Mar Biol DOI 10.1007/s00227-010-1563-9 ORIGINAL PAPER Inter-colony movements, at-sea behaviour and foraging in an immature seabird: results from GPS-PPT tracking, radio-tracking and stable isotope analysis Stephen C. Votier · W. James Grecian · Samantha Patrick · Jason Newton Received: 15 June 2010 / Accepted: 6 October 2010  Springer-Verlag 2010 Abstract Seabird populations contain large numbers of immatures––in some instances comprising >50% of the fully grown adults in the population. These birds are signiWcant components of marine food webs and may contribute to compensatory recruitment and dispersal, but remain severely understudied. Here, we use GPS-PTTs, radio-tracking and analysis of stable carbon (13C) and nitrogen (15N) isotopes to investigate the movements and foraging ecology of immature seabirds. Our study focussed on immature northern gannets Morus bassanus aged 2–4 attending non-breeding aggregations alongside a large breeding colony. GPS-PTT tracking of Wve birds revealed that immatures have the ability to disperse widely during the breeding season, with some individuals potentially prospecting at other colonies. Overall, however, immatures were faithful to the colony of capture. During returns to the focal colony, immatures acted as central place foragers, conducted looping and commuting Xights, and analysis of the variance in Wrst-passage time revealed evidence of area-restricted search (ARS) behaviour. In addition, stable carbon (13C) and nitrogen (15N) isotope analyses indicate that immatures were isotopically segregated from breeders. Our Wndings provide insights into the foraging, prospecting and dispersal behaviour of immature seabirds, which may have important implications for understanding seabird ecology and conservation. Communicated by M. E. Hauber. S. C. Votier (&) · W. J. Grecian · S. Patrick University of Plymouth, Plymouth PL4 8AA, UK e-mail: stephen.votier@plymouth.ac.uk J. Newton Natural Environment Research Council (NERC) Life Sciences Mass Spectrometry Facility, Scottish Universities Environmental Research Centre, Ranking Avenue, East Kilbride G75 0QF, UK Introduction Populations of long-lived iteroparous vertebrates often contain large numbers of immature animals that are either physiologically unable to reproduce or do not breed for other reasons (Weimerskirch 1992). This age class represents a signiWcant component of many populations, but because research tends to focus on breeding individuals, there is a genuine paucity of information on the ecology of immatures. Seabirds tend to live for a long time and have prolonged periods of immaturity (Weimerskirch 2002). As a consequence, their populations often contain large numbers of immature birds, which are important for a number of reasons. Firstly, because they are abundant, in some instances comprising >50% of the fully grown adult population (Klomp and Furness 1992), they represent major consumers in marine ecosystems (Brooke 2004) and may compete with breeding birds as well as other marine predators. Secondly, as future recruits, this age class is important for long-term population persistence, and compensatory recruitment by immatures may be important for oVsetting stochastic mortality in other parts of the population (Votier et al. 2008b). Thirdly, because immature seabirds tend to show higher rates of dispersal compared with breeders (Huyvaert and Anderson 2004), they may play a key role in population processes. Dispersal is an important factor leading to gene Xow and, moreover, because seabirds tend to breed at high densities in a small number of widely spaced locations, dispersal among colonies may be important to ensure population persistence in the face of global change (Kokko and Lopez-Sepulcre 2006). Yet despite the implications that immatures are signiWcant components of marine food webs and have relevance for buVering the eVects of global change, they remain severely understudied. 123 Mar Biol Although immature seabirds spend long periods of their Wrst few years at-sea, among many species, non-breeders aggregate alongside the main nesting areas in ‘club-sites’. Attendance at these sites appears to be important in the recruitment process (Hatchwell and Birkhead 1991; Klomp and Furness 1992) and, because they come to land, provides a research opportunity to study this age class. Previous work on immature seabirds has used individually identiWable Weld marks (Halley and Harris 1993; Halley et al. 1995) and transponders (Dittmann et al. 2005) attached to known-age birds to study club-site attendance both within and among colonies. While these approaches have revealed much about the dispersal and recruitment patterns of immature seabirds, they may be biased by resighting heterogeneity or limited to relatively small spatial scales. Molecular markers are also invaluable for studying dispersal (McCoy et al. 2005) but this approach will fail to capture movements that do not lead to gene Xow (i.e. prospecting). Moreover, neither mark–recapture nor studies of population genetics reveal anything about the at-sea movements or foraging behaviour of immatures. Advances in bio-logging technology have been instrumental in unravelling the foraging strategies and at-sea behaviour of seabirds (Ropert-Coudert and Wilson 2005), but, although Xedgling seabirds have been tracked (Weimerskirch et al. 2006; Ismar et al. 2010), thus far there have been no tracking studies of immature seabirds attending non-breeding clubsites, to our knowledge. A useful complement to tracking technology is the analysis of stable isotopes of carbon (13C and 12C) and nitrogen (15N and 14N) to characterise diet, trophic relationships and foraging tactics (Inger and Bearhop 2008). Gradients in these isotope ratios (expressed as 13C and 15N) are reXected in the tissues of consumers in a predictable fashion and provide a robust method for characterising the isotopic niche of individual animals (Bearhop et al. 2004). In the context of the present study, this technique oVers the opportunity to investigate whether breeding seabirds are isotopically segregated from immatures, assuming comparable physiologies. The goal of the current paper is to better understand the at-sea behaviour, foraging and colony attendance of immature northern gannets Morus bassanus (hereafter gannet), using a combination of GPS-PTT satellite transmitters and remote radio-receiver logging, as well as the analysis of stable isotope ratios in blood. Gannets are medium-ranging seabirds that nest in a small number of densely populated colonies (Mitchell et al. 2004) and like many seabirds show delayed sexual maturity: median age of Wrst breeding is Wve (Nelson 2002). Yet unlike most seabirds, they can be aged on the basis of plumage morphology up to age 4 (Nelson 2002). We take advantage of this pattern and study birds aged 2–4 years, caught while attending non-breeding club-sites alongside a large breed- 123 ing population. The aims of our study were to (1) study dispersal ability and prospecting behaviour of immature gannets using GPS-PTTs; (2) monitor attendance at the colony of capture using GPS-PTTs and radio-tracking; (3) characterise at-sea foraging behaviour using GPS-PTTs and; (4) examine whether breeders and immatures show foraging niche segregation by analysing 13C and 15N values in blood. Methods Fieldwork was conducted at the third largest northern gannet colony in the World, Grassholm, Wales, UK (51°43⬘N, 05°28⬘W), where several thousand pre-breeding immature gannets gather alongside »40,000 pairs of breeders. During June–July 2009, 31 immature gannets were caught using a brass noose on the end of a pole. While attendance of clubsites strongly suggests immaturity, some failed breeders also use these areas (Klomp and Furness 1992). Therefore, we only captured birds aged 2–4 year based on plumage (Nelson 2002). First-year birds do not return to the colony (Nelson 2002). To enable a comparison between the isotopic niche of immatures with breeders, we also caught 27 chick-guarding birds over the same period. GPS-PTTs and radio-tracking To determine Wne-scale movements of immatures away from the colony, we attached 40-g battery-powered LC4 GPS-Platform Terminal Transmitters (GPS-PTTs) (Microwave Telemetry, Inc.) on the base of the tail using cable ties and Tesa® tape to Wve birds (1£ two-year-old and 4£ three-year-old). Satellite transmitters were programmed to obtain an hourly GPS Wx, which was relayed through the Argos system every 48 h. Individual gannet movements were reconstructed based primarily upon GPS data, but we also incorporated Argos location estimates (Wx strength LC A, 0, 1, 2, 3) (Coyne and Godley 2005). Tracks for immatures were compared with GPS tracks obtained from breeding gannets tracked at Grassholm during 2006 (Votier et al. 2010). We examined for prospecting behaviour, deWned simply as a GPS Wx at another gannet colony. We used Wrst-passage time (FPT) analysis (Fauchald and Tveraa 2003) to examine at-sea search behaviours. FPT is the time it takes an individual to travel across a circle of given radius (Fauchald and Tveraa 2003). Log-transformed variance in FPT was plotted against the circle radius r, and according to Fauchald and Tveraa (2003), a peak in variance is representative of area-restricted search (ARS) behaviour. ARS indicates an increase in turning rate or decrease in speed and can be used to restrict searching Mar Biol behaviour to areas of high foraging success (Weimerskirch et al. 2007). ARS is also associated with an increase in diving behaviour by gannets (Hamer et al. 2009). Although this behaviour has been described for a number of seabird species (Weimerskirch et al. 2007; Hamer et al. 2009), it is currently unknown whether immature seabirds adopt a similar searching strategy. Since the accuracy of tracking data may restrict the ability to detect ARS behaviours (Pinaud 2008), this analysis was conducted only using GPS Wxes. Moreover, gannets rest on the water at night (Votier et al. 2010), which could lead to spurious identiWcation of ARS behaviours, so only Wxes obtained during daylight hours were included. To monitor regular attendance patterns at Grassholm, we attached 10-g TW-3 tail-mounted radio tags (Biotrack, Wareham, UK) to a further 20 immatures (4£ two-yearold, 13£ three-year-old and 3£ four-year-old) and recorded the presence/absence using a DataSika datalogging receiver attached to a 173-MHz omnidirectional antenna with an above-ground range of 3–6 km, during 26th June–29th July (late incubation and chick rearing for breeding birds), and again from 14 to 24th August (chick rearing and chick Xedging for breeding birds). The aerial was located to be in line of sight with all of the main areas of the gannet colony. We were interested in assessing the degree to which immatures returned repeatedly to club-sites on Grassholm, as has been shown for some other seabirds (Halley et al. 1995). Stable isotope analysis We sampled approximately 0.2 ml of blood from the tarsal vein using 25-gauge needles under licence from the UK Home OYce. Blood samples were separated into plasma and red blood cells (RBC), using a centrifuge, within 2–3 h of sampling and then stored frozen until preparation for analysis. Prior to stable isotope analysis, RBC were freezedried, homogenised and »0.7 mg was weighed into a tin cup. Analyses were conducted at the East Kilbride Node of the Natural Environment Research Council Life Sciences Mass Spectrometry Facility via continuous Xow isotope ratio mass spectrometry using a Costech (Milan, Italy) ECS 4010 elemental analyser interfaced with a Thermo Electron (Bremen, Germany) Delta XP mass spectrometer. Isotope ratios are reported as -values and expressed as ‰ according to the equation X = [(Rsample/Rstandard) ¡ 1] £ 1000, where X is 13C or 15N and R is the corresponding ratio 13C/ 12 C or 15N/14N and Rstandard is the ratio of the international references PDB for carbon and AIR for nitrogen. The standard deviation of multiple analyses of an internal gelatin standard in each experiment was better than 0.2‰ for 15N and 0.1‰ for 13C. 15N shows a stepwise enrichment of 3–5‰ at each trophic level. 13C values also increase with each trophic level (»1‰) but are also a function of a number of spatial gradients (such as latitude, inshore vs. oVshore and demersal vs. pelagic). Here, we use stable isotopes to determine the degree to which breeders and immatures are segregated along these isotopic gradients. In doing so, we assume that any potential diVerences in assimilation eYciency or physiology between these age classes does not inXuence the relationship between isotope values in prey and blood. Results GPS-PTTs and radio-tracking The Wve individuals with GPS-PTTs were tracked for an average of 17 days (range 4–25). This produced an average of 12.19 (range 7–21) Wxes per day and birds travelled an average of 200.5 km per day (range 157.7–244.6 km) (Table 1). In general, immatures were far more widely dispersed compared with 25 adult breeding gannets tracked from the same location during June–July 2006 (Fig. 1): average trip duration of four immatures returning to Grassholm was 114.4 h (range 22.0–270.0) and distance travelled 1065.5 km (range 100.6–2828.9); average trip duration of breeders was 25.1 h (range 3.7–73.9) and distance travelled 370.5 km (range 70.5–1121.1) (Votier et al. 2010). Two of the satellite-tracked birds visited other gannet colonies in the United Kingdom, Ireland and France (Fig. 1 a, b), where 1–2 GPS Wxes were received from within another gannetry. In addition, one other bird (Fig. 1 d) was recorded within 7 km of Great Saltee, Ireland, suggesting prospecting. Each bird only visited another colony once. All but one of the satellite-tracked birds returned to Grassholm (Fig. 1a)––one bird returned on 5 dates, one on 3 dates and 2 birds returned just once (Fig. 1 a–e), with departure from the colony varying considerably (Table 1). However, the number of returns to the colony is most likely an underestimate since this will be positively correlated with the tracking period and one bird (Fig. 1a) was only tracked for 4 days. Immature gannets showed evidence of central place foraging with variable search behaviours. Peaks in variance of FPT from daylight Wxes indicated ARS search behaviours for three birds (2, 3 and 4) (Fig. 2), although ARS was not found on all trips (Table 1). Large-scale ARS ranged from 20 to 110 km and nested Wne-scale ARS ranged from 20 to 60 km. Birds 1 and 5 showed linear declines in log FPT as a function of spatial scale (Fig. 2) with no obvious peaks and therefore provided no evidence for ARS. During foraging trips, immatures showed a range of Xight behaviours including commuting (Fig. 1 c, d) and looping (Fig. 1 b, d). 123 Mar Biol Table 1 Descriptive statistics for GPS and Argos Wxes from individual immature gannets tracked from Grassholm, Wales, UK Bird number 1 2 3 4 5 Mean SD Age 3 3 3 3 2 – – GPS total Wxes 21 315 115 128 286 173 123.9 Argos total Wxes 8 65 80 34 50 47.4 27.9 LC 3 0 6 3 0 4 2.6 2.6 LC 2 0 9 6 4 6 5 3.3 LC 1 1 20 5 9 10 9 7.1 LC 0 4 17 12 14 17 12.8 5.4 LC A 3 13 14 7 13 10 4.8 Total no. of Wxes 29 380 195 162 336 220.4 141.0 Days operational 4 18 19 16 25 16.6 7.6 Mean GPS Wxes per day 4.8 17.2 8.0 7.9 11.44 9.9 4.7 Mean Argos Wxes per day 1.82 3.6 2.1 2.1 1.99 2.3 0.7 Mean of all Wxes per day 6.6 20.7 10.1 10.1 13.44 12.2 5.4 Max displacement 680.7 696.1 605.9 486.3 643.5 622.5 83.8 Total distance 1062.8 4485 3646.8 2728.3 3942.9 3173.2 1340.6 40.3 Distance travelled per day 241.1 244.6 189.2 169.8 157.7 200.5 Total number of returns to Grassholm 0 1 5 3 1 2 2 Mean duration of round trips away from Grassholm (hours) – 243 81 65.3 17 101.6 98.1 Trips with suYcient data to investigate ARS 1 2 3 2 1 1.8 0.8 ARS behaviour No Yes Yes Yes No – – – Trips with ARS 0 2 1 2 0 – Average large-scale ARS (km) – 20 110 80 – – – Average nested-scale ARS (km) – No 20 50 – – – Nineteen out of 20 immature gannets with radio-transmitters were recorded repeatedly at Grassholm during the 45 days following device deployment. The median number of days tagged birds were recorded at the colony showed an increase with age (Table 2), but this diVerence was not statistically signiWcant (Kruskal–Wallis: H2 = 1.903, P = 0.386). The number of consecutive days at Grassholm varied from one to nine (Table 2). Stable isotope analysis During the 2009 breeding season, breeders and immatures were isotopically segregated (MANOVA, Wilk’s Lambda, F2,55 = 24.992, P < 0.001), and in univariate analysis, both 13C (t-test, t56 = 4.407, P < 0.001) and 15N (t-test, t56 = 6.546, P < 0.001) were signiWcantly diVerent from one another (Fig. 3). Breeders showed less depleted 13C and higher 15N compared with immatures (Fig. 3). Discussion Here, we report for the Wrst time, to our knowledge, at-sea movements and prospecting behaviour of an immature sea- 123 bird, using GPS-PTTs. In addition, we provide details of colony attendance using radio-tracking and also compare the isotopic niche of immatures and breeders attending the same colony. The main Wndings are that immatures were highly vagile, travelling long distances at-sea and also visited diVerent gannet colonies indicating prospecting. Despite this, immature gannets showed Wdelity to the colony of capture and evidence of central place foraging. Stable isotopes strongly indicated that immatures and breeders are isotopically segregated. We found no apparent evidence of any device eVects–– upon release all birds were seen to Xy oV strongly, and remote sensing revealed that many returned to the colony. Previously, 65-g GPS loggers and 20-g radio-transmitters had no detectable eVect when deployed on gannets in the same way as the 40-g GPS satellite transmitters deployed here (Lewis et al. 2002; Garthe et al. 2007). However, deployment of satellite tags on Xedgling Australian gannets M. serrator indicates potentially negative impacts on survival (Ismar et al. 2010). Analysis of Argos data indicates that GPS-PTTs did not fail because of power loss, and therefore, devices were either lost during tail moult or because of death. We cannot distinguish between these two possibilities, but immature gannets undergo complex semi-continuous moult Mar Biol Fig. 1 GPS satellite-tracking of immature northern gannets caught at Grassholm, Wales. Star indicates the study colony, and circles are all other gannet colonies in the United Kingdom, Ireland and France. a–e At-sea movements and presumed prospecting behaviour of individual immature gannets caught at Grassholm 10th July and tracked until 3rd August 2009. DiVerent coloured tracks represent repeat trips after returning to Grassholm, and Wlled coloured circles represent visits to diVerent gannet colonies, and hatched Wlled circles represents birds within 10 km of another colony. f At-sea movements of 25 individual adult gannets (including 5 repeat tracks) breeding at Grassholm during June–July 2006 (from Votier et al. 2010). Note that compared with immatures, breeders tend to travel over a smaller area and do not visit other gannet colonies of the tail feathers at this time (Nelson 2002) so loss during moult is certainly a plausible explanation. Although our sample sizes are small, GPS-PTT satellitetracking showed that immature gannets caught on Grassholm in 2009 travelled extensively throughout the Irish Sea, Celtic Sea, NE Atlantic and the Bay of Biscay, moving much greater distances compared with breeders tracked from the same colony during 2006 (Fig. 1). Annual varia- tion in food availability may inXuence foraging patterns, and previous work suggests similar tightly constrained foraging trips for centrally placed breeding gannets across years (Hamer et al. 2000, 2001). Therefore, we suggest that the diVerences reported here are not simply artefacts of sampling in diVerent years. The extensive foraging ranges of immatures may be important for a number of reasons. First, it may reduce 123 Mar Biol Variance in first passage time 1.6 Bird 1 Bird 2 Bird 3 Bird 4 Bird 5 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0 25 50 75 100 125 150 175 200 225 250 275 300 325 350 375 400 425 450 475 500 0.0 Spatial scale (km) Fig. 2 Variance in Wrst-passage time, as a function of radius r, for Wve immature northern gannets. The Wgure provides examples of single tracks for each bird illustrating peaks in variance for three individuals (birds 2, 3 and 4), indicating area-restricted search (ARS) behaviour. Bird 4 also showed ARS at nested spatial scales. Birds 1 and 5 showed no evidence of ARS Table 2 Immature northern gannets show colony Wdelity during the breeding season Age Total days at colony Consecutive days at colony Individuals 2 8.5 (4–15) 1.5 (1–7) n=4 3 12.5 (5–20) 2 (1–7) n = 12 4 17 (7–30) 2 (1–9) n=3 All birds 11 (4–30) 2 (1–9) n = 19 Median (range) total number of days and consecutive number of days radio-tagged birds were recorded at Grassholm, Wales, following release. Data are grouped by age 2–4. Remote-sensing station operated during 26th June–29th July and 14–24th August 2009 (n = 45 days) 17 Breeders (n = 27) Immatures ( n = 31) δ15N (‰) 16 15 14 13 12 -19.0 -18.5 -18.0 -17.5 -17.0 δ13C (‰) Fig. 3 Mean (§SD) 13C and 15N values in red blood cells of immature and breeding northern gannets. Immatures are birds aged 2–4 years attending non-breeding club-sites and are isotopically segregated from chick-guarding breeders 123 intraspeciWc competition for food at the focal colony. Second, because immatures are thinly spread over a wide area, this may reduce their vulnerability to the eVects of stochastic mortality such as bycatch or pollution (e.g. oil spills, Votier et al. 2008b). During movements away from Grassholm, 2 of 5 satellite-tracked birds visited other gannet colonies (Fig. 1 a, b) or were close to other gannetries (Fig. 1d), indicating prospecting. This conWrms earlier evidence that immature seabirds move between colonies during the breeding season (Halley and Harris 1993). On Wrst examination, this extreme vagility might seem unsurprising given that seabirds of all ages travel long distances during seasonal migrations (Guilford et al. 2009). However, dispersal and migration are likely to be triggered by quite diVerent behavioural cues (Serrano and Tella 2003), such that movements during the inter-breeding period and movements during the breeding season will not necessarily correlate. Therefore, the dispersal potential described here could enable longdistance movement of seabirds to novel areas in the face of environmental change. Despite having the ability to move freely among colonies, immature gannets appear strongly tied to the colony of capture, which may limit dispersal propensity. Prospecting visits away from Grassholm were brief; two GPS Wxes were obtained from one bird visiting Great Saltee, County Wexford (Fig. 1b), while all other colony visits represented only a single GPS Wx. In contrast, 19 of 20 radio-tracked birds and 4 of 5 satellite-tracked birds returned to Grassholm (Table 2). Selecting an appropriate colony in which to breed is an important behavioural process and our Wndings indicate that while immature birds mainly attend the large colony at Grassholm, visits to other colonies enable individuals to gather social information from these sites (Boulinier et al. 2008). This behaviour can be interpreted as a type of bet-hedging––the attendance patterns suggest that these immatures will most likely recruit into the breeding population at Grassholm, yet may still recruit to other colonies they have visited in the region. Our Wndings provide an interesting insight into the process of colony selection in immature seabirds, but a major gap in our knowledge is the behaviour of birds in their Wrst year––these birds do not generally return to the breeding areas until their third calendar year. As well as persistent visits to the club-site on Grassholm, immatures showed evidence of central place foraging (Fig. 1)––something that has not previously been reported for non-breeding seabirds. Foraging in this way will mean more time and energy lost travelling between suitable patches compared with remaining at sea, which strongly suggests that attendance at club-sites is beneWcial. These beneWts might include information about the location and availability of food, increased chances of Wnding a mate or Mar Biol improved chances of obtaining a suitable nest site (Courchamp et al. 2008). Much focus has been placed upon protecting nesting sites on land (Mitchell et al. 2004) as well as foraging areas at-sea (Louzao et al. 2006) for eVective seabird conservation, yet the value of club-sites is generally overlooked. Non-breeding club-sites are typically adjacent to breeding colonies and by default may be aVorded protection, but should these aggregations occur away from the main breeding aggregations, they should be incorporated into management strategies. During foraging bouts away from the colony, immatures showed a range of search behaviours (Fig. 1, 2). In particular, 3 of 5 tracked birds showed evidence of ARS behaviour (Table 1, Fig. 2). This behaviour was not shown by all birds, or during all foraging trips, and the reason for this is unclear. While it is possible that ARS behaviour was not detected because of limitations in the tracking data (it was not possible to do this for all tracks because of sparseness), this seems unlikely given the GPS resolution and number of Wxes (Pinaud 2008). An alternative explanation is the use of this type of searching strategy only develops with foraging experience, which is related to age (Catry and Furness 1999). Two birds (Fig. 1 c, d) showed commuting Xights with ARS behaviours at the distal portion, suggesting that these pre-breeders have some knowledge of where suitable patches are (Weimerskirch 2007). Analysis of 13C and 15N in red blood cells represented dietary intake of the previous 2–5 weeks (Hobson 2005) and indicated signiWcant diVerences between breeders and immatures (Fig. 3). The reasons for these diVerences are not clear, but the signiWcantly higher 13C and 15N values of breeders are consistent with consumption of a higher proportion of Wshery discards (Votier et al. 2010) or consumption of larger Wsh (Votier et al. 2008a). Alternatively, these diVerences may arise because immatures and breeders forage in isotopically distinct regions as indicated by the tracking data (Fig. 1) or it may be a combination of both these factors. Nonetheless, the key Wnding here is that immatures were isotopically segregated from breeders (Fig. 3). This result suggests that even if immatures do forage in similar areas to breeders, which we cannot completely exclude due to our tracking data coming from two separate years, they consume isotopically diVerent prey there. Segregation between breeders and immatures may be an important consequence of intraspeciWc competition, as previously discussed, and enable such large aggregations of birds to utilise the same colonies. This work reveals that immature seabirds have the ability to disperse widely at-sea and move among diVerent colonies during the breeding season of their Wrst 2–3 years of life. Nevertheless, Wdelity to one large colony may potentially lessen dispersal potential and highlights the vulnerability of having large numbers of seabirds clumped at a small number of colonies. Persistent attendance at club-sites reveals their importance to immatures and their relevance to conservation. In addition, we found that immatures acted as central place foragers during the breeding season and showed variable foraging tactics––some birds searched at random while others showed commuting Xights followed by ARS behaviours. Moreover, we revealed isotopic segregation between breeders and immatures, which may be important for such large aggregations of birds to coexist. Studying the behaviour of immature seabirds, in particular during their Wrst year of life, is a key step in understanding seabird ecology and conservation. Acknowledgments We would like to thank Greg & Lisa Morgan, Tim Brooke at VentureJet, Anthony Bicknell, Valentina Lauria, Carrie Gunn, James Waggitt, Claudia Stauss and Simon Rundle for help in Weld, laboratory and for comment on the manuscript. The Royal Society for the Protection of Birds granted permission to work on Grassholm. Device attachment was conducted with permission of the Countryside Council for Wales. SCV was funded by a NERC New Investigators Grant (NE/G001014/1), WJG funded by a studentship from the Peninsula Research Institute for Marine Renewable Energy and SP funded by EU INTERREG Project CHARM III. References Bearhop S, Adams CE, Waldron S, Fuller RA, Macleod H (2004) Determining trophic niche width: a novel approach using stable isotope analysis. J Anim Ecol 73:1007–1012 Boulinier T, McCoy KD, Yoccoz NG, Gasparini J, Tveraa T (2008) Public information aVects breeding dispersal in a colonial bird: kittiwakes cue on neighbours. Biol Lett 4:538–540 Brooke MD (2004) Food consumption of the world’s seabirds. Proc R Soc B 271:S246–S248 Catry P, Furness RW (1999) The inXuence of adult age on territorial attendance by breeding great skuas Catharacta skua: an experimental study. J Avian Biol 30:399–406 Courchamp F, Ludek B, Gascoigne J (2008) Allee eVects in ecology and conservation. Oxford University Press, Oxford Coyne MS, Godley BJ (2005) Satellite Tracking and Analysis Tool (STAT): an integrated system for archiving, analyzing and mapping animal tracking data. Mar Ecol Prog Ser 301:1–7 Dittmann T, Zinsmeister D, Becker PH (2005) Dispersal decisions: common terns, Sterna hirundo, choose between colonies during prospecting. Anim Behav 70:13–20 Fauchald P, Tveraa T (2003) Using Wrst-passage time in the analysis of area-restricted search and habitat selection. Ecology 84:282–288 Garthe S, Montevecchi WA, Chapdelaine G, Rail JF, Hedd A (2007) Contrasting foraging tactics by northern gannets (Sula bassana) breeding in diVerent oceanographic domains with diVerent prey Welds. Mar Biol 151:687–694 Guilford T, Meade J, Willis J, Phillips RA, Boyle D, Roberts S, Collett M, Freeman R, Perrins CM (2009) Migration and stopover in a small pelagic seabird, the Manx shearwater PuYnus puYnus: insights from machine learning. Proc R Soc B 276:1215–1223 Halley DJ, Harris MP (1993) Intercolony movement and behaviour of immature guillemots Uria aalge. Ibis 135:264–270 Halley DJ, Harris MP, Wanless S (1995) Colony attendance patterns and recruitment in immature Common Murres (Uria aalge). Auk 112:947–957 123 Mar Biol Hamer KC, Phillips RA, Wanless S, Harris MP, Wood AG (2000) Foraging ranges, diets and feeding locations of gannets Morus bassanus in the North Sea: evidence from satellite telemetry. Mar Ecol Prog Ser 200:257–264 Hamer KC, Phillips RA, Hill JK, Wanless S, Wood AG (2001) Contrasting foraging strategies of gannets Morus bassanus at two North Atlantic colonies: foraging trip duration and foraging area Wdelity. Mar Ecol Prog Ser 224:283–290 Hamer KC, Humphreys EM, Magalhaes MC, Garthe S, Hennicke G, Peters G, Gremillet D, Wanless S (2009) Fine-scale foraging behaviour of a medium-ranging marine predator. J Anim Ecol 78:880–889 Hatchwell BJ, Birkhead TR (1991) Population-dynamics of common guillemots Uria-Aalge on Skomer Island, Wales. Ornis Scand 22:55–59 Hobson KA (2005) Using stable isotopes to trace long-distance dispersal in birds and other taxa. Divers Distrib 11:157–164 Huyvaert KP, Anderson DJ (2004) Limited dispersal by Nazca boobies Sula granti. J Avian Biol 35:46–53 Inger R, Bearhop S (2008) Applications of stable isotope analyses to avian ecology. Ibis 150:447–461 Ismar SMH, Hunter C, Lay K, Ward-Smith T, Wilson PR, Hauber ME (2010) A virgin Xight across the Tasman Sea? Satellite tracking of post-Xedging movement in the Australasian Gannet Morus serrator. J Ornithol 151:755–759 Klomp NI, Furness RW (1992) Non-breeders as a BuVer against Environmental-Stress––declines in numbers of great Skuas on Foula, Shetland, and prediction of future recruitment. J Appl Ecol 29:341–348 Kokko H, Lopez-Sepulcre A (2006) From individual dispersal to species ranges: perspectives for a changing world. Science 313:789–791 Lewis S, Benvenuti S, Dall’Antonia L, GriYths R, Money L, Sherratt TN, Wanless S, Hamer KC (2002) Sex-speciWc foraging behaviour in a monomorphic seabird. Proc R Soc B 269:1687–1693 Louzao M, Hyrenbach KD, Arcos JM, Abello P, De Sola LG, Oro D (2006) Oceanographic habitat of an endangered mediterranean procellariiform: implications for marine protected areas. Ecol Appl 16:1683–1695 McCoy KD, Boulinier T, Tirard C (2005) Comparative host-parasite population structures: disentangling prospecting and dispersal in 123 the black-legged kittiwake Rissa tridactyla. Mol Ecol 14:2825– 2838 Mitchell PI, Newton SF, RatcliVe N, Dun TE (2004) Seabird populations of Britain and Ireland. Christopher Helm, London Nelson B (2002) The atlantic gannet. Fenix Books Ltd, Norfolk Pinaud D (2008) Quantifying search eVort of moving animals at several spatial scales using Wrst-passage time analysis: eVect of the structure of environment and tracking systems. J Appl Ecol 45:91–99 Ropert-Coudert Y, Wilson RP (2005) Trends and perspectives in animal-attached remote sensing. Front Ecol Environ 3:437–444 Serrano D, Tella JL (2003) Dispersal within a spatially structured population of lesser kestrels: the role of spatial isolation and conspeciWc attraction. J Anim Ecol 72:400–410 Votier SC, Bearhop S, Fyfe R, Furness RW (2008a) Temporal and spatial variation in the diet of a marine top predator––links with commercial Wsheries. Mar Ecol Prog Ser 367:223–232 Votier SC, Birkhead TR, Oro D, Trinder M, Grantham MJ, Clark JA, McCleery RH, Hatchwell BJ (2008b) Recruitment and survival of immature seabirds in relation to oil spills and climate variability. J Anim Ecol 77:974–983 Votier SC, Bearhop S, Witt MJ, Inger R, Thompson D, Newton J (2010) Individual responses of seabirds to commercial Wsheries revealed using GPS tracking, stable isotopes and vessel monitoring systems. J Appl Ecol 47:487–497 Weimerskirch H (1992) Reproductive eVort in long-lived birds––agespeciWc patterns of condition, reproduction and survival in the wandering albatross. Oikos 64:464–473 Weimerskirch H (2002) Seabird demography and its relationship with the marine environment. In: Schreiber EA, Burger J (eds) Biology of marine birds. CRC Press, London, pp 115–135 Weimerskirch H (2007) Are seabirds foraging for unpredictable resources? Deep Sea Res Part 2 Top Stud Oceanogr 54:211–223 Weimerskirch H, Akesson S, Pinaud D (2006) Postnatal dispersal of wandering albatrosses Diomedea exulans: implications for the conservation of the species. J Avian Biol 37:23–28 Weimerskirch H, Pinaud D, Pawlowski F, Bost CA (2007) Does prey capture induce area-restricted search? A Wne-scale study using GPS in a marine predator, the wandering albatross. Am Nat 170:734–743