The association between continual, year-round hunting and bellowing rate of bison bulls during the rut

Management of ungulates

Many ecosystems are experiencing real (cervids in US, Europe, and Japan;) (Côté, 2005), or locally perceived (Asia, Tibetan wild ass (Equus kiang) (Schaller, 1998); South America, guanaco (Lama guanicoe) and vicuña (Vicugna vicugna)) (Baldi et al., 2004; RJ Sarno, pers. obs., 2005), increases in ungulate densities, and many of these systems require extensive management. Human hunting, either prescribed or poaching, is considered a form of predation risk that can divert time/energy from fitness-enhancing activities such as feeding, parental care, or mating (Lima & Dill, 1990; Lima, 1998). In fact, ungulates can show stronger behavioral responses to human hunting than to predation risk by large carnivores (Proffitt et al., 2009; Ciuti et al., 2012). Other potential fitness-decreasing activities associated with hunting include increased movement and frequency of changing groups in elk (Cervus elaphus; Cleveland et al., 2012; Proffitt et al., 2009), increased vigilance and movement in red deer (Cervus elaphus; Jayakody et al., 2008), and more frequent changes in habitat and space use in red deer, roe deer (Cervus capreolus), and European bison (Bison bonasus; Theuerkauf & Rouys, 2008). Flight initiation distance has also been documented to significantly increase in several hunted populations of artiodactyls (Stankowich, 2008).

Despite the ubiquity of harvesting ungulates for food and trophies (Proaktor, Coulson & Milner-Gulland, 2007; Ripple et al., 2015), an understanding of the population-wide effects of hunting on the behavior of survivors remains fragmentary. Events commonly associated with hunting such as firearm discharge, observing downed conspecifics, pursuit by motor vehicles, and/or the presence of humans in close proximity for extended periods of time may produce behavioral changes in survivors. Associations with any of these possibly traumatic events can lead to sensitization to human presence, which in turn, may increase vigilance or stress levels, or suppress functional behaviors such as grooming, feeding, or vocalizations (MacArthur, Geist & Johnston, 1982). To our knowledge, the potential impact of hunting on behaviors explicitly tied to male reproductive ecology (e.g., mate advertisement, threat displays to potential competitors, fighting) has not been investigated.

Ecology and management of bison

This gap in the literature is particularly important because population-level responses to human activities will be crucial to successful restoration and long-term conservation of the iconic plains bison (Bison bison bison). Initial observations indicated that bulls in hunted populations were less vocal than those in a protected population (RJ Sarno, pers. obs.). Therefore, we hypothesized that in a risky environment (i.e., human hunting), the best strategy for reproductively-active bulls may be to dampen vocalizations, because bellowing bulls may be targeted by hunters’ sooner than other animals. Bison form large aggregations during the breeding season (rut) when mature bulls join mixed-sex and age groups. Males exhibit a linear dominance hierarchy, with older, larger bulls dominant to smaller, younger bulls (Komers, Messier & Gates, 1994; Roden, Vervaecke & Van Elsacker, 2005). Dominant males temporarily consort with cows prior to or during estrus and attempt to keep all other bulls away by engaging in vocalizations, threat displays, and fights.

The most conspicuous and frequent vocalizations made by bison bulls during the rut are bellows. Bellows function to intimidate rival males (Berger & Cunningham, 1991; Berger & Cunningham, 1994), while bellow amplitude has been linked to male body condition (Wyman et al., 2008). Importantly, amplitude and bellow quality are directly tied to mating success (Wyman et al., 2008). Vocalizations by males of different species also provide information on female reproductive status (Semple & McComb, 2000), influence female mate choice (McComb, 1991; Reby et al., 2010), accelerate estrus (McComb, 1987), and deter rival males (Clutton-Brock & Albon, 1979; Bowyer & Kitchen, 1987) and predators (Tilson & Norton, 1981). Therefore, any changes in bellowing behavior due to perceived predation risk (Lima & Dill, 1990; Lima, 1998) could have profound consequences for bison reproductive ecology (Berger & Cunningham, 1994; Wyman et al., 2008).

Given the importance of bellows during the mating period, and the use of hunting as a management tool, it is essential that we understand the degree to which hunting, or associated events, influence mating behavior of survivors. Therefore, our primary objective was to use focal-animal observations to compare bellowing frequency of male bison in three populations during the rut (one protected and two hunted) in western South Dakota. In addition, we assessed the influence of possible predictor variables on the frequency of bellowing rate. We hypothesized that if continual, year-round hunting of bison alters bellowing rate, then bulls in hunted populations will bellow less frequently than bulls in protected populations during the rut.

Study Area

Hunting events

Hunts were conducted year-round in Slim Buttes Pasture and Yellow Bear Pasture with .270 or .25–06 caliber centrefire rifles. In each bison pasture there were three hunting events per month (T Ecoffey, United States Department of Agriculture, pers. comm., 2014). The number of animals taken and the duration of hunts varied. During ceremonial hunts, shooters usually approached bison in motor vehicles (2–3 vehicles with 4–5 people total) to within 50–100 m. Non-lactating cows and 2–3 year-old bulls were usually selected during hunts, though older bulls (greater than five years old) were also removed. During these short ceremonial hunts only one or two bison were taken, and most hunts lasted approximately two hours (Graham & Gingerich, 2013).

During non-ceremonial, commercial hunts, up to eight animals were removed in 6 h. Here, a mobile abattoir was utilized, whereby animals were first driven into a holding pasture where two people approached bison to within 25–50 m before discharging their rifles. One animal at a time was downed and carcasses were removed from the pasture and processed at which time shooters re-entered the pasture and repeated the procedure until eight bison had been removed. Remaining animals were then re-released into the pasture (T Ecoffey, United States Department of Agriculture, pers. comm., 2014).

Data collection

We defined bellowing rate as the total number of distinct bellows by focal bulls that were recorded by the observer during 15-min observations. Bellows were quantified using a stopwatch and a manual hand counter. From a stationary vehicle, we conducted 245, 15-min focal observations (Altmann, 1974) of mature bulls between dawn and dusk from 1 July to 13 August 2013 using Vortex binoculars and Leica spotting scopes. We conducted 62 observations among 17 groups in Badlands National Park, 101 observations among 34 groups in Yellow Bear pasture, and 79 observations among 25 groups in Slim Buttes pasture. Mature bulls were defined as bulls greater than four years old (when marked) or those accompanying a cow during the rut when not marked. Group size and composition (i.e., number of attendant males (mature bulls), adult females, juvenile males/females) was defined by the number of animals within 50 m of the focal bull. All mature bulls, whether bellowing or not, that were accompanying a cow were considered to be participating in the rut, and therefore could be included as focal animals. We minimized pseudo-replication by identifying animals based on tag numbers and natural markings. When bulls could not be distinguished by natural markings or ear tags, we visited different groups for each observation and focal individuals were selected haphazardly. We observed bison as close as we could safely approach them throughout the day (50 m), while attempting to avoid oversampling during any particular time of day. Bison did not appear to be bothered by the presence of stationary vehicles in any of the sites as cows and calves routinely passed by parked vehicles within 5 m. Bulls also passed by parked vehicles within 3–5 m while bellowing and fighting other bulls as well as accompanying cows. Therefore, we do not believe that the data were influenced by the presence of observers in motor vehicles. In order to standardize the time period that we observed the rut among pastures, we limited our observations to three weeks after the date of initially observing bellows by bulls in each study site. For example, bellows were first recorded on 6 July in Badlands National Park, 16 July in Yellow Bear pasture, and 17 July in Slim Buttes pasture. Therefore, data were collected in Badlands National Park until 2 August. In Yellow Bear and Slim Buttes pastures data were collected until 9 and 10 August, respectively. Consequently, observation day (described below in the analysis) indicated the progress of the rut (i.e., the number of days that had elapsed after the day that the first bellow was observed).

We characterized adult sex ratios and numbers of bison in each population (Badlands National Park: Population Size = 1,204, Density = 6.2/km², Adult Male:Female Ratio = 1:10; Yellow Bear Pasture: Population Size = 464, Density = 11.9, Adult Male:Female Ratio = 1:22; Slim Buttes Pasture: Population Size = 184, Density = 3.7, Adult Male:Female Ratio = 1:15) utilizing data collected by Badlands National Park and Oglala Sioux Parks and Recreation Authority (OSPRA) personnel during fall roundups. Each of the three study sites were considered independent as all pastures were surrounded by fences.

Ethics Statement

Sensitization can be considered an important indirect effect of hunting; therefore, we observed animals only in areas where bison were subjected to regular human visitation. Data collection did not involve restricted habitat or interference with other species, and was in compliance with institutional (Hofstra University IACUC # 13/14-5) and national guidelines for ethical conduct in the care and use of nonhuman animals in research. We obtained permission to conduct fieldwork in Badlands National Park (Permit #: BADL-2013-SCI-0009) and Pine Ridge Reservation.

Data analysis

Initially, a series of descriptive statistics of the data were generated. Due to high skew (3.7) and kurtosis (22.2) of bellowing rate, a series of attempted transformations, including square root, log, and numerous others, along with the application of the Johnson family of transformations were run. However, all failed to substantially improve the normality of the data. Therefore, we utilized non-parametric bivariate statistics along with Poisson regression, to appropriately model the distribution of bellowing rate. Additionally, scatterplots of bellowing rate and the continuous predictors (number of attendant males, number of adult females, group size, observation day, and hunting) were investigated in order to ensure linearity as well as the absence of outliers. Linearity was indicated in all cases, with a single extreme outlier found with respect to bellowing rate. However, the removal of this outlier failed to substantially change the results of all analyses conducted; therefore, no cases were removed from the analyses.

Spearman’s correlations were initially performed in order to determine the strength, significance, and direction of the relationship between bellowing rate and the predictor variables. An —r— greater than 0.75 was used as the threshold for indicating serious multicollinearity (Mun, 2008). We used a Mann–Whitney U test to compare bellowing rate between the two hunted and one protected population, while a Kruskal–Wallis ANOVA with multiple comparisons was used to assess any differences in bellowing rate among the three populations. Finally, two Poisson regressions were conducted in order to determine the extent to which the predictor variables impacted bellowing rate. These multivariate regressions served to determine the impact of each predictor on the outcome of bellowing rate, while holding all other predictors in the model constant. The initial model only included main effects. The second model included an interaction between the number of attendant males and hunting. All analyses were performed using Stata 13.1.

Implications for conservation

Human hunting can influence reproductive behavior of ungulates by skewing the sex ratio and/or by removing certain individuals from a population (Bonenfant et al., 2004; Mysterud et al., 2003). Such changes can profoundly impact mating dynamics (Mysterud, Coulson & Stenseth, 2002). Consequently, knowledge of how and why individuals may vary in their reproductive behavior under specific regimes of human hunting will be important for effective management and conservation. It is possible that bison in Yellow Bear and Slim Buttes pastures on Pine Ridge have become sensitized to continual, year-round hunting. Therefore, our results should be of particular conservation relevance to bison because bellows function as an intrasexual display to deter rival males during the mating season (Berger & Cunningham, 1994). Bellow amplitude and quality also have been directly associated with mating success (Wyman et al., 2006). Because vocalizations of male ungulates appear to impact mating success in a variety of ways, any changes in behavior could alter critical population-genetic parameters like effective population size (Wright, 1938) and the capacity for populations to exhibit adaptive evolution. Because mate competition is an important evolutionary process, the IUCN Bison Specialist Group emphasizes the importance that bison express the full complement of natural mating behaviors (Gates et al., 2010) and stresses consideration of possible genetic consequences of all management actions, especially those for smaller herds; bellowing behavior is one such component that should be considered. One could argue, however, that bison still express natural mating behavior even if they bellow at diminished rates. It would be risky to assume, however, that decreases in bellowing would be linear among all animals. Differences in age and physical condition likely impact stamina, and therefore, the propensity of males to bellow. Human hunting may exacerbate these differences.

Changes in mating behavior (e.g., propensity to advertise for mates, intimidate, and/or fight rival males) can alter breeding opportunities—or possibly even effectively remove certain individuals from the breeding pool. Population-wide responses to hunting, like fear, are probably transmitted more quickly in social species (Cromsigt et al., 2013) like bison, where many individuals (on occasion) can witness conspecifics being downed by humans at close range. Therefore, we suggest that researchers and managers consider the potential that continual, year-round human hunting has to induce behavioral changes in survivors of exploited populations, especially those behaviors that are linked to mating success via mate advertisement and acquisition. In addition to replicating field observations in hunted vs. non-hunted populations, field experiments could also be conducted. For example, bellowing bulls from hunted and non-hunted populations could be recorded, bellows could be categorized, and playback calls could be presented to bulls and cows in order to measure their reaction. A more experimental approach would involve a BACI design (Green, 1979), whereby one non-hunted population would be separated into two equal populations. Hunting could be introduced into one of the populations with the goal to observe any changes in bellowing behavior (including amplitude, rate, pitch, and timbre) of bulls during the rut. Ultimately, if reduced bellowing as a consequence of human hunting is pervasive, then wildlife managers may need to adjust hunting rate and duration, timing (season), and the time lag between hunting events in order to insure that bison are able to express their full repertoire of natural mating behaviors.