Relevant plans and projects

815             The plans and projects set out in Table 13.5   Open ▸ have been considered within the assessment of other projects and plans with potential for in-combination effects.

Table 13.5:
List of Other Developments with Potential for In-Combination Effects on Annex II Marine Mammal Features

Table 13.5: List of Other Developments with Potential for In-Combination Effects on Annex II Marine Mammal Features

 

13.3 Construction and Decommissioning

13.3.1  Underwater Noise

816             Increases in underwater noise associated with the construction (and decommissioning) of the Proposed Development have the potential to cause injury and disturbance to marine mammals. The assessment of impacts associated with underwater noise has been informed by subsea noise modelling, the scope of which was agreed through the Road Map process (see volume 3, appendix 10.2 of the Offshore EIA Report).

817             This section addresses the underwater noise effects associated with the construction and decommissioning phases of the Proposed Development. For each potential underwater noise effect, the nature of the effect is described, the source activities generating the effect and the potential changes to marine mammal receptors are outlined. Effects are categorised as permanent or temporary.

818             The subsequent sub-sections provide more information on each of these potential underwater noise effects and the sensitivity of the Annex II marine mammal features to these effects:

  • a summary of the relevant components of the Proposed Development, outlined in the maximum design scenario ( Table 13.10   Open ▸ ) and designed-in measures ( Table 13.11   Open ▸ );
  • an overview of the methodology/modelling/assessment undertaken to quantify and assess underwater noise effects on marine mammals (paragraph 819 et seq.); and
  • an overview of relevant marine mammal information gathered to aid the assessment (paragraph 776 et seq.).

Assessment methodology

819             Marine mammals, particularly cetaceans, are capable of generating and detecting sound (Au et al., 1974; Bailey et al., 2010) and are dependent on sound for many aspects of their lives (i.e. prey identification; predator avoidance; communication and navigation). Increases in anthropogenic noise may consequently lead to a potential effect within the marine environment (Parsons et al., 2008; Bailey et al., 2010). Richardson et al. (1995) described four zones of noise influence which vary with the distance from the source, including: audibility (sound is detected); masking (interfere with detection of sounds and communication); responsiveness (behavioural or physiological response) and injury/hearing loss (tissue damage in the ear).

820             For this study, it is the zones of injury (auditory) and disturbance (i.e. responsiveness) that are of concern (there is insufficient scientific evidence to properly evaluate masking).

821             The following sub-sections (paragraph 822 et seq.) provide context for the effects of auditory injury and behavioural disturbance in the Annex II marine mammal species concerned and summarise the relevant thresholds for onset of effects and describe the evidence base used to derive them. Subsequent sections (paragraph 834 et seq.) outline the approach taken to the modelling and quantification of underwater noise effects on Annex II marine mammal species during construction and decommissioning.

Auditory injury in marine mammals

822             Auditory injury in marine mammals can occur as either a permanent threshold shift (PTS), where there is no hearing recovery in the animal, or as a temporary threshold shift (TTS), where an animal can recover from the tissue damage. The ‘onset’ of TTS is deemed to be where there is a temporary elevation in the hearing threshold by 6 dB and is “the minimum threshold shift clearly larger than any day to day or session to session variation in a subject’s normal hearing ability”, and which “is typically the minimum amount of threshold shift that can be differentiated in most experimental conditions” (Southall et al., 2007). Since it is considered unethical to conduct experiments measuring PTS in animals, the onset of PTS was extrapolated from early experiments on TTS growth rates in chinchillas (Henderson and Hamernick, 1986) and is conservatively considered to occur where there is 40 dB of TTS (Southall et al., 2007). Whether such shifts in hearing would lead to loss of fitness will depend on several factors including the frequency range of the shift and the duty cycle of impulsive sounds. For example, if a shift occurs within a frequency band that lies outside of the main hearing sensitivity of the receiving animal, there may be a ‘notch’ in this band but potentially no effect on the animal’s ability to survive.

823             For the purposes of the assessment of potential injury, the emphasis is on PTS as the appropriate threshold due to the irreversible nature of the effect whereas TTS is temporary and reversible. A likely response of an animal exposed to noise levels that could induce TTS is to flee the ensonified area. It is therefore considered that there is also a behavioural response (disturbance) that overlaps with potential TTS ranges, and animals exposed to noise levels that have the potential to induce TTS are likely to actively avoid hearing damage by moving away from the ensonified area. Since derived thresholds for the onset of TTS are based on the smallest measurable shift in hearing, TTS thresholds are likely to be very precautionary and could result in overestimates of potential range of effect. In addition, the assumptions and limitations of subsea noise modelling (e.g. equal energy rule, reduced sound levels near the surface, conservative swim speeds, and use of impulsive sound thresholds at large ranges) also lead to a potential overestimation of range of effect. Notably, Hastie et al. (2019) found that during pile driving there were range dependant changes in signal characteristics with received sound losing its impulsive characteristics at ranges of several kilometres, especially beyond 10 km. For these reasons TTS is not considered a useful predictor of the potential impacts of underwater noise on marine mammals where ranges exceed more than c. 10 km and therefore, where this is the case (i.e. piling and UXO clearance) TTS is not included in the assessment in terms of injury. To supporting this reasoning a synthesis of the use of impulsive sound thresholds at large ranges is presented volume 3, appendix 10.1 of the Offshore EIA Report. Ranges for TTS were, however, modelled for completeness for all noise-related impacts and are presented in volume 3, appendix 10.1 of the Offshore EIA Report.

824             For marine mammals, injury thresholds are based on both linear (i.e. un-weighted) peak sound pressure levels (SPLpk) and marine mammal hearing-weighted cumulative sound exposure level (SELcum). The SELcum takes account of the cumulative sound received by an animal within the ensonified area over the entire piling sequence and is weighted by marine mammal hearing groups based on similarities in known or expected hearing capabilities (Southall et al., 2007). Marine mammal hearing groups are described in the latest guidance (Southall et al., 2019) as follows:

  • low frequency (LF) cetaceans (i.e. marine mammal species such as baleen whales with an estimated functional hearing range between 7 Hz and 35 kHz); minke whale is the marine mammal IEF in the LF cetacean group;
  • high-frequency (HF) cetaceans (i.e. marine mammal species such as dolphins, toothed whales, beaked whales and bottlenose whales with an estimated functional hearing range between 150 Hz and 160 kHz). Bottlenose dolphin is the marine mammal IEFs in the HF cetacean group;
  • very high-frequency (VHF) cetaceans (i.e. marine mammal species such as true porpoises, with an estimated functional hearing range between 275 Hz and 160 kHz). Harbour porpoise is the marine mammal IEF in the HF cetacean group; and
  • pinnipeds in water (PW) (i.e. true seals with an estimated functional hearing range between 50 Hz and 86 kHz). Grey seal and harbour seal are the marine mammal IEFs in the PW group.

825             Injury criteria are proposed in Southall et al. (2019) for both impulsive and non-impulsive (continuous) sound and are summarised in Table 13.6   Open ▸ and Table 13.7   Open ▸ .

 

Table 13.6:
Summary of PTS Criteria for Impulsive and Non-Impulsive Noise (SEL Thresholds in dB re 1 μPa2s and peak SPL thresholds in dB re 1 μPa)

Table 13.6: Summary of PTS Criteria for Impulsive and Non-Impulsive Noise (SEL Thresholds in dB re 1 μPa2s and peak SPL thresholds in dB re 1 μPa)

 

Table 13.7:
Summary of TTS Criteria for Impulsive and Non-impulsive Noise (SEL Thresholds in dB re 1 μPa2s and peak SPL Thresholds in dB re 1 μPa)

Table 13.7: Summary of TTS Criteria for Impulsive and Non-impulsive Noise (SEL Thresholds in dB re 1 μPa2s and peak SPL Thresholds in dB re 1 μPa)

 

826             To carry out exposure calculations (SELcum metric) the underwater noise modelling made a simplistic assumption that an animal would be exposed over a 24-hour period and that there would be no breaks in activity during this time. It was assumed that an animal would swim away from the noise source at the onset of activity at a constant rate and subsequently conservative species specific swim speeds were incorporated into the model (see Table 13.8   Open ▸ ) following agreement with statutory nature conservation bodies (swim speeds presented during Road Map Meeting 2 with no queries raised - see volume 3, appendix 10.3 of the Offshore EIA Report).

 

Table 13.8:
Swim Speeds Assumed for Exposure Modelling (SELcum) for Marine Mammal IEFs

Table 13.8: Swim Speeds Assumed for Exposure Modelling (SELcum) for Marine Mammal IEFs

 

Disturbance in marine mammals

827             Beyond the zone of injury, noise levels are such that they no longer result in physical injury but can result in disturbance to marine mammal behaviour. A marine mammal’s response to disturbance will depend on the individual and the context; previous experience and acclimatisation will affect whether an individual exhibits an aversive response to noise, particularly in a historically noisy area. Typically, a threshold approach has been adopted in offshore wind farm assessments in the UK to quantify the scale of the potential effect. For example, the United States (US) National Marine Fisheries Service (NMFS, 2005) define strong disturbance in all marine mammals as “Level B harassment” and for impulsive noise suggests a threshold of 160 dB re 1 μPa (root mean square (rms)). This threshold meets the criteria defined by JNCC (2010a) as a ‘non-trivial’ (i.e. significant) disturbance and is equivalent to the Southall et al., (2007) severity score of five or more on the behavioural response scale. Beyond this threshold, the behavioural responses are likely to become less severe (e.g. minor changes in speed, direction and/or dive profile, modification of vocal behaviour and minor changes in respiratory rate (Southall et al., 2007)). The NMFS guidelines suggest a precautionary level of 140 dB re 1 μPa (rms) to indicate the onset of low-level marine mammal disturbance effects for all mammal groups for impulsive sound (NMFS, 2005), although this is not considered likely to lead to a ‘significant’ disturbance response.

828             More recently, to illustrate the variation in behavioural responses of marine mammals, Graham et al. (2017) used empirical evidence collected during piling at the BOWL (Moray Firth, Scotland) to demonstrate that the probability of occurrence of harbour porpoise (measured as porpoise positive minutes) increased exponentially moving further away from the source. The study showed a 100% probability of disturbance at an (un-weighted) 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. The dose response thresholds tie in with the NMFS (2005) criteria since a mild behavioural response is suggested to occur at a threshold of 140 dB re 1 μPa (rms) which is equivalent of 130 dB 1 μPa2s where a small response (c. 10% of animals) would occur according to the dose response. Dose response is an accepted approach to understanding the behavioural effects from piling and has been applied at other UK offshore wind farms (for example Seagreen Alpha/Bravo and Hornsea Project Three).

829             For the assessment of potential impacts of piling noise, subsea noise modelling was undertaken using the dose-response approach with SELss contours modelled in 5 dB increments. For all other noise impacts, the simple threshold approach using the NMFS criteria (NMFS, 2005) was adopted. Disturbance criteria are presented in Table 13.9   Open ▸ .

 

Table 13.9:
Disturbance Criteria for Marine Mammals Used in Assessment

Table 13.9: Disturbance Criteria for Marine Mammals Used in Assessment

 

830             In applying these criteria it is possible to provide quantification of the magnitude of effects with respect to the spatial extent of disturbance and subsequently the number of animals potentially affected. There is, however, a note of caution associated with this approach. Southall et al. (2021) highlights that the challenges for developing a comprehensive set of empirically derived criteria for such a diverse group of animals are significant. Extensive data gaps have been identified (e.g. measurements of the effects of elevated noise on baleen whales) which mean that extrapolation from other species has been necessary. Sounds that disturb one species may, however, be irrelevant or inaudible to other species since there are broad differences in hearing across the frequency spectrum for different marine mammal hearing groups. Variance in responses even within a species are well documented to be context and sound-type specific (Ellison et al., 2012; Southall et al., 2017). In addition, the potential interacting and cumulative effects of multiple stressors (e.g. reduction in prey, noise and disturbance, contamination) is likely to influence the severity of responses (Lacy et al., 2017).

831             For these reasons, neither a threshold approach nor a dose-response function was provided in the original guidance (Southall et al., 2007) and subsequently the recent recommendations by Southall et al. (2021) also steer away from a single overarching approach. Instead, Southall et al. (2021) proposes a framework for developing probabilistic response functions for future studies. The paper suggests different contexts for characterising marine mammal responses for both free ranging and captive animals with distinctions made by sound sources (i.e. active sonar, seismic surveys, continuous/industrial noise and pile driving). Three parallel categories have been proposed within which a severity score from an acute (discrete) exposure can be allocated:

  • survival – defence, resting, social interactions and navigation;
  • reproduction – mating and parenting behaviours; and
  • foraging – search, pursuit, capture and consumption.

832             Marine mammals considered in this assessment vary biologically and therefore have different ecological requirements that may affect their sensitivity to disturbance. To illustrate this point we can compare the differences between the two seal species identified as key biological receptors in the baseline. Grey seals are capital breeders (foraging to build up stored fat reserves for lactation) and often make long foraging trips from haul-outs. In contrast, harbour seals are income breeders (feeding throughout the pupping season) and making shorter foraging trips from haul-outs.

833             In summary, Southall et al (2021) clearly highlights the caveats associated with simple, one-size-fits-all threshold approaches that could lead to errors in disturbance assessments. Although approach presented in paragraph 827 et seq. is based on the best scientific evidence currently available, the quantification of effects should the interpreted with caution.

Summary underwater noise modelling

Piling

834             Pile driving during the construction phase of the Proposed Development has the potential to result in elevated levels of underwater noise that are detectable by marine mammals above background levels and could result in injury or behavioural effects on Annex II marine mammal species. A detailed underwater noise modelling assessment has been carried out to investigate the potential for injury and behavioural effects on marine mammals as a result of piling (impulsive sounds), using the latest assessment criteria (see volume 3, appendix 10.1 of the Offshore EIA Report).

835             With respect to the SPLpk metric, the soft start initiation (see Table 13.11   Open ▸ ) is the most relevant noise source and period, as this is the range at which animals may potentially experience injury from the initial strike of the hammer, after which point it is assumed that they will move away from the noise source. Secondly, injury ranges were predicted for marine mammals exposed to impulsive noise from multiple hammer strikes over a prolonged period (i.e. using the SELcum metric); the assumption being that a marine mammal exposed to lower noise levels over a prolonged period (as it moves away from the source) could experience auditory injury. The maximum injury ranges for each species have been provided with reference to the largest impact range from the dual criteria approach, and a proposed marine mammal mitigation zone has been determined on the basis of the largest range across all species.

836             Taking a precautionary approach, in line with SNCBs advice as discussed during Road Map Meetings (volume 3, appendix 10.3 of the Offshore EIA Report) and via Scoping Opinion (Marine Scotland, 2022), the subsea noise assessment considered a range of different conversion factors (the amount of hammer energy converted into received sound by marine mammal receptors): 1% constant, 4% reducing to 0.5% and 10% reducing to 1%.

837             A detailed study was undertaken reviewing noise modelling methodologies across different UK offshore wind farms and investigating energy conversion factors for determining sound source levels during piling. Published literature on energy conversion factors were explored together with available noise measurements taken during offshore wind farm construction and the results presented as an evidence-based, peer-reviewed report (volume 3, appendix 10.1, of the Offshore EIA Report). The study recommended that the most representative and precautionary conversion factor was 4% reducing to 0.5% as piling progresses. However, a sensitivity assessment was also undertaken to compare the results of noise modelling for the three different conversion factors requested by consultees (volume 3, appendix 10.1, annex B of the Offshore EIA Report). Subsequently, considering the evidence-base and the results of the sensitivity assessment, a precautionary approach was adopted for the marine mammal assessment of effects whereby both a conversion factor of 4% reducing to 0.5% and the 1% constant throughout the piling period has been taken forward to the quantitative assessment for marine mammals. As requested by consultees, a third conversion factor of 10% reducing to 1% was also quantified with respect to effects on marine mammal receptors, although not taken forward to the assessment of effects as it was determined to be overly conservative and therefore not realistic. Volume 3, appendix 10.5 of the Offshore EIA Report presents a comparison of the numbers of animals affected for all three conversion factors scenarios.

838             The scenarios modelled were based on the absolute maximum hammer energy (4,000 kJ) and a realistic maximum hammer energy (3,000 kJ). The assessment has been carried out at two locations on opposite sides of the Proposed Development array area, chosen to represent extremes of location. The bathymetry of the site is relatively flat, therefore the two locations were selected to represent the points closest and furthest away from the shoreline. These are represented by the indicative wind turbine foundation locations wind turbine 40 and wind turbine 135 (used in the assessment of underwater noise impacts for all species, except bottlenose dolphin, as these represent the largest area of impact) or wind turbine 1 and wind turbine 179 (used in the assessment of underwater noise impacts for bottlenose dolphin due to proximity to the areas of high coastal density, see volume 2, chapter 10 of the Offshore EIA Report).

839             For piling at wind turbines it is assumed that two vessels would pile concurrently, and two scenarios were modelled in this respect:

  • separation distance of 1.78 km (minimum distance between foundations) would result in the greatest potential for injury since animals could be exposed to sound from both rigs at relatively high levels; and
  • separation distance of c. 50 km (maximum separation distance between vessels) would result in the maximum area of disturbance since the overlap between disturbed areas would be smaller compared to vessels piling close together.

840             Using the equation below (see volume 3, appendix 10.1 of the Offshore EIA Report), a broadband source level value was evaluated for the noise emitted during impact pile driving operation in each operation window.

SEL =

841             In this equation, β is the energy transmitted from the pile into the water column, E is the hammer energy employed in joules, C0 is the speed of sound in the water column, and ρ is the density of the water. From the SEL result calculated using the equation above, source-level spectra can also be calculated for different third octave frequency bands.

842             Following a noise modelling workshop to test sensitivities of different scenarios, the piling campaign was developed with a low hammer energy and slow initiation phase in order to provide designed-in measures to reduce the potential risk of injury to marine mammal receptors. Four scenarios were investigated in the subsea noise modelling assessment and are summarised as follows:

  • wind turbine foundations (piled jacket) maximum design scenario – up to 179 piled jacket foundations, with up to four legs per foundation and up to 2 x 5.5 m diameter piles per leg (1,432 piles) using an absolute maximum hammer energy of 4,000 kJ for the longest possible duration (up to ten hours);
  • wind turbine foundations (piled jacket) realistic design scenario – up to 179 piled jacket foundations, with up to four legs per foundation and up to 2 x 5.5 m diameter piles per leg (1,432 piles) using a realistic average maximum hammer energy of 3,000 kJ for a realistic maximum duration (up to nine hours);
  • OSP/Offshore convertor station platform foundations (jacket) maximum design scenario – using a maximum hammer energy of 4,000 kJ for a duration of up to eight hours; and
  • OSP/Offshore convertor station platform foundations (jacket) realistic design scenario – using a maximum hammer energy of 3,000 kJ for a duration of up to seven hours.

843             The marine mammal assessment was based on the maximum design scenario with piling at a maximum energy of 4,000kJ for both wind turbine foundations and OSP/Offshore convertor station platform foundations. However, since piling is unlikely to reach and maintain the absolute maximum hammer energy of 4,000 kJ at all locations, results for a realistic design scenario were also provided for context using an average maximum hammer energy of 3,000 kJ for both foundations. There will be a maximum of two piling events at any one time and subsea noise modelling assumed concurrent piling at two wind turbine foundations as a maximum design scenario. This was due to the distances between wind turbines (i.e. maximum spatial separation) as well as the longer duration of piling at wind turbine foundations compared to OSP/Offshore convertor station platform foundations. Installation does not, however, preclude concurrent piling at a wind turbine foundation and OSP/Offshore convertor station platform foundation but this scenario is captured in the maximum design scenario case for concurrent piling at two wind turbine foundations. Results presented here are therefore for concurrent piling at two wind turbine foundations and single piling at wind turbine or OSP/maximum design scenario foundations.

844             A number of conservative assumptions were adopted in the subsea noise model that resulted in a precautionary assessment (volume 3, appendix 10.1 of the Offshore EIA Report). These are summarised here:

  • the subsea noise modelling assumed that the maximum hammer energy would be reached and maintained for 195 minutes at all locations, whereas this is unlikely to be the case based on examples from other offshore wind farms For example, at the BOWL the mean actual hammer energy averages were considerably lower than the maximum adverse scenario assessed in the EIA Report and only six out of 86 asset locations reached maximum hammer energy (Beatrice, 2018);
  • the soft start procedure simulated does not allow for short pauses in piling (e.g. for realignment) and therefore the modelled SELcum is likely to be an overestimate since, in reality, these pauses will reduce the noise exposure that animals experience whilst fleeing;
  • the modelling assessment assumed that animals swim directly away from the noise source at constant and conservative average speeds based on published values (see volume 3, appendix 10.1 of the Offshore EIA Report). This is likely to lead to overestimates of the potential range of effect where animals exceed these speeds. For example, Otani et al. (2000) note that horizontal speed for harbour porpoise can be significantly faster than vertical speed and cite a maximum speed of 4.3 m/s. Similarly, Leatherwood et al. (1988) reported harbour porpoise swim speeds of approximately 6.2 m/s.
  • the use of the SELcum metric is described as an equal energy rule where exposures of equal energy are assumed to produce the same noise-induced threshold shift regardless of how the energy is distributed over time. This means that for intermittent noise, such as piling, the equal-energy rule overestimates the effects since the quiet periods between noise exposures will allow some recovery of hearing compared to continuous noise;
  • the model overestimates the noise exposure an animal receives since it does not account for any time that marine mammals spend at the surface and the reduced sound levels near the surface; and
  • due to a combination of factors (e.g. dispersion of the waveform, multiple reflections from sea surface and seafloor, and molecular absorption of high frequency energy), impulsive sounds are likely to transition into non-impulsive sounds at distance from the sound source with empirical evidence suggesting such shifts in impulsivity could occur markedly within 10 km from the sound source (Hastie et al., 2019) (see volume 3, appendix 10.1 of the Offshore EIA Report). Since the precise range at which this transition occurs is unknown, noise models still adopt the impulsive thresholds at all ranges which is likely to lead to an overestimate of effect ranges at larger distances (tens of kilometres) from the sound source.

845             A final scenario was modelled to include the use of an Acoustic Deterrent Device (ADD) activated for a period of 30 minutes prior to initiation of piling to illustrate the potential efficacy of using this as a secondary mitigation (for more details see volume 2, chapter 10). The injury scenarios with and without use of ADDs were suggested by NatureScot in their 2020 Berwick Bank Scoping Advice on 07 October 2020. Therefore, additional noise modelling was undertaken to determine whether the potential for injury to marine mammals would be reduced through the application of ADDs.

Dose response

846             Empirical evidence from monitoring at offshore wind farms during construction suggests that pile driving is unlikely to lead to 100% avoidance of all individuals exposed, and that there will be a proportional decrease in avoidance at greater distances from the pile driving source (Brandt et al., 2011). This was demonstrated at Horns Rev Offshore Wind Farm, where 100% avoidance occurred in harbour porpoises at up to 4.8 km from the piles, whilst at greater distances (10 km plus) the proportion of animals displaced reduced to < 50% (Brandt et al., 2011). Similarly, Graham et al. (2019) used empirical evidence collected during piling at the BOWL (Moray Firth, Scotland) to demonstrate that the probability of occurrence of harbour porpoise (measured as porpoise positive minutes) increased exponentially moving further away from the noise source. Importantly, Graham et al. (2019) demonstrated that the response of harbour porpoise to piling diminished over the piling phase such that, for a given received noise level or at a given distance from the source, there were more detections of animals at the last piling location compared to the first piling location.

847             Similarly, a telemetry study undertaken by Russell et al. (2016) investigating the behaviour of tagged harbour seals during pile driving at the Lincs Offshore 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 non-piling versus piling periods to show change in usage. The study found that there was a significant decrease during piling at predicted received SEL levels of between 142 dB and 151 dB re 1µPa2s.

848             A dose response curve was applied to this assessment to determine the number of animals that may potentially display a behavioural response to received noise levels during piling. Unweighted sound exposure level single strike (SELss) contours were plotted in 5 dB isopleths in decreasing increments from 180 dB to 120 dB re.1µPa2s using the highest modelled received noise level for 4% reducing to 0.5% conversion factor and 1% constant conversion factor.

849             To adopt the most precautionary approach, the dose response contours were plotted in Geographical Information System (GIS) for all modelled locations and the location selected for assessment was the one whereby the contours covered the greatest spatial area, thereby representing the maximum adverse scenario. The areas within each 5 dB isopleth were calculated from the spatial GIS map and a proportional expected response, derived from the dose response curve for each isopleth area, was used to calculate the number of animals potentially disturbed. These numbers were subsequently summed across all isopleths to estimate the total number of animals disturbed during piling. The number of animals predicted to respond was based on species specific densities as agreed with statutory consultees (volume 2, chapter 10 of the Offshore EIA Report).

850             For harbour porpoise the dose-response curve was applied from the first location modelled as shown by Graham et al. (2017) where the probability of response approaches zero at c. 120 dB SELss. In the absence of species-specific data for other cetacean species the same dose response curve was assumed to apply to all cetaceans in this assessment (see volume 2, chapter 10 of the Offshore EIA Report).

851             For harbour seal and grey seal the most appropriate dose response curve was derived from the Russell et al., (2017) study and has been previously applied to other Offshore Wind Farm assessments in the UK (e.g. Hornsea Project Three (GoBe, 2018a) and Seagreen Alpha/Bravo optimised design (Seagreen Wind Energy, 2018)). In this case the highest received level at which a response was detected was at 135 dB SELss with a zero probability of response measured at 130 dB SELss (see volume 2, chapter 10 of the Offshore EIA Report).

Conversion Factors

852             At the request of MS-LOT, a range of conversion factors - 1% constant, 4% reducing to 0.5% and 10% reducing to 1% - have been modelled with respect to how much of the hammer energy is converted into received sound. Based on a comprehensive peer-reviewed study, it was recommended that 4% reducing to 0.5% is most representative of a precautionary estimate of the conversion factor for the type of hammer to be used at the Proposed Development. A summary of the reasoning behind this conclusion is provided below with full detail given in the Subsea Noise Technical Report (see volume 3, appendix 10.1, Annex A of the Offshore EIA Report).

853             The study on conversion factors (volume 3, appendix 10.1, annex A of the Offshore EIA Report) found that theoretical values for representative conversion factors were likely to reach an upper limit of 1.5% for an above water hammer throughout a piling sequence with a conversion factor of 1% being typical throughout the majority of the piling (as estimated from in field measurements (e.g. Dahl and Reinhall, 2013)). The 1% constant conversion factor is therefore representative of this theoretical average and use of a constant conversion factor is typical of the approach adopted by previous UK offshore wind farm subsea noise assessments.

854             There are, however, likely to be differences in conversion factors depending on the type of hammer used. The use of a submersible hammer, as opposed to an above water hammer, can result in a conversion factor that varies with pile penetration depth. Since the piling at the Proposed Development is likely to involve a partially submersible hammer, the literature review explored the conversion factors that may be applicable in this situation. A key study cited in the review was by Lippert et al., (2017) where both modelled and measured data were used to estimate a conversion factor of between 2% and 0.5% for a partially submersible hammer. In this study the modelled and measured data were strongly correlated suggesting that the estimated conversion factors were very representative. Nevertheless, it was recognised that for the Lippert et al. (2017) study a significant proportion of the pile was above water at the start of the piling sequence which could have reduced the apparent conversion factor compared to a situation where the pile starts just above the water line. Assuming that the energy radiated into the water is approximately proportional to the length of pile which is exposed to the water then the conversion factor at the start of piling from the Lippert study can be estimated to be approximately 3.5% (see volume 3, appendix 10.1, annex A of the Offshore EIA Report). Thus, the 4% conversion factor requested by SNCBs is considered to be close to, but more precautionary, than the empirically derived value based on the Lippert et al., (2017) study.

855             The study on conversion factors (volume 3, appendix 10.1, annex A of the offshore EIA Report) found that a conversion factor of 10% was likely to be over precautionary and therefore more likely to lead to an overestimate of effect ranges, particularly considering the transition from impulsive to continuous noise over distance from the source. The 10% reducing conversion factor was based upon a study conducted at the BOWL for a fully submersible hammer which suggested that higher conversion factors were found for longer exposed lengths of pile towards the start of the piling and reduced to 1% as the pile penetrated further into the seabed (Thompson et al., 2020). However, there were large discrepancies between the noise modelling and real-world propagation particularly at further distances from the pile. By reanalysing the data from BOWL it was determined that at closer distances, the modelled and measured levels were closer in value and suggested a conversion factor closer to 5% rather than the 10% cited in the study.

856             Acknowledging that the conversion factor of 10% reducing to 1% as unrealistic and likely to be over precautionary, the sensitivity assessment found that for the peak pressure metric (SPLpk) the maximum injury ranges for all species were derived using the 1% conversion factor as opposed to the 4% reducing to 0.5% conversion factor. This is because the higher conversion rate for the 4% reducing to 0.5% conversion factor occurs when the hammer is at its lowest energy at the start of the piling sequence, so the highest estimated SPLpk levels are later in the piling sequence once the conversion factor has reduced. In contrast, with a constant 1% conversion factor throughout the piling sequence, the SPLpk ranges increase throughout the piling sequence with increasing hammer energy.

857             As previously, discounting the conversion factor of 10% reducing to 1% as over-precautionary for the cumulative exposure metric (SELcum), the maximum injury ranges for all species were derived using the 4% reducing to 0.5% conversion factor. Since the noise modelling for injury adopts a dual metric approach using both SPLpk and SELcum the most precautionary approach was to assess the greatest injury range using either metric and considering both the 1% throughout the piling period, and 4% reducing to 0.5% conversion factor. The maximum injury ranges were predicted using the 1% conversion factor throughout the piling period and were based on the SPLpk metric. The number of animals affected were subsequently estimated on this basis and differs by species hearing group. This was to ensure that, for mitigation purposes, the most precautionary approach was adopted.

858             In terms of behavioural effects, the 1% constant conversion factor was found to result in the highest SELss at any point over the piling sequence compared to the 4% reducing to 0.5% conversion factor and therefore resulted in the largest potential effect area (see volume 2, chapter 10 of the Offshore EIA Report). The reason for this is that the maximum SEL for the 1% constant scenario is at the end of the piling sequence, which is when the hammer energy is maximum (i.e. up to 4,000 kJ) because for a constant conversion factor of 1% the SEL will increase with increasing hammer energy (see volume 2, chapter 10 of the Offshore EIA Report). This is not the case for the 4% reducing to 0.5% scenario as in this instance, the highest SEL occurs during initiation as the 4% conversion factor at this point leads to a higher SELss than at any other point during the piling sequence (see volume 2, chapter 10 of the Offshore EIA Report). The SELss is an unweighted metric and therefore there is no difference in modelled contours by marine mammal hearing group.

859             Although not considered as part of the assessment of effects for the reasons described in paragraph 855, for completeness the dose-response contours were also plotted for the 10% reducing to 1% conversion factor to allow estimates of the numbers of animals potentially disturbed by this scenario. The results are presented in volume 3, appendix 10.5 of the Offshore EIA Report.

Summary of iPCoD modelling

860             There is limited understanding of how behavioural disturbance and auditory injury affect survival and reproduction in individual marine mammals and consequently how this translates into effects at the population level. The iPCoD model was developed using a process of expert elicitation to determine how physiological and behavioural changes affect individual vital rates (i.e. the components of individual fitness that affect the probability of survival, production of offspring, growth rate and offspring survival).

861             Expert elicitation is a widely accepted process in conservation science whereby the opinions of many experts are combined when there is an urgent need for decisions to be made but a lack of empirical data with which to inform them. In the case of iPCoD, the marine mammal experts were asked for their opinion on how changes in hearing resulting from PTS and behavioural disturbance (equivalent to a score of 5* or higher on the ‘behavioural severity scale’ described by Southall et al. (2007)) associated with offshore renewable energy developments affect calf and juvenile survival and the probability of giving birth (Harwood et al., 2014). Experts were asked to estimate values for two parameters which determine the shape of the relationships between the number of days of disturbance experienced by an individual and its vital rates, thus providing parameter values for functions that form part of the iPCoD model (Harwood et al., 2014).

862             The iPCoD model simulates the median population difference over time for a disturbed and an undisturbed population to provide comparison of the type of changes that could occur resulting from natural environmental variation, demographic stochasticity[10] and man-induced disturbance. The results are summarised in relation to the forecasted population size over time with forecasts made at certain timepoints (e.g. two, seven, 13, 19 and 25 years) after piling commences. In addition, the model calculates the ratio of the unimpacted to the impacted population size at these timepoints. A caveat of this model, however, is that the model does not account for density dependence and therefore the forecasts may be unrealistic as they assume that vital rates in the population will not alter as a result of density dependent factors (e.g. competition).

863             Whilst there are many limitations to this process, iPCoD was requested by statutory consultees as part of Road Map Meeting process as it represents the best available approach for the species considered in this assessment (volume 3, appendix 10.3 of the Offshore EIA Report). In addition, any uncertainties have been offset as far as possible by adopting a precautionary approach at all stages of the assessment from the maximum design parameters in the project envelope, conservatism in the subsea noise model and adoption of precautionary estimates to represent the densities of key species. Thus, the result from the iPCoD is considered to be inherently cautious and should be interpreted as such.

864             Given that Annex II marine mammals constituting the population of the designated site are a part of the wider populations, population modelling using iPCoD was carried out for MU populations from which individuals can be linked to respective SACs:

  • Harbour porpoise (Annex II species of the Southern North Sea SAC) and North Sea MU;
  • Bottlenose dolphin (Annex II species of the Moray Firth SAC) and Coastal East Scotland MU;
  • Harbour seal (Annex II species of the Firth of Tay and Eden Estuary SAC) and East Scotland MU; and
  • Grey seal (Annex II species of the Isle of May SAC and the Berwickshire and North Northumberland SAC) and East Scotland and North East Scotland MU.
Site investigation surveys

865             Several sonar-based survey types will potentially be used for the geophysical surveys, including multibeam echosounder (MBES), Side Scan Sonar (SSS), Single beam echo sounders (SBES) and sub-bottom sonar (SBS). 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, acts like a beam and is emitted in pulses. Sonar-based sources are considered as continuous (non-impulsive) because they generally compromise a single (or multiple discrete) frequency as opposed to a broadband signal with high kurtosis, high peak pressures and rapid rise times. Unlike the sonar-based surveys, the ultra high resolution seismic (UHRS) is likely to utilise a sparker, which produces an impulsive, broadband source signal.

866             Source levels for borehole drilling ahead of standard penetration testing are in a range of 142 dB to 145 dB re 1 µPa re 1 m (rms). SEL measurements conducted during core penetration tests (CPTs) showed that it is characterised by broadband sound with levels measured generally 20 dB above the acoustic ocean noise floor (see volume 3, appendix 10.1 of the Offshore EIA Report). For the purpose of assessment of effects, these sources are considered as impulsive sounds. Measurements of a vibro-core test (Reiser et al., 2011) show underwater source sound pressure levels of approximately 187 dB re 1 µPa re 1 m (rms). The vibro-core sound is considered to be continuous (non-impulsive).

867             Full description of the source noise levels for geophysical and geotechnical survey activities is provided in volume 3, appendix 10.1 of the Offshore EIA Report.

UXO clearance

868             Although the clearance of UXO prior to commencement of construction using low order techniques is the preferred option, there is a small risk UXO clearance could result in high order detonation. High order detonation has the potential to generate some of the highest peak sound pressures of all anthropogenic underwater sound sources (von Benda-Beckman et al., 2015), and are considered a high energy, impulsive sound source. The potential impacts of this activity will depend on noise source characteristics, the receptor species, distance from the sound source and noise attenuation within the environment.

869             Subsea noise modelling for UXO clearance (both low order and high order detonation) has been undertaken using the methodology described in Soloway and Dahl (2014), which provides a simple relationship between distance from an explosion and the weight of the charge (or equivalent trinitrotoluene (TNT) weight). 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. Marine mammal hearing weighted thresholds were compared by application of the frequency dependent weighting functions at each distance from the source. Based on findings presented in Robinson et al. (2020), noise modelling for low order techniques followed the same methodology as for high order detonation, with a smaller donor charge size. Full details of underwater noise modelling undertaken for UXO clearance is provided in volume 3, appendix 10.1 of the Offshore EIA Report.

870             Potential effects of underwater noise from high order UXO clearance on marine mammals include mortality, physical injury or auditory injury. The duration of effect for each UXO detonation is less than one second. Behavioural effects are therefore considered to be negligible in this context. TTS is presented as a temporary auditory injury but also represents a threshold for the onset of a fleeing response. Proposed Development specific underwater noise modelling was carried out using published and peer-reviewed criteria to determine the potential magnitude (range) of effect on marine mammal receptors. A project specific Marine Mammal Mitigation Plan (MMMP) will be developed in order to reduce the potential to experience injury and has been provided as a high-level draft at Application (volume 4, appendix 23) ( Table 13.11   Open ▸ ).

871             It is anticipated that up to 70 UXOs are likely to be found within the Proposed Development array area and Proposed Development export cable corridor, however, only 14 of these will require clearance. The maximum design scenario is based on experience of UXO clearance at Seagreen Offshore Wind Farm (in close proximity to the Proposed Development). For Seagreen, of the 20 UXOs estimated to be present for the purposes of the marine mammal risk assessment (Seagreen Wind Energy, 2021), only four (20%) were found to require clearance within the proposed development site, one of which was relocated rather than cleared by high order techniques (SSE pers. Comm.). The estimate of 70 UXOs for Berwick Bank Offshore Wind Farm was extrapolated from the same study carried out for Seagreen (Ordtek, 2017; Ordtek, 2019) and therefore it is considered likely that the number of UXOs requiring disposal will be significantly less than assessed here (i.e. based on the same proportion cleared for Seagreen, there may only be 14 UXOs requiring clearance for the Proposed Development). The precise details and locations of potential UXOs is unknown at the time of Application. During the UXO clearance campaign at Seagreen Offshore Wind Farm the maximum UXO size identified was 250 kg NEQ. Given that Seagreen Offshore Wind Farm is located approximately 4 km from the Proposed Development array area, a similar maximum size of munition is expected to be encountered in the same region. Therefore, for the purposes of this assessment, it has been assumed that the maximum design scenario is UXO size up to 300 kg (see Table 13.10   Open ▸ ). The maximum frequency is up to two detonations within 24 hours. The clearance activities will be tide and weather dependant as detonations will take place during daylight hours and slack water only. The aim is to enable clearance of at least one UXO per tide, during daylight hours only.

872             Low order techniques will be applied as the intended methodology for clearance of UXO. The technique uses a single charge of up to 80 g Net Explosive Quantity (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. Recent controlled experiments showed low-order clearance using deflagration to result in a substantial reduction in acoustic output over traditional high-order methods, with SPLpk and SELcum being typically significantly lower for the low order techniques 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 clearance 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. It is also possible that there will be residual explosive material remaining on the seabed following low order clearance. In this case, recovery will be performed, including the potential need of a small (500 g NEQ) ‘clearing shot’.

873             There is a small risk that a low order clearance could result in high order detonation of UXO. In addition, some UXOs may be deemed to be too unstable to warrant a low order approach and therefore for safety reasons would need to be cleared using high order methods. At Neart na Gaoithe Offshore Wind Farm in the Firth of Forth, a total of 53 items of UXO required detonation and four of the 37 (c. 10%) monitored UXO clearance events resulted in a high order detonation, largely as a result of the age, condition and type of munition (Seagreen Wind Energy, 2021). Therefore the assessment of potential impacts due to the underwater noise during UXO clearance and subsequent secondary mitigation is based on the maximum design scenario of high order detonation and is presented in paragraph 961 et seq. Additional, secondary mitigation measures will be agreed with the statutory conservation bodies post-Application discussions, and included as a part of the MMMP to ensure that the potential to experience injury by marine mammals is reduced for all clearance activities.