Christensen.indd 109Christensen & Falk 2001: Polar Research 20(1), 109–114 Status of the common eider breeding in the municipality of Avanersuaq (Thule), north-west Greenland Kim Diget Christensen & Knud Falk Late in the summers of 1997 and 1998 surveys of common eider (Somate- ria mollissima) colonies were conducted throughout Avanersuaq Munici- pality in north-west Greenland. Although old information from eider col- onies was available, these surveys provided the first thorough census of almost all colonies in the district, thereby improving the baseline data for assessing future population changes. The surveys were based on nest counts, and all but two colonies in the district were counted. In total, 3800 nests were counted, and an educated guess for the total common eider population in Avanersuaq Municipality, including the two inaccessible colonies, would be around 5000 pairs. Average clutch size was 3.74, simi- lar to or higher than in other areas of Greenland and high Arctic Canada, suggesting favourable conditions for eiders in the survey area. Com- parison with older data suggests that the breeding population in Avaner- suaq is stable. This observation contrasts with the declines observed in other parts of Greenland. Further studies into possible population-spe- cific migration routes and wintering areas as well as the direct and indi- rect effects of the intensive winter hunting in south-west Greenland are warranted. K. D. Christensen & K. Falk, Ornis Consult Ltd., Vesterbrogade 140A, 2, DK-1620 Copenhagen V, Denmark; K. Falk currently at Danish Polar Center, Strandgade 100H, DK-1401 Copenhagen K, Denmark. In the Arctic, hunting still provides a significant portion of the food for a household. In particular, seabird species are harvested widely, often causing national and international management concerns (CAFF 1996, 1997). In Greenland, considerable effort has been put into surveying breeding popu- lations of the most important seabird quarry, the thick-billed murre (Uria lomvia) (Kampp et al. 1994; Falk & Kampp 1997), while little is known of the current status of the second most hunted species, the common eider (Somateria mollis- sima). Approximately 80 000 common eiders are reported to be shot annually in Greenland, while the real figure might be much higher (Frich 1997). Available information suggests that a consider- able proportion of the harvested population is wintering birds that breed in Canada (Nunavut), but due to lack of data, the relative contribution of birds from colonies in Greenland cannot be assessed. The breeding eider population in west Greenland has been reduced considerably during the 20th century (Salomonsen 1967), and although only a few recent surveys have been conducted, the downward trend appears to continue in the most populated areas of west Greenland (Boert- mann et al. 1996). A substantial breeding pop- ulation decline has recently been found in an Research Note 110 Status of the common eider in Avanersuaq, north-west Greenland area with low human population density in west Greenland (Frich et al. 1997). Evidence of declin- ing eider populations (common and king eiders [S. spectabilis]) have also been reported from Arctic Canada (Cooch 1986; Reed & Erskine 1986; Dickson 1997; Robertson & Gilchrist 1998). The high hunting toll in certain areas, combined with the scant information on population status, has led managers to call for intensified surveys and improved population monitoring of eiders in the Arctic (CAFF 1997). Although not known for certain, it is highly unlikely that the southern parts of west Green- land—where most people live—should hold large, undiscovered eider colonies. However, according to existing information collated in the Greenland seabird colony database (DMU-AM & OC 1999), thinly populated north-west Greenland might hold large colonies. Therefore, field surveys were con- ducted in the Avanersuaq (Thule) Municipality in 1997 and 1998 to provide up-to-date informa- tion on the status of the common eider for use in resource management initiatives (CAFF 1997), and for establishing a basis for future population monitoring. Materials and methods Survey area All available data on colonial seabird populations in Greenland are compiled in the Greenland sea- bird colony database, which was used to localize existing eider colonies between 76º and 79º N in the Avanersuaq Municipality, north-west Green- land. A total of 14 eider colonies, all of which were on small groups of islands, were registered in the database for that region. In addition, detailed maps and aerial photographs were studied to identify other small islands that could be poten- tial breeding sites for eiders. Compared to other parts of Greenland, the survey area had only few small islands, and no archipelagos, so the poten- tial breeding sites for eiders are limited. This made surveys relatively straightforward, except for logistical constraints imposed by weather, ice and the distances involved. By means of outboard powered boats we attempted to visit all known colonies and poten- tial sites late in, or just after, the incubation period. In the survey area, sea ice breaks up only a few weeks before the eiders leave the colonies, providing only a narrow time window to reach the colonies by boat. Therefore, it was necessary to split the survey between two years. In 1997 (12–27 August) the effort was concentrated on the north- ernmost area, and in 1998 (29 July–19 August) the southernmost colonies were surveyed. Even so, the northernmost sites were inaccessible due to heavy drift ice in the 1997 survey period. Survey methods The survey was based on counts of nests. On small islands one or two observers counted all nests. To avoid double counting, each nest was marked with a piece of sugar (to dissolve with the next rain). On the bigger islands, flat areas with some vegetation were searched thoroughly; while steep slopes and areas devoid of vegetation were also searched since they were found to contain a few scattered nests. In one case, a relatively large island (in Three Sister Bees group) with many scattered nests, the survey was based on line transects. The nests were counted on two approximately 370 m transects starting on the lower north-western part and ending on the higher (15 m) south-eastern part of the island; these Fig. 1. Map of the known and surveyed sites in the Avanersuaq area, north-west Greenland (see numbers in Table 1); white areas on land indicate ice cap and glaciers. 111Christensen & Falk 2001: Polar Research 20(1), 109–114 transects were located within the area with eider nests (approximately 50 % of the island). When a nest was found, the perpendicular distance from the transect to the nest was measured. The number of nests/hectare was estimated employing the PC software “Distance” (Laake et al. 1993) using the half normal model. Common eiders make a depression in the ground for their nest, which is first lined with plant materials and then with down (Cramp 1977). Many nests are reused, leading to formation of a depression that probably remains for many years under high Arctic climatic conditions, with little rain and frozen ground most of the year. In some parts of the area local inhabitants collect eider down after the birds leave their nests. During the survey, many of the nests were found empty of down due to the collection, but still with quite fresh plant material in the bottom. Hence, in the survey we distinguished between: i) nests with eggs and/or young; ii) nests with down; iii) nests with fresh plant material; and iv) old nest cups. In the third category it is possible that new plant material could have been deposited by birds which were merely prospecting for nest sites. But since we only found category iii nests on islands, where we knew down collection had taken place, we assumed that these nests had been used within the breeding season. The first three categories were therefore considered as representing breed- ing pairs while the latter category was excluded. For nests with eggs, the clutch size was noted. Results In Avanersuaq Municipality, 12 of the 14 known sites were surveyed in the two years. Some sites comprised several islands, so the total number of islands surveyed added up to 21 (in the data- base, it is not consistent whether groups of islands are pooled into one record, or data from each island recorded separately). The two northern- most known sites could not be reached because of sea ice. The location of all colonies is shown in Fig. 1, and the number of nests counted in each Colony Site name No. of islands No. of islands Survey year No. of pairs Comments from older sources code a surveyed with eiders present 78001 Pikulleq 0 - - ? • “large colony” (1936)b (MacGary/Eider islands) • 5000 eggs taken annually (sites 78001 + 78002) (1980s)c • <2500 eggs taken annually (sites 78001 + 78002) (1980s)d 78002 Knorr Islands 0 - - ? • “large colony” (1936)b • see also comments above 78003 Sutherland Island 1 0 1997 0 • “large colony” (1936)b 77002 Hakluyt Island 1 1 1997-98 9 no comments (new site, this study) 76016 Qeqertaaraq 1 1 1997 105 • 50 - 66 (1987)e 76002-6 Booth Sound 7 5 1997 866 • 220 (1968)f (several islands) 76022 Three Sisters Bees 3 3 1998 349 • “large colony” (1936)b • “colony” (1988)c 76001 Manson Islands 4 3 1998 652 • 100 (1985)g 76019 Igannaq 1 1 1998 255 • 400 (1988)h (Dalrymple Rock) 76021 Qeqertaarsuit 1 1 1998 1572 • “large colony” (1936)b (Ederfugleøer) • 1000 (1988)h 76018 Nuulliarsuit 1 1 1998 24 no comments (new site, this study 76031 Igannaq 1 0 1998 0 • “large colony” (1936)b (Conical Rock) Total 21 16 3832 a Colony code according to the Greenland seabird colony database (DMU-AM & OC 1999) b Salomonsen (1950) c A. Rosing-Asvid (pers. comm.) d Born (1987) e K. Kampp (pers. comm.) f Salomonsen (unpubl. notes) g L. Wellander in Vaughan (1988) h M. Kofoed (pers. comm.) Table 1. Results of a common eider colony survey in Avanersuaq (Thule) Municipality, north-west Greenland, 1997-98. 112 Status of the common eider in Avanersuaq, north-west Greenland colony (or group of islands) summarized in Table 1. The 16 islands with evidence of breeding common eiders contained a total of about 3800 nests (Table 1). Three new breeding sites were identified: in the Manson Islands group (no. 76001 in Table 1) a small islet with 522 nests was found; a new colony (76018) with 24 nests was found near Thule Air Base (Pituffik); and on the large Hakluyt Island (77002) a few scattered nests were found. At two of the “known” sites no nests were found. According to local hunters, eiders breed on Sutherland Island in some years, but the only apparently suitable nesting area we found con- tained only a single possible old nest cup, so the island cannot hold a major colony. On Igannaq (76031), there was no suitable eider nesting habi- tat, so it seems unlikely that there has ever been an eider colony here; the site has probably been included in the database due to the observer’s confusion with another island of a similar name. The largest colonies were usually found in areas with vegetation on small (2 - 5 ha), low-lying islands; this was the case at the Booth Sound islands, Qeqertaarsuit, Manson Islands (522 of the 652 nests on a small island), and Three Sis- ters Bees (341 of the 349 nests on a small island). On Three Sisters Bees, the estimate of 341 pairs was based on counts of 46 nests using the line transect method. The larger islands (> 100 ha) usually only held relatively few nests, such as on Igannaq and in the Manson Islands (107 nests of 652). On the larger islands the eiders avoided steep bare areas and seemed to concentrate in areas with vegetation close to the shore. Most of the nests were found empty because the survey was conducted late in the season. In 1997, 59 % of the nests contained young, eggs or down and 41 % contained only plant material, but down collection had taken place on the island and therefore the nests with only plant material were assumed to represent breeding pairs. In 1998, none of the nests contained only plant material-- all nests contained young, eggs or down. This was probably because the 1998 survey was conducted earlier in the season, before down had been col- lected. In the Manson Islands, the average clutch size was 3.74 (s.d. = 1.53) among 239 nests with eggs. In three cases, the eiders were found nesting close to other bird species, potentially assisting in fending off mammalian predators such as the Arctic fox (Alopex lagopus). In the Manson Islands, eiders nested near greater snow geese (Anser c. caerulescens), a species that has been expanding its range in the Avanersuaq area in recent years (Boertmann 1994); on Three Sister Bees, Arctic terns (Sterna paradisaea) aggres- sively protected the main breeding island, and on Hakluyt Island the few eider nests were found near a colony of glaucous gulls (Larus hyper- boreus). Discussion The survey included all but two of the known breeding sites for common eiders in the Avaner- suaq Municipality, and revealed a total popula- tion of at least 3800 pairs. Apart from the two northernmost colonies (see below), the coverage was fairly complete. Eiders normally show high site tenacity, usually returning to previously used islands, and sometimes even the same nest cups (Reed 1975; Parker & Mehlum 1991). However, they may move depending on whether the particu- lar nesting island is connected by ice to the main- land early in the season, when the birds select their nest site based on “security concerns” related to access of predators, especially the Arctic fox (Mehlum 1991a; Parker & Mehlum 1991). Our survey was conducted over two seasons, with no repeated counts in both years at any sites, so it cannot be ruled out that parts of the “regional population” may have been overlooked or dou- ble-counted due to this approach. Nevertheless, movements from year to year would normally be short (Reed 1975; Swennen 1990), so since our surveys in 1997 and 1998 covered rather distinct areas, we consider that the possible bias caused by inter-annual movements was minimal. In Svalbard, the number of breeding pairs has been shown to be lower in years with late ice break-up (Mehlum 1991b). In our survey area, ice break-up was normal in the survey years, so the results should not be significantly affected by unusual ice conditions. The mean clutch size of 3.74 in Avanersuaq was higher than values (means between 3.12 and 3.44) found for the same subspecies (S. m. borealis) in areas of Baffin and Ellesmere islands in east- ern Canada, and higher than in Upernavik, north- west Greenland (3.31), and similar to findings of 3.82 from south-west Greenland (68° N) in 1997 and 1954 (Joensen & Preuss 1972; Prach et al. 1986; Frich et al. 1997; Salomonsen, 113Christensen & Falk 2001: Polar Research 20(1), 109–114 unpubl. notes). The eider populations of low Arctic Hudson Bay in Canada, however, lay large clutches, ranging between 4.0 and 5.6 eggs, pos- sibly because the birds of this subspecies (S. m. sedentaria) are resident so they may have more energy reserves available for egg produc- tion (Robertson & Gilchrist 1998; Robertson et al. 2001 [this issue]). Since common eiders show a seasonal decline in clutch size (Mehlum 1991b; Robertson 1995; Coulson 1999) our result might partly be biased by the late survey date, when only late, presumably smaller, clutches would be recorded. Even so, the clutch size was relatively large, suggesting that conditions for eiders in the area may have been favourable. No data are available from the two known (northern) colonies we could not survey, but archaeologists working in the area in 1996 and 1997 reported active nests (but no numbers) from MacGary Island (Bo Bendix, University of Copenhagen, pers. comm.). Before the late 1980s, local hunters have reported that up to 5000 eggs may have been harvested in some years (DMU-AM & OC 1999). Although such informa- tion is hard to evaluate, egg collection must have been substantial in some years. As the common eiders in the area lay 3 - 5 eggs, the reported egg collection suggests that more than 1000 nests may have occurred there. Hence, combined with our counts of 3800 nests, an educated guess of the total common eider population in Avanersuaq would be around 5000 pairs. Most of the records in the database on eiders in this area contain secondhand information or very crude estimates rather than actual counts, so we have a weak basis for evaluating popula- tion trends. The best available information stems from the group of islands in Booth Sound, where counts from 1968 indicated a total population of 220 nests, and brief visits in 1987 indicated only about a few hundred nests. Comparing these fig- ures to the 866 nests actually counted in 1997 sug- gests a real increase in this particular sub-area. In general, the results of the 1997-98 survey indi- cated colonies of similar size, or larger, than pre- vious information suggested—disregarding the dubious cases where “known” colonies could not be confirmed. We therefore consider that the pop- ulation in Avanersuaq may at least have remained stable, and avoided a dramatic decline like that seen in some areas further south in Greenland (Boertmann et al. 1996; Frich et al. 1997). To establish a firm platform for monitoring popula- tion trends would require several years of data to account for inter-annual variability— a task hard to accomplish in this large and remote region. But this survey has improved the quantitative information available assessing major population changes in the region. The possibly stable population in north-west Greenland raises the question whether the inten- sive winter hunting in south-west Greenland is actually the main reason for the declining popu- lations in both Greenland and Canada, as is cur- rently assumed. If eiders from different regions mix in mutual staging and wintering grounds with high hunting impact, a widespread—not a localized—population decline would be expected. Local declines, however, would be expected if a) eiders from different regions move to separate staging and wintering grounds with different hunting levels, or b) over-harvesting and/or dis- turbance occur mainly in the breeding areas. To identify migration routes and wintering grounds, satellite telemetry on females from breeding sites in low and high Arctic Greenland and eastern Canada would be desirable, as would be detailed studies on the direct and indirect impact of hunt- ing in the wintering areas. Acknowledgements.—We thank Kurt Thomsen and Kristian Eipe for safe dinghy navigation throughout the district, and Jonas Teilmann and Erik W. Born for assistance in the field- work. Jan Thrysøe and the Danish Liaison Officer at Thule Air Base kindly facilitated our transit stays there. The project was funded through a grant to Ornis Consult Ltd. from the DANCEA programme of the Danish Ministry of Environ- ment. Comments from Gregory J. Robertson and one anony- mous referee helped us improve this paper. References Boertmann, D. 1994: An annotated checklist to the birds of Greenland. Medd. Grønl. Biosci. 38, 1–63. Boertmann, D., Mosbech, A., Falk, K. & Kampp, K. 1996: Seabird colonies in western Greenland (60°-79°30’ N. lat.). NERI Technical Rep. 170. Copenhagen: National Environ- mental Research Institute. Born, E. W. 1987: Aspects of present-day maritime subsist- ence hunting in the Thule area, northwest Greenland. In L. Hacquebord & R. Vaughan (eds.): Between Greenland and America: cross-cultural contacts and the environment in the Baffin Bay area. Pp. 109–132. Groningen, The Nether- lands: University of Groningen. CAFF 1996: International murre conservation strategy and action plan. Arctic Environmental protection strategy. Akureyri, Iceland: Conservation of Arctic Flora and Fauna International Secretariat. CAFF 1997: Circumpolar eider conservation strategy and 114 Status of the common eider in Avanersuaq, north-west Greenland action plan. Akureyri, Iceland: Conservation of Arctic Flora and Fauna International Secretariat. Cooch, F. G. 1986: The numbers of nesting northern eiders in the West Foxe Islands, NWT, in 1956 and 1976. In A. Reed (ed.): Eider ducks in Canada. Can. Wildl. Serv. Rep. Ser. 47, 114–118. Ottawa. Coulson, J. C. 1999: Variation in clutch size of the common eider: a study based on 41 breeding seasons on Coquet Island, Northumberland, England. Waterbirds 22, 225–238. Cramp, S. (ed.) 1977: Handbook of the birds of Europe, the Middle East and North Africa. Vol. I. The birds of the west- ern palearctic. Oxford: Oxford University Press. Dickson, D. L. (ed.) 1997: King and common eiders of west- ern Canada. Can. Wildl. Serv. Occas. Pap. 94. Ottawa: Canadian Wildlife Service, Environment Canada. DMU-AM & OC 1999: Database over Grønlands havfugle- kolonier. (Greenland seabird colony database.) Copenha- gen: National Environmental Research Institute (Dept. of Arctic Environment) and Ornis Consult Ltd. Falk, K. & Kampp, K. 1997: A manual for monitoring thick- billed murre populations in Greenland. Tech. Rep. 8. Nuuk: Greenland Institute of Natural Resources. Frich, A. S. 1997: Ederfuglefangst i Grønland 1993. (The bag of common eiders in Greenland 1993.) Tech. Rep. 9. Nuuk: Greenland Institute of Natural Resources. Frich, A. S., Christensen, K. D. & Falk, K. 1997: Ederfugle- optællinger i Kangaatsiaq og Avanersuaq 1997. (Common eider counts in Kangaatsiaq and Avanersuaq.) Tech. Rep. 10. Nuuk: Greenland Institute of Natural Resources. Joensen, A. H. & Preuss, N. O. 1972: Report on the orni- thological expedition to northwest Greenland 1965. Medd. Grønl. 191(5), 1–55. Kampp, K., Nettleship, D. N. & Evans, P. G. H. 1994: Thick- billed murres of Greenland: status and prospects. In D. N. Nettleship et al. (eds): Seabirds on islands. Threats, case studies and action plans. BirdLife Conserv. Ser. 1. Pp. 133–154. Cambridge: BirdLife International. Laake, J. L., Buckland, S. T., Anderson, D. R. & Burnham, K. P. 1993: DISTANCE user’s guide. Ver. 2.0. Fort Collins, CO: Colorado Cooperative Fish & Wildlife Research Unit, Colorado State University. Mehlum, F. 1991a: Egg predation in a breeding colony of the common eider Somateria mollissima in Kongsfjorden, Svalbard. Nor. Polarinst. Skr. 195, 37–45. Mehlum, F. 1991b: Breeding population size of the common eider Somateria mollissima in Kongsfjorden, Svalbard, 1881-1987. Nor. Polarinst. Skr. 195, 21–29. Parker, H. & Mehlum, F. 1991: Influence of sea-ice on nest- ing density in the common eider Somateria mollissima in Svalbard. Nor. Polarinst. Skr. 195, 31–36. Prach, R. W., Smith, A. R. & Dzubin, A. 1986: Nesting of the common eider near the Hell Gate–Cardigan Strait polynya, 1980-1981. In A. Reed (ed.): Eider ducks in Canada. Can. Wildl. Serv. Rep. Ser. 47, 127–135. Ottawa. Reed, A. 1975: Migration, homing, and mortality of breed- ing female eiders Somateria mollissima dresseri of the St. Lawrence estuary, Quebec. Ornis Scand. 6, 41–47. Reed, A. & Erskine, A. J. 1986: Populations of the common eiders in the eastern North America: their size and status. In A. Reed (ed.): Eider ducks in Canada. Can. Wildl. Serv. Rep. Ser. 47, 156–162. Ottawa. Robertson, G. J. 1995: Annual variation in common eider egg size: effects of temperature, clutch size, laying date, and laying sequence. Can. J. Zool. 73, 1579–1587. Robertson, G. J. & Gilchrist, H. G. 1998: Evidence of popula- tion declines among common eiders breeding in the Belcher Islands, Northwest Territories. Arctic 51, 378–385. Robertson, G. J., Reed, A. & Gilchrist, H. G. 2001: Clutch, egg and body size variation among common eiders breed- ing in Hudson Bay, Canada. Polar Res. 20, 85–94. Salomonsen, F. unpubl. notebooks. University of Copenha- gen. Salomonsen, F. 1950: The birds of Greenland. Copenhagen: Munksgaard. Salomonsen, F. 1967: Fuglene på Grønland. (The birds of Greenland.) Copenhagen: Rhodos. Swennen, C. 1990: Dispersal and migratory movements of eiders Somateria mollissima breeding in the Netherlands. Ornis Scand. 21, 17–27. Vaughan, R. 1988: Birds of the Thule district, northwest Greenland. Arctic 51, 53–58.