1. Introduction
Predator-prey relations are an integral force in many communities and
are often key to understanding how communities function. Predators
occupy a wide spectrum of foraging strategies, ranging from generalists
to specialists. Individuals using a smaller subset of resources than the
population as a whole are defined as individual specialists (Van Valen,
1965) whereas individuals consuming a wider range of resources than used
on average by the population are defined as generalists (Hanski,
Hansson, & Henttonen, 1991). In some cases, preying on one species may
preclude consumption of a different species as there may be trade-offs
in skills required to utilize different resources (Arthur et al. 2016;
Wilson & Yoshimura 1994). Inter-individual differences in resource use
that are transient and the result of short-term choices in habitat or
hunting strategies are best described as intrapopulation feeding
diversity while permanent differences between individuals based on sex,
size, or personality are better described as individual specialization
(Van Valen 1965). Both types of inter-individual differences in resource
use need to be examined to understand predator-prey interactions.
The level of individual specialization and/or intrapopulation feeding
diversity can affect food web dynamics, responses to changes in prey
availability, and the accuracy of predictive models (Bolnick et al.,
2003). For example, in a population of bluegill sunfish Lepomis
macrochirus prior experience foraging on a single prey type increases
the likelihood of an individual using that resource, even when another
resource becomes more profitable (Werner, Mittelbach, & Hall,1981).
Further, theoretical models predict that changes in prey abundance in a
system with highly specialized individuals (i.e. slow to switch
preferred prey) are much more likely to have chaotic dynamics (Abrams &
Matsuda, 2004). In general, dynamics can be sensitive to small
variations in the speed of predators changing prey preferences (Abrams
& Matsuda, 2004), which would be affected by the predator’s level of
specialization. More generally, diversification within a population can
have significant impacts on ecosystems functions, such as prey community
structure (Harmon et al., 2009). In addition, differences in a single
species’ population structure can have larger impacts on community
composition than differences between species (Rudolf & Rasmussen,
2013). These findings imply that differences between individuals of the
same species can be important drivers of ecosystem functions (Harmon et
al., 2009, Rudolf & Rasmussen, 2013). Thus, including metrics of
variation in foraging decisions between individuals of the same
population in ecosystem studies provides a more clear and accurate
description of the system and ignoring them can be an oversimplification
of the ecological interactions in the community (Araújo, Bolnick, &
Layman, 2011; Bolnick et al., 2003; Bolnick et al., 2011; Dall, Bell,
Bolnick, & Ratnieks, 2012). Unfortunately, most foraging studies do not
describe the level of intraspecific variation in the predator
population.
Variation in foraging decisions within populations of predators is
difficult to describe empirically because they require observing a large
number of predation events in multiple individuals across many
ecological contexts. The empirical problems are even greater when
studying predators that forage in environments where it is difficult to
directly observe predation events (e.g., marine environments) and that
prey on a large diversity of taxonomically similar prey species that
make it difficult to determine which species has been consumed. Here, we
demonstrate how the application of an individual diet specialization
metric to a large set of molecular prey barcoding data from scat can be
used to describe intrapopulation feeding diversity by examining the
short-term variation in individual foraging decisions in a marine
predator. Our analysis allowed us to explore 1) correlations of
individual diet diversity with the sex of the predator and time of year
in which the predation occurred, and 2) to test whether short-term diet
diversity was related to the consumption of particular prey species and
their preferred habitat.
Harbor seals (Phoca vitulina , Linnaeus 1758) have the largest
worldwide distribution of any pinniped in coastal areas (Teilmann &
Galatius, 2018) and appear to have reached carrying capacity in the
Salish Sea (Jeffries, Huber, Calambokidis, & Laake, 2003; Olesiuk,
2009). Because harbor seals are abundant in the ecosystem and feed on a
wide range of species, they have significant impacts on prey populations
(Howard, Lance, Jeffries, & Acevedo-Gutiérrez, 2013; Lance, Chang,
Jeffries, Pearson, & Acevedo-Gutiérrez, 2012; Olesiuk, Bigg, Ellis,
Crockford, & Wigen, 1990). Some of their prey species are of high
conservation concern, such as Pacific salmon Oncorhynchus spp.,
rockfish Sebastes spp., and Pacific herring (Clupea
pallasii pallasii, Valenciennes, 1847) (Bjorland et al., 2015;
Bromaghin et al., 2013; Lance et al., 2012). . Harbor seal’s effect onOncorhynchus spp. is also of interest because they eat both
juvenile and adult individuals (Thomas, Nelson, Lance, Deagle, &
Trites, 2017). Eating juveniles may have considerable impacts on
populations of Chinook (Oncorhynchus tshawytscha, Walbaum 1792),
coho (Oncorhynchus kisutch, Walbaum 1792), and steelhead
(Oncorhynchus mykiss, Walbaum 1792), as survival during the first
several months at sea is believed to be the primary factor limiting
population abundance and productivity (Beamish et al., 2010; Kendall,
Marston, & Klungle, 2017; Neville, Beamish, & Chittenden, 2015).
Due to the large diversity of prey species that harbor seal populations
eat, the species has historically been considered a generalist predator
(Teilmann & Galatius, 2018). However, prey composition and foraging
dive behavior of harbor seals in the Salish Sea vary relative to
habitat, sex, and time of year (Lance et al., 2012; Olesiuk et al.,
1990; Wilson, Lance, Jeffries, & Acevedo-Gutiérrez, 2014). Harbor seals
also eat different types of prey depending on the type of environment in
which they forage. Scat samples from haul-outs located in estuaries have
higher prey diversity than those coming from outside estuaries (Lance et
al., 2012; Luxa & Acevedo-Gutiérrez, 2013). Further, males and females
consume different prey (Bjorland et al., 2015; Schwarz et al., 2018) and
have different foraging dive patterns (Wilson et al., 2014).
Specifically, females frequently perform longer and deeper foraging
dives than males, and more commonly consume benthic species (Schwarz et
al., 2018; Wilson et al., 2014).
These traits of high abundance and differences in diet and foraging
patterns between males and females, suggest that harbor seals could
display intrapopulation feeding diversity.. As such, ecosystem dynamics
with regard to the effect of harbor seals on prey species are likely
more complex than described in current models of the system which assume
consistent generalized behavior (e.g. Chasco et al. 2017, Howard et al.
2013). As harbor seals are the most abundant mammalian predator in the
Salish Sea, and prey on many species of economic and conservation
concern, an accurate understanding of their role in ecosystem dynamics
is important, for which high quality diet data are required. While
current bioenergetics based models are useful descriptors of harbor seal
consumption (e.g. Chasco et al. 2017, Howard et al. 2013), they can be
improved by including the effects of different foraging strategies
across sexes and individuals.
Obtaining high quality diet data from large, mobile organisms, such as
marine mammals, can be costly and time consuming (Rothstein, McLaughlin,
Acevedo-Gutiérrez, & Schwarz, 2017). Analysis of prey contents in scat
via metabarcoding is a relatively cheap, non-invasive, and time
efficient way to obtain large sample sizes with species-level taxonomic
resolution (Deagle et al., 2005; Deagle et al., 2019; Rothstein et al.,
2017; Tollit et al., 2009). However, to our knowledge, these molecular
techniques have not previously been used to quantify intrapopulation
feeding diversity at large spatial and temporal scales.
Here, we use molecular barcoding of prey DNA from scat in a novel way to
examine intrapopulation feeding diversity, in harbor seals by answering
the following questions: 1. How do the factors Sex, Time of year,
Location, and Year affect cross sectional estimation of a specialization
metric? 2. What prey items correlate with high levels of specialization
relative to sex and environment? To answer these questions, we collected
and analyzed scat from wild harbor seals in the Salish Sea. Diet of
harbor seals was determined from the scat using both molecular and
traditional techniques. Sex of the depositor was determined using
molecular techniques. Diet data were analyzed with a proportional
similarity index (Bolnick, Yang, Fordyce, Davis, & Svanbäck, 2002) to
describe the variation in individual foraging decisions in harbor seals.