Recommended bird survey methods to inform impact assessment of onshore windfarms

Updated: March 2025

1. Introduction

The purpose of this document is to define standards for bird survey at proposed wind farms on land in Scotland. Survey is usually required to inform turbine layout, species protection plans (particularly with specially protected species) and mitigation including habitat management plans. It also informs whether there may be a requirement for future monitoring. However, it should be noted that a pre-construction survey specifically aimed at the species and/or issues to be monitored post-consent will likely be required and this survey should use the same methods as those being used for the post-construction monitoring. Whilst this guidance is not prescriptive or able to cater for every possible scenario, if followed it should produce sufficient quantitative information to assess the potential effects of the development on birds. This guidance is not appropriate for repowering proposals but we have published specific advice on Repowering onshore wind farms: bird survey requirements on our website.

The guidance covers:

• Essential pre-survey requirements needed to define a site-specific survey programme.
• Detailed recommendations for survey methods.

Many of the survey methods and principles behind this guidance can also be applied to assessment of other types of development but this guidance is not suitable for the offshore and inshore elements of marine renewables projects.

1.1 Potential Impacts on Birds

Wind farms present three main potential risks to birds (Drewitt & Langston 2006, 2008; Band et al. 2007; Band, 2024):

• Direct habitat loss through construction of wind farm infrastructure.
• Displacement (sometimes called indirect habitat loss) if birds avoid the wind farm and its surrounding area due to turbine construction and operation. Displacement may include barrier effects in which birds are deterred from using normal routes to feeding or roosting grounds.
• Death through collision or interaction with turbine blades and other infrastructure.

For each of these risks, detailed knowledge of bird distribution and flight activity is necessary to predict the potential effects of the wind farm on birds.

2. Pre-Survey

2.1 Principles

2.1.1 EIA regulations

Most wind farm and wind turbine applications fall under the Environmental Impact Assessment (Scotland) Regulations 2017, which specify thresholds for EIA screening by a local authority of more than two turbines and/or turbines or other structures above 15m in height. We recommend that developers use the current Guidelines for Ecological Impact Assessment (EcIA) available on the CIEEM website when assessing impacts on birds arising from wind farm developments although these are not suitable for assessing the impacts on designated sites. Box 1 sets out information on how to deal with extensions and revised proposals rather than new applications.

2.1.2 Designated sites

There are three types of designated site for birds in Scotland:

• Site of Special Scientific Interest (SSSI) under the Wildlife and Countryside Act 1981 (as amended);
• Special Protection Areas (SPA) under the EC Wild Birds Directive; and
• Ramsar sites under the Ramsar Convention (all Ramsar sites are either SPAs or SSSIs).

A wind farm proposal must not adversely affect the interests of these designated sites. Requirements relating to an SPA are, however, much more stringent and are set out in our guidance Habitats Regulations Appraisal (HRA). A more comprehensive survey is likely to be required for these areas.

Developments out with the boundary of a designated site may affect the interest within it. This is of particular importance in relation to SPAs. Proposed wind farm sites which are outside the boundary of a site designated for bird interests must include this interest in any assessment of potential bird impacts when the proposal lies within the regular commuting or foraging distance of that bird interest. The distances over which such effects will occur vary with the ecology and behaviour of the species concerned. We have published guidance on our website Assessing connectivity with Special Protection Areas although the basic approach is also applicable to SSSIs.

Where there is a need to assess impacts on bird species which are not a qualifying interest of an SPA or SSSI, please refer to our guidance on the website on Assessing the significance of impacts on bird populations from onshore wind farms that do not affect protected areas.

2.1.3 Use skilled and licensed observers

The reliability of the assessment is dependent on the observers used to collect the underlying information. Using appropriately skilled and experienced observers is therefore essential.

Many wind farm proposals involve survey and monitoring of a range of Schedule 1 bird species which are specially protected under the Wildlife & Countryside Act 1981 (as amended). A licence is required from NatureScot to visit the nests or disturb these species during the breeding season although, in most cases, the level of survey required for wind farms does not require nests to be visited. Where the presence of breeding Schedule 1 species is known or suspected, all surveyors carrying out breeding bird surveys should be suitably licensed prior to fieldwork commencing. A Schedule 1 disturbance licence is generally not needed for Vantage Point (VP) survey work as locations chosen for this type of survey should not be at a distance close enough to disturb any breeding Schedule 1 bird species.

Schedules A1 (which protect nests all year round) and 1A (which prevents harassment of the species all year) of the Wildlife & Countryside Act apply in Scotland. Four raptor species are listed on one or both Schedules with white-tailed eagle and golden eagle on both A1 & 1A and hen harrier and red kite on 1A. Schedule 1A listing increases the protection given to individual birds and has particular relevance to roosting birds (hen harrier, red kite and white-tailed eagle regularly form communal roosts). Further guidance relating to A1 and 1A species is available on our website in Implications on Additional Protection for Hen Harriers, Red Kite and Golden Eagle under Schedule A1 & 1A of the Wildlife and Countryside Act 1981.

2.1.4 Cumulative impacts

Assessment of the cumulative impact arising from the development of multiple wind farms is required. Sites and species which should be included in any cumulative assessment must be identified as part of the essential preparatory work. Further guidance on our website in the guidance document Assessing the cumulative impacts of onshore wind farms on birds.

2.1.5 Determining the level of survey work required

The potential risk to birds varies with the size and location of the wind farm. For many species, collision and/or displacement risk may only be a serious problem if the wind farm is large. Larger developments will usually require a full EIA to assess potential impacts.

Smaller developments and even single turbines may also present a potential risk if placed in areas of high bird sensitivity, particularly on or near SPAs. Survey effort and assessment of the possible impacts of a proposed wind farm on birds should be in proportion to the scale of the wind farm and the bird interest in the area.

Some small-scale developments will either not require an EIA or, if they are only just over the thresholds, may require a limited environmental assessment. Typically, these very small developments (including ‘domestic’ turbines of 15m or less) are not considered to be of significant risk to birds, exceptions to this being proposals on or near designated sites is where the location ;is needed in many of these cases and simple mitigation such as relocating the turbine further away from a is very close to the nest site of a Schedule 1 bird. Assessment of existing bird data for the area may be all that is needed in many of these cases and simple mitigation such as relocating the turbine further away from a designated site or a Schedule 1 bird nest site may be sufficient to overcome any issues without the need for further survey. More detailed guidance on small scale projects not requiring an EIA is available in on our website in guidance on Assessing the impact of small scale wind energy proposals on the natural heritage.

The location and scale of the proposal, and sensitivity of the bird interest present will determine the target species and the duration of the survey period.

2.2 Process

2.2.1 Essential preparatory work

Prior to commencing survey work, an initial sift of information will help to form a view on the bird populations on a site and their likely sensitivity. This is the key factor in defining the survey requirements.

This is largely a desk-based stage during which existing information is sought and collated on:

1. birds
2. habitats, and thus typical bird communities; and
3. designated sites in the vicinity of the proposed development

The absence of data collated through these sources does NOT necessarily indicate that there is no interest which needs to be assessed.

In the absence of adequate existing information, e.g. no data, data greater than five years old, etc., information must be gathered on site to ascertain the likely bird interest. For example, a ‘Walkover’ method designed to record bird activity over large areas of ground relatively quickly can be used. The walk route should give representative coverage of all key habitats on the site and pay particular attention to features/habitats which may be of a potential ornithological importance, e.g. water bodies, woods, crags. Periodic scanning for birds and stops to listen for calls should be incorporated. Frequency of these walks will vary depending on the site, habitat and likely bird species present, but must cover all seasons.

3. Survey Methods

3.1 Background

Survey data to inform any proposal should have been collected within five years of planned submission, and within three years where species populations are rapidly changing.

There are two main broad survey types involved.

• Distribution and abundance surveys. These record numbers and distribution of breeding, wintering and migrant birds using the site and will allow the evaluation of a site’s importance and provide information to help quantify predicted impacts from disturbance and displacement.
• Vantage Point (VP) surveys. These comprise a series of watches from a fixed location to quantify the flight activity of birds at a proposed development site, providing data to estimate the collision risk. Further information on Collision Risk Modelling is on our website at Wind farm impacts on birds.

This guidance document is not exhaustive and there may well be occasions where novel or different survey methods are required. Where bird survey methods differ from those outlined in this guidance, we would recommend you raise this with NatureScot at pre-app, including a clear rationale for using a different approach.

3.2 Target Species

In most circumstances the target species will be limited to those species which are afforded a higher level of legislative protection. Some species may also be selected as a result of their behaviour which makes them more likely to be subject to impact from wind farms. There are three important species lists from which target species may be drawn:

• Annex 1 of the EC Birds Directive;
• Schedule 1 of the Wildlife & Countryside Act 1981; and
• Red-listed Birds of Conservation Concern.

Given the significant decline in gull populations in Scotland, all gull species should now also be considered as target species. Many gull species have large foraging ranges and may be within connectivity to SPAs a considerable distance away. See identifying theoretical connectivity with breeding site Special Protection Areas using breeding season foraging ranges for more details on connectivity distance.

Consideration should also be given to species identified locally as of conservation concern within Local Biodiversity Action Plans. Target species should be restricted to those likely to be affected by wind farms.

Local circumstances may indicate that survey information should also be acquired on other species, especially those of regional conservation concern. Such species are termed secondary species. Recording of secondary species is subsidiary to recording of target species. The list of any secondary species should also be determined at scoping stage.

Passerine species are not generally considered to be significantly impacted by wind farms. Additionally, care should be exercised when considering Red list species for inclusion in any survey. Attention should be paid to the criteria used for their listing, e.g. restricted or declining wintering populations, which may not be relevant for the proposed wind farm location.

Where designated sites are involved, the specific bird interest of these sites must be assessed. This includes species that are difficult to survey, especially using standard VP survey. Sections 2.1.2 and 3.8.2.1 set out further information on designated sites.

The survey programme should be sufficiently flexible to adapt to situations where one or more species unexpectedly present an issue, especially where these are not chosen as a target or secondary species and have a particularly high presence on the site.

3.3 Area of Survey Required

The survey area and design must adequately cover the entire development area, i.e. the largest possible layout, all the alternative layouts and ancillary structures and works. This includes access tracks; borrow pits, electrical substations and underground and overground grid connections.

Potential collision risk, habitat loss and displacement could affect birds out with the proposal site. The main breeding and wintering bird survey areas should therefore extend at least 500m beyond the development/ planning application boundary. For access tracks and grid connections, the survey area should be 500m either side of the proposed limits of variation of the route. However, depending on the species using the area, there may be a need for further species or species group-specific survey to establish nest, roost or display sites up to 6km from the proposed development site (See Annex 1).

Where there may be access issues or other land management or development activities that constrain the ability to survey the area properly, this should be discussed with us prior to any survey starting or as soon as such issues become apparent during the survey period. The same applies to other land management, e.g. forestry works, and development activities, e.g. small-scale hydro scheme works, taking place on or near the survey area which are likely to affect bird activity.

Where there are adjacent or overlapping wind farm proposals being surveyed simultaneously, we strongly recommend that the developers and consultants cooperate with each other over access and coordination of survey. This will avoid surveying the same area at the same time, people walking through another proposal's survey area during VP watches and a surveyor from one proposal using a VP survey location within the viewshed of another VP location for an overlapping proposal whilst it is being used. Where someone carrying out a survey unexpectedly encounters another surveyor in their area, e.g. during a VP survey watch, the survey should be suspended and the surveyors should report the situation to their employers. Agreement to share data for the overlap area where appropriate or a revised survey schedule that avoids overlapping visits should be sought. This will minimise potential disturbance issues and prevent survey work potentially being compromised.

3.4 Timing of Survey Visits

All bird species have varied seasonal and within day activity patterns. Survey design should be based around the times when birds are likely to be most active. It is important that survey methods are followed accurately (see Section 3.7 and 3.8 for more detail).

3.5 Duration of Survey Period

Survey work should span all times of the year when the target species are present. We recommend survey for a minimum of two years to allow for variation in bird use between years. This may not need to be specifically commissioned, as adequate site-specific information collected for other purposes up to five years old may be available to be used, e.g. black grouse lek data.

In recognition of the wind farm industry moving into more sensitive bird which may impact on the qualifying interests of designated sites, two years survey will be required unless it can be demonstrated by the developer that a shorter period of survey is sufficient. Examples of this may be where a lowland farmland site with lower bird interest or bird activity levels in potentially sensitive locations away from designated sites are much less than anticipated after one year or 18 months of survey. Individual elements of the survey programme should not be split over different years, e.g. breeding bird surveys should cover the whole breeding season of one year. Some birds also alternate the use of ‘traditional’ sites or locations for breeding or feeding between years and these factors should be considered in an assessment. Where proposed survey effort is less than two years, this should be fully justified and agreed with us prior to any survey work.

3.6 Control and Reference Sites

Post-construction monitoring is often a condition of consent for proposals greater than 50MW, which require consent under the Electricity Act. We previously recommended the Before After Control Impacts (BACI) method however it has since been shown that the practical difficulties of selecting a suitable comparison site often renders this method ineffective. We therefore no longer recommend this method. There are a suite of other options available for post-construction monitoring. If in doubt as to which one may be appropriate, please contact us during the scoping phase.

3.7 Distribution and Abundance Surveys

Survey techniques will vary according to species of interest, habitat and time of year. The bird interest at most proposed wind farms in Scotland will be adequately surveyed using the following techniques however, where proposed methods differ from those outlined here, the submission should explain clearly the reasons for this and demonstrate the quality of the resulting data. For other situations not covered below, please contact us.

Details on survey methods specific to species and species groups referred to below- are included in Annex 1.

3.7.1 Moorland breeding birds

The survey method outlined is suitable for many moorland and open country species including, waders, skuas, gulls and some wildfowl species. We generally do not recommend survey of moorland passerines. In previous guidance we recommended the Brown & Shepherd (1993) method but, based on recommendations set out in Calladine et al. (2009), we now require an adapted Brown & Shepherd method with four survey visits at least seven days apart. These should cover the whole breeding season between mid-April and early July, and be done between 8:30 hours and 18:00 hours. They should be carried out in a wind of Beaufort force four or less, and in dry weather.

Timing of breeding varies between species and geographically across Scotland and this should be borne in mind when planning survey. Information on these is published in the Bird Breeding Season Dates in Scotland document on the NatureScot website. Mistimed survey visits will fail to record many birds and will not be of an acceptable standard.

All survey visits should be undertaken in the same season; splitting survey visits between years is not acceptable, e.g. first visit in year one, second visit in year two.

This method is not appropriate for dotterel. More on this species can be found in Gilbert et al. (1998).

3.7.2 Raptors and short-eared owls

Recommended survey techniques are described in Hardey et al. (2013). The Area of Interest will exceed the standard 500m buffer to take account of wide-ranging birds. Species-specific distances are given in Annex 1.

The Scottish Raptor Monitoring Scheme (SRMS) and/or relevant local Raptor Study Group (RSG) branch (see Box 2) should be contacted to; 

• help coordinate surveys between raptor study group members and by wind farm surveyors to minimise disturbance to breeding birds, and/or
• seek historical contextual information, e.g. alternate nest sites and changes in distribution due to fluctuating prey base.

Note the most recent data held by the SRMS will be for the previous breeding season and the local RSG branch will hold the most up to date information for the current breeding season.

It is also important to note that areas without breeding pairs can be used by immature and non-breeding birds. These are important in supporting the wider population as they will include birds that will breed in future. Several raptor species, especially red kite, white-tailed eagle and hen harriers, form communal roosts mostly out with the breeding season. Any roost sites within 2km of a proposed wind farm site should be identified.

3.7.3 Breeding divers

Recommended survey methods for black-throated and red-throated divers are described in Gilbert et al. (1998), note once occupancy has been confirmed there is no requirement to revisit the lochan and occupied sites should be watched from a safe distance where possible. Both these species commute or can commute from their nest sites to feed on the sea or freshwater bodies and this needs to be considered when identifying the survey area (Annex 1; Table 1.6). They can change nest location between years and non-breeding birds can occupy suitable breeding lochs or lochans.

Although often considered to be segregated by loch size, studies have demonstrated that there is significant overlap between the species in loch sizes used. No loch or lochan should be discounted on size alone.

3.7.4 Woodland grouse

All leks for woodland grouse species should be identified within 1.5km of the proposed wind farm site.

The local black grouse study group or capercaillie project officer (see Box 2)) should be contacted to;

• help coordinate surveys to minimise disturbance to breeding birds, and/or
• seek historical contextual information.

For black grouse, known lek sites and other areas of suitable habitat, e.g. open moorland and upland farmland adjacent to woodland or forestry, woodland edges, tracks and clearings in plantations, should be surveyed. Recommended survey techniques are described in Gilbert et al. 1998.

Capercaillie survey methods are set out in Annex 2.

3.7.5 Woodland passerines

Survey of woodland passerines, especially in commercial conifer forest is generally not required. There are several Schedule 1 woodland passerine species, e.g. crested tit, redwing, which have restricted ranges within Scotland. These would only require a survey if the desk study identified that the proposal site is in a key area for the species. A reduced visit Common Bird Census type survey or, if necessary, a point count method (Gilbert et al 1998; Bibby et al, 2000) targeted for these species is recommended in such situations. Crossbill are listed on Schedule 1 and may need to be considered for proposals in commercial forestry in relation to species protection plans but any survey required would be undertaken prior to construction after consent for the proposal.

3.7.6 Nocturnal species, especially owls

Species which are exclusively or largely nocturnal pose particular problems for survey as activity and use of the site occurs mainly during restricted visibility. Nocturnal activity may pose greater hazards for birds at wind farms therefore nocturnal species should be surveyed. Breeding survey should be undertaken and should extend out to 500m beyond the proposed site for non-owl species and 1km for owls.

Survey methods have been published for owls (Hardey et al., 2009) and nightjar (Gilbert et al 1998). In Scotland, nightjars are largely restricted as a regular breeding bird to a few areas of Dumfries & Galloway (D&G) and are unlikely to be encountered elsewhere. A radio-tracking study on nightjars in D&G showed that home range size was variable, ranging from approximately 90 to 600ha (Spray, 2007). Owls and nightjar can be surveyed by listening for calling birds around dusk at appropriate times of year: early spring from February onwards for owls and May-July for nightjar.

For owls, late evening survey for calling juveniles in May-July can also be useful to detect successful pairs. Adults may also be active during this time. Surveys for owls can be complemented by signs of occupation, such as moulted feathers and pellets.

3.7.7 Lowland and farmland birds

Surveys of farmland passerines especially on more intensive arable habitat are generally not required. Where wetlands and water bodies or rough grazing occur, this is likely to introduce the presence of more important species such as breeding waders. Appropriate survey methods for lowland waders are found in Gilbert et al (1998).

Where a range of agricultural habitats occur, information should be gathered on crop and grazing rotations on and within the vicinity of the site. Management changes over time can influence distribution and usage of species across the site.

3.7.8 Wintering and migratory waterfowl, especially geese and swans

Disturbance or displacement of wintering and migrant waterfowl can occur on both roost sites and feeding areas, so surveys for both should be considered.

Substantial information on these species is already held by NatureScot, RSPB and BTO through specific surveys and national monitoring such as the BTO Wetland Bird Survey (WeBS) and the Goose and Swan Monitoring Programme (GSMP). However, these may not be up to date or may not cover all sites.

For greylag geese and pink-footed geese, feeding distribution surveys need only be undertaken when the survey area lies within known goose feeding areas (see Mitchell, 2012). For whooper swan, Greenland white- fronted geese, bean geese and barnacle geese feeding distribution surveys should be undertaken in areas of suitable habitat when the survey area lies within the core foraging distance of SPAs for these species or other major roosts unless it can be established from existing data that the area is not utilised for feeding. Table 2 in our guidance on Assessing connectivity with special protection areas details the species-specific wildfowl foraging ranges. We also hold some feeding distribution data for these species at a few sites.

Feeding distribution surveys of geese and swans should be carried out on a fortnightly basis where species are likely to be wintering, or on a weekly basis for sites where birds are likely to be present in the migration period only. The survey area should extend to 500m from the proposed development site. Feeding distribution surveys can be undertaken by road transects where suitable or by observations from vantage points. In either case it is necessary to ensure that all potential feeding habitat can be surveyed. Depending on topography this may require walking into fields to check blind areas.

In addition, searching the survey area for signs of wildfowl presence (counts of droppings) can help determine if feeding birds are using the wind farm site by night or on days prior to survey visits.

It is known that geese are particularly sensitive to disturbance on roost sites. As such any known roost sites within 1km of the proposed wind farm should be surveyed fortnightly. Methodology should follow Gilbert et al (1998). It should be noted that roost sites may not be restricted to permanent water bodies but may include temporary standing water, intertidal areas and other wetland habitat.

Where significant levels of goose flight activity are anticipated from vantage point surveys, survey of roost sites beyond 1km from the proposed wind farm site may also be necessary to provide contextual information on local population levels for assessing predicted collision risk from VP work.

3.7.9 Coastal species (also includes inland breeding gulls and terns)

During the breeding season, coastal environments may host aggregations of breeding birds, such as seabirds and sea duck, which may be vulnerable to habitat loss and collision with operational turbines. Surveys should focus on identifying such aggregations, especially those within 2km of any proposed development. It should be noted that seabirds, especially gulls, may use inland, coastal and inshore habitats for feeding and other behaviours and there may be regular long distance flight routes inland. Breeding aggregations of other species such as coastal waterfowl should also be considered, especially where they are known to fly between breeding, feeding and resting areas, and may pass through the survey area.

During the non-breeding period, assemblages of wintering waders and waterfowl as well as coastal gull and tern roosts may occur in areas where wind farms are proposed. Surveys should focus on identifying such assemblages, especially those within 2km of any development. Wintering water birds (waders, waterfowl and gulls) may show regular flight routes to and from nocturnal roosts, as well as movements associated with tidal cycles, and these must be factored into survey design if they are likely to use or pass through the survey area.

3.8 Vantage Point Survey

3.8.1 Background

Vantage point (VP) survey is designed to quantify the level of flight activity and its distribution over the survey area. Its primary purpose is to provide data for the Collision Risk Model (Band et al. 2007; Band 2024; NatureScot, 2024), which predicts mortalities from collision with turbines. Data can also be used to provide an overview of bird usage of the site, which may help to inform an overview of potential disturbance and displacement. However, the data gathered on target species other than those for Collision Risk Modelling (CRM) may be biased (see section 3.8.2.1). Where new above-ground grid connections are planned, the proposed connection route should be covered by VP observations to assess potential collision risk.

VP survey must not take place simultaneously with any other fieldwork on the site, as it may cause disturbance and invalidate the VP survey results.

3.8.2 Vantage Point Survey and Detectability

3.8.2.1 Distance

Detectability of birds to human observers declines with distance. This is particularly noticeable with smaller species but can also occur due to the species’ typical flight behaviour, e.g. merlin, some smaller breeding waders. Even larger species such as divers can show declines in detectability under 2km distance. Selection of target and secondary species for VP survey draws from the lists in section 3.2 but potential detectability issues should be borne in mind when selecting VP locations to ensure they are as appropriate as possible for the species to be surveyed (see section 3.8.4 for more information on VP selection).

In some situations, corrections using distance detection methods can be applied. Distance correction calculations require a reference level against which other distance bands can be compared and adjusted. This band is usually the closest to the observer but there may be occasions where this is not possible and other approaches are required. Correcting for distance detection may be problematic if species detection in the furthest distance bands leads to significant numbers of zero values for flight activity. This is because it is not clear whether there was no activity or a very low probability of detecting low levels of activity. As the outermost distance band contributes most to the overall area, a zero value can have a disproportionate effect on the collision risk prediction if a correction is applied to the zero value. Restricting distance detection to bands less than 1km may be advisable but will limit VP coverage. Other methods can also be used, e.g. point counts of wader flocks which can be converted to flux and used in the collision risk modelling. Where survey sites are small with very good VP coverage, and/or flight activity of target species is low and backed up by results of the distribution and abundance surveys, there is unlikely to be a need to undertake corrections.

Additionally, where designated sites, especially SPAs, are involved, there is a requirement to assess impacts on their qualifying interests. Whilst robust collision risk data may not be able to be collected for all qualifying species, any flight activity of qualifying species should be recorded. This will provide information to help inform the assessment, e.g. regular flight lines to/from the designated site may be detected. Note, however, that in some cases specific VPs may need to be established to assess whether there is connectivity to designated sites.

Annex 1 provides additional species and species group specific information for vantage point survey.

3.8.2.2 Terrain

Some areas may be hidden from view due to terrain and/or vegetation. Observers should re-use the exact VP location in successive watches as small changes in VP location can produce significantly different visible areas. Changing the VP location complicates collision risk analysis and assumptions cannot be made that viewshed (the view visible from the VP location) areas will be the same. It is therefore critical that the spatial coordinates of VP positions are measured to the highest level of accuracy using a GPS. The VP locations and associated viewsheds referred to in the environmental statement must be those used in the field.

Birds are often visible when the ground they are flying over is not. Thus, birds can sometimes be seen flying or soaring over hidden valleys and watersheds. Where the key purpose is to estimate the risk of collision with turbines, it is the visibility of the airspace to be occupied by the turbine rotors (the collision risk volume) that is of prime importance. Therefore, it is recommended that visibility be calculated using the least visible part of this airspace, i.e. an imaginary layer suspended at the lowermost height passed through by the rotor blade tips (typically about 20-30m above ground level). Predicting visibility at this level is a simple task using GIS however it should be noted that the baseline should take account of any forestry or other features that will potentially obstruct the view. For example, forestry may be 10-30m high and if viewshed height is taken as 20-30m ground level, the visible area could be overestimated if there is forestry within the viewshed. Being able to view all or most of the site to ground level can be helpful in gauging overall bird activity and usage of the site but is not as important as being able to view the collision risk volume.

3.8.3 Area of Flight Activity Survey

Information is collected during timed watches from strategic VPs. This should cover the defined survey area encompassing the proposed turbine envelope if known, or the maximum extent of potential turbine layouts. This should extend to 500m beyond the outermost proposed turbines to deal with inaccuracies of position for flight line observations. This will also reduce the risk of failing to record birds that use the wind farm area only occasionally.

3.8.4 Vantage Point Watch Selection

When selecting VPs, the aim should be to cover all the flight activity survey area such that no point is greater than 2km from a VP. Any exceptions should be agreed with us. It is very important that VPs are chosen to achieve maximum visibility with the minimum number of points. As detection of flight activity will decrease with distance, VPs should be located as close to the survey area as possible.

It is important to minimise the observer’s effect on bird behaviour. For this reason, VPs are best located outside the survey area where possible. To minimise disturbance, VPs should not be located near to sensitive sites for target species, i.e. nest, roost or lek sites. Observers should try to position themselves inconspicuously to minimise their effects on bird movements. Care also needs to be taken not to locate observation points in locations that may lie directly between the site and a roost or nest site of a key target species, as this can seriously influence the behaviour of birds to be surveyed.

Where VPs are located within the survey area, they should not be used simultaneously with other VP locations which overlook them as the presence of an observer either sitting at or moving to/from the VP may affect bird behaviour.

We recommend scanning an arc of up to 180^o from each VP. Larger arcs cannot be scanned efficiently. With proposals for up to three turbines it may be possible to observe the entire survey area from a single VP. In most cases, however, two or more VPs will be required. For example, an upland site in Scotland measuring around 10km² typically requires three or four VPs. The number of observers required to undertake watches will vary depending on the levels of target bird activity. If activity is predicted to be high and involves several target species, judgement should be exercised as to whether more than one observer may be required in order that all activity of target species can be recorded.

3.8.5 Vantage Point Watch Timings

Watches should be tailored to the ecology of the target bird species involved. This should provide a spread over the full daylight period available, from official local sunrise to sunset times. Watches should be spread across all calendar months when the species is present or likely to be so. The watches should be stratified according to the ecology of the target species present and should give a representative sample of site use.

Migration watches should take account of key periods for the target species to be surveyed, and reference should be made to known peaks in their migration and weather patterns that can produce larger scale movements.

Watches should be taken under conditions of good ground visibility (>2km) and can be during showery days providing showers are not too frequent or prolonged. The cloud base should be high enough to allow observation of the collision risk volume. Ideally such observations should be made in a range of wind conditions. This is particularly important in the case of soaring birds when wind direction and strength is likely to have a large effect on ranging behaviour. Some bird species will fly in poorer weather conditions, e.g. persistent rain/mist, very strong winds (Beaufort Force 7 or above), but useful observation in such conditions is often difficult and is not recommended. The likelihood of such activity should be borne in mind during the assessment though.

3.8.6 Vantage Point Watch Hours

The longer the overall survey period of VP survey, the more accurate and precise the sample of flight behaviour.

We recommend a minimum of 72 hours per VP location divided between seasons (36 hours breeding and 36 hours non-breeding) per year as a standard for species where vantage point survey is required. We expect that VP survey effort would be greater than this if the site is particularly sensitive. Where a high level of migration movement is considered likely, or is known, to occur, sampling within this period should be stratified to ensure adequate data collection across the spring and autumn periods. Additional survey is not always necessary but may be required to cover this adequately in large areas or areas of high activity. Note that some areas may be more heavily used in either spring or autumn, e.g. the West Water SPA pink-footed goose roost is much more heavily used in autumn and early winter than other periods, and this must be considered when designing the VP work. Where proposed effort is less than 72 hours, this should be fully justified and agreed with us prior to survey commencing.

Within each season, each part of the wind farm should be watched for at least 36 hours. If half of the proposed wind farm area has been watched for 36 hours, for example, and the other half has been watched for 36 hours with no overlap in visibility areas, then the proposed wind farm area has been watched for 36 hours, i.e. the time spent observing each part of the proposed wind farm, and not 72 hours, i.e. the total time spent in observation.

Observation time during a distribution survey does not count towards observation time conducted under VP watches: the two methods are not consistent in design or objective.

3.8.7 Vantage Point Watch Durations

We recommend that VP watches are conducted as a series of watches each of no more than three hours continuous duration at a time. They are designed to form a representative sample of bird flight activity and a sample of, for example 12 x three hour watches is better than fewer longer watches. There should be suitable breaks of at least 30 minutes between watches to minimise observer fatigue. Watches can be suspended and then resumed to take account of changes in visibility, e.g. fluctuations in the cloud base, passing rain shower or for the observer to rest. A combination of more than nine hours VP watches should not be carried out by the same observer(s) over the course of a single 24-hour period.

The time taken to move between VP locations must be factored into the survey schedule so that there is sufficient time for surveyors to move around the site without risking disturbance impacts on birds or, if a small team of surveyors is involved, to other VPs being watched. Similarly, there should be a short ‘settling in’ period of approximately 10 minutes at each VP before watches start to allow surveyor familiarisation and ensure any disturbance from observers moving around the site has passed.

3.8.8 Vantage Point Watch Recording

During each watch, two recording methods are used to record data: focal bird sampling for target species and activity summaries for secondary species (See Sections 3.2 and 3.8.2.1).

a. Focal Bird Sampling. The area in view is scanned until a target species is detected at which point it is followed until it ceases flying or is lost from view. The time the target bird was detected, and the flight duration, are recorded. The route followed is plotted in the field onto OS 1:25 000 scale maps (enlarged OS 1:25 000 scale may be useful in some cases). The bird’s flight height is estimated at the time of detection and then at 15 second intervals thereafter, using, for example, a count-down timer with an audible alarm. A 15 second interval is recommended as a practical compromise that aims to minimise dependency within data while maximising the sample of observations.

Flight heights should be classified into height bands, i.e. below the rotor-swept area, within the rotor-swept area and above the rotor-swept area, allowing for observer error. Where there is doubt over the size of turbines to be used, further height bands to reflect the possible turbine sizes can be included. It should benoted however that lots of narrow height bands within the rotor-swept area (10m or 20m increments) are difficult to assess accurately and are likely to increase potential observer error. If conditions allow a finer resolution of height bands, e.g. presence of topographical or structural features of known height, more detailed observations of flight height should be made. Use of a clinometer and range finder provides one means of determining flight height accurately. Observations of target species take priority over completion of activity summaries.

b. Activity Summaries. Each watch should be sub-divided into five-minute periods, at the end of which the number and activity of all secondary species observed should be recorded (target and secondary species to be targeted should be agreed with us at scoping stage prior to survey commencing). If a target species is being tracked at the end of a five-minute period, then the activity summary for that period should be abandoned and a new one started once observations of the target species have ended. Observation of target species takes priority over recording of secondary species. Note that the number of birds recorded should be the minimum number of individuals that could account for the activity observed.

Static and flying birds should be recorded separately. Observers should record perched birds and birds on water bodies once only on arrival at the VP, and the area or site used marked on a map. Thereafter only flying birds and newly noticed perching/swimming birds should be included in the activity summaries. This allows greater time for focal bird sampling, rather than repeated observations of the same static birds. Changes in weather should be recorded as frequently as necessary.

3.8.9 Vantage Point Watch Observational Error and Flight Height Estimation

Observers must be trained in recording of different heights from a distance, and a period of training and familiarisation should be conducted at a site prior to formal observations starting. Comparing observer estimates of the heights of objects against their known heights can be used in training and post-training in the estimation of error.

3.8.10 Recording

We recommend that data for target species are recorded on the forms in Annex 3 and mapped on 1:25 000 map(s). This must be completed for each VP watch, regardless of whether target species were recorded or not. Use different forms for different watches, i.e. do not combine data from different watches onto one form or map. Maps should clearly identify and cross-reference to the form information on:

• the location of the VP used;
• flight lines of target species and direction of flight. Use different colours and symbols for each species, including a key on the form; and
• reference number of each flight line corresponding to that on the form.

Use additional map(s) as appropriate to prevent confusion between overlapping flight lines.

4. Alternative Methods

For species which are active at dawn and dusk, or at night, other methods of recording or assessing activity need to be employed. Night vision/infra-red equipment and survey on moonlit nights can establish presence of nocturnal species or presence and direction of feeding/migration movements both by calls and by sight (although we accept following birds beyond short distances is almost impossible and that, for most species, nocturnal activity is likely to be underestimated in any attempted survey).

Use of automated sensing techniques such as radar have been used particularly in relation to offshore renewables and can be useful in assessing activity. Low-powered surveillance radar, as used in conventional marine navigation systems, offers the simplest means of tracking bird movements in two dimensions to provide an overview of the location and number of bird flight trajectories out to several kilometres (Desholm et al., 2006). This can provide useful supporting information on the volume and location of bird movements.

Such systems cannot discriminate between species of similar size and weight, e.g. common gull and golden plover produce similar echoes, therefore some observations in daylight are required to enable recorded echoes to be assigned to the relevant bird species. In addition, they can be used in conjunction with vertically mounted radar to detect flight heights. We recommend that radar is only used to assess sites where there is likely to be high nocturnal activity of important species, especially if an SPA qualifying species is potentially affected.

In Scotland, radar has been used in onshore cases in relation to assessing nocturnal movements of geese and golden plover. For some species it may be possible to estimate levels of nocturnal activity based on studies on these species. Based on a study in Kintyre of Greenland white-fronted geese (Walls et al., 2006), it has been estimated for geese species that regular nocturnal feeding activity movements can be accounted by for adding on 28% extra activity to observed VP data.

Playback techniques may be helpful in stimulating call responses during nocturnal survey and for other more difficult to detect species. Care should be taken to avoid excessive disturbance to the species, using short bursts of calls (15-30 seconds) several times during a period of 15-20 minutes at any one location. Playback should only be used where suitable habitat for the species is present, and on calm nights. Note that not all birds will respond to playback and that for Schedule 1 species a licence would be required to carry out this type of survey.

Various types of tagging (radio, GPS and satellite) are increasingly being used to investigate a range of bird behaviours. These methods are continuing to reveal important information on migration, foraging and ranging behaviour. Many ongoing tagging projects are publicly available via various websites and can provide useful information, e.g. routes of migrating geese and swans, movements of dispersing young raptors, etc.

Such tagging needs careful planning and can be logistically difficult depending on the species involved. It requires competent, experienced ringers/fieldworkers to trap and mark birds, with such tagging requiring approval from the BTO Ringing Unit’s Special Methods Technical Panel. A Schedule 1 licence may also be required, and a proposal will need further assessment if it involves an SPA qualifying species. Anyone considering undertaking such projects should first contact NatureScot.

For breeding golden eagles, whole range use was previously estimated using the PAT (Predicting Aquila Territory) model in a GIS (McLeod et al., 2002a, b). In recent years new insights into golden eagle ranging and territorial behaviour have become apparent through the research findings from satellite tagged birds in Scotland. This is now showing that the limitations of the PAT model are greater than previously considered and that some of the other assumptions in that model are not correct. This means that in many cases the predictions of percentage losses are unreliable.

The authors/developers of the PAT model have now published a new model, the GET (golden eagle territory), based on data from satellite tagged birds (Fielding et al., 2019), which highlights the importance of ridges to eagles. This is a topographical model and does not restrict itself to known territories. It has initially been developed using data from non-territorial satellite tagged birds but ongoing research tagging territorial adults is showing that the current model is a good proxy for territorial bird usage too. The GET model is publicly available and should be used. Our position on this change is set out on our website in our statement NatureScot statement on modelling to support the assessment of forestry and wind farm impacts on golden eagles. 

In some situations, other methods may be needed to help assess collision risk over the lifetime of the wind farm. For example, where large scale habitat change associated with the wind farm may alter activity levels, e.g. forestry clearance, or where a proposal lies adjacent to an unfavourable SPA and the flight activity of a qualifying species was much lower than would potentially be the case if the population recovers. In such cases where collision risk predictions from the VP survey data are likely to be underestimated, it may be possible to model future collision risk to a particular species based on known flight behaviour and time-activity budgets or by use of other modelling approaches. Information available on some of the key parameters required is likely to be limited depending on the species so the values used in such modelling need careful consideration. Such approaches will typically only be required in certain specific cases and usually after discussion with us. They are not a substitute for good quality field survey.

In some instances, assessing habitat quality, prey abundance and availability can be important in assessing impacts on raptors and can provide important context for survey results, e.g. low hen harrier activity correlated with low vole and/or pipit densities. Whilst some prey species and habitat quality are likely to be covered by other bird and habitat survey work for the proposal, assessing mammal abundance, i.e. voles, rabbit/hares, deer/sheep carrion etc., usually requires additional survey. We recommend that such surveys are undertaken where there is likely to be high activity of raptor species, or where SPA-qualifying raptor species are potentially affected.

5. Recording, Reporting and Presentation of Data

This section sets out what data should be reported to allow potential impacts to be assessed. This helps to both streamline the process and facilitate data use by other parties, for example for cumulative assessments. The following elements are required:

• All survey and assessment must be completed and included in the ES or EIA reports before formal submission of the proposal into the planning system.
• Information must be provided on the bird observation experience level of the staff employed on the field survey team.
• Details must be provided in tabular form for all forms of survey work conducted, including dates, times, observer(s), and weather. An appendix to the environmental statement containing a summary of all the VP survey watches and their results and worked examples of collision risk calculations must be provided to allow collision risk estimates to be independently checked. Examples of VP survey watch and survey visit summary tables are given in Annex 3.
• A summary table containing just the flight line data used in the collision risk modelling must be presented where there is a large amount of flight data involved.
• Flight line activity must be presented in relation to the proposed turbine(s) and application boundary location. Figures should not be cluttered and should allow easy assessment, i.e. not too many species or flights on one figure. If there are multiple species or years displayed on one figure then contrasting colours should be used for each
• The location of VP watch points and the area of visibility from each VP must be presented as a map or maps which show the arc (viewshed) in which the observations were conducted. Such viewshed figures should include details of altitudinal cut–off levels to allow assessment of coverage and whether any lower-level flight activity may have been missed. The map should also show the location of the proposed wind farm including turbine locations and its proximity to any designated sites where relevant.
• Locations of all survey records of birds/nest distribution must be clearly presented in map and tabular form. Agreements on the distribution, publicity and retention of data owned and supplied by external parties must be respected.
• Details of the assessment of impacts should be presented for each target or secondary species where impacts have been identified, even if the impact is deemed to be minor. Where SPA qualifying interest species are involved, the bar for what constitutes a ‘minor impact’ is much higher. The level of impacts often considered minor in non-SPA related cases may not be acceptable in SPA-related cases. Such estimates need to be checked by us and the consenting authority and may be invaluable for future cumulative assessments.

5.1 Confidential Annexes

We have published guidance on Environmental Statements and Annexes of Environmentally Sensitive Bird Information). Key points include:

• information included should be limited to Schedule 1 species and their breeding and/or communal roost locations;
• confidential annexes should not be used to keep non-sensitive information from the public; and
• environmentally sensitive confidential data should be in a separate annex from any other commercial in-confidence data which developers may share with us or consenting authorities.

We and consenting authorities are subject to the Freedom of Information (Scotland) Act 2002 and the Environmental Information (Scotland) Regulations 2004. Both have a presumption that any information held will be released to any member of the public who seeks it, except in certain circumstances and subject to restriction of information whose release might lead to harm to the species. It is therefore important that confidential annexes include only that information which it is necessary to keep confidential.

6. Further Reading

Convention on the Conservation of European Wildlife and Natural Habitats (2003). Windfarms and Birds: An analysis of the effects of windfarms on birds, and guidance on environmental assessment criteria and site selection issues. TPVS/ Inf (2003) 12. Convention on the Conservation of European Wildlife and Natural Habitats (Standing Committee), Council of Europe, Strasbourg

Erickson, W.P., Strickland, M.D., Johnson, G.D. & Kern, J.W. (2000) Examples of statistical methods to assess risk of impacts to birds from wind plants. In: Proceedings of the National Avian-Wind Power Planning Meeting III, San Diego, CA, May 1998, pp. 172-182. NWCC c/o RESOLVE Inc., Washington, DC & LGL Ltd., King City, Ontario.

Fielding, A. H. & Haworth, P. F. (2012). Edinbane Windfarm: Ornithological Monitoring. A review of the spatial use of the area by birds of prey. Haworth Conservation report to Vattenfall, March 2012.

Gauthreaux, S.A. (1996) Suggested practices for monitoring bird populations, movements and mortality in wind resource areas. In: Proceedings of the National Avian-Wind Power Planning Meeting II, Palm Springs, CA, 1995, pp. 80-110. NWCC c/o RESOLVE Inc., Washington, DC & LGL Ltd., King City, Ontario

Martin, G.R.(2011) Understanding bird collisions with man-made objects: a sensory ecology approach. Ibis: 153: 239–254.

Madsen J & Boertmann (2008) Animal behavioural adaptation to changing landscapes: spring-staging geese habituate to wind farms. Landscape Ecology 23:1007–1011.

Morrison, M.L. & Pollock, K.H. (2000) Development of a practical modelling framework for estimating the impact of wind technology on bird populations. In: Proceedings of the National Avian-Wind Power Planning Meeting III, San Diego, CA, May 1998, pp. 183-188. NWCC c/o RESOLVE Inc., Washington, DC & LGL Ltd., King City, Ontario.

Scottish Government (2014) Scottish Planning Policy 

Scottish Government (2011) Town and Country Planning (Environmental Impact Assessment) (Scotland) Regulations.

Scottish Government (2024) National Planning Framework 4.

If you have any comments or queries about this guidance, please contact Sophie Christman, email: Sophie.christman@nature.scot.

7. References

Anderson, R.L., Morrison, M., Sinclair, K. & Strickland, D. with Davis, H. & Kendall, W. (1999) Studying Wind Energy/Bird Interactions: A Guidance Document. National Wind Coordinating Committee, c/o RESOLVE, Washington DC. 

Band, W (2024) Using a collision risk model to assess bird collision risks for onshore wind farms. NatureScot Commissioned Research Report No. 909. Perth.

Bibby, C.J., Burgess, N.D. & Hill, D.A. & Mustoe, S. (2000) Bird Census Techniques (Second edition). Academic Press, London.

Brown, A.F. & Shepherd, K.B. (1993) A method for censusing upland breeding waders. Bird Study, 40: 189-195.

Calladine, J., Garner, G., Wernham, C. & Thiel, A. (2009) The influence of survey frequency on population estimates of moorland breeding birds. Bird Study, 56 (3), 381-388.

Calladine, J. & Morrison, N. (2013) Diurnal and nocturnal ranging behaviour by moorland breeding short-eared owls Asio flammeus in Scotland. Bird Study, 60 (1), 44-51.

Drewitt, A. L. & Langston, R.H.W. (2006). Assessing the impacts of wind farms on birds. Ibis. 148: 29–42.

Drewitt, A. L. & Langston, R.H.W. (2008) Collision effects of wind-power generators and other obstacles on birds. Annals of the New York Academy of Sciences, 1134(1): pp. 233-266(34).

Fielding, A.H., Haworth, P.F., Anderson, D., Benn, S., Dennis, R, Weston, E. & Whitfield, D.P. (2019). A simple topographic model to predict Golden Eagle Aquila chrysaetos space use during dispersal. Ibis, 162(2). 

Gilbert, G., Gibbons, D.W. & Evans, J. (1998) Bird Monitoring Methods. RSPB, Sandy.

Hardey, J., Crick, H., Wernham, C., Riley, H. & Thompson, D. (2013): Raptors: a field guide to survey and monitoring. 3^rd Edition, Edinburgh: The Stationery Office.

McLeod, D., Whitfield, D.P. & McGrady, M.J. (2002a) Improving prediction of golden eagle (Aquila chrysaetos) ranging in western Scotland, using GIS and terrain modelling. Journal of Raptor Research, 36 (1 Supplement): 72-79.

McLeod, D.R.A., Whitfield, D.P., Fielding, A.H., Haworth, P.F. & McGrady, M.J. (2002b) Predicting home range use by golden eagles Aquila chrysaetos in western Scotland. Avian Science, 2: 183-198.

Mitchell, C. 2012. Mapping the distribution of feeding Pink-footed and Iceland Greylag Geese in Scotland. Wildfowl & Wetlands Trust / Scottish Natural Heritage Report, Slimbridge. 

NatureScot, 2024. Guidance on using an updated collision risk model to assess bird collusions on onshore wind farms. NatureScot Guidance Note. 

Spray, S. (2007) Dumfries and Galloway Nightjar Radio Tracking Project: Final Report: Stuart Spray wildlife consultancy report to Dumfries & Galloway Nightjar Working Group.

Walls, R.J., Brown, M.B., Budgey, R. & Parnell, M. (2006) Bird Detection Radar as a tool for monitoring White-fronted goose, Anser albifrons flavirostris movements around the proposed Largie Wind Farm, Kintyre. CSL report for Eurus Energy UK Ltd.

ANNEXES 1-3