Modeling the Effects of Trophy Selection and Environmental Disturbance on a Simulated Population of African

KARYL L. WHITMAN,∗†§‡ ANTHONY M. STARFIELD,† HENLEY QUADLING,†‡ AND CRAIG PACKER† †Department of Ecology, Evolution & Behavior, , 1987 Upper Buford Circle, St. Paul, MN 55108, U.S.A. §Porini consulting, 18527 73rd Avenue NE, Kenmore, WA 98028, U.S.A. ‡D4D Technologies, 630 International Parkway #150, Richardson, TX 75081, U.S.A.

Abstract: Tanzania is a premier destination for trophy hunting of African lions (Panthera leo) and is home to the most extensive long-term study of unhunted lions. Thus, it provides a unique opportunity to apply data from a long-term field study to a conservation dilemma: How can a trophy-hunted species whose reproductive success is closely tied to social stability be harvested sustainably? We used an individually based, spatially explicit, stochastic model, parameterized with nearly 40 years of behavioral and demographic data on lions in the Serengeti, to examine the separate effects of trophy selection and environmental disturbance on the viability of a simulated population in response to annual harvesting. Female population size was sensitive to the harvesting of young males (≥3 years), whereas hunting represented a relatively trivial threat to population viability when the harvest was restricted to mature males (≥6 years). Overall model performance was robust to environmental disturbance and to errors in age assessment based on nose coloration as an index used to age potential trophies. Introducing an environmental disturbance did not eliminate the capacity to maintain a viable breeding population when harvesting only older males, and initially depleted populations recovered within 15–25 years after the disturbance to levels comparable to hunted populations that did not experience a catastrophic event. These results are consistent with empirical observations of lion resilience to environmental stochasticity.

Keywords: lion harvest, Panthera leo, population model, population viability, sustainable use, Serengeti, trophy hunting

Modelaci´on de los Efectos de la Selecci´on de Presas y de la Perturbaci´on Ambiental en una Poblaci´on Simulada de Leones Africanos Resumen: Tanzania es un destino preferido para la caza deportiva de leones Africanos (Panthera leo) y es el hogar del estudio de largo plazo mas´ extensivo de leones no cazados. Por lo tanto, proporciona una oportunidad unica´ para aplicar datos de un estudio de largo plazo a un dilema de conservacion:´ Como´ se puede aprovechar sustentablemente a una especie cineg´etica cuyo ´exito reproductivo esta´ estrechamente ligado a la estabilidad social? Utilizamos un modelo estocastico,´ espacialmente expl´ıcito y basado en individuos con datos conductuales y demograficos,´ de casi 40 anos,˜ de leones en el Serengeti, para examinar los efectos separados de la seleccion´ de trofeos y la perturbacion´ ambiental sobre la viabilidad de una poblacion´ simulada de leones en respuesta a la cacer´ıa anual. El tamano˜ de la poblacion´ de hembras fue sensible a la cacer´ıa de machos jovenes´ (≥3anos),˜ mientras que la cacer´ıa represento´ una amenaza relativamente trivial a la viabilidad poblacional cuando la cacer´ıa se restringio´ a los machos maduros (≥6anos).˜ El funcionamiento general del modelo fue robusto a la perturbacion´ ambiental y a errores en la estimacion´ de edades con base

∗email [email protected] Paper submitted May 18, 2006; revised manuscript accepted February 12, 2007 591 Conservation Volume 21, No. 3, 591–601 C 2007 Society for Conservation Biology DOI: 10.1111/j.1523-1739.2007.00700.x 592 Harvesting African Lions Whitman et al. en la coloracion´ de la nariz como ´ındice para estimar la edad de trofeos potenciales. La introduccion´ de la perturbacion´ ambiental no elimino´ la capacidad para mantener una poblacion´ reproductiva viable cuando se cazaba solo a machos maduros, y las poblaciones inicialmente reducidas se recuperaron entre 15 y 25 anos˜ despu´es de la perturbacion´ alcanzando niveles comparables a poblaciones cazadas que no experimentaron un evento catastrofico.´ Estos resultados son consistentes con observaciones emp´ıricas de la resiliencia de leones a la estocacidad ambiental.

Palabras Clave: cacer´ıa deportiva, cosecha de leones modelo poblacional, Panthera leo, Serengeti, viabilidad poblacional, uso sustentable

Introduction too crude to evaluate the extent to which lion numbers have declined over the past 20–30 years, but suitable habi- tat has declined, and lions no longer occupy the full extent One of the most important factors for the effective man- of their historic range (Nowell & Jackson 1996; Chardon- agement of a species is a thorough understanding of life- net 2002), particularly in unprotected areas (Bauer & Van history and behavioral characteristics that are integral Der Merwe 2004). to its reproductive success. Behavioral ecological stud- Early twentieth-century hunting expeditions removed ies therefore play a critical role in conservation science large numbers of African lions, as did the European live- because they define such behavior and would not be pos- stock ranchers that followed. Many of the historic hunt- sible without long-term demographic records or detailed ing grounds used by white big game hunters and settlers observations of individuals (Caro & Durant 1995). Nev- have since acquired some degree of protection (MacKen- ertheless, long-term data sets encompassing several gen- zie 1988), and lion hunting has ceased altogether in many erations are uncommon, even more so for species that areas. Human encroachment, reduced prey base, and di- exhibit socially complex behaviors and that are hunted rect human-caused mortality (e.g., problem-animal con- by humans for sport. Removing the primary breeders in trol) are presumably responsible for recent declines in a population can disrupt social organization and hamper lion abundance and distribution (Woodroffe & Ginsberg reproduction and recruitment, thus making it difficult to 1998; Harcourt et al. 2001; Chardonnet 2002). In Kenya predict the potential impact of commercial hunting. Long- for example, lion numbers have continued to decline out- term studies have been conducted on African lions in the side protected areas (Ogutu et al. 2005; L. Frank, personal Serengeti for nearly 40 years providing precise estimates communication 2005) despite a ban on trophy hunting for for demographic and behavioral parameters in a detailed nearly 30 years. simulation model that can predict the likely impact of Lion populations are particularly vulnerable to extir- management strategies (Whitman et al. 2004). pation when population sinks and edge effects are cre- Large carnivores have declined in abundance and their ated by large-scale human activity (Woodroffe & Ginsberg ranges have diminished substantially throughout Africa 1998). In areas where humans and lions exist in close (Frank & Woodroffe 2001; Woodroffe 2001). As rising hu- proximity, lions pose a persistent threat to rural people man populations encroach on wildlife habitat, carnivores and their livelihood (Frank 1998; Packer et al. 2005a). are becoming increasingly vulnerable to human-wildlife Because lions will readily scavenge, they are easily poi- conflict, especially adjacent to protected areas in Africa soned and trapped by poachers using long-line snares. In (Woodroffe & Ginsberg 1998; Harcourt et al. 2001). Hu- addition only a few protected areas are large enough to man population size is a significant threat to species vi- provide adequate amounts of suitable habitat to maintain ability (Herremans & Herremans-Tonnoeyr 2000; Luck viable populations (Harcourt et al. 2001; Loveridge et al. et al. 2004) and a substantial proportion of carnivore 2001). Villagers and governments regularly kill lions that mortality is directly attributable to anthropogenic factors leave protected areas, potentially creating a vacuum that (Woodroffe & Ginsberg 1998). Retaliation from farmers draws more animals from the “source” populations in the over livestock depredation is a leading cause of carnivore protected areas. mortality (Ogada et al. 2003; Treves & Karanth 2003). Increasing human demands for land have reduced suit- Based largely on educated guesses from scientific au- able wildlife habitat and challenged developing countries thorities (Chardonnet 2002; Bauer & Van Der Merwe to find alternative resource-management strategies. Tro- 2004) and extrapolation from known population densi- phy hunting by foreign tourists (Kock 1995; Lewis & ties combined with catch-per-unit effort estimates from Alpert 1997) brings money into areas that are unsuitable trophy hunters (Chardonnet 2002), the number of African for photographic tourism and provides an economic in- lions (Panthera leo) is estimated be between 16,500 and centive for locals to conserve wildlife (Child & Child 1990; 47,000 individuals. Earlier estimates of lion numbers are Eltringham 1994). Nevertheless, profit-oriented, short-

Conservation Biology Volume 21, No. 3, June 2007 Whitman et al. Harvesting African Lions 593 term gains from trophy hunting can increase the risk of Although we demonstrated that sustainable hunting of overexploitation (Lavigne et al. 1996). lions could be achieved by removing only older individ- Trophy hunting of lions is a multimillion dollar industry uals (Whitman et al. 2004), we did not test whether this in sub-Saharan Africa (Creel & Creel 1997; Overton, 1998; strategy is sensitive to environmental effects. The impact Whitman 2002). Tourists hunt lions in 10 African coun- of environmental stochasticity was homogenized in our tries: Botswana, Burkina Faso, Central African Republic, original model by averaging the year-to-year variation in Chad, Mozambique, Namibia, Republic of South Africa, the underlying data despite a catastrophic die-off in 1993– Tanzania, Zambia, and Zimbabwe. Tanzania yields about 1994 from a virulent canine distemper epidemic in our 17% of the top-10 lions reported to the Safari Club Interna- study population (Roelke-Parker et al. 1996). Social be- tional (SCI) and is the most expensive country in which to havior of some animals can influence the impact of eco- hunt lions (Whitman 2002). An average 21-day safari costs logical fluctuations even when environmental conditions over $75,000. In Tanzania lions contribute approximately are notably severe (Vucetich et al. 1997; Grimm et al. 10–13% of trophy fees despite accounting for only 2–4% 2003). To further examine the effects of environmental of the total number of animals taken in any given year perturbation, we examined the robustness of our model (PAWM 1995; Broomhead 1997a,b). A lion quota often against two additional influences: (1) the importance of promotes use of less-marketable species (Creel & Creel environmental stochasticity, by imposing an ecological 1997). The temporary ban on lion hunting in Botswana disaster before the onset of annual harvesting and (2) the between 2001–2005 is estimated to have cost the govern- inevitable inaccuracies of age estimation when it is based ment over $4.3 million/year (McGreal 2001). on a statistical correlate of age. Although the sport-hunting industry generates signifi- cant revenue and makes a positive contribution to the na- tional economy, a lack of transparency in hunting-block allocation, arbitrary division of hunting concessions and Lion-Hunting Policies in Tanzania concomitant increases in quotas, and unethical behav- ior by professional hunters all contribute to the risk of In Tanzania trophy hunting takes place on approximately overhunting (Overton 1998; Whitman 2002; Baldus & 250,000 km2 of government-owned land in game re- Cauldwell 2004). Given that lion populations may be par- serves, open areas, wildlife management areas, and game- ticularly sensitive to the removal of resident males (e.g., controlled areas. Hunting areas are divided into approxi- Packer et al. 1988; Yamazaki 1996; Whitman et al. 2004), mately 141 concessions used by approximately 42 sepa- questions have arisen concerning the sustainability and rate outfitters (Wildlife Division 2003). Imprecise bound- utility of trophy hunting as a conservation tool. Legal aries of many of the hunting concessions preclude an quotas are based on guess work (Severre 1995) and es- accurate measurement of the area of each, but in 1992 tablishing a reasonable offtake presumes knowledge of a the average hunting block was approximately 1370 km2 population that is often unavailable. Nevertheless, Whit- (Leader-Williams et al. 1996). Today the average size is man et al. (2004) recently demonstrated that lion hunting likely somewhat smaller because many areas have been can be biologically sustainable without the use of quotas subdivided since 1992. Trophy hunting of lions is limited by applying an age-based criterion to trophy selection. to males in Tanzania. In 1996 all but 2 of 131 concessions In a previous study we estimated the effects of trophy had a lion quota for their area, and the average quota for hunting on long-term population viability with varying those areas was 3.9 lions/concession (range 1–12, n = harvesting strategies for a simulated population (Whit- 129) (Broomhead 1997a). Less than 50% of the total lion man et al. 2004). We concluded that harvesting ≥5- to quota is met each year (236 lions out of a quota of 516 6-year-old males has the least impact on the population were shot in 1995: Broomhead 1997a,b). If we conser- and produces the greatest cumulative number and most vatively estimate that 10,000–15,000 lions reside in the consistent “crop” of high-quality males over time and runs whole of Tanzania, an annual harvest of about 200 lions virtually no risk of overhunting. Our data also showed represents <2% of the total population. that nose color can be used to estimate the age of po- The Wildlife Division of Tanzania mandates that for- tential trophy males before they are shot. We parame- eign tourists must use a licensed professional hunter as terized the model based on empirical data from nearly a guide and pay for a 21-day safari to shoot a lion (re- 40 years of ongoing behavioral and demographic studies gardless of the success or actual length of the safari). The conducted in the Serengeti (Packer et al. 1988; Packer government sets a flat rate of $850/day/client and a tro- et al. 1998; Packer et al. 2001). The lions’ complex so- phy fee of $2000 (PAWM 1995). Nevertheless, hunting cial behavior plays a significant role in cub survival, pride companies charge significantly higher rates for conces- formation, and recruitment (Packer 2000; Packer & Pusey sions that repute to have “quality” lions such as the re- 1983a,b, 1984), and infanticide by incoming males greatly serves surrounding . In these ar- increases the risk of extinction in simulated hunted pop- eas daily rates may reach $2800/day/client, trophy fees ulations (Whitman et al. 2004). may be as much as $5500, and companies may add a

Conservation Biology Volume 21, No. 3, June 2007 594 Harvesting African Lions Whitman et al.

5–25% surcharge to the trophy fee that goes to antipoach- cruitment into natal prides is contingent upon a specified ing and community development programs (Whitman maximum number of females (that can be temporarily 2002). Approximately 9% of all trophy fees paid to the exceeded by no more than 2 females) for each pride. Wildlife Division are supposed to be disbursed to lo- Subadult females that do not join their natal prides may cal communities through their district councils (Leader- search for vacant territories or are eliminated from the Williams et al. 1996, but see Overton 1998; Leader- model if none are available. The output of average age of Williams & Hutton 2005). At the minimum a one-person resident males, size range of male coalitions and subadult 21-day safari costs between $30,000–50,000, excluding cohorts, and male:female:cub ratios of a simulated pop- transportation, taxidermy, preparation and shipping of ulation at “equilibrium” realistically portrayed observed trophies, nonhunting observers, and gratuities (Whitman populations (Whitman et al. 2004). Each “population” of 2002). Baiting lions by means of a carcass is unofficially lions contains a maximum of 10 prides consisting of ≤ 9 permitted by order of the Director of Wildlife and shoot- females/ pride and roughly equates to numbers expected ing may take place only between one-half hour before in an area approximately 1000 km2 in East Africa (Schaller sunrise and sunset (Whitman 2002). 1972; Rudnai 1973) We established a standardized start- ing point for all simulations by using a population at equi- librium (in the absence of hunting) that began with an arbitrary set of individuals and an arbitrary age distribu- The Model tion. “Founding” populations reached equilibrium within 20–25 years, so we ran each simulation for 30 and 40 years We developed a comprehensive, spatially explicit model with 100 replicates. that tracks individuals through time (Quadling & Starfield To test for the effects of trophy hunting, we selected 2002; Whitman et al. 2004). Each territory or “pride” con- trophy males from our standardized population according sists of females and their dependent offspring that are to age (≥3yearsto≥6 years) and removed lions at ran- spatially linked within the population. To represent the dom with respect to social status (resident vs. nomad) for worst-case scenario, we modeled the impact of trophy annual quotas ranging from 0 to 20 males. Each replicate hunting on a closed population. Life-history behavior and began with identical age structure, reproductive history, demographic rates used in the model are defined by age pride affiliation, and spatial distribution. We harvested class (cubs <6 months, cubs 6–12 months, cubs 13–24 males at random (according to our selection criteria) ev- months, subadult males and females, adult males and fe- ery 6 months, although it was not always possible to sat- males) based on long-term studies of an unhunted popu- isfy the quota. lation of lions in Serengeti National Park and Ngorongoro We tested for the importance of environmental stochas- Crater (Packer et al. 1988; Packer et al. 1998; Packer et ticity by harvesting an annual quota of 0 to 20 males ac- al. 2001). Females may only produce cubs between 3–13 cording to age from a “recovering” population versus one years of age, and they breed only when their offspring that was stable at the onset of hunting. A recovering pop- from a previous litter reach 2 years of age or if an entire ulation began with only 5 of 10 territories occupied from litter perishes. Males are classified as subadults, nomads, the initial standardized population. or residents. Singleton males may join other solitary males The fleshy part of a lion nose accrues pigmentation or coalitions of two. Resident males may reside in up to linearly with age and can be used to estimate age accord- three prides at one time. Nomadic and subadult males do ing to a “nose pigmentation index” (NPI) (Whitman et not breed, but may move freely between prides a speci- al. 2004). We used nose coloration measurements from fied number of times per time step. A competition matrix 95 individuals of known age from the Serengeti (males (Starfield et al. 1981) based on social status, collective age, and females) and Ngorongoro Crater (females) (Whit- and coalition size determines whether males encounter- man et al. 2004) to group individuals into 40, 50, 60, ing each other fight, and subsequent survival if they do. and 70% “black” categories based on the proportion of An age-specific probability of infanticide following the dark pigmentation on their noses. (We excluded males takeover of prides by males determines cub survivorship. from Ngorongoro Crater because of bias shown in the At each time step (6 months), the model assesses in- original study.) These categories corresponded with ages dividual survival, updates ages of survivors, allows fe- 4.42, 5.02, 5.61, and 6.20 years old, respectively. We cal- males to produce cubs, categorizes 2-year-old males in culated the frequency of overestimating a lion’s age with each pride into subadult male groups, advances 3-year-old the NPI relative to its true age. Thus, for each age class males into nomadic groups, and either permits subadult we established a probability that a lion would be shot, females to join a pride or removes them from the pop- based on nose pigmentation to estimate age (Table 1). ulation. The model determines litter size for each eligi- We then tested for the effect of accidentally harvesting ble female with a random number from a distribution. an underage animal in our model according to these fre- Survivorship depends on observed survival rates for in- quencies by removing 0 to 20 lions annually from a stable dividuals of similar age, sex, and social status. Female re- population.

Conservation Biology Volume 21, No. 3, June 2007 Whitman et al. Harvesting African Lions 595

Table 1. The probability of a male lion being shot based on a specified percentage of nose coloration according to age class.∗

Probability (%) Age class 40 50 60 70

3 0.17 0.08 0.00 0.00 4 0.67 0.17 0.00 0.00 5 0.82 0.73 0.36 0.09 6 1.00 0.88 0.63 0.38 7 0.75 0.75 0.75 0.75 8 1.00 1.00 1.00 0.75 ∗Observed frequency of overestimation of lion age based on nose coloration measurements from 95 individuals of known age from Serengeti males and females and Ngorongoro Crater females.

Results

Female population size was highly sensitive to any quota taking >2 males/year when the harvest included lions as young as 3 years old (Fig. 1a). Irrespective of quota size, harvesting males at least 5 years old had a negligible effect on population viability. Placing age restrictions on lions shot increased the total number of males harvested after 30 years and increased the number of 5- and 6- year old trophies in the population by protecting young males (Fig. 1b & c). Harvesting older males did not reduce the population’s capacity to maintain a viable breeding population even when the population was initially depleted by an environ- mental perturbation, whereas shooting younger males sig- nificantly affected the outcome. Female population size after 30 years of harvesting differed very little between a population that was initially at equilibrium and one that was recovering from a perturbation event (Fig. 1a & 2). Restricting the harvest to older age classes permitted the initially depleted populations to recover to levels com- parable to stable populations within 15–25 years. Growth in female population size was resilient to harvesting when annual quota size was both moderate (4 males; Fig. 3) Figure 1. Effects of 30 years of lion trophy hunting as and high (8 males; Fig. 4). Although unhunted, recov- a function of hunting quota size and male age in a ering populations maintained a larger number of females hypothetical population. Average outcome after 100 through time in comparison with recovering hunted pop- simulation runs is shown from shooting males in four ulations, they grew in a similar trajectory when only males age groups (≥ 3, ≥4, ≥5, and ≥ 6 years old): (a) ≥5or≥6 years were targeted. Quota size was only impor- number of adult females after 30 years, (b) total tant when immature males (≤4 years) were taken. Extinc- number of males harvested, (c) total number of 5- to tion occurred at least once per 100 runs when quota size 6-year-old “trophies” harvested (adapted from exceeded more than two ≥3or≥4 year old males each Whitman et al. 2004). year; whereas extinction never occurred when hunting was restricted to males ≥5or≥6 years old. trend in our study population, the 60% rule could lead to In the absence of background demographic data, a few misclassifications of overestimating the target lion’s hunters who tried to follow our recommendations of a age, but there would also be a number of older animals 6-year age minimum could rely on nose coloration to esti- whose ages were underestimated (Table 1). The overall mate age. They shot only males whose noses were at least effect of this error was negligible (Fig. 5), and there was 60% pigmented (cf. Table 1; Whitman et al. 2004). Given no real risk of overexploiting the population if the hunt- the observed variability in the age and nose coloration ing industry were to enforce a 60% minimum. There is

Conservation Biology Volume 21, No. 3, June 2007 596 Harvesting African Lions Whitman et al.

B 90 Figure 2. Effects of 30 years of trophy hunting of male 80 lions on the number of adult females as a function of 70 the hunting quota size and male age for a population 60 recovering from a hypothetical environmental disturbance after 100 simulation runs (males shot in 50 four age groups: ≥3, ≥4, ≥5, and ≥6 years old). 40 30

20 an obvious difference in the degree of pigmentation be- Number of females tween 30% and 60% (Fig. 6), and even if hunters applied 10 a 50% rule (for 5-year-old lions), the long-term effect on 0 population size would be small. Consistent with shoot- 12345101520253050 ing males that are a minimum of 4 years of age, extinction Years never occurred when hunters only removed males with at C 90 least 40% black noses, although the effect on population viability would be much less desirable. 80 70 Discussion 60 50 Modeling 40 In testing for measurement errors in age assessments, 30 20 we found that our model was robust to minor misclas- Number of females sifications in trophy age, and the results of our analy- 10 sis suggest that populations remained most viable when 0 the harvest was limited to older males. We found no 12345101520253050 meaningful difference in population persistence between Years a simulated hunted population that was initially stable D 90 and one that was recovering from a perturbation. It is 80

70 −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ 60 Figure 3. Effect of environmental disturbance on 50 female population numbers through time as a 40 function of male age in a stable versus recovering 30 population. Female population size when hunters 20 shoot males that are (a) ≥3, (b) ≥4, (c) ≥5, and (d) Number of females 10 ≥6 years old. Average outcomes over 100 replicates for a stable population, a recovering population, and a 0 12345101520253050 recovering unhunted population are shown. The stable Years and recovering populations describe an annual hunting quota of four males. Recovering Stable Unhunted

Conservation Biology Volume 21, No. 3, June 2007 Whitman et al. Harvesting African Lions 597 well documented that environmental factors such as dis- A 90 ease (Roelke-Parker et al. 1996; Kissui & Packer 2004), 80 food availability (Packer & Pusey 1984, 1995), and rainfall 70 (Hanby & Bygott 1979; Packer et al. 2005a) have a con- 60 siderable impact on the mortality of lion cubs, but when 50 social stability is high, lions can be quite resilient to dis- turbance. For example, even after a catastrophic decline 40 of about 30% in the Serengeti population as a result of a 30 20 canine distemper virus epidemic in 1993–1994 (Roelke- females of Number Parker et al. 1996; Packer et al. 1999), the population 10 fully recovered within a few years (Packer et al. 2005b). 0 Further back in history, Serengeti lions were heavily ex- 1 2 3 4 5 101520253050 ploited by settlers before the establishment of the national Years park (Turner 1987). The carrying capacity of the Serengeti B 90 lions in those days was far lower due to the suppression 80 of the wildebeest population by rinderpest (Packer et al. 70

2005b). Yet lion numbers grew rapidly within 10 years 60 of the eradication of rinderpest in the 1960s, as soon as short-term ecological factors were favorable (Hanby et al. 50 1995; Packer et al. 2005a). 40 Populations that undergo repeated catastrophic events 30

in quick succession may never return to their carrying Number of females 20 capacity (Kissui & Packer 2004), and small, isolated pop- 10 ulations suffer additional risks from inbreeding (e.g., Pe- 0 terson et al. 1998), so trophy hunting is ill-advised in cer- 12345101520253050 tain circumstances. Rather than rely on the results of our Years study, conservation managers should take care to param- 90 eterize our model with data from their own populations C wherever possible. 80 70

Conservation Implications 60 In observed populations, overall pride size, resident male 50 tenure, and the size and age of male coalitions all differ 40 between hunted and unhunted populations (Yamazaki 30 1996; K.W., unpublished data). Our model suggests that 20 long-term harvesting can be conducted sustainably and Number of females the quality of trophies can be enhanced by improving 10 trophy selection. 0 12345101520253050 Older males tend to avoid baits set by hunters and limit their visits to short nocturnal bouts, whereas immature Years males readily scavenge and often rest near baits, poten- D tially making them more vulnerable to hunters (K.W.,per- sonal observation). Therefore it is important that hunters 70 −−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−−→ 60 50

Figure 4. Effect of environmental disturbance on 40 female population numbers through time as a 30 function of male age in a stable versus recovering 20 population. Female population size when hunters Number of females shoot males that are (a) ≥3, (b) ≥4, (c) ≥5, and (d) ≥6 10 years old. Average outcomes over 100 replicates for a 0 stable population, a recovering population, and a 12345101520253050 recovering unhunted population are shown. The stable Years and recovering populations describe an annual quota Recovering Stable Unhunted recovering of eight males.

Conservation Biology Volume 21, No. 3, June 2007 598 Harvesting African Lions Whitman et al.

Figure 5. Effects of 40 years of trophy hunting of male lions on female population size as a function of hunting quota size and male age. Average outcomes over 100 simulation runs from shooting males ≥3, ≥4, ≥5, and ≥6 years old are shown. Dotted lines show use of nose coloration as a measure of male age: ≥40% black, 4 years old; ≥50% black, 5 years old; and ≥60% black, 6 years old. be able to quickly and accurately determine the age of li- ons. The color of a lion’s nose can be quickly assessed with a nominal amount of experience. Professional hunters routinely use high-quality binoculars, making this simple rule of thumb easy to implement (G. Damm, personal communication). In comparison more-stringent restric- Figure 6. Lion nose tips typifying coloration tions exist for other big game such as Stone sheep (Ovis measurements: (a) 4-year-old Serengeti male with 30% dalli stonei) in British Columbia, where rams must be nose pigmented and (b) 7-year-old Serengeti female ≥8 years old and hunters are required to either count with 60% nose pigmentation. Measurements were growth rings or ensure that the tips of the horns rise taken following methods outlined in Whitman et al. above the bridge of the nose (British of Columbia Min- (2004). istry of Water, Land & Air Protection 2005). Professional hunters in Tanzania are currently using the nose pigmen- tation index in concert with other aging methods (Whit- man & Packer 2007) to reduce the likelihood of shooting to the northern half of the country (Overton 1998). In immature male lions (Legendary Adventures 2005). addition, hunting areas within Tanzania are generally lo- Legalized hunting can often improve wildlife survivor- cated immediately adjacent to national parks and provide ship and population viability by mitigating other anthro- a buffer zone that protects wildlife from people and peo- pogenic factors (see Leader-Williams & Hutton 2005 for ple from wildlife. With the control of problem animals on further reference). Some hunting companies actively pa- the rise, hunting’s capacity to reduce lion-human interac- trol their concessions for poachers and remove wire tions with local communities should not be underrated. snares that threaten not only lions, but vast numbers of One of the biggest issues to be addressed is ensuring that other wildlife. Likewise, because hunting typically takes financial benefits from hunting filter down to the affected place in marginal and tsetse-fly-infested areas unsuitable communities (Leader-Williams & Hutton 2005; Walpole & for other livelihoods, it provides the best economic incen- Thouless 2005). tive for protecting wildlife in those areas (Overton 1998). Control of problem animals, antagonistic killing, poach- Photographic tourism can generate approximately twice ing, and loss of habitat are more serious threats to lion the revenue, but it does so at the expense of greater im- conservation than legalized hunting. Control of problem pact on the environment in the form of cars, fuel, and animals represents the single greatest factor responsible other infrastructure (Overton 1998; Baldus & Cauldwell for lion decline outside protected areas today. Recently 2004). In Tanzania hunting distributes wealth country- in Kenya, for example, lion numbers in Nairobi National wide; photographic tourism tends to limit local benefits Park have declined from approximately 35 lions to 9 since

Conservation Biology Volume 21, No. 3, June 2007 Whitman et al. Harvesting African Lions 599

1998 due to Maasai spearing 27 out of 40 lions that strayed Long-term Research in Environmental Biology and Animal outside the park (Njumbi 2003). Behavior programs. We are grateful to G. Damm for con- Tourist hunting has enormous potential for minimizing tributing a hunter’s perspective to our work. We thank T. the costs of living with problem animals through eco- Gelatt, S. Thirgood, L. Frank, P.Lindsey,and an anonymous nomic compensation of local communities living near reviewer for their valuable comments on the manuscript. hunting areas (Conover 2002). Zimbabwe’s community- This material is based upon work supported by the Na- owned CAMPFIRE (Communal Areas Management Pro- tional Science Foundation under Grant No. 0343960. gramme for Indigenous Resources) contributes 50% of each license fee for hunting an elephant to the community Literature Cited as compensation for elephant damage and as an incentive Baldus, R. D. 2004. Lion conservation in Tanzania leads to serious to reduce illegal hunting (Maveneke 1996 [from Lewis & human–lion conflicts with a case study of a man-eating lion killing Jackson 2005]). Although no compensation scheme set 35 people. Tanzania wildlife discussion paper no. 41. Deutsche up to specifically address lion-human conflict outside a Gesellschaft fur¨ Tecnische Zusammenarbeit (GTZ), Dar-es-Salaam, Tanzania. hunting area currently exists, a program that compen- Baldus, R. D., and A. E. Cauldwell. 2004. Tourist hunting and its role in sated Maasai outside Nairobi National Park demonstrates development of wildlife management areas in Tanzania. Deutsche the conservation potential of such schemes. Friends of Gesellschaft fur¨ Tecnische Zusammenarbeit (GTZ), Dar-es-Salaam, Nairobi National Park were successful in halting Maa- Tanzania. sai killing of lions in response to livestock depredation Bauer, H., and S. Van Der Merwe. 2004. Inventory of free-ranging lions Panthera leo in Africa. Oryx 38: 26–31. in 2001 by compensating local herdsmen KsH$15000 British Columbia Ministry of Water, Land and Air Protection. 2005. Hunt- (US$200) for each cow or donkey killed by a lion and ing and trapping synopsis 2005–2006. Monday Tourism Publications, KsH$2500 (US$33) for each goat (Njumbi 2003). In 1 Victoria, British Columbia. Available from http://bchuntingregs.com year KSH$417,500 (US$5490) was paid to the Kitengela (accessed July 2005). community (Njumbi 2003). Unfortunately the program Broomhead, N. G. 1997a. Procedures used by the Tourist Hunting Of- fices of the Wildlife Department. Report 1. Wildlife Division, Ministry was short-lived due to lack of funds, and Maasai have of Natural Resources and Tourism, Dar es Salaam, Tanzania. since killed at least 10 lions since the program ended Broomhead, N. G. 1997b. Income arising from game hunting by tourists in 2002 (Njumbi 2003). Because wealthy foreign tourists year ended 30 June 1996. Report 2. Wildlife Division, Ministry of have demonstrated a willingness to pay for the privilege to Natural Resources and Tourism, Dar es Salaam, Tanzania. hunt lions, compensating individuals who suffer from lion Caro, T. M., and S. M. Durant. 1995. The importance of behavioral ecol- ogy for conservation biology: examples from Serengeti carnivores. attacks or livestock depredation by establishing a fund fi- Pages 451–472 in A. R. E. Sinclair and P.Arcese, editors. Serengeti II: nanced by hunters through a reasonable compensation dynamics, management, and conservation of an ecosystem. Univer- fee for each lion shot as a trophy might greatly enhance sity of Chicago Press, Chicago. a community’s tolerance for living near lions. Chardonnet, P. 2002. Conservation of the African Lion: Contribution If trophy hunting is to serve as a conservation tool to a status survey. International Foundation for the Conservation of Wildlife, Paris, and Conservation Force, Metairie, Louisiana. for lion populations outside protected areas, the industry Child, G., and B. Child. 1990. An historical perspective of sustainable needs to adopt an age minimum to ensure a sustainable wildlife utilization. Paper prepared for workshop no. 7. 18th IUCN offtake and to address the needs of the people most likely General Assembly. World Conservation Union, Gland, Switzerland. to be affected by living in close proximity to lions. The Conover, M. 2002. Resolving human-wildlife conflicts: the science of results from our model demonstrate that lion populations wildlife management. CRC Press, Boca Raton, Florida. Creel, S., and N. M. Creel. 1997. Lion density and population structure in are resilient to sport hunting of mature males and that en- the Selous Game Reserve: evaluation of hunting quotas and offtake. vironmental stochasticity does not affect the predictions African Journal of Ecology 35:83–93. of the model. An age-restricted offtake provides an easy- Eltringham, S. K. 1994. Can wildlife pay its way? Oryx 28:163–168. to-implement alternative to quotas despite some inherent Frank, L. G. 1998. Living with lions: carnivore conservation and live- error in its application. stock in Laikipia District, Kenya. U.S. Agency for International Devel- opment contract number 623–0247-C-00–3002-00. Mpala Research Centre, Nanyuki, Kenya. Frank, L. G., and R. Woodroffe. 2001. Behavior of carnivores in exploited and controlled populations. Pages 419–442 in J. L. Gittleman, S. M. Acknowledgments Funk, D. W. Macdonald, and R. K. Wayne, editors. Carnivore conser- vation. Cambridge University Press, Cambridge, United Kingdom. Grimm, V., N. Dorndorf, F. Frey-Roos, C. Wissel, T. Wyszomirski, and W. We thank the government of Tanzania for permis- Arnold. 2003. Modelling the role of social behavior in the persistence sion to conduct research. K.L.W. appreciates support of the alpine marmot Marmota marmota. Oikos 102:124–136. from Wildlife Conservation Society, Big Game Special Hanby, J. P., and J. D. Bygott. 1979. Population changes in lions and other Projects Foundation of Minnesota, John D. & Cather- predators. Pages 249–62 in A. R. E. Sinclair and M. Norton-Griffiths, ine T. MacArthur Foundation, Graduate School of the editors. Serengeti: dynamics of an ecosystem. University of Chicago Press, Chicago. University of Minnesota, Dayton-Wilkie Foundation, and Hanby, J. P., J. D. Bygott, and C. Packer. 1995. Ecology, demography and Global Wildlife Trust; C.P. acknowledges support by the behavior of lions in two contrasting habitats: Ngorongoro Crater and McKnight Foundation and National Science Foundation the Serengeti Plains. Pages 315–331 in A. R. E. Sinclair and P.Arcese,

Conservation Biology Volume 21, No. 3, June 2007 600 Harvesting African Lions Whitman et al.

editors. Serengeti II: dynamics, management, and conservation of an protection on the density and distribution of carnivores and their ecosystem. University of Chicago Press, Chicago. prey in the Mara ecosystem of Kenya. Journal of , London Harcourt, A. H., S. A. Parks, and R. Woodroffe. 2001. Human density as 265:281–293. an influence on species/area relationships: double jeopardy for small Overton, G. 1998. The good, the bad, and the ugly. Swara 20 & 21:22– African reserves? Biodiversity and Conservation 110:1011–1026. 24. Herremans, M., and D. Herremans-Tonnoeyr. 2000. Land use and the Packer, C. 2000. Infanticide is no fallacy. American Anthropologist conservation status of raptors in Botswana. Biological Conservation 102:829–830 94: 31–41. Packer, C., S. Altizer, M. Appel, E. Brown, J. Martenson, S. J. O’Brien, Kiss, A. 1990. Living with wildlife: wildlife resource management with M. Roelke-Parker, R. Hofmann-Lehmann, and H. Lutz. 1999. Viruses local participation in Africa. Technical paper 130. World Bank, Wash- of the Serengeti: patterns of infection and mortality in African lions. ington, D.C. Journal of Animal Ecology 68:1161–1178. Kissui, B. M., and C. Packer. 2004. Top-down population regulation of Packer, C., L. Herbst, A. E. Pusey, J. D. Bygott, S. J. Cairns, J. P.Hanby, and a top predator: lions in the Ngorongoro Crater. Proceedings Royal M. Borgerhoff-Mulder. 1988. Reproductive success of lions. Pages Society of London, Series B DOI:10.1098/rspb.2004.2797. 363–383 in T. H. Clutton-Brock, editor. Reproductive Success. Uni- Kock, R. A. 1995. Wildlife utilization: use it or lose it—a Kenyan per- versity of Chicago Press, Chicago. spective. Biodiversity and Conservation 4:241–256. Packer, C., D. Ikanda, B. Kissui, and H. Kushnir. 2005a. Lion attacks on Lavigne, D. M., C. J. Callaghan, and R. J. Smith. 1996. Sustainable uti- humans in Tanzania. Nature 436:927–928. lization: the lessons of history. Pages 250–265 in V. J. Taylor and N. Packer, C., R. Hilborn, A. Mosser, B. Kissui, M. Borner, G. Hopcraft, Dunstone, editors. The exploitation of mammal populations. Chap- J. Wilmshurst, S. Mduma, and A. R. E. Sinclair. 2005b. Ecological man & Hall, London. change, group territoriality and population dynamics in Serengeti Leader-Williams, N., J. A. Kayera, and G. L. Overton. 1996. Tourist hunt- lions. Science 307:390–393. ing in Tanzania. World Conservation Union, Gland, Switzerland. Packer, C., and A. E. Pusey. 1983a. Adaptations of female lions to in- Leader-Williams, N., and J. M. Hutton. 2005. Does extractive use provide fanticide by incoming males. The American Naturalist 121:716– opportunities to offset conflicts between people and wildlife? Pages 728. 140–161 in R. Woodroffe, S. Thirgood, and A. Rabinowitz, editors. Packer, C., and A. E. Pusey. 1983b. Male takeovers and female reproduc- People and wildlife: conflict or coexistence? Cambridge University tive parameters: a simulation of oestrus synchrony in lions (Panthera Press, Cambridge, United Kingdom. leo). Animal Behaviour 31:334–340. Legendary Adventures. 2005. Choosing the right lion. Legendary Packer, C., and A. E. Pusey. 1984. Infanticide in carnivores. Pages 31–42 Adventures, Houston, Texas. Available from http://www. in G. Hausfater and S. B. Hardy, editors. Infanticide: comparative and legendaryadventures.com/LAI-hunting/files/lionarticle.pdf (access- evolutionary perspectives. Aldine, New York. ed July 2005). Packer, C., and A. E. Pusey. 1995. The Lack clutch in a communal Lewis, D. M., and P. Alpert. 1997. Trophy hunting and wildlife conser- breeder: lion litter size is a mixed evolutionarily stable strategy. The vation in Zambia. Conservation Biology 11:59–68. American Naturalist 145:833–841. Lewis, D., and J. Jackson. 2005. Safari hunting and conservation on com- Packer, C., A. E. Pusey, and L. Eberly. 2001. Egalitarianism in female munal land in southern Africa. Pages 239–251 in R. Woodroffe, S. African lions. Science 293:690–693. Thirgood, and A. Rabinowitz, editors. People and wildlife: conflict or Packer, C., M. Tatar and D. A. Collins. 1998. Reproductive cessation in coexistence? Cambridge University Press, Cambridge, United King- female mammals. Nature 392:807–11. dom. Peterson, R. O., N. J. Thomas, J. M. Thurber, J. A. Vucetich, and T. A. Loveridge, A. J., T. Lynam, and D. W. Macdonald. 2001. Lion conserva- Waite. 1998. Population limitation and the wolves of Isle Royale. tion research, workshop 1: survey techniques. Wildlife Conservation Journal of Mammalogy 79:828–841. Research Unit, University of Oxford, Oxford, United Kingdom. Planning and Assessment for Wildlife Management (PAWM). 1995. Pages Luck, G. W., T. H. Ricketts, G. C. Daily, and M. Imhoff. 2004. Alleviating 63–76 in N. Leader-Williams, J. A. Kayera, and G. L. Overton, editors. spatial conflict between people and biodiversity. Proceedings of Tourist hunting in Tanzania. Tanzania. Ministry of Natural Resources the National Academy of Sciences of the United States of America and Tourism, Dar es Salaam, Tanzania 101:182–186. Quadling, H. S., and A. M. Starfield. 2002. Exploiting object-oriented MacKenzie, J. M. 1988. The empire of nature: hunting, conservation, programming structures in the quest for an individual-based lion and British imperialism. Manchester University Press, Manchester, population model with an attractive user interface. South African United Kingdom. Journal of Science 98:449–454. Maveneke, T. 1996. The elephant’s importance to CBCD: the CAMPFIRE Roelke-Parker, M. E., et al. 1996. A canine distemper virus epidemic in example. Pages 31–33 in Rural development and conservation in Serengeti lions (Panthera leo). Nature 379:441–445 Africa. African Resources Trust, Harare, Zimbabwe. Rudnai, J. 1973. Reproductive biology of lions (Panthera leo massaica McGreal, C. 2001. Lions face new threat: they’re rich, American and Neumann) in Nairobi National Park. East African Wilidlife Journal they’ve got guns. The Guardian Unlimited, 27 April. Available from 11:241–253. http://www.guardian.co.uk/international/story/03604479311,00. Safari Club International (SCI). 1997. African lion. Pages 12–27 in The html (accessed October 2006). SCI record book of trophy animals. XI edition, Volume 1, Africa. SCI, Nowell, K., and P. Jackson, editors. 1996. Wild cats, status survey Tuscon, Arizona. and conservation action plan. World Conservation Union, Gland, Schaller, G. B. 1972. The Serengeti lion. The University of Chicago Press, Switzerland. Chicago. Njumbi, S. 2003. The surviving lions need not perish. Daily Na- Severrre, E. L. M. 1995. Setting quotas for tourist hunting in Tanzania. tion (Nairobi, Kenya). Available from http://63.110.5.89/news/Daily Pages 52–53 in N. Leader-Williams, J. A. Kayera, and G. L. Overton, Nation/27062003/Comment/Comment270620033.html (accessed editors. Tourist hunting in Tanzania. Ministry of Natural Resources July 2005). and Tourism, Dar es Salaam, Tanzania. Ogada, M. O., R. Woodroffe, N. O. Oguge, and L. G. Frank. 2003. Limiting Starfield, A. M., P. R. Furniss, and G. L. Smuts. 1981. A model of lion depredation by African carnivores: the role of livestock husbandry. population dynamics as a function of social behavior. Pages 121– Conservation Biology 17:1521–1530. 134 in C. W.Fowler, editor. Dynamics of large mammal populations. Ogutu, J. O., N. Bhola, and R. Reid. 2005. The effects of pastoralism and Wiley-Interscience, New York.

Conservation Biology Volume 21, No. 3, June 2007 Whitman et al. Harvesting African Lions 601

Treves, A., and K. U. Karanth. 2003. Human-carnivore conflict and per- as game ranch animals. South African Veterinary Wildlife Group, spectives on carnivore management worldwide. Conservation Biol- Faculty of Veterinary Science, Onderstepoort, Republic of South ogy 17:1491–1499. Africa. Turner, M. 1987. My Serengeti years. W.W. Norton, New York Whitman, K. L., and C. Packer. 2007. A hunter’s guide to aging lions in Vucetich, J. A., R. O. Peterson, and T. A. Waite. 1997. Effects of social eastern and southern Africa, Safari Press, Longbeach, California. structure and prey dynamics on extinction risk in gray wolves. Con- Whitman, K., A. M. Starfield, H. S. Quadling, and C. Packer. 2004. servation Biology 11:957–965. Sustainable trophy hunting of African lions. Nature 428:175– Walpole, M. J., and C. R. Thouless. 2005. Increasing the value of wildlife 178. through nonconsumptive use: deconstructing the myths of eco- Wildlife Division. 2003. Reference manual for implementing guide- tourism and community based tourism in the tropics. Pages 122–139 lines for the designation and management of wildlife management in R. Woodroffe, S. Thirgood, and A. Rabinowitz, editors. People and (WMAs) in Tanzania. Ministry of Natural Resources and Tourism, wildlife: conflict or coexistence? Cambridge University Press, Cam- Dar es Salaam, Tanzania. bridge, United Kingdom. Woodroffe, R. 2001. Strategies for carnivore conservation: lessons from Whitman, K. L. 2002. Safari Hunting of Lions: A review of policies, contemporary extinctions. Pages 61–92 in J. L. Gittleman, S. M. practices and industry concerns. Pages 111–124 in H. O. De Waal, Frank, D. MacDonald, and R. K. Wayne, editors. Carnivore conser- editor. Proceedings of the 2nd meeting of the African Lion Work- vation. Cambridge University Press, Cambridge, United Kingdom. ing Group, 2002. World Conservation Union, African Lion Working Woodroffe, R., and J. R. Ginsberg. 1998. Edge effects and the ex- Group, Brandhof, Republic of South Africa. tinction of populations inside protected areas. Science 280:2126– Whitman, K. L., and C. Packer. 1997. The effect of sport hunting on 2128. the population dynamics of the African lion. Pages 177–183 in J. van Yamazaki, K. 1996. Social variation of lions in a male-depopulated area Heerden, editor. Proceedings of a symposium on lions and leopards in Zambia. Journal of Wildlife Management 60:490–497.

Conservation Biology Volume 21, No. 3, June 2007