1. Introduction

1. This technical report presents the results of a desktop study undertaken by Seiche Ltd. considering the potential effects of underwater noise on the marine environment from the development of the Ossian Array (hereafter simply referred to as the ‘Array’).
2. The Array, which is to be located within the site boundary, is located in the North Sea, approximately 80 km east of the Aberdeenshire coast ( Figure 1.1   Open ▸ ). The planned activities within the site boundary fall into four phases: pre-construction, construction, operation and maintenance, and decommissioning. Within each of these four phases, different underwater noise sources are identified. These noise sources are both continuous and intermittent in characteristics.
3. Noise is readily transmitted into the underwater environment and there is potential for the noise emissions from construction, operation and maintenance and decommissioning of the Array to adversely affect marine mammals and fish. At a close range from a noise source with high noise levels, permanent or temporary hearing damage may occur to marine species, while at a very close range gross physical trauma is possible. At wider ranges, the introduction of any additional noise could potentially cause short term behavioural changes, for example the ability of a species to communicate and to determine the presence of predators, food, underwater features and obstructions. It should be noted, however, despite the topic being an area of active research, current scientific literature is unclear on whether/how close range or short-term impacts may translate to long term population level impacts.
4. The primary purpose of this technical report is to present the likely distances at which the onset of potential auditory injury (i.e. Permanent Threshold Shifts (PTS) in hearing) and behavioural effects on different marine species may occur when exposed to the different anthropogenic noises that occur during different developmental phases of the Array. The results from this technical report have been used to inform the following chapters of the Array Environmental Impact Assessment (EIA) Report in order to determine the potential impact of underwater noise on marine species:

• volume 2, chapter 9: fish and shellfish ecology;
• volume 2, chapter 10: marine mammals; and
• volume 2, chapter 12: commercial fisheries.

5. Consequently, the sensitivity of species, magnitude of potential impact and significance of effect from underwater noise associated with the Array are addressed within the relevant chapters (volume 2, chapters 9, 10 and 12).
6. This technical report uses noise propagation models to calculate the impact ranges to marine mammals and fish for each phase of the Array. Key modelled sources include:

• clearance of Unexploded Ordnance (UXO);
• geophysical and geotechnical surveys;
• impact piling; and
• vessels and other non-impulsive sources.

Figure 1.1: Location of the Array

2.             Study Area

2. Study Area

7. No specific study area has been outlined for underwater noise as this is defined by the receptors and discussed within the relevant topics listed in paragraph 4.
8. The modelled area is approximately 146,000 km2 and covers the Array and an area extending up to 200 km from the site boundary up to land.
9. Bathymetry data used within the modelling was obtained from the General Bathymetric Chart of the Oceans (GEBCO). The GEBCO 2021 Grid is a global terrain model for ocean and land, providing elevation data, in metres, on a 15 arc-second interval grid.
10. To produce a representative sound speed profile, conductivity, temperature, and depth (CTD) data were obtained from the National Oceanic Atmospheric Administration (NOAA) service WODselect for the closest sample point to the development (NOAA, 2023).

3.             Acoustic Concepts and Terminology

3. Acoustic Concepts and Terminology

11. Noise travels through water as vibrations of the fluid particles in a series of pressure waves. These waves comprise a series of alternating compressions (positive pressure) and rarefactions (negative pressure). As noise consists of variations in pressure, the unit for measuring noise is usually referenced to a unit of pressure, the Pascal (Pa). The decibel (dB) is a logarithmic ratio scale used to communicate the large range of acoustic pressures that can be perceived or detected, with a known pressure amplitude chosen as a reference value (i.e. 0 dB). In the case of underwater noise, the reference value (Pref) is taken as 1 μPa, whereas the airborne noise is usually referenced to a pressure of 20 μPa. To convert from a sound pressure level referenced to 20 μPa to a sound pressure referenced to 1 μPa, a factor of 20 log (20/1) (i.e. 26 dB has to be added to the former quantity). Thus 60 dB re 20 μPa is the same as 86 dB re 1 μPa, although differences in sound speeds and different densities mean that the decibel level difference in sound intensity is much more than 26 dB when converting pressure from air to water. All underwater sound pressure levels in this report are quantified in dB re 1 μPa.
12. There are several descriptors used to characterise a sound wave. The difference between the lowest pressure variation (rarefaction) and the highest-pressure variation (compression) is called the peak-to-peak (or pk-pk) sound pressure level. The difference between the highest variation (either positive or negative) and the mean pressure is called the peak pressure level. Lastly, the Root Mean Square (rms) sound pressure level is used as a description of the average amplitude of the variations in pressure over a specific time window. Decibel values reported should always be quoted along with the Pref value employed during calculations. For example, the measured sound pressure level (SPLrms) value of a pulse may be reported as 100 dB re 1 µPa. These descriptions are shown graphically in Figure 3.1   Open ▸ .

Figure 3.1: Graphical Representation of Acoustic Wave Descriptors

13. The SPLrms is defined as:

14. The magnitude of the rms sound pressure level for an impulsive noise (such as airguns from a seismic survey source) will depend upon the integration time, T, used for the calculation (Madsen, 2005). It has become customary to utilise the T90 time period for calculating and reporting rms sound pressure levels[1]. This T90 time period is the interval over which the cumulative energy curve rises from 5% to 95% of the total energy and therefore contains 90% of the sound energy.
15. Another useful measure of noise used in underwater acoustics is the Sound Exposure Level (SEL). This descriptor is used as a measure of the total sound energy of an event or a number of events (e.g. over the course of a day) and is normalised to one second. This allows the total acoustic energy contained in events lasting a different amount of time to be compared on a like for like basis[2]. The SEL is defined as:

where is the integration time of the noise “event”, is the squared sound pressure at a time and is the reference time-integrated squared sound pressure of 1 µPa2s.

16. The frequency of the noise is the rate at which the acoustic oscillations occur in the medium (air/water) and is measured in cycles per second, or Hertz (Hz). When noise is measured in a way which approximates to how a human would perceive it using an A-weighting filter on a noise level meter, the resulting level is described in values of dBA. However, the hearing capability of marine species is not the same as humans, with marine mammals hearing over a wider range of frequencies and with a different sensitivity. It is therefore important to understand how an animal’s hearing varies over its entire frequency range to assess the effects of anthropogenic noise on marine mammals. Consequently, use can be made of frequency weighting scales (M-weighting) to determine the level of the noise in comparison with the auditory response of the animal concerned. A comparison between the typical hearing response curves for fish, humans and marine mammals is shown in Figure 3.2   Open ▸ [3].

Figure 3.2: Comparison Between Hearing Thresholds of Different Animals

17. The broadband acoustic power (i.e. containing all the possible frequencies) emitted by a noise source, measured/modelled at a location within the Array is generally split into and reported in a series of frequency bands. In marine acoustics, the spectrum is generally reported in standard one-third octave band frequencies, where an octave represents a doubling in noise frequency[4].
18. The source level is the sound pressure level of an equivalent and infinitesimally small version of the source (known as point source) at a hypothetical distance of 1 m from it. The source level is commonly used in combination with the Transmission Loss (TL) associated with the environment to obtain the Received Level (RL) at distances from (in the far field of) the source. The far field distance is chosen so that the behaviour of a distributed source[5] can be approximated to that of a point source. Source levels do not indicate the real sound pressure level at 1 m. TL at a frequency of interest is defined as the loss of acoustic energy as the signal propagates from a hypothetical (point) source location to the chosen receiver location. The TL is dependent on water depth, source depth, receiver depth, frequency, geology, and environmental conditions. The TL values are generally evaluated using an acoustic propagation model (various numerical methods exist) accounting for these dependencies.

The RL is the noise level of the acoustic signal recorded (or modelled) at a given location, that corresponds to the acoustic pressure/energy generated by a known active noise source. This considers the acoustic output of a source and is modified by propagation effects. This RL value is strongly dependant on the source, environmental properties, geological properties, and measurement location/depth. The RL is reported in dB either in rms or peak-to-peak sound pressure level (SPL), and SEL metrics, within the relevant one-third octave band frequencies. The RL is related to the SL as (where TL is the transmission loss of the acoustic energy within the survey region):

19. The directional dependence of the source signature and the variation of TL with azimuthal direction (which is strongly dependent on bathymetry) are generally combined and interpolated to report a two-Dimensional (2-D) plot of the RL around the chosen source point up to a chosen distance.

4.             Baseline

4. Baseline

20. Background or “ambient” underwater noise is created by several natural sources, such as rain, breaking waves, wind at the surface, seismic noise, biological noise, and thermal noise. Anthropogenic noises related to the Array activities can be either impulsive (pulsed) such as impact piling, or non-impulsive (continuous) such as ship engines, and the magnitude of the potential impact on marine species will depend heavily on these characteristics. Biological noise sources include marine mammals (using noise to communicate, build up an image of their environment and detect prey and predators) as well as certain fish and shrimp. Anthropogenic noise sources of noise in the marine environment include fishing boats, ships (non-impulsive), marine construction, seismic surveys, and leisure activities (all could be either impulsive or non-impulsive), all of which add to ambient background noise. Other anthropogenic noise within the vicinity of the Array will arise primarily from shipping, the offshore oil and gas industry, subsea geophysical and geotechnical surveys, and the offshore renewables industry.
21. Historically, research relating to both physiological effects and behavioural disturbance of noise on marine receptors has typically been based on determining the absolute noise level for the onset of that effect (whether presented as a single onset threshold or a dose-response/probabilistic function). Consequently, the available numerical criteria for assessing the effects of noise on marine mammals, fish, and shellfish, tend to be based on the absolute noise level criteria, rather than the difference between the baseline noise level and the noise being assessed (Southall et al., 2019).
22. Baseline or background noise levels vary significantly depending on multiple factors, such as seasonal variations and different sea states. Lack of long term measurements/noise data is a widely recognised gap in knowledge in relation to general noisescape and potential effects of human activities on marine species. Understanding the baseline noise level could therefore be valuable in enabling future studies to assess long term effects related to continuous noise levels over time in addition to activity specific effects such as masking, i.e. interfering with useful noises such as predator or prey activity. However, the value of establishing the precise baseline noise level is limited in relation to the current assessment methods due to the lack of available evidence-based studies on the effects of noise relative to background levels on marine receptors.

5.             Acoustic Assessment Criteria

5. Acoustic Assessment Criteria

23. Section 5 describes the background and criteria on which the assessment has been based. Agreement on the methodology was agreed with NatureScot via the Array EIA Scoping Report (volume 3, appendix 6.1) and post-Scoping consultation via email on 05 December 2023. Consultation relevant to underwater noise is included in volume 3, appendix 5.1 and responses to consultation have been addressed in volume 2, chapter 10.

5.1.        Introduction

5.1. Introduction

24. Underwater noise has the potential to affect marine species in different ways depending on its noise level and characteristics. Richardson et al. (1995) defined four zones of noise influence which vary with distance from the source and level. These are:

• The zone of audibility: this is the area within which the animal can detect the noise. Audibility itself does not implicitly mean that the noise will affect the marine mammal.
• The zone of masking: this is defined as the area within which noise can interfere with the detection of other noises such as communication or echolocation clicks. This zone is very hard to estimate due to a paucity of data relating to how marine mammals detect noise in relation to masking levels[6] (for example, humans can hear tones well below the numeric value of the overall noise level).
• The zone of responsiveness: this is defined as the area within which the animal responds either behaviourally or physiologically. The zone of responsiveness is usually smaller than the zone of audibility because, as stated previously, audibility does not necessarily evoke a reaction.
• The zone of injury/hearing loss: this is the area where the noise level is high enough to cause tissue damage in the ear. This can be classified as either a Temporary Threshold Shift (TTS) or PTS. At even closer ranges, and for very high intensity noise sources (e.g. underwater explosions), physical trauma or even death are possible.

25. For the study contained within this technical report, it is the zones of injury and disturbance (i.e. responsiveness) that are of interest (there is insufficient scientific evidence to properly evaluate masking). To determine the potential spatial range of injury and disturbance, a review has been undertaken of available evidence, including international guidance and scientific literature. The following sections summarise the relevant thresholds for onset of effects and describe the evidence base used to derive them.

5.2.        Injury (Physiological Damage) to Mammals

5.2. Injury (Physiological Damage) to Mammals

26. Noise propagation models can be constructed to allow the received noise level at different distances from the source to be calculated. To determine the potential consequence of these received levels on any marine mammals which might experience such noise emissions, it is necessary to relate the levels to known or estimated potential impact thresholds. The auditory injury (PTS/TTS) threshold criteria proposed by Southall et al. (2019) are based on a combination of unweighted peak pressure levels and mammal hearing weighted SEL. The hearing weighting function is designed to represent the frequency characteristics (bandwidth and noise level) for each group within which acoustic signals can be perceived and therefore assumed have auditory effects. The categories include:

• Low Frequency (LF) cetaceans: marine mammal species such as baleen whales (e.g. minke whale Balaenoptera acutorostrata).
• High Frequency (HF) cetaceans: marine mammal species such as dolphins, toothed whales, beaked whales and bottlenose whales (e.g. bottlenose dolphin Tursiops truncates and white-beaked dolphin Lagenorhynchus albirostris).
• Very High Frequency (VHF) cetaceans: marine mammal species such as true porpoises, river dolphins and pygmy/dwarf sperm whales and some oceanic dolphins, generally with auditory centre frequencies above 100 kHz) (e.g. harbour porpoise Phocoena phocoena).
• Phocid Carnivores in Water (PCW): true seals (e.g. harbour seal Phoca vitulina and grey seal Halichoreus grypus); hearing in air is considered separately in the group Phocid Carnivores in Air (PCA).
• Other Marine Carnivores in Water (OCW): including otariid pinnipeds (e.g. sea lions and fur seals), sea otters and polar bears; air hearing considered separately in the group Other Marine Carnivores in Air (OCA).

27. These weighting functions have therefore been used in this study and are shown in Figure 5.1   Open ▸ .

Figure 5.1: Hearing Weighting Functions for Pinnipeds and Cetaceans (Southall et al., 2019)

28. Auditory injury criteria proposed in Southall et al. (2019) are for two different types of noise as follows:

• Impulsive noises which are typically transient, brief (less than one second), broadband, and consist of high peak sound pressure with rapid rise time and rapid decay (ANSI, 1986 and 2005; NIOSH, 1998). This category includes noise sources such as seismic surveys, impact piling and underwater explosions.
• Non-impulsive noises which can be broadband, narrowband or tonal, brief or prolonged, continuous or intermittent and typically do not have a high peak sound pressure with rapid rise/decay time that impulsive noises do (ANSI, 1995; NIOSH, 1998). This category includes noise sources such as continuous running machinery, sonar, and vessels.

29. The criteria for impulsive and non-impulsive noise have been adopted for this study given the nature of the variety of noise source used during the various activities. The relevant criteria proposed by Southall et al. (2019) are as summarised in Table 5.1   Open ▸ and Table 5.2   Open ▸ .

Table 5.1: Summary of PTS Onset Acoustic Thresholds (Southall et al., 2019; Tables 6 and 7)

Table 5.2: Summary of TTS Onset Acoustic Thresholds (Southall et al., 2019; Tables 6 and 7)

5.3.        Disturbance to Marine Mammals

5.3. Disturbance to Marine Mammals

30. Beyond the area in which auditory injury may occur, effects on marine mammal behaviour are an important measure of potential impact. Non-trivial disturbance may occur when there is a risk of animals incurring sustained or chronic disruption of behaviour or when animals are displaced from an area, with subsequent redistribution being significantly different from that occurring due to natural variation.
31. To consider the possibility of disturbance resulting from the Array, it is necessary to consider:

• whether or not a noise can be detected/heard by an animal above background noise levels or level of acclimatisation above background levels;
• the likelihood that the noise could cause non-trivial disturbance;
• the likelihood that the sensitive animals will be exposed to that noise; and
• whether the number of animals exposed are likely to be significant at the population level.

32. Assessing these impacts is however a very difficult task due to the complex and variable nature of noise propagation, the variability of documented animal responses to similar levels of noise, and the availability of population estimates, and regional density estimates for all marine mammal species. Behavioural responses are widely recognised as being highly variable and context specific (Southall et al., 2007; 2019; 2021). Assessing the severity of such potential impacts and development of probability-based response functions continues to be an area of ongoing scientific research interest (Graham et al., 2019; Harris et al., 2018; Southall et al., 2021).
33. Southall et al. (2007) recommended that at the time the only feasible way to assess whether a specific noise could cause disturbance is to compare the circumstances of the situation with empirical studies. Joint Nature Conservation Committee (JNCC) guidance in the United Kingdom (UK) (JNCC, 2010) indicates that a score of five or more on the Southall et al. (2007) behavioural response severity scale could be significant. The more severe the response on the scale, the lower the amount of time that the animals will tolerate it before there could be adverse consequences to life functions, which would constitute a disturbance. The severity scale was revised in Southall et al. (2021), which included splitting severity assessment methods on captive studies from assessments on field studies. Behavioural responses related to field studies included impacts to survival, reproduction, and foraging.
34. Southall et al. (2007 and 2021) both present a summary of observed behavioural responses for various mammal groups exposed to different types of noise: continuous (non-pulsed) or impulsive (single or multiple pulsed).
35. Disturbance to marine mammals is discussed in more detail in volume 2, chapter 10.

5.3.1.    Continuous (Non-Pulsed, Non-Impulsive) Noise

5.3.1. Continuous (Non-Pulsed, Non-Impulsive) Noise

36. For non-pulsed noise (e.g. installation of pile foundations using drilling, vessels etc.), the lowest sound pressure level at which a score of five or more on the Southall et al. (2007) behavioural response severity scale occurs for low frequency cetaceans is 90 dB to 100 dB re 1 μPa (rms). However, this relates to a study involving only migrating gray whales Eschrichtius robustus. A study for minke whale showed a response score of three at a received level of 100 dB to 110 dB re 1 μPa (rms), with no higher severity score encountered for this species. For mid frequency cetaceans, a response score of eight was encountered at a received level of 90 dB to 100 dB re 1 μPa (rms), but this was for one mammal (a sperm whale Physeter macrocephalus) and might not be applicable for the species likely to be encountered in the vicinity of the Array. For Atlantic white-beaked dolphin, a response score of three was encountered for received levels of 110 dB to 120 dB re 1 μPa (rms), with no higher severity score encountered. For high frequency cetaceans such as bottlenose dolphins a number of individual responses with a response score of six are noted ranging from 80 dB re 1 μPa (rms) and upwards. There is a significant increase in the number of mammals responding at a response score of six once the received sound pressure level is greater than 140 dB re 1 μPa (rms).
37. It is worth noting that the above sound pressure levels are based on the rms sound pressure level metric, which was historically often reported in such studies. More recent studies often use other metrics such as the SEL and care must be taken not to directly compare noise levels quoted using different parameters (refer to section 3 for a discussion of these different metrics).
38. The National Marine Fisheries Service (NMFS) (2005) guidance sets the marine mammal Level B harassment threshold (analogous to disturbance) for continuous noise at 120  dB re 1 μPa (rms). This threshold is based on studies by Malme et al. (1984) which investigate the effects of noise from the offshore petroleum industry on migrating gray whale behaviour offshore Alaska. This value sits approximately mid-way between the range of values identified in Southall et al. (2007) for continuous noise but is lower than the value at which the majority of marine mammals responded at a response score of six (i.e. once the received rms sound pressure level is greater than 140 dB re 1 μPa). Considering the paucity and high level variation of data relating to onset of behavioural effects due to continuous noise, any ranges predicted using this number are likely to be probabilistic and potentially over precautionary.
39. It is worth nothing that the distinction between impulsive and non-impulsive noise was removed from Southall et al. (2021) as “some source types, such as airguns, may produce impulsive noises near the source and non-impulsive noises at greater ranges”. However, Southall et al. (2021) does not present thresholds for assessing disturbance, therefore the thresholds discussed in section 5.3.1 have been adopted.

5.3.2.    Impulsive (Pulsed) Noise

5.3.2. Impulsive (Pulsed) Noise

40. Southall et al. (2007) presents a summary of observed behavioural responses due to multiple pulsed noise, although the data is primarily based on responses to seismic exploration activities (rather than for piling). Although these datasets contain much relevant data for LF cetaceans, there is less data for MF or HF cetaceans within the document. Low frequency cetaceans, other than bowhead whales (Balaena mysticetus), were typically observed to respond significantly at 140 dB to 160 dB re 1 μPa (rms). Behavioural changes at these levels during multiple pulses may have included visible startle response, extended cessation or modification of vocal behaviour, brief cessation of reproductive behaviour or brief/minor separation of females and dependent offspring. The data available for MF cetaceans indicate that some significant response was observed at a SPL of 120 dB to 130 dB re 1 μPa (rms), although the majority of cetaceans in this category did not display behaviours of this severity until exposed to a level of 170 dB to 180 dB re 1 μPa (rms). Furthermore, other MF cetaceans within the same study were observed to have no behavioural response even when exposed to a level of 170 dB to 180 dB re 1 μPa (rms).
41. More recently, Graham et al. (2019) describes empirical evidence from piling at the Beatrice Offshore Wind Farm (Moray Firth, Scotland) used to derive a dose-response curve for harbour porpoise[7]. The unweighted single pulse SEL contours were plotted in 5 dB increments and applied the dose-response curve to estimate the number of animals that would be disturbed by piling within each stepped contour. The study shows a 100% probability of disturbance at an (unweighted) SEL of 180 dB re 1 μPa2s, 50% at 155 dB re 1 μPa2s and dropping to approximately 0% at an SEL of 120 dB re 1 μPa2s. This approach to understanding the behavioural effects from piling has been applied at other UK offshore wind farms (for example Seagreen Alpha/Bravo Environmental Statement Chapter 10 Marine Mammals (Seagreen Wind Energy, 2018), Hornsea Three Environmental Statement Volume 2 Chapter 4 Marine mammals (Orsted, 2020) and Awel y Môr Environmental Statement Volume 2, Chapter 7: Marine mammals (RWE, 2022)). Similar stepped/probability-based threshold criteria have been used on other studies such as for assessing the response of marine mammals to geophysical activities (e.g. Southall et al., 2017). The data were subsequently used to develop a dose-response curve. The assessment of behavioural response and disturbance is presented in volume 2, chapter 10.
42. Southall et al. (2007) suggested that there was a general paucity of data relating to the effects of noise on pinnipeds in particular. One study using ringed Pusa hispida, bearded Erignathus barbatus and spotted Phoca largha seals (Harris et al., 2001) found onset of a significant response at a received sound pressure level of 160 dB to 170 dB re 1 μPa (rms), although larger numbers of animals showed no response at noise levels of up to 180 dB re 1 μPa (rms). It is only at much higher sound pressure levels in the range of 190 dB to 200 dB re 1 μPa (rms) that significant numbers of seals were found to exhibit a significant response. For comparison, for non-pulsed noise one study elicited a significant response on a single harbour seal at a received level of 100 dB to 110 dB re 1 μPa (rms), although other studies found no response or non-significant reactions occurred at much higher received levels of up to 140 dB re 1 μPa (rms).
43. In more recent studies, and following a similar method to Graham et al. (2019), a telemetry study undertaken by Russell et al. (2016) investigating the behaviour of tagged harbour seals during pile driving at the Lincs Wind Farm in the Wash found that there was a proportional response at different received noise levels. Dividing the study area into a 5 km x 5 km grid, the authors modelled SELss levels and matched these to corresponding densities of harbour seals in the same grids during periods of non-piling versus piling to show change in usage. The study found that there was a significant decrease during piling activities at predicted received SEL levels of between 142 dB re 1 µPa2s and 151 dB re 1 µPa2s.
44. Southall et al. (2007) also noted that due to the uncertainty over whether HF cetaceans may perceive certain noises and due to the paucity of data, it was not possible to present any data on responses of HF cetaceans. However, Lucke et al. (2009) showed a single harbour porpoise consistently showed aversive behavioural reactions to pulsed noise at received SPL above 174 dB re 1 μPa (peak-to-peak) or a SEL of 145 dB re 1 μPa2s, equivalent to an estimated[8]  sound pressure level of 166 dB re 1 μPa rms.
45. There is much intra-category and perhaps intra-species variability in behavioural response. As such, a conservative approach should be taken to ensure that the most sensitive marine mammals remain protected.
46. The High Energy Seismic Survey (HESS) workshop on the effects of seismic (i.e. pulsed) noise on marine mammals (HESS, 1997) concluded that mild behavioural disturbance would most likely occur at rms noise levels greater than 140 dB re 1 μPa (rms). This workshop drew on studies by Richardson (1995) but recognised that there was some degree of variability in reactions between different studies and mammal groups. Consequently, for the purposes of this study, a precautionary level of 140 dB re 1 μPa (rms) is used to indicate the onset of low-level marine mammal disturbance effects for all mammal groups for impulsive noise.
47. The approach to be employed for the Array is therefore to plot unweighted single pulse SEL contours in 5 dB increments and apply the appropriate dose-response curve to estimate the number of animals that would be disturbed by noise from the piling within each stepped contour. For cetaceans, the dose-response curve will be applied from the Beatrice data (Graham et al., 2019) (refer to Figure 5.2   Open ▸ ), whilst for pinnipeds the dose-response curve will be applied using Whyte et al. (2020). Although the Whyte paper derives more recent response curves, these are only proposed for pinnipeds and hence the need to also include data from older sources for other key species.

Figure 5.2: The Probability of a Harbour Porpoise Response (24 hrs) in Relation to the Partial Contribution of Unweighted Received Single-Pulse SEL for the First Location Piled (Purple Line), the Middle Location (Green Line) and the Final Location Piled (Blue Line). Reproduced with Permission from Graham et al. (2019)

48. The dose-response approach is a widely accepted approach to assessing potential behavioural effects of noise from piling and has been applied at other recent UK offshore wind farms (e.g. Seagreen Alpha/Bravo, Awel y Môr, Hornsea Three and Hornsea Four).
49. For impulsive noise sources other than piling (e.g. UXO clearance, some geotechnical and geophysical surveys), this assessment adopts the NMFS (2005) Level B harassment threshold of 160 dB re 1 μPa (rms) for impulsive noise. Level B harassment is defined by NMFS (2005) as having the potential to disturb a marine mammal or marine mammal stock in the wild by causing disruption of behavioural patterns, including, but not limited to, migration, breathing, nursing, breeding, feeding, or sheltering but which does not have the potential to injure a marine mammal or marine mammal stock in the wild. This is similar to the JNCC (2010) description of non-trivial disturbance and has therefore been used as the basis for onset of behavioural change in the assessment.
50. For assessing the severity of behavioural response, the distinction between impulsive and non-impulsive noise was removed from Southall et al. (2021) as “some source types, such as airguns, may produce impulsive noises near the source and non-impulsive noises at greater ranges”. Southall et al. (2021) instead assigns categories to various sources based on the operational characteristics and applies revised severity assessments to selected studies in each category. For example, Table 7 within that paper details a number of observational studies of marine mammals and their responses to piling, with an indication of severity of response and in some cases a received level. However, Southall et al. (2021) does not present thresholds for assessing disturbance, therefore the thresholds discussed in Table 5.3   Open ▸ have been adopted for this study. The assessment of disturbance and behavioural response is presented in full in volume 2, chapter 10.
51. A recent position statement from Natural Resources Wales (NRW, 2023) presents a number of disturbance criteria specifically for assessing the impacts on harbour porpoise. This document recommends as a first instance using the 143 dB re 1 µPa2s SEL­ss contour as an indicator of disturbance from pile driving, as proposed by Tougaard (2021). This value is based on measurements undertaken to inform the changes to guidelines from the Danish Energy Agency, and the contour was seen to extend to 20 km to 30 km from the piling site.
52. It is important to understand that exposure to noise levels in excess of the behavioural change threshold stated above does not necessarily imply that the noise will result in significant disturbance. As noted previously, it is also necessary to assess the likelihood that the sensitive receptors will be exposed to that noise and whether the numbers exposed are likely to be significant at the population level.

Table 5.3: Disturbance Criteria for Marine Mammals Used in this Technical Report

53. There is, however, a considerable degree of uncertainty and variability in the onset of disturbance and therefore any disturbance ranges should be treated as potentially over precautionary. Another important consideration is that the majority of noise produced by project activities, with the exception of operational wind turbine noise, will be either temporary or transitory, as opposed to permanent and fixed. These important considerations are not taken into account in the noise modelling but will be assessed in the relevant marine ecology topic chapters.

5.4.        Injury and Disturbance to Fish

5.4. Injury and Disturbance to Fish

54. For fish, the most relevant criteria for injury effects are considered to be those contained in the Noise Exposure Guidelines for Fishes and Sea Turtles (Popper et al., 2014). These guidelines broadly group fish into the following categories based on their anatomy and the available information on hearing of other fish species with comparable anatomies:

• Group 1: fishes with no swim bladder or other gas chamber (e.g. elasmobranchs, flatfishes and lampreys). These species are less susceptible to barotrauma and are only sensitive to particle motion, not sound pressure. Basking shark Cetorhinus maximus, which do not have a swim bladder, also fall into this hearing group.
• Group 2: fishes with swim bladders but the swim bladder does not play a role in hearing (e.g. salmonids). These species are susceptible to barotrauma, although hearing only involves particle motion, not sound pressure.
• Group 3: fishes with swim bladders that are close, but not connected, to the ear (e.g. gadoids and eels). These fishes are sensitive to both particle motion and sound pressure and show a more extended frequency range than Groups 1 and 2, extending to about 500 Hz.
• Group 4: fishes that have special structures mechanically linking the swim bladder to the ear (e.g. clupeids such as herring, sprat and shads). These fishes are sensitive primarily to sound pressure, although they also detect particle motion. These species have a wider frequency range, extending to several kHz and generally show higher sensitivity to sound pressure than fishes in Groups 1, 2 and 3.
• Fish eggs and larvae: separated due to greater vulnerability and reduced mobility. Very few peer-reviewed studies report on the response of eggs and larvae to anthropogenic noise.

55. The guidelines set out criteria for injury effects due to different sources of noise. Those relevant to the Array are considered to be those for impulsive piling sources only, as non-impulsive sources were not considered to be a key potential impact and therefore were screened out of the guidance[9]. The criteria include a range of indices including SEL, rms and peak SPLs. Where insufficient data exist to determine a quantitative guideline value, the risk is categorised in relative terms as “high”, “moderate” or “low” at three distances from the source: “near” (i.e. in the tens of metres), “intermediate” (i.e. in the hundreds of metres) or “far” (i.e. in the thousands of metres). It should be noted that these qualitative criteria cannot differentiate between exposures to different noise levels and therefore all sources of noise, no matter how loud, would theoretically elicit the same assessment result. However, because the qualitative risks are generally qualified as “low”, with the exception of a moderate risk at “near” range (i.e. within tens of metres) for some types of hearing groups and impairment effects, this is not considered to be a significant issue with respect to determining the potential effect of noise on fish.
56. The injury threshold criteria used in this underwater noise assessment for impulsive piling are given in Table 5.4   Open ▸ . In the table, both peak and SEL criteria are unweighted. Physiological effects relating to injury criteria are described below (Popper et al., 2014; Popper and Hawkins, 2016):

• Mortality and potential mortal injury: either immediate mortality or tissue and/or physiological damage that is sufficiently severe (e.g. a barotrauma) that death occurs sometime later due to decreased fitness. Mortality has a direct effect upon animal populations, especially if it affects individuals close to maturity.
• Recoverable injury: Tissue and other physical damage or physiological effects, that are recoverable, but which may place animals at lower levels of fitness, may render them more open to predation, impaired feeding and growth, or lack of breeding success, until recovery takes place.
• TTS: Short term changes in hearing sensitivity may, or may not, reduce fitness and survival. Impairment of hearing may affect the ability of animals to capture prey and avoid predators, and also cause deterioration in communication between individuals affecting growth, survival, and reproductive success. After termination of a noise that causes TTS, normal hearing ability returns over a period that is variable, depending on many factors, including the intensity and duration of noise exposure.

Table 5.4: Criteria for Onset of Injury to Fish Due to Impulsive Piling (Popper et al., 2014)

57. The criteria used in this underwater noise assessment for non-impulsive piling and other continuous noise sources, such as vessels, are given in Table 5.5   Open ▸ . The only numerical criteria for these sources are for recoverable injury and TTS for Groups 3 and 4 Fish.

Table 5.5: Criteria for Onset of Injury to Fish Due to Non-Impulsive Noise (Popper et al., 2014)

58. The criteria used in this underwater noise assessment for explosives are given in Table 5.6   Open ▸ . It should be noted that there are no thresholds in Popper et al. (2014) in relation to eggs and larvae in terms of sound pressure.

Table 5.6: Criteria for Injury to Fish Due to Explosives (Popper et al., 2014)

59. It should be noted that there are no thresholds in Popper et al. (2014) in relation to noise from high frequency sonar (>10 kHz). This is because the hearing range of fish species falls well below the frequency range of high frequency sonar systems. Consequently, the effects of noise from high frequency sonar surveys on fish has not been conducted as part of this study, due to the frequency of the source being beyond the range of hearing and also due to the lack of any suitable thresholds.
60. Behavioural reaction of fish to noise has been found to vary between species based on their hearing sensitivity. Typically, fish sense noise via particle motion in the inner ear which is detected from noise-induced motions in the fish’s body (refer to section 9 for further details on particle motion). The detection of sound pressure is restricted to those fish which have air filled swim bladders; however, particle motion (induced by noise) can be detected by fish without swim bladders[10].
61. Highly sensitive species such as herring have elaborate specialisations of their auditory apparatus, known as an otic bulla – a gas filled sphere, connected to the swim bladder, which enhances hearing ability. The gas filled swim bladder in species groups such as cod and salmon may be involved in their hearing capabilities, so although there is no direct link to the inner ear, these species are able to detect lower noise frequencies and as such are considered to be of medium sensitivity to noise. Flat fish and elasmobranchs have no swim bladders and as such are considered to be relatively less sensitive to sound pressure.
62. The most recent criteria for disturbance are considered to be those contained in Popper et al. (2014) which set out qualitative criteria for disturbance due to different sources of noise. The risk of behavioural effects is categorised in relative terms as “high”, “moderate” or “low” at three distances from the source: “near” (i.e. in the tens of metres), “intermediate” (i.e. in the hundreds of metres) or “far” (i.e. in the thousands of metres), as shown in Table 5.7   Open ▸ .

Table 5.7: Criteria for Onset of Behavioural Effects in Fish for Impulsive and Non-Impulsive Noise (Popper et al., 2014)

63. It is important to note that the Popper et al. (2014) criteria for disturbance due to noise are qualitative rather than quantitative. Consequently, a source of noise of a particular type (e.g. piling) would be predicted to result in the same potential impact, no matter the level of noise produced or the propagation characteristics.
64. Therefore, the criteria presented in the Washington State Department of Transport (WSDOT) Biological Assessment Preparation for Transport Projects Advanced Training Manual (WSDOT, 2011) are also used in this assessment for predicting the distances at which behavioural effects may occur due to noise from impulsive piling. The manual suggests an unweighted sound pressure level of 150 dB re 1 μPa (rms) as the criterion for onset of behavioural effects, based on work by (Hastings, 2002). Sound pressure levels in excess of 150 dB re 1 μPa (rms) are expected to cause temporary behavioural changes, such as elicitation of a startle response, disruption of feeding, or avoidance of an area. The document notes that levels exceeding this threshold are not expected to cause direct permanent injury but may indirectly affect the individual fish (such as by impairing predator detection). It is important to note that this threshold is for onset of potential effects, and not necessarily an ‘adverse effect’ threshold.

5.5.        Use of Impulsive Noise Thresholds at Large Ranges

5.5. Use of Impulsive Noise Thresholds at Large Ranges

65. For any noise of a given amplitude and frequency content, impulsive noise has a greater potential to cause auditory injury than a similar magnitude non-impulsive noise (Southall et al., 2007; 2019; 2021; NMFS, 2018; von Benda-Beckmann et al., 2022). For highly impulsive noises such as those generated by impact piling, UXO detonations and seismic source arrays, the interaction with the seafloor and the water column is complex. In these cases, due to a combination of dispersion (i.e. where the waveform elongates), multiple reflections from the sea surface and seafloor and molecular absorption of high frequency energy, the noise is unlikely to still be impulsive in character once it has propagated some distance (Hastie et al., 2019; Martin et al., 2020; Southall et al., 2019; Southall, 2021). This transition in the acoustic characteristics therefore has implications with respect to which threshold values should be used (impulsive vs. non impulsive criteria) and, consequently, the distances at which potential injury effects may occur.
66. This acoustic wave elongation effect is particularly pronounced at larger ranges of several kilometres and, in particular, it is considered highly unlikely that predicted PTS or TTS ranges for impulsive noise which are found to be in the tens of kilometres are realistic (Southall, 2021). However, the precise range at which the transition from impulsive to non-impulsive noise occurs is difficult to define precisely, not least because the transition also depends on the response of the marine mammals’ ear. Consequently, there is currently no consensus as to the range at which this transition occurs or indeed the measure of impulsivity which can be used to determine which threshold should be applied (Southall, 2021). However, evidence for impact pile driving and seismic source arrays does indicate that some measures of impulsivity change markedly within 10 km of the source (Hastie et al., 2019). Additionally, the draft NMFS (2018) guidance suggested 3 km as a transition range, but this was removed from the final document.
67. The cross-over between impulsive and non-impulsive noise is an area of ongoing research and there are a number of potential methods for determining the cross-over point being investigated, such as the kurtosis metric, and the loss of high frequency energy from the spectrum (above 10 kHz, e.g. Southall, 2021). In the meantime, it is considered that any predicted injury ranges in the tens of kilometres are almost certainly an overly precautionary interpretation of existing criteria (Southall, 2021).
68. As disturbance ranges are likely to extend beyond the range at which injury (PTS or TTS) could occur, this transition from impulsive to continuous noise is likely to have even more impact on the disturbance range (e.g. Southall et al., 2021). For example, where dose-response relationships have been derived based on exposure to impulsive noises, particularly where these have been derived based on experiments relatively close to the impulsive source, then extrapolation of the dose-response relationship to larger ranges could be misleading. This is particularly true where the dose-response relationship has been derived using parameters such as unweighted single pulse SEL or rms(T90) SPL, which does not take into account the characteristics (e.g. frequency content of impulsivity) of the noise. Consequently, great caution should be used when interpreting potential disturbance ranges in the order of tens of kilometres. Where appropriate, these should be considered alongside an understanding of potential background noise levels in order to understand the distances at which noises related to an impulsive source may be detected.

6.             Source Noise Levels

6. Source Noise Levels

6.1.        General

6.1. General

69. Underwater noise source level is usually quantified using a dB scale with values generally referenced to 1 μPa pressure amplitude as if measured at a distance of 1 m from a hypothetical, infinitesimally small point source (sometimes referred to as the SL). This quantity is often referred to as an equivalent monopole source level. In practice, it is not usually possible to measure noise at 1 m from a large structure, which is more akin to a distributed noise source, but the source level metric allows comparisons and reporting of different source noise emissions on a like-for-like basis, as well as a standard input parameter for noise propagation models. In reality, for a large noise source such as a monopile, seismic source array or vessel, the source level value at this conceptual point at 1 m from the (theoretical, infinitesimally small) acoustic centre does not exist. Furthermore, the energy is distributed across the source and does not all emanate from this imagined acoustic centre point. Therefore, the stated sound pressure level at 1 m does not occur at any point in space for these large sources. In the acoustic near field (i.e. close to the source), the sound pressure level will be significantly lower than the value predicted by the SL.
70. A wealth of experimental data and literature-based information is available for quantifying the noise emission from different construction operations. This information, which allows us to predict with a good degree of accuracy the noise generated by a source at discrete frequencies in one-third octave bands, will be employed to characterise their acoustic emission in the underwater environment. Sections 6.2 to 6.7 detail the types of noise sources present during different construction activities, their potential signatures in different frequency bands, and acoustic levels.

6.2.        Types of Noise Sources

6.2. Types of Noise Sources

71. The noise sources and activities which were investigated during the development of this technical report are summarised in Table 6.1   Open ▸ .

Table 6.1: Summary of Noise Sources and Activities Included in the Underwater Noise Technical Report

72. Noise sources included in Table 6.1   Open ▸ are considered in more detail in the following sections.

6.3.        Pre-construction Phase

6.3. Pre-construction Phase

6.3.1.    Geophysical Surveys

6.3.1. Geophysical Surveys

73. Several sonar-like survey types will potentially be used for the pre-construction site investigation geophysical surveys. During the survey, a transmitter emits an acoustic signal directly toward the seabed (or alongside, at an angle to the seabed, in the case of side scan techniques). The equipment likely to be used can typically work at a range of signal frequencies, depending on the distance to the bottom and the required resolution. The signal is highly directional and acts as a beam, with the energy narrowly concentrated within a few degrees of the direction in which it is aimed. The signal is emitted in pulses, the length of which can be varied as per the survey requirements. The assumed pulse rate, pulse width and beam width used in the assessment are based on a review of typical units used in other similar surveys. It should be noted that sonar like survey sources are classed as non-impulsive noise because they generally comprise a single (or multiple discrete) frequency (e.g. a sine wave or swept sine wave) as opposed to a broadband signal with high kurtosis, high peak pressures and rapid rise times.
74. The characteristics assumed for each device modelled in this technical report are summarised in Table 6.2   Open ▸ . these sources are considered to be continuous (non-impulsive).

Table 6.2: Typical Sonar Based Survey Equipment Parameters Used in Assessment

75. The assumed pulse rate has been used to calculate the SEL, which is normalised to 1 s, from the rms sound pressure level. Directivity corrections were calculated based on the transducer dimensions and ping frequency and taken from manufacturer’s datasheets. It is important to note that directivity will vary significantly with frequency, but that these directivity values have been used in line with the modelling assumptions stated in Table 6.2   Open ▸ .
76. Directivity corrections have been applied to the source noise level data based on directivity characteristics for the proposed sources. Directivity factors were derived based on source take-off angle for an animal on the seabed. This results in a larger correction (reduction in level) due to directivity at distances further from the source than for receivers close to the source.
77. At distances closer to the source (i.e. less than the water depth), no directivity correction is made because the animal could be directly underneath the source. As the source to receiver range increases, the take-off angle between the source and animal becomes larger. Hence, when the range to source is large in comparison to the water depth, the effects of the source's directivity will have a much greater bearing on the received noise level. Once the range to source becomes larger than the water column depth then the source directivity effects will become increasingly more important.
78. Unlike the sonar like survey sources, the UHRS source is likely to utilise a sparker, which produces an impulsive, broadband source signal. The parameters used in the underwater noise modelling are summarised in Table 6.3   Open ▸ .

Table 6.3: Typical UHRS Survey Equipment Parameters Used in Assessment

6.3.2.    Geotechnical Surveys

6.3.2. Geotechnical Surveys

79. Source noise data for the proposed CPTs was reported by Erbe and McPherson (2017). In this report, the SEL measurements at two different sites in Western Australia at a measured distance of 10 m were presented. The signature is generally broadband in nature with levels measured generally 20 dB above the baseline noise levels. The report also refers to other paths for acoustic energy including direct air to water transmission and other multipath directions, which implied that measured noise level is strongly dependant on depth and range from the source. The third octave band SEL levels from the CPT extracted are presented in Table 6.4   Open ▸ .

Table 6.4: CPT Source Levels in Different Third Octave Band Frequencies (SEL Metric) Used for the Assessment (Erbe and McPherson, 2017)

80. Seismic CPT noise is classified as impulsive at source since it has a rapid rise time and a high peak sound pressure level of 220 dB re 1 µPa (pk), compared to a SEL of 189 dB re 1 µPa2s.
81. The seismic CPT test is typically conducted at various depths for each location every three to five minutes with between 10 and 20 strikes per depth.
82. It should be noted that if non-seismic CPT were to be used, the noise would be considered non-impulsive if it produced any noise at all, and therefore the assessment of seismic CPT is considered precautionary. As piston core and box core methods are lower in sound energy, CPT has been used as a maximum design scenario across all core measurement techniques.
83. Measurements of a vibro-core test show underwater source sound pressure levels of approximately 187 dB re 1 µPa re 1 m (rms) (Reiser et al., 2011). The SEL has been calculated based on a one hour sample time which, it is understood, is the typical maximum time required for each sample. The vessel would then move on to the next location and take the next sample with approximately one hour break between each operation. The vibro-core noise is considered to be continuous (non-impulsive).

Table 6.5: Vibro-Core Source Levels Used in the Assessment

84. The frequency spectral shape for vibro-coring is presented in Figure 6.1   Open ▸ .

Figure 6.1: Frequency Spectral Shape Used for Vibro-Coring

85. Source levels for borehole drilling was reported in Erbe and McPherson (2017), with source levels of 142 dB to 145 dB re 1 µPa re 1 m (rms). A set of one third octave band levels, calculated from the spectrum presented in the paper are shown in Figure 6.2   Open ▸ .

Figure 6.2: Borehole Drilling Source Level Spectrum Shape Used in the Assessment

86. As for other non-impulsive sources, the impact assessment criteria is the SEL metric for a receptor moving away from the source.

6.3.3.    UXO Clearance

6.3.3. UXO Clearance

87. The precise details and locations of potential UXOs is unknown at this time. For the purposes of this assessment, it has been assumed that the Maximum Design Scenario (MDS) will be clearance of UXO with a Net Explosive Quantity (NEQ) of 698 kg cleared by either low-order or high order techniques. Low-order techniques are not always possible and are dependent upon the individual situations surrounding each UXO. The value of 698 kg is derived from a site-specific desktop review undertaken by Ordtek (2022) to understand the potential maximum charge weight and likelihood based on past military activity.
88. There are a number of low-order and low-yield techniques available for the clearance of UXO, with the development of new techniques being a subject of ongoing research. For example, one such technique (deflagration) uses a single charge of 30 g to 80 g NEQ which is placed in close proximity to the UXO to target a specific entry point. When detonated, a shaped charge penetrates the casing of the UXO to introduce a small, clinical plasma jet into the main explosive filling. The intention is to excite the explosive molecules within the main filling to generate enough pressure to burst the UXO casing, producing a deflagration of the main filling and neutralising the UXO.
89. Recent controlled experiments showed low-order deflagration to result in a substantial reduction in acoustic output over traditional high order methods, with SPLpk and SEL being typically significantly lower for the deflagration of the same size munition, and with the acoustic output being proportional to the size of the shaped charge, rather than the size of the UXO itself (Robinson et al., 2020). Using this low-order deflagration method, the probability of a low order outcome is high; however, there is a small inherent risk with these clearance methods that the UXO will detonate or deflagrate violently resulting in higher noise level emissions.
90. It is possible that there will be residual explosive material remaining on the seabed following the use of low-order techniques for unexploded ordnance disposal. In this case, and only for debris of sufficient size to be a risk to fishing activities, recovery will be performed which includes the potential use of a small (500 g) ‘clearing shot’.
91. As a last resort, if it is not possible to carry out low-order or low-yield clearance techniques, it may be necessary to carry out a high order detonation of the UXO. The underwater noise modelling has been undertaken for a range of charge configurations as set out in Table 6.6   Open ▸ .

Table 6.6: Details of UXO and their Relevant Charge Sizes Employed for Modelling

92. The source levels for UXO are included within the terms for propagation modelling and are described in section 7.7.

6.3.4.    Vessels

6.3.4. Vessels

93. Use of vessels is addressed in section 6.7 for all phases of the Array.

6.4.        Construction Phase

6.4. Construction Phase

6.4.1.    Impact Piling

6.4.1. Impact Piling

94. The noise generated and radiated by a pile as it is driven into the ground is complex, due to the many components which constitute the generation and radiation mechanisms. Larger pile sizes can require a higher energy in order to drive them into the seabed, and different seabed and underlying substrate types can require use of different installation techniques including varying the hammer energies and the number of hammer strikes. In addition, the seabed characteristics can affect how noise propagates from the pile through the sub-surface geology, thus fundamentally affecting the acoustic field around the activity. The type of hammer method used (i.e. the force-impulse characteristics) can also affect the noise characteristics.

A useful measure of noise used in underwater acoustics is the SEL. This descriptor is used as a measure of the total noise energy of an event or a number of events (e.g. over the course of a day) and is normalised to one second. This allows the total acoustic energy contained in events lasting a different amount of time to be compared on a like for like basis. For impulsive noises it has become customary to utilise the T90 time period for calculating and reporting root mean squared (rms) sound pressure levels. This is the interval over which the cumulative energy curve rises from 5% to 95% of the total energy and therefore contains 90% of the noise energy.

95. The estimation of source levels for noise propagation modelling of piling is an important aspect of the noise modelling methodology. Ideally, use can be made of noise data measurement for similar piles, installed using a similar hammer in similar conditions. However, since noise modelling for proposed wind farms often proposes the use of piles and hammers for which there is no currently available measured data, it is often necessary to utilise an alternative method to estimate the source level inputs to the model. One such method used in some previous noise modelling assessments is the use of energy conversion factors, which involves estimating the proportion of the hammer energy which is transmitted into the water column as noise. However, the subject of noise generation due to impact piling is an active area of research and the evidence base is constantly being updated by new measurements, research and published papers. It is therefore important to ensure that the methodology used for determining the source levels of piling take into account the most recent research.
96. It is proposed to utilise scaling of measured data during pile driving for similar operations to the Array in order to determine source levels. The subject of noise generation due to impact piling is an active area of research and the evidence base is constantly being updated by new measurements, research and published papers. A recent peer-reviewed paper (von Pein et al., 2022) presents a methodology for the dependencies of the SEL on strike energy, diameter, ram weight, and water depth that can be used for scaling measured or computed SELs from one project to another. The method has been shown to be usable within practical ranges of accuracy, especially if the measurement uncertainties are taken into account. The paper suggests that scaling should be performed over either a small number of very similar piling situations or over a larger data set with according averaging. This is a recently published method for deriving the noise source level which provides a more scientifically robust method compared to using an energy conversion factor (the conversion factor method simply assumes that a percentage of the hammer energy is converted into noise irrespective of parameters such as pile size, water depth and hammer specifications). Since the von Pein et al. (2022) methodology takes into account several site-specific and pile-specific factors, in addition to hammer energy, and because it is based on a scientifically rigorous and peer reviewed study, it is therefore considered to be a significant improvement on the use of simple conversion factors alone. This methodology is further endorsed by the recent study undertaken by Jasco on behalf of Marine Scotland, which recommends use of scaling methods instead of those relying on conversion factors (Wood et al. 2023).
97. Using the equation below (von Pein et al. 2022), a broadband source level value is calculates for the noise emitted during impact pile driving operation in each operation window.

98. In this equation, E is the hammer energy employed in Joules, d is the pile diameter, mr is the ram mass in kg, h is the water depth in m, is the reflection coefficient and is the propagation angle (approximately 17° for a Mach wave generated by impact piling). The equation allows measured pile noise data from one site (denoted by subscript 0) to be scaled to another site (denoted by subscript 1).
99. To account for the pile penetration and use of submerged piling rigs, a correction is applied through the piling sequence based on Lippert et al. (2017) by considering the quotient of wetted pile length Lw and water depth hw using the following equation:

100. This methodology therefore considers the following factors:

• pile diameter;
• pile length;
• pile penetration;
• water depth;
• rated maximum hammer energy of the proposed hammer;
• hammer energy being used;
• ram mass for the hammer; and
• acoustical parameters of the soil and water.

101. The peak SPL can be calculated from SEL values via the empirical fitting between pile driving SEL and peak SPL data, given in Lippert et al. (2015), as:

SPLpk = .

102. Root mean square (rms) sound pressure levels were calculated assuming a typical T90 pulse duration (i.e. the period that contains 90% of the total cumulative noise energy) of 100 ms. It should be noted that in reality the rms T90 period will increase significantly with distance which means that any ranges based on rms sound pressure levels at ranges of more than a few kilometres are likely to be significant overestimates and should therefore be treated as highly conservative.
103. The piling scenarios for the Array include the following phases:

• initiation (including slow-start);
• soft start;
• ramp up; and
• full power piling.

104. These phases and the various associated parameters and durations are shown in Table 6.7   Open ▸ .
105. The impact piling scenarios that have been modelled for the Array are as follows:

• OSP foundations (piled jacket) using a maximum hammer energy of 4,400 kJ for a duration of up to eight hours, related to the MDS associated with the largest diameter OSP foundations.
• Floating wind turbine foundations anchor piles using a maximum hammer energy of 3,000 kJ for a duration of up to eight hours, related to the MDS associated with the largest diameter wind turbine anchors.

Table 6.7: Modelled Pile Installation Parameters (Pile Diameter 4.5 m for Both Pile Types)

6.4.2.    Drilled Pile Installation

6.4.2. Drilled Pile Installation

106. For drilled pile installation, source noise levels have been based on pile drilling for the Oyster 800 project (Kongsberg, 2011). The hydraulic rock breaking source noise levels are based on those measured by Lawrence (2016). The source levels used in the assessment are summarised in Table 6.8   Open ▸ .
107. Rotary drilling is non-impulsive in character and therefore the non-impulsive injury and behavioural thresholds have been adopted for the assessment.

Table 6.8: Drilled Pile Noise Source Levels Used in Assessment (Un-Weighted)

108. The other noise source potentially active during the construction phase are related to cable installation (i.e. trenching and cable laying activities), and their related operations such as the jack-up rigs. The source levels are presented in Table 6.9   Open ▸ .

Table 6.9: Source Levels for Other Sources

6.4.3.    Vessels

6.4.3. Vessels

109. Use of vessels is addressed in section 6.7 for all phases of the Array.

6.5.        Operation and Maintenance Phase

6.5. Operation and Maintenance Phase

6.5.1.    Operational Noise From wind Turbines

6.5.1. Operational Noise From wind Turbines

110. Operational noise due to the floating wind turbines and mooring lines is dealt with qualitatively in section 8.3.2.

6.5.2.    Geophysical Surveys

6.5.2. Geophysical Surveys

111. Routine geophysical surveys will be similar to the geophysical surveys already discussed for the pre-construction phase (refer to section 6.3).

6.5.3.    Routine Operation and Maintenance

6.5.3. Routine Operation and Maintenance

112. There are very few activities during the operations and maintenance phase that generate significant amounts of underwater noise. These noise generating activities are anticipated at this stage to be characterised by vessel movements.

6.5.4.    Vessels

6.5.4. Vessels

113. The potential for vessel use to create underwater noise is presented in section 6.7 for all phases of the Array.

6.6.        Decommissioning Phase

6.6. Decommissioning Phase

6.6.1.    Vessels

6.6.1. Vessels

114. Only the potential impact of noise from vessel activity has been included in the underwater noise assessment for the decommissioning phase of the Array. It should be noted that cavitation from the vessels themselves is likely to dominate the noisescape for other decommissioning activities (e.g. removal of subsea structures). The potential impact of vessels noise emissions is addressed in section 6.7 for all phases of the Array.

6.7.        Vessels (All Phases)

6.7. Vessels (All Phases)

115. The noise emissions from the types of vessels that may be used for the Array are quantified in Table 6.10   Open ▸ , based on a review of publicly available data. Noise from the vessels themselves (e.g. propeller, thrusters and sonar (if used)) primarily dominates the emission level, hence noise from activities such as seabed preparation, trenching and rock placement (if required) have not been included separately.
116. In Table 6.10   Open ▸ , SELs have been estimated for each source based on 24 hours continuous operation, although it is important to note that it is highly unlikely that any marine mammal or fish would stay at a stationary location or within a fixed radius of a vessel (or any other noise source) for 24 hours. Consequently, the acoustic modelling has been undertaken based on an animal swimming away from the source (or the source moving away from an animal).
117. Source noise levels for vessels depend on the vessel size and speed as well as propeller design and other factors. There can be considerable variation in noise magnitude and character between vessels even within the same class. Therefore, source data for the Array has been based on MDS assumptions (i.e. using noise data toward the higher end of the scale for the relevant class of ship as a proxy). In the case of the cable laying vessel, no publicly available information was available for a similar vessel and therefore measurements on a suction dredger using Dynamic Positioning (DP) thrusters were used as a proxy. This is considered an appropriate proxy because it is a similar size of vessel using dynamic positioning and therefore likely to have a similar acoustic footprint.

Table 6.10: Source Noise Data for Site Preparation, Construction, Operation and Maintenance and Decommissioning Vessels

7.             Propagation Modelling

7. Propagation Modelling

7.1.        Propagation of Noise Underwater

7.1. Propagation of Noise Underwater

118. As the distance from the noise source increases the level of received or recorded noise reduces, primarily due to the spreading of the noise energy with distance, in combination with attenuation due to absorption of noise energy by molecules in the water. This latter mechanism results in higher attenuation at higher frequency noise than for lower frequencies.
119. The way that the noise spreads (geometrical divergence) will depend upon several factors such as water column depth, pressure, temperature gradients, salinity as well as water surface and bottom (i.e. seabed) conditions. Thus, even for a given locality, there are temporal variations to the way that noise will propagate. However, in simple terms, the noise energy may spread out in a spherical pattern (close to the source) or a cylindrical pattern (much further from the source), although other factors mean that decay in noise energy may be somewhere between these two simplistic cases. The distance at which cylindrical spreading dominates is highly dependent on water depth. Noise propagation in shallow water depths will be dominated by cylindrical spreading as opposed to spherical spreading.
120. In acoustically shallow waters[11] in particular, the propagation mechanism is influenced by multiple interactions with the seabed and the water surface (Lurton, 2002; Etter, 2013; Urick, 1983; Brekhovskikh et al, 2003; Kinsler et al., 1999). Whereas in deeper waters, the noise will propagate further without encountering the surface or bottom of the sea (seabed).
121. At the sea surface, the majority of the noise is reflected into the water due to the difference in acoustic impedance (i.e. product of noise speed and density) between air and water. However, the scattering of noise at the surface of the sea can be an important factor in the propagation of noise. In an ideal case (i.e. for a perfectly smooth sea surface), the majority of noise energy will be reflected into the sea. However, for rough seas, much of the noise energy is scattered (e.g. Eckart, 1953; Fortuin, 1970; Marsh, Schulkin, and Kneale, 1961; Urick and Hoover, 1956). Scattering can also occur due to bubbles near the surface such as those generated by wind or fish or due to suspended solids in the water such as particulates and marine species. Scattering is more pronounced for higher frequencies than for low frequencies and is dependent on the sea state (i.e. wave height). However, the various factors affecting this mechanism are complex.
122. As surface scattering results in differences in reflected noise, its effect will be more apparent at longer ranges from the noise source and in acoustically shallow water (i.e. where there are multiple reflections between the source and receiver). The degree of scattering will depend upon the sea state/wind speed, water depth, frequency of the noise, temperature gradient, grazing angle and range from source. It should be noted that variations in propagation due to scattering will vary temporally within an area primarily due to different sea-states/wind speeds at different times. However, over shorter ranges (e.g. several hundred meters or less) the noise will experience fewer reflections and so the effect of scattering should not be significant.
123. When noise waves encounter the seabed, the amount of noise reflected will depend on the geoacoustic properties of the bottom (e.g. grain size, porosity, density, noise speed, absorption coefficient and roughness) as well as the grazing angle and frequency of the noise (Cole, 1965; Hamilton, 1970; Mackenzie, 1960; McKinney and Anderson, 1964; Etter, 2013; Lurton, 2002; Urick, 1983). Thus, seabeds comprising primarily mud or other acoustically soft sediments will reflect less noise than acoustically harder bottoms such as rock or sand. This will also depend on the profile of the bottom (e.g. the depth of the sediment layer and how the geoacoustic properties vary with depth below the seafloor). The effect is less pronounced at low frequencies (a few kHz and below). A scattering effect (similar to that which occurs at the surface) also occurs at the seabed (Essen, 1994; Greaves and Stephen, 2003; McKinney and Anderson, 1964; Kuo, 1992), particularly on rough substrates (e.g. pebbles).
124. The waveguide effect should also be considered, which defines the shallow water columns that do not allow the propagation of low frequency noise (Urick, 1983; Etter, 2013). The cut-off frequency of the lowest mode in a channel can be calculated based on the water depth and knowledge of the sediment geoacoustic properties but, for example, the cut-off frequency as a function of water depth (based on the equations set out in Urick, 1983) is shown in Figure 7.1   Open ▸ for a range of seabed types. Any noise below this frequency will not propagate far due to energy losses through multiple reflections.

Figure 7.1: Lower Cut-Off Frequency as a Function of Depth for a Range of Seabed Types

125. Changes in the water temperature and the hydrostatic pressure with depth mean that the speed of noise varies throughout the water column. This can lead to significant variations in noise propagation and can also lead to noise channels, particularly for high-frequency noise (Lurton 2002). Noise can propagate in a duct-like manner within these channels, effectively focussing the noise, and conversely, they can also lead to shadow zones. The frequency at which this occurs depends on the characteristics of the noise channel and since the temperature gradient can vary throughout the year there will be potential variation in noise propagation depending on the season.
126. Noise energy is also absorbed due to interactions at the molecular level converting the acoustic energy into heat (Urick 1983). This is another frequency-dependent effect with higher frequencies experiencing much higher losses than lower frequencies.

7.2.        Modelling Approach

7.2. Modelling Approach

127. There are several methods available for modelling the propagation of noise between a source and receiver ranging from very simple models which simply assume spreading effects according to a 10 log (R) or 20 log (R) relationship (as discussed above, and where R is the range from source) to full acoustic models (e.g. ray tracing, normal mode, parabolic equation, wavenumber integration and energy flux models). In addition, semi-empirical models are available, in which complexity and accuracy are somewhere in between these two extremes.
128. In choosing the correct propagation model to employ, it is important to ensure that it is fit for purpose and produces results with a suitable degree of accuracy for the application in question, taking into account the context, as detailed in “Monitoring Guidance for Underwater Noise in European Seas Part III”, National Physical Laboratory Guidance (Dekeling et al., 2014) and in Farcas et al. (2016). Thus, in some situations (e.g. low risk of auditory injury due to underwater noise, where range dependent bathymetry is not an issue, i.e. for non-impulsive noise) a simple (N log R) model might be sufficient, particularly where other uncertainties (such as uncertainties in source level or the impact thresholds) outweigh the uncertainties due to modelling. On the other hand, some situations (e.g. very high source levels, impulsive noise, complex source and propagation path characteristics, highly sensitive receivers, and low uncertainties in assessment criteria) warrant a more complex modelling methodology.
129. The first step in choosing a propagation model is therefore to examine these various factors, such as:

• balancing of errors/uncertainties;
• range dependant bathymetry;
• frequency dependence; and
• source characteristics.

7.3.        Modelling Approach for Vessles and Continous Sources

7.3. Modelling Approach for Vessles and Continous Sources

130. For the noise field model, relevant survey parameters were chosen based on a combination of data provided by the Applicant combined with the information gathered from the publicly available literature. These parameters were fed into an appropriate propagation model routine, in this case the Weston Energy Flux model (for more information refer to Weston, 1971; 1980a; 1980b), suited to the region and the frequencies of interest. The frequency-dependent loss of acoustic energy with distance (TL) values were then evaluated along different transects around the chosen source points. The frequencies of interest in the present study are from 20 Hz to 80 kHz, with different noise sources operating in different frequency bands.
131. The propagation loss is calculated using one of four regions, depending on the distance of the receiver location from the source, and related to the frequency and the seafloor conditions such as depth and its composition.
132. The spherical spreading region exists in the immediate vicinity of the source, which is followed by a region where the propagation follows a cylindrical spread out until the grazing angle is equal to the critical grazing angle. Above the critical grazing angle in the mode stripping region an additional loss factor is introduced which is due to seafloor reflection loss, where higher modes are attenuated faster due to their larger grazing angles. In the final region, the single-mode region, all modes but the lowest have been fully attenuated.

7.4.        Modelling Approach for Impact Piling

7.4. Modelling Approach for Impact Piling

133. In the case of offshore pile installation using an impact hammer, the noise source can be thought of as a “line source” extending through the water column (or in the case of installations using a submersible hammer, a line source through a lower portion of the water column). The hammer strike at the top of the pile produces a compression wave in the pile resulting in radial displacement of the pile walls which is transmitted into the surround media (water and sediments) as noise waves. These compressional waves travel through the pile at circa 5,000 m/s, resulting in a conically shaped wavefront in the water column.
134. Underwater acoustic propagation modelling for this project will be undertaken using a combined distributed line-source array normal mode model for low frequencies (<1 kHz) complimented by a line-source energy flux model for high frequencies (>1 kHz). The line source normal mode model is based on the KrakenC solver (Porter 2001) implemented for a line array over the pile length. The line-source energy flux model is based on an implementation of the energy flux model for a directional source set out in de Jong et al. (2019).
135. The normal-mode method involves solving a depth-dependent equation based on the assumption of a set of modes of vibration which are roughly akin to the modes of a vibrating string (Jensen, 1994). The complete acoustic field is constructed by summing up contributions of each of the modes weighted in accordance with the source depth. The KrakenC solver finds the normal modes in the complex wavenumber plane, which allows it to deal with elastic seabed layers, and to include the effects of leaky modes, making it a good choice for calculation of low frequencies for both close and long range noise fields. The method is, however, slow at higher frequencies and has therefore only been implemented for low frequencies (<1 kHz).
136. The line-source energy flux model (de Jong et al. 2019) used for higher frequencies includes the effect of directionality of the cone shaped wavefront associated with piling noise (circa 17 degrees). This results in damped cylindrical spreading at shorter ranges and mode stripping behaviour at more distant ranges. At even more distant ranges, once the ‘mode stripping’ has eliminated the contribution of all waveguide modes except the lowest mode, propagation is evaluated according to a single mode regime.
137. For estimation of propagation loss of acoustic energy at different distances away from the noise source location (in different directions), the following steps were considered:

• The bathymetry information around this chosen source points will be extracted from the GEBCO database in 72 different transects.
• A geoacoustic model of the different seafloor layers in the survey region will be calculated based on the British Geological Survey (BGS) borehole database and Emodnet sediment database.
• A calibrated line-source propagation model will be employed to estimate the transmission loss matrices for different frequencies of interest (from 25 Hz to 80 kHz) along the 72 different transects.
• Source levels for the line-source array will be determined based on a back-calculation from the received noise level and spectrum shape at 750 m (based on the scaling laws set out in von Pein et al. (2022).
• The calculated source level values will be combined with the transmission loss results to achieve a frequency and range dependant RL of acoustic energy around the chosen source position.
• The TTS and PTS potential impact distances for different marine mammal groups will be calculated using relevant metrics and weighting functions (from (Southall et al. 2019) and by employing a simplistic animal movement model (directly away from the noise source) where appropriate (as agreed with NatureScot on 27 September 2023; refer to section 7.6.1 for more detail). For assessing marine mammal disturbance using single pulse SEL and for assessing effects on fish, no frequency weighting is applied.

138. The approach to pile modelling using a line source array model has been consulted on with NatureScot and agreement was received via email on 05 December 2023 (pers. comms., 2023).
139. The level of detail presented in terms of noise modelling needs to be considered in relation to the level of uncertainty for animal injury and disturbance thresholds. Uncertainty in the noise level predictions will be higher over larger propagation distances (i.e. in relation to disturbance thresholds) and much lower over shorter distances (i.e. in relation to injury thresholds). Nevertheless, it is considered that the uncertainty in animal injury and disturbance thresholds is likely to be higher than uncertainty in noise predictions. This is further compounded by differences in individual animal response, sensitivity, and behaviour. It would therefore be wholly misleading to present any injury or disturbance ranges as a clear line beyond which no effect can occur, and it would be equally misleading to present any noise modelling results in such a way.

7.4.1.    Seiche Line Source Model Calibration

7.4.1. Seiche Line Source Model Calibration

140. The Seiche Ltd line array model has been benchmarked against the COMPILE benchmark workshop for numerical models for pile driving acoustics (Lippert et al., 2016). The COMPILE workshop included modelling results from a number of different organisations in an attempt to compare the performance of acoustic models for piling against pre-defined input parameters. The models included in the benchmarking exercise included those developed by Seoul National University (SNU), Netherlands Organisation for Applied Scientific Research (TNO), Hamburg University of Technology (TUHH), Jasco Applied Sciences, Curtin University and Bundeswehr Technical Centre for Ships and Naval Weapons, Maritime Technology and Research (WTD 71).
141. A comparison between the Seiche model and the benchmark workshop model results is presented in Figure 7.2   Open ▸ . The results of the benchmarking exercise show good correlation with the other models with the results most closely matching the TNO model. The Seiche model predicts slightly higher received levels compared to the other models at 20 km range for this particular benchmark scenario (10 m water depth, sand substrate). Nevertheless, it is considered that the results of the benchmarking exercise demonstrate a good degree of agreement with other noise propagation models for piling.

Figure 7.2: Comparison of Seiche Underwater Acoustic Model Against COMPILE Benchmarks

7.5.        Geo-acoustic and Noise-speed Input Parameters

7.5. Geo-acoustic and Noise-speed Input Parameters

142. Based on BGS core data in the vicinity of the Array, the geo-acoustic model is based on the parameters presented in Table 7.1   Open ▸ .

Table 7.1: Geo-Acoustic Model Used in Propagation Model

143. The sound speed profile has been based on the mean summer temperature and salinity profile for the region, as presented in Figure 7.3   Open ▸ . To produce a representative sound speed profile, conductivity, temperature, and depth (CTD) data was obtained from the NOAA service WODselect for the closest sample point to the development[12] (NOAA 2023).

Figure 7.3: Noise Speed Profile Used in Propagation Model

7.6.        Batch Processing

7.6. Batch Processing

144. To improve the performance and reduce the time taken to process and evaluate multiple TL calculations required for this study, Seiche Ltd.’s proprietary software was employed. This software iteratively evaluates the propagation modelling routine for the specified number of azimuthal bearings radiating from a source point, providing a fan of range-dependent TL curves departing from the noise source for each given frequency and receiver depth. In-house routines are then employed to interpolate the TL values across transects, to give an estimate of the noise field for the whole area around the source point.
145. Once the TL values were evaluated at the source points, in all azimuthal directions, and at all frequencies of interest for various sources, the results were then coupled with the corresponding SL values in third octave frequency bands. The combination of SL with TL data provided us with the third octave band RL at each point in the receiver grid (i.e. at each modelled range, depth, and azimuth of the receiver).
146. The received levels were evaluated for the SPLpk, SPLrms or SEL metric, for each source type, source location, and azimuthal transect to produce the associated TL. The broadband RL were then calculated for these metrics and from the third octave band results. The set of simulated RL transects were circularly interpolated to generate the broadband RL maps centred around each source point. Representations of these RLs are provided in volume 2, chapter 9 in the form of contour maps.
147. For impact piling, the far-field received peak sound pressure level was calculated from SEL values via the empirical fitting between pile driving SEL and peak SPL data.
148. RMS sound pressure levels were calculated assuming a typical T90 pulse duration for impact piling (i.e. the period that contains 90% of the total cumulative noise energy) of 100 ms. It should be noted that in reality, the rms T90 period will increase significantly with distance which means that any ranges based on rms sound pressure levels at ranges of more than a few kilometres are likely to be significant overestimates and should therefore be treated as highly conservative.

7.6.1.    Exposure calculations

7.6.1. Exposure calculations

149. As well as calculating the unweighted noise levels at various distances from different source, it is also necessary to calculate the received acoustic signal in terms of the SEL metric (where necessary and possible) for a marine mammal using the relevant hearing weighting functions. For different operations related noise sources, the numerical SEL value is equal to the SPL rms value integrated over a one second window as the sources are continuous and non-impulsive. These SEL values are employed for calculation of SELcum (cumulative SEL) metric for different marine mammal groups to assess potential impact ranges.
150. Simplified exposure modelling could assume that the animal is either static and at a fixed distance away from the noise source, or that the animal is swimming at a constant speed in a perpendicular direction away from a noise source. For fixed receiver calculations, it has generally been assumed (in literature) that an animal will stay at a known distance from the noise source for a period of 24 hours. As the animal does not move, the noise will be constant over the integration period of 24 hours (assuming the source does not change its operational characteristics over this time). This, however, would give an unrealistic level of exposure, as the animals are highly unlikely to remain stationary when exposed to loud noise, and are therefore expected to swim away from the source. The approximation used in these calculations, therefore, is that the animals move directly away from the source. Nevertheless, in the case of fish exposure, calculations have also been undertaken based on a static receiver assumption.
151. It should be noted that the noise exposure calculations are based on the simplistic assumption that the noise source is active continuously (or intermittently based on source activation timings) over a 24 hour period. The real world situation, however, is more complex. The SEL calculations presented in this study do not take any breaks in activity into account, such as repositioning of the piling vessel, or downtime due to mechanics, logistics or weather.
152. Furthermore, the noise criteria described in the Southall et al. (2019) guidelines assume that the animal does not recover hearing between periods of activity. It is likely that both the intervals between operations could allow some recovery from temporary hearing threshold shifts for animals exposed to the noise (von Benda-Beckmann et al. 2022) and, therefore, the assessment of sound exposure level is conservative.
153. In order to carry out the moving marine mammal calculation, it has been assumed that a mammal will swim away from the noise source at the onset of activities. For impulsive noises of piledriving the calculation considers each pulse to be established separately resulting in a series of discrete SEL values of decreasing magnitude (see Figure 7.4   Open ▸ ).

Figure 7.4: A Comparison of Discrete SEL Per Pulse, and Cumulative SEL Values

154. As an animal swims away from the noise source, the noise it experiences will become progressively lower (more attenuated); the cumulative SEL is derived by logarithmically adding the SEL to which the mammal is exposed as it travels away from the source. This calculation was used to estimate the approximate minimum start distance for an animal in order for it not to be exposed to sufficient noise energy to result in the onset of potential auditory injury. It should be noted that the noise exposure calculations are based on the simplistic assumption that the animal will continue to swim away at a fairly constant relative speed. The real world situation is more complex, and the animal is likely to move in a more complex manner: at varying speed and direction.
155. The assumed swim speeds for animals likely to be present across the Array are set out in Table 7.2   Open ▸ .

Table 7.2: Assessment Swim Speeds of Marine Mammals and Fish that are Likely to Occur Within the North Sea for the Purpose of Exposure Modelling

156. As an additional sensitivity analysis, modelling was also carried out for fish assuming a swim speed of 0 m/s (i.e. stationary).
157. To perform the cumulative exposure calculation, the first step is to parameterise the m-weighted sound exposure levels (or unweighted in the case of fish) for single strikes of a given energy via the 95th percentile line of best fit against the calculated received levels from the model. This function is then used to predict the exposure level for each strike in the planned hammer schedule (periods of slow start, ramp up and full power).
158. In addition to the single source pile driving, simplified situations of simultaneous pile driving from two piling rigs have been considered. The response has been approximated as moving directly away from the point on a line equidistant between the two sources. For simplicity, the sources are considered to be omnidirectional and the piling schedules (soft start, ramp up, etc.) are synchronised, entering each stage of the schedule at the same time.

7.7.        UXO Noise Modelling

7.7. UXO Noise Modelling

7.7.1.    High Order Detonation

7.7.1. High Order Detonation

159. Acoustic modelling for UXO clearance has been undertaken using the methodology described in Soloway and Dahl (2014) and Arons (1954). The equation provides a simple relationship between distance from an explosion and the weight of the charge (or equivalent trinitrotoluene (TNT) weight) but does not take into account bottom topography or sediment characteristics:

where W is the equivalent TNT charge weight and R is the distance from source to receiver.

160. Since the charge is assumed to be freely standing in mid-water, unlike a UXO which would be resting on the seabed and could potentially be buried, degraded or subject to other significant attenuation, this estimation of the source level can be considered conservative.
161. According to Soloway and Dahl (2014), the SEL can be estimated using the following equation:

Figure 7.5: Assumed Explosive Spectrum Shape Used to Estimate Hearing Weighting Corrections to SEL reproduced from Weston (1960)

162. In order to compare against the marine mammal hearing weighted thresholds, it is necessary to apply the frequency dependent weighting functions at each distance from the source. This was accomplished by determining a transfer function between unweighted and weighted SEL values at various distances based on an assumed spectrum shape (refer to Figure 7.5   Open ▸ ) and taking into account molecular absorption at various ranges. Furthermore, because there is potential for more than one UXO clearance event per day (a maximum of two per day is assumed) then it is also necessary to take this into account in the exposure calculation.

7.7.2.    Low Order Techniques

7.7.2. Low Order Techniques

163. According to Robinson et al. (2020), deflagration (a specific method of low order UXO clearance) results in a much lower amplitude of peak sound pressure than high order detonations. The study concluded that peak sound pressure during deflagration is due only to the size of the shaped charge used to initiate deflagration and, consequently, that the acoustic output can be predicted for deflagration if the size of the shaped charge is known.
164. Acoustic modelling for low order techniques (such as deflagration) has therefore been based on the methodology described in section 7.7.1 for high order detonations, using a smaller donor charge size ( Table 6.6   Open ▸ ).

8.             Noise Modelling Results

8. Noise Modelling Results

8.1.        Pre-construction Phase

8.1. Pre-construction Phase

165. The estimated ranges for auditory injury to marine mammals due to various proposed activities undertaken during the pre-construction site investigation surveying phase of the operations are presented in this section. These include geophysical and geotechnical survey activities, UXO clearance and supported vessel activities.
166. The potential ranges presented for injury and behavioural response are not a clearly delineated ‘line’ where an impact will occur on one side and not on the other. Potential impact is more probabilistic; in reality, dose dependency in PTS onset, individual variations, and uncertainties regarding behavioural response and swim speed/direction combine to create a probability field around the source location. Defining a single distance around this area of probability allows visualisation of the spatial extent of different source types and levels and allows comparison of the impacts on a like-for-like basis.

8.1.1.    Geophysical and Geotechnical Surveys

8.1.1. Geophysical and Geotechnical Surveys

167. Geophysical surveying includes many sonar like noise sources and the resulting injury and disturbance ranges for marine mammals are presented in Table 8.1   Open ▸ , based on a comparison to the non-impulsive thresholds set out in Southall et al. (2019). Table 8.2   Open ▸ presents the results for geotechnical investigations. CPT distances are based on a comparison to the Southall et al. (2019) thresholds for impulsive noise (with the distances presented in brackets for peak SPL thresholds) whereas borehole drilling and vibro-core results are compared against the non-impulsive thresholds. Borehole drilling source levels were reported as 142 dB to 145 dB re 1 µPa rms at 1 m, indicating little to no disturbance.
168. The potential impact distances from these operations vary based on their frequencies of operation and source levels and are rounded to the nearest 5 m. It should be noted that, for the sonar like survey sources, many of the injury ranges are limited to circa 75 m as this is the approximate water depth in the area. Sonar like systems have very strong directivity which effectively means that there is only potential for injury when a marine mammal is directly underneath the noise source. Once the animal moves outside of the main beam, there is significantly reduced potential for injury. The same is true in many cases for TTS where an animal is only exposed to enough energy to cause TTS when inside the direct beam of the sonar like source. For this reason, many of the TTS and PTS ranges are similar (i.e. limited by the depth of the water). Disturbance thresholds are as shown in Table 5.3   Open ▸ for impulsive and non-impulsive sources respectively, noting that impulsive sources have both a strong and a mild disturbance threshold.

Table 8.1: Potential Impact Ranges (m) for Marine Mammals During the Various Geophysical Investigation Activities Based on Comparison to Southall et al. (2019) SEL Thresholds

Table 8.2: Potential Impact Ranges for Geotechnical Site Investigation Activities Based on Comparison to Southall et al. (2019) SEL Thresholds (Comparison to Ranges for Peak SPL Where Threshold was Exceeded Shown in Brackets)

8.1.2.    Vessels

8.1.2. Vessels

169. The potential impact ranges for vessels are included in section 8.4, which summarises the vessel modelling results for all phases of the Array.

8.1.3.    UXO Clearance

8.1.3. UXO Clearance

170. The predicted injury ranges for low order disposal are presented in Table 8.3   Open ▸ and for high order detonation of UXOs in Table 8.4   Open ▸ . All UXO injury and disturbance ranges are based on a comparison to the relevant impulsive noise thresholds as set out in section 5.3.2.
171. It should be noted that, due to a combination of dispersion (i.e. where the waveform elongates), multiple reflections from the sea surface and seabed and molecular absorption of high frequency energy, the noise is unlikely to still be impulsive in character once it has propagated more than a few kilometres. Consequently, great caution should be used when interpreting any results with predicted injury ranges in the order of tens of kilometres. Furthermore, the modelling assumes that the UXO acts like a charge suspended in open water whereas in reality it is likely to be partially buried in the sediment. In addition, it is possible that the explosive material will have deteriorated over time meaning that the predicted noise levels are likely to be over-estimated. In combination, these factors mean that the results should be treated as precautionary potential impact ranges which are likely to be significantly lower than predicted.

Table 8.3: Potential Impact Ranges for Low Order and Low Yield UXO Clearance Activities

Table 8.4: Potential Impact Ranges for High Order Clearance of UXOs

8.2.        Construction Phase

8.2. Construction Phase

8.2.1.    Impact Piling

8.2.1. Impact Piling

172. The impact piling scenarios modelled were as follows:

• single piling – wind turbine foundation piles (3,000 kJ);
• single piling – OSP jacket piles (4,400 kJ);
• two concurrent piling event – wind turbine foundation piles (3,000 kJ); and
• two concurrent piling events – wind turbine foundation pile (3,000 kJ) and OSP jacket pile (4,400 kJ).

173. All cases are presented both with and without the use of 30 minutes of ADD prior to installation.
174. Impact ranges were modelled for all three locations, as shown in Figure 8.1   Open ▸ , however only the most adverse case injury ranges have been reported in this section. The 90th percentile of the transmission loss was used for the calculations, which is considered to be a worst case while taking into account outliers due to extreme changes in bathymetry and sea conditions, for example.

Figure 8.1: Locations Modelled Within the Ossian Array (Red Line)[13]

175. All impact piling injury ranges are based on a comparison to the relevant impulsive noise thresholds as set out in section 5. Disturbance effects are presented in volume 2, chapter 10 using the dose-response approach described in section 5.3.
176. The injury ranges for peak sound pressure are based on the noise from the maximum hammer energy over the entire installation.
177. It should be noted that peak sound pressure is a time domain parameter and does not necessarily add together to produce higher received peak sound pressure levels. Even if two piling hammers were to strike their piles synchronously (i.e. to the exact millisecond) the noise waves will arrive at different locations at different times. Consequently, the peak pressure ranges for simultaneous piling do not differ from the peak injury ranges identified for single piling spreads.
178. During impact piling the interaction with the seabed and the water column is complex. In these cases, a combination of dispersion (i.e. where the waveform shape elongates), and multiple reflections from the sea surface and bottom and molecular absorption of high frequency energy, the noise will lose its impulsive shape after some distance (generally in order of several kilometres).
179. An  article by Southall (2021) discusses this aspect in detail, and notes that “…when onset criteria levels were applied to relatively high-intensity impulsive sources (e.g. pile driving), TTS onset was predicted in some instances at ranges of tens of kilometres from the sources. In reality, acoustic propagation over such ranges transforms impulsive characteristics in time and frequency (see Hastie et al. 2019; Amaral et al. 2020; Martin et al., 2020). Changes to received signals include less rapid signal onset, longer total duration, reduced crest factor, reduced kurtosis, and narrower bandwidth (reduced high-frequency content). A better means of accounting for these changes can avoid overly precautionary conclusions, although how to do so is proving vexing”. The point is reinforced later in the discussion which points out that “…it should be recognised that the use of impulsive exposure criteria for receivers at greater ranges (tens of kilometres) is almost certainly an overly precautionary interpretation of existing criteria”.
180. Consequently, great caution should be used when interpreting any results with predicted injury ranges in the order of tens of kilometres.

8.2.2.    Single Piling

8.2.2. Single Piling

181. Distances are presented at which noise levels decrease to below PTS/TTS threshold values in terms of cumulative SEL and peak sound pressure level. It should be noted that the potential PTS/TTS ranges reduce significantly with the use of ADD because it is assumed that an animal swims away from the area for 30 minutes before being exposed to noise from piling, therefore significantly reducing its cumulative SEL for any given start range.
182. Distances are presented in Table 8.5   Open ▸ to Table 8.9   Open ▸ for wind turbine foundation pile installation at 3,000 kJ and Table 8.10   Open ▸ to Table 8.14   Open ▸ for OSP jacket pile installation at 4,400 kJ.