PILE DRIVING UNDERWATER NOISE ASSESSMENT...

PILE DRIVING UNDERWATER NOISE

ASSESSMENT, PROPOSED

BELL BAY PULP MILL

WHARF DEVELOPMENT

Prepared for:

Gunns Limited

Prepared by

R. D. McCauley and C.P. Salgado Kent

Centre for Marine Science and Technology

Curtin University

June 2008

CMST Job 730; CMST Report 2008-27

Curtin University; GPO Box U 1987; Perth 6845; Phone 08 9266 7380,

fax 08 9266 4799;

UUnnddeerrwwaatteerr NNooiissee IImmppaaccttss BBBBPPMMPP RReeppoorrtt 22000088

2

Executive summary Gunns Limited propose to carry out wharf construction approximately 18 km up the Tamar River in northern Tasmania. Around 190-236 piles are expected to be driven, with the final number dependant on the bedrock depth found at each pile location. Piles will be driven initially by vibratory piling where the pile is vibrated by a relatively low impact, high frequency ‘hammer’ to the maximum depth achievable. Impact piling using a hydraulic or diesel driven hammer will then be carried out to drive the pile the remaining distance required. Vibratory pile driving may produce high noise levels but these are known to be some 25 dB below those of impact piling and do not produce the high impulse signatures of impact pile driving which have been associated with serious physiological impacts on fish in other regions. Hence the review of environmental implications of underwater noise produced by wharf construction prepared has focused on impact pile driving only. Where the wharf is to be constructed the river is approximately 600 m wide although deeper parts of the river at 15-20 m depth extend only 540 m across the river from the wharf edge. The seabed across the river is of a layer of almost equal parts fine sand and clay/silt of thickness up to 40 m. The water column across the river for this study was considered to be uniformly mixed with constant 1507 ms-1 vertical sound speed profile. Using these environmental parameters and assuming: a constant water depth of 18 m; pile noise source at the seabed; and a uniform 20 m thick fine-sand/clay-silt layer over bedrock; the underwater sound transmission model Scooter was run at frequencies from 10 Hz to 1 kHz in 1 Hz steps at range a resolution of 0.5 m over 5-500 m in the horizontal plane and 0.5-18 m in the vertical plane. The resulting output of amplitude and phase corrections was applied as a transform function at each spatial point to a representative source pile driving signal to give predicted received pile driving signals. These received signals were then characterized for received sound exposure level, peak-peak level and mean squared pressure. Predicted pile driving signal parameters in the Tamar river followed trends measured from other pile driving sites around Australia. The modeling impact pile driving output showed that: • The estimated sound field across (and along the river) showed considerable variability with

spatial location, vertically and horizontally, with differences of up to 10 dB over a few m of vertical range and tens of m horizontal range, implying that fish in close proximity may receive considerably different sound loadings from a pile impact;

• For a single pile strike most fish will need to be within 10 m from the pile to suffer any serious physiological trauma resulting in injury;

• For a worst case scenario of continual pile hammering over four hours, stationary fish within 120 m of pile driving may receive cumulative sound loading equivalent to that at which serious physiological harm may occur, with this range dropping to 80 m for a more typical 30 minute period of pile hammering;

• For fish swimming steadily past the wharf during pile hammering episodes at slack water, cumulative sound loadings equivalent to that at which a single strike is believed to cause serious physical trauma may occur for fish passing within 50 m of the wharf. The range for the equivalent sound loading will decrease for current assisted fish or fish larger than 250 mm (the maximum size at which cumulative sound loadings were calculated);

• Fish within 60 m of a single pile strike may be exposed to peak-peak intensity sufficient to cause temporary hearing loss;

• Stationary fish within 200 m of continual piling over four hours may be exposed to a cumulative energy loading sufficient to cause temporary hearing loss;

• Fish swimming steadily past the wharf during pile hammering episodes at slack water within 100-300 m closest range to the wharf may receive sufficient sound loadings to cause temporary


3

hearing loss. The degree or presence of any temporary threshold shifts will depend on the fish species, swim speed and proximity to the wharf;

• It is unlikely that dolphins or seals will be exposed to pile driving sound intensities sufficient to cause any serious physiological damage;

• Marine mammals within 50 m of a single pile driving strike may receive a sufficient sound intensity to cause temporary hearing loss;

• Pile driving events will be audible in the river for several km up and downstream to animals with reasonable hearing capability;

• Marine mammals are unlikely to show overt behavioural responses to continual pile driving if they are more than a few hundred m off, and may show some initial curiosity;

• Fish are likely to show strong behavioural responses to pile driving noise all the way across the river and for at least 500 m up or downstream. Such responses could include avoidance or huddling near the seabed;

• Caged fish in a nearby aquaculture farm opposite the proposed wharf are unlikely to be exposed to single or cumulative sound loadings sufficient to cause any physical trauma or hearing damage. But the sound exposures received at the aquaculture farm will be within the bounds at which observable behavioural responses have been detected in caged fish.

Given the ranges for various impacts calculated and defined above, then impacts on marine mammals and seals can only be considered serious if the animals strays within 50 m of the impact pile driving activities. For fish the types of impacts may vary considerably depending on the species, its hearing capability, its habits, proximity to pile driving and if pile driving overlaps some critical part of its life cycle. Only fish in the immediate vicinity, perhaps within 10 m of a pile, may receive sufficient signal intensities to cause serious physiological trauma. Fish moving up or downstream will be exposed to brief episodes of high impact piling noise levels. For the worst case small fish traveling slowly may receive sufficient sound loadings from within 300 m of impact piling to begin to cause temporary shifts in hearing. Fish traveling faster will receive fewer high pile impact signals and thus will need to be closer than 300 m to begin to suffer any physiological impacts on their hearing. To simplify interpretation of impacts on wild fish three zones extending from piling are proposed: Zone 1 - Serious Physiological impacts – out to 10-20 m range from impact piling. This range bracket allows for an uncertainty or safety margin of times-2 to times-4. Within this range fish may suffer serious internal injuries. Zone 2 – Physiological impacts – from 20 m to 300 m from impact piling, with the severity of impact increasing as the range decreases to 20 m. At the outside boundary of this zone (300 m) some fish in certain circumstances (ie. fish stationary for long periods during continual impact piling) may begin to suffer hearing damage or temporary threshold shifts to continual impact piling. At the inside edge of the zone (20 m) most fish would be expected to suffer some form of hearing damage or temporary threshold shifts to continual impact piling. Zone 3 – Behavioural impacts – out to 500 m from impact piling – Within 500 m of impact piling fish will show some behavioural response to impact piling ranging from avoidance or startle response to an increased alertness.


4

Contents Executive summary ................................................................................................................................. 2 1 Introduction .......................................................................................................................................... 4

1.1 Background 4 1.2 Scope 4

2. Noise levels and units .......................................................................................................................... 5 3. Literature review - Impacts of pile driving noise on marine fauna ..................................................... 5

3.1 Invertebrates 5 3.2 Fish 6 3.3 Aquatic birds 13 3.4 Marine Mammals 13 3.5 Summary 16

4 Characterization of pile driving noise ................................................................................................ 17 4.1 Equipment proposed for the construction of the wharf 17 4.2 Sound signal produced by pile driving 18

5 Site characteristics and estimated noise levels ................................................................................... 20 5.1 Site characteristics 20 5.2 Modeling pile driving noise 22 5.3 Estimated underwater noise levels at nearby aquaculture facility 27 5.4 Summary 28

6 Assessment of Impacts ....................................................................................................................... 28 6.1 Critical frequencies 28 6.2 Serious physiological impacts 28 6.3 Temporary hearing impairment 31 6.4 Behavioural responses 32 6.5 Impacts at nearby aquaculture farm 33 6.6 Summary 33

7 Recommended mitigation measures................................................................................................... 34 8 References .......................................................................................................................................... 35

1 Introduction 1.1 Background Gunns Limited proposes to develop the Bell Bay Pulp Mill designed to extract fibre from wood. The location of the project will be at Bell Bay on the Tamar river, 36 km from Launceston and 18 km upstream of the Tamar river mouth. The Bell Bay Pulp Mill project (BBPMP) will include the construction of a wharf (barge unloading facility). Because of potential environmental impacts due to underwater noise associated with pile driving, Gunns Limited has chosen to undertake an assessment of environmental impacts on marine fauna likely to be found in the Tamar River, near the pile driving activities. This document has been prepared for this purpose. 1.2 Scope The construction of the Bell Bay Pulp Mill wharf requires driving of as many as 236 piles, although this is expected to be more like 190 piles, with the actual number dependant on the depth of bedrock found under the proposed wharf. Setting piles will involve either one or both of vibratory piling and impact piling. To assess the impacts from underwater noise produced during pile driving for this project, the following work is presented below; 1) a review of available literature on impacts of pile driving on marine fauna; 2) a description of the noise signatures of the proposed technologies and construction equipment; 3) predictions of transmission of noise based on underwater noise propagation modeling; and 4) a risk assessment for marine animals exposed to pile driving noise. Suggestions are also presented in this document for practical and achievable mitigation measures to


5

be taken during construction by verifying noise predictions, and for mitigating any environmental impacts.

2. Noise levels and units Underwater noise level units are confusing in the literature. Correct units are not always specified in available literature resulting in a difficult comparison of information presented. In underwater acoustics, a variety of units are used to define steady-state and impulsive signals, which include; mean square pressure (dB re 1µPa msp); peak pressure (dB re 1µPa p), peak to peak (dB re 1µPa2 pp), equivalent energy or sound exposure level (dB re 1µPa2.s SEL), and spectral level (dB re 1µPa2/Hz). The mean squared pressure is the decibel value of the mean of the squared pressure over a defined period of a signal. For steady signals the averaging time does not matter, for impulsive (short and sharp) signals the averaging time does matter significantly. Impulsive signals are better described by a measure of the amount of energy which passes through time and some measure of the signal peak amplitude (positive and/or negative). A measure of an impulsive signals energy commonly used is the sound exposure level (SEL) units of dB re 1µPa2.s. The convention for displaying frequency spectra is to reduce the bandwidth of the measurement (which may be across several Hz) to 1 Hz, which gives units of dB re 1µPa2/Hz. These are termed spectral level units and can be readily compared between sources as they offer a standardised frequency bandwidth for the measurement. When presenting noise levels for a source one can present the frequency spectra, in spectral level units (dB re 1µPa2/Hz), or the broadband noise (dB re 1µPa) which is the sum of energy across the frequency band of most energy in the signal. All units are specified in this document.

3. Literature review - Impacts of pile driving noise on marine fauna The use of sound for communication and detection in the marine environment is important for the livelihood and survival of marine animals. Marine animals depend on their hearing sensitivity to retain cohesion in groups, for echolocation (among marine mammals), to locate and capture food, for detection of predators, for sensing their physical and biological environment and for avoiding dangerous situations (including anthropogenic threats). Impacts of underwater noise to marine fauna vary depending upon the characteristics of the noise source (energy, frequency, etc.). Intense noise emitted into the environment has the capability of causing direct physiological damage to animals, while less intense noise can disrupt an animal’s ability to sense its environment and impair it so that the animal is unable to response effectively. Man made underwater noise covers a large range of frequencies, and the way in which a species is impacted by the sound depends on the frequency range it can hear, the level of sound (or energy) and its frequency spectrum. Both the sensitivity of hearing and the frequency range over which sound can be heard varies from species to species due to evolutionary diversification which is often driven by anatomical limitations on the types of signals which can be produced or detected or by the way an animal’s local environment constrains underwater sound transmission. To produce a species specific impact assessment, the hearing sensitivity of each species relevant to the Tamar River would need to be measured. By comprehensively reviewing past studies on hearing sensitivities of a broad range of taxonomic groups, a general indicator can be obtained of the level of sensitivity, hence expected impacts from underwater noise, for animals occurring in the Tamar River. The remainder of this section is therefore dedicated to reviewing past and recent studies on; 1) hearing sensitivity; and 2) known impacts of underwater noise with respect to broad taxonomic grouping (invertebrates, fish, birds, marine mammals). 3.1 Invertebrates


6

Studies on sensitivity to sound Information is scant on invertebrate sensitivity to sound, and the ecological and behavioural functions of sound receptors. For example, squid have demonstrated responses to sound which has been hypothesized to be related to their schooling nature (which requires synchronization and predator aversion mechanisms). Statocysts and/or proprioception (the sensing of movement of bodily tissue by acoustic energy) may be involved in the detection of sound. Information is even more scant on the sensitivity of sound by mollusks (i.e. clams, mussels, oysters, chiton, snails, slugs and limpets). Response to sound has been evident by changes in aggregations. Eradication of zebra muscles, for example, has been accomplished by using ultrasound (Donskoy et al. 1996). A study on the Ox-Heart Clam (Glossus humanus) has demonstrated sensitivity to vibrations and hypothesized that the sensitivity was related to sensing breaking waves on the incoming tide, to move with the tide (Frings 1964). Crustaceans sense sound through their chordotonal organs which are at the joint segments (acting as proprioceptors, or as highly specific mechanoreceptor organs Salmon et al. 1977) or via their statocyst organs. Many Cnidarians (jellyfish, anemones, hydra and corals), crustaceans and mollusks have statocysts which are organs consisting of a calcareous ‘statolith’ surrounded by sensory hair cells. Statocysts are considered to be organs that aid in equilibrium and which have been shown to provide a sense of hearing (Lovell et al 2005). Studies on marine invertebrate responses to noise A study on the impact of a single air gun on squid behaviour demonstrated its sensitivity to sound, in that alarm response (increased swimming behaviour) was observed for a steadily approaching air gun at received levels of 156-161 dB re 1µPa (msp), and strong startle responses at 174 dB re 1µPa (msp) (firing of ink from ink sacs; McCauley et al. 2003a). 3.2 Fish Studies on sensitivity to sound The variability in hearing sensitivity in fish is related to the physiology of the hearing anatomy of different species (see section below; Yan et al. 2000). Fish have been divided into two broad groups based on hearing sensitivity, ‘hearing specialists’ and ‘hearing generalists’. Distinctions between the groups are based on whether the species has specialised organs for improving sound reception. These two groups may serve as a general guideline for hearing sensitivity, but do not replace audiograms which accurately describe the hearing sensitivity of a species. Most fish have not yet been classified as hearing specialists or generalists. The variation among fishes in respect to sensitivity to sound is immense, and is in part due to the diversity of anatomical structures involved in detection (Popper and Fay, 1999). Fish that have morphological adaptations that link the otolithic hearing end organs to their swimbladders or a gas filled bullae are considered ‘hearing specialists’. Audiograms of ‘hearing specialists’ show high sensitivity to sounds with sound levels as low as 60 dB re 1µPa (msp to tones) across a broad frequency range into several kHz (Figure 1).


7

40

60

80

100

120

140

160

180

10 100 1000 10000Frequency (Hz)

Thre

shol

d (d

B re

1 u

Pa)

Carp (Cyprinus carpio; Popper1972)

Catfish (Ictalurus punctatus;Fay & Popper 1975)

Cubbyu (Equetus acuminatus;Tarolga & Wodinsky)

Flathead minnow (Pimephalespromelas; Scholik & Yan2001)Goldfish (Carassius auratus;Enger 1967)

Herring (Clupea harengus;Enger 1967)

Figure 1: Underwater audiograms of some hearing specialists (from Nedwell et al. 2004). Fish of the family Clupeoidea, which includes herring (i.e. Clupea harengus, Figure 1), anchovy (Engraulis australis), pilchard (Sardinops sagax) and sprat (Sprattus sprattus) are examples of hearing specialists having specialised auditory systems which include a structure called the prootic bulla (a gas-containing sphere evolved from the bones of the ear capsule – Figure 2; Blaxter, 1980; Nedwell et al. 2004). A membrane divides the bulla into an upper part containing fluid and a lower part containing gas. Movements of the bulla stimulate both the utricular macula and the lateral line, improving sound receptivity. Many fish have a swimbladder (rather than the prootic bulla of Clupeoidea) which is physically linked to the inner ear. The swimbladder is a gas-filled cavity that can act to transfer an impinging sound waves pressure information to the fish ear end organs or otolith systems. Examples of fish having their swimbladder linked to the inner ear are the Otophysi (which include mostly freshwater species), including the order Cypriniformes (goldfish, carp, and minnows; Popper & Fay, 1993).


8

Figure 2: Anatomical structures involved in fish hearing: A- General anatomy of the hearing sturucture (a=inner ear, b= coupling, and c=swimbladder); B- prootic bulla. Fish with the prootic bulla generally have higher sensitivity than those with a swimbladder, and those with a swimbladder usually have greater sensitivity than non-specialists with no swimbladder (Nedwell et al. 2004). Examples of species that have no direct coupling between the ear and the swimbladder and which fall into the group of ‘hearing generalists’ are the blue gourami (Trichogaster trichopterus) and the oyster toadfish (Opsanus tau). Finally, there is increasing evidence that some species are able to detect very high frequency sounds (Popper et al., 2005). A selection of hearing generalist fish audiograms are shown on Figure 3.

A B


9

40

60

80

100

120

140

160

180

10 100 1000 10000Frequency (Hz)

Thre

shol

d (d

B re

1 u

Pa)

African mouthbreeder (Tilapiamacrocephala; Fay & Popper1975)Blugill sunfish (Lepomismacrochirus; Scholik & Yan2002)

Cod (Gadus morhua; Offutt1974)

dab (Limanda limanda;Chapman & Sand 1974)

Damselfish (Eupomacentrusdorsopunicans; Myrberg &Spires 1980)

Beau-Gregory damselfish(Eupomacentrus leucostictus;Tavolga & Wodnsky 1963)

Figure 3: Underwater audiograms of fish (from Nedwell et al. 2004), where red arrows indicate hearing generalists, and blue hearing specialists. Among the groups of fish without any known audiograms of sound sensitivities, are the syngnathids (pipefish and seahorses) and the elasmobranches (sharks, skates, and rays). Many syngnathids have been documented to produce sound (loud clicks), suggesting that sound is important to them for communication in the aquatic environment (Fish and Mowbray, 1970; Bergert and Wainwright, 1997; Colson et al., 1998; Ripley, 2006). The function of clicks may be associated with mating, to coordinate spawning (to signal readiness and orientation of mates), or to advertise prey availability. Among these contexts, feeding clicks are the most widely noted. For two species of pipefish studied, peak frequency measurements were highest between 2650 – 3430 Hz for H. zosterae, and 1960 – 2370 Hz for H. erectus (Colson et al., 1998). Elasmobranches rely on low frequency sound (as well as electro-chemical receptors) to locate distressed prey (Myrberg 1978). Studies on impacts from noise The extent of potential noise impacts on fish is not comprehensively understood. It is known however, that intense impulsive signals such as those produced from pile drivers, can cause fish kills, and signals of a smaller magnitude can cause behavioural changes (Nedwell et al. 2004). High-intensity sounds may temporarily or permanently damage fish audition. However, damage to hearing by intense sound depends on the auditory threshold of the receiving species and will therefore vary from species to species (Popper and Fay 1973, 1993). The highly variable auditory sensitivity of fish, means that it is impossible to generalize on the impact of impulse signals from one species to another. While there are no studies yet conducted that have been dedicated to measuring mortality in relation to noise exposure levels, there are many observations from explosive and pile driving sources. Nedwell et al. (2004) observed that fish kills occurred at a distance of 400 m from an explosive source but did not occur where the estimated received peak level was only 134 dB re 1µPa, (however no other distances were assessed in this study). Greene and Moore (1995) found that the mechanical impact of a short duration pressure pulse such as an explosion was best correlated with


10

organ damage. For fish, gas oscillations induced by high sound pressure levels can cause the swim bladder to tear or rupture, as has been shown in response to explosive stimuli in several reports (e.g., Alpin 1947; Coker and Hollis 1950; Yelverton et al. 1975). Other structures within the body can also be affected by exposure to intense sound impulses because of their small size or dynamic characteristics. There is some evidence to suggest that sound at sufficiently high peak to peak pressure levels can generate bubbles from micronuclei in the blood and other tissues such as fat. In fish, blood vessels are particularly small in diameter so bubble growth by rectified diffusion (Crum and Mao 1996) at low frequencies could create an embolism and burst small capillaries to cause superficial bleeding. This type of bubble growth may also occur in the eyes of fish where the tissue might have high levels of gas saturations. Explosive impacts such as that mentioned above, have mainly been studied for fish response to impulse signals, but similar effects have also been observed in fish exposed to short range impact signals from pile driving (Hastings and Popper 2005). Several studies have attempted to quantify non-mortality injuries that resulted from pile driving (mainly in the gray literature; Hastings and Popper 2005, Marty 2004), but the degrees of damage in these studies are not readily quantifiable and comparable among studies. Other unpublished reports have attempted to observe the behavior of fish during pile driving activities. Feist et al. (1992) found that there were more fish schools in an area when there was no pile driving activity than when there was pile driving activity. None of these studies reported any other notable effects on the fish or their behavior. At the same time, these observations were opportunistic observations of free-swimming fish rather than on animals with known received sound exposures related to pile driving activity. There are no studies that have focused on long-term effects of exposure to pile driving sounds that may lead to delayed death, or to other changes in behavior that could affect the survival of individuals or of populations of fishes. Furthermore there are no studies examining responses of fishes outside of the pile driving / explosive “kill-zone”, effects that although may not be immediate, may have significant effects on fish populations. Non-mortality effects may include temporary injury that heals, injury that leads to a slow death (e.g., break down of tissues in some organ system), temporary or permanent hearing loss, or movement of fish away from feeding grounds due to high signal levels. Finally, it is important to consider the effects of cumulative exposures on mortality, physiology, and behavior. For example consideration of the effects of exposure to multiple impacts from pile driving and the time between signals (one every few seconds for example) need to be made. Another aspect of cumulative exposure that needs investigation is a larger temporal length of exposure to repeated signals (repeated exposures several hours, days, or weeks later). A study on hearing loss up to a few days after exposure from a seismic airgun has been conducted (Popper et al. unpublished). This study involved exposing three species of fish to sounds from a seismic airgun (an impulse sound). Peak sound levels ranged between 205 and 209 dB re 1µPa peak pressure. They exposed a hearing generalist (broad whitefish), a hearing specialist (lake chub), and a species that is intermediate in hearing (northern pike). They found that the hearing generalist had no significant effects from air gun exposure, the lake chub indicated the most effect in temporary threshold shift, and the northern pike showed a significant hearing loss but less than that of the lake chub. Lake chub and northern pike returned to their respective normal thresholds after 18 to 24 hours. McCauley et al (2003b) exposed pink snapper to a regime of approaching and departing air gun pulses and observed strong behavioural changes and evidence of massive hearing damage which did not fully show until 60 days after the impulse exposure.


11

Table 1: Impacts for various fish species (TTS = temporary threshold shift).

Species Level of Impact Sound Source Received Sound Level Reference

Lake chub (fish)

TTS 200, 400, 1600 Hz sonar signals

100 dB re 1µPa (msp) Popper et al. 2005

Fish Alarm response 20 cui Air gun 156-161 dB re 1µPa (msp)

McCauley et al. 2003a

Goldfish TTS 500 and 800 kHz tones

149 dB re 1µPa (msp) for 4 hrs

Popper and Clarke 1976

Flat fish and invertebrates

injury 217 dB re 1µPa (peak) Cudahy et al., 1998

Pink snapper Hearing damage Air gun Ensemble of signals up to 185 dB re 1µPa (msp)

McCauley et al (2003a & b)

It is difficult to extrapolate for pile driving noise from studies using other types of signals (e.g., pure tones, explosions or air guns) since each sound source has particular signal characteristics in terms of duration, rise and fall times, and frequency content (Yelverton et al. 1975; Hastings et al. 1996; McCauley et al. 2002). Thus, specific signal components that affect marine animals may be different. Furthermore sound exposure levels are highly dependent on the characteristics of the environment which will affect propagation of the sound produced. Sound pressure levels do not necessarily decrease monotonically with increasing distance from the pile for example, since it depends upon the propagation characteristics of the medium. When assessing impacts from noise it is advisable to measure noise levels within the proposed impact areas and / or to conduct propagation modelling exercises, in order to develop exposure metrics that may correlate with mortality and different types of impacts (including damage) observed in exposed animals. Fish Kills – Key variables that appear to control the physical interaction of sound with fishes include the size of the fish relative to the wavelength of sound, mass of the fish, anatomical variation, and location of the fish in the water column relative to the sound source. Most studies on fish kills have been related to explosive blast pressure waves consisting of an extremely high peak pressure with very rapid rise times (< 1 ms). Yelverton et al. (1975) exposed eight different species of fish, five with ducted swim bladders and three with non-ducted swim bladders to blasts. Fish sizes ranged from 0.02 g to 744 g body mass and included small and large animals from each species. The fish were exposed to blasts having high peak pressures with varying impulse lengths. Yelverton et al. (1975) found a direct correlation between body mass and the magnitude of the “impulse,” (characterized by the product of peak overpressure and the time it took the overpressure to rise and fall back to zero), which caused 50% mortality. Trasky (1976) also reported significant differences between adult fishes, and salmon and herring fry in the lethal blast overpressure from buried seismic charges.

Additional studies using explosives suggest that there is far more damaging to fishes with swim bladders than to species, such as flatfish, that do not have such air chambers (e.g., Baxter et al. 1982, Hastings and Popper 2005). It has also been shown that the effects on fish decline rapidly with distance from the explosion as the peak overpressure decreases and the impulse duration increases. Similarly, a study by Kearns and Boyd (1965) suggested that the extent of fish kill decreases with increasing distance of the fish from an air gun source, and another unpublished study indicated no mortality from seismic air gun shots at considerable distance (4000 m) from the source. There is evidence that the effects of explosions vary by species, even when all test fish have a swim bladder (Govoni et al. 2003). Based on these and other studies (i.e. Yelverton et al. 1975), it is clear that there is considerable variability in the effects of explosive blasts on fishes, and that the


12

variables include received sound energy, presence or absence of gas bubbles (i.e., swim bladder), mass of fish and perhaps body shape, and biomechanical properties of the swim bladder wall. Yelverton et al. (1975) suggested that a metric related to the amount of sound energy received, such as the sound exposure level rather than just peak pressure correlates with swim bladder and other tissue damage as well as mortality in fish. They concluded that peak pressure alone did not correlate with damage because peak pressure was kept constant and the impulse duration was varied or vice versa in their study. The injuries observed included swim bladder rupture, kidney damage, and liver damage. Govani et al. (2003) also concluded that the total energy in the sound wave, regardless of pressure polarity, was responsible for observed effects of submarine detonations on juvenile pinfish (Leiostomus xanthurus). Moreover, Stuhmiller et al. (1996) suggested that incidence of blast injury to the lung and lethality correlated with total energy in the wave normalized by lung volume in terrestrial animals. It has been suggested that the large negative overpressure characteristic of pile driving sounds may be more damaging to the swim bladder than the initial positive overpressure (Trasky 1976) because of the swimbladder expansion during the negative phase. Bailey et al. (1996), however, found that a sound pulse having a large positive peak overpressure was at least as damaging as one having a large negative peak overpressure of approximately the same level and time duration, to the lungs of mice submerged in water. Damage increased with magnitude of pressure incident at the lung, but histology showed no differences between the effects of leading positive or negative pressures. Mouse lungs had increasing hemorrhage with increasing exposure levels regardless of the polarity of the peak overpressure. These findings indicate that injury would correlate with the work done on the lung tissue, which would be equivalent to the total energy in the sound wave if delivered over a short enough time frame. Taking into consideration the limitations of any extrapolation when assessing level of fish kill, Hastings and Popper (2005) made a preliminary attempt at estimating levels of impact. Their reasoning was that if transient sounds, such as those produced by pile driving, could be characterized using a waveform similar to the ideal impulse sound (Friedlander wave; Hamernik and Hsueh 1991), then effects of pile driving on aquatic animals could potentially be extrapolated from data based on effects observed from exposure to other transient signals (e.g. explosives, air guns, sonic booms) or other transient waveforms that could be described by the Friedlander wave model. These estimates could provide a basis for developing interim guidance for exposure to sound from pile driving until more research is completed. Hastings & Popper (2005) show an approximation of a pile driving sound using a Friedlander wave, and compare the temporal characteristics, sound exposure spectral density, and cumulative pressure squared over time, respectively, for the idealized and actual pile driving sound and found that pile driving waves are very close in exposure characteristics, which indicate that the key characteristics for pile driving may be the peak positive and negative pressures and their time durations, which are combined to calculate the cumulative pressure squared and Sound Exposure Level (SEL). Thus Hastings & Popper (2005) suggest that a systematic approach to approximate pile-driving signals using mathematically modeled Friedlander type waves could provide a way to determine how data, which have been obtained in effects studies using blasts or other transient sources, relate to different pile driving scenarios. Based on the above methods and extrapolating from Yelveron et al. (1975), they came up with estimated impacts relative to Sound Pressure Level (SPL) shown below (Figure 4). These estimates are based broadly on a very limited set of fish species.


13

Figure 4: Estimated sound exposure level (SEL) that results in 50% mortality based on data for exposures to a single explosive sound as reported by Yelverton et al. (1975) and modeled as an ideal impulse wave. (Friedlander waveform as described by Hamernik and Hsueh 1991). Directly from Hastings and Popper (1995). Based on the studies that have been done, Hastings (in 2002) recommended 150 dB re 1µPa (msp) and 180 dB re 1µPa peak for impulsive sounds as the thresholds for protecting salmon against physiological damage (NOAA/USFWS 2005), and 150 dB re 1µPa (msp) level as the threshold for disturbance to salmon, bull trout, marbled murrelets. Based on their assessment, sound pressure levels in excess of 150 dB re 1µPa (msp) are expected to cause temporary behavioural changes, such as elicitation of a startle response, disruption of feeding, or avoidance of an area. The USFWS (2004) has also identified underwater threshold sound levels for foraging marbled murrelets. As with bull trout, the injury threshold remains at 180 dB re 1µPa peak. 3.3 Aquatic birds Studies of sensitivity to sound underwater Little information is available on the hearing of birds underwater. In general, it is thought that penguins are sensitive to frequencies over the range of 100 Hz to 15 kHz. A study of the hearing of the Blackfooted penguin (Sphenikcus demersus), in terms of the cochlear potentials showed best sensitivity in the region of 600 to 4000 Hz (Figure 5, Wever et al., 1969). Generally the hearing of penguins appears to be similar to the hearing of other smaller birds, with greatest sensitivity between 1000-3000 Hz. Studies of Impacts of Noise on birds No published studies to date, that the authors are aware of, investigate the impact of underwater noise on birds. 3.4 Marine Mammals Studies on sensitivity to sound Dolphins – Odontocetes (toothed whales and dolphins) are known to communicate at frequencies from 1 kHz to greater than 20 kHz and to use echolocation over the frequency range 1-150 kHz (pending species and their desired use of the sonar). Hearing in the bottlenose dolphin extends from at least 40-75 Hz to as high as 80-150 kHz with best sensitivity in the frequency range of ~15 kHz to 50 kHz (ie. see Figure 6).


14

Figure 5: Audiogram of blackfooted penguin (Sphenikcus demersus) in air (from Wever et al. 1967).

Figure 6: Underwater audiogram of the bottlenose dolphin (Tursiops spp.; figure from Nedwell et al. 2004). Seals – Due to the difficulty in working with marine mammals, many of the seal underwater audiograms identified in literature are from a single Northern fur seal (Callorhinus ursinus). In the audiograms below, at least one individual was an aged animal who had been in a zoo (Moore and Schusterman, 1987), hence the animal may not be representative for the species. The audiogram, however suggests that auditory sensitivity is greater at lower frequencies than for the bottlenose dolphin (Figure 7).


15

Figure 7: Underwater audiogram for the northern fur seal (Callorhinus ursinus; figure taken from Nedwell et al. 2004). Studies of impacts of underwater noise on marine mammals Many studies of underwater sound impacts on marine mammals record a response (behavioural impact) to a noise source without recording the received level of the noise. Therefore, availability of information on response to noise level is scarce. However, the effects of elevated noise levels on marine mammals is believed to include: • masking (which can lead behavioural responses); • behavioural response; • temporary threshold shift (TTS); • permanent threshold shift (PTS); • organ damage; and • death. Behavioural responses will generally indicate some level of disturbance (that may correlate with physiological impact). There are limited data on noise levels that cause TTS or PTS in marine mammals (Richardson et al. 1995). Richardson et al. 1995, extrapolated on levels required to cause PTS from information on human threshold levels where an impinging signal which was 80 dB above best hearing threshold caused PTS in humans (exposure of 8 hours a day over ~10 years). The National Marine Fisheries Service in the US (NMFS) considers that underwater Sound Pressure Level (SPL) above 190 dB re 1µPa msp (impulse) could cause temporary hearing impairment in harbour seals and sea lions and SPL above 180 dB re 1µPa msp (impulse) could cause temporary hearing impairment in whales (Vagle 2003). Based on this, the NMFS has established a safety zone of 180 dB re 1µPa msp (impulse) for grey whales. If marine mammals are found within the safety zone, pile driving will be delayed until the whales move out of the area (NMFS 2006). The injury threshold is identified as 180 dB re 1µPa msp

for cetaceans and 190 dB re 1µPa msp

for

pinnepeds. NMFS states that cetaceans should not be exposed to underwater noise exceeding 180 dB re 1µPa msp in order to avoid permanent physiological damage to hearing. The underwater


16

disturbance threshold for cetaceans is 160 dB re 1µPa msp for impulse noises and 120 dB re 1µPa

msp for non-impulse, continuous, industrial noises. These levels were set based on data on the effects of anthropomorphic noise on gray whale migration (from studies by Malme et al. 1984, as cited in NMFS 2006). In Australia the Environment and Protection Biodiversity and Conservation Act sets two levels of mitigation requirements for impulsive seismic signals, based one whether they will fall below or exceed 160 dB re 1µPa2.s (SEL) for 95% of the time at 1 km range. This threshold value is set as the EPBC Act considers it the threshold at which TTS will occur in marine mammals. 3.5 Summary Although each species has its unique hearing sensitivity, and many species fall outside of the average sensitivity range for its taxonomic group, it can be broadly said that the following sensitivities correspond to the groups listed below:

1) Invertebrates: have a measurable level of sensitivity to sound energy but little is known except that some species (some mollusks and crustaceans) appear to resond to low-frequency impulsive sounds;

2) Fish (Hearing Generalists and Specialists): highly variable sensitivity to sound energy, with highest sensitivity at mid frequency ranges (100 Hz to 1 kHz). Hearing generalists may have a narrower frequency range of sensitivity;

3) Penguins: although not much is known about the little penguin, it is likely to have its most sensitive hearing in agreement with the smaller birds – between 1 kHz and 4 kHz;

4) Odontocetes (dolphins and killer whales): odontocetes (or the toothed whales), including bottlenose dolphin, common dolphin and killer whale, have very sensitive hearing which is centred at high frequencies (10-100 kHz);

5) Seals: the Australian fur seal (based on measurements from the northern fur seal) is likely to have its greatest sensitivity at mid to high frequencies (3-20 kHz); and

A summary of hearing sensitivity from measured audiograms of the fish and marine mammals discussed earlier in this section is presented in Figure 8. The arrows within the figure indicate a general frequency range of sensitivity for these animals.


17

0

20

40

60

80

100

120

140

160

180

10 100 1000 10000 100000

Frequency (Hz)

dB re

1 u

PaAmerican shad (Alosa sapidissima; Mannet al. 1997)African mouthbreeder (Tilapiamacrocephala; Fay & Popper 1975)Blugill sunfish (Lepomis macrochirus;Scholik & Yan 2002)Bonefish (Albula vulpes; Tavolga 1974)

Carp (Cyprinus carpio; Popper 1972)

Catfish (Ictalurus punctatus; Fay & Popper1975)Clown knifefish (Notopterus chitala;Coombs & Popper 1982)Cod (Gadus morhua; Offutt 1974)

Cubbyu (Equetus acuminatus; Tarolga &Wodinsky)dab (Limanda limanda; Chapman & Sand1974)Damselfish (Eupomacentrusdorsopunicans; Myrberg & Spires 1980)Beau-Gregory damselfish (Eupomacentrusleucostictus; Tavolga & Wodnsky 1963)Flathead minnow (Pimephales promelas;Scholik & Yan 2001)Goldfish (Carassius auratus; Enger 1967)

Herring (Clupea harengus; Enger 1967)

Series9

Northern Fur seal (Callorhinus ursinus)

Bottlenose dolphin (Tursiops spp.)

Fish

Marine Mammals

FishBottlenose

Northern Fur

Figure 8: Summary of estimated hearing range based on audiograms of fifteen fish species and three marine mammals (from Nedwell et al. 2004).

4 Characterization of pile driving noise Impact pile driving was the primary noise source considered in this document. While the noise of vibratory piling may reach high levels it is not known to cause the potential serious physiological impacts which impact pile driving is known to produce. The authors have measurements which show that typically at any given range vibratory piling is on the order of 20-25 dB below highest mean squared pressure levels of impact pile driving. Thus impact pile driving only has been considered here. The machinery used in wharf construction will be located above water and while some of this noise energy may enter the water, the coupling efficiency is generally poor and so the noise levels transmitted into the river are considered to be of low to negligible levels. There will be small vessel traffic associated with wharf construction but thus would be no more than for any similar sized vessels operating elsewhere along the river and so was not considered. This section identifies attributes (technologies to be used, associated source characteristics, etc.) that are relevant to the assessment of the impacts of noise produced by pile driving for the construction of the proposed Bell Bay Pulp Mill wharf. The section draws on relevant studies in the literature which describe pile driving noise, as well as pile driving measurements made around Australia undertaken by the consultants during recent years. 4.1 Equipment proposed for the construction of the wharf Both diesel and vibratory hammers suspended from a crane are proposed for use during wharf construction. Vibratory hammers vibrate the pile into the sediment by use of an oscillating hammer placed on top of the pile. The vibratory action causes the sediment surrounding the pile to liquefy and allows the pile to be driven into the sediment with less force. In most cases piles cannot be


18

driven by vibratory hammers to their desired depth so an impact hammer (such as a diesel or hydraulic hammer) is used to finish driving the pile in. The wharf construction will include pile driving works consisting of piles placed at: a) each groyne end (approx 10 piles in each groyne); b) vertical and raking piles to support the suspended deck (approximately 214); and c) two berthing dolphins. The berthing dolphins consist of a concrete pile head on three piles with a fender attached. The normal method of installing piles is to initially sink them by vibratory driving until the substrate is sufficiently hard to require impact hammering, where upon impact pile driving is carried out until a predetermined load capacity is attained. The piling operation will be from a barge mounted crane moored over each pile location. As piles are driven extra sections will be welded on as the depth to founding rock is sought. Specific details of the pile dimensions and the types of vibratory and impact pile driving heads to be used were not available. 4.2 Sound signal produced by pile driving Impact pile driving – Pile driving sounds are impulsive signals with examples shown in Figure 9 and Figure 10 (authors data). The frequency bandwidth for most of the energy in pile driving sounds is typically below 1,000 Hz as shown in Figure 10 (lower plot). This energy content of pile driving signals overlaps the same hearing bandwidth of many species of marine animals, particularly fish.

Figure 9: Waveform of blows and multiple bounces produced by pile driving at a range of 303 m at Twofolds Bay, NSW (from McCauley et al. 2002). Vagle (2006) reports differing sound characteristics produced by pile driving which depended upon bottom type and pile type (cedar vs. steel) and size. The authors suggested that with harder bottoms such as gravel or hard clay and larger piles, noise levels increased. Although the study conducted by Vagle (2006) was limited, it shows evidence of differing noise signatures resulting from different pile, pile driver, and environmental attributes. Sound levels from various projects are summarised by Hastings and Popper (2005) below (Table 2), although the seafloor bottom type for each scenario has not been described.


19

Figure 10: Signal waveform, the cumulative energy, and the power spectra of a pile driving signal received at 303 m from Twofolds Bay, NSW (from McCauley et al. 2002). Table 2: Summary of Measured Underwater Sound Levels Near Marine Pile Driving, directly from Hastings and Popper (2005).

Pile Type Distance from Pile (m)

Peak Pressure (dB re 1µPa)

msp (impulse) Pressure (dB re 1µPa)

SEL (dB re 1µPa2-s)

--Various Projects Timber (12-in) Drop 10 177 165 157 CISS (12-in) Drop 10 177 165 152 Concrete (24-in) Impact (diesel) 10 188 176 166 Steel H-Type Impact (diesel) 10 190 175 -- CISS (12-in) Impact (diesel) 10 190 180 165 CISS (24-in) Impact (diesel) 10 203 190 178 CISS (30-in) Impact (diesel) 10 208 192 180 --Richmond-San Rafael Bridge CISS (66-in) Impact (diesel) 4 219 202 -- CISS (66-in) Impact (diesel) 10 210 195 -- CISS (66-in) Impact (diesel) 20 204 189 -- --Benicia-Martinez Bridge CISS (96-in) Impact (Hydraulic) 5 227 215 201 CISS (96-in) Impact (Hydraulic) 10 220 205 194 CISS (96-in) Impact (Hydraulic) 20 214 203 190 --SFOBB East Span CISS (96-in) Impact (Hydraulic) 25 212 198 188 CISS (96-in) Impact (Hydraulic) 50 212 197 188 CISS (96-in) Impact (Hydraulic) 100 204 192 180

For steel pipes of differing diameters driven into an area of mud/silt at two different sites in Canada, the sound energy levels below 10 kHz differed remarkably (Figure 11; Vagle, 2003). Vagle (2003) suggests that the excessive energy levels at the lower frequencies likely caused fish deaths observed in the study.


20

Figure 11: Spectra of a pile driving blow at Canada Place as compared to an equivalent blow in Sicamous Narrows (from Vagle, 2003).

5 Site characteristics and estimated noise levels 5.1 Site characteristics The proposed location of the wharf is shown on Figure 12. The wharf site is 18 km upstream. The seabed in the vicinity of the wharf has been described in Anon (2007) as primarily of fine sand or silt/clay down to several m depth. The depth of the sediment layer has been described in Anon (207) as “up to 40 m”, without specifying actual depths. The bathymetry at the wharf edge is for a water depth of around 15 m. On heading up or downstream parallel to shore from the wharf edge the bathymetry stays reasonably constant at close to 15 m depth. On moving across the river from the wharf edge the seabed drops to a maximum depth of over 20 m between 280-450 m from the wharf and begins shallowing at greater range. The maximum distance from the wharf across the river to the 10 m depth contour on the other side is around 540 m. The river opposite the wharf was considered well mixed due to the strong prevailing tidal regime and fully flushed by the ocean, although in strong winter rains the salinity would be expected to be lower than typical ocean values this far downstream. Based on the World Ocean Data Atlas 2005 data set (WODA) for mean water column parameters for Bass Strait water at the Tamar river mouth (considered as 41o S 146o 30’ E), the mean sound speed through the water column was calculated to be around 1507 ms-1 in summer and 1498 ms–1 in winter. There was little variability in vertical sound speed calculated down to the river bed during summer or winter.


21

Figure 12: Proposed location of BBPMP wharf (blue cross). The longitude bars shown are 146o E, the latitude bars are: top 41o 8’ S; and the horizontal line running through the image center is 41o 10’ S (taken off AUS chart 168).


22

5.2 Modeling pile driving noise The most significant parameter of impulsive noise, such as pile driving, potentially resulting in physiological damage in fish is the rapid onset of a high positive or negative peak pressure, followed almost immediately by a high peak pressure of the opposite sign. Several workers have estimated exposures required to damage fish by converting unit systems (see Figure 4 above from Popper and Hastings, 2005 for exposures to injure fish or Table 4 of inferred exposures required to damage marine mammals from Richardson et al. 1995). An impulsive signal’s initial peak pressure, the onset or rise-time of this initial peak, and the reverse polarity peak immediately following the initial peak (ie. peak-peak pressure), are critical parameters to be calculated when attempting to assess the potential of a pile driving signal to cause physiological damage. Unfortunately the typical sound transmission models available do not readily calculate the peak-peak signal parameters of the way a source signal changes with range. The sound transmission models rather calculate a phase and amplitude correction for a given source location in a certain environment, at defined points in space and discrete frequencies. The absolute value of these corrections is termed the transmission loss at the respective spatial point and is a measure of the loss of energy from the source at that spatial point for that frequency. Thus by running models at set frequency intervals and assuming sound transmission loss to be similar across bands spaced about frequencies the model is run at, then measures of the energy loss of a real source with range can be calculated. These loss estimates can be applied to the source level energy estimate to give the received energy level at the specified spatial point using the modeled environment. Since here we are interested in evaluating peak signal parameters as well as estimating received energy levels, then a different approach was used. We use the phase and amplitude correction given by the sound transmission model at discrete frequencies and spatial points and apply this to the Fast Fourier Transform (FFT) of a representative source signal at the appropriate frequency. The source signal FFT phase and amplitude values are adjusted at the respective frequencies the model was run at, a process termed applying a transfer function to the source signal. This process is repeated for as many frequencies as possible and the modified source FFT converted back to the received signal (an inverse FFT). We have used 1 Hz frequency steps across the frequency band from 10 Hz to 1 kHz (the model was thus run at 991 frequencies). Thus the process used here has reconstructed the pile driving signals at set spatial points about the pile and used the reconstructed signal to calculate the respective signal descriptors (primarily peak-peak levels). In order to be able to run a sound transmission model with sufficiently fine frequency resolution to enable a reasonable transform function to be applied to the source signature, some simplifications were required. Of several different types of sound transmission model available to us, the model SCOOTER has been found to be the most stable and reliable, whereas many of the other models we have (RAMS, RAMGEO, KRAKEN, Bellhop) either are not suited for the task or have pedantics and often give problems with particular scenarios. Since 991 runs of the model were required (one model run at each frequency) we have opted to use the model SCOOTER. This model requires a constant environment, hence a uniform depth profile, water column sound speed structure and seabed type. Thus we have used a uniform 18 m depth profile along a single track and assumed this as the bathymetry heading across the river from the wharf. While this is not the case it is a conservative (worst case) approximation as most piles will be driven in water depths of < 18 m which will involve slightly higher losses with range than would be the case if the pile were driven in at 18 m depth. The modeling output is also blurred slightly as the noise source from a pile being driven extends all along the pile, and may also extend into and below the seabed (ie. some noise energy of the pile being struck will radiate into the water column through shallow seabed layers). In order to run the model a single source location is required. Here we have placed the source at the seabed, or 18 m depth.


23

As mentioned above, the seabed layering near the wharf has been defined by Anon (2007) as being an almost equal mixture of fine sand and clay/silt with an unknown thickness across the river (“up to 40 m thick” near the wharf), lying over a dolerite rock base. We can assume this layer thins near the higher current areas of the river due to current scouring. Thus for the modeling we have assumed a uniform 20 m thickness of a mixture of fine sand, silt and clay sediment overlying the rock base. The geo-acoustic parameters used for the sediment layer were an average for sand and clay, and are listed in Table 3. The basement was considered to be rock with the geo-acoustic parameters of basalt. The water column properties have not been defined in any documentation so a profile as given by the mean 2005 WODA summer profile in the Tamar river mouth was used, giving a uniform 1507 ms-1 sound speed profile with water depth. The full layering used in the underwater sound modeling for pile driving signature prediction is given in Table 3. Table 3: Layering used in setting up the underwater sound transmission model. The parameters are: Cp= compressional wave speed ; Cs = shear waved speed ; ∝p = compressional wave attenuation; ∝s = shear wave attenuation Layer Thickness

(m) Density kg / m3

Cp ms-1 Cs ms-1 ∝p (dB / λ) ∝s (dB / λ)

Water column 18 1000 1507 (uniform)

0 0 0

Fine-sand / silt / clay

20 1700 1580 200 1.0 1.5

basement 2600 5000 2500 0.1 0.2

A further assumption was required in selection of a pile driving source waveform. The shortest range pile driving signal available to the authors was from 15 m range from a pile being driven. Here 17 signals were measured from steel piles being driven, in a uniform 13 m depth temperate environment over a fine mud/clay seabed. The signal from this batch with the greatest amplitude was selected as the source signal, with its waveform shown on Figure 13. Converting this signal back to a source level signal required knowing where the signal originated from along the pile. As this was not known and in reality the signal received at 15 m was most likely a combination of signals generated along the in-water pile length and a multitude of surface bounces, then the signal at 15 m was considered representative of the pile driving source signal shape. Only the initial impact pulse was used (first 0.2 s shown on Figure 13) and this adjusted for loss between the receiver and pile assuming spherical spreading (ie. multiply signal amplitude by range in m). While this was a further approximation in the modeling the resulting values are indicative of pile driving parameters with range.


24

Figure 13: Waveform of a pile driving signal measured at 15 m range by the authors. The reconstruction of pile driving signals was then carried out by: • Setting up the model SCOOTER with the seabed parameters listed above; • Placing the source (pile) depth at the seabed (18 m) • Using an 18 m uniform depth out to 500 m range sampled at 0.5 m increments (range and

depth, giving 36 depth and 991 range steps) • Running the model for frequencies between 10 and 1000 Hz in 1 Hz steps (991 frequencies). • Applying the resulting phase and amplitude transform function to the source signal for the 991

frequencies at each spatial point selected to give the estimated received signal • Characterising each calculated signal for a suite of impulse signal descriptors, as defined in

McCauley et al (2003a). The resulting grids of estimated peak-peak and SEL levels through the water column out to 500 m are shown on Figure 14. The results show spatial differences in received levels, which may be often significant at small spatial scales. These differences in received level may occur vertically or horizontally, with variations of up to ten dB on the scale of a few m vertically and 20 dB over 10’s of m horizontally. This implies that the precise location of fish will impact their propensity for suffering different types of physiological damage, with the probability that two fish metres apart may receive quite different pile driving exposures. As a check on the accuracy of modeling the results for peak-peak and SEL values calculated at 5 m depth were compared with several sets of similar values for measurements of pile driving signals available to the author, as shown on Figure 15. The calculated values agree well with the measured values out to the limit of range overlap between the measurement sets (200 m). To further reinforce the potential for small scale spatial scale differences in position resulting in large differences in received pile driving signals the full ensemble of curves of calculated SEL pile driving signals at each of the 36 depth increments are plotted with range on Figure 16. An envelope of received levels can be seen at any one range, emphasizing that typically 5-7 dB differences in levels may be experienced throughout the water column depending on the animals vertical position, and in extreme cases this may reach 15 dB. The lower received peak-peak level curve was present just below the water surface (this trend is evident on Figure 14). The ensemble of estimated received SEL curves were averaged through the water column at each range point, over 1.5 m depth


25

to the seafloor (to avoid bias from the near surface sound shadow), and is shown on Figure 17 with the 95% confidence limit curves.

Figure 14: Calculated underwater sound fields for pile driving noise across the river using the source waveform shown in Figure 13. The top panel is peak-peak levels and the lower panel sound exposure level units.


26

Figure 15: Calculated peak-peak (top) and SEL (bottom) pile driving levels (blue curve) along with measurements available to the authors (note measurements show mean values with 95% confidence limits as error bars).


27

Figure 16: Calculated pile driving SEL signal levels along water depths from 0.5 to 18 m depth in 0.5 m depth increments.

Figure 17: Mean trend (blue curve) of calculated pile driving SEL signal levels averaged through the water column from 1.5 to 18 m depth, with 95% confidence limits shown (red curves). 5.3 Estimated underwater noise levels at nearby aquaculture facility A commercial aquaculture facility exists across the river and slightly downstream of the proposed wharf construction site. At its closest point the boundary of the aquaculture farm was estimated at 650 m from the wharf outer edge with an unobstructed deep water path between the wharf and aquaculture farm. At its most distant point of a reasonable transmission path (before the water shallowed to less than 10 m depth) the aquaculture boundary was around 960 m off the wharf. Extrapolating the trend of the predicted depth-averaged SEL curve (+95% confidence limits) with range using a fitted curve of the form: RL = -15.24*log10(R) – 0.0086*R + 196.7, r2 = 0.9985 where RL is the received level in dB re 1µPa2.s, R is range in m and r2 is the correlation coefficient, gave an estimated level of 148 and 143 dB re 1µPa2.s at the closest and most distant


28

boundaries respectively. To estimate cumulative sound loadings (CSEL) at the aquaculture operation from multiple pile strikes of the same intensity is simply: CSEL = RL + 10*log10(N) Where CSEL is the cumulative sound loading, RL is the received level of one pile strike and N is the number of strikes. 5.4 Summary The impulse and vibratory pile driving equipment proposed for use in the Tamar River have most of their energy below 1 kHz (vibratory pile driving measured by authors but not reported here). The potential transmission of impact pile driving noise across the river was modeled using a worst case scenario where: • The highest level pile driving signal measured previously by the authors at the shortest range

(15 m range in a uniform depth water of 13 m) was adjusted for spherical transmission loss and used as a source signature for pile driving

• An environment of: constant 18 m water depth; a 20 m layer of fine sand / silt / clay overlying basement rock; and a uniform 1507 ms-1 sound speed through the water column was set up in the sound transmission model SCOOTER

• The model SCOOTER was run to retrieve amplitude and phase corrections at frequencies from 10 Hz to 1 kHz in 1 Hz steps, using a spatial grid of points encompassing 5-500 m in range at 0.5 m resolution and 0.5 – 18 m depth at 0.5 m resolution

• The appropriate amplitude and phase corrections were applied to the source pile driving signal at each spatial point in the grid to give an estimated received pile driving signal;

• Signal descriptors for calculated received signals at each spatial point were obtained The resulting estimates of pile driving signal levels agreed well with measured results from slightly different environments (primarily different water depths and seabed layering). The estimated sound field across (and along the river) showed considerable variability with spatial location, vertically and horizontally with differences of up to 10 dB over a few m of vertical range and tens of m horizontal range. The estimated mean received level SEL, mean squared pressure and peak-peak curves with range were calculated by averaging through the water column from 1.5 m to the seabed. These curves have been used in estimating ranges for various impacts below.

6 Assessment of Impacts 6.1 Critical frequencies The frequency spectra from the pile driving example 257 m from the source at Twofolds Bay (NSW, Figure 9 and Figure 10) shows highest sound levels between 100 Hz and 300 Hz although a significant level of noise (above ambient) occurs to frequencies above 1 kHz (Figure 13). Thus the frequency content of pile driving overlaps the frequencies of greatest sensitivity of many fish species (ie. Figure 3 and Figure 8). For toothed whales and seals while the energy content of pile driving does not overlap the region of their greatest hearing sensitivity, the pile driving energy does still have a reasonable overlap (ie. Figure 6 and Figure 7). 6.2 Serious physiological impacts For estimates of impacts of pile driving on fish the values given by Popper and Hastings (2005) are used to estimate the potential incidence of mortality. According to their predictions, fish of body mass over the range 10-1000 g will suffer 50% mortality when receiving a single impulse with SEL in the range 195-200 dB re 1µPa2.s. The expected maximum SEL to be received by any fish from the modeling work is less than 185 dB re 1µPa2.s (at 5 m from the pile, Figure 16), indicating a


29

single pile driving strike is unlikely to cause serious injury to any fish other than those possibly within 5-10 m of the pile (given uncertainties and assuming spherical spreading loss from 1 to 10 m). This assumes that the source signature used, which was previously measured from a typical pile used in wharf construction, is indicative of the signals to be generated during the proposed Gunns wharf construction. To receive an equivalent sound loading from multiple pile strikes requires adding the cumulative energy of each pile driving signal and assuming that the cumulative sum of energy has a similar impact to a single pile strike. This was done with the modeled mean estimated SEL curve with corresponding 95% confidence limits added, to determine the number of pile strikes required to reach a 195 dB re 1µPa2.s signal level. The resulting calculations are shown on Figure 18. The typical impact hammer heads used in driving wharf piles have measured repetition rates of 3-6 s (authors data, for example see Figure 9). An impact pile driving event can vary in length, with measured times for a pile ranging from 10 to 30 minutes. Typically during construction a sequence of piles, perhaps as many as six, are set up together and hammered in sequentially, with time breaks for various tasks between piles, tea/meal breaks or breakdowns. Assuming a worst case scenario of a highly efficient team working a three hour stretch of continuous piling at a 4 s hammer interval and with a 50% hammering duty cycle, gives 1350 strikes. This equates to around a 120 m range from the pile for an animal to experience an equivalent sound loading of 195 dB re 1µPa2.s. This assumes the animal remains at the location over the three hour period, which is unlikely given the strong tidal streams in this part of the Tamar river. A more probable working scenario is for 30 minutes of continuous work. Using a 3 s hammer rate this is 600 hammer events, giving a range out to around 80 m from the pile to experience a 195 dB re 1µPa2.s cumulative sound exposure. For mortality to occur from the cumulative sound energy of continuous piling assumes that an equivalent sound loading over an extended time period has the same impact as a single more intense impulse with the same total energy. Given the mechanics of the injury mechanism in fish from impulse signals this is considered unlikely, although we have calculated the number of piles required to produce these cumulative sound exposures.

Figure 18: Estimate of number of pile strikes with range required to produce an equivalent sound loading of 195 dB re 1µPa2.s, or that at which 50% mortality of fish in the range 10-1000 g may be expected. Calculations based on depth averaged SEL levels encompassing 95% of estimated values. For a fish swimming past the wharf area (ie. migrating up or down the river) we can estimate cumulative sound loading with range from the pile. If we assume slack water, so that the fish swims


30

at its endurance speed un-assisted by current then we can use estimates from Blaxter (1969) of endurance fish swimming speeds of 2-3 body lengths/s for most fish. If we assume a small fish (100 mm) migrating up or down the river then the trajectory in the river and a hammer rate can be used to estimate the total cumulative sound exposure received as it swims past the wharf area. This has been calculated for ranges from 10-400 m perpendicular to the wharf with each run starting 500 m from the closest point of approach (ie. a 1 km track), the fish assumed to swim on a constant heading and course, assuming a 3 s hammer rate and using the depth averaged (over 1.5 to 18 m depth) SEL experienced by 95% of fish throughout the water column. The results are shown on Figure 18. The swim times would range from 83 minutes for the slowest and smallest fish to 22 minutes for the largest and fastest fish used. Based on the estimates of Figure 19 then we can assume that only the slower and smaller fish swimming past (at slack water) and within perhaps 50 m of the pile driving event will receive cumulative sound loadings equivalent to that for a single event which is believed to cause 50% mortality. Note that increasing the start and end point ranges of the fish trajectory used in the above estimations will not change the results. For the high level impact types discussed it will only be a relatively small number of pile strikes about the shortest range to the pile which will contribute the bulk of the cumulative SEL received by the fish.

Figure 19: Calculated cumulative sound exposures for fish swimming past a pile being driven (see text for assumptions). For marine mammals the speculative Damage Risk Criteria (DRC, as inferred from auditory DRC values for humans) given by Richardson et al. (2005) for impulsive signals for marine mammals has been used to estimate potential risk of some permanent hearing damage. These values are presented in Table 4, in units of mean squared pressure. Mean squared pressure (msp) is a poor metric for impulsive signals, as to calculate mean squared pressure one needs to define an averaging time over the impulse, which means defining its start and end. This is almost never done adequately or stated in the literature, hence it is difficult to compare values presented in msp for impulse measures.


31

Table 4: Inferred Auditory Damage Risk Criteria for marine mammals exposed to noise pulses underwater (from Richardson et al., 1995) in units of mean squared pressure.

Number of pulses Speculative DRC (in dB re 1uPa) for marine mammals listening in water

Marine mammal with hearing threshold of 40 dB re 1uPa

Marine mammal with hearing threshold of 70 dB re 1uPa

100 long (>200 ms) 178 208

10 long (>200 ms) 183 213

1 long (>200 ms) 188 218

1 short (25 ms) 214 244

Using similar rationales to that defined above for fish but using mean squared pressure units averaged over the time taken for 90% of the pile drive energy to pass, the calculated depth averaged mean squared pressure with range for a single pile strike is shown on Figure 20. This curve has a strange shape beyond 450 m due to the signal length changing at around 450 m which alters the way the mean squared pressure is calculated. The depth averaged curve of SEL shown on Figure 16 is more indicative of the energy delivered by the pile strike at any range. Using the values given in Table 4 and assuming only toothed whales (dolphins) enter the river, implying use of the higher threshold category (hearing threshold of 70 dB re 1µPa) then even for 100 pulses the DRC criteria are not exceeded, hence it is unlikely that dolphins in the river will experience any permanent harm from the pile driving, assuming the modeled pile strike source waveform is indicative of that to be used.

Figure 20: Calculated depth averaged mean squared pressure with range for a single pile driving event. 6.3 Temporary hearing impairment For estimating hearing impairment in fishes the data from McCauley et al (2003a & b), in which pink snapper exposed to seismic survey noise suffered hearing damage, has been used. In this exposure the total cumulative energy received by the fish in the exposure which caused the damage presented was 187.5 dB re 1µPa2.s with the maximum level experienced 181.2 dB re 1µPa2.s (authors data). The maximum peak-peak pressure reached was 202 dB re 1µPa.


32

Using these values then we can expect hearing damage to occur in fish out to 60 m range from an individual pile strike (approximate range at which a peak-peak level of 202 dB re 1µPa was predicted to be reached, Figure 15, top panel). Using the cumulative energy of 187.5 dB re 1µPa2.s from McCauley et al (2003a&b) then we can estimate the number of pile strikes to reach this level (calculations as for Figure 18), to get the curve shown on Figure 21.

Figure 21: Estimated number of pile strikes to reach a cumulative sound exposure level of 187.5 dB re 1µPa2.s. Thus for a longer bout of hammering (30 minutes at 3 s hammer rate) we can expect stationary fish within 200 m to be at risk from hearing damage. Using the calculated cumulative exposures for fish swimming past at slack water perpendicular to the wharf (ie. Figure 19) we can expect some degree of temporary hearing impairment to occur for smaller fish passing within 300 m of the wharf assuming the pile driving continues over the pass-by (83 minutes for the full km swim but less for the critical period when maximum sound loading occurs). Larger fish swimming faster will need to be closer to the pile driving to experience the same cumulative sound loading. For example 250 mm long fish would need to be within 100-150 m to experience cumulative sound exposures sufficient to cause temporary hearing impairment. If the fish swim by assisted with a strong current then these ranges will drop to that for a single pile strike equivalent to 187.5 dB re 1µPa2.s, of 60 m, if the current assisted swim speed is fast enough. For marine mammals the current TTS onset criteria (which are only vaguely defined and in considerable dispute) are for temporary threshold shifts to occur from impulsive signals at mean squared pressure levels of 180 dB re 1µPa or above. From Figure 20 this would occur at < 50 m from a single pile strike. Applying a two-times safety margin we can assume a 100 m range for TTS onset for marine mammals. Since marine mammals are highly mobile and unlikely to linger near pile driving then calculations of the number of pile strikes required to reach an equivalent sound loading have not been made. 6.4 Behavioural responses Behavioural responses to pile driving will occur at much greater ranges than for the hearing impairment defined above. The pile driving signals are likely to be above natural ambient levels for perhaps several km up or downstream for an animal in the channel with an unobstructed deep water path to the pile being driven. Thus any animal moving up or down the channel with a reasonable hearing capability will be aware of pile driving activities well before it becomes a threat to them, assuming the pile driving does not commence while the animal is in close proximity.


33

For fish McCauley et al (2003a) determined that caged fish began to show well defined behavioural responses to oncoming air gun impulses at 145 dB re 1µPa2.s and that by the time the signal level reached 150 dB re 1µPa2.s the fish had huddled in the cage center. Thus we can expect that pile driving will create strong behavioural responses all across the river and perhaps out to 500 m up or downstream, based on the estimated depth averaged SEL curve shown on Figure 17, which had decayed to just above 150 dB re 1µPa2.s at 500 m. For marine mammals and seals there may be some initial curiosity to pile driving noise. The fact that the noise source is stationary and so does not imply any collision threat to them and its poor overlap with their hearing threshold curves suggests that behavioural responses will be minimal, other than potentially keeping a comfortable distance (perhaps several hundred m) off. 6.5 Impacts at nearby aquaculture farm The estimated pile driving levels received from a single pile strike were in the range 143-148 dB re 1µPa2.s depending on where the cage was sited within the aquaculture lease. For a single strike these received levels are below those required to cause any physiological damage but are at the lower bounds of levels at which behavioural responses have been observed from caged fish (McCauley et al 2003a). For the scenario of three hours of continual impact pile driving at a 50% duty cycle and four s strike rate, the cumulative sound exposure (CSEL) received at the aquaculture cages would be of the order of 174-179 dB re 1µPa2.s, below the level required for known hearing loss damage to occur and well below that at which serious physical effects occur from a single pile strike. For 30 minutes of continual piling at a three s repetition rate the CSEL received would be in the range 171 – 176 dB re 1µPa2.s again below the level required for any physiological harm. Thus if the source pile driving signal used in modeling is comparable with that generated during piling, then continual pile driving at the proposed wharf will be unlikely to cause physical injury to any caged fish, but it may cause some behavioural responses. The fact that the pile driving source will be stationary will be a mitigating factor in any behavioural response as it will imply a distant threat. 6.6 Summary • For a single pile strike most fish will need to be within perhaps 10 m from the pile to suffer

any serious physiological trauma resulting in injury • For a worst scenarios of continual pile hammering (three hours of continual hammering at a 4 s

hammer rate), stationary fish within 120 m of the pile will experience a cumulative sound loading equivalent to that from a single strike believed to cause serious physical trauma. Thus fish within 120 m may suffer some physical injury if exposed to three hours of continual hammering.

• For a typical pile hammering scenario (30 minutes of continual hammering at a 3 s hammer rate), stationary fish within 80 m of the pile will experience a cumulative sound loading equivalent to that from a single strike believed to cause serious physical trauma. Thus stationary fish within 80 m of piling may suffer some physical injury if exposed to thirty minutes of continual hammering.

• For fish swimming steadily past the wharf during pile hammering episodes at slack water, then cumulative sound loadings equivalent to that at which a single strike is believed to cause serious physical trauma, may occur for fish passing within 50 m of the wharf. Thus fish steadily swimming up or downstream at slack water within 50 m of piling may suffer some physical injury if exposed to continual hammering. The range for the equivalent sound loading


34

will decrease for current assisted fish or fish larger than 250 mm (the maximum size at which cumulative sound loadings were calculated).

• It is unlikely that dolphins or seals will be exposed to pile driving sound intensities sufficient to cause any serious physiological damage

• Fish within 60 m of a single pile strike may be exposed to a peak-peak intensity sufficient to cause temporary hearing loss.

• Stationary fish within 200 m of continual piling may be exposed to a cumulative energy loading sufficient to cause temporary hearing loss.

• Fish swimming steadily past the wharf during pile hammering episodes at slack water within 100-300 m closest range to the wharf may receive sufficient sound loadings to cause temporary hearing loss. The degree or presence of any temporary threshold shifts will depend on the fish species, swim speed and proximity to the wharf.

• Marine mammals within 50 m of pile driving may receive a sufficient sound intensity to cause temporary hearing loss. Using an uncertainty margin of two we can assume a 100 m range for TTS onset.

• Pile driving events will be audible to animals with reasonable hearing capability in the river for several km up and downstream of the wharf construction.

• Marine mammals are unlikely to show overt responses to continual pile driving if they are more than a few hundred m off, and may show some initial curiosity.

• Fish are likely to show strong behavioural responses to pile driving noise all the way across the river and for at least 500 m up or downstream. Such responses could include avoidance or huddling near the seabed.

• Caged fish in a nearby aquaculture farm opposite the proposed wharf are unlikely to be exposed to single or cumulative sound loadings sufficient to cause any physical trauma or hearing damage. But the sound exposures received at the aquaculture farm will be within the bounds at which observable behavioural responses have been detected in caged fish.

7 Recommended mitigation measures It is important to note that the findings here are indicative only, and the conclusions have been made based on the information available including an estimated source signature and sound transmission environment. There is a paucity of data on hearing sensitivity and impacts of noise on the physiology and behaviour of marine fauna and of pile driving measurements. The greatest potential impact from underwater noise during the BBPMP is likely to be from the continuous and long periods of impact pile driving (nominally set as longer than 30 minutes of continuous hammering, or 600 strikes assuming a 3 s hammer rate). Shorter periods of hammering will present less risk. The works should be managed appropriately (i.e. timed around sensitive periods for the most at-risk species, and other appropriate mitigation measures put in place). To minimize any impacts, the following mitigation procedures are recommended:


35

• An assessment of marine fauna that are economically or ecologically significant should be

made to identify the timing of occurrence and critical periods (such as spawning, and migration near the wharf area) so that pile driving works can be scheduled to not coincide with these critical periods.

• As a general guide, dredging and construction including pile driving should be avoided or suspended during sensitive periods for listed, protected or economically important species when a significant proportion of the population occurs or must pass through the Tamar River for critical spawning, or larval dispersal (and cannot do so due to noise construction).

• Underwater noise produced by pile driving should be recorded during wharf construction for monitoring purposes, and these guidelines re-assessed as necessary.

• If it is believed that marine mammals will frequent the area during operations then operations should be monitored by dedicated observers (working from land based platforms) to determine if animals are within 100 m of the pile driving (a safety factor of 2 for TTS)

• It is recommended that work should be suspended in the presence of marine (or endangered) mammals, which enter within a range of 100 m of pile driving operations

• Some form of underwater speaker is deployed before impact pile driving commences to broadcast a noxious noise signal (such as an alarm type signal) in an attempt to forewarn animals of impending pile driving. The type and range of signals are to be defined by an appropriate underwater acoustician.

• Because of the potential for behavioural impacts on caged fish in the nearby aquaculture farm then: 1) sound levels experienced at the nearby aquaculture farm are measured from impact pile driving; and 2) the growth and behaviour of fish held at the aquaculture farm are monitored well before, during and after wharf construction to ascertain if fish growth has been retarded or notable behavioural shifts occur during pile driving operations.

8 References Aplin, J.A. 1947. The effect of explosives on marine life. Calif. Fish and Game 33:23-30. Anon (2008) Bell Bay Pulp Mill Project, Wharf construction and existing bed sediment

contamination. EnviroGulf Consulting, Dubai, CR 001/4 prepared for Gunns. Baxter, L., E.E. Hayes, G.R. Hampson, and R.H. Backus. 1982. Mortality of fish subjected to

explosive shock as applied to oil well severance on George’s Bank. Woods Hole Oceanographic Institute, Woods Hole, Massachusetts 02543.

Bergert, B. A., and Wainwright, P. C. 1997. Morphology and kinematics of prey capture in the Syngnathid fishes Hippocampus erectus and Syngnathus floridae. Marine Biology. 127:563-570.

Blaxter, J.H.S. (1969). Swimming speeds of fish. FAO Fisheries Report 62(6):69-100 Blaxter, J.H.S. (1980). Fish Hearing. In: 'Oceanus, Senses of the Sea', 23(3), 27-33. Woods Hole. Bowden, A.R., Lane, M.R. and Martin, J.H., 2001, Triple Bottom Line Risk Management –

Enhancing Profit, Environmental Performance and Community Benefit, Wiley, New York.). Brill, R.L., Moore, P.W.B. & Dankiewicz, L.A. (2001). Assessment of dolphin (Tursiops sp.)

auditory sensitivity and hearing loss using jawphones. JASA, 109(4), 1717-1722. Bryant, P.J., Lafferty, C.M. and Lafferty, S.K. 1984. Reoccupation of Laguna Guerrero Negro, Baja

California, Mexico, by gray whales. P. 375-387. In: M.L. Jones, S.L. Swartz and S. Leatherwood (eds.), The gray whale Eschrichtius robustus. Academic Press, Orlando, FL. 600 Pp.

Chapman, C.J. & Sand, O. (1974). Field studies of hearing in two species of flatfish Pleuronectes Platessa (L.) and Limanda limanda (L.) (family Pleuronectidae). Comp. Biochem. Physiol., 47A, 371-385.


36

Coker, C.M., and E,H. Hollis. 1950. Fish mortality caused by a series of heavy explosions in Chesapeake Bay. J. Wildlife Management 14:435-444.

Colson, D. J., Patek, S. N., Brainerd, E. L., and Lewis, S. M. 1998. Sound production during feeding in Hippocampus seahorses (Syngnathidae). Environmental Biology of Fishes, 51: 221-229.

Coombs, S. & Popper, A.N. (1982). Structure and function of the auditory system in the clown knifefish, Notopterus chitala. J. Exp. Biol., 97:225-239.

Coutin, P. 2000. Port Phillip Bay fin fisheries - 1998. Marine and Freshwater Resources Institute: Queenscliff

Crum, L. A. and Mao, Y. 1996. Acoustically enhanced bubble growth at low frequencies and its implications for human diver and marine mammal safety. J. Acoustical Soc. Am. 99(5):2898-2907.

David, J. 2006. Channel Deepening Project Description. Supplementary Environmental Effects Statement. Maunsell Australia Pty Ltd. Pp. 41.

Dann, P., Cullen, M., and Weir, I. 1996. National Review of the Conservation Status and Management of Australian little penguin Colonies. Final Report for the Australian Nature Conservation Agency.

Donskoy, D.M., Ludyanskiy, M., and Wright, D.A. 1996. Effects of sound and ultrasound on Zebra Mussels. J. Acoust. Soc. Amer. 99(4):

Dunn, W., Goldsworthy, A., Glencross, D., & Charlton, K. 2001. Interactions between bottlenose dolphins and tour vessels in Port Phillip Bay, Victoria. Report to the Department of Natural Resources and Environment, Victoria, for the Sustainable Dolphin Tourism Program.

Enger, P. (1967). Hearing in herring. Comp. Biochem. Physiol., 22:527-538. ECC. 2000. Marine Coastal and Estuarine Investigation: final report. Environment Conservation

Council, ISBN: 0-646-39971-3. Fay, R.R. & Popper, A.N. 1975. Modes of stimulation of the teleost ear. J. Exp. Biol., 62, 370-387. Feist, B.E., J.J. Anderson, and R. Miyamoto. 1992. Potential impacts of pile driving on juvenile

pink (Oncorhynchus gorbuscha) and chum (O. keta) salmon behavior and distribution. FRI-UW-9603. Fisheries Resources Institute, University of Washington. Seattle.

Frings, H. 1964. ‘Problems and Prospects in Research on Marine Invertebrate Sound Production and Reception’ in: “Marine Bio-Acoustics” (Ed. Tavolga, W.N.).

Fry, D.H., and K.W. Cox. 1953. Observations on the effect of black powder explosions on fish life. Calif. Fish and Game 39(2):233-236

Fulton, W., Jenkins, G.P., and Walker, T.I. 2006. Literature Review of protected and listed marine fauna of Port Phillip Bay and identification of potential impacts of channel deepening. Department of Primary Industries Internal Report No 70.

Govoni, J.J., L.R. Settle, and M.A. West. 2003. Trauma to juvenile pinfish and spot inflicted by submarine detonations. J. Aquatic Animal Health 15:111-119.

Hall, J.D. and Johnson, C.S. (1972). Auditory thresholds of a killer whale Orcinus orca Linnaeus. JASA, 51(2), 515-517.

Hale, P. 2002. Interactions between vessels and dolphins in Port Phillip Bay. Final report. Department of Natural Resources and Environment.

Harris, M.P. & Norman, F.I. (1981) Distribution and status of coastal colonies of seabirds in Victoria. Memoirs of the National Museum of Victoria, 42, 89-106.

Hawkins, A.D. 1986. Underwater Sound and Fish Behaviour. In: 'The Behaviour of Teleost Fishes'. T.J. Pitcher (ed), 114-151. Croom Helm Ltd, Beckenham.

Harris, M.P. & Norman, F.I. (1981) Distribution and status of coastal colonies of seabirds in Victoria. Memoirs of the National Museum of Victoria, 42, 89-106.


37

Hastings, M.C., A.N. Popper, J.R. Finneran, and P.L. Lanford. 1996. Effects of underwater sound on hair cells of the inner ear and lateral line of the oscar (Astronotus ocellatus). J. Acoustical Soc. Am. 99(4):2576-2603.

Hastings, M. C. and Popper, A.N.. 2005. Effects of sound on fish. Subconsultants to Jones & Stokes Under California Department of Transportation Contract No. 43A0139. Report. Pp 82.

Hubbs, C.L., and A.B. Rechnitzer. 1952. Report on experiments designed to determine effects of underwater explosions on fish life. Calif. Fish Game 38: 333-365.

Jenkins, G.P., Plummer, A.J. and MacKinnon, L. 2006. Channel Deepening Supplementary Environmental Effects Statement – Aquaculture and Fisheries. Department of Primary Industries Report No. ?

Johnson, C.S. (1967). Sound detection thresholds in marine mammals. In W.N. Tavolga (ed), Marine bio-acoustics, vol. 2. Pergamon, Oxford, U.K.

Kearns, R. K., and F. C. Boyd. 1965. The effect of marine seismic exploration on fish populations in British Columbia coastal waters. Canadian Fish Culturist 34:3-26.

Koschinski. S., B. M. Culik., O. D. Henriksen., N. Tregenza., G. Ellis, C. Jansen and G. Kathe. 2003. Behavioural reactions of free ranging porpoises and seals to the noise of a simulated 2MW windpower generator. Marine Ecology-Progress Series 265: 263-273.

Ljungblad, D.K., Scoggins, P.D. & Gilmartin, W.G. (1982). Auditory thresholds of a captive Eastern Pacific bottle-nosed dolphin, Tursiops spp. JASA 72(6):1726-1729.

Lovell, J.M., Findlay, M.M., Moate, R.M., Yan, H.Y. (2005) The hearing abilities of the prawn Palaemon serratus. Comp. Biochem. Physiol. A140:89-100

Malme, , C.I., Miles, P.R., Clark, C.W. Tyack, P. and Bird, J.E. 1984. Investigation of the potential effects of underwater noise from petroleum industry activities on migrating gray whale behaviour/Phase II: January 1984 migration. BBN Minerals Manage. Serv., Anchorage, AK. NTIS. PB86-218377.

Mann, D.A., Lu, Z. & Popper, A.N. (1997). A clupeid fish can detect ultrasound. Nature, 48:341. McCartney BS, Stubbs AR (1971) Measurements of the acoustic target strengths of ®sh in dorsal

aspect including swimbladder resonance. J Sound Vib 15: 397-420 McCauley, R.D., Maggi, A.; Perry, M. and Siwabessey. 2002. Analysis of Underwater Noise

Produced by Pile Driving, Twofolds Bay NSW – Phase III,. Prepared for Baulderstone Hornibrook Pty. Ltd. from the Centre for Marine Science and Technology, Curtin University. CMST R2002-27.

McCauley, R.D., Fewtrell, J., Duncan, A.J., Jenner, C., Jenner, M-N., Penrose, J.D., Prince, R.I.T., Adhitya, A., Murdoch, J., McCabe, K. (2003a). Marine seismic surveys: analysis and propagation of air-gun signals; and effects of exposure on humpback whales, sea turtles, fishes and squid. In (Anon) Environmental implications of offshore oil and gas development in Australia: further research. Australian Petroleum Production Exploration Association, Canberra, pp 364-521

McCauley, R.D., Fewtrell, J., Popper, A.N., (2003b) High intensity anthropogenic sound damages fish ears. J. Acoust. Soc. Am. 113(1):638-642

McCauley, R.D. (2006) Ambient sea noise, estimation of hydro-hammer, pile driving and vessel noise transmission in the entrance and south channel, and pile driving measures in the Yarra River, Port Phillip Bay, Victoria, CMST Curtin University Report R2006-22. 43 pp. 33 Fig.

Myrberg, A.A. Jr. 1978. ‘Underwater Sound – Its effects on the Behavior of Sharks’ in “Sensory Biology of Sharks, Skates and Rays” (Eds: Edward S. Hodgson, Robert R. Matherson). Office of Naval Research.

Myrberg, A.A. Jr & Spires, J.Y. 1980. Hearing in damselfishes: an analysis of signal detection among closely related species. J. Comp. Physiol., 140, 135-144.


38

Moore, P.W.B. & Schusterman, R.J. 1987. Audiometric assessment of northern fur seals, Callorhinus ursinus. Marine Mammal Science, 3(1), 31-53.

Mustoe, S. 2006. Supplementary Environment Effects Statement Effects Statement. X12 Underwater Noise- Marine Biota.

Mustoe, S. 2006. Marine Mammals and Penguins Report. Port of Melbourne Channel Deepening Project Report No. ?

Nachtigall. P. E., J. L. Pawloski and W. W. L. Au. 2003. Temporary threshold shifts and recovery following noise exposure in Atlantic bottlenosed dolphins (Tursiops truncatus). Journal of

the Acoustical Society of America 113: 3425-3429. Nedwell, J.R.; Edwards, B., Turnpenny, A.W.H., and Gordon, J. 2004. Fish and Marine Mammal

Audiograms: A summary of available information. Subacoustech Report No: 534R0214. NMFS and NOAA. 2005. Endangered fish and wildlife; notice of intent to prepare an

environmental impact statement. Federal Register, 70(7): 1871-1875. Offutt, G.C. 1974. Structures for the detection of acoustic stimuli in the Atlantic codfish, Gadus

morhua. JASA, 56(2), 665-671. Petkovic, P. & Buchanan, C. 2002. Australian bathymetry and topography grid. [Digital Dataset].

Canberra: Geoscience Australia. http://www.ga.gov.au/general/technotes/20011023_32.jsp (Jan 2005).

Popov, V. & Supin, A. 1990. Electrophysiological studies of hearing in some cetaceans and a manatee. In ‘Sensory Abilities of Cetaceans’, 405-415. J. Thomas & R. Kastelein (eds). Plenum Press, N.Y.

Popper, A.N. 1972. Pure-tone auditory thresholds for the carp, Cyprinus carpio. JASA, 52(6) Part 2, 1714-1717.

Popper, A.N., and R.R Fay. 1973. Sound detection and processing by teleost fishes: critical review. J. Acoustical Soc. Am. 53(6):1515-1529.

Popper, A.N. & Fay, R.R. 1993. Sound detection and processing by fish: Critical review and major research questions. Brain, Behav., Evol., 41:14-38.

Popper, A. N., and Fay, R. R. 1999. "The auditory periphery in fishes." In Comparative Hearing: Fish and Amphibians, edited by R. R. Fay and A. N. Popper. Springer-Verlag, New York, pp. 43-100

Popper, A.N., M.E. Smith, P.A. Cott, B.W. Hanna, A.O. MacGillivray, M.E. Austin, and D.A. Mann. 2005. Effects of exposure to seismic airgun use on hearing of three fish species. J. Acoustical Soc. Am. 117(6): 3958-3971.

Popper, Arthur N., Thomas J. Carlson, Brandon L. Southall, and Roger L. Gentry. 2006. Interim Criteria for Injury of Fish Exposed to Pile Driving Operations: A White Paper.

Richardson, W.J., Wursig, B. and Greene, C.R. Jr. 1990. Reactions of bowhead whales, Balaena mysticetus, to drilling and dredging noise in the Canadian Beaufort Sea. Mar. Environ. Res. 29(2): 135-160.

Richardson, W. J., Jr. C.R. Green., R. Malme and C. I. Thomson. 1995. Marine mammals and noise (Academic, San Diego).

Ridgway, S. H., D. A. Carder., R. R. Smith., T. Kamolnick., C. E. Schlundt and W. R. Elsberry. 1997. Behavioural responses and temporary shift in masked hearing threshold of bottlenose dolphins Tursiops truncatus, to 1 second tones of 141 to 201 dB re 1µPa. Technical Report Number 1751, Naval Command Control and Ocean Surveillance Center, RDT&E Division, San Diego California.

Ripley, J.L. 2006. Effects of Environmental Factors on the Paternal Brood Pouch and Sound Production in Two Sympatric Pipefish Species from the Chincoteague Bay, Virginia. PhD Thesis. West Virginia University, Morgantown, West Virginia.


39

Salmon, M., Horch, K. and Hyatt, G.W. 1977. Barth's myochordotonal organ as a receptor for auditory and vibrational stimuli in fiddler crabs Uca pugilator and U. minax. Marine Behav. Physiol. 4:187-194.

Sand O, Enger PS (1973) Evidence for an auditory function of the swim bladder in the cod. J Exp Biol 59: 405±414

Schlundt, C. E., J. J. Finneran, D. A. Carder and S. H. Ridgway. 2000. Temporary shift in masked hearing thresholds of bottlenose dolphins, Tursiops truncatus and white whale Delphinapterus leucas, after exposureto intense tones. Journal of the Acoustical Society of America 107: 3496-3508.

Scholik, A.R. & Yan, H.Y. (2001). Effects of underwater noise on auditory sensitivity of a cyprinid fish. Hearing Research, 152:17-24.

Scholik, A.R. & Yan, H.Y. (2002). The effects of noise on the auditory sensitivity of the bluegill sunfish, Lepomis macrochirus. Comp Biochem Physiol A, 133:43-52.

Shallenberger, E.E. 1978. Activities possibly affecting the welfare of humpback whales. P. 81-85. In: K.S. Norris and R.R. Reeves (eds.), Report on a workshop on problems related to humpback whales (Megaptera novaeangliae) in Hawaii. MMC-77/03. Rep. From ?Sea Life Inc., Makapuu Pt., HI, for U.S. Mar. Mamm. Comm., Washington, DC. 90 p. NTIS PB-280794.

Sinclair Knight Merz. 2004. Channel Deepening Project Environment Effects Statement: Tourism and Recreation Study (Initial and Residual Risk Assessment Report).

Szymanski, M.D., Bain, D.E., Kiehl, K, Pennington, S., Wong, S. & Henry, K.R. (1999). Killer whale (Orcinus orca) hearing: Auditory brainstorm response and behavioral audiograms. JASA, 106(2): 1134-1141.

Tavolga, W.N. (1974). Sensory parameters in communication among coral reef fishes. Mt. Sinai J. Med., 41, 324-340.

Tavolga, W.N. & Wodinsky, J. (1963). Auditory capacities in fishes. Bull. Am. Mus. Nat. Hist., 126, 177-240.

Trasky, L. 1976. Environmental impact of seismic exploration and blasting in the aquatic environment. Alaska Dept. Fish and Game.

Vagle, S. 2006. On the impacts of pile-driving noise on marine life. Ocean Science and Productivity Division Institute of Ocean Sciences DFO/Pacific. Report. Pp 41.

Walker, T.I. 2002. Victorian shark catch. In 'Australian Shark Assessment Report for the Australian National Plan of Action for the Conservation and Management of Sharks'.

Ward, W.D. (ed.). 1968, Orioised danage-risk criterion for impulse noise (gun-fire). Committee on Hearing, Bioacoustics, and Biomechanics, Natl. Res. Counc. Natl. Acad. Sci., Washington, DC. 8 p.

Wever, E.G., Herman, P.N., Simmons, J.A., and Hertzler, D.R. 1969. Hearing in the Blackfooted Penguin, Spheniscus demersus, as represented by the cochlear potentials.Psychology. Vol. 63: 676:680.

Yan, H.Y., Fine, M.L., Horn, N.S., and Colón, W.E. 2000. Variability in the role of the gasbladder in fish audition. J. Comp. Physiol. A. 186: 435-445.

Yelverton, J.T., D.R. Richmond, W. Hicks, K. Saunders, and R. Fletcher. 1975. The relationship between fish size and their response to underwater blast. Topical Report DNA 3677T. Defense Nuclear Agency, Department of Defense, Washington, D.C.

Date post:	19-Aug-2020
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

PILE DRIVING UNDERWATER NOISE ASSESSMENT...

Documents