Distribution and abundance of Canadian High Arctic belugas , 1974-1979

We conducted >236,000 km of aerial surveys and some supplementary studies of belugas (Delphinapterus leucas) in the central and eastern Canadian High Arctic in 1974-79. Belugas that wintered in the “North Water” in Baffin Bay moved southwest into Lancaster Sound in April and early May. The main westward migration into Lancaster Sound occurred over a 2 to 3 week period during late June to late July. Estuaries along Somerset Island were occupied for <3 weeks from mid-July to mid-August. Little feeding occurred in estuaries. From mid-August until fall migration began in mid-September belugas occupied estuaries and offshore waters in Peel Sound. Fall migration eastward through Lancaster Sound was exclusively along the south coast of Devon Island, highly co-ordinated, and rapid; most of the population passed through the sound in <1 week. The whales then moved north along the east coast of Devon Island; some entered Jones Sound while others crossed directly to SE Ellesmere Island. Most calving occurred in July and early August; calving was not seen in estuaries and probably occurred offshore. Excluding calves, adults and yearlings formed 77% and 8.4%, respectively, of the population. The proportion of calves during mid-August was consistent with a triennial calving cycle. During late summer, belugas fed on coastal concentrations of polar cod (Boreogadus saida), under pan ice offshore (probably on cod), and in deep offshore waters. The size of the Canadian High Arctic population in the late 1970s was estimated to be at least 10,250 to 12,000 animals without allowing for animals that may have passed between surveys or that were below the surface at the time of the counts. Koski, W.R., Davis, R.A. and Finley, K.J. 2002. Distribution and abundance of Canadian High Arctic belugas, 1974-1979. NAMMCO Sci. Publ. 4:87-126.

D Di is st tr ri ib bu ut ti io on n a an nd d a ab bu un nd da an nc ce e o of f C Ca an na ad di ia an n H Hi ig gh h A Ar rc ct ti ic c b be el lu ug ga as s, , 1 19 97 74 4--1 19 97 79 9 William R. Koski 1 , Rolph 1 A. Davis and Kerwin J. Finley 1,2   1 LGL Limited environmental research associates, 22 Fisher Street, P.O.Box 280, King City, Ontario, Canada L7B 1A6 2 K.J. Finley Ecological Research, 10232 Summerset Place, Sidney, British Columbia V8L 4X2, Canada A AB BS ST TR RA AC CT T We conducted >236,000 km of aerial surveys and some supplementary studies of belugas (Delphinapterus leucas) in the central and eastern Canadian High Arctic in 1974-79.Belugas that wintered in the "North Water" in Baffin Bay moved southwest into Lancaster Sound in April and early May.The main westward migration into Lancaster Sound occurred over a 2 to 3 week period during late June to late July.Estuaries along Somerset Island were occupied for <3 weeks from mid-July to mid-August.Little feeding occurred in estuaries.From mid-August until fall migration began in mid-September belugas occupied estuaries and offshore waters in Peel Sound.Fall migration eastward through Lancaster Sound was exclusively along the south coast of Devon Island, highly co-ordinated, and rapid; most of the population passed through the sound in <1 week.The whales then moved north along the east coast of Devon Island; some entered Jones Sound while others crossed directly to SE Ellesmere Island.Most calving occurred in July and early August; calving was not seen in estuaries and probably occurred offshore.Excluding calves, adults and yearlings formed 77% and 8.4%, respectively, of the population.The proportion of calves during mid-August was consistent with a triennial calving cycle.During late summer, belugas fed on coastal concentrations of polar cod (Boreogadus saida), under pan ice offshore (probably on cod), and in deep offshore waters.The size of the Canadian High Arctic population in the late 1970s was estimated to be at least 10,250 to 12,000 animals without allowing for animals that may have passed between surveys or that were below the surface at the time of the counts.Koski, W.R., Davis, R.A. and Finley, K.J. 2002. Distribution and abundance of Canadian High Arctic belugas, 1974-1979. NAMMCO Sci. Publ. 4:87-126.

I IN NT TR RO OD DU UC CT TI IO ON N
B eluga whales (Delphinapterus leucas) that enter the central Canadian High Arctic in summer are part of the "Canadian High Arctic" population.This population is believed to winter off Western Greenland and in the "North Water" of northern Baffin Bay (Finley and Renaud 1980, Doidge and Finley 1993, Heide Jørgensen and Reeves 1996, Richard et al. 1998b).The Canadian High Arctic population is hunted by Canadian and Greenlandic Inuit, and surveys conducted in the late 1980s and early 1990s off Western Greenland suggested that it may have declined considerably over the last two decades, probably due to large catches in Western Greenland (Doidge and Finley 1993, Heide-Jørgensen 1994, Heide-Jørgensen and Reeves 1996).The latter authors estimated that the population that winters off Western Greenland had declined by as much as 62% from the early 1980s to the early 1990s.However, the numbers of belugas in summering areas in the central Canadian High Arctic do not appear to have declined as dramatically, if at all (Innes et al. 2002).It is unknown whether the apparent decline off Western Greenland was due to an actual decline in population, to changes in winter distribution, or to a combination of these two factors.
Prior to the early 1970s the size of this population had not been estimated.The first rough estimate of 10,000 was based on a reconnaissance survey of estuarine concentrations by Sergeant and Brodie (1975).The interest in hydrocarbon resources in the Canadian High Arctic during the 1970s led to several wide-ranging surveys of marine mammals.LGL Limited and the authors conducted most of these surveys in the central and eastern Canadian High Arctic in 1974-79.In this paper, we present information on the spring, summer and autumn distribution and migration routes and we estimate the size of the population in the late 1970s.These results are presented for comparison with later studies of the same beluga population (Smith et al. 1985, Martin and Smith 1992, Martin et al. 1993, Smith and Martin 1994, Richard et al. 1998a, Innes et al. 2002).

M ME ET TH HO OD DS S
A total of approximately 236,000 km of aerial surveys were conducted during 1974-79 in marine areas of the central and eastern Canadian High Arctic.Supplementary data were obtained from aerial photographs, by ground observations at four coastal vantage points, by examination of stomach contents of belugas killed by Inuit, and from incidental observations during various biological studies in 1974-79.

Aerial surveys
Surveys of coastal and (less intensively) offshore areas of the central Canadian High Arctic were conducted in 1974-77 and 1979(funded by the Polar Gas Project 1974-77 and Canada Department of Fisheries and Oceans in 1979); surveys of eastern Lancaster Sound were conducted in 1976 (funded by Norlands Petroleums Ltd.); and surveys of the eastern Canadian High Arctic were conducted in 1978-79 (funded by Petro-Canada Explorations, Inc).Surveys off SE Melville Island in 1977 were funded by Petro-Canada.Almost all surveys were conducted from twinengine, high-wing aircraft, usually a Cessna 337 in 1974-75 andDeHavilland Twin Otters in 1976-79.Some special surveys were conducted in helicopters.During most systematic surveys, Global Navigation Systems (GNS-200 in 1974-76, GNS-500 in 1978-79) or On-trac II (in 1977) were used for accurate navigation when landmarks were not visible, and a radar altimeter was used to maintain the desired survey altitude above sea level (ASL).During systematic surveys, sightings were tape-recorded and subsequently mapped for each 2-min (5 to 7.5 km) segment of transect.
During systematic (and most other) surveys, observers were seated in the co-pilot's (right front) seat and in a rear seat on the opposite side of the aircraft.A third observer was often present behind the front observer.Sightings within and beyond a designated transect width were distinguished; "on-transect" sightings were used to estimate the density of mammals per km 2 .Three main types of surveys were conducted: (1) low-altitude (30-50 m ASL) systematic surveys for both marine mammals and seabirds, (2) medium-altitude (90-150 m) systematic surveys for marine mammals, and (3) high-altitude systematic or reconnaissance surveys at variable heights (generally >150 m).Transect widths during such surveys were 400 m, 0.8-1.6 km, and 1.6 km or undefined, respectively.Ground speeds were generally 160-185 km/h, 220-240 km/h, and >220 km/h, respectively.During the second and third types of surveys we often interrupted the survey to circle and/or photograph noteworthy concentrations of mammals.
Observers do not see or record all animals that are present in a survey area.Some animals that 88 Belugas in the North Atlantic and the Russian Arctic are at the surface at the time of a survey are missed, and some animals are below the surface where they cannot be seen when the survey aircraft is present (Eberhardt 1978, Harwood et al. 1996, Innes et al. 2002).The data presented here do not include correction factors to account for animals that were missed by the primary observers for either reason, and so, are minimum estimates of the numbers that were present.The variable heights at which surveys were conducted and the purpose of the surveys also potentially affect the numbers of belugas that were recorded.During low altitude surveys (50 m ASL) there was a greater chance that some belugas were missed because observers attempted to record all birds and mammals that were present.
Transects along coasts and ice edges were usually centred 200 m from the interface during low-altitude surveys with transect width 400 m, and 400 to 1,000 m from the interface during medium-and high-altitude surveys.In 1976In , 1978In and 1979, we also conducted many lowaltitude "nearshore" surveys, which were parallel to and centred 1,200 m seaward of coasts and ice edges.

surveys
Low-altitude surveys (21,170 km) were conducted, mostly from 3-29 June and from 31 July-4 September, along coasts and ice edges from southern Boothia Peninsula north to western Devon Island and west to eastern Melville Island (west of Bathurst Island)(Fig.1, 2).We also obtained some coverage of offshore waters, especially in Barrow Strait (Finley et al. 1974, Davis et al. 1975).

surveys
In June and August, low-altitude surveys (20,275 km) were conducted of most marine areas that had been surveyed in 1974 (but not SE Prince of Wales Island or NW Devon Island).High-altitude surveys over open water (>4,215 km) included four surveys in July and August of 10 transects across Barrow Strait (Fig. 2); these surveys were designed to assess numbers of marine mammals offshore.From 20 July-25 August a helicopter (Hughes 500) was used to survey (3,000 km) whales concentrating along eastern Somerset Island, especially in Creswell Bay.Shore-based observations of marine mammals at Creswell Bay were also obtained (Finley 1976).

surveys
From 1 June-3 July, helicopters (Bell 206 and FH 1100) were used for 11 low-altitude surveys along ice edges and coasts along the northern edge of Barrow Strait, plus 3 surveys along the ice edge across western Barrow Strait (1,477 km, Fig. 1).
From 2 May-28 September, a standard route in eastern Lancaster Sound was surveyed weekly at low altitude (22 surveys, 28,635 km).The route consisted of (1) eight N-S transects across Lancaster Sound, (2) coastal or ice edge transects parallel to and 200 m from the north and south shores, and (3) nearshore transects 1.2 km from the north and south shores (Fig. 1, 2; Johnson et al. 1976, Davis et al. 1978b).Ferry flights also provided high-altitude reconnaissance of northern Admiralty Inlet and western Lancaster Sound.

surveys
On 19 April and 7-8 June, all isolated areas of open water along eastern Bathurst Island and northwestern Devon Island were surveyed for overwintering mammals (2,500 km, Fig. 1).Additional surveys were conducted in these and adjacent areas from 16 July-7 September (7,920 km., Fig. 2, Davis et al. 1978a).Reconnaissance flights (3,000 km) for summering mammals were conducted in Peel Sound, Prince Regent Inlet (both eastern and western sides) and Barrow Strait; vertical photographs of the major beluga summering concentrations were obtained.
From 3 July-28 August, parts of the east and south coasts of Melville Island were surveyed at 30 m and 100 m ASL (1,480 km, McLaren and Renaud 1977).

Fig. 2.
Main aerial survey routes during the summering and fall migration periods (August-October), 1974-79.The coasts, ice edges and offshore waters of a large area of NW Baffin Bay and adjacent channels were surveyed weekly, at low altitude, from 4 May-10 October (22 surveys, 60,678 km, Koski and Davis 1979).The heavily dotted line on Fig. 1 shows the route for the 11 surveys up to 11 July (see also Fig. 5); Fig. 2 shows the route for 8 surveys after 21 August.There were 3 surveys of a similar route in the intervening period.These low-level surveys for marine mammals and seabirds were supplemented with various mammal surveys, including: (1) twelve medium-altitude surveys from 12 May-2 August of a route over offshore Baffin Bay east of Baffin Island (7,528 km, route indicated by triangles on Fig. 1), (2) medium-altitude surveys of leads forming in Lancaster Sound and Prince Regent Inlet during break-up on 18-19 July (475 km), (3) medium-altitude surveys for summering mammals in Eclipse Sound, Navy Board Inlet and Admiralty Inlet (1,600 km), and (4) reconnaissance surveys around most of Devon Island on 13-15 September, during the peak of eastward migration of belugas (3,000 km).

Surveys
The coasts, ice edges and offshore waters of a large area of NW Baffin Bay and adjacent waters were surveyed weekly at 50 m and 100 m ASL from 9 May-22 July (31,260 km, Koski 1980).The hatched line on Fig. 1 shows the route for the 9 surveys up to 22 July.In addition, offshore surveys were conducted south and east of the area shown on Fig. 1 (see Fig. 11 in Koski and Davis 1994).
The autumn survey route was flown 12 times at 4-7 day intervals from 11 September-17 October (Fig. 2) (Koski and Davis 1980).Survey altitude was 90 m ASL.These surveys for both birds and marine mammals were supplemented with medium-altitude surveys (150 m ASL) for marine mammals during August and September including: (1) a survey along S. a Details in Finley and Johnston (1977).No data on 5, 9, 13, 14 Sept. b Details in Salter (1979).Included ten 24-h watches plus observations every 6 h on other dates.c Details in Koski andDavis (1979, 1980 ASL) were taken with color film (Kodak Ektachrome 200).

Shore-based watches
At various times we watched systematically for migrations of marine mammals past four coastal vantage points (Table 1).Weather permitting, crews of 2-3 people conducted continuous daylight watches with the aid of binoculars and spotting scopes at each of these locations.

Ice conditions
Ice strongly affects the seasonal distribution and migration patterns of High Arctic belugas (Vibe 1967).Their summer range is ice covered from about October to June, and their spring migration to these areas is influenced by the timing of break-up.Whales that linger too long on the summer range risk entrapment by newlyforming ice (Freeman 1968, Heide-Jørgensen et al. 2002).
In winter, the only areas of open water in the Canadian Arctic Archipelago during the years of our surveys were a few polynias near northwestern Devon Island and in Bellot Strait (Fig. 3).Most of the channels in the archipelago were covered by thick, landfast ice that persisted until July or early August.A few channels (Lancaster Sound, Prince Regent Inlet, Barrow Strait) contained temporary leads and areas of open water but usually were either completely covered by pack ice or, in some winters, 100% covered by landfast ice.Most of Baffin Bay was covered by "close pack ice" in winter.
Open-water areas in NW Baffin Bay occurred off Jones Sound, along SE Devon Island and across Smith Sound (Fig. 3).
Ice edges were a common feature through the winter, spring and early summer.They were the borders between landfast ice and either pack ice or open water.For most of the year an ice edge extended from NW Greenland across Smith Sound to Ellesmere Island and then south along the western margin of Baffin Bay, excluding whales from areas to the west and north (Fig. 3).However, open water or pack ice generally extended westward into eastern Jones Sound and into Lancaster Sound, even in winter.
As spring advanced, the proportion of Baffin Bay covered by pack ice gradually decreased.
In most years, a lead formed along the west coast of Greenland.were along the northern margin of the sound and only 14% were along the southern margin.Both margins were surveyed with equal intensity, so these percentages are comparable.During the main migration period (20 June-26 July), 94% of the whales within 1.6 km of coasts were along the northern margin of the sound.

1978
During three surveys 4-18 May, 400 to 600 belugas were seen during each survey (Table 2).On 4-7 May none were seen south of Lancaster Sound, and highest densities were along the Jones Sound ice edge.During mid-May, landfast ice prevented the whales from entering Lancaster Sound, and movements were both north and south along the coasts and ice edges.
An influx of belugas occurred prior to the 22-24 May (Table 2).Densities were high (16.8/km 2 ) along the Lancaster Sound ice edge.Distribution and movement patterns suggest that they had arrived from the north or northeast at latitudes north of 75º N. None were seen in off- shore Baffin Bay (Fig. 5) and few were seen south of Pond Inlet.The 455 belugas seen along the north coast of Bylot Island were swimming southeast.
The numbers recorded during surveys decreased substantially in late May (Table 2).The reason for the decline is not clear since the animals could not enter Lancaster Sound or move south of Pond Inlet where the coast was blocked by heavy pack ice.Our surveys included transects along and centred 200 m and 1,200 m from all coasts and ice edges; thus it is unlikely that the decline of over 1,200 animals was due to inadequate survey coverage in coastal areas.Some belugas must have moved either into offshore waters among the pack ice or north of our survey route.
We surveyed northern Baffin Bay on 31 May-1 June; only 349 belugas were recorded.We also surveyed the east coast of Ellesmere Island (281 belugas), the Smith Sound ice edge at 78º 45´ N (8 belugas), and the coast of Greenland south to Saunders Island (76º 35´ N) (29 belugas).The remaining 31 whales were sighted more than 25 km from shore along a grid of transects (180 m ASL) of total length 700 km (Fig. 1).Most of the offshore animals were moving north (74%) and west (10%); 16% were not moving.These results indicate that some belugas do move through offshore waters of extreme northern Baffin Bay.However, there was no evidence that the decline noted in our regular survey area in late May was due to a major northward movement into extreme northern Baffin Bay.It appears, instead, that many belugas moved into offshore waters when they were unable to migrate into Lancaster Sound.In fact, small numbers were noted along the western edge of the Baffin Bay pack ice off the entrance to Lancaster Sound on 1 and 16 June.
In addition to the weekly surveys, we conducted 10 surveys (some extending 250 km offshore) from 12 May -2 August over the pack ice in Baffin Bay to the east of NE Baffin Island (Fig. 1).These surveys covered a total of 6,022 km 2 , but only 15 belugas were seen.Similarly there was no evidence of a migration northwest along the east coast of Baffin Island where we conducted weekly surveys.This confirms that 97 NAMMCO Scientific Publications, Volume 4 most belugas entered Lancaster Sound from the northeast.
The numbers of belugas recorded increased steadily through June and a build-up at the Lancaster Sound ice edge began in mid-June (Table 2).The peak of migration occurred in early and mid-July; over 2,900 belugas (including animals adjacent to the transect strip) were recorded during an incomplete survey on 10-11 July.Nearly 1,700 whales (68.9/km 2 ) were present along the southern half of the Lancaster Sound ice edge; fog prevented surveys of the northern half.Fog hindered surveys until 20 July, and the ice edge began to break up sometime between 12 and 17 July.Belugas penetrated far into the newly developing leads; 728 were recorded in a lead about 150 km west of the ice edge on 17 July, and 320 were seen between northeastern Somerset Island and Maxwell Bay (250 km to the west) on 19 July.Few belugas (370) were recorded on our regular survey 21-26 July (Table 2), and the last westward migrants were noted during the 28 July-2 August survey.
In summary, the main movement of belugas into Lancaster Sound in 1978 occurred between 4 and 20 July.These whales apparently moved across Baffin Bay north of 74º 30´ N and then moved south along coasts and ice edges along Ellesmere and Devon islands to Lancaster Sound.Few belugas crossed Baffin Bay south of 74º 30´ N, or approached Lancaster Sound from the southeast.
1979 From 9 May-15 July weekly surveys were attempted of the route shown in Fig. 1.Fog often prevented completion of this route, particularly coastal and ice-edge transects (Table 3).The first complete survey was conducted 17-22 May and by this time at least 1,437 belugas were present (Fig. 6).Sightings of small numbers of belugas on offshore transects from 74º 30´-76º 00´ N, sightings of whales moving south along the ice edge northeast of Coburg Island (north of 76º), and the gradually increasing counts during each complete survey, all suggest that small numbers of belugas moved into the study area through offshore waters north of 74º 30´ during mid-May to late June.
Only a few belugas were sighted south of Lancaster Sound and the distribution of sightings suggests that those whales probably "overshot" Lancaster Sound after entering our study area from the northeast.No belugas were seen during extensive surveys of offshore Baffin Bay from 69º 30´-74º 30´ N although 29,931 to 38,988 narwhals (Monodon monoceros) were estimated to be present in offshore Baffin Bay during those same surveys (Koski and Davis 1994). Major

98
Belugas in the North Atlantic and the Russian Arctic 11 times from 1 June-3 July.Other channels in the central Canadian High Arctic were also surveyed during spring in 1974-75 (Fig. 1), but were either totally ice-covered or contained only isolated polynias where no whales were seen.

1974
In 1974, eastern Barrow Strait was ice free by April or early May; from then until late July a landfast ice edge extended across western Barrow Strait (position [3] on Fig. 3).We surveyed all coasts and ice edges around Barrow Strait on five occasions from 3-29 June; we also surveyed a small portion of the offshore waters (Fig. 1).4).
The main influx of belugas into the central Canadian High Arctic began in mid-July.Surveys (10% coverage) of eastern Barrow Strait on 17-18 July found a density of 0.27 belugas/km 2 in areas more than 4 km from shore (Fig. 7).This extrapolates to about 2,700 in these offshore waters; this is a minimum estimate because it does not include corrections to

Distribution
After the westward migration of belugas though eastern Lancaster Sound, the species was virtually absent from the eastern Canadian High Arctic.In 1976, we observed only 3 individuals during five summer surveys (31 July-7 September) of eastern Lancaster Sound; our route was as in Fig. 4, but transects along ice edges followed receding ice edges into bays.In 1978, only 14 were recorded during three surveys of NW Baffin Bay and eastern Lancaster Sound (Fig. 2) conducted 3 August-2 September.DIAND, pers.comm.) but we have no data on the numbers involved.
In 1979, the ice edge remained intact across Lancaster Sound until 2 August.Arrival in estuaries was not documented but must have been later than most years.During 17-20 August, 155 belugas were counted in Creswell Bay, 202 in or adjacent to Cunningham Inlet and 312 along the ice edge across Peel Sound.During the same period 2,200 (no corrections) belugas were estimated to have been present in offshore Barrow Strait and northern Prince Regent Inlet based on offshore transect surveys (Fig. 2).

Use of estuaries
Creswell Bay-1975 Belugas traditionally occupy Creswell Bay in large numbers.In 1975, they occupied the bay almost continuously from 20 July-16 August.
We conducted regular counts from a helicopter of the animals along the north shore and, less often, along the south shore of the bay.The numbers present were quite variable (Table 5) due to tide-related changes in distribution, movements between the north and south shores, and movements out of Creswell Bay.
The north and south coasts of the bay contain extensive foreshore flats and shoals.The shoals reach their maximum offshore extent of 6.5 km near the Creswell River and become narrower to the east and west.At low tide, many shoals are exposed and intervening areas are covered by less than 4 m of water.The water deepens abruptly beyond the foreshore flats and deep water extends throughout the central portion of the bay.Tides were unmeasured, but were probably similar to those at Fort Ross (peak amplitude = 2.9m), 75 km south of Creswell Bay.At high tide, belugas concentrated over the inner foreshore along the north coast of Creswell Bay.On 3 August the surface water temperature was 1.8º C at the edge of the foreshore flats off the mouth of Creswell River and 3.0-5.3ºC in areas about 3 km farther west.Twice daily, during low tide, the belugas dispersed along the outer edge of the foreshore area (Fig. 8).These tide-induced movements were regular, with one exception (see below), throughout the period that belugas occupied Creswell Bay.A notable exodus from Creswell Bay occurred at high tide in the late evening (22:00 CDT) of 3 August.Many whales moved east along the coast to Fury Point (Fig. 8).(30-40% coverage) off Fury Point; the next day only 344 belugas were found in the bay (Table 5).The pack ice could not be surveyed but we saw belugas in scattered small groups among the ice pans.The pan ice began to disperse on 5 August and large groups returned westward at 22:00 on that date.
Reasons for the exodus are unknown.Our observations confirmed that at least some of the whales from Creswell Bay were in the pack ice, but we did not obtain direct evidence of feeding under the ice.The presence of substantial amounts of ice near an estuarine concentration of belugas could pose a threat since the ice might blow into the area during high tide and trap the whales when the tide receded.However, belugas live among pack ice during most of the year and are unlikely to react to potential entrapment by ice until it actually enters the estuaries.
The numbers of belugas in Creswell Bay increased in August and reached a peak of at least 3,897 animals on 14 August (Table 5).All these animals had vacated the north coast of Creswell Bay by 16 August.
Other areas-1975 Day-to-day variation in numbers of belugas present in estuarine areas was noted not only at Creswell Bay, but at all bays and estuaries along Somerset Island in 1975 (Table 4).
Coastal areas along northern Somerset Island were occupied less consistently and for shorter periods than were areas along the east coast, probably due to the pack ice that frequently beset the north coast.There was apparently much movement among areas; for example, a grossly deformed adult photographed in Elwin Bay on 3 August was re-photographed in Creswell Bay on 11 August.

Annual variation in estuarine use
There was both within-year variation in use of estuarine summering areas by belugas and between-year differences.Our data for 1975 are presented in Our data from 1974-77 indicate that Peel Sound was consistently used by belugas in late summer.Landfast ice normally cleared from the sound in the first half of August and the whales penetrated into cracks along the east coast of Prince of Wales Island as breakup began.The return northbound migration was also along this coast.Coningham Bay, a shallow estuarine bay along SE Prince of Wales Island, was used by several hundred, or more, belugas in most years.This bay was occupied about a month later than were the estuaries used by summering belugas along northern and eastern Somerset Island.
Autumn migration Read and Stephansson (1976)    were 200 to 400 m from shore (observations by the front-seat observer are ignored here because of seat-related biases in detectability of whales).
In 1976, we did not follow the whales when they left Lancaster Sound.However, the facts that belugas were exclusively along the northern margin of the sound, and an opportunistic sighting of a herd of 735 moving north from NE Devon Island (Fig. 10A) on 27 September, suggest that they probably moved north along eastern Devon Island.

Timing of migration
The 1978 eastward migration through Lancaster Sound was first detected on 2 September when 400 belugas were sighted in Croker Bay, southern Devon Island, by personnel from LGL Limited aboard the research vessel MV Gulf Star.
During the peak eastward migration we surveyed the entire south coast of Devon Island on both 13 and 14 September.On 15 September, we surveyed the west coast of Devon Island north to Baring Bay (100% ice cover from there north), ice-free parts of the north coast of Devon Island, the south coast of Ellesmere Island from Bear Head to Grise Fiord, and the east coast of Devon Island.We conservatively estimate that of the 18,227 belugas sighted dur-ing the three days of surveys, 9,039 represented different animals (Fig. 10B, Table 7).Most of those along the south coasts of Devon and Ellesmere islands were moving east; and those along the east coast of Devon Island were moving north, with the exception of those in Bethune Inlet, which were dispersed and appeared to be feeding along the extensive glacier fronts in this area.
The concentration of belugas in Bethune Inlet persisted (1,597 on 21 September, 1,440 on 24 September, 258 on 3 October, 99 on 7 and 9 October), but numbers elsewhere in the study area declined dramatically (15 seen along the rest of the regular survey route on 18-22 September, 42 on 23-27 September, 288 on 28 September-4 October, and 26 on 7-10 October).

Migration route from Lancaster Sound
As in 1976, virtually all belugas migrating along the south coast of Devon Island in September 1978 were within 400 m of shore and most were within 200 m; this was true even for animals seen in bays or inlets (Fig. 11).The few exceptions to this pattern were animals in Bethune Inlet and at the mouth of Grise Fiord where their behaviour and dispersed distributions suggested that they were feeding.

Migration pattern in Jones Sound
We saw a few small groups of belugas moving north across the mouth of Jones Sound just west of Coburg Island (21 September-9 October, Fig. 11).However, few were seen there despite intensive survey coverage.We believe that a substantial number moved west along the north coast of Devon Island into Jones Sound and crossed to the south coast of Ellesmere Island.These animals then moved east along the south coast of Ellesmere Island.
Our aerial surveys in 1978, plus satellite imagery from 1974-77, show that the landfast ice in Jones Sound began to break up at the end of July and in early August.We surveyed the Jones Sound ice edge on three occasions from 21 July-15 August 1978 and no belugas were observed.We conclude that insignificant numbers of belugas entered Jones Sound from the east in either July or August.
Ice conditions prevented entry of belugas into Jones Sound from the west in July and early August unless some wintered in the few polynias that persisted through the winter in the central Arctic (Fig. 3).We surveyed these polynias in April 1977 and March 1978 (Fig. 1) and found no belugas.We also found no whales in the large polynia at the western end of Jones  Riewe (1977) documented in Jones Sound originated in Lancaster Sound and entered Jones Sound from the southeast in September.

Rate of movement
The fall migration through Lancaster Sound was rapid in both 1976 and 1978.Our consecutive surveys along the south coast of Devon Island on 13 and 14 September 1978 allowed an estimate of the speed of migration.A herd of 2,600 belugas was recorded at Cape Hurd, just west of Maxwell Bay, on 13 September (Fig. 10B).On 14 September, slightly less than 24 h later, the nearest large herds were considerably farther east-1,860 at Cape Bullen and 1,530 between Croker Bay and Dundas Harbour (Fig. 10B).If the whales closely followed the coast and entered the various inlets between these areas, they travelled 320 km (13.5 km/h) to Cape Bullen and/or 430 km (18.5 km/h) to the area just east of Croker Bay.If they travelled the most direct route, they moved about 140 km (5.9 km/h) and/or 200 km (8.6 km/h).Since almost all whales seen in September were within 400 m of the coast, the actual speed was probably closer to the 13.5-18.5km/h values.

1979
Aerial survey effort during the autumn migration period in 1979 was concentrated along NE

113
NAMMCO Scientific Publications, Volume 4 e) Baffin Island to document narwhal and bowhead migrations (Fig. 2).Thus the survey route did not include eastern Devon Island and the mouth of Jones Sound as in 1978.
Six surveys of eastern Lancaster Sound included part of the south coast of Devon Island during September and October (Fig. 12).During all surveys except the last survey on 16 October, a few hundred belugas were seen feeding along the glacier fronts in Maxwell Bay and small numbers were detected elsewhere (Fig. 12).A survey of the south coast of Devon Island on 19 and 20 September detected 142 and 280, respectively, belugas moving eastward between Maxwell and Croker bays.No large groups of strongly migrating whales were encountered during any of the surveys in 1979 but three groups of 200-300 belugas were seen along SE Devon Island during 24-26 September.The last sightings of belugas were during the 4-5 October survey when 313 were sighted; 257 of these were in Croker Bay and the remaining whales were seen along the coast east of there.
The coasts of eastern Devon Island and all of Jones Sound were surveyed on 14 September.A group of 60 belugas was seen swimming westward near Belcher Point on NE Devon Island and groups of 14 and 90 were seen swimming eastward along southern Ellesmere Island.These observations support our hypothesis that belugas seen passing Grise Fiord in autumn entered Jones Sound along NE Devon Island and circumnavigated all or part of Jones Sound.
In summary, the main beluga migration out of Lancaster Sound was not detected during 1979, but it was likely during late September.

Calving
We did not observe calving during our extensive surveys.However, we recorded the proportions of calves among groups sighted from early May to mid-October.In 1978, belugas could not penetrate through Lancaster Sound until mid-July and the first neonates were seen along the ice edge across eastern Lancaster Sound: one on 26 June, four on 3 July, and two on 11 July.In 1979, belugas could not penetrate through Lancaster Sound until 2 August and the first neonates (3 individuals) were seen along the Lancaster Sound ice edge on 30 June.Totals of 943 and 919 belugas were classified to age during May and 1-29 June 1979, respectively, and no neonates were seen.During surveys on 30 June-2 July, 8-15 July, and 22 July 1979, 3 of 409 (0.7%), 10 of 240 (4.3%) and 13 of 95 belugas (13.7%), respectively, classified to age class were neonates.
In 1975, the first neonate was identified on 13 July in a herd of belugas swimming along the south coast of Devon Island.Two neonates were observed in offshore Barrow Strait on 18 July and considerable numbers were present in herds along the north coast of Somerset Island in late July.In 1977, a herd of 1,000+ was noted at the entrance to Aston Bay at the northern end of Peel Sound on 20 July; of 314 classified to age, 18 (5.7%)were neonates.
On 24 July 1975, 4% of 385 belugas in Creswell Bay that were classified to age were neonates (Table 5).The proportion of neonates varied from 3% to 6% from 24 July -3 August; on the latter date most belugas moved out of Creswell Bay into offshore pan ice.When they returned on 5 August, the proportion of neonates was 10% (Table 5).By 14 August total numbers present had increased substantially, and about 12% were neonates.Despite careful observation, calving was not observed in Creswell Bay in 1975.

Age and sex composition
Data on the age composition of belugas were obtained during surveys along the ice edges during spring migration and during surveys of the estuaries during summer.A total of 2,596 belugas were classified during spring surveys in 1979: 76.3% were white adults, 1.0% were calves, 8.3% were yearlings and 14.4% were grey or brown subadults.The proportion of adults was relatively constant throughout the 9 complete surveys during the spring period (cv=2.8)but calves were not seen during the first 6 surveys.
Of the many belugas occupying summering areas in Creswell Bay in 1975, 64% of the postneonates were white adults and 36% were grey or brown immatures.This ratio was fairly uniform throughout the mid-July to mid-August

114
Belugas in the North Atlantic and the Russian Arctic period (Table 5).We suspect that some adult males were absent from the estuaries and the lower proportion of adult males in the estuaries accounted for the lower (64 vs 76) percentage of adults seen in Creswell Bay in summer vs in Baffin Bay in spring.
The proportion of neonates can vary among summering areas.On 3 August, a large herd (711) at Elwin Bay contained only 2% neonates while a herd at Creswell Bay contained 5% neonates.

Feeding
We examined the stomach contents of 22 hunter-killed belugas (Table 8).Polar cod was the principal food item, although smaller numbers of Arctic cod (Arctogadus sp.-presumably A. glacialis) were taken by whales near Grise Fiord in September 1978.
During their spring migration toward summering areas belugas concentrated along ice edges in Jones Sound, SE Devon Island and Lancaster Sound and fewer were seen along adjacent ice 115 NAMMCO Scientific Publications, Volume 4   edges across Glacier Strait and along NE Devon Island.It is likely that some feeding occurred in these areas.
Belugas that arrived in the central Arctic before their estuarine summering areas became icefree did not appear to concentrate or feed along ice edges.Few were seen along the ice edge across Barrow Strait in June 1975 (see Spring Migration).Intensive aerial and ice-based studies along the coasts and ice edges of northern Barrow Strait in June and early July 1976 showed that some belugas were present from 22 June onward, but they made no attempt to feed under the fast ice edge.
There appeared to be very little feeding during occupation of estuarine summering areas.The shallow foreshore area at Creswell Bay was a barren, ice-scoured environment with extremely few benthic animals or macrophytes (Sekerak et al. 1976).Significant benthic and macrophytic communities were first encountered in the deeper water at the edge of the foreshore area.Thus belugas could not have fed during high tide when they were in very shallow water; they did not appear to feed intensively at low tide when they were along the outer edge of the foreshore, although some dove among the kelp beds.The 13 stomachs from belugas killed by Inuit in Creswell Bay on 10 August 1975 indicated that little feeding had occurred recently (Table 8).The cod otoliths in these stomachs were very worn and had probably not been ingested recently.(300).Again the whales were accompanied by large numbers of fulmars and kittiwakes; the behaviour of the three species is aptly described as a "feeding frenzy".In 1976, the whales were diving beneath rotting landfast ice that remained in the bay.This activity forced many polar cod into shallow, ice-free water where they were taken by fulmars and kittiwakes.Thousands of cod were stranded in tide pools during low tide.On 6-7 September 1977, the 300 belugas were accompanied by about 65 harp seals, 10-20 ringed seals (Phoca hispida), two or three bearded seals (Erignathus barbatus) and several hundred fulmars and kittiwakes.The birds were observed catching polar cod, and we caught many cod in gill nets.No whale stomachs were available from either of the occasions described above, but all three stomachs from whales taken at Allen Bay on other occasions were full of polar cod (Table 8).
Feeding frenzies were also observed in Maxwell Bay on 21 September 1976; 500 belugas, 80 narwhals, 135 harp seals and many seabirds were present.A major feeding area along the glacier fronts in Bethune Inlet, SE Devon Island was heavily used in September of both 1976 and 1978.On 15 September 1978, 1,800 belugas were distributed throughout the inlet and appeared to be feeding; at least 30,000 kittiwakes were also present among the whales.Glacier fronts in Croker Bay were consistently used by several hundred belugas during mid-September to early October in 1979.

NAMMCO Scientific Publications, Volume 4
We observed a group of about 950 feeding belugas at the mouth of Grise Fiord on 15 September 1978.We could not determine what they were feeding on, but we obtained 6 stomachs from belugas that were killed by hunters near the same location three days earlier.All of those whales fed primarily on polar cod (Table 8).

Population size
Our estimate of the size of the Canadian High Arctic beluga population for the late 1970s comes from surveys during autumn 1978 as they migrated out of Lancaster Sound.During this period, movements were rapid and a large fraction of the population was counted during a 2 to 3 day period.
During our surveys on 13-15 September 1978, one observer (WRK) was present on all three days and RAD was present on 14 and 15 September.We circled (at 150-250 m ASL) each herd of belugas that was encountered.Both observers were on the same side of the aircraft; the aircraft circled until each observer was confident of his independent count of the animals.Most herds were compact and in clear, shallow water, and most animals did not disappear from sight during dives except when passing under ice pans.However, herds in Bethune Inlet were dispersed and disappeared during dives in the murky water off the glacier front; these groups were circled for about 22 minutes before observers were confident of their estimates.A total of 19 herds ranging in size from four to 2,156 animals were counted in this manner.The totals of the independent estimates by the two observers differed by only 1.6% (11,520 vs 11,340) and were not significantly different (Wilcoxon matched-pair signed rank test, T=55, n=15 unequal pairs, P>0.1).Six herds of 1,000+ whales were encountered.The independent estimates for these large herds differed by 1% to 14%.Thus, we believe that the estimates derived from the 13-15 September surveys were quite accurate.
We estimated that 9,039 different belugas were observed on 13-15 September.Other sightings made throughout the fall period were then assessed to determine whether they represented whales also seen on 13-15 September (Table 7).
Using conservative assumptions, a minimum estimate of the numbers of belugas involved in the 1978 fall migration is 10,242 (Table 7).Some groups may have passed on days when we had no survey coverage and some of the groups assumed to have been repeats may actually have included different whales.Also, correction factors to account for the diving behaviour of belugas during fall migration have not been included, but this was not a significant factor in the clear shallow waters where most groups of belugas were counted.The timing of migration along the W Greenland coast has been examined and summarised by Heide-Jørgensen (1994).The animals wintering off Western Greenland begin to leave the southernmost wintering areas (63º N) in February to early April and travel through Melville Bugt in June and early July.Thus, the early animals seen along coasts and ice edges of Lancaster and Jones sounds in April and early May of 1978-79 were probably whales that wintered in the "North Water".The movements into Lancaster Sound in late June-July appear to have followed the movements along Western Greenland.
Belugas probably followed leads between the landfast ice edge and the pack ice around Melville Bugt to northern Baffin Bay.This route differed from that used by bowhead whales which leave the Greenland coast at about 71º 30´ N and cross through the offshore pack ice (the "middle pack") to the Pond Inlet area (Brown 1868).Our data from extensive 118 Belugas in the North Atlantic and the Russian Arctic offshore surveys in 1979 indicated that belugas did not cross the "middle pack", but probably followed the Greenland coast into northern Baffin Bay and then moved southwest into the Lancaster Sound area north of 74º 30´ N. Sergeant (1973) thought that belugas occupied warm-water estuaries for the purpose of calving.Subsequent studies (Sergeant andHay 1979, this study, Smith et al. 1994) (Finley 1982, St. Aubin et al. 1990, Smith et al. 1992).

Summering
The use of specific estuaries varied markedly among years, and there was much movement among areas in the same year.Creswell Bay was one of two summering areas in the central Canadian High Arctic that were used consistently by large numbers of belugas every year; the other was Cunningham Inlet (Hay andMcClung 1974, this study, Smith et al. 1994).Studies in the 1990s indicate that belugas still occupy the same estuaries and follow the same sequence of occupation that we observed in the late 1970s (Martin et al. 1993, Smith and Martin 1994, Innes et al. 2002).
During any particular survey of estuaries in 1975-77, we were able to account for at most about half of the Canadian High Arctic population of belugas as estimated from surveys during the 1978 autumn migration period.Recent telemetry studies have confirmed that individual belugas do not remain in estuarine areas for extended periods of time and they spend much of the summer period feeding in deep offshore waters (Martin et al. 1993, Smith and Martin 1994, Richard et al. 1997).

Late summer
During late summer belugas moved into Peel Sound.They were found among offshore pack ice and in estuaries along the east coast of Prince of Wales Island.Feeding was seen on several occasions and was confirmed by feeding frenzies of birds and other marine mammals that accompanied the belugas.Movements out of Peel Sound appeared to follow the east coast of Prince of Wales Island.

Fall migration
The eastward fall migration of belugas through Lancaster Sound was exclusively along the south coast of Devon Island and was highly coordinated; most of the population moved through eastern Lancaster Sound in less than a week (19-22 September 1976, 13-15 September 1978), and some individual herds sustained high speeds.The proximate factors that triggered these mass movements are unknown but were presumably related to changes in weather systems that initiated the onset of ice formation in the central Canadian High Arctic channels.
From eastern Lancaster Sound, the animals moved north along the east coast of Devon Island.Some herds moved west into Jones Sound, crossed to the south coast of Ellesmere Island, and then moved east along that coast.Other herds crossed eastern Jones Sound to SE Ellesmere Island.It is not known how far north along the east coast of Ellesmere Island the whales travelled before crossing over to Greenland waters.Vibe (1967) recorded herds moving south past Neqe, on the Greenland coast at the southern end of Smith Sound, in September and October.It is not known whether these animals were migrants from the central Canadian Arctic or animals that summered in Smith Sound and Kane Basin.
Recent telemetry studies indicate that the fall migration patterns and behaviour of Canadian High Arctic belugas are similar to those observed during the 1970s (Martin et al. 1993, Smith and Martin 1994, Richard et al. 1998a).Belugas still migrate rapidly from summering areas, stop to feed in traditional feeding areas such as Croker Bay and along SE Devon Island, and follow the south and then east coasts of Devon Island toward Jones Sound.

NAMMCO Scientific Publications, Volume 4
Rates of movement during migration along Devon Island were extremely high (5.9-18.5 km/h), but are similar to the 12-14 km/h over a 3-h period estimated for belugas approaching Cornwallis Island on 10 September 1976 (see Late summer distribution).Martin et al. (1993) recorded a "straight-line" speed of 5.8 km/h for an adult female beluga during her easterly migration through Lancaster Sound.Allowing for the whale to have followed the coastline, the speed of this whale was in the middle to upper range of the speeds calculated by us.Richard et al. (1998a) used satellite transmitters to track six belugas migrating out of Lancaster Sound during September 1995.All of their whales had periods when rates of movement exceeded 8 km/h and two whales had maximum speeds between 10 and 14 km/h (they excluded two additional whales with "outlier" speeds of 16.7 to 27.5 km/h).Vladykov (1944) stated that belugas can attain speeds of 14 to 18 km/h while being chased, but they can sustain such speeds for only short periods.Kleinenberg et al. (1964) gave a speed of 3 to 9 km/h for calmly migrating belugas.Thus, our observed speeds of belugas migrating along the south coast of Devon Island are within the range of previously observed speeds, but they are unique in the length of time that the speeds were sustained.

Calving
Our observations of neonates suggest that most calving occurred during early July to early August which was slightly earlier than narwhals in the same area (Koski and Davis 1994).Although beluga neonates have been reported as early as 31 May in the Canadian High Arctic (Cosens and Dueck 1990) and as early as late March off West Greenland (Heide-Jørgensen et al. 1993), births before late June appear to be relatively rare in the Canadian High Arctic population.

Age and sex composition
Excluding neonates, about 77% of belugas in the Canadian High Arctic population were adults and 8.4% were yearlings during the late 1970s.Adult males seem to have been underrepresented in the estuaries during our surveys as has been noted by Smith et al. (1994) in Cunningham Inlet and the Nastapoka River delta in southeastern Hudson Bay.Allowing for under-representation of adult males, the proportion of neonates seen at the end of the calving season is consistent with a triennial reproductive cycle as noted by Sergeant (1973), Burns and Seaman (1986) and Heide-Jørgensen and Teilmann (1994).

Feeding
Belugas were present in the central Canadian Arctic channels for about two months in the summer and feeding apparently occurred throughout this period.Their principal food appeared to be polar cod, which is an important component of the diets of most seabirds and marine mammals in the Canadian High Arctic (Bradstreet 1980, 1982, Bradstreet et al. 1986, Welch et al. 1992).Belugas consistently utilised specific coastal feeding sites in late summer and during fall migration (e.g.Allen Bay, Croker Bay, Bethune Inlet).Use of these same areas has been documented by later studies by Welch et al. (1992), Smith and Martin (1994) and Richard et al. (1998a).In addition, belugas fed opportunistically in areas where polar cod occurred irregularly such as in inner Creswell Bay in 1976.
Belugas also fed offshore either under pack ice, where polar cod may have concentrated as the pack ice melted (e.g.Prince Regent Inlet off Creswell Bay, Peel Sound), or in deep offshore waters where they may have fed on polar cod, Greenland halibut or invertebrates (Martin et al. 1993).Recent telemetry studies of beluga movements during summer indicate that this offshore deep-diving behaviour is more common than previously thought (Martin et al. 1993, Smith and Martin 1994, Richard et al. 1997).
The only published data on the diet of belugas in the Canadian High Arctic are observations by Freeman (1968) on animals trapped by ice in Jones Sound, by Sergeant and Brodie (1975) of belugas feeding along a disintegrating ice edge in Maxwell Bay in early August 1973, and by Welch et al. (1992) on belugas feeding on a concentration of Polar cod near Resolute Bay during August 1985.The Jones Sound whales maintained small breathing holes in the ice from 25 November 1966 to 1 April 1967 and may not be representative because the animals 120 Belugas in the North Atlantic and the Russian Arctic were trapped by landfast ice.Of the 46 stomachs examined from hunter-killed whales, over half were empty and the rest contained only small amounts (about 0.5 l) of food, primarily polar cod and the mysid crustacean (Boreomysis nobilis).The belugas observed by Sergeant and Brodie (1975) were believed to be feeding on polar cod, although this could not be confirmed.The belugas of Welch et al. (1992) appeared to be feeding on dense schools of polar cod and returned to the same area to feed numerous times over a 4-day period.
Population size Sergeant and Brodie (1975) conducted aerial surveys in early August 1973; these "showed at least 10,000 animals…" in the High Arctic.Sergeant and Hay (1979) present the data from these (8,990 belugas estimated) and other surveys; they state (p.10) "the total in various years approaches but does not exceed 9,000 animals".
Our best estimate of the size of the High Arctic beluga population in the late 1970s comes from surveys conducted during the peak of fall migration when virtually all of the animals were present along the south and east coasts of Devon Island and the south coast of Ellesmere Island.Surveys of the east and south coasts of Devon Island on 19-22 September 1976 recorded 8,393 belugas (Fig. 10A).More complete surveys in 1978 found 9,039 animals on 13-15 September (Fig. 10B) and a total of at least 10,242 during the entire fall migration (Table 7).The latter estimate is negatively biased; we excluded many herds that might have been counted twice even though some were seen several days and up to 350 km apart.In addition, because of the high speed of fall migrants our weekly surveys probably missed some herds completely.Although correction factors to account for the diving behaviour of animals have not been incorporated into the estimate, we believe that any bias due to diving is small because of the behaviour of the animals during migration.We estimate that at least 10,250-12,000 belugas were involved in the fall migration out of the central Canadian Arctic in 1978.This estimate includes some, but probably not all, of the neonates born in July and early August of 1978August of . R.R.C.S. (1977) ) estimated that 15,500 belugas were present in eastern Lancaster Sound on 22 September 1976; however, they warned that the underlying assumptions for their extrapolation procedure (i.e., either random or uniform distribution throughout the area) were not met.They recorded 2,736 whales within 0.8 km of the south coast of Devon Island but none in offshore waters or along the southern margin of the sound; therefore, it was not appropriate to extrapolate densities to those latter areas.R.R.C.S. (1977), correctly, did not take the estimate of 15,500 seriously and it is only discussed because it has been uncritically cited.Smith et al. (1985) surveyed summering areas of the central Canadian High Arctic during July and August 1981.Their July estimate was 10,368 (6,264 to 18,564), which is essentially identical to our estimate from the 1978 fall migration period.Their August estimate was 8,802 (6,771 to 14,067).The Smith et al. (1985) estimates do not include corrections for whales missed by observers or for whales that were below the surface at the time of the survey.Innes et al. (2002) conducted a survey of the central Canadian High Arctic summering areas in 1996.Their estimate of 21,213 (95% CI 10,985 to 32,619) is unbiased and accounts for whales missed by observers, and the diving behaviour of whales.Their estimate corrected for perception bias was 10,347 (cv=0.28),which is only slightly lower than the two estimates of 10,368 and 8,802 by Smith et al. (1985) for the summer of 1981.
Recent winter surveys of belugas along Western Greenland suggest that the population wintering there may have declined by as much as 62% (Heide-Jørgensen and Reeves 1996).However, surveys conducted in the summering areas in 1981 (Smith et al. 1985) and 1996 (Innes et al. 2002) suggest that the decline, if one has occurred, is smaller than the winter surveys suggest.
In order to manage the Canadian High Arctic beluga population managers need to know either the population size or whether the population is increasing or decreasing.Surveys like those conducted by Innes et al. (2002) provide 121 NAMMCO Scientific Publications, Volume 4 122 Belugas in the North Atlantic and the Russian Arctic an unbiased estimate of population size but the confidence interval of the estimate is too broad to assess changes in population status over relatively short periods of time.For example, the 1981 aerial surveys conducted by Smith et al. (1985) give a similar point estimate of population size (10,400) as our 1978 surveys (10,242-12,000) but the CI of the Smith et al. (1985) estimate is 6,300-18,600.We are unable to place an upper limit on our estimate, but we counted at least 10,242 different whales during our survey so the lower limit is 10,242 (neither estimate includes correction factors to account for diving behaviour or for groups that might have been missed).Since the lower limit of population size is conservatively used for risk-averse management decisions, there is a need to repeat the census of the Canadian High Arctic beluga population using our methods.If the census were completed during a year when belugas from that population were radio tagged and had TDRs (time-depth-recorders), the estimate would benefit from greater certainty about the movements of specific groups between days and hence a better estimate of total numbers of animals in the census area.TDRs would provide data to correct counts for animals below the surface at the time of the counts and permit an estimate of population size and confidence intervals.
North Atlantic and the Russian Arctic

Fig. 1 .
Fig. 1.Main aerial survey routes during the overwintering and spring migration periods (March-July), 1974-79.Additional surveys of landfast ice in the Barrow Strait, Peel Sound and Viscount Melville Sound areas (west to Melville Island) are not shown.
account for whales missed by observers or to account for the diving behaviour of belugas.The major influx of belugas into the summering area at Creswell Bay, Somerset Island, occurred between 21 and 24 July; only 241 were seen there on 21 July but 3,376 were counted on the 24 th .Another survey of the offshore transects revealed that about 2,700 belugas (no corrections) were still present in eastern Barrow Strait on 27-28 July.Thus, the principal influx of belugas into the central Arctic occurred from 15-28 July in 1975.1976 In 1976, like 1974, eastern Barrow Strait was ice free by late May.During 11 surveys along the northern edge of the strait 1 June-3 July, we saw 222 belugas near SE Cornwallis Island on 22 June, 98 and 114 in the same area on 28 June and 3 July, and 123 near eastern Griffith Island on 28 June.None were seen during three earlier surveys (8, 19 and 29 June) along the ice edge between Griffith Island and NW Somerset Island.Thus, the appearance of belugas along the northern margin of Barrow Strait in 1976 100 Belugas in the North Atlantic and the Russian Arctic

Fig
Fig. 6.Distribution of belugas in NW Baffin Bay and eastern Lancaster Sound, 17-22 May 1979.Plotted as in Fig. 4.

Fig
Fig. 11a-e.(Continued on next page).Distribution of belugas sighted in eastern Lancaster Sound, along easternDevon Island and along eastern EllesmereIsland,  September and early  October 1978.Plotted as in Fig. 4.

Fig
Fig. 12. Distribution of belugas sighted in eastern Lancaster Sound and along eastern Devon Island, September 1979.Transect width was 0.8 km and density 0.2/km2 represents a sighting of a single whale; otherwise as plotted as in Fig. 4.
of items per stomach including empty stomachs.c Maximum no. of items in a single stomach.d Otoliths were used to identify fish; tabulated values are numbers of otoliths found (about twice the number of fish).e Present in trace amount.

Table 1 .
Shore-based watches for migrating marine mammals.For locations, see Fig.2.

Table 2 .
Numbers d Survey incomplete due to fog. e Ice-edge broke up between 12 and 20 July.
influxes of belugas entered Lancaster Sound during early July but bad weather prevented completion of surveys during early July to early August.Numbers along coasts and ice edges increased throughout the season but the Lancaster Sound and eastern Devon Island ice edges were not surveyed during the mid-July period.Densities on offshore transects in Lancaster Sound were 0.054/km 2 during 19-23 June, 0.202/km 2 during 28 June-2 July, and 0.234/km 2 during 8-15 July.In addition, the location of the ice edge moved west between 2 and 8 July so that the offshore area increased between the latter two surveys.Based on densities observed in offshore Lancaster Sound, the number of belugas present there increased from 483 during 19-23 June to 2,526 during 8-15 July.In summary, slightly larger numbers of belugas were present in NW Baffin Bay in mid-May of 1979 than in 1978.Small numbers of belugas entered the study area during May and June and the main movement appears to have occurred during July, but incomplete survey coverage did not permit accurate determination of the timing.Belugas entered the study area primarily through offshore waters north of 74º 30´.results complementary to those from eastern Lancaster Sound in the "normal ice year" of 1976.Also, in 1976 we surveyed the ice edges and coasts along the northern edge of Barrow Strait from SW Devon Island to Griffith Island,

Table 3 .
Numbers and densities of belugas recorded during aerial surveys of northwestern Baffin Bay and eastern Lancaster Sound, 9 May-15 July 1979.The survey route is the hatched line on Fig.1(see also Fig.6).Transect width was 0.4 km for surveys at 50 m ASL and 0.8 km for surveys at 90 m ASL.The length of survey (in km) was 1.25-2.5 times the value listed in this column, depending on the number of km of survey flown at each altitude during each survey period.
bExcluding whales seen adjacent to the transects.c Transects along coasts and ice-edges were centered 200 m off the interface; nearshore transects were 1200 m off the interface.d Survey of coasts and ice edges in and north of Lancaster Sound incomplete due to fog. e The ice-edge broke up on 2 August.

Table 4 .
Numbers and distribution of belugas in the central Canadian High Arctic, July -August 1975.
In 1976, the peak westward migration through Lancaster Sound occurred from late June to mid-July.However, the main summering estu-101 NAMMCO Scientific Publications, Volume 4 a Area numbers correspond to locations on Fig. 3. b '-' means not surveyed.cSurveyed on 26 July.dEstimated numbers based on 10% survey coverage (see Fig. 6).e Parentheses denote incomplete survey.ariesand bays remained ice covered until late July.On 14 July, many (900+) belugas were observed south of the landfast ice in the southeastern part of Creswell Bay (A.Clarke, cited in Finley and Johnston 1977).On 28 July, Clarke saw "thousands" from the edge of the landfast ice in Creswell Bay south to Bellot Strait.The ice had disintegrated sufficiently on 29 July to allow many whales to enter the northern side of Creswell Bay.In 1978, the migration through eastern Lancaster Sound was delayed by ice until sometime between 12 and 17 July.Ice remained in the summering areas until late July.Belugas arrived in Creswell Bay on 26 July (B.Kemper, 102 Belugas in the North Atlantic and the Russian Arctic

Table 5 .
Numbers and age composition of belugas in Creswell Bay, Somerset Island, July -August 1975.
a Surveys were conducted from a slow moving helicopter at an elevation of 90-150 m.An attempt was made to classify all animals within 400 m of the aircraft.Neonates were identified by pale coloration, relative size, and association with females.Immatures ranged from small dark-colored calves to large light brown or gray animals.b Counts at 04:45 and 11:55 (CDT); both are negatively biased due to turbid water.c This figure not considered typical because of small sample size.

Table 4
The status of belugas along the west coast of Brodeur Peninsula, Baffin Island, is poorly understood.Heyland (cited by Sergeant and Hay105NAMMCO Scientific Publications, Volume 4

Table 6 .
Distribution of belugas along the north and east coasts of SomersetIsland,  1976 and 1977.Area numbers correspond to locations on Fig.3.bBased on counts from aerial photographs (6x7 cm format).c1,500 were observed in Cunningham Inlet on 16 July 1977.1979)found 2,250 along this coast in 1973 and 857 in 1974; the survey dates and coverage are unpublished.We surveyed the entire east coast of Brodeur Peninsula (south to 72º 03' N) on 30 July 1977 and found only 315 belugas.
a 106 Belugas in the North Atlantic and the Russian Arctic 19 August.It appears that no more than a few hundred belugas penetrated into waters north and northwest of eastern Barrow Strait in late summer in most years.Peel Sound area The area used most consistently by belugas in late summer appeared to be Peel Sound and northern Franklin Strait.In 1974, we saw about 150 belugas along the landfast ice edge across northern Peel Sound on 3 August, and 500 in Coningham Bay along SE Prince of Wales Island on 3 September.In 1975, a herd of 169 swam strongly west along the north coast of Somerset Island toward Peel Sound on 19 August.A reconnaissance survey of the northern half of Peel Sound on 23 August 1975 found 200 belugas in an area of extensive pan ice off 107 NAMMCO Scientific Publications, Volume 4

Table 7 .
Summary of beluga sightings in the Canadian High Arctic during fall migration, 1978.See Fig. 10B for sighting locations.
Very few belugas were seen far offshore or anywhere south of Lancaster Sound during the fall 111 NAMMCO Scientific Publications, Volume 4 survey period (3 September-10 October).Only three were recorded more than 3 km from shore; they were swimming SE about 15 km north of Bylot Island on 3 October.Only two belugas were recorded south of this location during this period; they were swimming west along the south coast of Bylot Island on 7 October.None were observed passing Cape Adair (NE Baffin Island) during a bowhead whale (Balaena mysticetus) migration study 112 Belugas in the North Atlantic and the Russian Arctic Sound (Hell Gate and Cardigan Strait) during surveys in June 1974 or in June and July 1977.Our only record of belugas west or northwest of western Jones Sound is of 7 about 25 km SE of Cornwall Island on 14 August 1976.Thus, we conclude that few, if any, belugas entered Jones Sound from the west; we suspect that most of those that

Table 8 .
Diet of belugas in the Canadian High Arctic.
D DI IS SC CU US SS SI IO ON N