Introduction

Ecologists have long used niche theory to define the ecological niche of a species or population as a whole, assuming that all individuals are similarly affected and ecologically equivalent. However, many apparently generalized and specialized species and natural populations are in fact composed of ecologically heterogeneous individuals that repeatedly differ in foraging behaviour and use different subsets of the available resources (Bolnick et al. 2002; Bearhop et al. 2004; Bell et al. 2009). Such intra-specific variation in individual foraging strategies can be large and vary according to factors such as sex, age, morphology and individual specialization (Bolnick et al. 2003). Individual specialization refers to the use of a relatively limited fraction of the possible range of available resources, resulting in inter-individual niche variation, which could be consistent over short- and/or long-term periods, and is known to be widespread across a diverse set of taxa (Bolnick et al. 2003; Bell et al. 2009), including several seabird species (e.g. Woo et al. 2008; Votier et al. 2010a; Ceia et al. 2012). The inter-individual variation in resource use may have a strong impact on ecological processes and population dynamics, and there is evidence that such traits may reduce intra-specific competition among individuals promoting their ecological segregation (Bolnick et al. 2003; Patrick and Weimerskirch 2014a). However, its ecological implications at the individual and population level are still poorly known (Araújo et al. 2011), particularly for pelagic seabird species, as most of them spend the majority of their annual cycle at sea, often changing habitats and foraging strategies throughout their annual cycle (Mackley et al. 2010; Weimerskirch et al. 2014).

As marine top predators, seabirds are closely dependent on highly patchy distributed food resources they can find at sea, which are predictable at large spatial scales (100 s–1000 s of kilometres) (Weimerskirch 2007). To cope with these conditions, seabirds developed specific morphological characteristics and foraging strategies that enable them to conserve the maximum energy possible while foraging at sea (Nelson 1980). Therefore, individuals may retain different strategies in order to get food, leading to variability not only among species, but also within species and populations, and promoting disruptive selection and ultimately speciation (Bolnick 2004). Moreover, all seabirds are central-place foragers for at least part of the breeding season, which could potentially lead to partition in the use of available resources within and between species, according to the principle of competitive exclusion (Pianka 2011, e.g. Wakefield et al. 2013). Notwithstanding, it is not sufficient to document the foraging ecology of a species or a population as a whole, but it is crucial to study inter-individual variation in resource use within populations and to investigate its wider implications. This is important to understand trophic relationships between seabirds and the marine environment, in order to support theoretical ecological hypotheses and develop conservation strategies (Wakefield et al. 2011). Individual specialization does not refer only to behaviour associated with the foraging and feeding strategies of seabirds, but is also associated with the mate choice and territories at the colony sites, among other traits (Jouventin and Bried 2001). Nonetheless, this review is focusing on the foraging (i.e. searching for food) and feeding (i.e. diet and prey choice) strategies of seabirds, which are among the main traits directly related to the survival of the organism.

Typically, a given population or species could be considered as “specialist” or “generalist” according to the feeding habits of its conspecific individuals. A specialist population assumes that all its individuals are preying on the same food resources and have similar ecological niches, resulting in an overall small dietary niche width. On the other hand, generalist populations that show a large dietary niche width might be composed by generalist individuals all taking a wide range of food types (Type A generalization) or individuals each specializing on a different but narrow range of food types (Type B generalization) (Bearhop et al. 2004; Jaeger et al. 2009; Vander Zanden et al. 2010). However, the potential types of individual specialization in the foraging and feeding strategies of seabirds are not restricted to the diet consumed and may also include specialization in the foraging, searching and diving behaviour, site fidelity, consistency in the migratory and foraging routes and consistency in the non-breeding areas.

The fact that individuals of the same species or population may use different resources across time is crucial for constructing models of intra-specific competition, predator–prey interactions and food-web structure. However, understanding this issue using conventional approaches (e.g. pellets, regurgitates or stomach contents) requires laborious sampling of individuals over extended time periods, which is often logistically constrained, especially when working with pelagic seabirds as they spend most of their lifetime at sea. Determining at-sea distribution and habitat use using census at sea is also very challenging because seabirds are often patchily distributed over vast ocean areas. The use of stable isotope analyses (SIA) to infer diet and habitat selection is a powerful approach, as stable isotope ratios can integrate dietary information over different temporal scales depending on the tissue analysed (Inger and Bearhop 2008). For instance, blood retains information on diet up to the previous 3–4 weeks of sample collection (Hobson and Clark 1993) and feathers represent diet during the moult, unrelated to the sampling period since feather keratin is metabolically inert after synthesis (Quillfeldt et al. 2010; Ramos et al. 2011). Hence, the isotopic signature of different tissues provides information on the short- and long-term consistency in the isotopic niche of an individual (Ceia et al. 2014a). Logger devices (e.g. GPS, activity recorders, stomach temperature probes, geolocators, time-depth recorders, accelerometers) can be used simultaneously to obtain information on successive foraging trips, behaviour, foraging and feeding locations, and long-range movements of individual birds in the wild (Hunt and Wilson 2012). These techniques are particularly useful in the study of ecological niche mutually in space, time and trophic dimensions, especially when their use is combined (Navarro et al. 2013). These characteristics of SIA and tracking devices make them appealing to examine isotopic variance and differences in feeding behaviour within and among individuals and thus can be an effective way to investigate individual specialization (Newsome et al. 2007).

Although it is proven that individual specialization exists in many populations of seabirds (see Table 1), to our knowledge no studies summarized its incidence and investigated its wider implications. Therefore, the objectives of this review are to: (1) identify seabird populations with the incidence of individual specialization and (2) summarize all data gathered on this issue. For a better perception of the current knowledge worldwide, we summarized the documented incidence of individual specialization systematically by species, families and orders of seabirds, and ocean regions. Possible aspects on the ecological implications of individual specialization in seabirds based on what has been referenced are also discussed.

Table 1 Literature review (up to April 2015) of type of individual specialization in the foraging and feeding strategies of seabirds
Full size table

Methods

We reviewed a broad array of scientific publications, published in international scientific journals, to collate available data on individual specialization in the foraging and feeding strategies of seabirds. The literature sources were identified from a systematic search on the Internet and academic databases (e.g. ISI Web of Science and Scopus and ScienceDirect) up to April 2015, including a combination of the search term “seabird” or additional terms reflecting the main types of seabird “albatross”, “shearwater”, “petrel”, “prion”, “gull”, “kittiwake”, “tern”, “skua”, “jaeger”, “auk”, “murre”, “guillemot”, “puffin”, “penguin”, “gannet”, “booby”, “cormorant” or “shag”, and any of the terms “individual specializ(s)ation”, “foraging specializ(s)ation”, “consistency”, “route fidelity”, “foraging fidelity” or “site fidelity”. We also examined other studies otherwise known to authors of this review. Despite our extensive literature review, we acknowledge that there is a publication bias because researchers that did not find significant specialization within individuals (negative result) may not have reported their results. We focused this review on individual specialization in foraging and/or feeding strategies within seabird populations relying on the marine environment for at least part of the year, both in the short term (within season) and in the long term (across seasons and/or years). We excluded studies focusing on differences in foraging and feeding strategies between sexes (e.g. Phillips et al. 2011), colonies (e.g. Catry et al. 2011) or subcolonies (e.g. Ceia et al. 2015), and age classes (e.g. Lecomte et al. 2010), unrelated to the individual level and that could confound “true individual specialization”. Many seabird species are sexually dimorphic leading to different morphological characteristics that potentially promote sex-specific foraging and/or feeding specialization, and ideally, this should be controlled for before assessing the repeatability in foraging behaviour and diet (Ratcliffe et al. 2013). Likewise, we did not review studies focusing on the variability in the foraging and feeding strategies of seabirds from different colonies or subcolonies. For the most literature sources, we were unable to distinguish data in relation to age because many seabirds sampled in those studies were of undefined age class. However, it should be noted that comparing foraging and feeding strategies between sexes, colonies and age classes should provide insights into the degree of specialization in different groups.

During the search, we considered extant seabird species belonging to the orders, and respective families, of Procellariiformes (families: Diomedeidae, Procellariidae, Hydrobatidae and Oceanitidae), Charadriiformes (families: Laridae, Stercorariidae and Alcidae), Sphenisciformes (family: Spheniscidae), Suliformes (families: Sulidae, Phalacrocoracidae and Fregatidae) and Phaethontiformes (family: Phaethontidae). We excluded wader bird families belonging to the order of Charadriiformes from our search. Additionally, members of the family Anhingidae (Suliformes) were not considered because they are mainly found in freshwater lakes and streams. Thus, seabird populations considered in the present study belong to three groups according to the typical (i.e. year-round) marine zone exploited: “oceanic” (marine waters above open ocean, typically >200 m in depth), “neritic” (marine waters above continental shelf or around near-shore oceanic islands, typically <200 m in depth) and “coastal” (marine waters along coasts including estuaries, typically <8 km from the shoreline) (Croxall et al. 2012). The results were then summarized by orders, families and species or subspecies. We documented the type of individual specialization in the foraging and feeding strategies of seabirds identified in each study analysed, and we used FAO fishing area boundaries (http://www.fao.org/fishery/area/search/en) to classify the 19 core ocean regions corresponding to the study sites, which were then combined into ten main geographic areas to reduce some of the smaller-scale sampling biases. We followed the taxonomy of BirdLife International (2014).

The documented incidence of individual specialization

In total, we identified 81 scientific publications searching for any type of individual specialization in the foraging and feeding strategies of 42 seabird species (Table 1). Some publications examined individual specialization in different seabird species, years and/or study sites, resulting in a total of 94 records (hereafter studies), of which 82 identified any type of individual specialization (positive result) and only 12 documented little or no individual specialization (negative result). Although the database of published studies used in this review goes back to 1980, the great majority of the studies were published after the year 2000 (i.e. 89.4 %), highlighting the increasing focus on individual specialization in the foraging and feeding strategies of seabirds during the past 15 years. Overall, we found that most studies focused on the order Charadriiformes (n = 42) and that an intermediate number focused on Procellariiformes (n = 26) and Suliformes (n = 21). Five studies focused on Sphenisciformes and none in the order Phaethontiformes.

Evidences of individual specialization

It is clear that with the development of new technology, such as tracking devices and stable isotope analyses, it became possible to increase our knowledge on individual specialization of seabirds by investigating accurately and repeatedly their feeding ecology. To date, 40 species representing 11.7 % of the total extant seabird species (i.e. 341 seabird species, BirdLife International (2014)) exhibited some type of individual specialization in foraging and/or feeding strategies, but this proportion is certainly limited by existing studies. Our review shows that individual specialization is documented in all seabird orders (Fig. 1), with the exception of Phaethontiformes (possibly because no studies are available yet), which comprises only one family and three species, and hence potentially widespread among seabirds. No studies documented the incidence of individual specialization in the families Hydrobatidae and Oceanitidae (Procellariiformes), but this is not surprising as these Procellariiform families include species that are too small to study with currently available data-loggers. Additionally, we found no studies documenting individual specialization in the families Fregatidae (Suliformes) and Phaethontidae (Phaethontiformes; Fig. 2). The eight members of these two families (i.e. five frigatebird and three tropicbird species) are predominantly from tropical ocean regions, and there are no studies available in these species concerning individual specialization. In fact, our results showed that a greater number of studies were conducted in ocean regions at higher latitudes (i.e. in temperate and polar regions) than at lower latitudes (i.e. tropical regions; Fig. 3). This fact leads to ambiguity when drawing conclusions. If, on the one hand, there is a general gap in the study of individual specialization of seabirds at lower latitudes, on the other hand this trait could be more conspicuous and common at higher latitudes. Overall, our results reinforced the need for documenting either the presence or the absence of individual specialization in the studied seabird populations, particularly for tropical regions, because both positive and negative results are crucial to better understand this issue.

Fig. 1
figure 1

Number of seabird species in which any type of individual specialization in foraging and/or feeding strategies was positive and negatively documented, and total number of species for the five seabird orders, according to BirdLife International (2014). The fraction (%) between the number of studied species and the total number of species for each order is shown in the bars. The order of Charadriiformes does not comprise wading birds, and the order Suliformes does not comprise the family Anhingidae

Full size image
Fig. 2
figure 2

Number of seabird species in which any type of individual specialization in foraging and/or feeding strategies was positive and negatively documented, and total number of species for the 12 extant seabird families in accordance with BirdLife International (2014) (wading bird families and the family Anhingidae were not considered). The fraction (%) between the number of studied species and the total number of species for each family is shown in the bars

Full size image
Fig. 3
figure 3

Ocean regions showing the number of studies (black bars) and seabird species (grey bars) in which any type of individual specialization in the foraging and/or feeding strategies of seabirds was documented (both positive and negative results). The ten ocean regions were combined based on FAO fishing area boundaries

Full size image

We found few studies that documented little or no individual specialization in the foraging and feeding strategies of some seabird populations (Table 1b). Soanes et al. (2013) examined repeatability between the first and second foraging trips made by northern gannets Morus bassanus in the Northeast Atlantic and found no apparent short-term consistency to specific foraging areas (but see Wakefield et al.). Hamer et al. (2001) described contrasting foraging strategies of individual northern gannets at two Northeast Atlantic colonies; their results indicate a much lower degree of foraging area fidelity at one of the studied colonies. Ceia et al. (2014b) showed fluctuations on short-term consistency in the isotopic niche along a three-year study in a Procellariiform species, the Cory’s shearwater Calonectris borealis, at a Northeast Atlantic colony. Granadeiro et al. (2011) tracked 67 individual black-browed albatrosses Thalassarche melanophrys in order to study the interactions between individual birds and vessels in the Southwest Atlantic, and found that less than 5 % of the total birds returned to vessels with which they had interacted in a previous trip. Moreover, Granadeiro et al. (2014) found no long-term trophic consistency of individual black-browed albatrosses with regard to fisheries in the same region, suggesting that this species does not specialize in discards. Wanless et al. (1990) documented little foraging area fidelity in common murres Uria aalge, razorbills Alca torda and Atlantic puffins Fratercula arctica in the Northeast Atlantic. Votier et al. (2007) found high levels of annual variation in seabird consumption for great skuas Catharacta skua in two Northeast Atlantic colonies (Foula and Hermaness, UK) and Carneiro et al. (2014) found no clear evidence for specialist brown skuas Catharacta antarctica at South Georgia.

Fluctuating selection of individual specialization

Individual specialization in foraging strategies, diets, locations or niches of seabirds has been described in several seabird populations (Table 1a), but not in others (Table 1b). The studies reviewed in this work show variability in individual specialization within species and populations. Nonetheless, it remains unclear why there are such a high number of specialized and consistent individuals within some populations and not in others. Woo et al. (2008) suggest that the answer may be related to temporal changes in the predictability of resources, and Svanbäck and Persson (2004) showed that individual specialization in the perch Perca fluviatilis may fluctuate with population density through feedback mechanisms via resource levels. In fact, Ceia et al. (2014b) showed that individual variation in resources use during the short term may fluctuate over time (i.e. among years and periods) in a population of Cory’s shearwaters depending on resource availability (i.e. individual short-term consistency was higher during periods with a regular availability of food resources and lower when food availability was either abundant or scarce). This agrees with the theory that dietary specializations at the individual level can disappear under influence of prey availability, and hence competition, and a generalist strategy be pursued by most individuals (Svanbäck and Bolnick 2005). Therefore, fluctuations on individual specialization in seabirds may be related to temporal changes in the availability and predictability of resources and, thus, vulnerable to extrinsic factors, as suggested by Hamer et al. (2001), Woo et al. (2008) and Ceia et al. (2014b). Similarly, studies in wading birds such as oystercatchers Haematopus ostralegus (van de Pol et al. 2010) show that individuals with different strategies present differences in fitness between years, and therefore, a specific individual specialization may fluctuate depending on biotic and abiotic conditions; Catry et al. (2014) show that individual specialization in black-tailed godwits Limosa limosa islandica may fluctuate between years due to the frequency of specialists, competitor density, prey abundance and environmental conditions. As a consequence, birds display distinct degrees of individual specialization in diet and habitat use, indicating the presence of both generalists and specialists within a population (Catry et al. 2014). This allows individual feeding preferences and adjustments to spatial, seasonal and inter-annual variations in resources by exploiting different foraging niches (Moreno et al. 2009; Ramos et al. 2011; Ceia et al. 2014a). That is really the only way that a diversity of specialization can be maintained indefinitely within the population. Overall, these studies showed intra-individual variability in the foraging and feeding strategies of some seabird populations, at least during specific periods, suggesting that this trait could differ between species and/or populations or even oscillate in the same population over time (i.e. it may vary spatio-temporally among/within species and populations).

Ecological implications of individual specialization

Many seabird populations may present a high number of specialized individuals in the foraging and feeding strategies, but it remains unclear what are the ecological implications of individual specialization at the individual and population levels. Few studies addressed this question, and information on the implications of specialization is scarce (Araújo et al. 2011). During our literature review, some studies (e.g. Woo et al. 2008; Ceia et al. 2012) did not find any apparent differences in the fitness of specialists and generalists. However, some evidences show that individual specialization over short-time periods may confer an advantage in terms of birds’ condition status and breeding performance or foraging effort. For instance, Patrick and Weimerskirch (2014b) demonstrated that successful breeders of black-browed albatrosses had a considerably narrower niche width than unsuccessful breeders. Nevertheless, most studies showing some benefit in individual specialization were in seabird species with apparently broader niches such as skuas and gulls. Furness et al. (2006) demonstrated that the body condition, survival and breeding performance of individual great skuas may be improved by choosing to winter in areas where there are large fisheries providing discards. Anderson et al. (2009) found that individual specialization in the diet of brown skuas influenced timing of breeding and potentially the reproductive output of breeding pairs. Ceia et al. (2014a) showed that individual consistency was high and stable across two years of contrasting food availability in yellow-legged gulls, although birds differed in foraging behaviour and effort. Pierotti and Annett (1991) and Annett and Pierotti (1999) found that variation in diet choice was related to individual breeding performance in Arctic herring gulls Larus smithsonianus and western gulls Larus occidentalis, respectively. Similarly, Ronconi et al. (2014) showed that higher trophic level foraging in Arctic herring gulls was associated with better body condition. Watanuki (1992) showed that individuals of slaty-backed gulls Larus schistisagus specializing on seabird chick prey raised more fledglings than those delivering mostly fish. Overall, these studies suggest that the ecological implications of individual specialization at individual level, apparently more conspicuous in seabirds with broader species level niches, may lead to improvements in several ecological traits (especially in breeding performance) possibly due to substantial advantages in specialization on certain resources (e.g. higher energetic content, lower foraging effort). However, it should be noticed that these associations between individual specialization and fitness were found over short-time periods, and one breeding season (or part of it in many studies) is not enough to draw any conclusions about the evolutionary importance of individual specialization. As previously reported, such selection for specialization often fluctuates; thus, we should be critical when considering that specialists tend to have higher breeding performance than generalists over long-time periods. If there was really directional selection for specialization, then all individuals would quickly become specialists, promoting speciation.

The repeatability of behaviour is a widespread phenomenon in animals, but the proximate reasons for it are unresolved. Therefore, at this point, we place the question “why do individuals specialize?” Possibly there are many and different reasons, and the answer could be related to the underlying causes of specialization such as variation in morphology, physiology, personality or cognitive abilities. Many animals, including seabirds, are highly variable among individuals but highly repeatable within individuals for hormone levels (e.g. Seltmann et al. 2012; Elliott et al. 2014b), basal metabolism (e.g. Bech et al. 1999) and metabolic rate (e.g. Careau et al. 2008; Biro and Stamps 2010; Elliott et al. 2014a), personality (e.g. van Oers et al. 2004) or cognitive abilities (e.g. Kitaysky et al. 2006). Moreover, recent research suggests that colonies may be important as sources of social information allowing animals to make decisions based on information sharing from conspecifics (Wakefield et al. 2013; Evans et al. 2015). Nevertheless, whatever the ecological implications and the underlying causes of specialization, individual specialization within a seabird’s population has an important role on its ecology, foraging behaviour and dynamics, and may contribute to reduce intra-specific competition among individuals (Ceia et al. 2012; Patrick and Weimerskirch 2014a; Oppel et al. 2015), as suggested in other taxonomical animal groups (e.g. Svanbäck and Bolnick 2007, Vander Zanden et al. 2010, Matich et al. 2011).

Final remarks

Our review documented, for the first time, a full list of studies identifying individual specialization in the foraging and feeding strategies of seabirds, an area in which the knowledge is scarce for the general diversity of taxa and particularly for seabirds. Some degree of individual foraging specialization was previously documented in a large range of marine taxa, such as invertebrates (e.g. Burrows and Hughes 1991), fishes (e.g. Matich et al. 2011), sea turtles (e.g. Thomson et al. 2012), marine mammals (e.g. Tinker et al. 2008) and seabirds (this review), suggesting that this could be a widespread phenomenon.

Overall, this review demonstrates that inter-individual variation in resource use occurs within several populations and species of seabirds, which is necessary for natural selection, and hence adaptation. By extension, individual specialization may play an important role in the foraging dynamics of the populations by reducing, for instance, intra-specific competition, but further investigation is required in order to understand how ecological implications at individual level may be related to several traits such as reproductive success, body condition or foraging effort. Such implications though could vary according to species and spatio-temporal variation of the populations, and respective incidence of individual specialization.

Future research

In general, a small number of reviews on individual specialization have been published, and this is the first one specifically on seabirds. Our review documented its occurrence at around 12 % of the extant seabird species. However, this proportion is most likely limited by existing studies (especially in small or tropical seabird species), and many unanswered ecological questions require future research as trophic interactions among individual conspecifics and their relationships with the marine environment are particularly complex. Apart from documenting the existence of individual specialization, further studies should explore the timescale at which an individual’s specialization is consistent (i.e. during the short and long term), investigate the relative incidence in different populations, species and communities, determine the fitness implications of different strategies, and, finally, identify the mechanisms that generate inter-individual variation. Given the long history of diet analyses in seabird ecology, there are a substantial number of appropriate data sets that in conjunction with recent powerful techniques such as stable isotope analyses, logger devices and software would permit a wide range of theoretical hypotheses of niche variation to be tested and validated. However, for that, it is essential to document the negative results (i.e. studies in which individual specialization was tested in a given population but not validated). Then, it would be possible to explore in a robust meta-analysis the general veracity of this hypothesis, its spatio-temporal fluctuations (among/within species and populations) and the implications for fitness, fecundity and survival.