ANT 154BN-10 Predation

ANT 154BN-10 Predation - ANT 154B Lecture #10: Predation 3...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: ANT 154B Lecture #10: Predation 3 Feb 2011 Thursday, February 3, 2011 1 Predation 1. Predation on primates: general 2. Anti-predator strategies 3. Predation by primates: chimpanzees Thursday, February 3, 2011 2 Predation >1. Predation on primates: general 2. Anti-predator strategies 3. Predation by primates: chimpanzees Thursday, February 3, 2011 3 redation and antipredator strategies among primates. I describe patterns of ntipredator behavior and attempt to explain the variation among primate xa and among antipredator strategies. I use predation by chimpanzees on ed colobus and antipredator strategies by the colobus as a case study of how Predation primate prey species may respond to the threat of predation. EY WORDS: predation; antipredator behavior; colobus; chimpanzees; sociality. “During any given day, an animal may fail to obtain a meal and go hungry, or it may fail to obtain matings and thus realize no reproductive success, but in the long term, the day’s shortcomings may have minimal influence on lifetime fitness. Few failures, however, are as unforgiving as the failure to avoid a predator: being killed greatly decreases future fitness.” Lima and Dill (1990) INTRODUCTION The importance of predation as an evolutionary force that shapes anal behavior is not in dispute. Since Darwin, biologists have recognized atThursday, February 3, 2011 in nature must balance the need to find food and mates social animals 4 ften competing for both, with the necessity to avoid capture by a predator followed by unidentified predators (9.0%, n = 323), canids and hyaenids (7.0%, n = 253), reptiles (5.4%, n = 194), and small carnivores (3.3%, n = 118). Table 2.1 requires explanation lest the reader equate the number of predation events listed with the number of identified primate predator species. There is no direct cause and effect relationship between these two variables because the number of predation events is not random but, rather, the outcome of studies directed at specific primates or predators. Thus, the data on unsuccessful attacks, successful predations, and suspected predations by felids, raptors, canids and hyaenids, small carnivores, and reptiles are representative of those primates or predator species that Major predators 1000 900 Felids Raptors Canids & Hyaenids Small carnivores Reptiles Unidentified Predators Total Recorded Predations 800 700 600 500 400 300 200 100 0 Africa Madagascar Asia Neotropics F IGURE 2.1. Overall magnitude of recorded predation on primates (Data source: Hart, 2000) 3,592 instances of predation, Hart 2007 Thursday, February 3, 2011 5 2. A Biogeographical Analysis Successful Predations + Unsuccessful Attacks (Africa) 1,000 No. of Predations 35 800 600 400 200 0 Arb/Diur Variation by region 40 D. Hart Arb/Noct 47 n/a Terr/Diur n/a 806 1 353 Primate Wt < 2kg Primate Wt > 2kg F IGURE 2.4. Comparison of recorded predation on African primates weighing <2 kg and >2 kg in three ecological groups; n/a denotes no primate species exist in that category (Data source: Hart, 2000) Madagascar Region: Madagascar 200 180 160 No. of predations (Macaca), and arboreal-nocturnal primates under 2 kg (Tarsius, Nycticebus, Loris). The large diurnal species are preyed on by leopards, tigers, dholes, jackals, crocodiles, and snakes, but until recently there were so few studies on small, nocturnal Asian primates (Rasmussen, 1997) that only two incidents were available for examination at the time of this meta-analysis. There are three genera of prosimians in Asia, half the number found in Africa and Madagascar, but the current surge in field research on nocturnal Asian primates has greatly expanded knowledge about predation on these speciesBiogeographical Analysis 1999, 2003; 2. A (see Wiens and Zitzmann,33 Gursky, 2002, 2003, 2005; Lakshmi and Mohan, 2002; Nekaris, 2003; Nekaris and Region: Africa Jayewardene, 2004). 900 800 Successful Unsuccessful Africa The Neotropics 700 No. of Predations Successful Unsuccessful 140 120 100 80 60 40 20 0 s rs C H ani ya ds en & id s lid es U pr nid ed en at tif or ied s to til Fe S ca ma rn ll iv or es ap ep to rs C an H ya ids en & id s Sm ca a rn ll iv or es 38 D. Hart Region: Neotropics Region: Asia F IGURE 2.5. Comparison of successful and unsuccessful attacks by six categories of preda- F IGURE 2.2. Comparison of successful and unsuccessful attacks by six categories of predators in Madagascar (Data source: Hart, 2000) 300 450source: Hart, 2000) tors in Africa (Data Successful Successful Asia Neotropics No. of Predations Thursday, 87.9% (n = 29) of eagle prey were monkeys, and Struhsaker & Leakey (1990) 250 estimated this figure to be 83.7% (n = 41). Mitani et al., (2001) determined that primates composed the vast majority of crowned hawk-eagle prey items (82%, 200 n = 74) in the 150 Ngogo study site in Kibale during a 37-month study. At another 100 research site in the Kiwengoma Forest Reserve, Tanzania, the skeletal remains 100 found in one crowned hawk-eagle nest were “90% dominated by blue monkey” 50 (Msuya, 1993, p.120). The geographic range of this raptor is extensive throughout 50 the tropical belt of Africa. New research is finding that crowned hawk eagles exert 0 0 much the same predation pressure on monkeys across different parts of their range (cf. Mitani et al., 2001; Shultz, 2001, 2002). There are no arboreal-nocturnal primates weighing more than 2 kg in Africa, and there are no terrestrial-diurnal primates weighing less than 2 kg. (Of course, no terrestrial-nocturnal primates exist of any weight in any region.) Predation F IGURE 2.7. Comparison of successful and unsuccessful attacks by six categories of predawas recorded2.9.the remainingof successful andcategories identified by six categories of predaF IGURE in Comparison four ecological unsuccessful attacks in Figure 2.4. tors in Asia (Data source: Hart, 2000) The single dataNeotropics (Data source: Hart, 2000) tors in the point representing small, arboreal-diurnal primates refers to predation on the talapoin monkey (Miopithecus talapoin), the only African primate species in this category. The remaining three groups are dominated by guenons, mangabeys, and colobus in the arboreal-diurnal, over-2-kg category; arborealWith regard to currently available data on primate predators, canids and hyae- primates under 2 kg refer to galagos and lorisids; terrestrial-diurnal nocturnal nids are not heavily represented in any region. Nevertheless, Asian canids—the primates over 2 kg include apes and baboons. There Februarygolden jackal (Canis aureus) and the dhole (Cuon alpinus)—figure prominentlyare some interesting patterns that can be inferred from Figure 2.4. More 3, 2011 terrestrial as predators (Johnsingh, 1980; Newton, 1985; Stanford, 1989; D’Cunha, 1996; primate genera (n = 7) have evolved in Africa than other regions, and see Figure 2.7). Several Asian canids not previously considered primate predators 150 s rs C H ani ya ds en & id s es No. of Predations 400 250 Unsuccessful Unsuccessful small carnivores (specifically, the fossa, Cryptoprocta ferox) are important as pri350 may be a primate specialist. Studies of this raptor in the Kibale forest of Uganda mate predators. Indeed, more than half of the predation data for all four regions included200the small carnivore category of Table 2.1 refer to the fossa. This is found high percentages of primates in eagle diets; Skorupa (1989) noted that in 300 til es U pr nid ed en at tif or ied s R R R Figure 2.9 represents an overview of primate predation in the Neotropics. The 600 paucity 500 felid predation is readily apparent despite the fact that two large cat of species:400 jaguar (Panthera onca) and puma (Felis concolor), and four small felids: ocelot ( F . pardalis), jaguarundi ( F . yagouroundi), margay ( F . wiedii), and oncilla 300 ( F . tigrina), have been identified as primate predators. 200 A variety of small hawk and falcon species inhabit Central and South American forests. 100 Neotropical raptor species are twice as numerous as Old World species mainly because of the ubiquitous small forest falcons of the genus Micrastur. 0 Thiollay (1985) describes the hunting techniques of small rainforest hawks and falcons as a combination of active and inactive behaviors; sitting motionless Fe lid s ap R ep lid s U pr nid ed en at tif or ied s s to rs S ca ma rn ll iv or e S ca ma rn ll iv or es til e lid to s Pr nid ed en at tifi or ed s C hy ani ae ds ni & ds Fe R ap til R ep Fe R ap R ep U Hart 2007 6 Research on predation: complications Hard to observe Influenced by presence of observer Human disturbance alters predator densities Data anecdotal Important to separate rate from risk Thursday, February 3, 2011 7 Predation Expect: low predation rates on primates (b/c slow life histories imply low extrinsic mortality) Why? arboreality? (few predators, easy escape) gregariousness? Thursday, February 3, 2011 8 Predation rates on red colobus Gombe Kanyawara Ngogo % killed per year % adults 5-15 20-25 < 1? > 10? 15-53 Mahale 5? 30 Tai 10? 45 Thursday, February 3, 2011 9 0 12 8 9 42 1 13 7 9 38 5 17 9 15 64 0 26 11 8 70 0 4 6 17 31 0 6 4 3 0 6 4 4 2 2 3 13 0 9 4 17 0 10 7 15 3 30 23 55 7 50 28 44 0 54 30 36 Predation 20 34 on red colobus rates 48 135 162 177 19 17 gers, and (2) there were he chimpanzee territory tween groups between e. For example, the ratio ales was approximately es was 1:3 in most cases epresent between 0 and etween 0 and 16% and approximately 15 red nge of the chimpanzee an group size of 33.75, efore estimated that the was 506, 608, and 664 03, respectively. Using ange size February 3, 2011 ) (Teelen 2005 Thursday, ealed a similar size of Table 5 Off-take of red colobus monkeys from different years in percent of the total population of red colobus monkeys in the home range of the Ngogo chimpanzee community in 2001, 2002, and 2003, as well as off-take from the population of red colobus as determined with density data No. of kills/year 1995–98a 1995–99b 1998–99b 2000c 2001c 2002c a b c 2001 20.2 33.0 51.0 44.3 45.0 – 2002 16.8 27.5 42.5 36.9 37.5 53.0 2003 15.4 25.2 38.9 33.8 34.4 48.6 Density 16.7 27.3 42.1 36.6 37.2 52.6 102 167 258 224 228 322 Mitani and Watts (1999) Watts and Mitani (2002) J. Mitani and D. Watts, unpublished data Ngogo: Teelen 2008 10 Predation rates among regions 2. A Biogeographical Analysis Mean Estimated Predation Rate (%) 10 9 8 7 6 5 4 3 2 1 0 Africa Madagascar Region Asia Neotropics 43 F IGURE 2.12. Comparison of mean estimated predation rates by region (Data sources: Hart, 2000; Mitani et al., 2001; Shultz, 2003) TABLE 2.2. Estimated predation rates for primate weight and ecological groupings (Data sources: Hart, 2000; Mitani et al., 2001; Shultz, 2003). Primates < 2 kg Arboreal Diurnal (n = 7) Primates < 2 kg Arboreal Nocturnal (n = 4) Mean EPR 15.8% Range 8.6–25.0% Primates > 2 kg Arboreal Diurnal (n = 38) Mean EPR 5.4% Range 0–18.0% Primates > 2 kg Terrestrial Diurnal (n = 44) Mean EPR 4.4% Range 0–15.0% Mean EPR Thursday, February 3, 2011 7.0% Range 1.0–15.0% Hart 2007 11 Effects of chimpanzee hunting intensity on colobus demography J % immatures group size G 50 J 40 30 20 10 0 N edge N Center S G J G G J G J J G S edge Gombe; Stanford 1999 12 Chimpanzee community territory Thursday, February 3, 2011 Predation 1. Predation on primates: general >2. Anti-predator strategies 3. Predation by primates: chimpanzees Thursday, February 3, 2011 13 Group size Thursday, February 3, 2011 14 Predation rates decrease with increased group size h PATTERNS OF PREDATION O N predation in 1987 was estimated to be PRIMATES at least 45%, which is a n exceptionally high rate even for this p o p ~ l a t i o n . ~ ~ Given these problems, it is not surA final reason for t h e lack of docu- prising that there has been heavy relimentation of predation o n primates is a n c e o n i n d i r e c t e v i d e n c e a n d that such studies a r e likely t o involve o p p o r t u n i s t i c r e p o r t s of p r e d a t i o n great investment of time and energy in events and attempts. A s the data have return for relatively few d a t a points trickled in, however, interesting patbecause predators eat other animals terns have begun to emerge. besides primates. Primatologists a r e 1 ) Large primates should be inherunderstandably more eager to con- ently less vulnerable t o predation t h a n year). This may in fects of body size may b e stronger group size, if such species a r e legitim find a n association a n d estimated p r e groups within a po failed to find a sign 3b). The fact that e including ~ h i m p a b -t 0 301 a t r = -0.49, p < 0.01 -t 301 a t r = -0.40, c U m 0 t ! a a n 0 U 10 - .lFi c v) c .E c v) c W 0 0 1 2 3 4 5 oi 0 = I I 5 10 Ln ( G r o u p Size) B o d y size Figure 2. The reiationshipbetween estimated predation rate and group size (a) and body size (b), among Smaller primate sp From data in Cheney and Wrangham?, Goodman et ai.30 and Clutton-Brockand Harvey.34 n=31 non-ape primates Isbell (1994) Thursday, February 3, 2011 15 Mean group size ns the variance within a variy each taxonomic level, could to empty cells. Therefore, a sis (using dummy variables to ta) was used to examine the f variance explained at each examine different elements of analysis describes the general y without claiming signi®cance able 1). tion levels explain the largest in the demographic variables, nsistently accounts for more up characteristics do appear to size and composition appear ocal environment. Subsequent conducted at the population ns derived from these analyses ide some explanation for the composition, the number of d into two categories: unimale ns with groups containing an ss were coded as unimale, and ter than 1.2 males per group ale. The selection of 1.2 as the to 1.0, was to allow for the econd adult males occurring in population as a whole was RESULTS Group size Species subject to higher Overall, group size varied function of predation predation riskmean,asminimum and maximum livea in larger groups risk categories. The group sizes under different predation risks are shown in Fig. 1. There is a clear effect for group size to be larger 90 80 70 60 50 40 30 20 10 0 Low Medium Predation risk High Minimum Mean Maximum Fig. 1. Mean and standard error for the minimum, mean and maximum group sizes of populations under low, medium and high predation risk. Hill & Lee (2000) Thursday, February 3, 2011 16 Large groups can attract predators Stanford (2002) Thursday, February 3, 2011 17 No evidence that chimpanzee hunting -> larger colobus groups 50 - CHIMPS Jozani Abuko Tana + CHIMPS Gombe Tai Mahale Kanyawara Group size 40 30 20 10 0 0 G K T J M A T 100 200 300 Population density (per sq km) Thursday, February 3, 2011 18 Group composition Thursday, February 3, 2011 19 Variation in number of males 4 # adult males Alouatta Colobus Presbytis 1 1 # adult females 10 van Schaik & Horstermann (1994) Thursday, February 3, 2011 20 Variation by region 40 D. Hart 4 # adult males 1 1 38 D. Hart Alouatta Colobus (Macaca), and arboreal-nocturnal primates under 2 kg (Tarsius, Nycticebus, Loris). The large diurnal species are preyed on by leopards, tigers, dholes, jackals, crocodiles, and snakes, but until recently there were so few studies on small, nocturnal Asian primates (Rasmussen, 1997) that only two incidents were available for examination at the time of this meta-analysis. There are three genera of prosimians in Asia, half the number found in Africa and Madagascar, but the current surge in field research on nocturnal Asian primates has greatly expanded knowledge about predation on these species (see Wiens and Analysis 1999, 2003; 2. A Biogeographical Zitzmann, 33 Gursky, 2002, 2003, 2005; Lakshmi and Mohan, 2002; Nekaris, 2003; Nekaris and Region: Africa Jayewardene, 2004). 900 800 Africa: Colobus The Neotropics 700 No. of Predations Successful Unsuccessful Presbytis 10 # adult females Region: Asia Asia: Presbytis Successful Unsuccessful to rs C an H ya ids en & id s Sm ca a rn ll iv or es 300 250 Thursday, found high percentages of primates in eagle diets; Skorupa (1989) noted that 300 87.9% (n = 29) of eagle prey were monkeys, and Struhsaker & Leakey (1990) 250 estimated this figure to be 83.7% (n = 41). Mitani et al., (2001) determined that 150 primates composed the vast majority of crowned hawk-eagle prey items (82%, 200 n = 74) in the Ngogo study site in Kibale during a 37-month study. At another 100 150 research site in the Kiwengoma Forest Reserve, Tanzania, the skeletal remains found in one 100 crowned hawk-eagle nest were “90% dominated by blue monkey” 50 (Msuya, 1993,50 p.120). The geographic range of this raptor is extensive throughout the tropical belt of Africa. New research is finding that crowned hawk eagles exert 0 0 much the same predation pressure on monkeys across different parts of their range (cf. Mitani et al., 2001; Shultz, 2001, 2002). There are no arboreal-nocturnal primates weighing more than 2 kg in Africa, and there are no terrestrial-diurnal primates weighing less than 2 kg. (Of course, no terrestrial-nocturnal primates exist of any weight in any region.) Predation F IGURE 2.7. Comparison of successful and unsuccessful attacks by six categories of predawas recorded in Comparison of four ecologicalunsuccessful attacks by six categories of predaF IGURE 2.9. the remaining successful and categories identified in Figure 2.4. tors in Asia (Data source: Hart, 2000) The single data point representing small, arboreal-diurnal primates refers to pretors in the Neotropics (Data source: Hart, 2000) dation on the talapoin monkey (Miopithecus talapoin), the only African primate species in this category. The remaining three groups are dominated by guenons, mangabeys, and colobus in the arboreal-diurnal, over-2-kg category; arborealWith regard to currently available data on primate predators, canids and hyae- primates under 2 kg refer to galagos and lorisids; terrestrial-diurnal nocturnal nids are not heavily represented in any region. Nevertheless, Asian canids—the primates over 2 kg include apes and baboons. There Februarygolden jackal (Canis aureus) and the dhole (Cuon alpinus)—figure prominentlyare some interesting patterns that can be inferred from Figure 2.4. More 3, 2011 terrestrial as predators (Johnsingh, 1980; Newton, 1985; Stanford, 1989; D’Cunha, 1996; primate genera (n = 7) have evolved in Africa than other regions, and see Figure 2.7). Several Asian canids not previously considered primate predators 200 No. of Predations Region: Neotropics F IGURE 2.2. Comparison of successful and unsuccessful attacks by six categories of preda450 source: Hart, 2000) tors in Africa (Data Successful 400 Unsuccessful 350 may be a primate specialist. Studies of this raptor in the Kibale forest of Uganda Neotropics: Alouatta s rs C H ani ya ds en & id s es No. of Predations til es U pr nid ed en at tif or ied s R Figure 2.9 represents an overview of primate predation in the Neotropics. The 600 paucity of felid predation is readily apparent despite the fact that two large cat 500 species: jaguar (Panthera onca) and puma (Felis concolor), and four small felids: 400 ocelot ( F . pardalis), jaguarundi ( F . yagouroundi), margay ( F . wiedii), and oncilla 300 ( F . tigrina), have been identified as primate predators. 200 A variety of small hawk and falcon species inhabit Central and South American 100 forests. Neotropical raptor species are twice as numerous as Old World species mainly because of the ubiquitous small forest falcons of the genus Micrastur. 0 Thiollay (1985) describes the hunting techniques of small rainforest hawks and falcons as a combination of active and inactive behaviors; sitting motionless Fe lid s ap R ep lid s U pr nid ed en at tif or ied s s to rs S ca ma rn ll iv or e S ca ma rn ll iv or es til e lid to s Pr nid ed en at tifi or ed s C hy ani ae ds ni & ds Fe R ap til R ep Fe R ap R ep U Hart 2007 21 Low Medium Predation risk High Fig. 2. Mean number of females in unimale groups under low, medium and high predation risk. Mean residual + 95% CI as a positive linear funcfemales present in the , P < 0.001). Adding prethe regression analysis d in the number of males 27.24, P < 0.001). m multimale groups in r of females (t = 73.82, er group size (t = 75.18, groups are larger and sting that predation risk between these types of 0.6 0.4 0.2 0.0 –0.2 –0.4 –0.6 n= r trends to the total data roup size in the higher to small sample sizes, the uced (mean group size: group sizes: F2,16 = 4.53, m, high-low), P < 0.05; : F2,16 = 0.97, P > 0.40). show no variance in the p size changes, and thus up size associated with d through an increase in ig. 2). Female group size signi®cantly February 3,the Thursday, between 2011 (ANOVA: F2,23 = 3.79, 21 Low 37 Medium Predation risk 14 High Fig. 3. Mean and con®dence limits of residuals plotted from regression line of number of males on number of females for multimale groups in populations under low, medium and high predation risk. (r2 = 0.483, F1,70 = 67.33, P < 0.001). Predation risk again is a signi®cant factor in the overall regression (r2 = 0.518, F2,69 = 26.48, P < 0.001). Residuals from the overall regression line were compared between risk categories (Fig. 3), where residuals represent the differ- Hill & Lee (2000) 22 Less predation if more male defenders Stanford (2002) Thursday, February 3, 2011 23 Polyspecific associations Thursday, February 3, 2011 24 America folivore predation males per female polyspecific associations Thursday, February 3, 2011 Africa colobus C-H eagle chimps high common Asia langurs (none) low absent howlers harpy eagle high common 25 Birth synchrony Saimiri oerstedii Thursday, February 3, 2011 26 Birth synchrony 396 I I I xxx xx xCensus of Adult Females o= F- I x xxx III II x I 10- x II III x x x ~ x x x x x x x~o( x 8 I IIIII I I 6 4 2 1 5 9 1 3 ' 1 7 21 2 5 2 9 4 February I ,11, ,lift I I I I ....... 8 12 16 2 0 2 4 2 8 1 5 9 13 17 21 2 5 2 9 April DATE I ........ I,, 0 March N (8 F i g . 2. D i s t r i b u t i o n o f 1984 b i r t h s in t h e m a i n s t u d y t r o o p ( P T ) a n d t w o a d j a c e n t t r o o p s ( M W a n d CT). T i m i n g o f births was only precisely d e t e r m i n e d for the P T t r o o p . Censuses o f the adult females in t h e o t h e r two t r o o p s were c o n d u c t e d at irregular intervals indicated by an • The last birth in the C T t r o o p o c c u r r e d a b o u t 15 J u n e a n d is n o t f i g u r e d F i g . 3. t - t reproduct reproduct offspring number o of P<0.0 27 Boinski (1987) Thursday, February 3, 2011 Other correlates of predation pressure Thursday, February 3, 2011 28 rgy in oints imals s are con- o p p o r t u n i s t i c r e p o r t s of p r e d a t i o n events and attempts. A s the data have trickled in, however, interesting patterns have begun to emerge. 1 ) Large primates should be inherently less vulnerable t o predation t h a n Predation rates decrease with increased body size b h a n d estimated p r e d a t i o n r a t e across groups within a population of vervets failed to find a significant effect3*(Fig. 3b). The fact that evenlargerprimates, including ~himpanzees,~5-36e s u b ar , p < 0.01 -t 301 a t r = -0.40, p < 0.02 a a .E c v) c 3 4 5 oi 0 = I I - I 5 10 15 p Size) B o d y size ( k g ) ated predation rate and group size (a) and body size (b), among Smaller primate species (excluding apes). , Goodman et ai.30 and Clutton-Brockand Harvey.34 n=31 non-ape primates Isbell (1994) Thursday, February 3, 2011 29 Predation rates decrease 66 Evolutionary Anthropology closer to human settlements 8 h a 1001 806040- b c [r al m . s r = 0.95, I001 8060- p < 0.05 g P) w .* 0 .c o U C m 2 m U c 2 n U 2 7J a , .E" - CI Q 20- -; .a , w 40- 20- w 6 7 0 . I ' D i s t a n c e f r o m h u m a n s e t t l e m e n t (km) Figure 3 . The reiationship between estimated predation rate and distance to human habitation (a) an monkeys in Amboseli, K e n ~ a . 3 ~ 5 vervet groups ject to predation adds credence to the Thursday, February 3, 2011otion t h a t predation could have been n a strong selective force o n early homi- Familiarity with Isbellmay mean a n area (1994) knowing the locations of shelters a n30 d t h e patterns of individual predators, m re tr Factors thought to affect group size generally GS generally GS contest feeding competition (between groups) scramble feeding competition (within groups) predation infanticide disease parasites Thursday, February 3, 2011 31 “Predation/Intragroup Competition” Model larger groups smaller groups larger groups Van Schaik 1989; Sterck et al 1997 Thursday, February 3, 2011 32 Predation 1. Predation on primates: general 2. Anti-predator strategies >3. Predation by primates: chimpanzees Thursday, February 3, 2011 33 Explaining chimpanzee hunting Social factors male social bonding meat for sex Ecological factors nutrient shortfall nutrient surplus habitat types ecological factors favor cooperative hunting Gilby et al. 2006 Thursday, February 3, 2011 34 Explaining chimpanzee hunting level of analysis is important Probability that at least one male hunts month as an indicator 0.85 0.8 0.75 0.7 0.65 0.6 0.55 0.5 1 2 3 4 5 6 7 8 9 10 11 12 13 Number of adult male chimpanzees ty-level and focal-level e primary explanatory th simple (unadjusted) utcome either ‘hunt ed’ (yes/no), as approtentially confounding iables, we ran multiple with all interactions ber of swollen females tion type (woodland, ion. We refer to these e ran four MLR models: (2) focal-level hunting ccess (MLR 3) and (4) We reduced each model , removing interaction 5 (type III analysis). 3, 2011 Thursday, February We f significance. For all fo- Figure 1. Relation between the probability of hunting by at least one male and adult male party size. The regression line is from an unadjusted logistic regr es sion. Error ba rs represent 95% confidence intervals. hunting probability/group ratio ¼ 1.06, c2 ¼ 8:82, P ¼ 0.003; Fig. 1). The magnitude 1 and significance of this association increased after controlling for the confounding factors (see Methods; MLR 1: Gombe: Gilby et al. 2006 odds ratio ¼ 1.12, P < 0.0001; Table 2). There was no asso35 ciation between hunting by the focal male and adult male 2 Kanyawara 1996-2000. NUMBER OF HUNTS PER 1000 HOURS 61 hunts, 9278 hours r2 = 0.90 n = 11 P < 0.0001 Number of Adult + Adolescent Males Thursday, February 3, 2011 36 DISCUSSION We used 25 years of data from Gombe National Park, the longest running study of free-ranging chimpanzees, to examine male hunting behaviour. We focused upon hunts of red colobus monkeys, chimpanzees’ preferred prey. At least one male hunted in 69% of all recorded encounters with colobus monkeys, which is consistent with data from earlier studies at Gombe (71.5%, Stanford et al. 1994a; 1 0.9 Explaining chimpanzee hunting level of analysis is important Probability that at least one male hunter succeeded 0.8 0.7 0.6 0.5 0.4 0.3 the long-term data se evidence that this sam cerning the factors in This study is the fi simultaneously measu composition, vegetat hunting by specific c new insights into the ecological factors in th evidence to support so the results indicate ecological factors in predators. Social Factors Male social bonding hunting success rate/group Thursday, February 3, 2011 The male social-bon hunt to obtain meat 0.2 alliances and predicts male party size and h 0.1 2001). This prediction 0 1 2 3 4 5 6 7 8 9 10 11 12 13 parties are more likely is evidence that allies Number of adult male chimpanzees 2000) and are often to Figure 4. Relation between the probability of hunting success by at travel together exclusi least one male and adult male party size. The regression line is from indeed be more likely a n u n a d j u s t e d l o g is t i c r eg r e s s i o n . E r r o r ba r s r e p r e s e n t 9 5 % co n fi dence intervals. Gombe: a positive association Gilby et al. 2006 37 % of hunts with at least one kill 100 Gombe 80 Kanyawara 60 40 20 For each site: r2 = 0.70, n = 13, p < 0.001 0 0 2 4 6 8 10 12 Number of males in party (adult + adolescent) Thursday, February 3, 2011 38 analyses. The party-level hunting rate (hunts/encounters) was 0.69, compared to a focal hunting rate of 0.46 (Table 1). Fifty-three per cent of party-level hunts were successful, with a mean Æ SD of 1.76 Æ 1.1 kills per successful hunt (range 1–7; Table 1). The focal male’s success rate was 0.32 (Table 1). Explaining chimpanzee hunting Male Soci of analysis is important level al Bonding The odds that at least one male hunted increased significantly with adult male party size (unadjusted odds Table 1. Hunting totals Encounters Party level* Focal levely 1087 991 Hunts 751 453 Hunting rate 0.69 0.46 Table 2, Fig. 3). Similarly, there was a negative unadjusted association between hunting by the focal male and the number of swollen females (c2 ¼ 5:90, P ¼ 0.05). This neg2 ative relationship remained when we controlled for the confounding factors, but was tempered by the aforementioned interaction between adult male party size and swollen females. In parties with three adult males, the odds that the focal male hunted were 69% lower in parties with two or more swollen females than in parties with no swollen females (odds ratio ¼ 0.31, P ¼ 0.006; Table 3, Fig. 2). In parties with 10 adult males, however, there was no such association (P ¼ 0.53). Successful hunts 399 146 Success rate 0.53 0.32 Kills 703 151 Kill rate 1.76 1.03 *At least one male was present during all encounters and participated in all hunts, successful hunts and kills. yThe focal chimpanzee was an adult male in all encounters and participated in all hunts, successful hunts and kills. Gombe: Gilby et al. 2006 Thursday, February 3, 2011 39 Explaining chimpanzee hunting Social factors male social bonding: share meat with males in exchange for social support meat for sex: sharing with sexually receptive females in exchange for sex Thursday, February 3, 2011 40 Explaining chimpanzee hunting male social bonding 174 A N I M A L B E H A V I O U R , 72, 1 Probability that focal male hunted 0.6 0.5 0.4 0.3 0.2 0.1 0 1 2 0 Swollen females 2+ Swollen females 3 4 5 6 7 8 9 10 11 12 Number of adult male chimpanzees (including focal) 13 Table 4. Outp Parameter Adult male Swollen fe Vegetation Leaf/pith hunting rate for focal Thursday, February 3, 2011 Figure 2. Relation between the probability of hunting by the focal male, adult male party size and number of swollen females. The regr e s s io n l i n es a r e f r o m a G E E m ul t i pl e l o g i s t ic r e gr e s s i on ( MLR 2 , Table 3). Error bars represent 95% confidence intervals. Gombe: Gilby et Output from by at least o females, vege values are sho al. 2006 confounding factors were controlled for, this relationship significantly 41 estimate ¼ À Explaining chimpanzee hunting meat for sex Probability that at least one male hunted P ¼ 0.7; Table 5). Adult male party size was positively correlated with the number of monkeys killed per hunt (mixed effects linear regression: parameter estimate ¼ 0.12, F1,749 ¼ 91.58, P < 0.0001) and total kilograms of meat secured (parameter estimate ¼ 0.35, F1,749 ¼ 48.65, P < 0.0001). However, the number of kilograms available per male decreased hunting rate for group Thursday, February 3, 2011 tion between 19 monkeys were r of hunting wer level hunts, resp in the long-term In general, ho affect the overal tested. Consiste 1999–2002 data 1 by at least on 0.9 adult male pa P ¼ 0.0007), bu 0.8 (c2 ¼ 1:62, P ¼ 1 0.7 long-term stud 0.6 monthly leaf/p 0.5 hunting (party c2 ¼ 0:09, P ¼ 0 0.4 1 results from the 0.3 Gilby’s data set 0.2 woodland than 0 Swollen females 0.1 c2 ¼ 3:92, P ¼ 0 1 2 + Swollen females type on party-le 0 1 2 3 4 5 6 7 8 9 10 11 12 13 long-term data s tion type on p Number of adult male chimpanzees hunting. While Figure 3. Relation between the probability of hunting by at least one data sets indica male, adult male party size and number of swollen females. The relikely in woodl gression lines are from a logistic regression (MLR 1, Table 2). Error Gilby’s data bars represent 95% confidence intervals. Gombe: Gilby et al. 2006 set 42 Kanyawara 1996-2000. HUNTS / 1000 h No swollen female ≥ 1 swollen female P n.s. Number of Adult + Adolescent Males Thursday, February 3, 2011 43 Explaining chimpanzee hunting meat for sex 174 A N I M A L B E H A V I O U R , 72, 1 Probability that focal male hunted 0.6 0.5 0.4 0.3 0.2 0.1 0 1 2 0 Swollen females 2+ Swollen females 3 4 5 6 7 8 9 10 11 12 Number of adult male chimpanzees (including focal) 13 Table 4. Outp Parameter Adult male Swollen fe Vegetation Leaf/pith hunting rate for focal Thursday, February 3, 2011 Figure 2. Relation between the probability of hunting by the focal male, adult male party size and number of swollen females. The regr e s s io n l i n es a r e f r o m a G E E m ul t i pl e l o g i s t ic r e gr e s s i on ( MLR 2 , Table 3). Error bars represent 95% confidence intervals. Gombe: Gilby et Output from by at least o females, vege values are sho al. 2006 confounding factors were controlled for, this relationship significantly 44 estimate ¼ À Explaining chimpanzee hunting Social factors male social bonding: share meat with males in exchange for social support little support at Gombe, Kanyawara meat for sex: sharing with sexually receptive females in exchange for sex little support at Gombe, Kanyawara Gombe: Gilby et al. 2006; Kanyawara: Gilby et al. 2008 Thursday, February 3, 2011 45 Focal hunt Meat for sex Swollen females (0 / many) Party hunt Focal hunt Party kill Focal kill Nutrient shortfall Diet quality (low / high) Party hunt À Social factors do not explain chimpanzee hunting Focal hunt À Party kill Focal kill Nutrient surplus sults, organized by hypothesis Diet quality (low / high) Party hunt Focal hunt Dependent variable Party kill Predicted effect Focal kill Party hunt Focal hunt Party kill Focal kill þ þ þ Observed effect þ þ þ þ þ 0* þÀ þ þ þ þ þ À À þ þ þ þ þ Explanatory variable Adult males (few / many) Vegetation type Evergreen / Woodland Party hunt Focal hunt Swollen females (0 / many) Cooperative hunting Party hunt Focal / many) Adult males (few hunt Party kill Focal kill Party hunt Focal hunt the probability of hunting by at least one male, a result that Party kill agrees with previously published studies of chimpanzee Focal kill hunting behaviour (Stanford et al. 1994b; Mitani & Watts Evergreen / this seems 2001). At first, Woodland to supportPartymale social-bondthe hunt Focal hunting by speing hypothesis. However, examination ofhunt cific males (rather than by at least Partymale) casts doubt one kill Focal kill on this idea. Adult male party size did not affect hunting Adult focal 3, 2011 Party were fewer than by the males male, provided Thursday, February(few / many) that there kill Focal kill two swollen females present. Although there was a positive Party hunt 0 Focal hunt À 0 Party kill *There was a positive effect of adult males on hunting by the focal male, but only inÀ parties with 2þ swollen 0 females. 0 yWoodland hunts by the focal male were Focal kill to succeed, but only in parties À more likely with fewer than two swollen females. Diet quality (low / high) Diet quality (low / high) þ þ Party kill Focal kill þ Number of prey þ þ þ Total prey mass (kg) À Per capita prey mass (kg) À 0 0 þ 0 þ 0 swollen females and hunting (Stanford et al þ 0 found that hunting was significantly less lik þ 0 if swollen females were present. There are sev þ þ reasons for such contrasting results. First, we þ 0 years of data, compared to Stanford et al. ( þ examined þ relatively short period of just a þ y Gombe: Gilby theþdefinition of data. Second, we restrictedet al. 2006 þ þ males to those that were maximally tumesc 46 þ al. (1994b) included females th 0 Stanford et Explaining chimpanzee hunting Ecological factors nutrient shortfall: hunting supplements the diet nutrient surplus: more hunting when excess energy habitat types: more successful in open habitats ecological factors favor cooperative hunting: payoffs higher when cooperating than when hunting alone Gilby et al. 2006 Thursday, February 3, 2011 47 Explaining chimpanzee hunting Nutrient shortfall/surplus no effect on hunting rate for group Thursday, February 3, 2011 Gilby et al. 2006 48 Explaining chimpanzee hunting Nutrient shortfall/surplus no effect on hunting rate for focal Gilby et al. 2006 Thursday, February 3, 2011 49 Explaining chimpanzee hunting Habitat type more hunts/group in open habitat Thursday, February 3, 2011 Gilby et al. 2006 50 Explaining chimpanzee hunting Cooperation party size associated with total number of kills party size associated with total kilograms meat but individual meat consumption negatively correlated with party size Gilby et al. 2006 Thursday, February 3, 2011 51 Explaining chimpanzee hunting Ecological factors X nutrient shortfall: hunting supplements the diet X nutrient surplus: more hunting when excess energy √ habitat types: more successful in open habitats X ecological factors favor cooperative hunting: payoffs higher when cooperating than when hunting alone Gilby et al. 2006 Thursday, February 3, 2011 52 6 Table 6. Summary A N I M A L B E H A V I O U R , 72, 1 Hypothesis of results, organized by hypothesis Explanatory variable Dependent variable Predicted effect Observe Table 6. Summary of results, organized by hypothesis A N I M A L B E H A V I O U R , 72, 1 Meat for sex Male social bonding (few Party Hypothesis Explanatory variable Predicted hunting? Do ecologicalAdult malesvariable/ many) Dependenthunt factors explain chimpanzeeeffect þ Observed effect Focal hunt þ Male social bonding Adult males (few / many) þ 0* Swollen females (0 / many) Meat for sex Swollen females (0 s Table 6. Summary of results, organized by hypothesi/ many) Party kill Party hunt kill Focal Party hunt Focal hunt Focal hunt Focal hunt Party Dependent hunt Party kill variable Focal hunt Focal kill Party hunt Nutrient shortfall Hypothesis Male socialshortfall Nutrient bonding Meat for sex Diet quality (low Explanatory variable / high) Adult males (few // high) Diet quality (low many) Nutrient surplus Party kill hunt kill Party Party Focal kill hunt kill Focal Focal Party kill Vegetation Evergreen / Woodland Party huntParty hunt Nutrient shortfall type Diet quality (low / high) Focal kill Focal huntFocal hunt Party kill hunt kill Vegetation type Evergreen / Woodland Party Party Focal kill hunt kill Focal Focal Nutrient surplus huntingDiet quality (low / high) / many) Party kill Party huntParty kill Cooperative Adult males (few Focal kill Focal huntFocal kill Party kill kill Cooperative hunting Adult males (few / many) Party Number of prey Focal kill kill Focal X X √ Party huntFocal kill Party hunt Focal hunt Focal hunt Diet quality (low / high) Party huntParty hunt Party kill Swollen females (0 / many) Focal kill Focal huntFocal hunt Party kill þ þ Predicted effect þ þ þÀ þÀ þÀ þÀ þþ þþ Àþ Àþ Àþ Àþ þþ þþ þþ þþ þ þ þ þ þ þ À À þ þ þ þ þ þ þ þ þ þ þ þ þ þ 0* À À ÀObserved effect 0 À 0 þ0 0* 0 À0 À0 0 0 0 0 0 0 À À 0 0 0 0 0 0 Nutrient surplus Diet quality (low / high) 0 0 0 0 0 0 0 0 Vegetation type Number of prey Party huntPer capita prey mass (kg) Total prey Focal hunt mass (kg) Per capita prey mass (kg) Party kill *There was a positive effect of adult males on hunting byFocal focal male, but only in parties the kill Evergreen / Woodland X Total prey mass (kg) þ 0 0þ 0 þy 0þ 00 þ 0 þ þ found if swollen femalessignificantly less likely to severa that hunting was were present. There are occur agrees with behaviour (Stanford et al. 1994b; Mitani & Watts hunting previously published studies of chimpanzee *There was a Februaryeffect of adult et al. on hunting by the focal male, but swollen females were present. There are several possible positive 3, (2011 males 1994b; Mitani & Watts only in parties with 2þ swollen females. Thursday,behaviour Stanford 53 if hunting). At first, this seems to support the male social-bondreasons for such contrasting results. First, we exam 2001 yWoodland hunts by the focal male were more likely to succeed, but only in parties with fewer than two swollen females. reasons for such data, compared toFirst, we examined 25 2001). At first, this seems to support the maleof hunting by speyears of contrasting results. Stanford et al. (199 ing hypothesis. However, examination social-bond- *There was a positive effect of adult males on / many) the focal male, but only in parties with 2þ swollen females. Cooperative hunting Adult males (few hunting by Party kill þ þ yWoodland hunts by the focal male were more likely to succeed, but kill in parties with fewer than two swollen females. 0 Focal only þ Number of prey þ þ swollen females and hunting (Stanford et al. 19 the probability of hunting by at least one male, a result prey mass (kg) that Total þ þ Per capita prey mass (kg) that hunting was significantly1994blikely þ À found swollen females and hunting (Stanford et al. less ), we theagrees withof hunting bypublished studies result that probability previously at least one male, a of chimpanzee yWoodland hunts by the focal male were more likely to succeed, but only in parties with fewer than two swollen females. þþ þþ þþ 0þ þþ þÀ with 2þ swollen females. þ þy þ 0 þ þ À Gombe: Gilby et al. 2006 Importance of “impact hunters” Kanyawara: Gilby et al. 2008 Thursday, February 3, 2011 54 Take home messages 1. Predation is thought to be an important force shaping primate social systems; the evidence of this is limited– in part because studying predation is difficult. 2. Protection against predation is generally assumed to be a benefit of group living. 3. Primate prey species exhibit a variety of anti-predator strategies. 4. There is weak evidence in support of the hypothesis that social factors explain chimpanzee hunting. Among ecological factors examined to date, vegetation type appears to be the only consistent predictor of chimpanzee hunting (at least at Gombe). Thursday, February 3, 2011 55 Question to ponder Plot the relationship between the costs of scramble competition and group size (specifics are not important, but think about whether the line has a positive or negative slope, and whether it is straight or curved). Explain your logic in one sentence. On the same axes, plot the benefits from anti-predator protection as a function of group size (again focusing on slope and shape of the line). Explain your logic in one sentence. Indicate on your graph the predicted optimum group size. Explain your logic in one or two sentences. Finally, assume that infanticide is an important selective force in this species. Will optimum group size be smaller or larger than what you would predict in the absence of infanticide? Why? Thursday, February 3, 2011 56 ...
View Full Document

This note was uploaded on 04/05/2011 for the course ANT 154bn taught by Professor Debello during the Winter '10 term at UC Davis.

Ask a homework question - tutors are online