Post by warsaw on Feb 22, 2012 11:47:20 GMT -9
Fast carnivores and slow herbivores: differential foraging
strategies among grizzly bears in the Canadian Arctic
Mark A. Edwards • Andrew E. Derocher •
Keith A. Hobson • Marsha Branigan •
John A. Nagy
Abstract Categorizing animal populations by diet can
mask important intrapopulation variation, which is crucial
to understanding a species’ trophic niche width. To test
hypotheses related to intrapopulation variation in foraging
or the presence of diet specialization, we conducted stable
isotope analysis (d13C, d15N) on hair and claw samples
from 51 grizzly bears (Ursus arctos) collected from 2003
to 2006 in the Mackenzie Delta region of the Canadian
Arctic. We examined within-population differences in the
foraging patterns of males and females and the relationship
between trophic position (derived from d15N measurements)
and individual movement. The range of d15N values
in hair and claw (2.0–11.0%) suggested a wide niche width
and cluster analyses indicated the presence of three
foraging groups within the population, ranging from nearcomplete
herbivory to near-complete carnivory. We found
no linear relationship between home range size and trophic
position when the data were continuous or when grouped
by foraging behavior. However, the movement rate of
females increased linearly with trophic position. We used
multisource dual-isotope mixing models to determine the
relative contributions of seven prey sources within each
foraging group for both males and females. The mean bear
dietary endpoint across all foraging groups for each sex fell
toward the center of the mixing polygon, which suggested
relatively well-mixed diets. The primary dietary difference
across foraging groups was the proportional contribution of
herbaceous foods, which decreased for both males and
females from 42–76 to 0–27% and 62–81 to 0–44%,
respectively. Grizzlies of the Mackenzie Delta live in
extremely harsh conditions and identifying within-population
diet specialization has improved our understanding of
varying habitat requirements within the population.
Discussion
Intrapopulation variation in the diet of grizzly bears in the
Mackenzie Delta region supported our prediction that different
foraging groups would be present in the population
and that trophic position would be related to movement
rate. Importantly, we found evidence that differences in the
foraging behaviors were largely driven by the proportional
use of herbaceous food types. We also found that the
movement rate of females increased with trophic position,
providing support for the idea that search effort increases
with the level of carnivory. Because our analysis was
conducted using hair and claw tissues, which have known
temporal integration periods, and the study area presented a
constant isotopic baseline, we were confident that the isotopic
variation observed in grizzly tissues resulted from
individual-level diet specialization among sampled bears.
Grizzly bears have a wide niche width across different
populations (Hilderbrand et al. 1996; Mowat and Heard
2006); however, we found that even within populations,
individual diets ranged across several trophic levels. We
used a single diet–tissue discrimination value of ?2.6% to
predict bear diets from hair and claws, but this experimentally
derived value was based on high-protein diets and
not the omnivorous diet of wild grizzlies. However, a
recent estimate for the d13C diet–tissue discrimination
factor corresponding to a low-protein, herbivorous diet was
similar at ?2.9 ± 0.9% (Alves-Stanley and Worthy 2009).
Based on the results of feeding trials of American black
bears (U. americanus) conducted by Hilderbrand et al.
(1996), which showed low between-individual isotopic
variance, we are confident that the intrapopulation variation
in d15N that we recorded for wild bears was representative
of actual isotopic differences in diet and trophic
differences in foraging behavior, and was not due to
inherent population variance.
Diet specialization within populations may be driven by
age- or sex-related factors or differences among ecologically
heterogeneous individuals (Schoener 1986; Lima and
Magnusson 1998; Shine et al. 2002; Bolnick et al. 2003).
Because hair and claw samples included in our analysis
were only from adult bears, we conclude that the observed
diet specialization was not related to ontogenetic shifts, as
bears matured from juveniles to adults (Polischuk et al.
2001; Newsome et al. 2006). Equally, because three foraging
groups were identified for both males and females,
we can also conclude that the occurrence of diet specialization
was not limited by sexual dimorphism or body size
and the ability to secure and handle prey (Selander 1966;
Brown and Lasiewsk 1972). What did differ for male and
female bears was the proportional contributions of the
seven source food types to their diets, with males potentially
exploiting more animal protein, be it terrestrial,
avian or aquatic. For sexually dimorphic species like
grizzly bears, the nutritional needs of larger males are
greater than those for females, which can result
in increased carnivory for males compared to females
(Jacoby et al. 1999). Conversely, because of their smaller
size and reduced nutritional needs, females can select
poorer quality yet adequate food resources (Rode et al.
2006). Therefore, we suggest that the individual diet
specialization and trophic level variation that we observed
for male and female bears resulted from interindividual
differences in prey availability and foraging ability among
bears. Similar patterns have been observed in several
species. Svanback and Bolnick (2007) demonstrated that
the level of diet specialization within a population of
three-spine sticklebacks (Gasterosteus aculeatus) may
vary depending on changing ecological attributes, with
diet specialization increasing with the time it took to
detect a change in prey availability. Urton and Hobson
(2005) reported that the variable foraging behavior
observed among wolves (Canis lupus) resulted from differences
in the availability of foods specific to home
ranges, which resulted in isotopic variation and dietary
specialization among individuals. Also, a single population
of ring-tailed lemurs (Lemur catta) was divided into
three foraging groups based on diet specialization in their
use of available forage in habitats that ranged from forested
to open (Loudon et al. 2007).
It follows that along with variation in prey availability,
diet specialization may result from phenotypical trade-offs
due to individual-level morphological, physiological, and/
or behavioral attributes and experiences that allow different
individuals to be more effective at exploiting one type of
prey and less effective at exploiting another (Robinson
et al. 1996; Svanback and Bolnick 2005, 2007). The use of
all resources by all members within a population’s niche
width is thus reflected in these trade-offs to the extent that
subsets of diet specialization develop within the population
(Bolnick et al. 2003). If bears are foraging optimally, they
should maximize energy intake and may even ignore preferred
prey items when searching and handling time make
it more economical to seek alternate prey items (MacArthur
and Pianka 1966; Stephens and Krebs 1986). Optimal
foraging theory, therefore, provides added explanation for
individual-level diet specialization if individuals differ
phenotypically in their ability to exploit alternate prey
types, and these individuals are able to add different prey
types more effectively (Bolnick et al. 2003; Svanback and
Bolnick 2005). Therefore, where prey density is low, two
or more phenotypically different groups of consumers that
rely on divergent alternate prey types will increase the
population’s niche width (Svanback and Bolnick 2007).
The Mackenzie Delta is characterized by low productivity
and low availability of high-quality protein sources
(Hilderbrand et al. 1999), and bear densities in this region
are some of the lowest in North America (Nagy and
Haroldson 1990). For an omnivore like the grizzly bear
with a broad niche width and a high level of phenotypic
variation, it is not surprising that individual-level diet
specialization is more pronounced than for species with
narrower niche widths (Lister 1976; Roughgarden 1979;
Araujo et al. 2007). High levels of diet specialization and
patterns of greater intrapopulation variation in foraging
have been reported for other taxa with such characteristically
wide niche widths (Lister 1976; Roughgarden 1979;
Werner and Sherry 1987; Estes et al. 2003; Svanback and
Bolnick 2007). In a review by Bolnick et al. (2003), the
high occurrence of intrapopulation variation in foraging
behavior across taxa suggests that the presence of diet
specialization within populations may be a more general
pattern than previously considered.
Although we failed to find a significant relationship
between home range size and trophic position for the bears
of the Mackenzie Delta, our results did provide support for
the suggestion of Gittleman and Harvey (1982) and Mace
et al. (1983) that omnivores with a higher proportion of
animal protein in their diet should have higher rates of
movement as they search for low-density animal prey. More
herbivorous animals should have lower movement rates,
which suggest that they are able to meet their nutritional
needs by foraging slowly through a landscape of herbaceous
improving our ability to refine protocols to meet the needs
of focal groups and the population as a whole.
biology.ufl.edu/courses/zoo6927/2012Spring/bjorndal/readings/Specialization%20Edwards%20et%20al%202010.pdf
strategies among grizzly bears in the Canadian Arctic
Mark A. Edwards • Andrew E. Derocher •
Keith A. Hobson • Marsha Branigan •
John A. Nagy
Abstract Categorizing animal populations by diet can
mask important intrapopulation variation, which is crucial
to understanding a species’ trophic niche width. To test
hypotheses related to intrapopulation variation in foraging
or the presence of diet specialization, we conducted stable
isotope analysis (d13C, d15N) on hair and claw samples
from 51 grizzly bears (Ursus arctos) collected from 2003
to 2006 in the Mackenzie Delta region of the Canadian
Arctic. We examined within-population differences in the
foraging patterns of males and females and the relationship
between trophic position (derived from d15N measurements)
and individual movement. The range of d15N values
in hair and claw (2.0–11.0%) suggested a wide niche width
and cluster analyses indicated the presence of three
foraging groups within the population, ranging from nearcomplete
herbivory to near-complete carnivory. We found
no linear relationship between home range size and trophic
position when the data were continuous or when grouped
by foraging behavior. However, the movement rate of
females increased linearly with trophic position. We used
multisource dual-isotope mixing models to determine the
relative contributions of seven prey sources within each
foraging group for both males and females. The mean bear
dietary endpoint across all foraging groups for each sex fell
toward the center of the mixing polygon, which suggested
relatively well-mixed diets. The primary dietary difference
across foraging groups was the proportional contribution of
herbaceous foods, which decreased for both males and
females from 42–76 to 0–27% and 62–81 to 0–44%,
respectively. Grizzlies of the Mackenzie Delta live in
extremely harsh conditions and identifying within-population
diet specialization has improved our understanding of
varying habitat requirements within the population.
Discussion
Intrapopulation variation in the diet of grizzly bears in the
Mackenzie Delta region supported our prediction that different
foraging groups would be present in the population
and that trophic position would be related to movement
rate. Importantly, we found evidence that differences in the
foraging behaviors were largely driven by the proportional
use of herbaceous food types. We also found that the
movement rate of females increased with trophic position,
providing support for the idea that search effort increases
with the level of carnivory. Because our analysis was
conducted using hair and claw tissues, which have known
temporal integration periods, and the study area presented a
constant isotopic baseline, we were confident that the isotopic
variation observed in grizzly tissues resulted from
individual-level diet specialization among sampled bears.
Grizzly bears have a wide niche width across different
populations (Hilderbrand et al. 1996; Mowat and Heard
2006); however, we found that even within populations,
individual diets ranged across several trophic levels. We
used a single diet–tissue discrimination value of ?2.6% to
predict bear diets from hair and claws, but this experimentally
derived value was based on high-protein diets and
not the omnivorous diet of wild grizzlies. However, a
recent estimate for the d13C diet–tissue discrimination
factor corresponding to a low-protein, herbivorous diet was
similar at ?2.9 ± 0.9% (Alves-Stanley and Worthy 2009).
Based on the results of feeding trials of American black
bears (U. americanus) conducted by Hilderbrand et al.
(1996), which showed low between-individual isotopic
variance, we are confident that the intrapopulation variation
in d15N that we recorded for wild bears was representative
of actual isotopic differences in diet and trophic
differences in foraging behavior, and was not due to
inherent population variance.
Diet specialization within populations may be driven by
age- or sex-related factors or differences among ecologically
heterogeneous individuals (Schoener 1986; Lima and
Magnusson 1998; Shine et al. 2002; Bolnick et al. 2003).
Because hair and claw samples included in our analysis
were only from adult bears, we conclude that the observed
diet specialization was not related to ontogenetic shifts, as
bears matured from juveniles to adults (Polischuk et al.
2001; Newsome et al. 2006). Equally, because three foraging
groups were identified for both males and females,
we can also conclude that the occurrence of diet specialization
was not limited by sexual dimorphism or body size
and the ability to secure and handle prey (Selander 1966;
Brown and Lasiewsk 1972). What did differ for male and
female bears was the proportional contributions of the
seven source food types to their diets, with males potentially
exploiting more animal protein, be it terrestrial,
avian or aquatic. For sexually dimorphic species like
grizzly bears, the nutritional needs of larger males are
greater than those for females, which can result
in increased carnivory for males compared to females
(Jacoby et al. 1999). Conversely, because of their smaller
size and reduced nutritional needs, females can select
poorer quality yet adequate food resources (Rode et al.
2006). Therefore, we suggest that the individual diet
specialization and trophic level variation that we observed
for male and female bears resulted from interindividual
differences in prey availability and foraging ability among
bears. Similar patterns have been observed in several
species. Svanback and Bolnick (2007) demonstrated that
the level of diet specialization within a population of
three-spine sticklebacks (Gasterosteus aculeatus) may
vary depending on changing ecological attributes, with
diet specialization increasing with the time it took to
detect a change in prey availability. Urton and Hobson
(2005) reported that the variable foraging behavior
observed among wolves (Canis lupus) resulted from differences
in the availability of foods specific to home
ranges, which resulted in isotopic variation and dietary
specialization among individuals. Also, a single population
of ring-tailed lemurs (Lemur catta) was divided into
three foraging groups based on diet specialization in their
use of available forage in habitats that ranged from forested
to open (Loudon et al. 2007).
It follows that along with variation in prey availability,
diet specialization may result from phenotypical trade-offs
due to individual-level morphological, physiological, and/
or behavioral attributes and experiences that allow different
individuals to be more effective at exploiting one type of
prey and less effective at exploiting another (Robinson
et al. 1996; Svanback and Bolnick 2005, 2007). The use of
all resources by all members within a population’s niche
width is thus reflected in these trade-offs to the extent that
subsets of diet specialization develop within the population
(Bolnick et al. 2003). If bears are foraging optimally, they
should maximize energy intake and may even ignore preferred
prey items when searching and handling time make
it more economical to seek alternate prey items (MacArthur
and Pianka 1966; Stephens and Krebs 1986). Optimal
foraging theory, therefore, provides added explanation for
individual-level diet specialization if individuals differ
phenotypically in their ability to exploit alternate prey
types, and these individuals are able to add different prey
types more effectively (Bolnick et al. 2003; Svanback and
Bolnick 2005). Therefore, where prey density is low, two
or more phenotypically different groups of consumers that
rely on divergent alternate prey types will increase the
population’s niche width (Svanback and Bolnick 2007).
The Mackenzie Delta is characterized by low productivity
and low availability of high-quality protein sources
(Hilderbrand et al. 1999), and bear densities in this region
are some of the lowest in North America (Nagy and
Haroldson 1990). For an omnivore like the grizzly bear
with a broad niche width and a high level of phenotypic
variation, it is not surprising that individual-level diet
specialization is more pronounced than for species with
narrower niche widths (Lister 1976; Roughgarden 1979;
Araujo et al. 2007). High levels of diet specialization and
patterns of greater intrapopulation variation in foraging
have been reported for other taxa with such characteristically
wide niche widths (Lister 1976; Roughgarden 1979;
Werner and Sherry 1987; Estes et al. 2003; Svanback and
Bolnick 2007). In a review by Bolnick et al. (2003), the
high occurrence of intrapopulation variation in foraging
behavior across taxa suggests that the presence of diet
specialization within populations may be a more general
pattern than previously considered.
Although we failed to find a significant relationship
between home range size and trophic position for the bears
of the Mackenzie Delta, our results did provide support for
the suggestion of Gittleman and Harvey (1982) and Mace
et al. (1983) that omnivores with a higher proportion of
animal protein in their diet should have higher rates of
movement as they search for low-density animal prey. More
herbivorous animals should have lower movement rates,
which suggest that they are able to meet their nutritional
needs by foraging slowly through a landscape of herbaceous
improving our ability to refine protocols to meet the needs
of focal groups and the population as a whole.
biology.ufl.edu/courses/zoo6927/2012Spring/bjorndal/readings/Specialization%20Edwards%20et%20al%202010.pdf