Printer Friendly

Population Dynamics of a Bighorn Sheep (Ovis canadensis) Herd in the Southern Black Hills of South Dakota and Wyoming.

Introduction

Bighorn sheep (Ovis canadensis) once numbered in the millions across the western United States (Buechner, 1960; Berger, 1990); however, due to uncontrolled harvest and diseases introduced from domestic sheep, bighorn numbers began to plummet around the turn of the 20th century (Buechner, 1960; Berger, 1990). Managing agencies have attempted numerous re- introductions across the western United States since the mid-1900s with varying levels of success (Berger, 1990; Singer et al., 2000a; Hedrick, 2014).

In South Dakota native sheep were extirpated by the early 1900s (South Dakota Game, Fish, and Parks, 2007 [SDGFP]; Zimmerman, 2008). South Dakota Department of Game, Fish and Parks (SDGFP) began bighorn sheep re- introductions in the 1960s and experienced the same array of success as other agencies (Singer et al., 2000a; SDGFP, 2007). In the mid-2000s, the Custer State Park and Rapid Creek bighorn sheep herds began to decline due to pneumonia outbreaks (SDGFP, 2007; Smith et al., 2014a). The Elk Mountain herd, located in the southern Black Hills of South Dakota and Wyoming, declined in size between 2009 and 2012; the specific cause was unknown (J. Kanta, SDGFP, pers. comm.).

To better understand the decline in the Elk Mountain bighorn sheep population, our primary objectives were to evaluate survival and cause-specific mortality of all sex and age classes of bighorn sheep as well as estimate population size for the herd. More specifically, our objectives were to: determine annual survival rates of adult bighorn sheep, weekly survival rates and recruitment rates of lambs, determine cause-specific mortality of all classes of bighorn sheep, and estimate population size of the Elk Mountain herd.

METHODS

STUDY AREA

Our study area, Elk Mountain, is located in the southern Black Hills in western South Dakota and eastern Wyoming, U.S.A. (43[degrees]43' N latitude and 104[degrees]02' W latitude; Fig. 1). The study area encompassed approximately 18,600 ha. Elevations ranged from 1132 m to 1728 m above mean sea level, and topography consisted of rock outcrops, rolling hills, steep ridges, and gulches (Froiland, 1990). Herbaceous cover (Bromus spp., Poa annua) dominated the landscape at 54.7% (USDA GeoSpatialDataGateway, 2014), whereas shrub/scrub (Artemisia spp.) covered 26.8%, and evergreen forest (17.7%; Pinus ponderosa) comprised the majority of the remaining landscape. Average annual precipitation was 42 cm; mean temperatures ranged from a low of -11 C in January to a high of 32 C in July [National Oceanic and Atmospheric Administration (NOAA), 2014]. Climate values were based on data collected at the Newcastle, Wyoming, weather station from 1981-2010 (NOAA, 2014). Other ungulates in the study area included mule deer (Odocoileus hemicmus), white-tailed deer (O. virginianus), elk (Cervus elaphus), and pronghorn (Antilocapra americana). Potential predators on bighorn sheep at Elk Mountain included mountain lions (Puma concolor), bobcats (Lynx rufus; Parr et al., 2014), coyotes (Canis latrans), bald eagles (Haliaeetus leucocephalus), and golden eagles (Aquila chrysaetos).

ADULT CAPTURE AND DATA COLLECTION

From March 2012 to February 2014, we captured adult bighorn sheep inhabiting the Elk Mountain Region of South Dakota and Wyoming via drop net (Jessup et al., 1984; Kock et al., 1987) and helicopter net gun (Jacques et al., 2009). We aged females based on tooth eruption and wear (Hemming, 1969; Krausman and Bowyer, 2003) and males via horn annuli (Geist, 1966). We fitted adult rams (>4 y of age) and ewes (>1 y of age) with either very high frequency (VHF; Model M2520B, 154-155 MHz) or global positioning system (GPS; Model G2110B, 154-155 MHz) radio-collars (Advanced Telemetry Systems, Isanti, Minnesota). In addition to radio-collars, all captured ewes and immature rams (<3 y of age) received an ear tag for additional identification. We evaluated ewes for pregnancy status using a Bantam XLS portable ultrasound (E.I. Medical Imaging, Loveland, Colorado), and if pregnant, fitted them with temperature-activated vaginal implant transmitters (VITs; model M3930, Advanced Telemetry Systems, Isanti, Minnesota). We collected blood and various biological samples for disease testing and collected blood for estimating genetic diversity (Parr et al., 2016). Each individual sheep was weighed and released. In 2013 and 2014, we tested blood serum for pregnancy-specific protein B (Drew et al., 2001) to verify pregnancy status.

We relocated all radio-collared sheep a minimum of three to four times per week between March 2012--August 2014 using hand-held directional antennas (model RA-2AK, Telonics, Inc., Mesa, Arizona). From September 2014-May 2015, sheep were relocated once a month. We located mortality events within 12 h of receiving mortality signals; at the site, we recorded observations to determine cause of death. We categorized predation events on adults as either mountain lion or coyote predation based on bite marks, caching, plucking, blood, and consumption of carcass. In nonpredation events we collected samples of blood and sections of the liver, spleen, and lung. We then sent these samples to the Wyoming State Veterinary Lab (WSVL) for further analysis. In a few cases (n = 2), we did not sample sheep carcasses due to the high level of decay, and these were listed as "unknown" mortality events.

LAMB CAPTURE

Beginning mid-April each year, we located ewes daily to determine if VITs had been expelled via parturition. When expelled a team of one to two technicians tracked the ewe and VIT using hand-held directional antennas and attempted to locate the lamb (Smith et al., 20146). We captured lambs by hand, collared them with expandable VHF radio-collars (model M4210, Advanced Telemetry Systems), determined sex, weighed, and released them (Table 1). We aged neonates based on umbilicus condition, mobility, presence of afterbirth, and wet fur. We attempted to keep handling times <5 min. We monitored ewe/lamb pairs daily for 60 d postcapture of the neonate; thereafter, we located the pairs three to four times per week.

Similar to adults, we located all neonate mortalities within 12 h of receiving mortality signals. We recorded observations at the site to determine cause of death. We categorized predation events on lambs as either: mountain lion, bobcat (Parr et al., 2014), coyote, or unknown predator. In nonpredation events we performed an external examination for cause of death. We then removed the entire carcass and sent it to WSVL for further analysis. Based on evidence provided by WSVL, we categorized these neonate mortalities as either: starvation, disease, or other. All capture and handling procedures were approved by the South Dakota State University Animal Care and Use Committee (Approval Number 12-090A) and followed recommendations of the American Society of Mammalogists (Sikes et al., 2011).

We estimated survival using a known fate model in Program MARK (White and Burnham, 1999). We converted adult encounter histories to monthly encounter histories and censored individuals when we had no record of them within a given month (i.e., removed them from the at-risk group for that month; White and Burnham, 1999). We assigned mortalities to the month we located sheep; in cases where we did not have an exact mortality date, we used the average date between the last known live signal and the date of the mortality signal. We developed eight a priori models to investigate effects on annual ewe survival; variables investigated included year, age, VIT status, winter severity, and two temporal models [winter (Nov.-Apr.) vs summer (May-Oct.); parturition (Apr.-May) vs nonparturition (Jun.-Mar.); Table 2]. We developed a third temporal model post hoc to include the autumn period when adult sheep died due to a suspected bluetongue (BT) and/or epizootic hemorrhagic disease (EHD) outbreak. To investigate annual ram survival, we developed two models a priori-, a constant model and a temporal model that included harvest (Sep.-Dee.) versus nonharvest (Jan.-Aug.; Table 2) seasons. We ranked all models using Akaike's Information Criterion corrected for small sample size (AICc; Burnham and Anderson, 2002).

To estimate lamb survival, we converted lamb encounter histories to weekly encounter histories. We right-censored individuals when collars were located detached from the lamb emitting mortality signals without any evidence indicating cause of death (i.e., removed them from the at-risk group for the remainder of that biological year). We assigned deaths to the week when the date was known. We developed 10 models a priori (Table 2); intrinsic variables investigated included age at capture, mass at capture, year, sex, peak (date when 50% of known lambs were born +/- 3 d) vs. nonpeak (bom outside of peak interval; Smith et al, 2014a) and two temporal models. As Smith et al. (2014a) investigated neonate survival in the eastern Black Hills, we used their top model ([S.sub.wk1,2-8,>wks]) to compare survival rates and model fit for the Elk Mountain bighorn population to the Rapid Creek bighorn population. In addition we constructed a second temporal model ([S.sub.wk1,2-8,>wks]) to model the Elk Mountain population in the absence of pneumonia. We measured over dispersion by artificially inflating c for both ewe and lamb models (Bishop et al, 2008; Smith et al., 2014a). Recruitment was measured as the proportion of collared lambs to collared ewes. Lamb survival and recruitment were modeled through 26 w for a direct comparison between survival and recruitment rates.

In the autumns of 2013 and 2014, we drove along a standardized route to estimate lambiewe and ram:ewe ratios and to generate population estimates. We conducted four surveys using two technicians between 15 Oct and 30 Nov each year. Surveys occurred from 0800-1200 h, on days with < 2.5 cm of freshly fallen snow, wind speeds < 32.2 kph, and temperatures < 15.6 C. The standardized route indicated where surveyors were to stop and look (n = 10) for sheep using binoculars or spotting scopes and the length of time (avg = 7.5 min, SE = 2.63) spent at each location. We recorded all sheep observed, separated by gender classifications and horn classifications for rams. We recorded collared or ear-tagged sheep when their marked status could be determined. We excluded collared or ear-tagged sheep when their marked status could not be determined.

After we completed all surveys, we calculated Lincoln-Peterson (Kendall, 1999) and detection rate estimates based on survey results. We combined data across all four surveys to estimate population size using the Mark-Resight:Poisson model in Program MARK (McClintock and White, 2009). In addition we calculated lambda ([lambda]) with the equation [N.sub.t] = [N.sub.0] [e.sup.rt] during the study using these population estimates.

RESULTS

From March 2012 to February 2014, we captured and radio-collared 30 ewes (four at 1 y of age, three at 2 y of age, eight at 3 y of age, and 15 >4 y of age at first capture) and eight mature rams (>4 y of age). We captured an additional three immature rams (<3 y of age); these rams were fitted with ear tags but did not receive radio-collars. In 2012 we captured eight ewes; we believed seven were pregnant based on ultrasonography and deployed six VlTs. In 2013 we captured 22 ewes and deployed 17 VITs based on ultrasonography. In 2014 ultrasonography indicated that all 20 ewes captured were pregnant; we therefore deployed 20 VITs. Visual observations in 2012 and blood tested for pregnancy-specific protein B in 2013 and 2014 revealed actual pregnancy rates were 100% each year for all collared ewes, regardless of estimated age.

We captured a single lamb from a ewe without a VIT in 2012. We were unable to capture any lambs from ewes that received VITs in 2012 due to complete VIT failure (early expulsion, n = 2; battery failure, n = 4). In 2013 we were able to capture all 16 lambs from ewes that received VITs during capture and survived to parturition; one ewe that received a VIT died prior to parturition. In 2014 we successfully captured 15 lambs from ewes that received VITs. Two VITs were expelled prior to the lambing season. An additional two VITs were expelled during the lambing season but prior to parturition. We attempted to obtain daily visuals on these ewes following VIT expulsion but were unsuccessful in capturing the neonates. A single VIT in 2014 was retained through parturition, but we were unable to capture the neonate due to its mobility. We captured 32 lambs (18 males, 14 females) over the 3 y time span. Average birth mass was 5.12 kg (SD = 0.09; 2013 = 5.06 kg, SD = 0.11; 2014 = 5.19 kg, SD = 0.39) and there was no difference in lamb birth mass between the 2 y ([t.sub.26] = -0.76, P = 0.45). Male lambs were larger than females (males = 5.34 kg, SD= 0.13; females = 4.86 kg, SD = 0.07; [t.sub.25] = 3.15, P = 0.004). Capture age ranged from 0.125-2 d.

We documented nine adult mortality events from 2012 to 2014. Mountain lion predation accounted for three adult mortalities (two ewes, one ram). We categorized two ewe mortalities as unknown based on field observations. We sampled an additional two ewe carcasses in the field, but the samples were too autolyzed upon arrival at WSVL; therefore, we listed their cause of death as unknown. A mature ram was legally harvested during the 2014 season. A single ewe died due to capture related causes and was censored from survival analyses.

We documented 19 lamb mortalities from 2012 to 2014 (Table 1). We located 18 mortalities through lamb collars; a single lamb carcass was found and recovered by observing a collared ewe in 2012. We located seven predation mortalities (36.8%); we attributed predation mortalities to mountain lion (n = 2), coyote (n = 2), unknown predation (n = 2), and bobcat (n = 1). We sent 10 lamb carcasses to WSVL, and one lamb carcass was examined at the South Dakota State University Diagnostics Lab, Brookings, South Dakota. Three lambs died of starvation, accounting for 15.8% of the mortalities. Seven lambs died of "other" causes (chronic peritonitis/contagious ecthyma, reticulorumenitis, septicemia, trauma, perforated ulcers, unknown; Table 1). WSVL found pneumonia-related strains (Bibersteinia spp., Mannehaemia spp.) in seven lambs; however, pneumonia was not the ultimate cause of death in any of these cases. We found three separate lambs alive but abandoned over the 3 y; all three succumbed to death in captivity. When necropsies were performed on these lambs, two had underlying causes that would have killed the lamb regardless of our interactions (i.e., reticulorumenitis, perforated ulcers; Table 1). WSVL was unable to find any underlying cause of death for the third lamb; therefore, we believe human interaction caused this abandonment and removed the lamb from survival analyses. We found one lamb dead at 31 w of age due to unknown causes.

We considered {[S.sub.sep.oct]} as the top model for estimating annual ewe survival ([w.sub.i] = 0.72). Remaining models were >2 [DELTA][AIC.sub.c] units from this model, and the weight of evidence supporting this model was 2.6 times greater than all other models combined (Table 3). Furthermore, when c was artificially inflated to moderate ([??] = 2.0, [QAIC.sub.c] wt = 0.37) and extreme dispersion ([??] = 3.0, [QAIC.sub.c] wt = 0.26), {[S.sub.sep.oct]} continued to exhibit the lowest [QAIC.sub.c]. Monthly survival for September-October was 95.7% (95% CI = 0.89-0.98) and for all other months was 99.6% (95% CI = 0.98-0.99). The overall annual ewe survival rate was 88.1% (95% CI = 0.76-0.95).

We collared eight rams; five rams received VHF collars and three received GPS collars. One GPS ram (4.527) collar failed within 2 mo of being deployed on the animal; therefore, we censored him from analyses. A second GPS collar began to fail 8 mo after deployment; we were able to recapture this individual and manually remove the collar during the subsequent capture. As we did not attach ear tags to mature rams, we could not distinguish this ram from ram 4.52V once its collar began to fail; therefore, we right-censored this second ram from survival analysis (i.e., removed from the at-risk group for the remainder of that biological year) starting when its collar began to fail. Due to the small sample size of rams available for survival analysis (n = 7), our analyses were restricted. The top model for estimating annual ram survival was {[S.sub.constant]} with [w.sub.i] = 0.70. Model {[S.sub.harvest]} was within 2 [DELTA][AIC.sub.c] units of our top model ([DELTA][AIC.sub.c] = 1.67); however, the 95% confidence intervals included zero, indicating the parameter in this model was not significant. The overall annual ram survival was 85.1% (95% CI = 0.56-0.96).

We combined all lambs across years to evaluate the influences of covariates on overall lamb survival on Elk Mountain. We entered lamb encounter histories in a nonstaggered format. When evaluating the influences of covariates on overall lamb survival, {[S.sub.wk1,2-3,>3wks]} was the top model ([w.sub.i], = 0.99). Remaining models were all >2 [DELTA][AIC.sub.c], units from this top model (Table 4). When [??] was artificially inflated to 2.0 (moderate overdispersion, [QAIC.sub.c] wt = 0.98) and 3.0 (extreme overdispersion, [QAIC.sub.c] wt = 0.92): {[S.sub.wk1,2-3,>3wks]} remained the top model. Additionally, 95% confidence intervals of the (3 estimate did not contain 0. Our field season ended in 2014 when the youngest lamb was 26 w of age; therefore, we truncated weekly survival analysis to 26 w for all lambs. The overall probability of bighorn sheep lambs surviving to 26 weeks of age was 44.7% (95% CI = 0.28-0.62). Recruitment rates were similar (2013 = 0.36, 2014 = 0.33, SD = 0.02) between the 2 y.

We estimated the population size for bighorn sheep in 2012 based on the maximum number of sheep positively assigned to an age/sex class observed during routine relocations of radio-marked individuals throughout the course of the year. This generated a minimum population estimate of 80 (SE = 0.58) individual sheep. In 2013 Lincoln-Peterson, detection rates, and Program MARK estimates generated following the survey protocol resulted in population estimates of 96, 99, and 104 individuals, respectively. Due to estimate similarities, we calculated the mean of the three models to generate a population estimate of 100 (SE = 2.42) individuals (McClintock and White, 2009). In 2014 the Lincoln-Peterson, detection rates, and Program MARK estimates were 104, 128, and 111 individuals, respectively. We again calculated the mean of the three models due to their similarities to generate a population estimate of 115 (SE = 6.89) individuals. During the course of this study, X was estimated at 1.20.

DISCUSSION

Pregnancy rates, annual ewe and ram survival, overall lamb survival rates, recruitment rates, population estimates, and lambda estimates documented during the course of this study all indicated the Elk Mountain bighorn sheep herd was steadily increasing in size ([lambda] = 1.20). We found a 100% pregnancy rate across all years in this herd, including ewes aged at 1.5 y old at the time of their capture. In the eastern Black Hills, Smith et al. (2014a) found an average pregnancy rate of 93% (range = 91-97%). In other bighorn sheep populations, pregnancy rates for stable or increasing populations also were high [Thome et al, 1979 (87%); Festa-Bianchet, 1988 (92%); Cassirer and Sinclair, 2007 (94%)]. Others assumed a rate of 100% (Buechner, 1960; Hansen, 1965), but this was prior to analytical methods available for testing pregnancy rates.

An annual ewe survival rate of 88.1% was similar to other extant bighorn populations [Jorgenson et al., 1997 (95%); Singer et al., 20006 (89%); Cassirer and Sinclair, 2007 (91%)]. Similar to other studies, we had ewes die during the lambing season (Ross et al., 1997; Hayes et al., 2000; Cassirer and Sinclair, 2007) due to mountain lion predation. In both instances these ewes separated themselves from the remainder of their group, indicating they likely left the safety of the herd to find a location to give birth (Geist, 1971). As these movements are common in ungulates preparing for parturition, some predation loss due to a lack of group vigilance is expected (Geist, 1971; Ross et al., 1997). While many studies have found mountain lion predation caused bighorn sheep populations to decline dramatically (Wehausen, 1996; Ross et al, 1997; Hayes et al., 2000; Kamler et al., 2002), mountain lion predation in our study was minor and did not have population level effects on the herd.

All other ewe mortalities occurred during the autumn months, and we strongly suspect BT and/or EHD was involved based on the timing of the mortalities (i.e., during the autumn in a drought year) and the locations where ewes were found (i.e., at the bottom of ravines, next to standing water; Parr, 2015). Bluetongue and EHD can cause bighorn sheep mortalities (Robinson et al., 1967; Pasick et al., 2001; Noon et al., 2002), but most agree that BT/EHD mortalities in bighorn sheep are rare and insignificant at the population level (Robinson et al, 1967; Clark et al., 1993; Singer et al, 1997; Pasick et al, 2001; Noon et al, 2002). Prior to the initiation of this study, we hypothesized pneumonia would be the main disease encountered based on results from other bighorn sheep herds in the Black Hills (SDGFP, 2007; Smith et al, 2014a) and the western United States (Besser et al, 2012). Bluetongue and EHD are most commonly associated with white-tailed deer, mule deer, and to a lesser extent, pronghorn (Stair et al, 1968; Thomas and Trainer, 1970; Hoff and Trainer, 1973). Pneumonia deaths most commonly occur in the winter months in adult bighorn sheep and 1-2 mo after birth in lambs (Cassirer and Sinclair, 2007; Besser et al, 2012; Smith et al, 2014a). We did not experience any pneumonia-related mortalities during this study, and disease testing indicated this population had not been exposed to M. ovipneumoniae (Parr, 2015). Therefore, the autumn mortalities we encountered were unexpected.

Our findings indicate bighorn sheep may be more prone to BT/EHD than previously believed. Similar to human settlers bringing domestic livestock and diseases to the west during the turn of the 20th century (Buechner, 1960), the westward expansion of white-tailed deer also may introduce diseases to previously naive wildlife populations (VerCauteren and Hygnstrom, 2011). The expansion of cultivated fields, and consequently fragmented landscapes, has been implicated in the cause of the westward expansion of white-tailed deer (Brinkman et al, 2005; Paddock and Yabsley, 2007). Communal water sources may provide the link between white-tailed deer, naive bighorn sheep, and disease transfer.

Due to the small sample size of rams, we were unable to evaluate the influence of covariates, such as age, mass, or winter severity, on their survival. The temporal model we ran evaluating the influence of harvest on ram survival was not significant, likely due to the small sample size. However, our survival rate of 0.85 was similar to annual ram survival reported by Cassirer and Sinclair (2007; 0.84) and others [Jorgensen et al, 1997 (0.84); Singer et al., 2000b (0.91)] with stable or increasing populations. Additionally, observations of the herd indicated a large number of rams were present: ram:ewe ratios in 2013 and 2014 were 100:100 and 88:100, respectively (Parr, 2015).

Few studies have investigated neonate and lamb survival by capturing lambs at one day of age (Reading et al., 2009; Smith et al., 2014b) due to their inaccessibility in steep and rugged terrain (Spencer, 1943; Kennedy, 1948; Smith, 1954; Festa-Bianchet, 1988). As a result most studies and managing agencies have relied on visual observations of lambs with previously radio-marked ewes or lamb:ewe ratios to estimate lamb survival and recruitment (Douglas, 1993; Enk et al., 2001; Cassirer and Sinclair, 2007). These methods may lead to an overestimation of true survival and recruitment rates as they miss lambs that die prior to moving into nursery groups (Smith et al., 2014a, 2015) and also are unable to estimate pregnancy rates in the herd. In our study 63.2% (n = 12) of our lamb mortalities and abandonments occurred < 4 d postpartition, prior to the lambs being moved into nursery groups. Additionally, we found postlambing ewes tended to form groups based on their lamb's survival; i.e., ewes that lost lambs formed "lambless" groups and ewes with lambs formed multiple nursery groups. Ewes without lambs also made larger and more erratic movements during the summer months, whereas ewes with lambs were more likely to be located in known nursery areas (Parr, 2015). Ideally, all sheep in a herd would have an equal chance of being sampled. Large movements away from known nursery groups/locations lead to missing these "lambless" groups and overestimating the true number of lambs and their survival in studies using only visual observations or lamb:ewe ratios. As we collared lambs directly to measure their survival and did not rely on these traditional methods, this bias was avoided.

Our overall 26 wk lamb survival rate was 44.7%; other pneumonia-free populations have estimated rates as high as 76% (Cassirer and Sinclair, 2007). Smith et al. (2014a) estimated a much lower survival rate in the eastern Black Hills; among three subherds of bighorn sheep, only 2% of lambs survived to 1 y of age; when truncated to 26 wk for more direct comparison, 9% (95% CI = 0.03-0.20) of lambs survived. These subherds all experienced pneumonia die-offs in the last decade, and known or suspected pneumonia resulted in 45.7% of lamb mortalities during the course of study (Smith et al., 2014 a). When pneumonia or suspected pneumonia lamb mortalities were excluded from analyses, we compared ultimate causes of death: predation accounted for 52.6% of Smith et al.'s (2014a) mortalities compared to 38.9% in our study, starvation accounted for 21.1% and 16.7%, respectively, and other causes accounted for 26.3% and 44.4%, respectively.

Only one lamb surviving past 3 wk of age later died in our study; the remaining lamb collars either degraded or became entangled on a fence and removed, but we found no signs indicating these lambs died. Results of our study indicate that lamb collars have a major advantage in detecting early and cause-specific mortalities, with little negative effects on the animals, as was also documented by Smith et al. (2014a).

Recruitment across both years averaged 35%. Based on our collared lambs, this value is likely conservative due to censored collars (e.g., we were unable to identify individual lambs at 52 w of age without collars). When collaring lambs our goal was to minimize the time spent handling each neonate; hence, we did not fix each lamb with an ear tag or other "permanent" marking to identify the individual in the event of collar failure/loss. Finally, as the population estimates each year of the study indicated population growth and we found no cases of immigration (Parr, 2015), we believe the lamb survival rate at 26 wk is more indicative of actual recruitment rates.

We documented population growth over the course of this study and high ram:ewe ratios for the bighorn sheep herd at Elk Mountain. Additionally, high adult survival rates and recruitment of lambs into the population have contributed to population growth. Adult predation was minor and did not have population level effects on the herd. As the Elk Mountain herd recovered quickly from the 2010-2012 population decline, we believe this population can be used as a source herd for other translocations; however, we recommend periodic monitoring for disease testing to ensure pneumonia is not transferred to this population. We also recommend monitoring the population during the autumn for BT/ EHD outbreaks. We recommend running five to 10 surveys each autumn to estimate annual population size. Finally, in future studies capturing neonates, we recommend experimenting with more permanent markings on lambs, such as ear tags, to aid in individual identification in instances of collar failure or loss.

Acknowledgments.--Financial support for this project was prodded by Federal Aid to Wildlife Restoration (Study No. 7545) administered through South Dakota DeparUnent of Game, Fish and Parks, the Wyoming Community Foundation, Wyoming Governor's Big Game License Coalition, Bowhunters of Wyoming, Midwest Chapter of die Wild Sheep Foundauon, the Dacotah Chapter of Safari Club International, and Wyoming Chapter of the Wild Sheep Foundation. We thank South Dakota DeparUnent of Game, Fish and Parks, Wyoming Game and Fish Department, Quicksilver Air, and Native Ranges Helicopter Services for their assistance in capturing bighorn sheep. We diank J. Kanta and J. Sandrini for dieir assistance widi obtaining funding for the project. We diank the landowners around Elk Mountain for allowing access to their property and access to the mountain via their property. We thank B. Simpson for his assistance with data analyses. We thank K. Cudmore, S. Griffin, J. Broecher.J. Sandrini, J. Kanta, L. Wiechmann, K Robling, J. Doyle, 11. Juarez, T. Achterhof, J. Binfet, T. Caltrider, R. Amundson, I. Tator, and A. Lindbloom for their assistance capturing lambs. We also thank South Dakota Department of Game, Fish and Parks and Wyoming Game and Fish Department for their time and assistance running population surveys each autumn.

LITERATURE CITED

Bergf.r, J. 1990. Persistence of different-sized populations: an empirical assessment of rapid extinctions in bighorn sheep. Conseru. Biol., 4:91-98.

Besser, T. E., M. A. Highland, K. Baker, E. F. Cassirer, N.J. Anderson, J. M. Ramsey, K. Mansfield, D. L. Bri ning, P. Wolff, J. B. Smith and J. A. Jf.nks. 2012. Causes of pneumonia epizootics among bighorn sheep, western United States, 2008-2010. Emerg. Infect. Dis., 18:400-414.

Bishop, C. J., G. C. White and P. M. Lukacs. 2008. Evaluating dependence among mule deer siblings in fetal and neonatal survival analyses./ Wildl. Manage., 72:1085-1093.

Brinkman, T. J., G. S. Deperno.J. A. Jf.nks, B. S. Haroldson .and R. G. Osborn. 2005. Movement of female white-tailed deer: effects of climate and intensive row-crop agriculture. J. Wildl. Manage., 69:1099-1111.

Buechner, H. K. 1960. The bighorn sheep in the United States, its past, present, and future. Wildl. Monogr., 4:3-174.

Burnham, K. P. and D. R. Anderson. 2002. Model selection and multimodel inference: a practical information-theoretic approach. 2nd ed. Springer-Verlag, New York, New York, USA, 488 p.

Cassirer, E. F. and A. R. E. Sinclair. 2007. Dynamics of pneumonia in a bighorn sheep metapopulation. J. Wildl. Manage., 71:1080-1088.

Clark, R. K., W. M. Bowl, D. A. Jessup and L. F. Elliott. 1993. Survey of pathogen exposure among population clusters of bighorn sheep (Outs canadensis) in California. J. 7.oo. Wildl. Med., 24:48-53.

Douglas, C. L. 1993. Management model for predicting fall lamb:ewe ratios in desert bighorn sheep, Canyonlands National Park, Utah, p. 64-82. In: P. G. Rowlands, C. van Riper III, and M. K. Sogge, (eds). Proceedings of the first biennial conference on research in Colorado Plateau National Parks, Trans, and Proc. National Park Service Technical Report NPS/NRNAU/NRTP93/10, Washington, D.C., U.S.A.

Drew, M. L., V. C. Bleich, S. G. Torres and R. G. Sasser. 2001. Early pregnancy detection in mountain sheep using a pregnancy-specific protein B assay. Wildl. Soc. Bull., 29:1182-1185.

Enk, T. A., H. D. Picton andJ. S. Williams. 2001. Factors limiting a bighorn sheep population in Montana following a dieoff. Northwest Sri., 75:280-291.

Fest.a-Bianchet, M. 1988. Birthdate and survival in bighorn lambs (Ovis canadensis). J. Tool. London, 214:653-661.

Froiland, S. G. 1990. Natural history of the Black Hills and Badlands. The Center for Western Studies, South Dakota, U.S.A. 225 p.

Geist, V. 1966. Validity of horn segment counts in aging bighorn sheep. J Wildl. Manage., 30:634-635. --. 1971. Mountain sheep: a study in behavior and evolution. University of Chicago Press, Chicago, Illinois, U.S.A. 399 p.

Hansfn, C. G. 1965. Growth and development of desert bighorn sheep. J Wildl. Manage., 29:387-391.

Hayes, C. L., E. S. Rubin, M. C. Jorgensen, R. A. Botta and W. M. Boyce. 2000. Mountain lion predation of bighorn sheep in the peninsular ranges, California. J. Wildl. Manage., 64:954-959.

Hf.drick, P. W. 2014. Conservation genetics and the persistence and translocation of small populations: bighorn sheep populations as examples. Anim. Conseru., 17:106-114.

Hemming, J. E. 1969. Cemental deposition, tooth succession, and horn development as criteria of age in Dali sheep. J. Wildl. Manage., 33:552-558.

Hoff, G. L. and D. O. Trainer. 1973. Experimental infection in North American elk with epizootic hemorrhagic disease virus. J. Wildl. Dis., 9:129-132.

Jacques, C. N., J. A. Jf.nks, C. S. Deperno.J. D. Sievers, T. W. Grovenburg, T. J. Brinkman, C. C. Swanson and B. A. Stillings. 2009. Evaluating ungulate mortality associated with helicopter net-gun captures in the northern Great Plains. J Wildl. Manage., 73:1282-1291.

Jessup, D. A., W, E. Clark and R. C. Mohr. 1984. Capture of bighorn sheep: management recommendations. Wildlife Management Branch Administrative Report, California Department of Fish and Game, 84-1, Sacramento, California, U.S.A. 29 p.

Jorgenson, J. T., M. Festa-Bianchet, J. M. Gaillard and W. D. Wishart. 1997. Effects of age, sex, disease, and density on survival of bighorn sheep. Ecology, 78:1019-1032.

Kamler, J. F., R. M. Lee, J. C. deVos Jr., W. B. Ballard, and H. A. Whitlaw. 2002. Survival and cougar predation of translocated bighorn sheep in Arizona. J. Wildl. Manage., 66:1267-1272.

Kendall, W. L. 1999. Robustness of closed capture-recapture methods to violations of the closure assumption. Ecology, 80:2517-2525.

Kennedy, C. A. 1948. Golden eagle kills bighorn lamb. J Mammal., 29:68-69.

Kock, M. D., D. A. Jessup, R. K. Clark, C. E. Franti and R. A. Weaver. 1987. Capture methods in five subspecies of free-ranging bighorn sheep: an evaluation of drop-net, drive-net, chemical immobilization and the net-gun. J. Wildl. Dis., 23:634-640.

Krausman, P. R. and R. T. Bovver. 2003. Mountain sheep (Chiis canadensis And O. dalli), p. 1095-1115. In: G. A. Feldhamer, B. C. Thompson, J. A. Chapman, (eds). Wild mammals of North America: biology, management, and conservation. 2nd ed. Johns Hopkins University press, Baltimore, Maryland, U.S.A.

McClintock, B. T. and G. C. White. 2009. A less field-intensive robust design for estimating demographic parameters with mark-resight data. Ecology, 90:313-320.

National Oceanic and Atmospheric Administration. 2014. Data Tools: 1981-2010 Normals. <ww.ncdc. noaa.gov>. Accessed 13 August 2014.

Noon, T. H., S. L. Weschf., D. Cagle, D. G. Mead, E. J. Bicknell, G. A. Bradley, S. Riplog-Peterson, D. Edsall and C. Reggiardo. 2002. Hemorrhagic disease in bighorn sheep in Arizona. J. Wildl. Dis., 38:172-176.

Paddock, C. D. and M. J. Yybsley. 2007. Ecological havoc, the rise of white-tailed deer, and the emergence of Amblyomma amen'ranum-associated zoonoses in the United States. Carr. Top. Microbiol., 315:289-324.

Parr, B.L. 2015. Population parameters of a bighorn sheep herd inhabiting the Elk Mountain region of South Dakota and Wyoming. Thesis, South Dakota State University, Brookings, U.S.A. 133 p.

--, J. Kanta, J. Sandrini, D.J. Thompson, andJ. A. Jenks. 2014. Bobcat predation on bighorn lamb in the western Black Hills of South Dakota. Prairie Nat., 46:41-43.

--, R. L. Juarez andJ. A. Jenks. 2016. Assessing genetic variation of Rock Mountain bighorn sheep at Elk Mountain. Prairie Nat., 48:40-47.

P.asick, J., K. H andel, E. M. Zhou, A. Clavijo, J. Coates, Y. Robinson and B. Lincoln. 2001. Incursion of epiz.ootic hemorrhagic disease into the Okanagan Valley, British Columbia in 1999. Canadian Vet. J., 42:207-209.

Reading, R. P., D. Kenny, S. Amgalanb.aat.ar, A. DeNicoly and G. Wingard. 2009. Argali lamb (Ovis ammori) morphometric measurements and survivorship in Mongolia. Mammalia, 73:98-104.

Robinson, R. M., T. L. Hailey, C. W. Livingston .and J. W. Thomas. 1967. Bluetongue in the desert bighorn sheep. J. Wildl. Manage., 31:165-168.

Ross, P. 1., M. G. J.alkotzy and M. Festa-Bi.anc.het. 1997. Cougar predation on bighorn sheep in southwestern Alberta during winter. Can. J. Zool., 74:771-775.

Sikes, R. S., W. L. Gannon and the Animal Care and Use Committee of the American Society of Mammalogists. 2011. Guidelines of the American Society of Mammalogists for the use of wild mammals in research. J. Mummed., 92:235-253.

Singer, R. S., D. A. Jessup, 1. A. Gardner and W. M. Boyce. 1997. Pathogen exposure patterns among sympatric populations of bighorn sheep, mule deer and cattle. J. Wildl. Dis., 33:377-382.

Singer, F. J., C. M. Papouchis and K. K. Symonds. 2000n. Translocations as a tool for restoring populations of bighorn sheep. Restor. Ecol., 8:6-13.

--, E. Williams, M. W. Miller, L. C. Zeigenfuss. 2000b. Population growth, fecundity, and survivorship in recovering populations of bighorn sheep. Restor. Ecol., 8:75-84.

Smith, D. R. 1954. The bighorn sheep in Idaho. Idaho Department of Fish and Game Wildlife Bulletin, No. 1. Boise, Idaho, U.S.A. 156 p.

Smith, J. B., T. W. Grovenburg andJ. A.Jenks. 2015. Parturition and bed site selection of bighorn sheep at local and landscape scales .J. Wildl. Manage., 79:393-401.

--, J. A.Jenks, T. W. Grovenburg and R. W. Klaver. 2014a. Disease and predation: sorting out causes of a bighorn sheep (Ovis canadensis) decline. PLoSONE, 9(2): e88271. doi: 10.1371 / journal.pone.0088271. Accessed 20 March 2014.

--, D. P. Walsh, E. J. Goldstein, Z. D. Parsons, R. C. Karsch, J. R. Stiver, J. W. Cain 111, K.J. Raedeke and J, A.Jenks. 20146. Techniques for capturing bighorn sheep lambs. Wildl. Sor. Bull., 38:165174.

South Dakota Game, Fish and Parks. 2007. Rocky Mountain Bighorn Sheep Management Plan. South Dakota, U.S.A. 11 p.

Spencer, C. C. 1943. Notes of the life history of Rocky Mountain bighorn sheep in the Tarryall Mountains of Colorado. J Mammal., 24:1-11.

Stair, E. L., R. M. Robinson and L. P. Jones. 1968. Spontaneous bluetongue in Texas white-tailed deer. Vet. Path., 5:164-173.

Thomas, F. C. and D. O. Trainer. 1970. Bluetongue virus in white-tailed deer. Am.]. Vet. Res., 31:271-278.

Thorne, T., G. Butler, T. Varcalli, K. Becker and S. Hayden-Winc. 1979. The status, mortality and response to management of the bighorn sheep of Whiskey Mountain. Wyoming Game & Fish Department Wildlife Technical Report Number 7, Cheyenne, Wyoming, USA. 213 p.

United States Department or Agriculture GeoSpatialD.at.aGateway. 2014. <http://datagateway.nrcs. usda.gov/GDGOrder.aspx> Accessed 2 September 2014.

VerCauteren, K. C. and S. E. Hygnstrom. 2011. Managing white-tailed deer: Midwest North America, p. 501-535. In: D. G. Hewitt (ed). Biology and management of white-tailed deer. CRC Press, Boca Raton, Florida, U.S.A.

Wehausen, J. D. 1996. Effects of mountain lion predation on bighorn sheep in the Sierra Nevada and Granite Mountains of California. Wildl. Soc. Bull., 24:471-479.

White, G. C. and K. P. Burnham. 1999. Program MARK: survival estimation from populations of marked animals. Bird Study, 46:120-138.

Zimmerman, T. J. 2008. Evaluation of an augmentation of Rocky Mountain bighorn sheep at Badlands National Park, South Dakota. Dissertation, South Dakota State University, Brookings, U.S.A. 157 p.

Submitted 21 April 2017

Accepted 27 September 2017

BRYNN L. PARR (1), JOSHUA B. SMITH and JONATHAN A. JENKS

Department of Natural Resource Management, Smith Dakota State University, Brookings 57007

AND

DANIEL J. THOMPSON

Wyoming Game and Fish Department, Lander 82520

(1) Present address: Alaska Department of Fish and Game, Juneau 99811; e-mail: brynn.parr@alaska.

Caption: Figure 1.--The Elk Mountain bighorn sheep (Ovis Canadensis) study area, located in the southern Black Hills of western South Dakota and eastern Wyoming, U.S.A.
Table 1.--Demographics of neonatal bighorn sheep (Ovis canadensis)
captured on Elk Mountain in the southern Black Hills of South Dakota
and Wyoming, U.S.A., 2012-2014

Lamb                                                Weight
ID        Year    Ewe ID    Date of Birth    Sex     (kg)

1.524     2012     4.023      4/26/2012       M
2012-1    2012     4.103       Unknown3       M
4.940     2013     4.438      4/22/2013       F      4.77
4.741     2013     4.102      4/24/2013       F      4.78
4.001     2013     4.241      4/24/2013       F      4.25
4.880     2013     4.072      4/24/2010       F      4.79
4.780     2013     4.487      4/25/2013       F      4.70
4.720     2013     4.188      4/26/2013       M      5.10
4.800     2013     4.173      4/27/2013       M      5.03
4.840     2013     4.477      4/28/2013       F      4.88
4.700     2013     4.457      4/28/2013       F      4.97
4.562     2013     4.112      4/29/2013       M      5.51
4.040     2013     4.041      4/29/2013       F      4.74
4.800     2013     4.418       5/1/2013       M      5.52
4.001     2013     4.052       5/1/2013       M      5.60
4.92      2013     4.141       5/3/2013       M      5.10
4.021     2013     4.507      5/16/2013       M      5.12
4.700     2013     4.023      5/16/2013       M      6.04
4.200     2014     4.438      4/19/2014       M      3.98
4.780     2014     4.162      4/22/2014       M      5.75
4.028     2014     4.341      4/25/2016       M      6.15
4.079     2014     4.365      4/26/2014       M      5.20
4.860     2014     4.103      4/26/2014       F      5.50
4.580     2014     4.333      4/27/2014       M      5.12
4.179     2014     4.052      4/27/2014       F      5.00
4.640     2014     4.041      4/28/2014       F      5.00
4.129     2014     4.507      4/30/2014       F      5.08
4.840     2014     4.232       5/2/2014       F      4.62
4.129b    2014     4.141       5/3/2014       M      5.49
4.079b    2014     4.322       5/3/2014       F      5.00
4.621     2014     4.457      5/31/2014       M      5.48
4.281     2014     4.023       6/1/2014       M      5.94
4.148     2014     4.241       6/4/2014       M      4.58

Lamb
ID        Cause of Death

1.524     Mountain lion
2012-1    Chronic peritonitis/contagious ecthyma
4.940     Unknown predator
4.741
4.001
4.880
4.780     Coyote predation
4.720
4.800     Reticulorumenitis
4.840     Starvation
4.700     Unknown
4.562
4.040     Bobcat predation
4.800
4.001
4.92      Starvation
4.021     Unknown predator
4.700     Reticulorumenitis
4.200
4.780
4.028     Coyote predation
4.079     Septicemia
4.860     Starvation
4.580
4.179
4.640
4.129     Trauma
4.840     Mountain lion
4.129b    Omphalitis
4.079b    Abandoned
4.621
4.281     Perforated ulcers/aspirated rumen fluid
4.148

(a) Lamb carcass was observed by watching a collared ewe

Table 2.--Models constructed a priori to evaluate influences on annual
ewe, ram, and lamb bighorn sheep ((his Canadensis) survival on Elk
Mountain in the southern Black Hills of South Dakota and Wyoming,
U.S.A., 2012-2014

Model                         K (a)   Description

Ewes
{[S.sub.constant]}              1     Survival was constant
{[S.sub.winsev}                 2     Survival varied by previous
                                        winter's severity
{[S.sub.VIT]}                   2     Survival varied between ewes
                                        receiving VlTs and ewes
                                        without VITs
{[S.sub.apr-may]}               2     Survival varied between lambing
                                        season and non-lambing seasons
{[S.sub.season]}                2     Survival varied between winter
                                        (Nov-Apr) and summer (May-Oct)
{[S.sub.captage]}               2     Survival varied by ewe's age at
                                        capture
{[S.sub.year]}                  9     Survival varied by year
{[S.sub.T]}                    12     Survival varied bv month
{[S.sub.sep-oct]} (b)           2     Survival varied between fall
                                        (Sep-Oct) and the rest of the
                                        year (Nov-Aug)
Rams
{[S.sub.constant]}              1     Survival was constant
{[S.sub.harvest]}               2     Survival varied between harvest
                                        (Sep-Dec) and non-harvest
                                        (Jan-Aug)
Lambs
{[S.sub.constant]}              1     Survival was constant
{[S.sub.captage]}               2     Survival varied by neonate age
                                        at time of capture
{[S.sub.Sex]}                   2     Survival varied by sex
{[S.sub.weight]}                2     Survival varied by weight of
                                        neonate at time of capture
{[S.sub.year]}                  2     Survival varied by year
{[S.sub.peak]} (c,d)            2     Survival varied by timing of
                                        birth
{[S.sub.captage,weight]}        2     Survival varied by neonate
                                        weight and age at time of
                                        capture
{[S.sub.wk1,2-3,>3wks]}         3     Survival varied by age in 3
                                        stages
{[S.sub.wk1,2-8,>8wks]} (c)     3     Survival varied by age in 3
                                        stages
{[S.sub.weight wk1, 2-8,        4     Survival varied by weight and
  >8 wks]}                              age in 3 stages

(a) Number of parameters

(b) Model constructed post hoc based on data collected

(c) Models adapted from Smith et al. (2014) for comparison in the
Black Hills of South Dakota

(d) Peak = date when 50% of lambs were born + 3 d; nonpeak = lamb born
outside peak birthing period

Table 3.--Models of bighorn (Ovis canadensis) ewe annual survival on
Klk Mountain in the southern Black Hills of South Dakota and Wyoming,
U.S.A., 2012-2014 when c (a model term representing overdispersion)
was 1.0 (no dispersion)

Model (a)             [AIC.sub.c]   [AIC.sub.c]    [w.sub.i]
                          (b)           (b)           (b)

{[S.sub.sep-oct]}       63.237            0          0.720
{[S.sub.captage]}       66.619          3.381        0.133
{[S.sub.constant]}      69.229          5.992        0.036
{[S.sub.year]}          69.636          6.399        0.029
{[S.sub.apr-may]}       70.238          7.001        0.022
{[S.sub.season]}        70.377          7.140        0.020
{[S.sub.VIT]}           70.420          7.183        0.020
{[S.sub.winsev]}        70.599          7.362        0.018
{[S.sub.T]}             74.567         11.330        0.002

Model (a)             K (e)   Deviance

{[S.sub.sep-oct]}       2      59.217
{[S.sub.captage]}       9      62.598
{[S.sub.constant]}      1      67.222
{[S.sub.year]}          9      65.616
{[S.sub.apr-may]}       9      66.218
{[S.sub.season]}        9      66.357
{[S.sub.VIT]}           2      66.400
{[S.sub.winsev]}        2      66.579
{[S.sub.T]}            12      50.036

(a) Composition and description of models are listed in Table 2

(b) Akaike's Information Criterion corrected for small sample size
(Burnham and Anderson, 2002)

(c) Difference in AICc relative to min AICc

(d) Akaike wt (Burnham and Anderson, 2002)

(e) Number of parameters

Table 4.--Models of bighorn (Ovis canadensis) neonate survival on Elk
Mountain in the southern Black Hills of South Dakota and Wyoming,
U.S.A., 2012-2014 when [??] (a model term representing overdispersion)
was 1.0 (no dispersion)

Model (a)                        [AIC.sub.c]   [AIC.sub.c]   [w.sub.i]
                                     (b)           (c)          (d)

{[S.sub.wk1,2-3,>3wks]}             95.812          0          1.00
{[S.sub.wk1,2-8,>8wks]}            113.122       17.310         0.0
{[S.sub.weight,wk1,2-8,>8wks]}     113.694       17.882         0.0
{[S.sub.weight]}                   146.839       51.027         0.0
{[S.sub.constant]}                 146.974       51.162         0.0
{[S.sub.captage,weight]}           147.178       51.366         0.0
{[S.sub.captage]}                  148.151       52.338         0.0
{[S.sub.year]}                     148.748       52.936         0.0
{[S.sub.peak]}                     148.783       52.971         0.0
{[S.sub.Sex]}                      148.863       53.051         0.0

Model (a)                        K (e)    Deviance

{[S.sub.wk1,2-3,>3wks]}            3        89.759
{[S.sub.wk1,2-8,>8wks]}            3       107.069
{[S.sub.weight,wk1,2-8,>8wks]}     4       105.605
{[S.sub.weight]}                   2       142.813
{[S.sub.constant]}                 1       144.965
{[S.sub.captage,weight]}           3       141.124
{[S.sub.captage]}                  2       144.124
{[S.sub.year]}                     9       144.721
{[S.sub.peak]}                     2       144.757
{[S.sub.Sex]}                      2       144.837

(a) Composition and description of models are listed in Table 2

(b) Akaike's Information Criterion corrected for small sample size
(Burnham and Anderson, 2002)

(c) Difference in AICc relative to min AICc

(d) Akaike wt (Burnham and Anderson, 2002)

(e) Number of parameters
COPYRIGHT 2018 University of Notre Dame, Department of Biological Sciences
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Parr, Brynn L.; Smith, Joshua B.; Jenks, Jonathan A.; Thompson, Daniel J.
Publication:The American Midland Naturalist
Article Type:Report
Geographic Code:1U4SD
Date:Jan 1, 2018
Words:7899
Previous Article:Use of Fine-textured, Mineral-rich Soils by a Northern Flicker (Colaptes aura t us) in North central British Columbia.
Next Article:Weekly Summer Diet of Gray Wolves (Canis lupus) in Northeastern Minnesota.
Topics:

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters