The Extent, Cost and Control of Livestock Predation by Eagles with a Case Study on Black Eagles (Aquila Verreauxii) in the Karoo

j. RaptorRes. 33 (1) :67-72 ¸ 1999 The Raptor ResearchFoundation, Inc.

THE EXTENT, COST AND CONTROL OF LIVESTOCK PREDATION BY EAGLES WITH A CASE STUDY ON BLACK EAGLES (AQUILA VERREAUXII) IN THE KAROO

ROBERT A.G. DAVIES Mammal ResearchInstitute, University of Pretoria,Pretoria 0001, SouthAfrica

Intense persecutionof predators in the sheep- Bruns 1970, Skogland1974, Svendsen1980, Steyn farming Karoo region of South Africa has been 1982, O'Gara et al. 1983, Bergo 1987). Large ea- suggestedas the reasonfor irruptions of rock hy- gles are, therefore, capable of killing domestic rax (Procaviacapensis) which have caused signifi- sheep up to about half adult size, so the issueap- cant damage to vegetation resourcesin the past pears to be not whether they kill domesticlambs, (Thomas 1946, Kolbe 1967, Rubidge in Kolbe but to what extent does this occur and under what 1983). I carried out a 5-yrfield studyin and around circumstances. the Karoo National Park (KRNP) near Beaufort Most ranchers maintain that only small lambs Westto assessthe costsand benefitsof BlackEagles (<10-d old) are vulnerable to eagle attack (Arnold (Aquilaverreauxii) to Karoo farmers (Davies1994). 1954, Wiley and Bolen 1971, Palmer 1983). It is This paper summarizesthe harm that Black Eagles immature eaglesthat most often become involved may cause on farmland, and this is discussedin in livestockdepredations (Foster and Crisler 1979, relation to livestockpredation by eaglesworldwide, O'Gara 1978). The Golden Eagle is the chief of- and it appraisespossible management solutions for fender in the northern hemisphere. Only three problem situations. speciesof Haliaeetushave been recorded killing Predators can cause major lossesto livestock lambsin North America, Europe and Australia,but ranching operations (O'Gara et al. 1983). Preda- there is little evidence to support these accusations. tors may only remove a small percentageof sheep In Australia, lamb-killing by Wedge-tailedEagles flocks,but these fractionsmay amount to great fi- has been confirmed (Brooker and Ridpath 1980) nancial losswhen applied to nationwidelivestock and, in Africa Martial Eagles (plains), Black Eagles numbers (U.S. Fish and Wildlife Service 1978, Tex- (mountains) and CrownedEagles (forests) are reg- as Crop and LivestockReporting Service 1979, ularly accusedof killing lambs, with some confir- Lawson1989). Bountysystems, that resultedin the mation that they do. There is little overlapbetween destruction of millions of predators, have been in- the rangesof large, forest-dwellingeagles and the voked to avert such losses,but they have been major sheepfarming regionsof the world. largely ineffectual, and most have now been dis- Domesticlamb remains comprised 1.1% of the continued (Hey 1959). Large eagleshave not been 3586 prey items collected beneath Black Eagle exempt from bounty systemsand they have been nests in and around the KRNP, and, 1.6% of those persecutedin all the major sheep-farmingregions collected solely beneath farm nests. Domestic where they occur (Brown 1976). Kill rates of lambs comprised a similar proportion of 389 prey Wedge-tailed Eagles (Aquila audax) in Australia items delivered to nestsas monitored by time-lapse (Leopold and Wolfe 1970) were held to be the photography,and of sightingsof fresh prey on the highest for any large raptor worldwide (Brown nest. With an estimateof prey capture rate (173 1975), but they were much lower than kill ratesof prey/pair/yr), this indicatedthat a residentpair of Black and Martial (Polemaetusbellicosus) Eagles Black Eagles normally consumedthree lambs per claimed by farmers in parts of the Karoo in the year on Karoo farmland. In other regions of the 1960s (Siegfried1963). Cape, Black Eaglesmay consumemore domestic Has such persecutionbeen basedon any factual lambs since they comprised8% of prey remains evidenceof eagleskilling livestock?Golden (Aquila collected in the heavily-vegetatedEastern Cape chrysaetos)and Crowned (Stephanoaetuscoronatus) (Boshoff et al. 1991). Bolen (1975) observed that Eagleshave killed prey (including domesticsheep) livestockcomprised a greater portion of Golden weighing 20-50 kg (Lehti 1947, Cooper 1969, Eagle diet in areas of thick brush in Texas, where

67 68 EXPANDEDABSTRACTS VOL. 33, NO. 1 natural prey was highly abundant but presumably farmers spoken to reported no lamb lossesto ea- less vulnerable. While these collectionsof eagle gles;27% reported occasionallosses and 24% re- prey indicate that domestic livestockrepresent a ported significantlosses including a few eyewitness very small portion of large eagle diets,they should, accounts of kills. however,be interpreted cautiously.They only re- Twelve questionnairesurveys of ranchers (world- flect diet of breeding birds during the nestingsea- wide) indicated that most ranchers believed that son and we know that nonbreeding eagles cause predators removed nearly 7% of lambs born (9% more harm. The collectionsmay alsobe biasedbe- of lambs born were thought to die of other caus- causeremains of large animals like lambs tend to es)(Fig. 1). However,estimates of predation are of- decompose more slowly nor do they indicate ten exaggeratedin thesesurveys (Nesse et al. 1976, whether lambs delivered to nests were killed or Armentrout 1980, Boshoff 1980, Hewson 1981). scavenged(Matchett and O'Gara 1987). Lockie Field necropsies conducted on lamb carcasses (1964) showedthat only three of 10 assayablelamb found in lambing campsgave reliable data on caus- carcassesbrought to Golden Eagle nests were es of death and whether lambskilled by predators kdled by eagles. were dying from other causes(Rowley 1970, Wiley I observedresident Black Eagles for 55 d. During and Bolen 1971, Bowns et al. 1973, Brown 1976, this time, I did not see any attackson livestock,but Nesseet al. 1976, Tigner and Larson 1977, 1981, I observed the eagles scavenging on domestic O'Gara 1978, Wade and Bowns 1980). Massivesub- sheeptwice. During 80 d of observationon Merino cutaneoushemorrhaging surroundingirregularly- sheep as part of another study,I witnessedregular spaced talon punctures on the neck and upper attacks by an immature Martial Eagle on various back are prime indicators for eagle-killed lambs. natural prey, but the eagle ignored 40 vulnerable Carcassinspections are consideredto be the only Merino lambs which occasionallybedded directly realisticmethod for quantifyinglivestock predation beneath one of the eagle's roost trees. Birds, in- by eaglesand also provide good opportunitiesfor cluding raptors, are known to become 'imprinted' constructive interaction with ranchers. I only ex- or 'wedded' to particular food types if these are amined 23 carcasseson visitsto farmswhere eagles fed to the exclusionof all else during sensitivede- were allegedlykilling lambs.None of these lambs velopmental phasesin the nestling period (Wood- were found to have been killed eagles.I also ob- ford 1966, Rabinowitch 1969, Hess 1973, Immel- tained data from 44 studiesinvolving about 30 000 mann 1975). Infrequent predation on domestic necropsiesworldwide (Fig. 1). These results indi- lambs by eagles in the Karoo, despite high avail- cated that most ranching operations experience ability and vulnerability of this prey, may be ex- very low lossesto predation but at a few 'problem plained if most young eaglesare raised on a diet situation' ranchesvery high lossesprobably occur. of natural prey items. Greater incidence of eagle Necropsiesshow that predatorsremove only an av- predation on goat kids rather than sheep lambs erage of 4.9% of lambsborn as comparedto the has been attributed to poor attendanceby nannies averageof 13.7% lambs that die from nonviolent (Glover and Heugly 1970) and isolationin rough mortality factors. terrain (Nasset al. 1984), but goat kids bear a clos- Evidently ranching operationsin Australia and er resemblanceto juvenile antelopewhich are reg- especially South Africa and Scotland experience ularly killed by large eagles. relativelylow lossesof lambsto predators;whereas I interviewedfarmers during 30 visitsto farmsin North American operations experience relatively the central Karoo region, and some claimed ex- high losses(Fig. 2). This can be attributed to se- tremely high kill rates of eaglesin certain low-in- vere problems with coyotes(Canis latrans)in some come districts.Despite this, there was no evidence parts of the U.S., and to the fact that inviable lambs for a decline in BlackEagle numbersin the Karoo (already dying to other causes)were not distin- so there is probably a large 'floating population' guished in most North American estimatesof pre- of nonbreeders.The majority of farmers were predation. Certain investigationsprovided data on the pared to tolerateresident eagles until lamb killing relative involvement of different predators. It is was suspected.A minority of farmers took active clear that wild and domestic canids cause the most stepsto encourageeagles on their farms through damage to ranching operations.In South Africa, the provision of artificial nest sites and rehabilita- caracals(Felis caracal) cause significant losses, and tion of injured birds. Approximatelyhalf of the 37 in Australia feral pigs (Susscrofa) are significantly lVL•cH 1999 EXPANDED A•STP.•CTS 69

FREQUENCY HISTOGRAM FOR INVESTIGATIONS SHOWING PERCE NT LAMBSKILLED BY PREDATORS (as determinedbytwo methods)

40' 30 % investigations 2

10- mean=6,7%; median =5,3% 0 0 4 REPORTS

%lambs kille ,- 20:• .... NECROPSIESmean:4,9%;reedicon=2,1% Figure 1. Percentageof lambs born that are killed by predators based on 12 questionnairescompleted by ranchers and 44 investigationswith field necropsiesof dead lambs.

involved. Avian predation was less frequent than ical benefits. Population modeling of rock hyrax mammalian predation in all regions and lossesto suggeststhat the cost in grazing of increasedhyrax corvids exceeded those to eagles.Only in Scotland numbers in the absence of Black Eagles may be were eaglessignificantly involved in livestockpre- 150 times greater than the cost of lambs lost to dation, but this is based on only three studiesand eagles (Davies and Ferguson, in press). Eagles may overall lossesto predators were exceptionally low. also serve a valuable role on farmland by removing Eagleswere least involvedin livestockpredation in animals infected with rabies and other disease. South Africa. Black, Martial and Crowned Eagles It is certain that eagleshave causedproblems on are all highly territorial and are not inclined to some African farms in the past and will do so scavenge.There are no records of these eaglesbe- again. So far translocation of problem eagles has ing attracted in large numbers to an abundance of proved unsuccessful(Matchett and O'Gara 1987). dead or dying lambs, unlike Golden and Wedge- Five of eight translocated eagles in the Cape re- tailed Eagles. turned to their former ranges (Boshoff and Ver- The average Karoo farm accommodates 1534 non 1988), and 12 of 14 translocated eagles in small livestock which annually produce roughly North America returned to former ranges (Phillips 820 juveniles. Using the extensiveAustralian data et al. 1991). Harassment of problem eaglesby air- on neonatal mortality where Merino sheep are plane, explosivecharges and distresscall tapes did ranched under very similar conditions, 145 of not alleviate livestock losses in southwestern Mon- these juveniles are likely to die before weaning, tana, but human activityin the lambing paddo/:ks and predators are likely to be the prime causefor did help (Matchett and O'Gara 1987). Field trials 14 of these deaths (estimated cost R999 = of food aversion using lithium chloride in lamb US$289). Usingworldwide data, it is likely that ea- carcassesare not yet conclusive(Brand 1992), but gles would be responsiblefor the death of only trials on captive eaglesin South Africa are planned 0.32 lambs on the averageKaroo farm annually (es- and this method may still prove useful in the fu- timated cost R24 = US$7). Current South African ture. data would indicate lower rates of loss but are not A recent trend in ranch management has been yet considered sufficiently complete for such cal- that small improvementsin flock managementmay culations.This estimatedcost of having Black Ea- be far more beneficial to productivity than large- gles on the farm is negligible and should be borne scale and often unsuccessfulattempts to control by farmers in order to conserve a potentially vul- predators. Most farmers are well aware that the nerable species,especially in light of their ecolog- provision of sufficient food and shelter is most im- 70 EXPANDEDABSTRACTS VOL. 33, NO. 1

WORLDWIDEN,EOPSY SURVEYS

3,5O 3•0 250

191

EA CO WC DC F E OT

1423 15.9

EA CO WC DC

15704 2.46 100 5

EA CO WC DC FE OT

12660 2.25::o EA ICO 11'2/•J•19,1•11 DCI FE tOT I

WORLD299783.91 11.23 15.142.38 E A CO WC DC F E OT Figure2. A geographicalsummary of resultson neonatallamb mortality derived from field necropsysurveys. RSA = South Africa; UK = United Kingdom; AUS = Australia; NZ = New Zealand; USA = United States.Losses are expressedas % of lambsborn. American studies clump predation of bothviable (healthy) and inviable lambs, whereas only healthylambs are consideredin other regions.Histograms to the right showthe relativeinvolvement of different predators in total small stock losses.EA = eagles;CO = corvids;WC = wild canids;DC = domesticcanids; FE = felids;OT = other. Exactpercentage predation attributable to eaglesis givenin the last column. MARCH 1999 EXPANDED ABSTRACTS 71 portant for successfullambing. Livestock are no ARNOLD,L.W. 1954. The Golden Eagle and its economic longer gathered into 'kraals' for overnight protec- status.Circ. No. 27. U.S. FishWildl. Serv.,Washing- tion in the Karoo, but there has been a successful ton, DC. return to this tradition in parts of Namibia. By far BERGO,G. 1987. Eaglesas predatorson livestockand deer. Fauna NorvegicaSeries C. 10:95-102. the simplestand most effectivemanagement tech- BOLEN,E.G. 1975.Eagles and sheep:a viewpoint.J. Range nique employed in the Karoo to reduce lamb loss- Manage.28:11-17. es is to place a shepherdwith the flock during the BOSHOFF,A.E 1980. Some socio-economicaspects of a crucial lambing period. Farmers who employed bird of prey questionnairesurvey. S. Afz.J. Wildl.Res. shepherds reported no lossesto eagles. Where 10:71-81. flocks are not habituated to humans, scarecrows --, N.G. PALMER,G. AVERY,R.A.G. DAVIESAND MJ.F. may be used successfully(Matchett and O'Gara JArVIs. 1991. Biogeographicaland topographicalvar- 1987). Another simple management option cur- iation in the prey of the BlackEagle in the Cape Prov- rently employedby manyKaroo farmers is to move ince, South Africa. Ostrich 62:59-72. their lambing flocks away from areas of eagle ac- --AND C.J. VERNON. 1988. The translocation and homing ability of problem eagles.S. Aft.J. Wildl. Res tivitysuch as nests. This is especiallyimportant dur- 18:38-40. ing July and Augustin the Karoo (springlambing) BOWNS,J.E., J.W. DAVENPORT,J.F. Wo•kvd•O•N,D.B. NIEL- when eagle nestlingsare going through sensitive SON AND D.D. DWYER. 1973. Determination of cause developmentalphases and when dietary imprint- and magnitude of sheep lossesin southwesternUtah ing may occur.Removal of lamb carcassesfrom the Utah Sd. 34:34-37. lambing area prevents attraction of unwanted BRAND,D. 1992. The use of lithium-chloride as a preda- predators which may turn their attentions to live tion aversionagent for problem eagles in the Cape lambs when carrion becomes unavailable. If such Province and Orange Free State, South Africa. Gabar 7:51-55. management options fail to prevent livestockpre- BROOKER, M.G. AND M.G. RIDPATH. 1980. The diet of the dation by problem eagles, these birds should be Wedge-tailedEagle Aquila audaxin WesternAustralia. removed from the wild to prevent spread of a Aust. Wildl. Res. 7:433-452. lamb-killing trait in the population.Captive-breed- BROWN,J.E. 1976.Field criteriafor predatordamage as- ing programs,falconers and zoos can be consid- sessment. Utah Sci. 37:26-30. ered as destinations for such birds. BROWN,L.H. 1975. Brown on golden eagles'and sheep Simplyon the basisthat lamb killing by eaglesis Pages129-133 inJ.R. Murphy [EDs.], Population sta- very rare and can be dealt with, I would argue that tus of raptors. Proc. Conf. Raptor Conserv. Tech., large eagles are compatible with the open-range Raptor Res. Rep. No. 3. farming of small stock.Farmers must learn to tol- •. 1976. Eaglesof the world. Purnell, London,U.K. erate compatible predators, especiallyin the case BRUNS,E.H. 1970. Winter predation of Golden Eagles of vulnerable eagle populations.Conservationists and coyoteson pronghorn antelopes. Can. FieldNat 84:301-304. mustremain preparedto help farmersmanage any COOPER,A.B. 1969. Golden Eagle kills red deer calf. J problem situationsthat arise.Ecological benefits to Zool. (Lond.). 158:215-216. farmers (which are obviousfor black eaglesin the DAVIES,R.A.G. 1994. Black Eagle Aquila verreauxiipreda- Karoo) can be stressedto win the case for large tion on rock hyrax Procaviacapensis and other prey in eagles in sheep-farmingareas, but future survival the Karoo. Ph.D. thesis, Univ. of Pretoria, Pretoria, of wildlife should not have to depend on a signif- South Africa. icant financial gain to society,and education pro- DAVIES,R.A.G. ANDJ.W.H. FERGUSON(extended abstract gramsshould alsoaim to instill an aestheticappre- in press).The influence of predation by BlackEagles ciation of these predators. Aquila verreauxiion rock hyrax Procaviacapensis numbers in the arid Karoo. In Proceedingsof the Fifth ACKNOWLEDGMENTS World Conference on Birds of Prey, Midrand, South I would like to thank the Raptor ConservationGroup Africa, Aug. 1998. of the EndangeredWildlife Trust for paying my travel FOSTER,H.A. AND R.E. CPdSLER.1979. Evaluation of Gold- costs to the conference, and the conference organizers for coveringother expenses. en Eagle predation on domesticsheep, Temperance Creek-Snakesheep and goat allotment,Hell's Canyon LITERATURE CITED National Recreation Area, Oregon. USFWS, Animal ARMENTROUT,D. 1980. Environmental viewpoint. Pages Damage Control Programme, Portland, OR U.S.A. 6-9 in Proceedingsof the predator control summit. GLOVER,F.A. ANDL.G. HEUGLY.1970. Golden Eagle ecol- Texas Dept. Agric., Austin, TX U.S.A. ogy in WestTexas. Final rep. Natl. Audubon Soc.,Col- 72 EXPANDEDABSTRACTS VOL. 33, NO. 1

orado Co-operative Wildl. Res. Unit, Fort Collins, CO vironmental Conservation,Johannesburg, South Af- U.S.A. rica. HESS,E.H. 1973. Imprinting. Early experienceand the PHILLIPS,R.L., J.L. CUMMINGSAND J.D. BEPmY.1991. Re- developmentalpsychobiology of attachment.Van Nos- sponsesof breeding Golden Eagles to relocation trand, New York, NY U.S.A. Wildl. Soc. Bull. 19:430-434. HEWSON,R. 1981. The red fox Vulpesvulpes as a scavenger RABINOWITCH,g. 1969. The role of experiencein the de- and predator of sheep in West Scotland. Ph.D. disser- velopmentand retentionof seedpreferences in Zebra tation, University of Aberdeen, Aberdeen, Scotland. Finches. Behaviour 33:222-236. HEY, D. 1959. Conservationof birds in the province of ROWLEY,I. 1970.Lamb predationin Australia:incidence, the Cape of Good Hope. Ostrichsuppl. 3:23-26. predisposing conditions, and the identification of IMMELMaNN,K. 1975. Ecologicalsignificance of imprint- wounds. CSIRO Wildl. Res. 15:79-123. ing and early learning. Ann. Rev.Ecol. Syst. 6:15-37. SIEGFRIED,W.R. 1963. A preliminary report on Blackand KOLBE,F.E 1967. A plea for dassies.Afr. Wildl. 21:237- Martial Eaglesin the Laingsburgand PhilipstownDi- 248. visions.Cape Department of Nature Conservation,In- ß 1983. Dr. S.H. Rubidge, on the dassieproblem vestigationRep. No. 5, Johannesburg,South Africa. and "Lammervanger" controversy 1933-1954. The SI4OGI•ND,T. 1974.Villreinens habitatatferd, okologiske Naturalist 27:15-19. og sosialefaktorer. Utrykt progresjons-rapport til D I•. LAWSON,D. 1989. The effectsofpredators on sheepfarm- STEYN,P. 1982. Birds of prey of southern Africa. David ing in Natal SouthAfrica: an opinion survey.S. Afr.J. Philip, Cape Town, South Africa. Wildl. Res. 19:4-10. SVENDSEN,R.S. 1980. Kongeorn Aquila chrysaetosangrep LEHTI,R.A. 1947. The Golden Eagle attackingantelope. sau. Vat Fuglefauna3:20-26. J. Wildl. Manage. 11:348-349. TEXAS CROP AND LIVESTOCK REPORTING SERVICE. 1979. LEOPOLD,A.S. AND T.O. WOLFE. 1970. Food habits of nest- Texassheep and goat lossesand marketingpractices. ing Wedge-tailedEagles, Aquila audax, in South-east- Texas Dept. Agric., Austin, Texas. ern Australia. CSIRO Wildl. Res. 15:1-17. THOMAS,A.D. 1946. The Cape dassie.Aft. Wildl. 1:108- LOCKIE,J.D. 1964. The breeding densityof the Golden 111. Eagle and fox in relation to food supplyin Wester TIGNER,J.R. ANDG.E. LA•SON.1977. Sheeplosses on se- Ross, Scotland. Scott.Nat. 71:67-77. lected ranchesin southernWyoming. J. RangeMan- MATCHETT, M.R. AND B.W. O'GaRA. 1987. Methods of age.30:244-252. controlling Golden Eagle depredation on domestic -- AND--.. 1981. Golden Eagles:scavengers and sheepin southwesternMontana. J. RaptorRes. 21:85- predatorson domesticlambs. Pages 359-361 inJ.M. 94. Peek and P.D. Dalke [EDS.], 1981 Wildlife-Livestock NASS,R.D., G. LYNCHANDJ. THEADE. 1984. Circumstances RelationshipsSymposium: Proc. 10. For.Wildl. Range associatedwith predation rateson sheepand goats.J. Exp. Stn., Univ. Idaho, Moscow, ID U.S.A. RangeManage. 37:423-426. U.S. FISHAND WILDLIFE SERVICE. 1978. Predator damage NESSE,G.E., W.M. LONGHURSTAND W.E. HOWARD. 1976. in the west: a studyof coyote managementalterna- Predation and the sheep industryin California. Univ. tives. USDI, Fish and Wildl. Serv., Washington,DC of Calif. (Div.Agric. Sci.), Bulletin No. 1878,Berkeley, U.S.A. CA U.S.A. WADE,D.A. ANDJ.E. BOWNS.1980. Proceduresfor evalu- O'G•, B.W. 1978. Sheeppredation by GoldenEagles ating predationon livestockand wildlife. TexasAgric. in Montana. Proc.Vert. Pest Control Conf. 8:206-213. Ext. Serv., Texas A&M Univ., College Station, TX , K.C. BRAWLEY,J.R. MUNOZ AND D.R. HENNE. U.S.A. 1983. Predation on domestic sheep on a western WILEY,R.W. ANDE.G. BOLEN.1971. Eagle-livestockrela- Montana ranch. Wildl. Soc. Bull. 11:253-264. tionships: livestock carcasscensus and wound char- P•LMER,N.G. 1983. An investigationinto the BlackEagle acteristics. Southwest. Nat. 16:151-169. Aquilaverreauxii---small stock relationship in the Lad- WOODFOre),W.H. 1966. A manual of falconry. A. & C. ismith District. Cape Department of Nature and En- Black, London, U.K.