INTRODUCTION
Two pinnipeds from the Phocidae family inhabit the west coast of the Baja California (BC) Peninsula, Mexico: the harbour seal (Phoca vitulina) and the northern elephant seal (Mirounga angustirostris). The harbour seal is widely distributed from Alaska, USA to BC, Mexico (Thompson & Härkönen, Reference Thompson and Härkönen2008). The subspecies Phoca vitulina richardsi inhabits a number of islands along the BC Peninsula, including Coronados, Todos Santos, San Martin, San Jeronimo, San Benito, Cedros, Natividad, San Roque and Asuncion (Gallo-Reynoso & Aurioles-Gamboa, Reference Gallo-Reynoso and Aurioles-Gamboa1984). The coastal Washington and Oregon, USA, populations, as well as those inhabiting Washington's interior waterways, have reached their carrying capacity and are no longer increasing (Jeffries et al., Reference Jeffries, Huber, Calambokidis and Laake2003; Brown et al., Reference Brown, Wright, Riemer and Laake2005). The current abundance trend for Alaskan populations is largely unknown, although both decreases (Glacier Bay) and increases (Bristol Bay) have been reported (Allen & Angliss, Reference Allen and Angliss2012).
The northern elephant seal (Mirounga angustirostris) shows an increasing population trend, successfully rebounding from the verge of extinction toward the end of the 19th century (Campagna, Reference Campagna2008). The breeding distribution of the northern elephant seal spans ~15 islands located mainly in southern California, USA, and to a lesser extent off the west coast of the BC Peninsula, Mexico. In Mexico, the species inhabits several islands, including Coronado, San Martin, Guadalupe, San Benito, Cedros, Natividad, Todos Santos, San Jeronimo and Asuncion (Campagna, Reference Campagna2008; Franco-Ortiz, Reference Franco-Ortiz2012; Lowry et al., Reference Lowry, Condit, Hatfield, Allen, Berger, Morris, Le Boeuf and Reiter2014). Compared with the larger colonies in the USA and Canada, this is a marginal habitat at the southern limit of both species' distributions.
Both phocid species are generalists. At other latitudes (e.g. California, Washington and Alaska), harbour seals consume yellow fin goby (Acanthogobius flavimanus), coho salmon (Oncorhynchus kisutch), pink salmon (O. gorbuscha), walleye pollock (Theragra chalcogramma), capelin (Mallotus villosus), Pacific herring (Clupea pallasii), cephalopods (squid and octopus), crustaceans and other species (Pitcher, Reference Pitcher1980; Gibble, Reference Gibble2011; Lance et al., Reference Lance, Chang, Jeffries, Pearson and Acevedo-Gutiérrez2012; Luxa & Acevedo-Gutiérrez, Reference Luxa and Acevedo-Gutiérrez2013). On Natividad Island, BC, harbour seals consume Pacific red octopus (Octopus rubescens) and fishes like Mazatlan sole (Achirus mazatlanus), California needlefish (Strongylura exilis) and California lizardfish (Synodus lucioceps) (Elorriaga-Verplancken et al., Reference Elorriaga-Verplancken, Morales-Luna, Moreno-Sánchez and Mendoza-Salas2013b).
Foraging by northern elephant seals has been documented primarily among individuals from colonies in California. Its diet is composed of some 30 prey species, including cephalopods and myctophid fish (McGinnis & Schusterman, Reference McGinnis, Schusterman, Ridgway and Harrison1981; Condit & Le Boeuf, Reference Condit and Le Boeuf1984), as well as deep-sea squid (e.g. octopus squid and the boreal clubhook squid), fish (e.g. hake), and crustaceans (e.g. red pelagic crab) (Antonelis et al., Reference Antonelis, Lowry, DeMaster and Fiscus1987, Reference Antonelis, Lowry, Fiscus, Stewart, DeLong, Le Boeuf and Laws1994). It is difficult to assess the elephant seal diet using traditional methods like scat analysis or stomach lavage because they feed at sea for long periods of time and fast when they are on land to breed or moult (Condit & Le Boeuf, Reference Condit and Le Boeuf1984). However, telemetry analyses indicate oceanic foraging habits, particularly among adult females (Le Boeuf et al., Reference Le Boeuf, Crocker, Costa, Blackwell and Webb2000; Robinson et al., Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012), who make two annual foraging trips. The first is a short post-breeding migration (February–May), after which they return to their breeding islands for 5 or 6 weeks to moult. They then take a long post-moulting migration (June–January), returning to land to breed in December (Le Boeuf & Laws, Reference Le Boeuf, Laws, Le Boeuf and Laws1994; Robinson et al., Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012). During the post-moulting migration, females from California (Año Nuevo) and BC (San Benito) are found in foraging areas around 40–45°N. A smaller proportion of individuals from BC remain at lower latitudes, within ~500 km of San Benito, during their foraging migrations (Robinson et al., Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012).
Migratory patterns by elephant seals from colonies in the USA and Mexico have been documented using stable isotope analysis and telemetry (Le Boeuf et al., Reference Le Boeuf, Crocker, Costa, Blackwell and Webb2000; Aurioles-Gamboa et al., Reference Aurioles-Gamboa, Koch and Le Boeuf2006; Robinson et al., Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012). Unfortunately, very little information is available on harbour seals from BC, Mexico. This species is catalogued as non-migratory in areas adjacent to British Columbia, Canada, and California, USA (Spalding, Reference Spalding1964; Paulbitski & Maguire, Reference Paulbitski, Maguire and Poulter1972). Additionally, satellite-tracked adult female harbour seals from Glacier Bay, Alaska, have been observed making post-breeding migrations (Womble & Gende, Reference Womble and Gende2013).
This study is based on the analysis of stable isotopes of N and C, which is an efficient tool for determining foraging habits (Martínez del Río et al., Reference Martínez del Rio, Wolf, Carleton and Gannes2009). The relatively constant increments of δ15N and δ13C (~3–5‰ and ~0.5–2‰, respectively) can be traced from the base of the food web to the top predators (Minagawa & Wada, Reference Minagawa and Wada1984). This allows researchers to make inferences regarding both trophic position (δ15N) and habitat use (δ13C) (Newsome et al., Reference Newsome, del Rio, Bearhop and Phillips2007; Elorriaga-Verplancken et al., Reference Elorriaga-Verplancken, Aurioles-Gamboa, Newsome and Martínez-Díaz2013a). Since pups rely solely on their mothers for nutrition in utero and during lactation, tissues like lanugo – which starts to appear in some pinniped pups around the seventh month of gestation (Odell, Reference Odell1972) – reflect maternal foraging behaviour (e.g. Aurioles-Gamboa et al., Reference Aurioles-Gamboa, Koch and Le Boeuf2006; Newsome et al., Reference Newsome, Etnier and Aurioles-Gamboa2006). Isotope values reflect the prey consumed by an individual; similarly, some pup tissues – like hair, dental collagen or blood – have higher δ15N values than adults, as mothers catabolize fat and muscle to produce the milk consumed by their offspring, which in turn affects the isotope values of those offspring (Newsome et al., Reference Newsome, Etnier and Aurioles-Gamboa2006; Porras-Peters et al., Reference Porras-Peters, Aurioles-Gamboa, Cruz-Escalona and Koch2008; Habran et al., Reference Habran, Debier, Crocker, Houser, Lepoint, Bouquegneau and Das2010; Elorriaga-Verplancken et al., Reference Elorriaga-Verplancken, Morales-Luna, Moreno-Sánchez and Mendoza-Salas2013b).
In the North-east Pacific Ocean, spatial variation in δ15N and δ13C values at the base of food webs is negatively correlated with latitude (Altabet et al., Reference Altabet, Pilskaln, Thunell, Pride, Sigman, Chávez and Francois1999; McMahon et al., Reference McMahon, Hamady and Thorrold2013). These isotopic differences are strongly correlated with areas of denitrification and associated minimum oxygen zones at low–middle latitudes, with resulting increases in δ15N values at the base of the food chain (Wada & Hattori, Reference Wada and Hattori1991; Voss et al., Reference Voss, Dippner and Montoya2001). There is also an increase in temperature at lower latitudes, which decreases the mixing of CO2 (13C-depleted) with seawater, resulting in higher δ13C values at the base of the food chain (Goericke & Fry, Reference Goericke and Fry1994; Burton & Koch, Reference Burton and Koch1999; McMahon et al., Reference McMahon, Hamady and Thorrold2013).
The present study was carried out to evaluate the movements of female harbour seals on Natividad Island, a location representative of BC due to the species' abundance and relative stability throughout the year (Lubinsky, Reference Lubinsky2010). Stable isotope analysis was used to compare this species with the northern elephant seal during the same breeding season (February 2013) off the west coast of BC (Natividad Island and San Benito Archipelago; 60 km between both areas). The results of this study contribute to our ecological knowledge of both phocid species in Mexico, allowing us to define local foraging patterns; this new information complements what is known about these seals throughout their distribution.
METHODS
From 6–14 February 2013, lanugo samples were collected from 20 northern elephant seal pups (1–2 months old) on the San Benito Archipelago, BC, Mexico. Since these pups were moulting, immobilization or scissors were not required for this species, facilitating their sampling by hand (using gloves). A few days later (24–28 February 2013), hair samples were collected from 19 harbour seal pups (1–2 months old) on Natividad Island, BC, Mexico, some 60 km south of San Benito (Figure 1). Immobilization by two people was required for these pups; they were caught with nets and scissors were used to obtain the hair. All samples, for both species, were taken from the dorsal region of apparently healthy pups.
Fig. 1. Location of Natividad Island and the San Benito Archipelago on the west coast of BC, Mexico.
The fur samples were processed at the Chemistry Laboratory of the Centro Interdisciplinario de Ciencias Marinas (CICIMAR-IPN; Interdisciplinary Center for Marine Science). Samples were rinsed with distilled water and acetone/hexane (1:1), and homogenized in an agate mortar until a fine powder was obtained. Approximately 1 mg of each sample was weighed using an analytical microbalance with a precision of 0.001 mg. Samples were stored in 8 × 5 mm tin capsules and sent to the Earth and Planetary Sciences Stable Isotope Laboratory at the University of California in Santa Cruz (UCSC), where the δ15N and δ13C isotope values were determined using a Carlo Erba 1108 elemental analyser coupled to a ThermoFinnigan Delta Plus XP isotope ratio mass spectrometer, with an analytical precision of ±0.2‰ for both stable isotopes.
The stable isotope proportion of N and C is represented using delta (δ) notation. DeNiro & Epstein (Reference DeNiro and Epstein1978) proposed the following equation to determine this ratio:
$${{\rm \delta} ^{{\rm 13}}}{\rm C}\;{\rm or}\;{{\rm \delta} ^{{\rm 15}}}{\rm N} = {\rm 1000}\;\left[ {\left( {\displaystyle{{{R_{{\rm sample}}}} \over {{R_{{\rm standard}}}}}} \right) - {\rm 1}} \right]$$where δ15N and δ13C refer to the difference expressed in parts per thousand (‰) between the amount of 15N and 13C in the sample relative to the standard. R sample and R standard are the ratios of 15N/14N or 13C/12C in the sample and the standard, respectively. The international standards for these analyses are the PeeDee Belemnite formation (PDB) for carbon, with a value of 0.011‰, and atmospheric nitrogen (N2) for nitrogen, with a value of 0.004‰.
In this study, the isotopic niche characteristics of both phocids were measured using the Stable Isotope Bayesian Ellipses in R (SIBER) routine in SIAR, a package in the R programming environment. The variation in isotopic values was used to more accurately calculate the isotopic niche space of both species (Jackson et al., Reference Jackson, Inger, Parnell and Bearhop2011; R Development Core Team, 2011). This approach involves the use of Markov–Chain Monte Carlo simulations (bootstrapping) to construct ellipse parameters. Bivariate ellipses and convex hulls were used to delineate isotopic niche space (δ15N–δ13C 95% confidence interval ellipses for both pinniped species). Niche area and overlap were estimated based on 100,000 posterior draws of the Bayesian standard ellipse parameters. Isotopic data were incorporated into the posterior distribution, making Phoca and Mirounga niche space and area comparable. An additional non-parametric test (Mann–Whitney U) was performed to determine statistical isotopic differences between the two species.
RESULTS
For both stable isotopes (mean ± SD) (δ15N = 19.1 ± 0.3‰, δ13C = −15.4 ± 0.6‰), harbour seal pups had statistically higher values (δ15N: U = 61, P < 0.05; δ13C: U = 23, P < 0.05) than northern elephant seal pups (δ15N = 17.6 ± 1.3‰, δ13C = −17.2 ± 0.8‰).
The difference in variability between species was evident when observing the values for each stable isotope. Harbour seal δ15N values ranged from 18 to 19.5‰ (Δ = 1.5‰), while northern elephant seal δ15N values spanned 15.2 to 20‰ (Δ = 4.8‰). The δ13C values for harbour seals varied from −16.1 to −13.5‰ (Δ = 2.6‰), while those for northern elephant seals fell between −18.2 and −15.1‰ (Δ = 3.1‰).
The Bayesian analysis (SIBER) confirmed this distinction between phocids, with a low overlap value (0.4) and different isotopic spaces [Phoca: 1.7 (polygon) and 0.5 (ellipse); Mirounga: 4.2 (polygon) and 1.6 (ellipse)] (Figure 2).
Fig. 2. δ15N and δ13C breadth and overlap (SIBER analysis) for harbour seal (Phoca vitulina richardii) pups from Natividad Island and northern elephant seal (Mirounga angustirostris) pups from the San Benito Archipelago, BC. Large ellipse (Mirounga) = 1.6; small ellipse (Phoca) = 0.5. Overlap between phocids = 0.4.
DISCUSSION
Our study provides insights into the foraging behaviour of harbour seals from Natividad Island and northern elephant seals from the San Benito Archipelago; the former appears to be resident (local consumption of resources), while the latter is primarily migratory. The available information on elephant seal foraging behaviour at other latitudes indicates that this phocid includes several squid species in its diet; small sharks and other fish species have been documented occasionally as well (Campagna, Reference Campagna2008). If both phocids feed in ecosystems with similar base δ15N values, the higher δ15N values for elephant seals might reflect the high trophic position of sharks. However, the vast majority (85%) of elephant seals in this study had 15N-depleted values.
While comparing elephant seals from BC and California, Aurioles-Gamboa et al. (Reference Aurioles-Gamboa, Koch and Le Boeuf2006) pointed to a possible decrease of ~1‰ (δ15N and δ13C) for each 4–5° increase in latitude in the North-east Pacific Ocean. This proposed variation (or a similar one) likely means that our sampled harbour seals are resident at Natividad. In contrast, San Benito elephant seals could primarily migrate to northern foraging locations, probably near California. We cannot establish an accurate latitudinal gradient (‰) in this regard because: (1) we studied two different species, (2) isotopic variability exists within groups, and (3), because we did not analyse satellite-tracked individuals, we cannot yet confirm specific foraging areas by independent means.
Differences in foraging behaviour between phocids
The narrower isotopic niche (Newsome et al., Reference Newsome, del Rio, Bearhop and Phillips2007) of harbour seals relative to northern elephant seals provides additional support for our observation of regional foraging behaviour among harbour seals in the study area, at least during the 4 or 5 months prior to sample collection. The difference in isotopic variability between these two phocids may be related to the variety of foraging strategies they employ or to the elephant seal's consumption of prey from different latitudes (different prey or the same prey with different isotopic values). Based on the SIBER analysis, foraging segregation was more prominent among elephant seal pups, providing support for latitudinal separation of adult females of both species.
The δ15N values for the two phocids were the most divergent (difference between the maximum and minimum δ15N value); their δ13C values were less variable. Additionally, the δ13C variation among our elephant seal pups (SD = ±0.8‰) was similar to or greater than that reported by Aurioles-Gamboa et al. (Reference Aurioles-Gamboa, Koch and Le Boeuf2006) for elephant seal pups from San Benito (SD = ±0.9‰) and California (SD = ±0.4‰) based on fur samples, and by Habran et al. (Reference Habran, Debier, Crocker, Houser, Lepoint, Bouquegneau and Das2010) for adult female elephant seals from California (SD = ±~0.4–0.7‰) based on blood (red cells and serum) samples.
A few elephant seal isotopic values were close to those of the harbour seal group, suggesting a degree of similarity. This may be the result of Mirounga behaviour. Robinson et al. (Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012) used telemetry data to determine that 80% of tagged elephant seals (adult females) from San Benito fed in the same areas as the Año Nuevo (California) individuals during the post-moulting migration, and that a subset (20%) of individuals from San Benito stayed to feed locally, with important foraging activity around Natividad Island and other sites west of BC. All tagged San Benito elephant seals maintained their strategies (local or migratory) during the post-breeding season. However, they did not migrate as far north (Robinson et al., Reference Robinson, Costa, Crocker, Gallo-Reynoso, Champagne, Fowler, Goetsch, Goetz, Hassrick, Hückstadt, Kuhn, Maresh, Maxwell, McDonald, Peterson, Simmons, Teutschel, Villegas-Amtmann and Yoda2012), partially explaining the isotopic values reported by Aurioles-Gamboa et al. (Reference Aurioles-Gamboa, Koch and Le Boeuf2006), who argued that San Benito elephant seals feed south of California seals. The few elephant seal isotopic values that are similar to those of harbour seals may reflect this local foraging by Mirounga in BC. If these outliers are removed from our SIBER analysis, the overlap value between species is reduced from 0.4 to 0.
The elephant seal δ15N values reported here are similar to those reported for the same location based on analysis of pup hair (17.6‰ vs 17.7‰) (Aurioles-Gamboa et al., Reference Aurioles-Gamboa, Koch and Le Boeuf2006). However, our δ13C values were closer to those reported at Año Nuevo, CA (−17.2‰ vs −17.6‰) than values from San Benito (−16.1‰) (Aurioles-Gamboa et al. (Reference Aurioles-Gamboa, Koch and Le Boeuf2006). There are three possible explanations for this pattern, which are not mutually exclusive: (1) the elephant seals in our study fed in areas near where Año Nuevo individuals feed (δ13C), but at a higher trophic position (δ15N), (2) the elephant seals in our study fed at a trophic position similar to that previously reported for San Benito (δ15N), but they did so further out to sea (δ13C), or (3) the base isotopic values for San Benito were not the same during the 2002 (Aurioles-Gamboa et al., Reference Aurioles-Gamboa, Koch and Le Boeuf2006) and 2013 (this study) breeding seasons, complicating direct comparison between these studies.
Final remarks
It is important to continue gathering information regarding the ecology of these phocids in different segments of their populations; the Natividad and San Benito colonies in Mexico are located at the southern limit of these species' global distributions; thus, these populations exhibit characteristics (e.g. lower density of individuals) distinct from their counterparts at higher, and better studied, latitudes.
Based on existing data and inferences, we cannot entirely discard the possibility that other factors, such as differences in metabolism based on pup size, may influence the comparison of these two phocids. However, that factor apparently did not have an effect on this study or others where different pinniped species (phocids and otariids) were isotopically compared and related to different foraging habits in the North Pacific (e.g. Burton & Koch, Reference Burton and Koch1999; Hirons et al., Reference Hirons, Schell and Finney2001). We made inferences regarding the 4–5 months prior to sampling (Winter 2013) using keratinous tissue (lanugo or hair) samples from both species. Thus, we do not consider that comparing lanugo from elephant seals vs hair from harbour seals significantly biased our conclusions because their biochemical composition is essentially the same; however, we cannot entirely discard any effect (e.g. growth rate difference) in this regard. Our sample size per site was similar to that of previous studies (e.g. Porras-Peters et al., Reference Porras-Peters, Aurioles-Gamboa, Cruz-Escalona and Koch2008; Páez-Rosas & Aurioles-Gamboa, Reference Páez-Rosas and Aurioles-Gamboa2010; Elorriaga-Verplancken et al., Reference Elorriaga-Verplancken, Morales-Luna, Moreno-Sánchez and Mendoza-Salas2013b); however a higher number of samples should improve representativeness. Overall, further studies are needed in order to draw more robust conclusions regarding whether the isotopic values of these top predators reflect changes at the base of the food web, inter-individual behaviour, or changes in prey availability.
ACKNOWLEDGEMENTS
We thank the Secretaría de Medio Ambiente y Recursos Naturales (SEMARNAT; Secretariat of the Environment and Natural Resources) through the Dirección General de Vida Silvestre en México (Directorate General of Wildlife in Mexico) for granting us research permit SGPA/DGVS/11309/12. FREV thanks the Instituto Politécnico Nacional (IPN; National Polytechnic Institute) for the support received through the Programa de Contratación por Excelencia (Contracting Excellence Programme). We thank the Sociedad Cooperativa de Producción Pesquera Buzos y Pescadores de la Baja California SCL (Isla Natividad) (Cooperative Society of Fishing, Divers, and Fishermen of Baja California) and the Sociedad Cooperativa de Pescadores Nacionales de Abulón (Isla Cedros) (Cooperative Society of National Abalone Fishermen) for their support in the field, and Itzel Mendoza for her help extracting and preparing samples. We also thank Kristin Sullivan for polishing the English text.
FINANCIAL SUPPORT
Financial support was provided by the Consejo Nacional de Ciencia y Tecnología (CONACYT; National Council of Science and Technology) (grant number CB-181876) and the Instituto Politécnico Nacional (IPN; National Polytechnic Institute) (grant number SIP-20130944-IPN).
