIVE SUMMARY Sound is an important medium for communication and marine organisms have evolved to capitalize on the efficiency with which sound energy travels through water. Anthropogenic and natural sound sources contribute to ocean ambient noise, which can interfere with the use of this sensory modality by marine animals. Anthropogenic noise sources have been increasing steadily over recent decades largely due to coastal population growth, increased global transportation, and offshore industrialization. Understanding the potential impacts of anthropogenic noise requires the establishment of ambient acoustic baselines from which to measure change. Establishing baselines, especially in quiet areas still largely unaffected by anthropogenic stressors, is particularly crucial in the face of the expansion of offshore industries, increasing coastal population and growing reliance on the ocean for global transportation. Global demand for liquid natural gas (LNG), catalyzed primarily by a growing Asian market, and is expected to increase significantly in the next 20 years (Ministry of Energy and Mines, British Columbia). The geographic position of British Columbia relative to these markets, a growing supply of LNG and new technology for extraction and shipping situate British Columbia as a strong competitor in the lucrative market. The LNG industry could have many adverse impacts on these territories and ecosystems. The Kitimat Fjord System is slated for the development of these LNG export facilities increasing shipping traffic for the port and thus increasing ambient noise in the fjord system. The purpose of this study is to 1) quantify the existing sound levels in the area surrounding Gil Island and 2) identify potential source mechanisms in order to provide a baseline study of the acoustic environment in the Kitimat Fjord system prior to potential increases from LNG shipping. Overall, the acoustic environment is quiet relative to areas along the southern coast of British Columbia that are subject to more vessel traffic such as Georgia and Haro Strait. The median ambient noise levels reported here for the low, mid, high and broad bandwidths in the area off of Gil Island are 78, 83, 87 and 89 db re 1 µPa respectively, and are consistent with levels recorded at similar sites within the Kitimat Fjord system, Caamano, Kitimat, and Kitkiata inlet.

Analysis of the hourly Leq percentiles revealed no obvious diel trend in the data, though it is possible that a small diel trend may be obscured by high amplitude noise events generated by ships transiting the channel. The compression of the lower 50 percentile indicates that half of the time noise falls within 6 dB of the quietest ambient noise. Visual identification of loud sounds revealed that within the low and mid frequencies high amplitude event noises are

3 dominated by storm and surge noise as well as noises associated with vessel traffic. There is a strong vessel noise signal in the low and mid frequencies between the hours of 0800 and 1800. Specifically, 1400 and 1500 are dominated by vessel noise in all of the frequency bandwidths corroborating the strong peaks in the 99.9 percentiles at 1500 (Figure 13). The high and broad bandwidths are more dominated by vessel noise indicating potentially fewer higher frequency high amplitude event noises in the geophony and biophony of Kitimat Fjord system. Sounds generated by rain also contribute to the high-energy aspect of the ambient acoustic environment. In addition to the dominant sources, naturally occurring seismic activity and several sounds associated with marine mammals registered as some of the loudest sounds recorded during the deployment. These included humpback whale, killer whale and potential pinniped vocalizations. Though wind is a dominant source of noise in many acoustic environments, it was not strongly correlated to ambient noise in waters near Gil Island. There are multiple explanations for this; 1) wind speed data was taken from weather buoys ~90 km distance from acoustic mooring and may not be representative of wind conditions at the study site, 2) there may be a time lag between when wind speed increases and when sound levels increase due to the topography or fetch in the study site; 3) wind is simply not the dominant driver of ambient noise in waters surrounding Gil Island. This study provides baseline ambient acoustic levels for Kitimat Fjord system surrounding Gil Island and within a proposed shipping lane for future natural gas exports. The waters surrounding Gil Island are thought to be among the quietest soundscapes in British Columbia and increasing industrialization of the area is likely to increase the ambient noise and could have deleterious impacts on organisms that rely on the area for critical components of their life cycle. This area surrounding Gil Island is designated critical habitat for humpback whale populations and proposed for transient killer whales. The results of this analysis suggest that though the current ambient levels are low relative to busy shipping ports along the southern coast of British Columbia, vessel noise still contributes significantly to ambient noise levels and accounts for a majority of the highest intensity sounds over the duration of the deployment. It is recommended that further monitoring take place to fill seasonal and weather data gaps and determine diel or seasonal usage by acoustically sensitive species including marine mammals and fish. Determining the daily and seasonal habitat use could help with the designation of time/ area closures with limited vessel allowances. Vessel speed restrictions are also recommended as these are proven to reduce acoustic input from vessels, increase fuel efficiency and reduce the likelihood of ship strikes.

4 INTRODUCTION In a turbid and viscous environment, sound is an important medium for communication and marine organisms have evolved to capitalize on the efficiency with which sound energy travels through water. Anthropogenic and natural sound sources contribute to ocean ambient noise, which can interfere with the use of this sensory modality by marine animals. Sounds from the biophony, geophony and anthrophony combine to create a soundscape, and dominant sound sources within this soundscape may vary spatially, temporally and within frequency bands. Anthropogenic noise sources have been increasing steadily over recent decades largely due to coastal population growth, increased global transportation, offshore industrialization (Halpern et al., 2008; Shannon et al., 2015). Additionally, extractive industries such as oil and natural gas and mining are increasing their presence in offshore environments. These have the potential to contribute significantly to increases in overall ocean ambient noise, particularly in the lower frequencies, 20-1000 Hz (Hildebrand, 2009). A shipping traffic analysis of 2 decades of altimeter data estimated a global fourfold increase in shipping density between 1992 and 2012 (Tournadre, 2014). Not only does increased global shipping directly impact the atmosphere (Tournadre, 2014), it also increases the ambient noise levels (Hildebrand, 2009). In the last several decades, shipping alone is estimated to have increased background noise levels by 12dB (Hildebrand, 2009). There is growing concern about increased ambient noise levels and the impact noise has on marine organisms and ecosystems (Weilgart, 2007; Shannon et al., 2015; Williams et al., 2015). This concern has resulted in the designation of marine anthropogenic noise, specifically shipping noise, as a pollutant with the potential to affect marine ecosystems globally (EU 2008; Clark et al., 2009, Williams et al., 2014b; Merchant et al., 2015). As a result, management efforts have employed regulatory mechanisms that address noise as it affects individuals in a population or, more holistically, as a pollutant impacting the physical characteristics of the marine environment (Dolman and Jasny, 2015).

Understanding the potential impacts of anthropogenic noise requires the establishment of ambient acoustic baselines from which to measure change. Establishing baselines, especially in quiet areas still largely unaffected by anthropogenic stressors, is particularly crucial in the face of the expansion of offshore industries, increasing coastal population and growing reliance on the ocean for global transportation. Many of the long-term ambient acoustic datasets are recorded in deep-water environments (Andrew et al., 2002; McDonald et al., 2006; Chapman

5 and Price, 2011). Extrapolating these deep ocean trends in ambient noise to nearshore and coastal soundscapes is difficult due to the complexity of sound propagation in shallow water environments (Ferris, 1972). Therefore, coastal ambient noise datasets similar to the one presented here are crucial for understanding how changing ambient noise levels may impact coastal ecosystems, especially with increasing anthropogenic stressors such as shipping and extractive industries. Additionally, baseline acoustic datasets are imperative for understanding how local, coastal sound levels fit within a range of sound levels in a variety of coastal habitats (Haxel et al., 2013).

SHIPPING AS A SOURCE MECHANISM Merchant shipping vessels are the primary source of acoustic energy between the frequency band of 5 Hz - 1 kHz (Urick, 1983), and in some cases this source is responsible for up to 40 dB increases in ambient noise above local wind-driven noise conditions (Wales and Heitmeyer, 2002). There are multiple components of a vessel that produce noise and each behaves slightly differently. Propeller cavitation results from rapid changes in pressure in the water surrounding the rotating propeller. These changes result in the formation of vapor cavities that collapse when subject to higher pressure and produce a shock wave. This shock wave is what forms the broadband portion of the sounds produced by vessels (Wales and Heitmeyer, 2002). These are visible on the spectrogram as narrow band lines and are also referred to as blade lines (Wales and Heitmeyer, 2002). Blade lines are a result of a changing volume of water in the regions where the cavitation voids occur (Wales and Heitmeyer, 2002). Other components of the noise input include noises associated with rotational parts of the engine machinery. Different vessels produce different acoustic signatures based on their size, speed and vessel load (Hildebrand, 2009). Diesel engines, generators, pumps, fans or other rotating auxiliary equipment on board usually produce tonal peaks in the acoustic signature (Hildebrand, 2009). In coastal areas, fishing vessels and small craft also contribute to vessel noise and the ambient acoustic environment. Small watercraft may contribute to ambient noise in the mid-frequency range between 1 and 5 kHz and have source levels of 150 - 180 dB re 1 µPa at 1m (Erbe, 2002; Kipple and Gabriele 2003, 2004; Hildebrand, 2009) and cavitation noise from propellers can extend up to 10 kHz (Ross, 1976). Small craft may produce sounds up to and beyond 25 kHz (Kipple and Gabriele, 2003, 2004).

6 IMPACTS OF NOISE ON MARINE MAMMALS Sound is one of the most important sensory modalities for ocean animals, especially marine mammals, which have highly adapted hearing mechanisms for underwater listening and communication. Sound travels up to 5 times faster in the water than in air and it is often the most reliable sensory mechanism in a fluid, turbid and light restricted aquatic environment. The broadband nature of vessel traffic has implications for marine mammals and acoustically sensitive marine fauna, many of which vocalize and rely on hearing within these frequency bands. Marine mammals, as well as other aquatic animals, rely on sound for locating food, predator avoidance, short and long-range communication (Tyack, 2008). The best available science (reviewed in literature such as Shannon et al., 2015) confirms the fact that introduced anthropogenic sounds above ambient noise level, like broadband frequencies sounds from shipping and vessel traffic, can disrupt the normal behavior of marine mammals and can have auditory, behavioral and physiological impacts. These may include, but are not limited to increased physiological stress (Wright et al., 2007; Rolland et al., 2012), displacement (Castellote et al., 2012; Morton and Symonds, 2002), changes in foraging patterns and feeding behavior due to direct (Soto et al., 2006 or indirect (Popper et al., 2003) mechanisms, changes in vocalization or communication behavior (Blackwell et al., 2015; Castellote et al., 2012; Di Iorio and Clark, 2010; Parks et al., 2011) masking (Clark et al., 2009), loss of communication space (Hatch et al., 2012 and Williams et al., 2014a) and temporary or permanent threshold shifts (TTS and PTS) in the ability of the animals to hear (Lucke et al., 2009; Kastelein et al., 2013). In some cases shipping noise caused no immediate observable change in behavioral state but larger temporal and spatial scales might elucidate more chronic and cumulative impacts of anthropogenic noise (Croll et al., 2001). This highlights the important point that impact may exist even where observable and immediate changes in behavior do not occur. For a more detailed review of these impacts please see Nowacek et al., (2007); Weilgart (2007); Shannon et al., (2015). The state of the best available science points to the fact that these individual responses or impacts vary based on exposure levels, behavioral state of the individual, spatial and temporal characteristics of the sound and the ability of the individual to hear various frequencies. It is important to note that noise can and does have an impact on individual fitness (defined as individual reproductive success) and the structure of these ecological communities. This is well documented for a number of wildlife species (Shannon et al., 2015). However, the extent to

7 which these behavioral, auditory or physiological responses to noise changes the fitness of an individual marine mammal, has yet to be ascertained in a quantifiable manner. King et al. (2015) shows that for most wildlife, including marine and aquatic mammals, there is little to no empirical evidence to quantify the relationship between behavioral or physiological change and fitness.

STUDY SITE Global demand for liquid natural gas (LNG), catalyzed primarily by a growing Asian market, is expected to increase significantly in the next 20 years (Ministry of Energy and Mines, British Columbia). The geographic position of British Columbia relative to these markets, a growing supply of LNG and new technology for extraction and shipping situate British Columbia as a strong competitor in the lucrative market. However, the LNG proposals, which include pipelines, port terminals and shipping channels, physically traverse territories of at least 32 First Nations in British Columbia (Ministry of Aboriginal Relations and Reconciliation, 2015). The LNG industry could have many adverse impacts on these territories and ecosystems. British Columbia created a job initiative in 2011 in order to bolster local economies by creating new markets. As a part of this initiative, the British Columbia government set a goal to open three liquid natural gas facilities for processing and export (LNG Canada, 2015). Three of the existing twenty proposals are in the upper areas of the Kitimat Fjord System and they would require ships to transit through the important fjord ecosystem. The proposed tanker lane is routed through Gitga’at First Nation territory as well as through important habitat for fin whales, humpback whales and orcas (Figure 1). The most current export projects are in various stages of development and include Cedar LNG, Kitimat LNG, and LNG Canada. New LNG facilities mean increased vessel traffic in the Kitimat Fjord System and there is growing concern about the potential acoustic impacts of increased vessel traffic.

Figure 1. Map of the west coast of British Columbia showing the density of vessel traffic (in hours) within 5km by 5km grid cells for the year 2010. The map shows cumulative hours for fishing, government, merchant, passenger & cruise, pleasure & yacht, research, tanker and tug & service vessels. Darker blue colors represent higher traffic density. Vessel traffic data were accessed online from The BC Marine Conservation Analysis, Oil in Canadian Waters Research Group: Ron Pelot (MARIN, Dalhousie University) is the data custodian. Data were accessed from http://bcmca.ca.

9 PREVIOUS ACOUSTIC RESEARCH Previous acoustic research in the Kitimat Fjord system is limited. Cetacealab, a nonprofit organization dedicated to cetacean and acoustic research, has had a network of hydrophones deployed and recording since 2001, but no other long term and ongoing study exists. Williams et al., (2014a) conducted short-term acoustic research in nearby sites Caamano Sound, Kitkiata Inlet and Kitimat (Figure 2). They made underwater recordings June-October 2008 at Caamano Sound and August 2010 - September 2010 at Kitkiata Inlet and Kitimat. The purpose of the study was to determine the quality of the acoustic habitat for fin whales, humpback whales and orcas across 12 sites throughout the northern and southern British Columbia coastline. Their main results are summarized in Figure 2. The study also cites the Kitimat Fjord system as an area that experiences little large ship and tanker traffic relative to areas along the southern coast of British Columbia (Williams et al., 2014a).

Figure 2. Summary of the acoustic research conducted in the Kitimat Fjord system. Shade of the circles indicates the median ambient deployment noise level (in dB re: 1µPa) at each of the 12 sites. Relative size of the circle indicates the percentage of each species’ available communication space under the local median deployment ambient noise conditions relative to the communication space available during the quietest time periods. Source: Williams et al., 2014a.

10 SPECIES AT RISK ACT (SARA) AND CRITICAL HABITAT DESIGNATION The Douglas Channel area surrounding Gil Island is important habitat for four species protected under Canada’s Species at Risk Act (S.C. 2002, c. 29), transient killer whales, northern resident killer whales, humpback and fin whales. Within SARA, a critical habitat is defined as the “habitat that is necessary for the survival or recovery of a listed wildlife species and that is identified as the species’ critical habitat in the recovery strategy or in an action plan for the species” (S.C. 2002, c. 29). Marine mammals listed as threatened include the fin whale (Balaenoptera physalus) and humpback whale (Megaptera novaeangliae) and the Northeast Pacific transient killer whales and northern resident killer whales (Subsections 2(1), 42(2) and 68(2)). The Species at Risk Act (SARA, Section 37) requires the minister responsible for the species in question to create recovery strategies for listed endangered and threatened species. In January 2005 the North Pacific humpback whale was listed as threatened under SARA (Fisheries and Oceans Canada, 2013). Under the threat classification, acoustic disturbance relative risk to individual humpbacks and populations is unknown though acoustic disturbance risk is listed as low to moderate for their habitat. Specifically, the recovery strategy outlines that increased acoustic disturbance in humpback whale habitat, especially in fjord and channel systems, may cause displacement (Fisheries and Oceans Canada, 2013). Additionally, the bathymetric features of these deep fjord and channel systems may reduce sound absorption and dissipation making these habitats particularly vulnerable to acoustic pollution. Figure 3 outlines the four existing critical habitat areas, one of which overlaps significantly with the current study site near Gil Island (Fisheries and Oceans Canada, 2013). These areas have been identified as critical habitat for feeding and foraging as well as resting and socializing (Fisheries and Oceans Canada, 2013). The most current recovery strategy (Section 2.7.4.1) specifically highlights the importance of the acoustic environment and unimpeded space for detection of prey and feeding. Cooperative foraging strategies such as bubble net feeding are dependent on the ability of the group to communicate with conspecifics and intensive vessel traffic or increased vessel density are mentioned as two activities likely to result in the destruction of the designated Critical Habitat (Fisheries and Oceans Canada, 2013). It is the legal obligation, under SARA, to protect critical habitat and the physical components of it that are identified as important to the species in question and it is prohibited to cause destruction of these areas (SARA, 58(1)).

Figure 3. Locations of designated humpback whale critical habitat areas. Location (d) surrounds Gil Island and the site of hydrophone deployment for the present study. Source: DFO (2009).

Though the northern resident killer whale range extends up the northern coast of British Columbia, critical habitat has only been designated for Johnstone Strait and Queen Charlotte Strait when they frequent these areas from June to October (DFO, 2011). Their range during other times of the year is not well understood and efforts are underway to identify additional areas of critical habitat as well as the identification of sources of acoustic disturbance that may result in the destruction of critical habitat (DFO, 2011). In a report on the assessment of critical habitats for resident killer whales, Ford (2006) proposed that the area around Caamano Sound including Squally Channel and the area off of Gil Island were important for northern residents. In addition to the northern resident killer whales, efforts are in place to support the designation of critical habitat for transient killer whales. Due to their highly mobile nature and large-scale movements throughout the coastal range, their proposed critical habitat includes all Pacific coastal marine areas within 3nm of the most proximate shoreline (Figure 4, (DFO, 2013).

Figure 4. Proposed critical habitat for transient killer whales. Includes all areas within 3nm of coastline. Source: DFO (2014).

Currently, there is no designated critical habitat for fin whales and the number of fin whales occupying B.C. waters is unknown. Williams and Thomas (2007) estimate 496 (95% CI: 201- 1222, CV = 0.46) fin whales in Hecate Strait, Dixon Entrance and Queen Charlotte Sound. The document supporting the designation of critical habitat for large whales in B.C. also lists vessel traffic and changes to the acoustic environment as potential threats to the populations and their critical habitat (Nichol & Ford, 2012).

In summary, though the study site surrounding Gil Island is currently only designated critical habitat for humpback whales, it has been proposed previously as important habitat for transient and northern resident killer whales and fin whales. The literature and supporting documents from Fisheries and Oceans Canada all highlight the potential threats of vessel traffic and alteration of the ambient acoustic environments. However, scientific guidance is necessary to understand the level of acoustic disturbance that would constitute the destruction of a designated critical habitat (Federal Court, 2010). Thus, the purpose of this study is to 1) quantify the existing sound levels in the area surrounding Gil Island and 2) identify potential source mechanisms in order to provide a baseline study of the acoustic environment in the Kitimat Fjord system.

13 METHODS

ACOUSTIC RECORDING EQUIPMENT The Cetacealab research station is located on the southern tip of Gil Island in one of the most remote places along the northern coast of British Columbia. With permission from leaders of the Gitga’at First Nation, it became their mission to establish a network of hydrophones with the objective of collecting long-term data sets on the presence and abundance of whales in the Kitimat Fjord System. These acoustic datasets are also critical for understanding the ambient sound environment of the natural fjord system. The data presented in this report were collected using Ocean Sonics icListen High Frequency hydrophones at the locations shown in Figure 5. Data were sampled continuously with a sampling rate of 64 kHz with the exception of storm events when the power connection and transmitter were disrupted, especially during the winter months (Figure 6a). Though the data are seasonally biased, each hour of the data has relatively equal representation (Figure 6b). The hydrophone sensitivity curve was not flat over the desired frequency bandwidths (Figure 7a).

14 Figure 5. Map showing the study site and locations of the hydrophone deployments (yellow triangles). This study analyzed data collected from the Gil Island station. Proposed LNG shipping route is shown in orange.

The data were divided into three bands, low (20-100 Hz), mid (100 - 500 Hz), and high (500 Hz - 20 kHz) and a broadband frequency band (20 Hz - 25 kHz) with sensitivities of -170.5, -170, and -168.8 and -168.6 dB re 1 µPa respectively. All noise statistics reported have accuracy within 3dB (Figure 7c). The bandwidths were chosen based on anthropogenic influences from shipping noise as well as the vocalization ranges of fin, humpback and killer whales. Fin whale sound production is varied but occurs mostly in low frequencies (<100 Hz) (Watkins, 1981; Watkins et al., 1987; Sirovic et al., 2007). Humpback whales produce sound in the form of moans, grunts and pulse trains between 20-1900 Hz (Thompson et al., 1986). The vocalizations of Orcinus orca in the pacific west can be classified into clicks, whistles and pulsed calls (Ford, 1989). Whistles occur at frequencies of 1.5-18 kHz and pulse trains usually occur from 1-6 kHz

(Ford, 1989).

Figure 6. a) Total hours of acoustic recordings from Gil Island (1278) each month November 2014 - October 2015. The total hours of recordings possible within each month is represented by the dotted black line. b) Total hours of acoustic recordings from Gil Island (1278) in each hour of the day November 2014 - October 2015. The total hours of recording possible within each hour of the day over the course of the deployment is represented by the dotted black line.

Figure 7. a) icListen 1278 sensitivity curve showing the 25 kHz cut off for analysis b) icListen 1278 sensitivity curve showing the 20 Hz cut off for analysis. c) Boxplot of the range of hydrophone sensitivities within each bandwidth. Boxplots show median in red and the 25th and 75 quartiles (blue). Ranges reported are the interquartile ranges and represent the errors incurred by calibrating within each frequency band.

FIELD PROCEDURES To deploy the hydrophones a 40 liter bin was filled with concrete and a 2-inch poly pipe, 3-feet long was centered in the concrete. The icListen HF hydrophone was placed inside the pipe, held in place by short bungee cords, which prevent the hydrophone from touching the poly pipe. A protective sock was placed over the hydrophone before deployment to prevent particles hitting the transducer underwater. Hydrophones were deployed in nearshore areas at a depth of 70 feet. A hydrophone cable was securely fastened to hydrophone, and then fed through a 1-inch poly pipe for protection through the intertidal zone. The cable was connected to a land-based transmitter where solar panels and transmitting equipment (EHF5 UBIQUITI NANOBEAM M SERIES 5GHZ 25DBI) were placed in a tree with line of sight to a relay station. Here the signal was transmitted to Cetacealab and digitally recorded using Lucy, a software program built by Ocean Sonics and designed specifically for the icListen hydrophones.

17 DATA ANALYSIS

Acoustic data were transmitted from the relay station to the base research station on Gil Island, British Columbia. The data were digitized using Ocean Sonics Lucy software and analyzed using custom MatLab© scripts. The raw acoustic wav files were rewritten into 30-second file segments for subsequent analysis. Files were considered continuous if there were less than three seconds between consecutive file timestamps. Each 30-second file was filtered using a Butterworth 3rd order bandpass filter at the following frequency bandwidths: low (20 - 100 Hz), mid (100 - 500 Hz), high (500 Hz - 20 kHz), and broad (20 Hz - 25 kHz). As previously stated, these bandwidths were selected for their biological and anthropogenic relevance in addition to maintaining calibration-based errors below 3dB. Each bandpass filtered file was then calibrated by multiplying the waveform by the respective calibration factor.

Sound levels are reported in equivalent continuous sound level, Leq, which is an effective acoustic parameter for describing fluctuating sounds over time (Equation 1). The continuous equivalent sound level can be described as the total sound energy integrated over a given time duration. Thus, Leq is a measurement of a constant sound over a specified duration, which has the equivalent energy of the unsteady sound (Barber et al., 2011). This makes the Leq less susceptible to high intensity impulsive sounds than a root mean square (RMS) sound measurement.

2 2 Equation 1. Leq = 10 log10 (1/T p (t) / p 0 dt)

Leq was calculated directly from the filtered and calibrated waveform for each 30-second segment. Alternative segment lengths tested included 10, 60 and 300 seconds with the same processing as described above. These resulted in negligible differences between Leq levels calculated from the different segment lengths (< 1 dB re 1uPa, Figure 9), thus 30-second

18 segments were used.

Figure 9. The Leq percentile plots of the low frequency bandwidth (20-100 Hz) for segments lengths of 10, 30, 60 and 300 seconds.

The continuous equivalent sound levels for each 30 second segment were binned into hour long time periods and the 1st, 10th-90th (in tenths), and 99th percentile distributions were plotted for each bandwidth. The data could be visualized as a given percentile fluctuating over time with hourly resolution for each bandwidth and during different seasons of the year. This methodology allowed the examination of both cyclical and long-term changes to the acoustic environment (Seger et al., 2015). The exploration of different percentiles also allows for the determination of potential source sound mechanisms.

Continuous equivalent sound levels were used in determining diel cycles. The Leq measurements were binned into hour long time periods across all days and the percentile distributions were plotted as a function of the hour-of-day. The plots were then inspected for the presence of a diel cycle.

To identify potential acoustic source mechanisms, the 10 files with the highest levels within each hour were inspected manually for sound sources. Additionally, weather measurements from nearby weather buoys were used to aid in identifying potential source mechanisms associated

19 with weather (Figure 10). Pearson’s correlation coefficients were calculated for median hourly wind speed time series collected at the South Hecate, North Hecate, and Nanakwa shoal weather buoys and median hourly Leq values calculated using the acoustic data from the Gil Island site.

Figure 10. Map displays the hydrophone locations, weather buoys used to get wind speeds and the proposed shipping lanes.

20 RESULTS Data recorded from the hydrophone stationed at Gil Island are shown here. Data are continually recorded but this analysis is restricted to data collected from November 2014 to November 2015. It should be noted that the available data during the winter season are limited and all data are biased against extreme weather conditions when high wind speeds cause a disruption in the land-based transmitter and relay station. Figures show example spectrograms of probable sound source mechanisms for the ambient acoustic environment of the Kitimat Fjord system.

Figure 11. Spectrograms of marine mammals identified as high energy sound source mechanisms. A) humpback whale (nfft = 8192, 90% overlap, hann window), b) unknown, possibly pinniped (nfft = 8192, 90% overlap, hann window), c) killer whale whistles, clicks and biphonic calls (nfft = 1024, 90% overlap, hann window)

Figure 12. Spectrograms of a) vessel noise (nfft = 2048), b) fish (nfft = 10240), c) storm surge (nfft = 8192). All spectrograms calculated with 90% overlap and hann window.

The main goal of this analysis was to establish a baseline from which to measure potential future change in the soundscape due to increased vessel noise input. The selected percentiles of the Leq levels within each bandwidth are summarized in table 1. Overall, the acoustic environment is quiet relative to areas along the southern coast of British Columbia that are subject to more vessel traffic such as Georgia and Haro Strait (Williams et al., 2014a). Williams et al., (2014) report median ambient noise levels at these two locations of 90 - 96 dB re 1 µPa and the noise is mostly attributed to high-density vessel traffic. The median ambient noise levels reported here for the low, mid, high and broad bandwidths in the Douglas Channel off of Gil Island are 78, 83, 87 and 89 db re 1 µPa respectively, and are consistent with levels recorded at similar sites within the Kitimat Fjord system, Caamano, Kitimat, and Kitkiata inlet (Williams et al., 2014a). The purpose of the current study sought to quantify sound energy in bands relevant to marine mammals inhabiting the Kitimat Fjord system, namely fin whales, humpback whales and orcas. The bands vary in width, and the levels across the different

22 bandwidths are not directly comparable. They are comparable to the data collected at the other hydrophone locations and data collected under potential future increased vessel density conditions.

Table 1. Summary results of selected percentiles from the distribution of Leq values binned hourly over the duration of the deployment period. Frequency bandwidths are shown and Leq levels are reported in dB re 1 µPa.

Bandwidth Mean Max Min 99th 90th 50th 10th 1st

20 - 100 Hz 80 134 75 100 89 78 77 76

100- 500 Hz 85 133 75 106 97 83 78 77

500 Hz - 20 kHz 89 136 78 107 99 87 82 81

20 Hz - 25 kHz 91 136 82 109 102 89 85 83

Analysis of the hourly Leq percentiles revealed no obvious diel trend in the data (Figure 13), though it is possible that a small diel trend signal may be obscured by high amplitude noise events generated by ships transiting the channel. Across all frequency bandwidths the 1st, 10th and 50th percentiles are all compressed within 6 dB re 1 µPa. The lower three percentiles within the low bandwidth is compressed further within 2 dB re 1µPa. This compression of the lower 50 percentile indicates that half of the time noise falls within 6 dB of the quietest ambient noise (Fig & Table 1). The levels reported for the 90th and 99th percentiles across the bandwidths represent extreme noise events, or the levels exceeded only 10 and 1 percent of the time respectively. These show a slight trend with a slight peak between the hour of 0400 and 2100 in the low and broadband frequencies. In the mid, high and broad frequency bandwidth the 50th percentile also increases slightly after 1200. The most extreme 99.9 percentiles have large peaks at 1500 across all bandwidths. There are also several spikes during daylight hours indicating that these peaks may be indicative of ship or boat transit.

23 Figure 13. Percentiles of the Leq levels from the duration of the deployment binned into each hour of the day for each frequency bandwidth low, mid, high and broad bandwidths.

In order to determine the source mechanisms responsible for some of the extreme event peaks in the Leq percentile plots the 10 loudest files were selected for manual inspection within each hour of the day. These files represent the 10 loudest 30 seconds within each hour of the day over the course of the entire deployment (Figure 14). It should be noted that the 10 loudest files are not representative of all of the files above the 99th percentile but they do give a good indication of the existing source mechanisms in the ambient acoustic environment of the Kitimat Fjord System and specifically in the channel by Gil Island where the LNG shipping lane is proposed. In the low and mid frequencies the high amplitude event noises within the acoustic environment are dominated by storm and surge noise as well as noises associated with vessel traffic. There is a strong vessel noise signal in the low and mid frequencies between the hours of 0800 and 1800. Specifically, 1400 and 1500 are dominated by vessel noise in all of the frequency bandwidths corroborating the strong peaks in the 99.9 percentiles at 1500 (Figure 13). The high and broad bandwidths are more dominated by vessel noise indicating potentially fewer higher frequency high amplitude event noises in the geophony and biophony of Kitimat Fjord system. Sounds generated by rain also contribute to the high-energy aspect of the ambient acoustic environment. In addition to the dominant sources, naturally occurring seismic

24 activity and several sounds associated with marine mammals registered as some of the loudest sounds recorded during the deployment. These included humpback whale, killer whale and potential pinniped vocalizations. Sounds identified as unknown were artifacts of the deployment such as particles hitting the hydrophone unit. Other unknown sounds were low frequency fish or mammal like grunts. The quietest files were also visually inspected and consisted mostly of low density snapping shrimp noises. Though fin whale vocalizations are present in the dataset, none registered as the loudest sounds within the ambient acoustic environment. This indicates that they may be too intermittent or distant to contribute to the loudest sounds. In the absence of vessel traffic, the ambient acoustic environment would likely be dominated by abiotic source mechanisms such as rain, surge and storm events as well as biotic contributors such as marine mammals and fish.

Figure 14. Leq values from the duration of the deployment were binned and categorized by the hour of the day in which they occurred and by the corresponding frequency bandwidth. The corresponding ten loudest 30 second files within each bandwidth and for each hour of the day were manually examined to identify the sound source. Results are shown here as the percentage of source mechanisms within each hour of the day. Dark gray = vessel noise, dark green = storm / surge, light blue = rain, dark red = earthquake / seismic activity, purple = humpback, orange = fish, teal = orca and tan = unknown.

25 Wind produces ambient noise in the ocean through a mechanism known as wind-induced wave breaking. In coastal areas such as the Kitimat Fjord system shorter fetches and land topography constrain the degree to which wind may influence the ambient noise conditions. In order to examine the contribution of wind conditions to the ambient noise environment, hourly wind speed averages were obtained from three weather buoy stations, North and South Hecate Strait and Nanakwa shoal (station IDs 46183, 46185 and 46181 respectively) owned by Environment Canada. The buoys are approximately 90 km distance from the hydrophone mooring station. It is likely that the wind conditions inside the channel are slightly different than those existing in Hecate Strait so wind speed was averaged across all three buoys in an attempt to obtain a representative number. Median hourly wind speeds were correlated to median hourly Leq levels within each bandwidth (Figure 15). All Pearson’s correlation coefficients were significant (p < 0.0001) for low, mid, high and broadband Leq levels and the correlation coefficients are reported below (Figure 15). Low and mid frequency bands correlation coefficients suggest ambient noise is not correlated with wind speed. Though the relationship is significant, it is likely due to the large number of data points (n = 5958). The relationship in the low and mid frequencies is questionable because frequencies below 400 - 500 Hz are generally non-wind-dependent, though there are exceptions in shallow water (<46m) (Wenz, 1962). The high and broadband correlation coefficients (0.31 and 0.29 respectively) are suggestive of a weak positive relationship between Leq and wind speed though the spread of noise levels at a wind speed of zero knots suggest alternative sound source mechanisms as drivers of the ambient acoustic environment. The main caveat in this case is that the data are biased against high wind speed conditions when the land based data relay station fails in foul weather (pers. communication, Janie Wray). Collecting more acoustic data during periods of high wind speed and storms would help to elucidate the relationship. Additionally, lag correlating the data might result in a stronger relationship due to the fact that it takes time for wind to agitate surface water conditions.

Figure 15. The scatter plots show wind speed (knots) correlated with Leq values for each bandwidth, low, mid, high and broad. Corresponding correlation coefficients are also shown (r = 0.19, 0.22, 0.31 and 0.29 respectively for each bandwidth, low, mid, high broad).

DISCUSSION & RECOMMENDATIONS This study provides baseline ambient acoustic levels for Kitimat Fjord system surrounding Gil Island and within a proposed shipping lane for future natural gas exports. The waters surrounding Gil Island are thought to be among the quietest soundscapes in British Columbia (Erbe et al., 2012). Increasing industrialization of the area is likely to increase the ambient noise and could have deleterious impacts on organisms that rely on the area for critical components of their life cycle. This area surrounding Gil Island is designated critical habitat for humpback whale populations and proposed for transient killer whales. The results of this analysis suggest that though the current ambient levels are low relative to busy shipping ports along the southern coast of British Columbia, vessel noise still contributes significantly to ambient noise levels and accounts for a majority of the highest intensity sounds over the duration of the deployment. Currently, most of the vessel activity in the channel surrounding Gil Island is related to fisheries activities. An increase in vessel traffic due to LNG export could noticeably impact the acoustic

27 environment, especially considering the characteristics of a rocky substrate fjord that reflect rather than absorb acoustic energy. The results of Williams et al. (2014) indicate that the percentage of available communication space for acoustically sensitive species would decrease under increased vessel traffic conditions.

Other significant contributions to the ambient environment include abiotic weather related events such as rain, surge or storm events. Though wind is a dominant source of noise in many acoustic environments (Ross, 1976; Wenz, 1969), it was not strongly correlated to ambient noise in waters near Gil Island. There are multiple explanations for this; 1) wind speed data was taken from weather buoys ~90 km distance from acoustic mooring and may not be representative of wind conditions at the study site, 2) there may be a time lag between when wind speed increases and when sound levels increase due to the topography or fetch in the study site; 3) wind is simply not the dominant driver of ambient noise in waters surrounding Gil Island. Marine mammals that were identified and likely contribute transient, though high intensity, sounds in the acoustic environment and include the calls of humpbacks, whistles and clicks of killer whales, potential pinniped vocalizations, and fin whales. This suggests that in the absence of vessel traffic, both abiotic and biotic components of the ambient acoustic environment dominate. Diel cycle analysis showed no apparent diel trend such as cyclical trends created by snapping shrimp activity (Seger et al., 2015 or chorusing fish in other acoustic studies (Fish, 1964). Snapping shrimp were present in the acoustic data, though in low density, and it is possible diel trends in their activity was obscured by the high intensity event noises of vessel traffic. Alternative explanations for the pattern that was observed include 1) a lack of chorusing fish species in the area surrounding Gil Island, and 2) the shifting crepuscular time periods over the course of the year may be diluting a diel signal due to the way Leq values were binned for the diel cycle analysis.

Williams and Thomas (2007) conducted a study on abundance and distribution of marine mammals in coastal waters of British Columbia and found no fin whales in the Kitimat Fjord system. Four years later, Pilkington et al. (2011) published a study noting the occurrence of fin whales in areas surrounding Gil Island and Caamano Sound starting in year 2006. This suggests re-colonization of this area by historically harvested populations (Pilkington et al., 2011; Ashe et al., 2013). Previous estimates of modeled ship strike risk (Williams & O’Hara, 2010) used abundance and distribution estimates from the Williams and Thomas (2007) study

28 and so did not account for fin whale occurrence within the Kitimat Fjord system. Recovering fin whale populations may be at an increased risk, not only from the degradation of their acoustic environment, but also from vessel collision. A study by Castellote et al. (2012) suggests that male fin whales modify their song characteristics under increased background noise levels and during loud impulsive seismic survey noise there was evidence that the fin whales were displaced for an extended period of time. Another study, Castellote et al. (2010), showed that fin whales might leave an area for an extended period of time under increased ambient noise conditions. The study also suggests that processes of sensitization and habituation to increased ambient noise could play a role in reproductive success, especially when considering that fin whale song has been attributed to reproduction (Castellote et al., 2010; Watkins et al., 1987). Increased shipping in the Kitimat Fjord system and thus decreased acoustic space could have a direct impact on the reproductive success of the recently recolonizing fin whales of the inland British Columbia fjord systems. Though proposed, the area surrounding Gil Island has not yet been designated as critical habitat for fin whales and so their conservation status has no legal teeth in terms of their habitat.

Humpback whales and killer whales are also susceptible to acoustic habitat degradation. Dunlop et al. (2013) demonstrated that in times of high wind speed and high ambient noise humpback whales switched from vocal communication to surface based signaling. They also found evidence of the Lombard effect, or the increase in call amplitude in noise conditions (Dunlop et al., 2013). Lusseau et al. (2009) demonstrated the killer whales alter their behavior state when vessels are passing. For example, foraging behavior is disrupted and the killer whales switch behavior states to traveling in the presence of vessels (Lusseau et al., 2009). Jensen et al. (2009) showed that small vessels could reduce delphinid communication space in coastal deep-water environments. A recent study also demonstrated that noise radiated from ships not only elevated ambient low frequency (100-1000 Hz) ambient noise by 20-30 dB but also high frequency (10 kHz - 96 kHz) ambient noise by 5-13 dB (Veirs et al., 2015). This high frequency radiated noise extends well into the vocalization ranges of killer whales (Veirs et al., 2015). As noted previously, impacts may not be readily or immediately observable and a lack of an observed behavioral or disturbance response should not be mistaken for a lack of effect. Sub-lethal and cumulative impacts need to be better understood.

29 Moloney and Martin (2014) emphasize the importance of characterizing freshwater and marine soundscapes and highlight the Kitimat Fjord system, Dixon entrance and Queen Charlotte sound system as one of their recommended areas for study. They also emphasize that Canada needs a more complete regulatory framework for dealing with issues surrounding anthropogenic noise in the ocean. This study offers to fill at least a portion of these data gaps and more can be done. Well-defined metrics for soundscape measurement and analysis would streamline the process and facilitate data comparisons, especially across migratory species ranges. If the proposed LNG export facilities proceed, speed restrictions should be placed throughout the channel. Leaper et al. (2014) attributed an approximate 2dB reduction in ambient noise in the eastern Mediterranean to a reduction in vessel speeds. Reduced speeds are more fuel and have the added benefit of reducing the probability of fatal ship strikes, as shown for North Atlantic right whales in areas on the east coast important for feeding and reproduction (National Marine Fisheries Service, 2008; Laist et al., 2014). Implemented in 2008, the ship speed rule required all vessels to reduce speeds to 10 knots or slower seasonally in areas important for feeding and reproduction (Laist et al., 2014). It is recommended that acoustic monitoring continue in this area and efforts are made to complete seasonal gaps in the data and determine diel or seasonal usage by acoustically sensitive species including marine mammals and fish. Determining the daily and seasonal habitat use could help with the designation of time/ area closures with limited vessel allowances. Vessel speed restrictions are also recommended as these are proven to reduce acoustic input from vessels, increase fuel efficiency and reduce the likelihood of ship strikes. Propagation modeling is needed to better understand how increased vessel noise might behave in the fjord system and if LNG proposals move forward, efforts should be made to understand the noise conditions and animal behavior before, during and after port expansion. The Kitimat Fjord system is certainly among the quietest along the coast of British Columbia and maintaining the functionality of the acoustic environment for the species that rely on it is key to the coexistence of the ecosystem, communities and LNG export industry.

ACKNOWLEDGEMENTS Thank you to Janie Wray and Hermann Meuter of Cetacea Lab, the Gitga'at First Nation and the World Wildlife Fund for providing the acoustic dataset and guidance. Also thank you to my advisor, Dr. Douglas Nowacek and Will Cioffi for guidance and technical support.

REFERENCES Aguilar Soto, N., Johnson, M., Madsen, P. T., Tyack, P. L., Bocconcelli, A., & Fabrizio Borsani, J. (2006). Does intense ship noise disrupt foraging in deep-diving Cuvier’s beaked whales (Ziphius cavirostris)?. Marine Mammal Science, 22(3), 690-699.

Andrew, R. K., Howe, B. M., Mercer, J. A., & Dzieciuch, M. A. (2002). Ocean ambient sound: comparing the 1960s with the 1990s for a receiver off the California coast. Acoustics Research Letters Online, 3(2), 65-70.

Ashe, E., Wray, J., Picard, C. R., & Williams, R. (2013). Abundance and survival of Pacific humpback whales in a proposed critical habitat area. PloS one, 8(9), e75228.

Barber, J. R., Burdett, C. L., Reed, S. E., Warner, K. A., Formichella, C., Crooks, K. R., ... & Fristrup, K. M. (2011). Anthropogenic noise exposure in protected natural areas: estimating the scale of ecological consequences.Landscape ecology, 26(9), 1281-1295.

Blackwell, S. B., Nations, C. S., McDonald, T. L., Thode, A. M., Mathias, D., Kim, K. H., ... & Macrander, A. M. (2015). Effects of airgun sounds on bowhead whale calling rates: evidence for two behavioral thresholds. PloS one, 10(6), e0125720.

Castellote, M., Clark, C. W., & Lammers, M. O. (2010). Potential negative effects in the reproduction and survival on fin whales (Balaenoptera physalus) by shipping and airgun noise. International Whaling Commission report SC/62/E3–2010.

Castellote, M., Clark, C. W., & Lammers, M. O. (2012). Acoustic and behavioural changes by fin whales (Balaenoptera physalus) in response to shipping and airgun noise. Biological Conservation, 147(1), 115-122.

Chapman, N. R., & Price, A. (2011). Low frequency deep ocean ambient noise trend in the Northeast Pacific Ocean. The Journal of the Acoustical Society of America, 129(5), EL161-EL165.

Clark, C. W., Ellison, W. T., Southall, B. L., Hatch, L., Van Parijs, S. M., Frankel, A., & Ponirakis, D. (2009). Acoustic masking in marine ecosystems: intuitions, analysis, and implication. Marine Ecology Progress Series, 395, 201-222.

Croll, D. A., Clark, C. W., Calambokidis, J., Ellison, W. T., & Tershy, B. R. (2001). Effect of anthropogenic low-frequency noise on the foraging ecology of Balaenoptera whales. Animal Conservation, 4(01), 13-27.

DFO(Fisheries and Oceans Canada). 2013. Recovery Strategy for the North Pacific Humpback Whale (Megaptera novaeangliae) in Canada. Species at Risk Act Recovery Strategy Series. Fisheries and Oceans Canada, Ottawa. x + 67 pp.

31 DFO (Fisheries and Oceans Canada). 2009. Advice relevant to the identification of critical habitats for North Pacific Humpback Whales (Megaptera novaeangliae). DFO Can. Sci. Advis. Sec. Sci. Resp. 2009/016.

DFO (Fisheries and Oceans Canada). 2011. Recovery Strategy for the Northern and Southern Resident Killer Whales (Orcinus orca) in Canada. Species at Risk Act Recovery Strategy Series, Fisheries & Oceans Canada, Ottawa, ix + 80 pp. DFO (Fisheries and Oceans Canada). 2013. Information in Support of the Identification of Critical Habitat for Transient Killer Whales (Orcinus orca) off the West Coast of Canada. DFO Can. Sci. Advis. Sec. Sci. Advis. Rep. 2013/025.

Di Iorio, L., & Clark, C. W. (2010). Exposure to seismic survey alters blue whale acoustic communication. Biology letters, 6(1), 51-54.

Dolman, S. J., & Jasny, M. (2015). Evolution of Marine Noise Pollution Management. Aquatic Mammals, 41(4), 357.

Dunlop, R., Noad, M., & Cato, D. H. (2013, June). Modification of humpback whale social sound repertoire and vocal source levels with increased noise. In Proceedings of Meetings on Acoustics (Vol. 19, No. 1, p. 010058). Acoustical Society of America.

Erbe, C. (2002). Underwater noise of whale‐watching boats and potential effects on killer whales (Orcinus orca), based on an acoustic impact model.Marine mammal science, 18(2), 394-418.

Erbe, C. (2012). Effects of underwater noise on marine mammals. In The effects of noise on aquatic life (pp. 17-22). Springer New York.

EU (European Union).2008. Marine strategy framework directive. Directive 2008/56/EC, 17 Jun 2008, Official Journal of the European Union, L 164/19, 25 Jun 2008, part 3(8).Brussels, Belgium: EU.

Federal Court. (2010). Northern and southern resident killer whales (Orcinus orca) in Canada: critical habitat protection statement. In: 2010 FC 1233 127. Ecojustice (Ed.). Ottawa, Ontario, Canada: Federal Court of Appeal.

Ferris, R. H. (1972). Comparison of Measured and Calculated Normal‐Mode Amplitude Functions for Acoustic Waves in Shallow Water. The Journal of the Acoustical Society of America, 52(3B), 981- 989.

Fish, M. P. (1964). Biological sources of sustained ambient sea noise. Narragansett Marine Laboratory, University of Rhode Island.

Ford, J. K. (1989). Acoustic behaviour of resident killer whales (Orcinus orca) off Vancouver Island, British Columbia. Canadian Journal of Zoology, 67(3), 727-745.

Ford, J.K.B. 2006. An Assessment of critical habitats of resident killer whales in waters off the Pacific Coast of Canada. Canadian Science Advisory Secretariat, Research Document 2006/072.

32 Halpern, B. S., Walbridge, S., Selkoe, K. A., Kappel, C. V., Micheli, F., D'Agrosa, C., ... & Fujita, R. (2008). A global map of human impact on marine ecosystems. Science, 319(5865), 948-952.

Hatch, L. T., Clark, C. W., Van Parijs, S. M., Frankel, A. S., & Ponirakis, D. W. (2012). Quantifying loss of acoustic communication space for right whales in and around a US National Marine Sanctuary. Conservation Biology, 26(6), 983-994.

Haxel, J. H., Dziak, R. P., & Matsumoto, H. (2013). Observations of shallow water marine ambient sound: the low frequency underwater soundscape of the central Oregon coast. The Journal of the Acoustical Society of America,133(5), 2586-2596.

Hildebrand, J. A. (2009). Anthropogenic and natural sources of ambient noise in the ocean. Marine Ecology Progress Series, 395(5).

Jensen, F. H., Bejder, L., Wahlberg, M., Aguilar De Soto, N., Johnson, M. P., & Madsen, P. T. (2009). Vessel noise effects on delphinid communication. Marine Ecology Progress Series. 395, 161-175.

Kastelein, R. A., Gransier, R., & Hoek, L. (2013). Comparative temporary threshold shifts in a harbor porpoise and harbor seal, and severe shift in a seal. The Journal of the Acoustical Society of America, 134(1),13-16.

King, S. L., Schick, R. S., Donovan, C., Booth, C. G., Burgman, M., Thomas, L., & Harwood, J. (2015). An interim framework for assessing the population consequences of disturbance. Methods in Ecology and Evolution,6(10), 1150-1158.

Kipple, B., & Gabriele, C. (2003). Glacier Bay watercraft noise. Naval Surface Warfare Center Technical Report NSWCCD-71-TR-2003/522.

Kipple, B., & Gabriele, C. (2004). Glacier Bay watercraft noise–Noise characterization for tour, charter, private, and government vessels. Naval Surface Warfare Center, Bremerton, WA, Technical Report NSWCCDE-71-TR-2004/545, prepared for Glacier Bay National Park and Preserve, 1-47.

Laist, D. W., Knowlton, A. R., & Pendleton, D. (2014). Effectiveness of mandatory vessel speed limits for protecting North Atlantic right whales.Endangered Species Research, 23, 133-147.

Leaper, R., Renilson, M., & Ryan, C. (2014). Reducing underwater noise from large commercial ships: current status and future directions. Journal of Ocean Technology, 9(1).

LNG Canada. (2015). Project Overview. Retrieved from http://lngcanada.ca/wp- content/uploads/2015/09/R06939-LNG-Canada-Project-Overview-Factsheet-AW-v2-Web-1.pdf

Lucke, K., Siebert, U., Lepper, P. A., & Blanchet, M. A. (2009). Temporary shift in masked hearing thresholds in a harbor porpoise (Phocoena phocoena) after exposure to seismic airgun stimuli. The Journal of the Acoustical Society of America, 125(6), 4060-4070.

Lusseau, D., Bain, D. E., Williams, R., & Smith, J. C. (2009). Vessel traffic disrupts the foraging behavior of southern resident killer whales Orcinus orca.Endangered Species Research, 6(3), 211-221.

33 McDonald, M. A., Hildebrand, J. A., Wiggins, S. M., & Ross, D. (2008). A 50Year comparison of ambient ocean noise near San Clemente Island: A bathymetrically complex coastal region off Southern California. The journal of the Acoustical Society of America, 124(4), 1985-1992.

Merchant, N. D., Fristrup, K. M., Johnson, M. P., Tyack, P. L., Witt, M. J., Blondel, P., & Parks, S. E. (2015). Measuring acoustic habitats. Methods in Ecology and Evolution, 6(3), 257-265.

Ministry of Aboriginal Relations and Reconciliation. (2015 June 12) Factsheet: Partnering with First Nations on LNG development. BC Gov News. Retrieved from https://news.gov.bc.ca/factsheets/partnering-with-first-nations-on-lng-development

Ministry of Energy and Mines. Liquefied Natural Gas: A Strategy for B.C.’s Newest Industry. Retrieved from http://www.gov.bc.ca/ener/popt/down/liquefied_natural_gas_strategy.pdf

Moloney & Martin. (2014). Characterizing the Soundscapes of Canada's Oceans, Lakes and Rivers. Journal of Ocean Technology, 9(1), 27-33.

Morton, A. B., & Symonds, H. K. (2002). Displacement of Orcinus orca (L.) by high amplitude sound in British Columbia, Canada. ICES Journal of Marine Science: Journal du Conseil, 59(1), 71-80.

National Marine Fisheries Service (2008a) Final rule to implement speed restrictions to reduce the threat of ship collisions with North Atlantic right whales. Fed Regist 73: 60173−60191. Available at www. nmfs. noaa. gov/ pr/ pdfs/ fr/fr73-60173.pdf

Nichol, L.M. and J.K.B. Ford. 2012. Information relevant to the assessment of critical habitat for Blue, Fin, Sei and North Pacific Right Whales in British Columbia DFO Can. Sci. Advis. Sec. Res. Doc. 2011/137. vi + 31 p.

Nowacek, D. P., Thorne, L. H., Johnston, D. W., & Tyack, P. L. (2007). Responses of cetaceans to anthropogenic noise. Mammal Review, 37(2), 81-115.

Parks, S. E., Johnson, M., Nowacek, D., & Tyack, P. L. (2011). Individual right whales call louder in increased environmental noise. Biology Letters,7(1), 33-35.

Pilkington, J. (2011). Occurrence of Fin Whales (Balaenoptera physalus) in the Confined Channel Assessment Area Between Wright Sound and Caamano Sound from North Coast Cetacean Society Observations for the Period 2004 to 2011. Written Evidence Submission to the Joint Review Panel Assessing the Northern Gateway Pipeline Application. North Coast Cetacean Society.< http://www.acee-ceaa.gc.ca/050/documents/54929/54929E.pdf >

Popper, A. N., Fewtrell, J., Smith, M. E., & McCauley, R. D. (2003). Anthropogenic sound: effects on the behavior and physiology of fishes.Marine Technology Society Journal, 37(4), 35-40.

Rolland, R. M., Parks, S. E., Hunt, K. E., Castellote, M., Corkeron, P. J., Nowacek, D. P., ... & Kraus, S. D. (2012). Evidence that ship noise increases stress in right whales. Proceedings of the Royal Society of London B: Biological Sciences, 279(1737), 2363-2368.

Ross D. (1976). Mechanics of Underwater Noise (Pergamon, New York), p. 375.

34 Seger, K. D., Thode, A. M., Swartz, S. L., & Urbán, J. (2015). The ambient acoustic environment in Laguna San Ignacio, Baja California Sur, Mexico.The Journal of the Acoustical Society of America, 138(5), 3397-3410.

Shannon, G., McKenna, M. F., Angeloni, L. M., Crooks, K. R., Fristrup, K. M., Brown, E., ... & McFarland, S. (2015). A synthesis of two decades of research documenting the effects of noise on wildlife. Biological Reviews.

Širović, A., Hildebrand, J. A., & Wiggins, S. M. (2007). Blue and fin whale call source levels and propagation range in the Southern Ocean. The Journal of the Acoustical Society of America, 122(2), 1208-1215.

Species At Risk Act, § 29 (2002).

Species At Risk Act, § 29-56-60 (2002).

Thompson, P. O., Cummings, W. C., & Ha, S. J. (1986). Sounds, source levels, and associated behavior of humpback whales, Southeast Alaska. The Journal of the Acoustical Society of America, 80(3), 735-740.

Tournadre, J. (2014). Anthropogenic pressure on the open ocean: The growth of ship traffic revealed by altimeter data analysis. Geophysical Research Letters, 41(22), 7924-7932.

Tyack, P. L. (2008). Implications for marine mammals of large-scale changes in the marine acoustic environment. Journal of Mammalogy, 89(3), 549-558.

Urick, R. J. (1983). The noise background of the sea: ambient noise level.Principles of Underwater Sound. Peninsula Publishing, Los Altos, California, 202-236.

Veirs, S., Veirs, V., & Wood, J. D. (2016). Ship noise extends to frequencies used for echolocation by endangered killer whales. PeerJ, 4, e1657.

Wales, S. C., & Heitmeyer, R. M. (2002). An ensemble source spectra model for merchant ship-radiated noise. The Journal of the Acoustical Society of America, 111(3), 1211-1231.

Watkins, W. A. (1981). Activities and underwater sounds of fin whales [Balaenoptera physalus]. Scientific Reports of the Whales Research Institute (Japan).

Watkins, W. A., Tyack, P., Moore, K. E., & Bird, J. E. (1987). The 20‐Hz signals of finback whales (Balaenopteraphysalus). The Journal of the Acoustical Society of America, 82(6), 1901-1912.

Weilgart, L. S. (2007). A brief review of known effects of noise on marine mammals. International Journal of Comparative Psychology, 20(2).

Wenz, G. M. (1962). Acoustic ambient noise in the ocean: spectra and sources. The Journal of the Acoustical Society of America, 34(12), 1936-1956.

35 Wenz, G. M. (1969). Low-frequency deep-water ambient noise along the Pacific Coast of the United States. US Navy J. Underwater Acoust, 19(4), 423-444.

Williams, R., & Thomas, L. (2007). 15 Distribution and abundance of marine mammals in the coastal waters of British Columbia, Canada. J. Cetacean Res. Manage, 9(1), 15-28.

Williams, R., & O&rsquoHara, P. (2010). Modelling ship strike risk to fin, humpback and killer whales in British Columbia, Canada. Journal of Cetacean Research and Management, 11(1), 1-8.

Williams, R., Clark, C. W., Ponirakis, D., & Ashe, E. (2014a). Acoustic quality of critical habitats for three threatened whale populations. Animal Conservation, 17(2), 174-185.

Williams, R., Ashe, E., Blight, L., Jasny, M., & Nowlan, L. (2014b). Marine mammals and ocean noise: Future directions and information needs with respect to science, policy and law in Canada. Marine pollution bulletin, 86(1), 29-38.

Williams, R., Wright, A. J., Ashe, E., Blight, L. K., Bruintjes, R., Canessa, R., ... & Hammond, P. S. (2015). Impacts of anthropogenic noise on marine life: Publication patterns, new discoveries, and future directions in research and management. Ocean & Coastal Management, 115, 17-24.

Wright, A. J., Soto, N. A., Baldwin, A. L., Bateson, M., Beale, C. M., Clark, C., ... & Hatch, L. T. (2007). Anthropogenic noise as a stressor in animals: a multidisciplinary perspective. International Journal of Comparative Psychology, 20(2).