Policy Report and Recommendations of the National Goose Forum

Section 2 Status of goose populations breeding and wintering in Scotland and Population Viability Analyses

This chapter outlines what is known about the current status of goose populations in terms of population size and trends, and then deals with the Population Viability Analyses (PVAs) that SNH has undertaken for the NGF on the five main wintering goose populations, in an attempt to understand future trends in the populations, risks to longer term 'viability' and the implications of changed management.

2.1 Status of goose populations in Scotland

Most goose populations that regularly breed or winter (or both) in Scotland are stable or increasing (Kirby et al., 1999, Madsen et al. 1999), with a few possible exceptions, such as the Icelandic population of greylag geese. Current population sizes, trends and supporting information are shown in Annex 2.

2.1.1 Pink-footed geese (Anser brachyrhynchus)

The Icelandic population of pink-footed geese has shown a substantial increase in the last 30-40 years from a population of about 20-30,000 in the early 1950s, stabilising in the 1970s at around 80,000 and then increasing again to over 230,000. Birds breed in the interior of Iceland, where some range expansion has been seen as the population has increased. In winter most birds occupy low lying agricultural land from the Moray Firth down through Grampian, Tayside, Fife and into the Lothians and Borders. Large numbers also congregate in Lancashire and Norfolk. Current trends in the population are masked by substantial fluctuations in recent years (peaking at 260,000 individuals), but the three-year rolling average for the period 1994/95-1996/97 was 223,450 individuals.

2.1.2 Icelandic population of greylag geese (Anser anser)

Icelandic greylag geese for many years increased at a similar rate to pink-footed geese, though the trends diverged in the late 1970s when greylag goose numbers remained stable in contrast to pink-footed goose numbers. Although they breed in Iceland, greylag geese tend to concentrate in lower lying coastal areas, and are therefore more prone to shooting and hunting disturbance than pink-footed geese. In recent years estimates of numbers killed in Iceland have risen to about 36,500 (i.e. before birds leave for UK). In the UK birds winter on Orkney, and in low lying agricultural ground around the Moray Firth, Grampian, Tayside and Fife and into the Lothians and Borders. A few birds winter in northern England, although the great majority winter in Scotland. Population trends are believed to be either stable or declining and the population now stands at about 80,000 individuals. The three-year rolling average for the period 1994/95-1996/97 was 80,600 individuals

2.1.3 Greenland Population of barnacle geese (Branta leucopsis)

The Greenland population of barnacle geese has increased markedly in the last 30-40 years, and is now probably as high as it ever has been. The current population estimate is 40-45,000 individuals, which all breed in eastern Greenland (Madsen et al. 1999), although this estimate may need to be revised in the light of results from the full census of the entire wintering range in the spring of 1999. The bulk of the wintering population occurs on Islay: the three-year rolling average for the period 1996/97-1998/99 was 32134 individuals. There are smaller numbers at other west-coast island locations, ranging from Orkney in the north, through to the Outer Hebrides, and on Tiree and Coll. About 8,000 birds winter in Ireland.

2.1.4 Svalbard population of barnacle geese (Branta leucopsis)

This population, which breeds entirely on Svalbard, has also increased from a very low level in the 1940s (with perhaps as few as 300 birds) to the 1997/98 estimate of 24,000 birds. Almost all of the population winters around the Solway Firth, mainly on the Scottish side, though increasing numbers are using Rockcliffe and Burgh Marshes on the English side. Smaller numbers are found at a number of eastern coast localities, especially early on in the season (such as Loch of Strathbeg and Loch Leven). The current population was at one time thought to be stabilising, but has now apparently begun to increase again, which may be due to expansion into new breeding habitat (on Svalbard).

2.1.5 Greenland white-fronted geese (Anser albifrons flavirostris)

Greenland white fronted geese breed along the west coast of Greenland, migrate through Iceland, and winter along the west coast of Scotland and Ireland (apart from the substantial flock at the Wexford Slobs in south-eastern Ireland). Numbers have increased since the early 1980s to a peak of about 34,000 in 1995/96. However the population in spring 1998 was estimated to be about 30,000, with some 20,000 birds in Scotland and 10,000 in Ireland, and there is some suggestion that the population is stabilising (Cranswick et al. 1999). Within Great Britain most birds winter on Islay (the three year rolling average for the period 1996/97-1998/99 was 12267 individuals) although small numbers occur down the west coast from Caithness southwards, with substantial flocks on Tiree and Coll, and in Kintyre and Galloway.

2.1.6 Native greylag geese (Anser anser) and naturalised greylag geese (Anser anser)

Populations of native and naturalised greylag geese are small compared to the migratory greylag goose population, nor are they strictly migratory in the sense of overflying national boundaries. Native greylag geese , which formerly bred in wetlands widely across Britain, have been confined to parts of Caithness and Sutherland, Wester Ross and the Inner and Outer Hebrides for at least the past century. The most up-to-date population estimate of about 10,000 birds in 1997 is an increase on earlier estimates, and although this population is heavily shot, it is likely that it is slowly expanding and recovering some of its former range. The naturalised greylag goose population is often treated separately to the 'native' population, but many birds in the naturalised population originate from deliberate releases some years ago of birds taken from the native greylag goose population, hence the two populations cannot be genetically distinguished from each other. Significant numbers occur in southwest Scotland.

2.1.7 Taiga bean geese (Anser fabalis fabalis)

The only Scottish flock winters on the Slammanan Plateau. These birds are believed to be of the Taiga sub-species, which breeds in Scandinavia and north western Russia. This flock is very small, with only about 150 birds counted in 1998 _ (Cranswick et al. 1999). There is no good evidence on population trends.

2.1.8 Canada goose (Branta canadensis)

The Canada goose is an introduced species. Most of the population breeds and winters in England, though some birds undergo a moult migration to Scotland in a manner similar to the ancestral population in the USA. Scotland is thought to hold about 1,000 birds, and this number may be increasing.

2.2 Introduction to Population Viability Analyses (see Annex 3 for parameter values used in the risk assessment resource)

Although most geese found in Scotland are thought to have relatively healthy populations, variability from year to year in both birth and death rates means that longer term viability may not be guaranteed, especially if environmental conditions change in an adverse manner. Therefore, before recommending any management changes in relation to goose populations, the NGF considered it necessary to have available the best possible predictions of future population trends and the best assessment of the impacts of any such changes. Population Viability Analyses (PVA) were commissioned, and the remainder of this section introduces this technique and how it has been applied to five of the main wintering goose populations. The work undertaken is explained in greater detail in the collection of NGF papers and minutes which form the NGF background report, and PVA accounts for individual species have been published by SNH.

2.2.1 Background

The numbers of most animal populations change from year to year, and wintering geese are no exception. Reasons for such changes are many, but include variability in birth and death rates (demographic and environmentally induced variability) as well as other, intrinsic factors such as genetic drift. This uncertainty means that most species, even common ones, face the potential for long term decline towards extinction, although the risks for common species are generally thought to be quite low. The development of the theory and practice of Population Viability Analysis allows an understanding of how such uncertainty in the key demographic parameters from year to year affects long term population trends, and it can used to develop predictions over predetermined time periods. It must always be borne in mind that PVA is essentially an exercise in estimating probabilities, and that figures produced by such methods are probabilities of different population trajectories over time. Because PVA models are simulation models, they can also be used to examine the effect of changing one or more of the demographic parameters on the long term viability of the population. Another important use relates to the impact of catastrophic events where large numbers (often more than 50%) of the population die over a very short time period. Although the frequency and severity of such events is poorly known in animal populations, such events do occur, especially those arising from outbreaks of disease. Outbreaks of disease are known in wild geese, but their contribution to population dynamics is poorly understood. A recent and better known example is the phocine distemper virus which killed a very large percentage of the common seals occurring in the seas around the UK.

2.2.2 Application of PVA models to the management of wild species

Most PVA models have been applied to species of conservation concern, in an attempt to establish circumstances under which the population could either be enhanced or managed in such a way as to sustain a viable population in the longer term.

PVA models appear to have been used less frequently in setting management goals for populations either at favourable conservation status or for populations that are considered as 'problem' species. However, use of such models is increasing, and there are several such examples for North American geese. All these models have been used to test, theoretically, the effect of changes in management upon these populations and have not been used to determine 'target' population sizes (either 'minimum viable populations' or indeed, 'maximum tolerable populations').

Such models are still at an early stage of development, and to date there has been little practical or experimental validation of this approach. Hence this procedure, although offering valuable insights into population management must be viewed with some caution. The methodology used in the work commissioned for the NGF is described in more detail in Annex 4.

Within the USA a concept known as Adaptive Resource Management (ARM) is gaining acceptance, as a means of managing many animal populations. ARM is an iterative approach in which management policies may be treated as hypotheses and management actions as experiments. Based on the outcome of these experiments, policies and management actions may be changed or refined and further tested by another stage of management implemented as a quasi-experiment. This approach could be adopted in order to test and refine the Forum's management recommendations.

2.2.3 Work carried out to date on Scottish wintering goose populations

SNH has, in conjunction with Wildfowl and Wetlands Trust and latterly the Institute of Zoology, London, provided the results of PVAs on the five main wintering populations of geese of concern in relation to agriculture. These are:

• Greenland white-fronted geese
• Greenland population of barnacle geese
• Svalbard population of barnacle geese
• Icelandic/Greenlandic population of pink-footed geese
• Icelandic population of greylag geese

In addition, further work was carried out looking at the risk to four of the five populations under a broader range of conditions, including lower initial population sizes, as well as different levels of additional mortality (or reduced mortality for greylag geese) and different frequency and severity of catastrophic events. This work was also undertaken by the Institute of Zoology, London.

2.2.4 Main findings for each of the species populations

2.2.4.1 Pink-footed geese

• The current population of pink-footed geese, of about 235,000 individuals, is healthy and the long term likelihood of decline under current conditions is very small. The risk from any catastrophic events is also very low, except in the most extreme conditions modelled.
• During the period 1984-1992 the population grew at a rate of about 11% per annum. However, recent counts suggest a stabilisation or a slow down in the population growth rate. This means that any additional mortality of this magnitude would stabilise the population: any mortality in excess of this will lead to long term population decline. The risk of the population falling to 100,000 or below within the next 5 years under a 10% additional mortality is less than 2%, though this rises to over 70% over the 25 year period.
• The level of hunting mortality in Iceland is known with some precision. Similar figures are unavailable in the UK, and although the numbers shot under licence are known, these are probably a small fraction of the total annual bag shot by wildfowlers.

2.2.4.2 Icelandic greylag geese

• At about 80,000 the Icelandic population of greylag geese appears healthy, but at present the population is finely balanced between long term increase and long term decline. The predicted rate of increase in the population is only 1% per annum, though given the wide error limits on this, long term decline is a real possibility.
• A substantial proportion of birds (especially young birds) are killed in Iceland each year according to hunting bag records collected in Iceland. If this kill were to increase by 5% then the probability of a decline to less than 50,000 birds is 16% over 5 years and 90% over 25 years. Similarly if mortality is reduced by 5% then the long term potential for decline is reduced to almost zero.
• As the population is most sensitive to changes in adult and juvenile survival, then the level of hunting in Iceland and the UK is likely to be critical in determining the longer-term trends in this species population.

2.2.4.3 Greenland population of barnacle geese

• The current size of 40-45,000 barnacle geese in the Greenland population appears reasonably secure in the long term. The predicted rate of increase in the population of 3.6% per annum (Pettifor et al. 1999) suggests that under current conditions, the population will continue to grow, though rates of growth may vary between different wintering locations. While the rate of growth on Islay appears to have slowed, there is no evidence to suggest that the population is stabilising at an equilibrium level.
• The chance that under current conditions the population will fall below 20,000 in the short term (5 years) is negligible), nor does the risk increase appreciably over a longer 25 year time period, although this prediction is made in the absence of any catastrophic events occurring within this time period.
• Ignoring the likelihood of catastrophic events affecting the geese, the risk of long-term decline in the population to less than 20,000 individuals following an increase in mortality of up to 5% remains very low over 5 years (less than 1%), but increases to about 42% over 25 years. The risk is less than 1% over 5 years if mortality is increased by just 2.5%, though it increases to about 7% over 25 years. At 10% additional mortality, the risk of decline is about 8% over 5 years but rises to more than 97% over 25 years.
• As with many geese, the population appears most sensitive to changes in survival rather than to fluctuations in productivity.
• The aggregation of some 60-70% of this population into one wintering area, Islay, makes it vulnerable to chance catastrophic events.

2.2.4.4 Svalbard population of barnacle geese

• The current level of 20-23,000 barnacle geese breeding on Svalbard is relatively healthy (especially compared to the situation some 50 years ago), but the population is still sensitive to any decrease in survival.
• Although density dependence in productivity and survival has been found on the breeding grounds, it does not appear that this regulates the population as a whole, and at present the population is expanding, presumably because birds are still colonising new breeding habitat. Further increases in this population might be expected in future years.
• However, the aggregation of this population into one breeding area and one principal wintering area makes this population vulnerable to chance catastrophic events, such as adverse conditions on the breeding grounds, disease or adverse conditions during migration.
• As the population increases, it is possible that additional breeding and wintering areas will become established and increasingly important.

2.2.4.5 Greenland white-fronted goose

• The current population of Greenland white-fronted goose of 30,000 individuals (about 20,000 in Scotland) has increased to a level where its long-term viability is more secure than some 10-20 years ago. Although the population appears healthy at present and has been increasing at a rate of about 7.15%, the risk of short-term decline within 5 years to below 20,000 individuals under conditions of no additional mortality and no catastrophic events is about 4.5%.
• Modelling shows that the population is most sensitive to changes in survival rather than productivity, although it is noteworthy that the recent slight decline in Greenland white-fronted geese appears to be correlated with a decline in productivity (Cranswick et al 1999).
• If mortality is increased, then the risk of decline increases markedly. At a level of 2.5% additional mortality, the risk of decline over a five year period to a population of less than 20,000 is about 7% rising to about 16% over 25 years. The risk of decline increases to about 10% over a five year period for an additional mortality of 5% and nearly 32% over a 25 year period.
• The current population of Greenland white-fronted geese seems to have a favourable conservation status under the existing protective regime. However, the recent small decline in the wintering population in the UK and Eire strongly suggests that it is likely to be quite sensitive to any additional mortality imposed in any part of its breeding, passage or wintering range. Conversely, if the population expands further the situation will need to be reviewed in the future.
• The aggregation of around 40% of this population into one wintering area, Islay, makes it more vulnerable to chance catastrophic events.

2.2.5 Management recommendations for the wintering and breeding goose populations.

2.2.5.1 Icelandic pink-footed geese

• The Icelandic/Greenlandic population of pink-footed geese is currently healthy and could sustain a certain amount of additional shooting (hunting) pressure. As this species is a legitimate quarry species then there is no impediment to achieving this.
• For the close season especially during late winter and early spring, licences are already issued where appropriate, to prevent serious damage to crops. This system may continue, with the usual consultative approach in areas where populations within SPAs are at issue.
• It is imperative that continued monitoring of the population and of hunting mortality occurs. At present, the autumn roost count (in October) gives an acceptably accurate record of the population size. It may be necessary as part of any goose management scheme that includes pink-footed geese to undertake further monitoring in late winter and early spring, when distribution and habitat use changes from the pattern seen in early autumn.

Recommendation 1: Subject to improved monitoring of the population and of mortality from hunting, additional shooting of Icelandic pink-footed geese would be acceptable.

2.2.5.2 Icelandic greylag geese

• The current population of Icelandic greylag geese is healthy, but the potential for increased mortality either in Iceland or UK means that any action that increases greylag goose mortality should be avoided.
• For the close season, especially during late winter and early spring, licences are already issued where appropriate, to prevent serious damage to crops. This system may continue, with the usual consultative approach in areas where populations within SPAs are at issue.
• It is imperative that continued monitoring of the population and of hunting mortality occurs. At present, the autumn roost count (in November) gives an acceptably accurate record of the population size, although an examination of roost usage across the wintering range may be valuable to ascertain whether any significant concentrations of greylag geese are being missed in the autumn census counts. It may be necessary as part of any goose management scheme that includes greylag geese to undertake further work on monitoring in late winter and early spring, when distribution and habitat use changes from the pattern seen in early autumn.
• In view of the fact that greylag geese often feed with pink-footed geese, any increased mortality of pink-footed geese should not lead to increased mortality of greylag geese (where they occur together).

Recommendation 2: Increased mortality from hunting in the Icelandic greylag goose population should be avoided, and monitoring of current hunting mortality and population levels should be co-ordinated.

2.2.5.3 Greenland population of barnacle geese

• The Greenland population of barnacle geese is healthy and currently increasing. However, given the potential vulnerability from quite small changes in additional mortality, it is not recommended that the population be returned to the quarry list at this stage, though the position should be kept under review.
• Licensed shooting to prevent serious damage to crops may be undertaken, and could be used as an aid to any overall management across the range or in particular areas such as Islay. It is suggested that any additional level of additional mortality is set at a precautionary level of between 2.5%-5.0% (bearing in mind the figure of 3.6% required to stabilise the population at current growth rates).
• This global figure for mortality could either be spread across the range or targeted onto specific areas and flocks (bearing in mind international obligations and the network of protected areas such as SPAs already established for this population). One consequence of this additional mortality may be to encourage greater dispersal of birds into other areas, where impacts on agriculture may be less than on Islay at present. Wider dispersion in winter would also in principle enhance the probability of the population's viability in the longer term. Any such management strategy may also need to consider appropriate habitat management in some of the previously well-used locations, such as islands where grazing has been removed.
• Enhanced monitoring of existing populations and additional mortality must be undertaken as an essential part of the National Policy Framework. Monitoring of the whole population (currently undertaken on a five-yearly basis), should at least initially, be undertaken annually to allow refinement of the management recommendations.
• Local goose management groups may need to consider setting appropriate limits of acceptable change¹ for local barnacle goose populations, within which the management strategy would seek to regulate the overall population size and density.

Recommendation 3: Where required to support local management activity shooting of Greenland barnacle geese may be authorised, under licence. Initially additional mortality should not exceed 5% per annum of the current population.

Recommendation 4: Shooting may be licensed across the range of the population or targeted at specific areas, with a view to reducing local concentrations and encouraging wider geographic dispersal of the population in winter.

2.2.5.4 Svalbard population of barnacle geese

• The Svalbard population of barnacle geese is currently increasing and is looking increasingly healthy. At current population levels, a significant increase in shooting under licence cannot be justified. Further population viability analysis work and risk assessment would inform consideration of policy.
• Monitoring of the population is essential, and it may be necessary to move to a more formal annual co-ordinated count across the known wintering range (largely focused on the Solway at present). Currently goose data are obtained from an early count, and in recent years also a late winter count on and around the Solway Firth. Such counts could be co-ordinated with counts at other roost sites, currently undertaken as part of Wetland Bird Survey (WeBS).

Recommendation 5: Where necessary to prevent serious damage to agriculture, licensed shooting of Svalbard barnacle geese should continue to be permitted.

Recommendation 6: The position should be regularly reviewed in view of the improving conservation status of the population and further population viability analyses and risk assessment work should be commissioned for this population, to inform consideration of future changes in policy.

2.2.5.5 Greenland white-fronted geese

• The Greenland white-fronted goose population is healthy under the existing protective regime and although it has been increasing, recent counts show a slight decline in the population. Given the potential vulnerability of this sub-species from quite small changes in additional mortality it is not recommended that the population be returned to the quarry list.
• Given the vulnerability of the population outlined in previous sections, it is not recommended that an enhanced level of licensed shooting be resumed on this population. The risks to the population from additional mortality do not justify an increased level of shooting under licence, although the situation should be kept under review as further population data are gathered.

Recommendation 7: Where necessary to prevent serious damage to agriculture, licensed shooting of Greenland white-fronted geese should continue to be permitted. The position should be regularly reviewed.

2.2.5.6 Naturalised greylag geese

• The current population is small but increasing.

Recommendation 8: No alteration to the protective regime for native greylag geese and naturalised greylag geese is required.

2.2.5.7 Taiga bean geese

• The small size of this population makes it extremely vulnerable.

Recommendation 9: No alteration to the protective regime for Taiga bean geese is required.

2.2.5.8 Canada geese

• This species has not to date been a serious problem for agriculture except possibly in certain locations.

Recommendation 10: Canada geese may continue to be hunted or shot under licence where there is serious damage to agriculture.

¹ Limits of acceptable change (LACs) are well established means of determining the upper and lower limits within which populations are allowed to fluctuate. Populations falling below or rising above such limits may then be subject to particular management actions designed to either restore or limit numbers as appropriate.