Abstract
As part of the Offshore Renewables Joint Industry Programme (ORJIP) Offshore Wind Programme, GoBe Consultants Limited (GoBe) has undertaken environmental research and review in relation to a study to help inform the current understanding of the impact of piling during the construction of offshore wind farms (OWFs) upon herring (Clupea harengus) spawning.
OWF construction activities, such as pile driving, generate noise (both sound pressure and particle motion), which can disturb or injure fish. Due to potential effects, conditions incorporated into UK offshore renewables consents (such as Marine Licences issued by the Marine Management Organisation (MMO) or Marine Scotland Licensing Operations Team (MS-LOT)) may restrict construction activities during times when fish are considered to be most vulnerable to disturbance, such as during spawning and migration. Such restrictions can be temporal, spatial or technical in nature and are designed to minimise risk of disturbance or injury occurring to fish as well as to the population viability.
Inclusion of these conditions within consents ensures that development of OWFs can proceed, with controls over the way in which the development is undertaken to ensure impacts are limited in order to protect habitats, species and other key receptors. It is important to ensure that the established controls are appropriately formulated to deliver a protection benefit without unnecessarily burdening development (e.g. in terms of increased costs or time constraints). To achieve this there needs to be a sufficient level of understanding of the impact source - receptor pathway, the sensitivity of the receptor to the impact and evaluation of potential risk. There is also a need for consistency in the approach used to provide advice to the industry. Despite extensive academic literature and survey work, and their use in Environmental Impact Assessments (EIAs), uncertainty remains on both the accuracy of the fish spawning information currently available and the level / significance of impacts from piling activity on fish species.
Through the application of literature reviews, data gathering, data analysis and consultations, GoBe has completed a study looking at these aspects, with the following aims:
- Review and consolidate available data and information to define (where possible) UK populations, key spawning areas and key spawning periods for herring;
- Identify any gaps in our understanding of herring populations, spawning areas and periods;
- Define and gain acceptance with consenting authorities and experts on herring populations, spawning areas and periods, where data is available; and
- Discuss and understand how the current mitigation approaches are agreed by consenting authorities with the aim of ensuring mitigation is fit for purpose, for example in understanding what is required to ensure protection of the fish stocks, whilst also ensuring that restrictions are minimised and proportionate.
Population Information and OWFs
A review of literature and herring data sources has illustrated that there is a good understanding of the main spawning areas for the different herring stocks that are present within UK waters and that a broad-based identification of the key areas is possible. It is also possible to identify more specific spawning areas for most of these stocks through the combined use of historical fish sensitivity maps, more recent International Council for the Exploration of the Seas (ICES) International Herring Larvae Survey (IHLS) and International Bottom Trawl Surveys (IBTS) data and predicted seabed habitat types, along with other specific seabed habitat data. Through review of this combined information, the main spawning areas can be discerned. A notable outcome of this study is that through assessment of recent IHLS data (covering the last ten years) and analysis on a year to year basis as well as cumulatively, the comparison of these data with historical fish sensitivity maps produced by Coull et al. (1998) has enabled some refinement of the historical data. This may provide more confidence during EIA for both developers and regulators in assessing potential impact significance and mitigation options.
This study suggests that there is an ability to identify areas within the historical mapping (Coull et al., 1998) where spawning activity is focused within more defined spawning grounds. Even from larval data with drift that hasn’t been back-calculated to a specific location, it is possible to demonstrate that there are specific locations (spawning grounds) where spawning activity is focused within these historical spawning areas. This is illustrated by the ‘hot spots’ in the heat mapping.
The specific, discrete pockets ofspawning beds that herring use are, however, not so easily identified as they can change from year to year. This is due to the specific habitat and environmental conditions that herring require to enable successful spawning to take place. The literature review confirms that without undertaking intrusive grab sampling surveys during spawning periods, or by using drop down video surveying, it is difficult to locate or identify exact spawning bed locations (only identifiable through the presence of mats of eggs). As such, there are gaps in the current knowledge of where specific spawning beds are located for all of the stocks.
The use of the IHLS data to identify spawning grounds has been questioned due to larvae freely drifting away from spawning beds on the prevailing currents and so the confidence that these larval surveys are identifying actual spawning areas is reduced. However, the use of larvae to determine spawning areas has been accepted since 1957 (Parrish et al., 1957) and most recently the use of larvae was reviewed within the aggregate dredging industry for the BMAPA by MarineSpace (2013) where it was concluded that of all methods that can be used to determine spawning grounds, the IHLS data was the best indicator and is a direct measure of spawning where fish of length 0 – 11mm were caught. This study concluded that the methodology of using IHLS data to produce heat maps of spawning areas is widely accepted. Indeed, using this method to support further discussions on reducing herring restrictions for OWFs has recently been accepted in England by consenting authorities (as identified through this review).
Being able to take drift rates and apply back-calculation to larvae to identify spawning beds would provide an ability to be more definitive on the precise location of spawning beds. There are limitations with this method however. Attempts to undertake back-calculations of where larvae may have drifted from based upon published prevailing water current information, larvae age and growth rates have been undertaken (e.g. Beatrice Offshore Wind Farm Limited, 2016; Brown and May Marine, 2007) but these reports show how broad the resulting conclusions are, with very large areas being cited as potential spawning beds, with no exact location identification possible due to using estimates of current speeds and directions. From a review of literature, there is only one instance where the exact location of a spawning bed has been identified (Ballantray Bay in the Clyde) and this was via grab sampling, where herring eggs were incidentally located on the seabed (Parrish et al., 1959). It has been recognised since 1959 that due to changing environmental conditions and spawning beds being discrete pockets of suitable habitat (which can change from year to year) within a spawning ground or area, it is extremely difficult to identify the exact location of spawning beds unless they are recorded through grab sampling or by drop down video surveys.
There is a need to be proportionate in terms of risks posed by OWF developments that do not appear to affect the main spawning grounds within the Coull et al., 1998 mapping area, as refined by the heat mapping ‘hot spots’ versusthe OWF developmentsthat clearly do affect the ‘hotspots’. Given that the noise modelling undertaken for EIA is precautionary and that the fish sensitivity maps are also precautionary, then balanced consideration needs to be given to the level of potential effects that may result from OWF developments located on the fringes of spawning areas illustrated on the fish sensitivity maps and what the benefit of restricting piling at such developments might be in terms of safeguarding a herring population.
Although the herring stocks around the UK are identified within various scientific papers and within ICES advisory documents, there remains some uncertainty regarding the status of the West of Scotland and West Irish stocks and whether they are in fact components of one stock, or are genetically separate stocks. A review of literature suggests that genetic testing is underway to determine the status of these stocks, along with the relationship between the different stock components of the West of Scotland stock. These results are expected to be available in 2018.
The perceived impacts of piling on herring focus on spawning herring as opposed to other life stages such as eggs, larvae and juveniles. Perceived impacts are related to underwater sound pressure and particle motion, with there being more understanding of sound pressure and the potential effects upon herring than particle motion. Several studies have identified specific sound exposure criteria for fish and specific thresholds do exist for herring. There are no thresholds currently identified in terms of particle motion. The main effects are perceived to be behavioural changes that arise as a result of piling activities, with there being less concern regarding physiological effects due to the localised extents to which actual injury from levels of noise extends (a very small area around the piling so the risk of population level effect is minimal). Furthermore, sound pressure is considered to be of greater concern for herring as they are considered ‘hearing specialists’, with their swim bladder linked to the inner ear.
There is limited understanding of how behavioural changes affect herring populations (and other fish populations in general) and at what point the behavioural changes become significant and are unacceptable. Most research into this topic has been undertaken in laboratory or confined conditions and have not necessarily reflected conditions in the open sea.
Understanding Consenting Restrictions
A total of 19 OWF projects have been identified to date as having herring restrictions or other mitigation requirements for this species associated with marine licences. The reason that piling restrictions are applied is due to the potential effect of piling upon spawning adult herring and / or their behaviour. The conditions are not related to potential effects of piling noise on herring larvae, principally because of the assumption that the limited range of noise levels is considered to pose little if any risk to eggs or larvae.
Initially in England, blanket piling restrictions were put in place due to the use of a ‘design envelope’ approach for EIAs supporting consent applications. This was due to limited information on the actual final design of the OWFs being available at that time, leading to uncertainty in the actual level of potential impact risk associated with the individual projects and the application of a precautionary approach to ensure protection of spawning herring. Such restrictions would have caused considerable additional expense to developers at the time. As the offshore wind industry became more established, a more evidence based approach was adopted by The Planning Inspectorate, during the Development Consent Order (DCO) process, to piling restrictions with the introduction of flexibility within consent requirements and conditions to reduce or remove these restrictions, subject to the provision of further information and agreement with the MMO and their advisors. As a result of this, further discussions to reduce or remove restrictions have taken place in more recent years, with new evidence being submitted to support these discussions. This evidence has included further seabed sediment and habitat analysis, further specific herring larvae surveys and more refined engineering design information on what will actually be constructed at the OWF. However, the most recent licences to be granted for OWF (for example Hornsea Projects One and Two) have reverted back to the exclusion of this flexibility to further discuss removal of restrictions and have gone on to request further sediment grab sampling to identify whether a suitable spawning habitat is present or not in order to determine the need for further restrictions or mitigation for herring. For these two developments at Hornsea, the concern has changed from one of potential impacts from piling activity to potential impacts from dredging and disposal activities. The area of uncertainty, however, still remains the determination of the extent of suitable herring spawning ground habitat.
Within Scotland, the herring is listed as a Priority Marine Feature (PMF) and is given special consideration in line with the National Marine Plan. Three OWFs in Scottish waters have had herring conditions included within their marine licence. These conditions are for further herring larvae surveys to be undertaken in order to determine the need for further mitigation. This approach is different to that generally applied in English waters and for Beatrice Offshore Wind Farm, herring larvae surveys were undertaken post-consent along with the application of back calculations to determine that spawning grounds were not located within the modelled noise impact zone (which extends beyond the development area). Subsequently MS-LOT determined that no further mitigation would be required for herring.
It is important to recognise that regulators are required to make decisions according to the development site and design that the applicant applies for and the information that is available at that time. If the design envelope is broad, the assessment of impacts may therefore be greater than that which would arise in reality. The onus is on the applicant/developer to make their design envelope as realistic as possible which, in turn, allows any necessary restrictions to be applied at a proportionate level at the time of consent determination.
Regulations and Guidance
Through international and European consultation and literature review, it would appear that spawning herring have not been a concern for OWF development in non UK countries to date, with the main concern being related to the effects of underwater noise on marine mammals, particularly harbour porpoise in a European context. This limited concern seems to be related to the absence of herring spawning areas close to OWFs and this would appear to be by chance, rather than through strategic assessment of proposed OWF lease areas in relation to fish spawning grounds. As such, there is no specific guidance or regulations in place specifically to address herring in non-UK waters.
In the UK and the Republic of Ireland, there are some general regulations and guidance that relate to spawning fish populations as well as to sound exposure criteria and thresholds. In addition, the UK has a Marine Policy Statement in place, which sets out the process for UK countries to form and adopt National and Regional Marine Plans. It is through these Marine Plans that specific policies are / will be set for adoption during determination of OWF applications. For example, in Scotland, herring is listed as a PMF (with specific mention to spawning adults and juveniles in territorial waters) and therefore its conservation needs to be considered during the decision-making process.
Recommendations
The method of undertaking heat mapping of IHLS data (and IBTS data) to identify herring larvae hotspots and therefore determine the main herring spawning grounds is promoted in order to enable a more up to date appraisal of spawning grounds than that presented within the historical Coull et. al (1998) fish sensitivity maps. By using the heat mapping technique alongside the historical data, while overlaying seabed sediment and habitat information, the previous approach of relying upon the Coull et al. mapping can be further refined and modernised to provide more up to date information. This approach can be kept up to date each year by using the most recent 10 years of IHLS data, or five years worth of IBTS data. The benefit of using the 10 year data set (rather than just the most recent single year) is to ensure consideration of a robust data set accounting for inter-annual variation, whilst enabling a contemporary evaluation of current use of the spawning ground. This novel approach fits well when compared with the historical fish sensitivity maps and is an all-encompassing approach using all available evidence base as opposed to a series of older spawning maps. The technique can also be used for any fish species of conservation interest where larvae surveys are undertaken and available over a suitable time series.
One area of uncertainty is identification of the exact location where herring spawn (spawning beds). Attempts to undertake back-calculations of where larvae may have drifted from based upon published prevailing current information, larvae age and growth rates have been undertaken (e.g. Beatrice Offshore Wind Farm Limited, 2016; Brown and May Marine, 2007) but these reports show how broad the resulting conclusions are, with very large areas being cited as potential spawning beds and no exact identification available due to using estimates of current speeds and directions. Marine Space (2013) also looked at this within the aggregate dredging industry for the BMAPA (Section 2.2) and concluded that of all methods that can be used to determine spawning beds, the IHLS data was the best indicator and is a direct measure of spawning where fish of length 0 – 11mm were caught. From a review of literature, there is only one instance where the exact location of a spawning bed has been identified (Ballantray Bay in the Clyde) and this was via grab sampling, where herring eggs were incidentally located on the seabed (Parrish et al., 1959)).
It is recommended that the undertaking of back-calculations is examined in more detail as part of a future study. The approach to undertaking this exercise should be assessed in detail and build upon the earlier work undertaken for the Beatrice and Thanet Offshore Wind Farms. It is recommended that further review of the previous method used is undertaken and that this approach is developed further to provide a more robust methodology of back-calculation. This should include specific prevailing current modelling for the stock / stock component being analysed (e.g. Scottish Shelf Model) rather than utilising generic prevailing current information, which leads to a wide possible area of distribution being identified for drifting larvae. A thorough review of larval growth rates and also the movement of larvae within the water column should be undertaken, along with consideration of possible temporal (diurnal) components such as determining the time of day that the larvae tend to move.
By undertaking this more detailed back-calculation and then applying the IHLS heat mapping technique in combination with provision of seabed habitat data, it may be possible to further refine the historical fish sensitivity maps that are currently used to assist with licence determination and the identification of licence restrictions and other consent conditions.