Movements and site fidelity of harbour seals (Phoca vitulina) inKattegat, Denmark, with implications for the epidemiology of thephocine distemper virus

Rune Dietz1*, Jonas Teilmann1, Signe M. Andersen1,2, Frank Riget1, and Morten T. Olsen3

1Department of Bioscience, Aarhus University, Frederiksborgvej 399, DK-4000 Roskilde, Denmark2Institute of Biology, University of Southern Denmark, Campusvej 55, DK-5230 Odense M, Denmark3Department of Genetics, Microbiology and Toxicology, Stockholm University, S-106 91 Stockholm, Sweden

*Corresponding Author: tel: +45 2125 4035; fax: +45 8715 8690; e-mail: rdi@dmu.dk

Dietz, R., Teilmann, J., Andersen S. M. Riget, F., and Olsen, M. T. 2012. Movements and site fidelity of harbour seals (Phoca vitulina) in Kattegat,Denmark, with implications for the epidemiology of the phocine distemper virus – ICES Journal of Marine Science, doi:10.1093/icesjms/fss144.

Received 10 February 2012; accepted 17 July 2012.

Twenty-seven harbour seals were caught and tagged at the island of Anholt in central Kattegat, Denmark, the epicentre of the phocinedistemper virus (PDV) outbreaks in 1988 and 2002 that killed 50 –60% of the populations. The satellite tagging shows that harbourseals from Anholt moved widely across Kattegat with a maximum distance of 249 km from the tagging haul-out site. Overall, femalestravelled over a wider area compared with males [90% kernel home range (KHR) females, 5189 km2; males, 3293 km2). KHR calculatedfor yearlings (6414 km2) is larger than for subadults (2534 km2), which again is larger than for adult seals (1713 km2), showing a strongsite fidelity, indicating limited gene flow between haul-out sites. Distances moved and home range sizes increased across autumn,peaked in February–March, and decreased through spring. During the breeding season in spring, all seals were very stationaryaround Anholt. The onset of the PDV epizootics in 1988 and 2002 took place when the Anholt harbour seals congregate on theIsland during April. Anholt seal were also documented to have contact with infected seal locations at Hesselø, Læsø, and theSwedish west coast, although this contact takes place during winter prior to the documented summer outbreaks.

Keywords: age-related seasonal behaviour, area use, dispersal, distribution, harbour seal, haul-out site, kernel home range, PDV spreading,Phoca vitulina, satellite telemetry.

IntroductionThe management and conservation of marine species requires adetailed understanding of species movement behaviour and howthis may vary temporally, spatially, and individually. For marinespecies, such assessments are often difficult because direct observa-tions are hard to obtain and because generally little is known aboutthe physical and environmental features that promote or discour-age movements. Satellite telemetry provides a powerful techniquefor tracking the movements of marine species and has been usedwidely on pinnipeds, whales, turtles, fish, sharks, and seabirds(Block et al., 2011). Telemetry allows for a detailed contemporaryassessment of individual movement and space use patterns which,unlike most genetic approaches, provides the temporal and spatialresolution necessary for designation of local marine protectedareas.

The harbour seal (Phoca vitulina L.) is the most common sealspecies present in the cold temperate waters of the NorthAtlantic. They come ashore to give birth, rest, and moult (e.g.Bonner, 1972; Bigg, 1981). Populations of harbour seals are con-sidered non-migratory but, as a species, use a wide range of habi-tats across their geographical distribution (Bigg, 1981). Usuallythey are found as concentrated colonies on sandy beaches or asmore dispersed groups along rocky shores (Bigg, 1981). Theharbour seal forage on a wide range of cephalic and fish species.In Kattegat, it is one of only a few apex predators in the marineecosystem and may thus play a key role in maintaining local eco-system stability and productivity.

The harbour seals in Kattegat have experienced substantialpopulation declines at regular intervals over the past century(Olsen et al., 2010). From 1889, for 40 years the Danish and

# 2012 International Council for the Exploration of the Sea. Published by Oxford University Press. All rights reserved.For Permissions, please email: journals.permissions@oup.com

ICES Journal of

Marine ScienceICES Journal of Marine Science (2012), 69(10), 1–10. doi:10.1093/icesjms/fss144

ICES Journal of Marine Science Advance Access published October 24, 2012 by guest on June 5, 2016

http://icesjms.oxfordjournals.org/

Dow

nloaded from

Swedish harbour seals were targets of massive hunting campaignsto eradicate the species due to its competition with fisheries (see,for example, Andersen et al., 2007). The declines experiencedduring this period were so severe that increases in abundancefirst were accomplished with the initiation of total protectionand closed seal reserves in the late 1970s (Olsen et al., 2010).Within the past decades, the harbour seal populations of northernEurope have experienced repeated disease outbreaks caused by thephocine distemper virus (PDV) in 1988 and 2002 (Dietz et al.,1989; Harkonen et al., 2006) and an unknown pathogen in 2007,with high mortalities as a consequence (Harkonen et al., 2008).

The 1988 and 2002 PDV epidemics constitute the largest massmortalities ever observed in a marine mammal, resulting in thedeaths of . 23 000 harbour seals in 1988 and 30 000 in 2002 inthe Kattegat and Skagerrak and along North Sea coasts (Dietzet al., 2002; Harkonen et al., 2006). On both occasions, the epi-demics started at the Danish island of Anholt in central Kattegatand subsequently spread to adjacent colonies in a stepwise fashion.

In this study, we use information on the at-sea distribution ofsatellite-tagged seals to identify the areas used by harbour sealsfrom the haul-out site at Anholt and how habitat use and move-ment patterns may vary with sex, age, and season. Moreover, cov-ering the entire annual cycle of movements from the epicentre ofthe two PDV outbreaks and one of the most important harbourseal haul-out sites in Europe allows detailed inference of the rela-tionship between harbour seal movement patterns and the epi-demiology of PDV.

MethodsThe study areaThe island of Anholt is located in the central part of the Kattegat insouthern Scandinavia (Figure 1). The average water depth inKattegat is �20 m, with a maximum depth of � 100 m. The

deeper waters are located along the Swedish west coast in theeastern part of Kattegat. The waters around Anholt are mainlyshallow (down to 10 m) except for the area southeast of Anholtwhere water depth extends to 25 m. Sandy bottom occursmainly in the shallow areas, and mud dominates in deeper areasof the Kattegat (Harkonen, 1987). Tides are barely detectable inthe Kattegat, but strong wind-driven currents may occur.

During summer, harbour seals mainly haul-out on the easterntip of Anholt in the Totten seal sanctuary (56.738N; 11.668E,Figure 1–3). The seal sanctuary was established in 1990 andaccess is prohibited year-round (Jepsen, 2005). The seals mainlyprefer the eastern tip of Anholt during summer. Here a sealreserve yields protection when the island is visited by a highnumber of tourists (Andersen et al., 2011). During winter, theseals often haul-out near the northern shores of Anholt. The sealsite at Anholt has been surveyed during moult (late August) forthe past 25 years, and the 2010 estimate of harbour seals aroundAnholt (corrected for seals in the water) is 1800. Together withHesselø, this haul-out location has the highest number ofharbour seals in Denmark (Olsen et al., 2010; Teilmann et al.,2010).

Satellite telemetryNetting of the sealsChains of two to five 100 m nets were set 25–300 m from the shoreat 3–8 m depth. Green, blue, and black coloured monofilamentand twisted nylon nets with mesh sizes of 180 × 180 mm wereused with equal success. Twine size of 0.60 mm, a net depth andlength of 2.5 × 100 m, a float line with an uplift of 38 g m21,and a lead line with 23 g m21 weight were used.

The nets were ‘fishing’ continuously for up to 5 d and tendedregularly for entangled seals. Due to the large mesh, the floatline, and the thin lead line, the seals were able to put their head

Figure 1. Map showing the satellite tracks of the 27 harbour seals tagged in September 2005 and 2006, and in April 2008.

Page 2 of 10 R. Dietz et al.

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

through the mesh and rest at the surface until they were disen-tangled, tagged, and released.

In total, 27 harbour seals were captured during three fieldperiods. Ten seals were captured and tagged during 14–17September 2005, nine seals during 21–22 September 2006, andeight seals during 25–28 April 2008.

Handling of the sealsUpon capture, seals were lifted into the boat using a pole net,transported to shore, placed in a hoop net, and fixed on a speciallydesigned board with mounting straps. The fur was cleaned withacetone and the transmitter was attached to the fur using quick-setting two-component epoxy glue (Araldite 2012). During thehandling of the seals, the species, sex, length, circumference, andweight were recorded (Table 1). In addition, seals were taggedwith conventional flipper tags and freeze branded on each side,while hair and tissue biopsy samples (from the flipper tag hole)were taken for later contaminant and genetic analyses, respectively.

Transmitter typeSeals were instrumented with Argos position only tags (n ¼ 25,SPOT4 and 5, Wildlife Computers, Seattle, WA. USA) or Argos/Fastloc GPS tags (n ¼ 2, Wildlife Computers) all powered bytwo AA batteries. The tags were capable of providing 70 000 trans-missions and programmed to transmit 250–500 uplinks every dayor every second day depending on the period of tagging, providinga battery longevity of 140–280 d.

Data collection and analysisData on movements and transmitter status were collected via theArgos Location Service Plus system (Harris et al., 1990).Locations are classified by the Argos system into one of six locationclasses according to the level of accuracy (progressively higher un-certainty from 3, 2, 1, 0, A to B). Studies have shown that there canbe significant error in all location classes (up to several kilo-metres), but that even the low accuracy locations provide usefuland valid information if they are appropriately filtered (e.g.Sveegaard et al., 2011). Thus, all location classes were used inthe present study after filtering by an SAS routine, Argos FilterV7.02 (Douglas 2006). The filter is comparable with the R-basedSDA filter (speed, distance, and angle) tested by Freitas et al.(2008). The filter settings for this study were: maximum swimspeed, 10 km h21 (minrate ¼ 10), i.e. excluding swim speedbetween two locations .10 km h21. If, however, the distancebetween locations was , 5 km (maxredun ¼ 5), they were bothretained, because the swim speed calculations may be unrealisticdue to even smaller inaccuracies of closely spaced positions.Finally, positions were excluded if the angle between consecutivevector lines between the previous and following location was, 108 (ratecoef ¼ 10). All other settings were set as default. Themaps were generated using ArcMap (version 9.3). The bathymetri-cal depth contours are based on 18 resolution GEBCO data(version 1.00). Hawth’s Analysis Tools V3.27 was used as an exten-sion to ArcMap to generate track lines, kernel home range, andarea calculations. To avoid autocorrelation, only one locationwas subsampled from each of the days selected for the dutycycle, for the kernel home range analysis, and for the linearmixed effect model used on the distance from the tagginghaul-out site (see details below). This reduced the 8826 positionsto 2026 positions, that were distributed among either two sexes,two age groups, or four seasons, meaning that the home ranges

on average were based on a number . 500 observations, whichis ten times more that recommended by Seamann et al. (1999).The smoothing factor (bandwidth) was set to 20 000 and outputcell size to 1 km2. This was based on thorough inspection ofkernel contours during tests of alternate bandwidth as recom-mended by Beyer (2004), the creator of Hawth’s Analysis Tools.Detailed information on the filtering is provided in Sveegaardet al. (2011). A linear mixed effect model using maximum likeli-hood estimation was applied, with distance from the tagginghaul-out site as dependent variable, individual as a random orgrouping factor, and the fixed factors age group, sex, andmonth. Distance data were log-transformed prior to analyses toreduce skewness and to approximate normal distribution asrecommended in, for example, Zar (1996). The model was succes-sively reduced for non-significant factors at a 5% significance levelevaluated by the likelihood ratio test. The statistical analyses wereperformed with the software R (R Development Core Team, 2008).Excel 97 (SR2) and StatView V5.0.1 were used for basic statisticaldata presentations and graph presentations.

Data handlingThe seasonal categories and the exact date for these were definedby shifts in movement patterns (summer, 6 June–15 September;autumn, 16 September–13 December; winter, 14 December–21February; and spring, 22 February–5 June) taken from Figure 4.These categories were used in the subsequent statistical examin-ation. Distance data from the haul-out site were log(e)-transformed as the distribution was highly right skewed. The dis-tance data were analysed by a linear mixed effect model, with agegroup, season, and sex as fixed factors and seal individual asrandom grouping factor. The interaction factors between seasonand sex and between age group and sex were also included.However, excluding the interaction between age group and sexdid not result in a significantly worse model (log-likelihood, p ¼0.051), though very close. The data did not allow for testing ofthe interaction factor between season and age group due to lackof adult seal data during autumn and winter.

ResultsBasic information about the captured sealsOf the 27 harbour seals tagged in the present study the majoritywere yearlings (17) and subadults (8), whereas only two wereadults (Table 1). Age was based on weight and length informationfrom the literature (e.g. Pitcher and Calkins, 1979; Harkonen andHeide-Jørgensen, 1990). A similar age and sex distribution wastagged in each of the three years. The 27 seals were tracked for atotal of 3527 days (mean/seal ¼ 131 days; range ¼ 42–268days), providing 55 300 km of movements (mean ¼ 2048 km;range ¼ 584–4168 km) based on 8826 positions (after filtering).

Contact with other haul-out sites and regionsHarbour seals caught at Anholt move all over Kattegat with amaximum distance of 249 km from the tagging site at Anholtand average weekly distances of up to �85 km for yearlings,35 km for subadults, and 20 km for adults from the tagginghaul-out site (Figure 4). The seals tagged during autumn(14 September–9 June) showed a wider dispersal than the sealstagged during spring (25 April–9 September, Figures 4 and 5).Hence the seals had larger home ranges during the winter andvisited many sites in Kattegat (Figure 5; Tables 1 and 3). In

Movements and site fidelity of harbour seals in Kattegat, Denmark Page 3 of 10

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

Table 1. Basic biological and tagging information from the 27 harbour seals tagged in September 2005 and 2006 and April 2008 at Anholt.

Number PTT no.

Tracking

SexAgeclass

Denmark Sweden

Longevity(d)

Distance(km)

Length(cm) Skagen Hirsholmene Læsø Anholt Hesselø Gilleleje

SjællandsRev Tislarna Varberg

HallandsVadero

Tagged September 20051 200560265 268 2391 F Yearling 104 x x x2 200560266 69 878 M Yearling 107 x3 200560267 143 973 F Yearling 93 x4 200560268 251 1547 M Yearling 96 x x x x x x x5 200560269 78 1471 F Yearling 93 x x x x x x6 200560270 236 2278 M Yearling 102 x x7 200560271 42 1018 F Yearling 95 x x x8 200560272 242 2746 F Yearling 97 x x x9 200560273 266 2613 F Subadult 114 x10 200560274 152 2395 F Yearling 104 x x x xTagged September 200611 200608377 173 1509 F Subadult 116 x x12 200608378 137 1760 F Yearling 95 x x13 200608379 124 1735 F Yearling 96 x x14 200608380 116 709 M Subadult 116 x x15 200608381 161 1431 F Subadult 114 x x16 200637279 73 584 M Yearling 102 x x17 200637285 128 3434 F Yearling 95 x x x x x x18 200637287 75 973 F Yearling 99 x19 200637289 206 3536 M Subadult 108 xTagged April 200820 200867999 83 2649 M Yearling ? x x21 200837281 61 2882 M Subadult 102 x22 200808379 60 2533 F Yearling 102 x23 200837279 46 2481 M Subadult 118 x24 200868000 129 4168 F Yearling 101 x25 200868001 54 1523 M Subadult 118 x26 200808377 58 1025 F Adult 120 x27 200867998 96 4058 M Adult 140 xTagged September 2005 and 2006 (%) All 5 5 37 100 42 21 11 5 26 16

Yearling 7 7 43 100 43 29 14 7 36 21Subadult 20 100 40

Tagged April 2008 (%) All 13 100

Information on visits to neighbouring haul-out sites is also shown.

Page4

of10

R.

Dietz

etal.

by guest on June 5, 2016 http://icesjms.oxfordjournals.org/ Downloaded from

addition to Anholt, the preferred Danish areas (of which the ma-jority are well-known haul-out sites) for juveniles in declining im-portance were: Hesselø (43% of juveniles), Læsø (43%), Gilleleje

(29%), Sjællands Rev (14%), Hirsholmene (7%), and Skagen(7%; Table 1). Along the Swedish coast, yearlings visited threehaul-out sites: Varberg (36%), Hallands Vadero (21%), and

Figure 2. Map showing the positions of 27 harbour seals tagged in September 2005 and 2006, and in April 2008 at Anholt divided by sex.Males, green dots and green outline showing 90% kernel home range (3293 km2); and females, red dots and red outline (5189 km2).

Figure 3. Map showing the age-related distribution of the 27 tagged harbour seals. Dots indicate satellite locations and circles the 90% kernelhome ranges for the adults (1713 km2), subadults (2534 km2), and yearlings (6414 km2).

Movements and site fidelity of harbour seals in Kattegat, Denmark Page 5 of 10

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

Tislarna (7%; Table 1, Figure 1). There was a strong and significantcorrelation between the percentage of yearling (2005 and 2006)that visited a locality and the locality’s geographical distancefrom the tagging site at Anholt (r2 ¼ 0.595, F ¼ 10.27, p ¼

0.015; Figure 6). Subadults (n ¼ 5) showed a more restricted dis-persal, in which only two visited Hesselø and one visited Læsø(Table 1). No information was obtained from adult seals fromthe autumn tagging. The seals tagged during spring were all very

Figure 4. Age-specific seasonal site fidelity for harbour seals tagged at Anholt. Week number, month, and season are shown in relation to theweekly average distance from the tagging site at Anholt. The approximate distances to the closest haul-out sites are included on the secondy-axis.

Figure 5. Map showing the seasonal distribution of the 27 harbour seals tagged at Anholt separated into four seasons with locations (dots)and 90% kernel home ranges (outlines; summer, 1722 km2; autumn, 6885 km2; winter, 10 608 km2; spring, 5730 km2).

Page 6 of 10 R. Dietz et al.

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

stationary and only one individual out of the eight tagged visitedLæsø (see details below).

Distance from the tagging site relative to age, season, andsexAll class variables: age group, season, and sex, and the interactionbetween season and sex had a significant effect on distance movedfrom the tagging site, and imply that the sexes move differentlywithin seasons (Table 2). Yearlings showed the least site fidelityduring winter and spring (range means: 43.9 and 48.1 km) andfemales travelled further than males during autumn (30.4 vs.9.5 km), winter (47.1 vs. 27.6 km), and spring (49.1 vs. 45.4 km;Table 3). All sex and age groups showed strong site fidelityduring the summer breeding and moulting periods (average:9.7 km). Hence, during summer, all seals stayed at Anholt andthe surrounding waters, except for one subadult female(#200867999) which visited Læsø in mid July 2008.

Sex differences in the kernel home range distributionOverall, females travelled over a wider area compared with males(males, 3293 km2; females, 5189 km2; Figure 2). This movementpattern was mainly driven by yearlings in autumn and winterand the subadults in winter and spring (Table 3).

Age-related differences in distributionThe strength of site fidelity increases with age (especially fromyearlings to subadults), as seen on the age-separated plots onboth a geographic- and a distance-related scale (Figures 3and 4). Summarized over sex and season, the 90% kernel homeranges calculated for the yearlings (6414 km2) is larger than forsubadults (2534 km2), which again is larger than for adult seals(1713 km2). Likewise, yearlings move further away from theAnholt tagging haul-out site (25.5 km) than subadults(12.0 km), which again exceeds adult movements (8.9 km).However, the pattern of adults should be interpreted withcaution since only two seals were tracked and only in spring andsummer (see section on Data handling).

Overall seasonal distributionAs evident from the seasonal plots, all tagged harbour seals showthe strongest site fidelity to the tagging haul-out site on Anholtduring summer (week 19 until weeks 37–38), with a 90% kernelhome range of 1722 km2 and an average travelling distance of9.7 km from the tagging haul-out site (Figures 4 and 5; Table 3).Hereafter the seals move further away (average distance,22.6 km; 90% kernel home range, 6885 km2) during autumnuntil week 49, after which they range even further away from thetagging haul-out site during winter (average distance, 36.9 km;90% kernel home range, 10,608 km2; Figure 5; Table 3). Finally,during spring from week 7 to week 19, the seals gradually approachthe Anholt tagging haul-out site on average from 55 km to 10 km(average distance, 31.2 km; 90% kernel home range, 5730 km2).All visits to the Swedish coast, Skagerrak, and the northern coastof Sjælland took place during autumn and winter (Figure 5).

DiscussionSex-, age-, and seasonal-related differences in distributionOur findings that harbour seal sex and age groups move differentlywithin seasons have also been observed for other populations ofharbour seals. Lowry et al. (2001) found that the dispersal ofharbour seals tagged in Prince Williams Sound, Alaska, was signifi-cantly affected by sex and season. Mean monthly home rangesvaried from ,100 km2 to .1500 km2, and were smallest during

Table 2. Results of the linear mixed effect model conducted onthe distance (km) from the tagging site of a harmonized subsampleof 2026 positions of the total 8826 positions obtained from the 27harbour seals.

Source d.f. F-value p > F

Intercept 1 136.5 ,0.0001Age group 2 27.4 ,0.0001Season 3 46.2 ,0.0001Sex 1 1.9 0.1713Season × sex 4 41.7 ,0.0001Age group × sex 2 5.0 0.0067

Figure 6. Correlation between each haul-out site’s geographicaldistance from Anholt and the percentage of yearlings that visited thesite in 2005, 2006, and 2008. The trendline had an r2 ¼ 0.595 and wasdescribed by the formula y ¼ –0.496x + 62.64.

Table 3. Seasonal-, sex-, and age-related average distributiondistance from the tagging site (km) for the 27 harbour seals taggedat Anholt.

Season

Sexgroup Age group Spring Summer Autumn Winter

Yearround

Males Yearling 45.4 13.8 9.5 27.6 19.0Subadult 6.6 8.5 15.8 12.5 9.8Adult 8.4 13.7 12.6All age groups 19.0 10.7 12.3 17.7 13.1

Females Yearling 49.1 9.7 30.4 47.1 27.1Subadult 29.9 5.3 10.0 39.2 16.9Adult 4.3 3.1 3.3All age groups 42.9 8.7 25.7 45.4 24.3

Bothsexes

Yearling 48.1 10.5 26.7 43.9 25.5

Subadult 11.9 8.4 12.1 24.1 12.0Adult 7.0 9.4 8.9All age groups 31.2 9.7 22.6 36.9 20.0

Movements and site fidelity of harbour seals in Kattegat, Denmark Page 7 of 10

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

June–July. Females had larger mean home ranges than malesduring most of autumn and winter, and in some months up tofour times larger (Lowry et al., 2001). Females also tended tomove relatively more than males during autumn and winter inCalifornia (cf. Lowry et al., 2001) and in Moray Firth, Scotland(Thompson, 1993).

Movement patterns during the summer months have pre-viously been recorded by Harkonen and Harding (2001), whofound that female harbour seals in Skagerrak generally show stron-ger site fidelity during summer than males. Van Parijs et al. (1997)found that male harbour seals in Moray Firth, Scotland, travelledwidely in June, and after this restricted their ranges in early Julywhen females began foraging in late lactation. In that study,mean 7 days home ranges decreased from 65–480 km2 in Juneto 4–70 km2 in July. Home ranges of females in Moray Firthdecreased in size �2 weeks earlier than males, with the onset ofpupping (Thompson et al., 1994). A similar decrease in dispersalwas seen at Anholt prior to the pupping season, although thedata did not contain enough information to assess sex-related dif-ferences specifically during summer. In another study from MorayFirth, Thompson et al. (1996) estimated that 21 tagged harbourseals moved only to haul-out sites within a range of 75 km. Thesummer foraging areas were estimated for 31 harbour seals byTollit et al. (1998), where the majority of the seals foragedwithin 30 km of their haul-out site. In another study from south-east Scotland from the Tay and Eden estuaries, the seasonalhaul-out pattern of harbour seals was monitored (Sharples et al.,2009). Here the calculated probability of hauling out was highestduring June and July, which reflects a more stationary behaviour,whereas the lowest probability of hauling out was observed duringNovember, December, and January, which were the months whenthe Anholt seals moved furthest around. Similar results werereported from ten harbour seals tagged in the Kenmare River insouthwest Ireland (Cronin and McConnell, 2008). In this study,the haul-out frequency declined from October to January, wherethe lowest frequency was observed. Hereafter the frequenciesincreased again until April, from where a relatively stablehaul-out pattern was observed until July, where the tags werelost due to moulting. Another study from western Scotland con-ducted on the Isles of Skye, Islay, and Jura on adult harbourseals revealed a quite stationary pattern (Cunningham et al.,2009). This study showed that harbour seals generally remainedwithin a 25 km radius of haul-out sites, and only occasionally trav-elled .100 km away from the haul-out sites. The localized move-ments and the lack of variation with sex, age, and season in thestudy by Cunningham et al. (2009) contrasts with the finding ofour study and other studies (Thompson et al., 1998; Lowryet al., 2001). This difference may be due to the fact thatCunningham et al. (2009) only tagged adult seals, but may alsosuggest that food availability in the vicinity of the westernScotland haul-out sites was adequate throughout the year.

As suggested by, for example, Lowry et al. (2001) and Smallet al. (2005), the age or sex-related differences in movementsand home range sizes during late spring and summer are mostprobably attributed to behaviour associated with reproductionand moulting. At present, however, an explanation for the gender-specific difference in winter movements is still unclear.

The successively stronger site fidelity to the haul-out site foundfor older seals is consistent with information from other regionsand species. Lowry et al. (2001) found that juvenile harbourseals from Prince Williams Sound moved more than adults and

had larger home ranges. Movements were significantly affectedby age and month interactions. In all months, mean distancesbetween successively used haul-outs in Prince Williams Soundwere ,10 km for adult and ,20 km for juvenile seals. Mean dis-tance from haul-out to at-sea locations was 5–10 km for adultsand generally 10–25 km for juveniles (Lowry et al., 2001), whichis comparable with our findings from Anholt, of 12 and 26 kmfor yearlings and subadults, respectively, and 8.9 km for adultseals. Similar patterns have been observed in Skagerrak, whereyearlings generally exhibited larger home ranges than adults(Harkonen and Harding, 2001). These age-related differences inmovements probably reflect that the inexperienced yearlings andsubadults are going through a learning process in which theydevote substantial amounts of time (and resources) towards ex-ploring their habitat for prey and haul-out sites.

Regional differences in dispersalBoth tagging and genetic studies have documented regional differ-ences in the movement range of harbour seals. The present studyrevealed that seals at Anholt had much larger average home ranges(range: 1722–10 608 km2) than seals from Rødsand in the westernBaltic, with average home ranges (95% fixed kernel) of 215 km2

(range: 114–316 km2), but much smaller than harbour seals inthe Wadden Sea, which frequently make long-range foragingtrips into the North Sea (Dietz et al., 2003; Tougaard et al.,2008). These trends have also been supported by genetic studieswhere harbour seals in both the western Baltic and the Skagerrakwere much more restricted in their dispersal as measured bylevels of genetic differentiation than harbour seals in theKattegat (Olsen et al., in review). Still, although not as strong asbetween the two other regions, Kattegat seals did exhibit somedegree of genetic differentiation among haul-out sites, which isconsistent with the strong site fidelity during the breedingperiod recorded in the present study. The factors accounting forthe site fidelity may relate to spatial differences in availability ofhaul-out, feeding sites, hunting history, and disturbances.Although the literature seldom provides dispersal information ina uniform and comparable way, examples from France, Scotland,and Alaska show that movement patterns are very variable in dif-ferent regions and appear to correlate with local and regional dis-tributions of food resources. Average home ranges (95% fixedkernel) of harbour seals were quite low from France and estimatedto be 124 km2 (range: 58–358 km2) from Baie duMont-Saint-Michel and 80 km2 (range: 55–152 km2) from Baiedes Veys (Vincent et al., 2010). Harbour seals from PrinceWilliams Sound, Alaska showed a larger dispersal as juvenileharbour seals on average were estimated to have home ranges(percentage kernel not provided) of 3167 km2 (range: 485–54976 km2) and 1749 km2 (range: 311–16 970 km2) for adult seals(Lowry et al., 2001). Further information on regional homerange differences with reference to sex, age, and season was dis-cussed in previous sections. One could speculate as to whethersome of the tagged seals actually belonged to another breedingpopulation. However, the tagging took place at the start ofSeptember and late April for the end and start of the period,where the seals show a strong fidelity to their breeding area. Ifany of the long-lasting seals tagged in the autumn had comefrom other breeding populations they would probably haveentered these during the consecutive breeding period during thespring, which was not the case. Neither did any of the seals

Page 8 of 10 R. Dietz et al.

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

tagged in late April migrate to other haul-out sites. Therefore, it isquite unlikely that these seals came from another population.

Dispersal relative to previous PDV seal epizooticsIn both PDV incidences in 1988 and 2002, the first outbreaks wereobserved on the island of Anholt which is the primary reason whywe tracked the seals at this island. The first mortality from PDVdisease was reported on 12 April 1988 at Anholt (Dietz et al.,1989; Harkonen et al., 2006). The disease spread rapidly to otherseal haul-out sites, north (Læsø) and south (Hesselø) of Anholt,then further to the southwestern Kattegat (Samsø) and subse-quently to the west coast of Sweden. By the end of May 1988,the entire Kattegat harbour seal population was affected. Asimilar pattern of dispersal was apparent in 2002 (Harkonenet al., 2006). Again it started on Anholt, with the first dead sealsobserved on 4 May, �3 weeks later than in 1988. The start ofboth epizootics (12 April and 4 May) took place at the end ofthe spring season where the harbour seals change from a wide-spread dispersal to a more local and stationary behaviour. Thismeans that a harbour seal could have been infected somewherein Kattegat, and then, after the incubation time (5–12 d),started the epidemic at Anholt, where the disease spread rapidlyin the large and closely aggregated seal colony. The somewhatslower spreading of the disease in 2002 may be explained by theprogressively more stationary behaviour of harbour seals atAnholt during the spring.

The disease, however, continued to spread to the Baltic, theWadden Sea, and the Limfjord during summer, where theharbour seals were documented to be very stationary with astrong site fidelity. The spread of PDV to the UK took placefrom August to November when harbour seals, and especiallythe juveniles, move further away from Anholt. However, ourresults could not document migrations to either the Baltic,Limfjorden, the Wadden Sea, or the UK. This could either indicatethat other harbour seal groups not included in our tagging coulddisperse further or that another species such as the more widelydispersing grey seal (Halichoerus grypus) was involved in thespread of the disease. This is possible as a small group of greyseals are regularly observed at Anholt; however, their origin andmovements are unknown (Harkonen et al., 2007). The suddenjumps of the PDV to the Dutch Wadden Sea during both out-breaks also point to the grey seals, as this is the area of theWadden Sea with the most grey seals (Dietz et al., 1989;Harkonen et al., 2006).

In order to understand the full picture of the spread of PDV,additional information from harbour seals from other regions ofKattegat as well as for the grey seals is still required.

Management implicationsAlthough the strong site fidelity around the breeding periodobserved at Anholt could be a local phenomenon, several reportson restricted movement patterns in other harbour seal populationssuggest that site fidelity is a general characteristic of the harbourseal. This finding has several management implications: first, itstrongly indicates that harbour seals in southern Scandinavia com-prise several distinct population units both ecologically and genet-ically, with specific requirements in terms of management andconservation. Second, although harbour seals exhibit strong site fi-delity during the breeding period, contact with other localitiesdoes occur at other times of year. We found that the percentageof seals that visited a locality other than Anholt correlated strongly

with the locality’s distance from Anholt, suggesting that geograph-ical distance rather than preference for certain haul-out sites is themain factor driving movement between haul-out sites in Kattegat.Similar patterns have been observed in a genetic study (Olsen et al.,in review) and during the PDV epidemics in 1988 and 2002, wherethe virus spread outward in concentric circles from the epicentreon Anholt to neighbouring harbour seal localities in theKattegat–Skagerrak–Baltic area (Harkonen et al., 2006). Third,the strong site fidelity observed during summer suggests thatseals are not likely to travel between the haul-out sites duringlate August when the yearly aerial survey is conducted inDenmark and Sweden (e.g. Olsen et al., 2010; Teilmann et al.,2010). Thus, the estimates of population abundance and trendsobtained from these surveys can be regarded as reliable with noor only little bias introduced by movements of seals between local-ities during the survey period. Finally, because the observed fidelityto certain sites and marine areas appears relatively stable acrossseasons and years, the kernel home ranges calculated in thepresent study are likely to reflect marine and coastal regions ofspecial importance to the Anholt harbour seals foraging andbreeding in Kattegat. As such, they provide a good starting pointfor designating marine protected areas.

AcknowledgementsThe study was financed by Aage V. Jensens Foundations, theDanish Nature Agency (Danish Ministry of Environment),Gorenje, and the Danish Outdoor Council. Our colleagues SusiM. C. Edren, Jesper Møller, Rikke Danøe, Jakob Tougaard,Nikolaj I. Bech, Signe Sveegaard (all from the Department ofBioscience, Aarhus University), Minna B. Sørensen, MajaF. Rasmussen (Kattegatcentret), Morten Abildstrøm (Forest andNature Agency), Lee Miller (Odense University), Henrike Seibel(FTZ), Bjarni Mikkelsen, and the late Eyðfinnur Stefansson(Museum of Natural History Torshavn) and AqqaluRosing-Asvid (Greenland Institute of Natural Resources) areacknowledged for their help during the tagging operations. Thestudy was carried out with permission from the Danish NatureAgency (SNS-3446-00054 and SN 2001-34461/SN-0005) and theAnimal Welfare Division (Ministry of Justice, 2005/561-976).

ReferencesAndersen, S. M., Teilmann, J., Dietz, R., Schmidt, N. M., and Miller,

L. A. 2011. Behavioural responses of harbour seals tohuman-induced disturbances. Aquatic Conservation: Marine andFreshwater Ecosystems, in press.

Andersen, S. M., Teilmann, J., Harders, P. B., Hansen, E. H., andHjøllund, D. 2007. Diet of harbour seals and great cormorants inLimfjord, Denmark: interspecific competition and interactionwith fishery. ICES Journal of Marine Science, 64: 1235–1245.

Bigg, M. A. 1981. Harbour seal—Phoca vitulina and Phoca largha. InHandbook of Marine Mammals, pp. 1–27. Ed. by S. H. Ridgway,and R. J. Harrison. Academic Press: London.

Beyer, H. L. 2004. Hawth’s analysis tools for ArcGIS. http://www.spatialecology.com/htools/.

Block, B. A., Jonsen, I. D., Jorgensen, S. J., Winship, A. J., Shaffer, S. A.,Bograd, S. J., Hazen, E. L., et al. 2011. Tracking apex marine preda-tor movements in a dynamic ocean. Nature, 475: 86–90.

Bonner, W. N. 1972. The grey seal and common seal in Europeanwaters. Oceanografi and Marine Biology: Annual Review, 10:461–507.

Cronin, M. A., and McConnell, B. J. 2008. SMS seal: a new techniqueto measure haul out behaviour in marine vertebrates. Journal ofExperimental Marine Biology and Ecology, 362: 43–48.

Movements and site fidelity of harbour seals in Kattegat, Denmark Page 9 of 10

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from

Cunningham, L., Baxter, J. M., Boyd, I. L., Duck, C. D., Lonergan, M.,Moss, S. E., and McConnell, B. 2009. Harbour seal movements andhaul-out patterns: implications for monitoring and management.Aquatic Conservation: Marine and Freshwater Ecosystems, 19:398–407.

Dietz, R., Heide-Jørgensen, M. P., and Harkonen, T. 1989. Mass deathsof harbour seals (Phoca vitulina) in Europe. Ambio, 18: 258–264.

Dietz, R., Teilmann, J., Henriksen, O. D., and Laidre, K. 2003.Movements of seals from Rødsand seal sanctuary monitored bysatellite telemetry. Relative importance of the Nysted OffshoreWind Farm area to the seals. National Environmental ResearchInstitute, NERI Technical Report. 429: 44 pp.

Douglas, D. 2006. The Douglas Argos-Filter Algorithm. Version 7.03.http://alaska.usgs.gov/science/biology/spatial/pdfs/argosfilterv703_manual.pdf. U.S. Geological Survey, Anchorage, AK.

Freitas, C., Lydersen, C., Fedak, M. A., and Kovacs, K. M. 2008. Asimple new algorithm to filter marine mammal Argos locations.Marine Mammal Science, 24: 315–325.

Harkonen, T. J. 1987. Influence of feeding on haul-out patterns andsizes of sub-populations in harbour seals. Netherlands Journal ofSea Research, 21: 331–340.

Harkonen, T., Backlin, B. M., Barrett, T., Bergman, A., Corteyn, M.,Dietz, R., Harding, K. C., et al. 2008. Mass mortality in harbourseals and harbour porpoises caused by an unknown pathogen.Veterinary Record, 162: 555–556.

Harkonen, T., Brasseur, S., Teilmann, J., Vincent, C., Dietz, R., Abt, K.,Reijnders, P., et al. 2007. Status of grey seals along mainland Europefrom the Southwestern Baltic to France. NAMMCO ScientificPublications, 6: 57–68.

Harkonen, T., Dietz, R., Reijnders, P., Teilmann, J., Harding, K., Hall,A., Brasseur, S., et al. 2006. A review of the 1988 and 2002 phocinedistemper virus epidemics in European harbour seals. Diseases ofAquatic Organisms, 68: 115–130.

Harkonen, T., and Harding, K. C. 2001. Spatial structure of harbourseal populations and the implications thereof. Canadian Journalof Zoology, 79: 2115–2127.

Harkonen, T. J., and Heide-Jørgensen, M. P. 1990. Comparative lifehistories of East Atlantic and other harbor seal populations.Ophelia, 32: 211–236.

Harris, R. B., Fancy, S. G., Douglas, D. C., Garner, G. W., Amstrup,S. C., McCabe, T. R., and Pank, L. F. 1990. Tracking Wildlife bySatellite: Current Systems and Performance. United StatesDepartment of the Interior, Fish and Wildlife Service,Washington, DC 30: 52 pp.

Jepsen, P. U. 2005. Management Plan for Common Seal (Phocavitulina) and Grey Seal (Halichoerus grypus) in Denmark.http://www.naturstyrelsen.dk/Udgivelser/Aarstal/2005/Forvaltningsplansael.htm (In Danish). Ministry of Environment,Danish Nature Agency, Copenhagen, Denmark: 37 pp.

Lowry, L. F., Frost, K. J., Ver Hoef, J. M., and Delong, R. A. 2001.Movements of satellite-tagged subadult and adult harbor seals inPrince William Sound, Alaska. Marine Mammal Science, 17:835–861.

Olsen, M. T., Andersen, S. M., Teilmann, J., Dietz, R., Edren, S. M. C.,Linnet, A., and Harkonen, T. 2010. Status of the harbour seal(Phoca vitulina) in southern Scandinavia. NAMMCO ScientificPublications, 8: 77–94.

Olsen, M. T., Andersen, L. W., Dietz, R., Teilmann, J., Harkonen, T.,and Siegismund, H. in review. Remarkable fine-scale populationstructuring in a wide-spread marine mammal – Integratinggenetic and demographic data for the identification of Phoca vitu-lina populations and management units. Molecular Ecology.

Olsen, M. T., Teilmann, J., Dietz, R., Linnet, A., and Harkonen, T. inpreparation. Genetic variability of the harbour seal (Phoca vitulina)in southern Scandinavia.

Pitcher, K. W., and Calkins, D. G. 1979. Biology of the Harbor Seal,Phoca vitulina richardsi, in the Gulf of Alaska. US Department ofCommerce, NOAA, 19: 310 pp.

R Development Core Team. 2008. R: A language and Environment forStatistical Computing. R Foundation for Statistical Computing,Vienna.

Sharples, R. J., Mackenzie, M. L., and Hammond, P. S. 2009.Estimating seasonal abundance of a central place forager usingcounts and telemetry data. Marine Ecology-Progress Series, 378:289–298.

Small, R. J., Lowry, L. F., Hoef, J. M. V., Frost, K. J., Delong, R. A., andRehberg, M. J. 2005. Differential movements by harbor seal pups incontrasting Alaska environments. Marine Mammal Science, 21:671–694.

Seaman, D. E., Millspaugh, J. J., Kernohan, B. J., Brundige, G. C.,Raedeke, K. J., and Gitzen, R. A. 1999. Effects of sample size onkernel home range estimates. Journal of Wildlife Management,63: 739–747.

Sveegaard, S., Teilmann, J., Tougaard, J., Dietz, R., Mouritsen, K. N.,Desportes, G., and Siebert, U. 2011. High-density areas forharbor porpoises (Phocoena phocoena) identified by satellite track-ing. Marine Mammal Science, 27: 230–246.

Teilmann, J., Riget, F., and Harkonen, T. 2010. Optimizing surveydesign for Scandinavian harbour seals: population trend as an eco-logical quality element. ICES Journal of Marine Science, 67:952–958.

Thompson, P. M. 1993. Harbour seal movement patterns. Symposia ofthe Zoological Society of London, 66: 225–239.

Thompson, P. M., Mackay, A., Tollit, D. J., Enderby, S., andHammond, P. S. 1998. The influence of body size and sex on thecharacteristics of harbour seal foraging trips. Canadian Journal ofZoology, 76: 1044–1053.

Thompson, P. M., McConnell, B. J., Tollit, D. J., Mackay, A., Hunter,C., and Racey, P. A. 1996. Comparative distribution, movementsand diet of harbour and grey seals from the Moray Firth, N.E.Scotland. Journal of Applied Ecology, 33: 1572–1584.

Thompson, P. M., Miller, D., Cooper, R., and Hammond, P. S. 1994.Changes in the distribution and activity of female harbour sealsduring the breeding season; implications for their lactation strategyand mating patterns. Journal of Animal Ecology, 63: 24–30.

Tollit, D. J., Black, A. D., Thompson, P. M., Mackay, A., Corpe, H. M.,Wilson, B., van Parijs, S. M., et al. 1998. Variations in harbour sealPhoca vitulina diet and dive-depths in relation to foraging habitat.Journal of Zoology (London), 244: 209–222.

Tougaard, J., Teilmann, J., and Tougaard, S. 2008. Harbour seal spatialdistribution estimated from Argos satellite telemetry: overcomingpositioning errors. Endangered Species Research, 4: 113–122.

van Parijs, S. M., Thompson, P. M., Tollit, D. J., and Mackay, A. 1997.Distribution and activity of male harbour seals during the matingseason. Animal Behaviour, 54: 35–43.

Vincent, C., McConnell, B. J., Elder, J.-F., Gautier, G., and Ridoux, V.2010. Winter habitat use of harbour seals (Phoca vitulina) fittedwith FastlocTM GPS/GSM tags in two tidal bays in France.NAMMCO Sci. Publ, 8: 285–302.

Zar, J. H. 1996. Biostatistical Analysis. 3rd edn. Prentice Hall, NJ, USA.662p.

Handling editor: David Demer

Page 10 of 10 R. Dietz et al.

by guest on June 5, 2016http://icesjm

s.oxfordjournals.org/D

ownloaded from