Goose management policy in Scotland: 2010 review

2 The status of Scottish goose populations: conservation obligations, objectives and knowledge gaps

In this chapter we review the obligations that result from current conservation and hunting legislation, and the impact that regulation has had on goose populations in the UK to date. We then review the status and population trajectories of all Scottish goose populations, with particular focus on changes that have taken place since the inception of the current policy in 2000, and since the previous review of policy in 2005. We consider specifically whether any of the relevant 31 recommendations (Scottish Executive 2000, 2005) need altering in the light of our findings. We explore the interaction between goose policy and wider biodiversity policy in Scotland. Finally, we draw conclusions regarding gaps in knowledge that limit policy decisions at the current time, and suggest some priorities for addressing significant knowledge gaps.

2.1 The legislation governing goose management in the UK

There are a number of international treaties or conventions that have been orchestrated to ensure the conservation of birds and their associated habitats through international co-operation. These treaties have been transposed into European and country-based actions through European and domestic legislation, and have relevance for goose policy because they result in conservation obligations.

Information on the most relevant treaties and conventions is provided in Appendix B. Here we summarise the obligations and constraints that the various national and international legislative instruments and agreements place on the Scottish Government with respect to goose populations. The key legal instruments that dictate Scotland's conservation obligations for geese (see Appendix B for further details) are:

• The EU Birds Directive 1979 (referred to below as the Birds Directive);
• The EU Habitats Directive 1992 (referred to below as the Habitats Directive);
• The UK Wildlife and Countryside Act 1981 (referred to below as the WCA; implements the Birds Directive in the UK);
• The UK Conservation (Natural Habitats, &c.) Regulations 1994 (referred to below as the Conservation Regulations; implements the Habitats Directive in the UK);
• The Nature Conservation (Scotland) Act 2009 (amends various parts of the WCA specifically for Scotland).

In addition, the UK is a signatory to the African-Eurasian Waterbird Agreement 1999, under the Bonn Convention 1985 (referred to below as the AEWA), which provides some clear (and internationally accepted) interpretation and development of definitions (specifically for waterbirds including geese) of obligations that are only conceptualised in the Birds Directive and WCA. The UK is also a signatory to the Convention on Biological Diversity ( CBD, 1992, see Appendix B). International conventions and agreement are often not seen as legally binding in the same sense as the legislative instruments listed above (as there is no one formal mechanism for ensuring compliance from signatories). Nevertheless they confer important voluntary obligations on signatories, and pathways for settelment of disputes through arbitration do exist (see for example Chapter 5 in Fasham and Trumper 2001).

2.1.1 Underlying philosophy and generic conservation obligations

The following key principles are established through international conventions and agreements, and the international ( EU) and national legislation resulting from the conventions:

Protection of species

• All naturally occurring bird species in Scotland are protected by law against deliberate killing and the destruction of eggs and nests ( EU Birds Directive 1979 Articles 1, 5; and WCA 1981 as amended Section 1).
• All species are protected during the breeding season (which includes the period when migratory species are returning to their breeding grounds).

Hunting (use and take of species)

• The legislation protects all species and then defines specific circumstances under which species can be killed or taken (e.g. hunted or taken for other reasons).
• These specific circumstances depend on the range, numbers and vulnerability of individual species, which are used to place them on a series of Annexes (see Table 2.1).
• The EU Birds Directive (as implemented in the UK by the WCA 1981) makes the distinction between quarry species, which may killed or taken (hunted) during an open season, and non-quarry species that cannot be killed or taken except by special derogation/provision (Table 2.1).
• The EU Birds Directive deems that "hunting constitutes acceptable exploitation where certain limits are established and respected" and, as such, that "hunting must be compatible with maintenance of the population of these species at a satisfactory level". Species for which hunting is allowed are deemed suitable to place on the appropriate Annexes "because of their high population level, geographical distribution and reproductive rate within the Community as a whole".
• The EU Birds Directive prohibits methods of large-scale or non-selective killing as given in Annex IV (such as use of nets, traps, poison baits, semi-automatic or automatic weapons), which is translated into the UKWCA 1981.
• There is a general ban on marketing (live or dead birds) within the EU Birds Directive, with only specific species being made exceptions (because their "biological status permits"), and these exceptions being specific to the different regions of the EU. The UKWCA 1981 prohibits all sale of geese.

Exceptional provisions (derogations)

• Provision is made within the EU Birds Directive (Article 9) for Member States to except themselves or "derogate" from the overall protection clauses for a species for a number of reasons (including for the prevention of serious damage to crops, which is of the most relevance for goose policy). They may only do so where there is "no other satisfactory solution". This provision has been translated directly into the UKWCA 1981 (as amended) and into the AEWA.
• "No other satisfactory solution" has not been defined legally but there is general acceptance amongst conservation practitioners that the feasibility of use of alternatives to lethal killing (e.g. scaring techniques) is investigated fully before derogations are implemented. There is also no formal definition of "serious damage to crops…".
• In Scotland, licences have been issued for egg control (oiling) for Greylag Geese in the Western Isles on the basis of the prevention of serious damage to crops (as above; see also Sections 4.4.7 and 16.7.6). One conservation NGO stakeholder has questioned the legality of this use of derogation on the basis that it is fully grown birds not eggs that cause agricultural damage. Whilst there is no definitive answer to this legal question (in the absence of a test case), it is certainly the case that it is generally more difficult to make the direct link between the geographical location of goose damage to agriculture and eggs in a nest than it is to link fully grown geese feeding in specific fields to goose damage in those fields.

Sustainability

• The EU Birds Directive (Article 2) commits Member States to maintain the population of each naturally occurring bird species to "a level which corresponds in particular to ecological, scientific and cultural requirements, while taking account of economic and recreational requirements" or to "adapt the population of these species to that level". The Berne Convention (1982) confers similar responsibility on signatories. The EU Directive states that: "Conservation is aimed at the long-term protection and management of natural resources as an integral part of the heritage of the peoples of Europe. It makes it possible to control natural resources and governs their use on the basis of the measures necessary for the maintenance and adjustment of the natural balances between species as far as is reasonably possible".
• The Convention on the Conservation of Migratory Species of Wild Animals (Bonn Convention, 1979) sets a definition for the "conservation status of a migratory species" as "the sum of the influences acting on the migratory species that may affect its long-term distribution and abundance".
• The EU Habitats Directive (1992; Article 2) similarly goes on to define the "conservation status of a species" as "the sum of the influences acting on the species concerned that may affect long-term distribution and abundance of its population within the territory referred to in Article 2". It then defines the conservation of a species as "favourable" when: (i) "population dynamics data on the species concerned indicate that it is maintaining itself on a long-term basis as a viable component of its natural habitats"; (ii) "the natural range of the species is neither being reduced nor is likely to be reduced for the foreseeable future"; and (iii) "there is, and will probably continue to be, a sufficiently large habitat to maintain its populations on a long-term basis". Whilst this definition is not applied to any bird species in the Habitats Directive itself, it should be seen as indicative of a wider legislative definition of favourable conservation status and a qualification of obligations on maintaining population "level" stated in Article 2 of the Birds Directive (above).
• The Agreement on the Conservation of African-Eurasian Migratory Waterbirds ( AEWA, 2008; an agreement under the Bonn Convention), to which the UK is a signatory, sets the fundamental principle (Article II) that: "Parties shall take coordinated measures to maintain migratory waterbirds in a favourable conservation status or to restore them to such a status". Implicit in these statements is the need to maintain both numbers and range of migratory waterbirds covered by the Agreement (all migratory goose species in Scotland are covered; see Table 2.2).
• Under the general conservation measures specified in the AEWA (Article III), any use of migratory waterbirds (i.e. hunting) should be "based on an assessment of the best available knowledge of their ecology" and be "sustainable for the species as well as for the ecological systems that support them". Table 1 of the AEWA lists all the migratory goose species (and populations) according to their current population size and vulnerability (see Table 2.2 for categories). For species listed as category 'A', the AEWA Action Plan stipulates that signatories should: prohibit taking of these species or their eggs; prohibit any deliberate disturbance that would be of significance for their conservation; and prohibit trade. For species listed as category 'B', taking should be regulated to be sustainable (including regulation of timing of take, mode of take, establishing limits on take and controls to show these limits are observed, and prohibiting any trade in contravention of the regulations). The AEWA reflects the EU Birds Directive (and UKWCA 1981) by permitting exemptions to the above prohibitions for some specified purposes, one of which is to "prevent serious damage to crops, water and fisheries".

Non-native species

• With respect to non-native (introduced) species, Article 8(h) of the Convention on Biological Diversity calls on Contracting Parties to "prevent the introduction of, control or eradicate those alien species which threaten ecosystems, habitats or species.". In signing, the UK agreed to reduce significantly the current rate of biodiversity loss by 2010, including by reducing the threat of invasive non-native species.
• The AEWA general conservation measures (Article III) in relation to non-native species specify that signatories should "take all appropriate measures to prevent these species from becoming a potential threat to indigenous species", and particularly (from the AEWA Action Plan 2.5.3) other waterbird species listed in the AEWA Table 1.
• The UK legislation does not currently deal specifically with the control of non-native species in a consistent manner. Section 1 of the WCA prohibits the killing of any wild bird except under licence (and this applies equally to native and established non-native species; Fasham & Trumper 2001), and special conditions must apply for licences to be issued (e.g. to prevent damage to agriculture or to conserve other fauna or flora; as for native species above). The situation is different for e.g. introduced mammals, which are only protected against inhumane killing (see Fasham & Trumper 2001 for further consideration of these issues for non-native species).

Designated sites

• The creation of protected areas is required under the EU Birds Directive (Article 3) as one of a suite of measures to "preserve, maintain or re-establish a sufficient diversity and area of habitats for all the species of birds referred to in Annex 1".
• The UK response has been to establish a series of Special Protection Areas ( SPAs) for each appropriate species, and for the protection of wetlands of international importance (as stipulated in Article 4 of the Directive). These SPAs, and the equivalent Special Areas for Conservation ( SACs) designated under the EU (1992) Habitats Directive ( Appendix B), form the UK contribution to the EU 'Natura' site network.
• The principal criteria for designation of a site as an SPA are: that it supports 1% or more of the UK population of any Annex 1 species; or 1% of the biogeographical population of a regularly occurring migratory species; or it supports over 20,000 waterfowl in any season (see Appendix B for more information).
• It has been UK Government policy to underpin SPAs by first designating them as Sites of Special Scientific Interest ( SSSIs) under UK legislation. The latter aim to protect natural habitats and the species dependent on them through site-based mechanisms/agreements. This has meant that cropped habitats, including semi-improved and reseeded grasslands which are of high importance to geese as a foraging habitat, have largely been excluded from the SPA designation process to date, with the result that SPAs for geese have included roosting sites only, not the related goose feeding areas. However the EU Birds Directive does not preclude the designation of SPAs in what are considered artificial habitats. JNCC is currently funding a review (led by the BTO) of the SPA and Ramsar networks, which will include the development of a decision making tool for the inclusion of cropped habitats, for the following species of geese: Bean, Pink-footed, European White-fronted, Greenland White-fronted, Icelandic Greylag, Greenland Barnacle, Dark-bellied Brent, and East Atlantic Light-bellied Brent Goose. The draft tool is expected by spring 2011.
• There are SSSIs in Scotland designated under the WCA 1981 as amended that are not SPAs but have geese as notified features and therefore confer conservation obligations (e.g. those including native Greylag Geese as notified features).

Table 2.1 Status of Scottish goose populations under national and international legislation

Goose population	Bern Convention (1982) NOTE 1	Bonn Convention (1985) NOTE 2	EC Birds Directive (1979) NOTE 3	UK Wildlife and Countryside (1981) Act NOTE 4
Svalbard Barnacle	Appendix II	Appendix II	Annex I	Unscheduled
Greenland Barnacle	Appendix II	Appendix II	Annex I	Unscheduled
Greenland (Greater) White-fronted	Appendix III	Appendix II	Annex I	Unscheduled NOTE 5
Native Greylag	Appendix III	Appendix II	Annex II/1, Annex III/2	Schedule 1, Part II NOTE 6
Re-established Greylag	N/A	N/A	Annex II/1, Annex III/2	Schedule 2, Part I
Icelandic Greylag	Appendix III	Appendix II	Annex II/1, Annex III/2	Schedule 2, Part I
Pink-footed	Appendix III	Appendix II	Annex II/2	Schedule 2, Part I
Taiga Bean Goose	Appendix III	Appendix II	Annex II/1	Unscheduled
Canada Goose	N/A	N/A	Annex II/I	Schedule 2, Part I NOTE 7

NOTE 1 Bern Convention (1982)

Appendix II - Fully protected at all times
Appendix III - Protected but may be subject to regulated exploitation

NOTE 2 Bonn Convention (1985)

Appendix I - Species threatened with extinction
Appendix II - Species that need or would benefit from international cooperation

NOTE 3 EC Birds Directive (1979)

Annex I - Shall be the subject of special conservation measures concerning their habitat
Annex II/I - May be hunted throughout the EC
Annex II/2 - May be hunted in specified EU Member States only
Annex III/2 - May be legally sold, subject to consultation with the European Commission to assess whether sustainable

NOTE 4 UK Wildlife and Countryside (1981) Act as amended in Scotland

Unscheduled - Fully protected at all times
Schedule 1, Part II - Protected by special penalties during the close season
Schedule 2, Part I - May be killed or taken outside the close season
Schedule 2, Part II - May be killed or taken under an open general licence during the close season by an authorized person for a restricted number of purposes (including to protect crops)
No goose species may be sold

NOTE 5

In England and Wales, the European (Greater) White-fronted Goose is Schedule 2, Part 1

NOTE 6

The "Scottish" native Greylag Goose population is not explicitly recognized in the Act. Rather, the Act gives special protection to Greylags breeding in the "Outer Hebrides, Caithness & Sutherland and Wester Ross only". Because they cannot be separated in the field, naturalized Greylag Geese or any Icelandic Greylags staying to breed in these areas will also therefore be afforded the same degree of protection.

NOTE 7

In England the Canada Goose is on Schedule 2, Part II

Table 2.2 Conservation status of the main goose populations in Scotland

Species	Birds of Conservation Concern 3 NOTE 1	Species of European Conservation Concern NOTE 2	IUCN global red list criteria	AEWA status NOTE 3
Svalbard Barnacle	Amber (species) WL	Non SPEC E Secure	Least Concern (species)	B1 (formerly A2- downgraded in 2009)
Greenland Barnacle	Amber (species) WL	Non SPEC E Secure	Least Concern (Species)	B1
Greenland White-fronted	Red (race only) WL; WI; IUCN; SPEC	Non SPEC Secure	Least Concern (Species)	A2 NOTE 4 (formerly A3c- upgraded in 2009)
Native Greylag	Amber (species) WL; WI	Non SPEC Secure	Least Concern (Species)	B1
Icelandic Greylag	Amber (species) WL; WI	Non SPEC Secure	Least Concern (Species)	B1
Pink-footed	Amber (species) WL; WI	Non SPEC Secure	Least Concern (Species)	B2a
Taiga Bean	Red (race only) WR; IUCN; SPEC	Non SPEC Secure	Least Concern (Species)	B1

NOTE 1 BoCC3 designations

WL - Winter localisation at least 50% of wintering population found in 10 or fewer sites
WI - International importance - UK holds at least 20% of international population
WR - Rare non-breeder (fewer than 900 individuals)
IUCN - race awarded "pseudo-assessment" for BOCC3 purposes only
SPEC - race awarded "pseudo-assessment" for BOCC3 purposes only

NOTE 2 Species of European Conservation Concern designations

Non- SPEC E - Concentrated in Europe but with a favourable conservation status
non- SPEC - Not concentrated in Europe but with a favourable conservation status
SPEC 3 - Not concentrated in Europe but with an unfavourable conservation status

NOTE 3 AEWA listing

A1c: Populations which number less than around 10,000 individuals
A2: Populations numbering between around 10,000 and around 25,000 individuals
A3c: Populations numbering between 25,000 and 100,000 and considered to be at risk as a result of significant long term decline
B1 Populations numbering between 25,000 and around 100,000 (which do not fulfil the criteria for inclusion under A)
B2a Populations numbering between 25,000 and 100,000 and considered to be at risk as a result of concentration onto a small number of sites at any stage of their life cycle

NOTE 4 For this AEWA 'Column A' species, hunting could continue on a sustainable basis only where hunting is a long-established cultural practice ( AEWA Action Plan 2.1)

2.1.2 A practical summary of current conservation obligations with respect to geese in Scotland

It is clear from the above that Scotland's key obligations for all naturally occurring goose species are:

• To ensure maintenance of range and abundance; and
• To ensure sustainable use, for populations that may be hunted legitimately.

None of the legislative instruments or voluntary agreements precludes the management of populations to meet cultural and economic needs as well as conservation needs, so long as such management results in sustainable populations (implicit in this is the need for any management to be 'adaptive', such that it can be altered in response to population changes and external factors; see Section 2.7.1 for further consideration of the principles and practice of adaptive management as applied to wildlife populations). There is also strong legal obligation to ensure the provision of adequate information on which to base decisions about whether management is sustainable.

In practice, working definitions of range and abundance can be difficult to establish because they cannot simply be based on scientific evidence. Whilst not legally binding, the AEWA Agreement, and the status categories for waterbirds therein, provide some working population thresholds (e.g. in relation to those populations that should and should not be subject to quarry status) but these can only ever be a guide, as the extent to which any population can be hunted (or managed through control of numbers) sustainably will always depend on specific aspects of the population biology and ecology of that population. It is clear that legally any goose policy cannot allow goose populations to be driven to extinction. In practice, public preference and stakeholder pressure also come into play in attempting to agree acceptable range and abundance definitions. Our interviews with stakeholders suggested that there has been a reluctance to date amongst the stakeholder community to set target population levels (i.e. acceptable levels of abundance) for Scottish goose populations. Existing policy has quite correctly placed emphasis on population modelling approaches to guide and inform sustainable management to date (see Section 2.5). However, the information available for population modelling to date has been lacking in a number of areas, which we will explore in more detail in the sections that follow.

For non-native species of geese (the most widespread of which is the Canada Goose in Scotland), there are no such obligations under EU legislation to maintain range and abundance or to manage populations sustainably but domestic legislation dictates that any control of such populations still has to be justified on the basis of, for example, damage to agricultural or other nature conservation interests.

2.2 Hunting arrangements in the UK and reporting systems as they apply to geese

2.2.1 Shooting rights

Shooting rights belong to the owner of the land unless transferred to others. A landowner can authorise others to carry out shooting providing they meet other legal requirements (e.g. gun licensing). Wildfowling, as defined as the shooting of geese, ducks and waders below the high water mark, is part of the public right to recreation in Scotland.

2.2.2 Licencing arrangements for the possession of guns

In order to be in possession of either firearms or rifles, sportsmen or women are required by law to hold a shotgun or firearm certificate. These are granted following checks on medical history and liaison with local police. Renewal of licences is carried out on a five year basis and is subject to a fee. It is also possible to shoot in the UK without holding a certificate, as firearms and rifles can be borrowed from a licenced person under supervision and used on land with the landowner's permission. In order for a non- UK resident to shoot they must obtain a UK visitor's permit from the local police, which must be endorsed by a resident sponsor. They must also provide proof that they are legally allowed to be in possession of any firearm imported into the country (e.g. hold an EU Firearms Pass).

2.2.3 Goose quarry species

"Quarry species" (as listed as Schedule 2 part 1 of the Wildlife and Countryside Act 1981) can be shot during the "open season", which varies according to the species concerned. For ducks and geese that are listed as quarry, the open season in Britain (including Scotland) falls between September 1 ^st and January 31 ^st for inland areas and between September 1 ^st and February 20 ^th for areas below the high water mark. The use of artificial or dead decoys is permitted but the use of live decoys is strictly prohibited. There are three species of geese listed as quarry in Scotland: Canada, Greylag (all populations) and Pink-footed Goose.

2.2.4 Bag limits for quarry species and reporting systems for bag numbers

BASC recommend bag limits (per hunter per day) of two Greylag and five Pink-footed Goose: this is a voluntary code of practice for clients of goose guides rather than a legal requirement.

During the open season, quarry species can be taken in Scotland by hunters resident within Scotland, elsewhere in the UK, or hunters from abroad who shoot under a UK visitor permit. The proportions of geese shot by international hunters can be a significant proportion of the total shot. On Orkney, for example, the vast majority of wintering Greylag Geese are thought to be shot by European hunters, notably Italians ( BASC 2009).

There is no statutory requirement in the UK for hunters to report the numbers of quarry species shot during the open season. Consequently there is no mandatory bag reporting scheme in the UK and the numbers shot per quarry species per annum are not known. Under the EU Birds Directive, the UK is required to ensure that any hunting conforms to the "principles of wise use" and "ecologically balanced control of the species of birds concerned". Data on the hunting mortality of geese, in combination with other demographic information, are critical for management of goose populations under EU law. It could therefore be argued that the UK by not implementing a bag reporting system is failing to meet is international obligations, particularly for any declining goose populations.

The Federation of Associations for Hunting and Conservation of the European Union has launched an initiative to provide information on the bag scheme operating in Europe under a project called ARTEMIS ( http://www.artemis-face.eu/). The ultimate aim is to develop a coordinated, centralised databse of bag data across the EU that will facilitate analyses and interpretation. The UK is a part of this project through involvement by GWCT and BASC who co-ordinate the National Game Bird Census and the former WSS survey (see Section 2.7.3 for further information).

2.2.5 Out of season licences for shooting species causing agricultural damage

Licences to shoot (to kill) species during the closed season are issued by central Scottish Government Rural Payments and Inspections Directorate ( SGRPID) staff and can be administered via local SGRPID offices. In order for these licences to be granted, it must have been demonstrated that serious agricultural damage to crops has occurred and other (non-lethal) solutions (e.g. scaring techniques) have been tried and shown to be ineffective (Section 2.1). The basis for the issue of such licences for shooting geese is to support the use of non-lethal scaring techniques and they are not granted for the purpose of population control. Out of season licences impose a strict bag limit per species that can be shot and specify named locations where shooting can take place. The use of decoys is also strictly prohibited on the basis that the licences are granted to reduce agricultural damage not to draw in additional birds to control numbers. To date, such licences have been granted for Greylag, Pink-footed, Greenland Barnacle (Islay only) and Canada Geese in Scotland.

Licences can be issued either to the owner or occupier of the land, provided that they have been granted shooting rights. In addition to the named licence holder, an additional three persons can be authorised in writing to carry out shooting under any given licence. If the main licence holder does not personally carry out shooting, then up to four people can be authorised in writing to carry out the shooting. Changes in the person authorised can be made as long as the maximum number is not exceeded.

The licences issued for the shooting of Greenland Barnacle Goose, a non quarry species, on Islay have been issued by SGRPID staff in Edinburgh and administered in conjunction with locally based SNH staff. These licences are granted under special derogation from the EU Birds Directive on the grounds of preventing serious damage to crops. The use of a derogation in this instance has been controversial since conservation organisations have claimed that there is no evidence to prove that lethal shooting is more effective at protecting crops than non lethal scaring techniques. The methods employed for collecting count and shooting data on Islay, and the lack of an experimental design to address possible confounding influences, have precluded rigorous testing for differences in the effectiveness of lethal versus non-lethal scaring techniques to date (Douglas et al. 2009).

2.2.6 Reporting system for bag numbers for out of season licences

As a condition of the licence, reports that detail the numbers of each species that have been taken must be submitted annually by the licensee. There are currently some problems with the reporting of these figures:

• SGRPID has not been consistent when summarising information, which makes it difficult to look at trends in numbers taken for species over the years.
• There are inaccuracies within the SGRPID summary tables, which mean that figures often do not tally.
• Each year there appears to be a small proportion of licence holders failing to report back the numbers shot (thought by SGPRID staff to be largely from licencees who have applied for licences on a one off basis).

Reporting of bag numbers to SGRPID has been improved from 2009 and information is now also available at the scale of the individual Local Goose Management Schemes (in addition to at the local RPID office and national scales).

The numbers of birds shot under out of season licences are likely to be small compared to the numbers shot during the open season for the quarry species (Greylag, Pink-footed and possibly also Canada Goose), so in the absence of a current system for recording hunting bag data (Section 2.7.3), the tightening up of recording of numbers shot under licence is of a priority (and will do far less to improve predictive population modelling) than for non-quarry species. Greenland Barnacle Geese can only be shot under out of season licences and the special derogation (Section 2.2.5), and hence the accurate reporting of numbers taken is of high priority for this species.

2.2.7 Bag limits for out of season licences

The bag limits set by SNH for Icelandic Greylag and Icelandic Pink-footed Geese have been based on the numbers of birds shot in 2006, which was argued as being appropriate as numbers of geese were not declining in response to that level of shooting pressure. It was assumed that only 60% of the numbers set by the bag limits were actually shot. It is not possible to adapt bag limits according to changes in shooting levels because there is no comprehensive bag reporting scheme (Section 2.7.3). In addition, regional variation in the numbers shot by SGRPID area (see Figure 2.1 for current areas) were also taken into account in the calculations and the current bag limits are shown in Table 2.1. It is recognised in some regions that breeding native and re-established Greylag Geese pose problems for setting bag limits where they overlap in their distribution with Icelandic Greylag Geese, but to date this issue has not been resolved. There is no formal bag limit guidance for native Greylag Geese.

Table 2.9 Bag limits (per licence holder) for wintering quarry goose species (out of season licences).

SEERAD Area (see Figure 2.1	Icelandic Greylag Goose bag limit	Icelandic Pink-footed Goose bag limit	Comment
Northern (includes Northern Isles and Caithness & Sutherland)	10	5	Few Pink-footed Geese occur in this area
Highland	10	15
Grampian (includes Moray)	10	20
Central (Angus, Perth & Kinross)	5	10
South Eastern Scotland	5	5

Bag limits for Greenland Barnacle Geese on Islay were set in 2000 under guidance from SNH. PVA work carried out in1999 based on the Islay population (taken from data up to 1995) indicated that a shooting level of 3.6% of the population resulted in predicted population growth of zero (Pettifor et al. 1999). The NGF then agreed that an overall limit of 5% of the Islay population could be placed on licences each year. Bag limits for Barnacle Geese were then derived differently for non- SPA and SPA areas of Islay. In non- SPA areas, bag limits per farm unit were derived using the three-year average of goose count data, field size, and age of the grass crop. The age of grass crop was taken into account based on the findings of Percival (1996). For SPA areas, it was decided that a maximum of 1% of the overall total of feeding geese within each SPA could be shot (based on the three-year average goose count). This value is then broken down into individual farm units, taking into account the areas of fields in which scaring is to take place and the three-year mean of goose numbers.

Figure 2.1 Boundaries of SGRPID (formerly SEERAD) areas and locations of offices.

2.3 Historical changes in legislation and possible impacts on goose populations

The impacts of changes in regulation on goose populations and consequently on goose conservation status are difficult to prove as, at best, the evidence is correlative and changes in regulations are often temporally confounded with other factors. For example, changes in agricultural practices may have occurred at the same times as the impacts of legislative change were taking place. More specifically, in the 1970s there were major changes in agricultural subsidies that led to the widespread reseeding of former rough hill ground providing highly palatable and nutritious grass that geese were able to exploit (Mitchell et al. 2010), and these changes coincided with the introduction of the EU Birds Directive. These and other such changes will have impacted on goose populations alongside changes in legislation over the years. We have referred to reports and published papers on changes in goose populations, and these have identified the following acts and conventions implemented within the UK where there is some evidence for an impact of the change in regulation on goose populations. It is important to note, however, that any suggestions of legislative impact given below and not given as the sole reasons underpinning goose population changes (for the reasons given above).

2.3.1 National Parks and Access to the Countryside Act (1949)

National Nature Reserves ( NNRs) were first declared under the National Parks and Access to the Countryside Act (1949) and later reinforced by the Wildlife and Countryside Act (1981). Their designation places obligations upon those that own and manage them to do so in an appropriate manner. There are a total of 51 NNRs in Scotland at present.

The establishment of Caerlaverock NNR in 1957 is cited as helping the recovery of the Svalbard Barnacle Goose population (Owen and Black 1999; Mitchell et al. submitted). The NNR at Loch Druidibeg was designated because of the presence of breeding Native Greylag Geese and has also been credited with increases in the population (Mitchell et al. 2010).

2.3.2 Protection of Birds Act (1954)

Major reforms to UK hunting legislation came about with the Protection of Birds Act 1954. This resulted in a number of species being given full protection from shooting, including the Barnacle Goose (with no differentiation between the Greenland and Svalbard populations). There was however, a partial reversal of the act by an order passed in 1955, which permitted an open season for the Barnacle Goose from 1 December to 31 January for all islands west of 5°W (to allow shooting on Islay). Restrictions on hunting were also imposed by the use of a list of species that could be killed or taken outside the closed season only and those species included Bean, Canada, Pink-footed and White-fronted Geese (for the latter there was no differentiation between Greater and European). Greylag Geese were listed as birds, which along with their eggs, were protected by special penalties during the close season. The sale of wild geese was also permitted outwith the period of 28 ^th February and 31 ^st August. The act prohibited the taking of eggs from all wild bird populations.

As population numbers of Svalbard Barnacle Geese were shown to increase in response to a lack of shooting, the Protection of Birds Act 1954, in combination with a lack of shooting in Svalbard from 1995, have been acknowledged subsequently as being important steps in the recovery of the population (Owen and Black 1999). Populations of Icelandic Pink-footed Geese have also increased since the 1950s and this in part has been attributed to restrictions on shooting imposed by the Protection of Birds Act (Mitchell et al. 1999). Numbers of Greylag Geese, both Icelandic and native populations, were found to increase following the Protection of Birds Act and this again has been thought to be due to a lack of shooting pressure. However, we cannot conclude for certain the degree to which the 1954 Act contributed to increasing goose populations, not least because there are no comprehensive hunting bag data to monitor directly numbers of birds shot.

The banning of the taking of eggs has been cited as helping the native Greylag Goose population increase in numbers (Mitchell et al. 2010).

2.3.3 Protection of Birds (1967) Act

In 1967 there was an amendment to the Protection of Birds Act. This included an extension of the prohibition of the sale of wild geese to be in effect for the whole year. This was implemented at the behest of conservation organisations and WAGBI (the predecessor to BASC), due to concerns over the relatively low numbers of geese at that time.

2.3.4 European Birds Directive (1979)

Special Protection Areas ( SPAs) are designated under the EU Birds Directive and are designed to provide special protection for rare and vulnerable species as listed in Annex 1 of the Directive, and for the protection of wetlands of international importance. SPAs for geese are largely designated to confer protection on roost sites and staging areas, and only recently has consideration been given to the protection of feeding areas ( Appendix B, Section 2.6).

The recovery of the Icelandic Pink-footed and Greylag Goose populations has been attributed partly to the safeguard of their winter roosts through SPA designation (Mitchell et al. 1999; Mitchell and Sigfusson 1999).

2.3.5 Wildlife and Countryside (1981) Act

The Protection of Birds Act was largely superseded by the Wildlife and Countryside Act in 1981. Major changes included providing the Greenland Barnacle Goose and Greenland White-fronted Goose with full protection from shooting (European White-fronted Geese were not included). Greylag Geese were also added to the list of species that could be taken during the open season as a quarry species, along with Canada and Pink footed Geese. Conversely, special penalties were introduced against the taking of native Greylag Geese (as defined as those present in the Outer Hebrides, Caithness, Sutherland and Wester Ross) outwith the open season.

Greenland White-fronted Goose numbers rose subsequent to the implementation of the Wildlife and Countryside Act. The Act is therefore cited along with the hunting moratorium in Ireland (which was introduced in 1982 but was not fully implemented until 1989) as having a positive impact on the population of Greenland White-fronted Geese, before the major population declines began in the late 1990s (Fox et al. 1998).

2.4 Obligations and opportunities for goose policy within wider biodiversity policy

2.4.1 The UK Biodiversity Action Plan process

The UK Biodiversity Action Plan ( UKBAP; 1994) is the UK Government's response to the Convention on Biological Diversity (1993; see www.cbd.int/convention). UKBAP involves developing action plans for the most threatened species and habitats (Species Action Plans, SAP, and Habitat Action Plans, HAP), to outline the status of each species and the main threats faced. The plans include key objectives and a number of associated targets designed to help achieve recovery of these priority species and habitats, and to monitor success of delivery. At present there are 65 priority habitats and 1151 priority species at the UK level. UKHAPs are unlikely to be directly relevant to geese due to the lack of targeting actions. Therefore, although some of the priority habitats (see Appendix C) may support populations of geese, management is not aimed to benefit geese specifically. There are three priority goose species at the UK level (Biodiversity Reporting and Information Group, 2007): the European (Greater) White-fronted Goose, the Greenland (Greater) White-fronted Goose and Dark-bellied Brent Goose, although only the former has a UKSAP to date ( Appendix C). UKBAP therefore has only limited scope to deliver benefits for Scottish goose populations, either directly through the SAP (for Greenland Whitefront) or indirectly through HAPs.

In Scotland 26 of the 32 local authority areas also have their own Local Biodiversity Action Plans ( LBAPs). As for UKBAPs, they include the generation of HAPs and SAPs. Three of these have HAPs that mention specifically species of geese supported by these particular habitats, and include Greenland Barnacle, Greenland White-fronted, Greylag and Pink-footed Goose. As for UKHAPs, there are no specific targeting actions for geese however. Three Scottish LBAPS have written SAPs for the following species of geese: Greenland Barnacle, Greenland White-fronted and Bean Goose (see Appendix C). Objectives vary according to each LBAP but include maintaining the status at a local level and promoting tourism.

2.4.2 The Scottish Biodiversity Strategy

The Scottish Biodiversity Strategy ( SBS; Scottish Executive 2004) is Scotland's response to the Convention on Biological Diversity and its implementation through the UKBAP. One of the key outputs of the SBS has been the publication of the Scottish Biodiversity List, an inventory of flora, fauna and habitats that are considered to be of principle importance for biodiversity conservation. Three goose species have been selected for inclusion: Greater White-fronted, Barnacle and Bean Goose. In response to the SBS, a strategic approach to species management has been developed by SNH, referred to as the Species Action Framework ( SAF; see www.snh.gov.uk/protecting-scotlands-nature/species-action-framework/). The SAF identifies 32 species for inclusion on a Species Action List, for which targeted management development should be carried out.

The Greenland White-fronted Goose is the only goose species listed as requiring conservation action under the SAF, and a five year implementation plan running from 2007-2012 is now in effect for this species ( Appendix C). The Actions ( Appendix C) include "to ensure effective measures for managing potential conflict with agriculture (local goose management schemes) are retained and extended as appropriate".

There is a UKBAP Habitat Action Plan ( HAP) for machair habitat, for which SNH is the lead partner. Its principal aim is to restore improved machair grassland to traditional mixed management with no over-grazing, with concomitant reductions in stocking levels to avoid over-grazing of machair (and it has targets directed at a staged delivery of this aim ⁴). A consortium lead by RSPB has obtained EULIFE funding to run a project to address these HAB objectives for Natura sites in the Western Isles ( RSPB 2009a, b), including appropriate sites on Coll and Tiree and in the Uists. The Local Goose Management Schemes in these two areas will be supported by this funding for four years (until 2014).

2.4.3 The Scottish Rural Development Programme

The current Scottish Rural Development Programme ( SRDP) sets out how approximately £1.6bn of public expenditure will be deployed over the period 2007-2013 across a range of policy measures broadly relating to rural land management and rural development and split into three "Axes": Axis I relates to business competitiveness; Axis II to agri-environment; and Axis III to rural development.

Funding for these policy measures draws on a mix of Scottish and European Union ( EU) sources, with relative contributions and their distribution across different measures being guided by the EU's Rural Development Regulation ( RDR) and by domestic political priorities. The period beyond 2013 will be covered by a new SRDP, the details of which can not emerge until EU-level negotiations have confirmed the overall structure and funding of the Common Agricultural Policy ( CAP) moving forward.

Within the current SRDP, a number of separate schemes are used to distribute funding across land managers. These are: the Crofting Counties Agricultural Grant Scheme ( CCAGS); the Forestry Challenge Funds; the Food Processing, Marketing and Cooperation ( FPMC) scheme; the Less Favoured Area Support Scheme ( LFASS); LEADER, a Skills Development Scheme; and Rural Development Contracts ( RDCs), with the latter split further into Land Managers Options ( LMOs) and Rural Priorities ( RP). Some schemes sit wholly within one Axis whilst others span two or more axes.

Of the various schemes, RPs are potentially most directly relevant to goose management in that they offer some scope for detailed agri-envrionment management prescriptions. Moreover, they account for around 50% of SRDP expenditure and of this around 60% is already devoted to agri-environment measures under Axis II. Options under each of the Axes are listed in Appendix D. There is now expectation that SNH Natural Care Schemes, including individual SNH Management Agreements, will be funded through Rural Development Contracts-Rural Priorities in future, in particular for paying for the management of SSSIs and Natura ⁵ sites. Consequently, in future SNH may only be able to offer Management Agreements for specialised activities that are not available through Rural Priorities.

For Rural Priorities, the Scottish Rural Development Programme ( SRDP) divides Scotland into 11 regions, and each has set its own regional priorities under nine themes. Some of these relate to business competitiveness and community development, but other are linked explicitly to agri-environmental concerns. For example, the priorities under the theme of Biodiversity are designed to "focus on halting the loss of biodiversity and reversal of previous losses through management, conservation and enhancement". Also providing "protection of rare/endangered species, reducing threat from non-native/invasive species, achieving 'favourable condition' on SSSIs and increase in area of connected natural habitats are all encouraged."

There is a commitment to carry out the following actions, which have been adapted for each of the regions:

• Halt the loss in biodiversity and reverse previous losses through targeted action (this also falls under the Scottish Biodiversity Strategy) which can be achieved through specific management which restores, conserves or enhances Biodiversity Action Plans ( BAPs).
• Ensure that the special features on Scotland's nationally important nature sites are in favourable condition (95% by 2010).
• Carry out improved conservation management for viable populations of rare/and or endangered species that have been listed for priority action in the Species Action Framework for Scotland (this includes Greenland White-fronted Goose) and through targeted action identified in priority species action plans.
• Reduce the threat of non-native species through actions to eradicate or control target species.
• Increase the area of connected natural habitats and ecological features through collaboration between land managers to adopt a landscape scale, particularly in instances where the threat from climate change is at its highest or where beneficial species could increase their range.

Five of the Scottish SRDP regions have Local Goose Management Schemes operating within their boundaries. In addition, there are SNH goose Management Agreements in three of the regions (see Table 2.4). None of the biodiversity priorities in any of these 11 regions currently refer to geese however and any inclusion of goose objectives would probably be alongside rather than above other regional priorities. Moreover, no RP options currently posses the same degree of local flexibility of management prescriptions, payment rates or co-ordination across contiguous parcels of land deemed important in the current local goose management schemes. Hence the effectiveness of current RP options in addressing specific goose issues would be debatable.

Outwith Rural Priorities, other parts of the SRDP could also possibly be relevant to goose management. In particular, LMOs and CCAGS bear some similarities to RP options. However, despite accounting for around 10% of SRDP expenditure, LMOs generally involve only small sums per farm and relatively simple measures. As such they are unlikely to be able to address significant goose pressures in terms of payments or complex management prescriptions or indeed co-ordination of management across neighbouring land. Similarly, although CCAGS offers grants for (amongst other things) grassland improvement, the prescriptions are not tailored to goose management and the grant is for investment purposes rather than on-going management. Moreover, since eligibility is restricted to crofters, it would not be available in all areas affected by geese and even within crofting areas, co-ordination across neighbours might not be straightforward.

Hence, although existing grassland measures under RP, LMOs and CCAGS could be relevant to goose management, none is currently aimed specifically at geese. ⁶ They lack the flexibility of existing local goose schemes in terms of management prescriptions, payment rates and direct involvement of public sector employees. Importantly, they also lack an explicit prioritisation of goose management over competing calls on funding. Whilst such issues could possibly be addressed within the current or next SRDP, doing so would not be easy. The Forestry Challenge Funds, Food Processing and Marketing Scheme, and Skills Development Scheme are not considered relevant to goose management. LFASS is considered later, alongside the Single Farm Payment (see Chapter 7).

The practicalities in terms of funding arrangements and administrative issues for incorporating goose management schemes into the SRDP are discussed further elsewhere (Section 6 and 7). We conclude that, at present, the SRDP provides little opportunity for goose management.

Table 2.4 Existing goose management arrangements by SRDP region

SRDP region	LGMS	Other goose management arrangements
Northern Isles	South Walls	-
Highland	-	-
Grampian	Strathbeg	-
Tayside	-	-
Outer Hebrides	Uists	SNH individual Management Agreements (North Uist)
Argyll	Islay, Kintyre, and Coll and Tiree	SNH individual Management Agreements (Danna and Luig)
Forth	-	Slammanan Bean Goose Scheme
Clyde Valley	-	-
Ayrshire	-	-
Borders	-	-
Dumfries and Galloway	Solway	-

2.5 Status of Scottish goose populations and conservation obligations

The UK hosts large populations of geese that spend the winter within its border and has obligations to safeguard a number of goose populations and their associated habitats under national and international law (section 2.1). Moreover, several of these wintering populations are located mainly within Scotland, such that Scottish Government policy becomes paramount for their sustainable conservation management.

This section covers what are considered to be the main wintering goose populations in Scotland: Barnacle (Svalbard and Greenland populations), Greater White-fronted (Greenland population), Greylag (Icelandic population), Pink-footed (Icelandic population) and Bean (Taiga population) geese. Also included are the resident populations of Greylag (native and naturalised populations), as well as brief reference to the wintering populations of East Atlantic and Canadian Light-bellied Brent Geese and the naturalised populations of Canada, Snow and Bar-headed Geese. The Canada Goose is of particular importance in this context as it is likely to have increased in numbers since the 2005 policy review and may represent a future policy challenge. A number of other goose species can occur in very small numbers in Scotland, either due to escape from captivity or as vagrants with migrant geese, but these do not currently warrant further consideration as part of future goose policy.

Sources of biological information: numbers and demography

The policy decisions required to balance the needs to meet national and international conservation obligations but also sustainable economic growth need a sound evidence base in terms of goose population ecology. Various types of biological evidence are required: population sizes and trends; demographic information (survival, productivity, movements); knowledge of numbers hunted; and knowledge of other likely constraints on population size (perceived and known threats). The main monitoring programmes that operate for goose populations within the UK are the WWT/ JNCC/ SNH Goose and Swan Monitoring Programme, under which a number of single-species surveys are co-ordinated, and the BTO/ RSPB/ JNCC Wetland Bird Survey in association with WWT (see Appendix E for further details of these schemes). Population estimates and trends from these programmes have been further supported by the collation of demographic data (such as productivity, survival and age assessment) to help identify the underlying causes behind increases or decreases in populations.

Sources of biological information: Population Viability Analyses ( PVAs)

Population Viability Analysis ( PVA) is a well established technique in the management of animal populations. Although PVAs were initially designed to predict extinction risks and compare different management options of endangered or rare species (Brook et al. 2000), they are now being used for more common and widespread species, including geese. To date SNH have commissioned a number of PVAs for the following populations: Svalbard Barnacle, Greenland Barnacle, Icelandic Greylag, Icelandic Pinkfoot and Greenland White-fronted Goose. In addition there has recently been an assessment of the current state and analyses of future trends for the native Greylag Goose population.

The PVAs carried out for Scottish goose populations have generated predictions based on various scenarios including: (a) no changes in management or environmental conditions (baseline models); (b) changes in non-natural mortality rates (e.g. simulating variation in shooting pressure on the population); and (c) reducing productivity (e.g. simulating pressures at the breeding grounds). In general several key outputs are generated: the predicted population in 25 years; and the likelihood of a population falling below set threshold populations ⁷, which are referred to as quasi extinction thresholds. When constructing the models for geese, the relationship between population size and other demographic parameters (such as breeding success) has been tested for density dependence, and if none can be found, then model outputs have been generated for both density dependent and density independent population processes.

As the outputs of the population modelling are rather critical for making decisions about the feasibility of, and risks associated with, future management scenarios, there were several aspects of the current modelling available that were important to review:

• How well the modelling has performed to date in predicting population changes (by comparison with empirical count data);
• The sensitivity of the modelling to variation in the various input parameters (demographic rates; information on numbers shot); and
• The quality of the information (input parameters) available for each goose species (in particular the degree of uncertainty that they introduce into the model predictions) and hence priority gaps in knowledge.

It is important to raise some limitations of the population modelling that has been carried out to date at the outset. We found it difficult to make comparisons between the PVA outputs generated for each goose species, due the high degree of variation in how the results have been reported to date. There was a lack of reporting of samples sizes when estimating demographic parameters for inclusion into the PVA models, which meant it was not possible to evaluate the reliability of the estimates (i.e. the extent to which the population as a whole was represented). Moreover, the graphs generated for the model predictions were hard to read predictions from, and hence any values cited in the following sections of this report are limited to those quoted in the text within the PVA reports. There was also an issue of quoting only the median population estimates in 25 years but it was apparent from the graphs that there was often very large variation in the confidence limits of these values. Thus it was difficult for us to judge objectively how well predicted trajectories were supported by observed population changes during the period since the modelling was undertaken. It was not always clear how the sensitivity analyses had been performed, and how changes in demographic rates could affect the population projections.

There was also an issue in terms of how the results are used and related to policy. The PVA analyses sometimes lack a clear direction in terms of the desired end result (e.g. allow populations to increase, maintain the current levels or decrease the current numbers). This is of particular relevance for species which are continuing to increase in numbers resulting in increased damage to crops and/or increments in public expenditure through the Local Goose Management Schemes where payments are made based on goose numbers.

Changes since policy inception in 2000

In each of the species accounts that follow, we highlight important changes in numbers, demography or threats to populations that have occurred since the inception of current goose management policy in 2000. Changes in numbers since the inception of policy in 2000 are summarized in Table 2.5.

Changes since the 2005 review and recommendations

In each of the species accounts that follow, we highlight important changes in numbers, demography or threats to populations that have occurred since the 2005 policy review and assess the implications of these changes for the 31 recommendations in the 2005 review (Scottish Executive 2000, 2005) and future goose policy.

Current conservation and quarry status

A summary of the conservation and quarry status of each of the main Scottish goose species is provided in Tables 2.1 and 2.2. In addition to their status under EU and UK legislation and international agreements (as explained in Section 2.1), three other relevant conservation classifications are provided:

• Birds of Conservation Concern ( BOCC3, Eaton et al. 2009) is an assessment by a partnership of non-governmental organizations of the status of birds occurring in the UK, the Channel Islands and the Isle of Man. Most naturally occurring Scottish goose species are amber listed due to their relatively localised winter distributions, with two exceptions. The Greenland White-fronted Goose and the Taiga Bean Goose are both red listed because they qualify as Endangered and Vulnerable respectively under IUCN Red List criteria.
• The Species of European Conservation Concern ( SPEC, Birdlife International 2004) is an assessment made by BirdLife International of the European status of bird species. All goose species occurring naturally in Scotland are listed as 'Non- SPEC Secure' but the two populations of Barnacle Goose are flagged because they are both concentrated in Europe.
• The IUCN Red List identifies those species that are at a high risk of global extinction ( http://www.iucnredlist.org/). Although all goose species occurring naturally in Scotland are listed as of 'least concern', BOCC3 have carried out pseudo assessments of the races and have classified Greenland White-fronted Goose and Taiga Bean Goose as IUCN red listed races.

In each of the species accounts that follow, we report on any important changes to status since the 2005 review, and any implications of current status for future policy.

Table 2.5 Population changes over the period that the NGMRG has been in existence (2000-2010).

Species	Biogeographical NOTE 1			Scotland NOTE 1			Local NOTE 2
	Population (year)	Population (year)	Relative change	Population (year)	Population (year)	Relative change	Name	Population (year)	Population (year)	Change	Pattern of change
Svalbard Barnacle	24,000 app (99/00)	32,800 (09/10)	+ 25%	24,000 app (99/00)	32,800 (09/10)	+ 37%	Solway LGMS	NA	NA	NA	Likely increase
Greenland Barnacle	53,823 (98/99)	70,501 (07/08)	+ 31%	45,159 (98/99)	58,269 (07/08)	+ 29%	Islay ( WWT 5 year national census)	35,172 (98/99)	44,961 (07/08)	+ 28%	Increase
							Islay LGMS (international counts)	34,620 (00/01)	37,745 (09/10)	+ 8%	Increase but declines since 2007
							South Walls LGMS (scheme counts)	1,357 (00/01)	1,272 (09/10)	- 6%	Fairly stable
							Uists LGMS (scheme counts)	1,391 (01)	2,392 (10)	+ 71%	Increase with fluctuations
							Tiree and Coll LGMS	2,160 (00/01)	4,609 (09/10)	+ 88%	Increase with fluctuations
							Kintyre LGMS (scheme counts)	41 (00/01)	117 (09/10)	+ 285%	Increase
Greenland (Greater) White- Fronted	32,204 (99/00)	23,162 (08/09)	- 28%	19,017 (99/00)	12,505 (08/09)	- 34%	Islay LGMS (international counts)	12,277 (00/01)	6,005 (09/10)	- 51%	Decline
							Kintyre LGMS (scheme/ international counts)	3,600 (00/01)	3,360 (09/10)	- 7%	Stable
							Tiree and Coll LGMS	1,781 (00/01)	1,253 (09/10)	- 30%	Decrease with fluctuations
Native Greylag NOTE 3	10,000 (97)	34,500 (08/09)	+ 345%	10,000 (97)	34,500 (08/09)	+ 345%	Uists LGMS	4,991 (00)	6,098 (09)	+ 21%	Increase but decrease since 2007
							Coll and Tiree LGMS	3,214 (00)	4,091 (09)	2%	Increase but decline since 2006
Naturalised Greylag	19,000 (1991, UK only)	26,540 (2000, UK only)	39%	2,673 (91)	12,985 (08 /09)	+ 486%	Orkney ( RSPB counts)	110 (00)	10,000 (08)	+ 9090%	Increase
							Islay LGMS	Not known	1800 (09)		Increase
Icelandic Greylag NOTE 4	80,324 (00/01)	98,291 (08/09)	+ 22%	75,822 (00/01)	74,324 (08/09)	-2%	Orkney NOTE 5 ( WWTIGC)	14,414 (00/01)	58,349 (08/09)	Ca + 305%	Increase
							Kintyre LGMSNOTE 6	915	898	- 2%	Stable
Icelandic Pink- footed	245,349 (00/01)	351,118 (08/09)	+ 43%	223,903 (00/01)	248,153 (08/09) NOTE 8	+11%	Loch of Strathbeg	9,682 (April 00)	18,570 (March10)	NOTE 7	Fluctuating, no clear trend
Taiga Bean Goose	100000	70,000 to 90,000 (2006)	- 10 to -30%	ca 180 (00/01)	265 (08/09)	+ 47%	Slammanan	ca 180 (00/01)	265 (08/09)	+ 47%	Increase
Total	-	-	-	400,927	470,901	+ 17%		-	-	-	-

NOTE 1

Global and Scotland population data derived from: Svalbard Barnacle = Mitchell et al. (submitted), Griffin et al. 2009; Greenland Barnacle = Worden et al. 2004, Mitchell et al. 2008; Greenland (Greater) White Fronted = Fox et al. 2009, International Species Action Plan 2009; 2010 GSAG report; Native Greylag = Mitchell et al. 2000, Mitchell et al. 2010; Re-established Greylag = Brown and Dick 1991, Mitchell et al. 2010; Icelandic Greylag = Hearn 2002; Mitchell 2009; WWT pers.comm.; Pink-footed = Hearn 2002; Mitchell 2009; Taiga Bean Goose = AEWA 2002, Waterbird Estimates 2006.

NOTE 2

Local scheme population data were taken either from annual GSAG report or the local offices. Exceptions were the: Greenland Whitefronts on Islay (International annual census based on seasonal averages of the adopted counts used in the GSAG reports rather than the peak counts currently used by the LGMG); Svalbard Barnacle goose on Solway (National annual censuses); Taiga Bean goose on Slammanan from Maciver (2009); Native greylag (Trinder et al. 2009); Pink footed geese at Loch of Strathbeg (based on the max of the March and April roost counts provided by RSPB); Greylag geese on Orkney based on late autumn count peak ( RSPB) and on Islay data provided by John Armitage (Local Recorder).

NOTE 3

The numbers of Native Greylag Geese may also include naturalized Greylag Geese, particularly in the latter survey of 2008/2009.

NOTE 4

The biogeographical population counts for this species were already corrected by removing estimated numbers of resident (breeding) Greylags from the totals. Scottish population estimates were calculated from data provided from the national censuses of 2000/2001 and 2008/2009 (Hearn 2002; Mitchell 2009) and were adjusted to remove the estimated non-Icelandic Greylag population).

NOTE 5

Orkney totals are calculated after removing estimates of 1500 and 10,000 resident breeding Greylags from the counts for 2000/2001 and 2008/2009 respectively (based on WWTIGC data).

NOTE 6

May include breeding Greylag Geese (as no counts of breeding/summering birds are available for this area).

NOTE 7

The numbers of geese within the scheme are not directly counted and data are taken from the spring roost counts of March and April at Loch of Strathbeg, which fluctuate over the years (see Local Scheme accounts in Appendix G).

NOTE 8

Derivation of a true estimate of the 'Scottish wintering population' is problematic due to the high turnover of birds that move through Scotland during the winter. At present the counts in October are used to derive population estimates. WWT have suggested that counts carried out in January, outwith the autumn and spring movement, would be more appropriate to derive a Scottish specific estimate but there are very few years of data available to date .

2.5.1 Svalbard Barnacle Goose

Status and long-term change

Two distinct populations of Barnacle Geese occur in Scotland (excluding a small naturalised population): Greenland and Svalbard. Although these populations are morphologically indistinguishable from one another they are geographically isolated.

The Svalbard population of the Barnacle Goose winters mainly on the Solway Firth, particularly on the Scottish side, with very small numbers recorded in northeast Scotland during the early winter (which are occasionally known to stay for the whole winter period).

In 1948 the global (British) population was estimated as ca 300 birds, which was the lowest count ever on record (Owen and Black 1999). However, by the 1960s, the population had recovered to ca 2,500-3000 birds and since then the population has increased steadily (Figure 2.2) reaching an adopted population count of 32,800 individuals in the winter of 2009/2010 ( GSAG 2010).

Figure 2.2 Counts of the biogeographic Svalbard Barnacle Goose population on the Solway (Data provided by Eileen Rees, WWT. For details of counts see Section 16.3.3 and Appendix E).

Potential threats and hunting pressures

A catastrophic event (such as a disease outbreak or major pollution issue) at the restricted wintering area is considered the greatest risk to this population. Impacts of predation of eggs by Polar Bears on the breeding grounds have been raised recently. Anecdotally this is been shown to have a large impact at a localised level at the breeding sites (e.g. at one colony over 70% of nests were lost through predation, Prop and Wolters 2009). Concerns have also been raised over a possible offshore windfarm in the Solway Firth (Fred Olsen Renewables 2010).

Population modelling

The most recent PVA analyses were carried out on data taken from 1957 to 2003, at which time the population was estimated at ca 27,000 birds (Trinder et al. 2005). The results and key recommendations were reported in the last review in 2005. Since then the population has continued to increase and was estimated at just under 32,800 birds in the winter of 2008/2009 ( GSAG 2010). The 2008/2009 population is already higher than the density dependent predicted population estimate for 2028 of 29,000 individuals; the density independent estimate for 2028 is 63,000, assuming no change in demography or shooting pressures (see Table 2.6).

Table 2.6 Predicted population sizes and quasi extinction probabilities (for the thresholds 1000, 5000, 10000) for Svalbard Barnacle geese for baseline models and additional mortality models. Italics indicate model parameters that generated a stable population. Taken from Trinder et al. (2005). Predictions given are limited to those provided in the text of Trinder et al. (2005).

Model	Predicted population in 25 years (2028)		Quasi extinction probability
			Thresholdn 1000		Threshold 5000		Threshold 10,000
	DD	DI	DD	DI	DD	DI	DD	DI
Baseline No mortality	29,000	63,000	0%	0%	0%	0%	0%	0%
Extra mortality 350 p.a.	27,000	-	-	-	-	-	-	-
Extra mortality 1000 p.a.	22,500	-	<0.1%	1%	0.1%	2.5%	1%	6%
Extra mortality 1500 p.a.	-	27,000-	-	-	-	-	-	25%
Extra mortality 2000 p.a.	-	-	36%	42%	48%	48%	66%	55%
Extra mortality 3000 p.a.	-	-	99%	88%	99%	89%	99%	92%

The PVA needs to be updated for this species, to assess how the predicted population trajectory compares with current population estimates and assess any further evidence for the onset of density dependent processes. The updated analysis could benefit from the inclusion of data from over 10,000 individuals dating back to the 1960s that were excluded previously and the authors suggest could have improved the survival estimates.

Changes since policy inception in 2000

• The Scottish population of Svalbard Barnacle Goose has showed a sustained increase (of 37%) since 1999/2000 (Table 2.5; Figure 2.2).

Changes since the 2005 review and recommendations

• AEWA has recently revised the status of Svalbard Barnacle Goose, downgrading it in 2009 from category A2 to B1 to reflect the population now exceeding 25,000 individuals (Table 2.2). It now has the same AEWA status as Greenland Barnacle Goose, making a derogation to shoot for the purposes of mitigating agricultural damage (if required by future policy) equally as valid as for Greenland Barnacle Goose in terms of the conditions of AEWA.
• Both populations of Barnacle Goose remain on Annex I of the EU Birds Directive (as non-quarry species).
• Concerns have been raised very recently, however, over predation of Svalbard Barnacle Goose eggs by Polar Bears during the breeding season and over a possible offshore windfarm in the Solway Firth.
• International collaboration will be required to establish the impacts of Polar Bear predation on the Svalbard Barnacle Goose population.
• There is a need to update the population modelling for this species in order to: compare rigorously predicted population trajectories with observed changes; further investigate whether density dependence is acting; to take account of recent adverse changes in breeding success; and to potentially include additional data to improve estimates of annual survival rates.

2.5.2 Greenland Barnacle Goose

Status and long-term change

Greenland Barnacle Geese winter almost exclusively in Scotland and Ireland. The majority of the wintering population is hosted on a number of key sites, namely Islay, Coll and Tiree, and South Walls, Orkney.

In 1959, the global population (Britain and Ireland) was estimated at 8277 birds, of which 7000 were thought to occur in Scotland. Over the last 50 years, there has been a steady increase in numbers (Figure 2.3a), apart from a notable drop in numbers in 1983, which was followed by rapid recovery in the following census of 1988. The most recent comprehensive survey in 2008 estimated a global wintering population of 70,501, of which 58,269 (83%) were in Scotland (Mitchell et al. 2008). This represented a 23% increase in numbers within Scotland and a 25% increase at the global level since the last international survey in 2003. Most of the steady increase in population is attributable to the increased numbers on Islay (Figure 2.3a), which holds an estimated 63.8% of the total wintering population (Mitchell et al. 2008).

Figure 2.3a Counts of Greenland Barnacle Geese for the biogeographic, Scottish, Irish and Islay populations based on the International Censuses (Data provided by Carl Mitchell, WWT from WWT/ GSMP/National Parks and Wildlife Service in Ireland: see Appendix E).Note that counts for Islay here vary slightly compared to the international count data used for the local scheme reports (see Figure G1) because a different selection of seasonal counts is used for each purpose.

Figure 2.3b Counts of Greenland Barnacle Geese within local goose management scheme areas: Islay = seasonal average of adopted counts based on the international counts (see Section 16.1.3); Uists = February counts (see Section 16.7.2); South Walls = peak counts (see Section 16.5.3); Kintyre = peak counts (see Section16.2.3); and Coll and Tiree = peak counts (see Section 16.6.3) (Data sources :Islay and Uists data from 2010 GSAG report via Simon Cohen, SNH; South Walls data from LGMS annual reports; Coll and Tiree data from Ben Jones and John Bowler RSPB; see also Appendix G).

Potential threats and hunting pressures

Although the Barnacle Goose is not listed as a quarry species in the EU, a derogation exists under the Birds Directive to permit shooting of Greenland Barnacle Geese to prevent agricultural damage on Islay. Information on the numbers shot is collated annually by the Scottish Government and in 2009, a total of 656 Barnacle Geese were shot (based on bag limits, a further 93 birds may have yet to be reported). Greenland Barnacle Geese can also be shot in Iceland during the open season and an average of 1700 was shot each year between 1995 and 2002 (Trinder et al. 2005).

Population modelling

The most recent PVA analyses were carried out on data taken from 1966 to 2003 based on the Islay population, at which time numbers were estimated at ca 36,000 birds (Trinder et al. 2005). The results and key recommendations were reported in the last review in 2005. Since then the population on Islay has continued to increase, reaching a peak in the winter of 2006-2007 of over 50,000 birds ( GSAG, 2010). This already exceeds the density dependent predicted population of 34,000 for 2028; the predicted population size for 2028 from the density independent modelling is 100,000 individuals (see Table 2.7). Despite the PVA population predictions, there have been declines in the numbers on Islay over the three winters since 2007/2008, with the population estimated at ca 42,000 individuals in 2009/2010.

Table 2.7 Predicted population sizes and quasi extinction probabilities (for the thresholds 1000, 5000, 10000) for Greenland Barnacle geese for baseline models and additional mortality models. Italics indicate model parameters that generated a stable population. Taken from Trinder et al. (2005). Predictions given are limited to those provided in the text of Trinder et al. (2005).

Model	Predicted population in 25 years (2028)		Quasi extinction probability
			Threshold 1000		Threshold 5000		Threshold 10,000
	DD	DI	DD	DI	DD	DI	DD	DI
Baseline (500 shot in UK and 1,700 in Iceland)	34,000	100,000	0%	0%	<0.01%	0%	<0.01%	<0.01%
Extra mortality 500 p.a.	<30,000	-	<0.01%	<0.01%	<0.01%	<0.01%	2.5%	0.01%
Extra mortality 1500 p.a.	<10,000	35,000	19%	11%	27%	15%	42%	21%
Extra mortality 2500 p.a.	-	-	9%	57%	94%	62%	97%	67%

The PVA needs to be updated for this species, particularly as the current models for Islay (whether density dependent or independent) appear to predict population growth against empirical evidence for population decline over the past three winters. There is evidence that there have been recent changes to demographic rates (breeding success; below), the inclusion of which might improve the modelling predictions. Resighting information (from colour marked individuals from areas other than Islay) was not included in the PVAs, and inclusion of that information could improve the survival estimates (according to Trinder et al. 2005). Also, no attempt was made to differentiate between natural and shooting mortality, which should be addressed in future if possible.

Changes since policy inception in 2000

• The Scottish population of Greenland Barnacle Goose as a whole increased by 29% between 1998/1999 and the last full census in 2007/2008 (Table 2.5; Figure 2.3a).
• Local Goose Managament Scheme counts from the Uists and Kintyre have shown increases over the period since 2000, and numbers on South Walls have been fairly stable (Table 2.5; Figure 2.3b).
• Numbers on the Uists and Tiree and Coll have shown some increase since 2000 (Table 2.5; Figure 2.3b).
• Numbers on Islay have fallen over the three winters since 2007/2008 but were still ca 8% higher in 2009/2010 than in the winter of 2000/2001 (Table 2.5; Figure 2.3b).
• The status of the global population cannot be assessed fully until the next international survey takes place (currently programmed for 2013).
• There have also been possible declines in the last four years in the breeding success of the geese wintering on Islay ( GSAG 2010).

Changes since the 2005 review and recommendations

• Annual counts of Greenland Barnacle Geese on Islay have indicated a decrease over the last three years of more than 10,000 birds (Figure 2.3b; GSAG 2010). This could indicate either a true decrease in the global population or a change in geographic distribution.
• We have considered carefully the likelihood that geese could have moved away from Islay (due to scaring disturbance or for other reasons) to winter elsewhere. There is no evidence from ringing for significant mixing with the Svalbard population on the Solway (the last Islay ringed birds sighted there was a pair in February 2006 with only odd birds, less than annual, prior to that, L. Griffin, pers. comm.). Of the key Scottish wintering sites for Greenland Barnacles that are counted every year, trends over the last few years have been variable (Table 2.5; Figure 2.3b; WWT 2010). However, the last aerial survey of Ireland in 2008 showed a ca 35% increase in numbers there compared to the 2003 survey, some sites in Ireland have shown substantial recent increases (e.g. a doubling of birds wintering in Sligo from ca 2000 to 4000; C. Mitchell, pers. comm.) and only ca 50% of the Irish wintering population is watched intensively and checked for rings on an annual basis. In addition, the numbers of Greenland Barnacle Geese ringed on Islay have decreased in recent years (C. Mitchell, pers. comm.), which will reduce the probability of detecting Islay birds elsewhere even if they occur.
• The current numbers on Islay are still ca 8% higher than those reported in 2000/2001.
• The mean breeding success of Greenland Barnacle Geese on Islay, as represented by the percentage of young, over the period 2006-2009 was lower than the mean for 2001-2010 (with particularly low breeding success in 2006 and 2009; GSAG 2010).
• There is a need to update the population modelling for this population because: predicted population trajectories appear not to reflect observed population changes in the most recent years; breeding success may have changed (decreased); information on movements should be incorporated if sufficient to improve modelling; natural and shooting mortality should be treated separately if possible.
• There is a need to update the global and Scottish censuses for this population to check whether the population decreases on Islay are real decreases or a result of population redistribution.
• There has been an increased emphasis on goose scaring under the Islay Local Goose Management Scheme since 2005/2006, making it compulsory to shoot over all new reseeds. Evidence available for the winters of 2002/2003 to 2006/2007 suggested that the policy change resulted in both an increase in the numbers of fields shot over from 2005/2006 onwards (Table 3 in Douglas et al. 2009) and in an almost doubling of fields in which non-shooting scaring was implemented (Table 1 in Douglas et al. 2009).
• The current derogation allowing the lethal shooting of Greenland Barnacle Geese on Islay is perceived by some stakeholders as being contrary to the EU legislation because: it has not been demonstrated that lethal shooting is more effective at mitigating economic losses than shooting to scare or other non-shooting scaring methods (from the results of Douglas et al. 2009); and it involves geographical areas of Islay close to SPAs for Greenland Barnacle Geese (a conflict of conservation versus mitigation interests). Given the recent declines in numbers of Greenland Barnacle Geese on Islay, suspected recent declines in breeding success and uncertainties surrounding the results of the population modelling carried out to date, a precautionary approach would dictate reverting the derogation (see also 2.5.3, 2.7.6 and 2.8). At the very least, the population modelling and full census of this population should be updated and the degree of uncertainty in predicted population trajectories assessed explicitly. It is important to establish whether any redistribution of the population is occurring (e.g. to winter in parts of Ireland not surveyed annually).
• Serious consideration should also be given to previous suggestions to take more of an experimental approach to assessing the affects of scaring on Islay, either via direct counts (e.g. Douglas et al. 2009) or using grass growth as a proxi for goose numbers (e.g. RSPB pers. comm.).

2.5.3 Greenland (Greater) White-fronted Goose

Status and long-term change

There are two subspecies of the Greater White-fronted Goose that occur in the UK: the Greenland and the European (of which only the former is of relevance for Scotland).

The Greenland population of the Greater White-fronted Goose overwinters almost exclusively in the north and west of Scotland and in Ireland (Wexford Slobs in the south east, and and in the west and north Midlands of Ireland). Small numbers also known to occur in Norway. Apart from on Islay (the key site in Scotland), this species tends to occur in relatively small and isolated flocks along the west coast of Scotland (see www.greenlandwhitefront.org/gb-site-inventory).

Increases in the global population (Britain and Ireland) have been observed from 16,541 birds recorded in 1983 to a peak of 35,692 in 1999, when 62% of the population was wintering in Scotland (International Species Action Plan 2009). From 2000 onwards, numbers have declined steadily however (Figure 2.4a). The last published report based on counts in 2009 estimated a global population of 23,162, of which 54% wintered in Scotland (Fox et al. 2009). Of the 17 wintering sites holding fewer than 200 birds, eight currently show declining numbers and extinctions have occurred already at a further nine sites during the last 26 years ( GWGS, 2010).

In terms of Scottish goose populations, the Greenland White-fronted Goose is the only species to have a Species Action Plan under the AEWA, although the plan has yet to be finalised. The plan will be published as an AEWA Technical Report, although formal endorsement will not take place until Spring 2012 when the next meeting of the Parties is due to take place. The Greenland White-fronted Goose is 'red-listed' in Birds of Conservation Concern due to concerns over the status of its population (see Table 2.2).

Figure. 2.4a Counts of Greenland White-fronted Geese for the biogeographic, Scottish and Islay populations based on the International Census (Data provided by Tony Fox, GWGS: see Appendix E).Note that hollow symbols for the biogeographic population indicate years in which the counts were incomplete due to lack of coverage in Ireland and hence are minimum estimates. No data were collected in 2001 due to the outbreak of Foot and Mouth disease.

Figure. 2.4b Counts of Greenland White-fronted Geese for the biogeographic, Irish and Wexford populations based on the International Census (Data provided by Tony Fox, GWGS: see Appendix E).Note that hollow symbols for the biogeographic population indicate years in which the counts were incomplete due to lack of coverage in Ireland and hence are minimum estimates. No data were collected in 2001 due to the outbreak of Foot and Mouth disease.

Figure. 2.4c Counts of Greenland White-fronted Geese within local goose management scheme areas: Islay =seasonal average of adopted counts (see Section 16.1.3); Kintyre = peak counts (see Section 16.2.3); and Coll and Tiree = peak counts (see Section 16.6.3) (Data for local schemes were provided as follows: Islay data from 2010 GSAG report via Simon Cohen, SNH; Kintyre data from Hazel White, SNH; and Coll/Tiree data from Ben Jones and John Bowler RSPB).

Potential threats and hunting pressures

Changes in productivity have to date been accepted as the principle driver of population decline (International Species Action Plan 2009), and population declines have been attributed largely to changes on the breeding grounds.

The Canada Goose has recently expanded its range and is now found breeding extensively in the west of Greenland. Canada Geese are thought to behaviourally dominant to Greenland White-fronted Geese (Malecki et al. 2000 cited by Trinder 2010a). Likely interactions include displacement from breeding territories, although direct evidence is hard to provide particularly at a large scale (Fox et al. 2006). During the the wing moult period, when birds are very restricted in their movement, Greenland White-fronted Geese in the presence of Canada Geese can change from feeding on their preferred food type and have a lower food intake (Kristiansen and Jarrett 2002).

Breeding success has been shown to be consistently lower for the period 1995-2007 for the Greenland White-fronted Geese that winter on Islay and at Wexford Slobs, Ireland (Boyd and Fox 2008). This change is concurrent with change in the values of the Atlantic Multidecadal Oscillation Index ( AMO) to consistently positive values (resulting in both higher levels and higher annual variation in precipitation during the winter and spring, which at this time will fall as snow, at the breeding grounds). In other arctic nesting geese a negative relationship has been demonstrated between snow cover and breeding success (e.g. Pink-footed geese on Svalbard; Madsen et al. 2007). This is likely to be mediated through a delay to the onset of breeding, depleting the reserves of energy built up at the wintering grounds or reducing the availability of foraging sites. In extreme cases, where snow melt is insufficient, birds may abandon nesting attempts altogether (e.g. Red-breasted Geese in Siberia: Kostin and Mooij 1995). Although, there is no empirical snow cover data for the area of the breeding grounds in Greenland where Greenland White-fronted Goose breed, breeding success may have been reduced by a increase in snow cover (Boyd and Fox 2008) according to the mechanisms suggested for arctic-nesting geese in other areas (above).

Hunting of Greenland White-fronted Geese has been banned in the UK since 1982, apart from some shooting under licence between 1985 and 1989 in Wexford, Ireland. Similarly an embargo on hunting of Greenland Whitefronts was imposed in Iceland in 2006, where previously up to 3,300 birds were thought to be shot annually. More recent bag information has shown that up to 300 birds were shot illegally in the two years subsequent to the hunting ban in Iceland. Until recently up to 100-200 birds were also estimated to be shot in Greenland but a ban was also imposed in Greenland in 2009. Hunting of the Greater White-fronted goose is permitted in England and Wales but this is likely to have a minimum impact on the Greenland population.

It is agreed that reductions in breeding success, due to the above factors, are sufficient in magnitude to account for the majority of the decline in numbers shown in recent years, and numbers appear to have stablised somewhat (Figure 2.4)since the ban on shooting in Iceland in autumn 2006 (T. Fox pers. comm.). However, numbers on Islay have shown a disproportionately large decline compared to some other areas (Figure 2.4) and it is not possible for us to conclude definitively that changes in the wintering areas are not a contributory factor. It is possible that Greenland Whitefronts are experiencing competition from the increased numbers of Greenland Barnacle Geese on Islay, particularly given the increased use of intensively managed grasslands by the Whitefronts in the last 20 years and the faster intake rate of the smaller Barnacle Geese, which theoretically could provide them with a competitive advantage (T. Fox pers. comm.).

The Scottish wintering population of Greenland White-fronted Goose comprises a number of relatively small flocks (excluding the population on Islay). Greenland White-fronted Geese are known to be highly site faithful and flocks are thought to restrict their habitat use to a small number of fields. Therefore changes at a localised scale (e.g. alterations in farming practices at the scale of a farm unit) could have a significant impact on these small wintering flocks, potentially renders small populations them at high risk of extinctions. To date it has been difficult to make habitat management recommendations to benefit Greenland White-fronted Geese during the winter period. Traditionally the species foraged on natural wetlands, including oceanic blanket mires and raised bog systems, but over recent years the birds have been shown to also feed on intensively managed grasslands. This change is unlikely to be due to habitat destruction however. Flocks foraging solely on improved grassland and arable stubble during the winter have been shown to have higher breeding success compared to those exclusively feeding on bogland habitats (Fox et al. 2005). This was contrary to popular belief at the time that Greenland Whitefronts were likely to perform better when feeding in their traditional foraging habitat. An ongoing joint study between the Greenland White-fronted Goose Study Group, WWT and SNH will collate habitat use, land use and conservation issues for small wintering flock sites in Scotland. This work is not due to be completed until 2011 but it is anticipated that it will provide crucial information on how to best manage for these small and isolated populations of Greenland White-fronted Geese.

Other factors that have been identified as being of importance have little supporting evidence and are not discussed further at this stage: disturbance at either roost or feeding sites; collision with wind turbines or power lines; degradation of habitat through poor management /mining activities; disease and pollution (International Species Action Plan 2009).

Population modelling

The most recent PVA analyses were carried out on data collected from 1998 to 2007; the Scottish population was estimated at ca 13,000 in 2007 (Trinder 2010a). The baseline models, density dependent and density independent, both predicted a population increase over 25 years, with estimates of 25,000 and 30,000 respectively (see Table 2.8). The quasi extinction risks of the population falling below 10,000 for these simulations were 0 and 3% respectively. The most recent Scottish population estimate indicates that the population has continued to decline, however, with a population estimate of ca 12,500 individuals for the winter of 2008/2009 (Fox et al. 2009).

The modelling considered the impact of resuming higher levels of shooting (e.g. back to the levels when Icelanders were hunting Whitefronts in the years prior to 1996): the predicted populations from both the density dependent and density independent models were zero in 25 years. Even an extra mortality of 2000 and 1300 birds per annum, for density dependent and density independent models respectively, resulted in a quasi extinction risk to a threshold below 10,000 of 100%. Hence it was rightly recommended that the Icelandic hunting ban be kept in place.

The modelling also considered possible changes in the numbers of breeding pairs under two scenarios:

(i) due to the relatively low reproductive output of the population, which appears to have declined further over the last decade (possibly as a result of competition with Canada Geese in western Greenland during the post-breeding moult period); and

(ii) a converse scenario in which the area available for Greenland White-fronts to breed might actually increase if climate change were to result in the reduction of the extent of the Greenland ice-cap (one predicted scenario according to Trinder 2010a)

It was shown that if the density-dependent threshold number that can breed at the maximum rate was reduced by 800 breeding birds per annum, there was a 24% chance of the population falling below the threshold of 20,000 (compared to just 12% predicted from the baseline model). If the threshold number was increased by 800, then there was only a 7% probability of the population falling below 20,000.

Table 2.8 Predicted population sizes and quasi extinction probabilities (for the thresholds 10000, 20000) for Greenland White-fronted goose for baseline, additional mortality and reduced breeding success models. Taken from Trinder 2010a. Predictions given are limited to those provided in the text of Trinder 2010a.

Model	Predicted population in 25 years		Quasi extinction probability
			Threshold 10,000		Threshold 20,000
	DD	DI	DD	DI	DD	DI
Baseline (post moratorium - 300 shot)	25,000	30,000	0%	3%	12%	-
Extra mortality 1000 p.a.	-	-	15%	97%	-	-
Extra mortality 1300 pa.	-	-	-	-	-	100%
Extra mortality >2000 p.a.	-	-	100%	100%	100%	-
Extra mortality 3000 p.a.	0	0	-	-	-	-
Decrease density dependent threshold for birds breeding at maximum rate by 800 p.a	-	-	-	--	24%	-
Increase density dependent threshold for birds breeding at maximum rate by 800 p.a	-	-	-	-	7%	-

The modelling did consider the ban on shooting of Whitefronts in Iceland in autumn 2006 (prior to which ca 3300 birds were shot per annum). It is important to note, however, that at the time the modelling was carried out, 100-200 Whitefronts were still being shot on an annual basis in Greenland (Fox and Stroud 2002) but a ban was introduced in Greenland in 2009. This should now be reduced, with the exception of perhaps 100 birds that may be shot accidentally each year (Trinder 2010a). Sensitivity modelling carried out as part of the PVA analyses indicated that factors which influence adult survival will have the highest relative impact on the population growth rate (compared to juvenile survival and mean brood size). Therefore even relatively modest numbers of birds no longer being shot (in Greenland) on an annual basis may have an impact on the baseline population predictions and, in this case, predicted increases from the baseline model could actually be greater than those of Trinder (2010a).

Changes since policy inception in 2000

• The Scottish Greenland White-fronted Goose population has shown a steady decrease of 34% between 1999/2000 and 2008/2009 (Table 2.5; Figure 2.4a), with the global population showing a 28% decline over the same period. These declines had already started to occur in 1999.
• Numbers on Islay, which hosts the largest number of Whitefronts in Scotland, declined by 51% between 1999/2000 and 2008/2009, numbers on Tiree and Coll decreased by 30%, whilst numbers in Kintyre showed only a minor change over the same period (Table 2.5; Figure 2.4c). Numbers in Wexford (the other major concentration) also did not show any marked decline ( AEWA Species Action Plan 2010; Figure 2.4b).
• Numbers of Whitefronts in the smaller isolated flocks have continued to decline, although there is no evidence to suggest that the rates of decline have changed since 2000 (T. Fox pers. comm.).

Changes since the 2005 review and recommendations

• AEWA upgraded the status of Greenland White-fronted Goose in 2009 from A3c to A2, reflecting the fall in the population to below 25,000 individuals (see Table 2.2).
• Iceland introduced a ban on hunting in 2006 (prior to which ca 3300 birds were shot annually). Greenland introduced a ban on hunting in 2009 (prior to that date 100-200 individuals were thought to be shot annually. The bans on shooting in Iceland and Greenland need to be maintained while the population is still in decline and is of high conservation concern.
• There is a need to update the population modelling for this species because: predicted population trajectories (which predict increases) do not reflect observed population changes (continued decrease, particularly on Islay); less birds should now be being shot due to the shooting ban in Greenland and the models (adult survival estimates) could be particularly sensitive to this change.
• There has been an increased emphasis on goose scaring under the Islay Local Goose Management Scheme since 2005/2006, making it compulsory to shoot over all new reseeds. Evidence available for the winters of 2002/2003 to 2006/2007 suggested that this policy change resulted in both an increase in the numbers of fields shot over from 2005/2006 onwards (Table 3 in Douglas et al. 2009) and in an almost doubling of the number of fields in which non-shooting scaring was implemented (Table 1 in Douglas et al. 2009).
• Whilst there prevails a concensus view that declines in Greenland White-fronted Goose numbers are being driven largely by changes on the breeding grounds (potential competition with Canada Geese and/or increased snow cover) and the very recent decrease in hunting pressure (helping numbers to stabilize), the observed decline on Islay is not entirely consistent with the evidence (and the declining numbers on Islay since 2000/2001 have contributed disproportionately to the overall decline in numbers observed over the last 10 years). Competition with Greenland Barnacle Geese on Islay, and indirect effects of scaring under the Islay LGMS cannot be completely discounted as contributing to the Whitefront decline on Islay.
• Given the high conservation concern attached to this species (including the upgrading of AEWA status), it should be of highest conservation priority under future goose management policy.
• If taking a precautionary approach, it is not possible to conclude for certain that the scaring undertaken on Islay under the LGMS is not having a negative affect on the species' conservation (contravening the AEWA Action Plan as it applies to AEWA Column A species).
• However, contrary to this, if Whitefronts on Islay are experiencing increased competition from Barnacle Geese, appropriately targeted scaring of Barnacles could have a positive influence on the Whitefronts. We simply do not know how the two species interact on Islay, or how they respond to current scaring pressures.
• On Islay, we recommend that in future Greenland White-fronted Geese and Barnacle Geese receive specific and separate treatment in the LGMS objectives and recording protocols. In particular: numbers and species of geese should be recorded in fields immediately before they are shot over (which has not happened to date; Douglas et al. 2009); comprehensive field-by-field information should be collected on the implementation of non-shooting scaring devices, and the use of gas guns (which has not happened to date; Douglas et al. 2009); and more information should be collected and assessed regarding interactions between the two goose species.
• Serious consideration should also be given to previous suggestions to take more of an experimental approach to assessing the affects of scaring on Islay, either via direct counts (e.g. Douglas et al. 2009) or using grass growth as a proxi for goose numbers (e.g. RSPB pers. comm.).
• Greenland White-fronted Goose should also continue to be prioritized in terms of conservation objectives and information collection under other LGMSs where the species occurs (Kintyre).
• To fulfil obligations concerning maintenance of species range, future goose management policy should prioritise and support the positive conservation management of small flocks (and the current Greenland White-fronted Goose Study of land-use and conservation at small wintering sites in Scotland; GWGS April 2010).

2.5.4 Icelandic Greylag Goose

Status and long-term change

There are three populations of Greylag Goose that occur in the Scotland. Two of these are resident all year round: the native and naturalised/re-established populations (below). The third population, of the Icelandic Greylag goose, is only present in Scotland outside the breeding season. Morphologically birds from these populations are thought to be identical and in recent years it has become apparent that the ranges of the different populations may overlap in certain areas. The Icelandic population is by far the most abundant compared to the other sedentary populations.

The Icelandic population of Greylag Goose winters mainly in Britain and Ireland, with much smaller numbers occurring in south west Norway and the Faroe Islands (Mitchell et al.1999).

Numbers of Icelandic Greylag Geese have increased since 1960, when approximately 30,000 birds were estimated, until 1991, when a peak of ca 110,000 birds was attained. Subsequently there was a decline in numbers over the next decade, with an apparent low of ca 73,000 birds in the winter of 2002/2003 (Figure 2.5a). The low count in 2002/03 was likely to be due to the late arrival of birds from Iceland and the lack on inclusion of a number of sites in the survey in 2002 (Hearn 2004). Numbers since then have started to increase again (Figure 2.5a) and the most recent published UK population estimate of wintering birds in 2008/2009 was 98,291 birds, of which 74,324 occurred in Scotland (Mitchell 2009).

Figure 2.5a Counts of the biogeographical population of Icelandic Greylag Goose based on the International Census (Data provided by Carl Mitchell, WWT from WWT/ GSMP/Trinder et al. in press; see also Appendix E).Note that these data include counts from Iceland, Faroes, South-west Norway and Ireland from the late 1990s onwards. The data have been corrected from 1985 onwards to remove the resident population of birds on Orkney that were formerly included in the totals (following Trinder et al. in press). The annual local goose management scheme reports have a more basic (and less accurate) correction factor applied and WWT intend to adopt the new annual population estimates at the end of the year.

Figure 2.5b Numbers of wintering Greylag Geese on Orkney (Data provided by RSPB).Note that these counts include both wintering Icelandic and resident Greylag Geese.

Potential threats and hunting pressures

Previous estimates for the UK annual shooting bag total of Icelandic Greylag geese were ca 15,000-20,000 (Hart and Harradine 2003) and 20,000 -25,000 (Frederiksen 2002) were being shot around the start of the 21 ^st Century. There are likely to have been changes in the numbers shot in recent years due to the significant redistribution of wintering birds (concentrated on Orkney). Total numbers shot in the UK in recent years have been estimated as 10,000-15,000 per year, based on a reduction in the population exposed to shooting and the estimated annual shooting bag of 3,000-4,000 on Orkney (Trinder 2010b). In Iceland, the mean number of Greylag Geese shot annually between 1996 and 2000 was 36,608 (Frederiksen 2002).

There is current concern that hydropower developments in Iceland may reduce the extent of breeding habitat available for Greylag Geese. Although the actual impact of development on the availability of breeding habitat has not been assessed, it has been suggested that a single development which was recently finished could affect up to 2000 pairs (Trinder 2010b).

Population modelling

The most recent PVA analyses were carried out on data collected from to 1998 to 2007; the UK population was estimated at ca 100,000 individuals in 2007 (Trinder 2010b). For the purposes of the modelling, the UK population estimate for Icelandic Greylag Goose was corrected back to 1986, using an estimated growth rate of the Orkney resident population to allow subtraction of the resident birds from the total counts. Adjustments were also made to the estimates of numbers shot, as it was assumed likely that the population will experience lower shooting pressures; it was estimated that ca 8000 less birds per annum were being shot (compared to the 60,000 per annum that were thought to be shot in the late 1990s in the UK and abroad). Taking this as the baseline model (with 52,000 birds shot per annum in total), the predicted population in 25 years was 220,000. If shooting pressure resumed to the levels experienced in the 1990s however, the population could decline to 55,000 individuals and there was a 6% probability that the population would fall below 25,000 birds.

The impacts of the potential loss of breeding habitat due to concerns over hydropower developments in Iceland were also modelled (this was achieved by reducing the breeding population). The risk of the population falling below 50,000 was less than 1% until more than 2,000 pairs were prevented from breeding. However, the risk of decline below 50,000 birds increased to over 50% if more than 3,300 pairs were prevented from breeding (see Table 2.9). The risks of the population declining below 25,000 or 10,000 birds remained less than 1% until the number of pairs prevented from breeding exceeded 2,400 or 2,600 respectively (Trinder 2010b).

Table 2.9 Predicted population sizes and quasi extinction probabilities (for the thresholds Icelandic greylag geese for different shooting levels and reduction in breeding pairs. Predictions given are limited to those provided in the text of Trinder 2010b.

Model	Predicted population in 25 years	Quasi extinction probability
		Threshold 10,000	Threshold 25,000	Threshold 50,000
1990s shooting levels (60,000 p.a.)	55,000	-	6%	-
Reduced shooting levels (57,000 p.a.)	-	-	-	0.5%
Current shooting levels Å baseline (52,000 p.a. adjusted for Orkney shift)	220,000	-	5%	-
Reduced breeding (loss of >2000 pairs)	-	-	-	1%
Reduced breeding (loss of >3300 pairs)	-	-	--	30%
Reduced breeding (loss of >5000 pairs)	-	-	-	100%

The PVA modelling for this species is constrained by the lack of hunting bag data: inaccurate estimates of numbers shot (or survival rates) will have a large impact on the model predictions and the models must assume that hunting mortality is additive rather than compensatory. At the time of the modelling, it was estimated that the northward shift in the wintering area (from 1% of the population wintering on Orkney at the beginning of the 1990s to more than 60%), may have resulted in up to 15,000 less birds being shot each winter in the UK. However, this apparent reduction in shooting pressure had not manifested as an increase in annual survival rates at the time that the modelling was undertaken, which Trinder (2010b) suggested might be due to the short period since the shooting changes took place. However, the fewer data on survival collected in recent years may also have reduced the power to detect changes in survival rates.

Changes since policy inception in 2000

• The Scottish (and UK) population of Icelandic Greylag Goose has increased slowly since 2000, during a period when the biogeographical population increased by ca 22% (largely through increases in the Icelandic wintering component; Table 2.5; Figure 2.5a).
• Since policy inception there has also been a notable shift in the distribution of the Scottish wintering population. Previously, Icelandic Greylags occurred early in the winter in the north and east of Scotland and over the following couple of months they redistributed into the southern regions of Scotland and down into England (Mitchell and Sigfusson 1999). Over the last ten years, however, there has been a dramatic shift in the distribution, with a major proportion of the Icelandic Greylag Goose population now wintering on the archipelago of Orkney (Figure 2.5b).
• The most recent published figures report ca 58,349 Icelandic Greylags on Orkney in the winter of 2008/2009, the equivalent of over half the biogeographical population (Mitchell 2009) and almost 80% of the Scottish wintering population, and an increase of more than 300% since 2000/2001 (Table 2.5). The (unpublished) count on Orkney in the winter of 2009/2010 may be as high as 70,000 birds (Eric Meek pers. comm.).
• It has also been reported that some Icelandic Greylags are starting to overwinter in Iceland (Mitchell 2009).
• Counts of resident (during the breeding season) Greylag Geese on Orkney are not carried out comprehensively every year now however (Eric Meek, pers. comm.; Section 2.5.5), which is making it increasingly difficult to assess with accuracy the size of the Orkney (and Scottish) wintering population of Icelandic birds because birds of the two populations are indistinguishable in the field.

Changes since the 2005 review and recommendations

• The increasing aggregation of wintering birds on Orkney has continued since 2005, although conflicts with agriculture are still somewhat focused on problems during the breeding season caused by resident birds ( Chapter 4 and Appendix H Section 17.2).
• Because of this major shift in the distribution of wintering Icelandic birds, comprehensive counts of the Orkney resident (breeding season) population of Greylags (Section 2.7.3) are required in order to continue to assess numbers of wintering Icelandic birds with any degree of accuracy.
• With the increasing concentration of wintering Icelandic Greylags on Orkney, counting is becoming harder due to the remoteness/lack of accessibility of some parts of Orkney and methods of counting may need to change (e.g. it may be useful to investigate the feasibility and potential cost of aerial surveys on Orkney).
• The increasing concentration of the wintering population on Orkney may place the population at greater risk of a stochastic threat (such as a disease outbreak; see section 4.7.2), and this should be taken into account when deciding how to manage the population.
• Current population modelling is limited by the lack of hunting bag information.
• Better co-ordination between countries in carrying out the winter censuses is required to ensure accuracy of biogeographical population counts.
• Further collaboration with researchers working in Iceland is required to better quantify the potential loss of breeding habitat for Greylag Geese and the extent of risk to the biogeographica (and Scottish) population.

2.5.5 Native and naturalised Greylag Goose

Status and long-term change (native breeders)

The native Greylag is the only native breeding goose in the UK and is generally regarded as being sedentary (e.g. very few records exceeding 100 km from the breeding site; Mitchell 1999). At one time the resident native Greylag population was restricted to the Uists but these birds have since expanded their range to other areas of Scotland, including to Lewis and Harris, Coll andTiree, and the north and west regions of the mainland (Mitchell 1999). The most recent survey of breeding Greylag Geese in Scotland suggested that birds that occur north and west of the Great Glen Fault represent the native breeding population (Mitchell et al. 2010).

The lowest recorded numbers of native Greylag Geese occurred in the 1930s, when there was thought to be a population of only 500 birds on the Uists (Berry 1939 cited by Mitchell et al. submitted). By 1997, the population estimate of native Greylags was thought to be a minimum of 10,000 individuals (Mitchell et al. 2000), and by 2008/2009 it was estimated that there were 34,500 native Greylag Geese, representing a significant proportion of the total population estimate of 47,405 breeding Greylag Geese in Scotland (Mitchell et al. 2010).

Status and long-term change (naturalised population)

The presence of the naturalised Greylag Goose population is largely due to the efforts of wildfowlers who carried out a coordinated scheme to reintroduce the Greylag Goose into area of Scotland and England, using eggs and goslings collected from the native stock between the 1930s and 1970s. This was largely in response to the large decline in numbers of the native Greylag Goose at the start of the 20 ^th Century. The most recent survey of breeding Greylag Geese in Scotland suggested that birds which occur south and east of the Great Glen fault represent the naturalised breeding population (Mitchell et al. 2000). The naturalised Greylag Goose population of Scotland was estimated at 2,673 birds in 1991 (Brown and Dick 1992). In 2008/2009 this had increased to an estimated 12,985 individuals, of the combined Scottish breeding Greylag Goose population total of 47,405 (Mitchell et al. 2010).

The naturalised goose population of the UK was estimated in 1991 at 19,000 individuals (Delany 1992) but by 2000 the numbers had increased to 26,540 (Rowell et al. 2004). Survey coverage in Scotland was not particularly good in the former year however (see Appendix E)

Population threats and hunting pressure (native population)

On Tiree, hunting bag data have been collected since 1997 by SNH: the numbers shot show no overall trend and range from ca 200 to 1,700 per annum (although this upper estimate has been questioned subsequently; Trinder et al. 2009). Licences to shoot legally during the closed season were first issued on Tiree in 2004, which has had substantial effect on the numbers shot (Trinder et al. 2009). On the Uists, where bag data are also collated by SNH, the numbers shot increased steadily from the mid 1980s (when ca 100-300 birds were shot per annum) to ca 1,200-1,400 birds per annum in the late 1990s and early 2000s, with a figure of 1,606 reportedly shot in 2009 ( Appendix G, Section 16.7.5).

Population modelling (native population)

The modelling carried out for this species has not used PVA analyses per se but rather has predicted future trends assuming various scenarios for the two sub populations of Coll and Tiree, and the Uists (Trinder et al. 2009). Two types of population management were considered: shooting and the removal of eggs. In addition, models were run to consider two levels of population control: (i) population control imposed regardless of population size and; (ii) population control only imposed when the population exceeded a set threshold.

When population control was imposed with no reference to a lower threshold, the baseline models for both Tiree and Coll and the Uists predicted that the populations will continue to increase (see Table 2.10). In addition, the predicted population sizes were very sensitive to any form of population management, either causing the population to decline towards zero or display exponential growth. For example, the predicted median population for the Uists ranged from less than 1,000 individuals to over 10,000 by altering the annual shooting bag by only 10%. All models predicted that shooting was more efficient as a means of population regulation than the removal of broods. In general, shooting pressure has been shown to be higher on both Tiree and Coll and the Uists in more recent years, and if these raised shooting levels are sustained over the next 25 years then the populations are predicted to decline dramatically or even approach extinction (see Table 2.10).

If population control was only implemented above threshold population sizes, the predicted population sizes were far less sensitive to any form of population control. Consequently, it was recommended that in order to carry out effective management, such management should be varied according to the population size and growth rate. This would be consistent with an adaptive management approach to manage populations in future.

Table 2.10 Predicted population sizes and risks of falling below population of 1000. Figures for population control above a lower threshold were not given and cannot be presented for comparison. Predictions given are limited to those provided in the text of Trinder et al. 2009.

Model	Predicted population in 25 years	Risk of falling below 1000	Risk of falling below 2000
Tiree/Coll
Baseline Conditions 1998-2007	7,300	0.5%	-
Constant population control 15% population shot p.a. ˜ mean 1998-2007	<3,000*	-	-
Constant population control 31% shot p.a. ˜ mean 2005-2008	0**	100%	-
Uists
Baseline Conditions 1986-2003	15,700	-	0.1%
Constant population control 20% population shot p.a. ˜ mean 1986-2003	<6,500*	-	-
Constant population control 26.5% population shot pa. ˜ mean 2000, 2003, 2006, 2007	616	-	99%***

* population was stated to fall below current population
** predictions indicated that population could fall to zero around 2020.
*** this was the value cited for the risk of falling below 1800

The modelling for this species has assumed a lack of density dependence, whereby the shooting mortality is additive to natural mortality (increased shooting rates are assumed not to be compensated for a by a decrease in natural mortality). If there was compensation in natural mortality however, then the predicted declines in populations would not be as pronounced. The modelling has been limited by a lack of demographic data for large areas of the north and west of Scotland where the native Greylag Goose occurs, and modelling has not considered the naturalized population elsewhere in Scotland. There is also some evidence that colour-marked individuals have moved between the Uists and Coll and Tiree but information on movements is still relatively scant, particularly for other regions. It is not known how changes in intensification of shooting effort would affect emigration rates.

Figure 2.6 Population counts of Native Greylag Geese within two local goose management scheme areas: Uists = August counts (see Section 16.7.2); and Coll and Tiree = peak counts (see Section 16.6.3). Islay data are from the 2010 GSAG report via Simon Cohen, SNH; and Coll/Tiree data from Ben Jones and John Bowler, RSPB.

Changes since policy inception in 2000

• Populations of native and naturalised Greylag Geese are now thought to overlap but the true extent of this overlap is unknown. Moreover, the populations of resident native or naturalised geese are now known to overlap with the wintering population of Icelandic greylag, which presents a problem when trying to estimate their true numbers.
• The population estimates available for the native and naturalized population (north and west of the Great Glen Fault) indicate an increase in this population of >300% to ca 34,500 birds between1997 and 2008/2009 (Table 2.5). The naturalized population (south and east of the Great Glen Fault) increased by more than 450% over the same period to ca 13,000 birds.
• Numbers of native birds in areas with summer LGMSs have apparently either stabilised or started to decline in very recent years (Figure 2.6), which has raised some concerns over levels of shooting (see Appendix G).
• Numbers of naturalized birds have risen from 110 (2000) to ca 10,000 (2008/2009) on Orkney and increases have also occurred on Islay (to ca 1800 birds in 2009; Table 2.5).

Changes since the 2005 review and recommendations

• It is becoming increasingly difficult to differentiate between the native and naturalised populations because their ranges overlap, and the two 'populations' are not believed to be genetically distinct (although the empirical evidence for this is still scant). We concur with SNH and Mitchell et al. (2010) that the two 'populations' should now be treated as one (i.e. "native breeding").
• If the native and naturalized populations are to be treated as a single population in future (as we recommend, after Mitchell et al. 2010), and given the continued substantial increases in the numbers of the Native Greylag Goose, we recommend that the special protected status under the Wildlife and Countryside Act (additional penalty measures during the closed season; Schedule 1 Part II) of this species should be reviewed. This recommendation is based on the recognition that the range and numbers of native Greylag geese have increased markedly since 1981, when this legislation was introduced (although there is little evidence that it has ever had to be enforced).
• More regular surveys (during the breeding season) are now vital for monitoring changes in population size, particularly in areas where native breeding Greylags overlap with wintering populations of Icelandic Greylag Geese. For example, the size of the breeding population and nature of the terrain on Orkney make it very difficult to count comprehenisively on Orkney each year with the resources currently available (Eric Meek, RSPB pers. comm.).
• Demographic data collected from a range of sites would allow improved modelling of future trends and provide inputs into a PVA for the Scottish "native breeding" population (which should include the naturalised population in order to predict future population changes and thus the potential scale of future problems).
• The complete elimination of this species from some islands (particularly the west coast islands of Scotland; the preferred outcome for some stakeholders) would be at odds with UK obligations to maintain the range of this naturally occurring species.

2.5.6 Icelandic Pink-footed Goose

Status and long-term change

The Icelandic/Greenland Pink-footed goose winters almost entirely in Britain but small numbers are also now known to winter on the Faroe Islands, in Norway and in Ireland (Mitchell 1999). Within Scotland, these geese are concentrated in the east and south of the country.

The long term trend for the Icelandic/Greenland Pink-footed goose population is a steady increase since the 1960s (Figure 2.7), when ca 50,000 birds were estimated to winter in the UK. A notable rise in numbers occurred from the mid 1980s, and the most recent published UK population estimate (for 2008/09) was 351,118 birds, of which 248,153 were located in Scotland (Mitchell 2009).

Figure 2.7 Counts of the biogeographical population of Icelandic Pink-footed Goose from the International Census (Data provided by Carl Mitchell, WWT from WWT/ GSMP, see Appendix E).Note that derivaion of a true estimate of the 'Scottish wintering population' is problematic due to the high turnover of birds that move through Scotland during the winter. At present the counts in October are used to derive population estimates. WWT have suggested that counts carried out in January, outwith the autumn and spring movement, would be more appropriate to derive a Scottish specific estimate but there are very few years of data available to date.

Potential threats and hunting pressure

There have been concerns raised that Pink-footed Goose breeding habitat could be impacted by possible hydropower developments in Iceland. The extent to which this may have had an impact is likely to be minimal, as indicated by the last PVA analyses which suggested that a rather large proportion of the breeding grounds (>25%) would have to be impacted before a notable population declined would be observed (Trinder et al. 2005).

Hunting at the current levels appears to have little impact on the population of Pink- footed Goose. As there is no formal bag reporting system in the UK, the exact numbers of Pink-footed Geese that are shot each year are unknown. It has been estimated that ca 25,000 geese are shot each year in the UK (Hart and Harradine 2003; Frederiksen 2002). Based on mandatory bag data from 1995-2002, it is known that over 13,000 Pink-footed Geese are also shot annually in Iceland (Trinder et al. 2005). The shift of the UK wintering population of Icelandic Greylag goose further north and onto Orkney may mean that Pink-footed Geese become subject to higher shooting pressure within the UK in future (e.g. GSAG 2009).

Population modelling

The most recent PVA analyses were carried out on data collected from 1950 to 2002 (the UK population was estimated at ca 210,000 in 2002) and results were reported in the last review in 2005 (Trinder et al. 2005). The current estimate of the UK population (based on data from 2008/2009) is over 350,000 birds, which is already much higher than the predicted population estimate of 220,000 in around 2028 (based on the baseline model- see Table 2.11). The modelling also explored a range of scenarios for altering shooting pressures (Table 2.11).

Table 2.11 Predicted population sizes and quasi extinction probabilities (for the thresholds 10,000,20,000,30,000) for Icelandic Pink-footed geese for baseline models and additional mortality models. Predictions given are limited to those provided in the text of Trinder et al. 2005.

Model	Predicted population in 25 years	Quasi extinction probability
		Threshold 10,000	Threshold 25,000	Threshold 50,000
Reduced mortality (5000 p.a.)	277,335	0	0	<0.01%
Baseline (13,000 shot in Iceland and 25,000 in UK p.a.)	220,000	0%	<0.01%	2%
Extra mortality (5000 p.a.)	180,000	3%	5%	11%
Extra mortality (10000 p.a.)	109,000	24%	27%	34%
Extra mortality (15000 p.a.)	47,000	57%	60%	66%

Given the amount of time that has elapsed since the last PVA, and the current population estimates (which far exceed any of the predicted population estimates for 2028), the outputs have limited use in terms of setting possible management guidance for the next five years. Accurate predictions of population change for this quarry species will be dependent on high quality shooting bag information (Trinder et al. 2005).

Changes since policy inception in 2000

• The population of Iceland/Greenland Pink-footed Geese wintering in Scotland increased by 11% between 2000/2001 and 2008/2009, in the context of an increase of 43% in the biogeographica population over the same period.

Changes since the 2005 review and recommendations

• There is no evidence to support a change in conservation status as the population continues to increase (Figure 2.7); the species is already on the quarry list.
• Although the current levels of shooting appear to be sustainable, it is important to obtain improved UK hunting bag estimates for the Icelandic Pink-footed Goose population.
• Continued international collaboration on shooting mortality data is also required.
• There is a need to carry out updated PVA analyses, particularly as levels of shooting pressure within the UKmay increase for this species with the shift of the wintering range of Icelandic Greylag Geese northwards and away from mainland Scotland. However, we suggest that PVA model predictions are bound to remain uncertain until improved information on shooting take (hunting bag information) is available.

2.5.7 Taiga Bean Goose

Status and long-term change

There are five races of the Bean Goose, only two of which occur in Britain. Only one, the Taiga Bean goose, is dealt with as part of the current review (the other, the Tundra Bean Goose, occurs mainly in eastern England).

The main wintering grounds of the Taiga Bean Goose are in southern Sweden, Denmark and, to a lesser degree, in northern Germany and Poland (Madsen et al. 1999). Relatively few Taiga Bean Geese winter in the UK and they are restricted to two sites: the Slamannan Plateau, central Scotland; and the Yare Valley in Norfolk. Hence they are classified as a scarce winter visitor to the UK.

Numbers of Taiga Bean Geese on the Slamannan Plateau were recorded at ca 100 birds in the winter of 1985/1986. Since annual monitoring began, numbers have risen consistently until the mid 2000s, when numbers appeared to stabilise between 250 and 300 birds. The most recently published count was 265 in 2008/2009 (based on peak counts, Maciver 2009). Taiga Bean Goose numbers at the Yare Valley (England) in the 1960s were less than 50 birds but by the mid 1980s and early 1990s numbers had reached a peak of 400-500 individuals. Subsequently numbers have fallen to just over 100 birds in 2008/2009 (Mitchell 2010).

Potential threats and hunting pressure

The Scottish wintering population of Taiga Bean Geese has been relatively stable. The main risk is a catastrophic event, which could have a significant impact as the population is restricted to a localised area. Factors identified as risks to the Scottish population include disturbance and changes in land use (e.g. forestry, wind farms and opencast mining are all mentioned in the LBAPSAP for this population).

Population modelling

None available

Changes since policy inception in 2000

• Between 2000/2001 and 2008/2009 numbers in the Scottish wintering area (Slammanan Plateau) increased by 47% (180 to 265 individuals; Table 2.5).

Changes since the 2005 review and recommendations

• There is no evidence to suggest a change in the conservation status of the Scottish population.
• Due to the status of Taiga Bean Goose as a scarce winter visitor to the UK, future goose policy should continue to support positive conservation management of the Scottish flock centred on the Slamannan Plateau.

2.5.8 Brent Geese

East Atlantic Light-bellied Brent Goose

The main wintering grounds for the East Atlantic Light-bellied Brent Goose are in Denmark and on Lindisfarne, northern England but very small numbers occur regularly on the Inner Moray Firth, Inverness and the Eden Estuary in Fife (Mitchell et al. submitted). The UK population of the East Atlantic light-bellied Brent Goose has increased from 1959/1960, when less than 1000 birds were recorded, to reach a peak of just under 5000 birds in 2001/2002. The most recent estimate of the population was just under 3000 birds in 2008/2009. With respect to birds wintering in Scotland, the five year mean for the numbers wintering on the Inner Moray/Firth of Inverness and the Eden Estuary have been calculated as 47 and 28 respectively (Holt et al. 2009). There have been no notable changes in this population since the 2005 review.

Eastern Canadian Light-bellied Brent Goose

The main wintering grounds for the Eastern Canadian Light-bellied Brent Goose are almost wholly in Ireland, with smaller numbers reaching the Channel Islands and western France (Merne et al., 1999). In Scotland small numbers occur on passage in late winter and spring in the Hebrides and Dumfries and Galloway.

The UK and Irish population of the Eastern Canadian Light-bellied Brent Goose has increased from 11,900 in 1960/61, when the first complete census was undertaken to over 19,000 in the winter of 1999/2000 (Robinson, et al. 2004). However, since then numbers have increased to 35,000-40,000 in 2007 (Colhoun 2008). No Scottish sites hold internationally important numbers but sites may be used as temporary refuges in response to weather conditions (Robinson, et al. 2004). During the early 1990s, small numbers began to winter in Dumfries and Galloway, and since 2000, 20-50 (and up to ca 150) individuals have been counted regularly (overwintering) at Loch Ryan or Luce Bay, Dumfries and Galloway (Mitchell & Ogilvie 2007; Ian Bainbridge pers. Comm.). Overwintering birds may move between Strangford Lough, Northern Ireland and Dumfires and Galloway (there is limited ringing evidence for this; Mitchell & Ogilvie 2007). The five year mean for the numbers wintering on Loch Ryan (Dumfries and Galloway), Broadford Bay (Skye) and Loch Gruinart (Islay) are 52, 41 and 73 respectively (Calbrade et al. 2010). Recent winter counts at Loch Gruinart have been much lower and 20-25 birds are now recorded wintering at Loch Indaal (Islay). Autumn passage through Islay has increased from 9000 in 1960's to 20,000 at present and these are mostly recorded at Loch Gruinart (ap Rheinhalt et al. 2007). Small passage flocks are also reported on the Outer Hebrides from Harris to Barra.

2.5.9 Naturalised Canada Goose

Status and long-term change

This is an introduced species, which has naturalised extensively throughout the UK but notably within England. From the first population estimate of 3,906 Canada Geese in 1953 there has been a steady increase in numbers in the UK. By the 1991 survey, the population estimate had reached a peak of 63,581 birds, accompanied by large increases in range (Delany 1992). The most recent survey in 2000 estimated a Canada Goose population of 54,587 birds (14% decline relative to the 1991 survey). Most of the Canada Goose population is located in England and population estimates for Scotland are limited to two censuses only. In 1991, the Scottish population of Canada Geese was 1088 (Delany 1992) and by 2000 it had risen to 1,244 with the highest numbers found on the Beauly Firth and Cromarty Firth (Rowell et al. 2004). There has not been a more recent population estimate, but from anecdotal records and annual monitoring it is likely there has been a further increase in the Scottish population (Figure 2.8), and some stakeholders mentioned that flocks have been noted in a number of areas.

Figure 2.8 Annual trends (indices) in wintering (resident) Canada Geese numbers in Great Britain, Scotland, Wales, and England (Data provided by Neil Calbrade, BTO based on counts from the BTO/ RSPB/ JNCCWeBS in association with WWT; see Appendix E).Indexing techniques are used to track relative changes in numbers from incomplete data (when sites have not been visited an expected count is used based on the previous pattern of counts; only sites with over 50% coverage are used). Annual indices are expressed relative to the most recent year for data are available, which takes an arbitary value of 100. The trends presented here have been smoothed using generalized additive modelling.

Population modelling

None available

Changes since policy inception in 2000

• According to anecdotal records and annual monitoring of wintering numbers, the Canada Goose appears to be increasing in Scotland (Figure 2.8).
• During the last national ( UK) survey in 2000 the main concentrations of birds in Scotland were found on the Beauly and Cromarty Firths. Previously, these Canada Geese largely migrated from their breeding grounds in central England to moult in Scotland, and returned following moulting (Wernham et al. 2002).
• However, whether birds from these moulting flocks are staying in Scotland to breed, or others are moving into Scotland, provisional data collected as part of the BTO/ BWI/ SOC Bird Atlas 2007-11 project suggest large increases in the range of breeding Canada Geese in Scotland since the previous atlas survey in 1988-91, particularly in the central and mid-west of Scotland (D. Balmer BTO pers. comm.).
• The Wetland Bird Survey ( WeBS) index for the winter population of species shows a sustained increase of ca 40% between 2000/2001 and 2008/2009 (N. Calbrade BTO pers. comm.; Figure 2.8).

Changes since the 2005 review and recommendations

• Given the continued increase in the Scottish population and risks posed by it, future goose policy in Scotland should consider carefully and immediately whether Canada Goose should be added to Schedule 2 Part II of the 1981 Wildlife and Countryside Act (as passed in England in 2010). We recommend that at the very least the effects of adding Canada Goose to Schedule 2 Part II in England are reviewed in five years time to assess whether the legislative change has had any discernable effect on population levels in England (see also Section 2.7.6).

2.5.10 Naturalised Snow Goose

Status and long-term change

In 1991, a total of 140 Snow geese was found in the UK, of which only four were found in Scotland, on the island of Mull (Delany, 1993). In 2000, 86 Snow geese were located, of which less than five geese were in Scotland (Rowell et al. 2004). A small feral population is present on north-west Mull and Coll, with a high count of 57 on Mull in 1981. All breeding now occurs on Coll, where 24 adults and 10 young were recorded in 2002 (ap Rheinallt et al. 2007).

Changes since the 2005 review and recommendations

None

2.5.11 Naturalised Bar-headed Goose

Status and long-term change

A total of 85 Bar-headed geese was recorded in 1991 throughout the UK, with 11 records from Scotland (Castle Loch, Dumfries and Galloway, South Ronaldsay, Loch Tummel, Tayside and Tyninghame, Lothian; Delaney 1993). In 2000, a total of 83 Bar-headed geese was reported, of which six were found in Scotland at a single site (Castle Loch, Dumfries and Galloway) and two at a unnamed site in Tayside (Rowell et al. 2004). A small flock possibly exists near Oban, Argyll with 15 reported in 2002, 5 in 2004 and 4 in 2006 (ap Rheinallt et al. 2007).

Changes since the 2005 review and recommendations

None

2.6 Has the Scottish National Policy Framework succeeded in meeting conservation obligations for geese?

The National Policy Framework ( NPF) for goose management in Scotland came into force in 2000. Across the combined total population of geese in Scotland (wintering and resident) there has been a ca 17% increase in numbers since 2000, and a number of species have shown much larger increases (Table 2.5). This broad evidence suggests that policy has, in general terms, served Scottish geese well in terms of the first principle objective of the NPF to "Meet the UK's nature conservation obligations, within the context of wider biodiversity objectives" ⁸. The majority of stakeholders concur with this broad view (see national questionnaire responses on SG website). It is not possible to provide a clear picture of what the counterfactual would have been in terms of delivering conservation obligations however (i.e. what would have happened since 2000 in the complete absence of any goose policy over and above legal mechanisms). It is not possible to predict with any certainty to what extent the range of stakeholders with goose interests (conservation organisations, farmers, hunters) would have taken independent actions to rectify the problems that were perceived, and how this balance of actions would have affected the overall conservation status of the goose populations involved.

There are two exceptions to the general trend of increasing goose numbers in Scotland since the inception of current policy in 2000. The Greenland White-fronted Goose population has shown a steady decrease in the Scottish population of 34% since policy inception and its population was given elevated AEWA (conservation concern) status in 2009 (Section 2.5.3). This suggests that Scottish goose policy is not meeting its conservation obligations for this species, a view shared by a number of stakeholders including SNH⁹. The extent to which this population is limited by factors operating in the wintering area (Scotland) is unclear however. There was evidence for declines prior to 2000, and there is substantial evidence to suggest that factors outside of Scotland may be contributing to the decrease in this population (Section 2.5.3). GSAG has kept abreast of the research and evidence for this species, and supported the GWGS in development of a project to take forward improved management of small flocks in Scotland. We conclude, however, that the NPF could have done more to prioritise this species within policy, particularly within the Islay Local Goose Management Scheme (Section 2.5.3).

The Icelandic Greylag Goose population wintering in Scotland is at similar numbers in 2008/2009 to those found in 2000/2001 but a major shift in its wintering area has occurred (to Orkney; see Section 2.5.4). We conclude that the NPF has met conservation obligations for this species since 2000 but could have done more in terms of reacting to the changes in distribution, for example by better supporting the collection of appropriate evidence (e.g. count data) from Orkney to allow enhanced monitoring of this and the breeding season (resident) Greylag population (which has been monitored by RSPB and volunteers on Orkney for many years but is becoming increasingly difficult to count accurately because of rising numbers and a wider distribution on Orkney). This shortfall is a reflection of an overall weakness of the NPF (and NGMRG) in the extent to which it has been proactive in tackling emerging goose issues throughout Scotland (particularly those involving Greylag Geese) during the period since its inception ( Chapter 3).

The derogation to shoot Greenland Barnacle Geese on Islay has not been accepted by all stakeholders in Scotland as contributing appropriately to goose policy in terms of meeting the UK's nature conservation obligations (Section 2.5.2). We conclude that further evidence is emerging that supports the need to review the situation on Islay fully with respect to scaring, in terms of impacts on both Greenland Barnacle and Greenland White-fronted Geese (Sections 2.5.2 and 2.5.3).

The fact that the UKSPA network is incomplete for geese because it fails to protect goose feeding areas was highlighted by at least one stakeholder during the current review. This is now being addressed at a UK scale through a JNCC review (Section 2.1.1). The current NPF has offered some facility to enhance protection of feeding areas around SPAs where exceptional circumstances have prevailed (e.g. at Loch of Strathbeg through the LGMS there). The scaring and lethal shooting of Barnacle Geese close to SPAs on Islay could be viewed as contrary to meeting conservation obligations if the disturbance affects the feeding areas for geese roosting on the neighbouring SPAs. The JNCC decision tool for designating feeding areas associated with SPAs will be rather critical in this respect for clarifying the situation on Islay and more generally for goose policy in terms of defining legitimate management actions in future.

We conclude overall then that the Scottish Goose NPF has been only partially successful in meeting the UK's conservation obligations for geese.

2.7 Gaps in knowledge and suggested actions

The populations of most Scottish geese are now stable or increasing. Thus it is important that future policy development evaluates carefully the balance required between the first two fundamental objectives: on the one hand appropriate protection (to meet conservation obligations); and on the other hand active management (to mitigate economic loss). We later suggest that the Scottish Government takes a species-focussed approach to future policy (Section 7.4) because the balance of priorities between the two basic objectives of the current National Policy Framework varies between goose species. In Table 2.12 we set out recommendations for the overall level of priority that each of the main Scottish goose species should receive within future policy based on: (i) the current risk of Scotland defaulting on its conservation obligations and agreements for that species; and (ii) the likely future costs posed by each species to society (in terms of the likelihood of increases in conflicts with agriculture and thus increasing economic costs).

The current economic climate, and generally increasing number of geese in Scotland, also means that the third fundamental objective of policy (ensuring value for money) is increasing in importance. We now highlight the main gaps in knowledge that we have identified as limiting the Scottish Government's ability to make effective goose policy decisions currently, and attempt to prioritise these according to their importance and hence relative value for money. The following sections should be read keeping the above considerations in mind.

Table 2.12 Recommendations for the categorization of the main Scottish goose species according to the degree to which they should receive priority under future Scottish goose management policy. Additional information in support of these recommendations is given in Section 2.5 for each species.

Population	Conservation obligations ( NPF Objective 1)		Mitigation against economic loss ( NPF Objective 2)		Overall recommended priority within future goose management policy NOTE 1
	Risk of defaulting on conservation obligations	Justification	Risk of increased future costs to society	Justification
Taiga Bean Goose	High	Risk of loss of range in UK	Low	Conflicts with agriculture are very localized and population is small	High
Greenland White-fronted Goose	High	Species is of highest conservation priority of all the Scottish goose species and population is declining	Low	Population is currently declining	High
Icelandic Pink-footed Goose	Low	Population is still increasing markedly	Medium	There are known to be some localized conflicts with agriculture outside current Strathbeg LGMS	Low
Icelandic Greylag Goose	Medium	Scottish population has stabilized and is becoming increasingly concentrated on Orkney increasing its vulnerability	Medium	Increased concentration on Orkney likely to increase conflicts with agriculture	Medium
Resident Greylag Goose	Low	Overall population is expanding except in areas with current population management	High	Expanding population and potential costs to other biodiversity as well as agriculture	High
Greenland Barnacle Goose	Medium	Decrease in Islay population is unexplained at present and restricted wintering area renders the population vulnerable	Medium	Population still increasing (with the exception of Islay)	Medium
Svalbard Barnacle Goose	Medium	Restricted wintering area renders the population vulnerable	Medium	Population still increasing	Medium
Canada Goose	None	Introduced species	Medium	Population likely to be increasing, with risk of future conflicts with agriculture and other biodiversity interests	Low

NOTE 1 - OVERALL PRIORITY

HIGH: The species scores "high" priority either in terms of the risk of defaulting on conservation obligations or the risk of increased costs to Scottish society

MEDIUM: The species scores "medium" priority in terms of both the risk of defaulting on conservation obligations and the risk of increased costs to Scottish society

LOW: The species scores one "medium" score for either the risk of defaulting on conservation obligations or the risk of increased costs to Scottish society, or lower priority scores for both criteria

2.7.1 Predicting future population sizes and adaptive management of populations

The current project was given in a clear remit, which stated that the current National Policy Framework "seeks to facilitate adoption of adaptive management techniques as a means to deliver social and economic sustainability of land management businesses in areas frequented by significant goose populations, while also ensuring compliance with legal obligations under the EU Birds Directive".

In the broadest sense, the concept of adaptive management involves the continual evolution (improvement) of management practices through time in response to periodic evaluation of the success of the management in delivering its agreed objectives. Also in the broadest sense the NPF (with NGMRG at its core; Chapter 3), with its regular stipulated review periods, has been designed to deliver this adaptive management role.

In a stricter conservation management sense, the term adaptive (harvest) management has been used to describe a process by which populations of a given species can be utilized for human needs (e.g. hunted) in a sustainable manner by ensuring that (Nichols et al. 2007):

• The objectives of management are clear (e.g. targets for acceptable population size are set according to ecological, socio-economic and cultural needs);
• A series of alternative management actions (scenarios) are agreed;
• A series of models are used to predict responses of the population to any off-take or harvest (e.g. by modelling population change and explicitly incorporating uncertainty); and
• A fit for purpose monitoring programme is in place to allow models to be benchmarked against observed population changes on a regular (often annual) basis, and management (e.g. off-take) is adjusted accordingly (in the light of lessons learned) to maintain a sustainable population.

Adaptive harvest management has been used by the US Fish and Wildlife Service as a means of sustainably allowing the hunting of ducks since the mid 1990s (Nichols et al. 1995; Williams and Johnson 1995; Johnson 2001; US Fish and Wildlife Service 2010).

More recently, the approach has been extended to goose management abroad (e.g. by the Canadian Wildlife Service for Greater Snow Goose; by the US Fish and Wildlife Service for Greater and Lesser Snow Geese and Ross's Goose; e.g. Bélanger and Lefebvre 2006; U.S. Fish and Wildlife Service 2007). The approach in its entirety has not received such wide use in Europe, although its use for the management of European duck populations has been recommended (Elmberg et al. 2006). There is now progress towards the development of an international adaptive management plan for the Svalbard Pink-footed Goose (under AEWA, through a trilateral agreement between the Netherlands, Germany and Denmark; J. Madsen, pers. comm.).

Nichols et al. (2007) provide a concise summary of the adaptive harvest management framework adopted in the US, its strengths, and lessons learned since its inception in the mid-1990s. They also consider and discuss some of the reasons put forward previously for not adopting a similar approach in Europe, and conclude that the arguments in favour of adaptive management are still compelling. We agree that there is much that could be taken from the adaptive harvest management approaches that have been used in North America to facilitiate better management of geese in Scotland but the resources available for the current review precluded a detailed review of systems in there.

We suggest that SG consider in more detail the US and Canadian approaches to adaptive management, and any lessons that have been learnt from their experiences, during future policy development.

2.7.2 Current state of modelling of Scottish goose populations

The current Scottish goose NPF (through the advice of NGMRG and GSAG) has already established a firm basis to predicting changes in Scottish goose populations (an appropriate part of any adaptive management process; Section 2.7.1) through recently commissioned and updated PVA and similar analyses for the key Scottish species (Section 2.5).

The largest omission from the modelling that has been carried out to date (Section 2.5) is a consistent, comprehensive and transparent treatment of the high degree of uncertainty in the modelling predictions, and thus the extent to which appropriate management decisions (e.g. to increase numbers shot) can be taken with an acceptable level of risk in terms of conservation obligations. In many cases the confidence limits surrounding predictions are very wide and, whilst these are often presented graphically, the level of uncertainty is not adequately reflected when the information is discussed and summarized in the reports (which would make it more fit for purpose in terms of adaptively managing populations). Similarly, future predictive modelling to inform national goose policy would benefit from clearer management scenarios at the outset (as is the requirement for effective adaptive management).

For a number of the Scottish goose populations for which modelling has been undertaken to date (Section 2.5), it is clear that population trajectories are particularly sensitive to adult survival rates (and hence also numbers shot) but, once again, the relative sensitivities of the population predictions to variation in the different model input parameters (e.g. productivity, annual survival, numbers shot) have not always been examined clearly and comprehensively. Whilst there are clear knowledge gaps in terms of some of these demographic and shooting parameters highlighted in the modelling reports, it is difficult to prioritise future data collection and make judgments about the relative value for money of specific monitoring enhancements (specifically in the applied context of goose management) when it is not clear to what extent these enhancements would benefit future model predictions for individual species.

For a number of the Scottish goose populations for which modelling has been undertaken to date, it is already apparent that observed population changes following modelling do not reflect the predicted population trajectories. At worst, in two cases (Greenland White-fronted Goose and Greenland Barnacle Goose), the models predicted population growth in contrast to observed population declines subsequent to the modelling. In such cases, it is important that a careful investigation is undertaken to establish the mostly likely cause of the mismatch and allow incorporation of this extra uncertainty when informing future management decisions.

We suggest that before any further PVA/population modelling work is commissioned, a clear standard framework is set up, which stipulates: (i) clear population/management objectives or scenarios; (ii) explicit and comprehensive treatment and reporting of uncertainty/risk; and (iii) clear and comprehensive assessment of the sensitivity of model predictions to variation in all input parameters. This would ensure that each new piece of modelling, in following the framework guidance, and will include critical assessment of model performance and thus render the modelling of greater utility for applied goose management purposes.

Table 2.13 Summary of priorities for continuing and enhancing collection of biological information on the main Scottish goose populations, based on a review by GSAG ( GSAG 2009) and updated to reflect the findings of the current review.

Population NOTE 1	Current monitoring	Priority for continuation of current monitoring	Additional monitoring recommendations (see Section 2.5 for each species for more information)	Priority for additional monitoring
Taiga Bean Goose	Monthly counts, annual population estimate	High	-	-
	Annual productivity	High	-	-
Greenland White-fronted Goose	Bi-annual international counts, annual population estimate	High	-	-
	Annual productivity estimates in core areas	High	Enhance coverage of productivity estimates to other parts of range	High
	Field/habitat use	Not monitored currently	Include spatial recording of habitat type in surveys whenever possible	High
	Field use in relation to scaring/interaction with Greenland Barnacle Goose	Not monitored currently	Enhance recording of field use in relation to scaring on Islay	High
Icelandic Pink-footed Goose	Bi-annual international counts, annual population estimate	High	-	-
	Annual productivity	High	-	-
Icelandic Greylag Goose	Annual international counts, annual population estimate	High	Enhanced monitoring of resident Greylag Geese on Orkney (as below) NOTE 3	High
	Annual productivity	High	-	-
Resident Greylag Goose	Annual counts of core areas	High	Expand core areas to include annual surveys on Orkney	High
	Periodic national census	High	-	-
	Annual productivity estimates in core areas	High	Expand core areas to include annual estimates on Orkney	High
Greenland Barnacle Goose	Annual counts of core areas	High	-	-
	Periodic national census	High	More regular national census (2- or 4-year cycle)	High
	Annual productivity estimates in core areas	High	-	-
	Field use in relation to scaring/interaction with Greenland White-fronted Goose	Not monitored currently	Enhance recording of field use in relation to scaring on Islay	High
Svalbard Barnacle Goose	Annual population estimate	High	-	-
	Annual productivity	High	-	-
Canada Goose (and other non-native species)	Annual index	High	-	-
	Periodic Census NOTE 2	Medium	-	-

NOTE 1

We have omitted the two Brent Goose populations from the table but concur with GSAG (2009) in that it is important that their numbers in Scotland continue to be monitored annually via WeBS.

NOTE 2

Consider including resident Greylag Geese outside of their core breeding range in the north and west of Scotland.

NOTE 3

This is also important on other areas where there is overlap in the wintering areas of Icelandic Greylag Geese and resident breeding Greylags (e.g. Shetland). It is seen as beneficial to coordinate censusing of Icelandic Greylags in Scotland with the census in Iceland.

2.7.3 Current quality of Scottish goose biological information to underpin population modelling and population management

Population estimates and productivity information

The 2005 review of national goose management policy recommended that GSAG agree a five-year plan for monitoring and research ( GSAG 2009). In general we agree with the recommendations of GSAG with respect to their priorities for the continuation of data collection on population size and productivity for the main Scottish goose populations (Table 2.13). Successful and cost-effective management of Scottish goose populations in future, and the careful balancing of the three fundamental objectives of the National Policy Framework, will depend on the modelling of management scenarios. Periodic population size assessments are critical in predicting future population trajectories and judging how well the predictive models perform (and the degree of risk involved in management decisions). In addition, many of the long-term surveys from which this information is obtained are carried out largely by volunteers (see Appendix E), and provide very good value for money. We would concur with the view of GSAG that because of this, those surveys that are carried out on an annual basis should continue to run annually, because otherwise volunteers may move away from goose monitoring in the period between surveys (for example if surveying were reduced from annual to biennial). Thus we generally score the continuation of population monitoring as high priority across all the main goose species (Table 2.13). Productivity information is often also collected by volunteers or at little extra cost. It is certainly cheaper to obtain information on productivity to inform population models than on survival rates (below), such that the availability of rigorous productivity information will serve to reduce some of the uncertainty in population prediction at minimal cost, and we place high priority on continuing to collect this data for most species (Table 2.13).

Information on survival and movements (from mark-recapture)

Mark-recapture data (from ringing and other forms of colour marking) provides important information on: (i) the annual survival (or mortality) rates of geese; and (ii) movements (the degree of interchange) of geese between different wintering areas. This type of information is essential for a number of purposes relating to goose management decisions and policy questions:

• Survival (mortality) estimates are important in population modelling, to predict future changes in populations and understand the effects of hunting (i.e. whether mortality from shooting is additive to other forms of (natural) mortality or whether compensation occurs);
• Information on immigration and emigration is important to allow survival estimates from resightings to be corrected for birds that have left the local population (otherwise local survival will be underestimated);
• Information on movements of geese is essential for ensuring that monitoring of numbers is carried out effectively and surveys are designed and improved appropriately if the distribution of the population changes through time.
• Information on movements of geese is essential for interpreting survey results if populations change (e.g. to assess whether decreases at particular sites are due to true change in breeding success or mortality, or due to a redistribution of the species in question).

Much ringing in the UK is carried out by volunteers, and there is an established network through which volunteers submit resightings of marked birds (see www.bto.org/ringing). Despite this established infrastructure and volunteer base, goose ringing is a specialist activity that requires considerable time commitment on the part of the volunteers and, at the very least, resources for professional coordination to ensure that allocation of ringing effort is as strategic as possible, and to encourage groups of volunteers to keep active (provided by WWT to date). The collection of resighting information also requires dedicated volunteers (a limited number of volunteers do the majority of this at present) and these volunteers need continual encouragement from a coordinating organization in order to maintain their interest and effort. Collection of ringing information currently represents good value for money because of the high volunteer effort involved but it is our belief that the value for money could be even greater if a small amount of additional support were available to the main coordinating organization ( WWT) to coordinate the ringing strategically.

Ringing information tends to be more costly to analyse than productivity information because in general it demands a more complex analytical process (rather than the relatively simple summarisation of data collected in the field for productivity estimates). The 2005 review of the NPF (Scottish Executive 2005) concluded that ring-resighting information was most in need of improvement for Native Greylag Goose and Greenland Barnacle Goose (at locations away from Islay).

The current review concludes that the lack of information on survival rates and/or movements between sites has limited the population modelling available to date for at least five of the main goose populations considered in this review: Greenland White-fronted Goose; Icelandic Greylag Goose; Native Greylag Goose; and Svalbard and Greenland Barnacle Geese (Section 2.5).

We suggest that following implementation of our suggestion to set out a standard framework to guide PVA/modelling analyses (Section 2.7.2), future PVAs should address explicitly the sensitivity of the modelling to the demographic input parameters (including survival rates). The results could then be used to further prioritise data collection by species more efficiently, both in terms of individual priority species and sample sizes required.

We suggest that ringing should be supported at least at current levels for Greenland White-fronted Goose, Icelandic Greylag Goose and Svalbard Barnacle Goose, and improved if possible for Native Greylag Goose and Greenland Barnacle Goose, until such time as PVAs/modelling can be revised to address sensitivity and uncertainty explicitly (as suggested above). In the event that funding is insufficient to support this, ringing of these species should be prioritized in accordance with the overall priority suggested in Table 2.12.

Information on numbers shot (quarry species)

The EU Birds Directive deems that "hunting constitutes acceptable exploitation where certain limits are established and respected" and, as such, that "hunting must be compatible with maintenance of the population of these species at a satisfactory level". Thus the UK has a legal obligation to ensure that any hunting (or take) is sustainable (Section 2.1). Information on numbers of birds hunted (termed hunting bag data) is important to allow the establishment of appropriate "limits" and ensure that hunting does not jeopardize the sustainability of populations. The population modelling carried out for Scottish goose species to date has shown that predicted population trajectories are most sensitive to survival rates for most populations (Section 2.5), and numbers of birds shot are necessary to allow density-dependent processes to be adequately reflected in the modelling (i.e. whether mortality from shooting is additive or compensatory). Improved information on numbers shot is therefore essential for managing the populations of any species that can be shot anywhere on their flyway.

To date, however, only estimates of bag data of quarry species have been available for use in the modelling (Section 2.5). Moreover, the notable shift in the wintering distribution of Icelandic Greylags (Section 2.5.4) has introduced more uncertainty over the UK bag estimates (because other grey goose species may experience increased shooting pressures in the absence of so many Icelandic Greylags south of Orkney). For the Icelandic Greylag Goose, shooting may be the most important factor determining the population numbers and hence status (Trinder 2010), and the level of shooting mortality will also have an important influence on the population trajectories of the other resident Greylag population and that of the Pink-footed Goose (Sections 2.5.5 and 2.5.6). A lack of accurate bag data has also meant that it is largely impossible to differentiate between natural and shooting mortality in the population models to date, which increases the uncertainty of model predictions.

We conclude that there is an urgent need to consider options for obtaining improved bag data within the UK. This was also a recommendation of the 2005 review ( Appendix A). The majority of stakeholders that contributed opinions to this review (those sitting on NGMRG and others) believe that hunting bag information is not currently adequate for goose management policy purposes (see national stakeholder questionnaire responses on the SG website). Bag data could be collected via voluntary or mandatory systems. We consider the strengths, weaknesses and constraints in implementing either approach in Table 2.14.

Table 2.14 Contrasting a voluntary versus mandatory system for collecting huntingbag data in Scotland: some strengths, weaknesses and constraints

Attribute	Voluntary scheme	Mandatory scheme
Geographical scale of operation I	Could be Scotland (or region) specific.	Would have to be UK (or Scotland) wide (due to governing legislation). A system in Scotland only would omit collecting data on Pink-footed Geese (many) and Icelandic Greylags (less) wintering south of the border.
Geographical scale of operation II	A Scotland- or region-specific scheme would not provide bag information from England for those species that also winter south of the border.	Would ensure data from the whole of Scotland (or preferably UK) were available.
Quality of data	Would need to be bench-marked initially and periodically against whole hunter population to minimize bias and ensure sample sizes sufficient to generate data of sufficient precision for population modelling.	Would be comprehensive (subject to a small number of non-returns perhaps) or could be carried out on a rigorous sample basis to reduce annual costs.
Inclusion of information from non-Scotland shooters hunting in Scotland	Foreign hunters not included explicitly in voluntary schemes to date (some returns from foreign hunters probably included in NGC but probably few if any in WSS.	Could ensure (via legislation) that all shooters hunting in the UK (or Scotland) were obliged to make bag returns.
Set-up requirements	NGC already runs but WSS has lapsed since 2002. Would need (i) bench-marking study; (ii) power analyses study; and (iii) set-up costs for BASC and costs to enhance survey after (i) and (ii) carried out.	Change to primary legislation to introduce a permit system with renewals on an annual basis. Expectation of proficiency testing seen as a possible barrier (below).
Experiences from abroad ( EU) (see Appendix F)	Few voluntary schemes abroad (Sweden; Bulgaria; France)	Many countries have mandatory schemes (Iceland and Denmark may beparticularly worthy of further investigation). Many countries have proficiency testing as an integral part of hunting legislation.
Species coverage	Sample survey unlikely to produce adequate samples (precision) for all quarry species.	Comprehensive mandatory system would ensure coverage of all quarry species (not just geese), giving added VFM and fulfiling wider obligations under EU legislation.

Following the 2005 review of goose management policy in Scotland (Scottish Executive 2005), a study was commissioned (Aebischer and Harradine 2007) to explore the potential for developing two voluntary hunting bag schemes - the National Gamebag Census ( NGC) and the Waterfowl Shooting Survey ( WSS), as coordinated by GWCT and BASC respectively - into a more comprehensive bag data scheme for waders, gamebirds, ducks and geese. For geese, using existing data from these two voluntary recording schemes, it was possible to derive species-specific annual bag indices for Pink-footed and Greylag Goose at both the UK and Scottish level, and for Canada Goose at the UK and English level only. We concur with the key recommendations of Aebischer and Harradine (2007) that in order to develop an effective voluntary bag scheme for the future it would be important to:

• Carry out an independent review of the extent to which the NGC and WSS were representative of the wider community of hunters (to ensure that trends from the data were also representative; and essential if absolute numbers shot are to be derived from such a scheme); and
• Carry out power analyses to determine the minimum samples sizes required for each of the species to be monitored by any future scheme, and to guide the likely precision of resulting estimates of overall numbers shot from such a scheme.

One potential benefit of a voluntary scheme would be the flexibility to operate it at a geographical scale smaller than the whole of the UK. It would be possible to implement a scheme at a country level for species that occur almost exclusively in Scotland (such as the resident Greylag Goose and now also the Icelandic Greylag Goose) or at a regional level (e.g. for species which tend to concentrate at key sites such as Icelandic Greylag geese on Orkney). Drawbacks with a voluntary scheme include the fact that BASC stopped the collation of data under the WSS in 2002, and there would therefore be new start-up costs if a future voluntary scheme involved WSS (which would be required for geese). A voluntary approach, although perhaps adequate to provide indices (patterns of change) in numbers shot, might be insufficient to provide absolute numbers shot with any degree of precision. Also, without a regular bench-marking exercise biases could occur through time that would render the sample non-representative of the population of hunters as a whole (leading to biased estimates of numbers shot).

Mandatory schemes for the reporting of hunting bag data are already in operation in a number of other European countries (see Appendix F: Table F3) and there is clearly scope for adopting similar approaches in the UK. The need to make fundamental changes to primary hunting legislation within the UK has been reviewed as a barrier to putting such a mandatory system in place to date. Most EU countries have annual systems for hunting permit renewal, however, and a number of them have bag returns linked to this (see Appendix F: Table F3). There is also a precedent set in a number (the majority) of other EU countries, where hunters must pass proficiency exams as a prerequisite to a shooting licence being granted. In order to implement an effective mandatory system, hunting permits would need to be renewed on an annual basis and the submission of bag data made a condition of renewal. In the UK, shotgun/firearm certificates are currently issued every five years with no such stipulation imposed on the licencee.

Iceland and Denmark were noted by stakeholders as having efficient and effective hunting bag collection systems. Iceland has operated and mandatory system since 1995. Hunting 'cards' (licences) are renewed on an annual basis and renewal is conditional on: (i) bag returns from the previous year being submitted; and (ii) payment of an administrative fee that covers the cost of the permit and bag return system (A. Peterson, pers. comm.). Over 90% of Icelandic hunters are thought to comply with these conditions (B. Palsson, pers. comm.). The high degree of compliance may be in part due to the way that the system operates: although the new permit application and bag return are submitted at the same time, the latter is anonymous in that it is separated immediately from the former upon receipt. We have been led to believe that this system has led to Icelandic hunters being very honest about numbers and species of geese shot (including accidental illegal shooting of non-quarry species; Sigfússon et al. 2003). The danish system is less comprehensive (based on wing returns and returns are believed to be obtained from 71% of Danish hunters and 54.5% of foreigners shooting in Denmark (for 2009/2010; J. Madsen, pers. comm.).

Reporting from international hunters shooting within the UK (Scotland) needs to be given a sufficiently high priority in any bag scheme that is introduced, as they may be responsible for shooting a substantial proportion of the total bag. Options for how this could be achieved should be considered carefully, including implementation via individual hunters versus via goose guides.

We conclude that the implementation of a rigorous system for collecting hunting bag information for quarry species in Scotland is essential if future policy is to manage goose populations effectively, and a successful balance between the three principal objectives of the NPF be achieved. We believe that without satisfactory population models (with a manageable level of uncertainty within population trajectories under various management scenarios), it will not be possible to manage the growing goose population without an unacceptable level of risk.

We suggest that further steps towards the implementation of a rigorous hunting bag reporting system in Scotland are taken without further delay. We recommend that further detailed information and critique on operation is obtained from countries that already have voluntary and mandatory systems (indicated in Table F3), to better inform decision making. There may be an opportunity to include domestic hunting bag reporting legislation in the new Wildlife and Natural Enviroment Bill that is currently under consideration¹⁰.

2.7.4 Bag limits

Out of season licences

The out of season bag limits for geese were set following guidance from SNH (Section 2.2.7). There appears to be little demand to increase the current bag limits for Greenland Barnacle, Pink-footed or Canada Goose, as the numbers shot under licence are generally less than the quota. There may be merit in revising these figures within the next five years however. There is a more urgent need to critically consider the bag limits for the Scottish populations of Greylag Goose.

At present, there is no official guidance for out of season bag limits for the Native Greylag Goose population, and figures are simply derived from the historical numbers shot. In part, this is due to the lack of regular monitoring data to provide reasonable population estimates (these are available only for 1997 and 2008/2009; Section 2.5.5), compounded by problems with the ever increasing spatial overlap with the naturalised Greylag Goose population. We propose that these two populations should now be treated collectively as the native breeding population of Greylag Geese and that it is critical that bag limits for this population are now set with some rigour. Improved monitoring of the breeding Greylag Goose population must be carried out to derive improved population estimates, which are necessary to inform decisions over sustainable bag limits for this population.

Bag limits for out of season shooting of Icelandic Greylag Geese are currently based on numbers shot in 2006, on the assumption that the population at the time was not declining in response to that level of shooting pressure. These figures have not been reviewed with the continuing redistribution of the wintering population to Orkney (Section 2.5.4). There are also other areas in which the resident and wintering populations of Greylags overlap (e.g. Kintyre). Hence when shooting Icelandic Greylags under an out of season licence, the resident population can also be targeted unintentionally. Regular monitoring of the breeding population is therefore of highest importance in areas that also hold a wintering population, so that counts can be used to correct the latter population estimates accordingly. These figures should then be used to adjust the annual quota (bag limits) for native breeding and wintering Icelandic Greylag Goose populations.

We suggest that improved monitoring of the wintering population of Icelandic Greylag Geese is ensured through improved counts of breeding Greylags in areas of overlap, and the use of the enhanced population estimates to check current bag limits and adjust them accordingly.

Quarry shooting

There appears to be little need to impose statutory bag limits in Scotland for quarry species at present. Also bag limits for quarry species are generally not in place for other countries within the EU (see Appendices F: Table F3). It is important to note however, that the numbers shot as quarry can be several orders of magnitude higher than those shot under out of season licences (e.g. for Pink-footed Geese 20,000-25,000 are thought to be shot during the open season compared with a bag limit for out of season licences of ca 700 birds in 2009). If numbers of quarry species do decline however, there will be an urgent need to consider setting bag limits and this will have to underpinned by an adequate bag reporting system (Section 2.7.3).

2.7.5 Sale of carcasses

Given the recovery of most goose populations within Scotland and the UK to levels higher than previously reported in the last century, calls have been made to review whether there could be scope for selling geese either for local use or as an important Scottish produce (enhancing the value of geese to local communities in Scotland). Only species listed on Annex III/2 of the EU Birds Directive may be sold (Table 2.1), and the sale of geese is permitted in a number of EU countries (see Appendix F- Table F3). The Wildlife and Countryside (1981) Act currently prohibits the sale of geese in the UK however. The extent to which sales are regulated varies between countries (see Appendix F- Table F3).

We suggest that:

• In the absence of careful regulation, there is a legitimate fear that making goose sales legal once more could lead to over-exploitation (either at a national or local level) as occurred previously.
• The lack of a bag reporting system for quarry species in the UK (to monitor numbers shot on an annual basis; Section 2.7.3) would render allowing carcass sales risky in terms of ensuring sustainable management of quarry goose populations, because population modelling to predict the likely effects on populations would involve too much uncertainty.
• It is not clear currently how much business opportunity there is for the sale of goose products in Scotland. However, in order to maintain populations sustainably, it would be important to ensure that the level of take was dictated by a formal adaptive management process (Section 2.7.1), and not by commercial demand.
• Robust systems would be required for tracking the origins of carcasses (and the species taken).
• Any businesses supported by the opportunity would have to be robust to any years when populations were not sufficient to allow hunting to take place.

However, given calls from some stakeholders in parts of Scotland for the complete removal of Greylag Geese and the reinstatement of permission to sell carcasses (Section 4.4), it would be prudent to consider whether allowing limited carcass sales would place higher local value on maintaining goose populations in the future.

We suggest that a scoping study is commissioned to investigate further: (i) the market opportunities for goose products (and other goose benefits) across Scotland, with a particular emphasis on Greylags on the Western and Northern Isles; and (ii) practical options for tracking the origins of goose carcasses were permission to sell them to be reinstated. However, we conclude that any move to reinstate carcass sales in Scotland in the absence of a comprehensive hunting bag reporting system would involve a high risk of defaulting on conservation obligations.

2.7.6 Non-native species

A large increase in the Scottish Canada Goose population is likely to have taken place since 2000 (Section 2.5.9). We believe that there is a real risk that this species could compete with native goose species and that this risk may be greater in Scotland than south of the border (the Canada Goose is perhaps more of an urban/sub-urban species in England). There may also be a future risk to the native Greylag population through hybridization. A recent review of the impacts of alien (non-native) birds in Europe (Kumschick & Nentwig 2010) identified Canada Goose as having the highest potential to cause environmental impact and economic impact within Europe (from a list of 26 bird species). The UK has international obligation to control the impacts of such non-native species under the CBD (Section 2.1) but any such control is likely to be costly (it would need to be carried out over a large part of the range of the species in the UK) and there would probably be public opposition to large scale control of numbers. However, in England an action has recently (2010) been taken to add Canada Goose to Schedule 2 Part II of the 1981 Wildlife and Countryside Act (to allow 'general licensing' for the species to be taken during the closed season). Whilst this action alone may not be sufficient to manage increasing numbers of Canada Geese, we suggest that it is a step in the right direction in terms of managing future risks posed by the species.

We suggest that future goose policy in Scotland should consider carefully and immediately whether Canada Goose should be added to Schedule 2 Part II of the 1981 Wildlife and Countryside Act (as passed in England in 2010). At the very least, future goose policy must keep a careful watching brief on the Scottish Canada Goose population, including maintaining the periodic national surveys of naturalized goose species. We also suggest that the effects of placing Canada Goose on the general licence in England are reviewed in five years time to assess whether the legislative change has had any discernable effect on population levels in England.

2.7.7 Derogation to shoot Barnacle Geese

Given the recent declines in numbers of Greenland Barnacle Geese on Islay, suspected recent declines in breeding success and uncertainties surrounding the results of the population modelling carried out to date (Section 2.5.2), a strictly precautionary approach would dictate reverting the derogation to shoot Barnacle Geese there. At the very least, the population modelling and full census of the population of this race should be updated and levels of uncertainty in population trajectories considered explicitly.

Whilst it appears that changes in Greenland White-fronted Goose numbers on Islay are being driven at least in part by negative changes on the breeding grounds (potential competition with Canada Geese and/or increased snow cover) and the very recent decrease in hunting pressure (allowing the population to stabilize), the observed decline on Islay is not entirely consistent with the evidence, and the declining numbers on Islay since 2000/2001 have contributed disproportionately to the overall decline in numbers observed over the last 10 years (Section 5.5.3). Competition with Greenland Barnacle Geese on Islay and indirect effects of scaring under the Islay LGMS cannot be completely discounted as contributing to the Whitefront decline on Islay. If taking a precautionary approach, it is not possible to conclude for certain that the scaring undertaken on Islay under the LGMS is not having a negative affect on the species' conservation (contravening the AEWA Action Plan as it applies to AEWA Column A species). However, contrary to this, if Whitefronts on Islay are experiencing increased competition from Barnacle Geese, appropriately targeted scaring of Barnacles could conceivably have a positive influence on the Whitefronts. We simply do not know how the two species interact on Islay, or how they respond to current scaring pressures (Section 2.5.3).

We note that the continued increase in the Svalbard Barnacle goose population (since the 2005 review) and the downgrading of its AEWA status (in 2009) mean that this population now has the same AEWA status as Greenland Barnacle Goose if a derogation were to be sought for the purposes of mitigating agricultural damage in future, although recent concerns over reduced breeding success would also have to be taken into consideration, as would establishing the degree of current uncertainty surrounding PVA modelling outputs.

We suggest that the derogation for shooting Greenland Barnacle Geese on Islay is reviewed and, at the very least, the population modelling and census information is updated and levels of uncertainty in population trajectories considered explicitly.

We suggest that in future Greenland White-fronted Geese and Barnacle Geese on Islay receive specific and separate treatment in the LGMS objectives and recording protocols. In particular: numbers and species of geese should be recorded in fields immediately before they are shot over; comprehensive field-by-field information should be collected on the implementation of non-shooting scaring devices, and the use of gas guns; and every opportunity should be taken to assemble more information regarding interactions between the two goose species.

We suggest that serious consideration be given to previous suggestions to take more of an experimental approach to assessing the affects of scaring on Islay, either via direct counts or using grass growth as a proxi for goose numbers, and a draft protocol with costings should be commissioned from an independent source. Given the concern over Whitefronts, an approach which records actual goose numbers in response to scaring for at least a sample of areas would be appropriate.

2.8 Implementation of the 31 existing goose management policy recommendations

Appendix A provides a summary of the 31 recommendations underpinning current goose management policy in Scotland (Scottish Executive 2000, 2005) and indicates the implications of the current review for those that relate to the conservation status of Scottish goose species.

2.9 Conclusions

For all naturally occurring species, Scotland ( UK) has a legal obligation to ensure: (i) maintenance of range and abundance; and (ii) sustainable use for populations that may be hunted legally. Legally it is clear that: (i) naturally occurring populations cannot be driven to extinction; and (ii) adaptive management of populations to meet cultural and economic as well as conservation needs is acceptable as long as resultant populations are sustainable (Section 2.1).

Working definitions of acceptable range and abundance can be difficult to establish however. For waterbirds specifically, AEWA provides indicative threshold population levels and there is a presumption against hunting populations that are listed as AEWA 'Column A'. In practice, public preference and stakeholder pressure also come into play in attempting to agree acceptable population levels for any species, and goose policy to date has, in our opinion, quite correctly moved towards the use of population models to inform the sustainable management of populations.

However, the information available for population modelling to date has been lacking in a number of areas (most importantly in accurate bag data for quarry species and adult survival information for a number of species; Section 2.7). In a number of cases, the predicted population trajectories from the modelling exercise have not reflected observed population changes but results from modelling have not always been presented in a way that allows uncertainty/risk to be assessed transparently. The sensitivity of population predictions to variation (uncertainty) in the various input parameters needs to be explored in more detail and used to prioritise future data collection to improve modelling predictions.

For each of the Scottish goose populations we reviewed changes that have occurred since the inception of current goose policy in 2000, and changes that have occurred since the previous review in 2005 (Section 2.5). For five of the seven main (naturally occurring) goose populations, the Scottish population as a whole is higher now than in 2000 (and range has at least been maintained): Svalbard Barnacle Goose; Greenland Barnacle Goose; Native Greylag Goose and naturalised Greylag Goose (which we now treat as a single population); Icelandic Pink-footed Goose; and Taiga Bean Goose. In view of these population trends, we conclude that goose policy over the last decade has been at least partially successful in meeting the first guiding policy objective to meet UK conservation obligations (with the proviso that we cannot say for certain how a lack of any over-arching goose policy would have affected the conservation status of these species).

The Scottish population of Greenland White-fronted Goose declined by 34% between 1999/2000 and 2008/2009. The part of the population that winters on Islay showed the largest decline (-51% between 2000/2001 and 2009/2010) but some other flocks also showed decreases and some of the smaller flocks have disappeared over recent decades (Section 2.5.3). Whilst there is strong evidence that factors outside Scotland (on the breeding grounds) are contributing to the decline, we conclude that policy could have done more for this species in terms of prioritising it within the Islay LGMS. The lack of understanding of the habitat requirements of Greenland White-fronted Goose, and of the interaction between this species and Greenland Barnacle Goose on Islay (including with respect to the effects of scaring) need to be addressed by treating the two species separately within the Islay LGMS and through improved recording protocols and potentially some experimental work.

The Scottish population of Icelandic Greylag Goose has remained approximately stable since 2000 but a large shift in the wintering distribution has occurred so that the majority of the Scottish population now winters on Orkney rather than further south (Section 2.5.4). We conclude that national goose policy could have done more to react to this change by ensuring the collection of better quality information to inform the changes (such as appropriate count data on Orkney). We see this shortfall as a reflection of an overall weakness of the current National Policy Framework in the extent to which it has the remit to be proactive in tackling emerging goose issues (See Chapter 3).

The part of the population of Greenland Barnacle Geese that winters on Islay (which is covered by a derogation allowing licensed shooting of this species to mitigate against economic losses) has shown a decline (of more than 10,000 birds) over the last three winters (Section 5.5.2). There is some evidence that breeding success has decreased and most evidence suggests that birds have not redistributed from Islay to winter elsewhere. However, given the degree of uncertainty surrounding the recent decline, we recommend that at the very least the population modelling and survey information for this species should be updated and the degree of uncertainty in the model predictions considered explicitly. Whether the derogation to shoot this species is justified should then be reviewed fully, including the possible impacts of the scaring (disturbance) on Greenland Whitefronts on Islay.