sign in sign up
<<
Home , Small population size

Management of Small Populations: Concepts Affecting the Recovery of Endangered Species Author(s): Gary E. Belovsky, John A. Bissonette, Raymond D. Dueser, Thomas C. Edwards, Jr., Christopher M. Luecke, Mark E. Ritchie, Jennifer B. Slade and Frederic H. Wagner Source: Wildlife Society Bulletin (1973-2006), Vol. 22, No. 2 (Summer, 1994), pp. 307-316 Published by: Wiley on behalf of the Wildlife Society Stable URL: http://www.jstor.org/stable/3783262 Accessed: 24-02-2016 16:01 UTC

REFERENCES Linked references are available on JSTOR for this article: http://www.jstor.org/stable/3783262?seq=1&cid=pdf-reference#references_tab_contents

You may need to log in to JSTOR to access the linked references.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at http://www.jstor.org/page/ info/about/policies/terms.jsp

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Wiley and Wildlife Society are collaborating with JSTOR to digitize, preserve and extend access to Wildlife Society Bulletin ( 1973-2006). http://www.jstor.org

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions Wildl. Soc. Bull. 22:307-316, 1994

""INMY OPINION .

MANAGEMENTOF SMALL POPULATIONS:CONCEPTS AFFECTINGTHE RECOVERYOF ENDANGEREDSPECIES

GARY E. BELOVSKY, ConservationBiology Program, Department of Fisheries and Wildlife and Center, Utah State University,Logan, UT 84322-5210

JOHN A. BISSONETTE, National Biological Survey,Utah Cooperative Fish and WildlifeResearch Unit and Conservation Biology Program, Department of Fisheries and Wildlife, Utah State University,Logan, UT 84322-5210

RAYMOND D. DUESER, ConservationBiology Program, Department of Fisheries and Wildlife, Utah State University,Logan, UT 84322-5210

THOMAS C. EDWARDS, JR., National Biological Survey, Utah Cooperative Fish and Wildlife Research Unit and ConservationBiology Program, Department of Fisheries and Wildlife,Utah State University,Logan, UT 84322-5210

CHRISTOPHER M. LUECKE, ConservationBiology Program, Department of Fisheries and Wild- life, Utah State University,Logan, UT 84322-5210

MARK E. RITCHIE, ConservationBiology Program, Department of Fisheries and Wildlife, Utah State University,Logan, UT 84322-5210

JENNIFER B. SLADE, Department of Fisheries and Wildlife, Utah State University,Logan, UT 84322-5210

FREDERIC H. WAGNER, ConservationBiology Program, Department of Fisheries and Wildlife and Ecology Center, Utah State University,Logan, UT 84322-5210 Key words: endangeredspecies, recovery plans, small populations

Based on surveyresponses from state agen- ever,we wishto cautionthat the elimination cies, Hayes (1991) examinedwhich mamma- of the initialcauses of endangerment,such as lian orderswere mostvulnerable to becoming habitatdestruction, may not remove the threat endangered,threatened, or ofspecial concern, to populationsand may notbe theprime con- and the perceivedcauses. Habitatdestruction cernin developingrecovery plans. was listedmost frequently as the cause of a Wildlifebiologists should work to preserve species' vulnerability,whereas overexploita- and restorehabitat for plants and animals,as- tionwas onlyconsidered important for some sumingthat many populationsof plantsand Carnivoraand Artiodactyla.Reviews of ex- animalsare threatenedas a resultof tinctionsthrough history have documented the lossand degradation.Unfortunately, although overridingsignificance of habitatdestruction the preservationof suitablehabitat is neces- (Hester 1967, Diamond 1984). Therefore, sary,it may no longerbe sufficientto ensure wildlifebiologists and conservationistshave the recoveryof smallpopulations because the long consideredhabitat destruction to be the appropriationof habitatand "lettingnature criticalinitial threat to plantand animalpop- takeits course" (i.e., managementby "benign ulations(Terborgh and Winter1980). How- neglect")may no longerbe sufficient;rather,

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions 308 Wildl. Soc. Bull. 22(2) 1994

innovativeand activemanagement may be endangeredspecies and their or re- needed.Especially when considering habitat coveryin NorthAmerica since European col- lossand degradation,we shouldinclude hu- onizationin the sixteenthcentury. Hester's man-inducedfactors, such as climatechange conclusionsparalleled those of Hayes' (1991) (e.g.,global warming) and habitatfragmen- survey of state agencies: has tation,as well as theelimination (e.g., gray been the major cause of endangerment,and wolf,Canis lupus, in theGreater Yellowstone overharvesting has been a secondarycause. )and introductionof species(e.g., However,we examinedthe traitsof endan- rainbowtrout, Oncorhynchus mykiss, into cut- geredspecies (and several subspecies, e.g., heath throattrout, 0. clarki,streams and lakes). hen [Tympanuchuscupido]) thathave either We and others(e.g., Ehrlich and Ehrlich survived(n = 34) to the present(still endan- 1981)fear that managers and agenciesmight geredor recovered) versus those that have gone become complacent,operating under the extinct(n = 34), as reportedby Hester(1967). premisethat the protection of a pieceof hab- Thesedata are crude,but simple statistical tests itat ensurespopulation persistence. Hayes' based on binomialestimates of the 95% con- (1991) surveysupports our concern,because fidenceintervals for percentages illustrate in- stateagencies never attributed population de- terestingcontrasts between the 2 groupsof spe- clinesto the special demography ofsmall pop- cies. Species names used as examplesfollow ulationsand onlyonce attributed declines to Hester's(1967) listingfor easy comparison. thespecial of smallpop- 1. ulations. Endangeredspecies that have survivedare representedby a greaterproportion of spe- Confusionbetween the initial causes of en- cies withlow reproductiverates (i.e., long dangermentand concernsfor the persistence generationtime, or small clutch or litter orrecovery of precarious populations was cau- size) thanthose that went extinct: 29% (15- tionedagainst by Ehrlich and Ehrlich(1981), 49%) versus3% (0-7%). For example,the becausethe causes for reduced populations may Haitianhutia (Plagiodontia hylaeum) with notbe thesame as thecauses for further de- its low reproductiverate survived,while clinesand possible extinction. Although recov- otherCaribbean island hutias with higher erycannot proceed without curtailing habitat reproductiverates went extinct (Geocapro- lossand degradation,once a populationbe- mys columbianus,G. ingrahami,Hexolo- comessmall its existence is threatenedeven bodon phenax, Isolobodon portoricensis, afterthe original causes of endangerment are and I. levir).Species with low reproductive controlled.The recovery or persistence ofsmall ratesoften are assumedto be morevulner- populationscannot occur without considering able to extinction,but thedata do thespecial demographic and genetictraits of notsup- port this. The observed smallpopulations. These considerations arethe opposite pattern mightbe explained if essenceof population viability analysis (Shaffer high reproductive ratescorrelate with high mortality rates (low 1987), interagencycooperation in manage- life then small ment(Salwasser et al. 1987),reserve design expectancy), populations (Diamond1986), and captivebreeding pro- mightbe moreprecarious, even though high grams(Thorne and Belitsky1989). reproductiverates permit populations to re- coverquickly (Belovsky 1987, Pimm 1991). Furthermore,high reproductive rates may HISTORICAL INSIGHTS produce populationsthat fluctuatemore Hester(1967) reviewedthe historicaland over time and periodicallyapproach very scientificliterature on mammalianand avian low numbers(Pimm 1991).

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions IN MY OPINION ... * Belovskyet al. 309

2. Endangeredspecies that have survivedand theassociation between extinction and small thosethat have gone extinctdid not differ areais verystrong. Many of the extinct species in subjectiveestimates of theirpopulation (<30%) reportedby Hester(1967) were nat- densities(number/area) at the timeof Eu- urallyconfined to smallareas, especially is- ropeancontact: 44% (26-63%) versus41% lands,and consequentlymay have had small (23-60%) had low densities.For example, populations,but manybecame restricted to thesea mink(Mustela macrodon)a species smallareas that could only support small pop- witha high densitywent extinct, whereas ulationsthrough habitat destruction by hu- the marten(Martes americana) a species mans(e.g., the heath hen). Habitat destruction witha low densitysurvived. Low popula- canoccur by either a largeloss of habitat area tion densityoften is assumed to make a or by changingthe spatial distribution ofthe speciesvulnerable to extinction due to Allee habitat,i.e., fragmentation. In either case, small Effects(e.g., difficultiesin findingmates), populationsoften are created. butno evidencefor this emerged; however, Thenumber of species becoming threatened thismight not be expectedfor several rea- orextinct has increased over time (Fig. 1). As sons.First, may not be in thepast, the initial threats to the criticalmeans of assessinga smallpop- are largelycaused by habitat destruction cre- ulationbecause a populationcan be large atingsmall populations, especially as species' evenif it is at a low densitywhen it inhabits originalranges become restricted tosubregions a large area. Second, it is possiblethat a (e.g.,the grizzly bear, Ursus arctos, in the con- numberof populations, even at low density, tinentalU.S.). However,Hester's (1967) his- thatare spreadover a wide area can create toricalreview provides a glimmerof hope for a "rescue-effect"from extinction by mi- thissituation. Sixty-eight species of birds and grantsmoving between the populations (see mammalsin North America have been threat- below). enedwith extinction since the sixteenth cen- 3. Endangeredspecies that went extinct dif- tury,and 50% of them have gone extinct. How- feredfrom surviving species in havingpop- ever, in the 14 cases wherepeople have ulationsrestricted to smallareas: 74% (60- attemptedto recoveran endangeredspecies, 88%) versus35% (20-56%). This is illus- only1 species(7%) hasgone extinct, but most tratedby thelarge number of islanddwell- havenot recovered to thepoint that they are ing speciesthat have gone extinct.The ex- notstill considered threatened over their entire tentof a species'habitat may have been the previousrange. Therefore, wildlife and fish- principaldeterminant of its vulnerability to eries managersconcerned with protecting extinction,because even high population threatenedspecies should be hopeful,but suc- densitymay not be associatedwith large cesswill depend on betterunderstanding the populationsize when the populationis re- specialdemographic and genetic traits of small strictedto a smallarea, and smallpopula- populations. tionsize predisposesa populationto extinc- tion(see below). A FUTURE WITH SMALL POPULATIONS The above contrastsare no doubtconfound- ed by body size,, and lifehistory In the past,biologists or conservationists attributesof the species (e.g., large animals and wishingto protecta specieswould argue for carnivorestend to have lower densities and will habitatprotection assuming that success in this be affectedby small area more than smaller activityand protection ofthe species from fur- animalsor , see below).Nonetheless, therdeleterious factors such as harvesting

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions 310 Wildl. Soc. Bull. 22(2) 1994

25 Ol CARIBBEAN w ISLANDSAFTER MAINLAND WJ SPANISH P_ ( z 20 COLONIZATION 0 w 0 c- 0 10 co_a.

0

CENTURY Fig. 1. The chronologyof human-caused endangerment of birdand mammalspecies in NorthAmerica since Europeancolonization (from Hester 1967). After an initialpulse of on Caribbeanislands after Spanish colonization,the number of endangeredspecies has increasedrapidly on theNorth American mainland.

would ensurepersistence of the species. Al- protectionof theselands is alwayslimited and thoughsetting aside habitatis theonly feasible land prices have increaseddramatically; (3) optionfor protecting ecological communities human demands on lands already protected and theirbiodiversity, this alone may no longer have increasedbecause of increasinghuman be sufficientto ensure the long-termpersis- populationsand theirresource demands; (4) tenceof the componentspecies' populations. protectedhabitats may no longerbe refuges Habitat protectionas the sole action now as exoticspecies are introduced(e.g., the zebra may be insufficientbecause: (1) the amount mussel,Dreissena polymorpha [Roberts 1990]), and distributionof undisturbedhabitat avail- species importantto the functioningof eco- able forprotection has been affected by human logicalcommunities are eliminated(e.g., gray activities;(2) publicfunds for the purchase and wolvesin Yellowstoneand theirpredation on

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions "IN MY OPINION... * Belovskyet al. 311

elk,Cervus elaphus), and globalenvironmen- habitatislands often decreases as greaterhu- tal changes modifypreserved (e.g., man disturbanceeither increases distances be- globalwarming); and (5) naturalsuccession in tweenhabitat islands or makesthe ocean sep- ecologicalcommunities changes habitat suit- aratingthe habitat islands less hospitable. These ability. theoreticalpredictions of extinctionvulnera- Simplyidentifying the patchwork of habitat bilityhave been empiricallysupported (e.g., fragmentshas becomea majoreffort (e.g., Gap bird populationson islands[Diamond 1984, Analysis[Scott et al. 1993]), but this action Pimmet al. 1988];mammals on mountaintops cannotguarantee any level of habitatprotec- [Brown1971]; fish in lakes[Magnuson 1991]). tion,nor can it guaranteeprotection of suffi- From a managementperspective, the iso- cientlylarge areas to ensurethe persistence of lationof unitsof preservedhabitat (fragmen- populations.Therefore, without the unlikely tation)has been demonstratedto cause local eventof settingaside largertracts of habitat or regionalextinction of populations(Harris thanwe alreadyhave (see Mannand Plummer 1984, Lovejoy et al. 1986). Newmark(1987) 1993),we maybe facedwith increasing num- reportedthat western U.S. nationalparks may bers of small populationsof many plant and be too small to supportsome mammalpopu- animal speciesthat are vulnerableto the de- lationsafter the populations were isolated and mographicand geneticproblems that reduce preventedfrom using previously suitable hab- persistence. itatsurrounding the parks.The persistenceof graywolves at Isle RoyaleNational Park, where a largeisland of pristinehabitat was set aside, CONCERNFOR SMALL POPULATIONS preywere sufficient, and wolveswere at max- Early wildlifemanagers and ecologistshad imum density(no. /area) just a decade ago, concernsabout "edge-effects" and "edge- and wouldnot seem to be a concern,but wolf num- core-sensitivespecies" (Leopold 1933) thatre- bersin thissmall population have declined and late to problemscreated by habitatloss and theymay become extinct (Rolf Peterson, Mich- fragmentation,as well as the large areas of igan TechnologicalUniv., Houghton,Mich., habitat required by some species (Leopold pers. commun.,1993). Plantsand animalsin 1949, Shelford1963). But the increasingpo- U.S. nationalparks have receivedmore pro- tentialfor the creationof small populations tectionthan most, but setting aside habitat may and how thisreduces population persistence notbe sufficientif the area needed forpopu- firstwas addressedformally in MacArthurand lation persistenceis not encompassedin the Wilson's(1967) pioneeringwork on islandbio- parkand thecontinuing habitat loss and frag- geography.They realized that habitatfrag- mentationsurrounding the park are not con- mentswere "islands" surrounded by an "ocean" sidered(Newmark 1987). ofinhospitable habitat; as islandsize decreases, a population'ssize diminishesbecause habitat THE SPECIAL BIOLOGY OF SMALL availabilitydeclines and the population'svul- POPULATIONS nerabilityto extinctionincreases. Further- Conservationbiologists provide 2 reasonsfor more,a population'svulnerability to extinction concernfor small populations when designing increasesas the distancebetween islands in- managementplans to fostertheir persistence creasesbecause individualsfrom one popula- and recovery(Gilpin and Soule 1986, Soule tion will be less likelyto immigrateto a de- 1987). clining populationand help rescue it from extinction(Brown and Kodric-Brown1977). 1. Genetic considerations dealing with The abilityof individualsto move between "foundereffects", ,and in-

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions 312 Wildl. Soc. Bull. 22(2) 1994

breedingdepression can reducegenetic 1, p = 0.99; at t = 10,p = 0.90; at t = 100, variabilityand increasethe expressionof p = 0.36). The probabilisticnature of pop- deleterioustraits. These factors reduce sur- ulationpersistence and theincreasing prob- vivaland reproduction as well as long-term abilityof extinctionas populationsize de- evolutionarypotential and adaptabilityto creasesare oftenconfused; some examples changingenvironmental conditions. of thisdifficulty are presentedbelow. 2. Demographicconsiderations emerge from randomvariations in birth and death rates. The geneticand demographiceffects are Theserates are averages of probabilistic oc- magnifiedby environmental stochasticity (e.g., currencesfor the individuals composing the annual variationin food, cover, etc.), cata- populationeven in a constantenvironment strophic events (e.g., epidemics,floods, etc.), (e.g.,probability that a femalewill give birth and interactionswith other species (e.g., com- to 0, 1, 2 ... offspringin a timeperiod; petitors,predators, diseases, etc.) (Soule 1987, probabilityofdying in a timeperiod [Raup Lande 1988).The morea population'scarrying 1991]).Whereas these probabilities produce capacityand birthand death ratesvary with theexpected birth and deathrates given environmentalfluctuations over time (Rough- sufficienttime or averagingover many in- garden 1979, Goodman 1987), the more fre- dividuals,the smaller a population ata point quent and severethe occurrenceof catastro- intime the greater the likelihood that it will phes (Ewens et al. 1987); the moresevere the exhibitbirth and death rates very different interactions with other species (MacArthur fromexpected values (e.g., the probability 1972), the more vulnerableis the small pop- ofgetting all headsin 2 cointosses is 0.52, ulation. butthe probability of gettingall headsin The above factorsare not the initialcause 5 tossesis 0.55). Consequently,small pop- fora population'ssmall size, but theymake it ulationsare morevulnerable to extinctionvulnerable to extinctionbecause population becausethey are more likely to experience declines easily can continueafter the original a "runof bad luck"(a seriesof periods with cause forsmall populationsize is eliminated lowbirth rates and highdeath rates). Fur- (i.e., the extinctionvortex discussed by Gilpin thermore,as the population becomes small- and Soule [1986]).Scientists need to studythe er,the likelihood of a runof bad luckin- dynamicsof small populations and providethis creases;this has led Gilpin and Soule (1986) informationtomanagers who can improvetheir to referto thisprocess as a "vortex". strategicand tacticalplanning for species pro- Givensufficient time the probability that tection. a finitepopulation will persist, not experi- encea runof bad luckthat leads to extinc- MANAGINGSMALL POPULATIONS tion,is infinitelysmall; this is analogousto theeventual loss of a completebankroll by A betterunderstanding of the theoretical a gambler("gambler ruin" analogy sensu concepts for small populationswill enable [Raup1991]). For example, assuming a con- managersto develop betterplans forspecies stantprobability ofpersistence/time period recoveryand persistence.These conceptspro- (s), whichignores the decreasing probabil- vide criticalinsights into managementques- ityas thepopulation becomes smaller dur- tions,such as: (1) how muchtotal habitat must inga runof bad luck,the cumulative prob- be provided?(2) how should the mosaic of abilityof persisting (p) rapidlyapproaches habitat units be arranged?and (3) willhabitat zeroas thenumber of timeperiods (t) in- protectionbe sufficientand, if not,what other creases(e.g., s = 0.99and p = 0.99t:at t = actionsare necessary?Failure to consider these

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions IN MY OPINION . . . * Belovskyet al. 313

factorswill doom many preservation plans to differfrom these estimates based on its unique failure. traits. Ideally,the propermanagement of small If theabove habitat guidelines cannot be populationsrequires biologists to haveknowl- satisfied,the populationcannot be benignly edgeof a population'scarrying capacity, birth neglected.Rather, active management must be anddeath rates, and other parameters such as: performedto makethe available habitat ca- (1) variabilityof birth and death rates caused pableof supporting a sufficiently large popu- bystochastic differences between individuals; lation or to assist small population persistence. (2) temporalvariability in carryingcapacity Anexample is providedby the recent demise and birthand deathrates caused by environ-of the remainingwild black-footedferret mentalfluctuations (e.g., annual changes); (3) (Mustelanigripes) population caused by ca- frequencyand severity of catastrophes; and (4) ninedistemper (Thorne and Belitsky1989). intensityofinteractions with other species that The diseaseis notdensity-dependent, but a affectbirth and deathrates (e.g., competitor- catastrophe that would have little impact on a induceddeclines in birthrates and survival). largerpopulation where some individuals may Byconsidering these factors (this does not nec- be immune,develop immunity, or notbe ex- essarilymean absolute measurement) manag- posed(Dobson and May 1986).The studyof erscan beginto askwhether enough and the smallpopulations tells us to expect this type of correctmix of habitats have been provided. If vulnerabilityand to considerthese situations insufficienthabitat has been provided and more in ourplanning. is unavailable,managers must enhance the Somewildlife biologists already are empir- chancesfor a species'persistence in theavail- icallyconsidering the minimum ablehabitat (e.g., increasing , neededfor persistence; while this is encour- predatorcontrol, supplementing populations, aging, there also is misunderstanding.Forex- etc.). ample,Berger (1990) foundthat desert big- Rudimentaryguidelines are available for as- hornsheep (Ovis canadensis) populations of sessingwhether sufficient habitat area has been 50 or lesswent extinct, while those of 100 or providedor is evenavailable (Belovsky 1987). morepersisted for 70 years.These results were Furthermore,these guidelines for managers questionedby Krausman et al. (1993),because havebeen validated using data for the extinc- theyknow of populations of 50 or lessin Ar- tionsof mammalson GreatBasin mountain- izona that have persisted.Berger (1993) re- topssince the end of the Pleistocene (Belovsky spondedby askingwhether these data were 1987).These guidelines include: (1) largerar- statisticallyvalid. Neither statistics nor Ber- eas are requiredto supportlarger species; (2) ger's(1990) inability to includeArizona pop- carnivoresneed larger areas than herbivores; ulation data may be theissue, rather the debate (3) largerareas are requiredin tropicalthan maybe mootbecause the dynamics of small temperateregions; (4) largerareas are required populationsinform us thatthere is nevera in foreststhan grasslands; and (5) largerareas guaranteeof persistence. Some populations less are requiredin morevariable environments than 50 willpersist and others that are greater thanless variableenvironments. While esti- than100 will disappear. The managementis- matesof required habitat area and population sueis how the probability ofpersistence changes sizeneeded by a speciesof a givenbody size withpopulation size and whatprobability of in eachof the above categories to persist for a persistenceis acceptable to society. Therefore, giventime (e.g., 100 years) with a givenprob- managerswill have to deal withuncertainty, ability(e.g., 95%) can be computed,a specific evenwhen sound strategies are adopted. species'habitat area and populationsize can How toemploy the dynamics of small pop-

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions 314 Wildl. Soc. Bull. 22(2) 1994

ulationsin management becomes even less clear variationin birthand deathrates over time when existingtheoretical models are used. (i.e., environmentalvariability). Environ- These modelsare simplisticand have notbeen mental variabilityenhances demographic validated(Boyce 1992, Stacey and Taper 1992). variabilityand is moreimportant, reducing This was a criticaldebate in developingthe persistencetimes by an orderof magnitude recoveryplan for the Northernspotted owl ormore than when demographic variability (Strixoccidentalis, Murphy and Noon 1992). onlyis considered(Boyce 1992,Stacey and Nonetheless,the theoretical models often pro- Taper 1992). Beier (1993) does notinclude vide resultsconsistent with observations (Be- theenvironmental variability referred to in lovsky1987, Thomas 1990) or withpopulation theoreticalmodels; therefore, we wouldex- sizes that have evoked concernsover persis- pecthis persistence times to be muchlonger tencefrom managers (Wilcove et al. 1993). thanthose from the theoretical models. Fur- Theoreticalmodels, even thoughthey are thermore,regardless of the rescue effects of simplisticand have notbeen validated,do pro- migrants,the 15-20 adult mountainlions vide certainrobust rules for the management predictedto guaranteepersistence also is ofsmall populations. If a morecomplex model expected,because this is thevalue predicted ormanagement plan violates one ofthese rules, by theoreticalmodels when environmental the complexmodel's predictions or the man- variabilityis not incorporated.Perhaps a agementplan must be reevaluated.For ex- complexmodel was unnecessaryand the ample,an innovativemanagement proposal to managementconclusions are toooptimistic. balance economic developmentand persis- tenceof a mountainlion (Felis concolor)pop- The theoreticalmodels also indicate the types ulationin a regionof Californiawas based on of data thatwill be needed to computeper- populationpersistence times (Beier 1993). Per- sistencetimes so thatsound management strat- sistencetimes were computed using a complex egies can be developed.For example,to esti- simulationmodel because existingtheoretical matedemographic variability in themountain modelswere thoughtto be too simplistic,es- lion case, observationsof a sufficientnumber peciallygiven their density-independent con- of individualsin a given year are needed to struction,and toprovide persistence times that estimatethe probabilityof survivaland prob- were too pessimistic(short). However, two of abilitiesof differentlevels of female repro- thesimulation model's predictions violated the ductive output. To estimateenvironmental rulesestablished by theoreticalmodels: variability,observations of a sufficientnumber ofindividuals over a sufficientnumber of years 1. The theoreticalmodels are called density- are needed to estimatethe variationin birth independent,because populationgrowth and death ratesbetween years. abruptlystops when carryingcapacity is Small populationsare an importantsubject attained,rather than birth and death rates fortoday's wildlife and fisheriesmanagers. This changing gradually as density increases is stilla new area of study;it will strainthe (Boyce 1992, Stacey and Taper 1992). adequacy of our knowledgeof each species, Therefore,Beier's (1993) claim thatthese modifyhow we studypopulations, and force models frequentlyproduce populations biologiststo expandtheir perspectives on man- largerthan carrying capacity is notpossible. agement.Considering these problemswhen 2. Theoreticalmodels employ two formsof developingmanagement plans may not re- variationin birthand death rates:random quire costly monitoringprograms by state variationin birthsand deathsbetween in- agencies.Rather, existing information on small dividuals(i.e., demographic variability); and populationscan be incorporatedinto recovery

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions IN MY OPINION . .. .* Belovskyet al. 315

plans,and continued research advances can be malianpersistence. Pages 35-58 in M. E. Soule, ed. Viable populationsfor conservation.Cam- usedto updateplans. This will require better bridgeUniv. Press, Cambridge, U.K. communicationbetween biologists working on BERGER, J. 1990. Persistenceof different-sizedpop- smallpopulations and thosedeveloping and ulations:an empiricalassessment of rapidextinc- tionin bighornsheep. Conserv. Biol. 4:91-98. implementingrecovery plans. . 1993. Persistenceof mountainsheep: meth- ods and statistics.Conserv. Biol. 7:219-220. IMPLICATIONS BOLEN, E. G. 1989. Conservationbiology, wildlife management,and spaceship earth. Wildl. Soc. Bull. Wildlifeand fisheriesmanagers no doubt 17:351-354. BOYCE, M. S. 1992. Populationviability analysis. Annu. willdevote more of their time to dealing with Rev. Ecol. Syst.23:481-506. smallpopulations and therecovery plans for BROWN, J. 1971. Mammalson mountaintops:non- threatenedspecies because the threat to species equilibriuminsular . Am. Nat. 105: throughoutthe worldis increasing(Wilson 467-478. - AND A. KODRIC-BROWN. 1977. Turnoverrates 1989).A hands-offpreservationist approach to in insularbiogeography: effect of immigration on themaintenance of biologicaldiversity may extinction.Ecology 58:445-449. no longerbe realistic,rather the maintenance DIAMOND, J. M. 1984. Historicextinctions: a Rosetta Stonefor understanding prehistoric extinctions. ofprecarious species' populations may require Pages 824-862in P. S. Martinand R. G. Klein, newand moreintensive management activi- eds. Quaternaryextinctions: a prehistoric revolu- ties.Understanding the demography and pop- tion.Univ. Arizona Press, Tucson. . 1986. The designof a naturereserve system ulationgenetics of small populations as studied forIndonesian New Guinea.Pages 485-503 in M. by conservationbiologists will provide some E. Soule,ed. Conservationbiology: the science of insightsfor this management. com- scarcityand diversity. Sinauer Assoc., Sunderland, However, Mass. municationshave been strained between more DOBSON,A. P., ANDR. M. MAY. 1986. Diseaseand traditionalwildlife and fisheries biologists and conservation.Pages 345-366 in M. E. Soule,ed. conservationbiologists, and cooperationhas Conservationbiology: the science of scarcityand diversity.Sinauer Assoc., Sunderland, Mass. been jeopardizedby the creationof profes- EDWARDS, T. C., JR. 1989. The WildlifeSociety and sionaldichotomies based upon different phi- theSociety for Conservation Biology: strange but losophies(consumptive versus nonconsump- unwillingbedfellows. Wildl. Soc. Bull. 17:340-343. EHRLICH, P., AND A. EHRLICH. 1981. Extinction-the tive)and levelsof investigation(population causesand consequencesof thedisappearance of versuscommunity versus ) species.Random House, New York,N.Y. 384pp. (Templeet al. 1988,Anonymous 1989, Bolen EWENS,W. J.,P. J.BROCKWELL, J. M. GANI,AND S. I. RESNICK. 1987. Minimumviable population size 1989,Edwards 1989, Gavin 1989, Hunter 1989, in thepresence of catastrophes. Pages 59-68 in M. Teer 1989,Wagner 1989). Nonetheless, it is E. Soule,ed. Viablepopulations for conservation. logicalthat these two groups should join forces CambridgeUniv. Press, Cambridge, U.K. GAVIN, T. A. 1989. What'swrong with the questions to solvethis pressing set of problems. we ask in wildliferesearch? Wildl. Soc. Bull. 17: Acknowledgments.-We appreciatere- 345-350. viewsof this manuscript by J. P. Hayes,W. C. GILPIN, M. E., AND M. E. SOULE. 1986. Minimum viablepopulations: processes of species extinction. McComb,R. Noss,and R. J.Steidl. Pages 345-366in M. E. Soule,ed. Conservation biology:the science of scarcityand diversity.Si- nauerAssoc., Sunderland, Mass. LITERATURE CITED GOODMAN, D. 1987. The demographyof chanceex- tinction.Pages 11-34 in M. E. Soule,ed. Viable ANONYMOUS. 1989. The futureof wildliferesources populationsfor conservation.Cambridge Univ. cannotafford strange or unwillingbedfellows. Press,Cambridge, U.K. Wildl.Soc. Bull. 17:343-344. HARRIS, L. D. 1984. The fragmentedforest: island BEIER, P. 1993. Determiningminimum habitat areas biogeographytheory and the preservation ofbiotic and habitatcorridors for cougars. Conserv. Biol. diversity.Univ. Chicago Press, Chicago, Ill. 211pp. 7:94-108. HAYES, J. P. 1991. How mammalsbecome endan- BELOVSKY,G. E. 1987. Extinctionmodels and mam- gered.Wildl. Soc. Bull. 19:210-215.

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions 316 Wildl. Soc. Bull. 22(2) 1994

HESTER, J. J. 1967. The agentof manin animalex- managingfor viable populations. Pages 159-174 tinctions.Pages 169-192in P. S. Martinand H. in M. E. Soule,ed. Viablepopulations for conser- E. Wright,Jr., eds. Pleistoceneextinctions: the vation.Cambridge Univ. Press, Cambridge, U.K. searchfor a cause.Yale Univ.Press, New Haven, SCOTT,M. J.,F. DAVIS,B. CSUTI,R. Noss,B. BUTTER- Conn. FIELD, C. GROVES, H. ANDERSON, S. CAICCO, F. HUNTER, M. L., JR. 1989. Aardvarksand Arcadia: two DYERCHIA, T. C. EDWARDS, JR., J. ULLIMAN, AND R. principlesof wildliferesearch. Wildl. Soc. Bull. G. WRIGHT. 1993. Gap analysis:a geographic 17:350-351. approachto protectionof biologicaldiversity. KRAUSMAN, P. R., R. C. ETCHBERGER, AND R. M. LEE. Wildl.Monogr. 123. 41pp. 1993. Persistenceof mountainsheep. Conserv. SHAFFER,M. 1987. Minimumviable populations, cop- Biol.7:219. ing withuncertainty. Pages 69-86 in M. E. Soule, LANDE, R. 1988. Geneticsand demographyin bio- ed. Viable populations for conservation.Cam- logicalconservation. Science 241:1,455-1,460. bridgeUniv. Press, Cambridge, U.K. LEOPOLD, A. 1933. Gamemanagement. C. Scribner's SHELFORD, V. E. 1963. Theecology of North America. Sons,New York,N.Y. 481pp. Univ.Ill. Press,Urbana, Ill. 610pp. . 1949. A sand countyalmanac. C. W. SOULE, M. E. 1987. Introduction.Minimum viable Schwartz,Oxford Univ. Press,New York,N.Y. populations,coping with uncertainty. Pages 1-10 246pp. in M. E. Soule,ed. Viablepopulations for conser- LOVEJOY, T. E., R. 0. BIERREGAARD, JR., A. B. RYLANDS, vation.Cambridge Univ. Press, Cambridge, U.K. J.R. MALCOLM, C. E. QUINTELA, L. H. HARPER,K. STACEY, P. B., AND M. TAPER. 1992. Environmental S. BROWN, JR., A. H. POWELL, G. V. N. POWELL, variationand thepersistence of small populations. H. 0. R. SCHUBART, AND M. B. HAYS. 1986. Edge Ecol. Appl.2:18-29. and othereffects of isolationon Amazonforest TEER, J.G. 1989. ConservationBiology-a bookre- fragments.Pages 345-366 in M. E. Soule,ed. Con- view.Wildl. Soc. Bull.17:337-339. servationbiology: the science of scarcityand di- TEMPLE, S. A.,E. G. BOLEN, M. E. SOULE, P. F. BRUSSARD, versity.Sinauer Assoc., Sunderland, Mass. H. SALWASSER,AND J.G. TEER. 1988. What'sso MAcARTHUR, R. H. 1972. Geographicalecology. newabout conservation biology? Trans. North Am. Harperand Row,New York,N.Y. 269pp. Wildl.and Nat. Resour.Conf. 53:609-612. , AND E. 0. WILSON. 1967. Thetheory of island TERBORGH,J., AND B. WINTER. 1980. Somecauses of biogeography.Princeton Univ. Press,Princeton, extinction.Pages 119-134in M. E. Souleand B. N.J.203pp. A. Wilcox,eds. Conservationbiology: an evolu- MAGNUSON,J. J. 1991. Fishand fisheries ecology. Ecol. tionary-ecologicalperspective. Sinauer Assoc., Appl.1: 13-26. Sunderland,Mass. MANN, C. C., AND M. L. PLUMMER. 1993. The high THOMAS, C. D. 1990. Whatdo real populationdy- costof biodiversity.Science 260:1,868-1,871. namicstell us aboutminimum viable population MURPHY, D. D., AND B. R. NOON. 1992. Integrating sizes? Conserv. Biol. 4:324-327. scientificmethods with habitat conservation plan- THORNE, E. T., AND D. W. BELITSKY. 1989. Captive ning:reserve design for Northern Spotted Owls. propagationand the current status of free-ranging Ecol. Appl.2:3-17. black-footedferrets in Wyoming.Pages 223-234 NEWMARK, W. D. 1987. A land-bridgeisland per- in U. S. Seal, E. T. Thorne,M. A. Bogan,and S. spectiveon mammalianextinctions in western H. Anderson,eds. Conservationbiology and the NorthAmerican parks. Nature 325:430-432. black-footedferret. Yale Univ.Press, New Haven, PiMM, S. L. 1991. The balanceof nature? Ecological Conn. issuesin theconservation of speciesand commu- WAGNER,F. H. 1989. Americanwildlife management nities.Univ. Chicago Press, Chicago, Ill. 434pp. at thecrossroads. Wildl. Soc. Bull.17:354-360. , L. JONES,AND J. M. DIAMOND. 1988. On the WILCOVE, D. S., M. MCMILLAN, AND K. C. WINSTON. riskof extinction.Am. Nat. 132:757-785. 1993. What exactlyis an endangeredspecies? An RAUP, D. M. 1991. Extinction:bad genesor bad luck? analysisof the U.S. EndangeredSpecies List: 1985- W. W. Nortonand Co., New York,N.Y. 210pp. 1991. Conserv. Biol. 7:87-93. ROBERTS, L. 1990. Zebra musselinvasion threatens WILSON, E. 0. 1989. Threatsto biodiversity. Sci. Amer. U.S. waters.Science 249:1,370-1,372. 261:108-116. ROUGHGARDEN,J. 1979. Theoryof population genet- ics and evolutionaryecology: an introduction.Received 11 August1992. MacmillanPubl. Co., New York,N.Y. 634pp. Accepted22 November1993. SALWASSER, H., C. SCHONEWALD-COX, AND R. BAKER. AssociateEditor: McComb. 1987. The role of interagencycooperation in

This content downloaded from 66.254.247.100 on Wed, 24 Feb 2016 16:01:40 UTC All use subject to JSTOR Terms and Conditions