First independent feeding of harp seal (Phoca groenlandica) and hooded seal (Cystophora cristata) pups in the Greenland Sea

Data were collected from harp seal (Phoca groenlandica) and hooded seal (Cystophora cristata) pups belonging to the Greenland Sea (or "West Ice") stocks in 1995-1997. Pups of both species were observed to feed independently shortly after weaning, and their first food was almost exclusively crustaceans. Parathemisto sp., particularly P libellula, dominated the diet of both the harp and the hooded seal pups, but the diet also contained sympagic amphipods of the genus Gammarus. Krill (Thysanoessa sp.) was of minor importance as food for seal pups in 1995, but occurred more frequently in the diet of both species in 1996 and 1997. Considerable niche overlap may suggest some interspecific competition between harp and hooded seal pups in the West Ice. Haug, T, Nilssen, K.T. and Lindblom, L. 2000. First independent feeding of harp seal (Phoca groenlandica) and hooded seal (Cystophora cristata) pups in the Greenland Sea. NAMMCO Sci. Publ. 2:29-39.


Introduction
Three stocks of harp seals (Phoca glVenlandica) inhabit the North Atlantic Ocean (Sergeant 1991).One of the stocks is distributed in the Greenland and Norwegian Seas, breeding and moulting in the Greenland Sea pack-ice near Jan Mayen (the West Ice).The West Ice stock breeds on the fringe of the winter ice (usually snow covered floes of approximately 1 m thickness) seaward of the heavier multi-year ice along the East Greenland coast in an area where water depths range between 1000 and 3000 m.The main pupping period is in late March (Sergeant 1991 , 0ritsland and0ien 1995).Hooded seals (Cystophora cristata) also breed in the West Ice area, mainly in mid March (Rasmussen 1960(Rasmussen , 0ritsland 1964(Rasmussen , 0ritsland and 0ien 1995)).The duration of the lactation period for harp seals in the West Ice is probably 10-12 days (Sivertsen 1941, Rasmussen 1957, Sergeant 1991).The period is assumed to be considerably shorter for hooded seals, possibly within a range of 3-5 days, as observed for the species in the Northwest Atlantic (Bowen et al. 1985, Kovacs andLavigne 1992).Since both harp and hooded seals are present in the same areas of the West Ice through overlapping breeding seasons, pups of the two species commence independent feeding in a situation that might result in competition.Adult harp seals have a diet that differs from that of hooded seals in the West Ice (Potelov et al. this volume).It is, therefore, of interest to establish whether the same is true for the pups.

Sampling
The appropriate period for collection of seal pup data was assumed to be the first month after weaning, i.e. in April.This is the main sea ling season in the West Ice, and the co llection of pups could therefore be carried out from Norwegian sealing vessels during the commercial hunting seasons of 1995, 1996and 1997. In 1995, com, commercial catches of pups were prohibited in the West Ice, and a special permit was obtained to take harp and hooded seal pups for scientific purposes.The permit stipulated that sampled pups must be weaned.Harp seal pups were therefore either in the process of shedding the white foetal coat ("ragged jackets", a stage which lasts 1-2 weeks) or completely moulted pups ("beaters", with a short spotted pelage) (see Sergeant 1991).The hooded seal pups shed their foetal hair in the uterus and are born with the shorthaired pelage that has given them the name "bluebacks" (Lavigne and Kovacs 1988), and samp les from this species were obtained from weaned bluebacks.After the 1995 season, commercial catches of weaned pups were permitted and pup samp les could therefore be obtained from commercial catches.
In 1995 , samp ling was carried out from two Two scientists collected biological samples on each vessel.For each pup sampled, notes were made on species, date, time of the day, position, sex and stage of development.The stomach and large intestine of each pup were removed, and frozen for subsequent ana lysis in the laboratory.

Biological collections and analyses
In the laboratory, the stomachs and intestines were thawed and then cut open.The stomach contents were removed and the intestinal contents were flushed out using fresh water.Crustaceans were separated from fish and identified to the lowest possible taxon accordi ng to Enckell (1980).A crude estimate of the number of crustaceans was obtained by counting carapaces.Otoliths were collected and identified to the lowest possible taxon using the keys developed by Hark6nen (1986).The number offish of each species was calculated by dividing the number of otoliths by two.

-
2) the relative frequency of occurrence of a prey category, calculated as a numerical fraction of the total number of all prey categori es (% numerical frequency).
Possible niche overlap between the two species was assessed using a multiplicative measure suggested by Pianka (1 973 ): where Pij and Pi k are the proportions of the ith resource used by the jth and "-1:h species, respectively.The measure is symmetric and gives a single overlap value between 0 (no overlap) and 1 (complete overlap) for a niche overlap pair.
For calculations of the relative frequencies of occurrence of prey and of niche overlaps, the observed prey taxa were combined in five different categories: Parath emisto spp.(at least tlu'ee species of thi s hyperiid amphipod genus), krill (mainly Thysanoessa spp., and to a very small extent M egany ctiphanes norvegicus), Gammarus sp.(mainly G. wilkitzkii), other crustaceans (Eusirus sp., Lyssianasida, unidentified crustaceans) and various items (fi sh, squid, other remains).

Harp seals
In 1995, stomachs and intestines of harp seal pups (ragged jackets as well as beaters) were examined in tlu'ee sequential periods from 10 to 30 April.In 1996 and 1997, however, samples were generally collected over a more restricted period, and included only beaters.The stomachs were usually empty, but the incidence of empty intestines was low (0% to 15%).The percentage of occurrence of fo urteen groups of prey is given in Table 1.Hyperiid amphipods of the genus Parathemisto occurred in 44% to 100% of the intestines, and P libellula was particularly common .Amphipods belonging to the genus Gammarus sp., in parti cular the species G. wilkitzkii, were also encountered fi-equently in 1995 and 1996, but more rarely in 1997.The prey group s des ignated Pa rathemisto sp. and Gammarus sp.mo st pro bably also include remains of P libellula and G. wilkitzkii, which could not be identif ied to species.Some other amphipods also occurred occasionally in the intestines, including Eusirus sp. and representatives of the fami ly Lysianass idae.Krill of genus Thysanoessa were rarely observed in the gastrointestinal system of harp seal pups in 199 5, while in 1996 and 1997 they occurred much more frequently (56% to 73 %).Contents from stomachs generally resembled those observed in the intestines.
The dominant groups of prey by relative numerical frequency is illustra ted by Fig. 4. Parathemisto sp., in all probability mostly P libellula, was the dominant prey item in stomachs and intestines of both ragged jackets and beaters in all sampling years, contributing with 64% to 90%.The contribution of the genus Gammarus was notable in 1995 (5 % to 25%), while krill contributed signi ficantly in 1996 and 1997 (1 0% to 29%).Other food items were of minor importance.

••
• .""." 1995 • -----1996 • ---1997 sumed.P libellula was the most frequent prey species, occurring in 10% to 25% of the stomachs and 46% to 87% of the intestines.In 1995 and 1996, G. wilkitzkii was also of some importance, occurring in 5% to 9% of the stomachs and 15% to 33% of the intestines.Co-occurrence of Parathemisto and Gammarus was observed in 20% of the 1995 and 1996 samples.In 1997, Gammarus was fo und in only 6% of the intestines and not at all in the stomachs.Krill of the genus Thysanoessa occurred quite frequently in blueback intestines in 1996 (31 %), but more rarely in 1995 and 1997.
In terms of numerical frequency of occurrence (Fig. 5), Parathemisto sp., mainly P libellula, dominated with a contribution of 82% to 95% to the blueback gastrointestinal contents in all sampling years.

DISCUSSION
This study demonstrates that harp seal pups are able to find prey and feed independently immediately after weaning, and that hooded sea l pups do so at least two weeks ater weaning.In both species, most stomachs were empty, but food remains were found in the intestines of nearly all the harp seal pups, and a large fraction of the hooded seal pups.The sign if icance of this early independent feeding as an energetic supplement is, however, not known.The large frequency of empty stomachs could be indicative of rapid digestion by the pups, but may also reflect a 32  - ----------------------------------------------------33 NAMMCO Scientific Publications, Volume 2 Examination of stomach and intestinal contents revealed that the first food of both harp and hooded seal pups was almost exclus ive ly restricted to crustaceans.The post-weaning diet of the harp seal pups was particularly characterized by amphipods of the genus Parathemisto.In 1995, krill appeared to be rather unimportant, while amphipods of the genus Gammarus, in particular G. wilkitzhi, contributed significantly to the diet.The situation was different in 1996 and 1997 when krill occurred much more frequently and Gammarus sp.more rarely.These changes in diet may reflect annual variation in the relative abundance of prey organi sms in the Greenland Sea.Certainly, the differences in sampling sites due to variations in ice configurations between the years in question (Fig. 3) may have been a contributory factor.Unfortunately, there is no information avai lable on prey abundance in the West Ice area.
Previous information about the feeding habits of ragged jackets and beaters from the West Ice is very scarce, although Rasmussen (1957) observed beaters returning from dives with polar cod in their mouths in April.Only a few individuals of polar cod were recorded in the diet of harp seal pups in our study.Adult West Ice harp seals are known to feed mainly on Parathemisto sp. both during spring and early summer (April-June) (Potelov et al. this volume).The importance of crustaceans, P libelluia in particular, in the diet is also reported for the Barents and White Sea stock of harp seals.Weaned pups from this stock start their first independent feeding in the southeastern Barents Sea, and have been observed to have an early post-weaning diet dominated by krill , but with considerable amounts also of Parathemisto sp.(Haug et al. 1996).Parathemisto sp. and krill are important for pups and for older seals as well in the Barents Sea in late summer and early autumn (July-September), whi le in October there is a shift in the diet from crustaceans to fish (Nilssen et al. 1995, 1998, Lindstr0m et al. 1998).Harp seals N=60 N=17 N=6 100 - --------------------------------------------------35 NAMMCO Scientific Publications, Vo lume 2 have been reported to consume amphipods, in particular P libel/ula , in considerable amounts also in the Arctic areas of Greenland and Canada (Sergeant 1973, Davis et al. 1980, Finley et al. 1990, Kapel 1995).
The initial blueback diet appeared to be very similar to that observed for the harp seal pups: a dominance of Parathemisto (almost exclusively P libel/ula) , with some contribution from Gammarus sp.(mainly G. wilkitzkii).Krill appeared in the 1996 and 1997 blueback diet, while it was absent in 1995, a situation similar to that observed for the harp seal pups.There is some evidence that the abundance of non-crustacean prey items such as polar cod and the cephalopod Gonatus fabri ci may increase  H0st (1948).Both polar cod and in particular G. fabricii are known to be important food items for adult hooded seals in the West Ice (Potelov et al. 2000), although cephalopods and crustaceans were predominant food items for young-of-theyear hooded seals in Southeast Greenland in early September 1991 (Kapel 1995).In other areas, hooded seals are known to be mainly piscivorous (Ross 1993 , KapeI1995) .
In the Northwest Atlantic, P libellula has been shown to be a dominant species in cold water ----------------------------------------------------Minke whales, hGip and hooded seals: Major predators in the North Atlantic ecosystem plankton communities in the upper 50 m of the water column (Percy 1993).The dominance of Parathemisto sp. in the diets of both harp and hooded seal pups in the West Ice confirms previous suggestions that these amphipods are important links in the food chain between herbivorous zooplankton and marine fish , birds and mammals in most Arctic waters (Dunbar 1942, 1946, 1957, Sergeant 1973, Davis et al. 1980, Bradstreet and Cross 1982, L0I1I1e and Gulliksen 1989, Aijad and Gj0sreter MS 1990, Mehlum and Gabrielsen 1993).In the Barents Sea, krill species of the genus Thysanoessa are also known to be important links between herbivorous zooplankton and fish, sea birds and mammals (Sakshaug et al. 1992), and the present investigation seems to indicate that krill may, at least occasionally, be of some importance as food for seal pups in the Greenland Sea as well.

-
G. wilkitzkii is an autochthonous sympagic amphipod of Arctic areas (Gulliksen and L01111e 1991), and its occurrence in the diet of West Ice harp and hooded seal pups suggests that the pups may perform their first feeding excursions at the ice-water interface.G. wilkitzkii is most abundant in the perennial sea ice zone areas (Gulliksen and L0nne 1991), and this may explain why the species was present in the diets of the seal pups in the multi year sea ice areas of the West Ice, while it is known to be absent from the diets of harp seal pups in the seasonal sea ice areas of the southeastern Barents Sea (Haug et al. 1996).Given its high degree of co-occurrence with G. wilkitzkii in the seal stomachs and/or intestines, and its frequent occurrence close to the surface of sea ice, P libellula may also have been taken by the seal pups as they foraged near the undersurface of the ice floes.However, P libellula is classified as belonging to the planktonic fauna, due both to its lack of physical association with sea ice and to its frequent pelagic occurrence in ice-free cold water areas (Gulliksen andL0nne 1991, L0nne andGulliksen 1991).This prey may therefore have been consumed by seals foraging in open waters and/or at greater depths.
The general impression of similarity, gained from both the frequency of occurrence (Tables 1  and 2) and relative frequency of occurrence (Figs 2 and 3) of prey items between the two seal species, was confirmed by the niche overlap values, which were very close to unity in all years of investigation.The high degree of overlap along the niche dimensions describing the diet indicates that the behaviour of harp and hooded seal pups during their first independent feeding in the West Ice must be quite similar.Given their coinciding geographical and temporal appearance in the area, some interspecific competition between the two species may be inferred.

Table I :
Frequencies of empty stomachs and intestines, and percentage occurrence of prey items in stomachs and/or intestines of West Ice harp seal pups (RJ = ragged jackets, B = beaters) captured in 1995 during the periods 10-16 April, 17-23 April and 24-30 Apri l, respectively, in 1996 during the period 19-23 April , and in 1997 during the period 22-24 April.= number of stomachs and/or intestines examined."Others" includes remains such as hair and algae.

Table 2 :
Frequencies of empty stomachs and intestines, and percentage occurrence of prey items in stomachs and/or intestines of West Ice hooded seal pups (bluebacks) captured in 1995 during the periods 3-9 April , 10-16 April, 17-23 April and 24-30 April, respectively, in 1996 during the period 9-23 April, and in 1997 during the period 17 -18 April.N = number of stomachs/intestines examined."Others" includes remai ns such as hair and algae.