Using stable carbon (Î´13C) and nitrogen (Î´15N) isotopes to infer ...

Color profile: Disabled Composite Default screen

1332

Using stable carbon (δ13C) and nitrogen (δ15N) isotopes to infer trophic relationships among black and grizzly bears in the upper Columbia River basin, British Columbia Keith A. Hobson, Bruce N. McLellan, and John G. Woods

Abstract: Ecological segregation of species is difficult to determine using conventional dietary analysis techniques. However, stable-isotope analysis may provide a convenient means of establishing trophic segregation of species and of groups of animals within a species in the same area. We measured stable carbon (δ13C) and nitrogen (δ15N) isotope values in hair of black bears (Ursus americanus) and grizzly bears (Ursus arctos) inhabiting the upper Columbia River basin in southeastern British Columbia, together with samples of potential foods ranging from plant material through invertebrates and ungulate meat. We found extensive overlap in both δ15N and δ13C values of hair from male grizzly bears and black bears of both sexes. Female grizzly bears, however, had lower δ15N values in their hair than the other groups of bears, indicating either less animal protein in their diet or a reliance on foods more depleted in 15N, possibly related to altitude. Our isotopic model generally confirmed a herbivorous diet for both bear species (a mean estimated plant contribution of 91%). Bears showing the highest δ15N values were those captured because they posed a management problem. We suggest that the slope of the relationship between tissue δ15N and δ13C values might provide a convenient means of evaluating the occurrence of consumption of animal protein in populations, regardless of local isotopic end-points for dietary samples. We examined three black bear cubs from dens and found them to be about a trophic level higher than adult females, reflecting their dependence on mother’s milk, a result generally confirmed by an analysis of eight mother–cub pairs from Minnesota. Our study demonstrates how stable-isotope analysis of bear tissue can be used to monitor the feeding habits of populations, as well as provide dietary histories that may reveal dietary specializations among individuals. Résumé : La ségrégation écologique entre les espèces est difficile à déterminer au moyen des techniques conventionnelles d’analyse des aliments, mais l’analyse des isotopes stables peut être un moyen pratique d’établir la ségrégation trophique entre espèces et entre groupes d’animaux de la même espèce vivant dans la même région. Nous avons mesuré les isotopes stables de carbone (*13C) et d’azote *15N) dans le poil d’Ours noirs (Ursus americanus) et de Grizzlis (Ursus arctos) du bassin supérieur du Columbia, dans le sud-est de la Colombie-Britannique, de même que des échantillons de nourriture potentielle, plantes, invertébrés ou viande d’ongulés. Nous avons observé un chevauchement important des valeurs de *15N et de *13C dans les poils de grizzlis mâles et d’ours noirs des deux sexes. Les grizzlis femelles avaient cependant des valeurs de *15N plus faibles dans leurs poils que dans les poils des autres groupes examinés, ce qui révèle que ces animaux consomment moins de protéines animales dans leur régime ou alors comptent sur des aliments appauvris en *15N, peut-être à cause de l’altitude. De façon générale, notre modèle isotopique a confirmé le recours à un régime alimentaire herbivore chez les deux espèces (la contribution moyenne des plantes au régime était de 91 %). Les ours qui avaient les valeurs de *15N les plus élevées représentaient les cas problèmes d’aménagement. La pente de la relation entre les isotopes *15N et *13C dans un tissu est sans doute un bon indice de la consommation de protéines animales chez les populations, sans tenir compte des valeurs isotopiques locales extrêmes dans les échantillons de nourriture. Nous avons examiné trois oursons de l’Ours noir dans leurs terriers et avons constaté qu’ils étaient de niveau trophique supérieur à celui des femelles adultes à cause de leur recours au lait maternel, une constatation généralement confirmée par l’analyse de huit paires mère–ourson du Minnesota. Notre étude démontre comment l’analyse des isotopes stables dans les tissus des ours peut servir à évaluer les habitudes alimentaires des populations, tout en mettant en lumière des aspects historiques de leur alimentation qui peuvent démontrer les spécialisations alimentaires de certains individus. [Traduit par la Rédaction]

1339

Received August 9, 1999. Accepted March 30, 2000. K.A. Hobson.1 Canadian Wildlife Service, 115 Perimeter Road, Saskatoon, SK S7N 0X4, Canada. B.N. McLellan. British Columbia Ministry of Forests, Box 9156, R.P.O. 3, Revelstoke, BC V0E 3K0, Canada. J.G. Woods. Mount Revelstoke and Glacier National Parks, Box 350, Revelstoke, BC V0E 2SO, Canada. 1

Author to whom all correspondence should be addressed (e-mail: [email protected]).

Can. J. Zool. 78: 1332–1339 (2000)

J:\cjz\cjz78\cjz-08\Z00-069.vp Thursday, July 27, 2000 11:03:39 AM

© 2000 NRC Canada


Hobson et al.

Introduction Hobson et al. Two bear species, the black bear (Ursus americanus) and grizzly bear (Ursus arctos), occur sympatrically throughout much of British Columbia (McLellan 1993). The black bear and its ancestors have been in North America for >3 million years and, being omnivores, they have filled a generalist niche (Kurten and Anderson 1980). The grizzly bear, another omnivorous ursid, radiated across most of North America as recently as 13 000 BP (Kurten and Anderson 1980). The morphological and ecological similarity between these species has stimulated interest in the mechanisms of resource partitioning (McLellan 1993; Aune and Kasworm 1994). Fundamental to understanding resource partitioning is documenting the dietary overlap of sympatric black and brown bears. However, our investigation of the diets of wild bears has been hampered by limitations inherent in conventional methods such as scat analysis (Mattson et al. 1991; Mattson and Reinhart 1995; McLellan and Hovey 1995; Jacoby et al. 1999). Unequal digestibility of foods and unequal detectability in the field complicate traditional methods, and without DNA analysis, it is often difficult to determine which individual, or even species, deposited a scat (McLellan and Hovey 1995). Recently, stable-isotope analysis has been used to infer relative trophic positions of black and grizzly bears and, more particularly, their use of plants, terrestrial meat, and anadromous salmon in C-3 (C3 plant) dominated biomes of western North America (Hilderbrand et al. 1996, 1999; Jacoby et al. 1999). Although limited to determining trophic level, this approach has advantages over conventional approaches and has been used successfully in previous ecological investigations (reviewed by Peterson and Fry 1987; Hobson 1999). The stable-isotope approach is based on the fact that naturally occurring stable-isotope ratios in consumer tissues can be related to those in consumers’ diets (DeNiro and Epstein 1978, 1981). Changes in, or fractionation of, stable nitrogen isotope ratios (15N/14N) occur with trophic level and are of the order of 3–4‰ (Peterson and Fry 1987; Hobson and Welch 1992). Thus, isotopic measurement of consumers’ tissues can reveal information about their ingested foods and about trophic level in systems that are relatively simple and do not involve multiple isotopic inputs. One such simple application involves determining the relative dependence of omnivores on plants and vertebrate or invertebrate tissues in systems in which stable nitrogen isotope values of diet plants are similar and do not overlap those in the tissues of local herbivores. Here we expect stable-isotope values in consumers’ tissues to be a reliable indicator of trophic level during the period of tissue growth in the case of metabolically inert structures such as hair or nail and during the period of dietary integration in the case of metabolically active tissues such as blood and muscle (Hobson 1999). We measured stable carbon and nitrogen isotope ratios in the hair of black and grizzly bears from the upper Columbia River basin in British Columbia, Canada. Our own experience had indicated that bears’ hair grows predominantly during summer and fall (May–October) and, hence, represents a dietary average during that period. However, we are aware that hair growth can show significant individual variation due to differences in nutrition (Jacoby et al. 1999). We were

1333

particularly interested in the effect of species and sex on the level of carnivory.

Methods Study site and field collection The study area in the upper Columbia River basin of British Columbia was previously described by Woods et al. (1999). Hair samples came from 123 bears captured for research purposes between 1994 and 1998 as part of a radio-tracking project. To reduce possible sampling biases, bears were captured using both vehicle- and helicopter-based trapping plus free-darting from a helicopter to ensure thorough coverage of the study area. In 10 cases we sampled the same individuals in successive years but considered these to be independent in our analyses. Hair samples were also obtained from 21 bears that were captured because they posed a management problem, i.e., they were feeding in garbage dumps or close to human habitation. The age of “research” and “management” bears ranged from 1.4 to 26.7 years, based on tooth-cementum analysis; for comparisons we used only bears >2 years of age, to ensure that their mother’s milk was insignificant in their diet. We also obtained hair samples from three black bear cubs and their mothers in our study area and from eight black bear mothers and their cubs while in dens in northern Minnesota in 1998–1999. These samples were collected to confirm isotopic patterns expected in nursing cubs versus adults in general. Other tissue samples were salvaged from mammals found dead within the study area. Tissue samples were collected from plant species known to be eaten by bears in this area. These samples were collected primarily from two mid-elevation locations (1000–1600 m) and pooled for analysis. Animal handling and specimen collecting were conducted under permits issued by Parks Canada and the B.C. Ministry of Environment, Lands and Parks. All statistical analyses were performed with an assumed significance level of p < 0.05.

Stable-isotope analyses Hair was cleaned of surface oils using repeated rinses of a 2:1 chloroform:methanol solution and then air-dried for 24 h. Plant and insect material and meat were cleaned in distilled water, freezedried, and then ground to a powder using a dental-amalgam mill. Approximately 1 mg of each sample was loaded into tin cups for isotopic measurement. Our mass spectrometer consisted of a Europa Robo Prep combustion system interfaced with a Europa Tracermat continuous-flow isotope-ratio mass spectrometer. Samples were analyzed as five unknowns separated by two albumin standards. Based on thousands of standard measurements, we estimate our analytical error to be ±0.1‰ for 13C measurements and ±0.3‰ for 15N measurements. We present our stable-isotope results in δ notation as the deviation from parts per thousand (‰) according to the following relationship: [1]

δX = [(Rsample / Rstandard) – 1] × 1000

where X is 13C or 15N and R is the corresponding ratio 13C/12C or 15 14 N/ N. Rstandard for 13C and 15N corresponds to the Pee Dee Belemnite (PDB) standard and AIR, respectively. When animals feed on two food types that differ isotopically, a simple two-source mixing model can be used to ascertain the relative dependence of a consumer on each dietary source. Here the average isotopic values of dietary alternatives are referred to as isotopic end-points. In the case of inland terrestrial bears with no access to marine protein (i.e., salmon), we were interested in the relative dependence of individuals on meat and plant material. The relative contributions of animal and plant protein assimilated in the tissues of bears can be derived as follows: © 2000 NRC Canada



1334

Can. J. Zool. Vol. 78, 2000 Table 1. Stable-isotope values for bear hair and samples of potential bear foods in the upper Columbia River basin, British Columbia, 1994–1996. δ15N (‰)

δ13C (‰) Sample Research bears Male black Female black Male grizzly Female grizzly Management bears Male black Female black Female grizzly Cubs in den Potential prey Spring beauty Fireweed (Epilobium angustifolium) Soapberry Glacier lily (Erythronium montanum) Elk (Cervus elaphus) Moose (Alces alces) Mountain goat Mule deer (Odocoileus hemionus) White-tailed deer Ants (Formicidae)

n

Mean ± SD

Range

Mean ± SD

61 25 26 17

–23.2±0.5 –22.9±0.5 –23.1±0.4 –23.3±0.5

–24.4 –24.1 –24.1 –24.1

–21.9 –21.7 –22.2 –22.4

3.1±0.7 3.4±1.0 3.1±0.8 2.1±1.1

1.3 1.4 1.9 0.2

9 1 7 3

–23.0±0.5 –22.6 –22.7±1.06 –24.5±0.5

–23.9 to –22.1

3.8±1.0 3.6 5.0±1.9 6.6±0.3

2.0 to 5.0

7 5 7 7 1 2 2 1 1 4

–27.7±1.1 –28.8±1.3 –28.1±0.9 –26.4±0.6 –25.5 –25.7 –23.7 –26.8 –25.9 –25.5±0.3

to to to to

–23.9 to –21.0 –24.0 to –24.9 –29.5 –30.8 –29.4 –27.2

to to to to

–26.4 –27.5 –26.7 –25.8

–25.8 to –25.5 –24.1 to –23.3

–25.8 to –25.1

–3.9±0.6 –1.1±2.0 0.1±0.6 –3.4±0.3 3.1 2.9 1.7 3.8 4.6 3.8±0.3

Range to to to to

4.8 5.6 4.7 4.3

2.8 to 7.7 6.4 to 6.8 –2.7 –3.2 –0.8 –4.6

to to to to

–4.5 1.4 1.9 –2.2

2.9 to 2.9 1.6 to 1.8

3.5 to 4.2

Note: Bear samples include replicates of some individuals taken at least 1 year apart. Ant samples are separate bulk samples representing several individuals. Ungulate samples are all meat. Plant samples correspond to portions of the plant consumed by bears.

[2]

δ 15Ndiet = Pa (δ 15Na) + Pp (δ 15Np)

where δ15Ndiet is the isotopic value of the diet, δ15Na is the isotopic value of animal tissue (including invertebrate prey), δ 15Np is the isotopic value of plant foods, Pa is the proportion of the assimilated diet derived from animal sources, and Pp is the proportion of the assimilated diet consisting of plant foods. Using the relationship Pp = 1 – Pa and solving for Pa, eq. 2 becomes [3]

Pa = (δ 15Ndiet – δ 15Np) / (δ 15Na – δ 15Np)

Hilderbrand et al. (1996) raised black bears in captivity on diets ranging from pure plant material (apples) to pure animal tissue (mule deer and salmon). They derived a linear relationship between δ15Ndiet and δ15N values of bear plasma that is probably appropriate for all bear tissues except adipose tissue. Generalizing the relationship derived by Hilderbrand et al. (1996) for plasma to bear hair, we substituted the assumed relationship [4]

δ 15Nhair = 4.76 + 0.91 (δ 15Ndiet)

into eq. 3 to obtain [5]

Pa = [(δ 15Nhair – 4.76) / 0.91 – δ 15Np] / (δ 15Na – δ 15Np)

Results Both bear species showed a considerable range in their stable-isotope values for hair (Table 1, Fig. 1). For black bears there was no relationship between δ15N and δ13C values (r2 = 0.03), whereas for grizzly bears a strong positive relationship was found (r2 = 0.63, p < 0.01) that was driven largely by high δ15N and δ13C values for the hair of a few management bears (Fig. 1).

The δ15N and δ13C values from research bears showed a significant interaction between species and sex (MANOVA, Wilk’s λ = 0.873, F[2,124] = 9.04, p < 0.001). Using multiple comparisons of means with α adjusted according to the Bonferroni criterion, we determined that male grizzly bears and black bears of both sexes had similar δ15N values but that female grizzly bears had lower δ15N values (Table 1). The δ13C values differed only between female grizzly bears and female black bears (Table 1). Of the management bears, only samples of female grizzly (n = 7) and male black (n = 9) bears were large enough to be included with research bears of the same species and sex in a MANOVA based on the reason for capture (management or research). There was a significant interaction between reason for capture and sex–species group (female grizzly bears, male black bears; Wilk’s λ = 0.847, F[2,89] = 8.06, p = 0.001). Multiple comparisons of means suggested that δ15N values for management bears of the two species were equal, but were higher than those for female grizzly or male black research bears. Once again, research male black bears had higher δ15N values than research female grizzly bears. There was no difference in δ13C values between management and research bears. Our small sample of hair from black bear cubs from dens in the upper Columbia River basin was more enriched in 15N and tended to be more depleted in 13C than samples from adults of this species (Table 1). We confirmed that this relationship generally held for δ15N values for hair from mother– cub groups sampled in dens in northern Minnesota (Table 2). In all the family groups, hair δ15N was enriched in cubs relative to their mothers (average of mean enrichment values per © 2000 NRC Canada



Hobson et al.

1335

Fig. 1. Distribution of δ15N and δ13C values for hair of black and grizzly bears inhabiting the upper Columbia River basin, British Columbia, 1994–1996.

group = 2.5 ± 1.2‰, n = 8). In six of eight mother–cub groups, hair δ13C was also enriched in cubs over that of their mothers (average of mean enrichment values per group = 1.4 ± 0.9‰, n = 6); however, in the other two groups, hair δ13C was depleted in cubs relative to their mothers (average of mean depletion values per group = –0.8‰, n = 2). While there was considerable between-group variance in mother– cub enrichment for both isotopes, within-group variance was remarkably low (Table 2). Potential foods of bears ranged widely in stable-isotope values (Table 1). In general, plant foods were more depleted in both 15N and 13C than herbivore tissues. Ants were more enriched in 15N than mountain goat, moose, and elk, likely

reflecting their more omnivorous diets. However, white-tailed deer were the most enriched. Mean δ15N values for plants (δ15Np) and meat (δ15Na) can be used to solve eq. 5. Based on the potential food values listed in Table 1, we estimated (unweighted) average values for δ15Na and δ15Np to be 3.3 and –2.1‰, respectively. Assuming further that negative estimates of Pa indicated no animal tissue in the diet and values exceeding 1 indicated 100% animal tissue (e.g., Hilderbrand et al. 1996; Jacoby et al. 1999), vertebrate or invertebrate foods contributed detectable amounts to the diet of 62% of the research male black bears, 72% of the research female black bears, and 46% of the research male grizzly bears, but only 18% of the research female grizzly bears. Using data © 2000 NRC Canada



1336

Can. J. Zool. Vol. 78, 2000 Table 2. Stable-isotope values for hair of black bear mother–cub groups sampled from dens in Minnesota in 1998–1999 and their relative differences. δ15N (‰)

δ13C (‰) Group

Mother

Cub

∆δ13C (‰)

Mother

Cub

∆δ15N (‰)

1 2

–22.5 –22.8 –22.4

4

–22.7

5

–22.7

6

–23.9

7

–22.8

8

–22.5

+1.9 +1.9 +2.5 +0.5 +0.4 +1.2 +1.0 +1.0 +0.2 +0.7 +0.3 –0.3 –0.3 –1.3 –1.1 +2.2 +1.5

3.9 3.7

3

–20.6 –20.9 –20.3 –21.9 –22.0 –21.5 –21.7 –21.7 –22.5 –22.0 –22.4 –24.2 –24.2 –24.1 –23.9 –20.3 –21.0

7.7 7.6 7.5 5.0 4.8 7.7 7.6 7.5 7.6 7.5 7.4 7.2 7.1 5.2 5.2 7.3 7.5

+3.8 +3.9 +3.8 +0.4 +0.2 +3.1 +3.0 +2.9 +2.4 +2.3 +2.2 +2.4 +2.3 +1.4 +1.4 +3.1 +3.3

based on individual bears (i.e., not averages as presented in Table 1), the average contribution of meat and (or) invertebrates to the diet was 8.3% for research male black bears, 15.0% for research female black bears, 9.9% for research male grizzly bears, and only 3.0% for research female grizzly bears. Our analyses suggested that 82% of the management bears had a detectable amount of meat and (or) invertebrates in their diet, the average contribution being 31%. The isotope-mixing model used here is obviously sensitive to choice of isotopic plant and animal dietary end-points. The above calculations are appropriately based on unweighted mean values for a number of species that we felt were potential dietary components in our study area. However, to demonstrate the sensitivity of the mixing model to choice of local end-point values, we present the extreme example of research male grizzly bears consuming exclusively (i) plants and meat with the highest mean δ15N values (i.e., soapberry (Shepherdia canadensis) and white-tailed deer (Odocoileus virginianus)), (ii) plants and meat with the lowest mean δ15N values (i.e., spring beauty (Claytonia lanceolata) and mountain goat (Oreamnos americanus)), (iii) plants with high and meat with low mean δ15N values (i.e., soapberry and mountain goat), and (iv) plants with low and meat with high mean δ15N values (i.e., spring beauty and white-tailed deer). For these four scenarios, the mean percent animal contribution to the diet (based on data from individual bears) was 0, 37.5, 0, and 24.7%, respectively.

Discussion Our isotope investigation indicates trophic segregation between female grizzly bears and all other bears of both species inhabiting the upper Columbia River basin. Few female grizzly bears consumed meat or invertebrates in amounts sufficient to be detected using our isotopic model. In our study area, female grizzly bears were found most often in remote, higher elevation habitats (B.N. McLellan and J.G.

4.6 4.6

5.2

4.8 3.8 4.2

Woods, unpublished data). Although the abundance of all mammals in the study area has not been documented, large mammals appeared to be relatively rare at higher elevations. There is some evidence that soil δ15N may become depleted with altitude (Mariotti et al. 1980) and that these patterns can be reflected in local food webs (Gröcke et al. 1997). We do not know if this is the case in our study area, but if it is, greater depletion of δ15N in female grizzly bears would not necessarily mean lower consumption of animal protein relative to bears inhabiting lower altitudes. Smaller mammals, such as marmots (Marmota caligata) and ground squirrels (Spermophilus columbianus) were common in these higher basins, and female bears have often been seen digging for them. Male grizzly bears, with their larger ranges (Woods et al. 1999), not only travel more but also are found at lower elevations more often than female grizzly bears (B.N. McLellan and J.G. Woods, unpublished data) and thus males may encounter large mammals more often. Isotopic measurements from ground squirrels at high altitudes, as well as from soils or plants along an altitudinal gradient, are required to fully resolve the question of whether female grizzly bears actually consume less animal protein, as is suggested by our isotopic approach. Ants are a major food for many black bear populations (Hatler 1972; MacHutchon 1989; Holcroft and Herrero 1991; Noyce et al. 1997), and adjacent to our study area they were the main food item during June and July (Raine and Kansas 1990). Grizzly bears also consume ants (Hamer and Herrero 1987; Mattson et al. 1991; McLellan and Hovey 1995), but these were not the dominant food during any season. Because a higher proportion of black bears had detectable amounts of animal protein in their diet than male or female grizzly bears, it is probable that ants, which are more ubiquitous than large mammals, made up a higher portion of the black bears’ diet in our study area and that consumption of ants is reflected in the δ15N values of black bear hair. As expected, management bears had more animal protein in their © 2000 NRC Canada



Hobson et al.

diet than research bears. These bears risked a close association with people to gain access to these high-quality foods. Recently, Jacoby et al. (1999) used stable carbon and nitrogen isotope analysis of hair and bone to investigate trophic relationships among populations of black and grizzly bears in several areas in the western United States and Alaska. Much of Glacier National Park and all of the Cabinet/Yaak of Montana are, like our study area, in the Columbia River drainage. Like the bears in our study, recent (post 1996) samples of subadult male and female grizzly (n = 13), adult female grizzly (n = 2), and black (n = 13) bears from these sites had low mean δ15N values (2.6–3.9‰; Jacoby et al. 1999; Table 1). However, adult males at these sites (n = 9) had higher mean δ15N values (4.5–5.2‰), and contemporary samples from adult males in the Greater Yellowstone and Blackfeet and Flathead Indian Reservations (n = 26) had considerably higher δ15N values (7.4–7.6‰). This indicates that adult males sampled by Jacoby et al. (1999), particularly those in Glacier National Park, had higher levels of animal protein in their diet than the bears in our sample. The isotopic model used by Jacoby et al. (1999) differed from ours in that they estimated plant isotope dietary endpoints based on herbivores in the area. Our investigation suggests that this may be inappropriate because ungulates may have fed on foods that differed isotopically from those preferred by bears. For example, while we found plant foods of bears with an average (unweighted) δ15N value of –2.1‰, our estimate, if it were based solely on herbivores in our area, would have suggested a value of –0.8‰. Using our value of δ15Np (–2.1‰) and the δ15Na value of 3.4‰ measured by Jacoby et al. (1999) for ungulate meat in their area, our isotope model provides generally higher estimates of animalprotein input for the Glacier National Park and Cabinet/ Yaak samples. Without a detailed isotopic survey of plant foods of bears from these areas, it is currently impossible to know which estimates of animal-protein input into the diets of these bears are correct. Importantly, our comparison simply serves to suggest that the isotopic models used are quite sensitive to the choice of isotopic end-point values, and it is clearly more appropriate to use estimates based on known bear foods than those based on more indirect methods. Certainly, estimates of the contribution of meat to the diets of bears across their range, as recently provided by Hilderbrand et al. (1999) and based on local herbivore end-points, should be interpreted cautiously. While such studies are useful in providing general trends, they may not provide useful absolute dietary estimates. This is a general problem in applying the stable-isotope technique to the reconstruction of animal diets in cases where individuals may have access to prey that varies considerably in isotope values within so-called dietary end-points. We have provided an example of research male grizzly bears from our study area involving four scenarios of populations feeding exclusively on components of plant and animal diets that represented combinations of high and low stable isotope values. This exercise, although representing a worst case scenario, illustrates the sensitivity of the models to field sampling of appropriate dietary items and our ultimate knowledge of how best to estimate dietary end-point values (Schwarcz 1991; Hobson 1993). The overlap between ants and ungulate meat in the δ15N values that we measured

1337

further underlines the need to interpret isotopic models with care. For example, currently we cannot distinguish isotopically between those bears that feed heavily on ants and those that feed heavily on ungulates, despite the obvious ramifications for ecological interpretation. The above discussion underlines the need for caution when using stable-isotope models in ecological research. In the case of bears that have access to both plant and animal foods, there is the additional problem of interpreting isotope data because nutritional pathways involving proteins and carbohydrates may be different (Ambrose and Norr 1993; Tieszen and Fagre 1993; Hobson and Stirling 1997). Hilderbrand et al. (1996) provided valuable evidence that the δ15N and δ13C values for blood plasma correspond closely to those for bear diets ranging in composition from 100% plant (i.e., apples) to 100% meat, and this has provided a firm empirical basis for the isotope models used here and in other studies. However, in cases where plant carbon derived from carbohydrates is used largely in the manufacture of lipids instead of carbon from plant proteins contributing to bear protein (e.g., plasma, hair, muscle), δ13C values in bear tissues will depend on tissue type (i.e., protein versus lipid) and the δ13C value of the plant macronutrient contributing to that tissue. Our isotopic analysis of bear hair provides information on the relative contributions of protein sources, rather than carbohydrate or lipid sources, to bears. Thus, we need to realize that estimates of meat in the diet may be overrated, depending on whether, nutritionally, the bear is primarily acquiring protein or storing fat. As with other applications of the stable-isotope technique, understanding animal nutritional ecology is crucial to interpreting isotope data (e.g., Hobson and Stirling 1997). If plant protein and carbohydrate differ in their δ13C values, then the overall plant δ13C value would depend on plant carbohydrate content, resulting in a broad distribution of plant δ13C values even though the range of plant δ15N values remained fairly narrow. In our study, we found a weak coupling between δ15N and δ13C values for a bear population that was consuming largely plant material. Including the bears that consumed animal protein created trophic enrichment in both 15N and 13C that drove the positive relationship for grizzly bears, probably because both carbon and nitrogen show positive isotopic fractionation with trophic level, although the effect is weaker with carbon. Interestingly, in their examination of stable isotope ratios across bear populations representing a broad range of dependence on meat, Hilderbrand et al. (1999) present a figure that shows much higher variance in δ13C and δ15N values for bear populations that are largely dependent on plants than for populations dependent on meat. We suggest that the strong positive correlation between δ15N and δ13C values in bear tissues indicates that the nutritional pathways of carbon and nitrogen are coupled (i.e., dietary δ13C values correspond closely to those for dietary protein) and that this is much more likely in populations consuming primarily meat than in those consuming primarily plant foods. Rather than relying heavily on isotopic models to infer meat consumption by bears within populations having access to both plants and meat, an examination of the dispersion of δ15N and δ13C values in bear tissues should reveal those individuals that rely to a greater extent on meat in © 2000 NRC Canada



1338

their diet. In our study, the grizzly bears shown in the upper right quadrant of the lower panel of Fig. 1 likely consumed substantial animal protein, probably meat. Nursing black bear cubs in dens generally showed isotopic enrichment in their hair compared with the hair of their mothers. This was expected, especially for 15N, because cubs depend exclusively on their mother’s milk for protein derived from parental tissues, which thus constitutes a trophic enrichment step. While few other studies have included an investigation of such effects in mother–offspring pairs, there is some evidence that maternal transfer of nutrients leads to isotopic enrichment in offspring. For example, the δ15N values in the muscle tissue of nursing northern fur seal (Callorhinus ursinus) pups were 1.9‰ higher than those of their mothers (Hobson et al. 1997), and the δ15N values of tooth annuli from Steller sea lions (Eumetopias jubatus) were higher during their first year of life (Hobson and Sease 1998). Fingernails from human infants were also found to be enriched in 15 N by 2.4‰ over those of their mothers (Fogel et al. 1989). Bocherens et al. (1994) inferred a nursing effect in decidual teeth of extinct cave bears (Ursus spelaeus) and, more recently, Nelson et al. (1998) speculated that enriched δ15N and depleted δ13C in the bone collagen of young versus adult cave bears was related in part to their dependence on mother’s milk. In the case of our analysis of hair from black bear mothers and cubs, it is possible that some decoupling of the parent–offspring isotopic-enrichment effect occurred for some groups because the isotope values for hair from the mothers only represent the diet during the period of hair growth and may not always correspond to the diet when nutrients are stored for transfer to cubs during nursing. In addition, bear milk consists primarily of protein and lipids, and the relative mobilization of these two sources of carbon to bear hair is expected to strongly influence the δ13C values of cub hair, since lipids are typically more depleted in 13C than proteins (Gearing 1991). More research into the factors determining isotopic relationships between the tissues of mothers and their nursing cubs would be extremely useful.

Acknowledgements Stable-isotope analyses were provided by G. Parry of the Soil Science Stable Isotope Facility at the University of Saskatchewan. Additional assistance in the preparation of stableisotope samples was provided by S. van Wilgenburg and P. Healy. Funding was provided by K.A.H. and the Canadian Wildlife Service. Bear hair was collected as part of the West Slopes Bear Research Project, jointly supported by the British Columbia Ministry of Environment, Lands, and Parks (Grizzly Bear Conservation Strategy); the British Columbia Ministry of Forests; the Columbia Basin Fish and Wildlife Compensation Program; Forest Renewal British Columbia; the Friends of Mount Revelstoke and Glacier National Parks; Parks Canada (Glacier, Kootenay, Mount Revelstoke, Yoho, Ottawa); the Southern British Columbia Guides and Guide Outfitters Association; the University of Alberta; and The University of British Columbia. In addition, we thank the individuals who assisted us in capturing research bears, and the Conservation Officers of the Ministry of Environment, Lands, and Parks who made our study of management bears possible. Roger Ramcharita assisted with the collection of

Can. J. Zool. Vol. 78, 2000

bear food samples and Kelly Stalker with hair sampling. Conan Phelan assisted with data-base management. Black bear hair samples from Minnesota were provided by Karen Noyce of the Minnesota Department of Natural Resources.

References Ambrose, S.H., and Norr, L. 1993. Experimental evidence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone collagen and carbonate. In Prehistoric human bone: archaeology at the molecular level. Edited by J. Lambert and G. Grupe. Springer-Verlag, Berlin. pp. 1–37. Aune, K.E., and Kasworm, W.F. 1994. Comparative ecology of black and grizzly bears on the Rocky Mountain Front, Montana. In Bears—Their Biology and Management: Proceedings of the Ninth International Conference on Bear Research and Management, Missoula, Mont., February 1992. Edited by J. Claar and P. Schullery. International Association for Bear Research and Management, Washington, D.C. pp. 365–374. Bocherens, H., Fizet, M., and Mariotti, A. 1994. Diet, physiology and ecology of fossil mammals as inferred by stable carbon and nitrogen isotope biogeochemistry: implications for Pleistocene bears. Palaeogeogr. Palaeclimatol. Palaeoecol. 107: 213–225. DeNiro, M.J., and Epstein, S. 1978. Influence of diet on the distribution of carbon isotopes in animals. Geochim. Cosmochim. Acta, 42: 495–506. DeNiro, M.J., and Epstein, S. 1981. Influence of diet on the distribution of nitrogen isotopes in animals. Geochim. Cosmochim. Acta, 45: 341–351. Fogel, M.L., Tuross, N., and Owsley, D.W. 1989. Annual report of the director of the Geophysical Laboratory, Carnegie Institute, Washington, 1988–1989. Geophysical Laboratory, Carnegie Institute, Washington, D.C. Gearing, J.N. 1991. The study of diet and trophic relationships through natural abundance 13C. In Carbon isotope techniques. Edited by D. Coleman and B. Fry. Academic Press, New York. pp. 201–218. Gröcke, D.R., Bocherens, H., and Mariotti, A. 1997. Annual rainfall and nitrogen isotope correlation in macropod collagen: applications as a paleoprecipitation indicator. Earth Planet. Sci. Lett. 153: 279–285. Hamer, D., and Herrero, S. 1987. Grizzly bear food and habitat in the front ranges of Banff National Park, Alberta. In Bears—Their Biology and Management: Proceedings of the 7th International Conference on Bear Research and Management, Williamsburg, Va., U.S.A., and Plitvice Lakes, Yugoslavia, February and March 1986. Edited by P. Zager. International Association of Bear Research and Management, Washington, D.C. pp. 199–213. [Available from Michael R. Pelton, Department of Forestry, Wildlife and Fisheries, The University of Tennessee, P.O. Box 1071, Knoxville, TN 37901-1071, U.S.A.] Hatler, D.F. 1972. Food habits of black bears in interior Alaska. Can. Field-Nat. 86: 17–31. Hilderbrand, G.V., Farley, S.D., Robbins, C.T., Hanley, T.A., Titus, K., and Servheen, C. 1996. Use of stable isotopes to determine diets of living and extinct bears. Can. J. Zool. 74: 2080–2088. Hilderbrand, G.V., Schwartz, C.C., Robbins, C.T., Jacoby, M.E., Hanley, T.A., Arthur, S.M., and Servheen, C. 1999. The importance of meat, particularly salmon, to body size, population productivity, and conservation of North American brown bears. Can. J. Zool. 77: 132–138. Hobson, K.A. 1993. Trophic relationships among high Arctic seabirds: insights from tissue-dependent stable-isotope models. Mar. Ecol. Progr. Ser. 95: 7–18. © 2000 NRC Canada



Hobson et al. Hobson, K.A. 1999. Tracing origins and migration of wildlife using stable isotopes: a review. Oecologia. 120: 314–326. Hobson, K.A., and Sease, J.L. 1998. Stable isotope analyses of tooth annuli reveal temporal dietary records: an example using stellar sea lions. Mar. Mamm. Sci. 14: 116–129. Hobson, K.A., and Stirling, I. 1997. Terrestrial foraging by polar bears during the ice free period in western Hudson Bay: metabolic pathways and limitations of the stable isotope approach. Mar. Mamm. Sci. 13: 359–367. Hobson, K.A., and Welch, H.E. 1992. Determination of trophic relationships within a high Arctic marine food web using stableisotope analysis. Mar. Ecol. Progr. Ser. 84: 9–18. Hobson, K.A., Sease, J.L., Merrick, R.L., and Piatt, J.F. 1997. Investigating trophic relationships of pinnipeds in Alaska and Washington using stable isotope ratios of nitrogen and carbon. Mar. Mamm. Sci. 13: 114–132. Holcroft, A.C., and Herrero, S. 1991. Black bear, Ursus americanus, food habits in southwestern Alberta. Can. Field-Nat. 105: 335–345. Jacoby, M.E., Hilderbrand, G.V., Servheen, C., Schwartz, C.C., and Arthur, S.M. 1999. Trophic relationships of brown and black bears in several western North American ecosystems. J. Wildl. Manag. 63: 921–929. Kurten, B., and Anderson, E. 1980. Pleistocene mammals of North America. Columbia University Press, New York. MacHutchon, A.G. 1989. Spring and summer food habits of black bears in the Pelly River Valley, Yukon, Canada. Northwest Sci. 63: 116–118. Mariotti, A., Pierre, D., Vedy, J.C., Bruckert, S., and Guillemot, J. 1980. The abundance of natural nitrogen-15 in the organic matter of soils along an altitudinal gradient. Catena, 7: 293–300. Mattson, D.J., and Reinhart, D.P. 1995. Influences of cutthroat trout (Oncorhynchus clarki) on behaviour and reproduction of Yellowstone grizzly bears (Ursus arctos), 1975–1989. Can. J. Zool. 73: 2072–2079. Mattson, D.J., Blanchard, B.M., and Knight, R.R. 1991. Food hab-

1339 its of Yellowstone grizzly bears, 1977–1987. Can. J. Zool. 69: 1619–1629. McLellan, B.N. 1993. Competition between black and grizzly bears as a natural population regulating factor. U.S. Dep. Agric. Tech. Rep. NPS/NRWR-93/12. pp. 111–116. McLellan, B.N., and Hovey, F.W. 1995. The diet of grizzly bears in the Flathead River drainage of southeastern British Columbia. Can. J. Zool. 73: 704–712. Nelson, D.E., Angerbjorn, A., Liden, K., and Turk, I. 1998. Stable isotopes and the metabolism of the European cave bear. Oecologia, 116: 177–181. Noyce, K.V., Kannowski, P.B., and Riggs, M.R. 1997. Black bears as ant-eaters: seasonal associations between bear myrmecophagy and ant ecology in north-central Minnesota. Can. J. Zool. 75: 1671–1686. Peterson, B.J., and Fry, B. 1987. Stable isotopes in ecosystem studies. Annu. Rev. Ecol. Syst. 18: 293–320. Raine, R.M., and Kansas, J.L. 1990. Black bear seasonal food habits and distribution by elevation in Banff National Park, Alberta. In Bears—Their Biology and Management: Proceedings of the 8th International Conference on Bear Research and Management, Victoria, B.C., 20–25 February 1989. Edited by L.M. Darling and W.R. Archibald. International Association for Bear Research and Management, Vancouver, B.C. pp. 297–304. [Available from Michael R. Pelton, Department of Forestry, Wildlife and Fisheries, The University of Tennessee, P.O. Box 1071, Knoxville, TN 37901-1071, U.S.A.] Schwarcz, H.P. 1991. Some theoretical aspects of isotope paleodiet studies. J. Archaeol. Sci. 18: 261–275. Tieszen, L.L., and Fagre, T. 1993. Effect of diet quality and composition on the isotopic composition of respiratory CO2, bone collagen, bioapatite, and soft tissues. In Prehistoric human bone: archaeology at the molecular level. Edited by J. Lambert and G. Grupe. Springer-Verlag, Berlin. pp. 121–155. Woods, J.G., Paetkau, D., Lewis, D., McLellan, B.N., Proctor, M., and Strobeck, C. 1999. Genetic tagging free-ranging black and brown bears. Wildlife Soc. Bull. 27: 616–627

© 2000 NRC Canada
