Content uploaded by David A S Rosen
Author content
All content in this area was uploaded by David A S Rosen on Jun 17, 2016
Content may be subject to copyright.
ARTICLE
Net energy gained by northern fur seals (Callorhinus ursinus) is
impacted more by diet quality than by diet diversity
Mariana Diaz Gomez, David A.S. Rosen, and Andrew W. Trites
Abstract: Understanding whether northern fur seals (Callorhinus ursinus (L., 1758)) are negatively affected by changes in prey
quality or diversity could provide insights into their on-going population decline in the central Bering Sea. We investigated how
six captive female fur seals assimilated energy from eight different diets consisting of four prey species (walleye pollock (Gadus
chalcogrammus Pallas, 1814, formerly Theragra chalcogrammus (Pallas, 1814)), Pacific herring (Clupea pallasii Valenciennes in Cuvier
and Valenciennes, 1847), capelin (Mallotus villosus (Müller, 1776)), and magister armhook squid (Berryteuthis magister (Berry, 1913)))
fed alone or in combination. Net energy was quantified by measuring fecal energy loss, urinary energy loss, and heat increment
of feeding. Digestible energy (95.9%–96.7%) was high (reflecting low fecal energy loss) and was negatively affected by ingested
mass and dietary protein content. Urinary energy loss (9.3%–26.7%) increased significantly for high-protein diets. Heat increment
of feeding (4.3%–12.4%) was significantly lower for high-lipid diets. Overall, net energy gain (57.9%–83.0%) was affected by lipid
content and varied significantly across diets. Mixed-species diets did not provide any energetic benefit over single-species diets.
Our study demonstrates that diet quality was more important in terms of energy gain than diet diversity. These findings suggest
that fur seals consuming low-quality prey in the Bering Sea would be more challenged to obtain sufficient energy to satisfy
energetic and metabolic demands, independent of high prey abundance.
Key words: northern fur seal, Callorhinus ursinus, net energy, mixed-species diets, diet quality.
Résumé : La détermination d’une éventuelle incidence négative de changements de la qualité ou de la diversité des proies sur
les otaries a
`fourrure (Callorhinus ursinus (L., 1758)) pourrait jeter un nouvel éclairage sur la baisse soutenue de leur population
dans le centre de la mer de Behring. Nous avons examiné comment six otaries a
`fourrure femelles en captivité assimilaient
l’énergie tirée de huit régimes alimentaires distincts composés de quatre espèces de proies (goberge de l’Alaska (Gadus chalco-
grammus Pallas, 1814, anciennement Theragra chalcogrammus (Pallas, 1814)), hareng du Pacifique (Clupea pallasii Valenciennes in
Cuvier et Valenciennes, 1847), capelan (Mallotus villosus (Müller, 1776)) et calmar rouge (Berryteuthis magister (Berry, 1913))) dispen-
sées seules ou en combinaison. L’énergie nette a été quantifiée en mesurant la perte énergétique fécale, la perte énergétique
urinaire et l’accroissement de la température dû a
`l’alimentation. L’énergie digestible (95,9 % – 96,7 %) était élevée (reflétant une
perte énergétique fécale faible) et négativement influencée par la masse ingérée et le contenu en protéines alimentaires. La perte
énergétique urinaire (9,3 % – 26,7 %) augmentait significativement pour les régimes a
`forte teneur en protéines. L’accroissement
de la température dû a
`l’alimentation (4,3 % – 12,4 %) était significativement plus faible pour les régimes a
`forte teneur en lipides.
Globalement, le gain d’énergie nette (57,9 % – 83,0 %) était influencé par la teneur en lipides et variait significativement selon le
régime alimentaire. Les régimes composés de plusieurs espèces ne conféraient aucun avantage énergétique par rapport aux
régimes composés d’une seule espèce. L’étude démontre que la qualité du régime alimentaire était plus importante en ce qui
concerne les gains d’énergie que sa variété. Ces constatations donnent a
`penser que les otaries a
`fourrure qui consomment des
proies de moins bonne qualité dans la mer de Behring auraient de la difficulté a
`obtenir assez d’énergie pour répondre a
`leurs
demandes énergétiques et métaboliques, quelle que l’abondance de ces proies. [Traduit par la Rédaction]
Mots-clés : otarie a
`fourrure, Callorhinus ursinus, énergie nette, régimes a
`plusieurs espèces, qualité du régime alimentaire.
Introduction
Energy is the currency of classical optimal foraging theory,
which postulates that foragers should maximize their energy gain
while minimizing the energetic costs of obtaining prey (Stephens
and Krebs 1986). Ecological theories of predator–prey relation-
ships also employ energy as a currency of profitability that should
be maximized (Barbosa and Castellanos 2005). Similarly, many
bioenergetic models assume that the gross energy contained
in prey translates directly to the energy gained by predators
(Grodzinski 1975). However, digestive processes can be energeti-
cally costly and can disproportionately distort the value of differ-
ent prey items. Bioenergetics recognize that net energy gain, i.e.,
the energy remaining after digestive processes have occurred, is
the true measure of energy available to fuel the predator’s physi-
ological demands (Lavigne et al. 1982). Energy transformation is a
dynamic process that depends on many factors, such as prey com-
position, the presence of enzymes, and the characteristics of the
consumer’s digestive system (Schneider and Flatt 1975). Hence,
the energy gained by a predator is a function of the predator’s
ability to search for and obtain food in a timely manner, as well as
their physiological capability to absorb digestive products.
Animals are constantly faced with the complex challenge of
regulating their energetic and nutritional intake in such a way
that their prey selection meets their optimal intake requirements,
while accounting for external (e.g., prey availability, environmental
Received 9 July 2015. Accepted 17 November 2015.
M. Diaz Gomez, D.A.S. Rosen, and A.W. Trites. Department of Zoology and Marine Mammal Research Unit, Institute for the Oceans and Fisheries,
Room 247, 2202 Main Mall, AERL, The University of British Columbia, Vancouver, BC V6T 1Z4, Canada.
Corresponding author: Mariana Diaz Gomez (email: m.diazgomez@oceans.ubc.ca).
123
Can. J. Zool. 94: 123–135 (2016) dx.doi.org/10.1139/cjz-2015-0143 Published at www.nrcresearchpress.com/cjz on 24 November 2015.
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
conditions, foraging costs) and internal (e.g., developmental
stage, nutritional state) factors (Raubenheimer and Simpson
1999). Pinnipeds are considered to be generalist predators, as re-
flected by the wide variety of prey species they typically consume
(Riedman 1990). However, the diversity of prey consumed may not
represent prey abundance and occurrences alone, but may also
reflect satisfying optimal intake requirements.
Diets composed of mixed-prey species that are nutritionally
complementary to each other have been theorized to enhance
both the kinetic production rate of digestive products and the
postdigestive utilization of nutrients (Penry and Jumars 1987;
Singer and Bernays 2003). For example, mixed-species diets report-
edly provide phocid seals with significantly greater levels of energy
intake than equivalent single-species diets (Goodman-Lowe et al.
1999;Trumble and Castellini 2005). Results from these and other
studies that fed mixed-species diets to a variety of predator species
are consistent with the theory that mixed-species diets provide the
consumer with significantly higher returns than single-species diets.
Understanding how changes in diet composition affect the en-
ergetic and nutritional budgets of animals is essential for making
accurate inferences about the impacts that changes in prey avail-
ability may have on populations. This is particularly true for de-
clining populations of threatened or endangered species that may
have difficulty obtaining sufficient energy and nutrients due to
reductions in the quantity (biomass), quality (energy density), or
diversity (numbers of species) of prey (Trites and Donnelly 2003;
Rosen 2009). The nutritional stress hypothesis has been suggested
as a mechanism to explain the recent population declines of many
species of seabirds and marine mammals inhabiting the central
Bering Sea and the Gulf of Alaska (Pitcher 1990;Trites and Larkin
1996;National Research Council 1996) as revealed by the relation-
ships between rates of declines and reductions in the quality and
diversity of prey available to them (Alverson 1992;Castellini 1993;
Decker et al. 1995;Merrick et al. 1997;Calkins et al. 1998;Rosen
and Trites 2000a;Trites et al. 2007).
Northern fur seals (Callorhinus ursinus (L., 1758)) inhabiting the
North Pacific Ocean and Bering Sea have declined dramatically in
the Eastern Pacific from 2.1 million in the late 1940s and early
1950s to !550 000 in 2014. At present, pup production is declining
!3.7% per year on the Pribilof Islands in the central Bering Sea
(Towell et al. 2014). Northern fur seals are known to change their
seasonal foraging behaviours in response to their changing ener-
getic needs, as well as the daily and seasonal movements of their
prey (Gentry and Kooyman 1986;Gentry 2002). Their main prey
species include juvenile walleye pollock (Gadus chalcogrammus
Pallas, 1814, formerly Theragra chalcogrammus (Pallas, 1814)), Atka
mackerel (Pleurogrammus monopterygius (Pallas, 1810)), capelin
(Mallotus villosus (Müller, 1776)), Pacific herring (Clupea pallasii
Valenciennes in Cuvier and Valenciennes, 1847), and various squid
species (e.g., family Gonatidae) (Riedman 1990;Sinclair et al. 1994;
Call and Ream 2012). Shifts in the quantity and age class of fish
consumed by fur seals in the Bering Sea occurred in the late 1970s
(Swartzman and Haar 1983), concurrent with changes in the abun-
dance of fish stocks (National Marine Fisheries Service 1993). This
observed change in dietary intake has led to the hypothesis that
the caloric intake of fur seals has declined. Reductions in net
energy gains have obvious impacts on individual development,
survival, reproductive fitness, and ultimately, population growth
rates. However, the impact of the apparent dietary changes on the
net energy gained by individual northern fur seals is unknown.
Previous controlled feeding studies have investigated aspects of
how northern fur seals digest different single-species diets. For
instance, Miller (1978) and Fadely et al. (1990) investigated the dry
matter digestibility (the proportion of dry matter in food that is
absorbed) of different single-species diets consumed by northern
fur seals. These studies provided basic information on the diges-
tive efficiency of fur seals, but did not examine critical pathways
of energy transformation and assimilation. Nor did they examine
the digestibility of mixed-species diets, which are more represen-
tative of what wild fur seals consume.
The goal of our study was to investigate the efficiency of energy
transformation and absorption by six captive female northern fur
seals fed eight different diets. A secondary goal was to test the
theory that mixed-species diets provide greater energetic gain
than single-species diets. Experimental diets were composed of
four prey species of varying compositions (fed alone or in combi-
nation). The fur seals’ complete digestive pathway was calculated
to quantify net energy gain by measuring three pathways of diges-
tive energy loss: fecal energy loss, urinary energy loss, and the
heat increment of feeding. In addition, we also investigated po-
tential short-term physiological changes in the fur seals’ metabo-
lism due to dietary shifts. Results from our study are an essential
step in understanding the relationship between diets and individ-
ual energy budgets of northern fur seals, and aid in evaluating
whether dietary shifts are negatively affecting the energetic status
of northern fur seals in the North Pacific and Bering Sea.
Materials and methods
Animals
Experiments were conducted throughout November 2012 to
June 2013 on six female northern fur seals that were 4.5 years of
age, with a body mass of 19.5–28.9 kg at the start of the study. The
fur seals were captured as pups (approximately 4 months old) in
October 2008 from St. Paul Island, Alaska, USA. Subsequently, the
fur seals were housed at The University of British Columbia’s
Marine Mammal Energetics and Nutrition Laboratory, located at
the Vancouver Aquarium (Vancouver, British Columbia, Canada).
All experimental manipulations were in accordance with the
guidelines of The University of British Columbia Animal Care
Committee (#A10-0342) and the Canadian Council on Animal
Care. The fur seals’ standard diet consisted of thawed Pacific her-
ring and market squid (Loligo opalescens Berry, 1911), supplemented
with vitamins, fed three times a day. The fur seals had access to
continuous-flow seawater pools (with adjacent haul-out space)
that reflected local ocean temperatures during the experimental
period (8.6–10.6 °C). Fur seals were weighed daily on a platform
scale (±0.02 kg) prior to feeding.
Test diets and experimental design
The fur seals were subject to eight test diets that were hand-fed
by trainers three times a day to ensure the schedule of food intake
was consistent across trials. Experimental diets lasted 3 weeks and
were composed of four key prey items that fur seals encounter in
the wild: Pacific herring, walleye pollock, capelin, and magister
armhook squid (Berryteuthis magister (Berry, 1913)), fed alone or in
combinations (Table 1). Animals were previously exposed to her-
ring and capelin, but not to pollock or magister armhook squid.
The amount of fish consumed by the fur seals was recorded daily.
The different prey items were chosen to represent a range of
proximate compositions and energy densities (Table 1). The aim
was for the fur seals to be fed at a constant level of gross energy
intake (GEI) that approximated maintenance levels, such that the
fur seals were neither gaining nor losing body mass (Kleiber 1975).
As maintenance energy requirements varied between fur seals, a
separate target GEI was predetermined for each fur seal at the
start of each feeding trial. These target GEIs were also adjusted
with observed changes in body mass during the experiment to try
to ensure body mass was held constant across all diets. The energy
(GEI) required for maintenance was estimated to be between
11 500 and 12 500 kJ·d
−1
. This resulted in differences in the
amounts of ingested mass by the individual fur seals per diet.
The eight test diets consisted of (1) herring only, (2) pollock only,
(3) capelin only, (4) herring + pollock (50% by energy), (5) herring +
capelin (50%), (6) pollock + capelin (50%), (7) herring (batch B) +
124 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
magister armhook squid (50%), and (8) herring + pollock + capelin
(33%). Quantities of each prey type in the mixed-species diets were
balanced to provide equal levels of gross energy (resulting in dif-
ferent amounts of ingested mass). All six of the fur seals were
subject to all test diets, except for diet 7 that lasted 2 weeks and
was only consumed by four of the fur seals due to a shortage of
magister armhook squid. Additionally, all of the diets used her-
ring from the same batch (batch A) with the exception of the
magister armhook squid diet that used a different batch of her-
ring (batch B; Table 1). The fur seals were divided into three treat-
ment groups with two fur seals per group and test diets were
randomly assigned to each group to counter any potential effects
due to seasonality. Each feeding trial was conducted over a three-
week period and consisted of three phases: acclimation occurred
during the first week of feeding, fecal samples were collected
during the second week, and metabolic trials (detailed below)
were conducted during the third week.
Fecal sample collections
During the second week of each trial, fecal samples were col-
lected several times a day from the bottom of the holding pool or
haul-out area for subsequent digestibility analyses. Animals were
held in pools according to diet groups, but fecal samples from
individual fur seals were distinguished by using coloured micro-
grit markers (Micro Tracers Inc., 1370 Van Dyke Avenue, San Fran-
cisco, California, USA). Approximately 5–6 g of micro-grits were
fed via gel capsules inserted into the opercular cavity of the fish
over the first two feedings of the day. Fecal samples were collected
and date, time, mass, and fur seal identity were noted (by color)
for each sample. Samples were frozen in sealed plastic bags at
−20 °C until further analysis.
Metabolic rate measurements
The resting metabolic rate (RMR), added thermoregulation cost
(TC), and heat increment of feeding (HIF%) for each fur seal on
each diet were measured via open-circuit respirometry. During
the measurements of RMR and TC the animals were in a fasted
state (>16 h after their last meal), as well as during the initial
metabolic rate baseline for HIF (further explained below). Also, for
the entire duration of metabolic measurements, animal behav-
iour, ambient air temperature, and metabolic chamber tempera-
ture were noted every 5 min.
RMR is the total energy used by animals to perform vital bodily
functions while in a relaxed and postabsorptive state (Kleiber
1975). RMR was measured for each fur seal on the last day of each
feeding trial. The fur seals voluntarily entered a custom 340 L
metabolic chamber (dimensions: 0.92 m ×0.61 m ×0.61 m) where
rates of oxygen consumption (V
˙O
2
) and carbon dioxide production
(V
˙CO
2
) were measured while the animal rested in ambient air. V
˙O
2
and V
˙CO
2
were measured by continuously drawing ambient air
through the metabolic chamber at a set rate of 125 L·min
−1
using a
Sable Systems Model 500H Mass Flow Controller that continu-
ously corrected the flow to standard temperature and pressure
(Sable Systems, Las Vegas, Nevada, USA). Subsamples of excurrent
air were then desiccated using anhydrous Drierite (Hammond
Drierite, Xenia, Ohio, USA) and were analyzed afterwards to quan-
tify O
2
and CO
2
concentrations the chamber via Sable Systems
FC-1B and CA-1B gas analyzers, respectively (Sable Systems, Las
Vegas, Nevada, USA). Gas concentrations were monitored and re-
corded to a portable computer every 0.5 s using Sable Systems’
Expedata software. Ambient air baselines at the start and end of
each trial were used to account for any system drift. Changes in O
2
and CO
2
concentrations compared with ambient air baselines
were converted into V
˙O
2
(Withers 1977). RMR was determined as
the lowest continuous mean V
˙O
2
maintained for 20 min during
the last 30 min of the 45 min trial.
The potential additional cost of thermoregulation in cold water
(TC) was measured immediately following the RMR measurement.
After completion of the RMR measurement, data recording was
paused and the chamber was filled roughly two-thirds full with
continuously flowing 2 °C water. When the fur seal was partially
submerged, the metabolic trial resumed and the animal’s meta-
bolic rate was measured while in water for an additional 30 min.
TC, i.e., the metabolic rate while in the water, was determined as
the lowest continuous mean V
˙O
2
consumption maintained for
!10 min during the last 20 min of the 30 min trial. TC was calcu-
lated as the change in metabolic rate compared with the previ-
ously measured RMR in air.
The heat increment of feeding (HIF%) is the increase in metab-
olism resulting from the mechanical and chemical digestion of a
recent meal. Measurements of HIF% were conducted during the
last week of each diet trial. RMR was initially measured (as previ-
ously described) and after completion of the 30 min trial to obtain
an RMR baseline, data collection was then temporarily paused
while the fur seal was fed a meal of known energetic content
(6294.7 ± 1202.2 kJ, approximately half of their daily GEI) while
remaining inside the metabolic chamber. Data collection then
resumed and V
˙O
2
was monitored for about 5–6 h afterwards to
capture the entire postprandial rise in metabolism. V
˙O
2
was then
converted into rates of energy utilization (1 L O
2
= 20.1 kJ; Blaxter
1989) to quantify the energetic cost of HIF%. Ultimately, HIF% was
calculated as the total increase in energy utilized above RMR,
expressed as a percentage of the GEI of the ingested meal.
Fish prey and fecal laboratory analysis
Fish and fecal samples were analyzed in-house (see below) and
additional samples were sent to a commercial laboratory (SGS
Canada Inc., Burnaby, British Columbia, Canada) for quality con-
trol (i.e., to provide a correction factor for in-house measure-
ments). At least 10 samples of each of the prey items were analyzed
by SGS Canada Inc. for proximate composition (moisture, lipid,
and protein), energy density, and manganese (Mn
2+
) concentra-
tion (Table 1). An additional 10 samples of each of the fish items
were similarly analyzed in-house for the same analyses (see
below).
Three separate fecal samples per fur seal per diet were selected
for analysis; the majority of the samples were collected 24 h apart
Table 1. Mean proximate composition (percent crude protein and percent lipid content), mean energy density, mean manganese (Mn
2+
) concen-
tration, mean (±SD) body size (measured as length), and mean (±SD) body mass of a subsample of four species of prey (n= 12 of each) experimen-
tally fed to six female northern fur seals (Callorhinus ursinus).
Experimental prey
Water
(%)
Total lipid
(%)
Crude protein
(%)
Energy density
(kJ·g
−1
)
Mn
2+
(ppm)
Fish length
(cm)
Fish mass
(g)
Pacific herring (Clupea pallasii)
Batch A (main source) 68.5 41.6 51.4 24.3 5.1 19.9±1.5 93.0±20.8
Batch B (magister armhook squid diet only) 69.2 37.0 53.6 22.9 5.5 18.5±0.6 64.0±6.7
Walleye pollock (Gadus chalcogrammus) 75.3 32.8 57.5 22.1 2.4 24.5±2.2 134.0±33.2
Capelin (Mallotus villosus) 82.6 4.0 81.6 15.2 2.9 15.0±1.0 24.0±5.4
Magister armhook squid (Berryteuthis magister) 71.3 44.3 46.7 23.2 2.8 ——
Note: Proximate composition, energy density, and Mn
2+
concentration measured on a dry-weight basis.
Diaz Gomez et al. 125
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
from each other. From these fecal samples, 16 representative sam-
ples were sent to SGS Canada Inc. to be analyzed for Mn
2+
concen-
tration, while 138 were analyzed in-house. Fecal and fish samples
were thawed, ground (fish only), and two duplicate subsamples
(!25 g each) were then placed in polycarbonate vials and weighed
to the nearest milligram. Fecal and fish samples were refrozen
and subsequently freeze-dried for 36 h to a constant mass (Freeze
dryer Freezone 6; Labconco Corp., Kansas City, Missouri, USA).
Dried samples were reweighed to determine dry matter and water
content.
Homogenized dried samples were used to measure proximate
composition, energy density, and Mn
2+
concentration. Through-
out all laboratory analyses, sample variation of ≤5% was con-
sidered acceptable between subsample replicates. Total energy
density of fecal and fish samples was determined by combustion
of duplicates of 1 g of dried sample using an oxygen bomb calo-
rimeter (6400 Automatic isoperibol calorimeter; Parr Instrument
Company, Moline, Illinois, USA). Total crude protein of fish and
fecal samples was determined by measuring total Kjeldahl ni-
trogen (TKN) of duplicates of approximately 0.2 g of dried samples
with the addition of 2 Kjeltabs Cu catalyst tablets (FOSS, Eden
Prairie, Minnesota, USA), using the Kjeldahl method (AOAC 1990)
via spectrophotometric flow injection analyzer (FOSS FIAstar 5000
TKN analyzer unit; FOSS, Eden Prairie, Minnesota, USA) measured
at 590 nm. Nitrogen concentration was multiplied by a factor of
6.25 to determine total crude protein, based on the assumption
that 100 g of crude protein contains 16 g of nitrogen (Robbins
1993). Lipid contents from duplicate samples of approximately 2 g
of feces and 1.5 g of fish were measured using a modified Bligh–
Dyer technique (Bligh and Dyer 1959). It is important to note that
most of the fish and fecal samples used in our experiment were
relatively high in lipid (>2%), which in some studies has led to
underestimated lipid content (Iverson et al. 2001;Budge et al.
2006). However, our in-house lipid and protein laboratory analy-
ses were corroborated against the SGS Laboratory results to
ensure the accuracy of our techniques. Both protein and lipid
contents of samples were expressed as a percentage of total dried
samples.
Fish and fecal Mn
2+
concentrations were determined through
wet oxidation of duplicates of dried 0.4 g of fish and 0.2 g of fecal
samples. Concentrations were determined by using an atomic
absorption spectrophotometer (Perkin-Elmer 2380; 279.5 nm
wavelength, slit width 0.2 nm, oxidizing air-acetylene flame;
Perkin-Elmer, Montréal, Quebec, Canada). Standard curves were
generated with a Mn
2+
standard stock solution of 1.0 ppm MnNO
3
by serial dilutions to approximate 0.02, 0.04, 0.06, 0.08, 0.10, 0.20,
and 0.40 Mn
2+
concentrations (ppm).
Digestibility calculations
Laboratory results for the fish and fecal samples per animal per
diet were averaged together to provide a single value for an entire
diet trial for each individual fur seal. All calculations were done on
a dry-matter basis.
Calculations of digestibility efficiency require a means of deter-
mining the amount of prey “represented” by a fecal sample. Nat-
urally occurring Mn
2+
content in fish prey and feces has been
widely utilized as an inert marker (given its low biological require-
ments) to quantify digestibility of fur seals and other pinnipeds in
previous studies (e.g., Fadely et al. 1990;Lawson et al. 1997;Rosen
and Trites 2000b). However, the low concentrations of Mn
2+
in the
prey samples led to unacceptable levels of variance in the in-house
analyses; therefore, only the Mn
2+
results obtained from SGS Lab-
oratories were used.
GEI was calculated by multiplying ingested mass by the energy
density of the prey items, in proportion to the amount fed of each
experimental diet.
DMD% is the relative assimilation of dry materials and was cal-
culated as the change in concentration of Mn
2+
concentration
between diet and feces:
(1) DMD (%) !
!
1"Ci
Cf
"
×100
where Cis the concentration of Mn
2+
in diet (i) and feces (f)
(Schneider and Flatt 1975).
Digestible energy (DE%), i.e., the amount of energy assimilated,
was calculated as
(2) DE (%)!
!
1"Ci×Ef
Cf×Ei
"
×100
where Eis the energy density of the ingested diet (i) and feces (f)
(Mårtensson et al. 1994).
Fecal energy loss (FEL%), i.e., the inverse of DE%, was calculated as
(3) FEL (%)!1"DE%
To calculate urinary energy loss (UEL) per day, apparent digest-
ible nitrogen intake (ANI) was first calculated as
(4) ANI (g· d"1)
!total crude protein consumed ×digestibility of crude protein
6.25
where digestibility of crude protein was calculated as
(5) Protein digestibility (%)!
!
1"Ci×Pf
Cf×Pi
"
×100
where Pis the crude protein content of the ingested diet (i) and
feces (f)(Mårtensson et al. 1994).
UEL was calculated with the following formula based on data
from Keiver et al. (1984):
(6) UEL (kJ· d"1)!(6.128 ×ANI #14.737) ×4.186
The estimated energetic content in urine was then represented
as a proportion of DE% (rather than GEI) since urinary losses are
proportional to absorbed nitrogen and independent of what is
lost in the feces (Dierauf and Gulland 2001). While UEL% is most
accurately reported as a proportion of DE%, UEL% was also calcu-
lated as a proportion of GEI for data analysis only to keep statisti-
cal analysis consistent across all variables.
Metabolizable energy (ME%), i.e., the energy that remains avail-
able after accounting for the energy lost through the excreta,
expressed as a percentage of GEI, was calculated as
(7) ME (%)!
#
GEI "(FEL #UEL)
GEI
$
×100
Net energy (NE) is the total energy gained by fur seals after
accounting for the energy lost through excreta and through the
HIF%. NE% is this value expressed as a percentage of GEI and was
calculated as
(8) NE (%)!
#
GEI "(FEL #UEL #HIF)
GEI
$
×100
126 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
Testing effect of diet on digestibility and bioenergetics
Statistical differences in digestive and physiological parameters
attributable to diet type were determined via linear mixed-effect
(LME) models using R version 3.1.2 statistical software (R Core
Team 2014). Models were fitted using maximum-likelihood esti-
mates as required for LME model comparison using the package
nlme (Pinheiro et al. 2015). LME models were built in a stepwise
fashion to assess the ability of the fixed factors to explain differ-
ences in the response variables, such that models containing
fixed-effect factors hierarchically nested within the null model
(lacking a fixed-effect factor) were compared against the null
model and models with fewer fixed effects by likelihood ratio
tests (LRT) and by comparison of Akaike’s information criterion
(AIC) values. LME models accounted for repeated measures and
variability within and among animals by treating Animal ID as a
random effect, which also allowed inferences from the sample
population to be applied to their wild counterparts (Pinheiro
and Bates 2000;Crawley 2007;Zuur et al. 2009;Galecki and
Burzykowski 2013).
The first step in the data analysis was to determine the statisti-
cal influence of diet type on the following response variables:
percent change in body mass, dry matter digestibility, digestible
energy, fecal energy loss, urinary energy loss, heat increment of
feeding, and net energy gain. Due to the relatedness between
response variables, we examined each relationship indepen-
dently, and diet was always the only fixed factor tested in this
initial analysis. To investigate the nature of significant differences
between response variables and diets, a post hoc Tukey’s contrasts
simultaneous test for general linear hypotheses was used after
fitting the separate models.
When preliminary analysis demonstrated that diet type was a
significant factor, subsequent analyses explored which compo-
nents of the diet were at the root of the relationship by testing a
number of relevant fixed effects, but excluding diet type. Fixed
effects that were tested as potential model factors included food
mass intake (kg·d
−1
), gross energy intake (kJ·d
−1
), diet lipid intake
(%·d
−1
dry-weight), diet protein intake (%·d
−1
dry-weight), and lipid
to protein intake ratio. All models were compared against the null
model using an LRT test. Models with the same response variable
but different dependent variables in this analysis were compared
by AIC values as described by Pinheiro and Bates (2000) and
Crawley (2007), by selecting the lowest AIC and most parsimoni-
ous model (i.e., more complex models were tested against those
with fewer fixed effects). The selected best-fit model contained the
factor that best explained the trends observed in the response
variables and thus fitted the data the most accurately while ful-
filling the assumptions of the LME models (Pinheiro and Bates
2000;Crawley 2007;Zuur et al. 2009;Galecki and Burzykowski
2013).
To test whether mixed-species diets provided a greater than
expected digestibility efficiency, expected DMD% and expected
DE% of mixed-species diets were calculated (except for herring
and magister armhook squid diet). The expected digestibility of
energy for mixed-species diets (except for herring + magister arm-
hook squid diet) was calculated as a weighted mean from the
observed DE% of the single-species diet counterparts, propor-
tional to the energy densities of each individual prey species in
the diet (see Forster 1999) according to
(9) Expected DE (%)
!(massF1 ×energyF1 ×DE%F1)#(massF2 ×energyF2 ×DE%F2)
(massF1 ×energyF1 #massF2 ×energyF2)
Similarly, expected DMD% for the mixed diets were calculated
from the observed DMD% from the relevant single-species diet
counterparts, weighed proportional to the ingested mass (dry-
weight) of each component species. Statistical differences be-
tween expected and observed DMD% and DE% were determined
using a Welch two-sample ttest.
Statistical differences in metabolic measurements (specifically,
resting metabolic rate (RMR) and the added thermoregulation
cost (TC)) attributable to changes in diet were determined via
LME models in the same manner as previously explained. One-
sample ttests were also used to determine whether the added
cost of TC was significantly different from zero. Preliminary anal-
ysis resulted in data from one of the fur seals (ME08) being con-
sidered an outlier because it failed to fulfill the assumptions of the
LME models when included in the analysis (see also Dalton et al.
2014). Excluding this animal from all RMR and TC metabolic data
analysis resulted in all LME models meeting the assumptions of
normality of the random effect and of the residual errors and
homogeneity of the variance (Pinheiro and Bates 2000;Crawley
2007;Zuur et al. 2009;Galecki and Burzykowski 2013).
Results
Changes in body mass
The overall mean body mass of the fur seals at the start of each
of the diet trials was 23.3 ± 0.4 kg (mean ± SD; Table 2). Despite
minor changes in body mass while on the different diets (ranging
from herring and pollock diet +1.3% ± 3.3%, to herring and magis-
ter armhook squid diet −2.3% ± 2.3%), there were no significant
differences in percent body mass change due to diet (%) (F
[33]
= 0.4,
p= 0.9).
Prey item and dietary characteristics
Proximate composition, energy density, and Mn
2+
concentra-
tion differed among the four experimental prey items (dry-weight
basis; Table 1). Overall, magister armhook squid had the highest
lipid content (44.3%), while capelin had the lowest (4.0%). Con-
versely, capelin had the highest protein content (81.6%), while
magister armhook squid had the lowest (46.7%). Herring (batch A)
Table 2. Body mass of six captive female northern fur seals (Callorhinus ursinus) at the start of the feeding trial and ingested mass (wet) of the eight
experimental diets with their respective proximate composition, energy density, and manganese (Mn
2+
) concentration.
Diet
Fur seal
body mass (kg)
Ingested diet
mass (kg) Water (%)
Total
lipid (%)
Protein
(%)
Energy density
(kJ·g
−1
) Mn
2+
(ppm)
Pacific herring (Clupea pallasii) 23.9±3.5 1.6±0.3 68.5±3.6 38.0±0.01 47.1±0.01 24.3±0.01 5.1±0.01
Walleye pollock (Gadus chalcogrammus) 23.1±3.1 2.3±0.3 75.3±1.3 35.8±0.01 62.8±0.01 22.1±0.01 2.4±0.01
Capelin (Mallotus villosus) 23.2±3.3 3.3±0.5 82.6±1.4 3.3±0.01 67.6±0.01 15.2±0.01 2.9±0.01
Herring + pollock 23.1±3.2 2.0±0.1 72.4±0.04 37.0±0.01 54.6±0.1 23.1±0.01 3.8±0.02
Herring + capelin 23.0±2.7 2.9±0.3 79.0±0.04 15.9±0.1 60.2±0.08 18.7±0.04 3.7±0.01
Herring + magister armhook squid
(Berryteuthis magister)
23.8±3.2 2.4±0.2 70.3±0.01 43.2±0.1 57.2±2.8 23.1±0.01 4.3±0.02
Pollock + capelin 22.9±2.8 3.0±0.5 79.9±0.5 15.9±2.2 65.8±0.3 18.3±0.5 2.7±0.03
Herring + pollock + capelin 23.1±2.8 2.6±0.5 77.9±0.06 21.0±0.2 60.7±0.9 19.7±0.05 3.4±0.01
Note: Proximate composition, energy density, and Mn
2+
concentration measured on a dry-weight basis. Values are reported as means ± SD.
Diaz Gomez et al. 127
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
had the highest energy density (24.3 kJ·g
−1
), while capelin had the
lowest (15.2 kJ·g
−1
).
For the eight experimental diets, proximate compositions (dry-
weight basis) and energy density also differed significantly
(Table 2). Lipid content varied from 3.3% ± 0.01% (capelin diet) to
43.2% ± 0.1% (herring and magister armhook squid diet) (LRT =
265.3, p< 0.001). Diet protein content also differed significantly,
ranging from 47.1% ± 0.01% (herring diet) to 67.6% ± 0.01% (capelin
diet) (LRT = 366.9, p< 0.001), while energy density ranged from
15.2 ± 0.01 kJ·g
−1
(capelin diet) to 24.3 ± 0.01 kJ·g
−1
(herring diet)
(LRT = 266.7, p< 0.001). As the eight diets were balanced for GEI at
maintenance levels, ingested mass also differed with diet (wet-
weight) (LRT = 76.9, p< 0.001). Mean ingested mass ranged from
1.6 ± 0.3 kg (herring-only diet) to 3.3 ± 0.5 kg (capelin-only diet)
(Table 2).
Despite the fact that GEI was targeted to be within a specific
range of daily intake (i.e., 11 500–12 500 kJ·d
−1
) regardless of diet
type, daily GEI differed significantly across diets (LRT = 61.1,
p< 0.001; Table 3) due to two anomalies. The first was that the
animals refused to eat sufficient levels of capelin (!3.3 kg), which
resulted in GEI being significantly lower while consuming capelin
than for other diets (8712.9 ± 1409.7 kJ·d
−1
). The second was that
the herring and magister armhook squid diet had the highest GEI
(15 866.0 ± 1 426.8 kJ·d
−1
), a byproduct of an attempt to maximize
the use of the available magister armhook squid because the fur
seals showed high enthusiasm to its consumption. Surprisingly,
these differences in GEI (and related differences in net energy
gain) did not result in statistically significant changes in body
mass.
Changes in digestibility and bioenergetics due to changes
in diet
There were significant differences in DMD% among the experi-
mental diets (LRT = 38.0, p< 0.001; Table 3). DMD% was signifi-
cantly lower for pollock (88.2% ± 1.1%) than most other diets, and
highest for the herring and magister armhook squid diet (92.3% ±
0.1%). DMD% also decreased significantly with increased protein
content of diets (%) (LRT = 9.9, p= 0.002; Fig. 1).
Fecal energy density was significantly different across diets,
with the pollock diet having the lowest fecal energy density (7.7 ±
1.1 kJ·g
−1
) and the herring and magister armhook squid diet having
the highest (12.9 ± 0.9 kJ·g
−1
) (LRT = 56.5, p< 0.001). However, to
calculate total daily FEL%, these data were combined with the
Mn
2+
data and the prey energy density data (see eq. 3). The Mn
2+
of
the fecal samples ranged from 20.7 ± 3.3 ppm for the pollock diet
to 60.9 ± 9.7 ppm for the herring diet.
FEL%, expressed as a percentage of GEI, ranged from 3.1% ± 0.3%
to 4.1% ± 0.6% and was significantly different across diets (LRT =
19.6, p= 0.006; Table 3). The lowest FEL% was for the herring-only
diet and the highest FEL% was for the pollock and capelin diet.
Both the protein content (%) and mean ingested mass significantly
affected FEL%, such that increases in proportion of protein (LRT =
9.5, p= 0.002) and ingested mass (LRT = 9.4, p= 0.002) resulted in
increased FEL%.
Similarly, DE%, which is the inverse of FEL%, was observed to be
high overall and differed significantly by diet (LRT = 19.6, p= 0.006;
Table 3). DE% ranged from 95.9% ± 0.7% for the pollock and capelin
diet to 96.9% ± 0.3% for the herring-only diet. It was also inversely
related to both mean ingested mass (LRT = 9.4, p= 0.002) and
protein content of diets (%) (LRT = 9.5, p= 0.002), such that in-
creased intake in either ingested mass or protein resulted in de-
creased DE% (Fig. 2).
Both DMD% and DE% reflect digestive efficiencies, but the for-
mer is a measure of dry-matter digestibility, while the latter is
determined on an energetic basis. Although there was a signifi-
cant positive relationship between DMD% and DE% across all eight
experimental diets (DE% = 0.21·DMD% + 77.8, r
2
= 0.38) (LRT = 21.6,
p< 0.001), the slope was 0.2 and therefore the measures are not
interchangeable.
UEL%, expressed as percentage of GEI, ranged from 8.9% ± 0.1 to
22.0% ± 0.2% and was significantly different across diets (LRT =
247.9, p< 0.001; Table 3). The lowest UEL% was from the herring-
only diet and the highest UEL% was from the capelin diet. There
was a significant interaction between protein content (%) and lipid
content (%) among experimental diets such that the interaction of
both factors together affected UEL% significantly more than each
factor separately. UEL% increased with increases in protein con-
tent and decreased with increases in lipid content (LRT = 62.8,
p< 0.001).
ME% available to the fur seals was calculated by the subtraction
of fecal and urinary energy losses from the GEI (Table 3). ME%
differed significantly by diet (LRT = 197.5, p< 0.001). The lowest
amount of energy available was 70.3% ± 1.5% from the capelin-only
diet, while the highest was 87.2% ± 0.2% from the herring-only
diet. ME% increased significantly with increasing lipid content
of diets (%) (LRT = 133.1, p< 0.001).
HIF% differed significantly by experimental diet (LRT = 36.9,
p< 0.001; Table 3). HIF% was significantly greater while consuming
the capelin-only diet (12.4% ± 2.0%) and the least costly while con-
suming the herring-only diet (4.3% ± 1.0%). Furthermore, HIF%
varied significantly with the lipid content of the diets (%), where
HIF% decreased as lipid content increased (LRT = 15.3, p= 0.001;
Fig. 3).
Total net energy gain (kJ·d
−1
) increased significantly with in-
creases in GEI (kJ·d
−1
) across experimental diets with a positive
relationship (NE = 1.04·GEI – 3043.9, r
2
= 0.93). However, NE% as a
proportion of GEI also differed by diet (LRT = 122.3, p< 0.001;
Table 3). NE% was lowest while consuming the capelin diet, where
animals retained only 57.9% ± 2.6% of the ingested energy, and
highest when consuming the herring diet, where they retained
83.0% ± 1.0%. Lipid content in the diets (%) was a significant factor
in determining NE%, such that fur seals retained the most NE%
from fattier diets (LRT = 77.7, p< 0.001; Fig. 4).
Table 3. Dry matter digestibility (DMD%), gross energy intake (GEI), fecal energy loss (FEL%), digestible energy (DE%), apparent digestible nitrogen
intake (ANI), urinary energy loss (UEL%), metabolizable energy (ME%), heat increment of feeding (HIF%), and net energy (NE%) of six captive female
northern fur seals (Callorhinus ursinus) across the eight experimental diets.
Diet DMD% GEI (kJ·d
−1
) FEL% DE% ANI (g·d
−1
) UEL% ME% HIF% NE%
Pacific herring (Clupea pallasii) 91.5±1.1 12 135.7 ± 2 412.5 3.1±0.3 96.9±0.3 39.4±7.9 9.9±0.1 87.2±0.2 4.3±1.0 83.0±1.0
Walleye pollock (Gadus chalcogrammus) 88.2±1.1 12 688.6 ± 1 570.4 3.7±0.5 96.3±0.5 50.9±6.1 10.3±0.1 86.4±0.5 6.5±3.8 80.0±3.5
Capelin (Mallotus villosus) 90.2±1.9 8 712.9 ± 1 409.7 4.0±0.8 96.0±0.8 72.3±11.6 26.7±1.9 70.3±1.5 12.4±2.0 57.9±2.6
Herring + pollock 90.1±0.4 12 482.5 ± 787.9 3.5±0.7 96.5±0.7 45.3±2.7 10.1±0.1 86.8±0.7 7.1±2.3 79.7±2.8
Herring + capelin 90.8±1.2 11 301.8 ± 1 245.9 3.5±0.4 96.5±0.4 65.3±6.8 18.6±0.1 78.5±0.3 7.9±3.0 70.6±3.1
Herring + magister armhook squid
(Berryteuthis magister)
92.3±0.1 15 866.0 ± 1 426.8 3.9±0.2 96.1±0.2 55.1±4.8 9.3±0.1 87.1±0.1 6.0±1.5 81.1±1.5
Pollock + capelin 88.6±2.2 11 118.7 ± 1 613.2 4.1±0.6 95.9±0.7 66.3±10.9 18.1±1.2 78.5±1.1 6.9±2.0 71.6±1.2
Herring + pollock + capelin 91.0±1.0 11 472.6 ± 2 184.1 3.3±0.4 96.7±0.4 59.5±11.5 15.8±0.2 81.4±0.4 5.2±1.1 76.2±1.0
Note: All digestibility measures are expressed as a proportion of GEI, except for UEL% which is expressed as a proportion of DE%. Values are reported as means ±SD.
128 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
Changes on digestive efficiency due to diet mixing
Comparisons between observed DMD% and expected DMD% of
mixed-species diets (based on calculations from observed DMD%
of their single-species diet components) showed no significant
changes in DMD% due to diet mixing for any of the diets (p> 0.05).
Similarly, expected DE% values were not significantly different for
any of the mixed-species diets when compared with their respec-
tive observed DE% (p> 0.05). Therefore, diet mixing did not pro-
vide a significant advantage to fur seals to better assimilate either
dry matter or energy.
Effect of diet on metabolism
The mean mass-specific resting metabolic rate (RMR) while the
fur seals were resting in ambient air across all diets was 10.0 ±
3.4 mL O
2
·kg
−1
·min
−1
. While mean mass-specific RMR ranged from
8.2 ± 4.3 mL O
2
·kg
−1
·min
−1
for the pollock diet to 11.9 ± 4.6 mL
O
2
·kg
−1
·min
−1
for the herring and magister armhook squid diet, it
did not significantly differ among the experimental diets (F
[26]
=
2.2, p= 0.07). Mean mass-specific metabolic rate while the fur seals
were partially submerged in 2 °C water was 17.1 ± 4.1 mL
O
2
·kg
−1
·min
−1
, and this ranged from 13.8 ± 4.4 mL O
2
·kg
−1
·min
−1
Fig. 1. Changes in dry matter digestibility (DMD%) with the eight experimental diets tested in six captive female northern fur seals
(Callorhinus ursinus). Diets are arranged accordingly from low to high protein content (%), as denoted above the diet labels. Each box represents
the median (thick horizontal line), first and third quartiles (“hinges”), and 95% confidence intervals of the median (“notches”). Data for each
diet trial are from fur seals, with the exception of the herring and squid diet which was only consumed by four of the animals. Letters above
boxes indicate significant differences between diets.
Fig. 2. Changes in digestible energy (DE%) over the eight experimental diets tested in six captive female northern fur seals (Callorhinus ursinus).
Diets are arranged accordingly from low to high mean ingested mass (wet-weight) during experimental trials, as denoted above the diet
labels. Each box represents one diet trial for the fur seals, with the exception of the herring and squid diet which was only consumed by four
of the animals. Numbers above individual boxes indicate the protein content (%) for each of the experimental diets. Letters above boxes
indicate significant differences between diets.
Diaz Gomez et al. 129
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
for the pollock diet to 18.9 ± 5.4 mL O
2
·kg
−1
·min
−1
for the herring
and pollock diet. The added thermoregulation cost (TC) of being
partially submerged in 2 °C water, i.e., calculated as the mean
mass-specific amount of oxygen consumed above RMR, was 6.9 ±
3.3 mL O
2
·kg
−1
·min
−1
. The added TC ranged from 5.6 ± 3.3 to 8.1 ±
3.1 mL O
2
·kg
−1
·min
−1
, where the lowest rate was for the pollock
diet and the highest rate for capelin. Also, TC was found to be
significantly different from zero (p< 0.05) for all diets, with the
exception of the fur seals on the herring and magister armhook
squid diet (p= 0.05). The latter exception is likely the result of the
smaller sample size for the TC trial, since only data from three out
of the four animals consuming the diet was collected. However,
TC was not significantly different across experimental diets (F
[26]
=
1.2, p= 0.4).
Discussion
In most broad classifications, food with a high energy density is
considered to be of “high quality”, implying that it readily pro-
vides sufficient energy to its predator. The chemical energy in-
gested via food is defined as a consumer’s gross energy intake
(GEI) and is derived from the breakdown of its individual compo-
nents. For fish, this is a product of their lipid and protein content.
It has been estimated that 1 g of crude lipid contains 37.7 kJ of
energy, whereas 1 g of protein provides 17.8 kJ (Blaxter 1989).
However, the net energy gain (NE%), i.e., the biologically useful
energy available to the consumer after food has been broken
down and assimilated, is different. While NE% is roughly propor-
tional to GEI, it is affected by various factors such as composition
Fig. 3. Changes in heat increment of feeding (HIF%) of the eight experimental diets consumed by six captive female northern fur seals
(Callorhinus ursinus). Diets are arranged accordingly from low to high lipid content (%) in the experimental diets, as denoted above the diet
labels. Each box represents one diet trial for the fur seals, with the exception of the herring and squid diet which was only consumed by four
of the animals. Letters above individual boxes indicate significant differences between diets.
Fig. 4. Changes in net energy gain (NE%) from the eight experimental diets tested in six captive female northern fur seals (Callorhinus ursinus).
Diets are arranged accordingly from low to high lipid content (%) in the experimental diets, as denoted above the diet labels. Each box
represents one diet trial for the fur seals, with the exception of herring and squid diet which was only consumed by four of the animals.
Letters above individual boxes indicate significant differences between diets.
130 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
of diet, the level of food intake, and nutritional status (Schneider
and Flatt 1975). It is therefore necessary to empirically assess en-
ergy loss throughout the digestive process of an animal to deter-
mine and understand the NE% benefit of a particular diet.
Our study is the first to measure the complete pathway of en-
ergy transformation for an otariid, including digestible energy
(DE%), the heat increment of feeding (HIF%), metabolizable energy
(ME%), and net energy gained (NE%). Furthermore, unlike previous
studies with fur seals and most other pinnipeds, our study com-
pared digestive efficiencies of mixed-species diets. Overall, our
results showed that low energy density prey items, i.e., those nor-
mally classified as “low quality“, yielded significantly less NE% to
the fur seals than would be predicted solely on the basis of GEI.
This was largely due to the lower digestibility of protein vs. lipid,
compounded by the negative effect of required increased prey
mass intake. Furthermore, contrary to theory, there appeared to
be no benefit in terms of energy digestibility associated with con-
suming mixed-species diets.
Changes in digestibility and bioenergetics due to changes
in diet
In the past, many studies have quantified the dry matter digest-
ibility (DMD%), i.e., a measure of the proportion of indigestible to
digestible dry matter in food, as a proxy for energetic digestibility
in pinnipeds. The fur seal’s DMD% in our study was high and
varied significantly across diets (Table 3). DMD% was lowest for
walleye pollock (Fig. 1), due to a combination of its high protein
content and the fact that pollock has large bony structures com-
pared with the other prey consumed, making it more challenging
to digest.
Dry matter digestibility values in our study were consistent
with previous pinniped studies (see Table 3 in Rosen and Trites
2000b). For example, our DMD values for the pollock diet (88.2% ±
1.1%) were similar to those of Miller (1978), who reported that the
DMD% for northern fur seals on a pollock diet (86.6%–90%) were
lower than for diets of herring capelin, or squid. While the highest
DMD% in our study was for the herring and magister armhook
squid diet (92.3% ± 0.1%), the DMD of our herring-only diet (91.5% ±
1.1%) was comparable with previous fur seal studies by both Miller
(1978; 91.6%–93%) and Fadely et al. (1990; 90%) (Table 3). Even
though DMD% values are often reported, they are not informative
with respect to the energy absorbed from the various diets (Rosen
and Trites 2000b). Previous studies on northern fur seals did not
extend their research beyond the measurement of DMD%, which
has meant that the energy transformation pathway for fur seals
has been unclear.
Our values of digestible energy (DE%) were generally high across
diets (Table 3) and were comparable with previous pinniped stud-
ies (see Table 3 in Rosen and Trites 2000b), as well as other carniv-
orous terrestrial mammals consuming either meat or fish
(Barbiers et al. 1982;Best 1985;Pritchard and Robbins 1990). The
similarly high DE% among these carnivorous species was expected
because they are all characterized by relatively simple stomachs
and short intestinal tracts (Stevens and Hume 1995).
However, digestible energy (and its converse, fecal energy loss)
was not constant across diets for the fur seals and was negatively
affected by increases in both protein content of the diet and in-
gested mass (Fig. 2). The significant decrease in DE% with increas-
ing protein content of the diet may be explained by the fact that,
among all of the components in food, the breakdown and assim-
ilation of proteins to obtain energy takes the most time and
effort (Blaxter 1989). Protein molecules are long chains with
strong bonds, which require great mechanical and chemical effort
to break down, and require more time to digest (Blaxter 1989;
Stevens and Hume 1995). This means that diets higher in protein
content would have higher digestive costs and would provide less
DE%. This decrease in DE% values with increasing nitrogen intake
has been consistently observed in other pinniped species (Keiver
et al. 1984;Ronald et al. 1984). DE% was also significantly affected
by increases in ingested mass. This decrease in the efficiency of
the digestive process with increases in food consumption levels
has been confirmed in other species (Schneider and Flatt 1975),
and is due in part to a decrease in chemical and mechanical effi-
ciency (because of the higher food bolus), as well as the increased
energy required to produce more fecal waste (both of which con-
tribute to decreases in DE% gain).
The protein content in our experimental diets also affected the
fur seal’s urinary energy loss (UEL%), whereby the diets with the
greatest protein content had the greatest UEL% (Tables 2,3). Lipid
content of the diet also had a significant interaction along with
protein content, which affected UEL%. This may be partly attrib-
utable to the complementary relationship between lipid and pro-
tein in food, such that lipid-rich diets are low in protein content
and vice versa. It is suspected that protein was the primary driver
of the relationship, as the breakdown of protein produces more
wastes than the breakdown of other dietary components. The
primary waste product is ammonia, which is a toxic byproduct
that must be transformed to urea to be eliminated. The UEL%
values of the fur seals were comparable with those of previous
studies with pinnipeds (Parsons 1977;Ashwell-Erickson and
Elsner 1981;Keiver et al. 1984;Ronald et al. 1984;Goodman-Lowe
et al. 1999). However, it is interesting to note that the UEL% from
diets containing capelin (including the capelin-only diet) was un-
expectedly high.
Urinary energy loss is notoriously difficult to directly measure
from complete urine collection in large mammals. In our study,
UEL% was estimated from the apparent digestible nitrogen intake
of each diet, using equations generated from previous studies
with phocid seals (Keiver et al. 1984;Goodman-Lowe et al. 1999).
These phocid studies found that the measured energy density of
urine was higher (leading to higher UEL% values) than if calcu-
lated solely from energetic values per gram of nitrogen as urea
(i.e., 22.6 kJ·g
−1
of nitrogen; Keiver et al. 1984). This suggests that
an unidentified component that was not of nitrogenous origin
within the urine contributed to the energetic content of the sam-
ples (Keiver et al. 1984). As a result, our estimated UEL% were
approximately 1.5 times higher than estimates based solely on
nitrogen content.
While the values for urinary energy loss in our study may seem
higher than previous estimates, it is worth noting that most pre-
vious UEL% studies in pinnipeds have been undertaken using her-
ring diets that had a relatively low protein content (lipid-rich). An
exception was a study where harbour seals (Phoca vitulina L., 1758)
fed a pollock-only diet (90.6% protein) had a UEL% that was
1.5 times higher than when the seals were fed only herring (56.3%
protein; Ashwell-Erickson and Elsner 1981). Similarly, the UEL% of
the fur seals fed the capelin diet (67.6% protein) was 2.7 times
greater than the herring-only diet (47.1% protein) (Table 3), further
demonstrating the high cost of disposal of nitrogenous waste
products from protein sources.
Few past pinniped studies have measured metabolizable energy
(ME%), the energy remaining after accounting for the energy that
is lost through the excreta. Most of these were obtained when the
animals were fed single-species diets and ranged from 82.7% to
92.5% for herring, from 85.9% to 89.4% for pollock, and from 78.3%
for squid (Parsons 1977;Ashwell-Erickson and Elsner 1981;Keiver
et al. 1984;Ronald et al. 1984;Costa 1988). The ME% of the fur seals
in our study across all experimental diets fit well within these
previous pinnipeds studies (Table 3) and with ME% values from
other carnivorous terrestrial mammals consuming either fish diet
(Pritchard and Robbins 1990) or mammal meat diet (Davison et al.
1978).
Overall, metabolizable energy values for the fur seals were sig-
nificantly positively correlated to lipid content of the diet. In con-
trast, Goodman-Lowe et al. (1999) reported that diets higher in
protein and lower in lipid content provided the greatest ME% to
Diaz Gomez et al. 131
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
Hawaiian monk seals (Monachus schauinslandi Matschie, 1905).
However, it is important to note that their comparisons were not
calculated as proportions of the gross energy intake and that their
data supports the same pattern found in our study when recalcu-
lated appropriately.
Of all the digestive processes, one of the most studied and best
understood is the heat increment of feeding (HIF%), also known as
the specific dynamic action of feeding (Jobling 1983). HIF% across
a wide range of vertebrate and invertebrate taxa has been found to
depend upon various features of the ingested meal (composition,
type, size, temperature), characteristics of the animal (body size,
sex and age), and of the environment where the animal is found
(ambient temperature) (Blaxter 1989;Secor 2009). In mammals,
the main factors that affect HIF% are the consumer’s body mass,
the energetic content of the food, and the ingested mass. This
latter factor can account for about 90% of the variation in some
animal’s HIF% (Secor 2009).
Among pinnipeds, HIF% is known to range from 4.7% to 16.8% of
GEI when animals are fed herring-only diets, 5.7% for pollock
diets, and from 11.5% to 13.0% for capelin diets (see Table 3 in
Rosen and Trites 1997). Estimates of heat increment of feeding for
the fur seals in our study (Table 3) are comparable with other
pinniped values reported by Rosen and Trites (1997). Overall HIF%
was significantly affected by lipid content in the diet, where the
diets with the higher lipid content required the least amount of
energy to digest (Fig. 3). This coincides with the fact that the
specific dynamic action of proteins is 32% and only 16% for lipids
(Forbes and Swift 1944). For example, the fur seal’s HIF% for the
capelin-only diet was significantly higher than the other diets,
most likely due to its high protein content. It should also be noted
that the ingested mass of the capelin diet (to attain an equivalent
GEI) was significantly higher than the other diets (Table 2); a factor
that has also been observed to increase HIF% overall.
Mixed-species diets were predicted to have lower heat incre-
ment of feeding costs than single-species diets (Forbes and Swift
1944). However, our test to quantify the cost of the HIF% of mixed-
species diets with a pinniped showed that mixed-species diets do
not lower the cost of HIF% (Fig. 3;Table 3). There were thus no
energetic savings due to eating more than one prey species to-
gether in terms of the costs of digestion.
The energy remaining in the energy transformation pathway
after accounting for the cost of heat increment of feeding is the
net energy, which ranged from 57.9% to 83.0% for the fur seals
(Table 3). The only other study to estimate NE% on a pinniped
was with harbour seals, which reported NE% of 80.0%–80.2%
(Ashwell-Erickson and Elsner 1981). While the high NE% for some
of the fur seal diets agrees with other carnivorous terrestrial
mammals (83.5% ± 5.3%) and birds (83.4%) (Robbins 1993), some of
our diets yielded surprisingly lower estimates.
The differences in net energy gain across our experimental di-
ets was influenced by their lipid content, such that the highest
NE% was for the herring diet and the lowest NE% was for the
capelin diet (Table 3). Similar to these findings, Fisher et al. (1992)
reported that walruses (Odobenus rosmarus (L., 1758)) feeding on
lipid-rich herring diets had a higher apparent digestibility of lip-
ids compared with those feeding on clam diets (who subsequently
had a higher energetic gain). The findings from our fur seals and
from walrus (Fisher et al. 1992) indicate that marine mammals are
particularly adapted for high-lipid diets, given that the energetic
digestibility and NE% return is significantly higher from fattier
diets than from leaner diets (Fig. 4).
Robbins (1993) recognized that the amount of food that an ani-
mal must ingests to meet a fixed energetic requirement should be
directly proportional to the losses in digestion and metabolism.
However, as demonstrated by our results, the amount of food that
an animal must consume to meet energetic requirements should
be reconsidered in terms of the net energy gain from the food
rather than in terms of the gross energy density. For example, the
estimated amounts of capelin required to meet the fur seals’ en-
ergy requirements based upon NE% were twice the amount of fish
(!6.0 kg) compared with estimates calculated on GEI alone (Fig. 5).
In contrast, the amount of herring required based on NE% would
only be 20% more than estimates based on GEI due to herring
being more digestible (Fig. 5). This example highlights how the
cost of energy transformation and the variability of the digestive
losses can exaggerate the differences in the gross quality of the
diet.
While the net energy gain by the fur seals was linearly related to
the gross energy intake, our study demonstrated that this rela-
tionship was driven by the lipid content of the diets, which was
less costly to process and provided the fur seals with a greater
energetic return per gram. These results emphasize how NE% de-
pends upon the chemical nature of prey and how the assimilation
of these individual components can impact an animal’s energy
budget. It is nonetheless also important to recognize that the
efficiency with which prey are assimilated is dynamic over the
Fig. 5. Changes in required mass intake (kg) to sustain maintenance energetic level (12 000 kJ·d
–1
) with the energy density (wet-basis) of the
eight experimental diets (kJ·d
–1
) tested in six captive female northern fur seals (Callorhinus ursinus). Data are presented as mean values of the
gross energy intake (GEI) in black circles and net energy gain (NE) in gray squares. The figure represents the differences between food intake
levels required to meet energetic needs calculated on gross energy content of prey vs. the food intake level that take into account digestive
energy losses (i.e., NE).
132 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
course of an animal’s life and that it depends on the energetic,
nutritional, and mineral and vitamin requirements of each stage
of development (Reid et al. 1980).
Changes on digestive efficiency due to diet mixing
Mixed-species diets, i.e., those that consist of various prey items
that differ qualitatively in their composition, are believed to pro-
vide a greater energetic benefit to the consumer than equivalent
single-species diets (Penry and Jumars 1987;Singer and Bernays
2003). However, this theory has only been tested on a few pin-
niped species. Experiments on captive harbour seals reported
30%–40% higher digestible energy (DE%) values from mixed-species
diets than from single-species diets (Trumble and Castellini 2005).
While a similar difference was reported for the metabolizable energy
(ME%) of Hawaiian monk seals (Goodman-Lowe et al. 1999), these
differences were actually the result of differences in GEI and not
proportional gains in ME%.
Our findings do not support the mixed-species diet theory. The
observed DE% and DMD% values of the mixed-species diets ap-
proximated the mean values of their single-species DE% and
DMD% constituents and did not surpass them (Figs. 1,2;Table 3).
Similar results have been reported for harbour seals where the
DE% of the mixed-species diet (30% herring, 30% capelin, 30% pol-
lock, and 10% market squid) did not appear to be different from
the herring-only diets (Yamamoto et al. 2009). While our results
contradict the belief that mixed-species diets provide a significant
advantage to fur seals to better assimilate either dry matter or
energy, this does not imply that diet diversity is not important.
Rather, the mixing of prey species is a fundamental component of
foraging strategies and has ecological implications for predator–
prey interactions, as well as ecological benefits overall (Stephens
and Krebs 1986;Singer and Bernays 2003;Barbosa and Castellanos
2005).
Effect of diet on metabolism
Northern fur seals are known to change their foraging behav-
iour and dietary intake between seasons as they migrate through
the North Pacific and Bering Sea (Kajimura 1984;Gentry and
Kooyman 1986;Gentry 2002). Significant seasonal differences in
resting metabolic rates (RMR) have also been identified in female
fur seals (Dalton et al. 2014), but it is unclear how, if at all, these
two seasonal changes are related. Some have suggested that diet
quality can potentially affect physiological processes not directly
associated with the digestive process (Cruz-Neto and Bozinovic
2004). This is consistent with Steller sea lions (Eumetopias jubatus
(Schreber, 1776)) significantly depressing their resting metabo-
lism when consuming insufficient levels of low-energy diets
(Rosen and Trites 1999). However, the dietary changes in our study
had no impact on the fur seal’s RMR. Nonetheless, mass-specific
RMR values for the fur seals were within the range of RMR of other
otariid and phocid seals (Miller 1978;Lavigne et al. 1986;Donohue
et al. 2000;Dalton et al. 2014).
The fur seals in our study had thermoregulation cost (TC) rates
that were 1.7 times higher when partially submerged in 2 °C water
compared with metabolic costs when resting in ambient air. Sim-
ilarly, Costa and Gentry (1986) found that the at-sea metabolic rate
of fur seals (lactating and nonlactating) was 1.8 times the on-shore
fasting metabolic rate. This increase in metabolism in 2 °C water
is similar to that reported for northern fur seal pups (Liwanag
2010;Rosen and Trites 2014). While the fur seals in our study
exhibited a significant metabolic increase while in cold water, TC
did not differ across the experimental diets.
While diet did not directly affect resting metabolic rates or
thermoregulation costs, it is possible that RMR and TC affect prey
consumption by wild fur seals. Seasonal changes in energy intake
requirements, i.e., due to seasonal requirements for growth or
activity, could induce changes in diet to better fulfill those needs.
In many marine mammal species, seasonal changes in energy
requirements coincide with natural predictable changes in prey
abundance or quality. Such is the case for pregnant or lactating
harp seals (Pagophilus groenlandicus (Erxleben, 1777)), sea otters
(Enhydra lutris (L., 1758)), and Atlantic spotted dolphins (Stenella
frontalis (G. Cuvier, 1829)) that vary their prey consumption accord-
ing to their reproductive condition (Ronald and Healey 1981;
Riedman et al. 1988;Malinowski and Herzing 2015).
Energetic implications of consuming pollock
In our study, we chose prey species that allowed us to investi-
gate the effects of prey composition (or quality) on digestibility
and net energy gain. Our findings suggest that when proper
amounts of fish are available to fur seals, the quality of the diet is
the major factor in determining the capacity of different prey to
meet the fur seals’ energetic requirements. These conclusions
support optimal diet model predictions, where Estabrook and
Dunham (1976) contended that small changes in the relative value
of prey can be more effective in changing a predator’s optimal diet
than small changes in the relative abundance of the potential
prey. Another model of optimal digestion further indicates that
the rate of efficiency of absorption of digestive products and the
rate of egestion of usable organic matter both increase with food
quality (Dade et al. 1990). This is important given that the quality
of different prey species available to fur seals and other top pred-
ators likely differs significantly with both time of year and devel-
opmental stage (Van Pelt et al. 1997;Logerwell and Schaufler 2005;
Vollenweider et al. 2011).
Measuring the digestive efficiency and net energy gain of north-
ern fur seals consuming various prey species is important for
evaluating whether the fur seal’s current diet in the wild is nega-
tively impacting their energy budgets. This entails using represen-
tative prey items that free-ranging fur seals may encounter or fish
of comparable compositions. For example, the pollock we used
was of relatively “high quality” and yielded a high net energy gain.
However, the quality of pollock that the fur seals encounter in the
wild differs considerably from this. Sinclair et al. (1994) reported
that 65% of the fish in the stomachs of northern fur seals consisted
of age-0 walleye pollock and another 31% were of age-1 pollock. It
appears that the pollock we fed our fur seals was about age-2
based on mean body size (Table 1;Buckley and Livingston 1994;
Kimura 2008). While whole-body proximate composition of young
walleye pollock fluctuates across seasons, on average, young pol-
lock range from 3.7 to 4.8 kJ·g
−1
, from 2.3% to 3.2% lipid, and from
14.1% to 15.4% protein (wet-basis) (Van Pelt et al. 1997;Logerwell
and Schaufler 2005;Vollenweider et al. 2011). These values are
similar to the proximate composition values of the capelin used in
our study, which had the highest HIF% cost and also exceedingly
low NE% gain, and would be classified as “low-quality” prey
(Table 2;Table 3).
Miller (1978) estimated that a wild fur seal of median mass
(23 kg) require a daily energy intake of !16 300 kJ·d
−1
. Combining
the digestibility results from our study and the documented qual-
ity of the pollock that fur seals are currently consuming, fur seals
foraging in the Bering Sea would need to consume !6.2 kg·d
−1
of
fish (!27% of their body mass) to obtain the required amount of
energy, of which at least 4 kg would be pollock. However, our
study suggests that free-ranging fur seals may be physically chal-
lenged to consume such amounts of fish, given that the fur seals
in our study refused to eat more than 4 kg·d
−1
of the lower quality
fish (capelin). However, further research is required to specifically
test such satiation limits (see Rosen et al. 2012;Calkins et al. 2013).
Our findings suggest that fur seals consuming primarily young
pollock of poor quality could be nutritionally stressed despite
there being a high biomass of pollock available to them. The
higher digestive cost of processing large amounts of food with
high-protein and low-energy content (such as young pollock)
would increase the likelihood of the fur seals gaining less than the
energy they ultimately require. Deriving sufficient energy from
Diaz Gomez et al. 133
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
low-quality prey sources becomes even more critical if stock den-
sities are diminished, but the potential effect of changes in prey
quality on nutritional status are independent of this consider-
ation. Our study adds support to the hypothesis that the domi-
nance of juvenile pollock in the current diet of wild fur seals in the
Bering Sea is likely detrimental to their population health and
reproductive fitness. Our results therefore have implications for
the management of northern fur seals and the adult pollock fish-
ery in the Eastern Bering Sea.
Conclusions
In summary, our study demonstrates that northern fur seals
attained significant differences in net energy gain across experi-
mental diets. These differences were driven by both the high en-
ergetic cost of protein digestion and the significantly higher
energetic return of fattier diets. This highlights the importance of
considering the individual digestion of each component of a diet
to understand how fur seals obtain energy from particular prey
items. Our study also highlights how differences in gross prey
quality between prey items become exaggerated during the
course of digestion. In addition, our results contradict the theory
that mixed-species diets provide an energetic advantage to fur
seals over single-species diets. Furthermore, there was no effect of
dietary changes on secondary metabolic costs of the fur seals,
such as resting metabolism or the cost of thermoregulation. Col-
lectively, our findings indicate that northern fur seals assimilate a
higher proportion of the energy contained in high-quality (lipid-
rich) prey, and a lower proportion of the energy contained in
lower quality prey such as pollock, particularly because of the
higher digestive cost associated with handling large amounts of
such low-quality prey. Our study therefore adds support to the
nutritional stress hypothesis by demonstrating the extent to
which changes in prey quality result in proportionally larger
changes in net energy gain. This has implications for population
health, reproductive fitness, and determining why fur seal popu-
lations are declining in the central Bering Sea.
Acknowledgements
We are grateful to the Marine Mammal Department of the Van-
couver Aquarium, as well as the marine mammal trainers and
veterinary staff who ensured data collection (notably B. Lasby, and
Dr. M. Haulena). We are also thankful for the help provided by the
technicians of the Marine Mammal Research Unit (notably
M. Davis) and the Marine Mammal Energetics and Nutrition Lab-
oratory. We particularly wish to thank I. Forster and M. Rowshan-
deli from the Marine Ecosystems and Aquaculture Division
(MEAD, Fisheries and Oceans Canada) for their expert guidance
with the laboratory analysis of samples and insightful discussions
and suggestions over the course of this study. We also thank
C.J. Brauner and students for their laboratory support, as well as
the anonymous reviewers for their constructive comments and
suggestions. Funding was provided by grants from the Natural
Sciences and Engineering Research Council of Canada and the
National Oceanographic and Atmospheric Administration to the
North Pacific Universities Marine Mammal Research Consortium
through the North Pacific Marine Science Foundation.
References
Alverson, D.L. 1992. A review of commercial fisheries and the Steller sea lion
(Eumetopias jubatus): the conflict arena. Rev. Aquat. Sci. 6(3–4): 203–256.
AOAC. 1990. Official methods of analysis. 15th ed. Association of Official Analyt-
ical Chemists, Arlington, Va.
Ashwell-Erickson, S.M., and Elsner, R. 1981. The energy cost of free existence of
Bering Sea harbour and spotted seals. In The Eastern Bering Sea shelf: ocean-
ography and resources. Edited by D.W. Hood and J.A. Calder. University of
Washington Press, Washington. pp. 869–899.
Barbiers, R.B., Vosburgh, L.M., Ku, P.K., and Ullrey, D.E. 1982. Digestive efficien-
cies and maintenance energy requirements of captive wild Felidae: cougar
(Felis concolor); leopard (Panthera pardus); lion (Panthera leo); and tiger (Panthera
tigris). J. Zoo. Anim. Med. 13(1): 32–37. doi:10.2307/20094560.
Barbosa, P., and Castellanos, I. 2005. Ecology of predator–prey interactions.
Oxford University Press, Inc., New York.
Best, R.C. 1985. Digestibility of ringed seals by the polar bear. Can. J. Zool. 63(5):
1033–1036. doi:10.1139/z85-155.
Blaxter, S.K.L. 1989. Energy metabolism in animals and man. Cambridge Univer-
sity Press, Cambridge, UK.
Bligh, E.G., and Dyer, W.J. 1959. A rapid method of total lipid extraction and
purification. Can. J. Biochem. Physiol. 37(8): 911–917. doi:10.1139/o59-099.
PMID:13671378.
Buckley, T.W., and Livingston, P.A. 1994. A bioenergetics model of walleye pollock
(Theragra chalcogramma)intheEasternBeringSea:structureanddocumentation.
U.S. Department of Commerce, NOAA Tech. Memo. NMFS-AFSC-37.
Budge, S.M., Iverson, S.J., and Koopman, H.N. 2006. Studying trophic ecology in
marine ecosystems using fatty acids: a primer on analysis and interpretation.
Mar. Mamm. Sci. 22(4): 759–801. doi:10.1111/j.1748-7692.2006.00079.x.
Calkins, D.G., Becker, E.F., and Pitcher, K.W. 1998. Reduced body size of female
Steller sea lions from a declining population in the Gulf of Alaska. Mar.
Mamm. Sci. 14(2): 232–244. doi:10.1111/j.1748-7692.1998.tb00713.x.
Calkins, D.G., Atkinson, S., Mellish, J.-A., Waite, J.N., and Carpenter, J.R. 2013.
The pollock paradox: juvenile Steller sea lions experience rapid growth on
pollock diets in fall and spring. J. Exp. Mar. Biol. Ecol. 441: 55– 61. doi:10.1016/
j.jembe.2013.01.011.
Call, K.A., and Ream, R.R. 2012. Prey selection of subadult male northern fur
seals (Callorhinus ursinus) and evidence of dietary niche overlap with adult
females during the breeding season. Mar. Mamm. Sci. 28(1): 1–15. doi:10.1111/
j.1748-7692.2011.00463.x.
Castellini, M.A. 1993. Is it food? Addressing marine mammal and seabird de-
clines. Workshop summary. Alaska Sea Grant Rep. No. 93-01, University of
Alaska, Fairbanks, Alaska.
Costa, D.P. 1988. Assessment of the impact of the California sea lion and north-
ern elephant seal on commercial fisheries. California Sea Grant: Biennial
Report of Completed Projects 1984-86 Rep. No. 024, University of California,
Santa Cruz. pp. 36–43.
Costa, D.P., and Gentry, R.L. 1986. Free-ranging energetics of northern fur seals.
In Fur seals: maternal strategies on land and sea. Edited by R.L. Gentry and G.L.
Kooyman. Princeton University Press, Princeton, N.J. pp. 79–101.
Crawley, M.J. 2007. The R book. Wiley, Chichester, UK.
Cruz-Neto, A.P., and Bozinovic, F. 2004. The relationship between diet quality
and basal metabolic rate in endotherms: insights from intraspecific analysis.
Physiol. Biochem. Zool. 77(6): 877–889. doi:10.1086/425187. PMID:15674763.
Dade, W.B., Jumars, P.A., and Penry, D.L. 1990. Supply-side optimization: maxi-
mizing absorptive rates. In Behavioral mechanisms of food selection. Edited
by R.N. Hughes. Springer, Berlin, Heidelberg. pp. 531–556.
Dalton, A.J.M., Rosen, D.A.S., and Trites, A.W. 2014. Broad thermal capacity
facilitates the primarily pelagic existence of northern fur seals (Callorhinus
ursinus). Mar. Mamm. Sci. 30(3): 994–1013. doi:10.1111/mms.12103.
Davison, R.P., Mautz, W.W., Hayes, H.H., and Holter, J.B. 1978. The efficiency of
food utilization and energy requirements of captive female fishers. J. Wildl.
Manage. 42(4): 811–821. doi:10.2307/3800771.
Decker, M.B., Hunt, G.L., Jr., and Byrd, G.V., Jr. 1995. The relationships among
sea-surface temperature, the abundance of juvenile walleye pollock (Theragra
chalcogramma), and the reproductive performance and diets of seabirds at the
Pribilof Islands, southeastern Bering Sea. In Climate change and northern
fish populations. Edited by R.J. Beamish. Can. Spec. Publ. Fish. Aquat. Sci. No.
121. pp. 425–437.
Dierauf, L., and Gulland, F.M.D. 2001. CRC Handbook of marine mammal
medicine: health, disease, and rehabilitation. 2nd ed. CRC Press, Boca Raton, Fla.
Donohue, M.J., Costa, D.P., Goebel, M.E., and Baker, J.D. 2000. The ontogeny of
metabolic rate and thermoregulatory capabilities of northern fur seal,
Callorhinus ursinus, pups in air and water. J. Exp. Biol. 203(6): 1003–1016. PMID:
10683160.
Estabrook, G.F., and Dunham, A.E. 1976. Optimal diet as a function of absolute
abundance, relative abundance, and relative value of available prey. Am. Nat.
110(973): 401–413. doi:10.1086/283076.
Fadely, B.S., Worthy, G.A.J., and Costa, D.P. 1990. Assimilation efficiency of
northern fur seals determined using dietary manganese. J. Wildl. Manag.
54(2): 246–251. doi:10.2307/3809037.
Fisher, K.I., Stewart, R.E.A., Kastelein, R.A., and Campbell, L.D. 1992. Apparent
digestive efficiency in walruses (Odobenus rosmarus)fedherring(Clupea harengus)
and clams (Spisula sp.). Can. J. Zool. 70(1): 30–36. doi:10.1139/z92-005.
Forbes, E.B., and Swift, R.W. 1944. Associative dynamic effects of protein, carbo-
hydrate and fat. Science, 99(2580): 476–478. doi:10.1126/science.99.2580.476.
PMID:17802323.
Forster, I.P. 1999. A note on the method of calculating digestibility coefficients
of nutrients provided by single ingredients to feeds of aquatic animals.
Aquacult. Nutr. 5(2): 143–145. doi:10.1046/j.1365-2095.1999.00082.x.
Galecki, A., and Burzykowski, T. 2013. Linear mixed-effects models using R: a
step-by-step approach. 2013 ed. Springer, New York.
Gentry, R.L. 2002. Northern fur seal (Callorhinus ursinus). In Encyclopedia of ma-
rine mammals. 2nd ed. Edited by W.F. Perrin, B. Würsig, and J.G.M. Thewissen.
Academic Press, New York. pp. 813–817.
Gentry, R.L., and Kooyman, G.L. 1986. Fur seals: maternal strategies on land and
at sea. Princeton University Press, Princeton, N.J.
Goodman-Lowe, G.D., Carpenter, J.R., and Atkinson, S. 1999. Assimilation effi-
134 Can. J. Zool. Vol. 94, 2016
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
ciency of prey in the Hawaiian monk seal (Monachus schauinslandi). Can. J. Zool.
77(4): 653–660. doi:10.1139/z98-238.
Grodzinski, W. 1975. Methods for ecological bioenergetics. Blackwell Scientific
Publications, Oxford, UK.
Iverson, S.J., Lang, S.L.C., and Cooper, M.H. 2001. Comparison of the Bligh and
Dyer and Folch methods for total lipid determination in a broad range of
marine tissue. Lipids, 36(11): 1283–1287. doi:10.1007/s11745-001-0843-0. PMID:
11795862.
Jobling, M. 1983. Towards an explanation of specific dynamic action (SDA). J. Fish
Biol. 23(5): 549–555. doi:10.1111/j.1095-8649.1983.tb02934.x.
Kajimura, H. 1984. Opportunistic feeding of the northern fur seal, Callorhinus
ursinus, in the eastern North Pacific Ocean and Eastern Bering Sea. U.S. De-
partment of Commerce, National Oceanic and Atmospheric Administration,
National Marine Fisheries Service, Washington, D.C.
Keiver, K.M., Ronald, K., and Beamish, F.W.H. 1984. Metabolizable energy re-
quirements for maintenance and faecal and urinary losses of juvenile harp
seals (Phoca groenlandica). Can. J. Zool. 62(5): 769–776. doi:10.1139/z84-110.
Kimura, D.K. 2008. Extending the von Bertalanffy growth model using explan-
atory variables. Can. J. Fish Aquat. Sci. 65(9): 1879–1891. doi:10.1139/F08-091.
Kleiber, M. 1975. The fire of life: an introduction to animal energetics.
R.E. Krieger Publishing Co., New York.
Lavigne, D.M., Barchard, W., Innes, S., and Øritsland, N.A. 1982. Pinniped bioen-
ergetics. In Mammals in the seas: small cetaceans, seals, sirens, and sea otters.
FAO, Rome. pp. 191–235.
Lavigne, D.M., Innes, S., Worthy, G.A.J., Kovacs, K.M., Schmitz, O.J., and
Hickie, J.P. 1986. Metabolic rates of seals and whales. Can. J. Zool. 64(2):
279–284. doi:10.1139/z86-047.
Lawson, J.W., Miller, E.H., and Noseworthy, E. 1997. Variation in assimilation
efficiency and digestive efficiency of captive harp seals (Phoca groenlandica) on
different diets. Can. J. Zool. 75(8): 1285–1291. doi:10.1139/z97-152.
Liwanag, H.E.M. 2010. Energetic costs and thermoregulation in northern fur seal
(Callorhinus ursinus) pups: the importance of behavioral strategies for thermal
balance in furred marine mammals. Physiol. Biochem. Zool. 83(6): 898–910.
doi:10.1086/656426. PMID:20950169.
Logerwell, E.A., and Schaufler, L.E. 2005. New data on proximate composition
and energy density of Steller sea lion (Eumetopias jubatus) prey fills seasonal
and geographic gaps in existing information. Aquat. Mamm. 31(1): 62–82.
doi:10.1578/AM.31.1.2005.62.
Malinowski, C.R., and Herzing, D.L. 2015. Prey use and nutritional differences
between reproductive states and age classes in Atlantic spotted dolphins
(Stenella frontalis) in the Bahamas. Mar. Mamm. Sci. 31(4): 1471–1493. doi:10.
1111/mms.12238.
Mårtensson, P.E., Nordøy, E.S., and Blix, A.S. 1994. Digestibility of krill (Euphausia
superba and Thysanoessa sp.) in minke whales (Balaenoptera acutorostrata) and
crabeater seals (Lobodon carcinophagus). Br. J. Nutr. 72(5): 713–716. doi:10.1079/
BJN19940073. PMID:7826994.
Merrick, R.L., Chumbley, M.K., and Byrd, G.V. 1997. Diet diversity of Steller sea
lions (Eumetopias jubatus) and their population decline in Alaska: a potential
relationship. Can. J. Fish Aquat. Sci. 54(6): 1342–1348. doi:10.1139/f97-037.
Miller, L.K. 1978. Energetics of the northern fur seal in relation to climate and
food resources of the Bering Sea. U.S. Marine Mammal Commission, Wash-
ington, D.C.
National Marine Fisheries Service. 1993. Conservation plan for the northern fur
seal (Callorhinus ursinus). National Marine Fisheries Service, Seattle, Wash.
National Research Council. 1996. The Bering Sea ecosystem. National Academies
Press, Washington, D.C.
Parsons, J.L. 1977. Metabolic studies on ringed seals (Phoca hispida). M.Sc. thesis,
University of Guelph, Guelph, Ont.
Penry, D.L., and Jumars, P.A. 1987. Modeling animal guts as chemical reactors.
Am. Nat. 129(1): 69–96. doi:10.1086/284623.
Pinheiro, J.C., and Bates, D.M. 2000. Mixed-effects models in S and S-PLUS.
Springer-Verlag, New York.
Pinheiro, J.C., Bates, D.M., DebRoy, S., et al. 2015. nlme: linear and nonlinear
mixed-effects models. R Package. Available from https://cran.r-project.org/
web/packages/nlme/ [accessed 1 October 2014].
Pitcher, K.W. 1990. Major decline in number of harbor seals, Phoca vitulina
richardsi, on Tugidak Island, Gulf of Alaska. Mar. Mamm. Sci. 6(2): 121–134.
doi:10.1111/j.1748-7692.1990.tb00234.x.
Pritchard, G.T., and Robbins, C.T. 1990. Digestive and metabolic efficiencies of
grizzly and black bears. Can. J. Zool. 68(8): 1645–1651. doi:10.1139/z90-244.
R Core Team. 2014. R: a language and environment for statistical computing.
R foundation for statistical computing, Vienna, Austria. Available from
https://www.r-project.org/.
Raubenheimer, D., and Simpson, S.J. 1999. Integrating nutrition: a geometrical
approach. In Proceedings of the 10th International Symposium on
Insect–Plant Relationships. Edited by S.J. Simpson, A.J. Mordue, and J. Hardie.
Springer, the Netherlands. pp. 67–82.
Reid, J.T., White, O.D., Anrique, R., and Fortin, A. 1980. Nutritional energetics of
livestock: some present boundaries of knowledge and future research needs.
J. Anim. Sci. 51(6): 1393–1415.
Riedman, M. 1990. The pinnipeds: seals, sea lions, and walruses. University of
California Press, Los Angeles.
Riedman, M., Staedler, M., and Estes, J.A. 1988. Behavioral and biological studies
on the California sea otter in the northern part of its range. Research plan,
Monterey Bay Aquarium, Monterey Bay, Calif.
Robbins, C.T. 1993. Wildlife feeding and nutrition. Academic Press, San Diego,
Calif.
Ronald, K., and Healey, P.J. 1981. Harp seal, Phoca groenlandica.In: Handbook of
marine mammals. Edited by S.H. Ridgway and J. Harrison. Academic Press,
London, UK. pp. 55–87.
Ronald, K., Keiver, K.M., Beamish, F.W.H., and Frank, R. 1984. Energy require-
ments for maintenance and faecal and urinary losses of the grey seal
(Halichoerus grypus). Can. J. Zool. 62(6): 1101–1105. doi:10.1139/z84-160.
Rosen, D.A.S. 2009. Steller sea lions Eumetopias jubatus and nutritional stress:
evidence from captive studies. Mammal Rev. 39(4): 284–306. doi:10.1111/j.1365-
2907.2009.00150.x.
Rosen, D.A.S., and Trites, A.W. 1997. Heat increment of feeding in Steller sea
lions, Eumetopias jubatus. Comp. Biochem. Physiol. Part A Physiol. 118(3): 877–
881. doi:10.1016/S0300-9629(97)00039-X.
Rosen, D.A.S., and Trites, A.W. 1999. Metabolic effects of low-energy diet on
Steller sea lions, Eumetopias jubatus. Physiol. Biochem. Zool. 72(6): 723–731.
doi:10.1086/316705. PMID:10603336.
Rosen, D.A.S., and Trites, A.W. 2000a. Pollock and the decline of Steller sea lions:
testing the junk-food hypothesis. Can. J. Zool. 78(7): 1243–1250. doi:10.1139/
z00-060.
Rosen, D.A.S., and Trites, A.W. 2000b. Digestive efficiency and dry-matter digest-
ibility in Steller sea lions fed herring, pollock, squid, and salmon. Can. J. Zool.
78(2): 234–239. doi:10.1139/z99-201.
Rosen, D.A.S., and Trites, A.W. 2014. Thermal limits in young northern fur seals,
Callorhinus ursinus. Mar. Mamm. Sci. 30(3): 1014–1028. doi:10.1111/mms.12097.
Rosen, D.A.S., Young, B.L., and Trites, A.W. 2012. Rates of maximum food intake
in young northern fur seals (Callorhinus ursinus) and the seasonal effects of
food intake on body growth. Can. J. Zool. 90(1): 61–69. doi:10.1139/z11-112.
Schneider, B.H., and Flatt, W.P. 1975. The evaluation of feeds through digestibil-
ity experiments. University of Georgia Press, Athens.
Secor, S.M. 2009. Specific dynamic action: a review of the postprandial metabolic
response. J. Comp. Physiol. B. 179(1): 1–56. doi:10.1007/s00360-008-0283-7.
PMID:18597096.
Sinclair, E., Loughlin, T., and Pearcy, W. 1994. Prey selection by northern fur
seals (Callorhinus ursinus) in the eastern Bering Sea. Fish Bull. (Wash., D.C.),
92(1): 144–156.
Singer, M.S., and Bernays, E.A. 2003. Understanding omnivory needs a behav-
ioral perspective. Ecology, 84(10): 2532–2537. doi:10.1890/02-0397.
Stephens, D.W., and Krebs, J.R. 1986. Foraging theory. Princeton University
Press, Princeton, N.J.
Stevens, C.E., and Hume, I.D. 1995. Comparative physiology of the vertebrate
digestive system. 2nd ed. Cambridge University Press, Cambridge, UK.
Swartzman, G.L., and Haar, R.T. 1983. Interactions between fur seal populations
and fisheries in the Bering Sea. Fish Bull. (Wash., D.C.), 81(1): 121–132.
Towell, R., Ream, R.R., Bengtson, J., and Sterling, J. 2014. 2014 northern fur seal
pup production and adult male counts on the Pribilof Islands, Alaska. Alaska
Fisheries Science Center, National Marine Mammal Laboratory. Available
from https://www.afsc.noaa.gov/nmml/PDF/2014-nfs-pup-counts-pribs.pdf
[accessed 9 November 2015].
Trites, A.W., and Donnelly, C.P. 2003. The decline of Steller sea lions Eumetopias
jubatus in Alaska: a review of the nutritional stress hypothesis. Mammal Rev.
33(1): 3–28. doi:10.1046/j.1365-2907.2003.00009.x.
Trites, A.W., and Larkin, P.A. 1996. Changes in the abundance of Steller sea lions
(Eumetopias jubatus) in Alaska from 1956 to 1992: how many were there?
Aquat. Mamm. 22(3): 153–166.
Trites, A.W., Calkins, D.G., and Winship, A.J. 2007. Diets of Steller sea lions
(Eumetopias jubatus) in Southeast Alaska, 1993–1999. Fish Bull. (Wash., D.C.),
105(2): 234–248.
Trumble, S.J., and Castellini, M.A. 2005. Diet mixing in an aquatic carnivore, the
harbour seal. Can. J. Zool. 83(6): 851–859. doi:10.1139/z05-069.
Van, Pelt, T.I., Piatt, J.F., Lance, B.K., and Roby, D.D. 1997. Proximate composition
and energy density of some north pacific forage fishes. Comp. Biochem.
Physiol. Part A Physiol. 118(4): 1393–1398. doi:10.1016/S0300-9629(97)00240-5.
Vollenweider, J.J., Heintz, R.A., Schaufler, L., and Bradshaw, R. 2011. Seasonal
cycles in whole-body proximate composition and energy content of forage
fish vary with water depth. Mar. Biol. 158(2): 413–427. doi:10.1007/s00227-010-
1569-3. PMID:24391256.
Withers, P.C. 1977. Measurement of VO
2
, VCO
2
, and evaporative water loss with
a flow-through mask. J. Appl. Physiol. 42(1): 120–123. PMID:833070.
Yamamoto, K.S., Carpenter, J.R., Atkinson, S., et al. 2009. Validation of acid
insoluble ash (AIA) as an internal biomarker for digestibility studies in har-
bor seals (Phoca vitulina). Am. Soc. Anim. Sci. 60: 199–203.
Zuur, A., Ieno, E.N., and Walker, N. 2009. Mixed effects models and extensions in
ecology with R. 2009 ed. Springer, New York.
Diaz Gomez et al. 135
Published by NRC Research Press
Can. J. Zool. Downloaded from www.nrcresearchpress.com by University of British Columbia on 02/17/16
For personal use only.
