Behavioral Ecology Advance Access originally published online on October 19, 2006
Behavioral Ecology 2007 18(1):12-20; doi:10.1093/beheco/arl066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
The influence of conspecifics and predation risk on the vigilance of elk (Cervus elaphus) in Yellowstone National Park
a Department of Biological Sciences, Idaho State University, Pocatello, ID, USA b Department of Biological Sciences, Clemson University, Clemson, SC, USA
Address correspondence to M.A. Lung, who is now at Department of Natural and Environmental Sciences, Western State College, Gunnison, CO 81231, USA. E-mail: mlung{at}western.edu.
Received 19 May 2006; revised 20 July 2006; accepted 2 August 2006.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Vigilance in socially foraging animals (e.g., elk) may serve to detect predators (i.e., reduce predation risk) or to monitor the behavior of conspecifics (i.e., reduce conspecific risk and/or increase reproductive benefits). These potential benefits and costs of vigilance may not be equal among different genders and age classes foraging together. We observed male and female elk from 3 age classes (yearlings, nonreproductive, reproductive) during 2 seasons (calving season, breeding season) that vary in social interactions (i.e., conspecific risk) and in 2 regions of Yellowstone National Park that varied in predation risk due to density of wolves. This study was designed to determine the potential functional benefit of vigilance across a range of herd sizes. If vigilance serves to monitor the behavior of conspecifics, we expected it to increase in the fall breeding season along with aggressive behaviors, regardless of changes in predation risk. If vigilance serves to detect predators, we expected it to increase in regions with wolves regardless of changes in conspecific risk. Adult male vigilance and aggression increased during the fall breeding season, but yearling males had significantly lower levels of vigilance. Adult female vigilance increased during the calving season, with increasing predator encounter risk, decreasing herd size, and edge position within the herd. Yearling female vigilance decreased with herd size but was lower than adult females. We conclude that the primary benefit of vigilance for male elk is to monitor conspecifics, but the primary benefit of vigilance for female elk is to detect potential predators.
Key words: elk behavior, predation risk, ungulate behavior, vigilance.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
One of the most important benefits of group living is antipredatory vigilance, and many previous studies have found vigilance to vary with predation risk, group size, sex, and social status (Bertram 1978
; Pulliam and Caraco 1984; Lima and Dill 1990; Roberts 1996). However, vigilance may also be beneficial in monitoring the behavior of conspecifics, especially when differences in rank are established and reinforced with acts of aggression or when foraging information is gained by direct observation (Quenette 1990; Beauchamp 2001). Like antipredatory vigilance, the benefits of social monitoring vigilance are also likely to be influenced by group size, sex, and social status (Treves 2000; Robinette and Ha 2001). In some cases, antipredatory and social monitoring are likely to influence vigilance in the same direction. For example, juveniles may be more vigilant than adults because of an increased risk of predation or an increased risk of conspecific aggression (Alberts 1994; Yaber and Herrera 1994). In other cases, these factors may influence vigilance in opposite directions. For example, antipredatory vigilance might decrease with increasing group size, whereas social monitoring vigilance might increase (Beauchamp 2001). These differences in the predicted influence of social factors can help us to distinguish the functional significance of vigilance and its role in the maintenance of social groups (Hirsch 2002).
Vigilant behaviors have largely been attributed to variation in the risk of predation (Pulliam 1973; Lima 1987); however, it is becoming evident that other benefits such as gaining knowledge of food patches, keeping track of mates, or avoiding conspecific aggression may also be important reasons for vigilance (Elgar 1989; Roberts 1996; Beauchamp 2001; Robinette and Ha 2001). In animals with complex social relationships, vigilance may be directed at conspecifics to reduce risks associated with proximity to conspecifics (conspecific risk). Evidence for social influences on vigilance exists for birds (Caraco 1979; Knight SK and Knight RL 1986; Waite 1988a, 1988b; Artiss and Martin 1995; Robinette and Ha 2001), rodents (Monaghan and Metcalfe 1985; Roberts 1988; Yaber and Herrera 1994; Tchabovsky et al. 2001), and primates (Caine and Marra 1988; Baldellou and Henzi 1992; Alberts 1994; Hirsch 2002).
Ungulates are a particularly good model system to study the interaction between antipredatory and social monitoring hypotheses (Caro et al. 2004), for vigilance has been found to vary with predation pressure (Hunter and Skinner 1998; Laundré et al. 2001; Wolff and Van Horn 2003), sex/social status (Lipetz and Bekoff 1982; Frid 1997; Laundré et al. 2001), herd size (Underwood 1982; Dehn 1990; Bednekoff and Ritter 1994; Childress and Lung 2003), distance to obstructive cover (Berger 1978; Frid 1997), and distance to nearest neighbor (Underwood 1982). The majority of these studies have assumed that vigilance is primarily, if not exclusively, for the benefit of reducing predation risk rather than reducing the cost of intraspecific aggressive interactions or the loss of potential mates. Many ungulates, in addition, are also known to have well-developed social hierarchies where intraspecific aggression determines dominance rank and reproductive success varies greatly among males (Geist 1974; Clutton-Brock et al. 1982; Freeman et al. 1992).
Recently, several studies have examined the impact of the gray wolf (Canis lupus) reintroduction in Yellowstone National Park on the behavior of their ungulate prey, elk (Cervus elaphus) and bison (Bison bison) (Laundré et al. 2001; Mech et al. 2001; Childress and Lung 2003; Eberhardt et al. 2003; Wolff and Van Horn 2003; Smith et al. 2004; Creel and Winnie 2005), and their indirect effect on the plant community (Ripple et al. 2001; Beschta 2003; Ripple and Beschta 2004; Creel et al. 2005). In general, both bison and elk increase vigilance in the regions of the park with a high probability of encounter with wolves (Laundré et al. 2001; Childress and Lung 2003; Wolff and Van Horn 2003). However, the influence of herd size, age–sex class, herd composition, position in the herd, distance from cover, and nearest neighbor distance were highly variable and not always consistent with the hypothesis that vigilance serves only to reduce the risk of predation. During the calving season, the greatest difference in vigilance was observed between males and females of reproductive age (Laundré et al. 2001; Childress and Lung 2003). Females with newborn calves spent >25% of time scanning, whereas reproductive males spent <10% of time scanning. This result is consistent with the age–sex classes most often killed by wolves during the study years (calves > females > males; see Smith et al. 2000, 2004; Mech et al. 2001), but it does not explain why male vigilance does not vary with either predator encounter risk or herd size (Childress and Lung 2003; Wolff and Van Horn 2003).
In this study, we compared the predictions of 2 nonexclusive hypotheses (social monitoring vs. antipredatory) for the vigilant behavior exhibited by elk in Yellowstone National Park to determine if the benefit of vigilance differed between the sexes. We predicted that antipredatory vigilance should increase 1) in regions of high predator encounter risk, 2) with reproductive status (yearlings—lowest, reproductive adults—highest), 3) with deceasing herd size, 4) with position in the herd (middle—lowest, edge—highest), and 5) with increasing distance to cover. Alternatively, we predicted that social monitoring vigilance should increase 1) with season (spring—lowest, fall—highest), 2) with reproductive status (yearlings—lowest, reproductive adults—highest), and 3) with decreasing nearest neighbor distance. We used a stepwise general linear model (GLM) and analyses of covariance (ANCOVAs) to determine the best predictor variables of percentage of time spent scanning for each sex and reproductive status separately to identify if one or both functional benefits of vigilance were important.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Subjects and study area
Elk are a particularly interesting study system for assessing the interaction between antipredator and social monitoring vigilance. Because of their complex seasonal social interactions, risk associated with conspecifics (conspecific risk) varies with season, age, sex, and dominance rank. In addition, elk in Yellowstone National Park, since the reintroduction of wolves, are exposed to predation risk that varies by region (predator encounter risk) and age–sex classes (attack/capture risk).
Differences in conspecific risk
Elk live in herds that vary seasonally in size, composition, and degree of social interactions (Geist 1974; Clutton-Brock et al. 1982). In the spring during the calving season, adult females with and without yearlings and newborn calves form "nursery herds." Adult males are solitary or form small groups (<10). In the fall during the breeding season, males establish dominance hierarchies and dominant males maintain and defend "harems" of adult females with and without yearlings and calves. Subordinate males form small "bachelor" herds and occasionally try to sneak copulations with harem females. Dominant males and females enjoy large differences in reproductive success over subordinates (Gibson and Guiness 1980; Clutton-Brock 1986). Aggressive interactions are much more common during the breeding season and are usually initiated by dominant individuals (Appleby 1983; Thouless 1990). Thus, the influence of conspecifics on individual elk (conspecific risk) should vary by season, sex, age, and dominance rank.
Differences in predator encounter risk
The probability of encountering a predator is primarily dependent upon the density and distribution of predators (Lima and Dill 1990). We observed elk in 2 regions of Yellowstone National Park—Norris Basin and the Northern Range (Figure 1). During our study period (1999–2001), these 2 regions differed in their relative use by wolves (Norris Basin—low wolf use; Northern Range—high wolf use). Wolf use was estimated by calculating wolf density and by using kernel estimators as an indicator of wolf distribution. Wolf density was calculated by dividing the number of wolves in a pack by pack territory size that encompassed or overlapped with our study areas (Table 1). Yellowstone Wolf Project records the number of wolves per pack, maintains collars on approximately 30% of wolves in the park, and determines wolf pack territories by calculating 95% minimum convex polygons of collared wolves (Smith et al. 2004). During the first 2 years of the study, no wolf packs were observed in the Norris Basin, but during the last 2 years, the territory of one wolf pack (Nez Perce) overlapped with the study region. However, due to the size of the territory, the density of wolves (with a mean pack size of 20) remained low (0.12 wolves per square kilometer). In contrast, all observations of elk in the Northern Range were made within wolf pack territories. During the study period, at least 3 wolf packs occurred in this study region with a mean density of 0.36 wolves per square kilometer.
|
|
To supplement estimations of wolf density, which provides a relatively uniform picture of predator locations, it was also important to estimate distribution of wolves within territories through kernel estimators. Distribution of wolves was calculated using kernel estimates (50% and 75%) of all radio locations of collared wolves (Figure 1). For our study, all wolf radio locations (east/north UTM—NAD 27) between 1999 and 2001 were entered onto a raster map of Yellowstone National Park. During the study years, Yellowstone Wolf Project collected radio locations of collared wolves at least 4 times per month. It is estimated that approximately 30% of wolves in a pack are collared (Smith et al. 2004
). This confirmed that elk observations in Norris Basin between 1999 and 2001 fell in areas with low probability of encounter with wolves, whereas all observations of elks in the Northern Range were made in areas with intermediate to high distribution of wolves.
In addition, this gradient of risk is supported by data on elk kills and predator sightings (Smith 1997, 1998; Smith et al. 1999, 2000, 2004; Mech et al. 2001; Childress and Lung 2003). For example, during 3 years of study equaling almost 1000 h of observations, no wolves were observed in Norris Basin, but wolves were observed weekly in the Northern range.
Differences in attack/capture risk
Elk in Yellowstone National Park during this study period also differed in their vulnerability to predators due to season, sex, and age (attack/capture risk) with newborn calves being the most vulnerable and adult males the least vulnerable. In the spring, newborn calves are prey for predators including wolves, grizzly bears (Ursus arctos), and coyotes (Canis latrans) (Gunther and Renkin 1990; Gese and Grothe 1995) and are the most common prey of wolves (Smith et al. 2000). By the fall, however, calves are sufficiently swift to avoid predation by grizzlies and coyotes. Adult elk are prey of wolves but not usually grizzly bears or coyotes. Adult females and yearlings are killed by wolves during all seasons, but adult males are taken more often during late winter than during the other seasons (Mech et al. 2001).
In summary, we observed elk during 2 seasons (calving season, May–June; breeding season, September–November) in Yellowstone National Park. During the 2 observation seasons, we watched elk in 2 regions of the park that were at the extremes of predation risk (Northern Range—high encounter risk; Norris Basin—low encounter risk). These designations correspond to the "wolf use" and "non-wolf-use" areas of Laundré et al. (2001) and the high and low encounter risk areas of Childress and Lung (2003). A more detailed description of these geographic regions, the size and composition of their elk herds, and the type and frequency of their potential predators can be found in Childress and Lung (2003).
General methods
A total of 574 focal observations (302 females, 272 males) were made between 16 May and 26 June (calving season) and between 23 September and 5 November (breeding season) over 3 years (1999–2001). A herd was defined as a group of elk with a nearest neighbor distance of not more than 100 m regardless of their behavioral state. Individuals more than 100 m from another elk were not considered part of that herd. Elk were observed from alongside the park roads using binoculars (7 x 35) and spotting scopes (32 x 82) whenever they were visible and active between the hours of 0530–2130. For each herd observed, we recorded the date, time of day (before 1000, 1000–1400, after 1400 h), season (spring/fall), regional encounter risk (low/high), total herd size, distance to nearest cover (5, 10, 50, 100, 500, 1000 m), distance to the road (5, 10, 50, 100, 500, 1000 m), and the number of individuals in each sex and age class. We also recorded each focal individual's position in the herd (edge/middle) and distance to the nearest neighbor (1, 2–5, 5–10, >10 body lengths) at the beginning of each observation. Distance to the road was considered a nuisance variable to control for the impact of park visitors on elk vigilance near the road.
For the analysis of focal activity patterns and calculations of vigilance, elk were divided into 3 age classes within each sex. Females were defined as reproductive when a calf was present, nonreproductive when no calf was present, and yearling when noticeably smaller. There are some difficulties discerning female yearlings but generally during the fall for those female elk born in the previous spring. In most cases, those female elk recorded as yearlings were probably calves born that year. In some cases, it is possible to recognize females that were born the previous spring. During the spring, yearling females are easily discernable as they are half the size of adult females. Research on age–size dimensions in elk shows adult females average 310 kg with a 135 cm shoulder height. There are no height data for yearling females but spring yearlings average 140–160 kg and fall yearlings 220 kg (Hudson et al. 2002). To reduce misclassification of elk, we generally watch an elk herd for 15–30 min prior to recording observations. This allows us to recognize yearlings as they move in relation to other animals. To our knowledge, there are no recent studies that present size dimension data for Yellowstone elk.
Males were defined as reproductive when they had 5–6 antler points per side in the spring or when observed with a harem of females in the fall. Males were defined as nonreproductive when they had 2–4 antler points per side in the spring or when observed without a harem of females in the fall. Males were defined as yearlings when they had 1 antler point per side.
Observational methods
Elk activity was divided into 6 behavioral states: feeding, scanning, traveling, grooming, aggression, and resting. Feeding was defined as standing or walking slowly with the head below the level of the shoulder. Scanning was standing with the head at or above the shoulder level. Traveling was walking, trotting, or running with the head at or above the shoulder level. Grooming was licking or scratching oneself or another. Aggression was kicking, biting, or charging another elk with head fully raised. Resting was any behavior while lying on the ground. Scanning behavior was used as our estimate of vigilance. We recognize that animals have subtle behaviors associated with aggression (e.g., stare, ears back), but these are difficult to discern in the field as separate from scanning. We also recognize that animals may monitor conspecifics with head down while foraging and during other activities such as traveling. In addition, animals may be engaged in other activities while scanning (e.g., chewing, ruminating). However, scanning is the most common behavior used for evaluation in vigilance studies. A comparison of the trade-off between scanning and feeding behaviors can be found in Childress and Lung (2003).
Individual vigilance was estimated by a focal animal sampling rule (Martin and Bateson 1993) with behaviors continuously recorded for 15 min. Observations less than 3 min were excluded from the analysis. Focal individuals were haphazardly selected from the herd based on their sex and age class. In order to reduce the probability that the same individual elk was observed more than once, only 1–3 individuals were observed in each herd from different age classes and herds were only revisited on a future date if they contained more than 10 individuals. The length and number of scans per time observed was recorded and used to calculate average percentage of time scanning. An analysis of scan frequency and scan duration can be found in Childress and Lung (2003).
Statistical methods
In order to test the hypothesis that male and female vigilance is influenced by similar variables, we performed a GLM on each sex using SYSTAT 10's GLM procedure. We used a backwards stepwise model-building procedure with tolerance of 0.1 and a probability to enter and remove of 0.05 to minimize the impact of collinear main effects (Wilkinson 1996). The potential main effect variables included reproductive status (reproductive, nonreproductive, yearling), season (spring, fall), encounter risk (low, high), position in the herd (edge, middle), nearest neighbor distance (4 levels), distance to cover (6 levels), distance to road (6 levels), and time of day (3 levels). Dependent variables—percentage of time scanning and percentage of time aggression were arcsin square-root transformed, and the independent variable—herd size was natural log transformed in order to meet the assumptions of normality of residuals and homogeneity of variances. The analysis of the percentage of time aggression was performed only on those individuals that showed aggression in order to meet the assumption of homogeneity of variances.
In order to examine the interactions between significant main effects, we conducted 6 ANCOVAs on each sex—reproductive class using SYSTAT 10's ANCOVA procedure. Herd size (natural log transformed) was the covariate, and the main effects were season (spring, fall) and encounter risk (low, high). The dependent variable, percentage of time scanning, was arcsin square-root transformed in order to meet the assumptions of normality of residuals and homogeneity of variances.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The GLM for percentage of time scanning among female elk found that reproductive status, encounter risk, position in the herd, and herd size were all significant predictor variables of vigilance (Table 2). For males, however, only reproductive status and season were significant predictor variables. The models explained 24.5% of the variance for female vigilance and 23.3% for male vigilance.
|
Female vigilance marginally decreased with herd size (Table 2), but herd size had no effect on male vigilance (Figure 2). Both female and male vigilance was significantly higher (Table 2) for reproductive adults than for nonreproductive adults or yearlings (Figure 3). Male vigilance was significantly higher in the fall breeding season than in the spring calving season (Table 2, Figure 4b), but season had widely different effects on female vigilance (Figure 5). Female vigilance was significantly higher in high predator encounter risk regions (Table 2), but high encounter risk had no effect on male vigilance (Figure 5). Female vigilance was significantly lower for individuals in the middle of a herd (Table 2), but position in the herd had no effect on male vigilance.
|
|
|
|
Because reproductive status was highly significant for both males and females, we conducted individual ANCOVAs for each sex—reproductive class to look for significant interactions between the main effects of herd size, season, and encounter risk. The vigilance of reproductive females was significantly influenced by season (Table 3) with mothers spending twice as much time vigilant in the spring than fall (Figure 5A). The vigilance of nonreproductive females, however, decreased significantly with increasing herd size (Table 3) but was not influenced by season (Figure 5B).
|
Although these patterns are very different, both are consistent with antipredatory benefits of vigilance. Reproductive females with newborn calves in the spring must remain vigilant for predators in both high and low wolf encounter regions because ubiquitous coyotes and bears are also important predators of calves. However, once the calves are 6 months old, these "fall yearlings" depend less on their mother's vigilance. This explains why reproductive female vigilance decreases (Figure 5A), whereas yearling vigilance increases in the fall (Figure 5C,F). For both of these groups of females, season is a significant main effect (Table 3). The vigilance levels of female nonreproductives decreases significantly (Table 3) with increasing herd size and with decreasing wolf encounter risk (Figure 5B). This suggests that their vigilance serves primarily to decrease their own risk of predation by wolves. Overall, female vigilance is influenced by seasonal changes in the vulnerability of calves, regional differences in wolf encounter rates, and the antipredatory benefits of large herds.
The vigilance of reproductive males was significantly influenced by season (Table 3) with a 3- to 4-fold increase in the fall over their low spring levels (Figure 5D). The same pattern is also present for nonreproductive males (Table 3) with a 2-fold increase in vigilance in the fall (Figure 5E). Male aggression (percentage of time active) increased significantly in the fall breeding season (F1,42 = 11.75, P = 0.001) from 1% in the spring calving season to >7% in the fall breeding season (Figure 4a). Female aggression was unrelated to season or any other main effect variables. An increase in male vigilance and aggression in the fall is consistent with the predictions of the social monitoring hypothesis because adult males are competing with one another for access to the females and keeping track of harem females once obtained. In addition, we tested the influence of nearest neighbor distance on aggression. We found differences existed only for males and that aggression in the fall rut was marginally significant (F = 2.916, degrees of freedom [df] = 3,25; P = 0.053) for nonreproductive males and for yearling males in the spring (F = 3.11, df = 3,21; P = 0.048). Reproductive male vigilance would potentially benefit in detecting the approach of "sneaker" males or wandering females moving away from the harem. However, this data suggests that nonreproductive males and yearlings may benefit from vigilance by maintaining distances to avoid aggression.
Nonreproductive males have significantly lower levels of vigilance (Table 3) in larger herds. This pattern is consistent with the antipredatory benefit of vigilance but may be confounded with the differences in herd size associated with season. In the spring, nonreproductive males forage in herds with a median size of 8 individuals, but in the fall, the median herd size is down to 4 individuals. Thus, the increase in fall vigilance in nonreproductive males could be due to increased predation risk of smaller herds or the increased risk of aggression from conspecific males. Yearling male vigilance, like yearling female vigilance, increases in the fall in high wolf encounter areas (Figure 5F) consistent with the antipredatory benefit of vigilance.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results of this study suggest that vigilant behaviors in elk are influenced by both predation risk and conspecific risk. Variation in female vigilance was more consistent with predictions of the antipredatory hypothesis, whereas variation in male vigilance was more consistent with predictions of the social monitoring hypothesis (Table 4). Females increase vigilance with decreasing herd size and increasing predator encounter risk, whereas males, especially breeders with a harem of females, increase both vigilance and aggression during the fall mating season.
Reproductive status and season
Both female and male reproductive adults had higher vigilance than nonreproductive adults and yearlings, potentially for very different reasons. Reproductive females had higher levels of vigilance in the spring (
28% time scanning) than in the fall (16% time scanning). This high level of vigilance in the spring was observed even when a female's calf was safely resting out of sight. By fall, these 6-month-old calves (fall yearlings) were spending approximately 15% of time scanning, whereas their mothers reduced their time spent scanning by one-half their spring levels. These observations suggest that antipredatory vigilance in reproductive females is important primarily to increase offspring survival. What is surprising is the nonsignificant difference between the vigilance levels in low and high wolf encounter risk regions. One reason why this might be the case is that elk calves are preyed upon by coyotes and bears as well as wolves and these predators are found throughout Yellowstone in both the low and high wolf encounter risk regions.
Both reproductive and nonreproductive males maintained extremely high levels of vigilance only during the fall season (45% time scanning—reproductive males, 22% time scanning—nonreproductive males). The lower level of vigilance in nonreproductive males suggests that a proportion of reproductive male vigilance may serve to monitor the position and behavior of harem females. If a harem female strays away from the herd, the dominant male quickly heads her off and attempts to move her back toward the other females. If a subordinate male approaches, the dominant male moves quickly to intercept the intruder before he reaches the herd. Reproductive and nonreproductive males showed low levels of vigilance (10% time scanning) in the spring season suggesting that social monitoring vigilance in adult males is important primarily to monitor conspecifics and, potentially, increase their mating success during the rut. In the fall, adult males also showed a significant increase in aggression from 1% to 7% of their time active.
This corresponding increase in both vigilant and aggressive behaviors in the fall corroborates a previous study documenting the seasonal change in herd size and social behaviors of male elk (Weckerly 2001). In a polygynous system of harem holders, male reproductive success is limited by access to females and severe competition occurs between males for this access (Trivers 1972). During the rut, male elk engage in aggressive contests to establish dominance, reproductive success is correlated with competitive ability, and dominant males enjoy large increases in reproductive success (Clutton-Brock et al. 1979; Gibson and Guiness 1980; Wolff and Van Horn 2003). Reproductive success in female elk is generally limited by access to food resources or other factors that reduce the probability of calf survival (Iason et al. 1986). During our study years, reproductive male elk in Yellowstone National Park are at low risk of predation during spring, summer, and fall, whereas reproductive female elk must protect newborn calves from predation by wolves, grizzly bears, and coyotes (Houston 1978; Gunther and Renkin 1990; Gese and Grothe 1995; Mech et al. 2001; Smith al. 2003). Although both male and female reproductive elk occasionally show high levels of vigilance, the benefit of their vigilance appears to be fundamentally different.
Herd size, encounter risk, and position in the herd
Nonreproductive female vigilance increased with decreasing group size and in regions with high wolf encounter rates (18% time scanning) over regions with low wolf encounter rates (8% time scanning). Both factors are consistent with the antipredatory benefit of female vigilance. Male vigilance, however, was not influenced by either of these factors. This suggests that predation risk does not influence male vigilance as much as it influences female vigilance. One alternative explanation is that differences between Norris Basin and Lamar Valley, other than predation risk, might account for the lack of difference in male vigilance. Norris Basin is wetter with different species of grass and has smaller meadows surrounded by lodgepole pine forests. We believe, however, that these differences would favor an increase in vigilance for both males and females in Norris over the more open Lamar Valley. Our observations of elk in small meadows along Soda Butte Creek, similar to those in Norris Basin, suggest that the proximity to wolves matters much more than other habitat features that vary between our study sites. Female vigilance increased with edge position in the herd, but male vigilance did not. This is also consistent with our previous results that females are sensitive to changes in predation risk but males are not.
As mentioned earlier, only young elk calves are vulnerable to coyotes and grizzly bears (Gunther and Renkin 1990; Gese and Grothe 1995), and thus only reproductive female vigilance would be expected to be influenced by these predators. Coyotes and grizzly bears are commonly sighted in both study regions and thus probably represent small but equal risks between the study regions.
The antipredatory benefit of vigilance has very broad support in the literature on ungulate behavior (Caro et al. 2004). Underwood (1982) found that vigilance among 5 species of African antelope was correlated with body mass, such that the smallest species spent the greatest proportion of time being vigilant. Four of the 5 species increased vigilance with decreasing group size, but impalas, the smallest of the 5, showed the opposite pattern. Hunter and Skinner (1998) found that both impalas and wildebeest increased vigilance in response to increased predation risk from reintroduced lions. Their vigilance increased with decreasing herd size, edge position in the herd, and the presence of offspring but, unlike our study, was similar for both adult males and females. Although position in herd was a predictor variable for female elk vigilance, the biological meaning of these results is not clear in this system. For one, our estimate of position (middle or edge) was very coarse and to gather any true meaning it clearly needs to be more specifically defined (Stankowich 2003). In addition, position in herd as an index of predation risk may be less important for coursing predators (e.g., wolves) than it is for stalking predators (e.g., lions, cheetahs). Evidence and personal observations suggest that wolves select prey (either during the hunt or prior) based on age and/or health (Huggard 1992a, 1992b; Mech et al. 2001). It may be that these individuals are often positioned on the edge, but this is not known. Our results, and those of corroborating studies, suggest that female vigilance in Yellowstone National Park may indeed have an antipredatory benefit (Laundre et al. 2001; Childress and Lung 2003; Wolff and Van Horn 2003).
One plausible reason that elk might show such dramatic differences between the sexes is the high degree of sexual dimorphism in body mass. If elk males attain a size refuge from wolf predation that females are unable to attain, this could lead to differential selection regimes for each sex. This seems rather unlikely given that wolves also prey on adult moose and bison that are much larger than male elk. An alternative explanation is that the lifetime reproductive success of males favors vigilance only during the mating season and feeding during the nonreproductive season. The importance of male body condition entering the rut may leave males with few options regarding the vigilance/feeding trade-off during the nonreproductive season. Female lifetime reproductive success is less likely to be under such extreme foraging constraints (Clutton-Brock et al. 1982). Fitness in females may be maximized by maximizing offspring survival and future reproductive opportunities.
Antipredatory versus social monitoring vigilance
Social monitoring vigilance might be more important than antipredatory vigilance in primate troops (Treves 2000) and mixed foraging flocks of birds (Beauchamp 2001). Hirsch (2002) found that brown capuchin monkeys in low predation risk populations showed increased vigilance with increased conspecific density, an observation consistent with the social monitoring hypothesis. Robinette and Ha (2001) found that northwestern crows increased vigilance with increased flock size presumably due to increased scrounging opportunities. These examples illustrate a couple of potential benefits of social monitoring vigilance. What is unusual about our results is that different members of the same elk herd appear to have very different benefits as a result of increasing vigilance. The next step would be to explore why animals monitor conspecifics. This potentially has important consequences for herds that vary in composition of males and females, but our previous work has found that female vigilance levels are independent of herd composition (Childress and Lung 2003). Female vigilance is determined by reproductive status, position in the herd, predator encounter risk, and herd size but not by the overall vigilance of the herd.
|
|
ACKNOWLEDGEMENTS |
|---|
We thank our many wonderful field assistants for their long hours of data collection and data entry, C. Holmes, M. Ptacek, J. Wiklund, P. Young, J. Rupp, and D. Buddy. We offer a special thanks to Drs J. Laundré and L. Hernández for sharing their expertise, advice, and field locations. The manuscript has benefited from the constructive comments of M. Ptacek and E. Keeley. We also thank Doug Smith and the Yellowstone wolf project for the wolf data and assistance in this area. Funding for this research was provided by a Graduate Student Research Committee grant from Idaho State University (MAL) and a Faculty Research Committee grant from Idaho State University (MJC). Our thanks to the Yellowstone Center for Resources and to the National Park Service.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Alberts SC. (1994) Vigilance in young baboons: effects of habitat, age, sex, and maternal rank on glance rate. Anim Behav 47:749–755.[CrossRef]
Appleby MC. (1983) Competition in a red deer stag social group: rank, age, and relatedness of opponents. Anim Behav 31:913–918.[CrossRef]
Artiss T and Martin K. (1995) Male vigilance in white-tailed ptarmigan (Lagopus leucurus): mate guarding or predator detection? Anim Behav 49:1249–1258.[CrossRef]
Baldellou M and Henzi SP. (1992) Vigilance, predator detection and the presence of supernumerary males in vervet monkey groups. Anim Behav 43:451–461.[CrossRef]
Beauchamp G. (2001) Should vigilance always decrease with group size? Behav Ecol Sociobiol 51:47–52.[CrossRef]
Bednekoff PA and Ritter R. (1994) Vigilance in Nxai Pan springbok, Antidorcas marsupialis. Behaviour 129:1–11.
Berger J. (1978) Group size, foraging, and antipredator ploys: an analysis of bighorn sheep decisions. Behav Ecol Sociobiol 4:91–99.
Bertram BCR. (1978) Living in groups: predators and prey. In Krebs JR and Davies NB (Eds.). Behavioral ecology: an evolutionary approach(Sinauer Associates Inc., Sunderland (MA)) pp. 261–298.
Beschta RL. (2003) Cottonwoods, elk, and wolves in the Lamar Valley of Yellowstone National Park. Ecol Appl 13:1295–1309.
Caine NG and Marra SL. (1988) Vigilance and social organization in two species of primates. Anim Behav 36:897–904.[CrossRef]
Caraco T. (1979) Time budgeting and group size: a test of a theory. Ecology 60:618–627.[CrossRef]
Caro TM, Graham CM, Stoner CJ, Vargas JK. (2004) Adaptive significance of antipredator behaviour in artiodactyls. Anim Behav 67:205–228.[CrossRef]
Childress MJ and Lung MA. (2003) Predation risk, gender, and the group size effect: does elk vigilance depend upon the vigilance of conspecifics? Anim Behav 66:389–398.[CrossRef]
Clutton-Brock TH. (1986) Great expectations: dominance, breeding success, and offspring sex ratios in red deer. Anim Behav 34:460–471.[CrossRef]
Clutton-Brock TH, Albon S, Gibson R, Guiness F. (1979) The logical stag: adaptive aspects fighting in red deer (Cervus elaphus). Anim Behav 27:211–225.[CrossRef]
Clutton-Brock TH, Guiness FE, Albon SD. (1982) Red deer: behavior and ecology of two sexes(University of Chicago Press, Chicago (IL)).
Creel S and Winnie JR. (2005) Responses of elk herd size to fine-scale spatial and temporal variation in the risk of predation by wolves. Anim Behav 69:1181–1189.[CrossRef]
Creel S, Winnie J, Maxwell B, Hamlin K, Creel M. (2005) Elk alter habitat selection as an antipredator response to wolves. Ecology 86:123387–3397.
Dehn MM. (1990) Vigilance for predators: detection and dilution effects. Behav Ecol Sociobiol 26:337–342.
Eberhardt LL, Garrott RA, Smith DW, White PJ, Peterson RO. (2003) Assessing the impact of wolves on ungulate prey. Ecol Appl 13:776–783.
Elgar MA. (1989) Predator vigilance and group size in mammals and birds: a critical review of the empirical evidence. Biol Rev 64:13–33.[Medline]
Freeman LC, Freeman SC, Romney AK. (1992) The implications of social status for dominance hierarchies in red deer. Anim Behav 44:239–245.[CrossRef]
Frid A. (1997) Vigilance by female Dall's sheep: interactions between predation risk factors. Anim Behav 53:799–808.[CrossRef]
Geist V. (1974) On the relationship of social evolution and ecology in ungulates. Am Zool 14:205–220.
Gese EM and Grothe S. (1995) Analysis of coyote predation on deer and elk during winter in Yellowstone National Park, Wyoming. Am Midl Nat 133:36–43.[CrossRef]
Gibson RM and Guiness FE. (1980) Differential reproductive success in red deer stags. J Anim Ecol 49:193–208.
Gunther KA and Renkin RA. (1990) Grizzly bear predation on elk calves and other fauna of Yellowstone National Park. Int Conf Bear Res Manage 8:329–334.
Hirsch BT. (2002) Social monitoring and vigilance behavior in brown capuchin monkeys (Cebus apella). Behav Ecol Sociobiol 52:458–464.[CrossRef]
Houston DB. (1978) Elk as winter-spring food for carnivores in northern Yellowstone National Park. J Appl Ecol 15:653–661.[CrossRef]
Hudson RJ, Haigh JC, Bubenik AB. (2002) Physical and physiological adaptations. In Toweill DE and Thomas JW (Eds.). North American elk: ecology and management(Smithsonian Institution Press, Washington (DC)) pp. 199–257.
Huggard DJ. (1992a) Prey selectivity of wolves in Banff National Park. I. Prey species. Can J Zool 71:130–139.
Huggard DJ. (1992b) Prey selectivity of wolves in Banff National Park. II. Age, sex, and condition of elk. Can J Zool 71:140–147.
Hunter LTB and Skinner JD. (1998) Vigilance behavior in African ungulates: the role of predation pressure. Behaviour 135:195–211.
Iason GR, Duck CD, Clutton-Brock TH. (1986) Grazing and reproductive success of red deer: the effect of local enrichment by gull colonies. J Anim Ecol 55:507–515.[CrossRef]
Knight SK and Knight RL. (1986) Vigilance patterns of bald eagles (Haliaeetus leucocephalus) feeding in groups. Auk 103:263–272.
Laundré JW, Hernández L, Altendorf KB. (2001) Wolves, elk, bison: reestablishing the "landscape of fear" in Yellowstone National Park, USA. Can J Zool 79:1401–1409.[CrossRef]
Lima SL. (1987) Vigilance while feeding and its relation to the risk of predation. J Theor Biol 124:303–316.[CrossRef]
Lima SL and Dill LM. (1990) Behavioral decisions made under the risk of predation: a review and prospectus. Can J Zool 68:619–634.
Lipetz VE and Bekoff M. (1982) Group size and vigilance in pronghorns. Z Tierpsychol 58:203–216.
Martin P and Bateson P. (1993) Measuring behaviour: an introductory guide 2nd ed (Cambridge University Press, Cambridge).
Mech LD, Smith DW, Murphy KM, McNulty DR. (2001) Winter severity and wolf predation on a formerly wolf-free elk herd. J Wildl Manage 65:998–1003.
Monaghan P and Metcalfe NB. (1985) Group foraging in wild brown hares: effects of resource distribution and social status. Anim Behav 33:993–999.[CrossRef]
Pulliam HR. (1973) On the advantages of flocking. J Theor Biol 38:419–422.[CrossRef][Web of Science][Medline]
Pulliam HR and Caraco T. (1984) Living in groups: is there an optimal group size? In Krebs JR and Davies NB (Eds.). Behavioural ecology: an evolutionary approach(Blackwell Scientific, Cambridge) pp. 122–147.
Quenette PY. (1990) Functions of vigilance behavior in mammals: a review. Acta oecol 11:801–818.
Ripple WJ and Beschta RL. (2004) Wolves and ecology of fear: can predation risk structure ecosystems? Bioscience 54:755–766.[CrossRef]
Ripple WJ, Larsen EJ, Renkin RA, Smith DW. (2001) Trophic cascades among wolves, elk and aspen on Yellowstone National Park's northern range. Biol Conserv 102:227–234.[CrossRef]
Roberts G. (1996) Why individual vigilance declines as group size increases. Anim Behav 51:1077–1086.[CrossRef]
Roberts SC. (1988) Social influences on vigilance in rabbits. Anim Behav 36:905–913.[CrossRef]
Robinette RL and Ha JC. (2001) Social and ecological factors influencing vigilance in northwestern crows, Corvus caurinus. Anim Behav 62:447–452.[CrossRef]
Smith DW. (1997) 1996–97 annual report. Yellowstone wolf project(Yellowstone Institute for Research, Yellowstone National Park (WY)).
Smith DW. (1998) Yellowstone wolf project: annual report 1997(National Park Service, Yellowstone Center for Resources, Yellowstone National Park (WY)).
Smith DW. (2003) Yellowstone wolf project: annual report 2002(National Park Service, Yellowstone Center for Resources, Yellowstone National Park (WY)).
Smith DW, Drummer TD, Murphy KM, Guernsey DS, Evans SB. (2004) Winter prey selection and estimation of wolf kill rates in Yellowstone National Park, 1995–2000. J Wildl Manage 68:153–166.[CrossRef]
Smith DW, Mech LD, Meagher M, Clark WE, Jaffee R, Phillips MK, Mack JA. (2000) Wolf—bison interactions in Yellowstone National Park. J Mamm 81:1128–1135.[CrossRef]
Smith DW, Murphy KM, Guernsey DS. (1999) Yellowstone wolf report: annual report 1998(National Park Service, Yellowstone Center for Resources, Yellowstone National Park (WY)).
Stankowich T. (2003) Marginal predation methodologies and the importance of predator preferences. Anim Behav 66:589–599.[CrossRef]
Tchabovsky AV, Popov SV, Krasnov BR. (2001) Intra- and interspecific variation in vigilance and foraging of two gerbillid rodents, Rhombomys opimus and Psammomys obesus: the effects of social environment. Anim Behav 62:965–972.[CrossRef]
Thouless CR. (1990) Feeding competition between grazing red deer hinds. Anim Behav 40:105–111.[CrossRef]
Treves A. (2000) Theory and method in studies of vigilance and aggregation. Anim Behav 60:711–727.[CrossRef][Web of Science][Medline]
Trivers RL. (1972) Parental investment and sexual selection. In Campbell B (Ed.). Sexual selection and the descent of man(Aldine, Chicago (IL)) pp. 14–32.
Underwood R. (1982) Vigilance behaviour in grazing African antelope. Behaviour 79:81–107.
Waite TA. (1988a) Vigilance in the white-breasted nuthatch: effects of dominance and sociality. Auk 104:429–434.
Waite TA. (1988b) Dominance specific vigilance in the tufted tit-mouse: effects of social context. Condor 89:932–935.
Weckerly FW. (2001) Are large male Roosevelt elk less social because of aggression? J Mamm 82:414–421.[CrossRef]
Wilkinson L. (1996) SYSTAT 10. Vol. 1. Statistics I(Stanford University Press, Stanford (CA)) pp. 663.
Wolff JO and Van Horn T. (2003) Vigilance and foraging patterns of American elk during the rut in habitats with and without predators. Can J Zool 81:266–271.[CrossRef]
Yaber MC and Herrera EA. (1994) Vigilance, group size, and social status in capybaras. Anim Behav 48:1301–1307.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||