Temporal variation in trophic relationships among three congeneric penguin species breeding in sympatry

Abstract

Penguins are a monophyletic group in which many species are found breeding sympatrically, raising questions regarding how these species coexist successfully. Here, the isotopic niche of three sympatric pygoscelid penguin species was investigated at Powell Island, South Orkney Islands, during two breeding seasons (austral summers 2013–2014 and 2015–2016). Measurements of carbon (δ13C) and nitrogen (δ15N) stable isotope ratios were obtained from blood (adults) or feather (chicks) samples collected from Adélie Pygoscelis adeliae, chinstrap P. antarctica, and gentoo P. papua penguins. Isotopic niche regions (a proxy for the realized trophic niches) were computed to provide estimates of the trophic niche width of the studied species during the breeding season. The isotopic niche regions of adults of all three species were similar, but gentoo chicks had noticeably wider isotopic niches than the chicks of the other two species. Moderate to strong overlap in isotopic niche among species was found during each breeding season and for both age groups, suggesting that the potential for competition for shared food sources was similar during the two study years, although the actual level of competition could not be determined owing to the lack of data on resource abundance. Clear interannual shifts in isotopic niche were seen in all three species, though of lower amplitude for adult chinstrap penguins. These shifts were due to variation in carbon, but not nitrogen, isotopic ratios, which could indicate either a change in isotopic signature of their prey or a switch to an alternative food web. The main conclusions of this study are that (1) there is a partial overlap in the isotopic niches of these three congeneric species and that (2) they responded similarly to changes that likely occurred at the base of their food chain between the 2 years of the study.

Full text

3660
|
Ecology a nd Evolution . 2018;8:3660–3674.
www.ecolevol.org
Received: 30 March 2 017
|
Revised: 12 Ja nuary 2018
|
Accepted: 27 January 2 018
DOI: 10.100 2/ece3.39 37
ORIGINAL RESEARCH
Temporal variation in trophic relationships among three
congeneric penguin species breeding in sympatry
Arnaud Tarroux1,2 | Christian Lydersen1 | Philip N. Trathan3 | Kit M. Kovacs1
This is an op en access article under t he terms of t he Creat ive Commons Attr ibutio n License, which pe rmits use, dist ribution and rep roduc tion in any m edium,
provide d the orig inal work is proper ly cited .
© 2018 The Aut hors. Ecology an d Evolution published by John Wiley & Sons Ltd .
1Norwegian Polar Institute, Fram Centre,
Tromsø, Nor way
2Norwegian Institute for Nature Research,
Fram Centre, Tromsø, Norway
3British Antarctic Sur vey, High Cros s,
Cambr idge, UK
Correspondence
Arnaud Tarroux, Norwegian Instit ute for
Nature Re search , Fram Cent re, Tromsø,
Nor way.
Email: arnaud.tarroux@nina.no
Funding information
Norges Fo rskningsråd , Grant /Award
Number: 222798/E10
Abstract
Penguins are a monophyletic group in which many species are found breeding sym-
patrically, raising questions regarding how these species coexist successfully. Here,
the isotopic niche of three sympatric pygoscelid penguin species was investigated at
Powell Island, South Orkney Islands, during two breeding seasons (austral summers
2013–2014 and 2015–2016). Me as urement s of carbon (δ13C) and nit ro gen (δ15N) sta-
ble isotope ratios were obtained from blood (adults) or feather (chicks) samples col-
lected from Adélie Pygoscelis adeliae, chinstrap P. antarctica, and gentoo P. p apu a
penguins. Isotopic niche regions (a proxy for the realized trophic niches) were com-
puted to provide estimates of the trophic niche width of the studied species during
the breeding season. The isotopic niche regions of adults of all three species were
similar, but gentoo chicks had noticeably wider isotopic niches than the chicks of the
other two species. Moderate to strong overlap in isotopic niche among species was
found during each breeding season and for both age groups, suggesting that the po-
tential for competition for shared food sources was similar during the two study
years, although the actual level of competition could not be determined owing to the
lack of data on resource abundance. Clear interannual shifts in isotopic niche were
seen in all three species, though of lower amplitude for adult chinstrap penguins.
These shifts were due to variation in carbon, but not nitrogen, isotopic ratios, which
could indicate either a change in isotopic signature of their prey or a switch to an al-
ternative food web. The main conclusions of this study are that (1) there is a partial
overlap in the isotopic niches of these three congeneric species and that (2) they re-
sponded similarly to changes that likely occurred at the base of their food chain be-
tween the 2 years of the study.
KEYWORDS
Antarc tic krill, Euphausia superba, interspecific competition, isotopic niche, pygoscelid
penguin, stable isotopes
1 | INTRODUCTION
The coexistence of sympatric species and the extent to which
their ecological niches overlap are fundamental issues in both
theoretical and applied ecology (Silvertown, 200 4; Vellend,
2010). Phylogenetically close species, which have less differen-
tiated functional traits (i.e., more overlapping ecological niches,
sensu Hutchinson, 1957), have traditionally been thought to pose

|
3661
TARROUX eT Al .
strong competition for one another when they co- occur (“phylo-
genetic limiting similarity hypothesis”; Adler, HilleRisLambers, &
Levine, 2007; Violle, Nemergut, Pu, & Jiang, 2011). However, this
assumption has recently been challenged by theoretical and ex-
perimental studies on communities of primar y producers showing
that species’ phylogenic distances and coexistence can be unre-
lated (Fritschie, Cardinale, Alexandrou, & Oakley, 2014; Godoy,
Kraft, & Levine, 2014). Among high- trophic- level consumers, such
as seabirds, interspecific competition among closely related and
morphologically similar species can be buffered by subtle behav-
ioral adjustments which reduce their ecological overlap, for exam-
ple, using distinct foraging habitats or resources (Barger, Young,
Will, Ito, & Kitaysky, 2016; Robertson et al., 2014). Additionally,
the co- occurrence of several closely related species can result in
positive interactions such as the sharing of high- quality informa-
tio n ab ou t wh er e re sources are (Angu it a & Sime on e, 2016; Sridhar
et al., 2012). Ecological relationships among closely related spe-
cies are thus not necessarily purely competitive, which can make
understanding them quite challenging.
Penguins are a monophyletic group in which many species are
found breeding sympatrically at several sites in the sub- Antarc tic
islands (Forcada, Trathan, Reid, Murphy, & Croxall, 2006; Lynch,
Fagan, Naveen, Trivelpiece, & Trivelpiece, 2012; Niemandt et al.,
2016; Paterson, Wallis, Kennedy, & Gray, 2014; Trathan, Croxall,
& Murphy, 1996). The co- occurrence of these closely related spe-
cies has long raised questions regarding what degree of competi-
tion takes place and how these species successfully coexist (Lynnes,
Reid, Croxall, & Trathan, 2002; Trivelpiece & Volkman, 1979; White
& Conroy, 1975). This applies particularly to congeneric species
such as the three pygoscelids—Adélie Pygoscelis adeliae, chinstrap
P. antarctica, and gentoo P. papua penguins, which share similar
breeding and foraging ecologies (Hinke et al., 2015; Lynnes et al.,
2002; Negrete et al., 2017). In the South Shetland and South Orkney
Islands, these three species are found breeding sympatrically at high
densities (Lev y et al., 2016; Wilson, 2010). It is thought that they
achieve coexistence through fine- tuned ecological segregation
mechanisms. Such mechanisms can involve temporal separation
of chick- rearing periods among species (Lynch, Fagan et al., 2012;
Trivelpiece, Trivelpiece, & Volkman, 1987), use of spatially distinct
foraging habitats in two or three dimensions (Cimino, Moline, Fraser,
Patterson- Fraser, & Oliver, 2016; White & Conroy, 1975; Wilson,
2010), or specialized feeding on different types of prey when shar-
ing the same areas (Hinke et al., 2015; Negrete et al., 2017; Polito
et al., 2015). Such closely related species can be expected to reduce
the overlap in their ecological niches particularly when resources are
limited, as has been obser ved for instance between Adélie and chin-
strap penguins (Lynnes et al., 20 02).
Among pygoscelids, the breeding distribution of gentoo pen-
guins reaches much further north than that of the more ice- tolerant
Adélie and chinstrap penguins, but there is an overlap in all three
species’ breeding distributions between 54°S and 65°S (Ancel,
Beaulieu, & Gilbert, 2013; Black, 2016). Within these overlap areas,
breeding times or segregated foraging areas might serve to minimize
direct competition. Individual species do show some flexibility. For
example, the breeding phenolog y of gentoo penguins varies widely
throughout their breeding range, with later laying dates at more
southern latitudes (Black, 2016; Levy et al., 2016). Gentoo penguins
also tend to forage closer to shore and deeper in the water column
than chinstrap or Adélie penguins (Cimino et al., 2016; Lynnes et al.,
2002; Trivelpiece et al., 1987). Despite these differences, and be-
cause of their strong reliance on Antarc tic krill Euphausia superba
(hereafter simply referred to as krill) as a food source (Ratcliffe
& Trathan, 2011), all three species are listed by the CCAMLR
Ecosystem Monitoring Program (CCAMLR 2007) as sentinels of
change in critical components of the Southern Ocean food web.
Furthermore, penguin species in general are particularly sensitive to,
and thus good indicators of, the oceanographic conditions prevailing
near their breeding site, as they respond rapidly to fluctuations in
resource abundance during the breeding season, through changes
in reproductive success within a single breeding season as well as
potential short- term changes in population size (Boersma, 2008;
Browne, Lalas, Mattern, & Van Heezik, 2011; Trathan et al., 2015).
A better understanding of the mechanisms shaping their respective
ecological niches, and how these vary in the context of ecosystem
modification through climate change (Miller, Kappes, Trivelpiece, &
Trivelpiece, 2010; Polito et al., 2015), would help strengthen their
value as biological indicators.
Measurements of stable isotopes of carbon (δ13C) and nitrogen
(δ15N) in consumers’ tissue reflect those of their prey and of the
relative proportion of each prey in the consumers’ diet (DeNiro &
Epstein, 1978, 1981; Kelly, 20 00). Isotopic ratios obtained through
a single sampling event can provide dietary information integrated
over a period of time that depends on the tissue analyzed as well
as the species considered, ranging for instance from a few days in
blood plasma to several weeks in red blood cells (Cherel, Connan,
Jaeger, & Richard, 2014; Hobson & Clark, 1993). Southern Ocean
marine predators such as penguins breed in remote areas and feed
at sea, making them challenging to sample regularly for dietary in-
takes. The use of isotopic measurements that directly integrate their
average diet over the past days or weeks can thus prove particularly
useful for these species. Recently developed statistical approaches,
such as Bayesian niche ellipses (Jackson, Inger, Parnell, & Bearhop,
2010; Swanson et al., 2015), have increased the potential for more
refined studies of the trophic niche both at population- and individ-
ual levels. These methods allow the use of individual isotopic ratios
within a given population or group to estimate an n- dimensional
isotopic niche (n depending on the number of isotopes used). The
isotopic niche can subsequently be interpreted as a proxy for the
realized trophic niche, thereby providing valuable information on
the part of a species’ ecological niche that relates to the use of food
resources (Newsome, Martínez del Rio, Bearhop, & Phillips, 2007;
Yeakel, Bhat, Elliot t Smith, & Newsome, 2016). Pygoscelid penguins
can feed at various trophic levels, ranging from low- trophic- level
zooplankton characterized by lower δ15N values, such as krill, to
squid and fish that are characterized by higher δ15N values (Negrete
et al., 2017). This gradient in δ15N values allows discrimination

3662
|
TA RROUX eT A l.
am on g ind ividu als/po pul at ion s fe edi ng most ly on kril l vs thos e fee d-
ing mostly on fish/squid species (Juares, Santos, Mennucci, Coria, &
Mariano- Jelicich, 2016; Polito, Lynch, Naveen, & Emslie, 2011).
Herein, the isotopic niche of Adélie, chinstrap, and gentoo pen-
guins breeding at Powell Island, South Orkney Islands, was inves-
tigated during two nonconsecutive breeding seasons (2013–2014
and 2015–2016). Our main objectives were firstly to investigate
interannual variation/stability in the isotopic niche of each species,
and secondly if variation did occur, to determine whether it affected
all three species similarly. Specific ally, using measurements of car-
bon and nitrogen stable isotope ratios obtained from tissue samples
collected from both adults and chicks, this study (1) quantifies the
isotopic niche width and interindividual variation in isotopic ratios
in pygoscelids during part of their breeding season, (2) assesses the
potential for competition by measuring the interspecific overlap in
isotopic niche, and (3) provides an interspecific comparison of the
occurrence of an interannual shift in isotopic niche. Based on their
phylogenetic relatedness, a strong overlap among the isotopic niches
of the three species was expected. Interannual variation in isotopic
niche has been shown to occur concurrently in pygoscelid penguins
in other areas (Negrete et al., 2017), and it was thus expected that
potential changes in isotopic niche would be reflected similarly in all
three species.
2 | MATERIALS AND METHODS
2.1 | Study site and sample collection
This study focussed on Adélie, chinstrap, and gentoo penguins at
breeding colonies on Powell Island (60.73°S, 45.02°W), in the South
Orkney Islands (Figure 1), during the breeding seasons of 2013–
2014 and 2015–2016 (hereafter 2014 and 2016, respectively). The
three species were sampled during each season (Table S1). In the
early 1980s, the overall population sizes for Powell Island and the
adjacent islets were estimated to be ca. 16,750 Adélie, 28,100 chin-
strap, and 8,000 gentoo penguins; more recent estimations are not
available for this site (Harris et al., 2015; Poncet & Poncet, 1985).
2.2 | Sampling for isotopic analyses
Blood sampling of adult penguins took place between (earliest) 22
December and (latest) 9 February in each field season upon their
returns from foraging trips. Approximately 1.5 ml of whole blood
was collected from the brachial vein into a heparinized tube during
each field season, with samples collected during 2016 being cen-
trifuged at ca. 6,700 g during 10 minutes to separate out plasma
and red blood cells (RBC). Some plasma samples in 2016 were too
small to be processed and analyzed for stable isotopes (Table S1).
Each blood component was then stored in 95% ethanol in a separate
sterile tube until later analysis. During 2014, it was not possible to
centrifuge blood; thus, the entire (whole blood) sample was stored in
the same manner. Ideally, different tissues should not be compared
directly. However, because whole blood is highly enriched in RBC,
one can safely assume that both whole blood and RBC yield similar
dietary information through carbon and nitrogen isotopic analyses
(Hobson, Schell, Renouf, & Noseworthy, 1996). Therefore, whole
blood and RBC samples were pooled into one single group (blood) in
all analyses and figures. In the African penguin Spheniscus demersus,
the half- life of the 15N isotope was estimated to range from 7.6 days
in plasma to 14.3 days in red blood cells (Barquete, Strauss, & Ryan,
2013). Blood and plasma isotopic ratios do integrate diet ary infor-
mation over partially overlapping time windows, but herein, it is
thus assumed that the measured isotopic ratios integrated dietary
information principally over a period of 1–2 weeks for plasma and
2–4 weeks for blood. In order to account for potential confounding
factors, the occurrence of an intraseasonal trend in isotopic ratios
was examined in the two species for which the temporal coverage of
the sampling was long enough within one breeding season to allow
testing (chinstrap and gentoo penguins). Using simple linear regres-
sions, only slight temporal trends in isotopic ratios were detected
(all absolute trends <0.4‰/month; see details in Figures S1 & S2).
Therefore, stationarity of isotopic ratios was assumed throughout
each breeding period in all isotopic niche analyses, and the results
presented here were assumed to be representative of the average
isotopic ratios in the entire month of Januar y. Down and contour
feathers (hereafter feathers) were collected from chicks in early
Februar y, during both seasons, except for chinstrap penguins, which
were sampled only in 2014 (Figure S3). Being natur ally built sequen-
tially, down and feather tissues integrate dietary information dur-
ing the early and late stages of chick growth, respectively (Browne
et al., 2011). Both down and feathers were collected simultane-
ously on each individual with a certain amount of overlap in isotopic
FIGURE1 The South Orkney Islands host large breeding
populations of Adélie, chinstrap, and gentoo penguins (Pygoscelis
adeliae, P. antarctica, and P. papua, respectively). The study was
conducted during the austral summers of 2013–2014 and 2015–
2016 on Powell Island, where adults and chicks from each species
were sampled for isotopic analyses. Continent (Scambos, Haran,
Fahnestock, Painter, & Bohlander, 2007) and bathymetric (Dickens
et al., 2014) data and are shown only for descriptive purposes

|
3663
TARROUX eT Al .
ratios of late- grown down and early- grown feathers being expected.
Subsequent analyses therefore focused only on feather isotopic ra-
tios , al th ough data bas ed on both tissue s ar e presented for com pa ra-
tive purposes.
2.3 | Sample preparation
In the laborator y, all blood and plasma samples were frozen at
−80°C for 24 hr before being freeze- dried for 48 hr, while down
and feather samples were kept dry. Feather samples were washed
in an ultrasound bath for 20 min before fur ther processing, to re-
move dust and other particles. Samples were then powdered using a
ball- mill grinder (blood/plasma) or clipped with fine scissors (down/
feather). Some samples were treated to remove lipids (see Section
2.4 below). A small aliquot (target weight 0.4 mg) of each sample
was encapsulated into a tin shell before being combusted using a
Flash EA 1112 elemental analyzer (Thermo Scientific, Milan, Italy)
coupled to a Delta- V Advantage isotope ratio mass spectrometer via
a ConFlo IV interface (Thermo Fisher Scientific, Bremen, Germany).
Stable isotope ratios of carbon (δ13C) and nitrogen (δ15N) are ex-
pressed as ‰ of the deviation from isotopic ratios of international
standards (Hobson, Piatt, & Pitocchelli, 1994). Acetanilide (Thermo
Scientific) and peptone (Sigma- Aldrich) were used as internal stand-
ards and calibrated based on international standards supplied by the
International Atomic Energy Agency (IAEA, Vienna, Austria). All mass
spectrometr y analyses were run at the laboratory of the Lit toral
Environment and Societies (LIENS) research group at University of La
Rochelle, France. The overall measurement precision was evaluated
by duplicating a random subset of samples (Jardine & Cunjak, 2005).
The mean absolute difference between duplicates was 0.11‰ (95%
CI = [0.09; 0.13], n = 102) and 0.10‰ (95% CI = [0.09; 0.12], n = 72),
respectively, for δ13C and δ15N, both measures being well within the
analytical precision measures provided by the laboratory (<0.15‰
for both isotopes).
2.4 | Lipid correction
Lipids in tissues can bias δ13C values and dietar y interpretation
(Logan et al., 2008; Tarroux et al., 2010); high lipid content in animal
tissue alters the mass ratio of carbon over nitrogen (C:N ratio), with
ratios >4.0 typically indicating significant amounts of lipids (Post
et al., 2007). In order to remove surface lipids, down and feather
samples were washed using 2:1 chloroform–methanol as solvent and
then rinsed in methanol following the method of Jaeger et al. (2013).
To develop li pi d correc ti on metho ds suited to th is stu dy sys te m, nor-
malization equations were fitted based on a subset of plasma sam-
ples for which δ13C was measured before and after chemical lipid
removal (Wilson, Chanton, Balmer, & Nowacek, 2014). First, lipids
were chemically extracted from 46 samples through two successive
rinses with 2:1 chloroform–methanol as solvent. Then, normaliza-
tion equations were estimated by regressing the difference in δ13C
between lipid- extracted and bulk plasma samples on the C:N ratio
of the bulk samples, using nonlinear least square regression (Ehrich
et al., 2010). All δ13C values of plasma samples were thus corrected
(Table S2) using the normalization equation that best fitted the data
(Table S3). All δ15N values were left uncorrected as δ15N is not af-
fected by lipid content (Yurkowski, Hussey, Semeniuk, Ferguson, &
Fisk, 2015). The C:N ratios of whole blood samples were all <3.6,
confirming that lipid normalization was not necessary (Table S3).
2.5 | Statistical analyses
The isotopic data used in the analyses are available from the
Norwegian Polar Institute’s data repository https://doi.org/10.21334/
npolar.2018.5aadb005. All data were processed and analyzed in R
3.2. 5 (R Development Core Team 2017). Th e normali zation eq uations
for carbon isotopic dat a were deter mined using the funct ion nls from
package stat s. Average isotopic ratios were compared amon g species
by means of ANOVAs using function aov from the package st ats. The
analyses related to the isotopic data and niche computations were
conducted using the script from Turner, Collyer, and Krabbenhoft
(2010) and the package nicheROVER (Swanson et al., 2015).
For a given year and age class, the relative location of each species
within the two- dimensional isotopic space was compared by comput-
ing the Euclidean distance (DIST) among centroids. Additionally, the
mean distance to ce nt ro id (MDC), an in de x of tro ph ic diversity wit hin
a given group (i.e., dispersion), was computed and compared among
species (within year and age class), among years (within species and
age class), and among age classes (within species and year; Layman,
Ar ri ng ton, Mont an a, & Pos t, 20 07 ). All contr as ts were tested st at is ti-
cally against the null hypothesis that dif ference in DIST or MDC was
equal to zero (e.g., for DIST, testing that two species’ centroids are
in the same isotopic area), through residual permutation procedures
(RPP; Turner et al., 2010), using n = 9,999 permutations.
In order to compare the isotopic niche among years and species,
data were plotted using isotopic biplots and niche region (Nr) com-
puted for each year. To calculate credible intervals around the param-
eter estimates, 10,000 elliptical projections (random ellipses) of Nr
were drawn randomly from the posterior distributions. For a given
group of individuals, Nr corresponds to the portion of a multidimen-
sional isotopic space (two- dimensional in this study) where the prob-
ability of finding any individual from that group is equal to a given,
user- defined threshold (Swanson et al., 2015). For each year, 95%
was used as the threshold defining the global isotopic niche, and the
area of the two- dimensional 95% Nr (Nrarea) was used as a measure of
the trophic niche width. The overlap between the isotopic niches of
two species is defined as the probability of an individual drawn ran-
dom ly from a given spe cies being found in the Nr of the othe r species.
The niche overlap is therefore asymmetrical; overlap between spe-
cies A and B is not directly equivalent to overlap between species B
and A, depending on how evenly each group uses it s own niche area
(Swanson et al., 2015). Tissue- specific discrimination fac tors have not
yet been determined in Adélie or chinstrap penguins, and only feath-
ers have been investigated in gentoo penguins (Polito, Abel, Tobias, &
Emslie, 2011). Herein, the direct comparison of the isotopic niches of
th e th ree spec ie s relie s on the as su mpt io n tha t th e di et–ti ssu e is ot opi c

3664
|
TA RROUX eT A l.
discrimination factors are similar for all three species. While this as-
sumption is currently unverifiable, results from a study on different
pengu in spec ies sug ge st that it is re asona bl e, when stu dy- s pe cific dis-
crimination factors cannot be determined, to use an average value for
wild fish- eating birds (Cherel, Hobson, & Hassani, 2005).
Complementary to the niche overlap estimation, the amplitude
and direction of temporal isotopic shifts from 2014 to 2016, rep-
resented as two- dimensional vectors in the isotopic space, were
compared statistically among species and age class, again using RPP
(Turner et al., 2010), with n = 9,999 permutations.
3 | RESULTS
3.1 | Isotopic niche width and interindividual
variation in isotopic ratios
In a dult s, the 95 % Nr are a (her eaf te r Nrarea) ra nge d from 0 .9 to 2. 3 ‰2
and from 1.8 to 2.7‰2, for blood and plasma, respec tively (Table 1,
Figure 2). Plasma samples were not collected in 2014, and results
from 2016 are thus presented for comparative purposes only, but
are not discussed further. For blood, the Nrarea of gentoo penguins
in 2014 was smallest, while that of chinstrap penguins in 2014 was
largest (Table 1). There was little variation in isotopic ratios within
individual years, species, or tissues, although for gentoo penguins,
the variation was three times as high for δ15N (SD = 0.6‰) compared
to δ13C (SD = 0.2‰) in 2016. For chinstrap penguins, on the other
hand, variation along the carbon axis was higher, especially in 2014
(SD = 0.6‰). In adults, MDC was generally small (≤0.52; Table 2)
and did not vary significantly among species, except in 2014 when
chinstrap penguins had higher MDCs than Adélie penguins (differ-
ence = 0.20 ‰, p = .030; Table 2 & 3) and in 2016 when gentoo pen-
guins had higher MDC than Adélie penguins (difference = 0.17 ‰,
p = .043; Tables 2 & 3). There was no interannual variation in MDC
detected in adults of any of the species studied (all p- values <.001).
In chicks, there were also only slight interspecific differences
in isotopic niche width based on either down or feathers (Table 1,
Figure 3). For both tissues (feather and down), Nrarea values were gen-
erally smaller in chicks than in adults (blood and plasma), resulting in
more contracted isotopic niches. However, gentoo penguins had the
widest isotopic niches. This was particularly accentuated in 2016 due
to one indi vid ua l tha t wa s cle arl y dif fer ent fro m oth er indi vid ual s, with
notably higher δ13C and δ15N values (Figure 3). Both the down and
feather samples with the highest values corresponded to the same
individual and showed a similar dif ference from the rest of the group,
indicating that this was probably not due to an analytical art ifa ct . This
was also confirmed by running a duplicate analysis on the downsam-
ple. In gentoo chicks, MDC was over twice as large as in Adélie and
chinstrap penguins (Tables 2 & 4). For comparative purposes, when
excluding that individual from the analyses, the Nrarea was up to thre e
times narrower. In chicks’ feather, the Nrarea then decreased from
4.4‰ (95% credible interval = [2.3; 8.1]) to 1.3‰ (95% credible inter-
val = [0.7; 2.5]). In chicks’ down, when removing that individual from
the calculations, the Nrarea decreased from 4.9‰ (95% credible inter-
val = [2.7; 8.8]) to 2.3‰ (95% credible interval = [1.2; 4.5]).
3.2 | Isotopic niche overlap among species
Overall, chinstrap penguins were most unique, being situated further
apart (i.e., DIST values among species significantly different from
zero) from the two other species in the isotopic space in both years;
twice as much in 2016 (Figure 2, Table 3). The mean overlap among Nr
of adults was large with an average of 48% over both years (Figure 2,
Table 5). The isotopic niche of adult gentoo penguins generally had
the highest overlap with those of Adélie or chins trap penguins during
both years, ranging from 46% to 8 4% (Table 5). Contrastingly, adult
chinstrap penguins had the lowest overlap with the two other species
also during both years, ranging from 12% to 44% (Table 5).
In chicks, the average δ13C in feathers was significantly differ-
ent among Adélie and gentoo penguins both in 2014 (ANOVA; F2,
68 = 105.00, p < .001) and 2016 (ANOVA; F1, 17 = 42.38, p < .001).
Gentoo penguins had the highest δ13C and δ15N (Figure 3) in both
years. Specifically, in 2014, the average δ13C of gentoo penguins
Adults Chicks
Blood Plasma Down Feather
2016
Adélie 1.2a [0.7; 2.1] 1.8b [1.0; 3.2] 0.9c [0.5; 1.7] 0.9e [0.5; 1.6]
Chinstrap 1.8a [1.3; 2.4] 2.3b [1.7; 3.2] – –
Gentoo 1.4a [0.9; 2.0] 2.7b [1.6; 4.5] 4.9d [2.7; 8.8] 4.4f [2.3; 8.1]
2014
Adélie 1.2a [0.7; 2.0] –1.1c [0.7; 1.6] 0.8e [0.5; 1.2]
Chinstrap 2.3a [1.4; 3.8] –0.6c [0.4; 0.9] 0.6e [0.4; 0.8]
Gentoo 0.9a [0.5; 1.4] –2.0c,d [1.3; 3.0] 3.3f [2.2; 4.9]
“Blood” stands for “whole blood” or “red blood cells” (see Section 2 for details). “Feather” stands for
“contour feather”. Superscript let ters are identical among Nr areas that are not statistically different
from each other, within each tissue (i.e., when their 95% credible inter vals intersect). Nr areas are in
‰2 and were es timated based on 95% random ellipses (see Section 2 for details).
TABLE1 Mean area (95% credible
interval) of the isotopic niche region (Nr)
per breeding season, species, age class,
and tissue in pygoscelid penguins from
Powell Island, South Orkney Islands

|
3665
TARROUX eT Al .
was +1.1‰ (95% CI = [0.9; 1.3]) higher than that of Adélie penguins
and +0.9‰ (95% CI = [0.7; 1.1]) higher than that of chinstrap pen-
guins (Figure 3). In 2016, gentoo penguins’ average δ13C was 1.4‰
(95% CI = [0.9; 1.9]) higher than that of Adélie penguins. Similar
differences were found for δ15N. In 2014, the average δ15N of gen-
too penguins was +1.5‰ (95% CI = [1.2; 1.8]) higher than that of
Adélie penguins and +1.2‰ (95% CI = [0.9; 1.5]) higher than that of
chinstrap penguins (Figure 3). In 2016, the average δ15N of gentoo
penguins was +1.4‰ (95% CI = [0.7; 2.1]) higher than that of Adélie
penguins. Comparable differences were detected when exam-
ining isotopic ratios from down tissue, though of lower amplitude
(Figure 3). Overall, chicks from all three species occupied different
isotopic spaces in both years, with gentoo penguin’s chicks being
situated fur thest apart (Figure 3, Table 4). As a result, the Nr of
chicks generally showed less overlap (range: 0.8%–52.1%; Table 5)
among the three species than that of the adult s (Figure 2). The Nr
of gentoo penguin’s chicks overlapped those of Adélie or chinstrap
penguins by <5% (Table 5). However, the Nr of Adélie and chinstrap
penguins overlapped each other quite considerably, up to 52.1%
(Table 5).
3.3 | Interannual variation in isotopic niche
A decrease in δ13C occurred in all three species from 2014 to 2016,
both in adults and chicks (Figures 2 and 3), resulting in a shif t of
the Nr along the carbon axis ranging from 0.7‰ in adult chinstrap
FIGURE2 Interannual variation in
niche regions areas (Nrarea; represented
by 95% random ellipses) based on δ13C
and δ15N of blood and plasma from adult
Adélie, chinstrap, and gentoo penguins
from Powell Island, South Orkney Islands,
in 2014 (orange) and 2016 (black). “Blood”
stands for “whole blood” or “red blood
cells” (see Section 2 for details). Plasma
δ13C values are normalized to account for
lipid content (see Section 2 for det ails).
Empty circles and error bars show the
mean (±SD) isotopic ratios. Density curves
for each isotope are drawn marginally
along the corresponding axis
n =15
n = 16
n = 15
n = 12
n = 46
n = 22
n = 10
n = 33
n = 15

3666
|
TA RROUX eT A l.
penguins to 1.9‰ in adult Adélie penguins (Figure 4). This led to a
complete discrimination (i.e., 0% overlap) of the Nr from each year
in the isotopic space (Figures 2and 3), except for adult chinstrap
penguins where a more limited shift in δ13C generated substantially
overlapping Nr (mean overlap of Nr2016/2014 = 87.7%, 95% credible
interval = [64.7; 99.4]). In contrast, temporal shifts in δ15N were
≤0.3‰ for both adults and chicks (Figure 4). Overall, the amplitude
of the isotopic shift was significantly different from zero in adults
and chicks of all species (Figure 4; all p values <.001), but the ampli-
tude of the shift was more than twice as large for adult Adélie and
gentoo penguins compared to chinstrap adults (Figure 4, Table 6).
The direction of the shift in the isotopic space was similar among all
species and age classes (Figure 4, Table 6).
4 | DISCUSSION
Our study shows that closely related species breeding in sympatry
can have overlapping isotopic niches that can undergo similar vari-
ation through time, both in terms of amplitude and direction of the
isotopic shift. More specifically, two main findings emerged from this
study. Firstly, all three pygoscelid species had similar isotopic niche
region (Nr) during the breeding season. A moderate to strong over-
lap was measured in the isotopic niches, and thus assumed in the
trophic niches, of the three species, both in adults and chicks. This
indicates that the various pygoscelid species feed, at least partly, on
the same prey species in the waters around Powell Island. However,
the Nr of gentoo penguins was characterized by greater variation in
trophic levels with a variance in δ15N up to four times larger in adults.
This could be a consequence of a more diverse diet among individual
gentoo penguins compared to Adélie or chinstrap penguins and sup-
porting what has been found for that species at other study sites
(Camprasse, Cherel, Bustamante, Arnould, & Bost, 2017; Lescroel,
Ridoux, & Bost, 2004; Polito et al., 2015; Ratcliffe & Trathan, 2011);
this assertion is also borne out by diet samples collected at nearby
Signy Island (BAS unpublished data; Figure 1). Secondly, a clear
systemic shift in the isotopic niche of all three penguin species oc-
curred between 2014 and 2016, in both adults and chicks. This shift
was caused almost entirely by a decrease in δ13 C, while δ15N values
remained very similar in both years, coincidentally indicating that all
three penguin species maintained a remarkably stable trophic level
between these 2 years.
4.1 | Isotopic niche width and interindividual
variation in isotopic ratios
Irrespective of the species, δ15N measured in this study were gen-
erally moderately high, which is consistent with the contribution of
prey of higher trophic level to the diet (e.g., fish or squid; Negrete
et al., 2017) compared to that measured in other studies and areas.
For example, stomach content analysis on chinstrap penguins from
Bouvetøya described a diet composed of <1% fish during three non-
consecutive sampling years (Niemandt et al., 2016). Conversely, in the
South Shetland Islands, Polito et al. (2015) found that fish contributed
substantially to the diet of chinstrap and gentoo penguins alike, the
latter having a diet of up to 50% fish. These authors further described
that the δ15N values of both species were strongly and positively cor-
related to the estimated relative proportion of fish in the diet. It is
therefore likely that the δ15N values seen in the current study that
are in the high end of the range also correspond to higher input of
fish into the penguins’ diet than those in the low end of the range.
However, the absence of isotopic data on prey prevents a more pre-
cise estim at io n of the relative con tr ibution of fish vs krill, fo r ex ample,
using Bayesian mixing models (e.g., siar; Parnell & Jackson, 2010).
Gentoo penguins showed stronger individual variation in their
isotopic ratios than the two other species, with ranges in δ15N va r yin g
from 1.3‰ in the plasma of adults to 3.6‰ in chick feathers. A diet
based on larger and older krill could induce an increase in the δ15N
of consumers, because krill tend to increase their isotopic ratios at a
rate of 0.07‰/mm as they grow (Polito, Reiss, Trivelpiece, Patterson,
& Emslie, 2013). However, this would not suffice to explain all of the
observed variation in the current study. It would additionally require
that individuals with higher δ15N values had shifted their diet exclu-
sively to larger krill. Instead, the variation in δ15N values documented
in this study suggest s a population- level diet spanning several tro-
phic levels, that is, a more varied diet involving stronger reliance on
prey of higher trophic level, such as fish or squid. Coincidentally, this
shows that, while the estimation of isotopic Nr areas constitutes a
powe rfu l and infor mativ e to ol in tro phic ec ol ogy, this app roa ch might
be misleading when used in isolation from other approaches to as-
sessing trophic relationships, such as comparing intrapopulation vari-
an ce s in isotopic rat ios (Be arh op , Ad ams , Wald ro n, Fuller, & Mac leo d,
2004). During the period considered in the present study (i.e., for
whole blood: 2–4 weeks), individuals with high δ15N values were
feeding more consistently at higher trophic levels, presumably on
fish or squid (Miller et al., 2010). Fish and squid in the Scotia Sea and
near the Antarctic Peninsula are characterized by relatively high δ15N
values >8.0‰ (Negrete et al., 2017; Polito, Lynch et al., 2011; Polito,
Trivelpiece et al., 2011). In comparison, average δ15N values of krill
TABLE2 Summary of the interspecific differences in mean
distance to centroid (MDC, in ‰) between adult and chick
pygoscelid penguins from Powell Island, South Orkney Islands
Adults Chicks Empirical p- value
2014
Adélie 0.31 0.29 .794
Chinstrap 0.51 0.23 <.0 01
Gentoo 0.35 0.65 .003
2016
Adélie 0.35 0. 27 .222
Chinstrap 0.40 – –
Gentoo 0. 52 0.79 .340
Results based on isotopic ratios in blood (adults) and feather (chicks).
Empirical p- values estimated from permutations procedures (see
Section 2 for details) are in bold when significant at α = 0.05.

|
3667
TARROUX eT Al .
in the same region are typically <4.0‰ (review in Polito et al., 2013).
Specific preser vation methods used in this study can prevent a di-
rect, quantitative comparison of the absolute isotopic ratios to those
fro m oth er studies. Neverthel ess, th e larger var iance alon g the nitro-
gen axis for gentoo penguins shows that there is high heterogeneity
in the dieta r y ha bi ts of ge nt oo pengui ns at a po pu la ti on lev el . Gentoo
penguins in other regions have also been shown to display greater
foraging flexibility during the breeding season than closely related
species (Lescroel et al., 2004; Miller, Karnovsky, & Trivelpiece, 2009;
Polito et al., 2015; but see Juares et al., 2016; Negrete et al., 2017).
Individual variation in δ15N was also high in gentoo chicks, similar
to their adult conspecifics. When not considering the gentoo chick
that had higher δ15N than the others in 2016 (Figure 3), the Nrarea
and range in δ15N values still remained higher for gentoo chicks as
a group that year. This confirmed a generally more diversified iso-
topic niche for this species. In contrast, Adélie and chinstrap chicks
showed much lower interindividual variation. Remarkably, the isoto-
pic niche region of gentoo chicks also exhibited only marginal overlap
with those of the two other species and indic ated a diet at a higher
trophic level. This contrasted with the pattern observed in gentoo
adults. This could indicate stronger trophic segregation in gentoo
chicks and suggests that some gentoo adults might feed their chicks
with different prey than those they themselves feed on. Chick provi-
sioning with different prey than that eaten by adults has been docu-
me nt ed in oth er pe ng uin spec ie s and can in cr eas e ch ick s’ grow th ra te
when prey of higher quality are provided (Cherel, 2008). Individual
FIGURE 3 Interannual variation in
niche regions areas (Nrarea; represented
by 95% random ellipses) based on δ13C
and δ15N of feather and down from chick
Adélie, chinstrap, and gentoo penguins
from Powell Island, South Orkney Islands,
in 2014 (orange) and 2016 (black).
“Feather” stands for “contour feather”.
Empty circles and error bars show the
mean (± SD) isotopic ratios. Densit y
curves for each isotope are drawn
marginally along the corresponding axis.
The two arrows indicate the individual
points causing an increase in the Nrarea of
gentoo penguins in 2016
n = 22
n = 25
n = 24
n = 10
n = 10
n = 23
n = 25
n = 24
n = 10
n = 9

3668
|
TA RROUX eT A l.
specialization on particular prey types can occur in gentoo penguins
(Waluda, Hill, Peat, & Trathan, 2016), which is possibly a mechanism
that could hel p bu ffe r intr as pecif ic com pe tition. The res ul ts fr om thi s
study suggest that this mechanism might also apply to chick provi-
sioning; further investigation of this hypothesis is warranted.
4.2 | Isotopic niche overlap among species
This study adds to the growing literature supporting a potentially
high level of trophic overlap in adult pygoscelid penguins (Gorman,
2015; Juares et al., 2016; Miller et al., 2010; Trivelpiece et al., 1987).
This pattern was somehow moderated in chinstrap penguins, whose
isotopic niches showed lower overlap with the other species’ iso-
topic niches in both years. The reliance of all species on the same
trophic level, presumably predominantly on krill (Niemandt et al.,
2016; Ratcliffe & Trathan, 2011), was clear in the present study for
both years. At nearby breeding sites in the South Orkney Islands,
some studies have found that krill dominates the diet of Adélie and
chinstrap penguins, with estimated contributions generally over
90%, while fish seem to dominate in the diet of gentoo penguins
TABLE3 Mean Euclidean distance between species’ centroids
(DIST, in ‰; upper triangular matrices) and interspecific difference
in mean distance to centroid (MDC , in ‰; lower triangular matrices,
shaded) based on isotopic ratios in blood of adult pygoscelid
penguins from Powell Island, South Orkney Islands
Adélie Chinstrap Gentoo
2014
Adélie –0.46 (<0.001) 0.14 (0.506)
Chinstrap 0.20 (0.030) –0.36 (0.008)
Gentoo 0.0 4 (0.727) 0.17 (0.090) –
2016
Adélie –0.93 (<0.001) 0.24 (0.162)
Chinstrap 0.05 (0.513) –0.71 (<0.001)
Gentoo 0.17 (0.043) 0.11 (0.051) –
Empirical p- values estimated from permutations procedures are in pa-
rentheses (see Sec tion 2 for det ails) and in bold when signif icant at
α = 0.05.
Adélie Chinstrap Gentoo
Adults
2014
Adélie –24.3 [3.7; 58.9] 52.6 [29.8; 76.8]
Chinstrap 12.0 [2.2; 37.3] –18.8 [8.7; 32.5]
Gentoo 55.4 [34.8; 80.3] 46.0 [23.7; 68.0] -
2016
Adélie –64.0 [33.3; 92.8] 70.4 [46.3; 91.4]
Chinstrap 44.2 [20.9; 71.4] –31.9 [18.1; 50.0]
Gentoo 84.0 [61.3; 98.4] 72.5 [47.0; 94.3] –
Chicks
2014
Adélie –39.6 [19.0; 64.4] 5.5 [0.0; 50.4]
Chinstrap 52.1 [25.8; 81.0] –35.1 [0.5; 88.1]
Gentoo 0.8 [0.0; 3.5] 2.9 [0.4; 7.7] –
2016
Adélie – – 29.2 [0.4; 86.9]
Gentoo 4.7 [0.2; 15.5] – –
Overlap is expressed as the % probability of an individual from species A (rows) to be found in the Nr
of species B (columns; see Section 2 for details). Results are based on carbon and nitrogen isotopic
ratios measured in blood (adults, Figure 1) and feather (chicks, Figure 2).
TABLE5 Mean isotopic niche overlap
[95% credible inter val] in pygoscelid
penguins from Powell Island, South
Orkney Islands
TABLE4 Mean Euclidean distance between species’ centroids
(DIST, in ‰; upper triangular matrices and interspecific difference
in mean distance to centroid (MDC , in ‰; lower triangular matrices,
shaded) based on isotopic ratios in feather of chick pygoscelid
penguins from Powell Island, South Orkney Islands
Adélie Chinstrap Gentoo
2014
Adélie –0.36 (0.003) 1.8 8 (<0.001)
Chinstrap 0.06 (0.451) –1.51 (<0.001)
Gentoo 0.36 (<0.001) 0.42 (<0.0 01) –
2016
Adélie – – 2.00 (<0.001)
Chinstrap –– –
Gentoo 0.52 (0.036) – –
Empirical p- values estimated from permutations procedures are in pa-
rentheses (see Sec tion 2 for det ails) and in bold when signif icant at
α = 0.05.

|
3669
TARROUX eT Al .
(Lynnes, Reid, & Croxall, 200 4; Rombolá, Marschof f, & Coria, 20 06;
White & Conroy, 1975). In other regions, the pattern seems to be
similar for Adélie and chinstrap penguins (i.e., a diet almost entirely
composed of krill) but much more variable for gentoo penguins,
although fish seem to always contribute substantially to their diet
(Bengt son, Croll, & Goebel, 1993; Lescroel et al., 2004; Miller et al.,
2010; Polito et al., 2015).
The main difference between the isotopic niches of chinstrap
vs Adélie and gentoo penguins was the large variation in individual
δ13C in the former species. Variability in δ13C values in marine or-
ganisms can be associated with distance from shore and whether
the organism feeds in the pelagic (Cherel & Hobson, 2007; Hobson
et al., 1994; Kopp, Lefebvre, Cachera, Villanueva, & Ernande, 2015)
or benthic food webs. Chinstrap penguins appeared to use a wider
range of foraging habitats than the t wo other species, although this
was not directly reflected in their individual δ13C values. Indeed,
concurrent with this study on Powell Island, chinstrap penguins in-
strumented with GPS log gers showed a clear relationship between a
strong coastal downwelling signal during the 2016 season and their
movements, foraging up to 100 km farther offshore compared to
birds tracked in 2014 (A. D. Lowther, P. N. Trathan, A. Tarroux, C.
Lydersen, & K. M. Kovacs, in review). Krill are not passive organisms;
they can move against currents, as well as migrating vertically over
considerable depth ranges (Murphy et al., 1998; Tarling & Thorpe,
2014), while feeding upon diatoms which are passively transported.
Consequently, the variation in δ13C values detected in 2016 cou-
pled with relatively stable δ15N values might reflect some penguins
looking for krill which in turn were searching for diatoms that were
passively advected away from the shelf via coastally driven oceano-
graphic processes. Regardless of the mechanism driving greater δ13C
variability during 2016, the present study’s results show clearly that
pygoscelid penguins at Powell Island depended on similar trophic-
level prey during both years and that the isotopic shift was likely
due to a shift in the carbon sources at the base of the penguins’ food
chain, rather than a change in prey species.
The observed asymmetry in isotopic niche overlaps between
chinstrap penguins on the one hand and Adélie and gentoo pen-
guins on the other hand hints at behavioral mechanisms in chin-
strap penguins that could potentially mitigate their competition
with other pygoscelid species, when resources are limiting. Such
similarities in prey used by pygoscelid penguins that forage in dif-
ferent habitats have been observed in other areas such as the
South Shetland Islands (Kokubun, Takahashi, Mori, Watanabe,
& Shin, 2010), as well as in other congeneric penguin species
(Cherel, Hobson, Guinet, & Vanpe, 2007). Despite large intrapop-
ulation variation in the δ15N measured in adult gentoo penguins,
their isotopic niches overlapped substantially with those of both
Adélie and chinstrap penguins, especially in 2014. Gentoo pen-
guin individuals thus mostly foraged within the trophic niche of
their congeneric neighbors, while only a small proportion of indi-
viduals were feeding on different prey. Overall, such findings may
have important implications in term of conservation, given their
potential consequences on the respective population dynamics of
each species. Local gentoo penguin populations, being composed
of individuals that target different prey species, might be better
FIGURE4 Comparison of the shift s in mean isotopic ratios in
blood (whole blood or red blood cells of adults, continuous arrows)
and feather (contour feathers of chicks, dashed arrows) in pygoscelid
penguins between 2014 (orange circles) and 2016 (black circles) on
Powell Island. The shifts are represented as vectors in the two-
dimensional isotopic space, based on the data from Figures 2 and 3
TABLE6 Absolute differences in isotopic shift amplitude (in ‰, upper triangular matrix) and direction (in degrees, lower triangular
matrix, shaded) among species and age classes in pygoscelid penguins from Powell Island, South Orkney Islands
Adults Chicks
Adélie Chinstrap Gentoo Adélie Gentoo
Adults
Adélie –1.20 (<0.001) 0.24 (0.218) 0.13 (0.531) 0.45 (0.032)
Chinstrap 1.4 (0.879) –0.96 (<0.001) 1.07 (<0.001) 0.75 (<0.001)
Gentoo 4.6 (0.621) 3.2 (0.688) –0.10 (0.604) 0. 21 (0.281)
Chicks
Adélie 2.8 (0.776) 1.4 (0.870) 1.8 (0.845) –0.31 (0.135)
Gentoo 5.2 (0.605) 3.8 (0.662) 0.6 (0.952) 2.4 (0.8 08) –
The isotopic shift was measured between 2014 and 2016 (Figure 3).
Empirical p- values estimated from permutations procedures are in parentheses (see Section 2 for details) and in bold when significant at α = 0.05.

3670
|
TA RROUX eT A l.
able to adjust their foraging tactics to potential changes in prey
availability in the future compared to Adélie and chinstrap pen-
guins, despite an apparently limited ability to forage farther from
the colony in gentoo penguins (Wilson, 2010). There is also evi-
dence that interfere nce competition among pygoscelids occurs at
least to some degree during years of low prey abundance, induc-
ing for example spatial segregation of foraging areas and lower
reproductive success (Lynnes et al., 2002). Therefore, the abun-
dance level of prey generally available to pygoscelid species at a
given breeding site but also their relative ability to adjust their
diet or exclude each other from their foraging areas have direct
consequences on their short- term reproductive success and thus
also on longer- term population dynamics (Lynnes et al., 2004). At
a regional scale, populations of the various pygoscelid spe cie s are
experiencing differing trends (Trathan, Lynch, & Fraser, 2016);
Adélie penguin populations are generally increasing in most re-
gi ons of the Ant arc tic ap ar t from th e Pen ins ula wh ere po pul ati ons
have declined in recent years, but now are more stable (BirdLife
International 2016; Fountain et al., 2016); chinstrap penguins are
stab le or in de c lin e at ma ny locat ions (B ird Life I n tern atio nal 201 6);
while gentoo penguins are generally increasing, particularly in
the Peninsula region (BirdLife International 2016). Interspecific
differences in the ability to cope with changes in environmental
factors that are ultimately linked to the abundance of food re-
sources, suc h as se a ic e extent an d dur at io n (R om bo lá , Mar sc hoff,
& Coria, 2003; Trathan et al., 1996), could at least partly explain
such trends. The recent population trends at Powell Island are
not currently known (Harris et al., 2015; Poncet & Poncet, 1985);
however, over the past decades at the neighboring Signy Island,
the number of breeding pairs of Adélie and chinstrap penguins
has steadily decreased, while the number of gentoo penguins has
increased (Dunn et al., 2016). This contrasts with the global pop-
ulation trends for chinstrap and gentoo penguins and emphasizes
the need for more detailed local studies given the spatially het-
erogeneous response of individual populations (Hinke, Salwicka,
Trivelpiece, Watters, & Trivelpiece, 2007; Lynch, Naveen, Trathan,
& Fagan, 2012).
4.3 | Interannual variation in isotopic niche
A clear alteration of the isotopic niche occurred bet ween 2014
and 2016 in all three penguin species for both adults and chicks.
Although of lesser amplitude in chinstrap penguins, this isotopic
change was reflected similarly in adults and chicks of all species alike,
based almost entirely on a negative shift in δ13C, while the δ15N val-
ues remained stable wit hin all species. Several explanations are pos-
sible for these results. Firstly, all three species may have acquired
resources from a different food web in 2016, for example feeding
in more pelagic waters and on different species. However, the fact
that δ15N values remained virtually unchanged makes the possibil-
ity of a clear shift in prey species unlikely (Juares et al., 2016), un-
less the new prey was at a ver y similar trophic level to those eaten
in the first period. An alternate explanation is the observed shift
could be the result of an environmental change between the 2 years,
which might have affected the base levels of the food web, that is,
the isotopic signature of either phytoplankton or particulate organic
matter. However, it is not possible to confirm this latter hypothesis
owing to the lack of complementary isotopic data from lower trophic
levels. Independent from the origin of the change (shift in prey or
change in the baseline isotopic levels), the results of this study show
that all species reacted to this change similarly, although the change
was weaker in chinstrap penguins. Indeed, the isotopic niche of chin-
strap penguins in 2016 was nearly completely confined within that
of 2014. This demonstrates that, in 2016, chinstrap penguins exclu-
sively exploited a lesser part of the isotopic niche that they were
using in 2014, simply contracting their isotopic niche.
4.4 | Conclusion and limitations
Competition for food resources among pygoscelid penguins is ex-
pected to be par ticularly strong owing to their phylogenic and eco-
logical proximity (Wilson, 2010). Using stable isotope analyses, this
study showed th at all three pygoscelid species had partially overlap-
ping isotopic niches, which could be interpreted as a likely overlap
in their realized trophic niches. These results support findings from
previous studies showing that a high degree of reliance on the same
prey species may be buffered by fine- scale behavioral adjustments
leading to the partitioning among pygoscelid penguins of their avail-
able foraging habitat (Cimino et al., 2016; Wilson, 2010). Such ad-
justments in foraging behavior, in combination with subtle variation
in prey selection among the three species (this study; Polito et al.,
2015) and distinct breeding phenologies (Ancel et al., 2013; Black,
2016), appear to be sufficient to allow the co- occurrence of all three
species breeding in sympatr y and in relatively high numbers in the
South Orkney Islands area. It is important to note that owing to the
lack of data on resource availability it was not possible to assess the
actual degree of competition among the three species in the cur-
rent study. However, isotopic niche overlap, as a proxy of the trophic
niche overlap, informs us about the potential for competition (Hinke
et al., 2015). Importantly, our results and interpretations rely on the
assumption that there are no physiological differences among in-
dividuals and species that could bias the measurements of isotopic
ratios.
As stated plainly by Boersma (20 08), “Life is not likely to get
easier for penguins”: future environmental changes affecting the
Southern Ocean’s food web have the potential to disrupt the deli-
cate trophic equilibrium among these species, for instance through
changes in abundance of their main prey, krill (Flores et al., 2012;
Lynnes et al., 200 4; Melbourne- Thomas et al., 2016). If krill abun-
dance was to decline drastically in the near future, the ecologic al
similarity among pygoscelids could lead to high levels of competition
for food resources (Miller et al., 2010), with uncertain outcomes.
There are data suggesting that, in such a scenario, owing to their
greater ecological flexibility, gentoo penguin populations may cope
better than their congeners in the Antarctic Peninsula area (Carlini
et al., 2009; Levy et al., 2016; Lima & Estay, 2013; Lynch, Naveen

|
3671
TARROUX eT Al .
et al., 2012; Trivelpiece et al., 1987). Understanding how congeneric
species breeding in sympatry can adapt to such changes is achiev-
able through individual- based studies of their respective isotopic
niches that also integrate fluctuations of their isotopic environment
and the dynamics of their foraging patterns at fine spatiotemporal
scales and ultimately determining the consequences on their repro-
ductive success and survival. The interpretation of the result s from
the current study is limited by the absence of isotopic data on prey.
It relies solely on the interpretation of differences in isotopic ratios
among consumers. Furthermore, the resolution provided by a t wo-
dimensional isotopic space might not be sufficient to detect changes
or differences in small amplitude in the isotopic niches. Using a third
isotope such as sulfur (34S/32S) might provide valuable complemen-
tary information (Bradshaw et al., 2017; Rubenstein & Hobson,
2004). Finally, complementary techniques of diet reconstruction,
such as stomach content analysis (Polito et al., 2011c), should ideally
be performed to confirm the trends suggested herein and to allow
for the interpretation of any potential subtle changes in diet.
ACKNOWLEDGMENTS
This work is par t of a bilateral Norwegian–United Kingdom re-
search program funded by the Norwegian Research Council
(Project No 222798/E10—Krill- dominated Ecosystem Dynamics
in the Scotia Sea); additional funding was provided by the
Norwegian Polar Institute and the British Ant arctic Survey
(Natural Environment Research Council, UK). Fieldwork and logis-
tical support were provided by the RRS James Clark Ross and sup-
ply boats to the Norwegian fishery operating in the Powell Island
area. All capture and handling procedures were carried out in ac-
cordance with the requirements of both the Norwegian Animal
Research Authorit y and the British Ant arctic Sur vey Ani mal Ethics
Committee. Permits for animal handling were provided by the UK
Government. We thank all of the people who helped with sam-
ple collection at Powell Island in the two study years (Andrew
Lowther, Iain Staniland, and Catrin Thomas). We also thank our
colleagues at the Marine Research Institute in Bergen (Drs Olav
Rune Godø, Bjørn K raf ft, an d Georg S karet) for support of various
types, Dr Andrew Lowther for comments on early drafts of the
manuscript, and Dr Heidi Ahonen (NPI) for helping with sample
processing and encapsulations, as well as Drs Gaël Guillou and
Benoit Lebreton (LI ENSs) for con du ct in g the isot op ic ana ly se s. We
thank two anonymous reviewers whose constructive comments
helped improve this manuscript.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTIONS
All authors participated in the study design. KK and PT contributed
to the field planning and logistics and to the laboratory analyses. PT
was part of the field team on Powell in 2014. AT prepared the tis-
sue samples in the laboratory, formatted the raw isotopic data, ran
the statistical analyses, produced the figures and wrote the manu-
script. All authors commented upon the manuscript in multiple draft
rounds, thus contributing critically to the development and produc-
tion of the manuscript.
ORCID
Arnaud Tarroux http://orcid.org/0000-0001-8306-6694
REFERENCES
Adler, P. B., HilleRisLambers, J., & Levine, J. M. (2007). A niche
for neutrality. Ecology Letters, 10, 95–104. https://doi.
org /10.1111/j.1461-0 248 .2 00 6. 00 996 .x
Ancel , A. , Be aul ieu, M., & Gi lber t, C. (2013). The dif fe rent breedin g strat-
egies of penguins: A review. Comptes Rendus Biologies, 336, 1–12.
https://doi.org/10.1016/j.crvi.2013.02.002
Anguit a, C., & Simeone, A. (2016). The shifting roles of intrinsic traits
in determining seasonal feeding flock composition in seabirds.
Behavioral Ecology, 27, 501–511. https://doi.org/10.1093/beheco/
arv180
Barger, C. P., Young, R. C., Will, A., Ito, M., & Kitaysky, A. S. (2016).
Resource partitioning between sympatric seabird species in-
creases during chick- rearing. Ecosphere, 7, e01447. https://doi.
org /10.10 02 /ecs 2.14 47
Barquete, V., Strauss, V., & Ryan, P. G. (2013). Stable isotope turnover in
blood and claws: A case study in captive African Penguins. Journal of
Experimental Marine Biology and Ecology, 448, 121–127. https://doi.
org/10.1016/j.jembe.2013.06.021
Bearhop, S., Adams, C. E., Waldron, S., Fuller, R. A., & Macleod, H. (200 4).
Determining trophic niche width: A novel approach using stable iso-
tope analysis. Journal of Animal Ecology, 73, 1007–1012. ht tps://doi.
org/10.1111/j.0021-8790.2004.00861.x
Bengt son, J. L., Croll, D. A., & Goebel, M. E. (1993). Diving behavior of
chinstrap penguins at seal- island. Antarctic Science, 5, 9–15.
BirdLife International 2016. The IUCN Red List of Threatened Species
2016. http://www.iucnredlist.org. Accessed on 29 March 2017
Black, C. E. (2016). A comprehensive review of the phenolog y of Pygoscelis
penguins. Polar Biology, 39, 405–4 32. ht tps://doi.org/10.1007/
s00 300-015-1807-8
Boersma, P. D. (2008). Penguins as marine sentinels. BioScience, 58, 5 97–
607. https://doi.o rg/10.1641/b580707
Bradshaw, P. J., Broderick, A. C., Carreras, C., Inger, R., Fuller, W., Snape,
R., … Godley, B. J. (2017). Satellite tracking and stable isotope anal-
ysis highlight differential recruitment among foraging areas in green
turtles. Marine Ecology Progress Series, 582, 201–214. https://doi.
org/10.3354/meps12297
Browne, T., Lalas, C., Mattern, T., & Van Heezik, Y. (2011). Chick star-
vation in yellow- eyed penguins: Evidence for poor diet quality
and selec tive provisioning of chicks from conventional diet anal-
ysis and stable isotopes. Austral Ecology, 36, 99–108. https://doi.
org /10.1111/j.1442-9993.2010 .02125 .x
Camprasse, E . C. M ., Cherel, Y., Bus tamante, P., Arnould, J. P. Y., & Bost,
C. A. (2017). Intra- and inter- individual variation in the foraging ecol-
ogy of a generalist subantarctic seabird, the gentoo penguin. Marine
Ecology Progress Series, 578, 227–242. https://doi.org/10.3354/
meps12151
Carlini, A. R., Coria, N. R., Santos, M. M ., Negrete, J., Juares, M. A., &
Daneri, G. A . (2009). Responses of Pygoscelis adeliae a nd P. pap ua p op -
ulations to environmental changes at Isla 25 de Mayo (King George

3672
|
TA RROUX eT A l.
Island). Polar Biology, 32, 1427–1433 . ht tps://doi.o rg /10.1007/
s00300-009-0637-y
CCAMLR . 2007. CCAMLR Ecosystem Mon itoring Programme: standard
methods - updated 2014. CCAMLR.
Cherel, Y. (2008). Isotopic niches of emperor and Adélie penguins in
Adélie Land, Antarctica. Marine Biology, 154, 813–821. ht tps://doi.
org/10.1007/s00227-008-0974-3
Cherel, Y., Connan, M., Jaeger, A., & Richard, P. (2014). Seabird year- round
and historical feeding ecology: Blood and feather δ13C and δ15N val-
ues document foraging plasticity of small sympatric petrels. Marine
Ecology Progress Series, 505, 267–280. https://doi.org/10.3354/
meps10795
Cherel, Y., & Hobson, K. A. (2007). Geographical variation in carbon sta-
ble isotope signatures of marine predator s: A tool to investigate their
foraging areas in the Southern Ocean. Marine Ecology Progress Series,
329, 281–287. https://doi.org/10.3354/meps329281
Cherel, Y., Hobson, K. A., Guinet, C., & Vanpe, C. (2007). Stable iso-
topes document seasonal changes in trophic niches and winter
foraging individual specialization in diving predators from the
Southern Ocean. Journal of Animal Ecology, 76, 826–836. https://doi.
org /10.1111/j.1365 -2656 .2 007.0123 8. x
Ch er el, Y., Ho bs on, K. A ., & Hass an i, S. (20 05). Iso to pic dis cr im inat ion be-
tween food and blood and feather s of captive penguins: implications
for dietar y studies in the wild. Physiological & Biochemical Zoology, 78,
106–115. ht tps://doi.o rg/10 .1086/425202
Cimino, M. A., Moline, M. A., Fraser, W. R., Patterson-Fraser, D. L., &
Oliver, M. J. (2016). Climate- driven sympatry may not lead to for-
aging competition between congeneric top- predators. Scientific
Reports, 6, 18820. https://doi.org/10.1038/srep18820
DeNiro, M. J., & Epstein, S. (1978). Influence of diet on the distribution
of carbon isotopes in animals. Geochimica et Cosmochimica Acta, 42,
495–506. https://doi.org/10.1016/0016-7037(78)90199-0
DeNiro, M. J., & Epstein, S . (1981). Influence of diet on the distribution
of nitrogen isotopes in animals. Geochimica et Cosmochimica Acta, 45,
341–351. https://doi.org/10.1016/0016-7037(81)90244-1
Dickens, W. A., Graham, A . G. C., Smith, J. A ., Dowdeswell, J. A ., Larter,
R. D., Hillenbrand, C.-D., … Kuhn, G. (2014). A new bathymetric
compilation for the South Orkney Islands region, Antarctic Peninsula
(49°–39°W to 64°–59°S): Insights into the glacial development of the
continental shelf. Geochemistry, Geophysics, Geosystems, 15, 2494–
2514. https://doi.org/10.10 02/2014GC005323
Dunn, M. J., Jackson, J. A., Adlard, S., Lynnes, A. S., Briggs, D. R., Fox, D.,
& Waluda, C. M . (2016). Population size and decadal trends of three
penguin species nesting at Signy Island, South Orkney Islands. PLoS
ONE, 11, e0164025. https://doi.org/10.1371/journal.pone.0164025
Ehrich, D., Tarroux, A., Stien, J., Lecomte, N ., Killengreen, S., Berteaux,
D., & Yoccoz, N. G. (2010). Stable isotope analysis: Modelling
lipid normalization for muscle and eggs from Arctic mammals
and birds. Methods in Ecology and Evolution, 2, 66–76. https://doi.
org /10.1111/j.2041-210X .2010 .0 00 47.x
Flores, H., Atkinson , A ., Kawaguchi, S., Krafft, B., Milinevsky, G., Nicol,
S., … Werner, T. (2012). Impact of climate change on Antarctic krill.
Marine Ecology Progress Series, 458, 1–19. https://doi.org/10.3354/
meps09831
Forcada, J., Trathan, P. N., Reid, K., Murphy, E. J., & Croxall, J. P. (2006).
Contrasting population changes in sympatric penguin species in as-
sociation with climate warming. Global Change Biology, 12, 411–423.
https://doi. org/10.1111/j .1365-248 6. 20 06 .01108.x
Fountain, A . G., Saba, G., Adams, B., Doran, P., Fraser, W., Gooseff, M.,
… Virginia, R. A. (2016). The impact of a large- scale climate event on
Antarctic ecosystem processes. BioScience, 66, 848–863. https://doi.
org /10.1093/bio sci/biw 110
Fritschie, K. J., Cardinale, B. J., Alexandrou, M. A., & Oakley, T. H. (2014).
Evolutionary histor y and the strength of species interac tions: Testing
the phylogenetic limiting similarity hypothesis. Ecology, 95, 140 7–
1417. http s://doi .or g/10 .1890/13- 098 6.1
Godoy, O., Kraft, N. J. B., & Levine, J. M. (2014). Phylogenetic relatedness
and the determinants of competitive outcomes. Ecology Letters, 17,
836–844. https://doi.org/10.1111/ele.12289
Gorman, K. B. (2015). Integrative studies of Southern Ocean food-webs and
pygoscelis penguin demography: Mechanisms of population response to
environmental change. PhD Dissertation. British Columbia, Canada:
Simon Fraser Univer sity.
Harris, C. M., Lorenz, K., Fishpool, L. D. C., Lascelles, B., Cooper, J., Coria,
N. R., … Woehler, E. J. (2015). Im portant bird areas in Antarctica 2015.
Cambridge: BirdLife International and Environmental Research &
Assessment Ltd..
Hinke, J. T., Polito, M. J., G oebel, M. E., Jarvis, S., Reiss, C. S., Thorrold,
S. R., … Watters, G. M. (2015). Spatial and isotopic niche partition-
ing during winter in chinstrap and Adélie penguins from the South
Shetland Islands. Ecosphere, 6, 1–32. https://doi.org/10.1890/
E S14 - 0 0 28 7.1
Hinke, J. T., Salwicka, K., Trivelpiece, S. G., Watters, G. M., & Trivelpiece,
W. Z. (2007). Divergent responses of Pygoscelis penguins reveal a
common environmental driver. Oecologia, 153, 8 45. https://doi.
org/10.10 07/s004 42-007-0781-4
Hobson, K. A., & Clark, R. G. (1993). Turnover of 13C in cellular and
plasma fractions of blood - implications for nondestructive sam-
pling in avian dietar y studies. Auk, 110, 63 8–641. https://doi.
org/10.2307/4088430
Hobson, K. A., Piat t, J. F., & Pitocchelli, J. (1994). Using stable isotopes to
determine seabird trophic relationships. Journal of Animal Ecology, 63,
786–798. https://doi.org/10.2307/5256
Hobson, K. A., Schell, D. M., Renouf, D., & Noseworthy, E. (1996). Stable
carbon and nitrogen isotopic fractionation between diet and tissues
of captive seals: Implications for dietary reconstruc tions involving
marine mammals. Canadian Journal of Fisheries and Aquatic Sciences,
53, 528–533. https://doi.org/10.1139/f95-209
Hutchinson, G. E. (1957). Population studies - animal ecology and demogra-
phy - concluding remarks. Paper presented to the Cold Spring Harbor
Symposia on Quantitative Biolog y.
Jackson, A . L ., Inger, R., Parnell, A. C., & Bearhop, S. (2010). Comparing
isotopic niche widths among and within communities: SIBER – Stable
Isotope Bayesian Ellipses in R. Journal of Animal Ecology, 595–602,
https://doi. org/10.1111/j .1365-2656.2011.018 06 .x
Jaeger, A., Jaquemet, S., Phillips, R., Wanless, R., Richard, P., & Cherel, Y.
(2013). Stable isotopes document inter- and intra- specific variation
in feeding ecology of nine large southern Procellariiformes. Marine
Ecology Progress Series, 490, 255–266. https://doi.org/10.3354/
meps10436
Jardine, T. D., & Cunjak, R. A. (2005). Analytical error in stable iso-
tope ecolog y. Oecologia, 144 , 528–533 . https://doi.org /10.1007/
s00442-005-0013-8
Juares, M. A., Santos, M., Mennucci, J. A., Coria, N. R., & Mariano-Jelicich,
R. (2016). Diet composition and foraging habitats of Adelie and gen-
too penguins in three different stages of their annual cycle. Marine
Biology, 163, 105. https://doi.org/10.1007/s00227-016-2886-y
Kelly, J. F. (2000). Stable isotopes of carbon and nitrogen in the study of
avian and mammalian trophic ecology [Review]. Canadian Journal of
Zoology, 78, 1–27. ht tps://doi.org/10.1139/z99-165
Kokubun, N., Takahashi, A., Mori, Y., Watanabe, S., & Shin, H.-C. (2010).
Comparison of diving behavior and foraging habitat use between
chinstrap and gentoo penguins breeding in the South Shetland
Islands, Antarctica. Marine Biology, 157, 811–825. https://doi.
org/10.1007/s00227-009-1364-1
Kopp, D., Lefebvre, S., Cachera, M., Villanueva, M. C., & Ernande, B.
(2015). Reorganization of a marine trophic network along an in-
shore–offshore gradient due to stronger pelagic–benthic coupling

|
3673
TARROUX eT Al .
in coast al areas. Progress in Oceanography, 130, 157–171. https://doi.
org/10.1016/j.pocean. 2014.11.001
Layman , C. A., A rring ton, D. A., Montana, C. G., & Post , D. M. (2007).
Can stable isotope ratios provide for communit y- wide mea-
sures of trophic structure? Ecology, 88, 42–48. https://doi.
org/10.1890/0012-9658(2007)88[42:CSIRPF]2.0.CO;2
Lescroel, A., Ridoux, V., & Bost, C. A. (2004). Spatial and temporal varia-
tion in the diet of the gentoo penguin (Pygoscelis papua) at Kerguelen
Islands. Polar Biology, 27, 20 6–216. htt ps://doi.org /10.10 07/
s00300-003-0571-3
Le vy, H. , Clu cas , G. V., Ro ger s, A. D., Le ach é, A . D. , Cib oro wsk i, K . L ., P oli to,
M. J., … Hart, T. (2016). Population structure and phylogeography of
the gentoo penguin (Pygoscelis papua) across the Scotia Arc. Ecolog y
and Evolution, 6, 1834–1853. https://doi.org/10.1002/ece3.1929
Lima, M., & Estay, S. A. (2013). Warming effects in the western Ant arctic
Peninsula ecosystem: The role of population dynamic models for ex-
plaining and predicting penguin trends. Population Ecology, 55, 557–
565. https://doi.org /10.1007/s10144-013-0386-1
Logan, J. M., Jardine, T. D., Miller, T. J., Bunn, S. E., Cunjak, R. A., &
Lutcava ge , M. E. (2 008). Lipi d co rr ec ti on s in car bon an d ni tr og en sta -
ble isotope analyses: Comparison of chemical extraction and mod-
elling methods. Journal of Animal Ecology, 77, 838–846. https://doi.
org /10.1111/j.1365 -2656 .2 00 8.01394. x
Lynch, H. J., Fagan, W. F., Naveen, R., Trivelpiece, S. G., & Trivelpiece,
W. Z. (2012). Differential advancement of breeding phenology in
response to climate may alter staggered breeding among sympatric
pygoscelid penguins. Marine Ecology Progress Series, 454, 135–145.
https://doi.org/10.3354/meps09252
Lynch, H. J., Naveen, R., Trathan, P. N., & Fagan, W. F. (2012). Spatially
integrated assessment reveals widespread changes in penguin pop-
ulations on the Antarctic Peninsula. Ecology, 93, 1367–1377. https://
doi.org/10.1890/11-1588.1
Lynnes, A. S., Reid, K., & Croxall, J. P. (2004). Diet and reproductive suc-
cess of Adélie and chinstrap penguins: Linking response of predator s
to prey population dynamics. Polar Biology, 27, 544–554. https://doi.
org/10.1007/s00300-004-0617-1
Lynnes, A. S., Reid, K., Croxall, J. P., & Trathan, P. N. (2002). Conflict or co-
existence? Foraging distribution an d competition for prey between
Adelie and chinstrap penguins. Marine Biology, 141, 1165–1174.
Melbourne-Thomas, J., Corney, S. P., Trebilco, R., Meiners, K. M., Stevens,
R. P., Kawaguchi, S., … Constable, A . J. (2016). Under ice habitats for
Antarctic krill larvae: Could less mean more under climate warm-
ing? Geophysical Research Letters, 43, 10322–10327. https://doi.
org /10.1002/2016GL 0708 46
Miller, A. K., Kappes, M. A., Trivelpiece, S. G ., & Trivelpiece, W. Z. (2010).
Foraging- niche separation of breeding gentoo and chinstrap pen-
guins, South Shetland Islands, Antarctica. Condor, 112, 683–695.
https://doi.org/10.1525/cond.2010.090221
Miller, A. K., Karnovsky, N. J., & Trivelpiece, W. Z. (2009). Flexible forag-
ing strategies of gentoo penguins Pygoscelis papua over 5 years in
the South Shetland Islands, Antarctica. Marine Biology, 156, 2 52 7–
2537. https://doi.org/10.1007/s00227-009-1277-z
Murphy, E. J., Watkins, J. L., Reid, K., Trathan, P. N., Everson, I., Croxall, J.
P., … Hofmann, E. (1998). Interannual variability of the South Georgia
marine ecosystem: Biologic al and physical sources of variation in the
abundance of krill. Fisheries Oceanography, 7, 381–390. https://doi.
org /10.1046/j .1365-2419.1998. 00 081.x
Negrete, P., Sallaberry, M., Barceló, G., Maldonado, K., Perona, F., McGill,
R. A. R., … Sabat, P. (2017). Temporal variation in isotopic composi-
tion of Pygoscelis penguins at A rdley Island, Antarctic: are for aging
habits impacted by environmental change? Polar Biology, 40, 903–
916. https://doi.org/10.1007/s00300-016-2017-8
Newsome, S. D., Martíne z del Rio, C., Bearhop, S., & Phillips, D. L. (2007).
A niche for isotopic ecology. Frontiers in Ecology and the Environment,
5, 429–436. https://doi.org/10.1890/060150.1
Niemandt, C., Kovacs, K. M., Lydersen, C., Dyer, B. M., Isak sen, K.,
Hofmeyr, G. J. G., … de Bruyn, P. J. N. (2016). Chinstrap and maca-
roni penguin diet and demography at Nyrøysa, Bouvetøya. Antarctic
Science, 28, 91–100. https://doi.org /10.1017/S09541020150 00504
Parnell, A ., Jackson , A . L . 2010. SIAR: Stable Isotope Analysis in R. ver-
sion R package version 4.0.2 ., accessed at http://cran.r-project.org/
web/packages/siar
Paterson, A. M ., Wallis, G. P., Kennedy, M., & Gray, R. D. (2014).
Behavioural evolution in penguins does not reflect phylogeny.
Cladistics, 30, 243–259. https://doi.org/10.1111/cla.12040
Po lito , M. , A bel, S ., Tobia s, C. , & Ems lie, S . (20 11) . Diet ar y isot opic d i scri m-
ination in gentoo penguin (Pygoscelis papua) feathers. Polar Biology,
34, 1057–1063. http s://doi.o rg/10.1007/s003 00 -011-0966 -5
Polito, M. J., Lynch , H. J., Naveen, R., & Emsli e, S. D. (2 011). Stable isotopes
reveal regional heterogeneity in the pre- breeding distribution and
diets of sympatric ally breeding Pygoscelis spp . pe ngu ins . Marine Ecology
Progress Series, 421, 265–277. https://doi.org/10.3354/meps08 863
Polito, M. J., Reiss, C. S., Trivelpiece, W. Z., Patterson, W. P., & Emslie, S.
D. (2013). Stable isotopes identify an ontogenetic niche expansion in
Antarctic krill (Euphausia superba) from the South Shetland Islands ,
Antarctica. Marine Biology, 16 0, 1311–132 3. ht tp s://doi. or g/10 .10 07/
s00227-013-2182-z
Polito, M. J., Trivelpiece, W. Z., Karnovsky, N. J., Ng, E., Patterson, W. P.,
& Emslie, S. D. (2011). Integr ating stomach content and stable iso-
tope analyses to quantify the diets of pygoscelid penguins. PLoS ONE,
6, e26642. https://doi.org/10.1371/journal.pone.0026642
Polito, M. J., Trivelpiece, W. Z., Patterson, W. P., Karnovsky, N. J., Reiss,
C. S., & Ems lie , S. D. (2015). Contr ast ing special ist and ge neralist pat-
terns facilitate foraging niche partitioning in sympatric populations
of Pygoscelis penguins. Marine Ecology Progress Series, 519, 221–237.
https://doi.org/10.3354/meps11095
Poncet, S., & Poncet, J. (1985). British Antarctic Survey Bulletin. Br itish
Antarctic Survey Bulletin, 68, 71–81.
Post, D. M., Layman, C. A., Arrington, D. A., Takimoto, G., Quat trochi, J.,
& Montana, C . G. (2007). Getting to the fat of the matter: Models,
methods and assumptions for dealing with lipids in stable iso-
tope analyses. Oecologia, 152, 179–189. ht tps://doi.org/10.1007/
s00442-006-0630-x
R Development Core Team. 2017. R: A language and environment for
statistical computing. version 3.4.0. Accessed at http://www.r-proj-
ect.org
Ratclif fe, N., & Trathan, P. (2011). A review of the diet and at- sea dis-
tribution of penguins breeding within the CAMLR Convention Area
(submitted to the 2008 Joint CCAMLR- IWC Workshop). CCAMLR
Science, 18, 75–114.
Robertson, G. S., Bolton, M., Grecian, W. J., Wilson, L . J., Davies, W.,
& Monaghan, P. (2014). Resource partitioning in three conge-
neric sympatrically breeding seabirds: Foraging areas and prey
utilization. The Auk, 131, 434–446. https://doi.org/10.1642/
AU K-1 3- 24 3 .1
Rombolá, E., Marschoff, E., & Coria, N. (2003). Comparative study of the
effec ts of the late pack- ice break- off on chinstrap and Adélie pen-
guins’ diet and reproductive success at Laurie Island, South Orkney
Islands, Antarctica. Polar Biology, 26, 41–48 . ht tps: //doi. or g/10.10 07/
s00300-002-0444-1
Rombolá, E., Marschoff, E., & Coria, N. (2006). Interannual study of
Chinstrap penguin’s diet and reproductive success at Laurie Island,
South Ork ney Islands, Antarc tica. Polar Biology, 29, 502–509. ht tps://
doi.org/10.1007/s00300-005-0081-6
Rubenstein, D. R ., & Hobson, K. A. (2004). From birds to butter-
flies: Animal movement patterns and stable isotopes. Tre nds
in Ecology & Evolution, 19, 256–263. https://doi.org/10.1016/j.
tree.2004.03.017
Scambos, T. A., Haran, T. M., Fahnestock, M. A., Painter, T. H., &
Bohlander, J. (2007). MODIS- based Mosaic of Antarctica (MOA)

3674
|
TA RROUX eT A l.
data sets: Continent- wide surface morphology and snow grain
size. Remote Sensing of Environment, 111, 242–257. https://doi.
org/10.1016/j.rse.2006.12.020
Silvertown, J. (2004). Plant coexistence and the niche. Tre nds in
Ecology & Evolution, 19, 605–611. https://doi.org/10.1016/j.
tree.2004.09.003
Sridhar, H., Srinivasan, U., Askins, R. A., Canales-Delgadillo, J. C., Chen,
C.-C., Ewert, D. N., … Shanker, K. (2012). Positive relationships be-
tween association strengt h and phenotypic similarity characterize
the assembly of mixed- species bird flock s worldwide. The American
Naturalist, 180, 777–790. https://doi.org/10.1086/668012
Swanson, H. K., Lysy, M., Power, M., Stasko, A. D., Johnson, J. D., &
Reist, J. D. (2015). A new probabilis tic method for quantifying
n- dimensional ecological niches and niche overlap. Ecolog y, 96,
318–324. https://doi.org/10.1890/14-0235.1
Ta rl in g, G . A . , & T h or pe , S. E . (2 01 4 ) . In st an ta n eo us m o ve me nt o f k r il l sw ar ms
in the Antarctic Circumpolar Current. Limnology and Oceanography, 59,
872–886. https://doi.org/10.4319/lo.2014.59.3.0872
Tarroux, A., Ehrich, D., Lecomte, N., Jardine, T. D., Bêt y, J., &
Berteaux, D. (2010). Sensitivity of stable isotope mixing models
to variation in isotopic ratios: Evaluating consequences of lipid ex-
traction. Methods in Ecology and Evolution, 1, 231–241. https://doi.
org /10.1111/j.2041-210X .2010 .0 00 33 .x
Trathan, P. N., Croxall, J. P., & Murphy, E. J. (1996). Dynamics of Antarctic
penguin populations in relation to inter- annual variability in sea ice distri-
bution. Polar Biology, 16 , 321–330. https://doi.org/10.1007/bf02342178
Trathan, P. N., García-Borboroglu, P., Boersma, D., Bost, C.-A., Crawford,
R. J. M., Crossin, G. T., … W ienecke, B. (2015). Pollution, habi-
tat loss, fishing , and climate change as critical threats to penguins.
Conservation Biology, 29, 31–41. https://doi.org /10.1111/cobi.12349
Trathan, P. N., Lynch, H. J., & Fraser, W. R. (2016) Changes in pen-
guin distribution over the Antarctic Peninsula and Scotia Arc.
Antarctic Environments Portal, online publication 1, https://www.
environments.aq/emerging-issues/changes-in-penguin-distribu-
tion-over-the-antarctic-peninsula-and-scotia-arc/
Trivelpiece, W. Z., Trivelpiece, S. G., & Volkman, N. J. (1987). Ecological
segregation of Adelie, Gentoo, and Chinstrap Penguins at King
George Island, Antarctica. Ecology, 68, 351–361. https://doi.
org /10. 2307/1939266
Trivelpiece, W., & Volkman, N. J. (1979). Nest- site competition bet ween
Adelie and Chinstr ap Penguins: An ecological interpretation. The
Auk, 96, 675–681.
Turner, T. F., Collyer, M. L., & K rabbenhoft , T. J. (2010). A general
hypothesis- testing framework for stable isotope ratios in ecological
studies. Ecolog y, 91, 2227–2233. https://doi.org/10.1890/09-1454.1
Vellend, M. (2010). Conceptual synthesis in community ecology.
Quarterly Review of Biology, 85, 183–206. https://doi.org/10.10 86/
652373
Violle, C., Nemergut, D. R., Pu, Z., & Jiang, L. (2011). Phylogenetic limit-
ing similarity and competitive exclusion. Ecology Letters, 14, 782–787.
https://doi. org/10.1111/j .1461-0248.2011.016 4 4.x
Waluda, C. M., Hill, S. L., Peat , H. J., & Trathan, P. N. (2016). Long term
variability in the diet and repro ductive perfor mance of penguins at
Bird Island, South Georgia. Marine Biology, 164, 39.
White, M. G., & Conroy, J. W. H. (1975). A spects of competition be-
tween pygoscelid penguins at Signy Island, S outh Orkney Islands.
Ibis, 117, 371–373. ht tp s://doi .org/10.1111/ j.1474-919X.1975.
tb0 4224 .x
Wilson, R. P. (2010). Resource par tition ing and niche hyper- volume over-
lap in free- living Pygoscelid penguins. Functional Ecology, 24, 646–
657. https://doi.org/10.1111/j.1365-2435.2009.01654.x
Wilson, R. M., Chanton, J. P., Balmer, B. C ., & Nowacek, D. P. (2014). An
evaluation of lipid extraction techniques for interpretation of car-
bon and nitrogen isotope values in bottlenose dolphin (Tursiops trun-
catus) skin tissue. Marine Mammal Science, 30, 85–103. https://doi.
org /10.1111/mms.12 018
Yeakel, J. D., Bhat, U., Elliott Smith, E. A., & Newsome, S. D. (2016).
Exploring the isotopic niche: Isotopic variance, physiological incor-
poration, and the temporal dynamics of foraging. Frontiers in Ecolog y
and Evolution, 4, 1. https://doi.org/10.3389/fevo.2016.0 00 01
Yurkowski, D. J., Hussey, N. E., Semeniuk, C., Ferguson, S. H., & Fisk,
A. T. (2015). Effects of lipid extraction and the utility of lipid nor-
malization models on δ13C and δ15N values in Arctic marine mam-
mal tissues. Polar Biology, 38, 131–143. http s://doi.org /10.1007/
s0 03 0 0 - 0 14 -1 57 1-1
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the
supporting information tab for this ar ticle.
How to cite this article: Tarroux A, Lydersen C, Trathan P,
Kovacs KM. Temporal variation in trophic relationships among
three congeneric penguin species breeding in sympatry. Ecol
Evol. 2018;8:3660–3674. https://doi.org/10.1002/ece3.3937