The distribution and diel movements of Brunnich’s Guillemot Uria lomvia in ice covered waters in the Barents Sea, February/March 1987 VIDAR BAKKEN Bakken, V . 1990: The distribution and die1 movements of Brunnich’s Guillcmot Uria ~omoiu in icc covered waters in the Barents Sea, February/March 1987. Polar Research 8 , 55-59. The distribution of Briinnich’s Guillemot in ice covered waters and near the marginal ice zonc in thc southern part of the Barents Sca was mapped from ship and helicopter in February/Mdrch 1987. High densities of Briinnich’s Guillcmot (up to 1,300 ind./km*) were found in ice leads. The density of birds was especially high over shallow hanks where the sea depth was 4&80 m. A diel movement was also recorded. In the evening the birds left the leads and flew south. Next morning they returned to fccd in the open leads. How far they migrated is uncertain, but possibly they flew down to the open sea or to leads close to the marginal ice zone. The migration may have been a mcans of avoiding to becomc trapped if leads closed after dark. Vidar Rakken, Norsk Polarinsriturt. P . O . Box 158, N-1.330 Oslo Lufthaon, Norway; October I989 (revised February 1990). The Brunnich’s Guillemot is one of the most numerous seabird species in the Barents Sea. Breeding colonies are found on the Norwegian coast (only in small numbers), in the Svalbard area and in the Soviet territories (Frans Josef Land, Novaja Zemlja and the Murman coast) (Fig. 1) (Norderhaug et al. 1977; Golovkin 1984; Mehlum & Fjeld 1987). More than 600,000 pairs breed in the Svalbard area (Mehlum & Fjeld 1987; Bakken & Mehlum 1988; Knutsen et al. 1988; Kempf & Sittler 1988), while according to Golovkin (1984), the total breeding population in the Soviet territories is more than one million pairs. Ringing recoveries have shown that Briinnich’s Guillemots breeding on the west coast of Spits- bergen and on the Murman coast migrate towards waters southwest of Greenland in winter (Nor- derhaug et al. 1977; Kampp 1988). Despite that more than 50,000 Brunnich’s Guillemots have been ringed on Novaja Zemlja (Uspenski 1958), only a few have been recovered from Greenland (Salomonsen 1967,1971; Kampp 1988), and many probably stay south of the ice covered areas in the eastern part of the Barents Sea during the winter (Uspenski 1958). Counts of seabirds in the southern Barents Sea in 1986-1988 have shown that the Brunnich’s Guillemot was a common species south of the ice covered waters in winter (Anker-Nilssen et al. 1988). In early spring, high densities of Brunnich’s Guillemot have been recorded in leads near the breeding colonies at Svalbard (Norsk Polar- institutt unpublished), but if they also exploit the ice covered waters in winter is unknown. In the Bering Sea high densities of guillemots have been observed in March, both in leads and near the marginal ice zone (Irving et al. 1970). BJ0rne ya 1 3 LAND Fig. 1. Map of the Barents Sea. 56 V . Bakken In this paper I present results of counts of Brunnich's Guillemots in the ice covered waters in the Barents Sea in February/March 1987, during which period a diurnal migration was recorded. The possible causes for this migration are discussed. The ice conditions in the Barents Sea Most of the ice in the Barents Sea is formed locally (Vinje 1985). During the winter, it gradu- ally extends southwards, and surrounds the breed- ing colonies in the northern part of the Barents Sea (Svalbard, Frans Josef Land and Novaja Zemlja). However, the extent of the ice cover vanes greatly from year to year (Fig. 2). The ice drift is influenced mostly by the wind, and the occurrence and distribution of open leads are highly dependent on wind speed and direction (Vinje 1985). Southerly winds compact the ice and close the leads, while northerly winds have the opposite effect. There are no areas within the ice covered waters which are permanently open during the winter period. DECEMBER JANUARY FEBRUARY MARCH Fig. 2. Maximum and minimum extension of sea ice in the Barents Sea in December-March based on long-term Russian observations (after USSR Ministry of Defence 1980). Methods Counts were made from K.V. 'Nordkapp', a Norwegian coastguard vessel, and from a heli- copter which was based on the ship. Ship-based counts were made in daytime from one side of the bridge (16 m a.s.1) by a method described by Tasker et al. (1984). When the ship was anchored, counts-of flying guillemots were made in the morn- ing and evening from an observation platform (26 m a.s.1.). Counts from the helicopter were made at a speed of c. 90 knots and at a height of c. 200 feet. Two observers, one on each side, recorded birds seen in a 200 m wide (100 m on each side of the helicopter) transect. 18 71 9 0 0 0 0 0 D L" d ln a ln a00 e oe D L" d M 2 M a00 e oe 7 8 .O 11 .o 78 .O 17 .O 16 .O 15 .o 14 .O 9 9 9 9 9 9 s ln 0 ln L" a m 16 .O 11 .o 16 .O 1s .o l i . 0 0 NONE @ 1.0 - 10.0 0 0.1 - 0.3 Q 10.0 - 30.0 Q 0.1 - 1.0 0 10.0 - 100.0 0 1 . 0 - 1.0 MORE ?HAN 100.0 Fig. 3. Densities (number of birds per km2) of Briinnich's Guillemots observed from the boat (top) and helicopter (bottom) in the ice covered waters and close to the marginal ice zone in the Barents Sea, February/March 1987. Distribution and die1 movements of Briinnich's Guillemot 57 The areas covered by the ship and helicopter surveys are shown in Fig. 3. The data presented were collected from 27 February to 8 March. Results The distribution of Briinnich's Guillemot was patchy in the ice covered waters, with lower den- sities of birds in the eastern part of the study area (Fig. 3 ) . The highest densities (up t o 1,300 birds/ km2) were seen in leads south and west of Hopen (Fig. 4). In this area there are shallow banks with sea depths from 40 t o 80 m, as compared to depths of c. 150 to 300m in the eastern sector. The distance from this area to the southerly marginal ice zone was about 150km. Open leads were found throughout the area investigated. In the western part of the study area, between Hopen and B j m m y a (Fig. 4), we noted a distinct migration of Briinnich's Guillemots every morn- ing and evening. Just before dusk, the guillemots left the leads and flew south. In the morning they returned, and stayed in the leads for 7-10 hours. At night, there were almost no guillemots left in the area. The main northerly migration took place between 0700 and 1000 hrs. and the southerly migration between 1400 and 1700 hrs. (GMT). Numbers flying past one side of the ship were counted three times (twice in the morning and once in the evening) (Table 1). The total number of birds migrating was much higher than shown in Table 1. During the survey, which lasted 14 days, the ice distribution changed greatly, with a general eastward shift of the western edge of the marginal ice zone (Fig. 4). This change also influenced the distance from the main area, where the guillemots were observed southwest of Hopen, to the mar- ginal ice zone. In the last part of the survey, the distance to the western marginal ice zone was about the same as t o the southern marginal ice zone (Fig. 4). One night during this period the Fig. 4. The study area with the positions of the marginal ice zone at the start and at the end of the survey. The arrows indicate where and in which direction a diurnal migration of Briinnich's Guillemots was registered. The arrows pointing north are morning migrations and the arrow pointing south is an evening migration. Thc dotted area indicates where the highest densities of Briinnich's Guillemots in leads were observed. crew on the bridge observed 'large flocks' of Briin- nich's Guiliemot in the ship's search-light in big leads close to the western marginal ice zone. Because no migration observations were made the previous evening or the next morning, I do not know if there also was a shift in the migration direction. Discussion The high densities (up t o 1,300/km2) of Briin- nich's Guillemots show that the southern parts of the ice covered waters of the Barents Sea were important for this species in FebruarylMarch. When passing the breeding colonies on Hopen and Bj@rncbya in the helicopter, no birds were observed on the breeding ledges which were Table 1. Three counts of migrating Brunnich's Guillemots observed on one side of the ship in the Barents Sea, February/March 1987. Date Time (GMT) Position Number of migrating birds Direction 27 Feb. 0740-1140 74"40'N 21"56'E 6,440 North/northwest 04 Mar. 14W1600 76"20'N 25"OO'E 7,903 South/southeast 06 Mar. 070C-1000 76"05'N 23"40'E 10,564 North/northwest 58 V. Bakken covered with snow. Hence, the occurrence of the guillemots in the ice covered waters was probably not connected to the use of the colony prior to the breeding season. In the Barents Sea, the formation of leads is almost totally dependent on the wind (Vinje 1985). Meteorological data from the southern part of the Barents Sea during a 30 year period show that the predominant wind-direction in winter is from the north-east (Steffensen 1982). Leads form on the leeward side of the island (T. Vinje pers. comm.); hence, the probability of finding open leads southwest of Hopen in winter is high. Here tides probably also influence their formation, particularly in shallow areas where currents may be faster (T. Vinje pers. comm.). The area may thus be of great importance for Briinnich’s Guillemots when the sea is otherwise covered by ice. Leads may similarly form south of the islands further north (Spitsbergen and Edgeoya), but if the birds have to spend the night in open water the distance to the marginal ice zone is much longer. We have few data showing where the birds spent the night, but two observations indicate that the birds migrated to the marginal ice zone or to big open leads close to it. Birds observed on a northerly morning migration near Bjornoya were only a few km from the marginal ice zone (Fig. 4). The night observation of ‘large flocks’ in open water near the western marginal ice zone also suggests a daily migration to this area. The evening migration from the ice covered waters could be caused by the danger of being trapped in the ice at night when leads may close. Because of small wingloading (Ruppel 1975), guillemots need considerable room to take flight from the water. During other surveys in ice covered waters in the Barents Sea, I have observed Brunnich’s Guillemots trapped in small leads and being unable to take off. In strong winds Brunnich’s Guillemots may manage to get alight from an icefloe, but in most cases they fail (per- sonal observation). This die1 movement was only observed over a four day period when the ship stayed south and west of Hopen, and I do not know if this migration is the regular pattern throughout the winter. If the birds depend on light for feeding in the ice covered waters, it will probably be impossible to exploit these areas in December and January because of the dark period. At Hopen the sun is below the horizon from 2 November to 9 Febru- ary. A similar die1 movement for Briinnich’s Guillemots in ice covered waters has not pre- viously been recorded, but there are earlier obser- vations of high densities of guillemots near the marginal ice zone and in leads elsewhere (Irving et al. 1970; Divoky 1979; Bradstreet 1980, 1982). The profit of feeding in the leads must have been high in relation to a daily return flight of up to 300 km (assuming they flew from the marginal ice zone). During our survey, the feeding con- ditions in the leads were good. Stomach analyses of 30 Brunnich’s Guillemots shot in the leads showed that the staple food was the amphipod Parathemisto libellula and the Polar Cod Bor- eogadus saida (F. Mehlum & G. Gabrielsen pers. comm.). The mean body weight of the birds was 1,066 grams (SD = 113, n = 30), which is among the highest recorded during the year i n the Sval- bard area (F. Mehlum & G. Gabrielsen pers. comm. ; personal observation). Stomachs of birds shot in open water near the marginal ice zone in the same time period contained mostly Polar Cod and amphipods, but the samples were well digested (Erikstad 1990), suggesting that the birds had caught the food earlier (and possibly else- where). Gulliksen & Lonne (1989) have shown that the sympagic fauna in this area is rich in Polar Cod and amphipods, and these resources are perhaps easier to catch for the birds than prey in the open sea at this time of year. This may explain why the guillemots migrated into the ice covered waters each day. Acknowbdgemenn. -Thanks are due to Geir Wing Gabrielsen and Egil Soglo who helpcd counting birds. I am grateful to Geir Wing Gabrielsen and Fridtjof Mehlum for letting me use their unpublished data and for thcir comments on the manuscript. 1 am also grateful to Kjell Einar Erikstad, Rob Barrett and an anonymous referee for valuable comments. Rob Barrett also improved the English. Many thanks to the crew and helicopter pilots on the coastguard vcssel K/V ’Nordkapp’ for their help and hospitality. ‘This study was financed by the Norwegian Department of Oil and Energy as a part of an environmental impact analysis in connection with planned oil/gas drilling in the southern part of the Barents Sea. References Anker-Nilssen, T.. Bakken, V. & Strann, K-B. 1988: Kon- sekvensanalyse olje/sjchfugl ved petroleumsvirksomhet i Barentshavet s0r for 74”30’N. V i l f r p p o r f 46. 98 pp. +Appendix. Distribution and die1 movements of Brunnich's Guillemot 59 Bakken, V. & Mchlum, F . 1988: AKUP-Sluttrapport. Sjdfugl- undersvtkelscr nord for N7C/Bjvtrndya. Norsk Polarinst. Rapportserir 4 4 . 127 pp.+Appendix. Bradstreet, M. S . W . 19x0: Thick-billed murres and black guillemots in the Barrow Strait, N.W.T., during spring: diets and food availability along ice edges. Cun. J . Z o o / . 58, 2120- 2140. Bradstreet, M. S . W . 1982: Occurrcncc, Habitat Use, and Behaviour of Scahirds, marine Mammals, and Arctic Cod at the Pond Inlct Ice Edge. Arctic 35, 28-40, Divoky, G . J . 1979: Sea ice as a factor in seabird distribution and ecology in the Beaufort, Chukchi, and Bering Seas, Pp. 7-19 in Bartonek, J. C. & Nettleship, D . N. (eds.): Conservation of Marine Birds of northern North Amcrica. Fkh Wildl. Seru. Res. Rep. 11. Erikatad, K . E. 1990: Winter diets of 4 seabird species in the Barenls Sea after a crash in thc capelin stock. Polur Biol. (in prcss). Golovkin, A. N . 1984: Seahirds Nesting in the USSR: The Status and Protection of Populations. Pp. 473-486 in Croxall, J . P . , Evans, P. G . € I . & Schreiber, R . W . (eds.): Staius and Conseriiation of the World's Seabirds. K B P Techn. Publ. 2 . Gullikscn, B. & Ldnne, 0. J . 1989: Distribution, abundance, and ccological importance of marine sympagic fauna in the Arctic. Rapp. P.-v, Reun. Cons. inr. Explor. mer 188, 133- 138. Irving, L . , McRoy, C. P. & Burns, J . J . 1970: Birds observed during a cruise in the ice-covered Bering Sea in March 1968. Condor 72, lllL112. Kampp, K . 1988: Migration and winter ranges of Brunnich's Guillemots Uria lomuiu breeding or occurring in Greenland. Dunsk O m . Tidsskr. 82, 117-130. Kempf, C. & Sittler, B. 1988: Ccnsus of breeding seabirds on the northwest coast of Svalbard 1973 and 1978. Polar Reseurrh 6 , 195-203. Knutsen, L. O . , Fjeld, P. E. & Olsson, 0. 1988: Sjdfugl- undersvtkelser pA 0st-Spitsbergen, Svalbard, mcd kon- sckvensvurdering av aktivitet p i Haketangen. Pp. 79-1 14 in Prestrud, P. & 0ntsland. N. A. (eds.): Miljdunders~kelser i tilknytning ti1 petroleumsvirksomhet p i Svalbard 19x7. Norsk Polarinst. Rapportserie 41. Mchlum, F . & Fjeld, P. E. 1987: Catalogue of scahird colonies in Svalbard. Norsk Polarinst. Rapportserie 35. 222 pp. Nordcrhaug, M., Brun, E . & Mollen, G . U . 1977: Barents- havcts sjdfuglressurser. Norsk Polarinst. Meddelelser 104. 119 pp. Ruppcl, G . 1975: Fugle - Fm-digheter, Flukt og Fart. Lademann. 184 pp. Salomonscn, F. 1967: Fu,qIeiie p6 Gr@nlurrd. Khodos. 341 pp. Salomonsen, F. 1971: Recoveries in Grccnland of birds ringed abroad. Medd. o m Grgnlund 191. 52 pp. Stcffcnsen, E . 1982: The Climatc at Norwegian Arctic Stations. Klimu 5 . 44 pp. Tasker, M. L . , Hope Jones, P., Dixon, T. J . & Blake, B . F. 1984: Counting seabirds at sca from ships: a review of mcthods employed and a suggestion for a standardized approach. Auk 101, 561-577. Uspenski, S . M . 1958: The bird bazaars of Novaya Zemlja. Canadian Wildl. Seru. Translations of Rllssian Game Reports 4 . Ottawa. 159 pp. USSR Ministry of Defence (Gorskov, S . G. & Faleev, V. 1. (cds.)) 1980: Atlas of the Oceans. The Arctic Ocean. Vinje, T . 1985: Drift, composition, morphologhy, and distri- bution of the sea ice fields in the Barents Sca. Norsk Polarinst. Skrifter 179C. 26 pp.