2018 Changes in migration and wintering patterns of Greylag Geese Anser anser from southernmost Sweden during three decades

The recent increase of the Greylag Goose Anser anser population has caused complaints about crop damage. In response to this, the Nordic Collegium for Wildlife Re- search started a neck-banding program to elucidate the migration and movement patterns. During 1984–2009, 2639 Greylag Geese were marked in a breeding area in SW Scania, southernmost Sweden. Up to and including 2012, 15 296 re-sightings were reported from outside the breeding area (in addition to more than 100 000 from the breeding area). In the early years of the study, the majority of the geese migrated to winter in southernmost Spain, staging in The Netherlands. In later years, hardly any individuals migrated to Spain, the majority stay- ing in The Netherlands with an increasing proportion wintering in southern Sweden. From 1986 to 2017, the mean latitude for re-sightings changed ten degrees to the north. About 13% of the geese changed winter quarters between two consecutive winters. Moreover, the south-ward migration occurred much later in the autumn and the geese returned to the breeding areas about one month earlier in spring.


Introduction
Like most other goose populations in Europe, the Greylag Goose Anser anser has shown a marked increase in numbers in recent decades (Nilsson et al. 1999, Kampe-Persson 2002, 2010, Fox et al. 2010, Nilsson 2013, Fox & Madsen 2017. The populations in mainland north-western Europe (the East Atlantic Flyway population) increased from about 200 000 in the early 1990s to more than 600 000 twenty years later (Fox et al. 2010) and to 960 000 in the most recent report (Fox & Madsen 2017). This very marked increase in the population of Greylag Geese, as well as in other goose species, that started in the 1970s led to many complaints about farmland crop damages (Buij et al. 2017). Therefore, research projects including neck-banding of Greylag Geese and other species were started in many countries to increase the knowledge about the movement patterns and the migrations of different goose species. In the Nordic countries, a Greylag Goose project was initiated in 1984 (Andersson et al. 2001). The project was initially aimed for a five-year period but continued during the following years. In SW Scania, in southernmost Sweden, marking stopped after ORNIS SVECICA 28: 19-38, 2018 Changes in migration and wintering patterns of Greylag Geese Anser anser from southernmost Sweden during three decades Förändringar i flyttnings-och övervintringsvanor hos grågäss Anser anser från sydligaste Sverige under trettio år

LEIF NILSSON & HAKON KAMPE-PERSSON
The recent increase of the Greylag Goose Anser anser population has caused complaints about crop damage. In response to this, the Nordic Collegium for Wildlife Research started a neck-banding program to elucidate the migration and movement patterns. During 1984During -2009During , 2639 Greylag Geese were marked in a breeding area in SW Scania, southernmost Sweden. Up to and including 2012, 15 296 re-sightings were reported from outside the breeding area (in addition to more than 100 000 from the breeding area). In the early years of the study, the majority of the geese migrated to winter in southernmost Spain, staging in The Netherlands. In later years, hardly any individuals migrated to Spain, the majority staying in The Netherlands with an increasing proportion wintering in southern Sweden. From 1986 to 2017, the mean latitude for re-sightings changed ten degrees to the north. About 13% of the geese changed winter quarters between two consecutive winters. Moreover, the southward migration occurred much later in the autumn and the geese returned to the breeding areas about one month earlier in spring.
Traditionally, the Greylag Geese in mainland north-west Europe migrated along the east Atlantic coast via staging areas in The Netherlands to winter quarters in Spain, mainly the Guadalquivir Marismas in the south-west of the country (Nilsson et al. 1999, Andersson et al. 2001. Already during the early part of the study period, changes in the migration patterns and staging habits were recorded (Nilsson et al. 1999, Andersson et al. 2001, Nilsson 2006, Nilsson 2013). This change resulted in fewer geese from southern Sweden migrating to south-western Spain. Instead new traditions were established with wintering areas in northern Spain and further north in The Netherlands. Some Greylag Geese even started to winter in Sweden.
In the present study, we analyse the data from the neck-banding project in SW Scania for the entire study period 1984-2012. A former major analysis (Andersson et al. 2001) also covered neck-banding at other sites in Sweden but these projects were terminated earlier and the re-sightings from them are not analysed in this paper. https://doi.org/10. 34080/os.v28.19521 Material and methods

Marking area
The marking (i.e., neck-banding and ringing) area for this study was situated in south-western Sweden (SW Scania, Figure 1). The main marking efforts were made at four (later two) lakes in a relatively small area in an undulating open agricultural landscape with small patches of woodland in the western lake area (Figure 1). The marking sites were all eutrophic lakes with rich reed beds Phragmites australis along parts of the shores. Some of the lakes have small islands suitable for breeding geese. Grasslands and at one lake a golf course offered good feeding conditions for the goose families. A description of the area is found in Nilsson & Persson (1994). The breeding population of Greylag Geese increased from 120 pairs in 1985 to a peak of 1340 pairs in 2004, and thereafter the population decreased and in 2012 there were 820 pairs the area. Various aspects of the breeding ecology and dynamics of this population have been studied (Nilsson & Persson 1992, 1994, 2001a, 2001b, Nilsson et al. 1997, 2002, Nilsson 1998. During the first years of the study, geese were also neck-banded at two lakes in the eastern lake area of Scania ( Figure 1).
In SW Scania, due to all other studies of the Greylag Geese, all major sites for the species were checked for the occurrence of marked individuals at least on a weekly basis during the breeding season (arrival to early summer). The same applies to the coastal staging sites in the local region ( Figure  1) during late summer and early autumn when families from the study area concentrated here.

Catching and marking
In SW Scania, breeding families were caught on the breeding sites when the goslings were about six weeks old and the parents were unable to fly due to wing moult. The families were rounded up when feeding on grasslands, pastures or a golf course and driven into nets, which were mounted beforehand (Persson 1994a, 2000a, Kampe-Persson 2004. We kept the geese in large jute sacks when they awaited handling. The geese were marked with blue neck collars of UV-resistant laminated plastic. The collars had three-digit codes engraved with the first character larger and the two following smaller and in right angel to the first one. Letters and numbers most liable to misidentification were omitted. In good light conditions, it was possible to read the codes at a distance of 500-600 m. To make it possible to follow the individuals in the study area also in the case they lost their neck-collars, a proportion of the Greylag Geese were also marked with coloured leg rings with an engraved code. All individuals were also marked with metal rings from the Swedish national ringing scheme. For further information on the Nordic marking scheme, see Andersson et al. (2001). The mean annual retention rate of neckcollars in this study was 93.1±1.6 % for males and 98.8±0.5 % for females (Persson 2000b and unpubl.) Material During the years 1984-2008, a total of 2 639 Greylag Geese, 685 moulting birds and 1 954 goslings, were neck-banded in the study area in SW Scania and included in the present study. Most moulting birds were breeding adults but there were also about 20 non-breeders, most of them from the local population.
When one re-sighting per country and month was counted, in all 15 296 re-sightings of the neck-banded Greylag Geese were reported from other areas than the local study area up to and including year 2012 (Table 1). In addition to the re-sightings spec- Figure 1. Map of South Sweden with the position of the main study areas inserted. Neck-banding sites are marked with red dots. The main neck-banding and ringing was undertaken in the West Lakes area, where the most intensive studies were conducted during the entire period. Some markings were made in the East Lakes area during the first years of the study. Foteviken at the coast is an important staging area for Greylag Geese from the West Lakes area and was together with the lake area regularly searched for neck-banded geese. Karta över södra Sverige med undersökningsområdena markerade. Märklokaler har markerats med röda prickar. Huvuddelen av märkningarna ägde rum i det västra sjöområdet där mer intensiva undersökningar gjordes under hela perioden. Några märkningar genomfördes i det östra sjöområdet i början av undersökningsperioden. Foteviken vid kusten är ett viktigt område för gässen från det västra sjöområdet och genomsöktes tillsammans med sjöområdet regelbundet efter märkta gäss. ified in this table, 100 050 re-sightings were obtained locally in the study area in SW Scania in connection with other studies during the same time period (Nilsson & Persson 1991, 1992, 1994, Nilsson & Kampe-Persson 2017.
Re-sightings from areas outside the study area in SW Scania were obtained through a large network of voluntary observers that checked the flocks of geese for the occurrence of neck-banded individuals. One of the authors (HKP) carried out several research projects on marked Greylag Geese in Spain during the winters 1985/1986-2001/2002, also visiting staging/wintering areas in The Netherlands, France and Portugal (Persson 1993, 1996a, 1997, Kampe-Persson 2002. Since 2007 a website (www.geese.org) has been created on which observers can enter their readings of goose rings. On this website a lot of re-sightings, also from the past, were entered by observers through google-maps. The website encourages volunteers to report sightings, because they can check online where the bird they have observed has been ringed and which other re-sightings of the same bird that have been made.

Distribution of staging and wintering Greylag Geese
The general migration pattern of the Greylag Geese marked in SW Scania in the early part of the study period has been described in Andersson et al. (2001). In the present study, we therefore concentrate on comparing the migration patterns from the later years with that of the early part of the study period to elucidate changes in migration habits.
The majority of the Greylag Geese seen outside the study area in SW Scania have been reported from The Netherlands and Spain (Figures 2-4), whereas much fewer have been reported from other countries (Table 1). During the first years of the project, up to 50% of the neck-banded individuals reported in a single year were seen in The Netherlands at least once, whereas the highest annual percentage reported from Spain was somewhat more than 35%. In both cases, the proportion of neck-banded individuals reported have decreased over the years, partly as an effect of changes in the migration pattern as discussed below, but other factors such as the intensity of search is also a factor of importance. It should be kept in mind that the goose populations increased dramatically during this time-period, so it was more difficult to find neck-banded geese with the same fieldwork effort. The Guadalquivir Marismas was intensively searched for marked geese up to 2002 but not thereafter. This can have influenced the number of marked geese reported from this wintering area in later years.
During the summer, most reports of neck-banded Greylag Geese were from the study area with some records from other parts of South Sweden ( Figure  4). In the early period, a number of non-breeding Greylag Geese from SW Scania were reported from the moulting area in Oostvaardersplassen in Flevoland in The Netherlands during late May to early July, whereas only single observations were obtained from The Netherlands in these months    1984-1990 1990-1996 1996-2002 2002-2008 1984-1990 1990-1996 1996-2002 2002-2008 Western Western Germany        1985-1989 1990-1994 1995-1999 2000-2005 2006- 1985-1989 1990-1994 1995-1999 2000-2005 2006- 1985-1989 1990-1994 1995-1999 2000-2005 2006- 1985-1989 1990-1994 1995-1999 2000-2005 2006- 1985-1989 1990-1994 1995-1999 2000-2005 2006-2012  Fördelning på latitud (olika färger) för avlästa grågäss halsmärkta i sydvästra Skåne under fem olika tidsperioder (1985-2012). during later years (Figure 3). Numbers moulting at Oostvaardersplassen decreased already during the 1990s, when a new moulting tradition for south Swedish Greylags was established on the island of Saltholm between south Sweden and Denmark (Fox et al. 1995) and hardly any Greylag Geese from SW Scania went to Oostwaardersplassen anymore. Thus, during a period in the 1990s (due to intensive fieldwork on Saltholm in connection with the impact assessment for the bridge between Sweden and Denmark) more observations of marked Greylag Geese from SW Scania were reported from Denmark in June than from any other month (Figure 3). Unfortunately, the intensive studies on Saltholm stopped after year 2000 and there is therefore no information about geese from Scania there in later years. However, Greylag Geese still go there to moult as numbers counted on the moulting site was 46 500 in 2011 (Michael Flink pers. comm.). In later years, also a number of Greylag Geese marked in SW Scania were reported as moulting in Lake Hornborgasjön (Nilsson & Hermansson in prep.).
In April, most reports were from the breeding areas with some individuals lingering around in the north, fewer were found south of Sweden in the second period than during the first period (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994). In May, the moult migration to Oostvaardersplassen in The Netherlands was apparent in the first time-period, whereas no geese were reported from this area in May during the second period.
Outside the main flyway from south-west Sweden, only few reports have been obtained of marked Greylag Geese from other areas. As is seen from the maps small numbers have been seen regularly in Britain and there are a few observations further to the east in Europe. A few individuals from southwest Sweden established a breeding population in Britain, probably somewhere in Scotland and migrating to wintering areas in England (cf. Insley 1997).
The maps (Figure 4) clearly show that marked changes in the migration pattern and wintering areas of the Greylag Geese from south-west Scania have occurred over the years of the study. These changes became still clearer when the distribution of the re-sightings was analysed according to latitudinal distribution ( Figure 5). The differences are most prominent for December and January, when almost all re-sightings were from The Netherlands and Spain during the first period (1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996)(1997)(1998)(1999), with no re-sightings reported from the marking area. tumn, the majority of the Greylag Geese remained in SW Scania close to the marking areas, but there was a shift in the local distribution with a concentration to coastal areas (Nilsson & Persson 1992, Nilsson 2013, Nilsson & Kampe-Persson 2017. Smaller numbers stayed in Denmark, The Netherlands and northern Germany during this time of the year (Figures 3-5). Re-sightings from The Netherlands in late summer were restricted to the early part of the study period, when Greylag Geese from Scania still moulted at Oostwaardersplassen. Most German re-sightings during August and September were from Mecklenburg-Vorpommern. October and November was the main migration period for the Greylag Geese from SW Scania during the early years and a large proportion of the neck-banded birds was reported from The Netherlands. In these months, marked differences were found between the different time periods. During the early part of the study an appreciable proportion of the geese were already reaching south-western Spain in October and the staging and wintering areas in The Netherlands and Spain dominated over the areas in the north, for example Sweden, in November. In the later years, the majority of the Greylag Geese remained in Sweden during this time of the year. Small numbers were found in The Netherlands in October but hardly any reached Spain in that month. The picture was similar for November in the late period with much fewer south Swedish Greylag Geese reaching Spain compared to in the earlier years.
December and January are typical winter months and Greylag Geese were found all along the flyway from SW Scania to southern Spain. When comparing the two time-periods for these months it has been a clear shift in the distribution, especially during January. Within Spain, two areas were important during the first time-period, the Guadalquivir Marismas in the south-west and Villafáfila in the north, whereas very few geese were found in the former area in January during 2000-2012. Moreover, Greylag Geese from SW Scania were reported from a number of new sites in Spain during the latter period. It is also clear that more Greylag Geese were on migration northwards in January during the latter time-period compared to the first period. Re-sightings from South Sweden were more common in later years. Over the study period, there was a northward change of the mean latitude for January re-sightings of marked Greylag Geese from Scania ( Figure 6).
In February, Greylag Geese were found along the entire flyway and spring migration northwards was In the second period, an increasing proportion of the birds were found wintering in Sweden. In the same time period, the numbers wintering in Spain decreased markedly. Over the period 1986-2012, the mean latitude for January re-sightings changed about ten degrees northwards, the trend being highly significant (Figure 6). This change towards more northerly wintering areas continued during the entire study period.
When the study started only few Greylag Geese were found in SW Scania during the January counts. In the same way there were no winter observations of neck-banded Greylag Geese from Scania during the first years of the study. Beginning in 1991, small numbers of marked Greylag Geese remained in SW Scania in most winters, staying mainly close to the coast at Foteviken. Suddenly, in 2005, this number increased and 25% of all marked Greylag Geese seen that year were seen in Scania during the winter (Figure 5a-c), with more that 30% of the neck-banded geese staying in the winter of 2009.
Over the years, 13% of the re-sighted Greylag Geese changed winter quarter between two consecutive winters ( Table 2). The highest frequency of change was seen for the two Spanish winter areas (14% and 15%, respectively), whereas only 8% of those wintering in SW Netherlands changed winter area between two consecutive seasons. Clearly, the number of changes shown in Table 2 are biased to the low side as all geese were not reported in the winter in all years (not being seen). In addition to the 29 changes recorded between two consecutive years a further 27 changes were noted for geese that were not recorded in consecutive winters. The majority of the changes were from a southern winter area to a more northern winter area (Table 3) but there were six geese changing from the Netherlands to Spain. There is, however, a possibility that some of these birds were late migrants staging in the Netherlands not being seen later the same year in Spain.

Phenology of migration
The monthly distribution of reported re-sightings of neck-banded Greylag Geese from different countries gives a good picture of the phenology of the migration for the species through the main staging and wintering areas (Figures 3, 4). For Denmark, the picture was markedly dominated by the number of Greylag Geese reported from the moulting site at Saltholm in May-June during a number of years in the 1990s. Very few observations were reported from Denmark during November-February, but generally, the number of reported re-sightings from Denmark was quite small. Apparently, most Greylag Geese just passed over Denmark on their migration south.
The picture for Western Germany is quite different from the Danish one. During the first years, few Swedish Greylag Geese were reported from Germany outside the migration periods, but in the latter part of the study period, a number of observations of marked Swedish Greylag Geese were reported from the winter months and early spring. There are some reports of moulting Swedish Greylag Geese from the German west coast in May-June. As mentioned before, the largest number of resightings of Scanian Greylag Geese were reported from The Netherlands. During the early years, the highest numbers were reported from the autumn period, with much smaller number of re-sightings reported from the other months. Over the years, there was a change in the pattern with the highest counts reported from October 1984-1990, in November in 1990-1996 and in December for the last two periods (1996)(1997)(1998)(1999)(2000)(2001)(2002)(2003)(2004)(2005)(2006)(2007)(2008).
France was mainly a passage country for the Greylag Geese from SW Scania (Figure 2). Very few observations were reported before October. Peak number of observations were noted in November and February (during the first period March). Few geese from Scania stayed in France over the winter. For further details on the occurrence of Greylag Geese in France, see Nilsson et al. (2013). The majority of the Greylag Geese from SW Scania migrating to Spain apparently made a direct flight from the Netherlands to Spain (Persson 1993(Persson , 1994b. We assume that it was the extremely high hunting pressure in France (Persson 1999) made the geese reluctant to stage there Spain is the southernmost wintering area for the Greylag Geese from SW Scania even if a few individuals have been recorded in Morocco. During the first two periods (1984)(1985)(1986)(1987)(1988)(1989)(1990)(1991)(1992)(1993)(1994)(1995)(1996), many Greylag Geese were seen in Spain but especially in the last period (2002-2008; Figure 3), the number of observations was small. The data from Spain is a little biased as one of the authors (HKP) made intensive field work in the area up to 2002. The data from Spain clearly show a change in the wintering phenology. In the first part of the study Greylag Geese from SW Scania arrived in October, but later the arrival was in November.
In The Netherlands, there is a great interest in reading and reporting neck-bands and other marks on birds and this therefore offers a very good opportunity to follow changes in the timing of the  Table 3. Shifts of winter quarters by neck-banded Greylag Geese Anser anser from SW Scania separated between those seen on different winter areas in two consecutive years (Table 2) and those seen on the second winter quarter later (=year of change unknown).

Byte av vinterkvarter för halsmärkta grågäss från SV Skåne, separat för dem som setts två på varandra följande år (Tabell 2) och dem som setts i det nya vinterområdet senare (år för bytet okänt).
Greylag Goose migration as a very large proportion of the geese stay here for longer or shorter periods during the autumn migration. Thus, calculating annual median dates for the first observation of neck-banded Greylag Geese in The Netherlands give a good indication of changes in the time-table of autumn migration, which is much better than comparing last-sightings of the birds in the staging areas in Scania. During autumn, the Greylag Geese in Scania are dispersed over a large area for feeding during the day and it is difficult to establish with certainty when they leave the region on migration. Between 1987 and 2011 there has been a significant trend towards a later arrival of the marked birds in The Netherlands (Figure 7), the median  1986 1989 1992 1995 1998 2001 2004 2007 2010 Median arrival date (ten day periods from August 1) Year /År Median arrival date (ten day periods from January 1) Year/År Figure 8. Median arrival dates in spring in SW Scania, South Sweden in different years during ten-day periods from January 1st for Greylag Geese Anser anser marked in the area in earlier years.  1986-1990 1991-1995 1996-2000 2001-2005 April arrival date for the last few years being about one month later (early November) compared to the start of the study (early October). The timing of spring arrival of the Greylag Geese from SW Scania has also changed significantly during the study period (Figure 8), and the median arrival date back to Scania during the last few years is now ca. one month earlier than at the beginning of the study. For Greylag Geese known from winter re-sightings to winter in Spain, about 20% were back in the breeding areas in Scania in February during 1986-1990 compared to more than 40% in 2001-2005 (Figure 9).

Discussion
The reason why only few birds used Denmark as a staging area in the autumn, just passing over the country on the migration to the staging areas in The Netherlands, may be that the distance is short and easy to complete in a few hours. In the same way, few of the marked Scanian Greylag Geese used France as a staging area on the migration from The Netherlands to the wintering areas in Spain ). The distance from The Netherlands to south-western Spain was usually covered by a non-stop flight (Persson 1993(Persson , 1994b. The changes in the migration and wintering habits of the Greylag Geese from south-west Scania shown in the analysis of neck-banded individuals in this contribution are also reflected in the results from the ordinary goose counts (Nilsson 2013).
Before year 2000, hardly any Greylag Geese stayed in Scania over the winter. However, after 2000 the numbers wintering in southern Sweden increased markedly with a peak count in 2008 of about 50 000 Greylag Geese. This is about 25% of the total count of the species in the special survey in September the preceding autumn (Nilsson 2013, Nilsson & Haas 2016. The departure of the geese of the neck-banded population leaving the country was later during the study period. Thus, both October and November counts increased over the years and a larger proportion of the Greylag Geese counted in September stayed in October and November. This increase in the number of geese staying longer in the autumn was noted before the start of the wintering tradition of Greylag Geese in Sweden (Nilsson 2013). On the European scale, these changes in winter distribution of Greylag Geese have been discussed by Ramo et al. (2015).
This type of northern range shift has been reported for other goose species both from Sweden and from other countries. In Spain, the Tundra Bean Goose Anser fabalis rossicus declined from perhaps 200 000 birds in the 1930s to none in the mid-1990s (Persson & Urdiales 1995). Whether this was due to deterioration of conditions in Spain, or improvements at wintering areas closer to breeding areas (or a combination of both) is not known. In North America, there are several examples of how improved conditions along the migration routes have shifted the winter distribution northwards in both the Canada Goose Branta canadensis and the Snow Goose Anser caerulescens (for references, see Elmberg et al. 2014).
In Sweden, the Taiga Bean Goose Anser fabalis fabalis breeding in the north has also changed migration and wintering habits (Nilsson 1984, 2011, 2013, Nilsson & Persson 1984 staging further north in Sweden during autumn and to a large extent wintering in south Sweden and/or Denmark and not migrating to The Netherlands and Western Germany any more. The timing of migration has also changed and the Taiga Bean Geese arrive earlier to the spring staging areas in northern Sweden nowadays compared to in the late seventies (Nilsson & Persson 1984, Nilsson unpublished). In one of the most important staging and wintering areas for this taxon, NE Scania, the geese nowadays arrive much later in autumn and leave much earlier in spring than some decades ago (Kampe-Persson 2014).
The changes in winter distribution and migration pattern in the Greylag Goose have been quite fast occurring over a short time period. This is, how-ever, not unique for the Greylag Goose but has been found in several other goose species, showing the large flexibility in the migration behaviour of geese. In Sweden, the number of autumn staging Barnacle Geese Branta leucopsis has increased markedly in the last few years. Actually, there has been a change in the entire migration strategy for the Russian population of the species related to changes in the availability of staging sites in the Baltic (Eichhorn et al. 2009). Other fast changes in the migration pattern seen in Sweden is the occurrence of larger numbers of Tundra Bean Geese in recent years (Nilsson 2013, Kampe-Persson 2014 and a higher number of staging Pink-footed Geese Anser brachyrhynchus in south-central Sweden in autumn and spring (Nilsson 2013).
Several duck species have also changed the winter distribution northwards. In Sweden, Nilsson & Haas (2016) found such changes in the winter distribution of several species as an effect of milder winters. Lehikoinen et al. (2013) and Pavon-Jordan et al. (2015) found similar changes when comparing duck census data from a number of countries.
Besides climate change, changes in the availability of suitable crops for feeding can be a factor contributing to the changes in winter distribution and migration habits for the geese. For the Greylag Geese, the later migration from Sweden coincided with changes in the feeding habits of the geese in Scania (Nilsson & Persson 1998, Nilsson & Kampe-Persson 2013, as the geese starting to utilize fields with leftover sugar beet in the autumn. The agricultural landscape of South Sweden nowadays offers the geese a succession of good feeding opportunities over the entire season. This change in the availability of food in combination with milder winters and less snow cover creates better possibilities for the geese to stay in South Sweden over the winter. Changes in agriculture and suitable crops for feeding was probably also an important factor behind the change in autumn staging habits for the Taiga Bean Geese in Sweden.
Changes in the migration habits of the Greylag Geese from SW Scania over the years were not only found in the regular migrations to and from the wintering areas. Young, non-breeding geese migrate to traditional moulting sites in early summer (Salomonsen 1968). In the early years of the study, the Greylag Geese from Scania migrated to Flevoland in The Netherlands for moulting (Andersson et al. 2001) but as the population of Greylag Geese in Western Europe increased and the moulting site in Flevoland reached its carrying capacity (Zijlstra et al. 1991), Greylag Geese from South Sweden started to change their moult migration. Eventually, a large proportion of Greylag Geese from Scania were found moulting on the neighbouring island of Saltholm in the Øresund between Sweden and Denmark (Fox et al. 1995. When the moulting tradition for Scanian Greylag Geese started on Saltholm is not known because the systematic work on the island only started in 1994, when a number of marked geese were found here.