Deer-Predator Relationships: a Review of Recent North American Studies with Emphasis on Mule and Black-Tailed Deer

Deer-Predator Relationships: A Review of Recent North American Studies with Emphasis on Mule and Black-Tailed Deer

Warren B. Ballard; Daryl Lutz; Thomas W. Keegan; Len H. Carpenter; James C. deVos, Jr.

Wildlife Society Bulletin, Vol. 29, No. 1. (Spring, 2001), pp. 99-115.

Stable URL: http://links.jstor.org/sici?sici=0091-7648%28200121%2929%3A1%3C99%3ADRAROR%3E2.0.CO%3B2-I

Wildlife Society Bulletin is currently published by Alliance Communications Group.

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at http://www.jstor.org/about/terms.html. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at http://www.jstor.org/journals/acg.html.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is an independent not-for-profit organization dedicated to and preserving a digital archive of scholarly journals. For more information regarding JSTOR, please contact [email protected].

http://www.jstor.org Wed Apr 18 12:38:59 2007 Predation B DEER-PREDATOR RELATIONSHIPS 99 Deer-predator relationships: a review of recent North American studies with emphasis on mule and black-tailed deer

lirrren B. Ballarc(. Dnr7.l Lutz, Thomas ll: Keega11, Lerr H. Carpenter. ccnd Jarnos C. de Vo.s,Jr.

Abstract In recent years mule (Odocoileus hemionus) and black-tailed (0.h, columbianus) deer appear to have declined in many areas of the western United States and Canada, causing concern for population welfare and continued uses of the deer resource. Causes of the decline have not been identified, but predation by coyotes (Canis latrans), mountain lions (Puma concolor), and wolves (Canis lupus) has been proposed as one of many factors. We reviewed results of published studies conducted since the mid-1 970s concerning predator-deer relationships to determine whether predation could be a factor in the apparent deer population declines and whether there was evidence that predator control could be a viable management tool to restore deer populations. We reviewed 17 published studies concerning mule deer. We found only 4 published studies of the effects of predation on black-tailed deer. A larger database existed for white-tailed deer (0. virginianus), with 19 studies examining effects of predation on white-tailed deer. Study results were confounded by numerous factors. A deer population's relationship to habitat carrying capacity was crucial to the impacts of predation. Deer populations at or near carrying capacity did not respond to predator removal experiments. When deer populations appeared limited by predation and such populations were well below forage carrying capacity, deer mortality was reduced significantly when predator populations were reduced. Only one study, however, demonstrated that deer population increases resulted in greater harvests, although considerable data indicated that wolf control resulted in greater harvests of moose (Alces alces) and caribou (Rangifer tarandus). The most convincing evidence for deer population increases occurred when small enclosures (2-39 km2) were used. Our review suggests that predation by coyotes, mountain lions, or wolves may be a significant mortality factor in some areas under certain conditions. Relation to habitat carrying capacity, weather, human use patterns, number and type of predator species, and habitat alterations all affect predator-prey relationships. Only through intensive radiotelemetry and manipulative studies can predation be rdentified as a major limiting factor. When it is identified, deer managers face crucial decisions.

Address for LVarren B. Ballard: Department of Range, Wilcllife, and Fisheries Management, Texas Tech University; Box 12125, Lubbock, TX 79409. USA; e-mail: [email protected]. Address for Daryl Lutz: Wyoming Game and Fish Department, 3030 Energy Lane, Suite 100, Caper, LVY 82604. USA. Address ior Thomas LV. Keegan: Oregon Department oi Fish and Ll'ildlife, P. 0. Box 59, Port- land, OR 97207, USA. Address for Len H. Carpenter: Wildlife Management Institute, 4015 Cheney Drive, Fort Collins, CO 80526, USA. Address for James C. deVos, Ir.: Research Branch, Arizona Game and Fish Department, 2221 West Greenway Road, Phoenix, AZ 85023, USA. Present atldress ior Keegan: LVashington Department of Fish and L21ildlife. 600 Capitol North, Olympia, LVA 98.501, USA.

Wildlife Society Bulletin 2001, 29(1):99-115 Peer edited 100 Wtrnlife Society Bulletin 2001,29(1):99-1 1 5

Reductions in predator densities have occurred only on relatively small study areas (2-1 80 km*) where predators were identified as a major limiting factor and deer populations were \yell below forage carrying capacity (an important criterion). Thus a problem of scale, methods used to kill predators, benefit:cost ratios, results to hunters, and public acceptance are primary considerations. Methods of predator control available to deer managers have been severely restricted and current methods may not be feasible over large areas when and if predation becon~esa problem. Public acceptance of predator reduction programs is essential for predator-prey management, but may not be achievable given current public attitudes toward predators. We identified several recommendations and research needs based on our review of the literature given current social and political limitations.

Key words black-tailed deer, carrying capacity, coyote, mountain lion, mule deer, population management, predation management, predators, wolf

Wildlife management agencies in the western dation on ungulates, indicated that a selective IJnited States and Canada are concerned about an review of the literature could reinforce almost any apparent decline of mule deer (Ollocoileus view 011 the role of predation. He reviewed 3 1 stud- henzionus) and black-tailed deer (0, h. colum- ies that indicated predation was a limiting or regu- bium~~)populations over large portions of western lating influence and 27 studies indicating that pre- North America (Western Association of Fish and dation was not limiting (Connolly 19-8). However, Wildlife Agencies, mule Deer Committee. 1998 degree of docun~entation varied widely among unpublished report). Western deer populations studies. He concluded that predators acting in con- have been described as very volatile, with major cert with weather, disease, and habitat changes cycles of high and low populations (Denney 1976). could have important effects on prey numbers. Herds apparently began increasing in the 1920s. Since Connollj-'s review, scientists ha\-e continued peaked in the late 1940s to early 1960s, declined to debate whether predation is a significant regu- during the 1960s to mid-l970s, increased during lating factor on ungulate populations (Messier the 1980s. and then declined during the 1990s 1991, Sinclair 1991. Skogland 1991, Boutin 1992, (Denney 1976. Hurley and Unsworth 1998). Solne Van Ballenberghe and Ballartl 1994). Because of investigators indicated that deer populations in increased interest in relationships between preda- some areas have been declining since the 1960s tion and deer populations. we reviewed available (Workman and Low 1976. Schneegas and Bumstead literature concerning deer-predator relationships 1977, Bleich and Taylor 1998). and sought to draw conclusions regarding effects of Numerous factors could be responsible for deer predation on mule and black-tailed deer popula- declines. iilcluding habitat loss or change, severe tions and, based upon our assessment. make appro- weather (e.g., drought, deep snow), starvation, priate recommendations for additional research changes in age and sex structure, disease, predation. and management. competition with livestock and wildlife species such as elk (Cervus eluphus), hunting, and interactions of these factors (Wallmo 1981, Halls 1984, Whittaker Methods and Lindzey 1999). Recently, some members of the We focused our review on studies conducted public (e.g.. Barsness 1998) and some biologists since the mid-1970s and, where applicable to deer. (Gasaway et al. 1992) indicated that predation maj- included some studies summarized by Connolly be largely responsible for declines or lack of ungu- (1 9'8,198 1). We used selected abstracting services late population recovery and that predator control and searched for literature pertaining to deer-pred- may be necessary to restore sollle populations to ator relationships. We searched all major biological greater levels. However, empirical evidence exists and wildlife journals and reliewed literature cita- only for moose (Alces nlces). caribou (Rangfer tions within articles for additional references We tamndus), and one black-tailed deer population, and purposefully excluded predator diet studies this hypothesis has not been tested for mule deer. because these do not allow assessment of effects of Connolly (1978), in his review of effects of pre- predation on prej populations. Deer-predator relationships Ballard et al. 101

Several authors indicated that confusion exists in one cause of mortality or introduction of a new the predator-prey literature because biologists cause may or may not increase total mortality rate have used the terms regulation and limitation inter- depending upon whether there is additivity or changeably and the role of predation in ungulate compensation of mortality causes. For additive population dynamics was unclear (Messier 1991, mortality, additional risk of death does not cause Sinclair 1991, Skogland 1991, Boutin 1992, Dale et reductions in other forms of mortality but rather al. 1994, Van Ballenberghe and Ballard 1994). We increases overall mortality rate. For compensatory use the terms limiting and regulating factors fol- mortality, additional risk of death causes a reduc- lowing definitions by Messier (1991). By definition, tion in other forms of mortality so that overall mor- any mortality factor that reduces rate of population tality either does not change or is less than it would growth is a limiting factor. This defrnition can be if additive. Generally, when an ungulate popula- include density-dependent and density-independ- tion is at habitat carrying capacity (K, Macnab ent factors. However, the point at which an ungu- 1985), mortality is thought to be entirely compen- late population is in approximate equilibrium with satory and becomes increasingly additive the far- its long-term natality and mortality factors is regu- ther below K the population is, until theoretically lation and this equilibrium relies on density- all mortality is additive. It is important to note that dependent factors. In other words, the magnitude on some ranges in very variable environments, any of impact depends upon density, more impact at mortality factor may shift from one end of the addi- greater density or vice versa (inversely density- tive-compensatory scale to the other and back dependent). again (Mackie et al. 1998). Bartmann et a1. (1992) defined compensatory and Numerous studies have documented the follow- additive mortality by explaining that an increase in ing species as potential deer predators: gray wolf (Canis lupus), mountain lion (Puma concolor), coyote (Canis latrans), bobcat (Lym rufus), black bear (Ursus americanus), and grizzly bear (U arc- tos). Wolves, mountain lions, and bobcats are obligate carnivores, meaning they must kill prey species to survive, whereas species such as coyotes, grizzly bears, and black bears are considered facultative carnivores in that, although they can and do kill prey, their diets consist of a diversity of items including mast and vegetation.

" d a?t-% Introduction to deer population dynamics Comprehension of annual population dynamics of a particular deer herd is essential to understanding the potential importance of limiting factors such as predation. Number of animals within a deer herd is a function of births, deaths, and factors that affect them. In theory, numbers of births and deaths and their impact on a deer herd depend on the herd's relationship to K (Macnab 1985). Because field biologists rarely are able to determine exactly what constitutes K, they use indices related to deer condition or browse utilization. A deer population at or above K should produce relatively fewer fawns than a population below K, and mortality from all factors should be relatively great, State agencies have expressed concerns about apparent declines in mule deer populations across the western United the population is If a deer States. Photo courtesy of Wyoming Game and Fish Department. herd greatly exceeds K and over-utilizes forage 102 Wildltfe Society Bulletin 2001,29(1):99- 1 15 resources, individuals should be in poor physical many deer ranges. individuals have improved phys- condition, birth rates should be low, and mortality ical condition during the growing season and then rates should be greater. A deer herd that remains suffer declines in body condition during the 11011- above K ultimately will damage its food resources, growing season. If severe winters or droughts per- the deer population will decline. and when the sist, then mortality from starvation increases and herd recovers (i.e.,to a lesser level than previously can cause a significant population decline. Also. held), habitat carrying capacity will likely be drought results in lower-quality habitat (e.g.. less reduced from previous levels. fawning cover. poor nutrition, reductiotls in alter- Habitat carrying capacity is an important con- nate prey species, lack of water), potentially expos- cept with many implications to evaluating preda- ing fawns and adults to increased mortality from tor-prey relationships. Prey populatioil status in predation. These effects can occur at any popula- relation to K determines how mortality factors act tion density (i.e.,they are density-independent) and on a population. \V%en a deer herd is at K, deaths affect rate of increase of a particular deer herd. equal recruited offspring. l'he mortality cause (e.g., Most mortality in any ungulate population occurs predation) is inconsequential, because once preda- among the youngest age classes. Most of these tion is removed or reduced, other mortality factors mortalities usually occur immediately following will replace it. However, the farther a population is birth or during mid- to late winter, when deer are in below habitat carrying capacity, the more different relatively poor condition. Yearling deer (i.e.. I to 2 mortality factors combine to retard population years of age) probably experience mortality rates increases or even cause declitles. In other words, at intermediate between those of fawns and adults, K. mortality factors are compensatory (i.e., they but some studies indicate that yearling sunrival can replace each other so that total mortality remains be similar to that of adults (Wl~iteet al. 1987). Adult constant). but when populations are well below K, deer usually have low mortality rates. with mortali- each mortality source adds to total mortality and ty rates increasing among older animals. When mortality factors are termed additive. In reality. number of deaths exceeds number of surviving unless populations are well below K. these types of young entering a herd, the herd declines; converse- mortality are operating somewhere between com- ly, when number of surviving young exceeds total plete additivity and complete compensation. mortalities, the herd increases. Identification of additive or compensatory mortali- Predation is relevant to all of the above factors. If ty is often difficult to determine in field situations. losses to predators are high, then the deer herd's Numerous biologists believe compensatory mor- relationship to K can dictate importance of this par- tality is usually density-dependent. Deer birth rates ticular mortality factor. Therefore, if the deer popu- remain relatively high over a wide range of densi- lation is near K, predator removal will do little to ties, and only when densities become excessive are increase population numbers because such mortal- there density-dependent declines in reproduction. ity mill be replaced by other mortality factors and, Therefore. as deer herds approach carrying capaci- if not replaced, will result in deer overpopulation ty, reducing one form of mortality will only result in that could harm habitat and may result in a popula- that mortality factor being replaced by another. For tion crash. In contrast, if predation were suppress- example. at relatively high deer densities where ing a deer population at lou densities (i.e., deer winter severity is the only major mortality factor, population w-ell below K), then predator control doe hunting before a harsh winter would be con- may allow a deer population to increase or increase sidered a compensatory form of mortality. Without at a greater rate until compensatory factors take hunting, losses to hunting wodd be replaced large- over. Biologists continue to debate whether preda- ly by the severe winter weather so that total mor- tion is a major regulatory or limiting factor of ungu- tality for the year would remain the same. On the late populations. other hand, if such hunting occurred before a mild winter, it could be additive mortality. How-ever. deer herds also are affected by other factors. Current theories on effects of Persistent drought appears to reduce habitat car- predation on ungulates rying capacity on summer and winter deer ranges Current theories on the role of predation in the and is reflected in poorer body condition (Kucera population dynamics of ungulates have focused on 1988,Taylor 1996) and lesser neonatal survival. On 4 models: low-density equilibria, multiple stable Deer-predator relationships Ballard et al. 103 states,stable-limitcycles,and recurrent fluctuations bers, prey populations return to low densities. (Boutin 1992,Van Ballenberghe and Ballard 1994, Such systems typically are characterized by multi- Ballard andVan Ballenberghe 1997) These theories ple species of principal predators and principal have been the subject of intense debates focused prey on relationships among moose, wolves, grizzly ,Wultiple stable states. Ungulate populations are bears, and black bears because these species con- regulated by density-dependent predation at low tinue to exist largely in ecosystems not impacted by densities until either natural phenomena or predator humans An understanding of predator-prey rela- control reduce predator populations, at which time tionships under natural conditions is necessary ungulate densities reach K and prey become regu- before we can understand systems that have been lated by food competition (Figure 1). Food compe- affected by humans. Descriptions of the 4 models tition regulates ungulate population growth at the were provided by Ballard and Van Ballenberghe greater equilibrium even when predator popula- (1997), but as yet no model has been identified tions return to former levels. Such systems also are which explains ungulate-predator relationships characterized by multiple species of principal pred- under all conditions, and the models have not been ators and prey. The term predator-pit, which is tested on mule or black-tailed deer. often misused, refers to a narrow band of densities Low-densitjg equilibria. Under this model, ungu- between upper and lower equilibrium points late populations are regulated by density-depend- where ungulates can not increase because of ent predation at low densities for long periods density-dependent predation. (Figure 1). Ungulates remain at low densities until Stable-limit cycles. Ungulate populations fitting either natural phenomena or predator control this model exhibit regular cycles lasting from 30 to allows population growth. Food competition is 40 years (Figure 1).Weather conditions (e.g.,severe not important because ungulate densities never winters or drought) influence viability and survival reach K. When predators recover from low num- of neonates and also influence adult survival. Predation is density-independent during popula- vegetatve carrylng capaclty tion increases and inverse------AJreG~on. ellher densrtpdependent Food comoetltion ly density-dependent dur- or nversely I un~rnportant infrequent moose densty-dependent ing population declines. popuatlon increase > Increase only after .-- Moose populatlon predator control returns to Ungulate density, wea- u u precontrol low- Low-denslty denslty equll~bi~um ther, and forage interact 8 8 equll~briumregulated when predator by denslty.dependent populat~on to regulate populations. $ Mean long-term dens~ty New mean long-term - . $ predation recovers Such systems have been density-eve of equl~brlurnnot characterized as having pred~ctable I 1 one principal predator Years Years and one principal prey Recurrent Fluctuations Low-Density Equilibrium species. Recurrent$uctuations.

vegetatve carryng capacty vegetatve carrying capac~ty Ungulate populations are characterized by fluctuat-

Increase due to ing densities that are not .-- predator reducton increased vunerablty m and nverselydens~ty in equilibrium (Figure 1). uG) Changes in ungulate den- 8 Increase due to sities occur because of m~dwlnters which 2 changes in weather, food 1st equlbrium regulated by den sty^ qualit). and quantit): and dependent predation human harvest, but pre- Years Years dation is the primary fac- MultipleEquilibrium Stable Limit Cycle tor that most often limits ungulate density. Figure 1. Conceptual models of ungulate population regulation by predation: low-density equilibria, multiple stable states, stable-limit cycles, and recurrent iluctuations (from Ballard is and Van Ballenberghe 1997). density-dependent at high

Deer-predator relationships Ballard et al. 105

5 2 .... A' deer populations (Table 1). 3 2 8 b. Unfortunately, inconsistencies zk 3 b. ,,-gg 0. f. C 0 b. 2,'.e .- 0. - LC and significant variations among - - 37- 02' 7 3 C P '.e - -0cc .-0 .-2 FLY F,L studies because of relationships 3 % >; 2: z Zg -9. .z < 5 7- 3 +. c.;c- 2x2 2 $LX ; 2:.Ern to habitat carrying capacit); dif- 2 2 2~ ,_,b. mk =: 5" 0. r- b t 3 LY- 5" 7o css cCmxC fering weather patterns, and the 0 5 - s .E .E ~3% $ .g ? L L~ ? q5$r5=- I=== "- short-termnature of most studies L m 5 7 0 Ezf?+sg v, p22 2 * 5: CT; limit their usefulness in assessing % g 33; 49; 3; 311 2 sY+x a s -C CL overall importance of predation. 0 L -.- 0 c 0 Often small sample sizes limited -z 2 ,X ,X 2 ? ... - -c 2 usefiilness of studies because of 2% n w IZ1 7 their low statistical power to 3 5 z.g .y .-0 .-5s 5 j: - -- E Ek .- F 6 ~2 -.- E -.- 4 E p % actually detect significant differ- ? .? ..-- % .. % 52 c o -- &.: Q o < <" L Q ences. Also, many studies that -.- 2 simply determined causes of ,- r L 3 5 .$ ,G LY 2 2 : mortality with no additional " - .- -2 -0 -- 2 9 L?. m L --- 5 2 5 7d cc 5 2 c%Z 2 2 information inform us only that CL -c g - - predation may or may not be an % 2-2 9S - important mortality factor, but E 0 's. 2 - d c\ I 5"-0 LC say nothing about whether pre- -50- ? y 5 2 'Z -. A, . 2%. z r- dation significantly limits or reg- .: 3 -2 5 L C 2 .-0 ulates populatioll size. However, x .% -.-. u L - &. -. L2- 5% 3 z" 2 $3 our review of case histories has - 2 3 2 2 L g uncovered some notable pat- z a L - 0 7 - terns. $5 X x 5 15 .-,1 T Fetal rates/adult mule deer doe '2-r, ' 3 " u 2 '+ 5 have been relatively high - 28 Q (1.O- 1.9 fetuses/doe) over a zi .; E - r. -r5. .- C < x X X X X X wide distribution of deer densi- 35% +: 3 sac ,- .- % ,g L ties that ranged from well below 5 2~ J;"L K to habitat carrying capacity v, 0 %.$ E - 5 3- Most fawn mortality occurred 53 d ?- ri- N* 7 r. # 3= cob, .z T, a ern 7 ^C immediately following parturi- L +. ,2 3;- * 'i r-l tion and during mid- to late win- E -4 -G N -; 2kS - Y E &-" E ter, but in some studies substan- % ,zm 3 >> E Y V -+- CO c tial fawn mortality occurred in p72 Eo"? - .F& g 5 .CT;N3$Ln -C - =.C z.? r- 4 N =: i late summer. Coyote predation r. c.l FWN I\ -, -C .- Egz -5 has been implicated as a signifi- 3 z .z m GY 0 dii S 22 cant cause of mortality in most ,&? $2 ,g + ?. ?.* 525 .~~u- , mule deer studies that were 2%% B 7 5 5% 2%,$ - &zg- 6 5 c ;s;c U-c! v .-2: 2 - 2 based on either radiotelemetry Tj y2 9 7 E 3rc 0 5 5 or experimental manipulation of u 0 '- ,0 'aC - - - a .- - predator populations where pre- ,_= - -2 f S j i x mr,> Ty .- x x + 2 5 .z C $ 2 20 i ++ 2 $ Q dation was thought to be a prob- zE2 "- L -. -- &g 3 lem. Although bobcats kill deer, no study has implicated them as Synthesis of effects of predation on a major source of deer mortality. However, our deer review focused on studies of predator-prey relationships in which the subject of predation was When compared to other North American deer addressed or was a primary study objective. The species, relatively few studies have beell conducted literature is full of general ecological studies in on effects of predation on mule and black-tailed which predation was not considered a problem and 106 Wildlife Society Bulletin 2001,29(1):99- 115

the deer population had been cropped prior to exceeding K. Bartmann et a1. (1992) evaluated effects of coyote predation on mule deer fawn survival during winter where the population was at or near K. At this level, removal of coyotes resulted in fawns dying of starvation rather than predation. Predation mortality was compensatory at the reported mule deer densities and coyote removal had no impact on fawn survival. In other words, deer killed by coyotes were predisposed to die of other factors because the population was at or above K. Other Most mortality in ungulate populations occurs among the studies where deer populations appeared to be youngest age classes. Photo courtesy of George Andrejko, Ari- well below K indicated that although neonates may zona Game and Fish Department. have weighed less and may have been more vulnerable due to drought conditions, they appeared subsequently was not studied. We did not review to not be predisposed to death. these studies. No studies have implicated predation Two studies have recently evaluated effects of by mountain lions as a major mortality factor of mountain lion predation on mule deer populations, neonatal fawns, but this may be because this aspect but only one of these involved experimental reduc- has not been studied with radiotelemetry. tion of lion numbers to improve fawn survival. However, mountain lions have been implicated as Logan et al. (1996) studied effects of a mountain major predators of adult deer. lion translocation experiment on mule deer sur- Variable weather obviously changes impacts of vival during a drought. Although the study lacked predation through changes in forage and cover an untreated area, they compared deer survival (Smith and LeCount 1979,Teer et al. 1991), changes before and after lion removal and concluded that in alternate prey densities (Hamlin et al. 1984), and lion predation was a compensatory form of mortal- impacts on deer physical condition that influence ity because of drought conditions. A concurrent vulnerability to predation (Unsworth et al. 1999). evaluation of an untreated area may have shed addi- However, it appears that effects of weather may be tional light on effects of mountain lion predation. modified by predator densities, prey vulnerability, Bleich and Taylor (1998) suggested that mountain and by a deer population's relationship to K. lion predation may have been regulating mule deer Perhaps the best examples of this come from herds in the unpredictable environments of the predator-free enclosure studies. western Great Basin of California and Nevada. Smith and LeCount (1979) demonstrated that Effects of predation on black-tailed deer appear predation was a major limiting factor in their more pronounced because of the predator species Arizona study as the deer population was well involved. Wolves are the principal predator of below K and the deer population in a predator-free black-tailed deer in British Columbia and Alaska. In enclosure increased from about 3-5 to 18 these systems, wolves have effectively eliminated deer/km2. Fawn survival was greater during wet coyotes as serious predators of deer, whereas in years than dry years, but overall, the predator-free northeastern portions of the continent where enclosure population averaged 30 fawns/100 does wolves have been eliminated, coyotes have greater than populations outside. However, deer replaced wolves as effective predators of white- densities also increased outside the enclosure once tailed deer (Ballard et al. 1999). Two experiments favorable weather conditions returned. Teer et al. evaluated effects of wolf predation on black-tailed (1991) reported a similar experiment for white- deer on islands. Wolves introduced to a small island tailed deer in south Texas with similar results. In where they had not existed previously caused the both studies, protected deer populations increased deer population to decline to very low levels (Klein to levels far greater inside than outside enclosures 1995). McNay and Voller (1995) found mountain and then declined because deer numbers exceeded lions and wolves to be significant predators of adult food supply. These authors did not know whether female deer on Vancouver Island. Atkinson and Janz greater deer densities could have been sustained if (1994) experimentally reduced wolf numbers and Deer-predator relationships Ballard et al. 107 documented large increases in fawn survival when vide hunting and viewing opportunities, the dis- the deer population was well below K. A relation- tinction between these terms may not matter (Van ship exists between presence of wolves and deer Ballenberghe and Ballard 1994). If management of numbers on islands in southeast Alaska. On islands a significant limiting factor can result in increased south of Frederick Sound where wolves occur, deer deer harvest, then managers must be able to identi- numbers are low; deer numbers are substantially fy conditions under which such factors impact greater on islands north of Frederick Sound, where deer. Conditions that allow predation to become wolves are absent (Smith et al. 1987). Smith et al. an important limiting factor are poorly understood (1987) suggested that the prolonged recovery (i.e., in natural ecosystems with a full complement of 25 years) of deer south of Frederick Sound may be predators. In very altered ecosystems, such as due to wolf predation. Numerous studies of white- those where mule, black-tailed, and white-tailed tailed deer demonstrated that fawn survival on rel- deer exist, there is even more confusion about how atively small areas (2-194 km2) can be increased and when predation can become an important with large budgets by removing predators when mortality factor. Some insight can be gained by predation has been identified as a major limiting examining conditions under which predation has factor and when deer numbers are well below K. become an important factor on other species of Biologists continue to debate whether predation ungulates. is a regulatory or a limiting factor (Sinclair 1991, Numerous studies in arctic ecosystems, where Skogland 1991, Boutin 1992,Van Ballenberghe and presumably natural systems continue to function Ballard 1994), but to wildlife managers who are with minimal impacts by man, suggest that preda- responsible for managing deer populations to pro- tion becomes an important mortality factor when severe weather or human over-harvest initially cause population declines and then mortality from predation either retards or prevents population recovery (Gasaway et al. 1983, 1992; Van Ballenberghe 1985; Ballard et al. 1987, 1997; Bergerud and Ballard 1988,1989). Connolly (1978, 1981) and our review suggest that predation does not cause ungulate population declines, although some studies implicated predation as the causative factor for moose and caribou population declines (Gasaway et al. 1992, Seip 1992). In altered ecosystems such as those in the contiguous United States, and the Southwest in particular, prey populations may not respond or behave as those in unaltered or natural systems. In such systems Van Ballenberghe and Ballard (1994) concluded that predation may or may not be strongly limiting or regulatory, depending on degree of human influence on predators, prey, and habitat and on presence and relative densities of all species of predators. These conditions may explain why, in many deer populations, predation is not an important mortality factor or, where predation is the most important limiting factor,why population growth is not impacted. Elimination of major predators such as wolves and grizzly bears, livestock grazing, competition from livestock and other big game species, loss and fragmentation of Coyote predation has been implicated as a significant cause of habitats, and other major human influences alter mortality in most mule deer studies that were based on either relationships among predators, habitat, weather, radiotelemetry or experimental manipulation of predator populations where predation was thought to be a problem. Photo cour- and harvest by humans. Major changes in predator tesy of George Andrejko, Arizona Game and Fish Department. species composition also appear to alter how 108 Wildlife Society Bulletin 2001,29(1):99- 115 predators affect ungulate populations such as mule deer. Biologists continue to argue whether predator kill rates depend on density of prey species or on other factors. Kill rates appear to depend on whether the predator species is an obligate or facultative carnivore. For obligate predators, kill rates appear fairly consistent over a wide range of prey densities (Ballard and Van Ballenberghe 1998). Kill rates of facultative carnivores such as bears and perhaps coyotes have been thought or assumed to be independent of prey densities. However, kill rates by facultative carnivores may actually be more dependent on prey density than kill rates of obligate carnivores, particularly for predation on neonates. Solomon (1949) and Holling (1959) first proposed the idea of a functional response of predators to prey densities. Holling (1965) experiment- ed with consumption of nonmovable sawfly (Neodiprion sertifer) cocoons or dog biscuits by small mammals. Small mammals continued eating these items to the point of satiation. Predation on neonate ungulates may fit this pattern because they are largely stationary for several days, whereas predation on adults probably does not demonstrate a strong functional predator response. However, this may change when adults become vulnerable due to weather conditions. Predation by wolves and Additional experimental long-term research, particularly on mountain lions fits the scenario of obligate carni- coyote, mountain lion, and black bear predation, is needed to vores, whereas predation by bears and coyotes may clarify the role of predation on deer. Photo courtesy of Arizona Game and Fish Department. fit that of facultative carnivores. Black bears appear to have greater reproductive rates and survival when neonate ungulates are avail- has been identified, managers must estimate the able as prey (Schwartz and Franzmann 1991). deer population's relationship to habitat carrying Perhaps the same is true for facultative carnivores capacity to determine whether predator control like coyotes. Like bears, coyotes appear to have the may be warranted. ability to switch to other food resources, such as As mentioned earlier, few managers know or can fruits, and thus are not dependent on ungulates for measure K, but several indices of relationship to K survival. Such a system complicates attempts to are available. Low natality rates, low fawn:doe make generalized statements about effects of pre- ratios, poor body condition,high utilization of avail- dation on deer. able forage, and high deer population densities Several criteria exist that may help identify when should provide a reasonable indication that a deer predation has become an important mortality fac- population may be at, near, or above habitat carry- tor. First, predation must be identified as an impor- ing capacity. Unfavorable weather conditions may tant source of mortality. Seasonal or long-term alter this relationship, and if predator reduction changes in fawn:doe ratios can indicate when most were conducted, then managers should expect a losses are occurring, but can not be used to deter- depressed prey response until weather conditions mine causes for changes. Simple changes in become more favorable. If predation has been fawn:doe ratios can not be used to determine identified as a major limiting factor and the deer whether predation is a limiting factor. Only population is well below K, other important factors through intensive studies can predation be identi- must be considered before initiating predator fied as a significant mortality factor. If predation reduction. Deer-predator relationships Ballard et al. 109

Costs and benefits of predator pared to harvests predicted without wolf control. Economic benefits were realized by Alaska and reduction programs hunting-related businesses profited with increased Historically, 3 methods were used to reduce pred- game availability. Ballard and Stephenson (1982) ator populations: poisoning, trapping, and shooting indicated that an average of between $770 and from helicopters and fured-wing aircraft. The toxi- $873, excluding personnel costs, was spent for each cant Compound 1080 appeared to be relatively wolf harvested by aerial shooting in south-central effective to reduce predator numbers, but its use on Alaska, whereas Reid and Janz (1995) estimated public lands was banned in 1972 (Connolly 1978) that resident deer hunters on Vancouver Island, because of public distaste for government killing of British Columbia, received a $5.90 benefit for every predators and deaths of many nontarget species $1.OO spent on wolf control. Even fewer data exist (Hamlin 1997). concerning benefit:cost ratios for control programs Trapping also has been a popular method to involving mule or white-tailed deer. reduce predator numbers. However, to significant- The Western States and Provinces Deer and Elk ly reduce predator numbers, trapping must extend Workshop (R. M. Lee, Arizona Game and Fish well beyond the point of economic feasibility for Department, unpublished data) sumeyed western most trappers. Further, trapping is effective only in states and provinces to determine whether they small areas (Hamlin 1997). Small reductions in practiced predator reduction to increase wild ungu- predator numbers are unlikely to have any lasting late populations and determine annual expendi- effect on predator populations. For example, tures. At the time of the survey, only British Connolly and Longhurst (1975) estimated that coy- Columbia indicated wolves, mountain lions, and otes could easily withstand annual harvests of 70%, bears were important predators of mule deer and and even with 75% harvest, coyote populations that they intended to provide more liberal predator could persist for 50 years. Harvest levels for wolves hunting and trapping seasons in areas where pre- must exceed 50% to cause significant population dation was suppressing deer populations. They also declines (Ballard et al. 1997). indicated that wolf reduction may be essential to Hamlin (1997) indicated that managers had used maintain ungulate populations at prescribed levels 2 approaches to reduce coyote numbers. The first in some areas. Alaska, California, Colorado, Idaho, was to reduce overall numbers over large areas and Montana, Nevada, New Mexico, Oregon, Texas, and the second was to use selective control on individ- Washington reported they did not have predator uals or small populations that likely accounted for reduction programs to benefit big-game species, most prey mortality. Trapping over large areas has but they did have programs for problem wildlife or been largely unsuccessful in achieving desired depredation reduction for livestock or other agri- results of increasing game populations. Selective culture programs. Arizona was one of the few harvests immediately prior to denning may, howev- states that had an active predator reduction pro- er, reduce neonatal mortality in small areas (Hamlin gram to reduce coyote populations to improve 1997). He concluded that substantial knowledge of pronghorn (Antilocapra americana) fawn survival coyote territories and denning areas was essential for trapping to be effective and that intensive effort beyond levels resulting from fur price incentives was necessary to reduce losses of game to coyote predation. Aerial shooting is probably the most effective method to reduce predator numbers, but social acceptability is low (Boertje et al. 1996). Relatively few data exist on benefit:cost ratios of predator control programs, and few studies have actually demonstrated that reductions in predators actually resulted in increased human harvests. Notable exceptions were provided by Boertje et al. (1996), who indicated that wolf control and mild weather in harvest Deep snow makes deer more vulnerable to predation. Photo thousand additional moose and caribou, as com- by Len Carpenter. 110 Wikiqe Society Bulletin 2001,29(1):W-1 15 in specific areas; annual expenditures were approxi- Michigan attempted to ban spring bear hunting; in mately $22,900. Only7 Colorado and Montana indi- Oregon an initiative attempted to ban bear baiting cated annual costs for depredation reduction and hounding for bears and mountain lions. Alaska ($120,000 and $60,000, respectively). Since the voters banned use of aircraft to harvest wolves in 1997 survey,predator reduction programs have been 1998,but this initiative was recently overturned by- initiated in Alaska, Nevada, New Mexico, and Utah. the Alaska state legislature. Most individuals (75%) Recently, Utah approved predator management in the IJnited States stronglj7 or moderately approve plans for 15 management units to try to reduce of legal hunting and believe hunting should contin- mountain lions and coyotes and thus benefit mule ue to be legal (Duda et al. 1998). particularly if hunt- deer populations (Bates and Welch 1999). ing is for food, to manage game populations, or to Bodenchuk (United States Department of control animal populations. Regardless, certain Agriculture [USDA] ,Animal Plant Health Inspection methods of and reasons for hunting or trapping are Service,Wildlife Services, unpublished letter to Mr. clearly not accepted by many members of the pub- Jim Karpowitz dated 1 June 1999) summarized lic. However, the public will apparently accept costs and benefits associated with predator control trapping and predator control under certain cir- activities in 3 units (Henry Mountains, North cumstances. For example, although most respon- Bookcliffs, and Pahvant). His analysis included sev- dents to an Illinois survey disapproved of trapping, eral important assumptions: increases in fawn sur- it received the most approval when involved with vival and deer herd numbers were entirely due to animal damage control (71%). animal population predator reduction, costs included all predator control (70"/0),or biological research (63%. reduction for the deer program in addition to 50% Responsive Management 1994). In Utah, there was of costs associated with domestic livestock depre- strong support for and strong opposition to use of dation control, and the civil value of a deer in Utah predator control to protect game populations was $300 Shooting from fixed-wing aircraft and among the general populace (Krannich and Teel helicopter in addition to ground shooting and trap- 1999). However, among hunting license buyers ping were used. Benefit:cost ratios for the 3 areas there was moderate support for controlling preda- ranged from 11 to 23:1. However, Bates and Welch tors, whereas the general public was somewhat (1999) summarized the status of the 15 deer man- neutral on the issue (Krannich and Teel 1999). agement areas after 3 years of predator control and Messmer et al. (1999) measured public response to found that treated and untreated areas increased or predator control of mid-sized carnivores to remained stable and that results were equivocal. enhance avian recruitment. They found that the Bodenchuk (LTSDA.Animal Plant Health Inspection public was more prone to support predator reduc- Service, Wildlife Services, letter to Mr. Jim tions in avian populations if such control actions Karpowitz dated 1 June 1999) admitted that preda- were surgical in nature rather than widespread. tor control activities were not entirely responsible Clearly, wildlife managers need Inore information for the increases in fawn survival because hunting concerning public attitudes on large predator pressure on bucks had decreased and that wet reductions to favor game species. years had improved habitat and deer survival, but that "predator control projects have their place in wildlife management." Conclusions Numerous factors will dictate whether predator reduction may be warranted. These include public Human dimensions of predator acceptance, scale, methodology biological rele- management vance, and relationship to habitat carrying capacity. Trapping as a method of harvesting fiirbearers for Public acceptance of predator reduction has recreational use and for economic return appears changed drastically during the past 4 decades. to have become increasingly unpopular with the Public attitudes toward wildlife have changed along general public, particularly when leghold traps with changes in human population distribution, were involved. Recent ballot initiatives have education, and economic status. A proportion of banned or proposed banning trapping and using the human populace will not accept predator snares in Arizona, California, Colorado, and reduction, regardless of the reason (e.g., endan- Massachusetts (Andelt et al. 1999). An initiative in gered species conservation and particularly for Deer-predator relationships Ballard et al. 111 production of animals for sport harvest). Alaska some predator reductions may be acceptable if probably bas the best management data to support such programs are viewed as focused, site-specific predator reduction in limited areas,but public con- operations rather than broad-scale programs cover- troversy has stalled or canceled programs that were ing large areas. justified biologically (see Stephenson et al. 1995). Most studies we reviewed were relatively short- Strong public support would be necessary to initi- term and were conducted in relatively small areas, ate predator reduction,even if biologically justified. and only a few actually demonstrated increased Managers have a poor understanding of how the fawn recruitment and subsequent larger harvests public views predator reductioil programs and by humans (i.e.,wolf reductions). Also, conditions under what conditions such programs might be that led to a particular deer population being limit- acceptable. ed by predation were poorly documented. Scale of a predator removal program also will ulti- Additional experimental long-term research, partic- mately contribute to success of the program and ularly on coyote,mountain lion, and black bear pre- must be addressed by wildlife managers. To date,all dation, is needed to clariQ the role of predation on research (excluding that conducted on wolves) deer. An experimental approach is necessary concerning predator removal programs to enhance whereby deer population performance in relation deer populations has been conducted on small to predator removal is monitored in manipulated areas (i.e.,<1,000km2). Although managers would and unmanipulated areas. Such experiments like to affect ungulate populations over large areas, should be conducted over sufficient time such that recent research suggests that the public is more severe and favorable weather conditions occur. likely to accept predator control if it is conducted Confounding factors such as other predators. in small areas. human harvests, and alternate prey species also Current bans on use of poisons restrict reduction must be measured to allow proper assessment. methods to aerial shooting (from fixed-wing air- Managers also need to document conditions craft or helicopter), trapping, or ground shooting. under which predation becomes a significant limit- Trapping and ground shooting are relatively inef- ing factor,identify conditions under which predator fective by themselves to reduce predator densities. control should be implemented, and determine Thus, although aerial shooting is likely the most when control should be ended. Deer density in rela- effective method, it also is the most expensive tion to K appears to be a key consideration. Deer method. Aerial shooting in large areas to enhance densities vary widely over the range of deer distri- ungulates at the population level will probably be bution, but managers should qualitatively assess the cost-prohibitive unless specific areas are identified. relationship of individual deer herds to habitat car- For example. Smith et al. (1986) documented a rying capacity. The lower a population in relation to 400% increase in a pronghorn population at K,the greater the likelihood that predator reduction Anderson Mesa,Arizona. Helicopter gunning was would result in measurable increases in sun7ival and used during March through May to kill coyotes on herd numbers. Managers then need to determine a 490-km2 area. Although they estimated that only when most mortalities occur and whether preda- 22-296 of the coyote population was removed tion is an important cause. These types of informa- each year, they speculated that removal of 30% of tion can be obtained by using methods varying from breeding females and disruption of denning activi- relatively inexpensive examination of autumn and ties may have had a disproportionate positive effect winter fawn:doe ratios to expensive methods such on the pronghorn population. They examined ben- as capturing and monitoring neonate fawns through efit:cost ratios and schedules for coyote reduction a biological year to determine causes of mortality. and found that the greatest benefit:cost ratio (132) Large losses immediately following parturition are occurred when reduction was conducted once usually indicative of popdations experiencing high every 2 years. If managers could identify similar mortality due to predation. Managers then need to types of areas,such as fawning concentration areas decide the scale of control in relation to resources for mule or black-tailed deer, then intensive preda- available and public acceptance. However. deter- tor reduction could be feasible just prior to fawn- mining when to halt predator control programs is ing,assuming that predation is a significant limiting equally important. factor and the prey population is below K. The literature has many examples in which pred- Available human dimensions research suggests that ator reduction programs resulted in increased deer 1 1 2 WildIife Society Bulleti??200 1,29(1):99-115 survival and populations quickly increased. only to ulations sufficiently to be effective. and exceed habitat carrying capacity. Managers should 4)where control efforts were on large-scale halt predator control well before deer populatiotls areas. reach these levels and ensure that harvests by Numerous important factors must be considered hunters are adequate to stabilize deer numbers. prior to making a decision on whether to imple- Perhaps one of the greatest needs is documenta- ment a predator control program. Minimally, the tion of costs and benefits under varying levels of following steps should be in place: predator reduction. Although several studies demonstrated increases in deer survival in relative- 1) a management plan that identifies the follow- Iy small areas because of predator reduction, only a ing: current status of mule deer populations few studies on moose and caribou and one on and population objective clesired from the black-tailed deer actually documented that predator control project, desired reilloval increased survival was eventually passed on to goals for the predator species, timing and hunters. The scale and level of predator manage- method of removal efforts, scale of removal ment and public acceptance will likely determine effort, and what other limiting factors may be whether predator control is a viable management playing a role in depressing mule deer popu- tool. Lastly, managers need a better understanding lations; of how the general public reacts to their programs: 2)an adaptive management plan that sets moni- also needed is additional research on the human toring criteria that would resdt in evaluation dimensions of predator control. of predator and prej populations and identifies thresholds at which predator colltrol will Recommendations be eliminated or modified. The relationship between predators and their One failure in much of the research that has been prey is a very complex issue. The literature we completed on the utility of predator control to reviewed is eq~rivocal;in some cases predator con- improve mule deer populations is a lack of an ade- trol appeared to be useful in improving deer popu- quate experimental design. Also, additional lations and in some cases it was not. Some similar- research is needed, particularly in social aspects ities from cases in which predator control appeared related to predator control. To assist in improving to be effective are: the decision-making process related to predator management. we believe the following research is 1) predator control was implemented when the vitally needed deer populations were below habitat carrying capacity, 1) Experimental, long-tern1 research, particularly 2) predation was identified as a limiting factor, on coyote. mountain lion, and black bear pre- 3) control efforts reduced predator populations dation, to clari* the role of predation on deer. enough to yield results (e.g., expected to be An experimental approach is necessary approximately 70% of a local coyote popula- whereby deer population performance in tion). relation to predator removal is monitored it1 4) control efforts were timed to be most effec- manipulated and urlmanipulated areas. Such tive (just prior to predator or prey reproduc- experiments should be conducted over a suf- tion), and ficient period of time that severe and favor- 5) control occurred at a focused scale (generally able weather conditions occur. Confounding <1,000 km2 [259 mi2]). factors such as other predators, human harvests. buffer species densities. and habitat Cotlversely, there were similarities where preda- conditions also must be measured to allow tor control was not effective or its effectiveness at appropriate interpretation of the study improving mule deer populations could not be results. measured. These included: 2) Well-designed research to measure social atti- 1) when mule deer populations were at or near tudes toward various aspects of predator con- habitat carrying capacity. trol programs. Minimally. we believe research 2) when predation was not a key limiting factor, should develop a better understanding of 3) where control failed to reduce predator pop- variation in public attitudes on methods of Deer-predator relationships Ballard et al. 113

reda at or control, timing-. of reda at or control. RATES,B., AXD M. WELCH.1999. Managing mule deer recovery prey species thresholds that would call for through the use of predator management plans. Proceedings of the 1999 deer-elk workshop. Ctah Division of Wildlife implementation of a predator control pro- Resources. 4-6 March 1999,Salt Lake City. USA. gram, and of a predator pro- BEASOM.5. L. 1974a. Relationships between predator removal gram. and white-tailed deer net productivity. Journal of Wildlife 3)Research should include an analysis of Management 38:854-859. costbenefit ratios of any control efforts. BLISOM,S. L. 19'4b. Intensive short-term predator removal as a game management tool. Transactions of the North American Wildlife and Natural Resources Conference 39:230-240. Acknoulledgments. This manuscript was greatly BERGERLD.A.T.,.IUD W. B. B.DAI.IARII.1988. Wolf predation on cari- improved by constructive criticism provided by J. bou: the Nelchina herd case histor!. a different interpreta- R. Heffelfinger,\! C. Bleich,G. C.White,H.A.Whitlaw, tion. Journal of Wildlife Alanagement 52:344-35'. T. Beck. J. Bowman, J. Ellenberger, G. Erickson, B. BERGERLr).A.T..soW.B. BALLIRD.1989,Wolf predation on Nelchi- na caribou: a reply Journal of Wildlife Management 53: I. R. M. Glasgow, Hatter, Kahn, Kirchhoff, T. 251-259. Thompson, B.Wakeling,M.Welch,and D.Weybright. BLEIC-H.\: C.. AXI) T. J. T,~YLOR.1998. Survi\rorship and cause- This isTexasTech University College of Agricultural specific mortality in five populations of mule deer. Great Sciences and Natural Resources technical publica- Basin Naturalist 58:265-2'2. tion T-9-853. BOERTJI:.R. D.. P.V~AI.KENBLRC..XL) M.E. M(-N.i\: 1996. Increases in moose, caribou. and wolves following wolf control in Alaska. Journal of Wildlife Management 60:47'4-489. BOLTIX.S. 1992. Predation and moose population dymamics: a Literature cited critique. Journal of Wildlife Management 56: 116-12'. ANDI'IJ.WE. R. L. PHII.IIPS. R. H. SCIIZIIUT..INDR. H. G11.1. 1999. Trap- C.~ROLL.B. K...IuDD. L. BROUS.19''. Factors affecting neonatal ping furbearers: an overview of the biological and social fawn survival in southern-central Texas. Journal of Wildlife issues surrounding a public policy controversy. Wildlife Soci- Management 41:63-69. ety Bulletin 2': 53-64. C~YK~IIY,G. E. 19'8. Predators and predator control. Pages ArK~nsoh,K.T. .AN) D.U: J.*-z. 1994. Effect of wolf control on 369-394 in J. L. Schmidt and D. L. Gilbert. editors. Big Game black-tailed deer in the NimpkishValleyon lJancouverIsland. of North America. Stackpole. Harrisburg, Pennsylvania,LEA. Ministry of Environment, Lands and Parks. Wildlife Bulletin COKNOLLYG. E. 1981. Limiting factors and population regula- Number B-'3. Nanaimo. British Columbia, Canada. tion. Pages 245-285 in 0.C.Wallmo,editor. Mule and black- ALSTI~.D. D.,PJ. URXES~.AUDM. L.W~LFE.19". The influence of tailed deer of North America. University of Nebraska, Lin- predator control on two adjacent wintering deer herds. coln, USA. Great Basin Naturalist 3-: 101- 102. Cou~o~i,~.G. E...IxI)W hi. L~K~~HLTRST.1975. The effects of con- BALLIRI).W.B..L.Lk.Al-~~~.R. KRAT<\IAN, D.J.&ED. AND S. G. FAUCY trol on coyote populations. University of California.Division 199'. Ecology of wolves in relation to a migratory caribou of Agricultural Science. Bulletin 1872.Davis,USA. herd in northwest Alaska. Wildlife hlonographs 135 COOK.R. S.,M.W-HITE,D.~.TR~INER..LUDW C.GLUENER.1971. Mor- BALL~RII.UC B..ASI) R. 0.STEPHEXSON.1982. UJolf control-take tality of young white-tailed deer fawns in southTexas. Jour- some and leave some. Aces 18:2'6-300. nal of Wildlife Management 35:4'-56. B.~II.IRD,W: B., IT. A. WHITI.A\V,S.J. Y011(;. R. A. JEXKIUS.AUU G. J. D.II.E.R. U!. L. G. An.~\ls,bun R. T. RO\\YER.1994. Functional FORBES.1999. Predation and survival of white-tailed deer response of wolves preying on barren-ground caribou in a faurns in north-central New Brunswick. Journal of Wildlife multiple-prey ecosystem. Journal of Animal Ecolol~~.63: Management 63:5'4-579. 644-652. RAII.\RI).W,B..J. S.WHIT\IA~..*ADC. L. GARDNCR.1987. Ecology of DEIZLEI.R. N. 19-6. Regulations and mule cleer harvest-politi- an exploited wolf population in south-centralAkaska. Wildlife cal and biological management. Pages 85-92 in G. W Work- Monographs 98. luan and J. B. Low, editors. Mule deer decline in the West: a BALLIKD.~:B..ASDVVANBALLENBERGHE.19g7. Predator-prey rela- syniposium. Utah State University, Logan. USA. tionships. Pages 2472'3 in A. %: Franzmann and C. C. DEY~Iu(;. C.A. 1989. Mortality of adult male white-tailed deer Schwartz. editors. Ecology and management of the North in south Texas. Journal of Wildlife Management 53:513-523. American moose. Smithsonian Institution. Washington, D.C.. Dlc~~\so\.T.G.. G. E. W,IMPIER.G. W GARNER.AYD C. D. SI~IPSOK. ITSh. 1980. Mortality of desrrt mule deer fawns in Pecos County BAI.I~\RI).M. I).. I~I) VAXBALLEXBERGHE.1998. Moose-predator Texas. Proceedings Western Association of Fish and Wildlife relationships: research and management needs, Alces 34: Agencies 60:581-592. 91-105. Dr D.A.51. D..S.J.BISSEI.L.ANDK. C.Y~rhc;. 1998. Wildlife and the BARSAESS~J.1998. The new killers!:wolves, cougars,and coyotes American mind: public opinion 011 and attitudes toward fish take a bite out of your hunting. Field & Stream 20:58-65. and wildlife management. Responsive Management. Har- BART~IANN.R. &I,.G. C:. W-HITE,.%XI) L. H. CARPESTER.1992. Com- risonburg.Virginia USA. pensatory mortality in a Colorado mule deer population. GARLER.G.W 1976. Mortality of white-tailed deer fax-ns in the Wildlife Monographs 121 Wichita Mountains. Oklahoma. Dissertation. Oklahoma State B.~I.I,sH.W.S. 1978. Mortality of white-tailed deer fawns in the University, Stillwater. ITSA. Wichita Mountains. Comanche Count): Oklahoma. Part 11. G.IRNFR,G. W.J.A. MORRISOU,4x1)J. C. LEWIS.1976. Mortality of Thesis. Oklahoma State IJniversity,Stillwater.USA. n-hire-taileddeer fawns in the Wichita Mountains.Oklahoma. 1 14 Wildlife Society Bulleth?2001,29(1):99-1 15

Proceedings Annual Conference Southeastern Association resources. lltah State University. Logan. lJSA. Fish and Wildlife Agencies 13:493-iO6. KI.(.ER~.T.E. 1988. Ecology and population dynamics of mule GAS.AWA\,\VC., R. D. BOEKTIE.D.V (;R~N,AARD.D. (;. KELIE'IHO~SF,R. deer in the eastern Sierra Nevada. California. Dissertation. 0.SI.EPHEUSO\.ASDD. G. LAI~SEX,1992. The role of predation Llniversit) of California.Berkeley-, 1-SA. it1 limiting moose at low delisities in rUaska and Yukon and KLkticr., K. E .AN L. D. MECH.1994. Wolf and hear predation on implications for conservation. Wildlife Monographs 120. white-tailed deer Pam-11s in northe;lstern klinnesota. Canadi- G.\sA\\iu:W (1..K 0 SIEI'HI:~~OS.J. L. DAVIS.P E. K. SHFPHERD..\UI)0. a11 Journal of Zoology -2: 1i5-- 1565. E. BIRRIS. 1983. Interrelationships of n-olves. prey and man LECOIs7.A. lo--. Causes of fawn mortality Arizona Game and in interior hlaska. Wildlife Monographs 84. Fish Department. Federal hid to Wildlife Restorarioti Report. H.11 1.s. L. K.. editor. 1981. White-tailed deer ecology and man- W-'8-R.Vliork Plan 2. Job 11. Plioenix.Arizona. 1'SA agement. Stackpole. Harrisburg. Pennsylvania. USA. LOG.%>.K.A..L. L. S\\ E&VOK.T.K. RI.TH.AUI) 41. G. HOKSOC~~ER.1996. KUII.IS.K. L. l9c)' Survival and home range fidelity of coyotes Cougars of the San Aidres hlounrains. New hlexico. New in Montana: i~nplicatio~isfor control. Intermo~~ntainJournal Mexico Department of Fish and Game. Federal hid to Wildlife of Sciences 5:62--2. Restoration Fitla1 Report. Project \V-128-R. Santa Fe. ITSh. H.&\IIIv.K. L..&>I) R.J. LW&(:~~~.1989. Xlule deer in the hlissouri ~IALKIE.R.J.. K. L. H.&ac~r~..w~)D. F. P.K. 1982. Xlule deer. Pages lGver Breaks. Montana: a study of population dynamics in a 862-8'- in J.A. Chapman and G.A. Feldhamer, editors. Wild fluctuating environment. Montana Fish.Wildlife. and Parks, mammals of North Anerica. Biology management, and eco- Wildlife Division. Federal Aid to W~ldlifeRestoration Final nomics. The Johns Hopkins Llniversit!: Baltimore. xlaryland. Report. Project \V-120-R-'-18, Program Sumber 1. Study BG- L-SA. 1.O.JobNumbers 2 and 3, Helena. IISA. A~ACKII:. R. J., D. E P.K. K. L. H.\\lll\i,431) G. L DL\EK.1998. Ecolo- HA,\II.IU.K. I... 5.J. RILLLD. RRIII,A. R.\K Doo~).AVL) R. J. ,M.K:KIE g! and management of mule deer ant1 white-tailed deer in 1981. Relationships among ITILIL~ deer fdwn mortality. col-- bIontana. Mclotitana Fish.W~ldlifeand Parks.\Vildlife Division. ores. and alternate prey species. Journal of Wildlife Slanage- Frderal Aid to Wildlife Restoration Report. Project W-1 20-R. ment 48: 489-199. Helena. L-Sh. H.~I,.IER.1.W: 1988. Effects of wolf predation on recruitment of 11u xu%.J. 1985. Carrying capacity and related slippery shibho- black-tailed deer on northeastern Vancourer Island. Wildlife leths. Wildlife Society Bulletin 13:10.i-410. Branch. Slinistn of Environn~ent.WildlifeReport Number R- MCNAV.R. S.. nnJ. S~.\'OLLER,1995. hlortality causes and survival 23,Victoria. British (:olumbia, Canada. estiniates for adult female Colum1,ian black-tailed deer. Jour- K~ER.I. \V.. .&XI)D. \X. J.\uTz. 1991. Apparent demographic 1ra1 of \Vildlife Management 59: 138- 116. changes in black-tailed deer associated n-ith m-olf control (111 MECH.L. D.. .&XI) l? D. K~RUI.19--. Role of the m701f in a deer northern VT:~ncouverIsland. <:anadian Journal of Zoology '2: decline in the Superior National l:orest. Ynited Srates Forest 8'8-884, Service. Research Paper NC-148, St. Paul. 4linnesota. IISA. HEI-~EII:IU(;ER. J. R. 1989. Tlie effects of intensive coyote control MESSIER.F. 1991. The significance of limiting and regulating fac- on post-rut mortality of buck white-tailed deer. Thesis.Texas tors on the demography of moose and white-tailed deer. A&I 1Tniversity,Kingsrille. 1SA. Journal of Animal Ecolog! 60: 3"-393. HEFF~I.I:IU(.I.II.J. R.. S. L BE\SOM..&h~ C.A. l)~\iO[-x(;. 1990. The ~IESINER.T A.. 31.W.BKLSSOS.D. REIII.R,+.XI> D. Cr. HE\VI.IT.1999. effects of intensive coyote control on post-rut mortality of Vnited Srates public attitudes regarding predators and their male m-hite-taileddeer. Pages 3i-4i itz P R. Krnusman and N. nxlnagenlent to enhance avian recruitment. W~ldlifeSociety S. Smith. editors. hknaging wildlife in the Southwest. Ari- Bulletin 2':-5-85 zona Chapter of the Wildlife Society,Phoenix. CSA. NEL>o\.M.E.. A\D L. D. >II.(.H.1986n. Mortality of white-tailed HoILI\~,.C. S. 19i9. The components of predation as revealed deer in northeastern Alinnesota. Journal of Wildlife Wanage- by a study of small mammal predation of the European pine ment 50:691-698. sa~wfly Calladial1 Entoniology 91: 293-320. XI:[sou. M. E...\vr) L. D. \IE(:H. 19860. Relationship between snow Hal LI\<,.C. S. 1965. The functional responses of predators to depth and gray wolf predation on white-tailed deer. Journal prey density and its role in mimicry and population regula- of Wrildlife 3lanagement i0: 4'1- i-4. tion. Entomological Society of Canada. Alemoirs 4i: 140. RESPOU~.IVEM\KAGE\IE~T. 1994. Illinois residents' opinions and H~RIt'~.A~..AsD J.W.LTU>WORTH. 1998. So~~theastmule deer ecol- attitudes regarding trapping. fi~rhunting. and furbearer man- ogy. Study Uumber 1: Study Plan Development. Idaho agement. Report prepared for the Illi~ioisDepartment of Department of Fish and Game. Federal hid to Wildlife (:onserration. Responsive Alanagemcnt. Harrisonburg. Vir- Restoration Report. Project\V-160-R-25.Subproject 51. Boise. ginia. USA. L'Sh. REII).R...\XI) I) J\\z. 1995. Econon~icevaluation of Vancouver KII,.J. G...iun M. WHITE.1985. Population dytianiics of white- Island wolf control. Pages 515-521 ill L. N. Carbyn. S. H. railed deer (Odocoilr~~sz~i~ginic~r~~rs) on the Vl'elder Wildlife Fritts, and D. R. Seip. editors. Ecolog! and conser\ation of Refilge.Texas. Soutliwester~iNaturalist 30: 105-1 18. wolves in a changing world. Canadian Circumpolar Institute. KLEIY.D. R. 1995. Tlie introduction, increase. and demise of 1-niversity of Alberta. Edmonton. Canada. \voh-es on Coronation Island.Alaska. Pages 2-i-280 in L. \I. S.&I\\ASSER.H. 19'4. Coyote scats as an indicator of the time of Carhyn. S. 13. Fritts, and D. R. Seip, editors. Ecology and con- fin-n mortality in the ru'orth Kings deer herd. California Fish serwtion of wolves in a changing norld. Canadian Circunl- and Game 60:84-8'. polar Institute. L-niversio of Alberta. Edmonton. Canada. S\L\\A~\EK.H.. S.A. HOLE.A\II G.A.AsH(.R~FT.19-8. Fawn produc- KR~N\I(:H.R. S..IUD T L.?'EEL.1999. Attit~desand opinions about tion and survival in the North Kings River deer herd. Cali- wildlife resource conditions and management in Lirah: results tornia Fish and Game 64: 38-52. of a 1998 statewide general public atid license purchaser sur- SCIIXEE(,AS.E. R.. .I\IL) R. S. BI\ISTMD.lo'-. Decline of western 'e!: Institute for social science research on natural mule deer populations: probable cause. tentative solution. Deer-predator relationships Ballard et al. 115

Proceedings of Annual Conference of Western Association of WHITE,G. C., R.A. GAR ROT^, R. M. BARTMANN,L. H. CARPENTER,AND A. State Game and Fish Commissioners 57: 218-237. WALLDREGE.1987. Survival of mule deer in northwest Col- SCHWARTZ,C. C.,AND A.W FRANZMANN.1991. Interrelationship of orado. Journal of Wildlife Management 51:852-859. black bears to moose and forest succession in the northern WHITLAW,H. A,, W. B. BALLARD,D. L. SABINE,S. J.YOUNG, R. A. JENKINS, coniferous forest. Wildliife Monographs 113. AND G. J. FORBES.1998. Survival and cause-specitic mortality SEIP,D. R. 1992. Factors limiting woodland caribou populations rates of adult white-tailed deer in New Brunswick. Journal of and their interrelationships with wolves and moose in south- Wildlife Management 62: 1335-1341. eastern British Columbia. Canadian Journal of Zoology 70: WHITIAKER,D. G.,AND E G. LINDZFY.1999. Effect of coyote preda- 1494-1503. tion on early fawn survival in sympatric deer species. SINCIAIR,A.R. E. 1991. Science and the practice of wildlife man- Wildlife Society Bulletin 27: 256-262. agement. Journal of Wildlife Management 55: 767-773. WOFXMAN,G.W.,AND J. B. Low, editors. 1976. Mule deer decline SKOGLAND,T.1991. What are the effects of predators on large in the West: a symposium. Utah State University, Logan, USA. ungulate populations? Oikos 61:401-411. SMITH,C.A., E. L.YOUNG,C. R. LAND,AND K. l? BOVEE.1987. Preda- tor-induced limitations on deer population growth in South- Warren B. Ballard (photo) received his B.S. in fish east Alaska. Alaska Department of Fish and Game, Federal Aid and wildlife management to Wildlife Restoration Final Report, Project W-22-4,W-22-5, from New Mexico State and W-22-6, Juneau, USA. University, his M.S. in SMITH,R. H., AND A. LECOUNT.1979. Some factors affecting sur- environmental biology vival of desert mule deer fawns. Journal of Wildlife Manage- from Kansas State Univer- ment 43: 657-665. sity, and his Ph.D. in SMITH,R. H., D. J. NEW,AND N. G. WOOLSF(. 1986. Pronghorn wildlife science from the response to coyote control-a benefit:cost analysis. Wildlife University of Arizona. He Society Bulletin 14:226-231. currently is associate pro- fessor in the Department SOLOMON,M. E. 1949. The natural control of animal populations. of Range, Wildlife, and Journal ofAnimal Ecology 18: 1-35. Fisheries Management at STEIGERS,W D., JR.,AND J. T. FUNDERS.1980. Mortality and move- Texas Tech University. ments of mule deer fawns in Washington. Journal of Wildlife His professional interests Management 44: 381 -388. focus on predator-prey STEPHENSON,R. O., W B. BALLARD,C. A. SMITH,AND K. RICHARDSON. relationships and popula- 1995. Wolf biology and management in Alaska, 1981-1992. tion dynamics of carni- Pages 43-54 in L. N. Carbyn, S. H. Fritts, and D. R. Seip, editors. vores and ungulates. Ecology and conservation of wolves in a changing world. Daryl Lutz received his B.S. in wildlife management from the university of Wyoming Canadian Circumpolar Institute, University of Alberta, and his M.S. in wildlife biology from Humbolt State University. Edmonton, Canada. He currently serves as wildlife biologist supervisor for the STOUT,G. G. 1982. Effects of coyote reduction on white-tailed Casper Region of Wyoming Game and Fish Department. His deer productivity on Fort Sill, Oklahoma. Wildlife Society current interests center on research and management of big- Bulletin 10: 329-332. game species. fiomas W. Keegan is manager of the Upland TAYLOR,T.J. 1996. Condition and reproductive performance of Game and Furbearer Section of the Washington Department of female mule deer in the central Sierra Nevada. California Fish and Wildlife. He received his B.S. in natural resources Department of Fish and Game, Bishop, USA. from Cornell University, his M.S. in wildlife from Louisiana TEER,J. G., D. L. DRAWE,T.L. BLANKENSHIP,%! EANDELT, R. S. COOK,J. G. State University, and his Ph.D. in wildlife science from Oregon State University. His current professional interests center KIE,E E KNOWLTON,ANDM.WHITE. 1991. Deer and coyotes: the around upland and big-game management in the western Unit- Welder experiments. Transactions of the North American ed States. Len H. Carpenter is the southwest field representa- Wildlife and Natural Resources Conference 56: 550-560. tive for the Wildlife Management Institute. He received his B.S. TEMPLE,L. 1982. Mule deer research. New Mexico Department deeree from Colorado State Universitv and his Ph.D. in ranae" of Game and Fish, Federal Aid Project, New Mexico W-124-R- scGnce from Colorado State University. His principal profes- 5/Job 2, Santa Fe, USA. sional interests concern research and management of large TRAINER,C. E., J. C. LEMOS,T. KLsTNER,VC! C. LIGHTFOOT,AND D. E. mammals and their habitats and human dimension aspects of TOWEILL.1981. Mortality of mule deer fawns in southeastern wildlife policy and management. James C. deVos, Jr. received Oregon, 1968-1979. Oregon Department of Fish and his B.S. degree in wildlife management from Arizona State Uni- versity. He is currently the research branch chief for Arizona Wildlife,Wildlife Research Report Number 10, Portland, USA. Game and Fish Department. His professional interests focus on UNSWORTH,J. W., D. E PAC,G. C. WHITE,ANDR. M. BARTMANN.1999. management of wildlife in the Southwest and wildlife diseases. Mule deer survival in Colorado, Idaho, and Montana. Journal of Wildlife Management 63: 315-326. VAN BALLENBERGHE,V. 1985. Wolf predation on caribou: the Nelchina case history. Journal of Wildlife Management 49: 71 1-720. VANBAI.LENBERGHE,~,AND W. B. BALLARD. 1994. Limitation and regulation of moose populations: the role of predation. Canadi- an Journal of Zoology 72: 2071-2077. WALLMO,0.C., editor. 1981. Mule and black-tailed deer of North America. University of Nebraska, Lincoln, USA.