Seabird Breeding Colonies in East and North Greenland: A Baseline

This paper presents the results of a number of aircraftand boat-based surveys for seabird breeding colonies in East and North Greenland carried out in the period 2003 to 2018 and gives the first comprehensive overview of the distribution and size of the seabird breeding colonies in this remote and mainly uninhabited region. Seventeen seabird species breed in approximately 800 sites distributed very unevenly along the coasts, with high concentrations at the polynyas and long stretches with very few breeding seabirds. Climate changes are in full progress in East and North Greenland, especially affecting the sea ice regime, and seabirds are expected to respond to these changes in different ways. For example, since the 1980s, Common Eiders (Somateria mollissima) have extended their breeding range more than two latitudinal degrees towards the north, now reaching the northernmost land on Earth. Lesser Black-backed Gulls (Larus fuscus) and Great Cormorants (Phalacrocorax carbo) have immigrated, and Sabine’s Gulls (Xema sabini) have increased and extended their range. Besides presenting survey results, this report may also serve as a baseline for future studies of the abundance of breeding seabirds in East and North Greenland.


INTRODUCTION
Knowledge of the distribution and population sizes of breeding seabirds is important for management and protection not only of coastal areas, but also for the adjacent open sea exploited by foraging seabirds (Thaxter et al., 2012;Gaston et al., 2013). Such knowledge of seabird distribution and population size has been widely used for the regulation of petroleum exploration in West Greenland, where the knowledge base is extensive, largely because of the efforts of Finn Salomonsen (1979a, b) and many surveys carried out in the 1990s (Boertmann et al., 1996). In East and North Greenland, similar knowledge is much less comprehensive, mainly due to logistical constraints. Firstly, the infrastructure is very limited in this region, hosting only 6% of the Greenland population (totalling 56 000 inhabitants in 2014) in only two small areas (around Tasiilaq at 65˚37′ N and Ittoqqortoormiit at 70˚30′ N). Secondly, the coasts have been, until recently, difficult to access because of the continuous presence of drift ice, even during the short summer season.
This lack of knowledge has been addressed over the past 20 years with increased research activity. This paper presents an outline of the results obtained on the distribution and abundance of breeding colonial seabirds in the region covering East and North Greenland (Figs. 1,2). Our aim is to provide a reliable baseline that can be used to assess future changes in the seabird abundance and distribution in this region.
Regular monitoring of seabird colonies takes place only at the Thick-billed Murre (Uria lomvia) colonies at Kap Brewster and Raffles Ø near the town of Ittoqqortoormiit, known also as Scoresbysund (Falk and Kampp, 1998;Merkel, 2016;Egevang et al., 2017). However, the monitoring program at the Zackenberg research facility also includes occasional surveying of seabird colonies (Arctic Tern [Sterna paradisaea], Sabine's Gull [Xema sabini], and Common Eider [Somateria mollissima]) near the site; that is, at the Daneborg military station and on the island of Sandøen (Hansen et al., 2007).

This account covers the coasts of East and North
Greenland from Ikerasassuaq (Prins Christian Sund) in the south (60˚04′ N, 43˚13′ W) to the Humboldt Glacier in the northwest (79˚35′ N, 67˚15′ W). The area comprises the former municipalities of Tasiilaq and Ittoqqortoormiit (today part of the Sermersooq Municipality), the entire national park in East and North Greenland, and Washington Land. Hence, it includes the northernmost coasts in the world at 83˚40′ N.
In 2005, the Greenland government planned to launch a round of licensing for petroleum exploration in the waters off Northeast Greenland. This plan initiated a number of environmental background studies with the aim to provide data for the preparation of a strategic environmental impact assessment, which was carried out by the Danish Centre for Energy and Environment (DCE) at Aarhus University and the Greenland Institute of Natural Resources (Boertmann and Mosbech, 2012). These studies included two aerial surveys for seabirds in the coastal environment Boertmann and Nielsen, 2010). In 2008, the Greenland Institute of Natural Resources also carried out an aerial survey, primarily of Common Eider colonies in Southeast Greenland, to acquire data for the management of the Greenland eider populations . In addition, boat-based surveys of colonial seabirds were launched in 2014 and 2016, which covered the mainly unexplored (regarding seabirds) coast of Southeast Greenland (Boertmann andRosing-Asvid, 2014, 2017). All data collected by DCE and the Greenland Institute of Natural Resources is compiled in the Greenland Seabird Colony Register (Boertmann et al., 2010b). Table 1 gives an overview of the surveys.
Besides these studies supported by the Greenland and Danish governments, the French non-governmental organization, Groupe de Recherche en Écologie Arctique (GREA), has since 1979 performed almost annual ecological studies, including surveys of breeding seabirds, along the coasts of East Greenland between 69˚ N and 77˚ N. Since 2003, GREA also visited and surveyed different areas between 80˚ N and 83˚40′ N (Kap Morris Jesup), including regular visits to Station Nord (81˚37′ N) and several visits to the Henrik Krøyer Holme islands (80˚39′N) in the biologically rich Northeast Water polynya (Gilg et al., 2003(Gilg et al., , 2011. The most comprehensive colony surveys by GREA occurred in 2004 and 2015, when the schooner Tara (https://oceans.taraexpeditions.org) was used as an observation platform (Gilg et al., 2005). To date, with the exception of their specific research on Ivory Gulls (Pagophila eburnea) (Gilg et al., 2009(Gilg et al., , 2010Yannic et al., 2016), none of the GREA seabird surveys has ever been published.
Data of the 17 colonial seabird species breeding in East and North Greenland from the Greenland Seabird Colony Register and GREA were combined. The selected species are all colonial breeders in the region, except for the Parasitic Jaeger (Stercorarius parasiticus), but we have included it here because it does breed in colonies in other areas of the North Atlantic (Cramp and Simmons, 1982). Many of the other species, such as the Glaucous Gull (Larus hyperboreus), Common Eider, and Arctic Tern, also breed in solitary pairs and have been included here. Typical non-colonial species, such as Red-breasted Merganser (Mergus serrator) and Longtailed Duck (Clangula hyemalis), were omitted.
The survey methods used from the different observation platforms-ships and aircraft-are described in the post-survey reports (Gilg et al., 2005;Boertmann and Nielsen, 2010;Merkel et al., 2010;Boertmann andRosing-Asvid, 2014, 2017). Coasts and potential breeding sites were surveyed by flying over or sailing parallel to the coast ("total counts," see Laursen et al., 2008) and in some cases, specific sites were sought out and surveyed. Many colonies were also photographed for later estimation of breeding seabird numbers. Both during the aerial and ship-based surveys, at least two observers were actively searching the coasts.
Survey results for species originally recorded as individuals are converted to pairs by applying a conversion factor of 0.7 (Birkhead, 1978;Harris, 1989), as done in Canada in a comparable seabird status paper (Gaston et al., 2012). This conversion was not done for Black Guillemot (Cepphus grylle) (see below). Unless otherwise stated, all bird numbers in this paper refer to the most recent survey.
The estimates of the total seabird populations in East and North Greenland are based on the sum of the most recent survey results, including solitary pairs. Correction for the fraction of the unsurveyed coasts is also considered for widespread species such as the Black Guillemot and the Glaucous Gull, and expert judgement is applied to some degree. The fraction of unsurveyed coasts, excluding glacier coasts, were calculated from measurements of coastlines in GIS (MapInfo).
Since most of the colonies have only been described recently and visited once, assessing population trends is difficult for this region. Furthermore, successive surveys when available are sometimes difficult to compare because of changing observation conditions (e.g., in platform typeship vs. aircraft). However, as mentioned above, dedicated monitoring takes place at two colonies (Merkel, 2016) for which population trends will be published separately, and a more comprehensive analysis of population trends in the region covered by the GREA surveys is planned.
Site names are usually given either in Greenlandic or Danish following toponymy of available topographic maps. For a few more significant and internationally recognized sites, English names are used.

RESULTS
Overall, 71% of the coastline of East and North Greenland has been surveyed since 2003, with some regional variation (Table 2). Along these coasts, 799 breeding sites for 17 seabird species are recorded (Tables 3 and 4, Figs. 1, 2). From these sites, the combined databases contain 2191 records. Depending on its original description, we sometimes define a colony as including sub-colonies on nearby islands; in other cases, nearby island colonies are defined as separate colonies. The new data collected since 2003 more than sextupled the number of known seabird breeding sites in the covered region (Table 3). Figures 3 -12 present species maps with the location of all known seabird breeding colonies and sites in the covered region.

Species Accounts
Northern Fulmar (Fulmarus glacialis): Ten sites with breeding (or presumed breeding) Northern Fulmars are known (Fig. 3). However, at three of these sites (Dunholm, Hvalros Ø, Home Forland), Northern Fulmars have only been observed once on land despite several subsequent visits, and it is doubtful if these sites represent regular breeding sites. All the stable colonies are located on steep cliffs and at the polynyas (i.e., at the Northeast Water polynya and the mouth of Scoresby Sound). The colonies at the Northeast Water polynya are distributed over several cliffs and were discovered in 1907 during the Danmark Ekspedition (Manniche, 1910). Surveyed in 1993, the estimated number of breeding pairs at these colonies was  then 1475 Møller, 1995, 1997). All the other colonies are small with less than 150 pairs, and the total number in the region probably does not exceed 2000 pairs. Great Cormorant (Phalacrocorax carbo): Only three breeding colonies are known, and all are from cliffs in the Sermilik (Isortoq) area west of Tasiilaq (Fig. 3). An additional possible colony was reported by a local source around 2010; however, during the 2014 survey, it was determined that the location was a roosting site without nests. Only one of the colonies has been surveyed twice, and it held 22 nests in 2005 and 12 in 2014. The total population in the region apparently numbers fewer than 50 pairs. According to local sources, the colonies were established in the period 2000 -05, and there may be more colonies in the region. This small population is very isolated; the nearest breeding populations are located in Iceland (600 km) and in West Greenland (1500 km apart; if following the shortest distance over water).
Common Eider (Somateria mollissima): In East Greenland, Common Eiders usually breed in small colonies on low islands, but solitary nesting eiders can also be found on both islands and infrequently on mainland coasts. Scattered colonies have been located all along the coasts of the region from 61˚30′ N in the south to northernmost Greenland and in Washington Land in the west, where in 1999 Bennike and Feilberg (2004) observed single females with chicks, and Andreassen (1999) reported a small colony (Fig. 4). There are, however, long stretches of coastline where breeding has not yet been observed, such as the Blosseville Kyst and in most of North Greenland. In total, 116 colony sites are known; in addition to these are the solitary breeders, of which known sites are shown on Figure 4. The largest colonies are found on the mainland at two manned stations in Northeast Greenland (Danmarkshavn and Daneborg), where the birds place their nests near tethered sledge dogs (Meltofte, 1978). More than 3100 pairs have been reported at Daneborg (result of photocounts in 2012; O. Gilg, unpubl. data) and up to 400 at Danmarkshavn (U. Capito, pers. comm. 2009), while all other colonies range between 2 and 300 pairs.
In 2008, Merkel et al. (2010) estimated the population of breeding eiders in Southeast Greenland at 1600 -3200 pairs. Based on an aerial survey of pre-breeding birds in Northeast Greenland in May 2008, the breeding population here was estimated to be at least 13 000 pairs . The combined most recent survey results from all the colonies report many fewer (6000 pairs), which indicates many unknown breeding sites or perhaps a large non-breeding segment of the population.
Parasitic Jaeger (Stercorarius parasiticus): The Parasitic Jaeger is not strictly colonial in the region covered by this paper, and many solitary breeding pairs remain unrecorded. In total, 25 sites with breeding jaegers in East Greenland and one from Washington Land in North Greenland are registered in the two databases. Two pairs were observed at two sites (one with a distance of more than 1 km between pairs) and three pairs at only one site (Fig. 5).
In East Greenland, single birds have been observed as far north as the Northeast Water polynya . The population in East and North Greenland is very small, even considering that many breeding sites remain unrecorded, and probably does not exceed 200 pairs. Sabine's Gull (Xema sabini): Sabine's Gulls are known from 34 breeding sites in the region covered by this report (Fig. 6). The southernmost site is located on the coast of Jameson Land and the northernmost site on the island north of Bliss Bugt at 83˚38′ N. In contrast to the situation in Arctic Canada and Alaska (Stenhouse et al., 2001;Norment et al., 2015), Sabine's Gulls usually breed in well-defined colonies in East and North Greenland, probably because most colonies are located on small islands . At the few mainland sites, the colonies seem less well defined. The sum of the most recent survey results adds up to 470 pairs, with the largest colonies at Kilen, Renskaeret (at Danmarkshavn), and Sandøen, with a maximum of 150 pairs at all three, although not in the same season. The colony size fluctuates widely from year to year, and often many non-breeding adults attend the colonies (Forchhammer and Maagaard, 1991).
Lesser Black-backed Gull (Larus fuscus): This species immigrated to West Greenland in the 1990s and quickly established a large and widespread population (Boertmann, 2008;Boertmann and Frederiksen, 2016). and 2015 surveys. These surveys revealed that the species was well-established in Southeast Greenland. Several small colonies (up to 35 pairs) and solitary pairs in colonies of other gulls were found between 64˚ N and 68˚ N and farther north on the coast, just south of the Scoresby Sound polynya, with the main part located in the Tasiilaq region (Fig. 5). Eighteen breeding sites are now known from the region, with a total of 50 -60 pairs.
Iceland Gull (Larus glaucoides): This gull species has a low-Arctic distribution in Greenland. In East Greenland, 58 breeding colonies are known with the northernmost at 69˚30′ N and the largest concentration in the Tasiilaq area (Fig. 6). These colonies are relatively small with up to 55 pairs. The most recent survey results for all colonies add up to 750 pairs, but there are likely many unrecorded colonies along the unsurveyed coasts in Southeast Greenland.
Glaucous Gull (Larus hyperboreus): In total, 403 breeding sites for Glaucous Gulls are recorded from East and North Greenland (Fig. 7). These include solitary pairs (n = 139), either alone or at colonies of other species (n = 48), and colonies with up to 100 pairs (average 8 pairs). Breeding sites are found along all coasts, as far north as 83˚30′ N in the east and in Washington Land in the west, with concentrations at Tasiilaq in Scoresby Sound, at the Sirius Water, in Dove Bugt, and to a lesser extent at the Northeast Water polynya. Besides at the Northeast Water polynya, very few colonies have been located north of 78˚ N. The most recent surveys add up to 2200 pairs. As many solitary pairs are overlooked and considerable numbers may hide in the unsurveyed coasts-especially in the inner fjords, this figure represents a minimum estimate of the population size.
Great Black-backed Gull (Larus marinus): The range of this species, which is widespread in West Greenland, is restricted to two regions in East Greenland (the Tasiilaq area and the Scoresby Sound Fjord complex; Fig. 6), and the number of breeding pairs is very low. In Tasiilaq, only one breeding site was located during the surveys in 2008, 2014, and 2016. From the Scoresby Sound Fjord area, the Great Black-backed Gull is known from six sites (four at colony sites with other species) in the area south of Kap Brewster and inside the fjord complex (Gilg et al., 2005;Boertmann and Frederiksen, 2016). The total population hardly exceeds 20 pairs. Ross's Gull (Rhodostethia rosea): This species has been found breeding, or presumed breeding, at three sites in Northeast Greenland-all at the Northeast Water polynya and its associated shore leads (Fig. 5). Two of the sites, Henrik Krøyer Holme and Kilen, can be described as seabird colonies where a single pair of Ross's Gulls was found among Arctic Terns, Sabine's Gulls, and (on Henrik Krøyer Holme) Ivory Gulls (Gilg et al., 2003;Egevang and Boertmann, 2008). The third breeding site included a solitary pair (Hjort, 1980). Their presence on the breeding sites is apparently not constant, as several researchers failed to find the species on the main site (Henrik Krøyer Holme) in recent years (Gilg et al., , 2011M. Frederiksen, pers. comm. 2018).
Black-legged Kittiwake (Rissa tridactyla): Thirty colony sites are known from East Greenland (Fig. 8); of these, 10 were without birds during the most recent surveys. The combined survey results add up to a total of 4800 pairs. The colony size varies from 3 to 850 pairs, with the two largest at the Scoresby Sound polynya and the Northeast Water polynya. Labansen et al. (2010)  local people) had only empty nests and had been without birds for several years.
Ivory Gull (Pagophila eburnea): In total, 43 colony sites are known from East and North Greenland (Fig. 9). The majority are located on the coasts of the Northeast Water polynya and adjacent shore leads, while some are also found inland on island nunataks (isolated rocks protruding above the glacier surface). Far from this stronghold, a few colonies have been found in the region between Tasiilaq and Scoresby Sound (Wright and Matthews, 1980;Merkel et al., 2010) and in the Dove Bugt region (Meltofte et al., 1981). Gilg et al. (2009) also list a number of unconfirmed breeding sites recorded in North Greenland before 1925, including two sites in Washington Land. Ivory Gull colonies were specifically sought out in 2007, 2008, and 2009, when a total of 32 colony sites (including several new ones) were surveyed Gilg et al., 2009;Boertmann and Nielsen, 2010;Merkel et al., 2010). The size of the colonies ranges from one pair to approximately 200 pairs. The sum of the most recent surveys adds up to 1300 pairs, while Gilg et al. (2009) calculated a breeding population of ca. 1800 birds, and commented that the total Greenland population could possibly be higher than 4000 birds (2000 pairs). Thirteen colony sites in the stronghold near Station Nord were surveyed in both 2008 and 2009. Of these, six sites were occupied both years, four only in 2008, two only in 2009, and one in neither of the two years. In the southern part of the distribution range, five sites (one new) were surveyed during the same years and only two of them were occupied. In the Dove Bugt region two sites are known, one with very irregular occupation (most recently in 2004) and one probably an error (H. Meltofte, pers. comm. 2018).
On Figure 9, two areas (Køge Bugt south of Tasiilaq and part of Blosseville Kyst) are indicated by a question mark. Here, adult birds are regularly observed in coastal waters near productive glaciers and breeding colonies may exist inland.
Arctic Tern (Sterna paradisaea): In total, 216 breeding sites are known from East and North Greenland (Fig. 10). Of these, 25 are located in Southeast Greenland, 29 inside the Scoresby Sound Fjord, 158 in Northeast Greenland, and four in North Greenland. Besides these colonies, solitary Arctic Terns are also found breeding along coasts (on gravel beaches or in river deltas) or at inland lakes (usually on small islets). Most of the colonies were small, with fewer than 70 pairs (n = 166) during the most recent survey; 25 colonies were occupied by 70 -700 pairs, while three other colonies were occupied by more than 700 pairs (up to 1100 pairs). The latter are primarily located in Northeast Greenland between 73˚ N and 75˚ N. Extensive coastlines of East and North Greenland are, however, without any colonies (e.g., the Blosseville Kyst, the coast between Tasiilaq and Kangerlussuaq, and between Germania Land and Hovgaard Ø). Based on the most recent survey, the total number of Arctic Terns in these colonies adds up to 12 000 pairs. Black Guillemot (Cepphus grylle): The Black Guillemot is widespread and numerous in the region south of 72˚ N, where 183 colonies are recorded and where the number of observed birds at each site varies between 1 and 300 (Fig. 11). North of this latitude, colonies are very scarce (n = 16) and small, with recorded maxima of 60 individuals at Hvalros Ø and 20 at Mallemukfjeldet, both sites located at polynyas. From Washington Land, only two breeding sites are known, but observations of birds close to the coast indicate that more breeding sites may exist (Andreassen, 1999;Bennike and Feilberg, 2004). The total number of individuals recorded at the most recent survey adds up to 8000. However, since the number of observed individuals in this species fluctuates through the day and season (Ewins, 1985;Andersen et al., 2009), this figure may only reflect an order of magnitude. Moreover, it is certainly below the actual numbers in the region, as many colonies may hide along unsurveyed coasts, especially in Southeast Greenland.
Thick-billed Murre (Uria lomvia): There are only two large Thick-billed Murre colonies in East Greenland, both of which are located at the Scoresby Sound polynya a breeding site on Steward Ø south of Kap Brewster, but no murres have been observed at this site during subsequent surveys reported in this paper.
Little Auk (Alle alle): This species breeds in huge colonies at coasts bordering the Scoresby Sound polynya (Fig. 12). The total number of breeding birds was roughly estimated at 3.5 million pairs in 1985 (Kampp et al., 1986(Kampp et al., , 1987. Figure 12 only gives an overview of the positions of colonies, as their population size is practically unknown. Observations of Little Auks at Kap Dalton in 1933 and Hvalros Ø in 1987 were then considered as an indication of breeding (Degerbøl and Møhl-Hansen, 1935;Stemmerik, 1990), but we failed to find evidence of breeding in recent years (and only saw non-breeding birds at sea) despite several visits at both sites. Breeding has also been suspected on Shannon and nearby islands (Meltofte et al., 1981), but we did not find any trace of breeding birds here in 2004 and 2008 (Gilg et al., 2005;. Atlantic Puffin (Fratercula arctica): Meltofte (1976) mentioned Atlantic Puffins breeding on Raffles Ø and observations of birds at Kap Brewster, both sites at the Scoresby Sound polynya. Since then, one puffin was observed flying off the cliff at Raffles Ø in 2004 (Gilg et al., 2005), and a few birds were seen on (and below) the cliffs of Kap Brewster in 2004 and 2015. Farther north, one individual was observed on Hvalros Ø flying from the cliff in 2004. It is therefore likely that a few pairs breed at these sites located at the two southern polynyas in East Greenland. There is, moreover, an observation that indicates breeding in the southern part of the region: Glahder (1993) observed an adult carrying fish towards the coast just to the south of the region covered by this account.

Total Numbers
In Table 4, the breeding populations of colonial seabirds in East and North Greenland are estimated and compared to previously published estimates from West Greenland, which span almost the same latitudinal range. The number of breeding sites and the number of breeding birds are both much lower in East and North Greenland, where fewer species breed. A simple calculation of colony density in the region results in two colonies per 100 km of coast (excluding glacier coasts) compared to four colonies per 100 km in West Greenland. A rough estimate of the total number of seabirds breeding in East and North Greenland suggests that they are nearly 10 times less abundant than in West Greenland (20 times less when excluding the Little Auks; Table 4).

DISCUSSION
Although the survey effort has been intensive since 2003, large parts of the study region remain unsurveyed for breeding seabirds (Table 2). These parts include most of North Greenland between Washington Land and I.P. Koch Fjord and the region between Germania Land and Hovgaard Ø. The probability of finding significant colonies of Black-legged Kittiwake, Thick-billed Murre, and Little Auk in these parts is, however, low because the sea off these regions is mainly ice covered during the breeding season. Another gap in our survey effort is the interior part of many fjord systems (except between 72˚ N and 73˚30′ N where the GREA expeditions have surveyed). Commuting (between nests and foraging areas) murres and Blacklegged Kittiwakes would have been noticed when passing by the fjord mouths if large colonies were located in the inner fjord systems; however, many gulls-especially Glaucous and Iceland Gulls-Common Eiders, and Arctic Terns may hide in these unsurveyed parts.
There may also be more Ivory Gull colonies than reported here, as the extensive nunatak areas where a few colonies have been located, have only been surveyed sporadically. The observations of Ivory Gulls along some coasts with calving glaciers, such as Køge Bugt and Blosseville Kyst (Boertmann andRosing-Asvid, 2014, 2017) could indicate such overlooked breeding sites (Fig. 9).

Range Changes
There is no doubt that the Common Eider has expanded its range towards the north in recent decades. Nesting or paired eiders have been located at Station Nord (solitary) and on an island as far north as 83˚38′ N; further, females with chicks were observed in 2008 in outer Independence Fjord (82˚30′ N). These sites were visited by keen  Kampp et al. (1987). 2 Middle of range given by Petersen et al. (2015).
ornithologists in the 1970s and 1980s (Håkansson et al., 1981;Hjort, 1986;Hjort et al., 1983, 1988, Bennike and Kelly, 1986) and they did not report Common Eiders. The northernmost breeding record until then was at the coasts of the Northeast Water polynya at 81˚ N (Hjort et al., 1983(Hjort et al., , 1988. The Sabine's Gull population has definitely increased in Northeast Greenland during recent decades; new colonies have been found at many sites where ornithologists in the 1960s and 1970s did not report them (e.g., Meltofte et al., 1981), and many "old" colonies now hold much higher numbers than previously recorded (e.g., Meltofte, 1975). It is also noteworthy that small colonies (2 -3 pairs) were found far inland (up to 40 km) in the Danmark Fjord area and on the island at 83˚38′ N, in 2008. Salomonsen (1967) indicated the northern limit for breeding Iceland Gulls in East Greenland at 68˚ N, and understandably, given the great distance from the gull's known range, disregarded the record of a colony in Northeast Greenland at 75˚ N (Pedersen, 1934;Boertmann, 1994). Whether the colonies found north of 68˚ N in 2008 and 2016 are the result of a range expansion or just of increased observation effort is not known.
Historical sources mention the Great Black-backed Gull as a scarce breeder in the Tasiilaq area (Helms, 1926;Salomonsen, 1967) where its status has not changed during recent decades. The other breeding region, south of Kap Brewster, seems to be a recent colonization, first recorded in 2004 (Gilg et al., 2005). Previous surveys in 1974 did not mention this easily recognizable species (Meltofte, 1976). Both this species and the Lesser Black-backed Gull are expected to increase and extend their ranges in East Greenland in the future.
It is well known that Ivory Gulls have low site fidelity, moving among colony sites from year to year (de Korte and Volkov, 1993), and that they can establish colonies at temporary substrates such as gravel-covered ice floes or icebergs (Boertmann et al., 2010a;Nachtsheim et al., 2016) or at new sites. For example, in 2018 three new sites were located 18, 21, and 24 km from the historical colony at Station Nord (M. Frederiksen, pers. comm. 2018). Wellknown colony sites can also be without breeding birds for years, such as the largest colony in Greenland (Henrik Krøyer Holme), which was empty in 2009, 2017, and 2018, or can be occupied but totally unproductive in some years due to unsuitable weather conditions, such as rain storms (Yannic et al., 2014).
During the most recent surveys, 22 colonies of Arctic Tern had no breeding birds (Fig. 10). Arctic Terns can skip breeding in some years, for example, because of bad weather conditions in the spring or the threat of predation from foxes stuck on the islands after breakup of the sea ice Egevang and Frederiksen, 2011). Hence, at least in Northeast Greenland, these empty colonies cannot be taken as an indication of a population decline. In Southeast Greenland in 2014 on the other hand, local people argued that predation from landlocked polar bears was the main reason for non-breeding and that this had occurred for several years. In this region, a polar bearinduced decline of Arctic Terns cannot be excluded (see also Prop et al., 2015).

The East Greenland Ecosystem
The marked discrepancy in density and abundance of breeding seabirds in East and North Greenland compared to West Greenland is assumed to be linked to the different oceanographic conditions. The continental shelf off West Greenland has many local upwelling events, permanent open water in the southern part, and earlier breakup of sea ice farther north, all factors contributing to high primary production (Buch, 2002;Laidre et al., 2008). Off East Greenland, a strong current conveys nutrient-poor and cold water (including large quantities of drift ice) from the Polar Basin over the continental shelf-an outflow shelf, sensu Carmack and Wassmann (2006). This largely inhibits primary production and planktonic blooms, except in the few polynyas and probably along the shelf break. The summer sea ice off East and North Greenland is, however, retreating (Laidre et al., 2015;Stroeve and Notz, 2018), and seabirds have already responded, as exemplified by the recent breeding range extension of gulls, Arctic Terns, and Common Eiders to the northernmost land on Earth.
When looking at the distribution of all the seabird breeding colonies in East and North Greenland, one region stands out, both in terms of diversity and in bird numbers: the coasts of the Scoresby Sound polynya. This is the only region where huge colonies of Little Auks, roughly estimated at 3.5 million pairs in 1985 (Kampp et al., 1987), are found. The only colonies of Thick-billed Murre and the largest of Black-legged Kittiwake are also found there. The coast just south of Kap Brewster (to Rømer Fjord) also hosts Arctic Terns, Common Eiders, and several gull colonies. In total, 11 colonial seabird species breed here. The only other area in Greenland that can match this richness in seabirds is the Avanersuaq area of Northwest Greenland, which holds even more breeding seabirds (Boertmann and Mosbech, 1998;Burnham et al., 2012;Merkel et al., 2014) and which also borders the North Water polynya (Hastrup et al., 2018).
The two other major polynyas of East and North Greenland, the Northeast Water polynya (with breeding Northern Fulmar, Common Eider, Sabine's Gull, Blacklegged Kittiwake, Ivory Gull, Arctic Tern, and Black Guillemot) and the Sirius Water (with Black-legged Kittiwake, Arctic Tern, Sabine's Gull, and Common Eider) support many fewer species and breeding pairs, although still high in diversity and density compared to the general coastlines of the region.
Why these two polynyas are less rich in seabird abundance is unknown. Since the coastal morphology is optimal for breeding seabirds at both polynyas, the explanation probably comes from different oceanographic features. At least in the Northeast Water, a strong benthic-pelagic coupling and low densities of copepods (Hobson et al., 1995;Hirsche and Kwasniewski, 1997;Karnovsky et al., 2007) apparently limit the availability of food for most of the colonial seabirds . This is certainly not the case in the Scoresby Sound polynya, but biological and physical oceanographic data are lacking from this important polynya (and from the Sirius Water), except for some recent studies on copepod abundance carried out in relation to studies on Little Auk foraging strategies (Fort et al., 2013;Amélineau et al., 2016).
Locally, small areas with early sea ice breakup, for example created by tidal currents, also hold seabird colonies, although much smaller than at the large polynyas. Such areas include the northern outlet of Dove Bugt (with breeding Common Eider, Sabine's Gull, Black-legged Kittiwake, and Arctic Tern), the mouth of Kangerlussuaq (Common Eider, Arctic Tern, and Black Guillemot), Timmiarmiut Fjord (Common Eider, and Arctic Tern) and the archipelago of Umiivik (Common Eider).
The sea ice condition in spring and summer is another significant factor determining the distribution of seabird colonies in East and North Greenland; in contrast to the polynyas, the ice here both prevents access to foraging waters and delays the primary production in the water column. But the summer sea ice off East Greenland is currently decreasing, both in duration and extent (Laidre et al., 2015;Stroeve and Notz, 2018;Box et al., 2019). This decrease is illustrated by the ship-based surveys off the Blosseville Kyst in 2004 and 2016, which were performed in completely ice-free waters, whereas expeditions in 1899, 1900, 1932, and 1980 struggled with dense sea ice on exactly the same dates in July (Amdrup, 1902;Mikkelsen, 1933;Andersen, 1981).
Other coastlines stand out for their lack of seabird colonies. The most significant are the coast between Germania Land and Hovgaard Ø, where no colonies are located. Their absence can of course be partly attributed to the low survey effort there. However, ice conditions certainly play a role: the Norske Øer Ice Barrier, which is a wide (up to 100 km), semi-permanent ice shelf, covers all waters east of the islands throughout the year (Schneider and Budeus, 1997); the Jøkel Bugt between the mainland and the islands is also permanently covered by large, stationary, tabular icebergs. The ice shelf has, however, proven less stable in recent years (Sneed and Hamilton, 2016), and the potential for establishing new seabird colonies here is high.
Very few colonies are known from the western part of North Greenland, between I.P. Koch Fjord and Humboldt Glacier, and all are located in Washington Land where archaeologists, botanists, and geologists (with ornithology as a hobby) recorded birds in 1999 (Andreassen, 1999;Bennike and Feilberg, 2004). The surveys reported in the present account did not cover this area, but previous ornithological activity did not record any colonies between I.P. Koch Fjord and Washington Land (Håkansson et al., 1981;Bennike and Kelly, 1986). A few historical observations indicate that Black Guillemot may occur in Hall Land (Dietz and Andersen, 1984), and there may be additional colonies of Common Eider and Arctic Tern, especially along the coasts of Washington Land.

Climate Change
The ongoing decline in ice cover is expected to continue, and depending on which greenhouse gas emission scenarios will prevail, the summer sea ice at the North Pole may disappear within less than two decades (Overland and Wang, 2013). The world's northernmost seabird colony with Arctic Tern, Ivory Gull, Sabine's Gull, and Common Eider was located at 83˚38′ N in 2008, in an area where no colonial seabirds were breeding in the 1970s and 1980s. The eiders have moved more than two latitudinal degrees north to reach this site since the 1980s, and the potential for establishment of new colonies of these species along the coasts of North Greenland is high.
All but a few of the seabird species breeding in East and North Greenland will probably benefit from these changes-more open water and hence more, easier, and earlier feeding opportunities. The Ivory Gull, however, will lose its primary feeding habitat, the summer sea ice, and is expected to decline in Greenland and elsewhere. It is currently red listed globally as Near Threatened (NT) (BirdLife International, 2018a). It is also red listed on a national level in Greenland (Vulnerable, VU) and Norway (Vulnerable, VU) (Boertmann and Bay, 2018;Henriksen and Hilmo, 2015). Russia and Canada have not adopted the IUCN (International Union of Nature Conservation) red list system, but maintain similar lists on which the species is listed as "rare" and "endangered," respectively (Iliashenko and Iliashenko, 2000;COSEWIC, 2006).
Comprehensive surveys of Ivory Gull colonies have only been achieved twice in Greenland in recent decades. The small survey effort, in combination with the species ability to shift between breeding sites (de Korte and Volkov, 1993; and results reported above), makes it difficult to assess population trends for Greenland. However, the observation that three out of five colony sites in Southeast Greenland were empty in 2008 and 2009 is concerning (see also Gilg et al., 2009). This observation, and the information from Canada where the breeding population has declined by more than 70% within just 20 years (Gilchrist and Mallory, 2005), calls for a survey of all the breeding sites in Greenland. Such a survey was conducted in 2019, alongside similar surveys throughout the species breeding range (cf. the International Ivory Gull Conservation Strategy and Action Plan; Gilchrist et al., 2008).
Another species dependent on sea ice and ice edges for feeding during the breeding season is the Little Auk (e.g., Bradstreet, 1982;Lønne and Gabrielsen, 1992;Jakubas et al., 2013). Studies in East Greenland indicate that Little Auks are more flexible in their foraging behavior than previously assumed and can also fledge their chicks on a diet based on open-water prey (Gremillet et al., 2012;Amélineau et al., 2016). In the long run, however, there may be concerns for the species' status in the Scoresby Sound polynya. A northwards range shift could be likely if feeding conditions in the nearby sea allow, since nesting habitats are at least abundant there.
Thick-billed Murres are also dependent on the timing of sea ice breakup and the occurrence of ice edges (Laidre et al., 2008). Since their numbers in the two breeding colonies are rapidly declining, partly because of harvest (Merkel et al., 2014) and likely also because of other overall oceanographic processes affecting birds in the wintering area (Descamps et al., 2013, Frederiksen et al., 2019, there is strong concern for their future. Finally, new species may become established in the region. The most likely candidate is the Great Skua (Stercorarius skua). The species has since 2003 repeatedly been observed under conditions that strongly suggest breeding at several sites in East Greenland: Kilen (up to three supposed pairs), Nordostrundingen, and Henrik Krøyer Holme in the Northeast Water polynya area (Gilg et al., 2003;P. Clausen, pers. comm. 2006;Boertmann and Nielsen, 2010;M. Frederiksen, pers. comm. 2018) and at the northern Blosseville Kyst (Gilg et al., 2003;Boertmann and Nielsen, 2010;Boertmann and Rosing-Asvid, 2017). But breeding has so far not been confirmed.
The Northern Gannet (Morus bassanus) is another potential new breeder. The mackerel (Scomber scombrus) has recently moved into Southeast Greenland waters where a large fishery has developed (Jansen et al., 2016). As mackerel is an important food item for gannets (Cramp and Simmons, 1977), they may follow this prey and establish breeding colonies in Southeast Greenland or perhaps in Southwest Greenland first, matching the situation from the Barents Sea (Barrett et al., 2017). Whether Atlantic Puffin establish a larger and more stable breeding population is questionable as the population in the North Atlantic is, in general, not thriving (BirdLife International, 2017), although the Svalbard population appears to be doing well (Henriksen and Hilmo, 2015). Other likely candidates from neighbouring areas-Iceland and West Greenland-include Razorbill (Alca torda) and Herring Gull (Larus argentatus), both rare visitors to East Greenland. In a longer perspective, the Leach's Storm Petrel (Hydrobates leucorhous) may also be a candidate as a new breeding bird in Southeast (or Southwest) Greenland, although the population of this species is also currently declining (BirdLife International, 2018b).

Conservation
A very large part of the region described in the present study is included in the Northeast Greenland National Park (Figs. 1, 2), where the seabird colonies are protected and traffic-related disturbance regulated. Moreover, two areas inside the national park and two just south of the park are designated as Ramsar sites, which are wetlands of international importance and protected from habitat changes under the Ramsar Convention on Wetlands (Fig. 2). The remaining colonies, including the important Little Auk and Thick-billed Murre colonies, as well as other seabird colonies in Southeast Greenland, are only protected by the national nature protection law, which regulates traffic in and near colony sites. All species are also protected from hunting in the breeding season, a regulation relevant only in the two inhabited regions of East Greenland.
In recent years, there has been focus on vulnerable and important marine areas in the Arctic, both on a national level and internationally. The Arctic Council has designated Arctic marine areas of heightened ecological and cultural significance, which may lead to protection from the impacts of shipping (AMAP/CAFF/SDWG, 2013). In 2010, the Danish and Greenlandic ministers of environment requested an identification of ecologically valuable and sensitive marine areas in relation to shipping in Greenlandic waters (Christensen et al., 2012). These two initiatives resulted in the designation of a number of marine areas as candidates for protection, including the Northeast Water polynya, the Sirius Water, and the Scoresby Sound polynya, all areas of importance for breeding seabirds. In addition, the IUCN has designated the Northeast Water polynya and the Scoresby Sound polynya as sites of potential outstanding universal value (Speer et al., 2017).
The formal protection of seabird colonies in East and North Greenland can therefore be regarded as adequate, but regulation should be reinforced in a better way, especially for the rapidly declining Thick-billed Murre colonies, which have suffered (and perhaps still do) from illegal overharvesting.
Besides presenting the first comprehensive analysis of all available survey results, this status paper hopefully will help to assess future climate-induced changes on the breeding seabird populations in East and North Greenland, including indirect impacts resulting from human activities (e.g., tourism, onshore mining, offshore oil exploitation, shipping) facilitated by these ongoing rapid changes (e.g., warmer climate and less sea ice). number of private and public grants over the past 40 years. We cannot list them all, but the Tara Ocean Foundation (especially Etienne Bourgois, Agnès Bourgois and Romain Troublé), Columbia Sportswear, Post Greenland (Allan Pertti Frandsen), Frederik Paulsen, l'Agence Nationale de la Recherche (grant ILETOP ANR-16-CE34-0005) and, since 2011, the French Polar Institute IPEV (program Interactions 1036) deserve special mention for their long-term and invaluable support. Among the many ornithologists and naturalists who joined these private expeditions, a few also deserve special thanks for their significant help in the field or for sharing their data: Eric Bushel, Christian Drone, Corinne Eckert, Vladimir Gilg, Christian Hag, Vadim Heuacker, Jacques Inset, Johannes Lang, Pierre Leguesdron, Freddy Marivaux, Brigitte Sabari, Raphaël Sané, Renaud Scheiffler, Benoît Sittler, Jean-Pierre Wiest, and Glenn Yannic.
Finally, we wish to thank all the scientists and travellers who provided data to the Greenland Seabird Colony Register, the Danish Environment Protection Agency, which partly funded the preparation of this paper, and three anonymous reviewers, who commented on and improved the manuscript.