Scholarly article on topic 'The impact of feral horses on pronghorn behavior at water sources'

The impact of feral horses on pronghorn behavior at water sources Academic research paper on "History and archaeology"

CC BY-NC-ND
0
0
Share paper
Academic journal
Journal of Arid Environments
OECD Field of science
Keywords
{" Antilocapra americana " / " Equus caballus " / "Interference competition" / "Behavioral change" / "Vigilance behavior" / "Water source"}

Abstract of research paper on History and archaeology, author of scientific article — Amy M.J. Gooch, Steven L. Petersen, Gail H. Collins, Tom S. Smith, Brock R. McMillan, et al.

Abstract Feral horses (Equus callabus) occur throughout the world on all continents except Antarctica. In North America, feral horses occupy 31.6 million acres throughout western North America. Throughout their range, feral horses often share habitat with American pronghorn (Antilocapra americana). Since horses are larger and more aggressive than pronghorn, they are considered socially dominant. In the Great Basin of western North America, pronghorn often access water sources where horses occur since habitat preferences are similar. If pronghorn are excluded where water is used by both species, pronghorn fitness may be impaired, especially during dry or droughty periods. The purpose of this study was to investigate interference competition between pronghorn and feral horses at water sources within the Great Basin. We observed horses and pronghorn at high-use water sources and recorded all occurrences and outcomes of pronghorn/horse interactions. We assessed differences in pronghorn behavior in the presence or absence of horses. Pronghorn invested more time on vigilance behavior and less time foraging or drinking in the presence of horses than in their absence. Nearly half of pronghorn/horse interactions resulted in pronghorn exclusion from water. We conclude that as feral horse numbers increase, competition for water will subsequently increase.

Academic research paper on topic "The impact of feral horses on pronghorn behavior at water sources"

Contents lists available at ScienceDirect

Journal of Arid Environments

journal homepage: www.elsevier.com/locate/jaridenv

The impact of feral horses on pronghorn behavior at water sources

Amy M.J. Gooch a, Steven L. Petersen a' *, Brock R. McMillan a, Dennis L. Eggett c

Gail H. Collins b, Tom S. Smith

CrossMark

a Department of Plant and Wildlife Sciences, 4105 LSB, Brigham Young University, Provo, UT, United States b U.S. Fish and Wildlife Service, Sheldon-Hart Mountain National Wildlife Refuge Complex, Lakeview, OR, United States c Department of Statistics, 223 TMCB, Brigham Young University, Provo, UT, United States

ARTICLE INFO

Article history: Received 11 May 2016 Received in revised form 23 September 2016 Accepted 29 November 2016 Available online 12 December 2016

Keywords:

Antilocapra americana Equus caballus Interference competition Behavioral change Vigilance behavior Water source

ABSTRACT

Feral horses (Equus callabus) occur throughout the world on all continents except Antarctica. In North America, feral horses occupy 31.6 million acres throughout western North America. Throughout their range, feral horses often share habitat with American pronghorn (Antilocapra americana). Since horses are larger and more aggressive than pronghorn, they are considered socially dominant. In the Great Basin of western North America, pronghorn often access water sources where horses occur since habitat preferences are similar. If pronghorn are excluded where water is used by both species, pronghorn fitness may be impaired, especially during dry or droughty periods. The purpose of this study was to investigate interference competition between pronghorn and feral horses at water sources within the Great Basin. We observed horses and pronghorn at high-use water sources and recorded all occurrences and outcomes of pronghorn/horse interactions. We assessed differences in pronghorn behavior in the presence or absence of horses. Pronghorn invested more time on vigilance behavior and less time foraging or drinking in the presence of horses than in their absence. Nearly half of pronghorn/horse interactions resulted in pronghorn exclusion from water. We conclude that as feral horse numbers increase, competition for water will subsequently increase.

© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND

license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

1. Introduction

Feral (free-roaming) horses (Equus caballus) inhabit 18 countries across all continents except Antarctica (Beever et al., 2008). In semiarid rangelands of western North America, feral horses may compete with pronghorn (Antilocapra americana) and other sym-patric native species for limited water, food resources, and habitat (Meeker, 1979; Beever and Brussard, 2004; O'Gara and Yoakum, 2004; Beever and Herrick, 2006; Nevada Department of Wildlife, (2008); U.S. Fish and Wildlife Service, 2008; Davies et al., 2014; Zeigenfuss et al., 2014; Perry et al., 2015; Hall et al., 2016). According to Perry et al. (2015), aggressive interactions occur between horses and elk (Cervus elaphus) as water availability decreases. Several studies have examined forage overlap between horses and pronghorn to assess the potential for exploitative competition between these species (McInnis and Vavra, 1987; Meeker, 1979; Smith et al., 1998). However, little information is available regarding the

* Corresponding author. E-mail address: steven_petersen@byu.edu (S.L. Petersen).

potential for interference competition between feral horses and pronghorn.

Interference competition most often occurs when resources are not only limited but concentrated spatially. This resource concentration can create an aggregation of animals where direct contact between competing species is more common, increasing the likelihood of interference behavior (Valeix et al., 2008; Atwood et al., 2011). In arid and semi-arid ecosystems, water sources are ideal locations to investigate the influence of interference competition on mammal individuals or populations since they represent a scarce and concentrated resource (Valeix et al., 2008; Atwood et al., 2011).

Interference competition has been observed between individual feral horses at water (Stevens, 1988; Perry et al., 2015). In a study of interactions with desert bighorn sheep (Ovis canadensis nelsoni), domestic horses were experimentally placed near water sources which resulted in no direct aggression, however, the mere presence of horses resulted in a 76% decline in bighorn use of water holes at those locations (Ostermann-Kelm et al., 2008). Others have reported observing aggressive interactions in which horses chased pronghorn from water sources (Nevada Department of Wildlife,

http://dx.doi.org/10.1016/j.jaridenv.2016.11.012

0140-1963/© 2016 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/40/).

(2008)). In contrast, while observing water sources within our same study area, Meeker (1979) concluded there was no evidence of negative interactions between horses and pronghorn and suggested that such interactions were positive with pronghorn possibly using horses as a predator avoidance tactic.

Feral horses have been found to be typically dominant in their social interactions with native Great Basin ungulates, due to their large size (typically 1.3-1.6 m tall at the shoulder and weighing 315-450 kg) and often aggressive behavior (Berger, 1985). Even solitary horses have been shown to displace groups of deer (Odo-coileus hemionus), California bighorn sheep (Ovis canadensis), and pronghorn (Berger, 1985). Social dominance is key in interference competition, with subordinate species often displaced by larger, more aggressive competitors (Valeix et al., 2007). The expected response by the subordinate species would be a change in behaviors (Atwood et al., 2011). Interference competition can occur directly in interactions between species through aggression which results in flight response (Ping et al., 2011; Vahl et al., 2005). Interference competition can also occur indirectly through changes in activities when other species are present, such as avoidance or increased vigilance behaviors (Ping et al., 2011; Vahl et al., 2005; Valeix et al., 2008).

We evaluated direct and indirect interference competition between feral horses and pronghorn at water sources with the assumption that pronghorn would be the subordinate species. We assessed the behavioral responses of pronghorn during horse interactions by determining 1) changes in pronghorn foraging and vigilance behaviors in the presence of horses and 2) the outcomes of pronghorn-horse interactions.

2. Methods

2.1. Study area

The Sheldon National Wildlife Refuge (Refuge, SNWR), administered by the US Fish and Wildlife Service (FWS), is located in northwest Nevada. It stretches across 232,694 ha (575,000 acres) of high desert sagebrush-steppe habitat (41.867995, -119.015493). The SNWR ranged in elevation from 1280 to 2223 m with an annual precipitation of 15-33 cm, and 2011-2012 summer temperatures ranging from -0.6 °C to 39.4 °C. For the duration of this study, the SNWR supported large numbers of sympatric pronghorn and feral horses. Minimum population estimates of pronghorn and horses on the SNWR during this study were estimated at 1727 and 1,236, respectively (Collins, 2010). Water resources on the SNWR were scarce and typical of high-desert habitats, and the majority of available surface water was in the form of small, isolated springs and seeps. Prior to this study, nearly 80% of the Refuge's springs and 44% of stream reaches had been classified as receiving heavy to severe use by horses (U.S. Fish and Wildlife Service, 2008). All livestock grazing was eliminated from SNWR in the mid-1990s.

2.2. Behavior observations

We collected data on pronghorn behavioral responses during interactions with horses via field observation. We deployed 22 remote cameras at representative sites in summer 2011 to assess relative horse and pronghorn use; based upon that information we selected three field observation sites (site 138 [West Catnip], site 142 [WRB_142], and site 32 [Little Buckaroo]) in proximity to spring-fed riparian systems (Appendix A, Fig. 1). We chose these sites because they regularly maintained available surface water from spring through fall, and supported both high horse and pronghorn use. We made observations from June to September 2012. We randomly assigned observation order, observing sites in

one-week intervals before relocating to the next site. We observed each site once within a three-week block before observing any site again, repeated four times for a total of 12 weeks of observations.

We made observations in teams of two observers from 0700 to 1900 h, with alternating 6 h morning or evening "half' days and 12 h "full" days to prevent observer fatigue. Observers were concealed within a camouflaged blind 150-300 m away from the riparian habitat. Binoculars and spotting scopes were used for observing animals. We documented animal behaviors within a 500 m radius of the riparian habitat, and any animals outside of this observation area were not counted in our scans. We defined riparian habitat as the area within 5 m of riparian vegetation or water (Ostermann-Kelm et al., 2008). All other areas were considered uplands habitat and outside the scope of this paper.

We divided pronghorn observations within riparian habitats into group visits, or "observation sessions." An observation session began when the first pronghorn of a group entered the riparian habitat, and ended when the last pronghorn left and none other entered for at least 1 min. Solitary pronghorn constituted a group of one. We defined a horse/pronghorn interaction as any time one species came close enough to the other that we could observe the outward signs of one reacting to the others' presence, while acknowledging that unseen responses may have occurred at even greater distances (the overt response distance) (Smith et al., 2012). Based on preliminary observations, we chose 120 m as the overt response distance.

We used a scan sampling method (Altmann, 1974) to record pronghorn behaviors. During each observation day, we scanned the area every 15 min, recording the time of day and all pronghorn and/ or horses present. When pronghorn and/or horses were spotted we recorded their estimated distance from water, location, and group size. Additionally, we recorded each individual pronghorn's sex and estimated age (young [<1 year old] vs adult), distance from nearest horses (in meters), and behavior (defined below). Whenever pronghorn entered a riparian area, we began scan sampling every minute. However, whenever horses were within a distance that interactions could occur with pronghorn, we sampled every 30 s. We recorded all occurrences of aggression or alarm by either horses or pronghorn during interactions (Altmann, 1974). We considered animals to be in the same group if they were within 100 m of each other (Ostermann-Kelm et al., 2008). Since the majority of the pronghorn in our study were not uniquely marked, we made no attempt to identify individual animals, but groups were assigned temporary identification numbers as long as they were visible for recording purposes.

Based on preliminary observations recorded in 2011 we determined that horse and pronghorn interactions within riparian habitats often occurred too quickly for direct field observation, particularly when animals were in large groups. Subsequently, we recorded all observations of pronghorn within the riparian habitats via video recorder for later viewing and analysis, supplemented with a digital voice recorder as a backup in case of video failure. Videos were later analyzed using Noldus Observer XT® software.

2.3. Pronghorn behavior

We created a pronghorn ethogram (Appendix A), with mutually exclusive main primary behaviors (i.e. standing, walking, running, or bedding) and modifiers to further define those primary behaviors (e.g. standing/vigilant or standing/foraging). We then selected and analyzed two specific behavior modifiers: vigilance and foraging. Vigilance was defined as any time the pronghorn had its head above shoulder height while scanning the surroundings, and foraging was defined as any time an individual was consuming vegetation, drinking, or searching with their head below shoulder

Fig. 1. Maps show the location of the Sheldon National Wildlife Refuge in northwest Nevada (top left and bottom). The 2012 behavioral observation sites were located throughout much of the Sheldon National Wildlife Refuge (top right).

height. (Gavin and Komers, 2006; Appendix A). 2.4. Outcomes of interactions

For each pronghorn-horse interaction, we recorded the date and time, which species initiated the interaction (horse or pronghorn), the type of approach (e.g. walking or running in, and walking directly towards or to the side of the animals already in the riparian); individual pronghorn age (young [<1 year old]) vs. adult), sex, and group size; individual horse age (young [<1 year old]) vs.

adult), and group size; and the minimum distance between horses and pronghorn during the interaction measured in meters. We defined each interaction as neutral, negative, or positive. We categorized an interaction as neutral when pronghorn exhibited no outwardly detectable, costly negative reactions to the presence of horses (i.e. no detectable reaction, or minor reactions categorized as: raising their heads, moving a few meters but remaining in the riparian area, or leaving the riparian area but quickly returning). We categorized an interaction as negative when pronghorn left the riparian area as horses entered or immediately after, or when

pronghorn left some time after horses entered but in direct response to horse actions (e.g. aggression, activity, or approach). We also considered an interaction negative when the pronghorn stopped foraging and drinking due to horse activity, and did not resume foraging or drinking while there were horses present (Valeix et al., 2007). We categorized an interaction as positive if pronghorn waited outside the riparian area until horses entered first, and then entered the riparian area along with the horses.

Video and voice recordings were used to provide observations that are otherwise difficult to code on standardized data sheets during observations. Having both forms made it possible to accurately and efficiently collect and analyze data which is often too complex to capture on data sheets. These data were also used to validate the quality of data collected on datasheets.

2.5. Data analysis

We used logistic regression to explore the potential effects of site, pronghorn cohort, the presence of pronghorn young, prong-horn sex (adults), pronghorn group size, horse group size, time of day, distance to horses (from 0 to 120 m), and the presence of horses on pronghorn behavior. We selected 5 categories of prong-horn cohorts for analysis: 1) solitary adults, 2) groups of all adult females, 3) groups of adult females with young, 4) groups of all adult males, and 5) mixed sex/age groups (Ping et al., 2011). Additional variables included the presence of pronghorn young (yes, no), and sex (male, female, unknown). We selected 3 time categories: morning (0700-1059 h), day (1100-1459 h), and evening (1500—1900 h). We included a variable representing the presence of horses (yes, no) within the interaction distance (120 m) as well as the distance between pronghorn and the nearest group of horses (in meters).

We ran a model for each behavior of interest (vigilance and foraging). All variables were counted as fixed effects except site, which was a random effects model. We used backwards selection to eliminate the non-significant variables and identify the best model. We made no attempt to identify individuals, but since individuals were likely sampled multiple times over the course of each observation session, the data were blocked by observation session to correct for this lack of independence. We considered individual pronghorn observation sessions as a statistical unit. We performed these analyses using SAS statistical software. We considered a p-value <0.05 significant.

3. Results

Between June 25 and September 14,2012, we collected a total of 492 observation hours, averaging 164 h per site (n = 3 sites). We recorded a total of 166 pronghorn observation sessions, 20% of which included interactions with horses accessing riparian areas (n = 34), and documented 12,186 instances of pronghorn behavior (x = 4062 instances per site; Table 1). We excluded individual observations in which pronghorn behaviors were influenced by

Table 1

Summary of observations for three sites within the Sheldon National Wildlife Refuge, Nevada. Information included the number of hours of observation, the number of hours analyzed and the total number of pronghorn observations.

West Catnip WRB_142 Little Buckaroo Total

Hours of Observation 162 162 168 492

Hours Analyzed 42 24 36 102

Pronghorn Observations 8940 1566 1680 12,186

observer presence, and we also excluded whole observation sessions which consisted mostly of observer-biased observations. Due to time constraints, we subsampled the data collected from the video and voice recorders from June 25—July 14 (n = 102 total observation hours; x = 34 observation hours per site; Table 2).

3.1. Behavior analysis

Distance to horses and horse presence were determined to be highly correlated variables, therefore only the significant variable (distance to horses) was included. Neither site nor horse group size were found to be significant variables, and thus, were also not included in the model. The best model for vigilance included the variables pronghorn sex (F-value = 51.52, p < 0.0001), with young being vigilant less of the time than adult males (p < 0.0001) or adult females (p < 0.0001); time of day (F-value = 51.09, p < 0.0001), with pronghorn being vigilant more often in the morning than in the day (p < 0.0001) or afternoon (p < 0.0001); distance from horses (F-value = 13.49, p = 0.0002), with the probability of pronghorn vigilance decreasing 0.3% for each 1 m increase in distance from horses; pronghorn cohort (F-value = 6.52, p < 0.0001), with groups of adult females with young being less vigilant than groups of adult females without young (p = 0.0469, groups of adult males (p = 0.0031), mixed sex/age groups (p = 0.0041), or solitary adult pronghorn (p = 0.0016); and horse group size (F-value = 5.51, p = 0.0189), meaning that for each additional horse, there was a 2.3% increase in the probability of pronghorn vigilance.

Distance from horses was the most important variable influencing foraging behaviors. The influence of horse group size was not significant, and it was not included in the model (Table 3). Distance to horses and horse presence/absence were highly correlated variables. Both were initially included in each model, but the less significant variable of the two was removed. This was consistently the "horse presence" variable. The best model for foraging included the variables distance from horses (F-value = 67.28, p < 0.0001), with the probability of foraging increasing 0.6% for each 1 m increase in distance from horses; time of day (F-value = 17.05, p > 0.0001), with pronghorn in the day foraging less often than pronghorn in the morning (p < 0.0001); pronghorn sex (F-value = 6.55, p = 0.0002), with adult females spending more time foraging than either adult males (p = 0.0087) or young (p = 0.0187); and pronghorn cohort (F-value = 2.49, p = 0.0412), with groups of adult females with young foraging less than mixed sex/age groups (p = 0.0493).

3.2. Outcomes of interactions

During recorded interactions (n = 34) between pronghorn and horses, pronghorn individuals within groups responded in a consistent manner 88% of the time. Thus, we summarized prong-horn responses based on group response. In those instances, where all individuals did not respond similarly, we used the response of the majority of pronghorn in the group as the group response. Of the pronghorn/horse interactions we recorded, 41% (n = 14) were negative, 3% were positive (n = 1), and 47% (n = 16) were neutral. Three additional interactions were classified as "unknown," due to other factors (e.g. coyotes) where we could not confidently attribute pronghorn responses to the presence of horses. All negative interactions were classified as costly as they resulted in the pronghorn being prevented from drinking (n = 1) or leaving the riparian area entirely which prevented them from drinking (n = 13).

Horses initiated 47% of interactions (n = 16) and pronghorn initiated 53% (n = 18). Of the horse-initiated interactions, 75% (n = 12) were negative, 19% (n = 3) were neutral, and 6% (n = 1)

Table 2

Significant results from the differences of least squares means from all variables included in the foraging and vigilance behavior models. Adjustment for multiple comparisons: Tukey-Kramer.

Foraging

Variable Type Variable 1 Variable 2 Estimate SE t-value Adj p-value

Cohort Female with Young Mixed -0.1243 0.04548 -2.73 0.0493

Time of Day Afternoon Day 0.1957 0.08395 2.33 0.0517

Time of Day Day Morning -0.231 0.03962 -5.83 <0.0001

Pronghorn Age/Sex Adult Male 0.2921 0.09225 3.17 0.0084

Pronghorn Age/Sex Adult Young 0.2425 0.08715 2.78 0.0277

Pronghorn Age/Sex Female Male 0.1705 0.05402 3.16 0.0087

Pronghorn Age/Sex Female Young 0.121 0.0415 2.91 0.0187

Vigilance

Variable Type Variable 1 Variable 2 Estimate SE t-value Adj p-value

Cohort All Female Female with Young 0.204 0.07416 2.75 0.0469

Cohort All Male Female with Young 0.4591 0.128 3.59 0.0031

Cohort Female with Young Mixed -0.1574 0.04485 -3.51 0.0041

Cohort Female with Young Solitary -0.2588 0.06887 -3.76 0.0016

Time of Day Afternoon Morning -0.4194 0.08417 -4.98 <0.0001

Time of Day Day Morning -0.362 0.03942 -9.18 <0.0001

Pronghorn Age/Sex Adult Female -0.4601 0.09292 -4.95 <0.0001

Pronghorn Age/Sex Adult Male -0.5474 0.09813 -5.58 <0.0001

Pronghorn Age/Sex Female Young 0.4399 0.04158 10.58 <0.0001

Pronghorn Age/Sex Male Young 0.5272 0.05379 9.8 <0.0001

Table 3

Effect of distance to horses and horse group size on pronghorn vigilance and foraging behavior.

Foraging

Estimate Standard error t-value p-value

Distance to horses 0.0063 0.0008 8.2 <0.0001

Vigilance Estimate Standard error t-value p-value

Distance to horses Horse Group Size -0.003 0.0225 0.0007 0.0096 -3.67 2.35 0.0002 0.0189

were positive. An initiation was defined as the species making the approach was the initiator (e.g. if horses were already in the riparian area, and pronghorn approached the horses, then pronghorn were the initiators. Of the pronghorn-initiated interactions, 11% (n = 2) were negative, 72% (n = 13) were neutral, 17% (n = 3) were unknown, and none were positive. We also documented 6 instances of horse aggression during which pronghorn were present. In 50% of the instances (n = 3), horses were directly aggressive towards pronghorn, and resulted in a negative outcome (e.g., horses chased pronghorn from the area). For the remaining instances (n = 3), the horses were aggressive towards other targets such as other horses or coyotes. In two of those instances, the outcome of the aggression, although not directed at the pronghorn, was still negative for the pronghorn present. The third instance was neutral where the individual pronghorn initially fled the area following horse aggression, but then returned.

4. Discussion

Our work supports the findings of others (e.g. Berger, 1985; Meeker, 1979) that direct aggression is rare, however feral horses were found to be affecting pronghorn vigilance and foraging behaviors at water sources through interference competition. This is in contrast to conclusions by Meeker (1979), however, both the feral horse and pronghorn populations occupying the Refuge during the mid-1970s were substantially smaller, averaging less than half (42%, x = 521 horses) and less than a third (29%, x = 502 pronghorn) of recent population estimates, respectively (G. Collins,

unpublished data). It is reasonable to expect that increases in numbers and density of both species would result in increases in competition for scarce resources. As such, we observed increased vigilance and decreased foraging behaviors by pronghorn in the presence of feral horses. Further, nearly half of the pronghorn/horse interactions we observed were negative and resulted in pronghorn being excluded from the water source as a result of horse activity.

Directly measuring the consequences of these interactions on pronghorn water consumption and fitness was outside the scope of this study, therefore we cannot ascribe significance to these negative interactions. However, since about 40% of interactions resulted in pronghorn exclusion from water, these pronghorn/horse interactions are likely associated with some costs of fleeing (the cost of leaving the water source prematurely and the energy expended on departure; Frid and Dill, 2002) for pronghorn. Increased vigilance behaviors in the presence of horses represented opportunity costs, as seen by the decrease in time spent foraging and drinking, and may result in decreased consumption of water and forage (Frid and Dill, 2002; Wakefield and Attum, 2006). These effects could have detrimental impacts on pronghorn fitness and population dynamics, particularly under adverse conditions when surface water availability is limited and monopolized by horses.

Previous observations of horse and pronghorn interactions have suggested that interference competition at water in North America was unlikely (Meeker, 1979; Berger, 1985). However, our study supports findings by others that interference competition may occur between feral horses and native wildlife (Ostermann-Kelm et al., 2008). Hall et al. (2016) found that water sources experienced lower wildlife richness and diversity when associated with horse use. They suggest that horses have a competitive advantage over smaller wildlife species. Subsequently, native wildlife species visited water sources less frequently reducing the ability of wildlife to access limited water on arid environments. Horses have also been shown to displace even larger bodied animals such as elk in areas with water limitations (Perry et al., 2015). These data suggest that aggressive interactions by horses will increase as water becomes more limited. Competition has the potential to reduce the fitness of pronghorn and other native wildlife populations in areas with limited resources which are also impacted by high concentrations of feral horses. These impacts should be carefully considered when making management decisions.

Acknowledgments

We appreciate the Sheldon National Wildlife Refuge for lodging, transportation, and staff support. We appreciate the Sheldon NWR, Brigham Young University, and the Nevada Department of Wildlife (Michael Cox) for providing research funding. We appreciate Dr. Randy Larsen from Brigham Young University for his statistical consultation and assistance. We also appreciate the many technicians and volunteers who helped with data collection and lab analysis. We thank Brian Day, manager of the Sheldon NWR, for assistance with logistics and housing. The findings and conclusions in this article are those of the authors and do not necessarily represent the views of the U.S. Fish and Wildlife Service. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U.S. Government.

Appendix A. Pronghorn Ethogram

We composed a pronghorn ethogram, separating behaviors into main primary behaviors and secondary modifiers. Primary behaviors described the activity state of a pronghorn: standing, walking, running, or bedding. Secondary modifiers further defined the primary behaviors, and were defined as follows:

• Vigilance was defined as "any time the pronghorn had its head above shoulder height while scanning the surroundings" (Childress and Lung, 2003).

• Foraging was defined as: "Any time an individual [was] consuming vegetation or searching with their head below shoulder height." (Lipetz and Bekoff, 1982). Whether the animal was consuming forage or drinking could not always be distinguished, so these two behaviors were combined.

• Linked urination-defecation was defined as a sequence of behaviors used by territorial bucks, with the sequence proceeding as follows: (1) Sniff and paw the ground, (2) Stretch and urinate in the scrape, and (3) Squat and defecate (Kitchen, 1974).

• Marking/thrashing was defined as when a "male rubs its cheek scent gland on vegetation or thrashes vegetation with its horns" (Maher, 1991).

• Travelling was defined as "walking, trotting, or running with head at shoulder level" (Childress and Lung, 2003).s

• Chasing/following was defined as a dominance-oriented social interaction in which the dominant animal drives (at a walk) or chases (at a lope or run) the subordinate animal (Kitchen, 1974).

• Forcing up was defined as "Adominant pronghorn ... forcing up a bedded subordinate by approaching, lowering its head, and sniffing and/or butting the reclining individual" (Kitchen, 1974).

• Sparring was defined as any interaction between pronghorn individuals that involved touching horns, ranging from brief horn contact to actual fights.

• Herding was defined as males, positioned to the rear of females, with lowered heads walking towards the females, and/or blocking the flight path of the females in a broadside threat position (Kitchen, 1974).

• Courting was defined as the any of the progressive behaviors in the courtship sequence, which involves rump sniffing, chest touching, attempted mounting, and mounting (Kitchen, 1974)

• Other social behavior

• Other behavior.

References

Altmann, J., 1974. Observational study of behavior-sampling techniques. Behaviour 49, 227-267.

Atwood, T.C., Fry, T.L., Leland, B.R., 2011. Partitioning of anthropogenic watering sites by desert carnivores. J. Wildl. Manag. 75,1609-1615.

Beever, E.A., Brussard, P.F., 2004. Community- and landscape-level responses of reptiles and small mammals to feral-horse grazing in the Great Basin. J. Arid Environ. 59, 271-297.

Beever, E.A., Herrick, J.E., 2006. Effects of feral horses in Great Basin landscapes on soils and ants: direct and indirect mechanisms. J. Arid Environ. 66, 96-112.

Beever, E.A., Tausch, R.J., Thogmartin, W.E., 2008. Multi-scale responses of vegetation to removal of horse grazing from Great Basin (USA) mountain ranges. Plant Ecol. 196,163-184.

Berger, J., 1985. Interspecific interactions and dominance among wild Great Basin ungulates. J. Mammal. 66, 571 -573.

Childress, M.J., Lung, M.A., 2003. Predation risk, gender and the group size effect: does elk vigilance depend upon the behaviour of conspecifics? Anim. Behav. 66, 389-398.

Collins, G., 2010. Result of the July 2010 Aerial Survey of Sheldon National Wildlife Refuge. U.S. Fish and Wildlife Service, Lakeview, OR, p. 6 unpublished report.

Davies, K.W., Collins, G., Boyd, C.S., 2014. Effects of feralfree-roaming horses on semi-arid rangeland ecosystems: an example from the sagebrush steppe. Ecosphere 5 (10), 127. http://dx.doi.org/10.1890/ES14-0017L1.

Frid, A., Dill, L.M., 2002. Human-caused disturbance stimuli as a form of predation risk. Conserv. Ecol. 6 (11). Available from:http://www.consecol.org/vol6/iss1/ art11/.

Gavin, S.D., Komers, P.E., 2006. Do pronghorn (Antilocapra americana) perceive roads as a predation risk? Can. J. Zoology 84,1775-1780.

Hall, L.K., Larsen, R.T., Westover, M.D., Day, C.C., Knight, R.N., McMillan, B.R., 2016. Influence of exotic horses on the use of water by communities of native wildlife in a semi-arid environment. J. Arid Environ. 127, 100-105. http://dx.doi.org/ 10.1016/j.jaridenv.2015.11.008.

Kitchen, D.W., 1974. Social behavior and ecology of the pronghorn. Wildl. Monogr. 38,1-96.

Lipetz, V.E., Bekoff, M., 1982. Group size and vigilance in pronghorn. Z. für Tierp-sychol. 58, 203-216.

Maher, C.R., 1991. Activity budgets and mating systems of male pronghorn antelope at Sheldon National Wildlife Refuge, Nevada. J. Mammal. 72, 739-744.

McInnis, M.L., Vavra, M., 1987. Dietary relationships among feral horses, cattle, and pronghorn in southeastern Oregon. J. Range Manag. 40, 60-66.

Meeker, J.O., 1979. Interactions between Pronghorn Antelope and Feral Horses in Northwestern Nevada. Thesis. University of Nevada, Reno, Nevada, p. 101.

Nevada Department of Wildlife, 2008-2009. Big Game Status, p. 110. Reno, Nevada.

Ostermann-Kelm, S.D., Atwill, E.A., Rubin, E.S., Jorgensen, M.C., Boyce, W.M., 2008. Interactions between feral horses and desert bighorn sheep at water. J. Mammal. 89, 459-466.

O'Gara, B.W., Yoakum, J.D., 2004. Pronghorn: Ecology and Management. University Press of Colorado, Boulder, CO, p. 903.

Perry, et al., 2015. Dominance of a natural water source by feral horses. Southwest. Nat. 60, 390-393.

Ping, X.G., Li, C.W., Jiang, Z.G., Liu, W.H., Zhu, H.B., 2011. Interference competition and group size effect in sika deer (Cervus nippon) at salt licks. Acta Ethologica 14, 43-49.

Smith, C., Valdez, R., Holechek, J.L., Zwank, P.J., Cardenas, M., 1998. Diets of native and non-native ungulates in southcentral New Mexico. Southwest. Nat. 43, 163-169.

Smith, T.S., Oyster, O., Partridge, S.D., Martin, I.E., Sisson, A., 2012. Assessing american black bear response to human activity at kenai Fjords National Park, Alaska. Ursus 23, 179-191.

Stevens, F., 1988. Contests between bands of feral horses for access to fresh water: the resident wins. Anim. Behav. 36,1051-1053.

U.S. Fish and Wildlife Service, 2008. Horse and Burro Management at Sheldon National Wildlife Refuge: Revised. final environmental assessment. Lakeview, OR, 153 pp.

Vahl, W.K., van der Meer, J., Weissing, F.J., van Dullemen, D., Piersma, T., 2005. The mechanisms of interference competition: two experiments on foraging waders. Behav. Ecol. 16, 845-855.

Valeix, M., Chamaille-Jammes, S., Fritz, H., 2007. Interference competition and temporal niche shifts: elephants and herbivore communities at waterholes. Oecologia 153, 739-748.

Valeix, M., Fritz, H., Matsika, R., Matsvimbo, F., Madzikanda, H., 2008. The role of water abundance, thermoregulation, perceived predation risk and interference competition in water access by African herbivores. Afr. J. Ecol. 46, 402-410.

Wakefield, S., Attum, O., 2006. The effects of human visits on the use of a waterhole by endangered ungulates. J. Arid Environ. 65, 668-672.

Zeigenfuss, L.C., Schoenecker, K.A., Ransom, J.I., Ignizio, D.A., Mask, K., 2014. Influence of nonnative and native ungulate biomass and seasonal precipitation on vegetation production in a great basin ecosystem. West. North Am. Nat. 74 (3), 286-298.