1. Introduction

  1. Berwick Bank Wind Farm Limited (BBWFL), a wholly owned subsidiary of SSE Renewables Limited (hereafter be referred to as ‘the Applicant’), is proposing the development of the Berwick Bank Wind Farm (hereafter referred to as the ‘Proposed Development’), an offshore wind farm off the east coast of Scotland. The Proposed Development array area is located in the outer Firth of Forth and Forth of Tay, approximately 37.8 km east of the Scottish Borders coastline (St Abb’s Head) and 47.6 km from the East Lothian coastline. The Proposed Development array area will be connected to a SP Energy Networks (SPEN) substation at Branxton via a Proposed Development export cable corridor.
  2. An Environmental Impact Assessment was (EIA) was carried out to determine the potential effects of the Proposed Development on sensitive marine mammal receptors from a range of different impacts. A key impact assessed was the potential for elevations in subsea noise during piling activities to lead to injury and behavioural disturbance to individuals. Subsea noise modelling was conducted to predict the potential spatial scale of the effect. In particular, for behavioural disturbance, the assessment predicted that the elevations in subsea noise leading to disturbance could extend over a considerable area and potentially affect a large number of individuals of the key species identified within the marine mammal study area.
  3. Population modelling was therefore carried out to determine the potential for a short to medium term effects (piling could occur over a total duration of 372 days intermittently within a 52 month piling period during the eight year offshore construction timeframe) to result in long term population level effects on any species. The interim Population Consequences of Disturbance (iPCoD) model[1] (developed by Sea Mammal Research Unit (SMRU) Consulting, collaborating with a team of researchers at the University of St Andrews), was adopted to simulate the potential changes in the population over time and is explained within this report.

1.2. iPCoD

  1. The iPCoD model simulates the changes in a population over time, for both a disturbed and an undisturbed population. This provides a comparison of the type of changes that could occur resulting from natural environmental variation, demographic stochasticity (i.e. variability in population growth rates) and disturbance (Harwood et al., 2014; King et al., 2015).
  2. The iPCoD model is based on expert elicitation, 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 (Donovan et al., 2016). In the case of the iPCoD model, the marine mammal experts were asked for their opinion on how changes in hearing resulting from Permanent Threshold Shift (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). Following the initial development of the iPCoD model a study was undertaken to update the transfer functions on the effects of PTS and disturbance on the probability of survival and giving birth to a viable young for harbour porpoise, harbour seal and grey seal (again via expert elicitation) (Booth and Heinis, 2018; Booth et al., 2019). The iPCoD model has been updated in light of additional work undertaken since it was originally launched.
  3. A potential limitation of the iPCoD model is that no form of density dependence has been incorporated due to the uncertainties as to how this may occur. As discussed in Harwood et al. (2014), the concept of density-dependence is fundamental to understanding how animal populations respond to a reduction in their size. In population biology, density-dependant factors, such as resource availability or competition for space, can limit population growth. If the population declines, these factors no longer become limiting and therefore, for the remaining individuals in a population, there is likely to be an increase in survival rate and reproduction. This then allows the population to expand back to previous levels at which density-dependant factors become limiting again (i.e. population remains at carrying capacity). The limitations for assuming a simple linear ratio between the maximum net productivity level and carrying capacity have been highlighted by Taylor and Master (1993) as simple models demonstrate that density dependence is likely to involve several biological parameters which themselves have biological limits (e.g. fecundity and survival). For UK populations of harbour porpoise (and other marine mammal species) however, there is no published evidence for density dependence and therefore, density dependence assumptions are not currently included within the iPCoD protocol.

2. Methodology

2.1. Piling Parameters

2.1.1.    Maximum Design Scenario

  1. The maximum design scenario for piling at the Proposed Development assumes that 5.5 m diameter piled jacket foundations will be installed using a maximum hammer energy of 4,000 kJ. This represents the absolute maximum energy likely to be required at any point across foundation installation. Taken as an average, the maximum hammer energy is likely to be no greater than 3,000 kJ. For the purposes of population modelling, the assessment focussed only on the absolute maximum of 4,000 kJ as this represented the maximum adverse design (as agreed by consultees at Marine Mammal Road Map Meeting 3, 18 January 2022).
  2. Piling will be required at up to 179 wind turbine foundations and ten offshore substation platform (OSP)/Offshore convertor station platform foundations. The maximum design scenario was based on concurrent piling at wind turbine foundations with the largest separation between piling locations. Although piling could occur concurrently at a wind turbine and OSP/Offshore convertor station platform foundation these locations would be closer together compared to two wind turbine foundations. Therefore, piling at OSPs/Offshore convertor station platforms was considered as a single piling event and modelled as a separate operation within iPCoD but not coincident with concurrent piling at the wind turbine foundations (since this would represent three concurrent piling events which is not proposed as part of the Proposed Development design). Using the maximum number of hours of piling per pile, the number of piles likely to be installed within 24 hours and the number of concurrent installation vessels, it was possible to estimate the maximum number of days (24 hours) within which piling could occur on the basis of two piling operations:
  • 287 piling days (concurrent vessel) for the 179 wind turbines; and
  • 85 piling days (single vessel) for the ten OSPs/Offshore convertor station platforms.
    1. It is estimated that piling activity at the Proposed Development will take place in three campaigns and an indicative piling construction schedule is provided in Table 2.1   Open ▸ . Piling could potentially take place at any point within the foundation installation phases; however, for the purposes of developing the piling programme for iPCoD (a requirement of the model) an indicative programme has been developed based on a realistic installation approach. Therefore, within each campaign, a realistic scenario has been assumed where there are nine months of piling followed by 12 months where jackets are installed over the piles.  

2.1.2.    Noise Modelling (Conversion Factors)

  1. Subsea noise modelling was undertaken to predict the potential spatial scale of the effect of subsea noise. Potential injury, in the form of a PTS was determined using published and peer reviewed thresholds developed by Southall et al. (2019) for the dual metrics un-weighted peak Sound Pressure Levels (SPLpk) and marine mammal hearing-weighted cumulative Sound Exposure Level (SELcum). For behaviour disturbance a dose-response approach was undertaken using the metric single strike Sound Exposure Level (SELss) with contours modelled in 5 dB increments based on Graham et al. (2017). A full description of subsea noise modelling is provided in volume 3, appendix 10.1 and summarised in section 10.11.1 of volume 2, chapter 10. Further to discussion via the marine mammal Road Map process, the subsea noise modelling investigated the sensitivity of using different conversion factors to determine the amount of energy converted into received sound. In this respect, three conversion factors were modelled: 10% reducing to 1% as piling progresses, 4% reducing to 0.5% as piling progresses and a constant conversion factor of 1% throughout piling.
  2. A detailed study of existing literature was undertaken by Seiche Ltd, including exploration of published data from pile driving at other wind farms. Subsequently, the subsea noise modelling report recommended that the 4% reducing to 0.5% conversion factor was an appropriate conservative approach. This was evaluated alongside the 1% conversion factor in the full marine mammal assessment of effects.
  3. Whilst 10% reducing to 1% was not included in the marine mammal assessment of effects (as it was determined to be overly conservative and therefore an inaccurate representation of potential impact), for completeness the results of all conversion factor scenarios have been analysed and the estimated numbers of animals potentially affected for all scenarios are presented in an appendix to the marine mammal chapter (volume 3, appendix 10.5).
  4. For the purposes of population modelling, all three conversion factors were included to provide a comparison. For reasons described above, only the results of the 4% reducing to 0.5% conversion factor and 1% constant conversion factor have been taken forward to present in the marine mammal chapter (volume 3, appendix 10.5). The iPCoD modelling results are presented in order of the largest potential quantitative effect to the smallest:
  • 10% reducing to 1% conversion factor;
  • 1% constant conversion factor throughout the piling period; and
  • 4% reducing to 0.5% conversion factor.
    1. Note that in terms of behavioural effects, the 1% constant conversion factor was found to result in a higher SEL at any point over the piling sequence compared to the 4% reducing to 0.5% conversion factor and therefore led to a larger potential effect area (see Figure 10.4 in volume 2, chapter 10 and further explanation in volume 4, appendix 10.5).
Table 2.1:
Indicative Piling Construction Programme

Table 2.1: Indicative Piling Construction Programme

2.2. Key Species

  1. Key species to be included in the population modelling were discussed as part of the marine mammal Road Map consultation process and stakeholders requested inclusion of the following species:
  • harbour porpoise Phocoena phocoena;
  • bottlenose dolphin Tursiops truncatus;
  • minke whale Balaenoptera acutorostrata;
  • grey seal Halichoerus grypus; and
  • harbour seal Phoca vitulina.
    1. The first version of the iPCoD model was considered to be suitable for all species above with the exception of harbour seal since data on trends in the abundance of harbour seal were limited when iPCoD was initially developed. Subsequent count data has, however, allowed a better understanding of the demographics of this population. In the Firth of Tay and Eden Estuary Special Area of Conservation (SAC), the population declined between 2002 and 2017 by 18.6%, however, data from the 2016 counts suggested that the SAC represents only 15% of the East of Scotland Seal Management Area (SMA). When including counts from the wider Firth of Forth, the total East of Scotland SMA population appears to be more stable in recent years. Despite the potential limitations of the model for harbour seal, the consultees requested (25 February 2022) that this species was included in the population modelling due to concerns over the historic decline of harbour seal on the east coast of Scotland.

2.3. Model Inputs

  1. The iPCoD model v5.2[2] was set up using the program R v4.1.2 (2021) with RStudio as the user interface. To enable the iPCoD model to be run, the following data were provided:
  • demographic parameters for the key species;
  • user specified input parameters:

           vulnerable subpopulations; and

           residual days of disturbance.

  • number of animals predicted to experience PTS and/or disturbance during piling; and
  • estimated piling schedule during the proposed construction programme.

2.3.2.    Demographic Parameters

  1. Demographic parameters for the key species assessed in the population model are presented in Table 2.2   Open ▸ .

 

Table 2.2:
Demographic Parameters Recommended for Each Species for the Relevant Management Unit (MU)/SMAs (Sinclair et al., 2019)

Table 2.2: Demographic Parameters Recommended for Each Species for the Relevant Management Unit (MU)/SMAs (Sinclair et al., 2019)

 

2.3.3.    Reference Populations

  1. MU populations and vulnerable sub-populations were specified in the model as reference populations against which the effects (i.e. number of animals suffering PTS/disturbed) were assessed. The MUs and vulnerable subpopulations were agreed with stakeholders as part of the Road Map process (25 February 2022). Vulnerable subpopulations were requested for harbour porpoise and minke whale only. The results of the assessment using vulnerable subpopulations should, however, be interpreted with caution as the relevant area used to delineate the subpopulation (SCANS-III block R) are survey units rather than representing a biologically meaningful area. Table 2.3   Open ▸ provides the reference populations used in the iPCoD.

 

Table 2.3:
Reference Populations Used in the iPCoD

Table 2.3: Reference Populations Used in the iPCoD

1 The offshore EIA Report considers the reference population as East Scotland plus Northeast England MU, however, further to discussions with NatureScot and Marine Scotland Licencing Operations Team during the Marine Mammal Road Map consultation it was requested that the iPCoD model was run against the East Scotland population only.

 

2.3.4.    Residual Days Disturbance

  1. Empirical evidence from constructed wind farms (e.g. Graham et al., 2019; Brandt et al., 2011) suggests that the detection of animals returns to baseline levels in the hours following a disturbance from piling and therefore, for the most part, it can be assumed that the disturbance occurs only on the day (24 hours) that piling takes place. Due to the potential duration of piling occurring at the Proposed Development (up to 10 hours for installation of a single wind turbine jacket pile and up to five piles installed per 24 hours using two vessels), piling could occur for most of the 24 hour period. Therefore, the number of residual days of disturbance has, conservatively, been selected as one meaning that the model assumes that disturbance occurs on the day of piling and persists for a period of 24 hours after piling has ceased.

2.3.5.    Number of Animals (PTS/Disturbance)

  1. The number of animals predicted to experience PTS and/or disturbance was based on the density values provided as part of the baseline assessment (volume 3, appendix 10.2). For each species studied, the density values – including a mean and a maximum - were provided and these were used to quantify the number of animals affected, based on the modelled noise contours. For the purposes of this population modelling, the maximum density values were adopted to provide a conservative assessment ( Table 2.4   Open ▸ ).

 

Table 2.4:
Maximum Density Values Applied to the Calculation of Number of Animals Potentially Affected and Taken Forward for the iPCoD Model

Table 2.4: Maximum Density Values Applied to the Calculation of Number of Animals Potentially Affected and Taken Forward for the iPCoD Model

 

  1. The number of animals predicted to be injured or disturbed were calculated using these maximum densities and were estimated from the piling locations that gave rise to the largest potential impact ranges. Therefore, the highest numbers of animals potentially affected at any one time are assessed.
  2. For all scenarios, mitigation will be applied (see volume 2, chapter 10), including:
  • pre-start monitoring using Marine Mammal Observers (MMOs) and Passive Acoustic Monitoring (PAM);
  • Acoustic Deterrent Device (ADD) for a period of 30 minutes prior to commencement of piling;
  • low energy hammer initiation;
  • soft start for a period of 30 minutes; and
  • a gradual ramp up to full hammer.
    1. With these measures in place, the residual number of individuals potentially affected by PTS was zero for all species. The exception to this was for the scenario of 10% reducing to 1% conversion factor where for minke whale, a residual estimate of one individual could potentially experience PTS during piling.
    2. The total number of individuals affected by piling at any one time are provided in the Table 2.5   Open ▸ and represent the number disturbed (with exception of minke whale for the 10% reducing to 1% conversion factor scenario). Where residual PTS is predicted after application of mitigation measures, as outlined in paragraph 23, this is show in parenthesis.

 

Table 2.5:
Estimated Number of Animals Predicted to be Disturbed at any one Time During Piling Using Different Conversion Factors

Table 2.5: Estimated Number of Animals Predicted to be Disturbed at any one Time During Piling Using Different Conversion Factors

 

Piling schedule

  1. The piling schedule was developed from the project design envelope which provided an estimate of the number of days piling for the wind turbine and OSP/Offshore convertor station platform foundations within a defined piling phase, which is scheduled to take place within an overall offshore piling construction window of March 2026 to October 2028 ( Table 2.1   Open ▸ ).
  2. A total of 287 days (24-hour periods) on which piling could occur (based on the maximum design scenario) was estimated for concurrent piling at the wind turbines. A total of 85 days of piling (24-hour periods) on which piling could occur was estimated for single piling at the OSPs/Offshore convertor station platforms. The number of piling days was allocated evenly across months ( Table 2.6   Open ▸ ). The scenario of number of consecutive days piling followed by non-piling days was considered to be typical of a piling construction programme which would allow for weather downtime, breakdowns and/or return of vessel to port.
  3. The first two time points in the model were selected to coincide with key periods of the piling schedule. Subsequent time points were selected up to year 25 as follows:
  • time point 4: end of first two piling campaigns which run sequentially between 2026 and 2027;
  • time point 8: end of third piling campaign which ends December 2031;
  • time point 13: 13 years after the start of the offshore construction phase;
  • time point 19: 19 years after the start of the offshore construction phase; and
  • time point 25: 25 years after the start of the offshore construction phase.

 

Table 2.6:
Piling Schedule Assessed within the iPCoD Model

Table 2.6: Piling Schedule Assessed within the iPCoD Model

 

2.4. Cumulative Projects

  1. Population modelling was run for cumulative scenarios based on the scheduling of offshore construction for projects within the relevant study areas for each species. For harbour porpoise and minke whale the cumulative assessment considered the MU reference populations only and not the vulnerable subpopulations (defined within SCANS block R) as cumulative projects fell outside this SCANS block and therefore this subpopulation was not relevant with respect to the cumulative assessment. Details of piling schedules were unknown as offshore wind farm assessments typically only provide indicative offshore construction times ( Table 2.7   Open ▸ ). The maximum design scenario for each project was based on the maximum adverse consented or proposed design for each project.

 

Table 2.7:
Indicative Offshore Construction Schedules for Each of the Cumulative Projects

Table 2.7: Indicative Offshore Construction Schedules for Each of the Cumulative Projects

P = Indicative piling campaign at the Proposed Development

 

  1. The iPCoD model was set up as described above in terms of the demographic parameters (section 2.3.2), reference populations (section 2.3.3) and with the same days of residual disturbance specified (section 2.3.4). The number of animals affected for each of the key species and number of days on which piling occurred was taken from the maximum design scenario for each of the projects and has been referenced in the following sections. As piling schedules were unknown, the piling days were spread evenly throughout the offshore construction phases shown in Table 2.7   Open ▸ .
  2. Time points in the model were selected to coincide with the following periods:
  • time point 2: start of 2023, piling commences at four projects;
  • time point 3: start of 2024, piling continues with a total of eight projects potentially piling;
  • time point 4: start of 2025, piling continues with a total of seven projects potentially piling;
  • time point 5: start of 2026, piling continues at six projects plus start of offshore construction phase at the Proposed Development (just prior to start of piling at the Proposed Development);
  • time point 7: start of year 2028, piling continues at cumulative projects and is completed after the first two piling campaigns at the Proposed Development;
  • time point 11: start of year 2032, piling continues at cumulative projects and is completed after the third piling campaign at the Proposed Development;
  • time point 19: start of year 2040, 8 years after completion of piling at all projects; and
  • time point 25: start of year 2046, 14 years after completion of piling at all projects.
    1. For the cumulative projects only the 1% conversion factor was modelled for the Proposed Development as this represented the maximum spatial effect range compared to the 4% reducing conversion factor. A conversion factor of 10% reducing was not used as this was deemed to be unrepresentative (see section 2.1.2).

2.4.2.    Harbour Porpoise

  1. Cumulative projects for harbour porpoise were considered across the regional marine mammal study area which encompassed the northern North Sea. A summary of the number of harbour porpoise affected and number of piling days for each of cumulative projects is provided below ( Table 2.8   Open ▸ ).

 

Table 2.8:
Summary of Cumulative Projects Included in iPCoD for Harbour Porpoise

Table 2.8: Summary of Cumulative Projects Included in iPCoD for Harbour Porpoise

 

2.4.3.    Bottlenose Dolphin

  1. Cumulative projects for bottlenose dolphin were considered across the north-east of Scotland which encompassed the region between the northern part of Moray Firth to the southern part of the Firth of Forth. A summary of the number of bottlenose dolphin affected and number of piling days for each of cumulative projects is provided below ( Table 2.9   Open ▸ ).

 

Table 2.9:
Summary of Cumulative Projects Included in iPCoD for Bottlenose Dolphin

Table 2.9: Summary of Cumulative Projects Included in iPCoD for Bottlenose Dolphin

2.4.4.    Minke Whale

  1. Cumulative projects for minke whale were considered across the regional marine mammal study area which encompassed the northern North Sea. A summary of the number of minke affected and number of piling days for each of cumulative projects is provided below ( Table 2.10   Open ▸ ).

 

Table 2.10:
Summary of Cumulative Projects Included in iPCoD for Minke Whale

Table 2.10: Summary of Cumulative Projects Included in iPCoD for Minke Whale

1 The number of days of piling is based on the 2020 Seagreen 1A Piling Strategy (Seagreen Wind Energy Ltd, 2020) as the number of piled foundations has been reduced since the original EIA (Seagreen Wind Energy Ltd, 2012).

2 The number of minke whale potentially disturbed at any one time is based on impacts of piling at Seagreen Bravo presented in the original EIA (Seagreen Wind Energy Ltd, 2012), as it represents the worst-case number when compared with numbers presented in later documents.

 

2.4.5.    Grey Seal

  1. Cumulative projects for grey seal were considered across the north-east of Scotland which encompassed the region between the northern part of Moray Firth to the southern part of the Firth of Forth. A summary of the number of grey seals affected and number of piling days for each of cumulative projects is provided below ( Table 2.11   Open ▸ ).
Table 2.11:
Summary of Cumulative Projects Included in iPCoD for Grey Seal

Table 2.11: Summary of Cumulative Projects Included in iPCoD for Grey Seal

1 The number of days of piling is based on the 2020 Seagreen 1A Piling Strategy (Seagreen Wind Energy Ltd, 2020) as the number of piled foundations has been reduced since the original EIA (Seagreen Wind Energy Ltd, 2012).

2 The number of grey seal potentially disturbed at any one time is based on impacts of piling at Seagreen Bravo presented in the original EIA (Seagreen Wind Energy Ltd, 2012), as it represents the worst-case number when compared with numbers presented in later documents.

 

2.4.6.    Harbour Seal

  1. Cumulative projects for harbour seal were considered across the north-east of Scotland which encompassed the region between the northern part of Moray Firth to the southern part of the Firth of Forth. A summary of the number of harbour seal affected and number of piling days for each of cumulative projects is provided below ( Table 2.12   Open ▸ ).

 

Table 2.12:
Summary of Cumulative Projects Included in iPCoD for Harbour Seal

Table 2.12 Summary of Cumulative Projects Included in iPCoD for Harbour Seal

1 The number of days of piling is based on the 2020 Seagreen 1A Piling Strategy (Seagreen Wind Energy Ltd, 2020) as the number of piled foundations has been reduced since the original EIA (Seagreen Wind Energy Ltd, 2012).

2 The number of harbour seal potentially disturbed at any one time is based on impacts of piling at Seagreen Bravo presented in the original EIA (Seagreen Wind Energy Ltd, 2012), as it represents the worst-case number when compared with numbers presented in later documents.

 

2.5. Summary of Scenarios Modelled in iPCoD

  1. Table 2.13   Open ▸ presents a summary of the scenarios modelled through iPCoD for each species for the Proposed Development alone and for cumulative projects.

 

Table 2.13:
Summary of Scenarios Modelled for Each Species in iPCoD for the Proposed Development

Table 2.13: Summary of Scenarios Modelled for Each Species in iPCoD for the Proposed Development