Damselfish in Distress:
an exploration of context-dependent stress responses
of coral reef fish exposed to anthropogenic noise
Submitted by Emily Francesca Armstrong-Smith to the University of Exeter as a thesis for the degree of Masters by Research in Biological Sciences, May 2016.
This thesis is available for Library use on the understanding that it is copyright material and that no quotation from the thesis may be published without proper acknowledgement.
I certify that all material in this thesis which is not my own work has been identified and that no material has previously been submitted and approved for the award of a degree by this or any other University.
(Signature) ......
Abstract
This research examines the stress response of two coral reef fish species subject to the prevalent stressor anthropogenic noise, and examines how that response may change in different contexts. Literature on anthropogenic impacts has focused almost exclusively on the response of organisms to single stressors, however modern marine organisms face multiple stressors simultaneously; this research advances our understanding of this multiple stressor interaction. These studies tested two primary hypotheses: 1, there will be a significant difference in the stress response of fish to anthropogenic noise between field and laboratory studies; 2, fish subjected to limited feeding will exhibit a heightened stress response in anthropogenic noise conditions but not in ambient conditions. Using two important coral reef fish species (Chromis viridis and Acanthrochromis polyacanthus), experiments were conducted in the field and the laboratory and employed multiple methods to test physiological functions indicative of a stress response to anthropogenic noise. The findings from the research show that, while qualitatively similar, the stress response can be overstated when considering results from laboratory trials, compared with field trials. This suggests that the impact of experimental situation on fish physiological stress responses is more complex than previously thought, and highlights the importance of accurate validation of results before extrapolation.
This research also found fish subject to suboptimal dietary conditions are more likely to exhibit a heightened stress response when exposed to anthropogenic noise. A pronounced and/or prolonged stress response can be detrimental to organisms, both in terms of physiology and behaviour. This illustrates how vulnerable organisms in today’s oceans are even more at risk from impacts such as anthropogenic noise. Overall findings from this research offer insights
2 into the multi-stressor environment marine life currently face, and provide solid evidence for the inclusion of anthropogenic noise – a comparatively easy stressor to regulate and mitigate – in environmental management plans.
3 Acknowledgements
I would like to thank my supervisor, Steve Simpson, for his continual support and encouragement throughout the entirety of this project. Thanks to Charles
Tyler and Andy Radford for their support and advice, both remotely during fieldwork and directly throughout my writing up. Immense thanks go to Mark
McCormick for the Lizard Island opportunity; it was a truly amazing experience and would not have been possible without his support and advice. To Mark
Meekan, for invaluable academic support and advice, and for driving the boat for hours on end - it was all worth it! To both Mark Meekan and Steve, thank you for the unrelenting comedic relief on the boat and for teaching me to fish - thanks also to Mrs C and Roger Walker for their contributions, you all made field work an absolute blast. Immense thanks to Emily Walker for assistance and support during field trials, and many thanks to Ryan R, Rachael W, and Bridie G for valuable discussions. Thanks to the team at LIRS for their support and assistance, and to Jan Shears and the ARC team at Exeter for their help during my lab work in the UK. Thanks to CO for support during writing up! The biggest thanks to Lauren Fear – for support, assistance, advice and encouragement, for always being there to enjoy the good times and more for being there to help when things didn’t quite go to plan! I couldn’t have asked for a better dive buddy and travel companion. Finally, to Linsey Armstrong-Smith – for unwavering support and for constantly encouraging me to ‘go for it’ – thank you.
4 Table of Contents
ABSTRACT 2
ACKNOWLEDGEMENTS 4
TABLE OF CONTENTS 5
LIST OF TABLES AND FIGURES 6
DEFINITIONS AND ABBREVIATIONS 8
1. INTRODUCTION 10
2. LITERATURE REVIEW 13
2.1 ANTHROPOGENIC IMPACTS 13
2.2 ANTHROPOGENIC NOISE IN THE MARINE ENVIRONMENT 22
2.3 LIVING WITH STRESSORS 33
3. STUDY AIMS AND OBJECTIVES 43
4. CONTEXT-DEPENDENT RESPONSES TO ANTHROPOGENIC NOISE 45
4.1 INTRODUCTION 45
4.2 MATERIALS AND METHODS 50
4.3 RESULTS 57
4.4 DISCUSSION 59
5. CONDITION-DEPENDENT RESPONSES TO ANTHROPOGENIC NOISE 65
5.1 INTRODUCTION 65
5.2 MATERIALS AND METHODS 71
5.3 RESULTS 76
5.4 DISCUSSION 77
6. CONCLUSION 85
7. APPENDICES 89
8. BIBLIOGRAPHY 96
5 List of tables and figures
Table 1: Sample sizes for Chromis viridis and Acanthochromis polyacanthus for fish in test and control conditions in different experimental situations.
Figure 1: Frequency range of the most common anthropogenic noises and their overlap with the hearing ranges of important marine taxa. Taken from
Slabbekoorn, et al., 2010.
Figure 2: A simplified depiction of the fish stress response. The three primary categories stressors can be sorted into, and the three levels of stress responses that fish experience, with examples and potential interactions. Reproduced from
Barton, 2002.
Figure 3: Spectral content shown in acoustic pressure and particle acceleration domains for original field, field playback, and lab playback recordings of boat noise and ambient conditions.
Figure 4: Spectrograms for original field, field playback and lab playback recordings of boat noise and ambient conditions.
Figure 5a-d: Mean opercular beat rate for difference in two minutes pre- exposure/two minutes post-exposure and opercular beat gradient change over first five minutes post-exposure for Chromis viridis and Acanthochromis polyacanthus, in test and control conditions in different experimental situations.
Figure 6a-b: Metabolic rate for Chromis viridis and Acanthochromis polyacanthus, in test and control conditions in different experimental situations.
Figure 7: Mean opercular beat rate over two minutes post exposure for
Acanthochromis polyacanthus in test and control conditions in different dietary groups.
6 Figure 8: Mean opercular beat rate at different points during exposure period for Acanthochromis polyacanthus in test and control conditions in different dietary groups.
Figure 9: Opercular beat rate gradient change over first five minutes post- exposure for Acanthochromis polyacanthus in test and control conditions in different dietary groups.
Figure 10: Metabolic rate for Acanthochromis polyacanthus in test and control conditions in different dietary groups.
7 Definitions and abbreviations
Acoustic: relating to sound or the sense of hearing
Acoustic impedance: opposition to the flow of sound through a surface or medium
Anthropogenic: caused or produced by humans, originating of human activity
Attenuation: a measure of the energy loss of sound propagation in the media
Bioacoustics: the branch of acoustics concerned with sounds produced by or affecting living organisms, especially as relating to communication
Cilia: minute hair-like organelles that line the surfaces of certain cells to aid in locomotion (beating in rhythmic waves) or detection of vibration (due to enervation by received sound)
Corticosteroid: any of the steroid hormones produced by the adrenal gland; produced in reaction to stress and raising metabolic function dBa: Abbreviation of A-weighted decibels; an expression of the relative loudness of sounds in air as perceived by the human ear
Epinephrine (or adrenaline): a hormone that increases rate of blood circulation, breathing, and carbohydrate metabolism, prepares muscles for exertion
Frequency: the number of cycles of a repetitive waveform per second. A low frequency sound may sound like a low rumble, a high frequency sound could be a shrill whistle or squeak
Habituation: the diminishing of an innate response to a frequently repeated stimulus
Norepinephrine (another term for noradrenaline): a hormone that acts to constrict blood vessels and dilate bronchi, and also acts as a central and peripheral neurotransmitter
8 Otolith: the small calcareous bodies in the inner ear of vertebrates, involved in sensing gravity and movement
Soundscape: the auditory environment that surrounds an organism, the component sounds of an environment
Stress: the response of an organism to any demand placed on it such that it causes an extension of a physiological state beyond the normal resting state
Stressor: a chemical or biological agent, environmental condition, external stimulus or an event that causes stress to an organism
Temporary Threshold Shift: a temporary shift in the auditory threshold, resulting in temporary hearing loss. May occur after exposure to a high level of noise
ARC: Aquatic Resources Centre
FBA: Field boat trials, in Australia
FPA: Field playback trials, in Australia
GBR: Great Barrier Reef
GBRMPA: Great Barrier Reef Marine Park Authority
HPA: hypothalamic pituitary axis
LIRS: Lizard Island Research Station
LPA: Laboratory playback trials, in LIRS, Australia
LPE: Laboratory playback trials, in Exeter, UK
MR: metabolic rate
OBR: opercular beat rate
SNS: sensory nervous system
SPL: sound pressure level
UK: United Kingdom
UoE: University of Exeter
9 1. Introduction
In 1956, French explorer Jacques Cousteau invited the public to experience the serenity then associated with life under the sea in his documentary, ‘The Silent
World’. This world portrayed on screen, however, is a very different picture from the oceans we know and are coming to understand today: the silent splendor that Cousteau showed us is now, in reality, little more than a idealistic notion.
The oceans are struggling, with human activity putting a growing strain on the ocean and it’s resources, forcing marine life to adapt rapidly in order to survive
(Pandolfi, 2015; Poloczanska, et al., 2013).
Anthropogenic impacts are far reaching and can be devastating (Halpern, et al.,
2008). In this thesis the impacts of one such pollutant will be considered, anthropogenic noise, and it’s interactions with other common stressors marine organisms currently face. Anthropogenic noise pertaining from (predominantly) industrial shipping, motorboats and offshore construction projects is a global threat and one that impedes on marine organism’s life histories, both short- and long-term (Hildebrand, 2009; Wright, et al., 2007). To successfully study these impacts, and consider potential future threats anthropogenic noise may cause, our current knowledge of this threat will first be reviewed in Chapter 2, considering the growth of human activity following the industrial revolution and the subsequent effects on wildlife, both terrestrial and aquatic. In considering the many impacts of anthropogenic noise the issue of stress will be studied more deeply, since stress can be both physically and physiologically detrimental to individuals, with consequences for populations. By understanding the stress mechanism, and in turn the range of functions stress can affect, allows us to deepen our understanding of how harmful anthropogenic noise really can be.
10 Building on my review of the literature, experiments were then devised and conducted to address evident knowledge gaps the field of research of anthropogenic noise impacts. In combining study of the effects of anthropogenic noise with other relevant factors (in this case nutrition), this work is hopefully able to go some way to filling these knowledge gaps and broadening our understanding. It currently seems that anthropogenic impacts are largely unavoidable, however the majority of research thus far has been one- dimensional by only considering single stressors in isolation; perhaps we need to first understand how these impacts may interact before we can take meaningful steps to mitigate them.
Deliberating over this idea led first to calling into question the applicability of some methodologies used in research into impacts of environmental stressors, and so this work begins by studying the validity of using laboratory experiments to explain and predict phenomena in the real world (Chapter 3). As the oceans are changing rapidly, scientists have to try to predict what will happen if human activity continues along the current trajectory. However the question often ignored is whether the extrapolation of data from laboratory experiments is a reliable method considering the marked differences between controlled and confined experimental conditions compared to conditions in the field.
Moving forward, this work then goes on to consider the impacts of multiple stressors (in this case, nutrition) on fish health and well-being (Chapter 4). A well-known anthropogenic impact (noise) was combined with the lack of available food resources (poor diet), a very real threat for marine organisms today. It is important to consider how stressors may interact and impact upon one another, as marine life face a multitude of these stressors on a daily basis
11 in the real world. Only with a solid understanding of the impacts and interactions can we try to mitigate detrimental effects.
To summarize, findings were considered in context with current knowledge within the field and addressed the knowledge areas that urgently need developing further. Our time now is perhaps best spent trying to understand the most harmful of anthropogenic impacts, and working towards easing pressure on the oceans from those most easy to mitigate, including anthropogenic noise.
12 2. Literature Review
2.1 Anthropogenic impacts
Since early human evolution, man has been interacting with and affecting flora and fauna, predominantly to his benefit. Modern-day industrialization and development in many economic sectors has led to a boom in anthropogenic activity, bringing with it some changes for good, but many to the detriment of ecosystems around the globe (Goudie, 2013). One such negative impact, and currently one most discussed, is climate change. Climate change is a clear example where human activity has altered the course of nature (Oreskes,
2004), and put into motion a chain of events that will no doubt continue to reverberate throughout ecosystems for centuries to come (Thomas, et al., 2004;
Karl & Trenberth, 2003). Some other key anthropogenic influences, the impacts of which we are only now beginning to truly understand, include marine debris, including plastics and chemicals (Gall & Thompson, 2015). These factors may not necessarily attract the same media coverage as climate change but they can (and some would argue, do) have significant adverse impacts in the environment (Leu, et al., 2008).
As our understanding of anthropogenic impacts on the environment expands, it is becoming more obvious how industrial development is impacting adversely on the natural world; exacerbating current issues and possibly fostering new ones (Dobson, et al., 1997). Despite advances in our scientific knowledge of our impacts on the environment in the last 50 years, human action to remediate for these effects has fallen woefully short of what is needed. Unless fundamental change occurs in human habits, researchers within the field predict further irreversible damage to the environment (Hallegatte, 2009). One emerging topic that is starting to gain attention because of its possible environmental impacts is
13 anthropogenic noise. Noise is now recognized as a pervasive pollutant in both terrestrial and aquatic systems. Economic boom has brought with it many industrial developments, and such growth has led to global expansions in transport and construction (Goudie, 2013) and these actions can permanently alter an acoustic setting of a habitat and impact a wide range of organisms within.
2.1.1 Bioacoustics – the physics of sound
Bioacoustics includes the study of sound production, propagation and reception in animals, including humans. Sound can be defined as the ‘Oscillation in pressure, stress, particle displacement, particle velocity, etc., propagated in a medium with internal forces (e.g., elastic or viscous), or the superposition of such propagated oscillation’ (Acoustical Society of America, 2013). Put more simply, sound waves are vibrations that propagate away from a noise source, causing a wave throughout a medium until they reach an auditory receptor – for example, ears. The speed of sound depends on the medium through which the waves are travelling (through air and water sound travels in longitudinal waves), and in a dry air medium sound travels at 343.2 m/s. This is nearly five times more slowly than in water, where the speed of sound is ~1500 m/s.
Understanding the movement of sound is imperative to understanding the impact of anthropogenic sounds on an environment, and to understanding how different organisms receive and perceive sound. In terrestrial systems, background noise levels are largely measured in decibels (dB re 20 µPa), providing a relatively easy way to quantify and compare different noise sources.
Much legislation (for example, mitigation regarding air traffic) is imposed through setting dB limits, as it is one of the easiest metrics to measure and control. Despite this benefit on a larger scale, at a more specific species level
14 the need for a different classification system arises, due to varying responses to different frequencies. The physical reception of sound in any hearing organism is limited by a frequency range – for humans, this is between 20 Hz to 20,000
Hz (20 kHz), with the upper limit lowering with age. Frequency ranges in hearing differ between animals, and subsequently anthropogenic noises do not affect all animals within an ecosystem equally (see Elert, 2015).
The detection of sound can be split into two domains: sound pressure and particle motion. Sound pressure is the difference in a given medium between average local pressure and pressure carried by a sound wave, and is measured in Pascals (Pa). Sound pressure level (SPL) is a logarithmic measure of pressure of a sound relative to a reference value, and is reported as decibels
(dB re. 1 µPa). The decibel scale allows for easier transition between different hearing ranges and sound levels, and is in general agreement with physiological and psychological study of sound (SCENIHR, 2008).
The second acoustic domain is particle motion: a modality of particular importance to marine species (see 2.2.1). Particle motion is the back and forth motion of particles making up the medium, and is described by displacement
(x), velocity (y) and acceleration (a) of the particles, in units of dB re. 1m s-2.
Understanding the different aspects of sound, and having the right tools to measure them effectively is the first vital step when considering anthropogenic noise impacts. Clarity on the hearing abilities and processes of different species, which I will consider in more detail further in, is also important as it allows us to understand which noises are most likely to affect different animals and why. Acoustic procedures such as SONAR have been proposed as non- invasive methods to assess marine biodiversity, but the true impacts of the noise are not fully understood. The field of bioacoustics, and in particular
15 underwater bioacoustics, is one that is developing rapidly, and it is important now to make sure we understand the implications of different acoustic actions to ensure we do not cause undue harm to the environment.
2.1.2 Sources of anthropogenic noise in the environment
It is hard to imagine the predominantly agricultural environment that existed before the industrial revolution. Noise pollution, now often considered on a par with air and water pollution in terms of detriment to human health and well being, was of little importance before the advent of accessible motor transport, commercial airline operation and wide-scale urban development (Goines &
Hagler, 2007). The overall considerations of anthropogenic noise impacts within terrestrial systems have a substantial geographical bias, with research focusing predominantly on North American and European ecosystems. This focus on temperate species has caused a gap in our knowledge of tropical species responses, and the bulk of time dedicated to studying nations already well developed could perhaps now be better spent looking at how less developed landscapes respond to the addition of industrial noise (Blickley & Patricelli,
2010).
The transportation sector is the most pervasive source of anthropogenic noise, with vehicle miles in the UK up almost 20% from 1994 to 2014 (Department for
Transport, 2015), and overall air transport movements increased 100-fold over the past six decades (Rutherford, 2011). Specifically within transport, road traffic noise varies with a number of factors, including speed, type of vehicle and number of vehicles (Washington State Department of Transportation,
2012). The majority of research into terrestrial noise has focused on roads and its traffic, and many of the research techniques and results can be applied to other anthropogenic noise sources due to the similar nature of road noise.
16 Although road surface areas are relatively small, their ecological effects can be far reaching and can also have serious effects on many temporal patterns within nature (Forman & Deblinger, 2001). Similarly, railway noise, though less pervasive than the road network, can still impact on surrounding wildlife and in terms of livestock, and can cause economic and social impacts (Oertli, 2013).
The other major source of anthropogenic noise originates largely from industrial and social development. Human development activity will always create noise, from early construction to completion and the daily operation of the site, be it urban development, production factories or resource extraction. The nature of many industrial sources, for example military bases, factories, mines and wind farms means the resulting noise is usually more localized than transport networks but nevertheless may pose a significant threat to the surrounding wildlife (Gerges & Sehrndt, 1995). A noise source that poses a dilemma is that of wind turbine farms, as on the one hand wind farms offer a plausible solution to the issue of depleting fossil fuels and reduction in CO2 emissions through adopting renewable energy, but on they other they have strong environmental impacts, particularly through noise production. Both the mechanical and aerodynamic noise is a low frequency sound and can act as a nuisance and/or a disturbance, alongside potentially disrupting activity routines (Rogers, et al.,
2002). Similar effects (often with further physiological impacts) come about from noise from military bases, which is predominantly low frequency and can be both chronic (such as engines, machinery) and acute (e.g. gunshots, explosions) (Larkin, et al., 1996). The extent to which industry noise can infiltrate the environmental soundscape is unlimited, and the predominantly low frequency, far-reaching sound from factories has the potential to seriously negatively affect animals and humans.
17 Although measures are put in place to mitigate noise levels and control noise exposure, the sound from human activities is ultimately changing the acoustic landscape and bringing with it impacts at both the individual and population level (Barber, et al., 2010). Furthermore, these human activities often differ from ambient noise sources, in terms of both acoustic frequencies and other characteristics such as intervals, timing and regularity (or lack thereof)
(Hildebrand J. A., 2009), and so can preset novel acoustic conditions with a range of impacts. The majority of anthropogenic noise sources are now irreversibly embedded into human lifestyles; in research we must now make sure we fully understand each noise source, and take the correct steps to mitigate the noise and reduce the environmental impacts of anthropogenic stressors.
2.1.3 Impacts of anthropogenic noise on terrestrial systems
From what has already been discussed, it is clear that animals face disruption from a large number of anthropogenic noise sources at potentially any hour of the day. Although very early studies considered noise impacts on human welfare, a more recent focus of study has been the impact on other animal species, including various bird species. Birds lend themselves to study both in field and laboratory settings, and much is already known about how birds use the acoustic soundscape and how they communicate using acoustic signals. As the field develops, researchers are now starting to consider noise impacts on mammals and invertebrates, and the larger implications of any more permanent effects of noise.
Anthropogenic noise, for example from road traffic, is at the very least an annoyance to all in it’s vicinity, and this can range from a mild disturbance to a physiological reaction typical of a stress response (Ouis, 2001). For humans,
18 noise from the transport sector often embeds itself into everyday life, and can impact sleep patterns and quality (Finegold, et al., 1994), leading to increased chance of cardiovascular irregularities (Griefahn, et al., 2008) and Ischemic heart disease (Babisch, 2014). In an industrial setting, chronic noise exposure can lead to elevated cortisol levels (Melamed & Bruhis, 1996), increased fatigue and irritability, anxiety and depression stemming from constant noise disruption
(Melamed, et al., 1992), and in more extreme cases hearing problems and hearing loss (Atmaca, et al., 2005) as a result of noise levels exceeding the recommended maximum dBA. If basic behavioural and physiological problems can persist in humans subject to unwanted background noise, it is no wonder anthropogenic noise impacts (unlimited by geographical bounds) also threaten other species in ecosystems around the world.
Anthropogenic noise (again, from transport in particular) has been found to impact bird species throughout their life histories. Birdsong, a communicative behaviour that depends upon audition, is impacted heavily by noise from the transport sector – peak traffic times cause a shift in timing of birdsong activity in
Sturnus unicolor and Passer domesticus, typically the dawn chorus (Arroyo-
Solis, et al., 2013), the minimum song frequencies of many species increase
(Dowling, et al., 2011) and birds sing at a higher sound level (Brumm, 2004) and a higher pitch (Potvin, et al., 2011) to try and overcome the anthropogenic background noise. Reproductive success is also severely affected, partially a result of the impedance to communication – female Parus major laid smaller clutches in noisier areas and there was a negative effect of noise on number of fledglings (Halfwerk, et al., 2011). One recent paper has found anthropogenic noise can affect telomere length in developing Passer domesticus, suggesting that early-life noise exposure could at a minimum have negative lasting impacts
19 on young (Meillere, et al., 2015). Finally, chronic anthropogenic noise (largely industrial sources) can affect various songbird population densities, both in forest (Bayne, et al., 2008) and urban (Blickley, et al., 2012) environments.
There is more scope for this research field to develop, with many questions regarding physiological effects and stress responses remaining unanswered.
However, the impacts clearly carry important implications for bird species, and if noise masks communication and reduces reproductive success, entire populations could be at risk of decline.
Although birds may be some of more obvious animals when considering impact of noise, research has also found detrimental impacts on terrestrial mammals.
Around wind turbines, California ground squirrels (Spermophilus beecheyi) exhibit more vigilant behaviour and perceive potential threats as risks for longer than when compared to squirrels in non-noisy areas (Rabin, et al., 2006). Wind turbine noise also compromises squirrel vocalization behaviour, which is often used to alert conspecifics of a predator. Similarly, mule deer (Odocoileus hemionus) have been observed to pause more frequently during ruminations and adopt a greater level of auditory vigilance in accordance with local anthropogenic noise features (Lynch, et al., 2015). Though it could be said that this heightened awareness is a good thing as it means the animal is better protected and prepared for attack, with a chronic noise source the constant vigilance could needlessly draw energy away from other processes such as feeding and ultimately be detrimental to the animal. Here, anthropogenic noise is not just a distraction but an added variable animals must consider and work around in their everyday lives. One final consideration is the effect of anthropogenic noise on domestic livestock – often located well within the acoustic range of high frequency rail networks; many animals suffer from the
20 repeated passing of commercial and industrial trains. Factors from communication, to masking of predators, to a much delayed startle response have been found in cattle in rural areas (Hanson, 2008). Here, impacts on communication and subsequent populations do not just constitute an environmental issue, but a social and economic one for people who depend on livestock for financial and nutritional returns.
A taxonomic group largely overlooked in the study of anthropogenic noise impacts is the invertebrates. As one review (Morley, et al., 2013) discusses, only two papers out of 83 considering the impacts of anthropogenic noise on terrestrial species looked at the effect on invertebrates. Invertebrates play a crucial role within ecosystems, and reduction of species diversity or abundance would affect plant diversity and ecosystem processes (Mulder, et al., 1999), with knock on effects on larger fauna. Despite the current dearth in literature, research so far has found comparable effects of anthropogenic noise in insects as seen in birds and mammals. Grasshoppers (Chorthippus biguttulus), that use auditory communication for courtship, produce songs of a higher frequency when in an area of low-frequency noise that would mask or overlap with their mating signals (Lampe, et al., 2013), and cicadas (Cryptotympana takasagona) also shifted their songs to higher frequencies when in higher noise levels
(Shieh, et al., 2012). This potential disruption to communication and mating rituals could affect various species populations and, in turn, whole ecosystems.
These examples highlight the need for more research in this field, as the impacts of noise could be severe across many species.
21 2.2 Anthropogenic noise in the marine environment
The earliest roots of sound in the marine environment can be traced back to
Leonardo Da Vinci in 1490, who is attributed with saying ‘if you cause your ship to stop and place the head of a long tube in the water and place the outer extremity to your ear, you will hear ships at a great distance from you.’ (Urick,
1996). Though ships in Da Vinci’s era would have been very different from the average vessel today, this observation is a good starting point from which other ideas and understanding can advance. Noise pertaining from ships is of course an idea we now accept with no hesitation, and the theories that man-made actions cause noise and damage to organisms underwater is now very well established. Despite this, research has tended to center on other anthropogenic impacts in our oceans, such as ocean acidification and warming. Consequently, our understanding of the effects of anthropogenic noise is lagging behind our knowledge in other fields. The effects of anthropogenic noise are however now starting to gather momentum, and many governments, having recognized the true pervasive nature of anthropogenic noise, are introducing legislation to mitigate for the effects. These actions are a step in the right direction, but it must be recognized there is still a long way to go before we can be confident enough is being done to protect marine organisms.
The first experiment to study the speed of sound underwater was conducted in
1826 at Lake Geneva. Colladon and Sturm measured the elapsed time between a flash of light and the striking of a bell underwater (Sherman & Butler, 2007).
Amazingly, the figure they generated for the speed of sound was within 2% of currently accepted values and soon after this, ‘The Theory of Sound’ was written, which laid down the modern acoustic theory we know and work with today. Underwater technology developments experienced a boom in progress
22 following the sinking of the Titanic, and World War I, largely for marine exploration following the basic ideas of echolocation. The first submarines brought with them a primitive version of SONAR, and developments in this technology throughout WWII and the Cold War allowed the progression in our understanding of underwater acoustics, both theoretically and practically, that we have today.
The primary sources, and biggest contributors, of anthropogenic noise to the marine environment are shipping and industrial action (e.g. pile-driving, the process by which the foundations for offshore wind turbines are drilled into the sea bed). As will be discussed, these anthropogenic noises can affect marine animals in many different ways, from inducing alterations in the endocrine systems associated with the stress response to reduced foraging efficiency and disrupted communication. Many of the impacts of anthropogenic noise have the ability to have effects beyond individuals, at levels of populations and ecosystems - for example, disturbed patterns of communication could lead to reduced social interaction within a population and subsequent reduced mating.
The impacts of anthropogenic noise in the oceans should not be underestimated, especially given the known threats that are now well recognized for the terrestrial environment. Within this section, we will briefly explain the propagation of sound in water (and how it differs from the medium of air) and consider the importance of acoustic cues in the marine environment,, then discuss some of the principal sources of marine noise and study in more detail how they can impact on marine organisms. It is important to think now about how these different noise sources may be having adverse effects at all stages of animal life histories, and what can be done to alleviate these detrimental consequences.
23 2.2.1 Sound propagation in the marine environment
Having introduced the basic properties of sound and its movement (see 2.1.1.), let us now discuss the science of underwater acoustics. In the more particle- dense marine environment, sound propagates further and faster than in air
(Urick, 1996), and sound can be propagated in oceans for much greater distances than many other forms of energy – detectable ranges being up to several thousand kilometers (Stephens, 1970). Sound velocity exceeds that of air by a factor of around 4.4 (approx. 1500 m/s in seawater) and velocity increases further with increasing hydrostatic pressure (depth) - although at depths up to about 100m (relevant here), speed can be regarded as constant as the water temperature fluctuations are small enough to cause only minute variations in speed (Kuttruff, 2007).
As sound waves travel, their intensity diminishes over distance, however properties of certain media can exacerbate this further. A liquid medium, such as seawater, has a very low viscosity and thus has a high thermal consumption of energy, causing sound attenuation within water. Added to this is the scattering and reflection of sound, for which the seabed and surface both act as boundaries - the large impedance contrast between air and water means that the surface acts as a near perfect reflector for sounds below 1kHz and the resultant acoustic impedance of the sea is magnitudes greater than that of air or other gas media (Stephens, 1970). All of these qualities contribute to the overall movement of sound underwater; the high velocity coupled with high acoustic impedance suggest that anthropogenic noises could persist and spread throughout our oceans.
Underwater acoustic pressure is measured using a hydrophone that measures pressure fluctuations, which are then converted to SPL and reported in root
24 mean square (RMS) acoustic pressure referenced against one micro Pascal
(written dB re 1 µPa). Because sound intensity is very different in water, and original air references were set with the hearing thresholds of human adults in mind (dB re 20 µPa), a different reference system is used when discussing sound levels in water (University of Rhode Island, 2004). Particle motion that is considered to play a very important role in how marine animals detect sound can also be measured underwater using an accelerometer; the combination of the two measuring devices allows for comparison of particle acceleration and pressure auditory thresholds within species, and offers deeper resolution into the movement and propagation of sound waves underwater.
2.2.2 Fish hearing and importance of marine acoustics
It has long been known that fish have the capacity to hear, although the importance of acoustic cues in the marine environment have only been recognized more recently. The fish inner ear is responsible for both vestibular and auditory functions, and is responsive to acceleratory and vibrational stimuli in addition to auditory stimuli (Tavolga, et al., 2012). Otoliths (bones within the inner ear) and cilia are primarily responsible for detecting sound waves (as vibrations) – otoliths are more dense than the rest of the fish’s body and they move more slowly in response to sound waves. The difference in the motion between the bones and the body displace the cilia within the inner ear and enervate sensory cells that produce signals transferred by the brainstem to the brain (Schellart & Popper, 2012). Hearing sensitivity may also be linked with the fish swim bladder, which changes in volume in response to different water pressures and passing sound waves (Schulz-Mirbach, et al., 2012) and can feed information back to the cilia and inner ear mechanisms (Blaxter, 1981). It has been suggested that dependent on size, location or even presence of swim
25 bladder, fish can fall on a broad spectrum from high particle motion sensitivity to high-pressure sensitivity (Popper & Fay, 2011) – such examples include purely motion sensitive species like elasmobranchs (with no swim bladder), compared with otophysi (e.g. goldfish, Carassius spp, catfish, Ictalurus spp.) that can detect small changes in pressure due to a connection between the swim bladder and inner ear (Fay & Popper, 2012).
Different marine taxa use acoustic cues and signals in a variety of ways, from communication, learning, settlement orientation and protection from predation.
Cetaceans can use acoustic cues to coordinate group behaviour (Lammers &
Whitlow, 2003), to communicate with conspecifics and with their young (Edds-
Walton, 1997) and for navigation and orientation (Allen, 2013), while pinnipeds use sound for recognition of individuals in a social setting (Insley, et al., 2003) and for development and learning (Schusterman, 2008). Pelagic larval reef fish and crustaceans use acoustic cues for orientation towards settlement sites
(Simpson, et al., 2005; Jeffs, et al., 2003), as do juveniles and adults when relocating habitats (Radford, et al., 2011; Simpson, et al., 2008). Fish can also use acoustic signals for communication (McCauley & Cato, 2000) and some use auditory stimuli to deduce information about proximity and location of predators (Remage-Healey, et al., 2006).
Since marine animals use acoustic cues for many different purposes, the potential masking of these noises through anthropogenically derived action could have a serious consequences – in particular, the masking of settlement cues for pelagic larval stage reef fish, as the future of entire populations depend on whether these early life stages can find suitable habitat to settle. Similarly, the masking of communication cues for marine megafauna could have a detrimental impact on individual interactions and development. Here,
26 consideration is now given to how anthropogenic noises interact with the acoustic soundscape, and how they may be affecting marine populations worldwide.
2.2.3 Sources of marine anthropogenic noise
Many marine anthropogenic noise sources mirror those found in the terrestrial environment, stemming largely from transport and construction sectors. It is important to note that biotic factors contribute to the overall oceanic soundscape also, with ambient noise frequencies ranging from 1Hz to 100kHz dependent on wind and wave movements (Hildebrand, 2009). On average however, ambient noise falls into the ‘medium’ frequency range (500Hz to
25kHz) and owing to greater attenuation cannot propagate very far – these sounds are therefore most relevant to local ecosystems but do not influence the soundscape of more distant habitats (Hildebrand, 2009).
As previously discussed, marine exploration methods have advanced tremendously over the past century. Such growth brought with it many methods still in use today, for example SONAR used for navigation and object detection in water. Both SONAR and seismic exploration methods fall into low (10Hz to
500Hz) to medium frequency ranges (Hildebrand, 2009), and are used largely for military surveillance, despite also having commercial (detection and classification of objects) and scientific (detection of living organisms and particles) uses (Simmonds & MacLennan, 2005). The advent of new technologies has also brought about the development of acoustic deterrent devices, used to modify behaviour and discourage marine mammals from approaching fishing gear and fish farms (Pepper, et al., 2004). Although anthropogenic noises can negatively impact marine life, these devices are used to try and reduce marine mammal bycatch and improve the fishing industry.
27 Figure 1. Showing the frequency range of the most common anthropogenic noises and their overlap with the hearing ranges of important marine taxa. Taken from (Slabbekoorn, et al., 2010).
Developments in marine industries and associated equipment increasingly contribute to anthropogenic noise levels. Offshore wind farms typically have moderate source levels of low frequency noise during operation (Thomsen, et al., 2006), but during their construction – that involves driving of steel piles up to
5m in diameter – this is one of the most intense anthropogenic noise sources in the open ocean. Pile driving is a low frequency, high source level noise that can spread far with little attenuation (Matuschek & Betke, 2009; Bailey, et al., 2010).
As seen in Figure 1, pile driving and operation of offshore wind farms fall within the same range as the majority of anthropogenic sound sources, and these sound levels overlap with the hearing sensitivities of a number of key marine species (see 2.2.4 for more on impacts of noise). Similarly, offshore oil drilling activity produces predominantly low frequency sound (Blackwell, et al., 2004)
28 and is a near constant noise, potentially permanently affecting nearby populations. Mitigation methods such as using an air bubble curtain (to absorb sound energy and inhibit wave transmission through the medium) have been trialed to reduce noise levels from piling, but little benefit has been found thus far for locally affected animals (Wursig, et al., 2000), and so these anthropogenic pollutants still pose a threat to marine life.
Despite the pervasive nature of industrial activity, shipping is still considered to be the dominant source of anthropogenic noise in the marine environment and recent data show that shipping meets approximately 85% of global demands for transport (DNV, 2012). At predominantly low frequencies, shipping can contribute to background noise over very large areas meaning individual vessels are sometimes spatially and temporally indistinguishable within the overall traffic noise (Hildebrand, 2009). This chronic background noise overlaps with the hearing ranges of many important species, and with a predicted doubling of shipping activity by 2030 will rise further over the coming decades
(Lloyd's Marine Register, 2015). Finally, smaller recreational boat noise poses a problem in many shallower waters (Simmonds, et al., 2003), a product primarily of small-scale fishing, transportation, and recreation and tourism industries.
Recreational boat noise has been found to be the primary contributor to environmental noise levels in coastal waters, and despite the episodic nature of passing boats, the noise can be chronic at peak times within shipping lanes
(Haviland-Howell, et al., 2007). One major source of smaller boat noise stems from marine tourism. These vessels produce predominantly mid-frequency level sounds, and higher vessel speeds produce greater noise levels (Hildebrand,
2009). In popular tourism areas such as Cairns and the Whitsunday Islands within the Great Barrier Reef, acoustically pervasive boats make multiple
29 journeys a day to meet the tourism demands on the reef (Great Barrier Reef
Marine Park Authority, 2009). These two tourism hotspots make up just 7% of the GBR (Harriott, 2002), which calls into question how much damage and destruction is being concentrated on such a small percentage area and what is the true extent to which tourism is impacting upon local marine populations.
This list of marine anthropogenic noise sources is by no means exhaustive, but highlights the primary acoustic pollutants that can affect marine animals. The rate of development of shipping over the past few decades, combined with the need to implement sustainable alternatives to fossil fuels, suggests that noise sources within the oceans will continue to increase. Although these industries bring global benefit, both financially and sometimes environmentally, it is vital to consider how these advantages (that predominantly benefit humans) impact on animals. Thus further innovation in mitigation approaches and technologies is needed, grounded in the latest scientific evidence on impacts noise.
2.2.4 Known impacts of marine anthropogenic noise
Impacts of noise can be split into two main categories: physiological and behavioural impacts. Due to experimental constraints, especially for marine mammals, much more is known about behavioural responses to noise, as measuring physiological responses often causes stress that confounds the results. Nevertheless, we understand that marine anthropogenic noise can have serious detrimental effects and can persist in all areas of an organism’s life.
For various cetacean species, anthropogenic sounds in the ocean can impact on surfacing and diving patterns, and can invoke changes in type and timing of many vocalizations (Nowacek, et al., 2007). Blue whales (Balaenoptera musculus) are less likely to call when exposed to SONAR, but will call for longer in the presence of ships (Melcon, et al., 2012). These changes are most likely
30 linked to the frequency of the source noise, and could carry long-term implications for vocal behaviour within cetacean groups, affecting foraging efficiency and mating opportunities through impacts on social interactions
(Weilgart, 2007). SONAR can also cause mass whale strandings in a number of cetacean species (Balcomb, 2001), and has been associated with inducing ‘gas and fat embolic syndrome’ in beaked whales (Ziphiidae spp.) (Fernandez, et al.,
2005). Noise can also cause an increase in stress hormone levels in cetaceans
(Romano, et al., 2011) and has even been observed to cause temporary hearing threshold shifts (Schlundt, et al., 2000) following intense sound exposure. The reliance on sound in marine mammal for many processes means that anthropogenic noise will have detrimental effects on their life functions, as seen from implications for reproduction to accelerated aging and sickness
(Wright, et al., 2007).
In fish, impacts of anthropogenic noise span induction of stress, growth and reproduction effects, to hearing loss. Noise may induce increased cortisol levels
(Wysocki, et al., 2006) and increased metabolic and ventilation rates (Simpson, et al., 2014) when exposed to noise; suggesting a stress response, if prolonged, can impact survival. A study on largemouth bass (Micropterus salmoides) also found a dramatic increase in heart rate and decrease on stroke swimming rate during exposure to short bursts (up to 60 sec) of boat noise
(Graham & Cooke, 2008), demonstrating that even small boats can have potentially destructive ecological and environmental impacts.
Exposure to high-level noise can invoke temporary threshold shifts in hearing specialist species such as the otophysii (Amoser & Ladich, 2003), and can even cause hearing loss following prolonged exposure as shown in Carassius auratus (Smith, et al., 2003). Since many species use acoustic cues for
31 navigation and orientation, and a small number are also vocal, this implies potential problems for communication and navigation (Vasconcelos, et al.,
2007). This would be particularly important for pelagic larval stage fish, which may not be able to settle to appropriate habitat if natural reef sounds are masked by anthropogenic noise (Simpson, et al., 2004).
Anthropogenic noise can also cause behavioural impacts in fish. Pile-driving noise has been found to negatively impact foraging and anti-predator responses, while also causing an increase in aggression at different levels within hierarchies of a species (Bruintjes & Radford, 2013). Exposure to anthropogenic noise also causes a deviation from usual schooling behaviour in
Bluefin tuna (Thunnus thynnus), which could affect their migration routes and subsequent spawning and feeding behaviours (Sara, et al., 2007), and in swimming behaviour and speed in cod (Handegard, et al., 2003). These behavioural and social changes can have knock on effects for overall species survival, including maintaining sufficient food resources and ensuring reproduction success.
The impacts of anthropogenic noise are clearly extensive, and range from short-term, immediate effects to potentially life-altering responses. The fact that a stressor such as noise, with no biochemical foundations, can have such a large scope of impacts shows that any alteration of the environment, no matter how seemingly small or circumstantial, can have serious impacts and affect entire life histories and species. However, although noise can be highly detrimental, it is also one of the more simple stressors to mitigate, and we must remember this when moving forward and considering how to tackle impacts of this anthropogenic pollution.
32 2.3 Living with stressors
Stress is an important factor in the lives of all animals. Stress can influence many aspects of an organism’s life and can take hold in a variety of ways, from the routine to the potentially destructive. The very definition of stress could satisfy an entire section, especially if considering opinions from differing schools of thought, notably between the physical, life and social sciences. It is not within the scope of this thesis to spend a long time reflecting on the relative merits and faults of each school; instead this work will draw upon the facts from each discipline to explain stress in a manner that is relevant to this research. Stress is a phenomenon that enables organisms to cope and adapt to their environment, and is a life process central to survival. However, chronic stress can cause adverse health impacts, and it is here we will endeavor to explain the stress response from a bio-physiological perspective: how we can measure an individuals stress response, to separate impacts of stress into long and short- term effects, and to quantify why these consequences matter. Finally, we will conclude with brief comments on the still unknown aspects of the effects of stress, with particular reference to fish, and potential future directions of study.
2.3.1 The stress response
Hans Selye first introduced the idea of stress to the scientific community in
1936, suggesting ‘stress…is the most meaningful subject for humanity I can think of’ (Szabo, et al., 2012). Although Selye’s original theories have been developed and refined considerably, this quote puts into perspective the true importance of understanding and monitoring stress. A concrete definition of stress seems difficult to derive – indeed entire books have been written on the subject and have still reached no conclusion (Fink, 2010) –– but for ease of reference henceforth I will take stress to mean ‘the response of an organism to
33 any demand placed on it such that it causes an extension of a physiological state beyond the normal resting state’ (Iwama, et al., 2011). This best encompasses the idea of stressors as we understand them, and the stress response in terms of principally physiological change. After decades of extensive further research, the three main phases of stress proposed by Selye are still accepted today: first, the ‘fight or flight’ response; second, the stage of adaption (to combat the first excitable stage); and third, a period of exhaustion provided the exposure to stress was prolonged. These definitions came at a time when the idea that stress could affect well-being or illness was considered naïve, although we now know that there are many such links. The majority of terms used in discussing stress suggest a predominantly human phenomenon, however the accepted notion of a deviation from homeostasis can be used to explain stress for any organism (Chrousos & Gold, 1992). Furthermore, the biochemical foundations for a stress response apply to any animal, and can be simplified to comprise two major systems, the sympathetic nervous system
(SNS) and the hypothalamic-pituitary-adrenal (HPA) axis. Both are activated very rapidly following exposure to an acute stressor, or steadily escalate in response to a stressor that develops over a longer time period.
SNS activation can be broken down into direct and indirect measures, the former including increased heart rate and blood pressure, the latter being increased catecholamine plasma concentrations (e.g. epinephrine, norepinephrine). HPA axis activation (e.g. corticosteroid concentrations) data offers more information into the severity of a stressor, and can be used individually or collectively alongside SNS data. It is important to note here that these responses may arise as a result of other activity or even arousal, and are not always wholly indicative of a stressor - therefore caution must be taken to
34 discernably review the data and determine whether the changed responses are the product of a stress response or something completely different (Wright, et al., 2007).
The SNS is responsible for the implication of many physiological changes during a ‘fight or flight’ response (including accelerated heart rate, constricted blood vessels and perspiration), and is one of two parts of the autonomic nervous system, a division that influences internal organs. The HPA axis, however, is a part of the neuroendocrine system that regulates many body processes (such as stress, digestion, and the immune system) but does not control particular organs, more the hormones within them. The responses of both of these systems invoke different responses, which for ease of reference and understanding can be split into primary (corticosteroid and catecholamine release), secondary (metabolic changes, ventilation changes, haematocrit changes, osmoregulatory disturbance, immune function change) and tertiary responses (whole performance, growth, swimming capacity, feeding and aggression changes) (Barton, 2002). Many of these responses can be invoked independently, but may also occur as a result of a higher response – for example, corticosteroid release could induce metabolic level changes (McKay &
Cidlowski, 2003), which in turn could affect swimming capacity.
The best way to understand how a stressor might affect an animal is to measure a variety of different responses, and try to use a whole-system approach to help understand how certain responses may interact with and underpin others. In this way, one gains insight into multiple processes at different physiological and behavioural levels, providing information on the cohesive impact of a stressor, and providing understanding on ways to alleviate detrimental effects.
35 Primary Response(s) Chemical Stressors e.g. increase in e.g. pollution, low hormone levels oxygen
Physical Secondary Response(s) Stressors e.g. metabolic changes e.g. capture, (increases in glucose and/or handling lactate); changes in immune function
Tertiary Response(s) Perceived Stressors e.g. changes in whole-animal e.g. stimuli evoking a health (growth, reproduction, startle response, such as disease resistance); sound, presence of behavioural changes (feeding, predator aggression)
Figure 2. A simplified depiction of the fish stress response. The three primary categories stressors can be sorted into, and the three levels of stress responses that fish experience, with examples and potential interactions. (Reproduced from Barton, 2002)
2.3.2 Stress responses in fish
The primary, secondary and tertiary responses outlined above are applicable to the stress response in fish and serve as a valuable tool when categorizing and understanding different reactions to stressors (Barton, 2002). The response of different species of fish to stressors can be highly variable, and often parallels the many adaptations among fish species for life in different environments
(Bonga, 1997). However, the properties of the fish stress response remain similar to that of terrestrial vertebrates, in particular the stimulation of oxygen uptake, reallocation of energy away from growth and reproduction, and suppressive effects on immune functions (Bonga, 1997).
Differences in stress responses in fish often relate to the environmental challenges that they face. For example, ocean acidification can physiologically change the surface epithelia and gill permeability, leading to disturbances in bodily water and mineral levels. The unavoidable exposure of fish to many aquatic pollutants can damage the extensive gill network, and explains why
36 marine stressors with biochemical roots evoke such cohesive stress responses, with a whole host of toxic effects (Bonga, 1997).
The release and circulation of corticosteroids is a primary level stress response in fish. Plasma cortisol is one of the most direct and candid responses following
HPA axis activation, and is also one of the most potentially damaging factors following a stress response (Pickering, 1992). Although many other hormones may be seen in responses to a stressor, plasma cortisol is still the ‘stress indicator of choice’ and continues to be at the forefront of physiological stress research (Iwama, et al., 2011). Catecholamines (for example epinephrine) are sometimes also measured but levels in the blood do not generally change unless under the influence of a severe physiological stressor (Perry & Bernier,
1999).
Other metabolic and osmoregulatory (secondary) changes are often monitored as a more straightforward and effective way of ascertaining the impact of a stressor. These changes occur as a result of primary endocrine variations, but are generally less invasive than testing for elevated levels of corticosteroids or catecholamine (Fabbri, et al., 1998). Tertiary responses can also be used to monitor longer-term effects of a stressor; changes at this level are reflected in functions such as growth, reproduction and the immune system. These whole system and behavioural measures are perceptive indicators of many downstream physiological changes, and can also be interpreted within an ecological context (Iwama, et al., 2011).
The severity of a response is usually correlated with the length of exposure and/or the magnitude of the stressor (Iwama, et al., 2011), although some research has found the contrary to be true for disturbances in metabolic rate
(Mazeaud, et al., 1977). As responses cascade in a hierarchical fashion, there
37 can be a delay in secondary and tertiary responses (Arends, et al., 1999), and some latter responses such as effects on the immune system may not manifest until days or even weeks after exposure (Schreck, et al., 2001). Fish ultimately undergo the three stages of response to compensate for the difficulties imposed by the stressor and to survive, and this will often involve the reallocation of energy to vital processes during moments of concentrated stress. The production of glucose (partly through increased metabolic activity) aids with this energy provision, and substrates are quickly delivered to the brain, gills and sometimes muscles to help combat the stressor (Iwama, 1998). Although immediately beneficial to the animal, this diverts energy away from processes such as food digestion and growth, and in prolonged circumstances can deplete energy from liver and muscle glycogen stores (De Boeck, et al., 2000).
Larval fish as young as 2-weeks post hatching demonstrate corticosteroid responses to stressors, indicating that the SNS and HPA axis can be responsive even before exogenous feeding begins (Barry, et al., 1995). The two-week hypo-responsive period immediately following hatching may allow physiological functions to remain constant and allow basic development without complication (Romero, 2004). However, response to stress at such an early stage carries serious implications for fish in today’s polluted marine environments, where they are regularly exposed to anthropogenic stressors.
Stress in fish can manifest in many ways; the responses outlined above are just the most common and straightforward responses. The interaction of the SNS with different parts of the body means that the primary stress pathway can impact on a number of processes, from the most basic physiological functions such as ventilation, to routine processes such as digestion, through to vital behaviour including antipredator response. Stress is important and should be
38 considered across all fish research, from the study of anthropogenic stressors to aquaculture development. Clearly the pervasive effects of stress should not be underestimated or ignored.
2.3.3 Impacts of stress
Stress can manifest in a fish in many different ways and initiate multiple pathways of response. Many of these responses allow the animal to deal with a situation, but there are also costs of stress, which can be divided into short- and long-term impacts. While some changes are only detrimental during exposure to the stressor and/or for a short time period afterwards, others may cause persistent harm long after the stressor goes away. Additionally, inherently chronic stressors (such as ocean warming or ocean acidification) may provoke what seems like habituation in an organism (where a strong stress response can no longer be detected) whereas in reality the fish has undergone a physiological shift to levels above the (non-stressed) baseline.
Physiological changes associated with a stress response are initiated by the release of corticosteroids and catecholamines, followed by changes to metabolic and osmoregulatory function. Negative impacts of elevated cortisol levels have long been established in fish research. For example, cortisol can suppress the immune system and cause mortality through increased susceptibility to bacterial and fungal diseases (Pickering & Pottinger, 1989), inhibit reproduction through reduced gonadal development (in comparison with non-stressed counterparts) and reduced expression of reproductive hormones in both sexes (Carragher, et al., 1989), and have severe negative effects on appetite, leading to changes in condition factor and rate of growth (Gregory &
Wood, 1999). Raised cortisol levels can also suppress levels of heat shock
39 proteins, important for monitoring and assisting the establishment of cell proteins and vital in aiding recovery to physiological stress (Basu, et al., 2001).
When fish experience raised cortisol levels for a prolonged period, detrimental effects are often proportionally more severe. In a study on Atlantic Salmon
(Salmo salar), as time exposure to a stressor increased, white blood cell expression (important for fighting diseases) decreased (Fast, et al., 2008); clearly further weakening the immune system. Another found that aggressive behaviour (important for defense) in Rainbow Trout (Oncorhynchus mykiss) was increasingly inhibited as cortisol levels increased (Overli, et al., 2002), which could result in weaker defenses and increased risk of predation. Studies that exposed fish (male Common Carp, Cyprinus carpio, female Salmon,
Oncorhynchus kisutch) to long term cortisol treatments found implications for growth and development, suggesting fish exposed to continual stressors throughout development may be at risk of suppressed pubertal hormones and subsequent stunted growth of reproductive organs (Consten, et al., 2001;
Stratholt, et al., 1997), which could cause serious complications in later life for individual reproduction with consequences for population stability. Exposure to stress can also affect offspring, with maternal cortisol levels directly affecting size and morphology of larval progeny (McCormick, 1998), demonstrating how effects of stress can transcend generations and impact larval survival despite no direct experience of the stressor.
Reproduction in fish can also be affected through changes in metabolism; for example reallocation of energy to maintain homeostasis in Rainbow Trout can disrupt the timing of mating and reproduction (Schreck, et al., 2001). Metabolic disruption can lead to exhaustion in chub fish (Leuciscus cephalus), through anaerobic respiration and increased glycogen levels (Lackner, et al., 1988), and
40 if fish are chronically affected can have serious costs in maintenance (Lankford, et al., 2005) and growth rates, even beyond apparent habituation (Jentoft, et al.,
2005).
Stress can impact physiological processes including growth, reproduction and immune function via biochemical functions that occur within the body and underpin these functions. However, stressors can also impact fish behaviour by acting as a distractor, diverting attention away from important tasks and decisions towards the perceived stimuli of a stressor (Chan & Blumstein, 2011).
Stress can elevate levels of aggression, and in African cichlids (Astatotilapia burtoni) cause displaced aggression in non-dominant members and disruption of social hierarchies (Clement, 2005). Indeed the outcome of rank-order fights in rainbow trout may be correlated with plasma cortisol levels (Pottinger & Carrick,
2001). Stressors including noise can also reduce foraging success in terms of live prey consumed and efficiency in terms of food handling errors and discrimination between food and non-food items (Voellmy, et al., 2014). Various stressors, both chemical (Wibe, et al., 2001) and situational (Voellmy, et al.,
2014), can also have a profound effect on antipredator response. Following a predatory attack, the presence of a stressor can make animals likely to return to feeding more rapidly than their unstressed counterparts (Lima, 1998), with a reduced vigilance that carries serious impacts for survival in many species.
Although it is clear that stressors can affect many aspects of life, there are still many influences that are poorly understood; thus the study of impacts of stress across taxa is a field with substantial scope for growth. For fish, much remains unknown about how environmental stressors can affect future generations. For example consideration of epigenetic effects is only just beginning to receive attention. Furthermore, future research should explore different responses (in
41 particular, secondary and tertiary responses) to understand how they interact and influence each other.
42 3. Study aims and objectives
Within the academic study of marine biology, relatively little is understood about responses to noise in different contexts and how noise may interact with other stressors. The aim of this Masters research was to study the stress response of coral reef fish to anthropogenic noise, and to test whether these responses to noise are context-dependent – that is, do they differ with changing environments and external factors. To address this, it was first considered whether reported impacts of noise were affected by experimental conditions.
The bulk of anthropogenic noise research is conducted in a laboratory setting.
Although this environment allows for careful control of experimental variables and easy manipulation of different conditions, results are subject to a methodological bias that makes it difficult to extrapolate to the real world.
The first study (Chapter 4) used a combination of fieldwork and laboratory work, and both wild type fish and aquarium trade fish, to assess the stress response of fish to noise in different experimental contexts. Data was compared using fish collected directly from coral reefs (most closely aligned with the real-life environment and thus most easily extrapolated) with data from fish obtained through the aquarium trade (the furthest removed from a real-life setting).
Experiments also compared responses to real sources of anthropogenic noise
(small motorboats) with playback of recordings of the same boats in field and laboratory conditions. Using two widespread Damselfish species
(Pomacentridae), this study tests the hypotheses that there will be a significant difference in the stress response of fish between field and laboratory studies, with a noticeably heightened stress response in laboratory studies due to a lack of a familiar environment, and aims to answer the crucial question of whether lab-based studies, using artificial stressors, have any ecological validity.
43 The second study (Chapter 5) considered whether impacts of noise were influenced by dietary condition. Experiments were designed to test whether another common stressor– not having sufficient feeding opportunities – affected the response to anthropogenic noise. In current conditions, fish are simultaneously subjected to multiple stressors, but most research considers each stressor in isolation. To study the interaction of two common stressors, condition was manipulated through limited feeding with anthropogenic noise and tested the hypotheses that all fish will show an increased stress response when exposed to motorboat noise, and fish subject to limited feeding will exhibit a heightened stress response in anthropogenic noise conditions but not in ambient conditions. Using experiments carried out in the field with wild type fish, this study examined the interaction between different stressors, and explored how fish may respond to stress in different contexts. Discovering the mechanisms behind these stressors will deepen our scientific understanding, and it is hoped that these findings will provide a starting point from which similar studies can branch out.
44 4. Context-dependent responses to anthropogenic noise
4.1 Introduction
Ecological studies often carry a trade-off between field work and laboratory work – the former best replicating the real-world environment and yielding the most directly applicable conclusions; the latter allowing for the greatest control over extraneous variables - and it is this control over variables that often proffers the most statistically robust results (Halfwerk & Slabbekoorn, 2015).
For the majority of marine studies, the choice between field and laboratory work is taken out of the researchers’ hands – for example, it is not practical to study large groups of social marine mammals in the laboratory, or to replicate predicted future ocean acidification conditions in a field setting. Select fields (in particular in marine biology, fields including (but not limited to) reef fish social interactions and predator/prey interactions, for example) offer the chance to choose either, however, this is often still at the detriment or compromise of certain ideals. Using multiple experiments in the field and in the laboratory, and with wild-type fish and aquarium trade fish, this study explored the effects of different experimental situations to test whether laboratory experiments differ from reality when considering anthropogenic impacts, more specifically anthropogenic noise.
Anthropogenic noise
Noise in the ocean has been growing steadily over the past few decades as a result of increased industrial, transportation and recreational activity, and could now be causing ecosystem-wide impacts (Slabbekoorn, et al., 2010). Recent research into how anthropogenic noise affects marine life suggests that scientists could be underestimating the widespread influence on wildlife due to the many hidden costs of noise exposure, including long-term and broad spatial 45 changes (Francis & Barber, 2013). Examples of these changes include disruption to foraging routines (Miksis-Olds et al., 2007) and abandonment of habitat (Weilgart, 2007) following exposure to ship noise.
Acoustic signals and cues are used by a range of taxa for all different purposes, from cetaceans that use noise as a means for communication and learning
(Harrison & King, 1965), to pelagic larvae of reef fish and crustaceans that use acoustic cues for orientation towards settlement sites (Jeffs et al., 2003;
Simpson, et al., 2005). When anthropogenic noises mask the natural sounds of reefs or low-frequency acoustic communication between marine animals
(Vasconcelos, et al., 2007), this may have implications for survival, and at the very least can induce a stress response with a negative impact on the animal.
The combined nature of anthropogenic noise (with the need to often accurately but repeatedly replicate the sound) and physiological study (where close monitoring of the animal is often necessary) lends itself to laboratory work. The process of playback of anthropogenic noise involves careful preparation; the track has to be recorded from a real noise source, and then reconstructed using specific software. Following this, the track is played into the experimental area and re-recorded and modified to meet a uniform level – creating a track that attempts to replicate sound-pressure waves as accurately as possible (Wale, et al., 2013). Reflections of sound-pressure waves from the enclosure walls must be considered, alongside vibrations within the tank caused by the speaker itself
(OGP, 2008). Although significant progress has been made in this research field over the past couple of decades to refine the playback of noise in the laboratory, controlled experiments still come with logistical and scientific problems, not least designing an accurate experiment with high enough
46 statistical power and the subsequent challenge to predict long-term population effects (Tyack, 2009) – the definitive aim of scientific endeavor.
Merits of field experiments
Despite the wealth of scientific data readily available today, very few anthropogenic-noise studies have compared results from field- and laboratory- based work; those that have, although useful, are somewhat dated and so do not use sophisticated and contemporary methods generally expected in publications today. It is surprising that such little empirical research exists, especially considering the increasing cause for concern in these relatively new fields – ocean acidification, marine plastics and anthropogenic noise to name but a few. The current nature of fieldwork lends itself largely to observational e.g. (Baird & Whitehead, 2000; Eckert, et al., 2008) and analytical (statistics carried out on pre-existing data sets) work e.g. (Halpern & Warner, 2002), rather than the manipulation of species or surroundings. Nevertheless, slightly older studies are useful to our initial understanding and can provide a building block to work from, and fieldwork often presents the ideal opportunity to most closely approach the true ecological context. Talling (1960) considered marine plankton in a laboratory and field setting, and found photosynthetic characteristics to be in general agreement Another study found comparable results in metal-induced physiological effects between laboratory and field studies (Larsson, et al., 1985). The matched findings when considering physiological responses seem logical; we would expect basic processes to remain unchanged in a laboratory setting provided extraneous variables have been adequately controlled. Behaviour, however, may differ in a laboratory setting, where visual, auditory, olfactory and gustatory signals are likely to differ from a real-life marine environment. One study that applied laboratory-based
47 learning to a field environment saw slight behavioural changes in response to varying environmental contexts, but parallel physiological and morphological patterns (Black, et al., 2005). However, at the basic level, behaviour of the reef fish species Eupomacentrus partitus (Bicolor damselfish) barely changed in the laboratory environment when compared with field observations (Myrberg,
1972). Though these studies do not directly consider anthropogenic disturbances, it is important to clarify how fluctuations in an organism’s environment may or may not affect the individual’s physiology and behaviour. It could be argued that behaviour is consistent provided the routine of an organism is not disrupted; indeed more complex behaviours such as predator– prey interactions and habitat selection remain consistent between field observations and laboratory experiments (Main, 1987). It should be noted that in this last study, the author attributed the close behavioural correspondence between laboratory and field to the presence of the natural habitat of the study species. This highlights the importance of a precise laboratory set-up, and suggest that laboratory work on less well-studied species introduces the risk of a departure from findings of relevance in the real world.
Merits of laboratory studies
Laboratory-based studies also have several advantages, perhaps the most important of which is the opportunity to singly address different parts of a research question. The cost associated with fieldwork is also a large factor: travel, accommodation, equipment and subsistence is often a limitation that can direct researchers towards laboratory studies on familiar soil. Since many species are now easy and inexpensive to procure, and animal husbandry of marine species is significantly developed (Pillay & Kutty, 2005), this allows housing and testing of model species. A laboratory environment offers the
48 opportunity to manipulate isolated variables that can generate predictions for future conditions (Ridgwell, et al., 2009), and can also provide the opportunity to understand more complex processes arising from field observations (Grant &
Brown, 1999). However, the removal from a real-life environment invariably brings with it a degree of ambiguity, and extrapolation of data to the real word is a major limitation to laboratory studies (Mackenzie, et al., 1990).
Damselfish (Pomacentridae) make up almost 50% of coral reef fish biomass and are auditory species known to utilize acoustic reef cues and respond to anthropogenic noise (Simpson, et al., 2004). Blue-green damselfish Chromis viridis and spiny chromis Acanthachromis polyacanthus are easy to capture and house at Lizard Island, Australia and are also available through aquarium suppliers in the UK. Both species are important to the overall productivity of reef communities (Klumpp, et al., 1987), making them ecologically relevant species.
This study used two physiological responses, metabolic rate and opercular beat rate, in the two species in a mixture of laboratory and field-based environments to test the real-world validity of laboratory experiments and examine whether impacts of playback of noise results differ from impacts of real noise sources.
For these experiments, the noise was that of small motorboats with outboard engines, or playback of recordings of the same boats. Such boats are commonly used for tourism in shallower waters, thus they best replicate accurate anthropogenic exposure for coral reef fish. Both field and laboratory studies were conducted in Australia with wild-caught fish, and subsequent laboratory studies were conducted in Exeter (UK) using fish obtained through the aquarium trade. The study tested the hypothesis there would be a significant difference in metabolic rate and opercular beat rate between field
49 and laboratory studies (H1), and that opercular beat rate and metabolic rate will be noticeably higher in laboratory studies (H2) due to the high level of particle motion generated by the speakers in a confined space coupled with the lack of a familiar environment.
4.2 Materials and Methods
4.2.1 Lizard Island Research Station, Australia
4.2.1.1 Test subjects
Adult C. viridis (mean ± SE: 4.79 ± 0.05 cm, standard length) and A. polyacanthus (mean ± SE: 7.35 ± 0.11 cm, standard length) were collected on
SCUBA from shallow reefs (3–6 m) around Lizard Island Research Station
(LIRS; S 14° 40′, E 145° 28′), Northern Great Barrier Reef, Australia, using clove oil (Munday & Wilson, 1997) and hand nets (C. viridis), and barrier nets and hand nets (A. polyacanthus). Subjects were housed at LIRS in six species- specific glass tanks (45 x 67 x 62 cm) with constant flow-through of fresh seawater. All tanks were enriched and given shelter with equivalent pieces of dead coral and plastic tubing. Tanks were protected from the elements but exposed to natural dawn and dusk. Fish were fed a mixed diet of live Artemia and adult food pellets (Perla MP, Skretting, Verona, Italy) three times daily. All husbandry and handling was in line with Australian standards as outlined by the
Great Barrier Reef Marine Park Authority (GBRMPA), and all work carried out with permission and ethical approval from: Lizard Island Research Station,
Great Barrier Reef Marine Park Authority, James Cook University (A2081),
Australian Institute of Marine Science and University of Exeter (2013/247).
50 4.2.1.2 Noise protocol
4.2.1.3 Field: boat noise
In the ‘field: boat’ noise trials in Australia (to be referred to as FBA), fish (C. viridis and A. polyacanthus) were exposed to real boat noise at Casuarina
Beach, LIRS. Ambient reef noise was used as a control condition in all experiments. To reduce pseudo-replication, three different research station boats (5-m-long aluminium hulls with 30 hp Suzuki outboard motors) were used, all driven 40–200 m away throughout the trials. Boat and ambient noise recordings were taken from a kayak moored using an anchor without a chain to avoid unwanted noise and were made 1m above the seabed for 5 min. Acoustic pressure was recorded using an omnidirectional hydrophone (HiTech HTI 96-
MIN, inbuilt preamplifier; manufacturer calibrated sensitivity -164.3 dB re 1
V/µPa; frequency range 2 Hz to 30 kHz, calibrated by manufacturers, High Tech
Inc., Gulfport MS) and a digital recorder (PCM M10, 48kHz sampling rate, Sony
Corporation, Toyko, Japan). Particle acceleration was measured using a calibrated triaxial accelerometer (M20L; sensitivity following a curve over the frequency range 0–3 kHz; calibrated by manufacturers; Geospectrum
Technologies, Dartmouth, Canada) and a digital 4-track recorder (Boss BR-800,
44.1 kHz sampling rate, Roland Corporation, Los Angeles, CA). Recording levels used with each set-up were calibrated using pure sine wave signals from a function generator with a measured voltage recorded in line on an oscilloscope.
Recordings were analysed using audio software Avisoft SASLab Pro Software
(version 5.1.17, Avisoft, Berlin, Germany), and MATLAB (v2010a, MA, USA).
Fast-Fourier Transformation was used to calculate power spectral density for comparison of sound levels for each treatment across the frequency range 0–
51 3000 Hz (Fig. 3). Spectrograms of examples of the two noise conditions are also shown for comparison (Figs. 4a-b).
4.2.1.4 Field: playback of noise
In the ‘field: playback’ trials (to be referred to as FPA), fish (C. viridis and A. polyacanthus) were exposed to playback of boat noise on Casuarina Beach,
LIRS. Playback was via a sound system, which employed the same basic set- up used in previous studies assessing effects of anthropogenic noise on fish
(Simpson, et al., 2015) and consisted of an MP3 player (Archos 1 Vision,
Archos, Igny, France), an amplifier (18W, frequency response range 40-
20,000Hz, Kemo-Electronic Gmbh), a battery (12v 7.2 AH sealed lead-acid) and an underwater speaker (Lubell Labs University Sound EV UW-30, frequency response range 100-10,000Hz). Ambient (control) noise playback files were created using daytime recordings from three LIRS reefs (Casuarina, Horseshoe,
Big Vickys, see above for recording protocol). These were combined with recordings of three different boats (driven 40–200 m away) using Audacity
(version 2.0.5 for Mac OSX) for the boat noise treatment. To reduce pseudo- replication, nine unique tracks were created for both ambient and boat noise conditions and assigned to control or test fish respectively at random (note all tracks were used an equal amount of times). Before experiments began, noise playbacks were recorded and recordings were analysed (as above). Playback using speakers in both natural settings and in tanks alters the characteristics of the original recordings. However, analysis of spectral content and sound levels showed that characteristics of the original recordings were at least partially retained in playback and that these characteristics differed between playback of ambient and motorboat noise in both the field and in tanks. For comparison, spectral content of acoustic pressure and particle acceleration of ambient and
52 boat noise are shown together in Fig. 3 followed by spectrograms of examples of the two sound treatments (Figs. 4c-d).
4.2.1.5 Lab: playback of noise
In the ‘lab: playback’ trials in Australia (to be referred to as LPA), fish (only C. viridis in this case due to logistical constraints) were exposed to playback of boat noise in plastic tanks (45 x 67 x 62 cm) in the Lizard Island wetlab facility.
The same boat and ambient recordings as used in the ‘field: playback’ trials were allocated at random to fish in either the test or control condition respectively. Before experiments began, noise playbacks were recorded and recordings were analysed (as above). For comparison, spectral content of acoustic pressure and particle acceleration of ambient and boat noise are shown together in Fig. 3a-b followed by spectrograms of examples of the two noise conditions (Figs. 4e-f). MP3 sound files were played into the experimental tank via a sound system (as above, see 4.2.1.1).
4.2.2 Aquatic Resources Centre, University of Exeter, UK
4.2.2.1 Test subjects
Adult C. viridis (mean ± SE: 4.19 ± 0.05 cm, standard length) and A. polyacanthus (mean ± SE: 5.91 ± 0.07 cm, standard length) were obtained from the Tropical Marine Centre in the UK and housed in the Aquatic Resources
Centre (ARC) at the University of Exeter (Exeter, UK). Fish were held in two species-specific glass tanks (81 x 43 x 61 cm, water depth 20 cm) in artificial seawater (salinity 35pp) at 26°C with a simulated dawn at 0800 hours and dusk at 2000 hours. Both tanks were given equal habitat of coral and tubing. Fish were fed a mixed diet of live Artemia and juvenile food pellets (Perla, MP,
Skretting, Verona, Italy) three times daily. All storage and handling was in line
53 with European Union standards as outlined by the Animals Scientific
Procedures Act (ASPA) (Home Office, 2012).
4.2.2.2 Noise playback protocol
4.2.2.3 Lab: playback of noise
In the ‘lab: playback’ trials in Exeter (to be referred to as LPE), fish were exposed to playback of recordings of boat (test) and ambient (control) noise.
The playback protocol was identical to that in Australia, using the same playback tracks (see 4.2.1.1).
4.2.3 Experimental protocol for measuring physiological impacts of noise
Experiments at LIRS took place from October to December 2014; experiments at the ARC (Exeter) took place in May 2015. Single fish were placed in individual sealed experimental containers (up to six containers at any one time,
1 L, 115 x 115 x 160 mm, Sistema Plastics) painted white on the exterior to remove external visual disturbances, but to allow light in. Care was taken to reduce exterior noise. In FBA and FPA trials, activity not pertaining to the experiment took place a minimum of 300 m away (Simpson, et al., 2016). In
LPA and LPE trials, external noises were controlled and reduced as much as possible: experimental plastic tanks (45 x 67 x 62 cm, water depth 44cm) were placed on 5 cm insulation foam, and underwater speakers were positioned on the bottom of tanks on insulation foam to minimize vibrations from noise playback. In all trials, fish were given a minimum of 30 min to recover from handling stress and to acclimate to their container; during this time, containers were left open and water refreshed regularly to avoid oxygen depletion. Fish experienced ambient noise during acclimation. The container was then transferred to the experimental area (either in the field or in the tank), where the
54 water was refreshed one final time and the container was sealed. After each trial, fish were transferred to a separate holding tank and the container was washed out with fresh seawater in preparation for the next trial.
4.2.3.1 Opercular Beat Rate
This experimental paradigm was designed to test for an effect of boat noise or playback of boat noise on ventilation, measured as the opercular beat rate
(OBR). A total of 271 fish were used (Table 1). In each trial, single fish were sealed into individual containers (see 2.3) and were exposed to ambient noise for 10 min, then either boat or ambient noise for 20 min. Fish were filmed for the duration of the trial using video cameras (GoPro HERO 3, 3+, 4, GoPro Inc.,
California, US) attached to the lid of the container and facing inwards.
4.2.3.2 Metabolic Rate
This experimental paradigm was designed to test for an effect of boat noise or playback of boat noise on metabolic rate, measured as the rate of oxygen uptake. Fish were starved for 12 h before experiments began. A total of 331 fish were used (Table 1). In each trial, single fish were sealed into individual containers (see 2.3) and exposed to 10 min of ambient noise and then either boat (test) or ambient (control) noise for 20 min. After 20 min, the container was removed from the experimental arena, and the lid was taken off. The fish was immediately removed from the tub and the oxygen concentration of the water was measured using an oxygen probe (Dissolved Oxygen and Temperature
Meter HI 9164, Hanna Instruments Inc., Woonsocket, RI). These concentrations were compared to measurements of oxygen levels in the water that was used to fill the containers at the start of each experiment.
55 4.2.4 Statistical Analysis
All statistical analyses were conducted with SPSS version 22.0 (IBM Inc., New
York, US).
4.2.4.1 Opercular Beat Rate
Video recordings were exported and analysed without sound (to avoid observer effects) in QuickTime version 10.4 (Apple Inc., California, US). OBR was counted using a clicker for 15-s subsections for the 2 min prior to exposure, the initial 5 min of exposure, then for 15-s subsections at 10 and 20 min after the start of either the test or control exposure. Mean OBR was calculated for the 2 min prior to exposure (pre1&2), the 2 min post-exposure (post1&2) and the 10th minute post-exposure (post10). The differences in OBR between pre1&2 and post1&2, and between pre1&2 and post10 were analysed using an ANOVA, allowing for variance in size (length), testing for an effect of noise (boat - test; ambient - control), situation (FBA, FPA, LPA, LPE) and their interaction. The rate of the change of OBR function (hereafter referred to as gradient change of
OBR) was calculated over the 5 min following sound exposure and also analysed using an ANOVA, with the same fixed factors.
4.2.4.2 Metabolic Rate
Measurements from the oxygen probe were converted to metabolic rate (MR) values (M02, mmol/g/hr) for each fish over the 30-min exposure period and analysed using a two-way ANOVA, testing for an effect of noise (ambient, boat), situation (FBA, FPA, LPA, LPE) and their interaction.
56 4.3 Results
4.3.1 Opercular beat rate
4.3.1.1 Chromis viridis
The OBR change between pre1&2 and post1&2 was significantly affected by the interaction between noise treatment and experimental situation (ANOVA:
F3,157=5.43, p < 0.001; noise: F1,157=1654.12 p < 0.001; situation: F3,157=5.57, p
< 0.001), but not by fish length (F1,157=0.03, p=0.86). There was a greater change in OBR in fish exposed to playback of boat noise than ambient noise in lab conditions (Figure 5a).
The OBR change between pre1&2 and post10 was significantly affected by noise treatment (ANOVA, F1,157=9.92, p=0.002), but not by experimental situation (F3,157=1.30, p=0.278), interaction between the two (F3,157=1.15, p=0.331), or length (F1,157=0.48, p=0.49). There was a greater change in OBR when fish were exposed to boat noise than ambient noise.
The gradient change of OBR over 5 min was significantly affected by noise treatment (ANOVA, F1,157=2043.44, p < 0.001), but not experimental situation
(F3,157=1.96, p=0.123), the interaction between the two (F3, 157=1.82, p=0.145), or length (F1,157=0.60, p=0.44). There was a greater decrease of OBR gradient in fish exposed to boat noise than ambient noise (Figure 5b).
4.3.1.2 Acanthochromis polyacanthus
The OBR change between pre1&2 and post1&2 was significantly affected by the interaction of noise treatment and experimental situation, and by noise
(ANOVA: F2,113=3.77, p=0.03; noise: F1,113=1653.41, p < 0.001), but not experimental situation (F2,113=2.85, p=0.06) or length (F1,113=0.43, p=0.51).
57 There was a greater change in OBR in fish exposed to playback of boat noise than ambient noise in lab conditions (Figure 5c).
The OBR change between pre1&2 and post10 was significantly affected by noise treatment (ANOVA: F1,113=16.14, p < 0.001), but not by experimental situation (F2,113=2.32, p=0.10), interaction between the two (F2,113=1.02, p=0.37), or length (F1,113=0.88, p=0.35). There was a greater change in OBR when fish were exposed to boat noise than ambient noise.
The gradient change of OBR over 5 min was significantly affected by the interaction between noise treatment and experimental situation (ANOVA: F2,
113=4.40, p=0.015; noise: F1, 113=1436.59, p < 0.001; situation: F2, 113=3.54, p=0.032), but not of length (F1,113=0.13, p=0.72). There was a greater decrease of OBR gradient in fish exposed to boat noise than ambient noise (Figure 5d).
4.3.2 Metabolic Rate
4.3.2.1 Chromis viridis
Metabolic rate was significantly affected by the interaction between noise treatment and situation (ANOVA: F3,191=12.51, p < 0.001; noise: F3,191=606.00, p < 0.001; situation: F1,191=30.56, p < 0.001). There was a greater increase in metabolic rate in fish exposed to playback of boat noise than ambient noise in the field (Figure 6a).
4.3.2.2 Acanthochromis polyacanthus
Metabolic rate was significantly affected by the interaction between noise treatment and situation (ANOVA: F2,138=4.77, p=0.01; noise: F2,138=304.1, p <
0.001; situation: F1,138=10.85, p < 0.001). There was a greater increase in metabolic rate in fish exposed to playback of boat noise than ambient noise in lab conditions (Figure 6b).
58 4.4 Discussion
This study found clear similarities in the physiological responses observed to noise in the field and lab based trials across both test species. This is the first time that a four-stage transfer test has been conducted to test the ecological validity of aquarium-based studies on captive animals with respect to the impact of anthropogenic noise. There were some subtler differences between the trials depending on the experimental situations however in all seven experiments testing opercular beat rate (OBR) and metabolic rate (MR) across two species, noise had a significant effect by way of increased OBR and MR.
Results from the MR assessment suggest that experimental situation can have an impact on the physiological response. Lab/field based situation was significant for both species, however the aquarium C. viridis were less affected than their wild counterparts in the lab playback studies (LPA vs. LPE), whereas aquarium A. polyacanthus were more affected compared with the wild type. The differences between situations for OBR were also variable; for example, in the disparity in OBR gradient change between species. It is possible that differences in OBR and MR stem from fundamental species differences, but the inconsistency between which situation drove the strongest response is relatively minor compared to the clear and consistent effect of noise (when compared with ambient controls) across all experiments.
Increase in OBR & MR
Noise treatment clearly affected both the OBR and MR for both study species.
This supports long-established knowledge of anthropogenic noise as a stressor
(Simpson, et al., 2015; Williams, et al., 2015), and the difference in OBR even
10 min into exposure suggests fish perceive atypical auditory stimuli as a potential threat for a sustained period of time. Inconsistencies in OBR and MR
59 (for example, the eventual return to baseline of OBR but the seemingly consistently high MR) are only vaguely understood (see Priede, 1985 for more detail). Though OBR certainly increases with metabolic demand, it is known that fish can regulate oxygen uptake in ways other than through ventilation rate
(Kristensen, 1983): further research is needed to understand this potentially complex relationship. In the case of anthropogenic noise, a sustained response could be beneficial to survival – marine organisms may hear a boat or ship approaching from a distance, and stay alert until they can be sure they are no longer at risk (a response already observed in marine mammals; Ridgway, et al., 2006). Although potentially keeping the fish out of danger, this physiological state of alertness requires energy to be redirected from elsewhere (for example, appetite suppression and food digestion; Bernier, 2006) and when prolonged may be detrimental to the organism (Van Weerd & Komen, 1998). This theory is supported by the high MR rate in test conditions. It is worth considering using intermittent respirometry for future trials, a system that combines the best aspects of closed respirometry (as in this study) while reducing some of the problems that are associated with flow-through respirometry (Svendsen, et al.,
2016). This method would allow further insight into the physiological processes and direct comparison with OBR fluctuations over time.
Species differences
C. viridis and A. polyacanthus trials returned slightly different results for a small amount of tests. For the difference in OBR between pre1&2 and post1&2, experimental situation did not have a significant effect on A. polyacanthus
(although p = 0.06) and for gradient change of OBR there was no significant effect of situation or interaction term for C. viridis, whereas their counterparts did produce statistically significant results. One possible reason for such
60 discrepancies is the lack of LPA trials for A. polyacanthus, which were not completed due to logistical constraints in the fieldwork. The resulting mismatch of datasets for the two species could explain the slight differences in results (i.e.
N = 4 vs. N = 3 experimental situations), and if included, could strengthen the subtler comparisons that are currently different but not significant. This final set of LPA trials for A. polyacanthus would be straightforward enough to carry out at a later date, and could serve to make the data and corresponding findings complete.
There were also some differences between the species in the assessment of
MR. In ambient conditions, the baseline MR of C. viridis was higher than that of
A. polyacanthus. While movement was not officially quantified, general observations were that fish were predominantly sedentary throughout trials; therefore physical energy expenditure does not explain these differences. The
A. polyacanthus (7.35 ± 0.11 cm SE) were larger than C. viridis (4.79 ± 0.05 cm
SE) and, although MR calculations took the weight and length of fish into account, it is possible that the larger size corresponds to a more mature fish with advanced regulation over physiological functions (Pottinger, et al., 1995).
Based on average size, many of the C. viridis could still be in a late juvenile phase and may show different physiological responses to stressors at different stages in the life cycle. Previous work has found variance in MR among different damselfish species, especially between species from differing habitats and with different feeding habits (Clarke, 1992; Litsios, et al., 2012). It would be valuable to replicate these experiments with other species to develop a more generalised understanding.
61 Disparity in field and laboratory results
Though there may not be immediate cause for concern, the findings from this study do show that fish respond differently to anthropogenic stressors in a laboratory compared to a real-life environment - it is now important to consider any further variables that may have had an influence throughout the trials. The results suggest that laboratory studies may over exaggerate the stress response (especially the case for OBR) by 10–20%, which could lead to overly zealous conclusions and mitigation strategies. Although external variables were tightly controlled, the difference in the stress response could stem from the additional stress of being in an unfamiliar environment in the laboratory, and could also be influenced by insufficient holding and acclimation time leading up to trials. For those fish in field trials, boat noise was predominantly the only stressor (though confinement for metabolic rate measurements may have contributed as a stressor, this was equal across all experiments) and their acclimation and holding took place in a familiar field setting, with authentic ambient background noise, whereas trial subjects experienced confined tanks on a flow through system and playback of ambient noise.
For laboratory trials, tank acoustics were unavoidably different from the field set-up, and although the recording of boat noise was taken from the same distance as fish in field trials, it should be remembered that playback levels within a tank come with complications (OGP, 2008) and could lead to a heightened stress response. Reassuringly, there is little difference between fish that experienced real boat noise and those that experienced playback of boat noise in the field. This would suggest the effects seen in response to playback tracks are representative of the real life stressor, and that the addition of noise within the tank (with reflections of sound waves and vibrations to consider from
62 the speaker) is what is influencing the fish stress response. Furthermore, there was no difference in OBR change between pre1&2 and post10 for either species in any situation – this suggests that any adverse effects from laboratory trials only have an impact initially, and the overall response over time is unchanged.
Finally, aquarium trade fish seem to respond to noise more strongly when compared to wild-type fish responses. Their maturation in a different setting and subsequent inexperience with anthropogenic noise and regular stressors is the most likely explanation for the variance in their stress response. In all trials that took place in Australia, wild-type fish were not naïve to noise. However, this is beneficial as it best represents how fish will respond and react in the natural environment. These findings highlight the importance of using species appropriate to the experimental aims, and allowing for adequate acclimation time to housing and experimental set-up.
Further study
As previously mentioned, this study would benefit from further research to explore the relationship between OBR and MR and examine how they can influence each other but also differ from each other. Completing LPA trials for A. polyacanthus would also fill the missing gap and allow for resolute comparisons between the field and the laboratory experiments, and wild type and aquarium trade fish in both species. Addressing certain experimental shortfalls, for example allowing further holding/acclimation time, or ensuring all fish are fully mature, would help develop this work. Replicating all trials with further damselfish species, although not pressing, would allow further insight into the differences in metabolic rate – perhaps with intermittent respirometry to gain better resolution into MR fluctuation over time. Finally, having established a
63 difference in field and laboratory studies, it would now be beneficial to spend the time looking at some primary physiological responses – for example, cortisol levels over time. Understanding these most fundamental stress responses, and how they differ over time and in different scenarios, will strengthen our current knowledge and application of findings with regards to anthropogenic stressors.
Concluding comments
Although it could be the case that discrepancies between field and laboratory results can be explained largely by differing acoustics and divergently bred fish, these results should not be taken lightly. A huge proportion of marine scientific study takes place in the laboratory, allowing for tight control of extraneous variables and reduced experimental costs, but these findings show that the results from such studies should also receive some validation in the field. These data can perhaps now be available as a reference on which we can base and develop our understanding of real-life environments. Moving forward, it is vitally important to consider the degree to which a laboratory study replicates the natural environment, and take these results into account when considering extrapolation and wider application of laboratory-based findings. Laboratory work contributes a great proportion of information to scientific research, and is by far the most practical method to carry out experiments that consider, for example, long-term exposure to a stressor while still allowing detailed monitoring and tight control of variables. It is neither practical nor necessary to suggest the dismissal of laboratory study findings (Slabbekoorn, 2016), but rather it is important to remember to bear in mind potential discrepancies and the subsequent wider ecosystem implications.
64 5. Condition-dependent responses to anthropogenic noise
5.1 Introduction
Organisms continually adapt to survive. Global and local changes in the environment, due to both anthropogenic and biological drivers, are forcing adaptation of taxa in three directions – either individuals may alter their behaviour and physiology to acclimate to new conditions and remain in the same locations, they may move elsewhere to find more suitable habitat, or they may adapt genetically over time to attempt to continue to survive. Failure to follow one of these three options may lead to a struggle to survive and the possibility of facing extinction. In addition to many chronic pressures (e.g. ocean warming, ocean acidification), marine animals often find themselves simultaneously facing acute stressors such as predation or having to find adequate food. The combination of these acute and chronic stressors has been considered indirectly in a variety of literature (Matthaei & Lange, 2016), but such studies tend to employ the acute problems merely as a tool to measure overall organism response to the chronic. This study will directly test how two temporally variant stressors can interact, and discuss how these context- dependent stress responses can impact on life history.
A Context-Dependent Response
Efficient and appropriate stress responses are vital for any organism, and the ability to judge a stressor and respond suitably can often mean the difference between survival and mortality (Mesa, 1994). Interestingly, many marine organisms exhibit behaviours symptomatic of habituation when exposed to chronic stressors (Cyr & Romero, 2009), suggesting that over time they regulate their response to a perceived threat – though it is important to bear in mind there could be other explanations for these ‘habituation’ responses. This is a 65 particularly important consideration for anthropogenic noise – when exposed to noise, fish such as the lined seahorse (Hippocampus erectus) will become immediately stressed but will start to regulate their hormonal response over time, provided the stressor (in this case, noise) remains constant (Anderson, et al., 2011). Unfortunately, organisms in today’s oceans face multiple threats, both abiotic and anthropogenic, and research that considers a response to one stressor at a time is (although undeniably useful in it’s own way) not wholly applicable in a ‘real-life’ situation. When multiple stressors within a single environment interact, it may be the case that organisms face a trade-off between which they perceive as the most (or least, in terms of regulation) threatening at that time. More specifically, a stress response of an individual will depend largely on the broader environmental context within which the stressors are presented. Context-dependent stressors have been found to impact organism growth (Pottinger, 2006) and schooling behaviour (Hoare, et al.,
2004), in addition to longer lasting effects on juvenile size and physiology
(McCormick, 2006, Mesa, 1994). Variance in environmental conditions, for example oxygen pressure, can also induce a stressful response, alongside reduced performance in swimming (Herbert & Steffensen, 2005) and displays of aggression (Verbeek, et al., 2007). Failure to perform well in the above tasks carries knock-on implications for survival – for example, delayed responses to schooling may leave an individual more vulnerable to predation. Understanding context-dependent responses is essential to predict population-level impacts of multiple stressors, and the scope for adaptation to changing environments.
The Stress Response
The stress response in fish can be defined in terms of primary, secondary and tertiary responses (Barton, 2002). The primary response, generally quantified
66 as the beginning of the hypothalamus–pituitary–interrenal (HPI) axis (Bernier &
Peter, 2001), ends with the release of catecholamines and subsequently corticosteroids (most commonly epinephrine and cortisol respectively). These
‘stress hormones’ allow the organism to deal with a stressor effectively by redirecting energy from non-essential activities, such as growth and reproduction, at that point in time (Bonga, 1997). If this response is sustained, however, it can be detrimental: continuously elevated cortisol levels can impact on disease resistance (Pickering & Pottinger, 1989), growth and reproduction
(Iwama, et al., 2011). Similarly, a secondary stress response such as altered metabolic rate can allow the organism to enter anaerobic respiration for a burst of energy to escape the threat (Steffens, 1989). However, over a longer time period, anaerobic respiration is not sustainable and a consistently elevated metabolic rate can have knock-on affects for blood-sugar concentrations, erythrocyte content and ATP levels (Wells, et al., 1984) - the balance of which are vital to maintaining homeostasis. Finally, tertiary responses (as outlined by
Barton, 2002) can persist in the presence of a consistent stressor. Changes to growth and development, for example, are cascading effects from primary and secondary responses such as elevated cortisol, consistently high catecholamines and imbalanced blood chemistry (Van Weerd & Komen, 1998).
The effects of stress can clearly impact on life history, with consequences for future generations (Schreck, et al., 2001). Thus, it is important to consider how organisms may respond to different stressors, and whether current marine organisms are able to recover from, or tolerate sub-optimal performance, when exposed to various stressors.
67 Reduced Diet
Current anthropogenic stressors are impacting on food resources in multiple ways, from potential prey species shifting out of range (Booth, et al., 2011), through invasive or alien species outcompeting for resources (Albins & Hixon,
2013), to the mistaken ingestion of marine plastics (Rochman, et al., 2013).
When considering the many other stressors marine organisms currently face, struggling to find adequate food could pose a chronic and persistent problem to many animals. Though the detrimental effects of a reduced or manipulated diet are well-established, a dearth of literature currently exists on the direct interaction between a suboptimal diet and an additional anthropogenic stressor.
Inadequate food resources play a key role in the death of larval stage fish (Platt, et al., 2003), and can indirectly affect fish mortality through stunted growth in early stages (Lasker, et al., 1970). Deficiency of certain vitamins has been found to impact on fish life histories in different ways: a lack of rhodopsin (which stems from vitamin A) has been linked with insufficient melanin production and subsequent albinism (Nakamura, et al., 1986), insufficient vitamin C can affect larval resistance to infection (Merchie, et al., 1997), and insufficient essential amino acids (linked with protein in the diet) can impact growth and development
(Ronnestad, et al., 1999). With such clear detrimental effects, it is likely that malnourished fish subjected to environmental stressors, such as anthropogenic noise, will fare much worse than well fed counterparts.
Anthropogenic noise
Anthropogenic stressors pose a serious threat to marine fish communities and fisheries around the world (Halpern, et al., 2007). Anthropogenic noise is a relatively understudied stressor, despite potentially significant adverse impacts on fish and global fisheries. Noise from road and air traffic and construction is
68 known to have negative impacts on many terrestrial systems, overwhelming the auditory space of terrestrial animals and affecting critical processes spanning communication to predator avoidance (Francis, et al., 2009). In the higher density medium of the marine environment, sound propagates further and five times faster than in air (Urick, 1996), bringing with it the potential for much more significant impacts.
Only relatively recently has marine anthropogenic noise been identified a potential major stressor to fish, despite parallels with many terrestrial anthropogenic noise sources, such as transport (by ships) and construction
(using pile-driving). Other noise derived from underwater explosions, smaller boat noise (through fishing, recreation and tourism) and sonar (Hildebrand,
2009) can bring further harm to aquatic life through shock waves and temporary hearing loss. Anthropogenic noise has been shown to affect marine organisms at all stages of the life-cycle; it masks the natural noise of a reef, impeding the larval stages of fish and crustaceans from recruiting (Simpson, et al., 2005;
Jeffs, et al., 2003) and masking communication between marine organisms
(Vasconcelos, et al., 2007). Furthermore, noise can negatively impact fish through disruption to foraging routines (Miksis-Olds, et al., 2007), habitat abandonment (Weilgart, 2007), and through direct physiological impacts
(Simpson, et al., 2016; Simpson, et al., 2015).
Anthropogenic noise is now generated on a global scale, with cargo shipping rates expected to increase 2–3 times by 2025, and development of offshore renewable energy causing significant acoustic disturbance during construction of wind, wave and tidal turbines (Hawkins, et al., 2015). All of this noise adds to the aquatic soundscape, potentially masking natural sounds, with direct behavioural and physiological impacts that could lead to ecosystem-wide
69 impacts (Slabbekoorn, et al., 2010). Though much is understood about the prevalence and impacts of anthropogenic noise, very little is known about how it’s impact is affected by other stressors. As seen, lack of dietary resources is a growing problem in today’s oceans, and if fish are undernourished when they are exposed to noise, they could potentially suffer more severely, with implications for survival.
Damselfish (Pomacentridae) comprise up to 50% of the biomass in coral reef fish communities and are major contributors to the overall productivity of a reef ecosystem (Klumpp, et al., 1987). Previous work using physiological assays has established responses to anthropogenic noise in several damselfish species
(Simpson, et al., 2016). In a recent study on European eels (Anguilla anguilla) a correlational effect was found in the impact of noise on fish with differing levels of condition (Purser, et al., 2016) (indicating differences in feeding success)
(Jones, 1986); this study examined the interaction between a stress response and relative differences in well-being: combining a sub-optimal diet (with associated loss of condition) and anthropogenic noise over time. This experiment focused on the spiny chromis Acanthochromis polyacanthus, (a species that is easy to capture and house in captivity at Lizard Island (Great
Barrier Reef)) and manipulated fish diet over time to test for a varying stress response to anthropogenic noise. Experimental methodology employed two measures of secondary stress responses: opercular beat rate (OBR) and routine metabolic rate (MR). Monitoring OBR and MR allows non-invasive methods to test for a stress response, and is a method that can be easily adapted for use in field situations, allowing experimental exposure of fish to real sources of marine anthropogenic noise (motorboats). Not only are motorboats
70 readily available and easy to manipulate (when compared to ship and pile- driving noise, for example), they represent the noise A. polyacanthus are most likely to regularly be exposed to. This study tested the hypothesis that all fish would show an increase in OBR and MR (H1 and H2) in response to motorboat noise due to stress, and that there would be no difference in baseline OBR and
MR for fish tested in ambient noise conditions (H5 and H6 respectively) as fish would become acclimatised to modified dietary conditions. Experiments also tested the hypothesis that fish maintained on a reduced diet would exhibit a heightened stress response in OBR and MR (H5 and H6 respectively) when exposed to anthropogenic noise compared to fish on full diets.
5.2 Materials and Methods
5.2.1 Test subjects
Adult A. polyacanthus (mean ± SE standard length: 7.80 ± 0.09 cm; mass:
18.81 ± 0.65g) were collected by SCUBA divers with hand and barrier nets on shallow reefs (3-6 m) around Lizard Island Research Station (LIRS; S 14° 40′, E
145° 28′), Northern Great Barrier Reef, Australia. Subjects were housed in the flow-through aquarium system at LIRS, in six plastic tanks (45 x 67 x 62 cm, water depth 44cm). All tanks were enriched and given shelter with equivalent pieces of dead coral and plastic tubing. All husbandry and handling was in line with Australian standards as outlined by the Great Barrier Reef Marine Park
Authority (GBRMPA), and all work carried out with permission and ethical approval from: Lizard Island Research Station, Great Barrier Reef Marine Park
Authority, James Cook University (A2081), Australian Institute of Marine
Science and University of Exeter (2013/247).
71 5.2.2 Diet manipulation protocol
This experiment tested the effect of two diet conditions: high fed and low fed
(hereafter referred to as diet HF and LF). To track change in condition, initial fish length and mass were measured, and fish were individually tagged using elastomer before undergoing diet manipulation. Fish were then split equally
(N=24) across six tanks (three each for diet HF and LF treatments) and were maintained on the two diets for 10 days. Fish in diet HF were fed ad libitum three times daily; fish in diet LF were fed 3–5% of body mass daily (minimum recommended amount for experimental fish size and weight), distributed over three feeding sessions. All fish were fed a diet of fish feed pellets (Perla MP,
Skretting, Verona, Italy). Fish were re-measured (length and mass; as above) at the end of the 10-day diet manipulation.
5.2.3 Boat noise protocol
In all trials, fish were exposed to real boat noise on Casuarina Beach, LIRS.
Ambient reef noise was used as a control condition. To reduce pseudoreplication three different research station boats (5-m-long aluminium hulls with 30 hp Suzuki outboard motors) were used; boats were driven 40–50 m away. Boat and ambient-noise recordings were taken from a kayak moored using an anchor without a chain to avoid unwanted noise and were made 1m above the seabed for 5 min. Acoustic pressure was recorded using an omnidirectional hydrophone (HiTech HTI 96-MIN, inbuilt preamplifier; manufacturer calibrated sensitivity -164.3 dB re 1 V/µPa; frequency range 2 Hz to 30 kHz, calibrated by manufacturers, High Tech Inc., Gulfport MS) suspended in the centre of the trial area and a digital recorder (PCM M10,
48kHz sampling rate, Sony Corporation, Toyko, Japan). Particle acceleration was measured using a calibrated triaxial accelerometer (M20L; sensitivity
72 following a curve over the frequency range 0–3 kHz; calibrated by manufacturers; Geospectrum Technologies, Dartmouth, Canada) and a digital
4-track recorder (Boss BR-800, 44kHz sampling rate, Roland Corporation, Los
Angeles, CA). Recording levels used with each set-up were calibrated using pure sine wave signals from a function generator with a measured voltage recorded in line on an oscilloscope.
Recordings were analysed using audio software Audacity (version 2.0.5 for Mac
OSX), Avisoft SASLab Pro Software (version 5.1.17, Avisoft, Berlin, Germany), and MATLAB (v2010a, MA, USA). Fast-Fourier Transformation was used to calculate power spectral density for comparison of sound levels for each treatment across the frequency range 0–3000 Hz (Fig. 3). Spectrograms of examples of the two noise conditions are also shown for comparison (Figs. 4a- b).
5.2.4 Experimental protocol
All experiments took place at LIRS from October to December 2014. Single fish were placed in individual sealed experimental containers (up to six individual containers in any one experiment, 1 l, 115 x 115 x 160 mm, Sistema Plastics) painted with white paint to remove external visual disturbances but still allow light to enter. Care was taken to reduce exterior noise; activity not pertaining to the experiment took place a minimum of 300 m away (Simpson, et al., 2016). In all trials, fish were given a minimum of 30 min to recover from handling stress and to acclimatise to their container; during this time containers were left open and water refreshed regularly to avoid oxygen depletion. Fish experienced ambient noise during acclimatisation. The container was then transferred to the experimental area in the field, where water was refreshed one final time and the container then sealed. After each trial, fish length and mass was measured to
73 calculate condition and control for mass in MR calculations; fish were then transferred to a separate holding tank for release, and the container was washed out in preparation for the next trial.
5.2.4.1 Opercular Beat Rate
This field-based experiment was designed to allow assessment of effects of motorboat (test condition) noise on OBR, with ambient noise used as a control condition. A total of 89 fish were used (noise/diet HF, n = 23, ambient/diet HF, n
= 22, noise/diet LF, n = 22, ambient/diet LF, n = 22). Variability in n for OBR stems from technical errors when filming fish. In each trial, single fish were sealed into individual containers (see 2.4) and were exposed to either motorboat (test) or ambient (control) noise for 20 min. Fish were filmed for the duration of the trial using video cameras (GoPro HERO 3, 3+, 4, GoPro Inc.,
California, US) facing inwards from the lid of the container.
5.2.4.2 Metabolic Rate
Concurrently with OBR, this experiment also tested for an effect of boat noise on (MR). To avoid variation in MR due to digestion (Jobling, 1981), fish were starved for 12 h before experiments began. A total of 92 fish were used
(noise/diet HF, n = 23, ambient/diet HF, n = 23, noise/diet LF, n = 23, ambient/diet LF, n = 23). In each trial, single fish were sealed into individual containers (see 2.4) and exposed to either motorboat (test) or ambient (control) noise for 20 min. After 20 min, the container was removed from the field, and the lid was taken off. The fish was immediately removed from the tub and the oxygen concentration of the water was measured using an oxygen probe
(Dissolved Oxygen and Temperature Meter HI 9164, Hanna Instruments Inc.,
Woonsocket, RI), and compared to the oxygen concentration of the seawater
74 used to fill the containers at the start of the experiment to determine oxygen uptake.
5.2.5 Statistical Analysis
All statistical analyses were conducted using SPSS version 22.0 (IBM Inc., New
York, US).
5.2.5.1 Diet manipulation
Fulton’s Condition Index1 (Fulton, 1904) was used to calculate the condition of individual fish before and after diet manipulation, with paired-samples t-tests used to test for any change in condition following manipulation for fish in LF diet, and for change in condition for fish in HF diet. There was no significant difference in fish condition between the two groups before diet manipulation began (independent-samples t-test: t88 = 0.764, df = 86, p = 0.447).
5.2.5.2 Opercular Beat Rate
Trial video recordings were exported and analysed without sound (to avoid observer effects during review) in QuickTime version 10.4 (Apple Inc.,
California, US). Opercular beats were counted using a clicker for 15-s subsections for the initial 5 min of exposure, then for 15-s subsections at 10 and
20 min – preliminary experiments showed the greatest change on OBR over the first 5 min, and gradual/little change from 10 min onwards (Armstrong-Smith, et al., unpublished). Mean OBR was calculated for minutes one and two and analysed using an ANOVA, allowing for variance in size (length). The rate of change of OBR (hereafter referred to as gradient change of OBR) was
1 � = ! , Where K = Fulton’s condition factor, M = mass of fish and L = total !! length of fish
75 calculated over the 5 min following sound exposure and analysed using an
ANOVA.
5.2.5.3 Metabolic Rate
Oxygen probe concentration readings were converted to MR values (MO2, mmol/g/hr) for each fish over the 20-min exposure period and analysed in an
ANOVA.
5.3 Results
5.3.1 Effects of condition
LF fish condition was significantly affected following the period of diet manipulation (paired t-test; t44=-4.95, p < 0.001). HF fish condition was not affected (t45=0.23, p = 0.82).
5.3.2 Opercular Beat Rate
Mean OBR over 2 min was significantly affected by noise (ANOVA:
F1,88=459.48, p < 0.001) and diet (F1,88=39.43, p < 0.001) but not by the interaction between the two (F1,88=3.59, p=0.06). There was a greater change in
OBR in LF fish exposed to boat noise than ambient noise (Figure 7).
The gradient change of OBR over 5 min was significantly affected by the interaction between noise and diet (ANOVA: F1,88=30.43, p < 0.001; noise:
F1,88= 389.10, p < 0.001; diet: F1,88=4.88, p=0.03). There was a larger decrease in OBR gradient in HF fish exposed to boat noise than ambient noise (Figure 8,
9).
5.3.3 Metabolic Rate
Metabolic rate was significantly affected by the interaction between noise and diet (ANOVA: F1,91=4.61, p=0.03; noise: F1,91=45.07, p < 0.001; diet:
76 F1,91=14.38, p < 0.001). There was an increase in metabolic rate in LF fish exposed to boat noise than ambient noise (Figure 10).
5.4 Discussion
This study found that anthropogenic noise and a reduced diet both individually affect stress levels, and that the combination of both stressors further exacerbates the stress response. Despite this, it seems that fish may have a
‘stress threshold’, which they do not exceed for a perceived threat such as noise, as both HF and LF diet fish OBR peaked around the same point following the introduction of noise (perhaps the results would be different with olfactory or visual cues). It is interesting that it is noise that differentiates these stress responses – low and high fed fish exhibited the same stress response in terms of OBR and MR in test conditions, however in control conditions baseline stress levels were much higher in low fed fish. This suggests that a poor diet may impact on day-to-day activities, but in response to a threat fish are able to redirect energies as required and react appropriately regardless of physical condition. The field would now benefit from further study into the long term effects of diminished diet, as although short term responses do not seem affected, it is unclear how reduced diet may affect physiological processes, growth, reproduction etc.
Increase in OBR and MR
Opercular Beat Rate (OBR) was similar in test conditions regardless of diet group, but fish in control conditions exhibited different responses dependent on their preceding diet. Furthermore, the regulation of OBR (a potential insight into habituation to such stressors) over time did not differ in test conditions but in control conditions, fish in diet LF were initially more stressed and took longer for their OBR to recover back to baseline levels. After the 20-min exposure period,
77 fish did not differ in OBR, regardless of diet or treatment group, suggesting diet manipulation did not affect baseline stress levels but merely the ability to respond to new stressors. By contrast, MR did differ according to diet group, and those in diet HF had a lower MR in both test and control conditions, with the difference amplified in the noise treatment. These seemingly conflicting
OBR and MR results suggest that fish are undergoing an underlying physiological response with regards to differing diets. Various studies have attempted to quantify the relationship between OBR and MR (Lucas, 1994; van
Rooij & Videler, 1996), and Dalla Valle et al. (2003) highlighted the various ways in which fish can regulate oxygen uptake other than through ventilation rate. This may explain how fish respiratory rate seems to decrease but MR stays high when exposed to certain anthropogenic noises, although further research would be needed to better understand this relationship with regards to a stress response.
Previous studies that have manipulated, reduced or substituted essential components in fish diets have noted changes to oxidative stress responses
(Pascual, et al., 2003), and baseline plasma cortisol (Montero, et al., 2003) and glucose levels (Vijayan & Moon, 1992). Vijayan & Moon (1992) also found that food-deprived fish mobilized high levels of glycogen stores to accommodate the energetic demands of a stressor. This supports the above theory of an underlying physiological change and indicates that fish redirect energy stores to deal with the most immediately threatening perceived stressor. The low fed group would have had fewer glycogen stores following a period of food depravation (Van Dijk, et al., 2005) and may therefore sustain a higher MR to meet the energetic demands. Further study, perhaps employing intermittent
78 respirometry could yield better answers as to the metabolic rate and regulation over time in response to different stressors.
Habituation to stressors
The nature of closed respirometry makes it difficult to comment on potential changes in MR during the 20-min exposure, however monitoring OBR throughout the exposure period provides insight into how a fish may change its response to the acoustic stressor over time. In this study, OBR returned to a baseline level by the end of the 20-min test period in all treatment groups regardless of diet and noise exposure, thus OBR seems to be continually regulated over time. Two plausible explanations for this would be either fish are unable to sustain an increased OBR for a prolonged amount of time, or fish show some habituation to the noise. Although previous work has linked fatigue with mortality (Wood, et al., 1983), Wood’s study considered severe exercise over a much longer time period. While swimming behaviour was not quantified, fish were predominantly sedentary throughout trials - making it unlikely fish in this experiment reduced OBR to avoid imminent risk. Habituation, however, could be termed as the attribution of anthropogenic noise to an acoustic setting
(Wahlberg & Westerberg, 2005), and despite potential benefits involved with suppressing stress levels (and thus short-lived negative physiological effects), could be a disadvantage in the long term. The grouping of sounds within the acoustic scene (Fay, 1998) may leave fish less able to detect other predator cues, and prolonged exposure can lead to reduced hearing sensitivity and even hearing loss (Smith, et al., 2003). Furthermore, the allostatic load from continuous stressors (such as noise or reduced diet) can impair fish ability to carry out routine tasks (Schreck, 2000) and although fish may appear to have returned to a baseline level in their physiological condition, this does not
79 necessarily equate to behaviour being unaffected (Purser & Radford, 2011). It would be beneficial to study the physiological response to multiple stressors over a longer time period to determine the cause of varying OBR and MR, and tease apart the likelihood of habituation, tolerance, and physical incapacity through fatigue.
Implications of findings
The physiological toll of prolonged stress, alongside the potential desensitization of organisms to important acoustic cues (e.g. of predators and conspecifics), indicates potential impacts on fitness. With the added pressure of food-deprivation (where, even in ambient conditions, under-fed fish showed physiological impacts compared with their well-fed equivalents), it is clear that the individual behavioural and physiological responses of fish to these stressors are exacerbated when they are combined.
As previously highlighted, assessing context-dependent responses will help in understanding how basic processes interact, and will help explain if (and how) animals prioritize their responses to different stressors. The results here suggest fish prioritize regulating the stressor with the most immediate perceived threat, in this case noise. This is supported by the similar return to baseline
OBR in both diet groups.
Much recent research has explored effects of a range of individual stressors, and tested how, in more extreme cases, these stressors select for tolerant species, thus changing the abundance and distribution of species within communities. These knock-on effects can propagate through ecological networks and ultimately affect entire ecosystems (Breitburg & Riedel, 2005).
The effect of multiple stressors can be far greater than the additive effect of each individual impact; with each change at the physiological or ecological level
80 increasing the severity of the impact of the next stressor, with some responses and recoveries (in particular in the endocrine system) being delayed, with knock-on effects that combine with any subsequent stressors (Peterson, et al.,
2003). Here, it is likely that the effects of diet manipulation on the physiological system exacerbated the effect of the next stressor – in this case, the increased
MR in test conditions.
Although all experimental trials are partially limited by their very nature in terms of how to extrapolate implications to the real world, every effort was made here to improve the scope for extrapolation. For example, real boats were used to generate noise (therefore the response measured is an accurate representation of the wild-type fish) and although the reduced diet aspect is not wholly characteristic of coral reefs, there are factors that threaten reefs, which can ultimately reduce resources and food availability (Fabry, et al., 2008; Lo-Yat, et al., 2011). It is therefore highly likely the stressors employed in this study will interact and impact on marine life. The chronic nature of anthropogenic noise
(especially boat noise) and environmentally-degraded ecosystems therefore pose a real threat to marine organisms, and could cause impacts in both immediate and longer-term timescales,
Limitations of the study
Some of the difficulties often encountered in the extrapolation of data and application to real-life situations have been avoided here, by conducting the study in the field and by using wild fish. Fish were not naïve to noise (though each individual was only exposed to the trial set up once), but this actually represents real-life situations more accurately as present-day fish in the oceans will have been exposed to many sources of anthropogenic noise from the very earliest stages of their development.
81 In this experiment, fish were subjected to handling stress, and fish were given
30 min for acclimatisation to the experimental conditions. Some studies have found recovery from handling stress time to vary from a few hours (Artigas, et al., 2005) to two weeks (Pickering, et al., 1982). It is possible therefore that fish had not made a complete recovery from handling before being subjected to noise. When studying the stress response, it is challenging to design an experiment that does not impact on an animal’s stress levels, and while care was taken here, there remain elements within the experimental design that may influence the stress response. Ultimately, all fish received the same handling and acclimation so any negative effects of incomplete recovery should have been equal across all subjects and the impact been controlled for in the experimental design, such that any stress responses observed are likely to be conservative compared to those from fish in natural conditions.
Finally, a number of extraneous variables were controlled for (e.g. recovery from handling, effect of external visual disturbances on OBR and MR, effect of food on MR). It would be logistically very difficult to carry out such experiments on fish in their natural environment without introducing some external variables, and in doing so one would lose the opportunity to measure the stress response in response to carefully controlled exposures. One alteration the study would benefit from, both methodically and to further scientific understanding, would be the use of intermittent respirometry (Svendsen, et al., 2016). Advancements in the field will facilitate this soon, as at present the equipment needed for intermittent methods is difficult to move into the field, and the noise from the equipment further confounds any studies involving anthropogenic noise.
82 Future research
This study would greatly benefit from supplementary work to understand the interaction between OBR and MR, and how they change over time. Considering how these stressors interact over a longer time period would offer more insight into the responses to threats of varying danger. Testing of primary responses, in particular corticosteroids, would offer a higher level of detail about the processes underpinning the stress response.
It would be interesting to consider the results of this study in the context of a species with social hierarchies, for example the clownfish (Amphiprion spp.), where a struggle for food can occur naturally. Plasma cortisol levels can vary between subordinate fish and their dominant counterparts (Overli, et al., 1999)
(Pottinger & Carrick, 2001) – this, when coupled with sub-optimal levels of food could invoke a similar response to noise as seen here.
Future work within this field should now look to identify appropriate additive stressor effects (i.e. tease apart individual effects and then consider their interaction), and consider impacts and interactions on multiple sensory modalities to gain a more wholesome ecological understanding of the impact of stressors and anthropogenic pollutants (Halfwerk & Slabbekoorn, 2015). With this, it will be possible to ease stressors faced by marine species through focusing on individual stressors that are more straightforward to regulate. For example, it would be very difficult to ensure each fish is well fed, but it is easy to reduce anthropogenic noise in areas of high environmental pressure.
Concluding comments
This study supports the knowledge that fish are affected by anthropogenic noise, and provides robust evidence that food deprivation in marine fish alters
83 the stress response. Although theories to explain the variance in OBR and MR have been considered, more research is needed in this area before unequivocal conclusions can be made. The increased fitness demands of heightened OBR and MR can be severely detrimental to an organism, and the potential exaggeration of this when in low-fed conditions should be of great concern.
These impacts could scale up to population-level effects in a broad range of taxa, and ecological, social and economic implications for reefs should not be underestimated. Looking ahead, it is important now to delve deeper into the interactions of multiple stressors on a more basic physiological level, as this information could help counteract stressors and relieve some of the pressure currently inflicted on marine ecosystems.
84 6. Conclusion
Before embarking upon reflection of these findings, let us first revisit the broader research aims: to study the stress response of coral reef fish to anthropogenic noise, and to test whether these responses to noise are context- dependent, that is, do they differ with changing environments and external factors. As has been discovered, while qualitatively similar this research found that there were some differences between results from laboratory based studies and those from field work. It is possible and indeed probable that other factors are influencing these data, but this study highlights that while laboratory experiments can provide useful indicators of responses in the real world, caution should be taken as field and laboratory work proffer differences, which could lead to diverse conclusions and subsequent incorrect extrapolation and assumptions. Further to this, experiments found that reduced availability of food resources (and associated lack of fish condition) can influence the stress response in fish exposed to anthropogenic noise – quantifying, for the first time, the pronounced negative consequences for fish in sub-optimal dietary circumstances that are then exposed to anthropogenic noise.
The experimental approaches used to assess the impacts of anthropogenic noise within this work ranged from field experiments with real noise sources and wild fish, to laboratory studies using playback of noise and fish sourced through the aquarium trade. The varied stress response across all experimental groups, both species, and both studies, highlight the importance of considering the context within which the stressor (in this case, anthropogenic noise) has been presented. The first study tested the validity of experimental testing, comparing data from fieldwork and laboratory work. Although the stress response was generally comparable – all fish had an elevated stress response to noise, and
85 all fish seemed to regulate this response over time – certain differences, such as the initial strength of the response or the time taken to regulate physiological functions differed, and could cause data to be misinterpreted, with the stress response either exaggerated or underestimated dependent on which data are being considered. As the bulk of anthropogenic noise research to date has been conducted in a laboratory setting, forthcoming research should now consider the implications of these findings, and potentially utilize these data as a baseline for reference and comparison in further trials. Anthropogenic noise is a mounting issue (Burke, et al., 2011), and the ongoing debate for it to be considered within global framework for change means that it has never been more vital to accurately report and understand its impacts.
The second study within this work considered the stress response induced by anthropogenic noise in conjunction with another stressor, specifically dietary conditions. This combination of two such pertinent issues goes some way towards addressing current gaps in our understanding of anthropogenic impacts. Research thus far has focused largely on the impact of a single stressor at any one time, but in reality marine organisms are subject concurrently to multiple stressors, meaning it is vital for us to understand how they interact and impact upon marine life histories before we try to mitigate these pressures. The experiments here suggest that a sub-optimal diet can induce an exacerbated effect on the stress response and negatively influence the regulation of physiological function in response to the stress of anthropogenic noise. Future experiments are needed to study this in more detail: to tease apart the influence of individual stressors and perhaps try to establish the ‘stress threshold’ where an organism’s ability to regulate their stress response becomes impaired.
86 The data from both studies show an overall diminished stress response over time despite continual exposure to an anthropogenic noise source. This suggests fish can gain toleraance to noise, such that over time a stressor such as boat noise would have potentially less of an impact. This is certainly something to explore in the future; but it remains unknown whether the potential for adaptation and habituation is positive or negative for coral reef inhabitants
(Bejder, et al., 2009). It should be noted that a diminished external response does not necessarily correlate to a reduction in fitness consequences, as it is well established that noise even at a most basic level can act as a distraction and reduce survivorship in predator–prey interactions (Simpson, et al., 2016).
Future research should look towards testing the scale of impact anthropogenic noise both independently, and combined with multiple stressors, can have on a range of coral reef species at key life-history stages. Elevated stress levels can impact development in larval stage fish (Schreck, et al., 2001), and knock-on effects have been seen in progeny of fish subject to stressful environments
(McCormick, 2006) - both of which could stem from anthropogenic impacts. The implications of these findings, especially in coral reef areas subject to high motorboat use (for example, the Great Barrier Reef), could be particularly grave for fitness consequences and overall demography of reef fish populations.
Both studies highlight how anthropogenic noise in the marine environment has the potential to impact fish physiology and in turn, influence coral reef fish fitness. A range of stressors (far wider than considered within this body of work) threatens coral reefs, but reefs continue to play an important global role, both economically and socially. Coral reefs generate revenue for many countries through tourism, and are the main source of both food and livelihoods by way of fisheries for over 500 million people (Burke, et al., 2011). It has never been
87 more pertinent to effectively manage both worldwide and localized stressors if coral reefs are to survive the destructive forces of global climate change predicted fpr the coming decades. With many issues (such as coral bleaching) being largely outside of our control (Ainsworth, et al., 2016), it is even more important to include anthropogenic noise – a comparatively easy stressor to regulate and mitigate – in environmental management plans.
This work highlights the importance of considering the context of a stress response, and goes someway to addressing key knowledge gaps in our understanding of both anthropogenic noise and overall responses of fish to stressors. By testing and comparing different methodologies, this research attempted to reiterate and reinforce the value of reflecting at every stage of research, and not just accepting finding from contemporary studies at face value. By combining multiple stressors, experiments tried to better emulate the challenging situations that marine life finds itself facing on a now unrelenting basis. The global climate changes predicted for our oceans paint a grim picture where marine life will be forced in one of two directions: adapt to remain resilient or diminish. Rachel Carson, one pioneer of the global environmental movement, puts it rather eloquently: ‘The more clearly we can focus our attention on the wonders and realities of the universe about us, the less taste we shall have for destruction.’ It is now time for the global population to take action and contribute to saving our oceans from anthropogenic threats – otherwise we are looking at a very sorry future indeed.
88 7. Appendices
Table 1. Sample sizes for individual experimental groups, for Opercular Beat
Rate and Metabolic Rate experiments.
Opercular Beat Rate
C. viridis A. polyacanthus
Test Control Test Control
FBA 18 18 17 18 FPA 22 20 19 20
LPA 21 17 - -
LPE 21 21 19 20
Metabolic Rate
C. viridis A. polyacanthus
Test Control Test Control
FBA 24 24 23 22
FPA 24 24 22 24
LPA 24 24 - -
LPE 24 24 24 24
89 A
B
Figure 3. Spectral content shown in both (a) acoustic pressure and (b) particle acceleration domains for original field recordings of boat noise and ambient conditions, playback of ambient and boat-noise recordings in the field, and playback of ambient and boat-noise recordings in experimental tanks. Mean power spectral density of 1 min of each sound condition is shown. Sounds were analysed in MATLAB v2010a, FFT length = sampling frequency (48 kHz for sound pressure and 44.1 kHz for particle acceleration, both result in 1 Hz bands).
90
A B
C D
E F
Figure 4. Spectrogram of (a) field ambient, (b) field boat, (c) field playback ambient, (d) field playback boat, (e) lab playback ambient, (f) lab playback boat noise/noise playback taken over 10 seconds across 22 Hz (FFT analysis: Hann evaluation window, 50% overlap, FFT size 1024).
91 50 50 Test A Test 21 C 21 Control 40 Control 40 19 18 22 17 19
30 30
20 20
10 10 Change in OBR (bpm)
Change in OBR (bpm) 20 18 20 17 21 18 20 0 0 FBA FPA FPA LPA LPA LPE LPE FBA FBA FPA FPA LPE LPE 2 2 18 0 0 18 20 17 21 20 20 -2 -2
-4 -4
-6 -6
-8 -8
22 Gradient Change OBR (bpm) Gradient Change OBR (bpm) -10 18 Test -10 17 19 21 21 B Test Control D 19 Control -12 -12
Figure 5. Impacts of boat noise on opercular beat rate. a) Difference in OBR for
C. viridis between post1&2 and pre1&2. b) Gradient (rate of) change in OBR for
C. viridis for five minutes after exposure. c) Difference in OBR for A.
polyacanthus between post1&2 and pre1&2. d) Gradient (rate of) change in
OBR for A. polyacanthus for five minutes after exposure. Error bars show one
standard error.
92
40
24 35 Test A 24 30 Control 24 25
20 24 15
MR(mmol/g/hr) 24 10 24 24 24 5 0 FBA FPA LPA LPE
16
14 24 Test B
12 23 Control 10 22 8
6
MR(mmol/g/hr) 4 24 24 2 22
0 FBA FPA LPE
Figure 6. Impacts of boat noise on metabolic rate. a) C. viridis. b) A. polyacanthus. Error bars show one standard error.
93 160 22 Test
150 23 Control
140
130 22 OBR (bpm) 120 22
110
100 HF LF
Figure 7. Mean OBR for A. polyacanthus for two minutes post noise exposure, showing motorboat and ambient control conditions in two dietary groups. Error bars show one standard error.
170
160 LF Test
150 LF Control
140 HF Test
130 HF Control OBR (bpm) 120
110
100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (mins) Figure 8. Mean OBR at different time points for test and control groups of A. polyacanthus in two dietary conditions. Error bars show one standard error.
94 HF LF 0
-2 22
-4 22 -6
-8 Test
Gradient Change OBR (bpm) -10 22 Control 23 -12
Figure 9. Gradient change in OBR between test and control conditions for A. polyacanthus in two dietary groups. Error bars show one standard error.
10 Test 9 23 8 Control 7 6 23 5
4 23 MR(mmol/g/hr) 3 23 2 1 0 HF LF
Figure 10. Metabolic rate between test and control conditions for A. polyacanthus in two dietary groups. Error bars show one standard error.
95 8. Bibliography
Acoustical Society of America. (2013). Acoustical Terminology. New York:
American National Standard Institute.
Ainsworth, T. D., Heron, S. F., Ortiz, J. C., Mumby, P. J., Grech, A., Ogawa, D.,
et al. (2016). Climate change disables coral bleaching protection on the
Great Barrier Reef. Science , 352 (6283), 338-342.
Albins, M. A., & Hixon, M. A. (2013). Worst case scenario: potential long-term
effects of invasive predatory lionfish (Pterois volitans) on Atlantic and
Caribbean coral-reef communities. Environmental Biology of Fishes , 96 (10),
1151-1157.
Allen, A. N. (2013). An investigation of the roles of geomagnetic and acoustic
cues in whale navigation and orientation. WHOI Theses.
Amoser, S., & Ladich, F. (2003). Diversity in noise-induced temporary hearing
loss in otophysine fishes. Journal of the Acoustical Society of America , 113
(4), 2170-2179.
Anderson, P. A., Berzins, I. K., Fogarty, F., Hamlin, H. J., & Guillette, L. J.
(2011). Sound, stress, and seahorses: the consequences of a noisy
environment to animal health. Aquaculture , 311 (1-4), 129-138.
Arends, R. J., Mancera, J. M., Munoz, J. L., Wendelaar Bonga, S. E., & Flik, G.
(1999). The stress response of the gilthead sea bream (Sparus aurata L.) to
air exposure and confinement. Journal of Endocrinology , 163 (1), 149-157.
Arroyo-Solis, A., Castillo, M., Figueroa, E., Lopez-Sanchez, J. L., &
Slabbekoorn, H. (2013). Experimental evidence for an impact of
anthropogenic noise on dawn chorus timing in urban birds. Journal of Avian
Biology , 44 (3), 288-296.
96 Artigas, M. L., Skjaeraasen, J. E., Utne-Palm, A. C., & Nilsen, T. (2005).
Recovery from handling stress in Gadus morhua. Journal of Fish Biology ,
67, 384-391.
Atmaca, E., Peker, I., & Altin, A. (2005). Industrial noise and its effects on
humans. Polish Journal of Environmental Studies , 14 (6), 721-726.
Babisch, W. (2014). Updated exposure-response relationship between road
traffic noise and coronary heart diseases: a meta-analysis. Noise and Health
, 16 (68), 1-9.
Bailey, H., Senior, B., Simmons, D., Rusin, J., Picken, G., & Thompson, P. M.
(2010). Assessing underwater noise levels during pile-driving at an offshore
windfarm and it's potential effects on marine mammals. Marine Pollution
Bulletin , 60 (6), 888-897.
Baird, R. W., & Whitehead, H. (2000). Social organization of mammal-eating
killer whales: group stability and dispersal patterns. Canadian Journal of
Zoology , 78 (12), 2096-2105.
Balcomb KC, C. D.-8. (2001). A mass stranding of cetaceans caused by naval
sonar in the Bahamas. Bahamas Journal of Science , 5 (1), 2-12.
Barber, J. R., Crooks, K. R., & Fristrup, K. M. (2010). The costs of chronic noise
exposure for terrestrial organisms. Trends in Ecology and Evolution , 25 (3),
180-9.
Barry, T. P., Malison, J. A., Held, J. A., & Parrish, J. J. (1995). Ontogeny of the
cortisol stress response in larval rainbow trout. General and Comparative
Endocrinology , 97 (1), 57-65.
Barton, B. A. (2002). Stress in fishes: a diversity of responses with particular
reference to changes in circulating corticosteroids. Integrative and
Comparative Biology , 43 (3), 517-525.
97 Basu, N., Nakano, T., Grau, E. G., & Iwama, G. K. (2001). The effects of cortisol
on heat shock protein 70 levels in two fish species. General and Comparative
Endcrinology , 124 (1), 97-105.
Bayne, E. M., Habib, L., & Boutin , S. (2008). Impacts of chronic anthropogenic
noise from energy-sector activity on abundance of songbirds in the boreal
forest. Conservation Biology , 22 (5), 1186-1193.
Bejder, L., Samuels, A., Whitehead, H., Finn, H., & Allen, S. (2009). Impact
assessment research : use and misuse of habituation, sensitisation and
tolerance in describing wildlife responses to anthropogenic stimuli. Marine
Ecology Progress Series , 395, 117-185.
Bernier, N. J. (2006). The corticotropin-releasing factor system as a mediator of
the appetite-suppressing effects of stress in fish. General and Comparative
Endocrinology , 146 (1), 45-55.
Bernier, N. J., & Peter, R. E. (2001). The hypothalamic-pituitary-interrenal axis
and the control of food intake in teleost fish. Comparative Biochemistry and
Physiology Part B: Biochemistry and Molecular Biology , 129 (2-3), 639-644.
Black, M. P., Moore, B., Canario, A. V., Ford, D., Reavis, R. H., & Grober, M. S.
(2005). Reproduction in context: Field testing a laboratory model of socially
controlled sex change in Lythrypnus dalli (Gilbert). Journal of Experimental
Marine Biology and Ecology , 318 (2), 127-143.
Blackwell, S. B., Greene, C. R., & Richardson, W. J. (2004). Drilling and
operational sounds from an oil production island in the ice-covered Beaufort
Sea. The Journal of the Acoustical Socety of America , 116 (5), 3199-3211.
Blaxter, J. H. (1981). The Swimbladder and Hearing. In W. N. Tavolga, A. N.
Popper, & R. R. Fay, Hearing and Sound Communication in Fishes. New
York: Springer.
98 Blickley, J. L., Blackwood, D., & Patricelli, G. (2012). Experimental evidence for
the effects of chronic anthropogenic noise on abundance of greater sage-
grouse at leks. Conservation Biology , 26 (3), 461-471.
Blickley, J., & Patricelli, G. (2010). Impacts of anthropogenic noise on wildlife:
research priorities for the development of standards and mitigation. Journal
of International Wildlife Law and Policy , 13 (4), 274-292.
Bonga, S. W. (1997). The stress response in fish. Physiological Reviews , 77
(3), 591-625.
Booth, D. J., Bond, N., & Macreadie, P. (2011). Detecting range shifts among
Australian fishes in response to climate change. Marine and Freshwater
Research , 62 (9), 1323-1650.
Breitburg, D. L., & Riedel, G. F. (2005). Multiple Stressors in Marine Systems. In
M. E. Soule, Marine Conservation Biology: The Science of Maintaining the
Sea's Biodiversity. Island Press.
Bruintjes, R., & Radford, A. (2013). Context-dependent impacts of
anthropogenic noise on individual and social behaviour in a cooperatively
breeding fish. Animal Behaviour , 85 (6), 1343-1349
Brumm, H. (2004). The impact of environmental noise on song amplitude in a
territorial bird. Journal of Animal Ecology , 73 (3), 434-440.
Burke, L., Reytar, K., Spalding, M., & Perry, A. (2011). Reefs at Risk Revisited.
World Resources Institute.
Carragher, J. F., Sumpter, J. P., Pottinger, T. G., & Pickering, A. D. (1989). The
deleterious effects of cortisol implantation on reproductive function in two
species of trout, Salmo trutta L. and Salmo gairdneri Richardson . General
and Comparative Endocrinology , 76 (2), 310-321.
99 Chan, A. A., & Blumstein, D. T. (2011). Attention, noise, and implications for
wildlife conservation and management. Applied Animal Behaviour Science ,
131 (1-2), 1-7.
Chrousos, G. P., & Gold, P. W. (1992). The concepts of stress and stress
system disorders. Overview of physical and behavioural homeostasis.
Journal of the American Medical Association , 268 (2), 200.
Clarke, R. D. (1992). Effects of microhabitat and metabolic rate on food intake,
growth and fecundity of two competing coral reef fishes. Coral Reefs , 11 (4),
199-205.
Clement, T., Parikh, V., Schrumpf, M., & Fernald, R. (2005). Behavioural coping
strategies in a cichlid fish: the role of social status and acute stress response
in direct and displaced aggression. Hormones and Behaviour , 47 (3), 336-
342.
Consten, D., Bogerd, J., Komen, J., Lambert, J., & Goos, H. (2001). Long-term
cortisol treatment inhibits pubertal development in male common carp,
Cyprinus carpio. Biology of Reproduction , 64 (4), 1063-1071.
Cyr, N., & Romero, L. M. (2009). Identifying hormonal habituation in field
studies of stress. General and Comparative Endocrinology , 161 (3), 295-
303.
Dalla Valle, A. Z., Rivas-Diaz, R., & Claireaux, G. (2003). Opercular differential
pressure as a predictor of metabolic oxygen demand in the starry flounder.
Journal of Fish Biology , 63 (6), 1578-1588.
De Boeck, G., Vlaeminck, A., Van der Linden, A., & Blust, R. (2000). The
energy metabolism of common carp (Cyprinus carpio) when exposed to salt
stress: an increase in energy expenditure or effects of starvation?
100 Physiological and Biochemical Zoology: Ecological and Evolutionary
Approaches , 73 (1), 102-111.
Department for Transport. (2015). Road Traffic Estimates: Great Britain 2014 .
Statistical Release, London.
DNV. (2012). Shipping 2020.
Dobson, A. P., Bradshaw, A. D., & Baker, A. J. (1997). Hopes for the future:
restoration ecology and conservation biology. Science , 277 (5325), 515-522.
Dowling, J. L., Luther, D. A., & Marra, P. P. (2011). Comparative effects of
urban development and anthropogenic noise on bird songs. Behavioural
Ecology , 23 (1), 201-209.
Eckert, S. A., Moore , J. E., Dunn, D. C., Sagarminaga van Buiten, R. S.,
Eckert, K. L., & Halpin, P. N. (2008). Modeling loggerhead turtle movement in
the mediterranean: importance of body size and oceanography. Ecological
Applications , 18 (2), 290-308.
Edds-Walton, P. L. (1997). Acoustic communication signals of Mysticete
whales. Bioacoustics: The International Journal of Animal Sound and its
Recording , 8 (1-2), 47-60.
Elert, G. (2015). The Physics Hypertextbook. Retrieved from
http://physics.info/sound/
Fabbri, E., Capuzzo, A., & Moon, T. W. (1998). The role of circulating
catecholamines in the regulation of fish metabolism: An overview.
Comparative Biochemistry and Physiology Part C: Pharmacology, Toxicology
and Endocrinology , 120 (2), 177-192.
Fabry, V. J., Seibel, B. A., Feely, R. A., & Orr, J. C. (2008). Impacts of ocean
acidification on marine fauna and ecosystem processes. ICES Journal of
Marine Science , 65 (3), 414-432.
101 Fast , M., Hosoya, S., Johnson, S., & Afonso , L. (2008). Cortisol response and
immune-related effects of Atlantic salmon (Salmo salar Linnaeus) subjected
to short- and long-term stress. Fish and Shellfish Immunology , 24 (2), 194-
204.
Fay, R. (1998). Auditory stream segregation in goldfish (Carassius auratus).
Hearing Research , 120 (1-2), 69-76.
Fay, R., & Popper, A. (2012). Fish hearing: new perspectives from two ‘senior’
bioacousticians. Brain, Behaviour and Evolution , 79 (4), 215-217.
Fernandez, A., Edwards, J., Rodriguez, F., Espinosa de los monteros, A.,
Herraez, P., Castro, P., et al. (2005). “Gas and fat embolic syndrome”
involving a mass stranding of beaked whales (family Ziphiidae) exposed to
anthropogenic sonar signals. Veterinary Pathology , 42 (4), 446-457.
Finegold, L. S., Harris, C. S., & von Gierke, H. E. (1994). Community
annoyance and sleep disturbance: updates criteria for assessing the impacts
of general transportation noise on people. Noise Control Engineering Journal
, 42 (1), 25-30.
Fink, G. (2010). Stress: Definition and History. In G. Fink, Stress Science:
Neuroendocrinology. Academic Press.
Forman, R., & Deblinger, R. (2001). The ecological road-effect zone of a
Massachusetts (USA) suburban highway. Conservation Biology , 14 (1), 36-
46.
Francis, C. D., Ortega, C. P., & Cruz, A. (2009). Noise pollution changes avian
communities and species interactions. Current Biology , 19 (16), 1415-1419.
Francis, C. D., & Barber, J. R. (2013). A framework for understanding noise
impacts on wildlife: an urgent conservation priority. Frontiers in Ecology and
the Environment , 11 (6), 305-313.
102 Fulton, T. W. (1904). The rate of growth of fishes. 22nd Annual Report of the
Fishery Board of Scotland , 3, 141-241.
Gall, S. C., & Thompson, R. C. (2015). The impact of debris on marine life.
Marine Pollution Bulletin , 92 (1-2), 170-179.
Gerges, S., & Sehrndt, G. (1995). Noise Sources. WHO.
Goines, L., & Hagler, L. (2007). Noise pollution: a modern plague. Southern
Medical Journal , 100 (3), 287-294.
Goudie, A. (2013). The human impact on the natural environment: past, present
and future (Seventh ed.). Chichester: John Wiley & Sons.
Graham, A., & Cooke, S. (2008). The effects of noise disturbance from various
recreational boating activities common to inland waters on the cardiac
physiology of a freshwater fish, the largemouth bass (Micropterus
salmoides). Aquatic Conservation: Marine and Freshwater Ecosystems , 18
(7), 1315-1324.
Grant, S. M., & Brown, J. A. (1999). Variation in condition of coastal
Newfoundland 0-group Atlantic cod (Gadus morhua): field and laboratory
studies using simple condition indices. Marine Biology , 133 (4), 611-620.
Great Barrier Reef Marine Park Authority. (2009). Great Barrier Reef Outlook
Report 2009: In brief. Townsville: Great Barrier Reef Marine Park Authority.
Gregory, T. R., & Wood, C. M. (1999). The effects of chronic plasma cortisol
elevation on the feeding behaviour, growth, competitive ability, and
swimming performance of juvenile rainbow trout. Physiological and
Biochemical Zoology: Ecological and Evolutionary Approaches , 72 (3), 286-
295.
Griefahn, B., Brode, P., Marks, A., & Basner, M. (2008). Autonomic arousals
related to traffic noise during sleep. Sleep , 31 (4), 569-577.
103 Halfwerk, W., & Slabbekoorn, H. (2015). Pollution going multimodal: the
complex impact of the human-altered sensory environment on animal
perception and performance. Biology Letters , 11 (4), 20141051.
Halfwerk, W., Holleman, L. J., Lessells, C. M., & Slabbekoorn, H. (2011).
Negative impact of traffic noise on avian reproductive success. Journal of
Applied Ecology , 48 (1), 210-219.
Hallegatte, S. (2009). Strategies to adapt to an uncertain climate change.
Global Environmental Change , 19 (2), 240-247.
Halpern, B. S., Selkoe, K. A., Micheli, F., & Kappel, C. V. (2007). Evaluating and
ranking the vulnerability of global marine ecosystems to anthropogenic
threats. Conservation Biology , 21 (5), 1301-1315.
Halpern, B. S., Walbridge, S., Selkoe, K. A., Kappel, C. V., Micheli, F.,
D'Agrosa, C. D., et al. (2008). A global map of human impact on marine
ecosystems. Science , 319 (5865), 948-952.
Halpern, B. S., & Warner, R. (2002). Marine reserves have rapid and lasting
effects. Ecology Letters , 5 (3), 361-366.
Handegard, N., Michalsen, K., & Tjostheim, D. (2003). Avoidance behaviour in
cod (Gadus morhua) to a bottom-trawling vessel. Aquatic Living Resources ,
16 (3), 265-270.
Hanson, C. E. (2008). High speed train noise effects on wildlife and domestic
livestock. In Noise and vibration mitigation for rail transportation systems.
Berlin: Springer Berlin Heidelberg.
Harriott, V. J. (2002). Marine tourism impacts and their management on the
Great Barrier Reef. CRC Reef Research Centre.
Harrison, R., & King, J. (1965). Marine Mammals (2 ed.). (R. Harrison, & J.
King, Eds.) London, UK: Hutchinson & co.
104 Haviland-Howell, G., Frankel, A. S., Powell, C. M., Bocconcelli, A., Herman, R.
L., & Sayigh, L. S. (2007). Recreational boating traffic: A chronic source of
anthropogenic noise in the Wilmington, North Carolina Intracoastal
Waterway. The Journal of the Acoustical Society of America , 122 (1), 151-
160.
Hawkins, A. D., Pembroke, A. E., & Popper, A. N. (2015). Information gaps in
understanding the effects of noise on fishes and invertebrates. Reviews in
Fish Biology and Fisheries , 25 (1), 39-64.
Herbert, N. A., & Steffensen, J. F. (2005). The response of Atlantic cod, Gadus
morhua, to progressive hypoxia: fish swimming speed and physiological
stress. Marine Biology , 147 (6), 1403-1412.
Hildebrand, J. A. (2009). Anthropogenic and natural sources of ambient noise in
the ocean. Marine Ecology Progress Series , 395, 5-20.
Hoare, D. J., Couzin, I. D., Godin, J. G., & Krause, J. (2004). Context-
dependent group size choice in fish. Animal Behaviour , 67 (1), 155-164.
Home Office. (2012, August 3). Animals (Scientific Procedures) Act 1986
Amendment Regulations. London, Great Britain.
Insley, S. J., Phillips, A. V., & Charrier, I. (2003). A review of social recognition
in pinnipeds. Aquatic Mammals , 29 (2), 181-201.
Iwama, G. K. (1998). Stress in Fish. Annals of the New York Academy of
Sciences , 851 (1), 3014-310.
Iwama, G. K., Pickering, A. D., Sumpter, J. P., & Schreck, C. B. (2011). Fish
Stress and Health in Aqauculture. Cambridge University Press.
Jeffs, A., Tolimieri, N., & Montgomery, J. (2003). Crabs on cue for the coast: the
use of underwater sound for orientation by pelagic crab stages. Marine and
freshwater research , 54 (7), 841-845.
105 Jentoft, S., Aastveit, A. H., Torjesen, P. A., & Andersen, O. (2005). Effects of
stress on growth, cortisol and glucose levels in non-domesticated Eurasian
perch (Perca fluviatilis) and domesticated rainbow trout (Oncorhynchus
mykiss). Comparative Biochemistry and Physiology Part A: Molecular &
Integrative Physiology , 141 (3), 257-358.
Jobling, M. (1981). The influences of feeding on the metabolic rate of fishes: a
short review. Journal of Fish Biology , 18 (4), 385-400.
Jones, G. P. (1986). Food availability affects growth in a coral reef fish.
Oecologia , 70 (1), 136-139.
Karl, T. R., & Trenberth, K. E. (2003). Modern global climate change. Science ,
302 (5651), 1719-1723.
Klumpp, D. W., McKinnon, D., & Daniel, P. (1987). Damselfish territories: zones
of high productivity on coral reefs. Marine Ecology Progress Series , 40, 41-
51.
Kristensen, E. (1983). Ventilation and oxygen uptake by three species of Nereis
(Annelida: Polychaeta). I. Effects of hypoxia. Marine Ecology Progress Series
, 12, 299-306.
Kuttruff, H. (2007). Acoustics, an Introduction. Oxford: Taylor & Francis.
Lackner, R., Wieser, W., Huber, M., & Dalla Via, J. (1988). Responses of
intermediary metabolism to acute handling stress and recovery in untrained
and trained Leuciscus cephalus (cyprinidae, teleostei). Journal of
Experimental Biology , 140 (1), 393-404.
Lammers, M. O., & Whitlow, W. L. (2003). Directionality in the whistles of
Hawaiian spinner dolphins (Stenella longirostris): a singal feature to cue
direction of movement? Marine Mammal Science , 19 (2), 249-264.
106 Lampe, U., Reinhold, K., & Schmoll, T. (2013). How grasshoppers respond to
road noise: Developmental plasticity and population differentiation in acoustic
signalling. Functional Ecology , 28 (3), 660-668.
Lankford, S. E., Adams, T. E., Miller, R. A., & Cech, J. J. (2005). The cost of
chronic stress: impacts of a nonhabituating stress response on metabolic
variables and swimming performance in sturgeon. Physiological and
Biochemical Zoology: Ecological and Evolutionary Approaches , 78 (4), 599-
609.
Larkin, R., Pater, L., & Tazik, D. (1996). Effects of military noise on wildlife. A
literature review.
Larsson, A., Haux, C., & Sjobeck, M. (1985). Fish physiology and metal
pollution: results and experiences from laboratory and field studies.
Ecotoxicology and Environmental Safety , 9 (3), 250-281.
Lasker, R., Feder, H. M., Theilacker, G. H., & May, R. C. (1970). Feeding,
growth, and survival of Engraulis mordax larvae reared in the laboratory.
Marine Biology , 5 (4), 345-353.
Leu, M., Hanser, S., & Knick, S. (2008). The human footprint in the west: a large
scale analysis of anthropogenic impacts. Ecological Applications , 18 (5),
1119-39.
Lima, S. L. (1998). Stress and decision making under the risk of predation:
recent developments from behavioural, reproductive and ecological
perspectives. In P. J. Slater, A. P. Moller, & M. Manfred, Advances in the
tstudy of behaviour: stress and behaviour. Academic Press.
Litsios, G., Pellissier, L., Forest, F., Lexer, C., Pearman, P. B., Zimmermann, N.
E., et al. (2012). Trophic specialisation influences the rate of environmental
107 niche evolution in damselfishes (Pomacentridae). Proceedings of the Royal
Society B: Biological Sciences , 279 (1743), 3662-3669 .
Lloyd's Marine Register. (2015). Global Marine Technology Trends 2030.
Southampton: University of Southampton.
Lo-Yat, A., Simpson, S. D., Meekan, M. G., Lecchini, D., Martinez, E., & Galzin,
R. (2011). Extreme climatic events reduce ocean productivity and larval
supply in a tropical reef ecosystem. Global Change Biology , 17 (4), 1695-
1702.
Lucas, M. C. (1994). Heart rate as an indicator of metabolic rate and activity in
adult Atlantic salmon, Salmo salar. Journal of Fish Biology , 44 (5), 889-903.
Lynch, E., Northrub, J. M., McKenna, M. F., Anderson Jr, C. R., Angeloni, L., &
Wittemyer, G. (2015). Landscape and anthropogenic features influence the
use of auditory vigilance by mule deer. Behavioural Ecology , 26 (1), 75-82.
Mackenzie, B. R., Leggett, W. C., & Peters, R. H. (1990). Estimating larval fish
ingestion rates: can laboratory derived values be reliably extrapolated to the
wild? Marine Ecology Progress Series , 67, 209-225.
Main, K. L. (1987). Predator avoidance in seagrass meadows: prey behaviour,
microhabitat selection, and cryptic colouration. Ecology , 68 (1), 170-180.
Matthaei, C. D., & Lange, K. (2016). Multiple stressor effects on freshwater fish:
a review and meta-analysis. In G. P. Closs, M. Krkosek, & J. D. Olden,
Conservation of Freshwater Fishes. Cambridge University Press.
Matuschek, R., & Betke, K. (2009). Measurements of construction noise during
pile driving of offshore research platforms and wind farms. Rotterdam:
NAG/DAGA.
108 Mazeaud, M. M., Mazeaud, F., & Donaldson, E. M. (1977). Primary and
secondary effects of stress in fish: some new data with a general review.
Transactions of the American Fisheries Society , 106 (3), 201-212.
McCauley, R. D., & Cato, D. H. (2000). Patterns of fish calling in a nearshore
environment in the Great Barrier Reef. Philosophical Transactions of the
Royal Society B , 355 (1401), 1289-1293.
McCormick, M. I. (1998). Behaviourally induced maternal stress in fish
influences progeny quality by a hormonal mechanism. Ecology , 79 (6),
1873-1883.
McCormick, M. I. (2006). Mothers matter: crowding leads to stressed mothers
and smaller offspring in marine fish. Ecology , 87 (5), 1104-1109.
McKay, L., & Cidlowski, J. (2003). Physiologic and pharmacologic effects of
corticosteroids. In D. W. Kufe, R. E. Pollock, & R. Weichselbaum, Cancer
Medicine-6. Decker.
Meillere, A., Brischoux, F., Ribout, C., & Angelier, F. (2015). Traffic noise
exposure affects telomere length in nestling house sparrows. Biology Letters
, 11 (9), 20150559.
Melamed, S., & Bruhis, S. (1996). The effects of chronic industrial noise
exposure on urinary cortisol, fatigue, and irritability: a controlled field
experiment. Journal of Occupational and Environmental Medicine , 38 (3),
252-256.
Melamed, S., Luz, J., & Green, M. S. (1992). Noise exposure, noise annoyance
and their relation to psychological distress, accident and sickness absence
among blue-collar workers - the Cordis Study. Israel Journal of Medical
Sciences , 28 (8), 629-635.
109 Melcon, M., Cummins, A., Kerosky, S., Roche, L., Wiggins, S., & Hildebrand, J.
(2012). Blue whales respond to anthropogenic noise. PLOS One , 7 (2),
e32681.
Merchie, G., Lavens, P., & Sorgeloos, P. (1997). Optimization of dietary vitamin
C in fish and crustacean larvae: a review. Aquaculture , 155 (1), 165-181.
Mesa, M. (1994). Effects of multiple acute stressors on the predator avoidance
ability and physiology of juvenile chinook salmon. Transactions of the
American Fisheries Society , 123 (5), 786-793.
Miksis-Olds, J. L., Donaghay, P. L., Miller, J. H., Tyack, P. L., & Nystuen, J. A.
(2007). Noise level correlates with manatee use of foraging habitats. The
Journal of the Acoustical Society of America , 121 (5), 3011.
Montero, D., Kalinowski, T., Obach, A., Robaina, L., Tort, L., Caballero, M. J., et
al. (2003). Vegetable lipid sources for gilthead seabream (Sparus aurata):
effects on fish health. Aquaculture , 225 (1), 353-370.
Morley, E. L., Jones, G., & Radford, A. N. (2013). The importance of
invertebrates when considering the impacts of anthopogenic noise.
Proceedings of the Royal Society B , 281 (1776).
Mulder, Koricheva, Huss-Danell, Hogberg , & Joshi. (1999). Insects affect
relationships between plant species richness and ecosystem processes.
Ecology Letters , 2 (4) .
Munday, P. L., & Wilson, S. K. (1997). Comparative efficacy of clove oil and
other chemicals in anaesthetization of Pomacentrus amboinensis, a coral
reef fish. Journal of Fish Biology , 51 (5), 931-938.
Myrberg, A. (1972). Ethology of the bicolor damselfish, Eupomacentrus partitus
(pisces: Pomacentridae): a comparative analysis of laboratory and field
behaviour. Animal Behaviour Monographs , 5 (3), 199-283.
110 Nakamura, K., Iida, H., & Nakano, H. (1986). Riboflavin in the skin of albinic
flatfish Liopsetta obscura. Bulletin of the Japanese Society of Scientific
Fisheries , 52 (12), 2207.
Nowacek, D., Thorne, L., Johnston, D., & Tyack, P. (2007). Responses of
cetaceans to anthropogenic noise. Mammal Review , 37 (2), 81-115.
Oertli, J. (2013). Noise and vibration mitigation for rail transportation systems.
(J. Nielsen, Ed.)
OGP. (2008). Fundamentals of underwater sound. Technical, International
Association of Oil and Gas Producers.
Oreskes, N. (2004). The scientific consensus on climate change. Science , 306
(3), 1686.
Ouis, D. (2001). Annoyance from road traffic noise: a review. Journal of
Environmental Psychology , 21 (1), 101-120.
Overli, O., Harris, C. A., & Winberg, S. (1999). Short-term effects of fights for
social dominance and the establishment of dominant-subordinate
relationships on brain monoamines and cortisol in rainbow trout. Brain,
Behaviour and Ecology , 54 (5), 263-275.
Overli, O., Kotzian , S., & Winberg, S. (2002). Effects of cortisol on aggression
and locomotor activity in rainbow trout. Hormones and Behaviour , 42 (1), 53-
61.
Pandolfi, J. M. (2015). Ecology: Deep and complex ways to survive bleaching.
Nature , 518 (7537), 43-44.
Pascual, P., Pedrajas, J. R., Toribio, F., Lopez-Barea, J., & Peinado, J. (2003).
Effect of food deprivation on oxidative stress biomarkers in fish (Sparus
aurata). Chemico-Biological Interaction , 145 (2), 191-199.
111 Pepper, P. A., Turner, V. L., Goodson, A. D., & Black, K. D. (2004). Source
levels and spectra emitted by three commercial aquaculture anti-predation
devices. Proc 7th European Conference on Underwater Acoustics.
Commission of the European Communities.
Perry, S. F., & Bernier, N. J. (1999). The acute humoral adrenergic stress
response in fish: facts and fiction. Aquaculture , 177 (1), 285-295.
Peterson, C. H., Rice, S. D., Short, J. W., Esler, D., Bodkin, J. L., Ballachey, B.
E., et al. (2003). Long term ecosystem response to the Exxon Valdez oil spill.
Science , 302 (5653), 2082-2086.
Pickering, A. D. (1992). Rainbow trout husbandry: management of the stress
response. Aquaculture , 100 (1-3), 125-139.
Pickering, A. D., & Pottinger, T. G. (1989). Stress responses and disease
resistance in salmonid fish: Effects of chronic elevation of plasma cortisol.
Fish Physiology and Biochemistry , 7 (1), 253-258.
Pickering, A. D., Pottinger, T. G., & Christie, P. (1982). Recovery of the brown
trout, Salmo trutta L., from acute handling stress: a time-course study.
Journal of Fish Biology , 20 (2), 229-244.
Pillay, T. V., & Kutty, M. N. (2005). Aquaculture: principles and practices.
Oxford: Blackwell Publishing.
Platt, T., Fuentes-Yaco, C., & Frank, K. T. (2003). Marine ecology: spring algal
bloom and larval fish survival. Nature , 423 (6938), 398-399.
Poloczanska, E. S., Brown, C. J., Sydeman, W. J., Kiessling, W., Schoeman, D.
S., Moore, P. J., et al. (2013). Global imprint of climate change on marine life.
Nature Climate Change , 3, 919-925.
Popper, A. N., & Fay, R. R. (2011). Rethinking sound detection by fishes.
Hearing Research , 273 (1-2), 25-36.
112 Pottinger, T. G. (2006). Context dependent differences in growth of two rainbow
trout (Oncorhynchus mykiss) lines selected for divergent stress
responsiveness. Aquaculture , 256, 140-147.
Pottinger, T. G., & Carrick, T. R. (2001). Stress responsiveness affects
dominant-subordinate relationships in rainbow trout. Hormones and
Behaviour , 40 (3), 419-427.
Pottinger, T. G., Balm, P. H., & Pickering, A. D. (1995). Sexual maturity modifies
the responsiveness of the pituitary-interrenal axis to stress in male rainbow
trout. General and Comparative Endocrinology , 98 (3), 311-320.
Pottinger, T., & Carrick, T. (2001). Stress responsiveness affects dominant–
subordinate relationships in rainbow trout. Hormones and Behaviour , 40 (3),
419-427.
Potvin, D. A., Parris, K. M., & Mulder, R. A. (2011). Geographically pervasive
effects of urban noise on frequency and syllable rate of songs and calls in
silvereyes (Zosterops lateralis). Proceedings of the Royal Society B , 278
(1717).
Priede, I. G. (1985). Metabolic scope in fishes. In P. Tytler, & P. Calow, Fish
Energetics. Springer.
Purser, J., & Radford, A. N. (2011). Acoustic noise induces attention shifts and
reduces foraging performance in three-spined sticklebacks (Gasterosteus
aculeatus). PLOS one , 6 (2), e17478.
Purser, J., Bruintjes, R., Simpson, S. D., & Radford, A. N. (2016). Condition-
dependent physiological and behavioural responses to anthropogenic noise.
Physiology and Behaviour , 155, 157-161.
113 Rabin, L. A., Coss, R. G., & Owings, D. H. (2006). The effects of wind turbines
on antipredator behaviour in California ground squirrels (Spermophilus
beecheyi). Biological Conservation , 131 (3), 410-420.
Radford, C. A., Stanley, J. A., Simpson, S. D., & Jeffs, A. G. (2011). Juvenile
coral reef fish use sound to locate habitats. Coral Reefs , 30 (2), 295-305.
Remage-Healey, L., Nowacek, D. P., & Bass, A. H. (2006). Dolphin foraging
sounds suppress calling and elevate stress hormone levels in a prey species,
the Gulf toadfish. Journal of Experimental Biology , 209, 4444-4451.
Ridgway, S., Carder, D., Finneran, J., Keogh, M., Kamolnick, T., Todd, M., et al.
(2006). Dolphin continuous auditory vigilance for five days. Journal of
Experimental Biology , 209, 3621-3628.
Ridgwell, A., Schmidt, D. N., Turley, C., Brownlee, C., Maldonado, M. T., Tortell,
P., et al. (2009). From laboratory manipulations to earth system models:
scaling calcification impacts of ocean acidification. Biogeosciences , 6 (11),
2611-2623.
Rochman, C. M., Hoh, E., Kurobe, R., & Teh, S. J. (2013). Ingested plastic
transfers hazardous chemicals to fish and induces hepatic stress. Scientific
Reports , 3, 3263.
Rogers, A., Manwell, J., & Wright, S. (2002). Wind Turbine Acoustic Noise.
Romano, T. A., Keogh, M. J., Kelly, C., Feng, P., Berk, L., Schlundt, C. E., et al.
(2011). Anthropogenic sound and marine mammal health: measures of the
nervous and immune systems before and after intense sound exposure.
Canadian Journal of Fisheries and Aquatic Sciences , 61 (7), 1124-1134.
Romero, L. M. (2004). Physiological stress in ecology: lessons from biomedical
research. Trends in Ecology & Evolution , 19 (5), 249-255.
114 Ronnestad, I., Thorsen, A., & Finn, R. N. (1999). Fish larval nutrition: a review
of recent advances in the roles of amino acids. Aquaculture , 177 (1), 201-
216.
Rutherford, T. (2011). Air Transport Statistics. London: House of Commons.
Sara, G., Dean, J. M., D'Amato, D., Buscaino, G., Oliveri, A., Genovese, S., et
al. (2007). Effect of boat noise on the behaviour of bluefin tuna Thunnus
thynnus in the Mediterranean Sea. Marine Ecology Progress Series , 331,
243-253.
SCENIHR. (2008, Sep). Potential health risks of exposure to noise from
personal music players and mobile phones including a music playing
function.
Schellart, N. A., & Popper, A. N. (2012). Evolution of the Auditory System in
Fish. In D. B. Webster, & R. R. Fay, The Evolutionary Biology of Hearing.
Springer Science and Business Media.
Schlundt, C. E., Finneran, J. J., Carder, D. A., & Ridgway, S. H. (2000).
Temporary shift in masked hearing thresholds of bottlenose dolphins,
Tursiops truncatus, and white whales, Delphinapterus leucas, after exposure
to intense tones. The Journal of the Acoustical Society of America , 107 (6),
3496-3508.
Schreck, C. B. (2000). Accumulation and long-term effects of stress in fish. In
G. P. Moberg, & J. A. Mench (Eds.), The Biology of Animal Stress:
assessment and implications for welfare. Wallingford: CAB International.
Schreck, C. B., Contreras-Sanchez, W., & Fitzpatrick, M. S. (2001). Effects of
stress on fish reproduction, gamete quality, and progeny. Aquaculture , 197
(1-4), 3-24.
115 Schulz-Mirbach, T., Metscher, B., & Ladich, F. (2012). Relationship between
swim bladder morphology and hearing abilities - a case study on Asian and
African Cichlids. PLOS One , 7 (8), e42292.
Schusterman, R. J. (2008). Vocal Learning in Mammals with Special Emphasis
on Pinnipeds. In K. Oller, & U. Gribel, The Evolution of Communicative
Flexibility: Complexity, Creativity and Adaptability in Human and Animal
Communication. Cambridge, MA: MIT Press.
Sherman, C. H., & Butler, J. L. (2007). Transducers and Arrays for Underwater
Sound. In The Underwater Acoustics Series. Springer Science and Business
Media.
Shieh, B., Liang, S., Chen, C., Loa, H., & Liao, C. (2012). Acoustic adaptations
to anthropogenic noise in the cicada Cryptotympana takasagona Kato
(Hemiptera: Cicadidae). Acta Ethologica , 15 (1), 33-38.
Simmonds, E. J., & MacLennan, D. N. (2005). Fiheries Acoustics: Theory and
Practice. Blackwell Science.
Simmonds, M., Dolman, S., & Weilgart, L. (2003). Oceans of Noise: A WDCS
Science Report. Marine Turtle Newsletter , 105 (25).
Simpson, S. D., Jeffs, A., Montgomery, J. C., McCauley, R. D., & Meekan, M.
G. (2008). Nocturnal relocation of adult and juvenile coral reef fishes in
response to reef noise. Coral Reefs , 27 (1), 97-104.
Simpson, S. D., Meekan, M. G., McCauley, R. D., & Jeffs, A. (2004). Attraction
of settlement-stage coral reef fishes to reef noise. Marine Ecology Progress
Series , 276, 263-268.
Simpson, S. D., Purser, J., & Radford, A. N. (2015). Anthropogenic noise
compromises antipredator behaviour in European eels. Global Change
Biology , 21 (2), 586-593.
116 Simpson, S. D., Radford, A. N., Nedelec, S. L., Ferrari, M. C., Chivers, D. P.,
McCormick, M. I., et al. (2016). Anthropogenic noise increases fish mortality
by predation. Nature Communications , 7, 10544.
Simpson, S. D., Meekan, M. G., Montgomery, J., McCauley, R. D., & Jeffs, A.
(2005). Homeward Sound. Science , 308 (5719), 221.
Slabbekoorn, H. (2016). Aiming for progress in understanding noise impacts on
fish: complementary need for indoor and outdoor studies. Advances in
Experimental Medicine and Biology , 875, 1057-1065
Slabbekoorn, H., Bouton, N., van Opzeeland, I., Coers, A., ten Cate, C., &
Popper, A. (2010). A noisy spring: the impact of globally rising underwater
sound levels on fish. Trends in Ecology and Evolution , 25 (7), 419-427.
Smith, A., Kane, M., & Popper, A. (2003). Noise-induced stress response and
hearing loss in goldfish (Carassius auratus). Journal of Experimental Biology
, 207 (3), 427-435.
Steffens, W. (1989). Principles of fish nutrition. Ellis Horwood Limited.
Stephens, R. W. (1970). Underwater Acoustics. London: Wiley.
Stratholt, M., Donaldson, E., & Liley, N. (1997). Stress induced elevation of
plasma cortisol in adult female coho salmon (Oncorhynchus kisutch), is
reflected in egg cortisol content, but does not appear to affect early
development. Aquaculture , 158 (1-2), 141-153.
Svendsen, M. B., Bushnell, P. G., & Steffensen, J. F. (2016). Design and setup
of intermittent-flow respirometry system for aquatic organisms. Journal of
Fish Biology , 88 (1), 26-50.
Szabo, S., Tache, Y., & Somogyi, A. (2012). The legacy of Hans Selye and the
origins of stress research: a retrospective 75 years after his landmark brief
"letter" to the editor of nature. Stress , 15 (5), 472-478.
117 Talling, J. F. (1960). Comparative laboratory and field studies of photosynthesis
by a marine planktonic diatom. Limnology and Oceanography , 5 (1), 62-77.
Tavolga, W. N., Popper, A. N., & Fay, R. R. (2012). Hearing and sound
communication in fishes. Proceedings in Life Sciences .
Thomas, C. D., Cameron, A., Green, R. E., Bakkenes, M., Beaumont, L. J.,
Collingham, Y. C., et al. (2004). Extinction risk from climate change. Nature
427, 145-148.
Thomsen, F., Ludermann, K., Kafemann, R., & Piper, W. (2006). Effects of
offshore wind farm noise on marine mammals and fish. Hamburg: COWRIE
Ltd.
Tyack, P. (2009). Acoustic playback experiments to study behavioural
responses of free-ranging marine animals to anthropogenic sound. Marine
Ecology Progress Series , 395, 187-200.
University of Rhode Island. (2004). Discovery of Sound in the Sea. Retrieved
Oct 2015, from http://www.dosits.org/science/soundsinthesea/airwater/
Urick, R. J. (1996). Principles of Underwater Sound (3rd Edition ed.). Peninsula.
Van Dijk, P. L., Hardewig, I., & Holker, F. (2005). Energy reserves during food
deprivation and compensatory growth in juvenile roach: the importance of
season and temperature. Journal of Fish Biology , 66 (1), 167-181. van Rooij, J. M., & Videler, J. J. (1996). Estimating oxygen uptake rate from
ventilation frequency in the reef fish Sparisoma viride. Marine Ecology
Progress Series , 132 (1-3), 31-41.
Van Weerd, J. H., & Komen, J. (1998). The effects of chronic stress on growth
in fish: a critical appraisal. Comparative Biochemistry and Physiology Part A:
Molecular & Integrative Physiology , 120 (1), 107-112.
118 Vasconcelos, R., Amorim, C., & Ladich, F. (2007). Effects of ship noise on the
detectability of communication signals in the Lusitanian toadfish. The Journal
of Experimental Biology , 210, 2104-2112.
Verbeek, P., Iwamoto, T., & Murakami, N. (2007). Differences in aggression
between wild-type and domesticated fighting fish are context dependent.
Animal Behaviour , 73 (1), 75-83.
Vijayan , M., & Moon, T. (1992). Acute handling stress alters hepatic glycogen
metabolism in food-deprived rainbow trout (Oncorhynchus mykiss). Canadian
Journal of Fisheries and Aquatic Studies , 49 (11). 2260-2266.
Voellmy, I. K., Purser, J., Flynn, D., Kennedy, P., Simpson, S. D., & Radford, A.
N. (2014). Acoustic noise reduces foraging success in two sympatric fish
species via different mechanisms. Animal Behaviour , 89, 191-198.
Wahlberg, M., & Westerberg, H. (2005). Hearing in fish and their reactions to
sounds from offshore wind farms. Marine Ecology Progress Series , 288,
295-309.
Wale, M., Simpson, S. D., & Radford, A. N. (2013). Noise negatively affects
foraging and antipredator behaviour in shore crabs. Animal Behaviour , 86
(1), 111-118.
Washington State Department of Transportation. (2012). 2011 Traffic Noise
Policy and Procedures.
Weilgart, L. S. (2007). The impacts of anthropogenic ocean noise on cetaceans
and implications for management. Canadian Journal of Zoology , 85 (11),
1091-1116.
Wells, R. M., Tetens, V., & Devries, A. L. (1984). Recovery from stress following
capture and anaesthesia of antarctic fish: haematology and blood chemistry.
Journal of Fish Biology , 25 (5), 567-576.
119 Wibe, A. E., Nordtug, T., & Jenssen, B. M. (2001). Effects of bis(tributyltin)oxide
on antipredator behaviour in threespine stickleback Gasterosteus aculeatus
L. Chemosphere , 44 (3), 475-481.
Williams, R., Wright, A. J., Ashe, E., Blight, L. K., Bruintjes, R., Canessa, R., et
al. (2015). Impacts of anthropogenic noise on marine life: Publication
patterns, new discoveries, and future directions in research and
management. Oceans and Coastal Management , 115, 17-24.
Wood, C. M., Turner, J., & Graham, M. S. (1983). Why do fish die after severe
exercise? Journal of Fish Biology , 22 (2), 189-201.
Wright, A., Soto, N. A., Baldwin, A. L., Bateson, M., Beale , C., Clark , C., et al.
(2007). Do marine mammals experience stress related to anthropogenic
noise? International Journal of Comparative Psychology , 20, 274-316.
Wursig, B., Greene, C. R., & Jefferson, T. A. (2000). Development of an air
bubble curtain to reduce underwater noise of percussive piling. Marine
Environmental Research , 49 (1), 79-93.
Wysocki, L. E., Dittami, J. P., & Ladich, F. (2006). Ship noise and cortisol
secretion in European freshwater fishes. Biological Conservation , 128 (4),
501-508.
120