Great Basin Naturalist Great Basin Naturalist

Volume 56 Number 4 Article 5

11-21-1996

Bighorn sheep response to ephemeral habitat fragmentation by Bighorn sheep response to ephemeral habitat fragmentation by

cattle cattle

John A. Bissonette Utah State University

Melanie J. Steinkamp Utah State University

Follow this and additional works at: https://scholarsarchive.byu.edu/gbn

Recommended Citation Recommended Citation Bissonette, John A. and Steinkamp, Melanie J. (1996) "Bighorn sheep response to ephemeral habitat fragmentation by cattle," Great Basin Naturalist: Vol. 56 : No. 4 , Article 5. Available at: https://scholarsarchive.byu.edu/gbn/vol56/iss4/5

This Article is brought to you for free and open access by the Western North American Naturalist Publications at BYU ScholarsArchive. It has been accepted for inclusion in Great Basin Naturalist by an authorized editor of BYU ScholarsArchive. For more information, please contact scholarsarchive@byu.edu, ellen_amatangelo@byu.edu.

Great Basin Naturalist 56(4}, e 1996, pp. 319-325

BIGHORN SHEEP RESPONSE TO EPHEMERALHABITAT FRAGMENTATION BY CATTLE

J. A. Bissonette! and Melanie J. Steinkampl,2

ABSTR.\cr.-We studied seasonal cattle grazing as an agent of ephemeral babitat fragmentation on a newly reintro­duced population of CaJirornia bighorn sheep (Dvis canadensis calijonlUJna) in Big Cottonwood Canyon, Idaho,1988--89. W'lj ljvl:Iluatw the h)'pothesls that bighorn sbet::p avoid cauJc. We documeut<.-u .slll;;~p W:;POl~ to the proximityto cattle by direct observation. The core areas used by bighorn and distances to escape terrain generally decreased ascattle moved closer to sheep. Li};ewise, sheep moved from cattle as cattle approached them. Severity or response weobserved is in marked contr<lSt with that reported for established bighorn populations, suggesting thal newly reinlro­duced bighorn sheep are more highly sensitive to the presence ofcattle.

Key words: bighorn sheep, cattle, disturbance, Idaho, Ovis canadensis.

Prior to the 20th century, California big­horn sheep were abundant in montane regionsof the western United States (Van Dyke et aI.1986). However, since 1840 population num­bers of bighorn sheep and their area of distrib­ution have decreased (Cowan 1940, Buechner1960). Disease, excessive hunting, activitiesassociated with mining, human disturbance,and pressure from livestock for resources andspace reportedly contributed to the extirpationof the subspecies from most of its range (Smith1954, Geist 1971, Graham 1971, Demarchi andMitchell 1973, Demarchi 1975, Trefethan1975, Van Dyke 1978, Smith et aI. 1988).

California bighorn sheep were once abun­dant in parts of southwestern Idabo; the lastobservations were recorded dUring the 1920s(Hanna 1978). The Idaho Department of Fishand Game (IFG) initiated reintroduction pro­grams of returning California bighorn to partsof their historic range in 1963. Thirty-eightsheep from the Cbilcotin River herd in BritishColumbia were transplanted into the drainagesof the East Fork of the Owyhee River between1963 and 1966 and have provided a base forsubsequent reintroductions. In 1967, 12 addi­tional bighorn were reintroduced into thenearby Little Jack's Creek drainage. Both pop­ulations were allowed to expand until 1980(Toweill1985). From 1980 to 1989, > 100 sbeepwere relocated to 5 different regions in south­ern Idabo.

Livestock pressures have been heavy onrangelands in the western United States thathistorically supported populations of bighornsheep (Mackie 1978). Seventy percent of thepublic land area in the 11 westernmost states isgrazed at least seasonally. Within Idaho range­land conditions varied. In 1986 surveys fromthe Owyhee range in Idaho reported 57% ofthe range in poor condition, 35% fair, and only5% in good condition (Bureau of Land Man­agement, Owyhee rangeland program sum­mary, Burley District, 10, HIes, 16 pp., 1986);while in 1982, 30% of the range was in poorcondition, 57% fair, and 18% in good condition(Bureau of Land Management, Twin Falls, landuse decisions summary and rangeland programsummary, Burley District, 10, files, 26 pp.,1982). Peiper (1988) reported that improve­ment in range condition has been slow since1973.

Bighorn sheep are more sensitive to landuses associated with development than mostnative ungulates (Andryk and lrby 1986). Addi­tionally, bighorn sheep are comparatively lessabundant, react adversely to disturbance, andoccupy habitats sensitive to cbange (Van Dykeet aI. 1986). Livestock activities on these sitescan negatively affect sheep through resourceexploitation (i.e., forage, space, cover, water) orbehaviorally (Geist 1971). On shared rangessocial intolerance may impose greater limita­tions on distribution and habitat use of bighorn

Iu.S. N~lion&l ijiQ!<Jgical Service, Ut:lb Cooper-mv", Fish and Wildlife Research Unit. Deputment of Fisher>e5 and Wildlife, C<.>lkge of Nuturul Res"un~',Utah Stale Un"'ers.I)'. Logan, ur 84321-5200.

Z~.5e(l1 address: U.S. Fish ilIId Wildlife Service, Bv.o: 2676. Vero Beach. FL 32961

319

320 GREAT BASIN NATURALIST [Volume 56

than competition for forage; however, biofogistsdisagree whether livestock impact bighornsheep spatial boundaries, limiting distribution.Wilson (1975) and Van Dyke et al. (1986) re­ported that bighorn show aversions to cattleand avoid them when unaccustomed to theirpresence on the range (Drewek 1970, Kornet1978), while others did not detcct reactionsbetween sheep and cattle (King 1985, Kingand Workman 1985). Analyses that test theavoidance of livestock by bigborn sheep arelimited.

Habitat fragmentation theory bas applica­tion to seasonal livestock grazing. Hahitat frag­mentation may he pennanent {e.g.. subdivisionconstruction} or ephemeral, as in seasonal live·stock grazing. Effects of permanent fragmenta­tion all. habitat use have received increasingattention in recent years; however, less is under·stood about effects of seasonal fragmentation.We postulated that areas used by bighorn sheepare fragmented during spring and summer bycattle on grazing allotments. An area may appearlarge but, due to fragmentation, have a muchsmaller useable area. If bighorn sheep avoidlivestock, the area available to tbem is reducedtemporarily as livestock graze seasonally insheep habitat, resulting in sheep exclusion fromareas of potential use. A population may be in­fluenced as sheep are restricted to smallerpatcbes of habitat and effects of density depen­dence are felt. In our study we wanted todetermine whether avoidance occurs, assess itseffect on habitat use by sheep, and considerhow avoidance, if it occurs, might influencefuture decisions for reintroductions.

STUDY AREA

We conducted the study in Big CottonwoodCanyon 16 km northwest of Oakley (CassiaCo.), Idaho. The canyon is approximately 18 kmlong, with Cottonwood Creek flowing to theoortheast through the canyon bottom. Eleva­tion of the canyon floor increases graduallyfrom 1400 to 2100 m. Average elevation gainfrom the canyon floor to the mesa top is 365 m.Canyon walls are steep and characterized by acombination of cliffs, boulder slopes, grass, andsbrub slopes. Woody vegetation includes four­wing salt brush (Atriplex cGnescens), spiny hop­sage (Grayia spinosa), low sage (Artemisia a,.bu­scu1a), horse brush (Tetradymia caneseens),rabbit brusb (Chrysothamnus nauseosus), blue-

bunch wheatgrass (Ag,.opy,.011 spicatum), andjuniper (juniperus occidentalis).

Big Cottonwood Canyon lies within the Saw­tooth National Forest and contains a cattle graz­ing alIobnent that is leased from late May untilearly October. This grdzing allobnent consistsof 5 pastures managed on a reverse-rotationbasis and supports 400 cows with calves. Mesassouth of the canyon contain another allobnentof 3 pastures; this allobnent is managed on adeferred-rotational system with 100 cow, withcalves. Permit dates for the Big Hollow allot­ment are late May to late October.

METHODS

Thirty-seven California bigborn sheep (19with radio-collars, 18 with pattern-c'Oded col­lars) were released into Big Cottonwood Canyonby the Idaho Departinent of Fisb and Gameduring December 1986, December 1987, andNovember 1988. Collars marked with differentdesigns in permanent ink allowed us to distin­guish between non-transmittered individuals.The population at the beginning of our 1stsummer field season (1988) was 23, 13 fromthe 1st reintroduction in 1986 and 10 from the2nd in 1987. Fourteen additional sheep werereleased in November 1988.

We recorded daily locations ofhighom sheepby visual observation li'om May to September1988 and June to September 1989. Telemetrywas used only to aid in locating radio-collaredbil¢1orn sheep. We conducted weekly visual sur­veys to locate any uncollared sheep not closeto collared individuals. Sheep were viewedfrom > 500 m using a spotling scope to reducechance of detection and disturbance. 11 wewere detected and sheep movement followed,we disregarded subsequent observations ofthose individuals for the remainder of tbe day.Every effort was made to identifY individualswithin groops. We determined individuals bycollar design or by telemetry frequency. Loca­tions were recorded in Universal TransverseMercator (UTM) coordinates. For each locationwe recorded group size and composition.

We defined escape terrain as broken habitaton which mountain sbeep may safely oubna­neuver or outdistance predators (Gionfriddoand Krausman 1985). Specifically, escape ter­rain may be chardCterized by a ruggednessindex as defined by Beasom et al. (1983), andterrain class and number of cliff faces > 120%

1996] SHEEP RESPONSE TO FRAGMENTATION 321

following Krausman and Leopold (1986). Forevery location we measured distance to escapeterrain using a range finder once sheep left thearea. We determined slope with a clinometer.We located cattle by hiking a systematic routeon foot 3-4 times/wk. With the exception ofgroup composition, data recorded for each cat­tle location were identical to sheep locations.We recorded cattle and sheep locations simul­taneously allowing sheep movements to be anal­yzed in response to cattle movement for thatspecific time. Data not taken during identicaltime periods were not used in paired analyses.

Even though a controlled test was not possi­ble, we wanted to observe the response of sheepwhen livestock were in proximity to sheep. On14 August 1989, 5 cows were moved directlyinto an area of continuous sheep use aud heldcontinuously for 40 h. Cattle were kept withinapproximately a 0.8-km2 area by 2 cowboys.Sheep response was observed and recorded.Cattle were watered every 5 h by removingthem from the group one at a time and takingthem to a trough in the bordering pasture.After 40 h all cattle were removed. We locatedsheep daily for the next 10 d.

We combined individual bighorn sheep loca­tions for each group for analysis with ProgramHome Range (Samuel et al. 1985); thus, eachlocation represented a group of bighorn sheep,not an individual. We used 95% harmonic meanmeasures of activity to estimate home rangesand core areas. We defined core areas as themaximum area where the observed utilizationdistribution as determined from the harmonicmean values was greater than a uniform uti­lization distribution (Samuel et al. 1985). Kol­mogoroy's test was used to determine if ob­selved use was significantly (P < 0.05) greaterthan expected. All comparisons were consid­ered significant at the 0.05 level. All data pointswere plotted at a scale of 1: 12,000.

We recognize that harmonic mean measureshave been criticized. Naef-Daenzer (1993) testedthe spatial resolution of the conventional har­monic mean measure and a bivariate Donnalkernel estimator with a new kernel estimatorhe developed. The harmonic mean estimatorgeneralized the distributions of 2 parallel gra­dients and estimated density at higher thanzero for areas containing no sample points.Worton (1989, 1995) and Boulanger and White(1990) have outlined some undesirable proper­ties of harmonic mean measures that were

eliminated from kernel estimators using appro­priate smoothing techniques. Specifically, withthe harmonic measure. estimates of zero areacan occur, and isopleths may include areaswith no sample points (Worton 1995). We hadno estimates of home range or core areas thatapproached or even came close to zero. Addi­tionally, the isopleths we generated were basedon tightly grouped locations of sheep, thusavoiding the problem of areas with no samplepoints. Finally, we did not employ interstudycomparisons, thus avoiding the onerous prob­lem of comparing hetween methods, therebyreducing the effect of inherent bias.

We plotted mean monthly home ranges andcore areas of sheep and cattle and then over­laid them to determine changes in size andlocation between consecutive months. Wemeasured avoidance by quantifying changes insize and location of bighorn sheep range andcore areas as cattle moved through bighornsheep habitat. Changes in location were deter­mined from harmonic means. We compareddata collected during the 1st and 2nd field sea­sons to determine whether range and coreareas were related to seasonal changes.

We calculated daily distances between big­horn sheep and cattle using UTM location co­ordinates. We defined consecutive locations aslocations taken 1 d apart. Only cattle and big­horn sheep paired locations recorded at thesame time were analyzed. Simple linear regres­sions were used to test for associations between3 variables: distance (m) between cattle and big­horn, distance sheep moved in response, anddistance from location of sheep to escape ter­rain. First, we tested sheep response to prox­imity of cattle; then we tested to determinewhether distance between sheep and escapeterrain was related to proximity of cattle.

RESULTS

Response of Bighorn Sheep to Cattle

Sheep range size did not change signifi­cantly in size or location (P < 0.05) from Juneto July in 1988 or 1989. Cattle were in adjacentpastures but because of topography were usu­ally not visible to sheep or the observers. Dur­ing August 1988, when cattle were moved toan allotment adjacent to areas receiving highsheep use, home range position shifted andrange size decreased (Table 1). In Septembersheep expanded their range, coincident with

322 GREAT BASIN NATURALIST [Volume 56

TABLE 1. Spatial responses ofhighorn sheep in Little Cottonwood Canyon, Idaho, to the proximity of cattle.

Core area Mean distance (m)Range

size Size %use % arcail c_sh e_tC SheepdDate (km') (km')

6/88 13.4 4.3 61.4 32.1 4019 101 16167/88 13.7 4.7 53.9 27.3 4045 86 12468/88' 5.0 1.5 59.0 42.9 2251 55 10466/89 1.1.4 4.7 57.5 40.0 4820 112 16987/89 13.5 1.5 67.0 40.0 5148 63 10088/89" 7.2 1.5 55.6 40.0 3346 56 1276

0.51' 11

"PerClJllt o{' total homo mngc area that coro area encompasseshMcan di~tancc hetween cattle and highornCMCl\n distance of sheep to c~capc terrain

the movement of cattle during late August intoa pasture adjacent to a high use sheep area.Sheep tended to concentrate into smaller coreareas in 1988 and 1989 as proximity to cattledecreased.

No significant change «3%) in core area ofbighorn sheep occurred behveen July andAugust 1989 prior to moving cattle close to big­horn. When cattle were moved purposefully towithin 800 m, bighorn sheep responded byimmediately vacating the area and creating anew distinct core area. Distances moved bybighorn sheep directly after movement of cat­tle into the sheep core area were 355% greaterthan daily sheep movements dming early August(3000 vs. 845 m, respectively). Sheep remainedtogether and stayed within 35 m of escape ter­rain for the following 9 d. This was the longesttime period during the study that sheep re­mained within 35 m of escape terrain. Distancesbetween cattle and bighorn sheep remained>4000 m for the following 5 d.

Response of Bighorn SheepRelative to Escape Terrain

As mean daily distance between cattle andsheep decreased, the mean distance hetweensheep and escape terrain tended to decrease.Core-area size appeared to be directly related(adjusted ,.2 = 0.81) to distance to escape ter­rain (Fig. 1); the closer to escape terrain, tbetighter sheep grouped together. A correlationmatrix, generated from these spatial data, addsfurther corroboration for the association (Table2). The mean daily distance that bighornmoved during the month was positively corre­lated (1'2 = 0.88) with increasing distance ofsheep to escape terrain.

dMean daily dis lance sheep moved during the monthCeattle plac€d in allotments dose to sheepfField experiment data

DISCUSSION

Hicks and Elder (1979) suggested that big­horn sheep were more likely to move greaterdistances when cattle were close, but were lesslikely to relocate when cattle were distant. Ourdata show increased movement by bighornsheep as cattle moved closer. When we movedcattle to within 800 m, bighorn left the area.Sheep response to cattle was much more ex­treme than at any other time or when com­pared to their behavior when confronted byhumans at other times during the field season.We were unable to differentiate between theeffect that cattle had and the potential effect ofthe personnel involved. We do not doubt thatpersonnel moving the cattle had an effect. Fur­thermore, the presence of both cattle and per­sonnel close to sheep may well have augmentedbighorn response nonlinearly. However, at othertimes when we accidentally alerted sheep dur­ing the study (n = 10), bighorn responded byrelocating much shorter distances (between872 and 1190 m). Additionally, their responsewas typically short-lived and they left the prox­imity of escape cover by the next day or sooner.Although both the proximity of cattle and per­sonnel influenced bighorn response, the impor­tant point is that extreme proximity evoked ahighly charged response. Even without ourintentional movement of cattle toward sheep,their increasing affinity for escape cover as cat­tle moved closer suggests strongly that live­stock were perceived as a threat.

Escape terrain is an important componentof good sheep habitat (McQuivey 1978, Leslieand Douglas 1979, Weyhausen 1980, Krausmanand Leopold 1986). We would have predicted

1996] SHEEP RESPONSE TO FRAGMENTATION 323

5

..2 -4-

~~ 3(I)

«~ 2

~0

1

•

00 50 100 150

DISTANCE TO ESCAPE TERRAIN (m)

Fig. L Relationship between size of core area ofbighorn sheep, Cottonwood Canyon, Idaho, and distanceto escape terrain, 1988-89.

that tighter grouping should result as sheepmoved farther from escape cover. However,our data show the direct opposite result, sug·gesting that when sheep move farther fromescape terrain, they do so under less threaten­ing situations. Selective pressures under theseconditions appear not to result in tighter groups.

The response of bighorn sheep to cattle weobserved is in contrast with bighorn sheep innational parks. In some parks sheep approachedhumans closely and were photographed fromcar windows (Van Dyke et al. 1986). Smith (1954)reported sheep eating from his hand, whereasothers reported that sheep unaccustomed topeople or cattle fled at the sight of humans orvehicles > 1600 m (Van Dyke et al. 1985). It

appears that newly reintroduced sheep are moresensitive to disturbance, perhaps resulting fromrecent transplant activities, and react differ­ently than do established, undisturbed popula­tions. Sheep reintroduced into Big CottonwoodCanyon were net.gunned from helicopters,blindfolded, and flown to a base. They thenhad blood drawn, were given inoculations,weighed, measured, placed into the back of acovered pickup with several conspecifics, andthen transported approximately 160 km and keptovernight in the vehicles. All were released thefollOwing morning into an area foreign to them.As a result of exposure to such activities, anydisturbance may more likely be viewed as atlrreat. In the Big Cottonwood Canyon popula­tion, alert-alarm behavior appears to be rein­forced yearly with each new group of reintro·duced animals. Age may also playa part; 55%of individuals released were <2 years of age.Heightened sensitivity and subsequent fre­quent reinforcement of alClt behaviors appearto characterize the population and may he ageneral phenomenon for newly reintroducedpopulations placed into new areas. Sensitivity ofthese populations to disturbance may diminishover time as populations become established.

Avoidance has implications for reintroduc­tions ofhighorn sheep. The total area of paten·tial habitat may not be used by sheep if live­stock are present. Ifcattle allotments remain inuse. it would appear wise to consider tbe possi­bility of ephemeral fragmentation by calllewhen goals for desired bighorn population sizesare developed. Goals should be consistent withtotal useable habitat. Control of disturbance forrecently reintroduced populations of bighornsheep is certainly appropriate.

TABLE 2. Correlation matrix for home range, core area, and mean distance v<'iriables for bighorn sheep in Big Cotton­wood Canyon, Idaho, 1988-89.

Core area Mean distance (m)

Range size Size % use % areaa c-sb e-tc Sh.,.,p'

Range size 1.0 0.694 0.234 -0.601 0.887 0.721 0.440Size 1.0 -0.380 -0.704 0.385 0.916 0.765%use 1.0 0.335 0.410 -0.144 -0.272% area 1.0 -0.220 -0.458 -0.266c-, 1.0 0.520 0.308

e-' 1.0 0.887Sheep 1.0

'Pe~nt of total home ranse area that CQU! area e?loompasse$b)'h:ao distance betwam Cl.ttle Qld bighomCMelU\ distaDce d ,beep 10 e:;cape terraio

324 GREAT BASIN NATURALIST [Volume 56

ACKNOWLEDGMENTS

We thank J. J. Beecham, W L. Bodie, T C.Edwards, D. G. Oman, H. G. Hudak, P R.Krausman, R. B. Smith, D. E. Toweill, and P J.Urness (deceased) for their advice and helpduring the project. We also thank the HaroldCranneys who allowed us to use their land andutilities. We extend a special posthumus thanksto D. Balph for his insight and wisdom. We alsothank the editor and associate editor of GBNand an anonymous reviewer for constructivecomments that helped us improve the manu­script. This study was funded by the Idaho De­partment of Fish and Game through the UtahCooperative Fisheries and Wildlife ResearchUnit (NBS) at Utah State University. We thankthe United States Forest Service for providingaerial photographs and topographical maps.The research proposal was evaluated by theAnimal Care Committee at Utah State for ad­herence to established animal care guidelines.Data were collected following acceptable fieldmethodology established by the AmericanSociety of Mammalogists (1987). The UnitedStates National Biological Service, the UtahDivision of Wildlife Resources, the WildlifeManagement Institute, and Utah State Univer­sity jointly support the Utah Cooperative Fishand Wildlife Research Unit and make ourresearch possible. We thank tl,em.

LITERATURE CITED

AMERICAN SOCIETY OF MAMMALOGISTS. 1987. Acceptable

field methods in mammalogy: preliminary guidelinesapproved by the American Society of Mammalogists.Journal of Mammalogy 65 (4, supplement). 18 pp.

ANDHYK, T. A.• AND L. R. IRlJY. 1986. Evaluation of a moun­tain sheep transplant in north-central Montana. Jour­nal of Environmental Management 24: 337-346.

BEASOM, S. L., E. P. WIGGERS, AND J. R. GIARDINO. 1983. Atechnique for assessing land surface ruggedness.Journal of Wildlife Management 47: 1163--1166.

BOULANGER, J. G., AND G. C. WHITE, 1990, A comparisonof home range e~timators using Monte Carlo simula­tion. Journal of Wildlife Management 54: 310-315,

BUEcHNEH, H. K. 1960, The bighorn sheep in the UnitedStates, its past, present, and future. Wildlife Mono­brraph 4. 174 pp.

COWAN, I. M. 1940. Distribution and variation in thenative sheep of North America. American MidlandNaturalist 24: 505-580.

DEMAHCHI, D. A. 1975. Report and recommendations ofthe workshop on California bighorn sheep. Pages143-163 in J. B. Trefethen, editor, The wild sheep inmodern North America. Winchester Press, New York.

DEMARCHI, D. A" AND H. B. MITCHELL. 1973. The Chil­colin River bighorn population. Canadian Field-Nat­uralist 87: 433-454.

DREWEK, J" JR. 1970. Population characteristics and be­havior of introduced bighorn sheep in Owyhee County,Idaho. Unpublished master's thesis, University ofIdaho, Moscow. 46 pp.

GEIST, V 1971. Mountain sheep: a study in behavior andevolution. University of Chicago Press, Chicago, IL.383 pp.

GIONFRIDDO, J. P., AND P. R. KRAUSMAN. 1985. Summerhabitat use by mountain sheep. Journal of WildlifeManagement 50: 331-336.

GRAHAM, H. 1971. Environmental analysis procedures forbighorn in the San Gabriel Mountains. Desert BighornCouncil Transactions: 15: 38-45.

HANNA, P 1978. Owyhee connection. Idaho Wildlife 1(2):7-9.

HICKS, L. L., AND J. M. ELDER. 1979. Human disturbanceof Sierra Nevada bighorn sheep. Journal of WildlifeManagement 43: 909-915.

KING, M. M. 1985. Behavioral response of desert bighornsheep to human harassment: a comparison of dis y

turbed and undisturbed populations. Unpublisheddoctoral dissertation, Utah State University, Logan.137 pp.

KING, M. M., AND C. w: WORKMAN. 1985. Response ofdesert bighorn sheep to human harassment: manage­ment implications. Transactions of the 51st NorthAmerican Wildlife and Natural Resources Confer­ence 51: 74-85.

KORNET, C. A. 1978. Status and habitat use of Californiabighorn sheep on Hart Mountain, Oregon. Unpub­lished master's thesis, Oregon State University, Cor­vallis. 49 pp.

KRAUSMAN, P. R, AND B. D. LEOPOLD. 1986. Habitat com­ponents for desert bighorn in the Harquallala Moun­tains, Arizona. Journal of Wildlife Management 50:504-508.

LESLIE, D. M., JR., AND C. L. DOUGLAS. 1979. Desert big­horn sheep of the River Mountains, Nevada. WildlifeMonographs 66: 1-56.

MCQUIVEY, R. P. 1978. The bighorn sheep of Nevada.Nevada Department of Fish and Game, BiologicalBulletin 6. 81 pp.

MACKIE, R J. 1978. Impacts oflivestock grazing on wild un­gulates. Transactions of the North American Wildlifeand Natural Resources Conference 43: 462-476.

NAEF-DAENZER, B. 1993. A new transmitter for small ani­mals and enhanced methods of home-range analysis.Journal of Wildlife Management 57: 680-689.

PEIFER, R. D. 1988. Grazing systems and management.Pages 9-24 in B. A. Buchanan, editor, Rangelands.University of New Mexico Press, Albuquerque.

SAMUEL, M. D., D. J. PIERCE, AND E. O. GARTON. 1985.Identifying areas of concentrated use within thehome range. Journal of Animal Ecology 54: 711-719.

SMITH, D. R 1954. The bighorn sheep in Idaho: its status,life history and management. Idaho Department ofFish and Game, Wildlife Bulletin 1. 154 pp.

SMITH. N., P. R KRAUSMAN, AND R E. KIRBY. 1988. Desertbighorn sheep: a guide to selected management prac­tices. U.S. Department of Interior, Biological Report.88.35 pp.

TOWEILL, D. 1985. The California bighorn sheep in Idaho.Pages 45-56 in Proceedings of the California BighornWorkshop, Reno, NV

1996] SHEEP RESPONSE TO FRAGMENTATION 325

TREFETHEN, J. B., EDITOR. 1975. The wild sheep in NorthAmerica. Winchester Press, NY. 302 pp.

VAN DYKE, W. A. 1978. Population characteristics andhabitat utilization of bighorn sheep, Steens Moun­tain, Oregon. Unpublished master's thesis, OregonState University, Corvallis. 87 pp.

VAN DYKE, W. A., A. SANDS, J. YOAKUM, A. POLENZ, AND J.BLAISDELL. 1986. Wildlife habitats in managed range­lands-the Great Basin of southeast Oregon-bighornsheep. 2nd edition. General Technical Report PNW­159.37 pp.

WEYHAUSEN, J. D. 1980. Sierra Nevada bighorn sheep: his­tory and population ecology. Unpublished doctoraldissertation, University of Michigan, Ann Arbor. 240pp.

WILSON, L. O. 1975. Report and recommendations of theDesert and Mexican bighorn sheep workshop. Pages110-131 in lB. Trefethan, editor, The wild sheep inmodem North America. Winchester Press, N.Y.

WORTON, B. J. 1989. Kernel methods for estimating the uti­lization distribution in home-range studies. Ecologym 164-168.

. 1995. Using Monte Carlo simulation to evaluate---'k'emel-based home range estimators. Journal of Wild­

life Management 59: 794-800.

Received 23 October 1995Accepted 3 April 1996