, Volume 178, Issue 1, pp 17–29 | Cite as

The cost of reproduction: differential resource specialization in female and male California sea otters

  • Emma A. Elliott SmithEmail author
  • Seth D. Newsome
  • James A. Estes
  • M. Tim Tinker
Special Topic: Individual-Level Niche Specialization


Intraspecific variation in behavior and diet can have important consequences for population and ecosystem dynamics. Here, we examine how differences in reproductive investment and spatial ecology influence individual diet specialization in male and female southern sea otters (Enhydra lutris nereis). We hypothesize that greater reproductive constraints and smaller home ranges of females lead to more pronounced intraspecific competition and increased specialization. We integrate stable carbon (δ13C) and nitrogen (δ15N) isotope analysis of sea otter vibrissae with long-term observational studies of five subpopulations in California. We define individual diet specialization as low ratios of within-individual variation (WIC) to total population niche width (TNW). We compare isotopic and observational based metrics of WIC/TNW for males and females to data on population densities, and movement patterns using both general linear and linear mixed-effects models. Consistent with our hypothesis, increasing population density is associated with increased individual diet specialization by females but not by males. Additionally, we find the amount of coastline in a sea otter’s home range positively related with individual dietary variability, with increased range span resulting in weaker specialization for both males and females. We attribute our results to sex-based differences in movement, with females needing to specialize in their small ranges to maximize energy gain, and posit that the paradigm of individual prey specialization in sea otters with increased intraspecific competition may be a pattern driven largely by females. Our work highlights a potentially broader role of sex in the mechanistic pressures promoting and maintaining diet specialization.


Individual diet specialization Stable isotopes Reproductive investment Spatial ecology Enhydra lutris 


A recent surge of studies has demonstrated that individual-level variation in dietary niche is both widespread and ecologically significant (Bolnick et al. 2003; Kondoh 2003; Araújo et al. 2011). Such specialization results from a complex interplay between ecological and morphological characteristics of consumers and resources, and requires several conditions. First, potential prey species must be both diverse and abundant enough to provide the “ecological opportunity” for a consumer to diversify its diet (Darimont et al. 2009; Araújo et al. 2011). Second, a strong ecological or physiological pressure is needed to induce individuals to stray away from preferred prey and focus on alternative resources (Bolnick et al. 2003; Araújo et al. 2011). Studies consistently find intraspecific competition increases the degree of individual specialization (Tinker et al. 2008a; Svanbäck and Bolnick 2007; Svanbäck and Persson 2009), while both interspecific competition and predation appear to dampen it (Peacor and Pfister 2006; Eklov and Svanbäck 2006; Bolnick et al. 2010). Lastly, behavioral, physiological, and/or morphological variation that create inter-individual differences in the ability to capture, process, or assimilate prey must be maintained over time (Bolnick et al. 2003; Araújo et al. 2011).

Sex can be an important potential source of intraspecific niche variation (Clutton-Brock 2007; Ruckstuhl 2007). In mammals, males and females tend to differ in size, shape, behavior, and most notably, reproductive investment. For example, most mammals are sexually dimorphic, with males typically larger than females (Ralls 1977; Lindenfors et al. 2007). Additionally, females necessarily must invest more resources in offspring, owing to the fundamental limitations of anisogamous reproduction (Bateman 1948; Trivers 1972; Clutton-Brock 2007). Mammalian lactation, for example, has been shown to double or triple postpartum energy demands (Millar 1977; Williams et al. 2007; Thometz et al. 2014). Consequently females are typically the “choosy” sex, as they risk losing substantial time and resources by mating with a “low quality” male (Trivers 1972; Lima and Dill 1990; Charnov 1993; Clutton-Brock 1991, 2007). Conversely, males are limited in reproductive success only by the number of individuals they successfully mate with, and, consequently, often compete strongly for access to females and provide little or no parental care (Bateman 1948; Clutton-Brock 1991, 2007).

With males and females experiencing fundamentally different evolutionary and energetic pressures, there is reason to expect different responses to the ecological conditions that contribute to the development of individual diet specialization. Because females are tied to their young—either physically during gestation, or by postpartum parental care—they often experience restrictions in movement and range of behavior (Trivers 1972; Kleiman 1977; Ralls 1977; Trillmich 1990; Grigione et al. 2002; Loveridge et al. 2009). Lower investment in offspring by males means they can afford to engage in such behaviors as movement between habitats/populations, and immigration into novel territories (Clutton-Brock 1991; Clutton-Brock et al. 2002). This sex-specific life history limitation is relatively common among mammals: in marine carnivores such as California sea lions (Zalophus californianus) income breeding females are restricted to foraging near rookeries while males wander (Trillmich 1990). Similar patterns hold for terrestrial carnivores such as African lions (Panthera leo), where non-dominant males roam great distances while females remain in smaller territories with their cubs (Stander 1991; Loveridge et al. 2009), and pumas (Puma concolor), where female home range shrinks during periods of cub care (Grigione et al. 2002). Thus, females may be unable to escape ecological pressures such as high degrees of intraspecific competition, whereas males may be free to move to areas where resources are more abundant (Clutton-Brock et al. 2002). Given the importance of intraspecific competition in driving patterns of individual diet specialization, it is reasonable to expect that females may be under stronger selection for specialization than males.

California sea otters (Enhydra lutris nereis) provide a useful system for exploring these concepts. Populations of sea otters differ significantly in density of individuals and degree of intraspecific resource competition, with corresponding variation in the degree of individual diet specialization (Estes et al. 2003; Tinker et al. 2008a). When sea otter population densities are low and food is abundant, individual adult females tend to target the most energy-rich prey and have similar, overlapping diets (Tinker et al. 2008a). As competitive pressure and thus resource limitation increases, these individuals tend to specialize on a few unique prey, so that dietary diversity of the population increases while dietary niche width of individuals remains small and may even decline (Tinker et al. 2008a, 2012). Additionally, male and female sea otters exhibit different patterns of movement and reproductive investment (Jameson 1989; Ralls et al. 1996; Tinker et al. 2008b). Females are solely responsible for parental care and are either in a continuous state of pregnancy or provisioning a dependent pup, which can increase daily energetic requirements by up to 96 % over the lactation period (Thometz et al. 2014). Adult female sea otters are thus energetically constrained by reproduction and show high site fidelity, rarely moving more than 20 km from the center of their home range [Loughlin 1980; Tinker et al. 2008b; US Geological Survey (USGS), unpublished data]. Males are not constrained by parental care, and often travel distances of 50–200 km throughout the subspecies range (Jameson 1989; Ralls et al. 1996; Tinker et al. 2008b). These differences in reproductive investments and movement patterns among the sexes may well contribute to differences in the degree of specialization.

Here, we explore the hypothesis that individual diet specialization in sea otters differs between the sexes. More specifically, we hypothesize that due to their inability to move away from conspecific competitors, female sea otters should be more sensitive to intraspecific pressures and thus display greater degrees of prey specialization in resource-limited areas. In contrast, males are less constrained energetically by reproduction and may have a reduced response to intraspecific pressures and exhibit a wider range of foraging strategies. We use a combination of carbon (δ13C) and nitrogen (δ15N) isotope values of sub-sampled vibrissae and long-term observational studies of five sea otter subpopulations in central and southern California to characterize individual diet specialization in males and females. Our work provides further insight into the topic of individual diet specialization by highlighting the importance of sexual selection in indirectly shaping these ecological patterns.

Materials and methods

We characterized the degree of individual diet specialization by males and females for five subpopulations of southern sea otters (Enhydra lutris nereis) along the coast of central and southern California. To maximize our statistical power in evaluating broad ecological patterns between the sexes we employed two independent metrics of specialization: field observations and stable isotope analysis. Our five subpopulations encompass most of the sea otter’s distribution in California, from Monterey Bay (MBY) in the north, through Big Sur (BSR) and San Luis Obispo County (SLO) in the center, and Santa Barbara Channel (SBC) and San Nicolas Island (SNI) in the south (Fig. 1). Physical habitats were generally similar at all sites, including rocky reefs, patchy areas of soft sediment, and dominated by canopy-forming kelps (Tinker et al. 2012, see SOM). Long-term records from each site included data on population densities, foraging behavior, diets and movements/home range size (USGS, unpublished data; Kage 2004; Tinker et al. 2006, 2008a, b; Newsome et al. 2009, 2010; Caswell et al. 2011; Staedler 2011; Tinker et al. 2012). We integrated these data sets with isotopic analysis of sea otter vibrissae sampled from animals at all five sites (Table 1). Details of the procedures for the capture, handling, radio-tagging, and telemetric/observational monitoring of study animals are provided in Tinker et al. (2006, 2008a); all collection activities were authorized by federal, state, and institutional permits issued to M. T. Tinker.
Fig. 1

Geographic location of study sites. The five sites are located along the Pacific coast of North America and represent localities that have been occupied by sea otters for different periods of time. Range limits are based on 2013 survey data. Site codes are in blue boxes and approximate central location of sites are marked by stars. Note the substantial differences in density among the different locations. The most recently occupied site is Santa Barbara Channel (SBC), occupied by females only since 2013. MBY Monterey Bay, BSR Big Sur, SLO San Luis Obispo County, SNI San Nicolas Island (color figure online)

Table 1

Summary of sample sizes and collection dates for observational and isotopic data


Vibrissae (isotopic analysis)

Dietary data

Range span










Monterey Bay (MBY)










Big Sur (BSR)










San Luis Obispo County (SLO)










Santa Barbara Channel (SBC)







San Nicolas Island (SNI)







Population density estimates obtained from all sites in 2013. N represents the number of individuals, not within-individual sampling. For example, with vibrissae, n represents the number of vibrissae from different individuals, not the total number of sub-sampled vibrissae segments

Collection of density, movement and observational dietary data

Where possible we obtained data on population densities, movement patterns and prey selection of males and females for each study location (Table 1). Population censuses are conducted each year by the US Geological Survey, California Department of Fish and Wildlife, Monterey Bay Aquarium, University of California-Santa Cruz and the US Fish and Wildlife Service using both aerial and shore-based survey techniques as developed by Estes and Jameson (1988). Here we used 2013 spring census (early May) data for all sites (Estes and Jameson 1988; Tinker et al. 2006). We obtained movement data from long-term observations of individuals for all sites except SNI, where appropriate data were not available (Table 1). We described spatial use patterns in terms of “linear range span,” defined as the distance of coastline (kilometers) encompassed by an individual’s movements within its annual home range. Individual home ranges were estimated by fitting a one-dimensional kernel density function to coastal resight locations (see Laidre et al. 2009 for details) and we calculated range span as the distance of coastline within the 90 % utilization probability contour. At four sites (MBY, BSR, SLO and SNI) we made shore-based observations of sea otter foraging bouts (as described in Tinker et al. 2012) in order to estimate diet composition, collecting longitudinal data from individual sea otters over 2- to 5-year periods (Table 1). At the time of this study, observational dietary data were not available from the SBC site as data collection was still in progress.

Stable isotope analysis

Previous work has demonstrated that stable isotope analysis of sea otter vibrissae is a robust proxy for long-term observational dietary data and an effective tool for studying individual- and population-level diet variation in this system (Newsome et al. 2009, 2010, 2012, 2015). Sea otters are particularly well suited to an isotopic approach as they consume a wide variety of functionally distinct macroinvertebrate prey in a system fueled by two sources of primary production (phytoplankton and macroalgae) that vary by 5–10 ‰ in δ13C (Fogel and Cifuentes 1993; Kelly 2000; Vanderklift and Ponsard 2003; Page et al. 2008; Newsome et al. 2009, 2010). For example, mussels and clams filter feed particulate organic matter that is a mixture of phytoplankton (low δ13C) and macroalgae (high δ13C), leading to lower δ13C values than macroalgae browsers and grazers such as sea urchins (Strongylocentrotus spp.) (Page et al. 2008; Newsome et al. 2009). Isotope values of consumers reflect these baseline differences, but change in predictable amounts with each increase in trophic level. The offsets between consumer’s tissues and that of its prey are commonly called trophic discrimination factors and, on average, are ~3.5 ‰ per trophic level for δ15N (Kelly 2000; Vanderklift and Ponsard 2003). Thus, red sea urchins (Strongylocentrotus franciscanus), which are primary consumers and eat macroalgae, have lower δ15N values by about 3–4 ‰ than carnivorous Dungeness crabs (Metacarcinus magister; Page et al. 2008; Newsome et al. 2009). This large degree of variation in prey functional group and thus isotope values enhances our ability to examine sea otter dietary ecology and individuality. In sea otters, trophic discrimination factors have been measured as +2.2 ± 0.7 and +3.5 ± 0.6 ‰ for δ13C and δ15N, respectively (Newsome et al. 2010). However, for our analyses it was not necessary to correct for trophic discrimination as we examined population vs. individual dietary variation rather than diet composition.

Sea otter vibrissae were cleaned to remove surface contaminants before isotopic analysis by rinsing with a 2:1 chloroform:methanol solvent mixture. Each whisker was sub-sampled from base to tip in 0.5- to 0.6-mg segments using nail clippers, and δ13C and δ15N values of each segment were measured separately using a Costech 4010 (Costech, Valencia, CA) elemental analyzer interfaced with a Thermo-Finnigan Delta Plus XL mass spectrometer (Thermo Scientific, Bremen) at the University of Wyoming Stable Isotope Facility (Laramie, WY). Isotopic results are expressed as δ values, where δ13C or δ15N = 1,000 [(R sam/R std)−1], and R sam and R std are the 13C:12C or 15N:14N ratios of the sample and standard, respectively. The internationally accepted standards are Vienna-Pee Dee belemnite limestone for C and atmospheric N2 for N. The units are expressed as parts per thousand, or per mil (‰). The standard deviation of organic references within a run was ≤0.2 ‰ for both δ13C and δ15N values. To control for the quality of vibrissae protein (keratin) we also measured the ratio of [C]/[N] for each subsample—these values were within the range of unaltered protein (Ambrose 1990).

Characterization of individual diet specialization

We applied the same theoretical (Roughgarden 1972; Bolnick et al. 2003) framework to both isotopic and observational data sets to characterize individual diet specialization of male and female sea otters. As per convention, we define a specialist as an individual whose niche is substantially smaller than that of the population as a whole (Bolnick et al. 2003). The total population niche width (TNW) is the sum of within- and between-individual components of variation (Roughgarden 1972; Newsome et al. 2009, 2012). The individual component of variation (WIC), reflects the average variance of resources within an individuals’ niche, whereas the between-individual component of variation (BIC), is indicative of differences among individuals (Roughgarden 1972). A common metric for evaluating the degree of individual diet specialization is the WIC/TNW ratio that ranges from 0 to 1 with low values indicating strong specialization within a population, and high values representing a more generalist population (Bolnick et al. 2003).

We first used the field observations of prey use to calculate metrics of individual diet specialization (WICO, BICO, TNWO) for males and females at MBY, BSR, SLO and SNI. Following methods described in detail by Tinker et al. (2012), we used a Monte Carlo simulation algorithm to estimate the proportional contribution of each prey to the diet, and then calculated niche statistics from the resulting matrix of individual prey frequencies using standard methods (Bolnick et al. 2002; Tinker et al. 2012). In order to correct for differing sample sizes between study sites and to estimate variances and 95 % confidence limits we used 1,000 bootstrap resamples of ten individuals from each study site (11 was the actual sample size for SNI, the smallest study site) (Tinker et al. 2012).

We next used sub-sampled sea otter vibrissae to calculate equivalent isotopic metrics of individual diet specialization; observation and isotopic statistics are distinguished with respective O or I subscripts –e.g., TNWI. These metabolically inert tissues provide a longitudinal ~1-year record of dietary variation when sub-sampled (Tyrrell et al. 2013). This serves as a robust proxy for WICI, whereas variation between individual vibrissae serves as a measure of BICI (Newsome et al. 2009). This isotopic approach has been previously shown in southern sea otters to be an accurate proxy for observed population level dietary niche, and degree of individuality (Newsome et al. 2009, 2010). We computed variance components (WICI, BICI, TNWI) using restricted maximum likelihood methods using the statistical program JMP (SAS Institute; version 10.0.2). Following Newsome et al. (2009), we estimated the variance components separately for δ13C and δ15N, but combined the results and report average values of proportions of variance explained. As with the observational dietary data, we computed mean niche statistic values and associated variance (SD) from 1,000 bootstrap resamples of ten individuals from each study site (Tinker et al. 2012).

For individual otters we characterized diet variability using the standard ellipse area (SEA), which represents the amount of space occupied by an individual in bivariate δ13C versus δ15N space. We calculated SEA for each individual sea otter using the Bayesian-based statistics library “Stable Isotope Bayesian Ellipses in R” (SIBER; Jackson et al. 2011), with sequential segments from each vibrissae as inputs into the model (16 ± 4 segments per individual). Results are in units of  per mil squared (‰2). We calculated mean SEA and SD for each individual from 10,000 iterations of the model. We then computed average SEA for males and females at each site.

Statistical methods

We examined the effect of sample size on our isotopic specialization metric (WICI/TNWI ratio) by computing a sampling saturation curve (Fig. 2). Using data from female sea otters at SLO, which was our largest isotopic sample size of 41 individuals (Table 1), we estimated WICI/TNWI using three to 25 individuals randomly sampled from the population without replacement, and repeated this process 1,000 times. We plotted the mean and variation in this estimate as a function of sample size, to evaluate the relative effect of sampling error as one reduces sample size.
Fig. 2

Influence of sample size on isotopic diet specialization results. The degree of diet specialization was characterized using the ratio of the within-individual component of variation (WICI) to total population niche width (TNWI); where subscript I indicates isotopic metrics. WICI/TNWI was estimated using data from female sea otters at SLO, which had the largest sample size of 41 individuals (Table 1), using from 3 to 25 individuals randomly sampled from the population without replacement, repeating this process 1,000 times. We plotted the mean estimate and variation in this estimate as a function of sample size, to evaluate the relative effect of sampling error as one reduces sample size. Our saturation curve suggests that as sample size decreases below ten individuals we have reduced confidence in our isotopic characterizations of dietary specialization. For sample sizes of five, for example, the variance in our estimate of WICI/TNWI increases by almost 50 %

We used general linear models, assuming a normal error distribution and estimated using least squares, to evaluate the relationship between WICI and TNWI, and to determine whether this relationship differed between the sexes. We compared alternative models within an information theoretic framework, evaluating relative support for models with one or both of two main effects: sex (a categorical variable) and TNWI (a continuous variable) as well as a model with an interaction between main effects. We also used general linear models to evaluate potential predictors of WICI/TNWI, including population density (a continuous variable, the number of sea otters per square kilometer of sub-tidal habitat from 0- to 40-m depth) and sex, as well as the interaction between sex and density. Note that density was used as a proxy for per capita food abundance, a relationship that has been well documented in previous analyses of these same populations (Tinker et al. 2008a, 2012; Newsome et al. 2009, 2012, 2015). In both of the above analyses we also evaluated support for a constant model (i.e., no significant effects or interactions), and the model having the lowest Akaike information criterion (AIC) score was considered to be the best-supported model. We then combined isotopic and observational data and repeated the comparison of general linear models to evaluate potential predictors of WIC/TNW, this time including a categorical variable for sampling method (isotopic vs. observational). We note that the observational data sets generally spanned 2–3 years of observations, approximately twice as long as the sample periods for isotopic data from vibrissae (Tyrrell et al. 2013).

To evaluate the relationships between range span, sex and diet diversity, we used linear mixed-effects models to test for effects of sex and range span on individual diet variability, that we measured as the SEA of individuals (Jackson et al. 2011), while accounting for the random effect of study site. We evaluated models with one or both main effects and with an interaction between sex and range span, as well as a constant model, and AIC scores were used to select the best-fit model. For this analysis we used data from adult sea otters only, to avoid confounding effects of age. General linear models, linear mixed models, and all resampling analyses were conducted using the statistics toolbox in MATLAB (Release 2014a, MathWorks, Inc.).


Individual diet specialization

Stable isotopes

Our saturation curve (Fig. 2) suggests that sample sizes below ten individuals reduce confidence in isotopic characterizations of individual diet specialization. For a sample size of six, the variance in our estimate of WICI/TNWI increases by almost 50 % (Fig. 2), and so the WICI/TNWI from SNI (which is based on seven males and six females) must be interpreted cautiously. However, we note that because of the extremely low population densities at SNI, our data actually represented 30 % of the population at the time of collection (2004; Table 1). Additionally, WICI/TNWI estimates for males at BSR (n = 6) should be interpreted with caution, and to a lesser extent the estimates for males at SLO (n = 8) (Tables 1, 2).
Table 2

Summary of observational data results


Densitya (otters km−2)

Range spanb (km)

Male ± SD

Female ± SD

Monterey Bay (MBY)


30.6 ± 39.7

19.1 ± 27.7

Big Sur (BSR)


49.4 ± 65.7

18.1 ± 10.4

San Luis Obispo County (SLO)


30.3 ± 26.7

22.5 ± 15.3

Santa Barbara Channel (SBC)


87.5 ± 95.2

24.2 ± 5.8

San Nicolas Island (SNI)


aDensity represents the average number of sea otters per square kilometer of benthic sub-tidal habitat (measured from the low tide line to the 40-m isobath) at each study site

bRange span represents the average number of kilometers of coastline between the northernmost and southernmost points of the 90 % home-range polygons (fit using fixed kernel density methods)

We find male and female sea otters exhibit different degrees and patterns of isotopic individual diet specialization among and within sites (Table 3). Females show greater overall isotopic niche space (TNWI) compared to males at all sites but SBC (Table 3). Females also show less variance in average individual niche width (WICI) across sites than males by approximately a factor of 2, with males ranging (± SD) from 0.185 ± 0.035 at SNI to 0.513 ± 0.116 at SLO, and females from 0.274 ± 0.057 at MBY to 0.417 ± 0.083 at SBC (Table 3). A comparison of general linear models shows strongest support for a model including an interaction between sex and TNWI in terms of their effects on WICI (Table 4): males exhibit a positive relationship between WICI and TNWI, while females show a slightly negative relationship (Fig. 3).
Table 3

Summary of observational and isotopic individual specialization metrics










Observational specialization

 Monterey Bay (MBY)

1.504 ± 0.119

2.230 ± 0.105

0.675 ± 0.047

1.185 ± 0.159

1.918 ± 0.108

0.617 ± 0.065

 Big Sur (BSR)

1.330 ± 0.113

2.063 ± 0.127

0.646 ± 0.052

0.840 ± 0.143

2.026 ± 0.165

0.413 ± 0.056

 San Luis Obispo County (SLO)

1.594 ± 0.097

2.047 ± 0.107

0.779 ± 0.043

1.261 ± 0.125

2.048 ± 0.185

0.618 ± 0.061

 Santa Barbara Channel (SBC)


 San Nicolas Island (SNI)

0.946 ± 0.066

1.131 ± 0.083

0.837 ± 0.031

0.761 ± 0.062

0.819 ± 0.062

0.928 ± 0.012








Isotopic specialization

 Monterey Bay (MBY)

0.274 ± 0.057

1.127 ± 0.205

0.249 ± 0.059

0.408 ± 0.178

0.848 ± 0.315

0.492 ± 0.134

 Big Sur (BSR)

0.346 ± 0.088

1.180 ± 0.251

0.301 ± 0.078

0.238 ± 0.051

0.610 ± 0.124

0.394 ± 0.064

 San Luis Obispo County (SLO)

0.399 ± 0.073

0.898 ± 0.147

0.455 ± 0.106

0.513 ± 0.116

0.841 ± 0.101

0.610 ± 0.115

 Santa Barbara Channel (SBC)

0.417 ± 0.083

0.600 ± 0.094

0.696 ± 0.090

0.454 ± 0.104

1.101 ± 0.265

0.436 ± 0.139

 San Nicolas Island (SNI)

0.340 ± 0.086

0.855 ± 0.123

0.402 ± 0.100

0.185 ± 0.035

0.527 ± 0.054

0.350 ± 0.053

Isotopic variance components use an average of stable C isotope ratio (δ13C) and stable N isotope ratio (δ15N) values. For both data sets, mean niche statistic values (and associated SDs) were calculated from 1,000 bootstrap resamples of ten individuals from each study site

WIC Within-individual component of variation, TNW total population niche width, subscript O Observational specialization metric, subscript I isotopic specialization metric

Table 4

Summary of results from general linear models (GLM) and linear mixed effects models (LMM)






Isotope data, GLM, n = 10


 WIC ~ 1



 WIC ~ 1 + TNW



 WIC ~ 1 + sex



 WIC ~ 1 + TNW × sex






Sex (M)








TNW × Sex (M)



Isotope data, GLM, n = 10


 WIC/TNW ~ 1



 WIC/TNW ~ 1 + density



 WIC/TNW ~ 1 + sex



 WIC/TNW ~ 1 + sex × density






Sex (M)








Sex(M) × density



Isotope + observational data, GLM, n = 18


 WIC/TNW ~ 1 + method



 WIC/TNW ~ 1 + method + density





 WIC/TNW ~ 1 + method + sex


Method (obs)



 WIC/TNW ~ 1 + method + sex × density





Isotope and movement data, LMM, n = 74


 SEA ~ 1 + (1|site)



 SEA ~ 1 + sex + (1|site)



 SEA ~ 1 + log(range) + (1|site)





 SEA ~ 1 + sex + log(range) + (1|site)





 SEA ~ 1 + sex + log(range) + sex × log(range) + (1|site)



Alternate models and their associated Akaike information criterion (AIC) values are shown, as well as parameter estimates (and SEs) for the best-supported models. For categorical effects, comparison effect level is identified in parentheses

SEA Standard ellipse area; for other abbreviations, see Table 3

aLowest AIC

Fig. 3

The average variation within individuals in isotopic space compared to that of the overall sea otter population for five California sites. Males represented by open triangles, females by filled circles. Trend lines represent predictions of a general linear model describing the relationship between WICI and TNWI, and detailed statistics, including parameter estimates, model-selection criteria and alternate models, are provided in Table 4. TNWI represents total isotopic niche width utilized by the population and is the sum of within (WICI) and between individual components (BICI) of variance. Areas of relative low and high degrees of individuality are shown by the right-hand scale; the bottom right corner of the graph is the region of highest degree of individual diet specialization where the total niche width is large, but variation within individual diets is low. Note that trends for males and females have opposite slopes; the shallow slope exhibited by females reflects that their total dietary niche space increases due to increased variance between individuals rather then by an increase in the variation within individuals. For abbreviations, see Fig. 2


Our observational foraging data are generally in agreement with isotopic data in terms of patterns of variation in diet specialization, with a few key differences. As with isotopic data, females display greater TNWO than males at most sites. However, the exception in this case is SLO, where male and female TNWO is nearly equivalent (Table 3). Additionally, females vary (±SD) by only 0.19 in WICO/TNWO values, from 0.646 ± 0.052 at BSR to 0.837 ± 0.031 at SNI, whereas males vary by 0.515 from 0.413 ± 0.056 at BSR to 0.928 ± 0.012 at SNI (Table 3).

Diet and population density

Males and females appear to differ in terms of the relationship between diet specialization and population density, although our results for isotopic and observational data are not entirely consistent. When considering just the isotope data, there is weak but significant support for an interaction between sex and density (Table 4) such that diet specialization increases with density for females (i.e., WICI/TNWI decreases) but not for males (Fig. 4a). When isotopic and observational data are combined the sex-density interaction is no longer supported (Table 4), although there is support for an increase in diet specialization with density for both sexes (i.e., a decrease in WICI/TNWI; Fig. 4b). Based on isotopic data, females show a greater degree of individual specialization (lower WICI/TNWI values) than males at the sites with highest population densities (MBY, BSR and SLO; Table 3); in contrast, observational data suggested that females are less specialized (higher WICO/TNWO) than their male counterparts at every site except SNI (Table 3).
Fig. 4

The influence of density on sea otter dietary individuality. Shown are data derived from isotopic analysis (a) and observational studies (b) for females (filled circles) and males (open triangles). Trend lines represent predictions of general linear models describing the relationship between WIC/TNW and density, and detailed statistics, including parameter estimates, model-selection criteria and alternate models, are provided in Table 4. Density was determined by annual aerial census counts performed by the US Geological Survey. The degree of dietary specialization was characterized using the WIC/TNW ratio where TNW is the sum of WIC and BIC. The relationship between isotopic (WICI/TNWI) and observational (WICO/TNWO) specialization and density was evaluated using general linear models, and Akaike information criterion (AIC) statistics, combining isotopic and observational data and including a categorical variable for sampling method. The solid line represents the best-fit function (see Table 4), with surrounding confidence intervals (dotted lines) estimated for females when isotopic and observational metrics were combined. The dashed line represents the best-fit function for males only from isotopic data, as there is some support for a sex/density interaction in this data set. For abbreviations, see Figs. 2 and 3

Isotopic specialization and range span

We find males and females exhibit substantially different patterns of movement within and among sites. Males have, on average, larger range spans than females at all four sites where data were available (Table 2), although there is also greater variation among individual males. The amount of coastline encompassed by an individual’s home range has a significant effect on diet variability: the best-supported model explaining variation in SEA includes a fixed effect for range span (log transformed) as well as a random effect for site (Table 4). The increase in diet variability as a function of range span is consistent across sites and applies to both males and females (Fig. 5).
Fig. 5

The influence of range span on sea otter dietary individuality. Individual diet variability was characterized using standard ellipse areas (SEA) run using Stable Isotope Bayesian Ellipses in R (SIBER), which represents the amount of space occupied by an individual in bivariate δ13C and δ15N space; SEA units are ‰2. Mean SEA for each individual was calculated from 10,000 iterations of the model. Range span is defined as the linear distance of coastline encompassed by an individuals movements within its annual home range; movement data were log transformed. The relationship between range span and diet diversity was evaluated using linear mixed-effects models, using AIC scores and study site as a random effect. Solid line represents the best-fit function with surrounding confidence intervals (dotted lines); detailed statistics, including parameter estimates, model-selection criteria and alternate models, are provided in Table 4. The amount of coastline within an individual’s home range has a significant effect on diet variability consistent across sites and sexes. For other abbreviations, see Fig. 1


Our findings highlight the importance of sex-based differences in reproductive investment and behavior, particularly movement, in driving dietary individuality in sea otters. Consistent with our hypothesis, we find differences for both isotopic and observational metrics in the extent and variability of individual diet specialization between males and females both within and among sites along the California coast. Isotopic data suggest females are more specialized than males at sites with high sea otter densities (MBY, SLO, BSR), and less specialized at the two low-density sites: SBC and SNI (Tables 2, 3; Fig. 5a). Our observational data exhibit a different pattern; male otters appear to be more specialized than females at MBY, BSR, and SLO (Tables 2, 3; Fig. 5b). One possible explanation for this difference is that prey-stealing (kleptoparasitism) by male sea otters (Riedman and Estes 1990) leads to greater diet diversity, and this is reflected in the isotopic data more than the observational data. This explanation seems unlikely however, since male kleptoparasitism was consistently recorded and accounted for in the analysis of observational data (Tinker et al. 2012), and was not found to have a significant effect on diet-diversity estimates for males. Moreover, if such a bias occurred it should affect high-density and low-density sites equally. Instead, we suspect this finding is associated with a spatial bias in observational data for highly mobile males. Specifically, many males have home ranges that are divided into several distinct “centers-of-use” distributed widely throughout the range (Ralls et al. 1996; Jameson 1989; Tinker et al. 2008b). Observational dietary data were typically collected from just one or two of these centers where observers had the best access for shore-based observation. In contrast, the isotopic data are integrated across the males’ full ranges, or at least everywhere they have foraged within the previous year (Tyrrell et al. 2013). For such mobile males, the isotopic data thus provide a less-biased sampling method and result in more reliable estimates of dietary diversity. In contrast, the much smaller range of females minimizes the potential for observational bias. Hence, we consider the isotopic data to correctly reflect a greater degree of specialization in females as compared to males at high-density sites where there is strong intraspecific competition.

Our results also suggest that male and female sea otters respond differently to increased intraspecific competition (Fig. 4), although we note that the model including a sex-density interaction received only marginally more support than a constant model (Table 4). Females tend to become more specialized with increasing population densities, a pattern supported by both isotopic and observational analyses (Fig. 4a, b). At the population level, TNW increases at higher density sites; however, the variability of individual female diets (WIC) remains unchanged or even decreases (Fig. 3). These opposing trends result in a positive relationship between density and female diet specialization indicated by lower WICO/TNWO and WICI/TNWI ratios (Table 4; Fig. 4a, b). This relationship does not exist for males, and based on the analysis of isotope data, which we believe provides the most reliable estimate of male diet diversity, TNWI and WICI are positively correlated (Fig. 3). This suggests that individual diet diversity of males tracks population-level diversity. While the statistical significance of this sex-based difference (Fig. 4) is marginal at best, we note that this is not surprising given a sample size of five sites and the complexities of our study system. It was for this reason that we employed two independent lines of evidence (isotopic and observational), which produced largely similar results in consideration of their respective strengths and biases. We therefore suggest that the previously reported pattern of density-dependent emergence of diet specialization in sea otters (Estes 2003; Tinker et al. 2008a, 2012) may be largely driven by the behavioral responses of females. Our results also suggest a potential explanation for this sex dependency: specifically, the contrasting spatial ecologies of male and female sea otters.

The suggested link between spatial, reproductive, and foraging ecology is a novel aspect of our work on intrapopulation diet variation. We provide the first empirical evidence that diet diversity in sea otters is explained in large degree by individual mobility (Table 4; Fig. 5), and thus the greater mobility of males may be the proximate cause of sex-based differences in diet specialization. In grey wolves (Canis lupis), landscape and subsequent resource heterogeneity have been shown to influence the competitive environment and potential resources available to individuals, leading to greater population-level dietary variation (Darimont et al. 2009). Similarly with sea otters, it is reasonable to expect that a wide-ranging male is more likely to encounter greater environmental heterogeneity and thus prey diversity than a less mobile female (Table 2). The smaller home ranges used by females (Table 2: Loughlin 1980, Tinker et al. 2008b) may reflect their greater reproductive constraints, as adult female sea otters are constantly in a state of pregnancy or lactating and provisioning a dependent pup after reaching sexual maturity at 2–4 years of age (Estes 1980; Jameson and Johnson 1993; Riedman et al.1994; Thometz et al. 2014). The cost of this parental care is tremendous, elevating a female’s daily energetic requirements by 17–96 % over the ~6-month lactation period (Thometz et al. 2014). In stark contrast, reproductive investment in male sea otters is limited to territory defense, which does not appear to incur high energetic costs (Loughlin 1980; Pearson et al. 2006). As measured by one study, there was no significant difference in body condition between adult males holding high- or low-quality territories (Pearson et al. 2006). Thus, adult females have a smaller energetic margin of error and must rely more on extensive knowledge of local prey patches with less latitude for exploratory movements to unfamiliar habitats. The result of high site fidelity for females is a consistent pattern of local resource competition, and in such a scenario there is likely strong selection to specialize to maximize energy gain (Tinker et al. 2012).

Consistent with this prediction, we find the amount of coastline encompassed by an individual sea otter’s home range has a significant effect on diet variability (Table 4; Fig. 5). We find females from the highest density sites (MBY and BSR) are the most specialized and have the smallest range spans (Tables 2, 3). At low-density sites (SBC), females are less constrained by intraspecific competition (Tinker et al. 2008a, 2012; Newsome et al. 2012), have larger range spans, and exhibit weaker individual specialization (Tables 2, 3). Males are not constrained by the same reproductive demands as females, and can thus escape local resource competition by moving elsewhere, thereby gaining access to episodically abundant prey in diverse habitats spread throughout the range. The vagabond nature of males could actually select against diet specialization, as generalists may be better equipped to take advantage of a broader array of habitats and prey types (Bolnick et al. 2003; Darimont et al. 2009).

Recent reviews have called for more studies focused on elucidating the relative importance of four identified ecological mechanisms in maintaining individual specialization in natural populations (Araújo et al. 2011). Ecological opportunity and intraspecific competition have been proposed and demonstrated to promote intrapopulation variation (Svanbäck and Bolnick 2007; Tinker et al. 2008a, b; Darimont et al. 2009; Svanbäck and Persson 2009; Araújo et al. 2011), whereas interspecific competition and predation are thought to dampen it (Eklov and Svanbäck 2006; Peacor and Pfister 2006; Bolnick et al. 2010). Here, we argue that females are more sensitive than males to intraspecific competition (Fig. 4), reflecting differing spatial scales of habitat use that are in turn driven by differing life history constraints. Intense reproductive investment (Thometz et al. 2014) likely prevents female sea otters from moving away from conspecifics as food resources become limiting, and thus there is strong selection to avoid competition by partitioning resources and becoming more specialized. This sex-based difference in specialization driven by intraspecific competition might well be a general pattern in nature, as females of many taxa are constrained energetically and behaviorally to varying degrees by reproductive investments (Trivers 1972; Trillmich 1990; Charnov 1993; Ruckstuhl 2007; Grigione et al. 2002). In particular, female mammals may be especially susceptible to intraspecific pressure as they are typically the sole providers of parental care and the physiological costs of producing young are quite high (Kleiman 1977; Ralls 1977; Charnov 1993; Lindenfors et al. 2007).

Sex-based differences in intrapopulation niche partitioning might also be expected to occur in response to predation, interspecific competition and ecological opportunity. For example, in laboratory experiments, female rats (Rattus norvegicus), guppies (Poecilia reticulate) and lizards (Anolis carolinensis) demonstrated more defensive behavioral patterns than males with exposure to predatory stimuli (Magurran and Nowak 1991; Klein et al. 1994; Irschick et al. 2005). This could lead to: (1) decreased individual diet specialization if other females behave similarly and converge on “predator-safe” resources or habitats, or (2) a feedback loop where similar predator-avoidance behavior by all females leads to an increase in intraspecific competition and thus stronger specialization (Eklov and Svanbäck 2006; Araújo et al. 2011). In the case of interspecific competition, males may be more able to escape areas of high competition and thus utilize a greater range of resources, resulting in decreased individual diet specialization (Bolnick et al. 2010; Araújo et al. 2011). Finally, differences between males and females in size, behavior and energetic demands may lead to different rank preferences of prey. In sea otters, females have been observed to switch diets depending on reproductive stage, consuming less common but calorie-rich prey such as urchins during pregnancy and switching to nutritionally poor but abundant prey items such as snails during lactation (Staedler 2011). These different preferences could potentially lead to the sexes responding divergently to the same set of variable resources, thereby promoting or limiting ecological opportunity between females and males (Bolnick et al. 2003).

Individual-level niche variation can have substantial impacts on population, community and ecosystem dynamics (Bolnick et al. 2003; Kondoh 2003; Araújo et al. 2011). For example, individual diet specialization can lead to: (1) decreased strength in trophic interactions, (2) individuals within the same population being subject to different selective pressures, and (3) populations that are buffered against ecological perturbations (Bolnick et al. 2003; Kondoh 2003). Intra- and interspecific competition, predation and ecological opportunity all play roles in determining the incidence and magnitude of such specialization (Eklov and Svanbäck 2006; Peacor and Pfister 2006; Svanbäck and Bolnick 2007; Tinker et al. 2008a; Darimont et al. 2009; Svanbäck and Persson 2009; Bolnick et al. 2010). Our work suggests that the sexes experience these mechanistic pressures differently, and that females may be under stronger selection to specialize under some scenarios. We expect this pattern will be especially evident in taxa with high reproductive costs and differences between males and females in terms of offspring investment/care that ultimately impact mobility. Future research into sex-based specialization will be necessary to fully understand the mechanisms governing intrapopulation niche variation, and its implications for individual populations and food web dynamics.

Author contribution statement

EES formulated the research questions with input from MTT, JAE and SDN. SDN and MTT provided data; EES, SDN and MTT conducted statistical analyses. EES wrote the manuscript with editorial advice from all co-authors.



We thank all members of the Tinker, Estes and Newsome labs for data collection, support and comments. Additionally, we thank F. A. Smith for invaluable assistance with the manuscript. We thank Luke Tyrrell, Kelli Blomberg, Ryan Jones, and Deborah Boro for laboratory assistance. We also thank the USGS Pacific Nearshore Project team members, especially Keith Miles and Liz Bowen (USGS), Heather Coletti (National Park Service), Tom Dean (Coastal Resource Associates), and USGS personnel Brenda Ballachey, Dan Monson, George Esslinger, Kim Kloecker, and Ben Weitzman for logistical and intellectual support. The manuscript also benefitted greatly from constructive criticism by C. A. Layman and two anonymous reviewers. Any use of trade, product, or firm names in this publication is for descriptive purposes only and does not imply endorsement by the US government.


  1. Ambrose SH (1990) Preparation and characterization of bone and tooth collagen for isotopic analysis. J Archaeol Sci 17:431–451CrossRefGoogle Scholar
  2. Araújo MS, Bolnick DI, Layman CA (2011) The ecological causes of individual specialisation. Ecol Lett 14:948–958. doi: 10.1111/j.1461-0248.2011.01662.x CrossRefPubMedGoogle Scholar
  3. Bateman AJ (1948) Intra-sexual selection in Drosophila. Heredity 2:349–368CrossRefPubMedGoogle Scholar
  4. Bolnick DI, Yang LH, Fordyce JA, Davis JM, Svanbäck R (2002) Measuring individual-level resource specialisation. Ecology 83:2936–2941CrossRefGoogle Scholar
  5. Bolnick DI, Svanbäck R, Fordyca JA, Yang LH, Davis JM, Hulsey D, Forister ML (2003) The ecology of individuals: incidence and implications of individual specialization. Am Nat 161:1–28. doi: 10.1086/343878 CrossRefPubMedGoogle Scholar
  6. Bolnick DI, Ingram T, Snowberg LK, Stutz WE, Lau OL, Paull JS (2010) Ecological release from interspecific competition leads to decoupled changes in population and individual niche widths. Proc R Soc Lond B Ser B 277:1789–1797CrossRefGoogle Scholar
  7. Caswell H, Neubert MG, Hunter CM (2011) Demography and dispersal: invasion speeds and sensitivity analysis in periodic and stochastic environments. Theor Ecol 4:407–421CrossRefGoogle Scholar
  8. Charnov EL (1993) Life history invariants: some explorations of symmetry in evolutionary ecology. Oxford University Press, OxfordGoogle Scholar
  9. Clutton-Brock TH (1991) The evolution of parental care. Princeton University Press, PrincetonGoogle Scholar
  10. Clutton-Brock TH (2007) Sexual selection in males and females. Science 318(5858):1882–1885. doi: 10.1126/science.1133311 CrossRefPubMedGoogle Scholar
  11. Clutton-Brock TH, Coulson TN, Milner-Gulland EJ, Thomson D, Armstrong HM (2002) Sex differences in emigration and mortality affect optimal management of deer populations. Nature 415:633–637CrossRefPubMedGoogle Scholar
  12. Darimont CT, Paquet PC, Reimchen TE (2009) Landscape heterogeneity and marine subsidy generate extensive intrapopulation niche diversity in a large terrestrial carnivore. J Anim Ecol 78:126–133CrossRefPubMedGoogle Scholar
  13. Eklov P, Svanbäck R (2006) Predation risk influences adaptive morphological variation in fish populations. Am Nat 167:440–452. doi: 10.1086/499544 CrossRefPubMedGoogle Scholar
  14. Estes JA (1980) Enhydra lutris. Mammal Sp 133:1–8CrossRefGoogle Scholar
  15. Estes JA, Jameson RJ (1988) A double survey estimate for sighting probability of sea otters in California. J Wildl Manage 52:70–76CrossRefGoogle Scholar
  16. Estes JA, Riedman ML, Staedler MM, Tinker MT, Lyon BE (2003) Individual variation in prey selection by sea otters: patterns, causes and implications. J Anim Ecol 72:144–155. doi: 10.1046/j.1365-2656.2003.00690.x CrossRefGoogle Scholar
  17. Fogel ML, Cifuentes LA (1993) Isotope fractionation during primary production. In: Engel MH, Macko SA (eds) Organic geochemistry. Springer, New York, pp 73-98 doi:  10.1007/978-1-4615-2890-6
  18. Grigione MM, Beier P, Hopkins RA, Neal D, Padley WD, Schonewald CM, Johnson ML (2002) Ecological and allometric determinants of home-range size for mountain lions (Puma concolor). Anim Conserv 5:317–324. doi: 10.1017/S1367943002004079 CrossRefGoogle Scholar
  19. Irschick DJ, Carlisle E, Elstrott J, Ramos M, Buckley C, VanHooydonck B, Meyers JAY, Herrel A (2005) A comparison of habitat use, morphology, clinging performance and escape behaviour among two divergent green anole lizard (Anolis carolinensis) populations. Biol J Linn Soc 85(2):223–234CrossRefGoogle Scholar
  20. Jackson AL, Inger R, Parnell AC, Bearhop S (2011) Comparing isotopic niche widths among and within communities: SIBER-Stable Isotope Bayesian Ellipses in R. J Anim Ecol 80:595–602CrossRefPubMedGoogle Scholar
  21. Jameson RJ (1989) Movements, home range and territories of male sea otters off central California. Mar Mamm Sci 5:159–172. doi: 10.1111/j.1748-7692.1989.tb00330.x CrossRefGoogle Scholar
  22. Jameson RJ, Johnson AM (1993) Reproductive characteristics of female sea otters. Mar Mamm Sci 9:156–167. doi: 10.1111/j.1748-7692.1993.tb00440.x CrossRefGoogle Scholar
  23. Kage AH (2004) Temporal and spatial variation in movement patterns of the California sea otter, Enhydra lutris nereis. MS Dissertation. University of California, Santa Cruz, CAGoogle Scholar
  24. Kelly JF (2000) Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. Can J Zool 78:1–27CrossRefGoogle Scholar
  25. Kleiman DG (1977) Monogamy in mammals. Q Rev Biol 52:39–69. doi: 10.1086/409721 CrossRefPubMedGoogle Scholar
  26. Klein SL, Lambert KG, Durr D, Schaefer T, Waring RE (1994) Influence of environmental enrichment and sex on predator stress response in rats. Physiol Behav 46:291–297. doi: 10.1016/0031-9384(94)90197-X CrossRefGoogle Scholar
  27. Kondoh M (2003) Foraging adaptation and the relationship between food-web complexity and stability. Science 299(5611):1388–1391CrossRefPubMedGoogle Scholar
  28. Laidre KL, Jameson RJ, Gurarie E, Jeffries SJ, Allen H (2009) Spatial habitat use patterns of sea otters in coastal Washington. J Mammal 90:906–917CrossRefGoogle Scholar
  29. Lima SL, Dill LM (1990) Behavioral decisions made under the risk of predation: a review and prospectus. Can J Zool 68:619–640. doi: 10.1139/z90-092 CrossRefGoogle Scholar
  30. Lindenfors P, Gittleman JL, Jones KE (2007) Sexual size dimorphism in mammals. In: Fairbairn D, Blanckenhorn W, Székely T (eds) Sex evolutionary studies of sexual size dimorphism. Oxford University Press, Oxford, pp 16–26Google Scholar
  31. Loughlin TR (1980) Home range and territoriality of sea otters near Monterey, California. J Wildl Manage 44:576–582. doi: 10.2307/3808005 CrossRefGoogle Scholar
  32. Loveridge AJ, Valeix M, Davidson Z, Murindagomo F, Fritz H, Macdonald DW (2009) Changes in home range size of African lions in relation to pride size and prey biomass in a semi-arid savanna. Ecography 32:953–962. doi: 10.1111/j.1600-0587.2009.05745.x Google Scholar
  33. Magurran AE, Nowak MA (1991) Another battle of the sexes: the consequences of sexual asymmetry in mating costs and predation risk in the guppy, Poecilla reticulata. Proc R Soc Lond B 246:31–38. doi: 10.1098/rspb.1991.01211471-2954 CrossRefGoogle Scholar
  34. Millar JS (1977) Adaptive features of mammalian reproduction. Evolution 31:370–386CrossRefGoogle Scholar
  35. Newsome SD, Tinker MT, Monson DH, Oftedal OT, Ralls K, Staedler MM, Fogel ML, Estes JA (2009) Using stable isotopes to investigate individual diet specialization in California sea otters (Enhydra lutris nereis). Ecology 90:961–974.CrossRefPubMedGoogle Scholar
  36. Newsome SD, Bentall GB, Tinker MT, Oftedal OT, Ralls K, Estes JA, Fogel ML (2010) Variation in δ13C and δ15N diet-vibrissae trophic discrimination factors in a wild population of California sea otters. Ecol Appl 20:1744–1752. doi: 10.1890/09-1502.1 CrossRefPubMedGoogle Scholar
  37. Newsome SD, Yeakel JD, Wheatley PV, Tinker MT (2012) Tools for quantifying isotopic niche space and dietary variation at the individual and population scale. J Mamm 93(2):329–341CrossRefGoogle Scholar
  38. Newsome SD, Tinker MT, Gill VA, Hoyt ZN, Doroff A, Nichol L, Bodkin JL (2015) The interaction of intraspecific competition and habitat on individual diet specialization: a near range-wide examination of sea otters. Oecologia. doi: 10.1007/s00442-014-3204-3
  39. Page HM, Reed DC, Brzezinski MA, Melack JM, Dugan JE (2008) Assessing the importance of land and marine sources or organic matter to kelp forest food webs. Mar Ecol Prog Ser 360:47–62CrossRefGoogle Scholar
  40. Peacor SD, Pfister CA (2006) Experimental and model analyses of the effects of competition on individual size variation in wood frog (Rana sylvatica) tadpoles. J Anim Ecol 75:990–999. doi: 10.1111/j.1365-2656.2006.01119.x CrossRefPubMedGoogle Scholar
  41. Pearson HC, Packard JM, Davis RW (2006) Territory quality of male sea otters in Prince William Sound, Alaska: relation to body and territory maintenance behaviors. Can J Zool 84:939–946. doi: 10.1139/Z06-083 CrossRefGoogle Scholar
  42. Ralls K (1977) Sexual dimorphism in mammals: avian models and unanswered questions. Am Nat 111:917–938. doi: 10.1086/283223 CrossRefGoogle Scholar
  43. Ralls K, Eagle TC, Siniff DB (1996) Movement and spatial use patterns of California sea otters. Can J Zool 74:1841–1849CrossRefGoogle Scholar
  44. Riedman M, Estes JA (1990) The sea otter (Enhydra lutris): behavior, ecology, and natural history. Biol Rep (USA) 90(14)Google Scholar
  45. Riedman ML, Estes JA, Staedler MM, Giles AA, Carlson DR (1994) Breeding patterns and reproductive success of California sea otters. J Wildl Manage 58:391–399. doi: 10.2307/3809308 CrossRefGoogle Scholar
  46. Roughgarden J (1972) Evolution of niche width. Am Nat 106(683):712Google Scholar
  47. Ruckstuhl KE (2007) Sexual segregation in vertebrates: proximate and ultimate causes. Int Comp Biol 47(2):245–257CrossRefGoogle Scholar
  48. Staedler MM (2011) Individual variation in maternal care and provisioning in the southern sea otter (Enhydra lutris nereis): causes and consequences of diet specialization in a top predator. MS dissertation, University of California, Santa CruzGoogle Scholar
  49. Stander PE (1991) Demography of lions in the Etosha national park, Namibia. Modoqua 18:1–9Google Scholar
  50. Svanbäck R, Bolnick DI (2007) Intraspecific competition drives increased resource use diversity within a natural population. Proc R Soc B 274:839–844. doi: 10.1098/rspb.2006.0198 CrossRefPubMedCentralPubMedGoogle Scholar
  51. Svanbäck R, Persson L (2009) Population density fluctuations change the selection gradient in Eurasian perch. Am Nat 173(4):507–516CrossRefPubMedGoogle Scholar
  52. Thometz NM, Tinker MT, Staedler MM, Mayer KA, Williams TM (2014) Energetic demands of immature sea otters from birth to weaning: implications for maternal costs, reproductive behavior and population-level trends. J Exp Biol 217(12):2053–2061CrossRefPubMedGoogle Scholar
  53. Tinker MT, Doak DF, Estes JA, Hatfield BB, Staedler MM, Bodkin JL (2006) Incorporating diverse data and realistic complexity into demographic estimation procedures for sea otters. Ecol Appl 16:2293–2312CrossRefPubMedGoogle Scholar
  54. Tinker MT, Bentall G, Estes JA (2008a) Food limitation leads to behavioral diversification and dietary specialization in sea otters. PNAS 105:560–565. doi: 10.1073/pnas.0709263105 CrossRefPubMedCentralPubMedGoogle Scholar
  55. Tinker MT, Doak DF, Estes JA (2008b) Using demography and movement behavior to predict range expansion of the southern sea otter. Ecol Appl 18:1781–1794CrossRefPubMedGoogle Scholar
  56. Tinker MT, Guimaraes PR, Novak M, Marquitti FMD, Bodkin JL, Staedler M, Bentall G, Estes JA (2012) Structure and mechanism of diet specialisation: testing models of individual variation in resource use with sea otters. Ecol Lett 15:475–483. doi: 10.1111/j.1461-0248.2012.01760.x CrossRefGoogle Scholar
  57. Trillmich F (1990) The behavioral ecology of maternal effort in fur seals and sea lions. Behavior 114:3–20. doi: 10.1163/156853990X00022 CrossRefGoogle Scholar
  58. Trivers RL (1972) Parental investment and sexual selection. In: Campbell B (ed) Sexual selection and the descent of man, 1871-1971. Aldine, Chicago, pp 136-179Google Scholar
  59. Tyrrell LP, Newsome SD, Fogel ML, Viens M, Bowden R, Murray MJ (2013) Vibrissae growth rates and trophic discrimination factors in captive southern sea otters (Enhydra lutris nereis). J Mammal 94:331–338. doi: 10.1644/12-MAMM-A-035.1 CrossRefGoogle Scholar
  60. Vanderklift MA, Ponsard S (2003) Sources of variation in consumer-diet δ15N enrichment: a meta-analysis. Oecologia 136:169–182CrossRefPubMedGoogle Scholar
  61. Williams TM, Rutishauser M, Long B, Fink T, Gafney J, Mostman-Liwanag H, Casper D (2007) Seasonal variability in otariid energetics: implications for the effects of predators on localized prey resources. Physiol Biochem Zool 80:433–443CrossRefPubMedGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2015

Authors and Affiliations

  • Emma A. Elliott Smith
    • 1
    Email author
  • Seth D. Newsome
    • 1
  • James A. Estes
    • 2
  • M. Tim Tinker
    • 3
  1. 1.Biology DepartmentUniversity of New MexicoAlbuquerqueUSA
  2. 2.Department of Ecology and Evolutionary BiologyUniversity of California Santa CruzSanta CruzUSA
  3. 3.US Geological Survey, Western Ecological Research CenterSanta CruzUSA

Personalised recommendations