Offshore wind developments - collision and displacement in petrels and shearwaters: literature review

6 Remaining evidence needs to inform assessment

The information below was compiled based on the literature review and workshop discussions. While some data are available for many of the parameters relevant to assessing the impacts of offshore wind development, including from studies in Scotland (see 'Catalogue of data sources'), there are some key evidence gaps remaining. While data are incomplete for all three species considered here, information for Leach's Storm-petrel in Scotland is generally more limited than for Manx Shearwater and European Storm-petrel. The order of knowledge gaps presented here and in section 8 is based on Table 6, which follows the trajectory for assessments of offshore wind farm impacts. Note that the Offshore Wind Strategic Monitoring and Research Forum (OWSMRF) is currently conducting a detailed review of the knowledge gaps and research recommendations relating to parameters required for PVA for Manx Shearwater and European Storm-petrel.

6.1 Detectability and diel variation in marine distributions

The use of Digital Aerial Surveys (DAS) is currently the most commonly recommended method for providing the baseline characterisation for wind farm assessments. There are several advantages to DAS, such as providing an audit trail and potentially causing less behavioural change in seabirds than vessel-based surveys. However, there remain a number of potential biases and these could be exacerbated by the behaviour and morphology of some procellariform species. These issues arise from detectability and identification and have not formed part of the main body of the current review as there has been scant work published that examines them. The issues with detectability are twofold: whether the size and flight characteristics of the species make them harder to detect and whether the nocturnal and crepuscular nature of some of the at-sea behaviours means that they are not captured by the survey flights that are restricted to certain daylight hours. All the procellariiform species covered by this review can be active throughout the day and night and with different levels of activity at different times. For example, for Manx Shearwater tracked from Skomer, diving occurred during the day and peaked in the evening (Shoji et al., 2016), while nocturnal foraging was observed from tracking of birds from High Island, Ireland (Kane et al., 2020). These diel variations in activity may mean that key activity periods are not picked up by the constrained timings of DAS. Even where birds have been detected it remains unclear whether morphologically similar species such as European and Wilson's Storm-petrel can be successfully identified to species level. Full consideration should be given to both detectability and species identification in relevant impact assessments. These aspects of DAS for marine ornithology surveys are currently being considered as part of a review being carried out by NatureScot's Scientific Advisory Committee.

6.2 Basic morphometric data

The basic morphometric data used in collision risk modelling (i.e. body length, wingspan) may vary geographically and is lacking for birds of all three species in Scotland. However, while Scotland-specific data would be useful, differences from birds outside of Scotland are unlikely to be large.

6.3 Flight data

The flight speed data currently available come from relatively coarse-resolution tracking data and will tend to be underestimates of the true travel speeds of birds, and measured flight height data are limited or non-existent for these species. There is a need to understand the extent of flight activity and flight heights in different weather conditions, at different times of day and whether flight heights change in response to turbines. While there has been some work on diurnal activity patterns for Manx Shearwater and European Storm-petrel, information for Leach's Storm-petrel is lacking. Flight heights when arriving at or departing/fledging from high elevation nesting sites are also unknown. Fledglings may have particularly poor flight control in the first few days after fledging, making them more vulnerable, but our knowledge of fledgling behaviour is poor.

6.4 Avoidance/attraction behaviour

Since currently operational wind farms overlap very little with the distributions of Manx Shearwaters, European Storm-petrels or Leach's Storm-petrels, very little is known about their avoidance or attraction in relation to offshore turbines, support vessels and associated lighting, infrastructure and activities. This means that many assessments of the impacts of such developments are based largely on expert opinion rather than empirical data. The ability of these species to detect rotating turbine blades is unknown but could inform mitigation options. Whether or not sound influences avoidance or attraction behaviour is also unknown. There are some data available on the energetic requirements of adults of all species to inform assessment of impacts of displacement (resulting from avoidance), but the review found no data in the energy requirements of chicks of European Storm-petrels or Fulmars.

6.5 Light attraction/disorientation

Critical knowledge gaps relate to light attraction and disorientation. Specific aspects include: the range over which light attraction of nocturnal Procellariiformes may occur (and therefore the size of the light catch basin for wind farms and related activities or infrastructure); the extent to which light attraction is exacerbated by particular meteorological conditions (e.g. fog, rain); the influence of wavelength and pattern of illumination (flashing/steady); the extent to which light attraction differentially affects adults and juveniles, and for how long after fledging juveniles may remain particularly susceptible to light attraction.

6.6 Diet

The level of impact caused by displacement of seabirds from foraging areas is related to the degree of dietary specialisation and the distribution of food resources. Food availability within wind farm developments may also influence the degree to which birds are attracted to the area. There have been very few studies of the diet of these procellariiform species, especially within Scotland, which means that prey distributions, and how prey distributions may change around wind farm developments, are poorly understood. A small amount of metabarcoding of Manx Shearwater diet samples from colonies in Wales has been conducted by the University of Oxford (K. Davies, pers. comm.), and similar work has been carried out at Cardiff University for European Storm-petrels from Mousa (Z. Deakin, pers. comm.), but the results of these studies are not yet published.

6.7 Apportioning impacts to protected colonies

Understanding the connectivity between specific offshore developments and SPA colonies is essential for apportioning impacts to colonies. While long-term data on distributions has been collected by vessel-based surveys, these data do not provide information on the provenance or age of the birds observed, and therefore the connectivity between SPA populations and Plan Option Areas is generally not known. Manx Shearwaters have been tracked extensively from colonies in Wales and Northern Ireland, but limited tracking data are available for Scottish colonies of all three species, and the marine habitat associations of these species in Scotland are therefore poorly understood. It is important to note that birds from colonies outside of Scotland (i.e. in Wales and Ireland) also use Scottish waters and need to be considered in apportioning assessments.

6.8 Evaluation of remaining evidence needs

The evidence needs relating to the key factors involved in the trajectory of assessment of impacts of offshore windfarms on protected colonies of procellariform seabirds are summarised and evaluated in Table 6.

The assessment trajectory commences with quantifying the baseline marine densities of the species of interest and concludes with a Population Viability Assessment for protected colonies, considering both collision and displacement impact pathways. Firstly, the key factors for each stage of the assessment trajectory have been scored ("medium" or "high") in terms of their relative importance within their respective assessment process. In the absence of a formal sensitivity analysis of all factors, we have scored those that may be considered to have an approximately linear effect on the outcome of their respective assessment process as having "medium" importance (e.g. the effect of body length, wingspan or flight speed on collision risk), and those which act in a non-linear manner (i.e. through the existence of a threshold or power relationship, such as flight height) as "high" importance. No factors were considered to have "low" importance. For each species we scored the level of uncertainty surrounding the estimation of each factor as "low", "medium" or "high", based on the availability of evidence identified in the literature review. The "evidence need" for each factor was then scored on the basis of both the importance of the factor, and the level of current uncertainty as shown in Table 5 below.

Finally, the tractability of conducting new research to fill the current evidence gap was assessed using expert judgement as "low", "medium" or "high", taking into consideration factors such as: (i) whether methodologies currently exist; (ii) have been widely used on these (or similar) species elsewhere, or (iii) whether further technological development would be required. Note that costs were not considered in the tractability factor.

*Note that OWSMRF is currently conducting a detailed review of the knowledge gaps relating to parameters required for PVA for Manx Shearwater and European Storm-petrel.